diff --git a/QAs/002d2729-df71-438d-8396-5895b349e8fd-mcq.json b/QAs/002d2729-df71-438d-8396-5895b349e8fd-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2ea9c101dc421a6b4484fd44d595af0538935329 --- /dev/null +++ b/QAs/002d2729-df71-438d-8396-5895b349e8fd-mcq.json @@ -0,0 +1,934 @@ +[ + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_12", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the white cup?", + "answer": "You put it on the edge of the table.", + "question_chinese": "我刚刚把白色的杯子放哪儿去了？", + "answer_chinese": "你放在桌沿边.", + "evidence": [ + { + "description": "Put the white cup near the outer edge of the table.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:02:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it on the edge of the table.", + "B": "You put it mid-table.", + "C": "You put it beside the sink.", + "D": "You set it down on the floor near the couch, away from the table edge.", + "E": "You placed it on a shelf above the microwave instead of leaving it on the table." + }, + "answer_label": "A", + "answer_text": "You put it on the edge of the table.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_12", + "correct_answer": "You put it on the edge of the table.", + "original_options": { + "A": "You put it on the edge of the table.", + "B": "You put it mid-table.", + "C": "You put it beside the sink.", + "D": "You set it down on the floor near the couch, away from the table edge.", + "E": "You placed it on a shelf above the microwave instead of leaving it on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 16 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_13", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the white cup?", + "answer": "You put it on the inner part of the table.", + "question_chinese": "我刚刚把白色的杯子放哪儿去了？", + "answer_chinese": "你放在了桌子上靠内的地方。", + "evidence": [ + { + "description": "Put the white cup near the outer edge of the table.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:40" + } + }, + { + "description": "Move the white cup from the outer edge of the table inward.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:31" + } + } + ], + "question_timestamps": "00:34:31", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You put it on the cabinet's top shelf.", + "B": "You put it on the countertop edge.", + "C": "You put it on the inner part of the table.", + "D": "You placed it on the back of the chair rather than setting it down on the table.", + "E": "You put it on the lower part of the shelf instead of on the table surface." + }, + "answer_label": "C", + "answer_text": "You put it on the inner part of the table.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_13", + "correct_answer": "You put it on the inner part of the table.", + "original_options": { + "A": "You put it on the cabinet's top shelf.", + "B": "You put it on the countertop edge.", + "C": "You put it on the inner part of the table.", + "D": "You placed it on the back of the chair rather than setting it down on the table.", + "E": "You put it on the lower part of the shelf instead of on the table surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 16 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put that red knife?", + "answer": "That red knife is on the table behind you and to your left.", + "question_chinese": "我把那把红色的刀放在哪儿了？", + "answer_chinese": "那把红色的刀在你左后方的桌子上。", + "evidence": [ + { + "description": "There is a red knife on the table.", + "timestamps": { + "start": "00:27:58", + "end": "00:28:28" + } + } + ], + "question_timestamps": "00:34:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's on the counter in front of you.", + "B": "It's on the shelf above the stove.", + "C": "That red knife is on the table behind you and to your left.", + "D": "That red knife is in a drawer beside the refrigerator, positioned to your right.", + "E": "That red knife is under the sink in the cabinet to your left, not on the table." + }, + "answer_label": "C", + "answer_text": "That red knife is on the table behind you and to your left.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_1", + "correct_answer": "That red knife is on the table behind you and to your left.", + "original_options": { + "A": "It's on the counter in front of you.", + "B": "It's on the shelf above the stove.", + "C": "That red knife is on the table behind you and to your left.", + "D": "That red knife is in a drawer beside the refrigerator, positioned to your right.", + "E": "That red knife is under the sink in the cabinet to your left, not on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 17 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_3", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put that red knife?", + "answer": "That red knife is directly behind you, on the table.", + "question_chinese": "我把那把红色的刀放在哪儿了？", + "answer_chinese": "那把红色的刀在你正后方，就在桌子上。", + "evidence": [ + { + "description": "There is a red knife on the table.", + "timestamps": { + "start": "00:27:58", + "end": "00:28:28" + } + } + ], + "question_timestamps": "00:35:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's next to the sink on the counter.", + "B": "It's inside the drawer by the stove.", + "C": "That red knife is lying on the floor by the door, not placed on the table.", + "D": "That red knife is under the chair in the corner rather than directly behind you.", + "E": "That red knife is directly behind you, on the table." + }, + "answer_label": "E", + "answer_text": "That red knife is directly behind you, on the table.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_3", + "correct_answer": "That red knife is directly behind you, on the table.", + "original_options": { + "A": "It's next to the sink on the counter.", + "B": "It's inside the drawer by the stove.", + "C": "That red knife is lying on the floor by the door, not placed on the table.", + "D": "That red knife is under the chair in the corner rather than directly behind you.", + "E": "That red knife is directly behind you, on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 15 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put that red knife?", + "answer": "The red knife is right on the table to your left, opposite you.", + "question_chinese": "我把那把红色的刀放在哪儿了？", + "answer_chinese": "那把红色的刀就在你的左对面桌子上。", + "evidence": [ + { + "description": "There is a red knife on the table.", + "timestamps": { + "start": "00:27:58", + "end": "00:28:28" + } + } + ], + "question_timestamps": "00:43:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's on the counter to your right.", + "B": "The red knife is right on the table to your left, opposite you.", + "C": "It's under the couch behind you.", + "D": "The red knife is on the shelf above you to your right, not on the table to your left.", + "E": "The red knife is next to the chair across from you rather than on the table opposite you." + }, + "answer_label": "B", + "answer_text": "The red knife is right on the table to your left, opposite you.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_2", + "correct_answer": "The red knife is right on the table to your left, opposite you.", + "original_options": { + "A": "It's on the counter to your right.", + "B": "The red knife is right on the table to your left, opposite you.", + "C": "It's under the couch behind you.", + "D": "The red knife is on the shelf above you to your right, not on the table to your left.", + "E": "The red knife is next to the chair across from you rather than on the table opposite you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 18 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_5", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the tap directly right now?", + "answer": "Yes, it's right above you to the left; you can reach it by stretching out your hand.", + "question_chinese": "我现在能直接够到水龙头吗？", + "answer_chinese": "可以，就在你左上方，伸手就可以碰到。", + "evidence": [ + { + "description": "The tap is right above the sink.", + "timestamps": { + "start": "00:19:03", + "end": "00:19:08" + } + } + ], + "question_timestamps": "00:23:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, it’s too far above and to the left for you to reach from here; you’d need to move closer first.", + "B": "Yes, it's right above you to the left; you can reach it by stretching out your hand." + }, + "answer_label": "B", + "answer_text": "Yes, it's right above you to the left; you can reach it by stretching out your hand.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_5", + "correct_answer": "Yes, it's right above you to the left; you can reach it by stretching out your hand.", + "original_options": { + "A": "No, it’s too far above and to the left for you to reach from here; you’d need to move closer first.", + "B": "Yes, it's right above you to the left; you can reach it by stretching out your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 23 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_11", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the tap directly right now?", + "answer": "No, the tap is far from you; you'll have to get up and walk to the left.", + "question_chinese": "我现在能直接够到水龙头吗？", + "answer_chinese": "不可以，水龙头离你很远，你得起身向左走过去才行。", + "evidence": [ + { + "description": "The tap is right above the sink.", + "timestamps": { + "start": "00:19:03", + "end": "00:19:08" + } + }, + { + "description": "Walk from the tap to the dining table.", + "timestamps": { + "start": "00:33:10", + "end": "00:33:19" + } + } + ], + "question_timestamps": "00:49:59", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, the tap is far from you; you'll have to get up and walk to the left.", + "B": "Yes, it's right above you to the left; you can reach it by stretching out your hand." + }, + "answer_label": "A", + "answer_text": "No, the tap is far from you; you'll have to get up and walk to the left.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_11", + "correct_answer": "No, the tap is far from you; you'll have to get up and walk to the left.", + "original_options": { + "A": "No, the tap is far from you; you'll have to get up and walk to the left.", + "B": "Yes, it's right above you to the left; you can reach it by stretching out your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_8", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get from the sink to here?", + "answer": "Turn right, walk two steps and you're there.", + "question_chinese": "我要怎么从洗手台走到这里？", + "answer_chinese": "向右转，走两步就到了。", + "evidence": [ + { + "description": "Walk from the sink to the stove.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:37" + } + } + ], + "question_timestamps": "00:34:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left, walk two steps.", + "B": "Turn right, walk three steps.", + "C": "Turn right, walk one step, and you should arrive without needing any more steps.", + "D": "Turn right, walk two steps and you're there.", + "E": "Turn left, walk three steps, and you’ll reach this spot from the sink." + }, + "answer_label": "D", + "answer_text": "Turn right, walk two steps and you're there.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_8", + "correct_answer": "Turn right, walk two steps and you're there.", + "original_options": { + "A": "Turn left, walk two steps.", + "B": "Turn right, walk three steps.", + "C": "Turn right, walk one step, and you should arrive without needing any more steps.", + "D": "Turn right, walk two steps and you're there.", + "E": "Turn left, walk three steps, and you’ll reach this spot from the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_4", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get from the sink to here?", + "answer": "Turn around, walk two steps backward and you're there.", + "question_chinese": "我要怎么从洗手台走到这里？", + "answer_chinese": "转身，向后走两步就到了。", + "evidence": [ + { + "description": "Walk from the sink to the dining table.", + "timestamps": { + "start": "00:33:11", + "end": "00:33:16" + } + } + ], + "question_timestamps": "00:46:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn around and take one step back.", + "B": "Turn around, walk two steps backward and you're there.", + "C": "Turn around and take three steps back.", + "D": "Turn around, take two steps forward, and you should arrive at the spot you mean.", + "E": "Turn left, then walk two steps backward until you reach the location you're asking about." + }, + "answer_label": "B", + "answer_text": "Turn around, walk two steps backward and you're there.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_4", + "correct_answer": "Turn around, walk two steps backward and you're there.", + "original_options": { + "A": "Turn around and take one step back.", + "B": "Turn around, walk two steps backward and you're there.", + "C": "Turn around and take three steps back.", + "D": "Turn around, take two steps forward, and you should arrive at the spot you mean.", + "E": "Turn left, then walk two steps backward until you reach the location you're asking about." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_10", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I need the cooked dough, where should I go to get it?", + "answer": "Walk forward three steps, open the rice cooker, and you can get it.", + "question_chinese": "我需要熟面团，我该去哪拿呢？", + "answer_chinese": "向前走三步，打开电饭煲就可以拿到。", + "evidence": [ + { + "description": "Take the dough from the rice cooker.", + "timestamps": { + "start": "00:35:50", + "end": "00:36:08" + } + } + ], + "question_timestamps": "00:54:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk forward three steps, open the rice cooker, and you can get it.", + "B": "Walk forward two steps and open it.", + "C": "Walk forward four steps and open it.", + "D": "Walk forward three steps, open the microwave, and you can pick up the cooked dough there.", + "E": "Walk forward three steps, open the oven, and you should be able to get the cooked dough." + }, + "answer_label": "A", + "answer_text": "Walk forward three steps, open the rice cooker, and you can get it.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_10", + "correct_answer": "Walk forward three steps, open the rice cooker, and you can get it.", + "original_options": { + "A": "Walk forward three steps, open the rice cooker, and you can get it.", + "B": "Walk forward two steps and open it.", + "C": "Walk forward four steps and open it.", + "D": "Walk forward three steps, open the microwave, and you can pick up the cooked dough there.", + "E": "Walk forward three steps, open the oven, and you should be able to get the cooked dough." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_9", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I need the cooked dough, where should I go to get it?", + "answer": "Walk forward a step, open the rice cooker, and you can get it.", + "question_chinese": "我需要熟面团，我该去哪拿呢？", + "answer_chinese": "向你的右边走一步，打开电饭煲就可以拿到", + "evidence": [ + { + "description": "Take the dough from the rice cooker.", + "timestamps": { + "start": "00:35:50", + "end": "00:36:08" + } + }, + { + "description": "Walk from the sink to the rice cooker.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:37" + } + } + ], + "question_timestamps": "00:59:09", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk forward two steps, then open it.", + "B": "Walk backward one step, then open it.", + "C": "Walk forward a step, open the rice cooker, and you can get it.", + "D": "Walk forward a step, open the microwave, and you can get the cooked dough from inside.", + "E": "Walk forward a step, open the oven, and you should be able to take the cooked dough." + }, + "answer_label": "C", + "answer_text": "Walk forward a step, open the rice cooker, and you can get it.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_9", + "correct_answer": "Walk forward a step, open the rice cooker, and you can get it.", + "original_options": { + "A": "Walk forward two steps, then open it.", + "B": "Walk backward one step, then open it.", + "C": "Walk forward a step, open the rice cooker, and you can get it.", + "D": "Walk forward a step, open the microwave, and you can get the cooked dough from inside.", + "E": "Walk forward a step, open the oven, and you should be able to take the cooked dough." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_6", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stalks of peeled vegetables are left now?", + "answer": "There are seven stalks left.", + "question_chinese": "现在，剥下来的青菜还有几根？", + "answer_chinese": "还有七根。", + "evidence": [ + { + "description": "Peel seven stalks of vegetables.", + "timestamps": { + "start": "00:13:12", + "end": "00:13:39" + } + } + ], + "question_timestamps": "00:13:44", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are six stalks left.", + "B": "There are eight stalks left.", + "C": "There are seven stalks left.", + "D": "There are nine stalks left after you finished preparing the peeled vegetables.", + "E": "There are five stalks left now, since you used up more than expected." + }, + "answer_label": "C", + "answer_text": "There are seven stalks left.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_6", + "correct_answer": "There are seven stalks left.", + "original_options": { + "A": "There are six stalks left.", + "B": "There are eight stalks left.", + "C": "There are seven stalks left.", + "D": "There are nine stalks left after you finished preparing the peeled vegetables.", + "E": "There are five stalks left now, since you used up more than expected." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_7", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stalks of peeled vegetables are left now?", + "answer": "There are none left, because they were all cut by you.", + "question_chinese": "现在，剥下来的青菜还有几根？", + "answer_chinese": "一根都没有了，因为都被你切了。", + "evidence": [ + { + "description": "Peel seven stalks of vegetables.", + "timestamps": { + "start": "00:13:12", + "end": "00:13:39" + } + }, + { + "description": "Chop up the seven stalks of vegetables.", + "timestamps": { + "start": "00:17:55", + "end": "00:19:17" + } + } + ], + "question_timestamps": "00:29:28", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There is one stalk left.", + "B": "There are two stalks left.", + "C": "There are three stalks left, because you only cut part of the peeled vegetables.", + "D": "There are four stalks left now, since you stopped before cutting all of them.", + "E": "There are none left, because they were all cut by you." + }, + "answer_label": "E", + "answer_text": "There are none left, because they were all cut by you.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_7", + "correct_answer": "There are none left, because they were all cut by you.", + "original_options": { + "A": "There is one stalk left.", + "B": "There are two stalks left.", + "C": "There are three stalks left, because you only cut part of the peeled vegetables.", + "D": "There are four stalks left now, since you stopped before cutting all of them.", + "E": "There are none left, because they were all cut by you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_14", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many buns are in the steamer now?", + "answer": "There is no bun.", + "question_chinese": "现在蒸笼里有几个包子？", + "answer_chinese": "没有包子。", + "evidence": [ + { + "description": "An empty steamer.", + "timestamps": { + "start": "00:34:34", + "end": "00:34:57" + } + } + ], + "question_timestamps": "00:38:41", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one bun.", + "B": "There is no bun.", + "C": "There are two buns.", + "D": "There are three buns in the steamer right now, sitting next to each other near the center.", + "E": "There are four buns sitting inside the steamer now, so it definitely is not empty anymore." + }, + "answer_label": "B", + "answer_text": "There is no bun.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_14", + "correct_answer": "There is no bun.", + "original_options": { + "A": "There is one bun.", + "B": "There is no bun.", + "C": "There are two buns.", + "D": "There are three buns in the steamer right now, sitting next to each other near the center.", + "E": "There are four buns sitting inside the steamer now, so it definitely is not empty anymore." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_15", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many buns are in the steamer now?", + "answer": "There is one bun.", + "question_chinese": "现在蒸笼里有几个包子？", + "answer_chinese": "有一个包子。", + "evidence": [ + { + "description": "An empty steamer.", + "timestamps": { + "start": "00:34:34", + "end": "00:34:57" + } + }, + { + "description": "Put one bun into the steamer.", + "timestamps": { + "start": "00:39:40", + "end": "00:40:00" + } + } + ], + "question_timestamps": "00:40:56", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are no buns.", + "B": "There are two buns.", + "C": "There are three buns in the steamer right now, arranged together near the middle of the tray.", + "D": "There are five buns inside the steamer at the moment, so it's much more than a single bun.", + "E": "There is one bun." + }, + "answer_label": "E", + "answer_text": "There is one bun.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_15", + "correct_answer": "There is one bun.", + "original_options": { + "A": "There are no buns.", + "B": "There are two buns.", + "C": "There are three buns in the steamer right now, arranged together near the middle of the tray.", + "D": "There are five buns inside the steamer at the moment, so it's much more than a single bun.", + "E": "There is one bun." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_16", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many buns are in the steamer now?", + "answer": "There are two buns.", + "question_chinese": "现在蒸笼里有几个包子？", + "answer_chinese": "有两个包子。", + "evidence": [ + { + "description": "Put one bun into the steamer.", + "timestamps": { + "start": "00:39:40", + "end": "00:40:00" + } + } + ], + "question_timestamps": "00:41:56", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one bun.", + "B": "There are three buns.", + "C": "There are four buns in the steamer right now, so the count is definitely more than two.", + "D": "There are two buns.", + "E": "There are five buns inside the steamer at the moment, which makes it far from just two buns." + }, + "answer_label": "D", + "answer_text": "There are two buns.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_16", + "correct_answer": "There are two buns.", + "original_options": { + "A": "There is one bun.", + "B": "There are three buns.", + "C": "There are four buns in the steamer right now, so the count is definitely more than two.", + "D": "There are two buns.", + "E": "There are five buns inside the steamer at the moment, which makes it far from just two buns." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "002d2729-df71-438d-8396-5895b349e8fd", + "video_length": "00:59:31", + "qid": "002d2729-df71-438d-8396-5895b349e8fd_17", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many buns are in the steamer now?", + "answer": "There are four buns.", + "question_chinese": "现在蒸笼里有几个包子？", + "answer_chinese": "有四个包子。", + "evidence": [ + { + "description": "Put one bun into the steamer.", + "timestamps": { + "start": "00:39:40", + "end": "00:40:00" + } + }, + { + "description": "Put the fourth bun into the steamer.", + "timestamps": { + "start": "00:43:50", + "end": "00:44:00" + } + }, + { + "description": "An empty steamer.", + "timestamps": { + "start": "00:34:34", + "end": "00:34:57" + } + } + ], + "question_timestamps": "00:44:29", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are three buns.", + "B": "There are five buns.", + "C": "There are four buns.", + "D": "There are six buns in the steamer at the moment, not four like you might expect.", + "E": "There are seven buns in the steamer right now, so it is more crowded than before." + }, + "answer_label": "C", + "answer_text": "There are four buns.", + "q_id": "002d2729-df71-438d-8396-5895b349e8fd_17", + "correct_answer": "There are four buns.", + "original_options": { + "A": "There are three buns.", + "B": "There are five buns.", + "C": "There are four buns.", + "D": "There are six buns in the steamer at the moment, not four like you might expect.", + "E": "There are seven buns in the steamer right now, so it is more crowded than before." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + } +] diff --git a/QAs/0031d268-818c-4ec4-a804-935be610a61a-mcq.json b/QAs/0031d268-818c-4ec4-a804-935be610a61a-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..dc8cef31aa85c84393d5b4f5f15462a941311bed --- /dev/null +++ b/QAs/0031d268-818c-4ec4-a804-935be610a61a-mcq.json @@ -0,0 +1,1443 @@ +[ + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_19", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the balcony after going out through this door?", + "answer": "After you go out through this door, turn left first, then right, and then go straight to reach the balcony.", + "question_chinese": "我从这个门出去后如何去到阳台？", + "answer_chinese": "你从这个门出去后先左转，再右转，然后直走可以到达阳台。", + "evidence": [ + { + "description": "You turned left after you went out.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:48" + } + }, + { + "description": "Then you turned right.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:52" + } + }, + { + "description": "You took a few steps forward and reached the balcony.", + "timestamps": { + "start": "00:00:52", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:17:39", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right, then left, then straight.", + "B": "Turn left, then left, then straight.", + "C": "After you go out through this door, turn right first, then left, and then go straight to the balcony.", + "D": "After you go out through this door, turn left first, then right, and then go straight to reach the balcony.", + "E": "After you go out through this door, turn left first, then left, and continue straight until you reach the balcony." + }, + "answer_label": "D", + "answer_text": "After you go out through this door, turn left first, then right, and then go straight to reach the balcony.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_19", + "correct_answer": "After you go out through this door, turn left first, then right, and then go straight to reach the balcony.", + "original_options": { + "A": "Turn right, then left, then straight.", + "B": "Turn left, then left, then straight.", + "C": "After you go out through this door, turn right first, then left, and then go straight to the balcony.", + "D": "After you go out through this door, turn left first, then right, and then go straight to reach the balcony.", + "E": "After you go out through this door, turn left first, then left, and continue straight until you reach the balcony." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 20 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_20", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the balcony after going out through this door?", + "answer": "You need to turn left first, then right, and go straight to reach the balcony.", + "question_chinese": "我从这个门出去后如何去到阳台？", + "answer_chinese": "你需要先左转，接着右转，直走就可以到达阳台。", + "evidence": [ + { + "description": "You turned left first after coming out of this door.", + "timestamps": { + "start": "00:28:01", + "end": "00:28:05" + } + }, + { + "description": "Then you turn right and go straight to the balcony.", + "timestamps": { + "start": "00:28:06", + "end": "00:28:08" + } + } + ], + "question_timestamps": "00:31:49", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right, then left, then straight.", + "B": "You need to turn left first, then right, and go straight to reach the balcony.", + "C": "Turn left, then left, then straight.", + "D": "You need to turn left first, then right, and then turn left to reach the balcony.", + "E": "You need to turn right first, then right, and go straight to reach the balcony." + }, + "answer_label": "B", + "answer_text": "You need to turn left first, then right, and go straight to reach the balcony.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_20", + "correct_answer": "You need to turn left first, then right, and go straight to reach the balcony.", + "original_options": { + "A": "Turn right, then left, then straight.", + "B": "You need to turn left first, then right, and go straight to reach the balcony.", + "C": "Turn left, then left, then straight.", + "D": "You need to turn left first, then right, and then turn left to reach the balcony.", + "E": "You need to turn right first, then right, and go straight to reach the balcony." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_5", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to walk from the room to the balcony?", + "answer": "You need to make two turns.", + "question_chinese": "我刚刚从房间走到阳台需要拐几次弯？", + "answer_chinese": "你需要拐两次弯。", + "evidence": [ + { + "description": "You turned the first corner.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:50" + } + }, + { + "description": "You turned the second corner.", + "timestamps": { + "start": "00:00:51", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You need to make two turns.", + "B": "You need to make one turn.", + "C": "You need to make three turns.", + "D": "You need to make four turns to get from the room to the balcony.", + "E": "You need to make five turns before you can reach the balcony." + }, + "answer_label": "A", + "answer_text": "You need to make two turns.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_5", + "correct_answer": "You need to make two turns.", + "original_options": { + "A": "You need to make two turns.", + "B": "You need to make one turn.", + "C": "You need to make three turns.", + "D": "You need to make four turns to get from the room to the balcony.", + "E": "You need to make five turns before you can reach the balcony." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 12 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_6", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to walk from the room to the balcony?", + "answer": "You need to make two turns.", + "question_chinese": "我刚刚从房间走到阳台需要拐几次弯？", + "answer_chinese": "你需要拐两次弯。", + "evidence": [ + { + "description": "You turned the first corner.", + "timestamps": { + "start": "00:28:00", + "end": "00:28:05" + } + }, + { + "description": "You turned the second corner.", + "timestamps": { + "start": "00:28:06", + "end": "00:28:09" + } + } + ], + "question_timestamps": "00:29:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You need to make one turn.", + "B": "You need to make three turns.", + "C": "You need to make two turns.", + "D": "You need to make four turns to walk from the room to the balcony.", + "E": "You need to make six turns before you arrive at the balcony." + }, + "answer_label": "C", + "answer_text": "You need to make two turns.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_6", + "correct_answer": "You need to make two turns.", + "original_options": { + "A": "You need to make one turn.", + "B": "You need to make three turns.", + "C": "You need to make two turns.", + "D": "You need to make four turns to walk from the room to the balcony.", + "E": "You need to make six turns before you arrive at the balcony." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 12 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_9", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What did I just put the pen on?", + "answer": "You just put the pen on the ground.", + "question_chinese": "我刚刚把笔放在了什么上面？", + "answer_chinese": "你刚刚把笔放在了地上。", + "evidence": [ + { + "description": "You put the pen on the ground.", + "timestamps": { + "start": "00:14:49", + "end": "00:14:58" + } + } + ], + "question_timestamps": "00:15:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it on the table.", + "B": "You put it on the shelf.", + "C": "You just put the pen on the ground.", + "D": "You just put the pen inside the drawer, not down on the floor.", + "E": "You just put the pen beside the chair rather than placing it on the ground." + }, + "answer_label": "C", + "answer_text": "You just put the pen on the ground.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_9", + "correct_answer": "You just put the pen on the ground.", + "original_options": { + "A": "You put it on the table.", + "B": "You put it on the shelf.", + "C": "You just put the pen on the ground.", + "D": "You just put the pen inside the drawer, not down on the floor.", + "E": "You just put the pen beside the chair rather than placing it on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 15 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_10", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What did I just put the pen on?", + "answer": "You just put the pen in the cardboard box in front.", + "question_chinese": "我刚刚把笔放在了什么上面？", + "answer_chinese": "你刚刚把笔放在了前面的纸箱里。", + "evidence": [ + { + "description": "You just put the pen in the cardboard box in front.", + "timestamps": { + "start": "00:27:58", + "end": "00:28:02" + } + } + ], + "question_timestamps": "00:28:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it on the desk.", + "B": "You put it in the drawer.", + "C": "You just put the pen in the backpack on the floor, not in the box in front.", + "D": "You just put the pen on the chair behind you instead of into the cardboard box ahead.", + "E": "You just put the pen in the cardboard box in front." + }, + "answer_label": "E", + "answer_text": "You just put the pen in the cardboard box in front.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_10", + "correct_answer": "You just put the pen in the cardboard box in front.", + "original_options": { + "A": "You put it on the desk.", + "B": "You put it in the drawer.", + "C": "You just put the pen in the backpack on the floor, not in the box in front.", + "D": "You just put the pen on the chair behind you instead of into the cardboard box ahead.", + "E": "You just put the pen in the cardboard box in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_7_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I had just walked out of the room, where was the cement bucket relative to me?", + "answer": "The cement bucket is behind you.", + "question_chinese": "我刚刚走出房间时，水泥桶在我的什么位置？", + "answer_chinese": "水泥桶在你的身后。", + "evidence": [ + { + "description": "The cement bucket is behind you.", + "timestamps": { + "start": "00:17:38", + "end": "00:17:41" + } + } + ], + "question_timestamps": "00:17:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The cement bucket is behind you.", + "B": "The cement bucket is in front of you.", + "C": "The cement bucket is to your left.", + "D": "The cement bucket is to your right when you walk out of the room.", + "E": "When you had just walked out, the cement bucket was in front of you, not behind you." + }, + "answer_label": "A", + "answer_text": "The cement bucket is behind you.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_7_T2", + "correct_answer": "The cement bucket is behind you.", + "original_options": { + "A": "The cement bucket is behind you.", + "B": "The cement bucket is in front of you.", + "C": "The cement bucket is to your left.", + "D": "The cement bucket is to your right when you walk out of the room.", + "E": "When you had just walked out, the cement bucket was in front of you, not behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 17 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_15", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the cement bucket on me when I just walked out of the room?", + "answer": "The cement bucket is behind you.", + "question_chinese": "我刚刚走出房间时，水泥桶在我的什么位置？", + "answer_chinese": "水泥桶在你的身后。", + "evidence": [ + { + "description": "When you just walked out of the room, the cement bucket was behind you.", + "timestamps": { + "start": "00:17:38", + "end": "00:17:40" + } + } + ], + "question_timestamps": "00:17:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The cement bucket is in front of you.", + "B": "The cement bucket is to your left.", + "C": "The cement bucket is to your right as you step out of the room.", + "D": "The cement bucket is behind you.", + "E": "The cement bucket is beside you near the doorway when you walk out." + }, + "answer_label": "D", + "answer_text": "The cement bucket is behind you.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_15", + "correct_answer": "The cement bucket is behind you.", + "original_options": { + "A": "The cement bucket is in front of you.", + "B": "The cement bucket is to your left.", + "C": "The cement bucket is to your right as you step out of the room.", + "D": "The cement bucket is behind you.", + "E": "The cement bucket is beside you near the doorway when you walk out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 13 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_7_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I had just walked out of the room, where was the cement bucket relative to me?", + "answer": "The cement bucket is in front of you.", + "question_chinese": "我刚刚走出房间时，水泥桶在我的什么位置？", + "answer_chinese": "水泥桶在你的前面。", + "evidence": [ + { + "description": "When you just walked out of the room, the cement bucket was front you.", + "timestamps": { + "start": "00:17:55", + "end": "00:17:57" + } + } + ], + "question_timestamps": "00:17:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The cement bucket is behind you.", + "B": "The cement bucket is in front of you.", + "C": "The cement bucket is to your left.", + "D": "The cement bucket is to your right when you have just walked out of the room.", + "E": "The cement bucket is above you as you exit the room, not directly in front." + }, + "answer_label": "B", + "answer_text": "The cement bucket is in front of you.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_7_T1", + "correct_answer": "The cement bucket is in front of you.", + "original_options": { + "A": "The cement bucket is behind you.", + "B": "The cement bucket is in front of you.", + "C": "The cement bucket is to your left.", + "D": "The cement bucket is to your right when you have just walked out of the room.", + "E": "The cement bucket is above you as you exit the room, not directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 15 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_16", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the cement bucket on me when I just walked out of the room?", + "answer": "The cement bucket is behind you.", + "question_chinese": "我刚刚走出房间时，水泥桶在我的什么位置？", + "answer_chinese": "水泥桶在你的身后。", + "evidence": [ + { + "description": "The cement bucket is behind you.", + "timestamps": { + "start": "00:24:23", + "end": "00:24:26" + } + } + ], + "question_timestamps": "00:24:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The cement bucket is in front of you.", + "B": "The cement bucket is behind you.", + "C": "The cement bucket is to your right.", + "D": "The cement bucket is above you as you step out of the room.", + "E": "The cement bucket is on the floor beside you when you walk out." + }, + "answer_label": "B", + "answer_text": "The cement bucket is behind you.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_16", + "correct_answer": "The cement bucket is behind you.", + "original_options": { + "A": "The cement bucket is in front of you.", + "B": "The cement bucket is behind you.", + "C": "The cement bucket is to your right.", + "D": "The cement bucket is above you as you step out of the room.", + "E": "The cement bucket is on the floor beside you when you walk out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 13, + 13 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_3", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the cement bucket when I was just kneeling?", + "answer": "The cement bucket is on your left.", + "question_chinese": "我刚刚跪着时，水泥桶在什么位置？", + "answer_chinese": "水泥桶在你的左侧。", + "evidence": [ + { + "description": "The cement bucket is on your left.", + "timestamps": { + "start": "00:08:54", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:09:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "On your right.", + "C": "The cement bucket is directly in front of you when you kneel down, not off to either side.", + "D": "The cement bucket is on your left.", + "E": "The cement bucket is on your right side while kneeling, close enough to reach without turning around." + }, + "answer_label": "D", + "answer_text": "The cement bucket is on your left.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_3", + "correct_answer": "The cement bucket is on your left.", + "original_options": { + "A": "In front of you.", + "B": "On your right.", + "C": "The cement bucket is directly in front of you when you kneel down, not off to either side.", + "D": "The cement bucket is on your left.", + "E": "The cement bucket is on your right side while kneeling, close enough to reach without turning around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 18, + 17 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_4", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the cement bucket when I was just kneeling?", + "answer": "The cement bucket is on your left.", + "question_chinese": "我刚刚跪着时，水泥桶在什么位置？", + "answer_chinese": "水泥桶在你的左侧。", + "evidence": [ + { + "description": "The cement bucket is on your left.", + "timestamps": { + "start": "00:20:53", + "end": "00:21:09" + } + } + ], + "question_timestamps": "00:21:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The cement bucket is on your right.", + "B": "The cement bucket is in front of you.", + "C": "The cement bucket is on your left.", + "D": "The cement bucket is behind you while you are kneeling down on the floor.", + "E": "The cement bucket is under the table when you are kneeling down nearby." + }, + "answer_label": "C", + "answer_text": "The cement bucket is on your left.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_4", + "correct_answer": "The cement bucket is on your left.", + "original_options": { + "A": "The cement bucket is on your right.", + "B": "The cement bucket is in front of you.", + "C": "The cement bucket is on your left.", + "D": "The cement bucket is behind you while you are kneeling down on the floor.", + "E": "The cement bucket is under the table when you are kneeling down nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 14, + 13 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tiles did I just lay?", + "answer": "You just laid one tile.", + "question_chinese": "我刚刚铺了几块瓷砖？", + "answer_chinese": "你刚刚铺了1块瓷砖。", + "evidence": [ + { + "description": "You have laid one tile.", + "timestamps": { + "start": "00:02:38", + "end": "00:03:51" + } + } + ], + "question_timestamps": "00:04:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You just laid two tiles.", + "B": "You just laid three tiles.", + "C": "You just laid four tiles in a row before stopping to check alignment.", + "D": "You just laid one tile.", + "E": "You just laid five tiles one after another, then paused to adjust them." + }, + "answer_label": "D", + "answer_text": "You just laid one tile.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_1", + "correct_answer": "You just laid one tile.", + "original_options": { + "A": "You just laid two tiles.", + "B": "You just laid three tiles.", + "C": "You just laid four tiles in a row before stopping to check alignment.", + "D": "You just laid one tile.", + "E": "You just laid five tiles one after another, then paused to adjust them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tiles did I just lay?", + "answer": "You just laid three tiles.", + "question_chinese": "我刚刚铺了几块瓷砖？", + "answer_chinese": "你刚刚铺了3块瓷砖。", + "evidence": [ + { + "description": "You have laid the first tile.", + "timestamps": { + "start": "00:02:38", + "end": "00:04:50" + } + }, + { + "description": "You have laid the second tile.", + "timestamps": { + "start": "00:05:30", + "end": "00:06:25" + } + }, + { + "description": "You have laid the third tile.", + "timestamps": { + "start": "00:07:54", + "end": "00:08:22" + } + } + ], + "question_timestamps": "00:11:22", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You just laid two tiles.", + "B": "You just laid three tiles.", + "C": "You just laid four tiles.", + "D": "You just laid five tiles before you finished that section of the floor.", + "E": "You just laid six tiles in total right before you stopped working." + }, + "answer_label": "B", + "answer_text": "You just laid three tiles.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_2", + "correct_answer": "You just laid three tiles.", + "original_options": { + "A": "You just laid two tiles.", + "B": "You just laid three tiles.", + "C": "You just laid four tiles.", + "D": "You just laid five tiles before you finished that section of the floor.", + "E": "You just laid six tiles in total right before you stopped working." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 12 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_13", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I go to put the rag back in the kitchen?", + "answer": "After you come out of this door of the room, turn left, walk straight for a while and then turn left again.", + "question_chinese": "我想要把抹布放回厨房应该怎么走？", + "answer_chinese": "你从房间的这个门出来后左转，直走一段距离后再左转。", + "evidence": [ + { + "description": "You turned left after you went out.", + "timestamps": { + "start": "00:31:53", + "end": "00:31:55" + } + }, + { + "description": "You turned left again.", + "timestamps": { + "start": "00:31:56", + "end": "00:31:58" + } + } + ], + "question_timestamps": "00:32:28", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Exit the room, turn right, then left.", + "B": "After you come out of this door of the room, turn left, walk straight for a while and then turn left again.", + "C": "Exit the room, turn left, then right.", + "D": "After you come out of the door, turn right, walk straight for a while, then turn right again.", + "E": "After you come out of the room, turn left, walk straight for a while, then turn right at the next turn." + }, + "answer_label": "B", + "answer_text": "After you come out of this door of the room, turn left, walk straight for a while and then turn left again.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_13", + "correct_answer": "After you come out of this door of the room, turn left, walk straight for a while and then turn left again.", + "original_options": { + "A": "Exit the room, turn right, then left.", + "B": "After you come out of this door of the room, turn left, walk straight for a while and then turn left again.", + "C": "Exit the room, turn left, then right.", + "D": "After you come out of the door, turn right, walk straight for a while, then turn right again.", + "E": "After you come out of the room, turn left, walk straight for a while, then turn right at the next turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 21 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_14", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I go to put the rag back in the kitchen?", + "answer": "After you leave the room, turn left first, and then turn left again.", + "question_chinese": "我想要把抹布放回厨房应该怎么走？", + "answer_chinese": "你在走出房间后先左转，接着再一次左转。", + "evidence": [ + { + "description": "You turned left once after leaving the room.", + "timestamps": { + "start": "00:31:53", + "end": "00:31:56" + } + }, + { + "description": "Then you turn left again and reach the kitchen.", + "timestamps": { + "start": "00:31:57", + "end": "00:33:07" + } + } + ], + "question_timestamps": "00:33:07", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right, then turn right again.", + "B": "Turn left, then turn right.", + "C": "After you leave the room, turn right first, then turn left when you reach the hallway corner.", + "D": "Once you exit the room, go straight a bit, then make a left turn to head back into the kitchen.", + "E": "After you leave the room, turn left first, and then turn left again." + }, + "answer_label": "E", + "answer_text": "After you leave the room, turn left first, and then turn left again.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_14", + "correct_answer": "After you leave the room, turn left first, and then turn left again.", + "original_options": { + "A": "Turn right, then turn right again.", + "B": "Turn left, then turn right.", + "C": "After you leave the room, turn right first, then turn left when you reach the hallway corner.", + "D": "Once you exit the room, go straight a bit, then make a left turn to head back into the kitchen.", + "E": "After you leave the room, turn left first, and then turn left again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 20 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_17", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get back to my room now?", + "answer": "You should turn around first and then go straight.", + "question_chinese": "我现在要回到房间应该怎么走？", + "answer_chinese": "你要先转身，然后直走。", + "evidence": [ + { + "description": "After you come out of the room, walk straight here.", + "timestamps": { + "start": "00:04:56", + "end": "00:04:59" + } + } + ], + "question_timestamps": "00:05:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go straight without turning around.", + "B": "Turn left, then go straight.", + "C": "You should turn around first, then take a right turn instead of continuing straight.", + "D": "Turn around, walk a few steps, and then turn left to reach your room.", + "E": "You should turn around first and then go straight." + }, + "answer_label": "E", + "answer_text": "You should turn around first and then go straight.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_17", + "correct_answer": "You should turn around first and then go straight.", + "original_options": { + "A": "Go straight without turning around.", + "B": "Turn left, then go straight.", + "C": "You should turn around first, then take a right turn instead of continuing straight.", + "D": "Turn around, walk a few steps, and then turn left to reach your room.", + "E": "You should turn around first and then go straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_18", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get back to my room now?", + "answer": "You should turn around first and then go straight.", + "question_chinese": "我现在要回到房间应该怎么走？", + "answer_chinese": "你应该先转身，再直走。", + "evidence": [ + { + "description": "After you came out of the room, you walked straight to this location.", + "timestamps": { + "start": "00:17:38", + "end": "00:17:42" + } + } + ], + "question_timestamps": "00:17:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go straight without turning around.", + "B": "Turn right, then go straight.", + "C": "You should turn around first, then turn left rather than walking straight ahead.", + "D": "You should turn around first and then go straight.", + "E": "Turn around, take a few steps forward, and then turn right to get back to your room." + }, + "answer_label": "D", + "answer_text": "You should turn around first and then go straight.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_18", + "correct_answer": "You should turn around first and then go straight.", + "original_options": { + "A": "Go straight without turning around.", + "B": "Turn right, then go straight.", + "C": "You should turn around first, then turn left rather than walking straight ahead.", + "D": "You should turn around first and then go straight.", + "E": "Turn around, take a few steps forward, and then turn right to get back to your room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 17 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_8_T1", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which direction is the orange bucket relative to me?", + "answer": "The orange bucket is in front of you.", + "question_chinese": "橙色的桶在我的什么方向？", + "answer_chinese": "橙色的桶在你的前面。", + "evidence": [ + { + "description": "The orange bucket is in front of you.", + "timestamps": { + "start": "00:04:56", + "end": "00:04:59" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The orange bucket is in front of you.", + "B": "The orange bucket is behind you.", + "C": "The orange bucket is to your left.", + "D": "The orange bucket is to your right side, not directly in front of you.", + "E": "The orange bucket is above you, so you would need to look up to see it." + }, + "answer_label": "A", + "answer_text": "The orange bucket is in front of you.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_8_T1", + "correct_answer": "The orange bucket is in front of you.", + "original_options": { + "A": "The orange bucket is in front of you.", + "B": "The orange bucket is behind you.", + "C": "The orange bucket is to your left.", + "D": "The orange bucket is to your right side, not directly in front of you.", + "E": "The orange bucket is above you, so you would need to look up to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 16 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_8_T2", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which direction is the orange bucket relative to me?", + "answer": "The orange bucket is behind you.", + "question_chinese": "橙色的桶在我的什么方向？", + "answer_chinese": "橙色的桶在你的后面", + "evidence": [ + { + "description": "The orange bucket is in behind of you.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:24" + } + } + ], + "question_timestamps": "00:05:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The orange bucket is in front of you.", + "B": "The orange bucket is to your right.", + "C": "The orange bucket is to your left side rather than behind you where you can’t see it.", + "D": "The orange bucket is above you, so it isn’t behind you on the ground.", + "E": "The orange bucket is behind you." + }, + "answer_label": "E", + "answer_text": "The orange bucket is behind you.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_8_T2", + "correct_answer": "The orange bucket is behind you.", + "original_options": { + "A": "The orange bucket is in front of you.", + "B": "The orange bucket is to your right.", + "C": "The orange bucket is to your left side rather than behind you where you can’t see it.", + "D": "The orange bucket is above you, so it isn’t behind you on the ground.", + "E": "The orange bucket is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 15 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_2_T2", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the cement bucket relative to me?", + "answer": "The cement bucket is on the left side.", + "question_chinese": "水泥桶在我什么位置？", + "answer_chinese": "水泥桶在你的左侧。", + "evidence": [ + { + "description": "The cement bucket is on your left.", + "timestamps": { + "start": "00:08:53", + "end": "00:09:09" + } + } + ], + "question_timestamps": "00:09:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The cement bucket is in front of you.", + "B": "The cement bucket is behind you.", + "C": "The cement bucket is on the left side.", + "D": "The cement bucket is to your right side, not on the left side of your body.", + "E": "The cement bucket is somewhere above you, not positioned on the left side near the ground." + }, + "answer_label": "C", + "answer_text": "The cement bucket is on the left side.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_2_T2", + "correct_answer": "The cement bucket is on the left side.", + "original_options": { + "A": "The cement bucket is in front of you.", + "B": "The cement bucket is behind you.", + "C": "The cement bucket is on the left side.", + "D": "The cement bucket is to your right side, not on the left side of your body.", + "E": "The cement bucket is somewhere above you, not positioned on the left side near the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 16, + 16 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_2_T1", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the cement bucket relative to me?", + "answer": "The cement bucket is behind you.", + "question_chinese": "水泥桶在我什么位置？", + "answer_chinese": "水泥桶在你的后面。", + "evidence": [ + { + "description": "The cement bucket is on behind of you.", + "timestamps": { + "start": "00:10:40", + "end": "00:10:45" + } + } + ], + "question_timestamps": "00:10:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The cement bucket is in front of you.", + "B": "The cement bucket is to your left.", + "C": "The cement bucket is to your right side rather than behind you where you’d have to turn around.", + "D": "The cement bucket is above you, so it isn’t located behind you at ground level.", + "E": "The cement bucket is behind you." + }, + "answer_label": "E", + "answer_text": "The cement bucket is behind you.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_2_T1", + "correct_answer": "The cement bucket is behind you.", + "original_options": { + "A": "The cement bucket is in front of you.", + "B": "The cement bucket is to your left.", + "C": "The cement bucket is to your right side rather than behind you where you’d have to turn around.", + "D": "The cement bucket is above you, so it isn’t located behind you at ground level.", + "E": "The cement bucket is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 16 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_7", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How far is the bathtub from me now?", + "answer": "The bathtub is five steps away from you.", + "question_chinese": "现在浴缸离我有几步远？", + "answer_chinese": "浴缸离你有五步的距离。", + "evidence": [ + { + "description": "The bathtub is about five steps away from you.", + "timestamps": { + "start": "00:12:40", + "end": "00:13:27" + } + } + ], + "question_timestamps": "00:13:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "The bathtub is four steps away.", + "B": "The bathtub is six steps away.", + "C": "The bathtub is three steps away from you, so you can reach it with only a short walk.", + "D": "The bathtub is eight steps away from you, meaning you’d need to walk farther to get there.", + "E": "The bathtub is five steps away from you." + }, + "answer_label": "E", + "answer_text": "The bathtub is five steps away from you.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_7", + "correct_answer": "The bathtub is five steps away from you.", + "original_options": { + "A": "The bathtub is four steps away.", + "B": "The bathtub is six steps away.", + "C": "The bathtub is three steps away from you, so you can reach it with only a short walk.", + "D": "The bathtub is eight steps away from you, meaning you’d need to walk farther to get there.", + "E": "The bathtub is five steps away from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_8", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How far is the bathtub from me now?", + "answer": "The bathtub is two steps away from you.", + "question_chinese": "现在浴缸离我有几步远？", + "answer_chinese": "浴缸离你有两步的距离。", + "evidence": [ + { + "description": "The bathtub is two steps away from you.", + "timestamps": { + "start": "00:12:07", + "end": "00:12:11" + } + } + ], + "question_timestamps": "00:18:55", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One step away from you.", + "B": "Three steps away from you.", + "C": "The bathtub is four steps away from you right now, so you would need to take four steps to reach it.", + "D": "The bathtub is two steps away from you.", + "E": "The bathtub is five steps away from you at the moment, requiring you to walk five steps to get there." + }, + "answer_label": "D", + "answer_text": "The bathtub is two steps away from you.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_8", + "correct_answer": "The bathtub is two steps away from you.", + "original_options": { + "A": "One step away from you.", + "B": "Three steps away from you.", + "C": "The bathtub is four steps away from you right now, so you would need to take four steps to reach it.", + "D": "The bathtub is two steps away from you.", + "E": "The bathtub is five steps away from you at the moment, requiring you to walk five steps to get there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 20 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_3_T1", + "qid_num": 25, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the yellow bucket relative to me?", + "answer": "The yellow bucket is to your left.", + "question_chinese": "黄色的桶在我什么位置？", + "answer_chinese": "黄色的桶在你左边。", + "evidence": [ + { + "description": "The yellow bucket is on your left.", + "timestamps": { + "start": "00:00:57", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In front of you.", + "B": "The yellow bucket is to your left.", + "C": "To your right.", + "D": "The yellow bucket is behind you, positioned slightly back from where you are standing now.", + "E": "The yellow bucket is above you, sitting higher than your eye level in your current viewpoint." + }, + "answer_label": "B", + "answer_text": "The yellow bucket is to your left.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_3_T1", + "correct_answer": "The yellow bucket is to your left.", + "original_options": { + "A": "In front of you.", + "B": "The yellow bucket is to your left.", + "C": "To your right.", + "D": "The yellow bucket is behind you, positioned slightly back from where you are standing now.", + "E": "The yellow bucket is above you, sitting higher than your eye level in your current viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 16 + ] + }, + { + "video_uid": "0031d268-818c-4ec4-a804-935be610a61a", + "video_length": "00:33:07", + "qid": "0031d268-818c-4ec4-a804-935be610a61a_3_T2", + "qid_num": 26, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the yellow bucket relative to me?", + "answer": "The yellow bucket is in front of you.", + "question_chinese": "黄色的桶在我什么位置？", + "answer_chinese": "黄色的桶在你前面", + "evidence": [ + { + "description": "The yellow bucket is on front of you.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The yellow bucket is to your right, on the right-hand side of your current position.", + "D": "The yellow bucket is above you, located higher than you in your current line of sight.", + "E": "The yellow bucket is in front of you." + }, + "answer_label": "E", + "answer_text": "The yellow bucket is in front of you.", + "q_id": "0031d268-818c-4ec4-a804-935be610a61a_3_T2", + "correct_answer": "The yellow bucket is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The yellow bucket is to your right, on the right-hand side of your current position.", + "D": "The yellow bucket is above you, located higher than you in your current line of sight.", + "E": "The yellow bucket is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 16 + ] + } +] diff --git a/QAs/00c450a5-49d0-4647-a45c-bf214cea169d-mcq.json b/QAs/00c450a5-49d0-4647-a45c-bf214cea169d-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d5a96691578d22445829694a0139554de1c9601c --- /dev/null +++ b/QAs/00c450a5-49d0-4647-a45c-bf214cea169d-mcq.json @@ -0,0 +1,757 @@ +[ + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_9", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand in front of the cutting board, which is closer to me, the scissors or the water glass?", + "answer": "The scissors.", + "question_chinese": "当我站在切菜板前时，哪个离我更近，是剪刀还是水杯？", + "answer_chinese": "剪刀。", + "evidence": [ + { + "description": "The scissors were closer to you, at about 10 cm to your left.", + "timestamps": { + "start": "00:09:08", + "end": "00:09:10" + } + }, + { + "description": "The water cup is in front of you on the right, about 40cm away from you.", + "timestamps": { + "start": "00:08:44", + "end": "00:08:47" + } + } + ], + "question_timestamps": "00:09:12", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The water glass.", + "B": "The scissors." + }, + "answer_label": "B", + "answer_text": "The scissors.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_9", + "correct_answer": "The scissors.", + "original_options": { + "A": "The water glass.", + "B": "The scissors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_10", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand in front of the cutting board, which is closer to me, the scissors or the water glass?", + "answer": "The glass.", + "question_chinese": "当我站在切菜板前时，哪个离我更近，是剪刀还是水杯？", + "answer_chinese": "水杯。", + "evidence": [ + { + "description": "The water cup is in front of you on the right, about 40cm away from you.", + "timestamps": { + "start": "00:08:44", + "end": "00:08:47" + } + }, + { + "description": "The scissors were placed on the workbench behind you, approximately 1.5 meters away from you.", + "timestamps": { + "start": "00:15:18", + "end": "00:15:23" + } + } + ], + "question_timestamps": "00:15:25", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The glass.", + "B": "The scissors." + }, + "answer_label": "A", + "answer_text": "The glass.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_10", + "correct_answer": "The glass.", + "original_options": { + "A": "The glass.", + "B": "The scissors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_14", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I wash the dishes just now?", + "answer": "You just washed the dishes in the sink on your right.", + "question_chinese": "我刚刚在哪里洗的碗？", + "answer_chinese": "你刚刚在你右侧的水槽洗的碗。", + "evidence": [ + { + "description": "You washed the dishes and spoons in the sink on your right.", + "timestamps": { + "start": "00:06:59", + "end": "00:07:08" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the sink on your left.", + "B": "In the sink in front of you.", + "C": "You just washed the dishes in the sink behind you, turning around to use that basin.", + "D": "You just washed the dishes in the sink on your right.", + "E": "You just washed the dishes in the sink in front of you, using that basin instead." + }, + "answer_label": "D", + "answer_text": "You just washed the dishes in the sink on your right.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_14", + "correct_answer": "You just washed the dishes in the sink on your right.", + "original_options": { + "A": "In the sink on your left.", + "B": "In the sink in front of you.", + "C": "You just washed the dishes in the sink behind you, turning around to use that basin.", + "D": "You just washed the dishes in the sink on your right.", + "E": "You just washed the dishes in the sink in front of you, using that basin instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 16 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_13", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I wash the dishes just now?", + "answer": "You just washed the dishes in the sink on your right.", + "question_chinese": "我刚刚在哪里洗的碗？", + "answer_chinese": "你刚刚在你右侧的水槽洗的碗。", + "evidence": [ + { + "description": "You washed the dishes and spoons in the sink on your right.", + "timestamps": { + "start": "00:06:59", + "end": "00:07:08" + } + } + ], + "question_timestamps": "00:25:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the sink behind you.", + "B": "In the sink in front of you.", + "C": "You just washed the dishes in the sink on your right.", + "D": "You just washed the dishes in the bowl to your left, not using the sink at all.", + "E": "You just washed the dishes in the pot on your left, using it as a makeshift basin." + }, + "answer_label": "C", + "answer_text": "You just washed the dishes in the sink on your right.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_13", + "correct_answer": "You just washed the dishes in the sink on your right.", + "original_options": { + "A": "In the sink behind you.", + "B": "In the sink in front of you.", + "C": "You just washed the dishes in the sink on your right.", + "D": "You just washed the dishes in the bowl to your left, not using the sink at all.", + "E": "You just washed the dishes in the pot on your left, using it as a makeshift basin." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 17, + 17 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_4", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route should I take to go to the refrigerator and get the ingredients?", + "answer": "Turn right and walk three steps straight. Turn left and walk 2 steps. Turn left again and walk 2 steps to reach the refrigerator.", + "question_chinese": "我应该走哪条路线去冰箱拿食材？", + "answer_chinese": "向右转并直行3步。向左转并走2步。再次向左转并走2步即可到达冰箱。", + "evidence": [ + { + "description": "From your current position, turn right. Then, walk straight for three steps.", + "timestamps": { + "start": "00:11:34", + "end": "00:11:36" + } + }, + { + "description": "After that, turn left and take two steps.", + "timestamps": { + "start": "00:11:36", + "end": "00:11:37" + } + }, + { + "description": "Finally, turn left again and take two more steps to reach the refrigerator.", + "timestamps": { + "start": "00:11:46", + "end": "00:11:48" + } + } + ], + "question_timestamps": "00:11:48", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, then walk forward.", + "B": "Turn right and walk two steps.", + "C": "Turn right three steps, then left 2 steps, then turn right and walk 2 steps.", + "D": "Turn right three steps straight, then left 1 step, then left again and walk 3 steps.", + "E": "Turn right and walk three steps straight. Turn left and walk 2 steps. Turn left again and walk 2 steps to reach the refrigerator." + }, + "answer_label": "E", + "answer_text": "Turn right and walk three steps straight. Turn left and walk 2 steps. Turn left again and walk 2 steps to reach the refrigerator.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_4", + "correct_answer": "Turn right and walk three steps straight. Turn left and walk 2 steps. Turn left again and walk 2 steps to reach the refrigerator.", + "original_options": { + "A": "Turn left, then walk forward.", + "B": "Turn right and walk two steps.", + "C": "Turn right three steps, then left 2 steps, then turn right and walk 2 steps.", + "D": "Turn right three steps straight, then left 1 step, then left again and walk 3 steps.", + "E": "Turn right and walk three steps straight. Turn left and walk 2 steps. Turn left again and walk 2 steps to reach the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_3", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route should I take to go to the refrigerator and get the ingredients?", + "answer": "Turn right and walk three steps straight. Turn left and walk 2 steps. Turn left again and walk 2 steps to reach the refrigerator.", + "question_chinese": "我应该走哪条路线去冰箱拿食材？", + "answer_chinese": "向右转并直行3步。向左转并走2步。再次向左转并走2步即可到达冰箱。", + "evidence": [ + { + "description": "From your current position, turn right. Then, walk straight for three steps.", + "timestamps": { + "start": "00:11:34", + "end": "00:11:36" + } + }, + { + "description": "After that, turn left and take two steps.", + "timestamps": { + "start": "00:11:36", + "end": "00:11:37" + } + }, + { + "description": "Finally, turn left again and take two more steps to reach the refrigerator.", + "timestamps": { + "start": "00:11:46", + "end": "00:11:48" + } + } + ], + "question_timestamps": "00:23:08", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right, then left, then right.", + "B": "Turn right and walk three steps straight. Turn left and walk 2 steps. Turn left again and walk 2 steps to reach the refrigerator.", + "C": "Turn left and walk three steps.", + "D": "Turn right and walk three steps straight, then turn right and walk two steps to the refrigerator.", + "E": "Turn right and walk three steps, then turn left for two steps and go straight one more step to the fridge." + }, + "answer_label": "B", + "answer_text": "Turn right and walk three steps straight. Turn left and walk 2 steps. Turn left again and walk 2 steps to reach the refrigerator.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_3", + "correct_answer": "Turn right and walk three steps straight. Turn left and walk 2 steps. Turn left again and walk 2 steps to reach the refrigerator.", + "original_options": { + "A": "Turn right, then left, then right.", + "B": "Turn right and walk three steps straight. Turn left and walk 2 steps. Turn left again and walk 2 steps to reach the refrigerator.", + "C": "Turn left and walk three steps.", + "D": "Turn right and walk three steps straight, then turn right and walk two steps to the refrigerator.", + "E": "Turn right and walk three steps, then turn left for two steps and go straight one more step to the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 21 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I holding in my hand now?", + "answer": "You are holding a bowl and a pan in your hand now.", + "question_chinese": "我手里现在拿着什么?", + "answer_chinese": "你手里现在拿着一个碗和一个平底锅。", + "evidence": [ + { + "description": "You are holding a bowl in your left hand and a frying pan in your right hand.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are holding a bowl only.", + "B": "You are holding a pan only.", + "C": "You are holding a plate and a pan in your hand now, not a bowl.", + "D": "You are holding a bowl and a spatula in your hand now instead of a pan.", + "E": "You are holding a bowl and a pan in your hand now." + }, + "answer_label": "E", + "answer_text": "You are holding a bowl and a pan in your hand now.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_11", + "correct_answer": "You are holding a bowl and a pan in your hand now.", + "original_options": { + "A": "You are holding a bowl only.", + "B": "You are holding a pan only.", + "C": "You are holding a plate and a pan in your hand now, not a bowl.", + "D": "You are holding a bowl and a spatula in your hand now instead of a pan.", + "E": "You are holding a bowl and a pan in your hand now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_12", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I holding in my hand now?", + "answer": "You are holding a bowl and a spatula in your hand.", + "question_chinese": "我手里现在拿着什么?", + "answer_chinese": "你手里拿着一个碗和一个铲子。", + "evidence": [ + { + "description": "You are holding a bowl in your left hand and a spatula in your right hand.", + "timestamps": { + "start": "00:11:31", + "end": "00:11:32" + } + } + ], + "question_timestamps": "00:11:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are holding a bowl and a spatula in your hand.", + "B": "You are holding a bowl only.", + "C": "You are holding a spatula only.", + "D": "You are holding a bowl and a whisk in your hand, not a spatula.", + "E": "You are holding a plate and a spatula in your hand, not a bowl." + }, + "answer_label": "A", + "answer_text": "You are holding a bowl and a spatula in your hand.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_12", + "correct_answer": "You are holding a bowl and a spatula in your hand.", + "original_options": { + "A": "You are holding a bowl and a spatula in your hand.", + "B": "You are holding a bowl only.", + "C": "You are holding a spatula only.", + "D": "You are holding a bowl and a whisk in your hand, not a spatula.", + "E": "You are holding a plate and a spatula in your hand, not a bowl." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_15", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I end up in this position after picking up the shovel?", + "answer": "After picking up the spatula from the sink, you turn around and walk straight to the stove.", + "question_chinese": "我拿起铲子后是怎么走到现在这个位置的?", + "answer_chinese": "你在水槽拿起铲子后,转身直行来到灶台面前。", + "evidence": [ + { + "description": "You pick up the shovel from the sink.", + "timestamps": { + "start": "00:10:23", + "end": "00:10:25" + } + }, + { + "description": "You turn around and walk straight to the stove.", + "timestamps": { + "start": "00:10:24", + "end": "00:10:28" + } + } + ], + "question_timestamps": "00:10:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Pick up the shovel, go sink.", + "B": "After picking up the spatula from the sink, you turn around and walk straight to the stove.", + "C": "Pick up spatula, go fridge.", + "D": "After picking up the shovel from the counter, you turn around and walk straight to the stove.", + "E": "After picking up the spatula from the sink, you turn around and walk straight to the fridge." + }, + "answer_label": "B", + "answer_text": "After picking up the spatula from the sink, you turn around and walk straight to the stove.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_15", + "correct_answer": "After picking up the spatula from the sink, you turn around and walk straight to the stove.", + "original_options": { + "A": "Pick up the shovel, go sink.", + "B": "After picking up the spatula from the sink, you turn around and walk straight to the stove.", + "C": "Pick up spatula, go fridge.", + "D": "After picking up the shovel from the counter, you turn around and walk straight to the stove.", + "E": "After picking up the spatula from the sink, you turn around and walk straight to the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 17 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_16", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I end up in this position after picking up the shovel?", + "answer": "Take the shovel, turn right and take two steps, then turn left and walk straight until you reach the stove.", + "question_chinese": "我拿起铲子后是怎么走到现在这个位置的?", + "answer_chinese": "你拿着铲子右转走两步，再左转直行到灶台面前。", + "evidence": [ + { + "description": "You turn right from the table, take two steps, then turn left and walk straight to the stove.", + "timestamps": { + "start": "00:11:34", + "end": "00:11:39" + } + } + ], + "question_timestamps": "00:21:39", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Take the shovel, turn right and take two steps, then turn left and walk straight until you reach the stove.", + "B": "Turn left and go to the stove.", + "C": "Turn right and go to the sink.", + "D": "Take the shovel, turn right and take three steps, then turn left and walk straight until you reach the stove.", + "E": "Take the shovel, turn right and take two steps, then turn right and walk straight until you reach the stove." + }, + "answer_label": "A", + "answer_text": "Take the shovel, turn right and take two steps, then turn left and walk straight until you reach the stove.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_16", + "correct_answer": "Take the shovel, turn right and take two steps, then turn left and walk straight until you reach the stove.", + "original_options": { + "A": "Take the shovel, turn right and take two steps, then turn left and walk straight until you reach the stove.", + "B": "Turn left and go to the stove.", + "C": "Turn right and go to the sink.", + "D": "Take the shovel, turn right and take three steps, then turn left and walk straight until you reach the stove.", + "E": "Take the shovel, turn right and take two steps, then turn right and walk straight until you reach the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 20 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Was I facing the window when I checked my phone?", + "answer": "Yes.", + "question_chinese": "我查看手机的时候是面向窗户的吗？", + "answer_chinese": "是的。", + "evidence": [ + { + "description": "The phone was on the windowsill. When I looked at my phone, I was facing the window.", + "timestamps": { + "start": "00:02:22", + "end": "00:02:31" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Was I facing the window when I checked my phone?", + "answer": "Yes.", + "question_chinese": "我查看手机的时候是面向窗户的吗？", + "answer_chinese": "是的。", + "evidence": [ + { + "description": "The phone was on the windowsill. When I looked at my phone, I was facing the window.", + "timestamps": { + "start": "00:16:18", + "end": "00:16:19" + } + } + ], + "question_timestamps": "00:17:21", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_2", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_5", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "There are 0 foil boxes on the marble countertop now, how many were there before putting the bread?", + "answer": "Two.", + "question_chinese": "现在大理石台面上有0个锡箔盒，那么放面包之前有多少个？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "There are two foil boxes on the marble countertop.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:09" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four foil boxes were on the marble countertop before you put the bread down.", + "E": "Five foil boxes were on the marble countertop before you placed the bread there." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_5", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four foil boxes were on the marble countertop before you put the bread down.", + "E": "Five foil boxes were on the marble countertop before you placed the bread there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "00c450a5-49d0-4647-a45c-bf214cea169d", + "video_length": "00:30:00", + "qid": "00c450a5-49d0-4647-a45c-bf214cea169d_6", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "There are 0 foil boxes on the marble countertop now, how many were there before putting the bread?", + "answer": "Two.", + "question_chinese": "现在大理石台面上有0个锡箔盒，那么放面包之前有多少个？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "There are two foil boxes on the marble countertop.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:09" + } + } + ], + "question_timestamps": "00:24:55", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "One foil box was on the marble countertop before you put the bread down.", + "E": "Three foil boxes were on the marble countertop before you placed the bread there." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "00c450a5-49d0-4647-a45c-bf214cea169d_6", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "One foil box was on the marble countertop before you put the bread down.", + "E": "Three foil boxes were on the marble countertop before you placed the bread there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + } +] diff --git a/QAs/01020-mcq.json b/QAs/01020-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..23f65058c46183cb299c5f4b629bc14bc4573c86 --- /dev/null +++ b/QAs/01020-mcq.json @@ -0,0 +1,588 @@ +[ + { + "video_uid": "01020", + "video_length": "00:01:10", + "qid": "01020_1_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What trajectory has the tricycle taken me through from its start until now?", + "answer": "After starting, the tricycle moved forward to the right, passed a small house, then entered a small road and drove straight.", + "question_chinese": "三轮车从启动到现在载着我经过了什么轨迹？", + "answer_chinese": "三轮车启动后，向右前方前行，经过一个小屋，然后进入了一条小路并直行。", + "evidence": [ + { + "description": "A man in a black jacket got on the tricycle, then the tricycle started.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:47" + } + }, + { + "description": "The tricycle moved forward to the right, passed a small house, then entered a small road.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:59" + } + }, + { + "description": "The tricycle drove straight on the small road.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went left at the start.", + "B": "You passed a fence first.", + "C": "After starting, the tricycle moved forward to the right, passed a small house, then entered a small road and drove straight.", + "D": "After starting, the tricycle moved forward to the right, passed a small fence, then drove straight.", + "E": "After starting, the tricycle moved forward to the right, passed a small house, then turned left onto the road." + }, + "answer_label": "C", + "answer_text": "After starting, the tricycle moved forward to the right, passed a small house, then entered a small road and drove straight.", + "q_id": "01020_1_T1", + "correct_answer": "After starting, the tricycle moved forward to the right, passed a small house, then entered a small road and drove straight.", + "original_options": { + "A": "You went left at the start.", + "B": "You passed a fence first.", + "C": "After starting, the tricycle moved forward to the right, passed a small house, then entered a small road and drove straight.", + "D": "After starting, the tricycle moved forward to the right, passed a small fence, then drove straight.", + "E": "After starting, the tricycle moved forward to the right, passed a small house, then turned left onto the road." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 19 + ] + }, + { + "video_uid": "01020", + "video_length": "00:01:10", + "qid": "01020_1_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What trajectory has the tricycle taken me through from its start until now?", + "answer": "After starting, the tricycle moved forward to the right, passed a small house, then entered a small road and drove straight.", + "question_chinese": "三轮车从启动到现在载着我经过了什么轨迹？", + "answer_chinese": "三轮车启动后，向右前方前行，经过一个小屋，然后进入了一条小路并直行。", + "evidence": [ + { + "description": "A man in a black jacket got on the tricycle, then the tricycle started.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:47" + } + }, + { + "description": "The tricycle moved forward to the right, passed a small house, then entered a small road.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:59" + } + }, + { + "description": "The tricycle drove straight on the small road.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went left at the start.", + "B": "You passed a fence first.", + "C": "After starting, the tricycle moved forward to the right, passed a small house, then entered a small road and drove straight.", + "D": "After starting, the tricycle moved forward to the right, passed a small fence, then drove straight.", + "E": "After starting, the tricycle moved forward to the right, passed a small house, then turned left onto the road." + }, + "answer_label": "C", + "answer_text": "After starting, the tricycle moved forward to the right, passed a small house, then entered a small road and drove straight.", + "q_id": "01020_1_T2", + "correct_answer": "After starting, the tricycle moved forward to the right, passed a small house, then entered a small road and drove straight.", + "original_options": { + "A": "You went left at the start.", + "B": "You passed a fence first.", + "C": "After starting, the tricycle moved forward to the right, passed a small house, then entered a small road and drove straight.", + "D": "After starting, the tricycle moved forward to the right, passed a small fence, then drove straight.", + "E": "After starting, the tricycle moved forward to the right, passed a small house, then turned left onto the road." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 19 + ] + }, + { + "video_uid": "01020", + "video_length": "00:01:10", + "qid": "01020_2_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have entered the back compartment of the tricycle now?", + "answer": "Two people.", + "question_chinese": "现在三轮车后厢里面进来了几个人？", + "answer_chinese": "两个人。", + "evidence": [ + { + "description": "You entered the back compartment of the tricycle.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:28" + } + }, + { + "description": "A man in a black jacket entered the back compartment of the tricycle.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:01:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One person.", + "B": "Three people.", + "C": "Four people have entered the back compartment of the tricycle so far.", + "D": "Five people have entered the back compartment of the tricycle up to now.", + "E": "Two people." + }, + "answer_label": "E", + "answer_text": "Two people.", + "q_id": "01020_2_T1", + "correct_answer": "Two people.", + "original_options": { + "A": "One person.", + "B": "Three people.", + "C": "Four people have entered the back compartment of the tricycle so far.", + "D": "Five people have entered the back compartment of the tricycle up to now.", + "E": "Two people." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "01020", + "video_length": "00:01:10", + "qid": "01020_2_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have entered the back compartment of the tricycle now?", + "answer": "Two people.", + "question_chinese": "现在三轮车后厢里面进来了几个人？", + "answer_chinese": "两个人。", + "evidence": [ + { + "description": "You entered the back compartment of the tricycle.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:28" + } + }, + { + "description": "A man in a black jacket entered the back compartment of the tricycle.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One person.", + "B": "Three people.", + "C": "Four people have entered the back compartment of the tricycle so far.", + "D": "Five people have entered the back compartment of the tricycle up to now.", + "E": "Two people." + }, + "answer_label": "E", + "answer_text": "Two people.", + "q_id": "01020_2_T2", + "correct_answer": "Two people.", + "original_options": { + "A": "One person.", + "B": "Three people.", + "C": "Four people have entered the back compartment of the tricycle so far.", + "D": "Five people have entered the back compartment of the tricycle up to now.", + "E": "Two people." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "01020", + "video_length": "00:01:10", + "qid": "01020_5_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Now, where is the man in the black jacket relative to the tricycle?", + "answer": "The man in the black jacket is to the left of the tricycle.", + "question_chinese": "现在，那个穿黑色外套的男性在三轮车的哪个方位？", + "answer_chinese": "穿黑色外套的男性在三轮车的左边。", + "evidence": [ + { + "description": "The man in the black jacket is to the left of the tricycle.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "He is to the right.", + "B": "The man in the black jacket is to the left of the tricycle.", + "C": "He is behind it.", + "D": "The man in the black jacket is in front of the tricycle, facing toward it.", + "E": "The man in the black jacket is behind the tricycle, a little offset to the right side." + }, + "answer_label": "B", + "answer_text": "The man in the black jacket is to the left of the tricycle.", + "q_id": "01020_5_T1", + "correct_answer": "The man in the black jacket is to the left of the tricycle.", + "original_options": { + "A": "He is to the right.", + "B": "The man in the black jacket is to the left of the tricycle.", + "C": "He is behind it.", + "D": "The man in the black jacket is in front of the tricycle, facing toward it.", + "E": "The man in the black jacket is behind the tricycle, a little offset to the right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 17 + ] + }, + { + "video_uid": "01020", + "video_length": "00:01:10", + "qid": "01020_5_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Now, where is the man in the black jacket relative to the tricycle?", + "answer": "The man in the black jacket is to the left of the tricycle.", + "question_chinese": "现在，那个穿黑色外套的男性在三轮车的哪个方位？", + "answer_chinese": "穿黑色外套的男性在三轮车的左边。", + "evidence": [ + { + "description": "The man in the black jacket is to the left of the tricycle.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "He is to the right.", + "B": "He is in front.", + "C": "The man in the black jacket is behind the tricycle, slightly farther back from it.", + "D": "The man in the black jacket is to the right of the tricycle, not directly beside the seat.", + "E": "The man in the black jacket is to the left of the tricycle." + }, + "answer_label": "E", + "answer_text": "The man in the black jacket is to the left of the tricycle.", + "q_id": "01020_5_T2", + "correct_answer": "The man in the black jacket is to the left of the tricycle.", + "original_options": { + "A": "He is to the right.", + "B": "He is in front.", + "C": "The man in the black jacket is behind the tricycle, slightly farther back from it.", + "D": "The man in the black jacket is to the right of the tricycle, not directly beside the seat.", + "E": "The man in the black jacket is to the left of the tricycle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 18 + ] + }, + { + "video_uid": "01020", + "video_length": "00:01:10", + "qid": "01020_4_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the white car now relative to me?", + "answer": "The white car is behind you.", + "question_chinese": "现在那辆白色小车在我的哪个方位？", + "answer_chinese": "白色小车在你的身后。", + "evidence": [ + { + "description": "The white car is behind you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The white car is in front of you.", + "B": "The white car is to your left.", + "C": "The white car is behind you.", + "D": "The white car is to your right, roughly alongside you as you move.", + "E": "The white car is beside you, staying level with your position rather than behind you." + }, + "answer_label": "C", + "answer_text": "The white car is behind you.", + "q_id": "01020_4_T1", + "correct_answer": "The white car is behind you.", + "original_options": { + "A": "The white car is in front of you.", + "B": "The white car is to your left.", + "C": "The white car is behind you.", + "D": "The white car is to your right, roughly alongside you as you move.", + "E": "The white car is beside you, staying level with your position rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 13, + 15 + ] + }, + { + "video_uid": "01020", + "video_length": "00:01:10", + "qid": "01020_4_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the white car now relative to me?", + "answer": "The white car is behind you.", + "question_chinese": "现在那辆白色小车在我的哪个方位？", + "answer_chinese": "白色小车在你的身后。", + "evidence": [ + { + "description": "The white car is behind you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The white car is behind you.", + "B": "The white car is in front of you.", + "C": "The white car is to your left.", + "D": "The white car is off to your right side as you face forward.", + "E": "The white car is ahead of you, slightly to one side as you look forward." + }, + "answer_label": "A", + "answer_text": "The white car is behind you.", + "q_id": "01020_4_T2", + "correct_answer": "The white car is behind you.", + "original_options": { + "A": "The white car is behind you.", + "B": "The white car is in front of you.", + "C": "The white car is to your left.", + "D": "The white car is off to your right side as you face forward.", + "E": "The white car is ahead of you, slightly to one side as you look forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 13, + 15 + ] + }, + { + "video_uid": "01020", + "video_length": "00:01:10", + "qid": "01020_3_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have been seen in total besides me now?", + "answer": "Two people.", + "question_chinese": "现在除了我以外，一共看见了几个人？", + "answer_chinese": "两个人。", + "evidence": [ + { + "description": "Saw a man in a black jacket.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Saw a woman in a yellow jacket.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero people.", + "B": "One person.", + "C": "Three people in total besides you have been seen so far in the video.", + "D": "Two people.", + "E": "Five people besides you have appeared across the video up to now in total." + }, + "answer_label": "D", + "answer_text": "Two people.", + "q_id": "01020_3_T1", + "correct_answer": "Two people.", + "original_options": { + "A": "Zero people.", + "B": "One person.", + "C": "Three people in total besides you have been seen so far in the video.", + "D": "Two people.", + "E": "Five people besides you have appeared across the video up to now in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "01020", + "video_length": "00:01:10", + "qid": "01020_3_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have been seen in total besides me now?", + "answer": "Two people.", + "question_chinese": "现在除了我以外，一共看见了几个人？", + "answer_chinese": "两个人。", + "evidence": [ + { + "description": "Saw a man in a black jacket.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Saw a woman in a yellow jacket.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One person.", + "B": "Three people.", + "C": "Four people besides you have been seen in total up to now.", + "D": "Two people.", + "E": "Six people in total besides you have shown up so far in the video." + }, + "answer_label": "D", + "answer_text": "Two people.", + "q_id": "01020_3_T2", + "correct_answer": "Two people.", + "original_options": { + "A": "One person.", + "B": "Three people.", + "C": "Four people besides you have been seen in total up to now.", + "D": "Two people.", + "E": "Six people in total besides you have shown up so far in the video." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + } +] diff --git a/QAs/01039-mcq.json b/QAs/01039-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e5a9e3255008c988e8c9019294800331b841548d --- /dev/null +++ b/QAs/01039-mcq.json @@ -0,0 +1,1301 @@ +[ + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_15", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hangings are on the wall?", + "answer": "Six.", + "question_chinese": "墙上有多少个挂件？", + "answer_chinese": "6", + "evidence": [ + { + "description": "I saw 4 hangings.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "4 plus 2", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "Four hangings are visible on the wall when you look over at it.", + "D": "Six.", + "E": "Eight hangings are displayed on the wall in total, not six at all." + }, + "answer_label": "D", + "answer_text": "Six.", + "q_id": "01039_15", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "Four hangings are visible on the wall when you look over at it.", + "D": "Six.", + "E": "Eight hangings are displayed on the wall in total, not six at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_16", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hangings are on the wall?", + "answer": "six.", + "question_chinese": "墙上有多少个挂件？", + "answer_chinese": "6", + "evidence": [ + { + "description": "I saw 4 hangings.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "4 plus 2", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "six.", + "D": "Three hangings are on the wall in total when you count them.", + "E": "Eight pieces are hanging on the wall altogether, more than six in total." + }, + "answer_label": "C", + "answer_text": "six.", + "q_id": "01039_16", + "correct_answer": "six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "six.", + "D": "Three hangings are on the wall in total when you count them.", + "E": "Eight pieces are hanging on the wall altogether, more than six in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_13", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hanging ornaments are there on the ceiling in total?", + "answer": "twenty-four.", + "question_chinese": "天花板上总共有多少个挂件？", + "answer_chinese": "24", + "evidence": [ + { + "description": "I saw 18 hangings", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "18 plus 6", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Twenty.", + "B": "Thirty.", + "C": "Fourteen hanging ornaments are visible on the ceiling in total right now.", + "D": "Twelve hanging ornaments are on the ceiling altogether when you count them.", + "E": "twenty-four." + }, + "answer_label": "E", + "answer_text": "twenty-four.", + "q_id": "01039_13", + "correct_answer": "twenty-four.", + "original_options": { + "A": "Twenty.", + "B": "Thirty.", + "C": "Fourteen hanging ornaments are visible on the ceiling in total right now.", + "D": "Twelve hanging ornaments are on the ceiling altogether when you count them.", + "E": "twenty-four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_14", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hanging ornaments are there on the ceiling in total?", + "answer": "Twenty-four.", + "question_chinese": "天花板上总共有多少个挂件？", + "answer_chinese": "24", + "evidence": [ + { + "description": "I saw 18 hangingss", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "18 plus 6", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Twenty.", + "B": "Thirty.", + "C": "Fourteen hanging ornaments are visible on the ceiling in total right now.", + "D": "Twelve hanging ornaments are on the ceiling altogether when you count them.", + "E": "Twenty-four." + }, + "answer_label": "E", + "answer_text": "Twenty-four.", + "q_id": "01039_14", + "correct_answer": "Twenty-four.", + "original_options": { + "A": "Twenty.", + "B": "Thirty.", + "C": "Fourteen hanging ornaments are visible on the ceiling in total right now.", + "D": "Twelve hanging ornaments are on the ceiling altogether when you count them.", + "E": "Twenty-four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_11", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I take just now?", + "answer": "Two.", + "question_chinese": "我刚才拐了多少个弯？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:40", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:00:44", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "You took four turns just now while moving through the space you were in.", + "E": "You made five turns in total during the movement you just did." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "01039_11", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "You took four turns just now while moving through the space you were in.", + "E": "You made five turns in total during the movement you just did." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 12 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_5_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I just make?", + "answer": "You just made 1 turn.", + "question_chinese": "我刚才拐了多少个弯？", + "answer_chinese": "你刚才拐了1个弯。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just made 0 turns.", + "B": "You just made 2 turns.", + "C": "You just made 3 turns while moving through the area and shifting directions a few times.", + "D": "You just made 4 turns as you navigated around, changing direction repeatedly before stopping.", + "E": "You just made 1 turn." + }, + "answer_label": "E", + "answer_text": "You just made 1 turn.", + "q_id": "01039_5_T1", + "correct_answer": "You just made 1 turn.", + "original_options": { + "A": "You just made 0 turns.", + "B": "You just made 2 turns.", + "C": "You just made 3 turns while moving through the area and shifting directions a few times.", + "D": "You just made 4 turns as you navigated around, changing direction repeatedly before stopping.", + "E": "You just made 1 turn." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 14 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_5_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I just make?", + "answer": "2.", + "question_chinese": "我刚才拐了多少个弯？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:40", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "1.", + "B": "2.", + "C": "3.", + "D": "You made 4 turns in that short stretch while continuing forward before the next moment.", + "E": "You made 5 turns before reaching the next point, changing direction several times in sequence." + }, + "answer_label": "B", + "answer_text": "2.", + "q_id": "01039_5_T2", + "correct_answer": "2.", + "original_options": { + "A": "1.", + "B": "2.", + "C": "3.", + "D": "You made 4 turns in that short stretch while continuing forward before the next moment.", + "E": "You made 5 turns before reaching the next point, changing direction several times in sequence." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_3", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many staff members did I just see?", + "answer": "Four.", + "question_chinese": "我刚才看到了多少个工作人员？", + "answer_chinese": "4", + "evidence": [ + { + "description": "I saw four staffs", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "You just saw five staff members in total as you looked around the area briefly.", + "D": "You just saw six staff members as you looked around, noticing additional people nearby.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "01039_3", + "correct_answer": "Four.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "You just saw five staff members in total as you looked around the area briefly.", + "D": "You just saw six staff members as you looked around, noticing additional people nearby.", + "E": "Four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_4", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many staff members did I just see?", + "answer": "six.", + "question_chinese": "我刚才看到了多少个工作人员？", + "answer_chinese": "6", + "evidence": [ + { + "description": "I saw four staffs", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": " I saw the fifth staff", + "timestamps": { + "start": "00:00:024", + "end": "00:00:25" + } + }, + { + "description": "I saw the sixth staff", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "six.", + "C": "Five.", + "D": "You just saw two staff members nearby while scanning the area for assistance.", + "E": "You just saw eight staff members clustered together as you looked around the space briefly." + }, + "answer_label": "B", + "answer_text": "six.", + "q_id": "01039_4", + "correct_answer": "six.", + "original_options": { + "A": "Four.", + "B": "six.", + "C": "Five.", + "D": "You just saw two staff members nearby while scanning the area for assistance.", + "E": "You just saw eight staff members clustered together as you looked around the space briefly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many green plants did I pass just now?", + "answer": "Four.", + "question_chinese": "我刚才经过了几个绿植？", + "answer_chinese": "4", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "the third one and the fourth one", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "You just walked past five green plants while continuing along the path for a moment.", + "D": "You just walked past six green plants along the way as you moved forward steadily.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "01039_2", + "correct_answer": "Four.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "You just walked past five green plants while continuing along the path for a moment.", + "D": "You just walked past six green plants along the way as you moved forward steadily.", + "E": "Four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_1_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many green plants did I just walk past?", + "answer": "You passed 4 green plants: the 1st, 2nd, and 3rd were to your left, and the 4th was to your right.", + "question_chinese": "我刚才经过了几个绿植？", + "answer_chinese": "你经过4个，第一第二第三个绿植在你左边，第四个绿植在你右边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "the third one and the fourth one", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You passed 4 green plants: the 1st, 2nd, and 3rd were to your left, and the 4th was to your right.", + "B": "You passed 2 green plants.", + "C": "You passed 3 green plants.", + "D": "You passed 5 green plants: the 1st-4th were on your left, and the 5th was on your right.", + "E": "You passed 6 green plants: the 1st-4th were to your left, and the 5th and 6th were to your right." + }, + "answer_label": "A", + "answer_text": "You passed 4 green plants: the 1st, 2nd, and 3rd were to your left, and the 4th was to your right.", + "q_id": "01039_1_T1", + "correct_answer": "You passed 4 green plants: the 1st, 2nd, and 3rd were to your left, and the 4th was to your right.", + "original_options": { + "A": "You passed 4 green plants: the 1st, 2nd, and 3rd were to your left, and the 4th was to your right.", + "B": "You passed 2 green plants.", + "C": "You passed 3 green plants.", + "D": "You passed 5 green plants: the 1st-4th were on your left, and the 5th was on your right.", + "E": "You passed 6 green plants: the 1st-4th were to your left, and the 5th and 6th were to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 20 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_1_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many green plants did I just walk past?", + "answer": "You just passed 9 green plants.", + "question_chinese": "我刚才经过了几个绿植？", + "answer_chinese": "你刚才经过了9个绿植", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:23", + "end": "00:00:27" + } + }, + { + "description": "the third one and the fourth one", + "timestamps": { + "start": "00:00:31", + "end": "00:00:34" + } + }, + { + "description": "the fifth one", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + }, + { + "description": "the sixth one and seventh one", + "timestamps": { + "start": "00:00:08", + "end": "00:00:13" + } + }, + { + "description": "the eighth one and the ninth one", + "timestamps": { + "start": "00:00:42", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just passed 7 green plants.", + "B": "You just passed 9 green plants.", + "C": "You just passed 8 green plants.", + "D": "You just passed 10 green plants while walking forward through the area without stopping.", + "E": "You just passed 11 green plants during that segment as you continued ahead at a steady pace." + }, + "answer_label": "B", + "answer_text": "You just passed 9 green plants.", + "q_id": "01039_1_T2", + "correct_answer": "You just passed 9 green plants.", + "original_options": { + "A": "You just passed 7 green plants.", + "B": "You just passed 9 green plants.", + "C": "You just passed 8 green plants.", + "D": "You just passed 10 green plants while walking forward through the area without stopping.", + "E": "You just passed 11 green plants during that segment as you continued ahead at a steady pace." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 17 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_7", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I hand to the staff?", + "answer": "card.", + "question_chinese": "我把什么东西递给了工作人员？", + "answer_chinese": "卡片", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:01", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "a ticket.", + "B": "card.", + "C": "a brochure.", + "D": "You handed the staff a small note you were holding when they asked to see it.", + "E": "You handed the staff your wallet for them to check before letting you proceed further." + }, + "answer_label": "B", + "answer_text": "card.", + "q_id": "01039_7", + "correct_answer": "card.", + "original_options": { + "A": "a ticket.", + "B": "card.", + "C": "a brochure.", + "D": "You handed the staff a small note you were holding when they asked to see it.", + "E": "You handed the staff your wallet for them to check before letting you proceed further." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 15 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_4_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I hand to the staff member?", + "answer": "You handed the card to a staff member, and the staff member was in front of you.", + "question_chinese": "我把什么东西递给了工作人员？", + "answer_chinese": "你把卡片递给了工作人员，工作人员在你面前。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:01", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:01:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You handed the card to a staff member, and the staff member was in front of you.", + "B": "You handed over a ticket.", + "C": "You handed over a receipt.", + "D": "You handed the card to the staff member standing behind you, not the one directly in front of you.", + "E": "You handed a paper slip to the staff member in front of you, rather than giving them a card." + }, + "answer_label": "A", + "answer_text": "You handed the card to a staff member, and the staff member was in front of you.", + "q_id": "01039_4_T1", + "correct_answer": "You handed the card to a staff member, and the staff member was in front of you.", + "original_options": { + "A": "You handed the card to a staff member, and the staff member was in front of you.", + "B": "You handed over a ticket.", + "C": "You handed over a receipt.", + "D": "You handed the card to the staff member standing behind you, not the one directly in front of you.", + "E": "You handed a paper slip to the staff member in front of you, rather than giving them a card." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_4_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I hand to the staff member?", + "answer": "You handed over the card.", + "question_chinese": "我把什么东西递给了工作人员？", + "answer_chinese": "你把卡", + "evidence": [ + { + "description": "I handed the card to the staff.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:01:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You handed over a ticket.", + "B": "You handed over a receipt.", + "C": "You handed over the card.", + "D": "You handed over a small paper slip instead of a card to the staff member.", + "E": "You handed the staff member your phone, holding it out in front of you for them to take." + }, + "answer_label": "C", + "answer_text": "You handed over the card.", + "q_id": "01039_4_T2", + "correct_answer": "You handed over the card.", + "original_options": { + "A": "You handed over a ticket.", + "B": "You handed over a receipt.", + "C": "You handed over the card.", + "D": "You handed over a small paper slip instead of a card to the staff member.", + "E": "You handed the staff member your phone, holding it out in front of you for them to take." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_5", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is my right hand holding?", + "answer": "cane.", + "question_chinese": "我的右手拿着什么？", + "answer_chinese": "拐杖", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "a book.", + "B": "an apple.", + "C": "cane.", + "D": "a sandwich you’re holding in your right hand, not a cane.", + "E": "a pencil in your right hand, which you’re gripping near the tip." + }, + "answer_label": "C", + "answer_text": "cane.", + "q_id": "01039_5", + "correct_answer": "cane.", + "original_options": { + "A": "a book.", + "B": "an apple.", + "C": "cane.", + "D": "a sandwich you’re holding in your right hand, not a cane.", + "E": "a pencil in your right hand, which you’re gripping near the tip." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_3_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I holding in my right hand?", + "answer": "A cane.", + "question_chinese": "我的右手拿着什么？", + "answer_chinese": "拐杖", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A card.", + "B": "A phone.", + "C": "A cane.", + "D": "A small paper ticket that you’re holding in your right hand instead of a cane.", + "E": "A folded receipt in your right hand, which you keep pinched between your fingers." + }, + "answer_label": "C", + "answer_text": "A cane.", + "q_id": "01039_3_T2", + "correct_answer": "A cane.", + "original_options": { + "A": "A card.", + "B": "A phone.", + "C": "A cane.", + "D": "A small paper ticket that you’re holding in your right hand instead of a cane.", + "E": "A folded receipt in your right hand, which you keep pinched between your fingers." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 14 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_3_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I holding in my right hand?", + "answer": "You were holding a card in your right hand.", + "question_chinese": "我的右手拿着什么？", + "answer_chinese": "你的右手拿着卡片", + "evidence": [ + { + "description": "I‘m holding a card in my hand.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:013" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You held a cane.", + "B": "You held a ticket.", + "C": "You were holding a card in your left hand, not in your right hand.", + "D": "You were holding a folded receipt in your right hand instead of a card.", + "E": "You were holding a card in your right hand." + }, + "answer_label": "E", + "answer_text": "You were holding a card in your right hand.", + "q_id": "01039_3_T1", + "correct_answer": "You were holding a card in your right hand.", + "original_options": { + "A": "You held a cane.", + "B": "You held a ticket.", + "C": "You were holding a card in your left hand, not in your right hand.", + "D": "You were holding a folded receipt in your right hand instead of a card.", + "E": "You were holding a card in your right hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_6_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hanging decorations did I see on the ceiling in total?", + "answer": "You saw 18 hanging decorations, on the ceiling in front of you.", + "question_chinese": "我看到天花板上总共有多少个挂件？", + "answer_chinese": "你看到18个挂件，挂件在你面前的天花板上。", + "evidence": [ + { + "description": "I saw 18 hangings", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 18 hanging decorations, on the ceiling in front of you.", + "B": "You saw 16 hanging decorations.", + "C": "You saw 20 hanging decorations.", + "D": "You saw 17 hanging decorations on the ceiling in front of you in total.", + "E": "You saw 19 hanging decorations on the ceiling in front of you overall." + }, + "answer_label": "A", + "answer_text": "You saw 18 hanging decorations, on the ceiling in front of you.", + "q_id": "01039_6_T1", + "correct_answer": "You saw 18 hanging decorations, on the ceiling in front of you.", + "original_options": { + "A": "You saw 18 hanging decorations, on the ceiling in front of you.", + "B": "You saw 16 hanging decorations.", + "C": "You saw 20 hanging decorations.", + "D": "You saw 17 hanging decorations on the ceiling in front of you in total.", + "E": "You saw 19 hanging decorations on the ceiling in front of you overall." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_6_T2", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hanging decorations did I see on the ceiling in total?", + "answer": "24.", + "question_chinese": "我看到天花板上总共有多少个挂件？", + "answer_chinese": "24", + "evidence": [ + { + "description": "I saw 18 hangingss", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "18 plus 6", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "24.", + "B": "23.", + "C": "26.", + "D": "You saw 22 hanging decorations on the ceiling in total, rather than 24 overall.", + "E": "You saw 25 hanging decorations on the ceiling overall, instead of 24 in total." + }, + "answer_label": "A", + "answer_text": "24.", + "q_id": "01039_6_T2", + "correct_answer": "24.", + "original_options": { + "A": "24.", + "B": "23.", + "C": "26.", + "D": "You saw 22 hanging decorations on the ceiling in total, rather than 24 overall.", + "E": "You saw 25 hanging decorations on the ceiling overall, instead of 24 in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_7_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hanging decorations can I see on the wall?", + "answer": "You saw 4 hanging decorations on the wall in total: 1 to your left, and 3 in front of you.", + "question_chinese": "我能看见墙上有多少个挂件？", + "answer_chinese": "你看见墙上一共4个挂件。一个挂件在你左边，三个挂件在你前面。", + "evidence": [ + { + "description": "I saw 4 hangings.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 3 hanging decorations total.", + "B": "You saw 4 hanging decorations on the wall in total: 1 to your left, and 3 in front of you.", + "C": "You saw 5 hanging decorations total.", + "D": "You saw 2 hanging decorations on the wall in total: 1 to your left and 1 in front of you.", + "E": "You saw 6 hanging decorations on the wall in total: 2 to your left and 4 in front of you." + }, + "answer_label": "B", + "answer_text": "You saw 4 hanging decorations on the wall in total: 1 to your left, and 3 in front of you.", + "q_id": "01039_7_T1", + "correct_answer": "You saw 4 hanging decorations on the wall in total: 1 to your left, and 3 in front of you.", + "original_options": { + "A": "You saw 3 hanging decorations total.", + "B": "You saw 4 hanging decorations on the wall in total: 1 to your left, and 3 in front of you.", + "C": "You saw 5 hanging decorations total.", + "D": "You saw 2 hanging decorations on the wall in total: 1 to your left and 1 in front of you.", + "E": "You saw 6 hanging decorations on the wall in total: 2 to your left and 4 in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 20 + ] + }, + { + "video_uid": "01039", + "video_length": "00:01:15", + "qid": "01039_7_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hanging decorations can I see on the wall?", + "answer": "6.", + "question_chinese": "我能看见墙上有多少个挂件？", + "answer_chinese": "6", + "evidence": [ + { + "description": "I saw 4 hangings.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "4 plus 2", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "4.", + "B": "8.", + "C": "6.", + "D": "Five hanging decorations are visible on the wall, spaced across the surface.", + "E": "Seven hanging decorations are on the wall, with several clustered closer together." + }, + "answer_label": "C", + "answer_text": "6.", + "q_id": "01039_7_T2", + "correct_answer": "6.", + "original_options": { + "A": "4.", + "B": "8.", + "C": "6.", + "D": "Five hanging decorations are visible on the wall, spaced across the surface.", + "E": "Seven hanging decorations are on the wall, with several clustered closer together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + } +] diff --git a/QAs/01042-mcq.json b/QAs/01042-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b5e90b22860b123ccfaa17b61046e16b0d47051b --- /dev/null +++ b/QAs/01042-mcq.json @@ -0,0 +1,697 @@ +[ + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I just take?", + "answer": "two.", + "question_chinese": "我刚才拐了几个弯？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:03", + "end": "00:01:08" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:32", + "end": "00:01:36" + } + } + ], + "question_timestamps": "00:01:37", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One turn.", + "B": "Three turns.", + "C": "You just took four turns before reaching the next straight stretch of hallway.", + "D": "two.", + "E": "You took five turns in total while moving through the route to this point." + }, + "answer_label": "D", + "answer_text": "two.", + "q_id": "01042_7", + "correct_answer": "two.", + "original_options": { + "A": "One turn.", + "B": "Three turns.", + "C": "You just took four turns before reaching the next straight stretch of hallway.", + "D": "two.", + "E": "You took five turns in total while moving through the route to this point." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I just take?", + "answer": "Three.", + "question_chinese": "我刚才拐了几个弯？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:03", + "end": "00:01:08" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:32", + "end": "00:01:36" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:01:45", + "end": "00:01:46" + } + } + ], + "question_timestamps": "00:01:47", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two turns.", + "B": "Four turns.", + "C": "Three.", + "D": "You just took one turn before continuing straight ahead down the corridor.", + "E": "You took five turns overall while navigating the path to reach this location." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "01042_8", + "correct_answer": "Three.", + "original_options": { + "A": "Two turns.", + "B": "Four turns.", + "C": "Three.", + "D": "You just took one turn before continuing straight ahead down the corridor.", + "E": "You took five turns overall while navigating the path to reach this location." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_1_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the wall?", + "answer": "Yes—you can. The wall is to your left.", + "question_chinese": "我可以摸到墙壁吗？", + "answer_chinese": "可以，墙在你的左方。", + "evidence": [ + { + "description": "You need to take a step to the left", + "timestamps": { + "start": "00:02:17", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:02:26", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t. The wall is to your left.", + "B": "Yes—you can. The wall is to your left." + }, + "answer_label": "B", + "answer_text": "Yes—you can. The wall is to your left.", + "q_id": "01042_1_T1", + "correct_answer": "Yes—you can. The wall is to your left.", + "original_options": { + "A": "No—you can’t. The wall is to your left.", + "B": "Yes—you can. The wall is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_1_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the wall?", + "answer": "No—you can’t. The wall is to your right.", + "question_chinese": "我可以摸到墙壁吗？", + "answer_chinese": "不可以，墙在你的右边。", + "evidence": [ + { + "description": "You need to take a step to the left", + "timestamps": { + "start": "00:02:17", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t. The wall is to your right.", + "B": "Yes—you can. The wall is to your right." + }, + "answer_label": "A", + "answer_text": "No—you can’t. The wall is to your right.", + "q_id": "01042_1_T2", + "correct_answer": "No—you can’t. The wall is to your right.", + "original_options": { + "A": "No—you can’t. The wall is to your right.", + "B": "Yes—you can. The wall is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_3_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I pass along the way?", + "answer": "Two: one is to your back-left and the other is to your back-right.", + "question_chinese": "我沿途经过了几个垃圾桶？", + "answer_chinese": "2个，在你的左后方和右后方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:07", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "Two: one is to your back-left and the other is to your back-right.", + "C": "One: to your back-left.", + "D": "Three: two are to your back-left and one is to your back-right.", + "E": "Four: two are to your back-left and two are to your back-right." + }, + "answer_label": "B", + "answer_text": "Two: one is to your back-left and the other is to your back-right.", + "q_id": "01042_3_T1", + "correct_answer": "Two: one is to your back-left and the other is to your back-right.", + "original_options": { + "A": "Zero.", + "B": "Two: one is to your back-left and the other is to your back-right.", + "C": "One: to your back-left.", + "D": "Three: two are to your back-left and one is to your back-right.", + "E": "Four: two are to your back-left and two are to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 4, + 12, + 12 + ] + }, + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_3_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I pass along the way?", + "answer": "One, and it’s to your back-left.", + "question_chinese": "我沿途经过了几个垃圾桶？", + "answer_chinese": "1个，在你的左后方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "Two: back-left and back-right.", + "C": "One, and it’s to your back-left.", + "D": "Three: one is to your back-left, one is behind you, and one is to your back-right.", + "E": "Four: two are to your back-left and two are to your back-right." + }, + "answer_label": "C", + "answer_text": "One, and it’s to your back-left.", + "q_id": "01042_3_T2", + "correct_answer": "One, and it’s to your back-left.", + "original_options": { + "A": "Zero.", + "B": "Two: back-left and back-right.", + "C": "One, and it’s to your back-left.", + "D": "Three: one is to your back-left, one is behind you, and one is to your back-right.", + "E": "Four: two are to your back-left and two are to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 4, + 16, + 12 + ] + }, + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_6_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boats did I see on the river?", + "answer": "Two, both to your left.", + "question_chinese": "我看到有几条船在河上？", + "answer_chinese": "2条，都在你的左边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:51", + "end": "00:00:52" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:03", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, both to your left.", + "B": "Zero.", + "C": "One, to your left.", + "D": "Three: two are to your left and one is to your right.", + "E": "Four boats are visible, all positioned to your left along the riverbank." + }, + "answer_label": "A", + "answer_text": "Two, both to your left.", + "q_id": "01042_6_T2", + "correct_answer": "Two, both to your left.", + "original_options": { + "A": "Two, both to your left.", + "B": "Zero.", + "C": "One, to your left.", + "D": "Three: two are to your left and one is to your right.", + "E": "Four boats are visible, all positioned to your left along the riverbank." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 4, + 12, + 12 + ] + }, + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_6_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boats did I see on the river?", + "answer": "Three: they’re to your right and to your back-left.", + "question_chinese": "我看到有几条船在河上？", + "answer_chinese": "3条，在你的右边和左后方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:51", + "end": "00:00:52" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:03", + "end": "00:01:04" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:02:05", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:02:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two boats.", + "B": "Four boats.", + "C": "Two: one is to your right, and the other is to your back-left.", + "D": "Four: two are to your right, with the other two behind you toward the back-left.", + "E": "Three: they’re to your right and to your back-left." + }, + "answer_label": "E", + "answer_text": "Three: they’re to your right and to your back-left.", + "q_id": "01042_6_T1", + "correct_answer": "Three: they’re to your right and to your back-left.", + "original_options": { + "A": "Two boats.", + "B": "Four boats.", + "C": "Two: one is to your right, and the other is to your back-left.", + "D": "Four: two are to your right, with the other two behind you toward the back-left.", + "E": "Three: they’re to your right and to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_5_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people wearing white clothes did I see?", + "answer": "Two: one is in front of you, and the other is to your back-right.", + "question_chinese": "我看见了多少个白色衣服的人？", + "answer_chinese": "2个，在你的前方和右后方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One person.", + "B": "Three people.", + "C": "One: the only person in white is directly in front of you.", + "D": "Three: one is in front of you, and two others are positioned to your back-right.", + "E": "Two: one is in front of you, and the other is to your back-right." + }, + "answer_label": "E", + "answer_text": "Two: one is in front of you, and the other is to your back-right.", + "q_id": "01042_5_T2", + "correct_answer": "Two: one is in front of you, and the other is to your back-right.", + "original_options": { + "A": "One person.", + "B": "Three people.", + "C": "One: the only person in white is directly in front of you.", + "D": "Three: one is in front of you, and two others are positioned to your back-right.", + "E": "Two: one is in front of you, and the other is to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 15 + ] + }, + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_5_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people wearing white clothes did I see?", + "answer": "Three, all behind you.", + "question_chinese": "我看见了多少个白色衣服的人？", + "answer_chinese": "3个，都在你的后方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:01:10", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:01:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two people.", + "B": "Four people.", + "C": "Two, both behind you as you continue forward along the path ahead of you.", + "D": "Three, all behind you.", + "E": "Five, all behind you, clustered together rather than spread out in front of you." + }, + "answer_label": "D", + "answer_text": "Three, all behind you.", + "q_id": "01042_5_T1", + "correct_answer": "Three, all behind you.", + "original_options": { + "A": "Two people.", + "B": "Four people.", + "C": "Two, both behind you as you continue forward along the path ahead of you.", + "D": "Three, all behind you.", + "E": "Five, all behind you, clustered together rather than spread out in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_4_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people wearing yellow clothes did I see along the way?", + "answer": "Two, both to your right.", + "question_chinese": "沿途我看见了多少个黄色衣服的人？", + "answer_chinese": "两个，在你的右方。", + "evidence": [ + { + "description": "the first one and the second one", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One person.", + "B": "Three people.", + "C": "Two, both to your right.", + "D": "One, to your right as you move past them along the way.", + "E": "Zero—you didn’t see anyone wearing yellow clothes on either side of you." + }, + "answer_label": "C", + "answer_text": "Two, both to your right.", + "q_id": "01042_4_T1", + "correct_answer": "Two, both to your right.", + "original_options": { + "A": "One person.", + "B": "Three people.", + "C": "Two, both to your right.", + "D": "One, to your right as you move past them along the way.", + "E": "Zero—you didn’t see anyone wearing yellow clothes on either side of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "01042", + "video_length": "00:02:38", + "qid": "01042_4_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people wearing yellow clothes did I see along the way?", + "answer": "Three: two are to your left, and one is behind you.", + "question_chinese": "沿途我看见了多少个黄色衣服的人？", + "answer_chinese": "3，两个都在你的左方，一个在后方，", + "evidence": [ + { + "description": "the first one and the second one", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:02:05", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:02:31", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two people.", + "B": "Four people.", + "C": "Two: one is to your left, and the other is behind you.", + "D": "Three: two are to your left, and one is behind you.", + "E": "Five: three are to your left, while the remaining two are behind you." + }, + "answer_label": "D", + "answer_text": "Three: two are to your left, and one is behind you.", + "q_id": "01042_4_T2", + "correct_answer": "Three: two are to your left, and one is behind you.", + "original_options": { + "A": "Two people.", + "B": "Four people.", + "C": "Two: one is to your left, and the other is behind you.", + "D": "Three: two are to your left, and one is behind you.", + "E": "Five: three are to your left, while the remaining two are behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + } +] diff --git a/QAs/01087-mcq.json b/QAs/01087-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3e7644e45f482f8cce4b36b6d2b4bad1619ff51e --- /dev/null +++ b/QAs/01087-mcq.json @@ -0,0 +1,284 @@ +[ + { + "video_uid": "01087", + "video_length": "00:01:18", + "qid": "01087_5", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I pass by the carrot?", + "answer": "Yes.", + "question_chinese": "我经过了胡萝卜吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "I passed carrots", + "timestamps": { + "start": "00:01:08", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "01087_5", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01087", + "video_length": "00:01:18", + "qid": "01087_2_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I pass by the carrots?", + "answer": "Yes—you passed it. The carrot is to your right.", + "question_chinese": "我经过了胡萝卜吗？", + "answer_chinese": "你是，胡萝卜在你的右边。", + "evidence": [ + { + "description": "I passed carrots", + "timestamps": { + "start": "00:01:08", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:01:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you passed it. The carrot is to your right.", + "B": "No—you didn’t pass it; the carrot stays to your right." + }, + "answer_label": "A", + "answer_text": "Yes—you passed it. The carrot is to your right.", + "q_id": "01087_2_T2", + "correct_answer": "Yes—you passed it. The carrot is to your right.", + "original_options": { + "A": "Yes—you passed it. The carrot is to your right.", + "B": "No—you didn’t pass it; the carrot stays to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01087", + "video_length": "00:01:18", + "qid": "01087_6", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I pass by the carrot?", + "answer": "Yes.", + "question_chinese": "我经过了胡萝卜吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "I passed carrots", + "timestamps": { + "start": "00:01:08", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "01087_6", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01087", + "video_length": "00:01:18", + "qid": "01087_2_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I pass by the carrots?", + "answer": "No—you didn’t. You walked past the tomato, and the tomato is to your right.", + "question_chinese": "我经过了胡萝卜吗？", + "answer_chinese": "你不是，你经过了番茄，番茄在你的右边。", + "evidence": [ + { + "description": "I didn't passed carrots.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you did. You walked past the carrots, and they are to your left.", + "B": "No—you didn’t. You walked past the tomato, and the tomato is to your right." + }, + "answer_label": "B", + "answer_text": "No—you didn’t. You walked past the tomato, and the tomato is to your right.", + "q_id": "01087_2_T1", + "correct_answer": "No—you didn’t. You walked past the tomato, and the tomato is to your right.", + "original_options": { + "A": "Yes—you did. You walked past the carrots, and they are to your left.", + "B": "No—you didn’t. You walked past the tomato, and the tomato is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "01087", + "video_length": "00:01:18", + "qid": "01087_3", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the girl in black clothes?", + "answer": "My left side.", + "question_chinese": "黑衣服女生在哪里？", + "answer_chinese": "我的左边", + "evidence": [ + { + "description": "My left side", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your right.", + "B": "Behind you.", + "C": "In front of you, slightly to the right as you look ahead.", + "D": "Behind you on the right side, near the area you just passed.", + "E": "My left side." + }, + "answer_label": "E", + "answer_text": "My left side.", + "q_id": "01087_3", + "correct_answer": "My left side.", + "original_options": { + "A": "On your right.", + "B": "Behind you.", + "C": "In front of you, slightly to the right as you look ahead.", + "D": "Behind you on the right side, near the area you just passed.", + "E": "My left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01087", + "video_length": "00:01:18", + "qid": "01087_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the girl in black clothes?", + "answer": "My left side.", + "question_chinese": "黑衣服女生在哪里？", + "answer_chinese": "我的左边", + "evidence": [ + { + "description": "My left side", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your right.", + "B": "My left side.", + "C": "Behind you.", + "D": "In front of you, slightly to the right as you look ahead.", + "E": "Behind you on the right side, near the area you just passed." + }, + "answer_label": "B", + "answer_text": "My left side.", + "q_id": "01087_4", + "correct_answer": "My left side.", + "original_options": { + "A": "On your right.", + "B": "My left side.", + "C": "Behind you.", + "D": "In front of you, slightly to the right as you look ahead.", + "E": "Behind you on the right side, near the area you just passed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 12, + 12 + ] + } +] diff --git a/QAs/01106-mcq.json b/QAs/01106-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2cb564827a5792fb29ba3ec632875a3fea72ca43 --- /dev/null +++ b/QAs/01106-mcq.json @@ -0,0 +1,1214 @@ +[ + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I touch the sculpture?", + "answer": "Yes.", + "question_chinese": "我可以碰到雕塑吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "I can", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01106_1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_1_T1", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I touch the sculpture?", + "answer": "Yes, you can touch the sculpture in front of you.", + "question_chinese": "我可以碰到雕塑吗？", + "answer_chinese": "是，你可以碰到面前的雕塑", + "evidence": [ + { + "description": "I can", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can touch the sculpture in front of you.", + "B": "No, you can't touch the sculpture in front of you." + }, + "answer_label": "A", + "answer_text": "Yes, you can touch the sculpture in front of you.", + "q_id": "01106_1_T1", + "correct_answer": "Yes, you can touch the sculpture in front of you.", + "original_options": { + "A": "Yes, you can touch the sculpture in front of you.", + "B": "No, you can't touch the sculpture in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_1_T2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I touch the sculpture?", + "answer": "No, you can't touch the sculpture in front of you.", + "question_chinese": "我可以碰到雕塑吗？", + "answer_chinese": "不，你不可以碰到面前的雕塑", + "evidence": [ + { + "description": "I can't", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can touch the sculpture in front of you.", + "B": "No, you can't touch the sculpture in front of you." + }, + "answer_label": "B", + "answer_text": "No, you can't touch the sculpture in front of you.", + "q_id": "01106_1_T2", + "correct_answer": "No, you can't touch the sculpture in front of you.", + "original_options": { + "A": "Yes, you can touch the sculpture in front of you.", + "B": "No, you can't touch the sculpture in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_2_T1", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the glass display stand?", + "answer": "No, you can't touch the glass display case in front of you.", + "question_chinese": "我可以触碰到玻璃展台吗？", + "answer_chinese": "不，你不可以碰到面前的玻璃展台", + "evidence": [ + { + "description": "It is three meters in front of me.", + "timestamps": { + "start": "00:00:53", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can touch the glass display case in front of you.", + "B": "No, you can't touch the glass display case in front of you." + }, + "answer_label": "B", + "answer_text": "No, you can't touch the glass display case in front of you.", + "q_id": "01106_2_T1", + "correct_answer": "No, you can't touch the glass display case in front of you.", + "original_options": { + "A": "Yes, you can touch the glass display case in front of you.", + "B": "No, you can't touch the glass display case in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_2_T2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the glass display stand?", + "answer": "Yes, you can touch the glass display case in front of you.", + "question_chinese": "我可以触碰到玻璃展台吗？", + "answer_chinese": "可以，你可以碰到面前的玻璃展台", + "evidence": [ + { + "description": "You walked straight for three meters and could reach it.", + "timestamps": { + "start": "00:00:57", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can touch the glass display case in front of you.", + "B": "No, you can't touch the glass display case in front of you." + }, + "answer_label": "A", + "answer_text": "Yes, you can touch the glass display case in front of you.", + "q_id": "01106_2_T2", + "correct_answer": "Yes, you can touch the glass display case in front of you.", + "original_options": { + "A": "Yes, you can touch the glass display case in front of you.", + "B": "No, you can't touch the glass display case in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_3", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I walk up to the glass display case?", + "answer": "Okay,it's in front of me.", + "question_chinese": "我可以走到玻璃展台吗？", + "answer_chinese": "可以，它在我的前面", + "evidence": [ + { + "description": "it's in front of me", + "timestamps": { + "start": "00:00:48", + "end": "00:00:49" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Okay,it's in front of me.", + "B": "No, it's too far away." + }, + "answer_label": "A", + "answer_text": "Okay,it's in front of me.", + "q_id": "01106_3", + "correct_answer": "Okay,it's in front of me.", + "original_options": { + "A": "Okay,it's in front of me.", + "B": "No, it's too far away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_4", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I walk up to the glass display case?", + "answer": "Okay, it's in front of me.", + "question_chinese": "我可以走到玻璃展台吗？", + "answer_chinese": "可以，它在我的前面", + "evidence": [ + { + "description": "it's in front of me.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:49" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, it's too far away.", + "B": "Okay, it's in front of me." + }, + "answer_label": "B", + "answer_text": "Okay, it's in front of me.", + "q_id": "01106_4", + "correct_answer": "Okay, it's in front of me.", + "original_options": { + "A": "No, it's too far away.", + "B": "Okay, it's in front of me." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_3_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sculptures have I touched?", + "answer": "You touched one sculpture in front of you.", + "question_chinese": "我摸了多少个雕塑？", + "answer_chinese": "你摸了面前的一个雕塑", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You touched none.", + "B": "You touched two sculptures.", + "C": "You touched one sculpture in front of you.", + "D": "You touched zero sculptures, so you never made contact with any sculpture in front of you.", + "E": "You touched three sculptures in front of you, one after another as you moved along the display." + }, + "answer_label": "C", + "answer_text": "You touched one sculpture in front of you.", + "q_id": "01106_3_T1", + "correct_answer": "You touched one sculpture in front of you.", + "original_options": { + "A": "You touched none.", + "B": "You touched two sculptures.", + "C": "You touched one sculpture in front of you.", + "D": "You touched zero sculptures, so you never made contact with any sculpture in front of you.", + "E": "You touched three sculptures in front of you, one after another as you moved along the display." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 17 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_7", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sculptures did I touch?", + "answer": "two.", + "question_chinese": "我摸了多少个雕塑？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:38", + "end": "00:01:39" + } + } + ], + "question_timestamps": "00:01:40", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four, total, as you brushed past two more sculptures while moving down the display.", + "D": "Six altogether, counting every separate sculpture you made contact with throughout the entire exhibit.", + "E": "two." + }, + "answer_label": "E", + "answer_text": "two.", + "q_id": "01106_7", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four, total, as you brushed past two more sculptures while moving down the display.", + "D": "Six altogether, counting every separate sculpture you made contact with throughout the entire exhibit.", + "E": "two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_3_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sculptures have I touched?", + "answer": "You touched two sculptures in front of you.", + "question_chinese": "我摸了多少个雕塑？", + "answer_chinese": "你摸了面前的两个雕塑", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:38", + "end": "00:01:39" + } + } + ], + "question_timestamps": "00:01:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You touched none.", + "B": "You touched one sculpture.", + "C": "You touched three sculptures in front of you, brushing each one briefly as you passed by.", + "D": "You touched two sculptures in front of you.", + "E": "You touched four sculptures in front of you, counting each separate sculpture you made contact with." + }, + "answer_label": "D", + "answer_text": "You touched two sculptures in front of you.", + "q_id": "01106_3_T2", + "correct_answer": "You touched two sculptures in front of you.", + "original_options": { + "A": "You touched none.", + "B": "You touched one sculpture.", + "C": "You touched three sculptures in front of you, brushing each one briefly as you passed by.", + "D": "You touched two sculptures in front of you.", + "E": "You touched four sculptures in front of you, counting each separate sculpture you made contact with." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 16 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_13", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I holding in my right hand?", + "answer": "Cane.", + "question_chinese": "我的右手拿着什么？", + "answer_chinese": "拐杖", + "evidence": [ + { + "description": "Cane", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Cane.", + "B": "Umbrella.", + "C": "Golf club.", + "D": "A tennis racket, held in your right hand as you walk forward.", + "E": "A baseball bat, gripped in your right hand while you move through the area." + }, + "answer_label": "A", + "answer_text": "Cane.", + "q_id": "01106_13", + "correct_answer": "Cane.", + "original_options": { + "A": "Cane.", + "B": "Umbrella.", + "C": "Golf club.", + "D": "A tennis racket, held in your right hand as you walk forward.", + "E": "A baseball bat, gripped in your right hand while you move through the area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 14 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_14", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I holding in my right hand?", + "answer": "Cane.", + "question_chinese": "我的右手拿着什么？", + "answer_chinese": "拐杖", + "evidence": [ + { + "description": "Cane", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A bottle.", + "B": "Cane.", + "C": "A pen.", + "D": "You are holding a small rock in your right hand, gripping it loosely as you move.", + "E": "You are holding a stick in your right hand, like a short branch rather than a cane." + }, + "answer_label": "B", + "answer_text": "Cane.", + "q_id": "01106_14", + "correct_answer": "Cane.", + "original_options": { + "A": "A bottle.", + "B": "Cane.", + "C": "A pen.", + "D": "You are holding a small rock in your right hand, gripping it loosely as you move.", + "E": "You are holding a stick in your right hand, like a short branch rather than a cane." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_6_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is my left hand touching?", + "answer": "What you touched was the arm of someone next to you.", + "question_chinese": "我的左手触碰的是什么？", + "answer_chinese": "你触碰的是身边人的手臂", + "evidence": [ + { + "description": "You touched the arm of the person beside you.", + "timestamps": { + "start": "00:00:57", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sculpture in front of you.", + "B": "What you touched was the arm of someone next to you.", + "C": "The glass display case.", + "D": "What you touched was the edge of the display case right in front of you with your left hand.", + "E": "What you touched was your cane as you adjusted your grip, so your left hand contacted it briefly." + }, + "answer_label": "B", + "answer_text": "What you touched was the arm of someone next to you.", + "q_id": "01106_6_T1", + "correct_answer": "What you touched was the arm of someone next to you.", + "original_options": { + "A": "The sculpture in front of you.", + "B": "What you touched was the arm of someone next to you.", + "C": "The glass display case.", + "D": "What you touched was the edge of the display case right in front of you with your left hand.", + "E": "What you touched was your cane as you adjusted your grip, so your left hand contacted it briefly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 19, + 18 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_6_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is my left hand touching?", + "answer": "What you touched was the sculpture in front of you.", + "question_chinese": "我的左手触碰的是什么？", + "answer_chinese": "你触碰的是面前的雕塑", + "evidence": [ + { + "description": "sculpture", + "timestamps": { + "start": "00:01:40", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your left hand touched the sculpture behind you.", + "B": "Your left hand touched the sculpture below you.", + "C": "Your left hand was touching the sculpture to your right, not the one positioned straight ahead.", + "D": "Your left hand made contact with the sculpture above you rather than the one directly in front.", + "E": "What you touched was the sculpture in front of you." + }, + "answer_label": "E", + "answer_text": "What you touched was the sculpture in front of you.", + "q_id": "01106_6_T2", + "correct_answer": "What you touched was the sculpture in front of you.", + "original_options": { + "A": "Your left hand touched the sculpture behind you.", + "B": "Your left hand touched the sculpture below you.", + "C": "Your left hand was touching the sculpture to your right, not the one positioned straight ahead.", + "D": "Your left hand made contact with the sculpture above you rather than the one directly in front.", + "E": "What you touched was the sculpture in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 17 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_4_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sculptures did I see that were roped off with a red rope?", + "answer": "In front of you, you saw one sculpture roped off with a red rope.", + "question_chinese": "我看到了几个被红色绳子围起来的雕塑", + "answer_chinese": "在你面前，你看到了一个被红色绳子围起来的雕塑", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw zero roped-off sculptures.", + "B": "You saw two roped-off sculptures.", + "C": "In front of you, you saw three sculptures roped off with a red rope.", + "D": "In front of you, you saw four sculptures roped off with a red rope.", + "E": "In front of you, you saw one sculpture roped off with a red rope." + }, + "answer_label": "E", + "answer_text": "In front of you, you saw one sculpture roped off with a red rope.", + "q_id": "01106_4_T2", + "correct_answer": "In front of you, you saw one sculpture roped off with a red rope.", + "original_options": { + "A": "You saw zero roped-off sculptures.", + "B": "You saw two roped-off sculptures.", + "C": "In front of you, you saw three sculptures roped off with a red rope.", + "D": "In front of you, you saw four sculptures roped off with a red rope.", + "E": "In front of you, you saw one sculpture roped off with a red rope." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_4_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sculptures did I see that were roped off with a red rope?", + "answer": "In front of you, you saw two sculptures roped off with a red rope.", + "question_chinese": "我看到了几个被红色绳子围起来的雕塑", + "answer_chinese": "在你面前，你看到了两个被红色绳子围起来的雕塑", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:03", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:01:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw zero roped-off sculptures.", + "B": "You saw one roped-off sculpture.", + "C": "In front of you, you saw three sculptures roped off with a red rope.", + "D": "In front of you, you saw two sculptures roped off with a red rope.", + "E": "In front of you, you saw four sculptures roped off with a red rope." + }, + "answer_label": "D", + "answer_text": "In front of you, you saw two sculptures roped off with a red rope.", + "q_id": "01106_4_T1", + "correct_answer": "In front of you, you saw two sculptures roped off with a red rope.", + "original_options": { + "A": "You saw zero roped-off sculptures.", + "B": "You saw one roped-off sculpture.", + "C": "In front of you, you saw three sculptures roped off with a red rope.", + "D": "In front of you, you saw two sculptures roped off with a red rope.", + "E": "In front of you, you saw four sculptures roped off with a red rope." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_7_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see?", + "answer": "You saw four people in front of you.", + "question_chinese": "我看到了多少个人？", + "answer_chinese": "你看到了面前的四个人", + "evidence": [ + { + "description": "I saw three people", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "I saw four people", + "timestamps": { + "start": "00:00:26", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw two people in front of you.", + "B": "You saw four people in front of you.", + "C": "You saw three people in front of you.", + "D": "You saw five people in front of you, standing near the area you were facing.", + "E": "You saw six people in front of you, gathered in the direction you were looking." + }, + "answer_label": "B", + "answer_text": "You saw four people in front of you.", + "q_id": "01106_7_T1", + "correct_answer": "You saw four people in front of you.", + "original_options": { + "A": "You saw two people in front of you.", + "B": "You saw four people in front of you.", + "C": "You saw three people in front of you.", + "D": "You saw five people in front of you, standing near the area you were facing.", + "E": "You saw six people in front of you, gathered in the direction you were looking." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 15 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_15", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see?", + "answer": "eight.", + "question_chinese": "我看到了多少个人？", + "answer_chinese": "8", + "evidence": [ + { + "description": "I saw three people", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "I saw four people", + "timestamps": { + "start": "00:00:26", + "end": "00:00:27" + } + }, + { + "description": "I saw eight people", + "timestamps": { + "start": "00:02:18", + "end": "00:02:19" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Six.", + "C": "You saw three people in total, not eight, based on the group visible in the scene.", + "D": "You saw ten people in total, counting everyone visible rather than just the closest group.", + "E": "eight." + }, + "answer_label": "E", + "answer_text": "eight.", + "q_id": "01106_15", + "correct_answer": "eight.", + "original_options": { + "A": "Five.", + "B": "Six.", + "C": "You saw three people in total, not eight, based on the group visible in the scene.", + "D": "You saw ten people in total, counting everyone visible rather than just the closest group.", + "E": "eight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_7_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see?", + "answer": "You saw eight people in front of you.", + "question_chinese": "我看到了多少个人？", + "answer_chinese": "你看到了面前的八个人", + "evidence": [ + { + "description": "I saw three people", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "I saw four people", + "timestamps": { + "start": "00:00:26", + "end": "00:00:27" + } + }, + { + "description": "I saw eight people", + "timestamps": { + "start": "00:02:18", + "end": "00:02:19" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw six people in front of you.", + "B": "You saw seven people in front of you.", + "C": "You saw eight people in front of you.", + "D": "You saw nine people in front of you, spread out across the area you were facing.", + "E": "You saw ten people in front of you, standing together in the direction ahead." + }, + "answer_label": "C", + "answer_text": "You saw eight people in front of you.", + "q_id": "01106_7_T2", + "correct_answer": "You saw eight people in front of you.", + "original_options": { + "A": "You saw six people in front of you.", + "B": "You saw seven people in front of you.", + "C": "You saw eight people in front of you.", + "D": "You saw nine people in front of you, spread out across the area you were facing.", + "E": "You saw ten people in front of you, standing together in the direction ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 14 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_11", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction am I standing in relative to the girl?", + "answer": "In front of.", + "question_chinese": "我站在女生的哪个方向？", + "answer_chinese": "前面", + "evidence": [ + { + "description": "In front of", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind the girl.", + "B": "To the left of the girl.", + "C": "In front of.", + "D": "Next to the girl, standing beside her rather than directly ahead of her as you face forward.", + "E": "Above the girl, positioned higher than her instead of facing her from the front." + }, + "answer_label": "C", + "answer_text": "In front of.", + "q_id": "01106_11", + "correct_answer": "In front of.", + "original_options": { + "A": "Behind the girl.", + "B": "To the left of the girl.", + "C": "In front of.", + "D": "Next to the girl, standing beside her rather than directly ahead of her as you face forward.", + "E": "Above the girl, positioned higher than her instead of facing her from the front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 6, + 17, + 14 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_5_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction am I relative to the girl?", + "answer": "In front of you.", + "question_chinese": "我站在女生的哪个方向？", + "answer_chinese": "前面", + "evidence": [ + { + "description": "In front of", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "To your right, not directly in front of you where you are facing.", + "E": "Above you, positioned higher up rather than directly in front of you at eye level." + }, + "answer_label": "A", + "answer_text": "In front of you.", + "q_id": "01106_5_T1", + "correct_answer": "In front of you.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "To your right, not directly in front of you where you are facing.", + "E": "Above you, positioned higher up rather than directly in front of you at eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 15 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_12", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction am I standing in relative to the girl?", + "answer": "In front of.", + "question_chinese": "我站在女生的哪个方向？", + "answer_chinese": "前面", + "evidence": [ + { + "description": "In front of", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind the girl.", + "B": "To her left.", + "C": "You are standing to the girl's right, facing the same general direction as her.", + "D": "In front of.", + "E": "You are positioned beside the girl, close enough to be next to her rather than ahead." + }, + "answer_label": "D", + "answer_text": "In front of.", + "q_id": "01106_12", + "correct_answer": "In front of.", + "original_options": { + "A": "Behind the girl.", + "B": "To her left.", + "C": "You are standing to the girl's right, facing the same general direction as her.", + "D": "In front of.", + "E": "You are positioned beside the girl, close enough to be next to her rather than ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 16 + ] + }, + { + "video_uid": "01106", + "video_length": "00:02:24", + "qid": "01106_5_T2", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction am I relative to the girl?", + "answer": "To your right.", + "question_chinese": "我站在女生的哪个方向？", + "answer_chinese": "右边", + "evidence": [ + { + "description": "right", + "timestamps": { + "start": "00:01:40", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your right.", + "B": "To your left.", + "C": "Behind you.", + "D": "You are in front of the girl, positioned ahead of her rather than off to either side.", + "E": "You are behind the girl, so she is in front of you instead of you being to her right." + }, + "answer_label": "A", + "answer_text": "To your right.", + "q_id": "01106_5_T2", + "correct_answer": "To your right.", + "original_options": { + "A": "To your right.", + "B": "To your left.", + "C": "Behind you.", + "D": "You are in front of the girl, positioned ahead of her rather than off to either side.", + "E": "You are behind the girl, so she is in front of you instead of you being to her right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 17, + 19 + ] + } +] diff --git a/QAs/01169-mcq.json b/QAs/01169-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..0a2fda426f57f879e0cfd8156e38f6c7ffdd8ac3 --- /dev/null +++ b/QAs/01169-mcq.json @@ -0,0 +1,1556 @@ +[ + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_5", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the charging area?", + "answer": "Three meters to my left.", + "question_chinese": "充电区在哪？", + "answer_chinese": "我的左手边三米处", + "evidence": [ + { + "description": "Three meters to my left", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Three meters to my right.", + "B": "Three meters to my left.", + "C": "Three meters in front of me.", + "D": "It is about three meters behind me, not off to my left side.", + "E": "The charging area is roughly five meters to my left, farther away than three meters." + }, + "answer_label": "B", + "answer_text": "Three meters to my left.", + "q_id": "01169_5", + "correct_answer": "Three meters to my left.", + "original_options": { + "A": "Three meters to my right.", + "B": "Three meters to my left.", + "C": "Three meters in front of me.", + "D": "It is about three meters behind me, not off to my left side.", + "E": "The charging area is roughly five meters to my left, farther away than three meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 15 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_6", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the charging area?", + "answer": "Three meters to my left.", + "question_chinese": "充电区在哪？", + "answer_chinese": "我的左手边三米处", + "evidence": [ + { + "description": "Three meters to my left", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Three meters to my right.", + "B": "One meter in front of you.", + "C": "Three meters to my left.", + "D": "It is about two meters behind you rather than to your left side.", + "E": "The charging area is around four meters to my left, slightly farther than three meters." + }, + "answer_label": "C", + "answer_text": "Three meters to my left.", + "q_id": "01169_6", + "correct_answer": "Three meters to my left.", + "original_options": { + "A": "Three meters to my right.", + "B": "One meter in front of you.", + "C": "Three meters to my left.", + "D": "It is about two meters behind you rather than to your left side.", + "E": "The charging area is around four meters to my left, slightly farther than three meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 15 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_3_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the charging area relative to me?", + "answer": "About 10 meters to your front-left.", + "question_chinese": "充电区相对于我的方位在哪？", + "answer_chinese": "我的左前方大概10米处。", + "evidence": [ + { + "description": "Three meters to my left", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "About 10 meters to your front-left.", + "B": "About 10 meters front-right.", + "C": "About 10 meters back-left.", + "D": "It is roughly ten meters directly in front of you, not offset toward your left.", + "E": "The charging area is about three meters to your left, much closer than ten meters away." + }, + "answer_label": "A", + "answer_text": "About 10 meters to your front-left.", + "q_id": "01169_3_T1", + "correct_answer": "About 10 meters to your front-left.", + "original_options": { + "A": "About 10 meters to your front-left.", + "B": "About 10 meters front-right.", + "C": "About 10 meters back-left.", + "D": "It is roughly ten meters directly in front of you, not offset toward your left.", + "E": "The charging area is about three meters to your left, much closer than ten meters away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_3_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the charging area relative to me?", + "answer": "About 3 meters to your left.", + "question_chinese": "充电区相对于我的方位在哪？", + "answer_chinese": "我的左手边三米处", + "evidence": [ + { + "description": "Three meters to my left", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "About 3 meters to your right.", + "B": "About 3 meters in front of you.", + "C": "About 3 meters to your left.", + "D": "It is about three meters behind you rather than located on your left side.", + "E": "The charging area is around ten meters to your left, farther away than about three meters." + }, + "answer_label": "C", + "answer_text": "About 3 meters to your left.", + "q_id": "01169_3_T2", + "correct_answer": "About 3 meters to your left.", + "original_options": { + "A": "About 3 meters to your right.", + "B": "About 3 meters in front of you.", + "C": "About 3 meters to your left.", + "D": "It is about three meters behind you rather than located on your left side.", + "E": "The charging area is around ten meters to your left, farther away than about three meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 16 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_15", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the staff member holding?", + "answer": "Airplane ticket.", + "question_chinese": "工作人员拿着什么？", + "answer_chinese": "机票", + "evidence": [ + { + "description": "Airplane ticket", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A passport.", + "B": "A train schedule.", + "C": "Airplane ticket.", + "D": "The staff member is holding a luggage tag, not an airline document like a ticket.", + "E": "The staff member is holding a travel brochure, something informational rather than a boarding pass." + }, + "answer_label": "C", + "answer_text": "Airplane ticket.", + "q_id": "01169_15", + "correct_answer": "Airplane ticket.", + "original_options": { + "A": "A passport.", + "B": "A train schedule.", + "C": "Airplane ticket.", + "D": "The staff member is holding a luggage tag, not an airline document like a ticket.", + "E": "The staff member is holding a travel brochure, something informational rather than a boarding pass." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 15 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_16", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the staff member holding?", + "answer": "Airplane ticket.", + "question_chinese": "工作人员拿着什么？", + "answer_chinese": "机票", + "evidence": [ + { + "description": "Airplane ticket", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A passport.", + "B": "A luggage tag.", + "C": "The staff member is holding a travel itinerary, not a single ticket for a flight.", + "D": "The staff member is holding a travel brochure, which is promotional material rather than proof of booking.", + "E": "Airplane ticket." + }, + "answer_label": "E", + "answer_text": "Airplane ticket.", + "q_id": "01169_16", + "correct_answer": "Airplane ticket.", + "original_options": { + "A": "A passport.", + "B": "A luggage tag.", + "C": "The staff member is holding a travel itinerary, not a single ticket for a flight.", + "D": "The staff member is holding a travel brochure, which is promotional material rather than proof of booking.", + "E": "Airplane ticket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 17 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_1_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How many passersby did I run into just now?", + "answer": "You encountered three people, and they were all directly in front of you.", + "question_chinese": "我刚才遇到了几位路人？", + "answer_chinese": "你遇到了三位，他们都在你的正前方。", + "evidence": [ + { + "description": "You encountered three of them.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You encountered two people.", + "B": "You encountered four people.", + "C": "You encountered one person, directly in front of you as you moved forward.", + "D": "You encountered five people, and all of them were directly in front of you in your path.", + "E": "You encountered three people, and they were all directly in front of you." + }, + "answer_label": "E", + "answer_text": "You encountered three people, and they were all directly in front of you.", + "q_id": "01169_1_T1", + "correct_answer": "You encountered three people, and they were all directly in front of you.", + "original_options": { + "A": "You encountered two people.", + "B": "You encountered four people.", + "C": "You encountered one person, directly in front of you as you moved forward.", + "D": "You encountered five people, and all of them were directly in front of you in your path.", + "E": "You encountered three people, and they were all directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 17 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_1_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How many passersby did I run into just now?", + "answer": "You encountered six pedestrians. They were all directly in front of you, and three of them entered your view from your front-left.", + "question_chinese": "我刚才遇到了几位路人？", + "answer_chinese": "你遇到了六位路人。他们都在你的正前方，其中有三位是从你的左前方进入到你的视野。", + "evidence": [ + { + "description": "Left hand", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You encountered six pedestrians. They were all directly in front of you, and three of them entered your view from your front-left.", + "B": "You encountered five pedestrians.", + "C": "You encountered seven pedestrians.", + "D": "You encountered six pedestrians, all directly in front of you, but three entered your view from your front-right.", + "E": "You encountered eight pedestrians, all directly in front of you, and only two of them came in from your front-left." + }, + "answer_label": "A", + "answer_text": "You encountered six pedestrians. They were all directly in front of you, and three of them entered your view from your front-left.", + "q_id": "01169_1_T2", + "correct_answer": "You encountered six pedestrians. They were all directly in front of you, and three of them entered your view from your front-left.", + "original_options": { + "A": "You encountered six pedestrians. They were all directly in front of you, and three of them entered your view from your front-left.", + "B": "You encountered five pedestrians.", + "C": "You encountered seven pedestrians.", + "D": "You encountered six pedestrians, all directly in front of you, but three entered your view from your front-right.", + "E": "You encountered eight pedestrians, all directly in front of you, and only two of them came in from your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 20 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_13", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to drink water, what should I do?", + "answer": "Walk forward five meters, then turn left and walk three meters, and finally get water.", + "question_chinese": "我想喝水怎么办？", + "answer_chinese": "往前走五米，然后往左走三米，最后接水", + "evidence": [ + { + "description": "Walk forward five meters, then turn left and walk three meters, and finally get water.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk forward four meters, then left.", + "B": "Walk forward five meters, then right.", + "C": "Walk forward six meters, turn left, walk three meters, and then get water.", + "D": "Walk forward five meters, turn left, walk two meters, and finally get water.", + "E": "Walk forward five meters, then turn left and walk three meters, and finally get water." + }, + "answer_label": "E", + "answer_text": "Walk forward five meters, then turn left and walk three meters, and finally get water.", + "q_id": "01169_13", + "correct_answer": "Walk forward five meters, then turn left and walk three meters, and finally get water.", + "original_options": { + "A": "Walk forward four meters, then left.", + "B": "Walk forward five meters, then right.", + "C": "Walk forward six meters, turn left, walk three meters, and then get water.", + "D": "Walk forward five meters, turn left, walk two meters, and finally get water.", + "E": "Walk forward five meters, then turn left and walk three meters, and finally get water." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 13 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_8_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What should I do if I want to drink water?", + "answer": "There is a water dispenser to your left; walk left about 3 meters to reach it.", + "question_chinese": "我想喝水怎么办？", + "answer_chinese": "你左边有饮水机，向左走大概三米处有饮水机。", + "evidence": [ + { + "description": "About three meters to the left, there is a water dispenser.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The dispenser is to your right.", + "B": "The dispenser is behind you.", + "C": "There is a water dispenser in front of you; walk left about 3 meters to reach it.", + "D": "There is a water dispenser to your left; walk left about 3 meters to reach it.", + "E": "There is a water dispenser to your right; walk left about 3 meters to reach it." + }, + "answer_label": "D", + "answer_text": "There is a water dispenser to your left; walk left about 3 meters to reach it.", + "q_id": "01169_8_T1", + "correct_answer": "There is a water dispenser to your left; walk left about 3 meters to reach it.", + "original_options": { + "A": "The dispenser is to your right.", + "B": "The dispenser is behind you.", + "C": "There is a water dispenser in front of you; walk left about 3 meters to reach it.", + "D": "There is a water dispenser to your left; walk left about 3 meters to reach it.", + "E": "There is a water dispenser to your right; walk left about 3 meters to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 16 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_8_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What should I do if I want to drink water?", + "answer": "There is a water dispenser in front of you. Walk forward 5 meters, then go left 3 meters, and finally get some water.", + "question_chinese": "我想喝水怎么办？", + "answer_chinese": "你面前有饮水机往前走五米，然后往左走三米，最后接水", + "evidence": [ + { + "description": "Walk forward five meters, then turn left and walk three meters, and finally get water.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The dispenser is behind you.", + "B": "There is a water dispenser in front of you. Walk forward 5 meters, then go left 3 meters, and finally get some water.", + "C": "The dispenser is to your right.", + "D": "There is a water dispenser in front of you. Walk forward 4 meters, then go left 3 meters, and get water.", + "E": "There is a water dispenser in front of you. Walk forward 5 meters, then go right 3 meters, and get some water." + }, + "answer_label": "B", + "answer_text": "There is a water dispenser in front of you. Walk forward 5 meters, then go left 3 meters, and finally get some water.", + "q_id": "01169_8_T2", + "correct_answer": "There is a water dispenser in front of you. Walk forward 5 meters, then go left 3 meters, and finally get some water.", + "original_options": { + "A": "The dispenser is behind you.", + "B": "There is a water dispenser in front of you. Walk forward 5 meters, then go left 3 meters, and finally get some water.", + "C": "The dispenser is to your right.", + "D": "There is a water dispenser in front of you. Walk forward 4 meters, then go left 3 meters, and get water.", + "E": "There is a water dispenser in front of you. Walk forward 5 meters, then go right 3 meters, and get some water." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 21, + 22 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_2_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I make?", + "answer": "You turned one corner.", + "question_chinese": "我拐了多少个弯？", + "answer_chinese": "你拐了一个弯。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You turned one corner.", + "B": "You turned zero corners.", + "C": "You turned two corners.", + "D": "You turned three corners along the route before you stopped walking at the final point.", + "E": "You turned four corners during the entire route before stopping and reaching your destination." + }, + "answer_label": "A", + "answer_text": "You turned one corner.", + "q_id": "01169_2_T1", + "correct_answer": "You turned one corner.", + "original_options": { + "A": "You turned one corner.", + "B": "You turned zero corners.", + "C": "You turned two corners.", + "D": "You turned three corners along the route before you stopped walking at the final point.", + "E": "You turned four corners during the entire route before stopping and reaching your destination." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_3", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns have I made?", + "answer": "Two.", + "question_chinese": "我拐了多少个弯？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:41", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You made four turns in total during the route before you stopped moving completely.", + "D": "You made five turns overall on the route before stopping at the end of the path.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "01169_3", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You made four turns in total during the route before you stopped moving completely.", + "D": "You made five turns overall on the route before stopping at the end of the path.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_2_T2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I make?", + "answer": "You made two turns.", + "question_chinese": "我拐了多少个弯？", + "answer_chinese": "你拐了两个弯。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:41", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You made one turn.", + "B": "You made two turns.", + "C": "You made three turns.", + "D": "You made zero turns along the way before stopping at the end.", + "E": "You made four turns during the route before you finally stopped walking." + }, + "answer_label": "B", + "answer_text": "You made two turns.", + "q_id": "01169_2_T2", + "correct_answer": "You made two turns.", + "original_options": { + "A": "You made one turn.", + "B": "You made two turns.", + "C": "You made three turns.", + "D": "You made zero turns along the way before stopping at the end.", + "E": "You made four turns during the route before you finally stopped walking." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which hand am I using to hold the staff?", + "answer": "Left hand.", + "question_chinese": "我是用哪只手拉着工作人员的？", + "answer_chinese": "左手", + "evidence": [ + { + "description": "Left hand", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right hand.", + "B": "Left arm.", + "C": "You are holding the staff with your right hand, not your left.", + "D": "Left hand.", + "E": "You keep the staff gripped in your right hand while your left stays free." + }, + "answer_label": "D", + "answer_text": "Left hand.", + "q_id": "01169_1", + "correct_answer": "Left hand.", + "original_options": { + "A": "Right hand.", + "B": "Left arm.", + "C": "You are holding the staff with your right hand, not your left.", + "D": "Left hand.", + "E": "You keep the staff gripped in your right hand while your left stays free." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which hand am I using to hold the staff?", + "answer": "Left hand.", + "question_chinese": "我是用哪只手拉着工作人员的？", + "answer_chinese": "左手", + "evidence": [ + { + "description": "Left hand", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left hand.", + "B": "Right hand.", + "C": "Right arm.", + "D": "You are holding the staff with your right hand rather than your left.", + "E": "You grip the staff in your right hand while your left hand isn’t holding it." + }, + "answer_label": "A", + "answer_text": "Left hand.", + "q_id": "01169_2", + "correct_answer": "Left hand.", + "original_options": { + "A": "Left hand.", + "B": "Right hand.", + "C": "Right arm.", + "D": "You are holding the staff with your right hand rather than your left.", + "E": "You grip the staff in your right hand while your left hand isn’t holding it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 16 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_5_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see?", + "answer": "You saw four chairs.", + "question_chinese": "我看见多少张椅子？", + "answer_chinese": "你看见了四张椅子。", + "evidence": [ + { + "description": "You saw four chairs.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two chairs.", + "B": "You saw three chairs.", + "C": "You saw five chairs in total, with two grouped together and three positioned nearby.", + "D": "You saw four chairs.", + "E": "You saw six chairs in total, spread across the area rather than in one cluster." + }, + "answer_label": "D", + "answer_text": "You saw four chairs.", + "q_id": "01169_5_T1", + "correct_answer": "You saw four chairs.", + "original_options": { + "A": "You saw two chairs.", + "B": "You saw three chairs.", + "C": "You saw five chairs in total, with two grouped together and three positioned nearby.", + "D": "You saw four chairs.", + "E": "You saw six chairs in total, spread across the area rather than in one cluster." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_5_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see?", + "answer": "You saw 16 chairs.", + "question_chinese": "我看见多少张椅子？", + "answer_chinese": "你看见16张椅子。", + "evidence": [ + { + "description": "There are sixteen chairs", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 14 chairs.", + "B": "You saw 15 chairs.", + "C": "You saw 17 chairs arranged in rows, not the number stated in the answer.", + "D": "You saw 18 chairs overall, with several lined up and the rest scattered nearby.", + "E": "You saw 16 chairs." + }, + "answer_label": "E", + "answer_text": "You saw 16 chairs.", + "q_id": "01169_5_T2", + "correct_answer": "You saw 16 chairs.", + "original_options": { + "A": "You saw 14 chairs.", + "B": "You saw 15 chairs.", + "C": "You saw 17 chairs arranged in rows, not the number stated in the answer.", + "D": "You saw 18 chairs overall, with several lined up and the rest scattered nearby.", + "E": "You saw 16 chairs." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_9_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was the staff member in front of me holding?", + "answer": "The staff member in front of you is holding a rolled-up golden-yellow jacket.", + "question_chinese": "我面前的工作人员拿着什么？", + "answer_chinese": "你面前的工作人员拿着金黄色卷起来的马褂。", + "evidence": [ + { + "description": "Airplane ticket", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A folded blue jacket.", + "B": "A black backpack.", + "C": "The staff member in front of you is holding a rolled-up golden-yellow jacket.", + "D": "The staff member in front of you is holding a folded jacket, but it isn’t golden-yellow.", + "E": "The staff member in front of you is holding a small bag rather than a rolled-up jacket." + }, + "answer_label": "C", + "answer_text": "The staff member in front of you is holding a rolled-up golden-yellow jacket.", + "q_id": "01169_9_T1", + "correct_answer": "The staff member in front of you is holding a rolled-up golden-yellow jacket.", + "original_options": { + "A": "A folded blue jacket.", + "B": "A black backpack.", + "C": "The staff member in front of you is holding a rolled-up golden-yellow jacket.", + "D": "The staff member in front of you is holding a folded jacket, but it isn’t golden-yellow.", + "E": "The staff member in front of you is holding a small bag rather than a rolled-up jacket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 17 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_9_T2", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was the staff member in front of me holding?", + "answer": "The staff member in front of you is holding a plane ticket.", + "question_chinese": "我面前的工作人员拿着什么？", + "answer_chinese": "你面前的工作人员拿着机票。", + "evidence": [ + { + "description": "Airplane ticket", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A passport.", + "B": "A visa.", + "C": "The staff member in front of you is holding an ID card, not a plane ticket.", + "D": "The staff member in front of you is holding a plane ticket.", + "E": "The staff member in front of you is holding a document folder instead of a plane ticket." + }, + "answer_label": "D", + "answer_text": "The staff member in front of you is holding a plane ticket.", + "q_id": "01169_9_T2", + "correct_answer": "The staff member in front of you is holding a plane ticket.", + "original_options": { + "A": "A passport.", + "B": "A visa.", + "C": "The staff member in front of you is holding an ID card, not a plane ticket.", + "D": "The staff member in front of you is holding a plane ticket.", + "E": "The staff member in front of you is holding a document folder instead of a plane ticket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_7_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was the man in front of me holding?", + "answer": "The man in front of you is holding a suitcase.", + "question_chinese": "我面前的男人拿着什么？", + "answer_chinese": "你面前的男人拿着行李箱", + "evidence": [ + { + "description": "Suitcase", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "Your position", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A backpack.", + "B": "A duffel bag.", + "C": "The man in front of you is holding a suitcase.", + "D": "The man in front of you is holding a backpack rather than pulling a suitcase.", + "E": "The man in front of you is carrying a duffel bag, not the suitcase mentioned." + }, + "answer_label": "C", + "answer_text": "The man in front of you is holding a suitcase.", + "q_id": "01169_7_T1", + "correct_answer": "The man in front of you is holding a suitcase.", + "original_options": { + "A": "A backpack.", + "B": "A duffel bag.", + "C": "The man in front of you is holding a suitcase.", + "D": "The man in front of you is holding a backpack rather than pulling a suitcase.", + "E": "The man in front of you is carrying a duffel bag, not the suitcase mentioned." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 15 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_7_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was the man in front of me holding?", + "answer": "The man in front of you is carrying a white tote bag.", + "question_chinese": "我面前的男人拿着什么？", + "answer_chinese": "你面前的男人挎着白色手提袋。", + "evidence": [ + { + "description": "The man in front of you is carrying a white handbag.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + }, + { + "description": "Your position", + "timestamps": { + "start": "00:00:34", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The man in front of you is carrying a white tote bag.", + "B": "A black backpack.", + "C": "A rolling suitcase.", + "D": "The man in front of you is carrying a black backpack instead of a white tote bag.", + "E": "The man in front of you is holding a suitcase handle rather than carrying a white tote." + }, + "answer_label": "A", + "answer_text": "The man in front of you is carrying a white tote bag.", + "q_id": "01169_7_T2", + "correct_answer": "The man in front of you is carrying a white tote bag.", + "original_options": { + "A": "The man in front of you is carrying a white tote bag.", + "B": "A black backpack.", + "C": "A rolling suitcase.", + "D": "The man in front of you is carrying a black backpack instead of a white tote bag.", + "E": "The man in front of you is holding a suitcase handle rather than carrying a white tote." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_9", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there?", + "answer": "sixteen.", + "question_chinese": "有多少张椅子？", + "answer_chinese": "16", + "evidence": [ + { + "description": "There are sixteen chairs", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Fifteen.", + "B": "sixteen.", + "C": "Seventeen.", + "D": "There are fifteen chairs in total, one fewer than sixteen arranged around the area.", + "E": "There are seventeen chairs altogether, one additional chair compared with a count of sixteen." + }, + "answer_label": "B", + "answer_text": "sixteen.", + "q_id": "01169_9", + "correct_answer": "sixteen.", + "original_options": { + "A": "Fifteen.", + "B": "sixteen.", + "C": "Seventeen.", + "D": "There are fifteen chairs in total, one fewer than sixteen arranged around the area.", + "E": "There are seventeen chairs altogether, one additional chair compared with a count of sixteen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_10", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there?", + "answer": "sixteen.", + "question_chinese": "有多少张椅子？", + "answer_chinese": "16", + "evidence": [ + { + "description": "There are sixteen chairs", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "sixteen.", + "B": "Fourteen.", + "C": "Eighteen.", + "D": "There are fourteen chairs counted overall, which is two fewer than sixteen in the scene.", + "E": "There are eighteen chairs in total, which is two more than sixteen chairs." + }, + "answer_label": "A", + "answer_text": "sixteen.", + "q_id": "01169_10", + "correct_answer": "sixteen.", + "original_options": { + "A": "sixteen.", + "B": "Fourteen.", + "C": "Eighteen.", + "D": "There are fourteen chairs counted overall, which is two fewer than sixteen in the scene.", + "E": "There are eighteen chairs in total, which is two more than sixteen chairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 13 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_7", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the woman in the gray clothes?", + "answer": "She was first to my front left, then in front of me, and finally to my front right.", + "question_chinese": "灰衣服的女人在哪里？", + "answer_chinese": "她先在我的左前方，然后在我的前方，最后在我的右前方", + "evidence": [ + { + "description": "my front left", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + }, + { + "description": "in front of", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + }, + { + "description": "my front right.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Front right, then front, then front left.", + "B": "Back left, then behind, then back right.", + "C": "She started to your front right, moved to directly in front of you, and ended at your front left.", + "D": "She began behind you on the left, then was behind you, and finally ended behind you on the right.", + "E": "She was first to my front left, then in front of me, and finally to my front right." + }, + "answer_label": "E", + "answer_text": "She was first to my front left, then in front of me, and finally to my front right.", + "q_id": "01169_7", + "correct_answer": "She was first to my front left, then in front of me, and finally to my front right.", + "original_options": { + "A": "Front right, then front, then front left.", + "B": "Back left, then behind, then back right.", + "C": "She started to your front right, moved to directly in front of you, and ended at your front left.", + "D": "She began behind you on the left, then was behind you, and finally ended behind you on the right.", + "E": "She was first to my front left, then in front of me, and finally to my front right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_8", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the woman in the gray clothes?", + "answer": "She was first to my front left, then in front of me, and finally to my front right.", + "question_chinese": "灰衣服的女人在哪里？", + "answer_chinese": "她先在我的左前方，然后在我的前方，最后在我的右前方", + "evidence": [ + { + "description": "my front left", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + }, + { + "description": "in front of", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + }, + { + "description": "my front right.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Front left, then behind, then front right.", + "B": "Front right, then front, then front left.", + "C": "She was first to your front left, then shifted to behind you, and ended to your front right.", + "D": "She was first to my front left, then in front of me, and finally to my front right.", + "E": "She started at your front right, moved to directly in front of you, and finished at your front left." + }, + "answer_label": "D", + "answer_text": "She was first to my front left, then in front of me, and finally to my front right.", + "q_id": "01169_8", + "correct_answer": "She was first to my front left, then in front of me, and finally to my front right.", + "original_options": { + "A": "Front left, then behind, then front right.", + "B": "Front right, then front, then front left.", + "C": "She was first to your front left, then shifted to behind you, and ended to your front right.", + "D": "She was first to my front left, then in front of me, and finally to my front right.", + "E": "She started at your front right, moved to directly in front of you, and finished at your front left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 19 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_4_T1", + "qid_num": 27, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the woman in gray clothes relative to me?", + "answer": "She was initially to your front-left.", + "question_chinese": "灰衣服的女人在我的哪个方位？", + "answer_chinese": "她先在我的左前方。", + "evidence": [ + { + "description": "my front left", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "She was initially to your front-left.", + "B": "Front-right.", + "C": "Back-left.", + "D": "She was initially positioned to your front right relative to where you were facing at the start.", + "E": "At the beginning, she was behind you on the left side rather than in front of you." + }, + "answer_label": "A", + "answer_text": "She was initially to your front-left.", + "q_id": "01169_4_T1", + "correct_answer": "She was initially to your front-left.", + "original_options": { + "A": "She was initially to your front-left.", + "B": "Front-right.", + "C": "Back-left.", + "D": "She was initially positioned to your front right relative to where you were facing at the start.", + "E": "At the beginning, she was behind you on the left side rather than in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 17 + ] + }, + { + "video_uid": "01169", + "video_length": "00:00:50", + "qid": "01169_4_T2", + "qid_num": 28, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the woman in gray clothes relative to me?", + "answer": "She was first to your front-left, then in front of you, and finally to your front-right.", + "question_chinese": "灰衣服的女人在我的哪个方位？", + "answer_chinese": "她先在我的左前方，然后在我的前方，最后在我的右前方", + "evidence": [ + { + "description": "my front left", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + }, + { + "description": "in front of", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + }, + { + "description": "my front right.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Front-left, then behind, then front-right.", + "B": "She was first to your front-left, then in front of you, and finally to your front-right.", + "C": "Front-left, then left, then front-right.", + "D": "She started to your front-left, moved behind you, and then ended up at your front-right.", + "E": "She began at your front-left, shifted to your left side, and finished at your front-right." + }, + "answer_label": "B", + "answer_text": "She was first to your front-left, then in front of you, and finally to your front-right.", + "q_id": "01169_4_T2", + "correct_answer": "She was first to your front-left, then in front of you, and finally to your front-right.", + "original_options": { + "A": "Front-left, then behind, then front-right.", + "B": "She was first to your front-left, then in front of you, and finally to your front-right.", + "C": "Front-left, then left, then front-right.", + "D": "She started to your front-left, moved behind you, and then ended up at your front-right.", + "E": "She began at your front-left, shifted to your left side, and finished at your front-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + } +] diff --git a/QAs/01209-mcq.json b/QAs/01209-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..46ed28065e26a568f4b7b36b0eb2c44e1e100a7d --- /dev/null +++ b/QAs/01209-mcq.json @@ -0,0 +1,559 @@ +[ + { + "video_uid": "01209", + "video_length": "00:04:47", + "qid": "01209_11", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Has the rabbit toy been on my right hand the whole time?", + "answer": "Yes.", + "question_chinese": "兔子玩偶一直在我的右手上吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "on my right hand", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "still on my right hand", + "timestamps": { + "start": "00:04:11", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:04:13", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01209_11", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01209", + "video_length": "00:04:47", + "qid": "01209_12", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Has the rabbit toy been on my right hand the whole time?", + "answer": "Yes.", + "question_chinese": "兔子玩偶一直在我的右手上吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "on my right hand", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "still on my right hand", + "timestamps": { + "start": "00:04:11", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:04:40", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01209_12", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01209", + "video_length": "00:04:47", + "qid": "01209_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many doors did you just enter?", + "answer": "Two.", + "question_chinese": "刚才进了几个门？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:12", + "end": "00:00:20" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:24", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You only entered one door before stopping to look around for a moment.", + "E": "You went through three doors in total during that short stretch of walking." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "01209_1", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You only entered one door before stopping to look around for a moment.", + "E": "You went through three doors in total during that short stretch of walking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "01209", + "video_length": "00:04:47", + "qid": "01209_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many doors did you just enter?", + "answer": "Two.", + "question_chinese": "刚才进了几个门？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:12", + "end": "00:00:20" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:24", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You entered just one door immediately before this moment, not two doors total.", + "D": "You passed through three doors in sequence during this segment, rather than only two.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "01209_2", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You entered just one door immediately before this moment, not two doors total.", + "D": "You passed through three doors in sequence during this segment, rather than only two.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "01209", + "video_length": "00:04:47", + "qid": "01209_13", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I just pass by?", + "answer": "Four.", + "question_chinese": "我刚刚经过了几个人？", + "answer_chinese": "4", + "evidence": [ + { + "description": "two people", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "three people", + "timestamps": { + "start": "00:00:28", + "end": "00:00:30" + } + }, + { + "description": "four people", + "timestamps": { + "start": "00:00:31", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:01:33", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "You passed by three people in that section of the walk, not four people.", + "D": "Four.", + "E": "You moved past five people in total along that section of the walk." + }, + "answer_label": "D", + "answer_text": "Four.", + "q_id": "01209_13", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "You passed by three people in that section of the walk, not four people.", + "D": "Four.", + "E": "You moved past five people in total along that section of the walk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "01209", + "video_length": "00:04:47", + "qid": "01209_14", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I just pass by?", + "answer": "four.", + "question_chinese": "我刚刚经过了几个人？", + "answer_chinese": "4", + "evidence": [ + { + "description": "two people", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "three people", + "timestamps": { + "start": "00:00:28", + "end": "00:00:30" + } + }, + { + "description": "four people", + "timestamps": { + "start": "00:00:31", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:01:40", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "four.", + "C": "Five.", + "D": "You only passed by two people during that brief walk-by, not four people total.", + "E": "You passed by five people altogether as you continued forward through the area." + }, + "answer_label": "B", + "answer_text": "four.", + "q_id": "01209_14", + "correct_answer": "four.", + "original_options": { + "A": "Two.", + "B": "four.", + "C": "Five.", + "D": "You only passed by two people during that brief walk-by, not four people total.", + "E": "You passed by five people altogether as you continued forward through the area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "01209", + "video_length": "00:04:47", + "qid": "01209_15", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I just take?", + "answer": "one.", + "question_chinese": "我刚才拐了多少个弯？", + "answer_chinese": "1", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:04:09", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:04:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "You made two turns during that movement sequence before continuing straight ahead afterward.", + "D": "You took three distinct turns in that sequence of movement before stopping or changing direction.", + "E": "one." + }, + "answer_label": "E", + "answer_text": "one.", + "q_id": "01209_15", + "correct_answer": "one.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "You made two turns during that movement sequence before continuing straight ahead afterward.", + "D": "You took three distinct turns in that sequence of movement before stopping or changing direction.", + "E": "one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "01209", + "video_length": "00:04:47", + "qid": "01209_16", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I just take?", + "answer": "Two.", + "question_chinese": "我刚才拐了多少个弯？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:04:09", + "end": "00:04:12" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:04:45", + "end": "00:04:46" + } + } + ], + "question_timestamps": "00:04:47", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You took only one turn in that segment, then proceeded straight the rest of the way.", + "E": "You made three turns overall during that short route segment before continuing forward again." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "01209_16", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You took only one turn in that segment, then proceeded straight the rest of the way.", + "E": "You made three turns overall during that short route segment before continuing forward again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 14 + ] + }, + { + "video_uid": "01209", + "video_length": "00:04:47", + "qid": "01209_5", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just pass by the red service desk?", + "answer": "Yes.", + "question_chinese": "我刚才经过红色的服务台了吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "I passed the red sevice desk.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:34", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01209_5", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01209", + "video_length": "00:04:47", + "qid": "01209_6", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just pass by the red service desk?", + "answer": "Yes.", + "question_chinese": "我刚才经过红色的服务台了吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "I passed the red service desk.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "01209_6", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/01211-mcq.json b/QAs/01211-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..faf7dd64cb120765c87ce9a2e70b9bb7bc0499bb --- /dev/null +++ b/QAs/01211-mcq.json @@ -0,0 +1,478 @@ +[ + { + "video_uid": "01211", + "video_length": "00:04:49", + "qid": "01211_4_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the red car?", + "answer": "No—you can’t. It’s about 5 meters to your front-left.", + "question_chinese": "我可以碰到红车吗？", + "answer_chinese": "不可以，它在你的左前方五米处", + "evidence": [ + { + "description": "it is five meters in front of your left.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t. It’s about 5 meters to your front-left.", + "B": "Yes—you can. It’s about 5 meters to your front-left." + }, + "answer_label": "A", + "answer_text": "No—you can’t. It’s about 5 meters to your front-left.", + "q_id": "01211_4_T1", + "correct_answer": "No—you can’t. It’s about 5 meters to your front-left.", + "original_options": { + "A": "No—you can’t. It’s about 5 meters to your front-left.", + "B": "Yes—you can. It’s about 5 meters to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "01211", + "video_length": "00:04:49", + "qid": "01211_4_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the red car?", + "answer": "No—you can’t. It’s about 5 meters to your front-right.", + "question_chinese": "我可以碰到红车吗？", + "answer_chinese": "不可以，它在你的右前方五米处", + "evidence": [ + { + "description": "it is five meters in front of your left.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. It’s about 5 meters to your front-right.", + "B": "No—you can’t. It’s about 5 meters to your front-right." + }, + "answer_label": "B", + "answer_text": "No—you can’t. It’s about 5 meters to your front-right.", + "q_id": "01211_4_T2", + "correct_answer": "No—you can’t. It’s about 5 meters to your front-right.", + "original_options": { + "A": "Yes—you can. It’s about 5 meters to your front-right.", + "B": "No—you can’t. It’s about 5 meters to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "01211", + "video_length": "00:04:49", + "qid": "01211_2_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was I relative to the driver?", + "answer": "To your back-right. The driver is to your front-right.", + "question_chinese": "我在司机的哪里？", + "answer_chinese": "右后方，司机在我的右前方。", + "evidence": [ + { + "description": "drive.r", + "timestamps": { + "start": "00:00:17", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-left.", + "B": "To your front-left.", + "C": "To your front-right, with the driver positioned behind you and slightly to your left.", + "D": "To your back-right. The driver is to your front-right.", + "E": "To your back-right, but the driver is actually to your back-left rather than in front of you." + }, + "answer_label": "D", + "answer_text": "To your back-right. The driver is to your front-right.", + "q_id": "01211_2_T1", + "correct_answer": "To your back-right. The driver is to your front-right.", + "original_options": { + "A": "To your back-left.", + "B": "To your front-left.", + "C": "To your front-right, with the driver positioned behind you and slightly to your left.", + "D": "To your back-right. The driver is to your front-right.", + "E": "To your back-right, but the driver is actually to your back-left rather than in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 17 + ] + }, + { + "video_uid": "01211", + "video_length": "00:04:49", + "qid": "01211_2_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was I relative to the driver?", + "answer": "To your back-right. The driver is to your front-left.", + "question_chinese": "我在司机的哪里？", + "answer_chinese": "右后方，司机在我的左前方。", + "evidence": [ + { + "description": "Rear right", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-left.", + "B": "To your front-right.", + "C": "To your back-right, but the driver was to your front-right instead of your front-left.", + "D": "To your front-left, with the driver also on your left side rather than across your view.", + "E": "To your back-right. The driver is to your front-left." + }, + "answer_label": "E", + "answer_text": "To your back-right. The driver is to your front-left.", + "q_id": "01211_2_T2", + "correct_answer": "To your back-right. The driver is to your front-left.", + "original_options": { + "A": "To your back-left.", + "B": "To your front-right.", + "C": "To your back-right, but the driver was to your front-right instead of your front-left.", + "D": "To your front-left, with the driver also on your left side rather than across your view.", + "E": "To your back-right. The driver is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 16 + ] + }, + { + "video_uid": "01211", + "video_length": "00:04:49", + "qid": "01211_3_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Was my rabbit plush on the cane?", + "answer": "Yes—it was placed to your left.", + "question_chinese": "我的兔子玩偶在拐杖上吗？", + "answer_chinese": "是，放在你的左边。", + "evidence": [ + { + "description": "The rabbit doll was on the cane", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—it wasn’t on the cane; it was off to your left.", + "B": "Yes—it was placed to your left." + }, + "answer_label": "B", + "answer_text": "Yes—it was placed to your left.", + "q_id": "01211_3_T1", + "correct_answer": "Yes—it was placed to your left.", + "original_options": { + "A": "No—it wasn’t on the cane; it was off to your left.", + "B": "Yes—it was placed to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "01211", + "video_length": "00:04:49", + "qid": "01211_3_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Was my rabbit plush on the cane?", + "answer": "Yes—it was placed to your right.", + "question_chinese": "我的兔子玩偶在拐杖上吗？", + "answer_chinese": "是，放在你的右边。", + "evidence": [ + { + "description": "The rabbit doll was on the cane", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:04:49", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—it was placed to your right.", + "B": "No—it wasn’t on the cane; it was off to your right." + }, + "answer_label": "A", + "answer_text": "Yes—it was placed to your right.", + "q_id": "01211_3_T2", + "correct_answer": "Yes—it was placed to your right.", + "original_options": { + "A": "Yes—it was placed to your right.", + "B": "No—it wasn’t on the cane; it was off to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "01211", + "video_length": "00:04:49", + "qid": "01211_1_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is in my right hand?", + "answer": "A bangle, down-right relative to your view.", + "question_chinese": "我的右手上有什么？", + "answer_chinese": "镯子，在视角的右下方。", + "evidence": [ + { + "description": "There was a bracelet on my right hand", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A bangle, down-left.", + "B": "A bangle, up-right.", + "C": "A bangle, down-right relative to your view.", + "D": "A bangle in your right hand, but it sits up-left relative to your view instead.", + "E": "A bangle in your right hand, positioned centered in front of you rather than down-right." + }, + "answer_label": "C", + "answer_text": "A bangle, down-right relative to your view.", + "q_id": "01211_1_T1", + "correct_answer": "A bangle, down-right relative to your view.", + "original_options": { + "A": "A bangle, down-left.", + "B": "A bangle, up-right.", + "C": "A bangle, down-right relative to your view.", + "D": "A bangle in your right hand, but it sits up-left relative to your view instead.", + "E": "A bangle in your right hand, positioned centered in front of you rather than down-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "01211", + "video_length": "00:04:49", + "qid": "01211_1_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is in my right hand?", + "answer": "A bangle, to your right.", + "question_chinese": "我的右手上有什么？", + "answer_chinese": "一个镯子，在视角的右边。", + "evidence": [ + { + "description": "There was a bracelet on my right hand", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A bangle, to your left.", + "B": "A bangle, up-left.", + "C": "A bangle in your right hand, but it appears down-right relative to your view instead.", + "D": "A bangle, to your right.", + "E": "A bangle in your right hand, held centered in front of you rather than to your right." + }, + "answer_label": "D", + "answer_text": "A bangle, to your right.", + "q_id": "01211_1_T2", + "correct_answer": "A bangle, to your right.", + "original_options": { + "A": "A bangle, to your left.", + "B": "A bangle, up-left.", + "C": "A bangle in your right hand, but it appears down-right relative to your view instead.", + "D": "A bangle, to your right.", + "E": "A bangle in your right hand, held centered in front of you rather than to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 15, + 17 + ] + }, + { + "video_uid": "01211", + "video_length": "00:04:49", + "qid": "01211_5_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Did a sanitation worker pass by me?", + "answer": "Yes, there is. It’s to your right now.", + "question_chinese": "环卫工人有经过我吗？", + "answer_chinese": "有,现在在我右边。", + "evidence": [ + { + "description": "He has passed me", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, there is. It’s to your right now.", + "B": "No, there isn’t." + }, + "answer_label": "A", + "answer_text": "Yes, there is. It’s to your right now.", + "q_id": "01211_5_T2", + "correct_answer": "Yes, there is. It’s to your right now.", + "original_options": { + "A": "Yes, there is. It’s to your right now.", + "B": "No, there isn’t." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "01211", + "video_length": "00:04:49", + "qid": "01211_5_T1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Did a sanitation worker pass by me?", + "answer": "Yes, there is. It’s in front of you now.", + "question_chinese": "环卫工人有经过我吗？", + "answer_chinese": "有，现在在我前方。", + "evidence": [ + { + "description": "He has passed me", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, there isn’t.", + "B": "Yes, there is. It’s in front of you now." + }, + "answer_label": "B", + "answer_text": "Yes, there is. It’s in front of you now.", + "q_id": "01211_5_T1", + "correct_answer": "Yes, there is. It’s in front of you now.", + "original_options": { + "A": "No, there isn’t.", + "B": "Yes, there is. It’s in front of you now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + } +] diff --git a/QAs/01218-mcq.json b/QAs/01218-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ff187d7c5b67f5e75c4ab323154018ddbba01500 --- /dev/null +++ b/QAs/01218-mcq.json @@ -0,0 +1,944 @@ +[ + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_8_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I just see?", + "answer": "Two of them, all in front of you.", + "question_chinese": "我刚才看到了几个人？", + "answer_chinese": "2个，都在你的前方", + "evidence": [ + { + "description": "two person", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One person, in front of you.", + "B": "Three people, in front of you.", + "C": "Two of them, all in front of you.", + "D": "Four people, all positioned in front of you as you look ahead.", + "E": "Zero people—you don’t see anyone in front of you at that moment." + }, + "answer_label": "C", + "answer_text": "Two of them, all in front of you.", + "q_id": "01218_8_T2", + "correct_answer": "Two of them, all in front of you.", + "original_options": { + "A": "One person, in front of you.", + "B": "Three people, in front of you.", + "C": "Two of them, all in front of you.", + "D": "Four people, all positioned in front of you as you look ahead.", + "E": "Zero people—you don’t see anyone in front of you at that moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 14 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_8_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I just see?", + "answer": "Three of them, all in front of you.", + "question_chinese": "我刚才看到了几个人？", + "answer_chinese": "3个，都在你的前方", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:21", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two people, in front of you.", + "B": "Three of them, all in front of you.", + "C": "Four people, in front of you.", + "D": "One person, standing in front of you as you look straight ahead.", + "E": "Five people, all visible in front of you when you look forward." + }, + "answer_label": "B", + "answer_text": "Three of them, all in front of you.", + "q_id": "01218_8_T1", + "correct_answer": "Three of them, all in front of you.", + "original_options": { + "A": "Two people, in front of you.", + "B": "Three of them, all in front of you.", + "C": "Four people, in front of you.", + "D": "One person, standing in front of you as you look straight ahead.", + "E": "Five people, all visible in front of you when you look forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 12 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_2_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get past the red roadblock?", + "answer": "Yes—you can. It’s directly to your left.", + "question_chinese": "我可以经过红色的路障吗？", + "answer_chinese": "可以，在你的正左边", + "evidence": [ + { + "description": "You need to walk two meters", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t. It’s directly to your left.", + "B": "Yes—you can. It’s directly to your left." + }, + "answer_label": "B", + "answer_text": "Yes—you can. It’s directly to your left.", + "q_id": "01218_2_T1", + "correct_answer": "Yes—you can. It’s directly to your left.", + "original_options": { + "A": "No—you can’t. It’s directly to your left.", + "B": "Yes—you can. It’s directly to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_2_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get past the red roadblock?", + "answer": "Yes—you can. It’s right to your back-left.", + "question_chinese": "我可以经过红色的路障吗？", + "answer_chinese": "可以，它就在你的左后方。", + "evidence": [ + { + "description": "You need to walk one meter", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. It’s right to your back-left.", + "B": "No—you can’t. It’s right to your back-left." + }, + "answer_label": "A", + "answer_text": "Yes—you can. It’s right to your back-left.", + "q_id": "01218_2_T2", + "correct_answer": "Yes—you can. It’s right to your back-left.", + "original_options": { + "A": "Yes—you can. It’s right to your back-left.", + "B": "No—you can’t. It’s right to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_3_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I holding in my right hand?", + "answer": "A crutch, to your right in the view.", + "question_chinese": "我的右手拿着什么？", + "answer_chinese": "拐杖，在视角的右边。", + "evidence": [ + { + "description": "I held a cane", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A crutch, to your left in the view.", + "B": "A crutch, to your back-right in the view.", + "C": "A crutch, to your right in the view.", + "D": "A crutch, positioned to your front-right in the view as you hold it.", + "E": "A crutch, visible to your back-left in the view while you hold it." + }, + "answer_label": "C", + "answer_text": "A crutch, to your right in the view.", + "q_id": "01218_3_T2", + "correct_answer": "A crutch, to your right in the view.", + "original_options": { + "A": "A crutch, to your left in the view.", + "B": "A crutch, to your back-right in the view.", + "C": "A crutch, to your right in the view.", + "D": "A crutch, positioned to your front-right in the view as you hold it.", + "E": "A crutch, visible to your back-left in the view while you hold it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 13, + 13 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_3_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I holding in my right hand?", + "answer": "A crutch, to your back-right in the view.", + "question_chinese": "我的右手拿着什么？", + "answer_chinese": "拐杖，在视角的右后方。", + "evidence": [ + { + "description": "I held a cane", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A crutch, to your left in the view.", + "B": "A crutch, to your back-right in the view.", + "C": "A crutch, to your back-left in the view.", + "D": "A crutch, positioned to your front-right in the view as you hold it.", + "E": "A crutch, shown to your right in the view while it’s in your hand." + }, + "answer_label": "B", + "answer_text": "A crutch, to your back-right in the view.", + "q_id": "01218_3_T1", + "correct_answer": "A crutch, to your back-right in the view.", + "original_options": { + "A": "A crutch, to your left in the view.", + "B": "A crutch, to your back-right in the view.", + "C": "A crutch, to your back-left in the view.", + "D": "A crutch, positioned to your front-right in the view as you hold it.", + "E": "A crutch, shown to your right in the view while it’s in your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 13, + 15 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_7", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people wearing hats did I see?", + "answer": "two.", + "question_chinese": "我看到了多少个人戴帽子？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:21", + "end": "00:00:22" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:34", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "I saw four people wearing hats during the walk, not counting anyone without a hat.", + "D": "I saw five people wearing hats altogether, including one who was farther away.", + "E": "two." + }, + "answer_label": "E", + "answer_text": "two.", + "q_id": "01218_7", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "I saw four people wearing hats during the walk, not counting anyone without a hat.", + "D": "I saw five people wearing hats altogether, including one who was farther away.", + "E": "two." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 13 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_8", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people wearing hats did I see?", + "answer": "Three.", + "question_chinese": "我看到了多少个人戴帽子？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:21", + "end": "00:00:22" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Only one person was wearing a hat when I looked around during the scene.", + "D": "You saw five people wearing hats altogether, including one near the edge of view.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "01218_8", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Only one person was wearing a hat when I looked around during the scene.", + "D": "You saw five people wearing hats altogether, including one near the edge of view.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_9_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where am I going to buy breakfast?", + "answer": "Go straight for 6 meters, then turn right and walk 2 meters to reach the breakfast shop to your front-right.", + "question_chinese": "我要去哪里买早餐？", + "answer_chinese": "先直走六米，再右转走两米，到达右前方的早餐店。", + "evidence": [ + { + "description": "Go straight for six meters", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "turn right and walk two meters", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go straight 5 meters.", + "B": "Turn right after 7 meters.", + "C": "Go straight for 6 meters, then turn right and walk 2 meters to reach the breakfast shop to your front-right.", + "D": "Go straight for 6 meters, turn left, then walk 2 meters to the breakfast shop on your front-left.", + "E": "Go straight 6 meters, turn right, then walk 3 meters to reach the breakfast shop to your front-right." + }, + "answer_label": "C", + "answer_text": "Go straight for 6 meters, then turn right and walk 2 meters to reach the breakfast shop to your front-right.", + "q_id": "01218_9_T1", + "correct_answer": "Go straight for 6 meters, then turn right and walk 2 meters to reach the breakfast shop to your front-right.", + "original_options": { + "A": "Go straight 5 meters.", + "B": "Turn right after 7 meters.", + "C": "Go straight for 6 meters, then turn right and walk 2 meters to reach the breakfast shop to your front-right.", + "D": "Go straight for 6 meters, turn left, then walk 2 meters to the breakfast shop on your front-left.", + "E": "Go straight 6 meters, turn right, then walk 3 meters to reach the breakfast shop to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 18 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_9_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where am I going to buy breakfast?", + "answer": "Go straight for 6 meters, then turn right and walk 2 meters to reach the breakfast shop to your front-right.", + "question_chinese": "我要去哪里买早餐？", + "answer_chinese": "先直走六米，再右转走两米，到达你右前方的早餐店", + "evidence": [ + { + "description": "Go straight for six meters", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "turn right and walk two meters to reach the breakfast shop.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go straight 4 meters.", + "B": "Turn right after 8 meters.", + "C": "Go straight for 6 meters, then turn left and walk 2 meters to reach the shop on your front-left.", + "D": "Go straight for 6 meters, then turn right and walk 2 meters to reach the breakfast shop to your front-right.", + "E": "Go straight for 5 meters, then turn right and walk 3 meters to reach the breakfast shop to your front-right." + }, + "answer_label": "D", + "answer_text": "Go straight for 6 meters, then turn right and walk 2 meters to reach the breakfast shop to your front-right.", + "q_id": "01218_9_T2", + "correct_answer": "Go straight for 6 meters, then turn right and walk 2 meters to reach the breakfast shop to your front-right.", + "original_options": { + "A": "Go straight 4 meters.", + "B": "Turn right after 8 meters.", + "C": "Go straight for 6 meters, then turn left and walk 2 meters to reach the shop on your front-left.", + "D": "Go straight for 6 meters, then turn right and walk 2 meters to reach the breakfast shop to your front-right.", + "E": "Go straight for 5 meters, then turn right and walk 3 meters to reach the breakfast shop to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 19, + 20 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_6_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How can I walk to the space between the white car and the black car?", + "answer": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "question_chinese": "我要怎么才能走到白车和黑车之间？", + "answer_chinese": "先直走，左转走两米，再右转走三米。白车和黑车在你的左前方。", + "evidence": [ + { + "description": "Go straight first", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "turn left and walk two meters", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "turn right and walk three meters", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then right.", + "B": "Turn right, then left.", + "C": "Go straight first, then turn left and walk 1 meter, then turn right and walk 3 meters front-left.", + "D": "Go straight first, then turn left and walk 3 meters, then turn right and walk 3 meters; cars are front-left.", + "E": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left." + }, + "answer_label": "E", + "answer_text": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "q_id": "01218_6_T1", + "correct_answer": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "original_options": { + "A": "Turn left, then right.", + "B": "Turn right, then left.", + "C": "Go straight first, then turn left and walk 1 meter, then turn right and walk 3 meters front-left.", + "D": "Go straight first, then turn left and walk 3 meters, then turn right and walk 3 meters; cars are front-left.", + "E": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 20 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_6_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How can I walk to the space between the white car and the black car?", + "answer": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "question_chinese": "我要怎么才能走到白车和黑车之间？", + "answer_chinese": "先直走，左转走两米，再右转走三米。白车和黑车在你的左前方1。", + "evidence": [ + { + "description": "Go straight first", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "turn left and walk two meters", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "turn right and walk three meters", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "B": "Turn right, then left.", + "C": "Walk 4 meters at the end.", + "D": "Go straight first, then turn left and walk 2 meters, then turn right and walk 4 meters; cars front-left.", + "E": "Go straight first, then turn right and walk 2 meters, then turn left and walk 3 meters; cars front-right." + }, + "answer_label": "A", + "answer_text": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "q_id": "01218_6_T2", + "correct_answer": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "original_options": { + "A": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "B": "Turn right, then left.", + "C": "Walk 4 meters at the end.", + "D": "Go straight first, then turn left and walk 2 meters, then turn right and walk 4 meters; cars front-left.", + "E": "Go straight first, then turn right and walk 2 meters, then turn left and walk 3 meters; cars front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 19, + 19 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_7_T2", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How should I walk to the space between the white car and the black car?", + "answer": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "question_chinese": "我该怎么走到白车和黑车之间？", + "answer_chinese": "先直走，左转走两米，再右转走三米。白车和黑车在你的左前方。", + "evidence": [ + { + "description": "Go straight first", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "turn left and walk two meters", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "turn right and walk three meters", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left twice.", + "B": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "C": "Walk only 2 meters at end.", + "D": "Go straight first, then turn left and walk 1 meter, then turn right and walk 3 meters; cars front-left.", + "E": "Go straight first, then turn left and walk 2 meters, then turn right and walk 2 meters; cars front-left." + }, + "answer_label": "B", + "answer_text": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "q_id": "01218_7_T2", + "correct_answer": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "original_options": { + "A": "Turn left twice.", + "B": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters. The white car and the black car are to your front-left.", + "C": "Walk only 2 meters at end.", + "D": "Go straight first, then turn left and walk 1 meter, then turn right and walk 3 meters; cars front-left.", + "E": "Go straight first, then turn left and walk 2 meters, then turn right and walk 2 meters; cars front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 6, + 19, + 19 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_7_T1", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How should I walk to the space between the white car and the black car?", + "answer": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters to reach your front-left.", + "question_chinese": "我该怎么走到白车和黑车之间？", + "answer_chinese": "先直走，左转走两米，再右转走三米到达你的左前方。", + "evidence": [ + { + "description": "Go straight first", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "turn left and walk two meters", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "turn right and walk three meters", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters to reach your front-left.", + "B": "Walk 4 meters at the end.", + "C": "Turn right, then left.", + "D": "Go straight first, then turn left and walk 1 meter, then turn right and walk 3 meters to reach front-left.", + "E": "Go straight first, then turn left and walk 2 meters, then turn left and walk 3 meters to reach front-left." + }, + "answer_label": "A", + "answer_text": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters to reach your front-left.", + "q_id": "01218_7_T1", + "correct_answer": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters to reach your front-left.", + "original_options": { + "A": "Go straight first, then turn left and walk 2 meters, then turn right and walk 3 meters to reach your front-left.", + "B": "Walk 4 meters at the end.", + "C": "Turn right, then left.", + "D": "Go straight first, then turn left and walk 1 meter, then turn right and walk 3 meters to reach front-left.", + "E": "Go straight first, then turn left and walk 2 meters, then turn left and walk 3 meters to reach front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 20, + 20 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_5_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the yellow bicycle relative to me?", + "answer": "The yellow bicycle is to your front-right.", + "question_chinese": "黄色的单车在我的哪里？", + "answer_chinese": "黄色单车在我的右前方。", + "evidence": [ + { + "description": "Go straight for three meters", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "turn right and go two meters", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "The yellow bicycle is directly to your right as you face forward, not angled ahead of you.", + "D": "The yellow bicycle is to your front-right.", + "E": "The yellow bicycle is behind you on the left side, so you would need to turn around to see it." + }, + "answer_label": "D", + "answer_text": "The yellow bicycle is to your front-right.", + "q_id": "01218_5_T1", + "correct_answer": "The yellow bicycle is to your front-right.", + "original_options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "The yellow bicycle is directly to your right as you face forward, not angled ahead of you.", + "D": "The yellow bicycle is to your front-right.", + "E": "The yellow bicycle is behind you on the left side, so you would need to turn around to see it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 20 + ] + }, + { + "video_uid": "01218", + "video_length": "00:00:52", + "qid": "01218_5_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the yellow bicycle relative to me?", + "answer": "The yellow bicycle is to your right.", + "question_chinese": "黄色的单车在我的哪里？", + "answer_chinese": "黄色单车在我的右方。", + "evidence": [ + { + "description": "Go straight for three meters", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "turn right and go two meters", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "Directly in front of you.", + "C": "The yellow bicycle is to your front-right, slightly ahead of your shoulder rather than beside you.", + "D": "The yellow bicycle is behind you to the right, so it would be over your right shoulder.", + "E": "The yellow bicycle is to your right." + }, + "answer_label": "E", + "answer_text": "The yellow bicycle is to your right.", + "q_id": "01218_5_T2", + "correct_answer": "The yellow bicycle is to your right.", + "original_options": { + "A": "To your left.", + "B": "Directly in front of you.", + "C": "The yellow bicycle is to your front-right, slightly ahead of your shoulder rather than beside you.", + "D": "The yellow bicycle is behind you to the right, so it would be over your right shoulder.", + "E": "The yellow bicycle is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 5, + 16, + 17 + ] + } +] diff --git a/QAs/01221-mcq.json b/QAs/01221-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ccc955e7d4642425b8d1e79c85d278b3cc23ad93 --- /dev/null +++ b/QAs/01221-mcq.json @@ -0,0 +1,807 @@ +[ + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_10", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I pass just now?", + "answer": "two.", + "question_chinese": "我刚才经过了多少个人？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You passed one person just now, so there was only a single passerby.", + "D": "two.", + "E": "You passed three people just now, meaning you went by a small group instead of just two." + }, + "answer_label": "D", + "answer_text": "two.", + "q_id": "01221_10", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You passed one person just now, so there was only a single passerby.", + "D": "two.", + "E": "You passed three people just now, meaning you went by a small group instead of just two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 17 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_9", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I pass just now?", + "answer": "Three.", + "question_chinese": "我刚才经过了多少个人？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You passed two people just now, so it was just a pair rather than three.", + "E": "You passed four people just now, indicating you went by a larger group than three." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "01221_9", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You passed two people just now, so it was just a pair rather than three.", + "E": "You passed four people just now, indicating you went by a larger group than three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_7_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cars did I pass by just now?", + "answer": "Two, all to your back-left.", + "question_chinese": "我刚才经过了多少辆车？", + "answer_chinese": "2辆，都在你的左后方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, all to your back-left.", + "B": "One, to your back-left.", + "C": "Three, to your back-left.", + "D": "You passed one car, and it was positioned to your back-left as you moved past.", + "E": "You passed three cars, all located to your back-left, not just two." + }, + "answer_label": "A", + "answer_text": "Two, all to your back-left.", + "q_id": "01221_7_T2", + "correct_answer": "Two, all to your back-left.", + "original_options": { + "A": "Two, all to your back-left.", + "B": "One, to your back-left.", + "C": "Three, to your back-left.", + "D": "You passed one car, and it was positioned to your back-left as you moved past.", + "E": "You passed three cars, all located to your back-left, not just two." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 12 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_7_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cars did I pass by just now?", + "answer": "Three cars, all to your back-left.", + "question_chinese": "我刚才经过了多少辆车？", + "answer_chinese": "3辆车，都在你的左后方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:00:41", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two cars, back-left.", + "B": "Four cars, back-left.", + "C": "You passed two cars, and both were to your back-left as you went by.", + "D": "You passed four cars, all to your back-left, which is one more than three.", + "E": "Three cars, all to your back-left." + }, + "answer_label": "E", + "answer_text": "Three cars, all to your back-left.", + "q_id": "01221_7_T1", + "correct_answer": "Three cars, all to your back-left.", + "original_options": { + "A": "Two cars, back-left.", + "B": "Four cars, back-left.", + "C": "You passed two cars, and both were to your back-left as you went by.", + "D": "You passed four cars, all to your back-left, which is one more than three.", + "E": "Three cars, all to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 14 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_4_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What should I do if my phone is out of battery?", + "answer": "Walk forward 1 meter, then reach out to the right to borrow a power bank.", + "question_chinese": "我的手机没电了怎么办？", + "answer_chinese": "往前走一米，向右伸手借充电宝。", + "evidence": [ + { + "description": "Walk forward two meters", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + }, + { + "description": "reach to the right to borrow a power bank", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk 2 meters forward, then reach right.", + "B": "Walk forward 1 meter, then reach out to the right to borrow a power bank.", + "C": "Walk 1 meter forward, then reach left.", + "D": "Walk forward 1 meter, then reach straight ahead to borrow a power bank instead of reaching to the right.", + "E": "Walk forward 0 meters, then reach out to the right to borrow a power bank without taking a step forward." + }, + "answer_label": "B", + "answer_text": "Walk forward 1 meter, then reach out to the right to borrow a power bank.", + "q_id": "01221_4_T1", + "correct_answer": "Walk forward 1 meter, then reach out to the right to borrow a power bank.", + "original_options": { + "A": "Walk 2 meters forward, then reach right.", + "B": "Walk forward 1 meter, then reach out to the right to borrow a power bank.", + "C": "Walk 1 meter forward, then reach left.", + "D": "Walk forward 1 meter, then reach straight ahead to borrow a power bank instead of reaching to the right.", + "E": "Walk forward 0 meters, then reach out to the right to borrow a power bank without taking a step forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 20 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_4_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What should I do if my phone is out of battery?", + "answer": "Walk forward 2 meters, then reach out to the right to borrow a power bank.", + "question_chinese": "我的手机没电了怎么办？", + "answer_chinese": "往前走两米，向右伸手借取充电宝。", + "evidence": [ + { + "description": "Walk forward two meters", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + }, + { + "description": "reach to the right to borrow a power bank", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk 1 meter forward, then reach right.", + "B": "Walk 2 meters forward, then reach left.", + "C": "Walk forward 2 meters, then reach out to the right to borrow a power bank.", + "D": "Walk forward 3 meters, then reach out to the right to borrow a power bank after moving farther ahead.", + "E": "Walk forward 2 meters, then reach straight ahead to borrow a power bank rather than reaching to the right." + }, + "answer_label": "C", + "answer_text": "Walk forward 2 meters, then reach out to the right to borrow a power bank.", + "q_id": "01221_4_T2", + "correct_answer": "Walk forward 2 meters, then reach out to the right to borrow a power bank.", + "original_options": { + "A": "Walk 1 meter forward, then reach right.", + "B": "Walk 2 meters forward, then reach left.", + "C": "Walk forward 2 meters, then reach out to the right to borrow a power bank.", + "D": "Walk forward 3 meters, then reach out to the right to borrow a power bank after moving farther ahead.", + "E": "Walk forward 2 meters, then reach straight ahead to borrow a power bank rather than reaching to the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_3_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see the clerk holding in their hand?", + "answer": "Drinks, directly to your right.", + "question_chinese": "我看到店员的手里拿着什么？", + "answer_chinese": "饮料，在你的正右方。", + "evidence": [ + { + "description": "He is holding drinks", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Drinks, directly to your right.", + "B": "Drinks, directly to your left.", + "C": "Drinks, to your back-right.", + "D": "The clerk held drinks off to your left side as you faced them.", + "E": "The clerk held drinks slightly behind you on the right, not directly beside you." + }, + "answer_label": "A", + "answer_text": "Drinks, directly to your right.", + "q_id": "01221_3_T1", + "correct_answer": "Drinks, directly to your right.", + "original_options": { + "A": "Drinks, directly to your right.", + "B": "Drinks, directly to your left.", + "C": "Drinks, to your back-right.", + "D": "The clerk held drinks off to your left side as you faced them.", + "E": "The clerk held drinks slightly behind you on the right, not directly beside you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 14 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_3_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see the clerk holding in their hand?", + "answer": "Drinks, to your back-right.", + "question_chinese": "我看到店员的手里拿着什么？", + "answer_chinese": "饮料，在你的右后方。", + "evidence": [ + { + "description": "He held drinks", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Drinks, directly to your right.", + "B": "Drinks, to your back-right.", + "C": "Drinks, to your back-left.", + "D": "The clerk was holding drinks in front of you on the right side.", + "E": "You saw the clerk holding drinks to your left side rather than behind you." + }, + "answer_label": "B", + "answer_text": "Drinks, to your back-right.", + "q_id": "01221_3_T2", + "correct_answer": "Drinks, to your back-right.", + "original_options": { + "A": "Drinks, directly to your right.", + "B": "Drinks, to your back-right.", + "C": "Drinks, to your back-left.", + "D": "The clerk was holding drinks in front of you on the right side.", + "E": "You saw the clerk holding drinks to your left side rather than behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 14 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_1_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How do I throw away the trash?", + "answer": "Throw it away in the trash can to your front-left.", + "question_chinese": "我要怎么丢垃圾？", + "answer_chinese": "去你左前方的垃圾桶丢", + "evidence": [ + { + "description": "The can is in front of me", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use the trash can front-right.", + "B": "Use the trash can behind you.", + "C": "Throw it away in the trash can to your back-left, not the one in front of you.", + "D": "Put the trash into the can to your front-right, then step away once it drops in.", + "E": "Throw it away in the trash can to your front-left." + }, + "answer_label": "E", + "answer_text": "Throw it away in the trash can to your front-left.", + "q_id": "01221_1_T2", + "correct_answer": "Throw it away in the trash can to your front-left.", + "original_options": { + "A": "Use the trash can front-right.", + "B": "Use the trash can behind you.", + "C": "Throw it away in the trash can to your back-left, not the one in front of you.", + "D": "Put the trash into the can to your front-right, then step away once it drops in.", + "E": "Throw it away in the trash can to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 16 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_1_T1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How do I throw away the trash?", + "answer": "Throw it away in the trash can to your back-left.", + "question_chinese": "我要怎么丢垃圾？", + "answer_chinese": "去你左后方的垃圾桶丢", + "evidence": [ + { + "description": "The can is in front of me", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use the trash can front-left.", + "B": "Use the trash can back-right.", + "C": "Throw it away in the trash can directly behind you, instead of the one off to the left.", + "D": "Throw it away in the trash can to your back-left.", + "E": "Toss it into the trash can to your front-right, not the one located behind you on the left." + }, + "answer_label": "D", + "answer_text": "Throw it away in the trash can to your back-left.", + "q_id": "01221_1_T1", + "correct_answer": "Throw it away in the trash can to your back-left.", + "original_options": { + "A": "Use the trash can front-left.", + "B": "Use the trash can back-right.", + "C": "Throw it away in the trash can directly behind you, instead of the one off to the left.", + "D": "Throw it away in the trash can to your back-left.", + "E": "Toss it into the trash can to your front-right, not the one located behind you on the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_6_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How do I get to the other side?", + "answer": "Walk forward 3 meters, turn left and cross the road, and you’ll arrive at your front-left.", + "question_chinese": "我要怎么去对面？", + "answer_chinese": "往前走三米，左转过马路，到达你的左前方。", + "evidence": [ + { + "description": "Walk ten meters forward", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk forward 2 meters, turn left.", + "B": "Walk forward 4 meters, turn left.", + "C": "Walk forward 3 meters, turn right and cross the road, arriving at your front-right.", + "D": "Walk forward 3 meters, turn left and cross the road, but you’ll end up at your front-right.", + "E": "Walk forward 3 meters, turn left and cross the road, and you’ll arrive at your front-left." + }, + "answer_label": "E", + "answer_text": "Walk forward 3 meters, turn left and cross the road, and you’ll arrive at your front-left.", + "q_id": "01221_6_T2", + "correct_answer": "Walk forward 3 meters, turn left and cross the road, and you’ll arrive at your front-left.", + "original_options": { + "A": "Walk forward 2 meters, turn left.", + "B": "Walk forward 4 meters, turn left.", + "C": "Walk forward 3 meters, turn right and cross the road, arriving at your front-right.", + "D": "Walk forward 3 meters, turn left and cross the road, but you’ll end up at your front-right.", + "E": "Walk forward 3 meters, turn left and cross the road, and you’ll arrive at your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 18 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_6_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How do I get to the other side?", + "answer": "Walk forward 10 meters, turn left and cross the crosswalk, and you’ll arrive at your front-left.", + "question_chinese": "我要怎么去对面？", + "answer_chinese": "往前走十米，左转过斑马线，到达你的左前方。", + "evidence": [ + { + "description": "Walk ten meters forward", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "turn left at the crosswalk", + "timestamps": { + "start": "00:00:26", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:00:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk forward 9 meters, turn left.", + "B": "Walk forward 11 meters, turn left.", + "C": "Walk forward 10 meters, turn right and cross the crosswalk, arriving at your front-right.", + "D": "Walk forward 10 meters, turn left and cross the crosswalk, and you’ll arrive at your front-left.", + "E": "Walk forward 10 meters, turn left and cross the crosswalk, but you’ll arrive at your front-right." + }, + "answer_label": "D", + "answer_text": "Walk forward 10 meters, turn left and cross the crosswalk, and you’ll arrive at your front-left.", + "q_id": "01221_6_T1", + "correct_answer": "Walk forward 10 meters, turn left and cross the crosswalk, and you’ll arrive at your front-left.", + "original_options": { + "A": "Walk forward 9 meters, turn left.", + "B": "Walk forward 11 meters, turn left.", + "C": "Walk forward 10 meters, turn right and cross the crosswalk, arriving at your front-right.", + "D": "Walk forward 10 meters, turn left and cross the crosswalk, and you’ll arrive at your front-left.", + "E": "Walk forward 10 meters, turn left and cross the crosswalk, but you’ll arrive at your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 17 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_2_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the yellow cart relative to me?", + "answer": "It’s to the left of the refrigerator, at your front-right.", + "question_chinese": "黄色的小推车在我的哪里？", + "answer_chinese": "在冰箱的左边，你的右前方。", + "evidence": [ + { + "description": "To the left of the refrigerator", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to the left of the refrigerator, at your front-right.", + "B": "It’s at your front-left.", + "C": "It’s at your back-right.", + "D": "It’s to the right of the refrigerator, and it sits at your front-right as you face forward.", + "E": "It’s to the left of the refrigerator, but it’s located behind you on the right side." + }, + "answer_label": "A", + "answer_text": "It’s to the left of the refrigerator, at your front-right.", + "q_id": "01221_2_T2", + "correct_answer": "It’s to the left of the refrigerator, at your front-right.", + "original_options": { + "A": "It’s to the left of the refrigerator, at your front-right.", + "B": "It’s at your front-left.", + "C": "It’s at your back-right.", + "D": "It’s to the right of the refrigerator, and it sits at your front-right as you face forward.", + "E": "It’s to the left of the refrigerator, but it’s located behind you on the right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "01221", + "video_length": "00:00:47", + "qid": "01221_2_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the yellow cart relative to me?", + "answer": "It’s to your back-right.", + "question_chinese": "黄色的小推车在我的哪里？", + "answer_chinese": "在你的右后方。", + "evidence": [ + { + "description": "To the left of the refrigerator", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your back-left.", + "B": "It’s to your front-right.", + "C": "It was positioned at your front-left, not behind you, when you last looked in that direction.", + "D": "The yellow cart was directly behind you rather than angled off to your right side.", + "E": "It’s to your back-right." + }, + "answer_label": "E", + "answer_text": "It’s to your back-right.", + "q_id": "01221_2_T1", + "correct_answer": "It’s to your back-right.", + "original_options": { + "A": "It’s to your back-left.", + "B": "It’s to your front-right.", + "C": "It was positioned at your front-left, not behind you, when you last looked in that direction.", + "D": "The yellow cart was directly behind you rather than angled off to your right side.", + "E": "It’s to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + } +] diff --git a/QAs/01227-mcq.json b/QAs/01227-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9a6f4027e638ce64fc9512f7fc3e632bf7b263c8 --- /dev/null +++ b/QAs/01227-mcq.json @@ -0,0 +1,863 @@ +[ + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_2_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which side of the girl in white clothes was I on just now?", + "answer": "To your right. The girl in white was to your left.", + "question_chinese": "我刚才在白色衣服女生的哪一边？", + "answer_chinese": "右边，白衣女生在我的左方。", + "evidence": [ + { + "description": "On her right", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "To your right. The girl in white was to your left.", + "C": "Behind you.", + "D": "To your left; the girl in white was to your right the whole time.", + "E": "In front of you, with the girl in white positioned directly behind you." + }, + "answer_label": "B", + "answer_text": "To your right. The girl in white was to your left.", + "q_id": "01227_2_T2", + "correct_answer": "To your right. The girl in white was to your left.", + "original_options": { + "A": "To your left.", + "B": "To your right. The girl in white was to your left.", + "C": "Behind you.", + "D": "To your left; the girl in white was to your right the whole time.", + "E": "In front of you, with the girl in white positioned directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 14, + 13 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_2_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which side of the girl in white clothes was I on just now?", + "answer": "In front of you. The girl in white was behind you.", + "question_chinese": "我刚才在白色衣服女生的哪一边？", + "answer_chinese": "前方，白衣女生在你的后方。", + "evidence": [ + { + "description": "On her right", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you. The girl in white was behind you.", + "B": "Behind you.", + "C": "To your right.", + "D": "To your left, while the girl in white was positioned to your right.", + "E": "In front of you, and the girl in white stayed in front of you as well." + }, + "answer_label": "A", + "answer_text": "In front of you. The girl in white was behind you.", + "q_id": "01227_2_T1", + "correct_answer": "In front of you. The girl in white was behind you.", + "original_options": { + "A": "In front of you. The girl in white was behind you.", + "B": "Behind you.", + "C": "To your right.", + "D": "To your left, while the girl in white was positioned to your right.", + "E": "In front of you, and the girl in white stayed in front of you as well." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 16 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_4_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I make in total just now?", + "answer": "One—you turned to your right-hand side.", + "question_chinese": "我刚才总共拐了多少个弯？", + "answer_chinese": "1个，往你的右手边拐弯。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero—you didn’t turn.", + "B": "Two—you turned twice.", + "C": "Three—you turned right, then left, then right again, ending up facing forward afterward.", + "D": "One—you turned to your right-hand side.", + "E": "Two—you turned to your left-hand side and then turned back to the right." + }, + "answer_label": "D", + "answer_text": "One—you turned to your right-hand side.", + "q_id": "01227_4_T2", + "correct_answer": "One—you turned to your right-hand side.", + "original_options": { + "A": "Zero—you didn’t turn.", + "B": "Two—you turned twice.", + "C": "Three—you turned right, then left, then right again, ending up facing forward afterward.", + "D": "One—you turned to your right-hand side.", + "E": "Two—you turned to your left-hand side and then turned back to the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 14 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_4_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I make in total just now?", + "answer": "Two—first you turned to your left, then you turned to your right.", + "question_chinese": "我刚才总共拐了多少个弯？", + "answer_chinese": "2，先是往你的左边转，然后往右边转。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero—you didn’t turn.", + "B": "One—you turned left once.", + "C": "Two—first you turned to your left, then you turned to your right.", + "D": "Three—first you turned left, then right, then left again before continuing straight.", + "E": "Four—you turned left, then right, then left, then right in sequence." + }, + "answer_label": "C", + "answer_text": "Two—first you turned to your left, then you turned to your right.", + "q_id": "01227_4_T1", + "correct_answer": "Two—first you turned to your left, then you turned to your right.", + "original_options": { + "A": "Zero—you didn’t turn.", + "B": "One—you turned left once.", + "C": "Two—first you turned to your left, then you turned to your right.", + "D": "Three—first you turned left, then right, then left again before continuing straight.", + "E": "Four—you turned left, then right, then left, then right in sequence." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 12 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_5_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk into the elevator just now?", + "answer": "First you walked 2 meters to the left, then went right into the elevator, and then turned left.", + "question_chinese": "我刚才是怎么走进电梯的？", + "answer_chinese": "先向左走两米，再向右走进电梯，然后左转了。", + "evidence": [ + { + "description": "First walk two meters to the left", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + }, + { + "description": "turn right to enter the elevator.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walked left, then entered straight.", + "B": "Went right first, then entered.", + "C": "First you walked 3 meters to the left, then went right into the elevator, then turned left.", + "D": "First you walked 2 meters to the left, then went right into the elevator, and then turned left.", + "E": "First you walked 2 meters to the left, then went left into the elevator, and then turned right." + }, + "answer_label": "D", + "answer_text": "First you walked 2 meters to the left, then went right into the elevator, and then turned left.", + "q_id": "01227_5_T1", + "correct_answer": "First you walked 2 meters to the left, then went right into the elevator, and then turned left.", + "original_options": { + "A": "Walked left, then entered straight.", + "B": "Went right first, then entered.", + "C": "First you walked 3 meters to the left, then went right into the elevator, then turned left.", + "D": "First you walked 2 meters to the left, then went right into the elevator, and then turned left.", + "E": "First you walked 2 meters to the left, then went left into the elevator, and then turned right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_5_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk into the elevator just now?", + "answer": "First you walked 2 meters to the left, then went right and entered the elevator in front of you.", + "question_chinese": "我刚才是怎么走进电梯的？", + "answer_chinese": "先向左走两米，再向右走进你前方的电梯。", + "evidence": [ + { + "description": "First walk two meters to the left", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + }, + { + "description": "turn right to enter the elevator.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walked left, then entered on your left.", + "B": "Went right without moving left first.", + "C": "First you walked 1 meter to the left, then went right and entered the elevator in front of you.", + "D": "First you walked 2 meters to the left, then went right and entered the elevator behind you.", + "E": "First you walked 2 meters to the left, then went right and entered the elevator in front of you." + }, + "answer_label": "E", + "answer_text": "First you walked 2 meters to the left, then went right and entered the elevator in front of you.", + "q_id": "01227_5_T2", + "correct_answer": "First you walked 2 meters to the left, then went right and entered the elevator in front of you.", + "original_options": { + "A": "Walked left, then entered on your left.", + "B": "Went right without moving left first.", + "C": "First you walked 1 meter to the left, then went right and entered the elevator in front of you.", + "D": "First you walked 2 meters to the left, then went right and entered the elevator behind you.", + "E": "First you walked 2 meters to the left, then went right and entered the elevator in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 17 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_1_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Did I just touch the button?", + "answer": "Yes—the button is to your front-left.", + "question_chinese": "我刚才有没有碰到按钮？", + "answer_chinese": "有，按钮在左前方。", + "evidence": [ + { + "description": "I pressed the button just now", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—the button is to your front-left.", + "B": "No—you didn’t touch it; the button is to your front-left." + }, + "answer_label": "A", + "answer_text": "Yes—the button is to your front-left.", + "q_id": "01227_1_T1", + "correct_answer": "Yes—the button is to your front-left.", + "original_options": { + "A": "Yes—the button is to your front-left.", + "B": "No—you didn’t touch it; the button is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_1_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Did I just touch the button?", + "answer": "Yes—the button is to your front-right.", + "question_chinese": "我刚才有没有碰到按钮？", + "answer_chinese": "有，按钮在右前方。", + "evidence": [ + { + "description": "I pressed the button just now", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:59", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you didn’t touch it; the button is to your front-right.", + "B": "Yes—the button is to your front-right." + }, + "answer_label": "B", + "answer_text": "Yes—the button is to your front-right.", + "q_id": "01227_1_T2", + "correct_answer": "Yes—the button is to your front-right.", + "original_options": { + "A": "No—you didn’t touch it; the button is to your front-right.", + "B": "Yes—the button is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_6_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see just now?", + "answer": "4, in front of you.", + "question_chinese": "我刚才看见了多少人？", + "answer_chinese": "4个，在你的前方。", + "evidence": [ + { + "description": "I saw four person", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "4, in front of you.", + "B": "3, in front of you.", + "C": "5, in front of you.", + "D": "2 people, standing together in front of you as you looked ahead.", + "E": "6 people, all gathered in front of you as you looked ahead." + }, + "answer_label": "A", + "answer_text": "4, in front of you.", + "q_id": "01227_6_T2", + "correct_answer": "4, in front of you.", + "original_options": { + "A": "4, in front of you.", + "B": "3, in front of you.", + "C": "5, in front of you.", + "D": "2 people, standing together in front of you as you looked ahead.", + "E": "6 people, all gathered in front of you as you looked ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_6_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see just now?", + "answer": "Ten, in front of you.", + "question_chinese": "我刚才看见了多少人？", + "answer_chinese": "10，在你的前方。", + "evidence": [ + { + "description": "There is one person", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I saw 5 people", + "timestamps": { + "start": "00:00:19", + "end": "00:00:20" + } + }, + { + "description": "I saw 10 people", + "timestamps": { + "start": "00:00:26", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:00:29", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "9, in front of you.", + "B": "11, in front of you.", + "C": "8 people, grouped in front of you where you were looking just now.", + "D": "12 people, all visible in front of you at the same moment.", + "E": "Ten, in front of you." + }, + "answer_label": "E", + "answer_text": "Ten, in front of you.", + "q_id": "01227_6_T1", + "correct_answer": "Ten, in front of you.", + "original_options": { + "A": "9, in front of you.", + "B": "11, in front of you.", + "C": "8 people, grouped in front of you where you were looking just now.", + "D": "12 people, all visible in front of you at the same moment.", + "E": "Ten, in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 12 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_8_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see wearing white clothes?", + "answer": "One, to your left.", + "question_chinese": "我看到有多少个人穿了白色衣服？", + "answer_chinese": "1个，在我的左边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero, to your left.", + "B": "One, to your left.", + "C": "Two, to your left.", + "D": "Three people, to your left, were wearing white clothes when you looked just now.", + "E": "Four people, all to your left, had on white clothing when you looked just now." + }, + "answer_label": "B", + "answer_text": "One, to your left.", + "q_id": "01227_8_T1", + "correct_answer": "One, to your left.", + "original_options": { + "A": "Zero, to your left.", + "B": "One, to your left.", + "C": "Two, to your left.", + "D": "Three people, to your left, were wearing white clothes when you looked just now.", + "E": "Four people, all to your left, had on white clothing when you looked just now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_8_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see wearing white clothes?", + "answer": "Four people, all to your left.", + "question_chinese": "我看到有多少个人穿了白色衣服？", + "answer_chinese": "4个人，都在我的左方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I saw four people in white", + "timestamps": { + "start": "00:00:27", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three people, all to your left.", + "B": "Five people, all to your left.", + "C": "Four people, all to your left.", + "D": "Two people, both to your left, were wearing white clothes just now.", + "E": "Six people, all positioned to your left, had white clothes on just now." + }, + "answer_label": "C", + "answer_text": "Four people, all to your left.", + "q_id": "01227_8_T2", + "correct_answer": "Four people, all to your left.", + "original_options": { + "A": "Three people, all to your left.", + "B": "Five people, all to your left.", + "C": "Four people, all to your left.", + "D": "Two people, both to your left, were wearing white clothes just now.", + "E": "Six people, all positioned to your left, had white clothes on just now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 13 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_3_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I see indicator lights on the elevator buttons?", + "answer": "Yes—it's to your left.", + "question_chinese": "我看到电梯按钮有指示灯吗？", + "answer_chinese": "有，在你的左方。", + "evidence": [ + { + "description": "There are two lights", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—it's to your left.", + "B": "No—you didn’t see any indicator lights on the elevator buttons." + }, + "answer_label": "A", + "answer_text": "Yes—it's to your left.", + "q_id": "01227_3_T1", + "correct_answer": "Yes—it's to your left.", + "original_options": { + "A": "Yes—it's to your left.", + "B": "No—you didn’t see any indicator lights on the elevator buttons." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_3_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I see indicator lights on the elevator buttons?", + "answer": "Yes—it's to your back-right.", + "question_chinese": "我看到电梯按钮有指示灯吗？", + "answer_chinese": "有，在你的右后方。", + "evidence": [ + { + "description": "There are two lights", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you didn’t see any indicator lights on the elevator buttons.", + "B": "Yes—it's to your back-right." + }, + "answer_label": "B", + "answer_text": "Yes—it's to your back-right.", + "q_id": "01227_3_T2", + "correct_answer": "Yes—it's to your back-right.", + "original_options": { + "A": "No—you didn’t see any indicator lights on the elevator buttons.", + "B": "Yes—it's to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_7_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many elevator buttons did I see lit up?", + "answer": "Two—the buttons were to your back-right.", + "question_chinese": "我看到电梯的按钮亮了多少个？", + "answer_chinese": "2，按钮在你的右后方。", + "evidence": [ + { + "description": "the second one", + "timestamps": { + "start": "00:00:16", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One—the button was to your back-right.", + "B": "Two—the buttons were to your back-right.", + "C": "Three—the buttons were to your back-right.", + "D": "Zero—no elevator buttons were lit up to your back-right when you checked.", + "E": "Four—the lit elevator buttons were to your back-right when you looked over." + }, + "answer_label": "B", + "answer_text": "Two—the buttons were to your back-right.", + "q_id": "01227_7_T2", + "correct_answer": "Two—the buttons were to your back-right.", + "original_options": { + "A": "One—the button was to your back-right.", + "B": "Two—the buttons were to your back-right.", + "C": "Three—the buttons were to your back-right.", + "D": "Zero—no elevator buttons were lit up to your back-right when you checked.", + "E": "Four—the lit elevator buttons were to your back-right when you looked over." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 13 + ] + }, + { + "video_uid": "01227", + "video_length": "00:00:59", + "qid": "01227_7_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many elevator buttons did I see lit up?", + "answer": "Two—the buttons were in front of you and to your left.", + "question_chinese": "我看到电梯的按钮亮了多少个？", + "answer_chinese": "2个，按钮在你的前方和左方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:16", + "end": "00:00:18" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two—the buttons were in front of you and to your left.", + "B": "One—the button was in front-left.", + "C": "Three—the buttons were in front-left.", + "D": "Zero—no elevator buttons were lit up in front of you or to your left.", + "E": "Four—the lit elevator buttons were in front of you and to your left." + }, + "answer_label": "A", + "answer_text": "Two—the buttons were in front of you and to your left.", + "q_id": "01227_7_T1", + "correct_answer": "Two—the buttons were in front of you and to your left.", + "original_options": { + "A": "Two—the buttons were in front of you and to your left.", + "B": "One—the button was in front-left.", + "C": "Three—the buttons were in front-left.", + "D": "Zero—no elevator buttons were lit up in front of you or to your left.", + "E": "Four—the lit elevator buttons were in front of you and to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 14 + ] + } +] diff --git a/QAs/01235-mcq.json b/QAs/01235-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7294af61c080b97f9357d3eb620835a90aa0d310 --- /dev/null +++ b/QAs/01235-mcq.json @@ -0,0 +1,923 @@ +[ + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_2_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see just now?", + "answer": "Two—one is in front of you and one is to your right.", + "question_chinese": "我刚才见到了多少个人？", + "answer_chinese": "2，在你的前方和右方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One person in front of you.", + "B": "Three people to your right.", + "C": "Zero—you didn’t see anyone just now, so there weren’t any people in front of you or to your right.", + "D": "Four—one is in front of you, and three are positioned to your right side.", + "E": "Two—one is in front of you and one is to your right." + }, + "answer_label": "E", + "answer_text": "Two—one is in front of you and one is to your right.", + "q_id": "01235_2_T1", + "correct_answer": "Two—one is in front of you and one is to your right.", + "original_options": { + "A": "One person in front of you.", + "B": "Three people to your right.", + "C": "Zero—you didn’t see anyone just now, so there weren’t any people in front of you or to your right.", + "D": "Four—one is in front of you, and three are positioned to your right side.", + "E": "Two—one is in front of you and one is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 22, + 15 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_2_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see just now?", + "answer": "Three people, all to your right.", + "question_chinese": "我刚才见到了多少个人？", + "answer_chinese": "3个人，都在你的右方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "the third oen", + "timestamps": { + "start": "00:00:31", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three people, all to your right.", + "B": "Two people to your right.", + "C": "Four people to your right.", + "D": "One person, standing to your right with no one else nearby on that side.", + "E": "Five people, all clustered to your right rather than spread out in front of you." + }, + "answer_label": "A", + "answer_text": "Three people, all to your right.", + "q_id": "01235_2_T2", + "correct_answer": "Three people, all to your right.", + "original_options": { + "A": "Three people, all to your right.", + "B": "Two people to your right.", + "C": "Four people to your right.", + "D": "One person, standing to your right with no one else nearby on that side.", + "E": "Five people, all clustered to your right rather than spread out in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_6_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many red things did I see?", + "answer": "Two, both in front of you.", + "question_chinese": "我看到了多少个红色的东西？", + "answer_chinese": "2，都在你的前方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:05", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, in front of you.", + "B": "Three, in front of you.", + "C": "Zero—you didn’t see any red things, so there was nothing red in front of you.", + "D": "Four, all positioned in front of you instead of split to either side.", + "E": "Two, both in front of you." + }, + "answer_label": "E", + "answer_text": "Two, both in front of you.", + "q_id": "01235_6_T2", + "correct_answer": "Two, both in front of you.", + "original_options": { + "A": "One, in front of you.", + "B": "Three, in front of you.", + "C": "Zero—you didn’t see any red things, so there was nothing red in front of you.", + "D": "Four, all positioned in front of you instead of split to either side.", + "E": "Two, both in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 13 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_6_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many red things did I see?", + "answer": "Three—some are in front of you, and some are to your back-right.", + "question_chinese": "我看到了多少个红色的东西？", + "answer_chinese": "3，都在你的前方和右后方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:05", + "end": "00:01:06" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:01:47", + "end": "00:01:48" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, in front of you.", + "B": "Two—one in front, one back-right.", + "C": "Four—some are in front of you, and the rest are positioned to your back-right.", + "D": "Three—some are in front of you, and some are to your back-right.", + "E": "Five—there are a few in front of you, with the remaining red items to your back-right." + }, + "answer_label": "D", + "answer_text": "Three—some are in front of you, and some are to your back-right.", + "q_id": "01235_6_T1", + "correct_answer": "Three—some are in front of you, and some are to your back-right.", + "original_options": { + "A": "One, in front of you.", + "B": "Two—one in front, one back-right.", + "C": "Four—some are in front of you, and the rest are positioned to your back-right.", + "D": "Three—some are in front of you, and some are to your back-right.", + "E": "Five—there are a few in front of you, with the remaining red items to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 17 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_1_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where did I see the red pinwheel?", + "answer": "About 10 meters to your right.", + "question_chinese": "我看到红色的风车在哪里？", + "answer_chinese": "在我右方十米处", + "evidence": [ + { + "description": "Ten meters in front of me", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "About 8 meters to your right.", + "B": "About 12 meters to your right.", + "C": "Roughly 6 meters to your right, not as far as the distant objects on that side.", + "D": "About 10 meters to your right.", + "E": "Around 15 meters to your right, farther down the right side than you first noticed." + }, + "answer_label": "D", + "answer_text": "About 10 meters to your right.", + "q_id": "01235_1_T1", + "correct_answer": "About 10 meters to your right.", + "original_options": { + "A": "About 8 meters to your right.", + "B": "About 12 meters to your right.", + "C": "Roughly 6 meters to your right, not as far as the distant objects on that side.", + "D": "About 10 meters to your right.", + "E": "Around 15 meters to your right, farther down the right side than you first noticed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_1_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where did I see the red pinwheel?", + "answer": "About 10 meters in front of you.", + "question_chinese": "我看到红色的风车在哪里？", + "answer_chinese": "在我面前十米处", + "evidence": [ + { + "description": "Ten meters in front of me", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "About 8 meters in front of you.", + "B": "About 12 meters in front of you.", + "C": "Roughly 6 meters in front of you, closer than the far end of the path.", + "D": "Around 15 meters in front of you, farther ahead than the nearby foreground area.", + "E": "About 10 meters in front of you." + }, + "answer_label": "E", + "answer_text": "About 10 meters in front of you.", + "q_id": "01235_1_T2", + "correct_answer": "About 10 meters in front of you.", + "original_options": { + "A": "About 8 meters in front of you.", + "B": "About 12 meters in front of you.", + "C": "Roughly 6 meters in front of you, closer than the far end of the path.", + "D": "Around 15 meters in front of you, farther ahead than the nearby foreground area.", + "E": "About 10 meters in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 14 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_3_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Did I see the man in black always on the left side of the horse?", + "answer": "Yes. The man is on the left side of the horse and also on your left side.", + "question_chinese": "我看到黑衣男人一直在马的左侧吗？", + "answer_chinese": "是的，男人在马的左侧和我的左侧。", + "evidence": [ + { + "description": "left", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:00:54", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. The man is on the left side of the horse and also on your left side.", + "B": "No. The man is on the right side of the horse and also on your right side." + }, + "answer_label": "A", + "answer_text": "Yes. The man is on the left side of the horse and also on your left side.", + "q_id": "01235_3_T2", + "correct_answer": "Yes. The man is on the left side of the horse and also on your left side.", + "original_options": { + "A": "Yes. The man is on the left side of the horse and also on your left side.", + "B": "No. The man is on the right side of the horse and also on your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_3_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Did I see the man in black always on the left side of the horse?", + "answer": "No. He is now to your front-left, in front-left of both you and the horse.", + "question_chinese": "我看到黑衣男人一直在马的左侧吗？", + "answer_chinese": "不是，他现在在我和马的左前方。", + "evidence": [ + { + "description": "left", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + }, + { + "description": "in front of the horse", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. He is always on the left side of the horse.", + "B": "No. He is now to your front-left, in front-left of both you and the horse." + }, + "answer_label": "B", + "answer_text": "No. He is now to your front-left, in front-left of both you and the horse.", + "q_id": "01235_3_T1", + "correct_answer": "No. He is now to your front-left, in front-left of both you and the horse.", + "original_options": { + "A": "Yes. He is always on the left side of the horse.", + "B": "No. He is now to your front-left, in front-left of both you and the horse." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_4_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How do I get the black bag?", + "answer": "Walk forward five meters, then reach out your right hand.", + "question_chinese": "我要怎样拿到黑色的包？", + "answer_chinese": "往前走五米，向右伸手", + "evidence": [ + { + "description": "alk five meters forward", + "timestamps": { + "start": "00:01:38", + "end": "00:01:39" + } + }, + { + "description": "reach out to the right", + "timestamps": { + "start": "00:01:40", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:01:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk forward four meters, then reach right.", + "B": "Walk forward six meters, then reach right.", + "C": "Walk forward five meters, then reach out your right hand.", + "D": "Walk forward five meters, then reach out your left hand to grab the black bag.", + "E": "Walk forward five meters, then stretch out both hands together to pick up the black bag." + }, + "answer_label": "C", + "answer_text": "Walk forward five meters, then reach out your right hand.", + "q_id": "01235_4_T2", + "correct_answer": "Walk forward five meters, then reach out your right hand.", + "original_options": { + "A": "Walk forward four meters, then reach right.", + "B": "Walk forward six meters, then reach right.", + "C": "Walk forward five meters, then reach out your right hand.", + "D": "Walk forward five meters, then reach out your left hand to grab the black bag.", + "E": "Walk forward five meters, then stretch out both hands together to pick up the black bag." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_4_T1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How do I get the black bag?", + "answer": "Walk forward five meters, then reach out your right hand.", + "question_chinese": "我要怎样拿到黑色的包？", + "answer_chinese": "往前走五米，向右伸手", + "evidence": [ + { + "description": "alk five meters forward", + "timestamps": { + "start": "00:01:38", + "end": "00:01:39" + } + }, + { + "description": "reach out to the right", + "timestamps": { + "start": "00:01:40", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:01:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk forward three meters, then reach right.", + "B": "Walk forward five meters, then reach out your right hand.", + "C": "Walk forward seven meters, then reach right.", + "D": "Walk forward five meters, then reach out your left hand to take the black bag.", + "E": "Walk forward five meters, then use both hands to reach out and grab the black bag." + }, + "answer_label": "B", + "answer_text": "Walk forward five meters, then reach out your right hand.", + "q_id": "01235_4_T1", + "correct_answer": "Walk forward five meters, then reach out your right hand.", + "original_options": { + "A": "Walk forward three meters, then reach right.", + "B": "Walk forward five meters, then reach out your right hand.", + "C": "Walk forward seven meters, then reach right.", + "D": "Walk forward five meters, then reach out your left hand to take the black bag.", + "E": "Walk forward five meters, then use both hands to reach out and grab the black bag." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_5_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What should I do to ride the horse?", + "answer": "First lift your leg and step on the stirrup, then get on the horse, and then pull the reins.", + "question_chinese": "我骑马要怎么做？", + "answer_chinese": "先抬腿踩脚蹬，然后上马，接着拉缰绳", + "evidence": [ + { + "description": "lift your leg to step on the stirrup", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + }, + { + "description": "get on the horse", + "timestamps": { + "start": "00:00:49", + "end": "00:00:50" + } + }, + { + "description": "pull the reins.", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Pull the reins first.", + "B": "Mount before stepping on the stirrup.", + "C": "First lift your leg and step on the stirrup, then get on the horse, and then pull the reins.", + "D": "First pull the reins, then step on the stirrup, and finally get on the horse.", + "E": "First step on the stirrup, then pull the reins, and only after that get on the horse." + }, + "answer_label": "C", + "answer_text": "First lift your leg and step on the stirrup, then get on the horse, and then pull the reins.", + "q_id": "01235_5_T1", + "correct_answer": "First lift your leg and step on the stirrup, then get on the horse, and then pull the reins.", + "original_options": { + "A": "Pull the reins first.", + "B": "Mount before stepping on the stirrup.", + "C": "First lift your leg and step on the stirrup, then get on the horse, and then pull the reins.", + "D": "First pull the reins, then step on the stirrup, and finally get on the horse.", + "E": "First step on the stirrup, then pull the reins, and only after that get on the horse." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 17 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_5_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What should I do to ride the horse?", + "answer": "First lift your leg and step on the stirrup in front of you, then mount the horse on your left, and then pull the reins.", + "question_chinese": "我骑马要怎么做？", + "answer_chinese": "先抬腿踩你前方脚蹬，然后上左手边的马，接着拉缰绳。", + "evidence": [ + { + "description": "lift your leg to step on the stirrup", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + }, + { + "description": "get on the horse", + "timestamps": { + "start": "00:00:49", + "end": "00:00:50" + } + }, + { + "description": "pull the reins.", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Mount from the right side.", + "B": "Step on the stirrup behind you.", + "C": "First pull the reins, then step on the stirrup in front of you, and then mount on your left.", + "D": "First lift your leg and step on the stirrup in front of you, then mount the horse on your left, and then pull the reins.", + "E": "First lift your leg and step on the stirrup in front of you, then pull the reins, then mount on your left." + }, + "answer_label": "D", + "answer_text": "First lift your leg and step on the stirrup in front of you, then mount the horse on your left, and then pull the reins.", + "q_id": "01235_5_T2", + "correct_answer": "First lift your leg and step on the stirrup in front of you, then mount the horse on your left, and then pull the reins.", + "original_options": { + "A": "Mount from the right side.", + "B": "Step on the stirrup behind you.", + "C": "First pull the reins, then step on the stirrup in front of you, and then mount on your left.", + "D": "First lift your leg and step on the stirrup in front of you, then mount the horse on your left, and then pull the reins.", + "E": "First lift your leg and step on the stirrup in front of you, then pull the reins, then mount on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 22 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the horse?", + "answer": "White.", + "question_chinese": "马是什么颜色的？", + "answer_chinese": "白色", + "evidence": [ + { + "description": "It is white", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Brown.", + "B": "White.", + "C": "Gray.", + "D": "Chestnut brown, with a darker mane and tail visible as you look at the horse.", + "E": "Pale blue, almost grayish under the light, making the coat seem tinted rather than white." + }, + "answer_label": "B", + "answer_text": "White.", + "q_id": "01235_1", + "correct_answer": "White.", + "original_options": { + "A": "Brown.", + "B": "White.", + "C": "Gray.", + "D": "Chestnut brown, with a darker mane and tail visible as you look at the horse.", + "E": "Pale blue, almost grayish under the light, making the coat seem tinted rather than white." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the horse?", + "answer": "White.", + "question_chinese": "马是什么颜色的？", + "answer_chinese": "白色", + "evidence": [ + { + "description": "It is white", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Black.", + "B": "Brown.", + "C": "Gray, with a lighter coat than the mane, so it doesn’t look fully white overall.", + "D": "Chestnut, more reddish-brown than dark, with a warmer tone instead of a white coat.", + "E": "White." + }, + "answer_label": "E", + "answer_text": "White.", + "q_id": "01235_2", + "correct_answer": "White.", + "original_options": { + "A": "Black.", + "B": "Brown.", + "C": "Gray, with a lighter coat than the mane, so it doesn’t look fully white overall.", + "D": "Chestnut, more reddish-brown than dark, with a warmer tone instead of a white coat.", + "E": "White." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 14 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_5", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the saddle?", + "answer": "Black.", + "question_chinese": "马鞍是什么颜色的？", + "answer_chinese": "黑色", + "evidence": [ + { + "description": "It is black", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Brown.", + "B": "Red.", + "C": "Black.", + "D": "Blue, standing out against the horse’s coat, making the saddle look bright rather than black.", + "E": "White, making it look lighter than the rest of the tack when you look closely." + }, + "answer_label": "C", + "answer_text": "Black.", + "q_id": "01235_5", + "correct_answer": "Black.", + "original_options": { + "A": "Brown.", + "B": "Red.", + "C": "Black.", + "D": "Blue, standing out against the horse’s coat, making the saddle look bright rather than black.", + "E": "White, making it look lighter than the rest of the tack when you look closely." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "01235", + "video_length": "00:02:08", + "qid": "01235_6", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the saddle?", + "answer": "Black.", + "question_chinese": "马鞍是什么颜色的？", + "answer_chinese": "黑色", + "evidence": [ + { + "description": "It is black", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Blue.", + "B": "Red.", + "C": "Brown, looking like worn leather rather than dark fabric, so it doesn’t appear black.", + "D": "Black.", + "E": "White, appearing bright and contrasting strongly with the horse, making the saddle look light." + }, + "answer_label": "D", + "answer_text": "Black.", + "q_id": "01235_6", + "correct_answer": "Black.", + "original_options": { + "A": "Blue.", + "B": "Red.", + "C": "Brown, looking like worn leather rather than dark fabric, so it doesn’t appear black.", + "D": "Black.", + "E": "White, appearing bright and contrasting strongly with the horse, making the saddle look light." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + } +] diff --git a/QAs/01251-mcq.json b/QAs/01251-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5925b5f04db001c23498891908749f7b482e292f --- /dev/null +++ b/QAs/01251-mcq.json @@ -0,0 +1,888 @@ +[ + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_7_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I pass by just now?", + "answer": "One person, and that person is to your right.", + "question_chinese": "我刚才经过了多少个人？", + "answer_chinese": "1个人，那个人在你右边。", + "evidence": [ + { + "description": "I passed s person", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero people passed you.", + "B": "Two people, both to your right.", + "C": "One person, and that person is to your right.", + "D": "Three people passed by, and all of them were positioned to your right as you went by.", + "E": "Four people went past you just now, and they were grouped on your right-hand side." + }, + "answer_label": "C", + "answer_text": "One person, and that person is to your right.", + "q_id": "01251_7_T1", + "correct_answer": "One person, and that person is to your right.", + "original_options": { + "A": "Zero people passed you.", + "B": "Two people, both to your right.", + "C": "One person, and that person is to your right.", + "D": "Three people passed by, and all of them were positioned to your right as you went by.", + "E": "Four people went past you just now, and they were grouped on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 17, + 15 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_7_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I pass by just now?", + "answer": "Five, and they’re all behind you.", + "question_chinese": "我刚才经过了多少个人？", + "answer_chinese": "5，都在我的后方。", + "evidence": [ + { + "description": "I passed a perrson", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + }, + { + "description": "I passed three people", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "I passed five people", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four, all behind you.", + "B": "Six, all behind you.", + "C": "Three people, and they’re all behind you as you move forward past them.", + "D": "Five, and they’re all behind you.", + "E": "Seven people, and they’re all behind you after you walk by them." + }, + "answer_label": "D", + "answer_text": "Five, and they’re all behind you.", + "q_id": "01251_7_T2", + "correct_answer": "Five, and they’re all behind you.", + "original_options": { + "A": "Four, all behind you.", + "B": "Six, all behind you.", + "C": "Three people, and they’re all behind you as you move forward past them.", + "D": "Five, and they’re all behind you.", + "E": "Seven people, and they’re all behind you after you walk by them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 13 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_4_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to the courier pickup station?", + "answer": "First walk straight for five meters, then turn left and walk two meters.", + "question_chinese": "我怎么去快递驿站？", + "answer_chinese": "先直走五米，然后左转走两米", + "evidence": [ + { + "description": "Go straight for five meters", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go straight four meters, then left two.", + "B": "Go straight five meters, then right two.", + "C": "Walk straight for six meters, turn left, and continue for two meters to reach the pickup station.", + "D": "First go straight for five meters, then turn left and walk three meters to the courier pickup station.", + "E": "First walk straight for five meters, then turn left and walk two meters." + }, + "answer_label": "E", + "answer_text": "First walk straight for five meters, then turn left and walk two meters.", + "q_id": "01251_4_T1", + "correct_answer": "First walk straight for five meters, then turn left and walk two meters.", + "original_options": { + "A": "Go straight four meters, then left two.", + "B": "Go straight five meters, then right two.", + "C": "Walk straight for six meters, turn left, and continue for two meters to reach the pickup station.", + "D": "First go straight for five meters, then turn left and walk three meters to the courier pickup station.", + "E": "First walk straight for five meters, then turn left and walk two meters." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_4_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to the courier pickup station?", + "answer": "First walk straight for five meters, then turn right and walk two meters.", + "question_chinese": "我怎么去快递驿站？", + "answer_chinese": "先直走五米，然后右转走两米", + "evidence": [ + { + "description": "Go straight for five meters", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "turn left and go two meters", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First walk straight for five meters, then turn right and walk two meters.", + "B": "Go straight four meters, then right two.", + "C": "Go straight five meters, then left two.", + "D": "Walk straight for six meters, then turn right and walk two meters to the courier pickup station.", + "E": "First walk straight for five meters, then turn right and continue for three meters to reach the station." + }, + "answer_label": "A", + "answer_text": "First walk straight for five meters, then turn right and walk two meters.", + "q_id": "01251_4_T2", + "correct_answer": "First walk straight for five meters, then turn right and walk two meters.", + "original_options": { + "A": "First walk straight for five meters, then turn right and walk two meters.", + "B": "Go straight four meters, then right two.", + "C": "Go straight five meters, then left two.", + "D": "Walk straight for six meters, then turn right and walk two meters to the courier pickup station.", + "E": "First walk straight for five meters, then turn right and continue for three meters to reach the station." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_8_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I ship a package?", + "answer": "Walk straight for five meters into the parcel station, then hand the package to the staff on your right.", + "question_chinese": "我怎么寄快递？", + "answer_chinese": "直走五米进驿站，把快递交给右边工作人员。", + "evidence": [ + { + "description": "Go straight for five meters", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk straight four meters, then hand it left.", + "B": "Walk straight six meters, then hand it right.", + "C": "Walk straight for five meters into the parcel station, then hand the package to the staff on your left.", + "D": "Walk straight for five meters into the parcel station, then give the package to the staff directly in front of you.", + "E": "Walk straight for five meters into the parcel station, then hand the package to the staff on your right." + }, + "answer_label": "E", + "answer_text": "Walk straight for five meters into the parcel station, then hand the package to the staff on your right.", + "q_id": "01251_8_T2", + "correct_answer": "Walk straight for five meters into the parcel station, then hand the package to the staff on your right.", + "original_options": { + "A": "Walk straight four meters, then hand it left.", + "B": "Walk straight six meters, then hand it right.", + "C": "Walk straight for five meters into the parcel station, then hand the package to the staff on your left.", + "D": "Walk straight for five meters into the parcel station, then give the package to the staff directly in front of you.", + "E": "Walk straight for five meters into the parcel station, then hand the package to the staff on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 21 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_8_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I ship a package?", + "answer": "Walk straight for five meters, then turn left into the parcel station and hand the package to the staff.", + "question_chinese": "我怎么寄快递？", + "answer_chinese": "直走五米，然后左转进驿站，把快递交给工作人员。", + "evidence": [ + { + "description": "Go straight for five meters", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "turn left into the post station", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + }, + { + "description": "hand the package to the staff.", + "timestamps": { + "start": "00:00:49", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk straight four meters, then turn left in.", + "B": "Walk straight five meters, then turn right in.", + "C": "Walk straight for six meters, then turn left into the parcel station and hand the package to the staff.", + "D": "Walk straight for five meters, then turn left into the parcel station and hand the package to the staff.", + "E": "Walk straight for five meters, turn left into the parcel station, and hand the package to the staff on your right." + }, + "answer_label": "D", + "answer_text": "Walk straight for five meters, then turn left into the parcel station and hand the package to the staff.", + "q_id": "01251_8_T1", + "correct_answer": "Walk straight for five meters, then turn left into the parcel station and hand the package to the staff.", + "original_options": { + "A": "Walk straight four meters, then turn left in.", + "B": "Walk straight five meters, then turn right in.", + "C": "Walk straight for six meters, then turn left into the parcel station and hand the package to the staff.", + "D": "Walk straight for five meters, then turn left into the parcel station and hand the package to the staff.", + "E": "Walk straight for five meters, turn left into the parcel station, and hand the package to the staff on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 21 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_5_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What have I held in my hand?", + "answer": "The cane and the paper bag are on your right-hand side.", + "question_chinese": "我的手上拿过什么？", + "answer_chinese": "拐杖和纸袋,在你的右手边。", + "evidence": [ + { + "description": "Cane", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "paper bag", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The cane and the paper bag are on your right-hand side.", + "B": "Cane and paper bag: left.", + "C": "Cane and paper bag: in front.", + "D": "The cane and the paper bag are behind you, not on your right-hand side.", + "E": "The cane and the paper bag are on your left-hand side, rather than to your right." + }, + "answer_label": "A", + "answer_text": "The cane and the paper bag are on your right-hand side.", + "q_id": "01251_5_T1", + "correct_answer": "The cane and the paper bag are on your right-hand side.", + "original_options": { + "A": "The cane and the paper bag are on your right-hand side.", + "B": "Cane and paper bag: left.", + "C": "Cane and paper bag: in front.", + "D": "The cane and the paper bag are behind you, not on your right-hand side.", + "E": "The cane and the paper bag are on your left-hand side, rather than to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 16 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_5_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What have I held in my hand?", + "answer": "The cane and the paper bag are to your right.", + "question_chinese": "我的手上拿过什么？", + "answer_chinese": "拐杖和纸袋，在右边。", + "evidence": [ + { + "description": "Cane", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "paper bag", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Cane and paper bag: left.", + "B": "The cane and the paper bag are to your right.", + "C": "Cane and paper bag: behind.", + "D": "The cane and the paper bag are in front of you, not to your right side.", + "E": "The cane and the paper bag are to your front-left, rather than directly to your right." + }, + "answer_label": "B", + "answer_text": "The cane and the paper bag are to your right.", + "q_id": "01251_5_T2", + "correct_answer": "The cane and the paper bag are to your right.", + "original_options": { + "A": "Cane and paper bag: left.", + "B": "The cane and the paper bag are to your right.", + "C": "Cane and paper bag: behind.", + "D": "The cane and the paper bag are in front of you, not to your right side.", + "E": "The cane and the paper bag are to your front-left, rather than directly to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_3_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see the woman holding in her hand?", + "answer": "A package delivery, to your right.", + "question_chinese": "我看到女人的手上拿着什么？", + "answer_chinese": "快递，在你的右方。", + "evidence": [ + { + "description": "She held a few packaegs", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A package delivery, to your left.", + "B": "A package delivery, behind you.", + "C": "A package delivery, in front of you as you face her, not to your right side.", + "D": "A package delivery, to your front-right, not directly at your side like it is to your right.", + "E": "A package delivery, to your right." + }, + "answer_label": "E", + "answer_text": "A package delivery, to your right.", + "q_id": "01251_3_T2", + "correct_answer": "A package delivery, to your right.", + "original_options": { + "A": "A package delivery, to your left.", + "B": "A package delivery, behind you.", + "C": "A package delivery, in front of you as you face her, not to your right side.", + "D": "A package delivery, to your front-right, not directly at your side like it is to your right.", + "E": "A package delivery, to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_3_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see the woman holding in her hand?", + "answer": "A package delivery, to your right.", + "question_chinese": "我看到女人的手上拿着什么？", + "answer_chinese": "快递，在你的右方", + "evidence": [ + { + "description": "She held a few packages", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A package delivery, to your left.", + "B": "A package delivery, in front of you.", + "C": "A package delivery, behind you, out of your direct view, rather than on your right side.", + "D": "A package delivery, to your right.", + "E": "A package delivery, to your front-right rather than beside you, so it isn’t directly to your right." + }, + "answer_label": "D", + "answer_text": "A package delivery, to your right.", + "q_id": "01251_3_T1", + "correct_answer": "A package delivery, to your right.", + "original_options": { + "A": "A package delivery, to your left.", + "B": "A package delivery, in front of you.", + "C": "A package delivery, behind you, out of your direct view, rather than on your right side.", + "D": "A package delivery, to your right.", + "E": "A package delivery, to your front-right rather than beside you, so it isn’t directly to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 18 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_2_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I see?", + "answer": "Four, to your front-left.", + "question_chinese": "我看见了多少个垃圾桶？", + "answer_chinese": "4个，在你的左前方。", + "evidence": [ + { + "description": "I saw four trash cans", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, to your front-left.", + "B": "Three, to your front-left.", + "C": "Four, to your front-left.", + "D": "Five, clustered together to your front-left near the curb line, not just four cans.", + "E": "Six, all visible to your front-left, forming a longer row than expected, not four." + }, + "answer_label": "C", + "answer_text": "Four, to your front-left.", + "q_id": "01251_2_T1", + "correct_answer": "Four, to your front-left.", + "original_options": { + "A": "Two, to your front-left.", + "B": "Three, to your front-left.", + "C": "Four, to your front-left.", + "D": "Five, clustered together to your front-left near the curb line, not just four cans.", + "E": "Six, all visible to your front-left, forming a longer row than expected, not four." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_2_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I see?", + "answer": "Four, to your back-left.", + "question_chinese": "我看见了多少个垃圾桶？", + "answer_chinese": "4个，在你的左后方。", + "evidence": [ + { + "description": "I saw four trash cans", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, to your back-left.", + "B": "Three, to your back-left.", + "C": "Five, positioned to your back-left as a small cluster behind you, not just four cans.", + "D": "Four, to your back-left.", + "E": "Six, lined up to your back-left, extending farther than the main group, not four." + }, + "answer_label": "D", + "answer_text": "Four, to your back-left.", + "q_id": "01251_2_T2", + "correct_answer": "Four, to your back-left.", + "original_options": { + "A": "Two, to your back-left.", + "B": "Three, to your back-left.", + "C": "Five, positioned to your back-left as a small cluster behind you, not just four cans.", + "D": "Four, to your back-left.", + "E": "Six, lined up to your back-left, extending farther than the main group, not four." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_6_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Was the cane in my right hand the whole time?", + "answer": "No, it’s in your left hand.", + "question_chinese": "拐杖一直在我的右手吗？", + "answer_chinese": "不，他在你的左手。", + "evidence": [ + { + "description": "left hand", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, it’s in your right hand the whole time.", + "B": "No, it’s in your left hand." + }, + "answer_label": "B", + "answer_text": "No, it’s in your left hand.", + "q_id": "01251_6_T2", + "correct_answer": "No, it’s in your left hand.", + "original_options": { + "A": "Yes, it’s in your right hand the whole time.", + "B": "No, it’s in your left hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_6_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Was the cane in my right hand the whole time?", + "answer": "No—right now it’s in your right hand.", + "question_chinese": "拐杖一直在我的右手吗？", + "answer_chinese": "不是的，目前它在你的右手。", + "evidence": [ + { + "description": "right hand", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "left hand", + "timestamps": { + "start": "00:00:52", + "end": "00:00:53" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—right now it’s in your right hand.", + "B": "Yes, it stayed in your right hand the whole time." + }, + "answer_label": "A", + "answer_text": "No—right now it’s in your right hand.", + "q_id": "01251_6_T1", + "correct_answer": "No—right now it’s in your right hand.", + "original_options": { + "A": "No—right now it’s in your right hand.", + "B": "Yes, it stayed in your right hand the whole time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the streetlight?", + "answer": "Straight ahead, three meters from me.", + "question_chinese": "路灯在哪里？", + "answer_chinese": "在正前方，距离我三米", + "evidence": [ + { + "description": "Straight ahead, three meters from me", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Straight ahead, two meters from me.", + "B": "Straight ahead, four meters from me.", + "C": "Straight ahead, three meters from me.", + "D": "Slightly to the right, about three meters away from where I'm standing.", + "E": "Slightly to the left, three meters from me, not directly in front." + }, + "answer_label": "C", + "answer_text": "Straight ahead, three meters from me.", + "q_id": "01251_1", + "correct_answer": "Straight ahead, three meters from me.", + "original_options": { + "A": "Straight ahead, two meters from me.", + "B": "Straight ahead, four meters from me.", + "C": "Straight ahead, three meters from me.", + "D": "Slightly to the right, about three meters away from where I'm standing.", + "E": "Slightly to the left, three meters from me, not directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 12 + ] + }, + { + "video_uid": "01251", + "video_length": "00:00:57", + "qid": "01251_2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the streetlight?", + "answer": "Straight ahead, two meters from me.", + "question_chinese": "路灯在哪里？", + "answer_chinese": "在正前方，距离我两米", + "evidence": [ + { + "description": "Straight ahead, two meters from me", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Straight ahead, two meters from me.", + "B": "Straight ahead, three meters from you.", + "C": "To the left, one meter from you.", + "D": "To the right, five meters from you, farther than arm’s reach, not straight ahead.", + "E": "Near the corner, four meters from you, not straight in front, and not two meters away." + }, + "answer_label": "A", + "answer_text": "Straight ahead, two meters from me.", + "q_id": "01251_2", + "correct_answer": "Straight ahead, two meters from me.", + "original_options": { + "A": "Straight ahead, two meters from me.", + "B": "Straight ahead, three meters from you.", + "C": "To the left, one meter from you.", + "D": "To the right, five meters from you, farther than arm’s reach, not straight ahead.", + "E": "Near the corner, four meters from you, not straight in front, and not two meters away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 16 + ] + } +] diff --git a/QAs/01257-mcq.json b/QAs/01257-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1f58e3c6b04223c4db3ea298a170d3a66c72a2b6 --- /dev/null +++ b/QAs/01257-mcq.json @@ -0,0 +1,870 @@ +[ + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_6_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see just now?", + "answer": "Four people, all in front of you.", + "question_chinese": "我刚才看见了多少个人？", + "answer_chinese": "4个人，都在你的前方。", + "evidence": [ + { + "description": "I saw three people", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I saw four people", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three people, all in front of you.", + "B": "Five people, all in front of you.", + "C": "Two people, both standing in front of you near the same spot.", + "D": "Four people, all in front of you.", + "E": "Six people, all gathered in front of you rather than spread out around you." + }, + "answer_label": "D", + "answer_text": "Four people, all in front of you.", + "q_id": "01257_6_T2", + "correct_answer": "Four people, all in front of you.", + "original_options": { + "A": "Three people, all in front of you.", + "B": "Five people, all in front of you.", + "C": "Two people, both standing in front of you near the same spot.", + "D": "Four people, all in front of you.", + "E": "Six people, all gathered in front of you rather than spread out around you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 14 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_6_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see just now?", + "answer": "4, to your right and behind you.", + "question_chinese": "我刚才看见了多少个人？", + "answer_chinese": "4，在你的右边和后边。", + "evidence": [ + { + "description": "I saw three people", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I saw four people", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three, to your right and behind you.", + "B": "Five, to your right and behind you.", + "C": "4, to your right and behind you.", + "D": "Two, both positioned to your right and slightly behind you, not in front of you.", + "E": "Six, all located to your right and behind you, not in front of you." + }, + "answer_label": "C", + "answer_text": "4, to your right and behind you.", + "q_id": "01257_6_T1", + "correct_answer": "4, to your right and behind you.", + "original_options": { + "A": "Three, to your right and behind you.", + "B": "Five, to your right and behind you.", + "C": "4, to your right and behind you.", + "D": "Two, both positioned to your right and slightly behind you, not in front of you.", + "E": "Six, all located to your right and behind you, not in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 14 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_5", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the piglet?", + "answer": "No, you need to move forward three meters.", + "question_chinese": "我摸得到小猪吗？", + "answer_chinese": "不可以，要往前走三米", + "evidence": [ + { + "description": "you need to move forward three meters.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can touch the piglet from where you are.", + "B": "No, you need to move forward three meters." + }, + "answer_label": "B", + "answer_text": "No, you need to move forward three meters.", + "q_id": "01257_5", + "correct_answer": "No, you need to move forward three meters.", + "original_options": { + "A": "Yes, you can touch the piglet from where you are.", + "B": "No, you need to move forward three meters." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_6", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the piglet?", + "answer": "No, you need to walk forward three meters.", + "question_chinese": "我摸得到小猪吗？", + "answer_chinese": "不可以，你要往前走三米", + "evidence": [ + { + "description": "you need to walk forward three meters.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you need to walk forward three meters.", + "B": "Yes, you can touch it without moving." + }, + "answer_label": "A", + "answer_text": "No, you need to walk forward three meters.", + "q_id": "01257_6", + "correct_answer": "No, you need to walk forward three meters.", + "original_options": { + "A": "No, you need to walk forward three meters.", + "B": "Yes, you can touch it without moving." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_4_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Was my right hand holding the baby bottle the whole time?", + "answer": "No—the baby bottle is to your right.", + "question_chinese": "我的右手一直拿着奶瓶吗？", + "answer_chinese": "不是的,奶瓶在你的右边。", + "evidence": [ + { + "description": "Empty hand.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—the baby bottle is in your right hand the whole time.", + "B": "No—the baby bottle is to your right." + }, + "answer_label": "B", + "answer_text": "No—the baby bottle is to your right.", + "q_id": "01257_4_T2", + "correct_answer": "No—the baby bottle is to your right.", + "original_options": { + "A": "Yes—the baby bottle is in your right hand the whole time.", + "B": "No—the baby bottle is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_4_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Was my right hand holding the baby bottle the whole time?", + "answer": "Yes—right now the baby bottle is in your right hand.", + "question_chinese": "我的右手一直拿着奶瓶吗？", + "answer_chinese": "是，目前奶瓶在你的右手。", + "evidence": [ + { + "description": "right hand", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + }, + { + "description": "both hand", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—right now the baby bottle is in your right hand.", + "B": "No—right now the baby bottle isn’t in your right hand." + }, + "answer_label": "A", + "answer_text": "Yes—right now the baby bottle is in your right hand.", + "q_id": "01257_4_T1", + "correct_answer": "Yes—right now the baby bottle is in your right hand.", + "original_options": { + "A": "Yes—right now the baby bottle is in your right hand.", + "B": "No—right now the baby bottle isn’t in your right hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_3_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many piglets did I see?", + "answer": "One piglet, in front of you.", + "question_chinese": "我看到了多少只小猪？", + "answer_chinese": "1只小猪，在你的前方。", + "evidence": [ + { + "description": "I saw one piglets", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One piglet, in front of you.", + "B": "Zero piglets, in front of you.", + "C": "Two piglets, in front of you.", + "D": "Three piglets, all visible in front of you at the same time.", + "E": "Four piglets, all in front of you rather than off to the sides." + }, + "answer_label": "A", + "answer_text": "One piglet, in front of you.", + "q_id": "01257_3_T2", + "correct_answer": "One piglet, in front of you.", + "original_options": { + "A": "One piglet, in front of you.", + "B": "Zero piglets, in front of you.", + "C": "Two piglets, in front of you.", + "D": "Three piglets, all visible in front of you at the same time.", + "E": "Four piglets, all in front of you rather than off to the sides." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 13 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_3_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many piglets did I see?", + "answer": "Five—the piglets are all in front of you.", + "question_chinese": "我看到了多少只小猪？", + "answer_chinese": "5，小猪都在你的前方。", + "evidence": [ + { + "description": "I saw one piglet", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "I saw five piglets", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four piglets, all in front.", + "B": "Six piglets, all in front.", + "C": "Three piglets are all in front of you, grouped together in the same area.", + "D": "Seven piglets are all in front of you, not spread out behind you.", + "E": "Five—the piglets are all in front of you." + }, + "answer_label": "E", + "answer_text": "Five—the piglets are all in front of you.", + "q_id": "01257_3_T1", + "correct_answer": "Five—the piglets are all in front of you.", + "original_options": { + "A": "Four piglets, all in front.", + "B": "Six piglets, all in front.", + "C": "Three piglets are all in front of you, grouped together in the same area.", + "D": "Seven piglets are all in front of you, not spread out behind you.", + "E": "Five—the piglets are all in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_1_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the sign that says “Piglets Nursing”?", + "answer": "It’s in the pigpen, to your front-left.", + "question_chinese": "我看到“小猪喂奶”的牌子在哪里？", + "answer_chinese": "在猪圈里，，在你的左前方", + "evidence": [ + { + "description": "In the pigpen", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the pigpen, front-right.", + "B": "In the pigpen, back-left.", + "C": "It’s in the pigpen, slightly behind you on the left side.", + "D": "It’s in the pigpen, to your front-left.", + "E": "You saw it inside the pigpen, off to your right rather than front-left." + }, + "answer_label": "D", + "answer_text": "It’s in the pigpen, to your front-left.", + "q_id": "01257_1_T2", + "correct_answer": "It’s in the pigpen, to your front-left.", + "original_options": { + "A": "In the pigpen, front-right.", + "B": "In the pigpen, back-left.", + "C": "It’s in the pigpen, slightly behind you on the left side.", + "D": "It’s in the pigpen, to your front-left.", + "E": "You saw it inside the pigpen, off to your right rather than front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_1_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the sign that says “Piglets Nursing”?", + "answer": "It’s in the pigpen, to your left.", + "question_chinese": "我看到“小猪喂奶”的牌子在哪里？", + "answer_chinese": "在猪圈里，在你的左边", + "evidence": [ + { + "description": "In the pigpen", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the pigpen, to your right.", + "B": "In the pigpen, front-left.", + "C": "It’s in the pigpen, to your left.", + "D": "It’s in the pigpen, behind you on the left side near the back gate.", + "E": "You saw the “Piglets Nursing” sign in the pigpen, positioned ahead and to your right." + }, + "answer_label": "C", + "answer_text": "It’s in the pigpen, to your left.", + "q_id": "01257_1_T1", + "correct_answer": "It’s in the pigpen, to your left.", + "original_options": { + "A": "In the pigpen, to your right.", + "B": "In the pigpen, front-left.", + "C": "It’s in the pigpen, to your left.", + "D": "It’s in the pigpen, behind you on the left side near the back gate.", + "E": "You saw the “Piglets Nursing” sign in the pigpen, positioned ahead and to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 15, + 15 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_5_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see?", + "answer": "Four, to your right.", + "question_chinese": "我看见了多少个人？", + "answer_chinese": "4，在你的右边。", + "evidence": [ + { + "description": "I saw three people", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I saw four people", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three, to your right.", + "B": "Five, to your right.", + "C": "Four, to your right.", + "D": "Two people were on your right side when you looked over toward the group.", + "E": "Six people were grouped on your right as you looked over toward them." + }, + "answer_label": "C", + "answer_text": "Four, to your right.", + "q_id": "01257_5_T2", + "correct_answer": "Four, to your right.", + "original_options": { + "A": "Three, to your right.", + "B": "Five, to your right.", + "C": "Four, to your right.", + "D": "Two people were on your right side when you looked over toward the group.", + "E": "Six people were grouped on your right as you looked over toward them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 13 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_5_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see?", + "answer": "4.", + "question_chinese": "我看见了多少个人？", + "answer_chinese": "4", + "evidence": [ + { + "description": "I saw three people", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I saw four people", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "4.", + "B": "3.", + "C": "5.", + "D": "Only 2 people were visible in total when you looked around the area.", + "E": "There were 6 people in total, not four, spread out around you." + }, + "answer_label": "A", + "answer_text": "4.", + "q_id": "01257_5_T1", + "correct_answer": "4.", + "original_options": { + "A": "4.", + "B": "3.", + "C": "5.", + "D": "Only 2 people were visible in total when you looked around the area.", + "E": "There were 6 people in total, not four, spread out around you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_7_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I feed the piglets milk?", + "answer": "Pick up the baby bottle, walk 3 meters forward, and feed the piglet milk.", + "question_chinese": "我要怎么小猪喂奶？", + "answer_chinese": "拿奶瓶，往前走三米，喂小猪喝奶", + "evidence": [ + { + "description": "Pick up the milk bottle", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk 2 meters forward with the bottle.", + "B": "Walk 4 meters forward with the bottle.", + "C": "Pick up the baby bottle, walk 1 meter forward, then feed the piglet milk.", + "D": "Pick up the baby bottle, walk 5 meters forward, and then feed the piglet milk.", + "E": "Pick up the baby bottle, walk 3 meters forward, and feed the piglet milk." + }, + "answer_label": "E", + "answer_text": "Pick up the baby bottle, walk 3 meters forward, and feed the piglet milk.", + "q_id": "01257_7_T2", + "correct_answer": "Pick up the baby bottle, walk 3 meters forward, and feed the piglet milk.", + "original_options": { + "A": "Walk 2 meters forward with the bottle.", + "B": "Walk 4 meters forward with the bottle.", + "C": "Pick up the baby bottle, walk 1 meter forward, then feed the piglet milk.", + "D": "Pick up the baby bottle, walk 5 meters forward, and then feed the piglet milk.", + "E": "Pick up the baby bottle, walk 3 meters forward, and feed the piglet milk." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 15 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_7_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I feed the piglets milk?", + "answer": "Just hold the baby bottle; the piglet is in front of you.", + "question_chinese": "我要怎么小猪喂奶？", + "answer_chinese": "持有奶瓶即可，小猪在你的前方。", + "evidence": [ + { + "description": "Pick up the milk bottle", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + }, + { + "description": "walk forward three meters", + "timestamps": { + "start": "00:00:21", + "end": "00:00:22" + } + }, + { + "description": "feed the piglet", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Hold the bottle; it’s behind you.", + "B": "Hold the bottle; it’s to your left.", + "C": "Just hold the baby bottle; the piglet is off to your right side, not in front.", + "D": "Just hold the baby bottle; the piglet is in front of you.", + "E": "Just hold the baby bottle in place; the piglet is behind you so you need to turn around." + }, + "answer_label": "D", + "answer_text": "Just hold the baby bottle; the piglet is in front of you.", + "q_id": "01257_7_T1", + "correct_answer": "Just hold the baby bottle; the piglet is in front of you.", + "original_options": { + "A": "Hold the bottle; it’s behind you.", + "B": "Hold the bottle; it’s to your left.", + "C": "Just hold the baby bottle; the piglet is off to your right side, not in front.", + "D": "Just hold the baby bottle; the piglet is in front of you.", + "E": "Just hold the baby bottle in place; the piglet is behind you so you need to turn around." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 18 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_3", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is a pig?", + "answer": "Black and pink.", + "question_chinese": "猪是什么颜色的？", + "answer_chinese": "黑色和粉色", + "evidence": [ + { + "description": "Black and pink", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Black and pink.", + "B": "Pink and white.", + "C": "Black and gray.", + "D": "It’s mostly brown with some pink areas on its body and legs.", + "E": "The pig is pink overall, with darker spots rather than black and pink coloring." + }, + "answer_label": "A", + "answer_text": "Black and pink.", + "q_id": "01257_3", + "correct_answer": "Black and pink.", + "original_options": { + "A": "Black and pink.", + "B": "Pink and white.", + "C": "Black and gray.", + "D": "It’s mostly brown with some pink areas on its body and legs.", + "E": "The pig is pink overall, with darker spots rather than black and pink coloring." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 14 + ] + }, + { + "video_uid": "01257", + "video_length": "00:01:02", + "qid": "01257_4", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is a pig?", + "answer": "Black and pink.", + "question_chinese": "猪是什么颜色的？", + "answer_chinese": "黑色和粉色", + "evidence": [ + { + "description": "Black and pink", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Black and white.", + "B": "Pink and white.", + "C": "It’s white and gray, not the black-and-pink coloring you saw.", + "D": "Black and pink.", + "E": "The pig is red and black in color, rather than black and pink." + }, + "answer_label": "D", + "answer_text": "Black and pink.", + "q_id": "01257_4", + "correct_answer": "Black and pink.", + "original_options": { + "A": "Black and white.", + "B": "Pink and white.", + "C": "It’s white and gray, not the black-and-pink coloring you saw.", + "D": "Black and pink.", + "E": "The pig is red and black in color, rather than black and pink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 13 + ] + } +] diff --git a/QAs/01285-mcq.json b/QAs/01285-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7cc3d9fc35602f28bad503b367714e201a0e16f2 --- /dev/null +++ b/QAs/01285-mcq.json @@ -0,0 +1,884 @@ +[ + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_1_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the rabbit plush relative to me?", + "answer": "On your right-hand side.", + "question_chinese": "兔子玩偶在我的哪里？", + "answer_chinese": "在我的右手边", + "evidence": [ + { + "description": "It was on my right hand", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:01:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your left-hand side.", + "B": "Directly in front of you.", + "C": "It was behind you, slightly off to your left when you looked around.", + "D": "It sat to your back-left, so you would need to turn left and back to face it.", + "E": "On your right-hand side." + }, + "answer_label": "E", + "answer_text": "On your right-hand side.", + "q_id": "01285_1_T2", + "correct_answer": "On your right-hand side.", + "original_options": { + "A": "On your left-hand side.", + "B": "Directly in front of you.", + "C": "It was behind you, slightly off to your left when you looked around.", + "D": "It sat to your back-left, so you would need to turn left and back to face it.", + "E": "On your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 17 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_1_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the rabbit plush relative to me?", + "answer": "To your right.", + "question_chinese": "兔子玩偶在我的哪里？", + "answer_chinese": "在我的右方", + "evidence": [ + { + "description": "It was on my right hand", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:02:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "Straight ahead of you.", + "C": "It was behind you, so you would have to turn around to see it.", + "D": "To your right.", + "E": "It was to your back-left, positioned behind you and a little to the left." + }, + "answer_label": "D", + "answer_text": "To your right.", + "q_id": "01285_1_T1", + "correct_answer": "To your right.", + "original_options": { + "A": "To your left.", + "B": "Straight ahead of you.", + "C": "It was behind you, so you would have to turn around to see it.", + "D": "To your right.", + "E": "It was to your back-left, positioned behind you and a little to the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 14 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_4_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Have I been standing to the boy’s left the whole time?", + "answer": "Yes. The boy is to your right, and you are to his left.", + "question_chinese": "我一直站在男生的左边吗？", + "answer_chinese": "是的，男生在我的右方，我在男生左边。", + "evidence": [ + { + "description": "on his left", + "timestamps": { + "start": "00:00:21", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. The boy is to your right, and you are to his left.", + "B": "No. The boy is to your left, and you are to his right." + }, + "answer_label": "A", + "answer_text": "Yes. The boy is to your right, and you are to his left.", + "q_id": "01285_4_T2", + "correct_answer": "Yes. The boy is to your right, and you are to his left.", + "original_options": { + "A": "Yes. The boy is to your right, and you are to his left.", + "B": "No. The boy is to your left, and you are to his right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_4_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Have I been standing to the boy’s left the whole time?", + "answer": "No. The boy is to your right, and you are in front of him.", + "question_chinese": "我一直站在男生的左边吗？", + "answer_chinese": "不是，男生在我的右方，我在男生的前方。", + "evidence": [ + { + "description": "on his left", + "timestamps": { + "start": "00:00:21", + "end": "00:00:22" + } + }, + { + "description": "in front of him", + "timestamps": { + "start": "00:02:20", + "end": "00:02:21" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. The boy is to your right, and you are behind him.", + "B": "No. The boy is to your right, and you are in front of him." + }, + "answer_label": "B", + "answer_text": "No. The boy is to your right, and you are in front of him.", + "q_id": "01285_4_T1", + "correct_answer": "No. The boy is to your right, and you are in front of him.", + "original_options": { + "A": "Yes. The boy is to your right, and you are behind him.", + "B": "No. The boy is to your right, and you are in front of him." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_8_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sculptures have I touched?", + "answer": "One, to your left.", + "question_chinese": "我摸了多少个雕塑？", + "answer_chinese": "1个，在你的左边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, to your left.", + "B": "Zero, none touched.", + "C": "Two, both on your left.", + "D": "Three, all to your left, touched at separate moments as you moved along the display.", + "E": "Four, all on your left, with your hand contacting each one briefly during the interaction." + }, + "answer_label": "A", + "answer_text": "One, to your left.", + "q_id": "01285_8_T1", + "correct_answer": "One, to your left.", + "original_options": { + "A": "One, to your left.", + "B": "Zero, none touched.", + "C": "Two, both on your left.", + "D": "Three, all to your left, touched at separate moments as you moved along the display.", + "E": "Four, all on your left, with your hand contacting each one briefly during the interaction." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 5, + 15, + 15 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_8_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sculptures have I touched?", + "answer": "Three, in front of you and to your back-right.", + "question_chinese": "我摸了多少个雕塑？", + "answer_chinese": "3个，在你的前方和右后方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:57", + "end": "00:00:58" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:01:46", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, in front of you.", + "B": "Three, in front of you and to your back-right.", + "C": "Two, to your back-right.", + "D": "Four, with some in front of you and the rest located to your back-right.", + "E": "Five, spread between directly in front of you and clustered toward your back-right side." + }, + "answer_label": "B", + "answer_text": "Three, in front of you and to your back-right.", + "q_id": "01285_8_T2", + "correct_answer": "Three, in front of you and to your back-right.", + "original_options": { + "A": "One, in front of you.", + "B": "Three, in front of you and to your back-right.", + "C": "Two, to your back-right.", + "D": "Four, with some in front of you and the rest located to your back-right.", + "E": "Five, spread between directly in front of you and clustered toward your back-right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 14 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_7_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Which parts of the sculpture did I touch?", + "answer": "The chest, arms, and waist are in front of you and to your back-right, respectively.", + "question_chinese": "我摸了雕塑的哪些部位？", + "answer_chinese": "胸膛，手臂和腰，分别在你的前方和右后方。", + "evidence": [ + { + "description": "Chest", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "arms", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "waist", + "timestamps": { + "start": "00:00:46", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:00:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Chest, arms, and legs.", + "B": "Head, arms, and waist.", + "C": "The chest, arms, and waist are in front of you and to your back-left, respectively.", + "D": "The chest, arms, and waist are behind you and to your back-right, respectively.", + "E": "The chest, arms, and waist are in front of you and to your back-right, respectively." + }, + "answer_label": "E", + "answer_text": "The chest, arms, and waist are in front of you and to your back-right, respectively.", + "q_id": "01285_7_T1", + "correct_answer": "The chest, arms, and waist are in front of you and to your back-right, respectively.", + "original_options": { + "A": "Chest, arms, and legs.", + "B": "Head, arms, and waist.", + "C": "The chest, arms, and waist are in front of you and to your back-left, respectively.", + "D": "The chest, arms, and waist are behind you and to your back-right, respectively.", + "E": "The chest, arms, and waist are in front of you and to your back-right, respectively." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 13 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_7_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Which parts of the sculpture did I touch?", + "answer": "Chest, arms, and waist. They are in front of you and to your back-right.", + "question_chinese": "我摸了雕塑的哪些部位？", + "answer_chinese": "胸膛，手臂和腰。在你的前方和右后方。", + "evidence": [ + { + "description": "Chest", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "arms", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "waist", + "timestamps": { + "start": "00:00:46", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Chest, arms, and waist. They are in front of you and to your back-right.", + "B": "Chest, arms, and legs.", + "C": "Head, arms, and waist.", + "D": "Chest, arms, and waist. They are in front of you and to your back-left.", + "E": "Chest, arms, and waist. They are behind you and to your front-right." + }, + "answer_label": "A", + "answer_text": "Chest, arms, and waist. They are in front of you and to your back-right.", + "q_id": "01285_7_T2", + "correct_answer": "Chest, arms, and waist. They are in front of you and to your back-right.", + "original_options": { + "A": "Chest, arms, and waist. They are in front of you and to your back-right.", + "B": "Chest, arms, and legs.", + "C": "Head, arms, and waist.", + "D": "Chest, arms, and waist. They are in front of you and to your back-left.", + "E": "Chest, arms, and waist. They are behind you and to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 12 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_5_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Did I see the boy keep touching the statue with his left hand the whole time?", + "answer": "Yes, the boy is to your right.", + "question_chinese": "我看到男生一直用左手摸雕塑吗？", + "answer_chinese": "是，男生在我的右方。", + "evidence": [ + { + "description": "left hand", + "timestamps": { + "start": "00:00:31", + "end": "00:00:32" + } + }, + { + "description": "both hand", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:58", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, the boy is to your right.", + "B": "No, the boy used both hands, and he is to your right." + }, + "answer_label": "A", + "answer_text": "Yes, the boy is to your right.", + "q_id": "01285_5_T1", + "correct_answer": "Yes, the boy is to your right.", + "original_options": { + "A": "Yes, the boy is to your right.", + "B": "No, the boy used both hands, and he is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_5_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Did I see the boy keep touching the statue with his left hand the whole time?", + "answer": "No. The boy used both hands, and he is to your right.", + "question_chinese": "我看到男生一直用左手摸雕塑吗？", + "answer_chinese": "不是，男生用了双手，而他在你的右方。", + "evidence": [ + { + "description": "left hand", + "timestamps": { + "start": "00:00:31", + "end": "00:00:32" + } + }, + { + "description": "both hand", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:01:26", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, he kept touching the statue with his left hand the whole time, and he is to your right.", + "B": "No. The boy used both hands, and he is to your right." + }, + "answer_label": "B", + "answer_text": "No. The boy used both hands, and he is to your right.", + "q_id": "01285_5_T2", + "correct_answer": "No. The boy used both hands, and he is to your right.", + "original_options": { + "A": "Yes, he kept touching the statue with his left hand the whole time, and he is to your right.", + "B": "No. The boy used both hands, and he is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 19 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_3_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see in the second statue’s hand?", + "answer": "A basketball, to your back-right.", + "question_chinese": "我看到第二个雕塑手上是什么？", + "answer_chinese": "篮球，在你的右后方。", + "evidence": [ + { + "description": "Basketball was on its hand", + "timestamps": { + "start": "00:01:07", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A basketball, to your right.", + "B": "A basketball, to your back-right.", + "C": "A basketball, to your back-left.", + "D": "A basketball in its hand, positioned to your front-right as you look around.", + "E": "A sculpture held in the second statue’s hand, located to your front-left from where you stood." + }, + "answer_label": "B", + "answer_text": "A basketball, to your back-right.", + "q_id": "01285_3_T2", + "correct_answer": "A basketball, to your back-right.", + "original_options": { + "A": "A basketball, to your right.", + "B": "A basketball, to your back-right.", + "C": "A basketball, to your back-left.", + "D": "A basketball in its hand, positioned to your front-right as you look around.", + "E": "A sculpture held in the second statue’s hand, located to your front-left from where you stood." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 17 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_3_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see in the second statue’s hand?", + "answer": "A basketball, to your right.", + "question_chinese": "我看到第二个雕塑手上是什么？", + "answer_chinese": "篮球,在我的右方。", + "evidence": [ + { + "description": "Basketball was on its hand", + "timestamps": { + "start": "00:01:06", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A basketball, to your right.", + "B": "A basketball, to your left.", + "C": "A basketball, to your back-right.", + "D": "A basketball in its hand, sitting to your front-right relative to your viewpoint.", + "E": "A sculpture in the second statue’s hand, placed to your front-left as you faced forward." + }, + "answer_label": "A", + "answer_text": "A basketball, to your right.", + "q_id": "01285_3_T1", + "correct_answer": "A basketball, to your right.", + "original_options": { + "A": "A basketball, to your right.", + "B": "A basketball, to your left.", + "C": "A basketball, to your back-right.", + "D": "A basketball in its hand, sitting to your front-right relative to your viewpoint.", + "E": "A sculpture in the second statue’s hand, placed to your front-left as you faced forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 16 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_2_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is in front of me?", + "answer": "A sculpture, to your front-left.", + "question_chinese": "我面前的是什么？", + "answer_chinese": "雕塑，在你的左前方。", + "evidence": [ + { + "description": "Sculpture was in front of me", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A sculpture, to your front-left.", + "B": "A sculpture, to your front-right.", + "C": "A sculpture, to your back-left.", + "D": "A sculpture and a person, directly in front of you as you look ahead.", + "E": "A basketball behind you, positioned to your back-right from where you are standing." + }, + "answer_label": "A", + "answer_text": "A sculpture, to your front-left.", + "q_id": "01285_2_T2", + "correct_answer": "A sculpture, to your front-left.", + "original_options": { + "A": "A sculpture, to your front-left.", + "B": "A sculpture, to your front-right.", + "C": "A sculpture, to your back-left.", + "D": "A sculpture and a person, directly in front of you as you look ahead.", + "E": "A basketball behind you, positioned to your back-right from where you are standing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_2_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is in front of me?", + "answer": "A sculpture and a person, in front of you.", + "question_chinese": "我面前的是什么？", + "answer_chinese": "雕塑和人，在你的前方。", + "evidence": [ + { + "description": "Sculpture was in front of me", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A sculpture, to your front-left.", + "B": "A basketball, to your back-right.", + "C": "A sculpture and a person, but they are behind you rather than in front.", + "D": "A sculpture and a person located in back of you, not directly ahead in your view.", + "E": "A sculpture and a person, in front of you." + }, + "answer_label": "E", + "answer_text": "A sculpture and a person, in front of you.", + "q_id": "01285_2_T1", + "correct_answer": "A sculpture and a person, in front of you.", + "original_options": { + "A": "A sculpture, to your front-left.", + "B": "A basketball, to your back-right.", + "C": "A sculpture and a person, but they are behind you rather than in front.", + "D": "A sculpture and a person located in back of you, not directly ahead in your view.", + "E": "A sculpture and a person, in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_6_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the man in black relative to me?", + "answer": "First, it was to your right.", + "question_chinese": "黑衣男人在我的哪里？", + "answer_chinese": "先在你的右边", + "evidence": [ + { + "description": "on my right", + "timestamps": { + "start": "00:00:29", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First, it was to your left.", + "B": "First, it was in front of you.", + "C": "First, it was behind you before you turned and faced each other briefly.", + "D": "First, it was to your left, then moved in front as you changed where you were looking.", + "E": "First, it was to your right." + }, + "answer_label": "E", + "answer_text": "First, it was to your right.", + "q_id": "01285_6_T2", + "correct_answer": "First, it was to your right.", + "original_options": { + "A": "First, it was to your left.", + "B": "First, it was in front of you.", + "C": "First, it was behind you before you turned and faced each other briefly.", + "D": "First, it was to your left, then moved in front as you changed where you were looking.", + "E": "First, it was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 17 + ] + }, + { + "video_uid": "01285", + "video_length": "00:02:36", + "qid": "01285_6_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the man in black relative to me?", + "answer": "First it was to your right, then it was to your left.", + "question_chinese": "黑衣男人在我的哪里？", + "answer_chinese": "先在你的右边，后在你的左边", + "evidence": [ + { + "description": "on my right", + "timestamps": { + "start": "00:00:29", + "end": "00:00:30" + } + }, + { + "description": "on my right", + "timestamps": { + "start": "00:01:40", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First it was to your left.", + "B": "First it was to your right again.", + "C": "First it was to your left, then it stayed on your left without switching sides.", + "D": "First it was to your right, then it was to your left.", + "E": "First it was to your left, then it moved to your right instead of ending up on your left." + }, + "answer_label": "D", + "answer_text": "First it was to your right, then it was to your left.", + "q_id": "01285_6_T1", + "correct_answer": "First it was to your right, then it was to your left.", + "original_options": { + "A": "First it was to your left.", + "B": "First it was to your right again.", + "C": "First it was to your left, then it stayed on your left without switching sides.", + "D": "First it was to your right, then it was to your left.", + "E": "First it was to your left, then it moved to your right instead of ending up on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 19 + ] + } +] diff --git a/QAs/01286-mcq.json b/QAs/01286-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fc61c6977c99ea863fa9a3e7db004fda1d1eaa0a --- /dev/null +++ b/QAs/01286-mcq.json @@ -0,0 +1,1244 @@ +[ + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_15", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Have I always been behind the boy?", + "answer": "No.", + "question_chinese": "我一直在男生的后方吗？", + "answer_chinese": "不是", + "evidence": [ + { + "description": "Left", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "Behind", + "timestamps": { + "start": "00:00:42", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "01286_15", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_16", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Have I always been behind the boy?", + "answer": "No.", + "question_chinese": "我一直在男生的后方吗？", + "answer_chinese": "不是", + "evidence": [ + { + "description": "Left", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "Behind", + "timestamps": { + "start": "00:00:42", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01286_16", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_6_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I touch the statue's face just now?", + "answer": "You touched the face of the sculpture in front of you once.", + "question_chinese": "我刚才摸了多少次雕像的脸？", + "answer_chinese": "你摸了一次面前的雕塑的脸", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:00:37", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Twice.", + "B": "Three times.", + "C": "You touched the face of the sculpture in front of you twice in quick succession.", + "D": "You didn’t touch the statue’s face at all, keeping your hands away from it the whole time.", + "E": "You touched the face of the sculpture in front of you once." + }, + "answer_label": "E", + "answer_text": "You touched the face of the sculpture in front of you once.", + "q_id": "01286_6_T1", + "correct_answer": "You touched the face of the sculpture in front of you once.", + "original_options": { + "A": "Twice.", + "B": "Three times.", + "C": "You touched the face of the sculpture in front of you twice in quick succession.", + "D": "You didn’t touch the statue’s face at all, keeping your hands away from it the whole time.", + "E": "You touched the face of the sculpture in front of you once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 19 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_13", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I just touch the statue's face?", + "answer": "two.", + "question_chinese": "我刚才摸了多少次雕像的脸？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:00:38", + "end": "00:00:39" + } + }, + { + "description": "the second time", + "timestamps": { + "start": "00:00:40", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "one.", + "B": "two.", + "C": "three.", + "D": "You touched the statue’s face four times in total, returning to it repeatedly after pulling your hand away.", + "E": "You touched the statue’s face five times, briefly tapping it several times before stopping." + }, + "answer_label": "B", + "answer_text": "two.", + "q_id": "01286_13", + "correct_answer": "two.", + "original_options": { + "A": "one.", + "B": "two.", + "C": "three.", + "D": "You touched the statue’s face four times in total, returning to it repeatedly after pulling your hand away.", + "E": "You touched the statue’s face five times, briefly tapping it several times before stopping." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 19, + 15 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_6_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I touch the statue's face just now?", + "answer": "You touched the face of the sculpture in front of you twice.", + "question_chinese": "我刚才摸了多少次雕像的脸？", + "answer_chinese": "你摸了两次面前的雕塑的脸", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:00:37", + "end": "00:00:38" + } + }, + { + "description": "the second time", + "timestamps": { + "start": "00:00:40", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You touched the face of the sculpture in front of you twice.", + "B": "Once.", + "C": "Three times.", + "D": "You touched the face of the sculpture in front of you only once before pulling your hand back.", + "E": "You didn’t touch the statue’s face at all, and your hands never made contact with it." + }, + "answer_label": "A", + "answer_text": "You touched the face of the sculpture in front of you twice.", + "q_id": "01286_6_T2", + "correct_answer": "You touched the face of the sculpture in front of you twice.", + "original_options": { + "A": "You touched the face of the sculpture in front of you twice.", + "B": "Once.", + "C": "Three times.", + "D": "You touched the face of the sculpture in front of you only once before pulling your hand back.", + "E": "You didn’t touch the statue’s face at all, and your hands never made contact with it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 18, + 18 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_5", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the boy's hand?", + "answer": "Yes.", + "question_chinese": "我可以碰到男生的手吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "I can touch it if I move my hand a little forward.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "01286_5", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_2_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the boy's hand?", + "answer": "Yes, you can touch the boy's hand to your right.", + "question_chinese": "我可以碰到男生的手吗？", + "answer_chinese": "是，你可以碰到右边男生的手", + "evidence": [ + { + "description": "I can touch it if I move my hand a little forward.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can touch the boy's hand to your right.", + "B": "No, you can't touch the boy's hand to your right." + }, + "answer_label": "A", + "answer_text": "Yes, you can touch the boy's hand to your right.", + "q_id": "01286_2_T1", + "correct_answer": "Yes, you can touch the boy's hand to your right.", + "original_options": { + "A": "Yes, you can touch the boy's hand to your right.", + "B": "No, you can't touch the boy's hand to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_2_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the boy's hand?", + "answer": "No, you can't touch the boy's hand to your right.", + "question_chinese": "我可以碰到男生的手吗？", + "answer_chinese": "不，你不可以碰到右边男生的手", + "evidence": [ + { + "description": "I stood up and walked forward, avoiding touching the boy's hand.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can touch the boy's hand to your right.", + "B": "No, you can't touch the boy's hand to your right." + }, + "answer_label": "B", + "answer_text": "No, you can't touch the boy's hand to your right.", + "q_id": "01286_2_T2", + "correct_answer": "No, you can't touch the boy's hand to your right.", + "original_options": { + "A": "Yes, you can touch the boy's hand to your right.", + "B": "No, you can't touch the boy's hand to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_6", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the boy's hand?", + "answer": "Yes.", + "question_chinese": "我可以碰到男生的手吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "I can touch it if I move my hand a little forward.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "01286_6", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_7_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I relative to the boy?", + "answer": "You are to the left of the boy.", + "question_chinese": "我在男生的哪个方位", + "answer_chinese": "你在男生的左边", + "evidence": [ + { + "description": "Left", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are in front of the boy.", + "B": "You are behind the boy.", + "C": "You are to the right of the boy, standing slightly off to his side as you face him.", + "D": "You are to the left of the boy.", + "E": "You are directly in front of the boy, positioned so he is behind you when you look ahead." + }, + "answer_label": "D", + "answer_text": "You are to the left of the boy.", + "q_id": "01286_7_T2", + "correct_answer": "You are to the left of the boy.", + "original_options": { + "A": "You are in front of the boy.", + "B": "You are behind the boy.", + "C": "You are to the right of the boy, standing slightly off to his side as you face him.", + "D": "You are to the left of the boy.", + "E": "You are directly in front of the boy, positioned so he is behind you when you look ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 18 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_7_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I relative to the boy?", + "answer": "You are behind the boy.", + "question_chinese": "我在男生的哪个方位", + "answer_chinese": "你在男生的后面", + "evidence": [ + { + "description": "Behind", + "timestamps": { + "start": "00:00:42", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:00:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are in front of the boy.", + "B": "You are to the left of the boy.", + "C": "You are to the right of the boy, standing alongside him rather than directly behind him.", + "D": "You are behind the boy.", + "E": "You are in front of the boy, facing him so he is behind you from your viewpoint." + }, + "answer_label": "D", + "answer_text": "You are behind the boy.", + "q_id": "01286_7_T1", + "correct_answer": "You are behind the boy.", + "original_options": { + "A": "You are in front of the boy.", + "B": "You are to the left of the boy.", + "C": "You are to the right of the boy, standing alongside him rather than directly behind him.", + "D": "You are behind the boy.", + "E": "You are in front of the boy, facing him so he is behind you from your viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 17 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_1_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which sculptures have I touched?", + "answer": "You touched the small dog sculpture in front of you.", + "question_chinese": "我摸过什么雕塑？", + "answer_chinese": "你摸过面前的小狗雕塑", + "evidence": [ + { + "description": "puppy sculpture", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You touched the dog sculpture behind you.", + "B": "You touched the dog sculpture to your left.", + "C": "You touched the small dog sculpture to your right, reaching out from your side to make contact.", + "D": "You touched the small dog sculpture in front of you.", + "E": "You touched the small dog sculpture behind you, turning slightly before placing your hand on it." + }, + "answer_label": "D", + "answer_text": "You touched the small dog sculpture in front of you.", + "q_id": "01286_1_T1", + "correct_answer": "You touched the small dog sculpture in front of you.", + "original_options": { + "A": "You touched the dog sculpture behind you.", + "B": "You touched the dog sculpture to your left.", + "C": "You touched the small dog sculpture to your right, reaching out from your side to make contact.", + "D": "You touched the small dog sculpture in front of you.", + "E": "You touched the small dog sculpture behind you, turning slightly before placing your hand on it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 16 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_1_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which sculptures have I touched?", + "answer": "You touched the small dog sculpture and the celebrity sculpture in front of you.", + "question_chinese": "我摸过什么雕塑？", + "answer_chinese": "你摸过面前的小狗雕塑和名人雕塑", + "evidence": [ + { + "description": "puppy sculpture", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "celebrity sculpture", + "timestamps": { + "start": "00:00:37", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You touched both sculptures behind you.", + "B": "You touched both sculptures to your left.", + "C": "You touched the small dog and celebrity sculptures to your right, reaching across to feel them.", + "D": "You touched the small dog and celebrity sculptures above you, lifting your hand to make contact.", + "E": "You touched the small dog sculpture and the celebrity sculpture in front of you." + }, + "answer_label": "E", + "answer_text": "You touched the small dog sculpture and the celebrity sculpture in front of you.", + "q_id": "01286_1_T2", + "correct_answer": "You touched the small dog sculpture and the celebrity sculpture in front of you.", + "original_options": { + "A": "You touched both sculptures behind you.", + "B": "You touched both sculptures to your left.", + "C": "You touched the small dog and celebrity sculptures to your right, reaching across to feel them.", + "D": "You touched the small dog and celebrity sculptures above you, lifting your hand to make contact.", + "E": "You touched the small dog sculpture and the celebrity sculpture in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 16 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_7", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is my right hand holding?", + "answer": "Walking stick.", + "question_chinese": "我的右手拿着什么？", + "answer_chinese": "拐杖", + "evidence": [ + { + "description": "Cane", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Umbrella.", + "B": "Baseball bat.", + "C": "You are holding a tennis racket in your right hand while standing near the display.", + "D": "Walking stick.", + "E": "You are holding an umbrella in your right hand while standing still." + }, + "answer_label": "D", + "answer_text": "Walking stick.", + "q_id": "01286_7", + "correct_answer": "Walking stick.", + "original_options": { + "A": "Umbrella.", + "B": "Baseball bat.", + "C": "You are holding a tennis racket in your right hand while standing near the display.", + "D": "Walking stick.", + "E": "You are holding an umbrella in your right hand while standing still." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 12 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_8", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is my right hand holding?", + "answer": "Walking stick.", + "question_chinese": "我的右手拿着什么？", + "answer_chinese": "拐杖", + "evidence": [ + { + "description": "Cane", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:01:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walking stick.", + "B": "Tennis racket.", + "C": "Water bottle.", + "D": "You are holding a pencil case in your right hand while looking at the sculptures.", + "E": "You are holding a paperweight in your right hand as you move." + }, + "answer_label": "A", + "answer_text": "Walking stick.", + "q_id": "01286_8", + "correct_answer": "Walking stick.", + "original_options": { + "A": "Walking stick.", + "B": "Tennis racket.", + "C": "Water bottle.", + "D": "You are holding a pencil case in your right hand while looking at the sculptures.", + "E": "You are holding a paperweight in your right hand as you move." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 12 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_3_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is my left hand touching right now?", + "answer": "The small dog sculpture in front of you.", + "question_chinese": "我的左手现在碰的是什么？", + "answer_chinese": "你面前的小狗雕塑", + "evidence": [ + { + "description": "the puppy sculpture in front of you", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The small dog sculpture behind you.", + "B": "The small dog sculpture to your left.", + "C": "Your left hand is touching the small dog sculpture to your right right now.", + "D": "Your left hand is touching the small dog sculpture above you, with your arm raised to reach it.", + "E": "The small dog sculpture in front of you." + }, + "answer_label": "E", + "answer_text": "The small dog sculpture in front of you.", + "q_id": "01286_3_T1", + "correct_answer": "The small dog sculpture in front of you.", + "original_options": { + "A": "The small dog sculpture behind you.", + "B": "The small dog sculpture to your left.", + "C": "Your left hand is touching the small dog sculpture to your right right now.", + "D": "Your left hand is touching the small dog sculpture above you, with your arm raised to reach it.", + "E": "The small dog sculpture in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 18 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_3_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is my left hand touching right now?", + "answer": "The celebrity sculpture in front of you.", + "question_chinese": "我的左手现在碰的是什么？", + "answer_chinese": "你面前的名人雕塑", + "evidence": [ + { + "description": "the celebrity sculpture in front of you", + "timestamps": { + "start": "00:00:39", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The celebrity sculpture behind you.", + "B": "The celebrity sculpture to your left.", + "C": "Your left hand is touching the celebrity sculpture beside you, not the one directly in front.", + "D": "The celebrity sculpture in front of you.", + "E": "You are touching the celebrity sculpture off to your right, rather than the sculpture positioned ahead." + }, + "answer_label": "D", + "answer_text": "The celebrity sculpture in front of you.", + "q_id": "01286_3_T2", + "correct_answer": "The celebrity sculpture in front of you.", + "original_options": { + "A": "The celebrity sculpture behind you.", + "B": "The celebrity sculpture to your left.", + "C": "Your left hand is touching the celebrity sculpture beside you, not the one directly in front.", + "D": "The celebrity sculpture in front of you.", + "E": "You are touching the celebrity sculpture off to your right, rather than the sculpture positioned ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_12", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see?", + "answer": "seven.", + "question_chinese": "我看到了多少个人？", + "answer_chinese": "7", + "evidence": [ + { + "description": "three people", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "five people", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + }, + { + "description": "seven people", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "Eight.", + "C": "You saw six people in total during the scene as you looked around the area.", + "D": "You counted eight people overall as you moved through the scene and scanned around.", + "E": "seven." + }, + "answer_label": "E", + "answer_text": "seven.", + "q_id": "01286_12", + "correct_answer": "seven.", + "original_options": { + "A": "Six.", + "B": "Eight.", + "C": "You saw six people in total during the scene as you looked around the area.", + "D": "You counted eight people overall as you moved through the scene and scanned around.", + "E": "seven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_11", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see?", + "answer": "Five.", + "question_chinese": "我看到了多少个人？", + "answer_chinese": "5", + "evidence": [ + { + "description": "three people", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "five people", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "You only saw four people over the whole clip from start to finish.", + "E": "You saw six people in total as you moved through the scene and looked around." + }, + "answer_label": "C", + "answer_text": "Five.", + "q_id": "01286_11", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "You only saw four people over the whole clip from start to finish.", + "E": "You saw six people in total as you moved through the scene and looked around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_9", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sculptures do I see now?", + "answer": "Two.", + "question_chinese": "现在我看到了多少个雕塑？", + "answer_chinese": "2", + "evidence": [ + { + "description": "I saw a sculpture", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "I saw two sculptures", + "timestamps": { + "start": "00:00:37", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You can see three sculptures at the same time right now in front of you.", + "E": "There is only one sculpture visible in your current view at this moment." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "01286_9", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You can see three sculptures at the same time right now in front of you.", + "E": "There is only one sculpture visible in your current view at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 13 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_10", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sculptures do I see now?", + "answer": "seven.", + "question_chinese": "现在我看到了多少个雕塑？", + "answer_chinese": "7", + "evidence": [ + { + "description": "I saw a sculpture", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "I saw two sculptures", + "timestamps": { + "start": "00:00:37", + "end": "00:00:38" + } + }, + { + "description": "I saw seven sculptures", + "timestamps": { + "start": "00:00:54", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Eight.", + "C": "You are looking at five sculptures in the display right now, all visible at once.", + "D": "You can count eight sculptures visible across the scene at this moment.", + "E": "seven." + }, + "answer_label": "E", + "answer_text": "seven.", + "q_id": "01286_10", + "correct_answer": "seven.", + "original_options": { + "A": "Five.", + "B": "Eight.", + "C": "You are looking at five sculptures in the display right now, all visible at once.", + "D": "You can count eight sculptures visible across the scene at this moment.", + "E": "seven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 12 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_3", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the boy in front of me wearing on his hand?", + "answer": "Beaded bracelet.", + "question_chinese": "面前的男生手上戴着什么？", + "answer_chinese": "手串", + "evidence": [ + { + "description": "Bracelet", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A metal watch.", + "B": "A leather wristband.", + "C": "He is wearing a fabric glove on his hand instead of any bracelet.", + "D": "The boy has a silicone ring on his hand rather than a bracelet.", + "E": "Beaded bracelet." + }, + "answer_label": "E", + "answer_text": "Beaded bracelet.", + "q_id": "01286_3", + "correct_answer": "Beaded bracelet.", + "original_options": { + "A": "A metal watch.", + "B": "A leather wristband.", + "C": "He is wearing a fabric glove on his hand instead of any bracelet.", + "D": "The boy has a silicone ring on his hand rather than a bracelet.", + "E": "Beaded bracelet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 13 + ] + }, + { + "video_uid": "01286", + "video_length": "00:01:32", + "qid": "01286_4", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the boy in front of me wearing on his hand?", + "answer": "Beaded bracelet.", + "question_chinese": "面前的男生手上戴着什么？", + "answer_chinese": "手串", + "evidence": [ + { + "description": "Bracelet", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Beaded bracelet.", + "B": "A silver bracelet.", + "C": "A woven string bracelet.", + "D": "He is wearing a metal chain bracelet on his hand instead of beads.", + "E": "The boy has a leather strap bracelet on his hand, not a beaded one." + }, + "answer_label": "A", + "answer_text": "Beaded bracelet.", + "q_id": "01286_4", + "correct_answer": "Beaded bracelet.", + "original_options": { + "A": "Beaded bracelet.", + "B": "A silver bracelet.", + "C": "A woven string bracelet.", + "D": "He is wearing a metal chain bracelet on his hand instead of beads.", + "E": "The boy has a leather strap bracelet on his hand, not a beaded one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 14 + ] + } +] diff --git a/QAs/01303-mcq.json b/QAs/01303-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fe971f829ffa0b25ee8df0f10d40cd6899d7b73b --- /dev/null +++ b/QAs/01303-mcq.json @@ -0,0 +1,770 @@ +[ + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_5_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I take an escalator?", + "answer": "You did not take the escalator.", + "question_chinese": "我有没有上扶梯？", + "answer_chinese": "你没有上扶梯。", + "evidence": [ + { + "description": "The escalator is right in front of you.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you took the escalator.", + "B": "You did not take the escalator." + }, + "answer_label": "B", + "answer_text": "You did not take the escalator.", + "q_id": "01303_5_T1", + "correct_answer": "You did not take the escalator.", + "original_options": { + "A": "Yes, you took the escalator.", + "B": "You did not take the escalator." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_9", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I get on the escalator?", + "answer": "Yes.", + "question_chinese": "我有没有上扶梯？", + "answer_chinese": "有", + "evidence": [ + { + "description": "I got on the escalator", + "timestamps": { + "start": "00:00:45", + "end": "00:00:46" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01303_9", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_5_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I take an escalator?", + "answer": "Yes.", + "question_chinese": "我有没有上扶梯？", + "answer_chinese": "有", + "evidence": [ + { + "description": "I got on the escalator", + "timestamps": { + "start": "00:00:45", + "end": "00:00:46" + } + } + ], + "question_timestamps": "00:00:54", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01303_5_T2", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_2_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the fire extinguisher?", + "answer": "Right in front of you.", + "question_chinese": "我看见灭火器在哪里？", + "answer_chinese": "在你的正前面。", + "evidence": [ + { + "description": "Left side of the escalator", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right behind you.", + "B": "Right to your left.", + "C": "Right in front of you.", + "D": "It was positioned to your right rather than directly in front of you at eye level.", + "E": "You spotted it above your line of sight rather than straight in front of you." + }, + "answer_label": "C", + "answer_text": "Right in front of you.", + "q_id": "01303_2_T1", + "correct_answer": "Right in front of you.", + "original_options": { + "A": "Right behind you.", + "B": "Right to your left.", + "C": "Right in front of you.", + "D": "It was positioned to your right rather than directly in front of you at eye level.", + "E": "You spotted it above your line of sight rather than straight in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 15 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_2_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the fire extinguisher?", + "answer": "To the left of the escalator in front of you.", + "question_chinese": "我看见灭火器在哪里？", + "answer_chinese": "你面前的扶梯左边", + "evidence": [ + { + "description": "Left side of the escalator", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To the right of the escalator.", + "B": "Behind the escalator.", + "C": "To the left of the escalator in front of you.", + "D": "It was to the left of the escalator behind you, not the one directly ahead.", + "E": "You saw it to the left of an escalator off to your right rather than the escalator in front." + }, + "answer_label": "C", + "answer_text": "To the left of the escalator in front of you.", + "q_id": "01303_2_T2", + "correct_answer": "To the left of the escalator in front of you.", + "original_options": { + "A": "To the right of the escalator.", + "B": "Behind the escalator.", + "C": "To the left of the escalator in front of you.", + "D": "It was to the left of the escalator behind you, not the one directly ahead.", + "E": "You saw it to the left of an escalator off to your right rather than the escalator in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 3, + 15, + 19 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_1_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the man wearing a black mask?", + "answer": "To the right of the woman in a black mask at your front-left.", + "question_chinese": "我看见黑口罩男人在哪里？", + "answer_chinese": "在你的左前方黑口罩女人的右边", + "evidence": [ + { + "description": "To the right of the woman with the black mask", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To the right of the woman in a black mask at your front-left.", + "B": "To her left at front-left.", + "C": "To her right at front-right.", + "D": "He was to the right of the woman in a black mask, but she was positioned at your front-right.", + "E": "You saw him to the right of the masked woman when she was behind you on the left side." + }, + "answer_label": "A", + "answer_text": "To the right of the woman in a black mask at your front-left.", + "q_id": "01303_1_T2", + "correct_answer": "To the right of the woman in a black mask at your front-left.", + "original_options": { + "A": "To the right of the woman in a black mask at your front-left.", + "B": "To her left at front-left.", + "C": "To her right at front-right.", + "D": "He was to the right of the woman in a black mask, but she was positioned at your front-right.", + "E": "You saw him to the right of the masked woman when she was behind you on the left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_1_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the man wearing a black mask?", + "answer": "Behind the woman in a black mask directly in front of you.", + "question_chinese": "我看见黑口罩男人在哪里？", + "answer_chinese": "在你的正前方黑口罩女人的后面。", + "evidence": [ + { + "description": "Behind the woman wearing a black mask.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of her.", + "B": "To her right at front-left.", + "C": "He was behind the woman in a black mask, but she was directly behind you instead of in front.", + "D": "You saw him behind the masked woman, though she was in back of you rather than directly ahead.", + "E": "Behind the woman in a black mask directly in front of you." + }, + "answer_label": "E", + "answer_text": "Behind the woman in a black mask directly in front of you.", + "q_id": "01303_1_T1", + "correct_answer": "Behind the woman in a black mask directly in front of you.", + "original_options": { + "A": "In front of her.", + "B": "To her right at front-left.", + "C": "He was behind the woman in a black mask, but she was directly behind you instead of in front.", + "D": "You saw him behind the masked woman, though she was in back of you rather than directly ahead.", + "E": "Behind the woman in a black mask directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 19, + 18 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_3_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people in front of me were wearing red clothes?", + "answer": "No one in front of you is wearing red.", + "question_chinese": "我面前有多少个人穿红衣？", + "answer_chinese": "你面前没有人穿红衣服。", + "evidence": [ + { + "description": "Everyone is wearing black clothes.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One person.", + "B": "Two people.", + "C": "No one in front of you is wearing red.", + "D": "There was one person in front of you wearing red clothing at that moment.", + "E": "You had two people ahead of you who were dressed in red." + }, + "answer_label": "C", + "answer_text": "No one in front of you is wearing red.", + "q_id": "01303_3_T2", + "correct_answer": "No one in front of you is wearing red.", + "original_options": { + "A": "One person.", + "B": "Two people.", + "C": "No one in front of you is wearing red.", + "D": "There was one person in front of you wearing red clothing at that moment.", + "E": "You had two people ahead of you who were dressed in red." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 12 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_3_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people in front of me were wearing red clothes?", + "answer": "There is 1 person in front of you wearing red.", + "question_chinese": "我面前有多少个人穿红衣？", + "answer_chinese": "你面前有1个人穿红衣。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:21", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero people.", + "B": "There is 1 person in front of you wearing red.", + "C": "Two people.", + "D": "There were no people in front of you wearing red clothes at the time.", + "E": "You saw two people in front of you dressed in red clothing." + }, + "answer_label": "B", + "answer_text": "There is 1 person in front of you wearing red.", + "q_id": "01303_3_T1", + "correct_answer": "There is 1 person in front of you wearing red.", + "original_options": { + "A": "Zero people.", + "B": "There is 1 person in front of you wearing red.", + "C": "Two people.", + "D": "There were no people in front of you wearing red clothes at the time.", + "E": "You saw two people in front of you dressed in red clothing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 12 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_3", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the fire extinguisher?", + "answer": "Left side of the escalator.", + "question_chinese": "灭火器在哪里？", + "answer_chinese": "扶梯左边", + "evidence": [ + { + "description": "Left side of the escalator", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right side of the escalator.", + "B": "In front of the escalator.", + "C": "It is placed on the floor next to the escalator, closer to the right-hand side.", + "D": "You can find it mounted behind the escalator, positioned along the wall near the base.", + "E": "Left side of the escalator." + }, + "answer_label": "E", + "answer_text": "Left side of the escalator.", + "q_id": "01303_3", + "correct_answer": "Left side of the escalator.", + "original_options": { + "A": "Right side of the escalator.", + "B": "In front of the escalator.", + "C": "It is placed on the floor next to the escalator, closer to the right-hand side.", + "D": "You can find it mounted behind the escalator, positioned along the wall near the base.", + "E": "Left side of the escalator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_4", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the fire extinguisher?", + "answer": "Left side of the escalator.", + "question_chinese": "灭火器在哪里？", + "answer_chinese": "扶梯左边", + "evidence": [ + { + "description": "Left side of the escalator", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right side of the escalator.", + "B": "Left side of the escalator.", + "C": "In front of the escalator.", + "D": "It is on the floor behind the escalator, slightly offset from the center.", + "E": "It is positioned above the escalator, attached near the ceiling rather than at ground level." + }, + "answer_label": "B", + "answer_text": "Left side of the escalator.", + "q_id": "01303_4", + "correct_answer": "Left side of the escalator.", + "original_options": { + "A": "Right side of the escalator.", + "B": "Left side of the escalator.", + "C": "In front of the escalator.", + "D": "It is on the floor behind the escalator, slightly offset from the center.", + "E": "It is positioned above the escalator, attached near the ceiling rather than at ground level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 15 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_4_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the aunt in yellow clothes relative to me?", + "answer": "To your right.", + "question_chinese": "黄色衣服的阿姨在我的什么方位？", + "answer_chinese": "在你的右方。", + "evidence": [ + { + "description": "She is wearing a hat", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "In front of you.", + "C": "To your right.", + "D": "She is behind you, positioned slightly off to one side as you look ahead.", + "E": "She is above you, appearing higher up relative to your viewpoint rather than beside you." + }, + "answer_label": "C", + "answer_text": "To your right.", + "q_id": "01303_4_T2", + "correct_answer": "To your right.", + "original_options": { + "A": "To your left.", + "B": "In front of you.", + "C": "To your right.", + "D": "She is behind you, positioned slightly off to one side as you look ahead.", + "E": "She is above you, appearing higher up relative to your viewpoint rather than beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 15 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_4_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the aunt in yellow clothes relative to me?", + "answer": "The auntie in yellow is behind you.", + "question_chinese": "黄色衣服的阿姨在我的什么方位？", + "answer_chinese": "黄色衣服的阿姨在你的后方。", + "evidence": [ + { + "description": "The position of the lady in yellow clothes", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The auntie in yellow is behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "She is to your right, standing beside you as you face forward.", + "E": "She is above you, positioned higher than your eye level in the scene." + }, + "answer_label": "A", + "answer_text": "The auntie in yellow is behind you.", + "q_id": "01303_4_T1", + "correct_answer": "The auntie in yellow is behind you.", + "original_options": { + "A": "The auntie in yellow is behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "She is to your right, standing beside you as you face forward.", + "E": "She is above you, positioned higher than your eye level in the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 13 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the man in the black mask?", + "answer": "To the right of the woman with the black mask.", + "question_chinese": "黑口罩男人在哪里？", + "answer_chinese": "黑口罩女人的右边", + "evidence": [ + { + "description": "To the right of the woman with the black mask", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of her.", + "B": "Behind the woman with the black mask.", + "C": "To the left of the woman with the black mask, not on her right side as you look at them.", + "D": "To the right of the woman with the black mask.", + "E": "He is positioned behind the woman with the black mask, slightly offset rather than standing beside her." + }, + "answer_label": "D", + "answer_text": "To the right of the woman with the black mask.", + "q_id": "01303_1", + "correct_answer": "To the right of the woman with the black mask.", + "original_options": { + "A": "In front of her.", + "B": "Behind the woman with the black mask.", + "C": "To the left of the woman with the black mask, not on her right side as you look at them.", + "D": "To the right of the woman with the black mask.", + "E": "He is positioned behind the woman with the black mask, slightly offset rather than standing beside her." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 20, + 17 + ] + }, + { + "video_uid": "01303", + "video_length": "00:00:55", + "qid": "01303_2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the man in the black mask?", + "answer": "To the right of the woman with the black mask.", + "question_chinese": "黑口罩男人在哪里？", + "answer_chinese": "黑口罩女人的右边", + "evidence": [ + { + "description": "To the right of the woman with the black mask", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To the left of her.", + "B": "Behind her.", + "C": "To the right of the woman with the black mask.", + "D": "He is in front of the woman with the black mask, positioned between her and the camera.", + "E": "He is to the right of the woman with the black hat, not the woman wearing the black mask." + }, + "answer_label": "C", + "answer_text": "To the right of the woman with the black mask.", + "q_id": "01303_2", + "correct_answer": "To the right of the woman with the black mask.", + "original_options": { + "A": "To the left of her.", + "B": "Behind her.", + "C": "To the right of the woman with the black mask.", + "D": "He is in front of the woman with the black mask, positioned between her and the camera.", + "E": "He is to the right of the woman with the black hat, not the woman wearing the black mask." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 2, + 17, + 19 + ] + } +] diff --git a/QAs/01401-mcq.json b/QAs/01401-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5dd8a3be65443ceba391dea98709029797460ab7 --- /dev/null +++ b/QAs/01401-mcq.json @@ -0,0 +1,1125 @@ +[ + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which turns did I just take?", + "answer": "two.", + "question_chinese": "我刚才拐了几个弯？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:27", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:29", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "No turns.", + "B": "Three.", + "C": "You took four turns in total during the route, changing direction repeatedly as you moved forward.", + "D": "You made five turns overall, turning at nearly every junction you encountered along the way.", + "E": "two." + }, + "answer_label": "E", + "answer_text": "two.", + "q_id": "01401_1", + "correct_answer": "two.", + "original_options": { + "A": "No turns.", + "B": "Three.", + "C": "You took four turns in total during the route, changing direction repeatedly as you moved forward.", + "D": "You made five turns overall, turning at nearly every junction you encountered along the way.", + "E": "two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 1, + 16, + 15 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which turns did I just take?", + "answer": "Three.", + "question_chinese": "我刚才拐了几个弯？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:27", + "end": "00:00:28" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:00:37", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You only took one turn during the route, continuing straight for the rest of the movement.", + "E": "You took five turns in total, making several directional changes across the path you traveled." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "01401_2", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You only took one turn during the route, continuing straight for the rest of the movement.", + "E": "You took five turns in total, making several directional changes across the path you traveled." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_15", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I just see?", + "answer": "Three.", + "question_chinese": "我刚才看见了多少人？", + "answer_chinese": "3", + "evidence": [ + { + "description": "one person", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "three people", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You just saw two people standing nearby in the area you were looking at.", + "E": "You just saw four people standing nearby as you looked around the space." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "01401_15", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You just saw two people standing nearby in the area you were looking at.", + "E": "You just saw four people standing nearby as you looked around the space." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_16", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I just see?", + "answer": "Four.", + "question_chinese": "我刚才看见了多少人？", + "answer_chinese": "4", + "evidence": [ + { + "description": "one person", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "three people", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "four people", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "You just saw three people together near you as you looked forward a moment ago.", + "E": "You just saw five people in front of you while you were moving through the area." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "01401_16", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "You just saw three people together near you as you looked forward a moment ago.", + "E": "You just saw five people in front of you while you were moving through the area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I pass just now?", + "answer": "Three.", + "question_chinese": "我刚才经过了几个人？", + "answer_chinese": "3", + "evidence": [ + { + "description": "I passed by one person", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + }, + { + "description": "I passed by three people", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You just passed two people as you walked by them in the space ahead of you.", + "E": "You just passed four people while moving forward and walking past them a moment ago." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "01401_3", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You just passed two people as you walked by them in the space ahead of you.", + "E": "You just passed four people while moving forward and walking past them a moment ago." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I pass just now?", + "answer": "four.", + "question_chinese": "我刚才经过了几个人？", + "answer_chinese": "4", + "evidence": [ + { + "description": "I passed by one person", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + }, + { + "description": "I passed by three people", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "I passed by four people", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "four.", + "B": "Three.", + "C": "Five.", + "D": "You just passed three people while moving forward and walking past them a moment ago.", + "E": "You just passed five people moments ago as you continued moving forward through the area." + }, + "answer_label": "A", + "answer_text": "four.", + "q_id": "01401_4", + "correct_answer": "four.", + "original_options": { + "A": "four.", + "B": "Three.", + "C": "Five.", + "D": "You just passed three people while moving forward and walking past them a moment ago.", + "E": "You just passed five people moments ago as you continued moving forward through the area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_6_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many doors did I just go through?", + "answer": "You just went through 1 door.", + "question_chinese": "我刚才进了几个门？", + "answer_chinese": "你刚才进了1个门", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You went through 0 doors.", + "B": "You went through 2 doors.", + "C": "You just went through 0 doors and stayed in the same area without passing any doorway.", + "D": "You just went through 2 doors, entering and then leaving a space in quick succession.", + "E": "You just went through 1 door." + }, + "answer_label": "E", + "answer_text": "You just went through 1 door.", + "q_id": "01401_6_T2", + "correct_answer": "You just went through 1 door.", + "original_options": { + "A": "You went through 0 doors.", + "B": "You went through 2 doors.", + "C": "You just went through 0 doors and stayed in the same area without passing any doorway.", + "D": "You just went through 2 doors, entering and then leaving a space in quick succession.", + "E": "You just went through 1 door." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_6_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many doors did I just go through?", + "answer": "1.", + "question_chinese": "我刚才进了几个门？", + "answer_chinese": "1", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "1.", + "B": "2.", + "C": "3.", + "D": "You just went through 2 doors, passing one doorway and then another right afterward.", + "E": "You just went through 3 doors in quick succession while moving ahead through the area." + }, + "answer_label": "A", + "answer_text": "1.", + "q_id": "01401_6_T1", + "correct_answer": "1.", + "original_options": { + "A": "1.", + "B": "2.", + "C": "3.", + "D": "You just went through 2 doors, passing one doorway and then another right afterward.", + "E": "You just went through 3 doors in quick succession while moving ahead through the area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_7", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did my right hand touch?", + "answer": "Door curtain.", + "question_chinese": "我的右手碰到了什么？", + "answer_chinese": "门帘", + "evidence": [ + { + "description": "I bumped into the door curtain", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Door handle.", + "B": "Window frame.", + "C": "Your right hand touched the table surface while you were standing close to it.", + "D": "Your right hand touched a wall decoration near you as you moved past the wall.", + "E": "Door curtain." + }, + "answer_label": "E", + "answer_text": "Door curtain.", + "q_id": "01401_7", + "correct_answer": "Door curtain.", + "original_options": { + "A": "Door handle.", + "B": "Window frame.", + "C": "Your right hand touched the table surface while you were standing close to it.", + "D": "Your right hand touched a wall decoration near you as you moved past the wall.", + "E": "Door curtain." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_4_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did my right hand touch?", + "answer": "Your right hand touched the plastic strip curtain. The strip curtain was in front of you.", + "question_chinese": "我的右手碰到了什么？", + "answer_chinese": "你的右手碰到了塑料门帘，门帘在你的正前方。", + "evidence": [ + { + "description": "I bumped into the door curtain", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your right hand touched the plastic strip curtain. The strip curtain was in front of you.", + "B": "Plastic strip curtain behind you.", + "C": "Plastic strip curtain to your left.", + "D": "Your right hand touched the plastic strip curtain, and the strip curtain was to your right.", + "E": "Your right hand touched the plastic strip curtain, and the strip curtain was above you." + }, + "answer_label": "A", + "answer_text": "Your right hand touched the plastic strip curtain. The strip curtain was in front of you.", + "q_id": "01401_4_T2", + "correct_answer": "Your right hand touched the plastic strip curtain. The strip curtain was in front of you.", + "original_options": { + "A": "Your right hand touched the plastic strip curtain. The strip curtain was in front of you.", + "B": "Plastic strip curtain behind you.", + "C": "Plastic strip curtain to your left.", + "D": "Your right hand touched the plastic strip curtain, and the strip curtain was to your right.", + "E": "Your right hand touched the plastic strip curtain, and the strip curtain was above you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_4_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did my right hand touch?", + "answer": "You touched a card.", + "question_chinese": "我的右手碰到了什么？", + "answer_chinese": "你碰到了卡。", + "evidence": [ + { + "description": "my right hand touches the card.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You touched a cane.", + "B": "You touched a strip curtain.", + "C": "You touched a card.", + "D": "You touched an umbrella with your right hand while standing near the entrance area.", + "E": "You touched a slip of paper with your right hand, then pulled your hand back immediately." + }, + "answer_label": "C", + "answer_text": "You touched a card.", + "q_id": "01401_4_T1", + "correct_answer": "You touched a card.", + "original_options": { + "A": "You touched a cane.", + "B": "You touched a strip curtain.", + "C": "You touched a card.", + "D": "You touched an umbrella with your right hand while standing near the entrance area.", + "E": "You touched a slip of paper with your right hand, then pulled your hand back immediately." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 16 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_12", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see?", + "answer": "two.", + "question_chinese": "我看见了几张椅子？", + "answer_chinese": "2", + "evidence": [ + { + "description": "There are two chairs", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "two.", + "C": "Three.", + "D": "You saw 4 chairs in total as you looked around the area.", + "E": "You saw 5 chairs altogether, spread out across the space in front of you." + }, + "answer_label": "B", + "answer_text": "two.", + "q_id": "01401_12", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "two.", + "C": "Three.", + "D": "You saw 4 chairs in total as you looked around the area.", + "E": "You saw 5 chairs altogether, spread out across the space in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_7_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see?", + "answer": "You saw 3 chairs: a wheelchair, a wooden chair, and a bench where a girl was sitting.", + "question_chinese": "我看见了几张椅子？", + "answer_chinese": "你看见了3张椅子，一张轮椅，一张木头椅子，还有一张女生坐的长椅", + "evidence": [ + { + "description": "There are two chairs", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 3 chairs: a wheelchair, a wooden chair, and a bench where a girl was sitting.", + "B": "You saw 2 chairs.", + "C": "You saw 4 chairs.", + "D": "You saw 1 chair: just a wheelchair, with no other chairs visible nearby.", + "E": "You saw 5 chairs: a wheelchair, a wooden chair, and three benches in the area." + }, + "answer_label": "A", + "answer_text": "You saw 3 chairs: a wheelchair, a wooden chair, and a bench where a girl was sitting.", + "q_id": "01401_7_T2", + "correct_answer": "You saw 3 chairs: a wheelchair, a wooden chair, and a bench where a girl was sitting.", + "original_options": { + "A": "You saw 3 chairs: a wheelchair, a wooden chair, and a bench where a girl was sitting.", + "B": "You saw 2 chairs.", + "C": "You saw 4 chairs.", + "D": "You saw 1 chair: just a wheelchair, with no other chairs visible nearby.", + "E": "You saw 5 chairs: a wheelchair, a wooden chair, and three benches in the area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 15 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_7_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see?", + "answer": "You saw 2.", + "question_chinese": "我看见了几张椅子？", + "answer_chinese": "你看见了2张", + "evidence": [ + { + "description": "There are two chairs", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 0.", + "B": "You saw 1.", + "C": "You saw 3 chairs as you scanned the area in front of you.", + "D": "You saw 4 chairs in total as you moved through the space.", + "E": "You saw 2." + }, + "answer_label": "E", + "answer_text": "You saw 2.", + "q_id": "01401_7_T1", + "correct_answer": "You saw 2.", + "original_options": { + "A": "You saw 0.", + "B": "You saw 1.", + "C": "You saw 3 chairs as you scanned the area in front of you.", + "D": "You saw 4 chairs in total as you moved through the space.", + "E": "You saw 2." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 12 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_5_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I walked in through the door, what did my hand touch?", + "answer": "Your hand touched the strip curtain. The strip curtain was in front of you.", + "question_chinese": "我进门时手触碰到了什么？", + "answer_chinese": "你的手触碰到了门帘，门帘在你面前。", + "evidence": [ + { + "description": "I bumped into the door curtain", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your hand touched a cane.", + "B": "Your hand touched the strip curtain. The strip curtain was in front of you.", + "C": "Your hand touched a card.", + "D": "Your hand touched the strip curtain, but it was positioned to your left as you entered.", + "E": "Your hand touched the strip curtain, and it was behind you after you stepped through the doorway." + }, + "answer_label": "B", + "answer_text": "Your hand touched the strip curtain. The strip curtain was in front of you.", + "q_id": "01401_5_T2", + "correct_answer": "Your hand touched the strip curtain. The strip curtain was in front of you.", + "original_options": { + "A": "Your hand touched a cane.", + "B": "Your hand touched the strip curtain. The strip curtain was in front of you.", + "C": "Your hand touched a card.", + "D": "Your hand touched the strip curtain, but it was positioned to your left as you entered.", + "E": "Your hand touched the strip curtain, and it was behind you after you stepped through the doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_5_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I walked in through the door, what did my hand touch?", + "answer": "Your hand touched a cane. The cane was in your hand.", + "question_chinese": "我进门时手触碰到了什么？", + "answer_chinese": "你的手触碰到了拐杖。拐杖在你手里。", + "evidence": [ + { + "description": "I touched the walking stick.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your hand touched a card.", + "B": "Your hand touched a cane. The cane was in your hand.", + "C": "Your hand touched the strip curtain.", + "D": "Your hand touched a slip of paper, which you briefly held between your fingers.", + "E": "Your hand touched an umbrella handle as you moved through the doorway into the area." + }, + "answer_label": "B", + "answer_text": "Your hand touched a cane. The cane was in your hand.", + "q_id": "01401_5_T1", + "correct_answer": "Your hand touched a cane. The cane was in your hand.", + "original_options": { + "A": "Your hand touched a card.", + "B": "Your hand touched a cane. The cane was in your hand.", + "C": "Your hand touched the strip curtain.", + "D": "Your hand touched a slip of paper, which you briefly held between your fingers.", + "E": "Your hand touched an umbrella handle as you moved through the doorway into the area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 15 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_8_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was the person I met holding in their right hand?", + "answer": "The first person you saw was holding an umbrella in their right hand. That first person was to your front-left.", + "question_chinese": "我遇见的人右手拿的是什么？", + "answer_chinese": "你看见的第一个人右手拿着雨伞，第一个人在你的左前方。", + "evidence": [ + { + "description": "She is holding an umbrella", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They held nothing in their right hand.", + "B": "They held a slip of paper.", + "C": "The first person you saw was holding an umbrella in their right hand. That first person was to your front-left.", + "D": "The first person you saw was holding an umbrella in their left hand, standing to your front-left.", + "E": "The first person you saw was holding an umbrella in their right hand, but they were positioned to your back-left." + }, + "answer_label": "C", + "answer_text": "The first person you saw was holding an umbrella in their right hand. That first person was to your front-left.", + "q_id": "01401_8_T1", + "correct_answer": "The first person you saw was holding an umbrella in their right hand. That first person was to your front-left.", + "original_options": { + "A": "They held nothing in their right hand.", + "B": "They held a slip of paper.", + "C": "The first person you saw was holding an umbrella in their right hand. That first person was to your front-left.", + "D": "The first person you saw was holding an umbrella in their left hand, standing to your front-left.", + "E": "The first person you saw was holding an umbrella in their right hand, but they were positioned to your back-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 20 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_8_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was the person I met holding in their right hand?", + "answer": "The second person you ran into had nothing in their right hand, and the third person was holding a slip of paper in their right hand. The second and third people were to your left.", + "question_chinese": "我遇见的人右手拿的是什么？", + "answer_chinese": "我遇见的第二个人右手没拿东西，第三个人右手拿着单子。第二和第三人在你的左边。", + "evidence": [ + { + "description": "the third person is holding the list.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The second person you ran into had nothing in their right hand, and the third person was holding a slip of paper in their right hand. The second and third people were to your left.", + "B": "Both had nothing in their right hand.", + "C": "Both were holding a slip of paper.", + "D": "The second person held a slip of paper, while the third person had nothing in their right hand.", + "E": "The second had nothing and the third had paper, but both were to your right." + }, + "answer_label": "A", + "answer_text": "The second person you ran into had nothing in their right hand, and the third person was holding a slip of paper in their right hand. The second and third people were to your left.", + "q_id": "01401_8_T2", + "correct_answer": "The second person you ran into had nothing in their right hand, and the third person was holding a slip of paper in their right hand. The second and third people were to your left.", + "original_options": { + "A": "The second person you ran into had nothing in their right hand, and the third person was holding a slip of paper in their right hand. The second and third people were to your left.", + "B": "Both had nothing in their right hand.", + "C": "Both were holding a slip of paper.", + "D": "The second person held a slip of paper, while the third person had nothing in their right hand.", + "E": "The second had nothing and the third had paper, but both were to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 15 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_3_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the bucket relative to me?", + "answer": "The bucket was directly in front of you.", + "question_chinese": "水桶在我的什么位置？", + "answer_chinese": "水桶在你的正前方", + "evidence": [ + { + "description": "buckets is in front of me.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The bucket was behind you.", + "B": "The bucket was to your left.", + "C": "The bucket was off to your right side, not directly in front of you.", + "D": "The bucket was above you, positioned overhead rather than straight ahead from your viewpoint.", + "E": "The bucket was directly in front of you." + }, + "answer_label": "E", + "answer_text": "The bucket was directly in front of you.", + "q_id": "01401_3_T1", + "correct_answer": "The bucket was directly in front of you.", + "original_options": { + "A": "The bucket was behind you.", + "B": "The bucket was to your left.", + "C": "The bucket was off to your right side, not directly in front of you.", + "D": "The bucket was above you, positioned overhead rather than straight ahead from your viewpoint.", + "E": "The bucket was directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 14 + ] + }, + { + "video_uid": "01401", + "video_length": "00:00:47", + "qid": "01401_3_T2", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the bucket relative to me?", + "answer": "The bucket was behind you.", + "question_chinese": "水桶在我的什么位置？", + "answer_chinese": "水桶在你的后面", + "evidence": [ + { + "description": "buckets is behind me.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The bucket was in front of you.", + "B": "The bucket was to your right.", + "C": "The bucket was behind you.", + "D": "The bucket was directly to your left side instead of being behind you.", + "E": "The bucket was above you, not located behind you at your back." + }, + "answer_label": "C", + "answer_text": "The bucket was behind you.", + "q_id": "01401_3_T2", + "correct_answer": "The bucket was behind you.", + "original_options": { + "A": "The bucket was in front of you.", + "B": "The bucket was to your right.", + "C": "The bucket was behind you.", + "D": "The bucket was directly to your left side instead of being behind you.", + "E": "The bucket was above you, not located behind you at your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 12 + ] + } +] diff --git a/QAs/01409-mcq.json b/QAs/01409-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a9672554acf9dc5629e72cc845f2467c1de3c3bf --- /dev/null +++ b/QAs/01409-mcq.json @@ -0,0 +1,334 @@ +[ + { + "video_uid": "01409", + "video_length": "00:00:49", + "qid": "01409_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Since going up the stairs, how many stools have appeared within my field of vision?", + "answer": "one.", + "question_chinese": "从上楼梯到现在我的视线范围内出现了几个凳子？", + "answer_chinese": "1个", + "evidence": [ + { + "description": "There is a stool in the kitchen.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two stools.", + "B": "Three stools.", + "C": "Five stools appeared within your field of vision after you went up the stairs.", + "D": "Eight stools came into view since you went up the stairs and looked around.", + "E": "one." + }, + "answer_label": "E", + "answer_text": "one.", + "q_id": "01409_1", + "correct_answer": "one.", + "original_options": { + "A": "Two stools.", + "B": "Three stools.", + "C": "Five stools appeared within your field of vision after you went up the stairs.", + "D": "Eight stools came into view since you went up the stairs and looked around.", + "E": "one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "01409", + "video_length": "00:00:49", + "qid": "01409_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Since going up the stairs, how many stools have appeared within my field of vision?", + "answer": "four.", + "question_chinese": "从上楼梯到现在我的视线范围内出现了几个凳子？", + "answer_chinese": "4个", + "evidence": [ + { + "description": "There are 3 more stools on the second floor, making a total of 4 stools.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + }, + { + "description": "There are 1 stool on the first floor.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "four.", + "B": "Three stools.", + "C": "Five stools.", + "D": "Two stools have appeared within your field of vision since going up the stairs.", + "E": "Six stools came into view after you went up the stairs and scanned the area." + }, + "answer_label": "A", + "answer_text": "four.", + "q_id": "01409_2", + "correct_answer": "four.", + "original_options": { + "A": "four.", + "B": "Three stools.", + "C": "Five stools.", + "D": "Two stools have appeared within your field of vision since going up the stairs.", + "E": "Six stools came into view after you went up the stairs and scanned the area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "01409", + "video_length": "00:00:49", + "qid": "01409_11", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which floor is the dog on now?", + "answer": "second floor.", + "question_chinese": "狗狗现在在几楼？", + "answer_chinese": "2楼", + "evidence": [ + { + "description": "The dog was just by my side.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "First floor.", + "B": "second floor.", + "C": "Third floor.", + "D": "The dog is on the ground floor right now, not up on the second level.", + "E": "The dog is down in the basement now instead of being on the second floor." + }, + "answer_label": "B", + "answer_text": "second floor.", + "q_id": "01409_11", + "correct_answer": "second floor.", + "original_options": { + "A": "First floor.", + "B": "second floor.", + "C": "Third floor.", + "D": "The dog is on the ground floor right now, not up on the second level.", + "E": "The dog is down in the basement now instead of being on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "01409", + "video_length": "00:00:49", + "qid": "01409_12", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which floor is the dog on now?", + "answer": "first floor.", + "question_chinese": "狗狗现在在几楼？", + "answer_chinese": "1楼", + "evidence": [ + { + "description": "The dog ran from the second floor to the first floor.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:30" + } + }, + { + "description": "I didn't see the dog on the second floor.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:36 " + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "first floor.", + "B": "Second floor.", + "C": "Basement level.", + "D": "Third floor, not the first, where you ended up after going upstairs earlier during the walk.", + "E": "Second floor, after going up the stairs a moment ago from the first level." + }, + "answer_label": "A", + "answer_text": "first floor.", + "q_id": "01409_12", + "correct_answer": "first floor.", + "original_options": { + "A": "first floor.", + "B": "Second floor.", + "C": "Basement level.", + "D": "Third floor, not the first, where you ended up after going upstairs earlier during the walk.", + "E": "Second floor, after going up the stairs a moment ago from the first level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 14 + ] + }, + { + "video_uid": "01409", + "video_length": "00:00:49", + "qid": "01409_13", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which areas of the whole room can you see now?", + "answer": "Kitchen, living room, stairs.", + "question_chinese": "现在你可以看到整个房间的哪几个区域？", + "answer_chinese": "厨房、客厅、楼梯", + "evidence": [ + { + "description": "Straight ahead are the stairs.The kitchen is on the left.Next to the kitchen is the living room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Living room and stairs.", + "B": "Kitchen and stairs.", + "C": "You can see the living room, hallway, and stairs, but not the kitchen area.", + "D": "Kitchen, living room, stairs.", + "E": "You can see the kitchen, living room, and hallway, while the stairs are out of view." + }, + "answer_label": "D", + "answer_text": "Kitchen, living room, stairs.", + "q_id": "01409_13", + "correct_answer": "Kitchen, living room, stairs.", + "original_options": { + "A": "Living room and stairs.", + "B": "Kitchen and stairs.", + "C": "You can see the living room, hallway, and stairs, but not the kitchen area.", + "D": "Kitchen, living room, stairs.", + "E": "You can see the kitchen, living room, and hallway, while the stairs are out of view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 16 + ] + }, + { + "video_uid": "01409", + "video_length": "00:00:49", + "qid": "01409_14", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which areas of the whole room can you see now?", + "answer": "Bedroom, living room.", + "question_chinese": "现在你可以看到整个房间的哪几个区域？", + "answer_chinese": "卧室、客厅", + "evidence": [ + { + "description": "Ahead is the bedroom.Next to it is the living room", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Bedroom and hallway.", + "B": "Living room and kitchen.", + "C": "You can see the bedroom and dining room now, not both the bedroom and living room.", + "D": "You can see the living room and hallway from here, while the bedroom isn’t visible.", + "E": "Bedroom, living room." + }, + "answer_label": "E", + "answer_text": "Bedroom, living room.", + "q_id": "01409_14", + "correct_answer": "Bedroom, living room.", + "original_options": { + "A": "Bedroom and hallway.", + "B": "Living room and kitchen.", + "C": "You can see the bedroom and dining room now, not both the bedroom and living room.", + "D": "You can see the living room and hallway from here, while the bedroom isn’t visible.", + "E": "Bedroom, living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 16 + ] + } +] diff --git a/QAs/01434-mcq.json b/QAs/01434-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..15a158080f2d903d5648911fa3df8a2dc5690ae2 --- /dev/null +++ b/QAs/01434-mcq.json @@ -0,0 +1,902 @@ +[ + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_1_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I take the elevator?", + "answer": "Right now you’re riding the elevator for the first time, and the elevator is behind you.", + "question_chinese": "我坐了几次电梯？", + "answer_chinese": "你现在在坐第一次的电梯，电梯在你的后面。", + "evidence": [ + { + "description": "I came up by escalator", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took the elevator twice.", + "B": "You took the elevator three times.", + "C": "In total, you took the elevator twice, and the elevator is directly in front of you right now.", + "D": "Right now you’re riding the elevator for the second time, and the elevator is behind you.", + "E": "Right now you’re riding the elevator for the first time, and the elevator is behind you." + }, + "answer_label": "E", + "answer_text": "Right now you’re riding the elevator for the first time, and the elevator is behind you.", + "q_id": "01434_1_T1", + "correct_answer": "Right now you’re riding the elevator for the first time, and the elevator is behind you.", + "original_options": { + "A": "You took the elevator twice.", + "B": "You took the elevator three times.", + "C": "In total, you took the elevator twice, and the elevator is directly in front of you right now.", + "D": "Right now you’re riding the elevator for the second time, and the elevator is behind you.", + "E": "Right now you’re riding the elevator for the first time, and the elevator is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 17 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_1_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I take the elevator?", + "answer": "In total, you took the elevator twice, and the elevator is to your left.", + "question_chinese": "我坐了几次电梯？", + "answer_chinese": "你一共坐了两次电梯，电梯在你的左边。", + "evidence": [ + { + "description": "I came up by escalator", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "I took the elevator down to the subway", + "timestamps": { + "start": "00:02:16", + "end": "00:02:23" + } + } + ], + "question_timestamps": "00:02:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In total, you took the elevator twice, and the elevator is to your left.", + "B": "Once, and it’s to your left.", + "C": "Three times, and it’s to your left.", + "D": "In total, you took the elevator once, and the elevator is behind you now.", + "E": "In total, you took the elevator three times, and the elevator is behind you on your right." + }, + "answer_label": "A", + "answer_text": "In total, you took the elevator twice, and the elevator is to your left.", + "q_id": "01434_1_T2", + "correct_answer": "In total, you took the elevator twice, and the elevator is to your left.", + "original_options": { + "A": "In total, you took the elevator twice, and the elevator is to your left.", + "B": "Once, and it’s to your left.", + "C": "Three times, and it’s to your left.", + "D": "In total, you took the elevator once, and the elevator is behind you now.", + "E": "In total, you took the elevator three times, and the elevator is behind you on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 14, + 17 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_7_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What am I holding in my hand?", + "answer": "Right now you’re holding a guide cane in your hand, and the guide cane is to your right.", + "question_chinese": "我手上拿的是什么？", + "answer_chinese": "你现在手上拿的是一根导盲杖，导盲杖在你的右边。", + "evidence": [ + { + "description": "What you are holding now is a guide stick.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:26", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you’re holding a guide cane in your hand, and the guide cane is to your right.", + "B": "Right now you’re holding your ID card in your hand, and the ID card is to your left." + }, + "answer_label": "A", + "answer_text": "Right now you’re holding a guide cane in your hand, and the guide cane is to your right.", + "q_id": "01434_7_T1", + "correct_answer": "Right now you’re holding a guide cane in your hand, and the guide cane is to your right.", + "original_options": { + "A": "Right now you’re holding a guide cane in your hand, and the guide cane is to your right.", + "B": "Right now you’re holding your ID card in your hand, and the ID card is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 19 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_7_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What am I holding in my hand?", + "answer": "Right now you’re holding your ID card in your hand, and the ID card is to your left.", + "question_chinese": "我手上拿的是什么？", + "answer_chinese": "你现在手上拿的是你的身份证，身份证在你的左边。", + "evidence": [ + { + "description": "What you are holding now is your ID card.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you’re holding a guide cane in your hand, and the guide cane is to your right.", + "B": "Right now you’re holding your ID card in your hand, and the ID card is to your left." + }, + "answer_label": "B", + "answer_text": "Right now you’re holding your ID card in your hand, and the ID card is to your left.", + "q_id": "01434_7_T2", + "correct_answer": "Right now you’re holding your ID card in your hand, and the ID card is to your left.", + "original_options": { + "A": "Right now you’re holding a guide cane in your hand, and the guide cane is to your right.", + "B": "Right now you’re holding your ID card in your hand, and the ID card is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 19 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_9_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What am I doing right now?", + "answer": "Right now you’re going through the security scanner, and the security scanner is directly in front of you.", + "question_chinese": "我现在在干嘛？", + "answer_chinese": "你现在在过安检机，安检机在你的正前方。", + "evidence": [ + { + "description": "You are going through the security machine now.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Going through scanner behind.", + "B": "Going through scanner left.", + "C": "Right now you’re going through the security scanner, and the security scanner is directly in front of you.", + "D": "Right now you’re going through the security scanner, and it ends up directly behind you as you pass.", + "E": "Right now you’re going through the security scanner, and the scanner is directly to your right while walking through." + }, + "answer_label": "C", + "answer_text": "Right now you’re going through the security scanner, and the security scanner is directly in front of you.", + "q_id": "01434_9_T1", + "correct_answer": "Right now you’re going through the security scanner, and the security scanner is directly in front of you.", + "original_options": { + "A": "Going through scanner behind.", + "B": "Going through scanner left.", + "C": "Right now you’re going through the security scanner, and the security scanner is directly in front of you.", + "D": "Right now you’re going through the security scanner, and it ends up directly behind you as you pass.", + "E": "Right now you’re going through the security scanner, and the scanner is directly to your right while walking through." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 20 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_9_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What am I doing right now?", + "answer": "Right now you’re going through the ticket gate, and the ticket gate is directly in front of you.", + "question_chinese": "我现在在干嘛？", + "answer_chinese": "你现在在过闸机，闸机在你的正前方。", + "evidence": [ + { + "description": "You are crossing the gate.", + "timestamps": { + "start": "00:02:03", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:02:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Going through gate behind.", + "B": "Going through gate left.", + "C": "Right now you’re going through the ticket gate, and it ends up directly behind you as you pass.", + "D": "Right now you’re going through the ticket gate, and the gate is directly to your right while walking through.", + "E": "Right now you’re going through the ticket gate, and the ticket gate is directly in front of you." + }, + "answer_label": "E", + "answer_text": "Right now you’re going through the ticket gate, and the ticket gate is directly in front of you.", + "q_id": "01434_9_T2", + "correct_answer": "Right now you’re going through the ticket gate, and the ticket gate is directly in front of you.", + "original_options": { + "A": "Going through gate behind.", + "B": "Going through gate left.", + "C": "Right now you’re going through the ticket gate, and it ends up directly behind you as you pass.", + "D": "Right now you’re going through the ticket gate, and the gate is directly to your right while walking through.", + "E": "Right now you’re going through the ticket gate, and the ticket gate is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 20 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_8_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is in front of me?", + "answer": "In front of you is the security scanner.", + "question_chinese": "我面前的是什么？", + "answer_chinese": "你面前的是安检机", + "evidence": [ + { + "description": "You're looking at a security machine.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you is the security scanner.", + "B": "In front of you is the security scanner.", + "C": "To your left is the security scanner.", + "D": "To your right is the security scanner, not directly in front of you right now.", + "E": "Above you is the security scanner, rather than being in front of you." + }, + "answer_label": "B", + "answer_text": "In front of you is the security scanner.", + "q_id": "01434_8_T2", + "correct_answer": "In front of you is the security scanner.", + "original_options": { + "A": "Behind you is the security scanner.", + "B": "In front of you is the security scanner.", + "C": "To your left is the security scanner.", + "D": "To your right is the security scanner, not directly in front of you right now.", + "E": "Above you is the security scanner, rather than being in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 13 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_8_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is in front of me?", + "answer": "In front of you is the ticket gate.", + "question_chinese": "我面前的是什么？", + "answer_chinese": "你面前的是闸机。", + "evidence": [ + { + "description": "The gate is in front of you.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:02:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you is the ticket gate.", + "B": "To your left is the ticket gate.", + "C": "To your right is the ticket gate, not directly in front of you right now.", + "D": "Above you is the ticket gate, rather than being in front of you.", + "E": "In front of you is the ticket gate." + }, + "answer_label": "E", + "answer_text": "In front of you is the ticket gate.", + "q_id": "01434_8_T1", + "correct_answer": "In front of you is the ticket gate.", + "original_options": { + "A": "Behind you is the ticket gate.", + "B": "To your left is the ticket gate.", + "C": "To your right is the ticket gate, not directly in front of you right now.", + "D": "Above you is the ticket gate, rather than being in front of you.", + "E": "In front of you is the ticket gate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 13 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_9", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the subway escalator located in for me now?", + "answer": "Left side.", + "question_chinese": "现在乘地铁下去的扶梯位于我的什么方位？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The escalator is currently in front of me", + "timestamps": { + "start": "00:00:51", + "end": "00:00:53" + } + }, + { + "description": "The escalator is currently on my left", + "timestamps": { + "start": "00:01:16", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Left side.", + "B": "Right side.", + "C": "In front of you.", + "D": "It's on your right as you face the platform, not on the left.", + "E": "The escalator is behind you near the right side, so you’d need to turn around." + }, + "answer_label": "A", + "answer_text": "Left side.", + "q_id": "01434_9", + "correct_answer": "Left side.", + "original_options": { + "A": "Left side.", + "B": "Right side.", + "C": "In front of you.", + "D": "It's on your right as you face the platform, not on the left.", + "E": "The escalator is behind you near the right side, so you’d need to turn around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 13, + 16 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_10", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the subway escalator located in for me now?", + "answer": "Back side.", + "question_chinese": "现在乘地铁下去的扶梯位于我的什么方位？", + "answer_chinese": "后方", + "evidence": [ + { + "description": "The escalator is currently in front of me", + "timestamps": { + "start": "00:00:51", + "end": "00:00:53" + } + }, + { + "description": "The escalator is currently on my left", + "timestamps": { + "start": "00:01:16", + "end": "00:01:22" + } + }, + { + "description": "The escalator is currently on my left, the elevator entrance is on my right", + "timestamps": { + "start": "00:02:08", + "end": "00:02:12" + } + } + ], + "question_timestamps": "00:02:16", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Front side.", + "B": "Back side.", + "C": "Right side.", + "D": "It's off to your left-front, so it isn't located behind you right now.", + "E": "The escalator is ahead near the center area, not at the back side behind you." + }, + "answer_label": "B", + "answer_text": "Back side.", + "q_id": "01434_10", + "correct_answer": "Back side.", + "original_options": { + "A": "Front side.", + "B": "Back side.", + "C": "Right side.", + "D": "It's off to your left-front, so it isn't located behind you right now.", + "E": "The escalator is ahead near the center area, not at the back side behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_5", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have I encountered so far?", + "answer": "one.", + "question_chinese": "现在我已经遇到了几个人？", + "answer_chinese": "1个", + "evidence": [ + { + "description": "I encountered 1 person", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "one.", + "D": "You've encountered two people so far during your walk through the station.", + "E": "So far, you’ve run into three different people while moving through the area." + }, + "answer_label": "C", + "answer_text": "one.", + "q_id": "01434_5", + "correct_answer": "one.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "one.", + "D": "You've encountered two people so far during your walk through the station.", + "E": "So far, you’ve run into three different people while moving through the area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_6", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have I encountered so far?", + "answer": "Five.", + "question_chinese": "现在我已经遇到了几个人？", + "answer_chinese": "5个", + "evidence": [ + { + "description": "I encountered 1 person", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "I encountered 3 people", + "timestamps": { + "start": "00:00:18", + "end": "00:00:21" + } + }, + { + "description": "I encountered 1 person", + "timestamps": { + "start": "00:00:45", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:51", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "You have encountered four people in total up to this point in the station concourse.", + "D": "Five.", + "E": "Altogether, you’ve seen six people so far as you moved through the station." + }, + "answer_label": "D", + "answer_text": "Five.", + "q_id": "01434_6", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "You have encountered four people in total up to this point in the station concourse.", + "D": "Five.", + "E": "Altogether, you’ve seen six people so far as you moved through the station." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_7", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to exit the station now?", + "answer": "Walk straight to the end, the exit is forward on both left and right sides.", + "question_chinese": "现在我想出站应该怎么移动？", + "answer_chinese": "直走到底左右两侧往前是出站口", + "evidence": [ + { + "description": "The exit is in front left of me, the security check is in front right", + "timestamps": { + "start": "00:00:38", + "end": "00:00:40" + } + }, + { + "description": "After entering the ticket check, turning right is the elevator", + "timestamps": { + "start": "00:02:10", + "end": "00:02:14" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left to exit.", + "B": "Go back to the exit.", + "C": "Walk straight to the end, but the exit is behind you on both left and right sides.", + "D": "Walk straight a bit, then take the right corner; the exit is around the bend on your right.", + "E": "Walk straight to the end, the exit is forward on both left and right sides." + }, + "answer_label": "E", + "answer_text": "Walk straight to the end, the exit is forward on both left and right sides.", + "q_id": "01434_7", + "correct_answer": "Walk straight to the end, the exit is forward on both left and right sides.", + "original_options": { + "A": "Turn left to exit.", + "B": "Go back to the exit.", + "C": "Walk straight to the end, but the exit is behind you on both left and right sides.", + "D": "Walk straight a bit, then take the right corner; the exit is around the bend on your right.", + "E": "Walk straight to the end, the exit is forward on both left and right sides." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 18 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_8", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to exit the station now?", + "answer": "Turn right to take the elevator to the ticket hall level, after exiting the elevator turn right and walk a few steps, then turn left to the exit.", + "question_chinese": "现在我想出站应该怎么移动？", + "answer_chinese": "向右转去去坐直梯到售票大厅层出电梯右转走几步再往左走是出站口", + "evidence": [ + { + "description": "The exit is in front left of me, the security check is in front right", + "timestamps": { + "start": "00:00:38", + "end": "00:00:40" + } + }, + { + "description": "After entering the ticket check, turning right is the elevator", + "timestamps": { + "start": "00:02:10", + "end": "00:02:14" + } + }, + { + "description": "After exiting the elevator, turning right is the train waiting area", + "timestamps": { + "start": "00:02:22", + "end": "00:02:26" + } + } + ], + "question_timestamps": "00:02:47", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left to the elevator.", + "B": "Turn right, then right again.", + "C": "Turn right to take the elevator up, then turn left after exiting and walk a few steps to the exit.", + "D": "Turn right to take the elevator to the ticket hall level, after exiting the elevator turn right and walk a few steps, then turn left to the exit.", + "E": "Take the elevator by turning right, then after you get out turn right and continue to the exit." + }, + "answer_label": "D", + "answer_text": "Turn right to take the elevator to the ticket hall level, after exiting the elevator turn right and walk a few steps, then turn left to the exit.", + "q_id": "01434_8", + "correct_answer": "Turn right to take the elevator to the ticket hall level, after exiting the elevator turn right and walk a few steps, then turn left to the exit.", + "original_options": { + "A": "Turn left to the elevator.", + "B": "Turn right, then right again.", + "C": "Turn right to take the elevator up, then turn left after exiting and walk a few steps to the exit.", + "D": "Turn right to take the elevator to the ticket hall level, after exiting the elevator turn right and walk a few steps, then turn left to the exit.", + "E": "Take the elevator by turning right, then after you get out turn right and continue to the exit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 18 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_11", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans have I seen so far?", + "answer": "two.", + "question_chinese": "现在我看见了几个垃圾桶？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "There are two trash cans in front left", + "timestamps": { + "start": "00:00:13", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "two.", + "C": "Three.", + "D": "You have seen one trash can so far while moving through the station.", + "E": "So far, you’ve spotted three trash cans in total along your path." + }, + "answer_label": "B", + "answer_text": "two.", + "q_id": "01434_11", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "two.", + "C": "Three.", + "D": "You have seen one trash can so far while moving through the station.", + "E": "So far, you’ve spotted three trash cans in total along your path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "01434", + "video_length": "00:02:47", + "qid": "01434_12", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans have I seen so far?", + "answer": "Four.", + "question_chinese": "现在我看见了几个垃圾桶？", + "answer_chinese": "4个", + "evidence": [ + { + "description": "There are two trash cans in front left", + "timestamps": { + "start": "00:00:13", + "end": "00:00:16" + } + }, + { + "description": "There are two trash cans far ahead", + "timestamps": { + "start": "00:02:25", + "end": "00:02:27" + } + } + ], + "question_timestamps": "00:02:47", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "You have seen three trash cans so far during your time in the station.", + "D": "Four.", + "E": "Up to now, you've counted five trash cans placed along the route you walked." + }, + "answer_label": "D", + "answer_text": "Four.", + "q_id": "01434_12", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "You have seen three trash cans so far during your time in the station.", + "D": "Four.", + "E": "Up to now, you've counted five trash cans placed along the route you walked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + } +] diff --git a/QAs/01435-mcq.json b/QAs/01435-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..311011053058118756c5a0c301983d67547d3a1c --- /dev/null +++ b/QAs/01435-mcq.json @@ -0,0 +1,1047 @@ +[ + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_9", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned since exiting the station?", + "answer": "two times.", + "question_chinese": "从出站口到现在我已经转了几次弯？", + "answer_chinese": "2次", + "evidence": [ + { + "description": "Turned left once", + "timestamps": { + "start": "00:02:58", + "end": "00:03:00" + } + }, + { + "description": "Turned left again", + "timestamps": { + "start": "00:03:02", + "end": "00:03:04" + } + } + ], + "question_timestamps": "00:03:06", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One time.", + "B": "Three times.", + "C": "You have turned three times since exiting the station, changing direction at three different points.", + "D": "two times.", + "E": "You have turned four times since leaving the station, making four separate direction changes along the way." + }, + "answer_label": "D", + "answer_text": "two times.", + "q_id": "01435_9", + "correct_answer": "two times.", + "original_options": { + "A": "One time.", + "B": "Three times.", + "C": "You have turned three times since exiting the station, changing direction at three different points.", + "D": "two times.", + "E": "You have turned four times since leaving the station, making four separate direction changes along the way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 17 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_10", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned since exiting the station?", + "answer": "five times.", + "question_chinese": "从出站口到现在我已经转了几次弯？", + "answer_chinese": "5次", + "evidence": [ + { + "description": "Turned left once", + "timestamps": { + "start": "00:02:58", + "end": "00:03:00" + } + }, + { + "description": "Turned left again", + "timestamps": { + "start": "00:03:02", + "end": "00:03:04" + } + }, + { + "description": "Turned left again to go outside, then turned right once and walked a few steps to the stairs, then turned left", + "timestamps": { + "start": "00:03:08", + "end": "00:03:12" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "five times.", + "B": "Four times.", + "C": "Six times.", + "D": "You have turned three times since exiting the station, with three distinct direction changes along your route.", + "E": "You have turned seven times since leaving the station, making seven separate turns before now." + }, + "answer_label": "A", + "answer_text": "five times.", + "q_id": "01435_10", + "correct_answer": "five times.", + "original_options": { + "A": "five times.", + "B": "Four times.", + "C": "Six times.", + "D": "You have turned three times since exiting the station, with three distinct direction changes along your route.", + "E": "You have turned seven times since leaving the station, making seven separate turns before now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 15 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_8_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were in the elevator I took just now?", + "answer": "There were originally three people in the elevator in front of you.", + "question_chinese": "刚才我坐的电梯里有几个人？", + "answer_chinese": "你面前的电梯里原有三人。", + "evidence": [ + { + "description": "There are 2 people in the elevator plus me makes 5 people", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There were originally two people.", + "B": "There were originally four people.", + "C": "There were originally five people in the elevator in front of you when you got on.", + "D": "There were originally three people in the elevator in front of you.", + "E": "There was originally one person in the elevator in front of you before you entered." + }, + "answer_label": "D", + "answer_text": "There were originally three people in the elevator in front of you.", + "q_id": "01435_8_T2", + "correct_answer": "There were originally three people in the elevator in front of you.", + "original_options": { + "A": "There were originally two people.", + "B": "There were originally four people.", + "C": "There were originally five people in the elevator in front of you when you got on.", + "D": "There were originally three people in the elevator in front of you.", + "E": "There was originally one person in the elevator in front of you before you entered." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_8_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were in the elevator I took just now?", + "answer": "There are five people in the elevator in front of you.", + "question_chinese": "刚才我坐的电梯里有几个人？", + "answer_chinese": "你面前的电梯里有5个人", + "evidence": [ + { + "description": "There are 4 people in the elevator plus me makes 5 people", + "timestamps": { + "start": "00:02:49", + "end": "00:02:52" + } + } + ], + "question_timestamps": "00:03:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are four people.", + "B": "There are six people.", + "C": "There are five people in the elevator in front of you.", + "D": "There are three people in the elevator in front of you as you look inside right now.", + "E": "There are seven people in the elevator in front of you at the moment you approach it." + }, + "answer_label": "C", + "answer_text": "There are five people in the elevator in front of you.", + "q_id": "01435_8_T1", + "correct_answer": "There are five people in the elevator in front of you.", + "original_options": { + "A": "There are four people.", + "B": "There are six people.", + "C": "There are five people in the elevator in front of you.", + "D": "There are three people in the elevator in front of you as you look inside right now.", + "E": "There are seven people in the elevator in front of you at the moment you approach it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_15", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I turn when entering the elevator just now?", + "answer": "Turn right.", + "question_chinese": "刚才进直梯我是怎么转进电梯的？", + "answer_chinese": "右转", + "evidence": [ + { + "description": "Turned right to enter the elevator", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left.", + "B": "Go straight in.", + "C": "Turn slightly left as you enter, keeping your body facing the elevator doors.", + "D": "Turn right.", + "E": "Turn around before entering the elevator, then step in facing the opposite direction." + }, + "answer_label": "D", + "answer_text": "Turn right.", + "q_id": "01435_15", + "correct_answer": "Turn right.", + "original_options": { + "A": "Turn left.", + "B": "Go straight in.", + "C": "Turn slightly left as you enter, keeping your body facing the elevator doors.", + "D": "Turn right.", + "E": "Turn around before entering the elevator, then step in facing the opposite direction." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 13 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_16", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I turn when entering the elevator just now?", + "answer": "Turn left.", + "question_chinese": "刚才进直梯我是怎么转进电梯的？", + "answer_chinese": "左转", + "evidence": [ + { + "description": "Turned left to enter the elevator", + "timestamps": { + "start": "00:02:44", + "end": "00:02:47" + } + } + ], + "question_timestamps": "00:03:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right.", + "B": "Go straight in.", + "C": "Turn slightly right as you enter the elevator, aligning yourself with the control panel.", + "D": "Turn left.", + "E": "Turn around before entering the elevator so you step in while facing back toward the hallway." + }, + "answer_label": "D", + "answer_text": "Turn left.", + "q_id": "01435_16", + "correct_answer": "Turn left.", + "original_options": { + "A": "Turn right.", + "B": "Go straight in.", + "C": "Turn slightly right as you enter the elevator, aligning yourself with the control panel.", + "D": "Turn left.", + "E": "Turn around before entering the elevator so you step in while facing back toward the hallway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 14, + 16 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_7", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move to take the elevator after getting off the subway?", + "answer": "First turn right and walk a few steps, then turn left and walk around the elevator a few steps, then turn left to the elevator entrance.", + "question_chinese": "我下地铁后要坐直梯应该怎么移动", + "answer_chinese": "先右转走几步再左转绕电梯走几步再左转至电梯口", + "evidence": [ + { + "description": "The elevator is in front right of me, and the elevator direction is the same as the direction I got off the subway", + "timestamps": { + "start": "00:02:33", + "end": "00:02:36" + } + } + ], + "question_timestamps": "00:02:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go straight to the elevator.", + "B": "First turn right and walk a few steps, then turn left and walk around the elevator a few steps, then turn left to the elevator entrance.", + "C": "Turn left, then left again.", + "D": "First turn right, then right around the elevator, then right to enter the elevator.", + "E": "First turn left, walk a few steps, then turn right around the elevator, then left to enter." + }, + "answer_label": "B", + "answer_text": "First turn right and walk a few steps, then turn left and walk around the elevator a few steps, then turn left to the elevator entrance.", + "q_id": "01435_7", + "correct_answer": "First turn right and walk a few steps, then turn left and walk around the elevator a few steps, then turn left to the elevator entrance.", + "original_options": { + "A": "Go straight to the elevator.", + "B": "First turn right and walk a few steps, then turn left and walk around the elevator a few steps, then turn left to the elevator entrance.", + "C": "Turn left, then left again.", + "D": "First turn right, then right around the elevator, then right to enter the elevator.", + "E": "First turn left, walk a few steps, then turn right around the elevator, then left to enter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 17 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_4_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I get off the subway, how should I move to take the elevator?", + "answer": "First turn right and walk a few steps, then turn left and walk around the elevator for a few steps, then turn left again to the elevator entrance.", + "question_chinese": "我下地铁后要坐直梯应该怎么移动？", + "answer_chinese": "你先右转走几步再左转绕电梯走几步再左转至电梯口", + "evidence": [ + { + "description": "The elevator is in front right of me, and the elevator direction is the same as the direction I got off the subway", + "timestamps": { + "start": "00:02:33", + "end": "00:02:36" + } + } + ], + "question_timestamps": "00:02:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First turn right and walk a few steps, then turn left and walk around the elevator for a few steps, then turn left again to the elevator entrance.", + "B": "Go straight to the elevator.", + "C": "Turn left, then right.", + "D": "First turn right, then turn right around the elevator, then turn right to the elevator entrance.", + "E": "First turn left, walk a few steps, then turn left around the elevator, then turn right to enter." + }, + "answer_label": "A", + "answer_text": "First turn right and walk a few steps, then turn left and walk around the elevator for a few steps, then turn left again to the elevator entrance.", + "q_id": "01435_4_T1", + "correct_answer": "First turn right and walk a few steps, then turn left and walk around the elevator for a few steps, then turn left again to the elevator entrance.", + "original_options": { + "A": "First turn right and walk a few steps, then turn left and walk around the elevator for a few steps, then turn left again to the elevator entrance.", + "B": "Go straight to the elevator.", + "C": "Turn left, then right.", + "D": "First turn right, then turn right around the elevator, then turn right to the elevator entrance.", + "E": "First turn left, walk a few steps, then turn left around the elevator, then turn right to enter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 18 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_4_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I get off the subway, how should I move to take the elevator?", + "answer": "The elevator is in front of you—go straight from where you are.", + "question_chinese": "我下地铁后要坐直梯应该怎么移动？", + "answer_chinese": "直梯在你的前方，从你现在的位置直行。", + "evidence": [ + { + "description": "The elevator is in front right of me, and the elevator direction is the same as the direction I got off the subway", + "timestamps": { + "start": "00:02:33", + "end": "00:02:36" + } + } + ], + "question_timestamps": "00:03:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left to reach the elevator.", + "B": "Turn right to reach the elevator.", + "C": "The elevator is behind you, so turn around and walk straight from where you got off.", + "D": "The elevator is in front of you—go straight from where you are.", + "E": "The elevator is off to your left; walk a few steps forward, then turn left to it." + }, + "answer_label": "D", + "answer_text": "The elevator is in front of you—go straight from where you are.", + "q_id": "01435_4_T2", + "correct_answer": "The elevator is in front of you—go straight from where you are.", + "original_options": { + "A": "Turn left to reach the elevator.", + "B": "Turn right to reach the elevator.", + "C": "The elevator is behind you, so turn around and walk straight from where you got off.", + "D": "The elevator is in front of you—go straight from where you are.", + "E": "The elevator is off to your left; walk a few steps forward, then turn left to it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_5_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I get out of the elevator, how should I move to take the subway?", + "answer": "The subway is in front of you. After exiting the elevator, go straight a few steps—the subway is right ahead.", + "question_chinese": "我下直梯后要坐地铁应该怎么移动？", + "answer_chinese": "地铁在你面前，出电梯直走几步，地铁就在前方。", + "evidence": [ + { + "description": "The subway direction I need to take is in front left of me", + "timestamps": { + "start": "00:00:37", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left after exiting the elevator.", + "B": "Turn right after leaving the elevator.", + "C": "The subway is behind you; step out of the elevator, turn around, and walk straight to it.", + "D": "The subway is in front of you. After exiting the elevator, go straight a few steps—the subway is right ahead.", + "E": "The subway is to your right; walk forward a few steps, then turn right to reach it." + }, + "answer_label": "D", + "answer_text": "The subway is in front of you. After exiting the elevator, go straight a few steps—the subway is right ahead.", + "q_id": "01435_5_T2", + "correct_answer": "The subway is in front of you. After exiting the elevator, go straight a few steps—the subway is right ahead.", + "original_options": { + "A": "Turn left after exiting the elevator.", + "B": "Turn right after leaving the elevator.", + "C": "The subway is behind you; step out of the elevator, turn around, and walk straight to it.", + "D": "The subway is in front of you. After exiting the elevator, go straight a few steps—the subway is right ahead.", + "E": "The subway is to your right; walk forward a few steps, then turn right to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_5_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I get out of the elevator, how should I move to take the subway?", + "answer": "The subway is to your left. After exiting the elevator, go straight a few steps, then turn left and walk a few steps to take the subway.", + "question_chinese": "我下直梯后要坐地铁应该怎么移动？", + "answer_chinese": "地铁在你左边，出电梯直走几步再左转走几步就可以乘坐地铁", + "evidence": [ + { + "description": "The subway direction I need to take is in front left of me", + "timestamps": { + "start": "00:00:37", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:02:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go straight to the subway.", + "B": "Turn right to reach the subway.", + "C": "The subway is to your left. After exiting the elevator, go straight a few steps, then turn left and walk a few steps to take the subway.", + "D": "The subway is in front of you; after exiting the elevator, keep walking straight a few steps to it.", + "E": "The subway is behind you; step out, turn around, and walk a few steps to take it." + }, + "answer_label": "C", + "answer_text": "The subway is to your left. After exiting the elevator, go straight a few steps, then turn left and walk a few steps to take the subway.", + "q_id": "01435_5_T1", + "correct_answer": "The subway is to your left. After exiting the elevator, go straight a few steps, then turn left and walk a few steps to take the subway.", + "original_options": { + "A": "Go straight to the subway.", + "B": "Turn right to reach the subway.", + "C": "The subway is to your left. After exiting the elevator, go straight a few steps, then turn left and walk a few steps to take the subway.", + "D": "The subway is in front of you; after exiting the elevator, keep walking straight a few steps to it.", + "E": "The subway is behind you; step out, turn around, and walk a few steps to take it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 17 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_3_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From exiting the subway station to getting into a taxi, can I make it in 50 steps?", + "answer": "Yes—you can. The taxi is in front of you.", + "question_chinese": "我从出了地铁站到坐上出租车能50步到达吗？", + "answer_chinese": "可以，出租车在你的前方。", + "evidence": [ + { + "description": "From exiting the door until now, I have walked more than 10 steps and haven't reached the taxi yet", + "timestamps": { + "start": "00:03:08", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:03:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. The taxi is in front of you.", + "B": "No—you can't. The taxi is far away in front of you." + }, + "answer_label": "A", + "answer_text": "Yes—you can. The taxi is in front of you.", + "q_id": "01435_3_T2", + "correct_answer": "Yes—you can. The taxi is in front of you.", + "original_options": { + "A": "Yes—you can. The taxi is in front of you.", + "B": "No—you can't. The taxi is far away in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_3_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From exiting the subway station to getting into a taxi, can I make it in 50 steps?", + "answer": "No, the taxi was far away in front of you.", + "question_chinese": "我从出了地铁站到坐上出租车能50步到达吗？", + "answer_chinese": "不可以，出租车在你很远的前方。", + "evidence": [ + { + "description": "From exiting the door until now, I have walked more than 10 steps and haven't reached the taxi yet", + "timestamps": { + "start": "00:03:08", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:03:32", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. The taxi is in front of you.", + "B": "No, the taxi was far away in front of you." + }, + "answer_label": "B", + "answer_text": "No, the taxi was far away in front of you.", + "q_id": "01435_3_T1", + "correct_answer": "No, the taxi was far away in front of you.", + "original_options": { + "A": "Yes—you can. The taxi is in front of you.", + "B": "No, the taxi was far away in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_3", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which floor am I on now?", + "answer": "B3.", + "question_chinese": "现在我在几楼？", + "answer_chinese": "-3楼", + "evidence": [ + { + "description": "Now I am in the elevator, the elevator is going to -3rd floor", + "timestamps": { + "start": "00:00:07", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "B2.", + "B": "B4.", + "C": "You are on B1, one level above B2, not down on the B3 basement level.", + "D": "B3.", + "E": "You are on the 1st basement level, closer to street level than B3." + }, + "answer_label": "D", + "answer_text": "B3.", + "q_id": "01435_3", + "correct_answer": "B3.", + "original_options": { + "A": "B2.", + "B": "B4.", + "C": "You are on B1, one level above B2, not down on the B3 basement level.", + "D": "B3.", + "E": "You are on the 1st basement level, closer to street level than B3." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 13 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_4", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which floor am I on now?", + "answer": "1st floor.", + "question_chinese": "现在我在几楼？", + "answer_chinese": "1楼", + "evidence": [ + { + "description": "Now I am in the elevator, the elevator is going to 1st floor", + "timestamps": { + "start": "00:02:51", + "end": "00:02:53" + } + }, + { + "description": "Now I am in the elevator, the elevator is going to 2nd floor", + "timestamps": { + "start": "00:03:05", + "end": "00:03:07" + } + } + ], + "question_timestamps": "00:03:09", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Ground floor.", + "B": "1st floor.", + "C": "Second floor.", + "D": "You are on the basement level, below the street-level entrance, not up on the 1st floor.", + "E": "You are on the third floor, several levels above where you entered, not on the 1st floor." + }, + "answer_label": "B", + "answer_text": "1st floor.", + "q_id": "01435_4", + "correct_answer": "1st floor.", + "original_options": { + "A": "Ground floor.", + "B": "1st floor.", + "C": "Second floor.", + "D": "You are on the basement level, below the street-level entrance, not up on the 1st floor.", + "E": "You are on the third floor, several levels above where you entered, not on the 1st floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_11", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people pushing carts have I encountered so far?", + "answer": "one.", + "question_chinese": "现在我已经碰到了几个推小推车的人？", + "answer_chinese": "1个", + "evidence": [ + { + "description": "There is a person pulling a cart in the elevator", + "timestamps": { + "start": "00:00:30", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four people pushing carts in total so far, which is more than the one you encountered.", + "D": "Five people pushing carts so far, meaning you would have passed several cart pushers already.", + "E": "one." + }, + "answer_label": "E", + "answer_text": "one.", + "q_id": "01435_11", + "correct_answer": "one.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four people pushing carts in total so far, which is more than the one you encountered.", + "D": "Five people pushing carts so far, meaning you would have passed several cart pushers already.", + "E": "one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_12", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people pushing carts have I encountered so far?", + "answer": "Two.", + "question_chinese": "现在我已经碰到了几个推小推车的人？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "There is a person pulling a cart in the elevator", + "timestamps": { + "start": "00:00:30", + "end": "00:00:33" + } + }, + { + "description": "There is a mother pushing a stroller", + "timestamps": { + "start": "00:00:56", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Four people pushing carts so far, including two you passed close by in the hallway.", + "E": "Five people pushing carts so far, with several encountered while moving between the aisles." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "01435_12", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Four people pushing carts so far, including two you passed close by in the hallway.", + "E": "Five people pushing carts so far, with several encountered while moving between the aisles." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_1_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many subway staff members have guided me so far?", + "answer": "There are two staff members in front of you guiding you right now.", + "question_chinese": "现在有多少个地铁工作人员为我带了路？", + "answer_chinese": "你面前现在有两个工作人员为你带路。", + "evidence": [ + { + "description": "Now there is a staff member taking me to the subway", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "Now changed to another staff member", + "timestamps": { + "start": "00:00:47", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One staff member.", + "B": "There are two staff members in front of you guiding you right now.", + "C": "Three staff members.", + "D": "There is one staff member in front of you guiding you right now.", + "E": "There are three staff members in front of you guiding you right now." + }, + "answer_label": "B", + "answer_text": "There are two staff members in front of you guiding you right now.", + "q_id": "01435_1_T2", + "correct_answer": "There are two staff members in front of you guiding you right now.", + "original_options": { + "A": "One staff member.", + "B": "There are two staff members in front of you guiding you right now.", + "C": "Three staff members.", + "D": "There is one staff member in front of you guiding you right now.", + "E": "There are three staff members in front of you guiding you right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 13 + ] + }, + { + "video_uid": "01435", + "video_length": "00:03:32", + "qid": "01435_1_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many subway staff members have guided me so far?", + "answer": "Altogether, there were 3 subway staff members in front of you guiding you.", + "question_chinese": "现在有多少个地铁工作人员为我带了路？", + "answer_chinese": "你面前累计有3个地铁工作人员为你带路", + "evidence": [ + { + "description": "Now there is a staff member taking me to the subway", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "Now changed to another staff member", + "timestamps": { + "start": "00:00:47", + "end": "00:00:50" + } + }, + { + "description": "Now changed to another staff member", + "timestamps": { + "start": "00:03:29", + "end": "00:03:30" + } + } + ], + "question_timestamps": "00:03:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Altogether, there were 3 subway staff members in front of you guiding you.", + "B": "Two staff members.", + "C": "Four staff members.", + "D": "Altogether, there were 2 subway staff members in front of you guiding you.", + "E": "Altogether, there were 4 subway staff members in front of you guiding you." + }, + "answer_label": "A", + "answer_text": "Altogether, there were 3 subway staff members in front of you guiding you.", + "q_id": "01435_1_T1", + "correct_answer": "Altogether, there were 3 subway staff members in front of you guiding you.", + "original_options": { + "A": "Altogether, there were 3 subway staff members in front of you guiding you.", + "B": "Two staff members.", + "C": "Four staff members.", + "D": "Altogether, there were 2 subway staff members in front of you guiding you.", + "E": "Altogether, there were 4 subway staff members in front of you guiding you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 13 + ] + } +] diff --git a/QAs/01436-mcq.json b/QAs/01436-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b487f68af580b81e4d2b0cc87d9de9256cd372ab --- /dev/null +++ b/QAs/01436-mcq.json @@ -0,0 +1,203 @@ +[ + { + "video_uid": "01436", + "video_length": "00:02:07", + "qid": "01436_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many electric bicycles have I passed by so far?", + "answer": "three.", + "question_chinese": "现在我已经经过了多少辆电动自行车？", + "answer_chinese": "3辆", + "evidence": [ + { + "description": "There are 3 electric bicycles on the left side", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "three.", + "B": "Two.", + "C": "Four.", + "D": "You have passed by two electric bicycles so far on your route.", + "E": "You have passed by four electric bicycles so far along your route." + }, + "answer_label": "A", + "answer_text": "three.", + "q_id": "01436_1", + "correct_answer": "three.", + "original_options": { + "A": "three.", + "B": "Two.", + "C": "Four.", + "D": "You have passed by two electric bicycles so far on your route.", + "E": "You have passed by four electric bicycles so far along your route." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "01436", + "video_length": "00:02:07", + "qid": "01436_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many electric bicycles have I passed by so far?", + "answer": "Four.", + "question_chinese": "现在我已经经过了多少辆电动自行车？", + "answer_chinese": "4辆", + "evidence": [ + { + "description": "There are 3 electric bicycles on the left side", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + }, + { + "description": "There is 1 electric bicycle on the left side", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "You have passed by three electric bicycles so far while moving forward.", + "E": "You have passed by five electric bicycles so far while moving forward." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "01436_2", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "You have passed by three electric bicycles so far while moving forward.", + "E": "You have passed by five electric bicycles so far while moving forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "01436", + "video_length": "00:02:07", + "qid": "01436_10", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the napkin directly if I want to use it now?", + "answer": "Yes.", + "question_chinese": "现在我想用餐巾纸我能直接够到吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "There are toothpicks on the far right side of the table", + "timestamps": { + "start": "00:00:19", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01436_10", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01436", + "video_length": "00:02:07", + "qid": "01436_9", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the napkin directly if I want to use it now?", + "answer": "No.", + "question_chinese": "现在我想用餐巾纸我能直接够到吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "There is no napkin at my seat position, there are napkins on the table in front of me", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "01436_9", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/01442-mcq.json b/QAs/01442-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5f2f21609ecb90ab41337ce71179b67a41db24ae --- /dev/null +++ b/QAs/01442-mcq.json @@ -0,0 +1,859 @@ +[ + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Who has helped me from the beginning until now?", + "answer": "Staff of Bawang Tea Princess.", + "question_chinese": "从一开始到现在有谁帮助我了？", + "answer_chinese": "霸王茶姬的工作人员", + "evidence": [ + { + "description": "Staff of Bawang Tea Princess", + "timestamps": { + "start": "00:00:36", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Staff of Bawang Tea King.", + "B": "Crew of Bawang Tea Empress.", + "C": "You have been helped by staff of Bawang Tea King from the beginning until now.", + "D": "Staff of Bawang Tea Princess.", + "E": "You have been helped by a group from Bawang Tea Master throughout the entire time so far." + }, + "answer_label": "D", + "answer_text": "Staff of Bawang Tea Princess.", + "q_id": "01442_7", + "correct_answer": "Staff of Bawang Tea Princess.", + "original_options": { + "A": "Staff of Bawang Tea King.", + "B": "Crew of Bawang Tea Empress.", + "C": "You have been helped by staff of Bawang Tea King from the beginning until now.", + "D": "Staff of Bawang Tea Princess.", + "E": "You have been helped by a group from Bawang Tea Master throughout the entire time so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_6", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Who has helped me from the beginning until now?", + "answer": "A taxi driver and Staff of Bawang Tea Princess.", + "question_chinese": "从一开始到现在有谁帮助我了？", + "answer_chinese": "出租车司机和霸王茶姬的工作人员", + "evidence": [ + { + "description": "A taxi driver helped me.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:28" + } + }, + { + "description": "The staff at Bawang Chaji helped me.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:02:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A motorcycle driver and Bawang Tea Princess staff.", + "B": "A taxi driver and the coffee shop staff.", + "C": "A taxi driver and staff from Bawang Tea Garden helped you from the beginning all the way until now.", + "D": "A bus driver and the staff at Bawang Tea Princess have been the ones helping you throughout.", + "E": "A taxi driver and Staff of Bawang Tea Princess." + }, + "answer_label": "E", + "answer_text": "A taxi driver and Staff of Bawang Tea Princess.", + "q_id": "01442_6", + "correct_answer": "A taxi driver and Staff of Bawang Tea Princess.", + "original_options": { + "A": "A motorcycle driver and Bawang Tea Princess staff.", + "B": "A taxi driver and the coffee shop staff.", + "C": "A taxi driver and staff from Bawang Tea Garden helped you from the beginning all the way until now.", + "D": "A bus driver and the staff at Bawang Tea Princess have been the ones helping you throughout.", + "E": "A taxi driver and Staff of Bawang Tea Princess." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 17 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_13", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Approximately how long has it been since we started walking along the roadside?", + "answer": "nine minutes.", + "question_chinese": "从沿路边开始走到现在大概过了多久？", + "answer_chinese": "9分钟", + "evidence": [ + { + "description": "I know through my phone that it is now 4:08 PM.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "I know through my phone that it is now 4:17 PM.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:01:21", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Seven minutes.", + "B": "nine minutes.", + "C": "Eleven minutes.", + "D": "About eight minutes have passed since you started walking along the roadside.", + "E": "Roughly ten minutes have passed since you began walking beside the road." + }, + "answer_label": "B", + "answer_text": "nine minutes.", + "q_id": "01442_13", + "correct_answer": "nine minutes.", + "original_options": { + "A": "Seven minutes.", + "B": "nine minutes.", + "C": "Eleven minutes.", + "D": "About eight minutes have passed since you started walking along the roadside.", + "E": "Roughly ten minutes have passed since you began walking beside the road." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_12", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Approximately how long has it been since we started walking along the roadside?", + "answer": "twenty-four minutes.", + "question_chinese": "从沿路边开始走到现在大概过了多久？", + "answer_chinese": "24分钟", + "evidence": [ + { + "description": "I know through my phone that it is now 4:08 PM.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "I know through my phone that it is now 4:32 PM.", + "timestamps": { + "start": "00:01:49", + "end": "00:01:50" + } + }, + { + "description": "I know through my phone that it is now 4:17 PM.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:01:51", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Fourteen minutes.", + "B": "Thirty-two minutes.", + "C": "twenty-four minutes.", + "D": "About eighteen minutes have passed since you started walking along the roadside.", + "E": "Roughly twenty minutes have passed since you began walking beside the road." + }, + "answer_label": "C", + "answer_text": "twenty-four minutes.", + "q_id": "01442_12", + "correct_answer": "twenty-four minutes.", + "original_options": { + "A": "Fourteen minutes.", + "B": "Thirty-two minutes.", + "C": "twenty-four minutes.", + "D": "About eighteen minutes have passed since you started walking along the roadside.", + "E": "Roughly twenty minutes have passed since you began walking beside the road." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Since the stone pier, how many times have I turned around until now?", + "answer": "once.", + "question_chinese": "从石墩开始到现在我转身了几次？", + "answer_chinese": "1次", + "evidence": [ + { + "description": "I turned left once.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Never.", + "B": "Twice.", + "C": "once.", + "D": "You turned around three times since the stone pier up to now.", + "E": "You’ve turned around four times since passing the stone pier until now." + }, + "answer_label": "C", + "answer_text": "once.", + "q_id": "01442_3", + "correct_answer": "once.", + "original_options": { + "A": "Never.", + "B": "Twice.", + "C": "once.", + "D": "You turned around three times since the stone pier up to now.", + "E": "You’ve turned around four times since passing the stone pier until now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_16", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Since the stone pier, how many times have I turned around until now?", + "answer": "four times.", + "question_chinese": "从石墩开始到现在我转身了几次？", + "answer_chinese": "4次", + "evidence": [ + { + "description": "At the bus stop seat, he turned to the left again and sat down.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:49" + } + }, + { + "description": "I turned around and got in the car.", + "timestamps": { + "start": "00:01:31", + "end": "00:01:36" + } + }, + { + "description": "I turned to face the road.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:23" + } + }, + { + "description": "I turned around and walked back toward the car.", + "timestamps": { + "start": "00:01:21", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:01:41", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "four times.", + "C": "Seven times.", + "D": "You turned around three times since the stone pier up to now.", + "E": "You’ve turned around five times since passing the stone pier until now." + }, + "answer_label": "B", + "answer_text": "four times.", + "q_id": "01442_16", + "correct_answer": "four times.", + "original_options": { + "A": "Two times.", + "B": "four times.", + "C": "Seven times.", + "D": "You turned around three times since the stone pier up to now.", + "E": "You’ve turned around five times since passing the stone pier until now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_8_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many people are in front of me?", + "answer": "There are three people in front of you.", + "question_chinese": "我前面有多少个人？", + "answer_chinese": "你前面有三个人。", + "evidence": [ + { + "description": "There are three people in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There is one boy in front of you.", + "B": "There are three people in front of you." + }, + "answer_label": "B", + "answer_text": "There are three people in front of you.", + "q_id": "01442_8_T2", + "correct_answer": "There are three people in front of you.", + "original_options": { + "A": "There is one boy in front of you.", + "B": "There are three people in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_8_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many people are in front of me?", + "answer": "There is one boy in front of you.", + "question_chinese": "我前面有多少个人？", + "answer_chinese": "你前面有一位男生。", + "evidence": [ + { + "description": "There is a boy in front of you.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:51" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There is one boy in front of you.", + "B": "There are three people in front of you." + }, + "answer_label": "A", + "answer_text": "There is one boy in front of you.", + "q_id": "01442_8_T1", + "correct_answer": "There is one boy in front of you.", + "original_options": { + "A": "There is one boy in front of you.", + "B": "There are three people in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_7_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I throw the trash directly into the trash can from where I am?", + "answer": "Yes—you can. The trash can is to your right.", + "question_chinese": "我可以直接丢垃圾在垃圾桶里吗？", + "answer_chinese": "可以，垃圾桶在我的右边", + "evidence": [ + { + "description": "The trash is very close to the bus station; I'm at the bus stop now.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:38", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. The trash can is to your right.", + "B": "No—you can't. The trash can is to your right." + }, + "answer_label": "A", + "answer_text": "Yes—you can. The trash can is to your right.", + "q_id": "01442_7_T2", + "correct_answer": "Yes—you can. The trash can is to your right.", + "original_options": { + "A": "Yes—you can. The trash can is to your right.", + "B": "No—you can't. The trash can is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_7_T1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I throw the trash directly into the trash can from where I am?", + "answer": "No—you can't. The trash can is to your front-right.", + "question_chinese": "我可以直接丢垃圾在垃圾桶里吗？", + "answer_chinese": "不可以，垃圾桶在你的右前方。", + "evidence": [ + { + "description": "It's not allowed because there's no trash can.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:01:31", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. The trash can is far behind you.", + "B": "No—you can't. The trash can is to your front-right." + }, + "answer_label": "B", + "answer_text": "No—you can't. The trash can is to your front-right.", + "q_id": "01442_7_T1", + "correct_answer": "No—you can't. The trash can is to your front-right.", + "original_options": { + "A": "Yes—you can. The trash can is far behind you.", + "B": "No—you can't. The trash can is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_9", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Walking along the road, I passed by several cars parked on the roadside.", + "answer": "two.", + "question_chinese": "沿马路走过来，我经过了几辆在路旁停的车？", + "answer_chinese": "2辆", + "evidence": [ + { + "description": "Passed by two white cars", + "timestamps": { + "start": "00:00:21", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "You passed three parked cars along the roadside while walking down the road.", + "D": "two.", + "E": "You passed four parked cars lined up along the roadside as you walked." + }, + "answer_label": "D", + "answer_text": "two.", + "q_id": "01442_9", + "correct_answer": "two.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "You passed three parked cars along the roadside while walking down the road.", + "D": "two.", + "E": "You passed four parked cars lined up along the roadside as you walked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_8", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Walking along the road, I passed by several cars parked on the roadside.", + "answer": "three.", + "question_chinese": "沿马路走过来，我经过了几辆在路旁停的车？", + "answer_chinese": "3辆", + "evidence": [ + { + "description": "I passed by two white cars.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:32" + } + }, + { + "description": "I passed by a green taxi.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "three.", + "B": "Two.", + "C": "Four.", + "D": "You passed two parked cars along the roadside while walking down the road.", + "E": "You passed four cars parked on the roadside while walking down the road." + }, + "answer_label": "A", + "answer_text": "three.", + "q_id": "01442_8", + "correct_answer": "three.", + "original_options": { + "A": "three.", + "B": "Two.", + "C": "Four.", + "D": "You passed two parked cars along the roadside while walking down the road.", + "E": "You passed four cars parked on the roadside while walking down the road." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the door handle in the back seat now?", + "answer": "No.", + "question_chinese": "现在我能够到车后排的门把手吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "It's still a bit far from the door, I can't reach the doorknob.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:23", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01442_1", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the door handle in the back seat now?", + "answer": "Yes.", + "question_chinese": "现在我能够到车后排的门把手吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "Now you can touch the door handle inside the car.", + "timestamps": { + "start": "00:01:28", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:34", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "01442_2", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_5", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is KFC from my location now?", + "answer": "KFC is on my front right side.", + "question_chinese": "现在肯德基位于我的什么方位？", + "answer_chinese": "肯德基在我的前方右侧", + "evidence": [ + { + "description": "KFC is on my front right side.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "KFC is on my left.", + "B": "KFC is on my front right side.", + "C": "KFC is straight ahead.", + "D": "KFC is on my back-left side from where I’m standing now.", + "E": "KFC is on my right side, not in front, relative to my current position." + }, + "answer_label": "B", + "answer_text": "KFC is on my front right side.", + "q_id": "01442_5", + "correct_answer": "KFC is on my front right side.", + "original_options": { + "A": "KFC is on my left.", + "B": "KFC is on my front right side.", + "C": "KFC is straight ahead.", + "D": "KFC is on my back-left side from where I’m standing now.", + "E": "KFC is on my right side, not in front, relative to my current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 12, + 14 + ] + }, + { + "video_uid": "01442", + "video_length": "00:02:08", + "qid": "01442_4", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is KFC from my location now?", + "answer": "KFC is right behind me.", + "question_chinese": "现在肯德基位于我的什么方位？", + "answer_chinese": "肯德基在我的正后方", + "evidence": [ + { + "description": "KFC is right behind the station.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:29" + } + }, + { + "description": "I am at the station right now.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "KFC is in front of me.", + "B": "KFC is on my left side.", + "C": "KFC is on my front-right side from where I’m standing now.", + "D": "KFC is right behind me.", + "E": "KFC is behind me to the left, not directly behind my current position." + }, + "answer_label": "D", + "answer_text": "KFC is right behind me.", + "q_id": "01442_4", + "correct_answer": "KFC is right behind me.", + "original_options": { + "A": "KFC is in front of me.", + "B": "KFC is on my left side.", + "C": "KFC is on my front-right side from where I’m standing now.", + "D": "KFC is right behind me.", + "E": "KFC is behind me to the left, not directly behind my current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 13 + ] + } +] diff --git a/QAs/01443-mcq.json b/QAs/01443-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..90ac7251d8791589e4c765ad165ae06eac473748 --- /dev/null +++ b/QAs/01443-mcq.json @@ -0,0 +1,750 @@ +[ + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_9", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to go to the bathroom, can I get there within 10 steps?", + "answer": "Yes.", + "question_chinese": "如果我现在想去上厕所能在10步内到达吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "There is a bathroom here.", + "timestamps": { + "start": "00:01:13", + "end": "00:01:17" + } + }, + { + "description": "I walked about 8 steps.", + "timestamps": { + "start": "00:01:17", + "end": "00:01:19" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01443_9", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_10", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to go to the bathroom, can I get there within 10 steps?", + "answer": "No.", + "question_chinese": "如果我现在想去上厕所能在10步内到达吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "There is a bathroom here.", + "timestamps": { + "start": "00:01:13", + "end": "00:01:17" + } + }, + { + "description": "After passing by the side, I had already walked dozens of steps.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:01:31", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01443_10", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_2_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cars did I run into at the doorway?", + "answer": "You encountered one car. It passed in front of you and is now to your left.", + "question_chinese": "我在门口遇到了多少辆车？", + "answer_chinese": "你遇到了一辆车，他从你面前经过，它在你的左边。", + "evidence": [ + { + "description": "You come across a car.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero cars.", + "B": "You encountered one car. It passed in front of you and is now to your left.", + "C": "Two cars.", + "D": "You encountered two cars. The last passed in front of you and ended up to your left.", + "E": "You encountered three cars. The last one crossed in front of you and is now positioned to your left." + }, + "answer_label": "B", + "answer_text": "You encountered one car. It passed in front of you and is now to your left.", + "q_id": "01443_2_T2", + "correct_answer": "You encountered one car. It passed in front of you and is now to your left.", + "original_options": { + "A": "Zero cars.", + "B": "You encountered one car. It passed in front of you and is now to your left.", + "C": "Two cars.", + "D": "You encountered two cars. The last passed in front of you and ended up to your left.", + "E": "You encountered three cars. The last one crossed in front of you and is now positioned to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 19 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_2_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cars did I run into at the doorway?", + "answer": "You encountered two cars at the entrance. The last one passed in front of you and is now to your left.", + "question_chinese": "我在门口遇到了多少辆车？", + "answer_chinese": "你在门口遇到了两辆车，最后一辆从你面前经过，现在在你的左边。", + "evidence": [ + { + "description": "First.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:08" + } + }, + { + "description": "Second.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You encountered two cars at the entrance. The last one passed in front of you and is now to your left.", + "B": "One car.", + "C": "Three cars.", + "D": "You encountered one car at the entrance. It passed in front of you and is now to your left.", + "E": "You encountered four cars at the entrance. The last one passed in front of you and is now to your left." + }, + "answer_label": "A", + "answer_text": "You encountered two cars at the entrance. The last one passed in front of you and is now to your left.", + "q_id": "01443_2_T1", + "correct_answer": "You encountered two cars at the entrance. The last one passed in front of you and is now to your left.", + "original_options": { + "A": "You encountered two cars at the entrance. The last one passed in front of you and is now to your left.", + "B": "One car.", + "C": "Three cars.", + "D": "You encountered one car at the entrance. It passed in front of you and is now to your left.", + "E": "You encountered four cars at the entrance. The last one passed in front of you and is now to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 19, + 21 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_11", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the instrument cart now?", + "answer": "Yes.", + "question_chinese": "我现在可以够到仪器的推车吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The instrument cart is right to my right.", + "timestamps": { + "start": "00:01:28", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01443_11", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_12", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the instrument cart now?", + "answer": "No.", + "question_chinese": "我现在可以够到仪器的推车吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "I am already a distance away from the instrument vehicle and cannot directly reach it.", + "timestamps": { + "start": "00:01:36", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:01:42", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "01443_12", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_5", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What floor am I on now?", + "answer": "1st floor.", + "question_chinese": "我现在在几楼？", + "answer_chinese": "1楼", + "evidence": [ + { + "description": "After going to the hospital, I did not climb any stairs.", + "timestamps": { + "start": "00:00:29", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "1st floor.", + "B": "2nd floor.", + "C": "Basement level.", + "D": "3rd floor, after riding up and exiting the elevator a moment ago.", + "E": "Ground floor, since you never went up a level during the walk." + }, + "answer_label": "A", + "answer_text": "1st floor.", + "q_id": "01443_5", + "correct_answer": "1st floor.", + "original_options": { + "A": "1st floor.", + "B": "2nd floor.", + "C": "Basement level.", + "D": "3rd floor, after riding up and exiting the elevator a moment ago.", + "E": "Ground floor, since you never went up a level during the walk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_6", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What floor am I on now?", + "answer": "Third floor.", + "question_chinese": "我现在在几楼？", + "answer_chinese": "3楼", + "evidence": [ + { + "description": "Took the elevator to the third floor.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:11" + } + }, + { + "description": "After reaching the third floor, there are no stairs up or down.", + "timestamps": { + "start": "00:01:11", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "First floor.", + "B": "Third floor.", + "C": "Second floor.", + "D": "You are on the ground floor rather than up on the third level.", + "E": "You are on the fourth floor, one level higher than the third." + }, + "answer_label": "B", + "answer_text": "Third floor.", + "q_id": "01443_6", + "correct_answer": "Third floor.", + "original_options": { + "A": "First floor.", + "B": "Third floor.", + "C": "Second floor.", + "D": "You are on the ground floor rather than up on the third level.", + "E": "You are on the fourth floor, one level higher than the third." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_4_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many people in wheelchairs have I seen?", + "answer": "You met a person in a wheelchair; they are in front of you.", + "question_chinese": "我遇见过多少个坐轮椅的人？", + "answer_chinese": "你遇见一个坐轮椅的人，他在你前面", + "evidence": [ + { + "description": "You meet a person in a wheelchair.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No one.", + "B": "Two people.", + "C": "You met a person in a wheelchair, but they are behind you now.", + "D": "You met a person in a wheelchair; they are in front of you.", + "E": "You met two people in wheelchairs, and the last one you saw is to your left." + }, + "answer_label": "D", + "answer_text": "You met a person in a wheelchair; they are in front of you.", + "q_id": "01443_4_T2", + "correct_answer": "You met a person in a wheelchair; they are in front of you.", + "original_options": { + "A": "No one.", + "B": "Two people.", + "C": "You met a person in a wheelchair, but they are behind you now.", + "D": "You met a person in a wheelchair; they are in front of you.", + "E": "You met two people in wheelchairs, and the last one you saw is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 16 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_4_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many people in wheelchairs have I seen?", + "answer": "You met two people in wheelchairs; the last one you met is to your front-left.", + "question_chinese": "我遇见过多少个坐轮椅的人？", + "answer_chinese": "你遇见两个坐轮椅的人，最后遇见的人在你的左前方。", + "evidence": [ + { + "description": "First", + "timestamps": { + "start": "00:00:36", + "end": "00:00:40" + } + }, + { + "description": "Second", + "timestamps": { + "start": "00:01:18", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw one person in a wheelchair.", + "B": "You saw three people in wheelchairs.", + "C": "You met two people in wheelchairs; the last one you met is to your front-left.", + "D": "You met one person in a wheelchair, and the last one you met is to your front-left.", + "E": "You met four people in wheelchairs, and the last one you met is positioned to your front-left." + }, + "answer_label": "C", + "answer_text": "You met two people in wheelchairs; the last one you met is to your front-left.", + "q_id": "01443_4_T1", + "correct_answer": "You met two people in wheelchairs; the last one you met is to your front-left.", + "original_options": { + "A": "You saw one person in a wheelchair.", + "B": "You saw three people in wheelchairs.", + "C": "You met two people in wheelchairs; the last one you met is to your front-left.", + "D": "You met one person in a wheelchair, and the last one you met is to your front-left.", + "E": "You met four people in wheelchairs, and the last one you met is positioned to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 17 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_14", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the elevator from me now?", + "answer": "Left side.", + "question_chinese": "现在电梯在我的哪个方位？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "I turned left after exiting the elevator.", + "timestamps": { + "start": "00:01:08", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:01:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "In front of you.", + "C": "It’s to your right when you face the hallway, not on the left side.", + "D": "Left side.", + "E": "The elevator is straight ahead of you from here, rather than located on your left side." + }, + "answer_label": "D", + "answer_text": "Left side.", + "q_id": "01443_14", + "correct_answer": "Left side.", + "original_options": { + "A": "Right side.", + "B": "In front of you.", + "C": "It’s to your right when you face the hallway, not on the left side.", + "D": "Left side.", + "E": "The elevator is straight ahead of you from here, rather than located on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 15, + 16 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_13", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the elevator from me now?", + "answer": "rear left.", + "question_chinese": "现在电梯在我的哪个方位？", + "answer_chinese": "左后方", + "evidence": [ + { + "description": "After getting off the elevator, I turned left.", + "timestamps": { + "start": "00:01:08", + "end": "00:01:12" + } + }, + { + "description": "I kept going straight, with the elevator behind me to the left.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Front right.", + "B": "Rear right.", + "C": "It’s behind you but on the right side, not at the rear left.", + "D": "The elevator is in front of you on the left side, rather than behind-left.", + "E": "rear left." + }, + "answer_label": "E", + "answer_text": "rear left.", + "q_id": "01443_13", + "correct_answer": "rear left.", + "original_options": { + "A": "Front right.", + "B": "Rear right.", + "C": "It’s behind you but on the right side, not at the rear left.", + "D": "The elevator is in front of you on the left side, rather than behind-left.", + "E": "rear left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_8_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the doorway relative to me?", + "answer": "The entrance is to your right.", + "question_chinese": "门口在我的哪里？", + "answer_chinese": "门口在你的右边", + "evidence": [ + { + "description": "The doorway is on your right.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The entrance is to your right.", + "B": "The entrance is to your left.", + "C": "The entrance is in front of you.", + "D": "The entrance is behind you as you stand here, not positioned to your right side.", + "E": "The entrance is above you rather than on your right, so you’d need to look upward." + }, + "answer_label": "A", + "answer_text": "The entrance is to your right.", + "q_id": "01443_8_T1", + "correct_answer": "The entrance is to your right.", + "original_options": { + "A": "The entrance is to your right.", + "B": "The entrance is to your left.", + "C": "The entrance is in front of you.", + "D": "The entrance is behind you as you stand here, not positioned to your right side.", + "E": "The entrance is above you rather than on your right, so you’d need to look upward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 17 + ] + }, + { + "video_uid": "01443", + "video_length": "00:01:42", + "qid": "01443_8_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the doorway relative to me?", + "answer": "The doorway is in front of you.", + "question_chinese": "门口在我的哪里？", + "answer_chinese": "门口在你的前面", + "evidence": [ + { + "description": "The door is in front of you.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:01:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The doorway is behind you.", + "B": "The doorway is to your left.", + "C": "The doorway is to your right side as you face forward, not directly in front of you.", + "D": "The doorway is above you rather than in front, so you’d have to tilt your head up.", + "E": "The doorway is in front of you." + }, + "answer_label": "E", + "answer_text": "The doorway is in front of you.", + "q_id": "01443_8_T2", + "correct_answer": "The doorway is in front of you.", + "original_options": { + "A": "The doorway is behind you.", + "B": "The doorway is to your left.", + "C": "The doorway is to your right side as you face forward, not directly in front of you.", + "D": "The doorway is above you rather than in front, so you’d have to tilt your head up.", + "E": "The doorway is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 18 + ] + } +] diff --git a/QAs/01446-mcq.json b/QAs/01446-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7448c1bcc0d8dcc8d2590042aab64502e4b42832 --- /dev/null +++ b/QAs/01446-mcq.json @@ -0,0 +1,852 @@ +[ + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_15", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, I have gone through several areas of the market.", + "answer": "one.", + "question_chinese": "到现在我经过了市场的几个区域？", + "answer_chinese": "1个", + "evidence": [ + { + "description": "Up to now, I have passed through the rice district.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "one.", + "C": "Four.", + "D": "You have gone through two areas of the market so far, not just one.", + "E": "You have already passed through six areas of the market so far, rather than only one." + }, + "answer_label": "B", + "answer_text": "one.", + "q_id": "01446_15", + "correct_answer": "one.", + "original_options": { + "A": "Two.", + "B": "one.", + "C": "Four.", + "D": "You have gone through two areas of the market so far, not just one.", + "E": "You have already passed through six areas of the market so far, rather than only one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_16", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, I have gone through several areas of the market.", + "answer": "two.", + "question_chinese": "到现在我经过了市场的几个区域？", + "answer_chinese": "2", + "evidence": [ + { + "description": "I passed by the vegetable section again.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:05" + } + }, + { + "description": "I passed through the rice area again.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You have gone through one area of the market so far, not two.", + "D": "You have already gone through four areas of the market so far, rather than two.", + "E": "two." + }, + "answer_label": "E", + "answer_text": "two.", + "q_id": "01446_16", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You have gone through one area of the market so far, not two.", + "D": "You have already gone through four areas of the market so far, rather than two.", + "E": "two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_6_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I still want to buy chili peppers now, how should I move to get the chili peppers?", + "answer": "First turn around and walk straight along the stall, then turn right and keep walking forward to the end of this stall; it’ll be to your left.", + "question_chinese": "如果我现在还想买辣椒应该怎么移动去拿辣椒？", + "answer_chinese": "先掉头沿着摊位直走再右转一直往前走走到这个摊位的尽头，在你的左方。", + "evidence": [ + { + "description": "The chili is behind the cucumber.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:01:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around, then turn left.", + "B": "Turn around, then go straight.", + "C": "First turn around and walk straight along the stall, then turn right and keep walking forward to the end of this stall; it’ll be to your left.", + "D": "Turn around, go straight, then turn right; the chili peppers will be on your right.", + "E": "Walk straight without turning around, then turn right at the end; they’ll be on your left." + }, + "answer_label": "C", + "answer_text": "First turn around and walk straight along the stall, then turn right and keep walking forward to the end of this stall; it’ll be to your left.", + "q_id": "01446_6_T1", + "correct_answer": "First turn around and walk straight along the stall, then turn right and keep walking forward to the end of this stall; it’ll be to your left.", + "original_options": { + "A": "Turn around, then turn left.", + "B": "Turn around, then go straight.", + "C": "First turn around and walk straight along the stall, then turn right and keep walking forward to the end of this stall; it’ll be to your left.", + "D": "Turn around, go straight, then turn right; the chili peppers will be on your right.", + "E": "Walk straight without turning around, then turn right at the end; they’ll be on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_6_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I still want to buy chili peppers now, how should I move to get the chili peppers?", + "answer": "First turn around and walk straight along the stall, then turn right and keep walking forward to the end of this stall; you’ll end up to your back-left.", + "question_chinese": "如果我现在还想买辣椒应该怎么移动去拿辣椒？", + "answer_chinese": "先掉头沿着摊位直走再右转一直往前走走到这个摊位的尽头，去到你的左后方。", + "evidence": [ + { + "description": "The chili is behind the cucumber.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:57" + } + }, + { + "description": "After taking the cucumber, I went to the right to pay.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:10" + } + }, + { + "description": "I walked forward two steps and then turned left and walked a few steps.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:26" + } + }, + { + "description": "After paying the bill, I turned left.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:19" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right and walk to the end.", + "B": "Turn left, then walk to the end.", + "C": "Turn around and walk along the stall, then turn right; the chili peppers will be at your back-right.", + "D": "First turn around and walk straight along the stall, then turn right and keep walking forward to the end of this stall; you’ll end up to your back-left.", + "E": "Turn around and go straight along the stall, then turn left and continue forward; they’ll end up behind you." + }, + "answer_label": "D", + "answer_text": "First turn around and walk straight along the stall, then turn right and keep walking forward to the end of this stall; you’ll end up to your back-left.", + "q_id": "01446_6_T2", + "correct_answer": "First turn around and walk straight along the stall, then turn right and keep walking forward to the end of this stall; you’ll end up to your back-left.", + "original_options": { + "A": "Turn right and walk to the end.", + "B": "Turn left, then walk to the end.", + "C": "Turn around and walk along the stall, then turn right; the chili peppers will be at your back-right.", + "D": "First turn around and walk straight along the stall, then turn right and keep walking forward to the end of this stall; you’ll end up to your back-left.", + "E": "Turn around and go straight along the stall, then turn left and continue forward; they’ll end up behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 20 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_5_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to buy cucumbers right now, how should I move to get to the cucumbers?", + "answer": "Turn left and walk a few steps; it’ll be to your left.", + "question_chinese": "如果现在我还想买青瓜应该怎么移动去拿青瓜？", + "answer_chinese": "向左转走几步到你的左边。", + "evidence": [ + { + "description": "The chili is behind the cucumber.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:57" + } + }, + { + "description": "After taking the cucumber, I went to the right to pay.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right a few steps.", + "B": "Turn left and walk a few steps; it’ll be to your left.", + "C": "Walk straight a few steps.", + "D": "Turn left and walk a few steps; the cucumbers will be to your right instead of your left.", + "E": "Turn around first, then walk a few steps forward; the cucumbers will end up behind you." + }, + "answer_label": "B", + "answer_text": "Turn left and walk a few steps; it’ll be to your left.", + "q_id": "01446_5_T1", + "correct_answer": "Turn left and walk a few steps; it’ll be to your left.", + "original_options": { + "A": "Turn right a few steps.", + "B": "Turn left and walk a few steps; it’ll be to your left.", + "C": "Walk straight a few steps.", + "D": "Turn left and walk a few steps; the cucumbers will be to your right instead of your left.", + "E": "Turn around first, then walk a few steps forward; the cucumbers will end up behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 16 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_5_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to buy cucumbers right now, how should I move to get to the cucumbers?", + "answer": "Turn around, then turn right and walk a few steps; it’ll be to your back-left.", + "question_chinese": "如果现在我还想买青瓜应该怎么移动去拿青瓜？", + "answer_chinese": "回头向右转走几步到你的左后方。", + "evidence": [ + { + "description": "The chili is behind the cucumber.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:57" + } + }, + { + "description": "After taking the cucumber, I went to the right to pay.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:01:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around, then turn left.", + "B": "Turn around and walk straight ahead.", + "C": "Turn around, then turn right and walk a few steps; it’ll be to your back-left.", + "D": "Turn around, then turn right and walk a few steps; it’ll be to your back-right.", + "E": "Turn around, then turn right and take a few steps; the cucumbers will end up on your front-left." + }, + "answer_label": "C", + "answer_text": "Turn around, then turn right and walk a few steps; it’ll be to your back-left.", + "q_id": "01446_5_T2", + "correct_answer": "Turn around, then turn right and walk a few steps; it’ll be to your back-left.", + "original_options": { + "A": "Turn around, then turn left.", + "B": "Turn around and walk straight ahead.", + "C": "Turn around, then turn right and walk a few steps; it’ll be to your back-left.", + "D": "Turn around, then turn right and walk a few steps; it’ll be to your back-right.", + "E": "Turn around, then turn right and take a few steps; the cucumbers will end up on your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 18 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How much did I spend on groceries since I went to the market?", + "answer": "ten yuan.", + "question_chinese": "我买了多少钱的菜？", + "answer_chinese": "10元", + "evidence": [ + { + "description": "I bought vegetables for 10 yuan.", + "timestamps": { + "start": "00:01:17", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "five yuan.", + "B": "ten yuan.", + "C": "fifteen yuan.", + "D": "In total, you spent twenty yuan on groceries at the market since arriving.", + "E": "Altogether, you paid thirty yuan for groceries after you started shopping at the market." + }, + "answer_label": "B", + "answer_text": "ten yuan.", + "q_id": "01446_3", + "correct_answer": "ten yuan.", + "original_options": { + "A": "five yuan.", + "B": "ten yuan.", + "C": "fifteen yuan.", + "D": "In total, you spent twenty yuan on groceries at the market since arriving.", + "E": "Altogether, you paid thirty yuan for groceries after you started shopping at the market." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How much did I spend on groceries since I went to the market?", + "answer": "twenty-four yuan.", + "question_chinese": "我买了多少钱的菜？", + "answer_chinese": "24元", + "evidence": [ + { + "description": "I bought vegetables for 10 yuan.", + "timestamps": { + "start": "00:01:17", + "end": "00:01:21" + } + }, + { + "description": "I bought vegetables for 14 yuan.", + "timestamps": { + "start": "00:01:55", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "nineteen yuan.", + "B": "twenty-one yuan.", + "C": "twenty-four yuan.", + "D": "In total, you spent twenty-two yuan on groceries since you went to the market.", + "E": "Altogether, you paid eighteen yuan for your groceries after you arrived at the market." + }, + "answer_label": "C", + "answer_text": "twenty-four yuan.", + "q_id": "01446_4", + "correct_answer": "twenty-four yuan.", + "original_options": { + "A": "nineteen yuan.", + "B": "twenty-one yuan.", + "C": "twenty-four yuan.", + "D": "In total, you spent twenty-two yuan on groceries since you went to the market.", + "E": "Altogether, you paid eighteen yuan for your groceries after you arrived at the market." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kinds of vegetables did I touch?", + "answer": "eight.", + "question_chinese": "我摸了几种菜？", + "answer_chinese": "8", + "evidence": [ + { + "description": "I touched the Chinese cabbage, the onion, and the scallion.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:27" + } + }, + { + "description": "I touched the radish and the potato.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:32" + } + }, + { + "description": "I touched ginger and tomatoes.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:40" + } + }, + { + "description": "I touched the cucumber.", + "timestamps": { + "start": "00:00:51", + "end": "00:00:56" + } + } + ], + "question_timestamps": "00:01:16", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "seven.", + "B": "nine.", + "C": "You touched six different kinds of vegetables while shopping at the market today.", + "D": "In total, you handled five kinds of vegetables during your trip through the market.", + "E": "eight." + }, + "answer_label": "E", + "answer_text": "eight.", + "q_id": "01446_2", + "correct_answer": "eight.", + "original_options": { + "A": "seven.", + "B": "nine.", + "C": "You touched six different kinds of vegetables while shopping at the market today.", + "D": "In total, you handled five kinds of vegetables during your trip through the market.", + "E": "eight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kinds of vegetables did I touch?", + "answer": "ten.", + "question_chinese": "我摸了几种菜？", + "answer_chinese": "10", + "evidence": [ + { + "description": "I touched the Chinese cabbage, the onion, and the scallion.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:27" + } + }, + { + "description": "I touched the radish and the potato.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:32" + } + }, + { + "description": "I touched ginger and tomatoes.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:40" + } + }, + { + "description": "I touched the cucumber.", + "timestamps": { + "start": "00:00:51", + "end": "00:00:56" + } + }, + { + "description": "I touched the mushroom", + "timestamps": { + "start": "00:00:31", + "end": "00:00:33" + } + }, + { + "description": "I touched the tea tree mushrooms.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "nine.", + "B": "seven.", + "C": "You touched six different kinds of vegetables while shopping around the market today.", + "D": "ten.", + "E": "Altogether, you handled four kinds of vegetables during your trip through the market." + }, + "answer_label": "D", + "answer_text": "ten.", + "q_id": "01446_1", + "correct_answer": "ten.", + "original_options": { + "A": "nine.", + "B": "seven.", + "C": "You touched six different kinds of vegetables while shopping around the market today.", + "D": "ten.", + "E": "Altogether, you handled four kinds of vegetables during your trip through the market." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_13", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach out and touch the mushrooms now?", + "answer": "Yes.", + "question_chinese": "现在我能直接伸手碰到蘑菇吗?", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The mushroom is right next to me.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:34", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01446_13", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_14", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach out and touch the mushrooms now?", + "answer": "No.", + "question_chinese": "现在我能直接伸手碰到蘑菇吗?", + "answer_chinese": "不能", + "evidence": [ + { + "description": "Now I can reach out and grab the mushrooms directly.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:00" + } + }, + { + "description": "I took several steps forward.", + "timestamps": { + "start": "00:02:05", + "end": "00:02:09" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01446_14", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_4_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, which side of me is the blue cart on?", + "answer": "Behind you.", + "question_chinese": "现在蓝色推车在我的哪边？", + "answer_chinese": "你的后方。", + "evidence": [ + { + "description": "The cart is on my right.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "To your right, near where you’re facing at the moment right now.", + "E": "Beside you, close enough to reach without turning around at all right now." + }, + "answer_label": "B", + "answer_text": "Behind you.", + "q_id": "01446_4_T2", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "To your right, near where you’re facing at the moment right now.", + "E": "Beside you, close enough to reach without turning around at all right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 13, + 13 + ] + }, + { + "video_uid": "01446", + "video_length": "00:02:10", + "qid": "01446_4_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, which side of me is the blue cart on?", + "answer": "To your right.", + "question_chinese": "现在蓝色推车在我的哪边？", + "answer_chinese": "你的右方。", + "evidence": [ + { + "description": "The cart is on my right.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + }, + { + "description": "I turned around.The blue cart is behind me, and I walk straight ahead.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:16" + } + }, + { + "description": "I have been walking along the aisle, and now turning towards it, the blue cart is on my right.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your right.", + "B": "To your left.", + "C": "Behind you.", + "D": "The blue cart is behind you at the moment, not on your right side.", + "E": "It is in front of you right now, along the direction you are currently facing." + }, + "answer_label": "A", + "answer_text": "To your right.", + "q_id": "01446_4_T1", + "correct_answer": "To your right.", + "original_options": { + "A": "To your right.", + "B": "To your left.", + "C": "Behind you.", + "D": "The blue cart is behind you at the moment, not on your right side.", + "E": "It is in front of you right now, along the direction you are currently facing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 14, + 15 + ] + } +] diff --git a/QAs/01451-mcq.json b/QAs/01451-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..826fd8ef9484ed86e7049a9bc6652d3cba50167a --- /dev/null +++ b/QAs/01451-mcq.json @@ -0,0 +1,1096 @@ +[ + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_2_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take after getting out of the elevator?", + "answer": "Walk forward one meter, then turn left.", + "question_chinese": "我出电梯后的路径是什么", + "answer_chinese": "向前走一米，左转", + "evidence": [ + { + "description": "Walk forward one meter.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + }, + { + "description": "turn left", + "timestamps": { + "start": "00:00:40", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk forward, then turn right.", + "B": "Walk forward two meters, then turn left.", + "C": "Walk forward one meter, then turn right and continue straight down the hallway.", + "D": "Walk forward half a meter, then turn left and stop just outside the elevator doors.", + "E": "Walk forward one meter, then turn left." + }, + "answer_label": "E", + "answer_text": "Walk forward one meter, then turn left.", + "q_id": "01451_2_T1", + "correct_answer": "Walk forward one meter, then turn left.", + "original_options": { + "A": "Walk forward, then turn right.", + "B": "Walk forward two meters, then turn left.", + "C": "Walk forward one meter, then turn right and continue straight down the hallway.", + "D": "Walk forward half a meter, then turn left and stop just outside the elevator doors.", + "E": "Walk forward one meter, then turn left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 13, + 15 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_2_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take after getting out of the elevator?", + "answer": "Walk forward one meter, turn left, and go in through the front door.", + "question_chinese": "我出电梯后的路径是什么", + "answer_chinese": "向前走一米，左转，进家门", + "evidence": [ + { + "description": "Walk forward one meter.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + }, + { + "description": "turn left", + "timestamps": { + "start": "00:00:40", + "end": "00:00:41" + } + }, + { + "description": "Enter the house.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk forward, turn right, enter door.", + "B": "Walk forward one meter, turn left, and go in through the front door.", + "C": "Walk forward, turn left, use back door.", + "D": "Walk forward one meter, then turn left and stop without going through any doorway.", + "E": "Walk forward half a meter out of the elevator, turn left, and go in through the front door." + }, + "answer_label": "B", + "answer_text": "Walk forward one meter, turn left, and go in through the front door.", + "q_id": "01451_2_T2", + "correct_answer": "Walk forward one meter, turn left, and go in through the front door.", + "original_options": { + "A": "Walk forward, turn right, enter door.", + "B": "Walk forward one meter, turn left, and go in through the front door.", + "C": "Walk forward, turn left, use back door.", + "D": "Walk forward one meter, then turn left and stop without going through any doorway.", + "E": "Walk forward half a meter out of the elevator, turn left, and go in through the front door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 18 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_1_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take while riding the elevator?", + "answer": "Walk forward half a meter into the elevator, then turn around.", + "question_chinese": "我坐电梯的路径是什么", + "answer_chinese": "向前走半米进入电梯，向后转", + "evidence": [ + { + "description": "Walk half a meter forward and enter the elevator.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "Turn around.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Step in and press the button.", + "B": "Walk forward half a meter into the elevator, then turn around.", + "C": "Walk one meter, then turn left.", + "D": "Walk forward half a meter into the elevator, then turn left and face the side wall.", + "E": "Walk forward half a meter into the elevator, keep facing forward, and wait without turning around." + }, + "answer_label": "B", + "answer_text": "Walk forward half a meter into the elevator, then turn around.", + "q_id": "01451_1_T1", + "correct_answer": "Walk forward half a meter into the elevator, then turn around.", + "original_options": { + "A": "Step in and press the button.", + "B": "Walk forward half a meter into the elevator, then turn around.", + "C": "Walk one meter, then turn left.", + "D": "Walk forward half a meter into the elevator, then turn left and face the side wall.", + "E": "Walk forward half a meter into the elevator, keep facing forward, and wait without turning around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_1_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take while riding the elevator?", + "answer": "Walk forward half a meter into the elevator, turn around, and press the elevator button.", + "question_chinese": "我坐电梯的路径是什么", + "answer_chinese": "向前走半米进入电梯，向后转，按电梯按钮", + "evidence": [ + { + "description": "Walk half a meter forward and enter the elevator.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "Turn around.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "Press the elevator button.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Enter elevator, then turn around.", + "B": "Walk one meter, then turn left.", + "C": "Walk forward half a meter into the elevator, press the button, and then turn around afterward.", + "D": "Walk forward half a meter into the elevator, turn around, and press the elevator button.", + "E": "Walk forward half a meter into the elevator, turn around, but do not press any elevator button." + }, + "answer_label": "D", + "answer_text": "Walk forward half a meter into the elevator, turn around, and press the elevator button.", + "q_id": "01451_1_T2", + "correct_answer": "Walk forward half a meter into the elevator, turn around, and press the elevator button.", + "original_options": { + "A": "Enter elevator, then turn around.", + "B": "Walk one meter, then turn left.", + "C": "Walk forward half a meter into the elevator, press the button, and then turn around afterward.", + "D": "Walk forward half a meter into the elevator, turn around, and press the elevator button.", + "E": "Walk forward half a meter into the elevator, turn around, but do not press any elevator button." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many electrical rooms have I seen so far?", + "answer": "Two.", + "question_chinese": "我现在已经看到了几个电井？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "There is an electrical room behind me", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + }, + { + "description": "There is an electrical room in front of me", + "timestamps": { + "start": "00:00:30", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You have seen one electrical room so far during the walkthrough today.", + "E": "You have seen three electrical rooms so far while moving through the building." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "01451_9", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You have seen one electrical room so far during the walkthrough today.", + "E": "You have seen three electrical rooms so far while moving through the building." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_10", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many electrical rooms have I seen so far?", + "answer": "Three.", + "question_chinese": "我现在已经看到了几个电井？", + "answer_chinese": "3个", + "evidence": [ + { + "description": "There is an electrical room behind me", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + }, + { + "description": "There is an electrical room in front of me", + "timestamps": { + "start": "00:00:30", + "end": "00:00:33" + } + }, + { + "description": "There is also an electrical room on my right side", + "timestamps": { + "start": "00:00:36", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:41", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "You have only seen two electrical rooms so far, not three rooms in total.", + "D": "Three.", + "E": "Up to now, you have passed through four electrical rooms in total during your walk." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "01451_10", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "You have only seen two electrical rooms so far, not three rooms in total.", + "D": "Three.", + "E": "Up to now, you have passed through four electrical rooms in total during your walk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_9_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to go out now and take the elevator to the 1st floor, how should I move?", + "answer": "Go out the door, turn right, and walk a few steps to enter the elevator. Turn around and press the elevator button on the right side.", + "question_chinese": "我现在要出门坐电梯到1楼应该在怎么移动？", + "answer_chinese": "出门右转几步进电梯转过身按右侧的电梯按钮", + "evidence": [ + { + "description": "Enter the elevator, turn around and press the button", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + }, + { + "description": "Pressed the third button from top to bottom on the left to go to the 3rd floor", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left to reach the elevator.", + "B": "Press the button to your left.", + "C": "Go out the door, turn right, and walk a few steps to enter the elevator. Turn around and press the elevator button on the right side.", + "D": "Go out the door, turn left, walk a few steps, enter the elevator, then press the left-side button.", + "E": "Go out the door, turn right, enter the elevator, then press the button panel on the left side." + }, + "answer_label": "C", + "answer_text": "Go out the door, turn right, and walk a few steps to enter the elevator. Turn around and press the elevator button on the right side.", + "q_id": "01451_9_T2", + "correct_answer": "Go out the door, turn right, and walk a few steps to enter the elevator. Turn around and press the elevator button on the right side.", + "original_options": { + "A": "Turn left to reach the elevator.", + "B": "Press the button to your left.", + "C": "Go out the door, turn right, and walk a few steps to enter the elevator. Turn around and press the elevator button on the right side.", + "D": "Go out the door, turn left, walk a few steps, enter the elevator, then press the left-side button.", + "E": "Go out the door, turn right, enter the elevator, then press the button panel on the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_9_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to go out now and take the elevator to the 1st floor, how should I move?", + "answer": "Go out the door, turn right, and walk a few steps to enter the elevator. Turn around and press the elevator button on the right side; press the 5th one down from the top in the left column to go to the 1st floor.", + "question_chinese": "我现在要出门坐电梯到1楼应该在怎么移动？", + "answer_chinese": "出门右转几步进电梯转过身按右侧的电梯按钮，按左侧那列的从上到下数第5个可以到1楼", + "evidence": [ + { + "description": "Enter the elevator, turn around and press the button", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + }, + { + "description": "Pressed the third button from top to bottom on the left to go to the 3rd floor", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + }, + { + "description": "Got off the elevator and turned left to the home entrance", + "timestamps": { + "start": "00:00:35", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Press the right-column button instead.", + "B": "Press the 4th one down.", + "C": "Go out the door, turn right, and walk a few steps to enter the elevator. Turn around and press the elevator button on the right side; press the 5th one down from the top in the left column to go to the 1st floor.", + "D": "Go out the door, turn right, enter the elevator, then press the 5th button down in the right column.", + "E": "Go out the door, turn right, enter the elevator, then press the 6th one down from the top in the left column." + }, + "answer_label": "C", + "answer_text": "Go out the door, turn right, and walk a few steps to enter the elevator. Turn around and press the elevator button on the right side; press the 5th one down from the top in the left column to go to the 1st floor.", + "q_id": "01451_9_T1", + "correct_answer": "Go out the door, turn right, and walk a few steps to enter the elevator. Turn around and press the elevator button on the right side; press the 5th one down from the top in the left column to go to the 1st floor.", + "original_options": { + "A": "Press the right-column button instead.", + "B": "Press the 4th one down.", + "C": "Go out the door, turn right, and walk a few steps to enter the elevator. Turn around and press the elevator button on the right side; press the 5th one down from the top in the left column to go to the 1st floor.", + "D": "Go out the door, turn right, enter the elevator, then press the 5th button down in the right column.", + "E": "Go out the door, turn right, enter the elevator, then press the 6th one down from the top in the left column." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 22 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_4_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to put my shoes on the shoe rack right now, can I reach it directly?", + "answer": "No.", + "question_chinese": "我现在要把鞋放在鞋架上能直接够到吗？", + "answer_chinese": "不", + "evidence": [ + { + "description": "The shoe rack is on the right side of the elevator entrance，but you haven't gotten out of the elevator yet.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01451_4_T2", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_4_T1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to put my shoes on the shoe rack right now, can I reach it directly?", + "answer": "Yes.", + "question_chinese": "我现在要把鞋放在鞋架上能直接够到吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "The shoe rack is on the right side of the elevator entrance", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01451_4_T1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_8_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how many steps away am I from home right now?", + "answer": "You can get there within 5 steps.", + "question_chinese": "我现在距离家还有几步路", + "answer_chinese": "5步之内就能到", + "evidence": [ + { + "description": "Entering the elevator and turning around takes about 10 steps", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + }, + { + "description": "I walked about 6 steps after getting off the elevator", + "timestamps": { + "start": "00:00:33", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Within 3 steps.", + "B": "You can get there within 5 steps.", + "C": "Within 7 steps.", + "D": "You can get there within 4 steps, so it is slightly closer than five.", + "E": "You can get there within 6 steps, so it is a bit farther than five." + }, + "answer_label": "B", + "answer_text": "You can get there within 5 steps.", + "q_id": "01451_8_T2", + "correct_answer": "You can get there within 5 steps.", + "original_options": { + "A": "Within 3 steps.", + "B": "You can get there within 5 steps.", + "C": "Within 7 steps.", + "D": "You can get there within 4 steps, so it is slightly closer than five.", + "E": "You can get there within 6 steps, so it is a bit farther than five." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 15 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_8_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how many steps away am I from home right now?", + "answer": "You can get there within 1 step.", + "question_chinese": "我现在距离家还有几步路", + "answer_chinese": "1步之内就能到", + "evidence": [ + { + "description": "Entering the elevator and turning around takes about 10 steps", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + }, + { + "description": "I walked about 6 steps after getting off the elevator", + "timestamps": { + "start": "00:00:33", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:41", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can get there within 1 step.", + "B": "Within 0 steps.", + "C": "Within 2 steps.", + "D": "You can get there within 3 steps, meaning it takes a short walk to reach home.", + "E": "You can get there within 4 steps, so home is nearby but not right next to you." + }, + "answer_label": "A", + "answer_text": "You can get there within 1 step.", + "q_id": "01451_8_T1", + "correct_answer": "You can get there within 1 step.", + "original_options": { + "A": "You can get there within 1 step.", + "B": "Within 0 steps.", + "C": "Within 2 steps.", + "D": "You can get there within 3 steps, meaning it takes a short walk to reach home.", + "E": "You can get there within 4 steps, so home is nearby but not right next to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_7_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many rooms are in front of me?", + "answer": "There is a room in front of you.", + "question_chinese": "我的前方有几个房间？", + "answer_chinese": "你前方有一个房间", + "evidence": [ + { + "description": "After opening the door, you see a room.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There is a room in front of you.", + "B": "No rooms are in front of you.", + "C": "Two rooms are in front of you.", + "D": "There are three rooms in front of you, lined up ahead as you look forward.", + "E": "There are four rooms in front of you, so you would have multiple doorways ahead." + }, + "answer_label": "A", + "answer_text": "There is a room in front of you.", + "q_id": "01451_7_T2", + "correct_answer": "There is a room in front of you.", + "original_options": { + "A": "There is a room in front of you.", + "B": "No rooms are in front of you.", + "C": "Two rooms are in front of you.", + "D": "There are three rooms in front of you, lined up ahead as you look forward.", + "E": "There are four rooms in front of you, so you would have multiple doorways ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_13", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many rooms are in front of me?", + "answer": "Four.", + "question_chinese": "我的前方有几个房间？", + "answer_chinese": "4个", + "evidence": [ + { + "description": "There are 4 rooms in front of me", + "timestamps": { + "start": "00:00:45", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Two rooms are directly in front of you right now, not four.", + "E": "Six rooms are in front of you at the moment, lined up ahead." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "01451_13", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Two rooms are directly in front of you right now, not four.", + "E": "Six rooms are in front of you at the moment, lined up ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_7_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many rooms are in front of me?", + "answer": "There are four rooms in front of you.", + "question_chinese": "我的前方有几个房间？", + "answer_chinese": "你前方有四个房间", + "evidence": [ + { + "description": "There are 4 rooms in front of me", + "timestamps": { + "start": "00:00:45", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Two rooms are ahead.", + "B": "Five rooms are ahead.", + "C": "There are three rooms in front of you right now, not four.", + "D": "There are six rooms in front of you at the moment, more than four.", + "E": "There are four rooms in front of you." + }, + "answer_label": "E", + "answer_text": "There are four rooms in front of you.", + "q_id": "01451_7_T1", + "correct_answer": "There are four rooms in front of you.", + "original_options": { + "A": "Two rooms are ahead.", + "B": "Five rooms are ahead.", + "C": "There are three rooms in front of you right now, not four.", + "D": "There are six rooms in front of you at the moment, more than four.", + "E": "There are four rooms in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_3_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is my cane probing right now?", + "answer": "In front of you.", + "question_chinese": "我的拐杖现在在探我哪个方向的路", + "answer_chinese": "前方", + "evidence": [ + { + "description": "Front", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "In front of you.", + "C": "Behind you.", + "D": "Your cane is probing to your right at this moment, not forward.", + "E": "Your cane is probing behind you rather than straight ahead in front." + }, + "answer_label": "B", + "answer_text": "In front of you.", + "q_id": "01451_3_T2", + "correct_answer": "In front of you.", + "original_options": { + "A": "To your left.", + "B": "In front of you.", + "C": "Behind you.", + "D": "Your cane is probing to your right at this moment, not forward.", + "E": "Your cane is probing behind you rather than straight ahead in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_3_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is my cane probing right now?", + "answer": "To your left.", + "question_chinese": "我的拐杖现在在探我哪个方向的路", + "answer_chinese": "左边", + "evidence": [ + { + "description": "left", + "timestamps": { + "start": "00:00:37", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "To your right.", + "D": "Your cane is probing behind you right now, not toward the side.", + "E": "Your cane is probing straight ahead in front of you at the moment." + }, + "answer_label": "B", + "answer_text": "To your left.", + "q_id": "01451_3_T1", + "correct_answer": "To your left.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "To your right.", + "D": "Your cane is probing behind you right now, not toward the side.", + "E": "Your cane is probing straight ahead in front of you at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 13 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_11", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the elevator in for me now?", + "answer": "Back side.", + "question_chinese": "现在电梯在我的什么方位？", + "answer_chinese": "后方", + "evidence": [ + { + "description": "I walked straight a few steps after getting off the elevator", + "timestamps": { + "start": "00:00:33", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Front side.", + "B": "Left side.", + "C": "The elevator is on your right side from where you are now.", + "D": "The elevator is in front of you rather than positioned behind you.", + "E": "Back side." + }, + "answer_label": "E", + "answer_text": "Back side.", + "q_id": "01451_11", + "correct_answer": "Back side.", + "original_options": { + "A": "Front side.", + "B": "Left side.", + "C": "The elevator is on your right side from where you are now.", + "D": "The elevator is in front of you rather than positioned behind you.", + "E": "Back side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01451", + "video_length": "00:00:50", + "qid": "01451_12", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the elevator in for me now?", + "answer": "Left side.", + "question_chinese": "现在电梯在我的什么方位？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "I walked straight a few steps after getting off the elevator", + "timestamps": { + "start": "00:00:33", + "end": "00:00:36" + } + }, + { + "description": "I turned right", + "timestamps": { + "start": "00:00:39", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:00:44", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "Front side.", + "C": "Left side.", + "D": "The elevator is on your back side rather than to your left.", + "E": "The elevator is positioned to your right side from where you stand." + }, + "answer_label": "C", + "answer_text": "Left side.", + "q_id": "01451_12", + "correct_answer": "Left side.", + "original_options": { + "A": "Right side.", + "B": "Front side.", + "C": "Left side.", + "D": "The elevator is on your back side rather than to your left.", + "E": "The elevator is positioned to your right side from where you stand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + } +] diff --git a/QAs/01452-mcq.json b/QAs/01452-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8dd59e8950c6c61a2b0c066fe8afff39634929e4 --- /dev/null +++ b/QAs/01452-mcq.json @@ -0,0 +1,1022 @@ +[ + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_15", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which direction did I turn at the last intersection?", + "answer": "Turn left.", + "question_chinese": "我在上一个路口往哪转了？", + "answer_chinese": "左转", + "evidence": [ + { + "description": "I turned left at the intersection", + "timestamps": { + "start": "00:00:38", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:02:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left.", + "B": "Turn right.", + "C": "Go straight.", + "D": "Turn around and head back the way you came at the last intersection.", + "E": "Keep going straight through the intersection, then take the next right turn." + }, + "answer_label": "A", + "answer_text": "Turn left.", + "q_id": "01452_15", + "correct_answer": "Turn left.", + "original_options": { + "A": "Turn left.", + "B": "Turn right.", + "C": "Go straight.", + "D": "Turn around and head back the way you came at the last intersection.", + "E": "Keep going straight through the intersection, then take the next right turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_16", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which direction did I turn at the last intersection?", + "answer": "Turn left.", + "question_chinese": "我在上一个路口往哪转了？", + "answer_chinese": "左转", + "evidence": [ + { + "description": "I turned left at the intersection", + "timestamps": { + "start": "00:01:13", + "end": "00:01:21" + } + } + ], + "question_timestamps": "00:02:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right.", + "B": "Turn around.", + "C": "At the last intersection, you went straight rather than turning left there.", + "D": "Turn left.", + "E": "You turned right at the last intersection, not left as you proceeded." + }, + "answer_label": "D", + "answer_text": "Turn left.", + "q_id": "01452_16", + "correct_answer": "Turn left.", + "original_options": { + "A": "Turn right.", + "B": "Turn around.", + "C": "At the last intersection, you went straight rather than turning left there.", + "D": "Turn left.", + "E": "You turned right at the last intersection, not left as you proceeded." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is Adidas in for me now?", + "answer": "Left side.", + "question_chinese": "现在adidas在我的什么方位？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "Adidas is in front of me", + "timestamps": { + "start": "00:00:20", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "Behind you.", + "C": "Left side.", + "D": "It should be in front of you, not off to either side.", + "E": "You'd find it on your right as you face forward from where you are now." + }, + "answer_label": "C", + "answer_text": "Left side.", + "q_id": "01452_1", + "correct_answer": "Left side.", + "original_options": { + "A": "Right side.", + "B": "Behind you.", + "C": "Left side.", + "D": "It should be in front of you, not off to either side.", + "E": "You'd find it on your right as you face forward from where you are now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 15 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is Adidas in for me now?", + "answer": "Back side.", + "question_chinese": "现在adidas在我的什么方位？", + "answer_chinese": "后方", + "evidence": [ + { + "description": "Adidas is in front of me", + "timestamps": { + "start": "00:00:20", + "end": "00:00:22" + } + }, + { + "description": "Adidas is on my left side", + "timestamps": { + "start": "00:00:48", + "end": "00:00:52" + } + }, + { + "description": "I walked straight for a distance and then turned right", + "timestamps": { + "start": "00:01:00", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Front side.", + "B": "Left side.", + "C": "Adidas is in front of you, not behind where you are standing right now.", + "D": "Adidas is off to your left side from your current position rather than at your back.", + "E": "Back side." + }, + "answer_label": "E", + "answer_text": "Back side.", + "q_id": "01452_2", + "correct_answer": "Back side.", + "original_options": { + "A": "Front side.", + "B": "Left side.", + "C": "Adidas is in front of you, not behind where you are standing right now.", + "D": "Adidas is off to your left side from your current position rather than at your back.", + "E": "Back side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans have I seen so far?", + "answer": "Two.", + "question_chinese": "现在我已经看到几个垃圾桶了？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "There are two trash cans in front right of me", + "timestamps": { + "start": "00:00:25", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You have seen three trash cans in total so far in the video.", + "D": "Two.", + "E": "Up to this point, you've encountered only one trash can during the video." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "01452_3", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You have seen three trash cans in total so far in the video.", + "D": "Two.", + "E": "Up to this point, you've encountered only one trash can during the video." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans have I seen so far?", + "answer": "Three.", + "question_chinese": "现在我已经看到几个垃圾桶了？", + "answer_chinese": "3个", + "evidence": [ + { + "description": "There are two trash cans in front right of me", + "timestamps": { + "start": "00:00:25", + "end": "00:00:27" + } + }, + { + "description": "There is one trash can in the tasting area in front of me", + "timestamps": { + "start": "00:00:41", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:03:19", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "So far you've come across four trash cans altogether in the video.", + "D": "Three.", + "E": "You have only seen two trash cans up to this point in the footage." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "01452_4", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "So far you've come across four trash cans altogether in the video.", + "D": "Three.", + "E": "You have only seen two trash cans up to this point in the footage." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many strollers have I encountered so far?", + "answer": "one.", + "question_chinese": "现在我已经碰到了多少个婴儿推车？", + "answer_chinese": "1个", + "evidence": [ + { + "description": "There is one stroller in front of me", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "You've encountered two strollers in total so far during the walk today.", + "D": "one.", + "E": "Up to now, you have seen three strollers during your walk so far." + }, + "answer_label": "D", + "answer_text": "one.", + "q_id": "01452_11", + "correct_answer": "one.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "You've encountered two strollers in total so far during the walk today.", + "D": "one.", + "E": "Up to now, you have seen three strollers during your walk so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_12", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many strollers have I encountered so far?", + "answer": "two.", + "question_chinese": "现在我已经碰到了多少个婴儿推车？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "There is one stroller in front of me", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + }, + { + "description": "There is one stroller in front of me", + "timestamps": { + "start": "00:01:09", + "end": "00:01:12" + } + } + ], + "question_timestamps": "00:02:41", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "So far you've encountered three strollers in total during your walk today.", + "D": "Up to this point, you have only come across one stroller so far.", + "E": "two." + }, + "answer_label": "E", + "answer_text": "two.", + "q_id": "01452_12", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "So far you've encountered three strollers in total during your walk today.", + "D": "Up to this point, you have only come across one stroller so far.", + "E": "two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_13", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many intersections have I passed so far?", + "answer": "one.", + "question_chinese": "现在我已经经过了几个路口？", + "answer_chinese": "1个", + "evidence": [ + { + "description": "There is an intersection in front of me", + "timestamps": { + "start": "00:00:34", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "one.", + "D": "You have already passed two intersections so far on your route today.", + "E": "Up to now, you've gone through three intersections along your route so far." + }, + "answer_label": "C", + "answer_text": "one.", + "q_id": "01452_13", + "correct_answer": "one.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "one.", + "D": "You have already passed two intersections so far on your route today.", + "E": "Up to now, you've gone through three intersections along your route so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_14", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many intersections have I passed so far?", + "answer": "two.", + "question_chinese": "现在我已经经过了几个路口？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "There is an intersection in front of me", + "timestamps": { + "start": "00:00:34", + "end": "00:00:38" + } + }, + { + "description": "There is an intersection in front of me", + "timestamps": { + "start": "00:00:57", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:02:05", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "two.", + "C": "Three.", + "D": "So far, you've passed three intersections in total along your route today.", + "E": "Up to this point, you have only gone through one intersection so far." + }, + "answer_label": "B", + "answer_text": "two.", + "q_id": "01452_14", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "two.", + "C": "Three.", + "D": "So far, you've passed three intersections in total along your route today.", + "E": "Up to this point, you have only gone through one intersection so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_7", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many shops have I passed by so far?", + "answer": "two.", + "question_chinese": "现在我已经经过了多少个商铺？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "There is a Xuanzi shop in front of me", + "timestamps": { + "start": "00:00:38", + "end": "00:00:40" + } + }, + { + "description": "Adidas is on my left side", + "timestamps": { + "start": "00:00:50", + "end": "00:00:56" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One shop.", + "B": "Three shops.", + "C": "You have passed by three shops so far along the way today.", + "D": "two.", + "E": "So far, you have walked past only one shop on this route." + }, + "answer_label": "D", + "answer_text": "two.", + "q_id": "01452_7", + "correct_answer": "two.", + "original_options": { + "A": "One shop.", + "B": "Three shops.", + "C": "You have passed by three shops so far along the way today.", + "D": "two.", + "E": "So far, you have walked past only one shop on this route." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_8", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many shops have I passed by so far?", + "answer": "five.", + "question_chinese": "现在我已经经过了多少个商铺？", + "answer_chinese": "5个", + "evidence": [ + { + "description": "There is a Xuanzi shop in front of me", + "timestamps": { + "start": "00:00:38", + "end": "00:00:40" + } + }, + { + "description": "Adidas is on my left side", + "timestamps": { + "start": "00:00:50", + "end": "00:00:56" + } + }, + { + "description": "I passed by Huawei store again", + "timestamps": { + "start": "00:01:04", + "end": "00:01:10" + } + }, + { + "description": "I passed by Lukfook Jewelry and Heytea again", + "timestamps": { + "start": "00:01:27", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:01:34", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "five.", + "B": "Four shops.", + "C": "Six shops.", + "D": "You have passed by three shops so far during your walk today.", + "E": "So far, you have already walked past seven shops along the corridor." + }, + "answer_label": "A", + "answer_text": "five.", + "q_id": "01452_8", + "correct_answer": "five.", + "original_options": { + "A": "five.", + "B": "Four shops.", + "C": "Six shops.", + "D": "You have passed by three shops so far during your walk today.", + "E": "So far, you have already walked past seven shops along the corridor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_9", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to take the elevator now?", + "answer": "Turn left and go straight to reach the elevator.", + "question_chinese": "现在我想去坐直梯应该怎么移动？", + "answer_chinese": "左转直走可以到直梯", + "evidence": [ + { + "description": "The elevator is in front left of me", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right to the elevator.", + "B": "Go straight to the elevator.", + "C": "Turn right and continue straight until you reach the elevator doors inside.", + "D": "Walk straight ahead, then turn right at the corner to get to the elevator.", + "E": "Turn left and go straight to reach the elevator." + }, + "answer_label": "E", + "answer_text": "Turn left and go straight to reach the elevator.", + "q_id": "01452_9", + "correct_answer": "Turn left and go straight to reach the elevator.", + "original_options": { + "A": "Turn right to the elevator.", + "B": "Go straight to the elevator.", + "C": "Turn right and continue straight until you reach the elevator doors inside.", + "D": "Walk straight ahead, then turn right at the corner to get to the elevator.", + "E": "Turn left and go straight to reach the elevator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 14 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_10", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to take the elevator now?", + "answer": "Turn around, turn left and walk straight to the intersection, turn right and walk along the shops to the end, the elevator is on the right side.", + "question_chinese": "现在我想去坐直梯应该怎么移动？", + "answer_chinese": "掉头左转直走走到路口右转，沿着店铺走到底右侧是直梯", + "evidence": [ + { + "description": "The elevator is in front left of me", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + }, + { + "description": "I walked straight to the intersection and turned left", + "timestamps": { + "start": "00:00:31", + "end": "00:00:43" + } + }, + { + "description": "Walked straight forward and turned right", + "timestamps": { + "start": "00:00:57", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn around, turn left and walk straight to the intersection, turn right and walk along the shops to the end, the elevator is on the right side.", + "B": "The elevator is on your left.", + "C": "The elevator is straight ahead.", + "D": "After the shops, the elevator will be on your left side at the end.", + "E": "At the end of the shops, you will find the elevator directly in front of you." + }, + "answer_label": "A", + "answer_text": "Turn around, turn left and walk straight to the intersection, turn right and walk along the shops to the end, the elevator is on the right side.", + "q_id": "01452_10", + "correct_answer": "Turn around, turn left and walk straight to the intersection, turn right and walk along the shops to the end, the elevator is on the right side.", + "original_options": { + "A": "Turn around, turn left and walk straight to the intersection, turn right and walk along the shops to the end, the elevator is on the right side.", + "B": "The elevator is on your left.", + "C": "The elevator is straight ahead.", + "D": "After the shops, the elevator will be on your left side at the end.", + "E": "At the end of the shops, you will find the elevator directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 16 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_5", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach a place to sit down and rest within 10 steps now?", + "answer": "Yes.", + "question_chinese": "现在我想坐下来休息能在10步内到达吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "There is a place to sit in front left of me", + "timestamps": { + "start": "00:01:00", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "01452_5", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_6", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach a place to sit down and rest within 10 steps now?", + "answer": "No.", + "question_chinese": "现在我想坐下来休息能在10步内到达吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "There is a place to sit in front left of me", + "timestamps": { + "start": "00:01:00", + "end": "00:01:03" + } + }, + { + "description": "There is no place to sit and rest around me", + "timestamps": { + "start": "00:01:31", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:02:45", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01452_6", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_8_T2", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many people are directly in front of me right now?", + "answer": "Right now, there are two people directly in front of you.", + "question_chinese": "现在我的正前方有多少人？", + "answer_chinese": "现在你的正前方有两个人。", + "evidence": [ + { + "description": "Now there are two people in front of you.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Zero people.", + "B": "One person.", + "C": "Right now, there are two people directly in front of you.", + "D": "Right now, there are three people directly in front of you in total.", + "E": "Right now, there are four people directly in front of you in total." + }, + "answer_label": "C", + "answer_text": "Right now, there are two people directly in front of you.", + "q_id": "01452_8_T2", + "correct_answer": "Right now, there are two people directly in front of you.", + "original_options": { + "A": "Zero people.", + "B": "One person.", + "C": "Right now, there are two people directly in front of you.", + "D": "Right now, there are three people directly in front of you in total.", + "E": "Right now, there are four people directly in front of you in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "01452", + "video_length": "00:03:19", + "qid": "01452_8_T3", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many people are directly in front of me right now?", + "answer": "Right now, there is a person in front of you.", + "question_chinese": "现在我的正前方有多少人？", + "answer_chinese": "现在你的正前方有一个人。", + "evidence": [ + { + "description": "There is a woman in front of you now.", + "timestamps": { + "start": "00:02:03", + "end": "00:02:07" + } + } + ], + "question_timestamps": "00:02:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Zero people.", + "B": "Two people.", + "C": "Right now, there are three people directly in front of you at the moment.", + "D": "Right now, there is a person in front of you.", + "E": "Right now, there are four people directly in front of you at the moment." + }, + "answer_label": "D", + "answer_text": "Right now, there is a person in front of you.", + "q_id": "01452_8_T3", + "correct_answer": "Right now, there is a person in front of you.", + "original_options": { + "A": "Zero people.", + "B": "Two people.", + "C": "Right now, there are three people directly in front of you at the moment.", + "D": "Right now, there is a person in front of you.", + "E": "Right now, there are four people directly in front of you at the moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + } +] diff --git a/QAs/01453-mcq.json b/QAs/01453-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b0199be4d9a932e476805f761617959d0fe19043 --- /dev/null +++ b/QAs/01453-mcq.json @@ -0,0 +1,930 @@ +[ + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_16", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned in Haidilao so far?", + "answer": "two times.", + "question_chinese": "我现在在海底捞已经转了几次弯？", + "answer_chinese": "2次", + "evidence": [ + { + "description": "Turned left", + "timestamps": { + "start": "00:00:41", + "end": "00:00:43" + } + }, + { + "description": "I turned left at the intersection at this time", + "timestamps": { + "start": "00:00:46", + "end": "00:00:49" + } + } + ], + "question_timestamps": "00:00:51", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One time.", + "B": "Three times.", + "C": "You’ve turned one time so far inside Haidilao during your walk around the restaurant.", + "D": "You’ve turned three times so far while moving through Haidilao, before reaching your current spot.", + "E": "two times." + }, + "answer_label": "E", + "answer_text": "two times.", + "q_id": "01453_16", + "correct_answer": "two times.", + "original_options": { + "A": "One time.", + "B": "Three times.", + "C": "You’ve turned one time so far inside Haidilao during your walk around the restaurant.", + "D": "You’ve turned three times so far while moving through Haidilao, before reaching your current spot.", + "E": "two times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_17", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned in Haidilao so far?", + "answer": "three times.", + "question_chinese": "我现在在海底捞已经转了几次弯？", + "answer_chinese": "3次", + "evidence": [ + { + "description": "Turned left", + "timestamps": { + "start": "00:00:41", + "end": "00:00:43" + } + }, + { + "description": "I turned left at the intersection at this time", + "timestamps": { + "start": "00:00:46", + "end": "00:00:49" + } + }, + { + "description": "Turned right", + "timestamps": { + "start": "00:00:52", + "end": "00:00:53" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "three times.", + "C": "Four times.", + "D": "You’ve turned two times so far inside Haidilao while walking through the restaurant up to now.", + "E": "You’ve turned four times so far while walking around Haidilao, before arriving at your current location." + }, + "answer_label": "B", + "answer_text": "three times.", + "q_id": "01453_17", + "correct_answer": "three times.", + "original_options": { + "A": "Two times.", + "B": "three times.", + "C": "Four times.", + "D": "You’ve turned two times so far inside Haidilao while walking through the restaurant up to now.", + "E": "You’ve turned four times so far while walking around Haidilao, before arriving at your current location." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_12", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of meat have I eaten so far?", + "answer": "two pieces.", + "question_chinese": "我现在已经吃了几块肉？", + "answer_chinese": "2块", + "evidence": [ + { + "description": "There are two pieces of meat on the plate", + "timestamps": { + "start": "00:02:11", + "end": "00:02:14" + } + } + ], + "question_timestamps": "00:02:26", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "two pieces.", + "B": "One piece.", + "C": "Three pieces.", + "D": "You’ve eaten only one piece of meat so far, and you haven’t had a second piece yet.", + "E": "You’ve already eaten three pieces of meat up to this point, counting everything you swallowed earlier." + }, + "answer_label": "A", + "answer_text": "two pieces.", + "q_id": "01453_12", + "correct_answer": "two pieces.", + "original_options": { + "A": "two pieces.", + "B": "One piece.", + "C": "Three pieces.", + "D": "You’ve eaten only one piece of meat so far, and you haven’t had a second piece yet.", + "E": "You’ve already eaten three pieces of meat up to this point, counting everything you swallowed earlier." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 19, + 17 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_13", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of meat have I eaten so far?", + "answer": "four pieces.", + "question_chinese": "我现在已经吃了几块肉？", + "answer_chinese": "4块", + "evidence": [ + { + "description": "There are two pieces of meat on the plate", + "timestamps": { + "start": "00:02:11", + "end": "00:02:14" + } + }, + { + "description": "The waiter put another piece of meat", + "timestamps": { + "start": "00:02:36", + "end": "00:02:39" + } + }, + { + "description": "The waiter put another piece of meat", + "timestamps": { + "start": "00:03:36", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three pieces.", + "B": "Five pieces.", + "C": "You’ve eaten three pieces of meat so far, and you still haven’t reached four pieces yet.", + "D": "You’ve already eaten five pieces of meat up to now, including the most recent bite you finished.", + "E": "four pieces." + }, + "answer_label": "E", + "answer_text": "four pieces.", + "q_id": "01453_13", + "correct_answer": "four pieces.", + "original_options": { + "A": "Three pieces.", + "B": "Five pieces.", + "C": "You’ve eaten three pieces of meat so far, and you still haven’t reached four pieces yet.", + "D": "You’ve already eaten five pieces of meat up to now, including the most recent bite you finished.", + "E": "four pieces." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 18 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I go if I want to leave Haidilao now?", + "answer": "First turn around, then walk a few steps, turn right and keep walking straight to exit Haidilao.", + "question_chinese": "我现在想出海底捞应该怎么走", + "answer_chinese": "先向后转再走几步右转一直往前走可以出海底捞", + "evidence": [ + { + "description": "Entered Haidilao and walked straight first", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + }, + { + "description": "Turned left", + "timestamps": { + "start": "00:00:41", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left and walk straight out.", + "B": "Turn around, then turn left.", + "C": "First turn around, walk a few steps, turn left, and keep walking straight to leave Haidilao.", + "D": "First turn around, then walk a few steps, turn right and keep walking straight to exit Haidilao.", + "E": "Turn around first, walk a few steps, then turn right briefly before turning left and continuing straight out." + }, + "answer_label": "D", + "answer_text": "First turn around, then walk a few steps, turn right and keep walking straight to exit Haidilao.", + "q_id": "01453_1", + "correct_answer": "First turn around, then walk a few steps, turn right and keep walking straight to exit Haidilao.", + "original_options": { + "A": "Turn left and walk straight out.", + "B": "Turn around, then turn left.", + "C": "First turn around, walk a few steps, turn left, and keep walking straight to leave Haidilao.", + "D": "First turn around, then walk a few steps, turn right and keep walking straight to exit Haidilao.", + "E": "Turn around first, walk a few steps, then turn right briefly before turning left and continuing straight out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 18 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I go if I want to leave Haidilao now?", + "answer": "Stand up and turn right, walk a few steps, turn left at the intersection, then turn right at the next intersection, continue walking a few steps, turn right and keep walking straight to exit Haidilao.", + "question_chinese": "我现在想出海底捞应该怎么走", + "answer_chinese": "起身向右转走几步在路口左转再在下一个路口右转再继续走几步右转后一直往前走可以出海底捞", + "evidence": [ + { + "description": "Entered Haidilao and walked straight first", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + }, + { + "description": "Turned left", + "timestamps": { + "start": "00:00:41", + "end": "00:00:43" + } + }, + { + "description": "I turned left at the intersection at this time", + "timestamps": { + "start": "00:00:47", + "end": "00:00:50" + } + }, + { + "description": "Turned right", + "timestamps": { + "start": "00:00:51", + "end": "00:00:53" + } + }, + { + "description": "Walked a few more steps and chose the first table on the right to sit down", + "timestamps": { + "start": "00:00:54", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:04:17", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Stand up and turn left first.", + "B": "Stand up and turn right, walk a few steps, turn left at the intersection, then turn right at the next intersection, continue walking a few steps, turn right and keep walking straight to exit Haidilao.", + "C": "Turn right, then right again.", + "D": "Stand up and turn right, walk a few steps, turn right at the intersection, then continue straight to exit Haidilao.", + "E": "Stand up, turn right, walk a few steps, turn left at the first intersection, then turn left and go straight out." + }, + "answer_label": "B", + "answer_text": "Stand up and turn right, walk a few steps, turn left at the intersection, then turn right at the next intersection, continue walking a few steps, turn right and keep walking straight to exit Haidilao.", + "q_id": "01453_2", + "correct_answer": "Stand up and turn right, walk a few steps, turn left at the intersection, then turn right at the next intersection, continue walking a few steps, turn right and keep walking straight to exit Haidilao.", + "original_options": { + "A": "Stand up and turn left first.", + "B": "Stand up and turn right, walk a few steps, turn left at the intersection, then turn right at the next intersection, continue walking a few steps, turn right and keep walking straight to exit Haidilao.", + "C": "Turn right, then right again.", + "D": "Stand up and turn right, walk a few steps, turn right at the intersection, then continue straight to exit Haidilao.", + "E": "Stand up, turn right, walk a few steps, turn left at the first intersection, then turn left and go straight out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 21 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_14", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many more types of vegetables do I need to eat now?", + "answer": "four types.", + "question_chinese": "我现在还有多少种菜需要吃？", + "answer_chinese": "4种", + "evidence": [ + { + "description": "There is one plate of meat and three plates of vegetables on the table", + "timestamps": { + "start": "00:02:03", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three types.", + "B": "Five types.", + "C": "four types.", + "D": "You still need to eat three more types of vegetables right now, before you can be finished.", + "E": "You still need to eat five more types of vegetables at this point, based on what you’ve had already." + }, + "answer_label": "C", + "answer_text": "four types.", + "q_id": "01453_14", + "correct_answer": "four types.", + "original_options": { + "A": "Three types.", + "B": "Five types.", + "C": "four types.", + "D": "You still need to eat three more types of vegetables right now, before you can be finished.", + "E": "You still need to eat five more types of vegetables at this point, based on what you’ve had already." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 20 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_15", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many more types of vegetables do I need to eat now?", + "answer": "three types.", + "question_chinese": "我现在还有多少种菜需要吃？", + "answer_chinese": "3种", + "evidence": [ + { + "description": "One type of vegetable has been put into the hot pot", + "timestamps": { + "start": "00:03:49", + "end": "00:03:55" + } + }, + { + "description": "There are still 2 plates of vegetables on the table", + "timestamps": { + "start": "00:03:53", + "end": "00:03:56" + } + } + ], + "question_timestamps": "00:04:17", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "three types.", + "B": "Two types.", + "C": "Four types.", + "D": "You still need to eat two more types of vegetables right now, before you can be done.", + "E": "You still need to eat four more types of vegetables at this point, based on what you’ve had already." + }, + "answer_label": "A", + "answer_text": "three types.", + "q_id": "01453_15", + "correct_answer": "three types.", + "original_options": { + "A": "three types.", + "B": "Two types.", + "C": "Four types.", + "D": "You still need to eat two more types of vegetables right now, before you can be done.", + "E": "You still need to eat four more types of vegetables at this point, based on what you’ve had already." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 20 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_2_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far at Haidilao, how many yellow reminder signs have I seen?", + "answer": "One, to your right.", + "question_chinese": "现在我已经在海底捞看到了几个黄色提示牌？", + "answer_chinese": "1个，在你的右。", + "evidence": [ + { + "description": "There is a yellow warning sign at the entrance of Haidilao", + "timestamps": { + "start": "00:00:08", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero, to your right.", + "B": "Two, to your right.", + "C": "One, to your right.", + "D": "Three, all positioned along your right-hand side as you look around the area.", + "E": "Four, visible on your right, spread out so you notice them at different moments." + }, + "answer_label": "C", + "answer_text": "One, to your right.", + "q_id": "01453_2_T2", + "correct_answer": "One, to your right.", + "original_options": { + "A": "Zero, to your right.", + "B": "Two, to your right.", + "C": "One, to your right.", + "D": "Three, all positioned along your right-hand side as you look around the area.", + "E": "Four, visible on your right, spread out so you notice them at different moments." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_2_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far at Haidilao, how many yellow reminder signs have I seen?", + "answer": "One, to your back-right.", + "question_chinese": "现在我已经在海底捞看到了几个黄色提示牌？", + "answer_chinese": "1个，在你的右后方。", + "evidence": [ + { + "description": "There is a yellow warning sign at the entrance of Haidilao", + "timestamps": { + "start": "00:00:08", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero, to your back-right.", + "B": "One, to your back-right.", + "C": "Two, to your back-right.", + "D": "Three, located behind you on the right side when you turn your head slightly.", + "E": "Four, all appearing toward your back-right, so you’ve noticed several in that direction." + }, + "answer_label": "B", + "answer_text": "One, to your back-right.", + "q_id": "01453_2_T1", + "correct_answer": "One, to your back-right.", + "original_options": { + "A": "Zero, to your back-right.", + "B": "One, to your back-right.", + "C": "Two, to your back-right.", + "D": "Three, located behind you on the right side when you turn your head slightly.", + "E": "Four, all appearing toward your back-right, so you’ve noticed several in that direction." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_5", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many staff members have I encountered in Haidilao so far?", + "answer": "two.", + "question_chinese": "现在我已经在海底捞遇到多少个店员了？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "Encountered a staff member at the Haidilao entrance", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + }, + { + "description": "There is a staff member at the manicure area", + "timestamps": { + "start": "00:00:19", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "two.", + "D": "Six, since you’ve already interacted with multiple staff members during your time inside.", + "E": "Ten, because staff have come by repeatedly, so you’ve encountered many different employees." + }, + "answer_label": "C", + "answer_text": "two.", + "q_id": "01453_5", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "two.", + "D": "Six, since you’ve already interacted with multiple staff members during your time inside.", + "E": "Ten, because staff have come by repeatedly, so you’ve encountered many different employees." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_6", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many staff members have I encountered in Haidilao so far?", + "answer": "four.", + "question_chinese": "现在我已经在海底捞遇到多少个店员了？", + "answer_chinese": "4个", + "evidence": [ + { + "description": "Encountered a staff member at the Haidilao entrance", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + }, + { + "description": "There is a staff member at the manicure area", + "timestamps": { + "start": "00:00:19", + "end": "00:00:23" + } + }, + { + "description": "A waiter took my order, and another staff member brought me a doll", + "timestamps": { + "start": "00:01:03", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Five, because one additional staff member appeared after the first few encounters.", + "D": "four.", + "E": "Eight, since you’ve come across several different staff members as you moved around the restaurant." + }, + "answer_label": "D", + "answer_text": "four.", + "q_id": "01453_6", + "correct_answer": "four.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Five, because one additional staff member appeared after the first few encounters.", + "D": "four.", + "E": "Eight, since you’ve come across several different staff members as you moved around the restaurant." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 16 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_6_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "If I want to throw the napkin I used earlier into the trash can now, where should I go to pick up the used napkin?", + "answer": "The tissues on the table are to your front-left.", + "question_chinese": "现在我想把刚才我用过的餐巾纸丢进垃圾桶应该去哪里拿用过的餐巾纸？", + "answer_chinese": "桌子上的纸巾在你的左前方。", + "evidence": [ + { + "description": "After using the napkin, I put it on the left side of the tabel", + "timestamps": { + "start": "00:01:27", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:01:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The tissues are to your front-right.", + "B": "The tissues on the table are to your front-left.", + "C": "The tissues are behind you.", + "D": "The tissues on the table are to your back-left, so you’d turn around to reach them.", + "E": "The tissues on the table are directly in front of you, so you can grab them without turning." + }, + "answer_label": "B", + "answer_text": "The tissues on the table are to your front-left.", + "q_id": "01453_6_T2", + "correct_answer": "The tissues on the table are to your front-left.", + "original_options": { + "A": "The tissues are to your front-right.", + "B": "The tissues on the table are to your front-left.", + "C": "The tissues are behind you.", + "D": "The tissues on the table are to your back-left, so you’d turn around to reach them.", + "E": "The tissues on the table are directly in front of you, so you can grab them without turning." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 18 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_6_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "If I want to throw the napkin I used earlier into the trash can now, where should I go to pick up the used napkin?", + "answer": "To your front-left.", + "question_chinese": "现在我想把刚才我用过的餐巾纸丢进垃圾桶应该去哪里拿用过的餐巾纸？", + "answer_chinese": "你的左前方。", + "evidence": [ + { + "description": "After using the napkin, I put it on the left side of the table", + "timestamps": { + "start": "00:02:11", + "end": "00:02:13" + } + } + ], + "question_timestamps": "00:03:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-left.", + "B": "To your front-right.", + "C": "To your back-left.", + "D": "Directly in front of you, so you wouldn’t need to turn to pick it up.", + "E": "To your back-right, meaning you’d have to turn around before grabbing it." + }, + "answer_label": "A", + "answer_text": "To your front-left.", + "q_id": "01453_6_T1", + "correct_answer": "To your front-left.", + "original_options": { + "A": "To your front-left.", + "B": "To your front-right.", + "C": "To your back-left.", + "D": "Directly in front of you, so you wouldn’t need to turn to pick it up.", + "E": "To your back-right, meaning you’d have to turn around before grabbing it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 13 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_4_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to use a napkin right now, can I reach it directly?", + "answer": "Yes. The napkin is to your front-left.", + "question_chinese": "现在我想用餐巾纸能直接拿到吗？", + "answer_chinese": "能，餐巾纸在你的左前方。", + "evidence": [ + { + "description": "Napkins are placed on the left side of the table", + "timestamps": { + "start": "00:00:57", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. The napkin is to your front-left.", + "B": "No. The napkin is too far to reach directly." + }, + "answer_label": "A", + "answer_text": "Yes. The napkin is to your front-left.", + "q_id": "01453_4_T1", + "correct_answer": "Yes. The napkin is to your front-left.", + "original_options": { + "A": "Yes. The napkin is to your front-left.", + "B": "No. The napkin is too far to reach directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "01453", + "video_length": "00:04:17", + "qid": "01453_4_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to use a napkin right now, can I reach it directly?", + "answer": "Yes. The napkin is to your front-left.", + "question_chinese": "现在我想用餐巾纸能直接拿到吗？", + "answer_chinese": "能，餐巾纸在你的左前方。", + "evidence": [ + { + "description": "Napkins are placed on the left side of the table", + "timestamps": { + "start": "00:00:57", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:03:37", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. The napkin is too far to reach directly.", + "B": "Yes. The napkin is to your front-left." + }, + "answer_label": "B", + "answer_text": "Yes. The napkin is to your front-left.", + "q_id": "01453_4_T2", + "correct_answer": "Yes. The napkin is to your front-left.", + "original_options": { + "A": "No. The napkin is too far to reach directly.", + "B": "Yes. The napkin is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + } +] diff --git a/QAs/01469-mcq.json b/QAs/01469-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2d0d35d3d530dbedd3613f27fd68d3c1c4a1fedf --- /dev/null +++ b/QAs/01469-mcq.json @@ -0,0 +1,828 @@ +[ + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_5_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From the start until now, how have I moved?", + "answer": "You walked straight forward to the steps and went up them.", + "question_chinese": "从开始到现在我是怎么移动的？", + "answer_chinese": "我往前直走走到台阶上台阶.", + "evidence": [ + { + "description": "I kept walking forward", + "timestamps": { + "start": "00:00:01", + "end": "00:00:09" + } + }, + { + "description": "I walked up the steps", + "timestamps": { + "start": "00:00:31", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right and went up steps.", + "B": "You walked straight forward without stairs.", + "C": "You walked straight forward to the steps, but you went down them instead of going up.", + "D": "You walked forward to the steps, went up, then kept walking straight to an entrance and opened a door.", + "E": "You walked straight forward to the steps and went up them." + }, + "answer_label": "E", + "answer_text": "You walked straight forward to the steps and went up them.", + "q_id": "01469_5_T2", + "correct_answer": "You walked straight forward to the steps and went up them.", + "original_options": { + "A": "You turned right and went up steps.", + "B": "You walked straight forward without stairs.", + "C": "You walked straight forward to the steps, but you went down them instead of going up.", + "D": "You walked forward to the steps, went up, then kept walking straight to an entrance and opened a door.", + "E": "You walked straight forward to the steps and went up them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 19 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_5_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From the start until now, how have I moved?", + "answer": "You walked straight forward onto the steps, went up the steps, kept walking straight to the movie theater entrance, and pushed the door open to go in.", + "question_chinese": "从开始到现在我是怎么移动的？", + "answer_chinese": "我往前直走走到台阶上台阶继续直走到电影院门口并推门进入", + "evidence": [ + { + "description": "I kept walking forward", + "timestamps": { + "start": "00:00:01", + "end": "00:00:09" + } + }, + { + "description": "I walked up the steps", + "timestamps": { + "start": "00:00:31", + "end": "00:00:44" + } + }, + { + "description": "Climbed the steps and walked straight to the cinema entrance and pushed the door to enter", + "timestamps": { + "start": "00:01:06", + "end": "00:01:29" + } + }, + { + "description": "Staff guided me to enter and walk straight", + "timestamps": { + "start": "00:02:00", + "end": "00:02:04" + } + } + ], + "question_timestamps": "00:02:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You only went up the steps.", + "B": "You turned around and walked backward.", + "C": "You walked straight forward to the steps and went up them, but you did not reach any entrance or open a door.", + "D": "You walked straight forward onto the steps, went up the steps, kept walking straight to the movie theater entrance, and pushed the door open to go in.", + "E": "You went up the steps, then turned right to a nearby entrance instead of continuing straight and opening the door." + }, + "answer_label": "D", + "answer_text": "You walked straight forward onto the steps, went up the steps, kept walking straight to the movie theater entrance, and pushed the door open to go in.", + "q_id": "01469_5_T1", + "correct_answer": "You walked straight forward onto the steps, went up the steps, kept walking straight to the movie theater entrance, and pushed the door open to go in.", + "original_options": { + "A": "You only went up the steps.", + "B": "You turned around and walked backward.", + "C": "You walked straight forward to the steps and went up them, but you did not reach any entrance or open a door.", + "D": "You walked straight forward onto the steps, went up the steps, kept walking straight to the movie theater entrance, and pushed the door open to go in.", + "E": "You went up the steps, then turned right to a nearby entrance instead of continuing straight and opening the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 20 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_6_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From the start up to now, how many trash bins have appeared in my view?", + "answer": "One trash can appeared in front of you.", + "question_chinese": "从开始到现在我的视线中出现了多少个垃圾桶？", + "answer_chinese": "你的面前出现了1个垃圾桶", + "evidence": [ + { + "description": "One trash can appeared on the right side in my sight", + "timestamps": { + "start": "00:01:23", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero trash cans appeared.", + "B": "Two trash cans appeared.", + "C": "Three trash cans appeared in front of you during the walk so far, all visible at separate moments.", + "D": "One trash can appeared in front of you.", + "E": "Four trash cans came into your view from the start, with at least one appearing off to the side." + }, + "answer_label": "D", + "answer_text": "One trash can appeared in front of you.", + "q_id": "01469_6_T2", + "correct_answer": "One trash can appeared in front of you.", + "original_options": { + "A": "Zero trash cans appeared.", + "B": "Two trash cans appeared.", + "C": "Three trash cans appeared in front of you during the walk so far, all visible at separate moments.", + "D": "One trash can appeared in front of you.", + "E": "Four trash cans came into your view from the start, with at least one appearing off to the side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_6_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From the start up to now, how many trash bins have appeared in my view?", + "answer": "There are two trash cans in front of you.", + "question_chinese": "从开始到现在我的视线中出现了多少个垃圾桶？", + "answer_chinese": "你面前有2个垃圾桶", + "evidence": [ + { + "description": "One trash can appeared on the right side in my sight", + "timestamps": { + "start": "00:01:23", + "end": "00:01:26" + } + }, + { + "description": "One trash can appeared on the right side in my sight", + "timestamps": { + "start": "00:01:33", + "end": "00:01:36" + } + } + ], + "question_timestamps": "00:02:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are no trash cans.", + "B": "There is one trash can.", + "C": "There are two trash cans in front of you.", + "D": "There are three trash cans in front of you right now, spaced a short distance apart.", + "E": "There are four trash cans in front of you, with two closer and the others slightly farther away." + }, + "answer_label": "C", + "answer_text": "There are two trash cans in front of you.", + "q_id": "01469_6_T1", + "correct_answer": "There are two trash cans in front of you.", + "original_options": { + "A": "There are no trash cans.", + "B": "There is one trash can.", + "C": "There are two trash cans in front of you.", + "D": "There are three trash cans in front of you right now, spaced a short distance apart.", + "E": "There are four trash cans in front of you, with two closer and the others slightly farther away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_7_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From the start up to now, how many full turns have I made?", + "answer": "You didn’t turn in a circle.", + "question_chinese": "从开始到现在我转了几圈？", + "answer_chinese": "你没有转圈。", + "evidence": [ + { + "description": "IYou have been going straight.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You turned around once.", + "B": "You turned around twice.", + "C": "You made three full turns, completing a full circle each time before continuing forward.", + "D": "You completed four full turns in place, ending up facing forward after the last rotation.", + "E": "You didn’t turn in a circle." + }, + "answer_label": "E", + "answer_text": "You didn’t turn in a circle.", + "q_id": "01469_7_T2", + "correct_answer": "You didn’t turn in a circle.", + "original_options": { + "A": "You turned around once.", + "B": "You turned around twice.", + "C": "You made three full turns, completing a full circle each time before continuing forward.", + "D": "You completed four full turns in place, ending up facing forward after the last rotation.", + "E": "You didn’t turn in a circle." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_7_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From the start up to now, how many full turns have I made?", + "answer": "You turned around once.", + "question_chinese": "从开始到现在我转了几圈？", + "answer_chinese": "你转了一圈", + "evidence": [ + { + "description": "I turned one circle", + "timestamps": { + "start": "00:01:58", + "end": "00:02:03" + } + } + ], + "question_timestamps": "00:02:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You didn’t turn fully.", + "B": "You turned around twice.", + "C": "You turned around three times, completing full circles rather than just pivoting slightly.", + "D": "You turned around once.", + "E": "You turned around four times in total, making complete rotations before moving on." + }, + "answer_label": "D", + "answer_text": "You turned around once.", + "q_id": "01469_7_T1", + "correct_answer": "You turned around once.", + "original_options": { + "A": "You didn’t turn fully.", + "B": "You turned around twice.", + "C": "You turned around three times, completing full circles rather than just pivoting slightly.", + "D": "You turned around once.", + "E": "You turned around four times in total, making complete rotations before moving on." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 13 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_1_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many steps have I climbed so far?", + "answer": "You went up 2 of the steps in front of you.", + "question_chinese": "现在我已经上了多少个台阶？", + "answer_chinese": "你面前的台阶你上了2个", + "evidence": [ + { + "description": "The steps are right in front of you.", + "timestamps": { + "start": "00:00:29", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You climbed zero steps.", + "B": "You climbed one step.", + "C": "You went up 3 of the steps in front of you, reaching the third step before stopping.", + "D": "You went up 2 of the steps in front of you.", + "E": "You went up 4 of the steps in front of you, continuing past the first few risers." + }, + "answer_label": "D", + "answer_text": "You went up 2 of the steps in front of you.", + "q_id": "01469_1_T2", + "correct_answer": "You went up 2 of the steps in front of you.", + "original_options": { + "A": "You climbed zero steps.", + "B": "You climbed one step.", + "C": "You went up 3 of the steps in front of you, reaching the third step before stopping.", + "D": "You went up 2 of the steps in front of you.", + "E": "You went up 4 of the steps in front of you, continuing past the first few risers." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_1_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many steps have I climbed so far?", + "answer": "You went up 8 of the steps in front of you.", + "question_chinese": "现在我已经上了多少个台阶？", + "answer_chinese": "你面前的台阶你上了8个", + "evidence": [ + { + "description": "Now walked 8 steps", + "timestamps": { + "start": "00:00:33", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You went up 6 steps.", + "B": "You went up 7 steps.", + "C": "You went up 8 of the steps in front of you.", + "D": "You went up 9 of the steps in front of you, climbing one more step beyond eight.", + "E": "You went up 10 of the steps in front of you, reaching the tenth step before pausing." + }, + "answer_label": "C", + "answer_text": "You went up 8 of the steps in front of you.", + "q_id": "01469_1_T1", + "correct_answer": "You went up 8 of the steps in front of you.", + "original_options": { + "A": "You went up 6 steps.", + "B": "You went up 7 steps.", + "C": "You went up 8 of the steps in front of you.", + "D": "You went up 9 of the steps in front of you, climbing one more step beyond eight.", + "E": "You went up 10 of the steps in front of you, reaching the tenth step before pausing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_3_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "If I want to rest now, what options do I have?", + "answer": "One option: rest on the bench in front of you.", + "question_chinese": "现在我想休息有哪几种选择？", + "answer_chinese": "1种，在你面前的长椅上休息", + "evidence": [ + { + "description": "There is a bench on my right side", + "timestamps": { + "start": "00:00:44", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero options to rest.", + "B": "Two options to rest.", + "C": "Three options: you can rest on the bench in front of you, considering it as multiple separate choices.", + "D": "Four options: you have several resting choices, with the bench in front of you counted as more than one.", + "E": "One option: rest on the bench in front of you." + }, + "answer_label": "E", + "answer_text": "One option: rest on the bench in front of you.", + "q_id": "01469_3_T1", + "correct_answer": "One option: rest on the bench in front of you.", + "original_options": { + "A": "Zero options to rest.", + "B": "Two options to rest.", + "C": "Three options: you can rest on the bench in front of you, considering it as multiple separate choices.", + "D": "Four options: you have several resting choices, with the bench in front of you counted as more than one.", + "E": "One option: rest on the bench in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_3_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "If I want to rest now, what options do I have?", + "answer": "Two options: you can rest on the chair in front of you, or on the massage chair to your right.", + "question_chinese": "现在我想休息有哪几种选择？", + "answer_chinese": "两种，你可以在你面前的椅子上休息也可以在你右边的按摩椅上休息。", + "evidence": [ + { + "description": "On the left is a chair, and on the right is a massage chair.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:38", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two options: you can rest on the chair in front of you, or on the massage chair to your right.", + "B": "One option: the chair ahead.", + "C": "One option: the massage chair right.", + "D": "Three options: rest on the chair in front, the massage chair right, and a bench behind you.", + "E": "Zero options: there isn’t anywhere you can sit to rest right now in your immediate view." + }, + "answer_label": "A", + "answer_text": "Two options: you can rest on the chair in front of you, or on the massage chair to your right.", + "q_id": "01469_3_T2", + "correct_answer": "Two options: you can rest on the chair in front of you, or on the massage chair to your right.", + "original_options": { + "A": "Two options: you can rest on the chair in front of you, or on the massage chair to your right.", + "B": "One option: the chair ahead.", + "C": "One option: the massage chair right.", + "D": "Three options: rest on the chair in front, the massage chair right, and a bench behind you.", + "E": "Zero options: there isn’t anywhere you can sit to rest right now in your immediate view." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 17 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_2_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "If I want to go watch a movie now, at minimum how many more steps do I still need to climb?", + "answer": "You still need to go up four more steps in front of you.", + "question_chinese": "现在我想去看电影至少还要上多少个台阶？", + "answer_chinese": "还要再上你面前的4个台阶。", + "evidence": [ + { + "description": "The steps are right in front of you.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three more steps.", + "B": "Five more steps.", + "C": "You still need to climb three more steps in front of you before you can head to the movie.", + "D": "You still need to go up four more steps in front of you.", + "E": "You still need to go up five more steps directly in front of you to reach the movie area." + }, + "answer_label": "D", + "answer_text": "You still need to go up four more steps in front of you.", + "q_id": "01469_2_T1", + "correct_answer": "You still need to go up four more steps in front of you.", + "original_options": { + "A": "Three more steps.", + "B": "Five more steps.", + "C": "You still need to climb three more steps in front of you before you can head to the movie.", + "D": "You still need to go up four more steps in front of you.", + "E": "You still need to go up five more steps directly in front of you to reach the movie area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 19, + 19 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_2_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "If I want to go watch a movie now, at minimum how many more steps do I still need to climb?", + "answer": "You still need to go up two more steps in front of you.", + "question_chinese": "现在我想去看电影至少还要上多少个台阶？", + "answer_chinese": "还要再上你面前的2个台阶。", + "evidence": [ + { + "description": "The steps are right in front of you.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One more step.", + "B": "Three more steps.", + "C": "You still need to go up one more step in front of you to get there.", + "D": "You still need to climb three more steps straight ahead before you can watch a movie.", + "E": "You still need to go up two more steps in front of you." + }, + "answer_label": "E", + "answer_text": "You still need to go up two more steps in front of you.", + "q_id": "01469_2_T2", + "correct_answer": "You still need to go up two more steps in front of you.", + "original_options": { + "A": "One more step.", + "B": "Three more steps.", + "C": "You still need to go up one more step in front of you to get there.", + "D": "You still need to climb three more steps straight ahead before you can watch a movie.", + "E": "You still need to go up two more steps in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many more steps do I need to climb at least to go to the movies now?", + "answer": "Two.", + "question_chinese": "现在我想去看电影至少还要上多少个台阶？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "The cinema on the right requires climbing 2 more steps", + "timestamps": { + "start": "00:01:45", + "end": "00:01:59" + } + } + ], + "question_timestamps": "00:02:07", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You need at least four more steps to reach the movie area from where you are now.", + "E": "You need at least five more steps ahead before you can get to the movies from here." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "01469_2", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You need at least four more steps to reach the movie area from where you are now.", + "E": "You need at least five more steps ahead before you can get to the movies from here." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 17 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_4_T1", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to sit down and rest now, can I get there within 5 steps?", + "answer": "No—you can’t. The chair is 10 steps straight in front of you.", + "question_chinese": "现在我想坐下休息可以5步之内到达吗？", + "answer_chinese": "不可以，椅子在你的正前方10步远。", + "evidence": [ + { + "description": "There are seats on my right side", + "timestamps": { + "start": "00:00:44", + "end": "00:00:47" + } + }, + { + "description": "I walked forward many steps", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. The chair is only a few steps straight in front of you.", + "B": "No—you can’t. The chair is 10 steps straight in front of you." + }, + "answer_label": "B", + "answer_text": "No—you can’t. The chair is 10 steps straight in front of you.", + "q_id": "01469_4_T1", + "correct_answer": "No—you can’t. The chair is 10 steps straight in front of you.", + "original_options": { + "A": "Yes—you can. The chair is only a few steps straight in front of you.", + "B": "No—you can’t. The chair is 10 steps straight in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "01469", + "video_length": "00:02:07", + "qid": "01469_4_T2", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to sit down and rest now, can I get there within 5 steps?", + "answer": "Yes—you can. There are chairs in front of you and to your right.", + "question_chinese": "现在我想坐下休息可以5步之内到达吗？", + "answer_chinese": "可以，你的正前方和右方都有椅子。", + "evidence": [ + { + "description": "On the left is a chair, and on the right is a massage chair.", + "timestamps": { + "start": "00:01:44", + "end": "00:01:47" + } + }, + { + "description": "I walked forward many steps", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:49", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. There are chairs in front of you and to your right.", + "B": "No—you can’t. The chairs are farther than 5 steps away." + }, + "answer_label": "A", + "answer_text": "Yes—you can. There are chairs in front of you and to your right.", + "q_id": "01469_4_T2", + "correct_answer": "Yes—you can. There are chairs in front of you and to your right.", + "original_options": { + "A": "Yes—you can. There are chairs in front of you and to your right.", + "B": "No—you can’t. The chairs are farther than 5 steps away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + } +] diff --git a/QAs/01470-mcq.json b/QAs/01470-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f195e8a9866c37c5a28a878058419c52d2a4885d --- /dev/null +++ b/QAs/01470-mcq.json @@ -0,0 +1,584 @@ +[ + { + "video_uid": "01470", + "video_length": "00:01:22", + "qid": "01470_1_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many burners on the stovetop can I use right now?", + "answer": "One, on your right.", + "question_chinese": "我现在灶台上有几个灶可以使用？", + "answer_chinese": "1个，在你的右方。", + "evidence": [ + { + "description": "The left stove on the cooktop is occupied, the pot on the right stove can be used", + "timestamps": { + "start": "00:00:29", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:00:34", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero, on your right.", + "B": "Two, on your right.", + "C": "You can use two burners right now, both located on your right side of the stovetop.", + "D": "You can’t use any burners right now; none of the burners on your right are available.", + "E": "One, on your right." + }, + "answer_label": "E", + "answer_text": "One, on your right.", + "q_id": "01470_1_T1", + "correct_answer": "One, on your right.", + "original_options": { + "A": "Zero, on your right.", + "B": "Two, on your right.", + "C": "You can use two burners right now, both located on your right side of the stovetop.", + "D": "You can’t use any burners right now; none of the burners on your right are available.", + "E": "One, on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "01470", + "video_length": "00:01:22", + "qid": "01470_1_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many burners on the stovetop can I use right now?", + "answer": "One, directly to your right.", + "question_chinese": "我现在灶台上有几个灶可以使用？", + "answer_chinese": "1个，在你的正右方", + "evidence": [ + { + "description": "The left stove on the cooktop is occupied, the pot on the right stove can be used", + "timestamps": { + "start": "00:00:29", + "end": "00:00:32" + } + }, + { + "description": "The pot on the left stove on the cooktop has been removed, the pot on the right stove can be used", + "timestamps": { + "start": "00:00:45", + "end": "00:00:53" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, directly to your right.", + "B": "Zero, directly to your right.", + "C": "Two, directly to your right.", + "D": "You can use two burners right now, and they’re directly to your right on the stovetop.", + "E": "You can use four burners right now, all positioned directly to your right on the stovetop." + }, + "answer_label": "A", + "answer_text": "One, directly to your right.", + "q_id": "01470_1_T2", + "correct_answer": "One, directly to your right.", + "original_options": { + "A": "One, directly to your right.", + "B": "Zero, directly to your right.", + "C": "Two, directly to your right.", + "D": "You can use two burners right now, and they’re directly to your right on the stovetop.", + "E": "You can use four burners right now, all positioned directly to your right on the stovetop." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "01470", + "video_length": "00:01:22", + "qid": "01470_2_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Which steps of making yogurt have I completed so far?", + "answer": "I heated fresh milk, added granulated sugar while it was heating, turned off the heat once it boiled, let it sit to cool, then added yogurt and stirred until even, and left it for 3 hours. Right now, the yogurt is in front of you.", + "question_chinese": "现在我已经做了酸奶的哪些步骤？", + "answer_chinese": "我放入鲜牛奶加热期间加入白砂糖沸腾后关火，静置冷却后放入酸奶搅拌均匀，放3小时, 目前酸奶在我的前方。", + "evidence": [ + { + "description": "First put in fresh milk", + "timestamps": { + "start": "00:00:14", + "end": "00:00:17" + } + }, + { + "description": "Added white sugar", + "timestamps": { + "start": "00:00:35", + "end": "00:00:40" + } + }, + { + "description": "Boiled and turned off the heat", + "timestamps": { + "start": "00:00:42", + "end": "00:00:45" + } + }, + { + "description": "Let it cool, then added yogurt and stirred evenly", + "timestamps": { + "start": "00:00:51", + "end": "00:01:01" + } + }, + { + "description": "Let it sit for 3 hours", + "timestamps": { + "start": "00:01:09", + "end": "00:01:12" + } + } + ], + "question_timestamps": "00:01:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You left it for 1 hour.", + "B": "I heated fresh milk, added granulated sugar while it was heating, turned off the heat once it boiled, let it sit to cool, then added yogurt and stirred until even, and left it for 3 hours. Right now, the yogurt is in front of you.", + "C": "You left it for 2 hours.", + "D": "You heated fresh milk, stirred in yogurt, and left it for 2 hours; it is not ready yet.", + "E": "You boiled the milk, skipped adding sugar, then added yogurt and left it for 4 hours before checking." + }, + "answer_label": "B", + "answer_text": "I heated fresh milk, added granulated sugar while it was heating, turned off the heat once it boiled, let it sit to cool, then added yogurt and stirred until even, and left it for 3 hours. Right now, the yogurt is in front of you.", + "q_id": "01470_2_T2", + "correct_answer": "I heated fresh milk, added granulated sugar while it was heating, turned off the heat once it boiled, let it sit to cool, then added yogurt and stirred until even, and left it for 3 hours. Right now, the yogurt is in front of you.", + "original_options": { + "A": "You left it for 1 hour.", + "B": "I heated fresh milk, added granulated sugar while it was heating, turned off the heat once it boiled, let it sit to cool, then added yogurt and stirred until even, and left it for 3 hours. Right now, the yogurt is in front of you.", + "C": "You left it for 2 hours.", + "D": "You heated fresh milk, stirred in yogurt, and left it for 2 hours; it is not ready yet.", + "E": "You boiled the milk, skipped adding sugar, then added yogurt and left it for 4 hours before checking." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "01470", + "video_length": "00:01:22", + "qid": "01470_2_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Which steps of making yogurt have I completed so far?", + "answer": "I heated fresh milk, added granulated sugar while it was heating, turned off the heat once it boiled, let it sit to cool, then added yogurt and stirred until even, and left it for 3 hours. Right now, the yogurt is in front of you.", + "question_chinese": "现在我已经做了酸奶的哪些步骤？", + "answer_chinese": "我放入鲜牛奶加热期间加入白砂糖沸腾后关火，静置冷却后放入酸奶搅拌均匀，放3小时, 目前酸奶在我的前方。", + "evidence": [ + { + "description": "First put in fresh milk", + "timestamps": { + "start": "00:00:14", + "end": "00:00:17" + } + }, + { + "description": "Added white sugar", + "timestamps": { + "start": "00:00:35", + "end": "00:00:40" + } + }, + { + "description": "Boiled and turned off the heat", + "timestamps": { + "start": "00:00:42", + "end": "00:00:45" + } + }, + { + "description": "Let it cool, then added yogurt and stirred evenly", + "timestamps": { + "start": "00:00:51", + "end": "00:01:01" + } + }, + { + "description": "Let it sit for 3 hours", + "timestamps": { + "start": "00:01:09", + "end": "00:01:12" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "I heated fresh milk, added granulated sugar while it was heating, turned off the heat once it boiled, let it sit to cool, then added yogurt and stirred until even, and left it for 3 hours. Right now, the yogurt is in front of you.", + "B": "You only heated the milk.", + "C": "You added yogurt before cooling.", + "D": "You heated the fresh milk with sugar, then mixed in yogurt and left it to set for only 1 hour.", + "E": "You heated and sweetened the milk, but you haven’t added the yogurt yet; it’s still cooling on the counter." + }, + "answer_label": "A", + "answer_text": "I heated fresh milk, added granulated sugar while it was heating, turned off the heat once it boiled, let it sit to cool, then added yogurt and stirred until even, and left it for 3 hours. Right now, the yogurt is in front of you.", + "q_id": "01470_2_T1", + "correct_answer": "I heated fresh milk, added granulated sugar while it was heating, turned off the heat once it boiled, let it sit to cool, then added yogurt and stirred until even, and left it for 3 hours. Right now, the yogurt is in front of you.", + "original_options": { + "A": "I heated fresh milk, added granulated sugar while it was heating, turned off the heat once it boiled, let it sit to cool, then added yogurt and stirred until even, and left it for 3 hours. Right now, the yogurt is in front of you.", + "B": "You only heated the milk.", + "C": "You added yogurt before cooling.", + "D": "You heated the fresh milk with sugar, then mixed in yogurt and left it to set for only 1 hour.", + "E": "You heated and sweetened the milk, but you haven’t added the yogurt yet; it’s still cooling on the counter." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 21 + ] + }, + { + "video_uid": "01470", + "video_length": "00:01:22", + "qid": "01470_3_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many kettles have I seen in total?", + "answer": "One, on your left.", + "question_chinese": "现在我已经看到了多少个水壶？", + "answer_chinese": "1个，在你的左方。", + "evidence": [ + { + "description": "1 bottle.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:46" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero kettles.", + "B": "One, on your left.", + "C": "Two kettles.", + "D": "Three kettles total, with one positioned on your left side near the counter as you looked over.", + "E": "Four kettles altogether, and the most recent one you noticed was on your left." + }, + "answer_label": "B", + "answer_text": "One, on your left.", + "q_id": "01470_3_T1", + "correct_answer": "One, on your left.", + "original_options": { + "A": "Zero kettles.", + "B": "One, on your left.", + "C": "Two kettles.", + "D": "Three kettles total, with one positioned on your left side near the counter as you looked over.", + "E": "Four kettles altogether, and the most recent one you noticed was on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 14 + ] + }, + { + "video_uid": "01470", + "video_length": "00:01:22", + "qid": "01470_3_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many kettles have I seen in total?", + "answer": "One, in front of you.", + "question_chinese": "现在我已经看到了多少个水壶？", + "answer_chinese": "1个,在你的前方。", + "evidence": [ + { + "description": "1 bottle", + "timestamps": { + "start": "00:00:50", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero kettles.", + "B": "Two kettles.", + "C": "One, in front of you.", + "D": "Three kettles total, with one positioned in front of you near the workspace when you looked over.", + "E": "Four kettles altogether, and the most recent one you noticed was in front of you." + }, + "answer_label": "C", + "answer_text": "One, in front of you.", + "q_id": "01470_3_T2", + "correct_answer": "One, in front of you.", + "original_options": { + "A": "Zero kettles.", + "B": "Two kettles.", + "C": "One, in front of you.", + "D": "Three kettles total, with one positioned in front of you near the workspace when you looked over.", + "E": "Four kettles altogether, and the most recent one you noticed was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 15 + ] + }, + { + "video_uid": "01470", + "video_length": "00:01:22", + "qid": "01470_9", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the kettle directly if I want to get it now?", + "answer": "Yes.", + "question_chinese": "现在我想拿水壶可以直接够到吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The kettle is on the left side of the kitchen cooktop", + "timestamps": { + "start": "00:00:04", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "01470_9", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01470", + "video_length": "00:01:22", + "qid": "01470_10", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the kettle directly if I want to get it now?", + "answer": "No.", + "question_chinese": "现在我想拿水壶可以直接够到吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "The kettle is on the left side of the kitchen cooktop", + "timestamps": { + "start": "00:00:04", + "end": "00:00:08" + } + }, + { + "description": "I am now in the living room", + "timestamps": { + "start": "00:01:04", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01470_10", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01470", + "video_length": "00:01:22", + "qid": "01470_7", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I get the spoon from if I want to use it now?", + "answer": "Right side.", + "question_chinese": "现在我想用勺子，应该从哪里拿？", + "answer_chinese": "右侧", + "evidence": [ + { + "description": "The spoon is on the left side of the pot", + "timestamps": { + "start": "00:00:31", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "Left side.", + "C": "Middle of the table.", + "D": "Behind the faucet, next to the sink area where you last set things down.", + "E": "Underneath the placemat, where it’s partially hidden near the table setting." + }, + "answer_label": "A", + "answer_text": "Right side.", + "q_id": "01470_7", + "correct_answer": "Right side.", + "original_options": { + "A": "Right side.", + "B": "Left side.", + "C": "Middle of the table.", + "D": "Behind the faucet, next to the sink area where you last set things down.", + "E": "Underneath the placemat, where it’s partially hidden near the table setting." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 14, + 12 + ] + }, + { + "video_uid": "01470", + "video_length": "00:01:22", + "qid": "01470_8", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I get the spoon from if I want to use it now?", + "answer": "Left side.", + "question_chinese": "现在我想用勺子，应该从哪里拿？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The spoon is on the left side of the pot", + "timestamps": { + "start": "00:00:31", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "Left side.", + "C": "Front side.", + "D": "Back side, closer to where the items are placed farther from you.", + "E": "Top shelf, above the counter where you would normally store utensils and small tools." + }, + "answer_label": "B", + "answer_text": "Left side.", + "q_id": "01470_8", + "correct_answer": "Left side.", + "original_options": { + "A": "Right side.", + "B": "Left side.", + "C": "Front side.", + "D": "Back side, closer to where the items are placed farther from you.", + "E": "Top shelf, above the counter where you would normally store utensils and small tools." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + } +] diff --git a/QAs/01471-mcq.json b/QAs/01471-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..09f5d228a20fee375c0df9d0c94bec6fd5d7a324 --- /dev/null +++ b/QAs/01471-mcq.json @@ -0,0 +1,1036 @@ +[ + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_11", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many electric bicycles or bicycles are there from leaving the road to entering the workplace?", + "answer": "three.", + "question_chinese": "从离开马路到进入上班的地方有多少辆电动车或自行车？", + "answer_chinese": "3辆", + "evidence": [ + { + "description": "There are three electric bicycles at the unit entrance now", + "timestamps": { + "start": "00:01:05", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Five bicycles in total along that stretch from the road to the workplace entrance.", + "D": "three.", + "E": "Seven bicycles altogether between leaving the road and entering the workplace area." + }, + "answer_label": "D", + "answer_text": "three.", + "q_id": "01471_11", + "correct_answer": "three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Five bicycles in total along that stretch from the road to the workplace entrance.", + "D": "three.", + "E": "Seven bicycles altogether between leaving the road and entering the workplace area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 12 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_12", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many electric bicycles or bicycles are there from leaving the road to entering the workplace?", + "answer": "Four.", + "question_chinese": "从离开马路到进入上班的地方有多少辆电动车或自行车？", + "answer_chinese": "4辆", + "evidence": [ + { + "description": "There are three electric bicycles at the unit entrance now", + "timestamps": { + "start": "00:01:05", + "end": "00:01:07" + } + }, + { + "description": "There is another bicycle near the work window", + "timestamps": { + "start": "00:01:29", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:01:51", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "You saw five electric bicycles or bicycles before you entered the workplace area.", + "D": "Four.", + "E": "You saw six electric bicycles or bicycles on the way from leaving the road to entering the workplace." + }, + "answer_label": "D", + "answer_text": "Four.", + "q_id": "01471_12", + "correct_answer": "Four.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "You saw five electric bicycles or bicycles before you entered the workplace area.", + "D": "Four.", + "E": "You saw six electric bicycles or bicycles on the way from leaving the road to entering the workplace." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 18 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_9_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was coming home from work, which direction was the car parking area relative to me?", + "answer": "On the left side.", + "question_chinese": "我下班回家的时候，汽车的停车区在我的哪个方位？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "When going to work, the parking position is on my right side", + "timestamps": { + "start": "00:00:36", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the right side.", + "B": "Straight ahead.", + "C": "Behind you as you walked toward home from work the whole time.", + "D": "On the left side.", + "E": "On your front-right side while you were coming home from work today." + }, + "answer_label": "D", + "answer_text": "On the left side.", + "q_id": "01471_9_T1", + "correct_answer": "On the left side.", + "original_options": { + "A": "On the right side.", + "B": "Straight ahead.", + "C": "Behind you as you walked toward home from work the whole time.", + "D": "On the left side.", + "E": "On your front-right side while you were coming home from work today." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_9_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was coming home from work, which direction was the car parking area relative to me?", + "answer": "On the right side.", + "question_chinese": "我下班回家的时候，汽车的停车区在我的哪个方位？", + "answer_chinese": "右侧", + "evidence": [ + { + "description": "When going to work, the parking position is on my right side", + "timestamps": { + "start": "00:00:36", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the left side.", + "B": "Straight ahead.", + "C": "Behind you as you walked toward home from work the whole time.", + "D": "On your front-left side while you were coming home from work today.", + "E": "On the right side." + }, + "answer_label": "E", + "answer_text": "On the right side.", + "q_id": "01471_9_T2", + "correct_answer": "On the right side.", + "original_options": { + "A": "On the left side.", + "B": "Straight ahead.", + "C": "Behind you as you walked toward home from work the whole time.", + "D": "On your front-left side while you were coming home from work today.", + "E": "On the right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_13", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which side is the trash can on for me now?", + "answer": "Left side.", + "question_chinese": "现在垃圾桶在我的哪侧？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The trash can is in front of me", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "In front of you.", + "C": "Behind you, near the back of the room, not beside you now.", + "D": "On your front-right side as you are standing right here now today.", + "E": "Left side." + }, + "answer_label": "E", + "answer_text": "Left side.", + "q_id": "01471_13", + "correct_answer": "Left side.", + "original_options": { + "A": "Right side.", + "B": "In front of you.", + "C": "Behind you, near the back of the room, not beside you now.", + "D": "On your front-right side as you are standing right here now today.", + "E": "Left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 12, + 12 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_14", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which side is the trash can on for me now?", + "answer": "Right side.", + "question_chinese": "现在垃圾桶在我的哪侧？", + "answer_chinese": "右侧", + "evidence": [ + { + "description": "The trash can is in front of me", + "timestamps": { + "start": "00:01:03", + "end": "00:01:06" + } + }, + { + "description": "I walked straight toward the unit building for a few steps and turned right", + "timestamps": { + "start": "00:01:19", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "Left side.", + "C": "Behind you.", + "D": "In front of you, slightly off to the left as you face forward now.", + "E": "On your left-front side relative to where you are standing right now." + }, + "answer_label": "A", + "answer_text": "Right side.", + "q_id": "01471_14", + "correct_answer": "Right side.", + "original_options": { + "A": "Right side.", + "B": "Left side.", + "C": "Behind you.", + "D": "In front of you, slightly off to the left as you face forward now.", + "E": "On your left-front side relative to where you are standing right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 12 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many unit entrances have I passed so far?", + "answer": "two.", + "question_chinese": "现在我已经经过了几个单元口？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "I am currently at the unit entrance", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "There is another unit entrance", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You have passed four unit entrances so far on this route today.", + "D": "So far, you have gone by six unit entrances along the way.", + "E": "two." + }, + "answer_label": "E", + "answer_text": "two.", + "q_id": "01471_3", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You have passed four unit entrances so far on this route today.", + "D": "So far, you have gone by six unit entrances along the way.", + "E": "two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many unit entrances have I passed so far?", + "answer": "three.", + "question_chinese": "现在我已经经过了几个单元口？", + "answer_chinese": "3个", + "evidence": [ + { + "description": "I am currently at the unit entrance", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "There is another unit entrance", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + }, + { + "description": "There is another unit entrance", + "timestamps": { + "start": "00:00:50", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:01:51", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "You have passed one unit entrance so far on this route today.", + "D": "three.", + "E": "So far, you have already passed five unit entrances on your route." + }, + "answer_label": "D", + "answer_text": "three.", + "q_id": "01471_4", + "correct_answer": "three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "You have passed one unit entrance so far on this route today.", + "D": "three.", + "E": "So far, you have already passed five unit entrances on your route." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_5", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cars have appeared within my field of vision?", + "answer": "Six.", + "question_chinese": "现在我的实现范围内出现了几辆车？", + "answer_chinese": "6辆", + "evidence": [ + { + "description": "Three cars parked by the roadside", + "timestamps": { + "start": "00:00:23", + "end": "00:00:26" + } + }, + { + "description": "Three cars parked by the roadside", + "timestamps": { + "start": "00:00:38", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "Seven cars have appeared within your field of vision during the scene.", + "D": "Six.", + "E": "You have seen eight cars enter your field of view during the scene." + }, + "answer_label": "D", + "answer_text": "Six.", + "q_id": "01471_5", + "correct_answer": "Six.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "Seven cars have appeared within your field of vision during the scene.", + "D": "Six.", + "E": "You have seen eight cars enter your field of view during the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_6", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cars have appeared within my field of vision?", + "answer": "seven.", + "question_chinese": "现在我的实现范围内出现了几辆车？", + "answer_chinese": "7辆", + "evidence": [ + { + "description": "Three cars parked by the roadside", + "timestamps": { + "start": "00:00:23", + "end": "00:00:26" + } + }, + { + "description": "Three cars parked by the roadside", + "timestamps": { + "start": "00:00:38", + "end": "00:00:41" + } + }, + { + "description": "Another car parked by the roadside", + "timestamps": { + "start": "00:00:49", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:01:49", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "seven.", + "C": "Ten.", + "D": "You have only seen six cars so far during your walk today.", + "E": "You have already seen eight cars pass through your field of view so far." + }, + "answer_label": "B", + "answer_text": "seven.", + "q_id": "01471_6", + "correct_answer": "seven.", + "original_options": { + "A": "Five.", + "B": "seven.", + "C": "Ten.", + "D": "You have only seen six cars so far during your walk today.", + "E": "You have already seen eight cars pass through your field of view so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_5_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to go back to the entrance of my apartment building unit now, can I get back within 200 steps?", + "answer": "No—you need to climb over a wall. The wall is in front of you.", + "question_chinese": "现在我要回到我家的那个单元口能在200步内回去吗？", + "answer_chinese": "不能，你需要翻墙。墙在你前方。", + "evidence": [ + { + "description": "Currently walking more than 100 meters and passing through the middle unit building", + "timestamps": { + "start": "00:01:20", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you need to climb over a wall. The wall is in front of you.", + "B": "Yes—you can get back within 200 steps; there’s no wall blocking you, and the entrance is behind you." + }, + "answer_label": "A", + "answer_text": "No—you need to climb over a wall. The wall is in front of you.", + "q_id": "01471_5_T2", + "correct_answer": "No—you need to climb over a wall. The wall is in front of you.", + "original_options": { + "A": "No—you need to climb over a wall. The wall is in front of you.", + "B": "Yes—you can get back within 200 steps; there’s no wall blocking you, and the entrance is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_5_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to go back to the entrance of my apartment building unit now, can I get back within 200 steps?", + "answer": "No—you need to climb over a wall. The wall is behind you.", + "question_chinese": "现在我要回到我家的那个单元口能在200步内回去吗？", + "answer_chinese": "不能，你需要翻墙。墙在你身后。", + "evidence": [ + { + "description": "Currently walking more than 100 meters and passing through the middle unit building", + "timestamps": { + "start": "00:01:20", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:01:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can get back within 200 steps; there’s no wall blocking you, and the entrance is ahead of you.", + "B": "No—you need to climb over a wall. The wall is behind you." + }, + "answer_label": "B", + "answer_text": "No—you need to climb over a wall. The wall is behind you.", + "q_id": "01471_5_T1", + "correct_answer": "No—you need to climb over a wall. The wall is behind you.", + "original_options": { + "A": "Yes—you can get back within 200 steps; there’s no wall blocking you, and the entrance is ahead of you.", + "B": "No—you need to climb over a wall. The wall is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 21 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_7", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How should I move if I want to go home now?", + "answer": "Turn left and walk straight for about 150 meters.", + "question_chinese": "现在我要回家的话怎么移动？", + "answer_chinese": "左转一直往前走大概150米", + "evidence": [ + { + "description": "Walk straight along this road", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + }, + { + "description": "Turned left", + "timestamps": { + "start": "00:00:59", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:01:04", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right and walk straight.", + "B": "Go straight for about 150 meters.", + "C": "Turn left, then turn right shortly after, and continue straight until you reach home.", + "D": "Turn left and walk straight for about 150 meters.", + "E": "Turn right first, keep walking straight, and then turn left near the end to get home." + }, + "answer_label": "D", + "answer_text": "Turn left and walk straight for about 150 meters.", + "q_id": "01471_7", + "correct_answer": "Turn left and walk straight for about 150 meters.", + "original_options": { + "A": "Turn right and walk straight.", + "B": "Go straight for about 150 meters.", + "C": "Turn left, then turn right shortly after, and continue straight until you reach home.", + "D": "Turn left and walk straight for about 150 meters.", + "E": "Turn right first, keep walking straight, and then turn left near the end to get home." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 16 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_8", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How should I move if I want to go home now?", + "answer": "Turn around, climb through the window, walk a few steps, turn right, walk a few more steps, turn left to the road, then turn right and walk straight for about 150 meters.", + "question_chinese": "现在我要回家的话怎么移动？", + "answer_chinese": "掉头爬过窗户走几步右转再走几步左转到马路，再右转直走150m左右", + "evidence": [ + { + "description": "Walking along this road", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + }, + { + "description": "Turned left to the unit entrance", + "timestamps": { + "start": "00:01:17", + "end": "00:01:19" + } + }, + { + "description": "Turned left a few steps to the window", + "timestamps": { + "start": "00:01:23", + "end": "00:01:28" + } + }, + { + "description": "Turned left a few steps to the window", + "timestamps": { + "start": "00:01:38", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:01:49", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn around and climb through the window.", + "B": "Turn around, then head to the road.", + "C": "Turn around, climb through the window, then turn left, reach the road, turn right, and walk straight 150 meters.", + "D": "Turn around, climb through the window, turn right, reach the road, then turn left and continue straight about 150 meters.", + "E": "Turn around, climb through the window, walk a few steps, turn right, walk a few more steps, turn left to the road, then turn right and walk straight for about 150 meters." + }, + "answer_label": "E", + "answer_text": "Turn around, climb through the window, walk a few steps, turn right, walk a few more steps, turn left to the road, then turn right and walk straight for about 150 meters.", + "q_id": "01471_8", + "correct_answer": "Turn around, climb through the window, walk a few steps, turn right, walk a few more steps, turn left to the road, then turn right and walk straight for about 150 meters.", + "original_options": { + "A": "Turn around and climb through the window.", + "B": "Turn around, then head to the road.", + "C": "Turn around, climb through the window, then turn left, reach the road, turn right, and walk straight 150 meters.", + "D": "Turn around, climb through the window, turn right, reach the road, then turn left and continue straight about 150 meters.", + "E": "Turn around, climb through the window, walk a few steps, turn right, walk a few more steps, turn left to the road, then turn right and walk straight for about 150 meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 20 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_8_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I walked into a dead end now?", + "answer": "Once, on your left.", + "question_chinese": "现在我走到死角几次？", + "answer_chinese": "1次，在你的左方。", + "evidence": [ + { + "description": "Walked into a dead end", + "timestamps": { + "start": "00:01:01", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero times.", + "B": "Twice—left and behind you.", + "C": "Once, on your left.", + "D": "Three times—once on your left and twice behind you in different spots.", + "E": "Once, but it was on your right side rather than on your left." + }, + "answer_label": "C", + "answer_text": "Once, on your left.", + "q_id": "01471_8_T1", + "correct_answer": "Once, on your left.", + "original_options": { + "A": "Zero times.", + "B": "Twice—left and behind you.", + "C": "Once, on your left.", + "D": "Three times—once on your left and twice behind you in different spots.", + "E": "Once, but it was on your right side rather than on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 5, + 13, + 13 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_8_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I walked into a dead end now?", + "answer": "Once, behind you.", + "question_chinese": "现在我走到死角几次？", + "answer_chinese": "1次，在你后方。", + "evidence": [ + { + "description": "Walked into a dead end", + "timestamps": { + "start": "00:01:01", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero times.", + "B": "Once, in front of you.", + "C": "Twice—once behind you and once on your left during the route.", + "D": "Once, behind you.", + "E": "Three times—twice behind you and once on your left after turning around." + }, + "answer_label": "D", + "answer_text": "Once, behind you.", + "q_id": "01471_8_T2", + "correct_answer": "Once, behind you.", + "original_options": { + "A": "Zero times.", + "B": "Once, in front of you.", + "C": "Twice—once behind you and once on your left during the route.", + "D": "Once, behind you.", + "E": "Three times—twice behind you and once on your left after turning around." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 5, + 12, + 13 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans have I encountered so far?", + "answer": "two.", + "question_chinese": "现在我遇到了几个垃圾桶？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "There are two trash cans", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "two.", + "B": "One.", + "C": "Three.", + "D": "You have encountered four trash cans so far along the way today.", + "E": "You have already passed five trash cans during your walk up to now." + }, + "answer_label": "A", + "answer_text": "two.", + "q_id": "01471_1", + "correct_answer": "two.", + "original_options": { + "A": "two.", + "B": "One.", + "C": "Three.", + "D": "You have encountered four trash cans so far along the way today.", + "E": "You have already passed five trash cans during your walk up to now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01471", + "video_length": "00:01:51", + "qid": "01471_2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans have I encountered so far?", + "answer": "six.", + "question_chinese": "现在我遇到了几个垃圾桶？", + "answer_chinese": "6个", + "evidence": [ + { + "description": "There are two trash cans", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "There are two trash cans", + "timestamps": { + "start": "00:00:36", + "end": "00:00:38" + } + }, + { + "description": "There are two trash cans", + "timestamps": { + "start": "00:00:47", + "end": "00:00:49" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "six.", + "C": "Seven.", + "D": "Eight trash cans have been encountered so far along your route in the mall area.", + "E": "Nine trash cans have been encountered up to this point during your walk through the video." + }, + "answer_label": "B", + "answer_text": "six.", + "q_id": "01471_2", + "correct_answer": "six.", + "original_options": { + "A": "Five.", + "B": "six.", + "C": "Seven.", + "D": "Eight trash cans have been encountered so far along your route in the mall area.", + "E": "Nine trash cans have been encountered up to this point during your walk through the video." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + } +] diff --git a/QAs/01550-mcq.json b/QAs/01550-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..80af024965152838bf80e9bf0f144badd948a9e8 --- /dev/null +++ b/QAs/01550-mcq.json @@ -0,0 +1,1002 @@ +[ + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_11", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have passed by me since leaving the tactile paving?", + "answer": "six.", + "question_chinese": "从出盲道到现在有几个人从我身边经过？", + "answer_chinese": "6个", + "evidence": [ + { + "description": "6 people", + "timestamps": { + "start": "00:02:45", + "end": "00:02:48" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "Eight people have passed by you since leaving the tactile paving area.", + "D": "six.", + "E": "Ten people have gone past you after you left the tactile paving path." + }, + "answer_label": "D", + "answer_text": "six.", + "q_id": "01550_11", + "correct_answer": "six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "Eight people have passed by you since leaving the tactile paving area.", + "D": "six.", + "E": "Ten people have gone past you after you left the tactile paving path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_12", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have passed by me since leaving the tactile paving?", + "answer": "eleven.", + "question_chinese": "从出盲道到现在有几个人从我身边经过？", + "answer_chinese": "11个", + "evidence": [ + { + "description": "6 people", + "timestamps": { + "start": "00:02:45", + "end": "00:02:48" + } + }, + { + "description": "2 people", + "timestamps": { + "start": "00:02:53", + "end": "00:02:56" + } + }, + { + "description": "3 people", + "timestamps": { + "start": "00:01:12", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:02:56", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "eleven.", + "B": "Nine.", + "C": "Ten.", + "D": "Seven people have passed by you since leaving the tactile paving area.", + "E": "Thirteen people have gone past you after you left the tactile paving path." + }, + "answer_label": "A", + "answer_text": "eleven.", + "q_id": "01550_12", + "correct_answer": "eleven.", + "original_options": { + "A": "eleven.", + "B": "Nine.", + "C": "Ten.", + "D": "Seven people have passed by you since leaving the tactile paving area.", + "E": "Thirteen people have gone past you after you left the tactile paving path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_7", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned around since entering the mall?", + "answer": "two times.", + "question_chinese": "从进商场到现在我转身几次？", + "answer_chinese": "2次", + "evidence": [ + { + "description": "Turned left once after entering the mall", + "timestamps": { + "start": "00:02:24", + "end": "00:02:27" + } + }, + { + "description": "Turned left once after entering the store", + "timestamps": { + "start": "00:02:29", + "end": "00:02:31" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One time.", + "B": "Three times.", + "C": "Four times in total since you entered the mall and continued walking inside the corridors.", + "D": "two times.", + "E": "Five times overall since entering the mall and walking through the interior shopping area." + }, + "answer_label": "D", + "answer_text": "two times.", + "q_id": "01550_7", + "correct_answer": "two times.", + "original_options": { + "A": "One time.", + "B": "Three times.", + "C": "Four times in total since you entered the mall and continued walking inside the corridors.", + "D": "two times.", + "E": "Five times overall since entering the mall and walking through the interior shopping area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_8", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned around since entering the mall?", + "answer": "four times.", + "question_chinese": "从进商场到现在我转身几次？", + "answer_chinese": "4次", + "evidence": [ + { + "description": "Turned left once after entering the mall", + "timestamps": { + "start": "00:02:24", + "end": "00:02:27" + } + }, + { + "description": "Turned left once after entering the store", + "timestamps": { + "start": "00:02:29", + "end": "00:02:31" + } + }, + { + "description": "Turned around once in the store", + "timestamps": { + "start": "00:02:33", + "end": "00:02:37" + } + }, + { + "description": "Turned left once again in the store", + "timestamps": { + "start": "00:02:41", + "end": "00:02:43" + } + } + ], + "question_timestamps": "00:02:47", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three times.", + "B": "Five times.", + "C": "four times.", + "D": "Two times in total since you entered the mall and kept walking past the storefronts.", + "E": "Six times overall since entering the mall and walking through the indoor shopping corridors." + }, + "answer_label": "C", + "answer_text": "four times.", + "q_id": "01550_8", + "correct_answer": "four times.", + "original_options": { + "A": "Three times.", + "B": "Five times.", + "C": "four times.", + "D": "Two times in total since you entered the mall and kept walking past the storefronts.", + "E": "Six times overall since entering the mall and walking through the indoor shopping corridors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_3", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is Tasting in for me now?", + "answer": "Right side.", + "question_chinese": "现在塔斯汀在我的什么方向？", + "answer_chinese": "右侧", + "evidence": [ + { + "description": "Tasting is in front right of me", + "timestamps": { + "start": "00:00:53", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left side.", + "B": "In front of you.", + "C": "Behind you on your left side, not to your right as you are standing now.", + "D": "Straight ahead and slightly left, rather than on the right side where it actually is.", + "E": "Right side." + }, + "answer_label": "E", + "answer_text": "Right side.", + "q_id": "01550_3", + "correct_answer": "Right side.", + "original_options": { + "A": "Left side.", + "B": "In front of you.", + "C": "Behind you on your left side, not to your right as you are standing now.", + "D": "Straight ahead and slightly left, rather than on the right side where it actually is.", + "E": "Right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 15, + 15 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is Tasting in for me now?", + "answer": "Right back side.", + "question_chinese": "现在塔斯汀在我的什么方向？", + "answer_chinese": "右后侧", + "evidence": [ + { + "description": "Tasting is in front right of me", + "timestamps": { + "start": "00:00:53", + "end": "00:00:55" + } + }, + { + "description": "I walked around the shops and turned right once", + "timestamps": { + "start": "00:01:00", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Left front side.", + "C": "Directly behind you on the left side, not on the right back side.", + "D": "Right back side.", + "E": "On your right side and slightly in front, rather than behind you to the right." + }, + "answer_label": "D", + "answer_text": "Right back side.", + "q_id": "01550_4", + "correct_answer": "Right back side.", + "original_options": { + "A": "In front of you.", + "B": "Left front side.", + "C": "Directly behind you on the left side, not on the right back side.", + "D": "Right back side.", + "E": "On your right side and slightly in front, rather than behind you to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 13, + 15 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cyclists have I encountered on the tactile paving so far?", + "answer": "one.", + "question_chinese": "现在我在盲道上碰到了几个骑车的人？", + "answer_chinese": "1个", + "evidence": [ + { + "description": "A delivery rider on a bicycle came from the opposite direction", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "one.", + "C": "Three.", + "D": "Four cyclists have been encountered on the tactile paving so far during your walk outside.", + "E": "Five cyclists have been encountered on the tactile paving up to this point in the video." + }, + "answer_label": "B", + "answer_text": "one.", + "q_id": "01550_1", + "correct_answer": "one.", + "original_options": { + "A": "Two.", + "B": "one.", + "C": "Three.", + "D": "Four cyclists have been encountered on the tactile paving so far during your walk outside.", + "E": "Five cyclists have been encountered on the tactile paving up to this point in the video." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cyclists have I encountered on the tactile paving so far?", + "answer": "Two.", + "question_chinese": "现在我在盲道上碰到了几个骑车的人？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "A delivery rider on a bicycle came from the opposite direction", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + }, + { + "description": "A delivery rider on a bicycle came from the opposite direction", + "timestamps": { + "start": "00:01:03", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four cyclists have appeared on the tactile paving up to this point.", + "D": "Five cyclists in total have been encountered while you were on the tactile paving.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "01550_2", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four cyclists have appeared on the tactile paving up to this point.", + "D": "Five cyclists in total have been encountered while you were on the tactile paving.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_13", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many stores can I choose from if I want to repair my phone now?", + "answer": "one.", + "question_chinese": "现在我想去修手机的话我有几家店可以选择？", + "answer_chinese": "1个", + "evidence": [ + { + "description": "There is a mobile phone specialty store on the right that also provides repair services", + "timestamps": { + "start": "00:01:04", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "one.", + "D": "You can choose from two different stores to repair your phone right now.", + "E": "There are three stores available at the moment where you could get your phone repaired." + }, + "answer_label": "C", + "answer_text": "one.", + "q_id": "01550_13", + "correct_answer": "one.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "one.", + "D": "You can choose from two different stores to repair your phone right now.", + "E": "There are three stores available at the moment where you could get your phone repaired." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_14", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many stores can I choose from if I want to repair my phone now?", + "answer": "two.", + "question_chinese": "现在我想去修手机的话我有几家店可以选择？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "There is a mobile phone specialty store on the right that also provides repair services", + "timestamps": { + "start": "00:01:04", + "end": "00:01:07" + } + }, + { + "description": "There is a mobile phone repair shop in front right", + "timestamps": { + "start": "00:01:14", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Right now, you have three stores to choose from for phone repair.", + "D": "At this moment, only one store is available if you want to repair your phone.", + "E": "two." + }, + "answer_label": "E", + "answer_text": "two.", + "q_id": "01550_14", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Right now, you have three stores to choose from for phone repair.", + "D": "At this moment, only one store is available if you want to repair your phone.", + "E": "two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 15 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_5_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to grab the cake now, can I reach it directly with my hand?", + "answer": "No—you can’t reach it with your hand. You’re outside the cake shop.", + "question_chinese": "现在我想去拿蛋糕，可以直接用手够到吗？", + "answer_chinese": "你不能用手够到，你在蛋糕店外。", + "evidence": [ + { + "description": "you have not arrived at the cake shop yet.", + "timestamps": { + "start": "00:02:18", + "end": "00:02:26" + } + } + ], + "question_timestamps": "00:02:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t reach it with your hand. You’re outside the cake shop.", + "B": "Yes—you can reach it with your hand. You’re right at the cake counter." + }, + "answer_label": "A", + "answer_text": "No—you can’t reach it with your hand. You’re outside the cake shop.", + "q_id": "01550_5_T2", + "correct_answer": "No—you can’t reach it with your hand. You’re outside the cake shop.", + "original_options": { + "A": "No—you can’t reach it with your hand. You’re outside the cake shop.", + "B": "Yes—you can reach it with your hand. You’re right at the cake counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_9", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the cake directly with my hand if I want to get it now?", + "answer": "Yes.", + "question_chinese": "现在我想去拿蛋糕，可以直接用手够到吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The cake refrigerator is in front left of me", + "timestamps": { + "start": "00:02:31", + "end": "00:02:33" + } + }, + { + "description": "Turned around in front of the cake refrigerator", + "timestamps": { + "start": "00:02:33", + "end": "00:02:34" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "01550_9", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_5_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to grab the cake now, can I reach it directly with my hand?", + "answer": "Yes—you can. The cake is in your hand.", + "question_chinese": "现在我想去拿蛋糕，可以直接用手够到吗？", + "answer_chinese": "能，蛋糕在你手里", + "evidence": [ + { + "description": "The cake refrigerator is in front left of me", + "timestamps": { + "start": "00:02:31", + "end": "00:02:56" + } + } + ], + "question_timestamps": "00:02:58", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t. The cake is out of reach.", + "B": "Yes—you can. The cake is in your hand." + }, + "answer_label": "B", + "answer_text": "Yes—you can. The cake is in your hand.", + "q_id": "01550_5_T1", + "correct_answer": "Yes—you can. The cake is in your hand.", + "original_options": { + "A": "No—you can’t. The cake is out of reach.", + "B": "Yes—you can. The cake is in your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_15", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times has my white cane touched something so far?", + "answer": "three times.", + "question_chinese": "现在我的导盲杖已经碰到了几次东西？", + "answer_chinese": "3次", + "evidence": [ + { + "description": "Touched a delivery rider", + "timestamps": { + "start": "00:00:17", + "end": "00:00:19" + } + }, + { + "description": "Touched an electric bicycle", + "timestamps": { + "start": "00:00:20", + "end": "00:00:22" + } + }, + { + "description": "Touched a delivery rider", + "timestamps": { + "start": "00:00:29", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "three times.", + "C": "Four times.", + "D": "Your white cane has made contact with something two times so far.", + "E": "Up to now, your white cane has touched something four times in total." + }, + "answer_label": "B", + "answer_text": "three times.", + "q_id": "01550_15", + "correct_answer": "three times.", + "original_options": { + "A": "Two times.", + "B": "three times.", + "C": "Four times.", + "D": "Your white cane has made contact with something two times so far.", + "E": "Up to now, your white cane has touched something four times in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_16", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times has my white cane touched something so far?", + "answer": "seven times.", + "question_chinese": "现在我的导盲杖已经碰到了几次东西？", + "answer_chinese": "7次", + "evidence": [ + { + "description": "Touched a delivery rider", + "timestamps": { + "start": "00:00:17", + "end": "00:00:19" + } + }, + { + "description": "Touched an electric bicycle", + "timestamps": { + "start": "00:00:20", + "end": "00:00:22" + } + }, + { + "description": "Touched a delivery rider", + "timestamps": { + "start": "00:00:29", + "end": "00:00:31" + } + }, + { + "description": "Touched a pedestrian", + "timestamps": { + "start": "00:00:55", + "end": "00:00:58" + } + }, + { + "description": "Knocked over a horn", + "timestamps": { + "start": "00:02:00", + "end": "00:02:03" + } + }, + { + "description": "Touched a person", + "timestamps": { + "start": "00:02:23", + "end": "00:02:25" + } + }, + { + "description": "Touched a staff member", + "timestamps": { + "start": "00:02:41", + "end": "00:02:43" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five times.", + "B": "Eight times.", + "C": "Six times in total; you have made six separate cane contacts so far.", + "D": "Nine times overall; the cane has touched objects nine distinct times up to now.", + "E": "seven times." + }, + "answer_label": "E", + "answer_text": "seven times.", + "q_id": "01550_16", + "correct_answer": "seven times.", + "original_options": { + "A": "Five times.", + "B": "Eight times.", + "C": "Six times in total; you have made six separate cane contacts so far.", + "D": "Nine times overall; the cane has touched objects nine distinct times up to now.", + "E": "seven times." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_3_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How do I get to KFC from where I am now?", + "answer": "Turn left and walk straight a few steps to KFC; KFC is to your left.", + "question_chinese": "现在我要去肯德基怎么走？", + "answer_chinese": "左转直走几步到肯德基，肯德基在你的左边。", + "evidence": [ + { + "description": "KFC is in front right of me", + "timestamps": { + "start": "00:01:56", + "end": "00:01:59" + } + } + ], + "question_timestamps": "00:02:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right and walk a few steps.", + "B": "Go straight ahead to KFC.", + "C": "Turn left, then keep walking until KFC is directly in front of you, not off to the side.", + "D": "Turn left and walk straight a few steps to KFC; KFC is to your left.", + "E": "Turn left and walk straight for a bit; you should end up with KFC on your right side." + }, + "answer_label": "D", + "answer_text": "Turn left and walk straight a few steps to KFC; KFC is to your left.", + "q_id": "01550_3_T1", + "correct_answer": "Turn left and walk straight a few steps to KFC; KFC is to your left.", + "original_options": { + "A": "Turn right and walk a few steps.", + "B": "Go straight ahead to KFC.", + "C": "Turn left, then keep walking until KFC is directly in front of you, not off to the side.", + "D": "Turn left and walk straight a few steps to KFC; KFC is to your left.", + "E": "Turn left and walk straight for a bit; you should end up with KFC on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 18 + ] + }, + { + "video_uid": "01550", + "video_length": "00:03:10", + "qid": "01550_3_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How do I get to KFC from where I am now?", + "answer": "Walk straight a few steps to KFC; KFC is in front of you.", + "question_chinese": "现在我要去肯德基怎么走？", + "answer_chinese": "你直走几步到肯德基，肯德基在你面前。", + "evidence": [ + { + "description": "KFC is in front of me", + "timestamps": { + "start": "00:02:10", + "end": "00:02:14" + } + } + ], + "question_timestamps": "00:02:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk straight a few steps to KFC; KFC is in front of you.", + "B": "Turn left to reach KFC.", + "C": "Turn right to reach KFC.", + "D": "Walk straight a few steps, but KFC will be behind you, so you’ll need to turn around.", + "E": "Walk straight a few steps toward KFC; it should end up on your left rather than directly ahead." + }, + "answer_label": "A", + "answer_text": "Walk straight a few steps to KFC; KFC is in front of you.", + "q_id": "01550_3_T2", + "correct_answer": "Walk straight a few steps to KFC; KFC is in front of you.", + "original_options": { + "A": "Walk straight a few steps to KFC; KFC is in front of you.", + "B": "Turn left to reach KFC.", + "C": "Turn right to reach KFC.", + "D": "Walk straight a few steps, but KFC will be behind you, so you’ll need to turn around.", + "E": "Walk straight a few steps toward KFC; it should end up on your left rather than directly ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + } +] diff --git a/QAs/01551-mcq.json b/QAs/01551-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6d37cd2d5a2dc8defc49c05ee4d97638b2d55435 --- /dev/null +++ b/QAs/01551-mcq.json @@ -0,0 +1,810 @@ +[ + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_13", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stone piers have appeared in my sight since the beginning?", + "answer": "eleven.", + "question_chinese": "从开始到现在我的视线中已经出现了几个石墩吗？", + "answer_chinese": "11个", + "evidence": [ + { + "description": "There are 11 stone piers", + "timestamps": { + "start": "00:00:00", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "eleven.", + "B": "Ten.", + "C": "Twelve.", + "D": "You have seen nine stone piers in total since the beginning of the video so far.", + "E": "Since the beginning, you have counted thirteen stone piers appearing in your sight altogether." + }, + "answer_label": "A", + "answer_text": "eleven.", + "q_id": "01551_13", + "correct_answer": "eleven.", + "original_options": { + "A": "eleven.", + "B": "Ten.", + "C": "Twelve.", + "D": "You have seen nine stone piers in total since the beginning of the video so far.", + "E": "Since the beginning, you have counted thirteen stone piers appearing in your sight altogether." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 14 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_14", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stone piers have appeared in my sight since the beginning?", + "answer": "fifteen.", + "question_chinese": "从开始到现在我的视线中已经出现了几个石墩吗？", + "answer_chinese": "15个", + "evidence": [ + { + "description": "There are 11 stone piers", + "timestamps": { + "start": "00:00:00", + "end": "00:00:12" + } + }, + { + "description": "There are 4 more stone piers", + "timestamps": { + "start": "00:02:32", + "end": "00:02:36" + } + } + ], + "question_timestamps": "00:03:28", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Fourteen.", + "B": "Sixteen.", + "C": "fifteen.", + "D": "From the beginning until now, you have seen thirteen stone piers appear in your sight.", + "E": "Overall, there have been seventeen stone piers that appeared in your sight since the start." + }, + "answer_label": "C", + "answer_text": "fifteen.", + "q_id": "01551_14", + "correct_answer": "fifteen.", + "original_options": { + "A": "Fourteen.", + "B": "Sixteen.", + "C": "fifteen.", + "D": "From the beginning until now, you have seen thirteen stone piers appear in your sight.", + "E": "Overall, there have been seventeen stone piers that appeared in your sight since the start." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many operating street shops have I seen so far?", + "answer": "seven.", + "question_chinese": "现在我已经看到了街边有几个正在营业的铺子？", + "answer_chinese": "7间", + "evidence": [ + { + "description": "There are 4 operating shops on the right", + "timestamps": { + "start": "00:00:16", + "end": "00:00:40" + } + }, + { + "description": "There are 3 operating shops here", + "timestamps": { + "start": "00:02:06", + "end": "00:02:08" + } + } + ], + "question_timestamps": "00:02:12", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "seven.", + "B": "Six.", + "C": "Eight.", + "D": "So far, you have seen five operating street shops during your walk up to this point.", + "E": "Up to now, you have encountered ten operating street shops that were open and running." + }, + "answer_label": "A", + "answer_text": "seven.", + "q_id": "01551_3", + "correct_answer": "seven.", + "original_options": { + "A": "seven.", + "B": "Six.", + "C": "Eight.", + "D": "So far, you have seen five operating street shops during your walk up to this point.", + "E": "Up to now, you have encountered ten operating street shops that were open and running." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many operating street shops have I seen so far?", + "answer": "nine.", + "question_chinese": "现在我已经看到了街边有几个正在营业的铺子？", + "answer_chinese": "9间", + "evidence": [ + { + "description": "There are 4 operating shops on the right", + "timestamps": { + "start": "00:00:16", + "end": "00:00:40" + } + }, + { + "description": "There are 5 operating shops here", + "timestamps": { + "start": "00:02:06", + "end": "00:02:23" + } + } + ], + "question_timestamps": "00:03:28", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Eight.", + "B": "Ten.", + "C": "By this point, you have seen seven operating street shops in total during the video.", + "D": "So far in the video, you have counted twelve operating street shops that were open for business.", + "E": "nine." + }, + "answer_label": "E", + "answer_text": "nine.", + "q_id": "01551_4", + "correct_answer": "nine.", + "original_options": { + "A": "Eight.", + "B": "Ten.", + "C": "By this point, you have seen seven operating street shops in total during the video.", + "D": "So far in the video, you have counted twelve operating street shops that were open for business.", + "E": "nine." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 17 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_7", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sections of tactile paving have I walked on so far?", + "answer": "one section.", + "question_chinese": "现在我已经走了几段盲道？", + "answer_chinese": "1道", + "evidence": [ + { + "description": "Have walked one section of tactile paving at this time", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:01:45", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two sections.", + "B": "one section.", + "C": "Three sections.", + "D": "You have walked on four separate sections of tactile paving so far during the route.", + "E": "Up to this point, you have stepped onto five different tactile paving sections while moving forward." + }, + "answer_label": "B", + "answer_text": "one section.", + "q_id": "01551_7", + "correct_answer": "one section.", + "original_options": { + "A": "Two sections.", + "B": "one section.", + "C": "Three sections.", + "D": "You have walked on four separate sections of tactile paving so far during the route.", + "E": "Up to this point, you have stepped onto five different tactile paving sections while moving forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_8", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sections of tactile paving have I walked on so far?", + "answer": "two sections.", + "question_chinese": "现在我已经走了几段盲道？", + "answer_chinese": "2道", + "evidence": [ + { + "description": "Have walked one section of tactile paving at this time", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + }, + { + "description": "Have walked another section of tactile paving at this time", + "timestamps": { + "start": "00:02:18", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:03:02", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One section.", + "B": "Three sections.", + "C": "So far, you have walked over four distinct sections of tactile paving during your movement.", + "D": "two sections.", + "E": "Up to now, you have stepped on five separate tactile paving sections along the path." + }, + "answer_label": "D", + "answer_text": "two sections.", + "q_id": "01551_8", + "correct_answer": "two sections.", + "original_options": { + "A": "One section.", + "B": "Three sections.", + "C": "So far, you have walked over four distinct sections of tactile paving during your movement.", + "D": "two sections.", + "E": "Up to now, you have stepped on five separate tactile paving sections along the path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_15", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have I talked to so far?", + "answer": "Three.", + "question_chinese": "现在我已经跟多少个人交谈过了？", + "answer_chinese": "3个", + "evidence": [ + { + "description": "Asked a passerby for directions", + "timestamps": { + "start": "00:01:00", + "end": "00:01:06" + } + }, + { + "description": "Asked a milk tea shop staff for directions", + "timestamps": { + "start": "00:02:29", + "end": "00:02:32" + } + }, + { + "description": "Talked with a chestnut shop staff", + "timestamps": { + "start": "00:02:41", + "end": "00:02:43" + } + } + ], + "question_timestamps": "00:02:49", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "Only one person so far, not three, even though I have interacted with people around me.", + "E": "Five people in total up to now, counting everyone I spoke with before reaching this spot." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "01551_15", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "Only one person so far, not three, even though I have interacted with people around me.", + "E": "Five people in total up to now, counting everyone I spoke with before reaching this spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_16", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have I talked to so far?", + "answer": "four.", + "question_chinese": "现在我已经跟多少个人交谈过了？", + "answer_chinese": "4个", + "evidence": [ + { + "description": "Asked a passerby for directions", + "timestamps": { + "start": "00:01:00", + "end": "00:01:06" + } + }, + { + "description": "Asked a milk tea shop staff for directions", + "timestamps": { + "start": "00:02:29", + "end": "00:02:32" + } + }, + { + "description": "Talked with a chestnut shop staff", + "timestamps": { + "start": "00:02:41", + "end": "00:02:43" + } + }, + { + "description": "Talked with another chestnut shop staff", + "timestamps": { + "start": "00:03:04", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:03:18", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "Only two people so far, despite passing by others without actually speaking to them directly.", + "D": "four.", + "E": "Ten people in total up to now, including brief conversations and the earlier interactions before this moment." + }, + "answer_label": "D", + "answer_text": "four.", + "q_id": "01551_16", + "correct_answer": "four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "Only two people so far, despite passing by others without actually speaking to them directly.", + "D": "four.", + "E": "Ten people in total up to now, including brief conversations and the earlier interactions before this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 17 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to buy Mixue Ice City now?", + "answer": "Turn around and walk forward a few steps, Mixue Ice City is on the left.", + "question_chinese": "现在我想去买杯蜜雪冰城应该怎么移动？", + "answer_chinese": "转身掉头往前走几步，左侧是蜜雪冰城", + "evidence": [ + { + "description": "There is a Mixue Ice City on the right", + "timestamps": { + "start": "00:00:31", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left; it's on your right.", + "B": "Go straight ahead; it's on the left.", + "C": "Turn around and walk forward a few steps, Mixue Ice City is on the left.", + "D": "Turn around and walk forward a few steps, but Mixue Ice City will be on your right.", + "E": "Turn around and walk backward a few steps, and Mixue Ice City will appear on your left." + }, + "answer_label": "C", + "answer_text": "Turn around and walk forward a few steps, Mixue Ice City is on the left.", + "q_id": "01551_1", + "correct_answer": "Turn around and walk forward a few steps, Mixue Ice City is on the left.", + "original_options": { + "A": "Turn left; it's on your right.", + "B": "Go straight ahead; it's on the left.", + "C": "Turn around and walk forward a few steps, Mixue Ice City is on the left.", + "D": "Turn around and walk forward a few steps, but Mixue Ice City will be on your right.", + "E": "Turn around and walk backward a few steps, and Mixue Ice City will appear on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 17 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to buy Mixue Ice City now?", + "answer": "Turn left and walk along the street shops, Mixue Ice City is on the left.", + "question_chinese": "现在我想去买杯蜜雪冰城应该怎么移动？", + "answer_chinese": "左转沿着街边铺子走，左侧是蜜雪冰城", + "evidence": [ + { + "description": "There is a Mixue Ice City on the right", + "timestamps": { + "start": "00:00:31", + "end": "00:00:34" + } + }, + { + "description": "I turned left once", + "timestamps": { + "start": "00:01:53", + "end": "00:01:58" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right; it's on the right.", + "B": "Turn left and walk along the street shops, Mixue Ice City is on the left.", + "C": "Go straight; it's on the left.", + "D": "Turn left and walk along the street shops, but Mixue Ice City will be on your right.", + "E": "Turn around and walk toward the intersection, and Mixue Ice City will be straight ahead." + }, + "answer_label": "B", + "answer_text": "Turn left and walk along the street shops, Mixue Ice City is on the left.", + "q_id": "01551_2", + "correct_answer": "Turn left and walk along the street shops, Mixue Ice City is on the left.", + "original_options": { + "A": "Turn right; it's on the right.", + "B": "Turn left and walk along the street shops, Mixue Ice City is on the left.", + "C": "Go straight; it's on the left.", + "D": "Turn left and walk along the street shops, but Mixue Ice City will be on your right.", + "E": "Turn around and walk toward the intersection, and Mixue Ice City will be straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 15 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_11", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the chestnut shop within 10 steps now?", + "answer": "No.", + "question_chinese": "现在我能在10步以内到达板栗店吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "There are no shops within 10 steps around me", + "timestamps": { + "start": "00:01:44", + "end": "00:01:49" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01551_11", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_12", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the chestnut shop within 10 steps now?", + "answer": "Yes.", + "question_chinese": "现在我能在10步以内到达板栗店吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The chestnut shop is some distance ahead on my left, there is a milk tea shop staff not far in front of the chestnut shop", + "timestamps": { + "start": "00:02:24", + "end": "00:02:26" + } + }, + { + "description": "I am now talking with the milk tea shop staff", + "timestamps": { + "start": "00:02:28", + "end": "00:02:31" + } + } + ], + "question_timestamps": "00:02:34", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01551_12", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_9", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is McDonald's in for me now?", + "answer": "Left side.", + "question_chinese": "现在麦当劳在我的什么方向？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "There is a McDonald's on the left", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "Straight ahead.", + "C": "It's behind you on the right side, so you would need to turn around before walking toward it.", + "D": "It's in front of you, slightly to the right, not on the left side from where you are standing.", + "E": "Left side." + }, + "answer_label": "E", + "answer_text": "Left side.", + "q_id": "01551_9", + "correct_answer": "Left side.", + "original_options": { + "A": "Right side.", + "B": "Straight ahead.", + "C": "It's behind you on the right side, so you would need to turn around before walking toward it.", + "D": "It's in front of you, slightly to the right, not on the left side from where you are standing.", + "E": "Left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 19 + ] + }, + { + "video_uid": "01551", + "video_length": "00:03:28", + "qid": "01551_10", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is McDonald's in for me now?", + "answer": "Left back side.", + "question_chinese": "现在麦当劳在我的什么方向？", + "answer_chinese": "左后侧", + "evidence": [ + { + "description": "There is a McDonald's on the left", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + }, + { + "description": "I walked forward along the tactile paving for a distance", + "timestamps": { + "start": "00:00:52", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:01:14", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Left back side.", + "B": "Right back side.", + "C": "Left front side.", + "D": "Directly behind you on the right side, so you would need to turn around to see it clearly.", + "E": "In front of you and slightly to the left, meaning it is not behind you on the left." + }, + "answer_label": "A", + "answer_text": "Left back side.", + "q_id": "01551_10", + "correct_answer": "Left back side.", + "original_options": { + "A": "Left back side.", + "B": "Right back side.", + "C": "Left front side.", + "D": "Directly behind you on the right side, so you would need to turn around to see it clearly.", + "E": "In front of you and slightly to the left, meaning it is not behind you on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 18 + ] + } +] diff --git a/QAs/01552-mcq.json b/QAs/01552-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..888846ee67bdd211f2b2b93b2f2b6f017bbedc03 --- /dev/null +++ b/QAs/01552-mcq.json @@ -0,0 +1,909 @@ +[ + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_8_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which side of me is the entrance to Miniso right now?", + "answer": "The entrance of MINISO is to your front-left.", + "question_chinese": "现在名创优品的门口在我的哪侧？", + "answer_chinese": "名创优品的门口在你左前方。", + "evidence": [ + { + "description": "I will enter Miniso", + "timestamps": { + "start": "00:01:53", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:01:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "The entrance of MINISO is to your front-left.", + "D": "The entrance of MINISO is directly in front of you, not angled to your left.", + "E": "The entrance of MINISO is behind you on the left side as you stand there." + }, + "answer_label": "C", + "answer_text": "The entrance of MINISO is to your front-left.", + "q_id": "01552_8_T2", + "correct_answer": "The entrance of MINISO is to your front-left.", + "original_options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "The entrance of MINISO is to your front-left.", + "D": "The entrance of MINISO is directly in front of you, not angled to your left.", + "E": "The entrance of MINISO is behind you on the left side as you stand there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_8_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which side of me is the entrance to Miniso right now?", + "answer": "Behind you.", + "question_chinese": "现在名创优品的门口在我的哪侧？", + "answer_chinese": "后面", + "evidence": [ + { + "description": "I just entered Miniso", + "timestamps": { + "start": "00:01:58", + "end": "00:02:01" + } + } + ], + "question_timestamps": "00:02:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "Behind you.", + "D": "The entrance to MINISO is on your right side at the moment, not behind you.", + "E": "The entrance to MINISO is ahead of you, slightly to the left as you face forward." + }, + "answer_label": "C", + "answer_text": "Behind you.", + "q_id": "01552_8_T1", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "Behind you.", + "D": "The entrance to MINISO is on your right side at the moment, not behind you.", + "E": "The entrance to MINISO is ahead of you, slightly to the left as you face forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 16 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_3", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I passed by Jasmine Milk White shop so far?", + "answer": "one time.", + "question_chinese": "现在已经经过茉莉奶白店几次？", + "answer_chinese": "1次", + "evidence": [ + { + "description": "Passed by Jasmine Milk White shop once at this time", + "timestamps": { + "start": "00:00:19", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:00:58", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Three times.", + "C": "one time.", + "D": "You have passed by Jasmine Milk White shop two times so far during your walk.", + "E": "You have passed by Jasmine Milk White shop three times so far while moving through the mall." + }, + "answer_label": "C", + "answer_text": "one time.", + "q_id": "01552_3", + "correct_answer": "one time.", + "original_options": { + "A": "Two times.", + "B": "Three times.", + "C": "one time.", + "D": "You have passed by Jasmine Milk White shop two times so far during your walk.", + "E": "You have passed by Jasmine Milk White shop three times so far while moving through the mall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 17 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_4", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I passed by Jasmine Milk White shop so far?", + "answer": "two times.", + "question_chinese": "现在已经经过茉莉奶白店几次？", + "answer_chinese": "2次", + "evidence": [ + { + "description": "Passed by Jasmine Milk White shop once at this time", + "timestamps": { + "start": "00:00:19", + "end": "00:00:22" + } + }, + { + "description": "Entered Jasmine Milk White shop once at this time", + "timestamps": { + "start": "00:01:22", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:02:05", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "two times.", + "B": "One time.", + "C": "Three times.", + "D": "You have passed by Jasmine Milk White shop one time so far on your route.", + "E": "You have passed by Jasmine Milk White shop three times so far while walking around the mall." + }, + "answer_label": "A", + "answer_text": "two times.", + "q_id": "01552_4", + "correct_answer": "two times.", + "original_options": { + "A": "two times.", + "B": "One time.", + "C": "Three times.", + "D": "You have passed by Jasmine Milk White shop one time so far on your route.", + "E": "You have passed by Jasmine Milk White shop three times so far while walking around the mall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 17 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_9_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of items have I already bought in the mall?", + "answer": "Milk tea.", + "question_chinese": "现在我已经在商场买了哪几种东西？", + "answer_chinese": "奶茶", + "evidence": [ + { + "description": "I bought milk tea", + "timestamps": { + "start": "00:01:41", + "end": "00:01:43" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A hat.", + "B": "A bowl.", + "C": "You’ve already bought a hat in the mall, not any milk tea.", + "D": "Milk tea.", + "E": "You’ve already bought a bowl and a pan in the mall, not milk tea." + }, + "answer_label": "D", + "answer_text": "Milk tea.", + "q_id": "01552_9_T1", + "correct_answer": "Milk tea.", + "original_options": { + "A": "A hat.", + "B": "A bowl.", + "C": "You’ve already bought a hat in the mall, not any milk tea.", + "D": "Milk tea.", + "E": "You’ve already bought a bowl and a pan in the mall, not milk tea." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_9_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of items have I already bought in the mall?", + "answer": "You’ve already bought milk tea and a hat.", + "question_chinese": "现在我已经在商场买了哪几种东西？", + "answer_chinese": "你已经买了奶茶和帽子。", + "evidence": [ + { + "description": "I bought milk tea", + "timestamps": { + "start": "00:01:41", + "end": "00:01:43" + } + }, + { + "description": "I bought a hat.", + "timestamps": { + "start": "00:03:15", + "end": "00:03:20" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Milk tea only.", + "B": "You’ve already bought milk tea and a hat.", + "C": "A hat only.", + "D": "You’ve already bought milk tea and a bowl in the mall, not a hat.", + "E": "You’ve already bought a hat and a pan in the mall, rather than milk tea." + }, + "answer_label": "B", + "answer_text": "You’ve already bought milk tea and a hat.", + "q_id": "01552_9_T2", + "correct_answer": "You’ve already bought milk tea and a hat.", + "original_options": { + "A": "Milk tea only.", + "B": "You’ve already bought milk tea and a hat.", + "C": "A hat only.", + "D": "You’ve already bought milk tea and a bowl in the mall, not a hat.", + "E": "You’ve already bought a hat and a pan in the mall, rather than milk tea." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_7", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many children have I encountered in the mall so far?", + "answer": "four.", + "question_chinese": "现在我已经在商场遇到多少个小孩？", + "answer_chinese": "4个", + "evidence": [ + { + "description": "Encountered two children", + "timestamps": { + "start": "00:00:37", + "end": "00:00:42" + } + }, + { + "description": "Encountered two more children", + "timestamps": { + "start": "00:00:48", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:00:54", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "four.", + "C": "Five.", + "D": "You have encountered three children in the mall so far today, not four.", + "E": "You have encountered five children in the mall up to this point, not four." + }, + "answer_label": "B", + "answer_text": "four.", + "q_id": "01552_7", + "correct_answer": "four.", + "original_options": { + "A": "Three.", + "B": "four.", + "C": "Five.", + "D": "You have encountered three children in the mall so far today, not four.", + "E": "You have encountered five children in the mall up to this point, not four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_8", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many children have I encountered in the mall so far?", + "answer": "Five.", + "question_chinese": "现在我已经在商场遇到多少个小孩？", + "answer_chinese": "5个", + "evidence": [ + { + "description": "Encountered two children", + "timestamps": { + "start": "00:00:37", + "end": "00:00:42" + } + }, + { + "description": "Encountered two more children", + "timestamps": { + "start": "00:00:48", + "end": "00:00:52" + } + }, + { + "description": "Encountered one more child", + "timestamps": { + "start": "00:01:20", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "You have encountered four children in the mall so far today, not five.", + "E": "You have encountered six children in the mall up to this point, not five." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "01552_8", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "You have encountered four children in the mall so far today, not five.", + "E": "You have encountered six children in the mall up to this point, not five." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_5", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I asked for directions so far?", + "answer": "two times.", + "question_chinese": "现在我已经问过几次路了？", + "answer_chinese": "2次", + "evidence": [ + { + "description": "Asked for directions once", + "timestamps": { + "start": "00:00:43", + "end": "00:00:46" + } + }, + { + "description": "Asked for directions again", + "timestamps": { + "start": "00:01:07", + "end": "00:01:12" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "two times.", + "C": "Three times.", + "D": "You have asked for directions one time so far during this visit.", + "E": "You have asked for directions three times up to now during this visit." + }, + "answer_label": "B", + "answer_text": "two times.", + "q_id": "01552_5", + "correct_answer": "two times.", + "original_options": { + "A": "Once.", + "B": "two times.", + "C": "Three times.", + "D": "You have asked for directions one time so far during this visit.", + "E": "You have asked for directions three times up to now during this visit." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 13 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_6", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I asked for directions so far?", + "answer": "four times.", + "question_chinese": "现在我已经问过几次路了？", + "answer_chinese": "4次", + "evidence": [ + { + "description": "Asked for directions once", + "timestamps": { + "start": "00:00:43", + "end": "00:00:46" + } + }, + { + "description": "Asked for directions again", + "timestamps": { + "start": "00:01:07", + "end": "00:01:12" + } + }, + { + "description": "Asked for directions again", + "timestamps": { + "start": "00:01:42", + "end": "00:01:47" + } + }, + { + "description": "Asked for directions again", + "timestamps": { + "start": "00:03:23", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:03:39", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "four times.", + "B": "Three times.", + "C": "Five times.", + "D": "You have asked for directions three times so far during this visit.", + "E": "You have asked for directions five times up to this point today." + }, + "answer_label": "A", + "answer_text": "four times.", + "q_id": "01552_6", + "correct_answer": "four times.", + "original_options": { + "A": "four times.", + "B": "Three times.", + "C": "Five times.", + "D": "You have asked for directions three times so far during this visit.", + "E": "You have asked for directions five times up to this point today." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_9", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the shelf selling thermal water cups within 10 steps if I want to buy one now?", + "answer": "Yes.", + "question_chinese": "现在我想去买保温水杯，可以在10步内走到卖水杯的货架上吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "Thermal water cups are on the first shelf on the right after entering the door", + "timestamps": { + "start": "00:01:55", + "end": "00:01:59" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01552_9", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_10", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the shelf selling thermal water cups within 10 steps if I want to buy one now?", + "answer": "No.", + "question_chinese": "现在我想去买保温水杯，可以在10步内走到卖水杯的货架上吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "Thermal water cups are on the first shelf on the right after entering the door", + "timestamps": { + "start": "00:01:55", + "end": "00:01:59" + } + }, + { + "description": "I walked many steps into the store interior, the distance is relatively far beyond 10 steps range", + "timestamps": { + "start": "00:02:16", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:02:23", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "01552_10", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_6_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to buy a red hat now, which shelf level should I take it from?", + "answer": "You should go to the 1st level of the shelf to get it; the shelf is in front of you.", + "question_chinese": "现在我想去买红色的帽子，应该帽子的货架的第几层拿？", + "answer_chinese": "你应该去货架第1层拿，货架在你前面。", + "evidence": [ + { + "description": "The red hat is on the first level from bottom to top of the shelf", + "timestamps": { + "start": "00:02:40", + "end": "00:02:42" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "2nd shelf level.", + "B": "You should go to the 1st level of the shelf to get it; the shelf is in front of you.", + "C": "Top shelf level.", + "D": "Take it from the 2nd level of the shelf; the shelf is in front of you.", + "E": "Take it from the top level of the shelf; it should be directly in front of you." + }, + "answer_label": "B", + "answer_text": "You should go to the 1st level of the shelf to get it; the shelf is in front of you.", + "q_id": "01552_6_T2", + "correct_answer": "You should go to the 1st level of the shelf to get it; the shelf is in front of you.", + "original_options": { + "A": "2nd shelf level.", + "B": "You should go to the 1st level of the shelf to get it; the shelf is in front of you.", + "C": "Top shelf level.", + "D": "Take it from the 2nd level of the shelf; the shelf is in front of you.", + "E": "Take it from the top level of the shelf; it should be directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_6_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to buy a red hat now, which shelf level should I take it from?", + "answer": "The 2nd level; the shelf is in front of you.", + "question_chinese": "现在我想去买红色的帽子，应该帽子的货架的第几层拿？", + "answer_chinese": "第2层，货架在你面前。", + "evidence": [ + { + "description": "The red hat is on the second level from bottom to top of the shelf", + "timestamps": { + "start": "00:02:49", + "end": "00:02:53" + } + } + ], + "question_timestamps": "00:02:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "1st shelf level.", + "B": "The 2nd level; the shelf is in front of you.", + "C": "3rd shelf level.", + "D": "Take it from the 1st level of the shelf; the shelf is in front of you.", + "E": "Take it from the 3rd level of the shelf; it should be straight ahead of you." + }, + "answer_label": "B", + "answer_text": "The 2nd level; the shelf is in front of you.", + "q_id": "01552_6_T1", + "correct_answer": "The 2nd level; the shelf is in front of you.", + "original_options": { + "A": "1st shelf level.", + "B": "The 2nd level; the shelf is in front of you.", + "C": "3rd shelf level.", + "D": "Take it from the 1st level of the shelf; the shelf is in front of you.", + "E": "Take it from the 3rd level of the shelf; it should be straight ahead of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to eat ice cream now?", + "answer": "Turn around and walk forward a few steps, the ice cream shop is on the right.", + "question_chinese": "现在我想去吃冰淇淋的话怎么移动？", + "answer_chinese": "转身掉头往前走几步，右侧是冰淇淋店", + "evidence": [ + { + "description": "There is an ice cream shop on the left, in the shop before Jasmine Milk White shop", + "timestamps": { + "start": "00:00:04", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn around; the ice cream shop is left.", + "B": "Keep going straight; it’s on your right.", + "C": "Turn around and walk forward a few steps; the ice cream shop will be on your left.", + "D": "Turn around and walk forward a few steps, the ice cream shop is on the right.", + "E": "Turn around, walk forward a few steps, then cross over; the ice cream shop will be on the left side." + }, + "answer_label": "D", + "answer_text": "Turn around and walk forward a few steps, the ice cream shop is on the right.", + "q_id": "01552_1", + "correct_answer": "Turn around and walk forward a few steps, the ice cream shop is on the right.", + "original_options": { + "A": "Turn around; the ice cream shop is left.", + "B": "Keep going straight; it’s on your right.", + "C": "Turn around and walk forward a few steps; the ice cream shop will be on your left.", + "D": "Turn around and walk forward a few steps, the ice cream shop is on the right.", + "E": "Turn around, walk forward a few steps, then cross over; the ice cream shop will be on the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 20 + ] + }, + { + "video_uid": "01552", + "video_length": "00:03:39", + "qid": "01552_2", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to eat ice cream now?", + "answer": "Go out the door, turn right, and the next shop sells ice cream.", + "question_chinese": "现在我想去吃冰淇淋的话怎么移动？", + "answer_chinese": "出门右转隔壁店铺就可以买冰淇淋", + "evidence": [ + { + "description": "There is an ice cream shop on the left, in the shop before Jasmine Milk White shop", + "timestamps": { + "start": "00:00:04", + "end": "00:00:08" + } + }, + { + "description": "I am now at the Jasmine Milk White shop", + "timestamps": { + "start": "00:01:30", + "end": "00:01:36" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go out the door and turn left.", + "B": "Go out the door and walk straight.", + "C": "Go out the door, turn right, but the second shop down is the one that sells ice cream.", + "D": "Go out the door, turn left, and the next shop you reach is the ice cream place.", + "E": "Go out the door, turn right, and the next shop sells ice cream." + }, + "answer_label": "E", + "answer_text": "Go out the door, turn right, and the next shop sells ice cream.", + "q_id": "01552_2", + "correct_answer": "Go out the door, turn right, and the next shop sells ice cream.", + "original_options": { + "A": "Go out the door and turn left.", + "B": "Go out the door and walk straight.", + "C": "Go out the door, turn right, but the second shop down is the one that sells ice cream.", + "D": "Go out the door, turn left, and the next shop you reach is the ice cream place.", + "E": "Go out the door, turn right, and the next shop sells ice cream." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 17 + ] + } +] diff --git a/QAs/01553-mcq.json b/QAs/01553-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c4a40f9cacff7199b235a1e32ca2f307a01c27c0 --- /dev/null +++ b/QAs/01553-mcq.json @@ -0,0 +1,1039 @@ +[ + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_11", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to find a seat nearby now, can I reach a seat within 10 steps?", + "answer": "No.", + "question_chinese": "如果现在我想就近找位置坐下，可以10步到达位置吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "Currently the nearest seats are all occupied and relatively far away", + "timestamps": { + "start": "00:01:58", + "end": "00:02:01" + } + } + ], + "question_timestamps": "00:02:04", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01553_11", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_12", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to find a seat nearby now, can I reach a seat within 10 steps?", + "answer": "Yes.", + "question_chinese": "如果现在我想就近找位置坐下，可以10步到达位置吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "There are seats nearby and very close at this time", + "timestamps": { + "start": "00:03:07", + "end": "00:03:09" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01553_12", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_7_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, how many people are in line in front of me?", + "answer": "Right now, there are 2 people queued up in front of you.", + "question_chinese": "此时我前面排队多少人？", + "answer_chinese": "此时你前面排队2个人。", + "evidence": [ + { + "description": "There are two people ahead of you in the queue right now.", + "timestamps": { + "start": "00:02:03", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:02:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There are 0 people ahead.", + "B": "There is 1 person ahead.", + "C": "Right now, there are 3 people queued up in front of you.", + "D": "Right now, there are 2 people queued up in front of you.", + "E": "Right now, there are 4 people queued up in front of you, waiting ahead of you in line." + }, + "answer_label": "D", + "answer_text": "Right now, there are 2 people queued up in front of you.", + "q_id": "01553_7_T2", + "correct_answer": "Right now, there are 2 people queued up in front of you.", + "original_options": { + "A": "There are 0 people ahead.", + "B": "There is 1 person ahead.", + "C": "Right now, there are 3 people queued up in front of you.", + "D": "Right now, there are 2 people queued up in front of you.", + "E": "Right now, there are 4 people queued up in front of you, waiting ahead of you in line." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 18 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_7_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, how many people are in line in front of me?", + "answer": "Right now, there are 0 people queued up in front of you.", + "question_chinese": "此时我前面排队多少人？", + "answer_chinese": "此时我前面排队0个人。", + "evidence": [ + { + "description": "There are two people ahead of you in the queue right now.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:11" + } + } + ], + "question_timestamps": "00:02:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There is 1 person ahead.", + "B": "There are 2 people ahead.", + "C": "Right now, there are 0 people queued up in front of you.", + "D": "Right now, there are 3 people queued up in front of you.", + "E": "Right now, there are 4 people queued up in front of you, waiting ahead of you in line." + }, + "answer_label": "C", + "answer_text": "Right now, there are 0 people queued up in front of you.", + "q_id": "01553_7_T1", + "correct_answer": "Right now, there are 0 people queued up in front of you.", + "original_options": { + "A": "There is 1 person ahead.", + "B": "There are 2 people ahead.", + "C": "Right now, there are 0 people queued up in front of you.", + "D": "Right now, there are 3 people queued up in front of you.", + "E": "Right now, there are 4 people queued up in front of you, waiting ahead of you in line." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 18 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many electric bicycles have passed by me so far?", + "answer": "four.", + "question_chinese": "现在已经有几辆电动自行车车从我身边路过？", + "answer_chinese": "4辆", + "evidence": [ + { + "description": "Two electric bicycles passed by me", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "Two electric bicycles passed by me", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "four.", + "C": "Five.", + "D": "Eight electric bicycles have passed by you so far while you’ve been standing here.", + "E": "Ten electric bicycles have passed by you so far as you’ve been watching the street." + }, + "answer_label": "B", + "answer_text": "four.", + "q_id": "01553_1", + "correct_answer": "four.", + "original_options": { + "A": "Three.", + "B": "four.", + "C": "Five.", + "D": "Eight electric bicycles have passed by you so far while you’ve been standing here.", + "E": "Ten electric bicycles have passed by you so far as you’ve been watching the street." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many electric bicycles have passed by me so far?", + "answer": "Nine.", + "question_chinese": "现在已经有几辆电动自行车车从我身边路过？", + "answer_chinese": "9辆", + "evidence": [ + { + "description": "eight electric bicycles passed by me", + "timestamps": { + "start": "00:00:09", + "end": "00:00:31" + } + }, + { + "description": "One electric bicycle passed by me", + "timestamps": { + "start": "00:00:48", + "end": "00:00:51" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Eight.", + "B": "Ten.", + "C": "Seven electric bicycles have passed by you so far while you’ve been watching traffic here.", + "D": "Eleven electric bicycles have passed by you so far as you’ve been standing here.", + "E": "Nine." + }, + "answer_label": "E", + "answer_text": "Nine.", + "q_id": "01553_2", + "correct_answer": "Nine.", + "original_options": { + "A": "Eight.", + "B": "Ten.", + "C": "Seven electric bicycles have passed by you so far while you’ve been watching traffic here.", + "D": "Eleven electric bicycles have passed by you so far as you’ve been standing here.", + "E": "Nine." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_3_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned right so far?", + "answer": "So far, you have turned right 0 times, and the intersection is to your right.", + "question_chinese": "现在我已经右转了几次？", + "answer_chinese": "现在你已经右转了0次，路口在你右边。", + "evidence": [ + { + "description": "Turned right once at this moment", + "timestamps": { + "start": "00:01:00", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have turned right 1 time so far.", + "B": "You have turned right 2 times so far.", + "C": "So far, you have turned right 3 times, and the intersection is to your right.", + "D": "So far, you have turned right 4 times, and the intersection is still to your right.", + "E": "So far, you have turned right 0 times, and the intersection is to your right." + }, + "answer_label": "E", + "answer_text": "So far, you have turned right 0 times, and the intersection is to your right.", + "q_id": "01553_3_T2", + "correct_answer": "So far, you have turned right 0 times, and the intersection is to your right.", + "original_options": { + "A": "You have turned right 1 time so far.", + "B": "You have turned right 2 times so far.", + "C": "So far, you have turned right 3 times, and the intersection is to your right.", + "D": "So far, you have turned right 4 times, and the intersection is still to your right.", + "E": "So far, you have turned right 0 times, and the intersection is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 16 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_3_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned right so far?", + "answer": "So far, you have turned right 6 times, and the intersection is to your right.", + "question_chinese": "现在我已经右转了几次？", + "answer_chinese": "现在你已经右转了6次，路口在你右边。", + "evidence": [ + { + "description": "Turned right once at this moment", + "timestamps": { + "start": "00:01:19", + "end": "00:01:23" + } + }, + { + "description": "Turned right once at this moment", + "timestamps": { + "start": "00:01:58", + "end": "00:02:01" + } + }, + { + "description": "Turned right once at this moment, facing the cafeteria lady", + "timestamps": { + "start": "00:02:06", + "end": "00:02:10" + } + }, + { + "description": "Turned right once at this moment", + "timestamps": { + "start": "00:02:38", + "end": "00:02:42" + } + }, + { + "description": "Turned right once at this moment", + "timestamps": { + "start": "00:02:48", + "end": "00:02:52" + } + }, + { + "description": "Turned right once at this moment", + "timestamps": { + "start": "00:03:02", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:03:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have turned right 5 times so far.", + "B": "You have turned right 7 times so far.", + "C": "So far, you have turned right 4 times, and the intersection is to your right.", + "D": "So far, you have turned right 6 times, and the intersection is to your right.", + "E": "So far, you have turned right 8 times, and the intersection remains to your right." + }, + "answer_label": "D", + "answer_text": "So far, you have turned right 6 times, and the intersection is to your right.", + "q_id": "01553_3_T1", + "correct_answer": "So far, you have turned right 6 times, and the intersection is to your right.", + "original_options": { + "A": "You have turned right 5 times so far.", + "B": "You have turned right 7 times so far.", + "C": "So far, you have turned right 4 times, and the intersection is to your right.", + "D": "So far, you have turned right 6 times, and the intersection is to your right.", + "E": "So far, you have turned right 8 times, and the intersection remains to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 15 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_6", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned right so far?", + "answer": "six times.", + "question_chinese": "现在我已经右转了几次？", + "answer_chinese": "6次", + "evidence": [ + { + "description": "Turned right once at this moment", + "timestamps": { + "start": "00:01:19", + "end": "00:01:23" + } + }, + { + "description": "Turned right once at this moment", + "timestamps": { + "start": "00:01:58", + "end": "00:02:01" + } + }, + { + "description": "Turned right once at this moment, facing the cafeteria lady", + "timestamps": { + "start": "00:02:06", + "end": "00:02:10" + } + }, + { + "description": "Turned right once at this moment", + "timestamps": { + "start": "00:02:38", + "end": "00:02:42" + } + }, + { + "description": "Turned right once at this moment", + "timestamps": { + "start": "00:02:48", + "end": "00:02:52" + } + }, + { + "description": "Turned right once at this moment", + "timestamps": { + "start": "00:03:02", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:03:16", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Seven times.", + "B": "Five times.", + "C": "You have turned right eight times so far during this route today.", + "D": "six times.", + "E": "Up to now, you have turned right four times along your path." + }, + "answer_label": "D", + "answer_text": "six times.", + "q_id": "01553_6", + "correct_answer": "six times.", + "original_options": { + "A": "Seven times.", + "B": "Five times.", + "C": "You have turned right eight times so far during this route today.", + "D": "six times.", + "E": "Up to now, you have turned right four times along your path." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_5_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What staple food have I ordered so far?", + "answer": "By now you’ve ordered rice, and the rice is to your front-left.", + "question_chinese": "现在我已经点了什么主食？", + "answer_chinese": "现在你已经点了米饭，米饭在你左前方。", + "evidence": [ + { + "description": "There are three bowls of rice at this time", + "timestamps": { + "start": "00:02:54", + "end": "00:02:57" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’ve ordered shaomai so far.", + "B": "You ordered rice and shaomai.", + "C": "By now you’ve ordered rice, and the rice is to your front-right.", + "D": "By now you’ve ordered rice, and the rice is behind you on the left.", + "E": "By now you’ve ordered rice, and the rice is to your front-left." + }, + "answer_label": "E", + "answer_text": "By now you’ve ordered rice, and the rice is to your front-left.", + "q_id": "01553_5_T1", + "correct_answer": "By now you’ve ordered rice, and the rice is to your front-left.", + "original_options": { + "A": "You’ve ordered shaomai so far.", + "B": "You ordered rice and shaomai.", + "C": "By now you’ve ordered rice, and the rice is to your front-right.", + "D": "By now you’ve ordered rice, and the rice is behind you on the left.", + "E": "By now you’ve ordered rice, and the rice is to your front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 15 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_5_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What staple food have I ordered so far?", + "answer": "By now you’ve ordered rice and shaomai, and the rice and shaomai are to your front-left.", + "question_chinese": "现在我已经点了什么主食？", + "answer_chinese": "现在你已经点了米饭和烧麦，米饭和烧麦在你左前方。", + "evidence": [ + { + "description": "There are three bowls of rice at this time", + "timestamps": { + "start": "00:02:54", + "end": "00:02:57" + } + }, + { + "description": "Took a basket of shaomai at this time", + "timestamps": { + "start": "00:03:05", + "end": "00:03:09" + } + } + ], + "question_timestamps": "00:03:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’ve ordered only rice so far.", + "B": "By now you’ve ordered rice and shaomai, and the rice and shaomai are to your front-left.", + "C": "You’ve ordered only shaomai so far.", + "D": "By now you’ve ordered rice and shaomai, and both are to your front-right.", + "E": "By now you’ve ordered rice and shaomai, and the rice and shaomai are to your back-left." + }, + "answer_label": "B", + "answer_text": "By now you’ve ordered rice and shaomai, and the rice and shaomai are to your front-left.", + "q_id": "01553_5_T2", + "correct_answer": "By now you’ve ordered rice and shaomai, and the rice and shaomai are to your front-left.", + "original_options": { + "A": "You’ve ordered only rice so far.", + "B": "By now you’ve ordered rice and shaomai, and the rice and shaomai are to your front-left.", + "C": "You’ve ordered only shaomai so far.", + "D": "By now you’ve ordered rice and shaomai, and both are to your front-right.", + "E": "By now you’ve ordered rice and shaomai, and the rice and shaomai are to your back-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 17 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_10", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What staple food have I ordered so far?", + "answer": "Rice and Shaomai.", + "question_chinese": "现在我已经点了什么主食？", + "answer_chinese": "米饭和烧麦", + "evidence": [ + { + "description": "There are three bowls of rice at this time", + "timestamps": { + "start": "00:02:54", + "end": "00:02:57" + } + }, + { + "description": "Took a basket of shaomai at this time", + "timestamps": { + "start": "00:03:05", + "end": "00:03:09" + } + } + ], + "question_timestamps": "00:03:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Noodles and Shaomai.", + "B": "Rice and wontons.", + "C": "You have ordered rice with steamed buns as your staple items so far.", + "D": "You have ordered rice with spring rolls as your staple items so far.", + "E": "Rice and Shaomai." + }, + "answer_label": "E", + "answer_text": "Rice and Shaomai.", + "q_id": "01553_10", + "correct_answer": "Rice and Shaomai.", + "original_options": { + "A": "Noodles and Shaomai.", + "B": "Rice and wontons.", + "C": "You have ordered rice with steamed buns as your staple items so far.", + "D": "You have ordered rice with spring rolls as your staple items so far.", + "E": "Rice and Shaomai." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 13 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_7", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many dishes or soups have I ordered so far?", + "answer": "Three.", + "question_chinese": "现在我已经点了几道菜或汤？", + "answer_chinese": "3道", + "evidence": [ + { + "description": "There are three dishes at this time", + "timestamps": { + "start": "00:02:37", + "end": "00:02:40" + } + }, + { + "description": "There are three soups at this time", + "timestamps": { + "start": "00:02:54", + "end": "00:02:57" + } + } + ], + "question_timestamps": "00:02:57", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You have ordered five dishes or soups so far in total at this point.", + "E": "So far, you have ordered one dish or soup altogether during the meal." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "01553_7", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You have ordered five dishes or soups so far in total at this point.", + "E": "So far, you have ordered one dish or soup altogether during the meal." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_8", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many dishes or soups have I ordered so far?", + "answer": "six.", + "question_chinese": "现在我已经点了几道菜或汤？", + "answer_chinese": "6道", + "evidence": [ + { + "description": "There are three dishes at this time", + "timestamps": { + "start": "00:02:37", + "end": "00:02:40" + } + }, + { + "description": "There are three soups at this time", + "timestamps": { + "start": "00:02:54", + "end": "00:02:57" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "six.", + "C": "Four.", + "D": "You have ordered three dishes or soups so far in total at this point.", + "E": "So far, you have ordered only two dishes or soups altogether during the meal." + }, + "answer_label": "B", + "answer_text": "six.", + "q_id": "01553_8", + "correct_answer": "six.", + "original_options": { + "A": "Five.", + "B": "six.", + "C": "Four.", + "D": "You have ordered three dishes or soups so far in total at this point.", + "E": "So far, you have ordered only two dishes or soups altogether during the meal." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_2_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many intersections have I passed so far?", + "answer": "So far you’ve passed 0 intersections, and the intersection is to your right.", + "question_chinese": "现在我已经经过了几个路口？", + "answer_chinese": "现在你已经经过了0个路口，路口在我右边。", + "evidence": [ + { + "description": "There is an intersection on the right side at this moment", + "timestamps": { + "start": "00:00:15", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve passed 1 intersection so far.", + "B": "You’ve passed 2 intersections so far.", + "C": "So far you’ve passed 0 intersections, and the intersection is to your right.", + "D": "So far you’ve passed 1 intersection, and the intersection is to your right.", + "E": "So far you’ve passed 3 intersections, and the intersection is to your right." + }, + "answer_label": "C", + "answer_text": "So far you’ve passed 0 intersections, and the intersection is to your right.", + "q_id": "01553_2_T1", + "correct_answer": "So far you’ve passed 0 intersections, and the intersection is to your right.", + "original_options": { + "A": "You’ve passed 1 intersection so far.", + "B": "You’ve passed 2 intersections so far.", + "C": "So far you’ve passed 0 intersections, and the intersection is to your right.", + "D": "So far you’ve passed 1 intersection, and the intersection is to your right.", + "E": "So far you’ve passed 3 intersections, and the intersection is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 14 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_2_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many intersections have I passed so far?", + "answer": "By now, you’ve passed 4 intersections, and the intersections are to your right.", + "question_chinese": "现在我已经经过了几个路口？", + "answer_chinese": "现在你已经经过了4个路口，路口在我右边。", + "evidence": [ + { + "description": "There is an intersection on the right side at this moment", + "timestamps": { + "start": "00:00:16", + "end": "00:00:19" + } + }, + { + "description": "There is an intersection on the right side at this moment", + "timestamps": { + "start": "00:00:58", + "end": "00:01:00" + } + }, + { + "description": "There is an intersection on the front right side at this moment", + "timestamps": { + "start": "00:01:07", + "end": "00:01:10" + } + }, + { + "description": "There is an intersection on the right side at this moment", + "timestamps": { + "start": "00:01:12", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve passed 3 intersections by now.", + "B": "You’ve passed 5 intersections by now.", + "C": "By now, you’ve passed 2 intersections, and the intersections are to your right.", + "D": "By now, you’ve passed 4 intersections, and the intersections are to your right.", + "E": "By now, you’ve passed 6 intersections, and the intersections are to your right." + }, + "answer_label": "D", + "answer_text": "By now, you’ve passed 4 intersections, and the intersections are to your right.", + "q_id": "01553_2_T2", + "correct_answer": "By now, you’ve passed 4 intersections, and the intersections are to your right.", + "original_options": { + "A": "You’ve passed 3 intersections by now.", + "B": "You’ve passed 5 intersections by now.", + "C": "By now, you’ve passed 2 intersections, and the intersections are to your right.", + "D": "By now, you’ve passed 4 intersections, and the intersections are to your right.", + "E": "By now, you’ve passed 6 intersections, and the intersections are to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 14 + ] + }, + { + "video_uid": "01553", + "video_length": "00:03:16", + "qid": "01553_4", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many intersections have I passed so far?", + "answer": "four.", + "question_chinese": "现在我已经经过了几个路口？", + "answer_chinese": "4个", + "evidence": [ + { + "description": "There is an intersection on the right side at this moment", + "timestamps": { + "start": "00:00:16", + "end": "00:00:19" + } + }, + { + "description": "There is an intersection on the right side at this moment", + "timestamps": { + "start": "00:00:58", + "end": "00:01:00" + } + }, + { + "description": "There is an intersection on the front right side at this moment", + "timestamps": { + "start": "00:01:07", + "end": "00:01:10" + } + }, + { + "description": "There is an intersection on the right side at this moment", + "timestamps": { + "start": "00:01:12", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Five.", + "C": "four.", + "D": "You’ve passed three intersections so far, and another intersection is still ahead of you.", + "E": "You’ve passed seven intersections up to now, with the next intersection coming up on your left." + }, + "answer_label": "C", + "answer_text": "four.", + "q_id": "01553_4", + "correct_answer": "four.", + "original_options": { + "A": "Two.", + "B": "Five.", + "C": "four.", + "D": "You’ve passed three intersections so far, and another intersection is still ahead of you.", + "E": "You’ve passed seven intersections up to now, with the next intersection coming up on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 17 + ] + } +] diff --git a/QAs/01556-mcq.json b/QAs/01556-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..63a0662635798bfd54e2e3aea25fe8259c409407 --- /dev/null +++ b/QAs/01556-mcq.json @@ -0,0 +1,821 @@ +[ + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_8_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to go buy clothes right now, how should I go?", + "answer": "Turn right and walk a few steps; the clothing section is to your right.", + "question_chinese": "我现在想去买衣服应该怎么走？", + "answer_chinese": "右转走几步，衣服区在你的右边。", + "evidence": [ + { + "description": "The clothing area is on the right after entering the supermarket", + "timestamps": { + "start": "00:00:37", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left a few steps.", + "B": "Turn right and walk a few steps; the clothing section is to your right.", + "C": "Go straight ahead a few steps.", + "D": "Turn right and walk a few steps; the clothing section will be on your left.", + "E": "Turn right and walk a few steps, then turn left; the clothing section will be straight ahead." + }, + "answer_label": "B", + "answer_text": "Turn right and walk a few steps; the clothing section is to your right.", + "q_id": "01556_8_T1", + "correct_answer": "Turn right and walk a few steps; the clothing section is to your right.", + "original_options": { + "A": "Turn left a few steps.", + "B": "Turn right and walk a few steps; the clothing section is to your right.", + "C": "Go straight ahead a few steps.", + "D": "Turn right and walk a few steps; the clothing section will be on your left.", + "E": "Turn right and walk a few steps, then turn left; the clothing section will be straight ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 17 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_8_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to go buy clothes right now, how should I go?", + "answer": "Turn right and walk a few steps, then turn right again and you’ll be there. The clothing section is to your front-right.", + "question_chinese": "我现在想去买衣服应该怎么走？", + "answer_chinese": "右转走几步，再次右转就到了。衣服区在你的右前方。", + "evidence": [ + { + "description": "The clothing area is on the right after entering the supermarket", + "timestamps": { + "start": "00:00:37", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then turn right.", + "B": "Turn right once and stop.", + "C": "Turn right and walk a few steps, then turn right again and you’ll be there. The clothing section is to your front-right.", + "D": "Turn right and walk a few steps, then turn right again; the clothing section is to your back-right.", + "E": "Turn right and walk a few steps, then turn left; the clothing section is to your front-right." + }, + "answer_label": "C", + "answer_text": "Turn right and walk a few steps, then turn right again and you’ll be there. The clothing section is to your front-right.", + "q_id": "01556_8_T2", + "correct_answer": "Turn right and walk a few steps, then turn right again and you’ll be there. The clothing section is to your front-right.", + "original_options": { + "A": "Turn left, then turn right.", + "B": "Turn right once and stop.", + "C": "Turn right and walk a few steps, then turn right again and you’ll be there. The clothing section is to your front-right.", + "D": "Turn right and walk a few steps, then turn right again; the clothing section is to your back-right.", + "E": "Turn right and walk a few steps, then turn left; the clothing section is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_5", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which floor am I on now?", + "answer": "First floor.", + "question_chinese": "现在我在几楼？", + "answer_chinese": "一楼", + "evidence": [ + { + "description": "Entered the supermarket on the first floor", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Second floor.", + "B": "First floor.", + "C": "Basement level.", + "D": "You’re on the mezzanine level, not on the first floor right now.", + "E": "You’re on the lobby level right now, rather than being on the first floor." + }, + "answer_label": "B", + "answer_text": "First floor.", + "q_id": "01556_5", + "correct_answer": "First floor.", + "original_options": { + "A": "Second floor.", + "B": "First floor.", + "C": "Basement level.", + "D": "You’re on the mezzanine level, not on the first floor right now.", + "E": "You’re on the lobby level right now, rather than being on the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_6", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which floor am I on now?", + "answer": "Second floor.", + "question_chinese": "现在我在几楼？", + "answer_chinese": "二楼", + "evidence": [ + { + "description": "Entered the supermarket on the first floor", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "Went up to the second floor", + "timestamps": { + "start": "00:01:22", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:01:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You are on the first floor.", + "B": "Second floor.", + "C": "You are on the third floor.", + "D": "You are on the ground floor right now, not up on the second floor.", + "E": "You are on the fourth floor currently, rather than being on the second floor." + }, + "answer_label": "B", + "answer_text": "Second floor.", + "q_id": "01556_6", + "correct_answer": "Second floor.", + "original_options": { + "A": "You are on the first floor.", + "B": "Second floor.", + "C": "You are on the third floor.", + "D": "You are on the ground floor right now, not up on the second floor.", + "E": "You are on the fourth floor currently, rather than being on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_11", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned so far?", + "answer": "two times.", + "question_chinese": "现在我已经转了几次？", + "answer_chinese": "2次", + "evidence": [ + { + "description": "Turned right once", + "timestamps": { + "start": "00:00:07", + "end": "00:00:13" + } + }, + { + "description": "Turned left once", + "timestamps": { + "start": "00:00:14", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "two times.", + "B": "Zero times.", + "C": "One time.", + "D": "You’ve turned three times so far, including one extra turn after the last straight segment.", + "E": "You’ve turned four times up to now, with two additional turns after the second corner." + }, + "answer_label": "A", + "answer_text": "two times.", + "q_id": "01556_11", + "correct_answer": "two times.", + "original_options": { + "A": "two times.", + "B": "Zero times.", + "C": "One time.", + "D": "You’ve turned three times so far, including one extra turn after the last straight segment.", + "E": "You’ve turned four times up to now, with two additional turns after the second corner." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_12", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned so far?", + "answer": "six times.", + "question_chinese": "现在我已经转了几次？", + "answer_chinese": "6次", + "evidence": [ + { + "description": "Turned right once", + "timestamps": { + "start": "00:00:07", + "end": "00:00:13" + } + }, + { + "description": "Turned left once", + "timestamps": { + "start": "00:00:14", + "end": "00:00:18" + } + }, + { + "description": "Turned right once", + "timestamps": { + "start": "00:00:25", + "end": "00:00:28" + } + }, + { + "description": "Turned left once", + "timestamps": { + "start": "00:00:39", + "end": "00:00:44" + } + }, + { + "description": "Turned left once", + "timestamps": { + "start": "00:00:39", + "end": "00:00:44" + } + }, + { + "description": "Turned left once", + "timestamps": { + "start": "00:00:51", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "five times.", + "B": "six times.", + "C": "seven times.", + "D": "four times, counting only the major turns you have made while moving so far.", + "E": "eight times, counting every direction change you made while walking up to now." + }, + "answer_label": "B", + "answer_text": "six times.", + "q_id": "01556_12", + "correct_answer": "six times.", + "original_options": { + "A": "five times.", + "B": "six times.", + "C": "seven times.", + "D": "four times, counting only the major turns you have made while moving so far.", + "E": "eight times, counting every direction change you made while walking up to now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to buy snacks now?", + "answer": "Turn right and walk a few steps.", + "question_chinese": "现在我想买零食应该怎么移动？", + "answer_chinese": "向右转走几步", + "evidence": [ + { + "description": "The snack area is on the right", + "timestamps": { + "start": "00:02:25", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:02:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left and walk a few steps.", + "B": "Walk straight a few steps.", + "C": "Turn right and walk a long distance to reach the snacks from your current spot.", + "D": "Turn around first, then walk a few steps to where the snacks are located now.", + "E": "Turn right and walk a few steps." + }, + "answer_label": "E", + "answer_text": "Turn right and walk a few steps.", + "q_id": "01556_1", + "correct_answer": "Turn right and walk a few steps.", + "original_options": { + "A": "Turn left and walk a few steps.", + "B": "Walk straight a few steps.", + "C": "Turn right and walk a long distance to reach the snacks from your current spot.", + "D": "Turn around first, then walk a few steps to where the snacks are located now.", + "E": "Turn right and walk a few steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 15 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to buy snacks now?", + "answer": "Turn around and walk a few steps.", + "question_chinese": "现在我想买零食应该怎么移动？", + "answer_chinese": "向后转走几步", + "evidence": [ + { + "description": "The snack area is on the right", + "timestamps": { + "start": "00:02:25", + "end": "00:02:30" + } + }, + { + "description": "Turned left to go to the checkout counter", + "timestamps": { + "start": "00:02:38", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:02:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right and walk a few steps.", + "B": "Turn left and walk a few steps.", + "C": "Turn around and walk a few steps.", + "D": "Walk straight ahead a few steps without turning around first at the start.", + "E": "Turn around, then keep walking past the snack area instead of stopping after a few steps." + }, + "answer_label": "C", + "answer_text": "Turn around and walk a few steps.", + "q_id": "01556_2", + "correct_answer": "Turn around and walk a few steps.", + "original_options": { + "A": "Turn right and walk a few steps.", + "B": "Turn left and walk a few steps.", + "C": "Turn around and walk a few steps.", + "D": "Walk straight ahead a few steps without turning around first at the start.", + "E": "Turn around, then keep walking past the snack area instead of stopping after a few steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 16 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_5_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I using to carry the things I bought right now?", + "answer": "You used a basket. The basket is in the waiter’s hand, and the waiter is to your right.", + "question_chinese": "现在我用什么装买的东西？", + "answer_chinese": "你用篮子，篮子在服务员手里。服务员在你的右边。", + "evidence": [ + { + "description": "Currently carrying a basket", + "timestamps": { + "start": "00:01:53", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:01:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You used a cart.", + "B": "You are carrying items by hand.", + "C": "You used a basket, but you are holding it yourself rather than the waiter holding it.", + "D": "You used a basket. The basket is in the waiter’s hand, and the waiter is to your right.", + "E": "You used a basket, and it is in the waiter’s hand, but the waiter is positioned to your left." + }, + "answer_label": "D", + "answer_text": "You used a basket. The basket is in the waiter’s hand, and the waiter is to your right.", + "q_id": "01556_5_T1", + "correct_answer": "You used a basket. The basket is in the waiter’s hand, and the waiter is to your right.", + "original_options": { + "A": "You used a cart.", + "B": "You are carrying items by hand.", + "C": "You used a basket, but you are holding it yourself rather than the waiter holding it.", + "D": "You used a basket. The basket is in the waiter’s hand, and the waiter is to your right.", + "E": "You used a basket, and it is in the waiter’s hand, but the waiter is positioned to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 16, + 20 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_5_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I using to carry the things I bought right now?", + "answer": "You used a cart. The waiter is pushing the cart on your left.", + "question_chinese": "现在我用什么装买的东西？", + "answer_chinese": "你用推车，推车在你的左边被服务员推着。", + "evidence": [ + { + "description": "Load things into the cart.", + "timestamps": { + "start": "00:02:37", + "end": "00:02:41" + } + } + ], + "question_timestamps": "00:02:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You used a basket.", + "B": "You are holding them by hand.", + "C": "You used a cart, but the waiter is pushing it on your right side instead of your left.", + "D": "You used a basket, with the basket in the waiter’s hand, and the waiter is standing to your right.", + "E": "You used a cart. The waiter is pushing the cart on your left." + }, + "answer_label": "E", + "answer_text": "You used a cart. The waiter is pushing the cart on your left.", + "q_id": "01556_5_T2", + "correct_answer": "You used a cart. The waiter is pushing the cart on your left.", + "original_options": { + "A": "You used a basket.", + "B": "You are holding them by hand.", + "C": "You used a cart, but the waiter is pushing it on your right side instead of your left.", + "D": "You used a basket, with the basket in the waiter’s hand, and the waiter is standing to your right.", + "E": "You used a cart. The waiter is pushing the cart on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 18, + 20 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_7", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have given me directions so far?", + "answer": "one person.", + "question_chinese": "现在有几个人为我指路？", + "answer_chinese": "1个人", + "evidence": [ + { + "description": "A staff member told me to turn towards the supermarket direction", + "timestamps": { + "start": "00:00:19", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:38", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "two people.", + "B": "five people.", + "C": "one person.", + "D": "seven people have given you directions up to this point in the video so far.", + "E": "ten people have given you directions so far during your trip up to now." + }, + "answer_label": "C", + "answer_text": "one person.", + "q_id": "01556_7", + "correct_answer": "one person.", + "original_options": { + "A": "two people.", + "B": "five people.", + "C": "one person.", + "D": "seven people have given you directions up to this point in the video so far.", + "E": "ten people have given you directions so far during your trip up to now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_8", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have given me directions so far?", + "answer": "three people.", + "question_chinese": "现在有几个人为我指路？", + "answer_chinese": "3个人", + "evidence": [ + { + "description": "A staff member told me to turn towards the supermarket direction", + "timestamps": { + "start": "00:00:19", + "end": "00:00:23" + } + }, + { + "description": "A staff member took me to the elevator", + "timestamps": { + "start": "00:00:41", + "end": "00:00:42" + } + }, + { + "description": "A staff member took me to the vegetable area", + "timestamps": { + "start": "00:01:49", + "end": "00:02:01" + } + } + ], + "question_timestamps": "00:02:24", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "two people.", + "B": "five people.", + "C": "one person has given you directions so far during your walk up to now.", + "D": "six people have given you directions so far during your walk up to now.", + "E": "three people." + }, + "answer_label": "E", + "answer_text": "three people.", + "q_id": "01556_8", + "correct_answer": "three people.", + "original_options": { + "A": "two people.", + "B": "five people.", + "C": "one person has given you directions so far during your walk up to now.", + "D": "six people have given you directions so far during your walk up to now.", + "E": "three people." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_3", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the self-checkout area in for me now?", + "answer": "Left side.", + "question_chinese": "现在自助收银区在我的哪个方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "The self-checkout area is now on my left, in front of the manual checkout", + "timestamps": { + "start": "00:02:36", + "end": "00:02:38" + } + } + ], + "question_timestamps": "00:02:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left side.", + "B": "Right side.", + "C": "Straight ahead.", + "D": "Behind you, on the opposite side from where you are facing now.", + "E": "Farther ahead and slightly to the right, past the area in front of you." + }, + "answer_label": "A", + "answer_text": "Left side.", + "q_id": "01556_3", + "correct_answer": "Left side.", + "original_options": { + "A": "Left side.", + "B": "Right side.", + "C": "Straight ahead.", + "D": "Behind you, on the opposite side from where you are facing now.", + "E": "Farther ahead and slightly to the right, past the area in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "01556", + "video_length": "00:03:00", + "qid": "01556_4", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the self-checkout area in for me now?", + "answer": "Back side.", + "question_chinese": "现在自助收银区在我的哪个方向？", + "answer_chinese": "后方", + "evidence": [ + { + "description": "The self-checkout area is in front of the manual checkout", + "timestamps": { + "start": "00:02:36", + "end": "00:02:38" + } + }, + { + "description": "I have passed the manual checkout area and am checking out", + "timestamps": { + "start": "00:02:40", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Front side.", + "B": "Left side.", + "C": "It is in the right corner from where you are standing now.", + "D": "Back side.", + "E": "It is in the center area ahead of you, not behind you at this moment." + }, + "answer_label": "D", + "answer_text": "Back side.", + "q_id": "01556_4", + "correct_answer": "Back side.", + "original_options": { + "A": "Front side.", + "B": "Left side.", + "C": "It is in the right corner from where you are standing now.", + "D": "Back side.", + "E": "It is in the center area ahead of you, not behind you at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 15 + ] + } +] diff --git a/QAs/01562-mcq.json b/QAs/01562-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..10ef77367f05f3e56ae4e60aa1685b7f7572ca22 --- /dev/null +++ b/QAs/01562-mcq.json @@ -0,0 +1,1043 @@ +[ + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_5", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans have I seen on the road so far?", + "answer": "two.", + "question_chinese": "现在我在路上看到了几个垃圾桶？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "There are two trash cans across the street", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "two.", + "C": "Three.", + "D": "You've only passed one trash can along the road so far today.", + "E": "You've already seen three trash cans on the road up to this point." + }, + "answer_label": "B", + "answer_text": "two.", + "q_id": "01562_5", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "two.", + "C": "Three.", + "D": "You've only passed one trash can along the road so far today.", + "E": "You've already seen three trash cans on the road up to this point." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_6", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans have I seen on the road so far?", + "answer": "six.", + "question_chinese": "现在我在路上看到了几个垃圾桶？", + "answer_chinese": "6个", + "evidence": [ + { + "description": "There are two trash cans across the street", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "There is a green trash can ahead", + "timestamps": { + "start": "00:00:41", + "end": "00:00:43" + } + }, + { + "description": "There are 3 trash cans ahead", + "timestamps": { + "start": "00:01:50", + "end": "00:01:54" + } + } + ], + "question_timestamps": "00:02:41", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Eight.", + "C": "You have seen five trash cans on the road so far today.", + "D": "six.", + "E": "You have already passed eight trash cans while traveling along the road." + }, + "answer_label": "D", + "answer_text": "six.", + "q_id": "01562_6", + "correct_answer": "six.", + "original_options": { + "A": "Five.", + "B": "Eight.", + "C": "You have seen five trash cans on the road so far today.", + "D": "six.", + "E": "You have already passed eight trash cans while traveling along the road." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_2_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many electric bicycles have I encountered on the road so far?", + "answer": "You encountered one car; it's to your left.", + "question_chinese": "现在我在路上遇到了几辆电动自行车？", + "answer_chinese": "你遇到1辆车，它在你左边。", + "evidence": [ + { + "description": "There is a Meituan shared electric bicycle in front of me", + "timestamps": { + "start": "00:00:44", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You encountered zero car; it's to your left.", + "B": "You encountered two car; it's to your left.", + "C": "You encountered one car; it's to your left.", + "D": "You encountered three car; it's to your left as you continue down the road.", + "E": "You encountered four car so far, and the most recent one is still on your left." + }, + "answer_label": "C", + "answer_text": "You encountered one car; it's to your left.", + "q_id": "01562_2_T1", + "correct_answer": "You encountered one car; it's to your left.", + "original_options": { + "A": "You encountered zero car; it's to your left.", + "B": "You encountered two car; it's to your left.", + "C": "You encountered one car; it's to your left.", + "D": "You encountered three car; it's to your left as you continue down the road.", + "E": "You encountered four car so far, and the most recent one is still on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 14, + 16 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_2_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many electric bicycles have I encountered on the road so far?", + "answer": "You encountered three cars; the last one is to your front-left.", + "question_chinese": "现在我在路上遇到了几辆电动自行车？", + "answer_chinese": "你遇到了3辆车，最后一辆在你左前方。", + "evidence": [ + { + "description": "There is a Meituan shared electric bicycle in front of me", + "timestamps": { + "start": "00:00:44", + "end": "00:00:48" + } + }, + { + "description": "There is an electric bicycle in front of me", + "timestamps": { + "start": "00:01:33", + "end": "00:01:35" + } + }, + { + "description": "There is an electric bicycle in front left of me", + "timestamps": { + "start": "00:02:27", + "end": "00:02:31" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One electric bicycle.", + "B": "Two electric bicycles.", + "C": "You encountered four electric bicycles; the last one is to your front-left as you continue forward.", + "D": "You encountered three cars; the last one is to your front-left.", + "E": "You encountered five electric bicycles; the last one is positioned to your front-left at this moment." + }, + "answer_label": "D", + "answer_text": "You encountered three cars; the last one is to your front-left.", + "q_id": "01562_2_T2", + "correct_answer": "You encountered three cars; the last one is to your front-left.", + "original_options": { + "A": "One electric bicycle.", + "B": "Two electric bicycles.", + "C": "You encountered four electric bicycles; the last one is to your front-left as you continue forward.", + "D": "You encountered three cars; the last one is to your front-left.", + "E": "You encountered five electric bicycles; the last one is positioned to your front-left at this moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_4", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many electric bicycles have I encountered on the road so far?", + "answer": "four.", + "question_chinese": "现在我在路上遇到了几辆电动自行车？", + "answer_chinese": "4辆", + "evidence": [ + { + "description": "There is a Meituan shared electric bicycle in front of me", + "timestamps": { + "start": "00:00:44", + "end": "00:00:48" + } + }, + { + "description": "There is an electric bicycle in front of me", + "timestamps": { + "start": "00:01:33", + "end": "00:01:35" + } + }, + { + "description": "There is an electric bicycle in front right of me", + "timestamps": { + "start": "00:02:32", + "end": "00:02:35" + } + }, + { + "description": "There is an electric bicycle at the pharmacy entrance", + "timestamps": { + "start": "00:03:08", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:03:12", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "four.", + "B": "Two.", + "C": "Five.", + "D": "You have encountered two electric bicycles on the road so far today.", + "E": "You have encountered five electric bicycles while traveling along the road so far." + }, + "answer_label": "A", + "answer_text": "four.", + "q_id": "01562_4", + "correct_answer": "four.", + "original_options": { + "A": "four.", + "B": "Two.", + "C": "Five.", + "D": "You have encountered two electric bicycles on the road so far today.", + "E": "You have encountered five electric bicycles while traveling along the road so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_5_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned around so far?", + "answer": "So far, you've turned around to the right once.", + "question_chinese": "现在我已经掉头几次了？", + "answer_chinese": "现在你向右掉头了1次", + "evidence": [ + { + "description": "Turned around from the road towards the pharmacy", + "timestamps": { + "start": "00:00:36", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, you've turned around to the right once.", + "B": "No turns yet.", + "C": "Left once.", + "D": "So far, you've turned around twice; the last time you turned around to the left.", + "E": "So far, you've turned around twice; the last time you turned around to the right." + }, + "answer_label": "A", + "answer_text": "So far, you've turned around to the right once.", + "q_id": "01562_5_T1", + "correct_answer": "So far, you've turned around to the right once.", + "original_options": { + "A": "So far, you've turned around to the right once.", + "B": "No turns yet.", + "C": "Left once.", + "D": "So far, you've turned around twice; the last time you turned around to the left.", + "E": "So far, you've turned around twice; the last time you turned around to the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 15, + 15 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_5_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned around so far?", + "answer": "So far, you've turned around twice; the last time you turned around to the right.", + "question_chinese": "现在我已经掉头几次了？", + "answer_chinese": "现在你掉头了2次，最后一次你向右掉头。", + "evidence": [ + { + "description": "Turned around from the road towards the pharmacy", + "timestamps": { + "start": "00:00:37", + "end": "00:00:44" + } + }, + { + "description": "Turned around again after reaching the end of this road", + "timestamps": { + "start": "00:02:16", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:02:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Right once.", + "B": "So far, you've turned around twice; the last time you turned around to the right.", + "C": "No turns yet.", + "D": "So far, you've turned around twice; the last time you turned around to the left.", + "E": "So far, you've turned around three times; the last time you turned around to the right." + }, + "answer_label": "B", + "answer_text": "So far, you've turned around twice; the last time you turned around to the right.", + "q_id": "01562_5_T2", + "correct_answer": "So far, you've turned around twice; the last time you turned around to the right.", + "original_options": { + "A": "Right once.", + "B": "So far, you've turned around twice; the last time you turned around to the right.", + "C": "No turns yet.", + "D": "So far, you've turned around twice; the last time you turned around to the left.", + "E": "So far, you've turned around three times; the last time you turned around to the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 16 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_10", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned around so far?", + "answer": "two.", + "question_chinese": "现在我已经掉头几次了？", + "answer_chinese": "2次", + "evidence": [ + { + "description": "Turned around from the road towards the pharmacy", + "timestamps": { + "start": "00:00:36", + "end": "00:00:42" + } + }, + { + "description": "Turned around again after reaching the end of this road", + "timestamps": { + "start": "00:02:16", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:02:51", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "two.", + "C": "Three.", + "D": "You have turned around only once so far, and then continued walking forward afterward.", + "E": "You have turned around three times so far, alternating directions before continuing along the path." + }, + "answer_label": "B", + "answer_text": "two.", + "q_id": "01562_10", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "two.", + "C": "Three.", + "D": "You have turned around only once so far, and then continued walking forward afterward.", + "E": "You have turned around three times so far, alternating directions before continuing along the path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_7", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many shops have I passed by so far?", + "answer": "seven.", + "question_chinese": "现在我已经经过了多少个店铺？", + "answer_chinese": "7个", + "evidence": [ + { + "description": "There are 8 shops here, the studio is the 7th one", + "timestamps": { + "start": "00:00:58", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "Nine.", + "C": "You have passed by six shops so far along the route you have been walking.", + "D": "You have passed by nine shops so far, counting every storefront you went past on the way.", + "E": "seven." + }, + "answer_label": "E", + "answer_text": "seven.", + "q_id": "01562_7", + "correct_answer": "seven.", + "original_options": { + "A": "Six.", + "B": "Nine.", + "C": "You have passed by six shops so far along the route you have been walking.", + "D": "You have passed by nine shops so far, counting every storefront you went past on the way.", + "E": "seven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 17 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_8", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many shops have I passed by so far?", + "answer": "Eight.", + "question_chinese": "现在我已经经过了多少个店铺？", + "answer_chinese": "8个", + "evidence": [ + { + "description": "There are 8 shops here", + "timestamps": { + "start": "00:00:58", + "end": "00:01:00" + } + }, + { + "description": "I have reached the end of this road", + "timestamps": { + "start": "00:02:04", + "end": "00:02:07" + } + } + ], + "question_timestamps": "00:02:11", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Seven.", + "B": "Ten.", + "C": "You have passed by seven shops so far as you moved down the street.", + "D": "Eight.", + "E": "You have passed by ten shops so far, including each shopfront you walked past." + }, + "answer_label": "D", + "answer_text": "Eight.", + "q_id": "01562_8", + "correct_answer": "Eight.", + "original_options": { + "A": "Seven.", + "B": "Ten.", + "C": "You have passed by seven shops so far as you moved down the street.", + "D": "Eight.", + "E": "You have passed by ten shops so far, including each shopfront you walked past." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_13", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many choices do I have if I want to buy lamb or other parts of sheep now?", + "answer": "one.", + "question_chinese": "现在我想去买关于羊肉或者羊的其他部位我有几种选择？", + "answer_chinese": "一种", + "evidence": [ + { + "description": "There is a beef and lamb shop across the street", + "timestamps": { + "start": "00:00:08", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:01:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "You have two choices available right now for lamb or other sheep parts.", + "D": "You have three different choices at the moment if you want lamb or other sheep parts.", + "E": "one." + }, + "answer_label": "E", + "answer_text": "one.", + "q_id": "01562_13", + "correct_answer": "one.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "You have two choices available right now for lamb or other sheep parts.", + "D": "You have three different choices at the moment if you want lamb or other sheep parts.", + "E": "one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 16 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_14", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many choices do I have if I want to buy lamb or other parts of sheep now?", + "answer": "two.", + "question_chinese": "现在我想去买关于羊肉或者羊的其他部位我有几种选择？", + "answer_chinese": "2种", + "evidence": [ + { + "description": "There is a beef and lamb shop across the street", + "timestamps": { + "start": "00:00:08", + "end": "00:00:12" + } + }, + { + "description": "There is a lamb offal shop on the left", + "timestamps": { + "start": "00:02:34", + "end": "00:02:38" + } + } + ], + "question_timestamps": "00:03:28", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "two.", + "B": "One.", + "C": "Three.", + "D": "You have only one choice right now for lamb or other parts of sheep.", + "E": "You have three choices available now if you want to buy lamb or other sheep parts." + }, + "answer_label": "A", + "answer_text": "two.", + "q_id": "01562_14", + "correct_answer": "two.", + "original_options": { + "A": "two.", + "B": "One.", + "C": "Three.", + "D": "You have only one choice right now for lamb or other parts of sheep.", + "E": "You have three choices available now if you want to buy lamb or other sheep parts." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_11", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the noodle restaurant within 15 steps if I want to eat noodles now?", + "answer": "No.", + "question_chinese": "现在我想去吃面,15步之内能到吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "There is a noodle restaurant in the shops on the right, relatively far away", + "timestamps": { + "start": "00:00:36", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01562_11", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_12", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the noodle restaurant within 15 steps if I want to eat noodles now?", + "answer": "Yes.", + "question_chinese": "现在我想去吃面,15步之内能到吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "There is a noodle restaurant in the shops on the right, the noodle restaurant is close to the shed", + "timestamps": { + "start": "00:00:36", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:01:39", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "01562_12", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_15", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to send a package now?", + "answer": "Turn around and walk straight past the residential area entrance, Cainiao Guoguo is on the left side of the residential area entrance.", + "question_chinese": "现在我想去寄快递应该怎么移动？", + "answer_chinese": "掉头直走走过小区口，小区口的左侧就是菜鸟裹裹", + "evidence": [ + { + "description": "There is a Cainiao Guoguo at the residential area entrance", + "timestamps": { + "start": "00:02:08", + "end": "00:02:12" + } + }, + { + "description": "I turned around on this street at the residential area entrance", + "timestamps": { + "start": "00:02:16", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:02:38", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right and keep walking straight.", + "B": "Turn left past the entrance.", + "C": "Turn around and walk straight past the residential area entrance, Cainiao Guoguo is on the left side of the residential area entrance.", + "D": "Turn around and walk straight to the residential area entrance; Cainiao Guoguo is on the right side there.", + "E": "Turn around and walk straight past the residential area entrance; Cainiao Guoguo is on the right side of the entrance." + }, + "answer_label": "C", + "answer_text": "Turn around and walk straight past the residential area entrance, Cainiao Guoguo is on the left side of the residential area entrance.", + "q_id": "01562_15", + "correct_answer": "Turn around and walk straight past the residential area entrance, Cainiao Guoguo is on the left side of the residential area entrance.", + "original_options": { + "A": "Turn right and keep walking straight.", + "B": "Turn left past the entrance.", + "C": "Turn around and walk straight past the residential area entrance, Cainiao Guoguo is on the left side of the residential area entrance.", + "D": "Turn around and walk straight to the residential area entrance; Cainiao Guoguo is on the right side there.", + "E": "Turn around and walk straight past the residential area entrance; Cainiao Guoguo is on the right side of the entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 20 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_16", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move if I want to send a package now?", + "answer": "Exit the pharmacy, turn right and keep walking straight, pass the residential area entrance, Cainiao Guoguo is on the left side of the residential area entrance.", + "question_chinese": "现在我想去寄快递应该怎么移动？", + "answer_chinese": "出药店右转一直直走，走过小区口，小区口的左侧就是菜鸟裹裹", + "evidence": [ + { + "description": "There is a Cainiao Guoguo at the residential area entrance", + "timestamps": { + "start": "00:02:08", + "end": "00:02:12" + } + }, + { + "description": "I turned around on this street at the residential area entrance", + "timestamps": { + "start": "00:02:16", + "end": "00:02:25" + } + }, + { + "description": "I turned left into the pharmacy", + "timestamps": { + "start": "00:02:59", + "end": "00:03:05" + } + } + ], + "question_timestamps": "00:03:07", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Exit the pharmacy, then turn left.", + "B": "Exit the pharmacy and go straight ahead.", + "C": "Exit the pharmacy, turn right, walk straight, and Cainiao Guoguo will be on your right at the entrance.", + "D": "Exit the pharmacy, turn right, keep going, and you’ll find Cainiao Guoguo before you reach the residential area entrance.", + "E": "Exit the pharmacy, turn right and keep walking straight, pass the residential area entrance, Cainiao Guoguo is on the left side of the residential area entrance." + }, + "answer_label": "E", + "answer_text": "Exit the pharmacy, turn right and keep walking straight, pass the residential area entrance, Cainiao Guoguo is on the left side of the residential area entrance.", + "q_id": "01562_16", + "correct_answer": "Exit the pharmacy, turn right and keep walking straight, pass the residential area entrance, Cainiao Guoguo is on the left side of the residential area entrance.", + "original_options": { + "A": "Exit the pharmacy, then turn left.", + "B": "Exit the pharmacy and go straight ahead.", + "C": "Exit the pharmacy, turn right, walk straight, and Cainiao Guoguo will be on your right at the entrance.", + "D": "Exit the pharmacy, turn right, keep going, and you’ll find Cainiao Guoguo before you reach the residential area entrance.", + "E": "Exit the pharmacy, turn right and keep walking straight, pass the residential area entrance, Cainiao Guoguo is on the left side of the residential area entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 20 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the pharmacy in for me now?", + "answer": "Right side.", + "question_chinese": "现在药店在我的什么方位？", + "answer_chinese": "右侧", + "evidence": [ + { + "description": "The pharmacy is in front of me", + "timestamps": { + "start": "00:00:45", + "end": "00:00:49" + } + } + ], + "question_timestamps": "00:01:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left side.", + "B": "Straight ahead.", + "C": "Behind you on the left side as you face forward, not on your right side at all.", + "D": "Right side.", + "E": "On your left front side, just past where you are standing, rather than to your right." + }, + "answer_label": "D", + "answer_text": "Right side.", + "q_id": "01562_1", + "correct_answer": "Right side.", + "original_options": { + "A": "Left side.", + "B": "Straight ahead.", + "C": "Behind you on the left side as you face forward, not on your right side at all.", + "D": "Right side.", + "E": "On your left front side, just past where you are standing, rather than to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 16 + ] + }, + { + "video_uid": "01562", + "video_length": "00:03:28", + "qid": "01562_2", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the pharmacy in for me now?", + "answer": "Right back side.", + "question_chinese": "现在药店在我的什么方位？", + "answer_chinese": "右后方", + "evidence": [ + { + "description": "The pharmacy is in front of me", + "timestamps": { + "start": "00:00:45", + "end": "00:00:49" + } + }, + { + "description": "Walked forward for a distance", + "timestamps": { + "start": "00:01:37", + "end": "00:01:40" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Left back side.", + "B": "Right front side.", + "C": "Right back side.", + "D": "Straight behind you, not to the right, as you look from your current position.", + "E": "On your right side ahead of you, rather than behind your right shoulder." + }, + "answer_label": "C", + "answer_text": "Right back side.", + "q_id": "01562_2", + "correct_answer": "Right back side.", + "original_options": { + "A": "Left back side.", + "B": "Right front side.", + "C": "Right back side.", + "D": "Straight behind you, not to the right, as you look from your current position.", + "E": "On your right side ahead of you, rather than behind your right shoulder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 13 + ] + } +] diff --git a/QAs/01569-mcq.json b/QAs/01569-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1e85a270c9fd9393c1c3b97d578ff97bcb72e8cc --- /dev/null +++ b/QAs/01569-mcq.json @@ -0,0 +1,1060 @@ +[ + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_9", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned since the beginning?", + "answer": "four times.", + "question_chinese": "从开始到现在我转了几次弯？", + "answer_chinese": "4次", + "evidence": [ + { + "description": "I turned left once", + "timestamps": { + "start": "00:00:06", + "end": "00:00:13" + } + }, + { + "description": "I turned right once", + "timestamps": { + "start": "00:00:28", + "end": "00:00:34" + } + }, + { + "description": "I turned left once", + "timestamps": { + "start": "00:00:38", + "end": "00:00:42" + } + }, + { + "description": "I turned right once", + "timestamps": { + "start": "00:00:52", + "end": "00:00:56" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "three times.", + "B": "five times.", + "C": "You have turned two times since the beginning of the route, not four times in total.", + "D": "four times.", + "E": "You have turned six times in total since you started moving, rather than only four." + }, + "answer_label": "D", + "answer_text": "four times.", + "q_id": "01569_9", + "correct_answer": "four times.", + "original_options": { + "A": "three times.", + "B": "five times.", + "C": "You have turned two times since the beginning of the route, not four times in total.", + "D": "four times.", + "E": "You have turned six times in total since you started moving, rather than only four." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 15 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_10", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I turned since the beginning?", + "answer": "seven times.", + "question_chinese": "从开始到现在我转了几次弯？", + "answer_chinese": "7次", + "evidence": [ + { + "description": "I turned left once", + "timestamps": { + "start": "00:00:06", + "end": "00:00:13" + } + }, + { + "description": "I turned right once", + "timestamps": { + "start": "00:00:28", + "end": "00:00:34" + } + }, + { + "description": "I turned left once", + "timestamps": { + "start": "00:00:38", + "end": "00:00:42" + } + }, + { + "description": "I turned right once", + "timestamps": { + "start": "00:00:52", + "end": "00:00:56" + } + }, + { + "description": "I turned left once", + "timestamps": { + "start": "00:01:10", + "end": "00:01:15" + } + }, + { + "description": "I turned left once", + "timestamps": { + "start": "00:01:23", + "end": "00:01:28" + } + }, + { + "description": "I turned left once", + "timestamps": { + "start": "00:01:33", + "end": "00:01:36" + } + } + ], + "question_timestamps": "00:01:40", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "six times.", + "B": "seven times.", + "C": "eight times.", + "D": "You have turned five times since the beginning of the route, not seven times total.", + "E": "You took nine turns in total from the start up to now, not seven turns." + }, + "answer_label": "B", + "answer_text": "seven times.", + "q_id": "01569_10", + "correct_answer": "seven times.", + "original_options": { + "A": "six times.", + "B": "seven times.", + "C": "eight times.", + "D": "You have turned five times since the beginning of the route, not seven times total.", + "E": "You took nine turns in total from the start up to now, not seven turns." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What have I bought since entering the cafeteria?", + "answer": "Chicken steak rice.", + "question_chinese": "我从进食堂到现在买了什么？", + "answer_chinese": "鸡排饭", + "evidence": [ + { + "description": "I bought a chicken steak rice", + "timestamps": { + "start": "00:01:28", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:01:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Tomato beef noodles.", + "B": "Chicken burger meal.", + "C": "You bought chicken wings with soup after entering the cafeteria, not chicken steak rice.", + "D": "You purchased a chicken sandwich with fries since you entered the cafeteria, not chicken steak rice.", + "E": "Chicken steak rice." + }, + "answer_label": "E", + "answer_text": "Chicken steak rice.", + "q_id": "01569_7", + "correct_answer": "Chicken steak rice.", + "original_options": { + "A": "Tomato beef noodles.", + "B": "Chicken burger meal.", + "C": "You bought chicken wings with soup after entering the cafeteria, not chicken steak rice.", + "D": "You purchased a chicken sandwich with fries since you entered the cafeteria, not chicken steak rice.", + "E": "Chicken steak rice." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 16 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_8", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What have I bought since entering the cafeteria?", + "answer": "Chicken steak rice and tomato beef noodles.", + "question_chinese": "我从进食堂到现在买了什么？", + "answer_chinese": "鸡排饭和番茄牛肉拉面", + "evidence": [ + { + "description": "I bought a chicken steak rice", + "timestamps": { + "start": "00:01:28", + "end": "00:01:32" + } + }, + { + "description": "I bought a bowl of tomato beef noodles", + "timestamps": { + "start": "00:02:18", + "end": "00:02:23" + } + } + ], + "question_timestamps": "00:02:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Chicken steak rice only.", + "B": "Tomato beef noodles only.", + "C": "Chicken steak rice and tomato beef noodles.", + "D": "You bought chicken steak rice and then added Lanzhou lamian as the second item.", + "E": "You purchased tomato beef noodles along with a chicken burger dessert after entering the cafeteria." + }, + "answer_label": "C", + "answer_text": "Chicken steak rice and tomato beef noodles.", + "q_id": "01569_8", + "correct_answer": "Chicken steak rice and tomato beef noodles.", + "original_options": { + "A": "Chicken steak rice only.", + "B": "Tomato beef noodles only.", + "C": "Chicken steak rice and tomato beef noodles.", + "D": "You bought chicken steak rice and then added Lanzhou lamian as the second item.", + "E": "You purchased tomato beef noodles along with a chicken burger dessert after entering the cafeteria." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_8_T2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From the entrance on the 1st floor of the cafeteria, can I reach Lanzhou Lamian within 20 steps?", + "answer": "Yes.", + "question_chinese": "我从食堂1楼门口到兰州拉面能在20步之内到达吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "Lanzhou Ramen is at the innermost part of the cafeteria, far from the entrance, cannot be reached within 20 steps", + "timestamps": { + "start": "00:02:42", + "end": "00:02:43" + } + } + ], + "question_timestamps": "00:02:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "01569_8_T2", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_8_T1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From the entrance on the 1st floor of the cafeteria, can I reach Lanzhou Lamian within 20 steps?", + "answer": "No.", + "question_chinese": "我从食堂1楼门口到兰州拉面能在20步之内到达吗？", + "answer_chinese": "不", + "evidence": [ + { + "description": "Lanzhou Ramen is at the innermost part of the cafeteria, far from the entrance, cannot be reached within 20 steps", + "timestamps": { + "start": "00:02:42", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01569_8_T1", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_5", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which floor am I on now?", + "answer": "2nd floor.", + "question_chinese": "我现在在几楼？", + "answer_chinese": "2楼", + "evidence": [ + { + "description": "Went from the first floor to the second floor", + "timestamps": { + "start": "00:00:40", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "2nd floor.", + "B": "1st floor.", + "C": "3rd floor.", + "D": "You are on the 1st floor, near the cafeteria entrance area now.", + "E": "You are on the 4th floor, far above the level with the nearby food stalls." + }, + "answer_label": "A", + "answer_text": "2nd floor.", + "q_id": "01569_5", + "correct_answer": "2nd floor.", + "original_options": { + "A": "2nd floor.", + "B": "1st floor.", + "C": "3rd floor.", + "D": "You are on the 1st floor, near the cafeteria entrance area now.", + "E": "You are on the 4th floor, far above the level with the nearby food stalls." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 15 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_6", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which floor am I on now?", + "answer": "1st floor.", + "question_chinese": "我现在在几楼？", + "answer_chinese": "1楼", + "evidence": [ + { + "description": "Went from the second floor to the first floor", + "timestamps": { + "start": "00:01:46", + "end": "00:01:52" + } + } + ], + "question_timestamps": "00:02:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "2nd floor.", + "B": "Basement level.", + "C": "You are on the 3rd floor, one level above where you entered.", + "D": "1st floor.", + "E": "You are on the ground floor, not up on the first-floor dining level." + }, + "answer_label": "D", + "answer_text": "1st floor.", + "q_id": "01569_6", + "correct_answer": "1st floor.", + "original_options": { + "A": "2nd floor.", + "B": "Basement level.", + "C": "You are on the 3rd floor, one level above where you entered.", + "D": "1st floor.", + "E": "You are on the ground floor, not up on the first-floor dining level." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_6_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What route should I take right now to go buy pork rib rice?", + "answer": "Turn right first, then go through the door and keep walking straight ahead.", + "question_chinese": "我现在要去买排骨饭的路径是什么？", + "answer_chinese": "先右转再进门一直往前走", + "evidence": [ + { + "description": "Pork rib rice is on the left side after entering the door for a distance", + "timestamps": { + "start": "00:01:20", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left first.", + "B": "Turn right first, then go through the door and keep walking straight ahead.", + "C": "Go straight through the door.", + "D": "Turn left first, go through the door, then keep walking straight ahead.", + "E": "Turn right first, but don’t go through the door; keep walking straight outside instead." + }, + "answer_label": "B", + "answer_text": "Turn right first, then go through the door and keep walking straight ahead.", + "q_id": "01569_6_T2", + "correct_answer": "Turn right first, then go through the door and keep walking straight ahead.", + "original_options": { + "A": "Turn left first.", + "B": "Turn right first, then go through the door and keep walking straight ahead.", + "C": "Go straight through the door.", + "D": "Turn left first, go through the door, then keep walking straight ahead.", + "E": "Turn right first, but don’t go through the door; keep walking straight outside instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 5, + 12, + 15 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_6_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What route should I take right now to go buy pork rib rice?", + "answer": "Turn right first, then go through the door and keep walking straight ahead. The pork rib rice is on the left side.", + "question_chinese": "我现在要去买排骨饭的路径是什么？", + "answer_chinese": "先右转再进门一直往前走，排骨饭在左侧", + "evidence": [ + { + "description": "Pork rib rice is on the left side after entering the door for a distance", + "timestamps": { + "start": "00:01:20", + "end": "00:01:24" + } + }, + { + "description": "After buying the meal, I immediately turned around and left the cafeteria door", + "timestamps": { + "start": "00:01:32", + "end": "00:01:36" + } + } + ], + "question_timestamps": "00:01:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on your right side.", + "B": "It’s straight ahead.", + "C": "Turn right first, then go through the door and keep walking straight ahead. The pork rib rice is on the left side.", + "D": "Turn left first, then go through the door and keep walking straight ahead to find it.", + "E": "Turn right first, go through the door, and the pork rib rice will be on your right side." + }, + "answer_label": "C", + "answer_text": "Turn right first, then go through the door and keep walking straight ahead. The pork rib rice is on the left side.", + "q_id": "01569_6_T1", + "correct_answer": "Turn right first, then go through the door and keep walking straight ahead. The pork rib rice is on the left side.", + "original_options": { + "A": "It’s on your right side.", + "B": "It’s straight ahead.", + "C": "Turn right first, then go through the door and keep walking straight ahead. The pork rib rice is on the left side.", + "D": "Turn left first, then go through the door and keep walking straight ahead to find it.", + "E": "Turn right first, go through the door, and the pork rib rice will be on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 16, + 18 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_7_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get to Xiangyang Beef Noodles within 20 steps?", + "answer": "No.", + "question_chinese": "我能在20步之内到达襄阳牛肉面吗？", + "answer_chinese": "不", + "evidence": [ + { + "description": "The distance between Xiangyang beef noodles and the current Lanzhou beef noodles is more than twenty steps.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:16" + } + } + ], + "question_timestamps": "00:02:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "01569_7_T2", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_7_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get to Xiangyang Beef Noodles within 20 steps?", + "answer": "Yes.", + "question_chinese": "我能在20步之内到达襄阳牛肉面吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "Xiangyang Beef Noodles is not far from the entrance, can be reached within 20 steps", + "timestamps": { + "start": "00:02:44", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:02:47", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "01569_7_T1", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_3", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which side is the parking area on for me now?", + "answer": "Back side.", + "question_chinese": "现在停车区域在我的哪侧？", + "answer_chinese": "后方", + "evidence": [ + { + "description": "The parking area is on my left", + "timestamps": { + "start": "00:00:18", + "end": "00:00:22" + } + }, + { + "description": "I made a turn", + "timestamps": { + "start": "00:00:29", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Front side.", + "B": "Left side.", + "C": "It’s on your right side from where you’re standing now.", + "D": "Back side.", + "E": "The parking area is in front of you, not behind you at the moment." + }, + "answer_label": "D", + "answer_text": "Back side.", + "q_id": "01569_3", + "correct_answer": "Back side.", + "original_options": { + "A": "Front side.", + "B": "Left side.", + "C": "It’s on your right side from where you’re standing now.", + "D": "Back side.", + "E": "The parking area is in front of you, not behind you at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_4", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which side is the parking area on for me now?", + "answer": "Right side.", + "question_chinese": "现在停车区域在我的哪侧？", + "answer_chinese": "右侧", + "evidence": [ + { + "description": "The parking area is on my left", + "timestamps": { + "start": "00:00:18", + "end": "00:00:22" + } + }, + { + "description": "I made a turn", + "timestamps": { + "start": "00:00:29", + "end": "00:00:32" + } + }, + { + "description": "I went up the stairs and then turned left", + "timestamps": { + "start": "00:00:46", + "end": "00:00:53" + } + }, + { + "description": "Kept going up the stairs without turning", + "timestamps": { + "start": "00:01:03", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Left side.", + "B": "Right side.", + "C": "Straight ahead.", + "D": "The parking area is on your left as you face forward from where you are now.", + "E": "It’s behind you on the left, so you’d need to turn around to see the parking area." + }, + "answer_label": "B", + "answer_text": "Right side.", + "q_id": "01569_4", + "correct_answer": "Right side.", + "original_options": { + "A": "Left side.", + "B": "Right side.", + "C": "Straight ahead.", + "D": "The parking area is on your left as you face forward from where you are now.", + "E": "It’s behind you on the left, so you’d need to turn around to see the parking area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 19 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_15", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How should I go back to the starting point now?", + "answer": "Turn around, go down the stairs, turn right at the intersection, then turn left and walk forward to the next intersection, then turn right.", + "question_chinese": "现在我要回到出发的地方应该怎么走？", + "answer_chinese": "掉头下楼梯右转走到路口左转，再往前走到路口右转", + "evidence": [ + { + "description": "I turned left at the intersection", + "timestamps": { + "start": "00:00:10", + "end": "00:00:14" + } + }, + { + "description": "Walked straight for a distance, the cafeteria door is in front right of me", + "timestamps": { + "start": "00:00:22", + "end": "00:00:28" + } + }, + { + "description": "Went from the first floor to the second floor", + "timestamps": { + "start": "00:00:40", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left at the intersection.", + "B": "Skip the stairs and go straight.", + "C": "Turn around, go down the stairs, turn left at the intersection, then turn left and continue to the next intersection.", + "D": "Turn around and go down the stairs, then turn right, walk to the next intersection, and turn left.", + "E": "Turn around, go down the stairs, turn right at the intersection, then turn left and walk forward to the next intersection, then turn right." + }, + "answer_label": "E", + "answer_text": "Turn around, go down the stairs, turn right at the intersection, then turn left and walk forward to the next intersection, then turn right.", + "q_id": "01569_15", + "correct_answer": "Turn around, go down the stairs, turn right at the intersection, then turn left and walk forward to the next intersection, then turn right.", + "original_options": { + "A": "Turn left at the intersection.", + "B": "Skip the stairs and go straight.", + "C": "Turn around, go down the stairs, turn left at the intersection, then turn left and continue to the next intersection.", + "D": "Turn around and go down the stairs, then turn right, walk to the next intersection, and turn left.", + "E": "Turn around, go down the stairs, turn right at the intersection, then turn left and walk forward to the next intersection, then turn right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 18 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_16", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How should I go back to the starting point now?", + "answer": "Walk straight to the intersection and turn left, then walk forward to the next intersection and turn right.", + "question_chinese": "现在我要回到出发的地方应该怎么走？", + "answer_chinese": "直走到路口左转，再往前走到路口右转", + "evidence": [ + { + "description": "I turned left at the intersection", + "timestamps": { + "start": "00:00:10", + "end": "00:00:14" + } + }, + { + "description": "Walked straight for a distance, the cafeteria door is in front right of me", + "timestamps": { + "start": "00:00:22", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right at the first intersection.", + "B": "Turn left twice at intersections.", + "C": "Walk straight to the intersection and turn right, then walk forward to the next intersection and turn right.", + "D": "Walk straight to the intersection and turn left, then walk forward to the next intersection and turn right.", + "E": "Walk straight to the intersection and turn left, continue to the next intersection, then turn left again." + }, + "answer_label": "D", + "answer_text": "Walk straight to the intersection and turn left, then walk forward to the next intersection and turn right.", + "q_id": "01569_16", + "correct_answer": "Walk straight to the intersection and turn left, then walk forward to the next intersection and turn right.", + "original_options": { + "A": "Turn right at the first intersection.", + "B": "Turn left twice at intersections.", + "C": "Walk straight to the intersection and turn right, then walk forward to the next intersection and turn right.", + "D": "Walk straight to the intersection and turn left, then walk forward to the next intersection and turn right.", + "E": "Walk straight to the intersection and turn left, continue to the next intersection, then turn left again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 17 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many white cars have I encountered so far?", + "answer": "one.", + "question_chinese": "现在我遇到几辆白车？", + "answer_chinese": "1辆", + "evidence": [ + { + "description": "There is a white car ahead", + "timestamps": { + "start": "00:00:02", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "You have encountered two white cars so far during your route up to this point.", + "D": "You’ve seen three white cars so far, counting each one you passed along the way.", + "E": "one." + }, + "answer_label": "E", + "answer_text": "one.", + "q_id": "01569_1", + "correct_answer": "one.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "You have encountered two white cars so far during your route up to this point.", + "D": "You’ve seen three white cars so far, counting each one you passed along the way.", + "E": "one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "01569", + "video_length": "00:03:20", + "qid": "01569_2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many white cars have I encountered so far?", + "answer": "Six.", + "question_chinese": "现在我遇到几辆白车？", + "answer_chinese": "6辆", + "evidence": [ + { + "description": "There is a white car ahead", + "timestamps": { + "start": "00:00:02", + "end": "00:00:06" + } + }, + { + "description": "There is a white car on the left", + "timestamps": { + "start": "00:00:16", + "end": "00:00:20" + } + }, + { + "description": "There are four white cars on the left", + "timestamps": { + "start": "00:00:22", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:58", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "You have encountered five white cars so far during your route up to this point.", + "D": "You’ve seen seven white cars so far, counting each one you passed along the way.", + "E": "Six." + }, + "answer_label": "E", + "answer_text": "Six.", + "q_id": "01569_2", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "You have encountered five white cars so far during your route up to this point.", + "D": "You’ve seen seven white cars so far, counting each one you passed along the way.", + "E": "Six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + } +] diff --git a/QAs/04ac4c40-22fd-42aa-a7f0-ee597ffb7058-mcq.json b/QAs/04ac4c40-22fd-42aa-a7f0-ee597ffb7058-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..04fbc3f9f757eca0a40457bd2b11e577e5df568c --- /dev/null +++ b/QAs/04ac4c40-22fd-42aa-a7f0-ee597ffb7058-mcq.json @@ -0,0 +1,849 @@ +[ + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_9", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the kitchen now relative to me?", + "answer": "The kitchen is now on your right.", + "question_chinese": "厨房现在相对我是什么位置？", + "answer_chinese": "厨房现在在你的右侧。", + "evidence": [ + { + "description": "The kitchen is now on your right.", + "timestamps": { + "start": "00:11:05", + "end": "00:11:42" + } + } + ], + "question_timestamps": "00:14:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your left.", + "B": "Behind you.", + "C": "The kitchen is now on your right.", + "D": "The kitchen is now in front of you as you face forward from your current position.", + "E": "The kitchen is diagonally behind you to the left, so you’d need to turn to see it." + }, + "answer_label": "C", + "answer_text": "The kitchen is now on your right.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_9", + "correct_answer": "The kitchen is now on your right.", + "original_options": { + "A": "On your left.", + "B": "Behind you.", + "C": "The kitchen is now on your right.", + "D": "The kitchen is now in front of you as you face forward from your current position.", + "E": "The kitchen is diagonally behind you to the left, so you’d need to turn to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 16, + 18 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_10", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the kitchen now relative to me?", + "answer": "The kitchen is now in front of you.", + "question_chinese": "厨房现在相对我是什么位置？", + "answer_chinese": "厨房现在在你的前方。", + "evidence": [ + { + "description": "The kitchen is now in front of you.", + "timestamps": { + "start": "00:49:59", + "end": "00:50:06" + } + } + ], + "question_timestamps": "00:50:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your right.", + "B": "Behind you.", + "C": "The kitchen is now on your left as you face forward from where you are standing now.", + "D": "The kitchen is now in front of you.", + "E": "The kitchen is diagonally behind you to the right, so it isn’t directly in front of you." + }, + "answer_label": "D", + "answer_text": "The kitchen is now in front of you.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_10", + "correct_answer": "The kitchen is now in front of you.", + "original_options": { + "A": "On your right.", + "B": "Behind you.", + "C": "The kitchen is now on your left as you face forward from where you are standing now.", + "D": "The kitchen is now in front of you.", + "E": "The kitchen is diagonally behind you to the right, so it isn’t directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 17, + 18 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_17", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many eggs did I just crack?", + "answer": "You just cracked an egg.", + "question_chinese": "我刚刚打了几个蛋？", + "answer_chinese": "你刚刚打了一个蛋。", + "evidence": [ + { + "description": "You just cracked an egg.", + "timestamps": { + "start": "00:37:14", + "end": "00:37:21" + } + } + ], + "question_timestamps": "00:38:15", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You just cracked an egg.", + "B": "You just cracked two eggs.", + "C": "You just cracked three eggs.", + "D": "You just cracked four eggs, one after another, into the bowl in front of you.", + "E": "You just cracked five eggs in total before moving on to the next step." + }, + "answer_label": "A", + "answer_text": "You just cracked an egg.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_17", + "correct_answer": "You just cracked an egg.", + "original_options": { + "A": "You just cracked an egg.", + "B": "You just cracked two eggs.", + "C": "You just cracked three eggs.", + "D": "You just cracked four eggs, one after another, into the bowl in front of you.", + "E": "You just cracked five eggs in total before moving on to the next step." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_18", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many eggs did I just crack?", + "answer": "You just cracked three eggs.", + "question_chinese": "我刚刚打了几个蛋？", + "answer_chinese": "你刚刚打了三个蛋。", + "evidence": [ + { + "description": "You just cracked an egg.", + "timestamps": { + "start": "00:37:14", + "end": "00:37:21" + } + }, + { + "description": "You just laid the second egg.", + "timestamps": { + "start": "00:37:23", + "end": "00:37:28" + } + }, + { + "description": "You just laid the third egg.", + "timestamps": { + "start": "00:37:32", + "end": "00:37:38" + } + } + ], + "question_timestamps": "00:53:47", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You just cracked two eggs.", + "B": "You just cracked four eggs.", + "C": "You just cracked three eggs.", + "D": "You just cracked two eggs, cracking one less than the three you were aiming for.", + "E": "You just cracked four eggs, adding one extra egg beyond what you intended to crack." + }, + "answer_label": "C", + "answer_text": "You just cracked three eggs.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_18", + "correct_answer": "You just cracked three eggs.", + "original_options": { + "A": "You just cracked two eggs.", + "B": "You just cracked four eggs.", + "C": "You just cracked three eggs.", + "D": "You just cracked two eggs, cracking one less than the three you were aiming for.", + "E": "You just cracked four eggs, adding one extra egg beyond what you intended to crack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_10_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach and touch the sunglasses on the table?", + "answer": "You can reach the sunglasses to your front-right directly.", + "question_chinese": "我可以直接触碰到桌上的墨镜吗？", + "answer_chinese": "你可以直接触碰到，你右前方的墨镜。", + "evidence": [ + { + "description": "You can touch it directly.", + "timestamps": { + "start": "00:56:24", + "end": "00:56:26" + } + } + ], + "question_timestamps": "00:56:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the sunglasses to your front-right directly.", + "B": "No, you can’t reach the sunglasses directly; they’re too far in front-right." + }, + "answer_label": "A", + "answer_text": "You can reach the sunglasses to your front-right directly.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_10_T1", + "correct_answer": "You can reach the sunglasses to your front-right directly.", + "original_options": { + "A": "You can reach the sunglasses to your front-right directly.", + "B": "No, you can’t reach the sunglasses directly; they’re too far in front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_10_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach and touch the sunglasses on the table?", + "answer": "You can reach it directly; the sunglasses are by your right hand.", + "question_chinese": "我可以直接触碰到桌上的墨镜吗？", + "answer_chinese": "你可以直接触碰到，墨镜在你的右手边。", + "evidence": [ + { + "description": "You can touch it directly.", + "timestamps": { + "start": "00:56:24", + "end": "00:56:26" + } + }, + { + "description": "You can touch it directly.", + "timestamps": { + "start": "01:15:31", + "end": "01:15:52" + } + } + ], + "question_timestamps": "01:22:32", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can't reach it directly; the sunglasses are too far from your hands.", + "B": "You can reach it directly; the sunglasses are by your right hand." + }, + "answer_label": "B", + "answer_text": "You can reach it directly; the sunglasses are by your right hand.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_10_T2", + "correct_answer": "You can reach it directly; the sunglasses are by your right hand.", + "original_options": { + "A": "You can't reach it directly; the sunglasses are too far from your hands.", + "B": "You can reach it directly; the sunglasses are by your right hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_8_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "To get to the bedroom, how many steps do I need to go up?", + "answer": "You need to go past a staircase.", + "question_chinese": "我想要去到卧室需要经过几个楼梯？", + "answer_chinese": "你需要经过一个楼梯。", + "evidence": [ + { + "description": "You took a flight of stairs to reach this location from the bedroom.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:49" + } + } + ], + "question_timestamps": "00:29:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You need to go up two steps.", + "B": "You need to go upstairs first.", + "C": "You need to go past a staircase.", + "D": "You don’t need to go up any steps; the bedroom is on the same level as you.", + "E": "You need to go down a short set of steps before reaching the bedroom area." + }, + "answer_label": "C", + "answer_text": "You need to go past a staircase.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_8_T1", + "correct_answer": "You need to go past a staircase.", + "original_options": { + "A": "You need to go up two steps.", + "B": "You need to go upstairs first.", + "C": "You need to go past a staircase.", + "D": "You don’t need to go up any steps; the bedroom is on the same level as you.", + "E": "You need to go down a short set of steps before reaching the bedroom area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 15 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_8_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "To get to the bedroom, how many steps do I need to go up?", + "answer": "You need to go past a staircase in front of you.", + "question_chinese": "我想要去到卧室需要经过几个楼梯？", + "answer_chinese": "你需要经过一个楼梯，在你的前方。", + "evidence": [ + { + "description": "You took a flight of stairs to reach this location from the bedroom.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:49" + } + } + ], + "question_timestamps": "01:00:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Past staircase behind you.", + "B": "Past staircase to your left.", + "C": "You need to go past a staircase to your right before continuing toward the bedroom.", + "D": "You need to go past a staircase in front of you.", + "E": "You need to go down the staircase in front of you to reach the bedroom area." + }, + "answer_label": "D", + "answer_text": "You need to go past a staircase in front of you.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_8_T2", + "correct_answer": "You need to go past a staircase in front of you.", + "original_options": { + "A": "Past staircase behind you.", + "B": "Past staircase to your left.", + "C": "You need to go past a staircase to your right before continuing toward the bedroom.", + "D": "You need to go past a staircase in front of you.", + "E": "You need to go down the staircase in front of you to reach the bedroom area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 16 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_7", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the faucet directly now?", + "answer": "You can now touch the faucet directly.", + "question_chinese": "我现在可以直接触碰到水龙头吗？", + "answer_chinese": "你现在可以直接触碰到水龙头。", + "evidence": [ + { + "description": "You can now touch the faucet directly.", + "timestamps": { + "start": "00:06:02", + "end": "00:06:05" + } + } + ], + "question_timestamps": "00:06:06", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now touch the faucet directly.", + "B": "You cannot touch the faucet directly now." + }, + "answer_label": "A", + "answer_text": "You can now touch the faucet directly.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_7", + "correct_answer": "You can now touch the faucet directly.", + "original_options": { + "A": "You can now touch the faucet directly.", + "B": "You cannot touch the faucet directly now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_8", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the faucet directly now?", + "answer": "You can't touch the faucet directly now.", + "question_chinese": "我现在可以直接触碰到水龙头吗？", + "answer_chinese": "你现在无法直接触碰到水龙头。", + "evidence": [ + { + "description": "You can't touch the faucet directly now.", + "timestamps": { + "start": "00:47:42", + "end": "00:48:11" + } + } + ], + "question_timestamps": "00:48:37", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch the faucet directly now.", + "B": "You can't touch the faucet directly now." + }, + "answer_label": "B", + "answer_text": "You can't touch the faucet directly now.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_8", + "correct_answer": "You can't touch the faucet directly now.", + "original_options": { + "A": "You can touch the faucet directly now.", + "B": "You can't touch the faucet directly now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_3", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get to the computer desk if I want to use it now?", + "answer": "After you come out of the bedroom, turn right and go straight to reach your computer desk.", + "question_chinese": "我现在要使用电脑应该怎么去到电脑桌前？", + "answer_chinese": "你从卧室出来后向右转，直走即可到达电脑桌前。", + "evidence": [ + { + "description": "After you come out of the bedroom, turn right and go straight to reach your computer desk.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:01:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "After you come out of the bedroom, turn right and go straight to reach your computer desk.", + "B": "Exit the bedroom and turn left.", + "C": "Come out and go straight ahead.", + "D": "After leaving the bedroom, turn left and continue straight until you reach the computer desk area.", + "E": "Come out of the bedroom, go straight to the living room, then turn left to arrive at your computer desk." + }, + "answer_label": "A", + "answer_text": "After you come out of the bedroom, turn right and go straight to reach your computer desk.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_3", + "correct_answer": "After you come out of the bedroom, turn right and go straight to reach your computer desk.", + "original_options": { + "A": "After you come out of the bedroom, turn right and go straight to reach your computer desk.", + "B": "Exit the bedroom and turn left.", + "C": "Come out and go straight ahead.", + "D": "After leaving the bedroom, turn left and continue straight until you reach the computer desk area.", + "E": "Come out of the bedroom, go straight to the living room, then turn left to arrive at your computer desk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 20 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_4", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get to the computer desk if I want to use it now?", + "answer": "You need to go upstairs first. Walk along the corridor to the end and you will reach your computer desk.", + "question_chinese": "我现在要使用电脑应该怎么去到电脑桌前？", + "answer_chinese": "你要先上楼，沿着过道走到尽头即可到达电脑桌前。", + "evidence": [ + { + "description": "You went downstairs and reached the living room.", + "timestamps": { + "start": "00:01:38", + "end": "00:01:53" + } + }, + { + "description": "You walked along the corridor to the staircase.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:27:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go downstairs to the desk.", + "B": "Take the first door on the left.", + "C": "Stay on this floor, turn left into the hallway, and stop halfway down at the desk.", + "D": "Go upstairs, but turn into the first room on your right instead of walking to the corridor’s end.", + "E": "You need to go upstairs first. Walk along the corridor to the end and you will reach your computer desk." + }, + "answer_label": "E", + "answer_text": "You need to go upstairs first. Walk along the corridor to the end and you will reach your computer desk.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_4", + "correct_answer": "You need to go upstairs first. Walk along the corridor to the end and you will reach your computer desk.", + "original_options": { + "A": "Go downstairs to the desk.", + "B": "Take the first door on the left.", + "C": "Stay on this floor, turn left into the hallway, and stop halfway down at the desk.", + "D": "Go upstairs, but turn into the first room on your right instead of walking to the corridor’s end.", + "E": "You need to go upstairs first. Walk along the corridor to the end and you will reach your computer desk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 16, + 19 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_3_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the sofa relative to me right now?", + "answer": "The sofa is now behind you.", + "question_chinese": "沙发现在相对我是什么位置？", + "answer_chinese": "沙发现在在你的后方。", + "evidence": [ + { + "description": "The sofa is now behind you.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:54" + } + } + ], + "question_timestamps": "00:02:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sofa is in front of you.", + "B": "The sofa is to your left.", + "C": "The sofa is now behind you.", + "D": "The sofa is slightly to your right, close enough that you could sit down without turning around.", + "E": "The sofa is ahead of you on your left side, so you’d need to move forward to reach it." + }, + "answer_label": "C", + "answer_text": "The sofa is now behind you.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_3_T1", + "correct_answer": "The sofa is now behind you.", + "original_options": { + "A": "The sofa is in front of you.", + "B": "The sofa is to your left.", + "C": "The sofa is now behind you.", + "D": "The sofa is slightly to your right, close enough that you could sit down without turning around.", + "E": "The sofa is ahead of you on your left side, so you’d need to move forward to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 20 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_3_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the sofa relative to me right now?", + "answer": "The sofa is now to your back-right.", + "question_chinese": "沙发现在相对我是什么位置？", + "answer_chinese": "沙发现在在你的右后方。", + "evidence": [ + { + "description": "The sofa is now behind you.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:54" + } + }, + { + "description": "The sofa is now behind you.", + "timestamps": { + "start": "00:35:44", + "end": "00:35:53" + } + } + ], + "question_timestamps": "00:36:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sofa is to your back-left.", + "B": "The sofa is to your front-right.", + "C": "The sofa is directly behind you, not off to either side at all.", + "D": "The sofa is now to your back-right.", + "E": "The sofa is off your left shoulder in front of you, so you’d face it by turning left." + }, + "answer_label": "D", + "answer_text": "The sofa is now to your back-right.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_3_T2", + "correct_answer": "The sofa is now to your back-right.", + "original_options": { + "A": "The sofa is to your back-left.", + "B": "The sofa is to your front-right.", + "C": "The sofa is directly behind you, not off to either side at all.", + "D": "The sofa is now to your back-right.", + "E": "The sofa is off your left shoulder in front of you, so you’d face it by turning left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 19 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_7_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the balcony relative to me right now?", + "answer": "The balcony is now behind you.", + "question_chinese": "阳台现在相对我是什么位置？", + "answer_chinese": "阳台现在在你的身后。", + "evidence": [ + { + "description": "The balcony is now behind you.", + "timestamps": { + "start": "00:19:34", + "end": "00:19:39" + } + } + ], + "question_timestamps": "00:19:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The balcony is in front of you.", + "B": "The balcony is now behind you.", + "C": "The balcony is to your left.", + "D": "The balcony is to your right side at the moment, so you’d turn right to face it.", + "E": "The balcony is ahead of you rather than behind, so walking forward would take you toward it." + }, + "answer_label": "B", + "answer_text": "The balcony is now behind you.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_7_T1", + "correct_answer": "The balcony is now behind you.", + "original_options": { + "A": "The balcony is in front of you.", + "B": "The balcony is now behind you.", + "C": "The balcony is to your left.", + "D": "The balcony is to your right side at the moment, so you’d turn right to face it.", + "E": "The balcony is ahead of you rather than behind, so walking forward would take you toward it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058", + "video_length": "01:30:24", + "qid": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_7_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the balcony relative to me right now?", + "answer": "The balcony is now in front of you.", + "question_chinese": "阳台现在相对我是什么位置？", + "answer_chinese": "阳台现在在你的身前。", + "evidence": [ + { + "description": "The balcony is now behind you.", + "timestamps": { + "start": "00:28:11", + "end": "00:28:18" + } + } + ], + "question_timestamps": "00:35:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The balcony is behind you.", + "B": "The balcony is to your right.", + "C": "The balcony is now in front of you.", + "D": "The balcony is to your back-right, so it sits diagonally behind your right shoulder.", + "E": "The balcony is off to your left side instead of ahead, so you’d turn left to look at it." + }, + "answer_label": "C", + "answer_text": "The balcony is now in front of you.", + "q_id": "04ac4c40-22fd-42aa-a7f0-ee597ffb7058_7_T2", + "correct_answer": "The balcony is now in front of you.", + "original_options": { + "A": "The balcony is behind you.", + "B": "The balcony is to your right.", + "C": "The balcony is now in front of you.", + "D": "The balcony is to your back-right, so it sits diagonally behind your right shoulder.", + "E": "The balcony is off to your left side instead of ahead, so you’d turn left to look at it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 20 + ] + } +] diff --git a/QAs/04f0a415-0da1-4f19-a705-19ca52931fc0-mcq.json b/QAs/04f0a415-0da1-4f19-a705-19ca52931fc0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2421fb5fa72ee8718772ca8a62ac73720af2d677 --- /dev/null +++ b/QAs/04f0a415-0da1-4f19-a705-19ca52931fc0-mcq.json @@ -0,0 +1,1266 @@ +[ + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_4", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the first piece of paper I removed from the packaging placed?", + "answer": "The first piece of paper you tore off from the packaging was placed to the right of the backing board.", + "question_chinese": "我从包装上拆下来的第一块纸片被放到了哪里？", + "answer_chinese": "你从包装上拆下来的第一块纸片被放到了垫板右边。", + "evidence": [ + { + "description": "The piece of paper was placed to the right of the backing board.\nThe piece of paper was placed to the right of the backing board.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:06:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The first piece of paper you tore off from the packaging was placed to the right of the backing board.", + "B": "It was placed to the left.", + "C": "It was placed in front.", + "D": "The first paper you removed was set behind the backing board, not on its right-hand side.", + "E": "You placed the first torn paper on the left side of the backing board, away from the right edge." + }, + "answer_label": "A", + "answer_text": "The first piece of paper you tore off from the packaging was placed to the right of the backing board.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_4", + "correct_answer": "The first piece of paper you tore off from the packaging was placed to the right of the backing board.", + "original_options": { + "A": "The first piece of paper you tore off from the packaging was placed to the right of the backing board.", + "B": "It was placed to the left.", + "C": "It was placed in front.", + "D": "The first paper you removed was set behind the backing board, not on its right-hand side.", + "E": "You placed the first torn paper on the left side of the backing board, away from the right edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 19 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_2_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the first piece of paper that I tore off from the packaging?", + "answer": "The first piece of paper you tore off the packaging was placed to the right of the backing board.", + "question_chinese": "我从包装上拆下来的第一块纸片被放到了哪里？", + "answer_chinese": "你从包装上拆下来的第一块纸片被放到了垫板右边。", + "evidence": [ + { + "description": "The piece of paper was placed to the right of the backing board.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:12:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put it to the left.", + "B": "You put it behind the board.", + "C": "The first piece of paper you tore off the packaging was placed to the right of the backing board.", + "D": "You placed the first torn piece of paper in front of the backing board rather than on the right side.", + "E": "The first paper you tore off ended up on the left of the backing board, not on its right." + }, + "answer_label": "C", + "answer_text": "The first piece of paper you tore off the packaging was placed to the right of the backing board.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_2_T2", + "correct_answer": "The first piece of paper you tore off the packaging was placed to the right of the backing board.", + "original_options": { + "A": "You put it to the left.", + "B": "You put it behind the board.", + "C": "The first piece of paper you tore off the packaging was placed to the right of the backing board.", + "D": "You placed the first torn piece of paper in front of the backing board rather than on the right side.", + "E": "The first paper you tore off ended up on the left of the backing board, not on its right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 19 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_2_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the first piece of paper that I tore off from the packaging?", + "answer": "The first piece of paper you tore off the packaging was placed at the very bottom edge of the backing board.", + "question_chinese": "我从包装上拆下来的第一块纸片被放到了哪里？", + "answer_chinese": "你从包装上拆下来的第一块纸片被放到了垫板的最下边。", + "evidence": [ + { + "description": "The first piece of paper you removed from the packaging was placed at the bottom of the pad.", + "timestamps": { + "start": "00:12:48", + "end": "00:12:49" + } + } + ], + "question_timestamps": "00:12:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was placed at the top edge.", + "B": "It was placed on the left side.", + "C": "The first piece of paper you tore off the packaging was placed at the very bottom edge of the backing board.", + "D": "The first piece of paper you tore off was placed along the very top edge of the backing board.", + "E": "You put the first torn paper along the right edge of the backing board, not at the bottom edge." + }, + "answer_label": "C", + "answer_text": "The first piece of paper you tore off the packaging was placed at the very bottom edge of the backing board.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_2_T1", + "correct_answer": "The first piece of paper you tore off the packaging was placed at the very bottom edge of the backing board.", + "original_options": { + "A": "It was placed at the top edge.", + "B": "It was placed on the left side.", + "C": "The first piece of paper you tore off the packaging was placed at the very bottom edge of the backing board.", + "D": "The first piece of paper you tore off was placed along the very top edge of the backing board.", + "E": "You put the first torn paper along the right edge of the backing board, not at the bottom edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_3_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many wooden boards did I take out just now?", + "answer": "You just took out a square wooden board from the table in front of you.", + "question_chinese": "我刚才拿出来了几个木板？", + "answer_chinese": "你刚才从你面前的桌子上拿出来一个方形木板上。", + "evidence": [ + { + "description": "You took out the first plank.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:06" + } + } + ], + "question_timestamps": "00:09:34", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took out two wooden boards.", + "B": "You took out three wooden boards.", + "C": "You just took out four wooden boards from the table in front of you.", + "D": "You just took out a square wooden board from the table in front of you.", + "E": "You just took out five wooden boards from the table in front of you." + }, + "answer_label": "D", + "answer_text": "You just took out a square wooden board from the table in front of you.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_3_T1", + "correct_answer": "You just took out a square wooden board from the table in front of you.", + "original_options": { + "A": "You took out two wooden boards.", + "B": "You took out three wooden boards.", + "C": "You just took out four wooden boards from the table in front of you.", + "D": "You just took out a square wooden board from the table in front of you.", + "E": "You just took out five wooden boards from the table in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_3_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many wooden boards did I take out just now?", + "answer": "You took three wooden boards from the table in front of you.", + "question_chinese": "我刚才拿出来了几个木板？", + "answer_chinese": "你从面前的桌子上拿出了三个木板。", + "evidence": [ + { + "description": "You took out the first plank.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:06" + } + }, + { + "description": "You took out the second and third wooden boards.", + "timestamps": { + "start": "00:04:11", + "end": "00:04:13" + } + } + ], + "question_timestamps": "00:14:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took one wooden board.", + "B": "You took two wooden boards.", + "C": "You took four wooden boards from the table in front of you.", + "D": "You took five wooden boards from the table in front of you.", + "E": "You took three wooden boards from the table in front of you." + }, + "answer_label": "E", + "answer_text": "You took three wooden boards from the table in front of you.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_3_T2", + "correct_answer": "You took three wooden boards from the table in front of you.", + "original_options": { + "A": "You took one wooden board.", + "B": "You took two wooden boards.", + "C": "You took four wooden boards from the table in front of you.", + "D": "You took five wooden boards from the table in front of you.", + "E": "You took three wooden boards from the table in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_6_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did my blue paper take as it moved from below the instruction manual to above the backing board?", + "answer": "You picked up the blue paper in your hand.", + "question_chinese": "我的蓝色的纸从说明书下方到垫板上方，经历了什么路径？", + "answer_chinese": "你将蓝色的纸拿在了手上。", + "evidence": [ + { + "description": "You held the blue paper in your hand.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You slid the blue paper upward.", + "B": "You picked up the blue paper in your hand.", + "C": "You moved it above the wooden board.", + "D": "You moved the blue paper from below the manual to above the backing board without picking it up.", + "E": "You lifted the blue paper, passed it below the wooden board, and then placed it above the backing board." + }, + "answer_label": "B", + "answer_text": "You picked up the blue paper in your hand.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_6_T2", + "correct_answer": "You picked up the blue paper in your hand.", + "original_options": { + "A": "You slid the blue paper upward.", + "B": "You picked up the blue paper in your hand.", + "C": "You moved it above the wooden board.", + "D": "You moved the blue paper from below the manual to above the backing board without picking it up.", + "E": "You lifted the blue paper, passed it below the wooden board, and then placed it above the backing board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_6_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did my blue paper take as it moved from below the instruction manual to above the backing board?", + "answer": "The blue paper was moved from below the instruction manual, passed above the wooden board, and was placed above the backing board.", + "question_chinese": "我的蓝色的纸从说明书下方到垫板上方，经历了什么路径？", + "answer_chinese": "蓝色的纸被从说明书下方经过木板上方放到垫板上方。", + "evidence": [ + { + "description": "The blue paper was moved from underneath the manual, over the wooden board, and placed on top of the mat.", + "timestamps": { + "start": "00:18:29", + "end": "00:18:34" + } + } + ], + "question_timestamps": "00:23:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It went under the wooden board.", + "B": "It started above the instruction manual.", + "C": "The blue paper was moved from below the instruction manual, passed above the wooden board, and was placed above the backing board.", + "D": "The blue paper was moved from below the instruction manual, passed below the wooden board, and was placed above the backing board.", + "E": "The blue paper was moved from below the instruction manual, passed above the wooden board, and was placed above the fronting board." + }, + "answer_label": "C", + "answer_text": "The blue paper was moved from below the instruction manual, passed above the wooden board, and was placed above the backing board.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_6_T1", + "correct_answer": "The blue paper was moved from below the instruction manual, passed above the wooden board, and was placed above the backing board.", + "original_options": { + "A": "It went under the wooden board.", + "B": "It started above the instruction manual.", + "C": "The blue paper was moved from below the instruction manual, passed above the wooden board, and was placed above the backing board.", + "D": "The blue paper was moved from below the instruction manual, passed below the wooden board, and was placed above the backing board.", + "E": "The blue paper was moved from below the instruction manual, passed above the wooden board, and was placed above the fronting board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 22 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_1_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where on the table is my transparent tape?", + "answer": "The clear tape is directly below your hand.", + "question_chinese": "我的透明胶带在桌子上哪个位置？", + "answer_chinese": "透明胶带在你的手的正下方。", + "evidence": [ + { + "description": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:10:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The clear tape is in front.", + "B": "The clear tape is behind your hand.", + "C": "The clear tape is directly below your hand.", + "D": "The clear tape is directly to the left of your hand on the table.", + "E": "The clear tape is directly to the right of your hand on the table." + }, + "answer_label": "C", + "answer_text": "The clear tape is directly below your hand.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_1_T1", + "correct_answer": "The clear tape is directly below your hand.", + "original_options": { + "A": "The clear tape is in front.", + "B": "The clear tape is behind your hand.", + "C": "The clear tape is directly below your hand.", + "D": "The clear tape is directly to the left of your hand on the table.", + "E": "The clear tape is directly to the right of your hand on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 14 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_1_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where on the table is my transparent tape?", + "answer": "The clear tape is in front of you, at the lower-left of the tissue, and to the right of the tweezers.", + "question_chinese": "我的透明胶带在桌子上哪个位置？", + "answer_chinese": "透明胶带在你面前纸巾左下方，镊子右边。", + "evidence": [ + { + "description": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:20" + } + }, + { + "description": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "timestamps": { + "start": "00:16:40", + "end": "00:16:44" + } + } + ], + "question_timestamps": "00:22:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s behind you on the table.", + "B": "It’s in front, lower-right of the tissue.", + "C": "The clear tape is in front of you, at the lower-left of the tissue, and to the right of the tweezers.", + "D": "The clear tape is in front of you, at the upper-left of the tissue, and to the right of the tweezers.", + "E": "The clear tape is in front of you, at the lower-left of the tissue, and to the left of the tweezers." + }, + "answer_label": "C", + "answer_text": "The clear tape is in front of you, at the lower-left of the tissue, and to the right of the tweezers.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_1_T2", + "correct_answer": "The clear tape is in front of you, at the lower-left of the tissue, and to the right of the tweezers.", + "original_options": { + "A": "It’s behind you on the table.", + "B": "It’s in front, lower-right of the tissue.", + "C": "The clear tape is in front of you, at the lower-left of the tissue, and to the right of the tweezers.", + "D": "The clear tape is in front of you, at the upper-left of the tissue, and to the right of the tweezers.", + "E": "The clear tape is in front of you, at the lower-left of the tissue, and to the left of the tweezers." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 21, + 21 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_6", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many holes are there on the first square wooden board I took out?", + "answer": "The first square wooden board you took out has three holes.", + "question_chinese": "我第一块拿出来的方形木板上有几个孔？", + "answer_chinese": "你第一块拿出来的方形木板上有3个孔。", + "evidence": [ + { + "description": "The first square wooden board has three holes.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:25" + } + } + ], + "question_timestamps": "00:09:34", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "It has one hole.", + "B": "The first square wooden board you took out has three holes.", + "C": "It has two holes.", + "D": "The first square wooden board you took out has four holes on its face.", + "E": "The first square wooden board you took out has seven holes visible on the surface." + }, + "answer_label": "B", + "answer_text": "The first square wooden board you took out has three holes.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_6", + "correct_answer": "The first square wooden board you took out has three holes.", + "original_options": { + "A": "It has one hole.", + "B": "The first square wooden board you took out has three holes.", + "C": "It has two holes.", + "D": "The first square wooden board you took out has four holes on its face.", + "E": "The first square wooden board you took out has seven holes visible on the surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_5", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many holes are there on the first square wooden board I took out?", + "answer": "The first square wooden board you took out has three holes.", + "question_chinese": "我第一块拿出来的方形木板上有几个孔？", + "answer_chinese": "你第一块拿出来的方形木板上有3个孔。", + "evidence": [ + { + "description": "The first square wooden board has three holes.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:25" + } + } + ], + "question_timestamps": "00:14:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "The first square wooden board you took out has three holes.", + "B": "It has one hole.", + "C": "It has two holes.", + "D": "The first square wooden board you took out has four holes on its face.", + "E": "The first square wooden board you took out has seven holes visible on the surface." + }, + "answer_label": "A", + "answer_text": "The first square wooden board you took out has three holes.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_5", + "correct_answer": "The first square wooden board you took out has three holes.", + "original_options": { + "A": "The first square wooden board you took out has three holes.", + "B": "It has one hole.", + "C": "It has two holes.", + "D": "The first square wooden board you took out has four holes on its face.", + "E": "The first square wooden board you took out has seven holes visible on the surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_7_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "The first time I used the ruler, what path did it take from being to the right of the hammer to being placed on top of the scissors?", + "answer": "When you used the ruler, you picked it up from the right of the hammer and moved it above the wooden board.", + "question_chinese": "我第一次使用尺子时，尺子被从锤子右边到被放到剪刀上经历了什么路径？", + "answer_chinese": "你使用尺子时，尺子被从锤子右边拿起来，经过木板上方。", + "evidence": [ + { + "description": "It was picked up from the right side of the hammer, passed over the wooden board.", + "timestamps": { + "start": "00:08:16", + "end": "00:08:24" + } + } + ], + "question_timestamps": "00:08:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved it below the board.", + "B": "You lifted it from the left.", + "C": "You picked it up right of the hammer and slid it under the wooden board before setting it down.", + "D": "You grabbed it from the hammer’s right side, moved it above the board, then set it beside the scissors.", + "E": "When you used the ruler, you picked it up from the right of the hammer and moved it above the wooden board." + }, + "answer_label": "E", + "answer_text": "When you used the ruler, you picked it up from the right of the hammer and moved it above the wooden board.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_7_T1", + "correct_answer": "When you used the ruler, you picked it up from the right of the hammer and moved it above the wooden board.", + "original_options": { + "A": "You moved it below the board.", + "B": "You lifted it from the left.", + "C": "You picked it up right of the hammer and slid it under the wooden board before setting it down.", + "D": "You grabbed it from the hammer’s right side, moved it above the board, then set it beside the scissors.", + "E": "When you used the ruler, you picked it up from the right of the hammer and moved it above the wooden board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 20 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_13", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "When I used the ruler for the first time, what path did the ruler take from the right side of the hammer to being placed on the scissors?", + "answer": "The first time you used the ruler, it was picked up from the right side of the hammer, passed over the wooden board, and placed on the scissors after measuring.", + "question_chinese": "我第一次使用尺子时，尺子被从锤子右边到被放到剪刀上经历了什么路径？", + "answer_chinese": "你第一次使用尺子时，尺子被从锤子右边拿起来，经过木板上方，量完东西后被放到剪刀上。", + "evidence": [ + { + "description": "It was picked up from the right side of the hammer, passed over the wooden board.", + "timestamps": { + "start": "00:08:16", + "end": "00:08:24" + } + }, + { + "description": "The ruler was placed on the scissors after measuring.", + "timestamps": { + "start": "00:08:44", + "end": "00:08:49" + } + } + ], + "question_timestamps": "00:12:52", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It passed under the wooden board.", + "B": "The first time you used the ruler, it was picked up from the right side of the hammer, passed over the wooden board, and placed on the scissors after measuring.", + "C": "It started from the hammer’s left side.", + "D": "You picked it up from the hammer’s right side, carried it around the wooden board, and left it beside the scissors.", + "E": "The ruler was taken from the right of the hammer, moved over the wooden board, and placed under the scissors." + }, + "answer_label": "B", + "answer_text": "The first time you used the ruler, it was picked up from the right side of the hammer, passed over the wooden board, and placed on the scissors after measuring.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_13", + "correct_answer": "The first time you used the ruler, it was picked up from the right side of the hammer, passed over the wooden board, and placed on the scissors after measuring.", + "original_options": { + "A": "It passed under the wooden board.", + "B": "The first time you used the ruler, it was picked up from the right side of the hammer, passed over the wooden board, and placed on the scissors after measuring.", + "C": "It started from the hammer’s left side.", + "D": "You picked it up from the hammer’s right side, carried it around the wooden board, and left it beside the scissors.", + "E": "The ruler was taken from the right of the hammer, moved over the wooden board, and placed under the scissors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 22, + 20 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_7_T2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "The first time I used the ruler, what path did it take from being to the right of the hammer to being placed on top of the scissors?", + "answer": "The first time you used the ruler, you picked it up from the right of the hammer, moved it above the wooden board, and after measuring, you put it on top of the scissors.", + "question_chinese": "我第一次使用尺子时，尺子被从锤子右边到被放到剪刀上经历了什么路径？", + "answer_chinese": "你第一次使用尺子时，尺子被从锤子右边拿起来，经过木板上方，量完东西后被放到剪刀上。", + "evidence": [ + { + "description": "It was picked up from the right side of the hammer, passed over the wooden board.", + "timestamps": { + "start": "00:08:16", + "end": "00:08:24" + } + }, + { + "description": "The ruler was placed on the scissors after measure things.", + "timestamps": { + "start": "00:08:44", + "end": "00:08:49" + } + } + ], + "question_timestamps": "00:21:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Moved it below the wooden board.", + "B": "Picked it up left of hammer.", + "C": "The first time you used the ruler, you picked it up from the right of the hammer, moved it above the wooden board, and after measuring, you put it on top of the scissors.", + "D": "Picked it up from the right of the hammer, moved it below the wooden board, then placed it on scissors.", + "E": "Picked it up from the left of the hammer, moved it above the wooden board, then placed it on scissors." + }, + "answer_label": "C", + "answer_text": "The first time you used the ruler, you picked it up from the right of the hammer, moved it above the wooden board, and after measuring, you put it on top of the scissors.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_7_T2", + "correct_answer": "The first time you used the ruler, you picked it up from the right of the hammer, moved it above the wooden board, and after measuring, you put it on top of the scissors.", + "original_options": { + "A": "Moved it below the wooden board.", + "B": "Picked it up left of hammer.", + "C": "The first time you used the ruler, you picked it up from the right of the hammer, moved it above the wooden board, and after measuring, you put it on top of the scissors.", + "D": "Picked it up from the right of the hammer, moved it below the wooden board, then placed it on scissors.", + "E": "Picked it up from the left of the hammer, moved it above the wooden board, then placed it on scissors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 20 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_8_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many wooden boards have I taken out of the plastic packaging?", + "answer": "So far, you haven’t taken any wooden boards out of the plastic packaging.", + "question_chinese": "目前为止，我从塑料包装里拿出来几块木板？", + "answer_chinese": "目前为止，你还没有从塑料包装里拿出来木板。", + "evidence": [ + { + "description": "You only took out the instruction manual from your pocket.", + "timestamps": { + "start": "00:00:29", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:04:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One wooden board.", + "B": "Two wooden boards.", + "C": "So far, you’ve taken one wooden board out of the plastic packaging and left the rest inside.", + "D": "So far, you’ve taken two wooden boards out of the plastic packaging, but the remaining boards are still sealed.", + "E": "So far, you haven’t taken any wooden boards out of the plastic packaging." + }, + "answer_label": "E", + "answer_text": "So far, you haven’t taken any wooden boards out of the plastic packaging.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_8_T2", + "correct_answer": "So far, you haven’t taken any wooden boards out of the plastic packaging.", + "original_options": { + "A": "One wooden board.", + "B": "Two wooden boards.", + "C": "So far, you’ve taken one wooden board out of the plastic packaging and left the rest inside.", + "D": "So far, you’ve taken two wooden boards out of the plastic packaging, but the remaining boards are still sealed.", + "E": "So far, you haven’t taken any wooden boards out of the plastic packaging." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 20 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_8_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many wooden boards have I taken out of the plastic packaging?", + "answer": "So far, you’ve taken two wooden boards out of the plastic packaging.", + "question_chinese": "目前为止，我从塑料包装里拿出来几块木板？", + "answer_chinese": "目前为止，你从塑料包装里拿出来两块木板。", + "evidence": [ + { + "description": "I have taken two wooden boards out of the plastic packaging.", + "timestamps": { + "start": "00:04:08", + "end": "00:04:13" + } + } + ], + "question_timestamps": "00:15:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero wooden boards.", + "B": "One wooden board.", + "C": "So far, you’ve taken three wooden boards out of the plastic packaging and kept the rest in the wrap.", + "D": "So far, you’ve taken two wooden boards out of the plastic packaging.", + "E": "So far, you’ve taken four wooden boards out of the plastic packaging, leaving none remaining inside." + }, + "answer_label": "D", + "answer_text": "So far, you’ve taken two wooden boards out of the plastic packaging.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_8_T1", + "correct_answer": "So far, you’ve taken two wooden boards out of the plastic packaging.", + "original_options": { + "A": "Zero wooden boards.", + "B": "One wooden board.", + "C": "So far, you’ve taken three wooden boards out of the plastic packaging and kept the rest in the wrap.", + "D": "So far, you’ve taken two wooden boards out of the plastic packaging.", + "E": "So far, you’ve taken four wooden boards out of the plastic packaging, leaving none remaining inside." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 20, + 17 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_10", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many wooden boards have I glued?", + "answer": "So far, you have applied glue to two wooden boards.", + "question_chinese": "目前为止，我给几块木板涂了胶水？", + "answer_chinese": "目前为止，你给2块木板涂了胶水。", + "evidence": [ + { + "description": "I applied glue to the first wooden board.", + "timestamps": { + "start": "00:14:13", + "end": "00:14:18" + } + }, + { + "description": "I applied glue to the second wooden board.", + "timestamps": { + "start": "00:14:27", + "end": "00:14:33" + } + } + ], + "question_timestamps": "00:14:35", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One wooden board.", + "B": "Three wooden boards.", + "C": "So far, you have applied glue to four wooden boards, spreading adhesive across each piece.", + "D": "So far, you have applied glue to five wooden boards, coating them before assembling anything.", + "E": "So far, you have applied glue to two wooden boards." + }, + "answer_label": "E", + "answer_text": "So far, you have applied glue to two wooden boards.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_10", + "correct_answer": "So far, you have applied glue to two wooden boards.", + "original_options": { + "A": "One wooden board.", + "B": "Three wooden boards.", + "C": "So far, you have applied glue to four wooden boards, spreading adhesive across each piece.", + "D": "So far, you have applied glue to five wooden boards, coating them before assembling anything.", + "E": "So far, you have applied glue to two wooden boards." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_9", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many wooden boards have I glued?", + "answer": "So far, you have applied glue to three wooden boards.", + "question_chinese": "目前为止，我给几块木板涂了胶水？", + "answer_chinese": "目前为止，你给3块木板涂了胶水。", + "evidence": [ + { + "description": "I applied glue to the first wooden board.", + "timestamps": { + "start": "00:14:13", + "end": "00:14:18" + } + }, + { + "description": "I applied glue to the second wooden board.", + "timestamps": { + "start": "00:14:27", + "end": "00:14:33" + } + }, + { + "description": "I applied glue to third wooden board.", + "timestamps": { + "start": "00:14:44", + "end": "00:14:48" + } + } + ], + "question_timestamps": "00:14:50", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One wooden board.", + "B": "Two wooden boards.", + "C": "So far, you have applied glue to four wooden boards, adding adhesive to another piece as well.", + "D": "So far, you have applied glue to five wooden boards, covering all of the boards you handled.", + "E": "So far, you have applied glue to three wooden boards." + }, + "answer_label": "E", + "answer_text": "So far, you have applied glue to three wooden boards.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_9", + "correct_answer": "So far, you have applied glue to three wooden boards.", + "original_options": { + "A": "One wooden board.", + "B": "Two wooden boards.", + "C": "So far, you have applied glue to four wooden boards, adding adhesive to another piece as well.", + "D": "So far, you have applied glue to five wooden boards, covering all of the boards you handled.", + "E": "So far, you have applied glue to three wooden boards." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_4_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in my room is the bag?", + "answer": "The clear plastic bag is at the top-right corner of the table in front of you.", + "question_chinese": "袋子在我房间里哪个位置？", + "answer_chinese": "透明袋子在你面前的桌子右上角。", + "evidence": [ + { + "description": "The transparent bag is on the upper right corner of the table.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:03" + } + } + ], + "question_timestamps": "00:09:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The clear plastic bag is at the top-right corner of the table in front of you.", + "B": "At the top-left of the table.", + "C": "At the bottom-right of the table.", + "D": "The clear plastic bag is on the table in front of you, but positioned at the bottom-left corner.", + "E": "The clear plastic bag is on the table in front of you, sitting near the center rather than the corner." + }, + "answer_label": "A", + "answer_text": "The clear plastic bag is at the top-right corner of the table in front of you.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_4_T2", + "correct_answer": "The clear plastic bag is at the top-right corner of the table in front of you.", + "original_options": { + "A": "The clear plastic bag is at the top-right corner of the table in front of you.", + "B": "At the top-left of the table.", + "C": "At the bottom-right of the table.", + "D": "The clear plastic bag is on the table in front of you, but positioned at the bottom-left corner.", + "E": "The clear plastic bag is on the table in front of you, sitting near the center rather than the corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 20 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_4_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in my room is the bag?", + "answer": "The bag is on the floor at the bottom-right corner of the table in front of you.", + "question_chinese": "袋子在我房间里哪个位置？", + "answer_chinese": "袋子在你面前桌子右下角地面上。", + "evidence": [ + { + "description": "The bag is on the floor at the lower right corner of the table.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:31" + } + }, + { + "description": "The bag is on the floor at the lower right corner of the table.", + "timestamps": { + "start": "00:09:34", + "end": "00:09:35" + } + } + ], + "question_timestamps": "00:13:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the floor under the table.", + "B": "On the floor at the table’s front-left.", + "C": "The bag is on the floor at the bottom-left corner of the table in front of you.", + "D": "The bag is on the floor at the bottom-right corner of the table behind you, near the table leg.", + "E": "The bag is on the floor at the bottom-right corner of the table in front of you." + }, + "answer_label": "E", + "answer_text": "The bag is on the floor at the bottom-right corner of the table in front of you.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_4_T1", + "correct_answer": "The bag is on the floor at the bottom-right corner of the table in front of you.", + "original_options": { + "A": "On the floor under the table.", + "B": "On the floor at the table’s front-left.", + "C": "The bag is on the floor at the bottom-left corner of the table in front of you.", + "D": "The bag is on the floor at the bottom-right corner of the table behind you, near the table leg.", + "E": "The bag is on the floor at the bottom-right corner of the table in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 17, + 19 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the transparent tape on the table?", + "answer": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "question_chinese": "透明胶带在桌子上哪个位置？", + "answer_chinese": "透明胶带在纸巾左下方，镊子右边。", + "evidence": [ + { + "description": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:20" + } + }, + { + "description": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "timestamps": { + "start": "00:16:40", + "end": "00:16:44" + } + } + ], + "question_timestamps": "00:20:48", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "B": "Above the tissue’s bottom-left corner.", + "C": "Right of the tweezers.", + "D": "The transparent tape is below the bottom right of the tissue, positioned to the left of the tweezers.", + "E": "The transparent tape sits below the bottom left of the tissue, but it is placed to the right of the tweezers." + }, + "answer_label": "A", + "answer_text": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_1", + "correct_answer": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "original_options": { + "A": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "B": "Above the tissue’s bottom-left corner.", + "C": "Right of the tweezers.", + "D": "The transparent tape is below the bottom right of the tissue, positioned to the left of the tweezers.", + "E": "The transparent tape sits below the bottom left of the tissue, but it is placed to the right of the tweezers." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 18, + 21 + ] + }, + { + "video_uid": "04f0a415-0da1-4f19-a705-19ca52931fc0", + "video_length": "00:25:01", + "qid": "04f0a415-0da1-4f19-a705-19ca52931fc0_2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the transparent tape on the table?", + "answer": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "question_chinese": "透明胶带在桌子上哪个位置？", + "answer_chinese": "透明胶带在纸巾左下方，镊子右边。", + "evidence": [ + { + "description": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:20" + } + }, + { + "description": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "timestamps": { + "start": "00:16:40", + "end": "00:16:44" + } + } + ], + "question_timestamps": "00:22:54", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Above the tissue’s top right.", + "B": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "C": "Beneath the tissue’s bottom right.", + "D": "The transparent tape is beside the bottom left of the notebook, positioned to the left of the scissors.", + "E": "The transparent tape is behind the middle left of the tissue, placed to the left of the ruler." + }, + "answer_label": "B", + "answer_text": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "q_id": "04f0a415-0da1-4f19-a705-19ca52931fc0_2", + "correct_answer": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "original_options": { + "A": "Above the tissue’s top right.", + "B": "The transparent tape is below the bottom left of the tissue, to the left of the tweezers.", + "C": "Beneath the tissue’s bottom right.", + "D": "The transparent tape is beside the bottom left of the notebook, positioned to the left of the scissors.", + "E": "The transparent tape is behind the middle left of the tissue, placed to the left of the ruler." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + } +] diff --git a/QAs/05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657-mcq.json b/QAs/05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a2ad5bfd2ef84c529298ffa58c2ca73202759e4d --- /dev/null +++ b/QAs/05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657-mcq.json @@ -0,0 +1,1527 @@ +[ + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_14", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many kinds of fruit did I buy at the fruit store before?", + "answer": "You bought one kind of fruit.", + "question_chinese": "我之前在水果店买了几种水果？", + "answer_chinese": "买了一种水果。", + "evidence": [ + { + "description": "Paid for a serving of bananas at the fruit store.", + "timestamps": { + "start": "00:23:54", + "end": "00:24:12" + } + } + ], + "question_timestamps": "00:26:32", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You bought two kinds of fruit.", + "B": "You bought one kind of fruit.", + "C": "You bought three kinds of fruit.", + "D": "You bought four kinds of fruit at the fruit store before, not just a single kind.", + "E": "You bought five kinds of fruit at the fruit store before, so there were multiple varieties." + }, + "answer_label": "B", + "answer_text": "You bought one kind of fruit.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_14", + "correct_answer": "You bought one kind of fruit.", + "original_options": { + "A": "You bought two kinds of fruit.", + "B": "You bought one kind of fruit.", + "C": "You bought three kinds of fruit.", + "D": "You bought four kinds of fruit at the fruit store before, not just a single kind.", + "E": "You bought five kinds of fruit at the fruit store before, so there were multiple varieties." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_13", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many kinds of fruit did I buy at the fruit store before?", + "answer": "You bought one kind of fruit.", + "question_chinese": "我之前在水果店买了几种水果？", + "answer_chinese": "买了一种水果。", + "evidence": [ + { + "description": "Paid for a serving of bananas at the fruit store.", + "timestamps": { + "start": "00:23:54", + "end": "00:24:12" + } + } + ], + "question_timestamps": "00:36:18", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You bought two kinds of fruit.", + "B": "You bought three kinds of fruit.", + "C": "You bought four kinds of fruit at the fruit store before, meaning more than one variety.", + "D": "You bought one kind of fruit.", + "E": "You bought five kinds of fruit at the fruit store before, so it wasn’t limited to one kind." + }, + "answer_label": "D", + "answer_text": "You bought one kind of fruit.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_13", + "correct_answer": "You bought one kind of fruit.", + "original_options": { + "A": "You bought two kinds of fruit.", + "B": "You bought three kinds of fruit.", + "C": "You bought four kinds of fruit at the fruit store before, meaning more than one variety.", + "D": "You bought one kind of fruit.", + "E": "You bought five kinds of fruit at the fruit store before, so it wasn’t limited to one kind." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 19 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from finishing the pizza to throwing the trash in the bin?", + "answer": "You first got up from the dining table, walked straight to the front left, came to the front of the trash can, threw the trash into the trash can, and continued to walk to the right.", + "question_chinese": "我从吃完披萨到把垃圾扔进垃圾桶，经过了什么路径？", + "answer_chinese": "你首先从餐桌起身，向左前方直走，来到垃圾桶前面，把垃圾扔进垃圾桶，继续向右走。", + "evidence": [ + { + "description": "Finished eating pizza, got up to clean up the trash.", + "timestamps": { + "start": "00:16:53", + "end": "00:18:01" + } + }, + { + "description": "Walk straight to the front left to the front of the trash can.", + "timestamps": { + "start": "00:18:04", + "end": "00:18:14" + } + }, + { + "description": "Throw the trash into the trash can and continue to the right.", + "timestamps": { + "start": "00:18:15", + "end": "00:18:18" + } + } + ], + "question_timestamps": "00:18:21", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You walked to the back right.", + "B": "You approached the bin from the side.", + "C": "You got up, walked to the front right, reached the trash can, threw the trash in, then walked left.", + "D": "You first got up from the dining table, walked straight to the front left, came to the front of the trash can, threw the trash into the trash can, and continued to walk to the right.", + "E": "You stood up, walked to the front left, stopped at the side of the trash can, then walked left." + }, + "answer_label": "D", + "answer_text": "You first got up from the dining table, walked straight to the front left, came to the front of the trash can, threw the trash into the trash can, and continued to walk to the right.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_1", + "correct_answer": "You first got up from the dining table, walked straight to the front left, came to the front of the trash can, threw the trash into the trash can, and continued to walk to the right.", + "original_options": { + "A": "You walked to the back right.", + "B": "You approached the bin from the side.", + "C": "You got up, walked to the front right, reached the trash can, threw the trash in, then walked left.", + "D": "You first got up from the dining table, walked straight to the front left, came to the front of the trash can, threw the trash into the trash can, and continued to walk to the right.", + "E": "You stood up, walked to the front left, stopped at the side of the trash can, then walked left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 19 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from finishing the pizza to throwing the trash in the bin?", + "answer": "You first got up from the dining table, walked straight to the front left, came to the front of the trash can, threw the trash into the trash can, and continued to walk to the right.", + "question_chinese": "我从吃完披萨到把垃圾扔进垃圾桶，经过了什么路径？", + "answer_chinese": "你首先从餐桌起身，向左前方直走，来到垃圾桶前面，把垃圾扔进垃圾桶，继续向右走。", + "evidence": [ + { + "description": "Finished eating pizza, got up to clean up the trash.", + "timestamps": { + "start": "00:16:53", + "end": "00:18:01" + } + }, + { + "description": "Walk straight to the front left to the front of the trash can.", + "timestamps": { + "start": "00:18:04", + "end": "00:18:14" + } + }, + { + "description": "Throw the trash into the trash can and continue to the right.", + "timestamps": { + "start": "00:18:15", + "end": "00:18:18" + } + } + ], + "question_timestamps": "00:19:20", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You walked diagonally to the back right.", + "B": "You threw it away in the compost bin.", + "C": "You first got up from the dining table, walked straight to the front left, came to the front of the trash can, threw the trash into the trash can, and continued to walk to the right.", + "D": "You stood up from the dining table, moved to the back left, threw the trash into the bin, then walked right.", + "E": "You left your seat, walked to the front left, tossed the trash in from the side, and then walked left." + }, + "answer_label": "C", + "answer_text": "You first got up from the dining table, walked straight to the front left, came to the front of the trash can, threw the trash into the trash can, and continued to walk to the right.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_2", + "correct_answer": "You first got up from the dining table, walked straight to the front left, came to the front of the trash can, threw the trash into the trash can, and continued to walk to the right.", + "original_options": { + "A": "You walked diagonally to the back right.", + "B": "You threw it away in the compost bin.", + "C": "You first got up from the dining table, walked straight to the front left, came to the front of the trash can, threw the trash into the trash can, and continued to walk to the right.", + "D": "You stood up from the dining table, moved to the back left, threw the trash into the bin, then walked right.", + "E": "You left your seat, walked to the front left, tossed the trash in from the side, and then walked left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 21, + 20 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from throwing away the trash to getting in the car?", + "answer": "After you threw away the trash, you walked straight to the right, came to a crossroad, continued straight until you reached the exit, walked from the exit to the side of the parked car, and finally got in the car.", + "question_chinese": "我从扔完垃圾到上车，经过了什么路径？", + "answer_chinese": "你扔完垃圾后向右直走，来到一个十字路口，继续直走，直到走到出口，从出口走到停车的车旁，最后上车。", + "evidence": [ + { + "description": "After throwing away the trash, continue to walk straight to the right to a crossroad.", + "timestamps": { + "start": "00:18:15", + "end": "00:18:35" + } + }, + { + "description": "Continue straight until you reach the exit.", + "timestamps": { + "start": "00:18:35", + "end": "00:19:09" + } + }, + { + "description": "Walk from the exit to the parked car and get in.", + "timestamps": { + "start": "00:19:10", + "end": "00:19:19" + } + } + ], + "question_timestamps": "00:19:24", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "After you threw away the trash, you walked straight to the right, came to a crossroad, continued straight until you reached the exit, walked from the exit to the side of the parked car, and finally got in the car.", + "B": "You turned left at the crossroad.", + "C": "You walked to the car’s front first.", + "D": "After throwing away the trash, you walked right to the junction, turned right, reached the gate, and got in.", + "E": "After you threw away the trash, you went straight to the exit, walked to the back of the car, and got in." + }, + "answer_label": "A", + "answer_text": "After you threw away the trash, you walked straight to the right, came to a crossroad, continued straight until you reached the exit, walked from the exit to the side of the parked car, and finally got in the car.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_3", + "correct_answer": "After you threw away the trash, you walked straight to the right, came to a crossroad, continued straight until you reached the exit, walked from the exit to the side of the parked car, and finally got in the car.", + "original_options": { + "A": "After you threw away the trash, you walked straight to the right, came to a crossroad, continued straight until you reached the exit, walked from the exit to the side of the parked car, and finally got in the car.", + "B": "You turned left at the crossroad.", + "C": "You walked to the car’s front first.", + "D": "After throwing away the trash, you walked right to the junction, turned right, reached the gate, and got in.", + "E": "After you threw away the trash, you went straight to the exit, walked to the back of the car, and got in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 19, + 22 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_4", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from throwing away the trash to getting in the car?", + "answer": "After you threw away the trash, you walked straight to the right, came to a crossroad, continued straight until you reached the exit, walked from the exit to the side of the parked car, and finally got in the car.", + "question_chinese": "我从扔完垃圾到上车，经过了什么路径？", + "answer_chinese": "你扔完垃圾后向右直走，来到一个十字路口，继续直走，直到走到出口，从出口走到停车的车旁，最后上车。", + "evidence": [ + { + "description": "After throwing away the trash, continue to walk straight to the right to a crossroad.", + "timestamps": { + "start": "00:18:15", + "end": "00:18:35" + } + }, + { + "description": "Continue straight until you reach the exit.", + "timestamps": { + "start": "00:18:35", + "end": "00:19:09" + } + }, + { + "description": "Walk from the exit to the parked car and get in.", + "timestamps": { + "start": "00:19:10", + "end": "00:19:19" + } + } + ], + "question_timestamps": "00:21:33", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turned left and headed to the exit.", + "B": "After you threw away the trash, you walked straight to the right, came to a crossroad, continued straight until you reached the exit, walked from the exit to the side of the parked car, and finally got in the car.", + "C": "You went right, then took a left turn.", + "D": "After tossing the trash, you walked right to the crossroad, turned left there, and entered the car.", + "E": "After you threw the trash away, you went straight right, reached the exit sooner, and got into the car from the front." + }, + "answer_label": "B", + "answer_text": "After you threw away the trash, you walked straight to the right, came to a crossroad, continued straight until you reached the exit, walked from the exit to the side of the parked car, and finally got in the car.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_4", + "correct_answer": "After you threw away the trash, you walked straight to the right, came to a crossroad, continued straight until you reached the exit, walked from the exit to the side of the parked car, and finally got in the car.", + "original_options": { + "A": "You turned left and headed to the exit.", + "B": "After you threw away the trash, you walked straight to the right, came to a crossroad, continued straight until you reached the exit, walked from the exit to the side of the parked car, and finally got in the car.", + "C": "You went right, then took a left turn.", + "D": "After tossing the trash, you walked right to the crossroad, turned left there, and entered the car.", + "E": "After you threw the trash away, you went straight right, reached the exit sooner, and got into the car from the front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 22 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many servings of raw meat did I just take in total?", + "answer": "You took two servings of raw meat in total.", + "question_chinese": "我刚才一共拿了几份生肉？", + "answer_chinese": "一共拿了两份生肉。", + "evidence": [ + { + "description": "Put a piece of meat in the shopping cart.", + "timestamps": { + "start": "00:33:07", + "end": "00:33:16" + } + }, + { + "description": "Put a piece of shredded meat in the shopping cart.", + "timestamps": { + "start": "00:33:32", + "end": "00:34:31" + } + } + ], + "question_timestamps": "00:34:32", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One serving.", + "B": "Three servings.", + "C": "You took four servings of raw meat in total, counting everything you grabbed in one trip.", + "D": "You took two servings of raw meat in total.", + "E": "You took five servings of raw meat in total, after picking up multiple portions before leaving." + }, + "answer_label": "D", + "answer_text": "You took two servings of raw meat in total.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_11", + "correct_answer": "You took two servings of raw meat in total.", + "original_options": { + "A": "One serving.", + "B": "Three servings.", + "C": "You took four servings of raw meat in total, counting everything you grabbed in one trip.", + "D": "You took two servings of raw meat in total.", + "E": "You took five servings of raw meat in total, after picking up multiple portions before leaving." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_12", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many servings of raw meat did I just take in total?", + "answer": "You took two servings of raw meat in total.", + "question_chinese": "我刚才一共拿了几份生肉？", + "answer_chinese": "一共拿了两份生肉。", + "evidence": [ + { + "description": "Put a piece of meat in the shopping cart.", + "timestamps": { + "start": "00:33:07", + "end": "00:33:16" + } + }, + { + "description": "Put a piece of shredded meat in the shopping cart.", + "timestamps": { + "start": "00:33:32", + "end": "00:34:31" + } + } + ], + "question_timestamps": "00:34:44", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One serving.", + "B": "Three servings.", + "C": "You took four servings of raw meat in total, counting everything you grabbed in one trip.", + "D": "You took six servings of raw meat in total, after picking up multiple portions before leaving.", + "E": "You took two servings of raw meat in total." + }, + "answer_label": "E", + "answer_text": "You took two servings of raw meat in total.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_12", + "correct_answer": "You took two servings of raw meat in total.", + "original_options": { + "A": "One serving.", + "B": "Three servings.", + "C": "You took four servings of raw meat in total, counting everything you grabbed in one trip.", + "D": "You took six servings of raw meat in total, after picking up multiple portions before leaving.", + "E": "You took two servings of raw meat in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_9", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of yogurt did I just put in the shopping cart in total?", + "answer": "You took a total of 4 boxes of yogurt.", + "question_chinese": "我刚才一共拿了几盒酸奶放进购物车？", + "answer_chinese": "一共拿了4盒酸奶。", + "evidence": [ + { + "description": "Put two boxes of green yogurt in the shopping cart.", + "timestamps": { + "start": "00:30:04", + "end": "00:30:22" + } + }, + { + "description": "Put two boxes of blue yogurt in the shopping cart.", + "timestamps": { + "start": "00:30:28", + "end": "00:30:45" + } + } + ], + "question_timestamps": "00:31:11", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three boxes.", + "B": "Five boxes.", + "C": "You took a total of 4 boxes of yogurt.", + "D": "You put six boxes of yogurt in the cart in total, after picking up extra boxes.", + "E": "You put seven boxes of yogurt in the cart in total, adding more than you first intended." + }, + "answer_label": "C", + "answer_text": "You took a total of 4 boxes of yogurt.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_9", + "correct_answer": "You took a total of 4 boxes of yogurt.", + "original_options": { + "A": "Three boxes.", + "B": "Five boxes.", + "C": "You took a total of 4 boxes of yogurt.", + "D": "You put six boxes of yogurt in the cart in total, after picking up extra boxes.", + "E": "You put seven boxes of yogurt in the cart in total, adding more than you first intended." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_10", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of yogurt did I just put in the shopping cart in total?", + "answer": "You took a total of 4 boxes of yogurt.", + "question_chinese": "我刚才一共拿了几盒酸奶放进购物车？", + "answer_chinese": "一共拿了4盒酸奶。", + "evidence": [ + { + "description": "Put two boxes of green yogurt in the shopping cart.", + "timestamps": { + "start": "00:30:04", + "end": "00:30:22" + } + }, + { + "description": "Put two boxes of blue yogurt in the shopping cart.", + "timestamps": { + "start": "00:30:28", + "end": "00:30:45" + } + } + ], + "question_timestamps": "00:32:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three boxes.", + "B": "Five boxes.", + "C": "You took a total of 4 boxes of yogurt.", + "D": "You put six boxes of yogurt in the cart in total, after picking up extra boxes.", + "E": "You put eight boxes of yogurt in the cart in total, adding several more boxes." + }, + "answer_label": "C", + "answer_text": "You took a total of 4 boxes of yogurt.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_10", + "correct_answer": "You took a total of 4 boxes of yogurt.", + "original_options": { + "A": "Three boxes.", + "B": "Five boxes.", + "C": "You took a total of 4 boxes of yogurt.", + "D": "You put six boxes of yogurt in the cart in total, after picking up extra boxes.", + "E": "You put eight boxes of yogurt in the cart in total, adding several more boxes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 15 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_17", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of seasoning powder did I just take?", + "answer": "one bottle.", + "question_chinese": "我刚才拿了几瓶调味粉？", + "answer_chinese": "1瓶。", + "evidence": [ + { + "description": "You put a bottle of seasoning powder in the shopping cart.", + "timestamps": { + "start": "00:35:19", + "end": "00:35:57" + } + } + ], + "question_timestamps": "00:36:49", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero bottles.", + "B": "Two bottles.", + "C": "one bottle.", + "D": "You took three bottles of seasoning powder, grabbing multiple containers instead of just one.", + "E": "You took four bottles of seasoning powder, filling your hands with several bottles at once." + }, + "answer_label": "C", + "answer_text": "one bottle.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_17", + "correct_answer": "one bottle.", + "original_options": { + "A": "Zero bottles.", + "B": "Two bottles.", + "C": "one bottle.", + "D": "You took three bottles of seasoning powder, grabbing multiple containers instead of just one.", + "E": "You took four bottles of seasoning powder, filling your hands with several bottles at once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_18", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of seasoning powder did I just take?", + "answer": "one bottle.", + "question_chinese": "我刚才拿了几瓶调味粉？", + "answer_chinese": "1瓶。", + "evidence": [ + { + "description": "You put a bottle of seasoning powder in the shopping cart.", + "timestamps": { + "start": "00:35:19", + "end": "00:35:57" + } + } + ], + "question_timestamps": "00:38:36", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero bottles.", + "B": "Two bottles.", + "C": "You took three bottles of seasoning powder, grabbing multiple containers instead of just one.", + "D": "one bottle.", + "E": "You took four bottles of seasoning powder, filling your hands with several bottles at once." + }, + "answer_label": "D", + "answer_text": "one bottle.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_18", + "correct_answer": "one bottle.", + "original_options": { + "A": "Zero bottles.", + "B": "Two bottles.", + "C": "You took three bottles of seasoning powder, grabbing multiple containers instead of just one.", + "D": "one bottle.", + "E": "You took four bottles of seasoning powder, filling your hands with several bottles at once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_9_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many different kinds of food did I just pick up?", + "answer": "You picked up 5 kinds. They’re on the counter in front of you.", + "question_chinese": "我刚才拿了几种吃的东西？", + "answer_chinese": "你拿了5种。在你前面的柜台上。", + "evidence": [ + { + "description": "You've picked up five items in total.", + "timestamps": { + "start": "00:22:10", + "end": "00:27:56" + } + } + ], + "question_timestamps": "00:28:02", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You picked up 4 kinds.", + "B": "You picked up 6 kinds.", + "C": "You picked up 3 kinds. They’re on the counter in front of you.", + "D": "You picked up 5 kinds. They’re on the counter in front of you.", + "E": "You picked up 7 kinds. They’re on the counter in front of you." + }, + "answer_label": "D", + "answer_text": "You picked up 5 kinds. They’re on the counter in front of you.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_9_T1", + "correct_answer": "You picked up 5 kinds. They’re on the counter in front of you.", + "original_options": { + "A": "You picked up 4 kinds.", + "B": "You picked up 6 kinds.", + "C": "You picked up 3 kinds. They’re on the counter in front of you.", + "D": "You picked up 5 kinds. They’re on the counter in front of you.", + "E": "You picked up 7 kinds. They’re on the counter in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_9_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many different kinds of food did I just pick up?", + "answer": "You picked up 6 types of items. They’re in the shopping cart in front of you.", + "question_chinese": "我刚才拿了几种吃的东西？", + "answer_chinese": "你拿了6种。在你前面的购物车里。", + "evidence": [ + { + "description": "You've picked up six items in total.", + "timestamps": { + "start": "00:22:24", + "end": "00:28:58" + } + } + ], + "question_timestamps": "00:29:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You picked up 4 types of items.", + "B": "You picked up 8 types of items.", + "C": "You picked up 5 types of items, and they’re all sitting in the shopping cart in front of you.", + "D": "You picked up 7 types of items, and they’re placed together in the shopping cart in front of you.", + "E": "You picked up 6 types of items. They’re in the shopping cart in front of you." + }, + "answer_label": "E", + "answer_text": "You picked up 6 types of items. They’re in the shopping cart in front of you.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_9_T2", + "correct_answer": "You picked up 6 types of items. They’re in the shopping cart in front of you.", + "original_options": { + "A": "You picked up 4 types of items.", + "B": "You picked up 8 types of items.", + "C": "You picked up 5 types of items, and they’re all sitting in the shopping cart in front of you.", + "D": "You picked up 7 types of items, and they’re placed together in the shopping cart in front of you.", + "E": "You picked up 6 types of items. They’re in the shopping cart in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 20 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_16", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many kinds of food did I eat before I came to the supermarket?", + "answer": "Two kinds.", + "question_chinese": "我在来超市之前吃了几种食物？", + "answer_chinese": "两种", + "evidence": [ + { + "description": "You ate a black boxed food and a pizza.", + "timestamps": { + "start": "00:05:40", + "end": "00:17:32" + } + } + ], + "question_timestamps": "00:18:37", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One kind.", + "B": "Two kinds.", + "C": "Three kinds.", + "D": "You ate four kinds of food before you came to the supermarket earlier.", + "E": "You ate five different kinds of food before heading to the supermarket." + }, + "answer_label": "B", + "answer_text": "Two kinds.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_16", + "correct_answer": "Two kinds.", + "original_options": { + "A": "One kind.", + "B": "Two kinds.", + "C": "Three kinds.", + "D": "You ate four kinds of food before you came to the supermarket earlier.", + "E": "You ate five different kinds of food before heading to the supermarket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_15", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many kinds of food did I eat before I came to the supermarket?", + "answer": "Two kinds.", + "question_chinese": "我在来超市之前吃了几种食物？", + "answer_chinese": "两种", + "evidence": [ + { + "description": "You ate a black boxed food and a pizza.", + "timestamps": { + "start": "00:05:40", + "end": "00:17:32" + } + } + ], + "question_timestamps": "00:26:27", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One kind.", + "B": "Four kinds.", + "C": "Two kinds.", + "D": "You ate three kinds of food before you came to the supermarket.", + "E": "You ate five kinds of food before arriving at the supermarket earlier today." + }, + "answer_label": "C", + "answer_text": "Two kinds.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_15", + "correct_answer": "Two kinds.", + "original_options": { + "A": "One kind.", + "B": "Four kinds.", + "C": "Two kinds.", + "D": "You ate three kinds of food before you came to the supermarket.", + "E": "You ate five kinds of food before arriving at the supermarket earlier today." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_10_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were in the vehicle I was riding in?", + "answer": "There were 2 people in the vehicle you were riding in—one in front of you and one to your front-left.", + "question_chinese": "我坐的车上有几个人？", + "answer_chinese": "你坐的车上有2个人。一个在你的正前方，一个在你的左前方。", + "evidence": [ + { + "description": "There is only one driver in sight, and yourself.and the other one.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:01:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There was 1 person in the vehicle.", + "B": "There were 3 people in the vehicle.", + "C": "There were 4 people in the vehicle you were riding in—one in front, one front-left, and two in back.", + "D": "There were 2 people in the vehicle you were riding in—one in front of you and one to your front-left.", + "E": "There were 0 people in the vehicle you were riding in—you were alone with no passengers around you." + }, + "answer_label": "D", + "answer_text": "There were 2 people in the vehicle you were riding in—one in front of you and one to your front-left.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_10_T1", + "correct_answer": "There were 2 people in the vehicle you were riding in—one in front of you and one to your front-left.", + "original_options": { + "A": "There was 1 person in the vehicle.", + "B": "There were 3 people in the vehicle.", + "C": "There were 4 people in the vehicle you were riding in—one in front, one front-left, and two in back.", + "D": "There were 2 people in the vehicle you were riding in—one in front of you and one to your front-left.", + "E": "There were 0 people in the vehicle you were riding in—you were alone with no passengers around you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 19 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_10_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were in the vehicle I was riding in?", + "answer": "There was 1 person in the vehicle you were riding in, to your front-left.", + "question_chinese": "我坐的车上有几个人？", + "answer_chinese": "你坐的车上有1个人。在你的左前方。", + "evidence": [ + { + "description": "There is only one driver in sight, and yourself.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:01:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There were 0 people in the vehicle.", + "B": "There were 2 people in the vehicle.", + "C": "There were 3 people in the vehicle you were riding in, with one passenger to your front-left.", + "D": "There were 4 people in the vehicle you were riding in, with multiple passengers around you in front and back.", + "E": "There was 1 person in the vehicle you were riding in, to your front-left." + }, + "answer_label": "E", + "answer_text": "There was 1 person in the vehicle you were riding in, to your front-left.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_10_T2", + "correct_answer": "There was 1 person in the vehicle you were riding in, to your front-left.", + "original_options": { + "A": "There were 0 people in the vehicle.", + "B": "There were 2 people in the vehicle.", + "C": "There were 3 people in the vehicle you were riding in, with one passenger to your front-left.", + "D": "There were 4 people in the vehicle you were riding in, with multiple passengers around you in front and back.", + "E": "There was 1 person in the vehicle you were riding in, to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 20 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_19", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were in the car I took when I came?", + "answer": "Two people.", + "question_chinese": "我来的时候坐的车上有几个人？", + "answer_chinese": "两个人。", + "evidence": [ + { + "description": "There is only one driver in sight, and yourself.", + "timestamps": { + "start": "00:00:00", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:05:47", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One person.", + "B": "Three people.", + "C": "Two people.", + "D": "Four people were in the car you took when you came over.", + "E": "Five people were in the car you took when you came to the location." + }, + "answer_label": "C", + "answer_text": "Two people.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_19", + "correct_answer": "Two people.", + "original_options": { + "A": "One person.", + "B": "Three people.", + "C": "Two people.", + "D": "Four people were in the car you took when you came over.", + "E": "Five people were in the car you took when you came to the location." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_20", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were in the car I took when I came?", + "answer": "Two people.", + "question_chinese": "我来的时候坐的车上有几个人？", + "answer_chinese": "两个人。", + "evidence": [ + { + "description": "There is only one driver in sight, and yourself.", + "timestamps": { + "start": "00:00:00", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:18:27", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero people.", + "B": "Four people.", + "C": "Five people were in the car you took when you came there.", + "D": "Two people.", + "E": "Eight people were in the car you took when you came to the place." + }, + "answer_label": "D", + "answer_text": "Two people.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_20", + "correct_answer": "Two people.", + "original_options": { + "A": "Zero people.", + "B": "Four people.", + "C": "Five people were in the car you took when you came there.", + "D": "Two people.", + "E": "Eight people were in the car you took when you came to the place." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_4_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many things have I bought in total so far?", + "answer": "So far, you’ve bought 1 item in total. It’s in the shopping cart in front of you.", + "question_chinese": "我现在一共买了几份东西？", + "answer_chinese": "你现在一共买了1份东西。在你面前的购物车里。", + "evidence": [ + { + "description": "Paid for a serving of bananas.", + "timestamps": { + "start": "00:23:54", + "end": "00:24:12" + } + } + ], + "question_timestamps": "00:24:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, you’ve bought 0 items total.", + "B": "So far, you’ve bought 4 items total.", + "C": "So far, you’ve bought 2 items in total, and they’re sitting in the shopping cart in front of you.", + "D": "So far, you’ve bought 1 item in total. It’s in the shopping cart in front of you.", + "E": "So far, you’ve bought 3 items in total, and they’re all in the shopping cart in front of you." + }, + "answer_label": "D", + "answer_text": "So far, you’ve bought 1 item in total. It’s in the shopping cart in front of you.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_4_T1", + "correct_answer": "So far, you’ve bought 1 item in total. It’s in the shopping cart in front of you.", + "original_options": { + "A": "So far, you’ve bought 0 items total.", + "B": "So far, you’ve bought 4 items total.", + "C": "So far, you’ve bought 2 items in total, and they’re sitting in the shopping cart in front of you.", + "D": "So far, you’ve bought 1 item in total. It’s in the shopping cart in front of you.", + "E": "So far, you’ve bought 3 items in total, and they’re all in the shopping cart in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 21, + 21 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_7", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items have I bought in total now?", + "answer": "You have bought three items in total.", + "question_chinese": "我现在一共买了几份东西？", + "answer_chinese": "一共买了三份东西。", + "evidence": [ + { + "description": "Paid for a serving of bananas.", + "timestamps": { + "start": "00:23:54", + "end": "00:24:12" + } + }, + { + "description": "Took two boxes of food from the clerk.", + "timestamps": { + "start": "00:26:43", + "end": "00:26:43" + } + } + ], + "question_timestamps": "00:26:47", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have bought two items in total.", + "B": "You have bought three items in total.", + "C": "You have bought four items in total.", + "D": "You have bought five items in total so far, counting everything you’ve purchased up to now.", + "E": "You have bought one item in total now, with only a single purchase completed so far." + }, + "answer_label": "B", + "answer_text": "You have bought three items in total.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_7", + "correct_answer": "You have bought three items in total.", + "original_options": { + "A": "You have bought two items in total.", + "B": "You have bought three items in total.", + "C": "You have bought four items in total.", + "D": "You have bought five items in total so far, counting everything you’ve purchased up to now.", + "E": "You have bought one item in total now, with only a single purchase completed so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_4_T2", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many things have I bought in total so far?", + "answer": "You’ve bought 3 items in total. They’re in the shopping cart on your right-hand side.", + "question_chinese": "我现在一共买了几份东西？", + "answer_chinese": "你一共买了3份东西。在你右手边的购物车里。", + "evidence": [ + { + "description": "Paid for a serving of bananas.", + "timestamps": { + "start": "00:23:54", + "end": "00:24:12" + } + }, + { + "description": "Took two boxes of food from the clerk.", + "timestamps": { + "start": "00:26:42", + "end": "00:26:43" + } + } + ], + "question_timestamps": "00:27:01", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve bought 2 items in total.", + "B": "You’ve bought 4 items in total.", + "C": "You’ve bought 1 item in total, and it’s sitting in the shopping cart on your right-hand side.", + "D": "You’ve bought 5 items in total, and they’re all in the shopping cart on your right-hand side.", + "E": "You’ve bought 3 items in total. They’re in the shopping cart on your right-hand side." + }, + "answer_label": "E", + "answer_text": "You’ve bought 3 items in total. They’re in the shopping cart on your right-hand side.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_4_T2", + "correct_answer": "You’ve bought 3 items in total. They’re in the shopping cart on your right-hand side.", + "original_options": { + "A": "You’ve bought 2 items in total.", + "B": "You’ve bought 4 items in total.", + "C": "You’ve bought 1 item in total, and it’s sitting in the shopping cart on your right-hand side.", + "D": "You’ve bought 5 items in total, and they’re all in the shopping cart on your right-hand side.", + "E": "You’ve bought 3 items in total. They’re in the shopping cart on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_8_T1", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different kinds of food have I eaten so far?", + "answer": "You ate 1 kind. It’s on the table in front of you.", + "question_chinese": "我现在吃了几种食物？", + "answer_chinese": "你吃了1种。在你前面的桌子上。", + "evidence": [ + { + "description": "You ate a black boxed food.", + "timestamps": { + "start": "00:05:40", + "end": "00:09:53" + } + } + ], + "question_timestamps": "00:09:55", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You ate 0 kinds.", + "B": "You ate 2 kinds.", + "C": "You ate 1 kind. It’s on the table in front of you.", + "D": "You ate 3 kinds, and they’re on the table in front of you where you can see them.", + "E": "You ate 4 kinds, with all of them placed on the table in front of you right now." + }, + "answer_label": "C", + "answer_text": "You ate 1 kind. It’s on the table in front of you.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_8_T1", + "correct_answer": "You ate 1 kind. It’s on the table in front of you.", + "original_options": { + "A": "You ate 0 kinds.", + "B": "You ate 2 kinds.", + "C": "You ate 1 kind. It’s on the table in front of you.", + "D": "You ate 3 kinds, and they’re on the table in front of you where you can see them.", + "E": "You ate 4 kinds, with all of them placed on the table in front of you right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 18 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_8_T2", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different kinds of food have I eaten so far?", + "answer": "You ate 2 kinds. They’re on the table on your right-hand side.", + "question_chinese": "我现在吃了几种食物？", + "answer_chinese": "你吃了2种。在你右手边的桌子上。", + "evidence": [ + { + "description": "You ate a black boxed food and a pizza.", + "timestamps": { + "start": "00:05:40", + "end": "00:17:32" + } + } + ], + "question_timestamps": "00:17:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You ate 1 kind.", + "B": "You ate 3 kinds.", + "C": "You ate 0 kinds, and there isn’t any food on the table on your right-hand side.", + "D": "You ate 2 kinds. They’re on the table on your right-hand side.", + "E": "You ate 4 kinds, and they’re on the table on your right-hand side beside you." + }, + "answer_label": "D", + "answer_text": "You ate 2 kinds. They’re on the table on your right-hand side.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_8_T2", + "correct_answer": "You ate 2 kinds. They’re on the table on your right-hand side.", + "original_options": { + "A": "You ate 1 kind.", + "B": "You ate 3 kinds.", + "C": "You ate 0 kinds, and there isn’t any food on the table on your right-hand side.", + "D": "You ate 2 kinds. They’re on the table on your right-hand side.", + "E": "You ate 4 kinds, and they’re on the table on your right-hand side beside you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_5", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one am I closer to now, the banana or the watermelon?", + "answer": "The watermelon is closer.", + "question_chinese": "我现在离香蕉和西瓜哪个更近？", + "answer_chinese": "西瓜更近一些。", + "evidence": [ + { + "description": "The banana and watermelon are about 10 meters apart.", + "timestamps": { + "start": "00:23:10", + "end": "00:23:20" + } + }, + { + "description": "You are now near the watermelon.", + "timestamps": { + "start": "00:23:19", + "end": "00:23:24" + } + } + ], + "question_timestamps": "00:23:25", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The watermelon is closer.", + "B": "The banana is closer." + }, + "answer_label": "A", + "answer_text": "The watermelon is closer.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_5", + "correct_answer": "The watermelon is closer.", + "original_options": { + "A": "The watermelon is closer.", + "B": "The banana is closer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657", + "video_length": "00:46:24", + "qid": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_6", + "qid_num": 27, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one am I closer to now, the banana or the watermelon?", + "answer": "The watermelon is closer.", + "question_chinese": "我现在离香蕉和西瓜哪个更近？", + "answer_chinese": "西瓜更近一些。", + "evidence": [ + { + "description": "The banana and watermelon are about 10 meters apart.", + "timestamps": { + "start": "00:23:10", + "end": "00:23:20" + } + }, + { + "description": "You are now near the watermelon.", + "timestamps": { + "start": "00:23:19", + "end": "00:23:24" + } + } + ], + "question_timestamps": "00:23:59", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The banana is closer.", + "B": "The watermelon is closer." + }, + "answer_label": "B", + "answer_text": "The watermelon is closer.", + "q_id": "05a6d5f7-e1d5-4af3-9ae8-f07d0ac12657_6", + "correct_answer": "The watermelon is closer.", + "original_options": { + "A": "The banana is closer.", + "B": "The watermelon is closer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + } +] diff --git a/QAs/05dc055d-e88a-4ab1-8474-e72b97a818b2-mcq.json b/QAs/05dc055d-e88a-4ab1-8474-e72b97a818b2-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..265d425ec38003972ae9a49ea62d533f7a58f97e --- /dev/null +++ b/QAs/05dc055d-e88a-4ab1-8474-e72b97a818b2-mcq.json @@ -0,0 +1,687 @@ +[ + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I taken the stairs so far?", + "answer": "So far, you have taken one flight of stairs.", + "question_chinese": "到目前为止我下过几次楼梯？", + "answer_chinese": "到目前为止，你下过一次楼梯。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:24:41", + "end": "00:25:08" + } + } + ], + "question_timestamps": "00:34:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have taken two flights.", + "B": "So far, you have taken one flight of stairs.", + "C": "You have taken three flights.", + "D": "So far, you have taken four flights of stairs across your route up to this point.", + "E": "So far, you have taken five flights of stairs in total during everything you’ve done so far." + }, + "answer_label": "B", + "answer_text": "So far, you have taken one flight of stairs.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_7", + "correct_answer": "So far, you have taken one flight of stairs.", + "original_options": { + "A": "You have taken two flights.", + "B": "So far, you have taken one flight of stairs.", + "C": "You have taken three flights.", + "D": "So far, you have taken four flights of stairs across your route up to this point.", + "E": "So far, you have taken five flights of stairs in total during everything you’ve done so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I taken the stairs so far?", + "answer": "So far, you have descended the stairs twice.", + "question_chinese": "到目前为止我下过几次楼梯？", + "answer_chinese": "到目前为止你下了两次楼梯。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:24:47", + "end": "00:25:10" + } + }, + { + "description": "Second time", + "timestamps": { + "start": "00:43:56", + "end": "00:44:14" + } + } + ], + "question_timestamps": "00:51:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You descended the stairs once.", + "B": "You descended the stairs three times.", + "C": "So far, you have descended the stairs twice.", + "D": "So far, you have ascended the stairs twice rather than going down them during your movement so far.", + "E": "So far, you have climbed the stairs four times, meaning you went up repeatedly instead of descending twice." + }, + "answer_label": "C", + "answer_text": "So far, you have descended the stairs twice.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_8", + "correct_answer": "So far, you have descended the stairs twice.", + "original_options": { + "A": "You descended the stairs once.", + "B": "You descended the stairs three times.", + "C": "So far, you have descended the stairs twice.", + "D": "So far, you have ascended the stairs twice rather than going down them during your movement so far.", + "E": "So far, you have climbed the stairs four times, meaning you went up repeatedly instead of descending twice." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 18 + ] + }, + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_5_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many stairway corners did I just pass?", + "answer": "You just passed one stairway corner.", + "question_chinese": "我刚刚经过了几个楼梯拐角？", + "answer_chinese": "你刚刚经过了一个楼梯拐角。", + "evidence": [ + { + "description": "You have gone around one corners.", + "timestamps": { + "start": "00:38:10", + "end": "00:38:23" + } + } + ], + "question_timestamps": "00:38:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just passed zero stairway corners.", + "B": "You just passed two stairway corners.", + "C": "You just passed three stairway corners as you turned twice and then rounded one more bend.", + "D": "You just passed one stairway corner.", + "E": "You just passed four stairway corners, looping through several landings before continuing up the stairs." + }, + "answer_label": "D", + "answer_text": "You just passed one stairway corner.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_5_T1", + "correct_answer": "You just passed one stairway corner.", + "original_options": { + "A": "You just passed zero stairway corners.", + "B": "You just passed two stairway corners.", + "C": "You just passed three stairway corners as you turned twice and then rounded one more bend.", + "D": "You just passed one stairway corner.", + "E": "You just passed four stairway corners, looping through several landings before continuing up the stairs." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_5_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many stairway corners did I just pass?", + "answer": "You just passed two stairway corners.", + "question_chinese": "我刚刚经过了几个楼梯拐角？", + "answer_chinese": "你刚刚经过了两个楼梯拐角。", + "evidence": [ + { + "description": "You have gone around two corners.", + "timestamps": { + "start": "00:38:00", + "end": "00:38:45" + } + } + ], + "question_timestamps": "01:07:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just passed one stairway corner.", + "B": "You just passed three stairway corners.", + "C": "You just passed zero stairway corners because you went straight up without turning at a landing.", + "D": "You just passed four stairway corners, making several tight turns as you climbed past multiple landings.", + "E": "You just passed two stairway corners." + }, + "answer_label": "E", + "answer_text": "You just passed two stairway corners.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_5_T2", + "correct_answer": "You just passed two stairway corners.", + "original_options": { + "A": "You just passed one stairway corner.", + "B": "You just passed three stairway corners.", + "C": "You just passed zero stairway corners because you went straight up without turning at a landing.", + "D": "You just passed four stairway corners, making several tight turns as you climbed past multiple landings.", + "E": "You just passed two stairway corners." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_1_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the safety helmet that I picked up before?", + "answer": "It’s in my lidless wooden crate, and the crate is to your left.", + "question_chinese": "我曾经拿起过的那个安全帽在哪里？", + "answer_chinese": "在我的无盖木箱子里，木箱在我左边。", + "evidence": [ + { + "description": "That safety helmet is still in the open wooden box on the right.", + "timestamps": { + "start": "00:31:12", + "end": "00:31:32" + } + } + ], + "question_timestamps": "00:33:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in my lidless wooden crate, and the crate is to your left.", + "B": "It’s in the crate in front.", + "C": "It’s in the crate to your right.", + "D": "It’s in my lidless wooden crate, but the crate is positioned directly behind you right now.", + "E": "It’s in my lidless wooden crate, and the crate is slightly above you on the higher landing." + }, + "answer_label": "A", + "answer_text": "It’s in my lidless wooden crate, and the crate is to your left.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_1_T1", + "correct_answer": "It’s in my lidless wooden crate, and the crate is to your left.", + "original_options": { + "A": "It’s in my lidless wooden crate, and the crate is to your left.", + "B": "It’s in the crate in front.", + "C": "It’s in the crate to your right.", + "D": "It’s in my lidless wooden crate, but the crate is positioned directly behind you right now.", + "E": "It’s in my lidless wooden crate, and the crate is slightly above you on the higher landing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 18 + ] + }, + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_1_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the safety helmet that I picked up before?", + "answer": "That hard hat is in the lidless wooden crate to your left.", + "question_chinese": "我曾经拿起过的那个安全帽在哪里？", + "answer_chinese": "那个安全帽在你左边的无盖木箱子里。", + "evidence": [ + { + "description": "That safety helmet is still in the open wooden box on the right.", + "timestamps": { + "start": "00:31:12", + "end": "00:31:32" + } + } + ], + "question_timestamps": "00:33:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in the crate in front.", + "B": "It’s in the crate to your right.", + "C": "That hard hat is in the lidless wooden crate to your left.", + "D": "That hard hat is in the lidless wooden crate behind you, near where you were standing earlier.", + "E": "That hard hat is in the lidless wooden crate above you, up on the next level." + }, + "answer_label": "C", + "answer_text": "That hard hat is in the lidless wooden crate to your left.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_1_T2", + "correct_answer": "That hard hat is in the lidless wooden crate to your left.", + "original_options": { + "A": "It’s in the crate in front.", + "B": "It’s in the crate to your right.", + "C": "That hard hat is in the lidless wooden crate to your left.", + "D": "That hard hat is in the lidless wooden crate behind you, near where you were standing earlier.", + "E": "That hard hat is in the lidless wooden crate above you, up on the next level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 16 + ] + }, + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_9", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many corners do I pass by each time I go up the stairs?", + "answer": "You have gone around two corners.", + "question_chinese": "我每次上楼梯时经过了几个拐角？", + "answer_chinese": "你经历了两个拐角。", + "evidence": [ + { + "description": "You have gone around two corners.", + "timestamps": { + "start": "00:38:00", + "end": "00:38:45" + } + } + ], + "question_timestamps": "00:51:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have gone around one corner.", + "B": "You have gone around three corners.", + "C": "You have gone around four corners, turning at several landings before reaching the next level.", + "D": "You have gone around zero corners since you climb straight up without changing direction on the stairs.", + "E": "You have gone around two corners." + }, + "answer_label": "E", + "answer_text": "You have gone around two corners.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_9", + "correct_answer": "You have gone around two corners.", + "original_options": { + "A": "You have gone around one corner.", + "B": "You have gone around three corners.", + "C": "You have gone around four corners, turning at several landings before reaching the next level.", + "D": "You have gone around zero corners since you climb straight up without changing direction on the stairs.", + "E": "You have gone around two corners." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_10", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many corners do I pass by each time I go up the stairs?", + "answer": "You have gone around two corners.", + "question_chinese": "我每次上楼梯时经过了几个拐角？", + "answer_chinese": "你经历了两个拐角。", + "evidence": [ + { + "description": "You have gone around two corners.", + "timestamps": { + "start": "00:38:00", + "end": "00:38:45" + } + } + ], + "question_timestamps": "01:07:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have gone around one corner.", + "B": "You have gone around four corners.", + "C": "You have gone around three corners, making an extra turn at an additional landing on the way up.", + "D": "You have gone around two corners.", + "E": "You have gone around zero corners because the stair route is straight with no landings to turn on." + }, + "answer_label": "D", + "answer_text": "You have gone around two corners.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_10", + "correct_answer": "You have gone around two corners.", + "original_options": { + "A": "You have gone around one corner.", + "B": "You have gone around four corners.", + "C": "You have gone around three corners, making an extra turn at an additional landing on the way up.", + "D": "You have gone around two corners.", + "E": "You have gone around zero corners because the stair route is straight with no landings to turn on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_4", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to get back the safety helmet I once picked up. How should I go about it?", + "answer": "You will turn around and walk straight to the open wooden box.", + "question_chinese": "我要取回我曾经拿起过的安全帽，我将要怎么走？", + "answer_chinese": "你将要向后转，直走至无盖木箱子旁。", + "evidence": [ + { + "description": "I threw the safety helmet into the open wooden box.", + "timestamps": { + "start": "00:05:36", + "end": "00:06:01" + } + }, + { + "description": "Current position and target position", + "timestamps": { + "start": "00:31:10", + "end": "00:31:38" + } + } + ], + "question_timestamps": "00:33:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You will turn left and walk straight ahead.", + "B": "Turn right and walk to the crate.", + "C": "You will walk straight first, then turn around and head to the wooden box.", + "D": "You will turn around, but walk straight to the closed wooden box instead of the open one.", + "E": "You will turn around and walk straight to the open wooden box." + }, + "answer_label": "E", + "answer_text": "You will turn around and walk straight to the open wooden box.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_4", + "correct_answer": "You will turn around and walk straight to the open wooden box.", + "original_options": { + "A": "You will turn left and walk straight ahead.", + "B": "Turn right and walk to the crate.", + "C": "You will walk straight first, then turn around and head to the wooden box.", + "D": "You will turn around, but walk straight to the closed wooden box instead of the open one.", + "E": "You will turn around and walk straight to the open wooden box." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 17 + ] + }, + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_3", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to get back the safety helmet I once picked up. How should I go about it?", + "answer": "You will turn right, walk to the stairwell, then take the stairs up to the top floor. Then turn right, walk a few steps, turn right again, walk to between the two blue rooms, then turn left and walk straight until you reach the sandy area. Then turn left and walk to the safety helmet.", + "question_chinese": "我要取回我曾经拿起过的安全帽，我将要怎么走？", + "answer_chinese": "你将要右转，走到楼梯口，然后上楼梯直到最顶层，然后右转，走几步再右转，走到两个蓝色房间中间再左转，直走至走上沙土地，然后左转走至安全帽旁。", + "evidence": [ + { + "description": "The safety helmet was placed on the metal products on the top layer.", + "timestamps": { + "start": "00:37:28", + "end": "00:39:09" + } + }, + { + "description": "My movement path", + "timestamps": { + "start": "00:41:25", + "end": "00:43:04" + } + }, + { + "description": "My movement path", + "timestamps": { + "start": "00:43:51", + "end": "00:44:13" + } + } + ], + "question_timestamps": "00:50:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left and go to the stairwell.", + "B": "Go upstairs, then turn left twice.", + "C": "You will turn right, walk to the stairwell, then take the stairs up to the top floor. Then turn right, walk a few steps, turn right again, walk to between the two blue rooms, then turn left and walk straight until you reach the sandy area. Then turn left and walk to the safety helmet.", + "D": "You will turn right, go to the stairwell, take the stairs up, then turn left toward the blue rooms.", + "E": "Turn right, reach the stairwell, go up to the top floor, then turn right once and walk straight." + }, + "answer_label": "C", + "answer_text": "You will turn right, walk to the stairwell, then take the stairs up to the top floor. Then turn right, walk a few steps, turn right again, walk to between the two blue rooms, then turn left and walk straight until you reach the sandy area. Then turn left and walk to the safety helmet.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_3", + "correct_answer": "You will turn right, walk to the stairwell, then take the stairs up to the top floor. Then turn right, walk a few steps, turn right again, walk to between the two blue rooms, then turn left and walk straight until you reach the sandy area. Then turn left and walk to the safety helmet.", + "original_options": { + "A": "Turn left and go to the stairwell.", + "B": "Go upstairs, then turn left twice.", + "C": "You will turn right, walk to the stairwell, then take the stairs up to the top floor. Then turn right, walk a few steps, turn right again, walk to between the two blue rooms, then turn left and walk straight until you reach the sandy area. Then turn left and walk to the safety helmet.", + "D": "You will turn right, go to the stairwell, take the stairs up, then turn left toward the blue rooms.", + "E": "Turn right, reach the stairwell, go up to the top floor, then turn right once and walk straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 18 + ] + }, + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_5", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are wearing safety helmets?", + "answer": "Three.", + "question_chinese": "有多少人戴着安全帽？", + "answer_chinese": "三人。", + "evidence": [ + { + "description": "A person removed the safety helmet.", + "timestamps": { + "start": "00:02:02", + "end": "00:02:18" + } + }, + { + "description": "Three people are wearing safety helmets.", + "timestamps": { + "start": "00:00:13", + "end": "00:01:30" + } + }, + { + "description": "There is a people wearing safety helmet who has not appeared before.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:30" + } + } + ], + "question_timestamps": "00:17:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "Only two people are wearing safety helmets in the scene right now.", + "E": "Five people have safety helmets on, and one more person is also wearing one." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_5", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "Only two people are wearing safety helmets in the scene right now.", + "E": "Five people have safety helmets on, and one more person is also wearing one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "05dc055d-e88a-4ab1-8474-e72b97a818b2", + "video_length": "01:07:43", + "qid": "05dc055d-e88a-4ab1-8474-e72b97a818b2_6", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are wearing safety helmets?", + "answer": "Two.", + "question_chinese": "有多少人戴着安全帽？", + "answer_chinese": "两人。", + "evidence": [ + { + "description": "Two people are wearing safety helmets.", + "timestamps": { + "start": "00:40:35", + "end": "00:41:47" + } + }, + { + "description": "No one is wearing a safety helmet.", + "timestamps": { + "start": "00:44:15", + "end": "00:44:21" + } + } + ], + "question_timestamps": "00:50:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Three people are wearing safety helmets, rather than only two in this scene.", + "E": "Four individuals are wearing safety helmets while one person is without one." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "05dc055d-e88a-4ab1-8474-e72b97a818b2_6", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Three people are wearing safety helmets, rather than only two in this scene.", + "E": "Four individuals are wearing safety helmets while one person is without one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + } +] diff --git a/QAs/06a51f5d-79ab-4b2c-adef-05aea1e06077-mcq.json b/QAs/06a51f5d-79ab-4b2c-adef-05aea1e06077-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5f15fbd8ad218464e403c2ed1abbe435043316c5 --- /dev/null +++ b/QAs/06a51f5d-79ab-4b2c-adef-05aea1e06077-mcq.json @@ -0,0 +1,898 @@ +[ + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_2_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go from the bedroom to the living room, how many turns do I need to make?", + "answer": "You need to make two turns.", + "question_chinese": "如果我想要从卧室到客厅需要拐几个弯？", + "answer_chinese": "你需要拐两个弯。", + "evidence": [ + { + "description": "You turned the first corner.", + "timestamps": { + "start": "00:03:31", + "end": "00:03:37" + } + }, + { + "description": "You turned the second corner.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:48" + } + } + ], + "question_timestamps": "00:03:58", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One turn.", + "B": "You need to make two turns.", + "C": "Three turns.", + "D": "You need to make one turn to reach the living room from the bedroom.", + "E": "You need to make three turns along the way before arriving in the living room." + }, + "answer_label": "B", + "answer_text": "You need to make two turns.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_2_T1", + "correct_answer": "You need to make two turns.", + "original_options": { + "A": "One turn.", + "B": "You need to make two turns.", + "C": "Three turns.", + "D": "You need to make one turn to reach the living room from the bedroom.", + "E": "You need to make three turns along the way before arriving in the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_2_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go from the bedroom to the living room, how many turns do I need to make?", + "answer": "2. Turn left, go straight, then turn right.", + "question_chinese": "如果我想要从卧室到客厅需要拐几个弯？", + "answer_chinese": "2，你需要朝你的左边拐一个弯，再直走，往右边拐一个弯。", + "evidence": [ + { + "description": "You turned the first corner.", + "timestamps": { + "start": "00:03:31", + "end": "00:03:37" + } + }, + { + "description": "You turned the second corner.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:48" + } + } + ], + "question_timestamps": "00:06:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "0. Go straight ahead.", + "B": "1. Turn right, then go straight.", + "C": "2. Turn left, go straight, then turn right.", + "D": "3. Turn left, go straight, turn right, then turn left once more to enter.", + "E": "4. Turn right, go straight, turn left, go straight, then turn right to reach the living room." + }, + "answer_label": "C", + "answer_text": "2. Turn left, go straight, then turn right.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_2_T2", + "correct_answer": "2. Turn left, go straight, then turn right.", + "original_options": { + "A": "0. Go straight ahead.", + "B": "1. Turn right, then go straight.", + "C": "2. Turn left, go straight, then turn right.", + "D": "3. Turn left, go straight, turn right, then turn left once more to enter.", + "E": "4. Turn right, go straight, turn left, go straight, then turn right to reach the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 14, + 17 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_7_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Relative to me, where did I just dump the vegetables from the bowl?", + "answer": "You just dumped the vegetables into the trash can behind you.", + "question_chinese": "我刚刚把盆子里的菜倒在相对于我的什么地方了？", + "answer_chinese": "你刚刚把菜倒进你后面的垃圾桶里了。", + "evidence": [ + { + "description": "You just poured the vegetables into the trash can.", + "timestamps": { + "start": "00:25:54", + "end": "00:26:05" + } + } + ], + "question_timestamps": "00:26:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the trash can in front.", + "B": "You just dumped the vegetables into the trash can behind you.", + "C": "Into the trash can to your left.", + "D": "You dumped the vegetables into the trash can to your right, not behind you.", + "E": "You emptied the bowl into the trash can directly in front of you near the counter." + }, + "answer_label": "B", + "answer_text": "You just dumped the vegetables into the trash can behind you.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_7_T1", + "correct_answer": "You just dumped the vegetables into the trash can behind you.", + "original_options": { + "A": "Into the trash can in front.", + "B": "You just dumped the vegetables into the trash can behind you.", + "C": "Into the trash can to your left.", + "D": "You dumped the vegetables into the trash can to your right, not behind you.", + "E": "You emptied the bowl into the trash can directly in front of you near the counter." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 16 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_7_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Relative to me, where did I just dump the vegetables from the bowl?", + "answer": "You just dumped the vegetables into the kitchen trash can behind you.", + "question_chinese": "我刚刚把盆子里的菜倒在相对于我的什么地方了？", + "answer_chinese": "你刚刚把菜倒进位于你身后的厨房的垃圾桶里了。", + "evidence": [ + { + "description": "You just poured the vegetables into the trash can.", + "timestamps": { + "start": "00:25:54", + "end": "00:26:05" + } + } + ], + "question_timestamps": "00:26:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the kitchen trash can in front.", + "B": "Into the kitchen trash can to your right.", + "C": "You just dumped the vegetables into the kitchen trash can behind you.", + "D": "You dumped the vegetables into the kitchen trash can to your left, not behind you.", + "E": "You emptied the bowl into the kitchen trash can in front of you, rather than behind you." + }, + "answer_label": "C", + "answer_text": "You just dumped the vegetables into the kitchen trash can behind you.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_7_T2", + "correct_answer": "You just dumped the vegetables into the kitchen trash can behind you.", + "original_options": { + "A": "Into the kitchen trash can in front.", + "B": "Into the kitchen trash can to your right.", + "C": "You just dumped the vegetables into the kitchen trash can behind you.", + "D": "You dumped the vegetables into the kitchen trash can to your left, not behind you.", + "E": "You emptied the bowl into the kitchen trash can in front of you, rather than behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 17 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_5_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator directly right now?", + "answer": "You can reach the refrigerator to your right directly right now.", + "question_chinese": "我现在可以直接触碰到冰箱吗？", + "answer_chinese": "你现在可以直接触碰到你右边的冰箱。", + "evidence": [ + { + "description": "You can now touch the refrigerator directly.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:30" + } + } + ], + "question_timestamps": "00:09:32", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the refrigerator to your right directly right now.", + "B": "You can’t reach the refrigerator directly right now; it’s to your back-right." + }, + "answer_label": "A", + "answer_text": "You can reach the refrigerator to your right directly right now.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_5_T1", + "correct_answer": "You can reach the refrigerator to your right directly right now.", + "original_options": { + "A": "You can reach the refrigerator to your right directly right now.", + "B": "You can’t reach the refrigerator directly right now; it’s to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_5_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator directly right now?", + "answer": "You can’t reach the refrigerator directly right now; it’s to your back-right.", + "question_chinese": "我现在可以直接触碰到冰箱吗？", + "answer_chinese": "你现在不可以直接触碰到冰箱，冰箱在你的右后边。", + "evidence": [ + { + "description": "You can now touch the refrigerator directly.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:30" + } + } + ], + "question_timestamps": "00:09:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the refrigerator to your right directly right now.", + "B": "You can’t reach the refrigerator directly right now; it’s to your back-right." + }, + "answer_label": "B", + "answer_text": "You can’t reach the refrigerator directly right now; it’s to your back-right.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_5_T2", + "correct_answer": "You can’t reach the refrigerator directly right now; it’s to your back-right.", + "original_options": { + "A": "You can reach the refrigerator to your right directly right now.", + "B": "You can’t reach the refrigerator directly right now; it’s to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_3_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach out and touch the guitar directly right now?", + "answer": "You can’t reach it directly right now. The guitar is to your back-left.", + "question_chinese": "我现在可以直接触碰到吉他吗？", + "answer_chinese": "你现在不可以直接触碰到。吉他在你的左后方。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:04:59", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:05:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach it directly right now. The guitar is to your back-left.", + "B": "You can reach it directly right now; the guitar is to your right." + }, + "answer_label": "A", + "answer_text": "You can’t reach it directly right now. The guitar is to your back-left.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_3_T1", + "correct_answer": "You can’t reach it directly right now. The guitar is to your back-left.", + "original_options": { + "A": "You can’t reach it directly right now. The guitar is to your back-left.", + "B": "You can reach it directly right now; the guitar is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_3_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach out and touch the guitar directly right now?", + "answer": "You can reach it directly right now; the guitar is to your right.", + "question_chinese": "我现在可以直接触碰到吉他吗？", + "answer_chinese": "你现在可以直接触碰到，吉他在你的右边。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:12:40", + "end": "00:12:46" + } + } + ], + "question_timestamps": "00:12:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach it directly right now. The guitar is to your back-left.", + "B": "You can reach it directly right now; the guitar is to your right." + }, + "answer_label": "B", + "answer_text": "You can reach it directly right now; the guitar is to your right.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_3_T2", + "correct_answer": "You can reach it directly right now; the guitar is to your right.", + "original_options": { + "A": "You can’t reach it directly right now. The guitar is to your back-left.", + "B": "You can reach it directly right now; the guitar is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_9_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach out and touch the robot vacuum directly right now?", + "answer": "You can reach it directly right now; it’s to your front-left.", + "question_chinese": "我现在可以直接触碰到扫地机器人吗？", + "answer_chinese": "你现在可以直接触碰到，它在你的左前方。", + "evidence": [ + { + "description": "You can't make direct contact now.", + "timestamps": { + "start": "00:09:52", + "end": "00:10:02" + } + } + ], + "question_timestamps": "00:10:04", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach it directly right now; it’s to your front-left.", + "B": "You can’t reach it directly right now; the robot vacuum is in the room to your back-right." + }, + "answer_label": "A", + "answer_text": "You can reach it directly right now; it’s to your front-left.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_9_T1", + "correct_answer": "You can reach it directly right now; it’s to your front-left.", + "original_options": { + "A": "You can reach it directly right now; it’s to your front-left.", + "B": "You can’t reach it directly right now; the robot vacuum is in the room to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_9_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach out and touch the robot vacuum directly right now?", + "answer": "You can’t reach it directly right now; the robot vacuum is in the room to your back-right.", + "question_chinese": "我现在可以直接触碰到扫地机器人吗？", + "answer_chinese": "你现在无法直接触碰到，扫地机器人在你右后方的房间里。", + "evidence": [ + { + "description": "You can't make direct contact now.", + "timestamps": { + "start": "00:09:52", + "end": "00:10:02" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach it directly right now; it’s to your front-left.", + "B": "You can’t reach it directly right now; the robot vacuum is in the room to your back-right." + }, + "answer_label": "B", + "answer_text": "You can’t reach it directly right now; the robot vacuum is in the room to your back-right.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_9_T2", + "correct_answer": "You can’t reach it directly right now; the robot vacuum is in the room to your back-right.", + "original_options": { + "A": "You can reach it directly right now; it’s to your front-left.", + "B": "You can’t reach it directly right now; the robot vacuum is in the room to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_6_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the faucet right now?", + "answer": "You can reach it directly right now.", + "question_chinese": "我现在可以直接触碰到水龙头吗？", + "answer_chinese": "你现在可以直接触碰到。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:17:05", + "end": "00:17:11" + } + } + ], + "question_timestamps": "00:17:44", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can't reach it directly right now.", + "B": "You can reach it directly right now." + }, + "answer_label": "B", + "answer_text": "You can reach it directly right now.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_6_T1", + "correct_answer": "You can reach it directly right now.", + "original_options": { + "A": "You can't reach it directly right now.", + "B": "You can reach it directly right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_6_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the faucet right now?", + "answer": "You can reach it directly. The faucet is to your front-left.", + "question_chinese": "我现在可以直接触碰到水龙头吗？", + "answer_chinese": "你可以直接触碰到。水龙头在你的左前方。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:21:51", + "end": "00:21:53" + } + } + ], + "question_timestamps": "00:22:14", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach it directly. The faucet is to your front-left.", + "B": "You can't reach it directly. The faucet is to your front-left." + }, + "answer_label": "A", + "answer_text": "You can reach it directly. The faucet is to your front-left.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_6_T2", + "correct_answer": "You can reach it directly. The faucet is to your front-left.", + "original_options": { + "A": "You can reach it directly. The faucet is to your front-left.", + "B": "You can't reach it directly. The faucet is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What position is the sofa relative to me now?", + "answer": "The sofa is on your left.", + "question_chinese": "沙发现在相对我是什么位置？", + "answer_chinese": "沙发在你的左侧。", + "evidence": [ + { + "description": "The sofa is on your left.", + "timestamps": { + "start": "00:04:47", + "end": "00:04:51" + } + } + ], + "question_timestamps": "00:04:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The sofa is in front of you.", + "B": "The sofa is on your left.", + "C": "The sofa is on your right.", + "D": "The sofa is directly behind you right now, not off to your left side.", + "E": "The sofa sits behind you, so you would have to turn around to face it." + }, + "answer_label": "B", + "answer_text": "The sofa is on your left.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_1", + "correct_answer": "The sofa is on your left.", + "original_options": { + "A": "The sofa is in front of you.", + "B": "The sofa is on your left.", + "C": "The sofa is on your right.", + "D": "The sofa is directly behind you right now, not off to your left side.", + "E": "The sofa sits behind you, so you would have to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 15 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What position is the sofa relative to me now?", + "answer": "The sofa is now behind you.", + "question_chinese": "沙发现在相对我是什么位置？", + "answer_chinese": "沙发现在在你的后方。", + "evidence": [ + { + "description": "The sofa is now behind you.", + "timestamps": { + "start": "00:10:33", + "end": "00:10:36" + } + } + ], + "question_timestamps": "00:10:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The sofa is in front of you.", + "B": "The sofa is on your left.", + "C": "The sofa is now behind you.", + "D": "The sofa is positioned to your right side right now, not behind you where you can’t see it.", + "E": "The sofa is off to your left at the moment, so it isn’t located behind you." + }, + "answer_label": "C", + "answer_text": "The sofa is now behind you.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_2", + "correct_answer": "The sofa is now behind you.", + "original_options": { + "A": "The sofa is in front of you.", + "B": "The sofa is on your left.", + "C": "The sofa is now behind you.", + "D": "The sofa is positioned to your right side right now, not behind you where you can’t see it.", + "E": "The sofa is off to your left at the moment, so it isn’t located behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 17 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_7", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the position of the wardrobe relative to me now?", + "answer": "The wardrobe is now on your right side.", + "question_chinese": "衣柜现在相对我是什么位置？", + "answer_chinese": "衣柜现在在你的右侧。", + "evidence": [ + { + "description": "The wardrobe is now on your right side.", + "timestamps": { + "start": "00:07:08", + "end": "00:07:11" + } + } + ], + "question_timestamps": "00:07:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your left.", + "B": "It is behind you.", + "C": "The wardrobe is positioned directly in front of you right now, facing the same direction you are.", + "D": "The wardrobe is now on your right side.", + "E": "The wardrobe sits slightly behind and to your left, not on the side you are currently facing." + }, + "answer_label": "D", + "answer_text": "The wardrobe is now on your right side.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_7", + "correct_answer": "The wardrobe is now on your right side.", + "original_options": { + "A": "It is on your left.", + "B": "It is behind you.", + "C": "The wardrobe is positioned directly in front of you right now, facing the same direction you are.", + "D": "The wardrobe is now on your right side.", + "E": "The wardrobe sits slightly behind and to your left, not on the side you are currently facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 17 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_8", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the position of the wardrobe relative to me now?", + "answer": "The wardrobe is now on your left.", + "question_chinese": "衣柜现在相对我是什么位置？", + "answer_chinese": "衣柜现在在你的左侧。", + "evidence": [ + { + "description": "The wardrobe is now on your left.", + "timestamps": { + "start": "00:15:14", + "end": "00:15:23" + } + } + ], + "question_timestamps": "00:15:29", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your right.", + "B": "It is in front of you.", + "C": "The wardrobe is positioned behind you now, so you would need to turn around to see it.", + "D": "The wardrobe is above you at the moment, higher than your eye level rather than beside you.", + "E": "The wardrobe is now on your left." + }, + "answer_label": "E", + "answer_text": "The wardrobe is now on your left.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_8", + "correct_answer": "The wardrobe is now on your left.", + "original_options": { + "A": "It is on your right.", + "B": "It is in front of you.", + "C": "The wardrobe is positioned behind you now, so you would need to turn around to see it.", + "D": "The wardrobe is above you at the moment, higher than your eye level rather than beside you.", + "E": "The wardrobe is now on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 17 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_17", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the balcony relative to me now?", + "answer": "The balcony is now in front of you.", + "question_chinese": "阳台相对我现在是什么位置？", + "answer_chinese": "阳台现在在你的前方。", + "evidence": [ + { + "description": "The balcony is now in front of you.", + "timestamps": { + "start": "00:23:08", + "end": "00:23:10" + } + } + ], + "question_timestamps": "00:23:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your right.", + "B": "It is behind you.", + "C": "The balcony is now in front of you.", + "D": "The balcony is located to your left side now, so it is not in your direct line of sight.", + "E": "The balcony sits behind and slightly below you at the moment, rather than straight ahead." + }, + "answer_label": "C", + "answer_text": "The balcony is now in front of you.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_17", + "correct_answer": "The balcony is now in front of you.", + "original_options": { + "A": "It is on your right.", + "B": "It is behind you.", + "C": "The balcony is now in front of you.", + "D": "The balcony is located to your left side now, so it is not in your direct line of sight.", + "E": "The balcony sits behind and slightly below you at the moment, rather than straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 19, + 15 + ] + }, + { + "video_uid": "06a51f5d-79ab-4b2c-adef-05aea1e06077", + "video_length": "00:30:00", + "qid": "06a51f5d-79ab-4b2c-adef-05aea1e06077_18", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the balcony relative to me now?", + "answer": "The balcony is now on your right.", + "question_chinese": "阳台相对我现在是什么位置？", + "answer_chinese": "阳台现在在你的右侧。", + "evidence": [ + { + "description": "The balcony is now on your right.", + "timestamps": { + "start": "00:28:38", + "end": "00:28:40" + } + } + ], + "question_timestamps": "00:28:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your left.", + "B": "It is behind you.", + "C": "The balcony is directly in front of you right now, not positioned to the side you are facing.", + "D": "The balcony is now on your right.", + "E": "The balcony is above you at the moment, higher than your current position rather than to your right." + }, + "answer_label": "D", + "answer_text": "The balcony is now on your right.", + "q_id": "06a51f5d-79ab-4b2c-adef-05aea1e06077_18", + "correct_answer": "The balcony is now on your right.", + "original_options": { + "A": "It is on your left.", + "B": "It is behind you.", + "C": "The balcony is directly in front of you right now, not positioned to the side you are facing.", + "D": "The balcony is now on your right.", + "E": "The balcony is above you at the moment, higher than your current position rather than to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 18 + ] + } +] diff --git a/QAs/0a01978c-e16d-4587-95f1-49efa3ab15d9-mcq.json b/QAs/0a01978c-e16d-4587-95f1-49efa3ab15d9-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e688db6efe4fef59212870ff274a6f7eb6293490 --- /dev/null +++ b/QAs/0a01978c-e16d-4587-95f1-49efa3ab15d9-mcq.json @@ -0,0 +1,1263 @@ +[ + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_3_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The refrigerator is now in front of you.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱现在在你的前面。", + "evidence": [ + { + "description": "The refrigerator is in front of you now.", + "timestamps": { + "start": "00:10:06", + "end": "00:10:08" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is now in front of you.", + "B": "It is behind you.", + "C": "It is on your left.", + "D": "The refrigerator is positioned on your right side now, so you would look right to see it.", + "E": "The refrigerator is above you at the moment, not at your eye level in front of you." + }, + "answer_label": "A", + "answer_text": "The refrigerator is now in front of you.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_3_T1", + "correct_answer": "The refrigerator is now in front of you.", + "original_options": { + "A": "The refrigerator is now in front of you.", + "B": "It is behind you.", + "C": "It is on your left.", + "D": "The refrigerator is positioned on your right side now, so you would look right to see it.", + "E": "The refrigerator is above you at the moment, not at your eye level in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 17 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_5", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator now relative to me?", + "answer": "The refrigerator is now on your left.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱现在在你的左侧。", + "evidence": [ + { + "description": "The refrigerator is now on your left.", + "timestamps": { + "start": "00:10:06", + "end": "00:10:18" + } + } + ], + "question_timestamps": "00:10:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is now on your left.", + "B": "It is on your right.", + "C": "It is behind you.", + "D": "The refrigerator is in front of you right now, directly ahead rather than to your left side.", + "E": "The refrigerator is positioned above you at the moment, higher up rather than beside you." + }, + "answer_label": "A", + "answer_text": "The refrigerator is now on your left.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_5", + "correct_answer": "The refrigerator is now on your left.", + "original_options": { + "A": "The refrigerator is now on your left.", + "B": "It is on your right.", + "C": "It is behind you.", + "D": "The refrigerator is in front of you right now, directly ahead rather than to your left side.", + "E": "The refrigerator is positioned above you at the moment, higher up rather than beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 15 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_3_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The fridge is on the left side of you now.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱现在在你的左侧。", + "evidence": [ + { + "description": "The refrigerator is now on your left.", + "timestamps": { + "start": "00:10:06", + "end": "00:10:19" + } + } + ], + "question_timestamps": "00:22:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is in front of you now.", + "B": "The fridge is on the left side of you now.", + "C": "The fridge is to your right now.", + "D": "The fridge is behind you now, so you would have to turn around to face it.", + "E": "The fridge is above you right now, higher than your eye level rather than beside you." + }, + "answer_label": "B", + "answer_text": "The fridge is on the left side of you now.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_3_T2", + "correct_answer": "The fridge is on the left side of you now.", + "original_options": { + "A": "The fridge is in front of you now.", + "B": "The fridge is on the left side of you now.", + "C": "The fridge is to your right now.", + "D": "The fridge is behind you now, so you would have to turn around to face it.", + "E": "The fridge is above you right now, higher than your eye level rather than beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 16 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_9", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the living room relative to me?", + "answer": "The living room is in front of you.", + "question_chinese": "客厅相对我是什么位置？", + "answer_chinese": "客厅在你的前方。", + "evidence": [ + { + "description": "The living room is in front of you.", + "timestamps": { + "start": "00:20:00", + "end": "00:20:05" + } + } + ], + "question_timestamps": "00:20:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is to your right.", + "B": "It is behind you.", + "C": "The living room is behind and slightly to your left, not straight ahead of your current position.", + "D": "The living room is in front of you.", + "E": "The living room is beyond the kitchen area now, so it is not directly in front of you." + }, + "answer_label": "D", + "answer_text": "The living room is in front of you.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_9", + "correct_answer": "The living room is in front of you.", + "original_options": { + "A": "It is to your right.", + "B": "It is behind you.", + "C": "The living room is behind and slightly to your left, not straight ahead of your current position.", + "D": "The living room is in front of you.", + "E": "The living room is beyond the kitchen area now, so it is not directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 18 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_10", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the living room relative to me?", + "answer": "The living room is on your right.", + "question_chinese": "客厅相对我是什么位置？", + "answer_chinese": "客厅在你的右侧。", + "evidence": [ + { + "description": "The living room is on your right.", + "timestamps": { + "start": "00:33:10", + "end": "00:33:13" + } + } + ], + "question_timestamps": "00:33:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The living room is on your left.", + "B": "The living room is on your right.", + "C": "The living room is behind you.", + "D": "The living room is straight ahead of you, just past the doorway.", + "E": "The living room is slightly behind you on the left side, near the doorway." + }, + "answer_label": "B", + "answer_text": "The living room is on your right.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_10", + "correct_answer": "The living room is on your right.", + "original_options": { + "A": "The living room is on your left.", + "B": "The living room is on your right.", + "C": "The living room is behind you.", + "D": "The living room is straight ahead of you, just past the doorway.", + "E": "The living room is slightly behind you on the left side, near the doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 14 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_3", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pancakes did I fry in total just now?", + "answer": "You just fried a total of two pancakes.", + "question_chinese": "我刚刚一共煎了几块饼？", + "answer_chinese": "你刚刚一共煎了2块饼。", + "evidence": [ + { + "description": "You've finished frying the first pancake.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:48" + } + }, + { + "description": "You've finished frying the second pancake.", + "timestamps": { + "start": "00:01:03", + "end": "00:05:15" + } + } + ], + "question_timestamps": "00:05:17", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You just fried a total of two pancakes.", + "B": "You fried one pancake.", + "C": "You fried three pancakes.", + "D": "You just fried a total of four pancakes during this cooking step.", + "E": "You just fried a total of five pancakes altogether during this cooking step." + }, + "answer_label": "A", + "answer_text": "You just fried a total of two pancakes.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_3", + "correct_answer": "You just fried a total of two pancakes.", + "original_options": { + "A": "You just fried a total of two pancakes.", + "B": "You fried one pancake.", + "C": "You fried three pancakes.", + "D": "You just fried a total of four pancakes during this cooking step.", + "E": "You just fried a total of five pancakes altogether during this cooking step." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_4", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pancakes did I fry in total just now?", + "answer": "You just fried a total of three pancakes.", + "question_chinese": "我刚刚一共煎了几块饼？", + "answer_chinese": "你刚刚一共煎了3块饼。", + "evidence": [ + { + "description": "You've finished frying the first pancake.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:48" + } + }, + { + "description": "You've finished frying the second pancake.", + "timestamps": { + "start": "00:01:08", + "end": "00:05:16" + } + }, + { + "description": "You've finished frying the third pancake.", + "timestamps": { + "start": "00:05:28", + "end": "00:11:53" + } + } + ], + "question_timestamps": "00:11:54", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You fried two pancakes.", + "B": "You fried four pancakes.", + "C": "You just fried a total of three pancakes.", + "D": "You just fried a total of five pancakes during this cooking step.", + "E": "You just fried a total of six pancakes altogether during this cooking step." + }, + "answer_label": "C", + "answer_text": "You just fried a total of three pancakes.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_4", + "correct_answer": "You just fried a total of three pancakes.", + "original_options": { + "A": "You fried two pancakes.", + "B": "You fried four pancakes.", + "C": "You just fried a total of three pancakes.", + "D": "You just fried a total of five pancakes during this cooking step.", + "E": "You just fried a total of six pancakes altogether during this cooking step." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_7", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles are there in the cabinet on the left that I just opened?", + "answer": "There are five bottles in the cabinet you just opened.", + "question_chinese": "我刚刚打开左边的柜子里面有几个瓶子？", + "answer_chinese": "你刚刚打开的柜子里有5个瓶子。", + "evidence": [ + { + "description": "There are five bottles in the cabinet you just opened.", + "timestamps": { + "start": "00:18:00", + "end": "00:18:05" + } + } + ], + "question_timestamps": "00:18:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are four bottles inside.", + "B": "There are six bottles inside.", + "C": "There are seven bottles in the cabinet you just opened on the left side.", + "D": "There are eight bottles in the cabinet on the left that you just opened.", + "E": "There are five bottles in the cabinet you just opened." + }, + "answer_label": "E", + "answer_text": "There are five bottles in the cabinet you just opened.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_7", + "correct_answer": "There are five bottles in the cabinet you just opened.", + "original_options": { + "A": "There are four bottles inside.", + "B": "There are six bottles inside.", + "C": "There are seven bottles in the cabinet you just opened on the left side.", + "D": "There are eight bottles in the cabinet on the left that you just opened.", + "E": "There are five bottles in the cabinet you just opened." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_8", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles are there in the cabinet on the left that I just opened?", + "answer": "There are six bottles in the cabinet you just opened.", + "question_chinese": "我刚刚打开左边的柜子里面有几个瓶子？", + "answer_chinese": "你刚刚打开的柜子里有6个瓶子。", + "evidence": [ + { + "description": "You just put a bottle in.", + "timestamps": { + "start": "00:35:15", + "end": "00:35:18" + } + }, + { + "description": "There were originally five bottles in the cabinet you just opened.", + "timestamps": { + "start": "00:17:58", + "end": "00:18:03" + } + } + ], + "question_timestamps": "00:35:23", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are five bottles inside.", + "B": "There are six bottles in the cabinet you just opened.", + "C": "There are seven bottles inside.", + "D": "There are four bottles in the cabinet you just opened on the left side.", + "E": "There are nine bottles in the cabinet on the left that you just opened." + }, + "answer_label": "B", + "answer_text": "There are six bottles in the cabinet you just opened.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_8", + "correct_answer": "There are six bottles in the cabinet you just opened.", + "original_options": { + "A": "There are five bottles inside.", + "B": "There are six bottles in the cabinet you just opened.", + "C": "There are seven bottles inside.", + "D": "There are four bottles in the cabinet you just opened on the left side.", + "E": "There are nine bottles in the cabinet on the left that you just opened." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_11", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pancakes did I put in the plastic bowl?", + "answer": "You just put five pancakes in a plastic bowl.", + "question_chinese": "我刚刚装了几块煎饼在塑料碗里？", + "answer_chinese": "你刚刚装了5块煎饼在塑料碗里。", + "evidence": [ + { + "description": "You have put five pancakes in a plastic bowl.", + "timestamps": { + "start": "00:28:50", + "end": "00:29:25" + } + } + ], + "question_timestamps": "00:29:27", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You put four pancakes in it.", + "B": "You put seven pancakes in it.", + "C": "You just put three pancakes in a plastic bowl after taking them off the pan.", + "D": "You just put five pancakes in a plastic bowl.", + "E": "You just put two pancakes into the plastic bowl after taking them off the pan." + }, + "answer_label": "D", + "answer_text": "You just put five pancakes in a plastic bowl.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_11", + "correct_answer": "You just put five pancakes in a plastic bowl.", + "original_options": { + "A": "You put four pancakes in it.", + "B": "You put seven pancakes in it.", + "C": "You just put three pancakes in a plastic bowl after taking them off the pan.", + "D": "You just put five pancakes in a plastic bowl.", + "E": "You just put two pancakes into the plastic bowl after taking them off the pan." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_12", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pancakes did I put in the plastic bowl?", + "answer": "You just put six pancakes in the bowl.", + "question_chinese": "我刚刚装了几块煎饼在塑料碗里？", + "answer_chinese": "你刚刚装了6块煎饼在碗里。", + "evidence": [ + { + "description": "You added another pancake to it.", + "timestamps": { + "start": "00:33:34", + "end": "00:33:41" + } + }, + { + "description": "You put five pancakes in the plastic bowl.", + "timestamps": { + "start": "00:28:48", + "end": "00:29:22" + } + } + ], + "question_timestamps": "00:34:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You put five pancakes in the bowl.", + "B": "You put seven pancakes in the bowl.", + "C": "You just put four pancakes in the bowl after removing them from the pan.", + "D": "You just put six pancakes in the bowl.", + "E": "You just put eight pancakes into the bowl after removing them from the pan." + }, + "answer_label": "D", + "answer_text": "You just put six pancakes in the bowl.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_12", + "correct_answer": "You just put six pancakes in the bowl.", + "original_options": { + "A": "You put five pancakes in the bowl.", + "B": "You put seven pancakes in the bowl.", + "C": "You just put four pancakes in the bowl after removing them from the pan.", + "D": "You just put six pancakes in the bowl.", + "E": "You just put eight pancakes into the bowl after removing them from the pan." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 14 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_7_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, how should I get from the kitchen to the living room?", + "answer": "Go out the door, turn left, and walk straight to reach the living room.", + "question_chinese": "我现在想要从厨房去到客厅应该怎么走？", + "answer_chinese": "你应该出门左转，直走即可到达客厅。", + "evidence": [ + { + "description": "You turned left after coming out of the kitchen.", + "timestamps": { + "start": "00:15:30", + "end": "00:15:32" + } + }, + { + "description": "You walked straight for a while and reached the living room.", + "timestamps": { + "start": "00:15:32", + "end": "00:15:36" + } + } + ], + "question_timestamps": "00:16:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go out the door and turn right.", + "B": "Go out the door, turn left, and walk straight to reach the living room.", + "C": "Turn left first, then go out.", + "D": "Go out the door, turn right, then walk straight to reach the living room.", + "E": "Go out the door, turn left, then make a right turn before walking straight into the living room." + }, + "answer_label": "B", + "answer_text": "Go out the door, turn left, and walk straight to reach the living room.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_7_T1", + "correct_answer": "Go out the door, turn left, and walk straight to reach the living room.", + "original_options": { + "A": "Go out the door and turn right.", + "B": "Go out the door, turn left, and walk straight to reach the living room.", + "C": "Turn left first, then go out.", + "D": "Go out the door, turn right, then walk straight to reach the living room.", + "E": "Go out the door, turn left, then make a right turn before walking straight into the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 18 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_7_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, how should I get from the kitchen to the living room?", + "answer": "First turn around to the kitchen doorway; after you go out, turn left first, then walk straight to reach the living room.", + "question_chinese": "我现在想要从厨房去到客厅应该怎么走？", + "answer_chinese": "你先向后转到厨房门口，出门后后先向左转，直走即可到达客厅。", + "evidence": [ + { + "description": "You turned left after coming out of the kitchen.", + "timestamps": { + "start": "00:15:28", + "end": "00:15:33" + } + }, + { + "description": "You walked straight to the living room.", + "timestamps": { + "start": "00:15:33", + "end": "00:15:35" + } + }, + { + "description": "The kitchen door is behind you.", + "timestamps": { + "start": "00:15:24", + "end": "00:15:28" + } + } + ], + "question_timestamps": "00:31:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right after leaving the kitchen.", + "B": "Go straight out into the living room.", + "C": "Turn around to the doorway, exit the kitchen, then turn right and walk straight to the living room.", + "D": "Face the kitchen doorway, walk out, turn left twice, and continue straight until you enter the living room.", + "E": "First turn around to the kitchen doorway; after you go out, turn left first, then walk straight to reach the living room." + }, + "answer_label": "E", + "answer_text": "First turn around to the kitchen doorway; after you go out, turn left first, then walk straight to reach the living room.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_7_T2", + "correct_answer": "First turn around to the kitchen doorway; after you go out, turn left first, then walk straight to reach the living room.", + "original_options": { + "A": "Turn right after leaving the kitchen.", + "B": "Go straight out into the living room.", + "C": "Turn around to the doorway, exit the kitchen, then turn right and walk straight to the living room.", + "D": "Face the kitchen doorway, walk out, turn left twice, and continue straight until you enter the living room.", + "E": "First turn around to the kitchen doorway; after you go out, turn left first, then walk straight to reach the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_15", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the faucet in the kitchen?", + "answer": "You can't touch the faucet directly.", + "question_chinese": "我能直接触碰到厨房的水龙头吗？", + "answer_chinese": "你不能直接触碰到水龙头。", + "evidence": [ + { + "description": "The faucet is on the left of your current position, about two steps away.", + "timestamps": { + "start": "00:17:20", + "end": "00:17:30" + } + } + ], + "question_timestamps": "00:24:35", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch the faucet directly.", + "B": "You can't touch the faucet directly." + }, + "answer_label": "B", + "answer_text": "You can't touch the faucet directly.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_15", + "correct_answer": "You can't touch the faucet directly.", + "original_options": { + "A": "You can touch the faucet directly.", + "B": "You can't touch the faucet directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_16", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the faucet in the kitchen?", + "answer": "You can't touch the faucet directly.", + "question_chinese": "我能直接触碰到厨房的水龙头吗？", + "answer_chinese": "你无法直接触碰到水龙头。", + "evidence": [ + { + "description": "The faucet is about three steps away from your location.", + "timestamps": { + "start": "00:28:25", + "end": "00:28:27" + } + } + ], + "question_timestamps": "00:28:28", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't touch the faucet directly.", + "B": "You can touch the faucet directly." + }, + "answer_label": "A", + "answer_text": "You can't touch the faucet directly.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_16", + "correct_answer": "You can't touch the faucet directly.", + "original_options": { + "A": "You can't touch the faucet directly.", + "B": "You can touch the faucet directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_8_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the kitchen faucet?", + "answer": "You can't directly reach the faucet.", + "question_chinese": "我能直接触碰到厨房的水龙头吗？", + "answer_chinese": "你不能直接触碰到水龙头。", + "evidence": [ + { + "description": "The faucet is on the left of your current position, about two steps away.", + "timestamps": { + "start": "00:17:20", + "end": "00:17:30" + } + } + ], + "question_timestamps": "00:28:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can directly reach the faucet.", + "B": "You can't directly reach the faucet." + }, + "answer_label": "B", + "answer_text": "You can't directly reach the faucet.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_8_T1", + "correct_answer": "You can't directly reach the faucet.", + "original_options": { + "A": "You can directly reach the faucet.", + "B": "You can't directly reach the faucet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_8_T2", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the kitchen faucet?", + "answer": "You can directly reach the faucet.", + "question_chinese": "我能直接触碰到厨房的水龙头吗？", + "answer_chinese": "你可以直接触碰到水龙头。", + "evidence": [ + { + "description": "The faucet is about one steps away from your location.", + "timestamps": { + "start": "00:35:00", + "end": "00:35:04" + } + } + ], + "question_timestamps": "00:35:06", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can directly reach the faucet.", + "B": "You can't directly reach the faucet." + }, + "answer_label": "A", + "answer_text": "You can directly reach the faucet.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_8_T2", + "correct_answer": "You can directly reach the faucet.", + "original_options": { + "A": "You can directly reach the faucet.", + "B": "You can't directly reach the faucet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_1", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the fork on the table?", + "answer": "You can touch the fork directly.", + "question_chinese": "我能直接触碰到桌上的叉子吗？", + "answer_chinese": "你能直接触碰到叉子。", + "evidence": [ + { + "description": "You put the fork in your hand on the plate on the table.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch the fork directly.", + "B": "You cannot touch the fork directly." + }, + "answer_label": "A", + "answer_text": "You can touch the fork directly.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_1", + "correct_answer": "You can touch the fork directly.", + "original_options": { + "A": "You can touch the fork directly.", + "B": "You cannot touch the fork directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_2", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the fork on the table?", + "answer": "You can directly touch the fork on the table.", + "question_chinese": "我能直接触碰到桌上的叉子吗？", + "answer_chinese": "你能直接触碰到桌上的叉子。", + "evidence": [ + { + "description": "You put the fork in your hand on the plate on the table.", + "timestamps": { + "start": "00:11:56", + "end": "00:11:58" + } + } + ], + "question_timestamps": "00:11:59", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can directly touch the fork on the table.", + "B": "You cannot directly touch the fork on the table." + }, + "answer_label": "A", + "answer_text": "You can directly touch the fork on the table.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_2", + "correct_answer": "You can directly touch the fork on the table.", + "original_options": { + "A": "You can directly touch the fork on the table.", + "B": "You cannot directly touch the fork on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_19", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the sofa on my left or right now?", + "answer": "The sofa is on your right.", + "question_chinese": "沙发现在在我的左侧还是右侧？", + "answer_chinese": "沙发在你的右侧。", + "evidence": [ + { + "description": "The sofa is on your right.", + "timestamps": { + "start": "00:15:45", + "end": "00:15:49" + } + } + ], + "question_timestamps": "00:15:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The sofa is on your right.", + "B": "The sofa is on your left." + }, + "answer_label": "A", + "answer_text": "The sofa is on your right.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_19", + "correct_answer": "The sofa is on your right.", + "original_options": { + "A": "The sofa is on your right.", + "B": "The sofa is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_10_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the sofa on the left side of me or on the right side of me right now?", + "answer": "The sofa is on the right side of you.", + "question_chinese": "沙发现在在我的左侧还是右侧？", + "answer_chinese": "沙发在你的右侧。", + "evidence": [ + { + "description": "The sofa is on your right.", + "timestamps": { + "start": "00:15:45", + "end": "00:15:49" + } + } + ], + "question_timestamps": "00:15:51", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sofa is on the right side of you.", + "B": "The sofa is on the left side of you." + }, + "answer_label": "A", + "answer_text": "The sofa is on the right side of you.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_10_T1", + "correct_answer": "The sofa is on the right side of you.", + "original_options": { + "A": "The sofa is on the right side of you.", + "B": "The sofa is on the left side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_10_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the sofa on the left side of me or on the right side of me right now?", + "answer": "The sofa is on the left side of you.", + "question_chinese": "沙发现在在我的左侧还是右侧？", + "answer_chinese": "沙发在你的左侧。", + "evidence": [ + { + "description": "The sofa is on your right.", + "timestamps": { + "start": "00:15:45", + "end": "00:15:49" + } + }, + { + "description": "The sofa is on your left.", + "timestamps": { + "start": "00:16:11", + "end": "00:16:13" + } + } + ], + "question_timestamps": "00:16:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sofa is on the left side of you.", + "B": "The sofa is on the right side of you." + }, + "answer_label": "A", + "answer_text": "The sofa is on the left side of you.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_10_T2", + "correct_answer": "The sofa is on the left side of you.", + "original_options": { + "A": "The sofa is on the left side of you.", + "B": "The sofa is on the right side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_9_T1", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Besides what I’m holding in my hand, what is the closest item to me on the table?", + "answer": "The closest item to you is the fork.", + "question_chinese": "除了我手上拿着的东西以外，桌子上距离我最近的物品是什么？", + "answer_chinese": "距离你最近的物品是叉子。", + "evidence": [ + { + "description": "The nearest item to you is the fork.", + "timestamps": { + "start": "00:34:49", + "end": "00:34:50" + } + } + ], + "question_timestamps": "00:34:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The lunch box.", + "B": "The spoon.", + "C": "The closest item to you is the fork.", + "D": "The closest item to you on the table is the lunch box.", + "E": "Besides what you're holding, the spoon is the nearest item to you on the table." + }, + "answer_label": "C", + "answer_text": "The closest item to you is the fork.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_9_T1", + "correct_answer": "The closest item to you is the fork.", + "original_options": { + "A": "The lunch box.", + "B": "The spoon.", + "C": "The closest item to you is the fork.", + "D": "The closest item to you on the table is the lunch box.", + "E": "Besides what you're holding, the spoon is the nearest item to you on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 12, + 15 + ] + }, + { + "video_uid": "0a01978c-e16d-4587-95f1-49efa3ab15d9", + "video_length": "00:37:16", + "qid": "0a01978c-e16d-4587-95f1-49efa3ab15d9_9_T2", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Besides what I’m holding in my hand, what is the closest item to me on the table?", + "answer": "The closest item to you is the lunch box.", + "question_chinese": "除了我手上拿着的东西以外，桌子上距离我最近的物品是什么？", + "answer_chinese": "距离你最近的物品是饭盒。", + "evidence": [ + { + "description": "The nearest item to you is the fork.", + "timestamps": { + "start": "00:34:49", + "end": "00:34:50" + } + }, + { + "description": "The nearest item to you is the lunch box.", + "timestamps": { + "start": "00:35:39", + "end": "00:35:45" + } + } + ], + "question_timestamps": "00:35:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fork.", + "B": "The spoon.", + "C": "Besides what you're holding, the fork is the closest item on the table to you.", + "D": "The closest item to you is the lunch box.", + "E": "Besides what you're holding in your hand, the spoon is the closest item on the table." + }, + "answer_label": "D", + "answer_text": "The closest item to you is the lunch box.", + "q_id": "0a01978c-e16d-4587-95f1-49efa3ab15d9_9_T2", + "correct_answer": "The closest item to you is the lunch box.", + "original_options": { + "A": "The fork.", + "B": "The spoon.", + "C": "Besides what you're holding, the fork is the closest item on the table to you.", + "D": "The closest item to you is the lunch box.", + "E": "Besides what you're holding in your hand, the spoon is the closest item on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + } +] diff --git a/QAs/0a569c6b-2818-410c-8eff-9b0290b2404c-mcq.json b/QAs/0a569c6b-2818-410c-8eff-9b0290b2404c-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1e43bf7464b46c21e7c4ae7a374e3a46f50fc6e5 --- /dev/null +++ b/QAs/0a569c6b-2818-410c-8eff-9b0290b2404c-mcq.json @@ -0,0 +1,931 @@ +[ + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_6", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the red paper fish from the scissors when I'm doing origami?", + "answer": "When you are folding paper, the red paper fish is thirty centimeters away from the scissors.", + "question_chinese": "我在折纸时，红色纸鱼距离剪刀有多远？", + "answer_chinese": "你在折纸时，红色纸鱼距离剪刀三十厘米。", + "evidence": [ + { + "description": "The red paper fish is on my left, and the scissors are on my right.", + "timestamps": { + "start": "00:04:31", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:05:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Ten centimeters away.", + "B": "Fifty centimeters away.", + "C": "When you are folding paper, the red paper fish is ten centimeters away from the scissors.", + "D": "When you are folding paper, the red paper fish is seventy centimeters away from the scissors.", + "E": "When you are folding paper, the red paper fish is thirty centimeters away from the scissors." + }, + "answer_label": "E", + "answer_text": "When you are folding paper, the red paper fish is thirty centimeters away from the scissors.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_6", + "correct_answer": "When you are folding paper, the red paper fish is thirty centimeters away from the scissors.", + "original_options": { + "A": "Ten centimeters away.", + "B": "Fifty centimeters away.", + "C": "When you are folding paper, the red paper fish is ten centimeters away from the scissors.", + "D": "When you are folding paper, the red paper fish is seventy centimeters away from the scissors.", + "E": "When you are folding paper, the red paper fish is thirty centimeters away from the scissors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_7", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the red paper fish from the scissors when I'm doing origami?", + "answer": "When folding paper, the red paper fish is 50 centimeters away from the scissors in a straight line.", + "question_chinese": "我在折纸时，红色纸鱼距离剪刀有多远？", + "answer_chinese": "你在折纸时，红色纸鱼距离剪刀直线距离50厘米。", + "evidence": [ + { + "description": "Red paper fish in my hand, scissors in front of me.", + "timestamps": { + "start": "00:22:03", + "end": "00:22:13" + } + } + ], + "question_timestamps": "00:22:15", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "47 centimeters away.", + "B": "53 centimeters away.", + "C": "When folding paper, the red paper fish is 50 centimeters away from the scissors in a straight line.", + "D": "When folding paper, the red paper fish is 49 centimeters away from the scissors in a straight line.", + "E": "When folding paper, the red paper fish is 51 centimeters away from the scissors in a straight line." + }, + "answer_label": "C", + "answer_text": "When folding paper, the red paper fish is 50 centimeters away from the scissors in a straight line.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_7", + "correct_answer": "When folding paper, the red paper fish is 50 centimeters away from the scissors in a straight line.", + "original_options": { + "A": "47 centimeters away.", + "B": "53 centimeters away.", + "C": "When folding paper, the red paper fish is 50 centimeters away from the scissors in a straight line.", + "D": "When folding paper, the red paper fish is 49 centimeters away from the scissors in a straight line.", + "E": "When folding paper, the red paper fish is 51 centimeters away from the scissors in a straight line." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 18 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_9", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I want to make a paper fish model, how many usable paper fish are there now?", + "answer": "The currently usable paper fish are eight.", + "question_chinese": "我想做纸鱼模型，现在能用的纸鱼有几个？", + "answer_chinese": "现在能用的纸鱼有八个。", + "evidence": [ + { + "description": "There are four uncut paper fish and four cut ones.", + "timestamps": { + "start": "00:01:45", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Nine paper fish.", + "B": "Ten paper fish.", + "C": "The currently usable paper fish are eleven in total, ready to use right now.", + "D": "The currently usable paper fish are eight.", + "E": "The currently usable paper fish are twelve at the moment, ready to use." + }, + "answer_label": "D", + "answer_text": "The currently usable paper fish are eight.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_9", + "correct_answer": "The currently usable paper fish are eight.", + "original_options": { + "A": "Nine paper fish.", + "B": "Ten paper fish.", + "C": "The currently usable paper fish are eleven in total, ready to use right now.", + "D": "The currently usable paper fish are eight.", + "E": "The currently usable paper fish are twelve at the moment, ready to use." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 13 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_8", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I want to make a paper fish model, how many usable paper fish are there now?", + "answer": "The currently usable paper fish are six.", + "question_chinese": "我想做纸鱼模型，现在能用的纸鱼有几个？", + "answer_chinese": "现在能用的纸鱼有六个。", + "evidence": [ + { + "description": "There are four uncut paper fish and four cut ones.", + "timestamps": { + "start": "00:01:45", + "end": "00:02:00" + } + }, + { + "description": "I've already glued a paper fish, and there are seven left.", + "timestamps": { + "start": "00:09:50", + "end": "00:10:02" + } + } + ], + "question_timestamps": "00:19:34", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three paper fish.", + "B": "Four paper fish.", + "C": "The currently usable paper fish are six.", + "D": "The currently usable paper fish are seven in total, ready to use right now.", + "E": "The currently usable paper fish are eight at the moment, ready to use." + }, + "answer_label": "C", + "answer_text": "The currently usable paper fish are six.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_8", + "correct_answer": "The currently usable paper fish are six.", + "original_options": { + "A": "Three paper fish.", + "B": "Four paper fish.", + "C": "The currently usable paper fish are six.", + "D": "The currently usable paper fish are seven in total, ready to use right now.", + "E": "The currently usable paper fish are eight at the moment, ready to use." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 13 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_7_T3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is on my blue mat?", + "answer": "There are some paper scraps on your blue mat, and the blue mat is directly in front of you.", + "question_chinese": "我的蓝色垫子上有什么？", + "answer_chinese": "你的蓝色垫子上有一些纸屑，蓝色垫子在你的正前方。", + "evidence": [ + { + "description": "There are some scraps of paper on your blue mat.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There are some paper scraps on your blue mat, and the blue mat is directly in front of you.", + "B": "Only the blue mat is there.", + "C": "A red paper goldfish is on it.", + "D": "There is a pair of scissors on your blue mat, with the mat sitting directly behind you.", + "E": "There are paper scraps scattered on the blue mat, but the mat is positioned off to your left." + }, + "answer_label": "A", + "answer_text": "There are some paper scraps on your blue mat, and the blue mat is directly in front of you.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_7_T3", + "correct_answer": "There are some paper scraps on your blue mat, and the blue mat is directly in front of you.", + "original_options": { + "A": "There are some paper scraps on your blue mat, and the blue mat is directly in front of you.", + "B": "Only the blue mat is there.", + "C": "A red paper goldfish is on it.", + "D": "There is a pair of scissors on your blue mat, with the mat sitting directly behind you.", + "E": "There are paper scraps scattered on the blue mat, but the mat is positioned off to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 18 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_7_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is on my blue mat?", + "answer": "There are a pair of scissors and a carving knife on your blue mat, and the blue mat is to your right.", + "question_chinese": "我的蓝色垫子上有什么？", + "answer_chinese": "你的蓝色垫子上有一把剪刀和一把雕刻刀,蓝色垫子在你的右边。", + "evidence": [ + { + "description": "On your blue mat are a pair of scissors and an engraving knife.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There are a pair of scissors and a carving knife on your blue mat, and the blue mat is to your right.", + "B": "Only scissors are on the mat.", + "C": "Only a carving knife is on it.", + "D": "There are scissors and a carving knife on your blue mat, but the mat is in front of you.", + "E": "There are scissors and a carving knife on your blue mat, and the mat is positioned to your left." + }, + "answer_label": "A", + "answer_text": "There are a pair of scissors and a carving knife on your blue mat, and the blue mat is to your right.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_7_T1", + "correct_answer": "There are a pair of scissors and a carving knife on your blue mat, and the blue mat is to your right.", + "original_options": { + "A": "There are a pair of scissors and a carving knife on your blue mat, and the blue mat is to your right.", + "B": "Only scissors are on the mat.", + "C": "Only a carving knife is on it.", + "D": "There are scissors and a carving knife on your blue mat, but the mat is in front of you.", + "E": "There are scissors and a carving knife on your blue mat, and the mat is positioned to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 19 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_7_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is on my blue mat?", + "answer": "There is a red paper goldfish on your blue mat, and the blue mat is under your right hand.", + "question_chinese": "我的蓝色垫子上有什么？", + "answer_chinese": "你的蓝色垫子上有一只红色的纸金鱼，蓝色垫子在你的右手下。", + "evidence": [ + { + "description": "There is a red paper goldfish on your blue mat.", + "timestamps": { + "start": "00:01:17", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Paper scraps are on your blue mat.", + "B": "A pair of scissors is on your mat.", + "C": "There is a red paper goldfish on your blue mat, and the blue mat is under your right hand.", + "D": "There is a red paper goldfish on your blue mat, and the blue mat is under your left hand.", + "E": "There is a red paper goldfish on your blue mat, but the blue mat is directly in front of you." + }, + "answer_label": "C", + "answer_text": "There is a red paper goldfish on your blue mat, and the blue mat is under your right hand.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_7_T2", + "correct_answer": "There is a red paper goldfish on your blue mat, and the blue mat is under your right hand.", + "original_options": { + "A": "Paper scraps are on your blue mat.", + "B": "A pair of scissors is on your mat.", + "C": "There is a red paper goldfish on your blue mat, and the blue mat is under your right hand.", + "D": "There is a red paper goldfish on your blue mat, and the blue mat is under your left hand.", + "E": "There is a red paper goldfish on your blue mat, but the blue mat is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 20 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_13", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "I want to put the ruler back in place, what should I do?", + "answer": "You should place the ruler on the book to your left.", + "question_chinese": "我要把尺子放回原位，应该怎么做？", + "answer_chinese": "你要把尺子放到你左侧的书本上。", + "evidence": [ + { + "description": "The ruler was originally placed on the book on the left.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Place the ruler on the desk.", + "B": "Put the ruler on the shelf.", + "C": "You should place the ruler back on the table behind you, where it was resting earlier.", + "D": "You should place the ruler on the book to your left.", + "E": "You should set the ruler down on the floor beside you to return it to its spot." + }, + "answer_label": "D", + "answer_text": "You should place the ruler on the book to your left.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_13", + "correct_answer": "You should place the ruler on the book to your left.", + "original_options": { + "A": "Place the ruler on the desk.", + "B": "Put the ruler on the shelf.", + "C": "You should place the ruler back on the table behind you, where it was resting earlier.", + "D": "You should place the ruler on the book to your left.", + "E": "You should set the ruler down on the floor beside you to return it to its spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_14", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "I want to put the ruler back in place, what should I do?", + "answer": "You should place the ruler on the book to your left.", + "question_chinese": "我要把尺子放回原位，应该怎么做？", + "answer_chinese": "你要把尺子放到你左侧的书本上。", + "evidence": [ + { + "description": "The ruler was originally placed on the book on the left.", + "timestamps": { + "start": "00:21:40", + "end": "00:22:30" + } + } + ], + "question_timestamps": "00:22:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Place the ruler on the paper.", + "B": "You should place the ruler on the book to your left.", + "C": "Put the ruler on the right book.", + "D": "You should place the ruler on the desk to your left instead of returning it to the book.", + "E": "You should place the ruler on the shelf to your left, not on the book where it belongs." + }, + "answer_label": "B", + "answer_text": "You should place the ruler on the book to your left.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_14", + "correct_answer": "You should place the ruler on the book to your left.", + "original_options": { + "A": "Place the ruler on the paper.", + "B": "You should place the ruler on the book to your left.", + "C": "Put the ruler on the right book.", + "D": "You should place the ruler on the desk to your left instead of returning it to the book.", + "E": "You should place the ruler on the shelf to your left, not on the book where it belongs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_5", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Now is the scissors in front/left/right of me?", + "answer": "Now the scissors are on your right.", + "question_chinese": "现在剪刀在我的前/左/右？", + "answer_chinese": "现在剪刀在你的右边。", + "evidence": [ + { + "description": "Scissors are placed on the right side and have not been used yet.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:00:59", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Now the scissors are on your right.", + "B": "Now the scissors are in front of you.", + "C": "Now the scissors are on your left.", + "D": "Now the scissors are directly behind you, not positioned on your right side anymore.", + "E": "Now the scissors sit in front of you on the work surface rather than to your right." + }, + "answer_label": "A", + "answer_text": "Now the scissors are on your right.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_5", + "correct_answer": "Now the scissors are on your right.", + "original_options": { + "A": "Now the scissors are on your right.", + "B": "Now the scissors are in front of you.", + "C": "Now the scissors are on your left.", + "D": "Now the scissors are directly behind you, not positioned on your right side anymore.", + "E": "Now the scissors sit in front of you on the work surface rather than to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 17 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_12", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Now is the scissors in front/left/right of me?", + "answer": "The scissors are in front of you now.", + "question_chinese": "现在剪刀在我的前/左/右？", + "answer_chinese": "现在剪刀在你的前面。", + "evidence": [ + { + "description": "Scissors are placed on the right side and have not been used yet.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:58" + } + }, + { + "description": "After using the scissor, I placed it in front of me.", + "timestamps": { + "start": "00:12:35", + "end": "00:13:50" + } + } + ], + "question_timestamps": "00:13:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The scissors are in front of you now.", + "B": "The scissors are to your left now.", + "C": "The scissors are to your right now.", + "D": "The scissors are behind you now, not directly in front of you as you face forward.", + "E": "The scissors are above you now, so you’d need to look up to spot them." + }, + "answer_label": "A", + "answer_text": "The scissors are in front of you now.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_12", + "correct_answer": "The scissors are in front of you now.", + "original_options": { + "A": "The scissors are in front of you now.", + "B": "The scissors are to your left now.", + "C": "The scissors are to your right now.", + "D": "The scissors are behind you now, not directly in front of you as you face forward.", + "E": "The scissors are above you now, so you’d need to look up to spot them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_3", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many paper fish do I have in my hand now?", + "answer": "Now you have a paper fish in your hand.", + "question_chinese": "现在我的手上有几个纸鱼？", + "answer_chinese": "现在你的手上有一个纸鱼。", + "evidence": [ + { + "description": "I'm gluing the first paper fish.", + "timestamps": { + "start": "00:06:24", + "end": "00:06:48" + } + } + ], + "question_timestamps": "00:06:48", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "No paper fish.", + "B": "Two paper fish.", + "C": "Now you have a paper fish in your hand.", + "D": "Now you have three paper fish in your hand, held together as a small stack.", + "E": "Now you have four paper fish in your hand, all gathered and gripped at once." + }, + "answer_label": "C", + "answer_text": "Now you have a paper fish in your hand.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_3", + "correct_answer": "Now you have a paper fish in your hand.", + "original_options": { + "A": "No paper fish.", + "B": "Two paper fish.", + "C": "Now you have a paper fish in your hand.", + "D": "Now you have three paper fish in your hand, held together as a small stack.", + "E": "Now you have four paper fish in your hand, all gathered and gripped at once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_4", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many paper fish do I have in my hand now?", + "answer": "Now you have two paper fish in your hands.", + "question_chinese": "现在我的手上有几个纸鱼？", + "answer_chinese": "现在你的手上有两个纸鱼。", + "evidence": [ + { + "description": "I'm gluing the first paper fish.", + "timestamps": { + "start": "00:06:24", + "end": "00:06:48" + } + }, + { + "description": "I'm gluing the second paper fish.", + "timestamps": { + "start": "00:15:10", + "end": "00:15:46" + } + } + ], + "question_timestamps": "00:16:12", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Now you have one paper fish.", + "B": "Now you have three paper fish.", + "C": "Now you have four paper fish in your hands right now, after picking them up.", + "D": "Now you have two paper fish in your hands.", + "E": "Now you have five paper fish in your hands, holding them together before doing anything else." + }, + "answer_label": "D", + "answer_text": "Now you have two paper fish in your hands.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_4", + "correct_answer": "Now you have two paper fish in your hands.", + "original_options": { + "A": "Now you have one paper fish.", + "B": "Now you have three paper fish.", + "C": "Now you have four paper fish in your hands right now, after picking them up.", + "D": "Now you have two paper fish in your hands.", + "E": "Now you have five paper fish in your hands, holding them together before doing anything else." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_10", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many uncut paper fish are there now?", + "answer": "There are four uncut paper fish now.", + "question_chinese": "现在有几个没有剪的纸鱼？", + "answer_chinese": "现在有四个没有剪的纸鱼。", + "evidence": [ + { + "description": "My left hand is holding four uncut paper fish.", + "timestamps": { + "start": "00:01:43", + "end": "00:01:53" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three uncut paper fish now.", + "B": "There are four uncut paper fish now.", + "C": "There are five uncut paper fish now.", + "D": "There are six uncut paper fish now, with the rest already cut and set aside.", + "E": "There are seven uncut paper fish now, while the remaining ones have been trimmed already." + }, + "answer_label": "B", + "answer_text": "There are four uncut paper fish now.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_10", + "correct_answer": "There are four uncut paper fish now.", + "original_options": { + "A": "There are three uncut paper fish now.", + "B": "There are four uncut paper fish now.", + "C": "There are five uncut paper fish now.", + "D": "There are six uncut paper fish now, with the rest already cut and set aside.", + "E": "There are seven uncut paper fish now, while the remaining ones have been trimmed already." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_11", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many uncut paper fish are there now?", + "answer": "There are three uncut paper fish now.", + "question_chinese": "现在有几个没有剪的纸鱼？", + "answer_chinese": "现在有三个没有剪的纸鱼。", + "evidence": [ + { + "description": "I cut out a paper fish, leaving three uncut.", + "timestamps": { + "start": "00:12:27", + "end": "00:13:48" + } + } + ], + "question_timestamps": "00:13:48", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three uncut paper fish now.", + "B": "There are two uncut paper fish now.", + "C": "There are four uncut paper fish now.", + "D": "There are five uncut paper fish now, and only a few have been cut so far.", + "E": "There are seven uncut paper fish now, leaving fewer fish that have been cut already." + }, + "answer_label": "A", + "answer_text": "There are three uncut paper fish now.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_11", + "correct_answer": "There are three uncut paper fish now.", + "original_options": { + "A": "There are three uncut paper fish now.", + "B": "There are two uncut paper fish now.", + "C": "There are four uncut paper fish now.", + "D": "There are five uncut paper fish now, and only a few have been cut so far.", + "E": "There are seven uncut paper fish now, leaving fewer fish that have been cut already." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 15 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the ruler so far?", + "answer": "So far you have used the ruler once.", + "question_chinese": "目前为止我使用了几次尺子？", + "answer_chinese": "目前为止你使用了一次尺子。", + "evidence": [ + { + "description": "I used a ruler for the first time.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:51" + } + } + ], + "question_timestamps": "00:00:51", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have used the ruler twice.", + "B": "So far you have used the ruler once.", + "C": "You have used the ruler three times.", + "D": "So far you have used the ruler four times, lining it up repeatedly to measure your paper pieces.", + "E": "So far you have used the ruler five times, using it again each time you checked the length." + }, + "answer_label": "B", + "answer_text": "So far you have used the ruler once.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_1", + "correct_answer": "So far you have used the ruler once.", + "original_options": { + "A": "You have used the ruler twice.", + "B": "So far you have used the ruler once.", + "C": "You have used the ruler three times.", + "D": "So far you have used the ruler four times, lining it up repeatedly to measure your paper pieces.", + "E": "So far you have used the ruler five times, using it again each time you checked the length." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + }, + { + "video_uid": "0a569c6b-2818-410c-8eff-9b0290b2404c", + "video_length": "00:23:18", + "qid": "0a569c6b-2818-410c-8eff-9b0290b2404c_2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the ruler so far?", + "answer": "So far you have used the ruler twice.", + "question_chinese": "目前为止我使用了几次尺子？", + "answer_chinese": "目前为止你使用了两次尺子。", + "evidence": [ + { + "description": "This is the first time using a ruler.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:51" + } + }, + { + "description": "This is the second time using a ruler.", + "timestamps": { + "start": "00:20:23", + "end": "00:21:54" + } + } + ], + "question_timestamps": "00:21:54", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "So far you have used the ruler twice.", + "B": "You have used the ruler once.", + "C": "You have used the ruler three times.", + "D": "So far you have used the ruler four times, measuring and rechecking the spacing along the paper.", + "E": "So far you have used the ruler five times, putting it down and picking it up for repeated measurements." + }, + "answer_label": "A", + "answer_text": "So far you have used the ruler twice.", + "q_id": "0a569c6b-2818-410c-8eff-9b0290b2404c_2", + "correct_answer": "So far you have used the ruler twice.", + "original_options": { + "A": "So far you have used the ruler twice.", + "B": "You have used the ruler once.", + "C": "You have used the ruler three times.", + "D": "So far you have used the ruler four times, measuring and rechecking the spacing along the paper.", + "E": "So far you have used the ruler five times, putting it down and picking it up for repeated measurements." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 19 + ] + } +] diff --git a/QAs/0b637191-00b7-4fd8-af37-d26d80dc59ce-mcq.json b/QAs/0b637191-00b7-4fd8-af37-d26d80dc59ce-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6f66a0bd83ee2c7f34caa4b5ff19739b7d2656c1 --- /dev/null +++ b/QAs/0b637191-00b7-4fd8-af37-d26d80dc59ce-mcq.json @@ -0,0 +1,1168 @@ +[ + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_6_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to check out, how should I get there?", + "answer": "You should turn left and then go straight to the checkout counter.", + "question_chinese": "如果我想结账，我应该怎么走呢？", + "answer_chinese": "你应该左转然后直行到收银台。", + "evidence": [ + { + "description": "You went to the checkout counter to pay.", + "timestamps": { + "start": "00:20:28", + "end": "00:21:29" + } + } + ], + "question_timestamps": "00:21:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight.", + "B": "Go straight without turning.", + "C": "You should turn left and then go straight to the checkout counter.", + "D": "You should turn right first, then walk straight ahead until you reach the checkout counter.", + "E": "You should turn left, pass a shelf, then turn right to arrive at the checkout counter." + }, + "answer_label": "C", + "answer_text": "You should turn left and then go straight to the checkout counter.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_6_T1", + "correct_answer": "You should turn left and then go straight to the checkout counter.", + "original_options": { + "A": "Turn right, then go straight.", + "B": "Go straight without turning.", + "C": "You should turn left and then go straight to the checkout counter.", + "D": "You should turn right first, then walk straight ahead until you reach the checkout counter.", + "E": "You should turn left, pass a shelf, then turn right to arrive at the checkout counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 16 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_6_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to check out, how should I get there?", + "answer": "You should turn around, go straight, and then turn left to the checkout counter.", + "question_chinese": "如果我想结账，我应该怎么走呢？", + "answer_chinese": "你应该向后转直行然后左转到收银台。", + "evidence": [ + { + "description": "You went to the checkout counter to pay.", + "timestamps": { + "start": "00:20:28", + "end": "00:21:29" + } + } + ], + "question_timestamps": "00:22:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around, then turn right.", + "B": "Turn left, then go straight.", + "C": "You should turn around, walk straight ahead, and then turn right to reach the checkout counter.", + "D": "You should turn around, go straight past a shelf, then turn right toward the checkout counter.", + "E": "You should turn around, go straight, and then turn left to the checkout counter." + }, + "answer_label": "E", + "answer_text": "You should turn around, go straight, and then turn left to the checkout counter.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_6_T2", + "correct_answer": "You should turn around, go straight, and then turn left to the checkout counter.", + "original_options": { + "A": "Turn around, then turn right.", + "B": "Turn left, then go straight.", + "C": "You should turn around, walk straight ahead, and then turn right to reach the checkout counter.", + "D": "You should turn around, go straight past a shelf, then turn right toward the checkout counter.", + "E": "You should turn around, go straight, and then turn left to the checkout counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_13", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many snacks did I buy in total?", + "answer": "In total, you purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "question_chinese": "我一共买了多少食物呢？", + "answer_chinese": "你一共买了七包零食，两杯饮料，两个锡纸包的食物。", + "evidence": [ + { + "description": "You purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food on the table.", + "timestamps": { + "start": "00:15:10", + "end": "00:15:18" + } + } + ], + "question_timestamps": "00:18:36", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You bought six packs of snacks.", + "B": "You bought eight packs of snacks.", + "C": "In total, you purchased six packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "D": "In total, you purchased eight packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "E": "In total, you purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food." + }, + "answer_label": "E", + "answer_text": "In total, you purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_13", + "correct_answer": "In total, you purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "original_options": { + "A": "You bought six packs of snacks.", + "B": "You bought eight packs of snacks.", + "C": "In total, you purchased six packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "D": "In total, you purchased eight packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "E": "In total, you purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_14", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many snacks did I buy in total?", + "answer": "In total, you purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "question_chinese": "我一共买了多少食物呢？", + "answer_chinese": "你一共买了七包零食，两杯饮料，两个锡纸包的食物。", + "evidence": [ + { + "description": "You purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food on the table.", + "timestamps": { + "start": "00:15:10", + "end": "00:15:18" + } + } + ], + "question_timestamps": "00:22:42", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Six snack packs total.", + "B": "Eight snack packs total.", + "C": "In total, you purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "D": "In total, you purchased six packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "E": "In total, you purchased seven packs of snacks, two beverages, and three pieces of foil-wrapped food." + }, + "answer_label": "C", + "answer_text": "In total, you purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_14", + "correct_answer": "In total, you purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "original_options": { + "A": "Six snack packs total.", + "B": "Eight snack packs total.", + "C": "In total, you purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "D": "In total, you purchased six packs of snacks, two beverages, and two pieces of foil-wrapped food.", + "E": "In total, you purchased seven packs of snacks, two beverages, and three pieces of foil-wrapped food." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_3_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How can I get to the beverage area?", + "answer": "After you entered the store, you walked straight ahead and then turned left.", + "question_chinese": "我怎样才能到饮料区？", + "answer_chinese": "你进入商店直走然后左转。", + "evidence": [ + { + "description": "You are now at the entrance of the store.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk straight, then turn right.", + "B": "After you entered the store, you walked straight ahead and then turned left.", + "C": "Turn left immediately after entering.", + "D": "After you entered the store, you turned left first, then walked straight ahead to the beverage area.", + "E": "After you entered the store, you walked straight ahead, passed one aisle, then turned right for beverages." + }, + "answer_label": "B", + "answer_text": "After you entered the store, you walked straight ahead and then turned left.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_3_T1", + "correct_answer": "After you entered the store, you walked straight ahead and then turned left.", + "original_options": { + "A": "Walk straight, then turn right.", + "B": "After you entered the store, you walked straight ahead and then turned left.", + "C": "Turn left immediately after entering.", + "D": "After you entered the store, you turned left first, then walked straight ahead to the beverage area.", + "E": "After you entered the store, you walked straight ahead, passed one aisle, then turned right for beverages." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_3_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How can I get to the beverage area?", + "answer": "Turn left now, then go straight to reach the drinks section.", + "question_chinese": "我怎样才能到饮料区？", + "answer_chinese": "你现在左转然后直走就可以到饮料区。", + "evidence": [ + { + "description": "You are now in the snack area.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:07:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left now, then go straight to reach the drinks section.", + "B": "Turn right now, then go straight.", + "C": "Go straight first, then turn left.", + "D": "Turn left now, then take the first right to reach the drinks section at the end.", + "E": "Turn right now, then keep walking straight past the next aisle to reach the drinks section." + }, + "answer_label": "A", + "answer_text": "Turn left now, then go straight to reach the drinks section.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_3_T2", + "correct_answer": "Turn left now, then go straight to reach the drinks section.", + "original_options": { + "A": "Turn left now, then go straight to reach the drinks section.", + "B": "Turn right now, then go straight.", + "C": "Go straight first, then turn left.", + "D": "Turn left now, then take the first right to reach the drinks section at the end.", + "E": "Turn right now, then keep walking straight past the next aisle to reach the drinks section." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_5", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to buy drinks. Where are they located in the mall?", + "answer": "The beverage section is on the left side of the mall entrance.", + "question_chinese": "我想去买饮料，它在商场的哪个位置呢？", + "answer_chinese": "饮料区在商场入口处的左侧。", + "evidence": [ + { + "description": "The beverage section is on the left side of the mall entrance.", + "timestamps": { + "start": "00:04:35", + "end": "00:04:44" + } + } + ], + "question_timestamps": "00:08:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the right side of the entrance.", + "B": "The beverage section is on the left side of the mall entrance.", + "C": "Straight ahead from the entrance.", + "D": "The beverage section is across from the mall entrance, on the opposite side from where you came in.", + "E": "The beverage section is upstairs near the mall entrance, not on the ground level where you entered." + }, + "answer_label": "B", + "answer_text": "The beverage section is on the left side of the mall entrance.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_5", + "correct_answer": "The beverage section is on the left side of the mall entrance.", + "original_options": { + "A": "On the right side of the entrance.", + "B": "The beverage section is on the left side of the mall entrance.", + "C": "Straight ahead from the entrance.", + "D": "The beverage section is across from the mall entrance, on the opposite side from where you came in.", + "E": "The beverage section is upstairs near the mall entrance, not on the ground level where you entered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 17 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_6", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to buy drinks. Where are they located in the mall?", + "answer": "The beverage section is on the left side of the mall entrance.", + "question_chinese": "我想去买饮料，它在商场的哪个位置呢？", + "answer_chinese": "饮料区在商场入口处的左侧。", + "evidence": [ + { + "description": "The beverage section is on the left side of the mall entrance.", + "timestamps": { + "start": "00:04:35", + "end": "00:04:44" + } + } + ], + "question_timestamps": "00:09:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the right side of the entrance.", + "B": "The beverage section is on the left side of the mall entrance.", + "C": "Straight ahead from the entrance.", + "D": "The beverage section is just past the mall entrance, directly ahead before you reach any side aisles.", + "E": "The beverage section is upstairs by the mall entrance, so you need to go up a level first." + }, + "answer_label": "B", + "answer_text": "The beverage section is on the left side of the mall entrance.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_6", + "correct_answer": "The beverage section is on the left side of the mall entrance.", + "original_options": { + "A": "On the right side of the entrance.", + "B": "The beverage section is on the left side of the mall entrance.", + "C": "Straight ahead from the entrance.", + "D": "The beverage section is just past the mall entrance, directly ahead before you reach any side aisles.", + "E": "The beverage section is upstairs by the mall entrance, so you need to go up a level first." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 17, + 18 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_7", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I'm looking for peanut jar snacks. Where should I find them?", + "answer": "The peanut jar is on the second shelf on the right side of the entrance.", + "question_chinese": "我想去找PEANUT Jar的零食，我应该在哪找到呢？", + "answer_chinese": "PEANUT Jar在入口右手第二个货架上。", + "evidence": [ + { + "description": "The peanut jar is on the second shelf on the right side of the entrance.", + "timestamps": { + "start": "00:08:17", + "end": "00:08:27" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the second shelf to the left.", + "B": "On the top shelf to the right.", + "C": "The peanut jar is on the second shelf on the right side of the entrance.", + "D": "The peanut jar is on the bottom shelf on the right side of the entrance, near the floor.", + "E": "The peanut jar is on the second shelf in the middle section, not on the right side by the entrance." + }, + "answer_label": "C", + "answer_text": "The peanut jar is on the second shelf on the right side of the entrance.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_7", + "correct_answer": "The peanut jar is on the second shelf on the right side of the entrance.", + "original_options": { + "A": "On the second shelf to the left.", + "B": "On the top shelf to the right.", + "C": "The peanut jar is on the second shelf on the right side of the entrance.", + "D": "The peanut jar is on the bottom shelf on the right side of the entrance, near the floor.", + "E": "The peanut jar is on the second shelf in the middle section, not on the right side by the entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 20 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_8", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I'm looking for peanut jar snacks. Where should I find them?", + "answer": "The peanut jar is on the second shelf on the right side of the entrance.", + "question_chinese": "我想去找PEANUT Jar的零食，我应该在哪找到呢？", + "answer_chinese": "PEANUT Jar在入口右手第二个货架上。", + "evidence": [ + { + "description": "The peanut jar is on the second shelf on the right side of the entrance.", + "timestamps": { + "start": "00:08:17", + "end": "00:08:27" + } + } + ], + "question_timestamps": "00:20:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The peanut jar is on the second shelf on the right side of the entrance.", + "B": "On the third shelf to the right.", + "C": "On the first shelf to the right.", + "D": "The peanut jar is on the second shelf on the right side, but farther down the aisle past the entrance area.", + "E": "The peanut jar is on the second shelf on the left side just past the entrance, not on the right." + }, + "answer_label": "A", + "answer_text": "The peanut jar is on the second shelf on the right side of the entrance.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_8", + "correct_answer": "The peanut jar is on the second shelf on the right side of the entrance.", + "original_options": { + "A": "The peanut jar is on the second shelf on the right side of the entrance.", + "B": "On the third shelf to the right.", + "C": "On the first shelf to the right.", + "D": "The peanut jar is on the second shelf on the right side, but farther down the aisle past the entrance area.", + "E": "The peanut jar is on the second shelf on the left side just past the entrance, not on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 20 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_4_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to find the PEANUT Jar snack—how should I get there?", + "answer": "It’s on the shelf behind you.", + "question_chinese": "我想去找PEANUT Jar的零食，我应该怎么走呢？", + "answer_chinese": "你现在向后转的货架上。", + "evidence": [ + { + "description": "The peanut jar is on the second shelf on the right side of the entrance.", + "timestamps": { + "start": "00:08:17", + "end": "00:08:27" + } + } + ], + "question_timestamps": "00:08:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the shelf in front.", + "B": "It’s on the shelf to your left.", + "C": "It’s on the shelf to your right, so you should turn and look along the right-hand shelves.", + "D": "It’s on the shelf above you, so you need to look up at the higher row of shelves.", + "E": "It’s on the shelf behind you." + }, + "answer_label": "E", + "answer_text": "It’s on the shelf behind you.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_4_T1", + "correct_answer": "It’s on the shelf behind you.", + "original_options": { + "A": "On the shelf in front.", + "B": "It’s on the shelf to your left.", + "C": "It’s on the shelf to your right, so you should turn and look along the right-hand shelves.", + "D": "It’s on the shelf above you, so you need to look up at the higher row of shelves.", + "E": "It’s on the shelf behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 18, + 19 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_4_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to find the PEANUT Jar snack—how should I get there?", + "answer": "Stand up from the bench, turn around, and walk straight to the second shelf.", + "question_chinese": "我想去找PEANUT Jar的零食，我应该怎么走呢？", + "answer_chinese": "你从板凳上起来，向后转直走到第二个货架。", + "evidence": [ + { + "description": "The peanut jar is on the second shelf on the right side of the entrance.", + "timestamps": { + "start": "00:08:17", + "end": "00:08:27" + } + } + ], + "question_timestamps": "00:20:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left and go to the first shelf.", + "B": "Walk straight to the top shelf.", + "C": "Stand up from the bench, turn around, and walk straight to the second shelf.", + "D": "Stand up, face forward, and walk to the first shelf instead of the second one.", + "E": "Get up from the bench, turn left, then follow the aisle until you reach the far shelf." + }, + "answer_label": "C", + "answer_text": "Stand up from the bench, turn around, and walk straight to the second shelf.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_4_T2", + "correct_answer": "Stand up from the bench, turn around, and walk straight to the second shelf.", + "original_options": { + "A": "Turn left and go to the first shelf.", + "B": "Walk straight to the top shelf.", + "C": "Stand up from the bench, turn around, and walk straight to the second shelf.", + "D": "Stand up, face forward, and walk to the first shelf instead of the second one.", + "E": "Get up from the bench, turn left, then follow the aisle until you reach the far shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 15, + 17 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_3", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I'm looking for chocolate cookies. In what direction are they from me?", + "answer": "The chocolate cookies are at the end of the shelf on your left.", + "question_chinese": "我想找到巧克力曲奇，它在我的什么方位呢？", + "answer_chinese": "巧克力曲奇在你左手边的货架尽头。", + "evidence": [ + { + "description": "The chocolate cookies are at the end of the second shelf on the right-hand side from the entrance.", + "timestamps": { + "start": "00:01:36", + "end": "00:01:48" + } + } + ], + "question_timestamps": "00:02:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "End of shelf right.", + "B": "Middle of left shelf.", + "C": "They’re at the end of the shelf straight ahead of you, not on the left side.", + "D": "They’re behind you on the end of a nearby shelf, so you need to turn around first.", + "E": "The chocolate cookies are at the end of the shelf on your left." + }, + "answer_label": "E", + "answer_text": "The chocolate cookies are at the end of the shelf on your left.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_3", + "correct_answer": "The chocolate cookies are at the end of the shelf on your left.", + "original_options": { + "A": "End of shelf right.", + "B": "Middle of left shelf.", + "C": "They’re at the end of the shelf straight ahead of you, not on the left side.", + "D": "They’re behind you on the end of a nearby shelf, so you need to turn around first.", + "E": "The chocolate cookies are at the end of the shelf on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_4", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I'm looking for chocolate cookies. In what direction are they from me?", + "answer": "The chocolate cookies are at the end of the shelf on your right.", + "question_chinese": "我想找到巧克力曲奇，它在我的什么方位呢？", + "answer_chinese": "巧克力曲奇就在你的右边的货架的尽头。", + "evidence": [ + { + "description": "The chocolate cookies are at the end of the second shelf on the right-hand side from the entrance.", + "timestamps": { + "start": "00:01:36", + "end": "00:01:48" + } + } + ], + "question_timestamps": "00:02:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "End of shelf left.", + "B": "Middle of right shelf.", + "C": "The chocolate cookies are at the end of the shelf on your right.", + "D": "They’re at the end of the shelf straight in front of you, rather than on your right side.", + "E": "They’re behind you at the end of the shelf, so you’d need to turn around to see them." + }, + "answer_label": "C", + "answer_text": "The chocolate cookies are at the end of the shelf on your right.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_4", + "correct_answer": "The chocolate cookies are at the end of the shelf on your right.", + "original_options": { + "A": "End of shelf left.", + "B": "Middle of right shelf.", + "C": "The chocolate cookies are at the end of the shelf on your right.", + "D": "They’re at the end of the shelf straight in front of you, rather than on your right side.", + "E": "They’re behind you at the end of the shelf, so you’d need to turn around to see them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 20 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I'm looking for a snack called Hi Chow. What route should I take?", + "answer": "You should turn around, walk past one shelf, and then turn back.", + "question_chinese": "我想找名为 Hi chow 的零食，我的路线应该是怎么样的呢？", + "answer_chinese": "你应该向后转，然后走过一个货架，再回头。", + "evidence": [ + { + "description": "Hi Chow snacks are on this shelf.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should turn around, walk past one shelf, and then turn back.", + "B": "Turn left, pass one shelf.", + "C": "Go straight past two shelves.", + "D": "Turn around, walk past two shelves, then turn back when you reach the next aisle opening.", + "E": "Head straight, pass one shelf, then turn around and continue back toward where you started." + }, + "answer_label": "A", + "answer_text": "You should turn around, walk past one shelf, and then turn back.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_1", + "correct_answer": "You should turn around, walk past one shelf, and then turn back.", + "original_options": { + "A": "You should turn around, walk past one shelf, and then turn back.", + "B": "Turn left, pass one shelf.", + "C": "Go straight past two shelves.", + "D": "Turn around, walk past two shelves, then turn back when you reach the next aisle opening.", + "E": "Head straight, pass one shelf, then turn around and continue back toward where you started." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I'm looking for a snack called Hi Chow. What route should I take?", + "answer": "You should turn left, then move forward past two shelves, and continue straight to the left until you reach the end of the shelf.", + "question_chinese": "我想找名为 Hi chow 的零食，我的路线应该是怎么样的呢？", + "answer_chinese": "你应该向左转，然后向前跨越两个货架再向左直走到货架的尽头。", + "evidence": [ + { + "description": "Hi Chow snacks are on this shelf.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:12:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go forward.", + "B": "Pass three shelves, then continue left.", + "C": "Turn left, walk forward past two shelves, then continue straight to the right until you reach the shelf’s end.", + "D": "You should turn left, then move forward past two shelves, and continue straight to the left until you reach the end of the shelf.", + "E": "Turn left first, move forward past one shelf, and keep going left until you reach the start of the shelf." + }, + "answer_label": "D", + "answer_text": "You should turn left, then move forward past two shelves, and continue straight to the left until you reach the end of the shelf.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_2", + "correct_answer": "You should turn left, then move forward past two shelves, and continue straight to the left until you reach the end of the shelf.", + "original_options": { + "A": "Turn right, then go forward.", + "B": "Pass three shelves, then continue left.", + "C": "Turn left, walk forward past two shelves, then continue straight to the right until you reach the shelf’s end.", + "D": "You should turn left, then move forward past two shelves, and continue straight to the left until you reach the end of the shelf.", + "E": "Turn left first, move forward past one shelf, and keep going left until you reach the start of the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 20 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_7_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How much food am I holding right now?", + "answer": "Right now, you are holding seven packs of snacks and two cups of drinks.", + "question_chinese": "我现在拿了多少食物呢？", + "answer_chinese": "你现在拿了七包零食，两杯饮料。", + "evidence": [ + { + "description": "Taking two foil-wrapped foods.", + "timestamps": { + "start": "00:09:50", + "end": "00:10:07" + } + } + ], + "question_timestamps": "00:18:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Right now, you are holding seven packs of snacks and two cups of drinks.", + "B": "Seven packs and one cup.", + "C": "Six packs and two cups.", + "D": "Right now, you are holding seven packs of snacks and three cups of drinks.", + "E": "Right now, you are holding eight packs of snacks and two cups of drinks." + }, + "answer_label": "A", + "answer_text": "Right now, you are holding seven packs of snacks and two cups of drinks.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_7_T1", + "correct_answer": "Right now, you are holding seven packs of snacks and two cups of drinks.", + "original_options": { + "A": "Right now, you are holding seven packs of snacks and two cups of drinks.", + "B": "Seven packs and one cup.", + "C": "Six packs and two cups.", + "D": "Right now, you are holding seven packs of snacks and three cups of drinks.", + "E": "Right now, you are holding eight packs of snacks and two cups of drinks." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_7_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How much food am I holding right now?", + "answer": "Right now, you are holding seven packs of snacks, two cups of drinks, and two foil-wrapped foods.", + "question_chinese": "我现在拿了多少食物呢？", + "answer_chinese": "你现在拿了七包零食，两杯饮料，两个锡纸包的食物。", + "evidence": [ + { + "description": "You purchased seven packs of snacks, two beverages, and two pieces of foil-wrapped food on the table.", + "timestamps": { + "start": "00:15:10", + "end": "00:15:18" + } + } + ], + "question_timestamps": "00:22:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Six packs, two cups, two foils.", + "B": "Right now, you are holding seven packs of snacks, two cups of drinks, and two foil-wrapped foods.", + "C": "Seven packs, three cups, two foils.", + "D": "Right now, you are holding eight packs of snacks, two cups of drinks, and two foil-wrapped foods.", + "E": "Right now, you are holding seven packs of snacks, two cups of drinks, and three foil-wrapped foods." + }, + "answer_label": "B", + "answer_text": "Right now, you are holding seven packs of snacks, two cups of drinks, and two foil-wrapped foods.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_7_T2", + "correct_answer": "Right now, you are holding seven packs of snacks, two cups of drinks, and two foil-wrapped foods.", + "original_options": { + "A": "Six packs, two cups, two foils.", + "B": "Right now, you are holding seven packs of snacks, two cups of drinks, and two foil-wrapped foods.", + "C": "Seven packs, three cups, two foils.", + "D": "Right now, you are holding eight packs of snacks, two cups of drinks, and two foil-wrapped foods.", + "E": "Right now, you are holding seven packs of snacks, two cups of drinks, and three foil-wrapped foods." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_5_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the food in the oven relative to me?", + "answer": "The food in the oven is on your right side.", + "question_chinese": "烤箱里的食品相对于我的位置是什么呢？", + "answer_chinese": "烤箱里的食品在你的右侧。", + "evidence": [ + { + "description": "The food in the oven is on your right.", + "timestamps": { + "start": "00:09:41", + "end": "00:09:43" + } + } + ], + "question_timestamps": "00:09:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The food in the oven is on your right side.", + "B": "On your left.", + "C": "Straight ahead.", + "D": "The food in the oven is behind you, so you would need to turn around to face it.", + "E": "The food in the oven is slightly to your left, not on your right side where you’re looking." + }, + "answer_label": "A", + "answer_text": "The food in the oven is on your right side.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_5_T2", + "correct_answer": "The food in the oven is on your right side.", + "original_options": { + "A": "The food in the oven is on your right side.", + "B": "On your left.", + "C": "Straight ahead.", + "D": "The food in the oven is behind you, so you would need to turn around to face it.", + "E": "The food in the oven is slightly to your left, not on your right side where you’re looking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 18, + 19 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_5_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the food in the oven relative to me?", + "answer": "The food in the oven is in front of you.", + "question_chinese": "烤箱里的食品相对于我的位置是什么呢？", + "answer_chinese": "烤箱里的食品在你的前面。", + "evidence": [ + { + "description": "The food in the oven is in front of you.", + "timestamps": { + "start": "00:09:46", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:10:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The food in the oven is in front of you.", + "D": "The food in the oven is to your right as you face the oven opening.", + "E": "The food in the oven is above you, positioned higher than your eye level." + }, + "answer_label": "C", + "answer_text": "The food in the oven is in front of you.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_5_T1", + "correct_answer": "The food in the oven is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The food in the oven is in front of you.", + "D": "The food in the oven is to your right as you face the oven opening.", + "E": "The food in the oven is above you, positioned higher than your eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 14 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_9", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of food in the oven relative to the cashier?", + "answer": "The baked goods are on the right side of the checkout counter.", + "question_chinese": "烤箱里的食品相对于收银台的位置是什么呢？", + "answer_chinese": "烤箱里的食品在收银台右侧。", + "evidence": [ + { + "description": "The baked goods are on the right side of the checkout counter.", + "timestamps": { + "start": "00:09:41", + "end": "00:10:33" + } + } + ], + "question_timestamps": "00:12:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the left side of the checkout counter.", + "B": "Behind the checkout counter.", + "C": "The baked goods are in front of the checkout counter, facing the cashier directly.", + "D": "The baked goods are on the far side of the checkout counter, opposite where the cashier stands.", + "E": "The baked goods are on the right side of the checkout counter." + }, + "answer_label": "E", + "answer_text": "The baked goods are on the right side of the checkout counter.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_9", + "correct_answer": "The baked goods are on the right side of the checkout counter.", + "original_options": { + "A": "On the left side of the checkout counter.", + "B": "Behind the checkout counter.", + "C": "The baked goods are in front of the checkout counter, facing the cashier directly.", + "D": "The baked goods are on the far side of the checkout counter, opposite where the cashier stands.", + "E": "The baked goods are on the right side of the checkout counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 4, + 14, + 17 + ] + }, + { + "video_uid": "0b637191-00b7-4fd8-af37-d26d80dc59ce", + "video_length": "00:22:44", + "qid": "0b637191-00b7-4fd8-af37-d26d80dc59ce_10", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of food in the oven relative to the cashier?", + "answer": "The baked goods are on the right side of the checkout counter.", + "question_chinese": "烤箱里的食品相对于收银台的位置是什么呢？", + "answer_chinese": "烤箱里的食品在收银台右侧。", + "evidence": [ + { + "description": "The baked goods are on the right side of the checkout counter.", + "timestamps": { + "start": "00:09:41", + "end": "00:10:33" + } + } + ], + "question_timestamps": "00:22:39", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the left side of the counter.", + "B": "In front of the checkout counter.", + "C": "The baked goods are behind the checkout counter, positioned closer to the cashier’s side.", + "D": "The baked goods are on the right side of the checkout counter.", + "E": "The baked goods are on a shelf above the checkout counter, higher than the cashier’s workspace." + }, + "answer_label": "D", + "answer_text": "The baked goods are on the right side of the checkout counter.", + "q_id": "0b637191-00b7-4fd8-af37-d26d80dc59ce_10", + "correct_answer": "The baked goods are on the right side of the checkout counter.", + "original_options": { + "A": "On the left side of the counter.", + "B": "In front of the checkout counter.", + "C": "The baked goods are behind the checkout counter, positioned closer to the cashier’s side.", + "D": "The baked goods are on the right side of the checkout counter.", + "E": "The baked goods are on a shelf above the checkout counter, higher than the cashier’s workspace." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 17 + ] + } +] diff --git a/QAs/0b703793-134a-4889-bc40-c19328d8f7cd-mcq.json b/QAs/0b703793-134a-4889-bc40-c19328d8f7cd-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..000b02537bbc6cde490ec271b69e082433e999c9 --- /dev/null +++ b/QAs/0b703793-134a-4889-bc40-c19328d8f7cd-mcq.json @@ -0,0 +1,1161 @@ +[ + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_5", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doorways did I pass through when I went from the kitchen to the garden to get the ball?", + "answer": "You went through two doors, first the kitchen door, and then the garden's patio doors.", + "question_chinese": "从厨房移动到花园去拿球，我穿过了多少个门？", + "answer_chinese": "你穿过了两个门，首先是厨房门，然后是花园的落地门。", + "evidence": [ + { + "description": "I pass through the door of kitchen.", + "timestamps": { + "start": "00:32:25", + "end": "00:32:28" + } + }, + { + "description": "I opened and passed through the door to the garden.", + "timestamps": { + "start": "00:33:04", + "end": "00:33:10" + } + } + ], + "question_timestamps": "00:33:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You went through one door.", + "B": "You went through three doors.", + "C": "You went through two doors, first the kitchen door, and then the garden's patio doors.", + "D": "You went through one door, using the kitchen door that leads straight out to the garden.", + "E": "You went through three doors: the kitchen door, then an interior doorway, and finally the garden exit." + }, + "answer_label": "C", + "answer_text": "You went through two doors, first the kitchen door, and then the garden's patio doors.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_5", + "correct_answer": "You went through two doors, first the kitchen door, and then the garden's patio doors.", + "original_options": { + "A": "You went through one door.", + "B": "You went through three doors.", + "C": "You went through two doors, first the kitchen door, and then the garden's patio doors.", + "D": "You went through one door, using the kitchen door that leads straight out to the garden.", + "E": "You went through three doors: the kitchen door, then an interior doorway, and finally the garden exit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_20", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doorways did I pass through when I went from the kitchen to the garden to get the ball?", + "answer": "You went through two doors, first the kitchen door, and then the garden's patio doors.", + "question_chinese": "从厨房移动到花园去拿球，我穿过了多少个门？", + "answer_chinese": "你穿过了两个门，首先是厨房门，然后是花园的落地门。", + "evidence": [ + { + "description": "I pass through the door of kitchen.", + "timestamps": { + "start": "00:32:25", + "end": "00:32:28" + } + }, + { + "description": "I opened and passed through the door to the garden.", + "timestamps": { + "start": "00:33:04", + "end": "00:33:10" + } + } + ], + "question_timestamps": "00:42:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You went through one door.", + "B": "You went through three doors.", + "C": "You went through four doors: the kitchen door, then two interior doorways, and finally the patio doors.", + "D": "You went through two doors, first the kitchen door, and then the garden's patio doors.", + "E": "You went through five doors, passing multiple interior doorways before reaching the garden patio doors." + }, + "answer_label": "D", + "answer_text": "You went through two doors, first the kitchen door, and then the garden's patio doors.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_20", + "correct_answer": "You went through two doors, first the kitchen door, and then the garden's patio doors.", + "original_options": { + "A": "You went through one door.", + "B": "You went through three doors.", + "C": "You went through four doors: the kitchen door, then two interior doorways, and finally the patio doors.", + "D": "You went through two doors, first the kitchen door, and then the garden's patio doors.", + "E": "You went through five doors, passing multiple interior doorways before reaching the garden patio doors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 15 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_6", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make from the current location to the main door?", + "answer": "You need to make one turn.", + "question_chinese": "从现在的位置到大门，我需要拐几个弯？", + "answer_chinese": "你需要拐一个弯。", + "evidence": [ + { + "description": "Go around one corner to reach the main door.", + "timestamps": { + "start": "00:32:25", + "end": "00:32:35" + } + } + ], + "question_timestamps": "00:32:38", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You need two turns.", + "B": "You need to make one turn.", + "C": "You need three turns.", + "D": "You need four turns, changing direction repeatedly before you reach the main door.", + "E": "You need five turns, weaving through several corners on the way to the main door." + }, + "answer_label": "B", + "answer_text": "You need to make one turn.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_6", + "correct_answer": "You need to make one turn.", + "original_options": { + "A": "You need two turns.", + "B": "You need to make one turn.", + "C": "You need three turns.", + "D": "You need four turns, changing direction repeatedly before you reach the main door.", + "E": "You need five turns, weaving through several corners on the way to the main door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 15 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_9", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make from the current location to the main door?", + "answer": "You need to make three turns.", + "question_chinese": "从现在的位置到大门，我需要拐几个弯？", + "answer_chinese": "你需要拐3个弯。", + "evidence": [ + { + "description": "Turn the first corner to reach the hallway.", + "timestamps": { + "start": "00:36:18", + "end": "00:36:25" + } + }, + { + "description": "Turn the second corner.", + "timestamps": { + "start": "00:36:26", + "end": "00:36:30" + } + }, + { + "description": "Go around one corner to reach the main door.", + "timestamps": { + "start": "00:32:31", + "end": "00:32:36" + } + } + ], + "question_timestamps": "00:36:30", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You need one turn.", + "B": "You need to make three turns.", + "C": "You need two turns.", + "D": "You need four turns, making an extra change of direction before reaching the main door.", + "E": "You need five turns, taking a more winding route with multiple direction changes to the main door." + }, + "answer_label": "B", + "answer_text": "You need to make three turns.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_9", + "correct_answer": "You need to make three turns.", + "original_options": { + "A": "You need one turn.", + "B": "You need to make three turns.", + "C": "You need two turns.", + "D": "You need four turns, making an extra change of direction before reaching the main door.", + "E": "You need five turns, taking a more winding route with multiple direction changes to the main door." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_12", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I place the empty glass vase before I left the kitchen?", + "answer": "You put it on the far right of the windowsill.", + "question_chinese": "在我离开厨房之前，我把空的玻璃花瓶放哪了？", + "answer_chinese": "你把它放在了窗台的最右边。", + "evidence": [ + { + "description": "The vase was filled halfway with water and then put on the far right of the windowsill.", + "timestamps": { + "start": "00:08:42", + "end": "00:09:04" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the left side of the windowsill.", + "B": "In the middle of the windowsill.", + "C": "You put it on the far right of the windowsill.", + "D": "You placed it on the countertop near the sink, not on the windowsill before leaving.", + "E": "You set it inside the cabinet below the sink, putting it away before you left the kitchen." + }, + "answer_label": "C", + "answer_text": "You put it on the far right of the windowsill.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_12", + "correct_answer": "You put it on the far right of the windowsill.", + "original_options": { + "A": "On the left side of the windowsill.", + "B": "In the middle of the windowsill.", + "C": "You put it on the far right of the windowsill.", + "D": "You placed it on the countertop near the sink, not on the windowsill before leaving.", + "E": "You set it inside the cabinet below the sink, putting it away before you left the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 17 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_13", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I place the empty glass vase before I left the kitchen?", + "answer": "You put it on the far right of the windowsill of the kitchen.", + "question_chinese": "在我离开厨房之前，我把空的玻璃花瓶放哪了？", + "answer_chinese": "你把它放在了厨房的窗台的最右边。", + "evidence": [ + { + "description": "The vase was filled halfway with water and then put on the far right of the windowsill.", + "timestamps": { + "start": "00:08:42", + "end": "00:09:04" + } + } + ], + "question_timestamps": "01:06:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the left side of the windowsill.", + "B": "On the kitchen countertop near the sink.", + "C": "You placed it in the middle of the windowsill rather than on the far right side.", + "D": "You put it on the far right of the windowsill of the kitchen.", + "E": "You set it on the kitchen table near the center, not on the windowsill at all." + }, + "answer_label": "D", + "answer_text": "You put it on the far right of the windowsill of the kitchen.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_13", + "correct_answer": "You put it on the far right of the windowsill of the kitchen.", + "original_options": { + "A": "On the left side of the windowsill.", + "B": "On the kitchen countertop near the sink.", + "C": "You placed it in the middle of the windowsill rather than on the far right side.", + "D": "You put it on the far right of the windowsill of the kitchen.", + "E": "You set it on the kitchen table near the center, not on the windowsill at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_19", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the refrigerator from my current location?", + "answer": "Walk straight ahead, turn left at the end to enter the kitchen, and the refrigerator is near the doorway.", + "question_chinese": "如何从我现在的位置到达冰箱？", + "answer_chinese": "向前直走，尽头左转进入厨房，冰箱就在门口附近。", + "evidence": [ + { + "description": "The refrigerator in the kitchen is near the doorway.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:40" + } + }, + { + "description": "A long hallway, with the kitchen on the left at the end if you walk forward.", + "timestamps": { + "start": "00:09:20", + "end": "00:09:24" + } + } + ], + "question_timestamps": "00:09:28", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right into the kitchen; it's nearby.", + "B": "Go straight; the fridge is far inside.", + "C": "Walk straight ahead, turn right at the end, then look for the refrigerator away from the doorway.", + "D": "Walk straight ahead, turn left at the end to enter the kitchen, and the refrigerator is near the doorway.", + "E": "Walk straight ahead, turn left into the kitchen, but the refrigerator is deeper inside, not near the entrance." + }, + "answer_label": "D", + "answer_text": "Walk straight ahead, turn left at the end to enter the kitchen, and the refrigerator is near the doorway.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_19", + "correct_answer": "Walk straight ahead, turn left at the end to enter the kitchen, and the refrigerator is near the doorway.", + "original_options": { + "A": "Turn right into the kitchen; it's nearby.", + "B": "Go straight; the fridge is far inside.", + "C": "Walk straight ahead, turn right at the end, then look for the refrigerator away from the doorway.", + "D": "Walk straight ahead, turn left at the end to enter the kitchen, and the refrigerator is near the doorway.", + "E": "Walk straight ahead, turn left into the kitchen, but the refrigerator is deeper inside, not near the entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_17", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the refrigerator from my current location?", + "answer": "Turn left, walk two steps forward, and the refrigerator will be on your right.", + "question_chinese": "如何从我现在的位置到达冰箱？", + "answer_chinese": "向左转，再向前走两步，冰箱就在你的右手边。", + "evidence": [ + { + "description": "The refrigerator in the kitchen is near the doorway.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:27:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right and walk two steps.", + "B": "Go straight ahead to the kitchen corner.", + "C": "Turn left and walk about four steps forward, then the refrigerator should be on your left side.", + "D": "Turn left, walk two steps forward, and the refrigerator will be on your right.", + "E": "Turn around and walk a few steps forward, and the refrigerator will end up in front of you." + }, + "answer_label": "D", + "answer_text": "Turn left, walk two steps forward, and the refrigerator will be on your right.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_17", + "correct_answer": "Turn left, walk two steps forward, and the refrigerator will be on your right.", + "original_options": { + "A": "Turn right and walk two steps.", + "B": "Go straight ahead to the kitchen corner.", + "C": "Turn left and walk about four steps forward, then the refrigerator should be on your left side.", + "D": "Turn left, walk two steps forward, and the refrigerator will be on your right.", + "E": "Turn around and walk a few steps forward, and the refrigerator will end up in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 18 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_18", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the refrigerator from my current location?", + "answer": "Turn around, walk about four steps forward, and the refrigerator will be on your right.", + "question_chinese": "如何从我现在的位置到达冰箱？", + "answer_chinese": "向后转，向前走四步左右，冰箱就在你的右手边。", + "evidence": [ + { + "description": "The refrigerator in the kitchen is near the doorway.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:31:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn around; the fridge is on your left.", + "B": "Turn right and walk three steps forward.", + "C": "Turn around, walk about four steps forward, and the refrigerator will be on your right.", + "D": "Turn around, walk about four steps forward, but the refrigerator will be on your left instead of your right.", + "E": "Turn around and walk about six steps forward, and the refrigerator will be straight ahead rather than to your right." + }, + "answer_label": "C", + "answer_text": "Turn around, walk about four steps forward, and the refrigerator will be on your right.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_18", + "correct_answer": "Turn around, walk about four steps forward, and the refrigerator will be on your right.", + "original_options": { + "A": "Turn around; the fridge is on your left.", + "B": "Turn right and walk three steps forward.", + "C": "Turn around, walk about four steps forward, and the refrigerator will be on your right.", + "D": "Turn around, walk about four steps forward, but the refrigerator will be on your left instead of your right.", + "E": "Turn around and walk about six steps forward, and the refrigerator will be straight ahead rather than to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 20 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_10", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I put my phone?", + "answer": "You put it on the kitchen table, above the jar.", + "question_chinese": "我把手机放在哪里了？", + "answer_chinese": "你放在了厨房的桌子上，罐子的上方。", + "evidence": [ + { + "description": "The phone was placed on the jar on the kitchen table.", + "timestamps": { + "start": "00:10:45", + "end": "00:11:15" + } + } + ], + "question_timestamps": "00:31:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the kitchen table beside the jar.", + "B": "On the kitchen counter near the jar.", + "C": "You put it on the kitchen table, above the jar.", + "D": "You left it on the kitchen table below the jar, not above it where you can see it easily.", + "E": "You put it on the kitchen table to the right of the jar rather than directly above the jar." + }, + "answer_label": "C", + "answer_text": "You put it on the kitchen table, above the jar.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_10", + "correct_answer": "You put it on the kitchen table, above the jar.", + "original_options": { + "A": "On the kitchen table beside the jar.", + "B": "On the kitchen counter near the jar.", + "C": "You put it on the kitchen table, above the jar.", + "D": "You left it on the kitchen table below the jar, not above it where you can see it easily.", + "E": "You put it on the kitchen table to the right of the jar rather than directly above the jar." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_11", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I put my phone?", + "answer": "The phone is now in your left hand, above the book you are holding.", + "question_chinese": "我把手机放在哪里了？", + "answer_chinese": "现在手机在你的左手手中，你拿着的书本的上方。", + "evidence": [ + { + "description": "The phone was picked up and is now held above the book in the hand.", + "timestamps": { + "start": "00:32:06", + "end": "00:32:20" + } + }, + { + "description": "The phone was placed on the jar on the kitchen table.", + "timestamps": { + "start": "00:10:45", + "end": "00:11:15" + } + } + ], + "question_timestamps": "00:32:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In your right hand, above the book.", + "B": "The phone is now in your left hand, above the book you are holding.", + "C": "In your left hand, below the book.", + "D": "The phone is now in your right hand, resting below the book you are holding.", + "E": "The phone is tucked under the book in your left hand rather than held above it." + }, + "answer_label": "B", + "answer_text": "The phone is now in your left hand, above the book you are holding.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_11", + "correct_answer": "The phone is now in your left hand, above the book you are holding.", + "original_options": { + "A": "In your right hand, above the book.", + "B": "The phone is now in your left hand, above the book you are holding.", + "C": "In your left hand, below the book.", + "D": "The phone is now in your right hand, resting below the book you are holding.", + "E": "The phone is tucked under the book in your left hand rather than held above it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_8", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the microwave from my current location?", + "answer": "Yes. The microwave is to your left.", + "question_chinese": "我现在的位置能碰到微波炉吗？", + "answer_chinese": "是。微波炉在你左边。", + "evidence": [ + { + "description": "The microwave is on the kitchen table.", + "timestamps": { + "start": "00:25:56", + "end": "00:26:06" + } + } + ], + "question_timestamps": "00:26:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No. The microwave is in the kitchen.", + "B": "Yes. The microwave is to your left." + }, + "answer_label": "B", + "answer_text": "Yes. The microwave is to your left.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_8", + "correct_answer": "Yes. The microwave is to your left.", + "original_options": { + "A": "No. The microwave is in the kitchen.", + "B": "Yes. The microwave is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_7", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the microwave from my current location?", + "answer": "No, the microwave is in the kitchen.", + "question_chinese": "我现在的位置能碰到微波炉吗？", + "answer_chinese": "不，微波炉在厨房。", + "evidence": [ + { + "description": "The microwave is on the kitchen table.", + "timestamps": { + "start": "00:25:56", + "end": "00:26:06" + } + } + ], + "question_timestamps": "01:14:52", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, the microwave is in the kitchen.", + "B": "Yes. The microwave is to your left." + }, + "answer_label": "A", + "answer_text": "No, the microwave is in the kitchen.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_7", + "correct_answer": "No, the microwave is in the kitchen.", + "original_options": { + "A": "No, the microwave is in the kitchen.", + "B": "Yes. The microwave is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get that large bag of blocks now?", + "answer": "No, the large bag of blocks is behind the door.", + "question_chinese": "我现在能拿到那一大袋积木吗？", + "answer_chinese": "不行，积木在门的后面。", + "evidence": [ + { + "description": "The large bag of blocks is on the storeroom floor.", + "timestamps": { + "start": "00:34:05", + "end": "00:34:08" + } + } + ], + "question_timestamps": "00:34:09", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can reach the large bag of blocks now.", + "B": "No, the large bag of blocks is behind the door." + }, + "answer_label": "B", + "answer_text": "No, the large bag of blocks is behind the door.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_2", + "correct_answer": "No, the large bag of blocks is behind the door.", + "original_options": { + "A": "Yes, you can reach the large bag of blocks now.", + "B": "No, the large bag of blocks is behind the door." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_3", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get that large bag of blocks now?", + "answer": "No, the blocks are on the storeroom floor.", + "question_chinese": "我现在能拿到那一大袋积木吗？", + "answer_chinese": "不行，积木在储藏室的地板上。", + "evidence": [ + { + "description": "The large bag of blocks is on the storeroom floor.", + "timestamps": { + "start": "00:34:04", + "end": "00:34:09" + } + } + ], + "question_timestamps": "00:59:26", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, the blocks are on the storeroom floor.", + "B": "Yes, you can reach the large bag of blocks now." + }, + "answer_label": "A", + "answer_text": "No, the blocks are on the storeroom floor.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_3", + "correct_answer": "No, the blocks are on the storeroom floor.", + "original_options": { + "A": "No, the blocks are on the storeroom floor.", + "B": "Yes, you can reach the large bag of blocks now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of dough are still on the kitchen table now?", + "answer": "There are two.", + "question_chinese": "现在厨房的桌上还有几个面团？", + "answer_chinese": "有两个。", + "evidence": [ + { + "description": "There are two pieces of dough left on the table.", + "timestamps": { + "start": "00:25:29", + "end": "00:26:00" + } + } + ], + "question_timestamps": "00:26:17", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two.", + "B": "There is one.", + "C": "There are three.", + "D": "There are four pieces of dough still on the kitchen table now.", + "E": "There are five pieces of dough still sitting on the kitchen table right now." + }, + "answer_label": "A", + "answer_text": "There are two.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_1", + "correct_answer": "There are two.", + "original_options": { + "A": "There are two.", + "B": "There is one.", + "C": "There are three.", + "D": "There are four pieces of dough still on the kitchen table now.", + "E": "There are five pieces of dough still sitting on the kitchen table right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 14 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_4", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of dough are still on the kitchen table now?", + "answer": "There is no dough left because you put it in the refrigerator.", + "question_chinese": "现在厨房的桌上还有几个面团？", + "answer_chinese": "没有面团了，因为被你放入了冰箱。", + "evidence": [ + { + "description": "The dough was wrapped in plastic wrap and put in the refrigerator. There is no dough on the table now.", + "timestamps": { + "start": "00:26:30", + "end": "00:26:50" + } + }, + { + "description": "There are two pieces of dough left on the table.", + "timestamps": { + "start": "00:25:36", + "end": "00:26:00" + } + }, + { + "description": "There is no dough on the table.", + "timestamps": { + "start": "00:27:50", + "end": "00:28:10" + } + } + ], + "question_timestamps": "00:43:13", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There is one piece left.", + "B": "There are two pieces left.", + "C": "There are three pieces of dough left because you left them sitting on the kitchen table.", + "D": "There are five pieces of dough left because you never moved any of it into the refrigerator.", + "E": "There is no dough left because you put it in the refrigerator." + }, + "answer_label": "E", + "answer_text": "There is no dough left because you put it in the refrigerator.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_4", + "correct_answer": "There is no dough left because you put it in the refrigerator.", + "original_options": { + "A": "There is one piece left.", + "B": "There are two pieces left.", + "C": "There are three pieces of dough left because you left them sitting on the kitchen table.", + "D": "There are five pieces of dough left because you never moved any of it into the refrigerator.", + "E": "There is no dough left because you put it in the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_3_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Relative to me, where is the dining table?", + "answer": "The dining table is directly in front of you.", + "question_chinese": "相对于我的位置，餐桌在哪里？", + "answer_chinese": "餐桌在你的正前方。", + "evidence": [ + { + "description": "The dining table is directly in front of you.", + "timestamps": { + "start": "00:01:06", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Directly behind you.", + "B": "Directly to your left.", + "C": "The dining table is directly in front of you.", + "D": "The dining table is directly to your right relative to where you are standing now.", + "E": "The dining table is above you relative to your position, not in front of you." + }, + "answer_label": "C", + "answer_text": "The dining table is directly in front of you.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_3_T2", + "correct_answer": "The dining table is directly in front of you.", + "original_options": { + "A": "Directly behind you.", + "B": "Directly to your left.", + "C": "The dining table is directly in front of you.", + "D": "The dining table is directly to your right relative to where you are standing now.", + "E": "The dining table is above you relative to your position, not in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 15 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_3_T1", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Relative to me, where is the dining table?", + "answer": "The dining table is behind you.", + "question_chinese": "相对于我的位置，餐桌在哪里？", + "answer_chinese": "餐桌在你的背后。", + "evidence": [ + { + "description": "The dining table is behind you.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:03:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The dining table is directly to your right relative to where you are standing right now.", + "D": "The dining table is behind you.", + "E": "The dining table is above you relative to your position, rather than behind you." + }, + "answer_label": "D", + "answer_text": "The dining table is behind you.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_3_T1", + "correct_answer": "The dining table is behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The dining table is directly to your right relative to where you are standing right now.", + "D": "The dining table is behind you.", + "E": "The dining table is above you relative to your position, rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 14 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_15", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the dish soap next to the sink from where I'm standing?", + "answer": "Yes, the dish soap is right next to the faucet on your right, within your reach.", + "question_chinese": "站在我所处的位置，可以拿到水槽旁边的洗洁精吗？", + "answer_chinese": "可以，洗洁精就在你右手边的水龙头旁边，你伸手就能拿到。", + "evidence": [ + { + "description": "The dish soap is under the windowsill, next to the sink.", + "timestamps": { + "start": "00:06:55", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:25:29", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, the dish soap is next to the sink but it's out of your reach from where you're standing.", + "B": "Yes, the dish soap is right next to the faucet on your right, within your reach." + }, + "answer_label": "B", + "answer_text": "Yes, the dish soap is right next to the faucet on your right, within your reach.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_15", + "correct_answer": "Yes, the dish soap is right next to the faucet on your right, within your reach.", + "original_options": { + "A": "No, the dish soap is next to the sink but it's out of your reach from where you're standing.", + "B": "Yes, the dish soap is right next to the faucet on your right, within your reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 19 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_14", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the dish soap next to the sink from where I'm standing?", + "answer": "Yes, the dish soap is behind you; you can reach it by turning around.", + "question_chinese": "站在我所处的位置，可以拿到水槽旁边的洗洁精吗？", + "answer_chinese": "能，洗洁精在你的背后，转身就可以拿到。", + "evidence": [ + { + "description": "The dish soap is under the windowsill, next to the sink.", + "timestamps": { + "start": "00:06:55", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:32:03", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, the dish soap is behind you; you can reach it by turning around.", + "B": "No, the dish soap is behind you and you can't reach it from here." + }, + "answer_label": "A", + "answer_text": "Yes, the dish soap is behind you; you can reach it by turning around.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_14", + "correct_answer": "Yes, the dish soap is behind you; you can reach it by turning around.", + "original_options": { + "A": "Yes, the dish soap is behind you; you can reach it by turning around.", + "B": "No, the dish soap is behind you and you can't reach it from here." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "0b703793-134a-4889-bc40-c19328d8f7cd", + "video_length": "01:17:10", + "qid": "0b703793-134a-4889-bc40-c19328d8f7cd_16", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the dish soap next to the sink from where I'm standing?", + "answer": "No, the dish soap is in the kitchen.", + "question_chinese": "站在我所处的位置，可以拿到水槽旁边的洗洁精吗？", + "answer_chinese": "不可以，洗洁精在厨房。", + "evidence": [ + { + "description": "Yes, the dish soap is behind you; you can reach it by turning around.", + "timestamps": { + "start": "00:06:55", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:55:34", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, the dish soap is in the kitchen.", + "B": "Yes, the dish soap is next to the sink, so you can reach it." + }, + "answer_label": "A", + "answer_text": "No, the dish soap is in the kitchen.", + "q_id": "0b703793-134a-4889-bc40-c19328d8f7cd_16", + "correct_answer": "No, the dish soap is in the kitchen.", + "original_options": { + "A": "No, the dish soap is in the kitchen.", + "B": "Yes, the dish soap is next to the sink, so you can reach it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + } +] diff --git a/QAs/0b9ee926-00f1-4b22-9e83-b664c5e465e4-mcq.json b/QAs/0b9ee926-00f1-4b22-9e83-b664c5e465e4-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c9e41f3cfaf9193e08c7984fb87a12b3691e40ca --- /dev/null +++ b/QAs/0b9ee926-00f1-4b22-9e83-b664c5e465e4-mcq.json @@ -0,0 +1,766 @@ +[ + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_5", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the coffee cup or the white plastic bottle?", + "answer": "The coffee cup is closer to me.", + "question_chinese": "咖啡杯和白色塑料瓶哪个离我更近？", + "answer_chinese": "咖啡杯离我更近。", + "evidence": [ + { + "description": "The location of the coffee cup and the white plastic bottle", + "timestamps": { + "start": "00:00:27", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:17:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The white plastic bottle is closer to me.", + "B": "The coffee cup is closer to me." + }, + "answer_label": "B", + "answer_text": "The coffee cup is closer to me.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_5", + "correct_answer": "The coffee cup is closer to me.", + "original_options": { + "A": "The white plastic bottle is closer to me.", + "B": "The coffee cup is closer to me." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_6", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the coffee cup or the white plastic bottle?", + "answer": "The white plastic bottle is closer to me.", + "question_chinese": "咖啡杯和白色塑料瓶哪个离我更近？", + "answer_chinese": "白色塑料瓶离我更近。", + "evidence": [ + { + "description": "The coffee cup is in the bedroom.", + "timestamps": { + "start": "00:30:07", + "end": "00:30:08" + } + }, + { + "description": "A white plastic bottle is in the room where I am now.", + "timestamps": { + "start": "00:32:15", + "end": "00:32:17" + } + } + ], + "question_timestamps": "00:34:30", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The coffee cup is closer to me.", + "B": "The white plastic bottle is closer to me." + }, + "answer_label": "B", + "answer_text": "The white plastic bottle is closer to me.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_6", + "correct_answer": "The white plastic bottle is closer to me.", + "original_options": { + "A": "The coffee cup is closer to me.", + "B": "The white plastic bottle is closer to me." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_9", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the restroom next, how should I go?", + "answer": "You will turn around, walk out of the room, then turn right and walk out of the room again, then turn left and walk to the nearest door, turn left into the room, then walk straight, passing two doors, and you will reach the bathroom.", + "question_chinese": "如果我接下来想去洗手间，我将会怎么走？", + "answer_chinese": "你将要向后转，走出房间，然后向右转，走出房间，然后向左转，走到最近的房门，左转进入房间，然后直走，经过两扇门就到了洗手间。", + "evidence": [ + { + "description": "My movement trajectory to this room.", + "timestamps": { + "start": "00:00:15", + "end": "00:01:38" + } + } + ], + "question_timestamps": "00:17:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left, then head to the nearest door.", + "B": "You will turn around, walk out of the room, then turn right and walk out of the room again, then turn left and walk to the nearest door, turn left into the room, then walk straight, passing two doors, and you will reach the bathroom.", + "C": "Turn right, then go through two doors.", + "D": "Turn around, exit the room, then turn right again and continue straight past one door to the bathroom.", + "E": "Turn around, walk out, turn right, then turn right to the nearest door; go left in and pass two doors." + }, + "answer_label": "B", + "answer_text": "You will turn around, walk out of the room, then turn right and walk out of the room again, then turn left and walk to the nearest door, turn left into the room, then walk straight, passing two doors, and you will reach the bathroom.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_9", + "correct_answer": "You will turn around, walk out of the room, then turn right and walk out of the room again, then turn left and walk to the nearest door, turn left into the room, then walk straight, passing two doors, and you will reach the bathroom.", + "original_options": { + "A": "Turn left, then head to the nearest door.", + "B": "You will turn around, walk out of the room, then turn right and walk out of the room again, then turn left and walk to the nearest door, turn left into the room, then walk straight, passing two doors, and you will reach the bathroom.", + "C": "Turn right, then go through two doors.", + "D": "Turn around, exit the room, then turn right again and continue straight past one door to the bathroom.", + "E": "Turn around, walk out, turn right, then turn right to the nearest door; go left in and pass two doors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 20 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_10", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the restroom next, how should I go?", + "answer": "You will turn right and walk straight through two doors to reach the restroom.", + "question_chinese": "如果我接下来想去洗手间，我将会怎么走？", + "answer_chinese": "你将要向右转，直走穿过两扇门就到了洗手间。", + "evidence": [ + { + "description": "Motion Trajectory 2", + "timestamps": { + "start": "00:30:18", + "end": "00:31:28" + } + }, + { + "description": "Motion Trajectory 1", + "timestamps": { + "start": "00:00:21", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:34:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You will turn right and walk straight through two doors to reach the restroom.", + "B": "Turn left through two doors.", + "C": "Turn right through three doors.", + "D": "Turn right and go straight through two doors, but you will end up in the kitchen instead of the restroom.", + "E": "Turn right and walk straight, but stop after the first door instead of going through two." + }, + "answer_label": "A", + "answer_text": "You will turn right and walk straight through two doors to reach the restroom.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_10", + "correct_answer": "You will turn right and walk straight through two doors to reach the restroom.", + "original_options": { + "A": "You will turn right and walk straight through two doors to reach the restroom.", + "B": "Turn left through two doors.", + "C": "Turn right through three doors.", + "D": "Turn right and go straight through two doors, but you will end up in the kitchen instead of the restroom.", + "E": "Turn right and walk straight, but stop after the first door instead of going through two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 16 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_14", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I last entered the bedroom, there were several large trash bins in the bedroom.", + "answer": "One.", + "question_chinese": "我上一次进入卧室时，卧室里有几个大垃圾桶", + "answer_chinese": "一个", + "evidence": [ + { + "description": "One.", + "timestamps": { + "start": "00:00:36", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:17:15", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four large trash bins were in the bedroom when you last entered.", + "D": "One.", + "E": "Five large trash bins were in the bedroom the last time you entered." + }, + "answer_label": "D", + "answer_text": "One.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_14", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four large trash bins were in the bedroom when you last entered.", + "D": "One.", + "E": "Five large trash bins were in the bedroom the last time you entered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_13", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I last entered the bedroom, there were several large trash bins in the bedroom?", + "answer": "Zero.", + "question_chinese": "我上一次进入卧室时，卧室里有几个大垃圾桶？", + "answer_chinese": "0个。", + "evidence": [ + { + "description": "There was no large trash can in the bedroom.", + "timestamps": { + "start": "00:58:20", + "end": "00:59:28" + } + } + ], + "question_timestamps": "01:09:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "Three large trash bins were in the bedroom when you last entered.", + "E": "Four large trash bins were in the bedroom the last time you entered." + }, + "answer_label": "A", + "answer_text": "Zero.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_13", + "correct_answer": "Zero.", + "original_options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "Three large trash bins were in the bedroom when you last entered.", + "E": "Four large trash bins were in the bedroom the last time you entered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_4_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I throw the paper towel last time?", + "answer": "You threw the tissue into the big trash can in the living room behind you.", + "question_chinese": "我上次把纸巾丢到哪里了？", + "answer_chinese": "你把纸巾丢到你身后客厅大垃圾桶里了。", + "evidence": [ + { + "description": "You threw the tissues into the large trash can.", + "timestamps": { + "start": "00:35:40", + "end": "00:36:31" + } + } + ], + "question_timestamps": "00:50:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the trash can ahead.", + "B": "In the trash can to your left.", + "C": "You threw the tissue into the big trash can in the living room behind you.", + "D": "You threw the tissue into the big trash can in the living room to your right.", + "E": "You threw the tissue into the big trash can in the living room in front of you, not behind you." + }, + "answer_label": "C", + "answer_text": "You threw the tissue into the big trash can in the living room behind you.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_4_T1", + "correct_answer": "You threw the tissue into the big trash can in the living room behind you.", + "original_options": { + "A": "In the trash can ahead.", + "B": "In the trash can to your left.", + "C": "You threw the tissue into the big trash can in the living room behind you.", + "D": "You threw the tissue into the big trash can in the living room to your right.", + "E": "You threw the tissue into the big trash can in the living room in front of you, not behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 16, + 20 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_4_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I throw the paper towel last time?", + "answer": "You threw the tissue into the trash can in the middle of the room behind you.", + "question_chinese": "我上次把纸巾丢到哪里了？", + "answer_chinese": "你把纸巾丢到了你后面的房间中央来的垃圾桶里了。", + "evidence": [ + { + "description": "You threw the tissues into the large trash can.", + "timestamps": { + "start": "00:35:40", + "end": "00:36:31" + } + } + ], + "question_timestamps": "00:55:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the trash can in front of you.", + "B": "Into the trash can to your left.", + "C": "You tossed the paper towel into the trash can in the middle of the room, in front of you.", + "D": "You threw the tissue into the trash can in the middle of the room, slightly to your right.", + "E": "You threw the tissue into the trash can in the middle of the room behind you." + }, + "answer_label": "E", + "answer_text": "You threw the tissue into the trash can in the middle of the room behind you.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_4_T2", + "correct_answer": "You threw the tissue into the trash can in the middle of the room behind you.", + "original_options": { + "A": "Into the trash can in front of you.", + "B": "Into the trash can to your left.", + "C": "You tossed the paper towel into the trash can in the middle of the room, in front of you.", + "D": "You threw the tissue into the trash can in the middle of the room, slightly to your right.", + "E": "You threw the tissue into the trash can in the middle of the room behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 18 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_2_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The last time I saw the large black trash bin, where was it?", + "answer": "It’s in the bedroom, next to the low cabinet with the coffee cup on it.", + "question_chinese": "我最后一次看见黑色的大垃圾桶时，黑色的大垃圾桶在哪里？", + "answer_chinese": "在卧室里，在放着咖啡杯的矮柜的旁边", + "evidence": [ + { + "description": "It was in the bedroom, next to the low cabinet where the coffee cup is placed on top.", + "timestamps": { + "start": "00:14:16", + "end": "00:14:33" + } + } + ], + "question_timestamps": "00:17:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was in the hallway.", + "B": "It was in the kitchen.", + "C": "It was in the bedroom, but placed across from the low cabinet with the coffee cup on it.", + "D": "It’s in the bedroom, next to the low cabinet with the coffee cup on it.", + "E": "It was in the bedroom, positioned to the left of the low cabinet that had the coffee cup on top." + }, + "answer_label": "D", + "answer_text": "It’s in the bedroom, next to the low cabinet with the coffee cup on it.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_2_T2", + "correct_answer": "It’s in the bedroom, next to the low cabinet with the coffee cup on it.", + "original_options": { + "A": "It was in the hallway.", + "B": "It was in the kitchen.", + "C": "It was in the bedroom, but placed across from the low cabinet with the coffee cup on it.", + "D": "It’s in the bedroom, next to the low cabinet with the coffee cup on it.", + "E": "It was in the bedroom, positioned to the left of the low cabinet that had the coffee cup on top." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 20 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_2_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The last time I saw the large black trash bin, where was it?", + "answer": "It’s to the right of the low cabinet that has the coffee cup on it.", + "question_chinese": "我最后一次看见黑色的大垃圾桶时，黑色的大垃圾桶在哪里？", + "answer_chinese": "它在放着咖啡杯的矮柜的右边", + "evidence": [ + { + "description": "It was in the bedroom, next to the low cabinet where the coffee cup is placed on top.", + "timestamps": { + "start": "00:14:16", + "end": "00:14:33" + } + } + ], + "question_timestamps": "00:22:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was to your left.", + "B": "It’s to the right of the low cabinet that has the coffee cup on it.", + "C": "It was in front of you.", + "D": "It was behind the low cabinet that has the coffee cup on it, rather than on the right side.", + "E": "It was to the left of the low cabinet with the coffee cup, near its front corner." + }, + "answer_label": "B", + "answer_text": "It’s to the right of the low cabinet that has the coffee cup on it.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_2_T1", + "correct_answer": "It’s to the right of the low cabinet that has the coffee cup on it.", + "original_options": { + "A": "It was to your left.", + "B": "It’s to the right of the low cabinet that has the coffee cup on it.", + "C": "It was in front of you.", + "D": "It was behind the low cabinet that has the coffee cup on it, rather than on the right side.", + "E": "It was to the left of the low cabinet with the coffee cup, near its front corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 17 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_11", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the trash can now?", + "answer": "You cannot touch the trash can.", + "question_chinese": "我现在能接触到垃圾桶吗？", + "answer_chinese": "你不能接触到垃圾桶。", + "evidence": [ + { + "description": "The trash can is in a position you cannot reach.", + "timestamps": { + "start": "00:10:05", + "end": "00:11:51" + } + } + ], + "question_timestamps": "00:17:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot touch the trash can.", + "B": "You can reach the trash can." + }, + "answer_label": "A", + "answer_text": "You cannot touch the trash can.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_11", + "correct_answer": "You cannot touch the trash can.", + "original_options": { + "A": "You cannot touch the trash can.", + "B": "You can reach the trash can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_6_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the trash bin right now?", + "answer": "There isn’t a trash can in front of you, so you can’t reach a trash can.", + "question_chinese": "我现在能接触到垃圾桶吗？", + "answer_chinese": "你面前没有垃圾桶，你不能接触到垃圾桶。", + "evidence": [ + { + "description": "The trash can is in a position you cannot reach.", + "timestamps": { + "start": "00:31:17", + "end": "00:31:33" + } + } + ], + "question_timestamps": "00:34:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There is a trash can in front of you, so you can reach it.", + "B": "There isn’t a trash can in front of you, so you can’t reach a trash can." + }, + "answer_label": "B", + "answer_text": "There isn’t a trash can in front of you, so you can’t reach a trash can.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_6_T2", + "correct_answer": "There isn’t a trash can in front of you, so you can’t reach a trash can.", + "original_options": { + "A": "There is a trash can in front of you, so you can reach it.", + "B": "There isn’t a trash can in front of you, so you can’t reach a trash can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_6_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the trash bin right now?", + "answer": "You can reach the trash can in front of you.", + "question_chinese": "我现在能接触到垃圾桶吗？", + "answer_chinese": "你可以接触到你面前的垃圾桶。", + "evidence": [ + { + "description": "The trash can is in a position you can reach.", + "timestamps": { + "start": "00:10:05", + "end": "00:11:51" + } + } + ], + "question_timestamps": "00:35:14", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the trash can in front of you.", + "B": "There isn’t a trash can in front of you, so you can’t reach it." + }, + "answer_label": "A", + "answer_text": "You can reach the trash can in front of you.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_6_T1", + "correct_answer": "You can reach the trash can in front of you.", + "original_options": { + "A": "You can reach the trash can in front of you.", + "B": "There isn’t a trash can in front of you, so you can’t reach it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the shovel I often use?", + "answer": "The spade is next to the coffee cup.", + "question_chinese": "我经常拿的铲刀在哪里？", + "answer_chinese": "铲刀在咖啡杯的旁边", + "evidence": [ + { + "description": "The spade is next to the coffee cup", + "timestamps": { + "start": "00:29:49", + "end": "00:30:08" + } + } + ], + "question_timestamps": "00:34:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The spade is under the coffee cup.", + "B": "The spade is next to the coffee cup.", + "C": "The spade is behind the coffee cup.", + "D": "The spade is on the low cabinet, but it’s set far away from the coffee cup.", + "E": "The spade is near the coffee cup, but it is not placed directly next to it." + }, + "answer_label": "B", + "answer_text": "The spade is next to the coffee cup.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_2", + "correct_answer": "The spade is next to the coffee cup.", + "original_options": { + "A": "The spade is under the coffee cup.", + "B": "The spade is next to the coffee cup.", + "C": "The spade is behind the coffee cup.", + "D": "The spade is on the low cabinet, but it’s set far away from the coffee cup.", + "E": "The spade is near the coffee cup, but it is not placed directly next to it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4", + "video_length": "01:09:05", + "qid": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the shovel I often use?", + "answer": "The spade is between the coffee cup and the tissue box.", + "question_chinese": "我经常拿的铲刀在哪里？", + "answer_chinese": "铲刀在咖啡杯和纸巾盒中间。", + "evidence": [ + { + "description": "The spade is between the coffee cup and the tissue box.", + "timestamps": { + "start": "00:50:16", + "end": "00:50:22" + } + } + ], + "question_timestamps": "00:50:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The spade is between the coffee cup and the tissue box.", + "B": "The spade is next to the coffee cup.", + "C": "The spade is next to the tissue box.", + "D": "The spade is to the right of the coffee cup and not positioned between the coffee cup and the tissue box.", + "E": "The spade is closer to the tissue box, sitting off to one side rather than between the coffee cup and it." + }, + "answer_label": "A", + "answer_text": "The spade is between the coffee cup and the tissue box.", + "q_id": "0b9ee926-00f1-4b22-9e83-b664c5e465e4_1", + "correct_answer": "The spade is between the coffee cup and the tissue box.", + "original_options": { + "A": "The spade is between the coffee cup and the tissue box.", + "B": "The spade is next to the coffee cup.", + "C": "The spade is next to the tissue box.", + "D": "The spade is to the right of the coffee cup and not positioned between the coffee cup and the tissue box.", + "E": "The spade is closer to the tissue box, sitting off to one side rather than between the coffee cup and it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 21, + 21 + ] + } +] diff --git a/QAs/0d2231f9-d67a-4df0-b36f-40cbf374ed03-mcq.json b/QAs/0d2231f9-d67a-4df0-b36f-40cbf374ed03-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4a98e18c6d50a0d9791024e030fd2af5fb81aacf --- /dev/null +++ b/QAs/0d2231f9-d67a-4df0-b36f-40cbf374ed03-mcq.json @@ -0,0 +1,97 @@ +[ + { + "video_uid": "0d2231f9-d67a-4df0-b36f-40cbf374ed03", + "video_length": "00:54:10", + "qid": "0d2231f9-d67a-4df0-b36f-40cbf374ed03_1_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I remember bringing the set square here—can I reach it right now?", + "answer": "Yes. The set square is to your left.", + "question_chinese": "我记得我拿了三角板到这里来，我现在能够到这个东西吗？", + "answer_chinese": "是的，三角板就在你的左边。", + "evidence": [ + { + "description": "The set square is on your left.", + "timestamps": { + "start": "00:00:29", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. The set square is to your left.", + "B": "No. The set square is to your right." + }, + "answer_label": "A", + "answer_text": "Yes. The set square is to your left.", + "q_id": "0d2231f9-d67a-4df0-b36f-40cbf374ed03_1_T1", + "correct_answer": "Yes. The set square is to your left.", + "original_options": { + "A": "Yes. The set square is to your left.", + "B": "No. The set square is to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "0d2231f9-d67a-4df0-b36f-40cbf374ed03", + "video_length": "00:54:10", + "qid": "0d2231f9-d67a-4df0-b36f-40cbf374ed03_1_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I remember bringing the set square here—can I reach it right now?", + "answer": "No, you can’t reach it—you’re already on the ground.", + "question_chinese": "我记得我拿了三角板到这里来，我现在能够到这个东西吗？", + "answer_chinese": "不，你够不到，你已经在地面上了", + "evidence": [ + { + "description": "The set square is on your left.", + "timestamps": { + "start": "00:00:29", + "end": "00:00:34" + } + }, + { + "description": "You climbed down using the ladder.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:37" + } + } + ], + "question_timestamps": "00:04:48", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t reach it—you’re already on the ground.", + "B": "Yes, you can reach it—you’re already on the ground." + }, + "answer_label": "A", + "answer_text": "No, you can’t reach it—you’re already on the ground.", + "q_id": "0d2231f9-d67a-4df0-b36f-40cbf374ed03_1_T2", + "correct_answer": "No, you can’t reach it—you’re already on the ground.", + "original_options": { + "A": "No, you can’t reach it—you’re already on the ground.", + "B": "Yes, you can reach it—you’re already on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + } +] diff --git a/QAs/0dcf7a01-8bbc-484a-b093-ff6d7095cfd4-mcq.json b/QAs/0dcf7a01-8bbc-484a-b093-ff6d7095cfd4-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5ad1b87128dae57b309e27228874600486de7bce --- /dev/null +++ b/QAs/0dcf7a01-8bbc-484a-b093-ff6d7095cfd4-mcq.json @@ -0,0 +1,422 @@ +[ + { + "video_uid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4", + "video_length": "00:35:35", + "qid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_6", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get to the power strip?", + "answer": "No.", + "question_chinese": "我可以拿到电源板吗？", + "answer_chinese": "不。", + "evidence": [ + { + "description": "The power strip is placed in the cabinet in the warehouse.", + "timestamps": { + "start": "00:24:47", + "end": "00:24:58" + } + } + ], + "question_timestamps": "00:25:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "No.", + "B": "Yes.", + "C": "Maybe.", + "D": "Yes, you can reach it if you move closer to the power strip first.", + "E": "Yes, you can get to the power strip right now without having to move anywhere else." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_6", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes.", + "C": "Maybe.", + "D": "Yes, you can reach it if you move closer to the power strip first.", + "E": "Yes, you can get to the power strip right now without having to move anywhere else." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4", + "video_length": "00:35:35", + "qid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I cut this silvery-white iron plate with a saw?", + "answer": "One time.", + "question_chinese": "我用锯子锯了多少次这个银白色的铁板？", + "answer_chinese": "1次。", + "evidence": [ + { + "description": "First time sawing.", + "timestamps": { + "start": "00:02:29", + "end": "00:03:04" + } + } + ], + "question_timestamps": "00:03:07", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Six times.", + "C": "You have cut the silvery-white iron plate with a saw five times.", + "D": "You have cut the silvery-white iron plate with a saw ten times in total.", + "E": "One time." + }, + "answer_label": "E", + "answer_text": "One time.", + "q_id": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_1", + "correct_answer": "One time.", + "original_options": { + "A": "Two times.", + "B": "Six times.", + "C": "You have cut the silvery-white iron plate with a saw five times.", + "D": "You have cut the silvery-white iron plate with a saw ten times in total.", + "E": "One time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4", + "video_length": "00:35:35", + "qid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I cut this silvery-white iron plate with a saw?", + "answer": "Four times.", + "question_chinese": "我用锯子锯了多少次这个银白色的铁板？", + "answer_chinese": "四次。", + "evidence": [ + { + "description": "First time sawing.", + "timestamps": { + "start": "00:02:29", + "end": "00:03:04" + } + }, + { + "description": "Second sawing.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:24" + } + }, + { + "description": "The third sawing.", + "timestamps": { + "start": "00:03:51", + "end": "00:04:14" + } + }, + { + "description": "The fourth saw.", + "timestamps": { + "start": "00:04:35", + "end": "00:05:01" + } + } + ], + "question_timestamps": "00:18:13", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three times.", + "B": "Six times.", + "C": "Four times.", + "D": "You have cut the silvery-white iron plate with a saw five times.", + "E": "You have cut the silvery-white iron plate with a saw seven times overall." + }, + "answer_label": "C", + "answer_text": "Four times.", + "q_id": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_2", + "correct_answer": "Four times.", + "original_options": { + "A": "Three times.", + "B": "Six times.", + "C": "Four times.", + "D": "You have cut the silvery-white iron plate with a saw five times.", + "E": "You have cut the silvery-white iron plate with a saw seven times overall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4", + "video_length": "00:35:35", + "qid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_9", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I repaired the white iron sheet?", + "answer": "Six times.", + "question_chinese": "我维修了白色铁片多少次呢？", + "answer_chinese": "六次。", + "evidence": [ + { + "description": "First time.", + "timestamps": { + "start": "00:02:23", + "end": "00:03:13" + } + }, + { + "description": "Second time.", + "timestamps": { + "start": "00:03:47", + "end": "00:04:14" + } + } + ], + "question_timestamps": "00:08:39", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five times.", + "B": "Six times.", + "C": "Two times.", + "D": "You have repaired the white iron sheet eight times in total so far.", + "E": "You have repaired the white iron sheet ten times altogether up to now." + }, + "answer_label": "B", + "answer_text": "Six times.", + "q_id": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_9", + "correct_answer": "Six times.", + "original_options": { + "A": "Five times.", + "B": "Six times.", + "C": "Two times.", + "D": "You have repaired the white iron sheet eight times in total so far.", + "E": "You have repaired the white iron sheet ten times altogether up to now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4", + "video_length": "00:35:35", + "qid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_10", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I repaired the white iron sheet?", + "answer": "Four times.", + "question_chinese": "我维修了白色铁片多少次呢？", + "answer_chinese": "四次。", + "evidence": [ + { + "description": "First time.", + "timestamps": { + "start": "00:02:23", + "end": "00:03:13" + } + }, + { + "description": "Second time.", + "timestamps": { + "start": "00:03:47", + "end": "00:04:14" + } + }, + { + "description": "Used a drill to smooth the patch.", + "timestamps": { + "start": "00:17:27", + "end": "00:18:04" + } + }, + { + "description": "Used a drill to smooth the patch again.", + "timestamps": { + "start": "00:19:57", + "end": "00:20:30" + } + } + ], + "question_timestamps": "00:20:33", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three times.", + "B": "Five times.", + "C": "Six times, counting the first fix and five later touch-ups on the same white iron sheet.", + "D": "Four times.", + "E": "Two times, once early on and once again after the sheet loosened and needed another repair." + }, + "answer_label": "D", + "answer_text": "Four times.", + "q_id": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_10", + "correct_answer": "Four times.", + "original_options": { + "A": "Three times.", + "B": "Five times.", + "C": "Six times, counting the first fix and five later touch-ups on the same white iron sheet.", + "D": "Four times.", + "E": "Two times, once early on and once again after the sheet loosened and needed another repair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4", + "video_length": "00:35:35", + "qid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_8", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I get a wrench to repair my bike?", + "answer": "On the platform of the bicycle rack at your feet.", + "question_chinese": "我能去哪里获取维修单车用的扳手呢？", + "answer_chinese": "在你脚边的单车架的平台上。", + "evidence": [ + { + "description": "On the platform of the bicycle rack at your feet.", + "timestamps": { + "start": "00:09:46", + "end": "00:13:58" + } + } + ], + "question_timestamps": "00:14:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the tool cabinet drawer.", + "B": "On the platform of the bicycle rack at your feet.", + "C": "On the shelf to your left.", + "D": "Under the workbench in the toolbox beside your feet, where the other hand tools are kept.", + "E": "By the shed door next to the bicycle rack, near where you entered to work on the bike." + }, + "answer_label": "B", + "answer_text": "On the platform of the bicycle rack at your feet.", + "q_id": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_8", + "correct_answer": "On the platform of the bicycle rack at your feet.", + "original_options": { + "A": "In the tool cabinet drawer.", + "B": "On the platform of the bicycle rack at your feet.", + "C": "On the shelf to your left.", + "D": "Under the workbench in the toolbox beside your feet, where the other hand tools are kept.", + "E": "By the shed door next to the bicycle rack, near where you entered to work on the bike." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 18 + ] + }, + { + "video_uid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4", + "video_length": "00:35:35", + "qid": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_7", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I get a wrench to repair my bike?", + "answer": "In the top drawer of the workbench.", + "question_chinese": "我能去哪里获取维修单车用的扳手呢？", + "answer_chinese": "工作台的最高层抽屉里。", + "evidence": [ + { + "description": "The wrench is in the drawer.", + "timestamps": { + "start": "00:23:28", + "end": "00:23:38" + } + } + ], + "question_timestamps": "00:23:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the bottom cabinet drawer.", + "B": "On the shelf above the workbench.", + "C": "In the side pocket of the bicycle bag, tucked with the other small repair items.", + "D": "In the top drawer of the workbench.", + "E": "In the toolbox next to the garage door, where you grabbed tools earlier for the bike." + }, + "answer_label": "D", + "answer_text": "In the top drawer of the workbench.", + "q_id": "0dcf7a01-8bbc-484a-b093-ff6d7095cfd4_7", + "correct_answer": "In the top drawer of the workbench.", + "original_options": { + "A": "In the bottom cabinet drawer.", + "B": "On the shelf above the workbench.", + "C": "In the side pocket of the bicycle bag, tucked with the other small repair items.", + "D": "In the top drawer of the workbench.", + "E": "In the toolbox next to the garage door, where you grabbed tools earlier for the bike." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + } +] diff --git a/QAs/0e66de5b-08e2-4cbb-8f2d-0028168218b0-mcq.json b/QAs/0e66de5b-08e2-4cbb-8f2d-0028168218b0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e4ae8a380f5848da8449e7a75935f3c9e85ca8f2 --- /dev/null +++ b/QAs/0e66de5b-08e2-4cbb-8f2d-0028168218b0-mcq.json @@ -0,0 +1,1446 @@ +[ + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the tape measure relative to me?", + "answer": "The tape measure is on your right.", + "question_chinese": "卷尺相对我是什么位置？", + "answer_chinese": "卷尺在你的右侧。", + "evidence": [ + { + "description": "The tape measure is on your right.", + "timestamps": { + "start": "00:04:09", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:04:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The tape measure is on your left.", + "B": "The tape measure is on your right.", + "C": "The tape measure is in front of you.", + "D": "The tape measure is behind you, not on the side you’re currently facing.", + "E": "The tape measure is under the desk near your feet, rather than out to your right." + }, + "answer_label": "B", + "answer_text": "The tape measure is on your right.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_1", + "correct_answer": "The tape measure is on your right.", + "original_options": { + "A": "The tape measure is on your left.", + "B": "The tape measure is on your right.", + "C": "The tape measure is in front of you.", + "D": "The tape measure is behind you, not on the side you’re currently facing.", + "E": "The tape measure is under the desk near your feet, rather than out to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 14, + 16 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_1_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the tape measure relative to me?", + "answer": "The tape measure is now on the right side of you.", + "question_chinese": "卷尺相对我是什么位置？", + "answer_chinese": "卷尺现在在你的右侧。", + "evidence": [ + { + "description": "The tape measure is on your right.", + "timestamps": { + "start": "00:04:09", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:04:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The tape measure is now on your left.", + "B": "The tape measure is now behind you.", + "C": "The tape measure is now in front of you, not off to either side as you stand there.", + "D": "The tape measure is now on the right side of you.", + "E": "The tape measure is now under the desk near your feet, rather than on your right side." + }, + "answer_label": "D", + "answer_text": "The tape measure is now on the right side of you.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_1_T1", + "correct_answer": "The tape measure is now on the right side of you.", + "original_options": { + "A": "The tape measure is now on your left.", + "B": "The tape measure is now behind you.", + "C": "The tape measure is now in front of you, not off to either side as you stand there.", + "D": "The tape measure is now on the right side of you.", + "E": "The tape measure is now under the desk near your feet, rather than on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 17 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_1_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the tape measure relative to me?", + "answer": "The tape measure is now on the left side of you.", + "question_chinese": "卷尺相对我是什么位置？", + "answer_chinese": "卷尺现在在你的左侧。", + "evidence": [ + { + "description": "The tape measure is on your left.", + "timestamps": { + "start": "00:16:46", + "end": "00:17:25" + } + } + ], + "question_timestamps": "00:17:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The tape measure is now on your right.", + "B": "The tape measure is now behind you.", + "C": "The tape measure is now in front of you, not positioned along your left side anymore.", + "D": "The tape measure is now on the left side of you.", + "E": "The tape measure is now under the desk by your feet, rather than sitting to your left." + }, + "answer_label": "D", + "answer_text": "The tape measure is now on the left side of you.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_1_T2", + "correct_answer": "The tape measure is now on the left side of you.", + "original_options": { + "A": "The tape measure is now on your right.", + "B": "The tape measure is now behind you.", + "C": "The tape measure is now in front of you, not positioned along your left side anymore.", + "D": "The tape measure is now on the left side of you.", + "E": "The tape measure is now under the desk by your feet, rather than sitting to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 17 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_19", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can now relative to me?", + "answer": "The trash can is now behind you.", + "question_chinese": "垃圾桶现在相对我是什么位置？", + "answer_chinese": "垃圾桶现在在你的后方。", + "evidence": [ + { + "description": "The trash can is now behind you.", + "timestamps": { + "start": "00:35:37", + "end": "00:35:39" + } + } + ], + "question_timestamps": "00:35:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The trash can is now ahead of you.", + "B": "The trash can is now behind you.", + "C": "The trash can is now to your left.", + "D": "The trash can is now on your right, not positioned behind you where you can’t see it.", + "E": "The trash can is now far in front of you, not behind you where it would be out of view." + }, + "answer_label": "B", + "answer_text": "The trash can is now behind you.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_19", + "correct_answer": "The trash can is now behind you.", + "original_options": { + "A": "The trash can is now ahead of you.", + "B": "The trash can is now behind you.", + "C": "The trash can is now to your left.", + "D": "The trash can is now on your right, not positioned behind you where you can’t see it.", + "E": "The trash can is now far in front of you, not behind you where it would be out of view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 20 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_20", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can now relative to me?", + "answer": "The trash can is now on your right.", + "question_chinese": "垃圾桶现在相对我是什么位置？", + "answer_chinese": "垃圾桶现在在你的右侧。", + "evidence": [ + { + "description": "The trash can is now on your right.", + "timestamps": { + "start": "00:35:37", + "end": "00:35:40" + } + } + ], + "question_timestamps": "00:39:36", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The trash can is now on your left.", + "B": "The trash can is now ahead of you.", + "C": "The trash can is now behind you, so you’d need to turn around to reach it.", + "D": "The trash can is now beside the table, not on your right side where you’re standing.", + "E": "The trash can is now on your right." + }, + "answer_label": "E", + "answer_text": "The trash can is now on your right.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_20", + "correct_answer": "The trash can is now on your right.", + "original_options": { + "A": "The trash can is now on your left.", + "B": "The trash can is now ahead of you.", + "C": "The trash can is now behind you, so you’d need to turn around to reach it.", + "D": "The trash can is now beside the table, not on your right side where you’re standing.", + "E": "The trash can is now on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 17 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_5_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I finished drinking water, where did I put the cup?", + "answer": "After you finished drinking water, you threw the cup into the trash can to your right.", + "question_chinese": "我刚刚喝完水把杯子放在什么地方了？", + "answer_chinese": "你喝完水把杯子扔进你右边的垃圾桶里了。", + "evidence": [ + { + "description": "You threw the cup into the trash can after drinking water and the trash is on your right now.", + "timestamps": { + "start": "00:27:05", + "end": "00:27:11" + } + } + ], + "question_timestamps": "00:27:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "After you finished drinking water, you threw the cup into the trash can to your right.", + "B": "Into the trash can in front of you.", + "C": "Into the trash can to your left.", + "D": "After drinking, you tossed the cup into the trash can behind you instead of to your right.", + "E": "Once you finished the water, you placed the cup on the counter in front of you instead of trashing it." + }, + "answer_label": "A", + "answer_text": "After you finished drinking water, you threw the cup into the trash can to your right.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_5_T1", + "correct_answer": "After you finished drinking water, you threw the cup into the trash can to your right.", + "original_options": { + "A": "After you finished drinking water, you threw the cup into the trash can to your right.", + "B": "Into the trash can in front of you.", + "C": "Into the trash can to your left.", + "D": "After drinking, you tossed the cup into the trash can behind you instead of to your right.", + "E": "Once you finished the water, you placed the cup on the counter in front of you instead of trashing it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 20 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_5_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I finished drinking water, where did I put the cup?", + "answer": "After you finished drinking water, you threw the cup into the trash can to your left.", + "question_chinese": "我刚刚喝完水把杯子放在什么地方了？", + "answer_chinese": "你喝完水把杯子扔进你左边的垃圾桶里了。", + "evidence": [ + { + "description": "You threw the cup into the trash can after drinking water.", + "timestamps": { + "start": "00:27:05", + "end": "00:27:11" + } + }, + { + "description": "The trash is on your left now.", + "timestamps": { + "start": "00:27:41", + "end": "00:27:51" + } + } + ], + "question_timestamps": "00:27:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the trash can to your right.", + "B": "Into the trash can in front of you.", + "C": "After you finished drinking, you threw the cup into the trash can behind you, not the one on your left.", + "D": "After you finished drinking water, you threw the cup into the trash can to your left.", + "E": "Once you were done with the water, you tossed the cup into the trash can to your right." + }, + "answer_label": "D", + "answer_text": "After you finished drinking water, you threw the cup into the trash can to your left.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_5_T2", + "correct_answer": "After you finished drinking water, you threw the cup into the trash can to your left.", + "original_options": { + "A": "Into the trash can to your right.", + "B": "Into the trash can in front of you.", + "C": "After you finished drinking, you threw the cup into the trash can behind you, not the one on your left.", + "D": "After you finished drinking water, you threw the cup into the trash can to your left.", + "E": "Once you were done with the water, you tossed the cup into the trash can to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 20, + 18 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_10", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the cup just after drinking water?", + "answer": "You threw the cup into the trash can after drinking water.", + "question_chinese": "我刚刚喝完水把杯子放在什么地方了？", + "answer_chinese": "你喝完水把杯子扔进垃圾桶里了。", + "evidence": [ + { + "description": "You threw the cup into the trash can after drinking water.", + "timestamps": { + "start": "00:27:05", + "end": "00:27:11" + } + } + ], + "question_timestamps": "00:43:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You threw the cup into the trash can after drinking water.", + "B": "You placed it on the counter.", + "C": "You dropped it onto the floor.", + "D": "Right after drinking water, you set the cup down beside the refrigerator instead of throwing it away.", + "E": "After you took a drink, you left the cup near the window rather than putting it in the trash." + }, + "answer_label": "A", + "answer_text": "You threw the cup into the trash can after drinking water.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_10", + "correct_answer": "You threw the cup into the trash can after drinking water.", + "original_options": { + "A": "You threw the cup into the trash can after drinking water.", + "B": "You placed it on the counter.", + "C": "You dropped it onto the floor.", + "D": "Right after drinking water, you set the cup down beside the refrigerator instead of throwing it away.", + "E": "After you took a drink, you left the cup near the window rather than putting it in the trash." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_5", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I put my mobile phone just after I took it out?", + "answer": "You put your mobile phone back in your pocket.", + "question_chinese": "我刚刚拿出手机后把它放在哪里了？", + "answer_chinese": "你把手机放回口袋里了。", + "evidence": [ + { + "description": "You put your mobile phone back in your pocket.", + "timestamps": { + "start": "00:30:03", + "end": "00:30:08" + } + } + ], + "question_timestamps": "00:30:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it on the table.", + "B": "You put it inside your bag.", + "C": "After taking it out, you set your mobile phone on the kitchen counter instead of putting it away.", + "D": "Just after you took it out, you rested your mobile phone on your lap rather than in your pocket.", + "E": "You put your mobile phone back in your pocket." + }, + "answer_label": "E", + "answer_text": "You put your mobile phone back in your pocket.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_5", + "correct_answer": "You put your mobile phone back in your pocket.", + "original_options": { + "A": "You put it on the table.", + "B": "You put it inside your bag.", + "C": "After taking it out, you set your mobile phone on the kitchen counter instead of putting it away.", + "D": "Just after you took it out, you rested your mobile phone on your lap rather than in your pocket.", + "E": "You put your mobile phone back in your pocket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_6", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I put my mobile phone just after I took it out?", + "answer": "You put your mobile phone back in your pocket.", + "question_chinese": "我刚刚拿出手机后把它放在哪里了？", + "answer_chinese": "你把手机放回口袋里了。", + "evidence": [ + { + "description": "You put your mobile phone back in your pocket.", + "timestamps": { + "start": "00:30:03", + "end": "00:30:08" + } + } + ], + "question_timestamps": "00:33:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it inside your bag.", + "B": "You put it on the counter.", + "C": "After taking it out, you placed your mobile phone on the table beside you instead of putting it away.", + "D": "You put your mobile phone back in your pocket.", + "E": "Right after you took it out, you set your mobile phone on the chair’s armrest rather than in your pocket." + }, + "answer_label": "D", + "answer_text": "You put your mobile phone back in your pocket.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_6", + "correct_answer": "You put your mobile phone back in your pocket.", + "original_options": { + "A": "You put it inside your bag.", + "B": "You put it on the counter.", + "C": "After taking it out, you placed your mobile phone on the table beside you instead of putting it away.", + "D": "You put your mobile phone back in your pocket.", + "E": "Right after you took it out, you set your mobile phone on the chair’s armrest rather than in your pocket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 21 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_17", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles were placed on the left when I just walked into the store?", + "answer": "When you just walked into the store, there was a bottle on the left.", + "question_chinese": "我刚刚走进店里时左边放着几个瓶子？", + "answer_chinese": "你刚刚走进店里时左边放着一个瓶子。", + "evidence": [ + { + "description": "When you just walked into the store, there was a bottle on the left.", + "timestamps": { + "start": "00:09:11", + "end": "00:09:12" + } + } + ], + "question_timestamps": "00:09:14", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There were two bottles on the left.", + "B": "When you just walked into the store, there was a bottle on the left.", + "C": "There were three bottles on the left.", + "D": "When you just walked into the store, there were four bottles placed on the left side.", + "E": "When you just walked into the store, there were five bottles lined up on the left." + }, + "answer_label": "B", + "answer_text": "When you just walked into the store, there was a bottle on the left.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_17", + "correct_answer": "When you just walked into the store, there was a bottle on the left.", + "original_options": { + "A": "There were two bottles on the left.", + "B": "When you just walked into the store, there was a bottle on the left.", + "C": "There were three bottles on the left.", + "D": "When you just walked into the store, there were four bottles placed on the left side.", + "E": "When you just walked into the store, there were five bottles lined up on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_18", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles were placed on the left when I just walked into the store?", + "answer": "When you just walked into the store, there was a bottle on the left.", + "question_chinese": "我刚刚走进店里时左边放着几个瓶子？", + "answer_chinese": "你刚刚走进店里时左边放着一个瓶子。", + "evidence": [ + { + "description": "When you just walked into the store, there was a bottle on the left.", + "timestamps": { + "start": "00:35:22", + "end": "00:35:23" + } + } + ], + "question_timestamps": "00:35:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There were two bottles on the left.", + "B": "There were four bottles on the left.", + "C": "When you just walked into the store, there was a bottle on the left.", + "D": "When you just walked into the store, there were three bottles positioned on the left.", + "E": "When you just walked into the store, there were five bottles placed on the left side." + }, + "answer_label": "C", + "answer_text": "When you just walked into the store, there was a bottle on the left.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_18", + "correct_answer": "When you just walked into the store, there was a bottle on the left.", + "original_options": { + "A": "There were two bottles on the left.", + "B": "There were four bottles on the left.", + "C": "When you just walked into the store, there was a bottle on the left.", + "D": "When you just walked into the store, there were three bottles positioned on the left.", + "E": "When you just walked into the store, there were five bottles placed on the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_9_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I first walked into the shop, where were those bottles relative to me?", + "answer": "The bottle that was there when you first walked into the shop is to your left.", + "question_chinese": "我刚刚走进店里时放着的几个瓶子在相对于我的什么位置？", + "answer_chinese": "你刚刚走进店里时放着的一个瓶子在你的左边。", + "evidence": [ + { + "description": "When you just walked into the store, there was a bottle on the left.", + "timestamps": { + "start": "00:09:11", + "end": "00:09:12" + } + } + ], + "question_timestamps": "00:09:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "They were to your right.", + "B": "The bottle that was there when you first walked into the shop is to your left.", + "C": "They were in front of you.", + "D": "The bottle that was there when you first walked into the shop is behind you.", + "E": "The bottle that was there when you first walked into the shop is to your left and slightly behind you." + }, + "answer_label": "B", + "answer_text": "The bottle that was there when you first walked into the shop is to your left.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_9_T1", + "correct_answer": "The bottle that was there when you first walked into the shop is to your left.", + "original_options": { + "A": "They were to your right.", + "B": "The bottle that was there when you first walked into the shop is to your left.", + "C": "They were in front of you.", + "D": "The bottle that was there when you first walked into the shop is behind you.", + "E": "The bottle that was there when you first walked into the shop is to your left and slightly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 20 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_9_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I first walked into the shop, where were those bottles relative to me?", + "answer": "The bottle is behind you.", + "question_chinese": "我刚刚走进店里时放着的几个瓶子在相对于我的什么位置？", + "answer_chinese": "瓶子在你的后面。", + "evidence": [ + { + "description": "When you just walked into the store, there was a bottle on the left.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:23" + } + } + ], + "question_timestamps": "00:09:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The bottle is in front of you.", + "B": "The bottle is behind you.", + "C": "The bottle is to your left.", + "D": "The bottle is to your right when you first enter the shop and look around.", + "E": "The bottle is above you near the top shelf as you walk into the shop." + }, + "answer_label": "B", + "answer_text": "The bottle is behind you.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_9_T2", + "correct_answer": "The bottle is behind you.", + "original_options": { + "A": "The bottle is in front of you.", + "B": "The bottle is behind you.", + "C": "The bottle is to your left.", + "D": "The bottle is to your right when you first enter the shop and look around.", + "E": "The bottle is above you near the top shelf as you walk into the shop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 15 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_6_T2", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What should I do to turn on the electric fan?", + "answer": "Reach out your right hand to the electric fan switch and turn it on.", + "question_chinese": "我想打开电风扇应该怎么做？", + "answer_chinese": "你伸出右手够到电风扇的按钮，打开按钮即可。", + "evidence": [ + { + "description": "The electric fan is on the left side of the door.", + "timestamps": { + "start": "00:09:25", + "end": "00:09:30" + } + } + ], + "question_timestamps": "00:09:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use your left hand to switch it on.", + "B": "Flip the fan switch with your finger.", + "C": "Walk to the right side of the doorway first, then turn on the electric fan switch.", + "D": "Reach out your right hand to the electric fan switch and turn it on.", + "E": "Turn around and step back a little before reaching for the electric fan switch to turn it on." + }, + "answer_label": "D", + "answer_text": "Reach out your right hand to the electric fan switch and turn it on.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_6_T2", + "correct_answer": "Reach out your right hand to the electric fan switch and turn it on.", + "original_options": { + "A": "Use your left hand to switch it on.", + "B": "Flip the fan switch with your finger.", + "C": "Walk to the right side of the doorway first, then turn on the electric fan switch.", + "D": "Reach out your right hand to the electric fan switch and turn it on.", + "E": "Turn around and step back a little before reaching for the electric fan switch to turn it on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 18 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_6_T1", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What should I do to turn on the electric fan?", + "answer": "You should first walk to the left side of the doorway, then turn on the electric fan switch.", + "question_chinese": "我想打开电风扇应该怎么做？", + "answer_chinese": "你应该先走到门口的左侧，然后打开电风扇的按钮。", + "evidence": [ + { + "description": "The electric fan is on the left side of the door.", + "timestamps": { + "start": "00:09:40", + "end": "00:09:42" + } + } + ], + "question_timestamps": "00:15:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk to the right side first.", + "B": "Go straight ahead first.", + "C": "You should first walk to the left side of the doorway, then turn on the electric fan switch.", + "D": "You should first walk behind the doorway area, then turn on the electric fan switch.", + "E": "You should first walk to the front side of the doorway, then turn on the electric fan switch." + }, + "answer_label": "C", + "answer_text": "You should first walk to the left side of the doorway, then turn on the electric fan switch.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_6_T1", + "correct_answer": "You should first walk to the left side of the doorway, then turn on the electric fan switch.", + "original_options": { + "A": "Walk to the right side first.", + "B": "Go straight ahead first.", + "C": "You should first walk to the left side of the doorway, then turn on the electric fan switch.", + "D": "You should first walk behind the doorway area, then turn on the electric fan switch.", + "E": "You should first walk to the front side of the doorway, then turn on the electric fan switch." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 15, + 18 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_7_T1", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the electric fan right now?", + "answer": "You can directly reach the electric fan right now.", + "question_chinese": "我现在能够直接触碰到电风扇吗？", + "answer_chinese": "你现在可以直接触碰到电风扇。", + "evidence": [ + { + "description": "The electric fan is on your right.", + "timestamps": { + "start": "00:09:26", + "end": "00:09:35" + } + } + ], + "question_timestamps": "00:09:38", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can't directly reach the electric fan right now.", + "B": "You can directly reach the electric fan right now." + }, + "answer_label": "B", + "answer_text": "You can directly reach the electric fan right now.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_7_T1", + "correct_answer": "You can directly reach the electric fan right now.", + "original_options": { + "A": "You can't directly reach the electric fan right now.", + "B": "You can directly reach the electric fan right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_13", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the electric fan directly now?", + "answer": "You can't touch the electric fan directly now.", + "question_chinese": "我现在能够直接触碰到电风扇吗？", + "answer_chinese": "你现在无法直接触碰到电风扇。", + "evidence": [ + { + "description": "The electric fan is on the left side of the door, some distance from where you are now.", + "timestamps": { + "start": "00:18:01", + "end": "00:18:05" + } + } + ], + "question_timestamps": "00:21:32", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't touch the electric fan directly now.", + "B": "You can touch the electric fan directly now." + }, + "answer_label": "A", + "answer_text": "You can't touch the electric fan directly now.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_13", + "correct_answer": "You can't touch the electric fan directly now.", + "original_options": { + "A": "You can't touch the electric fan directly now.", + "B": "You can touch the electric fan directly now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_7_T2", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the electric fan right now?", + "answer": "You can't directly reach the electric fan right now.", + "question_chinese": "我现在能够直接触碰到电风扇吗？", + "answer_chinese": "你现在无法直接触碰到电风扇。", + "evidence": [ + { + "description": "The electric fan is on the left side of the door, some distance from where you are now.", + "timestamps": { + "start": "00:27:40", + "end": "00:28:01" + } + } + ], + "question_timestamps": "00:28:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can't directly reach the electric fan right now.", + "B": "You can directly reach the electric fan right now." + }, + "answer_label": "A", + "answer_text": "You can't directly reach the electric fan right now.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_7_T2", + "correct_answer": "You can't directly reach the electric fan right now.", + "original_options": { + "A": "You can't directly reach the electric fan right now.", + "B": "You can directly reach the electric fan right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_2_T2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the red plastic stool right now?", + "answer": "You can't directly reach the red plastic stool.", + "question_chinese": "我现在能直接触碰到红色塑料凳吗？", + "answer_chinese": "你不能直接触碰到红色塑料凳。", + "evidence": [ + { + "description": "You can't touch the red plastic stool directly.", + "timestamps": { + "start": "00:26:54", + "end": "00:26:55" + } + } + ], + "question_timestamps": "00:26:58", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can directly reach the red plastic stool right now.", + "B": "You can't directly reach the red plastic stool." + }, + "answer_label": "B", + "answer_text": "You can't directly reach the red plastic stool.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_2_T2", + "correct_answer": "You can't directly reach the red plastic stool.", + "original_options": { + "A": "You can directly reach the red plastic stool right now.", + "B": "You can't directly reach the red plastic stool." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_4", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the red plastic stool directly now?", + "answer": "You can't touch the red plastic stool directly.", + "question_chinese": "我现在能直接触碰到红色塑料凳吗？", + "answer_chinese": "你无法直接触碰到红色塑料凳。", + "evidence": [ + { + "description": "You can't touch the red plastic stool directly.", + "timestamps": { + "start": "00:26:54", + "end": "00:26:55" + } + } + ], + "question_timestamps": "00:27:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't touch the red plastic stool directly.", + "B": "You can touch the red plastic stool directly now." + }, + "answer_label": "A", + "answer_text": "You can't touch the red plastic stool directly.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_4", + "correct_answer": "You can't touch the red plastic stool directly.", + "original_options": { + "A": "You can't touch the red plastic stool directly.", + "B": "You can touch the red plastic stool directly now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_2_T1", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the red plastic stool right now?", + "answer": "You can directly reach the red plastic stool.", + "question_chinese": "我现在能直接触碰到红色塑料凳吗？", + "answer_chinese": "你可以直接触碰到红色塑料凳。", + "evidence": [ + { + "description": "You can't touch the red plastic stool directly.", + "timestamps": { + "start": "00:26:54", + "end": "00:26:55" + } + } + ], + "question_timestamps": "00:27:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can directly reach the red plastic stool.", + "B": "You can't directly reach the red plastic stool right now." + }, + "answer_label": "A", + "answer_text": "You can directly reach the red plastic stool.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_2_T1", + "correct_answer": "You can directly reach the red plastic stool.", + "original_options": { + "A": "You can directly reach the red plastic stool.", + "B": "You can't directly reach the red plastic stool right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_3_T1", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the tool I used to pound the steel balls, relative to me?", + "answer": "The tools are to your left.", + "question_chinese": "我用来捶打钢珠的工具在相对于我的什么位置？", + "answer_chinese": "工具在你的左边。", + "evidence": [ + { + "description": "The tool used to hit steel balls is on your left.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:01:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The tools are in front of you.", + "B": "The tools are behind you.", + "C": "The tools are to your right side, not positioned to your left.", + "D": "The tools are located behind you instead of being placed off to your left.", + "E": "The tools are to your left." + }, + "answer_label": "E", + "answer_text": "The tools are to your left.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_3_T1", + "correct_answer": "The tools are to your left.", + "original_options": { + "A": "The tools are in front of you.", + "B": "The tools are behind you.", + "C": "The tools are to your right side, not positioned to your left.", + "D": "The tools are located behind you instead of being placed off to your left.", + "E": "The tools are to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 12, + 14 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_3_T2", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the tool I used to pound the steel balls, relative to me?", + "answer": "The tools are to your right.", + "question_chinese": "我用来捶打钢珠的工具在相对于我的什么位置？", + "answer_chinese": "工具在你的右边。", + "evidence": [ + { + "description": "The tool used to hit steel balls is on your right.", + "timestamps": { + "start": "00:03:03", + "end": "00:03:38" + } + } + ], + "question_timestamps": "00:32:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The tools are to your right.", + "B": "The tools are to your left.", + "C": "The tools are in front of you.", + "D": "The tools are behind you, not to your right side at all.", + "E": "The tools are positioned to your left rather than on your right side." + }, + "answer_label": "A", + "answer_text": "The tools are to your right.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_3_T2", + "correct_answer": "The tools are to your right.", + "original_options": { + "A": "The tools are to your right.", + "B": "The tools are to your left.", + "C": "The tools are in front of you.", + "D": "The tools are behind you, not to your right side at all.", + "E": "The tools are positioned to your left rather than on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 12, + 13 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_15", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the position of the white stool relative to me now?", + "answer": "The white stool is in front of you.", + "question_chinese": "白色凳子现在相对我是什么位置？", + "answer_chinese": "白色凳子在你的前方。", + "evidence": [ + { + "description": "The white stool is in front of you.", + "timestamps": { + "start": "00:16:28", + "end": "00:16:29" + } + } + ], + "question_timestamps": "00:16:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The white stool is behind you.", + "B": "The white stool is to your left.", + "C": "The white stool is on your right side, not directly in front of you.", + "D": "The white stool is in front of you.", + "E": "The white stool is positioned behind you instead of being in front of you." + }, + "answer_label": "D", + "answer_text": "The white stool is in front of you.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_15", + "correct_answer": "The white stool is in front of you.", + "original_options": { + "A": "The white stool is behind you.", + "B": "The white stool is to your left.", + "C": "The white stool is on your right side, not directly in front of you.", + "D": "The white stool is in front of you.", + "E": "The white stool is positioned behind you instead of being in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 14 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_16", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the position of the white stool relative to me now?", + "answer": "The white stool is now behind you.", + "question_chinese": "白色凳子现在相对我是什么位置？", + "answer_chinese": "白色凳子现在在你的后方。", + "evidence": [ + { + "description": "You got up from the white stool and walked out of the door.", + "timestamps": { + "start": "00:24:56", + "end": "00:25:00" + } + }, + { + "description": "You just sat on the white stool.", + "timestamps": { + "start": "00:19:50", + "end": "00:20:01" + } + } + ], + "question_timestamps": "00:25:04", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The white stool is in front of you.", + "B": "The white stool is to your right.", + "C": "The white stool is now behind you.", + "D": "The white stool is beside you, not positioned behind you right now.", + "E": "The white stool is located in front of you rather than behind your position." + }, + "answer_label": "C", + "answer_text": "The white stool is now behind you.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_16", + "correct_answer": "The white stool is now behind you.", + "original_options": { + "A": "The white stool is in front of you.", + "B": "The white stool is to your right.", + "C": "The white stool is now behind you.", + "D": "The white stool is beside you, not positioned behind you right now.", + "E": "The white stool is located in front of you rather than behind your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 12, + 14 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_7", + "qid_num": 27, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the position of the red plastic stool relative to me now?", + "answer": "The red plastic stool is on your left.", + "question_chinese": "红色塑料凳现在相对我是什么位置？", + "answer_chinese": "红色塑料凳在你的左侧。", + "evidence": [ + { + "description": "The red plastic stool is on your left.", + "timestamps": { + "start": "00:35:45", + "end": "00:35:55" + } + } + ], + "question_timestamps": "00:35:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The red plastic stool is on your right.", + "B": "The red plastic stool is on your left.", + "C": "The red plastic stool is behind you.", + "D": "The red plastic stool is behind you rather than positioned on your left side.", + "E": "The red plastic stool is under the table, not sitting on your left." + }, + "answer_label": "B", + "answer_text": "The red plastic stool is on your left.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_7", + "correct_answer": "The red plastic stool is on your left.", + "original_options": { + "A": "The red plastic stool is on your right.", + "B": "The red plastic stool is on your left.", + "C": "The red plastic stool is behind you.", + "D": "The red plastic stool is behind you rather than positioned on your left side.", + "E": "The red plastic stool is under the table, not sitting on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 13 + ] + }, + { + "video_uid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0", + "video_length": "00:43:17", + "qid": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_8", + "qid_num": 28, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the position of the red plastic stool relative to me now?", + "answer": "The red plastic stool is behind you.", + "question_chinese": "红色塑料凳现在相对我是什么位置？", + "answer_chinese": "红色塑料凳在你的后方。", + "evidence": [ + { + "description": "The red plastic stool is behind you.", + "timestamps": { + "start": "00:36:00", + "end": "00:36:13" + } + } + ], + "question_timestamps": "00:39:59", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The red plastic stool is to your left.", + "B": "The red plastic stool is to your right.", + "C": "The red plastic stool is behind you.", + "D": "The red plastic stool is beside you, not placed behind you at the moment.", + "E": "The red plastic stool is positioned in front of you rather than behind your current position." + }, + "answer_label": "C", + "answer_text": "The red plastic stool is behind you.", + "q_id": "0e66de5b-08e2-4cbb-8f2d-0028168218b0_8", + "correct_answer": "The red plastic stool is behind you.", + "original_options": { + "A": "The red plastic stool is to your left.", + "B": "The red plastic stool is to your right.", + "C": "The red plastic stool is behind you.", + "D": "The red plastic stool is beside you, not placed behind you at the moment.", + "E": "The red plastic stool is positioned in front of you rather than behind your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 14, + 16 + ] + } +] diff --git a/QAs/0ed86ac8-75ac-48f4-985e-ac5b991d6b37-mcq.json b/QAs/0ed86ac8-75ac-48f4-985e-ac5b991d6b37-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..026f790600747a43062bff45d2fce1faacb8ad2b --- /dev/null +++ b/QAs/0ed86ac8-75ac-48f4-985e-ac5b991d6b37-mcq.json @@ -0,0 +1,1339 @@ +[ + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_6_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I put the phone on the table, can I reach the phone from where I am now?", + "answer": "If you put the phone on the table, you can’t reach the phone right now.", + "question_chinese": "如果我把手机放在桌子上，那我在现在能够到手机吗？", + "answer_chinese": "如果你把手机放在桌子上，你现在不能够到手机。", + "evidence": [ + { + "description": "The sofa is about three meters away from the TV cabinet.", + "timestamps": { + "start": "00:32:30", + "end": "00:32:40" + } + } + ], + "question_timestamps": "00:32:44", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "If you put the phone on the table, you can reach the phone right now.", + "B": "If you put the phone on the table, you can’t reach the phone right now." + }, + "answer_label": "B", + "answer_text": "If you put the phone on the table, you can’t reach the phone right now.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_6_T1", + "correct_answer": "If you put the phone on the table, you can’t reach the phone right now.", + "original_options": { + "A": "If you put the phone on the table, you can reach the phone right now.", + "B": "If you put the phone on the table, you can’t reach the phone right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_6_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I put the phone on the table, can I reach the phone from where I am now?", + "answer": "If you put the phone on the table, you can reach the phone right now.", + "question_chinese": "如果我把手机放在桌子上，那我在现在能够到手机吗？", + "answer_chinese": "如果你把手机放在桌子上，你现在够到手机。", + "evidence": [ + { + "description": "The sofa is three meters away from the TV cabinet.", + "timestamps": { + "start": "00:32:30", + "end": "00:32:40" + } + }, + { + "description": "You are now beside the table", + "timestamps": { + "start": "00:50:38", + "end": "00:50:41" + } + } + ], + "question_timestamps": "01:10:53", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "If you put the phone on the table, you can reach the phone right now.", + "B": "If you put the phone on the table, you cannot reach the phone right now." + }, + "answer_label": "A", + "answer_text": "If you put the phone on the table, you can reach the phone right now.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_6_T2", + "correct_answer": "If you put the phone on the table, you can reach the phone right now.", + "original_options": { + "A": "If you put the phone on the table, you can reach the phone right now.", + "B": "If you put the phone on the table, you cannot reach the phone right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_11", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I put my phone on the TV stand, can I reach it from the sofa?", + "answer": "If you put your phone on the TV stand, you won't be able to reach it from the sofa.", + "question_chinese": "如果我把手机放在电视柜上，那我在沙发上能够到手机吗？", + "answer_chinese": "如果你把手机放在电视柜上，你在沙发上不能够到手机。", + "evidence": [ + { + "description": "The sofa is about three meters away from the TV cabinet.", + "timestamps": { + "start": "00:32:30", + "end": "00:32:40" + } + } + ], + "question_timestamps": "00:32:44", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "If you put your phone on the TV stand, you'll be able to reach it from the sofa.", + "B": "If you put your phone on the TV stand, you won't be able to reach it from the sofa." + }, + "answer_label": "B", + "answer_text": "If you put your phone on the TV stand, you won't be able to reach it from the sofa.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_11", + "correct_answer": "If you put your phone on the TV stand, you won't be able to reach it from the sofa.", + "original_options": { + "A": "If you put your phone on the TV stand, you'll be able to reach it from the sofa.", + "B": "If you put your phone on the TV stand, you won't be able to reach it from the sofa." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_12", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I put my phone on the TV stand, can I reach it from the sofa?", + "answer": "If you put your phone on the TV stand, you won't be able to reach it from the sofa.", + "question_chinese": "如果我把手机放在电视柜上，那我在沙发上能够到手机吗？", + "answer_chinese": "如果你把手机放在电视柜上，你在沙发上不能够到手机。", + "evidence": [ + { + "description": "The sofa is three meters away from the TV cabinet.", + "timestamps": { + "start": "00:32:30", + "end": "00:32:40" + } + } + ], + "question_timestamps": "01:10:53", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "If you put your phone on the TV stand, you won't be able to reach it from the sofa.", + "B": "If you put your phone on the TV stand, you'll be able to reach it from the sofa." + }, + "answer_label": "A", + "answer_text": "If you put your phone on the TV stand, you won't be able to reach it from the sofa.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_12", + "correct_answer": "If you put your phone on the TV stand, you won't be able to reach it from the sofa.", + "original_options": { + "A": "If you put your phone on the TV stand, you won't be able to reach it from the sofa.", + "B": "If you put your phone on the TV stand, you'll be able to reach it from the sofa." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_9_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was doing push-ups, how many chairs were in the scene?", + "answer": "When you were doing push-ups, there were 4 chairs in the scene.", + "question_chinese": "当我在做俯卧撑时，场景里有多少张椅子？", + "answer_chinese": "当你在做俯卧撑时，场景里有4张椅子。", + "evidence": [ + { + "description": "There are four chairs in front of me.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:03:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There were 2 chairs.", + "B": "There were 6 chairs.", + "C": "When you were doing push-ups, there were 4 chairs in the scene.", + "D": "When you were doing push-ups, there were 3 chairs in the scene.", + "E": "When you were doing push-ups, there were 5 chairs in the scene." + }, + "answer_label": "C", + "answer_text": "When you were doing push-ups, there were 4 chairs in the scene.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_9_T1", + "correct_answer": "When you were doing push-ups, there were 4 chairs in the scene.", + "original_options": { + "A": "There were 2 chairs.", + "B": "There were 6 chairs.", + "C": "When you were doing push-ups, there were 4 chairs in the scene.", + "D": "When you were doing push-ups, there were 3 chairs in the scene.", + "E": "When you were doing push-ups, there were 5 chairs in the scene." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_9_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was doing push-ups, how many chairs were in the scene?", + "answer": "When you were doing push-ups, there were 5 chairs in the scene.", + "question_chinese": "当我在做俯卧撑时，场景里有多少张椅子？", + "answer_chinese": "当你在做俯卧撑时，场景里有5张椅子。", + "evidence": [ + { + "description": "There are five chairs in the living room.", + "timestamps": { + "start": "00:03:29", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:40:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There were 3 chairs.", + "B": "There were 7 chairs.", + "C": "When you were doing push-ups, there were 4 chairs in the scene.", + "D": "When you were doing push-ups, there were 5 chairs in the scene.", + "E": "When you were doing push-ups, there were 6 chairs in the scene." + }, + "answer_label": "D", + "answer_text": "When you were doing push-ups, there were 5 chairs in the scene.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_9_T2", + "correct_answer": "When you were doing push-ups, there were 5 chairs in the scene.", + "original_options": { + "A": "There were 3 chairs.", + "B": "There were 7 chairs.", + "C": "When you were doing push-ups, there were 4 chairs in the scene.", + "D": "When you were doing push-ups, there were 5 chairs in the scene.", + "E": "When you were doing push-ups, there were 6 chairs in the scene." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_15", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the water?", + "answer": "You just put the water in the kettle.", + "question_chinese": "我刚刚把水放在了哪里？", + "answer_chinese": "你刚刚把水放在了水壶里。", + "evidence": [ + { + "description": "I filled the water bottle with water.", + "timestamps": { + "start": "00:22:28", + "end": "00:22:32" + } + } + ], + "question_timestamps": "00:22:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put the water in the sink.", + "B": "You put the water in a glass.", + "C": "You just put the water down on the counter beside where you were working.", + "D": "You just put the water in the kettle.", + "E": "You just poured the water into a cup and left it sitting on the counter." + }, + "answer_label": "D", + "answer_text": "You just put the water in the kettle.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_15", + "correct_answer": "You just put the water in the kettle.", + "original_options": { + "A": "You put the water in the sink.", + "B": "You put the water in a glass.", + "C": "You just put the water down on the counter beside where you were working.", + "D": "You just put the water in the kettle.", + "E": "You just poured the water into a cup and left it sitting on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 15 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_16", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the water?", + "answer": "You just put the water in the pot.", + "question_chinese": "我刚刚把水放在了哪里？", + "answer_chinese": "你刚刚把水放在了锅里。", + "evidence": [ + { + "description": "After I finished filling the water, I poured it into the pot.", + "timestamps": { + "start": "00:22:50", + "end": "00:23:18" + } + } + ], + "question_timestamps": "00:23:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put the water in the sink.", + "B": "You put the water in a pitcher.", + "C": "You just put the water into the measuring cup instead of pouring it into anything to cook.", + "D": "You just put the water in the pot.", + "E": "You just poured the water into a bowl and left it there before moving on." + }, + "answer_label": "D", + "answer_text": "You just put the water in the pot.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_16", + "correct_answer": "You just put the water in the pot.", + "original_options": { + "A": "You put the water in the sink.", + "B": "You put the water in a pitcher.", + "C": "You just put the water into the measuring cup instead of pouring it into anything to cook.", + "D": "You just put the water in the pot.", + "E": "You just poured the water into a bowl and left it there before moving on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 15 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_4_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the meat?", + "answer": "You just put the meat into the basin.", + "question_chinese": "我刚刚把肉放在了哪里？", + "answer_chinese": "你刚刚把肉放进了盆子里。", + "evidence": [ + { + "description": "After I washed the meat, I put it into the basin.", + "timestamps": { + "start": "00:16:45", + "end": "00:16:51" + } + } + ], + "question_timestamps": "00:19:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put the meat into the pot.", + "B": "You put the meat on the counter.", + "C": "You just put the meat into a bowl and left it there before doing anything else.", + "D": "You just put the meat into the basin.", + "E": "You just put the meat into the sink to rinse it before continuing with preparation." + }, + "answer_label": "D", + "answer_text": "You just put the meat into the basin.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_4_T1", + "correct_answer": "You just put the meat into the basin.", + "original_options": { + "A": "You put the meat into the pot.", + "B": "You put the meat on the counter.", + "C": "You just put the meat into a bowl and left it there before doing anything else.", + "D": "You just put the meat into the basin.", + "E": "You just put the meat into the sink to rinse it before continuing with preparation." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 15 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_7", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the meat?", + "answer": "You just put the meat into the basin.", + "question_chinese": "我刚刚把肉放在了哪里？", + "answer_chinese": "你刚刚把肉放进了盆子里。", + "evidence": [ + { + "description": "After I washed the meat, I put it into the basin.", + "timestamps": { + "start": "00:16:45", + "end": "00:16:51" + } + } + ], + "question_timestamps": "00:24:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just put the meat in the fridge.", + "B": "You just put the meat on the countertop.", + "C": "You just put the meat into the basin.", + "D": "You just put the meat inside the drawer, not in any container on the counter.", + "E": "You just set the meat on the stove area after moving it away from the sink." + }, + "answer_label": "C", + "answer_text": "You just put the meat into the basin.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_7", + "correct_answer": "You just put the meat into the basin.", + "original_options": { + "A": "You just put the meat in the fridge.", + "B": "You just put the meat on the countertop.", + "C": "You just put the meat into the basin.", + "D": "You just put the meat inside the drawer, not in any container on the counter.", + "E": "You just set the meat on the stove area after moving it away from the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 16 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_4_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the meat?", + "answer": "You just put the meat into the pot.", + "question_chinese": "我刚刚把肉放在了哪里？", + "answer_chinese": "你刚刚把肉放进了锅里。", + "evidence": [ + { + "description": "After I washed the meat, I put it into the basin.", + "timestamps": { + "start": "00:16:45", + "end": "00:16:51" + } + }, + { + "description": "You just put the meat into the pot.", + "timestamps": { + "start": "00:19:23", + "end": "00:19:36" + } + } + ], + "question_timestamps": "00:24:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You just put the meat into the basin.", + "B": "You just put the meat on the countertop.", + "C": "You just put the meat into the fridge to keep it cold before cooking.", + "D": "You just put the meat into the pot.", + "E": "You just placed the meat into a bowl on the counter instead of putting it in the pot." + }, + "answer_label": "D", + "answer_text": "You just put the meat into the pot.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_4_T2", + "correct_answer": "You just put the meat into the pot.", + "original_options": { + "A": "You just put the meat into the basin.", + "B": "You just put the meat on the countertop.", + "C": "You just put the meat into the fridge to keep it cold before cooking.", + "D": "You just put the meat into the pot.", + "E": "You just placed the meat into a bowl on the counter instead of putting it in the pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 14, + 18 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_3", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where can I find more onions?", + "answer": "You can find a bag of onions in the cabinet.", + "question_chinese": "我想要更多的洋葱可以去哪里找？", + "answer_chinese": "你可以在柜子里找到一袋洋葱。", + "evidence": [ + { + "description": "There is a bag of onions in the cabinet.", + "timestamps": { + "start": "00:10:18", + "end": "00:10:21" + } + } + ], + "question_timestamps": "00:10:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the drawer.", + "B": "On the kitchen counter.", + "C": "You can find more onions on a pantry shelf rather than inside the cabinet.", + "D": "You can find more onions in the fruit basket on the counter, not stored in a cabinet.", + "E": "You can find a bag of onions in the cabinet." + }, + "answer_label": "E", + "answer_text": "You can find a bag of onions in the cabinet.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_3", + "correct_answer": "You can find a bag of onions in the cabinet.", + "original_options": { + "A": "In the drawer.", + "B": "On the kitchen counter.", + "C": "You can find more onions on a pantry shelf rather than inside the cabinet.", + "D": "You can find more onions in the fruit basket on the counter, not stored in a cabinet.", + "E": "You can find a bag of onions in the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 17 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_4", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where can I find more onions?", + "answer": "You can find a bag of onions in the cabinet.", + "question_chinese": "我想要更多的洋葱可以去哪里找？", + "answer_chinese": "你可以在柜子里找到一袋洋葱。", + "evidence": [ + { + "description": "There is a bag of onions in the cupboard.", + "timestamps": { + "start": "00:10:17", + "end": "00:10:21" + } + } + ], + "question_timestamps": "00:25:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You can find a bag of onions in the cabinet.", + "B": "On the kitchen counter.", + "C": "In the fruit basket.", + "D": "You can find a bag of onions inside the refrigerator drawer rather than in the cabinet.", + "E": "You can find a bag of onions on the pantry shelf instead of inside the cabinet." + }, + "answer_label": "A", + "answer_text": "You can find a bag of onions in the cabinet.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_4", + "correct_answer": "You can find a bag of onions in the cabinet.", + "original_options": { + "A": "You can find a bag of onions in the cabinet.", + "B": "On the kitchen counter.", + "C": "In the fruit basket.", + "D": "You can find a bag of onions inside the refrigerator drawer rather than in the cabinet.", + "E": "You can find a bag of onions on the pantry shelf instead of inside the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_9", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I stand in front of the mirror, which is closer to me, the sofa or the table？", + "answer": "You stand in front of the mirror, and the table is closer to you.", + "question_chinese": "我站在镜子前，沙发和桌子哪一个更靠近我？", + "answer_chinese": "你站在镜子前，桌子更靠近你。", + "evidence": [ + { + "description": "The table is right next to the mirror.", + "timestamps": { + "start": "00:18:50", + "end": "00:18:56" + } + } + ], + "question_timestamps": "00:19:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You stand in front of the mirror, and the table is closer to you.", + "B": "You stand in front of the mirror, and the sofa is closer to you." + }, + "answer_label": "A", + "answer_text": "You stand in front of the mirror, and the table is closer to you.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_9", + "correct_answer": "You stand in front of the mirror, and the table is closer to you.", + "original_options": { + "A": "You stand in front of the mirror, and the table is closer to you.", + "B": "You stand in front of the mirror, and the sofa is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_10", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I stand in front of the mirror, which is closer to me, the sofa or the table？", + "answer": "You stand in front of the mirror, and the table is closer to you.", + "question_chinese": "我站在镜子前，沙发和桌子哪一个更靠近我？", + "answer_chinese": "你站在镜子前，桌子更靠近你。", + "evidence": [ + { + "description": "The mirror is next to the table.", + "timestamps": { + "start": "00:18:50", + "end": "00:18:56" + } + } + ], + "question_timestamps": "00:44:36", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You stand in front of the mirror, and the sofa is closer to you.", + "B": "You stand in front of the mirror, and the table is closer to you." + }, + "answer_label": "B", + "answer_text": "You stand in front of the mirror, and the table is closer to you.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_10", + "correct_answer": "You stand in front of the mirror, and the table is closer to you.", + "original_options": { + "A": "You stand in front of the mirror, and the sofa is closer to you.", + "B": "You stand in front of the mirror, and the table is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_5_T1", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is nearer to me, the sofa or the table?", + "answer": "The table is closer to you.", + "question_chinese": "沙发和桌子哪一个更靠近我？", + "answer_chinese": "桌子更靠近你。", + "evidence": [ + { + "description": "The table is right next to the mirror.", + "timestamps": { + "start": "00:18:50", + "end": "00:18:56" + } + } + ], + "question_timestamps": "00:19:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The table is closer to you.", + "B": "The sofa is closer to you." + }, + "answer_label": "A", + "answer_text": "The table is closer to you.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_5_T1", + "correct_answer": "The table is closer to you.", + "original_options": { + "A": "The table is closer to you.", + "B": "The sofa is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_5_T2", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is nearer to me, the sofa or the table?", + "answer": "The sofa is closer to you.", + "question_chinese": "沙发和桌子哪一个更靠近我？", + "answer_chinese": "沙发更靠近你。", + "evidence": [ + { + "description": "The sofa is closer to you", + "timestamps": { + "start": "00:26:26", + "end": "00:28:31" + } + } + ], + "question_timestamps": "00:44:36", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sofa is closer to you.", + "B": "The table is closer to you." + }, + "answer_label": "A", + "answer_text": "The sofa is closer to you.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_5_T2", + "correct_answer": "The sofa is closer to you.", + "original_options": { + "A": "The sofa is closer to you.", + "B": "The table is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_1_T1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many garlic cloves have I chopped so far?", + "answer": "So far, you have chopped four cloves of garlic.", + "question_chinese": "现在我切了几颗蒜？", + "answer_chinese": "现在你切了四颗蒜。", + "evidence": [ + { + "description": "I took out four cloves of garlic.", + "timestamps": { + "start": "00:15:15", + "end": "00:15:19" + } + } + ], + "question_timestamps": "00:15:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You chopped three cloves.", + "B": "You chopped five cloves.", + "C": "So far, you have chopped two cloves of garlic, not four, based on what you've prepared.", + "D": "Up to now, you have chopped six cloves of garlic, which is more than the four you think.", + "E": "So far, you have chopped four cloves of garlic." + }, + "answer_label": "E", + "answer_text": "So far, you have chopped four cloves of garlic.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_1_T1", + "correct_answer": "So far, you have chopped four cloves of garlic.", + "original_options": { + "A": "You chopped three cloves.", + "B": "You chopped five cloves.", + "C": "So far, you have chopped two cloves of garlic, not four, based on what you've prepared.", + "D": "Up to now, you have chopped six cloves of garlic, which is more than the four you think.", + "E": "So far, you have chopped four cloves of garlic." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 18 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_1_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many garlic cloves have I chopped so far?", + "answer": "So far, you have chopped five cloves of garlic.", + "question_chinese": "现在我切了几颗蒜？", + "answer_chinese": "现在你切了五颗蒜。", + "evidence": [ + { + "description": "After cutting four cloves of garlic, I took another clove of garlic.", + "timestamps": { + "start": "00:16:00", + "end": "00:16:09" + } + } + ], + "question_timestamps": "00:59:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, you have chopped five cloves of garlic.", + "B": "You chopped four cloves.", + "C": "You chopped six cloves.", + "D": "So far, you have chopped three cloves of garlic, not five, based on your progress.", + "E": "Up to now, you have chopped seven cloves of garlic, which is more than five." + }, + "answer_label": "A", + "answer_text": "So far, you have chopped five cloves of garlic.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_1_T2", + "correct_answer": "So far, you have chopped five cloves of garlic.", + "original_options": { + "A": "So far, you have chopped five cloves of garlic.", + "B": "You chopped four cloves.", + "C": "You chopped six cloves.", + "D": "So far, you have chopped three cloves of garlic, not five, based on your progress.", + "E": "Up to now, you have chopped seven cloves of garlic, which is more than five." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_19", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Now I want to go up the stairs, how should I go?", + "answer": "You should walk to your front left.", + "question_chinese": "现在我想上楼梯应该怎么走？", + "answer_chinese": "你应该往你的左前方走。", + "evidence": [ + { + "description": "The stairs are in front of me to the left.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk to your front right.", + "B": "You should walk to your front left.", + "C": "Walk straight ahead.", + "D": "You should move toward your back left to reach the stairs from where you're standing.", + "E": "Head to your back right first, then adjust your direction once you see the stair entrance." + }, + "answer_label": "B", + "answer_text": "You should walk to your front left.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_19", + "correct_answer": "You should walk to your front left.", + "original_options": { + "A": "Walk to your front right.", + "B": "You should walk to your front left.", + "C": "Walk straight ahead.", + "D": "You should move toward your back left to reach the stairs from where you're standing.", + "E": "Head to your back right first, then adjust your direction once you see the stair entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 15, + 16 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_20", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Now I want to go up the stairs, how should I go?", + "answer": "The passage on the left leads out, then turn left.", + "question_chinese": "现在我想上楼梯应该怎么走？", + "answer_chinese": "左边的通道走出去，然后左拐。", + "evidence": [ + { + "description": "The stairs are on the left side of the road going out.", + "timestamps": { + "start": "00:13:17", + "end": "00:13:21" + } + } + ], + "question_timestamps": "00:13:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Take the right passage, then turn right.", + "B": "The left passage leads out, then turn right.", + "C": "Go through the right passage to get out first, then make a left turn afterward.", + "D": "Use the left passage that leads up, and then turn right to reach the stairway.", + "E": "The passage on the left leads out, then turn left." + }, + "answer_label": "E", + "answer_text": "The passage on the left leads out, then turn left.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_20", + "correct_answer": "The passage on the left leads out, then turn left.", + "original_options": { + "A": "Take the right passage, then turn right.", + "B": "The left passage leads out, then turn right.", + "C": "Go through the right passage to get out first, then make a left turn afterward.", + "D": "Use the left passage that leads up, and then turn right to reach the stairway.", + "E": "The passage on the left leads out, then turn left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 15 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_13", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many trash cans have I encountered now?", + "answer": "Now you have encountered two trash cans.", + "question_chinese": "现在我遇到了几个垃圾桶？", + "answer_chinese": "现在你遇到了两个垃圾桶。", + "evidence": [ + { + "description": "I passed by two trash cans.", + "timestamps": { + "start": "00:09:36", + "end": "00:09:46" + } + } + ], + "question_timestamps": "00:09:49", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You encountered zero trash cans.", + "B": "You encountered three trash cans.", + "C": "Now you have encountered two trash cans.", + "D": "Now you have encountered four trash cans, which is more than the two you recall seeing.", + "E": "At this point you have encountered six trash cans along the way, not just two." + }, + "answer_label": "C", + "answer_text": "Now you have encountered two trash cans.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_13", + "correct_answer": "Now you have encountered two trash cans.", + "original_options": { + "A": "You encountered zero trash cans.", + "B": "You encountered three trash cans.", + "C": "Now you have encountered two trash cans.", + "D": "Now you have encountered four trash cans, which is more than the two you recall seeing.", + "E": "At this point you have encountered six trash cans along the way, not just two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_14", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many trash cans have I encountered now?", + "answer": "Now you have encountered three trash cans.", + "question_chinese": "现在我遇到了几个垃圾桶？", + "answer_chinese": "现在你遇到了三个垃圾桶。", + "evidence": [ + { + "description": "There are two large trash bins next to the kitchen counter, and there is a food waste bin next to the stove.", + "timestamps": { + "start": "00:22:57", + "end": "00:23:07" + } + } + ], + "question_timestamps": "00:23:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You encountered zero trash cans.", + "B": "You encountered four trash cans.", + "C": "Now you have encountered six trash cans, which is more than the three you remember.", + "D": "At this point you have encountered eight trash cans along your route, not three.", + "E": "Now you have encountered three trash cans." + }, + "answer_label": "E", + "answer_text": "Now you have encountered three trash cans.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_14", + "correct_answer": "Now you have encountered three trash cans.", + "original_options": { + "A": "You encountered zero trash cans.", + "B": "You encountered four trash cans.", + "C": "Now you have encountered six trash cans, which is more than the three you remember.", + "D": "At this point you have encountered eight trash cans along your route, not three.", + "E": "Now you have encountered three trash cans." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_2_T1", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the TV on my left or my right right now?", + "answer": "It’s to your right.", + "question_chinese": "现在电视在我的左/右？", + "answer_chinese": "在你的右边。", + "evidence": [ + { + "description": "I stood up and turned left.", + "timestamps": { + "start": "00:02:37", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:02:44", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It’s to your right.", + "B": "It’s to your left." + }, + "answer_label": "A", + "answer_text": "It’s to your right.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_2_T1", + "correct_answer": "It’s to your right.", + "original_options": { + "A": "It’s to your right.", + "B": "It’s to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_2_T2", + "qid_num": 27, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the TV on my left or my right right now?", + "answer": "It’s to your left.", + "question_chinese": "现在电视在我的左/右？", + "answer_chinese": "在你的左边。", + "evidence": [ + { + "description": "The television is now on my left.", + "timestamps": { + "start": "00:05:42", + "end": "00:05:45" + } + } + ], + "question_timestamps": "00:56:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It’s to your right.", + "B": "It’s to your left." + }, + "answer_label": "B", + "answer_text": "It’s to your left.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_2_T2", + "correct_answer": "It’s to your left.", + "original_options": { + "A": "It’s to your right.", + "B": "It’s to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_2", + "qid_num": 28, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Facing the desk, is the TV now on my left/right?", + "answer": "On your right.", + "question_chinese": "面对桌子，现在电视在我的左/右？", + "answer_chinese": "在你的右边。", + "evidence": [ + { + "description": "I stood up and turned left.", + "timestamps": { + "start": "00:02:37", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:02:44", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "On your left.", + "B": "On your right." + }, + "answer_label": "B", + "answer_text": "On your right.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_2", + "correct_answer": "On your right.", + "original_options": { + "A": "On your left.", + "B": "On your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37", + "video_length": "01:10:53", + "qid": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_5", + "qid_num": 29, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Facing the desk, is the TV now on my left/right?", + "answer": "On your right.", + "question_chinese": "面对桌子，现在电视在我的左/右？", + "answer_chinese": "在你的右边。", + "evidence": [ + { + "description": "The television is on the right side of the table.", + "timestamps": { + "start": "00:02:35", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:56:18", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "On your right.", + "B": "On your left." + }, + "answer_label": "A", + "answer_text": "On your right.", + "q_id": "0ed86ac8-75ac-48f4-985e-ac5b991d6b37_5", + "correct_answer": "On your right.", + "original_options": { + "A": "On your right.", + "B": "On your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + } +] diff --git a/QAs/0fedfa42-bfda-47c1-8f59-2c91b2c229fe-mcq.json b/QAs/0fedfa42-bfda-47c1-8f59-2c91b2c229fe-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..833860bb1464921fdcdebe325635538c852dea32 --- /dev/null +++ b/QAs/0fedfa42-bfda-47c1-8f59-2c91b2c229fe-mcq.json @@ -0,0 +1,563 @@ +[ + { + "video_uid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe", + "video_length": "01:02:13", + "qid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_10", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times have I used the faucet so far?", + "answer": "Four times.", + "question_chinese": "到目前为止，我用了多少次水龙头？", + "answer_chinese": "四次。", + "evidence": [ + { + "description": "Twice", + "timestamps": { + "start": "00:02:49", + "end": "00:03:04" + } + }, + { + "description": "Add one more time", + "timestamps": { + "start": "00:13:48", + "end": "00:14:01" + } + }, + { + "description": "Add one more time", + "timestamps": { + "start": "00:15:00", + "end": "00:15:25" + } + } + ], + "question_timestamps": "00:15:30", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four times.", + "B": "Two times.", + "C": "Six times.", + "D": "You’ve used the faucet three times so far, turning it on and off on three separate occasions.", + "E": "You have used the faucet five times so far, counting each separate time you turned it on." + }, + "answer_label": "A", + "answer_text": "Four times.", + "q_id": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_10", + "correct_answer": "Four times.", + "original_options": { + "A": "Four times.", + "B": "Two times.", + "C": "Six times.", + "D": "You’ve used the faucet three times so far, turning it on and off on three separate occasions.", + "E": "You have used the faucet five times so far, counting each separate time you turned it on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 17 + ] + }, + { + "video_uid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe", + "video_length": "01:02:13", + "qid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_9", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times have I used the faucet so far?", + "answer": "Five times.", + "question_chinese": "到目前为止，我用了多少次水龙头？", + "answer_chinese": "五次。", + "evidence": [ + { + "description": "Add one more time", + "timestamps": { + "start": "00:23:36", + "end": "00:23:49" + } + }, + { + "description": "Used twice", + "timestamps": { + "start": "00:02:50", + "end": "00:03:10" + } + }, + { + "description": "Add one more time", + "timestamps": { + "start": "00:15:13", + "end": "00:15:29" + } + }, + { + "description": "Add one more time", + "timestamps": { + "start": "00:13:30", + "end": "00:13:50" + } + } + ], + "question_timestamps": "00:31:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Six times.", + "C": "You have used the faucet once so far, with only a single on-and-off use.", + "D": "Five times.", + "E": "You’ve used the faucet zero times so far; you haven’t turned it on at all yet." + }, + "answer_label": "D", + "answer_text": "Five times.", + "q_id": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_9", + "correct_answer": "Five times.", + "original_options": { + "A": "Two times.", + "B": "Six times.", + "C": "You have used the faucet once so far, with only a single on-and-off use.", + "D": "Five times.", + "E": "You’ve used the faucet zero times so far; you haven’t turned it on at all yet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 18 + ] + }, + { + "video_uid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe", + "video_length": "01:02:13", + "qid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_15", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many seasonings have I used so far?", + "answer": "One kind.", + "question_chinese": "到目前为止，我用了多少种调料？", + "answer_chinese": "一种。", + "evidence": [ + { + "description": "An additional one has been added.", + "timestamps": { + "start": "00:10:27", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:16:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "No types.", + "B": "One kind.", + "C": "Two kinds.", + "D": "You’ve used three different seasonings so far, adding three separate kinds during cooking.", + "E": "You’ve used four seasonings so far, counting four distinct kinds added to the food." + }, + "answer_label": "B", + "answer_text": "One kind.", + "q_id": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_15", + "correct_answer": "One kind.", + "original_options": { + "A": "No types.", + "B": "One kind.", + "C": "Two kinds.", + "D": "You’ve used three different seasonings so far, adding three separate kinds during cooking.", + "E": "You’ve used four seasonings so far, counting four distinct kinds added to the food." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe", + "video_length": "01:02:13", + "qid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_16", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many seasonings have I used so far?", + "answer": "Two seasonings.", + "question_chinese": "到目前为止，我用了多少种调料？", + "answer_chinese": "两种调料。", + "evidence": [ + { + "description": "An additional seasoning has been added.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:40" + } + }, + { + "description": "Repeated a seasoning.", + "timestamps": { + "start": "00:18:21", + "end": "00:18:40" + } + }, + { + "description": "An additional seasoning has been added.", + "timestamps": { + "start": "00:20:40", + "end": "00:21:00" + } + } + ], + "question_timestamps": "00:31:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "No seasonings.", + "B": "Four seasonings.", + "C": "You’ve used five seasonings so far, meaning five different kinds have been added.", + "D": "Two seasonings.", + "E": "You’ve used ten seasonings so far, counting ten distinct seasoning types used in total." + }, + "answer_label": "D", + "answer_text": "Two seasonings.", + "q_id": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_16", + "correct_answer": "Two seasonings.", + "original_options": { + "A": "No seasonings.", + "B": "Four seasonings.", + "C": "You’ve used five seasonings so far, meaning five different kinds have been added.", + "D": "Two seasonings.", + "E": "You’ve used ten seasonings so far, counting ten distinct seasoning types used in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe", + "video_length": "01:02:13", + "qid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I am going to get apples. How should I go?", + "answer": "Turn around and walk to the end of the room, then turn left and walk straight out through the kitchen door until you reach the end, then turn left again and walk to the wooden table to get the apple on the table.", + "question_chinese": "我将要去取苹果，我将要怎么走？", + "answer_chinese": "向后转走到房间尽头，然后左转直走出厨房门，直到走到尽头，然后左转走到木桌子旁即可取到桌子上的苹果。", + "evidence": [ + { + "description": "The path from the kitchen to the place where the apples are kept.", + "timestamps": { + "start": "00:24:37", + "end": "00:25:00" + } + } + ], + "question_timestamps": "00:31:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right through the kitchen door.", + "B": "Go straight to the dining table.", + "C": "Turn around and walk to the end of the room, then turn left and walk straight out through the kitchen door until you reach the end, then turn left again and walk to the wooden table to get the apple on the table.", + "D": "Turn around, walk to the end, turn right through the kitchen door, then turn left to the wooden table.", + "E": "Turn around, reach the end of the room, turn left out the kitchen door, then turn right to the table." + }, + "answer_label": "C", + "answer_text": "Turn around and walk to the end of the room, then turn left and walk straight out through the kitchen door until you reach the end, then turn left again and walk to the wooden table to get the apple on the table.", + "q_id": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_3", + "correct_answer": "Turn around and walk to the end of the room, then turn left and walk straight out through the kitchen door until you reach the end, then turn left again and walk to the wooden table to get the apple on the table.", + "original_options": { + "A": "Turn right through the kitchen door.", + "B": "Go straight to the dining table.", + "C": "Turn around and walk to the end of the room, then turn left and walk straight out through the kitchen door until you reach the end, then turn left again and walk to the wooden table to get the apple on the table.", + "D": "Turn around, walk to the end, turn right through the kitchen door, then turn left to the wooden table.", + "E": "Turn around, reach the end of the room, turn left out the kitchen door, then turn right to the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 20 + ] + }, + { + "video_uid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe", + "video_length": "01:02:13", + "qid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_4", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I am going to get apples. How should I go?", + "answer": "Turn around and walk to the end of the room, then turn left and walk straight out through the kitchen door until you reach the end, then turn left again and walk to the wooden table to get the apple on the table.", + "question_chinese": "我将要去取苹果，我将要怎么走？", + "answer_chinese": "向后转走到房间尽头，然后左转直走出厨房门，直到走到尽头，然后左转走到木桌子旁即可取到桌子上的苹果。", + "evidence": [ + { + "description": "The path from the kitchen to the apple storage area.", + "timestamps": { + "start": "00:24:30", + "end": "00:24:53" + } + } + ], + "question_timestamps": "01:02:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn around and walk to the end of the room, then turn left and walk straight out through the kitchen door until you reach the end, then turn left again and walk to the wooden table to get the apple on the table.", + "B": "Turn right at the end.", + "C": "Head straight to the counter.", + "D": "Turn around, walk to the end, turn left through the kitchen door, then turn right to the wooden table.", + "E": "Turn around and go to the end, turn right through the kitchen door, then turn left and stop at the table." + }, + "answer_label": "A", + "answer_text": "Turn around and walk to the end of the room, then turn left and walk straight out through the kitchen door until you reach the end, then turn left again and walk to the wooden table to get the apple on the table.", + "q_id": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_4", + "correct_answer": "Turn around and walk to the end of the room, then turn left and walk straight out through the kitchen door until you reach the end, then turn left again and walk to the wooden table to get the apple on the table.", + "original_options": { + "A": "Turn around and walk to the end of the room, then turn left and walk straight out through the kitchen door until you reach the end, then turn left again and walk to the wooden table to get the apple on the table.", + "B": "Turn right at the end.", + "C": "Head straight to the counter.", + "D": "Turn around, walk to the end, turn left through the kitchen door, then turn right to the wooden table.", + "E": "Turn around and go to the end, turn right through the kitchen door, then turn left and stop at the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 21 + ] + }, + { + "video_uid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe", + "video_length": "01:02:13", + "qid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the pot now?", + "answer": "You can touch the pot.", + "question_chinese": "我现在触摸得到锅吗？", + "answer_chinese": "你可以触摸到锅。", + "evidence": [ + { + "description": "You can touch the pot.", + "timestamps": { + "start": "00:30:24", + "end": "00:30:50" + } + } + ], + "question_timestamps": "00:31:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch the pot.", + "B": "You cannot touch the pot." + }, + "answer_label": "A", + "answer_text": "You can touch the pot.", + "q_id": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_1", + "correct_answer": "You can touch the pot.", + "original_options": { + "A": "You can touch the pot.", + "B": "You cannot touch the pot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe", + "video_length": "01:02:13", + "qid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the pot now?", + "answer": "You cannot touch the pot.", + "question_chinese": "我现在触摸得到锅吗？", + "answer_chinese": "你无法触摸到锅。", + "evidence": [ + { + "description": "There is no pot here.", + "timestamps": { + "start": "00:44:20", + "end": "00:44:50" + } + } + ], + "question_timestamps": "00:46:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot touch the pot.", + "B": "You can touch the pot." + }, + "answer_label": "A", + "answer_text": "You cannot touch the pot.", + "q_id": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_2", + "correct_answer": "You cannot touch the pot.", + "original_options": { + "A": "You cannot touch the pot.", + "B": "You can touch the pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe", + "video_length": "01:02:13", + "qid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_5", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the faucet in?", + "answer": "The faucet is to your left rear.", + "question_chinese": "水龙头在我的什么方向？", + "answer_chinese": "水龙头在你的左后方。", + "evidence": [ + { + "description": "The location of the faucet and the white cutting board", + "timestamps": { + "start": "00:17:30", + "end": "00:17:36" + } + } + ], + "question_timestamps": "00:31:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The faucet is in front of you.", + "B": "The faucet is to your right rear.", + "C": "The faucet is to your left front, slightly ahead of where you are standing.", + "D": "The faucet is to your left rear.", + "E": "The faucet is directly behind you, centered along the counter line rather than off to the side." + }, + "answer_label": "D", + "answer_text": "The faucet is to your left rear.", + "q_id": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_5", + "correct_answer": "The faucet is to your left rear.", + "original_options": { + "A": "The faucet is in front of you.", + "B": "The faucet is to your right rear.", + "C": "The faucet is to your left front, slightly ahead of where you are standing.", + "D": "The faucet is to your left rear.", + "E": "The faucet is directly behind you, centered along the counter line rather than off to the side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 17 + ] + }, + { + "video_uid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe", + "video_length": "01:02:13", + "qid": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_6", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the faucet in?", + "answer": "The faucet is to your left.", + "question_chinese": "水龙头在我的什么方向？", + "answer_chinese": "水龙头在你的左方。", + "evidence": [ + { + "description": "The location of the faucet and the transparent bowl.", + "timestamps": { + "start": "00:43:54", + "end": "00:44:12" + } + } + ], + "question_timestamps": "00:46:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The faucet is to your right.", + "B": "The faucet is to your left.", + "C": "The faucet is in front of you.", + "D": "The faucet is behind you, slightly to the right, not on your left side.", + "E": "The faucet is to your right rear, so you would need to turn back to face it." + }, + "answer_label": "B", + "answer_text": "The faucet is to your left.", + "q_id": "0fedfa42-bfda-47c1-8f59-2c91b2c229fe_6", + "correct_answer": "The faucet is to your left.", + "original_options": { + "A": "The faucet is to your right.", + "B": "The faucet is to your left.", + "C": "The faucet is in front of you.", + "D": "The faucet is behind you, slightly to the right, not on your left side.", + "E": "The faucet is to your right rear, so you would need to turn back to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 17 + ] + } +] diff --git a/QAs/10a7fd5a-169d-4b0c-ba8b-825af7bd2117-mcq.json b/QAs/10a7fd5a-169d-4b0c-ba8b-825af7bd2117-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1d6c3056b9fa6f57c8d921781aa9e22b955d5663 --- /dev/null +++ b/QAs/10a7fd5a-169d-4b0c-ba8b-825af7bd2117-mcq.json @@ -0,0 +1,708 @@ +[ + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_15", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up until now, how many things have I put in the refrigerator?", + "answer": "One.", + "question_chinese": "到现在为止，我放了多少样东西在冰箱？", + "answer_chinese": "一。", + "evidence": [ + { + "description": "You put the pizza from the counter into the fridge.", + "timestamps": { + "start": "00:28:40", + "end": "00:28:42" + } + } + ], + "question_timestamps": "00:28:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "You have put two things in the refrigerator so far, not just a single item.", + "D": "Up to now, you placed three separate items into the refrigerator during your actions.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_15", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "You have put two things in the refrigerator so far, not just a single item.", + "D": "Up to now, you placed three separate items into the refrigerator during your actions.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_16", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up until now, how many things have I put in the refrigerator?", + "answer": "Two.", + "question_chinese": "到现在为止，我放了多少样东西在冰箱？", + "answer_chinese": "二。", + "evidence": [ + { + "description": "You put the pizza from the counter into the fridge.", + "timestamps": { + "start": "00:28:40", + "end": "00:28:41" + } + }, + { + "description": "After putting down the pizza, you took something from another counter and placed it in the fridge.", + "timestamps": { + "start": "00:28:43", + "end": "00:28:45" + } + } + ], + "question_timestamps": "00:28:50", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Zero.", + "C": "You have put one thing in the refrigerator up to this point, not two items.", + "D": "Two.", + "E": "So far, you have placed three different items into the refrigerator rather than only two." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_16", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Zero.", + "C": "You have put one thing in the refrigerator up to this point, not two items.", + "D": "Two.", + "E": "So far, you have placed three different items into the refrigerator rather than only two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_7", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get from where I was just now to the kitchen?", + "answer": "After you left your workstation, you turned left, went straight down the hallway, then turned right into the kitchen.", + "question_chinese": "我从刚刚的位置是怎么走到厨房的?", + "answer_chinese": "你离开工位后左转,沿走廊直行,然后右转进入厨房。", + "evidence": [ + { + "description": "You turn left first.", + "timestamps": { + "start": "00:06:04", + "end": "00:06:08" + } + }, + { + "description": "Then walk straight along the corridor.", + "timestamps": { + "start": "00:06:09", + "end": "00:06:11" + } + }, + { + "description": "Finally, turn right into the kitchen.", + "timestamps": { + "start": "00:06:11", + "end": "00:06:12" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turned right into the kitchen.", + "B": "You went straight into the kitchen.", + "C": "After leaving your workstation, you turned right, walked down the hallway, then turned left into the kitchen.", + "D": "After you left your workstation, you turned left, went down the hallway, and entered the kitchen without turning right.", + "E": "After you left your workstation, you turned left, went straight down the hallway, then turned right into the kitchen." + }, + "answer_label": "E", + "answer_text": "After you left your workstation, you turned left, went straight down the hallway, then turned right into the kitchen.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_7", + "correct_answer": "After you left your workstation, you turned left, went straight down the hallway, then turned right into the kitchen.", + "original_options": { + "A": "You turned right into the kitchen.", + "B": "You went straight into the kitchen.", + "C": "After leaving your workstation, you turned right, walked down the hallway, then turned left into the kitchen.", + "D": "After you left your workstation, you turned left, went down the hallway, and entered the kitchen without turning right.", + "E": "After you left your workstation, you turned left, went straight down the hallway, then turned right into the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_8", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get from where I was just now to the kitchen?", + "answer": "After you opened the door, you went straight down the hallway, then turned right into the kitchen.", + "question_chinese": "我从刚刚的位置是怎么走到厨房的?", + "answer_chinese": "你开门后沿走廊直行,然后右转进入厨房。", + "evidence": [ + { + "description": "You first go straight along the corridor.", + "timestamps": { + "start": "00:23:53", + "end": "00:23:55" + } + }, + { + "description": "Then turn right into the kitchen.", + "timestamps": { + "start": "00:23:55", + "end": "00:23:57" + } + } + ], + "question_timestamps": "00:24:02", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "After you opened the door, you went straight down the hallway, then turned right into the kitchen.", + "B": "You turned left into the kitchen.", + "C": "You turned right immediately into the kitchen.", + "D": "After you opened the door, you went down the hallway and then turned left into the kitchen.", + "E": "After opening the door, you walked straight down the hallway, then continued past the kitchen without turning right." + }, + "answer_label": "A", + "answer_text": "After you opened the door, you went straight down the hallway, then turned right into the kitchen.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_8", + "correct_answer": "After you opened the door, you went straight down the hallway, then turned right into the kitchen.", + "original_options": { + "A": "After you opened the door, you went straight down the hallway, then turned right into the kitchen.", + "B": "You turned left into the kitchen.", + "C": "You turned right immediately into the kitchen.", + "D": "After you opened the door, you went down the hallway and then turned left into the kitchen.", + "E": "After opening the door, you walked straight down the hallway, then continued past the kitchen without turning right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 18 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_3_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks did I just pour out?", + "answer": "You didn’t pour out any drinks just now. The drinks are below you.", + "question_chinese": "我刚刚倒掉了多少瓶饮料？", + "answer_chinese": "你刚刚没有倒掉饮料。饮料在你的下面。", + "evidence": [ + { + "description": "There was a bottle of drink in the sink that you picked up", + "timestamps": { + "start": "00:06:13", + "end": "00:06:16" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You didn’t pour out any drinks just now. The drinks are below you.", + "B": "You poured out one bottle.", + "C": "You poured out two bottles.", + "D": "You just poured out one bottle of drinks, and the drinks are in front of you.", + "E": "You just poured out three bottles of drinks, with the spilled drinks pooling below you." + }, + "answer_label": "A", + "answer_text": "You didn’t pour out any drinks just now. The drinks are below you.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_3_T1", + "correct_answer": "You didn’t pour out any drinks just now. The drinks are below you.", + "original_options": { + "A": "You didn’t pour out any drinks just now. The drinks are below you.", + "B": "You poured out one bottle.", + "C": "You poured out two bottles.", + "D": "You just poured out one bottle of drinks, and the drinks are in front of you.", + "E": "You just poured out three bottles of drinks, with the spilled drinks pooling below you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_3_T4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks did I just pour out?", + "answer": "You just poured out one bottle of drinks. The drinks are below you.", + "question_chinese": "我刚刚倒掉了多少瓶饮料？", + "answer_chinese": "你刚刚倒掉了一瓶饮料。饮料在你的下面。", + "evidence": [ + { + "description": "There was a bottle of drink in the sink that you picked up and poured out.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:26" + } + } + ], + "question_timestamps": "00:06:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You poured out zero bottles.", + "B": "You poured out two bottles.", + "C": "You just poured out three bottles of drinks, and the drinks are below you.", + "D": "You just poured out one bottle of drinks. The drinks are below you.", + "E": "You just poured out four bottles of drinks, with the drinks spilled below you." + }, + "answer_label": "D", + "answer_text": "You just poured out one bottle of drinks. The drinks are below you.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_3_T4", + "correct_answer": "You just poured out one bottle of drinks. The drinks are below you.", + "original_options": { + "A": "You poured out zero bottles.", + "B": "You poured out two bottles.", + "C": "You just poured out three bottles of drinks, and the drinks are below you.", + "D": "You just poured out one bottle of drinks. The drinks are below you.", + "E": "You just poured out four bottles of drinks, with the drinks spilled below you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_3_T3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks did I just pour out?", + "answer": "You just poured out two bottles of drinks. The drinks are in front of you.", + "question_chinese": "我刚刚倒掉了多少瓶饮料？", + "answer_chinese": "你刚刚倒掉了两瓶饮料。饮料在你的前面。", + "evidence": [ + { + "description": "There was a bottle of drink in the sink that you picked up and poured out.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:26" + } + }, + { + "description": "There was a bottle of drink on the table that you picked up and poured out.", + "timestamps": { + "start": "00:06:29", + "end": "00:06:40" + } + } + ], + "question_timestamps": "00:06:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You poured out zero bottles.", + "B": "You poured out one bottle.", + "C": "You just poured out two bottles of drinks. The drinks are in front of you.", + "D": "You just poured out three bottles of drinks, and the drinks are in front of you.", + "E": "You just poured out four bottles of drinks, with the drinks spread out in front of you." + }, + "answer_label": "C", + "answer_text": "You just poured out two bottles of drinks. The drinks are in front of you.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_3_T3", + "correct_answer": "You just poured out two bottles of drinks. The drinks are in front of you.", + "original_options": { + "A": "You poured out zero bottles.", + "B": "You poured out one bottle.", + "C": "You just poured out two bottles of drinks. The drinks are in front of you.", + "D": "You just poured out three bottles of drinks, and the drinks are in front of you.", + "E": "You just poured out four bottles of drinks, with the drinks spread out in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_6_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was just pouring the food, could I reach the spoon directly?", + "answer": "You could reach the spoon directly. The spoon is in front of you.", + "question_chinese": "我刚刚倒食物的时候，能直接够到勺子吗？", + "answer_chinese": "你能直接够到勺子。勺子在你的前面。", + "evidence": [ + { + "description": "You can reach the spoon directly.", + "timestamps": { + "start": "00:24:43", + "end": "00:24:44" + } + } + ], + "question_timestamps": "00:24:46", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn’t reach the spoon directly. The spoon is in front of you.", + "B": "You could reach the spoon directly. The spoon is in front of you." + }, + "answer_label": "B", + "answer_text": "You could reach the spoon directly. The spoon is in front of you.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_6_T2", + "correct_answer": "You could reach the spoon directly. The spoon is in front of you.", + "original_options": { + "A": "You couldn’t reach the spoon directly. The spoon is in front of you.", + "B": "You could reach the spoon directly. The spoon is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_6_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was just pouring the food, could I reach the spoon directly?", + "answer": "You couldn’t reach the spoon directly. The spoon is to your left.", + "question_chinese": "我刚刚倒食物的时候，能直接够到勺子吗？", + "answer_chinese": "你不能直接够到勺子。勺子在你的左边。", + "evidence": [ + { + "description": "There are no spoons on the table.", + "timestamps": { + "start": "00:24:55", + "end": "00:25:00" + } + } + ], + "question_timestamps": "00:25:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn’t reach the spoon directly. The spoon is to your left.", + "B": "You could reach the spoon directly. The spoon is to your left." + }, + "answer_label": "A", + "answer_text": "You couldn’t reach the spoon directly. The spoon is to your left.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_6_T1", + "correct_answer": "You couldn’t reach the spoon directly. The spoon is to your left.", + "original_options": { + "A": "You couldn’t reach the spoon directly. The spoon is to your left.", + "B": "You could reach the spoon directly. The spoon is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_19", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which route am I taking right now?", + "answer": "You are leaving the kitchen, going straight, heading toward the bathroom.", + "question_chinese": "我现在正在走什么路线?", + "answer_chinese": "你正从厨房出发,直行,朝着厕所方向移动。", + "evidence": [ + { + "description": "You went straight and turned left into the bathroom.", + "timestamps": { + "start": "00:31:30", + "end": "00:31:34" + } + } + ], + "question_timestamps": "00:31:39", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are heading toward the kitchen.", + "B": "You are leaving the bathroom.", + "C": "You are leaving the kitchen, turning left, and heading down the hall toward the living room.", + "D": "You are walking straight from the kitchen but continuing past the bathroom toward the next room.", + "E": "You are leaving the kitchen, going straight, heading toward the bathroom." + }, + "answer_label": "E", + "answer_text": "You are leaving the kitchen, going straight, heading toward the bathroom.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_19", + "correct_answer": "You are leaving the kitchen, going straight, heading toward the bathroom.", + "original_options": { + "A": "You are heading toward the kitchen.", + "B": "You are leaving the bathroom.", + "C": "You are leaving the kitchen, turning left, and heading down the hall toward the living room.", + "D": "You are walking straight from the kitchen but continuing past the bathroom toward the next room.", + "E": "You are leaving the kitchen, going straight, heading toward the bathroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 16 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_20", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which route am I taking right now?", + "answer": "You are leaving the bathroom, going straight, and moving towards the kitchen.", + "question_chinese": "我现在正在走什么路线?", + "answer_chinese": "你正从厕所出发，直行，朝着厨房方向移动。", + "evidence": [ + { + "description": "You left the bathroom, went straight, and entered the kitchen.", + "timestamps": { + "start": "00:32:20", + "end": "00:32:24" + } + } + ], + "question_timestamps": "00:32:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are leaving the bathroom, turning left.", + "B": "You are leaving the bathroom, turning right.", + "C": "You are leaving the bathroom, going straight, and moving towards the kitchen.", + "D": "You are leaving the bathroom, turning left, and heading down the hallway toward the living room.", + "E": "You are leaving the bathroom, going straight, but moving away from the kitchen toward the bedroom instead." + }, + "answer_label": "C", + "answer_text": "You are leaving the bathroom, going straight, and moving towards the kitchen.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_20", + "correct_answer": "You are leaving the bathroom, going straight, and moving towards the kitchen.", + "original_options": { + "A": "You are leaving the bathroom, turning left.", + "B": "You are leaving the bathroom, turning right.", + "C": "You are leaving the bathroom, going straight, and moving towards the kitchen.", + "D": "You are leaving the bathroom, turning left, and heading down the hallway toward the living room.", + "E": "You are leaving the bathroom, going straight, but moving away from the kitchen toward the bedroom instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_1", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Relative to my current position, where are the sunglasses?", + "answer": "The sunglasses are on the top right of you, on the table.", + "question_chinese": "相对我现在的位置，墨镜在哪里？", + "answer_chinese": "墨镜在你的右上方，桌子上。", + "evidence": [ + { + "description": "There is a pair of sunglasses on the table, next to the bottle.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They are on the top left.", + "B": "They are on the bottom right.", + "C": "The sunglasses are on the top right of you, on the table.", + "D": "The sunglasses are on the top left of you, resting on the table beside your arm.", + "E": "The sunglasses are on the bottom right of you, sitting on the table near the edge." + }, + "answer_label": "C", + "answer_text": "The sunglasses are on the top right of you, on the table.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_1", + "correct_answer": "The sunglasses are on the top right of you, on the table.", + "original_options": { + "A": "They are on the top left.", + "B": "They are on the bottom right.", + "C": "The sunglasses are on the top right of you, on the table.", + "D": "The sunglasses are on the top left of you, resting on the table beside your arm.", + "E": "The sunglasses are on the bottom right of you, sitting on the table near the edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117", + "video_length": "00:34:51", + "qid": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Relative to my current position, where are the sunglasses?", + "answer": "The sunglasses are directly in front of me.", + "question_chinese": "相对我现在的位置，墨镜在哪里？", + "answer_chinese": "墨镜在我的正前方。", + "evidence": [ + { + "description": "There is a pair of sunglasses on the table, next to the table.t", + "timestamps": { + "start": "00:04:15", + "end": "00:04:19" + } + } + ], + "question_timestamps": "00:04:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They are off to your left.", + "B": "They are off to your right.", + "C": "The sunglasses are behind you, slightly to the left, not in your forward view.", + "D": "The sunglasses are directly in front of me.", + "E": "The sunglasses are in front of you but far to the right, closer to the desk edge." + }, + "answer_label": "D", + "answer_text": "The sunglasses are directly in front of me.", + "q_id": "10a7fd5a-169d-4b0c-ba8b-825af7bd2117_2", + "correct_answer": "The sunglasses are directly in front of me.", + "original_options": { + "A": "They are off to your left.", + "B": "They are off to your right.", + "C": "The sunglasses are behind you, slightly to the left, not in your forward view.", + "D": "The sunglasses are directly in front of me.", + "E": "The sunglasses are in front of you but far to the right, closer to the desk edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 17 + ] + } +] diff --git a/QAs/11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a-mcq.json b/QAs/11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ebdabd01182661ac520586e4b8c4bede5a79e79e --- /dev/null +++ b/QAs/11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a-mcq.json @@ -0,0 +1,1293 @@ +[ + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_17", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to place the level ruler closer to the welding machine, how far should I move it?", + "answer": "You can move the level ruler about one.5 meters to the lower right.", + "question_chinese": "如果我想让水平尺和焊接机放得近一些,应该移动多远?", + "answer_chinese": "你可以把水平尺向右下方移动1.5m左右。", + "evidence": [ + { + "description": "They are currently about 1.5 meters apart, so you can move the level ruler about 1.5 meters to the right and downwards.", + "timestamps": { + "start": "00:03:55", + "end": "00:04:06" + } + } + ], + "question_timestamps": "00:04:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You can move the level ruler about one.5 meters to the lower right.", + "B": "Move it 1.5 meters lower left.", + "C": "Move it 1.0 meter lower right.", + "D": "Move the level ruler about 1.5 meters to the lower left to bring it nearer the welding machine.", + "E": "Move the level ruler roughly 2 meters down and slightly right so it ends up closer to the welding machine." + }, + "answer_label": "A", + "answer_text": "You can move the level ruler about one.5 meters to the lower right.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_17", + "correct_answer": "You can move the level ruler about one.5 meters to the lower right.", + "original_options": { + "A": "You can move the level ruler about one.5 meters to the lower right.", + "B": "Move it 1.5 meters lower left.", + "C": "Move it 1.0 meter lower right.", + "D": "Move the level ruler about 1.5 meters to the lower left to bring it nearer the welding machine.", + "E": "Move the level ruler roughly 2 meters down and slightly right so it ends up closer to the welding machine." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 20 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_18", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to place the level ruler closer to the welding machine, how far should I move it?", + "answer": "You can move the welding machine downwards by about one.7 meters.", + "question_chinese": "如果我想让水平尺和焊接机放得近一些,应该移动多远?", + "answer_chinese": "你可以把焊接机向下移动1.7m左右。", + "evidence": [ + { + "description": "Their current vertical distance is approximately 1.7 meters, so you can move the welding machine downwards by about 1.7 meters.", + "timestamps": { + "start": "00:22:34", + "end": "00:22:35" + } + } + ], + "question_timestamps": "00:22:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You can move the welding machine downwards by about one.7 meters.", + "B": "Move it 1.7 meters upward.", + "C": "Move it 1.0 meter downward.", + "D": "You should move the welding machine about 1.7 meters upward instead of downward to make it closer.", + "E": "You should move the welding machine about 2 meters downward, not 1.7, to place it closer to the level ruler." + }, + "answer_label": "A", + "answer_text": "You can move the welding machine downwards by about one.7 meters.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_18", + "correct_answer": "You can move the welding machine downwards by about one.7 meters.", + "original_options": { + "A": "You can move the welding machine downwards by about one.7 meters.", + "B": "Move it 1.7 meters upward.", + "C": "Move it 1.0 meter downward.", + "D": "You should move the welding machine about 1.7 meters upward instead of downward to make it closer.", + "E": "You should move the welding machine about 2 meters downward, not 1.7, to place it closer to the level ruler." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 21 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_6", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was welding on the top floor, which was closer to me, the angle grinder or the level ruler?", + "answer": "The angle grinder is closer to you.", + "question_chinese": "当我在顶楼焊接的时候,角磨机和水平尺哪个离我更近?", + "answer_chinese": "角磨机离你更近。", + "evidence": [ + { + "description": "The angle grinder is closer to you, with a horizontal distance of about 5cm. The level ruler is to your right front, with a distance of about 75cm.", + "timestamps": { + "start": "00:15:48", + "end": "00:15:59" + } + } + ], + "question_timestamps": "00:16:04", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The angle grinder is closer to you.", + "B": "The level ruler is closer to you." + }, + "answer_label": "A", + "answer_text": "The angle grinder is closer to you.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_6", + "correct_answer": "The angle grinder is closer to you.", + "original_options": { + "A": "The angle grinder is closer to you.", + "B": "The level ruler is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_5", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was welding on the top floor, which was closer to me, the angle grinder or the level ruler?", + "answer": "The angle grinder is closer to you.", + "question_chinese": "当我在顶楼焊接的时候,角磨机和水平尺哪个离我更近?", + "answer_chinese": "角磨机离你更近。", + "evidence": [ + { + "description": "The angle grinder is closer to you, with a horizontal distance of about 5cm. The level ruler is to your right front, with a distance of about 75cm.", + "timestamps": { + "start": "00:15:48", + "end": "00:15:59" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The angle grinder is closer to you.", + "B": "The level ruler is closer to you." + }, + "answer_label": "A", + "answer_text": "The angle grinder is closer to you.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_5", + "correct_answer": "The angle grinder is closer to you.", + "original_options": { + "A": "The angle grinder is closer to you.", + "B": "The level ruler is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_6_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I put down the spirit level, which direction was it relative to me?", + "answer": "The level is behind you.", + "question_chinese": "当我放下水平尺时，它在我的哪个方向？", + "answer_chinese": "水平尺在你的后面。", + "evidence": [ + { + "description": "The level is behind you.", + "timestamps": { + "start": "00:06:50", + "end": "00:06:53" + } + } + ], + "question_timestamps": "00:06:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The level is in front of you.", + "B": "The level is behind you.", + "C": "The level is to your left.", + "D": "The level is to your right, close to your side when you put it down.", + "E": "The level is ahead of you on the floor, a little farther than your feet after you set it down." + }, + "answer_label": "B", + "answer_text": "The level is behind you.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_6_T1", + "correct_answer": "The level is behind you.", + "original_options": { + "A": "The level is in front of you.", + "B": "The level is behind you.", + "C": "The level is to your left.", + "D": "The level is to your right, close to your side when you put it down.", + "E": "The level is ahead of you on the floor, a little farther than your feet after you set it down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 20 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_6_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I put down the spirit level, which direction was it relative to me?", + "answer": "The level is on the left side of you.", + "question_chinese": "当我放下水平尺时，它在我的哪个方向？", + "answer_chinese": "水平尺在你的左侧。", + "evidence": [ + { + "description": "The level ruler is on the ground to your left.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:49" + } + } + ], + "question_timestamps": "01:01:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The level is on the left side of you.", + "B": "It was on your right.", + "C": "It was in front of you.", + "D": "You set the spirit level down behind you, not to either side of your body.", + "E": "When you put it down, the spirit level ended up directly in front of you on the work surface." + }, + "answer_label": "A", + "answer_text": "The level is on the left side of you.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_6_T2", + "correct_answer": "The level is on the left side of you.", + "original_options": { + "A": "The level is on the left side of you.", + "B": "It was on your right.", + "C": "It was in front of you.", + "D": "You set the spirit level down behind you, not to either side of your body.", + "E": "When you put it down, the spirit level ended up directly in front of you on the work surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 19 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_13", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first put down the level ruler, which direction was it pointing towards me?", + "answer": "The level ruler is on your left.", + "question_chinese": "当我第一次放下水平尺时，它在我的哪个方向？", + "answer_chinese": "水平尺在你的左侧。", + "evidence": [ + { + "description": "The level ruler is on the ground to your left.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:49" + } + } + ], + "question_timestamps": "00:00:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The level ruler is on your left.", + "B": "It was on your right.", + "C": "It was behind you.", + "D": "The level ruler was placed in front of you rather than off to your left side.", + "E": "When you first set it down, the level ruler ended up behind you instead of to your left." + }, + "answer_label": "A", + "answer_text": "The level ruler is on your left.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_13", + "correct_answer": "The level ruler is on your left.", + "original_options": { + "A": "The level ruler is on your left.", + "B": "It was on your right.", + "C": "It was behind you.", + "D": "The level ruler was placed in front of you rather than off to your left side.", + "E": "When you first set it down, the level ruler ended up behind you instead of to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 18 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_14", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first put down the level ruler, which direction was it pointing towards me?", + "answer": "The level ruler is on your left.", + "question_chinese": "当我第一次放下水平尺时，它在我的哪个方向？", + "answer_chinese": "水平尺在你的左侧。", + "evidence": [ + { + "description": "The level ruler is on the ground to your left.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:49" + } + } + ], + "question_timestamps": "01:01:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "It was on your right.", + "C": "You first put the level ruler down behind you, not on the left side.", + "D": "When you placed it initially, the level ruler ended up in front of you rather than to your left.", + "E": "The level ruler is on your left." + }, + "answer_label": "E", + "answer_text": "The level ruler is on your left.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_14", + "correct_answer": "The level ruler is on your left.", + "original_options": { + "A": "It was in front of you.", + "B": "It was on your right.", + "C": "You first put the level ruler down behind you, not on the left side.", + "D": "When you placed it initially, the level ruler ended up in front of you rather than to your left.", + "E": "The level ruler is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 19 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_11", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I drunk water in total?", + "answer": "You drank water twice in total.", + "question_chinese": "我一共喝了几次水？", + "answer_chinese": "你一共喝了两次水。", + "evidence": [ + { + "description": "You drink water for the first time.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:06" + } + }, + { + "description": "You drink water for the second time.", + "timestamps": { + "start": "00:14:08", + "end": "00:14:21" + } + } + ], + "question_timestamps": "00:14:26", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You drank water twice in total.", + "B": "You drank water once.", + "C": "You drank water three times.", + "D": "In total, you took four separate drinks of water across the video.", + "E": "Overall, you drank water six times, with multiple sips counted as separate drinking moments." + }, + "answer_label": "A", + "answer_text": "You drank water twice in total.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_11", + "correct_answer": "You drank water twice in total.", + "original_options": { + "A": "You drank water twice in total.", + "B": "You drank water once.", + "C": "You drank water three times.", + "D": "In total, you took four separate drinks of water across the video.", + "E": "Overall, you drank water six times, with multiple sips counted as separate drinking moments." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 12, + 14 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_12", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I drunk water in total?", + "answer": "You drank water three times in total.", + "question_chinese": "我一共喝了几次水？", + "answer_chinese": "你一共喝了三次水。", + "evidence": [ + { + "description": "You drink water for the first time.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:06" + } + }, + { + "description": "You drink water for the second time.", + "timestamps": { + "start": "00:14:08", + "end": "00:14:21" + } + }, + { + "description": "You are drinking water for the third time.", + "timestamps": { + "start": "00:28:32", + "end": "00:28:55" + } + } + ], + "question_timestamps": "00:28:58", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You drank water once.", + "B": "You drank water three times in total.", + "C": "You drank water five times.", + "D": "In total, you drank water seven times throughout the entire sequence of actions.", + "E": "Overall, you drank water nine times, counting each distinct time you went back to drink." + }, + "answer_label": "B", + "answer_text": "You drank water three times in total.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_12", + "correct_answer": "You drank water three times in total.", + "original_options": { + "A": "You drank water once.", + "B": "You drank water three times in total.", + "C": "You drank water five times.", + "D": "In total, you drank water seven times throughout the entire sequence of actions.", + "E": "Overall, you drank water nine times, counting each distinct time you went back to drink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 15 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_19", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I changed the working attachments on the angle grinder?", + "answer": "You changed it once in total.", + "question_chinese": "我一共换了几次角磨机上的工作附件？", + "answer_chinese": "你一共换了一次。", + "evidence": [ + { + "description": "You replaced the working attachment on the angle grinder with a cutting blade.", + "timestamps": { + "start": "00:05:26", + "end": "00:05:33" + } + } + ], + "question_timestamps": "00:05:38", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You changed it twice.", + "B": "You changed it three times.", + "C": "You changed it once in total.", + "D": "You switched the angle grinder attachment four separate times over the course of the task.", + "E": "In total, you changed the working attachment five times, replacing it repeatedly between steps." + }, + "answer_label": "C", + "answer_text": "You changed it once in total.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_19", + "correct_answer": "You changed it once in total.", + "original_options": { + "A": "You changed it twice.", + "B": "You changed it three times.", + "C": "You changed it once in total.", + "D": "You switched the angle grinder attachment four separate times over the course of the task.", + "E": "In total, you changed the working attachment five times, replacing it repeatedly between steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 14 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_20", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I changed the working attachments on the angle grinder?", + "answer": "You changed it three times in total.", + "question_chinese": "我一共换了几次角磨机上的工作附件？", + "answer_chinese": "你一共换了三次。", + "evidence": [ + { + "description": "You replaced the working attachment on the angle grinder with a cutting blade.", + "timestamps": { + "start": "00:05:26", + "end": "00:05:33" + } + }, + { + "description": "You replaced the working attachment on the angle grinder with a grinding disc.", + "timestamps": { + "start": "00:07:27", + "end": "00:07:34" + } + }, + { + "description": "You've changed the working attachment on the angle grinder to a cutting blade again.", + "timestamps": { + "start": "00:21:03", + "end": "00:21:11" + } + } + ], + "question_timestamps": "00:21:16", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You changed it four times.", + "B": "You changed it three times in total.", + "C": "You changed it five times.", + "D": "Altogether, you swapped the angle grinder’s working attachment six times during the workflow.", + "E": "You changed the working attachments seven times in total, making several quick swaps between uses." + }, + "answer_label": "B", + "answer_text": "You changed it three times in total.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_20", + "correct_answer": "You changed it three times in total.", + "original_options": { + "A": "You changed it four times.", + "B": "You changed it three times in total.", + "C": "You changed it five times.", + "D": "Altogether, you swapped the angle grinder’s working attachment six times during the workflow.", + "E": "You changed the working attachments seven times in total, making several quick swaps between uses." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_4", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I use the angle grinder in total?", + "answer": "You used the angle grinder once in total.", + "question_chinese": "我一共用了几次角磨机？", + "answer_chinese": "你一共用了一次角磨机。", + "evidence": [ + { + "description": "You used the angle grinder once in total.", + "timestamps": { + "start": "00:05:41", + "end": "00:06:05" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used it twice.", + "B": "You used it three times.", + "C": "Across the whole video, you used the angle grinder four different times for separate cuts or grinds.", + "D": "You used the angle grinder once in total.", + "E": "In total, you used the angle grinder five times, returning to it repeatedly throughout the task." + }, + "answer_label": "D", + "answer_text": "You used the angle grinder once in total.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_4", + "correct_answer": "You used the angle grinder once in total.", + "original_options": { + "A": "You used it twice.", + "B": "You used it three times.", + "C": "Across the whole video, you used the angle grinder four different times for separate cuts or grinds.", + "D": "You used the angle grinder once in total.", + "E": "In total, you used the angle grinder five times, returning to it repeatedly throughout the task." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 16 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_3", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I use the angle grinder in total?", + "answer": "You used the angle grinder a total of four times.", + "question_chinese": "我一共用了几次角磨机？", + "answer_chinese": "你一共用了四次角磨机。", + "evidence": [ + { + "description": "You used the angle grinder once in total.", + "timestamps": { + "start": "00:05:41", + "end": "00:06:05" + } + }, + { + "description": "This is the second time you use the angle grinder.", + "timestamps": { + "start": "00:06:55", + "end": "00:07:22" + } + }, + { + "description": "This is the third time you have used an angle grinder.", + "timestamps": { + "start": "00:07:58", + "end": "00:08:25" + } + }, + { + "description": "This is the fourth time you have used the angle grinder.", + "timestamps": { + "start": "00:21:15", + "end": "00:21:19" + } + } + ], + "question_timestamps": "00:21:21", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used it three times.", + "B": "You used it five times.", + "C": "You used the angle grinder a total of two times during the entire task, then stopped using it.", + "D": "You used the angle grinder a total of six times overall, returning to it repeatedly throughout the sequence.", + "E": "You used the angle grinder a total of four times." + }, + "answer_label": "E", + "answer_text": "You used the angle grinder a total of four times.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_3", + "correct_answer": "You used the angle grinder a total of four times.", + "original_options": { + "A": "You used it three times.", + "B": "You used it five times.", + "C": "You used the angle grinder a total of two times during the entire task, then stopped using it.", + "D": "You used the angle grinder a total of six times overall, returning to it repeatedly throughout the sequence.", + "E": "You used the angle grinder a total of four times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_7", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was sitting on the stairs just now, could I reach the hammer directly?", + "answer": "You can't reach the hammer.", + "question_chinese": "我刚才坐在楼梯上的时候,能直接够到锤子吗?", + "answer_chinese": "你不能够到锤子。", + "evidence": [ + { + "description": "You can't reach it. The hammer is about 80cm away from you, on the ground in front of you on the right.", + "timestamps": { + "start": "00:30:00", + "end": "00:30:06" + } + } + ], + "question_timestamps": "00:30:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can reach the hammer.", + "B": "You can't reach the hammer." + }, + "answer_label": "B", + "answer_text": "You can't reach the hammer.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_7", + "correct_answer": "You can't reach the hammer.", + "original_options": { + "A": "You can reach the hammer.", + "B": "You can't reach the hammer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_8", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was sitting on the stairs just now, could I reach the hammer directly?", + "answer": "You can't reach the hammer.", + "question_chinese": "我刚才坐在楼梯上的时候,能直接够到锤子吗?", + "answer_chinese": "你不能够到锤子。", + "evidence": [ + { + "description": "The hammer is approximately 4.5 meters vertically away from you, on the ground of the first floor.", + "timestamps": { + "start": "00:47:42", + "end": "00:47:46" + } + }, + { + "description": "You are on the platform between the second and third floors.", + "timestamps": { + "start": "01:19:19", + "end": "01:19:24" + } + } + ], + "question_timestamps": "01:19:26", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't reach the hammer.", + "B": "You can reach the hammer." + }, + "answer_label": "A", + "answer_text": "You can't reach the hammer.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_8", + "correct_answer": "You can't reach the hammer.", + "original_options": { + "A": "You can't reach the hammer.", + "B": "You can reach the hammer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_10", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route am I currently taking?", + "answer": "You are moving downstairs from the top floor towards the first floor.", + "question_chinese": "我现在正在走什么路线?", + "answer_chinese": "你正从顶楼出发，下楼，朝着一楼的方向移动。", + "evidence": [ + { + "description": "You start from the top floor.", + "timestamps": { + "start": "00:01:51", + "end": "00:01:56" + } + }, + { + "description": "You are going downstairs.", + "timestamps": { + "start": "00:02:11", + "end": "00:02:24" + } + }, + { + "description": "You have arrived at the first floor.", + "timestamps": { + "start": "00:02:36", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:02:44", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You are heading upstairs to the top floor.", + "B": "You are walking across the first floor hallway.", + "C": "You are moving downstairs from the top floor towards the first floor.", + "D": "You are moving downstairs from the first floor toward the ground level, continuing down the staircase.", + "E": "You are coming up from the lower level and climbing the stairs toward the top floor." + }, + "answer_label": "C", + "answer_text": "You are moving downstairs from the top floor towards the first floor.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_10", + "correct_answer": "You are moving downstairs from the top floor towards the first floor.", + "original_options": { + "A": "You are heading upstairs to the top floor.", + "B": "You are walking across the first floor hallway.", + "C": "You are moving downstairs from the top floor towards the first floor.", + "D": "You are moving downstairs from the first floor toward the ground level, continuing down the staircase.", + "E": "You are coming up from the lower level and climbing the stairs toward the top floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 16 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_9", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route am I currently taking?", + "answer": "You are walking from outside the house, ascending the steps to enter, climbing the stairs, and moving towards the top floor.", + "question_chinese": "我现在正在走什么路线?", + "answer_chinese": "你正从房子外面出发,上台阶进屋，上楼,朝着顶楼方向移动。", + "evidence": [ + { + "description": "You start from outside the house and ascend the steps to enter the house.", + "timestamps": { + "start": "00:52:41", + "end": "00:52:56" + } + }, + { + "description": "You are climbing the stairs.", + "timestamps": { + "start": "00:53:09", + "end": "00:53:21" + } + }, + { + "description": "You have reached the top floor.", + "timestamps": { + "start": "00:53:47", + "end": "00:53:53" + } + } + ], + "question_timestamps": "00:53:55", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You are leaving the house and going outside.", + "B": "You are heading downstairs toward the first floor.", + "C": "You are walking from outside the house, ascending the steps to enter, climbing the stairs, and moving towards the top floor.", + "D": "You are already inside on the first floor, going down the stairs toward the ground level instead.", + "E": "You are approaching from outside, stepping in briefly, then turning back and descending the stairs again." + }, + "answer_label": "C", + "answer_text": "You are walking from outside the house, ascending the steps to enter, climbing the stairs, and moving towards the top floor.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_9", + "correct_answer": "You are walking from outside the house, ascending the steps to enter, climbing the stairs, and moving towards the top floor.", + "original_options": { + "A": "You are leaving the house and going outside.", + "B": "You are heading downstairs toward the first floor.", + "C": "You are walking from outside the house, ascending the steps to enter, climbing the stairs, and moving towards the top floor.", + "D": "You are already inside on the first floor, going down the stairs toward the ground level instead.", + "E": "You are approaching from outside, stepping in briefly, then turning back and descending the stairs again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 16 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_3_T2", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Right now, can I directly reach the hammer?", + "answer": "You can reach the hammer; the hammer is directly in front of you.", + "question_chinese": "现在,我能直接够到锤子吗?", + "answer_chinese": "你能够到锤子，锤子在你的正前方。", + "evidence": [ + { + "description": "You can reach the hammer, it's not far from you.", + "timestamps": { + "start": "00:26:10", + "end": "00:26:13" + } + } + ], + "question_timestamps": "00:26:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the hammer; it’s too far away.", + "B": "You can reach the hammer; the hammer is directly in front of you." + }, + "answer_label": "B", + "answer_text": "You can reach the hammer; the hammer is directly in front of you.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_3_T2", + "correct_answer": "You can reach the hammer; the hammer is directly in front of you.", + "original_options": { + "A": "You can’t reach the hammer; it’s too far away.", + "B": "You can reach the hammer; the hammer is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_3_T1", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Right now, can I directly reach the hammer?", + "answer": "You can’t reach the hammer; the hammer is to your left.", + "question_chinese": "现在,我能直接够到锤子吗?", + "answer_chinese": "你不能够到锤子，锤子在你的左边。", + "evidence": [ + { + "description": "You can't reach it. The hammer is about 80cm away from you, on the ground in front of you on the right.", + "timestamps": { + "start": "00:30:00", + "end": "00:30:06" + } + } + ], + "question_timestamps": "00:30:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the hammer; the hammer is to your left.", + "B": "You can reach the hammer; the hammer is to your left." + }, + "answer_label": "A", + "answer_text": "You can’t reach the hammer; the hammer is to your left.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_3_T1", + "correct_answer": "You can’t reach the hammer; the hammer is to your left.", + "original_options": { + "A": "You can’t reach the hammer; the hammer is to your left.", + "B": "You can reach the hammer; the hammer is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_2_T1", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me, the angle grinder or the spirit level?", + "answer": "The level is closer to you, and the level is directly in front of you.", + "question_chinese": "现在,角磨机和水平尺哪个离我更近?", + "answer_chinese": "水平尺离你更近，水平尺在你的正前方。", + "evidence": [ + { + "description": "The level is closer to you.", + "timestamps": { + "start": "00:11:15", + "end": "00:11:17" + } + } + ], + "question_timestamps": "00:11:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The angle grinder is closer to you, and the angle grinder is directly in front of you.", + "B": "The level is closer to you, and the level is directly in front of you." + }, + "answer_label": "B", + "answer_text": "The level is closer to you, and the level is directly in front of you.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_2_T1", + "correct_answer": "The level is closer to you, and the level is directly in front of you.", + "original_options": { + "A": "The angle grinder is closer to you, and the angle grinder is directly in front of you.", + "B": "The level is closer to you, and the level is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_2_T2", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me, the angle grinder or the spirit level?", + "answer": "The angle grinder is closer to you, and the angle grinder is directly in front of you.", + "question_chinese": "现在,角磨机和水平尺哪个离我更近?", + "answer_chinese": "角磨机离你更近，角磨机在你的正前方。", + "evidence": [ + { + "description": "The angle grinder is closer to you, with a horizontal distance of about 5cm. The level ruler is to your right front, with a distance of about 75cm.", + "timestamps": { + "start": "00:15:48", + "end": "00:15:59" + } + } + ], + "question_timestamps": "00:16:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The spirit level is closer to you, and the spirit level is directly in front of you.", + "B": "The angle grinder is closer to you, and the angle grinder is directly in front of you." + }, + "answer_label": "B", + "answer_text": "The angle grinder is closer to you, and the angle grinder is directly in front of you.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_2_T2", + "correct_answer": "The angle grinder is closer to you, and the angle grinder is directly in front of you.", + "original_options": { + "A": "The spirit level is closer to you, and the spirit level is directly in front of you.", + "B": "The angle grinder is closer to you, and the angle grinder is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_7_T1", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "In which direction is the blue water bottle relative to me?", + "answer": "The blue water bottle is to your left.", + "question_chinese": "蓝色水瓶在我的哪个方向？", + "answer_chinese": "蓝色水瓶在你的左边。", + "evidence": [ + { + "description": "The blue water bottle is on your left", + "timestamps": { + "start": "00:29:16", + "end": "00:29:19" + } + } + ], + "question_timestamps": "00:29:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to your right.", + "B": "It is in front of you.", + "C": "The blue water bottle is behind you, not on your left side where you would reach first.", + "D": "The blue water bottle is above you, so you would need to look up to see it clearly.", + "E": "The blue water bottle is to your left." + }, + "answer_label": "E", + "answer_text": "The blue water bottle is to your left.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_7_T1", + "correct_answer": "The blue water bottle is to your left.", + "original_options": { + "A": "It is to your right.", + "B": "It is in front of you.", + "C": "The blue water bottle is behind you, not on your left side where you would reach first.", + "D": "The blue water bottle is above you, so you would need to look up to see it clearly.", + "E": "The blue water bottle is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 18 + ] + }, + { + "video_uid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a", + "video_length": "01:25:10", + "qid": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_7_T2", + "qid_num": 24, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "In which direction is the blue water bottle relative to me?", + "answer": "The blue water bottle is in front of you.", + "question_chinese": "蓝色水瓶在我的哪个方向？", + "answer_chinese": "蓝色水瓶在你的前面。", + "evidence": [ + { + "description": "The blue water bottle is in front of you.", + "timestamps": { + "start": "00:58:06", + "end": "00:58:09" + } + } + ], + "question_timestamps": "00:58:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The blue water bottle is to your right, positioned off to the side rather than ahead.", + "D": "The blue water bottle is in front of you.", + "E": "The blue water bottle is above you, so it is not located in the space directly in front." + }, + "answer_label": "D", + "answer_text": "The blue water bottle is in front of you.", + "q_id": "11e56f1e-baef-4fe8-ad0f-a7ec8788ea3a_7_T2", + "correct_answer": "The blue water bottle is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The blue water bottle is to your right, positioned off to the side rather than ahead.", + "D": "The blue water bottle is in front of you.", + "E": "The blue water bottle is above you, so it is not located in the space directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 18 + ] + } +] diff --git a/QAs/15f1a422-327d-4dab-b14d-7003544c5817-mcq.json b/QAs/15f1a422-327d-4dab-b14d-7003544c5817-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..963deab978230ef7c92789359c0a5a6f7d8ce117 --- /dev/null +++ b/QAs/15f1a422-327d-4dab-b14d-7003544c5817-mcq.json @@ -0,0 +1,1841 @@ +[ + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_5_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the kitchen paper to my left or to my right?", + "answer": "The kitchen paper towels are to your left.", + "question_chinese": "厨房用纸在我的左/右？", + "answer_chinese": "厨房用纸在你的左边。", + "evidence": [ + { + "description": "The kitchen paper is on my left side.", + "timestamps": { + "start": "00:13:49", + "end": "00:13:53" + } + } + ], + "question_timestamps": "00:15:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The kitchen paper towels are to your left.", + "B": "The kitchen paper towels are by your right hand." + }, + "answer_label": "A", + "answer_text": "The kitchen paper towels are to your left.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_5_T1", + "correct_answer": "The kitchen paper towels are to your left.", + "original_options": { + "A": "The kitchen paper towels are to your left.", + "B": "The kitchen paper towels are by your right hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_5_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the kitchen paper to my left or to my right?", + "answer": "The kitchen paper towels are by your right hand.", + "question_chinese": "厨房用纸在我的左/右？", + "answer_chinese": "厨房用纸在你的右手边。", + "evidence": [ + { + "description": "The kitchen paper is on my right hand side.", + "timestamps": { + "start": "00:13:49", + "end": "00:13:53" + } + } + ], + "question_timestamps": "00:16:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The kitchen paper towels are to your left.", + "B": "The kitchen paper towels are by your right hand." + }, + "answer_label": "B", + "answer_text": "The kitchen paper towels are by your right hand.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_5_T2", + "correct_answer": "The kitchen paper towels are by your right hand.", + "original_options": { + "A": "The kitchen paper towels are to your left.", + "B": "The kitchen paper towels are by your right hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the scene?", + "answer": "There are two chairs in the scene.", + "question_chinese": "场景里一共出现了几把椅子？", + "answer_chinese": "场景里一共出现了两把椅子。", + "evidence": [ + { + "description": "There are two armchairs in the corner of the living room.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three chairs.", + "B": "There are four chairs.", + "C": "There are five chairs in the scene, spread out rather than just a pair.", + "D": "There are two chairs in the scene.", + "E": "There are six chairs in the scene, so more than two are visible at once." + }, + "answer_label": "D", + "answer_text": "There are two chairs in the scene.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_3", + "correct_answer": "There are two chairs in the scene.", + "original_options": { + "A": "There are three chairs.", + "B": "There are four chairs.", + "C": "There are five chairs in the scene, spread out rather than just a pair.", + "D": "There are two chairs in the scene.", + "E": "There are six chairs in the scene, so more than two are visible at once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the scene?", + "answer": "There are seven chairs in the scene.", + "question_chinese": "场景里一共出现了几把椅子？", + "answer_chinese": "场景里一共出现了七把椅子。", + "evidence": [ + { + "description": "There are two armchairs in the corner of the living room.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:20" + } + }, + { + "description": "There are two empty wooden chairs by the kitchen table, and the girl is sitting on one of them.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:07" + } + }, + { + "description": "There is an empty wooden chair and a black soft chair next to the sink.", + "timestamps": { + "start": "00:01:14", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:21:07", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are seven chairs in the scene.", + "B": "There are five chairs.", + "C": "There are eight chairs.", + "D": "There are four chairs in the scene, which is fewer than the total of seven.", + "E": "There are nine chairs in the scene, meaning you can count more than seven present." + }, + "answer_label": "A", + "answer_text": "There are seven chairs in the scene.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_4", + "correct_answer": "There are seven chairs in the scene.", + "original_options": { + "A": "There are seven chairs in the scene.", + "B": "There are five chairs.", + "C": "There are eight chairs.", + "D": "There are four chairs in the scene, which is fewer than the total of seven.", + "E": "There are nine chairs in the scene, meaning you can count more than seven present." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_1_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sofas are in the living room, and where are they relative to me?", + "answer": "There is a sofa in the living room, in front of you.", + "question_chinese": "客厅里有几个沙发，在我的什么位置？", + "answer_chinese": "客厅里有一个沙发，在我的前面。", + "evidence": [ + { + "description": "There is a couch full of clothes in front of me.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "A sofa is behind you.", + "B": "Two sofas are in front of you.", + "C": "There are two sofas in the living room, and both are positioned behind you.", + "D": "There is a sofa in the living room, in front of you.", + "E": "There are three sofas in the living room, arranged in front of you rather than just one." + }, + "answer_label": "D", + "answer_text": "There is a sofa in the living room, in front of you.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_1_T2", + "correct_answer": "There is a sofa in the living room, in front of you.", + "original_options": { + "A": "A sofa is behind you.", + "B": "Two sofas are in front of you.", + "C": "There are two sofas in the living room, and both are positioned behind you.", + "D": "There is a sofa in the living room, in front of you.", + "E": "There are three sofas in the living room, arranged in front of you rather than just one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 14, + 17 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_1_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sofas are in the living room, and where are they relative to me?", + "answer": "There is a sofa in the living room, behind you.", + "question_chinese": "客厅里有几个沙发，在我的什么位置？", + "answer_chinese": "客厅里有一个沙发，在我的后面。", + "evidence": [ + { + "description": "There is a couch full of clothes in front of me.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:01:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One sofa, in front of you.", + "B": "Two sofas, behind you.", + "C": "There is a sofa in the living room, behind you.", + "D": "There are two sofas in the living room, one behind you and the other in front of you.", + "E": "There are three sofas in the living room, all positioned in front of you as you face the room." + }, + "answer_label": "C", + "answer_text": "There is a sofa in the living room, behind you.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_1_T1", + "correct_answer": "There is a sofa in the living room, behind you.", + "original_options": { + "A": "One sofa, in front of you.", + "B": "Two sofas, behind you.", + "C": "There is a sofa in the living room, behind you.", + "D": "There are two sofas in the living room, one behind you and the other in front of you.", + "E": "There are three sofas in the living room, all positioned in front of you as you face the room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 18, + 19 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_11", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand at the foot of the bed in the bedroom, what is the straight-line distance between me and the mirror?", + "answer": "When you stand at the foot of the bed in the bedroom, the straight-line distance between you and the mirror is one meter.", + "question_chinese": "当我站在卧室的床尾，我和镜子的直线距离是多少？", + "answer_chinese": "当你站在卧室的床尾，你和镜子的直线距离是一米。", + "evidence": [ + { + "description": "The mirror is on the wall opposite the bed.", + "timestamps": { + "start": "00:07:10", + "end": "00:07:17" + } + } + ], + "question_timestamps": "00:08:29", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Half a meter.", + "B": "When you stand at the foot of the bed in the bedroom, the straight-line distance between you and the mirror is one meter.", + "C": "Three meters.", + "D": "Standing at the foot of the bed, the mirror is about two and a half meters away in a straight line.", + "E": "When you're at the foot of the bed, the mirror sits roughly five meters away straight ahead." + }, + "answer_label": "B", + "answer_text": "When you stand at the foot of the bed in the bedroom, the straight-line distance between you and the mirror is one meter.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_11", + "correct_answer": "When you stand at the foot of the bed in the bedroom, the straight-line distance between you and the mirror is one meter.", + "original_options": { + "A": "Half a meter.", + "B": "When you stand at the foot of the bed in the bedroom, the straight-line distance between you and the mirror is one meter.", + "C": "Three meters.", + "D": "Standing at the foot of the bed, the mirror is about two and a half meters away in a straight line.", + "E": "When you're at the foot of the bed, the mirror sits roughly five meters away straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 21, + 17 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_15", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I stand by the piano, can I reach the TV?", + "answer": "When you stand by the piano, you can't reach the TV.", + "question_chinese": "当我站在钢琴旁边，我能够到电视吗？", + "answer_chinese": "当你站在钢琴旁边，你不能够到电视。", + "evidence": [ + { + "description": "The TV is five meters away from the piano in a straight line.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:17" + } + } + ], + "question_timestamps": "00:11:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you stand by the piano, you can reach the TV.", + "B": "When you stand by the piano, you can't reach the TV." + }, + "answer_label": "B", + "answer_text": "When you stand by the piano, you can't reach the TV.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_15", + "correct_answer": "When you stand by the piano, you can't reach the TV.", + "original_options": { + "A": "When you stand by the piano, you can reach the TV.", + "B": "When you stand by the piano, you can't reach the TV." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_14", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I stand by the piano, can I reach the TV?", + "answer": "When you stand by the piano, you can't reach the TV.", + "question_chinese": "当我站在钢琴旁边，我能够到电视吗？", + "answer_chinese": "当你站在钢琴旁边，你不能够到电视。", + "evidence": [ + { + "description": "The TV is five meters away from the piano in a straight line.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:17" + } + } + ], + "question_timestamps": "00:12:17", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you stand by the piano, you can reach the TV.", + "B": "When you stand by the piano, you can't reach the TV." + }, + "answer_label": "B", + "answer_text": "When you stand by the piano, you can't reach the TV.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_14", + "correct_answer": "When you stand by the piano, you can't reach the TV.", + "original_options": { + "A": "When you stand by the piano, you can reach the TV.", + "B": "When you stand by the piano, you can't reach the TV." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_3_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I went from the small sofa to the bedroom, how many doors did I see?", + "answer": "From the small sofa to the bedroom, you first saw three doors.", + "question_chinese": "我从小沙发去到卧室看到了几扇门？", + "answer_chinese": "你从小沙发去到卧室先看到了三扇门。", + "evidence": [ + { + "description": "I see the hallway from the small sofa to the bedroom has three doors.", + "timestamps": { + "start": "00:06:38", + "end": "00:06:43" + } + } + ], + "question_timestamps": "00:06:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One door.", + "B": "Two doors.", + "C": "From the small sofa to the bedroom, you first saw four doors as you moved through the space.", + "D": "From the small sofa to the bedroom, you first saw five doors before continuing toward the bedroom.", + "E": "From the small sofa to the bedroom, you first saw three doors." + }, + "answer_label": "E", + "answer_text": "From the small sofa to the bedroom, you first saw three doors.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_3_T1", + "correct_answer": "From the small sofa to the bedroom, you first saw three doors.", + "original_options": { + "A": "One door.", + "B": "Two doors.", + "C": "From the small sofa to the bedroom, you first saw four doors as you moved through the space.", + "D": "From the small sofa to the bedroom, you first saw five doors before continuing toward the bedroom.", + "E": "From the small sofa to the bedroom, you first saw three doors." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 17 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_3_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I went from the small sofa to the bedroom, how many doors did I see?", + "answer": "From the small sofa to the bedroom, you saw a total of four doors.", + "question_chinese": "我从小沙发去到卧室看到了几扇门？", + "answer_chinese": "你从小沙发去到卧室一共看到了四扇门。", + "evidence": [ + { + "description": "The hallway from the small sofa to the bedroom has four doors.", + "timestamps": { + "start": "00:06:38", + "end": "00:06:49" + } + } + ], + "question_timestamps": "00:23:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "From the small sofa to the bedroom, you saw a total of four doors.", + "B": "Two doors total.", + "C": "Three doors total.", + "D": "From the small sofa to the bedroom, you saw a total of five doors along the way.", + "E": "From the small sofa to the bedroom, you saw a total of six doors as you traveled through." + }, + "answer_label": "A", + "answer_text": "From the small sofa to the bedroom, you saw a total of four doors.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_3_T2", + "correct_answer": "From the small sofa to the bedroom, you saw a total of four doors.", + "original_options": { + "A": "From the small sofa to the bedroom, you saw a total of four doors.", + "B": "Two doors total.", + "C": "Three doors total.", + "D": "From the small sofa to the bedroom, you saw a total of five doors along the way.", + "E": "From the small sofa to the bedroom, you saw a total of six doors as you traveled through." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 18 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_5", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From the small sofa to the bedroom, how many doors did I see in total?", + "answer": "You saw a total of four doors when you went from the small sofa to the bedroom.", + "question_chinese": "我从小沙发去到卧室要一共看到了几扇门？", + "answer_chinese": "你从小沙发去到卧室要一共看到了四扇门。", + "evidence": [ + { + "description": "The hallway from the small sofa to the bedroom has four doors.", + "timestamps": { + "start": "00:06:38", + "end": "00:06:49" + } + } + ], + "question_timestamps": "00:07:49", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three doors total.", + "B": "Five doors total.", + "C": "You saw a total of two doors when you went from the small sofa to the bedroom.", + "D": "You saw a total of six doors when you went from the small sofa to the bedroom.", + "E": "You saw a total of four doors when you went from the small sofa to the bedroom." + }, + "answer_label": "E", + "answer_text": "You saw a total of four doors when you went from the small sofa to the bedroom.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_5", + "correct_answer": "You saw a total of four doors when you went from the small sofa to the bedroom.", + "original_options": { + "A": "Three doors total.", + "B": "Five doors total.", + "C": "You saw a total of two doors when you went from the small sofa to the bedroom.", + "D": "You saw a total of six doors when you went from the small sofa to the bedroom.", + "E": "You saw a total of four doors when you went from the small sofa to the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_6", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From the small sofa to the bedroom, how many doors did I see in total?", + "answer": "You saw a total of four doors when you went from the small sofa to the bedroom.", + "question_chinese": "我从小沙发去到卧室要一共看到了几扇门？", + "answer_chinese": "你从小沙发去到卧室要一共看到了四扇门。", + "evidence": [ + { + "description": "The hallway from the small sofa to the bedroom has four doors.", + "timestamps": { + "start": "00:06:38", + "end": "00:06:49" + } + } + ], + "question_timestamps": "00:23:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You saw three doors total.", + "B": "You saw five doors total.", + "C": "You saw a total of two doors when you went from the small sofa to the bedroom.", + "D": "You saw a total of four doors when you went from the small sofa to the bedroom.", + "E": "You saw a total of six doors on your way from the small sofa to the bedroom." + }, + "answer_label": "D", + "answer_text": "You saw a total of four doors when you went from the small sofa to the bedroom.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_6", + "correct_answer": "You saw a total of four doors when you went from the small sofa to the bedroom.", + "original_options": { + "A": "You saw three doors total.", + "B": "You saw five doors total.", + "C": "You saw a total of two doors when you went from the small sofa to the bedroom.", + "D": "You saw a total of four doors when you went from the small sofa to the bedroom.", + "E": "You saw a total of six doors on your way from the small sofa to the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_6_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the straight-line distance between me and the bedroom mirror?", + "answer": "When you’re standing at the foot of the bed in the bedroom, your straight-line distance to the mirror is 1 meter.", + "question_chinese": "我和卧室镜子的直线距离是多少？", + "answer_chinese": "当你站在卧室的床尾，你和镜子的直线距离是一米。", + "evidence": [ + { + "description": "The mirror is on the wall opposite the bed.", + "timestamps": { + "start": "00:07:10", + "end": "00:07:17" + } + } + ], + "question_timestamps": "00:07:22", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It is 2 meters away.", + "B": "It is 3 meters away.", + "C": "When you’re standing at the foot of the bed in the bedroom, your straight-line distance to the mirror is 1 meter.", + "D": "When you’re standing at the foot of the bed in the bedroom, your straight-line distance to the mirror is 4 meters.", + "E": "When you’re standing at the foot of the bed in the bedroom, your straight-line distance to the mirror is 0 meters." + }, + "answer_label": "C", + "answer_text": "When you’re standing at the foot of the bed in the bedroom, your straight-line distance to the mirror is 1 meter.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_6_T1", + "correct_answer": "When you’re standing at the foot of the bed in the bedroom, your straight-line distance to the mirror is 1 meter.", + "original_options": { + "A": "It is 2 meters away.", + "B": "It is 3 meters away.", + "C": "When you’re standing at the foot of the bed in the bedroom, your straight-line distance to the mirror is 1 meter.", + "D": "When you’re standing at the foot of the bed in the bedroom, your straight-line distance to the mirror is 4 meters.", + "E": "When you’re standing at the foot of the bed in the bedroom, your straight-line distance to the mirror is 0 meters." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 22, + 22 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_6_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the straight-line distance between me and the bedroom mirror?", + "answer": "When you’re standing next to the cabinet, your straight-line distance to the mirror is 2 meters.", + "question_chinese": "我和卧室镜子的直线距离是多少？", + "answer_chinese": "当你站在柜子旁边，你和镜子的直线距离是两米。", + "evidence": [ + { + "description": "The mirror is on the wall opposite the bed.", + "timestamps": { + "start": "00:07:10", + "end": "00:07:17" + } + } + ], + "question_timestamps": "00:07:49", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It is 1 meter away.", + "B": "It is 3 meters away.", + "C": "When you’re standing next to the cabinet, your straight-line distance to the mirror is 0 meters.", + "D": "When you’re standing next to the cabinet, your straight-line distance to the mirror is 2 meters.", + "E": "When you’re standing next to the cabinet, your straight-line distance to the mirror is 4 meters." + }, + "answer_label": "D", + "answer_text": "When you’re standing next to the cabinet, your straight-line distance to the mirror is 2 meters.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_6_T2", + "correct_answer": "When you’re standing next to the cabinet, your straight-line distance to the mirror is 2 meters.", + "original_options": { + "A": "It is 1 meter away.", + "B": "It is 3 meters away.", + "C": "When you’re standing next to the cabinet, your straight-line distance to the mirror is 0 meters.", + "D": "When you’re standing next to the cabinet, your straight-line distance to the mirror is 2 meters.", + "E": "When you’re standing next to the cabinet, your straight-line distance to the mirror is 4 meters." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_9_T1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I walk to get to the bathroom?", + "answer": "Just turn right— the bathroom is by your right hand.", + "question_chinese": "我想去厕所应该怎么走？", + "answer_chinese": "你向右转即可，你右手边就是厕所。", + "evidence": [ + { + "description": "The toilet is in the first room down the hallway behind the fridge.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:01:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left to reach it.", + "B": "Go straight ahead first.", + "C": "Just turn right— the bathroom is by your right hand.", + "D": "Walk forward a few steps, then turn left and enter the first doorway on your right.", + "E": "Head toward your front-left, turn right, and go into the first room on your left." + }, + "answer_label": "C", + "answer_text": "Just turn right— the bathroom is by your right hand.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_9_T1", + "correct_answer": "Just turn right— the bathroom is by your right hand.", + "original_options": { + "A": "Turn left to reach it.", + "B": "Go straight ahead first.", + "C": "Just turn right— the bathroom is by your right hand.", + "D": "Walk forward a few steps, then turn left and enter the first doorway on your right.", + "E": "Head toward your front-left, turn right, and go into the first room on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 15 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_9_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I walk to get to the bathroom?", + "answer": "Walk toward your front-right, turn left, then go into the first room on your right.", + "question_chinese": "我想去厕所应该怎么走？", + "answer_chinese": "往右前方走，左转，进去右手边第一个房间。", + "evidence": [ + { + "description": "The toilet is in the first room down the hallway behind the fridge.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:15:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Enter the first room on your left.", + "B": "Go into the room straight ahead.", + "C": "Walk toward your front-right, turn left, then go into the first room on your right.", + "D": "Walk toward your front-right, turn left, then go into the second room on your right.", + "E": "Walk toward your front-right, turn right, then enter the first room on your right." + }, + "answer_label": "C", + "answer_text": "Walk toward your front-right, turn left, then go into the first room on your right.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_9_T2", + "correct_answer": "Walk toward your front-right, turn left, then go into the first room on your right.", + "original_options": { + "A": "Enter the first room on your left.", + "B": "Go into the room straight ahead.", + "C": "Walk toward your front-right, turn left, then go into the first room on your right.", + "D": "Walk toward your front-right, turn left, then go into the second room on your right.", + "E": "Walk toward your front-right, turn right, then enter the first room on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 14 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_4_T1", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want eggs, where should I go to get them?", + "answer": "If you want eggs, you should get them from the fridge—the fridge is to your right.", + "question_chinese": "我想要鸡蛋应该去哪里拿？", + "answer_chinese": "你想要鸡蛋应该去冰箱拿，冰箱在我的右边。", + "evidence": [ + { + "description": "There are two boxes of eggs on the top shelf of the refrigerator.", + "timestamps": { + "start": "00:13:45", + "end": "00:13:48" + } + } + ], + "question_timestamps": "00:13:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Get them from the pantry.", + "B": "Get them from the countertop.", + "C": "If you want eggs, you should get them from the fridge—the fridge is to your left.", + "D": "If you want eggs, you should get them from the fridge—the fridge is to your right.", + "E": "If you want eggs, you should get them from the fridge—the fridge is behind you." + }, + "answer_label": "D", + "answer_text": "If you want eggs, you should get them from the fridge—the fridge is to your right.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_4_T1", + "correct_answer": "If you want eggs, you should get them from the fridge—the fridge is to your right.", + "original_options": { + "A": "Get them from the pantry.", + "B": "Get them from the countertop.", + "C": "If you want eggs, you should get them from the fridge—the fridge is to your left.", + "D": "If you want eggs, you should get them from the fridge—the fridge is to your right.", + "E": "If you want eggs, you should get them from the fridge—the fridge is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_8", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I go to get eggs?", + "answer": "You should get the eggs from the fridge if you want them.", + "question_chinese": "我想要鸡蛋应该去哪里拿？", + "answer_chinese": "你想要鸡蛋应该去冰箱拿。", + "evidence": [ + { + "description": "There are two boxes of eggs on the top shelf of the refrigerator.", + "timestamps": { + "start": "00:13:45", + "end": "00:13:48" + } + } + ], + "question_timestamps": "00:14:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should get the eggs from the fridge if you want them.", + "B": "Get them from the pantry.", + "C": "Get them from the freezer.", + "D": "You should get the eggs from the countertop if you want them instead of looking in the fridge.", + "E": "You should get the eggs from the vegetable drawer if you want them, not from the fridge." + }, + "answer_label": "A", + "answer_text": "You should get the eggs from the fridge if you want them.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_8", + "correct_answer": "You should get the eggs from the fridge if you want them.", + "original_options": { + "A": "You should get the eggs from the fridge if you want them.", + "B": "Get them from the pantry.", + "C": "Get them from the freezer.", + "D": "You should get the eggs from the countertop if you want them instead of looking in the fridge.", + "E": "You should get the eggs from the vegetable drawer if you want them, not from the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_7", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I go to get eggs?", + "answer": "You should get the eggs from the fridge if you want them.", + "question_chinese": "我想要鸡蛋应该去哪里拿？", + "answer_chinese": "你想要鸡蛋应该去冰箱拿。", + "evidence": [ + { + "description": "There are two boxes of eggs on the top shelf of the refrigerator.", + "timestamps": { + "start": "00:13:45", + "end": "00:13:48" + } + } + ], + "question_timestamps": "00:26:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Get them from the pantry.", + "B": "Get them from the freezer.", + "C": "You should get the eggs from a bowl on the kitchen counter if you want them.", + "D": "You should get the eggs from the fridge if you want them.", + "E": "You should get the eggs from the basket on the table by the door if you want them." + }, + "answer_label": "D", + "answer_text": "You should get the eggs from the fridge if you want them.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_7", + "correct_answer": "You should get the eggs from the fridge if you want them.", + "original_options": { + "A": "Get them from the pantry.", + "B": "Get them from the freezer.", + "C": "You should get the eggs from a bowl on the kitchen counter if you want them.", + "D": "You should get the eggs from the fridge if you want them.", + "E": "You should get the eggs from the basket on the table by the door if you want them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_4_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want eggs, where should I go to get them?", + "answer": "If you want eggs, you should get them from the fridge—the fridge is behind you.", + "question_chinese": "我想要鸡蛋应该去哪里拿？", + "answer_chinese": "你想要鸡蛋应该去冰箱拿，冰箱在我的后面。", + "evidence": [ + { + "description": "There are two boxes of eggs on the top shelf of the refrigerator.", + "timestamps": { + "start": "00:13:45", + "end": "00:13:48" + } + } + ], + "question_timestamps": "00:27:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use the fridge in front.", + "B": "Use the fridge to your left.", + "C": "Get the eggs from the fridge; it should be off to your right as you face forward.", + "D": "Head to the fridge for eggs; you’ll find it in front of you rather than behind you.", + "E": "If you want eggs, you should get them from the fridge—the fridge is behind you." + }, + "answer_label": "E", + "answer_text": "If you want eggs, you should get them from the fridge—the fridge is behind you.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_4_T2", + "correct_answer": "If you want eggs, you should get them from the fridge—the fridge is behind you.", + "original_options": { + "A": "Use the fridge in front.", + "B": "Use the fridge to your left.", + "C": "Get the eggs from the fridge; it should be off to your right as you face forward.", + "D": "Head to the fridge for eggs; you’ll find it in front of you rather than behind you.", + "E": "If you want eggs, you should get them from the fridge—the fridge is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 18 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_19", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cabinets are there in my bedroom?", + "answer": "Your bedroom has three cabinets.", + "question_chinese": "我的卧室有几个柜子？", + "answer_chinese": "你的卧室有三个柜子。", + "evidence": [ + { + "description": "There are two wardrobes in the bedroom.", + "timestamps": { + "start": "00:06:57", + "end": "00:07:04" + } + }, + { + "description": "There is a dressing table opposite the bed.", + "timestamps": { + "start": "00:07:05", + "end": "00:07:15" + } + } + ], + "question_timestamps": "00:09:11", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two cabinets.", + "B": "Four cabinets.", + "C": "Your bedroom has five cabinets lined up along the wall near the bed.", + "D": "Your bedroom has one cabinet only, not three, in the bedroom area by the bed.", + "E": "Your bedroom has three cabinets." + }, + "answer_label": "E", + "answer_text": "Your bedroom has three cabinets.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_19", + "correct_answer": "Your bedroom has three cabinets.", + "original_options": { + "A": "Two cabinets.", + "B": "Four cabinets.", + "C": "Your bedroom has five cabinets lined up along the wall near the bed.", + "D": "Your bedroom has one cabinet only, not three, in the bedroom area by the bed.", + "E": "Your bedroom has three cabinets." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_20", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cabinets are there in my bedroom?", + "answer": "Your bedroom has three cabinets.", + "question_chinese": "我的卧室有几个柜子？", + "answer_chinese": "你的卧室有三个柜子。", + "evidence": [ + { + "description": "There is a dressing table opposite the bed.", + "timestamps": { + "start": "00:07:15", + "end": "00:07:50" + } + }, + { + "description": "There are two wardrobes in the bedroom.", + "timestamps": { + "start": "00:06:57", + "end": "00:07:04" + } + } + ], + "question_timestamps": "00:25:09", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two cabinets.", + "B": "Four cabinets.", + "C": "Your bedroom has three cabinets.", + "D": "Your bedroom has five cabinets in total, not three, across the whole wall.", + "E": "Your bedroom has one cabinet in the room, not three cabinets, near the bed wall." + }, + "answer_label": "C", + "answer_text": "Your bedroom has three cabinets.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_20", + "correct_answer": "Your bedroom has three cabinets.", + "original_options": { + "A": "Two cabinets.", + "B": "Four cabinets.", + "C": "Your bedroom has three cabinets.", + "D": "Your bedroom has five cabinets in total, not three, across the whole wall.", + "E": "Your bedroom has one cabinet in the room, not three cabinets, near the bed wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_10_T1", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cabinet did I see in the bedroom?", + "answer": "You saw two cabinets in the bedroom.", + "question_chinese": "我看见卧室有几个柜子？", + "answer_chinese": "你看见卧室有两个柜子。", + "evidence": [ + { + "description": "There are two wardrobes in the bedroom.", + "timestamps": { + "start": "00:06:57", + "end": "00:07:03" + } + } + ], + "question_timestamps": "00:07:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two cabinets in the bedroom.", + "B": "One cabinet.", + "C": "Three cabinets.", + "D": "You saw four cabinets in the bedroom during the walkthrough, not just two.", + "E": "You saw zero cabinets in the bedroom when you looked around the room." + }, + "answer_label": "A", + "answer_text": "You saw two cabinets in the bedroom.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_10_T1", + "correct_answer": "You saw two cabinets in the bedroom.", + "original_options": { + "A": "You saw two cabinets in the bedroom.", + "B": "One cabinet.", + "C": "Three cabinets.", + "D": "You saw four cabinets in the bedroom during the walkthrough, not just two.", + "E": "You saw zero cabinets in the bedroom when you looked around the room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_10_T2", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cabinet did I see in the bedroom?", + "answer": "You saw three cabinets in the bedroom.", + "question_chinese": "我看见卧室有几个柜子？", + "answer_chinese": "你看见卧室有三个柜子。", + "evidence": [ + { + "description": "There is a dressing table opposite the bed.", + "timestamps": { + "start": "00:07:15", + "end": "00:07:50" + } + }, + { + "description": "There are two wardrobes in the bedroom.", + "timestamps": { + "start": "00:06:57", + "end": "00:07:04" + } + } + ], + "question_timestamps": "00:25:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two cabinets.", + "B": "You saw three cabinets in the bedroom.", + "C": "Four cabinets.", + "D": "You saw five cabinets in the bedroom while you were there, not three.", + "E": "You saw one cabinet in the bedroom, not three cabinets in total." + }, + "answer_label": "B", + "answer_text": "You saw three cabinets in the bedroom.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_10_T2", + "correct_answer": "You saw three cabinets in the bedroom.", + "original_options": { + "A": "Two cabinets.", + "B": "You saw three cabinets in the bedroom.", + "C": "Four cabinets.", + "D": "You saw five cabinets in the bedroom while you were there, not three.", + "E": "You saw one cabinet in the bedroom, not three cabinets in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_17", + "qid_num": 27, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I stood in front of the refrigerator and wanted to go to the toilet, how should I go?", + "answer": "Go straight ahead to the right, turn left, and enter the first room on the right.", + "question_chinese": "我站在冰箱前，想去上厕所应该怎么走？", + "answer_chinese": "往右前方走，左转，进去右手边第一个房间。", + "evidence": [ + { + "description": "The toilet is in the first room down the hallway behind the fridge.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:04:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go straight ahead to the right, turn left, and enter the first room on the right.", + "B": "Turn right, then enter the second room.", + "C": "Go left first, then turn left inside.", + "D": "Go straight ahead to the right, turn right, and enter the first room on the right.", + "E": "Go straight ahead to the right, turn left, then enter the second room on the right instead." + }, + "answer_label": "A", + "answer_text": "Go straight ahead to the right, turn left, and enter the first room on the right.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_17", + "correct_answer": "Go straight ahead to the right, turn left, and enter the first room on the right.", + "original_options": { + "A": "Go straight ahead to the right, turn left, and enter the first room on the right.", + "B": "Turn right, then enter the second room.", + "C": "Go left first, then turn left inside.", + "D": "Go straight ahead to the right, turn right, and enter the first room on the right.", + "E": "Go straight ahead to the right, turn left, then enter the second room on the right instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_18", + "qid_num": 28, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I stood in front of the refrigerator and wanted to go to the toilet, how should I go?", + "answer": "Go straight ahead to the right, turn left, and enter the right side.", + "question_chinese": "我站在冰箱前，想去上厕所应该怎么走？", + "answer_chinese": "往右前方走，左转，进去右手边第一个房间。", + "evidence": [ + { + "description": "The toilet is in the first room down the hallway behind the fridge.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:15:59", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go straight left, then turn right.", + "B": "Turn right and enter the left side.", + "C": "Go straight ahead to the right, turn right, and enter the right side instead of turning left.", + "D": "Go straight ahead to the right, turn left, and enter the right side.", + "E": "Go straight ahead to the right, turn left, but enter the left side rather than the right side." + }, + "answer_label": "D", + "answer_text": "Go straight ahead to the right, turn left, and enter the right side.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_18", + "correct_answer": "Go straight ahead to the right, turn left, and enter the right side.", + "original_options": { + "A": "Go straight left, then turn right.", + "B": "Turn right and enter the left side.", + "C": "Go straight ahead to the right, turn right, and enter the right side instead of turning left.", + "D": "Go straight ahead to the right, turn left, and enter the right side.", + "E": "Go straight ahead to the right, turn left, but enter the left side rather than the right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 18 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_12", + "qid_num": 29, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I'm standing next to the small sofa, how should I go to get the guitar?", + "answer": "Go straight, then pass the stairs and go to the door on the left at the end of the corridor to enter.", + "question_chinese": "我站在小沙发旁边，要去拿吉他应该怎么走？", + "answer_chinese": "直走然后经过楼梯去到走廊尽头左边的门进去。", + "evidence": [ + { + "description": "The guitar is placed by the wall in the bedroom.", + "timestamps": { + "start": "00:07:27", + "end": "00:07:29" + } + }, + { + "description": "The hallway from the small sofa to the bedroom has four doors.", + "timestamps": { + "start": "00:06:43", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:13:57", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight; take the door on the right.", + "B": "Turn left and enter the first door.", + "C": "Go straight, pass the stairs, then take the door on the right at the corridor’s end to enter.", + "D": "Go straight, then pass the stairs and go to the door on the left at the end of the corridor to enter.", + "E": "Go straight to the stairs, turn right before passing them, and enter the door at the end." + }, + "answer_label": "D", + "answer_text": "Go straight, then pass the stairs and go to the door on the left at the end of the corridor to enter.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_12", + "correct_answer": "Go straight, then pass the stairs and go to the door on the left at the end of the corridor to enter.", + "original_options": { + "A": "Go straight; take the door on the right.", + "B": "Turn left and enter the first door.", + "C": "Go straight, pass the stairs, then take the door on the right at the corridor’s end to enter.", + "D": "Go straight, then pass the stairs and go to the door on the left at the end of the corridor to enter.", + "E": "Go straight to the stairs, turn right before passing them, and enter the door at the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 17 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_13", + "qid_num": 30, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I'm standing next to the small sofa, how should I go to get the guitar?", + "answer": "Go straight, then pass the stairs and go to the door on the left at the end of the corridor to enter.", + "question_chinese": "我站在小沙发旁边，要去拿吉他应该怎么走？", + "answer_chinese": "直走然后经过楼梯去到走廊尽头左边的门进去。", + "evidence": [ + { + "description": "The hallway from the small sofa to the bedroom has four doors.", + "timestamps": { + "start": "00:06:43", + "end": "00:06:50" + } + }, + { + "description": "The guitar is placed by the wall in the bedroom.", + "timestamps": { + "start": "00:07:27", + "end": "00:07:29" + } + } + ], + "question_timestamps": "00:25:36", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right into the nearest door.", + "B": "Go left past the stairs.", + "C": "Go straight, stop before the stairs, then enter the first door on your right in the corridor.", + "D": "Walk straight down the corridor, pass the stairs, then take the door on the right at the very end.", + "E": "Go straight, then pass the stairs and go to the door on the left at the end of the corridor to enter." + }, + "answer_label": "E", + "answer_text": "Go straight, then pass the stairs and go to the door on the left at the end of the corridor to enter.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_13", + "correct_answer": "Go straight, then pass the stairs and go to the door on the left at the end of the corridor to enter.", + "original_options": { + "A": "Turn right into the nearest door.", + "B": "Go left past the stairs.", + "C": "Go straight, stop before the stairs, then enter the first door on your right in the corridor.", + "D": "Walk straight down the corridor, pass the stairs, then take the door on the right at the very end.", + "E": "Go straight, then pass the stairs and go to the door on the left at the end of the corridor to enter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 19 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_8_T1", + "qid_num": 31, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the TV?", + "answer": "When you’re standing next to the piano, you can’t reach the TV.", + "question_chinese": "我能够到电视吗？", + "answer_chinese": "当你站在钢琴旁边，你不能够到电视。", + "evidence": [ + { + "description": "The TV is five meters away from the piano in a straight line.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:16" + } + } + ], + "question_timestamps": "00:11:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you’re standing next to the piano, you can’t reach the TV.", + "B": "When you’re standing next to the piano, you can reach the TV." + }, + "answer_label": "A", + "answer_text": "When you’re standing next to the piano, you can’t reach the TV.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_8_T1", + "correct_answer": "When you’re standing next to the piano, you can’t reach the TV.", + "original_options": { + "A": "When you’re standing next to the piano, you can’t reach the TV.", + "B": "When you’re standing next to the piano, you can reach the TV." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_8_T2", + "qid_num": 32, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the TV?", + "answer": "You can reach the TV. It’s by your right hand—just reach out and you’ll be able to touch it.", + "question_chinese": "我能够到电视吗？", + "answer_chinese": "你可以够到电视，它在你的右手边，你伸手就能够到。", + "evidence": [ + { + "description": "The TV is on your right, and you can reach it with your hand.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:20" + } + } + ], + "question_timestamps": "00:11:22", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the TV; it’s too far to touch from where you’re standing.", + "B": "You can reach the TV. It’s by your right hand—just reach out and you’ll be able to touch it." + }, + "answer_label": "B", + "answer_text": "You can reach the TV. It’s by your right hand—just reach out and you’ll be able to touch it.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_8_T2", + "correct_answer": "You can reach the TV. It’s by your right hand—just reach out and you’ll be able to touch it.", + "original_options": { + "A": "You can’t reach the TV; it’s too far to touch from where you’re standing.", + "B": "You can reach the TV. It’s by your right hand—just reach out and you’ll be able to touch it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_7_T1", + "qid_num": 33, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I walk to go get the guitar?", + "answer": "You were standing next to the small sofa, walked straight ahead, passed the stairs, and went into the door at the end of the hallway on the left.", + "question_chinese": "我要去拿吉他应该怎么走？", + "answer_chinese": "我站在小沙发旁边，直走然后经过楼梯去到走廊尽头左边的门进去。", + "evidence": [ + { + "description": "The guitar is placed by the wall in the bedroom.", + "timestamps": { + "start": "00:07:27", + "end": "00:07:29" + } + }, + { + "description": "The hallway from the small sofa to the bedroom has four doors.", + "timestamps": { + "start": "00:06:43", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:07:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Enter the door on the right.", + "B": "Turn left at the stairs.", + "C": "You were standing next to the small sofa, walked straight ahead, passed the stairs, and went into the door at the end of the hallway on the left.", + "D": "You started by the small sofa, went past the stairs, and entered the first door on your right in the hallway.", + "E": "From the small sofa, you walked straight, stopped at the stairs, then turned right into the nearby doorway." + }, + "answer_label": "C", + "answer_text": "You were standing next to the small sofa, walked straight ahead, passed the stairs, and went into the door at the end of the hallway on the left.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_7_T1", + "correct_answer": "You were standing next to the small sofa, walked straight ahead, passed the stairs, and went into the door at the end of the hallway on the left.", + "original_options": { + "A": "Enter the door on the right.", + "B": "Turn left at the stairs.", + "C": "You were standing next to the small sofa, walked straight ahead, passed the stairs, and went into the door at the end of the hallway on the left.", + "D": "You started by the small sofa, went past the stairs, and entered the first door on your right in the hallway.", + "E": "From the small sofa, you walked straight, stopped at the stairs, then turned right into the nearby doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 18 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_7_T2", + "qid_num": 34, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I walk to go get the guitar?", + "answer": "Turn right and take one step forward, and you can reach the guitar.", + "question_chinese": "我要去拿吉他应该怎么走？", + "answer_chinese": "你右转向前走一步就可以拿到吉他。", + "evidence": [ + { + "description": "The guitar is placed by the wall in the bedroom.", + "timestamps": { + "start": "00:07:27", + "end": "00:07:32" + } + } + ], + "question_timestamps": "00:07:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left one step to reach it.", + "B": "Go straight two steps forward.", + "C": "Turn right, take two steps forward, and you’ll be able to grab the guitar from there.", + "D": "Turn left, then walk forward a couple of steps until you’re by the guitar and can pick it up.", + "E": "Turn right and take one step forward, and you can reach the guitar." + }, + "answer_label": "E", + "answer_text": "Turn right and take one step forward, and you can reach the guitar.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_7_T2", + "correct_answer": "Turn right and take one step forward, and you can reach the guitar.", + "original_options": { + "A": "Turn left one step to reach it.", + "B": "Go straight two steps forward.", + "C": "Turn right, take two steps forward, and you’ll be able to grab the guitar from there.", + "D": "Turn left, then walk forward a couple of steps until you’re by the guitar and can pick it up.", + "E": "Turn right and take one step forward, and you can reach the guitar." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 17, + 20 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_9", + "qid_num": 35, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Facing the microwave, is the kitchen paper towel on my left/right?", + "answer": "Facing the microwave, the kitchen paper is on your left.", + "question_chinese": "面对微波炉，厨房用纸在我的左/右？", + "answer_chinese": "面对微波炉，厨房用纸在你的左边。", + "evidence": [ + { + "description": "Kitchen paper is placed on the left side of the microwave.", + "timestamps": { + "start": "00:13:49", + "end": "00:13:53" + } + } + ], + "question_timestamps": "00:15:29", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Facing the microwave, the kitchen paper is on your right.", + "B": "Facing the microwave, the kitchen paper is on your left." + }, + "answer_label": "B", + "answer_text": "Facing the microwave, the kitchen paper is on your left.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_9", + "correct_answer": "Facing the microwave, the kitchen paper is on your left.", + "original_options": { + "A": "Facing the microwave, the kitchen paper is on your right.", + "B": "Facing the microwave, the kitchen paper is on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "15f1a422-327d-4dab-b14d-7003544c5817", + "video_length": "00:30:00", + "qid": "15f1a422-327d-4dab-b14d-7003544c5817_10", + "qid_num": 36, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Facing the microwave, is the kitchen paper towel on my left/right?", + "answer": "Facing the microwave, the kitchen paper is on your left.", + "question_chinese": "面对微波炉，厨房用纸在我的左/右？", + "answer_chinese": "面对微波炉，厨房用纸在你的左边。", + "evidence": [ + { + "description": "Kitchen paper is placed on the left side of the microwave.", + "timestamps": { + "start": "00:13:49", + "end": "00:13:53" + } + } + ], + "question_timestamps": "00:15:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Facing the microwave, the kitchen paper is on your right.", + "B": "Facing the microwave, the kitchen paper is on your left." + }, + "answer_label": "B", + "answer_text": "Facing the microwave, the kitchen paper is on your left.", + "q_id": "15f1a422-327d-4dab-b14d-7003544c5817_10", + "correct_answer": "Facing the microwave, the kitchen paper is on your left.", + "original_options": { + "A": "Facing the microwave, the kitchen paper is on your right.", + "B": "Facing the microwave, the kitchen paper is on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + } +] diff --git a/QAs/18a3840b-7463-43c4-9aa9-b1d8e486fa84-mcq.json b/QAs/18a3840b-7463-43c4-9aa9-b1d8e486fa84-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ccdceb286d17a64d7e96c4a457bb736e1abe0d5b --- /dev/null +++ b/QAs/18a3840b-7463-43c4-9aa9-b1d8e486fa84-mcq.json @@ -0,0 +1,1466 @@ +[ + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_6", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When you last saw the rear compartment door, was it open or closed?", + "answer": "The rear cargo door was closed.", + "question_chinese": "上次看见后车厢门时，它是开着的还是关着的？", + "answer_chinese": "后车厢门是关闭的。", + "evidence": [ + { + "description": "The rear cargo door is closed.", + "timestamps": { + "start": "00:12:00", + "end": "00:12:30" + } + } + ], + "question_timestamps": "00:16:55", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The rear cargo door was open.", + "B": "The rear cargo door was closed." + }, + "answer_label": "B", + "answer_text": "The rear cargo door was closed.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_6", + "correct_answer": "The rear cargo door was closed.", + "original_options": { + "A": "The rear cargo door was open.", + "B": "The rear cargo door was closed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_5", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When you last saw the rear compartment door, was it open or closed?", + "answer": "The rear cargo door was closed.", + "question_chinese": "上次看见后车厢门时，它是开着的还是关着的？", + "answer_chinese": "后车厢门是关闭的。", + "evidence": [ + { + "description": "The rear cargo door is closed.", + "timestamps": { + "start": "00:12:00", + "end": "00:12:30" + } + } + ], + "question_timestamps": "00:33:31", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The rear cargo door was closed.", + "B": "The rear cargo door was open." + }, + "answer_label": "A", + "answer_text": "The rear cargo door was closed.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_5", + "correct_answer": "The rear cargo door was closed.", + "original_options": { + "A": "The rear cargo door was closed.", + "B": "The rear cargo door was open." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_9", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What has been the motion trajectory of the miter saw so far?", + "answer": "It was lifted, carried out of the front door to the right, down the steps by the front door, and placed on a folding workbench.", + "question_chinese": "到目前为止斜切锯的运动轨迹是怎样的？", + "answer_chinese": "它被抬起，向右被抬出家门，沿家门口的台阶向下，被放在折叠式工作凳上", + "evidence": [ + { + "description": "It was lifted, carried out of the front door to the right, down the steps by the front door, and placed on a folding workbench.", + "timestamps": { + "start": "00:02:51", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:16:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was lifted, carried out of the front door to the right, down the steps by the front door, and placed on a folding workbench.", + "B": "It stayed on the workbench.", + "C": "It was carried to the left.", + "D": "It was lifted, taken out the front door to the left, walked down the steps, and set on the ground.", + "E": "It was picked up, carried out the front door to the right, but placed on a table instead of a folding workbench." + }, + "answer_label": "A", + "answer_text": "It was lifted, carried out of the front door to the right, down the steps by the front door, and placed on a folding workbench.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_9", + "correct_answer": "It was lifted, carried out of the front door to the right, down the steps by the front door, and placed on a folding workbench.", + "original_options": { + "A": "It was lifted, carried out of the front door to the right, down the steps by the front door, and placed on a folding workbench.", + "B": "It stayed on the workbench.", + "C": "It was carried to the left.", + "D": "It was lifted, taken out the front door to the left, walked down the steps, and set on the ground.", + "E": "It was picked up, carried out the front door to the right, but placed on a table instead of a folding workbench." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 22 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_10", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What has been the motion trajectory of the miter saw so far?", + "answer": "It was lifted, carried out of the house to the right, down the steps near the front door, and placed on a folding work stool.", + "question_chinese": "到目前为止斜切锯的运动轨迹是怎样的？", + "answer_chinese": "它被抬起，向右被抬出家门，沿家门口的台阶向下，被放在折叠式工作凳上。", + "evidence": [ + { + "description": "It was lifted, carried out of the front door to the right, down the steps by the front door, and placed on a folding workbench.", + "timestamps": { + "start": "00:02:51", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:32:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It went out to the left.", + "B": "It was lifted, carried out of the house to the right, down the steps near the front door, and placed on a folding work stool.", + "C": "It stayed inside the house.", + "D": "It was lifted, carried out of the house to the left, down the steps, and set on the ground.", + "E": "It was lifted, carried out to the right, taken down a ramp by the driveway, and set on a workbench." + }, + "answer_label": "B", + "answer_text": "It was lifted, carried out of the house to the right, down the steps near the front door, and placed on a folding work stool.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_10", + "correct_answer": "It was lifted, carried out of the house to the right, down the steps near the front door, and placed on a folding work stool.", + "original_options": { + "A": "It went out to the left.", + "B": "It was lifted, carried out of the house to the right, down the steps near the front door, and placed on a folding work stool.", + "C": "It stayed inside the house.", + "D": "It was lifted, carried out of the house to the left, down the steps, and set on the ground.", + "E": "It was lifted, carried out to the right, taken down a ramp by the driveway, and set on a workbench." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 20 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_4_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I find the wire that was coiled into loops?", + "answer": "In the kitchen, you found the coiled red electrical wire below you under the step ladder.", + "question_chinese": "我从哪里找到了缠绕成圈状的电线？", + "answer_chinese": "你在厨房里的人字梯下方找到了缠绕成圈的红电线", + "evidence": [ + { + "description": "I found the red coiled wire under the ladder in the kitchen at home.", + "timestamps": { + "start": "00:04:40", + "end": "00:05:01" + } + } + ], + "question_timestamps": "00:05:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "In the kitchen, you found the coiled red electrical wire to your left under the step ladder.", + "D": "In the kitchen, you found the coiled red electrical wire to your right under the step ladder.", + "E": "In the kitchen, you found the coiled red electrical wire below you under the step ladder." + }, + "answer_label": "E", + "answer_text": "In the kitchen, you found the coiled red electrical wire below you under the step ladder.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_4_T1", + "correct_answer": "In the kitchen, you found the coiled red electrical wire below you under the step ladder.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "In the kitchen, you found the coiled red electrical wire to your left under the step ladder.", + "D": "In the kitchen, you found the coiled red electrical wire to your right under the step ladder.", + "E": "In the kitchen, you found the coiled red electrical wire below you under the step ladder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 17, + 17 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_8", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I find the wire coiled into a loop?", + "answer": "You found the red wires coiled under the A-frame ladder in the kitchen, and the green wires in the cabinet next to the A-frame ladder.", + "question_chinese": "我从哪里找到了缠绕成圈状的电线？", + "answer_chinese": "你在厨房里的人字梯下方找到了缠绕成圈的红电线，在人字梯旁边的柜子里找到了绿电线", + "evidence": [ + { + "description": "I found the red coiled wire under the ladder in the kitchen at home.", + "timestamps": { + "start": "00:04:40", + "end": "00:05:01" + } + }, + { + "description": "I found the green wires coiled up in a cabinet next to the step ladder in the kitchen.", + "timestamps": { + "start": "00:06:50", + "end": "00:07:31" + } + } + ], + "question_timestamps": "00:16:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Both sets were under the ladder.", + "B": "You found the red wires coiled under the A-frame ladder in the kitchen, and the green wires in the cabinet next to the A-frame ladder.", + "C": "Both sets were in the cabinet.", + "D": "You found the red wires coiled in the cabinet by the A-frame ladder, and the green wires under the ladder.", + "E": "You found the red wires coiled under the A-frame ladder in the kitchen, and the green wires under the sink cabinet." + }, + "answer_label": "B", + "answer_text": "You found the red wires coiled under the A-frame ladder in the kitchen, and the green wires in the cabinet next to the A-frame ladder.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_8", + "correct_answer": "You found the red wires coiled under the A-frame ladder in the kitchen, and the green wires in the cabinet next to the A-frame ladder.", + "original_options": { + "A": "Both sets were under the ladder.", + "B": "You found the red wires coiled under the A-frame ladder in the kitchen, and the green wires in the cabinet next to the A-frame ladder.", + "C": "Both sets were in the cabinet.", + "D": "You found the red wires coiled in the cabinet by the A-frame ladder, and the green wires under the ladder.", + "E": "You found the red wires coiled under the A-frame ladder in the kitchen, and the green wires under the sink cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 21 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_4_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I find the wire that was coiled into loops?", + "answer": "In the kitchen, you found the coiled red electrical wire below you under the step ladder, and you found the green electrical wire in the cabinet next to the step ladder.", + "question_chinese": "我从哪里找到了缠绕成圈状的电线？", + "answer_chinese": "你在厨房里的人字梯下方找到了缠绕成圈的红电线，在人字梯旁边的柜子里找到了绿电线", + "evidence": [ + { + "description": "I found the red coiled wire under the ladder in the kitchen at home.", + "timestamps": { + "start": "00:04:40", + "end": "00:05:01" + } + }, + { + "description": "I found the green wires coiled up in a cabinet next to the step ladder in the kitchen.", + "timestamps": { + "start": "00:06:50", + "end": "00:07:31" + } + } + ], + "question_timestamps": "00:33:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Red wire was in front.", + "B": "In the kitchen, you found the coiled red electrical wire below you under the step ladder, and you found the green electrical wire in the cabinet next to the step ladder.", + "C": "Red wire was behind you.", + "D": "In the kitchen, you found the coiled red wire under the step ladder, and the green wire on the floor nearby.", + "E": "In the kitchen, you found the coiled red wire under the step ladder, and the green wire in a cabinet across." + }, + "answer_label": "B", + "answer_text": "In the kitchen, you found the coiled red electrical wire below you under the step ladder, and you found the green electrical wire in the cabinet next to the step ladder.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_4_T2", + "correct_answer": "In the kitchen, you found the coiled red electrical wire below you under the step ladder, and you found the green electrical wire in the cabinet next to the step ladder.", + "original_options": { + "A": "Red wire was in front.", + "B": "In the kitchen, you found the coiled red electrical wire below you under the step ladder, and you found the green electrical wire in the cabinet next to the step ladder.", + "C": "Red wire was behind you.", + "D": "In the kitchen, you found the coiled red wire under the step ladder, and the green wire on the floor nearby.", + "E": "In the kitchen, you found the coiled red wire under the step ladder, and the green wire in a cabinet across." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 21 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_2_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of items did I take from the car?", + "answer": "You handled a spray can, a wooden board, a nail gun, and nail gun nails.", + "question_chinese": "我从车上拿过哪几种东西？", + "answer_chinese": "你拿过喷雾剂、木板、钉枪、钉枪钉", + "evidence": [ + { + "description": "Aerosol", + "timestamps": { + "start": "00:03:51", + "end": "00:04:03" + } + }, + { + "description": "Wooden board", + "timestamps": { + "start": "00:05:53", + "end": "00:06:12" + } + }, + { + "description": "Nail gun and nail gun nails", + "timestamps": { + "start": "00:11:43", + "end": "00:12:01" + } + } + ], + "question_timestamps": "00:12:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Just the wooden board.", + "B": "A spray can and board.", + "C": "You handled a spray can, a wooden board, and a nail gun, but no nails for it.", + "D": "You handled a spray can, a wooden board, a nail gun, and nail gun nails.", + "E": "You handled a spray can, a wooden board, nail gun nails, and a hacksaw, but you did not take a nail gun." + }, + "answer_label": "D", + "answer_text": "You handled a spray can, a wooden board, a nail gun, and nail gun nails.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_2_T2", + "correct_answer": "You handled a spray can, a wooden board, a nail gun, and nail gun nails.", + "original_options": { + "A": "Just the wooden board.", + "B": "A spray can and board.", + "C": "You handled a spray can, a wooden board, and a nail gun, but no nails for it.", + "D": "You handled a spray can, a wooden board, a nail gun, and nail gun nails.", + "E": "You handled a spray can, a wooden board, nail gun nails, and a hacksaw, but you did not take a nail gun." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 22 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_2_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of items did I take from the car?", + "answer": "You handled a spray can, a wooden board, a nail gun, nail gun nails, and a hacksaw.", + "question_chinese": "我从车上拿过哪几种东西？", + "answer_chinese": "你拿过喷雾剂、木板、钉枪、钉枪钉和钢锯。", + "evidence": [ + { + "description": "hacksaw", + "timestamps": { + "start": "00:31:00", + "end": "00:31:11" + } + }, + { + "description": "Aerosol", + "timestamps": { + "start": "00:03:51", + "end": "00:04:03" + } + }, + { + "description": "Wooden board", + "timestamps": { + "start": "00:05:53", + "end": "00:06:12" + } + }, + { + "description": "Nail gun and nail gun nails", + "timestamps": { + "start": "00:11:43", + "end": "00:12:01" + } + } + ], + "question_timestamps": "00:33:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Only the nail gun.", + "B": "A board and a hacksaw.", + "C": "You handled a spray can, a wooden board, a nail gun, and nail gun nails, but you never took a hacksaw.", + "D": "You handled a spray can, a wooden board, and a hacksaw, but you did not take a nail gun or its nails.", + "E": "You handled a spray can, a wooden board, a nail gun, nail gun nails, and a hacksaw." + }, + "answer_label": "E", + "answer_text": "You handled a spray can, a wooden board, a nail gun, nail gun nails, and a hacksaw.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_2_T1", + "correct_answer": "You handled a spray can, a wooden board, a nail gun, nail gun nails, and a hacksaw.", + "original_options": { + "A": "Only the nail gun.", + "B": "A board and a hacksaw.", + "C": "You handled a spray can, a wooden board, a nail gun, and nail gun nails, but you never took a hacksaw.", + "D": "You handled a spray can, a wooden board, and a hacksaw, but you did not take a nail gun or its nails.", + "E": "You handled a spray can, a wooden board, a nail gun, nail gun nails, and a hacksaw." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 21, + 22 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_9_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I went out and went to the trunk to get something, what path did I take?", + "answer": "You took a cigarette from the toolbox to your left at the doorway, picked up the white wooden board from the ground, and after going down the stairs you walked toward the white car.", + "question_chinese": "我出门后去车后厢拿东西，经过了什么路径", + "answer_chinese": "你在门口左边的工具箱上拿了烟，拿起地上白色的木板，下楼梯后走向白色的车", + "evidence": [ + { + "description": "You picked up the cigarettes from the toolbox on the left side of the door.", + "timestamps": { + "start": "00:48:50", + "end": "00:48:52" + } + }, + { + "description": "Pick up the white wooden board on the ground.", + "timestamps": { + "start": "00:49:00", + "end": "00:49:01" + } + }, + { + "description": "After going downstairs, walk towards the white car.", + "timestamps": { + "start": "00:49:15", + "end": "00:49:20" + } + } + ], + "question_timestamps": "00:49:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the toolbox on your right.", + "B": "You took a cigarette from the toolbox to your left at the doorway, picked up the white wooden board from the ground, and after going down the stairs you walked toward the white car.", + "C": "You went straight to the car.", + "D": "You took a cigarette from the toolbox to your left, then went back inside instead of heading to the car.", + "E": "You grabbed a cigarette at the doorway and walked toward the white car without picking up the wooden board." + }, + "answer_label": "B", + "answer_text": "You took a cigarette from the toolbox to your left at the doorway, picked up the white wooden board from the ground, and after going down the stairs you walked toward the white car.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_9_T1", + "correct_answer": "You took a cigarette from the toolbox to your left at the doorway, picked up the white wooden board from the ground, and after going down the stairs you walked toward the white car.", + "original_options": { + "A": "From the toolbox on your right.", + "B": "You took a cigarette from the toolbox to your left at the doorway, picked up the white wooden board from the ground, and after going down the stairs you walked toward the white car.", + "C": "You went straight to the car.", + "D": "You took a cigarette from the toolbox to your left, then went back inside instead of heading to the car.", + "E": "You grabbed a cigarette at the doorway and walked toward the white car without picking up the wooden board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 19 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_9_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I went out and went to the trunk to get something, what path did I take?", + "answer": "You took a cigarette from the toolbox to your left at the doorway, picked up the white wooden board from the ground, and after going down the stairs you walked toward the white car. After grabbing the items, you walked to the right.", + "question_chinese": "我出门后去车后厢拿东西，经过了什么路径", + "answer_chinese": "你在门口左边的工具箱上拿了烟，拿起地上白色的木板，下楼梯后走向白色的车，拿完东西后向右走", + "evidence": [ + { + "description": "You picked up the cigarettes from the toolbox on the left side of the door.", + "timestamps": { + "start": "00:48:50", + "end": "00:48:51" + } + }, + { + "description": "Pick up the white wooden board on the ground.", + "timestamps": { + "start": "00:49:00", + "end": "00:49:01" + } + }, + { + "description": "After going downstairs, walk towards the white car.", + "timestamps": { + "start": "00:49:15", + "end": "00:49:20" + } + } + ], + "question_timestamps": "00:49:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Afterward, you walked left.", + "B": "Afterward, you walked straight ahead.", + "C": "You took a cigarette from the toolbox to your left at the doorway, picked up the white wooden board from the ground, and after going down the stairs you walked toward the white car. After grabbing the items, you walked to the right.", + "D": "You took a cigarette from the toolbox on your right, grabbed the board, went downstairs, then turned right.", + "E": "You walked down the stairs toward the white car, grabbed the items, and then turned left instead of right." + }, + "answer_label": "C", + "answer_text": "You took a cigarette from the toolbox to your left at the doorway, picked up the white wooden board from the ground, and after going down the stairs you walked toward the white car. After grabbing the items, you walked to the right.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_9_T2", + "correct_answer": "You took a cigarette from the toolbox to your left at the doorway, picked up the white wooden board from the ground, and after going down the stairs you walked toward the white car. After grabbing the items, you walked to the right.", + "original_options": { + "A": "Afterward, you walked left.", + "B": "Afterward, you walked straight ahead.", + "C": "You took a cigarette from the toolbox to your left at the doorway, picked up the white wooden board from the ground, and after going down the stairs you walked toward the white car. After grabbing the items, you walked to the right.", + "D": "You took a cigarette from the toolbox on your right, grabbed the board, went downstairs, then turned right.", + "E": "You walked down the stairs toward the white car, grabbed the items, and then turned left instead of right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 19 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_1_T1", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I pick up the wooden board from?", + "answer": "You took the wooden board off the car in front of you.", + "question_chinese": "我是从哪里取的木板？", + "answer_chinese": "你从面前的车上取下了木板。", + "evidence": [ + { + "description": "You take the board from the car.", + "timestamps": { + "start": "00:05:51", + "end": "00:06:13" + } + } + ], + "question_timestamps": "00:16:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Off the car behind you.", + "B": "Off the car to your left.", + "C": "You took the wooden board off the car to your right while standing near the front area.", + "D": "You grabbed the wooden board from the car behind you, reaching back instead of taking it from the front.", + "E": "You took the wooden board off the car in front of you." + }, + "answer_label": "E", + "answer_text": "You took the wooden board off the car in front of you.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_1_T1", + "correct_answer": "You took the wooden board off the car in front of you.", + "original_options": { + "A": "Off the car behind you.", + "B": "Off the car to your left.", + "C": "You took the wooden board off the car to your right while standing near the front area.", + "D": "You grabbed the wooden board from the car behind you, reaching back instead of taking it from the front.", + "E": "You took the wooden board off the car in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 19 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_1_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I pick up the wooden board from?", + "answer": "You picked up the wooden board from the ground in front of you.", + "question_chinese": "我是从哪里取的木板？", + "answer_chinese": "你从面前的地上拿起的木板。", + "evidence": [ + { + "description": "The wooden board you picked up from the ground.", + "timestamps": { + "start": "00:49:00", + "end": "00:49:01" + } + } + ], + "question_timestamps": "00:49:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the ground behind you.", + "B": "From the ground to your left.", + "C": "You picked up the wooden board from the ground to your right, not from directly in front of you.", + "D": "You picked up the wooden board from the ground in front of you.", + "E": "You lifted the wooden board from the ground behind you after turning around to face the opposite direction." + }, + "answer_label": "D", + "answer_text": "You picked up the wooden board from the ground in front of you.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_1_T2", + "correct_answer": "You picked up the wooden board from the ground in front of you.", + "original_options": { + "A": "From the ground behind you.", + "B": "From the ground to your left.", + "C": "You picked up the wooden board from the ground to your right, not from directly in front of you.", + "D": "You picked up the wooden board from the ground in front of you.", + "E": "You lifted the wooden board from the ground behind you after turning around to face the opposite direction." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 18 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_15", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach my phone without moving my feet now?", + "answer": "You cannot get your phone.", + "question_chinese": "我现在能在不移动脚步的情况下拿到手机吗？", + "answer_chinese": "你不能拿到手机。", + "evidence": [ + { + "description": "The mobile phone is on the tool storage box at the higher end of the steps in front of the house.", + "timestamps": { + "start": "00:48:55", + "end": "00:49:07" + } + }, + { + "description": "You are now at the lower end of the steps in front of your home", + "timestamps": { + "start": "00:50:09", + "end": "00:50:15" + } + } + ], + "question_timestamps": "00:50:21", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can reach your phone without moving your feet.", + "B": "You cannot get your phone." + }, + "answer_label": "B", + "answer_text": "You cannot get your phone.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_15", + "correct_answer": "You cannot get your phone.", + "original_options": { + "A": "You can reach your phone without moving your feet.", + "B": "You cannot get your phone." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_16", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach my phone without moving my feet now?", + "answer": "You cannot get your phone right now.", + "question_chinese": "我现在能在不移动脚步的情况下拿到手机吗？", + "answer_chinese": "你现在不能拿到手机。", + "evidence": [ + { + "description": "The mobile phone is on the tool storage box at the doorstep.", + "timestamps": { + "start": "00:48:35", + "end": "00:49:11" + } + }, + { + "description": "You are now in the kitchen.", + "timestamps": { + "start": "01:07:06", + "end": "01:07:10" + } + } + ], + "question_timestamps": "01:07:32", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You cannot get your phone right now.", + "B": "You can reach your phone without moving." + }, + "answer_label": "A", + "answer_text": "You cannot get your phone right now.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_16", + "correct_answer": "You cannot get your phone right now.", + "original_options": { + "A": "You cannot get your phone right now.", + "B": "You can reach your phone without moving." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_6_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my phone?", + "answer": "On the toolbox in front of you.", + "question_chinese": "我的手机在哪里？", + "answer_chinese": "在你面前的工具箱上", + "evidence": [ + { + "description": "The mobile phone is on the tool storage box in front of the doorstep.", + "timestamps": { + "start": "00:48:49", + "end": "00:48:50" + } + } + ], + "question_timestamps": "00:49:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the toolbox behind you.", + "B": "On the toolbox to your left.", + "C": "Your phone is on the toolbox to your right, not on the one directly in front of you.", + "D": "Your phone is resting on a toolbox behind you, so you would need to turn around to see it.", + "E": "On the toolbox in front of you." + }, + "answer_label": "E", + "answer_text": "On the toolbox in front of you.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_6_T1", + "correct_answer": "On the toolbox in front of you.", + "original_options": { + "A": "On the toolbox behind you.", + "B": "On the toolbox to your left.", + "C": "Your phone is on the toolbox to your right, not on the one directly in front of you.", + "D": "Your phone is resting on a toolbox behind you, so you would need to turn around to see it.", + "E": "On the toolbox in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 19 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_6_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my phone?", + "answer": "The phone is to your left, on the toolbox in front of the doorway.", + "question_chinese": "我的手机在哪里？", + "answer_chinese": "手机在你的左边，门口前的工具箱上", + "evidence": [ + { + "description": "The mobile phone is on the tool storage box in front of the doorstep.", + "timestamps": { + "start": "00:48:56", + "end": "00:49:10" + } + } + ], + "question_timestamps": "00:50:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The phone is to your left, on the toolbox in front of the doorway.", + "B": "The phone is in front of you.", + "C": "The phone is to your right.", + "D": "The phone is behind you, sitting on the toolbox in front of the doorway rather than to your left.", + "E": "The phone is to your right on the toolbox by the doorway, not on the left-hand side." + }, + "answer_label": "A", + "answer_text": "The phone is to your left, on the toolbox in front of the doorway.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_6_T2", + "correct_answer": "The phone is to your left, on the toolbox in front of the doorway.", + "original_options": { + "A": "The phone is to your left, on the toolbox in front of the doorway.", + "B": "The phone is in front of you.", + "C": "The phone is to your right.", + "D": "The phone is behind you, sitting on the toolbox in front of the doorway rather than to your left.", + "E": "The phone is to your right on the toolbox by the doorway, not on the left-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 17 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I get my first piece of wood from?", + "answer": "You took the board off the car.", + "question_chinese": "我第一次取木板是从哪里取的？", + "answer_chinese": "你从车上取下了木板。", + "evidence": [ + { + "description": "You take the board from the car.", + "timestamps": { + "start": "00:05:51", + "end": "00:06:13" + } + } + ], + "question_timestamps": "00:16:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took it from the ground.", + "B": "You grabbed it from the workbench.", + "C": "You took the board off the car.", + "D": "You pulled the first board out of the storage room before carrying it over to where you were working.", + "E": "You removed the first piece of wood from the toolbox and then brought it back to your work area." + }, + "answer_label": "C", + "answer_text": "You took the board off the car.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_1", + "correct_answer": "You took the board off the car.", + "original_options": { + "A": "You took it from the ground.", + "B": "You grabbed it from the workbench.", + "C": "You took the board off the car.", + "D": "You pulled the first board out of the storage room before carrying it over to where you were working.", + "E": "You removed the first piece of wood from the toolbox and then brought it back to your work area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I get my first piece of wood from?", + "answer": "You took the board off the car.", + "question_chinese": "我第一次取木板是从哪里取的？", + "answer_chinese": "你从车上取下了木板。", + "evidence": [ + { + "description": "You take the board from the car.", + "timestamps": { + "start": "00:05:51", + "end": "00:06:13" + } + } + ], + "question_timestamps": "00:33:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took it off the shelf.", + "B": "You grabbed it from the shed.", + "C": "You retrieved the first piece of wood from the garage and carried it out to start working.", + "D": "You pulled the board down from the attic and brought it over to where you were standing.", + "E": "You took the board off the car." + }, + "answer_label": "E", + "answer_text": "You took the board off the car.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_2", + "correct_answer": "You took the board off the car.", + "original_options": { + "A": "You took it off the shelf.", + "B": "You grabbed it from the shed.", + "C": "You retrieved the first piece of wood from the garage and carried it out to start working.", + "D": "You pulled the board down from the attic and brought it over to where you were standing.", + "E": "You took the board off the car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_11", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the mobile phone?", + "answer": "The mobile phone is on the tool storage box in front of the doorstep.", + "question_chinese": "手机在哪里？", + "answer_chinese": "手机在家门口前的工具收纳箱上。", + "evidence": [ + { + "description": "The mobile phone is on the tool storage box in front of the doorstep.", + "timestamps": { + "start": "00:48:56", + "end": "00:49:10" + } + } + ], + "question_timestamps": "00:51:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The mobile phone is on the tool storage box in front of the doorstep.", + "B": "It’s on the kitchen counter.", + "C": "It’s under the coffee table.", + "D": "The mobile phone is in the drawer beside the workbench, tucked away rather than left out.", + "E": "The mobile phone is on the bookshelf above the living room floor, not near the doorstep." + }, + "answer_label": "A", + "answer_text": "The mobile phone is on the tool storage box in front of the doorstep.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_11", + "correct_answer": "The mobile phone is on the tool storage box in front of the doorstep.", + "original_options": { + "A": "The mobile phone is on the tool storage box in front of the doorstep.", + "B": "It’s on the kitchen counter.", + "C": "It’s under the coffee table.", + "D": "The mobile phone is in the drawer beside the workbench, tucked away rather than left out.", + "E": "The mobile phone is on the bookshelf above the living room floor, not near the doorstep." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_12", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the mobile phone?", + "answer": "The mobile phone is on the tool storage box in front of the doorstep.", + "question_chinese": "手机在哪里？", + "answer_chinese": "手机在家门口前的工具收纳箱上。", + "evidence": [ + { + "description": "The mobile phone is on the tool storage box in front of the doorstep.", + "timestamps": { + "start": "00:48:56", + "end": "00:49:10" + } + } + ], + "question_timestamps": "01:07:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s in the kitchen drawer.", + "B": "It’s on the living room shelf.", + "C": "The mobile phone is under the couch in the corner of the bedroom, not sitting on any box.", + "D": "The mobile phone is on the tool storage box in front of the doorstep.", + "E": "The mobile phone is at the bottom of the stairs next to the umbrella stand, away from the doorstep." + }, + "answer_label": "D", + "answer_text": "The mobile phone is on the tool storage box in front of the doorstep.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_12", + "correct_answer": "The mobile phone is on the tool storage box in front of the doorstep.", + "original_options": { + "A": "It’s in the kitchen drawer.", + "B": "It’s on the living room shelf.", + "C": "The mobile phone is under the couch in the corner of the bedroom, not sitting on any box.", + "D": "The mobile phone is on the tool storage box in front of the doorstep.", + "E": "The mobile phone is at the bottom of the stairs next to the umbrella stand, away from the doorstep." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_13", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me now, the mobile phone or the miter saw?", + "answer": "The miter saw is closer to you.", + "question_chinese": "现在手机和斜切锯哪个离我更近？", + "answer_chinese": "斜切锯离你更近。", + "evidence": [ + { + "description": "The miter saw is located at the lower end of the steps in front of the house.", + "timestamps": { + "start": "00:48:58", + "end": "00:49:20" + } + }, + { + "description": "The mobile phone is on the storage box at the higher end of the steps near the doorstep.", + "timestamps": { + "start": "00:48:35", + "end": "00:49:09" + } + } + ], + "question_timestamps": "00:50:59", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The miter saw is closer to you.", + "B": "The mobile phone is closer to you." + }, + "answer_label": "A", + "answer_text": "The miter saw is closer to you.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_13", + "correct_answer": "The miter saw is closer to you.", + "original_options": { + "A": "The miter saw is closer to you.", + "B": "The mobile phone is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_14", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me now, the mobile phone or the miter saw?", + "answer": "Your phone is closer to you.", + "question_chinese": "现在手机和斜切锯哪个离我更近？", + "answer_chinese": "手机离你更近。", + "evidence": [ + { + "description": "The miter saw is located at the lower end of the steps in front of the house.", + "timestamps": { + "start": "00:48:58", + "end": "00:49:20" + } + }, + { + "description": "The mobile phone is on the storage box at the higher end of the steps near the doorstep.", + "timestamps": { + "start": "00:48:35", + "end": "00:49:09" + } + } + ], + "question_timestamps": "01:07:32", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Your phone is closer to you.", + "B": "The miter saw is closer to you." + }, + "answer_label": "A", + "answer_text": "Your phone is closer to you.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_14", + "correct_answer": "Your phone is closer to you.", + "original_options": { + "A": "Your phone is closer to you.", + "B": "The miter saw is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_3_T1", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the white car relative to me?", + "answer": "In front of you.", + "question_chinese": "白车在我的哪个方位", + "answer_chinese": "你的前方", + "evidence": [ + { + "description": "The white car is parked on the road in front of the house.", + "timestamps": { + "start": "00:49:01", + "end": "00:49:02" + } + } + ], + "question_timestamps": "00:49:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "In front of you.", + "D": "To your right, not directly in front of you where you’re facing.", + "E": "Above you, rather than on the ground ahead in your line of sight." + }, + "answer_label": "C", + "answer_text": "In front of you.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_3_T1", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "In front of you.", + "D": "To your right, not directly in front of you where you’re facing.", + "E": "Above you, rather than on the ground ahead in your line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 13 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_3_T2", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the white car relative to me?", + "answer": "To your right.", + "question_chinese": "白车在我的哪个方位", + "answer_chinese": "你的右边", + "evidence": [ + { + "description": "The white car is parked on the road in front of the house.", + "timestamps": { + "start": "00:49:01", + "end": "00:49:02" + } + } + ], + "question_timestamps": "00:50:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your right.", + "D": "To your left, not on your right-hand side as you’re oriented.", + "E": "Above you, rather than positioned beside you at ground level on your right." + }, + "answer_label": "C", + "answer_text": "To your right.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_3_T2", + "correct_answer": "To your right.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your right.", + "D": "To your left, not on your right-hand side as you’re oriented.", + "E": "Above you, rather than positioned beside you at ground level on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 12, + 13 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_17", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the front door open or closed?", + "answer": "The front door of the car is closed.", + "question_chinese": "车前门是打开的还是关闭的？", + "answer_chinese": "车前门是关闭的。", + "evidence": [ + { + "description": "The front door of the car is closed.", + "timestamps": { + "start": "00:48:58", + "end": "00:49:12" + } + } + ], + "question_timestamps": "00:52:05", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The front door of the car is open.", + "B": "The front door of the car is closed." + }, + "answer_label": "B", + "answer_text": "The front door of the car is closed.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_17", + "correct_answer": "The front door of the car is closed.", + "original_options": { + "A": "The front door of the car is open.", + "B": "The front door of the car is closed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84", + "video_length": "01:07:32", + "qid": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_18", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the front door open or closed?", + "answer": "The front door of the car is closed.", + "question_chinese": "车前门是打开的还是关闭的？", + "answer_chinese": "车前门是关闭的。", + "evidence": [ + { + "description": "The front door of the car is closed.", + "timestamps": { + "start": "00:48:58", + "end": "00:49:12" + } + } + ], + "question_timestamps": "01:07:32", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The front door of the car is closed.", + "B": "The front door of the car is open." + }, + "answer_label": "A", + "answer_text": "The front door of the car is closed.", + "q_id": "18a3840b-7463-43c4-9aa9-b1d8e486fa84_18", + "correct_answer": "The front door of the car is closed.", + "original_options": { + "A": "The front door of the car is closed.", + "B": "The front door of the car is open." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + } +] diff --git a/QAs/18c05110-1ef1-4da2-8b59-c80bf0cc9751-mcq.json b/QAs/18c05110-1ef1-4da2-8b59-c80bf0cc9751-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d9664b4c98b1c0f877c9197b2a21b4a4a2ccad4d --- /dev/null +++ b/QAs/18c05110-1ef1-4da2-8b59-c80bf0cc9751-mcq.json @@ -0,0 +1,1390 @@ +[ + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_3_T2", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was stirring the food in the pot just now, could I reach the banana?", + "answer": "You couldn’t reach the banana; it was to your back-left.", + "question_chinese": "我刚才在搅拌锅里食物时，能拿到香蕉吗？", + "answer_chinese": "你不能拿到香蕉，香蕉在你左后方。", + "evidence": [ + { + "description": "The banana is on your left, about 10cm away from the microwave.", + "timestamps": { + "start": "00:10:40", + "end": "00:11:41" + } + }, + { + "description": "The banana is on your left, about 10cm away from the microwave.", + "timestamps": { + "start": "00:11:45", + "end": "00:13:00" + } + } + ], + "question_timestamps": "00:16:42", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You could reach the banana; it was on your front-left.", + "B": "You couldn’t reach the banana; it was to your back-left." + }, + "answer_label": "B", + "answer_text": "You couldn’t reach the banana; it was to your back-left.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_3_T2", + "correct_answer": "You couldn’t reach the banana; it was to your back-left.", + "original_options": { + "A": "You could reach the banana; it was on your front-left.", + "B": "You couldn’t reach the banana; it was to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_3_T1", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was stirring the food in the pot just now, could I reach the banana?", + "answer": "When you were just stirring the food in the pot, you could reach the banana. It was on your left-hand side, about 10 cm from the microwave, next to it.", + "question_chinese": "我刚才在搅拌锅里食物时，能拿到香蕉吗？", + "answer_chinese": "你刚才在搅拌锅里食物时，能拿到香蕉，香蕉就在你的左手边，微波炉旁边大概10cm的地方。", + "evidence": [ + { + "description": "The banana is on your left, about 10cm away from the microwave.", + "timestamps": { + "start": "00:10:40", + "end": "00:11:41" + } + }, + { + "description": "The banana is on your left, about 10cm away from the microwave.", + "timestamps": { + "start": "00:11:45", + "end": "00:13:00" + } + } + ], + "question_timestamps": "00:27:57", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were just stirring the food in the pot, you could reach the banana. It was on your left-hand side, about 10 cm from the microwave, next to it.", + "B": "You couldn’t reach the banana; it was to your back-left." + }, + "answer_label": "A", + "answer_text": "When you were just stirring the food in the pot, you could reach the banana. It was on your left-hand side, about 10 cm from the microwave, next to it.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_3_T1", + "correct_answer": "When you were just stirring the food in the pot, you could reach the banana. It was on your left-hand side, about 10 cm from the microwave, next to it.", + "original_options": { + "A": "When you were just stirring the food in the pot, you could reach the banana. It was on your left-hand side, about 10 cm from the microwave, next to it.", + "B": "You couldn’t reach the banana; it was to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_14", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Could I reach for the banana while I was stirring the food in it", + "answer": "Yes, you can.", + "question_chinese": "我刚才在搅拌锅里食物时，能拿到香蕉吗？", + "answer_chinese": "是的，你可以", + "evidence": [ + { + "description": "The banana is on your left, about 10cm away from the microwave.", + "timestamps": { + "start": "00:10:40", + "end": "00:11:41" + } + }, + { + "description": "The banana is on your left, about 10cm away from the microwave.", + "timestamps": { + "start": "00:11:45", + "end": "00:13:00" + } + } + ], + "question_timestamps": "00:40:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "answer_label": "A", + "answer_text": "Yes, you can.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_14", + "correct_answer": "Yes, you can.", + "original_options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_7_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the washed knife just now?", + "answer": "You just put the washed knife on the cutting board on the sink to your left.", + "question_chinese": "我刚才把洗完的刀放在了哪里？", + "answer_chinese": "你刚才把洗完的刀放在你左边的洗手台上的菜板上。", + "evidence": [ + { + "description": "You just placed the washed knife on the chopping board on the sink.", + "timestamps": { + "start": "00:28:20", + "end": "00:28:22" + } + } + ], + "question_timestamps": "00:28:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the sink in front of you.", + "B": "On the sink to your right.", + "C": "You just put the washed knife on the cutting board on the sink to your left.", + "D": "You put the washed knife on the cutting board on the sink in front of you, not on your left side.", + "E": "You placed the washed knife on the cutting board on the sink to your right, beside the basin." + }, + "answer_label": "C", + "answer_text": "You just put the washed knife on the cutting board on the sink to your left.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_7_T1", + "correct_answer": "You just put the washed knife on the cutting board on the sink to your left.", + "original_options": { + "A": "On the sink in front of you.", + "B": "On the sink to your right.", + "C": "You just put the washed knife on the cutting board on the sink to your left.", + "D": "You put the washed knife on the cutting board on the sink in front of you, not on your left side.", + "E": "You placed the washed knife on the cutting board on the sink to your right, beside the basin." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 21, + 18 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_11", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the washed knife just now?", + "answer": "You just placed the washed knife on the shelf next to the sink.", + "question_chinese": "我刚才把洗完的刀放在了哪里？", + "answer_chinese": "你刚才把洗完的刀放在洗手台旁边的置物架上。", + "evidence": [ + { + "description": "I just placed the washed knife on the shelf next to the sink.", + "timestamps": { + "start": "00:28:37", + "end": "00:28:38" + } + } + ], + "question_timestamps": "00:36:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the counter beside the stove.", + "B": "On the rack near the window.", + "C": "You placed the washed knife on the shelf above the dishwasher, away from the sink area.", + "D": "You just placed the washed knife on the shelf next to the sink.", + "E": "You set the washed knife on the table next to the fridge, not on the shelf by the sink." + }, + "answer_label": "D", + "answer_text": "You just placed the washed knife on the shelf next to the sink.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_11", + "correct_answer": "You just placed the washed knife on the shelf next to the sink.", + "original_options": { + "A": "On the counter beside the stove.", + "B": "On the rack near the window.", + "C": "You placed the washed knife on the shelf above the dishwasher, away from the sink area.", + "D": "You just placed the washed knife on the shelf next to the sink.", + "E": "You set the washed knife on the table next to the fridge, not on the shelf by the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 19 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_7_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the washed knife just now?", + "answer": "You just put the washed knife on the storage rack next to the sink to your left.", + "question_chinese": "我刚才把洗完的刀放在了哪里？", + "answer_chinese": "你刚才把洗完的刀放在你左边的洗手台旁边的置物架上。", + "evidence": [ + { + "description": "I just placed the washed knife on the shelf next to the sink.", + "timestamps": { + "start": "00:28:37", + "end": "00:28:38" + } + } + ], + "question_timestamps": "00:42:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the rack in front of you.", + "B": "You just put the washed knife on the storage rack next to the sink to your left.", + "C": "On the rack to your right.", + "D": "You put the washed knife on the storage rack next to the sink behind you, not to your left.", + "E": "You placed the washed knife on the storage rack next to the sink in front of you rather than to your left." + }, + "answer_label": "B", + "answer_text": "You just put the washed knife on the storage rack next to the sink to your left.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_7_T2", + "correct_answer": "You just put the washed knife on the storage rack next to the sink to your left.", + "original_options": { + "A": "On the rack in front of you.", + "B": "You just put the washed knife on the storage rack next to the sink to your left.", + "C": "On the rack to your right.", + "D": "You put the washed knife on the storage rack next to the sink behind you, not to your left.", + "E": "You placed the washed knife on the storage rack next to the sink in front of you rather than to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 22 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_9", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was washing things just now, was the gray shelf on my left or right side?", + "answer": "When you were washing things just now, the gray shelf was on your left side.", + "question_chinese": "我刚才洗东西的时候，灰色置物架在我的左手边还是右手边？", + "answer_chinese": "你刚才洗东西的时候，灰色置物架在你的左手边。", + "evidence": [ + { + "description": "When I were washing things just now, the gray shelf was on my left side.", + "timestamps": { + "start": "00:20:18", + "end": "00:24:05" + } + }, + { + "description": "When I were washing things just now, the gray shelf was on my left side.", + "timestamps": { + "start": "00:26:03", + "end": "00:26:33" + } + }, + { + "description": "When I were washing things just now, the gray shelf was on my left side.", + "timestamps": { + "start": "00:27:10", + "end": "00:29:49" + } + } + ], + "question_timestamps": "00:33:11", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "When you were washing things just now, the gray shelf was on your left side.", + "B": "When you were washing things just now, the gray shelf was on your right side." + }, + "answer_label": "A", + "answer_text": "When you were washing things just now, the gray shelf was on your left side.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_9", + "correct_answer": "When you were washing things just now, the gray shelf was on your left side.", + "original_options": { + "A": "When you were washing things just now, the gray shelf was on your left side.", + "B": "When you were washing things just now, the gray shelf was on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_15", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was washing things just now, was the gray shelf on my left or right side?", + "answer": "When you were washing things just now, the gray shelf was on your left side.", + "question_chinese": "我刚才洗东西的时候，灰色置物架在我的左手边还是右手边？", + "answer_chinese": "你刚才洗东西的时候，灰色置物架在你的左手边。", + "evidence": [ + { + "description": "When I were washing things just now, the gray shelf was on my left side.", + "timestamps": { + "start": "00:20:18", + "end": "00:24:05" + } + }, + { + "description": "When I were washing things just now, the gray shelf was on my left side.", + "timestamps": { + "start": "00:26:03", + "end": "00:26:33" + } + }, + { + "description": "When I were washing things just now, the gray shelf was on my left side.", + "timestamps": { + "start": "00:27:10", + "end": "00:29:49" + } + } + ], + "question_timestamps": "00:44:09", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "When you were washing things just now, the gray shelf was on your left side.", + "B": "When you were washing things just now, the gray shelf was on your right side." + }, + "answer_label": "A", + "answer_text": "When you were washing things just now, the gray shelf was on your left side.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_15", + "correct_answer": "When you were washing things just now, the gray shelf was on your left side.", + "original_options": { + "A": "When you were washing things just now, the gray shelf was on your left side.", + "B": "When you were washing things just now, the gray shelf was on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_6_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How should I get to the laundry room from where I am now?", + "answer": "If you turn around, you’ll be facing the laundry room.", + "question_chinese": "我现在该怎么前往洗衣房？", + "answer_chinese": "向后转就是洗衣房。", + "evidence": [ + { + "description": "The location of the laundry room", + "timestamps": { + "start": "00:33:30", + "end": "00:33:32" + } + } + ], + "question_timestamps": "00:33:41", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Turn left and go straight.", + "B": "Turn right and go straight.", + "C": "If you walk forward without turning, you’ll reach the laundry room ahead of you.", + "D": "If you turn around, you’ll be facing the laundry room.", + "E": "Turn around, then take a right and continue straight until you reach the laundry room." + }, + "answer_label": "D", + "answer_text": "If you turn around, you’ll be facing the laundry room.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_6_T2", + "correct_answer": "If you turn around, you’ll be facing the laundry room.", + "original_options": { + "A": "Turn left and go straight.", + "B": "Turn right and go straight.", + "C": "If you walk forward without turning, you’ll reach the laundry room ahead of you.", + "D": "If you turn around, you’ll be facing the laundry room.", + "E": "Turn around, then take a right and continue straight until you reach the laundry room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_6_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How should I get to the laundry room from where I am now?", + "answer": "You should turn right and then go straight.", + "question_chinese": "我现在该怎么前往洗衣房？", + "answer_chinese": "你应该向右转然后直走。", + "evidence": [ + { + "description": "The location of the laundry room", + "timestamps": { + "start": "00:33:30", + "end": "00:33:32" + } + } + ], + "question_timestamps": "00:35:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Turn left and go straight.", + "B": "Turn around to face it.", + "C": "Go straight first, then turn right; that route gets you to the laundry room from here.", + "D": "Turn right, but stop before going straight; the laundry room won’t be straight ahead afterward.", + "E": "You should turn right and then go straight." + }, + "answer_label": "E", + "answer_text": "You should turn right and then go straight.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_6_T1", + "correct_answer": "You should turn right and then go straight.", + "original_options": { + "A": "Turn left and go straight.", + "B": "Turn around to face it.", + "C": "Go straight first, then turn right; that route gets you to the laundry room from here.", + "D": "Turn right, but stop before going straight; the laundry room won’t be straight ahead afterward.", + "E": "You should turn right and then go straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_8_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How can I get the flamingo-shaped inflatable swim ring?", + "answer": "You can reach it if you turn left.", + "question_chinese": "我该怎样去拿到火烈鸟造型的充气游泳圈？", + "answer_chinese": "你向左转就可以拿到。", + "evidence": [ + { + "description": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "timestamps": { + "start": "00:02:48", + "end": "00:03:01" + } + } + ], + "question_timestamps": "00:03:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right to reach it.", + "B": "You can reach it if you turn left.", + "C": "Turn around to reach it.", + "D": "You can reach it if you turn around and walk straight toward it from where you are standing.", + "E": "You can reach it by turning right and taking a few steps forward to get close enough." + }, + "answer_label": "B", + "answer_text": "You can reach it if you turn left.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_8_T1", + "correct_answer": "You can reach it if you turn left.", + "original_options": { + "A": "Turn right to reach it.", + "B": "You can reach it if you turn left.", + "C": "Turn around to reach it.", + "D": "You can reach it if you turn around and walk straight toward it from where you are standing.", + "E": "You can reach it by turning right and taking a few steps forward to get close enough." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_8_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How can I get the flamingo-shaped inflatable swim ring?", + "answer": "You can reach it if you turn around and walk straight.", + "question_chinese": "我该怎样去拿到火烈鸟造型的充气游泳圈？", + "answer_chinese": "你向后转直走就可以拿到。", + "evidence": [ + { + "description": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "timestamps": { + "start": "00:02:48", + "end": "00:03:01" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left to reach it.", + "B": "Turn right to reach it.", + "C": "You can reach it if you turn around and walk straight.", + "D": "You can reach it if you turn left and walk straight a short distance instead of turning around.", + "E": "You can reach it by turning right and walking straight ahead, rather than turning around first." + }, + "answer_label": "C", + "answer_text": "You can reach it if you turn around and walk straight.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_8_T2", + "correct_answer": "You can reach it if you turn around and walk straight.", + "original_options": { + "A": "Turn left to reach it.", + "B": "Turn right to reach it.", + "C": "You can reach it if you turn around and walk straight.", + "D": "You can reach it if you turn left and walk straight a short distance instead of turning around.", + "E": "You can reach it by turning right and walking straight ahead, rather than turning around first." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 16 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_10", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many washing machines are there in the laundry room?", + "answer": "There are two washing machines in the laundry room.", + "question_chinese": "洗衣房有几个洗衣机？", + "answer_chinese": "洗衣房有两个洗衣机。", + "evidence": [ + { + "description": "There are two washing machines in the laundry room.", + "timestamps": { + "start": "00:30:30", + "end": "00:33:32" + } + } + ], + "question_timestamps": "00:36:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one washing machine.", + "B": "There are two washing machines in the laundry room.", + "C": "There are three washing machines.", + "D": "There are four washing machines in the laundry room, lined up along the wall.", + "E": "There are five washing machines in the laundry room, not just a pair side by side." + }, + "answer_label": "B", + "answer_text": "There are two washing machines in the laundry room.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_10", + "correct_answer": "There are two washing machines in the laundry room.", + "original_options": { + "A": "There is one washing machine.", + "B": "There are two washing machines in the laundry room.", + "C": "There are three washing machines.", + "D": "There are four washing machines in the laundry room, lined up along the wall.", + "E": "There are five washing machines in the laundry room, not just a pair side by side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_16", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many washing machines are there in the laundry room?", + "answer": "There are two washing machines in the laundry room.", + "question_chinese": "洗衣房有几个洗衣机？", + "answer_chinese": "洗衣房有两个洗衣机。", + "evidence": [ + { + "description": "There are two washing machines in the laundry room.", + "timestamps": { + "start": "00:30:30", + "end": "00:33:32" + } + } + ], + "question_timestamps": "00:43:58", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one washing machine.", + "B": "There are two washing machines in the laundry room.", + "C": "There are three washing machines.", + "D": "There are four washing machines in the laundry room, arranged in a row.", + "E": "There are five washing machines in the laundry room, more than the two you might expect." + }, + "answer_label": "B", + "answer_text": "There are two washing machines in the laundry room.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_16", + "correct_answer": "There are two washing machines in the laundry room.", + "original_options": { + "A": "There is one washing machine.", + "B": "There are two washing machines in the laundry room.", + "C": "There are three washing machines.", + "D": "There are four washing machines in the laundry room, arranged in a row.", + "E": "There are five washing machines in the laundry room, more than the two you might expect." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 16 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_12", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the inflatable swimming ring shaped like a flamingo located in the room?", + "answer": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "question_chinese": "火烈鸟造型的充气游泳圈在房间里哪个位置？", + "answer_chinese": "火烈鸟造型的充气游泳圈在沙发和纸箱中间。", + "evidence": [ + { + "description": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "timestamps": { + "start": "00:02:48", + "end": "00:03:01" + } + } + ], + "question_timestamps": "00:22:36", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "B": "It’s in front of the sofa.", + "C": "It’s beside the cardboard box.", + "D": "The flamingo-shaped inflatable ring is placed on the sofa rather than between the sofa and the box.", + "E": "The inflatable flamingo ring is located next to the sofa, not positioned in the gap by the cardboard box." + }, + "answer_label": "A", + "answer_text": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_12", + "correct_answer": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "original_options": { + "A": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "B": "It’s in front of the sofa.", + "C": "It’s beside the cardboard box.", + "D": "The flamingo-shaped inflatable ring is placed on the sofa rather than between the sofa and the box.", + "E": "The inflatable flamingo ring is located next to the sofa, not positioned in the gap by the cardboard box." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 19 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_18", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the inflatable swimming ring shaped like a flamingo located in the room?", + "answer": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "question_chinese": "火烈鸟造型的充气游泳圈在房间里哪个位置？", + "answer_chinese": "火烈鸟造型的充气游泳圈在沙发和纸箱中间。", + "evidence": [ + { + "description": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "timestamps": { + "start": "00:02:48", + "end": "00:03:01" + } + } + ], + "question_timestamps": "00:43:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "B": "It is behind the sofa.", + "C": "It is in front of the box.", + "D": "The flamingo-shaped swimming ring is leaning against the cardboard box, not near the sofa.", + "E": "The inflatable flamingo ring is placed to the side of the sofa, with the cardboard box farther away." + }, + "answer_label": "A", + "answer_text": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_18", + "correct_answer": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "original_options": { + "A": "The inflatable swimming ring in the shape of a flamingo is located between the sofa and the cardboard box.", + "B": "It is behind the sofa.", + "C": "It is in front of the box.", + "D": "The flamingo-shaped swimming ring is leaning against the cardboard box, not near the sofa.", + "E": "The inflatable flamingo ring is placed to the side of the sofa, with the cardboard box farther away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 14, + 18 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_5_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the gray storage rack on my left-hand side or my right-hand side right now?", + "answer": "When you were washing things just now, the gray storage rack was on your left-hand side.", + "question_chinese": "现在灰色置物架在我的左手边还是右手边？", + "answer_chinese": "你刚才洗东西的时候，灰色置物架在你的左手边。", + "evidence": [ + { + "description": "When I were washing things just now, the gray shelf was on my left side.", + "timestamps": { + "start": "00:20:18", + "end": "00:24:05" + } + }, + { + "description": "When I were washing things just now, the gray shelf was on my left side.", + "timestamps": { + "start": "00:26:03", + "end": "00:26:33" + } + } + ], + "question_timestamps": "00:27:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were washing things just now, the gray storage rack was on your right-hand side.", + "B": "When you were washing things just now, the gray storage rack was on your left-hand side." + }, + "answer_label": "B", + "answer_text": "When you were washing things just now, the gray storage rack was on your left-hand side.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_5_T2", + "correct_answer": "When you were washing things just now, the gray storage rack was on your left-hand side.", + "original_options": { + "A": "When you were washing things just now, the gray storage rack was on your right-hand side.", + "B": "When you were washing things just now, the gray storage rack was on your left-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_5_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the gray storage rack on my left-hand side or my right-hand side right now?", + "answer": "The gray storage rack is on your right-hand side.", + "question_chinese": "现在灰色置物架在我的左手边还是右手边？", + "answer_chinese": "灰色置物架在你的右手边。", + "evidence": [ + { + "description": "When I were washing things just now, the gray shelf was on my left side.", + "timestamps": { + "start": "00:20:18", + "end": "00:24:05" + } + }, + { + "description": "When I were washing things just now, the gray shelf was on my left side.", + "timestamps": { + "start": "00:26:03", + "end": "00:26:33" + } + }, + { + "description": "The grey storage rack is on your right hand side.", + "timestamps": { + "start": "00:27:10", + "end": "00:29:49" + } + } + ], + "question_timestamps": "00:29:53", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The gray storage rack is on your left-hand side.", + "B": "The gray storage rack is on your right-hand side." + }, + "answer_label": "B", + "answer_text": "The gray storage rack is on your right-hand side.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_5_T1", + "correct_answer": "The gray storage rack is on your right-hand side.", + "original_options": { + "A": "The gray storage rack is on your left-hand side.", + "B": "The gray storage rack is on your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_4", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of seasoning have I opened so far?", + "answer": "So far, you have opened a bottle of seasoning.", + "question_chinese": "目前为止，我开了几瓶调味料？", + "answer_chinese": "目前为止，你开了一瓶调味料。", + "evidence": [ + { + "description": "I opened the first bottle of seasoning.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:16" + } + } + ], + "question_timestamps": "00:06:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "So far, you have opened none.", + "B": "So far, you have opened a bottle of seasoning.", + "C": "So far, you have opened two bottles.", + "D": "So far, you have opened three bottles of seasoning, not just a single bottle.", + "E": "So far, you have opened four bottles of seasoning, meaning you have unsealed several different containers." + }, + "answer_label": "B", + "answer_text": "So far, you have opened a bottle of seasoning.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_4", + "correct_answer": "So far, you have opened a bottle of seasoning.", + "original_options": { + "A": "So far, you have opened none.", + "B": "So far, you have opened a bottle of seasoning.", + "C": "So far, you have opened two bottles.", + "D": "So far, you have opened three bottles of seasoning, not just a single bottle.", + "E": "So far, you have opened four bottles of seasoning, meaning you have unsealed several different containers." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 16 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_5", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of seasoning have I opened so far?", + "answer": "So far, I have opened two bottles of seasoning.", + "question_chinese": "目前为止，我开了几瓶调味料？", + "answer_chinese": "目前为止，我开了2瓶调味料。", + "evidence": [ + { + "description": "I opened the first bottle of seasoning.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:16" + } + }, + { + "description": "I opened the second bottle of seasoning.", + "timestamps": { + "start": "00:06:37", + "end": "00:06:41" + } + } + ], + "question_timestamps": "00:06:44", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "So far, I have opened one bottle.", + "B": "So far, I have opened three bottles.", + "C": "So far, I have opened two bottles of seasoning.", + "D": "So far, I have opened four bottles of seasoning, which is more than the two you opened.", + "E": "So far, I have opened five bottles of seasoning, so the number is higher than a pair." + }, + "answer_label": "C", + "answer_text": "So far, I have opened two bottles of seasoning.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_5", + "correct_answer": "So far, I have opened two bottles of seasoning.", + "original_options": { + "A": "So far, I have opened one bottle.", + "B": "So far, I have opened three bottles.", + "C": "So far, I have opened two bottles of seasoning.", + "D": "So far, I have opened four bottles of seasoning, which is more than the two you opened.", + "E": "So far, I have opened five bottles of seasoning, so the number is higher than a pair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_7", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoons have I washed so far?", + "answer": "So far, you have washed one spoon.", + "question_chinese": "目前为止，我洗了几把勺子？", + "answer_chinese": "目前为止，你洗了1把勺子。", + "evidence": [ + { + "description": "I'm washing the first spoon.", + "timestamps": { + "start": "00:20:20", + "end": "00:20:30" + } + } + ], + "question_timestamps": "00:20:32", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "So far, you have washed two spoons.", + "B": "So far, you have washed three spoons.", + "C": "So far, you have washed four spoons, with each one rinsed and cleaned before setting it aside.", + "D": "So far, you have washed five spoons, finishing them one by one before moving on to anything else.", + "E": "So far, you have washed one spoon." + }, + "answer_label": "E", + "answer_text": "So far, you have washed one spoon.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_7", + "correct_answer": "So far, you have washed one spoon.", + "original_options": { + "A": "So far, you have washed two spoons.", + "B": "So far, you have washed three spoons.", + "C": "So far, you have washed four spoons, with each one rinsed and cleaned before setting it aside.", + "D": "So far, you have washed five spoons, finishing them one by one before moving on to anything else.", + "E": "So far, you have washed one spoon." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_8", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoons have I washed so far?", + "answer": "So far, you have washed two spoons.", + "question_chinese": "目前为止，我洗了几把勺子？", + "answer_chinese": "目前为止，你洗了2把勺子。", + "evidence": [ + { + "description": "I'm washing the first spoon.", + "timestamps": { + "start": "00:20:20", + "end": "00:20:30" + } + }, + { + "description": "I'm washing the second spoon.", + "timestamps": { + "start": "00:20:31", + "end": "00:20:38" + } + } + ], + "question_timestamps": "00:20:42", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "So far, you have washed zero spoons.", + "B": "So far, you have washed one spoon.", + "C": "So far, you have washed three spoons, cleaning each thoroughly before putting it down.", + "D": "So far, you have washed four spoons, continuing the same wash-and-rinse routine for each.", + "E": "So far, you have washed two spoons." + }, + "answer_label": "E", + "answer_text": "So far, you have washed two spoons.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_8", + "correct_answer": "So far, you have washed two spoons.", + "original_options": { + "A": "So far, you have washed zero spoons.", + "B": "So far, you have washed one spoon.", + "C": "So far, you have washed three spoons, cleaning each thoroughly before putting it down.", + "D": "So far, you have washed four spoons, continuing the same wash-and-rinse routine for each.", + "E": "So far, you have washed two spoons." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 15 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_3", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many things have I used to open bottle caps?", + "answer": "So far, you have used two things to open the bottle cap.", + "question_chinese": "目前为止，我用了几样东西开瓶盖？", + "answer_chinese": "目前为止，你用了两样东西开瓶盖。", + "evidence": [ + { + "description": "This is the first thing you've used.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:33" + } + }, + { + "description": "This is the second thing you've used.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:00:41", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "So far, you have used one thing.", + "B": "So far, you have used three things.", + "C": "So far, you have used four things to open the bottle cap, switching tools after each attempt.", + "D": "So far, you have used two things to open the bottle cap.", + "E": "So far, you have used five things to open the bottle cap, trying multiple options before it finally worked." + }, + "answer_label": "D", + "answer_text": "So far, you have used two things to open the bottle cap.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_3", + "correct_answer": "So far, you have used two things to open the bottle cap.", + "original_options": { + "A": "So far, you have used one thing.", + "B": "So far, you have used three things.", + "C": "So far, you have used four things to open the bottle cap, switching tools after each attempt.", + "D": "So far, you have used two things to open the bottle cap.", + "E": "So far, you have used five things to open the bottle cap, trying multiple options before it finally worked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 19 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_2", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many things have I used to open bottle caps?", + "answer": "So far, you have used three things to open bottle caps.", + "question_chinese": "目前为止，我用了几样东西开瓶盖？", + "answer_chinese": "目前为止，你用了三样东西开瓶盖。", + "evidence": [ + { + "description": "This is the first thing you've used.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:33" + } + }, + { + "description": "This is the second thing you've used", + "timestamps": { + "start": "00:00:35", + "end": "00:00:39" + } + }, + { + "description": "This is the third thing you've used.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "So far, you have used two things.", + "B": "So far, you have used four things.", + "C": "So far, you have used five things to open bottle caps, changing what you used across different tries.", + "D": "So far, you have used six things to open bottle caps, cycling through several options before succeeding.", + "E": "So far, you have used three things to open bottle caps." + }, + "answer_label": "E", + "answer_text": "So far, you have used three things to open bottle caps.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_2", + "correct_answer": "So far, you have used three things to open bottle caps.", + "original_options": { + "A": "So far, you have used two things.", + "B": "So far, you have used four things.", + "C": "So far, you have used five things to open bottle caps, changing what you used across different tries.", + "D": "So far, you have used six things to open bottle caps, cycling through several options before succeeding.", + "E": "So far, you have used three things to open bottle caps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 17 + ] + }, + { + "video_uid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751", + "video_length": "00:44:09", + "qid": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_1", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many things have I used to open bottle caps?", + "answer": "So far, you have used four things to open bottle caps.", + "question_chinese": "目前为止，我用了几样东西开瓶盖？", + "answer_chinese": "目前为止，你用了四样东西开瓶盖。", + "evidence": [ + { + "description": "This is the first thing you've used.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:33" + } + }, + { + "description": "This is the second thing you've used.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:39" + } + }, + { + "description": "This is the third thing you've used.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:50" + } + }, + { + "description": "This is the fourth thing you've used.", + "timestamps": { + "start": "00:00:52", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "So far, you have used four things to open bottle caps.", + "B": "So far, you have used three things.", + "C": "So far, you have used five things.", + "D": "So far, you have used six things to open bottle caps, using a different item each time you tried.", + "E": "So far, you have used seven things to open bottle caps, attempting several alternatives before getting them open." + }, + "answer_label": "A", + "answer_text": "So far, you have used four things to open bottle caps.", + "q_id": "18c05110-1ef1-4da2-8b59-c80bf0cc9751_1", + "correct_answer": "So far, you have used four things to open bottle caps.", + "original_options": { + "A": "So far, you have used four things to open bottle caps.", + "B": "So far, you have used three things.", + "C": "So far, you have used five things.", + "D": "So far, you have used six things to open bottle caps, using a different item each time you tried.", + "E": "So far, you have used seven things to open bottle caps, attempting several alternatives before getting them open." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 18 + ] + } +] diff --git a/QAs/18c12e08-e269-4ba7-bb5c-095bcaf0537c-mcq.json b/QAs/18c12e08-e269-4ba7-bb5c-095bcaf0537c-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..be55d65d3c1a8bde4a9ba5c6b129bf5355f5d38c --- /dev/null +++ b/QAs/18c12e08-e269-4ba7-bb5c-095bcaf0537c-mcq.json @@ -0,0 +1,1136 @@ +[ + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_6_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many plates have I washed?", + "answer": "You washed three plates, and after washing them you put them on your right.", + "question_chinese": "到目前为止，我已经洗到了多少个盘子？", + "answer_chinese": "你洗了三个盘子，你洗好把他们放在右边。", + "evidence": [ + { + "description": "First", + "timestamps": { + "start": "00:04:00", + "end": "00:04:16" + } + }, + { + "description": "The second and third", + "timestamps": { + "start": "00:08:23", + "end": "00:09:48" + } + } + ], + "question_timestamps": "00:09:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed two plates.", + "B": "You washed four plates.", + "C": "You washed one plate, and after washing it you put it on your right.", + "D": "You washed three plates, and after washing them you put them on your right.", + "E": "You washed five plates, and after washing them you put them on your right." + }, + "answer_label": "D", + "answer_text": "You washed three plates, and after washing them you put them on your right.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_6_T2", + "correct_answer": "You washed three plates, and after washing them you put them on your right.", + "original_options": { + "A": "You washed two plates.", + "B": "You washed four plates.", + "C": "You washed one plate, and after washing it you put it on your right.", + "D": "You washed three plates, and after washing them you put them on your right.", + "E": "You washed five plates, and after washing them you put them on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_6_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many plates have I washed?", + "answer": "You've already washed six plates; they're all to your right.", + "question_chinese": "到目前为止，我已经洗到了多少个盘子？", + "answer_chinese": "你已经洗了六个盘子，它们都在你的右边。", + "evidence": [ + { + "description": "First", + "timestamps": { + "start": "00:04:00", + "end": "00:04:16" + } + }, + { + "description": "The second and third", + "timestamps": { + "start": "00:08:23", + "end": "00:09:48" + } + }, + { + "description": "The fourth and fifth", + "timestamps": { + "start": "00:11:26", + "end": "00:12:35" + } + }, + { + "description": "The sixth", + "timestamps": { + "start": "00:12:42", + "end": "00:13:01" + } + } + ], + "question_timestamps": "00:13:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You've already washed four plates.", + "B": "You've already washed five plates.", + "C": "You've already washed six plates; they're all to your right.", + "D": "You've already washed seven plates; they're all stacked to your right on the counter.", + "E": "You've already washed eight plates; they're all placed to your right after you finish rinsing them." + }, + "answer_label": "C", + "answer_text": "You've already washed six plates; they're all to your right.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_6_T1", + "correct_answer": "You've already washed six plates; they're all to your right.", + "original_options": { + "A": "You've already washed four plates.", + "B": "You've already washed five plates.", + "C": "You've already washed six plates; they're all to your right.", + "D": "You've already washed seven plates; they're all stacked to your right on the counter.", + "E": "You've already washed eight plates; they're all placed to your right after you finish rinsing them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_11", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates have I washed so far?", + "answer": "You have washed six plates.", + "question_chinese": "到目前为止，我已经洗到了多少个盘子？", + "answer_chinese": "你已经洗了六个盘子。", + "evidence": [ + { + "description": "First", + "timestamps": { + "start": "00:04:00", + "end": "00:04:16" + } + }, + { + "description": "The second and third", + "timestamps": { + "start": "00:08:23", + "end": "00:09:48" + } + }, + { + "description": "The fourth and fifth", + "timestamps": { + "start": "00:11:26", + "end": "00:12:35" + } + }, + { + "description": "The sixth", + "timestamps": { + "start": "00:12:42", + "end": "00:13:01" + } + } + ], + "question_timestamps": "00:24:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have washed seven plates.", + "B": "You have washed eight plates.", + "C": "You have washed six plates.", + "D": "You have washed nine plates, finishing them before moving on to the next task.", + "E": "You have washed ten plates, completing the washing for each plate in the set." + }, + "answer_label": "C", + "answer_text": "You have washed six plates.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_11", + "correct_answer": "You have washed six plates.", + "original_options": { + "A": "You have washed seven plates.", + "B": "You have washed eight plates.", + "C": "You have washed six plates.", + "D": "You have washed nine plates, finishing them before moving on to the next task.", + "E": "You have washed ten plates, completing the washing for each plate in the set." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_12", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates have I washed so far?", + "answer": "You have washed six plates.", + "question_chinese": "到目前为止，我已经洗到了多少个盘子？", + "answer_chinese": "你以及洗了六个盘子。", + "evidence": [ + { + "description": "First", + "timestamps": { + "start": "00:04:00", + "end": "00:04:16" + } + }, + { + "description": "The second and third", + "timestamps": { + "start": "00:08:23", + "end": "00:09:48" + } + }, + { + "description": "The fourth and fifth", + "timestamps": { + "start": "00:11:26", + "end": "00:12:35" + } + }, + { + "description": "The sixth", + "timestamps": { + "start": "00:12:42", + "end": "00:13:01" + } + } + ], + "question_timestamps": "00:31:11", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have washed five plates.", + "B": "You have washed seven plates.", + "C": "You have washed four plates so far, finishing the last one just before you moved away from the sink.", + "D": "You have washed six plates.", + "E": "You have washed eight plates in total up to now, counting the ones you rinsed and set aside to dry." + }, + "answer_label": "D", + "answer_text": "You have washed six plates.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_12", + "correct_answer": "You have washed six plates.", + "original_options": { + "A": "You have washed five plates.", + "B": "You have washed seven plates.", + "C": "You have washed four plates so far, finishing the last one just before you moved away from the sink.", + "D": "You have washed six plates.", + "E": "You have washed eight plates in total up to now, counting the ones you rinsed and set aside to dry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 20 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_4_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to now, how many times have I stopped at or passed by the computer?", + "answer": "You have stopped in front of the computer or passed by it three times; the last time, the computer was behind you.", + "question_chinese": "到目前为止，我有多少次停下来或经过电脑呢？", + "answer_chinese": "你已经停在电脑前或经过电脑三次了，最后一次电脑在你的后面。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:00:21", + "end": "00:00:30" + } + }, + { + "description": "Second time", + "timestamps": { + "start": "00:01:12", + "end": "00:01:20" + } + }, + { + "description": "Third time", + "timestamps": { + "start": "00:01:30", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:01:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have stopped in front of the computer or passed by it three times; the last time, the computer was behind you.", + "B": "You passed the computer twice.", + "C": "You passed the computer four times.", + "D": "You stopped or passed it twice; the last time, the computer was in front of you.", + "E": "You stopped or passed it five times; the last time, the computer was to your left." + }, + "answer_label": "A", + "answer_text": "You have stopped in front of the computer or passed by it three times; the last time, the computer was behind you.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_4_T1", + "correct_answer": "You have stopped in front of the computer or passed by it three times; the last time, the computer was behind you.", + "original_options": { + "A": "You have stopped in front of the computer or passed by it three times; the last time, the computer was behind you.", + "B": "You passed the computer twice.", + "C": "You passed the computer four times.", + "D": "You stopped or passed it twice; the last time, the computer was in front of you.", + "E": "You stopped or passed it five times; the last time, the computer was to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_4_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to now, how many times have I stopped at or passed by the computer?", + "answer": "You have stopped in front of the computer or passed by it four times; the last time, the computer was in front of you.", + "question_chinese": "到目前为止，我有多少次停下来或经过电脑呢？", + "answer_chinese": "你已经停在电脑前或经过电脑四次了，最后一次电脑在你的前面。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:00:21", + "end": "00:00:30" + } + }, + { + "description": "Second time", + "timestamps": { + "start": "00:01:12", + "end": "00:01:20" + } + }, + { + "description": "Third time", + "timestamps": { + "start": "00:01:33", + "end": "00:01:40" + } + }, + { + "description": "Fourth time", + "timestamps": { + "start": "00:01:48", + "end": "00:02:41" + } + } + ], + "question_timestamps": "00:02:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You passed the computer three times.", + "B": "You have stopped in front of the computer or passed by it four times; the last time, the computer was in front of you.", + "C": "You passed the computer five times.", + "D": "You have stopped in front of the computer or passed by it six times; the last time, it was behind you.", + "E": "You have stopped in front of the computer or passed by it two times; the last time, it was to your right." + }, + "answer_label": "B", + "answer_text": "You have stopped in front of the computer or passed by it four times; the last time, the computer was in front of you.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_4_T2", + "correct_answer": "You have stopped in front of the computer or passed by it four times; the last time, the computer was in front of you.", + "original_options": { + "A": "You passed the computer three times.", + "B": "You have stopped in front of the computer or passed by it four times; the last time, the computer was in front of you.", + "C": "You passed the computer five times.", + "D": "You have stopped in front of the computer or passed by it six times; the last time, it was behind you.", + "E": "You have stopped in front of the computer or passed by it two times; the last time, it was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 22 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_7", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I stopped or passed by the computer so far?", + "answer": "You have stopped in front of or passed by the computer four times.", + "question_chinese": "到目前为止，我有多少次停下来或经过电脑呢？", + "answer_chinese": "你已经停在电脑前或经过电脑四次了。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:00:21", + "end": "00:00:30" + } + }, + { + "description": "Second time", + "timestamps": { + "start": "00:01:12", + "end": "00:01:20" + } + }, + { + "description": "Third time", + "timestamps": { + "start": "00:01:33", + "end": "00:01:40" + } + }, + { + "description": "Fourth time", + "timestamps": { + "start": "00:01:48", + "end": "00:02:41" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You passed it three times.", + "B": "You passed it five times.", + "C": "You have stopped in front of or passed by the computer two times since you started moving around the area.", + "D": "You have stopped in front of or passed by the computer six times in total, including quick walk-bys without stopping.", + "E": "You have stopped in front of or passed by the computer four times." + }, + "answer_label": "E", + "answer_text": "You have stopped in front of or passed by the computer four times.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_7", + "correct_answer": "You have stopped in front of or passed by the computer four times.", + "original_options": { + "A": "You passed it three times.", + "B": "You passed it five times.", + "C": "You have stopped in front of or passed by the computer two times since you started moving around the area.", + "D": "You have stopped in front of or passed by the computer six times in total, including quick walk-bys without stopping.", + "E": "You have stopped in front of or passed by the computer four times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 20 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_8", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I stopped or passed by the computer so far?", + "answer": "You have stopped in front of or passed by the computer four times.", + "question_chinese": "到目前为止，我有多少次停下来或经过电脑呢？", + "answer_chinese": "你已经停在电脑前或经过电脑四次了。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:00:21", + "end": "00:00:30" + } + }, + { + "description": "Second time", + "timestamps": { + "start": "00:01:12", + "end": "00:01:20" + } + }, + { + "description": "Third time", + "timestamps": { + "start": "00:01:33", + "end": "00:01:40" + } + }, + { + "description": "Fourth time", + "timestamps": { + "start": "00:01:48", + "end": "00:02:41" + } + } + ], + "question_timestamps": "00:15:51", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have stopped in front of or passed by the computer four times.", + "B": "You passed it three times.", + "C": "You passed it five times.", + "D": "You have stopped in front of or passed by the computer two times so far during your movements in the room.", + "E": "You have stopped in front of or passed by the computer six times up to now, counting each brief pass as one." + }, + "answer_label": "A", + "answer_text": "You have stopped in front of or passed by the computer four times.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_8", + "correct_answer": "You have stopped in front of or passed by the computer four times.", + "original_options": { + "A": "You have stopped in front of or passed by the computer four times.", + "B": "You passed it three times.", + "C": "You passed it five times.", + "D": "You have stopped in front of or passed by the computer two times so far during your movements in the room.", + "E": "You have stopped in front of or passed by the computer six times up to now, counting each brief pass as one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 22 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_16", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the microwave or the frying pan?", + "answer": "The microwave is closer to you.", + "question_chinese": "微波炉和平底锅哪个离我更近？", + "answer_chinese": "微波炉离你更近。", + "evidence": [ + { + "description": "The frying pan is in the corner of the cabinet.", + "timestamps": { + "start": "00:07:38", + "end": "00:07:53" + } + }, + { + "description": "The microwave is on the edge of the cabinet near the computer.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:25" + } + } + ], + "question_timestamps": "00:31:11", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The frying pan is closer to you.", + "B": "The microwave is closer to you." + }, + "answer_label": "B", + "answer_text": "The microwave is closer to you.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_16", + "correct_answer": "The microwave is closer to you.", + "original_options": { + "A": "The frying pan is closer to you.", + "B": "The microwave is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_10", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to play computer games now. How should I go about it?", + "answer": "You should turn left, walk straight to the first wooden table, sit down, and play on the computer.", + "question_chinese": "我现在想去玩电脑，我该怎么走？", + "answer_chinese": "你应该向左转，直走到第一个木桌子前坐下，玩电脑。", + "evidence": [ + { + "description": "You leave in front of the computer, turn right, and walk straight to the faucet.", + "timestamps": { + "start": "00:03:15", + "end": "00:03:28" + } + } + ], + "question_timestamps": "00:15:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right and sit at the table.", + "B": "Walk straight and sit down.", + "C": "Turn left, pass the first wooden table, sit at the second table, and play on the computer.", + "D": "Turn left, walk to the first wooden table, stand by it, and play without sitting down.", + "E": "You should turn left, walk straight to the first wooden table, sit down, and play on the computer." + }, + "answer_label": "E", + "answer_text": "You should turn left, walk straight to the first wooden table, sit down, and play on the computer.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_10", + "correct_answer": "You should turn left, walk straight to the first wooden table, sit down, and play on the computer.", + "original_options": { + "A": "Turn right and sit at the table.", + "B": "Walk straight and sit down.", + "C": "Turn left, pass the first wooden table, sit at the second table, and play on the computer.", + "D": "Turn left, walk to the first wooden table, stand by it, and play without sitting down.", + "E": "You should turn left, walk straight to the first wooden table, sit down, and play on the computer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 17, + 16 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_9", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to play computer games now. How should I go about it?", + "answer": "You should turn left and walk straight to the computer.", + "question_chinese": "我现在想去玩电脑，我该怎么走？", + "answer_chinese": "你应该向左转，直走到电脑旁", + "evidence": [ + { + "description": "The computer is now on the table to your left.", + "timestamps": { + "start": "00:20:05", + "end": "00:20:32" + } + } + ], + "question_timestamps": "00:23:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right to the computer.", + "B": "Turn left to the nearest table.", + "C": "Turn left, walk straight past the computer, then stop and play from the far side.", + "D": "Turn left, walk straight until you reach the wall, then turn around to use the computer.", + "E": "You should turn left and walk straight to the computer." + }, + "answer_label": "E", + "answer_text": "You should turn left and walk straight to the computer.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_9", + "correct_answer": "You should turn left and walk straight to the computer.", + "original_options": { + "A": "Turn right to the computer.", + "B": "Turn left to the nearest table.", + "C": "Turn left, walk straight past the computer, then stop and play from the far side.", + "D": "Turn left, walk straight until you reach the wall, then turn around to use the computer.", + "E": "You should turn left and walk straight to the computer." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_5", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Based on my position, which direction is my computer in?", + "answer": "Your computer is to the left of you.", + "question_chinese": "根据我站的位置，我的电脑在我的哪个方向？", + "answer_chinese": "你的电脑在你的左方。", + "evidence": [ + { + "description": "The computer is to your left.", + "timestamps": { + "start": "00:20:05", + "end": "00:20:42" + } + } + ], + "question_timestamps": "00:23:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is to your right.", + "B": "It is directly ahead.", + "C": "Your computer is to the left of you.", + "D": "Your computer is behind you, so you would need to turn around to face it.", + "E": "Your computer is on your right side near the table, not on the left." + }, + "answer_label": "C", + "answer_text": "Your computer is to the left of you.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_5", + "correct_answer": "Your computer is to the left of you.", + "original_options": { + "A": "It is to your right.", + "B": "It is directly ahead.", + "C": "Your computer is to the left of you.", + "D": "Your computer is behind you, so you would need to turn around to face it.", + "E": "Your computer is on your right side near the table, not on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 14 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_6", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Based on my position, which direction is my computer in?", + "answer": "The computer is to your left.", + "question_chinese": "根据我站的位置，我的电脑在我的哪个方向？", + "answer_chinese": "电脑在你的左方。", + "evidence": [ + { + "description": "The computer is to your left.", + "timestamps": { + "start": "00:20:05", + "end": "00:20:42" + } + } + ], + "question_timestamps": "00:31:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The computer is to your right.", + "B": "The computer is straight ahead.", + "C": "The computer is to your left.", + "D": "The computer is behind you, so you need to turn around to reach it.", + "E": "The computer is on your right side near the desk rather than on your left." + }, + "answer_label": "C", + "answer_text": "The computer is to your left.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_6", + "correct_answer": "The computer is to your left.", + "original_options": { + "A": "The computer is to your right.", + "B": "The computer is straight ahead.", + "C": "The computer is to your left.", + "D": "The computer is behind you, so you need to turn around to reach it.", + "E": "The computer is on your right side near the desk rather than on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 15 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_3", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the fire extinguisher?", + "answer": "The fire extinguisher is to the left of the microwave and is hung on the wall.", + "question_chinese": "灭火器在哪？", + "answer_chinese": "灭火器在微波炉左方，并且挂在墙上", + "evidence": [ + { + "description": "The fire extinguisher is to the left of the microwave and is hung on the wall.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is right of the microwave.", + "B": "The fire extinguisher is to the left of the microwave and is hung on the wall.", + "C": "It is above the microwave.", + "D": "The fire extinguisher is to the left of the microwave, but it is sitting on the counter, not hung up.", + "E": "The fire extinguisher is beside the microwave, mounted low near the baseboard instead of on the wall." + }, + "answer_label": "B", + "answer_text": "The fire extinguisher is to the left of the microwave and is hung on the wall.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_3", + "correct_answer": "The fire extinguisher is to the left of the microwave and is hung on the wall.", + "original_options": { + "A": "It is right of the microwave.", + "B": "The fire extinguisher is to the left of the microwave and is hung on the wall.", + "C": "It is above the microwave.", + "D": "The fire extinguisher is to the left of the microwave, but it is sitting on the counter, not hung up.", + "E": "The fire extinguisher is beside the microwave, mounted low near the baseboard instead of on the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 17 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_4", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the fire extinguisher?", + "answer": "The fire extinguisher is next to the microwave and hanging on the wall.", + "question_chinese": "灭火器在哪？", + "answer_chinese": "灭火器在微波炉旁边，并且挂在墙上", + "evidence": [ + { + "description": "The fire extinguisher is to the left of the microwave and is hung on the wall.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:15:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is next to the oven.", + "B": "It is by the fridge.", + "C": "The fire extinguisher is above the microwave, attached under a cabinet rather than hanging on the wall.", + "D": "The fire extinguisher is next to the microwave, but it is sitting on the counter instead of mounted.", + "E": "The fire extinguisher is next to the microwave and hanging on the wall." + }, + "answer_label": "E", + "answer_text": "The fire extinguisher is next to the microwave and hanging on the wall.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_4", + "correct_answer": "The fire extinguisher is next to the microwave and hanging on the wall.", + "original_options": { + "A": "It is next to the oven.", + "B": "It is by the fridge.", + "C": "The fire extinguisher is above the microwave, attached under a cabinet rather than hanging on the wall.", + "D": "The fire extinguisher is next to the microwave, but it is sitting on the counter instead of mounted.", + "E": "The fire extinguisher is next to the microwave and hanging on the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 18 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the faucet on or off now?", + "answer": "The faucet is left on.", + "question_chinese": "现在水龙头是开的还是关的？", + "answer_chinese": "水龙头是开着的。", + "evidence": [ + { + "description": "The faucet is left on.", + "timestamps": { + "start": "00:00:29", + "end": "00:01:13" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The faucet is left on.", + "B": "The faucet is turned off." + }, + "answer_label": "A", + "answer_text": "The faucet is left on.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_1", + "correct_answer": "The faucet is left on.", + "original_options": { + "A": "The faucet is left on.", + "B": "The faucet is turned off." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the faucet on or off now?", + "answer": "The faucet is turned off.", + "question_chinese": "现在水龙头是开的还是关的？", + "answer_chinese": "水龙头是关的。", + "evidence": [ + { + "description": "The faucet is turned off.", + "timestamps": { + "start": "00:14:50", + "end": "00:15:05" + } + } + ], + "question_timestamps": "00:15:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The faucet is turned off.", + "B": "The faucet is left on." + }, + "answer_label": "A", + "answer_text": "The faucet is turned off.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_2", + "correct_answer": "The faucet is turned off.", + "original_options": { + "A": "The faucet is turned off.", + "B": "The faucet is left on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_13", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the backpack?", + "answer": "The package is on a chair without a handrail near the staircase entrance on the first floor, but it is somewhat away from the staircase entrance.", + "question_chinese": "背包在哪里？", + "answer_chinese": "背包在一楼楼梯口旁边的离楼梯口较远的无扶手的椅子上。", + "evidence": [ + { + "description": "The package is on a chair without a handrail near the staircase entrance on the first floor, but it is somewhat away from the staircase entrance.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the floor by the staircase entrance.", + "B": "On a handrail chair near the stairs.", + "C": "The package is on a chair without a handrail near the staircase entrance on the first floor, but it is somewhat away from the staircase entrance.", + "D": "It is on a chair without a handrail right at the staircase entrance on the first floor.", + "E": "The package is on a chair without a handrail on the first floor, but it is far from the staircase entrance." + }, + "answer_label": "C", + "answer_text": "The package is on a chair without a handrail near the staircase entrance on the first floor, but it is somewhat away from the staircase entrance.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_13", + "correct_answer": "The package is on a chair without a handrail near the staircase entrance on the first floor, but it is somewhat away from the staircase entrance.", + "original_options": { + "A": "On the floor by the staircase entrance.", + "B": "On a handrail chair near the stairs.", + "C": "The package is on a chair without a handrail near the staircase entrance on the first floor, but it is somewhat away from the staircase entrance.", + "D": "It is on a chair without a handrail right at the staircase entrance on the first floor.", + "E": "The package is on a chair without a handrail on the first floor, but it is far from the staircase entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 21 + ] + }, + { + "video_uid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c", + "video_length": "00:31:12", + "qid": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_14", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the backpack?", + "answer": "The backpack is on a chair without a handrail near the staircase entrance on the first floor, and the chair is relatively far from the staircase entrance.", + "question_chinese": "背包在哪里？", + "answer_chinese": "背包在一楼楼梯口旁边的离楼梯口较远的无扶手的椅子上。", + "evidence": [ + { + "description": "The package is on a chair without a handrail near the staircase entrance on the first floor, but it is somewhat away from the staircase entrance.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:31:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On a chair beside the staircase entrance.", + "B": "On the floor near the staircase entrance.", + "C": "The backpack is on a chair without a handrail near the staircase entrance on the second floor.", + "D": "The backpack is on a chair with a handrail near the staircase entrance on the first floor, relatively far away.", + "E": "The backpack is on a chair without a handrail near the staircase entrance on the first floor, and the chair is relatively far from the staircase entrance." + }, + "answer_label": "E", + "answer_text": "The backpack is on a chair without a handrail near the staircase entrance on the first floor, and the chair is relatively far from the staircase entrance.", + "q_id": "18c12e08-e269-4ba7-bb5c-095bcaf0537c_14", + "correct_answer": "The backpack is on a chair without a handrail near the staircase entrance on the first floor, and the chair is relatively far from the staircase entrance.", + "original_options": { + "A": "On a chair beside the staircase entrance.", + "B": "On the floor near the staircase entrance.", + "C": "The backpack is on a chair without a handrail near the staircase entrance on the second floor.", + "D": "The backpack is on a chair with a handrail near the staircase entrance on the first floor, relatively far away.", + "E": "The backpack is on a chair without a handrail near the staircase entrance on the first floor, and the chair is relatively far from the staircase entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 20 + ] + } +] diff --git a/QAs/18e84829-901a-414d-8a2b-d1d2b3244db7-mcq.json b/QAs/18e84829-901a-414d-8a2b-d1d2b3244db7-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d95b6d737c6a43aa640740ad80e4d2c92f6e61f6 --- /dev/null +++ b/QAs/18e84829-901a-414d-8a2b-d1d2b3244db7-mcq.json @@ -0,0 +1,549 @@ +[ + { + "video_uid": "18e84829-901a-414d-8a2b-d1d2b3244db7", + "video_length": "00:30:00", + "qid": "18e84829-901a-414d-8a2b-d1d2b3244db7_10", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I entered through the red door just now, was the staircase on my left or right?", + "answer": "Just now, after you entered through the red door, the stairs were on your left.", + "question_chinese": "刚才我从红色的门进去后，楼梯在我的左边还是右边？", + "answer_chinese": "刚才你从红色的门进去后，楼梯在你的左边。", + "evidence": [ + { + "description": "After I entered through the red door, the stairs were on my left.", + "timestamps": { + "start": "00:25:09", + "end": "00:25:11" + } + } + ], + "question_timestamps": "00:27:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Just now, after you entered through the red door, the stairs were on your right.", + "B": "Just now, after you entered through the red door, the stairs were on your left." + }, + "answer_label": "B", + "answer_text": "Just now, after you entered through the red door, the stairs were on your left.", + "q_id": "18e84829-901a-414d-8a2b-d1d2b3244db7_10", + "correct_answer": "Just now, after you entered through the red door, the stairs were on your left.", + "original_options": { + "A": "Just now, after you entered through the red door, the stairs were on your right.", + "B": "Just now, after you entered through the red door, the stairs were on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "18e84829-901a-414d-8a2b-d1d2b3244db7", + "video_length": "00:30:00", + "qid": "18e84829-901a-414d-8a2b-d1d2b3244db7_15", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I entered through the red door just now, was the staircase on my left or right?", + "answer": "Just now, after you entered through the red door, the stairs were on your left.", + "question_chinese": "刚才我从红色的门进去后，楼梯在我的左边还是右边？", + "answer_chinese": "刚才你从红色的门进去后，楼梯在你的左边。", + "evidence": [ + { + "description": "After I entered through the red door, the stairs were on my left.", + "timestamps": { + "start": "00:25:09", + "end": "00:25:11" + } + } + ], + "question_timestamps": "00:27:56", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Just now, after you entered through the red door, the stairs were on your right.", + "B": "Just now, after you entered through the red door, the stairs were on your left." + }, + "answer_label": "B", + "answer_text": "Just now, after you entered through the red door, the stairs were on your left.", + "q_id": "18e84829-901a-414d-8a2b-d1d2b3244db7_15", + "correct_answer": "Just now, after you entered through the red door, the stairs were on your left.", + "original_options": { + "A": "Just now, after you entered through the red door, the stairs were on your right.", + "B": "Just now, after you entered through the red door, the stairs were on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "18e84829-901a-414d-8a2b-d1d2b3244db7", + "video_length": "00:30:00", + "qid": "18e84829-901a-414d-8a2b-d1d2b3244db7_1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the apron in the room?", + "answer": "The apron is hanging on the door across from the table.", + "question_chinese": "围裙在房间哪个地方？", + "answer_chinese": "围裙挂在桌子对面的门上。", + "evidence": [ + { + "description": "The apron is hanging on the door across from the table.", + "timestamps": { + "start": "00:03:01", + "end": "00:03:13" + } + }, + { + "description": "The apron is hanging on the door across from the table.", + "timestamps": { + "start": "00:10:29", + "end": "00:10:33" + } + } + ], + "question_timestamps": "00:15:01", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Hanging on the door beside the table.", + "B": "The apron is hanging on the door across from the table.", + "C": "Hanging on the chair next to the table.", + "D": "The apron is hanging on the wall next to the table, not on the door across from it.", + "E": "The apron is hanging on a cabinet near the table, rather than on the door directly opposite the table." + }, + "answer_label": "B", + "answer_text": "The apron is hanging on the door across from the table.", + "q_id": "18e84829-901a-414d-8a2b-d1d2b3244db7_1", + "correct_answer": "The apron is hanging on the door across from the table.", + "original_options": { + "A": "Hanging on the door beside the table.", + "B": "The apron is hanging on the door across from the table.", + "C": "Hanging on the chair next to the table.", + "D": "The apron is hanging on the wall next to the table, not on the door across from it.", + "E": "The apron is hanging on a cabinet near the table, rather than on the door directly opposite the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 19 + ] + }, + { + "video_uid": "18e84829-901a-414d-8a2b-d1d2b3244db7", + "video_length": "00:30:00", + "qid": "18e84829-901a-414d-8a2b-d1d2b3244db7_2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the apron in the room?", + "answer": "The apron is hanging on the door across from the table.", + "question_chinese": "围裙在房间哪个地方？", + "answer_chinese": "围裙挂在桌子对面的门上。", + "evidence": [ + { + "description": "The apron is hanging on the door across from the table.", + "timestamps": { + "start": "00:03:01", + "end": "00:03:13" + } + }, + { + "description": "The apron is hanging on the door across from the table.", + "timestamps": { + "start": "00:10:29", + "end": "00:10:33" + } + } + ], + "question_timestamps": "00:20:09", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Hanging on the door beside the table.", + "B": "Hanging on the wall near the table.", + "C": "The apron is hanging on the door next to the table, not on the door across from the table.", + "D": "The apron is hanging on a chair close to the table, instead of hanging on the opposite door.", + "E": "The apron is hanging on the door across from the table." + }, + "answer_label": "E", + "answer_text": "The apron is hanging on the door across from the table.", + "q_id": "18e84829-901a-414d-8a2b-d1d2b3244db7_2", + "correct_answer": "The apron is hanging on the door across from the table.", + "original_options": { + "A": "Hanging on the door beside the table.", + "B": "Hanging on the wall near the table.", + "C": "The apron is hanging on the door next to the table, not on the door across from the table.", + "D": "The apron is hanging on a chair close to the table, instead of hanging on the opposite door.", + "E": "The apron is hanging on the door across from the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 18 + ] + }, + { + "video_uid": "18e84829-901a-414d-8a2b-d1d2b3244db7", + "video_length": "00:30:00", + "qid": "18e84829-901a-414d-8a2b-d1d2b3244db7_4", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from my desk to my colleague's side?", + "answer": "You started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and you could see your colleague.", + "question_chinese": "我从工作台的桌子出发到同事身边，经历了什么路径？", + "answer_chinese": "你从工作台的桌子出发向前直走，到挂着围裙的门口左转就可以看见同事。", + "evidence": [ + { + "description": "I started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and I saw my colleague.", + "timestamps": { + "start": "00:10:28", + "end": "00:10:39" + } + } + ], + "question_timestamps": "00:14:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You turned right at the apron door.", + "B": "You went straight to your colleague.", + "C": "You started at the desk at the workbench, reached the door with an apron on, then turned right to find your colleague.", + "D": "From the desk at the workbench you walked to the apron door, then continued straight ahead until you reached your colleague.", + "E": "You started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and you could see your colleague." + }, + "answer_label": "E", + "answer_text": "You started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and you could see your colleague.", + "q_id": "18e84829-901a-414d-8a2b-d1d2b3244db7_4", + "correct_answer": "You started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and you could see your colleague.", + "original_options": { + "A": "You turned right at the apron door.", + "B": "You went straight to your colleague.", + "C": "You started at the desk at the workbench, reached the door with an apron on, then turned right to find your colleague.", + "D": "From the desk at the workbench you walked to the apron door, then continued straight ahead until you reached your colleague.", + "E": "You started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and you could see your colleague." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 22, + 21 + ] + }, + { + "video_uid": "18e84829-901a-414d-8a2b-d1d2b3244db7", + "video_length": "00:30:00", + "qid": "18e84829-901a-414d-8a2b-d1d2b3244db7_5", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from my desk to my colleague's side?", + "answer": "You started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and you saw your colleague.", + "question_chinese": "我从工作台的桌子出发到同事身边，经历了什么路径？", + "answer_chinese": "你从工作台的桌子出发向前直走，到挂着围裙的门口左转就可以看见同事。", + "evidence": [ + { + "description": "I started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and I saw my colleague.", + "timestamps": { + "start": "00:10:28", + "end": "00:10:39" + } + } + ], + "question_timestamps": "00:20:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You turned right at the apron door.", + "B": "You kept going straight past the door.", + "C": "You started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and you saw your colleague.", + "D": "You started at the workbench desk, went to the apron door, then turned right and saw your colleague.", + "E": "You began at the workbench desk, reached the apron door, then went straight ahead instead of turning left." + }, + "answer_label": "C", + "answer_text": "You started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and you saw your colleague.", + "q_id": "18e84829-901a-414d-8a2b-d1d2b3244db7_5", + "correct_answer": "You started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and you saw your colleague.", + "original_options": { + "A": "You turned right at the apron door.", + "B": "You kept going straight past the door.", + "C": "You started from the desk at the workbench and walked straight ahead to the door with an apron on, then turned left and you saw your colleague.", + "D": "You started at the workbench desk, went to the apron door, then turned right and saw your colleague.", + "E": "You began at the workbench desk, reached the apron door, then went straight ahead instead of turning left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "18e84829-901a-414d-8a2b-d1d2b3244db7", + "video_length": "00:30:00", + "qid": "18e84829-901a-414d-8a2b-d1d2b3244db7_12", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take when I left my work desk and went out?", + "answer": "You go straight ahead from the desk at work, turn left at the door with an apron hanging, and go straight to the door at the end to exit.", + "question_chinese": "我从工作的桌子出门，经历了什么路径？", + "answer_chinese": "你从工作的桌子旁边向前直走，到挂着围裙的门口左转，直走到尽头的门出去。", + "evidence": [ + { + "description": "I walked straight ahead from the desk at work, turned left at the door with an apron on.", + "timestamps": { + "start": "00:24:42", + "end": "00:24:52" + } + }, + { + "description": "I went straight to the door at the end to exit.", + "timestamps": { + "start": "00:24:45", + "end": "00:24:52" + } + } + ], + "question_timestamps": "00:27:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You go straight ahead from the desk at work, turn left at the door with an apron hanging, and go straight to the door at the end to exit.", + "B": "Turn right at the apron door.", + "C": "Exit through the side door.", + "D": "You head straight from the desk, turn right at the apron door, then continue straight to exit.", + "E": "You walk straight from the desk, turn left at a door with a coat hanging, and exit at the end." + }, + "answer_label": "A", + "answer_text": "You go straight ahead from the desk at work, turn left at the door with an apron hanging, and go straight to the door at the end to exit.", + "q_id": "18e84829-901a-414d-8a2b-d1d2b3244db7_12", + "correct_answer": "You go straight ahead from the desk at work, turn left at the door with an apron hanging, and go straight to the door at the end to exit.", + "original_options": { + "A": "You go straight ahead from the desk at work, turn left at the door with an apron hanging, and go straight to the door at the end to exit.", + "B": "Turn right at the apron door.", + "C": "Exit through the side door.", + "D": "You head straight from the desk, turn right at the apron door, then continue straight to exit.", + "E": "You walk straight from the desk, turn left at a door with a coat hanging, and exit at the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 20 + ] + }, + { + "video_uid": "18e84829-901a-414d-8a2b-d1d2b3244db7", + "video_length": "00:30:00", + "qid": "18e84829-901a-414d-8a2b-d1d2b3244db7_16", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take when I left my work desk and went out?", + "answer": "You go straight ahead from the desk at work, turn left at the door with an apron hanging, and go straight to the door at the end to exit.", + "question_chinese": "我从工作的桌子出门，经历了什么路径？", + "answer_chinese": "你从工作的桌子旁边向前直走，到挂着围裙的门口左转，直走到尽头的门出去。", + "evidence": [ + { + "description": "I walked straight ahead from the desk at work, turned left at the door with an apron on.", + "timestamps": { + "start": "00:24:42", + "end": "00:24:52" + } + }, + { + "description": "I went straight to the door at the end to exit.", + "timestamps": { + "start": "00:24:45", + "end": "00:24:52" + } + } + ], + "question_timestamps": "00:29:19", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right at the apron door.", + "B": "Turn left at the jacket door.", + "C": "You go straight ahead from the desk at work, turn left at the door with an apron hanging, and go straight to the door at the end to exit.", + "D": "You go straight from the desk, turn left at a door with a bag hanging, then exit at the end.", + "E": "You head straight from the desk, turn left at the apron door, then turn left again before exiting." + }, + "answer_label": "C", + "answer_text": "You go straight ahead from the desk at work, turn left at the door with an apron hanging, and go straight to the door at the end to exit.", + "q_id": "18e84829-901a-414d-8a2b-d1d2b3244db7_16", + "correct_answer": "You go straight ahead from the desk at work, turn left at the door with an apron hanging, and go straight to the door at the end to exit.", + "original_options": { + "A": "Turn right at the apron door.", + "B": "Turn left at the jacket door.", + "C": "You go straight ahead from the desk at work, turn left at the door with an apron hanging, and go straight to the door at the end to exit.", + "D": "You go straight from the desk, turn left at a door with a bag hanging, then exit at the end.", + "E": "You head straight from the desk, turn left at the apron door, then turn left again before exiting." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 18 + ] + }, + { + "video_uid": "18e84829-901a-414d-8a2b-d1d2b3244db7", + "video_length": "00:30:00", + "qid": "18e84829-901a-414d-8a2b-d1d2b3244db7_11", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the gloves when I was cutting the fabric ?", + "answer": "When you were cutting the fabric, you were able to get the gloves between the tape and the box containing scissors.", + "question_chinese": "我裁剪布料时，我能拿到手套吗？", + "answer_chinese": "你裁剪布料时，你能拿到手套，手套在胶带和装有剪刀的盒子中间。", + "evidence": [ + { + "description": "The gloves between the tape and the box containing scissors.", + "timestamps": { + "start": "00:21:32", + "end": "00:21:45" + } + }, + { + "description": "The glove is between the tape and the box containing scissors.Another glove is near the box containing scissors.", + "timestamps": { + "start": "00:21:48", + "end": "00:22:08" + } + } + ], + "question_timestamps": "00:25:11", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "When you were cutting the fabric, you were able to get the gloves between the tape and the box containing scissors.", + "B": "No, they were out of reach.", + "C": "Yes, but behind the scissors box.", + "D": "When you were cutting the fabric, you could reach the gloves, but they were on the other side of the scissors box.", + "E": "When you were cutting the fabric, you could get the gloves, but they were next to the tape rather than between items." + }, + "answer_label": "A", + "answer_text": "When you were cutting the fabric, you were able to get the gloves between the tape and the box containing scissors.", + "q_id": "18e84829-901a-414d-8a2b-d1d2b3244db7_11", + "correct_answer": "When you were cutting the fabric, you were able to get the gloves between the tape and the box containing scissors.", + "original_options": { + "A": "When you were cutting the fabric, you were able to get the gloves between the tape and the box containing scissors.", + "B": "No, they were out of reach.", + "C": "Yes, but behind the scissors box.", + "D": "When you were cutting the fabric, you could reach the gloves, but they were on the other side of the scissors box.", + "E": "When you were cutting the fabric, you could get the gloves, but they were next to the tape rather than between items." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 22 + ] + }, + { + "video_uid": "18e84829-901a-414d-8a2b-d1d2b3244db7", + "video_length": "00:30:00", + "qid": "18e84829-901a-414d-8a2b-d1d2b3244db7_7", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the gloves when I was cutting the fabric ?", + "answer": "When you were cutting the fabric, you were able to get the gloves between the tape and the box containing scissors.", + "question_chinese": "我裁剪布料时，我能拿到手套吗?", + "answer_chinese": "你裁剪布料时，你能拿到手套，手套在胶带和装有剪刀的盒子中间。", + "evidence": [ + { + "description": "The gloves between the tape and the box containing scissors.", + "timestamps": { + "start": "00:21:32", + "end": "00:21:45" + } + } + ], + "question_timestamps": "00:25:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "No, they were out of reach.", + "B": "Yes, but behind the tape.", + "C": "When you were cutting the fabric, you were able to get the gloves between the tape and the box containing scissors.", + "D": "When you were cutting the fabric, you could reach the gloves, but they were behind the scissors box.", + "E": "When you were cutting the fabric, you could get the gloves, but they were across from the scissors box." + }, + "answer_label": "C", + "answer_text": "When you were cutting the fabric, you were able to get the gloves between the tape and the box containing scissors.", + "q_id": "18e84829-901a-414d-8a2b-d1d2b3244db7_7", + "correct_answer": "When you were cutting the fabric, you were able to get the gloves between the tape and the box containing scissors.", + "original_options": { + "A": "No, they were out of reach.", + "B": "Yes, but behind the tape.", + "C": "When you were cutting the fabric, you were able to get the gloves between the tape and the box containing scissors.", + "D": "When you were cutting the fabric, you could reach the gloves, but they were behind the scissors box.", + "E": "When you were cutting the fabric, you could get the gloves, but they were across from the scissors box." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 19 + ] + } +] diff --git a/QAs/1a7e6155-3613-4f55-865c-00e196db0da0-mcq.json b/QAs/1a7e6155-3613-4f55-865c-00e196db0da0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9b713634ee667d626d14a69ec37d7a8e28d14876 --- /dev/null +++ b/QAs/1a7e6155-3613-4f55-865c-00e196db0da0-mcq.json @@ -0,0 +1,1371 @@ +[ + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_1_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the refrigerator on the left side of me or on the right side of me?", + "answer": "The refrigerator is on the left side of you.", + "question_chinese": "冰箱在我的左侧还是右侧？", + "answer_chinese": "冰箱在你的左侧。", + "evidence": [ + { + "description": "The refrigerator is on your left.", + "timestamps": { + "start": "00:27:56", + "end": "00:27:57" + } + } + ], + "question_timestamps": "00:27:59", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The refrigerator is on the right side of you.", + "B": "The refrigerator is on the left side of you." + }, + "answer_label": "B", + "answer_text": "The refrigerator is on the left side of you.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_1_T1", + "correct_answer": "The refrigerator is on the left side of you.", + "original_options": { + "A": "The refrigerator is on the right side of you.", + "B": "The refrigerator is on the left side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_1_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the refrigerator on the left side of me or on the right side of me?", + "answer": "The fridge is behind you.", + "question_chinese": "冰箱在我的左侧还是右侧？", + "answer_chinese": "冰箱在你的后方。", + "evidence": [ + { + "description": "The refrigerator is on behind of you.", + "timestamps": { + "start": "00:28:00", + "end": "00:28:03" + } + } + ], + "question_timestamps": "00:28:06", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The fridge is behind you.", + "B": "The refrigerator is on the left side of you." + }, + "answer_label": "A", + "answer_text": "The fridge is behind you.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_1_T2", + "correct_answer": "The fridge is behind you.", + "original_options": { + "A": "The fridge is behind you.", + "B": "The refrigerator is on the left side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_7", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the kitchen now relative to me?", + "answer": "The kitchen is now on your right.", + "question_chinese": "厨房现在相对我是什么位置？", + "answer_chinese": "厨房现在在你的右侧。", + "evidence": [ + { + "description": "The kitchen is now in front of you.", + "timestamps": { + "start": "00:23:43", + "end": "00:23:44" + } + } + ], + "question_timestamps": "00:23:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The kitchen is behind you.", + "B": "The kitchen is to your left.", + "C": "The kitchen is directly in front of you, past where you are currently standing.", + "D": "The kitchen is now on your right.", + "E": "The kitchen is below you, as if it were on a lower level than you." + }, + "answer_label": "D", + "answer_text": "The kitchen is now on your right.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_7", + "correct_answer": "The kitchen is now on your right.", + "original_options": { + "A": "The kitchen is behind you.", + "B": "The kitchen is to your left.", + "C": "The kitchen is directly in front of you, past where you are currently standing.", + "D": "The kitchen is now on your right.", + "E": "The kitchen is below you, as if it were on a lower level than you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 15 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_8", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the kitchen now relative to me?", + "answer": "The kitchen is now in front of you.", + "question_chinese": "厨房现在相对我是什么位置？", + "answer_chinese": "厨房现在在你的前方。", + "evidence": [ + { + "description": "The kitchen is now in front of you.", + "timestamps": { + "start": "00:23:43", + "end": "00:23:44" + } + } + ], + "question_timestamps": "00:23:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The kitchen is to your left.", + "B": "The kitchen is behind you.", + "C": "The kitchen is now in front of you.", + "D": "The kitchen is diagonally to your right, not straight ahead from where you are standing.", + "E": "The kitchen is directly beside you, positioned to the side rather than in front." + }, + "answer_label": "C", + "answer_text": "The kitchen is now in front of you.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_8", + "correct_answer": "The kitchen is now in front of you.", + "original_options": { + "A": "The kitchen is to your left.", + "B": "The kitchen is behind you.", + "C": "The kitchen is now in front of you.", + "D": "The kitchen is diagonally to your right, not straight ahead from where you are standing.", + "E": "The kitchen is directly beside you, positioned to the side rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 14 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_6_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which direction is the child relative to me?", + "answer": "The child is in front of you.", + "question_chinese": "小孩在我的什么方向？", + "answer_chinese": "小孩在你前方。", + "evidence": [ + { + "description": "You turned right immediately after leaving the restaurant.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The child is behind you.", + "B": "The child is to your left.", + "C": "The child is in front of you.", + "D": "The child is to your right, not directly ahead of you at all.", + "E": "The child is above you, so you must look upward to see them clearly." + }, + "answer_label": "C", + "answer_text": "The child is in front of you.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_6_T2", + "correct_answer": "The child is in front of you.", + "original_options": { + "A": "The child is behind you.", + "B": "The child is to your left.", + "C": "The child is in front of you.", + "D": "The child is to your right, not directly ahead of you at all.", + "E": "The child is above you, so you must look upward to see them clearly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 14 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_6_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which direction is the child relative to me?", + "answer": "The child is behind you.", + "question_chinese": "小孩在我的什么方向？", + "answer_chinese": "小孩在你后面。", + "evidence": [ + { + "description": "The kid is on behind of you.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The child is in front of you.", + "B": "The child is to your right.", + "C": "The child is to your left, not positioned behind you at all.", + "D": "The child is above you, so they are not behind you at all.", + "E": "The child is behind you." + }, + "answer_label": "E", + "answer_text": "The child is behind you.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_6_T1", + "correct_answer": "The child is behind you.", + "original_options": { + "A": "The child is in front of you.", + "B": "The child is to your right.", + "C": "The child is to your left, not positioned behind you at all.", + "D": "The child is above you, so they are not behind you at all.", + "E": "The child is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 13 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_5_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put my phone?", + "answer": "You just put the phone on the cup, and the phone is in front of you.", + "question_chinese": "我刚刚把手机放在了什么位置？", + "answer_chinese": "你刚刚把手机放在杯子上，手机在你前面。", + "evidence": [ + { + "description": "You just put your mobile phone on cup.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:39" + } + } + ], + "question_timestamps": "00:08:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the cup behind you.", + "B": "On the cup to your left.", + "C": "You just put the phone on the cup, and the phone is in front of you.", + "D": "You just put the phone on the cup, but it is positioned to your right.", + "E": "You just put the phone on the cup, and it is behind you rather than in front." + }, + "answer_label": "C", + "answer_text": "You just put the phone on the cup, and the phone is in front of you.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_5_T2", + "correct_answer": "You just put the phone on the cup, and the phone is in front of you.", + "original_options": { + "A": "On the cup behind you.", + "B": "On the cup to your left.", + "C": "You just put the phone on the cup, and the phone is in front of you.", + "D": "You just put the phone on the cup, but it is positioned to your right.", + "E": "You just put the phone on the cup, and it is behind you rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 17 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_5_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put my phone?", + "answer": "You just put the phone on the dining table, and the phone is in front of you.", + "question_chinese": "我刚刚把手机放在了什么位置？", + "answer_chinese": "你刚刚把手机放在了餐桌上,手机在你前面。", + "evidence": [ + { + "description": "You just put your mobile phone on the dining table.", + "timestamps": { + "start": "00:08:40", + "end": "00:08:41" + } + } + ], + "question_timestamps": "00:08:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the dining table behind you.", + "B": "On the dining table to your left.", + "C": "You just put the phone on the dining table, but it is located to your right.", + "D": "You just put the phone on the dining table, and it is above you rather than ahead.", + "E": "You just put the phone on the dining table, and the phone is in front of you." + }, + "answer_label": "E", + "answer_text": "You just put the phone on the dining table, and the phone is in front of you.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_5_T1", + "correct_answer": "You just put the phone on the dining table, and the phone is in front of you.", + "original_options": { + "A": "On the dining table behind you.", + "B": "On the dining table to your left.", + "C": "You just put the phone on the dining table, but it is located to your right.", + "D": "You just put the phone on the dining table, and it is above you rather than ahead.", + "E": "You just put the phone on the dining table, and the phone is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_10", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put my mobile phone?", + "answer": "You just put your mobile phone on the dining table.", + "question_chinese": "我刚刚把手机放在了什么位置？", + "answer_chinese": "你刚刚把手机放在了餐桌上。", + "evidence": [ + { + "description": "You just put your mobile phone on the dining table.", + "timestamps": { + "start": "00:08:40", + "end": "00:08:42" + } + } + ], + "question_timestamps": "00:27:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just put your mobile phone on the dining table.", + "B": "You put it on the coffee table.", + "C": "You put it in your backpack.", + "D": "You just placed your mobile phone under the couch instead of setting it on the dining table.", + "E": "You just left your mobile phone on the kitchen counter rather than on the dining table." + }, + "answer_label": "A", + "answer_text": "You just put your mobile phone on the dining table.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_10", + "correct_answer": "You just put your mobile phone on the dining table.", + "original_options": { + "A": "You just put your mobile phone on the dining table.", + "B": "You put it on the coffee table.", + "C": "You put it in your backpack.", + "D": "You just placed your mobile phone under the couch instead of setting it on the dining table.", + "E": "You just left your mobile phone on the kitchen counter rather than on the dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 16 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_8_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I just walked into the kitchen, how many times did I open the refrigerator?", + "answer": "You just went into the kitchen and opened the refrigerator 0 times; the refrigerator is to your left.", + "question_chinese": "我刚刚走进厨房打开了几次冰箱？", + "answer_chinese": "你刚刚走进厨房打开过零次冰箱，冰箱在你左边。", + "evidence": [ + { + "description": "You opened the refrigerator once.", + "timestamps": { + "start": "00:24:20", + "end": "00:24:36" + } + } + ], + "question_timestamps": "00:24:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just went into the kitchen and opened the refrigerator 0 times; the refrigerator is to your left.", + "B": "You opened it 1 time.", + "C": "You opened it 2 times.", + "D": "You just went into the kitchen and opened the refrigerator 3 times; the refrigerator is to your left.", + "E": "You just went into the kitchen and opened the refrigerator 4 times; the refrigerator is to your left." + }, + "answer_label": "A", + "answer_text": "You just went into the kitchen and opened the refrigerator 0 times; the refrigerator is to your left.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_8_T1", + "correct_answer": "You just went into the kitchen and opened the refrigerator 0 times; the refrigerator is to your left.", + "original_options": { + "A": "You just went into the kitchen and opened the refrigerator 0 times; the refrigerator is to your left.", + "B": "You opened it 1 time.", + "C": "You opened it 2 times.", + "D": "You just went into the kitchen and opened the refrigerator 3 times; the refrigerator is to your left.", + "E": "You just went into the kitchen and opened the refrigerator 4 times; the refrigerator is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_8_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I just walked into the kitchen, how many times did I open the refrigerator?", + "answer": "You just went into the kitchen and opened the refrigerator once; the refrigerator is to your left.", + "question_chinese": "我刚刚走进厨房打开了几次冰箱？", + "answer_chinese": "你刚刚走进厨房打开过一次冰箱，冰箱在你左边。", + "evidence": [ + { + "description": "You opened the refrigerator once.", + "timestamps": { + "start": "00:24:20", + "end": "00:24:36" + } + } + ], + "question_timestamps": "00:24:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened it 0 times.", + "B": "You just went into the kitchen and opened the refrigerator once; the refrigerator is to your left.", + "C": "You opened it 2 times.", + "D": "You just went into the kitchen and opened the refrigerator 3 times; the refrigerator is to your left.", + "E": "You just went into the kitchen and opened the refrigerator 4 times; the refrigerator is to your left." + }, + "answer_label": "B", + "answer_text": "You just went into the kitchen and opened the refrigerator once; the refrigerator is to your left.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_8_T2", + "correct_answer": "You just went into the kitchen and opened the refrigerator once; the refrigerator is to your left.", + "original_options": { + "A": "You opened it 0 times.", + "B": "You just went into the kitchen and opened the refrigerator once; the refrigerator is to your left.", + "C": "You opened it 2 times.", + "D": "You just went into the kitchen and opened the refrigerator 3 times; the refrigerator is to your left.", + "E": "You just went into the kitchen and opened the refrigerator 4 times; the refrigerator is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_15", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I just walked into the kitchen and opened the refrigerator?", + "answer": "You just walked into the kitchen and opened the refrigerator once.", + "question_chinese": "我刚刚走进厨房打开了几次冰箱？", + "answer_chinese": "你刚刚走进厨房打开过一次冰箱。", + "evidence": [ + { + "description": "You opened the refrigerator once.", + "timestamps": { + "start": "00:24:20", + "end": "00:24:36" + } + } + ], + "question_timestamps": "00:24:49", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You opened it twice.", + "B": "You opened it three times.", + "C": "You just walked into the kitchen and opened the refrigerator once.", + "D": "You just walked into the kitchen and opened the refrigerator six times.", + "E": "You just walked into the kitchen and opened the refrigerator eight times." + }, + "answer_label": "C", + "answer_text": "You just walked into the kitchen and opened the refrigerator once.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_15", + "correct_answer": "You just walked into the kitchen and opened the refrigerator once.", + "original_options": { + "A": "You opened it twice.", + "B": "You opened it three times.", + "C": "You just walked into the kitchen and opened the refrigerator once.", + "D": "You just walked into the kitchen and opened the refrigerator six times.", + "E": "You just walked into the kitchen and opened the refrigerator eight times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 12, + 12 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_16", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I just walked into the kitchen and opened the refrigerator?", + "answer": "You just walked into the kitchen and opened the refrigerator once.", + "question_chinese": "我刚刚走进厨房打开了几次冰箱？", + "answer_chinese": "你刚刚走进厨房打开过一次冰箱。", + "evidence": [ + { + "description": "You opened the refrigerator once.", + "timestamps": { + "start": "00:24:20", + "end": "00:24:36" + } + } + ], + "question_timestamps": "00:33:42", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Three times.", + "C": "You walked into the kitchen and opened the refrigerator twice, doing the full enter-and-open action both times.", + "D": "You walked into the kitchen and opened the refrigerator three times, repeating the same entry and opening sequence each time.", + "E": "You just walked into the kitchen and opened the refrigerator once." + }, + "answer_label": "E", + "answer_text": "You just walked into the kitchen and opened the refrigerator once.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_16", + "correct_answer": "You just walked into the kitchen and opened the refrigerator once.", + "original_options": { + "A": "Twice.", + "B": "Three times.", + "C": "You walked into the kitchen and opened the refrigerator twice, doing the full enter-and-open action both times.", + "D": "You walked into the kitchen and opened the refrigerator three times, repeating the same entry and opening sequence each time.", + "E": "You just walked into the kitchen and opened the refrigerator once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 18, + 20 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I just walked into the kitchen, was the refrigerator on my left or right?", + "answer": "The refrigerator is on your left.", + "question_chinese": "我刚刚走进厨房时，冰箱在我的左侧还是右侧？", + "answer_chinese": "冰箱在你的左侧。", + "evidence": [ + { + "description": "The refrigerator is on your left.", + "timestamps": { + "start": "00:27:56", + "end": "00:27:58" + } + } + ], + "question_timestamps": "00:27:59", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The refrigerator is on your right.", + "B": "The refrigerator is on your left." + }, + "answer_label": "B", + "answer_text": "The refrigerator is on your left.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_2", + "correct_answer": "The refrigerator is on your left.", + "original_options": { + "A": "The refrigerator is on your right.", + "B": "The refrigerator is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I just walked into the kitchen, was the refrigerator on my left or right?", + "answer": "The refrigerator is on your left.", + "question_chinese": "我刚刚走进厨房时，冰箱在我的左侧还是右侧？", + "answer_chinese": "冰箱在你的左侧。", + "evidence": [ + { + "description": "The refrigerator is on your left.", + "timestamps": { + "start": "00:27:56", + "end": "00:27:58" + } + } + ], + "question_timestamps": "00:28:03", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The refrigerator is on your left.", + "B": "The refrigerator is on your right." + }, + "answer_label": "A", + "answer_text": "The refrigerator is on your left.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_1", + "correct_answer": "The refrigerator is on your left.", + "original_options": { + "A": "The refrigerator is on your left.", + "B": "The refrigerator is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_11", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get to the kitchen to wash the plates?", + "answer": "You should come out of the restaurant first, turn right and then turn right again at the first door you encounter.", + "question_chinese": "我想把盘子拿到厨房清洗应该怎么走？", + "answer_chinese": "你应该先从餐厅出来，右转之后在遇到的第一个门时再次右转。", + "evidence": [ + { + "description": "You turned right immediately after leaving the restaurant.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:23" + } + }, + { + "description": "You turned right again and reached the kitchen.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:15:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Exit, turn left, then left again.", + "B": "You should come out of the restaurant first, turn right and then turn right again at the first door you encounter.", + "C": "Exit, turn left, then right.", + "D": "Come out of the restaurant, turn left, then take the first door you encounter on your right.", + "E": "Come out of the restaurant first, turn right, then turn left at the first door you come to." + }, + "answer_label": "B", + "answer_text": "You should come out of the restaurant first, turn right and then turn right again at the first door you encounter.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_11", + "correct_answer": "You should come out of the restaurant first, turn right and then turn right again at the first door you encounter.", + "original_options": { + "A": "Exit, turn left, then left again.", + "B": "You should come out of the restaurant first, turn right and then turn right again at the first door you encounter.", + "C": "Exit, turn left, then right.", + "D": "Come out of the restaurant, turn left, then take the first door you encounter on your right.", + "E": "Come out of the restaurant first, turn right, then turn left at the first door you come to." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 18 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_12", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get to the kitchen to wash the plates?", + "answer": "You should come out of the restaurant first, turn right and then turn right again at the first door you encounter.", + "question_chinese": "我想把盘子拿到厨房清洗应该怎么走？", + "answer_chinese": "你应该先从餐厅出来，右转之后在遇到的第一个门时再次右转。", + "evidence": [ + { + "description": "You turned right immediately after leaving the restaurant.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:23" + } + }, + { + "description": "You turned right again and reached the kitchen.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:45:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Exit, turn left, then left again.", + "B": "Exit, turn right, then left.", + "C": "Come out of the restaurant, turn left, then take the first door you encounter on your right.", + "D": "You should come out of the restaurant first, turn right and then turn right again at the first door you encounter.", + "E": "Come out of the restaurant first, turn right, then turn left at the first door you come to." + }, + "answer_label": "D", + "answer_text": "You should come out of the restaurant first, turn right and then turn right again at the first door you encounter.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_12", + "correct_answer": "You should come out of the restaurant first, turn right and then turn right again at the first door you encounter.", + "original_options": { + "A": "Exit, turn left, then left again.", + "B": "Exit, turn right, then left.", + "C": "Come out of the restaurant, turn left, then take the first door you encounter on your right.", + "D": "You should come out of the restaurant first, turn right and then turn right again at the first door you encounter.", + "E": "Come out of the restaurant first, turn right, then turn left at the first door you come to." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 18 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_7_T1", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I see myself in the mirror directly right now?", + "answer": "Right now you can’t see yourself in the mirror; the mirror is in front of you.", + "question_chinese": "我现在可以直接照到镜子吗？", + "answer_chinese": "你现在无法照到镜子，镜子在你前面。", + "evidence": [ + { + "description": "You are at a certain distance from the mirror, and the mirror is not directly in front of you.", + "timestamps": { + "start": "00:43:00", + "end": "00:43:02" + } + } + ], + "question_timestamps": "00:43:04", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can’t see yourself in the mirror; the mirror is in front of you.", + "B": "Right now you can see yourself in the mirror; the mirror is in front of you." + }, + "answer_label": "A", + "answer_text": "Right now you can’t see yourself in the mirror; the mirror is in front of you.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_7_T1", + "correct_answer": "Right now you can’t see yourself in the mirror; the mirror is in front of you.", + "original_options": { + "A": "Right now you can’t see yourself in the mirror; the mirror is in front of you.", + "B": "Right now you can see yourself in the mirror; the mirror is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_14", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I look directly into the mirror now?", + "answer": "You can't look directly in the mirror now.", + "question_chinese": "我现在可以直接照到镜子吗？", + "answer_chinese": "你现在无法直接照到镜子。", + "evidence": [ + { + "description": "You are at a certain distance from the mirror, and the mirror is not directly in front of you.", + "timestamps": { + "start": "00:43:00", + "end": "00:43:02" + } + } + ], + "question_timestamps": "00:43:05", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't look directly in the mirror now.", + "B": "You can look directly in the mirror now." + }, + "answer_label": "A", + "answer_text": "You can't look directly in the mirror now.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_14", + "correct_answer": "You can't look directly in the mirror now.", + "original_options": { + "A": "You can't look directly in the mirror now.", + "B": "You can look directly in the mirror now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_13", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I look directly into the mirror now?", + "answer": "You can't look directly in the mirror now.", + "question_chinese": "我现在可以直接照到镜子吗？", + "answer_chinese": "你现在无法直接照到镜子。", + "evidence": [ + { + "description": "You are at a certain distance from the mirror, and the mirror is not directly in front of you.", + "timestamps": { + "start": "00:43:00", + "end": "00:43:02" + } + } + ], + "question_timestamps": "00:43:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can look directly in the mirror now.", + "B": "You can't look directly in the mirror now." + }, + "answer_label": "B", + "answer_text": "You can't look directly in the mirror now.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_13", + "correct_answer": "You can't look directly in the mirror now.", + "original_options": { + "A": "You can look directly in the mirror now.", + "B": "You can't look directly in the mirror now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_7_T2", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I see myself in the mirror directly right now?", + "answer": "Right now you can see yourself in the mirror directly; the mirror is in front of you.", + "question_chinese": "我现在可以直接照到镜子吗？", + "answer_chinese": "你现在可以直接照到镜子，镜子在你前面。", + "evidence": [ + { + "description": "You are at a certain distance from the mirror, and the mirror is not directly in front of you.", + "timestamps": { + "start": "00:56:29", + "end": "00:56:31" + } + } + ], + "question_timestamps": "00:56:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can’t see yourself in the mirror directly; the mirror is behind you.", + "B": "Right now you can see yourself in the mirror directly; the mirror is in front of you." + }, + "answer_label": "B", + "answer_text": "Right now you can see yourself in the mirror directly; the mirror is in front of you.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_7_T2", + "correct_answer": "Right now you can see yourself in the mirror directly; the mirror is in front of you.", + "original_options": { + "A": "Right now you can’t see yourself in the mirror directly; the mirror is behind you.", + "B": "Right now you can see yourself in the mirror directly; the mirror is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_5", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the rice cooker directly now?", + "answer": "You can now touch the rice cooker directly.", + "question_chinese": "我现在能够直接触碰到电饭煲吗？", + "answer_chinese": "你现在可以直接触碰到电饭煲。", + "evidence": [ + { + "description": "The rice cooker is in the kitchen.", + "timestamps": { + "start": "00:02:59", + "end": "00:03:00" + } + }, + { + "description": "Your current location is the living room.", + "timestamps": { + "start": "00:30:10", + "end": "00:30:11" + } + } + ], + "question_timestamps": "00:30:15", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can now touch the rice cooker directly.", + "B": "You cannot touch the rice cooker directly now." + }, + "answer_label": "A", + "answer_text": "You can now touch the rice cooker directly.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_5", + "correct_answer": "You can now touch the rice cooker directly.", + "original_options": { + "A": "You can now touch the rice cooker directly.", + "B": "You cannot touch the rice cooker directly now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_6", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the rice cooker directly now?", + "answer": "You can't touch the rice cooker directly now.", + "question_chinese": "我现在能够直接触碰到电饭煲吗？", + "answer_chinese": "你现在不能直接触碰到电饭煲。", + "evidence": [ + { + "description": "The rice cooker is in the kitchen.", + "timestamps": { + "start": "00:02:59", + "end": "00:03:00" + } + }, + { + "description": "Your current location is the living room.", + "timestamps": { + "start": "00:30:10", + "end": "00:30:11" + } + } + ], + "question_timestamps": "00:31:11", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can touch the rice cooker directly now.", + "B": "You can't touch the rice cooker directly now." + }, + "answer_label": "B", + "answer_text": "You can't touch the rice cooker directly now.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_6", + "correct_answer": "You can't touch the rice cooker directly now.", + "original_options": { + "A": "You can touch the rice cooker directly now.", + "B": "You can't touch the rice cooker directly now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_2_T1", + "qid_num": 26, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I reach the cleaning cloth?", + "answer": "You can reach the rag; it’s to your left.", + "question_chinese": "我能接触到抹布吗？", + "answer_chinese": "你能接触到抹布，抹布在你左边。", + "evidence": [ + { + "description": "You can not touch the dishcloth.", + "timestamps": { + "start": "00:00:29", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can’t reach the rag.", + "B": "You can reach the rag to your right.", + "C": "You can reach the rag, but it’s positioned in front of you rather than off to your left.", + "D": "You can reach the rag easily, yet it’s behind you, not sitting on the left side.", + "E": "You can reach the rag; it’s to your left." + }, + "answer_label": "E", + "answer_text": "You can reach the rag; it’s to your left.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_2_T1", + "correct_answer": "You can reach the rag; it’s to your left.", + "original_options": { + "A": "You can’t reach the rag.", + "B": "You can reach the rag to your right.", + "C": "You can reach the rag, but it’s positioned in front of you rather than off to your left.", + "D": "You can reach the rag easily, yet it’s behind you, not sitting on the left side.", + "E": "You can reach the rag; it’s to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 19, + 17 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_2_T2", + "qid_num": 27, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I reach the cleaning cloth?", + "answer": "You can’t reach the rag; it’s in front of you.", + "question_chinese": "我能接触到抹布吗？", + "answer_chinese": "你不能接触到抹布，抹布在你前方。", + "evidence": [ + { + "description": "You can not touch the dishcloth.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:01:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can’t reach the rag; it’s in front of you.", + "B": "You can reach the rag.", + "C": "You can’t reach it.", + "D": "You can reach the rag; it’s in front of you, so it’s within arm’s length.", + "E": "You can reach the rag even though it’s to your right, because it’s close enough to grab." + }, + "answer_label": "A", + "answer_text": "You can’t reach the rag; it’s in front of you.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_2_T2", + "correct_answer": "You can’t reach the rag; it’s in front of you.", + "original_options": { + "A": "You can’t reach the rag; it’s in front of you.", + "B": "You can reach the rag.", + "C": "You can’t reach it.", + "D": "You can reach the rag; it’s in front of you, so it’s within arm’s length.", + "E": "You can reach the rag even though it’s to your right, because it’s close enough to grab." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 19 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_18", + "qid_num": 28, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Apart from the sofa I'm sitting on now, what is the nearest item to me?", + "answer": "The nearest thing to you is a box of masks.", + "question_chinese": "除了我现在坐的沙发以外，离我最近的物品是什么？", + "answer_chinese": "离你最近的是一盒口罩。", + "evidence": [ + { + "description": "The nearest thing to you is a box of masks.", + "timestamps": { + "start": "00:44:53", + "end": "00:44:54" + } + } + ], + "question_timestamps": "00:44:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A table lamp.", + "B": "A stack of books.", + "C": "The nearest thing to you is a box of masks.", + "D": "The nearest item to you is a chair cushion, sitting closer than anything else nearby.", + "E": "The nearest item to you is a basket of toys, right next to you compared with other objects." + }, + "answer_label": "C", + "answer_text": "The nearest thing to you is a box of masks.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_18", + "correct_answer": "The nearest thing to you is a box of masks.", + "original_options": { + "A": "A table lamp.", + "B": "A stack of books.", + "C": "The nearest thing to you is a box of masks.", + "D": "The nearest item to you is a chair cushion, sitting closer than anything else nearby.", + "E": "The nearest item to you is a basket of toys, right next to you compared with other objects." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 18 + ] + }, + { + "video_uid": "1a7e6155-3613-4f55-865c-00e196db0da0", + "video_length": "01:04:01", + "qid": "1a7e6155-3613-4f55-865c-00e196db0da0_17", + "qid_num": 29, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Apart from the sofa I'm sitting on now, what is the nearest item to me?", + "answer": "The nearest item to you is the throw pillow.", + "question_chinese": "除了我现在坐的沙发以外，离我最近的物品是什么？", + "answer_chinese": "离你最近的物品是抱枕。", + "evidence": [ + { + "description": "The nearest thing to you is a box of masks.", + "timestamps": { + "start": "00:44:53", + "end": "00:44:54" + } + } + ], + "question_timestamps": "00:44:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The nearest item to you is the blanket.", + "B": "The nearest item to you is the throw pillow.", + "C": "The nearest item to you is the footstool.", + "D": "The nearest item to you is the armrest cushion, sitting closer than the pillow on the sofa.", + "E": "The nearest item to you is the large rug, right in front of you by the sofa edge." + }, + "answer_label": "B", + "answer_text": "The nearest item to you is the throw pillow.", + "q_id": "1a7e6155-3613-4f55-865c-00e196db0da0_17", + "correct_answer": "The nearest item to you is the throw pillow.", + "original_options": { + "A": "The nearest item to you is the blanket.", + "B": "The nearest item to you is the throw pillow.", + "C": "The nearest item to you is the footstool.", + "D": "The nearest item to you is the armrest cushion, sitting closer than the pillow on the sofa.", + "E": "The nearest item to you is the large rug, right in front of you by the sofa edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 18 + ] + } +] diff --git a/QAs/1aa2aa12-436f-48b4-853d-ae7b03ebed53-mcq.json b/QAs/1aa2aa12-436f-48b4-853d-ae7b03ebed53-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..861b2a9537eb709f9043ce53d7e77f63350166be --- /dev/null +++ b/QAs/1aa2aa12-436f-48b4-853d-ae7b03ebed53-mcq.json @@ -0,0 +1,883 @@ +[ + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_8_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go get four more packaging boxes with green ends later, how should I walk there?", + "answer": "You’re going to turn around, go down the stairs, walk forward to the corner under the eaves, then turn left and go straight to the pile of packing boxes with green ends; then you’ll be able to get the packing boxes. The green packing boxes are at the left turn in front of you downstairs.", + "question_chinese": "如果过一会儿我想再拿四个两端颜色为绿色的包装盒，我将要怎么走？", + "answer_chinese": "你将要向后转，走下楼梯，往前走到屋檐拐角，然后左转，直走到两端为绿色的包装盒堆旁，即可拿到包装盒。绿色包装盒在你楼下前方左转处。", + "evidence": [ + { + "description": "Walk straight from the green packaging boxes, turn right at the corner, go upstairs and come here.", + "timestamps": { + "start": "00:17:17", + "end": "00:17:40" + } + } + ], + "question_timestamps": "00:22:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left at the bottom of the stairs.", + "B": "Turn right at the bottom of the stairs.", + "C": "Turn around, go downstairs, but turn right at the corner under the eaves to reach the boxes.", + "D": "Turn around, go down the stairs, walk to the corner under the eaves, then continue straight without turning left.", + "E": "You’re going to turn around, go down the stairs, walk forward to the corner under the eaves, then turn left and go straight to the pile of packing boxes with green ends; then you’ll be able to get the packing boxes. The green packing boxes are at the left turn in front of you downstairs." + }, + "answer_label": "E", + "answer_text": "You’re going to turn around, go down the stairs, walk forward to the corner under the eaves, then turn left and go straight to the pile of packing boxes with green ends; then you’ll be able to get the packing boxes. The green packing boxes are at the left turn in front of you downstairs.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_8_T2", + "correct_answer": "You’re going to turn around, go down the stairs, walk forward to the corner under the eaves, then turn left and go straight to the pile of packing boxes with green ends; then you’ll be able to get the packing boxes. The green packing boxes are at the left turn in front of you downstairs.", + "original_options": { + "A": "Turn left at the bottom of the stairs.", + "B": "Turn right at the bottom of the stairs.", + "C": "Turn around, go downstairs, but turn right at the corner under the eaves to reach the boxes.", + "D": "Turn around, go down the stairs, walk to the corner under the eaves, then continue straight without turning left.", + "E": "You’re going to turn around, go down the stairs, walk forward to the corner under the eaves, then turn left and go straight to the pile of packing boxes with green ends; then you’ll be able to get the packing boxes. The green packing boxes are at the left turn in front of you downstairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 19 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_8_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go get four more packaging boxes with green ends later, how should I walk there?", + "answer": "You’re going to turn right, walk to the corner under the eaves, then turn left and go straight to the pile of packing boxes with green ends; then you’ll be able to get the packing boxes. The packing boxes are to the left of the room corner on your right.", + "question_chinese": "如果过一会儿我想再拿四个两端颜色为绿色的包装盒，我将要怎么走？", + "answer_chinese": "你将要向右转，走到屋檐拐角，然后左转，直走到两端为绿色的包装盒堆旁，即可拿到包装盒。包装盒在你右边房角左边的位置。", + "evidence": [ + { + "description": "You walked straight from the pile of green packaging boxes, turned right at the corner, and walked straight to the current location.", + "timestamps": { + "start": "00:17:17", + "end": "00:17:40" + } + } + ], + "question_timestamps": "00:24:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left and head to the corner.", + "B": "You’re going to turn right, walk to the corner under the eaves, then turn left and go straight to the pile of packing boxes with green ends; then you’ll be able to get the packing boxes. The packing boxes are to the left of the room corner on your right.", + "C": "Go straight to the corner under the eaves.", + "D": "Turn right, walk to the corner under the eaves, then turn right and go straight to the boxes.", + "E": "Turn right to the corner under the eaves, but the boxes are to the right of that room corner." + }, + "answer_label": "B", + "answer_text": "You’re going to turn right, walk to the corner under the eaves, then turn left and go straight to the pile of packing boxes with green ends; then you’ll be able to get the packing boxes. The packing boxes are to the left of the room corner on your right.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_8_T1", + "correct_answer": "You’re going to turn right, walk to the corner under the eaves, then turn left and go straight to the pile of packing boxes with green ends; then you’ll be able to get the packing boxes. The packing boxes are to the left of the room corner on your right.", + "original_options": { + "A": "Turn left and head to the corner.", + "B": "You’re going to turn right, walk to the corner under the eaves, then turn left and go straight to the pile of packing boxes with green ends; then you’ll be able to get the packing boxes. The packing boxes are to the left of the room corner on your right.", + "C": "Go straight to the corner under the eaves.", + "D": "Turn right, walk to the corner under the eaves, then turn right and go straight to the boxes.", + "E": "Turn right to the corner under the eaves, but the boxes are to the right of that room corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 19 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_9", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where was I when I used the power drill last time?", + "answer": "You were between two blue metal frames on the first floor, next to the first-floor staircase.", + "question_chinese": "我上次用电钻时我在哪里？", + "answer_chinese": "你在一楼两个蓝色金属制支架中间，在一楼楼梯旁。", + "evidence": [ + { + "description": "You were between two blue metal frames on the first floor, next to the first-floor staircase.", + "timestamps": { + "start": "00:13:58", + "end": "00:14:49" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You were near the service elevator.", + "B": "You were between two blue metal frames on the first floor, next to the first-floor staircase.", + "C": "You were beside the main entrance.", + "D": "You were between two blue metal frames on the second floor, next to the staircase up there.", + "E": "You were between two blue metal frames on the first floor, but closer to the maintenance room door than the stairs." + }, + "answer_label": "B", + "answer_text": "You were between two blue metal frames on the first floor, next to the first-floor staircase.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_9", + "correct_answer": "You were between two blue metal frames on the first floor, next to the first-floor staircase.", + "original_options": { + "A": "You were near the service elevator.", + "B": "You were between two blue metal frames on the first floor, next to the first-floor staircase.", + "C": "You were beside the main entrance.", + "D": "You were between two blue metal frames on the second floor, next to the staircase up there.", + "E": "You were between two blue metal frames on the first floor, but closer to the maintenance room door than the stairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 21 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_10", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where was I when I used the power drill last time?", + "answer": "You were on the open-air wooden floor of the second floor of the house.", + "question_chinese": "我上次用电钻时我在哪里？", + "answer_chinese": "你在房子二楼的露天木地板上。", + "evidence": [ + { + "description": "You were on the open-air wooden floor of the second floor of the house.", + "timestamps": { + "start": "00:22:09", + "end": "00:23:02" + } + } + ], + "question_timestamps": "00:24:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You were on the third floor.", + "B": "You were on an enclosed wooden floor.", + "C": "You were on the open-air wooden floor of the second floor of the house.", + "D": "You were on the tiled concrete floor of the second-floor balcony rather than the open-air wooden floor.", + "E": "You were on a semi-enclosed wooden area on the second floor, not out on the open-air section." + }, + "answer_label": "C", + "answer_text": "You were on the open-air wooden floor of the second floor of the house.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_10", + "correct_answer": "You were on the open-air wooden floor of the second floor of the house.", + "original_options": { + "A": "You were on the third floor.", + "B": "You were on an enclosed wooden floor.", + "C": "You were on the open-air wooden floor of the second floor of the house.", + "D": "You were on the tiled concrete floor of the second-floor balcony rather than the open-air wooden floor.", + "E": "You were on a semi-enclosed wooden area on the second floor, not out on the open-air section." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 17 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_15", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I am going to get back the red set square. How should I go?", + "answer": "You will turn left, walk to the stairs on the first floor, and then you can get the red triangle ruler.", + "question_chinese": "我将要拿回红色三角尺，我该怎么走？", + "answer_chinese": "你将要向左转，走到一楼的楼梯旁，就可以拿到红色三角尺了。", + "evidence": [ + { + "description": "A red set square is on the stairs to your left.", + "timestamps": { + "start": "00:07:49", + "end": "00:07:55" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You will turn right to the stairs.", + "B": "You will turn left, walk to the stairs on the first floor, and then you can get the red triangle ruler.", + "C": "You will go to the second-floor stairs.", + "D": "You will turn left, walk past the stairs on the first floor, and then get the red triangle ruler.", + "E": "You will turn left, walk to the stairs on the second floor, and then you can get the red triangle ruler." + }, + "answer_label": "B", + "answer_text": "You will turn left, walk to the stairs on the first floor, and then you can get the red triangle ruler.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_15", + "correct_answer": "You will turn left, walk to the stairs on the first floor, and then you can get the red triangle ruler.", + "original_options": { + "A": "You will turn right to the stairs.", + "B": "You will turn left, walk to the stairs on the first floor, and then you can get the red triangle ruler.", + "C": "You will go to the second-floor stairs.", + "D": "You will turn left, walk past the stairs on the first floor, and then get the red triangle ruler.", + "E": "You will turn left, walk to the stairs on the second floor, and then you can get the red triangle ruler." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 21 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_16", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I am going to get back the red set square. How should I go?", + "answer": "You will turn around, walk out of the small room, then turn left and walk straight to the corner, then turn right and walk to the stairs where you can get the red triangular ruler.", + "question_chinese": "我将要拿回红色三角尺，我该怎么走？", + "answer_chinese": "你将要向后转，走出小房间，然后左转，直走至拐角，然后右转，走到楼梯旁就可以拿到红色三角尺了。", + "evidence": [ + { + "description": "Start from beside the set square, walk straight to the corner, turn left, walk straight to the side of the small room, turn right and enter, and you will reach your current location.", + "timestamps": { + "start": "00:14:45", + "end": "00:15:57" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right after leaving the small room.", + "B": "Turn left twice to reach the stairs.", + "C": "You will turn around, walk out of the small room, then turn left and walk straight to the corner, then turn right and walk to the stairs where you can get the red triangular ruler.", + "D": "Turn around, exit the small room, turn left to the corner, then turn left and walk to the stairs.", + "E": "Turn around, walk out of the small room, turn left to the corner, then turn right toward the elevator to get it." + }, + "answer_label": "C", + "answer_text": "You will turn around, walk out of the small room, then turn left and walk straight to the corner, then turn right and walk to the stairs where you can get the red triangular ruler.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_16", + "correct_answer": "You will turn around, walk out of the small room, then turn left and walk straight to the corner, then turn right and walk to the stairs where you can get the red triangular ruler.", + "original_options": { + "A": "Turn right after leaving the small room.", + "B": "Turn left twice to reach the stairs.", + "C": "You will turn around, walk out of the small room, then turn left and walk straight to the corner, then turn right and walk to the stairs where you can get the red triangular ruler.", + "D": "Turn around, exit the small room, turn left to the corner, then turn left and walk to the stairs.", + "E": "Turn around, walk out of the small room, turn left to the corner, then turn right toward the elevator to get it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 22 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_4", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I last saw the green ladder, was there someone on it?", + "answer": "The last time you saw the green ladder, there was someone on it.", + "question_chinese": "我最后一次看到绿色人字梯时，绿色人字梯上是否有人？", + "answer_chinese": "你最后一次看到绿色人字梯时，绿色人字梯上有人。", + "evidence": [ + { + "description": "The last time you saw the green ladder, there was someone on it.", + "timestamps": { + "start": "00:15:32", + "end": "00:15:41" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The last time you saw the green ladder, there was someone on it.", + "B": "The last time you saw the green ladder, there was not someone on it." + }, + "answer_label": "A", + "answer_text": "The last time you saw the green ladder, there was someone on it.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_4", + "correct_answer": "The last time you saw the green ladder, there was someone on it.", + "original_options": { + "A": "The last time you saw the green ladder, there was someone on it.", + "B": "The last time you saw the green ladder, there was not someone on it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_3_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "The last time I saw the green stepladder, was anyone on it?", + "answer": "The last time you saw the green step ladder, there was someone on it. The green step ladder was to your right.", + "question_chinese": "我最后一次看到绿色人字梯时，绿色人字梯上是否有人？", + "answer_chinese": "你最后一次看到绿色人字梯时，绿色人字梯上有人。绿色人字梯在你右边。", + "evidence": [ + { + "description": "The last time you saw the green ladder, there was someone on it.", + "timestamps": { + "start": "00:17:25", + "end": "00:17:35" + } + } + ], + "question_timestamps": "00:18:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The last time you saw the green step ladder, there was someone on it. The green step ladder was to your right.", + "B": "The last time you saw the green step ladder, there was no one on it. The green step ladder was to your right." + }, + "answer_label": "A", + "answer_text": "The last time you saw the green step ladder, there was someone on it. The green step ladder was to your right.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_3_T2", + "correct_answer": "The last time you saw the green step ladder, there was someone on it. The green step ladder was to your right.", + "original_options": { + "A": "The last time you saw the green step ladder, there was someone on it. The green step ladder was to your right.", + "B": "The last time you saw the green step ladder, there was no one on it. The green step ladder was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 23 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_3_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "The last time I saw the green stepladder, was anyone on it?", + "answer": "The last time you saw the green step ladder, there was no one on it. The green step ladder was in front of you.", + "question_chinese": "我最后一次看到绿色人字梯时，绿色人字梯上是否有人？", + "answer_chinese": "你最后一次看到绿色人字梯时，绿色人字梯上没有人。绿色人字梯在你正前方。", + "evidence": [ + { + "description": "The last time you saw the green ladder, there was no someone on it.", + "timestamps": { + "start": "00:27:00", + "end": "00:27:08" + } + } + ], + "question_timestamps": "00:27:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The last time you saw the green step ladder, there was no one on it. The green step ladder was in front of you.", + "B": "The last time you saw the green step ladder, there was someone on it. The green step ladder was in front of you." + }, + "answer_label": "A", + "answer_text": "The last time you saw the green step ladder, there was no one on it. The green step ladder was in front of you.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_3_T1", + "correct_answer": "The last time you saw the green step ladder, there was no one on it. The green step ladder was in front of you.", + "original_options": { + "A": "The last time you saw the green step ladder, there was no one on it. The green step ladder was in front of you.", + "B": "The last time you saw the green step ladder, there was someone on it. The green step ladder was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 23 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch a piece of wood to be processed now?", + "answer": "You can't feel the board now.", + "question_chinese": "我现在可以触摸一块待加工的木板吗？", + "answer_chinese": "你现在摸不到木板。", + "evidence": [ + { + "description": "You start from the nearest board to your current position and it takes about seven to eight seconds to reach your current location.", + "timestamps": { + "start": "00:15:47", + "end": "00:15:58" + } + }, + { + "description": "There are no untreated wooden boards here.", + "timestamps": { + "start": "00:15:58", + "end": "00:15:59" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't feel the board now.", + "B": "You can feel the board now." + }, + "answer_label": "A", + "answer_text": "You can't feel the board now.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_2", + "correct_answer": "You can't feel the board now.", + "original_options": { + "A": "You can't feel the board now.", + "B": "You can feel the board now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_2_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch a piece of wood waiting to be processed right now?", + "answer": "You can’t touch the wooden board right now. The wooden board is on the grass to your right.", + "question_chinese": "我现在可以触摸一块待加工的木板吗？", + "answer_chinese": "你现在摸不到木板。木板在你右边的草地上。", + "evidence": [ + { + "description": "You start from the nearest board to your current position and it takes about seven to eight seconds to reach your current location.", + "timestamps": { + "start": "00:15:47", + "end": "00:15:58" + } + }, + { + "description": "There are no untreated wooden boards here.", + "timestamps": { + "start": "00:15:58", + "end": "00:15:59" + } + } + ], + "question_timestamps": "00:16:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t touch the wooden board right now. The wooden board is on the grass to your right.", + "B": "You can touch the wooden board right now. The wooden board is on the grass to your right." + }, + "answer_label": "A", + "answer_text": "You can’t touch the wooden board right now. The wooden board is on the grass to your right.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_2_T1", + "correct_answer": "You can’t touch the wooden board right now. The wooden board is on the grass to your right.", + "original_options": { + "A": "You can’t touch the wooden board right now. The wooden board is on the grass to your right.", + "B": "You can touch the wooden board right now. The wooden board is on the grass to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_2_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch a piece of wood waiting to be processed right now?", + "answer": "You can touch the wooden board right now. The wooden board is in your hand.", + "question_chinese": "我现在可以触摸一块待加工的木板吗？", + "answer_chinese": "你现在能摸到木板。木板在你手上。", + "evidence": [ + { + "description": "You are holding the wooden planks.", + "timestamps": { + "start": "00:17:47", + "end": "00:18:00" + } + }, + { + "description": "There are no untreated wooden boards here.", + "timestamps": { + "start": "00:15:58", + "end": "00:15:59" + } + } + ], + "question_timestamps": "00:18:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t touch the wooden board right now. The wooden board is in your hand.", + "B": "You can touch the wooden board right now. The wooden board is in your hand." + }, + "answer_label": "B", + "answer_text": "You can touch the wooden board right now. The wooden board is in your hand.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_2_T2", + "correct_answer": "You can touch the wooden board right now. The wooden board is in your hand.", + "original_options": { + "A": "You can’t touch the wooden board right now. The wooden board is in your hand.", + "B": "You can touch the wooden board right now. The wooden board is in your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_11", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Are the two transparent boxes next to the transparent box where the black small object is located covered?", + "answer": "They are covered.", + "question_chinese": "现在在黑色小物件所在透明盒子旁边的那两个透明盒子是否被盖住了？", + "answer_chinese": "它们被盖住了。", + "evidence": [ + { + "description": "They are covered.", + "timestamps": { + "start": "00:07:39", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "They are covered.", + "B": "They are not covered." + }, + "answer_label": "A", + "answer_text": "They are covered.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_11", + "correct_answer": "They are covered.", + "original_options": { + "A": "They are covered.", + "B": "They are not covered." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_12", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Are the two transparent boxes next to the transparent box where the black small object is located covered?", + "answer": "Only one is not covered.", + "question_chinese": "现在在黑色小物件所在透明盒子旁边的那两个透明盒子是否被盖住了？", + "answer_chinese": "只有一个没被盖住。", + "evidence": [ + { + "description": "Only one is not covered.", + "timestamps": { + "start": "00:14:39", + "end": "00:14:48" + } + }, + { + "description": "You opened one of the boxes.", + "timestamps": { + "start": "00:09:39", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "They are both covered.", + "B": "Only one is not covered." + }, + "answer_label": "B", + "answer_text": "Only one is not covered.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_12", + "correct_answer": "Only one is not covered.", + "original_options": { + "A": "They are both covered.", + "B": "Only one is not covered." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_8", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the meter stick without moving my feet now?", + "answer": "You cannot get the meter stick.", + "question_chinese": "现在我可以在不移动脚步的情况下拿到米尺吗？", + "answer_chinese": "你不能拿到米尺。", + "evidence": [ + { + "description": "Starting from the miter saw, you need to turn left and walk straight to the small room to reach your current location.", + "timestamps": { + "start": "00:14:49", + "end": "00:15:57" + } + }, + { + "description": "The tape measure is on the miter saw.", + "timestamps": { + "start": "00:06:40", + "end": "00:06:47" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You cannot get the meter stick.", + "B": "You can get the meter stick." + }, + "answer_label": "A", + "answer_text": "You cannot get the meter stick.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_8", + "correct_answer": "You cannot get the meter stick.", + "original_options": { + "A": "You cannot get the meter stick.", + "B": "You can get the meter stick." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_7", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the meter stick without moving my feet now?", + "answer": "You can get the meter stick.", + "question_chinese": "现在我可以在不移动脚步的情况下拿到米尺吗？", + "answer_chinese": "你可以拿到米尺。", + "evidence": [ + { + "description": "The tape measure is to your lower right, on the miter saw.", + "timestamps": { + "start": "00:23:47", + "end": "00:23:59" + } + } + ], + "question_timestamps": "00:24:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot get the meter stick.", + "B": "You can get the meter stick." + }, + "answer_label": "B", + "answer_text": "You can get the meter stick.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_7", + "correct_answer": "You can get the meter stick.", + "original_options": { + "A": "You cannot get the meter stick.", + "B": "You can get the meter stick." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_6", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the red set square or the miter saw?", + "answer": "Now the red set square is closer to you.", + "question_chinese": "现在红色三角尺和斜切锯哪个离我更近？", + "answer_chinese": "现在红色三角尺离你更近。", + "evidence": [ + { + "description": "Now the red set square is closer to you.", + "timestamps": { + "start": "00:07:39", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The miter saw is closer to you.", + "B": "Now the red set square is closer to you." + }, + "answer_label": "B", + "answer_text": "Now the red set square is closer to you.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_6", + "correct_answer": "Now the red set square is closer to you.", + "original_options": { + "A": "The miter saw is closer to you.", + "B": "Now the red set square is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53", + "video_length": "00:32:30", + "qid": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_5", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the red set square or the miter saw?", + "answer": "The miter saw is closer to you.", + "question_chinese": "现在红色三角尺和斜切锯哪个离我更近？", + "answer_chinese": "斜切锯离你更近。", + "evidence": [ + { + "description": "To get from the red triangle ruler to your current location, you must pass through the miter saw.", + "timestamps": { + "start": "00:14:39", + "end": "00:15:57" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The red set square is closer to you.", + "B": "The miter saw is closer to you." + }, + "answer_label": "B", + "answer_text": "The miter saw is closer to you.", + "q_id": "1aa2aa12-436f-48b4-853d-ae7b03ebed53_5", + "correct_answer": "The miter saw is closer to you.", + "original_options": { + "A": "The red set square is closer to you.", + "B": "The miter saw is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + } +] diff --git a/QAs/1aa426b2-4d83-4b56-9d3c-4616adc464c8-mcq.json b/QAs/1aa426b2-4d83-4b56-9d3c-4616adc464c8-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..417b9ff01ad0a3bbf7c8edca62d926d6de881f2f --- /dev/null +++ b/QAs/1aa426b2-4d83-4b56-9d3c-4616adc464c8-mcq.json @@ -0,0 +1,877 @@ +[ + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_4", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the bookshelf directly now?", + "answer": "No, you can't.", + "question_chinese": "如果我想把茶几上的本子放到书架需要经过什么路径?", + "answer_chinese": "书从当前茶几位置,需向左移动,然后右转，到达书架所在位置。", + "evidence": [ + { + "description": "The bookshelf is about 3 meters away from you, on your upper left.", + "timestamps": { + "start": "00:14:20", + "end": "00:14:21" + } + } + ], + "question_timestamps": "00:14:26", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "answer_label": "B", + "answer_text": "No, you can't.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_4", + "correct_answer": "No, you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_5", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the bookshelf directly now?", + "answer": "The book needs to be moved from its current position on the coffee table to the left, and then turned right to reach its destination on the bookshelf.", + "question_chinese": "如果我想把茶几上的本子放到书架需要经过什么路径?", + "answer_chinese": "书从当前茶几位置,需向左移动,然后右转，到达书架所在位置。", + "evidence": [ + { + "description": "The bookshelf is located in the corner of the room. To reach it, you need to move from the coffee table to the left and then turn right.", + "timestamps": { + "start": "00:14:15", + "end": "00:14:20" + } + } + ], + "question_timestamps": "00:38:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The book needs to be moved from its current position on the coffee table to the left, and then turned right to reach its destination on the bookshelf.", + "B": "No, you can't." + }, + "answer_label": "A", + "answer_text": "The book needs to be moved from its current position on the coffee table to the left, and then turned right to reach its destination on the bookshelf.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_5", + "correct_answer": "The book needs to be moved from its current position on the coffee table to the left, and then turned right to reach its destination on the bookshelf.", + "original_options": { + "A": "The book needs to be moved from its current position on the coffee table to the left, and then turned right to reach its destination on the bookshelf.", + "B": "No, you can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_8", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first put down my phone, in which direction was it facing?", + "answer": "It's on your left front.", + "question_chinese": "当我放下手机时，它在我的哪个方向？", + "answer_chinese": "它在你的左前方。", + "evidence": [ + { + "description": "The phone is on the edge of the coffee table.", + "timestamps": { + "start": "00:14:50", + "end": "00:14:54" + } + } + ], + "question_timestamps": "00:15:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's on your right front.", + "B": "It's on your left back.", + "C": "It's positioned directly in front of you, slightly to the right from where you first set it down.", + "D": "It's on your left front.", + "E": "It's behind you on the right side when you first put it down, facing away from your body." + }, + "answer_label": "D", + "answer_text": "It's on your left front.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_8", + "correct_answer": "It's on your left front.", + "original_options": { + "A": "It's on your right front.", + "B": "It's on your left back.", + "C": "It's positioned directly in front of you, slightly to the right from where you first set it down.", + "D": "It's on your left front.", + "E": "It's behind you on the right side when you first put it down, facing away from your body." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_9", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first put down my phone, in which direction was it facing?", + "answer": "It's on your right front.", + "question_chinese": "当我放下手机时，它在我的哪个方向？", + "answer_chinese": "它在你的右前方。", + "evidence": [ + { + "description": "You are facing the drum set. The phone is on the cabinet, beside the drum set.", + "timestamps": { + "start": "00:46:01", + "end": "00:46:02" + } + } + ], + "question_timestamps": "00:46:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's on your left front.", + "B": "It's on your right side.", + "C": "It's set down directly in front of you, not toward your right front position.", + "D": "It's placed behind you on the right, so it isn't facing your right front.", + "E": "It's on your right front." + }, + "answer_label": "E", + "answer_text": "It's on your right front.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_9", + "correct_answer": "It's on your right front.", + "original_options": { + "A": "It's on your left front.", + "B": "It's on your right side.", + "C": "It's set down directly in front of you, not toward your right front position.", + "D": "It's placed behind you on the right, so it isn't facing your right front.", + "E": "It's on your right front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_10", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take to pick up the poster just now?", + "answer": "You get up from the sofa, walk around the coffee table, and go straight to the cabinet.", + "question_chinese": "我刚才去拿海报走了什么路线?", + "answer_chinese": "你从沙发起身,绕过茶几,直走到柜子前。", + "evidence": [ + { + "description": "The cabinet is directly in front of the sofa, separated by a coffee table.", + "timestamps": { + "start": "00:19:18", + "end": "00:19:29" + } + }, + { + "description": "You bypass the coffee table and walk to the cabinet.", + "timestamps": { + "start": "00:31:00", + "end": "00:31:05" + } + }, + { + "description": "The poster is in the cabinet.", + "timestamps": { + "start": "00:31:08", + "end": "00:31:10" + } + } + ], + "question_timestamps": "00:31:10", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You went straight to the cabinet.", + "B": "You walked around the ottoman first.", + "C": "You get up from the sofa, walk around the coffee table, and go straight to the cabinet.", + "D": "You got up from the armchair, circled the coffee table, and then walked straight to the cabinet.", + "E": "You stood up from the sofa, went around the side table instead, and headed to the cabinet afterward." + }, + "answer_label": "C", + "answer_text": "You get up from the sofa, walk around the coffee table, and go straight to the cabinet.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_10", + "correct_answer": "You get up from the sofa, walk around the coffee table, and go straight to the cabinet.", + "original_options": { + "A": "You went straight to the cabinet.", + "B": "You walked around the ottoman first.", + "C": "You get up from the sofa, walk around the coffee table, and go straight to the cabinet.", + "D": "You got up from the armchair, circled the coffee table, and then walked straight to the cabinet.", + "E": "You stood up from the sofa, went around the side table instead, and headed to the cabinet afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_11", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take to pick up the poster just now?", + "answer": "You bypass the coffee table and walk to the cabinet.", + "question_chinese": "我刚才去拿海报走了什么路线?", + "answer_chinese": "你从墙壁前右转，直走三步走到茶几前。", + "evidence": [ + { + "description": "The poster is on the floor next to the coffee table.", + "timestamps": { + "start": "00:38:12", + "end": "00:38:16" + } + }, + { + "description": "After measuring the wall, you walked to the coffee table to pick up the poster.", + "timestamps": { + "start": "00:38:14", + "end": "00:38:20" + } + } + ], + "question_timestamps": "00:38:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You walked straight to the cabinet.", + "B": "You went around the coffee table.", + "C": "You walked to the cabinet but took a wide detour instead of bypassing the coffee table directly.", + "D": "You bypass the coffee table and walk to the cabinet.", + "E": "You approached the cabinet after circling the coffee table rather than simply bypassing it on your way." + }, + "answer_label": "D", + "answer_text": "You bypass the coffee table and walk to the cabinet.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_11", + "correct_answer": "You bypass the coffee table and walk to the cabinet.", + "original_options": { + "A": "You walked straight to the cabinet.", + "B": "You went around the coffee table.", + "C": "You walked to the cabinet but took a wide detour instead of bypassing the coffee table directly.", + "D": "You bypass the coffee table and walk to the cabinet.", + "E": "You approached the cabinet after circling the coffee table rather than simply bypassing it on your way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_16", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many instruments did I play just now?", + "answer": "You played a musical instrument.", + "question_chinese": "我刚才演奏了几种乐器？", + "answer_chinese": "你演奏了一种乐器。", + "evidence": [ + { + "description": "You played the drum set.", + "timestamps": { + "start": "00:46:29", + "end": "00:46:36" + } + } + ], + "question_timestamps": "00:46:36", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You played no instruments.", + "B": "You played two instruments.", + "C": "You played three different instruments back to back, switching between them during the moment.", + "D": "You played several instruments, moving from one to another rather than sticking with a single one.", + "E": "You played a musical instrument." + }, + "answer_label": "E", + "answer_text": "You played a musical instrument.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_16", + "correct_answer": "You played a musical instrument.", + "original_options": { + "A": "You played no instruments.", + "B": "You played two instruments.", + "C": "You played three different instruments back to back, switching between them during the moment.", + "D": "You played several instruments, moving from one to another rather than sticking with a single one.", + "E": "You played a musical instrument." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 16 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_17", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many instruments did I play just now?", + "answer": "You played two kinds of musical instruments.", + "question_chinese": "我刚才演奏了几种乐器？", + "answer_chinese": "你演奏了两种乐器。", + "evidence": [ + { + "description": "You played the drum set.", + "timestamps": { + "start": "00:46:28", + "end": "00:46:36" + } + }, + { + "description": "You played the piano.", + "timestamps": { + "start": "00:52:43", + "end": "00:53:02" + } + } + ], + "question_timestamps": "00:53:02", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You played one instrument.", + "B": "You played three instruments.", + "C": "You played four different kinds of musical instruments instead of playing only two kinds.", + "D": "You played two kinds of musical instruments.", + "E": "You didn't play any musical instruments at all, so it wasn't two kinds of musical instruments." + }, + "answer_label": "D", + "answer_text": "You played two kinds of musical instruments.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_17", + "correct_answer": "You played two kinds of musical instruments.", + "original_options": { + "A": "You played one instrument.", + "B": "You played three instruments.", + "C": "You played four different kinds of musical instruments instead of playing only two kinds.", + "D": "You played two kinds of musical instruments.", + "E": "You didn't play any musical instruments at all, so it wasn't two kinds of musical instruments." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 16 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_6", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pens did I use to draw just now?", + "answer": "One pen.", + "question_chinese": "我刚才用了几支笔画画？", + "answer_chinese": "一支笔。", + "evidence": [ + { + "description": "You picked up a pen from the coffee table and started drawing.", + "timestamps": { + "start": "00:14:19", + "end": "00:14:24" + } + } + ], + "question_timestamps": "00:14:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two pens.", + "B": "Three pens.", + "C": "You used two pens to draw, switching once between them during the drawing.", + "D": "You used four pens to draw, changing pens several times while making different marks.", + "E": "One pen." + }, + "answer_label": "E", + "answer_text": "One pen.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_6", + "correct_answer": "One pen.", + "original_options": { + "A": "Two pens.", + "B": "Three pens.", + "C": "You used two pens to draw, switching once between them during the drawing.", + "D": "You used four pens to draw, changing pens several times while making different marks.", + "E": "One pen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_7", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pens did I use to draw just now?", + "answer": "Two pens.", + "question_chinese": "我刚才用了几支笔画画？", + "answer_chinese": "两支笔。", + "evidence": [ + { + "description": "You picked up a pen from the coffee table and started drawing.", + "timestamps": { + "start": "00:14:19", + "end": "00:14:24" + } + }, + { + "description": "You changed another pen to draw.", + "timestamps": { + "start": "00:14:27", + "end": "00:14:31" + } + } + ], + "question_timestamps": "00:29:27", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One pen.", + "B": "Three pens.", + "C": "Two pens.", + "D": "You used four pens to draw, rotating through them instead of using exactly two pens.", + "E": "You used five pens to draw, picking up new pens multiple times rather than just two." + }, + "answer_label": "C", + "answer_text": "Two pens.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_7", + "correct_answer": "Two pens.", + "original_options": { + "A": "One pen.", + "B": "Three pens.", + "C": "Two pens.", + "D": "You used four pens to draw, rotating through them instead of using exactly two pens.", + "E": "You used five pens to draw, picking up new pens multiple times rather than just two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_12", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many posters did I paste just now?", + "answer": "You pasted two posters.", + "question_chinese": "我刚才粘贴了几张海报？", + "answer_chinese": "你粘贴了两张海报。", + "evidence": [ + { + "description": "You pasted a white poster.", + "timestamps": { + "start": "00:39:41", + "end": "00:39:54" + } + }, + { + "description": "You pasted a blue poster.", + "timestamps": { + "start": "00:41:37", + "end": "00:42:04" + } + } + ], + "question_timestamps": "00:42:06", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You pasted one poster.", + "B": "You pasted two posters.", + "C": "You pasted three posters.", + "D": "You pasted four posters in total, not two posters, placing more than a pair on the surface.", + "E": "You pasted five posters altogether, so the total was greater than two posters." + }, + "answer_label": "B", + "answer_text": "You pasted two posters.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_12", + "correct_answer": "You pasted two posters.", + "original_options": { + "A": "You pasted one poster.", + "B": "You pasted two posters.", + "C": "You pasted three posters.", + "D": "You pasted four posters in total, not two posters, placing more than a pair on the surface.", + "E": "You pasted five posters altogether, so the total was greater than two posters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 13 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_13", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many posters did I paste just now?", + "answer": "You pasted three posters.", + "question_chinese": "我刚才粘贴了几张海报？", + "answer_chinese": "你粘贴了三张海报。", + "evidence": [ + { + "description": "You pasted another poster with a white border.", + "timestamps": { + "start": "00:44:07", + "end": "00:44:16" + } + }, + { + "description": "You pasted a white poster.", + "timestamps": { + "start": "00:39:41", + "end": "00:39:54" + } + }, + { + "description": "You pasted a blue poster.", + "timestamps": { + "start": "00:41:37", + "end": "00:42:04" + } + } + ], + "question_timestamps": "00:44:16", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You pasted two posters.", + "B": "You pasted four posters.", + "C": "You pasted two posters, placing them side by side on the surface just now.", + "D": "You pasted three posters.", + "E": "You pasted four posters in total, finishing the last one moments before looking back." + }, + "answer_label": "D", + "answer_text": "You pasted three posters.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_13", + "correct_answer": "You pasted three posters.", + "original_options": { + "A": "You pasted two posters.", + "B": "You pasted four posters.", + "C": "You pasted two posters, placing them side by side on the surface just now.", + "D": "You pasted three posters.", + "E": "You pasted four posters in total, finishing the last one moments before looking back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_18", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I holding in my hand now?", + "answer": "What am I holding in my hand now?", + "question_chinese": "我手里现在拿着什么？", + "answer_chinese": "你手里拿着本子和笔。", + "evidence": [ + { + "description": "You are holding a notebook and a pen in your hand.", + "timestamps": { + "start": "00:18:49", + "end": "00:18:59" + } + } + ], + "question_timestamps": "00:18:59", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "What am I holding in my hand now?", + "B": "You are holding a cup.", + "C": "You are holding a phone.", + "D": "You are holding a bottle and a notebook in your hand right now, not a cup or spoon.", + "E": "You are holding a book and a marker in your hand now, rather than a phone and wallet." + }, + "answer_label": "A", + "answer_text": "What am I holding in my hand now?", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_18", + "correct_answer": "What am I holding in my hand now?", + "original_options": { + "A": "What am I holding in my hand now?", + "B": "You are holding a cup.", + "C": "You are holding a phone.", + "D": "You are holding a bottle and a notebook in your hand right now, not a cup or spoon.", + "E": "You are holding a book and a marker in your hand now, rather than a phone and wallet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_19", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I holding in my hand now?", + "answer": "You have a mobile phone in your hand.", + "question_chinese": "我手里现在拿着什么？", + "answer_chinese": "你手里拿着手机。", + "evidence": [ + { + "description": "You are holding a mobile phone in both hands and searching for something.", + "timestamps": { + "start": "00:45:40", + "end": "00:45:43" + } + } + ], + "question_timestamps": "00:45:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You have a tablet in your hand.", + "B": "You have a mobile phone in your hand.", + "C": "You have a remote control in your hand.", + "D": "You have a portable music player in your hand, small enough to grip in one hand.", + "E": "You have a handheld gaming device in your hand, with controls on both sides of the screen." + }, + "answer_label": "B", + "answer_text": "You have a mobile phone in your hand.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_19", + "correct_answer": "You have a mobile phone in your hand.", + "original_options": { + "A": "You have a tablet in your hand.", + "B": "You have a mobile phone in your hand.", + "C": "You have a remote control in your hand.", + "D": "You have a portable music player in your hand, small enough to grip in one hand.", + "E": "You have a handheld gaming device in your hand, with controls on both sides of the screen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 17 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_2_T2", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the bookshelf right now?", + "answer": "Yes, you can. The bookshelf is directly in front of you.", + "question_chinese": "我现在能直接够到书架吗？", + "answer_chinese": "是的，你能，书架在你的正前方", + "evidence": [ + { + "description": "This decoration is on the upper left of the bookshelf.", + "timestamps": { + "start": "00:34:47", + "end": "00:34:50" + } + }, + { + "description": "You are measuring the wall on the right side of the decoration, directly above the bookshelf, which is about 10cm away from you and in front of you.", + "timestamps": { + "start": "00:35:31", + "end": "00:35:40" + } + } + ], + "question_timestamps": "00:35:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can. The bookshelf is directly in front of you.", + "B": "No, you can't. The bookshelf is too far away to reach from where you are standing." + }, + "answer_label": "A", + "answer_text": "Yes, you can. The bookshelf is directly in front of you.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_2_T2", + "correct_answer": "Yes, you can. The bookshelf is directly in front of you.", + "original_options": { + "A": "Yes, you can. The bookshelf is directly in front of you.", + "B": "No, you can't. The bookshelf is too far away to reach from where you are standing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8", + "video_length": "00:58:24", + "qid": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_2_T3", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the bookshelf right now?", + "answer": "From your current position at the coffee table, move left, then turn right to get to the bookshelf where the book is.", + "question_chinese": "我现在能直接够到书架吗？", + "answer_chinese": "书从当前茶几位置,需向左移动,然后右转，到达书架所在位置。", + "evidence": [ + { + "description": "The bookshelf is located in the corner of the room. To reach it, you need to move from the coffee table to the left and then turn right.", + "timestamps": { + "start": "00:14:15", + "end": "00:14:20" + } + } + ], + "question_timestamps": "00:38:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can. The bookshelf is within arm’s reach right now.", + "B": "From your current position at the coffee table, move left, then turn right to get to the bookshelf where the book is." + }, + "answer_label": "B", + "answer_text": "From your current position at the coffee table, move left, then turn right to get to the bookshelf where the book is.", + "q_id": "1aa426b2-4d83-4b56-9d3c-4616adc464c8_2_T3", + "correct_answer": "From your current position at the coffee table, move left, then turn right to get to the bookshelf where the book is.", + "original_options": { + "A": "Yes, you can. The bookshelf is within arm’s reach right now.", + "B": "From your current position at the coffee table, move left, then turn right to get to the bookshelf where the book is." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + } +] diff --git a/QAs/1bec800a-c3cf-431f-bf0a-7632ad53bcb7-mcq.json b/QAs/1bec800a-c3cf-431f-bf0a-7632ad53bcb7-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a11b66572c13cfb3c3004d00c00a22906ff5a491 --- /dev/null +++ b/QAs/1bec800a-c3cf-431f-bf0a-7632ad53bcb7-mcq.json @@ -0,0 +1,1114 @@ +[ + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_3_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Before I reached the place where I pumped up the bike tire, how many cars did I see?", + "answer": "You saw a small car to your right.", + "question_chinese": "在我到达给自行车轮胎充气的地方之前，我看到了几辆小汽车？", + "answer_chinese": "你看到了一辆小汽车，在你的右边。", + "evidence": [ + { + "description": "You saw a car.", + "timestamps": { + "start": "00:03:59", + "end": "00:04:01" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw no cars.", + "B": "You saw a small car to your right.", + "C": "You saw two cars.", + "D": "You saw one car to your left before reaching the spot where you pumped up the bike tire.", + "E": "You saw three cars total: one to your left, one ahead of you, and one behind you." + }, + "answer_label": "B", + "answer_text": "You saw a small car to your right.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_3_T1", + "correct_answer": "You saw a small car to your right.", + "original_options": { + "A": "You saw no cars.", + "B": "You saw a small car to your right.", + "C": "You saw two cars.", + "D": "You saw one car to your left before reaching the spot where you pumped up the bike tire.", + "E": "You saw three cars total: one to your left, one ahead of you, and one behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 17 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_5", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Before I reached the place where I was supposed to inflate the bicycle tires, I saw a few cars?", + "answer": "You saw two cars.", + "question_chinese": "在我到达给自行车轮胎充气的地方之前，我看到了几辆小汽车？", + "answer_chinese": "你看到了两辆小汽车。", + "evidence": [ + { + "description": "You saw a car.", + "timestamps": { + "start": "00:03:59", + "end": "00:04:01" + } + }, + { + "description": "You saw the second car.", + "timestamps": { + "start": "00:04:14", + "end": "00:04:17" + } + } + ], + "question_timestamps": "00:06:48", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You saw two cars.", + "B": "You saw one car.", + "C": "You saw three cars.", + "D": "You saw four cars before reaching the place to inflate the bicycle tires, spread out along the road.", + "E": "You saw five cars in total before you arrived, with several passing as you continued forward." + }, + "answer_label": "A", + "answer_text": "You saw two cars.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_5", + "correct_answer": "You saw two cars.", + "original_options": { + "A": "You saw two cars.", + "B": "You saw one car.", + "C": "You saw three cars.", + "D": "You saw four cars before reaching the place to inflate the bicycle tires, spread out along the road.", + "E": "You saw five cars in total before you arrived, with several passing as you continued forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 16 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_3_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Before I reached the place where I pumped up the bike tire, how many cars did I see?", + "answer": "You saw two small cars: one passed in the opposite direction on your right, and the other drove along the road in front of you.", + "question_chinese": "在我到达给自行车轮胎充气的地方之前，我看到了几辆小汽车？", + "answer_chinese": "你看到了两辆小汽车，一辆是在你右方对向经过，一辆是从你面前的马路经过。", + "evidence": [ + { + "description": "You saw a car.", + "timestamps": { + "start": "00:03:59", + "end": "00:04:01" + } + }, + { + "description": "You saw the second car.", + "timestamps": { + "start": "00:04:14", + "end": "00:04:17" + } + } + ], + "question_timestamps": "00:21:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one small car.", + "B": "You saw three small cars.", + "C": "You saw two small cars: one passed in the opposite direction on your right, and the other drove along the road in front of you.", + "D": "You saw zero cars before reaching the place where you pumped up the bike tire, only an empty road.", + "E": "You saw four small cars: one passed opposite on your right, and three drove along the road ahead." + }, + "answer_label": "C", + "answer_text": "You saw two small cars: one passed in the opposite direction on your right, and the other drove along the road in front of you.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_3_T2", + "correct_answer": "You saw two small cars: one passed in the opposite direction on your right, and the other drove along the road in front of you.", + "original_options": { + "A": "You saw one small car.", + "B": "You saw three small cars.", + "C": "You saw two small cars: one passed in the opposite direction on your right, and the other drove along the road in front of you.", + "D": "You saw zero cars before reaching the place where you pumped up the bike tire, only an empty road.", + "E": "You saw four small cars: one passed opposite on your right, and three drove along the road ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 18 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_13", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Before I picked up the stapler, was it closer to the computer or the pen holder?", + "answer": "The stapler is closer to the pen holder.", + "question_chinese": "在我拿起订书机之前，它离电脑更近还是笔筒更近？", + "answer_chinese": "订书机离笔筒更近。", + "evidence": [ + { + "description": "The stapler is about three centimeters away from the pen holder and about forty centimeters away from the computer.", + "timestamps": { + "start": "00:18:08", + "end": "00:18:11" + } + } + ], + "question_timestamps": "00:18:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The stapler is closer to the pen holder.", + "B": "The stapler is closer to the computer." + }, + "answer_label": "A", + "answer_text": "The stapler is closer to the pen holder.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_13", + "correct_answer": "The stapler is closer to the pen holder.", + "original_options": { + "A": "The stapler is closer to the pen holder.", + "B": "The stapler is closer to the computer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_14", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Before I picked up the stapler, was it closer to the computer or the pen holder?", + "answer": "The stapler is closer to the computer.", + "question_chinese": "在我拿起订书机之前，它离电脑更近还是笔筒更近？", + "answer_chinese": "订书机离电脑更近。", + "evidence": [ + { + "description": "The stapler is about five centimeters away from the computer and about fifty centimeters away from the pen holder.", + "timestamps": { + "start": "00:18:33", + "end": "00:18:35" + } + } + ], + "question_timestamps": "00:18:36", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The stapler is closer to the computer.", + "B": "The stapler is closer to the pen holder." + }, + "answer_label": "A", + "answer_text": "The stapler is closer to the computer.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_14", + "correct_answer": "The stapler is closer to the computer.", + "original_options": { + "A": "The stapler is closer to the computer.", + "B": "The stapler is closer to the pen holder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_5_T1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I finished using the printer, how far away was I from the printer?", + "answer": "You are about 150 cm away from the printer, and it’s behind you.", + "question_chinese": "当我使用完打印机时,离打印机有多远?", + "answer_chinese": "你离打印机大约150cm,在你的后面。", + "evidence": [ + { + "description": "The printer is on your right, you can put the paper in directly.", + "timestamps": { + "start": "00:12:50", + "end": "00:12:54" + } + }, + { + "description": "You went around the printer from its left side, then walked away with your back to the printer.", + "timestamps": { + "start": "00:15:05", + "end": "00:15:10" + } + } + ], + "question_timestamps": "00:15:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "About 90 cm away, behind you.", + "B": "About 200 cm away, behind you.", + "C": "You are about 120 cm away from the printer, and it’s behind you when you finish.", + "D": "You are about 150 cm away from the printer, and it’s behind you.", + "E": "You are about 250 cm away from the printer, and it remains behind you after you’re done." + }, + "answer_label": "D", + "answer_text": "You are about 150 cm away from the printer, and it’s behind you.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_5_T1", + "correct_answer": "You are about 150 cm away from the printer, and it’s behind you.", + "original_options": { + "A": "About 90 cm away, behind you.", + "B": "About 200 cm away, behind you.", + "C": "You are about 120 cm away from the printer, and it’s behind you when you finish.", + "D": "You are about 150 cm away from the printer, and it’s behind you.", + "E": "You are about 250 cm away from the printer, and it remains behind you after you’re done." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_5_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I finished using the printer, how far away was I from the printer?", + "answer": "You are about 8 meters from the printer, and it is behind you.", + "question_chinese": "当我使用完打印机时,离打印机有多远?", + "answer_chinese": "你离打印机大约8m,在你的后面。", + "evidence": [ + { + "description": "You left from the printer’s on the right side, then walked a dozen or so steps.", + "timestamps": { + "start": "00:17:02", + "end": "00:17:30" + } + } + ], + "question_timestamps": "00:17:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "About 6 meters away, behind you.", + "B": "About 10 meters away, behind you.", + "C": "You are about 7 meters from the printer, and it is behind you when you finish.", + "D": "You are about 9 meters from the printer, and it is still behind you afterward.", + "E": "You are about 8 meters from the printer, and it is behind you." + }, + "answer_label": "E", + "answer_text": "You are about 8 meters from the printer, and it is behind you.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_5_T2", + "correct_answer": "You are about 8 meters from the printer, and it is behind you.", + "original_options": { + "A": "About 6 meters away, behind you.", + "B": "About 10 meters away, behind you.", + "C": "You are about 7 meters from the printer, and it is behind you when you finish.", + "D": "You are about 9 meters from the printer, and it is still behind you afterward.", + "E": "You are about 8 meters from the printer, and it is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_9", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand in front of the round table, how far is it from the printer?", + "answer": "You are about 50cm away from the printer, on your right.", + "question_chinese": "当我站在圆桌前时,离打印机有多远?", + "answer_chinese": "你离打印机大约50cm,在你的右边。", + "evidence": [ + { + "description": "The printer is on your right, you can put the paper in directly.", + "timestamps": { + "start": "00:12:50", + "end": "00:12:54" + } + } + ], + "question_timestamps": "00:13:38", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You are about 50cm away from the printer, on your right.", + "B": "About 30 cm away, on your left.", + "C": "About 90 cm away, behind you.", + "D": "You are about 70 cm away from the printer, and it is positioned in front of you.", + "E": "You are about 40 cm away from the printer, and it sits off to your left side." + }, + "answer_label": "A", + "answer_text": "You are about 50cm away from the printer, on your right.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_9", + "correct_answer": "You are about 50cm away from the printer, on your right.", + "original_options": { + "A": "You are about 50cm away from the printer, on your right.", + "B": "About 30 cm away, on your left.", + "C": "About 90 cm away, behind you.", + "D": "You are about 70 cm away from the printer, and it is positioned in front of you.", + "E": "You are about 40 cm away from the printer, and it sits off to your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 17 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_10", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand in front of the round table, how far is it from the printer?", + "answer": "You are about 50cm away from the printer, on your right.", + "question_chinese": "当我站在圆桌前时,离打印机有多远?", + "answer_chinese": "你离打印机大约50cm,在你的右边。", + "evidence": [ + { + "description": "The printer is on your right, you can put the paper in directly.", + "timestamps": { + "start": "00:12:35", + "end": "00:12:39" + } + } + ], + "question_timestamps": "00:26:33", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "About 30 cm away, on your left.", + "B": "You are about 50cm away from the printer, on your right.", + "C": "About 70 cm away, in front.", + "D": "You are about 100 cm away from the printer, and it is located to your left.", + "E": "You are about 40 cm away from the printer, and it is positioned behind you." + }, + "answer_label": "B", + "answer_text": "You are about 50cm away from the printer, on your right.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_10", + "correct_answer": "You are about 50cm away from the printer, on your right.", + "original_options": { + "A": "About 30 cm away, on your left.", + "B": "You are about 50cm away from the printer, on your right.", + "C": "About 70 cm away, in front.", + "D": "You are about 100 cm away from the printer, and it is located to your left.", + "E": "You are about 40 cm away from the printer, and it is positioned behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 15 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_3", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I take from my home to where I parked my bike?", + "answer": "You took four turns.", + "question_chinese": "我从家里走到我停自行车的地方，一共拐了几个弯？", + "answer_chinese": "你拐了四个弯。", + "evidence": [ + { + "description": "You took the first turn.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:24" + } + }, + { + "description": "You took the second turn.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:09" + } + }, + { + "description": "You took the third turn.", + "timestamps": { + "start": "00:02:48", + "end": "00:02:58" + } + }, + { + "description": "You took the last turn.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:12" + } + } + ], + "question_timestamps": "00:03:16", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You took two turns.", + "B": "You took three turns.", + "C": "You took four turns.", + "D": "You took five turns from your home before reaching the place you parked your bike.", + "E": "You took six turns in total on the way from home to where you left your bike." + }, + "answer_label": "C", + "answer_text": "You took four turns.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_3", + "correct_answer": "You took four turns.", + "original_options": { + "A": "You took two turns.", + "B": "You took three turns.", + "C": "You took four turns.", + "D": "You took five turns from your home before reaching the place you parked your bike.", + "E": "You took six turns in total on the way from home to where you left your bike." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_4", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I take from my home to where I parked my bike?", + "answer": "You took the last turn.", + "question_chinese": "我从家里走到我停自行车的地方，一共拐了几个弯？", + "answer_chinese": "你没有拐弯。", + "evidence": [ + { + "description": "You parked your bike right in front of your house, without having to make a turn.", + "timestamps": { + "start": "00:28:40", + "end": "00:28:45" + } + } + ], + "question_timestamps": "00:28:48", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You took the first turn.", + "B": "You took the second turn.", + "C": "You took the last turn.", + "D": "You took the second left turn on the way from your home to where you parked your bike.", + "E": "You took the third turn during the route from home before you reached the bike parking spot." + }, + "answer_label": "C", + "answer_text": "You took the last turn.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_4", + "correct_answer": "You took the last turn.", + "original_options": { + "A": "You took the first turn.", + "B": "You took the second turn.", + "C": "You took the last turn.", + "D": "You took the second left turn on the way from your home to where you parked your bike.", + "E": "You took the third turn during the route from home before you reached the bike parking spot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_6_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I used the printer just now, how many times did I swipe my card?", + "answer": "You swiped your card once.", + "question_chinese": "我刚刚使用打印机时，刷了几次卡？", + "answer_chinese": "你刷了一次卡。", + "evidence": [ + { + "description": "You swiped your card for the first time.", + "timestamps": { + "start": "00:13:00", + "end": "00:13:01" + } + } + ], + "question_timestamps": "00:14:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You swiped your card twice.", + "B": "You swiped your card three times.", + "C": "You swiped your card zero times before printing anything at the machine.", + "D": "You swiped your card four times in a row to authenticate and start the print job.", + "E": "You swiped your card once." + }, + "answer_label": "E", + "answer_text": "You swiped your card once.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_6_T1", + "correct_answer": "You swiped your card once.", + "original_options": { + "A": "You swiped your card twice.", + "B": "You swiped your card three times.", + "C": "You swiped your card zero times before printing anything at the machine.", + "D": "You swiped your card four times in a row to authenticate and start the print job.", + "E": "You swiped your card once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 16 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_11", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I swipe the card when I just used the printer?", + "answer": "You swiped the card twice.", + "question_chinese": "我刚刚使用打印机时，刷了几次卡？", + "answer_chinese": "你刷了两次卡。", + "evidence": [ + { + "description": "You swiped your card for the first time.", + "timestamps": { + "start": "00:13:00", + "end": "00:13:01" + } + }, + { + "description": "You swiped the card for the second time.", + "timestamps": { + "start": "00:15:45", + "end": "00:15:54" + } + } + ], + "question_timestamps": "00:15:57", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You swiped the card twice.", + "B": "You swiped the card once.", + "C": "You swiped the card three times.", + "D": "You swiped the card four times while trying to get the printer to accept it.", + "E": "You swiped the card five times because it didn’t read properly on the first attempts." + }, + "answer_label": "A", + "answer_text": "You swiped the card twice.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_11", + "correct_answer": "You swiped the card twice.", + "original_options": { + "A": "You swiped the card twice.", + "B": "You swiped the card once.", + "C": "You swiped the card three times.", + "D": "You swiped the card four times while trying to get the printer to accept it.", + "E": "You swiped the card five times because it didn’t read properly on the first attempts." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_6_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I used the printer just now, how many times did I swipe my card?", + "answer": "You swiped your card twice.", + "question_chinese": "我刚刚使用打印机时，刷了几次卡？", + "answer_chinese": "你刷了两次卡。", + "evidence": [ + { + "description": "You swiped your card for the first time.", + "timestamps": { + "start": "00:13:00", + "end": "00:13:01" + } + }, + { + "description": "You swiped the card for the second time.", + "timestamps": { + "start": "00:13:45", + "end": "00:13:54" + } + } + ], + "question_timestamps": "00:31:22", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You swiped your card twice.", + "B": "You swiped your card once.", + "C": "You swiped your card three times.", + "D": "You swiped your card four times to log in and release the print job.", + "E": "You didn’t swipe your card at all when you used the printer just now." + }, + "answer_label": "A", + "answer_text": "You swiped your card twice.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_6_T2", + "correct_answer": "You swiped your card twice.", + "original_options": { + "A": "You swiped your card twice.", + "B": "You swiped your card once.", + "C": "You swiped your card three times.", + "D": "You swiped your card four times to log in and release the print job.", + "E": "You didn’t swipe your card at all when you used the printer just now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 15 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_15", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans of oats are there on the shelf before I pick up one?", + "answer": "There are eight cans of oats on the shelf.", + "question_chinese": "我拿起一罐燕麦之前，货架上有几罐燕麦？", + "answer_chinese": "货架上有8罐燕麦。", + "evidence": [ + { + "description": "You picked up a can of oats.", + "timestamps": { + "start": "00:24:18", + "end": "00:24:22" + } + }, + { + "description": "There are still seven cans of oats on the shelf.", + "timestamps": { + "start": "00:24:23", + "end": "00:24:25" + } + } + ], + "question_timestamps": "00:24:28", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are six cans of oats.", + "B": "There are twelve cans of oats.", + "C": "There are eight cans of oats on the shelf.", + "D": "There are five cans of oats on the shelf before you pick one up.", + "E": "There are three cans of oats on the shelf before you reach out and take one." + }, + "answer_label": "C", + "answer_text": "There are eight cans of oats on the shelf.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_15", + "correct_answer": "There are eight cans of oats on the shelf.", + "original_options": { + "A": "There are six cans of oats.", + "B": "There are twelve cans of oats.", + "C": "There are eight cans of oats on the shelf.", + "D": "There are five cans of oats on the shelf before you pick one up.", + "E": "There are three cans of oats on the shelf before you reach out and take one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 16 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_16", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans of oats are there on the shelf before I pick up one?", + "answer": "There are ten cans of oats on the shelf.", + "question_chinese": "我拿起一罐燕麦之前，货架上有几罐燕麦？", + "answer_chinese": "货架上有10罐燕麦。", + "evidence": [ + { + "description": "There is a jar of oats on the third shelf from the bottom.", + "timestamps": { + "start": "00:24:30", + "end": "00:24:31" + } + }, + { + "description": "On the last shelf, there are seven cans on the outside and two cans on the inside.", + "timestamps": { + "start": "00:24:35", + "end": "00:24:40" + } + } + ], + "question_timestamps": "00:24:52", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are six cans of oats.", + "B": "There are twelve cans of oats.", + "C": "There are ten cans of oats on the shelf.", + "D": "There are five cans of oats on the shelf before you pick up one can.", + "E": "There are three cans of oats on the shelf before you grab one from the row." + }, + "answer_label": "C", + "answer_text": "There are ten cans of oats on the shelf.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_16", + "correct_answer": "There are ten cans of oats on the shelf.", + "original_options": { + "A": "There are six cans of oats.", + "B": "There are twelve cans of oats.", + "C": "There are ten cans of oats on the shelf.", + "D": "There are five cans of oats on the shelf before you pick up one can.", + "E": "There are three cans of oats on the shelf before you grab one from the row." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I am facing the mirror now. Is the washing machine on my left, right, or behind me?", + "answer": "The washing machine is behind you.", + "question_chinese": "我现在面向镜子，洗衣机在我的左边，右边还是后面？", + "answer_chinese": "洗衣机在你的后面。", + "evidence": [ + { + "description": "The washing machine is opposite the toilet, facing the mirror.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The washing machine is to your left.", + "B": "The washing machine is to your right.", + "C": "The washing machine is in front of you when facing the mirror, not directly behind you.", + "D": "The washing machine is behind you.", + "E": "The washing machine is above you while you look at the mirror, rather than behind your back." + }, + "answer_label": "D", + "answer_text": "The washing machine is behind you.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_1", + "correct_answer": "The washing machine is behind you.", + "original_options": { + "A": "The washing machine is to your left.", + "B": "The washing machine is to your right.", + "C": "The washing machine is in front of you when facing the mirror, not directly behind you.", + "D": "The washing machine is behind you.", + "E": "The washing machine is above you while you look at the mirror, rather than behind your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I am facing the mirror now. Is the washing machine on my left, right, or behind me?", + "answer": "The washing machine is behind you.", + "question_chinese": "我现在面向镜子，洗衣机在我的左边，右边还是后面？", + "answer_chinese": "洗衣机在你的后面。", + "evidence": [ + { + "description": "A washing machine appears in the mirror, behind you.", + "timestamps": { + "start": "00:30:16", + "end": "00:30:18" + } + } + ], + "question_timestamps": "00:30:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The washing machine is to your left.", + "B": "The washing machine is behind you.", + "C": "The washing machine is to your right.", + "D": "When you are facing the mirror, the washing machine is in front of you, not behind you.", + "E": "Facing the mirror, the washing machine is positioned to your left, instead of directly behind you." + }, + "answer_label": "B", + "answer_text": "The washing machine is behind you.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_2", + "correct_answer": "The washing machine is behind you.", + "original_options": { + "A": "The washing machine is to your left.", + "B": "The washing machine is behind you.", + "C": "The washing machine is to your right.", + "D": "When you are facing the mirror, the washing machine is in front of you, not behind you.", + "E": "Facing the mirror, the washing machine is positioned to your left, instead of directly behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_1_T1", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Based on where I am right now, where is the washing machine relative to me?", + "answer": "The washing machine is behind you.", + "question_chinese": "根据我现在所在的位置，洗衣机在我的什么方位？", + "answer_chinese": "洗衣机在你的后面。", + "evidence": [ + { + "description": "The washing machine is opposite the toilet, facing the mirror.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The washing machine is behind you.", + "B": "The washing machine is in front of you.", + "C": "The washing machine is to your left.", + "D": "The washing machine is to your right relative to where you are standing right now.", + "E": "The washing machine is above you, not behind, from where you are currently standing." + }, + "answer_label": "A", + "answer_text": "The washing machine is behind you.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_1_T1", + "correct_answer": "The washing machine is behind you.", + "original_options": { + "A": "The washing machine is behind you.", + "B": "The washing machine is in front of you.", + "C": "The washing machine is to your left.", + "D": "The washing machine is to your right relative to where you are standing right now.", + "E": "The washing machine is above you, not behind, from where you are currently standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 14 + ] + }, + { + "video_uid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7", + "video_length": "00:43:25", + "qid": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_1_T2", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Based on where I am right now, where is the washing machine relative to me?", + "answer": "The washing machine is to your right.", + "question_chinese": "根据我现在所在的位置，洗衣机在我的什么方位？", + "answer_chinese": "洗衣机在你的右边。", + "evidence": [ + { + "description": "The washing machine door appeared to your front-right.", + "timestamps": { + "start": "00:30:12", + "end": "00:30:13" + } + } + ], + "question_timestamps": "00:30:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The washing machine is to your left.", + "B": "The washing machine is in front of you.", + "C": "The washing machine is behind you, not on your right side where you’re facing now.", + "D": "The washing machine is to your right.", + "E": "The washing machine is above you relative to your current position, rather than beside you." + }, + "answer_label": "D", + "answer_text": "The washing machine is to your right.", + "q_id": "1bec800a-c3cf-431f-bf0a-7632ad53bcb7_1_T2", + "correct_answer": "The washing machine is to your right.", + "original_options": { + "A": "The washing machine is to your left.", + "B": "The washing machine is in front of you.", + "C": "The washing machine is behind you, not on your right side where you’re facing now.", + "D": "The washing machine is to your right.", + "E": "The washing machine is above you relative to your current position, rather than beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 15 + ] + } +] diff --git a/QAs/1c7426ad-b6ba-4494-897c-537dd4a201e9-mcq.json b/QAs/1c7426ad-b6ba-4494-897c-537dd4a201e9-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2f35a2620b056bcb738f58016fa00cbc859869d2 --- /dev/null +++ b/QAs/1c7426ad-b6ba-4494-897c-537dd4a201e9-mcq.json @@ -0,0 +1,433 @@ +[ + { + "video_uid": "1c7426ad-b6ba-4494-897c-537dd4a201e9", + "video_length": "01:09:46", + "qid": "1c7426ad-b6ba-4494-897c-537dd4a201e9_3_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to use the vernier caliper, where will I take it from?", + "answer": "You’re going to pick up the vernier caliper from in front of you.", + "question_chinese": "我想用游标卡尺，我将从哪里拿？", + "answer_chinese": "你将从你的前面拿游标卡尺。", + "evidence": [ + { + "description": "I placed the vernier caliper on the workbench.", + "timestamps": { + "start": "00:53:00", + "end": "00:53:24" + } + } + ], + "question_timestamps": "00:53:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’ll take it from behind you.", + "B": "You’re going to pick up the vernier caliper from in front of you.", + "C": "You’ll take it from your left.", + "D": "You’re going to pick up the vernier caliper from your right side instead of straight ahead.", + "E": "You’re going to pick up the vernier caliper from above you, not from the area in front." + }, + "answer_label": "B", + "answer_text": "You’re going to pick up the vernier caliper from in front of you.", + "q_id": "1c7426ad-b6ba-4494-897c-537dd4a201e9_3_T1", + "correct_answer": "You’re going to pick up the vernier caliper from in front of you.", + "original_options": { + "A": "You’ll take it from behind you.", + "B": "You’re going to pick up the vernier caliper from in front of you.", + "C": "You’ll take it from your left.", + "D": "You’re going to pick up the vernier caliper from your right side instead of straight ahead.", + "E": "You’re going to pick up the vernier caliper from above you, not from the area in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "1c7426ad-b6ba-4494-897c-537dd4a201e9", + "video_length": "01:09:46", + "qid": "1c7426ad-b6ba-4494-897c-537dd4a201e9_6", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to use a vernier caliper, where should I get it from?", + "answer": "You will take the vernier caliper from the workbench.", + "question_chinese": "我想用游标卡尺，我将从哪里拿？", + "answer_chinese": "你将从工作台上拿游标卡尺。", + "evidence": [ + { + "description": "I placed the vernier caliper on the workbench.", + "timestamps": { + "start": "00:53:00", + "end": "00:53:24" + } + } + ], + "question_timestamps": "00:53:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You will take it from the drawer.", + "B": "You will get it from the shelf.", + "C": "You will take the vernier caliper from the workbench.", + "D": "You will grab the vernier caliper from the counter rather than from the main work surface.", + "E": "You will pick up the vernier caliper from the floor instead of taking it off the work surface." + }, + "answer_label": "C", + "answer_text": "You will take the vernier caliper from the workbench.", + "q_id": "1c7426ad-b6ba-4494-897c-537dd4a201e9_6", + "correct_answer": "You will take the vernier caliper from the workbench.", + "original_options": { + "A": "You will take it from the drawer.", + "B": "You will get it from the shelf.", + "C": "You will take the vernier caliper from the workbench.", + "D": "You will grab the vernier caliper from the counter rather than from the main work surface.", + "E": "You will pick up the vernier caliper from the floor instead of taking it off the work surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "1c7426ad-b6ba-4494-897c-537dd4a201e9", + "video_length": "01:09:46", + "qid": "1c7426ad-b6ba-4494-897c-537dd4a201e9_3_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to use the vernier caliper, where will I take it from?", + "answer": "You’re going to pick up the vernier caliper from your right.", + "question_chinese": "我想用游标卡尺，我将从哪里拿？", + "answer_chinese": "你将从右边拿游标卡尺。", + "evidence": [ + { + "description": "I placed the vernier caliper on the workbench.", + "timestamps": { + "start": "01:08:40", + "end": "01:08:45" + } + } + ], + "question_timestamps": "01:08:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’re going to pick up the vernier caliper from your right.", + "B": "From your left.", + "C": "From behind you.", + "D": "You’re going to pick up the vernier caliper from in front of you rather than from your right side.", + "E": "You will take the vernier caliper from the workbench instead of picking it up from your right." + }, + "answer_label": "A", + "answer_text": "You’re going to pick up the vernier caliper from your right.", + "q_id": "1c7426ad-b6ba-4494-897c-537dd4a201e9_3_T2", + "correct_answer": "You’re going to pick up the vernier caliper from your right.", + "original_options": { + "A": "You’re going to pick up the vernier caliper from your right.", + "B": "From your left.", + "C": "From behind you.", + "D": "You’re going to pick up the vernier caliper from in front of you rather than from your right side.", + "E": "You will take the vernier caliper from the workbench instead of picking it up from your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 20, + 17 + ] + }, + { + "video_uid": "1c7426ad-b6ba-4494-897c-537dd4a201e9", + "video_length": "01:09:46", + "qid": "1c7426ad-b6ba-4494-897c-537dd4a201e9_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where will I put the saw after I finish using it?", + "answer": "You will place the saw on the workbench.", + "question_chinese": "我用完锯子后，将会把锯子放在哪里？", + "answer_chinese": "你将把锯子放在工作台上。", + "evidence": [ + { + "description": "You placed the saw on the workbench.", + "timestamps": { + "start": "00:07:30", + "end": "00:07:50" + } + } + ], + "question_timestamps": "00:34:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You will put the saw on the floor.", + "B": "You will place the saw in the toolbox.", + "C": "You will store the saw under the shelf instead of putting it onto the work surface.", + "D": "You will place the saw on the workbench.", + "E": "You will hang the saw on the wall rather than setting it down on a workbench." + }, + "answer_label": "D", + "answer_text": "You will place the saw on the workbench.", + "q_id": "1c7426ad-b6ba-4494-897c-537dd4a201e9_4", + "correct_answer": "You will place the saw on the workbench.", + "original_options": { + "A": "You will put the saw on the floor.", + "B": "You will place the saw in the toolbox.", + "C": "You will store the saw under the shelf instead of putting it onto the work surface.", + "D": "You will place the saw on the workbench.", + "E": "You will hang the saw on the wall rather than setting it down on a workbench." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 16 + ] + }, + { + "video_uid": "1c7426ad-b6ba-4494-897c-537dd4a201e9", + "video_length": "01:09:46", + "qid": "1c7426ad-b6ba-4494-897c-537dd4a201e9_1_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The first time I picked up the hammer, where did I take it from?", + "answer": "The first time, you picked up the hammer from the left edge of the workbench. Now the workbench is behind you.", + "question_chinese": "我第一次拿锤子时，从哪里拿的锤子？", + "answer_chinese": "第一次你从工作台左侧边缘拿到了锤子，现在工作台在你的后面。", + "evidence": [ + { + "description": "I got the hammer from the left edge of the workbench.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:14" + } + }, + { + "description": "The workbench is behind you now.", + "timestamps": { + "start": "00:10:40", + "end": "00:10:47" + } + } + ], + "question_timestamps": "00:10:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the workbench center.", + "B": "From the right edge.", + "C": "The first time, you picked up the hammer from the left edge of the workbench. Now the workbench is behind you.", + "D": "First time, you picked it from the left edge of the workbench; now the bench is in front of you.", + "E": "First time, you picked it from the left edge of the workbench; now the bench is to your right." + }, + "answer_label": "C", + "answer_text": "The first time, you picked up the hammer from the left edge of the workbench. Now the workbench is behind you.", + "q_id": "1c7426ad-b6ba-4494-897c-537dd4a201e9_1_T1", + "correct_answer": "The first time, you picked up the hammer from the left edge of the workbench. Now the workbench is behind you.", + "original_options": { + "A": "From the workbench center.", + "B": "From the right edge.", + "C": "The first time, you picked up the hammer from the left edge of the workbench. Now the workbench is behind you.", + "D": "First time, you picked it from the left edge of the workbench; now the bench is in front of you.", + "E": "First time, you picked it from the left edge of the workbench; now the bench is to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 19 + ] + }, + { + "video_uid": "1c7426ad-b6ba-4494-897c-537dd4a201e9", + "video_length": "01:09:46", + "qid": "1c7426ad-b6ba-4494-897c-537dd4a201e9_1_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The first time I picked up the hammer, where did I take it from?", + "answer": "The first time, you picked up the hammer from the left edge of the workbench, and the workbench was in front of you.", + "question_chinese": "我第一次拿锤子时，从哪里拿的锤子？", + "answer_chinese": "第一次你从工作台左侧边缘拿到了锤子，工作台在你的前方。", + "evidence": [ + { + "description": "I got the hammer from the left edge of the workbench and the workbench is in front of you now.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:14" + } + } + ], + "question_timestamps": "01:09:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the right edge of the workbench.", + "B": "From the middle of the workbench.", + "C": "The first time, you picked up the hammer from the left edge of the workbench, and the workbench was in front of you.", + "D": "The first time, you picked up the hammer from the right edge of the workbench, with it in front of you.", + "E": "The first time, you picked up the hammer from the left edge of the workbench, but it was behind you." + }, + "answer_label": "C", + "answer_text": "The first time, you picked up the hammer from the left edge of the workbench, and the workbench was in front of you.", + "q_id": "1c7426ad-b6ba-4494-897c-537dd4a201e9_1_T2", + "correct_answer": "The first time, you picked up the hammer from the left edge of the workbench, and the workbench was in front of you.", + "original_options": { + "A": "From the right edge of the workbench.", + "B": "From the middle of the workbench.", + "C": "The first time, you picked up the hammer from the left edge of the workbench, and the workbench was in front of you.", + "D": "The first time, you picked up the hammer from the right edge of the workbench, with it in front of you.", + "E": "The first time, you picked up the hammer from the left edge of the workbench, but it was behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 21, + 20 + ] + }, + { + "video_uid": "1c7426ad-b6ba-4494-897c-537dd4a201e9", + "video_length": "01:09:46", + "qid": "1c7426ad-b6ba-4494-897c-537dd4a201e9_2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I get the hammer from when I first took it?", + "answer": "I got the hammer from the left edge of the workbench.", + "question_chinese": "我第一次拿锤子时，时从哪里拿的锤子？", + "answer_chinese": "我从工作台左侧边缘拿到了锤子。", + "evidence": [ + { + "description": "I got the hammer from the left edge of the workbench.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:14" + } + } + ], + "question_timestamps": "00:17:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I got the hammer from the left edge of the workbench.", + "B": "From the right edge.", + "C": "From the middle of the workbench.", + "D": "I got the hammer from the front edge of the workbench, closest to where I was standing.", + "E": "I got the hammer from the back edge of the workbench, farther away from me at the time." + }, + "answer_label": "A", + "answer_text": "I got the hammer from the left edge of the workbench.", + "q_id": "1c7426ad-b6ba-4494-897c-537dd4a201e9_2", + "correct_answer": "I got the hammer from the left edge of the workbench.", + "original_options": { + "A": "I got the hammer from the left edge of the workbench.", + "B": "From the right edge.", + "C": "From the middle of the workbench.", + "D": "I got the hammer from the front edge of the workbench, closest to where I was standing.", + "E": "I got the hammer from the back edge of the workbench, farther away from me at the time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 17, + 18 + ] + }, + { + "video_uid": "1c7426ad-b6ba-4494-897c-537dd4a201e9", + "video_length": "01:09:46", + "qid": "1c7426ad-b6ba-4494-897c-537dd4a201e9_1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I get the hammer from when I first took it?", + "answer": "I got the hammer from the left edge of the workbench.", + "question_chinese": "我第一次拿锤子时，时从哪里拿的锤子？", + "answer_chinese": "我从工作台左侧边缘拿到了锤子。", + "evidence": [ + { + "description": "I got the hammer from the left edge of the workbench.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:14" + } + } + ], + "question_timestamps": "01:09:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the right edge.", + "B": "I got the hammer from the left edge of the workbench.", + "C": "From the center of the workbench.", + "D": "You got the hammer from the front edge of the workbench, nearest to you when you reached for it.", + "E": "You got the hammer from the back corner of the workbench, not from the left edge." + }, + "answer_label": "B", + "answer_text": "I got the hammer from the left edge of the workbench.", + "q_id": "1c7426ad-b6ba-4494-897c-537dd4a201e9_1", + "correct_answer": "I got the hammer from the left edge of the workbench.", + "original_options": { + "A": "From the right edge.", + "B": "I got the hammer from the left edge of the workbench.", + "C": "From the center of the workbench.", + "D": "You got the hammer from the front edge of the workbench, nearest to you when you reached for it.", + "E": "You got the hammer from the back corner of the workbench, not from the left edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 19, + 16 + ] + } +] diff --git a/QAs/1cd5d583-afc5-424b-a56c-55a250e80a30-mcq.json b/QAs/1cd5d583-afc5-424b-a56c-55a250e80a30-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..32ec94cc25e6d788923ca92795c023981f239bff --- /dev/null +++ b/QAs/1cd5d583-afc5-424b-a56c-55a250e80a30-mcq.json @@ -0,0 +1,696 @@ +[ + { + "video_uid": "1cd5d583-afc5-424b-a56c-55a250e80a30", + "video_length": "00:42:08", + "qid": "1cd5d583-afc5-424b-a56c-55a250e80a30_16", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many black tubes have I used so far to make the fence?", + "answer": "You used eleven black tubes.", + "question_chinese": "到目前为止我用了多少根黑色管子用于制作围栏？", + "answer_chinese": "你用了11根黑色管子。", + "evidence": [ + { + "description": "You used 11 black tubes.", + "timestamps": { + "start": "00:03:27", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:10:32", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used nine black tubes.", + "B": "You used twelve black tubes.", + "C": "You used ten black tubes in total so far to assemble the fence.", + "D": "You used thirteen black tubes overall while building the fence up to this point.", + "E": "You used eleven black tubes." + }, + "answer_label": "E", + "answer_text": "You used eleven black tubes.", + "q_id": "1cd5d583-afc5-424b-a56c-55a250e80a30_16", + "correct_answer": "You used eleven black tubes.", + "original_options": { + "A": "You used nine black tubes.", + "B": "You used twelve black tubes.", + "C": "You used ten black tubes in total so far to assemble the fence.", + "D": "You used thirteen black tubes overall while building the fence up to this point.", + "E": "You used eleven black tubes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "1cd5d583-afc5-424b-a56c-55a250e80a30", + "video_length": "00:42:08", + "qid": "1cd5d583-afc5-424b-a56c-55a250e80a30_15", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many black tubes have I used so far to make the fence?", + "answer": "You used twenty-three black tubes.", + "question_chinese": "到目前为止我用了多少根黑色管子用于制作围栏？", + "answer_chinese": "你用了23根黑色管子。", + "evidence": [ + { + "description": "You used 11 black tubes.", + "timestamps": { + "start": "00:03:27", + "end": "00:04:12" + } + }, + { + "description": "You have used another 12 black tubes.", + "timestamps": { + "start": "00:22:05", + "end": "00:23:14" + } + } + ], + "question_timestamps": "00:31:36", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You used twenty-one black tubes.", + "B": "You used twenty-five black tubes.", + "C": "You used twenty-two black tubes so far while making the fence sections.", + "D": "You used twenty-six black tubes in total up to this point in the fence build.", + "E": "You used twenty-three black tubes." + }, + "answer_label": "E", + "answer_text": "You used twenty-three black tubes.", + "q_id": "1cd5d583-afc5-424b-a56c-55a250e80a30_15", + "correct_answer": "You used twenty-three black tubes.", + "original_options": { + "A": "You used twenty-one black tubes.", + "B": "You used twenty-five black tubes.", + "C": "You used twenty-two black tubes so far while making the fence sections.", + "D": "You used twenty-six black tubes in total up to this point in the fence build.", + "E": "You used twenty-three black tubes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 15 + ] + }, + { + "video_uid": "1cd5d583-afc5-424b-a56c-55a250e80a30", + "video_length": "00:42:08", + "qid": "1cd5d583-afc5-424b-a56c-55a250e80a30_12", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to saw the wooden board in front of me into two pieces, how should I carry the board?", + "answer": "You will walk straight to the back right towards the staircase entrance, go down the stairs, walk straight to the miter saw, and then you can use the miter saw to cut the board into two pieces.", + "question_chinese": "如果我想把面前的木板锯成两段，我将会带着木板怎么走？", + "answer_chinese": "你将会向右后方直走到楼梯口，下楼梯，直走到斜切锯处，然后你可以用斜切锯将木板切成两段。", + "evidence": [ + { + "description": "I walked straight from the miter saw to the staircase, went up the stairs, then walked straight towards the front right to reach my current location.", + "timestamps": { + "start": "00:08:48", + "end": "00:09:09" + } + }, + { + "description": "I can cut the wooden board into two pieces with a miter saw.", + "timestamps": { + "start": "00:07:06", + "end": "00:07:41" + } + } + ], + "question_timestamps": "00:10:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk to the back left staircase.", + "B": "Head to the front right staircase.", + "C": "Walk straight toward the staircase entrance on the back left, go downstairs, then continue to the miter saw to cut it.", + "D": "Turn left and go to the staircase entrance, but stay upstairs and cut the board there instead of going down.", + "E": "You will walk straight to the back right towards the staircase entrance, go down the stairs, walk straight to the miter saw, and then you can use the miter saw to cut the board into two pieces." + }, + "answer_label": "E", + "answer_text": "You will walk straight to the back right towards the staircase entrance, go down the stairs, walk straight to the miter saw, and then you can use the miter saw to cut the board into two pieces.", + "q_id": "1cd5d583-afc5-424b-a56c-55a250e80a30_12", + "correct_answer": "You will walk straight to the back right towards the staircase entrance, go down the stairs, walk straight to the miter saw, and then you can use the miter saw to cut the board into two pieces.", + "original_options": { + "A": "Walk to the back left staircase.", + "B": "Head to the front right staircase.", + "C": "Walk straight toward the staircase entrance on the back left, go downstairs, then continue to the miter saw to cut it.", + "D": "Turn left and go to the staircase entrance, but stay upstairs and cut the board there instead of going down.", + "E": "You will walk straight to the back right towards the staircase entrance, go down the stairs, walk straight to the miter saw, and then you can use the miter saw to cut the board into two pieces." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 20 + ] + }, + { + "video_uid": "1cd5d583-afc5-424b-a56c-55a250e80a30", + "video_length": "00:42:08", + "qid": "1cd5d583-afc5-424b-a56c-55a250e80a30_11", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to saw the wooden board in front of me into two pieces, how should I carry the board?", + "answer": "I will start from where I am now, turn left, walk straight to the staircase entrance, go down the stairs, and walk straight to the miter saw.I can cut the wooden board into two pieces with a miter saw.", + "question_chinese": "如果我想把面前的木板锯成两段，我将会带着木板怎么走？", + "answer_chinese": "我将从我现在的位置开始，向左转，直走到楼梯入口处，下楼后直走到圆锯那里。我可以用圆锯把木板切成两块。", + "evidence": [ + { + "description": "I will start from where I am now, turn left, walk straight to the staircase entrance, go down the stairs, and walk straight to the miter saw.", + "timestamps": { + "start": "00:06:37", + "end": "00:06:51" + } + }, + { + "description": "I can cut the wooden board into two pieces with a miter saw.", + "timestamps": { + "start": "00:07:12", + "end": "00:07:42" + } + } + ], + "question_timestamps": "00:21:04", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I will start from where I am now, turn left, walk straight to the staircase entrance, go down the stairs, and walk straight to the miter saw.I can cut the wooden board into two pieces with a miter saw.", + "B": "Turn right and head to the stairs.", + "C": "Turn around and walk backward.", + "D": "Start where you are, turn right, go downstairs, then walk straight to the workbench area instead.", + "E": "Turn left to the staircase, go upstairs, then look for a hand saw instead of using the miter saw." + }, + "answer_label": "A", + "answer_text": "I will start from where I am now, turn left, walk straight to the staircase entrance, go down the stairs, and walk straight to the miter saw.I can cut the wooden board into two pieces with a miter saw.", + "q_id": "1cd5d583-afc5-424b-a56c-55a250e80a30_11", + "correct_answer": "I will start from where I am now, turn left, walk straight to the staircase entrance, go down the stairs, and walk straight to the miter saw.I can cut the wooden board into two pieces with a miter saw.", + "original_options": { + "A": "I will start from where I am now, turn left, walk straight to the staircase entrance, go down the stairs, and walk straight to the miter saw.I can cut the wooden board into two pieces with a miter saw.", + "B": "Turn right and head to the stairs.", + "C": "Turn around and walk backward.", + "D": "Start where you are, turn right, go downstairs, then walk straight to the workbench area instead.", + "E": "Turn left to the staircase, go upstairs, then look for a hand saw instead of using the miter saw." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 19 + ] + }, + { + "video_uid": "1cd5d583-afc5-424b-a56c-55a250e80a30", + "video_length": "00:42:08", + "qid": "1cd5d583-afc5-424b-a56c-55a250e80a30_7_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What places have I put the electric drill down?", + "answer": "You previously placed the electric drill on the wooden floor next to the second-floor railing, on the second-floor railing, on the grass beside the stairs on the first floor, on a stand on the second floor, and on the wood board on that stand that was being worked on. All of these spots were in front of you.", + "question_chinese": "我把电钻都放到过哪些地方？", + "answer_chinese": "你曾经把电钻放到过二楼栏杆旁的木地板上、二楼栏杆上、一楼楼梯旁的草地上、二楼上的一个支架上以及在支架上的待加工的木板上。都在你的前方。", + "evidence": [ + { + "description": "I placed the power drill on the second-floor rack where the wood to be processed is.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:23" + } + }, + { + "description": "I put the power drill on the hardwood floor next to the railing on the second floor.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:00" + } + }, + { + "description": "I put the power drill on the wooden floor next to the railing on the second floor.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:18" + } + }, + { + "description": "I put the power drill on the wooden floor next to the railing on the second floor.", + "timestamps": { + "start": "00:02:37", + "end": "00:02:50" + } + }, + { + "description": "I put the power drill on the wooden floor next to the railing on the second floor.", + "timestamps": { + "start": "00:02:54", + "end": "00:02:59" + } + }, + { + "description": "I put the power drill on the wooden floor next to the railing on the second floor.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:04" + } + }, + { + "description": "I put the power drill on the wooden floor next to the railing on the second floor.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:18" + } + }, + { + "description": "I put the power drill on the wooden floor next to the railing on the second floor.", + "timestamps": { + "start": "00:04:46", + "end": "00:04:56" + } + }, + { + "description": "I put the power drill on the stand on the second-floor wooden floor.", + "timestamps": { + "start": "00:05:53", + "end": "00:06:00" + } + }, + { + "description": "I put the power drill on the wooden railing of the second floor.", + "timestamps": { + "start": "00:04:58", + "end": "00:05:10" + } + }, + { + "description": "I put the power drill on the railing of the second floor.", + "timestamps": { + "start": "00:13:55", + "end": "00:14:00" + } + }, + { + "description": "I put the power drill on the grass next to the stairs on the first floor.", + "timestamps": { + "start": "00:28:50", + "end": "00:29:53" + } + } + ], + "question_timestamps": "00:31:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You previously placed the electric drill on the wooden floor next to the second-floor railing, on the second-floor railing, on the grass beside the stairs on the first floor, on a stand on the second floor, and on the wood board on that stand that was being worked on. All of these spots were in front of you.", + "B": "Only on the second-floor railing.", + "C": "Only on the stand upstairs.", + "D": "You placed the electric drill on the wooden floor by the second-floor railing, and later on the grass beside the first-floor stairs.", + "E": "You placed the electric drill on the second-floor railing, then on the wooden floor beside it, and finally on the stand upstairs." + }, + "answer_label": "A", + "answer_text": "You previously placed the electric drill on the wooden floor next to the second-floor railing, on the second-floor railing, on the grass beside the stairs on the first floor, on a stand on the second floor, and on the wood board on that stand that was being worked on. All of these spots were in front of you.", + "q_id": "1cd5d583-afc5-424b-a56c-55a250e80a30_7_T1", + "correct_answer": "You previously placed the electric drill on the wooden floor next to the second-floor railing, on the second-floor railing, on the grass beside the stairs on the first floor, on a stand on the second floor, and on the wood board on that stand that was being worked on. All of these spots were in front of you.", + "original_options": { + "A": "You previously placed the electric drill on the wooden floor next to the second-floor railing, on the second-floor railing, on the grass beside the stairs on the first floor, on a stand on the second floor, and on the wood board on that stand that was being worked on. All of these spots were in front of you.", + "B": "Only on the second-floor railing.", + "C": "Only on the stand upstairs.", + "D": "You placed the electric drill on the wooden floor by the second-floor railing, and later on the grass beside the first-floor stairs.", + "E": "You placed the electric drill on the second-floor railing, then on the wooden floor beside it, and finally on the stand upstairs." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 22, + 22 + ] + }, + { + "video_uid": "1cd5d583-afc5-424b-a56c-55a250e80a30", + "video_length": "00:42:08", + "qid": "1cd5d583-afc5-424b-a56c-55a250e80a30_7_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What places have I put the electric drill down?", + "answer": "You previously placed the electric drill on the wooden floor next to the second-floor railing, on the second-floor railing, on the grass beside the stairs on the first floor, on a stand on the second floor, and on the wood board on that stand that was being worked on. These locations were to your right and behind you, respectively.", + "question_chinese": "我把电钻都放到过哪些地方？", + "answer_chinese": "你曾经把电钻放到过二楼栏杆旁的木地板上、二楼栏杆上、一楼楼梯旁的草地上、二楼上的一个支架上以及在支架上的待加工的木板上。分别在你的右方和后方。", + "evidence": [ + { + "description": "I placed the power drill on the second-floor rack where the wood to be processed is.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:23" + } + }, + { + "description": "I put the power drill on the hardwood floor next to the railing on the second floor.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:00" + } + }, + { + "description": "I put the power drill on the wooden floor next to the railing on the second floor.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:18" + } + }, + { + "description": "I put the power drill on the wooden floor next to the railing on the second floor.", + "timestamps": { + "start": "00:02:54", + "end": "00:02:59" + } + }, + { + "description": "I put the power drill on the wooden floor next to the railing on the second floor.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:04" + } + }, + { + "description": "I put the power drill on the wooden floor next to the railing on the second floor.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:18" + } + }, + { + "description": "I put the power drill on the wooden floor next to the railing on the second floor.", + "timestamps": { + "start": "00:04:46", + "end": "00:04:56" + } + }, + { + "description": "I put the power drill on the stand on the second-floor wooden floor.", + "timestamps": { + "start": "00:05:53", + "end": "00:06:00" + } + }, + { + "description": "I put the power drill on the wooden railing of the second floor.", + "timestamps": { + "start": "00:04:58", + "end": "00:05:10" + } + }, + { + "description": "I put the power drill on the railing of the second floor.", + "timestamps": { + "start": "00:13:55", + "end": "00:14:00" + } + }, + { + "description": "I put the power drill on the grass next to the stairs on the first floor.", + "timestamps": { + "start": "00:28:50", + "end": "00:29:53" + } + } + ], + "question_timestamps": "00:31:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Only on the second-floor railing.", + "B": "You previously placed the electric drill on the wooden floor next to the second-floor railing, on the second-floor railing, on the grass beside the stairs on the first floor, on a stand on the second floor, and on the wood board on that stand that was being worked on. These locations were to your right and behind you, respectively.", + "C": "On the grass beside the stairs.", + "D": "You put the drill on the wooden floor by the railing and later on the stand, both in front of you.", + "E": "You set it on the second-floor railing, then on the wood board on the stand, all to your left." + }, + "answer_label": "B", + "answer_text": "You previously placed the electric drill on the wooden floor next to the second-floor railing, on the second-floor railing, on the grass beside the stairs on the first floor, on a stand on the second floor, and on the wood board on that stand that was being worked on. These locations were to your right and behind you, respectively.", + "q_id": "1cd5d583-afc5-424b-a56c-55a250e80a30_7_T2", + "correct_answer": "You previously placed the electric drill on the wooden floor next to the second-floor railing, on the second-floor railing, on the grass beside the stairs on the first floor, on a stand on the second floor, and on the wood board on that stand that was being worked on. These locations were to your right and behind you, respectively.", + "original_options": { + "A": "Only on the second-floor railing.", + "B": "You previously placed the electric drill on the wooden floor next to the second-floor railing, on the second-floor railing, on the grass beside the stairs on the first floor, on a stand on the second floor, and on the wood board on that stand that was being worked on. These locations were to your right and behind you, respectively.", + "C": "On the grass beside the stairs.", + "D": "You put the drill on the wooden floor by the railing and later on the stand, both in front of you.", + "E": "You set it on the second-floor railing, then on the wood board on the stand, all to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 21, + 19 + ] + }, + { + "video_uid": "1cd5d583-afc5-424b-a56c-55a250e80a30", + "video_length": "00:42:08", + "qid": "1cd5d583-afc5-424b-a56c-55a250e80a30_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I last saw the wooden staircase, was there anyone else on it?", + "answer": "There was no one else on the wooden staircase.", + "question_chinese": "我最后一次看到木楼梯时，木楼梯上是否有其他人？", + "answer_chinese": "木楼梯上没有其他人。", + "evidence": [ + { + "description": "No one else was standing on the wooden staircase", + "timestamps": { + "start": "00:10:12", + "end": "00:10:17" + } + } + ], + "question_timestamps": "00:10:32", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "There was no one else on the wooden staircase.", + "B": "There was someone else on the wooden staircase." + }, + "answer_label": "A", + "answer_text": "There was no one else on the wooden staircase.", + "q_id": "1cd5d583-afc5-424b-a56c-55a250e80a30_5", + "correct_answer": "There was no one else on the wooden staircase.", + "original_options": { + "A": "There was no one else on the wooden staircase.", + "B": "There was someone else on the wooden staircase." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "1cd5d583-afc5-424b-a56c-55a250e80a30", + "video_length": "00:42:08", + "qid": "1cd5d583-afc5-424b-a56c-55a250e80a30_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I last saw the wooden staircase, was there anyone else on it?", + "answer": "There were other people on the wooden staircase.", + "question_chinese": "我最后一次看到木楼梯时，木楼梯上是否有其他人？", + "answer_chinese": "木楼梯上有其它人。", + "evidence": [ + { + "description": "Apart from you, there was one person going down the stairs.", + "timestamps": { + "start": "00:37:11", + "end": "00:37:13" + } + } + ], + "question_timestamps": "00:42:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "There were other people on the wooden staircase.", + "B": "There was no one else on the wooden staircase." + }, + "answer_label": "A", + "answer_text": "There were other people on the wooden staircase.", + "q_id": "1cd5d583-afc5-424b-a56c-55a250e80a30_6", + "correct_answer": "There were other people on the wooden staircase.", + "original_options": { + "A": "There were other people on the wooden staircase.", + "B": "There was no one else on the wooden staircase." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "1cd5d583-afc5-424b-a56c-55a250e80a30", + "video_length": "00:42:08", + "qid": "1cd5d583-afc5-424b-a56c-55a250e80a30_1_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The electric drill first appeared on the second step of the wooden stairs connecting the first and second floors. Where was it the last time I saw it?", + "answer": "That chainsaw is on the grass next to the wooden staircase, and it’s now to your front-right.", + "question_chinese": "最开始出现在连接一楼和二楼的木楼梯第二级台阶上的电钻在我最后一次看到它时它在哪里？", + "answer_chinese": "那把电锯在木质楼梯旁边的草地上，现在在你的右前方。", + "evidence": [ + { + "description": "You used that power drill and threw it onto the grass beside the wooden staircase", + "timestamps": { + "start": "00:28:39", + "end": "00:29:53" + } + }, + { + "description": "That chainsaw is still on the grass beside the wooden staircase.", + "timestamps": { + "start": "00:31:20", + "end": "00:31:34" + } + } + ], + "question_timestamps": "00:31:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was to your front-left.", + "B": "It was directly behind you.", + "C": "It was on the grass beside the wooden staircase, but positioned to your back-left when you last saw it.", + "D": "It was on the grass next to the staircase and sat slightly to your right, not out in front.", + "E": "That chainsaw is on the grass next to the wooden staircase, and it’s now to your front-right." + }, + "answer_label": "E", + "answer_text": "That chainsaw is on the grass next to the wooden staircase, and it’s now to your front-right.", + "q_id": "1cd5d583-afc5-424b-a56c-55a250e80a30_1_T1", + "correct_answer": "That chainsaw is on the grass next to the wooden staircase, and it’s now to your front-right.", + "original_options": { + "A": "It was to your front-left.", + "B": "It was directly behind you.", + "C": "It was on the grass beside the wooden staircase, but positioned to your back-left when you last saw it.", + "D": "It was on the grass next to the staircase and sat slightly to your right, not out in front.", + "E": "That chainsaw is on the grass next to the wooden staircase, and it’s now to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "1cd5d583-afc5-424b-a56c-55a250e80a30", + "video_length": "00:42:08", + "qid": "1cd5d583-afc5-424b-a56c-55a250e80a30_1_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The electric drill first appeared on the second step of the wooden stairs connecting the first and second floors. Where was it the last time I saw it?", + "answer": "That chainsaw is on the grass next to the wooden staircase, and it’s now to your right.", + "question_chinese": "最开始出现在连接一楼和二楼的木楼梯第二级台阶上的电钻在我最后一次看到它时它在哪里？", + "answer_chinese": "那把电锯在木质楼梯旁边的草地上，现在在你的右方。", + "evidence": [ + { + "description": "You used that power drill and threw it onto the grass beside the wooden staircase", + "timestamps": { + "start": "00:28:39", + "end": "00:29:53" + } + }, + { + "description": "That chainsaw is still on the grass beside the wooden staircase.", + "timestamps": { + "start": "00:31:20", + "end": "00:31:34" + } + } + ], + "question_timestamps": "00:35:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "That chainsaw is on the grass next to the wooden staircase, and it’s now to your right.", + "B": "It was to your left.", + "C": "It was behind you.", + "D": "It was on the grass beside the wooden staircase, but it was positioned in front of you, not at your side.", + "E": "It was still on the second step of the stairs, directly in front of you the last time you saw it." + }, + "answer_label": "A", + "answer_text": "That chainsaw is on the grass next to the wooden staircase, and it’s now to your right.", + "q_id": "1cd5d583-afc5-424b-a56c-55a250e80a30_1_T2", + "correct_answer": "That chainsaw is on the grass next to the wooden staircase, and it’s now to your right.", + "original_options": { + "A": "That chainsaw is on the grass next to the wooden staircase, and it’s now to your right.", + "B": "It was to your left.", + "C": "It was behind you.", + "D": "It was on the grass beside the wooden staircase, but it was positioned in front of you, not at your side.", + "E": "It was still on the second step of the stairs, directly in front of you the last time you saw it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 21, + 21 + ] + } +] diff --git a/QAs/1def75d5-bd39-4638-8b3d-a4e1c2ceefeb-mcq.json b/QAs/1def75d5-bd39-4638-8b3d-a4e1c2ceefeb-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..0a034a2cc91e98ffd6290119f1393a14bd588fa3 --- /dev/null +++ b/QAs/1def75d5-bd39-4638-8b3d-a4e1c2ceefeb-mcq.json @@ -0,0 +1,1160 @@ +[ + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_6_T2", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I want to add water to the glass bottle—how do I go get the water bottle?", + "answer": "You can turn around and pick up the water bottle on your right hand side at the kitchen entrance.", + "question_chinese": "我想往玻璃瓶中加水，我该怎么去拿水瓶?", + "answer_chinese": "你可以向后转拿取厨房门口你右手边的水瓶。", + "evidence": [ + { + "description": "You added a bottle of water and placed it on the left side of the door.", + "timestamps": { + "start": "00:07:58", + "end": "00:08:09" + } + } + ], + "question_timestamps": "00:08:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Pick it up on your left.", + "B": "Grab it in front of you.", + "C": "You can turn around and pick up the water bottle on your right hand side at the kitchen entrance.", + "D": "Turn around at the kitchen entrance and pick up the water bottle on your left-hand side instead.", + "E": "Walk forward into the kitchen entrance and take the water bottle from directly ahead of you." + }, + "answer_label": "C", + "answer_text": "You can turn around and pick up the water bottle on your right hand side at the kitchen entrance.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_6_T2", + "correct_answer": "You can turn around and pick up the water bottle on your right hand side at the kitchen entrance.", + "original_options": { + "A": "Pick it up on your left.", + "B": "Grab it in front of you.", + "C": "You can turn around and pick up the water bottle on your right hand side at the kitchen entrance.", + "D": "Turn around at the kitchen entrance and pick up the water bottle on your left-hand side instead.", + "E": "Walk forward into the kitchen entrance and take the water bottle from directly ahead of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_6_T1", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I want to add water to the glass bottle—how do I go get the water bottle?", + "answer": "Go through the hallway on your right hand side and enter the door on the left; one bottle was placed by your left foot, and the other bottle was placed by your right foot.", + "question_chinese": "我想往玻璃瓶中加水，我该怎么去拿水瓶?", + "answer_chinese": "从你的右手边经过走廊，进入左边的门，其中一个瓶子放在了左脚边，另一个瓶子放在了右脚边。", + "evidence": [ + { + "description": "You added a bottle of water and placed it on the left side of the door.", + "timestamps": { + "start": "00:07:58", + "end": "00:08:09" + } + }, + { + "description": "You added a bottle of water and placed it near the door.", + "timestamps": { + "start": "00:08:22", + "end": "00:08:26" + } + }, + { + "description": "You came here from the kitchen.", + "timestamps": { + "start": "00:08:27", + "end": "00:08:38" + } + } + ], + "question_timestamps": "00:08:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use the hallway on your left.", + "B": "Enter the door on the right.", + "C": "Go through the hallway on your right and enter the door on the right; both bottles are by your left foot.", + "D": "Go through the hallway on your right hand side and enter the door on the left; one bottle was placed by your left foot, and the other bottle was placed by your right foot.", + "E": "Go through the hallway straight ahead and enter the left door; both bottles are placed by your right foot." + }, + "answer_label": "D", + "answer_text": "Go through the hallway on your right hand side and enter the door on the left; one bottle was placed by your left foot, and the other bottle was placed by your right foot.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_6_T1", + "correct_answer": "Go through the hallway on your right hand side and enter the door on the left; one bottle was placed by your left foot, and the other bottle was placed by your right foot.", + "original_options": { + "A": "Use the hallway on your left.", + "B": "Enter the door on the right.", + "C": "Go through the hallway on your right and enter the door on the right; both bottles are by your left foot.", + "D": "Go through the hallway on your right hand side and enter the door on the left; one bottle was placed by your left foot, and the other bottle was placed by your right foot.", + "E": "Go through the hallway straight ahead and enter the left door; both bottles are placed by your right foot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 19 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_4_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "I want to use the thermometer—how do I get it?", + "answer": "The thermometer is on the stovetop, to your left of the pot.", + "question_chinese": "我想用一下温度计，我该怎么拿到？", + "answer_chinese": "温度计在灶台上锅的左边。", + "evidence": [ + { + "description": "You placed the thermometer to the left of the pot on the stove.", + "timestamps": { + "start": "00:01:31", + "end": "00:01:38" + } + } + ], + "question_timestamps": "00:03:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to the right of the pot.", + "B": "It’s in front of the pot.", + "C": "The thermometer is on the stovetop, positioned behind the pot rather than to its left side.", + "D": "The thermometer is on the stovetop, to your left of the pot.", + "E": "The thermometer is on the stovetop, sitting to the pot’s right side when you reach for it." + }, + "answer_label": "D", + "answer_text": "The thermometer is on the stovetop, to your left of the pot.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_4_T2", + "correct_answer": "The thermometer is on the stovetop, to your left of the pot.", + "original_options": { + "A": "It’s to the right of the pot.", + "B": "It’s in front of the pot.", + "C": "The thermometer is on the stovetop, positioned behind the pot rather than to its left side.", + "D": "The thermometer is on the stovetop, to your left of the pot.", + "E": "The thermometer is on the stovetop, sitting to the pot’s right side when you reach for it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 18 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_4_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "I want to use the thermometer—how do I get it?", + "answer": "The thermometer is on the counter to your left hand side.", + "question_chinese": "我想用一下温度计，我该怎么拿到？", + "answer_chinese": "温度计在你左手边的台子上", + "evidence": [ + { + "description": "The thermometer is on the table on your left.", + "timestamps": { + "start": "00:04:50", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:18:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the counter in front of you.", + "B": "On the counter to your right.", + "C": "The thermometer is on the counter to your left hand side.", + "D": "It’s on the counter directly ahead of you, near the edge where you were working.", + "E": "Look on the counter to your right-hand side, beside where you last set items down." + }, + "answer_label": "C", + "answer_text": "The thermometer is on the counter to your left hand side.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_4_T1", + "correct_answer": "The thermometer is on the counter to your left hand side.", + "original_options": { + "A": "On the counter in front of you.", + "B": "On the counter to your right.", + "C": "The thermometer is on the counter to your left hand side.", + "D": "It’s on the counter directly ahead of you, near the edge where you were working.", + "E": "Look on the counter to your right-hand side, beside where you last set items down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 15 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_3_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I take to move the thermometer from the cabinet to where it is now?", + "answer": "The thermometer was taken out of the cabinet, rinsed in the sink, put into the pot, and then placed to your left of the pot.", + "question_chinese": "我把温度计从柜子里到现在的位置经过了什么路径？", + "answer_chinese": "温度计被从柜子里拿了出来，被放入水池冲洗后被放进了锅里，然后被放在了锅的左边。", + "evidence": [ + { + "description": "The thermometer was taken out of the cabinet, rinsed in the sink, then put into the pot, and finally placed on the left side of the pot.", + "timestamps": { + "start": "00:00:05", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Cabinet to pot to counter.", + "B": "Cabinet, then straight to counter.", + "C": "The thermometer was taken out of the cabinet, rinsed in the sink, put into the pot, and then placed to your left of the pot.", + "D": "You took it from the cabinet, rinsed it in the sink, then placed it to your right of the pot.", + "E": "You removed it from the cabinet, skipped rinsing, put it in the pot, then set it in front of the pot." + }, + "answer_label": "C", + "answer_text": "The thermometer was taken out of the cabinet, rinsed in the sink, put into the pot, and then placed to your left of the pot.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_3_T2", + "correct_answer": "The thermometer was taken out of the cabinet, rinsed in the sink, put into the pot, and then placed to your left of the pot.", + "original_options": { + "A": "Cabinet to pot to counter.", + "B": "Cabinet, then straight to counter.", + "C": "The thermometer was taken out of the cabinet, rinsed in the sink, put into the pot, and then placed to your left of the pot.", + "D": "You took it from the cabinet, rinsed it in the sink, then placed it to your right of the pot.", + "E": "You removed it from the cabinet, skipped rinsing, put it in the pot, then set it in front of the pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 21 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_3_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I take to move the thermometer from the cabinet to where it is now?", + "answer": "The thermometer was taken out of the cabinet, rinsed in the sink, put into the pot, and then placed to your left of the pot. You picked up the thermometer and washed it clean, then turned left and walked straight to place it on this counter.", + "question_chinese": "我把温度计从柜子里到现在的位置经过了什么路径？", + "answer_chinese": "温度计被从柜子里拿了出来，被放入水池冲洗后被放进了锅里，然后被放在了锅的左边。你拿起了温度计并洗干净，左转后直走放在了这个台子上。", + "evidence": [ + { + "description": "The thermometer was taken out of the cabinet, rinsed in the sink, then put into the pot, and finally placed on the left side of the pot.", + "timestamps": { + "start": "00:00:05", + "end": "00:01:34" + } + }, + { + "description": "You picked up the thermometer and washed it clean. Then you turned left and walked straight, placing it on this table.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:50" + } + } + ], + "question_timestamps": "00:04:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Cabinet, sink, pot, then right.", + "B": "Cabinet straight to the counter.", + "C": "The thermometer was taken out of the cabinet, rinsed in the sink, put into the pot, and then placed to your left of the pot. You picked up the thermometer and washed it clean, then turned left and walked straight to place it on this counter.", + "D": "You took it from the cabinet, rinsed it in the sink, and placed it on the counter without using the pot.", + "E": "You removed it, rinsed it, put it in the pot, then walked right and set it on the counter." + }, + "answer_label": "C", + "answer_text": "The thermometer was taken out of the cabinet, rinsed in the sink, put into the pot, and then placed to your left of the pot. You picked up the thermometer and washed it clean, then turned left and walked straight to place it on this counter.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_3_T1", + "correct_answer": "The thermometer was taken out of the cabinet, rinsed in the sink, put into the pot, and then placed to your left of the pot. You picked up the thermometer and washed it clean, then turned left and walked straight to place it on this counter.", + "original_options": { + "A": "Cabinet, sink, pot, then right.", + "B": "Cabinet straight to the counter.", + "C": "The thermometer was taken out of the cabinet, rinsed in the sink, put into the pot, and then placed to your left of the pot. You picked up the thermometer and washed it clean, then turned left and walked straight to place it on this counter.", + "D": "You took it from the cabinet, rinsed it in the sink, and placed it on the counter without using the pot.", + "E": "You removed it, rinsed it, put it in the pot, then walked right and set it on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 19 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_9_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sponges in total did I see in the kitchen?", + "answer": "You saw four sponges in total.", + "question_chinese": "我看到厨房里一共有几块海绵？", + "answer_chinese": "你一共看见了四块海绵.", + "evidence": [ + { + "description": "Here are three pieces of sponge.", + "timestamps": { + "start": "00:05:21", + "end": "00:05:29" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two sponges total.", + "B": "You saw three sponges total.", + "C": "You saw five sponges in total across the kitchen while moving between the counter and sink.", + "D": "You saw four sponges in total.", + "E": "You saw six sponges in total, counting the ones by the sink and the extra ones on the counter." + }, + "answer_label": "D", + "answer_text": "You saw four sponges in total.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_9_T2", + "correct_answer": "You saw four sponges in total.", + "original_options": { + "A": "You saw two sponges total.", + "B": "You saw three sponges total.", + "C": "You saw five sponges in total across the kitchen while moving between the counter and sink.", + "D": "You saw four sponges in total.", + "E": "You saw six sponges in total, counting the ones by the sink and the extra ones on the counter." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 19 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_9_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sponges in total did I see in the kitchen?", + "answer": "You saw four sponges in total.", + "question_chinese": "我看到厨房里一共有几块海绵？", + "answer_chinese": "你一共看见了四块海绵.", + "evidence": [ + { + "description": "Here are three pieces of sponge.", + "timestamps": { + "start": "00:05:21", + "end": "00:05:29" + } + }, + { + "description": "Here is a piece of sponge.", + "timestamps": { + "start": "00:21:23", + "end": "00:21:28" + } + } + ], + "question_timestamps": "00:21:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two sponges total.", + "B": "You saw five sponges total.", + "C": "You saw four sponges in total.", + "D": "You saw three sponges in total while scanning the sink area and nearby counter space.", + "E": "You saw six sponges in total, including the cluster by the sink and the additional ones elsewhere." + }, + "answer_label": "C", + "answer_text": "You saw four sponges in total.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_9_T1", + "correct_answer": "You saw four sponges in total.", + "original_options": { + "A": "You saw two sponges total.", + "B": "You saw five sponges total.", + "C": "You saw four sponges in total.", + "D": "You saw three sponges in total while scanning the sink area and nearby counter space.", + "E": "You saw six sponges in total, including the cluster by the sink and the additional ones elsewhere." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_8_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sponges did I see by the sink?", + "answer": "You can see three sponges by the sink.", + "question_chinese": "我看到水池边有几块海绵？", + "answer_chinese": "你可以看到水池边有三块海绵。", + "evidence": [ + { + "description": "Here you can see two pieces of sponge.", + "timestamps": { + "start": "00:04:36", + "end": "00:04:42" + } + } + ], + "question_timestamps": "00:04:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can see one sponge.", + "B": "You can see two sponges.", + "C": "You can see three sponges by the sink.", + "D": "You can see four sponges by the sink, grouped together along the edge of the basin.", + "E": "You can see five sponges by the sink, with several stacked and others laid out nearby." + }, + "answer_label": "C", + "answer_text": "You can see three sponges by the sink.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_8_T2", + "correct_answer": "You can see three sponges by the sink.", + "original_options": { + "A": "You can see one sponge.", + "B": "You can see two sponges.", + "C": "You can see three sponges by the sink.", + "D": "You can see four sponges by the sink, grouped together along the edge of the basin.", + "E": "You can see five sponges by the sink, with several stacked and others laid out nearby." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_8_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sponges did I see by the sink?", + "answer": "You can see three sponges by the sink.", + "question_chinese": "我看到水池边有几块海绵？", + "answer_chinese": "你可以看到水池边有三块海绵。", + "evidence": [ + { + "description": "Here you can see three pieces of sponge.", + "timestamps": { + "start": "00:05:24", + "end": "00:05:29" + } + } + ], + "question_timestamps": "00:06:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can see one sponge.", + "B": "You can see three sponges by the sink.", + "C": "You can see four sponges.", + "D": "You can see two sponges by the sink, sitting next to each other near the faucet.", + "E": "You can see five sponges by the sink, spread out along the counter beside the basin." + }, + "answer_label": "B", + "answer_text": "You can see three sponges by the sink.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_8_T1", + "correct_answer": "You can see three sponges by the sink.", + "original_options": { + "A": "You can see one sponge.", + "B": "You can see three sponges by the sink.", + "C": "You can see four sponges.", + "D": "You can see two sponges by the sink, sitting next to each other near the faucet.", + "E": "You can see five sponges by the sink, spread out along the counter beside the basin." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_5_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where should I find the washed white towel to use it?", + "answer": "The white towel is in your hand.", + "question_chinese": "我要使用被洗过的白色毛巾应该在哪里找到他？", + "answer_chinese": "白色毛巾在你的手里。", + "evidence": [ + { + "description": "You washed a white towel.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:05" + } + } + ], + "question_timestamps": "00:06:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The white towel is in your hand.", + "B": "On the counter in front.", + "C": "Hanging by the sink.", + "D": "You should find the washed white towel on the counter to your right, near where you rinsed items.", + "E": "Look for the washed white towel on the shelf behind you, close to the area where you set drying items." + }, + "answer_label": "A", + "answer_text": "The white towel is in your hand.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_5_T2", + "correct_answer": "The white towel is in your hand.", + "original_options": { + "A": "The white towel is in your hand.", + "B": "On the counter in front.", + "C": "Hanging by the sink.", + "D": "You should find the washed white towel on the counter to your right, near where you rinsed items.", + "E": "Look for the washed white towel on the shelf behind you, close to the area where you set drying items." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 20 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_5_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where should I find the washed white towel to use it?", + "answer": "Go out through the door behind you; the towel is hanging on the shelf outside.", + "question_chinese": "我要使用被洗过的白色毛巾应该在哪里找到他？", + "answer_chinese": "从你身后的门出去，毛巾被晾在外面的架子上。", + "evidence": [ + { + "description": "Go out through the door behind you. The towels are hung on the rack outside.", + "timestamps": { + "start": "00:06:17", + "end": "00:06:22" + } + }, + { + "description": "You washed a white towel.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:05" + } + } + ], + "question_timestamps": "00:07:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go out the door in front.", + "B": "Go out the door to your right.", + "C": "Go out through the door behind you; the towel is hanging on the shelf outside.", + "D": "Go out through the door to your left; check the shelf outside for the towel.", + "E": "Step outside through the door in front, and look for the towel hanging on the shelf." + }, + "answer_label": "C", + "answer_text": "Go out through the door behind you; the towel is hanging on the shelf outside.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_5_T1", + "correct_answer": "Go out through the door behind you; the towel is hanging on the shelf outside.", + "original_options": { + "A": "Go out the door in front.", + "B": "Go out the door to your right.", + "C": "Go out through the door behind you; the towel is hanging on the shelf outside.", + "D": "Go out through the door to your left; check the shelf outside for the towel.", + "E": "Step outside through the door in front, and look for the towel hanging on the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 16 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_2_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How do I get back to nearby the living-room table?", + "answer": "Walk backward out through the door, then turn right and go straight; the table will be on your right hand side.", + "question_chinese": "我该怎么回到客厅的桌子附近？", + "answer_chinese": "往后走出门然后右转然后直走，桌子在你的右手边.", + "evidence": [ + { + "description": "Go out the door, then turn right, and then go straight. The table is on your right side.", + "timestamps": { + "start": "00:08:28", + "end": "00:08:40" + } + }, + { + "description": "You turn right from the table, pass through the corridor, and then turn left to arrive here now.", + "timestamps": { + "start": "00:08:45", + "end": "00:08:54" + } + } + ], + "question_timestamps": "00:16:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left after the door.", + "B": "Walk forward through the door.", + "C": "Walk backward out through the door, then turn left and go straight; the table will be on your left side.", + "D": "Back out through the doorway, turn right, then stop quickly; the table should end up directly in front of you.", + "E": "Walk backward out through the door, then turn right and go straight; the table will be on your right hand side." + }, + "answer_label": "E", + "answer_text": "Walk backward out through the door, then turn right and go straight; the table will be on your right hand side.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_2_T1", + "correct_answer": "Walk backward out through the door, then turn right and go straight; the table will be on your right hand side.", + "original_options": { + "A": "Turn left after the door.", + "B": "Walk forward through the door.", + "C": "Walk backward out through the door, then turn left and go straight; the table will be on your left side.", + "D": "Back out through the doorway, turn right, then stop quickly; the table should end up directly in front of you.", + "E": "Walk backward out through the door, then turn right and go straight; the table will be on your right hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 20 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How can I get back to the table near the living room?", + "answer": "Go out the door, then turn right, and then go straight. The table is on your right side.", + "question_chinese": "我该怎么回到客厅的桌子附近？", + "answer_chinese": "往后走出门然后右转然后直走，桌子在你的右手边.", + "evidence": [ + { + "description": "Go out the door, then turn right, and then go straight. The table is on your right side.", + "timestamps": { + "start": "00:08:28", + "end": "00:08:40" + } + }, + { + "description": "You turn right from the table, pass through the corridor, and then turn left to arrive here now.", + "timestamps": { + "start": "00:08:45", + "end": "00:08:54" + } + } + ], + "question_timestamps": "00:21:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go straight.", + "B": "Go out the door, then turn right, and then go straight. The table is on your right side.", + "C": "Go straight, then turn right.", + "D": "Go out the door, turn right, then turn left; you should find the table on your left side.", + "E": "Exit through the door, turn left, and continue straight until the table appears on your right side." + }, + "answer_label": "B", + "answer_text": "Go out the door, then turn right, and then go straight. The table is on your right side.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_2", + "correct_answer": "Go out the door, then turn right, and then go straight. The table is on your right side.", + "original_options": { + "A": "Turn left, then go straight.", + "B": "Go out the door, then turn right, and then go straight. The table is on your right side.", + "C": "Go straight, then turn right.", + "D": "Go out the door, turn right, then turn left; you should find the table on your left side.", + "E": "Exit through the door, turn left, and continue straight until the table appears on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_2_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How do I get back to nearby the living-room table?", + "answer": "Turn around and then walk straight; the table will be on your right hand side.", + "question_chinese": "我该怎么回到客厅的桌子附近？", + "answer_chinese": "向后转然后直走，桌子在你的右手边.", + "evidence": [ + { + "description": "Go out the door, then turn right, and then go straight. The table is on your right side.", + "timestamps": { + "start": "00:08:28", + "end": "00:08:40" + } + }, + { + "description": "You turn right from the table, pass through the corridor, and then turn left to arrive here now.", + "timestamps": { + "start": "00:08:45", + "end": "00:08:54" + } + } + ], + "question_timestamps": "00:21:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around and then walk straight; the table will be on your right hand side.", + "B": "Just walk straight ahead.", + "C": "Turn around; it’s on your left.", + "D": "Turn around and then walk straight; the table will be on your left hand side.", + "E": "Turn around, walk straight a bit, and you should end up with the table directly in front of you." + }, + "answer_label": "A", + "answer_text": "Turn around and then walk straight; the table will be on your right hand side.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_2_T2", + "correct_answer": "Turn around and then walk straight; the table will be on your right hand side.", + "original_options": { + "A": "Turn around and then walk straight; the table will be on your right hand side.", + "B": "Just walk straight ahead.", + "C": "Turn around; it’s on your left.", + "D": "Turn around and then walk straight; the table will be on your left hand side.", + "E": "Turn around, walk straight a bit, and you should end up with the table directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 15, + 19 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_7_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many water bottles with blue caps did I see?", + "answer": "You saw a total of three.", + "question_chinese": "有几个蓝色盖子的水瓶被我看见了？", + "answer_chinese": "你看见了一共有三个。", + "evidence": [ + { + "description": "Here are two water bottles placed here.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:12" + } + }, + { + "description": "Here is one water bottle placed here.", + "timestamps": { + "start": "00:08:23", + "end": "00:08:25" + } + } + ], + "question_timestamps": "00:08:29", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw a total of three.", + "B": "You saw a total of one.", + "C": "You saw a total of two.", + "D": "You saw a total of four water bottles with blue caps in the scene.", + "E": "You saw a total of five water bottles with blue caps in all." + }, + "answer_label": "A", + "answer_text": "You saw a total of three.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_7_T1", + "correct_answer": "You saw a total of three.", + "original_options": { + "A": "You saw a total of three.", + "B": "You saw a total of one.", + "C": "You saw a total of two.", + "D": "You saw a total of four water bottles with blue caps in the scene.", + "E": "You saw a total of five water bottles with blue caps in all." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_8", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many water bottles with blue caps did I see?", + "answer": "There are a total of five.", + "question_chinese": "有几个蓝色盖子的水瓶被我看见了？", + "answer_chinese": "一共有五个。", + "evidence": [ + { + "description": "Here are two water bottles placed here.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:12" + } + }, + { + "description": "Here is one water bottle placed here.", + "timestamps": { + "start": "00:08:23", + "end": "00:08:25" + } + }, + { + "description": "Here are two bottles.", + "timestamps": { + "start": "00:08:31", + "end": "00:08:33" + } + } + ], + "question_timestamps": "00:08:59", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are a total of six.", + "B": "There are a total of seven.", + "C": "There are a total of five.", + "D": "There are a total of eight water bottles with blue caps in the scene.", + "E": "There are a total of nine water bottles with blue caps altogether." + }, + "answer_label": "C", + "answer_text": "There are a total of five.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_8", + "correct_answer": "There are a total of five.", + "original_options": { + "A": "There are a total of six.", + "B": "There are a total of seven.", + "C": "There are a total of five.", + "D": "There are a total of eight water bottles with blue caps in the scene.", + "E": "There are a total of nine water bottles with blue caps altogether." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 12 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_7_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many water bottles with blue caps did I see?", + "answer": "You saw a total of five.", + "question_chinese": "有几个蓝色盖子的水瓶被我看见了？", + "answer_chinese": "你看见了一共有五个。", + "evidence": [ + { + "description": "Here are two water bottles placed here.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:12" + } + }, + { + "description": "Here is one water bottle placed here.", + "timestamps": { + "start": "00:08:23", + "end": "00:08:25" + } + }, + { + "description": "There are two water bottles here.", + "timestamps": { + "start": "00:08:32", + "end": "00:08:34" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw a total of three.", + "B": "You saw a total of five.", + "C": "You saw a total of four.", + "D": "You saw a total of six water bottles with blue caps in the scene.", + "E": "You saw a total of seven water bottles with blue caps overall." + }, + "answer_label": "B", + "answer_text": "You saw a total of five.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_7_T2", + "correct_answer": "You saw a total of five.", + "original_options": { + "A": "You saw a total of three.", + "B": "You saw a total of five.", + "C": "You saw a total of four.", + "D": "You saw a total of six water bottles with blue caps in the scene.", + "E": "You saw a total of seven water bottles with blue caps overall." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 12 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_1_T1", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "I’m facing the stove now—where can I grab the thermometer from?", + "answer": "You took the thermometer out of the cabinet to your left.", + "question_chinese": "现在我面对灶台，我从哪里可以拿到温度计？", + "answer_chinese": "你从你左手边的柜子里拿出来了温度计。", + "evidence": [ + { + "description": "The thermometer is in the second cabinet on the left-hand side, from left to right.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took the thermometer out of the cabinet to your left.", + "B": "From the cabinet to your right.", + "C": "From the cabinet in front of you.", + "D": "You took the thermometer out of the cabinet behind you instead of the left one.", + "E": "You grabbed the thermometer from the cabinet above you, not from the one to your left." + }, + "answer_label": "A", + "answer_text": "You took the thermometer out of the cabinet to your left.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_1_T1", + "correct_answer": "You took the thermometer out of the cabinet to your left.", + "original_options": { + "A": "You took the thermometer out of the cabinet to your left.", + "B": "From the cabinet to your right.", + "C": "From the cabinet in front of you.", + "D": "You took the thermometer out of the cabinet behind you instead of the left one.", + "E": "You grabbed the thermometer from the cabinet above you, not from the one to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 16 + ] + }, + { + "video_uid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb", + "video_length": "00:22:32", + "qid": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_1_T2", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "I’m facing the stove now—where can I grab the thermometer from?", + "answer": "You took the thermometer out of the cabinet to your left.", + "question_chinese": "现在我面对灶台，我从哪里可以拿到温度计？", + "answer_chinese": "你从你左手边的柜子里拿出来了温度计。", + "evidence": [ + { + "description": "The thermometer is in the second cabinet on the left-hand side, from left to right.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:29" + } + }, + { + "description": "You placed the thermometer to the left of the pot on the stove.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:44" + } + } + ], + "question_timestamps": "00:04:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the cabinet to your right.", + "B": "You took the thermometer out of the cabinet to your left.", + "C": "From the cabinet in front of you.", + "D": "You grabbed the thermometer from the cabinet behind you, then brought it over to the stove.", + "E": "You took the thermometer out of the cabinet above you, not the one beside the stove." + }, + "answer_label": "B", + "answer_text": "You took the thermometer out of the cabinet to your left.", + "q_id": "1def75d5-bd39-4638-8b3d-a4e1c2ceefeb_1_T2", + "correct_answer": "You took the thermometer out of the cabinet to your left.", + "original_options": { + "A": "From the cabinet to your right.", + "B": "You took the thermometer out of the cabinet to your left.", + "C": "From the cabinet in front of you.", + "D": "You grabbed the thermometer from the cabinet behind you, then brought it over to the stove.", + "E": "You took the thermometer out of the cabinet above you, not the one beside the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 16 + ] + } +] diff --git a/QAs/1f216b72-21a2-4463-b5ae-13bffd779b4d-mcq.json b/QAs/1f216b72-21a2-4463-b5ae-13bffd779b4d-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..62284eb34e522290b3b99966c4ba59c32283f6e3 --- /dev/null +++ b/QAs/1f216b72-21a2-4463-b5ae-13bffd779b4d-mcq.json @@ -0,0 +1,1353 @@ +[ + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I stood in the repair room, how many drawers were on the floor?", + "answer": "When you stand in the repair room, there are five drawers on the floor.", + "question_chinese": "当我站在修理间，地上有几个抽屉？", + "answer_chinese": "当你站在修理间，地上有五个抽屉。", + "evidence": [ + { + "description": "There are five drawers on the left side of the locker.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:21" + } + } + ], + "question_timestamps": "00:04:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "When you stand in the repair room, there are five drawers on the floor.", + "B": "There are four drawers on the floor.", + "C": "There are six drawers on the floor.", + "D": "When you stand in the repair room, you can count seven drawers scattered across the floor.", + "E": "When you stand in the repair room, there are three drawers on the floor near the work area." + }, + "answer_label": "A", + "answer_text": "When you stand in the repair room, there are five drawers on the floor.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_1", + "correct_answer": "When you stand in the repair room, there are five drawers on the floor.", + "original_options": { + "A": "When you stand in the repair room, there are five drawers on the floor.", + "B": "There are four drawers on the floor.", + "C": "There are six drawers on the floor.", + "D": "When you stand in the repair room, you can count seven drawers scattered across the floor.", + "E": "When you stand in the repair room, there are three drawers on the floor near the work area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I stood in the repair room, how many drawers were on the floor?", + "answer": "When you stand in the repair room, there are six drawers on the floor.", + "question_chinese": "当我站在修理间，地上有几个抽屉？", + "answer_chinese": "当你站在修理间，地上有六个抽屉。", + "evidence": [ + { + "description": "There are five drawers on the left side of the locker.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:21" + } + }, + { + "description": "There is a drawer to the right of the locker.", + "timestamps": { + "start": "00:27:20", + "end": "00:27:37" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are five drawers on the floor.", + "B": "There are seven drawers on the floor.", + "C": "When you stand in the repair room, there are six drawers on the floor.", + "D": "When you stand in the repair room, there are four drawers lying on the floor in view.", + "E": "When you stand in the repair room, you see eight drawers on the floor spread around the room." + }, + "answer_label": "C", + "answer_text": "When you stand in the repair room, there are six drawers on the floor.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_2", + "correct_answer": "When you stand in the repair room, there are six drawers on the floor.", + "original_options": { + "A": "There are five drawers on the floor.", + "B": "There are seven drawers on the floor.", + "C": "When you stand in the repair room, there are six drawers on the floor.", + "D": "When you stand in the repair room, there are four drawers lying on the floor in view.", + "E": "When you stand in the repair room, you see eight drawers on the floor spread around the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_8_T2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was standing next to the storage rack, where was the car relative to me?", + "answer": "When you were standing next to the storage rack, the car was to your right.", + "question_chinese": "当我站在置物架旁边，汽车相对你在什么位置？", + "answer_chinese": "当你站在置物架旁边，汽车在你的右边。", + "evidence": [ + { + "description": "The car is on your right.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were standing next to the storage rack, the car was to your right.", + "B": "When you were standing next to the storage rack, the car was to your left." + }, + "answer_label": "A", + "answer_text": "When you were standing next to the storage rack, the car was to your right.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_8_T2", + "correct_answer": "When you were standing next to the storage rack, the car was to your right.", + "original_options": { + "A": "When you were standing next to the storage rack, the car was to your right.", + "B": "When you were standing next to the storage rack, the car was to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_8_T1", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was standing next to the storage rack, where was the car relative to me?", + "answer": "When you were standing next to the storage rack, the car was in front of you.", + "question_chinese": "当我站在置物架旁边，汽车相对你在什么位置？", + "answer_chinese": "当你站在置物架旁边，汽车在你的前边。", + "evidence": [ + { + "description": "The car is on front of you.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:50" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were standing next to the storage rack, the car was behind you.", + "B": "When you were standing next to the storage rack, the car was in front of you." + }, + "answer_label": "B", + "answer_text": "When you were standing next to the storage rack, the car was in front of you.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_8_T1", + "correct_answer": "When you were standing next to the storage rack, the car was in front of you.", + "original_options": { + "A": "When you were standing next to the storage rack, the car was behind you.", + "B": "When you were standing next to the storage rack, the car was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_16", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand next to the shelf, which is closer to me, the lawn mower or the car?", + "answer": "When you stand next to the shelf, the lawn mower is closer to you.", + "question_chinese": "当我站在置物架旁边，除草机和汽车哪个离我更近？", + "answer_chinese": "当你站在置物架旁边，除草机离你更近。", + "evidence": [ + { + "description": "The lawn mower is three meters away from the storage shelf in a straight line.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:46" + } + }, + { + "description": "The car is seven meters away from the shelf in a straight line.", + "timestamps": { + "start": "00:01:45", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:06:16", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "When you stand next to the shelf, the lawn mower is closer to you.", + "B": "When you stand next to the shelf, the car is closer to you." + }, + "answer_label": "A", + "answer_text": "When you stand next to the shelf, the lawn mower is closer to you.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_16", + "correct_answer": "When you stand next to the shelf, the lawn mower is closer to you.", + "original_options": { + "A": "When you stand next to the shelf, the lawn mower is closer to you.", + "B": "When you stand next to the shelf, the car is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_15", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand next to the shelf, which is closer to me, the lawn mower or the car?", + "answer": "When you stand next to the shelf, the lawn mower is closer to you.", + "question_chinese": "当我站在置物架旁边，除草机和汽车哪个离我更近？", + "answer_chinese": "当你站在置物架旁边，除草机离你更近。", + "evidence": [ + { + "description": "The lawn mower is three meters away from the storage shelf in a straight line.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:46" + } + }, + { + "description": "The car is seven meters away from the shelf in a straight line.", + "timestamps": { + "start": "00:01:45", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:20:11", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "When you stand next to the shelf, the lawn mower is closer to you.", + "B": "When you stand next to the shelf, the car is closer to you." + }, + "answer_label": "A", + "answer_text": "When you stand next to the shelf, the lawn mower is closer to you.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_15", + "correct_answer": "When you stand next to the shelf, the lawn mower is closer to you.", + "original_options": { + "A": "When you stand next to the shelf, the lawn mower is closer to you.", + "B": "When you stand next to the shelf, the car is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_13", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I'm facing the car, how do I get to the restroom?", + "answer": "When facing the car, you should turn around and walk straight to enter the first door to wash your hands.", + "question_chinese": "当我面对汽车时，要去洗手应该怎么走？", + "answer_chinese": "当你面对汽车时，要去洗手应该向后转直走进去第一道门。", + "evidence": [ + { + "description": "The sink is in the compartment after opening the first door.", + "timestamps": { + "start": "00:02:52", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:04:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left and go through the last door.", + "B": "Turn right and enter the first door.", + "C": "When facing the car, you should turn around and walk straight to enter the first door to wash your hands.", + "D": "When facing the car, walk straight ahead and use the second door you pass to reach the restroom.", + "E": "When facing the car, turn around, take a right, then enter the first door to find the restroom." + }, + "answer_label": "C", + "answer_text": "When facing the car, you should turn around and walk straight to enter the first door to wash your hands.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_13", + "correct_answer": "When facing the car, you should turn around and walk straight to enter the first door to wash your hands.", + "original_options": { + "A": "Turn left and go through the last door.", + "B": "Turn right and enter the first door.", + "C": "When facing the car, you should turn around and walk straight to enter the first door to wash your hands.", + "D": "When facing the car, walk straight ahead and use the second door you pass to reach the restroom.", + "E": "When facing the car, turn around, take a right, then enter the first door to find the restroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 18 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_14", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I'm facing the car, how do I get to the restroom?", + "answer": "When facing the car, you should turn around and walk straight to enter the first door to wash your hands.", + "question_chinese": "当我面对汽车时，要去洗手应该怎么走？", + "answer_chinese": "当你面对汽车时，要去洗手应该向后转直走进去第一道门。", + "evidence": [ + { + "description": "The sink is in the compartment after opening the first door.", + "timestamps": { + "start": "00:02:52", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:15:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left and enter the first door.", + "B": "Turn right and enter the second door.", + "C": "When facing the car, turn around, walk straight, and go through the second door to reach it.", + "D": "When facing the car, you should turn around and walk straight to enter the first door to wash your hands.", + "E": "When facing the car, turn right, walk straight past the first door, and enter the next one." + }, + "answer_label": "D", + "answer_text": "When facing the car, you should turn around and walk straight to enter the first door to wash your hands.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_14", + "correct_answer": "When facing the car, you should turn around and walk straight to enter the first door to wash your hands.", + "original_options": { + "A": "Turn left and enter the first door.", + "B": "Turn right and enter the second door.", + "C": "When facing the car, turn around, walk straight, and go through the second door to reach it.", + "D": "When facing the car, you should turn around and walk straight to enter the first door to wash your hands.", + "E": "When facing the car, turn right, walk straight past the first door, and enter the next one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_3", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I face the shelf, is the car in front/left/right of me?", + "answer": "When you face the shelf, the car is on your left.", + "question_chinese": "当我面对置物架的时候，汽车在我的前/左/右？", + "answer_chinese": "当你面对置物架的时候，汽车在你的左边。", + "evidence": [ + { + "description": "The car is on my left.", + "timestamps": { + "start": "00:00:55", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:01:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "When you face the shelf, the car is on your left.", + "B": "The car is in front of you.", + "C": "The car is on your right.", + "D": "When you face the shelf, the car is directly behind you rather than to either side.", + "E": "When you face the shelf, the car sits ahead of you near the shelf, not off to the left." + }, + "answer_label": "A", + "answer_text": "When you face the shelf, the car is on your left.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_3", + "correct_answer": "When you face the shelf, the car is on your left.", + "original_options": { + "A": "When you face the shelf, the car is on your left.", + "B": "The car is in front of you.", + "C": "The car is on your right.", + "D": "When you face the shelf, the car is directly behind you rather than to either side.", + "E": "When you face the shelf, the car sits ahead of you near the shelf, not off to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 19 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_4", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I face the shelf, is the car in front/left/right of me?", + "answer": "When you face the shelf, the car is behind you.", + "question_chinese": "当我面对置物架的时候，汽车在我的前/左/右？", + "answer_chinese": "当你面对置物架的时候，汽车在你的后边。", + "evidence": [ + { + "description": "The car is on my left.", + "timestamps": { + "start": "00:00:55", + "end": "00:01:04" + } + }, + { + "description": "I walked around to the other side of the shelf.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:28" + } + }, + { + "description": "I stood on the side of the shelf .", + "timestamps": { + "start": "00:05:25", + "end": "00:05:36" + } + } + ], + "question_timestamps": "00:07:36", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The car is in front of you.", + "B": "The car is on your left.", + "C": "When you face the shelf, the car is behind you.", + "D": "When you face the shelf, the car is to your right side, not behind you.", + "E": "When you face the shelf, the car is positioned ahead of you near the shelf rather than behind." + }, + "answer_label": "C", + "answer_text": "When you face the shelf, the car is behind you.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_4", + "correct_answer": "When you face the shelf, the car is behind you.", + "original_options": { + "A": "The car is in front of you.", + "B": "The car is on your left.", + "C": "When you face the shelf, the car is behind you.", + "D": "When you face the shelf, the car is to your right side, not behind you.", + "E": "When you face the shelf, the car is positioned ahead of you near the shelf rather than behind." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 18 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_6", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors did I pass through from the repair room to the living room?", + "answer": "You passed through two doors from the repair room to the living room.", + "question_chinese": "我从修理间到客厅经过了几扇门？", + "answer_chinese": "你从修理间到客厅经过了两扇门。", + "evidence": [ + { + "description": "I passed through the first door.", + "timestamps": { + "start": "00:02:50", + "end": "00:03:06" + } + }, + { + "description": "I passed through the second door.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:04:28", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You passed through one door.", + "B": "You passed through two doors from the repair room to the living room.", + "C": "You passed through three doors.", + "D": "You passed through four doors from the repair room to the living room.", + "E": "You passed through five doors on the way from the repair room into the living room." + }, + "answer_label": "B", + "answer_text": "You passed through two doors from the repair room to the living room.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_6", + "correct_answer": "You passed through two doors from the repair room to the living room.", + "original_options": { + "A": "You passed through one door.", + "B": "You passed through two doors from the repair room to the living room.", + "C": "You passed through three doors.", + "D": "You passed through four doors from the repair room to the living room.", + "E": "You passed through five doors on the way from the repair room into the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 16 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_5", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors did I pass through from the repair room to the living room?", + "answer": "You passed through two doors from the repair room to the living room.", + "question_chinese": "我从修理间到客厅经过了几扇门？", + "answer_chinese": "你从修理间到客厅经过了两扇门。", + "evidence": [ + { + "description": "I passed through the second door.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:10" + } + }, + { + "description": "I passed through the first door.", + "timestamps": { + "start": "00:02:50", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:17:01", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You passed through one door.", + "B": "You passed through three doors.", + "C": "You passed through four doors from the repair room to the living room.", + "D": "You passed through five doors on the way from the repair room into the living room.", + "E": "You passed through two doors from the repair room to the living room." + }, + "answer_label": "E", + "answer_text": "You passed through two doors from the repair room to the living room.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_5", + "correct_answer": "You passed through two doors from the repair room to the living room.", + "original_options": { + "A": "You passed through one door.", + "B": "You passed through three doors.", + "C": "You passed through four doors from the repair room to the living room.", + "D": "You passed through five doors on the way from the repair room into the living room.", + "E": "You passed through two doors from the repair room to the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 16 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_9", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the shelf now?", + "answer": "You can't reach the shelf now.", + "question_chinese": "我现在能摸到置物架吗？", + "answer_chinese": "你现在不能摸到置物架。", + "evidence": [ + { + "description": "I am in the room now.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:07" + } + } + ], + "question_timestamps": "00:03:43", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't reach the shelf now.", + "B": "You can reach the shelf now." + }, + "answer_label": "A", + "answer_text": "You can't reach the shelf now.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_9", + "correct_answer": "You can't reach the shelf now.", + "original_options": { + "A": "You can't reach the shelf now.", + "B": "You can reach the shelf now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_10", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the shelf now?", + "answer": "You can reach the shelf now.", + "question_chinese": "我现在能摸到置物架吗？", + "answer_chinese": "你现在能摸到置物架。", + "evidence": [ + { + "description": "I'm standing next to the shelf.", + "timestamps": { + "start": "00:10:25", + "end": "00:11:00" + } + } + ], + "question_timestamps": "00:22:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can reach the shelf now.", + "B": "You can't reach the shelf now." + }, + "answer_label": "A", + "answer_text": "You can reach the shelf now.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_10", + "correct_answer": "You can reach the shelf now.", + "original_options": { + "A": "You can reach the shelf now.", + "B": "You can't reach the shelf now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_6_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was standing in the living room, which direction was the fan relative to me?", + "answer": "When you were standing in the living room, the fan was in front of you.", + "question_chinese": "我站在客厅时，风扇在我的哪个方向？", + "answer_chinese": "你站在客厅时，风扇在你的前面。", + "evidence": [ + { + "description": "The fan in front of you.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:03:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The fan was behind you.", + "B": "The fan was to your left.", + "C": "When you were standing in the living room, the fan was to your right rather than in front of you.", + "D": "When you were standing in the living room, the fan was in front of you.", + "E": "When you were standing in the living room, the fan was off to your left, not positioned ahead of you." + }, + "answer_label": "D", + "answer_text": "When you were standing in the living room, the fan was in front of you.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_6_T1", + "correct_answer": "When you were standing in the living room, the fan was in front of you.", + "original_options": { + "A": "The fan was behind you.", + "B": "The fan was to your left.", + "C": "When you were standing in the living room, the fan was to your right rather than in front of you.", + "D": "When you were standing in the living room, the fan was in front of you.", + "E": "When you were standing in the living room, the fan was off to your left, not positioned ahead of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 20 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_6_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was standing in the living room, which direction was the fan relative to me?", + "answer": "When you were standing in the living room, the fan was to your right.", + "question_chinese": "我站在客厅时，风扇在我的哪个方向？", + "answer_chinese": "你站在客厅时，风扇在你的右边。", + "evidence": [ + { + "description": "The fan is on your right.", + "timestamps": { + "start": "00:03:30", + "end": "00:03:33" + } + } + ], + "question_timestamps": "00:03:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The fan was in front of you.", + "B": "When you were standing in the living room, the fan was to your right.", + "C": "The fan was to your left.", + "D": "When you were standing in the living room, the fan was behind you, not to either side.", + "E": "When you were standing in the living room, the fan was above you, positioned overhead rather than beside you." + }, + "answer_label": "B", + "answer_text": "When you were standing in the living room, the fan was to your right.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_6_T2", + "correct_answer": "When you were standing in the living room, the fan was to your right.", + "original_options": { + "A": "The fan was in front of you.", + "B": "When you were standing in the living room, the fan was to your right.", + "C": "The fan was to your left.", + "D": "When you were standing in the living room, the fan was behind you, not to either side.", + "E": "When you were standing in the living room, the fan was above you, positioned overhead rather than beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 19 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_3_T1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I make direct contact with the table?", + "answer": "You can’t directly reach the table; the table is to your front-left.", + "question_chinese": "我能直接接触到桌子吗？", + "answer_chinese": "你不能直接接触到桌子，桌子在你左前方。", + "evidence": [ + { + "description": "you can not touch the table directly.", + "timestamps": { + "start": "00:03:13", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:03:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can directly reach the table.", + "B": "The table is to your left.", + "C": "You can’t directly reach the table; the table is to your front-right, just out of reach.", + "D": "You can directly reach the table; it sits slightly to your left instead of front-left.", + "E": "You can’t directly reach the table; the table is to your front-left." + }, + "answer_label": "E", + "answer_text": "You can’t directly reach the table; the table is to your front-left.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_3_T1", + "correct_answer": "You can’t directly reach the table; the table is to your front-left.", + "original_options": { + "A": "You can directly reach the table.", + "B": "The table is to your left.", + "C": "You can’t directly reach the table; the table is to your front-right, just out of reach.", + "D": "You can directly reach the table; it sits slightly to your left instead of front-left.", + "E": "You can’t directly reach the table; the table is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 15 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_3_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I make direct contact with the table?", + "answer": "You can directly reach the table; the table is to your left.", + "question_chinese": "我能直接接触到桌子吗？", + "answer_chinese": "你能直接接触到桌子，桌子在你左边。", + "evidence": [ + { + "description": "you can touch the table directly/", + "timestamps": { + "start": "00:03:18", + "end": "00:03:23" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can directly reach the table; the table is to your left.", + "B": "You can’t reach the table.", + "C": "The table is behind you.", + "D": "You can directly reach the table; the table is to your front-left rather than directly to your left.", + "E": "You can’t directly reach the table; the table is to your left but positioned too far away." + }, + "answer_label": "A", + "answer_text": "You can directly reach the table; the table is to your left.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_3_T2", + "correct_answer": "You can directly reach the table; the table is to your left.", + "original_options": { + "A": "You can directly reach the table; the table is to your left.", + "B": "You can’t reach the table.", + "C": "The table is behind you.", + "D": "You can directly reach the table; the table is to your front-left rather than directly to your left.", + "E": "You can’t directly reach the table; the table is to your left but positioned too far away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 18 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_9_T1", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the car relative to me?", + "answer": "At this moment, the car is in front of you.", + "question_chinese": "此时，汽车在我哪个位置？", + "answer_chinese": "此时，汽车在你前方。", + "evidence": [ + { + "description": "The car is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:3" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "At this moment, the car is in front of you.", + "B": "At this moment, the car is behind you." + }, + "answer_label": "A", + "answer_text": "At this moment, the car is in front of you.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_9_T1", + "correct_answer": "At this moment, the car is in front of you.", + "original_options": { + "A": "At this moment, the car is in front of you.", + "B": "At this moment, the car is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_9_T2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the car relative to me?", + "answer": "At this moment, the car is behind you.", + "question_chinese": "此时，汽车在我哪个位置？", + "answer_chinese": "此时，汽车在你后面。", + "evidence": [ + { + "description": "The car is behind of you.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "At this moment, the car is in front of you.", + "B": "At this moment, the car is behind you." + }, + "answer_label": "B", + "answer_text": "At this moment, the car is behind you.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_9_T2", + "correct_answer": "At this moment, the car is behind you.", + "original_options": { + "A": "At this moment, the car is in front of you.", + "B": "At this moment, the car is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_7_T1", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which direction is the car relative to me?", + "answer": "At this moment, the car is in front of you.", + "question_chinese": "此时汽车在我的哪个方向？", + "answer_chinese": "此时汽车在你的前面。", + "evidence": [ + { + "description": "The car is on front of you.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:54" + } + } + ], + "question_timestamps": "00:02:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "At this moment, the car is in front of you.", + "B": "The car is behind you.", + "C": "The car is to your left.", + "D": "At this moment, the car is to your right, positioned beside you rather than ahead.", + "E": "At this moment, the car is above you, not located in your forward line of sight." + }, + "answer_label": "A", + "answer_text": "At this moment, the car is in front of you.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_7_T1", + "correct_answer": "At this moment, the car is in front of you.", + "original_options": { + "A": "At this moment, the car is in front of you.", + "B": "The car is behind you.", + "C": "The car is to your left.", + "D": "At this moment, the car is to your right, positioned beside you rather than ahead.", + "E": "At this moment, the car is above you, not located in your forward line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_7_T2", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which direction is the car relative to me?", + "answer": "At this moment, the car is behind you.", + "question_chinese": "此时汽车在我的哪个方向？", + "answer_chinese": "此时汽车在你的后面。", + "evidence": [ + { + "description": "The car is on behind of you.", + "timestamps": { + "start": "00:02:53", + "end": "00:02:55" + } + } + ], + "question_timestamps": "00:02:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The car is in front of you.", + "B": "The car is to your right.", + "C": "At this moment, the car is to your left, positioned beside you rather than behind.", + "D": "At this moment, the car is above you, not located behind your current position.", + "E": "At this moment, the car is behind you." + }, + "answer_label": "E", + "answer_text": "At this moment, the car is behind you.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_7_T2", + "correct_answer": "At this moment, the car is behind you.", + "original_options": { + "A": "The car is in front of you.", + "B": "The car is to your right.", + "C": "At this moment, the car is to your left, positioned beside you rather than behind.", + "D": "At this moment, the car is above you, not located behind your current position.", + "E": "At this moment, the car is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 14 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_7", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the straight-line distance between me and the shelf now?", + "answer": "The shelf is now seven meters away from you in a straight line.", + "question_chinese": "现在置物架离我直线距离是多少？", + "answer_chinese": "现在置物架离你直线距离是七米。", + "evidence": [ + { + "description": "I stood at the door.", + "timestamps": { + "start": "00:02:33", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:02:53", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "The shelf is five meters away.", + "B": "The shelf is nine meters away.", + "C": "The shelf is now four meters away from you in a straight line, closer than you are now.", + "D": "The shelf is now seven meters away from you in a straight line.", + "E": "The shelf is now eight meters away from you in a straight line, slightly farther than seven." + }, + "answer_label": "D", + "answer_text": "The shelf is now seven meters away from you in a straight line.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_7", + "correct_answer": "The shelf is now seven meters away from you in a straight line.", + "original_options": { + "A": "The shelf is five meters away.", + "B": "The shelf is nine meters away.", + "C": "The shelf is now four meters away from you in a straight line, closer than you are now.", + "D": "The shelf is now seven meters away from you in a straight line.", + "E": "The shelf is now eight meters away from you in a straight line, slightly farther than seven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_8", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the straight-line distance between me and the shelf now?", + "answer": "The shelf is now one meter away from you in a straight line.", + "question_chinese": "现在置物架离我直线距离是多少？", + "answer_chinese": "现在置物架离你直线距离是一米。", + "evidence": [ + { + "description": "I stood next to the shelf painting the chair.", + "timestamps": { + "start": "00:26:14", + "end": "00:26:23" + } + } + ], + "question_timestamps": "00:28:21", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Half a meter away.", + "B": "Two meters away.", + "C": "The shelf is now three meters away from you in a straight line.", + "D": "The shelf is now five meters away from you in a straight line.", + "E": "The shelf is now one meter away from you in a straight line." + }, + "answer_label": "E", + "answer_text": "The shelf is now one meter away from you in a straight line.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_8", + "correct_answer": "The shelf is now one meter away from you in a straight line.", + "original_options": { + "A": "Half a meter away.", + "B": "Two meters away.", + "C": "The shelf is now three meters away from you in a straight line.", + "D": "The shelf is now five meters away from you in a straight line.", + "E": "The shelf is now one meter away from you in a straight line." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 13, + 13 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_17", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When entering the living room, do you see the sofa or the computer desk first?", + "answer": "Enter the living room and first see the sofa.", + "question_chinese": "进入客厅，先看到沙发还是先看到电脑桌？", + "answer_chinese": "进入客厅，先看到沙发。", + "evidence": [ + { + "description": "The sofa is at the entrance of the living room.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:11" + } + }, + { + "description": "The computer desk is next to the balcony door.", + "timestamps": { + "start": "00:03:11", + "end": "00:03:18" + } + } + ], + "question_timestamps": "00:07:16", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Enter the living room and first see the computer desk.", + "B": "Enter the living room and first see the sofa." + }, + "answer_label": "B", + "answer_text": "Enter the living room and first see the sofa.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_17", + "correct_answer": "Enter the living room and first see the sofa.", + "original_options": { + "A": "Enter the living room and first see the computer desk.", + "B": "Enter the living room and first see the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "1f216b72-21a2-4463-b5ae-13bffd779b4d", + "video_length": "00:30:00", + "qid": "1f216b72-21a2-4463-b5ae-13bffd779b4d_18", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When entering the living room, do you see the sofa or the computer desk first?", + "answer": "Enter the living room and first see the sofa.", + "question_chinese": "进入客厅，先看到沙发还是先看到电脑桌？", + "answer_chinese": "进入客厅，先看到沙发。", + "evidence": [ + { + "description": "The sofa is at the entrance of the living room.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:11" + } + }, + { + "description": "The computer desk is next to the balcony door.", + "timestamps": { + "start": "00:03:11", + "end": "00:03:18" + } + } + ], + "question_timestamps": "00:28:50", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Enter the living room and first see the computer desk.", + "B": "Enter the living room and first see the sofa." + }, + "answer_label": "B", + "answer_text": "Enter the living room and first see the sofa.", + "q_id": "1f216b72-21a2-4463-b5ae-13bffd779b4d_18", + "correct_answer": "Enter the living room and first see the sofa.", + "original_options": { + "A": "Enter the living room and first see the computer desk.", + "B": "Enter the living room and first see the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + } +] diff --git a/QAs/20bd786f-d222-4b0e-ad2c-a2381929cf78-mcq.json b/QAs/20bd786f-d222-4b0e-ad2c-a2381929cf78-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a4b60cddd56cbb9610282502490fe84b54fd2670 --- /dev/null +++ b/QAs/20bd786f-d222-4b0e-ad2c-a2381929cf78-mcq.json @@ -0,0 +1,1089 @@ +[ + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_20", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I sit on the stool, which is closer to me, scissors or colored pens?", + "answer": "When you sit on the stool, the scissors are closer to you.", + "question_chinese": "当我坐在凳子上，剪刀和水彩笔哪个离我更近？", + "answer_chinese": "当你坐在凳子上，剪刀离你更近。", + "evidence": [ + { + "description": "The scissors are five centimeters away from me, and the markers are twenty centimeters away from me.", + "timestamps": { + "start": "00:08:19", + "end": "00:08:24" + } + } + ], + "question_timestamps": "00:13:05", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you sit on the stool, the scissors are closer to you.", + "B": "When you sit on the stool, the colored pens are closer to you." + }, + "answer_label": "A", + "answer_text": "When you sit on the stool, the scissors are closer to you.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_20", + "correct_answer": "When you sit on the stool, the scissors are closer to you.", + "original_options": { + "A": "When you sit on the stool, the scissors are closer to you.", + "B": "When you sit on the stool, the colored pens are closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_19", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I sit on the stool, which is closer to me, scissors or colored pens?", + "answer": "When you sit on the stool, the scissors are closer to you.", + "question_chinese": "当我坐在凳子上，剪刀和水彩笔哪个离我更近？", + "answer_chinese": "当你坐在凳子上，剪刀离你更近。", + "evidence": [ + { + "description": "The scissors are five centimeters away from me, and the markers are twenty centimeters away from me.", + "timestamps": { + "start": "00:08:19", + "end": "00:08:24" + } + } + ], + "question_timestamps": "00:23:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you sit on the stool, the scissors are closer to you.", + "B": "When you sit on the stool, the colored pens are closer to you." + }, + "answer_label": "A", + "answer_text": "When you sit on the stool, the scissors are closer to you.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_19", + "correct_answer": "When you sit on the stool, the scissors are closer to you.", + "original_options": { + "A": "When you sit on the stool, the scissors are closer to you.", + "B": "When you sit on the stool, the colored pens are closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_6_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was sitting on the stool, how many times did I use the scissors?", + "answer": "You used it once. The scissors were by your right hand.", + "question_chinese": "当我坐在凳子上，我用了几次剪刀？", + "answer_chinese": "你使用了一次。剪刀在你的右手边。", + "evidence": [ + { + "description": "You used the scissors once.", + "timestamps": { + "start": "00:12:43", + "end": "00:13:08" + } + } + ], + "question_timestamps": "00:14:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You didn’t use the scissors.", + "B": "You used the scissors twice.", + "C": "You used it three times, and the scissors were by your right hand while you were sitting.", + "D": "You used it once, but the scissors were by your left hand when you were sitting on the stool.", + "E": "You used it once. The scissors were by your right hand." + }, + "answer_label": "E", + "answer_text": "You used it once. The scissors were by your right hand.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_6_T1", + "correct_answer": "You used it once. The scissors were by your right hand.", + "original_options": { + "A": "You didn’t use the scissors.", + "B": "You used the scissors twice.", + "C": "You used it three times, and the scissors were by your right hand while you were sitting.", + "D": "You used it once, but the scissors were by your left hand when you were sitting on the stool.", + "E": "You used it once. The scissors were by your right hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 19 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_6_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was sitting on the stool, how many times did I use the scissors?", + "answer": "You used them twice—one was placed by your left hand, and one was by your right hand.", + "question_chinese": "当我坐在凳子上，我用了几次剪刀？", + "answer_chinese": "你使用了两次。一把放在了你的左手边，一把在右手边。", + "evidence": [ + { + "description": "You used the scissors once.", + "timestamps": { + "start": "00:12:43", + "end": "00:13:08" + } + }, + { + "description": "You used the scissors once agine.", + "timestamps": { + "start": "00:22:10", + "end": "00:22:18" + } + } + ], + "question_timestamps": "00:26:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You used them twice—one was placed by your left hand, and one was by your right hand.", + "B": "You used them once.", + "C": "You used them three times.", + "D": "You used them twice, but both were placed by your left hand and none was by your right hand.", + "E": "You used them twice, but none was placed by your left hand and both were by your right hand." + }, + "answer_label": "A", + "answer_text": "You used them twice—one was placed by your left hand, and one was by your right hand.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_6_T2", + "correct_answer": "You used them twice—one was placed by your left hand, and one was by your right hand.", + "original_options": { + "A": "You used them twice—one was placed by your left hand, and one was by your right hand.", + "B": "You used them once.", + "C": "You used them three times.", + "D": "You used them twice, but both were placed by your left hand and none was by your right hand.", + "E": "You used them twice, but none was placed by your left hand and both were by your right hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 19, + 19 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_13", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "While I was sitting on the stool, the colored pencils were in front of/left to me？", + "answer": "When you sit on the stool, the watercolor pens are in front of you.", + "question_chinese": "当我坐在凳子上，水彩笔在我的前/左？", + "answer_chinese": "当你坐在凳子上，水彩笔在你的前面。", + "evidence": [ + { + "description": "The watercolor pens are placed in front of me.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:04:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They are to your left.", + "B": "They are behind you.", + "C": "When you sit on the stool, the watercolor pens are in front of you.", + "D": "When you sit on the stool, the watercolor pens are to your left rather than directly in front of you.", + "E": "When you sit on the stool, the watercolor pens are behind you, not in front of you as you face forward." + }, + "answer_label": "C", + "answer_text": "When you sit on the stool, the watercolor pens are in front of you.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_13", + "correct_answer": "When you sit on the stool, the watercolor pens are in front of you.", + "original_options": { + "A": "They are to your left.", + "B": "They are behind you.", + "C": "When you sit on the stool, the watercolor pens are in front of you.", + "D": "When you sit on the stool, the watercolor pens are to your left rather than directly in front of you.", + "E": "When you sit on the stool, the watercolor pens are behind you, not in front of you as you face forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 20, + 21 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_14", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "While I was sitting on the stool, the colored pencils were in front of/left to me？", + "answer": "When you sit on the stool, the watercolor pens are in front of you.", + "question_chinese": "当我坐在凳子上，水彩笔在我的前/左？", + "answer_chinese": "当你坐在凳子上，水彩笔在你的前面。", + "evidence": [ + { + "description": "The watercolor pens are placed in front of me.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:26:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They were on your left.", + "B": "When you sit on the stool, the watercolor pens are in front of you.", + "C": "They were behind you.", + "D": "While sitting on the stool, the watercolor pens were to your right, not in front of you.", + "E": "When you sat on the stool, the watercolor pens were beside you rather than directly in front." + }, + "answer_label": "B", + "answer_text": "When you sit on the stool, the watercolor pens are in front of you.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_14", + "correct_answer": "When you sit on the stool, the watercolor pens are in front of you.", + "original_options": { + "A": "They were on your left.", + "B": "When you sit on the stool, the watercolor pens are in front of you.", + "C": "They were behind you.", + "D": "While sitting on the stool, the watercolor pens were to your right, not in front of you.", + "E": "When you sat on the stool, the watercolor pens were beside you rather than directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 17 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_7_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I'm sitting on the stool, how do I get to the sofa?", + "answer": "To get to the sofa, you needed to stand up, turn left, and walk about two steps. The sofa was to your left.", + "question_chinese": "当我坐在凳子上，要去沙发那里怎么走？", + "answer_chinese": "你去沙发要站起身左转走过去大概两步。沙发在你的左边。", + "evidence": [ + { + "description": "The sofa is on the left side of the table.", + "timestamps": { + "start": "00:23:48", + "end": "00:23:52" + } + } + ], + "question_timestamps": "00:23:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To get to the sofa, you needed to stand up, turn left, and walk about two steps. The sofa was to your left.", + "B": "Stand up and turn right.", + "C": "Stand up and walk three steps.", + "D": "From the stool, stand up, turn right, and walk about two steps to reach the sofa on your right.", + "E": "Stand up, turn left, then walk about one step; the sofa should be just to your left." + }, + "answer_label": "A", + "answer_text": "To get to the sofa, you needed to stand up, turn left, and walk about two steps. The sofa was to your left.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_7_T1", + "correct_answer": "To get to the sofa, you needed to stand up, turn left, and walk about two steps. The sofa was to your left.", + "original_options": { + "A": "To get to the sofa, you needed to stand up, turn left, and walk about two steps. The sofa was to your left.", + "B": "Stand up and turn right.", + "C": "Stand up and walk three steps.", + "D": "From the stool, stand up, turn right, and walk about two steps to reach the sofa on your right.", + "E": "Stand up, turn left, then walk about one step; the sofa should be just to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 17 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_18", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I sit on the stool, how do I get to the sofa?", + "answer": "You need to stand up and turn left to go to the sofa.", + "question_chinese": "当我坐在凳子上，要去沙发那里怎么走？", + "answer_chinese": "你去沙发要站起身左转。", + "evidence": [ + { + "description": "The sofa is on the left side of the table.", + "timestamps": { + "start": "00:23:48", + "end": "00:23:52" + } + } + ], + "question_timestamps": "00:25:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Stand up and turn right.", + "B": "Stand up and walk straight.", + "C": "You need to stand up and turn around to head toward the sofa from the stool.", + "D": "You should stay seated, then slide to your left to reach the sofa without turning.", + "E": "You need to stand up and turn left to go to the sofa." + }, + "answer_label": "E", + "answer_text": "You need to stand up and turn left to go to the sofa.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_18", + "correct_answer": "You need to stand up and turn left to go to the sofa.", + "original_options": { + "A": "Stand up and turn right.", + "B": "Stand up and walk straight.", + "C": "You need to stand up and turn around to head toward the sofa from the stool.", + "D": "You should stay seated, then slide to your left to reach the sofa without turning.", + "E": "You need to stand up and turn left to go to the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_7_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I'm sitting on the stool, how do I get to the sofa?", + "answer": "To get to the sofa, you needed to stand up and turn left. When you stood up, the sofa was behind you.", + "question_chinese": "当我坐在凳子上，要去沙发那里怎么走？", + "answer_chinese": "你去沙发要站起身左转。当你起身的时候沙发在你的后方。", + "evidence": [ + { + "description": "The sofa is on the left side of the table.", + "timestamps": { + "start": "00:23:48", + "end": "00:23:52" + } + } + ], + "question_timestamps": "00:38:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Stand up and turn right.", + "B": "To get to the sofa, you needed to stand up and turn left. When you stood up, the sofa was behind you.", + "C": "Stand up and face forward.", + "D": "To reach the sofa, you needed to stand up and turn left, and the sofa was in front of you.", + "E": "To get to the sofa, stand up, turn left, and walk about two steps; the sofa was to your left." + }, + "answer_label": "B", + "answer_text": "To get to the sofa, you needed to stand up and turn left. When you stood up, the sofa was behind you.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_7_T2", + "correct_answer": "To get to the sofa, you needed to stand up and turn left. When you stood up, the sofa was behind you.", + "original_options": { + "A": "Stand up and turn right.", + "B": "To get to the sofa, you needed to stand up and turn left. When you stood up, the sofa was behind you.", + "C": "Stand up and face forward.", + "D": "To reach the sofa, you needed to stand up and turn left, and the sofa was in front of you.", + "E": "To get to the sofa, stand up, turn left, and walk about two steps; the sofa was to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 20 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_17", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the iron when I sit on the chair?", + "answer": "When you sit on the chair, you can't reach the iron.", + "question_chinese": "当我坐在椅子上，我能够到熨斗吗？", + "answer_chinese": "当你坐在椅子上，你不能够到熨斗。", + "evidence": [ + { + "description": "The iron is on the table to the right.", + "timestamps": { + "start": "00:23:40", + "end": "00:23:48" + } + } + ], + "question_timestamps": "00:33:06", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you sit on the chair, you can't reach the iron.", + "B": "When you sit on the chair, you can reach the iron." + }, + "answer_label": "A", + "answer_text": "When you sit on the chair, you can't reach the iron.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_17", + "correct_answer": "When you sit on the chair, you can't reach the iron.", + "original_options": { + "A": "When you sit on the chair, you can't reach the iron.", + "B": "When you sit on the chair, you can reach the iron." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_16", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the iron when I sit on the chair?", + "answer": "When you sit on the chair, you can't reach the iron.", + "question_chinese": "当我坐在椅子上，我能够到熨斗吗？", + "answer_chinese": "当你坐在椅子上，你不能够到熨斗。", + "evidence": [ + { + "description": "The iron is on the table to the right.", + "timestamps": { + "start": "00:23:40", + "end": "00:23:48" + } + } + ], + "question_timestamps": "00:44:55", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you sit on the chair, you can reach the iron.", + "B": "When you sit on the chair, you can't reach the iron." + }, + "answer_label": "B", + "answer_text": "When you sit on the chair, you can't reach the iron.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_16", + "correct_answer": "When you sit on the chair, you can't reach the iron.", + "original_options": { + "A": "When you sit on the chair, you can reach the iron.", + "B": "When you sit on the chair, you can't reach the iron." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_11", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color fabric did I take from the box in the lower right corner?", + "answer": "You took the flesh-pink fabric from the box in the lower right corner.", + "question_chinese": "我在右下角的箱子里拿了什么颜色的布料？", + "answer_chinese": "你在右下角的箱子里拿了肉粉的布料。", + "evidence": [ + { + "description": "You took the flesh-pink fabric from the box in the lower right corner.", + "timestamps": { + "start": "00:24:15", + "end": "00:24:35" + } + } + ], + "question_timestamps": "00:30:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took the yellow fabric.", + "B": "You took the dark green fabric.", + "C": "You took the bright blue fabric from the box in the lower right corner.", + "D": "You took the flesh-pink fabric from the box in the lower right corner.", + "E": "You took the red and white fabric from the box in the lower right corner." + }, + "answer_label": "D", + "answer_text": "You took the flesh-pink fabric from the box in the lower right corner.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_11", + "correct_answer": "You took the flesh-pink fabric from the box in the lower right corner.", + "original_options": { + "A": "You took the yellow fabric.", + "B": "You took the dark green fabric.", + "C": "You took the bright blue fabric from the box in the lower right corner.", + "D": "You took the flesh-pink fabric from the box in the lower right corner.", + "E": "You took the red and white fabric from the box in the lower right corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 15 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_12", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color fabric did I take from the box in the lower right corner?", + "answer": "You took the flesh-pink fabric from the box in the lower right corner.", + "question_chinese": "我在右下角的箱子里拿了什么颜色的布料？", + "answer_chinese": "你在右下角的箱子里拿了肉粉的布料。", + "evidence": [ + { + "description": "You took the flesh-pink fabric from the box in the lower right corner.", + "timestamps": { + "start": "00:24:15", + "end": "00:24:35" + } + } + ], + "question_timestamps": "00:42:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took the green fabric.", + "B": "You took the blue fabric.", + "C": "You took the white fabric from the box in the lower right corner.", + "D": "You took the striped fabric from the box in the lower right corner.", + "E": "You took the flesh-pink fabric from the box in the lower right corner." + }, + "answer_label": "E", + "answer_text": "You took the flesh-pink fabric from the box in the lower right corner.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_12", + "correct_answer": "You took the flesh-pink fabric from the box in the lower right corner.", + "original_options": { + "A": "You took the green fabric.", + "B": "You took the blue fabric.", + "C": "You took the white fabric from the box in the lower right corner.", + "D": "You took the striped fabric from the box in the lower right corner.", + "E": "You took the flesh-pink fabric from the box in the lower right corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_2_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was sitting on the chair, how many pairs of scissors were on the table?", + "answer": "When you were sitting on the chair, there were two pairs of scissors on the table—one in front of you and one by your right hand.", + "question_chinese": "我坐在椅子上的时候，桌子上有几把剪刀？", + "answer_chinese": "你坐在椅子上的时候，桌子上有两把剪刀。一把在你的前面，一把在你的右手边。", + "evidence": [ + { + "description": "There is a pair of silver scissors on the table.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:35" + } + }, + { + "description": "I took out a big black pair of scissors again.", + "timestamps": { + "start": "00:11:38", + "end": "00:12:19" + } + } + ], + "question_timestamps": "00:24:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "When you were sitting on the chair, there were two pairs of scissors on the table—one in front of you and one by your right hand.", + "B": "One pair of scissors.", + "C": "Three pairs of scissors.", + "D": "There was one pair of scissors on the table, sitting directly in front of you.", + "E": "There were four pairs of scissors on the table, one near you and three spread along the far edge." + }, + "answer_label": "A", + "answer_text": "When you were sitting on the chair, there were two pairs of scissors on the table—one in front of you and one by your right hand.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_2_T1", + "correct_answer": "When you were sitting on the chair, there were two pairs of scissors on the table—one in front of you and one by your right hand.", + "original_options": { + "A": "When you were sitting on the chair, there were two pairs of scissors on the table—one in front of you and one by your right hand.", + "B": "One pair of scissors.", + "C": "Three pairs of scissors.", + "D": "There was one pair of scissors on the table, sitting directly in front of you.", + "E": "There were four pairs of scissors on the table, one near you and three spread along the far edge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 19 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_2_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was sitting on the chair, how many pairs of scissors were on the table?", + "answer": "When you were sitting on the chair, there was a pair of scissors on the table. It was to your front-right.", + "question_chinese": "我坐在椅子上的时候，桌子上有几把剪刀？", + "answer_chinese": "你坐在椅子上的时候，桌子上有一把剪刀。剪刀在你的右前方。", + "evidence": [ + { + "description": "There is a pair of silver scissors on the table.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:24:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "No scissors were on the table.", + "B": "When you were sitting on the chair, there was a pair of scissors on the table. It was to your front-right.", + "C": "Two pairs of scissors.", + "D": "There were two pairs of scissors on the table, one in front of you and one by your right hand.", + "E": "There were three pairs of scissors on the table, one front-right, one centered, and one to your left." + }, + "answer_label": "B", + "answer_text": "When you were sitting on the chair, there was a pair of scissors on the table. It was to your front-right.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_2_T2", + "correct_answer": "When you were sitting on the chair, there was a pair of scissors on the table. It was to your front-right.", + "original_options": { + "A": "No scissors were on the table.", + "B": "When you were sitting on the chair, there was a pair of scissors on the table. It was to your front-right.", + "C": "Two pairs of scissors.", + "D": "There were two pairs of scissors on the table, one in front of you and one by your right hand.", + "E": "There were three pairs of scissors on the table, one front-right, one centered, and one to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 20, + 18 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_5", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I sit on the chair, is the black scissors on my left/right?", + "answer": "When you sit on the chair, the black scissors are on your right.", + "question_chinese": "我坐在椅子上的时候，黑色剪刀在我的左/右？", + "answer_chinese": "你坐在椅子上的时候，黑色剪刀在你的右边。", + "evidence": [ + { + "description": "I put the black scissors on the right.", + "timestamps": { + "start": "00:12:10", + "end": "00:12:18" + } + } + ], + "question_timestamps": "00:16:34", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you sit on the chair, the black scissors are on your left.", + "B": "When you sit on the chair, the black scissors are on your right." + }, + "answer_label": "B", + "answer_text": "When you sit on the chair, the black scissors are on your right.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_5", + "correct_answer": "When you sit on the chair, the black scissors are on your right.", + "original_options": { + "A": "When you sit on the chair, the black scissors are on your left.", + "B": "When you sit on the chair, the black scissors are on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_6", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I sit on the chair, is the black scissors on my left/right?", + "answer": "When you sit on the chair, the black scissors are on your left.", + "question_chinese": "我坐在椅子上的时候，黑色剪刀在我的左/右？", + "answer_chinese": "你坐在椅子上的时候，黑色剪刀在你的左边。", + "evidence": [ + { + "description": "I put the black scissors on the right.", + "timestamps": { + "start": "00:12:10", + "end": "00:12:18" + } + }, + { + "description": "I took the black scissors to the left.", + "timestamps": { + "start": "00:20:35", + "end": "00:20:37" + } + } + ], + "question_timestamps": "00:38:38", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "When you sit on the chair, the black scissors are on your left.", + "B": "When you sit on the chair, the black scissors are on your right." + }, + "answer_label": "A", + "answer_text": "When you sit on the chair, the black scissors are on your left.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_6", + "correct_answer": "When you sit on the chair, the black scissors are on your left.", + "original_options": { + "A": "When you sit on the chair, the black scissors are on your left.", + "B": "When you sit on the chair, the black scissors are on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_9", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I go to get more adhesive bandages?", + "answer": "You need more bandages, go get them from the table on the right.", + "question_chinese": "我要拿更多的百粘布要去哪里？", + "answer_chinese": "你要更多的百粘布要去右边的桌子上拿。", + "evidence": [ + { + "description": "The white cloth is placed on the table to the right.", + "timestamps": { + "start": "00:23:25", + "end": "00:23:30" + } + } + ], + "question_timestamps": "00:26:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Get them from the left drawer.", + "B": "Get them from the shelf above.", + "C": "You need more bandages, go get them from the drawer on the left side of the table.", + "D": "You need more bandages, go get them from the table on the right.", + "E": "You need more bandages, go get them from the shelf near the sink rather than the table." + }, + "answer_label": "D", + "answer_text": "You need more bandages, go get them from the table on the right.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_9", + "correct_answer": "You need more bandages, go get them from the table on the right.", + "original_options": { + "A": "Get them from the left drawer.", + "B": "Get them from the shelf above.", + "C": "You need more bandages, go get them from the drawer on the left side of the table.", + "D": "You need more bandages, go get them from the table on the right.", + "E": "You need more bandages, go get them from the shelf near the sink rather than the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_10", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I go to get more adhesive bandages?", + "answer": "You need more bandages, go get them from the table on the right.", + "question_chinese": "我要拿更多的百粘布要去哪里？", + "answer_chinese": "你要更多的百粘布要去右边的桌子上拿。", + "evidence": [ + { + "description": "The white cloth is placed on the table to the right.", + "timestamps": { + "start": "00:23:25", + "end": "00:23:30" + } + } + ], + "question_timestamps": "00:38:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Get them from the left shelf.", + "B": "Get them from the right drawer.", + "C": "You need more bandages, go get them from the counter in the kitchen area instead of the table.", + "D": "You need more bandages, go get them from the floor in the living room, not from any table.", + "E": "You need more bandages, go get them from the table on the right." + }, + "answer_label": "E", + "answer_text": "You need more bandages, go get them from the table on the right.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_10", + "correct_answer": "You need more bandages, go get them from the table on the right.", + "original_options": { + "A": "Get them from the left shelf.", + "B": "Get them from the right drawer.", + "C": "You need more bandages, go get them from the counter in the kitchen area instead of the table.", + "D": "You need more bandages, go get them from the floor in the living room, not from any table.", + "E": "You need more bandages, go get them from the table on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_7", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there anything in the little dog's hand on the yellow fabric?", + "answer": "A puppy on yellow fabric, holding something in its hand.", + "question_chinese": "黄色布料上的小狗，手里有拿着东西吗？", + "answer_chinese": "黄色布料上的小狗，手里有拿着东西。", + "evidence": [ + { + "description": "A puppy on yellow fabric holding a toothbrush in one hand and toothpaste in the other.", + "timestamps": { + "start": "00:22:46", + "end": "00:22:50" + } + } + ], + "question_timestamps": "00:23:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A puppy on yellow fabric, holding something in its hand.", + "B": "Nothing in its paw.", + "C": "It’s holding a stick.", + "D": "A puppy on yellow fabric, holding a small ball in its paw instead.", + "E": "A puppy on yellow fabric, with empty paws, not holding anything in its hand." + }, + "answer_label": "A", + "answer_text": "A puppy on yellow fabric, holding something in its hand.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_7", + "correct_answer": "A puppy on yellow fabric, holding something in its hand.", + "original_options": { + "A": "A puppy on yellow fabric, holding something in its hand.", + "B": "Nothing in its paw.", + "C": "It’s holding a stick.", + "D": "A puppy on yellow fabric, holding a small ball in its paw instead.", + "E": "A puppy on yellow fabric, with empty paws, not holding anything in its hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 14 + ] + }, + { + "video_uid": "20bd786f-d222-4b0e-ad2c-a2381929cf78", + "video_length": "00:44:55", + "qid": "20bd786f-d222-4b0e-ad2c-a2381929cf78_8", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there anything in the little dog's hand on the yellow fabric?", + "answer": "A puppy on yellow fabric, holding nothing in its paws.", + "question_chinese": "黄色布料上的小狗，手里有拿着东西吗？", + "answer_chinese": "黄色布料上的小狗，手里没拿着东西。", + "evidence": [ + { + "description": "A puppy on yellow fabric holding a toothbrush in one hand and toothpaste in the other.", + "timestamps": { + "start": "00:22:46", + "end": "00:22:50" + } + }, + { + "description": "I took the things from the puppy's hand.", + "timestamps": { + "start": "00:24:10", + "end": "00:24:13" + } + } + ], + "question_timestamps": "00:28:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It’s holding something in its paws.", + "B": "It’s holding a stick.", + "C": "A puppy on yellow fabric, holding nothing in its paws.", + "D": "A puppy on yellow fabric, holding treats in its paws instead of having empty paws.", + "E": "A puppy on yellow fabric, gripping a small toy in its paws rather than holding nothing." + }, + "answer_label": "C", + "answer_text": "A puppy on yellow fabric, holding nothing in its paws.", + "q_id": "20bd786f-d222-4b0e-ad2c-a2381929cf78_8", + "correct_answer": "A puppy on yellow fabric, holding nothing in its paws.", + "original_options": { + "A": "It’s holding something in its paws.", + "B": "It’s holding a stick.", + "C": "A puppy on yellow fabric, holding nothing in its paws.", + "D": "A puppy on yellow fabric, holding treats in its paws instead of having empty paws.", + "E": "A puppy on yellow fabric, gripping a small toy in its paws rather than holding nothing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 16 + ] + } +] diff --git a/QAs/21abd7c0-2527-487a-a249-36d47408bae0-mcq.json b/QAs/21abd7c0-2527-487a-a249-36d47408bae0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..94ba5e9fe5ee6502c0c38866da61157784c678bd --- /dev/null +++ b/QAs/21abd7c0-2527-487a-a249-36d47408bae0-mcq.json @@ -0,0 +1,673 @@ +[ + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I assembled vacuum cleaners so far?", + "answer": "You assembled the vacuum cleaner twice.", + "question_chinese": "到现在为止，我一共组装了多少次吸尘器？", + "answer_chinese": "你组装了两次吸尘器。", + "evidence": [ + { + "description": "You took apart the vacuum cleaner, emptied the dirt, and assembled it again.", + "timestamps": { + "start": "00:09:55", + "end": "00:10:05" + } + }, + { + "description": "You are assembling a vacuum cleaner for the first time.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:10:35", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You assembled the vacuum cleaner once.", + "B": "You assembled the vacuum cleaner twice.", + "C": "You assembled the vacuum cleaner three times.", + "D": "You assembled the vacuum cleaner four times in total so far, counting each complete assembly you finished.", + "E": "You assembled the vacuum cleaner five times overall, including each time you put the main parts together." + }, + "answer_label": "B", + "answer_text": "You assembled the vacuum cleaner twice.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_7", + "correct_answer": "You assembled the vacuum cleaner twice.", + "original_options": { + "A": "You assembled the vacuum cleaner once.", + "B": "You assembled the vacuum cleaner twice.", + "C": "You assembled the vacuum cleaner three times.", + "D": "You assembled the vacuum cleaner four times in total so far, counting each complete assembly you finished.", + "E": "You assembled the vacuum cleaner five times overall, including each time you put the main parts together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 17 + ] + }, + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I assembled vacuum cleaners so far?", + "answer": "You have assembled the vacuum cleaner three times.", + "question_chinese": "到现在为止，我一共组装了多少次吸尘器？", + "answer_chinese": "你组装了三次吸尘器。", + "evidence": [ + { + "description": "You assembled a vacuum cleaner for the last time.", + "timestamps": { + "start": "00:28:03", + "end": "00:28:10" + } + }, + { + "description": "You took apart the vacuum cleaner, emptied the dirt, and assembled it again.", + "timestamps": { + "start": "00:09:55", + "end": "00:10:05" + } + }, + { + "description": "You are assembling a vacuum cleaner for the first time.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:28:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have assembled the vacuum cleaner twice.", + "B": "You have assembled the vacuum cleaner four times.", + "C": "You have assembled the vacuum cleaner three times.", + "D": "You have assembled the vacuum cleaner five times so far, based on each full build you completed.", + "E": "You have assembled the vacuum cleaner six times in total up to now, counting every completed assembly." + }, + "answer_label": "C", + "answer_text": "You have assembled the vacuum cleaner three times.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_8", + "correct_answer": "You have assembled the vacuum cleaner three times.", + "original_options": { + "A": "You have assembled the vacuum cleaner twice.", + "B": "You have assembled the vacuum cleaner four times.", + "C": "You have assembled the vacuum cleaner three times.", + "D": "You have assembled the vacuum cleaner five times so far, based on each full build you completed.", + "E": "You have assembled the vacuum cleaner six times in total up to now, counting every completed assembly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 17 + ] + }, + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_13", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to move the hand sanitizer closer to the bottle on the right, how far should I move it?", + "answer": "Move the hand sanitizer ten centimeters to the right.", + "question_chinese": "如果我想让洗手液和右边的瓶子放得近一些,应该移动多远?", + "answer_chinese": "洗手液向右移动十厘米。", + "evidence": [ + { + "description": "They are currently about fifteen centimeters apart. You can move the hand sanitizer to the right by about ten centimeters.", + "timestamps": { + "start": "00:17:35", + "end": "00:17:41" + } + } + ], + "question_timestamps": "00:17:46", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Move it five centimeters right.", + "B": "Move it ten centimeters left.", + "C": "Move the hand sanitizer fifteen centimeters to the right to bring it closer to the bottle on the right.", + "D": "Move the hand sanitizer ten centimeters to the right.", + "E": "Move the hand sanitizer twenty centimeters to the right so it lines up closer to the bottle on the right." + }, + "answer_label": "D", + "answer_text": "Move the hand sanitizer ten centimeters to the right.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_13", + "correct_answer": "Move the hand sanitizer ten centimeters to the right.", + "original_options": { + "A": "Move it five centimeters right.", + "B": "Move it ten centimeters left.", + "C": "Move the hand sanitizer fifteen centimeters to the right to bring it closer to the bottle on the right.", + "D": "Move the hand sanitizer ten centimeters to the right.", + "E": "Move the hand sanitizer twenty centimeters to the right so it lines up closer to the bottle on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 20 + ] + }, + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_14", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to move the hand sanitizer closer to the bottle on the right, how far should I move it?", + "answer": "Move the hand sanitizer ten centimeters to the right.", + "question_chinese": "如果我想让洗手液和右边的瓶子放得近一些,应该移动多远?", + "answer_chinese": "洗手液向右移动十厘米。", + "evidence": [ + { + "description": "They are currently about fifteen centimeters apart. You can move the hand sanitizer to the right by about ten centimeters.", + "timestamps": { + "start": "00:17:35", + "end": "00:17:41" + } + } + ], + "question_timestamps": "00:20:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Move the hand sanitizer ten centimeters to the right.", + "B": "Move it six centimeters right.", + "C": "Move it eight centimeters right.", + "D": "Move the hand sanitizer twelve centimeters to the right to close the gap to the bottle on the right.", + "E": "Move the hand sanitizer fifteen centimeters to the right so it sits closer to the bottle on the right." + }, + "answer_label": "A", + "answer_text": "Move the hand sanitizer ten centimeters to the right.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_14", + "correct_answer": "Move the hand sanitizer ten centimeters to the right.", + "original_options": { + "A": "Move the hand sanitizer ten centimeters to the right.", + "B": "Move it six centimeters right.", + "C": "Move it eight centimeters right.", + "D": "Move the hand sanitizer twelve centimeters to the right to close the gap to the bottle on the right.", + "E": "Move the hand sanitizer fifteen centimeters to the right so it sits closer to the bottle on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following just now?", + "answer": "You started from the living room, walked straight for four steps, turned right, and moved toward the kitchen.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你从客厅出发直走四步，右转，朝着厨房方向移动。", + "evidence": [ + { + "description": "Take four steps straight ahead, turn right, and then take five steps straight ahead to the kitchen.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From living room, turned left toward dining room.", + "B": "From hallway, turned right toward kitchen.", + "C": "You started in the dining room, walked straight for three steps, turned right, and moved toward the kitchen.", + "D": "You started from the living room, walked straight for four steps, turned right, and moved toward the kitchen.", + "E": "You started from the living room, walked straight for five steps, turned right, and then moved toward the kitchen." + }, + "answer_label": "D", + "answer_text": "You started from the living room, walked straight for four steps, turned right, and moved toward the kitchen.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_1", + "correct_answer": "You started from the living room, walked straight for four steps, turned right, and moved toward the kitchen.", + "original_options": { + "A": "From living room, turned left toward dining room.", + "B": "From hallway, turned right toward kitchen.", + "C": "You started in the dining room, walked straight for three steps, turned right, and moved toward the kitchen.", + "D": "You started from the living room, walked straight for four steps, turned right, and moved toward the kitchen.", + "E": "You started from the living room, walked straight for five steps, turned right, and then moved toward the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 18, + 19 + ] + }, + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following just now?", + "answer": "You started from the hallway and moved toward the bathroom.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你从楼道出发，朝着厕所方向移动。", + "evidence": [ + { + "description": "You are carrying a vacuum cleaner from the hallway towards the bathroom.", + "timestamps": { + "start": "00:15:18", + "end": "00:15:28" + } + } + ], + "question_timestamps": "00:15:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From kitchen toward dining room.", + "B": "From living room toward garage.", + "C": "You left the bathroom and went toward the front door instead of starting in the hallway.", + "D": "You started at the stairs and headed to the utility closet rather than going from hallway to bathroom.", + "E": "You started from the hallway and moved toward the bathroom." + }, + "answer_label": "E", + "answer_text": "You started from the hallway and moved toward the bathroom.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_2", + "correct_answer": "You started from the hallway and moved toward the bathroom.", + "original_options": { + "A": "From kitchen toward dining room.", + "B": "From living room toward garage.", + "C": "You left the bathroom and went toward the front door instead of starting in the hallway.", + "D": "You started at the stairs and headed to the utility closet rather than going from hallway to bathroom.", + "E": "You started from the hallway and moved toward the bathroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many dolls were there on the sofa when I first entered the room?", + "answer": "One.", + "question_chinese": "我刚进房间时，沙发上有几个玩偶？", + "answer_chinese": "一个。", + "evidence": [ + { + "description": "There is a black doll on the side of the sofa near the door.", + "timestamps": { + "start": "00:18:43", + "end": "00:18:48" + } + } + ], + "question_timestamps": "00:18:50", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "There were four dolls on the sofa when you first entered the room.", + "D": "One.", + "E": "There were five dolls sitting on the sofa at the moment you first walked in." + }, + "answer_label": "D", + "answer_text": "One.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_11", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "There were four dolls on the sofa when you first entered the room.", + "D": "One.", + "E": "There were five dolls sitting on the sofa at the moment you first walked in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_12", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many dolls were there on the sofa when I first entered the room?", + "answer": "Two.", + "question_chinese": "我刚进房间时，沙发上有几个玩偶？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "There is a black doll on the side of the sofa near the door.", + "timestamps": { + "start": "00:18:43", + "end": "00:18:48" + } + }, + { + "description": "There is also a black doll on the other side of the sofa.", + "timestamps": { + "start": "00:18:59", + "end": "00:19:08" + } + } + ], + "question_timestamps": "00:19:11", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Zero.", + "C": "Three.", + "D": "There were six dolls on the sofa when you first entered the room.", + "E": "There were seven dolls on the sofa at the time you first walked into the room." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_12", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "Zero.", + "C": "Three.", + "D": "There were six dolls on the sofa when you first entered the room.", + "E": "There were seven dolls on the sofa at the time you first walked into the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 16 + ] + }, + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_9", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I find more clothes?", + "answer": "You can look for it on the clothesline.", + "question_chinese": "我可以去哪找更多的衣服？", + "answer_chinese": "你可以去晾衣架上找。", + "evidence": [ + { + "description": "There are many clothes on the clothesline.", + "timestamps": { + "start": "00:12:55", + "end": "00:12:58" + } + } + ], + "question_timestamps": "00:13:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the drying rack.", + "B": "Inside the wardrobe.", + "C": "You can look for it hanging on the porch railing near where you were standing.", + "D": "You can look for it hanging on the balcony area instead of on the clothesline.", + "E": "You can look for it on the clothesline." + }, + "answer_label": "E", + "answer_text": "You can look for it on the clothesline.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_9", + "correct_answer": "You can look for it on the clothesline.", + "original_options": { + "A": "On the drying rack.", + "B": "Inside the wardrobe.", + "C": "You can look for it hanging on the porch railing near where you were standing.", + "D": "You can look for it hanging on the balcony area instead of on the clothesline.", + "E": "You can look for it on the clothesline." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 15 + ] + }, + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_10", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I find more clothes?", + "answer": "You can look on the clothesline and in the wardrobe.", + "question_chinese": "我可以去哪找更多的衣服？", + "answer_chinese": "你可以去晾衣架上和衣柜里找。", + "evidence": [ + { + "description": "There are also many clothes in the wardrobe of the room.", + "timestamps": { + "start": "00:19:43", + "end": "00:19:46" + } + }, + { + "description": "There are many clothes on the clothesline.", + "timestamps": { + "start": "00:12:58", + "end": "00:13:03" + } + } + ], + "question_timestamps": "00:21:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Only on the clothesline.", + "B": "Only in the wardrobe.", + "C": "You can look in the wardrobe first, then check the dresser nearby for any extra clothes.", + "D": "You can look on the clothesline and in the wardrobe.", + "E": "You can check the clothesline, and if it’s empty, search around the room for folded clothes." + }, + "answer_label": "D", + "answer_text": "You can look on the clothesline and in the wardrobe.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_10", + "correct_answer": "You can look on the clothesline and in the wardrobe.", + "original_options": { + "A": "Only on the clothesline.", + "B": "Only in the wardrobe.", + "C": "You can look in the wardrobe first, then check the dresser nearby for any extra clothes.", + "D": "You can look on the clothesline and in the wardrobe.", + "E": "You can check the clothesline, and if it’s empty, search around the room for folded clothes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_5", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Compared to my current position, where is the chair?", + "answer": "The chair is directly to your left.", + "question_chinese": "相对我现在的位置，椅子在哪里？", + "answer_chinese": "椅子在你的正左方。", + "evidence": [ + { + "description": "There is a chair on your left.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:03" + } + } + ], + "question_timestamps": "00:05:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Directly to your right.", + "B": "Directly behind you.", + "C": "The chair is directly to your left.", + "D": "The chair is straight ahead of you, centered in your line of sight.", + "E": "The chair is slightly behind and to your right, not on your left side." + }, + "answer_label": "C", + "answer_text": "The chair is directly to your left.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_5", + "correct_answer": "The chair is directly to your left.", + "original_options": { + "A": "Directly to your right.", + "B": "Directly behind you.", + "C": "The chair is directly to your left.", + "D": "The chair is straight ahead of you, centered in your line of sight.", + "E": "The chair is slightly behind and to your right, not on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 13, + 14 + ] + }, + { + "video_uid": "21abd7c0-2527-487a-a249-36d47408bae0", + "video_length": "00:28:11", + "qid": "21abd7c0-2527-487a-a249-36d47408bae0_6", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Compared to my current position, where is the chair?", + "answer": "The chair is directly in front of you.", + "question_chinese": "相对我现在的位置，椅子在哪里？", + "answer_chinese": "椅子在你的正前方。", + "evidence": [ + { + "description": "There is a chair in front of you.", + "timestamps": { + "start": "00:08:51", + "end": "00:08:52" + } + } + ], + "question_timestamps": "00:08:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Directly to your left.", + "B": "Directly behind you.", + "C": "The chair is directly in front of you.", + "D": "The chair is just to your right, slightly off your forward direction rather than straight ahead.", + "E": "The chair is behind you and a bit to the left, so you’d need to turn around to see it." + }, + "answer_label": "C", + "answer_text": "The chair is directly in front of you.", + "q_id": "21abd7c0-2527-487a-a249-36d47408bae0_6", + "correct_answer": "The chair is directly in front of you.", + "original_options": { + "A": "Directly to your left.", + "B": "Directly behind you.", + "C": "The chair is directly in front of you.", + "D": "The chair is just to your right, slightly off your forward direction rather than straight ahead.", + "E": "The chair is behind you and a bit to the left, so you’d need to turn around to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 21 + ] + } +] diff --git a/QAs/21e0e93d-b97e-40a1-83bf-9267448231fd-mcq.json b/QAs/21e0e93d-b97e-40a1-83bf-9267448231fd-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6566a1455b688bdfabc328b5b0f628d1a67cf8e5 --- /dev/null +++ b/QAs/21e0e93d-b97e-40a1-83bf-9267448231fd-mcq.json @@ -0,0 +1,707 @@ +[ + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_13", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I turn left or right when entering the convenience store just now?", + "answer": "You turned right.", + "question_chinese": "刚才我是左转还是右转走进便利店？", + "answer_chinese": "你右转了", + "evidence": [ + { + "description": "Turned right when entering the convenience store", + "timestamps": { + "start": "00:00:22", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You turned left.", + "B": "You turned right." + }, + "answer_label": "B", + "answer_text": "You turned right.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_13", + "correct_answer": "You turned right.", + "original_options": { + "A": "You turned left.", + "B": "You turned right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_14", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I turn left or right when entering the convenience store just now?", + "answer": "You turned left.", + "question_chinese": "刚才我是左转还是右转走进便利店？", + "answer_chinese": "你左转了", + "evidence": [ + { + "description": "Turned left when entering the convenience store", + "timestamps": { + "start": "00:06:50", + "end": "00:06:55" + } + } + ], + "question_timestamps": "00:07:36", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You turned right.", + "B": "You turned left." + }, + "answer_label": "B", + "answer_text": "You turned left.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_14", + "correct_answer": "You turned left.", + "original_options": { + "A": "You turned right.", + "B": "You turned left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_1_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Just now, what area of the convenience store did I pass through first?", + "answer": "The first areas you passed were the vegetable section and the chocolate section. The vegetable section was to your right, and the chocolate section was to your left.", + "question_chinese": "刚才我最先经过便利店的什么区域？", + "answer_chinese": "你最先经过了蔬菜区和巧克力区。蔬菜区在你的右边，巧克力区在你的左边。", + "evidence": [ + { + "description": "Right side is vegetables, left side is chocolate", + "timestamps": { + "start": "00:00:30", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:00:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You first passed the vegetable section.", + "B": "The first areas you passed were the vegetable section and the chocolate section. The vegetable section was to your right, and the chocolate section was to your left.", + "C": "You first passed the chocolate section.", + "D": "The first areas you passed were the vegetable and chocolate sections, with vegetables on your left and chocolate on your right.", + "E": "You first went through the vegetable and chocolate sections, but the vegetable section was in front of you, not to the right." + }, + "answer_label": "B", + "answer_text": "The first areas you passed were the vegetable section and the chocolate section. The vegetable section was to your right, and the chocolate section was to your left.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_1_T1", + "correct_answer": "The first areas you passed were the vegetable section and the chocolate section. The vegetable section was to your right, and the chocolate section was to your left.", + "original_options": { + "A": "You first passed the vegetable section.", + "B": "The first areas you passed were the vegetable section and the chocolate section. The vegetable section was to your right, and the chocolate section was to your left.", + "C": "You first passed the chocolate section.", + "D": "The first areas you passed were the vegetable and chocolate sections, with vegetables on your left and chocolate on your right.", + "E": "You first went through the vegetable and chocolate sections, but the vegetable section was in front of you, not to the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 22 + ] + }, + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_1_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Just now, what area of the convenience store did I pass through first?", + "answer": "The first area you passed was the vegetable section. The vegetable section was behind you.", + "question_chinese": "刚才我最先经过便利店的什么区域？", + "answer_chinese": "你最先经过了蔬菜区。蔬菜区在你的后边。", + "evidence": [ + { + "description": "Right side is vegetables", + "timestamps": { + "start": "00:00:30", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The first area was the chocolate section.", + "B": "The vegetable section was in front.", + "C": "The first area you passed was the vegetable section. The vegetable section was behind you.", + "D": "The first area you passed was the vegetable section, and it was to your left rather than behind you.", + "E": "The first areas you passed were the vegetable and chocolate sections, with vegetables on the right and chocolate on the left." + }, + "answer_label": "C", + "answer_text": "The first area you passed was the vegetable section. The vegetable section was behind you.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_1_T2", + "correct_answer": "The first area you passed was the vegetable section. The vegetable section was behind you.", + "original_options": { + "A": "The first area was the chocolate section.", + "B": "The vegetable section was in front.", + "C": "The first area you passed was the vegetable section. The vegetable section was behind you.", + "D": "The first area you passed was the vegetable section, and it was to your left rather than behind you.", + "E": "The first areas you passed were the vegetable and chocolate sections, with vegetables on the right and chocolate on the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 21 + ] + }, + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I talked with the staff so far?", + "answer": "Three.", + "question_chinese": "现在我已经和店员交谈几次了？", + "answer_chinese": "3次", + "evidence": [ + { + "description": "Talking with staff", + "timestamps": { + "start": "00:02:46", + "end": "00:02:53" + } + }, + { + "description": "Checking out at the staff counter", + "timestamps": { + "start": "00:15:00", + "end": "00:15:15" + } + }, + { + "description": "Checking out at the staff counter", + "timestamps": { + "start": "00:18:00", + "end": "00:18:15" + } + } + ], + "question_timestamps": "00:20:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You talked with the staff twice in total by now, not three times.", + "E": "You have spoken with the staff four different times so far in total, not three." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_5", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You talked with the staff twice in total by now, not three times.", + "E": "You have spoken with the staff four different times so far in total, not three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I talked with the staff so far?", + "answer": "Four.", + "question_chinese": "现在我已经和店员交谈几次了？", + "answer_chinese": "4次", + "evidence": [ + { + "description": "Talking with staff", + "timestamps": { + "start": "00:02:46", + "end": "00:02:53" + } + }, + { + "description": "Checking out at the staff counter", + "timestamps": { + "start": "00:15:00", + "end": "00:15:15" + } + }, + { + "description": "Checking out at the staff counter", + "timestamps": { + "start": "00:18:00", + "end": "00:18:15" + } + }, + { + "description": "Now talking with staff", + "timestamps": { + "start": "00:25:34", + "end": "00:25:44" + } + } + ], + "question_timestamps": "00:25:44", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You have talked with the staff two times so far, not four.", + "E": "You have talked with the staff five times up to now, rather than four." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_6", + "correct_answer": "Four.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You have talked with the staff two times so far, not four.", + "E": "You have talked with the staff five times up to now, rather than four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_3_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many laps around the convenience store have I made by now?", + "answer": "You had already walked around 4 laps. You were inside the convenience store.", + "question_chinese": "现在我已经逛了便利店多少圈？", + "answer_chinese": "你已经逛了4圈。你现在在便利店里面。", + "evidence": [ + { + "description": "Have walked one lap around the convenience store now", + "timestamps": { + "start": "00:00:31", + "end": "00:01:00" + } + }, + { + "description": "Have walked two laps around the convenience store now", + "timestamps": { + "start": "00:01:01", + "end": "00:01:19" + } + }, + { + "description": "Have walked three laps around the convenience store now", + "timestamps": { + "start": "00:01:33", + "end": "00:02:12" + } + }, + { + "description": "Have walked the fourth lap around the convenience store now", + "timestamps": { + "start": "00:06:50", + "end": "00:10:33" + } + } + ], + "question_timestamps": "00:12:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You walked 3 laps inside.", + "B": "You walked 5 laps inside.", + "C": "You had already walked around 2 laps, and you were inside the convenience store.", + "D": "You had already walked around 6 laps, and you were still inside the convenience store.", + "E": "You had already walked around 4 laps. You were inside the convenience store." + }, + "answer_label": "E", + "answer_text": "You had already walked around 4 laps. You were inside the convenience store.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_3_T2", + "correct_answer": "You had already walked around 4 laps. You were inside the convenience store.", + "original_options": { + "A": "You walked 3 laps inside.", + "B": "You walked 5 laps inside.", + "C": "You had already walked around 2 laps, and you were inside the convenience store.", + "D": "You had already walked around 6 laps, and you were still inside the convenience store.", + "E": "You had already walked around 4 laps. You were inside the convenience store." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_3_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many laps around the convenience store have I made by now?", + "answer": "You had already walked around 5 laps. The convenience store was in front of you.", + "question_chinese": "现在我已经逛了便利店多少圈？", + "answer_chinese": "你已经逛了5圈。便利店在你的前面。", + "evidence": [ + { + "description": "Have walked one lap around the convenience store now", + "timestamps": { + "start": "00:00:31", + "end": "00:01:00" + } + }, + { + "description": "Have walked two laps around the convenience store now", + "timestamps": { + "start": "00:01:01", + "end": "00:01:19" + } + }, + { + "description": "Have walked three laps around the convenience store now", + "timestamps": { + "start": "00:01:33", + "end": "00:02:12" + } + }, + { + "description": "Have walked the fourth lap around the convenience store now", + "timestamps": { + "start": "00:06:50", + "end": "00:10:33" + } + }, + { + "description": "Have walked the fifth lap around the convenience store now", + "timestamps": { + "start": "00:11:00", + "end": "00:12:00" + } + } + ], + "question_timestamps": "00:19:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You walked 4 laps.", + "B": "You walked 6 laps.", + "C": "You had already walked around 3 laps, and the convenience store was in front of you.", + "D": "You had already walked around 5 laps. The convenience store was in front of you.", + "E": "You had already walked around 7 laps, and the convenience store was in front of you." + }, + "answer_label": "D", + "answer_text": "You had already walked around 5 laps. The convenience store was in front of you.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_3_T1", + "correct_answer": "You had already walked around 5 laps. The convenience store was in front of you.", + "original_options": { + "A": "You walked 4 laps.", + "B": "You walked 6 laps.", + "C": "You had already walked around 3 laps, and the convenience store was in front of you.", + "D": "You had already walked around 5 laps. The convenience store was in front of you.", + "E": "You had already walked around 7 laps, and the convenience store was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_11", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the tomatoes directly if I want to get them now?", + "answer": "Yes.", + "question_chinese": "现在我想拿西红柿可以直接够到吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "There are tomatoes on the right side after entering the convenience store", + "timestamps": { + "start": "00:00:30", + "end": "00:00:37" + } + }, + { + "description": "Now shopping in the convenience store", + "timestamps": { + "start": "00:06:48", + "end": "00:06:52" + } + } + ], + "question_timestamps": "00:06:56", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_11", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_12", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the tomatoes directly if I want to get them now?", + "answer": "No.", + "question_chinese": "现在我想拿西红柿可以直接够到吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "There are tomatoes on the right side after entering the convenience store", + "timestamps": { + "start": "00:00:30", + "end": "00:00:37" + } + }, + { + "description": "The checkout counter is far from the fruit area", + "timestamps": { + "start": "00:16:04", + "end": "00:16:12" + } + } + ], + "question_timestamps": "00:17:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_12", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_9", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the checkout counter located for me now?", + "answer": "Back side.", + "question_chinese": "现在收银台在我的什么位置？", + "answer_chinese": "后方", + "evidence": [ + { + "description": "When I was in the aisle, the checkout counter was on my left, I now turn right into the right area of the convenience store", + "timestamps": { + "start": "00:00:29", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Front side.", + "B": "Left side.", + "C": "Back side.", + "D": "It's on the front side of you, positioned ahead rather than behind where you are standing now.", + "E": "It's on your right side at the moment, not on the back side relative to your current position." + }, + "answer_label": "C", + "answer_text": "Back side.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_9", + "correct_answer": "Back side.", + "original_options": { + "A": "Front side.", + "B": "Left side.", + "C": "Back side.", + "D": "It's on the front side of you, positioned ahead rather than behind where you are standing now.", + "E": "It's on your right side at the moment, not on the back side relative to your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "21e0e93d-b97e-40a1-83bf-9267448231fd", + "video_length": "00:27:25", + "qid": "21e0e93d-b97e-40a1-83bf-9267448231fd_10", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the checkout counter located for me now?", + "answer": "Right side.", + "question_chinese": "现在收银台在我的什么位置？", + "answer_chinese": "右方", + "evidence": [ + { + "description": "When I was in the aisle, the checkout counter was on my left, I now turn right into the right area of the convenience store", + "timestamps": { + "start": "00:00:29", + "end": "00:00:35" + } + }, + { + "description": "Entering the right side of the convenience store is the fruit and vegetable area", + "timestamps": { + "start": "00:00:37", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:14:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "Left side.", + "C": "Front side.", + "D": "It's on the back side behind you, rather than positioned to your right as you face forward.", + "E": "It's located straight ahead in front of you, not on the right side from your current viewpoint." + }, + "answer_label": "A", + "answer_text": "Right side.", + "q_id": "21e0e93d-b97e-40a1-83bf-9267448231fd_10", + "correct_answer": "Right side.", + "original_options": { + "A": "Right side.", + "B": "Left side.", + "C": "Front side.", + "D": "It's on the back side behind you, rather than positioned to your right as you face forward.", + "E": "It's located straight ahead in front of you, not on the right side from your current viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + } +] diff --git a/QAs/24ef2f0c-9301-4158-84af-c1d5b7c6ed8a-mcq.json b/QAs/24ef2f0c-9301-4158-84af-c1d5b7c6ed8a-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8e2400b2dc3cdec7efab3b8cf84e10512651b18b --- /dev/null +++ b/QAs/24ef2f0c-9301-4158-84af-c1d5b7c6ed8a-mcq.json @@ -0,0 +1,1473 @@ +[ + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_5_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take after coming out of the bathroom?", + "answer": "So far, you walked out of the bathroom and took one step to the right.", + "question_chinese": "我从浴室出来经历了什么路径？", + "answer_chinese": "到目前为止，你从浴室出门，向右走了一步。", + "evidence": [ + { + "description": "I went out of the bathroom and took a step to the right.", + "timestamps": { + "start": "00:08:52", + "end": "00:08:55" + } + } + ], + "question_timestamps": "00:08:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took one step left.", + "B": "You took one step forward.", + "C": "So far, you walked out of the bathroom and took one step behind you instead of to the right.", + "D": "So far, you walked out of the bathroom and took one step to the left, not to the right.", + "E": "So far, you walked out of the bathroom and took one step to the right." + }, + "answer_label": "E", + "answer_text": "So far, you walked out of the bathroom and took one step to the right.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_5_T1", + "correct_answer": "So far, you walked out of the bathroom and took one step to the right.", + "original_options": { + "A": "You took one step left.", + "B": "You took one step forward.", + "C": "So far, you walked out of the bathroom and took one step behind you instead of to the right.", + "D": "So far, you walked out of the bathroom and took one step to the left, not to the right.", + "E": "So far, you walked out of the bathroom and took one step to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_5_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take after coming out of the bathroom?", + "answer": "You walked out of the bathroom, first took one step to the right, then went straight left to return to the bedroom.", + "question_chinese": "我从浴室出来经历了什么路径？", + "answer_chinese": "你从浴室出门，先向右走一步，再向左直走回到卧室。", + "evidence": [ + { + "description": "I went out of the bathroom and took a step to the right.", + "timestamps": { + "start": "00:08:52", + "end": "00:08:55" + } + }, + { + "description": "I turned left and walked straight back to the bedroom.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:24:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked out of the bathroom, first took one step to the right, then went straight left to return to the bedroom.", + "B": "You stepped left, then went straight.", + "C": "You stepped forward, then went left.", + "D": "You walked out of the bathroom, first took one step to the left, then went straight left to return to the bedroom.", + "E": "You walked out of the bathroom, first took one step to the right, then went straight right instead of straight left." + }, + "answer_label": "A", + "answer_text": "You walked out of the bathroom, first took one step to the right, then went straight left to return to the bedroom.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_5_T2", + "correct_answer": "You walked out of the bathroom, first took one step to the right, then went straight left to return to the bedroom.", + "original_options": { + "A": "You walked out of the bathroom, first took one step to the right, then went straight left to return to the bedroom.", + "B": "You stepped left, then went straight.", + "C": "You stepped forward, then went left.", + "D": "You walked out of the bathroom, first took one step to the left, then went straight left to return to the bedroom.", + "E": "You walked out of the bathroom, first took one step to the right, then went straight right instead of straight left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 21 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_10", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from the bathroom to the bedroom?", + "answer": "You went out of the bathroom, took a step to the right, then turned left and walked straight back to the bedroom.", + "question_chinese": "我从浴室回到卧室经历了什么路径？", + "answer_chinese": "你从浴室出门，先向右走一步，再向左直走回到卧室。", + "evidence": [ + { + "description": "I went out of the bathroom and took a step to the right.", + "timestamps": { + "start": "00:08:52", + "end": "00:08:55" + } + }, + { + "description": "I turned left and walked straight back to the bedroom.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:14:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Step left, then turn left.", + "B": "You went out of the bathroom, took a step to the right, then turned left and walked straight back to the bedroom.", + "C": "Step right, then turn right.", + "D": "You went out of the bathroom, took a step to the left, then turned right and walked straight back to the bedroom.", + "E": "You went out of the bathroom, took a step to the right, then turned right and walked straight back to the bedroom." + }, + "answer_label": "B", + "answer_text": "You went out of the bathroom, took a step to the right, then turned left and walked straight back to the bedroom.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_10", + "correct_answer": "You went out of the bathroom, took a step to the right, then turned left and walked straight back to the bedroom.", + "original_options": { + "A": "Step left, then turn left.", + "B": "You went out of the bathroom, took a step to the right, then turned left and walked straight back to the bedroom.", + "C": "Step right, then turn right.", + "D": "You went out of the bathroom, took a step to the left, then turned right and walked straight back to the bedroom.", + "E": "You went out of the bathroom, took a step to the right, then turned right and walked straight back to the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 22, + 22 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_11", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from the bathroom to the bedroom?", + "answer": "You went out of the bathroom,took a step to the right,then turned left and walked straight back to the bedroom.", + "question_chinese": "我从浴室回到卧室经历了什么路径？", + "answer_chinese": "你从浴室出门，先向右走一步，再向左直走回到卧室。", + "evidence": [ + { + "description": "I went out of the bathroom and took a step to the right.", + "timestamps": { + "start": "00:08:52", + "end": "00:08:55" + } + }, + { + "description": "I turned left and walked straight back to the bedroom.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:24:44", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Step left, then turn right.", + "B": "Step forward, then turn left.", + "C": "You went out of the bathroom, took a step to the right, then turned right and walked straight back to the bedroom.", + "D": "You went out of the bathroom,took a step to the right,then turned left and walked straight back to the bedroom.", + "E": "You went out of the bathroom, took a step to the left, then turned left and walked straight back to the bedroom." + }, + "answer_label": "D", + "answer_text": "You went out of the bathroom,took a step to the right,then turned left and walked straight back to the bedroom.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_11", + "correct_answer": "You went out of the bathroom,took a step to the right,then turned left and walked straight back to the bedroom.", + "original_options": { + "A": "Step left, then turn right.", + "B": "Step forward, then turn left.", + "C": "You went out of the bathroom, took a step to the right, then turned right and walked straight back to the bedroom.", + "D": "You went out of the bathroom,took a step to the right,then turned left and walked straight back to the bedroom.", + "E": "You went out of the bathroom, took a step to the left, then turned left and walked straight back to the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 22, + 22 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_2_T2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I just clipped the card onto the mirror frame—how far is the card from me now?", + "answer": "The card is about 40 cm away from you right now.", + "question_chinese": "我刚才把卡片卡在了镜子边框上，卡片现在离我有多远？", + "answer_chinese": "卡片现在离你大概40厘米。", + "evidence": [ + { + "description": "The card is about 40cm away from the rack where clothes are hanging.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:01:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "About 38 cm away.", + "B": "The card is about 40 cm away from you right now.", + "C": "About 42 cm away.", + "D": "The card is about 39 cm away from you right now, slightly closer than forty centimeters.", + "E": "The card is about 41 cm away from you right now, a bit farther than forty centimeters." + }, + "answer_label": "B", + "answer_text": "The card is about 40 cm away from you right now.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_2_T2", + "correct_answer": "The card is about 40 cm away from you right now.", + "original_options": { + "A": "About 38 cm away.", + "B": "The card is about 40 cm away from you right now.", + "C": "About 42 cm away.", + "D": "The card is about 39 cm away from you right now, slightly closer than forty centimeters.", + "E": "The card is about 41 cm away from you right now, a bit farther than forty centimeters." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_2_T1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I just clipped the card onto the mirror frame—how far is the card from me now?", + "answer": "The card is about 200 cm away from you right now.", + "question_chinese": "我刚才把卡片卡在了镜子边框上，卡片现在离我有多远？", + "answer_chinese": "卡片现在离你大概200厘米。", + "evidence": [ + { + "description": "The card is about 40cm away from the rack where clothes are hanging.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:24" + } + }, + { + "description": "Facing the mirror, you turn around and take three steps back.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:43" + } + } + ], + "question_timestamps": "00:01:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The card is about 200 cm away from you right now.", + "B": "About 198 cm away.", + "C": "About 202 cm away.", + "D": "The card is about 199 cm away from you right now, just under two meters from where you are.", + "E": "The card is about 201 cm away from you right now, just over two meters from your current position." + }, + "answer_label": "A", + "answer_text": "The card is about 200 cm away from you right now.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_2_T1", + "correct_answer": "The card is about 200 cm away from you right now.", + "original_options": { + "A": "The card is about 200 cm away from you right now.", + "B": "About 198 cm away.", + "C": "About 202 cm away.", + "D": "The card is about 199 cm away from you right now, just under two meters from where you are.", + "E": "The card is about 201 cm away from you right now, just over two meters from your current position." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 19 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_4", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I just stuck the card on the frame of the mirror. How far is the card from the wardrobe now?", + "answer": "The card is now about 40cm away from the rack where clothes are hanging.", + "question_chinese": "我刚才把卡片卡在了镜子边框上，卡片现在离衣柜有多远？", + "answer_chinese": "卡片现在离挂着衣服的架子大概40cm。", + "evidence": [ + { + "description": "The card is about 40cm away from the rack where clothes are hanging.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:08:59", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "About 30cm from the wardrobe.", + "B": "Roughly 50cm from the wardrobe.", + "C": "The card sits around 25cm away from the wardrobe where the clothes are hanging.", + "D": "It’s approximately 60cm from the wardrobe, leaving a wider gap than before.", + "E": "The card is now about 40cm away from the rack where clothes are hanging." + }, + "answer_label": "E", + "answer_text": "The card is now about 40cm away from the rack where clothes are hanging.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_4", + "correct_answer": "The card is now about 40cm away from the rack where clothes are hanging.", + "original_options": { + "A": "About 30cm from the wardrobe.", + "B": "Roughly 50cm from the wardrobe.", + "C": "The card sits around 25cm away from the wardrobe where the clothes are hanging.", + "D": "It’s approximately 60cm from the wardrobe, leaving a wider gap than before.", + "E": "The card is now about 40cm away from the rack where clothes are hanging." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_16", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I just stuck the card on the frame of the mirror. How far is the card from the wardrobe now?", + "answer": "The card is now about 40cm away from the rack where clothes are hanging.", + "question_chinese": "我刚才把卡片卡在了镜子边框上，卡片现在离衣柜有多远？", + "answer_chinese": "卡片现在离挂着衣服的架子大概40cm。", + "evidence": [ + { + "description": "The card is about 40cm away from the rack where clothes are hanging.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:14:49", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "The card is now about 40cm away from the rack where clothes are hanging.", + "B": "About 20cm from the wardrobe.", + "C": "Around 55cm from the wardrobe.", + "D": "The card is now roughly 30cm away from the wardrobe where clothes are hanging.", + "E": "It’s closer to 65cm from the wardrobe, not at that mid-range distance." + }, + "answer_label": "A", + "answer_text": "The card is now about 40cm away from the rack where clothes are hanging.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_16", + "correct_answer": "The card is now about 40cm away from the rack where clothes are hanging.", + "original_options": { + "A": "The card is now about 40cm away from the rack where clothes are hanging.", + "B": "About 20cm from the wardrobe.", + "C": "Around 55cm from the wardrobe.", + "D": "The card is now roughly 30cm away from the wardrobe where clothes are hanging.", + "E": "It’s closer to 65cm from the wardrobe, not at that mid-range distance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_6_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I just put the red cup on the countertop—how far is the red cup from the green kettle now?", + "answer": "You just placed the red cup on the countertop, and it’s about 15 cm from the green kettle.", + "question_chinese": "我刚才把红色的杯子放在台面上，现在红色杯子离绿色水壶有多远？", + "answer_chinese": "你刚才把红色的杯子放在台面上，红色水杯离绿色水壶大概15厘米。", + "evidence": [ + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:22:37", + "end": "00:23:30" + } + }, + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:24:14", + "end": "00:24:15" + } + }, + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:24:47", + "end": "00:24:49" + } + } + ], + "question_timestamps": "00:24:55", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "About 13 cm away.", + "B": "Roughly 17 cm away.", + "C": "You just placed the red cup on the countertop, and it’s about 12 cm from the green kettle.", + "D": "You set the red cup down on the countertop, leaving it around 18 cm from the green kettle.", + "E": "You just placed the red cup on the countertop, and it’s about 15 cm from the green kettle." + }, + "answer_label": "E", + "answer_text": "You just placed the red cup on the countertop, and it’s about 15 cm from the green kettle.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_6_T1", + "correct_answer": "You just placed the red cup on the countertop, and it’s about 15 cm from the green kettle.", + "original_options": { + "A": "About 13 cm away.", + "B": "Roughly 17 cm away.", + "C": "You just placed the red cup on the countertop, and it’s about 12 cm from the green kettle.", + "D": "You set the red cup down on the countertop, leaving it around 18 cm from the green kettle.", + "E": "You just placed the red cup on the countertop, and it’s about 15 cm from the green kettle." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 18 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_6_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I just put the red cup on the countertop—how far is the red cup from the green kettle now?", + "answer": "You didn’t move the cup. The red cup is still about 15 cm from the green kettle.", + "question_chinese": "我刚才把红色的杯子放在台面上，现在红色杯子离绿色水壶有多远？", + "answer_chinese": "你没有移动过杯子，现在红色水杯离绿色水壶依旧大概15厘米。", + "evidence": [ + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:22:37", + "end": "00:23:30" + } + }, + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:24:14", + "end": "00:24:15" + } + }, + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:24:47", + "end": "00:24:49" + } + } + ], + "question_timestamps": "00:29:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You didn’t move the cup. The red cup is still about 15 cm from the green kettle.", + "B": "It’s still about 13 cm away.", + "C": "It’s still about 17 cm away.", + "D": "You didn’t move the cup, so the red cup remains about 14 cm from the green kettle.", + "E": "Since you never moved it, the red cup is still roughly 16 cm from the green kettle." + }, + "answer_label": "A", + "answer_text": "You didn’t move the cup. The red cup is still about 15 cm from the green kettle.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_6_T2", + "correct_answer": "You didn’t move the cup. The red cup is still about 15 cm from the green kettle.", + "original_options": { + "A": "You didn’t move the cup. The red cup is still about 15 cm from the green kettle.", + "B": "It’s still about 13 cm away.", + "C": "It’s still about 17 cm away.", + "D": "You didn’t move the cup, so the red cup remains about 14 cm from the green kettle.", + "E": "Since you never moved it, the red cup is still roughly 16 cm from the green kettle." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 17 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_12", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I placed the red cup on the table just now, how far was the red cup from the green kettle?", + "answer": "When you just placed the red cup on the table, the red water cup was about 15cm away from the green kettle.", + "question_chinese": "我刚才把红色的杯子放在台面上的时候，红色杯子离绿色水壶有多远？", + "answer_chinese": "你刚才把红色的杯子放在台面上的时候，红色水杯离绿色水壶大概15cm。", + "evidence": [ + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:22:37", + "end": "00:23:30" + } + }, + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:24:14", + "end": "00:24:15" + } + }, + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:24:47", + "end": "00:24:49" + } + } + ], + "question_timestamps": "00:27:19", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "About 10cm away.", + "B": "When you just placed the red cup on the table, the red water cup was about 15cm away from the green kettle.", + "C": "About 25cm away.", + "D": "When you placed the red cup on the table, it was about 20cm away from the green kettle.", + "E": "At the moment you set it on the table, the red water cup sat roughly 35cm from the green kettle." + }, + "answer_label": "B", + "answer_text": "When you just placed the red cup on the table, the red water cup was about 15cm away from the green kettle.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_12", + "correct_answer": "When you just placed the red cup on the table, the red water cup was about 15cm away from the green kettle.", + "original_options": { + "A": "About 10cm away.", + "B": "When you just placed the red cup on the table, the red water cup was about 15cm away from the green kettle.", + "C": "About 25cm away.", + "D": "When you placed the red cup on the table, it was about 20cm away from the green kettle.", + "E": "At the moment you set it on the table, the red water cup sat roughly 35cm from the green kettle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 20 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_17", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I placed the red cup on the table just now, how far was the red cup from the green kettle?", + "answer": "When you just placed the red cup on the table, the red water cup was about 15cm away from the green kettle.", + "question_chinese": "我刚才把红色的杯子放在台面上的时候，红色杯子离绿色水壶有多远？", + "answer_chinese": "你刚才把红色的杯子放在台面上的时候，红色水杯离绿色水壶大概15cm。", + "evidence": [ + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:22:37", + "end": "00:23:30" + } + }, + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:24:14", + "end": "00:24:15" + } + }, + { + "description": "The red water cup was about 15cm away from the green kettle.", + "timestamps": { + "start": "00:24:47", + "end": "00:24:49" + } + } + ], + "question_timestamps": "00:29:54", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "About 10cm away.", + "B": "When you just placed the red cup on the table, the red water cup was about 15cm away from the green kettle.", + "C": "About 30cm away.", + "D": "When you just placed the red cup on the table, it was about 20cm away from the green kettle.", + "E": "Right after you set it down on the table, the red water cup was roughly 25cm from the green kettle." + }, + "answer_label": "B", + "answer_text": "When you just placed the red cup on the table, the red water cup was about 15cm away from the green kettle.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_17", + "correct_answer": "When you just placed the red cup on the table, the red water cup was about 15cm away from the green kettle.", + "original_options": { + "A": "About 10cm away.", + "B": "When you just placed the red cup on the table, the red water cup was about 15cm away from the green kettle.", + "C": "About 30cm away.", + "D": "When you just placed the red cup on the table, it was about 20cm away from the green kettle.", + "E": "Right after you set it down on the table, the red water cup was roughly 25cm from the green kettle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 19, + 20 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_7", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I finished washing things just now, is the towel I used to dry them on my left or right?", + "answer": "The towel you used to dry your things was on your left.", + "question_chinese": "我刚才洗完东西后，擦干东西用的毛巾是在我的左边还是右边？", + "answer_chinese": "你刚才洗完东西后，擦干东西用的毛巾在你的左边。", + "evidence": [ + { + "description": "The towel I used to dry my laundry was on my left.", + "timestamps": { + "start": "00:07:27", + "end": "00:08:40" + } + } + ], + "question_timestamps": "00:19:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The towel you used to dry your things was on your right.", + "B": "The towel you used to dry your things was on your left." + }, + "answer_label": "B", + "answer_text": "The towel you used to dry your things was on your left.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_7", + "correct_answer": "The towel you used to dry your things was on your left.", + "original_options": { + "A": "The towel you used to dry your things was on your right.", + "B": "The towel you used to dry your things was on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_8", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I finished washing things just now, is the towel I used to dry them on my left or right?", + "answer": "The towel you used to dry your things was on your left.", + "question_chinese": "我刚才洗完东西后，擦干东西用的毛巾是在我的左边还是右边？", + "answer_chinese": "你刚才洗完东西后，擦干东西用的毛巾在你的左边。", + "evidence": [ + { + "description": "The towel I used to dry my laundry was on my left.", + "timestamps": { + "start": "00:07:27", + "end": "00:08:40" + } + } + ], + "question_timestamps": "00:24:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The towel you used to dry your things was on your right.", + "B": "The towel you used to dry your things was on your left." + }, + "answer_label": "B", + "answer_text": "The towel you used to dry your things was on your left.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_8", + "correct_answer": "The towel you used to dry your things was on your left.", + "original_options": { + "A": "The towel you used to dry your things was on your right.", + "B": "The towel you used to dry your things was on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_8_T2", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the black cup in the kitchen?", + "answer": "Yes. When you were washing dishes in the kitchen, you could reach the black cup on the windowsill above the sink.", + "question_chinese": "我可以拿到厨房的黑色杯子吗？", + "answer_chinese": "可以，你在厨房洗碗时，能拿到窗台上的黑色杯子，它在洗手台上方的窗台上。", + "evidence": [ + { + "description": "The black cup was located on the windowsill above the sink.", + "timestamps": { + "start": "00:14:10", + "end": "00:14:24" + } + }, + { + "description": "The black cup was located on the windowsill above the sink.", + "timestamps": { + "start": "00:14:47", + "end": "00:14:58" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. When you were washing dishes in the kitchen, you couldn’t reach the black cup on the windowsill above the sink.", + "B": "Yes. When you were washing dishes in the kitchen, you could reach the black cup on the windowsill above the sink." + }, + "answer_label": "B", + "answer_text": "Yes. When you were washing dishes in the kitchen, you could reach the black cup on the windowsill above the sink.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_8_T2", + "correct_answer": "Yes. When you were washing dishes in the kitchen, you could reach the black cup on the windowsill above the sink.", + "original_options": { + "A": "No. When you were washing dishes in the kitchen, you couldn’t reach the black cup on the windowsill above the sink.", + "B": "Yes. When you were washing dishes in the kitchen, you could reach the black cup on the windowsill above the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 22 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_8_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the black cup in the kitchen?", + "answer": "No. The black cup is on the windowsill above the sink.", + "question_chinese": "我可以拿到厨房的黑色杯子吗？", + "answer_chinese": "不可以，黑色杯子在洗手台上方的窗台上。", + "evidence": [ + { + "description": "The black cup was located on the windowsill above the sink.", + "timestamps": { + "start": "00:14:10", + "end": "00:14:24" + } + }, + { + "description": "The black cup was located on the windowsill above the sink.", + "timestamps": { + "start": "00:14:47", + "end": "00:14:58" + } + }, + { + "description": "You took three steps to the right from the sink.", + "timestamps": { + "start": "00:14:58", + "end": "00:15:05" + } + } + ], + "question_timestamps": "00:15:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. The black cup is on the windowsill above the sink.", + "B": "Yes. Even though it’s on the windowsill above the sink, you can still reach the black cup from where you stand." + }, + "answer_label": "A", + "answer_text": "No. The black cup is on the windowsill above the sink.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_8_T1", + "correct_answer": "No. The black cup is on the windowsill above the sink.", + "original_options": { + "A": "No. The black cup is on the windowsill above the sink.", + "B": "Yes. Even though it’s on the windowsill above the sink, you can still reach the black cup from where you stand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 22 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_4_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the towel I use to dry things to my left or to my right?", + "answer": "After you finished washing things just now, the towel you use to dry items is to your left.", + "question_chinese": "我擦干东西用的毛巾是在我的左边还是右边？", + "answer_chinese": "你刚才洗完东西后，擦干东西用的毛巾在你的左边。", + "evidence": [ + { + "description": "The towel I used to dry my laundry was on my left.", + "timestamps": { + "start": "00:07:27", + "end": "00:08:35" + } + } + ], + "question_timestamps": "00:08:42", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "After you finished washing things just now, the towel you use to dry items is to your left.", + "B": "After you finished washing things just now, the towel you use to dry items is to your right." + }, + "answer_label": "A", + "answer_text": "After you finished washing things just now, the towel you use to dry items is to your left.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_4_T2", + "correct_answer": "After you finished washing things just now, the towel you use to dry items is to your left.", + "original_options": { + "A": "After you finished washing things just now, the towel you use to dry items is to your left.", + "B": "After you finished washing things just now, the towel you use to dry items is to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_4_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the towel I use to dry things to my left or to my right?", + "answer": "The towel you use to dry items is to your right right now.", + "question_chinese": "我擦干东西用的毛巾是在我的左边还是右边？", + "answer_chinese": "你用来擦干东西用的毛巾现在在你的右边。", + "evidence": [ + { + "description": "The towel I used to dry my laundry was on my left.", + "timestamps": { + "start": "00:07:27", + "end": "00:08:40" + } + } + ], + "question_timestamps": "00:08:54", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The towel you use to dry items is to your right right now.", + "B": "The towel you use to dry items is to your left right now." + }, + "answer_label": "A", + "answer_text": "The towel you use to dry items is to your right right now.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_4_T1", + "correct_answer": "The towel you use to dry items is to your right right now.", + "original_options": { + "A": "The towel you use to dry items is to your right right now.", + "B": "The towel you use to dry items is to your left right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_5", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many things have I put on the sink so far?", + "answer": "So far, you have put three things on the sink.", + "question_chinese": "目前为止，我放了几个东西在洗手台上？", + "answer_chinese": "目前为止，你放了3个东西在洗手台上。", + "evidence": [ + { + "description": "I put one things on the sink.", + "timestamps": { + "start": "00:06:23", + "end": "00:06:26" + } + }, + { + "description": "I put two more things on the sink.", + "timestamps": { + "start": "00:06:27", + "end": "00:06:31" + } + } + ], + "question_timestamps": "00:06:33", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have put two things on the sink.", + "B": "You have put four things on the sink.", + "C": "So far, you have put five things on the sink in total.", + "D": "Up to now, you have placed six things on the sink altogether.", + "E": "So far, you have put three things on the sink." + }, + "answer_label": "E", + "answer_text": "So far, you have put three things on the sink.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_5", + "correct_answer": "So far, you have put three things on the sink.", + "original_options": { + "A": "You have put two things on the sink.", + "B": "You have put four things on the sink.", + "C": "So far, you have put five things on the sink in total.", + "D": "Up to now, you have placed six things on the sink altogether.", + "E": "So far, you have put three things on the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 12, + 12 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_6", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many things have I put on the sink so far?", + "answer": "So far, you have put five things on the sink.", + "question_chinese": "目前为止，我放了几个东西在洗手台上？", + "answer_chinese": "目前为止，你放了5个东西在洗手台上。", + "evidence": [ + { + "description": "I put one thing on the sink.", + "timestamps": { + "start": "00:06:23", + "end": "00:06:26" + } + }, + { + "description": "I put two more things on the sink.", + "timestamps": { + "start": "00:06:27", + "end": "00:06:31" + } + }, + { + "description": "I put two more things on the sink.", + "timestamps": { + "start": "00:06:31", + "end": "00:06:40" + } + } + ], + "question_timestamps": "00:06:43", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have put four things on the sink.", + "B": "You have put six things on the sink.", + "C": "So far, you have put seven things on the sink in total.", + "D": "So far, you have put five things on the sink.", + "E": "Up to now, you have placed eight things on the sink altogether." + }, + "answer_label": "D", + "answer_text": "So far, you have put five things on the sink.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_6", + "correct_answer": "So far, you have put five things on the sink.", + "original_options": { + "A": "You have put four things on the sink.", + "B": "You have put six things on the sink.", + "C": "So far, you have put seven things on the sink in total.", + "D": "So far, you have put five things on the sink.", + "E": "Up to now, you have placed eight things on the sink altogether." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 12, + 12 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_14", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls are there on the countertop next to the sink so far?", + "answer": "So far, there are two bowls on the countertop next to the sink.", + "question_chinese": "目前为止，洗手台旁边的台面上有几个碗？", + "answer_chinese": "目前为止，洗手台旁边的台面上有2个碗。", + "evidence": [ + { + "description": "I took out the first bowl and placed it on the table.", + "timestamps": { + "start": "00:21:47", + "end": "00:21:50" + } + }, + { + "description": "I took out the second bowl and placed it on the table.", + "timestamps": { + "start": "00:21:54", + "end": "00:21:56" + } + } + ], + "question_timestamps": "00:21:58", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "So far, there are two bowls on the countertop next to the sink.", + "B": "There is one bowl there.", + "C": "There are three bowls there.", + "D": "So far, there are four bowls on the countertop next to the sink.", + "E": "Up to now, there are five bowls on the countertop beside the sink." + }, + "answer_label": "A", + "answer_text": "So far, there are two bowls on the countertop next to the sink.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_14", + "correct_answer": "So far, there are two bowls on the countertop next to the sink.", + "original_options": { + "A": "So far, there are two bowls on the countertop next to the sink.", + "B": "There is one bowl there.", + "C": "There are three bowls there.", + "D": "So far, there are four bowls on the countertop next to the sink.", + "E": "Up to now, there are five bowls on the countertop beside the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_13", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls are there on the countertop next to the sink so far?", + "answer": "So far, there are three bowls on the countertop next to the sink.", + "question_chinese": "目前为止，洗手台旁边的台面上有几个碗？", + "answer_chinese": "目前为止，洗手台旁边的台面上有3个碗。", + "evidence": [ + { + "description": "I took out the first bowl and placed it on the table.", + "timestamps": { + "start": "00:21:47", + "end": "00:21:50" + } + }, + { + "description": "I took out the second bowl and placed it on the table.", + "timestamps": { + "start": "00:21:54", + "end": "00:21:56" + } + }, + { + "description": "I took out the third bowl and placed it on the table.", + "timestamps": { + "start": "00:21:57", + "end": "00:22:00" + } + } + ], + "question_timestamps": "00:22:04", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are two bowls there.", + "B": "There are four bowls there.", + "C": "So far, there are three bowls on the countertop next to the sink.", + "D": "So far, there are five bowls on the countertop next to the sink.", + "E": "Up to now, there are six bowls on the countertop beside the sink." + }, + "answer_label": "C", + "answer_text": "So far, there are three bowls on the countertop next to the sink.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_13", + "correct_answer": "So far, there are three bowls on the countertop next to the sink.", + "original_options": { + "A": "There are two bowls there.", + "B": "There are four bowls there.", + "C": "So far, there are three bowls on the countertop next to the sink.", + "D": "So far, there are five bowls on the countertop next to the sink.", + "E": "Up to now, there are six bowls on the countertop beside the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_9", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the empty photo frame in the bedroom?", + "answer": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "question_chinese": "空相框在卧室的哪个位置？", + "answer_chinese": "空相框在挂着衣服的架子上面，两幅画的中间。", + "evidence": [ + { + "description": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:15:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the nightstand beside the bed.", + "B": "On the dresser next to the mirror.", + "C": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "D": "It is on the desk in the corner, sitting beside a small stack of books.", + "E": "You left it on the wall above the bed, positioned between two shelves." + }, + "answer_label": "C", + "answer_text": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_9", + "correct_answer": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "original_options": { + "A": "On the nightstand beside the bed.", + "B": "On the dresser next to the mirror.", + "C": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "D": "It is on the desk in the corner, sitting beside a small stack of books.", + "E": "You left it on the wall above the bed, positioned between two shelves." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 13 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_3", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the empty photo frame in the bedroom?", + "answer": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "question_chinese": "空相框在卧室的哪个位置?", + "answer_chinese": "空相框在挂着衣服的架子上面，两幅画的中间。", + "evidence": [ + { + "description": "The empty photo frame is on the shelf with clothes hanging, in the middle of two\npaintings.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:22:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the desk beside a lamp.", + "B": "On the floor near the closet.", + "C": "It is on the nightstand by the bed, tucked under a small clock.", + "D": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "E": "The frame is on the wall above the trophy shelves, placed right beside a mirror." + }, + "answer_label": "D", + "answer_text": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_3", + "correct_answer": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "original_options": { + "A": "On the desk beside a lamp.", + "B": "On the floor near the closet.", + "C": "It is on the nightstand by the bed, tucked under a small clock.", + "D": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "E": "The frame is on the wall above the trophy shelves, placed right beside a mirror." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 15 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_1_T1", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the empty photo frame relative to me?", + "answer": "The empty picture frame is on the shelf in front of you, on the rack where clothes are hanging, between the two paintings.", + "question_chinese": "空相框在相对于我的哪个位置?", + "answer_chinese": "空相框在你的前方的挂着衣服的架子上面，两幅画的中间。", + "evidence": [ + { + "description": "The empty photo frame is on the shelf with clothes hanging, in the middle of two paintings.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:01:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is on the shelf behind you.", + "B": "The empty picture frame is on the shelf in front of you, on the rack where clothes are hanging, between the two paintings.", + "C": "It is on the shelf to your left.", + "D": "The empty picture frame is on a shelf behind you on the clothes rack, between the two paintings.", + "E": "The empty picture frame is on the shelf to your right on the rack with hanging clothes, between the paintings." + }, + "answer_label": "B", + "answer_text": "The empty picture frame is on the shelf in front of you, on the rack where clothes are hanging, between the two paintings.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_1_T1", + "correct_answer": "The empty picture frame is on the shelf in front of you, on the rack where clothes are hanging, between the two paintings.", + "original_options": { + "A": "It is on the shelf behind you.", + "B": "The empty picture frame is on the shelf in front of you, on the rack where clothes are hanging, between the two paintings.", + "C": "It is on the shelf to your left.", + "D": "The empty picture frame is on a shelf behind you on the clothes rack, between the two paintings.", + "E": "The empty picture frame is on the shelf to your right on the rack with hanging clothes, between the paintings." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 20 + ] + }, + { + "video_uid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a", + "video_length": "00:30:00", + "qid": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_1_T2", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the empty photo frame relative to me?", + "answer": "The empty picture frame is on the shelf behind you, on the rack where clothes are hanging.", + "question_chinese": "空相框在相对于我的哪个位置?", + "answer_chinese": "空相框在你后方的挂着衣服的架子上面。", + "evidence": [ + { + "description": "The empty photo frame is on the shelf with clothes hanging, in the middle of two\npaintings.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:01:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you on the shelf.", + "B": "The empty picture frame is on the shelf behind you, on the rack where clothes are hanging.", + "C": "To your left on the shelf.", + "D": "The empty picture frame is on the shelf to your right, on the rack where clothes are hanging.", + "E": "It is on the shelf above you, positioned on the rack where clothes are hanging." + }, + "answer_label": "B", + "answer_text": "The empty picture frame is on the shelf behind you, on the rack where clothes are hanging.", + "q_id": "24ef2f0c-9301-4158-84af-c1d5b7c6ed8a_1_T2", + "correct_answer": "The empty picture frame is on the shelf behind you, on the rack where clothes are hanging.", + "original_options": { + "A": "In front of you on the shelf.", + "B": "The empty picture frame is on the shelf behind you, on the rack where clothes are hanging.", + "C": "To your left on the shelf.", + "D": "The empty picture frame is on the shelf to your right, on the rack where clothes are hanging.", + "E": "It is on the shelf above you, positioned on the rack where clothes are hanging." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 15 + ] + } +] diff --git a/QAs/24fa7505-bafd-418d-8bed-62e614085fa9-mcq.json b/QAs/24fa7505-bafd-418d-8bed-62e614085fa9-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c0261f7ca08d7770b7ecb6951dae81e42b03a9a9 --- /dev/null +++ b/QAs/24fa7505-bafd-418d-8bed-62e614085fa9-mcq.json @@ -0,0 +1,1386 @@ +[ + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_15", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the warehouse?", + "answer": "The warehouse is on the left of the main gate.", + "question_chinese": "仓库在哪里？", + "answer_chinese": "仓库在大门的左边。", + "evidence": [ + { + "description": "You walked out of the gate.", + "timestamps": { + "start": "00:54:55", + "end": "00:54:57" + } + }, + { + "description": "The door of the warehouse.", + "timestamps": { + "start": "00:55:16", + "end": "00:55:18" + } + }, + { + "description": "The warehouse.", + "timestamps": { + "start": "00:55:22", + "end": "00:55:23" + } + } + ], + "question_timestamps": "00:58:56", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "On the right of the main gate.", + "B": "Behind the main gate.", + "C": "The warehouse is in front of the main building, facing the entrance area near the gate.", + "D": "The warehouse is on the left of the main gate.", + "E": "It is next to the main office, set back behind the gate along the same frontage." + }, + "answer_label": "D", + "answer_text": "The warehouse is on the left of the main gate.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_15", + "correct_answer": "The warehouse is on the left of the main gate.", + "original_options": { + "A": "On the right of the main gate.", + "B": "Behind the main gate.", + "C": "The warehouse is in front of the main building, facing the entrance area near the gate.", + "D": "The warehouse is on the left of the main gate.", + "E": "It is next to the main office, set back behind the gate along the same frontage." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 4, + 16, + 16 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_16", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the warehouse?", + "answer": "The warehouse is on the left of the main gate.", + "question_chinese": "仓库在哪里？", + "answer_chinese": "仓库在大门的左边。", + "evidence": [ + { + "description": "You walked out of the gate.", + "timestamps": { + "start": "00:54:55", + "end": "00:54:57" + } + }, + { + "description": "The door of the warehouse.", + "timestamps": { + "start": "00:55:16", + "end": "00:55:18" + } + }, + { + "description": "The warehouse.", + "timestamps": { + "start": "00:55:22", + "end": "00:55:23" + } + } + ], + "question_timestamps": "01:02:56", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Behind the main gate.", + "B": "In front of the main building.", + "C": "The warehouse is on the left of the main gate.", + "D": "The warehouse is next to the main office, positioned along the same side of the entrance area.", + "E": "It sits beside the loading dock, just off the approach to the main building." + }, + "answer_label": "C", + "answer_text": "The warehouse is on the left of the main gate.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_16", + "correct_answer": "The warehouse is on the left of the main gate.", + "original_options": { + "A": "Behind the main gate.", + "B": "In front of the main building.", + "C": "The warehouse is on the left of the main gate.", + "D": "The warehouse is next to the main office, positioned along the same side of the entrance area.", + "E": "It sits beside the loading dock, just off the approach to the main building." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 17, + 14 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_11", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the two lawn mowers I pushed just now?", + "answer": "You pushed the lawn mower straight ahead and then turned right into the warehouse.", + "question_chinese": "刚才我推着两个除草机去的运动轨迹是什么？", + "answer_chinese": "你推着除草机先直走然后右拐进了仓库。", + "evidence": [ + { + "description": "You pushed the lawn mower straight ahead.", + "timestamps": { + "start": "00:20:36", + "end": "00:20:45" + } + }, + { + "description": "You then turned right into the warehouse.", + "timestamps": { + "start": "00:20:45", + "end": "00:20:47" + } + } + ], + "question_timestamps": "00:27:24", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Straight ahead, then left into the warehouse.", + "B": "Straight ahead, then right into the parking lot.", + "C": "You pushed the lawn mower diagonally forward and then turned right as you entered the warehouse.", + "D": "You backed the lawn mower up first, then turned right and went into the warehouse.", + "E": "You pushed the lawn mower straight ahead and then turned right into the warehouse." + }, + "answer_label": "E", + "answer_text": "You pushed the lawn mower straight ahead and then turned right into the warehouse.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_11", + "correct_answer": "You pushed the lawn mower straight ahead and then turned right into the warehouse.", + "original_options": { + "A": "Straight ahead, then left into the warehouse.", + "B": "Straight ahead, then right into the parking lot.", + "C": "You pushed the lawn mower diagonally forward and then turned right as you entered the warehouse.", + "D": "You backed the lawn mower up first, then turned right and went into the warehouse.", + "E": "You pushed the lawn mower straight ahead and then turned right into the warehouse." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 15 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_12", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the two lawn mowers I pushed just now?", + "answer": "You pushed the lawn mower straight ahead and then turned right into the warehouse.", + "question_chinese": "刚才我推着两个除草机去的运动轨迹是什么？", + "answer_chinese": "你推着除草机先直走然后右拐进了仓库。", + "evidence": [ + { + "description": "You pushed the lawn mower straight ahead.", + "timestamps": { + "start": "00:20:36", + "end": "00:20:45" + } + }, + { + "description": "You then turned right into the warehouse.", + "timestamps": { + "start": "00:20:45", + "end": "00:20:47" + } + } + ], + "question_timestamps": "00:33:24", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You pushed the lawn mower straight ahead and then turned right into the warehouse.", + "B": "Straight ahead, then left into the garden.", + "C": "Straight ahead, then right into the parking lot.", + "D": "You pushed the lawn mower diagonally forward and then turned left as you went into the warehouse.", + "E": "You pushed the lawn mower straight ahead, then turned right toward the driveway instead of entering the warehouse." + }, + "answer_label": "A", + "answer_text": "You pushed the lawn mower straight ahead and then turned right into the warehouse.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_12", + "correct_answer": "You pushed the lawn mower straight ahead and then turned right into the warehouse.", + "original_options": { + "A": "You pushed the lawn mower straight ahead and then turned right into the warehouse.", + "B": "Straight ahead, then left into the garden.", + "C": "Straight ahead, then right into the parking lot.", + "D": "You pushed the lawn mower diagonally forward and then turned left as you went into the warehouse.", + "E": "You pushed the lawn mower straight ahead, then turned right toward the driveway instead of entering the warehouse." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 18 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_7_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From my point of view, where is the warehouse?", + "answer": "From your point of view, the warehouse is to the front-left of the two new machines. The warehouse is to your front-right.", + "question_chinese": "在我的视角，仓库在哪里？", + "answer_chinese": "在你的视角，仓库在两辆新机器的左前方。仓库在你的右前方。", + "evidence": [ + { + "description": "The warehouse is on the left front of the two new machines.", + "timestamps": { + "start": "00:55:10", + "end": "00:55:13" + } + } + ], + "question_timestamps": "00:55:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It's front-right of the machines.", + "B": "From your point of view, the warehouse is to the front-left of the two new machines. The warehouse is to your front-right.", + "C": "It's behind you.", + "D": "From your point of view, the warehouse is front-right of the two new machines and to your front-left.", + "E": "From your point of view, the warehouse is behind the two new machines, and it ends up directly behind you." + }, + "answer_label": "B", + "answer_text": "From your point of view, the warehouse is to the front-left of the two new machines. The warehouse is to your front-right.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_7_T2", + "correct_answer": "From your point of view, the warehouse is to the front-left of the two new machines. The warehouse is to your front-right.", + "original_options": { + "A": "It's front-right of the machines.", + "B": "From your point of view, the warehouse is to the front-left of the two new machines. The warehouse is to your front-right.", + "C": "It's behind you.", + "D": "From your point of view, the warehouse is front-right of the two new machines and to your front-left.", + "E": "From your point of view, the warehouse is behind the two new machines, and it ends up directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 18, + 20 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_7_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From my point of view, where is the warehouse?", + "answer": "From your point of view, the warehouse is to the left of the main gate. The warehouse is to your right.", + "question_chinese": "在我的视角，仓库在哪里？", + "answer_chinese": "在你的视角，仓库在大门的左边。仓库在你的右边。", + "evidence": [ + { + "description": "You walked out of the gate.", + "timestamps": { + "start": "00:54:55", + "end": "00:54:57" + } + }, + { + "description": "The door of the warehouse.", + "timestamps": { + "start": "00:55:16", + "end": "00:55:18" + } + }, + { + "description": "The warehouse.", + "timestamps": { + "start": "00:55:22", + "end": "00:55:23" + } + } + ], + "question_timestamps": "00:55:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From your point of view, the warehouse is to the left of the main gate. The warehouse is to your right.", + "B": "It's to your left.", + "C": "It's right of the main gate.", + "D": "From your point of view, the warehouse is to the right of the main gate and also to your left.", + "E": "From your point of view, the warehouse is behind the main gate, so it sits behind you as well." + }, + "answer_label": "A", + "answer_text": "From your point of view, the warehouse is to the left of the main gate. The warehouse is to your right.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_7_T1", + "correct_answer": "From your point of view, the warehouse is to the left of the main gate. The warehouse is to your right.", + "original_options": { + "A": "From your point of view, the warehouse is to the left of the main gate. The warehouse is to your right.", + "B": "It's to your left.", + "C": "It's right of the main gate.", + "D": "From your point of view, the warehouse is to the right of the main gate and also to your left.", + "E": "From your point of view, the warehouse is behind the main gate, so it sits behind you as well." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 20, + 19 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_17", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can?", + "answer": "The trash can is right in front of the gate.", + "question_chinese": "垃圾桶在哪里？", + "answer_chinese": "垃圾桶正对着大门。", + "evidence": [ + { + "description": "The trash can is right in front of the gate.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:07:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Just left of the gate.", + "B": "The trash can is right in front of the gate.", + "C": "Just behind the gate.", + "D": "The trash can is slightly to the right side of the gate, not directly in front of it.", + "E": "The trash can sits a few steps back from the gate, closer to the fence line than the entrance." + }, + "answer_label": "B", + "answer_text": "The trash can is right in front of the gate.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_17", + "correct_answer": "The trash can is right in front of the gate.", + "original_options": { + "A": "Just left of the gate.", + "B": "The trash can is right in front of the gate.", + "C": "Just behind the gate.", + "D": "The trash can is slightly to the right side of the gate, not directly in front of it.", + "E": "The trash can sits a few steps back from the gate, closer to the fence line than the entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 19 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_18", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can?", + "answer": "The trash can is right in front of the gate.", + "question_chinese": "垃圾桶在哪里？", + "answer_chinese": "垃圾桶正对着大门。", + "evidence": [ + { + "description": "The trash can is right in front of the gate.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:12:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To the left of the gate.", + "B": "Behind the gate.", + "C": "The trash can is positioned off to the right of the gate, not centered in front of it.", + "D": "The trash can is right in front of the gate.", + "E": "The trash can is a short distance past the gate area, tucked along the fence rather than at the entrance." + }, + "answer_label": "D", + "answer_text": "The trash can is right in front of the gate.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_18", + "correct_answer": "The trash can is right in front of the gate.", + "original_options": { + "A": "To the left of the gate.", + "B": "Behind the gate.", + "C": "The trash can is positioned off to the right of the gate, not centered in front of it.", + "D": "The trash can is right in front of the gate.", + "E": "The trash can is a short distance past the gate area, tucked along the fence rather than at the entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 3, + 18, + 20 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_13", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the electric drill on the workbench?", + "answer": "It's red and green.", + "question_chinese": "工作台上的电钻是什么颜色的？", + "answer_chinese": "是红色和绿色的。", + "evidence": [ + { + "description": "It's red and green.", + "timestamps": { + "start": "00:22:59", + "end": "00:23:00" + } + } + ], + "question_timestamps": "00:34:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's blue and yellow.", + "B": "It's black and orange.", + "C": "It's red and green.", + "D": "It's green and black, with the darker color on the handle and green on the body.", + "E": "It's red and blue, with a red body and blue accents around the grip and trigger." + }, + "answer_label": "C", + "answer_text": "It's red and green.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_13", + "correct_answer": "It's red and green.", + "original_options": { + "A": "It's blue and yellow.", + "B": "It's black and orange.", + "C": "It's red and green.", + "D": "It's green and black, with the darker color on the handle and green on the body.", + "E": "It's red and blue, with a red body and blue accents around the grip and trigger." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_14", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the electric drill on the workbench?", + "answer": "It's red and green.", + "question_chinese": "工作台上的电钻是什么颜色的？", + "answer_chinese": "是红色和绿色的。", + "evidence": [ + { + "description": "It's red and green.", + "timestamps": { + "start": "00:22:59", + "end": "00:23:00" + } + } + ], + "question_timestamps": "00:37:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's blue and orange.", + "B": "It's red and green.", + "C": "It's black and white.", + "D": "It's yellow and purple, with the bright yellow casing and purple detailing near the handle.", + "E": "It's gray and brown, with a gray body and a brown grip section on the handle." + }, + "answer_label": "B", + "answer_text": "It's red and green.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_14", + "correct_answer": "It's red and green.", + "original_options": { + "A": "It's blue and orange.", + "B": "It's red and green.", + "C": "It's black and white.", + "D": "It's yellow and purple, with the bright yellow casing and purple detailing near the handle.", + "E": "It's gray and brown, with a gray body and a brown grip section on the handle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where can I find a rag?", + "answer": "It can be found in the box on the shelf behind the workbench.", + "question_chinese": "我从哪里能找到抹布？", + "answer_chinese": "在工作台后面的架子上的箱子里可以找到。", + "evidence": [ + { + "description": "The workbench.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:26" + } + }, + { + "description": "The box.", + "timestamps": { + "start": "00:03:29", + "end": "00:03:31" + } + } + ], + "question_timestamps": "00:06:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In a bag beside the workbench.", + "B": "In a bin near the window.", + "C": "It can be found in a box on the shelf above the workbench, not on the shelf behind it.", + "D": "It can be found in the box on the shelf behind the workbench.", + "E": "It can be found in the box on the lower shelf in front of the workbench, instead of behind the workbench." + }, + "answer_label": "D", + "answer_text": "It can be found in the box on the shelf behind the workbench.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_1", + "correct_answer": "It can be found in the box on the shelf behind the workbench.", + "original_options": { + "A": "In a bag beside the workbench.", + "B": "In a bin near the window.", + "C": "It can be found in a box on the shelf above the workbench, not on the shelf behind it.", + "D": "It can be found in the box on the shelf behind the workbench.", + "E": "It can be found in the box on the lower shelf in front of the workbench, instead of behind the workbench." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 21 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where can I find a rag?", + "answer": "It can be found in the box on the shelf behind the workbench.", + "question_chinese": "我从哪里能找到抹布？", + "answer_chinese": "在工作台后面的架子上的箱子里可以找到。", + "evidence": [ + { + "description": "The workbench.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:26" + } + }, + { + "description": "The box.", + "timestamps": { + "start": "00:03:29", + "end": "00:03:31" + } + } + ], + "question_timestamps": "00:09:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In the basket near the window.", + "B": "It can be found in the box on the shelf behind the workbench.", + "C": "In the drawer above the desk.", + "D": "It can be found in a box on the shelf next to the workbench, not on the shelf behind it.", + "E": "It can be found in a box under the workbench area, rather than in a box on the shelf behind it." + }, + "answer_label": "B", + "answer_text": "It can be found in the box on the shelf behind the workbench.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_2", + "correct_answer": "It can be found in the box on the shelf behind the workbench.", + "original_options": { + "A": "In the basket near the window.", + "B": "It can be found in the box on the shelf behind the workbench.", + "C": "In the drawer above the desk.", + "D": "It can be found in a box on the shelf next to the workbench, not on the shelf behind it.", + "E": "It can be found in a box under the workbench area, rather than in a box on the shelf behind it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 21 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_5_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tires does the car I repaired have?", + "answer": "The car you repaired has two tires. The tires are below you.", + "question_chinese": "我修理的这辆车装有几个轮胎？", + "answer_chinese": "你修理的这辆车装有两个轮胎。轮胎在你的下面。", + "evidence": [ + { + "description": "The second one.", + "timestamps": { + "start": "00:15:39", + "end": "00:15:40" + } + }, + { + "description": "The first one.", + "timestamps": { + "start": "00:14:57", + "end": "00:15:00" + } + } + ], + "question_timestamps": "00:15:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The car you repaired has one tire.", + "B": "The car you repaired has four tires.", + "C": "The car you repaired has three tires, and they are positioned below you as you look down.", + "D": "The car you repaired has zero tires, so there are no tires below you to see.", + "E": "The car you repaired has two tires. The tires are below you." + }, + "answer_label": "E", + "answer_text": "The car you repaired has two tires. The tires are below you.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_5_T2", + "correct_answer": "The car you repaired has two tires. The tires are below you.", + "original_options": { + "A": "The car you repaired has one tire.", + "B": "The car you repaired has four tires.", + "C": "The car you repaired has three tires, and they are positioned below you as you look down.", + "D": "The car you repaired has zero tires, so there are no tires below you to see.", + "E": "The car you repaired has two tires. The tires are below you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_5_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tires does the car I repaired have?", + "answer": "The car you repaired has four tires. The tires are below you.", + "question_chinese": "我修理的这辆车装有几个轮胎？", + "answer_chinese": "你修理的这辆车装有四个轮胎。轮胎在你的下面。", + "evidence": [ + { + "description": "The second one.", + "timestamps": { + "start": "00:15:38", + "end": "00:15:49" + } + }, + { + "description": "The first one.", + "timestamps": { + "start": "00:14:57", + "end": "00:15:38" + } + } + ], + "question_timestamps": "00:15:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The car you repaired has two tires.", + "B": "The car you repaired has six tires.", + "C": "The car you repaired has three tires, and the tires are located below you in your view.", + "D": "The car you repaired has four tires. The tires are below you.", + "E": "The car you repaired has five tires, with all of the tires positioned below you as you work." + }, + "answer_label": "D", + "answer_text": "The car you repaired has four tires. The tires are below you.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_5_T1", + "correct_answer": "The car you repaired has four tires. The tires are below you.", + "original_options": { + "A": "The car you repaired has two tires.", + "B": "The car you repaired has six tires.", + "C": "The car you repaired has three tires, and the tires are located below you in your view.", + "D": "The car you repaired has four tires. The tires are below you.", + "E": "The car you repaired has five tires, with all of the tires positioned below you as you work." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_9", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tires does the car I repaired have?", + "answer": "It is equipped with two tires.", + "question_chinese": "我修理的这辆车装有几个轮胎？", + "answer_chinese": "装有两个轮胎。", + "evidence": [ + { + "description": "The second one.", + "timestamps": { + "start": "00:15:39", + "end": "00:15:40" + } + }, + { + "description": "The first one.", + "timestamps": { + "start": "00:14:57", + "end": "00:15:00" + } + } + ], + "question_timestamps": "00:19:28", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "It is equipped with one tire.", + "B": "It is equipped with two tires.", + "C": "It is equipped with four tires.", + "D": "It is equipped with three tires, not the amount you were expecting after the repair.", + "E": "It is equipped with five tires, making it more than a typical setup for that car." + }, + "answer_label": "B", + "answer_text": "It is equipped with two tires.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_9", + "correct_answer": "It is equipped with two tires.", + "original_options": { + "A": "It is equipped with one tire.", + "B": "It is equipped with two tires.", + "C": "It is equipped with four tires.", + "D": "It is equipped with three tires, not the amount you were expecting after the repair.", + "E": "It is equipped with five tires, making it more than a typical setup for that car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_10", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tires does the car I repaired have?", + "answer": "It is equipped with two tires.", + "question_chinese": "我修理的这辆车装有几个轮胎？", + "answer_chinese": "装有两个轮胎。", + "evidence": [ + { + "description": "The second one.", + "timestamps": { + "start": "00:15:39", + "end": "00:15:40" + } + }, + { + "description": "The first one.", + "timestamps": { + "start": "00:14:57", + "end": "00:15:00" + } + } + ], + "question_timestamps": "00:23:48", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "It is equipped with one tire.", + "B": "It is equipped with four tires.", + "C": "It is equipped with three tires, which is more than what you repaired onto it.", + "D": "It is equipped with two tires.", + "E": "It is equipped with six tires, giving it a larger total than a standard car." + }, + "answer_label": "D", + "answer_text": "It is equipped with two tires.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_10", + "correct_answer": "It is equipped with two tires.", + "original_options": { + "A": "It is equipped with one tire.", + "B": "It is equipped with four tires.", + "C": "It is equipped with three tires, which is more than what you repaired onto it.", + "D": "It is equipped with two tires.", + "E": "It is equipped with six tires, giving it a larger total than a standard car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_8_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to find the trash can—where is it?", + "answer": "You want to find the trash can. The trash can is to the right of the broom. The trash can is to your front-right.", + "question_chinese": "我要找到垃圾桶，它在哪里？", + "answer_chinese": "你要找到垃圾桶，垃圾桶在扫帚的右边。垃圾桶在你的右前方。", + "evidence": [ + { + "description": "The trash can is on your left front.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your front-left.", + "B": "It’s to your back-right.", + "C": "You want to find the trash can. The trash can is to the right of the broom. The trash can is to your front-right.", + "D": "The trash can is to the left of the broom, and it sits in your front-right area.", + "E": "The trash can is to the right of the broom, but it is located behind you on the right." + }, + "answer_label": "C", + "answer_text": "You want to find the trash can. The trash can is to the right of the broom. The trash can is to your front-right.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_8_T2", + "correct_answer": "You want to find the trash can. The trash can is to the right of the broom. The trash can is to your front-right.", + "original_options": { + "A": "It’s to your front-left.", + "B": "It’s to your back-right.", + "C": "You want to find the trash can. The trash can is to the right of the broom. The trash can is to your front-right.", + "D": "The trash can is to the left of the broom, and it sits in your front-right area.", + "E": "The trash can is to the right of the broom, but it is located behind you on the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_8_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to find the trash can—where is it?", + "answer": "You want to find the trash can. The trash can is in front of the main gate. The trash can is to your front-right.", + "question_chinese": "我要找到垃圾桶，它在哪里？", + "answer_chinese": "你要找到垃圾桶，垃圾桶在大门的前面。垃圾桶在你的右前方。", + "evidence": [ + { + "description": "The trash can is right in front of the gate.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your front-left.", + "B": "You want to find the trash can. The trash can is in front of the main gate. The trash can is to your front-right.", + "C": "It’s to your back-left.", + "D": "The trash can is behind the main gate, and it is positioned to your front-right.", + "E": "The trash can is in front of the main gate, but it sits to your back-right instead." + }, + "answer_label": "B", + "answer_text": "You want to find the trash can. The trash can is in front of the main gate. The trash can is to your front-right.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_8_T1", + "correct_answer": "You want to find the trash can. The trash can is in front of the main gate. The trash can is to your front-right.", + "original_options": { + "A": "It’s to your front-left.", + "B": "You want to find the trash can. The trash can is in front of the main gate. The trash can is to your front-right.", + "C": "It’s to your back-left.", + "D": "The trash can is behind the main gate, and it is positioned to your front-right.", + "E": "The trash can is in front of the main gate, but it sits to your back-right instead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_7", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I need to use a broom. Where is the broom?", + "answer": "The broom is leaning against the gate.", + "question_chinese": "我需要使用扫帚，扫帚在哪里？", + "answer_chinese": "扫帚在大门旁靠着。", + "evidence": [ + { + "description": "The broom is leaning against the gate.", + "timestamps": { + "start": "00:13:11", + "end": "00:13:12" + } + } + ], + "question_timestamps": "00:21:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s under the porch steps.", + "B": "It’s leaning on the fence post.", + "C": "The broom is leaning against the gate.", + "D": "The broom is resting on a counter top nearby rather than propped up at the gate.", + "E": "The broom is hanging inside a storage shed, not leaning against the gate outside." + }, + "answer_label": "C", + "answer_text": "The broom is leaning against the gate.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_7", + "correct_answer": "The broom is leaning against the gate.", + "original_options": { + "A": "It’s under the porch steps.", + "B": "It’s leaning on the fence post.", + "C": "The broom is leaning against the gate.", + "D": "The broom is resting on a counter top nearby rather than propped up at the gate.", + "E": "The broom is hanging inside a storage shed, not leaning against the gate outside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 14 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_8", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I need to use a broom. Where is the broom?", + "answer": "The broom is leaning against the gate.", + "question_chinese": "我需要使用扫帚，扫帚在哪里？", + "answer_chinese": "扫帚在大门旁靠着。", + "evidence": [ + { + "description": "The broom is leaning against the gate.", + "timestamps": { + "start": "00:13:11", + "end": "00:13:12" + } + } + ], + "question_timestamps": "00:25:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s by the sofa.", + "B": "The broom is leaning against the gate.", + "C": "It’s in the garage.", + "D": "The broom is stored inside a closet near the stairs, instead of being left at the gate.", + "E": "The broom is placed under the dining table, rather than leaning against the gate outside." + }, + "answer_label": "B", + "answer_text": "The broom is leaning against the gate.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_8", + "correct_answer": "The broom is leaning against the gate.", + "original_options": { + "A": "It’s by the sofa.", + "B": "The broom is leaning against the gate.", + "C": "It’s in the garage.", + "D": "The broom is stored inside a closet near the stairs, instead of being left at the gate.", + "E": "The broom is placed under the dining table, rather than leaning against the gate outside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 15 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_19", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I find white paper?", + "answer": "It can be found on the top shelf on the left side of the gate.", + "question_chinese": "白纸在哪里可以找到？", + "answer_chinese": "在大门左边的货架最上面可以找到。", + "evidence": [ + { + "description": "There are shelves on the left side of the gate.", + "timestamps": { + "start": "00:51:47", + "end": "00:51:49" + } + }, + { + "description": "There is white paper on the top of the shelf.", + "timestamps": { + "start": "00:51:52", + "end": "00:51:54" + } + } + ], + "question_timestamps": "00:57:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the top shelf by the door.", + "B": "On the middle shelf left of the gate.", + "C": "It is on the top shelf, but on the right side of the gate instead of the left.", + "D": "Look on the top shelf to the left of the cabinet, not on the left side of the gate.", + "E": "It can be found on the top shelf on the left side of the gate." + }, + "answer_label": "E", + "answer_text": "It can be found on the top shelf on the left side of the gate.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_19", + "correct_answer": "It can be found on the top shelf on the left side of the gate.", + "original_options": { + "A": "On the top shelf by the door.", + "B": "On the middle shelf left of the gate.", + "C": "It is on the top shelf, but on the right side of the gate instead of the left.", + "D": "Look on the top shelf to the left of the cabinet, not on the left side of the gate.", + "E": "It can be found on the top shelf on the left side of the gate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 19 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_20", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I find white paper?", + "answer": "It can be found on the top shelf on the left side of the gate.", + "question_chinese": "白纸在哪里可以找到？", + "answer_chinese": "在大门左边的货架最上面可以找到。", + "evidence": [ + { + "description": "There are shelves on the left side of the gate.", + "timestamps": { + "start": "00:51:47", + "end": "00:51:49" + } + }, + { + "description": "There is white paper on the top of the shelf.", + "timestamps": { + "start": "00:51:52", + "end": "00:51:54" + } + } + ], + "question_timestamps": "00:59:54", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It can be found on the top shelf on the left side of the gate.", + "B": "On the middle shelf, right side.", + "C": "On the bottom shelf by the door.", + "D": "It is on the top shelf, but it sits on the right side of the wall rather than near the gate.", + "E": "Check the drawers and the desk beside the window; that is where the white paper is kept." + }, + "answer_label": "A", + "answer_text": "It can be found on the top shelf on the left side of the gate.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_20", + "correct_answer": "It can be found on the top shelf on the left side of the gate.", + "original_options": { + "A": "It can be found on the top shelf on the left side of the gate.", + "B": "On the middle shelf, right side.", + "C": "On the bottom shelf by the door.", + "D": "It is on the top shelf, but it sits on the right side of the wall rather than near the gate.", + "E": "Check the drawers and the desk beside the window; that is where the white paper is kept." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 17 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_5", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did you get the red funnel?", + "answer": "It was obtained from your workbench.", + "question_chinese": "红色的漏斗是从哪里拿的？", + "answer_chinese": "是从你的工作台上拿到的。", + "evidence": [ + { + "description": "It was obtained from your workbench.", + "timestamps": { + "start": "00:05:49", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:11:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was obtained from your workbench.", + "B": "From under the kitchen sink.", + "C": "From the storage closet.", + "D": "You took the red funnel from the garage shelf when you were gathering supplies.", + "E": "You found it inside the bathroom cabinet and then brought it back to your work area." + }, + "answer_label": "A", + "answer_text": "It was obtained from your workbench.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_5", + "correct_answer": "It was obtained from your workbench.", + "original_options": { + "A": "It was obtained from your workbench.", + "B": "From under the kitchen sink.", + "C": "From the storage closet.", + "D": "You took the red funnel from the garage shelf when you were gathering supplies.", + "E": "You found it inside the bathroom cabinet and then brought it back to your work area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 16 + ] + }, + { + "video_uid": "24fa7505-bafd-418d-8bed-62e614085fa9", + "video_length": "01:03:27", + "qid": "24fa7505-bafd-418d-8bed-62e614085fa9_6", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did you get the red funnel?", + "answer": "It was obtained from your workbench.", + "question_chinese": "红色的漏斗是从哪里拿的？", + "answer_chinese": "是从你的工作台上拿到的。", + "evidence": [ + { + "description": "It was obtained from your workbench.", + "timestamps": { + "start": "00:05:49", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:17:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was obtained from your workbench.", + "B": "From the tool rack beside you.", + "C": "From the toolbox on the floor.", + "D": "You picked up the red funnel from the storage cabinet nearby, not from the workbench itself.", + "E": "You pulled it out from a nearby drawer, not from the workbench itself." + }, + "answer_label": "A", + "answer_text": "It was obtained from your workbench.", + "q_id": "24fa7505-bafd-418d-8bed-62e614085fa9_6", + "correct_answer": "It was obtained from your workbench.", + "original_options": { + "A": "It was obtained from your workbench.", + "B": "From the tool rack beside you.", + "C": "From the toolbox on the floor.", + "D": "You picked up the red funnel from the storage cabinet nearby, not from the workbench itself.", + "E": "You pulled it out from a nearby drawer, not from the workbench itself." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 13 + ] + } +] diff --git a/QAs/29e45749-f8d5-42c7-af09-7fbb3d6697e9-mcq.json b/QAs/29e45749-f8d5-42c7-af09-7fbb3d6697e9-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2e75afbec02f916d25de5c131a31034f5049aa2d --- /dev/null +++ b/QAs/29e45749-f8d5-42c7-af09-7fbb3d6697e9-mcq.json @@ -0,0 +1,1025 @@ +[ + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_17", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothes have I folded?", + "answer": "You folded two pieces.", + "question_chinese": "我叠好了多少件衣服", + "answer_chinese": "你叠好了两件", + "evidence": [ + { + "description": "You folded two pieces.", + "timestamps": { + "start": "00:21:55", + "end": "00:22:29" + } + } + ], + "question_timestamps": "00:22:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You folded one piece.", + "B": "You folded three pieces.", + "C": "You folded four pieces of clothes in total during the folding task.", + "D": "You folded five pieces of clothing altogether before you stopped folding completely.", + "E": "You folded two pieces." + }, + "answer_label": "E", + "answer_text": "You folded two pieces.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_17", + "correct_answer": "You folded two pieces.", + "original_options": { + "A": "You folded one piece.", + "B": "You folded three pieces.", + "C": "You folded four pieces of clothes in total during the folding task.", + "D": "You folded five pieces of clothing altogether before you stopped folding completely.", + "E": "You folded two pieces." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_18", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothes have I folded?", + "answer": "You folded twenty-two pieces of clothing.", + "question_chinese": "我叠好了多少件衣服", + "answer_chinese": "你叠好了22件衣服", + "evidence": [ + { + "description": "You folded two pieces.", + "timestamps": { + "start": "00:21:55", + "end": "00:22:29" + } + }, + { + "description": "You folded 22 pieces of clothing.", + "timestamps": { + "start": "00:22:30", + "end": "00:28:05" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You folded twenty-two pieces of clothing.", + "B": "You folded twelve pieces of clothing.", + "C": "You folded eighteen pieces of clothing.", + "D": "You folded five pieces of clothing in total, not a large batch.", + "E": "You folded thirty pieces of clothing altogether during the entire folding session." + }, + "answer_label": "A", + "answer_text": "You folded twenty-two pieces of clothing.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_18", + "correct_answer": "You folded twenty-two pieces of clothing.", + "original_options": { + "A": "You folded twenty-two pieces of clothing.", + "B": "You folded twelve pieces of clothing.", + "C": "You folded eighteen pieces of clothing.", + "D": "You folded five pieces of clothing in total, not a large batch.", + "E": "You folded thirty pieces of clothing altogether during the entire folding session." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 12 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_11", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do with this woman?", + "answer": "You danced with her and also sorted out clothes together.", + "question_chinese": "我和这个女人做了什么", + "answer_chinese": "你跟她跳了舞还一起整理了衣服", + "evidence": [ + { + "description": "You danced with her.", + "timestamps": { + "start": "00:15:00", + "end": "00:15:25" + } + }, + { + "description": "You and her tidy up the clothes.", + "timestamps": { + "start": "00:21:58", + "end": "00:22:26" + } + } + ], + "question_timestamps": "00:22:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went shopping with her.", + "B": "You painted with her.", + "C": "You danced with her and also sorted out clothes together.", + "D": "You chatted with her and then prepared a meal together afterward at home.", + "E": "You played games with her and also helped arrange furniture in the room." + }, + "answer_label": "C", + "answer_text": "You danced with her and also sorted out clothes together.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_11", + "correct_answer": "You danced with her and also sorted out clothes together.", + "original_options": { + "A": "You went shopping with her.", + "B": "You painted with her.", + "C": "You danced with her and also sorted out clothes together.", + "D": "You chatted with her and then prepared a meal together afterward at home.", + "E": "You played games with her and also helped arrange furniture in the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 13 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_12", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do with this woman?", + "answer": "You cooked with her.", + "question_chinese": "我和这个女人做了什么", + "answer_chinese": "你跟她一起做饭了", + "evidence": [ + { + "description": "You danced with her.", + "timestamps": { + "start": "00:15:00", + "end": "00:15:25" + } + }, + { + "description": "You and her tidy up the clothes.", + "timestamps": { + "start": "00:21:58", + "end": "00:22:26" + } + }, + { + "description": "You cooked with her.", + "timestamps": { + "start": "00:28:30", + "end": "00:30:00" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You made a smoothie with her.", + "B": "You cleaned the kitchen with her.", + "C": "You painted a room with her instead of cooking anything together for dinner.", + "D": "You cooked with her.", + "E": "You danced with her rather than working together on food preparation in the kitchen." + }, + "answer_label": "D", + "answer_text": "You cooked with her.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_12", + "correct_answer": "You cooked with her.", + "original_options": { + "A": "You made a smoothie with her.", + "B": "You cleaned the kitchen with her.", + "C": "You painted a room with her instead of cooking anything together for dinner.", + "D": "You cooked with her.", + "E": "You danced with her rather than working together on food preparation in the kitchen." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 14 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_15", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How do I get to the bedside?", + "answer": "You can go straight first, then turn right, and finally go straight.", + "question_chinese": "我如何到床旁边", + "answer_chinese": "你可以先直走再右拐最后直走", + "evidence": [ + { + "description": "You can go straight first, then turn right, and finally go straight.", + "timestamps": { + "start": "00:21:15", + "end": "00:21:30" + } + } + ], + "question_timestamps": "00:22:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn left.", + "B": "Turn right, then go straight.", + "C": "Go straight first, turn left at the corner, then continue straight until you reach the bedside.", + "D": "Head straight ahead, make a right turn, then veer slightly left instead of continuing straight.", + "E": "You can go straight first, then turn right, and finally go straight." + }, + "answer_label": "E", + "answer_text": "You can go straight first, then turn right, and finally go straight.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_15", + "correct_answer": "You can go straight first, then turn right, and finally go straight.", + "original_options": { + "A": "Go straight, then turn left.", + "B": "Turn right, then go straight.", + "C": "Go straight first, turn left at the corner, then continue straight until you reach the bedside.", + "D": "Head straight ahead, make a right turn, then veer slightly left instead of continuing straight.", + "E": "You can go straight first, then turn right, and finally go straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_16", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How do I get to the bedside?", + "answer": "You can go straight.", + "question_chinese": "我如何到床旁边", + "answer_chinese": "你可以直走", + "evidence": [ + { + "description": "You can go straight first, then turn right, and finally go straight.", + "timestamps": { + "start": "00:21:15", + "end": "00:21:30" + } + }, + { + "description": "You can go straight.", + "timestamps": { + "start": "00:28:27", + "end": "00:28:33" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right immediately.", + "B": "Go straight, then turn left.", + "C": "Take a couple of steps forward, then turn right to reach the bedside instead of continuing straight.", + "D": "Walk straight briefly, then stop and turn around before moving forward again toward the bedside.", + "E": "You can go straight." + }, + "answer_label": "E", + "answer_text": "You can go straight.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_16", + "correct_answer": "You can go straight.", + "original_options": { + "A": "Turn right immediately.", + "B": "Go straight, then turn left.", + "C": "Take a couple of steps forward, then turn right to reach the bedside instead of continuing straight.", + "D": "Walk straight briefly, then stop and turn around before moving forward again toward the bedside.", + "E": "You can go straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 5, + 17, + 15 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_6", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the cabinet?", + "answer": "You opened it zero times.", + "question_chinese": "我开了多少次柜子", + "answer_chinese": "你打开了0次", + "evidence": [ + { + "description": "You opened it 0 times.", + "timestamps": { + "start": "00:00:00", + "end": "00:07:29" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You opened it once.", + "B": "You opened it twice.", + "C": "You opened the cabinet three times over the course of the task.", + "D": "You opened it zero times.", + "E": "You opened the cabinet four times, closing it each time before reopening it." + }, + "answer_label": "D", + "answer_text": "You opened it zero times.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_6", + "correct_answer": "You opened it zero times.", + "original_options": { + "A": "You opened it once.", + "B": "You opened it twice.", + "C": "You opened the cabinet three times over the course of the task.", + "D": "You opened it zero times.", + "E": "You opened the cabinet four times, closing it each time before reopening it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_5", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the cabinet?", + "answer": "You opened the cabinet once.", + "question_chinese": "我开了多少次柜子", + "answer_chinese": "你打开了一次柜子", + "evidence": [ + { + "description": "You opened it 0 times.", + "timestamps": { + "start": "00:00:00", + "end": "00:07:29" + } + }, + { + "description": "You opened the cabinet once.", + "timestamps": { + "start": "00:09:38", + "end": "00:09:56" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You opened it zero times.", + "B": "You opened the cabinet once.", + "C": "You opened it twice.", + "D": "You opened the cabinet three times while getting items and then closing it again each time.", + "E": "You opened the cabinet four times during the sequence, reopening it after closing it between attempts." + }, + "answer_label": "B", + "answer_text": "You opened the cabinet once.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_5", + "correct_answer": "You opened the cabinet once.", + "original_options": { + "A": "You opened it zero times.", + "B": "You opened the cabinet once.", + "C": "You opened it twice.", + "D": "You opened the cabinet three times while getting items and then closing it again each time.", + "E": "You opened the cabinet four times during the sequence, reopening it after closing it between attempts." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_8", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I do when I opened the refrigerator?", + "answer": "You took out two kinds of ingredients.", + "question_chinese": "我打开冰箱做了什么", + "answer_chinese": "你取出了两种食材", + "evidence": [ + { + "description": "You took out a carrot.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:28" + } + }, + { + "description": "You took out an apple.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:16" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You grabbed a drink.", + "B": "You only looked inside.", + "C": "You took out ingredients from the fridge and set them aside on a shelf rather than taking two kinds out.", + "D": "You took out two kinds of ingredients.", + "E": "You removed a few items from the refrigerator, then put them back without separating them into two kinds." + }, + "answer_label": "D", + "answer_text": "You took out two kinds of ingredients.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_8", + "correct_answer": "You took out two kinds of ingredients.", + "original_options": { + "A": "You grabbed a drink.", + "B": "You only looked inside.", + "C": "You took out ingredients from the fridge and set them aside on a shelf rather than taking two kinds out.", + "D": "You took out two kinds of ingredients.", + "E": "You removed a few items from the refrigerator, then put them back without separating them into two kinds." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 18 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_7", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I do when I opened the refrigerator?", + "answer": "You took and put some things.", + "question_chinese": "我打开冰箱做了什么", + "answer_chinese": "你拿放了一些东西", + "evidence": [ + { + "description": "You took out a carrot.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:28" + } + }, + { + "description": "You took out an apple.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:16" + } + }, + { + "description": "You put the sliced apples in.", + "timestamps": { + "start": "00:09:56", + "end": "00:10:00" + } + }, + { + "description": "You picked up a bottle of sauce, found it empty, and put it back.", + "timestamps": { + "start": "00:12:35", + "end": "00:12:47" + } + }, + { + "description": "You put the mixed ingredients in the refrigerator.", + "timestamps": { + "start": "00:13:45", + "end": "00:13:54" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You only took things out.", + "B": "You took and put some things.", + "C": "You just put items inside.", + "D": "You took items out of the refrigerator and kept them out, without placing anything back inside.", + "E": "You rearranged what was in the refrigerator by moving items around, but you did not take and put things." + }, + "answer_label": "B", + "answer_text": "You took and put some things.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_7", + "correct_answer": "You took and put some things.", + "original_options": { + "A": "You only took things out.", + "B": "You took and put some things.", + "C": "You just put items inside.", + "D": "You took items out of the refrigerator and kept them out, without placing anything back inside.", + "E": "You rearranged what was in the refrigerator by moving items around, but you did not take and put things." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 19 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_4", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the chopped carrots?", + "answer": "You poured it into the box containing the first ingredient and then put it in the oven.", + "question_chinese": "我把切好的胡萝卜放在了哪里", + "answer_chinese": "你把它倒入装第一个食材的盒子里然后放进了烤箱", + "evidence": [ + { + "description": "You poured it into the box containing the first ingredient and then put it in the oven.", + "timestamps": { + "start": "00:05:20", + "end": "00:06:05" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You left them on the counter.", + "B": "You put them in the fridge.", + "C": "You poured the chopped carrots into the box with the first ingredient, then set the box on the counter.", + "D": "You poured it into the box containing the first ingredient and then put it in the oven.", + "E": "You poured the chopped carrots into a bowl with the first ingredient and then put it in the oven." + }, + "answer_label": "D", + "answer_text": "You poured it into the box containing the first ingredient and then put it in the oven.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_4", + "correct_answer": "You poured it into the box containing the first ingredient and then put it in the oven.", + "original_options": { + "A": "You left them on the counter.", + "B": "You put them in the fridge.", + "C": "You poured the chopped carrots into the box with the first ingredient, then set the box on the counter.", + "D": "You poured it into the box containing the first ingredient and then put it in the oven.", + "E": "You poured the chopped carrots into a bowl with the first ingredient and then put it in the oven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_3", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the chopped carrots?", + "answer": "You poured it into the box containing the first ingredient and then put it in the oven.", + "question_chinese": "我把切好的胡萝卜放在了哪里", + "answer_chinese": "你把它倒入装第一个食材的盒子里然后放进了烤箱", + "evidence": [ + { + "description": "You poured it into the box containing the first ingredient and then put it in the oven.", + "timestamps": { + "start": "00:05:20", + "end": "00:06:05" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You poured it into the box containing the first ingredient and then put it in the oven.", + "B": "You put it in the fridge.", + "C": "You left it on the counter.", + "D": "You poured the chopped carrots into the box with the first ingredient, then placed it on a shelf.", + "E": "You poured the chopped carrots into a bowl containing the first ingredient and then put it in the oven." + }, + "answer_label": "A", + "answer_text": "You poured it into the box containing the first ingredient and then put it in the oven.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_3", + "correct_answer": "You poured it into the box containing the first ingredient and then put it in the oven.", + "original_options": { + "A": "You poured it into the box containing the first ingredient and then put it in the oven.", + "B": "You put it in the fridge.", + "C": "You left it on the counter.", + "D": "You poured the chopped carrots into the box with the first ingredient, then placed it on a shelf.", + "E": "You poured the chopped carrots into a bowl containing the first ingredient and then put it in the oven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_19", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the kitchen knife?", + "answer": "After you washed it, you put it on the stove.", + "question_chinese": "我把菜刀放在了哪里", + "answer_chinese": "你洗完之后把它放在了灶台上", + "evidence": [ + { + "description": "After you washed it, you put it on the stove.", + "timestamps": { + "start": "00:19:06", + "end": "00:20:00" + } + } + ], + "question_timestamps": "00:22:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it in the drawer.", + "B": "After you washed it, you put it on the stove.", + "C": "You put it in the sink.", + "D": "After washing it, you set it on the counter beside the stove, not directly on the stove.", + "E": "After you washed it, you placed it in the cupboard to dry instead of leaving it out on the stove." + }, + "answer_label": "B", + "answer_text": "After you washed it, you put it on the stove.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_19", + "correct_answer": "After you washed it, you put it on the stove.", + "original_options": { + "A": "You put it in the drawer.", + "B": "After you washed it, you put it on the stove.", + "C": "You put it in the sink.", + "D": "After washing it, you set it on the counter beside the stove, not directly on the stove.", + "E": "After you washed it, you placed it in the cupboard to dry instead of leaving it out on the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 20 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_20", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the kitchen knife?", + "answer": "You put it in the box.", + "question_chinese": "我把菜刀放在了哪里", + "answer_chinese": "你把它放在了盒子里", + "evidence": [ + { + "description": "After you washed it, you put it on the stove.", + "timestamps": { + "start": "00:19:06", + "end": "00:20:00" + } + }, + { + "description": "You put it in the box.", + "timestamps": { + "start": "00:29:00", + "end": "00:29:45" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You put it in the drawer.", + "B": "You put it on the counter.", + "C": "You put it in the box.", + "D": "You put it near the cutting board, leaving it out rather than putting it away in a container.", + "E": "You slid it under the table for a moment instead of placing it inside any box." + }, + "answer_label": "C", + "answer_text": "You put it in the box.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_20", + "correct_answer": "You put it in the box.", + "original_options": { + "A": "You put it in the drawer.", + "B": "You put it on the counter.", + "C": "You put it in the box.", + "D": "You put it near the cutting board, leaving it out rather than putting it away in a container.", + "E": "You slid it under the table for a moment instead of placing it inside any box." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 16 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_13", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the detergent?", + "answer": "You can.", + "question_chinese": "我能够到洗洁精吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:15:45", + "end": "00:20:30" + } + } + ], + "question_timestamps": "00:22:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_13", + "correct_answer": "You can.", + "original_options": { + "A": "You cannot.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_14", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the detergent?", + "answer": "You can.", + "question_chinese": "我能够到洗洁精吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:15:45", + "end": "00:20:30" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:28:46", + "end": "00:28:52" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You cannot." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_14", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_2", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach that green thing?", + "answer": "You can.", + "question_chinese": "我能够到那个绿色的东西吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:04:40", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_2", + "correct_answer": "You can.", + "original_options": { + "A": "You cannot.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9", + "video_length": "00:30:00", + "qid": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_1", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach that green thing?", + "answer": "You can.", + "question_chinese": "我能够到那个绿色的东西吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:04:40", + "end": "00:05:00" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:08:19", + "end": "00:09:00" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:09:43", + "end": "00:09:55" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You cannot." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "29e45749-f8d5-42c7-af09-7fbb3d6697e9_1", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + } +] diff --git a/QAs/2a6b3d10-8da9-4f0e-a681-59ba48a55dbf-mcq.json b/QAs/2a6b3d10-8da9-4f0e-a681-59ba48a55dbf-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..106e1668dca3fcd2b62d979eb910f7bad58b37d4 --- /dev/null +++ b/QAs/2a6b3d10-8da9-4f0e-a681-59ba48a55dbf-mcq.json @@ -0,0 +1,422 @@ +[ + { + "video_uid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf", + "video_length": "00:27:47", + "qid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_7", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following?", + "answer": "You turned around and walked straight three steps to the sink.", + "question_chinese": "我刚刚走什么路线？", + "answer_chinese": "你转身，直走三步到洗手池旁边。", + "evidence": [ + { + "description": "Turn around from the table and take three steps straight to the sink.", + "timestamps": { + "start": "00:16:08", + "end": "00:16:11" + } + } + ], + "question_timestamps": "00:16:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You turned left to the sink.", + "B": "You turned around and walked straight three steps to the sink.", + "C": "You turned around, two steps to sink.", + "D": "You turned around and walked straight three steps to the counter instead of the sink.", + "E": "You turned back and walked straight four steps, stopping at the sink area after overshooting your route." + }, + "answer_label": "B", + "answer_text": "You turned around and walked straight three steps to the sink.", + "q_id": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_7", + "correct_answer": "You turned around and walked straight three steps to the sink.", + "original_options": { + "A": "You turned left to the sink.", + "B": "You turned around and walked straight three steps to the sink.", + "C": "You turned around, two steps to sink.", + "D": "You turned around and walked straight three steps to the counter instead of the sink.", + "E": "You turned back and walked straight four steps, stopping at the sink area after overshooting your route." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 17 + ] + }, + { + "video_uid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf", + "video_length": "00:27:47", + "qid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_8", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following?", + "answer": "You turn away from the stove and take three steps straight to the table.", + "question_chinese": "我刚刚走什么路线？", + "answer_chinese": "你从灶台旁边转身，直走三步到桌子旁边。", + "evidence": [ + { + "description": "Turn around from the stove and take three steps straight to the table.", + "timestamps": { + "start": "00:20:53", + "end": "00:20:55" + } + } + ], + "question_timestamps": "00:20:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You turn away from the stove and take three steps straight to the table.", + "B": "You turned toward the stove.", + "C": "Four steps to the table.", + "D": "You turn away from the stove and take three steps straight to the counter, not the table.", + "E": "You turn away from the stove and take three steps, but head diagonally instead of straight." + }, + "answer_label": "A", + "answer_text": "You turn away from the stove and take three steps straight to the table.", + "q_id": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_8", + "correct_answer": "You turn away from the stove and take three steps straight to the table.", + "original_options": { + "A": "You turn away from the stove and take three steps straight to the table.", + "B": "You turned toward the stove.", + "C": "Four steps to the table.", + "D": "You turn away from the stove and take three steps straight to the counter, not the table.", + "E": "You turn away from the stove and take three steps, but head diagonally instead of straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf", + "video_length": "00:27:47", + "qid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_1_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Before I started cutting the ingredients, was the knife to my left or to my right?", + "answer": "The knife is to your right.", + "question_chinese": "我开始切食材之前，刀在我的左边还是右边？", + "answer_chinese": "刀在你的右边。", + "evidence": [ + { + "description": "There is a knife on the edge of the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The knife is to your right.", + "B": "The knife is to your left." + }, + "answer_label": "A", + "answer_text": "The knife is to your right.", + "q_id": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_1_T1", + "correct_answer": "The knife is to your right.", + "original_options": { + "A": "The knife is to your right.", + "B": "The knife is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf", + "video_length": "00:27:47", + "qid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_1_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Before I started cutting the ingredients, was the knife to my left or to my right?", + "answer": "The knife is on your right-hand side.", + "question_chinese": "我开始切食材之前，刀在我的左边还是右边？", + "answer_chinese": "刀在你的右手边。", + "evidence": [ + { + "description": "There is a knife on the edge of the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:09:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The knife is on your left-hand side.", + "B": "The knife is on your right-hand side." + }, + "answer_label": "B", + "answer_text": "The knife is on your right-hand side.", + "q_id": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_1_T2", + "correct_answer": "The knife is on your right-hand side.", + "original_options": { + "A": "The knife is on your left-hand side.", + "B": "The knife is on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf", + "video_length": "00:27:47", + "qid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I want to use green peppers to cook. How many green peppers are still usable now?", + "answer": "A green pepper.", + "question_chinese": "我想用青椒做菜，现在还能用的青椒有多少根？", + "answer_chinese": "一根青椒。", + "evidence": [ + { + "description": "You just used a green pepper.", + "timestamps": { + "start": "00:02:36", + "end": "00:02:45" + } + }, + { + "description": "There is also a green pepper on the table.", + "timestamps": { + "start": "00:02:54", + "end": "00:02:58" + } + } + ], + "question_timestamps": "00:03:02", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "No green peppers.", + "B": "A green pepper.", + "C": "Two green peppers.", + "D": "Three green peppers are still usable now, so you can use any of them for cooking.", + "E": "Four green peppers are still usable now, meaning you have several left to cook with." + }, + "answer_label": "B", + "answer_text": "A green pepper.", + "q_id": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_3", + "correct_answer": "A green pepper.", + "original_options": { + "A": "No green peppers.", + "B": "A green pepper.", + "C": "Two green peppers.", + "D": "Three green peppers are still usable now, so you can use any of them for cooking.", + "E": "Four green peppers are still usable now, meaning you have several left to cook with." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 15 + ] + }, + { + "video_uid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf", + "video_length": "00:27:47", + "qid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I want to use green peppers to cook. How many green peppers are still usable now?", + "answer": "Zero.", + "question_chinese": "我想用青椒做菜，现在还能用的青椒有多少根？", + "answer_chinese": "零。", + "evidence": [ + { + "description": "You used the last green pepper on the table.", + "timestamps": { + "start": "00:09:39", + "end": "00:09:55" + } + }, + { + "description": "You just used a green pepper.", + "timestamps": { + "start": "00:02:34", + "end": "00:02:46" + } + } + ], + "question_timestamps": "00:10:05", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three green peppers are still usable now, so you can choose from multiple peppers to cook.", + "D": "Zero.", + "E": "Five green peppers are still usable now, giving you plenty of peppers left to use." + }, + "answer_label": "D", + "answer_text": "Zero.", + "q_id": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_4", + "correct_answer": "Zero.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three green peppers are still usable now, so you can choose from multiple peppers to cook.", + "D": "Zero.", + "E": "Five green peppers are still usable now, giving you plenty of peppers left to use." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf", + "video_length": "00:27:47", + "qid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_11", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I need to put this plate back in its original position. Which way should I go?", + "answer": "You should turn around and take four steps backward to place the plate on the shelf.", + "question_chinese": "我要把这个盘子放回原位，应该怎么走？", + "answer_chinese": "你应该向后转走四步，把盘子放到碗架上。", + "evidence": [ + { + "description": "You took the plate, turned around, and walked four steps backward to your current position.", + "timestamps": { + "start": "00:21:55", + "end": "00:21:58" + } + } + ], + "question_timestamps": "00:22:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left to the shelf.", + "B": "You should turn around and take four steps backward to place the plate on the shelf.", + "C": "Two steps backward.", + "D": "You should turn around and take four steps forward to place the plate on the shelf, not backward.", + "E": "You should turn around, move four steps to your right, and then place the plate back on the shelf." + }, + "answer_label": "B", + "answer_text": "You should turn around and take four steps backward to place the plate on the shelf.", + "q_id": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_11", + "correct_answer": "You should turn around and take four steps backward to place the plate on the shelf.", + "original_options": { + "A": "Turn left to the shelf.", + "B": "You should turn around and take four steps backward to place the plate on the shelf.", + "C": "Two steps backward.", + "D": "You should turn around and take four steps forward to place the plate on the shelf, not backward.", + "E": "You should turn around, move four steps to your right, and then place the plate back on the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 18, + 19 + ] + }, + { + "video_uid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf", + "video_length": "00:27:47", + "qid": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_12", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I need to put this plate back in its original position. Which way should I go?", + "answer": "You should turn around and take five steps, pass through the wooden door, and then take two more steps to place the plate on the shel.f.", + "question_chinese": "我要把这个盘子放回原位，应该怎么走？", + "answer_chinese": "你应该向后转走5步，穿过木门，再走两步，把盘子放到碗架上。", + "evidence": [ + { + "description": "You turned right from the stove, passed through the wooden door, and took the plate from the shelf.", + "timestamps": { + "start": "00:21:47", + "end": "00:21:51" + } + } + ], + "question_timestamps": "00:22:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should turn around and take five steps, pass through the wooden door, and then take two more steps to place the plate on the shel.f.", + "B": "Take six steps first.", + "C": "Wooden door after four steps.", + "D": "You should turn around, take five steps, pass through the wooden door, then take three more steps to reach the shelf.", + "E": "You should turn around, take five steps, go through the blue door, then take two steps to the shelf." + }, + "answer_label": "A", + "answer_text": "You should turn around and take five steps, pass through the wooden door, and then take two more steps to place the plate on the shel.f.", + "q_id": "2a6b3d10-8da9-4f0e-a681-59ba48a55dbf_12", + "correct_answer": "You should turn around and take five steps, pass through the wooden door, and then take two more steps to place the plate on the shel.f.", + "original_options": { + "A": "You should turn around and take five steps, pass through the wooden door, and then take two more steps to place the plate on the shel.f.", + "B": "Take six steps first.", + "C": "Wooden door after four steps.", + "D": "You should turn around, take five steps, pass through the wooden door, then take three more steps to reach the shelf.", + "E": "You should turn around, take five steps, go through the blue door, then take two steps to the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 21, + 19 + ] + } +] diff --git a/QAs/2b8f76b9-e542-4413-9fbd-0f7710a39dfb-mcq.json b/QAs/2b8f76b9-e542-4413-9fbd-0f7710a39dfb-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fcff39f862da2428c612fea9c0f7eb511536071a --- /dev/null +++ b/QAs/2b8f76b9-e542-4413-9fbd-0f7710a39dfb-mcq.json @@ -0,0 +1,729 @@ +[ + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_13", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From my current location, what route should I take to go home?", + "answer": "You should turn around, then go straight along the path on the right.", + "question_chinese": "从我现在的位置回家，我要走什么路线呢？", + "answer_chinese": "你应该转身向后，然后沿着右边的路直行。", + "evidence": [ + { + "description": "You walked from the front door along the path to the park.", + "timestamps": { + "start": "00:01:21", + "end": "00:03:07" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go straight.", + "B": "Go straight without turning around.", + "C": "Turn around and follow the path on the left until you reach home.", + "D": "You should turn around, then go straight along the path on the right.", + "E": "Turn around, then take the middle path and continue straight to get back home." + }, + "answer_label": "D", + "answer_text": "You should turn around, then go straight along the path on the right.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_13", + "correct_answer": "You should turn around, then go straight along the path on the right.", + "original_options": { + "A": "Turn left, then go straight.", + "B": "Go straight without turning around.", + "C": "Turn around and follow the path on the left until you reach home.", + "D": "You should turn around, then go straight along the path on the right.", + "E": "Turn around, then take the middle path and continue straight to get back home." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_14", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From my current location, what route should I take to go home?", + "answer": "You should go straight along the path on the right.", + "question_chinese": "从我现在的位置回家，我要走什么路线呢？", + "answer_chinese": "你应该沿着右边的路直行。", + "evidence": [ + { + "description": "You walked from the front door along the path to the park.", + "timestamps": { + "start": "00:01:21", + "end": "00:03:07" + } + }, + { + "description": "You started walking back from the park.", + "timestamps": { + "start": "00:03:43", + "end": "00:04:35" + } + } + ], + "question_timestamps": "00:04:47", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left onto the left path.", + "B": "Turn around, then go straight.", + "C": "You should go straight along the path on the right.", + "D": "Walk straight along the left path rather than taking the right-side route home.", + "E": "Go straight along the middle path until you reach home, not the path on the right." + }, + "answer_label": "C", + "answer_text": "You should go straight along the path on the right.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_14", + "correct_answer": "You should go straight along the path on the right.", + "original_options": { + "A": "Turn left onto the left path.", + "B": "Turn around, then go straight.", + "C": "You should go straight along the path on the right.", + "D": "Walk straight along the left path rather than taking the right-side route home.", + "E": "Go straight along the middle path until you reach home, not the path on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 16 + ] + }, + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_4", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to walk the dog in the park now, what route should I take?", + "answer": "You should go straight to the kitchen, then turn left and continue straight, then turn right and go straight to the door. Once you open the door, continue forward and follow the left side.", + "question_chinese": "如果我现在想去公园里遛狗，我要走什么路线呢？", + "answer_chinese": "你应该直走到厨房，再左转直走，然后右转直走到门口，打开门口之后向前方，沿着左边走。", + "evidence": [ + { + "description": "You walked the dog in the park.", + "timestamps": { + "start": "00:02:36", + "end": "00:04:49" + } + }, + { + "description": "You returned home.", + "timestamps": { + "start": "00:05:32", + "end": "00:05:52" + } + }, + { + "description": "You went straight, turned left, and walked to the kitchen.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:03" + } + }, + { + "description": "You walked from the kitchen to the living room.", + "timestamps": { + "start": "00:07:15", + "end": "00:07:21" + } + } + ], + "question_timestamps": "00:07:23", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right after the kitchen.", + "B": "Go to the door directly.", + "C": "You should go straight to the kitchen, then turn left and continue straight, then turn right and go straight to the door. Once you open the door, continue forward and follow the left side.", + "D": "Go straight to the kitchen, turn left, then turn left again and go straight to the door.", + "E": "Head to the kitchen, turn right, keep going straight, then turn right to the door and follow the right side outside." + }, + "answer_label": "C", + "answer_text": "You should go straight to the kitchen, then turn left and continue straight, then turn right and go straight to the door. Once you open the door, continue forward and follow the left side.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_4", + "correct_answer": "You should go straight to the kitchen, then turn left and continue straight, then turn right and go straight to the door. Once you open the door, continue forward and follow the left side.", + "original_options": { + "A": "Turn right after the kitchen.", + "B": "Go to the door directly.", + "C": "You should go straight to the kitchen, then turn left and continue straight, then turn right and go straight to the door. Once you open the door, continue forward and follow the left side.", + "D": "Go straight to the kitchen, turn left, then turn left again and go straight to the door.", + "E": "Head to the kitchen, turn right, keep going straight, then turn right to the door and follow the right side outside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 21 + ] + }, + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_3", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to walk the dog in the park now, what route should I take?", + "answer": "You should turn left and go straight, then turn right and go straight to the door. Once you open the door, continue forward and follow the left side.", + "question_chinese": "如果我现在想去公园里遛狗，我要走什么路线呢？", + "answer_chinese": "你应该左转直走，然后右转直走到门口，打开门口之后向前方，沿着左边走。", + "evidence": [ + { + "description": "You walked the dog in the park.", + "timestamps": { + "start": "00:02:36", + "end": "00:04:49" + } + }, + { + "description": "You returned home.", + "timestamps": { + "start": "00:05:32", + "end": "00:05:52" + } + }, + { + "description": "You went straight, turned left, and walked to the kitchen.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:03" + } + } + ], + "question_timestamps": "00:08:07", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You should turn left and go straight, then turn right and go straight to the door. Once you open the door, continue forward and follow the left side.", + "B": "Turn right and go straight.", + "C": "Turn left, then turn left again.", + "D": "Turn left and go straight, then turn right to the door and follow the right side outside.", + "E": "Go straight to the kitchen first, then turn left, continue straight, turn right to the door, and head out." + }, + "answer_label": "A", + "answer_text": "You should turn left and go straight, then turn right and go straight to the door. Once you open the door, continue forward and follow the left side.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_3", + "correct_answer": "You should turn left and go straight, then turn right and go straight to the door. Once you open the door, continue forward and follow the left side.", + "original_options": { + "A": "You should turn left and go straight, then turn right and go straight to the door. Once you open the door, continue forward and follow the left side.", + "B": "Turn right and go straight.", + "C": "Turn left, then turn left again.", + "D": "Turn left and go straight, then turn right to the door and follow the right side outside.", + "E": "Go straight to the kitchen first, then turn left, continue straight, turn right to the door, and head out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 19 + ] + }, + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get the bag now, what route should I take?", + "answer": "You should turn around and walk back along the path.", + "question_chinese": "如果我现在想去拿袋子的话，我要走什么路线呢？", + "answer_chinese": "你应该转身然后沿着路径向后走。", + "evidence": [ + { + "description": "You obtained the bag at the park entrance.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:47" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go straight along the path.", + "B": "You should turn around and walk back along the path.", + "C": "Turn left and keep walking.", + "D": "Keep moving forward along the path instead of turning around to walk back.", + "E": "Turn right off the path and continue straight until you reach the bag’s location." + }, + "answer_label": "B", + "answer_text": "You should turn around and walk back along the path.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_1", + "correct_answer": "You should turn around and walk back along the path.", + "original_options": { + "A": "Go straight along the path.", + "B": "You should turn around and walk back along the path.", + "C": "Turn left and keep walking.", + "D": "Keep moving forward along the path instead of turning around to walk back.", + "E": "Turn right off the path and continue straight until you reach the bag’s location." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 15 + ] + }, + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get the bag now, what route should I take?", + "answer": "You should turn around and walk along the left path.", + "question_chinese": "如果我现在想去拿袋子的话，我要走什么路线呢？", + "answer_chinese": "你应该向后转身，沿着左边的路走。", + "evidence": [ + { + "description": "You obtained the bag at the park entrance.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:47" + } + } + ], + "question_timestamps": "00:05:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn around along the right path.", + "B": "Turn around and go straight.", + "C": "Turn around and take the right path instead of the left to reach the bag.", + "D": "Turn around and follow the path ahead without choosing the left branch at the split.", + "E": "You should turn around and walk along the left path." + }, + "answer_label": "E", + "answer_text": "You should turn around and walk along the left path.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_2", + "correct_answer": "You should turn around and walk along the left path.", + "original_options": { + "A": "Turn around along the right path.", + "B": "Turn around and go straight.", + "C": "Turn around and take the right path instead of the left to reach the bag.", + "D": "Turn around and follow the path ahead without choosing the left branch at the split.", + "E": "You should turn around and walk along the left path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 15 + ] + }, + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_4_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to keep wiping things—where should I go to find the kitchen paper?", + "answer": "You should go to the right side of the kitchen sink; it’s on your right.", + "question_chinese": "我想继续擦东西，我应该去哪里找到厨房纸？", + "answer_chinese": "你应该去厨房的洗手台右侧，在你的右侧。", + "evidence": [ + { + "description": "You got the kitchen paper from the right side of the kitchen sink.", + "timestamps": { + "start": "00:19:00", + "end": "00:19:11" + } + } + ], + "question_timestamps": "00:19:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go to the left side.", + "B": "Go behind the sink.", + "C": "Go to the right side of the kitchen sink; it’s to your front-right, not directly on your right.", + "D": "Go to the right side of the kitchen sink; it should be slightly behind you rather than on your right.", + "E": "You should go to the right side of the kitchen sink; it’s on your right." + }, + "answer_label": "E", + "answer_text": "You should go to the right side of the kitchen sink; it’s on your right.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_4_T1", + "correct_answer": "You should go to the right side of the kitchen sink; it’s on your right.", + "original_options": { + "A": "Go to the left side.", + "B": "Go behind the sink.", + "C": "Go to the right side of the kitchen sink; it’s to your front-right, not directly on your right.", + "D": "Go to the right side of the kitchen sink; it should be slightly behind you rather than on your right.", + "E": "You should go to the right side of the kitchen sink; it’s on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 19, + 20 + ] + }, + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_4_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to keep wiping things—where should I go to find the kitchen paper?", + "answer": "You should go to the right side of the kitchen sink; it’s to your back-right.", + "question_chinese": "我想继续擦东西，我应该去哪里找到厨房纸？", + "answer_chinese": "你应该去厨房的洗手台右侧，在你的右后方。", + "evidence": [ + { + "description": "You got the kitchen paper from the right side of the kitchen sink.", + "timestamps": { + "start": "00:19:00", + "end": "00:19:11" + } + } + ], + "question_timestamps": "00:25:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You should go to the right side of the kitchen sink; it’s to your back-right.", + "B": "It’s to your front-right.", + "C": "It’s to your back-left.", + "D": "Go to the left side of the kitchen sink; the paper is positioned to your back-right from there.", + "E": "Go to the right side of the kitchen sink; the paper is on your right side rather than back-right." + }, + "answer_label": "A", + "answer_text": "You should go to the right side of the kitchen sink; it’s to your back-right.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_4_T2", + "correct_answer": "You should go to the right side of the kitchen sink; it’s to your back-right.", + "original_options": { + "A": "You should go to the right side of the kitchen sink; it’s to your back-right.", + "B": "It’s to your front-right.", + "C": "It’s to your back-left.", + "D": "Go to the left side of the kitchen sink; the paper is positioned to your back-right from there.", + "E": "Go to the right side of the kitchen sink; the paper is on your right side rather than back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 19 + ] + }, + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_12", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I watered the plants?", + "answer": "You watered plants three times.", + "question_chinese": "我浇水了多少次？", + "answer_chinese": "你浇了三次水。", + "evidence": [ + { + "description": "You watered once.", + "timestamps": { + "start": "00:21:12", + "end": "00:21:13" + } + }, + { + "description": "You watered twice.", + "timestamps": { + "start": "00:21:22", + "end": "00:21:30" + } + } + ], + "question_timestamps": "00:21:40", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You watered plants three times.", + "B": "You watered plants twice.", + "C": "You watered plants four times.", + "D": "You watered the plants only twice so far, stopping after the second watering.", + "E": "You watered the plants four separate times, giving them an extra watering beyond the third." + }, + "answer_label": "A", + "answer_text": "You watered plants three times.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_12", + "correct_answer": "You watered plants three times.", + "original_options": { + "A": "You watered plants three times.", + "B": "You watered plants twice.", + "C": "You watered plants four times.", + "D": "You watered the plants only twice so far, stopping after the second watering.", + "E": "You watered the plants four separate times, giving them an extra watering beyond the third." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 15 + ] + }, + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_11", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I watered the plants?", + "answer": "You watered plants a total of seven times.", + "question_chinese": "我浇水了多少次？", + "answer_chinese": "你一共浇了七次水。", + "evidence": [ + { + "description": "You watered once.", + "timestamps": { + "start": "00:21:12", + "end": "00:21:13" + } + }, + { + "description": "You watered twice.", + "timestamps": { + "start": "00:21:22", + "end": "00:21:30" + } + }, + { + "description": "You watered three times.", + "timestamps": { + "start": "00:21:55", + "end": "00:22:00" + } + }, + { + "description": "You watered once.", + "timestamps": { + "start": "00:24:05", + "end": "00:24:14" + } + } + ], + "question_timestamps": "00:26:01", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You watered plants six times.", + "B": "You watered plants eight times.", + "C": "You watered plants a total of seven times.", + "D": "You watered the plants a total of six times, one fewer than seven overall.", + "E": "You watered the plants a total of eight times, adding one more watering past seven." + }, + "answer_label": "C", + "answer_text": "You watered plants a total of seven times.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_11", + "correct_answer": "You watered plants a total of seven times.", + "original_options": { + "A": "You watered plants six times.", + "B": "You watered plants eight times.", + "C": "You watered plants a total of seven times.", + "D": "You watered the plants a total of six times, one fewer than seven overall.", + "E": "You watered the plants a total of eight times, adding one more watering past seven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_5", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cups have I washed so far?", + "answer": "You washed two cups.", + "question_chinese": "现在我洗了多少个杯具？", + "answer_chinese": "你洗了两个杯子。", + "evidence": [ + { + "description": "You washed the first cup.", + "timestamps": { + "start": "00:07:29", + "end": "00:08:30" + } + }, + { + "description": "You washed the second cup.", + "timestamps": { + "start": "00:08:40", + "end": "00:08:51" + } + } + ], + "question_timestamps": "00:12:47", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You washed one cup.", + "B": "You washed three cups.", + "C": "You washed only one cup so far, leaving the rest unwashed at this point.", + "D": "You washed two cups.", + "E": "You washed three cups in total, cleaning one more cup than two." + }, + "answer_label": "D", + "answer_text": "You washed two cups.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_5", + "correct_answer": "You washed two cups.", + "original_options": { + "A": "You washed one cup.", + "B": "You washed three cups.", + "C": "You washed only one cup so far, leaving the rest unwashed at this point.", + "D": "You washed two cups.", + "E": "You washed three cups in total, cleaning one more cup than two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 12 + ] + }, + { + "video_uid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb", + "video_length": "00:26:01", + "qid": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_6", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cups have I washed so far?", + "answer": "You washed three cups.", + "question_chinese": "现在我洗了多少个杯具？", + "answer_chinese": "你洗了三个杯子。", + "evidence": [ + { + "description": "You washed the first cup.", + "timestamps": { + "start": "00:07:29", + "end": "00:08:30" + } + }, + { + "description": "You washed the second cup.", + "timestamps": { + "start": "00:08:40", + "end": "00:08:51" + } + }, + { + "description": "You washed the third cup.", + "timestamps": { + "start": "00:13:36", + "end": "00:13:46" + } + } + ], + "question_timestamps": "00:22:54", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You washed two cups.", + "B": "You washed four cups.", + "C": "You washed only two cups so far, not reaching three cups cleaned yet.", + "D": "You washed four cups in total, washing one additional cup beyond three.", + "E": "You washed three cups." + }, + "answer_label": "E", + "answer_text": "You washed three cups.", + "q_id": "2b8f76b9-e542-4413-9fbd-0f7710a39dfb_6", + "correct_answer": "You washed three cups.", + "original_options": { + "A": "You washed two cups.", + "B": "You washed four cups.", + "C": "You washed only two cups so far, not reaching three cups cleaned yet.", + "D": "You washed four cups in total, washing one additional cup beyond three.", + "E": "You washed three cups." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 12 + ] + } +] diff --git a/QAs/2bd1593c-ae88-417e-827f-3997201da229-mcq.json b/QAs/2bd1593c-ae88-417e-827f-3997201da229-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4412290d37fc564d79136b753a34b295b5041092 --- /dev/null +++ b/QAs/2bd1593c-ae88-417e-827f-3997201da229-mcq.json @@ -0,0 +1,789 @@ +[ + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_5_T2", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach out and touch the refrigerator?", + "answer": "Yes, the refrigerator is right next to you on your right, within arm's reach.", + "question_chinese": "在我现在的位置，伸手能碰到冰箱吗？", + "answer_chinese": "可以，冰箱就在你后面。", + "evidence": [ + { + "description": "Open the refrigerator and put some things inside.", + "timestamps": { + "start": "00:20:00", + "end": "00:20:25" + } + } + ], + "question_timestamps": "00:20:57", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, the refrigerator is not within arm's reach from where you are standing.", + "B": "Yes, the refrigerator is right next to you on your right, within arm's reach." + }, + "answer_label": "B", + "answer_text": "Yes, the refrigerator is right next to you on your right, within arm's reach.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_5_T2", + "correct_answer": "Yes, the refrigerator is right next to you on your right, within arm's reach.", + "original_options": { + "A": "No, the refrigerator is not within arm's reach from where you are standing.", + "B": "Yes, the refrigerator is right next to you on your right, within arm's reach." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_5_T1", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach out and touch the refrigerator?", + "answer": "Yes, the refrigerator is right next to you on your right, within arm's reach.", + "question_chinese": "在我现在的位置，伸手能碰到冰箱吗？", + "answer_chinese": "可以，冰箱就在你右边右手可及之处。", + "evidence": [ + { + "description": "Open the refrigerator and put some things inside.", + "timestamps": { + "start": "00:20:00", + "end": "00:20:25" + } + } + ], + "question_timestamps": "00:32:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, the refrigerator is not within arm's reach from where you are standing.", + "B": "Yes, the refrigerator is right next to you on your right, within arm's reach." + }, + "answer_label": "B", + "answer_text": "Yes, the refrigerator is right next to you on your right, within arm's reach.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_5_T1", + "correct_answer": "Yes, the refrigerator is right next to you on your right, within arm's reach.", + "original_options": { + "A": "No, the refrigerator is not within arm's reach from where you are standing.", + "B": "Yes, the refrigerator is right next to you on your right, within arm's reach." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_14", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there still plenty of cream in the large glass bowl?", + "answer": "Yes, the large glass bowl is half-full with cream.", + "question_chinese": "大的玻璃碗中还有充足的奶油吗？", + "answer_chinese": "有，大玻璃碗里装着半满的奶油。", + "evidence": [ + { + "description": "The large glass bowl contains cream.", + "timestamps": { + "start": "00:06:32", + "end": "00:06:37" + } + } + ], + "question_timestamps": "00:20:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, the large glass bowl is half-full with cream.", + "B": "No, only residual cream is left." + }, + "answer_label": "A", + "answer_text": "Yes, the large glass bowl is half-full with cream.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_14", + "correct_answer": "Yes, the large glass bowl is half-full with cream.", + "original_options": { + "A": "Yes, the large glass bowl is half-full with cream.", + "B": "No, only residual cream is left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_15", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there still plenty of cream in the large glass bowl?", + "answer": "No, only residual cream is left.", + "question_chinese": "大的玻璃碗中还有充足的奶油吗？", + "answer_chinese": "没有了，只剩下残留的奶油。", + "evidence": [ + { + "description": "The large glass bowl contains cream.", + "timestamps": { + "start": "00:06:32", + "end": "00:06:37" + } + }, + { + "description": "The remaining cream is scraped clean.", + "timestamps": { + "start": "00:27:20", + "end": "00:27:40" + } + } + ], + "question_timestamps": "00:33:32", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, the large glass bowl is half-full with cream.", + "B": "No, only residual cream is left." + }, + "answer_label": "B", + "answer_text": "No, only residual cream is left.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_15", + "correct_answer": "No, only residual cream is left.", + "original_options": { + "A": "Yes, the large glass bowl is half-full with cream.", + "B": "No, only residual cream is left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get from my current location to the location of the computer?", + "answer": "Walk three steps to your back-left and you will arrive.", + "question_chinese": "我如何从当前位置走到电脑所在的位置？", + "answer_chinese": "向你的左后方走三步即可到达。", + "evidence": [ + { + "description": "Walk from the sink to the computer.", + "timestamps": { + "start": "00:02:18", + "end": "00:02:24" + } + }, + { + "description": "The computer is above the table.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk three steps to your back-left and you will arrive.", + "B": "Walk three steps to your back-right.", + "C": "Walk two steps to your back-left.", + "D": "Walk three steps to your front-left, and you should reach the computer from there.", + "E": "Turn slightly to your left, then take four steps backward until you are at the computer." + }, + "answer_label": "A", + "answer_text": "Walk three steps to your back-left and you will arrive.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_3", + "correct_answer": "Walk three steps to your back-left and you will arrive.", + "original_options": { + "A": "Walk three steps to your back-left and you will arrive.", + "B": "Walk three steps to your back-right.", + "C": "Walk two steps to your back-left.", + "D": "Walk three steps to your front-left, and you should reach the computer from there.", + "E": "Turn slightly to your left, then take four steps backward until you are at the computer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 16 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get from my current location to the location of the computer?", + "answer": "No need to move. The computer is right above you.", + "question_chinese": "我如何从当前位置走到电脑所在的位置？", + "answer_chinese": "无需移动。电脑就在你的上方。", + "evidence": [ + { + "description": "The computer is above the table.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:18:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk straight ahead to reach it.", + "B": "Turn left; it's beside you.", + "C": "No need to move. The computer is right above you.", + "D": "Take a couple steps backward, and the computer will be behind you at that spot.", + "E": "Step to your right once and look forward; the computer should be right in front of you." + }, + "answer_label": "C", + "answer_text": "No need to move. The computer is right above you.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_1", + "correct_answer": "No need to move. The computer is right above you.", + "original_options": { + "A": "Walk straight ahead to reach it.", + "B": "Turn left; it's beside you.", + "C": "No need to move. The computer is right above you.", + "D": "Take a couple steps backward, and the computer will be behind you at that spot.", + "E": "Step to your right once and look forward; the computer should be right in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 17 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get from my current location to the location of the computer?", + "answer": "Take two small steps to your back-right.", + "question_chinese": "我如何从当前位置走到电脑所在的位置？", + "answer_chinese": "向你的右后方走两小步即可。", + "evidence": [ + { + "description": "Walk from the microwave to the computer.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:53" + } + }, + { + "description": "The computer is above the table.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:30:29", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Take two small steps to your back-left.", + "B": "Take one small step to your back-right.", + "C": "Take two small steps forward, keeping centered, and you will reach the computer’s position.", + "D": "Move diagonally forward-right with two short steps, then stop where the computer should be.", + "E": "Take two small steps to your back-right." + }, + "answer_label": "E", + "answer_text": "Take two small steps to your back-right.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_2", + "correct_answer": "Take two small steps to your back-right.", + "original_options": { + "A": "Take two small steps to your back-left.", + "B": "Take one small step to your back-right.", + "C": "Take two small steps forward, keeping centered, and you will reach the computer’s position.", + "D": "Move diagonally forward-right with two short steps, then stop where the computer should be.", + "E": "Take two small steps to your back-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 14 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_7", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where can I find more slices of bread?", + "answer": "You can get them by reaching out to your right side.", + "question_chinese": "我想要更多面包片去哪找？", + "answer_chinese": "你右手边伸手就可以搞到。", + "evidence": [ + { + "description": "There are many slices of bread on the cutting board.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:15" + } + } + ], + "question_timestamps": "00:10:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Reach out to your left side.", + "B": "Look straight ahead for them.", + "C": "You can get them by reaching out to your right side.", + "D": "Turn slightly to your left and reach forward; the slices should be within arm’s reach there.", + "E": "Check behind you by extending your hand backward; the extra slices are located there." + }, + "answer_label": "C", + "answer_text": "You can get them by reaching out to your right side.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_7", + "correct_answer": "You can get them by reaching out to your right side.", + "original_options": { + "A": "Reach out to your left side.", + "B": "Look straight ahead for them.", + "C": "You can get them by reaching out to your right side.", + "D": "Turn slightly to your left and reach forward; the slices should be within arm’s reach there.", + "E": "Check behind you by extending your hand backward; the extra slices are located there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 14 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_8", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where can I find more slices of bread?", + "answer": "They are to your back-left.", + "question_chinese": "我想要更多面包片去哪找？", + "answer_chinese": "你左后方。", + "evidence": [ + { + "description": "There are many slices of bread on the cutting board.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:15" + } + } + ], + "question_timestamps": "00:20:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They are directly in front of you.", + "B": "They are to your front-right.", + "C": "Look to your left side; the extra slices should be near you rather than behind you.", + "D": "They are to your back-left.", + "E": "Turn around and scan to your back-right; the bread slices are positioned in that direction." + }, + "answer_label": "D", + "answer_text": "They are to your back-left.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_8", + "correct_answer": "They are to your back-left.", + "original_options": { + "A": "They are directly in front of you.", + "B": "They are to your front-right.", + "C": "Look to your left side; the extra slices should be near you rather than behind you.", + "D": "They are to your back-left.", + "E": "Turn around and scan to your back-right; the bread slices are positioned in that direction." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 15 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_9", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where can I find more slices of bread?", + "answer": "They are behind you.", + "question_chinese": "我想要更多面包片去哪找？", + "answer_chinese": "在你后边。", + "evidence": [ + { + "description": "There are many slices of bread on the cutting board.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:15" + } + } + ], + "question_timestamps": "00:30:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They are in front of you.", + "B": "They are behind you.", + "C": "They are to your left.", + "D": "Look up above you; the slices are stored higher than your current line of sight.", + "E": "Turn slightly right and reach forward; the bread is located beside you, not behind." + }, + "answer_label": "B", + "answer_text": "They are behind you.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_9", + "correct_answer": "They are behind you.", + "original_options": { + "A": "They are in front of you.", + "B": "They are behind you.", + "C": "They are to your left.", + "D": "Look up above you; the slices are stored higher than your current line of sight.", + "E": "Turn slightly right and reach forward; the bread is located beside you, not behind." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 14 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_4", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the cabinet I opened for the first time?", + "answer": "It is behind you.", + "question_chinese": "我第一次打开的橱柜在我哪个方位？", + "answer_chinese": "在你后面。", + "evidence": [ + { + "description": "Open the cabinet for the first time.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:05:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is to your right.", + "C": "Turn to your left and you should be facing the cabinet you opened first.", + "D": "It is behind you.", + "E": "Look down and slightly forward; the first cabinet is located lower and ahead of you." + }, + "answer_label": "D", + "answer_text": "It is behind you.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_4", + "correct_answer": "It is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is to your right.", + "C": "Turn to your left and you should be facing the cabinet you opened first.", + "D": "It is behind you.", + "E": "Look down and slightly forward; the first cabinet is located lower and ahead of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 15 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_5", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the cabinet I opened for the first time?", + "answer": "It is above you.", + "question_chinese": "我第一次打开的橱柜在我哪个方位？", + "answer_chinese": "在你上面。", + "evidence": [ + { + "description": "Open the cabinet for the first time.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:27:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is below you.", + "B": "It is above you.", + "C": "It is behind you.", + "D": "Turn to your right and the cabinet you opened first should be at your side.", + "E": "Look straight ahead at eye level; the first cabinet is positioned in front of you, not overhead." + }, + "answer_label": "B", + "answer_text": "It is above you.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_5", + "correct_answer": "It is above you.", + "original_options": { + "A": "It is below you.", + "B": "It is above you.", + "C": "It is behind you.", + "D": "Turn to your right and the cabinet you opened first should be at your side.", + "E": "Look straight ahead at eye level; the first cabinet is positioned in front of you, not overhead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_6", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the cabinet I opened for the first time?", + "answer": "It is to your left.", + "question_chinese": "我第一次打开的橱柜在我哪个方位？", + "answer_chinese": "在你左边。", + "evidence": [ + { + "description": "Open the cabinet for the first time.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:30:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is to your left.", + "B": "It is to your right.", + "C": "It is behind you.", + "D": "It is directly in front of you when you face the counter, not off to either side.", + "E": "It is above you, positioned higher up so you have to look up to open it." + }, + "answer_label": "A", + "answer_text": "It is to your left.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_6", + "correct_answer": "It is to your left.", + "original_options": { + "A": "It is to your left.", + "B": "It is to your right.", + "C": "It is behind you.", + "D": "It is directly in front of you when you face the counter, not off to either side.", + "E": "It is above you, positioned higher up so you have to look up to open it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 16 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_10", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where is the small, round bowl-shaped glass container placed?", + "answer": "It is in the porcelain plate under your hand.", + "question_chinese": "现在，小号的圆碗形玻璃容器放在哪了？", + "answer_chinese": "在你手底下的瓷盘里。", + "evidence": [ + { + "description": "Put the glass bowl on the table.", + "timestamps": { + "start": "00:04:01", + "end": "00:04:04" + } + } + ], + "question_timestamps": "00:10:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is beside your hand.", + "B": "It is on the countertop.", + "C": "It is sitting on the countertop near the stove rather than inside any plate under your hand.", + "D": "It is tucked away in the drawer beneath your desk, not placed in the porcelain plate.", + "E": "It is in the porcelain plate under your hand." + }, + "answer_label": "E", + "answer_text": "It is in the porcelain plate under your hand.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_10", + "correct_answer": "It is in the porcelain plate under your hand.", + "original_options": { + "A": "It is beside your hand.", + "B": "It is on the countertop.", + "C": "It is sitting on the countertop near the stove rather than inside any plate under your hand.", + "D": "It is tucked away in the drawer beneath your desk, not placed in the porcelain plate.", + "E": "It is in the porcelain plate under your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "2bd1593c-ae88-417e-827f-3997201da229", + "video_length": "00:35:28", + "qid": "2bd1593c-ae88-417e-827f-3997201da229_11", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where is the small, round bowl-shaped glass container placed?", + "answer": "It is in the refrigerator behind you.", + "question_chinese": "现在，小号的圆碗形玻璃容器放在哪了？", + "answer_chinese": "在你身后的冰箱里。", + "evidence": [ + { + "description": "Put the glass bowl on the table.", + "timestamps": { + "start": "00:04:01", + "end": "00:04:04" + } + }, + { + "description": "Put the glass bowl into the refrigerator.", + "timestamps": { + "start": "00:19:30", + "end": "00:19:55" + } + } + ], + "question_timestamps": "00:29:11", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It is in the refrigerator behind you.", + "B": "It is on the counter.", + "C": "It is in a drawer.", + "D": "It is in the cabinet beside the stove, not inside the refrigerator behind you.", + "E": "It is on the shelf in front of you, placed out in the open rather than refrigerated." + }, + "answer_label": "A", + "answer_text": "It is in the refrigerator behind you.", + "q_id": "2bd1593c-ae88-417e-827f-3997201da229_11", + "correct_answer": "It is in the refrigerator behind you.", + "original_options": { + "A": "It is in the refrigerator behind you.", + "B": "It is on the counter.", + "C": "It is in a drawer.", + "D": "It is in the cabinet beside the stove, not inside the refrigerator behind you.", + "E": "It is on the shelf in front of you, placed out in the open rather than refrigerated." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 17 + ] + } +] diff --git a/QAs/2c0c6508-397f-4c48-aeb7-abc7a3cae8d1-mcq.json b/QAs/2c0c6508-397f-4c48-aeb7-abc7a3cae8d1-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..baff9e703e46ded93c60c5cb937a30ae1e8d6d70 --- /dev/null +++ b/QAs/2c0c6508-397f-4c48-aeb7-abc7a3cae8d1-mcq.json @@ -0,0 +1,1481 @@ +[ + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_7_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The fridge is on the left side of you now.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱现在在你的左侧。", + "evidence": [ + { + "description": "The refrigerator is now on your left.", + "timestamps": { + "start": "00:28:16", + "end": "00:28:22" + } + } + ], + "question_timestamps": "00:28:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is to your right.", + "C": "The fridge is behind you now, so you would need to turn around to face it.", + "D": "The fridge is directly in front of you at the moment, not positioned on your left side.", + "E": "The fridge is on the left side of you now." + }, + "answer_label": "E", + "answer_text": "The fridge is on the left side of you now.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_7_T1", + "correct_answer": "The fridge is on the left side of you now.", + "original_options": { + "A": "It is in front of you.", + "B": "It is to your right.", + "C": "The fridge is behind you now, so you would need to turn around to face it.", + "D": "The fridge is directly in front of you at the moment, not positioned on your left side.", + "E": "The fridge is on the left side of you now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_7_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The refrigerator is on the left side of you.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱在你的左侧。", + "evidence": [ + { + "description": "The refrigerator is now on your left.", + "timestamps": { + "start": "00:28:16", + "end": "00:28:22" + } + } + ], + "question_timestamps": "00:31:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is on the left side of you.", + "B": "It is to your right.", + "C": "It is in front of you.", + "D": "The refrigerator is behind you right now, meaning you have to turn around to reach it.", + "E": "The refrigerator is directly in front of you, so you can see it without turning left." + }, + "answer_label": "A", + "answer_text": "The refrigerator is on the left side of you.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_7_T2", + "correct_answer": "The refrigerator is on the left side of you.", + "original_options": { + "A": "The refrigerator is on the left side of you.", + "B": "It is to your right.", + "C": "It is in front of you.", + "D": "The refrigerator is behind you right now, meaning you have to turn around to reach it.", + "E": "The refrigerator is directly in front of you, so you can see it without turning left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_6_T2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was stir-frying just now, what was the closest object to me?", + "answer": "The closest thing to you is the rag. The rag is in front of your knees.", + "question_chinese": "刚刚我炒菜时，距离我最近的物品是什么？", + "answer_chinese": "距离你最近的是抹布。抹布在你的膝盖前面。", + "evidence": [ + { + "description": "The nearest thing to you is the rag.", + "timestamps": { + "start": "00:24:55", + "end": "00:25:01" + } + } + ], + "question_timestamps": "00:25:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Rag behind knees.", + "B": "Rag to your left.", + "C": "The closest thing to you is the rag, but it is positioned behind your knees instead of in front.", + "D": "The closest thing to you is the rag, located to the right of your knees rather than straight ahead.", + "E": "The closest thing to you is the rag. The rag is in front of your knees." + }, + "answer_label": "E", + "answer_text": "The closest thing to you is the rag. The rag is in front of your knees.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_6_T2", + "correct_answer": "The closest thing to you is the rag. The rag is in front of your knees.", + "original_options": { + "A": "Rag behind knees.", + "B": "Rag to your left.", + "C": "The closest thing to you is the rag, but it is positioned behind your knees instead of in front.", + "D": "The closest thing to you is the rag, located to the right of your knees rather than straight ahead.", + "E": "The closest thing to you is the rag. The rag is in front of your knees." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 19, + 19 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_6_T1", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was stir-frying just now, what was the closest object to me?", + "answer": "The closest thing to you is the kitchen knife, and the kitchen knife is to your left.", + "question_chinese": "刚刚我炒菜时，距离我最近的物品是什么？", + "answer_chinese": "距离你最近的是菜刀，菜刀在你的左边。", + "evidence": [ + { + "description": "The nearest thing to you is the rag.", + "timestamps": { + "start": "00:24:55", + "end": "00:25:01" + } + } + ], + "question_timestamps": "00:25:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to your right.", + "B": "It is in front of you.", + "C": "The closest thing to you is the kitchen knife, and the kitchen knife is to your left.", + "D": "The closest thing to you is the kitchen knife, and it is positioned behind you.", + "E": "The closest thing to you is the kitchen knife, but it is on your right side rather than your left." + }, + "answer_label": "C", + "answer_text": "The closest thing to you is the kitchen knife, and the kitchen knife is to your left.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_6_T1", + "correct_answer": "The closest thing to you is the kitchen knife, and the kitchen knife is to your left.", + "original_options": { + "A": "It is to your right.", + "B": "It is in front of you.", + "C": "The closest thing to you is the kitchen knife, and the kitchen knife is to your left.", + "D": "The closest thing to you is the kitchen knife, and it is positioned behind you.", + "E": "The closest thing to you is the kitchen knife, but it is on your right side rather than your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 20 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_12", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What was the nearest item to me when I was cooking just now?", + "answer": "The nearest thing to you is the rag.", + "question_chinese": "刚刚我炒菜时，距离我最近的物品是什么？", + "answer_chinese": "距离你最近的是抹布。", + "evidence": [ + { + "description": "The nearest thing to you is the rag.", + "timestamps": { + "start": "00:24:55", + "end": "00:25:01" + } + } + ], + "question_timestamps": "00:25:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The nearest thing is the cutting board.", + "B": "The nearest thing to you is the rag.", + "C": "The nearest thing is the frying pan.", + "D": "The nearest item to you was the spatula, closer than anything else while you were cooking.", + "E": "The nearest thing to you was a bowl of fruit sitting nearby, closer than the rag." + }, + "answer_label": "B", + "answer_text": "The nearest thing to you is the rag.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_12", + "correct_answer": "The nearest thing to you is the rag.", + "original_options": { + "A": "The nearest thing is the cutting board.", + "B": "The nearest thing to you is the rag.", + "C": "The nearest thing is the frying pan.", + "D": "The nearest item to you was the spatula, closer than anything else while you were cooking.", + "E": "The nearest thing to you was a bowl of fruit sitting nearby, closer than the rag." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_10_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the roll of toilet paper relative to me right now?", + "answer": "The toilet paper roll is on the left side of you.", + "question_chinese": "卷纸现在相对我是什么位置？", + "answer_chinese": "卷纸在你的左侧。", + "evidence": [ + { + "description": "The roll of paper is on the table above the refrigerator.", + "timestamps": { + "start": "00:20:51", + "end": "00:21:09" + } + } + ], + "question_timestamps": "00:29:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is on your right.", + "B": "It is in front of you.", + "C": "The toilet paper roll is on the left side of you.", + "D": "The toilet paper roll is behind you, closer to your back than to either side.", + "E": "The toilet paper roll is to your right, not on the left side where you are facing." + }, + "answer_label": "C", + "answer_text": "The toilet paper roll is on the left side of you.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_10_T1", + "correct_answer": "The toilet paper roll is on the left side of you.", + "original_options": { + "A": "It is on your right.", + "B": "It is in front of you.", + "C": "The toilet paper roll is on the left side of you.", + "D": "The toilet paper roll is behind you, closer to your back than to either side.", + "E": "The toilet paper roll is to your right, not on the left side where you are facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 17 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_10_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the roll of toilet paper relative to me right now?", + "answer": "The toilet paper roll is on the left side of you.", + "question_chinese": "卷纸现在相对我是什么位置？", + "answer_chinese": "卷纸在你的左侧。", + "evidence": [ + { + "description": "The roll of paper is on the table above the refrigerator.", + "timestamps": { + "start": "00:20:51", + "end": "00:21:09" + } + } + ], + "question_timestamps": "00:29:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is on your right.", + "B": "It is behind you.", + "C": "The toilet paper roll is in front of you, directly ahead rather than off to either side.", + "D": "The toilet paper roll is to your right side, opposite from the left side you’d expect.", + "E": "The toilet paper roll is on the left side of you." + }, + "answer_label": "E", + "answer_text": "The toilet paper roll is on the left side of you.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_10_T2", + "correct_answer": "The toilet paper roll is on the left side of you.", + "original_options": { + "A": "It is on your right.", + "B": "It is behind you.", + "C": "The toilet paper roll is in front of you, directly ahead rather than off to either side.", + "D": "The toilet paper roll is to your right side, opposite from the left side you’d expect.", + "E": "The toilet paper roll is on the left side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 17 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_5", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What position is the frying pan relative to you?", + "answer": "The pan is on your right.", + "question_chinese": "平底锅相对你是什么位置？", + "answer_chinese": "平底锅在你的右侧。", + "evidence": [ + { + "description": "The pan is on your right.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:31" + } + } + ], + "question_timestamps": "00:09:36", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The pan is on your left.", + "B": "The pan is in front of you.", + "C": "The pan is on your right.", + "D": "The pan is behind you, not positioned to your right side at the moment.", + "E": "The pan is slightly to your left, so it isn’t on your right relative to you." + }, + "answer_label": "C", + "answer_text": "The pan is on your right.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_5", + "correct_answer": "The pan is on your right.", + "original_options": { + "A": "The pan is on your left.", + "B": "The pan is in front of you.", + "C": "The pan is on your right.", + "D": "The pan is behind you, not positioned to your right side at the moment.", + "E": "The pan is slightly to your left, so it isn’t on your right relative to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 17 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_6", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What position is the frying pan relative to you?", + "answer": "The pan is on your right.", + "question_chinese": "平底锅相对你是什么位置？", + "answer_chinese": "平底锅在你的右侧。", + "evidence": [ + { + "description": "The pan is on your right.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:31" + } + } + ], + "question_timestamps": "00:14:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The pan is on your left.", + "B": "The pan is behind you.", + "C": "The pan is in front of you, rather than positioned on your right side.", + "D": "The pan is on your right.", + "E": "The pan is slightly to your left, so it is not on your right relative to you." + }, + "answer_label": "D", + "answer_text": "The pan is on your right.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_6", + "correct_answer": "The pan is on your right.", + "original_options": { + "A": "The pan is on your left.", + "B": "The pan is behind you.", + "C": "The pan is in front of you, rather than positioned on your right side.", + "D": "The pan is on your right.", + "E": "The pan is slightly to your left, so it is not on your right relative to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 17 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_3_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the frying pan relative to me?", + "answer": "The frying pan is on the right side of you.", + "question_chinese": "平底锅相对我是什么位置？", + "answer_chinese": "平底锅相对于你的右侧。", + "evidence": [ + { + "description": "The pan is on your right.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:31" + } + }, + { + "description": "The pan is on your right.", + "timestamps": { + "start": "00:09:32", + "end": "00:09:36" + } + } + ], + "question_timestamps": "00:09:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is on your left.", + "B": "It is in front of you.", + "C": "The frying pan is on the right side of you.", + "D": "The frying pan is behind you, not on the right side relative to your body.", + "E": "The frying pan is slightly to your left, so it isn’t located on your right side." + }, + "answer_label": "C", + "answer_text": "The frying pan is on the right side of you.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_3_T1", + "correct_answer": "The frying pan is on the right side of you.", + "original_options": { + "A": "It is on your left.", + "B": "It is in front of you.", + "C": "The frying pan is on the right side of you.", + "D": "The frying pan is behind you, not on the right side relative to your body.", + "E": "The frying pan is slightly to your left, so it isn’t located on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 17 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_3_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the frying pan relative to me?", + "answer": "The frying pan is on the right side of you.", + "question_chinese": "平底锅相对我是什么位置？", + "answer_chinese": "平底锅在你的右侧。", + "evidence": [ + { + "description": "The pan is on your right.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:31" + } + } + ], + "question_timestamps": "00:14:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is on your left.", + "B": "It is behind you.", + "C": "The frying pan is in front of you, not on the right side relative to you.", + "D": "The frying pan is on the right side of you.", + "E": "The frying pan is positioned to your left side, so it is not on your right side." + }, + "answer_label": "D", + "answer_text": "The frying pan is on the right side of you.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_3_T2", + "correct_answer": "The frying pan is on the right side of you.", + "original_options": { + "A": "It is on your left.", + "B": "It is behind you.", + "C": "The frying pan is in front of you, not on the right side relative to you.", + "D": "The frying pan is on the right side of you.", + "E": "The frying pan is positioned to your left side, so it is not on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 17 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_7", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cherry tomatoes are left in the bowl after I just cut them?", + "answer": "There are still sixteen cherry tomatoes left.", + "question_chinese": "我刚刚切了圣女果后，碗里还剩下几个圣女果？", + "answer_chinese": "还剩16个圣女果。", + "evidence": [ + { + "description": "There are still 11 cherry tomatoes left.", + "timestamps": { + "start": "00:16:43", + "end": "00:16:54" + } + } + ], + "question_timestamps": "00:16:56", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are still sixteen cherry tomatoes left.", + "B": "There are fourteen left.", + "C": "There are twenty left.", + "D": "There are still ten cherry tomatoes left in the bowl after you finished cutting them.", + "E": "There are still eight cherry tomatoes left in the bowl once you are done cutting them." + }, + "answer_label": "A", + "answer_text": "There are still sixteen cherry tomatoes left.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_7", + "correct_answer": "There are still sixteen cherry tomatoes left.", + "original_options": { + "A": "There are still sixteen cherry tomatoes left.", + "B": "There are fourteen left.", + "C": "There are twenty left.", + "D": "There are still ten cherry tomatoes left in the bowl after you finished cutting them.", + "E": "There are still eight cherry tomatoes left in the bowl once you are done cutting them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_8", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cherry tomatoes are left in the bowl after I just cut them?", + "answer": "There are still eleven cherry tomatoes left.", + "question_chinese": "我刚刚切了圣女果后，碗里还剩下几个圣女果？", + "answer_chinese": "还剩11个圣女果。", + "evidence": [ + { + "description": "There are still 11 cherry tomatoes left.", + "timestamps": { + "start": "00:16:43", + "end": "00:16:54" + } + } + ], + "question_timestamps": "00:21:39", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are five left.", + "B": "There are still eleven cherry tomatoes left.", + "C": "There are eight left.", + "D": "There are still fifteen cherry tomatoes left in the bowl after you just cut them.", + "E": "There are only three cherry tomatoes left in the bowl after you finished cutting them." + }, + "answer_label": "B", + "answer_text": "There are still eleven cherry tomatoes left.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_8", + "correct_answer": "There are still eleven cherry tomatoes left.", + "original_options": { + "A": "There are five left.", + "B": "There are still eleven cherry tomatoes left.", + "C": "There are eight left.", + "D": "There are still fifteen cherry tomatoes left in the bowl after you just cut them.", + "E": "There are only three cherry tomatoes left in the bowl after you finished cutting them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I just opened the refrigerator?", + "answer": "You just opened the refrigerator once.", + "question_chinese": "我刚刚打开了冰箱几次？", + "answer_chinese": "你刚刚打开冰箱一次。", + "evidence": [ + { + "description": "You just opened the refrigerator once.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You just opened the refrigerator once.", + "B": "You just opened the refrigerator twice.", + "C": "You just opened the refrigerator three times.", + "D": "You just opened the refrigerator two times during this moment, opening and closing the door again.", + "E": "You just opened the refrigerator four times in total, repeating the same open-door action several times." + }, + "answer_label": "A", + "answer_text": "You just opened the refrigerator once.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_1", + "correct_answer": "You just opened the refrigerator once.", + "original_options": { + "A": "You just opened the refrigerator once.", + "B": "You just opened the refrigerator twice.", + "C": "You just opened the refrigerator three times.", + "D": "You just opened the refrigerator two times during this moment, opening and closing the door again.", + "E": "You just opened the refrigerator four times in total, repeating the same open-door action several times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 16 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I just opened the refrigerator?", + "answer": "You just opened the refrigerator once.", + "question_chinese": "我刚刚打开了冰箱几次？", + "answer_chinese": "你刚刚打开了一次冰箱。", + "evidence": [ + { + "description": "You just opened the refrigerator once.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:10:48", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You just opened the refrigerator twice.", + "B": "You just opened the refrigerator four times.", + "C": "You just opened the refrigerator three times, reopening it after closing the door each time.", + "D": "You just opened the refrigerator once.", + "E": "You just opened the refrigerator five times in total, repeatedly pulling the door open again." + }, + "answer_label": "D", + "answer_text": "You just opened the refrigerator once.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_2", + "correct_answer": "You just opened the refrigerator once.", + "original_options": { + "A": "You just opened the refrigerator twice.", + "B": "You just opened the refrigerator four times.", + "C": "You just opened the refrigerator three times, reopening it after closing the door each time.", + "D": "You just opened the refrigerator once.", + "E": "You just opened the refrigerator five times in total, repeatedly pulling the door open again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 15 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_15", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the spoon?", + "answer": "You just put the spoon on the cutting board.", + "question_chinese": "我刚刚把勺子放在什么地方？", + "answer_chinese": "你刚刚把勺子放在砧板上了。", + "evidence": [ + { + "description": "You just put the spoon on the cutting board.", + "timestamps": { + "start": "00:33:30", + "end": "00:33:33" + } + } + ], + "question_timestamps": "00:33:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the kitchen counter.", + "B": "In the drawer near the fridge.", + "C": "You just put the spoon next to the pot on the stove, leaving it beside the cookware.", + "D": "You just put the spoon behind the blender on the shelf, tucked out of the way.", + "E": "You just put the spoon on the cutting board." + }, + "answer_label": "E", + "answer_text": "You just put the spoon on the cutting board.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_15", + "correct_answer": "You just put the spoon on the cutting board.", + "original_options": { + "A": "On the kitchen counter.", + "B": "In the drawer near the fridge.", + "C": "You just put the spoon next to the pot on the stove, leaving it beside the cookware.", + "D": "You just put the spoon behind the blender on the shelf, tucked out of the way.", + "E": "You just put the spoon on the cutting board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 17, + 16 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_16", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the spoon?", + "answer": "You just put the spoon on the cutting board.", + "question_chinese": "我刚刚把勺子放在什么地方？", + "answer_chinese": "你刚刚把勺子放在砧板上了。", + "evidence": [ + { + "description": "You just put the spoon on the cutting board.", + "timestamps": { + "start": "00:33:30", + "end": "00:33:33" + } + } + ], + "question_timestamps": "00:35:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just put the spoon on the cutting board.", + "B": "In the kitchen drawer.", + "C": "On the countertop.", + "D": "You just put the spoon next to the sink, leaving it on the edge by the basin.", + "E": "You just put the spoon behind the refrigerator, placing it out of sight near the back." + }, + "answer_label": "A", + "answer_text": "You just put the spoon on the cutting board.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_16", + "correct_answer": "You just put the spoon on the cutting board.", + "original_options": { + "A": "You just put the spoon on the cutting board.", + "B": "In the kitchen drawer.", + "C": "On the countertop.", + "D": "You just put the spoon next to the sink, leaving it on the edge by the basin.", + "E": "You just put the spoon behind the refrigerator, placing it out of sight near the back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 16 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_1_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many avocados have I cut so far?", + "answer": "You’ve cut one avocado now, and the avocado is directly in front of you.", + "question_chinese": "我现在切了几个牛油果?", + "answer_chinese": "你现在切了一个牛油果，牛油果在你的正前方。", + "evidence": [ + { + "description": "You only cut an avocado now.", + "timestamps": { + "start": "00:07:39", + "end": "00:09:16" + } + } + ], + "question_timestamps": "00:09:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve cut zero avocados so far.", + "B": "You’ve cut two avocados so far.", + "C": "You’ve cut zero avocados now, and the avocado is directly in front of you, still uncut.", + "D": "You’ve cut three avocados now, and the avocados are directly in front of you, grouped together.", + "E": "You’ve cut one avocado now, and the avocado is directly in front of you." + }, + "answer_label": "E", + "answer_text": "You’ve cut one avocado now, and the avocado is directly in front of you.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_1_T2", + "correct_answer": "You’ve cut one avocado now, and the avocado is directly in front of you.", + "original_options": { + "A": "You’ve cut zero avocados so far.", + "B": "You’ve cut two avocados so far.", + "C": "You’ve cut zero avocados now, and the avocado is directly in front of you, still uncut.", + "D": "You’ve cut three avocados now, and the avocados are directly in front of you, grouped together.", + "E": "You’ve cut one avocado now, and the avocado is directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_1_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many avocados have I cut so far?", + "answer": "You’ve cut two avocados now, and the avocados are directly in front of you.", + "question_chinese": "我现在切了几个牛油果?", + "answer_chinese": "你现在切了两个牛油果，牛油果在你的正前方。", + "evidence": [ + { + "description": "first one", + "timestamps": { + "start": "00:07:39", + "end": "00:09:18" + } + }, + { + "description": "second one", + "timestamps": { + "start": "00:09:51", + "end": "00:11:12" + } + } + ], + "question_timestamps": "00:11:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve cut one avocado so far.", + "B": "You’ve cut three avocados so far.", + "C": "You’ve cut zero avocados now, and the avocados are directly in front of you, untouched.", + "D": "You’ve cut two avocados now, and the avocados are directly in front of you.", + "E": "You’ve cut four avocados now, and the avocados are directly in front of you, spread out on the surface." + }, + "answer_label": "D", + "answer_text": "You’ve cut two avocados now, and the avocados are directly in front of you.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_1_T1", + "correct_answer": "You’ve cut two avocados now, and the avocados are directly in front of you.", + "original_options": { + "A": "You’ve cut one avocado so far.", + "B": "You’ve cut three avocados so far.", + "C": "You’ve cut zero avocados now, and the avocados are directly in front of you, untouched.", + "D": "You’ve cut two avocados now, and the avocados are directly in front of you.", + "E": "You’ve cut four avocados now, and the avocados are directly in front of you, spread out on the surface." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 20 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_3", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the trash can on the table directly now?", + "answer": "You can't make direct contact now.", + "question_chinese": "我现在可以直接触碰到桌上的垃圾桶吗？", + "answer_chinese": "你现在无法直接触碰到。", + "evidence": [ + { + "description": "The trash can is located near the door, at a certain distance from where you are now.", + "timestamps": { + "start": "00:13:29", + "end": "00:13:33" + } + } + ], + "question_timestamps": "00:13:37", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can make direct contact now.", + "B": "You can't make direct contact now." + }, + "answer_label": "B", + "answer_text": "You can't make direct contact now.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_3", + "correct_answer": "You can't make direct contact now.", + "original_options": { + "A": "You can make direct contact now.", + "B": "You can't make direct contact now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_4", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the trash can on the table directly now?", + "answer": "You can't touch the trash can directly.", + "question_chinese": "我现在可以直接触碰到桌上的垃圾桶吗？", + "answer_chinese": "你无法直接触碰到垃圾桶。", + "evidence": [ + { + "description": "The trash can is located near the door, at a certain distance from where you are now.", + "timestamps": { + "start": "00:13:29", + "end": "00:13:33" + } + } + ], + "question_timestamps": "00:13:48", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't touch the trash can directly.", + "B": "You can touch the trash can directly now." + }, + "answer_label": "A", + "answer_text": "You can't touch the trash can directly.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_4", + "correct_answer": "You can't touch the trash can directly.", + "original_options": { + "A": "You can't touch the trash can directly.", + "B": "You can touch the trash can directly now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_17", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the microwave oven directly now?", + "answer": "You can touch the microwave oven directly.", + "question_chinese": "我现在能直接触碰到微波炉吗？", + "answer_chinese": "你可以直接触碰到微波炉。", + "evidence": [ + { + "description": "You can touch the microwave oven directly.", + "timestamps": { + "start": "00:08:51", + "end": "00:08:55" + } + } + ], + "question_timestamps": "00:10:13", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch the microwave oven directly.", + "B": "You cannot touch the microwave oven directly." + }, + "answer_label": "A", + "answer_text": "You can touch the microwave oven directly.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_17", + "correct_answer": "You can touch the microwave oven directly.", + "original_options": { + "A": "You can touch the microwave oven directly.", + "B": "You cannot touch the microwave oven directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_18", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the microwave oven directly now?", + "answer": "You can touch the microwave oven directly.", + "question_chinese": "我现在能直接触碰到微波炉吗？", + "answer_chinese": "你可以直接触碰到微波炉。", + "evidence": [ + { + "description": "You can touch the microwave oven directly.", + "timestamps": { + "start": "00:08:51", + "end": "00:08:55" + } + } + ], + "question_timestamps": "00:16:37", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch the microwave oven directly.", + "B": "You cannot touch the microwave oven directly." + }, + "answer_label": "A", + "answer_text": "You can touch the microwave oven directly.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_18", + "correct_answer": "You can touch the microwave oven directly.", + "original_options": { + "A": "You can touch the microwave oven directly.", + "B": "You cannot touch the microwave oven directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_10", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the rag directly now?", + "answer": "You can't touch the rag directly.", + "question_chinese": "我现在能直接触碰到抹布吗？", + "answer_chinese": "你无法直接触碰到抹布。", + "evidence": [ + { + "description": "You can't touch the rag directly.", + "timestamps": { + "start": "00:24:14", + "end": "00:24:28" + } + } + ], + "question_timestamps": "00:24:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch the rag directly.", + "B": "You can't touch the rag directly." + }, + "answer_label": "B", + "answer_text": "You can't touch the rag directly.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_10", + "correct_answer": "You can't touch the rag directly.", + "original_options": { + "A": "You can touch the rag directly.", + "B": "You can't touch the rag directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_9", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the rag directly now?", + "answer": "You can touch the rag directly.", + "question_chinese": "我现在能直接触碰到抹布吗？", + "answer_chinese": "你可以直接触碰到抹布。", + "evidence": [ + { + "description": "You can't touch the rag directly.", + "timestamps": { + "start": "00:24:14", + "end": "00:24:28" + } + } + ], + "question_timestamps": "00:24:32", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch the rag directly.", + "B": "You cannot touch the rag directly." + }, + "answer_label": "A", + "answer_text": "You can touch the rag directly.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_9", + "correct_answer": "You can touch the rag directly.", + "original_options": { + "A": "You can touch the rag directly.", + "B": "You cannot touch the rag directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_9_T1", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the microwave right now?", + "answer": "Right now you can directly touch the microwave, and the microwave is to your left.", + "question_chinese": "我现在能触碰到微波炉吗？", + "answer_chinese": "你现在可以直接触碰到微波炉,微波炉在你的左边。", + "evidence": [ + { + "description": "You can touch the microwave oven directly.", + "timestamps": { + "start": "00:08:51", + "end": "00:08:53" + } + } + ], + "question_timestamps": "00:08:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can’t directly touch the microwave; the microwave is to your left.", + "B": "Right now you can directly touch the microwave, and the microwave is to your left." + }, + "answer_label": "B", + "answer_text": "Right now you can directly touch the microwave, and the microwave is to your left.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_9_T1", + "correct_answer": "Right now you can directly touch the microwave, and the microwave is to your left.", + "original_options": { + "A": "Right now you can’t directly touch the microwave; the microwave is to your left.", + "B": "Right now you can directly touch the microwave, and the microwave is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_9_T2", + "qid_num": 27, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the microwave right now?", + "answer": "Right now you can’t directly touch the microwave; the microwave is to your right.", + "question_chinese": "我现在能触碰到微波炉吗？", + "answer_chinese": "你现在不可以直接触碰到微波炉，微波炉在你的右边。", + "evidence": [ + { + "description": "You can't touch the microwave oven directly now", + "timestamps": { + "start": "00:11:37", + "end": "00:11:40" + } + } + ], + "question_timestamps": "00:11:42", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can’t directly touch the microwave; the microwave is to your right.", + "B": "Right now you can directly touch the microwave, and the microwave is to your right." + }, + "answer_label": "A", + "answer_text": "Right now you can’t directly touch the microwave; the microwave is to your right.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_9_T2", + "correct_answer": "Right now you can’t directly touch the microwave; the microwave is to your right.", + "original_options": { + "A": "Right now you can’t directly touch the microwave; the microwave is to your right.", + "B": "Right now you can directly touch the microwave, and the microwave is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_8_T1", + "qid_num": 28, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can on the tabletop relative to me?", + "answer": "The trash can on the tabletop is to your left.", + "question_chinese": "桌面上的垃圾桶在相对于我的哪里?", + "answer_chinese": "桌面上的垃圾桶在相对于你的左边。", + "evidence": [ + { + "description": "The trash can on the desktop is on the left relative to you.", + "timestamps": { + "start": "00:14:51", + "end": "00:14:54" + } + } + ], + "question_timestamps": "00:14:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s to your right.", + "C": "The trash can on the tabletop is directly in front of you, not off to your left.", + "D": "The trash can on the tabletop is to your right side, rather than being positioned to your left.", + "E": "The trash can on the tabletop is to your left." + }, + "answer_label": "E", + "answer_text": "The trash can on the tabletop is to your left.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_8_T1", + "correct_answer": "The trash can on the tabletop is to your left.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s to your right.", + "C": "The trash can on the tabletop is directly in front of you, not off to your left.", + "D": "The trash can on the tabletop is to your right side, rather than being positioned to your left.", + "E": "The trash can on the tabletop is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 18 + ] + }, + { + "video_uid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1", + "video_length": "00:35:57", + "qid": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_8_T2", + "qid_num": 29, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can on the tabletop relative to me?", + "answer": "The trash can on the tabletop is behind you.", + "question_chinese": "桌面上的垃圾桶在相对于我的哪里?", + "answer_chinese": "桌面上的垃圾桶在相对于你的后面。", + "evidence": [ + { + "description": "The trash can on the desktop is behind you.", + "timestamps": { + "start": "00:15:02", + "end": "00:15:08" + } + } + ], + "question_timestamps": "00:15:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s to your left.", + "C": "The trash can on the tabletop is behind you.", + "D": "The trash can on the tabletop is to your right, instead of being placed behind you.", + "E": "The trash can on the tabletop is in front of you on the tabletop, not located behind you." + }, + "answer_label": "C", + "answer_text": "The trash can on the tabletop is behind you.", + "q_id": "2c0c6508-397f-4c48-aeb7-abc7a3cae8d1_8_T2", + "correct_answer": "The trash can on the tabletop is behind you.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s to your left.", + "C": "The trash can on the tabletop is behind you.", + "D": "The trash can on the tabletop is to your right, instead of being placed behind you.", + "E": "The trash can on the tabletop is in front of you on the tabletop, not located behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 18 + ] + } +] diff --git a/QAs/2c27b5f1-4af6-49ad-a43c-3efb0c150868-mcq.json b/QAs/2c27b5f1-4af6-49ad-a43c-3efb0c150868-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..faf0bbee967986650b1a1aa2210451546b60b1b8 --- /dev/null +++ b/QAs/2c27b5f1-4af6-49ad-a43c-3efb0c150868-mcq.json @@ -0,0 +1,655 @@ +[ + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_2_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I face the sink, is the microwave on my left/right?", + "answer": "When you face the sink, the microwave is on your left.", + "question_chinese": "当我面对水槽时，微波炉在我的左/右？", + "answer_chinese": "当你面对水槽时，微波炉在你的左边。", + "evidence": [ + { + "description": "The microwave is to the left of the sink.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The microwave is on your right.", + "B": "The microwave is straight ahead.", + "C": "When you face the sink, the microwave is on your left.", + "D": "When you face the sink, the microwave is to your right, near the counter edge beside it.", + "E": "When you face the sink, the microwave is in front of you, slightly offset to the right." + }, + "answer_label": "C", + "answer_text": "When you face the sink, the microwave is on your left.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_2_T1", + "correct_answer": "When you face the sink, the microwave is on your left.", + "original_options": { + "A": "The microwave is on your right.", + "B": "The microwave is straight ahead.", + "C": "When you face the sink, the microwave is on your left.", + "D": "When you face the sink, the microwave is to your right, near the counter edge beside it.", + "E": "When you face the sink, the microwave is in front of you, slightly offset to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 17 + ] + }, + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_2_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I face the sink, is the microwave on my left/right?", + "answer": "When you face the sink, the microwave is on your left.", + "question_chinese": "当我面对水槽时，微波炉在我的左/右？", + "answer_chinese": "当你面对水槽时，微波炉在你的左边。", + "evidence": [ + { + "description": "The microwave is to the left of the sink.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:18:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The microwave is on your right.", + "B": "The microwave is behind you.", + "C": "When you face the sink, the microwave is on your right, not on your left.", + "D": "When you face the sink, the microwave is directly in front of you rather than off to either side.", + "E": "When you face the sink, the microwave is on your left." + }, + "answer_label": "E", + "answer_text": "When you face the sink, the microwave is on your left.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_2_T2", + "correct_answer": "When you face the sink, the microwave is on your left.", + "original_options": { + "A": "The microwave is on your right.", + "B": "The microwave is behind you.", + "C": "When you face the sink, the microwave is on your right, not on your left.", + "D": "When you face the sink, the microwave is directly in front of you rather than off to either side.", + "E": "When you face the sink, the microwave is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 19 + ] + }, + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I face the stove, how many pots are there on it?", + "answer": "When you face the stove, there are three pots on it.", + "question_chinese": "当我面对灶台，灶台上有几个锅？", + "answer_chinese": "当你面对灶台，灶台上有三个锅。", + "evidence": [ + { + "description": "There are three pots on the stove.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one pot on it.", + "B": "There are two pots on it.", + "C": "When you face the stove, there are three pots on it.", + "D": "When you face the stove, there are four pots on it, spaced across the burners.", + "E": "When you face the stove, there are five pots on it, with one sitting on each burner." + }, + "answer_label": "C", + "answer_text": "When you face the stove, there are three pots on it.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_1", + "correct_answer": "When you face the stove, there are three pots on it.", + "original_options": { + "A": "There is one pot on it.", + "B": "There are two pots on it.", + "C": "When you face the stove, there are three pots on it.", + "D": "When you face the stove, there are four pots on it, spaced across the burners.", + "E": "When you face the stove, there are five pots on it, with one sitting on each burner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I face the stove, how many pots are there on it?", + "answer": "When you face the stove, there is no pot on it.", + "question_chinese": "当我面对灶台，灶台上有几个锅？", + "answer_chinese": "当你面对灶台，灶台上没有锅。", + "evidence": [ + { + "description": "There are three pots on the stove.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + }, + { + "description": "I took a pot, washed it clean, and put it away.", + "timestamps": { + "start": "00:04:40", + "end": "00:05:04" + } + }, + { + "description": "I took the second pot.", + "timestamps": { + "start": "00:09:35", + "end": "00:09:38" + } + } + ], + "question_timestamps": "00:23:39", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There is one pot on it.", + "B": "There are two pots on it.", + "C": "When you face the stove, there are three pots on it, sitting across multiple burners.", + "D": "When you face the stove, there is no pot on it.", + "E": "When you face the stove, there are four pots on it, with the burners each holding one." + }, + "answer_label": "D", + "answer_text": "When you face the stove, there is no pot on it.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_2", + "correct_answer": "When you face the stove, there is no pot on it.", + "original_options": { + "A": "There is one pot on it.", + "B": "There are two pots on it.", + "C": "When you face the stove, there are three pots on it, sitting across multiple burners.", + "D": "When you face the stove, there is no pot on it.", + "E": "When you face the stove, there are four pots on it, with the burners each holding one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_10", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was standing in the kitchen, was the fork next to the knife?", + "answer": "When you stand in the kitchen, the fork is next to the knife.", + "question_chinese": "我站在厨房时，叉子在刀的旁边吗？", + "answer_chinese": "你站在厨房时，叉子在刀的旁边。", + "evidence": [ + { + "description": "I put the knife and fork next to the stove.", + "timestamps": { + "start": "00:06:04", + "end": "00:06:11" + } + } + ], + "question_timestamps": "00:08:39", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you stand in the kitchen, the fork is not next to the knife.", + "B": "When you stand in the kitchen, the fork is next to the knife." + }, + "answer_label": "B", + "answer_text": "When you stand in the kitchen, the fork is next to the knife.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_10", + "correct_answer": "When you stand in the kitchen, the fork is next to the knife.", + "original_options": { + "A": "When you stand in the kitchen, the fork is not next to the knife.", + "B": "When you stand in the kitchen, the fork is next to the knife." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_9", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was standing in the kitchen, was the fork next to the knife?", + "answer": "When you stand in the kitchen, the fork is not next to the knife.", + "question_chinese": "我站在厨房时，叉子在刀的旁边吗？", + "answer_chinese": "你站在厨房时，叉子不在刀的旁边。", + "evidence": [ + { + "description": "I put the fork into the pot and took it to the sink.", + "timestamps": { + "start": "00:09:34", + "end": "00:09:38" + } + } + ], + "question_timestamps": "00:14:22", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you stand in the kitchen, the fork is next to the knife.", + "B": "When you stand in the kitchen, the fork is not next to the knife." + }, + "answer_label": "B", + "answer_text": "When you stand in the kitchen, the fork is not next to the knife.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_9", + "correct_answer": "When you stand in the kitchen, the fork is not next to the knife.", + "original_options": { + "A": "When you stand in the kitchen, the fork is next to the knife.", + "B": "When you stand in the kitchen, the fork is not next to the knife." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_7_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I get more seasoning?", + "answer": "You can find the seasonings on the left side of the stove.", + "question_chinese": "我需要更多的调味料去哪里拿？", + "answer_chinese": "可以在灶台的左边和上面找到调味料。", + "evidence": [ + { + "description": "There are seasonings on the left side of the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "There are condiments on top of the stove.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:05:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the right side of the stove.", + "B": "You can find the seasonings on the left side of the stove.", + "C": "On the countertop beside the stove.", + "D": "You can find the seasonings in the cabinet above the stove, not on the left side.", + "E": "You can get more seasoning from under the sink area, stored below the counter nearby." + }, + "answer_label": "B", + "answer_text": "You can find the seasonings on the left side of the stove.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_7_T1", + "correct_answer": "You can find the seasonings on the left side of the stove.", + "original_options": { + "A": "On the right side of the stove.", + "B": "You can find the seasonings on the left side of the stove.", + "C": "On the countertop beside the stove.", + "D": "You can find the seasonings in the cabinet above the stove, not on the left side.", + "E": "You can get more seasoning from under the sink area, stored below the counter nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 15 + ] + }, + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_7_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I get more seasoning?", + "answer": "You can find the seasonings on the left side and above the stove.", + "question_chinese": "我需要更多的调味料去哪里拿？", + "answer_chinese": "可以在灶台的左边和上面找到调味料。", + "evidence": [ + { + "description": "There are seasonings on the left side of the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "There are condiments on top of the stove.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:20:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the right side below the counter.", + "B": "On the countertop by the microwave.", + "C": "You can find the seasonings on the left side and above the stove.", + "D": "You can find the seasonings on the left side of the stove, but down on the counter level.", + "E": "You can get more seasoning from the right side of the stove, stored above rather than left." + }, + "answer_label": "C", + "answer_text": "You can find the seasonings on the left side and above the stove.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_7_T2", + "correct_answer": "You can find the seasonings on the left side and above the stove.", + "original_options": { + "A": "On the right side below the counter.", + "B": "On the countertop by the microwave.", + "C": "You can find the seasonings on the left side and above the stove.", + "D": "You can find the seasonings on the left side of the stove, but down on the counter level.", + "E": "You can get more seasoning from the right side of the stove, stored above rather than left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_6", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bags of portioned food are placed next to the sink?", + "answer": "There are twelve bags of portioned food placed next to the sink.", + "question_chinese": "水槽旁边放了多少个分装了食物的袋子？", + "answer_chinese": "水槽旁边放了12个分装了食物的袋子。", + "evidence": [ + { + "description": "There are 12 bags of portioned food placed next to the sink.", + "timestamps": { + "start": "00:19:28", + "end": "00:19:34" + } + } + ], + "question_timestamps": "00:19:36", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are ten bags by the sink.", + "B": "There are fifteen bags by the sink.", + "C": "There are twelve bags of portioned food placed next to the sink.", + "D": "There are nine bags of portioned food placed next to the sink.", + "E": "There are fourteen bags of portioned food sitting together next to the sink." + }, + "answer_label": "C", + "answer_text": "There are twelve bags of portioned food placed next to the sink.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_6", + "correct_answer": "There are twelve bags of portioned food placed next to the sink.", + "original_options": { + "A": "There are ten bags by the sink.", + "B": "There are fifteen bags by the sink.", + "C": "There are twelve bags of portioned food placed next to the sink.", + "D": "There are nine bags of portioned food placed next to the sink.", + "E": "There are fourteen bags of portioned food sitting together next to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 13 + ] + }, + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_5", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bags of portioned food are placed next to the sink?", + "answer": "There are eleven bags of portioned food placed next to the sink.", + "question_chinese": "水槽旁边放了多少个分装了食物的袋子？", + "answer_chinese": "水槽旁边放了11个分装了食物的袋子。", + "evidence": [ + { + "description": "A bag containing food was placed by me on the stove.", + "timestamps": { + "start": "00:20:13", + "end": "00:20:16" + } + }, + { + "description": "There are 12 bags of portioned food placed next to the sink.", + "timestamps": { + "start": "00:19:28", + "end": "00:19:34" + } + } + ], + "question_timestamps": "00:21:59", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are ten bags by the sink.", + "B": "There are thirteen bags by the sink.", + "C": "There are eight bags of portioned food placed next to the sink.", + "D": "There are eleven bags of portioned food placed next to the sink.", + "E": "There are fifteen bags of portioned food placed next to the sink area." + }, + "answer_label": "D", + "answer_text": "There are eleven bags of portioned food placed next to the sink.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_5", + "correct_answer": "There are eleven bags of portioned food placed next to the sink.", + "original_options": { + "A": "There are ten bags by the sink.", + "B": "There are thirteen bags by the sink.", + "C": "There are eight bags of portioned food placed next to the sink.", + "D": "There are eleven bags of portioned food placed next to the sink.", + "E": "There are fifteen bags of portioned food placed next to the sink area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 13 + ] + }, + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_6_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Facing the sink, what should I do if I want to throw away garbage?", + "answer": "The trash can is on the right side of the sink.", + "question_chinese": "面对水槽，我想扔垃圾应该怎么做？", + "answer_chinese": "面对水槽，你想扔垃圾应该扔到右边的垃圾桶。", + "evidence": [ + { + "description": "The trash can is on the right side of the sink.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:03:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can is on the right side of the sink.", + "B": "The trash can is on the left.", + "C": "The trash can is behind the sink.", + "D": "The trash can is inside the cabinet beneath the sink area now.", + "E": "The trash can is up on the counter next to the sink." + }, + "answer_label": "A", + "answer_text": "The trash can is on the right side of the sink.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_6_T1", + "correct_answer": "The trash can is on the right side of the sink.", + "original_options": { + "A": "The trash can is on the right side of the sink.", + "B": "The trash can is on the left.", + "C": "The trash can is behind the sink.", + "D": "The trash can is inside the cabinet beneath the sink area now.", + "E": "The trash can is up on the counter next to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 12 + ] + }, + { + "video_uid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868", + "video_length": "00:23:50", + "qid": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_6_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Facing the sink, what should I do if I want to throw away garbage?", + "answer": "Facing the sink, you should throw the garbage into the trash can on the right.", + "question_chinese": "面对水槽，我想扔垃圾应该怎么做？", + "answer_chinese": "面对水槽，你想扔垃圾应该扔到右边的垃圾桶。", + "evidence": [ + { + "description": "The trash can is on the right side of the sink.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:12:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use the trash can left.", + "B": "Use the bin ahead.", + "C": "Facing the sink, toss the garbage into the recycling bin on your left side.", + "D": "Facing the sink, drop the garbage into a bag sitting on the counter behind you.", + "E": "Facing the sink, you should throw the garbage into the trash can on the right." + }, + "answer_label": "E", + "answer_text": "Facing the sink, you should throw the garbage into the trash can on the right.", + "q_id": "2c27b5f1-4af6-49ad-a43c-3efb0c150868_6_T2", + "correct_answer": "Facing the sink, you should throw the garbage into the trash can on the right.", + "original_options": { + "A": "Use the trash can left.", + "B": "Use the bin ahead.", + "C": "Facing the sink, toss the garbage into the recycling bin on your left side.", + "D": "Facing the sink, drop the garbage into a bag sitting on the counter behind you.", + "E": "Facing the sink, you should throw the garbage into the trash can on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 15 + ] + } +] diff --git a/QAs/2c365418-7598-48e2-a2c7-6c1885a351f4-mcq.json b/QAs/2c365418-7598-48e2-a2c7-6c1885a351f4-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4c0029ecfab8c94a220dc7aa4d8f7d0f305e314c --- /dev/null +++ b/QAs/2c365418-7598-48e2-a2c7-6c1885a351f4-mcq.json @@ -0,0 +1,680 @@ +[ + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_5", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to look for a tape measure, what should my movement trajectory be?", + "answer": "You should turn around and walk straight to the colleague in the red shirt.", + "question_chinese": "如果我想去找卷尺的话，我的运动轨迹应该是怎么样的呢？", + "answer_chinese": "你应该向后转，然后直走红色衣服的同事那里。", + "evidence": [ + { + "description": "You took a tape measure from the table where the tools are arranged in the room.", + "timestamps": { + "start": "00:07:23", + "end": "00:07:30" + } + }, + { + "description": "The tape measure is being held by a colleague in a red shirt.", + "timestamps": { + "start": "00:07:04", + "end": "00:07:14" + } + } + ], + "question_timestamps": "00:07:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right to the colleague.", + "B": "You should turn around and walk straight to the colleague in the red shirt.", + "C": "Walk forward to the colleague.", + "D": "Turn around and head straight to the colleague holding the tape dispenser instead.", + "E": "Turn left and walk over to the colleague in a maroon shirt by the toolbox." + }, + "answer_label": "B", + "answer_text": "You should turn around and walk straight to the colleague in the red shirt.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_5", + "correct_answer": "You should turn around and walk straight to the colleague in the red shirt.", + "original_options": { + "A": "Turn right to the colleague.", + "B": "You should turn around and walk straight to the colleague in the red shirt.", + "C": "Walk forward to the colleague.", + "D": "Turn around and head straight to the colleague holding the tape dispenser instead.", + "E": "Turn left and walk over to the colleague in a maroon shirt by the toolbox." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 15 + ] + }, + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_6", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to look for a tape measure, what should my movement trajectory be?", + "answer": "You should turn left and walk straight to the workpiece.", + "question_chinese": "如果我想去找卷尺的话，我的运动轨迹应该是怎么样的呢？", + "answer_chinese": "你应该向左转身，然后直走到工件的地方。", + "evidence": [ + { + "description": "The tape measure is right above the workpiece.", + "timestamps": { + "start": "00:10:16", + "end": "00:10:22" + } + } + ], + "question_timestamps": "00:10:36", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right and walk to the toolbox.", + "B": "Turn around and walk to the shelf.", + "C": "Turn left and walk straight to the chair instead of the workpiece.", + "D": "You should turn left and walk straight to the workpiece.", + "E": "Turn right and walk straight over to the desk rather than the workpiece." + }, + "answer_label": "D", + "answer_text": "You should turn left and walk straight to the workpiece.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_6", + "correct_answer": "You should turn left and walk straight to the workpiece.", + "original_options": { + "A": "Turn right and walk to the toolbox.", + "B": "Turn around and walk to the shelf.", + "C": "Turn left and walk straight to the chair instead of the workpiece.", + "D": "You should turn left and walk straight to the workpiece.", + "E": "Turn right and walk straight over to the desk rather than the workpiece." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 13 + ] + }, + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_7", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to look for more tools, what should my movement trajectory be?", + "answer": "You should turn around and take four steps straight back.", + "question_chinese": "如果我想找更多工具的话，我的运动轨迹应该是怎么样的呢？", + "answer_chinese": "你应该向后转，然后直走四步。", + "evidence": [ + { + "description": "You see a long row of tool cabinets and a tool table.", + "timestamps": { + "start": "00:08:44", + "end": "00:08:50" + } + } + ], + "question_timestamps": "00:08:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn around and take three steps back.", + "B": "Turn around and take five steps back.", + "C": "Turn around and take four steps to the right instead of straight back.", + "D": "Turn left and take two steps forward rather than turning around and stepping back.", + "E": "You should turn around and take four steps straight back." + }, + "answer_label": "E", + "answer_text": "You should turn around and take four steps straight back.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_7", + "correct_answer": "You should turn around and take four steps straight back.", + "original_options": { + "A": "Turn around and take three steps back.", + "B": "Turn around and take five steps back.", + "C": "Turn around and take four steps to the right instead of straight back.", + "D": "Turn left and take two steps forward rather than turning around and stepping back.", + "E": "You should turn around and take four steps straight back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 14 + ] + }, + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_8", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to look for more tools, what should my movement trajectory be?", + "answer": "You should turn right 15 degrees and walk straight for eight steps.", + "question_chinese": "如果我想找更多工具的话，我的运动轨迹应该是怎么样的呢？", + "answer_chinese": "你应该右转15度，然后直走八步。", + "evidence": [ + { + "description": "You see a long row of tool cabinets and a tool table.", + "timestamps": { + "start": "00:08:44", + "end": "00:08:50" + } + } + ], + "question_timestamps": "00:29:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left 15 degrees and walk eight steps.", + "B": "Turn right 15 degrees and walk six steps.", + "C": "Turn right about 30 degrees, then walk straight forward for eight steps.", + "D": "Turn right 15 degrees and walk straight for ten steps instead of eight.", + "E": "You should turn right 15 degrees and walk straight for eight steps." + }, + "answer_label": "E", + "answer_text": "You should turn right 15 degrees and walk straight for eight steps.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_8", + "correct_answer": "You should turn right 15 degrees and walk straight for eight steps.", + "original_options": { + "A": "Turn left 15 degrees and walk eight steps.", + "B": "Turn right 15 degrees and walk six steps.", + "C": "Turn right about 30 degrees, then walk straight forward for eight steps.", + "D": "Turn right 15 degrees and walk straight for ten steps instead of eight.", + "E": "You should turn right 15 degrees and walk straight for eight steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 12, + 13 + ] + }, + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many jacks have I seen in total?", + "answer": "You have seen two jack.", + "question_chinese": "我一共看到过多少个千斤顶？", + "answer_chinese": "你看到过两个千斤顶。", + "evidence": [ + { + "description": "The girl in red is wiping a jack.", + "timestamps": { + "start": "00:04:53", + "end": "00:05:10" + } + }, + { + "description": "Your hand has also touched a jack.", + "timestamps": { + "start": "00:05:15", + "end": "00:05:20" + } + } + ], + "question_timestamps": "00:05:26", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have seen one jack.", + "B": "You have seen three jacks.", + "C": "You have seen four jacks in total so far, counting every time you came across them.", + "D": "You have seen two jack.", + "E": "You have seen five jacks altogether, including the ones you noticed earlier and later in the scene." + }, + "answer_label": "D", + "answer_text": "You have seen two jack.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_1", + "correct_answer": "You have seen two jack.", + "original_options": { + "A": "You have seen one jack.", + "B": "You have seen three jacks.", + "C": "You have seen four jacks in total so far, counting every time you came across them.", + "D": "You have seen two jack.", + "E": "You have seen five jacks altogether, including the ones you noticed earlier and later in the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many jacks have I seen in total?", + "answer": "You have seen three jacks.", + "question_chinese": "我一共看到过多少个千斤顶？", + "answer_chinese": "你看到过三个千斤顶。", + "evidence": [ + { + "description": "The girl in red is wiping a jack.", + "timestamps": { + "start": "00:04:53", + "end": "00:05:10" + } + }, + { + "description": "Your hand has also touched a jack.", + "timestamps": { + "start": "00:05:15", + "end": "00:05:20" + } + } + ], + "question_timestamps": "00:23:13", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have seen three jacks.", + "B": "You have seen two jacks.", + "C": "You have seen four jacks.", + "D": "You have seen five jacks in total, adding up each jack you encountered across the whole video.", + "E": "You have seen six jacks altogether, counting all instances from earlier moments through the end of the clip." + }, + "answer_label": "A", + "answer_text": "You have seen three jacks.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_2", + "correct_answer": "You have seen three jacks.", + "original_options": { + "A": "You have seen three jacks.", + "B": "You have seen two jacks.", + "C": "You have seen four jacks.", + "D": "You have seen five jacks in total, adding up each jack you encountered across the whole video.", + "E": "You have seen six jacks altogether, counting all instances from earlier moments through the end of the clip." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_5_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just walk from the garage to the living room?", + "answer": "Starting from the garage door, you walked straight through the kitchen in front of you, then turned left to face the living room.", + "question_chinese": "我刚刚从车库到客厅的行走轨迹是？", + "answer_chinese": "你从车库门出发先直走穿过面前的厨房，然后左转面向客厅。", + "evidence": [ + { + "description": "You start from the garage door, walk straight through the kitchen in front of you, then turn left to face the living roo.m", + "timestamps": { + "start": "00:20:20", + "end": "00:20:27" + } + } + ], + "question_timestamps": "00:21:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right into the living room.", + "B": "You turned left before the kitchen.", + "C": "Starting from the garage door, you walked straight through the kitchen, then turned right to enter the living room.", + "D": "Starting from the garage door, you walked straight through the kitchen in front of you, then turned left to face the living room.", + "E": "Starting from the garage door, you turned left first and then continued through the kitchen to reach the living room." + }, + "answer_label": "D", + "answer_text": "Starting from the garage door, you walked straight through the kitchen in front of you, then turned left to face the living room.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_5_T2", + "correct_answer": "Starting from the garage door, you walked straight through the kitchen in front of you, then turned left to face the living room.", + "original_options": { + "A": "You turned right into the living room.", + "B": "You turned left before the kitchen.", + "C": "Starting from the garage door, you walked straight through the kitchen, then turned right to enter the living room.", + "D": "Starting from the garage door, you walked straight through the kitchen in front of you, then turned left to face the living room.", + "E": "Starting from the garage door, you turned left first and then continued through the kitchen to reach the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 20 + ] + }, + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_5_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just walk from the garage to the living room?", + "answer": "Your route was: go straight through the kitchen first, then turn left into the living room.", + "question_chinese": "我刚刚从车库到客厅的行走轨迹是？", + "answer_chinese": "你的轨迹就是先直走穿过厨房，然后左转来到客厅。", + "evidence": [ + { + "description": "You walk from the garage to the kitchen.", + "timestamps": { + "start": "00:20:21", + "end": "00:20:23" + } + }, + { + "description": "You walk from the kitchen to the living room.", + "timestamps": { + "start": "00:20:23", + "end": "00:20:28" + } + } + ], + "question_timestamps": "00:24:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right into the living room.", + "B": "You turned left before going straight.", + "C": "Your route was: go straight through the kitchen first, then turn left into the living room.", + "D": "Your route was: go straight through the kitchen first, then turn right into the living room.", + "E": "Your route was: turn left into the kitchen first, then continue straight to reach the living room." + }, + "answer_label": "C", + "answer_text": "Your route was: go straight through the kitchen first, then turn left into the living room.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_5_T1", + "correct_answer": "Your route was: go straight through the kitchen first, then turn left into the living room.", + "original_options": { + "A": "You turned right into the living room.", + "B": "You turned left before going straight.", + "C": "Your route was: go straight through the kitchen first, then turn left into the living room.", + "D": "Your route was: go straight through the kitchen first, then turn right into the living room.", + "E": "Your route was: turn left into the kitchen first, then continue straight to reach the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 17 + ] + }, + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_11", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the wooden board I took out?", + "answer": "You have placed all the wooden boards you took out onto the workpiece.", + "question_chinese": "我取出来的木板放在了哪里？", + "answer_chinese": "你拿出来的木板都被放在工件上了。", + "evidence": [ + { + "description": "You have placed all the wooden boards you took out onto the workpiece.", + "timestamps": { + "start": "00:06:53", + "end": "00:07:07" + } + }, + { + "description": "You have placed all the wooden boards you took out onto the workpiece.", + "timestamps": { + "start": "00:05:50", + "end": "00:05:56" + } + } + ], + "question_timestamps": "00:08:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You have placed all the wooden boards you took out onto the workpiece.", + "B": "You put them on the countertop.", + "C": "You put them on the shelf.", + "D": "You placed all the wooden boards you took out onto the floorboard instead of onto the workpiece.", + "E": "You put all the wooden boards you took out into the cabinet, keeping them off the workpiece." + }, + "answer_label": "A", + "answer_text": "You have placed all the wooden boards you took out onto the workpiece.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_11", + "correct_answer": "You have placed all the wooden boards you took out onto the workpiece.", + "original_options": { + "A": "You have placed all the wooden boards you took out onto the workpiece.", + "B": "You put them on the countertop.", + "C": "You put them on the shelf.", + "D": "You placed all the wooden boards you took out onto the floorboard instead of onto the workpiece.", + "E": "You put all the wooden boards you took out into the cabinet, keeping them off the workpiece." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_12", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the wooden board I took out?", + "answer": "You have placed all the wooden boards you took out on the floor.", + "question_chinese": "我取出来的木板放在了哪里？", + "answer_chinese": "你拿出来的木板都被放在地上了。", + "evidence": [ + { + "description": "You have placed all the wooden boards you took out on the floor.", + "timestamps": { + "start": "00:25:53", + "end": "00:26:39" + } + } + ], + "question_timestamps": "00:29:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put them on the counter.", + "B": "You put them under the table.", + "C": "You placed all the wooden boards you took out beside the wall rather than leaving them on the floor.", + "D": "You placed all the wooden boards you took out in the garage area instead of putting them down on the floor.", + "E": "You have placed all the wooden boards you took out on the floor." + }, + "answer_label": "E", + "answer_text": "You have placed all the wooden boards you took out on the floor.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_12", + "correct_answer": "You have placed all the wooden boards you took out on the floor.", + "original_options": { + "A": "You put them on the counter.", + "B": "You put them under the table.", + "C": "You placed all the wooden boards you took out beside the wall rather than leaving them on the floor.", + "D": "You placed all the wooden boards you took out in the garage area instead of putting them down on the floor.", + "E": "You have placed all the wooden boards you took out on the floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 21 + ] + }, + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_3", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I find wooden boards?", + "answer": "You can find the pile of wooden planks in the center of the room.", + "question_chinese": "我能在哪里找到木板呢", + "answer_chinese": "你能在房间中心找到木板堆。", + "evidence": [ + { + "description": "You saw the pile of wooden planks in the center of the room.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:29" + } + } + ], + "question_timestamps": "00:05:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the corner of the room.", + "B": "Near the window in the room.", + "C": "You can find the pile of wooden planks on a shelf above the table, not in the middle of the room.", + "D": "You can find the wooden boards stored in a drawer under the desk rather than in a central pile.", + "E": "You can find the pile of wooden planks in the center of the room." + }, + "answer_label": "E", + "answer_text": "You can find the pile of wooden planks in the center of the room.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_3", + "correct_answer": "You can find the pile of wooden planks in the center of the room.", + "original_options": { + "A": "In the corner of the room.", + "B": "Near the window in the room.", + "C": "You can find the pile of wooden planks on a shelf above the table, not in the middle of the room.", + "D": "You can find the wooden boards stored in a drawer under the desk rather than in a central pile.", + "E": "You can find the pile of wooden planks in the center of the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 19 + ] + }, + { + "video_uid": "2c365418-7598-48e2-a2c7-6c1885a351f4", + "video_length": "00:30:00", + "qid": "2c365418-7598-48e2-a2c7-6c1885a351f4_4", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I find wooden boards?", + "answer": "You can find the pile of wooden planks on the left side and in the center of the room.", + "question_chinese": "我能在哪里找到木板呢", + "answer_chinese": "你能在房间的左侧和中间找到木板堆。", + "evidence": [ + { + "description": "You saw the pile of wooden planks in the center of the room.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:29" + } + }, + { + "description": "You also saw a pile of wooden planks on the left side of the room.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:10" + } + } + ], + "question_timestamps": "00:27:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the right side, centered.", + "B": "You can find the pile of wooden planks on the left side and in the center of the room.", + "C": "Behind you, centered.", + "D": "You can find the pile of wooden planks in front of you and in the center of the room.", + "E": "You can find the pile of wooden planks on the right side while still roughly in the center of the room." + }, + "answer_label": "B", + "answer_text": "You can find the pile of wooden planks on the left side and in the center of the room.", + "q_id": "2c365418-7598-48e2-a2c7-6c1885a351f4_4", + "correct_answer": "You can find the pile of wooden planks on the left side and in the center of the room.", + "original_options": { + "A": "On the right side, centered.", + "B": "You can find the pile of wooden planks on the left side and in the center of the room.", + "C": "Behind you, centered.", + "D": "You can find the pile of wooden planks in front of you and in the center of the room.", + "E": "You can find the pile of wooden planks on the right side while still roughly in the center of the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 19, + 21 + ] + } +] diff --git a/QAs/2c78909f-ff5c-4a70-b5ae-2d74e11fbe93-mcq.json b/QAs/2c78909f-ff5c-4a70-b5ae-2d74e11fbe93-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e82529083650afacdb531b2d5f3c66ea04ad2ad0 --- /dev/null +++ b/QAs/2c78909f-ff5c-4a70-b5ae-2d74e11fbe93-mcq.json @@ -0,0 +1,750 @@ +[ + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_4_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was kneading flour, was the stool on my left or my right?", + "answer": "When you were kneading the flour, the stool was to your left.", + "question_chinese": "当我在揉面粉时，凳子在我的左/右？", + "answer_chinese": "当你在揉面粉时，凳子在你的左边。", + "evidence": [ + { + "description": "The stool is placed on my left.", + "timestamps": { + "start": "00:05:54", + "end": "00:05:59" + } + } + ], + "question_timestamps": "00:15:39", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were kneading the flour, the stool was to your right.", + "B": "When you were kneading the flour, the stool was to your left." + }, + "answer_label": "B", + "answer_text": "When you were kneading the flour, the stool was to your left.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_4_T1", + "correct_answer": "When you were kneading the flour, the stool was to your left.", + "original_options": { + "A": "When you were kneading the flour, the stool was to your right.", + "B": "When you were kneading the flour, the stool was to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_4_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was kneading flour, was the stool on my left or my right?", + "answer": "Left.", + "question_chinese": "当我在揉面粉时，凳子在我的左/右？", + "answer_chinese": "左", + "evidence": [ + { + "description": "The stool is placed on my left.", + "timestamps": { + "start": "00:05:54", + "end": "00:05:59" + } + } + ], + "question_timestamps": "00:38:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right.", + "B": "Left." + }, + "answer_label": "B", + "answer_text": "Left.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_4_T2", + "correct_answer": "Left.", + "original_options": { + "A": "Right.", + "B": "Left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_4", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand in front of the table, which is closer to me, the spoon or the pen?", + "answer": "When you stand in front of the table, the spoon is closer to you.", + "question_chinese": "当我站在桌子前，勺子和笔哪个离我更近？", + "answer_chinese": "当你站在桌子前，勺子离你更近。", + "evidence": [ + { + "description": "The spoon is next to my right hand.", + "timestamps": { + "start": "00:05:48", + "end": "00:05:51" + } + } + ], + "question_timestamps": "00:07:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you stand in front of the table, the pen is closer to you.", + "B": "When you stand in front of the table, the spoon is closer to you." + }, + "answer_label": "B", + "answer_text": "When you stand in front of the table, the spoon is closer to you.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_4", + "correct_answer": "When you stand in front of the table, the spoon is closer to you.", + "original_options": { + "A": "When you stand in front of the table, the pen is closer to you.", + "B": "When you stand in front of the table, the spoon is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_5", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand in front of the table, which is closer to me, the spoon or the pen?", + "answer": "When you stand in front of the table, the pen is closer to you.", + "question_chinese": "当我站在桌子前，勺子和笔哪个离我更近？", + "answer_chinese": "当你站在桌子前，笔离你更近。", + "evidence": [ + { + "description": "The pen is next to the notebook on the table.", + "timestamps": { + "start": "00:05:48", + "end": "00:05:51" + } + }, + { + "description": "After using the spoon, I placed it in the metal basin to the left front.", + "timestamps": { + "start": "00:07:22", + "end": "00:08:17" + } + } + ], + "question_timestamps": "00:15:31", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "When you stand in front of the table, the spoon is closer to you.", + "B": "When you stand in front of the table, the pen is closer to you." + }, + "answer_label": "B", + "answer_text": "When you stand in front of the table, the pen is closer to you.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_5", + "correct_answer": "When you stand in front of the table, the pen is closer to you.", + "original_options": { + "A": "When you stand in front of the table, the spoon is closer to you.", + "B": "When you stand in front of the table, the pen is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_9_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many white towels did I hang out to dry in total?", + "answer": "You hung up eight white towels to dry in total.", + "question_chinese": "我一共晾了几条白色毛巾？", + "answer_chinese": "你一共晾了八条白色毛巾。", + "evidence": [ + { + "description": "I hung a total of eight white towels.", + "timestamps": { + "start": "00:32:38", + "end": "00:35:03" + } + } + ], + "question_timestamps": "00:36:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Nine towels.", + "B": "You hung up eight white towels to dry in total.", + "C": "Seven towels.", + "D": "You hung up six white towels to dry in total on the line outside.", + "E": "You hung up ten white towels to dry in total after washing them." + }, + "answer_label": "B", + "answer_text": "You hung up eight white towels to dry in total.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_9_T1", + "correct_answer": "You hung up eight white towels to dry in total.", + "original_options": { + "A": "Nine towels.", + "B": "You hung up eight white towels to dry in total.", + "C": "Seven towels.", + "D": "You hung up six white towels to dry in total on the line outside.", + "E": "You hung up ten white towels to dry in total after washing them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_9_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many white towels did I hang out to dry in total?", + "answer": "You hung up eight white towels to dry in total.", + "question_chinese": "我一共晾了几条白色毛巾？", + "answer_chinese": "你一共把八条白色毛巾晾了。", + "evidence": [ + { + "description": "I hung a total of eight white towels.", + "timestamps": { + "start": "00:32:38", + "end": "00:35:03" + } + } + ], + "question_timestamps": "00:42:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Nine towels.", + "B": "Six towels.", + "C": "You hung up eight white towels to dry in total.", + "D": "You hung up seven white towels to dry in total on the line outside.", + "E": "You hung up ten white towels to dry in total after washing them." + }, + "answer_label": "C", + "answer_text": "You hung up eight white towels to dry in total.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_9_T2", + "correct_answer": "You hung up eight white towels to dry in total.", + "original_options": { + "A": "Nine towels.", + "B": "Six towels.", + "C": "You hung up eight white towels to dry in total.", + "D": "You hung up seven white towels to dry in total on the line outside.", + "E": "You hung up ten white towels to dry in total after washing them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_7_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was washing the towel, which was closer to me: the blue bucket or the red bucket with paint in it?", + "answer": "When you were washing the towel, the blue bucket was closer to you.", + "question_chinese": "我在洗毛巾时，蓝色的桶和装有颜料的红色桶，哪个离我更近？", + "answer_chinese": "你在洗毛巾时，蓝色的桶离你更近。", + "evidence": [ + { + "description": "The blue bucket is in front of me.", + "timestamps": { + "start": "00:26:37", + "end": "00:26:40" + } + } + ], + "question_timestamps": "00:34:21", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were washing the towel, the red bucket with paint in it was closer to you.", + "B": "When you were washing the towel, the blue bucket was closer to you." + }, + "answer_label": "B", + "answer_text": "When you were washing the towel, the blue bucket was closer to you.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_7_T1", + "correct_answer": "When you were washing the towel, the blue bucket was closer to you.", + "original_options": { + "A": "When you were washing the towel, the red bucket with paint in it was closer to you.", + "B": "When you were washing the towel, the blue bucket was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_7_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was washing the towel, which was closer to me: the blue bucket or the red bucket with paint in it?", + "answer": "The blue bucket was closer to you when you were washing the towel.", + "question_chinese": "我在洗毛巾时，蓝色的桶和装有颜料的红色桶，哪个离我更近？", + "answer_chinese": "蓝色的桶离你更近，在你洗毛巾的时候。", + "evidence": [ + { + "description": "The blue bucket is in front of me.", + "timestamps": { + "start": "00:26:37", + "end": "00:26:40" + } + } + ], + "question_timestamps": "00:42:42", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The red bucket with paint in it was closer to you when you were washing the towel.", + "B": "The blue bucket was closer to you when you were washing the towel." + }, + "answer_label": "B", + "answer_text": "The blue bucket was closer to you when you were washing the towel.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_7_T2", + "correct_answer": "The blue bucket was closer to you when you were washing the towel.", + "original_options": { + "A": "The red bucket with paint in it was closer to you when you were washing the towel.", + "B": "The blue bucket was closer to you when you were washing the towel." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_19", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the green clip when I hang the towel?", + "answer": "When you hung the towel, the green clip was fastened to the clothesline.", + "question_chinese": "我晾毛巾时，绿色的夹子在哪里？", + "answer_chinese": "你晾毛巾时，绿色的夹子夹在了绳子上。", + "evidence": [ + { + "description": "The green clip is clipped onto the rope.", + "timestamps": { + "start": "00:35:10", + "end": "00:35:12" + } + } + ], + "question_timestamps": "00:37:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Clipped to the towel edge.", + "B": "Fastened to a nearby rod.", + "C": "When you hung the towel, the green clip was secured to the window frame rather than the line.", + "D": "When you hung the towel, the green clip was attached to a wooden beam next to the clothesline.", + "E": "When you hung the towel, the green clip was fastened to the clothesline." + }, + "answer_label": "E", + "answer_text": "When you hung the towel, the green clip was fastened to the clothesline.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_19", + "correct_answer": "When you hung the towel, the green clip was fastened to the clothesline.", + "original_options": { + "A": "Clipped to the towel edge.", + "B": "Fastened to a nearby rod.", + "C": "When you hung the towel, the green clip was secured to the window frame rather than the line.", + "D": "When you hung the towel, the green clip was attached to a wooden beam next to the clothesline.", + "E": "When you hung the towel, the green clip was fastened to the clothesline." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_20", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the green clip when I hang the towel?", + "answer": "The green clip clamped onto the red and white checkered towel.", + "question_chinese": "我晾毛巾时，绿色的夹子在哪里？", + "answer_chinese": "绿色夹子夹到了红白格子毛巾上。", + "evidence": [ + { + "description": "The green clip is clipped onto the rope.", + "timestamps": { + "start": "00:35:10", + "end": "00:35:12" + } + } + ], + "question_timestamps": "00:43:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Clamped onto the striped towel.", + "B": "Clamped onto the spotted towel.", + "C": "The green clip clamped onto the red and white checkered towel.", + "D": "The green clip clamped onto the green and white checkered towel instead of the red and white one.", + "E": "The green clip clamped onto the red and white checkered shirt rather than the towel you hung." + }, + "answer_label": "C", + "answer_text": "The green clip clamped onto the red and white checkered towel.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_20", + "correct_answer": "The green clip clamped onto the red and white checkered towel.", + "original_options": { + "A": "Clamped onto the striped towel.", + "B": "Clamped onto the spotted towel.", + "C": "The green clip clamped onto the red and white checkered towel.", + "D": "The green clip clamped onto the green and white checkered towel instead of the red and white one.", + "E": "The green clip clamped onto the red and white checkered shirt rather than the towel you hung." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_8_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was hanging the towel out to dry, how far was I from the banana tree?", + "answer": "When you were hanging the towel to dry, you were 5 meters from the banana tree.", + "question_chinese": "我晾毛巾时，距离芭蕉树多远？", + "answer_chinese": "你晾毛巾时，距离芭蕉树五米远。", + "evidence": [ + { + "description": "The banana tree is planted in the ground outside the house.", + "timestamps": { + "start": "00:32:38", + "end": "00:32:42" + } + } + ], + "question_timestamps": "00:35:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "When you were hanging the towel to dry, you were 5 meters from the banana tree.", + "B": "6 meters away.", + "C": "4 meters away.", + "D": "When you were hanging the towel to dry, you were 7 meters from the banana tree.", + "E": "When you were hanging the towel to dry, you were 3 meters from the banana tree." + }, + "answer_label": "A", + "answer_text": "When you were hanging the towel to dry, you were 5 meters from the banana tree.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_8_T1", + "correct_answer": "When you were hanging the towel to dry, you were 5 meters from the banana tree.", + "original_options": { + "A": "When you were hanging the towel to dry, you were 5 meters from the banana tree.", + "B": "6 meters away.", + "C": "4 meters away.", + "D": "When you were hanging the towel to dry, you were 7 meters from the banana tree.", + "E": "When you were hanging the towel to dry, you were 3 meters from the banana tree." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_8_T2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was hanging the towel out to dry, how far was I from the banana tree?", + "answer": "You were 5 meters from the banana tree.", + "question_chinese": "我晾毛巾时，距离芭蕉树多远？", + "answer_chinese": "你距离芭蕉树五米远。", + "evidence": [ + { + "description": "The banana tree is planted in the ground outside the house.", + "timestamps": { + "start": "00:32:38", + "end": "00:32:42" + } + } + ], + "question_timestamps": "00:40:52", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "6 meters away.", + "B": "You were 5 meters from the banana tree.", + "C": "4 meters away.", + "D": "When you were hanging the towel out to dry, you were 7 meters from the banana tree.", + "E": "When you were hanging the towel out to dry, you were 3 meters from the banana tree." + }, + "answer_label": "B", + "answer_text": "You were 5 meters from the banana tree.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_8_T2", + "correct_answer": "You were 5 meters from the banana tree.", + "original_options": { + "A": "6 meters away.", + "B": "You were 5 meters from the banana tree.", + "C": "4 meters away.", + "D": "When you were hanging the towel out to dry, you were 7 meters from the banana tree.", + "E": "When you were hanging the towel out to dry, you were 3 meters from the banana tree." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_5_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I’m standing in front of the table—how do I walk to the stove?", + "answer": "You should turn left and then walk straight.", + "question_chinese": "我站在桌子前，要去灶台那里怎么走？", + "answer_chinese": "你要左转然后直走。", + "evidence": [ + { + "description": "The stove is on the left side of the table.", + "timestamps": { + "start": "00:15:35", + "end": "00:15:40" + } + } + ], + "question_timestamps": "00:17:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You should turn left and then walk straight.", + "B": "Turn right, then walk straight.", + "C": "Go straight without turning.", + "D": "You should turn right first, then continue straight until you reach the stove.", + "E": "From the table, turn left, take a couple steps, then turn right to face the stove." + }, + "answer_label": "A", + "answer_text": "You should turn left and then walk straight.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_5_T1", + "correct_answer": "You should turn left and then walk straight.", + "original_options": { + "A": "You should turn left and then walk straight.", + "B": "Turn right, then walk straight.", + "C": "Go straight without turning.", + "D": "You should turn right first, then continue straight until you reach the stove.", + "E": "From the table, turn left, take a couple steps, then turn right to face the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 16 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_5_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I’m standing in front of the table—how do I walk to the stove?", + "answer": "You need to turn left and then walk straight.", + "question_chinese": "我站在桌子前，要去灶台那里怎么走？", + "answer_chinese": "你需要左转然后直走。", + "evidence": [ + { + "description": "The stove is on the left side of the table.", + "timestamps": { + "start": "00:15:35", + "end": "00:15:40" + } + } + ], + "question_timestamps": "00:40:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right, then walk straight.", + "B": "Walk straight without turning.", + "C": "You need to turn right first, then keep walking straight to get to the stove.", + "D": "Turn left, walk a few steps, then turn right to line up with the stove.", + "E": "You need to turn left and then walk straight." + }, + "answer_label": "E", + "answer_text": "You need to turn left and then walk straight.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_5_T2", + "correct_answer": "You need to turn left and then walk straight.", + "original_options": { + "A": "Turn right, then walk straight.", + "B": "Walk straight without turning.", + "C": "You need to turn right first, then keep walking straight to get to the stove.", + "D": "Turn left, walk a few steps, then turn right to line up with the stove.", + "E": "You need to turn left and then walk straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 15 + ] + }, + { + "video_uid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93", + "video_length": "00:44:29", + "qid": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of ungrated coconut meat are on the table?", + "answer": "There is a piece of coconut meat on the table that hasn't been grated into shreds.", + "question_chinese": "桌子上有几块没有磨成丝的椰肉？", + "answer_chinese": "桌子上有一块没有磨成丝的椰肉。", + "evidence": [ + { + "description": "There is a piece of coconut meat on the table that hasn't been grated into shreds.", + "timestamps": { + "start": "00:01:09", + "end": "00:01:13" + } + } + ], + "question_timestamps": "00:03:24", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "None.", + "B": "There is a piece of coconut meat on the table that hasn't been grated into shreds.", + "C": "Two pieces.", + "D": "There are three pieces of coconut meat on the table that haven't been grated into shreds.", + "E": "There are four pieces of coconut meat on the table that still haven't been grated into shreds." + }, + "answer_label": "B", + "answer_text": "There is a piece of coconut meat on the table that hasn't been grated into shreds.", + "q_id": "2c78909f-ff5c-4a70-b5ae-2d74e11fbe93_1", + "correct_answer": "There is a piece of coconut meat on the table that hasn't been grated into shreds.", + "original_options": { + "A": "None.", + "B": "There is a piece of coconut meat on the table that hasn't been grated into shreds.", + "C": "Two pieces.", + "D": "There are three pieces of coconut meat on the table that haven't been grated into shreds.", + "E": "There are four pieces of coconut meat on the table that still haven't been grated into shreds." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 16, + 17 + ] + } +] diff --git a/QAs/2caa83ec-4e4f-426e-b177-2bbf6ca81326-mcq.json b/QAs/2caa83ec-4e4f-426e-b177-2bbf6ca81326-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1c0e73baad7c0ff0448ae812fd11fc2ef98809f2 --- /dev/null +++ b/QAs/2caa83ec-4e4f-426e-b177-2bbf6ca81326-mcq.json @@ -0,0 +1,796 @@ +[ + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the car when I'm in the room？", + "answer": "When you're in the room and need to get to the car, turn right and then continue turning right to exit the doorway.", + "question_chinese": "当我在房间里，要去到汽车旁边怎么走？", + "answer_chinese": "当你在房间，要去到车旁边应该右转，然后继续右转出门口。", + "evidence": [ + { + "description": "I sat on the chair, and the car was outside the door to my left rear.", + "timestamps": { + "start": "00:11:12", + "end": "00:11:16" + } + } + ], + "question_timestamps": "00:11:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "When you're in the room and need to get to the car, turn right and then continue turning right to exit the doorway.", + "B": "Turn left to exit.", + "C": "Go straight out the doorway.", + "D": "When you're in the room, turn left and then turn left again to get out through the doorway.", + "E": "When you're in the room, turn right once, then go straight through the doorway without turning again." + }, + "answer_label": "A", + "answer_text": "When you're in the room and need to get to the car, turn right and then continue turning right to exit the doorway.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_7", + "correct_answer": "When you're in the room and need to get to the car, turn right and then continue turning right to exit the doorway.", + "original_options": { + "A": "When you're in the room and need to get to the car, turn right and then continue turning right to exit the doorway.", + "B": "Turn left to exit.", + "C": "Go straight out the doorway.", + "D": "When you're in the room, turn left and then turn left again to get out through the doorway.", + "E": "When you're in the room, turn right once, then go straight through the doorway without turning again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 17 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the car when I'm in the room？", + "answer": "When you are in the room and need to go to the car, you should turn right to exit the door.", + "question_chinese": "当我在房间里，要去到汽车旁边怎么走？", + "answer_chinese": "当你在房间，要去到车旁边应该右转出门口。", + "evidence": [ + { + "description": "I'm standing next to the sofa, and now the car is outside the door to my right front.", + "timestamps": { + "start": "00:11:16", + "end": "00:11:17" + } + }, + { + "description": "I stood up from the stool and walked to the sofa.", + "timestamps": { + "start": "00:11:15", + "end": "00:11:17" + } + } + ], + "question_timestamps": "00:11:22", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left to exit the door.", + "B": "Go straight out the door.", + "C": "When you are in the room and need to go to the car, you should turn left to exit the door.", + "D": "When you are in the room and need to go to the car, you should turn right to exit the door.", + "E": "When you are in the room and need to go to the car, turn right first, then turn left before exiting." + }, + "answer_label": "D", + "answer_text": "When you are in the room and need to go to the car, you should turn right to exit the door.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_8", + "correct_answer": "When you are in the room and need to go to the car, you should turn right to exit the door.", + "original_options": { + "A": "Turn left to exit the door.", + "B": "Go straight out the door.", + "C": "When you are in the room and need to go to the car, you should turn left to exit the door.", + "D": "When you are in the room and need to go to the car, you should turn right to exit the door.", + "E": "When you are in the room and need to go to the car, turn right first, then turn left before exiting." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 21 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_4", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was coloring, how many paint palettes were on the stool to the right?", + "answer": "When you are coloring, there are ten paint dishes on the stool to the right.", + "question_chinese": "当我在涂颜色时，右边的凳子上有几个调色碟？", + "answer_chinese": "当你在涂颜色时，右边的凳子上有10个调色碟。", + "evidence": [ + { + "description": "There are ten palette dishes placed on the stool.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:03:18", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Eight paint dishes.", + "B": "Eleven paint dishes.", + "C": "When you are coloring, there are nine paint dishes on the stool to the right.", + "D": "When you are coloring, there are twelve paint dishes on the stool to the right.", + "E": "When you are coloring, there are ten paint dishes on the stool to the right." + }, + "answer_label": "E", + "answer_text": "When you are coloring, there are ten paint dishes on the stool to the right.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_4", + "correct_answer": "When you are coloring, there are ten paint dishes on the stool to the right.", + "original_options": { + "A": "Eight paint dishes.", + "B": "Eleven paint dishes.", + "C": "When you are coloring, there are nine paint dishes on the stool to the right.", + "D": "When you are coloring, there are twelve paint dishes on the stool to the right.", + "E": "When you are coloring, there are ten paint dishes on the stool to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_3", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was coloring, how many paint palettes were on the stool to the right?", + "answer": "When you are coloring, there are nine paint dishes on the stool to the right.", + "question_chinese": "当我在涂颜色时，右边的凳子上有几个调色碟？", + "answer_chinese": "当你在涂颜色时，右边的凳子上有9个调色碟。", + "evidence": [ + { + "description": "There are nine palette dishes placed on the stool.", + "timestamps": { + "start": "00:11:22", + "end": "00:11:26" + } + } + ], + "question_timestamps": "00:19:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Eight paint dishes.", + "B": "Ten paint dishes.", + "C": "When you are coloring, there are seven paint dishes on the stool to the right.", + "D": "When you are coloring, there are nine paint dishes on the stool to the right.", + "E": "When you are coloring, there are eleven paint dishes on the stool to the right." + }, + "answer_label": "D", + "answer_text": "When you are coloring, there are nine paint dishes on the stool to the right.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_3", + "correct_answer": "When you are coloring, there are nine paint dishes on the stool to the right.", + "original_options": { + "A": "Eight paint dishes.", + "B": "Ten paint dishes.", + "C": "When you are coloring, there are seven paint dishes on the stool to the right.", + "D": "When you are coloring, there are nine paint dishes on the stool to the right.", + "E": "When you are coloring, there are eleven paint dishes on the stool to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_12", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was coloring, which was closer to me, the red paint dish or the yellow paint dish?", + "answer": "When you are coloring, the red paint dish is closer to you.", + "question_chinese": "当我在涂颜色时，红色颜料碟和黄色颜料碟哪个离我更近？", + "answer_chinese": "当你在涂颜色时，红色颜料碟离你更近。", + "evidence": [ + { + "description": "The red pigment dish is closest to me.", + "timestamps": { + "start": "00:01:34", + "end": "00:01:37" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you are coloring, the red paint dish is closer to you.", + "B": "When you are coloring, the yellow paint dish is closer to you." + }, + "answer_label": "A", + "answer_text": "When you are coloring, the red paint dish is closer to you.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_12", + "correct_answer": "When you are coloring, the red paint dish is closer to you.", + "original_options": { + "A": "When you are coloring, the red paint dish is closer to you.", + "B": "When you are coloring, the yellow paint dish is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_11", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was coloring, which was closer to me, the red paint dish or the yellow paint dish?", + "answer": "When you are coloring, the yellow paint dish is closer to you.", + "question_chinese": "当我在涂颜色时，红色颜料碟和黄色颜料碟哪个离我更近？", + "answer_chinese": "当你在涂颜色时，黄色颜料碟离你更近。", + "evidence": [ + { + "description": "The red pigment dish is closest to me.", + "timestamps": { + "start": "00:01:34", + "end": "00:01:37" + } + }, + { + "description": "I moved the red paint dish to the right side of the table.", + "timestamps": { + "start": "00:05:26", + "end": "00:05:32" + } + } + ], + "question_timestamps": "00:16:29", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "When you are coloring, the red paint dish is closer to you.", + "B": "When you are coloring, the yellow paint dish is closer to you." + }, + "answer_label": "B", + "answer_text": "When you are coloring, the yellow paint dish is closer to you.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_11", + "correct_answer": "When you are coloring, the yellow paint dish is closer to you.", + "original_options": { + "A": "When you are coloring, the red paint dish is closer to you.", + "B": "When you are coloring, the yellow paint dish is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_6_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I face the sofa, is the tall wooden table to the left/right of the sofa?", + "answer": "When you face the sofa, the tall wooden table is on the right side of the sofa.", + "question_chinese": "当我面对沙发时，高的木桌在沙发的左/右？", + "answer_chinese": "当你面对沙发时，高的木桌在沙发的右边。", + "evidence": [ + { + "description": "The tall wooden table is on the right side of the sofa.", + "timestamps": { + "start": "00:11:16", + "end": "00:11:18" + } + } + ], + "question_timestamps": "00:12:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is on the left side.", + "B": "When you face the sofa, the tall wooden table is on the right side of the sofa.", + "C": "It is in front of the sofa.", + "D": "When you face the sofa, the tall wooden table is centered in front of the sofa.", + "E": "When you face the sofa, the tall wooden table sits behind the sofa rather than beside it." + }, + "answer_label": "B", + "answer_text": "When you face the sofa, the tall wooden table is on the right side of the sofa.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_6_T1", + "correct_answer": "When you face the sofa, the tall wooden table is on the right side of the sofa.", + "original_options": { + "A": "It is on the left side.", + "B": "When you face the sofa, the tall wooden table is on the right side of the sofa.", + "C": "It is in front of the sofa.", + "D": "When you face the sofa, the tall wooden table is centered in front of the sofa.", + "E": "When you face the sofa, the tall wooden table sits behind the sofa rather than beside it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_6_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I face the sofa, is the tall wooden table to the left/right of the sofa?", + "answer": "When you face the sofa, the tall wooden table is on the right side of the sofa.", + "question_chinese": "当我面对沙发时，高的木桌在沙发的左/右？", + "answer_chinese": "当你面对沙发时，高的木桌在沙发的右边。", + "evidence": [ + { + "description": "The tall wooden table is on the right side of the sofa.", + "timestamps": { + "start": "00:11:16", + "end": "00:11:18" + } + } + ], + "question_timestamps": "00:18:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is on the left.", + "B": "It is in front of the sofa.", + "C": "When you face the sofa, the tall wooden table is behind the sofa rather than off to either side.", + "D": "When you face the sofa, the tall wooden table is in front of the sofa, not off to either side.", + "E": "When you face the sofa, the tall wooden table is on the right side of the sofa." + }, + "answer_label": "E", + "answer_text": "When you face the sofa, the tall wooden table is on the right side of the sofa.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_6_T2", + "correct_answer": "When you face the sofa, the tall wooden table is on the right side of the sofa.", + "original_options": { + "A": "It is on the left.", + "B": "It is in front of the sofa.", + "C": "When you face the sofa, the tall wooden table is behind the sofa rather than off to either side.", + "D": "When you face the sofa, the tall wooden table is in front of the sofa, not off to either side.", + "E": "When you face the sofa, the tall wooden table is on the right side of the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 19, + 20 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_1_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I faced the black round table, how many clay sculptures were on the table?", + "answer": "When you face the black round table, there is a clay sculpture on itfour.", + "question_chinese": "当我面对黑色圆桌时，桌子上有几个泥塑？", + "answer_chinese": "当你面对黑色圆桌时，桌子上有四个泥塑。", + "evidence": [ + { + "description": "I'm painting a clay sculpture on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + }, + { + "description": "I took another clay sculpture and placed it on the table.", + "timestamps": { + "start": "00:16:03", + "end": "00:16:09" + } + }, + { + "description": "I took two more clay sculptures and put them on the table.", + "timestamps": { + "start": "00:19:11", + "end": "00:19:17" + } + } + ], + "question_timestamps": "00:22:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two clay sculptures.", + "B": "Three clay sculptures.", + "C": "When you face the black round table, there is a clay sculpture on itfour.", + "D": "When you face the black round table, there are five clay sculptures placed on top of it.", + "E": "When you face the black round table, there are no clay sculptures on the table at all." + }, + "answer_label": "C", + "answer_text": "When you face the black round table, there is a clay sculpture on itfour.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_1_T1", + "correct_answer": "When you face the black round table, there is a clay sculpture on itfour.", + "original_options": { + "A": "Two clay sculptures.", + "B": "Three clay sculptures.", + "C": "When you face the black round table, there is a clay sculpture on itfour.", + "D": "When you face the black round table, there are five clay sculptures placed on top of it.", + "E": "When you face the black round table, there are no clay sculptures on the table at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_1_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I faced the black round table, how many clay sculptures were on the table?", + "answer": "When you face the black round table, there is a clay sculpture on itfour.", + "question_chinese": "当我面对黑色圆桌时，桌子上有几个泥塑？", + "answer_chinese": "当你面对黑色圆桌时，桌子上有四个泥塑。", + "evidence": [ + { + "description": "I'm painting a clay sculpture on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + }, + { + "description": "I took another clay sculpture and placed it on the table.", + "timestamps": { + "start": "00:16:03", + "end": "00:16:09" + } + }, + { + "description": "I took two more clay sculptures and put them on the table.", + "timestamps": { + "start": "00:19:11", + "end": "00:19:17" + } + } + ], + "question_timestamps": "00:22:38", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero clay sculptures.", + "B": "Five clay sculptures.", + "C": "When you face the black round table, there is a clay sculpture on itfour.", + "D": "When you face the black round table, there are two clay sculptures sitting on the tabletop.", + "E": "When you face the black round table, there are three clay sculptures arranged on the table." + }, + "answer_label": "C", + "answer_text": "When you face the black round table, there is a clay sculpture on itfour.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_1_T2", + "correct_answer": "When you face the black round table, there is a clay sculpture on itfour.", + "original_options": { + "A": "Zero clay sculptures.", + "B": "Five clay sculptures.", + "C": "When you face the black round table, there is a clay sculpture on itfour.", + "D": "When you face the black round table, there are two clay sculptures sitting on the tabletop.", + "E": "When you face the black round table, there are three clay sculptures arranged on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_7_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far was the straight-line distance from me to the scissors when I just stood up from the chair?", + "answer": "When you just stood up from the stool, the straight-line distance to the scissors was one meter.", + "question_chinese": "我刚从凳子上站起来时，距离剪刀直线距离有多远？", + "answer_chinese": "你刚从凳子上站起来时，距离剪刀直线距离有一米。", + "evidence": [ + { + "description": "Scissors were placed on the long wooden table in front of me.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:14" + } + } + ], + "question_timestamps": "00:13:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "When you just stood up from the stool, the straight-line distance to the scissors was one meter.", + "B": "Half a meter.", + "C": "Two meters.", + "D": "When you just stood up from the stool, the straight-line distance to the scissors was three meters.", + "E": "When you just stood up from the stool, the straight-line distance to the scissors was one and a half meters." + }, + "answer_label": "A", + "answer_text": "When you just stood up from the stool, the straight-line distance to the scissors was one meter.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_7_T1", + "correct_answer": "When you just stood up from the stool, the straight-line distance to the scissors was one meter.", + "original_options": { + "A": "When you just stood up from the stool, the straight-line distance to the scissors was one meter.", + "B": "Half a meter.", + "C": "Two meters.", + "D": "When you just stood up from the stool, the straight-line distance to the scissors was three meters.", + "E": "When you just stood up from the stool, the straight-line distance to the scissors was one and a half meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 17, + 20 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_7_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far was the straight-line distance from me to the scissors when I just stood up from the chair?", + "answer": "When you just stood up from the stool, the straight-line distance to the scissors was one meter.", + "question_chinese": "我刚从凳子上站起来时，距离剪刀直线距离有多远？", + "answer_chinese": "你刚从凳子上站起来时，距离剪刀直线距离有一米。", + "evidence": [ + { + "description": "Scissors were placed on the long wooden table in front of me.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:14" + } + } + ], + "question_timestamps": "00:21:02", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "1.2 meters.", + "B": "2 meters.", + "C": "When you just stood up from the stool, the straight-line distance to the scissors was one meter.", + "D": "When you just stood up from the stool, the straight-line distance to the scissors was 1.5 meters.", + "E": "When you just stood up from the stool, the straight-line distance to the scissors was 1.3 meters." + }, + "answer_label": "C", + "answer_text": "When you just stood up from the stool, the straight-line distance to the scissors was one meter.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_7_T2", + "correct_answer": "When you just stood up from the stool, the straight-line distance to the scissors was one meter.", + "original_options": { + "A": "1.2 meters.", + "B": "2 meters.", + "C": "When you just stood up from the stool, the straight-line distance to the scissors was one meter.", + "D": "When you just stood up from the stool, the straight-line distance to the scissors was 1.5 meters.", + "E": "When you just stood up from the stool, the straight-line distance to the scissors was 1.3 meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 18, + 18 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_4_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, how many doors have I seen in the scene?", + "answer": "So far, you have seen three doors in the scene.", + "question_chinese": "目前为止，我在场景里看到了多少个门？", + "answer_chinese": "目前为止，你在场景里看到了三个门。", + "evidence": [ + { + "description": "I was sitting on the stool with a door next to me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "There is a door opposite the table with the paint palette.", + "timestamps": { + "start": "00:11:12", + "end": "00:11:14" + } + }, + { + "description": "There is a door on the right side of the sofa.", + "timestamps": { + "start": "00:11:14", + "end": "00:11:16" + } + } + ], + "question_timestamps": "00:17:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two doors.", + "B": "Four doors.", + "C": "So far, you have seen one door in the scene, not three.", + "D": "So far, you have seen five doors in the scene up to this point.", + "E": "So far, you have seen three doors in the scene." + }, + "answer_label": "E", + "answer_text": "So far, you have seen three doors in the scene.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_4_T1", + "correct_answer": "So far, you have seen three doors in the scene.", + "original_options": { + "A": "Two doors.", + "B": "Four doors.", + "C": "So far, you have seen one door in the scene, not three.", + "D": "So far, you have seen five doors in the scene up to this point.", + "E": "So far, you have seen three doors in the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326", + "video_length": "00:22:47", + "qid": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_4_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, how many doors have I seen in the scene?", + "answer": "So far, you have seen three doors in the scene.", + "question_chinese": "目前为止，我在场景里看到了多少个门？", + "answer_chinese": "目前为止，你在场景里看到了三个门。", + "evidence": [ + { + "description": "I was sitting on the stool with a door next to me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "There is a door opposite the table with the paint palette.", + "timestamps": { + "start": "00:11:12", + "end": "00:11:14" + } + }, + { + "description": "There is a door on the right side of the sofa.", + "timestamps": { + "start": "00:11:14", + "end": "00:11:16" + } + } + ], + "question_timestamps": "00:17:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One door.", + "B": "Five doors.", + "C": "So far, you have seen two doors in the scene, not three.", + "D": "So far, you have seen four doors in the scene up to this point.", + "E": "So far, you have seen three doors in the scene." + }, + "answer_label": "E", + "answer_text": "So far, you have seen three doors in the scene.", + "q_id": "2caa83ec-4e4f-426e-b177-2bbf6ca81326_4_T2", + "correct_answer": "So far, you have seen three doors in the scene.", + "original_options": { + "A": "One door.", + "B": "Five doors.", + "C": "So far, you have seen two doors in the scene, not three.", + "D": "So far, you have seen four doors in the scene up to this point.", + "E": "So far, you have seen three doors in the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + } +] diff --git a/QAs/2d076447-17ad-47ab-b846-7eab0bf79090-mcq.json b/QAs/2d076447-17ad-47ab-b846-7eab0bf79090-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1ed7236f6ca3855a418562af8c5417cb2d21fa44 --- /dev/null +++ b/QAs/2d076447-17ad-47ab-b846-7eab0bf79090-mcq.json @@ -0,0 +1,1588 @@ +[ + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "From the moment I entered the door until now, what is the trajectory of this red drill in my hand?", + "answer": "The red drill was carried by you as you entered the door and walked straight, passing the wooden table and the storage rack in turn, to the sink, then turned right in front of the sink, moved to the workbench with the green toolbox, and finally ended up under the workbench.", + "question_chinese": "从刚才进门到现在，我手上这把红色电钻的运动轨迹是怎样的？", + "answer_chinese": "红色电钻被你拿着进门后直走，依次经过木桌和杂物架，走到洗手台前，然后在洗手台前右转，运动到放着绿色工具箱的工作台前，最后到了工作台的下面。", + "evidence": [ + { + "description": "You entered the door with the red drill, walked straight past the wooden table and the storage rack, and went to the sink.", + "timestamps": { + "start": "00:41:18", + "end": "00:41:28" + } + }, + { + "description": "Turn right in front of the sink, then walk straight to the workbench with the green toolbox and lie down under the workbench.", + "timestamps": { + "start": "00:41:28", + "end": "00:41:46" + } + } + ], + "question_timestamps": "00:41:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turned left at the sink.", + "B": "The red drill was carried by you as you entered the door and walked straight, passing the wooden table and the storage rack in turn, to the sink, then turned right in front of the sink, moved to the workbench with the green toolbox, and finally ended up under the workbench.", + "C": "It ended on top of the workbench.", + "D": "You carried the red drill straight past the wooden table and storage rack to the sink, then turned right toward it.", + "E": "You entered carrying the red drill, went straight to the sink, turned right, and placed it beside the workbench." + }, + "answer_label": "B", + "answer_text": "The red drill was carried by you as you entered the door and walked straight, passing the wooden table and the storage rack in turn, to the sink, then turned right in front of the sink, moved to the workbench with the green toolbox, and finally ended up under the workbench.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_7", + "correct_answer": "The red drill was carried by you as you entered the door and walked straight, passing the wooden table and the storage rack in turn, to the sink, then turned right in front of the sink, moved to the workbench with the green toolbox, and finally ended up under the workbench.", + "original_options": { + "A": "You turned left at the sink.", + "B": "The red drill was carried by you as you entered the door and walked straight, passing the wooden table and the storage rack in turn, to the sink, then turned right in front of the sink, moved to the workbench with the green toolbox, and finally ended up under the workbench.", + "C": "It ended on top of the workbench.", + "D": "You carried the red drill straight past the wooden table and storage rack to the sink, then turned right toward it.", + "E": "You entered carrying the red drill, went straight to the sink, turned right, and placed it beside the workbench." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 19 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_3_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What has my movement path been from when I entered the door until now?", + "answer": "After you entered, you walked straight, passing the wooden table and then the storage rack, until you reached the sink. Then you turned right at the sink.", + "question_chinese": "我刚才从进门到现在的运动轨迹是怎样的？", + "answer_chinese": "你进门后直走，依次经过木桌和杂物架，走到洗手台前，然后在洗手台前右转", + "evidence": [ + { + "description": "After entering the door, walk straight past the wooden table and the storage rack to the sink.", + "timestamps": { + "start": "00:41:18", + "end": "00:41:28" + } + }, + { + "description": "Turn right in front of the sink", + "timestamps": { + "start": "00:41:28", + "end": "00:41:29" + } + } + ], + "question_timestamps": "00:41:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left at the sink.", + "B": "You stopped at the storage rack.", + "C": "After you entered, you walked straight, passing the wooden table and then the storage rack, until you reached the sink. Then you turned right at the sink.", + "D": "After entering, you walked straight past the wooden table and storage rack, reached the sink, then turned left.", + "E": "You entered, walked straight past the wooden table and storage rack to the sink, then continued straight without turning." + }, + "answer_label": "C", + "answer_text": "After you entered, you walked straight, passing the wooden table and then the storage rack, until you reached the sink. Then you turned right at the sink.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_3_T1", + "correct_answer": "After you entered, you walked straight, passing the wooden table and then the storage rack, until you reached the sink. Then you turned right at the sink.", + "original_options": { + "A": "You turned left at the sink.", + "B": "You stopped at the storage rack.", + "C": "After you entered, you walked straight, passing the wooden table and then the storage rack, until you reached the sink. Then you turned right at the sink.", + "D": "After entering, you walked straight past the wooden table and storage rack, reached the sink, then turned left.", + "E": "You entered, walked straight past the wooden table and storage rack to the sink, then continued straight without turning." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_5", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my trajectory from the door to now?", + "answer": "After you entered the door, you walked straight, passed the wooden table and the storage rack in turn, walked to the sink, then turned right in front of the sink and walked straight to the green toolbox.", + "question_chinese": "我刚才从进门到现在的运动轨迹是怎样的？", + "answer_chinese": "你进门后直走，依次经过木桌和杂物架，走到洗手台前，然后在洗手台前右转，直走到绿色工具箱前。", + "evidence": [ + { + "description": "After entering the door, walk straight past the wooden table and the storage rack to the sink.", + "timestamps": { + "start": "00:41:18", + "end": "00:41:28" + } + }, + { + "description": "Turn right in front of the sink, then walk straight to the green toolbox.", + "timestamps": { + "start": "00:41:28", + "end": "00:41:34" + } + } + ], + "question_timestamps": "00:41:36", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "After you entered the door, you walked straight, passed the wooden table and the storage rack in turn, walked to the sink, then turned right in front of the sink and walked straight to the green toolbox.", + "B": "You turned left at the sink.", + "C": "You never reached the sink.", + "D": "After entering, you walked straight past the wooden table and storage rack to the sink, then turned right and stopped.", + "E": "You entered, passed the wooden table and storage rack, reached the sink, then turned left toward the green toolbox." + }, + "answer_label": "A", + "answer_text": "After you entered the door, you walked straight, passed the wooden table and the storage rack in turn, walked to the sink, then turned right in front of the sink and walked straight to the green toolbox.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_5", + "correct_answer": "After you entered the door, you walked straight, passed the wooden table and the storage rack in turn, walked to the sink, then turned right in front of the sink and walked straight to the green toolbox.", + "original_options": { + "A": "After you entered the door, you walked straight, passed the wooden table and the storage rack in turn, walked to the sink, then turned right in front of the sink and walked straight to the green toolbox.", + "B": "You turned left at the sink.", + "C": "You never reached the sink.", + "D": "After entering, you walked straight past the wooden table and storage rack to the sink, then turned right and stopped.", + "E": "You entered, passed the wooden table and storage rack, reached the sink, then turned left toward the green toolbox." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 19 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_3_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What has my movement path been from when I entered the door until now?", + "answer": "After you entered, you walked straight, passing the wooden table and then the storage rack, until you reached the sink. Then you turned right at the sink and walked straight to the green toolbox.", + "question_chinese": "我刚才从进门到现在的运动轨迹是怎样的？", + "answer_chinese": "你进门后直走，依次经过木桌和杂物架，走到洗手台前，然后在洗手台前右转，直走到绿色工具箱前。", + "evidence": [ + { + "description": "After entering the door, walk straight past the wooden table and the storage rack to the sink.", + "timestamps": { + "start": "00:41:18", + "end": "00:41:28" + } + }, + { + "description": "Turn right in front of the sink, then walk straight to the green toolbox.", + "timestamps": { + "start": "00:41:28", + "end": "00:41:34" + } + } + ], + "question_timestamps": "00:41:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left at the sink.", + "B": "After you entered, you walked straight, passing the wooden table and then the storage rack, until you reached the sink. Then you turned right at the sink and walked straight to the green toolbox.", + "C": "You stopped at the sink.", + "D": "After entering, you walked past the wooden table and storage rack to the sink, turned right, but didn’t continue forward.", + "E": "You walked straight past the wooden table and storage rack to the sink, turned left, and continued to the green toolbox." + }, + "answer_label": "B", + "answer_text": "After you entered, you walked straight, passing the wooden table and then the storage rack, until you reached the sink. Then you turned right at the sink and walked straight to the green toolbox.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_3_T2", + "correct_answer": "After you entered, you walked straight, passing the wooden table and then the storage rack, until you reached the sink. Then you turned right at the sink and walked straight to the green toolbox.", + "original_options": { + "A": "You turned left at the sink.", + "B": "After you entered, you walked straight, passing the wooden table and then the storage rack, until you reached the sink. Then you turned right at the sink and walked straight to the green toolbox.", + "C": "You stopped at the sink.", + "D": "After entering, you walked past the wooden table and storage rack to the sink, turned right, but didn’t continue forward.", + "E": "You walked straight past the wooden table and storage rack to the sink, turned left, and continued to the green toolbox." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 21 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_9", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my trajectory from getting the drill from the car to entering the door?", + "answer": "After you picked up the drill from the car, you turned right and went up the stairs. After going up the stairs, you turned right and walked straight into the door.", + "question_chinese": "我刚才去车上拿电钻，到我进门的运动轨迹是怎样的？", + "answer_chinese": "你先从车上拿起电钻后，右转走上楼梯，上楼梯后右转，直走进门。", + "evidence": [ + { + "description": "After picking up the drill from the car, turn right and go up the stairs.", + "timestamps": { + "start": "00:41:01", + "end": "00:41:12" + } + }, + { + "description": "After going up the stairs, turn right and walk straight in the door.", + "timestamps": { + "start": "00:41:12", + "end": "00:41:20" + } + } + ], + "question_timestamps": "00:41:24", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turned left after the car.", + "B": "You went down the stairs.", + "C": "After picking up the drill from the car, you turned right, went up the stairs, then turned left into the door.", + "D": "After you picked up the drill from the car, you turned right and went up the stairs. After going up the stairs, you turned right and walked straight into the door.", + "E": "You picked up the drill, went up the stairs, turned right, and walked straight past the door instead of entering." + }, + "answer_label": "D", + "answer_text": "After you picked up the drill from the car, you turned right and went up the stairs. After going up the stairs, you turned right and walked straight into the door.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_9", + "correct_answer": "After you picked up the drill from the car, you turned right and went up the stairs. After going up the stairs, you turned right and walked straight into the door.", + "original_options": { + "A": "You turned left after the car.", + "B": "You went down the stairs.", + "C": "After picking up the drill from the car, you turned right, went up the stairs, then turned left into the door.", + "D": "After you picked up the drill from the car, you turned right and went up the stairs. After going up the stairs, you turned right and walked straight into the door.", + "E": "You picked up the drill, went up the stairs, turned right, and walked straight past the door instead of entering." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 20 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_1_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take just now while carrying the electric drill?", + "answer": "You first picked up the red electric drill from the workbench to your left, then walked to the green toolbox on the workbench to your right.", + "question_chinese": "我刚才拿着电钻经过了什么路径？", + "answer_chinese": "你先从左边的工作台拿起红色电钻，然后走到右边工作台上的绿色工具箱前", + "evidence": [ + { + "description": "Pick up the red drill from the workbench on the left.", + "timestamps": { + "start": "00:30:40", + "end": "00:30:43" + } + }, + { + "description": "Walk to the green toolbox on the workbench on the right.", + "timestamps": { + "start": "00:30:44", + "end": "00:30:51" + } + } + ], + "question_timestamps": "00:30:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked it up from in front.", + "B": "You first picked up the red electric drill from the workbench to your left, then walked to the green toolbox on the workbench to your right.", + "C": "You grabbed it from the right bench.", + "D": "You picked up the red drill from the workbench behind you, then walked to the green toolbox on the right.", + "E": "You lifted the red drill from the bench in front of you and moved to the green toolbox on your right." + }, + "answer_label": "B", + "answer_text": "You first picked up the red electric drill from the workbench to your left, then walked to the green toolbox on the workbench to your right.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_1_T1", + "correct_answer": "You first picked up the red electric drill from the workbench to your left, then walked to the green toolbox on the workbench to your right.", + "original_options": { + "A": "You picked it up from in front.", + "B": "You first picked up the red electric drill from the workbench to your left, then walked to the green toolbox on the workbench to your right.", + "C": "You grabbed it from the right bench.", + "D": "You picked up the red drill from the workbench behind you, then walked to the green toolbox on the right.", + "E": "You lifted the red drill from the bench in front of you and moved to the green toolbox on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 21 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I just take with the drill?", + "answer": "You first picked up the red drill from the workbench on the left, then walked to the green toolbox on the workbench on the right, took a drill bit out of the green toolbox and installed it on the drill.", + "question_chinese": "我刚才拿着电钻经过了什么路径？", + "answer_chinese": "你先从左边的工作台拿起红色电钻，然后走到右边工作台上的绿色工具箱前，从绿色工具箱里拿出一个钻头并装在电钻上。", + "evidence": [ + { + "description": "Pick up the red drill from the workbench on the left.", + "timestamps": { + "start": "00:30:40", + "end": "00:30:43" + } + }, + { + "description": "Walk to the green toolbox on the workbench on the right.", + "timestamps": { + "start": "00:30:44", + "end": "00:30:51" + } + }, + { + "description": "Take a drill bit from the green toolbox and install it on the drill.", + "timestamps": { + "start": "00:30:51", + "end": "00:31:05" + } + } + ], + "question_timestamps": "00:31:06", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You went left to the toolbox.", + "B": "You didn’t install a drill bit.", + "C": "You picked up the red drill on the right, then went to the green toolbox on the left and installed a bit.", + "D": "You took the red drill from the left workbench, walked to the green toolbox on the right, and left it there.", + "E": "You first picked up the red drill from the workbench on the left, then walked to the green toolbox on the workbench on the right, took a drill bit out of the green toolbox and installed it on the drill." + }, + "answer_label": "E", + "answer_text": "You first picked up the red drill from the workbench on the left, then walked to the green toolbox on the workbench on the right, took a drill bit out of the green toolbox and installed it on the drill.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_1", + "correct_answer": "You first picked up the red drill from the workbench on the left, then walked to the green toolbox on the workbench on the right, took a drill bit out of the green toolbox and installed it on the drill.", + "original_options": { + "A": "You went left to the toolbox.", + "B": "You didn’t install a drill bit.", + "C": "You picked up the red drill on the right, then went to the green toolbox on the left and installed a bit.", + "D": "You took the red drill from the left workbench, walked to the green toolbox on the right, and left it there.", + "E": "You first picked up the red drill from the workbench on the left, then walked to the green toolbox on the workbench on the right, took a drill bit out of the green toolbox and installed it on the drill." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 22, + 21 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_1_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take just now while carrying the electric drill?", + "answer": "You first picked up the red electric drill from the workbench to your left, then walked to the green toolbox on the workbench to your right, took out a drill bit from the green toolbox, and mounted it onto the drill.", + "question_chinese": "我刚才拿着电钻经过了什么路径？", + "answer_chinese": "你先从左边的工作台拿起红色电钻，然后走到右边工作台上的绿色工具箱前，从绿色工具箱里拿出一个钻头并装在电钻上。", + "evidence": [ + { + "description": "Pick up the red drill from the workbench on the left.", + "timestamps": { + "start": "00:30:40", + "end": "00:30:43" + } + }, + { + "description": "Walk to the green toolbox on the workbench on the right.", + "timestamps": { + "start": "00:30:44", + "end": "00:30:51" + } + }, + { + "description": "Take a drill bit from the green toolbox and install it on the drill.", + "timestamps": { + "start": "00:30:51", + "end": "00:31:05" + } + } + ], + "question_timestamps": "00:32:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked it up from the right workbench.", + "B": "You went to the toolbox first.", + "C": "You grabbed the drill from the workbench in front of you, then took a bit from the green toolbox.", + "D": "You picked up the drill from the left workbench, walked to the toolbox on your left, and mounted a bit.", + "E": "You first picked up the red electric drill from the workbench to your left, then walked to the green toolbox on the workbench to your right, took out a drill bit from the green toolbox, and mounted it onto the drill." + }, + "answer_label": "E", + "answer_text": "You first picked up the red electric drill from the workbench to your left, then walked to the green toolbox on the workbench to your right, took out a drill bit from the green toolbox, and mounted it onto the drill.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_1_T2", + "correct_answer": "You first picked up the red electric drill from the workbench to your left, then walked to the green toolbox on the workbench to your right, took out a drill bit from the green toolbox, and mounted it onto the drill.", + "original_options": { + "A": "You picked it up from the right workbench.", + "B": "You went to the toolbox first.", + "C": "You grabbed the drill from the workbench in front of you, then took a bit from the green toolbox.", + "D": "You picked up the drill from the left workbench, walked to the toolbox on your left, and mounted a bit.", + "E": "You first picked up the red electric drill from the workbench to your left, then walked to the green toolbox on the workbench to your right, took out a drill bit from the green toolbox, and mounted it onto the drill." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 19, + 20 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_7_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the pair of pliers with the blue handle that I just used?", + "answer": "The blue-handled pliers are on the countertop in front of you.", + "question_chinese": "我刚用的那把蓝柄的钳子在哪里？", + "answer_chinese": "蓝柄的钳子在你身前的台面上。", + "evidence": [ + { + "description": "The blue-handled pliers are on the counter in front of you.", + "timestamps": { + "start": "00:06:59", + "end": "00:07:00" + } + } + ], + "question_timestamps": "00:07:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They’re on the countertop behind you.", + "B": "The blue-handled pliers are on the countertop in front of you.", + "C": "They’re on the countertop to your left.", + "D": "The blue-handled pliers are on the countertop to your right, not directly in front of you.", + "E": "The blue-handled pliers are on the countertop behind you, set back from where you were working." + }, + "answer_label": "B", + "answer_text": "The blue-handled pliers are on the countertop in front of you.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_7_T2", + "correct_answer": "The blue-handled pliers are on the countertop in front of you.", + "original_options": { + "A": "They’re on the countertop behind you.", + "B": "The blue-handled pliers are on the countertop in front of you.", + "C": "They’re on the countertop to your left.", + "D": "The blue-handled pliers are on the countertop to your right, not directly in front of you.", + "E": "The blue-handled pliers are on the countertop behind you, set back from where you were working." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 16 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_7_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the pair of pliers with the blue handle that I just used?", + "answer": "The blue-handled pliers are on the countertop behind you.", + "question_chinese": "我刚用的那把蓝柄的钳子在哪里？", + "answer_chinese": "蓝柄的钳子在你身后的台面上。", + "evidence": [ + { + "description": "The blue-handled pliers are on the counter behind you.", + "timestamps": { + "start": "00:07:10", + "end": "00:08:12" + } + } + ], + "question_timestamps": "00:08:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the countertop in front of you.", + "B": "They’re on the countertop to your right.", + "C": "The blue-handled pliers are on the countertop behind you.", + "D": "The blue-handled pliers are on the countertop to your left, not behind you where you were standing.", + "E": "The blue-handled pliers are on the countertop in front of you, closer to the edge than the back." + }, + "answer_label": "C", + "answer_text": "The blue-handled pliers are on the countertop behind you.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_7_T1", + "correct_answer": "The blue-handled pliers are on the countertop behind you.", + "original_options": { + "A": "On the countertop in front of you.", + "B": "They’re on the countertop to your right.", + "C": "The blue-handled pliers are on the countertop behind you.", + "D": "The blue-handled pliers are on the countertop to your left, not behind you where you were standing.", + "E": "The blue-handled pliers are on the countertop in front of you, closer to the edge than the back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 18 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_8_T2", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pick up this red electric drill in my hand from?", + "answer": "You got it from the car outside.", + "question_chinese": "我手上这把红色的电钻是从哪里拿到的？", + "answer_chinese": "你从外面的车上拿到的。", + "evidence": [ + { + "description": "You took it out of the trunk.", + "timestamps": { + "start": "00:41:02", + "end": "00:41:03" + } + } + ], + "question_timestamps": "00:41:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the workbench.", + "B": "You got it from the car outside.", + "C": "From the sink area.", + "D": "You picked it up from the workbench beside the green toolbox, right before you moved toward the sink.", + "E": "You grabbed it from near the sink, then carried it past the wooden table and the storage rack." + }, + "answer_label": "B", + "answer_text": "You got it from the car outside.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_8_T2", + "correct_answer": "You got it from the car outside.", + "original_options": { + "A": "From the workbench.", + "B": "You got it from the car outside.", + "C": "From the sink area.", + "D": "You picked it up from the workbench beside the green toolbox, right before you moved toward the sink.", + "E": "You grabbed it from near the sink, then carried it past the wooden table and the storage rack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 18, + 18 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_8_T1", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pick up this red electric drill in my hand from?", + "answer": "You got it from the trunk of the car outside.", + "question_chinese": "我手上这把红色的电钻是从哪里拿到的？", + "answer_chinese": "从外面的车后尾箱上拿到的。", + "evidence": [ + { + "description": "Get the red drill from the car outside, then enter the door.", + "timestamps": { + "start": "00:40:52", + "end": "00:41:22" + } + } + ], + "question_timestamps": "00:41:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You got it from the car outside.", + "B": "You got it from inside the car.", + "C": "You got it from the trunk of the car outside.", + "D": "You picked up the red electric drill from the back seat inside the car, not from the trunk outside.", + "E": "You got the red electric drill from the car outside, but not from the trunk area specifically." + }, + "answer_label": "C", + "answer_text": "You got it from the trunk of the car outside.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_8_T1", + "correct_answer": "You got it from the trunk of the car outside.", + "original_options": { + "A": "You got it from the car outside.", + "B": "You got it from inside the car.", + "C": "You got it from the trunk of the car outside.", + "D": "You picked up the red electric drill from the back seat inside the car, not from the trunk outside.", + "E": "You got the red electric drill from the car outside, but not from the trunk area specifically." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 17 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_11", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the sink now?", + "answer": "No, you can't.", + "question_chinese": "我现在可以够到洗手台吗？", + "answer_chinese": "不能够到。", + "evidence": [ + { + "description": "The sink is about 1.5 meters to the left of the red toolbox.", + "timestamps": { + "start": "00:06:01", + "end": "00:06:07" + } + }, + { + "description": "You are now standing in front of the red toolbox.", + "timestamps": { + "start": "00:06:08", + "end": "00:06:17" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No, you can't.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_11", + "correct_answer": "No, you can't.", + "original_options": { + "A": "No, you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_6_T2", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the sink right now?", + "answer": "No.", + "question_chinese": "我现在可以够到洗手台吗？", + "answer_chinese": "不", + "evidence": [ + { + "description": "The sink is about 1.5 meters to the left of the red toolbox.", + "timestamps": { + "start": "00:06:01", + "end": "00:06:07" + } + }, + { + "description": "You are now standing near the red toolbox.", + "timestamps": { + "start": "00:06:08", + "end": "00:06:20" + } + } + ], + "question_timestamps": "00:06:25", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_6_T2", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_6_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the sink right now?", + "answer": "Yes.", + "question_chinese": "我现在可以够到洗手台吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "The sink is about 1.5 meters to the left of the red toolbox.", + "timestamps": { + "start": "00:06:01", + "end": "00:06:07" + } + }, + { + "description": "You can reach the sink now.", + "timestamps": { + "start": "00:10:14", + "end": "00:10:15" + } + } + ], + "question_timestamps": "00:10:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_6_T1", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_10_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I entered the door, which side of me was the wooden table on?", + "answer": "The wooden table is to your front-left.", + "question_chinese": "我进门的时候，木桌在我哪一侧？", + "answer_chinese": "木桌在你的左前侧。", + "evidence": [ + { + "description": "When you enter the door, the wooden table is on your front left.", + "timestamps": { + "start": "00:41:17", + "end": "00:41:18" + } + } + ], + "question_timestamps": "00:41:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Front-right.", + "B": "Back-left.", + "C": "The wooden table is directly in front of you as you enter.", + "D": "The wooden table is behind you on the right side when you come through the door.", + "E": "The wooden table is to your front-left." + }, + "answer_label": "E", + "answer_text": "The wooden table is to your front-left.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_10_T1", + "correct_answer": "The wooden table is to your front-left.", + "original_options": { + "A": "Front-right.", + "B": "Back-left.", + "C": "The wooden table is directly in front of you as you enter.", + "D": "The wooden table is behind you on the right side when you come through the door.", + "E": "The wooden table is to your front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 16 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_17", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I entered the door, which side was the wooden table on?", + "answer": "The wooden table is on your left.", + "question_chinese": "我进门的时候，木桌在我哪一侧？", + "answer_chinese": "木桌在你的左侧。", + "evidence": [ + { + "description": "When you enter the door, the wooden table is on your left.", + "timestamps": { + "start": "00:41:18", + "end": "00:41:24" + } + } + ], + "question_timestamps": "00:41:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your right.", + "B": "In front of you.", + "C": "The wooden table is behind you as soon as you walk in.", + "D": "The wooden table is on your left.", + "E": "The wooden table is off to your front-right when you enter through the door." + }, + "answer_label": "D", + "answer_text": "The wooden table is on your left.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_17", + "correct_answer": "The wooden table is on your left.", + "original_options": { + "A": "On your right.", + "B": "In front of you.", + "C": "The wooden table is behind you as soon as you walk in.", + "D": "The wooden table is on your left.", + "E": "The wooden table is off to your front-right when you enter through the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 14 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_10_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I entered the door, which side of me was the wooden table on?", + "answer": "The wooden table is on the left side of you.", + "question_chinese": "我进门的时候，木桌在我哪一侧？", + "answer_chinese": "木桌在你的左侧。", + "evidence": [ + { + "description": "When you enter the door, the wooden table is on your left.", + "timestamps": { + "start": "00:41:18", + "end": "00:41:24" + } + } + ], + "question_timestamps": "00:46:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your right side.", + "B": "In front of you.", + "C": "The wooden table is behind you, slightly to the right as you enter.", + "D": "The wooden table is positioned in your front-right area when you come through the door.", + "E": "The wooden table is on the left side of you." + }, + "answer_label": "E", + "answer_text": "The wooden table is on the left side of you.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_10_T2", + "correct_answer": "The wooden table is on the left side of you.", + "original_options": { + "A": "On your right side.", + "B": "In front of you.", + "C": "The wooden table is behind you, slightly to the right as you enter.", + "D": "The wooden table is positioned in your front-right area when you come through the door.", + "E": "The wooden table is on the left side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 15 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_3", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the sink or the green toolbox?", + "answer": "The green toolbox is closer to you.", + "question_chinese": "洗手台和绿色工具箱，哪个离我最近？", + "answer_chinese": "绿色工具箱离你更近。", + "evidence": [ + { + "description": "The sink is about 1.5 meters to the left of the red toolbox.", + "timestamps": { + "start": "00:39:37", + "end": "00:39:41" + } + }, + { + "description": "The green toolbox is about 5 meters to the right of the red toolbox.", + "timestamps": { + "start": "00:39:41", + "end": "00:39:47" + } + }, + { + "description": "You are now standing near the green toolbox.", + "timestamps": { + "start": "00:39:44", + "end": "00:39:50" + } + } + ], + "question_timestamps": "00:39:51", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The green toolbox is closer to you.", + "B": "The sink is closer to you." + }, + "answer_label": "A", + "answer_text": "The green toolbox is closer to you.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_3", + "correct_answer": "The green toolbox is closer to you.", + "original_options": { + "A": "The green toolbox is closer to you.", + "B": "The sink is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_2_T1", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Between the sink and the green toolbox, which one is closer to me?", + "answer": "The green toolbox.", + "question_chinese": "洗手台和绿色工具箱，哪个离我最近？", + "answer_chinese": "绿色工具箱", + "evidence": [ + { + "description": "The sink is about 1.5 meters to the left of the red toolbox.", + "timestamps": { + "start": "00:39:37", + "end": "00:39:41" + } + }, + { + "description": "The green toolbox is about 5 meters to the right of the red toolbox.", + "timestamps": { + "start": "00:39:42", + "end": "00:39:46" + } + }, + { + "description": "You are now standing near the green toolbox.", + "timestamps": { + "start": "00:39:47", + "end": "00:39:50" + } + } + ], + "question_timestamps": "00:39:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sink.", + "B": "The green toolbox.", + "C": "Neither one.", + "D": "They are about the same distance from where you are standing right now.", + "E": "The sink is closer to you, and the green toolbox is farther away." + }, + "answer_label": "B", + "answer_text": "The green toolbox.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_2_T1", + "correct_answer": "The green toolbox.", + "original_options": { + "A": "The sink.", + "B": "The green toolbox.", + "C": "Neither one.", + "D": "They are about the same distance from where you are standing right now.", + "E": "The sink is closer to you, and the green toolbox is farther away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_4", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the sink or the green toolbox?", + "answer": "The green toolbox is closer to you.", + "question_chinese": "洗手台和绿色工具箱，哪个离我最近？", + "answer_chinese": "绿色工具箱离你更近。", + "evidence": [ + { + "description": "The sink is about 1.5 meters to the left of the red toolbox.", + "timestamps": { + "start": "00:39:37", + "end": "00:39:41" + } + }, + { + "description": "The green toolbox is about 5 meters to the right of the red toolbox.", + "timestamps": { + "start": "00:39:41", + "end": "00:39:47" + } + }, + { + "description": "You are now standing near the green toolbox.", + "timestamps": { + "start": "00:39:44", + "end": "00:39:50" + } + } + ], + "question_timestamps": "00:39:59", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The green toolbox is closer to you.", + "B": "The sink is closer to you." + }, + "answer_label": "A", + "answer_text": "The green toolbox is closer to you.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_4", + "correct_answer": "The green toolbox is closer to you.", + "original_options": { + "A": "The green toolbox is closer to you.", + "B": "The sink is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_2_T2", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Between the sink and the green toolbox, which one is closer to me?", + "answer": "The sink.", + "question_chinese": "洗手台和绿色工具箱，哪个离我最近？", + "answer_chinese": "洗手台", + "evidence": [ + { + "description": "The sink is about 1.5 meters to the left of the red toolbox.", + "timestamps": { + "start": "00:39:37", + "end": "00:39:41" + } + }, + { + "description": "The green toolbox is about 5 meters to the right of the red toolbox.", + "timestamps": { + "start": "00:39:41", + "end": "00:39:47" + } + }, + { + "description": "You are now standing near the sink.", + "timestamps": { + "start": "00:40:18", + "end": "00:40:19" + } + } + ], + "question_timestamps": "00:40:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The green toolbox.", + "B": "They’re about the same distance.", + "C": "The sink.", + "D": "The green toolbox is closer to you than the sink is right now.", + "E": "Neither is particularly close; both the sink and the green toolbox are far from you." + }, + "answer_label": "C", + "answer_text": "The sink.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_2_T2", + "correct_answer": "The sink.", + "original_options": { + "A": "The green toolbox.", + "B": "They’re about the same distance.", + "C": "The sink.", + "D": "The green toolbox is closer to you than the sink is right now.", + "E": "Neither is particularly close; both the sink and the green toolbox are far from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 6, + 13, + 15 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_11_T1", + "qid_num": 27, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red toolbox relative to me?", + "answer": "In front of you.", + "question_chinese": "红色工具箱在我的哪个方位？", + "answer_chinese": "你的前面", + "evidence": [ + { + "description": "in front of you", + "timestamps": { + "start": "00:24:10", + "end": "00:24:12" + } + } + ], + "question_timestamps": "00:24:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "To your right, slightly off to the side of your body as you face forward.", + "D": "In front of you.", + "E": "Above you, positioned higher than your eye level and not on the same plane." + }, + "answer_label": "D", + "answer_text": "In front of you.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_11_T1", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "To your right, slightly off to the side of your body as you face forward.", + "D": "In front of you.", + "E": "Above you, positioned higher than your eye level and not on the same plane." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 14 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_11_T2", + "qid_num": 28, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red toolbox relative to me?", + "answer": "Behind you.", + "question_chinese": "红色工具箱在我的哪个方位？", + "answer_chinese": "你的后面", + "evidence": [ + { + "description": "behind you", + "timestamps": { + "start": "00:24:17", + "end": "00:24:30" + } + } + ], + "question_timestamps": "00:26:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your right.", + "C": "To your left, just off your shoulder as you face forward in the scene.", + "D": "Behind you.", + "E": "Above you, positioned higher than you are, as if placed on an upper shelf." + }, + "answer_label": "D", + "answer_text": "Behind you.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_11_T2", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your right.", + "C": "To your left, just off your shoulder as you face forward in the scene.", + "D": "Behind you.", + "E": "Above you, positioned higher than you are, as if placed on an upper shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 14 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_19", + "qid_num": 29, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red toolbox in relation to the sink?", + "answer": "The red toolbox is to the right of the sink.", + "question_chinese": "红色工具箱在洗手台的哪个方位？", + "answer_chinese": "红色工具箱在洗手台的右边。", + "evidence": [ + { + "description": "Where is the red toolbox in relation to the sink?", + "timestamps": { + "start": "00:24:17", + "end": "00:24:20" + } + } + ], + "question_timestamps": "00:24:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left of the sink.", + "B": "In front of the sink.", + "C": "The red toolbox is behind the sink, farther back from the edge.", + "D": "The red toolbox is above the sink, placed on a higher surface.", + "E": "The red toolbox is to the right of the sink." + }, + "answer_label": "E", + "answer_text": "The red toolbox is to the right of the sink.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_19", + "correct_answer": "The red toolbox is to the right of the sink.", + "original_options": { + "A": "Left of the sink.", + "B": "In front of the sink.", + "C": "The red toolbox is behind the sink, farther back from the edge.", + "D": "The red toolbox is above the sink, placed on a higher surface.", + "E": "The red toolbox is to the right of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 12, + 12 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_20", + "qid_num": 30, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red toolbox in relation to the sink?", + "answer": "The red toolbox is to the right of the sink.", + "question_chinese": "红色工具箱在洗手台的哪个方位？", + "answer_chinese": "红色工具箱在洗手台的右边。", + "evidence": [ + { + "description": "Where is the red toolbox in relation to the sink?", + "timestamps": { + "start": "00:24:17", + "end": "00:24:20" + } + } + ], + "question_timestamps": "00:30:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The red toolbox is to the right of the sink.", + "B": "To the left of the sink.", + "C": "Directly in front of the sink.", + "D": "The red toolbox is behind the sink, sitting farther back on the counter area.", + "E": "The red toolbox is above the sink, set on a shelf rather than on the counter." + }, + "answer_label": "A", + "answer_text": "The red toolbox is to the right of the sink.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_20", + "correct_answer": "The red toolbox is to the right of the sink.", + "original_options": { + "A": "The red toolbox is to the right of the sink.", + "B": "To the left of the sink.", + "C": "Directly in front of the sink.", + "D": "The red toolbox is behind the sink, sitting farther back on the counter area.", + "E": "The red toolbox is above the sink, set on a shelf rather than on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 16 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_9_T1", + "qid_num": 31, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pick up this red electric drill from?", + "answer": "You got it from the surface to your left.", + "question_chinese": "这把红色的电钻是我从哪里拿到的？", + "answer_chinese": "你从左边的台子上拿到的", + "evidence": [ + { + "description": "You got it from the table on the left.", + "timestamps": { + "start": "00:30:40", + "end": "00:30:41" + } + } + ], + "question_timestamps": "00:31:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You got it from the surface to your left.", + "B": "From the surface to your right.", + "C": "From the surface in front of you.", + "D": "You picked it up from the surface behind you before turning around.", + "E": "You took it from a surface above you, reaching upward to grab it." + }, + "answer_label": "A", + "answer_text": "You got it from the surface to your left.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_9_T1", + "correct_answer": "You got it from the surface to your left.", + "original_options": { + "A": "You got it from the surface to your left.", + "B": "From the surface to your right.", + "C": "From the surface in front of you.", + "D": "You picked it up from the surface behind you before turning around.", + "E": "You took it from a surface above you, reaching upward to grab it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 12, + 13 + ] + }, + { + "video_uid": "2d076447-17ad-47ab-b846-7eab0bf79090", + "video_length": "00:54:35", + "qid": "2d076447-17ad-47ab-b846-7eab0bf79090_9_T2", + "qid_num": 32, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pick up this red electric drill from?", + "answer": "You got it from the car outside.", + "question_chinese": "这把红色的电钻是我从哪里拿到的？", + "answer_chinese": "从外面的车上拿到的。", + "evidence": [ + { + "description": "You took it out of the trunk.", + "timestamps": { + "start": "00:41:02", + "end": "00:41:03" + } + } + ], + "question_timestamps": "00:41:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You got it from the car outside.", + "B": "You got it from the car inside.", + "C": "You got it from the sink.", + "D": "You picked up the drill from a work surface near you, not from the car outside.", + "E": "You retrieved it after going upstairs, grabbing it near the sink instead of from the car." + }, + "answer_label": "A", + "answer_text": "You got it from the car outside.", + "q_id": "2d076447-17ad-47ab-b846-7eab0bf79090_9_T2", + "correct_answer": "You got it from the car outside.", + "original_options": { + "A": "You got it from the car outside.", + "B": "You got it from the car inside.", + "C": "You got it from the sink.", + "D": "You picked up the drill from a work surface near you, not from the car outside.", + "E": "You retrieved it after going upstairs, grabbing it near the sink instead of from the car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 16 + ] + } +] diff --git a/QAs/2d29b45a-169b-453f-8050-98ec147b0ccf-mcq.json b/QAs/2d29b45a-169b-453f-8050-98ec147b0ccf-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1faf242c59d5b7f79f58e7c80bfbcedb3489e318 --- /dev/null +++ b/QAs/2d29b45a-169b-453f-8050-98ec147b0ccf-mcq.json @@ -0,0 +1,1311 @@ +[ + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_1_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Before I walked out of the bathroom, was the bathtub on my left side or my right side?", + "answer": "The bathtub is on the right side in front of you.", + "question_chinese": "在我走出浴室之前，浴缸在我的左侧还是右侧？", + "answer_chinese": "浴缸在你面前的右侧。", + "evidence": [ + { + "description": "The bathtub is on your right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:07:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The bathtub is on the right side in front of you.", + "B": "The bathtub is on your left side." + }, + "answer_label": "A", + "answer_text": "The bathtub is on the right side in front of you.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_1_T2", + "correct_answer": "The bathtub is on the right side in front of you.", + "original_options": { + "A": "The bathtub is on the right side in front of you.", + "B": "The bathtub is on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_1_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Before I walked out of the bathroom, was the bathtub on my left side or my right side?", + "answer": "The bathtub is on your left side.", + "question_chinese": "在我走出浴室之前，浴缸在我的左侧还是右侧？", + "answer_chinese": "浴缸在你的左侧。", + "evidence": [ + { + "description": "The bathtub is on your left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:18:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The bathtub is on the right side in front of you.", + "B": "The bathtub is on your left side." + }, + "answer_label": "B", + "answer_text": "The bathtub is on your left side.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_1_T1", + "correct_answer": "The bathtub is on your left side.", + "original_options": { + "A": "The bathtub is on the right side in front of you.", + "B": "The bathtub is on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_3", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go from the kitchen to the bathroom, how should I get there?", + "answer": "First, face the stairs and go upstairs, then turn right at the first door on your right.", + "question_chinese": "如果我要从厨房去到浴室应该怎么走？", + "answer_chinese": "你先面向楼梯并上楼，然后在右手边第一个门右转。", + "evidence": [ + { + "description": "You turned left and went down the stairs, then kept going straight.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:12" + } + }, + { + "description": "You came out of the bathroom door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "Finally, turn left and you will reach the kitchen.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:02:42", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Go downstairs, then turn left.", + "B": "Go upstairs, then turn left.", + "C": "Face the stairs and go upstairs, then turn left at the first door on your left.", + "D": "Face the stairs and go upstairs, then turn right at the second door on your right-hand side.", + "E": "First, face the stairs and go upstairs, then turn right at the first door on your right." + }, + "answer_label": "E", + "answer_text": "First, face the stairs and go upstairs, then turn right at the first door on your right.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_3", + "correct_answer": "First, face the stairs and go upstairs, then turn right at the first door on your right.", + "original_options": { + "A": "Go downstairs, then turn left.", + "B": "Go upstairs, then turn left.", + "C": "Face the stairs and go upstairs, then turn left at the first door on your left.", + "D": "Face the stairs and go upstairs, then turn right at the second door on your right-hand side.", + "E": "First, face the stairs and go upstairs, then turn right at the first door on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_4", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go from the kitchen to the bathroom, how should I get there?", + "answer": "First, face the stairs and go upstairs, then turn right at the first door on your right.", + "question_chinese": "如果我要从厨房去到浴室应该怎么走？", + "answer_chinese": "你先面向楼梯并上楼，然后在右手边第一个门右转。", + "evidence": [ + { + "description": "You turned left and went down the stairs, then kept going straight.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:12" + } + }, + { + "description": "You came out of the bathroom door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "Finally, turn left and you will reach the kitchen.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:10:49", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Go downstairs, then turn left.", + "B": "Walk straight, then turn left.", + "C": "First, face the stairs and go upstairs, then turn right at the first door on your right.", + "D": "Face the stairs and go upstairs, then turn left at the first door on your right-hand side.", + "E": "Face the stairs and go upstairs, then turn right at the last door on your right-hand side." + }, + "answer_label": "C", + "answer_text": "First, face the stairs and go upstairs, then turn right at the first door on your right.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_4", + "correct_answer": "First, face the stairs and go upstairs, then turn right at the first door on your right.", + "original_options": { + "A": "Go downstairs, then turn left.", + "B": "Walk straight, then turn left.", + "C": "First, face the stairs and go upstairs, then turn right at the first door on your right.", + "D": "Face the stairs and go upstairs, then turn left at the first door on your right-hand side.", + "E": "Face the stairs and go upstairs, then turn right at the last door on your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_2_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bathroom, how should I walk there?", + "answer": "The bathroom is behind you; you turned around and went into the bathroom.", + "question_chinese": "如果我要去到浴室应该怎么走？", + "answer_chinese": "浴室在你的后方，你向后转进入到浴室。", + "evidence": [ + { + "description": "You turned left and went down the stairs, then kept going straight.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:12" + } + }, + { + "description": "You came out of the bathroom door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "Finally, turn left and you will reach the kitchen.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bathroom is in front of you.", + "B": "The bathroom is to your left.", + "C": "The bathroom is to your right; you turn around and then walk into the bathroom.", + "D": "The bathroom is in front of you; you walk straight ahead and enter without turning around.", + "E": "The bathroom is behind you; you turned around and went into the bathroom." + }, + "answer_label": "E", + "answer_text": "The bathroom is behind you; you turned around and went into the bathroom.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_2_T2", + "correct_answer": "The bathroom is behind you; you turned around and went into the bathroom.", + "original_options": { + "A": "The bathroom is in front of you.", + "B": "The bathroom is to your left.", + "C": "The bathroom is to your right; you turn around and then walk into the bathroom.", + "D": "The bathroom is in front of you; you walk straight ahead and enter without turning around.", + "E": "The bathroom is behind you; you turned around and went into the bathroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 16 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_2_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bathroom, how should I walk there?", + "answer": "You’re facing the stairs and going upstairs. The stairs are directly in front of you, and then you turn right at the first door on your right-hand side.", + "question_chinese": "如果我要去到浴室应该怎么走？", + "answer_chinese": "你现在面向楼梯并上楼，楼梯在你正前方，然后在右手边第一个门右转。", + "evidence": [ + { + "description": "You turned left and went down the stairs, then kept going straight.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:12" + } + }, + { + "description": "You came out of the bathroom door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "Finally, turn left and you will reach the kitchen.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:02:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’re facing the stairs and going upstairs. The stairs are directly in front of you, and then you turn right at the first door on your right-hand side.", + "B": "Go downstairs, then turn left.", + "C": "Walk straight and take the first left.", + "D": "Go upstairs with the stairs on your left, then turn left at the first door on your left-hand side.", + "E": "Go upstairs with the stairs directly ahead, then turn left at the first door you reach on your left." + }, + "answer_label": "A", + "answer_text": "You’re facing the stairs and going upstairs. The stairs are directly in front of you, and then you turn right at the first door on your right-hand side.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_2_T1", + "correct_answer": "You’re facing the stairs and going upstairs. The stairs are directly in front of you, and then you turn right at the first door on your right-hand side.", + "original_options": { + "A": "You’re facing the stairs and going upstairs. The stairs are directly in front of you, and then you turn right at the first door on your right-hand side.", + "B": "Go downstairs, then turn left.", + "C": "Walk straight and take the first left.", + "D": "Go upstairs with the stairs on your left, then turn left at the first door on your left-hand side.", + "E": "Go upstairs with the stairs directly ahead, then turn left at the first door you reach on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 19, + 19 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_19", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the living room relative to me now?", + "answer": "The living room is now behind you.", + "question_chinese": "客厅现在相对我是什么位置？", + "answer_chinese": "客厅现在在你的后方。", + "evidence": [ + { + "description": "The living room is now behind you.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The living room is in front of you.", + "B": "The living room is to your left.", + "C": "The living room is to your right side right now, not behind you as you stand there.", + "D": "The living room is now behind you.", + "E": "The living room is above you at the moment, so you’d need to go up to reach it." + }, + "answer_label": "D", + "answer_text": "The living room is now behind you.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_19", + "correct_answer": "The living room is now behind you.", + "original_options": { + "A": "The living room is in front of you.", + "B": "The living room is to your left.", + "C": "The living room is to your right side right now, not behind you as you stand there.", + "D": "The living room is now behind you.", + "E": "The living room is above you at the moment, so you’d need to go up to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 19 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_10_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the living room relative to me right now?", + "answer": "The living room is in front of you right now.", + "question_chinese": "客厅现在相对我是什么位置？", + "answer_chinese": "客厅现在在你的前方。", + "evidence": [ + { + "description": "The living room is now in front of you.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The living room is in front of you right now.", + "B": "The living room is behind you.", + "C": "The living room is on your left.", + "D": "The living room is to your right right now, so you’d have to turn right to face it.", + "E": "The living room is above you right now, meaning you’d need to go upstairs to get there." + }, + "answer_label": "A", + "answer_text": "The living room is in front of you right now.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_10_T1", + "correct_answer": "The living room is in front of you right now.", + "original_options": { + "A": "The living room is in front of you right now.", + "B": "The living room is behind you.", + "C": "The living room is on your left.", + "D": "The living room is to your right right now, so you’d have to turn right to face it.", + "E": "The living room is above you right now, meaning you’d need to go upstairs to get there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 18 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_10_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the living room relative to me right now?", + "answer": "The living room is behind you right now.", + "question_chinese": "客厅现在相对我是什么位置？", + "answer_chinese": "客厅现在在你的后方。", + "evidence": [ + { + "description": "The living room is now behind you.", + "timestamps": { + "start": "00:19:05", + "end": "00:19:08" + } + } + ], + "question_timestamps": "00:19:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The living room is in front of you.", + "B": "The living room is behind you right now.", + "C": "The living room is to your left.", + "D": "The living room is to your right right now, not behind you where you would expect it.", + "E": "The living room is in front of you right now, so you’d reach it by moving forward." + }, + "answer_label": "B", + "answer_text": "The living room is behind you right now.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_10_T2", + "correct_answer": "The living room is behind you right now.", + "original_options": { + "A": "The living room is in front of you.", + "B": "The living room is behind you right now.", + "C": "The living room is to your left.", + "D": "The living room is to your right right now, not behind you where you would expect it.", + "E": "The living room is in front of you right now, so you’d reach it by moving forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 18 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_3_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I just put the tape measure?", + "answer": "You just put the tape measure down by your feet.", + "question_chinese": "我刚刚将卷尺放在什么位置上？", + "answer_chinese": "你刚才将卷尺放在了脚边。", + "evidence": [ + { + "description": "You just placed the tape measure beside your feet.", + "timestamps": { + "start": "00:03:43", + "end": "00:03:49" + } + } + ], + "question_timestamps": "00:03:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put it on the table.", + "B": "You put it to your left.", + "C": "You just placed the tape measure behind you, rather than setting it down near your feet.", + "D": "You set the tape measure on your right side, not down by your feet as you stand.", + "E": "You just put the tape measure down by your feet." + }, + "answer_label": "E", + "answer_text": "You just put the tape measure down by your feet.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_3_T1", + "correct_answer": "You just put the tape measure down by your feet.", + "original_options": { + "A": "You put it on the table.", + "B": "You put it to your left.", + "C": "You just placed the tape measure behind you, rather than setting it down near your feet.", + "D": "You set the tape measure on your right side, not down by your feet as you stand.", + "E": "You just put the tape measure down by your feet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_3_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I just put the tape measure?", + "answer": "You just put the tape measure on your right side.", + "question_chinese": "我刚刚将卷尺放在什么位置上？", + "answer_chinese": "你刚刚将卷尺放在你的右侧。", + "evidence": [ + { + "description": "You just placed the tape measure on your right side.", + "timestamps": { + "start": "00:11:47", + "end": "00:11:49" + } + } + ], + "question_timestamps": "00:12:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put it by your feet.", + "B": "You put it on your left side.", + "C": "You just put the tape measure behind you, not on your right side where you can grab it easily.", + "D": "You placed the tape measure in front of you, instead of keeping it on your right side.", + "E": "You just put the tape measure on your right side." + }, + "answer_label": "E", + "answer_text": "You just put the tape measure on your right side.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_3_T2", + "correct_answer": "You just put the tape measure on your right side.", + "original_options": { + "A": "You put it by your feet.", + "B": "You put it on your left side.", + "C": "You just put the tape measure behind you, not on your right side where you can grab it easily.", + "D": "You placed the tape measure in front of you, instead of keeping it on your right side.", + "E": "You just put the tape measure on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 17 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_8_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the broom right now?", + "answer": "You can touch the broom directly right now. The broom is behind you.", + "question_chinese": "我现在可以直接触碰到扫把吗？", + "answer_chinese": "你现在可以直接触碰到扫把。扫把在你身后", + "evidence": [ + { + "description": "You can now touch the broom directly.", + "timestamps": { + "start": "00:30:36", + "end": "00:30:37" + } + } + ], + "question_timestamps": "00:30:44", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can touch the broom directly right now. The broom is behind you.", + "B": "You can’t touch the broom directly right now. The broom is behind you." + }, + "answer_label": "A", + "answer_text": "You can touch the broom directly right now. The broom is behind you.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_8_T1", + "correct_answer": "You can touch the broom directly right now. The broom is behind you.", + "original_options": { + "A": "You can touch the broom directly right now. The broom is behind you.", + "B": "You can’t touch the broom directly right now. The broom is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_8_T2", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the broom right now?", + "answer": "You can’t touch the broom directly right now. The broom is next to you.", + "question_chinese": "我现在可以直接触碰到扫把吗？", + "answer_chinese": "你现在不可以直接触碰到扫把。扫把在你旁边。", + "evidence": [ + { + "description": "You can now touch the broom directly.", + "timestamps": { + "start": "00:30:36", + "end": "00:30:37" + } + } + ], + "question_timestamps": "00:35:44", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can touch the broom directly right now. The broom is next to you.", + "B": "You can’t touch the broom directly right now. The broom is next to you." + }, + "answer_label": "B", + "answer_text": "You can’t touch the broom directly right now. The broom is next to you.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_8_T2", + "correct_answer": "You can’t touch the broom directly right now. The broom is next to you.", + "original_options": { + "A": "You can touch the broom directly right now. The broom is next to you.", + "B": "You can’t touch the broom directly right now. The broom is next to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_5_T1", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the large blue trash bin right now?", + "answer": "Yes, you can touch it directly. The large blue trash can is directly in front of you.", + "question_chinese": "我现在可以直接触碰到蓝色大垃圾桶吗？", + "answer_chinese": "可以直接触碰到。蓝色大垃圾桶在你的正前方", + "evidence": [ + { + "description": "You can touch the big blue trash can directly.", + "timestamps": { + "start": "00:12:36", + "end": "00:12:38" + } + } + ], + "question_timestamps": "00:18:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t touch it directly right now. The large blue trash can is directly in front of you.", + "B": "Yes, you can touch it directly. The large blue trash can is directly in front of you." + }, + "answer_label": "B", + "answer_text": "Yes, you can touch it directly. The large blue trash can is directly in front of you.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_5_T1", + "correct_answer": "Yes, you can touch it directly. The large blue trash can is directly in front of you.", + "original_options": { + "A": "No, you can’t touch it directly right now. The large blue trash can is directly in front of you.", + "B": "Yes, you can touch it directly. The large blue trash can is directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_5_T2", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the large blue trash bin right now?", + "answer": "You can’t touch it directly right now. The large blue trash can is in the living room behind you.", + "question_chinese": "我现在可以直接触碰到蓝色大垃圾桶吗？", + "answer_chinese": "你现在无法直接触碰到。蓝色大垃圾桶在你身后的客厅里", + "evidence": [ + { + "description": "You can't touch the big blue trash can directly.", + "timestamps": { + "start": "00:26:40", + "end": "00:26:43" + } + } + ], + "question_timestamps": "00:26:49", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t touch it directly right now. The large blue trash can is in the living room behind you.", + "B": "Yes, you can touch it directly. The large blue trash can is in the living room behind you." + }, + "answer_label": "A", + "answer_text": "You can’t touch it directly right now. The large blue trash can is in the living room behind you.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_5_T2", + "correct_answer": "You can’t touch it directly right now. The large blue trash can is in the living room behind you.", + "original_options": { + "A": "You can’t touch it directly right now. The large blue trash can is in the living room behind you.", + "B": "Yes, you can touch it directly. The large blue trash can is in the living room behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_11", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the door handle directly now?", + "answer": "You can now touch the door handle directly.", + "question_chinese": "我现在可以直接触碰到门把手吗？", + "answer_chinese": "你现在可以直接触碰到门把手。", + "evidence": [ + { + "description": "You can now touch the door handle directly.", + "timestamps": { + "start": "00:13:18", + "end": "00:13:40" + } + } + ], + "question_timestamps": "00:13:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now touch the door handle directly.", + "B": "You cannot touch the door handle directly now." + }, + "answer_label": "A", + "answer_text": "You can now touch the door handle directly.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_11", + "correct_answer": "You can now touch the door handle directly.", + "original_options": { + "A": "You can now touch the door handle directly.", + "B": "You cannot touch the door handle directly now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_12", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the door handle directly now?", + "answer": "You can't touch the door handle directly now.", + "question_chinese": "我现在可以直接触碰到门把手吗？", + "answer_chinese": "你现在无法直接触碰到门把手。", + "evidence": [ + { + "description": "You can't touch the door handle directly now.", + "timestamps": { + "start": "00:20:49", + "end": "00:21:00" + } + } + ], + "question_timestamps": "00:21:09", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch the door handle directly now.", + "B": "You can't touch the door handle directly now." + }, + "answer_label": "B", + "answer_text": "You can't touch the door handle directly now.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_12", + "correct_answer": "You can't touch the door handle directly now.", + "original_options": { + "A": "You can touch the door handle directly now.", + "B": "You can't touch the door handle directly now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_7_T1", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From where I am now, how many stairs do I have to go past to get to the living room?", + "answer": "You didn’t go past any stairs.", + "question_chinese": "我现在要去到客厅经过几个楼梯？", + "answer_chinese": "你没有经过楼梯。", + "evidence": [ + { + "description": "You didn't go up any stairs.", + "timestamps": { + "start": "00:19:07", + "end": "00:19:11" + } + } + ], + "question_timestamps": "00:19:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You pass one staircase.", + "B": "You didn’t go past any stairs.", + "C": "You pass two staircases.", + "D": "You have to go past one staircase before you reach the living room from where you are now.", + "E": "You need to go past two separate staircases on the way to the living room from your current spot." + }, + "answer_label": "B", + "answer_text": "You didn’t go past any stairs.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_7_T1", + "correct_answer": "You didn’t go past any stairs.", + "original_options": { + "A": "You pass one staircase.", + "B": "You didn’t go past any stairs.", + "C": "You pass two staircases.", + "D": "You have to go past one staircase before you reach the living room from where you are now.", + "E": "You need to go past two separate staircases on the way to the living room from your current spot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_7_T2", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From where I am now, how many stairs do I have to go past to get to the living room?", + "answer": "You need to go past a staircase.", + "question_chinese": "我现在要去到客厅经过几个楼梯？", + "answer_chinese": "你需要经过一个楼梯。", + "evidence": [ + { + "description": "You need to go up a staircase.", + "timestamps": { + "start": "00:19:07", + "end": "00:19:11" + } + } + ], + "question_timestamps": "00:23:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You don’t pass any stairs.", + "B": "You pass two staircases.", + "C": "You need to go past a staircase.", + "D": "You have to go past the stairs twice before you can reach the living room from where you are now.", + "E": "From where you are, you avoid the staircase entirely and can reach the living room without passing it." + }, + "answer_label": "C", + "answer_text": "You need to go past a staircase.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_7_T2", + "correct_answer": "You need to go past a staircase.", + "original_options": { + "A": "You don’t pass any stairs.", + "B": "You pass two staircases.", + "C": "You need to go past a staircase.", + "D": "You have to go past the stairs twice before you can reach the living room from where you are now.", + "E": "From where you are, you avoid the staircase entirely and can reach the living room without passing it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 20, + 18 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_13", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many stairs do I need to take to get to the living room now?", + "answer": "You need to go up a staircase.", + "question_chinese": "我现在要去到客厅需要经过几个楼梯？", + "answer_chinese": "你需要经过一个楼梯。", + "evidence": [ + { + "description": "You need to go up a staircase.", + "timestamps": { + "start": "00:19:07", + "end": "00:19:11" + } + } + ], + "question_timestamps": "00:19:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You need to go down stairs.", + "B": "You don’t need any stairs.", + "C": "You need to go down a staircase to get to the living room from where you are now.", + "D": "You can reach the living room on the same level without taking any stairs at all.", + "E": "You need to go up a staircase." + }, + "answer_label": "E", + "answer_text": "You need to go up a staircase.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_13", + "correct_answer": "You need to go up a staircase.", + "original_options": { + "A": "You need to go down stairs.", + "B": "You don’t need any stairs.", + "C": "You need to go down a staircase to get to the living room from where you are now.", + "D": "You can reach the living room on the same level without taking any stairs at all.", + "E": "You need to go up a staircase." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 16 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_14", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many stairs do I need to take to get to the living room now?", + "answer": "You need to go up a staircase.", + "question_chinese": "我现在要去到客厅需要经过几个楼梯？", + "answer_chinese": "你需要经过一个楼梯。", + "evidence": [ + { + "description": "You need to go up a staircase.", + "timestamps": { + "start": "00:19:07", + "end": "00:19:11" + } + } + ], + "question_timestamps": "00:23:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "No stairs are needed.", + "B": "You need to go downstairs.", + "C": "You need to go up a staircase.", + "D": "You need to go down a staircase to reach the living room from your current position.", + "E": "You can get to the living room without using stairs, staying on the same floor the whole way." + }, + "answer_label": "C", + "answer_text": "You need to go up a staircase.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_14", + "correct_answer": "You need to go up a staircase.", + "original_options": { + "A": "No stairs are needed.", + "B": "You need to go downstairs.", + "C": "You need to go up a staircase.", + "D": "You need to go down a staircase to reach the living room from your current position.", + "E": "You can get to the living room without using stairs, staying on the same floor the whole way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 18 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_9_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "After I went up the stairs, which door can lead to the bathroom?", + "answer": "When you’re facing the stairs, the left side leads you to the bathroom.", + "question_chinese": "我走上楼梯以后，哪个门可以通往浴室？", + "answer_chinese": "你面向楼梯的左侧可以通往浴室。", + "evidence": [ + { + "description": "On your left side facing the stairs, you can reach the bathroom.", + "timestamps": { + "start": "00:18:57", + "end": "00:19:06" + } + } + ], + "question_timestamps": "00:34:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "When you’re facing the stairs, the left side leads you to the bathroom.", + "B": "The right side door.", + "C": "The door straight ahead.", + "D": "When you’re facing the stairs, the right side is the one that leads you to the bathroom.", + "E": "After going upstairs, the bathroom is reached by taking the door in front of you rather than the left side." + }, + "answer_label": "A", + "answer_text": "When you’re facing the stairs, the left side leads you to the bathroom.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_9_T1", + "correct_answer": "When you’re facing the stairs, the left side leads you to the bathroom.", + "original_options": { + "A": "When you’re facing the stairs, the left side leads you to the bathroom.", + "B": "The right side door.", + "C": "The door straight ahead.", + "D": "When you’re facing the stairs, the right side is the one that leads you to the bathroom.", + "E": "After going upstairs, the bathroom is reached by taking the door in front of you rather than the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 20 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_9_T2", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "After I went up the stairs, which door can lead to the bathroom?", + "answer": "The door on the right side leads you to the bathroom.", + "question_chinese": "我走上楼梯以后，哪个门可以通往浴室？", + "answer_chinese": "右侧的门可以让你通往浴室。", + "evidence": [ + { + "description": "The door on the right leads you to the bathroom.", + "timestamps": { + "start": "00:18:57", + "end": "00:19:06" + } + } + ], + "question_timestamps": "00:35:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The door on the right side leads you to the bathroom.", + "B": "The door on the left.", + "C": "The door in front.", + "D": "The bathroom is through the door to your left after you go up the stairs, not the right.", + "E": "After going upstairs, you reach the bathroom by taking the door behind you instead of the right-side door." + }, + "answer_label": "A", + "answer_text": "The door on the right side leads you to the bathroom.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_9_T2", + "correct_answer": "The door on the right side leads you to the bathroom.", + "original_options": { + "A": "The door on the right side leads you to the bathroom.", + "B": "The door on the left.", + "C": "The door in front.", + "D": "The bathroom is through the door to your left after you go up the stairs, not the right.", + "E": "After going upstairs, you reach the bathroom by taking the door behind you instead of the right-side door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 18 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_7", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the blue big trash can now relative to me?", + "answer": "The big blue trash can is in front of you.", + "question_chinese": "蓝色大垃圾桶现在相对我是什么位置？", + "answer_chinese": "蓝色大垃圾桶在你的前方。", + "evidence": [ + { + "description": "The big blue trash can is in front of you.", + "timestamps": { + "start": "00:04:58", + "end": "00:04:59" + } + } + ], + "question_timestamps": "00:05:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "The big blue trash can is in front of you.", + "C": "It is to your left.", + "D": "The big blue trash can is off to your right side rather than directly in front of you.", + "E": "The big blue trash can is positioned behind you now, not in front where you are facing." + }, + "answer_label": "B", + "answer_text": "The big blue trash can is in front of you.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_7", + "correct_answer": "The big blue trash can is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "The big blue trash can is in front of you.", + "C": "It is to your left.", + "D": "The big blue trash can is off to your right side rather than directly in front of you.", + "E": "The big blue trash can is positioned behind you now, not in front where you are facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 17 + ] + }, + { + "video_uid": "2d29b45a-169b-453f-8050-98ec147b0ccf", + "video_length": "00:35:48", + "qid": "2d29b45a-169b-453f-8050-98ec147b0ccf_8", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the blue big trash can now relative to me?", + "answer": "The big blue trash can is on your right.", + "question_chinese": "蓝色大垃圾桶现在相对我是什么位置？", + "answer_chinese": "蓝色大垃圾桶在你的右侧。", + "evidence": [ + { + "description": "The big blue trash can is on your right.", + "timestamps": { + "start": "00:12:06", + "end": "00:12:15" + } + } + ], + "question_timestamps": "00:12:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The big blue trash can is on your right.", + "B": "It’s on your left.", + "C": "It’s in front of you.", + "D": "The big blue trash can is directly behind you, not on your right side.", + "E": "The big blue trash can is beneath you, rather than positioned to your right." + }, + "answer_label": "A", + "answer_text": "The big blue trash can is on your right.", + "q_id": "2d29b45a-169b-453f-8050-98ec147b0ccf_8", + "correct_answer": "The big blue trash can is on your right.", + "original_options": { + "A": "The big blue trash can is on your right.", + "B": "It’s on your left.", + "C": "It’s in front of you.", + "D": "The big blue trash can is directly behind you, not on your right side.", + "E": "The big blue trash can is beneath you, rather than positioned to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 14 + ] + } +] diff --git a/QAs/2fa81cfe-e0d4-42e5-be39-e14287215bd2-mcq.json b/QAs/2fa81cfe-e0d4-42e5-be39-e14287215bd2-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..82296091f8976d5e73adc27ea92707eaa5bfb478 --- /dev/null +++ b/QAs/2fa81cfe-e0d4-42e5-be39-e14287215bd2-mcq.json @@ -0,0 +1,1061 @@ +[ + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_4_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the scissors?", + "answer": "You just hung the scissors on the wall. The wall is directly in front of you.", + "question_chinese": "我刚刚把剪刀放在什么地方？", + "answer_chinese": "你刚刚把剪刀挂在墙面上。墙在你的正前方。", + "evidence": [ + { + "description": "You just hung scissors on the wall.", + "timestamps": { + "start": "00:13:45", + "end": "00:13:49" + } + } + ], + "question_timestamps": "00:13:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the wall behind you.", + "B": "On the wall to your left.", + "C": "You hung the scissors on the wall, but that wall is directly to your right.", + "D": "You just hung the scissors on the wall. The wall is directly in front of you.", + "E": "You hung the scissors on the wall, and that wall is above you, not in front." + }, + "answer_label": "D", + "answer_text": "You just hung the scissors on the wall. The wall is directly in front of you.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_4_T1", + "correct_answer": "You just hung the scissors on the wall. The wall is directly in front of you.", + "original_options": { + "A": "On the wall behind you.", + "B": "On the wall to your left.", + "C": "You hung the scissors on the wall, but that wall is directly to your right.", + "D": "You just hung the scissors on the wall. The wall is directly in front of you.", + "E": "You hung the scissors on the wall, and that wall is above you, not in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_4_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the scissors?", + "answer": "You just put the scissors into the paper box, and the paper box is behind you.", + "question_chinese": "我刚刚把剪刀放在什么地方？", + "answer_chinese": "你刚刚把剪刀放在纸盒子里，纸盒子在你的后面。", + "evidence": [ + { + "description": "You just put scissors in the paper box.", + "timestamps": { + "start": "00:13:54", + "end": "00:13:56" + } + } + ], + "question_timestamps": "00:13:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the paper box in front.", + "B": "You just put the scissors into the paper box, and the paper box is behind you.", + "C": "In the paper box to your left.", + "D": "You put the scissors into the paper box, and the paper box is to your right.", + "E": "You put the scissors into the paper box, but the paper box is above you, not behind." + }, + "answer_label": "B", + "answer_text": "You just put the scissors into the paper box, and the paper box is behind you.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_4_T2", + "correct_answer": "You just put the scissors into the paper box, and the paper box is behind you.", + "original_options": { + "A": "In the paper box in front.", + "B": "You just put the scissors into the paper box, and the paper box is behind you.", + "C": "In the paper box to your left.", + "D": "You put the scissors into the paper box, and the paper box is to your right.", + "E": "You put the scissors into the paper box, but the paper box is above you, not behind." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_8", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the scissors?", + "answer": "You just put the scissors on the table.", + "question_chinese": "我刚刚把剪刀放在什么地方？", + "answer_chinese": "你刚刚把剪刀放在桌上。", + "evidence": [ + { + "description": "You just put the scissors on the table.", + "timestamps": { + "start": "00:20:25", + "end": "00:20:30" + } + } + ], + "question_timestamps": "00:20:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put the scissors in the drawer.", + "B": "You put the scissors on the floor.", + "C": "You just placed the scissors on the counter, not on the table you were using.", + "D": "You just put the scissors on the table.", + "E": "You just set the scissors down next to the pen, rather than putting them on the table." + }, + "answer_label": "D", + "answer_text": "You just put the scissors on the table.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_8", + "correct_answer": "You just put the scissors on the table.", + "original_options": { + "A": "You put the scissors in the drawer.", + "B": "You put the scissors on the floor.", + "C": "You just placed the scissors on the counter, not on the table you were using.", + "D": "You just put the scissors on the table.", + "E": "You just set the scissors down next to the pen, rather than putting them on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 17 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_13", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the box?", + "answer": "You just put the box on the stool.", + "question_chinese": "我刚刚把箱子放在什么位置上？", + "answer_chinese": "你刚刚把箱子放在凳子上了。", + "evidence": [ + { + "description": "You just put the box on the stool.", + "timestamps": { + "start": "00:25:00", + "end": "00:25:09" + } + } + ], + "question_timestamps": "00:25:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put the box on the shelf.", + "B": "You put the box in the closet.", + "C": "You just placed the box under the chair, not on top of the stool.", + "D": "You just set the box next to the desk, rather than placing it on the stool.", + "E": "You just put the box on the stool." + }, + "answer_label": "E", + "answer_text": "You just put the box on the stool.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_13", + "correct_answer": "You just put the box on the stool.", + "original_options": { + "A": "You put the box on the shelf.", + "B": "You put the box in the closet.", + "C": "You just placed the box under the chair, not on top of the stool.", + "D": "You just set the box next to the desk, rather than placing it on the stool.", + "E": "You just put the box on the stool." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 16 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_14", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the box?", + "answer": "You just put the box on the stool.", + "question_chinese": "我刚刚把箱子放在什么位置上？", + "answer_chinese": "你刚刚把箱子放在凳子上了。", + "evidence": [ + { + "description": "You just put the box on the stool.", + "timestamps": { + "start": "00:25:00", + "end": "00:25:09" + } + } + ], + "question_timestamps": "00:31:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put the box on the shelf.", + "B": "You put the box on the floor.", + "C": "You just placed the box beside the chair, not on the stool.", + "D": "You just put the box on the stool.", + "E": "You just put the box under the desk, instead of setting it on the stool." + }, + "answer_label": "D", + "answer_text": "You just put the box on the stool.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_14", + "correct_answer": "You just put the box on the stool.", + "original_options": { + "A": "You put the box on the shelf.", + "B": "You put the box on the floor.", + "C": "You just placed the box beside the chair, not on the stool.", + "D": "You just put the box on the stool.", + "E": "You just put the box under the desk, instead of setting it on the stool." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 15 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_11", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the stool directly now?", + "answer": "You can now make direct contact.", + "question_chinese": "我现在可以直接触碰到凳子吗。", + "answer_chinese": "你现在可以直接触碰到。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:17:16", + "end": "00:17:29" + } + } + ], + "question_timestamps": "00:17:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now make direct contact.", + "B": "You cannot now make direct contact." + }, + "answer_label": "A", + "answer_text": "You can now make direct contact.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_11", + "correct_answer": "You can now make direct contact.", + "original_options": { + "A": "You can now make direct contact.", + "B": "You cannot now make direct contact." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_6_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the stool directly right now?", + "answer": "Right now you can directly touch the stool; the stool is behind you.", + "question_chinese": "我现在可以直接触碰到凳子吗？", + "answer_chinese": "你现在可以直接触碰到凳子，凳子在你的后面。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:17:16", + "end": "00:17:29" + } + } + ], + "question_timestamps": "00:17:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can directly touch the stool; the stool is behind you.", + "B": "Right now you can’t directly touch the stool; the stool is behind you." + }, + "answer_label": "A", + "answer_text": "Right now you can directly touch the stool; the stool is behind you.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_6_T1", + "correct_answer": "Right now you can directly touch the stool; the stool is behind you.", + "original_options": { + "A": "Right now you can directly touch the stool; the stool is behind you.", + "B": "Right now you can’t directly touch the stool; the stool is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_12", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the stool directly now?", + "answer": "You can now make direct contact.", + "question_chinese": "我现在可以直接触碰到凳子吗。", + "answer_chinese": "你可以直接触碰到。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:32:45", + "end": "00:32:50" + } + } + ], + "question_timestamps": "00:32:52", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot now make direct contact.", + "B": "You can now make direct contact." + }, + "answer_label": "B", + "answer_text": "You can now make direct contact.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_12", + "correct_answer": "You can now make direct contact.", + "original_options": { + "A": "You cannot now make direct contact.", + "B": "You can now make direct contact." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_6_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the stool directly right now?", + "answer": "Right now you can’t directly touch the stool; the stool is behind you.", + "question_chinese": "我现在可以直接触碰到凳子吗？", + "answer_chinese": "你现在不可以直接触碰到凳子，凳子在你的后面。", + "evidence": [ + { + "description": "You can't touch the stool directly now.", + "timestamps": { + "start": "00:33:05", + "end": "00:33:07" + } + } + ], + "question_timestamps": "00:33:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can directly touch the stool; the stool is in front of you.", + "B": "Right now you can’t directly touch the stool; the stool is behind you." + }, + "answer_label": "B", + "answer_text": "Right now you can’t directly touch the stool; the stool is behind you.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_6_T2", + "correct_answer": "Right now you can’t directly touch the stool; the stool is behind you.", + "original_options": { + "A": "Right now you can directly touch the stool; the stool is in front of you.", + "B": "Right now you can’t directly touch the stool; the stool is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_9", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the table directly now?", + "answer": "You can't touch it directly.", + "question_chinese": "我现在可以直接触碰到桌子吗？", + "answer_chinese": "你无法直接触碰到。", + "evidence": [ + { + "description": "You can't touch it directly.", + "timestamps": { + "start": "00:13:30", + "end": "00:13:35" + } + } + ], + "question_timestamps": "00:13:37", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch it directly.", + "B": "You can't touch it directly." + }, + "answer_label": "B", + "answer_text": "You can't touch it directly.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_9", + "correct_answer": "You can't touch it directly.", + "original_options": { + "A": "You can touch it directly.", + "B": "You can't touch it directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_10", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the table directly now?", + "answer": "You can touch it directly.", + "question_chinese": "我现在可以直接触碰到桌子吗？", + "answer_chinese": "你可以直接触碰到。", + "evidence": [ + { + "description": "You can touch it directly.", + "timestamps": { + "start": "00:25:41", + "end": "00:25:46" + } + } + ], + "question_timestamps": "00:25:49", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't touch it directly.", + "B": "You can touch it directly." + }, + "answer_label": "B", + "answer_text": "You can touch it directly.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_10", + "correct_answer": "You can touch it directly.", + "original_options": { + "A": "You can't touch it directly.", + "B": "You can touch it directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_1_T3", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my knife?", + "answer": "Right now, your knife is on the table, and the table is in front of you.", + "question_chinese": "我的刀在哪里？", + "answer_chinese": "现在你的刀在桌子上，桌子在你的正前方。", + "evidence": [ + { + "description": "Now your knife is on the table, and the table is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the table behind you.", + "B": "Under the table in front of you.", + "C": "Right now, your knife is on the table, and the table is in front of you.", + "D": "Right now, your knife is next to the table, and the table is in front of you.", + "E": "Right now, your knife is on the table, but the table is behind you instead of in front." + }, + "answer_label": "C", + "answer_text": "Right now, your knife is on the table, and the table is in front of you.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_1_T3", + "correct_answer": "Right now, your knife is on the table, and the table is in front of you.", + "original_options": { + "A": "On the table behind you.", + "B": "Under the table in front of you.", + "C": "Right now, your knife is on the table, and the table is in front of you.", + "D": "Right now, your knife is next to the table, and the table is in front of you.", + "E": "Right now, your knife is on the table, but the table is behind you instead of in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 17, + 18 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_1_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my knife?", + "answer": "Right now, your knife is on the knife sharpener, and the knife sharpener is in front of you.", + "question_chinese": "我的刀在哪里？", + "answer_chinese": "现在你的刀在磨刀器上，磨刀器在你的正前方。", + "evidence": [ + { + "description": "Your knife is on the sharpener now.", + "timestamps": { + "start": "00:02:17", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:02:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the sharpener behind you.", + "B": "Right now, your knife is on the knife sharpener, and the knife sharpener is in front of you.", + "C": "Next to the sharpener in front.", + "D": "Right now, your knife is under the knife sharpener, and the knife sharpener is in front of you.", + "E": "Right now, your knife is on the table, and the table is in front of you, not on the sharpener." + }, + "answer_label": "B", + "answer_text": "Right now, your knife is on the knife sharpener, and the knife sharpener is in front of you.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_1_T2", + "correct_answer": "Right now, your knife is on the knife sharpener, and the knife sharpener is in front of you.", + "original_options": { + "A": "On the sharpener behind you.", + "B": "Right now, your knife is on the knife sharpener, and the knife sharpener is in front of you.", + "C": "Next to the sharpener in front.", + "D": "Right now, your knife is under the knife sharpener, and the knife sharpener is in front of you.", + "E": "Right now, your knife is on the table, and the table is in front of you, not on the sharpener." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 20 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_5", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the table now relative to me?", + "answer": "The table is now on your left.", + "question_chinese": "桌子现在相对我是什么位置？", + "answer_chinese": "桌子现在在你的左侧。", + "evidence": [ + { + "description": "The table is now on your left.", + "timestamps": { + "start": "00:09:46", + "end": "00:09:52" + } + } + ], + "question_timestamps": "00:09:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The table is in front of you.", + "B": "The table is on your right.", + "C": "The table is behind you now, so you’d need to turn around to face it.", + "D": "The table is now on your left.", + "E": "The table is straight ahead of you, positioned slightly off-center rather than to your left side." + }, + "answer_label": "D", + "answer_text": "The table is now on your left.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_5", + "correct_answer": "The table is now on your left.", + "original_options": { + "A": "The table is in front of you.", + "B": "The table is on your right.", + "C": "The table is behind you now, so you’d need to turn around to face it.", + "D": "The table is now on your left.", + "E": "The table is straight ahead of you, positioned slightly off-center rather than to your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 16 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_6", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the table now relative to me?", + "answer": "The table is now behind you.", + "question_chinese": "桌子现在相对我是什么位置？", + "answer_chinese": "桌子现在在你的后方。", + "evidence": [ + { + "description": "The table is now behind you.", + "timestamps": { + "start": "00:17:27", + "end": "00:17:30" + } + } + ], + "question_timestamps": "00:17:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The table is now behind you.", + "B": "The table is in front of you.", + "C": "The table is to your left.", + "D": "The table is now on your right side rather than behind you.", + "E": "The table is now beside you rather than positioned behind you at the moment." + }, + "answer_label": "A", + "answer_text": "The table is now behind you.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_6", + "correct_answer": "The table is now behind you.", + "original_options": { + "A": "The table is now behind you.", + "B": "The table is in front of you.", + "C": "The table is to your left.", + "D": "The table is now on your right side rather than behind you.", + "E": "The table is now beside you rather than positioned behind you at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 14 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_3", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the white trash can now relative to me?", + "answer": "The white trash can is on your right.", + "question_chinese": "白色垃圾桶现在相对我是什么位置？", + "answer_chinese": "白色垃圾桶在你的右侧。", + "evidence": [ + { + "description": "The white trash can is on your right.", + "timestamps": { + "start": "00:05:55", + "end": "00:05:59" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "The white trash can is on your right.", + "C": "On your left.", + "D": "The white trash can is directly behind you, just out of your immediate view.", + "E": "The white trash can is in front of you, a short distance away from where you are standing." + }, + "answer_label": "B", + "answer_text": "The white trash can is on your right.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_3", + "correct_answer": "The white trash can is on your right.", + "original_options": { + "A": "In front of you.", + "B": "The white trash can is on your right.", + "C": "On your left.", + "D": "The white trash can is directly behind you, just out of your immediate view.", + "E": "The white trash can is in front of you, a short distance away from where you are standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 18 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_2_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the white trash can relative to me right now?", + "answer": "The white trash can is on the right side of you.", + "question_chinese": "白色垃圾桶现在相对我是什么位置？", + "answer_chinese": "白色垃圾桶相对在你的右侧。", + "evidence": [ + { + "description": "The white trash can is on your right.", + "timestamps": { + "start": "00:05:55", + "end": "00:05:59" + } + } + ], + "question_timestamps": "00:06:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "On your left side.", + "C": "The white trash can is behind you, positioned on the opposite side from where you’re facing.", + "D": "The white trash can is in front of you, lined up with your current direction of travel.", + "E": "The white trash can is on the right side of you." + }, + "answer_label": "E", + "answer_text": "The white trash can is on the right side of you.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_2_T1", + "correct_answer": "The white trash can is on the right side of you.", + "original_options": { + "A": "In front of you.", + "B": "On your left side.", + "C": "The white trash can is behind you, positioned on the opposite side from where you’re facing.", + "D": "The white trash can is in front of you, lined up with your current direction of travel.", + "E": "The white trash can is on the right side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_4", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the white trash can now relative to me?", + "answer": "The white trash can is on your right.", + "question_chinese": "白色垃圾桶现在相对我是什么位置？", + "answer_chinese": "白色垃圾桶在你的右侧。", + "evidence": [ + { + "description": "The white trash can is on your right.", + "timestamps": { + "start": "00:13:23", + "end": "00:13:27" + } + } + ], + "question_timestamps": "00:13:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The white trash can is on your right.", + "D": "The white trash can is behind you, not in your current line of sight.", + "E": "The white trash can is in front of you, positioned ahead of where you are looking." + }, + "answer_label": "C", + "answer_text": "The white trash can is on your right.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_4", + "correct_answer": "The white trash can is on your right.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The white trash can is on your right.", + "D": "The white trash can is behind you, not in your current line of sight.", + "E": "The white trash can is in front of you, positioned ahead of where you are looking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 16 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_2_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the white trash can relative to me right now?", + "answer": "The white trash can is on the right side of you.", + "question_chinese": "白色垃圾桶现在相对我是什么位置？", + "answer_chinese": "白色垃圾桶在你的右侧。", + "evidence": [ + { + "description": "The white trash can is on your right.", + "timestamps": { + "start": "00:13:23", + "end": "00:13:27" + } + } + ], + "question_timestamps": "00:13:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "On your left side.", + "C": "The white trash can is behind you, on the side opposite your current facing direction.", + "D": "The white trash can is in front of you, slightly ahead as you look forward.", + "E": "The white trash can is on the right side of you." + }, + "answer_label": "E", + "answer_text": "The white trash can is on the right side of you.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_2_T2", + "correct_answer": "The white trash can is on the right side of you.", + "original_options": { + "A": "In front of you.", + "B": "On your left side.", + "C": "The white trash can is behind you, on the side opposite your current facing direction.", + "D": "The white trash can is in front of you, slightly ahead as you look forward.", + "E": "The white trash can is on the right side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_7_T2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the knife sharpener relative to me?", + "answer": "The knife sharpener is to your right.", + "question_chinese": "磨刀器在相对于我的哪里？", + "answer_chinese": "磨刀器在相对于你的右边。", + "evidence": [ + { + "description": "The sharpener is on the right relative to.", + "timestamps": { + "start": "00:02:23", + "end": "00:02:26" + } + } + ], + "question_timestamps": "00:02:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "In front of you.", + "C": "The knife sharpener is behind you, positioned away from your current viewpoint.", + "D": "The knife sharpener is to your right.", + "E": "The knife sharpener is above you, located higher than your eye level in the scene." + }, + "answer_label": "D", + "answer_text": "The knife sharpener is to your right.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_7_T2", + "correct_answer": "The knife sharpener is to your right.", + "original_options": { + "A": "To your left.", + "B": "In front of you.", + "C": "The knife sharpener is behind you, positioned away from your current viewpoint.", + "D": "The knife sharpener is to your right.", + "E": "The knife sharpener is above you, located higher than your eye level in the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 15 + ] + }, + { + "video_uid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2", + "video_length": "00:36:49", + "qid": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_7_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the knife sharpener relative to me?", + "answer": "The knife sharpener is directly in front of you.", + "question_chinese": "磨刀器在相对于我的哪里？", + "answer_chinese": "磨刀器在相对于你的正前方。", + "evidence": [ + { + "description": "The sharpener is right in front of you.", + "timestamps": { + "start": "00:04:40", + "end": "00:04:42" + } + } + ], + "question_timestamps": "00:04:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly behind you.", + "B": "Directly to your left.", + "C": "The knife sharpener is directly to your right, aligned with your side rather than ahead.", + "D": "The knife sharpener is directly above you, positioned higher up instead of at your front.", + "E": "The knife sharpener is directly in front of you." + }, + "answer_label": "E", + "answer_text": "The knife sharpener is directly in front of you.", + "q_id": "2fa81cfe-e0d4-42e5-be39-e14287215bd2_7_T1", + "correct_answer": "The knife sharpener is directly in front of you.", + "original_options": { + "A": "Directly behind you.", + "B": "Directly to your left.", + "C": "The knife sharpener is directly to your right, aligned with your side rather than ahead.", + "D": "The knife sharpener is directly above you, positioned higher up instead of at your front.", + "E": "The knife sharpener is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 15 + ] + } +] diff --git a/QAs/30b9ff64-51c8-4481-ba6a-733ea2060aef-mcq.json b/QAs/30b9ff64-51c8-4481-ba6a-733ea2060aef-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a4c05c35f0c3bb372677c2bcb126e00b4600b488 --- /dev/null +++ b/QAs/30b9ff64-51c8-4481-ba6a-733ea2060aef-mcq.json @@ -0,0 +1,1585 @@ +[ + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_11", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the spoon that I previously put in the drawer back where I took it out, where should I put it?", + "answer": "You should put it in the top grid drawer uder the microwave.", + "question_chinese": "如果我想把我之前放在抽屉里的勺子放回我拿出来的地方，我应该放在哪里？", + "answer_chinese": "你应该放在微波炉下方的最上面一层网格抽屉里。", + "evidence": [ + { + "description": "I took it from the top grid drawer uder the microwave.", + "timestamps": { + "start": "00:14:30", + "end": "00:14:50" + } + } + ], + "question_timestamps": "00:18:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put it in the top drawer by sink.", + "B": "Put it in the middle grid drawer.", + "C": "You should put it in the bottom grid drawer beside the refrigerator, where you store similar utensils.", + "D": "You should put it in the top grid drawer uder the microwave.", + "E": "You should place it in the top grid drawer near the sink, not the one located under the microwave." + }, + "answer_label": "D", + "answer_text": "You should put it in the top grid drawer uder the microwave.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_11", + "correct_answer": "You should put it in the top grid drawer uder the microwave.", + "original_options": { + "A": "Put it in the top drawer by sink.", + "B": "Put it in the middle grid drawer.", + "C": "You should put it in the bottom grid drawer beside the refrigerator, where you store similar utensils.", + "D": "You should put it in the top grid drawer uder the microwave.", + "E": "You should place it in the top grid drawer near the sink, not the one located under the microwave." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 19 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_6_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the spoon I previously put in the drawer back where I originally took it from, where should I put it?", + "answer": "You should put it in the first drawer of the stove in front of you.", + "question_chinese": "如果我想把我之前放在抽屉里的勺子放回我拿出来的地方，我应该放在哪里？", + "answer_chinese": "你应该放在你面前的灶台的第一层抽屉里。", + "evidence": [ + { + "description": "I took it from the top grid drawer uder the microwave.", + "timestamps": { + "start": "00:14:30", + "end": "00:14:50" + } + } + ], + "question_timestamps": "00:21:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it in the stove drawer behind you.", + "B": "You should put it in the first drawer of the stove in front of you.", + "C": "Put it in the stove drawer left.", + "D": "You should put it in the first drawer of the stove to your right, rather than the one in front of you.", + "E": "You should put it in the first drawer of the stove behind you, not the drawer located in front of you." + }, + "answer_label": "B", + "answer_text": "You should put it in the first drawer of the stove in front of you.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_6_T1", + "correct_answer": "You should put it in the first drawer of the stove in front of you.", + "original_options": { + "A": "Put it in the stove drawer behind you.", + "B": "You should put it in the first drawer of the stove in front of you.", + "C": "Put it in the stove drawer left.", + "D": "You should put it in the first drawer of the stove to your right, rather than the one in front of you.", + "E": "You should put it in the first drawer of the stove behind you, not the drawer located in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 22, + 21 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_6_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the spoon I previously put in the drawer back where I originally took it from, where should I put it?", + "answer": "You should put it in the top mesh drawer below the microwave to your left.", + "question_chinese": "如果我想把我之前放在抽屉里的勺子放回我拿出来的地方，我应该放在哪里？", + "answer_chinese": "你应该放在你左边的微波炉下方的最上面一层网格抽屉里。", + "evidence": [ + { + "description": "I took it from the top grid drawer under the microwave.", + "timestamps": { + "start": "00:14:30", + "end": "00:14:50" + } + } + ], + "question_timestamps": "00:22:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it in the bottom mesh drawer.", + "B": "Put it in the top mesh drawer above.", + "C": "You should put it in the top mesh drawer below the microwave in front of you, not to your left.", + "D": "You should put it in the top mesh drawer below the microwave to your left.", + "E": "You should put it in the top mesh drawer below the microwave to your right, where you first grabbed it." + }, + "answer_label": "D", + "answer_text": "You should put it in the top mesh drawer below the microwave to your left.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_6_T2", + "correct_answer": "You should put it in the top mesh drawer below the microwave to your left.", + "original_options": { + "A": "Put it in the bottom mesh drawer.", + "B": "Put it in the top mesh drawer above.", + "C": "You should put it in the top mesh drawer below the microwave in front of you, not to your left.", + "D": "You should put it in the top mesh drawer below the microwave to your left.", + "E": "You should put it in the top mesh drawer below the microwave to your right, where you first grabbed it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 20, + 20 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_3", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was eating, was the computer on my left or right?", + "answer": "When you were eating just now, the computer was on your left.", + "question_chinese": "我刚才在吃饭时，电脑在我的左边还是右边？", + "answer_chinese": "你刚才在吃饭时，电脑在你的左边。", + "evidence": [ + { + "description": "When I were eating, the computer was on my left.", + "timestamps": { + "start": "00:01:10", + "end": "00:02:04" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you were eating just now, the computer was on your right.", + "B": "When you were eating just now, the computer was on your left." + }, + "answer_label": "B", + "answer_text": "When you were eating just now, the computer was on your left.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_3", + "correct_answer": "When you were eating just now, the computer was on your left.", + "original_options": { + "A": "When you were eating just now, the computer was on your right.", + "B": "When you were eating just now, the computer was on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_2_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was eating just now, was the computer to my left or to my right?", + "answer": "When you were eating just now, the computer was to your left.", + "question_chinese": "我刚才在吃饭时，电脑在我的左边还是右边？", + "answer_chinese": "你刚才在吃饭时，电脑在你的左边。", + "evidence": [ + { + "description": "When I were eating, the computer was on my left.", + "timestamps": { + "start": "00:01:10", + "end": "00:02:04" + } + } + ], + "question_timestamps": "00:20:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were eating just now, the computer was to your right.", + "B": "When you were eating just now, the computer was to your left." + }, + "answer_label": "B", + "answer_text": "When you were eating just now, the computer was to your left.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_2_T2", + "correct_answer": "When you were eating just now, the computer was to your left.", + "original_options": { + "A": "When you were eating just now, the computer was to your right.", + "B": "When you were eating just now, the computer was to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_2_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was eating just now, was the computer to my left or to my right?", + "answer": "When you were eating just now, the computer was to your front-right.", + "question_chinese": "我刚才在吃饭时，电脑在我的左边还是右边？", + "answer_chinese": "你刚才在吃饭时，电脑在你的右前方。", + "evidence": [ + { + "description": "When I were eating, the computer was on my left.", + "timestamps": { + "start": "00:01:10", + "end": "00:02:04" + } + } + ], + "question_timestamps": "00:22:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were eating just now, the computer was to your left.", + "B": "When you were eating just now, the computer was to your front-right." + }, + "answer_label": "B", + "answer_text": "When you were eating just now, the computer was to your front-right.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_2_T1", + "correct_answer": "When you were eating just now, the computer was to your front-right.", + "original_options": { + "A": "When you were eating just now, the computer was to your left.", + "B": "When you were eating just now, the computer was to your front-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_1_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "The cup I used for drinking water during the meal—what path did it take from the cabinet to the dining table?", + "answer": "While you were eating, you took the cup down from the cabinet and set it on the sink-side table.", + "question_chinese": "我吃饭时用来喝水的杯子从橱柜里到餐桌上经历了什么路径？", + "answer_chinese": "你吃饭时杯子被你从橱柜拿下，放到洗手池桌子上。", + "evidence": [ + { + "description": "The cup was taken down from the cabinet by me, placed on the sink table.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You carried it straight to the dining table.", + "B": "You set it on the dining table first.", + "C": "While you were eating, you took the cup down from the cabinet and set it on the sink-side table.", + "D": "While you were eating, you took the cup down and set it directly on the dining table.", + "E": "While you were eating, you took the cup from the cabinet, placed it on the countertop, then brought it to the table." + }, + "answer_label": "C", + "answer_text": "While you were eating, you took the cup down from the cabinet and set it on the sink-side table.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_1_T1", + "correct_answer": "While you were eating, you took the cup down from the cabinet and set it on the sink-side table.", + "original_options": { + "A": "You carried it straight to the dining table.", + "B": "You set it on the dining table first.", + "C": "While you were eating, you took the cup down from the cabinet and set it on the sink-side table.", + "D": "While you were eating, you took the cup down and set it directly on the dining table.", + "E": "While you were eating, you took the cup from the cabinet, placed it on the countertop, then brought it to the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 22 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the cup I used to drink water from while eating go through from the cabinet to the dining table?", + "answer": "The cup was taken down from the cabinet by you, placed on the sink table, and then brought to the dining table by you.", + "question_chinese": "我吃饭时用来喝水的杯子从橱柜里到餐桌上经历了什么路径？", + "answer_chinese": "你吃饭时杯子被你从橱柜拿下，放到洗手池桌子上，又被你拿到餐桌上。", + "evidence": [ + { + "description": "The cup was taken down from the cabinet by me, placed on the sink table, and then brought to the dining table by me.", + "timestamps": { + "start": "00:00:37", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:15:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took it straight to the dining table.", + "B": "You set it by the kettle first.", + "C": "The cup was taken down from the cabinet by you, placed on the sink table, and then brought to the dining table by you.", + "D": "You took the cup down from the cabinet, set it on the countertop, and then carried it to the dining table.", + "E": "You took the cup from the cabinet, placed it near the refrigerator briefly, then brought it to the dining table." + }, + "answer_label": "C", + "answer_text": "The cup was taken down from the cabinet by you, placed on the sink table, and then brought to the dining table by you.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_1", + "correct_answer": "The cup was taken down from the cabinet by you, placed on the sink table, and then brought to the dining table by you.", + "original_options": { + "A": "You took it straight to the dining table.", + "B": "You set it by the kettle first.", + "C": "The cup was taken down from the cabinet by you, placed on the sink table, and then brought to the dining table by you.", + "D": "You took the cup down from the cabinet, set it on the countertop, and then carried it to the dining table.", + "E": "You took the cup from the cabinet, placed it near the refrigerator briefly, then brought it to the dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 21, + 20 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_1_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "The cup I used for drinking water during the meal—what path did it take from the cabinet to the dining table?", + "answer": "While you were eating, you took the cup down from the cabinet, set it on the sink-side table, and then moved it onto the dining table.", + "question_chinese": "我吃饭时用来喝水的杯子从橱柜里到餐桌上经历了什么路径？", + "answer_chinese": "你吃饭时杯子被你从橱柜拿下，放到洗手池桌子上，又被你拿到餐桌上。", + "evidence": [ + { + "description": "The cup was taken down from the cabinet by me,placed on the sink table, and then brought to the dining table by me.", + "timestamps": { + "start": "00:00:37", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:22:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "While you were eating, you took the cup down from the cabinet, set it on the sink-side table, and then moved it onto the dining table.", + "B": "You carried it straight to the dining table.", + "C": "You left it on the sink-side table.", + "D": "While you were eating, you took the cup down and set it directly onto the dining table without stopping.", + "E": "While you were eating, you took the cup from the cabinet, put it on the countertop, then moved it to the table." + }, + "answer_label": "A", + "answer_text": "While you were eating, you took the cup down from the cabinet, set it on the sink-side table, and then moved it onto the dining table.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_1_T2", + "correct_answer": "While you were eating, you took the cup down from the cabinet, set it on the sink-side table, and then moved it onto the dining table.", + "original_options": { + "A": "While you were eating, you took the cup down from the cabinet, set it on the sink-side table, and then moved it onto the dining table.", + "B": "You carried it straight to the dining table.", + "C": "You left it on the sink-side table.", + "D": "While you were eating, you took the cup down and set it directly onto the dining table without stopping.", + "E": "While you were eating, you took the cup from the cabinet, put it on the countertop, then moved it to the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 22 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_7_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in the room did I see the fruit?", + "answer": "The fruit is on the cabinet below and to the right of the range hood.", + "question_chinese": "我看见水果在房间上哪个位置？", + "answer_chinese": "水果在吸油烟机右下方的柜子上。", + "evidence": [ + { + "description": "The fruit is on the cabineton which is on the lower right side of the the range hood.", + "timestamps": { + "start": "00:21:30", + "end": "00:21:34" + } + }, + { + "description": "The fruit is on the cabineton which is on the lower right side of the the range hood.", + "timestamps": { + "start": "00:15:34", + "end": "00:15:37" + } + }, + { + "description": "The fruit is on the cabineton which is on the lower right side of the the range hood.", + "timestamps": { + "start": "00:15:05", + "end": "00:15:15" + } + } + ], + "question_timestamps": "00:27:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fruit is on the cabinet below and to the right of the range hood.", + "B": "The fruit is on the cabinet above.", + "C": "The fruit is above the range hood.", + "D": "The fruit is on the cabinet below and to the left of the range hood, not to the right.", + "E": "The fruit is on the cabinet above and to the right of the range hood, higher than where you noticed it." + }, + "answer_label": "A", + "answer_text": "The fruit is on the cabinet below and to the right of the range hood.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_7_T2", + "correct_answer": "The fruit is on the cabinet below and to the right of the range hood.", + "original_options": { + "A": "The fruit is on the cabinet below and to the right of the range hood.", + "B": "The fruit is on the cabinet above.", + "C": "The fruit is above the range hood.", + "D": "The fruit is on the cabinet below and to the left of the range hood, not to the right.", + "E": "The fruit is on the cabinet above and to the right of the range hood, higher than where you noticed it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 21 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_7_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in the room did I see the fruit?", + "answer": "The fruit is to the stove’s front-right.", + "question_chinese": "我看见水果在房间上哪个位置？", + "answer_chinese": "水果在灶台的右前方。", + "evidence": [ + { + "description": "The fruit is on the cabineton which is on the lower right side of the the range hood.", + "timestamps": { + "start": "00:15:05", + "end": "00:15:15" + } + }, + { + "description": "The fruit is on the cabineton which is on the lower right side of the the range hood.", + "timestamps": { + "start": "00:15:34", + "end": "00:15:37" + } + }, + { + "description": "The fruit is on the cabineton which is on the lower right side of the the range hood.", + "timestamps": { + "start": "00:21:30", + "end": "00:21:34" + } + } + ], + "question_timestamps": "00:28:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fruit is to the stove’s front-right.", + "B": "The fruit is to the stove’s front-left.", + "C": "The fruit is to the stove’s back-right.", + "D": "The fruit is to the stove’s back-left, tucked near the counter edge behind the burners.", + "E": "The fruit is on the cabinet below, slightly left of the range hood rather than in front of the stove." + }, + "answer_label": "A", + "answer_text": "The fruit is to the stove’s front-right.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_7_T1", + "correct_answer": "The fruit is to the stove’s front-right.", + "original_options": { + "A": "The fruit is to the stove’s front-right.", + "B": "The fruit is to the stove’s front-left.", + "C": "The fruit is to the stove’s back-right.", + "D": "The fruit is to the stove’s back-left, tucked near the counter edge behind the burners.", + "E": "The fruit is on the cabinet below, slightly left of the range hood rather than in front of the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 20 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_9_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in the bedroom did I see the wardrobe?", + "answer": "The wardrobe is next to the bedroom door.", + "question_chinese": "我看见衣柜在卧室哪个位置？", + "answer_chinese": "衣柜在卧室门旁边。", + "evidence": [ + { + "description": "The wardrobe is next to the bedroom door.", + "timestamps": { + "start": "00:28:34", + "end": "00:28:37" + } + } + ], + "question_timestamps": "00:29:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The wardrobe is next to the bedroom door.", + "B": "The wardrobe is by the bed.", + "C": "The wardrobe is opposite the bedroom door.", + "D": "The wardrobe is to the bed’s front-right, closer to the foot of the bed than the doorway.", + "E": "The wardrobe is to the bed’s back-left, set along the wall away from the bedroom door." + }, + "answer_label": "A", + "answer_text": "The wardrobe is next to the bedroom door.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_9_T2", + "correct_answer": "The wardrobe is next to the bedroom door.", + "original_options": { + "A": "The wardrobe is next to the bedroom door.", + "B": "The wardrobe is by the bed.", + "C": "The wardrobe is opposite the bedroom door.", + "D": "The wardrobe is to the bed’s front-right, closer to the foot of the bed than the doorway.", + "E": "The wardrobe is to the bed’s back-left, set along the wall away from the bedroom door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 17 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_9_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in the bedroom did I see the wardrobe?", + "answer": "The wardrobe is to the bed’s front-right.", + "question_chinese": "我看见衣柜在卧室哪个位置？", + "answer_chinese": "衣柜在床的右前方。", + "evidence": [ + { + "description": "The wardrobe is next to the bedroom door.", + "timestamps": { + "start": "00:28:34", + "end": "00:28:37" + } + } + ], + "question_timestamps": "00:29:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The wardrobe is to the bed’s front-left.", + "B": "The wardrobe is next to the bedroom door.", + "C": "The wardrobe is to the bed’s back-right, positioned nearer the wall behind the headboard.", + "D": "The wardrobe is to the bed’s front-right.", + "E": "The wardrobe is to the bed’s back-left, not on the front-right side you were facing." + }, + "answer_label": "D", + "answer_text": "The wardrobe is to the bed’s front-right.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_9_T1", + "correct_answer": "The wardrobe is to the bed’s front-right.", + "original_options": { + "A": "The wardrobe is to the bed’s front-left.", + "B": "The wardrobe is next to the bedroom door.", + "C": "The wardrobe is to the bed’s back-right, positioned nearer the wall behind the headboard.", + "D": "The wardrobe is to the bed’s front-right.", + "E": "The wardrobe is to the bed’s back-left, not on the front-right side you were facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 16 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_5_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many chairs did I see next to the dining table?", + "answer": "There are two chairs by the dining table.", + "question_chinese": "我看见餐桌边有几把椅子？", + "answer_chinese": "餐桌边有两把椅子。", + "evidence": [ + { + "description": "There are two chairs next to the dining table.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:02:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There is one chair by the dining table.", + "B": "There are three chairs by the dining table.", + "C": "There are two chairs by the dining table.", + "D": "There are four chairs by the dining table, with two on each long side of the table.", + "E": "There are zero chairs by the dining table; the table is clear with no chairs beside it." + }, + "answer_label": "C", + "answer_text": "There are two chairs by the dining table.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_5_T1", + "correct_answer": "There are two chairs by the dining table.", + "original_options": { + "A": "There is one chair by the dining table.", + "B": "There are three chairs by the dining table.", + "C": "There are two chairs by the dining table.", + "D": "There are four chairs by the dining table, with two on each long side of the table.", + "E": "There are zero chairs by the dining table; the table is clear with no chairs beside it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 17 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_5_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many chairs did I see next to the dining table?", + "answer": "There are four chairs by the dining table.", + "question_chinese": "我看见餐桌边有几把椅子？", + "answer_chinese": "餐桌边有四把椅子。", + "evidence": [ + { + "description": "There are four chairs next to the dining table.", + "timestamps": { + "start": "00:13:44", + "end": "00:13:50" + } + }, + { + "description": "There are four chairs next to the dining table.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:25:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are three chairs by the dining table.", + "B": "There are five chairs by the dining table.", + "C": "There are two chairs by the dining table, both positioned on the same side of it.", + "D": "There are four chairs by the dining table.", + "E": "There are six chairs by the dining table, arranged evenly around the table’s sides." + }, + "answer_label": "D", + "answer_text": "There are four chairs by the dining table.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_5_T2", + "correct_answer": "There are four chairs by the dining table.", + "original_options": { + "A": "There are three chairs by the dining table.", + "B": "There are five chairs by the dining table.", + "C": "There are two chairs by the dining table, both positioned on the same side of it.", + "D": "There are four chairs by the dining table.", + "E": "There are six chairs by the dining table, arranged evenly around the table’s sides." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 15 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_14", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the cup take from the grid drawer to the cabinet above the kettle?", + "answer": "The cup was taken out from the grid drawer, counting down from top to bottom, and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "question_chinese": "杯子被从网格抽屉到烧水壶上方柜子里，经历了什么路径？", + "answer_chinese": "杯子被从网格抽屉从上往下数第二层拿出来，经过烧水壶上方，放到烧水壶上方柜子的最下面一层。", + "evidence": [ + { + "description": "The cup was taken out from the grid drawer, counting down from top to bottom, and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "timestamps": { + "start": "00:16:06", + "end": "00:16:18" + } + }, + { + "description": "The cup was taken out from the grid drawer, counting down from top to bottom, and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "timestamps": { + "start": "00:16:38", + "end": "00:16:46" + } + }, + { + "description": "The cup was taken out from the grid drawer, counting down from top to bottom, and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "timestamps": { + "start": "00:16:50", + "end": "00:16:55" + } + } + ], + "question_timestamps": "00:23:25", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You moved it beside the kettle.", + "B": "The cup was taken out from the grid drawer, counting down from top to bottom, and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "C": "You placed it on the middle shelf.", + "D": "The cup was taken out from the grid drawer, passed behind the kettle, and set on the middle shelf above it.", + "E": "The cup was removed from the cabinet above the kettle, brought down past the kettle, and returned into the grid drawer." + }, + "answer_label": "B", + "answer_text": "The cup was taken out from the grid drawer, counting down from top to bottom, and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_14", + "correct_answer": "The cup was taken out from the grid drawer, counting down from top to bottom, and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "original_options": { + "A": "You moved it beside the kettle.", + "B": "The cup was taken out from the grid drawer, counting down from top to bottom, and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "C": "You placed it on the middle shelf.", + "D": "The cup was taken out from the grid drawer, passed behind the kettle, and set on the middle shelf above it.", + "E": "The cup was removed from the cabinet above the kettle, brought down past the kettle, and returned into the grid drawer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 21 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_15", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the cup take from the grid drawer to the cabinet above the kettle?", + "answer": "The cup was taken out from the grid drawer, counting dow from top to bottom. and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "question_chinese": "杯子被从网格抽屉到烧水壶上方柜子里，经历了什么路径？", + "answer_chinese": "杯子从网格抽屉从上往下数第二层拿出来，经过烧水壶上方，放到烧水壶上方柜子的最下面一层。", + "evidence": [ + { + "description": "The cup was taken out from the grid drawer, counting dow from top to bottom.\nand then passed over the kettle, finally was placed on the bottom layer of the\ncabinet above the kettle.", + "timestamps": { + "start": "00:16:06", + "end": "00:16:18" + } + }, + { + "description": "The cup was taken out from the grid drawer, counting dow from top to bottom.\nand then passed over the kettle, finally was placed on the bottom layer of the\ncabinet above the kettle.", + "timestamps": { + "start": "00:16:38", + "end": "00:16:46" + } + }, + { + "description": "The cup was taken out from the grid drawer, counting dow from top to bottom.\nand then passed over the kettle, finally was placed on the bottom layer of the\ncabinet above the kettle.", + "timestamps": { + "start": "00:16:50", + "end": "00:16:55" + } + } + ], + "question_timestamps": "00:27:40", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The cup was taken out from the grid drawer, counting dow from top to bottom. and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "B": "You counted up from bottom to top.", + "C": "You placed it on the top shelf.", + "D": "The cup was taken from the grid drawer, moved to the right of the kettle, and placed on the countertop below.", + "E": "The cup was taken out from the grid drawer, passed behind the kettle, and set on the middle shelf of the cabinet." + }, + "answer_label": "A", + "answer_text": "The cup was taken out from the grid drawer, counting dow from top to bottom. and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_15", + "correct_answer": "The cup was taken out from the grid drawer, counting dow from top to bottom. and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "original_options": { + "A": "The cup was taken out from the grid drawer, counting dow from top to bottom. and then passed over the kettle, finally was placed on the bottom layer of the cabinet above the kettle.", + "B": "You counted up from bottom to top.", + "C": "You placed it on the top shelf.", + "D": "The cup was taken from the grid drawer, moved to the right of the kettle, and placed on the countertop below.", + "E": "The cup was taken out from the grid drawer, passed behind the kettle, and set on the middle shelf of the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 22 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_4_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in my room is the fire extinguisher?", + "answer": "The fire extinguisher is directly in front of the dining table.", + "question_chinese": "灭火器在我房间哪个地方？", + "answer_chinese": "灭火器在餐桌的正前方。", + "evidence": [ + { + "description": "The fire extinguisher is on the wall to the left of the microwave oven.", + "timestamps": { + "start": "00:06:50", + "end": "00:08:13" + } + }, + { + "description": "The fire extinguisher is on the wall to the left of the microwave oven.", + "timestamps": { + "start": "00:08:23", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:13:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fire extinguisher is directly in front of the dining table.", + "B": "Behind the dining table.", + "C": "Left of the dining table.", + "D": "The fire extinguisher is directly behind the dining table, closer to the wall than the table itself.", + "E": "The fire extinguisher is on the wall to the left of the microwave, not positioned in front of the dining table." + }, + "answer_label": "A", + "answer_text": "The fire extinguisher is directly in front of the dining table.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_4_T1", + "correct_answer": "The fire extinguisher is directly in front of the dining table.", + "original_options": { + "A": "The fire extinguisher is directly in front of the dining table.", + "B": "Behind the dining table.", + "C": "Left of the dining table.", + "D": "The fire extinguisher is directly behind the dining table, closer to the wall than the table itself.", + "E": "The fire extinguisher is on the wall to the left of the microwave, not positioned in front of the dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 21 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_4_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in my room is the fire extinguisher?", + "answer": "The fire extinguisher is on the wall to the left of the microwave.", + "question_chinese": "灭火器在我房间哪个地方？", + "answer_chinese": "灭火器在微波炉左边的墙面上。", + "evidence": [ + { + "description": "The fire extinguisher is on the wall to the left of the microwave oven.", + "timestamps": { + "start": "00:06:50", + "end": "00:08:13" + } + }, + { + "description": "The fire extinguisher is on the wall to the left of the microwave oven.", + "timestamps": { + "start": "00:08:23", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:25:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the wall to the right.", + "B": "On the wall behind the microwave.", + "C": "The fire extinguisher is mounted on the wall just to the right of the microwave, near its edge.", + "D": "The fire extinguisher is on the wall above the microwave, centered so you can reach it easily.", + "E": "The fire extinguisher is on the wall to the left of the microwave." + }, + "answer_label": "E", + "answer_text": "The fire extinguisher is on the wall to the left of the microwave.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_4_T2", + "correct_answer": "The fire extinguisher is on the wall to the left of the microwave.", + "original_options": { + "A": "On the wall to the right.", + "B": "On the wall behind the microwave.", + "C": "The fire extinguisher is mounted on the wall just to the right of the microwave, near its edge.", + "D": "The fire extinguisher is on the wall above the microwave, centered so you can reach it easily.", + "E": "The fire extinguisher is on the wall to the left of the microwave." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_6", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the fire extinguisher in the room?", + "answer": "The fire extinguisher is on the wall to the left of the microwave oven.", + "question_chinese": "灭火器在房间哪个地方？", + "answer_chinese": "灭火器在微波炉左边的墙面上。", + "evidence": [ + { + "description": "The fire extinguisher is on the wall to the left of the microwave oven.", + "timestamps": { + "start": "00:06:50", + "end": "00:08:13" + } + }, + { + "description": "The fire extinguisher is on the wall to the left of the microwave oven.", + "timestamps": { + "start": "00:08:23", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:19:59", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the wall to the right.", + "B": "On the floor beside the microwave.", + "C": "The fire extinguisher is on the wall to the left of the microwave oven.", + "D": "The fire extinguisher is hanging on the wall directly above the microwave oven, not to either side.", + "E": "The fire extinguisher is mounted on the wall to the right of the microwave oven, close to the corner." + }, + "answer_label": "C", + "answer_text": "The fire extinguisher is on the wall to the left of the microwave oven.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_6", + "correct_answer": "The fire extinguisher is on the wall to the left of the microwave oven.", + "original_options": { + "A": "On the wall to the right.", + "B": "On the floor beside the microwave.", + "C": "The fire extinguisher is on the wall to the left of the microwave oven.", + "D": "The fire extinguisher is hanging on the wall directly above the microwave oven, not to either side.", + "E": "The fire extinguisher is mounted on the wall to the right of the microwave oven, close to the corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_7", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the fire extinguisher in the room?", + "answer": "The fire extinguisher is on the wall to the left of the microwave oven.", + "question_chinese": "灭火器在房间哪个地方？", + "answer_chinese": "灭火器在微波炉左边的墙面上。", + "evidence": [ + { + "description": "The fire extinguisher is on the wall to the left of the microwave oven.", + "timestamps": { + "start": "00:06:50", + "end": "00:08:13" + } + }, + { + "description": "The fire extinguisher is on the wall to the left of the microwave oven.", + "timestamps": { + "start": "00:08:23", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:25:01", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the wall to the right.", + "B": "Beside the kitchen door.", + "C": "The fire extinguisher is mounted on the wall above the counter, not positioned next to the microwave oven.", + "D": "The fire extinguisher is on the wall near the sink on the right side, away from the microwave area.", + "E": "The fire extinguisher is on the wall to the left of the microwave oven." + }, + "answer_label": "E", + "answer_text": "The fire extinguisher is on the wall to the left of the microwave oven.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_7", + "correct_answer": "The fire extinguisher is on the wall to the left of the microwave oven.", + "original_options": { + "A": "On the wall to the right.", + "B": "Beside the kitchen door.", + "C": "The fire extinguisher is mounted on the wall above the counter, not positioned next to the microwave oven.", + "D": "The fire extinguisher is on the wall near the sink on the right side, away from the microwave area.", + "E": "The fire extinguisher is on the wall to the left of the microwave oven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 18, + 19 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_3_T1", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the green cup take from being to my left to being directly across from me?", + "answer": "The green cup was moved from your left to the mouth of the person on the left.", + "question_chinese": "绿色杯子从我的左边到我的对面经历了什么路径？", + "answer_chinese": "绿色杯子被从你的左边拿到左边人的嘴边。", + "evidence": [ + { + "description": "The green cup was placed across from you.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:37" + } + }, + { + "description": "The green cup was taken from your left to the mouth of the person on your left.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:31" + } + } + ], + "question_timestamps": "00:03:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From your left to your right.", + "B": "From your left to your mouth.", + "C": "The green cup was moved from your left to the hand of the person across from you, then held there.", + "D": "The green cup was moved from your left to the mouth of the person on your right before being set down across.", + "E": "The green cup was moved from your left to the mouth of the person on the left." + }, + "answer_label": "E", + "answer_text": "The green cup was moved from your left to the mouth of the person on the left.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_3_T1", + "correct_answer": "The green cup was moved from your left to the mouth of the person on the left.", + "original_options": { + "A": "From your left to your right.", + "B": "From your left to your mouth.", + "C": "The green cup was moved from your left to the hand of the person across from you, then held there.", + "D": "The green cup was moved from your left to the mouth of the person on your right before being set down across.", + "E": "The green cup was moved from your left to the mouth of the person on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 22 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_5", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the green cup take from my left side to my opposite side?", + "answer": "The green cup was taken from your left to the mouth of the person on your left,and then placed across from you.", + "question_chinese": "绿色杯子从我的左边到我的对面经历了什么路径？", + "answer_chinese": "绿色杯子被从你的左边拿到左边人的嘴边，又被放到你的对面。", + "evidence": [ + { + "description": "The green cup was placed across from you.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:37" + } + }, + { + "description": "The green cup was taken from your left to the mouth of the person on your left.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:31" + } + } + ], + "question_timestamps": "00:09:11", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From your left to your right.", + "B": "The green cup was taken from your left to the mouth of the person on your left,and then placed across from you.", + "C": "From your left to the sink.", + "D": "The green cup was taken from your left to the hand of the person across from you, then placed by them.", + "E": "The green cup was taken from your left to the mouth of the person on your right, then set down near you." + }, + "answer_label": "B", + "answer_text": "The green cup was taken from your left to the mouth of the person on your left,and then placed across from you.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_5", + "correct_answer": "The green cup was taken from your left to the mouth of the person on your left,and then placed across from you.", + "original_options": { + "A": "From your left to your right.", + "B": "The green cup was taken from your left to the mouth of the person on your left,and then placed across from you.", + "C": "From your left to the sink.", + "D": "The green cup was taken from your left to the hand of the person across from you, then placed by them.", + "E": "The green cup was taken from your left to the mouth of the person on your right, then set down near you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 22 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_3_T2", + "qid_num": 24, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the green cup take from being to my left to being directly across from me?", + "answer": "The green cup was moved from your left to the mouth of the person on the left, and then it was set down across from you.", + "question_chinese": "绿色杯子从我的左边到我的对面经历了什么路径？", + "answer_chinese": "绿色杯子被从你的左边拿到左边人的嘴边，又被放到你的对面。", + "evidence": [ + { + "description": "The green cup was taken from my left to the mouth of the person on my left.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:31" + } + }, + { + "description": "The green cup was placed across from me.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:16:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From your left to your right.", + "B": "From your left to your mouth.", + "C": "The green cup was moved from your right to the mouth of the person on the left, then set down near you.", + "D": "The green cup was moved from your left to the mouth of the person on the left, and then it was set down across from you.", + "E": "The green cup was moved from your left to the mouth of the person across from you, then placed to your left." + }, + "answer_label": "D", + "answer_text": "The green cup was moved from your left to the mouth of the person on the left, and then it was set down across from you.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_3_T2", + "correct_answer": "The green cup was moved from your left to the mouth of the person on the left, and then it was set down across from you.", + "original_options": { + "A": "From your left to your right.", + "B": "From your left to your mouth.", + "C": "The green cup was moved from your right to the mouth of the person on the left, then set down near you.", + "D": "The green cup was moved from your left to the mouth of the person on the left, and then it was set down across from you.", + "E": "The green cup was moved from your left to the mouth of the person across from you, then placed to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 22 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_16", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the wardrobe located in the bedroom?", + "answer": "The wardrobe is next to the bedroom door.", + "question_chinese": "衣柜在卧室哪个位置？", + "answer_chinese": "衣柜在卧室门旁边。", + "evidence": [ + { + "description": "The wardrobe is next to the bedroom door.", + "timestamps": { + "start": "00:28:34", + "end": "00:28:37" + } + } + ], + "question_timestamps": "00:29:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Across from the bedroom window.", + "B": "Near the foot of the bed.", + "C": "The wardrobe is positioned beside the bed in the corner, rather than close to the bedroom door.", + "D": "The wardrobe is next to the bedroom door.", + "E": "The wardrobe is located against the wall by the window, with space between it and the bedroom door." + }, + "answer_label": "D", + "answer_text": "The wardrobe is next to the bedroom door.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_16", + "correct_answer": "The wardrobe is next to the bedroom door.", + "original_options": { + "A": "Across from the bedroom window.", + "B": "Near the foot of the bed.", + "C": "The wardrobe is positioned beside the bed in the corner, rather than close to the bedroom door.", + "D": "The wardrobe is next to the bedroom door.", + "E": "The wardrobe is located against the wall by the window, with space between it and the bedroom door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 18 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_18", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the wardrobe located in the bedroom?", + "answer": "The wardrobe is next to the bedroom door.", + "question_chinese": "衣柜在卧室哪个位置？", + "answer_chinese": "衣柜在卧室门旁边。", + "evidence": [ + { + "description": "The wardrobe is next to the bedroom door.", + "timestamps": { + "start": "00:28:34", + "end": "00:28:37" + } + } + ], + "question_timestamps": "00:29:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Across from the bedroom window.", + "B": "The wardrobe is next to the bedroom door.", + "C": "Behind the bedroom curtain.", + "D": "The wardrobe sits near the foot of the bed, not next to the bedroom door as you enter.", + "E": "The wardrobe is tucked into a corner of the bedroom, away from the doorway and entry path." + }, + "answer_label": "B", + "answer_text": "The wardrobe is next to the bedroom door.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_18", + "correct_answer": "The wardrobe is next to the bedroom door.", + "original_options": { + "A": "Across from the bedroom window.", + "B": "The wardrobe is next to the bedroom door.", + "C": "Behind the bedroom curtain.", + "D": "The wardrobe sits near the foot of the bed, not next to the bedroom door as you enter.", + "E": "The wardrobe is tucked into a corner of the bedroom, away from the doorway and entry path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 17 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_8", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many chairs are there at the dining table?", + "answer": "There are four chairs next to the dining table.", + "question_chinese": "餐桌边有几把椅子？", + "answer_chinese": "餐桌边有四把椅子。", + "evidence": [ + { + "description": "There are four chairs next to the dining table.", + "timestamps": { + "start": "00:13:44", + "end": "00:13:50" + } + }, + { + "description": "There are four chairs next to the dining table.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:22:17", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are three chairs.", + "B": "There are five chairs.", + "C": "There are four chairs next to the dining table.", + "D": "There are six chairs next to the dining table instead of four.", + "E": "There are seven chairs arranged around the dining table, not four chairs." + }, + "answer_label": "C", + "answer_text": "There are four chairs next to the dining table.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_8", + "correct_answer": "There are four chairs next to the dining table.", + "original_options": { + "A": "There are three chairs.", + "B": "There are five chairs.", + "C": "There are four chairs next to the dining table.", + "D": "There are six chairs next to the dining table instead of four.", + "E": "There are seven chairs arranged around the dining table, not four chairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "30b9ff64-51c8-4481-ba6a-733ea2060aef", + "video_length": "00:30:10", + "qid": "30b9ff64-51c8-4481-ba6a-733ea2060aef_9", + "qid_num": 28, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many chairs are there at the dining table?", + "answer": "There are four chairs next to the dining table.", + "question_chinese": "餐桌边有几把椅子？", + "answer_chinese": "餐桌边有四把椅子。", + "evidence": [ + { + "description": "There are four chairs next to the dining table.", + "timestamps": { + "start": "00:13:44", + "end": "00:13:50" + } + }, + { + "description": "There are four chairs next to the dining table.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:25:53", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are three chairs.", + "B": "There are four chairs next to the dining table.", + "C": "There are five chairs.", + "D": "There are six chairs next to the dining table, not four chairs.", + "E": "There are seven chairs placed by the dining table rather than four." + }, + "answer_label": "B", + "answer_text": "There are four chairs next to the dining table.", + "q_id": "30b9ff64-51c8-4481-ba6a-733ea2060aef_9", + "correct_answer": "There are four chairs next to the dining table.", + "original_options": { + "A": "There are three chairs.", + "B": "There are four chairs next to the dining table.", + "C": "There are five chairs.", + "D": "There are six chairs next to the dining table, not four chairs.", + "E": "There are seven chairs placed by the dining table rather than four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + } +] diff --git a/QAs/33fced0b-4bf6-4989-9174-cda7bb51ecef-mcq.json b/QAs/33fced0b-4bf6-4989-9174-cda7bb51ecef-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..307c6cd5cfc8c9be462c68d86db7bda21485f76f --- /dev/null +++ b/QAs/33fced0b-4bf6-4989-9174-cda7bb51ecef-mcq.json @@ -0,0 +1,1040 @@ +[ + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_13", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the vacuum cleaner located at home?", + "answer": "The vacuum cleaner is on the bottom step of the stairs.", + "question_chinese": "吸尘机在家里哪个位置？", + "answer_chinese": "吸尘机在楼梯的最下面的台阶上。", + "evidence": [ + { + "description": "The vacuum cleaner is on the bottom step of the stairs.", + "timestamps": { + "start": "00:10:41", + "end": "00:10:44" + } + } + ], + "question_timestamps": "00:22:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the top step.", + "B": "The vacuum cleaner is on the bottom step of the stairs.", + "C": "In the hallway.", + "D": "The vacuum cleaner is leaning against the wall in the hallway near the coat rack.", + "E": "The vacuum cleaner is tucked under the dining table in the dining area, not on the stairs." + }, + "answer_label": "B", + "answer_text": "The vacuum cleaner is on the bottom step of the stairs.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_13", + "correct_answer": "The vacuum cleaner is on the bottom step of the stairs.", + "original_options": { + "A": "On the top step.", + "B": "The vacuum cleaner is on the bottom step of the stairs.", + "C": "In the hallway.", + "D": "The vacuum cleaner is leaning against the wall in the hallway near the coat rack.", + "E": "The vacuum cleaner is tucked under the dining table in the dining area, not on the stairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 17 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_20", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the vacuum cleaner located at home?", + "answer": "The vacuum cleaner is on the bottom step of the stairs.", + "question_chinese": "吸尘机在家里哪个位置？", + "answer_chinese": "吸尘机在楼梯的最下面的台阶上。", + "evidence": [ + { + "description": "The vacuum cleaner is on the bottom step of the stairs.", + "timestamps": { + "start": "00:10:41", + "end": "00:10:44" + } + } + ], + "question_timestamps": "00:28:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the kitchen.", + "B": "In the hallway.", + "C": "The vacuum cleaner is on the top shelf of the closet rather than on the stairs.", + "D": "The vacuum cleaner is by the bathroom door near the towel rack, not on the bottom step.", + "E": "The vacuum cleaner is on the bottom step of the stairs." + }, + "answer_label": "E", + "answer_text": "The vacuum cleaner is on the bottom step of the stairs.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_20", + "correct_answer": "The vacuum cleaner is on the bottom step of the stairs.", + "original_options": { + "A": "In the kitchen.", + "B": "In the hallway.", + "C": "The vacuum cleaner is on the top shelf of the closet rather than on the stairs.", + "D": "The vacuum cleaner is by the bathroom door near the towel rack, not on the bottom step.", + "E": "The vacuum cleaner is on the bottom step of the stairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I hang the apron before using it?", + "answer": "The apron was hanging on the cabinet diagonally opposite the refrigerator.", + "question_chinese": "围裙在我使用前在哪里挂着？", + "answer_chinese": "围裙在冰箱斜对面橱柜上挂着。", + "evidence": [ + { + "description": "The apron was hanging on the cabinet diagonally opposite the refrigerator.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:08:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the hook by the stove.", + "B": "On the shelf by the sink.", + "C": "The apron was hanging on the wall behind the door, not on a cabinet across the kitchen.", + "D": "The apron was hanging on the cabinet diagonally opposite the refrigerator.", + "E": "The apron was draped over the counter across from the oven instead of the cabinet diagonally opposite the refrigerator." + }, + "answer_label": "D", + "answer_text": "The apron was hanging on the cabinet diagonally opposite the refrigerator.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_2", + "correct_answer": "The apron was hanging on the cabinet diagonally opposite the refrigerator.", + "original_options": { + "A": "On the hook by the stove.", + "B": "On the shelf by the sink.", + "C": "The apron was hanging on the wall behind the door, not on a cabinet across the kitchen.", + "D": "The apron was hanging on the cabinet diagonally opposite the refrigerator.", + "E": "The apron was draped over the counter across from the oven instead of the cabinet diagonally opposite the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I hang the apron before using it?", + "answer": "The apron was hanging on the cabinet diagonally opposite the refrigerator.", + "question_chinese": "围裙在我使用前在哪里挂着？", + "answer_chinese": "围裙在冰箱斜对面橱柜上挂着。", + "evidence": [ + { + "description": "The apron is hanging on the cabinet diagonally opposite the refrigerator.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:17:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the wall behind it.", + "B": "The apron was hanging on the cabinet diagonally opposite the refrigerator.", + "C": "On the chair beside it.", + "D": "The apron was hanging under the counter near the sink rather than on a cabinet across from the fridge.", + "E": "The apron was placed on the shelf next to the oven, not on the cabinet diagonally opposite the refrigerator." + }, + "answer_label": "B", + "answer_text": "The apron was hanging on the cabinet diagonally opposite the refrigerator.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_1", + "correct_answer": "The apron was hanging on the cabinet diagonally opposite the refrigerator.", + "original_options": { + "A": "On the wall behind it.", + "B": "The apron was hanging on the cabinet diagonally opposite the refrigerator.", + "C": "On the chair beside it.", + "D": "The apron was hanging under the counter near the sink rather than on a cabinet across from the fridge.", + "E": "The apron was placed on the shelf next to the oven, not on the cabinet diagonally opposite the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_7", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I use for cooking just now?", + "answer": "You just used three potatoes for cooking.", + "question_chinese": "我刚才做菜用了几个土豆?", + "answer_chinese": "你刚才做菜用了三个土豆。", + "evidence": [ + { + "description": "Three potatoes were used for cooking.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:20" + } + }, + { + "description": "Three potatoes were used for cooking.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:29" + } + } + ], + "question_timestamps": "00:10:28", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You just used three potatoes for cooking.", + "B": "You used one potato.", + "C": "You used two potatoes.", + "D": "You just used four potatoes for cooking, not three potatoes in total.", + "E": "You just used five potatoes for cooking instead of three during the preparation." + }, + "answer_label": "A", + "answer_text": "You just used three potatoes for cooking.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_7", + "correct_answer": "You just used three potatoes for cooking.", + "original_options": { + "A": "You just used three potatoes for cooking.", + "B": "You used one potato.", + "C": "You used two potatoes.", + "D": "You just used four potatoes for cooking, not three potatoes in total.", + "E": "You just used five potatoes for cooking instead of three during the preparation." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_6", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I use for cooking just now?", + "answer": "You just used three potatoes for cooking.", + "question_chinese": "我刚才做菜用了几个土豆？", + "answer_chinese": "你刚才做菜用了三个土豆。", + "evidence": [ + { + "description": "Three potatoes were used for cooking.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:20" + } + }, + { + "description": "Three potatoes were used for cooking.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:29" + } + } + ], + "question_timestamps": "00:18:35", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You used one potato.", + "B": "You used two potatoes.", + "C": "You just used four potatoes for cooking rather than three potatoes in total today.", + "D": "You just used three potatoes for cooking.", + "E": "You just used five potatoes for cooking this time, not three potatoes." + }, + "answer_label": "D", + "answer_text": "You just used three potatoes for cooking.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_6", + "correct_answer": "You just used three potatoes for cooking.", + "original_options": { + "A": "You used one potato.", + "B": "You used two potatoes.", + "C": "You just used four potatoes for cooking rather than three potatoes in total today.", + "D": "You just used three potatoes for cooking.", + "E": "You just used five potatoes for cooking this time, not three potatoes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 12 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many seasonings did I use for castrating the chicken leg just now?", + "answer": "You just used three different seasonings for castrating the chicken leg.", + "question_chinese": "我刚才阉鸡腿用了几种调味料？", + "answer_chinese": "你刚才阉鸡腿用了3种调味料。", + "evidence": [ + { + "description": "I used another one kind of seasoning for castrating the chicken leg.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:38" + } + }, + { + "description": "I just used one kind of seasoning for castrating the chicken leg.", + "timestamps": { + "start": "00:08:07", + "end": "00:08:15" + } + }, + { + "description": "I just used the third kind of seasoning for castrating the chicken leg.", + "timestamps": { + "start": "00:08:39", + "end": "00:08:53" + } + } + ], + "question_timestamps": "00:14:53", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used two seasonings.", + "B": "You used four seasonings.", + "C": "You just used three different seasonings for castrating the chicken leg.", + "D": "You just used one seasoning to castrate the chicken leg, keeping it simple with a single flavor.", + "E": "You just used five different seasonings for castrating the chicken leg, layering several flavors one after another." + }, + "answer_label": "C", + "answer_text": "You just used three different seasonings for castrating the chicken leg.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_11", + "correct_answer": "You just used three different seasonings for castrating the chicken leg.", + "original_options": { + "A": "You used two seasonings.", + "B": "You used four seasonings.", + "C": "You just used three different seasonings for castrating the chicken leg.", + "D": "You just used one seasoning to castrate the chicken leg, keeping it simple with a single flavor.", + "E": "You just used five different seasonings for castrating the chicken leg, layering several flavors one after another." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_12", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many seasonings did I use for castrating the chicken leg just now?", + "answer": "You iust used three different seasonings for castrating the chicken leg.", + "question_chinese": "我刚才阉鸡腿用了几种调味料？", + "answer_chinese": "你刚才阉鸡腿用了3种调味料。", + "evidence": [ + { + "description": "This is the third seasoning I have used.", + "timestamps": { + "start": "00:08:39", + "end": "00:08:53" + } + }, + { + "description": "This is the second seasoning I have used.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:38" + } + }, + { + "description": "This is the first seasoning I have used.", + "timestamps": { + "start": "00:08:07", + "end": "00:08:15" + } + } + ], + "question_timestamps": "00:24:47", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used one seasoning.", + "B": "You used four seasonings.", + "C": "You iust used two different seasonings for castrating the chicken leg, adding only a pair of flavors.", + "D": "You iust used seven different seasonings for castrating the chicken leg, mixing many flavors into the chicken leg.", + "E": "You iust used three different seasonings for castrating the chicken leg." + }, + "answer_label": "E", + "answer_text": "You iust used three different seasonings for castrating the chicken leg.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_12", + "correct_answer": "You iust used three different seasonings for castrating the chicken leg.", + "original_options": { + "A": "You used one seasoning.", + "B": "You used four seasonings.", + "C": "You iust used two different seasonings for castrating the chicken leg, adding only a pair of flavors.", + "D": "You iust used seven different seasonings for castrating the chicken leg, mixing many flavors into the chicken leg.", + "E": "You iust used three different seasonings for castrating the chicken leg." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_3", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to put the gloves back in their original position, where should I put them back?", + "answer": "You should put the gloves back in the plate above the microwave.", + "question_chinese": "我想把手套放回最开始的位置，我应该把手套放回哪里？", + "answer_chinese": "你应该把手套放回微波炉上方的盘子里。", + "evidence": [ + { + "description": "The gloves was in the plate above the microwave.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:17:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put them in the bowl beside the refrigerator.", + "B": "Put them in the drawer beneath the sink.", + "C": "You should put the gloves back in the plate above the microwave.", + "D": "You should put the gloves back in the bag near the countertop, where you kept them earlier.", + "E": "You should put the gloves back in the basket on the kitchen table, returning them to that spot." + }, + "answer_label": "C", + "answer_text": "You should put the gloves back in the plate above the microwave.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_3", + "correct_answer": "You should put the gloves back in the plate above the microwave.", + "original_options": { + "A": "Put them in the bowl beside the refrigerator.", + "B": "Put them in the drawer beneath the sink.", + "C": "You should put the gloves back in the plate above the microwave.", + "D": "You should put the gloves back in the bag near the countertop, where you kept them earlier.", + "E": "You should put the gloves back in the basket on the kitchen table, returning them to that spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 18 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_10", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to put the gloves back in their original position, where should I put them back?", + "answer": "You should put the gloves back in the plate above the microwave.", + "question_chinese": "我想把手套放回最开始的位置，我应该把手套放回哪里?", + "answer_chinese": "你应该把手套放回微波炉上方的盘子里。", + "evidence": [ + { + "description": "The gloves was in the plate above the microwave.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:26:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put them in the basket beside the refrigerator.", + "B": "Place them in the drawer under the sink.", + "C": "You should put the gloves back in the plate above the microwave.", + "D": "You should put the gloves back on the counter beside the stove, right where you can reach them quickly.", + "E": "You should put the gloves back in the bowl near the cutting board, where they were sitting before." + }, + "answer_label": "C", + "answer_text": "You should put the gloves back in the plate above the microwave.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_10", + "correct_answer": "You should put the gloves back in the plate above the microwave.", + "original_options": { + "A": "Put them in the basket beside the refrigerator.", + "B": "Place them in the drawer under the sink.", + "C": "You should put the gloves back in the plate above the microwave.", + "D": "You should put the gloves back on the counter beside the stove, right where you can reach them quickly.", + "E": "You should put the gloves back in the bowl near the cutting board, where they were sitting before." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 18 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_4", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take when I brought the potatoes from the cabinet directly across from the refrigerator to the sink?", + "answer": "You took the potatoes out of the storage box next to the drinks in the cabinet directly opposite the refrigerator, walked left to the cabinet, and then walked right to the sink.", + "question_chinese": "我把土豆从冰箱正对面柜子里拿到洗手池旁边经历了什么路径？", + "answer_chinese": "你把土豆从冰箱正对面柜子里饮料旁边置物盒里拿出来，向左走到橱柜旁边再向右走到洗手池旁边。", + "evidence": [ + { + "description": "I walked left to the cabinet, and then walked right to the sink.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:26" + } + }, + { + "description": "I took the potatoes out of the storage box next to the drinks in the cabinet directly opposite the refrigerator.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:09:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You walked right, then left to the sink.", + "B": "You walked straight to the sink.", + "C": "You took the potatoes from the cabinet opposite the refrigerator, walked right to the cabinet, then left to the sink.", + "D": "You took the potatoes from the storage box next to the drinks, walked left to the cabinet, then left to the sink.", + "E": "You took the potatoes out of the storage box next to the drinks in the cabinet directly opposite the refrigerator, walked left to the cabinet, and then walked right to the sink." + }, + "answer_label": "E", + "answer_text": "You took the potatoes out of the storage box next to the drinks in the cabinet directly opposite the refrigerator, walked left to the cabinet, and then walked right to the sink.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_4", + "correct_answer": "You took the potatoes out of the storage box next to the drinks in the cabinet directly opposite the refrigerator, walked left to the cabinet, and then walked right to the sink.", + "original_options": { + "A": "You walked right, then left to the sink.", + "B": "You walked straight to the sink.", + "C": "You took the potatoes from the cabinet opposite the refrigerator, walked right to the cabinet, then left to the sink.", + "D": "You took the potatoes from the storage box next to the drinks, walked left to the cabinet, then left to the sink.", + "E": "You took the potatoes out of the storage box next to the drinks in the cabinet directly opposite the refrigerator, walked left to the cabinet, and then walked right to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 20, + 22 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_5", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take when I brought the potatoes from the cabinet directly across from the refrigerator to the sink?", + "answer": "You took the potatoes out of the storage box next to the drinks in the cabinet directly opposite the refrigerator, walked left to the cabinet,and then walked right to the sink.", + "question_chinese": "我把土豆从冰箱正对面柜子里拿到洗手池旁边经历了什么路径？", + "answer_chinese": "你把土豆从冰箱正对面柜子里饮料旁边置物盒里拿出来，向左走到橱柜旁边再向右走到洗手池旁边。", + "evidence": [ + { + "description": "I took the potatoes out of the storage box next to the drinks in the cabinetdirectly opposite the refrigerator.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:20" + } + }, + { + "description": "I walked left to the cabinet,and then walked right to the sink.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:26:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took the potatoes out of the storage box next to the drinks in the cabinet directly opposite the refrigerator, walked left to the cabinet,and then walked right to the sink.", + "B": "You walked right, then left to the sink.", + "C": "You walked left, then straight to the sink.", + "D": "You took the potatoes from the storage box beside the fridge, walked right to the pantry, then left to the sink.", + "E": "You picked potatoes from the cabinet opposite the stove, walked left to the counter, then right to the sink." + }, + "answer_label": "A", + "answer_text": "You took the potatoes out of the storage box next to the drinks in the cabinet directly opposite the refrigerator, walked left to the cabinet,and then walked right to the sink.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_5", + "correct_answer": "You took the potatoes out of the storage box next to the drinks in the cabinet directly opposite the refrigerator, walked left to the cabinet,and then walked right to the sink.", + "original_options": { + "A": "You took the potatoes out of the storage box next to the drinks in the cabinet directly opposite the refrigerator, walked left to the cabinet,and then walked right to the sink.", + "B": "You walked right, then left to the sink.", + "C": "You walked left, then straight to the sink.", + "D": "You took the potatoes from the storage box beside the fridge, walked right to the pantry, then left to the sink.", + "E": "You picked potatoes from the cabinet opposite the stove, walked left to the counter, then right to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 21, + 19 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_15", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the mineral water bucket located at home?", + "answer": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "question_chinese": "矿泉水桶在家里哪个位置？", + "answer_chinese": "矿泉水桶在灭火器下方的蓝色水盆里。", + "evidence": [ + { + "description": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "timestamps": { + "start": "00:16:49", + "end": "00:16:56" + } + }, + { + "description": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "timestamps": { + "start": "00:18:44", + "end": "00:18:45" + } + } + ], + "question_timestamps": "00:22:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It is under the kitchen sink.", + "B": "It is on the counter by the microwave.", + "C": "The mineral water bucket is in the green container beside the washing machine, tucked next to it on the floor.", + "D": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "E": "The mineral water bucket is behind the sofa near the bookshelf, sitting out of the way in the living area." + }, + "answer_label": "D", + "answer_text": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_15", + "correct_answer": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "original_options": { + "A": "It is under the kitchen sink.", + "B": "It is on the counter by the microwave.", + "C": "The mineral water bucket is in the green container beside the washing machine, tucked next to it on the floor.", + "D": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "E": "The mineral water bucket is behind the sofa near the bookshelf, sitting out of the way in the living area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 20, + 20 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_14", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the mineral water bucket located at home?", + "answer": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "question_chinese": "矿泉水桶在家里哪个位置？", + "answer_chinese": "矿泉水桶在灭火器下方的蓝色水盆里。", + "evidence": [ + { + "description": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "timestamps": { + "start": "00:16:49", + "end": "00:16:56" + } + }, + { + "description": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "timestamps": { + "start": "00:18:44", + "end": "00:18:45" + } + } + ], + "question_timestamps": "00:29:57", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "B": "It is behind the sofa.", + "C": "It is on the counter near the refrigerator.", + "D": "The mineral water bucket is under the kitchen sink next to the broom, stored down in that cabinet space.", + "E": "The mineral water bucket is in the green drawer opposite the microwave, placed inside that drawer for storage." + }, + "answer_label": "A", + "answer_text": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_14", + "correct_answer": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "original_options": { + "A": "The mineral water bucket is in the blue basin below the fire extinguisher.", + "B": "It is behind the sofa.", + "C": "It is on the counter near the refrigerator.", + "D": "The mineral water bucket is under the kitchen sink next to the broom, stored down in that cabinet space.", + "E": "The mineral water bucket is in the green drawer opposite the microwave, placed inside that drawer for storage." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 19, + 18 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_16", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the mineral water bottle take from the right side of the sink to the sink?", + "answer": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "question_chinese": "矿泉水瓶被从洗手池右边到水槽经历了什么路径？", + "answer_chinese": "矿泉水瓶被从洗手池右边拿起来，经过玻璃水瓶上方，被放到水槽里。", + "evidence": [ + { + "description": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "timestamps": { + "start": "00:18:46", + "end": "00:18:53" + } + }, + { + "description": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "timestamps": { + "start": "00:18:54", + "end": "00:18:56" + } + } + ], + "question_timestamps": "00:26:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You set it on the counter.", + "B": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "C": "You placed it in the dish drainer.", + "D": "You picked it up from the right, moved beside the glass bottle, and set it on the counter.", + "E": "You lifted it from the right side, passed behind the glass water bottle, and left it next to the faucet." + }, + "answer_label": "B", + "answer_text": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_16", + "correct_answer": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "original_options": { + "A": "You set it on the counter.", + "B": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "C": "You placed it in the dish drainer.", + "D": "You picked it up from the right, moved beside the glass bottle, and set it on the counter.", + "E": "You lifted it from the right side, passed behind the glass water bottle, and left it next to the faucet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 20 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_21", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the mineral water bottle take from the right side of the sink to the sink?", + "answer": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "question_chinese": "矿泉水瓶被从洗手池右边到水槽经历了什么路径？", + "answer_chinese": "矿泉水瓶被从洗手池右边拿起来，经过玻璃水瓶上方，被放到水槽里。", + "evidence": [ + { + "description": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "timestamps": { + "start": "00:18:46", + "end": "00:18:53" + } + }, + { + "description": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "timestamps": { + "start": "00:18:54", + "end": "00:18:56" + } + } + ], + "question_timestamps": "00:32:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You put it in the dishwasher.", + "B": "You left it by the faucet.", + "C": "You lifted it from the right side, passed beside the glass water bottle, and set it on the counter.", + "D": "You moved it from the left side of the sink, passed behind the glass water bottle, and placed it by the faucet.", + "E": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink." + }, + "answer_label": "E", + "answer_text": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_21", + "correct_answer": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink.", + "original_options": { + "A": "You put it in the dishwasher.", + "B": "You left it by the faucet.", + "C": "You lifted it from the right side, passed beside the glass water bottle, and set it on the counter.", + "D": "You moved it from the left side of the sink, passed behind the glass water bottle, and placed it by the faucet.", + "E": "The mineral water bottle was picked up from the right side of the sink, passed over the glass water bottle, and placed in the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 22 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_19", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the black backpack at home?", + "answer": "The black backpack is on the gray sofa.", + "question_chinese": "黑色书包在家里哪个地方？", + "answer_chinese": "黑色书包在灰色沙发上。", + "evidence": [ + { + "description": "The black backpack is on the gray sofa.", + "timestamps": { + "start": "00:33:39", + "end": "00:33:43" + } + } + ], + "question_timestamps": "00:34:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The black backpack is on the gray sofa.", + "B": "On the blue rug.", + "C": "Under the white table.", + "D": "The black backpack is beside the green armchair in the living room.", + "E": "The black backpack is in the hallway closet, tucked away near the door." + }, + "answer_label": "A", + "answer_text": "The black backpack is on the gray sofa.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_19", + "correct_answer": "The black backpack is on the gray sofa.", + "original_options": { + "A": "The black backpack is on the gray sofa.", + "B": "On the blue rug.", + "C": "Under the white table.", + "D": "The black backpack is beside the green armchair in the living room.", + "E": "The black backpack is in the hallway closet, tucked away near the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "33fced0b-4bf6-4989-9174-cda7bb51ecef", + "video_length": "00:34:42", + "qid": "33fced0b-4bf6-4989-9174-cda7bb51ecef_22", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the black backpack at home?", + "answer": "The black backpack is on the gray sofa.", + "question_chinese": "黑色书包在家里哪个地方？", + "answer_chinese": "黑色书包在灰色沙发上。", + "evidence": [ + { + "description": "The black backpack is on the gray sofa.", + "timestamps": { + "start": "00:33:39", + "end": "00:33:43" + } + } + ], + "question_timestamps": "00:34:29", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Under the blue table.", + "B": "On the green chair.", + "C": "The black backpack is on the gray sofa.", + "D": "The black backpack is beside the white shelf, resting on the floor.", + "E": "The black backpack is near the brown rug, leaning against the furniture." + }, + "answer_label": "C", + "answer_text": "The black backpack is on the gray sofa.", + "q_id": "33fced0b-4bf6-4989-9174-cda7bb51ecef_22", + "correct_answer": "The black backpack is on the gray sofa.", + "original_options": { + "A": "Under the blue table.", + "B": "On the green chair.", + "C": "The black backpack is on the gray sofa.", + "D": "The black backpack is beside the white shelf, resting on the floor.", + "E": "The black backpack is near the brown rug, leaning against the furniture." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + } +] diff --git a/QAs/34f1149c-d8d5-467c-9e3c-3b59d6393f41-mcq.json b/QAs/34f1149c-d8d5-467c-9e3c-3b59d6393f41-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f6082db8611a214f8a3a8208c42279d9cf324c56 --- /dev/null +++ b/QAs/34f1149c-d8d5-467c-9e3c-3b59d6393f41-mcq.json @@ -0,0 +1,781 @@ +[ + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_10", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of things have I held in my hands so far?", + "answer": "You have held jump ropes, resistance band fitness equipment, and a mobile phone in your hands.", + "question_chinese": "到目前为止，我手里拿过哪几种东西？", + "answer_chinese": "你手里拿过跳绳、弹力带健身器材和手机。", + "evidence": [ + { + "description": "Skipping rope", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + }, + { + "description": "Resistance band fitness equipment and mobile phone", + "timestamps": { + "start": "00:06:57", + "end": "00:07:07" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Jump ropes and a mobile phone.", + "B": "Resistance bands and a water bottle.", + "C": "You have held jump ropes, resistance band fitness equipment, and a smartwatch in your hands.", + "D": "You have held jump ropes, resistance band fitness equipment, and a mobile phone in your hands.", + "E": "You have held resistance band fitness equipment, dumbbells, and a mobile phone in your hands." + }, + "answer_label": "D", + "answer_text": "You have held jump ropes, resistance band fitness equipment, and a mobile phone in your hands.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_10", + "correct_answer": "You have held jump ropes, resistance band fitness equipment, and a mobile phone in your hands.", + "original_options": { + "A": "Jump ropes and a mobile phone.", + "B": "Resistance bands and a water bottle.", + "C": "You have held jump ropes, resistance band fitness equipment, and a smartwatch in your hands.", + "D": "You have held jump ropes, resistance band fitness equipment, and a mobile phone in your hands.", + "E": "You have held resistance band fitness equipment, dumbbells, and a mobile phone in your hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_9", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of things have I held in my hands so far?", + "answer": "You have held jump ropes, resistance band fitness equipment, a mobile phone, and a yoga mat in your hands.", + "question_chinese": "到目前为止，我手里拿过哪几种东西？", + "answer_chinese": "你手里拿过跳绳、弹力带健身器材、手机和瑜伽垫。", + "evidence": [ + { + "description": "You have a skipping rope in your hand.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:16" + } + }, + { + "description": "You are holding a yoga mat in your hand.", + "timestamps": { + "start": "00:15:19", + "end": "00:15:36" + } + }, + { + "description": "You have a resistance band fitness equipment and a mobile phone in your hand.", + "timestamps": { + "start": "00:06:52", + "end": "00:07:02" + } + } + ], + "question_timestamps": "00:21:20", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You have held jump ropes, resistance band fitness equipment, a mobile phone, and a yoga mat in your hands.", + "B": "Jump ropes, bands, and a phone.", + "C": "A foam roller and a mat.", + "D": "You have held dumbbells, a fitness tracker, a smartphone, and a towel in your hands.", + "E": "You have held a yoga block, resistance bands, a tablet, and a jump rope in your hands." + }, + "answer_label": "A", + "answer_text": "You have held jump ropes, resistance band fitness equipment, a mobile phone, and a yoga mat in your hands.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_9", + "correct_answer": "You have held jump ropes, resistance band fitness equipment, a mobile phone, and a yoga mat in your hands.", + "original_options": { + "A": "You have held jump ropes, resistance band fitness equipment, a mobile phone, and a yoga mat in your hands.", + "B": "Jump ropes, bands, and a phone.", + "C": "A foam roller and a mat.", + "D": "You have held dumbbells, a fitness tracker, a smartphone, and a towel in your hands.", + "E": "You have held a yoga block, resistance bands, a tablet, and a jump rope in your hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_7_T2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I grab the white cloth bag without moving my feet?", + "answer": "You can reach the white cloth bag without moving your feet. The white cloth bag is in front of you.", + "question_chinese": "我可以在不移动脚步的情况下拿到白色布袋子吗？", + "answer_chinese": "你能在不移动脚步的情况下拿到白色布袋子。白色布袋子在你的前面。", + "evidence": [ + { + "description": "You can reach the white cloth bag without moving your feet.", + "timestamps": { + "start": "00:06:51", + "end": "00:06:54" + } + } + ], + "question_timestamps": "00:06:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the white cloth bag without moving your feet. The white cloth bag is in front of you.", + "B": "You can't reach the white cloth bag without moving your feet. The white cloth bag is behind you." + }, + "answer_label": "A", + "answer_text": "You can reach the white cloth bag without moving your feet. The white cloth bag is in front of you.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_7_T2", + "correct_answer": "You can reach the white cloth bag without moving your feet. The white cloth bag is in front of you.", + "original_options": { + "A": "You can reach the white cloth bag without moving your feet. The white cloth bag is in front of you.", + "B": "You can't reach the white cloth bag without moving your feet. The white cloth bag is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_7_T1", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I grab the white cloth bag without moving my feet?", + "answer": "You can't reach the white cloth bag without moving your feet. The white cloth bag is behind you.", + "question_chinese": "我可以在不移动脚步的情况下拿到白色布袋子吗？", + "answer_chinese": "你不能在不移动脚步的情况下拿到白色布袋子。白色布袋子在你的后面。", + "evidence": [ + { + "description": "You start from the white cloth bag and need to walk a certain distance to reach your current position.", + "timestamps": { + "start": "00:06:57", + "end": "00:06:59" + } + } + ], + "question_timestamps": "00:07:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the white cloth bag without moving your feet. The white cloth bag is in front of you.", + "B": "You can't reach the white cloth bag without moving your feet. The white cloth bag is behind you." + }, + "answer_label": "B", + "answer_text": "You can't reach the white cloth bag without moving your feet. The white cloth bag is behind you.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_7_T1", + "correct_answer": "You can't reach the white cloth bag without moving your feet. The white cloth bag is behind you.", + "original_options": { + "A": "You can reach the white cloth bag without moving your feet. The white cloth bag is in front of you.", + "B": "You can't reach the white cloth bag without moving your feet. The white cloth bag is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_11", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the skipping rope without moving my feet?", + "answer": "You cannot get the skipping rope.", + "question_chinese": "我现在在不移动脚步的情况下能拿到跳绳吗？", + "answer_chinese": "你无法拿到跳绳。", + "evidence": [ + { + "description": "Jumping rope at the boundary between grass and stone-paved ground", + "timestamps": { + "start": "00:18:49", + "end": "00:18:58" + } + } + ], + "question_timestamps": "00:21:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot get the skipping rope.", + "B": "You can get the skipping rope without moving your feet." + }, + "answer_label": "A", + "answer_text": "You cannot get the skipping rope.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_11", + "correct_answer": "You cannot get the skipping rope.", + "original_options": { + "A": "You cannot get the skipping rope.", + "B": "You can get the skipping rope without moving your feet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_12", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the skipping rope without moving my feet?", + "answer": "You can't get the skipping rope.", + "question_chinese": "我现在在不移动脚步的情况下能拿到跳绳吗？", + "answer_chinese": "你无法拿到跳绳。", + "evidence": [ + { + "description": "Jumping rope at the boundary between grass and stone-paved ground", + "timestamps": { + "start": "00:18:49", + "end": "00:18:58" + } + } + ], + "question_timestamps": "00:31:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can get the skipping rope without moving your feet.", + "B": "You can't get the skipping rope." + }, + "answer_label": "B", + "answer_text": "You can't get the skipping rope.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_12", + "correct_answer": "You can't get the skipping rope.", + "original_options": { + "A": "You can get the skipping rope without moving your feet.", + "B": "You can't get the skipping rope." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_4_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far am I from the grass right now?", + "answer": "You are about 1 meter from the grass right now. The grass is to your right.", + "question_chinese": "我现在离草地有多远？", + "answer_chinese": "你现在离草地有一米左右。草地在你的右边。", + "evidence": [ + { + "description": "Jumping rope on the fallen leaves at the junction of the grass and the stone-paved ground, with a yoga mat on the stone-paved ground.", + "timestamps": { + "start": "00:20:40", + "end": "00:21:16" + } + } + ], + "question_timestamps": "00:21:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are about 1.5 meters from the grass right now. The grass is to your left.", + "B": "You are about 1 meter from the grass right now. The grass is to your right." + }, + "answer_label": "B", + "answer_text": "You are about 1 meter from the grass right now. The grass is to your right.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_4_T2", + "correct_answer": "You are about 1 meter from the grass right now. The grass is to your right.", + "original_options": { + "A": "You are about 1.5 meters from the grass right now. The grass is to your left.", + "B": "You are about 1 meter from the grass right now. The grass is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_4_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far am I from the grass right now?", + "answer": "You are about 1.5 meters from the grass right now. The grass is to your left.", + "question_chinese": "我现在离草地有多远？", + "answer_chinese": "你现在离草地有大概一点五米左右。草地在你的左边。", + "evidence": [ + { + "description": "You are about one and a half meters away from the grass now.", + "timestamps": { + "start": "00:30:23", + "end": "00:30:30" + } + } + ], + "question_timestamps": "00:30:32", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are about 1.5 meters from the grass right now. The grass is to your left.", + "B": "You are about 1 meter from the grass right now. The grass is to your right." + }, + "answer_label": "A", + "answer_text": "You are about 1.5 meters from the grass right now. The grass is to your left.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_4_T1", + "correct_answer": "You are about 1.5 meters from the grass right now. The grass is to your left.", + "original_options": { + "A": "You are about 1.5 meters from the grass right now. The grass is to your left.", + "B": "You are about 1 meter from the grass right now. The grass is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the mobile phone now?", + "answer": "The phone is above the junction of the grass and the concrete ground to my left.", + "question_chinese": "手机现在在哪里？", + "answer_chinese": "手机在我左方草地与水泥地的交界处上面", + "evidence": [ + { + "description": "The phone is above the junction of the grass and the concrete ground to my left.", + "timestamps": { + "start": "00:10:32", + "end": "00:10:46" + } + } + ], + "question_timestamps": "00:10:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the grass to your left.", + "B": "The phone is above the junction of the grass and the concrete ground to my left.", + "C": "At the junction in front of you.", + "D": "The phone is above the junction of the grass and concrete, but it is on your right side.", + "E": "The phone is on the junction of the grass and the concrete ground to your left, not above it." + }, + "answer_label": "B", + "answer_text": "The phone is above the junction of the grass and the concrete ground to my left.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_1", + "correct_answer": "The phone is above the junction of the grass and the concrete ground to my left.", + "original_options": { + "A": "On the grass to your left.", + "B": "The phone is above the junction of the grass and the concrete ground to my left.", + "C": "At the junction in front of you.", + "D": "The phone is above the junction of the grass and concrete, but it is on your right side.", + "E": "The phone is on the junction of the grass and the concrete ground to your left, not above it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the mobile phone now?", + "answer": "It is in the white bag on the ground in front of you.", + "question_chinese": "手机现在在哪里？", + "answer_chinese": "它在你前方地上的白袋子里", + "evidence": [ + { + "description": "It's in the white bag", + "timestamps": { + "start": "00:20:26", + "end": "00:20:33" + } + } + ], + "question_timestamps": "00:21:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In a white bag behind you.", + "B": "In a white bag to your left.", + "C": "It is in the white bag on the ground to your right, not in front of you.", + "D": "It is in the white bag on the ground in front of you.", + "E": "It is inside the white bag, but the bag is on the ground behind you instead of ahead." + }, + "answer_label": "D", + "answer_text": "It is in the white bag on the ground in front of you.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_2", + "correct_answer": "It is in the white bag on the ground in front of you.", + "original_options": { + "A": "In a white bag behind you.", + "B": "In a white bag to your left.", + "C": "It is in the white bag on the ground to your right, not in front of you.", + "D": "It is in the white bag on the ground in front of you.", + "E": "It is inside the white bag, but the bag is on the ground behind you instead of ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 18 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_5", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the tree closest to me now?", + "answer": "The tree closest to you is in front of you to the left.", + "question_chinese": "现在离我最近的树在我的什么方向？", + "answer_chinese": "现在离你最近的树在你的左前方。", + "evidence": [ + { + "description": "The tree closest to you is to the left of the yoga mat.", + "timestamps": { + "start": "00:19:51", + "end": "00:20:01" + } + } + ], + "question_timestamps": "00:22:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The tree closest to you is in front of you to the left.", + "B": "In front of you to the right.", + "C": "Behind you to the left.", + "D": "The closest tree is straight ahead of you, not offset to the left side.", + "E": "The closest tree is off to your right in front of you, rather than in front-left." + }, + "answer_label": "A", + "answer_text": "The tree closest to you is in front of you to the left.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_5", + "correct_answer": "The tree closest to you is in front of you to the left.", + "original_options": { + "A": "The tree closest to you is in front of you to the left.", + "B": "In front of you to the right.", + "C": "Behind you to the left.", + "D": "The closest tree is straight ahead of you, not offset to the left side.", + "E": "The closest tree is off to your right in front of you, rather than in front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 14, + 16 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_6", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the tree closest to me now?", + "answer": "The tree closest to you is in front of you to the left.", + "question_chinese": "现在离我最近的树在我的什么方向？", + "answer_chinese": "现在离你最近的树在你的左前方。", + "evidence": [ + { + "description": "The tree closest to you is to the left of the yoga mat.", + "timestamps": { + "start": "00:19:51", + "end": "00:20:01" + } + } + ], + "question_timestamps": "00:30:33", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you to the right.", + "B": "Directly to your right.", + "C": "The closest tree is just behind you on the left, not in front-left of you.", + "D": "The closest tree is in front of you but slightly to the right, rather than in front-left.", + "E": "The tree closest to you is in front of you to the left." + }, + "answer_label": "E", + "answer_text": "The tree closest to you is in front of you to the left.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_6", + "correct_answer": "The tree closest to you is in front of you to the left.", + "original_options": { + "A": "In front of you to the right.", + "B": "Directly to your right.", + "C": "The closest tree is just behind you on the left, not in front-left of you.", + "D": "The closest tree is in front of you but slightly to the right, rather than in front-left.", + "E": "The tree closest to you is in front of you to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 4, + 15, + 17 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_7", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to the grass, the yoga mat or the skipping rope?", + "answer": "Jump rope is closer to grass than a yoga mat.", + "question_chinese": "瑜伽垫和跳绳哪个离草地更近？", + "answer_chinese": "跳绳比瑜伽垫更接近草地。", + "evidence": [ + { + "description": "Jumping rope on the fallen leaves at the junction of the grass and the stone-paved ground, with a yoga mat on the stone-paved ground.", + "timestamps": { + "start": "00:30:23", + "end": "00:30:34" + } + } + ], + "question_timestamps": "00:32:29", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Jump rope is closer to grass than a yoga mat.", + "B": "Jump rope is not closer to grass than a yoga mat." + }, + "answer_label": "A", + "answer_text": "Jump rope is closer to grass than a yoga mat.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_7", + "correct_answer": "Jump rope is closer to grass than a yoga mat.", + "original_options": { + "A": "Jump rope is closer to grass than a yoga mat.", + "B": "Jump rope is not closer to grass than a yoga mat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_8", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to the grass, the yoga mat or the skipping rope?", + "answer": "Jump rope is closer to grass than a yoga mat.", + "question_chinese": "瑜伽垫和跳绳哪个离草地更近？", + "answer_chinese": "跳绳比瑜伽垫更接近草地。", + "evidence": [ + { + "description": "Jumping rope on the fallen leaves at the junction of the grass and the stone-paved ground, with a yoga mat on the stone-paved ground.", + "timestamps": { + "start": "00:30:23", + "end": "00:30:34" + } + } + ], + "question_timestamps": "00:43:52", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Jump rope is closer to grass than a yoga mat.", + "B": "Jump rope is not closer to grass than a yoga mat." + }, + "answer_label": "A", + "answer_text": "Jump rope is closer to grass than a yoga mat.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_8", + "correct_answer": "Jump rope is closer to grass than a yoga mat.", + "original_options": { + "A": "Jump rope is closer to grass than a yoga mat.", + "B": "Jump rope is not closer to grass than a yoga mat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_15", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to the white cloth bag, the skipping rope or the elastic band fitness equipment?", + "answer": "The resistance band fitness equipment is closer to the white cloth bag.", + "question_chinese": "跳绳和弹力带健身器材哪个离白色布袋子更近。", + "answer_chinese": "弹力带健身器材与白色布袋子更近。", + "evidence": [ + { + "description": "The resistance band fitness equipment is closer to the white cloth bag.", + "timestamps": { + "start": "00:28:26", + "end": "00:28:52" + } + } + ], + "question_timestamps": "00:30:11", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The skipping rope is closer to the white cloth bag.", + "B": "The resistance band fitness equipment is closer to the white cloth bag." + }, + "answer_label": "B", + "answer_text": "The resistance band fitness equipment is closer to the white cloth bag.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_15", + "correct_answer": "The resistance band fitness equipment is closer to the white cloth bag.", + "original_options": { + "A": "The skipping rope is closer to the white cloth bag.", + "B": "The resistance band fitness equipment is closer to the white cloth bag." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41", + "video_length": "00:43:52", + "qid": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_16", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to the white cloth bag, the skipping rope or the elastic band fitness equipment?", + "answer": "The resistance band fitness equipment is closer to the white cloth bag.", + "question_chinese": "跳绳和弹力带健身器材哪个离白色布袋子更近。", + "answer_chinese": "弹力带健身器材更接近白色布袋子。", + "evidence": [ + { + "description": "The resistance band fitness equipment is closer to the white cloth bag.", + "timestamps": { + "start": "00:28:26", + "end": "00:28:52" + } + } + ], + "question_timestamps": "00:43:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The skipping rope is closer to the white cloth bag.", + "B": "The resistance band fitness equipment is closer to the white cloth bag." + }, + "answer_label": "B", + "answer_text": "The resistance band fitness equipment is closer to the white cloth bag.", + "q_id": "34f1149c-d8d5-467c-9e3c-3b59d6393f41_16", + "correct_answer": "The resistance band fitness equipment is closer to the white cloth bag.", + "original_options": { + "A": "The skipping rope is closer to the white cloth bag.", + "B": "The resistance band fitness equipment is closer to the white cloth bag." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + } +] diff --git a/QAs/35e004f4-ad64-4835-9358-38a58b431252-mcq.json b/QAs/35e004f4-ad64-4835-9358-38a58b431252-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8808ccef971175195546e828854136a2ba1722cd --- /dev/null +++ b/QAs/35e004f4-ad64-4835-9358-38a58b431252-mcq.json @@ -0,0 +1,1491 @@ +[ + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_6_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the refrigerator relative to me?", + "answer": "The refrigerator is to your left.", + "question_chinese": "冰箱在我的哪里？", + "answer_chinese": "冰箱在你的左边", + "evidence": [ + { + "description": "The fridge is on your left.", + "timestamps": { + "start": "00:10:16", + "end": "00:10:18" + } + } + ], + "question_timestamps": "00:10:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "The refrigerator is to your left.", + "C": "To your right.", + "D": "The refrigerator is directly in front of you as you face forward.", + "E": "The refrigerator is behind you, so you would need to turn around to see it." + }, + "answer_label": "B", + "answer_text": "The refrigerator is to your left.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_6_T2", + "correct_answer": "The refrigerator is to your left.", + "original_options": { + "A": "In front of you.", + "B": "The refrigerator is to your left.", + "C": "To your right.", + "D": "The refrigerator is directly in front of you as you face forward.", + "E": "The refrigerator is behind you, so you would need to turn around to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 15 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_6_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the refrigerator relative to me?", + "answer": "The refrigerator is in front of you.", + "question_chinese": "冰箱在我的哪里？", + "answer_chinese": "冰箱在你的前面", + "evidence": [ + { + "description": "The fridge is in front of you.", + "timestamps": { + "start": "00:12:35", + "end": "00:12:40" + } + } + ], + "question_timestamps": "00:12:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "The refrigerator is in front of you.", + "C": "Behind you.", + "D": "The refrigerator is to your right side, just off to the side of your view.", + "E": "The refrigerator is behind you, positioned so it is not in your forward line of sight." + }, + "answer_label": "B", + "answer_text": "The refrigerator is in front of you.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_6_T1", + "correct_answer": "The refrigerator is in front of you.", + "original_options": { + "A": "To your left.", + "B": "The refrigerator is in front of you.", + "C": "Behind you.", + "D": "The refrigerator is to your right side, just off to the side of your view.", + "E": "The refrigerator is behind you, positioned so it is not in your forward line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 15, + 16 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_15", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was stir-frying the sauce just now, which was closer to me, the garlic or the scallion?", + "answer": "Garlic is closer to you.", + "question_chinese": "刚刚我在炒酱料的时候，大蒜和葱哪个离我更近？", + "answer_chinese": "大蒜离你更近。", + "evidence": [ + { + "description": "The sauce is in the pot on the stove.", + "timestamps": { + "start": "00:11:56", + "end": "00:12:05" + } + }, + { + "description": "The remaining garlic is on the left side of the refrigerator, about three meters away from you.", + "timestamps": { + "start": "00:10:16", + "end": "00:10:17" + } + }, + { + "description": "The onions are farther away from the refrigerator than the garlic.", + "timestamps": { + "start": "00:12:42", + "end": "00:12:44" + } + } + ], + "question_timestamps": "00:12:48", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Garlic is not closer to you.", + "B": "Garlic is closer to you." + }, + "answer_label": "B", + "answer_text": "Garlic is closer to you.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_15", + "correct_answer": "Garlic is closer to you.", + "original_options": { + "A": "Garlic is not closer to you.", + "B": "Garlic is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_16", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was stir-frying the sauce just now, which was closer to me, the garlic or the scallion?", + "answer": "Garlic is closer to you.", + "question_chinese": "刚刚我在炒酱料的时候，大蒜和葱哪个离我更近？", + "answer_chinese": "大蒜离你更近。", + "evidence": [ + { + "description": "The sauce is in the pot on the stove.", + "timestamps": { + "start": "00:11:56", + "end": "00:12:05" + } + }, + { + "description": "The remaining garlic is on the left side of the refrigerator, about three meters away from you.", + "timestamps": { + "start": "00:10:16", + "end": "00:10:17" + } + }, + { + "description": "The onions are farther away from the refrigerator than the garlic.", + "timestamps": { + "start": "00:12:42", + "end": "00:12:44" + } + } + ], + "question_timestamps": "00:22:00", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Garlic is closer to you.", + "B": "Garlic is not closer to you." + }, + "answer_label": "A", + "answer_text": "Garlic is closer to you.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_16", + "correct_answer": "Garlic is closer to you.", + "original_options": { + "A": "Garlic is closer to you.", + "B": "Garlic is not closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_7_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many knives have I used?", + "answer": "You didn’t use a knife. The knife was in front of you.", + "question_chinese": "到目前为止，我用了多少把刀？", + "answer_chinese": "你没用刀，刀在你前面。", + "evidence": [ + { + "description": "You haven't used the knife. It's in front of you.", + "timestamps": { + "start": "00:02:47", + "end": "00:02:51" + } + } + ], + "question_timestamps": "00:02:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used 1 knife.", + "B": "You used 2 knives.", + "C": "You used 1 knife, and it was to your left while you were working.", + "D": "You didn’t use a knife. The knife was in front of you.", + "E": "You used 2 knives, both placed in front of you as you continued cooking." + }, + "answer_label": "D", + "answer_text": "You didn’t use a knife. The knife was in front of you.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_7_T1", + "correct_answer": "You didn’t use a knife. The knife was in front of you.", + "original_options": { + "A": "You used 1 knife.", + "B": "You used 2 knives.", + "C": "You used 1 knife, and it was to your left while you were working.", + "D": "You didn’t use a knife. The knife was in front of you.", + "E": "You used 2 knives, both placed in front of you as you continued cooking." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_7_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many knives have I used?", + "answer": "You used a knife, and it was in front of you.", + "question_chinese": "到目前为止，我用了多少把刀？", + "answer_chinese": "你用了一把刀，它在你前面。", + "evidence": [ + { + "description": "You used a kitchen knife to chop garlic.", + "timestamps": { + "start": "00:10:30", + "end": "00:10:51" + } + } + ], + "question_timestamps": "00:10:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You didn’t use a knife.", + "B": "You used 2 knives.", + "C": "You used 1 knife, and it was to your left when you were preparing food.", + "D": "You used 1 knife, and it was to your right, not in front of you.", + "E": "You used a knife, and it was in front of you." + }, + "answer_label": "E", + "answer_text": "You used a knife, and it was in front of you.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_7_T2", + "correct_answer": "You used a knife, and it was in front of you.", + "original_options": { + "A": "You didn’t use a knife.", + "B": "You used 2 knives.", + "C": "You used 1 knife, and it was to your left when you were preparing food.", + "D": "You used 1 knife, and it was to your right, not in front of you.", + "E": "You used a knife, and it was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 15, + 15 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_8_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the garlic or the scallions?", + "answer": "The garlic was closer to you; it was to your left.", + "question_chinese": "大蒜和葱哪个离我更近？", + "answer_chinese": "大蒜离你更近，它在你左边。", + "evidence": [ + { + "description": "The remaining garlic is on the left side of the refrigerator, about three meters away from you.", + "timestamps": { + "start": "00:10:16", + "end": "00:10:22" + } + } + ], + "question_timestamps": "00:10:24", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The garlic was closer to you; it was to your left.", + "B": "The scallions were closer to you; they were in front of you." + }, + "answer_label": "A", + "answer_text": "The garlic was closer to you; it was to your left.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_8_T2", + "correct_answer": "The garlic was closer to you; it was to your left.", + "original_options": { + "A": "The garlic was closer to you; it was to your left.", + "B": "The scallions were closer to you; they were in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_8_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the garlic or the scallions?", + "answer": "The scallion was closer to you; it was in front of you.", + "question_chinese": "大蒜和葱哪个离我更近？", + "answer_chinese": "葱离你更近，它在你前面。", + "evidence": [ + { + "description": "The scallions are closer to you.", + "timestamps": { + "start": "00:12:42", + "end": "00:12:44" + } + } + ], + "question_timestamps": "00:12:46", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The scallion was closer to you; it was in front of you.", + "B": "The garlic was closer to you; it was to your left." + }, + "answer_label": "A", + "answer_text": "The scallion was closer to you; it was in front of you.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_8_T1", + "correct_answer": "The scallion was closer to you; it was in front of you.", + "original_options": { + "A": "The scallion was closer to you; it was in front of you.", + "B": "The garlic was closer to you; it was to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_5", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I finish cooking, on which side of me should I place the pot?", + "answer": "The pot should be placed on the stove burner to your left.", + "question_chinese": "当我做完饭后，锅应该放在我的哪一侧？", + "answer_chinese": "锅应该放在你左侧的灶台上。", + "evidence": [ + { + "description": "On the stove on the left, there are three soup pots and one flat pan.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:01:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the burner in front.", + "B": "The pot should be placed on the stove burner to your left.", + "C": "On the counter to your right.", + "D": "Place the pot on the stove burner in front of you when you finish cooking.", + "E": "Set the pot on the table to your right instead of on the stove burner." + }, + "answer_label": "B", + "answer_text": "The pot should be placed on the stove burner to your left.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_5", + "correct_answer": "The pot should be placed on the stove burner to your left.", + "original_options": { + "A": "On the burner in front.", + "B": "The pot should be placed on the stove burner to your left.", + "C": "On the counter to your right.", + "D": "Place the pot on the stove burner in front of you when you finish cooking.", + "E": "Set the pot on the table to your right instead of on the stove burner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_6", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I finish cooking, on which side of me should I place the pot?", + "answer": "The pot should be placed on the stove to your left.", + "question_chinese": "当我做完饭后，锅应该放在我的哪一侧？", + "answer_chinese": "锅应该放在你左侧的灶台上。", + "evidence": [ + { + "description": "There are three soup pots and two flat pans on the stove to your left.", + "timestamps": { + "start": "00:26:22", + "end": "00:26:23" + } + } + ], + "question_timestamps": "00:26:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the stove in front.", + "B": "On the counter to your right.", + "C": "Put the pot on the table to your right after you finish cooking.", + "D": "Place the pot on the shelf above you, not on the stove to your left.", + "E": "The pot should be placed on the stove to your left." + }, + "answer_label": "E", + "answer_text": "The pot should be placed on the stove to your left.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_6", + "correct_answer": "The pot should be placed on the stove to your left.", + "original_options": { + "A": "On the stove in front.", + "B": "On the counter to your right.", + "C": "Put the pot on the table to your right after you finish cooking.", + "D": "Place the pot on the shelf above you, not on the stove to your left.", + "E": "The pot should be placed on the stove to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 15 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_19", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following just now?", + "answer": "You was leaving the kitchen, turning right, walking straight through the hallway, and moving towards the living room.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你正从厨房出发，右转，直行穿过走廊，朝着客厅方向移动。", + "evidence": [ + { + "description": "Carry the food out of the kitchen, turn right, walk straight through the hallway, and place the food on the dining table in the living room.", + "timestamps": { + "start": "00:37:37", + "end": "00:37:43" + } + } + ], + "question_timestamps": "00:37:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Toward the dining room.", + "B": "Toward the foyer.", + "C": "You was leaving the kitchen, turning right, walking straight through the hallway, and moving towards the living room.", + "D": "You was leaving the kitchen, turning left, walking through the hallway, and heading toward the living room.", + "E": "You was leaving the kitchen, turning right, then cutting into the dining room instead of continuing to the living room." + }, + "answer_label": "C", + "answer_text": "You was leaving the kitchen, turning right, walking straight through the hallway, and moving towards the living room.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_19", + "correct_answer": "You was leaving the kitchen, turning right, walking straight through the hallway, and moving towards the living room.", + "original_options": { + "A": "Toward the dining room.", + "B": "Toward the foyer.", + "C": "You was leaving the kitchen, turning right, walking straight through the hallway, and moving towards the living room.", + "D": "You was leaving the kitchen, turning left, walking through the hallway, and heading toward the living room.", + "E": "You was leaving the kitchen, turning right, then cutting into the dining room instead of continuing to the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 20 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_20", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following just now?", + "answer": "You was moving from the living room, straight through the hallway, towards the kitchen.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你正从客厅出发，直行穿过走廊，朝着厨房方向移动。", + "evidence": [ + { + "description": "You leave the dining table with an empty plate, walk straight down the hallway, turn left, and return to the kitchen.", + "timestamps": { + "start": "00:38:09", + "end": "00:38:13" + } + } + ], + "question_timestamps": "00:38:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Toward the dining room.", + "B": "From the kitchen to the living room.", + "C": "You was moving from the living room, turning right at the hallway, and heading toward the kitchen.", + "D": "You was moving from the hallway, straight through the living room, then heading toward the kitchen.", + "E": "You was moving from the living room, straight through the hallway, towards the kitchen." + }, + "answer_label": "E", + "answer_text": "You was moving from the living room, straight through the hallway, towards the kitchen.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_20", + "correct_answer": "You was moving from the living room, straight through the hallway, towards the kitchen.", + "original_options": { + "A": "Toward the dining room.", + "B": "From the kitchen to the living room.", + "C": "You was moving from the living room, turning right at the hallway, and heading toward the kitchen.", + "D": "You was moving from the hallway, straight through the living room, then heading toward the kitchen.", + "E": "You was moving from the living room, straight through the hallway, towards the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 17, + 16 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_3", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was standing next to the sink just now, could I reach the faucet directly?", + "answer": "You can reach the faucet directly.", + "question_chinese": "我刚才在水槽旁边的时候，能直接够到水龙头吗？", + "answer_chinese": "你能直接够到水龙头。", + "evidence": [ + { + "description": "The faucet is located between the two sinks.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:14" + } + }, + { + "description": "You placed the lid directly on the pot in the sink on the right, so you could reach the nearby faucet more easily.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can reach the faucet directly.", + "B": "You cannot reach the faucet directly." + }, + "answer_label": "A", + "answer_text": "You can reach the faucet directly.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_3", + "correct_answer": "You can reach the faucet directly.", + "original_options": { + "A": "You can reach the faucet directly.", + "B": "You cannot reach the faucet directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_4", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was standing next to the sink just now, could I reach the faucet directly?", + "answer": "You can reach the faucet directly.", + "question_chinese": "我刚才在水槽旁边的时候，能直接够到水龙头吗？", + "answer_chinese": "你能直接够到水龙头。", + "evidence": [ + { + "description": "The faucet is located between the two sinks.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:14" + } + }, + { + "description": "You placed the lid directly on the pot in the sink on the right, so you could reach the nearby faucet more easily.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:28:58", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can reach the faucet directly.", + "B": "You cannot reach the faucet directly." + }, + "answer_label": "A", + "answer_text": "You can reach the faucet directly.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_4", + "correct_answer": "You can reach the faucet directly.", + "original_options": { + "A": "You can reach the faucet directly.", + "B": "You cannot reach the faucet directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_8", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cloths did I use to pick up the soup pot just now?", + "answer": "You used two cloths to carry the soup pot.", + "question_chinese": "我刚才用了几块抹布来拿汤锅？", + "answer_chinese": "你用了两块抹布拿汤锅。", + "evidence": [ + { + "description": "Take a green cloth in your left hand and a blue cloth in your right hand to pick up the soup pot.", + "timestamps": { + "start": "00:06:35", + "end": "00:06:48" + } + } + ], + "question_timestamps": "00:06:57", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used one cloth.", + "B": "You used three cloths.", + "C": "You used four cloths to carry the soup pot, keeping your hands protected on both sides.", + "D": "You used no cloths to carry the soup pot, lifting it directly with your bare hands.", + "E": "You used two cloths to carry the soup pot." + }, + "answer_label": "E", + "answer_text": "You used two cloths to carry the soup pot.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_8", + "correct_answer": "You used two cloths to carry the soup pot.", + "original_options": { + "A": "You used one cloth.", + "B": "You used three cloths.", + "C": "You used four cloths to carry the soup pot, keeping your hands protected on both sides.", + "D": "You used no cloths to carry the soup pot, lifting it directly with your bare hands.", + "E": "You used two cloths to carry the soup pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_10", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cloths did I use to pick up the soup pot just now?", + "answer": "You used two cloths to carry the soup pot.", + "question_chinese": "我刚才用了几块抹布来拿汤锅？", + "answer_chinese": "你用了两块抹布拿汤锅。", + "evidence": [ + { + "description": "Take a green cloth in your left hand and a blue cloth in your right hand to pick up the soup pot.", + "timestamps": { + "start": "00:06:35", + "end": "00:06:48" + } + } + ], + "question_timestamps": "00:21:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used two cloths to carry the soup pot.", + "B": "You used one cloth.", + "C": "You used four cloths.", + "D": "You used five cloths to pick up the soup pot, wrapping extra fabric around the handles for grip.", + "E": "You used six cloths to grab the soup pot, stacking multiple layers to avoid any heat on your hands." + }, + "answer_label": "A", + "answer_text": "You used two cloths to carry the soup pot.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_10", + "correct_answer": "You used two cloths to carry the soup pot.", + "original_options": { + "A": "You used two cloths to carry the soup pot.", + "B": "You used one cloth.", + "C": "You used four cloths.", + "D": "You used five cloths to pick up the soup pot, wrapping extra fabric around the handles for grip.", + "E": "You used six cloths to grab the soup pot, stacking multiple layers to avoid any heat on your hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_3_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What am I doing?", + "answer": "You're cooking, and both the pot and the spatula are in front of you.", + "question_chinese": "我在干什么？", + "answer_chinese": "你在煮菜，锅和铲子都在你前面", + "evidence": [ + { + "description": "You are cooking.", + "timestamps": { + "start": "00:16:31", + "end": "00:16:35" + } + } + ], + "question_timestamps": "00:16:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You're cooking; the pot is behind you.", + "B": "You're cooking, and both the pot and the spatula are in front of you.", + "C": "You're cooking; the spatula is to your left.", + "D": "You're cooking, with the pot in front of you while the spatula sits off to your right side.", + "E": "You're cooking, and both the pot and the spatula are positioned to your left rather than in front." + }, + "answer_label": "B", + "answer_text": "You're cooking, and both the pot and the spatula are in front of you.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_3_T1", + "correct_answer": "You're cooking, and both the pot and the spatula are in front of you.", + "original_options": { + "A": "You're cooking; the pot is behind you.", + "B": "You're cooking, and both the pot and the spatula are in front of you.", + "C": "You're cooking; the spatula is to your left.", + "D": "You're cooking, with the pot in front of you while the spatula sits off to your right side.", + "E": "You're cooking, and both the pot and the spatula are positioned to your left rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 18 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_3_T2", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What am I doing?", + "answer": "You're washing the pot, and the pot is in front of you.", + "question_chinese": "我在干什么？", + "answer_chinese": "你在洗锅，锅在你前面", + "evidence": [ + { + "description": "You are washing the pot.", + "timestamps": { + "start": "00:17:58", + "end": "00:18:03" + } + } + ], + "question_timestamps": "00:18:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You're washing the pot behind you.", + "B": "You're washing the pot, and the pot is in front of you.", + "C": "You're washing the pot to your left.", + "D": "You're washing the pot, but it is positioned to your right side instead of directly in front.", + "E": "You're washing the pot, and it is not in front of you; it sits behind you near the sink area." + }, + "answer_label": "B", + "answer_text": "You're washing the pot, and the pot is in front of you.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_3_T2", + "correct_answer": "You're washing the pot, and the pot is in front of you.", + "original_options": { + "A": "You're washing the pot behind you.", + "B": "You're washing the pot, and the pot is in front of you.", + "C": "You're washing the pot to your left.", + "D": "You're washing the pot, but it is positioned to your right side instead of directly in front.", + "E": "You're washing the pot, and it is not in front of you; it sits behind you near the sink area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 20 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_1", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I wanted to know, relative to my current position, whether I had placed my phone on the left or on the right？", + "answer": "You placed your phone on your left side.", + "question_chinese": "我现在想回想一下，相对我现在的位置，我之前把手机放在了左边还是右边？", + "answer_chinese": "你把手机放在了你的左边。", + "evidence": [ + { + "description": "You are in the kitchen now, and your phone is on the counter to your left.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You placed your phone on your left side.", + "B": "You placed your phone on your right side." + }, + "answer_label": "A", + "answer_text": "You placed your phone on your left side.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_1", + "correct_answer": "You placed your phone on your left side.", + "original_options": { + "A": "You placed your phone on your left side.", + "B": "You placed your phone on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I wanted to know, relative to my current position, whether I had placed my phone on the left or on the right？", + "answer": "You placed your phone on your right side.", + "question_chinese": "我现在想回想一下，相对我现在的位置，我之前把手机放在了左边还是右边？", + "answer_chinese": "你把手机放在了你的右边。", + "evidence": [ + { + "description": "You left your phone on the bar counter earlier.", + "timestamps": { + "start": "00:41:31", + "end": "00:41:33" + } + }, + { + "description": "You are now in the living room, and the bar counter is on your right-hand side.", + "timestamps": { + "start": "00:41:44", + "end": "00:41:48" + } + } + ], + "question_timestamps": "00:41:53", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You placed your phone on your right side.", + "B": "You placed your phone on your left side." + }, + "answer_label": "A", + "answer_text": "You placed your phone on your right side.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_2", + "correct_answer": "You placed your phone on your right side.", + "original_options": { + "A": "You placed your phone on your right side.", + "B": "You placed your phone on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_5_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the place where I cook, relative to me?", + "answer": "The place where you cooked is to your left.", + "question_chinese": "煮菜的地方在我的哪里？", + "answer_chinese": "煮菜的地方在你的左边。", + "evidence": [ + { + "description": "On your left.", + "timestamps": { + "start": "00:06:35", + "end": "00:06:48" + } + } + ], + "question_timestamps": "00:06:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The place where you cooked is ahead.", + "B": "The place where you cooked is behind you.", + "C": "The place where you cooked is to your right as you face forward from your current position.", + "D": "The place where you cooked is to your left.", + "E": "The place where you cooked is above you, positioned higher than your current eye level." + }, + "answer_label": "D", + "answer_text": "The place where you cooked is to your left.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_5_T2", + "correct_answer": "The place where you cooked is to your left.", + "original_options": { + "A": "The place where you cooked is ahead.", + "B": "The place where you cooked is behind you.", + "C": "The place where you cooked is to your right as you face forward from your current position.", + "D": "The place where you cooked is to your left.", + "E": "The place where you cooked is above you, positioned higher than your current eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 15 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_5_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the place where I cook, relative to me?", + "answer": "The place where you cooked is in front of you.", + "question_chinese": "煮菜的地方在我的哪里？", + "answer_chinese": "煮菜的地方在你的前面", + "evidence": [ + { + "description": "The cooking area is in front of you.", + "timestamps": { + "start": "00:16:26", + "end": "00:16:29" + } + } + ], + "question_timestamps": "00:16:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The place where you cooked is behind you.", + "B": "The place where you cooked is in front of you.", + "C": "The place where you cooked is left.", + "D": "The place where you cooked is off to your right side, not directly in front of you.", + "E": "The place where you cooked is above your line of sight, rather than positioned in front of you." + }, + "answer_label": "B", + "answer_text": "The place where you cooked is in front of you.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_5_T1", + "correct_answer": "The place where you cooked is in front of you.", + "original_options": { + "A": "The place where you cooked is behind you.", + "B": "The place where you cooked is in front of you.", + "C": "The place where you cooked is left.", + "D": "The place where you cooked is off to your right side, not directly in front of you.", + "E": "The place where you cooked is above your line of sight, rather than positioned in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 18 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_7", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are there in the kitchen now?", + "answer": "There are a total of 5 pots.", + "question_chinese": "现在厨房一共有几口锅？", + "answer_chinese": "一共有5口锅。", + "evidence": [ + { + "description": "There are four pots on the stove.", + "timestamps": { + "start": "00:04:57", + "end": "00:05:02" + } + }, + { + "description": "There is also a pot in the sink.", + "timestamps": { + "start": "00:05:55", + "end": "00:05:57" + } + } + ], + "question_timestamps": "00:06:29", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are a total of 5 pots.", + "B": "There are a total of 4 pots.", + "C": "There are a total of 6 pots.", + "D": "There are a total of 3 pots in the kitchen now after you finish your recent cooking steps.", + "E": "There are a total of 7 pots in the kitchen now, with two more than the five you remember." + }, + "answer_label": "A", + "answer_text": "There are a total of 5 pots.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_7", + "correct_answer": "There are a total of 5 pots.", + "original_options": { + "A": "There are a total of 5 pots.", + "B": "There are a total of 4 pots.", + "C": "There are a total of 6 pots.", + "D": "There are a total of 3 pots in the kitchen now after you finish your recent cooking steps.", + "E": "There are a total of 7 pots in the kitchen now, with two more than the five you remember." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 19 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_9", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are there in the kitchen now?", + "answer": "There are a total of 7 pots.", + "question_chinese": "现在厨房一共有几口锅？", + "answer_chinese": "一共有7口锅。", + "evidence": [ + { + "description": "There are two pots in the sink.", + "timestamps": { + "start": "00:08:58", + "end": "00:09:00" + } + }, + { + "description": "There are three pots on the stove.", + "timestamps": { + "start": "00:09:06", + "end": "00:09:08" + } + }, + { + "description": "There is a pot on the ground.", + "timestamps": { + "start": "00:09:23", + "end": "00:09:24" + } + }, + { + "description": "I have a pot in my hand.", + "timestamps": { + "start": "00:09:25", + "end": "00:09:26" + } + } + ], + "question_timestamps": "00:09:28", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are a total of 6 pots.", + "B": "There are a total of 8 pots.", + "C": "There are a total of 7 pots.", + "D": "There are a total of 5 pots in the kitchen now, fewer than what you ended with previously.", + "E": "There are a total of 9 pots in the kitchen now, adding two more beyond the seven expected." + }, + "answer_label": "C", + "answer_text": "There are a total of 7 pots.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_9", + "correct_answer": "There are a total of 7 pots.", + "original_options": { + "A": "There are a total of 6 pots.", + "B": "There are a total of 8 pots.", + "C": "There are a total of 7 pots.", + "D": "There are a total of 5 pots in the kitchen now, fewer than what you ended with previously.", + "E": "There are a total of 9 pots in the kitchen now, adding two more beyond the seven expected." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_9_T2", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the cutting board relative to me?", + "answer": "It was in front of you.", + "question_chinese": "菜板在我的哪里？", + "answer_chinese": "它在你的前面。", + "evidence": [ + { + "description": "It's in front of you.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was behind you.", + "B": "It was in front of you.", + "C": "It was to your left.", + "D": "It was to your right, positioned beside you rather than directly ahead.", + "E": "It was above you, placed higher up instead of on the counter in front." + }, + "answer_label": "B", + "answer_text": "It was in front of you.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_9_T2", + "correct_answer": "It was in front of you.", + "original_options": { + "A": "It was behind you.", + "B": "It was in front of you.", + "C": "It was to your left.", + "D": "It was to your right, positioned beside you rather than directly ahead.", + "E": "It was above you, placed higher up instead of on the counter in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 12, + 14 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_9_T1", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the cutting board relative to me?", + "answer": "The cutting board was to your left.", + "question_chinese": "菜板在我的哪里？", + "answer_chinese": "菜板在你的左边。", + "evidence": [ + { + "description": "The cutting board is on your left.", + "timestamps": { + "start": "00:04:16", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:04:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The cutting board was in front of you.", + "B": "The cutting board was behind you.", + "C": "The cutting board was to your left.", + "D": "The cutting board was to your right, located on the opposite side of where you were standing.", + "E": "The cutting board was above you, positioned higher than your usual countertop work area." + }, + "answer_label": "C", + "answer_text": "The cutting board was to your left.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_9_T1", + "correct_answer": "The cutting board was to your left.", + "original_options": { + "A": "The cutting board was in front of you.", + "B": "The cutting board was behind you.", + "C": "The cutting board was to your left.", + "D": "The cutting board was to your right, located on the opposite side of where you were standing.", + "E": "The cutting board was above you, positioned higher than your usual countertop work area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 17, + 14 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_12", + "qid_num": 27, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the knife moving now?", + "answer": "The plate is being lifted from the cutting board, moving to the right, passing by the sink, and heading towards the drawer.", + "question_chinese": "这个刀现在正在往哪里移动?", + "answer_chinese": "盘子正从砧板位置被端起,向右移动,经过水槽,朝着抽屉方向移动。", + "evidence": [ + { + "description": "The plate was lifted from the cutting board, moved to the right, passed by the sink, and placed into the drawer.", + "timestamps": { + "start": "00:09:39", + "end": "00:09:42" + } + } + ], + "question_timestamps": "00:09:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is moving to the left.", + "B": "It is moving behind you.", + "C": "The plate is lifted, moving forward past the sink, heading to the drawer.", + "D": "The plate is being lifted, moving right past the drawer, then heading toward the sink.", + "E": "The plate is being lifted from the cutting board, moving to the right, passing by the sink, and heading towards the drawer." + }, + "answer_label": "E", + "answer_text": "The plate is being lifted from the cutting board, moving to the right, passing by the sink, and heading towards the drawer.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_12", + "correct_answer": "The plate is being lifted from the cutting board, moving to the right, passing by the sink, and heading towards the drawer.", + "original_options": { + "A": "It is moving to the left.", + "B": "It is moving behind you.", + "C": "The plate is lifted, moving forward past the sink, heading to the drawer.", + "D": "The plate is being lifted, moving right past the drawer, then heading toward the sink.", + "E": "The plate is being lifted from the cutting board, moving to the right, passing by the sink, and heading towards the drawer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 15 + ] + }, + { + "video_uid": "35e004f4-ad64-4835-9358-38a58b431252", + "video_length": "00:43:50", + "qid": "35e004f4-ad64-4835-9358-38a58b431252_11", + "qid_num": 28, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the knife moving now?", + "answer": "The plate is being lifted from the cutting board, moving to the right, passing by the sink, and heading towards the drawer.", + "question_chinese": "这个刀现在正在往哪里移动?", + "answer_chinese": "盘子正从砧板位置被端起,向右移动,经过水槽,朝着抽屉方向移动。", + "evidence": [ + { + "description": "The plate was lifted from the cutting board, moved to the right, passed by the sink, and placed into the drawer.", + "timestamps": { + "start": "00:09:37", + "end": "00:09:42" + } + } + ], + "question_timestamps": "00:09:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The plate is being lifted from the cutting board, moving to the right, passing by the sink, and heading towards the drawer.", + "B": "It is moving to the left.", + "C": "It is moving in front of you.", + "D": "The plate is being lifted from the cutting board, moving behind you, passing by the sink, and heading towards the drawer.", + "E": "The plate is being lifted from the cutting board, moving to the right, passing by the drawer, and heading towards the sink." + }, + "answer_label": "A", + "answer_text": "The plate is being lifted from the cutting board, moving to the right, passing by the sink, and heading towards the drawer.", + "q_id": "35e004f4-ad64-4835-9358-38a58b431252_11", + "correct_answer": "The plate is being lifted from the cutting board, moving to the right, passing by the sink, and heading towards the drawer.", + "original_options": { + "A": "The plate is being lifted from the cutting board, moving to the right, passing by the sink, and heading towards the drawer.", + "B": "It is moving to the left.", + "C": "It is moving in front of you.", + "D": "The plate is being lifted from the cutting board, moving behind you, passing by the sink, and heading towards the drawer.", + "E": "The plate is being lifted from the cutting board, moving to the right, passing by the drawer, and heading towards the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 22 + ] + } +] diff --git a/QAs/39d0559a-d1e8-4fca-b7f4-db87bbbbde7e-mcq.json b/QAs/39d0559a-d1e8-4fca-b7f4-db87bbbbde7e-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2667e37f978bb2be5a870da1f94de4ded78b1575 --- /dev/null +++ b/QAs/39d0559a-d1e8-4fca-b7f4-db87bbbbde7e-mcq.json @@ -0,0 +1,1636 @@ +[ + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_2_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The fridge is in front of you.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱在你的前方。", + "evidence": [ + { + "description": "The refrigerator is in front of you.", + "timestamps": { + "start": "00:06:23", + "end": "00:06:24" + } + } + ], + "question_timestamps": "00:06:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is behind you.", + "B": "The fridge is to your left.", + "C": "The fridge is in front of you.", + "D": "The fridge is to your right as you face forward, not directly in front of you.", + "E": "The fridge is off to your left side, slightly behind your shoulder rather than ahead." + }, + "answer_label": "C", + "answer_text": "The fridge is in front of you.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_2_T1", + "correct_answer": "The fridge is in front of you.", + "original_options": { + "A": "The fridge is behind you.", + "B": "The fridge is to your left.", + "C": "The fridge is in front of you.", + "D": "The fridge is to your right as you face forward, not directly in front of you.", + "E": "The fridge is off to your left side, slightly behind your shoulder rather than ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator now relative to me?", + "answer": "The refrigerator is in front of you.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱在你的前方。", + "evidence": [ + { + "description": "The refrigerator is in front of you.", + "timestamps": { + "start": "00:21:19", + "end": "00:21:20" + } + } + ], + "question_timestamps": "00:21:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is behind you.", + "B": "The refrigerator is to your right.", + "C": "The refrigerator is slightly to your left, not positioned directly in front of you.", + "D": "The refrigerator is in front of you.", + "E": "The refrigerator is off to your right-hand side and a bit behind you rather than straight ahead." + }, + "answer_label": "D", + "answer_text": "The refrigerator is in front of you.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_4", + "correct_answer": "The refrigerator is in front of you.", + "original_options": { + "A": "The refrigerator is behind you.", + "B": "The refrigerator is to your right.", + "C": "The refrigerator is slightly to your left, not positioned directly in front of you.", + "D": "The refrigerator is in front of you.", + "E": "The refrigerator is off to your right-hand side and a bit behind you rather than straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 17 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_2_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The fridge is to your back-left.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱在你的左后方。", + "evidence": [ + { + "description": "The refrigerator is in front of you.", + "timestamps": { + "start": "00:21:19", + "end": "00:21:20" + } + }, + { + "description": "You walked to the right of the fridge.", + "timestamps": { + "start": "00:54:03", + "end": "00:54:12" + } + } + ], + "question_timestamps": "00:54:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is behind you.", + "B": "The fridge is to your front-left.", + "C": "The fridge is to your back-right, not behind your left shoulder where you would expect it.", + "D": "The fridge is to your back-left.", + "E": "The fridge sits in front of you and slightly left, rather than back-left relative to your stance." + }, + "answer_label": "D", + "answer_text": "The fridge is to your back-left.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_2_T2", + "correct_answer": "The fridge is to your back-left.", + "original_options": { + "A": "The fridge is behind you.", + "B": "The fridge is to your front-left.", + "C": "The fridge is to your back-right, not behind your left shoulder where you would expect it.", + "D": "The fridge is to your back-left.", + "E": "The fridge sits in front of you and slightly left, rather than back-left relative to your stance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_19", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Do I need to turn if I walk from the stove to the store entrance?", + "answer": "You need to turn.", + "question_chinese": "如果我从灶台前走到店门口需要拐弯吗？", + "answer_chinese": "你需要拐弯。", + "evidence": [ + { + "description": "You made a turn.", + "timestamps": { + "start": "00:53:37", + "end": "00:54:13" + } + } + ], + "question_timestamps": "00:54:31", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You need to turn.", + "B": "You don’t need to turn." + }, + "answer_label": "A", + "answer_text": "You need to turn.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_19", + "correct_answer": "You need to turn.", + "original_options": { + "A": "You need to turn.", + "B": "You don’t need to turn." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_20", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Do I need to turn if I walk from the stove to the store entrance?", + "answer": "You need to turn.", + "question_chinese": "如果我从灶台前走到店门口需要拐弯吗？", + "answer_chinese": "你需要拐弯。", + "evidence": [ + { + "description": "You made a turn.", + "timestamps": { + "start": "00:53:37", + "end": "00:54:13" + } + } + ], + "question_timestamps": "01:00:33", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You need to turn.", + "B": "You don’t need to turn." + }, + "answer_label": "A", + "answer_text": "You need to turn.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_20", + "correct_answer": "You need to turn.", + "original_options": { + "A": "You need to turn.", + "B": "You don’t need to turn." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_10_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to walk to the shop entrance now, how should I go?", + "answer": "Turn around, take one step to the left, then take one step forward. After that, turn right and walk straight to reach the shop entrance.", + "question_chinese": "如果我现在要走到店门口需要怎么走？", + "answer_chinese": "你需要向后转，然后向左迈一步，接着向前走一步，右转直走即可到达店门口。", + "evidence": [ + { + "description": "You made a turn.", + "timestamps": { + "start": "00:53:37", + "end": "00:53:42" + } + } + ], + "question_timestamps": "00:53:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Walk straight ahead without turning.", + "B": "Turn around, take one step to the left, then take one step forward. After that, turn right and walk straight to reach the shop entrance." + }, + "answer_label": "B", + "answer_text": "Turn around, take one step to the left, then take one step forward. After that, turn right and walk straight to reach the shop entrance.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_10_T1", + "correct_answer": "Turn around, take one step to the left, then take one step forward. After that, turn right and walk straight to reach the shop entrance.", + "original_options": { + "A": "Walk straight ahead without turning.", + "B": "Turn around, take one step to the left, then take one step forward. After that, turn right and walk straight to reach the shop entrance." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_10_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to walk to the shop entrance now, how should I go?", + "answer": "Turn left, then walk straight to reach the shop entrance.", + "question_chinese": "如果我现在要走到店门口需要怎么走？", + "answer_chinese": "你需要左转，接着直走即可到达店门口。", + "evidence": [ + { + "description": "You made a turn.", + "timestamps": { + "start": "00:53:37", + "end": "00:54:13" + } + } + ], + "question_timestamps": "01:00:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Turn left, then walk straight to reach the shop entrance.", + "B": "Turn around, take one step to the left, then take one step forward. After that, turn right and walk straight to reach the shop entrance." + }, + "answer_label": "A", + "answer_text": "Turn left, then walk straight to reach the shop entrance.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_10_T2", + "correct_answer": "Turn left, then walk straight to reach the shop entrance.", + "original_options": { + "A": "Turn left, then walk straight to reach the shop entrance.", + "B": "Turn around, take one step to the left, then take one step forward. After that, turn right and walk straight to reach the shop entrance." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 25 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_4_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I take scallions from the bowl just now?", + "answer": "You just took scallions once. The bowl is to your right.", + "question_chinese": "我刚刚从碗里拿了几次葱？", + "answer_chinese": "你刚刚拿了一次葱。碗在你的右边", + "evidence": [ + { + "description": "You took the first handful of green onions from the bowl.", + "timestamps": { + "start": "00:10:14", + "end": "00:10:15" + } + } + ], + "question_timestamps": "00:10:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You just took scallions twice. The bowl is to your right.", + "B": "You just took scallions once. The bowl is to your right." + }, + "answer_label": "B", + "answer_text": "You just took scallions once. The bowl is to your right.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_4_T2", + "correct_answer": "You just took scallions once. The bowl is to your right.", + "original_options": { + "A": "You just took scallions twice. The bowl is to your right.", + "B": "You just took scallions once. The bowl is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_4_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I take scallions from the bowl just now?", + "answer": "So far, you’ve taken scallions from the bowl twice in total. The bowl is to your right.", + "question_chinese": "我刚刚从碗里拿了几次葱？", + "answer_chinese": "到目前为止，你一共从碗里拿了两次葱。碗在你的右边", + "evidence": [ + { + "description": "You took the first handful of green onions from the bowl.", + "timestamps": { + "start": "00:10:14", + "end": "00:10:15" + } + }, + { + "description": "You took the second handful of green onions from the bowl.", + "timestamps": { + "start": "00:10:15", + "end": "00:10:18" + } + } + ], + "question_timestamps": "00:10:25", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "So far, you’ve taken scallions from the bowl twice in total. The bowl is to your right.", + "B": "You just took scallions once. The bowl is to your right." + }, + "answer_label": "A", + "answer_text": "So far, you’ve taken scallions from the bowl twice in total. The bowl is to your right.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_4_T1", + "correct_answer": "So far, you’ve taken scallions from the bowl twice in total. The bowl is to your right.", + "original_options": { + "A": "So far, you’ve taken scallions from the bowl twice in total. The bowl is to your right.", + "B": "You just took scallions once. The bowl is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_15", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls of white rice did I just scoop for the guests?", + "answer": "You just scooped up a bowl of rice.", + "question_chinese": "我刚刚勺了几碗白米饭给客人？", + "answer_chinese": "你刚刚勺了一碗饭。", + "evidence": [ + { + "description": "You just scooped up a bowl of rice.", + "timestamps": { + "start": "00:39:40", + "end": "00:39:45" + } + } + ], + "question_timestamps": "00:39:49", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You scooped two bowls of rice.", + "B": "You scooped three bowls of rice.", + "C": "You just scooped up a bowl of rice.", + "D": "You just scooped up four bowls of white rice to serve the guests.", + "E": "You scooped two bowls of white rice and set them out for the guests." + }, + "answer_label": "C", + "answer_text": "You just scooped up a bowl of rice.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_15", + "correct_answer": "You just scooped up a bowl of rice.", + "original_options": { + "A": "You scooped two bowls of rice.", + "B": "You scooped three bowls of rice.", + "C": "You just scooped up a bowl of rice.", + "D": "You just scooped up four bowls of white rice to serve the guests.", + "E": "You scooped two bowls of white rice and set them out for the guests." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 14 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_16", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls of white rice did I just scoop for the guests?", + "answer": "You just scooped up a bowl of rice.", + "question_chinese": "我刚刚勺了几碗白米饭给客人？", + "answer_chinese": "你刚刚勺了一碗饭。", + "evidence": [ + { + "description": "You just scooped up a bowl of rice.", + "timestamps": { + "start": "00:39:40", + "end": "00:39:45" + } + } + ], + "question_timestamps": "00:49:59", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You scooped two bowls of rice.", + "B": "You scooped four bowls of rice.", + "C": "You just scooped up three bowls of white rice for the guests to eat.", + "D": "You scooped two bowls of white rice and handed them out to the guests.", + "E": "You just scooped up a bowl of rice." + }, + "answer_label": "E", + "answer_text": "You just scooped up a bowl of rice.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_16", + "correct_answer": "You just scooped up a bowl of rice.", + "original_options": { + "A": "You scooped two bowls of rice.", + "B": "You scooped four bowls of rice.", + "C": "You just scooped up three bowls of white rice for the guests to eat.", + "D": "You scooped two bowls of white rice and handed them out to the guests.", + "E": "You just scooped up a bowl of rice." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_7", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there any left in the bowl after I just took the scallions?", + "answer": "There's still something left in the bowl after you just took the scallions.", + "question_chinese": "我刚刚拿了葱后碗里还有剩吗？", + "answer_chinese": "你刚刚拿了葱后碗里还有剩下的。", + "evidence": [ + { + "description": "There's still something left in the bowl after you just took the scallions.", + "timestamps": { + "start": "00:10:14", + "end": "00:10:18" + } + } + ], + "question_timestamps": "00:10:37", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, there isn't anything left in the bowl after you just took the scallions.", + "B": "There's still something left in the bowl after you just took the scallions." + }, + "answer_label": "B", + "answer_text": "There's still something left in the bowl after you just took the scallions.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_7", + "correct_answer": "There's still something left in the bowl after you just took the scallions.", + "original_options": { + "A": "No, there isn't anything left in the bowl after you just took the scallions.", + "B": "There's still something left in the bowl after you just took the scallions." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_8", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there any left in the bowl after I just took the scallions?", + "answer": "There's still something left in the bowl after you just took the scallions.", + "question_chinese": "我刚刚拿了葱后碗里还有剩吗？", + "answer_chinese": "你刚刚拿了葱后碗里还有剩余的葱。", + "evidence": [ + { + "description": "There's still something left in the bowl after you just took the scallions.", + "timestamps": { + "start": "00:16:14", + "end": "00:16:18" + } + } + ], + "question_timestamps": "00:17:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "There's still something left in the bowl after you just took the scallions.", + "B": "No, there isn't anything left in the bowl after you just took the scallions." + }, + "answer_label": "A", + "answer_text": "There's still something left in the bowl after you just took the scallions.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_8", + "correct_answer": "There's still something left in the bowl after you just took the scallions.", + "original_options": { + "A": "There's still something left in the bowl after you just took the scallions.", + "B": "No, there isn't anything left in the bowl after you just took the scallions." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_9", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just give to the little boy at the window?", + "answer": "You just gave him Onions.", + "question_chinese": "我刚刚给了在窗口的小男孩什么东西？", + "answer_chinese": "你刚刚给了他洋葱。", + "evidence": [ + { + "description": "You gave the onion.", + "timestamps": { + "start": "00:13:59", + "end": "00:14:01" + } + } + ], + "question_timestamps": "00:14:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just gave him apples.", + "B": "You just gave him carrots.", + "C": "You just handed the little boy at the window a couple of oranges.", + "D": "You just gave him Onions.", + "E": "You gave the little boy at the window some lemons just now." + }, + "answer_label": "D", + "answer_text": "You just gave him Onions.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_9", + "correct_answer": "You just gave him Onions.", + "original_options": { + "A": "You just gave him apples.", + "B": "You just gave him carrots.", + "C": "You just handed the little boy at the window a couple of oranges.", + "D": "You just gave him Onions.", + "E": "You gave the little boy at the window some lemons just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 12 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_10", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just give to the little boy at the window?", + "answer": "You gave him Onions.", + "question_chinese": "我刚刚给了在窗口的小男孩什么东西？", + "answer_chinese": "你给了他洋葱。", + "evidence": [ + { + "description": "You gave the onion.", + "timestamps": { + "start": "00:13:59", + "end": "00:14:01" + } + } + ], + "question_timestamps": "00:27:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You gave him Onions.", + "B": "You gave him carrots.", + "C": "You gave him apples.", + "D": "You gave the little boy at the window some grapes just now.", + "E": "You handed the little boy at the window a few tomatoes just now." + }, + "answer_label": "A", + "answer_text": "You gave him Onions.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_10", + "correct_answer": "You gave him Onions.", + "original_options": { + "A": "You gave him Onions.", + "B": "You gave him carrots.", + "C": "You gave him apples.", + "D": "You gave the little boy at the window some grapes just now.", + "E": "You handed the little boy at the window a few tomatoes just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls of fried rice did I just fill?", + "answer": "You just filled a bowl of fried rice.", + "question_chinese": "我刚刚装了几碗炒饭？", + "answer_chinese": "你刚刚装了1碗炒饭。", + "evidence": [ + { + "description": "You have packed the first bowl of fried rice.", + "timestamps": { + "start": "00:02:51", + "end": "00:02:58" + } + } + ], + "question_timestamps": "00:03:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You filled two bowls of fried rice.", + "B": "You just filled a bowl of fried rice.", + "C": "You filled three bowls of fried rice.", + "D": "You just filled four bowls of fried rice and set them out to serve.", + "E": "You filled two bowls of fried rice and placed them on the counter." + }, + "answer_label": "B", + "answer_text": "You just filled a bowl of fried rice.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_1", + "correct_answer": "You just filled a bowl of fried rice.", + "original_options": { + "A": "You filled two bowls of fried rice.", + "B": "You just filled a bowl of fried rice.", + "C": "You filled three bowls of fried rice.", + "D": "You just filled four bowls of fried rice and set them out to serve.", + "E": "You filled two bowls of fried rice and placed them on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 13 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls of fried rice did I just fill?", + "answer": "You just filled two bowls of fried rice.", + "question_chinese": "我刚刚装了几碗炒饭？", + "answer_chinese": "你刚刚装了两碗炒饭。", + "evidence": [ + { + "description": "You have packed the first bowl of fried rice.", + "timestamps": { + "start": "00:02:51", + "end": "00:02:58" + } + }, + { + "description": "You have packed the second bowl of fried rice.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:53" + } + }, + { + "description": "You have packed the third bowl of fried rice.", + "timestamps": { + "start": "00:08:19", + "end": "00:08:26" + } + }, + { + "description": "You have packed the fourth bowl of fried rice.", + "timestamps": { + "start": "00:08:33", + "end": "00:08:37" + } + } + ], + "question_timestamps": "00:12:29", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You just filled one bowl of fried rice.", + "B": "You just filled three bowls of fried rice.", + "C": "You just filled four bowls of fried rice, not two, before setting them aside.", + "D": "You just filled two bowls of fried rice.", + "E": "You just filled five bowls of fried rice in total, finishing the last one moments ago." + }, + "answer_label": "D", + "answer_text": "You just filled two bowls of fried rice.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_2", + "correct_answer": "You just filled two bowls of fried rice.", + "original_options": { + "A": "You just filled one bowl of fried rice.", + "B": "You just filled three bowls of fried rice.", + "C": "You just filled four bowls of fried rice, not two, before setting them aside.", + "D": "You just filled two bowls of fried rice.", + "E": "You just filled five bowls of fried rice in total, finishing the last one moments ago." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 14, + 16 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_8_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls of white rice did I scoop for the guest?", + "answer": "You just scooped one bowl of rice.", + "question_chinese": "我勺了几碗白米饭给客人？", + "answer_chinese": "你刚刚勺了一碗饭。", + "evidence": [ + { + "description": "You just scooped up a bowl of rice.", + "timestamps": { + "start": "00:39:40", + "end": "00:39:45" + } + } + ], + "question_timestamps": "00:39:49", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just scooped zero bowls of rice.", + "B": "You just scooped one bowl of rice.", + "C": "You just scooped two bowls of rice.", + "D": "You just scooped three bowls of rice for the guest, not just one.", + "E": "You just scooped four bowls of rice total, finishing after the last scoop." + }, + "answer_label": "B", + "answer_text": "You just scooped one bowl of rice.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_8_T1", + "correct_answer": "You just scooped one bowl of rice.", + "original_options": { + "A": "You just scooped zero bowls of rice.", + "B": "You just scooped one bowl of rice.", + "C": "You just scooped two bowls of rice.", + "D": "You just scooped three bowls of rice for the guest, not just one.", + "E": "You just scooped four bowls of rice total, finishing after the last scoop." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 13 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_8_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls of white rice did I scoop for the guest?", + "answer": "You only just scooped one bowl of rice. So far, you haven’t scooped another bowl for the guest.", + "question_chinese": "我勺了几碗白米饭给客人？", + "answer_chinese": "你只有刚刚勺了一碗饭，到目前为止，你没有再勺碗饭给客人。", + "evidence": [ + { + "description": "You just scooped up a bowl of rice.", + "timestamps": { + "start": "00:39:40", + "end": "00:39:45" + } + } + ], + "question_timestamps": "00:49:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You only just scooped zero bowls of rice.", + "B": "You only just scooped two bowls of rice.", + "C": "You only just scooped one bowl of rice. So far, you haven’t scooped another bowl for the guest.", + "D": "You only just scooped three bowls of rice, and you haven’t scooped another bowl for the guest.", + "E": "You only just scooped four bowls of rice so far, and you haven’t scooped another bowl for the guest." + }, + "answer_label": "C", + "answer_text": "You only just scooped one bowl of rice. So far, you haven’t scooped another bowl for the guest.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_8_T2", + "correct_answer": "You only just scooped one bowl of rice. So far, you haven’t scooped another bowl for the guest.", + "original_options": { + "A": "You only just scooped zero bowls of rice.", + "B": "You only just scooped two bowls of rice.", + "C": "You only just scooped one bowl of rice. So far, you haven’t scooped another bowl for the guest.", + "D": "You only just scooped three bowls of rice, and you haven’t scooped another bowl for the guest.", + "E": "You only just scooped four bowls of rice so far, and you haven’t scooped another bowl for the guest." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 20 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_7_T1", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the red stool?", + "answer": "You can’t touch it directly.", + "question_chinese": "我可以直接触碰到红色凳子吗？", + "answer_chinese": "你无法直接触碰到。", + "evidence": [ + { + "description": "The red stool is some distance away from you.", + "timestamps": { + "start": "00:27:20", + "end": "00:27:25" + } + } + ], + "question_timestamps": "00:28:39", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can touch it directly—the red stool is to your right.", + "B": "You can’t touch it directly." + }, + "answer_label": "B", + "answer_text": "You can’t touch it directly.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_7_T1", + "correct_answer": "You can’t touch it directly.", + "original_options": { + "A": "You can touch it directly—the red stool is to your right.", + "B": "You can’t touch it directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_14", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the red stool directly?", + "answer": "You can't touch it directly.", + "question_chinese": "我可以直接触碰到红色凳子吗？", + "answer_chinese": "你无法直接触碰到。", + "evidence": [ + { + "description": "The red stool is some distance away from you.", + "timestamps": { + "start": "00:44:46", + "end": "00:45:14" + } + } + ], + "question_timestamps": "00:45:17", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch it directly.", + "B": "You can't touch it directly." + }, + "answer_label": "B", + "answer_text": "You can't touch it directly.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_14", + "correct_answer": "You can't touch it directly.", + "original_options": { + "A": "You can touch it directly.", + "B": "You can't touch it directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_7_T2", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the red stool?", + "answer": "You can touch it directly—the red stool is to your right.", + "question_chinese": "我可以直接触碰到红色凳子吗？", + "answer_chinese": "你可以直接触碰到，红色凳子在你右边。", + "evidence": [ + { + "description": "The red stool is some distance away from you.", + "timestamps": { + "start": "00:44:46", + "end": "00:45:14" + } + }, + { + "description": "You walked near the red stool.", + "timestamps": { + "start": "00:53:54", + "end": "00:54:10" + } + } + ], + "question_timestamps": "00:54:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can touch it directly—the red stool is to your right.", + "B": "You can’t touch it directly—the red stool is too far away." + }, + "answer_label": "A", + "answer_text": "You can touch it directly—the red stool is to your right.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_7_T2", + "correct_answer": "You can touch it directly—the red stool is to your right.", + "original_options": { + "A": "You can touch it directly—the red stool is to your right.", + "B": "You can’t touch it directly—the red stool is too far away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_3_T1", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I throw the trash directly into the blue trash can by the doorway right now?", + "answer": "You can throw the trash directly into the trash can right now, and you just threw away eggshells.", + "question_chinese": "我现在可以直接将垃圾扔到到门口的蓝色垃圾桶吗？", + "answer_chinese": "你现在可以直接将垃圾扔到垃圾桶中，并且你刚才扔了鸡蛋壳。", + "evidence": [ + { + "description": "You are separated from the trash can by a table.", + "timestamps": { + "start": "00:08:21", + "end": "00:08:55" + } + } + ], + "question_timestamps": "00:08:57", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, you can’t throw the trash into the blue trash can by the doorway.", + "B": "You can throw the trash directly into the trash can right now, and you just threw away eggshells." + }, + "answer_label": "B", + "answer_text": "You can throw the trash directly into the trash can right now, and you just threw away eggshells.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_3_T1", + "correct_answer": "You can throw the trash directly into the trash can right now, and you just threw away eggshells.", + "original_options": { + "A": "Right now, you can’t throw the trash into the blue trash can by the doorway.", + "B": "You can throw the trash directly into the trash can right now, and you just threw away eggshells." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_3_T2", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I throw the trash directly into the blue trash can by the doorway right now?", + "answer": "Right now, you can’t throw the trash into the trash can by the doorway.", + "question_chinese": "我现在可以直接将垃圾扔到到门口的蓝色垃圾桶吗？", + "answer_chinese": "你现在无法将垃圾扔到门口垃圾桶。", + "evidence": [ + { + "description": "You are separated from the trash can by a table.", + "timestamps": { + "start": "00:21:47", + "end": "00:21:50" + } + }, + { + "description": "You took two steps to the left.", + "timestamps": { + "start": "00:21:50", + "end": "00:21:59" + } + } + ], + "question_timestamps": "00:22:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, you can’t throw the trash into the trash can by the doorway.", + "B": "You can throw the trash directly into the blue trash can by the doorway right now." + }, + "answer_label": "A", + "answer_text": "Right now, you can’t throw the trash into the trash can by the doorway.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_3_T2", + "correct_answer": "Right now, you can’t throw the trash into the trash can by the doorway.", + "original_options": { + "A": "Right now, you can’t throw the trash into the trash can by the doorway.", + "B": "You can throw the trash directly into the blue trash can by the doorway right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_17", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator directly now?", + "answer": "You can now make direct contact.", + "question_chinese": "我现在可以直接触碰到冰箱吗？", + "answer_chinese": "你现在可以直接触碰到。", + "evidence": [ + { + "description": "The refrigerator is beside you.", + "timestamps": { + "start": "00:54:03", + "end": "00:54:29" + } + } + ], + "question_timestamps": "00:54:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now make direct contact.", + "B": "You cannot now make direct contact." + }, + "answer_label": "A", + "answer_text": "You can now make direct contact.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_17", + "correct_answer": "You can now make direct contact.", + "original_options": { + "A": "You can now make direct contact.", + "B": "You cannot now make direct contact." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_9_T1", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator directly right now?", + "answer": "Right now, you can touch the fridge directly.", + "question_chinese": "我现在可以直接触碰到冰箱吗？", + "answer_chinese": "你现在可以直接触碰到冰箱。", + "evidence": [ + { + "description": "The refrigerator is beside you.", + "timestamps": { + "start": "00:54:03", + "end": "00:54:29" + } + } + ], + "question_timestamps": "00:54:31", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, you can’t touch the fridge directly.", + "B": "Right now, you can touch the fridge directly." + }, + "answer_label": "B", + "answer_text": "Right now, you can touch the fridge directly.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_9_T1", + "correct_answer": "Right now, you can touch the fridge directly.", + "original_options": { + "A": "Right now, you can’t touch the fridge directly.", + "B": "Right now, you can touch the fridge directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_18", + "qid_num": 27, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator directly now?", + "answer": "You can now make direct contact.", + "question_chinese": "我现在可以直接触碰到冰箱吗？", + "answer_chinese": "你现在可以直接触碰到。", + "evidence": [ + { + "description": "The refrigerator is beside you.", + "timestamps": { + "start": "00:54:03", + "end": "00:54:29" + } + } + ], + "question_timestamps": "00:58:42", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot now make direct contact.", + "B": "You can now make direct contact." + }, + "answer_label": "B", + "answer_text": "You can now make direct contact.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_18", + "correct_answer": "You can now make direct contact.", + "original_options": { + "A": "You cannot now make direct contact.", + "B": "You can now make direct contact." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_9_T2", + "qid_num": 28, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator directly right now?", + "answer": "Right now, you can’t touch the fridge directly.", + "question_chinese": "我现在可以直接触碰到冰箱吗？", + "answer_chinese": "你现在不可以直接触碰到冰箱。", + "evidence": [ + { + "description": "The refrigerator is beside you.", + "timestamps": { + "start": "00:54:03", + "end": "00:54:29" + } + }, + { + "description": "You stepped back two steps.", + "timestamps": { + "start": "00:59:55", + "end": "01:00:45" + } + } + ], + "question_timestamps": "01:00:47", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, you can’t touch the fridge directly.", + "B": "Right now, you can touch the fridge directly." + }, + "answer_label": "A", + "answer_text": "Right now, you can’t touch the fridge directly.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_9_T2", + "correct_answer": "Right now, you can’t touch the fridge directly.", + "original_options": { + "A": "Right now, you can’t touch the fridge directly.", + "B": "Right now, you can touch the fridge directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_5", + "qid_num": 29, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the blue trash can at the door now?", + "answer": "You can't make direct contact now.", + "question_chinese": "我现在可以直接触碰到门口的蓝色垃圾桶吗？", + "answer_chinese": "你现在无法直接触碰到。", + "evidence": [ + { + "description": "You are separated from the trash can by a table.", + "timestamps": { + "start": "00:08:21", + "end": "00:08:25" + } + } + ], + "question_timestamps": "00:08:37", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can make direct contact now.", + "B": "You can't make direct contact now." + }, + "answer_label": "B", + "answer_text": "You can't make direct contact now.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_5", + "correct_answer": "You can't make direct contact now.", + "original_options": { + "A": "You can make direct contact now.", + "B": "You can't make direct contact now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_6", + "qid_num": 30, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the blue trash can at the door now?", + "answer": "You can't make direct contact now.", + "question_chinese": "我现在可以直接触碰到门口的蓝色垃圾桶吗？", + "answer_chinese": "你现在无法直接触碰到。", + "evidence": [ + { + "description": "You are separated from the trash can by a table.", + "timestamps": { + "start": "00:21:47", + "end": "00:21:50" + } + } + ], + "question_timestamps": "00:21:56", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't make direct contact now.", + "B": "You can make direct contact now." + }, + "answer_label": "A", + "answer_text": "You can't make direct contact now.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_6", + "correct_answer": "You can't make direct contact now.", + "original_options": { + "A": "You can't make direct contact now.", + "B": "You can make direct contact now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_6_T1", + "qid_num": 31, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the window relative to me right now?", + "answer": "The window is in front of you.", + "question_chinese": "窗口现在相对我是什么位置？", + "answer_chinese": "窗口在你的前方。", + "evidence": [ + { + "description": "When you face the stove, the window is behind you.", + "timestamps": { + "start": "00:13:55", + "end": "00:14:00" + } + } + ], + "question_timestamps": "00:14:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is behind you.", + "C": "The window is to your left.", + "D": "The window is to your right at the moment, not directly in front of your current viewpoint.", + "E": "The window is above you right now, positioned higher than your eye level as you stand." + }, + "answer_label": "A", + "answer_text": "The window is in front of you.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_6_T1", + "correct_answer": "The window is in front of you.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is behind you.", + "C": "The window is to your left.", + "D": "The window is to your right at the moment, not directly in front of your current viewpoint.", + "E": "The window is above you right now, positioned higher than your eye level as you stand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 16 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_11", + "qid_num": 32, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the window now relative to me?", + "answer": "The window is behind you.", + "question_chinese": "窗口现在相对我是什么位置？", + "answer_chinese": "窗口在你的后方。", + "evidence": [ + { + "description": "The window is behind you.", + "timestamps": { + "start": "00:14:00", + "end": "00:14:03" + } + } + ], + "question_timestamps": "00:22:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is to your right as you face forward now, not behind your current position.", + "D": "The window is behind you.", + "E": "The window is above you right now, positioned higher than your current eye level as you look." + }, + "answer_label": "D", + "answer_text": "The window is behind you.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_11", + "correct_answer": "The window is behind you.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is to your right as you face forward now, not behind your current position.", + "D": "The window is behind you.", + "E": "The window is above you right now, positioned higher than your current eye level as you look." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 17 + ] + }, + { + "video_uid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e", + "video_length": "01:01:16", + "qid": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_6_T2", + "qid_num": 33, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the window relative to me right now?", + "answer": "The window is behind you.", + "question_chinese": "窗口现在相对我是什么位置？", + "answer_chinese": "窗口在你的后方。", + "evidence": [ + { + "description": "The window is behind you.", + "timestamps": { + "start": "00:23:48", + "end": "00:23:53" + } + } + ], + "question_timestamps": "00:32:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is behind you.", + "C": "The window is to your left.", + "D": "The window is slightly to your right, just off your shoulder as you face forward.", + "E": "The window is overhead and in front of you, above your line of sight right now." + }, + "answer_label": "B", + "answer_text": "The window is behind you.", + "q_id": "39d0559a-d1e8-4fca-b7f4-db87bbbbde7e_6_T2", + "correct_answer": "The window is behind you.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is behind you.", + "C": "The window is to your left.", + "D": "The window is slightly to your right, just off your shoulder as you face forward.", + "E": "The window is overhead and in front of you, above your line of sight right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 16 + ] + } +] diff --git a/QAs/3a8774b9-ea4d-439e-a862-b29b29df999d-mcq.json b/QAs/3a8774b9-ea4d-439e-a862-b29b29df999d-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..da1bd2ee5a51208c671f415554dda3d5df9651ee --- /dev/null +++ b/QAs/3a8774b9-ea4d-439e-a862-b29b29df999d-mcq.json @@ -0,0 +1,1101 @@ +[ + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_11", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am standing, can I reach the cell phone?", + "answer": "Yes, you can reach it by extending your left hand to your left side.", + "question_chinese": "在我所站的位置，我伸手能拿到手机吗？", + "answer_chinese": "可以，向你身体左侧伸出你的左手就可以拿到。", + "evidence": [ + { + "description": "The cell phone is picked up from the left side of the sink and placed on the right side of the sink.", + "timestamps": { + "start": "00:38:12", + "end": "00:38:25" + } + } + ], + "question_timestamps": "00:45:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you can’t reach it; it’s out of arm’s reach on your left side.", + "B": "Yes, you can reach it by extending your left hand to your left side." + }, + "answer_label": "B", + "answer_text": "Yes, you can reach it by extending your left hand to your left side.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_11", + "correct_answer": "Yes, you can reach it by extending your left hand to your left side.", + "original_options": { + "A": "No, you can’t reach it; it’s out of arm’s reach on your left side.", + "B": "Yes, you can reach it by extending your left hand to your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_12", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am standing, can I reach the cell phone?", + "answer": "No, the cell phone is two small steps away from you on your left.", + "question_chinese": "在我所站的位置，我伸手能拿到手机吗？", + "answer_chinese": "不可以，手机在你左侧离你有两小步远。", + "evidence": [ + { + "description": "The cell phone is picked up from the left side of the sink and placed on the right side of the sink.", + "timestamps": { + "start": "00:38:12", + "end": "00:38:25" + } + }, + { + "description": "The cell phone is at a distance of two small steps to the left.", + "timestamps": { + "start": "00:58:07", + "end": "00:58:12" + } + } + ], + "question_timestamps": "01:04:45", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you can reach it; it’s right next to you on your left.", + "B": "No, the cell phone is two small steps away from you on your left." + }, + "answer_label": "B", + "answer_text": "No, the cell phone is two small steps away from you on your left.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_12", + "correct_answer": "No, the cell phone is two small steps away from you on your left.", + "original_options": { + "A": "Yes, you can reach it; it’s right next to you on your left.", + "B": "No, the cell phone is two small steps away from you on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_13", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am standing, can I reach the cell phone?", + "answer": "Yes, the cell phone is within reach of your right hand.", + "question_chinese": "在我所站的位置，我伸手能拿到手机吗？", + "answer_chinese": "可以，手机就在你的右手可触及范围内。", + "evidence": [ + { + "description": "The cell phone is picked up from the left side of the sink and placed on the right side of the sink.", + "timestamps": { + "start": "00:38:12", + "end": "00:38:25" + } + }, + { + "description": "The cell phone is at a distance of two small steps to the left.", + "timestamps": { + "start": "00:58:07", + "end": "00:58:12" + } + }, + { + "description": "Walk to the sink location with the trash can.", + "timestamps": { + "start": "01:05:00", + "end": "01:05:05" + } + } + ], + "question_timestamps": "01:05:08", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "No, you can’t reach it with your right hand from where you’re standing.", + "B": "Yes, the cell phone is within reach of your right hand." + }, + "answer_label": "B", + "answer_text": "Yes, the cell phone is within reach of your right hand.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_13", + "correct_answer": "Yes, the cell phone is within reach of your right hand.", + "original_options": { + "A": "No, you can’t reach it with your right hand from where you’re standing.", + "B": "Yes, the cell phone is within reach of your right hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_20", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get more white cups, how should I walk?", + "answer": "Walk two steps to your back-left, the cups are in the cabinet above your head.", + "question_chinese": "如果我要拿更多的白色杯子，我该怎么走？", + "answer_chinese": "向左后方走两步，杯子在你头顶的柜子里。", + "evidence": [ + { + "description": "The cup is taken from the cabinet and placed on the table.", + "timestamps": { + "start": "00:40:10", + "end": "00:40:24" + } + } + ], + "question_timestamps": "00:47:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk two steps back-right.", + "B": "Walk two steps to your back-left, the cups are in the cabinet above your head.", + "C": "Walk two steps forward-left.", + "D": "Walk one step to your back-left, then look for the cups in the cabinet just above your eye level.", + "E": "Walk three steps to your back-left and open the cabinet above you, since the cups aren’t directly overhead." + }, + "answer_label": "B", + "answer_text": "Walk two steps to your back-left, the cups are in the cabinet above your head.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_20", + "correct_answer": "Walk two steps to your back-left, the cups are in the cabinet above your head.", + "original_options": { + "A": "Walk two steps back-right.", + "B": "Walk two steps to your back-left, the cups are in the cabinet above your head.", + "C": "Walk two steps forward-left.", + "D": "Walk one step to your back-left, then look for the cups in the cabinet just above your eye level.", + "E": "Walk three steps to your back-left and open the cabinet above you, since the cups aren’t directly overhead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 19 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_19", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get more white cups, how should I walk?", + "answer": "Walk two steps to your front-left, the cups are in the cabinet above you and to your left.", + "question_chinese": "如果我要拿更多的白色杯子，我该怎么走？", + "answer_chinese": "向你的左前方走两步，杯子就在你左上方的柜子里。", + "evidence": [ + { + "description": "The cup is taken from the cabinet and placed on the table.", + "timestamps": { + "start": "00:40:10", + "end": "00:40:24" + } + }, + { + "description": "The cup next to the stove is picked up, filled with some water and poured into the pot, and finally placed next to the sink.", + "timestamps": { + "start": "00:51:02", + "end": "00:51:17" + } + }, + { + "description": "Walk from the sink to the staircase entrance.", + "timestamps": { + "start": "01:02:40", + "end": "01:02:55" + } + } + ], + "question_timestamps": "01:03:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Cabinet above you and right.", + "B": "Cabinet above you and ahead.", + "C": "Walk two steps to your front-left, then reach into the cabinet above you but positioned more toward your right.", + "D": "Walk two steps to your front-left, the cups are in the cabinet above you and to your left.", + "E": "Walk two steps to your front-left; the cups are in the cabinet above you but slightly in front of you." + }, + "answer_label": "D", + "answer_text": "Walk two steps to your front-left, the cups are in the cabinet above you and to your left.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_19", + "correct_answer": "Walk two steps to your front-left, the cups are in the cabinet above you and to your left.", + "original_options": { + "A": "Cabinet above you and right.", + "B": "Cabinet above you and ahead.", + "C": "Walk two steps to your front-left, then reach into the cabinet above you but positioned more toward your right.", + "D": "Walk two steps to your front-left, the cups are in the cabinet above you and to your left.", + "E": "Walk two steps to your front-left; the cups are in the cabinet above you but slightly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 20 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_18", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get more white cups, how should I walk?", + "answer": "Stand up, walk two steps to the left, the cups are in the cabinet above your head.", + "question_chinese": "如果我要拿更多的白色杯子，我该怎么走？", + "answer_chinese": "站起来，向左走两步，杯子在你头顶的柜子里。", + "evidence": [ + { + "description": "The cup is taken from the cabinet and placed on the table.", + "timestamps": { + "start": "00:40:10", + "end": "00:40:24" + } + }, + { + "description": "The cup next to the stove is picked up, filled with some water and poured into the pot, and finally placed next to the sink.", + "timestamps": { + "start": "00:51:02", + "end": "00:51:17" + } + }, + { + "description": "Walk from the sink to the staircase entrance.", + "timestamps": { + "start": "01:02:40", + "end": "01:02:55" + } + } + ], + "question_timestamps": "01:06:54", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Stand up, walk two steps to the left, the cups are in the cabinet above your head.", + "B": "Stand up, walk one step left.", + "C": "Stand up, walk two steps right.", + "D": "Stand up, walk three steps to the left, then grab the cups from a cabinet that isn’t directly above you.", + "E": "Stand up and walk two steps to the left, but the cups are on the counter rather than in an overhead cabinet." + }, + "answer_label": "A", + "answer_text": "Stand up, walk two steps to the left, the cups are in the cabinet above your head.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_18", + "correct_answer": "Stand up, walk two steps to the left, the cups are in the cabinet above your head.", + "original_options": { + "A": "Stand up, walk two steps to the left, the cups are in the cabinet above your head.", + "B": "Stand up, walk one step left.", + "C": "Stand up, walk two steps right.", + "D": "Stand up, walk three steps to the left, then grab the cups from a cabinet that isn’t directly above you.", + "E": "Stand up and walk two steps to the left, but the cups are on the counter rather than in an overhead cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 22 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_14", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get the white cup, how should I walk?", + "answer": "No need to walk, stand up, the cup is right in front of you.", + "question_chinese": "如果我要拿白色的杯子，我应该怎么走？", + "answer_chinese": "不用走，站起来，杯子就在你面前。", + "evidence": [ + { + "description": "The cup is taken from the cabinet and placed on the table.", + "timestamps": { + "start": "00:40:10", + "end": "00:40:24" + } + } + ], + "question_timestamps": "00:42:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "No need to walk, stand up, the cup is right in front of you.", + "B": "Take one step forward.", + "C": "Step slightly to your left.", + "D": "Stand up and take a small step backward, because the white cup is behind you rather than in front.", + "E": "Stand up and move to your right a little; the cup is off to the side instead of straight ahead." + }, + "answer_label": "A", + "answer_text": "No need to walk, stand up, the cup is right in front of you.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_14", + "correct_answer": "No need to walk, stand up, the cup is right in front of you.", + "original_options": { + "A": "No need to walk, stand up, the cup is right in front of you.", + "B": "Take one step forward.", + "C": "Step slightly to your left.", + "D": "Stand up and take a small step backward, because the white cup is behind you rather than in front.", + "E": "Stand up and move to your right a little; the cup is off to the side instead of straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 19, + 20 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_15", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get the white cup, how should I walk?", + "answer": "Walk one step to your back-right, the cup is on the table.", + "question_chinese": "如果我要拿白色的杯子，我应该怎么走？", + "answer_chinese": "向右后方走一步，杯子就在桌子上。", + "evidence": [ + { + "description": "The cup is taken from the cabinet and placed on the table.", + "timestamps": { + "start": "00:40:10", + "end": "00:40:24" + } + } + ], + "question_timestamps": "00:43:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk one step to your front-right.", + "B": "Walk one step to your back-right, the cup is on the table.", + "C": "Walk two steps to your back-right.", + "D": "Walk one step to your back-left, and you’ll find the cup on the table instead.", + "E": "Take one step to your front-left; the cup should be sitting on the table there." + }, + "answer_label": "B", + "answer_text": "Walk one step to your back-right, the cup is on the table.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_15", + "correct_answer": "Walk one step to your back-right, the cup is on the table.", + "original_options": { + "A": "Walk one step to your front-right.", + "B": "Walk one step to your back-right, the cup is on the table.", + "C": "Walk two steps to your back-right.", + "D": "Walk one step to your back-left, and you’ll find the cup on the table instead.", + "E": "Take one step to your front-left; the cup should be sitting on the table there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_16", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get the white cup, how should I walk?", + "answer": "Walk one step to your back-left, the cup is next to the sink.", + "question_chinese": "如果我要拿白色的杯子，我应该怎么走？", + "answer_chinese": "向你的左后方走一步，杯子就在水槽的旁边。", + "evidence": [ + { + "description": "The cup is taken from the cabinet and placed on the table.", + "timestamps": { + "start": "00:40:10", + "end": "00:40:24" + } + }, + { + "description": "The cup next to the stove is picked up, filled with some water and poured into the pot, and finally placed next to the sink.", + "timestamps": { + "start": "00:51:02", + "end": "00:51:17" + } + } + ], + "question_timestamps": "01:01:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk one step to your front-left.", + "B": "Walk one step to your back-right.", + "C": "Walk one step to your left, and the cup will be on the counter rather than by the sink.", + "D": "Walk one step to your back-left, the cup is next to the sink.", + "E": "Take two steps to your front-right; the cup should be in a drawer instead of next to the sink." + }, + "answer_label": "D", + "answer_text": "Walk one step to your back-left, the cup is next to the sink.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_16", + "correct_answer": "Walk one step to your back-left, the cup is next to the sink.", + "original_options": { + "A": "Walk one step to your front-left.", + "B": "Walk one step to your back-right.", + "C": "Walk one step to your left, and the cup will be on the counter rather than by the sink.", + "D": "Walk one step to your back-left, the cup is next to the sink.", + "E": "Take two steps to your front-right; the cup should be in a drawer instead of next to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_17", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get the white cup, how should I walk?", + "answer": "Walk three steps to your front-left, the cup is next to the sink.", + "question_chinese": "如果我要拿白色的杯子，我应该怎么走？", + "answer_chinese": "向你的左前方走三步，杯子就在水槽旁边。", + "evidence": [ + { + "description": "The cup is taken from the cabinet and placed on the table.", + "timestamps": { + "start": "00:40:10", + "end": "00:40:24" + } + }, + { + "description": "The cup next to the stove is picked up, filled with some water and poured into the pot, and finally placed next to the sink.", + "timestamps": { + "start": "00:51:02", + "end": "00:51:17" + } + }, + { + "description": "Walk from the sink to the staircase entrance.", + "timestamps": { + "start": "01:02:40", + "end": "01:02:55" + } + } + ], + "question_timestamps": "01:03:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Walk three steps to your front-right.", + "B": "Walk two steps to your front-left.", + "C": "Walk three steps to your front-left, the cup is next to the sink.", + "D": "Walk three steps to your back-left, and you’ll reach the cup under the table instead of the sink.", + "E": "Take four steps to your front-left; the cup will be by the sink but farther than you need to go." + }, + "answer_label": "C", + "answer_text": "Walk three steps to your front-left, the cup is next to the sink.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_17", + "correct_answer": "Walk three steps to your front-left, the cup is next to the sink.", + "original_options": { + "A": "Walk three steps to your front-right.", + "B": "Walk two steps to your front-left.", + "C": "Walk three steps to your front-left, the cup is next to the sink.", + "D": "Walk three steps to your back-left, and you’ll reach the cup under the table instead of the sink.", + "E": "Take four steps to your front-left; the cup will be by the sink but farther than you need to go." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 20 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_2_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the frying pan located?", + "answer": "The frying pan is on the stove along the outer edge of the stovetop in front of you, near the table edge.", + "question_chinese": "平底锅在什么位置？", + "answer_chinese": "平底锅在你面前的灶台外侧靠桌沿的炉灶上。", + "evidence": [ + { + "description": "The frying pan is taken out of the cabinet and placed on the outer burner.", + "timestamps": { + "start": "00:31:40", + "end": "00:31:45" + } + } + ], + "question_timestamps": "00:32:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on the stovetop behind you.", + "B": "It’s on the stovetop to your left.", + "C": "The frying pan is on the stove along the outer edge of the stovetop to your right, near the table edge.", + "D": "The frying pan is on the stove along the outer edge of the stovetop in front of you, near the table edge.", + "E": "The frying pan is on the stove along the outer edge of the stovetop behind you, near the table edge." + }, + "answer_label": "D", + "answer_text": "The frying pan is on the stove along the outer edge of the stovetop in front of you, near the table edge.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_2_T1", + "correct_answer": "The frying pan is on the stove along the outer edge of the stovetop in front of you, near the table edge.", + "original_options": { + "A": "It’s on the stovetop behind you.", + "B": "It’s on the stovetop to your left.", + "C": "The frying pan is on the stove along the outer edge of the stovetop to your right, near the table edge.", + "D": "The frying pan is on the stove along the outer edge of the stovetop in front of you, near the table edge.", + "E": "The frying pan is on the stove along the outer edge of the stovetop behind you, near the table edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 21, + 20 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_2_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the frying pan located?", + "answer": "It’s on the burner at the upper-right area of the stovetop in front of you.", + "question_chinese": "平底锅在什么位置？", + "answer_chinese": "在你面前的灶台的右上方的炉灶上。", + "evidence": [ + { + "description": "The frying pan is taken out of the cabinet and placed on the outer burner.", + "timestamps": { + "start": "00:31:40", + "end": "00:31:45" + } + }, + { + "description": "The frying pan is moved from the lower burner to the upper right burner.", + "timestamps": { + "start": "00:33:35", + "end": "00:33:42" + } + } + ], + "question_timestamps": "00:34:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on the burner at the upper-right area of the stovetop in front of you.", + "B": "It’s on the lower-left burner.", + "C": "It’s on the upper-left burner.", + "D": "It’s on the burner at the lower-right area of the stovetop in front of you, not the upper-right.", + "E": "It’s on the burner at the upper-right area of the stovetop to your left rather than in front of you." + }, + "answer_label": "A", + "answer_text": "It’s on the burner at the upper-right area of the stovetop in front of you.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_2_T2", + "correct_answer": "It’s on the burner at the upper-right area of the stovetop in front of you.", + "original_options": { + "A": "It’s on the burner at the upper-right area of the stovetop in front of you.", + "B": "It’s on the lower-left burner.", + "C": "It’s on the upper-left burner.", + "D": "It’s on the burner at the lower-right area of the stovetop in front of you, not the upper-right.", + "E": "It’s on the burner at the upper-right area of the stovetop to your left rather than in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 21 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_2_T3", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the frying pan located?", + "answer": "The frying pan is on the stove along the outer edge of the stovetop in front of you, near the table edge.", + "question_chinese": "平底锅在什么位置？", + "answer_chinese": "平底锅在你面前的灶台外侧靠桌沿的炉灶上。", + "evidence": [ + { + "description": "The frying pan is taken out of the cabinet and placed on the outer burner.", + "timestamps": { + "start": "00:31:40", + "end": "00:31:45" + } + }, + { + "description": "The frying pan is moved from the lower burner to the upper right burner.", + "timestamps": { + "start": "00:33:35", + "end": "00:33:42" + } + }, + { + "description": "The frying pan is moved back from the upper right burner to the lower burner.", + "timestamps": { + "start": "00:35:57", + "end": "00:36:01" + } + } + ], + "question_timestamps": "00:46:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on the stovetop behind you.", + "B": "It’s on the stovetop to your right.", + "C": "The frying pan is on the stove along the outer edge of the stovetop to your left, near the table edge.", + "D": "The frying pan is on the stove along the outer edge of the stovetop in front of you, near the table edge.", + "E": "The frying pan is on the stove along the outer edge of the stovetop behind you, near the table edge." + }, + "answer_label": "D", + "answer_text": "The frying pan is on the stove along the outer edge of the stovetop in front of you, near the table edge.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_2_T3", + "correct_answer": "The frying pan is on the stove along the outer edge of the stovetop in front of you, near the table edge.", + "original_options": { + "A": "It’s on the stovetop behind you.", + "B": "It’s on the stovetop to your right.", + "C": "The frying pan is on the stove along the outer edge of the stovetop to your left, near the table edge.", + "D": "The frying pan is on the stove along the outer edge of the stovetop in front of you, near the table edge.", + "E": "The frying pan is on the stove along the outer edge of the stovetop behind you, near the table edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 21, + 20 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_6", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the cell phone?", + "answer": "It is on the bed to your right.", + "question_chinese": "手机在什么位置？", + "answer_chinese": "在你右手边的床上.", + "evidence": [ + { + "description": "The cell phone is placed on the bed.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:00:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on the bed to your left.", + "B": "It is on the table to your right.", + "C": "It is on the shelf to your right, not on the bed.", + "D": "It is on the bed to your right.", + "E": "It is under the bed in front of you rather than on the bed to your right." + }, + "answer_label": "D", + "answer_text": "It is on the bed to your right.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_6", + "correct_answer": "It is on the bed to your right.", + "original_options": { + "A": "It is on the bed to your left.", + "B": "It is on the table to your right.", + "C": "It is on the shelf to your right, not on the bed.", + "D": "It is on the bed to your right.", + "E": "It is under the bed in front of you rather than on the bed to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 12, + 17 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_8", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the cell phone?", + "answer": "It is two steps away to your left.", + "question_chinese": "手机在什么位置？", + "answer_chinese": "在你左侧两步远处。", + "evidence": [ + { + "description": "The cell phone is placed on the bed.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:22" + } + }, + { + "description": "The cell phone is taken downstairs, used for a few minutes, and then placed on the table.", + "timestamps": { + "start": "00:00:34", + "end": "00:08:10" + } + }, + { + "description": "The cell phone is picked up and propped against the side of the water bottle.", + "timestamps": { + "start": "00:29:42", + "end": "00:30:11" + } + } + ], + "question_timestamps": "00:33:05", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "It is two steps away to your left.", + "B": "It is one step away to your left.", + "C": "It is two steps away to your right.", + "D": "It is two steps away in front of you, not two steps away to your left.", + "E": "It is three steps away to your left instead of two, so you need to go farther." + }, + "answer_label": "A", + "answer_text": "It is two steps away to your left.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_8", + "correct_answer": "It is two steps away to your left.", + "original_options": { + "A": "It is two steps away to your left.", + "B": "It is one step away to your left.", + "C": "It is two steps away to your right.", + "D": "It is two steps away in front of you, not two steps away to your left.", + "E": "It is three steps away to your left instead of two, so you need to go farther." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 17 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_10", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Now, in relation to the sink, which side is the cell phone on?", + "answer": "It is on the left side.", + "question_chinese": "现在，手机相对于水槽来说，在哪一边？", + "answer_chinese": "在左边。", + "evidence": [ + { + "description": "The cell phone is picked up and propped against the side of the water bottle.", + "timestamps": { + "start": "00:29:42", + "end": "00:30:11" + } + } + ], + "question_timestamps": "00:33:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on the right side.", + "B": "It is in front of the sink.", + "C": "It is positioned on the right side of the sink, closer to the edge of the counter.", + "D": "It is on the back side of the sink, sitting near the faucet area rather than beside it.", + "E": "It is on the left side." + }, + "answer_label": "E", + "answer_text": "It is on the left side.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_10", + "correct_answer": "It is on the left side.", + "original_options": { + "A": "It is on the right side.", + "B": "It is in front of the sink.", + "C": "It is positioned on the right side of the sink, closer to the edge of the counter.", + "D": "It is on the back side of the sink, sitting near the faucet area rather than beside it.", + "E": "It is on the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 18 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_9", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Now, in relation to the sink, which side is the cell phone on?", + "answer": "It is on the right side.", + "question_chinese": "现在，手机相对于水槽来说，在哪一边？", + "answer_chinese": "在右边。", + "evidence": [ + { + "description": "The cell phone is picked up and propped against the side of the water bottle.", + "timestamps": { + "start": "00:29:42", + "end": "00:30:11" + } + }, + { + "description": "The cell phone is picked up from the left side of the sink and placed on the right side of the sink.", + "timestamps": { + "start": "00:38:12", + "end": "00:38:25" + } + } + ], + "question_timestamps": "00:46:02", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It is on the left side.", + "B": "It is on the right side.", + "C": "It is in front of the sink.", + "D": "It is on the back side of the sink, not on either the left or right side.", + "E": "It is centered directly at the sink area, so it is not clearly on the right side." + }, + "answer_label": "B", + "answer_text": "It is on the right side.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_9", + "correct_answer": "It is on the right side.", + "original_options": { + "A": "It is on the left side.", + "B": "It is on the right side.", + "C": "It is in front of the sink.", + "D": "It is on the back side of the sink, not on either the left or right side.", + "E": "It is centered directly at the sink area, so it is not clearly on the right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 17 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of chicken are in the white plastic vegetable washing colander now?", + "answer": "There are six pieces of chicken in the white plastic vegetable washing colander now.", + "question_chinese": "现在白色塑料洗菜沥水盆里有多少块鸡肉？", + "answer_chinese": "现在白色塑料洗菜滤盆里有6块鸡肉。", + "evidence": [ + { + "description": "There is a plate of chicken on the table", + "timestamps": { + "start": "00:32:10", + "end": "00:32:30" + } + } + ], + "question_timestamps": "00:37:04", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are six pieces of chicken in the white plastic vegetable washing colander now.", + "B": "There are four pieces.", + "C": "There are eight pieces.", + "D": "There are five pieces of chicken in the white plastic vegetable washing colander now.", + "E": "There are three pieces of chicken in the white plastic vegetable washing colander now." + }, + "answer_label": "A", + "answer_text": "There are six pieces of chicken in the white plastic vegetable washing colander now.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_1", + "correct_answer": "There are six pieces of chicken in the white plastic vegetable washing colander now.", + "original_options": { + "A": "There are six pieces of chicken in the white plastic vegetable washing colander now.", + "B": "There are four pieces.", + "C": "There are eight pieces.", + "D": "There are five pieces of chicken in the white plastic vegetable washing colander now.", + "E": "There are three pieces of chicken in the white plastic vegetable washing colander now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "3a8774b9-ea4d-439e-a862-b29b29df999d", + "video_length": "01:13:06", + "qid": "3a8774b9-ea4d-439e-a862-b29b29df999d_2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of chicken are in the white plastic vegetable washing colander now?", + "answer": "There is no chicken left, because it was poured into the pot by you.", + "question_chinese": "现在白色塑料洗菜沥水盆里有多少块鸡肉？", + "answer_chinese": "已经没有鸡肉了，因为被你倒入了锅里。", + "evidence": [ + { + "description": "There is a plate of chicken on the table", + "timestamps": { + "start": "00:32:10", + "end": "00:32:30" + } + }, + { + "description": "The chicken is poured into the pot.", + "timestamps": { + "start": "00:51:52", + "end": "00:52:00" + } + } + ], + "question_timestamps": "00:57:55", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are two pieces left.", + "B": "There are four pieces left.", + "C": "There is no chicken left, because it was poured into the pot by you.", + "D": "There is still some chicken left because you only poured part of it into the pot.", + "E": "There is no chicken left because you transferred it into the frying pan instead of the pot." + }, + "answer_label": "C", + "answer_text": "There is no chicken left, because it was poured into the pot by you.", + "q_id": "3a8774b9-ea4d-439e-a862-b29b29df999d_2", + "correct_answer": "There is no chicken left, because it was poured into the pot by you.", + "original_options": { + "A": "There are two pieces left.", + "B": "There are four pieces left.", + "C": "There is no chicken left, because it was poured into the pot by you.", + "D": "There is still some chicken left because you only poured part of it into the pot.", + "E": "There is no chicken left because you transferred it into the frying pan instead of the pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + } +] diff --git a/QAs/3a9464f8-5025-4a79-ac94-9e3d51f1ef49-mcq.json b/QAs/3a9464f8-5025-4a79-ac94-9e3d51f1ef49-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ff67e2630f014b85b894a7b709545fd1cda2bd90 --- /dev/null +++ b/QAs/3a9464f8-5025-4a79-ac94-9e3d51f1ef49-mcq.json @@ -0,0 +1,1287 @@ +[ + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_10_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The fridge is on your right side right now.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱现在在你的右侧。", + "evidence": [ + { + "description": "The refrigerator is on your right.", + "timestamps": { + "start": "00:11:53", + "end": "00:12:08" + } + } + ], + "question_timestamps": "00:12:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is on your left.", + "B": "The fridge is on your right side right now.", + "C": "The fridge is behind you.", + "D": "The fridge is directly in front of you right now, not off to your right side.", + "E": "The fridge is behind you on the left side, so it is not to your right." + }, + "answer_label": "B", + "answer_text": "The fridge is on your right side right now.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_10_T1", + "correct_answer": "The fridge is on your right side right now.", + "original_options": { + "A": "The fridge is on your left.", + "B": "The fridge is on your right side right now.", + "C": "The fridge is behind you.", + "D": "The fridge is directly in front of you right now, not off to your right side.", + "E": "The fridge is behind you on the left side, so it is not to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 16 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_10_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The fridge was directly in front of you.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱在你的正前方。", + "evidence": [ + { + "description": "The refrigerator is on your right.", + "timestamps": { + "start": "00:17:05", + "end": "00:17:12" + } + } + ], + "question_timestamps": "00:17:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge was behind you.", + "B": "The fridge was directly in front of you.", + "C": "The fridge was to your left.", + "D": "The fridge was directly to your right, not positioned straight ahead of you.", + "E": "The fridge was off to your left side rather than being directly in front of you." + }, + "answer_label": "B", + "answer_text": "The fridge was directly in front of you.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_10_T2", + "correct_answer": "The fridge was directly in front of you.", + "original_options": { + "A": "The fridge was behind you.", + "B": "The fridge was directly in front of you.", + "C": "The fridge was to your left.", + "D": "The fridge was directly to your right, not positioned straight ahead of you.", + "E": "The fridge was off to your left side rather than being directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 16 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_20", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator now relative to me?", + "answer": "The refrigerator is on your right.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱在你的右侧。", + "evidence": [ + { + "description": "The refrigerator is on your right.", + "timestamps": { + "start": "00:17:05", + "end": "00:17:12" + } + } + ], + "question_timestamps": "00:23:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is on your left.", + "B": "The refrigerator is behind you.", + "C": "The refrigerator is directly in front of you, not positioned off to your right side.", + "D": "The refrigerator is behind you on the left, so it is not on your right.", + "E": "The refrigerator is on your right." + }, + "answer_label": "E", + "answer_text": "The refrigerator is on your right.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_20", + "correct_answer": "The refrigerator is on your right.", + "original_options": { + "A": "The refrigerator is on your left.", + "B": "The refrigerator is behind you.", + "C": "The refrigerator is directly in front of you, not positioned off to your right side.", + "D": "The refrigerator is behind you on the left, so it is not on your right.", + "E": "The refrigerator is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 15 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_7_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how many steps away is the refrigerator from me right now?", + "answer": "The fridge is one step away from you right now.", + "question_chinese": "冰箱现在距离我大约有几步远？", + "answer_chinese": "冰箱现在距离你有一步远。", + "evidence": [ + { + "description": "The refrigerator is now about one step away from you.", + "timestamps": { + "start": "00:17:35", + "end": "00:17:43" + } + } + ], + "question_timestamps": "00:17:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Zero steps away.", + "B": "Two steps away.", + "C": "The fridge is three steps away from you right now, not just a single step away.", + "D": "The fridge is four steps away from you right now, requiring a short walk to reach it.", + "E": "The fridge is one step away from you right now." + }, + "answer_label": "E", + "answer_text": "The fridge is one step away from you right now.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_7_T2", + "correct_answer": "The fridge is one step away from you right now.", + "original_options": { + "A": "Zero steps away.", + "B": "Two steps away.", + "C": "The fridge is three steps away from you right now, not just a single step away.", + "D": "The fridge is four steps away from you right now, requiring a short walk to reach it.", + "E": "The fridge is one step away from you right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_13", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the refrigerator from me now?", + "answer": "The refrigerator is now about two steps away from you.", + "question_chinese": "冰箱现在距离我大约有几步远？", + "answer_chinese": "冰箱现在距离你大约有两步。", + "evidence": [ + { + "description": "The refrigerator is now about two steps away from you.", + "timestamps": { + "start": "00:17:35", + "end": "00:17:52" + } + } + ], + "question_timestamps": "00:17:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "About one step away.", + "B": "The refrigerator is now about two steps away from you.", + "C": "About three steps away.", + "D": "The refrigerator is roughly four steps away from you right now, not just a couple of steps.", + "E": "The refrigerator is only about an arm’s length away from you, close enough to reach immediately." + }, + "answer_label": "B", + "answer_text": "The refrigerator is now about two steps away from you.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_13", + "correct_answer": "The refrigerator is now about two steps away from you.", + "original_options": { + "A": "About one step away.", + "B": "The refrigerator is now about two steps away from you.", + "C": "About three steps away.", + "D": "The refrigerator is roughly four steps away from you right now, not just a couple of steps.", + "E": "The refrigerator is only about an arm’s length away from you, close enough to reach immediately." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_7_T1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how many steps away is the refrigerator from me right now?", + "answer": "The fridge is two steps away from you right now.", + "question_chinese": "冰箱现在距离我大约有几步远？", + "answer_chinese": "冰箱现在距离你有两步远。", + "evidence": [ + { + "description": "The refrigerator is now about two steps away from you.", + "timestamps": { + "start": "00:17:35", + "end": "00:17:52" + } + } + ], + "question_timestamps": "00:17:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Zero steps away.", + "B": "One step away.", + "C": "The fridge is about three steps away from you right now, a little farther than you think.", + "D": "The fridge is two steps away from you right now.", + "E": "The fridge is around four steps away from you right now, not just a couple of steps." + }, + "answer_label": "D", + "answer_text": "The fridge is two steps away from you right now.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_7_T1", + "correct_answer": "The fridge is two steps away from you right now.", + "original_options": { + "A": "Zero steps away.", + "B": "One step away.", + "C": "The fridge is about three steps away from you right now, a little farther than you think.", + "D": "The fridge is two steps away from you right now.", + "E": "The fridge is around four steps away from you right now, not just a couple of steps." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_14", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the refrigerator from me now?", + "answer": "The refrigerator is now about two steps away from you.", + "question_chinese": "冰箱现在距离我大约有几步远？", + "answer_chinese": "冰箱现在距离你大约有两步。", + "evidence": [ + { + "description": "The refrigerator is now about two steps away from you.", + "timestamps": { + "start": "00:17:35", + "end": "00:17:52" + } + } + ], + "question_timestamps": "00:22:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is now about two steps away from you.", + "B": "About one step away.", + "C": "About three steps away.", + "D": "The refrigerator is now about four steps away from you, farther than a couple of steps.", + "E": "The refrigerator is now roughly five steps away from you, so you’d need to walk a bit." + }, + "answer_label": "A", + "answer_text": "The refrigerator is now about two steps away from you.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_14", + "correct_answer": "The refrigerator is now about two steps away from you.", + "original_options": { + "A": "The refrigerator is now about two steps away from you.", + "B": "About one step away.", + "C": "About three steps away.", + "D": "The refrigerator is now about four steps away from you, farther than a couple of steps.", + "E": "The refrigerator is now roughly five steps away from you, so you’d need to walk a bit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 18 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_9_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was peeling the radish just now, what ingredients were on the cutting board?", + "answer": "You saw mushrooms on the cutting board. The mushrooms are on the right side of the cutting board in front of you.", + "question_chinese": "刚刚我削萝卜时，案板上的食材有什么？", + "answer_chinese": "你看见案板上有蘑菇。蘑菇在你面前案板的右边。", + "evidence": [ + { + "description": "You see mushrooms on the chopping board.", + "timestamps": { + "start": "00:18:02", + "end": "00:18:07" + } + } + ], + "question_timestamps": "00:23:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw mushrooms on the cutting board. The mushrooms are on the right side of the cutting board in front of you.", + "B": "Carrots on the board.", + "C": "Onions on the board.", + "D": "You saw mushrooms on the cutting board, but they were positioned on the left side in front of you.", + "E": "You saw mushrooms on the cutting board, and they were placed on the right side but behind you." + }, + "answer_label": "A", + "answer_text": "You saw mushrooms on the cutting board. The mushrooms are on the right side of the cutting board in front of you.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_9_T1", + "correct_answer": "You saw mushrooms on the cutting board. The mushrooms are on the right side of the cutting board in front of you.", + "original_options": { + "A": "You saw mushrooms on the cutting board. The mushrooms are on the right side of the cutting board in front of you.", + "B": "Carrots on the board.", + "C": "Onions on the board.", + "D": "You saw mushrooms on the cutting board, but they were positioned on the left side in front of you.", + "E": "You saw mushrooms on the cutting board, and they were placed on the right side but behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 18 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_9_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was peeling the radish just now, what ingredients were on the cutting board?", + "answer": "You saw a red bell pepper on the cutting board. The red bell pepper is on the left side of the cutting board in front of you.", + "question_chinese": "刚刚我削萝卜时，案板上的食材有什么？", + "answer_chinese": "你看见案板上有红色彩椒。红色彩椒在你面前案板上的左边。", + "evidence": [ + { + "description": "You see red bell peppers on the cutting board.", + "timestamps": { + "start": "00:18:02", + "end": "00:18:07" + } + } + ], + "question_timestamps": "00:25:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A green pepper.", + "B": "A tomato.", + "C": "You saw a red bell pepper on the cutting board, but it was on the right side in front of you.", + "D": "You saw a red bell pepper on the cutting board. The red bell pepper is on the left side of the cutting board in front of you.", + "E": "You saw a red bell pepper on the cutting board, and it was on the left side but behind you." + }, + "answer_label": "D", + "answer_text": "You saw a red bell pepper on the cutting board. The red bell pepper is on the left side of the cutting board in front of you.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_9_T2", + "correct_answer": "You saw a red bell pepper on the cutting board. The red bell pepper is on the left side of the cutting board in front of you.", + "original_options": { + "A": "A green pepper.", + "B": "A tomato.", + "C": "You saw a red bell pepper on the cutting board, but it was on the right side in front of you.", + "D": "You saw a red bell pepper on the cutting board. The red bell pepper is on the left side of the cutting board in front of you.", + "E": "You saw a red bell pepper on the cutting board, and it was on the left side but behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 21, + 20 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_3_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I pour flour in total just now?", + "answer": "You just poured flour a total of two times. The flour is by your right hand.", + "question_chinese": "我刚刚一共倒了几次面粉？", + "answer_chinese": "你刚刚一共倒了两次面粉。面粉在你右手边。", + "evidence": [ + { + "description": "You poured flour for the first time.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:34" + } + }, + { + "description": "You poured the flour for the second time.", + "timestamps": { + "start": "00:00:38", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero times.", + "B": "One time.", + "C": "You just poured flour a total of three times, and the flour is still by your right hand.", + "D": "You just poured flour a total of four times, with the flour remaining by your right hand.", + "E": "You just poured flour a total of two times. The flour is by your right hand." + }, + "answer_label": "E", + "answer_text": "You just poured flour a total of two times. The flour is by your right hand.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_3_T1", + "correct_answer": "You just poured flour a total of two times. The flour is by your right hand.", + "original_options": { + "A": "Zero times.", + "B": "One time.", + "C": "You just poured flour a total of three times, and the flour is still by your right hand.", + "D": "You just poured flour a total of four times, with the flour remaining by your right hand.", + "E": "You just poured flour a total of two times. The flour is by your right hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 17 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_5", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I pour the flour in total just now?", + "answer": "You just collapsed three times in total.", + "question_chinese": "我刚刚一共倒了几次面粉？", + "answer_chinese": "你刚刚一共倒了三次。", + "evidence": [ + { + "description": "You poured flour for the first time.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:34" + } + }, + { + "description": "You poured the flour for the second time.", + "timestamps": { + "start": "00:00:38", + "end": "00:00:43" + } + }, + { + "description": "You poured the flour for the third time.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You just collapsed two times in total.", + "B": "You just collapsed four times in total.", + "C": "You just collapsed once, then did it one more time, for a total of two collapses.", + "D": "You just collapsed three times in total.", + "E": "You just collapsed four times overall, repeating the same action across four separate moments just now." + }, + "answer_label": "D", + "answer_text": "You just collapsed three times in total.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_5", + "correct_answer": "You just collapsed three times in total.", + "original_options": { + "A": "You just collapsed two times in total.", + "B": "You just collapsed four times in total.", + "C": "You just collapsed once, then did it one more time, for a total of two collapses.", + "D": "You just collapsed three times in total.", + "E": "You just collapsed four times overall, repeating the same action across four separate moments just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_3_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I pour flour in total just now?", + "answer": "You just poured a total of three times. The flour is by your right hand.", + "question_chinese": "我刚刚一共倒了几次面粉？", + "answer_chinese": "你刚刚一共倒了三次。面粉在你右手边。", + "evidence": [ + { + "description": "You poured flour for the first time.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:34" + } + }, + { + "description": "You poured the flour for the second time.", + "timestamps": { + "start": "00:00:38", + "end": "00:00:43" + } + }, + { + "description": "You poured the flour for the third time.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:09:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just poured a total of three times. The flour is by your right hand.", + "B": "One time.", + "C": "Two times.", + "D": "You just poured a total of four times, and the flour is by your right hand.", + "E": "You just poured a total of five times, with the flour still sitting by your right hand." + }, + "answer_label": "A", + "answer_text": "You just poured a total of three times. The flour is by your right hand.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_3_T2", + "correct_answer": "You just poured a total of three times. The flour is by your right hand.", + "original_options": { + "A": "You just poured a total of three times. The flour is by your right hand.", + "B": "One time.", + "C": "Two times.", + "D": "You just poured a total of four times, and the flour is by your right hand.", + "E": "You just poured a total of five times, with the flour still sitting by your right hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_5_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the refrigerator just now?", + "answer": "You just opened the fridge once. The fridge is to your back-right.", + "question_chinese": "我刚刚打开了几次冰箱？", + "answer_chinese": "你刚刚打开了一次冰箱。冰箱在你右后方。", + "evidence": [ + { + "description": "You opened the refrigerator once.", + "timestamps": { + "start": "00:11:58", + "end": "00:12:08" + } + } + ], + "question_timestamps": "00:16:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened it zero times.", + "B": "You just opened the fridge once. The fridge is to your back-right.", + "C": "You opened it twice.", + "D": "You just opened the fridge two times, even though it stayed to your back-right the whole time.", + "E": "You just opened the fridge three times, with the refrigerator remaining to your back-right throughout." + }, + "answer_label": "B", + "answer_text": "You just opened the fridge once. The fridge is to your back-right.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_5_T1", + "correct_answer": "You just opened the fridge once. The fridge is to your back-right.", + "original_options": { + "A": "You opened it zero times.", + "B": "You just opened the fridge once. The fridge is to your back-right.", + "C": "You opened it twice.", + "D": "You just opened the fridge two times, even though it stayed to your back-right the whole time.", + "E": "You just opened the fridge three times, with the refrigerator remaining to your back-right throughout." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 15 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_5_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the refrigerator just now?", + "answer": "You just opened the fridge three times. The fridge is to your back-right.", + "question_chinese": "我刚刚打开了几次冰箱？", + "answer_chinese": "你刚刚打开了三次冰箱。冰箱在你右后方。", + "evidence": [ + { + "description": "You opened the refrigerator once.", + "timestamps": { + "start": "00:11:58", + "end": "00:12:08" + } + }, + { + "description": "You opened the refrigerator.", + "timestamps": { + "start": "00:17:05", + "end": "00:17:15" + } + }, + { + "description": "You opened the refrigerator.", + "timestamps": { + "start": "00:17:16", + "end": "00:17:28" + } + } + ], + "question_timestamps": "00:17:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened it twice.", + "B": "You opened it four times.", + "C": "You just opened the fridge three times. The fridge is to your back-right.", + "D": "You just opened the fridge one time, with the refrigerator positioned to your back-right.", + "E": "You just opened the fridge five times, and the fridge was still to your back-right afterward." + }, + "answer_label": "C", + "answer_text": "You just opened the fridge three times. The fridge is to your back-right.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_5_T2", + "correct_answer": "You just opened the fridge three times. The fridge is to your back-right.", + "original_options": { + "A": "You opened it twice.", + "B": "You opened it four times.", + "C": "You just opened the fridge three times. The fridge is to your back-right.", + "D": "You just opened the fridge one time, with the refrigerator positioned to your back-right.", + "E": "You just opened the fridge five times, and the fridge was still to your back-right afterward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 16 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_7", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the spoon?", + "answer": "You put the spoon on the table.", + "question_chinese": "我刚刚把勺子放在什么位置上了？", + "answer_chinese": "你把勺子放在了桌上。", + "evidence": [ + { + "description": "You put the spoon on the table.", + "timestamps": { + "start": "00:01:45", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:03:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the drawer.", + "B": "On the counter.", + "C": "You put the spoon on the table.", + "D": "You put the spoon in the pot on the stove instead of leaving it out.", + "E": "You placed the spoon on the shelf above the microwave rather than on any table surface." + }, + "answer_label": "C", + "answer_text": "You put the spoon on the table.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_7", + "correct_answer": "You put the spoon on the table.", + "original_options": { + "A": "In the drawer.", + "B": "On the counter.", + "C": "You put the spoon on the table.", + "D": "You put the spoon in the pot on the stove instead of leaving it out.", + "E": "You placed the spoon on the shelf above the microwave rather than on any table surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_8", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the spoon?", + "answer": "You put the spoon on the plate.", + "question_chinese": "我刚刚把勺子放在什么位置上了？", + "answer_chinese": "你把勺子放在盘子里了。", + "evidence": [ + { + "description": "You put the spoon on the plate.", + "timestamps": { + "start": "00:10:45", + "end": "00:10:49" + } + } + ], + "question_timestamps": "00:12:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the countertop.", + "B": "Inside the pot.", + "C": "You put the spoon in the drawer rather than setting it on any dish.", + "D": "You put the spoon on the plate.", + "E": "You left the spoon beside the bowl instead of placing it directly on the plate." + }, + "answer_label": "D", + "answer_text": "You put the spoon on the plate.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_8", + "correct_answer": "You put the spoon on the plate.", + "original_options": { + "A": "On the countertop.", + "B": "Inside the pot.", + "C": "You put the spoon in the drawer rather than setting it on any dish.", + "D": "You put the spoon on the plate.", + "E": "You left the spoon beside the bowl instead of placing it directly on the plate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 15 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_6_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put my phone?", + "answer": "You just put the phone on the right side of the stove in front of you.", + "question_chinese": "我刚刚把手机放在什么位置上？", + "answer_chinese": "你刚刚把手机放在你面前灶台的右侧。", + "evidence": [ + { + "description": "You just placed your mobile phone on the right side of the stove.", + "timestamps": { + "start": "00:11:08", + "end": "00:11:20" + } + } + ], + "question_timestamps": "00:16:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you at the stove.", + "B": "You just put the phone on the right side of the stove in front of you.", + "C": "On the stove’s left side.", + "D": "You placed the phone on the right side of the stove but behind you, not in front.", + "E": "You set the phone on the counter beside the stove, rather than on the stove’s right side in front of you." + }, + "answer_label": "B", + "answer_text": "You just put the phone on the right side of the stove in front of you.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_6_T1", + "correct_answer": "You just put the phone on the right side of the stove in front of you.", + "original_options": { + "A": "Behind you at the stove.", + "B": "You just put the phone on the right side of the stove in front of you.", + "C": "On the stove’s left side.", + "D": "You placed the phone on the right side of the stove but behind you, not in front.", + "E": "You set the phone on the counter beside the stove, rather than on the stove’s right side in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 22 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_6_T2", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put my phone?", + "answer": "You just put the phone on the right side of the stove.", + "question_chinese": "我刚刚把手机放在什么位置上？", + "answer_chinese": "你刚刚把手机放在灶台的右侧。", + "evidence": [ + { + "description": "You just placed your mobile phone on the right side of the stove.", + "timestamps": { + "start": "00:11:08", + "end": "00:11:20" + } + } + ], + "question_timestamps": "00:23:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To the left of the stove.", + "B": "Behind the stove.", + "C": "You put the phone in front of the stove rather than on the stove’s right side.", + "D": "You just put the phone on the right side of the stove.", + "E": "You set the phone to your right of the stove, not specifically on the right side of it." + }, + "answer_label": "D", + "answer_text": "You just put the phone on the right side of the stove.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_6_T2", + "correct_answer": "You just put the phone on the right side of the stove.", + "original_options": { + "A": "To the left of the stove.", + "B": "Behind the stove.", + "C": "You put the phone in front of the stove rather than on the stove’s right side.", + "D": "You just put the phone on the right side of the stove.", + "E": "You set the phone to your right of the stove, not specifically on the right side of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 3, + 17, + 18 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the faucet directly now?", + "answer": "You can touch the faucet directly.", + "question_chinese": "我现在可以直接触碰到水龙头吗？", + "answer_chinese": "你可以直接触碰到水龙头。", + "evidence": [ + { + "description": "You can touch the faucet directly.", + "timestamps": { + "start": "00:02:09", + "end": "00:02:14" + } + } + ], + "question_timestamps": "00:02:17", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch the faucet directly.", + "B": "You cannot touch the faucet directly." + }, + "answer_label": "A", + "answer_text": "You can touch the faucet directly.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_1", + "correct_answer": "You can touch the faucet directly.", + "original_options": { + "A": "You can touch the faucet directly.", + "B": "You cannot touch the faucet directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the faucet directly now?", + "answer": "You can't touch the faucet directly.", + "question_chinese": "我现在可以直接触碰到水龙头吗？", + "answer_chinese": "你无法直接触碰到水龙头。", + "evidence": [ + { + "description": "The faucet is just one step away from you.", + "timestamps": { + "start": "00:09:56", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:10:44", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't touch the faucet directly.", + "B": "You can touch the faucet directly." + }, + "answer_label": "A", + "answer_text": "You can't touch the faucet directly.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_2", + "correct_answer": "You can't touch the faucet directly.", + "original_options": { + "A": "You can't touch the faucet directly.", + "B": "You can touch the faucet directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_8_T1", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the pot right now?", + "answer": "You can’t directly touch the pot right now. The pot is to your back-left.", + "question_chinese": "我现在可以直接触碰到锅吗？", + "answer_chinese": "你现在无法直接触碰到锅。锅在你左后方。", + "evidence": [ + { + "description": "You can't make direct contact now.", + "timestamps": { + "start": "00:23:50", + "end": "00:23:53" + } + } + ], + "question_timestamps": "00:24:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t directly touch the pot right now. The pot is to your back-left.", + "B": "You can directly touch the pot right now. The pot is in front of you." + }, + "answer_label": "A", + "answer_text": "You can’t directly touch the pot right now. The pot is to your back-left.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_8_T1", + "correct_answer": "You can’t directly touch the pot right now. The pot is to your back-left.", + "original_options": { + "A": "You can’t directly touch the pot right now. The pot is to your back-left.", + "B": "You can directly touch the pot right now. The pot is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_8_T2", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the pot right now?", + "answer": "You can directly touch the pot right now. The pot is in front of you.", + "question_chinese": "我现在可以直接触碰到锅吗？", + "answer_chinese": "你现在可以直接触碰到锅。锅在你面前。", + "evidence": [ + { + "description": "You can make direct contact now.", + "timestamps": { + "start": "00:26:47", + "end": "00:27:57" + } + } + ], + "question_timestamps": "00:28:01", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t directly touch the pot right now. The pot is to your back-left.", + "B": "You can directly touch the pot right now. The pot is in front of you." + }, + "answer_label": "B", + "answer_text": "You can directly touch the pot right now. The pot is in front of you.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_8_T2", + "correct_answer": "You can directly touch the pot right now. The pot is in front of you.", + "original_options": { + "A": "You can’t directly touch the pot right now. The pot is to your back-left.", + "B": "You can directly touch the pot right now. The pot is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_3", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the position of the stove relative to me now?", + "answer": "The stove is now on your left.", + "question_chinese": "灶台现在相对我是什么位置？", + "answer_chinese": "灶台现在在你的左侧。", + "evidence": [ + { + "description": "The stove is now on your left.", + "timestamps": { + "start": "00:06:40", + "end": "00:06:41" + } + } + ], + "question_timestamps": "00:06:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The stove is now on your left.", + "B": "The stove is now in front of you.", + "C": "The stove is now behind you.", + "D": "The stove is now on your right, not on your left where you’d expect it to be.", + "E": "The stove is now off to your back-right, so you’d need to turn around to face it." + }, + "answer_label": "A", + "answer_text": "The stove is now on your left.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_3", + "correct_answer": "The stove is now on your left.", + "original_options": { + "A": "The stove is now on your left.", + "B": "The stove is now in front of you.", + "C": "The stove is now behind you.", + "D": "The stove is now on your right, not on your left where you’d expect it to be.", + "E": "The stove is now off to your back-right, so you’d need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 18, + 18 + ] + }, + { + "video_uid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49", + "video_length": "00:31:01", + "qid": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_4", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the position of the stove relative to me now?", + "answer": "The stove is on your left.", + "question_chinese": "灶台现在相对我是什么位置？", + "answer_chinese": "灶台在你的左侧。", + "evidence": [ + { + "description": "The stove is now on your left.", + "timestamps": { + "start": "00:19:36", + "end": "00:19:37" + } + } + ], + "question_timestamps": "00:19:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The stove is in front of you.", + "B": "The stove is on your left.", + "C": "The stove is behind you.", + "D": "The stove is on your right, so you would have to look right to see it.", + "E": "The stove is positioned at your back-right, meaning it isn’t on your left anymore." + }, + "answer_label": "B", + "answer_text": "The stove is on your left.", + "q_id": "3a9464f8-5025-4a79-ac94-9e3d51f1ef49_4", + "correct_answer": "The stove is on your left.", + "original_options": { + "A": "The stove is in front of you.", + "B": "The stove is on your left.", + "C": "The stove is behind you.", + "D": "The stove is on your right, so you would have to look right to see it.", + "E": "The stove is positioned at your back-right, meaning it isn’t on your left anymore." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 15 + ] + } +] diff --git a/QAs/3b5cd89f-6f74-4077-85ca-9a64b1ffa394-mcq.json b/QAs/3b5cd89f-6f74-4077-85ca-9a64b1ffa394-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8c566947c5b5fc73dee5fd0f011b158f471b1461 --- /dev/null +++ b/QAs/3b5cd89f-6f74-4077-85ca-9a64b1ffa394-mcq.json @@ -0,0 +1,835 @@ +[ + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_11", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I put down the long-handled hoe last time, where was the long-handled hoe placed?", + "answer": "At that time, the long-handled hoe was leaning against the end of the black net in the right front of where you are now.", + "question_chinese": "上次我吧长柄锄头放下的时候，长柄锄头被放在哪里了？", + "answer_chinese": "当时长柄锄头正靠在你现在所处位置的右前方黑网尽头处。", + "evidence": [ + { + "description": "You leaned the hoe against the end of the black net.", + "timestamps": { + "start": "00:06:40", + "end": "00:07:03" + } + } + ], + "question_timestamps": "00:09:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Leaning at the black net’s side.", + "B": "Lying on the ground by the net.", + "C": "At that time, the long-handled hoe was leaning against the side of the black net to your left rear.", + "D": "At that time, the long-handled hoe was standing upright near the black net’s front corner, closer to your center.", + "E": "At that time, the long-handled hoe was leaning against the end of the black net in the right front of where you are now." + }, + "answer_label": "E", + "answer_text": "At that time, the long-handled hoe was leaning against the end of the black net in the right front of where you are now.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_11", + "correct_answer": "At that time, the long-handled hoe was leaning against the end of the black net in the right front of where you are now.", + "original_options": { + "A": "Leaning at the black net’s side.", + "B": "Lying on the ground by the net.", + "C": "At that time, the long-handled hoe was leaning against the side of the black net to your left rear.", + "D": "At that time, the long-handled hoe was standing upright near the black net’s front corner, closer to your center.", + "E": "At that time, the long-handled hoe was leaning against the end of the black net in the right front of where you are now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 20 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_12", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I put down the long-handled hoe last time, where was the long-handled hoe placed?", + "answer": "I leaned the long-handled hoe against the black net adjacent to the black basket.", + "question_chinese": "上次我吧长柄锄头放下的时候，长柄锄头被放在哪里了？", + "answer_chinese": "我把长柄锄头靠在黑色篮子旁边的黑网上。", + "evidence": [ + { + "description": "I leaned the long-handled hoe against the black net adjacent to the black basket.", + "timestamps": { + "start": "00:19:40", + "end": "00:20:08" + } + } + ], + "question_timestamps": "00:29:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Against the net’s far end.", + "B": "On the wooden floor by the net.", + "C": "I placed the long-handled hoe on the soil a short distance from the black net, away from the basket.", + "D": "I leaned the long-handled hoe against the black net adjacent to the black basket.", + "E": "I leaned the long-handled hoe against the black net, but it was not adjacent to the black basket." + }, + "answer_label": "D", + "answer_text": "I leaned the long-handled hoe against the black net adjacent to the black basket.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_12", + "correct_answer": "I leaned the long-handled hoe against the black net adjacent to the black basket.", + "original_options": { + "A": "Against the net’s far end.", + "B": "On the wooden floor by the net.", + "C": "I placed the long-handled hoe on the soil a short distance from the black net, away from the basket.", + "D": "I leaned the long-handled hoe against the black net adjacent to the black basket.", + "E": "I leaned the long-handled hoe against the black net, but it was not adjacent to the black basket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 18 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_16", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of agricultural tools have appeared so far?", + "answer": "A long-handled hoe was present.", + "question_chinese": "到目前为止出现了哪几种农具？", + "answer_chinese": "出现过长柄锄头。", + "evidence": [ + { + "description": "Long-handled hoe", + "timestamps": { + "start": "00:28:37", + "end": "00:28:59" + } + } + ], + "question_timestamps": "00:29:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A medium-sized hoe was present.", + "B": "A long-handled hoe was present.", + "C": "A small shovel was present.", + "D": "A short-handled rake was present during the activity, rather than a long-handled hoe.", + "E": "A medium-sized rake appeared earlier, and no long-handled hoe showed up in the scene." + }, + "answer_label": "B", + "answer_text": "A long-handled hoe was present.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_16", + "correct_answer": "A long-handled hoe was present.", + "original_options": { + "A": "A medium-sized hoe was present.", + "B": "A long-handled hoe was present.", + "C": "A small shovel was present.", + "D": "A short-handled rake was present during the activity, rather than a long-handled hoe.", + "E": "A medium-sized rake appeared earlier, and no long-handled hoe showed up in the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_15", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of agricultural tools have appeared so far?", + "answer": "Long-handled hoe and small garden hoe have appeared.", + "question_chinese": "到目前为止出现了哪几种农具？", + "answer_chinese": "出现过长柄锄头和小园艺锄。", + "evidence": [ + { + "description": "Long-handled hoe", + "timestamps": { + "start": "00:28:37", + "end": "00:28:59" + } + }, + { + "description": "Small Garden Hoe", + "timestamps": { + "start": "00:36:54", + "end": "00:36:59" + } + } + ], + "question_timestamps": "00:38:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Long-handled hoe and small garden hoe have appeared.", + "B": "Long-handled hoe and hand cultivator appeared.", + "C": "Long-handled spade and small garden hoe appeared.", + "D": "Short-handled hoe and small garden fork have appeared so far, not the long-handled hoe combination.", + "E": "Long-handled hoe and pruning knife have appeared, rather than the small garden hoe mentioned in the correct set." + }, + "answer_label": "A", + "answer_text": "Long-handled hoe and small garden hoe have appeared.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_15", + "correct_answer": "Long-handled hoe and small garden hoe have appeared.", + "original_options": { + "A": "Long-handled hoe and small garden hoe have appeared.", + "B": "Long-handled hoe and hand cultivator appeared.", + "C": "Long-handled spade and small garden hoe appeared.", + "D": "Short-handled hoe and small garden fork have appeared so far, not the long-handled hoe combination.", + "E": "Long-handled hoe and pruning knife have appeared, rather than the small garden hoe mentioned in the correct set." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 18 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_5", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many holes have I drilled in the ground with wood so far?", + "answer": "You have already dug seven holes in the land.", + "question_chinese": "到目前为止我用木头在土地上打过几个洞？", + "answer_chinese": "你已经在土地上打过了七个洞。", + "evidence": [ + { + "description": "Seven holes have already been dug in the land.", + "timestamps": { + "start": "00:07:34", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:09:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have dug six holes.", + "B": "You have already dug seven holes in the land.", + "C": "You have dug eight holes.", + "D": "Up to now, you have dug nine holes in the land using the wood to make them.", + "E": "So far, you have dug ten holes in the land, counting each drilled spot you made with the wood." + }, + "answer_label": "B", + "answer_text": "You have already dug seven holes in the land.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_5", + "correct_answer": "You have already dug seven holes in the land.", + "original_options": { + "A": "You have dug six holes.", + "B": "You have already dug seven holes in the land.", + "C": "You have dug eight holes.", + "D": "Up to now, you have dug nine holes in the land using the wood to make them.", + "E": "So far, you have dug ten holes in the land, counting each drilled spot you made with the wood." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_6", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many holes have I drilled in the ground with wood so far?", + "answer": "You dug eleven holes in the land.", + "question_chinese": "到目前为止我用木头在土地上打过几个洞？", + "answer_chinese": "你在土地上打了十一个洞。", + "evidence": [ + { + "description": "Seven holes have already been dug in the land.", + "timestamps": { + "start": "00:07:34", + "end": "00:09:05" + } + }, + { + "description": "Dug four more holes in the ground.", + "timestamps": { + "start": "00:30:03", + "end": "00:31:04" + } + } + ], + "question_timestamps": "00:38:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You dug ten holes.", + "B": "You dug twelve holes.", + "C": "You dug eleven holes in the land.", + "D": "By this point, you dug thirteen holes in the land while drilling into the ground with the wood.", + "E": "So far, you have dug fourteen holes in the land, counting each time you drilled into the ground with wood." + }, + "answer_label": "C", + "answer_text": "You dug eleven holes in the land.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_6", + "correct_answer": "You dug eleven holes in the land.", + "original_options": { + "A": "You dug ten holes.", + "B": "You dug twelve holes.", + "C": "You dug eleven holes in the land.", + "D": "By this point, you dug thirteen holes in the land while drilling into the ground with the wood.", + "E": "So far, you have dug fourteen holes in the land, counting each time you drilled into the ground with wood." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 20 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_9", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many green ribbons tied to black tubes have I seen?", + "answer": "So far, you have seen five green ribbons tied to black tubes.", + "question_chinese": "到目前为止，我看到了多少个被绑在黑管子的绿带子？", + "answer_chinese": "到目前为止你看到了五个被绑在黑管子上的绿带子。", + "evidence": [ + { + "description": "The first green ribbon", + "timestamps": { + "start": "00:04:49", + "end": "00:04:52" + } + }, + { + "description": "The second green ribbon", + "timestamps": { + "start": "00:04:55", + "end": "00:04:58" + } + }, + { + "description": "The third and fourth green ribbons", + "timestamps": { + "start": "00:06:20", + "end": "00:06:26" + } + }, + { + "description": "The fifth green ribbon", + "timestamps": { + "start": "00:06:35", + "end": "00:06:39" + } + } + ], + "question_timestamps": "00:09:40", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "So far, you have seen five green ribbons tied to black tubes.", + "B": "You have seen four ribbons.", + "C": "You have seen six ribbons.", + "D": "So far, you have seen seven green ribbons tied to black tubes during your walk around.", + "E": "Up to now, you have seen eight green ribbons tied to black tubes, counting every one you noticed." + }, + "answer_label": "A", + "answer_text": "So far, you have seen five green ribbons tied to black tubes.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_9", + "correct_answer": "So far, you have seen five green ribbons tied to black tubes.", + "original_options": { + "A": "So far, you have seen five green ribbons tied to black tubes.", + "B": "You have seen four ribbons.", + "C": "You have seen six ribbons.", + "D": "So far, you have seen seven green ribbons tied to black tubes during your walk around.", + "E": "Up to now, you have seen eight green ribbons tied to black tubes, counting every one you noticed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_10", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many green ribbons tied to black tubes have I seen?", + "answer": "So far, you have seen seven green ribbons tied to black tubes.", + "question_chinese": "到目前为止，我看到了多少个被绑在黑管子的绿带子？", + "answer_chinese": "到目前为止，你看到了七个被绑在黑管子的绿带子。", + "evidence": [ + { + "description": "The first green ribbon", + "timestamps": { + "start": "00:04:49", + "end": "00:04:52" + } + }, + { + "description": "The second green ribbon", + "timestamps": { + "start": "00:04:55", + "end": "00:04:58" + } + }, + { + "description": "The third and fourth green ribbons", + "timestamps": { + "start": "00:06:20", + "end": "00:06:26" + } + }, + { + "description": "The fifth green ribbon", + "timestamps": { + "start": "00:06:35", + "end": "00:06:39" + } + }, + { + "description": "The sixth green ribbon", + "timestamps": { + "start": "00:23:13", + "end": "00:23:17" + } + }, + { + "description": "The seventh green ribbon", + "timestamps": { + "start": "00:24:33", + "end": "00:24:38" + } + } + ], + "question_timestamps": "00:29:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have seen six ribbons.", + "B": "You have seen eight ribbons.", + "C": "So far, you have seen nine green ribbons tied to black tubes, based on what you noticed.", + "D": "Up to now, you have seen ten green ribbons tied to black tubes as you moved through the area.", + "E": "So far, you have seen seven green ribbons tied to black tubes." + }, + "answer_label": "E", + "answer_text": "So far, you have seen seven green ribbons tied to black tubes.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_10", + "correct_answer": "So far, you have seen seven green ribbons tied to black tubes.", + "original_options": { + "A": "You have seen six ribbons.", + "B": "You have seen eight ribbons.", + "C": "So far, you have seen nine green ribbons tied to black tubes, based on what you noticed.", + "D": "Up to now, you have seen ten green ribbons tied to black tubes as you moved through the area.", + "E": "So far, you have seen seven green ribbons tied to black tubes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_8", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the small wooden board after I used it to dig a hole in the ground last time?", + "answer": "The small wooden board was thrown by you to the position symmetric to the second hole you shot regarding the black pipe, on the grass.", + "question_chinese": "我上次使用小木板在地上打洞以后，小木板在哪里？", + "answer_chinese": "小木板被你丢到了你所打的第二个洞关于黑管对称的位置上，在草地上.", + "evidence": [ + { + "description": "The small wooden board was thrown by you to the position symmetric to the second hole you shot regarding the black pipe, on the grass.", + "timestamps": { + "start": "00:08:35", + "end": "00:09:12" + } + } + ], + "question_timestamps": "00:09:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is beside the last hole.", + "B": "It is near the black pipe.", + "C": "You left the small wooden board on the soil, close to the black pipe and next to the first hole.", + "D": "You threw the small wooden board onto the grass on the same side of the black pipe as the second hole.", + "E": "The small wooden board was thrown by you to the position symmetric to the second hole you shot regarding the black pipe, on the grass." + }, + "answer_label": "E", + "answer_text": "The small wooden board was thrown by you to the position symmetric to the second hole you shot regarding the black pipe, on the grass.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_8", + "correct_answer": "The small wooden board was thrown by you to the position symmetric to the second hole you shot regarding the black pipe, on the grass.", + "original_options": { + "A": "It is beside the last hole.", + "B": "It is near the black pipe.", + "C": "You left the small wooden board on the soil, close to the black pipe and next to the first hole.", + "D": "You threw the small wooden board onto the grass on the same side of the black pipe as the second hole.", + "E": "The small wooden board was thrown by you to the position symmetric to the second hole you shot regarding the black pipe, on the grass." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 21 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_7", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the small wooden board after I used it to dig a hole in the ground last time?", + "answer": "It was on the soil, in the direction away from the black pipe, of the last hole you dug in the land.", + "question_chinese": "我上次使用小木板在地上打洞以后，小木板在哪里？", + "answer_chinese": "它在你在土地上打的最后一个洞的远离黑管的方向，在泥土上。", + "evidence": [ + { + "description": "It was on the soil, in the direction away from the black pipe, of the last hole you dug in the land.", + "timestamps": { + "start": "00:30:21", + "end": "00:30:54" + } + } + ], + "question_timestamps": "00:38:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was on the soil, in the direction away from the black pipe, of the last hole you dug in the land.", + "B": "On the grass near the black pipe.", + "C": "Beside the second hole on soil.", + "D": "It was resting on the soil between the black pipe and the first hole you dug in the land.", + "E": "It was on the soil close to the black pipe, next to the last hole you dug in the land." + }, + "answer_label": "A", + "answer_text": "It was on the soil, in the direction away from the black pipe, of the last hole you dug in the land.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_7", + "correct_answer": "It was on the soil, in the direction away from the black pipe, of the last hole you dug in the land.", + "original_options": { + "A": "It was on the soil, in the direction away from the black pipe, of the last hole you dug in the land.", + "B": "On the grass near the black pipe.", + "C": "Beside the second hole on soil.", + "D": "It was resting on the soil between the black pipe and the first hole you dug in the land.", + "E": "It was on the soil close to the black pipe, next to the last hole you dug in the land." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 20 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_7_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I passed by or stopped in front of the hanging clothes?", + "answer": "You passed by the hanging clothes to your left two times.", + "question_chinese": "我经过或停留在被悬挂的衣服前几次了？", + "answer_chinese": "你经过了你左边悬挂的衣服2次。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:07:05", + "end": "00:07:14" + } + }, + { + "description": "The second time.", + "timestamps": { + "start": "00:19:41", + "end": "00:20:09" + } + } + ], + "question_timestamps": "00:24:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You passed them zero times.", + "B": "You passed by the hanging clothes to your left two times.", + "C": "You passed them one time.", + "D": "You passed by the hanging clothes to your left three times while moving through the area.", + "E": "You passed by or stopped in front of the hanging clothes to your left four times during your route." + }, + "answer_label": "B", + "answer_text": "You passed by the hanging clothes to your left two times.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_7_T2", + "correct_answer": "You passed by the hanging clothes to your left two times.", + "original_options": { + "A": "You passed them zero times.", + "B": "You passed by the hanging clothes to your left two times.", + "C": "You passed them one time.", + "D": "You passed by the hanging clothes to your left three times while moving through the area.", + "E": "You passed by or stopped in front of the hanging clothes to your left four times during your route." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 19 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_7_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I passed by or stopped in front of the hanging clothes?", + "answer": "So far, you have passed by or stopped in front of the hanging clothes to your left three times.", + "question_chinese": "我经过或停留在被悬挂的衣服前几次了？", + "answer_chinese": "到目前为止，你经过或停留在你左边被悬挂的衣服前三次了。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:07:05", + "end": "00:07:14" + } + }, + { + "description": "The second time.", + "timestamps": { + "start": "00:19:41", + "end": "00:20:09" + } + }, + { + "description": "Third time.", + "timestamps": { + "start": "00:20:50", + "end": "00:21:03" + } + } + ], + "question_timestamps": "00:29:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, you have passed by or stopped in front of the hanging clothes to your left three times.", + "B": "You did it one time.", + "C": "You did it two times.", + "D": "So far, you have passed by or stopped in front of the hanging clothes to your left four times.", + "E": "Up to now, you have passed by or stopped in front of the hanging clothes to your left five times." + }, + "answer_label": "A", + "answer_text": "So far, you have passed by or stopped in front of the hanging clothes to your left three times.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_7_T1", + "correct_answer": "So far, you have passed by or stopped in front of the hanging clothes to your left three times.", + "original_options": { + "A": "So far, you have passed by or stopped in front of the hanging clothes to your left three times.", + "B": "You did it one time.", + "C": "You did it two times.", + "D": "So far, you have passed by or stopped in front of the hanging clothes to your left four times.", + "E": "Up to now, you have passed by or stopped in front of the hanging clothes to your left five times." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 20 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the seed bag?", + "answer": "On the wooden floor next to the black net.", + "question_chinese": "种子袋在哪里？", + "answer_chinese": "在黑网旁边的木制地板上，在你现在位置的右前方", + "evidence": [ + { + "description": "The seed bag is on the wooden floor next to the black net.", + "timestamps": { + "start": "00:09:21", + "end": "00:09:39" + } + } + ], + "question_timestamps": "00:09:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the wooden floor by the blue net.", + "B": "On the wooden floor beside the black basket.", + "C": "On the wooden floor next to the black basket, right where you set it down near the netting.", + "D": "On the wooden floor next to the black net.", + "E": "On the wooden floor near the black net, but closer to the shelf than where you last looked." + }, + "answer_label": "D", + "answer_text": "On the wooden floor next to the black net.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_1", + "correct_answer": "On the wooden floor next to the black net.", + "original_options": { + "A": "On the wooden floor by the blue net.", + "B": "On the wooden floor beside the black basket.", + "C": "On the wooden floor next to the black basket, right where you set it down near the netting.", + "D": "On the wooden floor next to the black net.", + "E": "On the wooden floor near the black net, but closer to the shelf than where you last looked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394", + "video_length": "00:38:43", + "qid": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the seed bag?", + "answer": "In the jar next to the black basket, between the hoe and the black net.", + "question_chinese": "种子袋在哪里？", + "answer_chinese": "在黑色篮子旁边的罐子里，位于锄头和黑色网之间。", + "evidence": [ + { + "description": "In the jar next to the black basket, between the hoe and the black net.", + "timestamps": { + "start": "00:19:39", + "end": "00:20:02" + } + } + ], + "question_timestamps": "00:29:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Between the hoe and the rake.", + "B": "In the black basket.", + "C": "In the jar next to the black basket, between the hoe and the brown mat instead of the black net.", + "D": "In the jar next to the black basket, between the rake and the black net.", + "E": "In the jar next to the black basket, between the hoe and the black net." + }, + "answer_label": "E", + "answer_text": "In the jar next to the black basket, between the hoe and the black net.", + "q_id": "3b5cd89f-6f74-4077-85ca-9a64b1ffa394_2", + "correct_answer": "In the jar next to the black basket, between the hoe and the black net.", + "original_options": { + "A": "Between the hoe and the rake.", + "B": "In the black basket.", + "C": "In the jar next to the black basket, between the hoe and the brown mat instead of the black net.", + "D": "In the jar next to the black basket, between the rake and the black net.", + "E": "In the jar next to the black basket, between the hoe and the black net." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 20, + 15 + ] + } +] diff --git a/QAs/3b989129-a310-4aa9-85f5-e5e9a90df99b-mcq.json b/QAs/3b989129-a310-4aa9-85f5-e5e9a90df99b-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..43544a9939d80eef4812eb4fc1ddee6d8af98ee8 --- /dev/null +++ b/QAs/3b989129-a310-4aa9-85f5-e5e9a90df99b-mcq.json @@ -0,0 +1,1510 @@ +[ + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_3_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "There are four eggs in the refrigerator now—how many did I take out earlier?", + "answer": "Earlier, you took out five in total.", + "question_chinese": "冰箱现在有四个鸡蛋，之前我拿出去了几个？", + "answer_chinese": "之前你拿出去了五个。", + "evidence": [ + { + "description": "There were originally nine eggs in the fridge.", + "timestamps": { + "start": "00:12:36", + "end": "00:12:40" + } + }, + { + "description": "I took five eggs from the fridge.", + "timestamps": { + "start": "00:12:41", + "end": "00:12:50" + } + }, + { + "description": "I took an egg from the fridge.", + "timestamps": { + "start": "00:23:02", + "end": "00:23:10" + } + } + ], + "question_timestamps": "00:26:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Earlier, you took out three in total.", + "B": "Earlier, you took out four in total.", + "C": "Earlier, you took out six in total, leaving four eggs still sitting in the refrigerator afterward.", + "D": "Earlier, you took out seven in total, which would explain why only four eggs remain in the refrigerator now.", + "E": "Earlier, you took out five in total." + }, + "answer_label": "E", + "answer_text": "Earlier, you took out five in total.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_3_T1", + "correct_answer": "Earlier, you took out five in total.", + "original_options": { + "A": "Earlier, you took out three in total.", + "B": "Earlier, you took out four in total.", + "C": "Earlier, you took out six in total, leaving four eggs still sitting in the refrigerator afterward.", + "D": "Earlier, you took out seven in total, which would explain why only four eggs remain in the refrigerator now.", + "E": "Earlier, you took out five in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 19 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_5", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "There are four eggs in the fridge now. How many did I take out earlier?", + "answer": "You took out six before.", + "question_chinese": "冰箱现在有四个鸡蛋，之前我拿出去了几个？", + "answer_chinese": "之前你拿出去了六个。", + "evidence": [ + { + "description": "There were originally nine eggs in the fridge.", + "timestamps": { + "start": "00:12:36", + "end": "00:12:40" + } + }, + { + "description": "I took five eggs from the fridge.", + "timestamps": { + "start": "00:12:41", + "end": "00:12:50" + } + }, + { + "description": "There are only three eggs left in the fridge now.", + "timestamps": { + "start": "00:26:52", + "end": "00:26:55" + } + }, + { + "description": "I took an egg from the fridge.", + "timestamps": { + "start": "00:23:02", + "end": "00:23:10" + } + } + ], + "question_timestamps": "00:26:57", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You took out four before.", + "B": "You took out five before.", + "C": "You took out three before, and later four eggs were left in the fridge.", + "D": "You took out six before.", + "E": "You took out seven before, which is why the fridge now has four eggs remaining." + }, + "answer_label": "D", + "answer_text": "You took out six before.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_5", + "correct_answer": "You took out six before.", + "original_options": { + "A": "You took out four before.", + "B": "You took out five before.", + "C": "You took out three before, and later four eggs were left in the fridge.", + "D": "You took out six before.", + "E": "You took out seven before, which is why the fridge now has four eggs remaining." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_3_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "There are four eggs in the refrigerator now—how many did I take out earlier?", + "answer": "Earlier, you took out six in total.", + "question_chinese": "冰箱现在有四个鸡蛋，之前我拿出去了几个？", + "answer_chinese": "之前你一共拿出去了六个。", + "evidence": [ + { + "description": "There are only three eggs left in the fridge now.", + "timestamps": { + "start": "00:26:32", + "end": "00:26:55" + } + }, + { + "description": "I took an egg from the fridge.", + "timestamps": { + "start": "00:23:02", + "end": "00:23:10" + } + }, + { + "description": "I took five eggs from the fridge.", + "timestamps": { + "start": "00:12:41", + "end": "00:12:50" + } + }, + { + "description": "There were originally nine eggs in the fridge.", + "timestamps": { + "start": "00:12:36", + "end": "00:12:40" + } + } + ], + "question_timestamps": "00:29:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Earlier, you took out six in total.", + "B": "Earlier, you took out four in total.", + "C": "Earlier, you took out five in total.", + "D": "Earlier, you took out seven in total, so there would still be four eggs left in the refrigerator now.", + "E": "Earlier, you took out eight in total, which would match seeing only four eggs remaining in the refrigerator." + }, + "answer_label": "A", + "answer_text": "Earlier, you took out six in total.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_3_T2", + "correct_answer": "Earlier, you took out six in total.", + "original_options": { + "A": "Earlier, you took out six in total.", + "B": "Earlier, you took out four in total.", + "C": "Earlier, you took out five in total.", + "D": "Earlier, you took out seven in total, so there would still be four eggs left in the refrigerator now.", + "E": "Earlier, you took out eight in total, which would match seeing only four eggs remaining in the refrigerator." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 18 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_7", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are the oven mitts when I stand by the stove?", + "answer": "When you stand by the stove, the oven mitt is on the left side of the stove.", + "question_chinese": "当我站在灶台旁边时，隔热手套在哪里？", + "answer_chinese": "当你站在灶台旁边时，隔热手套在灶台的左边。", + "evidence": [ + { + "description": "The oven mitt gloves are on the left side of the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:06:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the right side of the stove.", + "B": "On the countertop beside the stove.", + "C": "When you stand by the stove, the oven mitt is on the floor directly in front of the stove.", + "D": "When you stand by the stove, the oven mitt is tucked under the stove near the wall.", + "E": "When you stand by the stove, the oven mitt is on the left side of the stove." + }, + "answer_label": "E", + "answer_text": "When you stand by the stove, the oven mitt is on the left side of the stove.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_7", + "correct_answer": "When you stand by the stove, the oven mitt is on the left side of the stove.", + "original_options": { + "A": "On the right side of the stove.", + "B": "On the countertop beside the stove.", + "C": "When you stand by the stove, the oven mitt is on the floor directly in front of the stove.", + "D": "When you stand by the stove, the oven mitt is tucked under the stove near the wall.", + "E": "When you stand by the stove, the oven mitt is on the left side of the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 17 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_6", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are the oven mitts when I stand by the stove?", + "answer": "When you stand by the stove, the oven mitt is on the right side of the stove.", + "question_chinese": "当我站在灶台旁边时，隔热手套在哪里？", + "answer_chinese": "当你站在灶台旁边时，隔热手套在灶台的右边。", + "evidence": [ + { + "description": "I took the oven mitts to the right side of the stove.", + "timestamps": { + "start": "00:17:43", + "end": "00:17:48" + } + }, + { + "description": "The oven mitt are on the left side of the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:29:37", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of the stove.", + "B": "Behind the stove.", + "C": "When you stand by the stove, the oven mitt is on the right side of the stove.", + "D": "When you stand by the stove, the oven mitt is to your left of the stove.", + "E": "When you stand by the stove, the oven mitt sits behind you rather than on either side." + }, + "answer_label": "C", + "answer_text": "When you stand by the stove, the oven mitt is on the right side of the stove.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_6", + "correct_answer": "When you stand by the stove, the oven mitt is on the right side of the stove.", + "original_options": { + "A": "In front of the stove.", + "B": "Behind the stove.", + "C": "When you stand by the stove, the oven mitt is on the right side of the stove.", + "D": "When you stand by the stove, the oven mitt is to your left of the stove.", + "E": "When you stand by the stove, the oven mitt sits behind you rather than on either side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 16, + 17 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_6_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "To get to the refrigerator, how many turns do I need to make?", + "answer": "To get to the fridge, you don't need to make any turns.", + "question_chinese": "我去冰箱那里需要转多少个弯？", + "answer_chinese": "你去冰箱那里不需要转弯。", + "evidence": [ + { + "description": "The fridge is to the left of the stove.", + "timestamps": { + "start": "00:12:30", + "end": "00:12:35" + } + } + ], + "question_timestamps": "00:20:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You need to make one turn.", + "B": "You need to make two turns.", + "C": "To get to the fridge, you don't need to make any turns.", + "D": "First you should turn around, then walk straight to the fridge from where you are standing.", + "E": "Turn left first and then continue straight until you reach the refrigerator." + }, + "answer_label": "C", + "answer_text": "To get to the fridge, you don't need to make any turns.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_6_T1", + "correct_answer": "To get to the fridge, you don't need to make any turns.", + "original_options": { + "A": "You need to make one turn.", + "B": "You need to make two turns.", + "C": "To get to the fridge, you don't need to make any turns.", + "D": "First you should turn around, then walk straight to the fridge from where you are standing.", + "E": "Turn left first and then continue straight until you reach the refrigerator." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 12 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_6_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "To get to the refrigerator, how many turns do I need to make?", + "answer": "You were standing in front of the stove, and to get to the fridge you need to make a turn.", + "question_chinese": "我去冰箱那里需要转多少个弯？", + "answer_chinese": "你站在灶台前，去冰箱那里需要转一个弯。", + "evidence": [ + { + "description": "The fridge is to the left of the stove.", + "timestamps": { + "start": "00:12:30", + "end": "00:12:35" + } + } + ], + "question_timestamps": "00:25:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You don't need to turn.", + "B": "You need to make two turns.", + "C": "You were in front of the stove, and you need to make two turns to reach the refrigerator.", + "D": "You were standing in front of the stove, and to get to the fridge you need to make a turn.", + "E": "You were in front of the stove, and you need to turn around first before heading to the refrigerator." + }, + "answer_label": "D", + "answer_text": "You were standing in front of the stove, and to get to the fridge you need to make a turn.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_6_T2", + "correct_answer": "You were standing in front of the stove, and to get to the fridge you need to make a turn.", + "original_options": { + "A": "You don't need to turn.", + "B": "You need to make two turns.", + "C": "You were in front of the stove, and you need to make two turns to reach the refrigerator.", + "D": "You were standing in front of the stove, and to get to the fridge you need to make a turn.", + "E": "You were in front of the stove, and you need to turn around first before heading to the refrigerator." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 19 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_1_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk to the sink?", + "answer": "First, you should turn around.", + "question_chinese": "我去水槽是怎么走的？", + "answer_chinese": "你先向后转。", + "evidence": [ + { + "description": "The sink is at the rear right of the stove.", + "timestamps": { + "start": "00:12:10", + "end": "00:12:17" + } + } + ], + "question_timestamps": "00:15:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go straight to the sink.", + "B": "Turn left, then walk forward.", + "C": "Walk forward first, then turn left and continue toward the sink until you reach it.", + "D": "First, you should turn around.", + "E": "First turn around, then take a right and continue straight to the sink." + }, + "answer_label": "D", + "answer_text": "First, you should turn around.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_1_T1", + "correct_answer": "First, you should turn around.", + "original_options": { + "A": "Go straight to the sink.", + "B": "Turn left, then walk forward.", + "C": "Walk forward first, then turn left and continue toward the sink until you reach it.", + "D": "First, you should turn around.", + "E": "First turn around, then take a right and continue straight to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 13 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_1_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk to the sink?", + "answer": "First, you should turn around, then go straight.", + "question_chinese": "我去水槽是怎么走的？", + "answer_chinese": "你先向后转然后直走。", + "evidence": [ + { + "description": "The sink is at the rear right of the stove.", + "timestamps": { + "start": "00:12:10", + "end": "00:12:17" + } + } + ], + "question_timestamps": "00:26:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First, go straight ahead.", + "B": "Turn around, then turn left.", + "C": "First, you should turn around, then take a right before walking forward to the sink.", + "D": "You should turn left first, and then go straight until you reach the sink.", + "E": "First, you should turn around, then go straight." + }, + "answer_label": "E", + "answer_text": "First, you should turn around, then go straight.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_1_T2", + "correct_answer": "First, you should turn around, then go straight.", + "original_options": { + "A": "First, go straight ahead.", + "B": "Turn around, then turn left.", + "C": "First, you should turn around, then take a right before walking forward to the sink.", + "D": "You should turn left first, and then go straight until you reach the sink.", + "E": "First, you should turn around, then go straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 14 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_17", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tables appeared in the scene while I was beating eggs?", + "answer": "When you were beating the eggs, a table appeared in the scene.", + "question_chinese": "我在打鸡蛋时，场景里出现了几张桌子？", + "answer_chinese": "你在打鸡蛋时，场景里出现了一张桌子。", + "evidence": [ + { + "description": "There is a table next to the shelf for bowls.", + "timestamps": { + "start": "00:13:20", + "end": "00:13:26" + } + } + ], + "question_timestamps": "00:16:34", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two tables appeared.", + "B": "No tables appeared.", + "C": "While you were beating the eggs, two tables appeared in the scene at the same time.", + "D": "When you were beating the eggs, a table appeared in the scene.", + "E": "While you were beating the eggs, three tables appeared in the scene during that moment." + }, + "answer_label": "D", + "answer_text": "When you were beating the eggs, a table appeared in the scene.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_17", + "correct_answer": "When you were beating the eggs, a table appeared in the scene.", + "original_options": { + "A": "Two tables appeared.", + "B": "No tables appeared.", + "C": "While you were beating the eggs, two tables appeared in the scene at the same time.", + "D": "When you were beating the eggs, a table appeared in the scene.", + "E": "While you were beating the eggs, three tables appeared in the scene during that moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 15 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_18", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tables appeared in the scene while I was beating eggs?", + "answer": "When you were beating the eggs, a table appeared in the scene.", + "question_chinese": "我在打鸡蛋时，场景里出现了几张桌子？", + "answer_chinese": "你在打鸡蛋时，场景里出现了一张桌子。", + "evidence": [ + { + "description": "There is a table next to the shelf for bowls.", + "timestamps": { + "start": "00:13:20", + "end": "00:13:26" + } + } + ], + "question_timestamps": "00:27:11", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero tables appeared.", + "B": "Two tables appeared.", + "C": "When you were beating the eggs, a table appeared in the scene.", + "D": "When you were beating the eggs, three tables appeared in the scene during that time.", + "E": "When you were beating the eggs, four tables appeared in the scene all at once." + }, + "answer_label": "C", + "answer_text": "When you were beating the eggs, a table appeared in the scene.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_18", + "correct_answer": "When you were beating the eggs, a table appeared in the scene.", + "original_options": { + "A": "Zero tables appeared.", + "B": "Two tables appeared.", + "C": "When you were beating the eggs, a table appeared in the scene.", + "D": "When you were beating the eggs, three tables appeared in the scene during that time.", + "E": "When you were beating the eggs, four tables appeared in the scene all at once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_9_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tables appeared in my scene?", + "answer": "When you were cracking eggs, a table appeared in the scene, and it was in front of you.", + "question_chinese": "我场景里出现了几张桌子？", + "answer_chinese": "你在打鸡蛋时，场景里出现了一张桌子，桌子在你前面。", + "evidence": [ + { + "description": "There is a table next to the shelf for bowls.", + "timestamps": { + "start": "00:13:20", + "end": "00:13:26" + } + } + ], + "question_timestamps": "00:16:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "When you were cracking eggs, a table appeared in the scene, and it was in front of you.", + "B": "No table appeared.", + "C": "Two tables appeared in front.", + "D": "When you were cracking eggs, a table appeared in the scene, but it was behind you.", + "E": "When you were cracking eggs, two tables appeared in the scene, and they were both behind you." + }, + "answer_label": "A", + "answer_text": "When you were cracking eggs, a table appeared in the scene, and it was in front of you.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_9_T1", + "correct_answer": "When you were cracking eggs, a table appeared in the scene, and it was in front of you.", + "original_options": { + "A": "When you were cracking eggs, a table appeared in the scene, and it was in front of you.", + "B": "No table appeared.", + "C": "Two tables appeared in front.", + "D": "When you were cracking eggs, a table appeared in the scene, but it was behind you.", + "E": "When you were cracking eggs, two tables appeared in the scene, and they were both behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 5, + 16, + 17 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_9_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tables appeared in my scene?", + "answer": "When you were tearing cheese, no table appeared in the scene.", + "question_chinese": "我场景里出现了几张桌子？", + "answer_chinese": "你在撕奶酪时场景里没有出现了张桌子。", + "evidence": [ + { + "description": "There is a table next to the shelf for bowls.", + "timestamps": { + "start": "00:13:20", + "end": "00:13:26" + } + } + ], + "question_timestamps": "00:27:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A table appeared in front.", + "B": "When you were tearing cheese, no table appeared in the scene.", + "C": "Two tables appeared in front.", + "D": "When you were tearing cheese, a table appeared in the scene, and it was behind you.", + "E": "When you were tearing cheese, two tables appeared in the scene, and they were both behind you." + }, + "answer_label": "B", + "answer_text": "When you were tearing cheese, no table appeared in the scene.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_9_T2", + "correct_answer": "When you were tearing cheese, no table appeared in the scene.", + "original_options": { + "A": "A table appeared in front.", + "B": "When you were tearing cheese, no table appeared in the scene.", + "C": "Two tables appeared in front.", + "D": "When you were tearing cheese, a table appeared in the scene, and it was behind you.", + "E": "When you were tearing cheese, two tables appeared in the scene, and they were both behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_3", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where should I go if I want more kitchen paper?", + "answer": "You should get it from the cabinet above the kitchen counter.", + "question_chinese": "我想要更多的厨房用纸应该去哪里？", + "answer_chinese": "你应该去料理台上方的橱柜拿。", + "evidence": [ + { + "description": "Kitchen paper towels are in the cabinet above the cooking counter.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:03" + } + } + ], + "question_timestamps": "00:11:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should get it from the cabinet above the kitchen counter.", + "B": "Get it from the cabinet by the sink.", + "C": "Get it from the cabinet near the stove.", + "D": "You should get it from the cabinet above the kitchen island, not the one above the counter.", + "E": "You should get it from the cabinet above the kitchen table rather than the cabinet above the counter." + }, + "answer_label": "A", + "answer_text": "You should get it from the cabinet above the kitchen counter.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_3", + "correct_answer": "You should get it from the cabinet above the kitchen counter.", + "original_options": { + "A": "You should get it from the cabinet above the kitchen counter.", + "B": "Get it from the cabinet by the sink.", + "C": "Get it from the cabinet near the stove.", + "D": "You should get it from the cabinet above the kitchen island, not the one above the counter.", + "E": "You should get it from the cabinet above the kitchen table rather than the cabinet above the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 18 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_2_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want more kitchen paper towels, where should I go?", + "answer": "You should go to the first cabinet in the bottom-left corner to get it.", + "question_chinese": "我想要更多的厨房用纸应该去哪里？", + "answer_chinese": "你应该去左下角的第一个橱柜拿。", + "evidence": [ + { + "description": "Kitchen paper towels are in the cabinet above the cooking counter.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:03" + } + } + ], + "question_timestamps": "00:12:015", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go to the bottom-right cabinet.", + "B": "Check the cabinet above you.", + "C": "You should go to the first cabinet in the bottom-left corner to get it.", + "D": "Go to the second cabinet in the bottom-left corner and take the paper towels from there.", + "E": "Walk to the cabinet under the sink on the left side and grab the kitchen paper towels." + }, + "answer_label": "C", + "answer_text": "You should go to the first cabinet in the bottom-left corner to get it.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_2_T1", + "correct_answer": "You should go to the first cabinet in the bottom-left corner to get it.", + "original_options": { + "A": "Go to the bottom-right cabinet.", + "B": "Check the cabinet above you.", + "C": "You should go to the first cabinet in the bottom-left corner to get it.", + "D": "Go to the second cabinet in the bottom-left corner and take the paper towels from there.", + "E": "Walk to the cabinet under the sink on the left side and grab the kitchen paper towels." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_2_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want more kitchen paper towels, where should I go?", + "answer": "You should go to the cabinet above you to your right to get it.", + "question_chinese": "我想要更多的厨房用纸应该去哪里？", + "answer_chinese": "你应该去右边上方的橱柜拿。", + "evidence": [ + { + "description": "Kitchen paper towels are in the cabinet above the cooking counter.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:03" + } + } + ], + "question_timestamps": "00:25:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You should go to the cabinet above you to your right to get it.", + "B": "Go to the cabinet above your left.", + "C": "Check the cabinet above you in front.", + "D": "Turn around and get the kitchen paper towels from the cabinet above you behind you.", + "E": "Reach up to the cabinet directly above you, not the one on your right, for paper towels." + }, + "answer_label": "A", + "answer_text": "You should go to the cabinet above you to your right to get it.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_2_T2", + "correct_answer": "You should go to the cabinet above you to your right to get it.", + "original_options": { + "A": "You should go to the cabinet above you to your right to get it.", + "B": "Go to the cabinet above your left.", + "C": "Check the cabinet above you in front.", + "D": "Turn around and get the kitchen paper towels from the cabinet above you behind you.", + "E": "Reach up to the cabinet directly above you, not the one on your right, for paper towels." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 17 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_5_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want more bowls, where should I go to get them?", + "answer": "You should go to the storage cabinet to your right to get it.", + "question_chinese": "我想要更多的碗应该去哪里拿？", + "answer_chinese": "应该去你的右边的收纳柜那里拿。", + "evidence": [ + { + "description": "The bowl was placed in the storage cabinet under the lower right side of the sink.", + "timestamps": { + "start": "00:12:50", + "end": "00:12:57" + } + } + ], + "question_timestamps": "00:14:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go to the cabinet in front.", + "B": "Go to the cabinet on your left.", + "C": "You should go to the storage cabinet behind you to get more bowls.", + "D": "Look up and take the bowls from the storage cabinet above you instead of the one to your right.", + "E": "You should go to the storage cabinet to your right to get it." + }, + "answer_label": "E", + "answer_text": "You should go to the storage cabinet to your right to get it.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_5_T1", + "correct_answer": "You should go to the storage cabinet to your right to get it.", + "original_options": { + "A": "Go to the cabinet in front.", + "B": "Go to the cabinet on your left.", + "C": "You should go to the storage cabinet behind you to get more bowls.", + "D": "Look up and take the bowls from the storage cabinet above you instead of the one to your right.", + "E": "You should go to the storage cabinet to your right to get it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 19 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_10", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I go to get more bowls?", + "answer": "You should get it from the storage cabinet under the lower right side of the sink.", + "question_chinese": "我想要更多的碗应该去哪里拿？", + "answer_chinese": "应该去水槽右下方的收纳柜那里拿。", + "evidence": [ + { + "description": "The bowl was placed in the storage cabinet under the lower right side of the sink.", + "timestamps": { + "start": "00:12:50", + "end": "00:12:57" + } + } + ], + "question_timestamps": "00:23:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the cabinet under the sink’s left.", + "B": "From the shelf above the sink.", + "C": "You should get it from the storage cabinet under the lower right side of the sink.", + "D": "You should get the bowls from the cabinet above the sink on the right side, not from below.", + "E": "You should take the bowls from the storage cabinet under the lower left side of the sink instead." + }, + "answer_label": "C", + "answer_text": "You should get it from the storage cabinet under the lower right side of the sink.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_10", + "correct_answer": "You should get it from the storage cabinet under the lower right side of the sink.", + "original_options": { + "A": "From the cabinet under the sink’s left.", + "B": "From the shelf above the sink.", + "C": "You should get it from the storage cabinet under the lower right side of the sink.", + "D": "You should get the bowls from the cabinet above the sink on the right side, not from below.", + "E": "You should take the bowls from the storage cabinet under the lower left side of the sink instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 18, + 18 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_5_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want more bowls, where should I go to get them?", + "answer": "You should first turn around, then walk straight and get it from the storage cabinet.", + "question_chinese": "我想要更多的碗应该去哪里拿？", + "answer_chinese": "你应该先向后转，然后直走到收纳柜里拿。", + "evidence": [ + { + "description": "The bowl was placed in the storage cabinet under the lower right side of the sink.", + "timestamps": { + "start": "00:12:50", + "end": "00:12:57" + } + } + ], + "question_timestamps": "00:23:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk straight without turning around.", + "B": "Turn left, then go to the cabinet.", + "C": "First turn around, then take a right and get the bowls from the storage cabinet.", + "D": "You should first turn around, then walk straight and get it from the storage cabinet.", + "E": "Turn around, walk straight past the cabinet, and get the bowls from the next storage cabinet." + }, + "answer_label": "D", + "answer_text": "You should first turn around, then walk straight and get it from the storage cabinet.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_5_T2", + "correct_answer": "You should first turn around, then walk straight and get it from the storage cabinet.", + "original_options": { + "A": "Walk straight without turning around.", + "B": "Turn left, then go to the cabinet.", + "C": "First turn around, then take a right and get the bowls from the storage cabinet.", + "D": "You should first turn around, then walk straight and get it from the storage cabinet.", + "E": "Turn around, walk straight past the cabinet, and get the bowls from the next storage cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 15, + 16 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_8_T1", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many cartons of milk did I use?", + "answer": "You used three cans.", + "question_chinese": "我用了几罐牛奶？", + "answer_chinese": "你用了三罐", + "evidence": [ + { + "description": "You poured a whole can of milk into the container.", + "timestamps": { + "start": "00:17:38", + "end": "00:20:40" + } + } + ], + "question_timestamps": "00:22:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You used one can.", + "B": "You used two cans.", + "C": "You used four cans altogether, not three, to finish the milk off completely.", + "D": "You used five cans in total, counting every milk carton you opened.", + "E": "You used three cans." + }, + "answer_label": "E", + "answer_text": "You used three cans.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_8_T1", + "correct_answer": "You used three cans.", + "original_options": { + "A": "You used one can.", + "B": "You used two cans.", + "C": "You used four cans altogether, not three, to finish the milk off completely.", + "D": "You used five cans in total, counting every milk carton you opened.", + "E": "You used three cans." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 12 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_8_T2", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many cartons of milk did I use?", + "answer": "You used four cans in total.", + "question_chinese": "我用了几罐牛奶？", + "answer_chinese": "你一共用了四罐", + "evidence": [ + { + "description": "You poured three cans of milk into the container.", + "timestamps": { + "start": "00:17:38", + "end": "00:20:40" + } + }, + { + "description": "You poured a whole can of milk into the container.", + "timestamps": { + "start": "00:20:46", + "end": "00:23:05" + } + } + ], + "question_timestamps": "00:25:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You used four cans in total.", + "B": "You used two cans in total.", + "C": "You used three cans in total.", + "D": "You used five cans in total, counting each carton you went through.", + "E": "You used six cans in total, not four, by the time you were done." + }, + "answer_label": "A", + "answer_text": "You used four cans in total.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_8_T2", + "correct_answer": "You used four cans in total.", + "original_options": { + "A": "You used four cans in total.", + "B": "You used two cans in total.", + "C": "You used three cans in total.", + "D": "You used five cans in total, counting each carton you went through.", + "E": "You used six cans in total, not four, by the time you were done." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 14 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_11", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I stand in front of the stove, how many turns do I need to take to get to the fridge?", + "answer": "You stand in front of the stove, and need to turn one more corner to get to the fridge.", + "question_chinese": "我站在灶台前，去冰箱那里需要转多少个弯？", + "answer_chinese": "你站在灶台前，去冰箱那里需要转多一个弯。", + "evidence": [ + { + "description": "The fridge is to the left of the stove.", + "timestamps": { + "start": "00:12:30", + "end": "00:12:35" + } + } + ], + "question_timestamps": "00:15:21", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You need zero more turns.", + "B": "You need two more turns.", + "C": "You stand in front of the stove, and need to turn one more corner to get to the fridge.", + "D": "You stand in front of the stove, and need to turn three more corners to get to the fridge.", + "E": "You stand in front of the stove, and need to turn four more corners before reaching the fridge." + }, + "answer_label": "C", + "answer_text": "You stand in front of the stove, and need to turn one more corner to get to the fridge.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_11", + "correct_answer": "You stand in front of the stove, and need to turn one more corner to get to the fridge.", + "original_options": { + "A": "You need zero more turns.", + "B": "You need two more turns.", + "C": "You stand in front of the stove, and need to turn one more corner to get to the fridge.", + "D": "You stand in front of the stove, and need to turn three more corners to get to the fridge.", + "E": "You stand in front of the stove, and need to turn four more corners before reaching the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 18 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_12", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I stand in front of the stove, how many turns do I need to take to get to the fridge?", + "answer": "You stand in front of the stove, and need to turn one more corner to get to the fridge.", + "question_chinese": "我站在灶台前，去冰箱那里需要转多少个弯？", + "answer_chinese": "你站在灶台前，去冰箱那里需要转多一个弯。", + "evidence": [ + { + "description": "The fridge is to the left of the stove.", + "timestamps": { + "start": "00:12:30", + "end": "00:12:35" + } + } + ], + "question_timestamps": "00:25:21", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You stand in front of the stove, and need to turn one more corner to get to the fridge.", + "B": "Two more corners.", + "C": "Three more corners.", + "D": "From the stove, you would need to turn two more corners before reaching the fridge.", + "E": "Standing at the stove, you must turn three more corners to arrive at the fridge." + }, + "answer_label": "A", + "answer_text": "You stand in front of the stove, and need to turn one more corner to get to the fridge.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_12", + "correct_answer": "You stand in front of the stove, and need to turn one more corner to get to the fridge.", + "original_options": { + "A": "You stand in front of the stove, and need to turn one more corner to get to the fridge.", + "B": "Two more corners.", + "C": "Three more corners.", + "D": "From the stove, you would need to turn two more corners before reaching the fridge.", + "E": "Standing at the stove, you must turn three more corners to arrive at the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_1", + "qid_num": 24, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I stood in front of the stove, how do I get to the sink?", + "answer": "You should turn around and then go straight.", + "question_chinese": "我站在灶台前，想去水槽怎么走？", + "answer_chinese": "你应该向后转然后直走。", + "evidence": [ + { + "description": "The sink is at the rear right of the stove.", + "timestamps": { + "start": "00:12:10", + "end": "00:12:17" + } + } + ], + "question_timestamps": "00:18:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn around, then go left.", + "B": "You should turn around and then go straight.", + "C": "Go straight without turning.", + "D": "Turn left from the stove, then continue straight until you reach the sink.", + "E": "Turn right first, then walk straight ahead to get to the sink." + }, + "answer_label": "B", + "answer_text": "You should turn around and then go straight.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_1", + "correct_answer": "You should turn around and then go straight.", + "original_options": { + "A": "Turn around, then go left.", + "B": "You should turn around and then go straight.", + "C": "Go straight without turning.", + "D": "Turn left from the stove, then continue straight until you reach the sink.", + "E": "Turn right first, then walk straight ahead to get to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 12 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_2", + "qid_num": 25, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I stood in front of the stove, how do I get to the sink?", + "answer": "You should turn around and then go straight.", + "question_chinese": "我站在灶台前，想去水槽怎么走？", + "answer_chinese": "你应该向后转然后直走。", + "evidence": [ + { + "description": "The sink is at the rear right of the stove.", + "timestamps": { + "start": "00:12:10", + "end": "00:12:17" + } + } + ], + "question_timestamps": "00:26:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn around, then go left.", + "B": "Turn right, then go straight.", + "C": "Step backward first, then go straight forward until you reach the sink.", + "D": "Turn left at the stove, then continue straight ahead to arrive at the sink.", + "E": "You should turn around and then go straight." + }, + "answer_label": "E", + "answer_text": "You should turn around and then go straight.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_2", + "correct_answer": "You should turn around and then go straight.", + "original_options": { + "A": "Turn around, then go left.", + "B": "Turn right, then go straight.", + "C": "Step backward first, then go straight forward until you reach the sink.", + "D": "Turn left at the stove, then continue straight ahead to arrive at the sink.", + "E": "You should turn around and then go straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 14 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_13", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many pots were on the stove?", + "answer": "There are a pot on the stove.", + "question_chinese": "灶台上有多少个锅？", + "answer_chinese": "灶台上有一个锅。", + "evidence": [ + { + "description": "I took a pot away.", + "timestamps": { + "start": "00:11:48", + "end": "00:12:17" + } + }, + { + "description": "There are two pots on the stove.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:12:18", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "No pots on the stove.", + "B": "There are a pot on the stove.", + "C": "Two pots on the stove.", + "D": "There are three pots on the stove, spaced across the burners in front of you.", + "E": "There are two pots sitting on the stove, one on each side." + }, + "answer_label": "B", + "answer_text": "There are a pot on the stove.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_13", + "correct_answer": "There are a pot on the stove.", + "original_options": { + "A": "No pots on the stove.", + "B": "There are a pot on the stove.", + "C": "Two pots on the stove.", + "D": "There are three pots on the stove, spaced across the burners in front of you.", + "E": "There are two pots sitting on the stove, one on each side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 12 + ] + }, + { + "video_uid": "3b989129-a310-4aa9-85f5-e5e9a90df99b", + "video_length": "00:30:13", + "qid": "3b989129-a310-4aa9-85f5-e5e9a90df99b_14", + "qid_num": 27, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many pots were on the stove?", + "answer": "There are two pots on the stove.", + "question_chinese": "灶台上有多少个锅？", + "answer_chinese": "灶台上有两个锅。", + "evidence": [ + { + "description": "There are two pots on the stove.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:17:44", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two pots on the stove.", + "B": "Three pots on the stove.", + "C": "Four pots on the stove.", + "D": "There are five pots on the stove, covering most of the burners.", + "E": "There are six pots on the stove, spread out across the top." + }, + "answer_label": "A", + "answer_text": "There are two pots on the stove.", + "q_id": "3b989129-a310-4aa9-85f5-e5e9a90df99b_14", + "correct_answer": "There are two pots on the stove.", + "original_options": { + "A": "There are two pots on the stove.", + "B": "Three pots on the stove.", + "C": "Four pots on the stove.", + "D": "There are five pots on the stove, covering most of the burners.", + "E": "There are six pots on the stove, spread out across the top." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + } +] diff --git a/QAs/3ce37765-7d70-4dd6-991c-a83286ba05b2-mcq.json b/QAs/3ce37765-7d70-4dd6-991c-a83286ba05b2-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e426e4131973db46395a3e77a3843ceed33d22b7 --- /dev/null +++ b/QAs/3ce37765-7d70-4dd6-991c-a83286ba05b2-mcq.json @@ -0,0 +1,1795 @@ +[ + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many kinds of rulers did I use in total?", + "answer": "three kinds.", + "question_chinese": "我一共使用了几种尺子？", + "answer_chinese": "3种。", + "evidence": [ + { + "description": "Used a tape measure.", + "timestamps": { + "start": "00:01:23", + "end": "00:02:41" + } + }, + { + "description": "Used a triangle ruler.", + "timestamps": { + "start": "00:02:45", + "end": "00:03:15" + } + }, + { + "description": "Used a straight ruler.", + "timestamps": { + "start": "00:03:47", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:04:32", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two kinds.", + "B": "Four kinds.", + "C": "three kinds.", + "D": "In total, you used five different kinds of rulers across the task.", + "E": "Overall, you ended up using only one kind of ruler the entire time." + }, + "answer_label": "C", + "answer_text": "three kinds.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_1", + "correct_answer": "three kinds.", + "original_options": { + "A": "Two kinds.", + "B": "Four kinds.", + "C": "three kinds.", + "D": "In total, you used five different kinds of rulers across the task.", + "E": "Overall, you ended up using only one kind of ruler the entire time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many kinds of rulers did I use in total?", + "answer": "three kinds.", + "question_chinese": "我一共使用了几种尺子？", + "answer_chinese": "3种。", + "evidence": [ + { + "description": "Used a tape measure.", + "timestamps": { + "start": "00:01:23", + "end": "00:02:41" + } + }, + { + "description": "Used a triangle ruler.", + "timestamps": { + "start": "00:02:45", + "end": "00:03:15" + } + }, + { + "description": "Used a straight ruler.", + "timestamps": { + "start": "00:03:47", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One kind.", + "B": "Two kinds.", + "C": "Altogether, you used four kinds of rulers during the whole activity overall.", + "D": "three kinds.", + "E": "Across everything you did, you used five different kinds of rulers in total." + }, + "answer_label": "D", + "answer_text": "three kinds.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_2", + "correct_answer": "three kinds.", + "original_options": { + "A": "One kind.", + "B": "Two kinds.", + "C": "Altogether, you used four kinds of rulers during the whole activity overall.", + "D": "three kinds.", + "E": "Across everything you did, you used five different kinds of rulers in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_6_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many long screws have I installed in total?", + "answer": "So far, you’ve installed 4 screws in total, and the screws are in front of you.", + "question_chinese": "我一共装了几颗长螺丝钉？", + "answer_chinese": "目前为止，你一共装了4颗，螺丝钉在你的正前方。", + "evidence": [ + { + "description": "So far, you've loaded four.", + "timestamps": { + "start": "00:13:13", + "end": "00:16:53" + } + } + ], + "question_timestamps": "00:16:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve installed 3 screws total.", + "B": "So far, you’ve installed 4 screws in total, and the screws are in front of you.", + "C": "You’ve installed 5 screws total.", + "D": "So far, you’ve installed 2 screws in total, and the screws are in front of you.", + "E": "So far, you’ve installed 6 screws in total, and the screws are in front of you." + }, + "answer_label": "B", + "answer_text": "So far, you’ve installed 4 screws in total, and the screws are in front of you.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_6_T2", + "correct_answer": "So far, you’ve installed 4 screws in total, and the screws are in front of you.", + "original_options": { + "A": "You’ve installed 3 screws total.", + "B": "So far, you’ve installed 4 screws in total, and the screws are in front of you.", + "C": "You’ve installed 5 screws total.", + "D": "So far, you’ve installed 2 screws in total, and the screws are in front of you.", + "E": "So far, you’ve installed 6 screws in total, and the screws are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_6_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many long screws have I installed in total?", + "answer": "You’ve installed 12 screws in total, and the screws are in front of you.", + "question_chinese": "我一共装了几颗长螺丝钉？", + "answer_chinese": "你一共装了12颗，螺丝钉在你的正前方。", + "evidence": [ + { + "description": "The third time is still four.", + "timestamps": { + "start": "00:25:34", + "end": "00:27:08" + } + }, + { + "description": "The second time is also four.", + "timestamps": { + "start": "00:21:20", + "end": "00:22:20" + } + }, + { + "description": "The first time was four.", + "timestamps": { + "start": "00:13:26", + "end": "00:16:53" + } + } + ], + "question_timestamps": "00:27:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve installed 12 screws in total, and the screws are in front of you.", + "B": "You’ve installed 11 long screws total.", + "C": "You’ve installed 13 long screws total.", + "D": "You’ve installed 10 screws in total, and the long screws are still on the ground in front of you.", + "E": "You’ve installed 14 screws in total, and the installed screws are positioned just behind you." + }, + "answer_label": "A", + "answer_text": "You’ve installed 12 screws in total, and the screws are in front of you.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_6_T1", + "correct_answer": "You’ve installed 12 screws in total, and the screws are in front of you.", + "original_options": { + "A": "You’ve installed 12 screws in total, and the screws are in front of you.", + "B": "You’ve installed 11 long screws total.", + "C": "You’ve installed 13 long screws total.", + "D": "You’ve installed 10 screws in total, and the long screws are still on the ground in front of you.", + "E": "You’ve installed 14 screws in total, and the installed screws are positioned just behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 16 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_11", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many long screws did I pre-install just now?", + "answer": "four.", + "question_chinese": "我刚才一共预装了几颗长螺丝钉？", + "answer_chinese": "4颗。", + "evidence": [ + { + "description": "Pre-installed two long screws on the right side.", + "timestamps": { + "start": "00:25:34", + "end": "00:25:41" + } + }, + { + "description": "Pre-installed two long screws on the left side.", + "timestamps": { + "start": "00:25:45", + "end": "00:25:57" + } + } + ], + "question_timestamps": "00:26:22", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "four.", + "B": "Three.", + "C": "Five.", + "D": "You pre-installed two long screws just now, then set the remaining screws aside in front of you.", + "E": "You pre-installed six long screws just now, and the rest are still uninstalled nearby." + }, + "answer_label": "A", + "answer_text": "four.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_11", + "correct_answer": "four.", + "original_options": { + "A": "four.", + "B": "Three.", + "C": "Five.", + "D": "You pre-installed two long screws just now, then set the remaining screws aside in front of you.", + "E": "You pre-installed six long screws just now, and the rest are still uninstalled nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 14 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_12", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many long screws did I pre-install just now?", + "answer": "four.", + "question_chinese": "我刚才一共预装了几颗长螺丝钉？", + "answer_chinese": "4颗。", + "evidence": [ + { + "description": "Pre-installed two long screws on the right side.", + "timestamps": { + "start": "00:25:34", + "end": "00:25:41" + } + }, + { + "description": "Pre-installed two long screws on the left side.", + "timestamps": { + "start": "00:25:45", + "end": "00:25:57" + } + } + ], + "question_timestamps": "00:26:41", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Seven.", + "C": "You pre-installed three long screws just now, then paused before adding any more screws.", + "D": "four.", + "E": "You pre-installed five long screws just now, with the remaining long screws not yet installed." + }, + "answer_label": "D", + "answer_text": "four.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_12", + "correct_answer": "four.", + "original_options": { + "A": "Two.", + "B": "Seven.", + "C": "You pre-installed three long screws just now, then paused before adding any more screws.", + "D": "four.", + "E": "You pre-installed five long screws just now, with the remaining long screws not yet installed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_17", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the knife just now?", + "answer": "The knife was placed in your right pants pocket.", + "question_chinese": "我刚才把小刀放在哪里了？", + "answer_chinese": "小刀放在了你右边的裤兜里。", + "evidence": [ + { + "description": "Put the knife back into the right pants pocket.", + "timestamps": { + "start": "00:31:12", + "end": "00:31:16" + } + } + ], + "question_timestamps": "00:31:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In your left pants pocket.", + "B": "In your back pants pocket.", + "C": "You put the knife into your jacket pocket instead of your pants pocket after picking it up.", + "D": "The knife was placed in your right pants pocket.", + "E": "You set the knife down on the counter beside you rather than putting it into a pocket." + }, + "answer_label": "D", + "answer_text": "The knife was placed in your right pants pocket.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_17", + "correct_answer": "The knife was placed in your right pants pocket.", + "original_options": { + "A": "In your left pants pocket.", + "B": "In your back pants pocket.", + "C": "You put the knife into your jacket pocket instead of your pants pocket after picking it up.", + "D": "The knife was placed in your right pants pocket.", + "E": "You set the knife down on the counter beside you rather than putting it into a pocket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_18", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the knife just now?", + "answer": "The knife was placed in your right pants pocket.", + "question_chinese": "我刚才把小刀放在哪里了？", + "answer_chinese": "小刀放在了你右边的裤兜里。", + "evidence": [ + { + "description": "Put the knife back into the right pants pocket.", + "timestamps": { + "start": "00:31:12", + "end": "00:31:16" + } + } + ], + "question_timestamps": "00:35:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In your left pants pocket.", + "B": "The knife was placed in your right pants pocket.", + "C": "In your back pants pocket.", + "D": "You placed the knife on the kitchen counter instead of putting it into any pocket.", + "E": "You dropped the knife into the kitchen sink rather than stowing it in your pants pocket." + }, + "answer_label": "B", + "answer_text": "The knife was placed in your right pants pocket.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_18", + "correct_answer": "The knife was placed in your right pants pocket.", + "original_options": { + "A": "In your left pants pocket.", + "B": "The knife was placed in your right pants pocket.", + "C": "In your back pants pocket.", + "D": "You placed the knife on the kitchen counter instead of putting it into any pocket.", + "E": "You dropped the knife into the kitchen sink rather than stowing it in your pants pocket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_9", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was the path I just took with this small wooden board?", + "answer": "You walked to the bottom of the ladder with the small wooden board and then climbed up the ladder.", + "question_chinese": "我刚才拿着这块小木板走过的路径是怎样的？", + "answer_chinese": "你拿着小木板走到梯子下面，然后爬上梯子。", + "evidence": [ + { + "description": "Carrying the small wooden board, walked forward to the bottom of the ladder.", + "timestamps": { + "start": "00:24:21", + "end": "00:24:25" + } + }, + { + "description": "Climbed up the ladder with the small wooden board.", + "timestamps": { + "start": "00:24:25", + "end": "00:24:33" + } + } + ], + "question_timestamps": "00:24:38", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You climbed the ladder first.", + "B": "You walked past the ladder.", + "C": "You walked to the top of the ladder with the small wooden board, then climbed back down.", + "D": "You carried the small wooden board to the ladder but stepped around it without climbing up.", + "E": "You walked to the bottom of the ladder with the small wooden board and then climbed up the ladder." + }, + "answer_label": "E", + "answer_text": "You walked to the bottom of the ladder with the small wooden board and then climbed up the ladder.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_9", + "correct_answer": "You walked to the bottom of the ladder with the small wooden board and then climbed up the ladder.", + "original_options": { + "A": "You climbed the ladder first.", + "B": "You walked past the ladder.", + "C": "You walked to the top of the ladder with the small wooden board, then climbed back down.", + "D": "You carried the small wooden board to the ladder but stepped around it without climbing up.", + "E": "You walked to the bottom of the ladder with the small wooden board and then climbed up the ladder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_10", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was the path I just took with this small wooden board?", + "answer": "You walked to the bottom of the ladder with the small wooden board and then climbed up the ladder.", + "question_chinese": "我刚才拿着这块小木板走过的路径是怎样的？", + "answer_chinese": "你拿着小木板走到梯子下面，然后爬上梯子。", + "evidence": [ + { + "description": "Carrying the small wooden board, walked forward to the bottom of the ladder.", + "timestamps": { + "start": "00:24:21", + "end": "00:24:25" + } + }, + { + "description": "Climbed up the ladder with the small wooden board.", + "timestamps": { + "start": "00:24:25", + "end": "00:24:33" + } + } + ], + "question_timestamps": "00:26:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went to the ladder’s top.", + "B": "You walked around the ladder.", + "C": "You walked to the bottom of the ladder with the small wooden board and then climbed up the ladder.", + "D": "You walked to the bottom of the ladder with the small wooden board, then climbed down the ladder.", + "E": "You carried the small wooden board beside the ladder and stepped past it without climbing up." + }, + "answer_label": "C", + "answer_text": "You walked to the bottom of the ladder with the small wooden board and then climbed up the ladder.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_10", + "correct_answer": "You walked to the bottom of the ladder with the small wooden board and then climbed up the ladder.", + "original_options": { + "A": "You went to the ladder’s top.", + "B": "You walked around the ladder.", + "C": "You walked to the bottom of the ladder with the small wooden board and then climbed up the ladder.", + "D": "You walked to the bottom of the ladder with the small wooden board, then climbed down the ladder.", + "E": "You carried the small wooden board beside the ladder and stepped past it without climbing up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 16 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_13", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get this long wooden board in my hand from?", + "answer": "It was taken from the truck carriage behind you.", + "question_chinese": "我手上这块长木板是从哪里拿过来的？", + "answer_chinese": "从你身后的货车车厢里拿过来的。", + "evidence": [ + { + "description": "Took the long wooden board out of the truck carriage.", + "timestamps": { + "start": "00:28:26", + "end": "00:28:43" + } + } + ], + "question_timestamps": "00:28:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It came from the truck cab.", + "B": "It was taken from the trailer nearby.", + "C": "It was taken from the truck carriage behind you.", + "D": "You picked it up from the storage area to your left, not from the truck carriage.", + "E": "It was retrieved from a shed behind the house, then carried with you afterward." + }, + "answer_label": "C", + "answer_text": "It was taken from the truck carriage behind you.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_13", + "correct_answer": "It was taken from the truck carriage behind you.", + "original_options": { + "A": "It came from the truck cab.", + "B": "It was taken from the trailer nearby.", + "C": "It was taken from the truck carriage behind you.", + "D": "You picked it up from the storage area to your left, not from the truck carriage.", + "E": "It was retrieved from a shed behind the house, then carried with you afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 14 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_16", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the yellow tape measure?", + "answer": "The yellow tape measure is in your right pants pocket.", + "question_chinese": "我把黄色卷尺放在哪里了？", + "answer_chinese": "黄色卷尺在你右边的裤兜里。", + "evidence": [ + { + "description": "Put the yellow tape measure back into the right pants pocket.", + "timestamps": { + "start": "00:28:22", + "end": "00:28:27" + } + } + ], + "question_timestamps": "00:29:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s in your left pants pocket.", + "B": "It’s inside your backpack.", + "C": "The yellow tape measure is on the kitchen counter, not tucked in your right pants pocket.", + "D": "You put the yellow tape measure in your car’s glove compartment instead of your right pants pocket.", + "E": "The yellow tape measure is in your right pants pocket." + }, + "answer_label": "E", + "answer_text": "The yellow tape measure is in your right pants pocket.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_16", + "correct_answer": "The yellow tape measure is in your right pants pocket.", + "original_options": { + "A": "It’s in your left pants pocket.", + "B": "It’s inside your backpack.", + "C": "The yellow tape measure is on the kitchen counter, not tucked in your right pants pocket.", + "D": "You put the yellow tape measure in your car’s glove compartment instead of your right pants pocket.", + "E": "The yellow tape measure is in your right pants pocket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 18 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_9_T2", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many times did I pick up the utility knife?", + "answer": "This is the first time you’ve picked it up; the knife is in your right hand.", + "question_chinese": "我拿了几次小刀？", + "answer_chinese": "这是你第一次拿，小刀在你的右手里。", + "evidence": [ + { + "description": "Take it for the first time", + "timestamps": { + "start": "00:03:24", + "end": "00:03:26" + } + } + ], + "question_timestamps": "00:03:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Twice; it’s in your right pocket.", + "B": "Three times; it’s in your right hand.", + "C": "You haven’t picked it up yet; the utility knife is still in your right pants pocket.", + "D": "This is the second time you’ve picked it up, and the knife is in your right hand now.", + "E": "This is the first time you’ve picked it up; the knife is in your right hand." + }, + "answer_label": "E", + "answer_text": "This is the first time you’ve picked it up; the knife is in your right hand.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_9_T2", + "correct_answer": "This is the first time you’ve picked it up; the knife is in your right hand.", + "original_options": { + "A": "Twice; it’s in your right pocket.", + "B": "Three times; it’s in your right hand.", + "C": "You haven’t picked it up yet; the utility knife is still in your right pants pocket.", + "D": "This is the second time you’ve picked it up, and the knife is in your right hand now.", + "E": "This is the first time you’ve picked it up; the knife is in your right hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 19 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_9_T1", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many times did I pick up the utility knife?", + "answer": "You’ve taken it out twice; the knife is in the right pocket of your khaki pants.", + "question_chinese": "我拿了几次小刀？", + "answer_chinese": "你拿出过两次，小刀在你的卡其色裤子的右边口袋里。", + "evidence": [ + { + "description": "second time", + "timestamps": { + "start": "00:31:10", + "end": "00:31:12" + } + }, + { + "description": "first time", + "timestamps": { + "start": "00:03:24", + "end": "00:03:26" + } + } + ], + "question_timestamps": "00:31:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Once; it’s in your right hand.", + "B": "Three times; it’s in your right pocket.", + "C": "You’ve only taken it out once; the knife is in the right pocket of your khaki pants.", + "D": "You’ve taken it out twice; the knife is in the right pocket of your khaki pants.", + "E": "You haven’t picked it up yet; the knife is still in the right pocket of your khaki pants." + }, + "answer_label": "D", + "answer_text": "You’ve taken it out twice; the knife is in the right pocket of your khaki pants.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_9_T1", + "correct_answer": "You’ve taken it out twice; the knife is in the right pocket of your khaki pants.", + "original_options": { + "A": "Once; it’s in your right hand.", + "B": "Three times; it’s in your right pocket.", + "C": "You’ve only taken it out once; the knife is in the right pocket of your khaki pants.", + "D": "You’ve taken it out twice; the knife is in the right pocket of your khaki pants.", + "E": "You haven’t picked it up yet; the knife is still in the right pocket of your khaki pants." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 19 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_3_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow cutting machines do I have?", + "answer": "You have one yellow cutting machine, and it’s to your right.", + "question_chinese": "我有几个黄色的切割机？", + "answer_chinese": "你现在有一个，黄色切割机在你的右边。", + "evidence": [ + { + "description": "There is a yellow cutting machine on the grass.", + "timestamps": { + "start": "00:38:10", + "end": "00:38:58" + } + } + ], + "question_timestamps": "00:39:02", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have zero yellow cutting machines.", + "B": "You have two yellow cutting machines.", + "C": "You have one yellow cutting machine, and it’s to your right.", + "D": "You have three yellow cutting machines; one is to your right, one in front of you, and one to your left.", + "E": "You have four yellow cutting machines; two are to your right and two are in front of you." + }, + "answer_label": "C", + "answer_text": "You have one yellow cutting machine, and it’s to your right.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_3_T1", + "correct_answer": "You have one yellow cutting machine, and it’s to your right.", + "original_options": { + "A": "You have zero yellow cutting machines.", + "B": "You have two yellow cutting machines.", + "C": "You have one yellow cutting machine, and it’s to your right.", + "D": "You have three yellow cutting machines; one is to your right, one in front of you, and one to your left.", + "E": "You have four yellow cutting machines; two are to your right and two are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 18 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_3_T2", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow cutting machines do I have?", + "answer": "You have two in total: one is to your right, and the other is in front of you.", + "question_chinese": "我有几个黄色的切割机？", + "answer_chinese": "你一共有两个。一个在你的右边，另一个在你的正前方。", + "evidence": [ + { + "description": "There is a yellow cutting machine on the grass.", + "timestamps": { + "start": "00:38:10", + "end": "00:38:58" + } + }, + { + "description": "There is a yellow cutting machine on the red toolbox behind the ladder.", + "timestamps": { + "start": "00:39:00", + "end": "00:39:06" + } + } + ], + "question_timestamps": "00:39:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have one yellow cutting machine.", + "B": "You have three yellow cutting machines.", + "C": "You have one in total, and it’s directly in front of you with no other yellow cutting machine nearby.", + "D": "You have two in total: one is to your right, and the other is in front of you.", + "E": "You have three in total: one to your right, one in front of you, and another behind you." + }, + "answer_label": "D", + "answer_text": "You have two in total: one is to your right, and the other is in front of you.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_3_T2", + "correct_answer": "You have two in total: one is to your right, and the other is in front of you.", + "original_options": { + "A": "You have one yellow cutting machine.", + "B": "You have three yellow cutting machines.", + "C": "You have one in total, and it’s directly in front of you with no other yellow cutting machine nearby.", + "D": "You have two in total: one is to your right, and the other is in front of you.", + "E": "You have three in total: one to your right, one in front of you, and another behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 18 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_1_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kind of ruler am I using right now?", + "answer": "Right now you’re using a tape measure, and the tape measure is directly in front of you.", + "question_chinese": "我现在在用什么尺子？", + "answer_chinese": "你现在在用卷尺，卷尺在你的正前方。", + "evidence": [ + { + "description": "You are using a tape measure now.", + "timestamps": { + "start": "00:01:23", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:02:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "A straight ruler.", + "B": "Right now you’re using a tape measure, and the tape measure is directly in front of you.", + "C": "A folding ruler.", + "D": "Right now you’re using a set square, and the set square is in your hand rather than in front of you.", + "E": "Right now you’re using a yardstick, and it’s positioned off to your right instead of directly ahead." + }, + "answer_label": "B", + "answer_text": "Right now you’re using a tape measure, and the tape measure is directly in front of you.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_1_T2", + "correct_answer": "Right now you’re using a tape measure, and the tape measure is directly in front of you.", + "original_options": { + "A": "A straight ruler.", + "B": "Right now you’re using a tape measure, and the tape measure is directly in front of you.", + "C": "A folding ruler.", + "D": "Right now you’re using a set square, and the set square is in your hand rather than in front of you.", + "E": "Right now you’re using a yardstick, and it’s positioned off to your right instead of directly ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 22, + 19 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_1_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kind of ruler am I using right now?", + "answer": "Right now you’re using a set square, and the set square is in your hand.", + "question_chinese": "我现在在用什么尺子？", + "answer_chinese": "你现在在用三角尺，三角尺在你的手上。", + "evidence": [ + { + "description": "Used a triangle ruler.", + "timestamps": { + "start": "00:02:45", + "end": "00:03:09" + } + } + ], + "question_timestamps": "00:03:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Right now you’re using a set square, and the set square is in your hand.", + "B": "A tape measure.", + "C": "A straight ruler.", + "D": "Right now you’re using a folding ruler, and the folding ruler is in your hand as you measure.", + "E": "Right now you’re using a yardstick, and the yardstick is to your right instead of in your hand." + }, + "answer_label": "A", + "answer_text": "Right now you’re using a set square, and the set square is in your hand.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_1_T1", + "correct_answer": "Right now you’re using a set square, and the set square is in your hand.", + "original_options": { + "A": "Right now you’re using a set square, and the set square is in your hand.", + "B": "A tape measure.", + "C": "A straight ruler.", + "D": "Right now you’re using a folding ruler, and the folding ruler is in your hand as you measure.", + "E": "Right now you’re using a yardstick, and the yardstick is to your right instead of in your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 19, + 19 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_19", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the yellow cutting machine below me now?", + "answer": "No.", + "question_chinese": "我现在能够得到下面的黄色切割机吗？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The yellow cutting machine is on the ground below you. You are now on a ladder, about 1 to 2 meters away from the ground.", + "timestamps": { + "start": "00:43:12", + "end": "00:43:26" + } + } + ], + "question_timestamps": "00:43:32", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_19", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_20", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the yellow cutting machine below me now?", + "answer": "No.", + "question_chinese": "我现在能够得到下面的黄色切割机吗？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The yellow cutting machine is on the ground below you. You are now on a ladder, about 1 to 2 meters away from the ground.", + "timestamps": { + "start": "00:43:12", + "end": "00:43:26" + } + } + ], + "question_timestamps": "00:45:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_20", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_10_T1", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the yellow cutting machine right now?", + "answer": "Yes—you can reach it; it’s right in front of you.", + "question_chinese": "我现在能够得到黄色切割机吗？", + "answer_chinese": "你能，它就在你的前面。", + "evidence": [ + { + "description": "You can. It's right in front of you.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:12" + } + } + ], + "question_timestamps": "00:10:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t reach it; it’s under the ladder.", + "B": "Yes—you can reach it; it’s right in front of you." + }, + "answer_label": "B", + "answer_text": "Yes—you can reach it; it’s right in front of you.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_10_T1", + "correct_answer": "Yes—you can reach it; it’s right in front of you.", + "original_options": { + "A": "No—you can’t reach it; it’s under the ladder.", + "B": "Yes—you can reach it; it’s right in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_10_T2", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the yellow cutting machine right now?", + "answer": "No—you can’t reach it; it’s under the ladder.", + "question_chinese": "我现在能够得到黄色切割机吗？", + "answer_chinese": "你不能，它在梯子下面。", + "evidence": [ + { + "description": "The yellow cutting machine is on the ground below you. You are now on a ladder, about 1 to 2 meters away from the ground.", + "timestamps": { + "start": "00:43:12", + "end": "00:43:26" + } + } + ], + "question_timestamps": "00:45:01", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t reach it; it’s under the ladder.", + "B": "Yes—you can reach it; it’s right in front of you." + }, + "answer_label": "A", + "answer_text": "No—you can’t reach it; it’s under the ladder.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_10_T2", + "correct_answer": "No—you can’t reach it; it’s under the ladder.", + "original_options": { + "A": "No—you can’t reach it; it’s under the ladder.", + "B": "Yes—you can reach it; it’s right in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_7", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Now, how many small wooden boards have I installed in total?", + "answer": "You have now installed a total of two small wooden boards.", + "question_chinese": "现在，我一共装了几块小木板？", + "answer_chinese": "现在你一共装了两块小木板。", + "evidence": [ + { + "description": "Installed the first small wooden board.", + "timestamps": { + "start": "00:11:48", + "end": "00:14:09" + } + }, + { + "description": "Installed the second small wooden board.", + "timestamps": { + "start": "00:20:32", + "end": "00:22:21" + } + } + ], + "question_timestamps": "00:22:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have now installed a total of two small wooden boards.", + "B": "You have installed one small wooden board.", + "C": "You have installed three small wooden boards.", + "D": "You have now installed a total of four small wooden boards, adding two more after the first placements.", + "E": "You have now installed a total of five small wooden boards, with several added beyond the initial two." + }, + "answer_label": "A", + "answer_text": "You have now installed a total of two small wooden boards.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_7", + "correct_answer": "You have now installed a total of two small wooden boards.", + "original_options": { + "A": "You have now installed a total of two small wooden boards.", + "B": "You have installed one small wooden board.", + "C": "You have installed three small wooden boards.", + "D": "You have now installed a total of four small wooden boards, adding two more after the first placements.", + "E": "You have now installed a total of five small wooden boards, with several added beyond the initial two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_8", + "qid_num": 28, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Now, how many small wooden boards have I installed in total?", + "answer": "You have now installed a total of two small wooden boards.", + "question_chinese": "现在，我一共装了几块小木板？", + "answer_chinese": "现在你一共装了两块小木板。", + "evidence": [ + { + "description": "Installed the first small wooden board.", + "timestamps": { + "start": "00:11:48", + "end": "00:14:09" + } + }, + { + "description": "Installed the second small wooden board.", + "timestamps": { + "start": "00:20:32", + "end": "00:22:21" + } + } + ], + "question_timestamps": "00:25:25", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have installed one small wooden board.", + "B": "You have installed three small wooden boards.", + "C": "You have now installed a total of four small wooden boards across the work area so far.", + "D": "You have now installed a total of two small wooden boards.", + "E": "You have now installed a total of five small wooden boards, finishing several more than before." + }, + "answer_label": "D", + "answer_text": "You have now installed a total of two small wooden boards.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_8", + "correct_answer": "You have now installed a total of two small wooden boards.", + "original_options": { + "A": "You have installed one small wooden board.", + "B": "You have installed three small wooden boards.", + "C": "You have now installed a total of four small wooden boards across the work area so far.", + "D": "You have now installed a total of two small wooden boards.", + "E": "You have now installed a total of five small wooden boards, finishing several more than before." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_3", + "qid_num": 29, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of wood have I sawn in total now?", + "answer": "three pieces.", + "question_chinese": "现在我一共锯了几块木头？", + "answer_chinese": "3块。", + "evidence": [ + { + "description": "Sawed the first piece of wood.", + "timestamps": { + "start": "00:09:38", + "end": "00:10:07" + } + }, + { + "description": "Sawed the second piece of wood.", + "timestamps": { + "start": "00:19:16", + "end": "00:19:36" + } + }, + { + "description": "Sawed the third piece of wood.", + "timestamps": { + "start": "00:23:13", + "end": "00:23:35" + } + } + ], + "question_timestamps": "00:23:50", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two pieces.", + "B": "Four pieces.", + "C": "You have sawn a total of five pieces of wood by this point in the sequence.", + "D": "You have sawn six pieces of wood in total now, counting every cut piece so far.", + "E": "three pieces." + }, + "answer_label": "E", + "answer_text": "three pieces.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_3", + "correct_answer": "three pieces.", + "original_options": { + "A": "Two pieces.", + "B": "Four pieces.", + "C": "You have sawn a total of five pieces of wood by this point in the sequence.", + "D": "You have sawn six pieces of wood in total now, counting every cut piece so far.", + "E": "three pieces." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_4", + "qid_num": 30, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of wood have I sawn in total now?", + "answer": "three pieces.", + "question_chinese": "现在我一共锯了几块木头？", + "answer_chinese": "3块。", + "evidence": [ + { + "description": "Sawed the first piece of wood.", + "timestamps": { + "start": "00:09:38", + "end": "00:10:07" + } + }, + { + "description": "Sawed the second piece of wood.", + "timestamps": { + "start": "00:19:16", + "end": "00:19:36" + } + }, + { + "description": "Sawed the third piece of wood.", + "timestamps": { + "start": "00:23:13", + "end": "00:23:35" + } + } + ], + "question_timestamps": "00:25:59", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two pieces.", + "B": "Four pieces.", + "C": "three pieces.", + "D": "You have sawn a total of five pieces of wood at this stage of the task.", + "E": "You have sawn six pieces of wood in total now, including the most recent one." + }, + "answer_label": "C", + "answer_text": "three pieces.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_4", + "correct_answer": "three pieces.", + "original_options": { + "A": "Two pieces.", + "B": "Four pieces.", + "C": "three pieces.", + "D": "You have sawn a total of five pieces of wood at this stage of the task.", + "E": "You have sawn six pieces of wood in total now, including the most recent one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 15 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_5_T2", + "qid_num": 31, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is in front of me right now?", + "answer": "Right now, there’s a large wooden board in front of you.", + "question_chinese": "现在，我的前面是什么？", + "answer_chinese": "现在你的前面是一块大木板。", + "evidence": [ + { + "description": "Now in front of you is a big board.", + "timestamps": { + "start": "00:24:16", + "end": "00:24:19" + } + } + ], + "question_timestamps": "00:24:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right now, there’s a large wooden board in front of you.", + "B": "A large wooden board behind you.", + "C": "A large wooden board to your left.", + "D": "Right now, there’s a large wooden board to your right rather than directly in front of you.", + "E": "Right now, there’s a large wooden board above you, not positioned straight ahead." + }, + "answer_label": "A", + "answer_text": "Right now, there’s a large wooden board in front of you.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_5_T2", + "correct_answer": "Right now, there’s a large wooden board in front of you.", + "original_options": { + "A": "Right now, there’s a large wooden board in front of you.", + "B": "A large wooden board behind you.", + "C": "A large wooden board to your left.", + "D": "Right now, there’s a large wooden board to your right rather than directly in front of you.", + "E": "Right now, there’s a large wooden board above you, not positioned straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 14 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_5_T1", + "qid_num": 32, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is in front of me right now?", + "answer": "Right now, there’s an orange ladder in front of you.", + "question_chinese": "现在，我的前面是什么？", + "answer_chinese": "现在，你的前面是一把橙色的梯子。", + "evidence": [ + { + "description": "Now, in front of you is an orange ladder.", + "timestamps": { + "start": "00:24:24", + "end": "00:24:25" + } + } + ], + "question_timestamps": "00:24:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "An orange ladder behind you.", + "B": "An orange ladder to your left.", + "C": "Right now, there’s an orange ladder to your right instead of directly in front of you.", + "D": "Right now, there’s an orange ladder above you, not placed straight ahead of you.", + "E": "Right now, there’s an orange ladder in front of you." + }, + "answer_label": "E", + "answer_text": "Right now, there’s an orange ladder in front of you.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_5_T1", + "correct_answer": "Right now, there’s an orange ladder in front of you.", + "original_options": { + "A": "An orange ladder behind you.", + "B": "An orange ladder to your left.", + "C": "Right now, there’s an orange ladder to your right instead of directly in front of you.", + "D": "Right now, there’s an orange ladder above you, not placed straight ahead of you.", + "E": "Right now, there’s an orange ladder in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 15 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_4_T1", + "qid_num": 33, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, which small wooden plank am I installing (what number is it)?", + "answer": "Right now you’re installing the first small wooden board, and the board is in front of you.", + "question_chinese": "现在，我装的是第几块小木板？", + "answer_chinese": "现在你装的是第一块小木板，木板在你的正前方。", + "evidence": [ + { + "description": "Installed the first small wooden board.", + "timestamps": { + "start": "00:11:48", + "end": "00:14:09" + } + } + ], + "question_timestamps": "00:14:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’re installing the second board.", + "B": "You’re installing the third board.", + "C": "Right now you’re installing the fourth small wooden board, and the board is in front of you.", + "D": "Right now you’re installing the first small wooden board, and the board is in front of you.", + "E": "Right now you’re installing the fifth small wooden board, with the board positioned in front of you." + }, + "answer_label": "D", + "answer_text": "Right now you’re installing the first small wooden board, and the board is in front of you.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_4_T1", + "correct_answer": "Right now you’re installing the first small wooden board, and the board is in front of you.", + "original_options": { + "A": "You’re installing the second board.", + "B": "You’re installing the third board.", + "C": "Right now you’re installing the fourth small wooden board, and the board is in front of you.", + "D": "Right now you’re installing the first small wooden board, and the board is in front of you.", + "E": "Right now you’re installing the fifth small wooden board, with the board positioned in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_4_T2", + "qid_num": 34, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, which small wooden plank am I installing (what number is it)?", + "answer": "Right now you’re installing the second small wooden board, and the board is in front of you.", + "question_chinese": "现在，我装的是第几块小木板？", + "answer_chinese": "现在你装的是第二块小木板，木板在你的正前方。", + "evidence": [ + { + "description": "Installed the second small wooden board.", + "timestamps": { + "start": "00:20:32", + "end": "00:21:36" + } + } + ], + "question_timestamps": "00:21:38", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’re installing the first board.", + "B": "You’re installing the third board.", + "C": "Right now you’re installing the fourth small wooden board, and the board is in front of you.", + "D": "Right now you’re installing the fifth small wooden board, and the board is in front of you.", + "E": "Right now you’re installing the second small wooden board, and the board is in front of you." + }, + "answer_label": "E", + "answer_text": "Right now you’re installing the second small wooden board, and the board is in front of you.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_4_T2", + "correct_answer": "Right now you’re installing the second small wooden board, and the board is in front of you.", + "original_options": { + "A": "You’re installing the first board.", + "B": "You’re installing the third board.", + "C": "Right now you’re installing the fourth small wooden board, and the board is in front of you.", + "D": "Right now you’re installing the fifth small wooden board, and the board is in front of you.", + "E": "Right now you’re installing the second small wooden board, and the board is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_2_T1", + "qid_num": 35, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of wood have I sawed so far?", + "answer": "You’ve sawed one piece now, and the wood is in your left hand.", + "question_chinese": "现在我锯了几块木头？", + "answer_chinese": "你现在锯了一块，木头在你的左手上，", + "evidence": [ + { + "description": "You saw a piece now.", + "timestamps": { + "start": "00:09:38", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:10:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve sawed zero pieces so far.", + "B": "You’ve sawed one piece now, and the wood is in your left hand.", + "C": "You’ve sawed two pieces so far.", + "D": "You’ve sawed three pieces so far, and the wood is in your left hand.", + "E": "You’ve sawed four pieces so far, and the wood is still being held in your left hand." + }, + "answer_label": "B", + "answer_text": "You’ve sawed one piece now, and the wood is in your left hand.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_2_T1", + "correct_answer": "You’ve sawed one piece now, and the wood is in your left hand.", + "original_options": { + "A": "You’ve sawed zero pieces so far.", + "B": "You’ve sawed one piece now, and the wood is in your left hand.", + "C": "You’ve sawed two pieces so far.", + "D": "You’ve sawed three pieces so far, and the wood is in your left hand.", + "E": "You’ve sawed four pieces so far, and the wood is still being held in your left hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 18 + ] + }, + { + "video_uid": "3ce37765-7d70-4dd6-991c-a83286ba05b2", + "video_length": "00:45:23", + "qid": "3ce37765-7d70-4dd6-991c-a83286ba05b2_2_T2", + "qid_num": 36, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of wood have I sawed so far?", + "answer": "You’ve sawed two pieces of wood so far, and the wood is in your left hand.", + "question_chinese": "现在我锯了几块木头？", + "answer_chinese": "你现在在锯了二块木头，木头在你的左手上。", + "evidence": [ + { + "description": "Sawed the second piece of wood.", + "timestamps": { + "start": "00:19:16", + "end": "00:19:36" + } + }, + { + "description": "first one", + "timestamps": { + "start": "00:09:50", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:19:38", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve sawed one piece so far.", + "B": "You’ve sawed three pieces so far.", + "C": "You’ve sawed two pieces of wood so far, and the wood is in your left hand.", + "D": "You’ve sawed one piece of wood so far, and you’re holding the wood in your right hand.", + "E": "You’ve sawed four pieces of wood so far, and the wood you’re working on is in your left hand." + }, + "answer_label": "C", + "answer_text": "You’ve sawed two pieces of wood so far, and the wood is in your left hand.", + "q_id": "3ce37765-7d70-4dd6-991c-a83286ba05b2_2_T2", + "correct_answer": "You’ve sawed two pieces of wood so far, and the wood is in your left hand.", + "original_options": { + "A": "You’ve sawed one piece so far.", + "B": "You’ve sawed three pieces so far.", + "C": "You’ve sawed two pieces of wood so far, and the wood is in your left hand.", + "D": "You’ve sawed one piece of wood so far, and you’re holding the wood in your right hand.", + "E": "You’ve sawed four pieces of wood so far, and the wood you’re working on is in your left hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 21 + ] + } +] diff --git a/QAs/3d3cddf9-7c61-43af-8b1a-46aa78983a5a-mcq.json b/QAs/3d3cddf9-7c61-43af-8b1a-46aa78983a5a-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2b2a737cb419aa1b065c3b61e52c91ca1a0e412b --- /dev/null +++ b/QAs/3d3cddf9-7c61-43af-8b1a-46aa78983a5a-mcq.json @@ -0,0 +1,977 @@ +[ + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_10_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was reading, how far away was the staircase from me in a straight line?", + "answer": "When you are reading a book, the stairs are two meters away from you in a straight line.", + "question_chinese": "当我在看书时，楼梯距离我直线距离有多远？", + "answer_chinese": "当你在看书时，楼梯距离你直线距离两米。", + "evidence": [ + { + "description": "The stairs are to my front left.", + "timestamps": { + "start": "00:44:00", + "end": "00:44:05" + } + } + ], + "question_timestamps": "00:47:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The stairs are one meter away.", + "B": "When you are reading a book, the stairs are two meters away from you in a straight line.", + "C": "The stairs are three meters away.", + "D": "When you are reading a book, the stairs are four meters away from you in a straight line.", + "E": "When you are reading a book, the stairs are zero meters away from you in a straight line." + }, + "answer_label": "B", + "answer_text": "When you are reading a book, the stairs are two meters away from you in a straight line.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_10_T1", + "correct_answer": "When you are reading a book, the stairs are two meters away from you in a straight line.", + "original_options": { + "A": "The stairs are one meter away.", + "B": "When you are reading a book, the stairs are two meters away from you in a straight line.", + "C": "The stairs are three meters away.", + "D": "When you are reading a book, the stairs are four meters away from you in a straight line.", + "E": "When you are reading a book, the stairs are zero meters away from you in a straight line." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_10_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was reading, how far away was the staircase from me in a straight line?", + "answer": "When you are reading a book, the stairs are two meters away from you in a straight line.", + "question_chinese": "当我在看书时，楼梯距离我直线距离有多远？", + "answer_chinese": "当你在看书时，楼梯距离你直线距离两米。", + "evidence": [ + { + "description": "The stairs are to my front left.", + "timestamps": { + "start": "00:44:00", + "end": "00:44:05" + } + } + ], + "question_timestamps": "01:11:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The stairs are one meter away.", + "B": "The stairs are four meters away.", + "C": "When you are reading a book, the stairs are two meters away from you in a straight line.", + "D": "When you are reading a book, the stairs are three meters away from you in a straight line.", + "E": "When you are reading a book, the stairs are zero meters away from you in a straight line." + }, + "answer_label": "C", + "answer_text": "When you are reading a book, the stairs are two meters away from you in a straight line.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_10_T2", + "correct_answer": "When you are reading a book, the stairs are two meters away from you in a straight line.", + "original_options": { + "A": "The stairs are one meter away.", + "B": "The stairs are four meters away.", + "C": "When you are reading a book, the stairs are two meters away from you in a straight line.", + "D": "When you are reading a book, the stairs are three meters away from you in a straight line.", + "E": "When you are reading a book, the stairs are zero meters away from you in a straight line." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_13", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I stood in front of the sink, the green plastic basin was in front/left of me?", + "answer": "When you stand in front of the sink, the green plastic basin is in front of you.", + "question_chinese": "当我站在水槽前，绿色塑料盆在我的前/左？", + "answer_chinese": "当你站在水槽前，绿色塑料盆在你的前面。", + "evidence": [ + { + "description": "A green plastic basin is placed in the sink.", + "timestamps": { + "start": "00:02:25", + "end": "00:02:28" + } + } + ], + "question_timestamps": "00:11:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "When you stand in front of the sink, the green plastic basin is in front of you.", + "B": "It is to your left.", + "C": "It is behind you.", + "D": "When you stand in front of the sink, the green plastic basin is slightly to your left.", + "E": "When you stand in front of the sink, the green plastic basin is behind you and a bit to the left." + }, + "answer_label": "A", + "answer_text": "When you stand in front of the sink, the green plastic basin is in front of you.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_13", + "correct_answer": "When you stand in front of the sink, the green plastic basin is in front of you.", + "original_options": { + "A": "When you stand in front of the sink, the green plastic basin is in front of you.", + "B": "It is to your left.", + "C": "It is behind you.", + "D": "When you stand in front of the sink, the green plastic basin is slightly to your left.", + "E": "When you stand in front of the sink, the green plastic basin is behind you and a bit to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 21 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_14", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I stood in front of the sink, the green plastic basin was in front/left of me?", + "answer": "When you stand in front of the sink, the green plastic basin is on your left.", + "question_chinese": "当我站在水槽前，绿色塑料盆在我的前/左？", + "answer_chinese": "当你站在水槽前，绿色塑料盆在你的左边。", + "evidence": [ + { + "description": "A green plastic basin is placed in the sink.", + "timestamps": { + "start": "00:02:25", + "end": "00:02:28" + } + }, + { + "description": "I cleaned the green plastic basin and put it on the shelf on the left.", + "timestamps": { + "start": "00:28:40", + "end": "00:28:53" + } + } + ], + "question_timestamps": "00:56:41", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is on your right.", + "C": "When you stand in front of the sink, the green plastic basin is on your left.", + "D": "When you stand in front of the sink, the green plastic basin is directly in front of you.", + "E": "When you stand in front of the sink, the green plastic basin is slightly to your right near the sink." + }, + "answer_label": "C", + "answer_text": "When you stand in front of the sink, the green plastic basin is on your left.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_14", + "correct_answer": "When you stand in front of the sink, the green plastic basin is on your left.", + "original_options": { + "A": "It is in front of you.", + "B": "It is on your right.", + "C": "When you stand in front of the sink, the green plastic basin is on your left.", + "D": "When you stand in front of the sink, the green plastic basin is directly in front of you.", + "E": "When you stand in front of the sink, the green plastic basin is slightly to your right near the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 20 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_5_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I walk from the stovetop to the laundry room, and what do I pass along the way?", + "answer": "To get from the stove to the laundry room, you should turn left and go straight, passing through three doors.", + "question_chinese": "我从灶台去洗衣房应该怎么走，经过什么？", + "answer_chinese": "你从灶台去洗衣房应该左转直走，经过三道门。", + "evidence": [ + { + "description": "The laundry room is the third room down the hallway.", + "timestamps": { + "start": "00:41:59", + "end": "00:42:08" + } + } + ], + "question_timestamps": "00:51:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, pass one door.", + "B": "Turn left, pass two doors.", + "C": "To get from the stove to the laundry room, you should turn left and go straight, passing through three doors.", + "D": "To get from the stove to the laundry room, you should turn right and go straight, passing through three doors.", + "E": "To get from the stove to the laundry room, you should turn left and go straight, passing through four doors." + }, + "answer_label": "C", + "answer_text": "To get from the stove to the laundry room, you should turn left and go straight, passing through three doors.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_5_T1", + "correct_answer": "To get from the stove to the laundry room, you should turn left and go straight, passing through three doors.", + "original_options": { + "A": "Turn left, pass one door.", + "B": "Turn left, pass two doors.", + "C": "To get from the stove to the laundry room, you should turn left and go straight, passing through three doors.", + "D": "To get from the stove to the laundry room, you should turn right and go straight, passing through three doors.", + "E": "To get from the stove to the laundry room, you should turn left and go straight, passing through four doors." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 20 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_5_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I walk from the stovetop to the laundry room, and what do I pass along the way?", + "answer": "To get from the stove to the laundry room, you should turn left and go straight, passing through three doors.", + "question_chinese": "我从灶台去洗衣房应该怎么走，经过什么？", + "answer_chinese": "你从灶台去洗衣房应该向你的左转直走，经过三道门。", + "evidence": [ + { + "description": "The laundry room is the third room down the hallway.", + "timestamps": { + "start": "00:41:59", + "end": "00:42:08" + } + } + ], + "question_timestamps": "01:08:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, pass one door.", + "B": "Turn left, pass two doors.", + "C": "To get from the stove to the laundry room, you should turn right and go straight, passing through three doors.", + "D": "To get from the stove to the laundry room, you should turn left and go straight, passing through three doors.", + "E": "To get from the stove to the laundry room, you should turn left and go straight, passing through five doors." + }, + "answer_label": "D", + "answer_text": "To get from the stove to the laundry room, you should turn left and go straight, passing through three doors.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_5_T2", + "correct_answer": "To get from the stove to the laundry room, you should turn left and go straight, passing through three doors.", + "original_options": { + "A": "Turn left, pass one door.", + "B": "Turn left, pass two doors.", + "C": "To get from the stove to the laundry room, you should turn right and go straight, passing through three doors.", + "D": "To get from the stove to the laundry room, you should turn left and go straight, passing through three doors.", + "E": "To get from the stove to the laundry room, you should turn left and go straight, passing through five doors." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 20 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_2_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to eat a banana, where should I go to get it?", + "answer": "If you want to eat a banana, you need to get it from the fridge.", + "question_chinese": "我想吃香蕉要去哪里拿？", + "answer_chinese": "你想吃香蕉要去冰箱拿。", + "evidence": [ + { + "description": "Bananas are placed in the refrigerator.", + "timestamps": { + "start": "00:04:06", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:05:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Get it from the freezer.", + "B": "Get it from the kitchen counter.", + "C": "If you want to eat a banana, you need to get it from the fruit bowl on the table.", + "D": "If you want to eat a banana, you need to get it from the pantry shelf instead of the fridge.", + "E": "If you want to eat a banana, you need to get it from the fridge." + }, + "answer_label": "E", + "answer_text": "If you want to eat a banana, you need to get it from the fridge.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_2_T1", + "correct_answer": "If you want to eat a banana, you need to get it from the fridge.", + "original_options": { + "A": "Get it from the freezer.", + "B": "Get it from the kitchen counter.", + "C": "If you want to eat a banana, you need to get it from the fruit bowl on the table.", + "D": "If you want to eat a banana, you need to get it from the pantry shelf instead of the fridge.", + "E": "If you want to eat a banana, you need to get it from the fridge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 20 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_2_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to eat a banana, where should I go to get it?", + "answer": "If you want to eat a banana, you need to get it from the fridge, on your left-hand side.", + "question_chinese": "我想吃香蕉要去哪里拿？", + "answer_chinese": "你想吃香蕉要去冰箱拿，在你的左手边。", + "evidence": [ + { + "description": "Bananas are placed in the refrigerator.", + "timestamps": { + "start": "00:04:06", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:23:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "If you want to eat a banana, you need to get it from the fridge, on your left-hand side.", + "B": "Go to the freezer on your left.", + "C": "Check the fridge on your right.", + "D": "Get it from the fridge directly in front of you, not from the one on your left side.", + "E": "You should grab it from the kitchen counter on your left-hand side instead of the fridge." + }, + "answer_label": "A", + "answer_text": "If you want to eat a banana, you need to get it from the fridge, on your left-hand side.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_2_T2", + "correct_answer": "If you want to eat a banana, you need to get it from the fridge, on your left-hand side.", + "original_options": { + "A": "If you want to eat a banana, you need to get it from the fridge, on your left-hand side.", + "B": "Go to the freezer on your left.", + "C": "Check the fridge on your right.", + "D": "Get it from the fridge directly in front of you, not from the one on your left side.", + "E": "You should grab it from the kitchen counter on your left-hand side instead of the fridge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 16 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_3_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many keychains did I see on the table?", + "answer": "There are three sets of keys on the table, to your right.", + "question_chinese": "我看到桌子上有几串钥匙？", + "answer_chinese": "桌子上有三串钥匙，在你的右边。", + "evidence": [ + { + "description": "There are three bunches of keys on the table.", + "timestamps": { + "start": "00:43:15", + "end": "00:43:23" + } + } + ], + "question_timestamps": "00:45:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are three sets of keys on the table, to your right.", + "B": "You saw two keychains.", + "C": "You saw four keychains.", + "D": "There are five sets of keys on the table to your right, not three as you might recall.", + "E": "Only one set of keys is on the table to your right, so you saw a single keychain." + }, + "answer_label": "A", + "answer_text": "There are three sets of keys on the table, to your right.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_3_T1", + "correct_answer": "There are three sets of keys on the table, to your right.", + "original_options": { + "A": "There are three sets of keys on the table, to your right.", + "B": "You saw two keychains.", + "C": "You saw four keychains.", + "D": "There are five sets of keys on the table to your right, not three as you might recall.", + "E": "Only one set of keys is on the table to your right, so you saw a single keychain." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 18 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_3_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many keychains did I see on the table?", + "answer": "There are three sets of keys on the table, directly in front of you.", + "question_chinese": "我看到桌子上有几串钥匙？", + "answer_chinese": "桌子上有三串钥匙，在你的正前方。", + "evidence": [ + { + "description": "There are three bunches of keys on the table.", + "timestamps": { + "start": "00:43:15", + "end": "00:43:23" + } + } + ], + "question_timestamps": "01:08:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two keychains.", + "B": "There are three sets of keys on the table, directly in front of you.", + "C": "You saw four keychains.", + "D": "There is one set of keys on the table directly in front of you, not three sets.", + "E": "There are five sets of keys on the table directly in front of you, so you saw five keychains." + }, + "answer_label": "B", + "answer_text": "There are three sets of keys on the table, directly in front of you.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_3_T2", + "correct_answer": "There are three sets of keys on the table, directly in front of you.", + "original_options": { + "A": "You saw two keychains.", + "B": "There are three sets of keys on the table, directly in front of you.", + "C": "You saw four keychains.", + "D": "There is one set of keys on the table directly in front of you, not three sets.", + "E": "There are five sets of keys on the table directly in front of you, so you saw five keychains." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 19 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_8_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the TV?", + "answer": "You can’t reach the TV; it’s to your back-right.", + "question_chinese": "我能够得到电视吗？", + "answer_chinese": "你不能够得着电视，它在你的右后方。", + "evidence": [ + { + "description": "The sofa and the TV are three meters apart in a straight line.", + "timestamps": { + "start": "00:39:03", + "end": "00:39:05" + } + } + ], + "question_timestamps": "00:42:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can’t reach the TV; it’s to your back-right.", + "B": "You can reach the TV.", + "C": "It’s to your left.", + "D": "You can reach the TV; it’s positioned to your front-right within arm’s length.", + "E": "You can’t reach the TV; it’s directly in front of you, not to your back-right." + }, + "answer_label": "A", + "answer_text": "You can’t reach the TV; it’s to your back-right.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_8_T1", + "correct_answer": "You can’t reach the TV; it’s to your back-right.", + "original_options": { + "A": "You can’t reach the TV; it’s to your back-right.", + "B": "You can reach the TV.", + "C": "It’s to your left.", + "D": "You can reach the TV; it’s positioned to your front-right within arm’s length.", + "E": "You can’t reach the TV; it’s directly in front of you, not to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_8_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the TV?", + "answer": "Right now, you can reach the TV to your left.", + "question_chinese": "我能够得到电视吗？", + "answer_chinese": "你现在可以够得着你左边的电视。", + "evidence": [ + { + "description": "The sofa and the TV are three meters apart in a straight line.", + "timestamps": { + "start": "00:39:03", + "end": "00:39:05" + } + } + ], + "question_timestamps": "00:58:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can reach it to your right.", + "B": "Right now, you can reach the TV to your left.", + "C": "You can’t reach the TV.", + "D": "Right now, you can reach the TV directly in front of you without needing to move closer.", + "E": "You can’t reach the TV because it’s positioned to your front-left and out of arm’s reach." + }, + "answer_label": "B", + "answer_text": "Right now, you can reach the TV to your left.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_8_T2", + "correct_answer": "Right now, you can reach the TV to your left.", + "original_options": { + "A": "You can reach it to your right.", + "B": "Right now, you can reach the TV to your left.", + "C": "You can’t reach the TV.", + "D": "Right now, you can reach the TV directly in front of you without needing to move closer.", + "E": "You can’t reach the TV because it’s positioned to your front-left and out of arm’s reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 19 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_7", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many chairs have I seen?", + "answer": "So far, you have seen three chairs.", + "question_chinese": "目前为止，我看到了几把椅子？", + "answer_chinese": "目前为止，你看到了三把椅子。", + "evidence": [ + { + "description": "There are three chairs around the table.", + "timestamps": { + "start": "00:38:35", + "end": "00:38:40" + } + } + ], + "question_timestamps": "00:38:53", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "So far, you have seen three chairs.", + "B": "So far, you have seen two chairs.", + "C": "So far, you have seen four chairs.", + "D": "So far, you have seen five chairs in total as you moved through the space.", + "E": "So far, you have seen only one chair, with no other chairs appearing afterward." + }, + "answer_label": "A", + "answer_text": "So far, you have seen three chairs.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_7", + "correct_answer": "So far, you have seen three chairs.", + "original_options": { + "A": "So far, you have seen three chairs.", + "B": "So far, you have seen two chairs.", + "C": "So far, you have seen four chairs.", + "D": "So far, you have seen five chairs in total as you moved through the space.", + "E": "So far, you have seen only one chair, with no other chairs appearing afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 14 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_8", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many chairs have I seen?", + "answer": "So far, you have seen five chairs.", + "question_chinese": "目前为止，我看到了几把椅子？", + "answer_chinese": "目前为止，你看到了五把椅子。", + "evidence": [ + { + "description": "There are three chairs around the table.", + "timestamps": { + "start": "00:38:35", + "end": "00:38:40" + } + }, + { + "description": "There is a chair on each side of the table.", + "timestamps": { + "start": "00:39:02", + "end": "00:39:45" + } + } + ], + "question_timestamps": "00:49:18", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "So far, you have seen five chairs.", + "B": "So far, you have seen four chairs.", + "C": "So far, you have seen six chairs.", + "D": "So far, you have seen three chairs in total during everything you’ve looked at so far.", + "E": "So far, you have seen two chairs, with the remaining chairs never appearing in your view." + }, + "answer_label": "A", + "answer_text": "So far, you have seen five chairs.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_8", + "correct_answer": "So far, you have seen five chairs.", + "original_options": { + "A": "So far, you have seen five chairs.", + "B": "So far, you have seen four chairs.", + "C": "So far, you have seen six chairs.", + "D": "So far, you have seen three chairs in total during everything you’ve looked at so far.", + "E": "So far, you have seen two chairs, with the remaining chairs never appearing in your view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_12", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I go to get scissors?", + "answer": "Scissors are placed in front of the toaster.", + "question_chinese": "要用剪刀应该去哪里拿？", + "answer_chinese": "要用剪刀应该去面包机前面拿。", + "evidence": [ + { + "description": "Scissors are placed in front of the toaster.", + "timestamps": { + "start": "00:34:15", + "end": "00:34:18" + } + } + ], + "question_timestamps": "00:35:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Scissors are beside the blender.", + "B": "Scissors are next to the fridge.", + "C": "Scissors can be found in the cabinet above the sink rather than anywhere near the toaster.", + "D": "Scissors are stored underneath the cutting board on the counter, not placed in front of the toaster.", + "E": "Scissors are placed in front of the toaster." + }, + "answer_label": "E", + "answer_text": "Scissors are placed in front of the toaster.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_12", + "correct_answer": "Scissors are placed in front of the toaster.", + "original_options": { + "A": "Scissors are beside the blender.", + "B": "Scissors are next to the fridge.", + "C": "Scissors can be found in the cabinet above the sink rather than anywhere near the toaster.", + "D": "Scissors are stored underneath the cutting board on the counter, not placed in front of the toaster.", + "E": "Scissors are placed in front of the toaster." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_11", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I go to get scissors?", + "answer": "You should go to the drawer under the kitchen counter to get the scissors.", + "question_chinese": "要用剪刀应该去哪里拿？", + "answer_chinese": "要用剪刀应该去灶台下面的抽屉拿。", + "evidence": [ + { + "description": "Scissors are placed in front of the toaster.", + "timestamps": { + "start": "00:34:15", + "end": "00:34:18" + } + }, + { + "description": "I put the scissors in the drawer under the stove.", + "timestamps": { + "start": "00:35:43", + "end": "00:35:51" + } + } + ], + "question_timestamps": "01:00:29", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go to the cabinet under the counter.", + "B": "Go to the drawer by the sink.", + "C": "You should check the drawer inside the kitchen island to find the scissors you need.", + "D": "You should go to the drawer under the kitchen counter to get the scissors.", + "E": "You should look on the shelf inside the pantry for the scissors instead of the counter drawer." + }, + "answer_label": "D", + "answer_text": "You should go to the drawer under the kitchen counter to get the scissors.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_11", + "correct_answer": "You should go to the drawer under the kitchen counter to get the scissors.", + "original_options": { + "A": "Go to the cabinet under the counter.", + "B": "Go to the drawer by the sink.", + "C": "You should check the drawer inside the kitchen island to find the scissors you need.", + "D": "You should go to the drawer under the kitchen counter to get the scissors.", + "E": "You should look on the shelf inside the pantry for the scissors instead of the counter drawer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 17 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_9_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which time was this for me going to the laundry room?", + "answer": "This is your first time going to the laundry room; it’s to your back-right.", + "question_chinese": "这是我第几次去洗衣房？", + "answer_chinese": "这是你第一次去洗衣房，它在你的右后方。", + "evidence": [ + { + "description": "I went to the laundry room to get the broom and dustpan.", + "timestamps": { + "start": "00:38:45", + "end": "00:38:57" + } + } + ], + "question_timestamps": "00:39:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "This is your second time.", + "B": "This is your first time going to the laundry room; it’s to your back-right.", + "C": "This is your third time.", + "D": "This is your second time going to the laundry room; it’s still to your back-right.", + "E": "This is your first time going to the laundry room, but it’s located to your front-left." + }, + "answer_label": "B", + "answer_text": "This is your first time going to the laundry room; it’s to your back-right.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_9_T1", + "correct_answer": "This is your first time going to the laundry room; it’s to your back-right.", + "original_options": { + "A": "This is your second time.", + "B": "This is your first time going to the laundry room; it’s to your back-right.", + "C": "This is your third time.", + "D": "This is your second time going to the laundry room; it’s still to your back-right.", + "E": "This is your first time going to the laundry room, but it’s located to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a", + "video_length": "01:12:25", + "qid": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_9_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which time was this for me going to the laundry room?", + "answer": "This is your second time going to the laundry room, and the laundry room is to your front-left.", + "question_chinese": "这是我第几次去洗衣房？", + "answer_chinese": "这是你第二次去洗衣房，洗衣房在你的左前方。", + "evidence": [ + { + "description": "I went to the laundry room to get the broom and dustpan.", + "timestamps": { + "start": "00:38:45", + "end": "00:38:57" + } + } + ], + "question_timestamps": "00:59:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "This is your first time.", + "B": "This is your third time.", + "C": "This is your second time going to the laundry room, and the laundry room is to your front-left.", + "D": "This is your second time going to the laundry room, but it’s actually to your back-right.", + "E": "This is your fourth time going to the laundry room, and it’s still to your front-left." + }, + "answer_label": "C", + "answer_text": "This is your second time going to the laundry room, and the laundry room is to your front-left.", + "q_id": "3d3cddf9-7c61-43af-8b1a-46aa78983a5a_9_T2", + "correct_answer": "This is your second time going to the laundry room, and the laundry room is to your front-left.", + "original_options": { + "A": "This is your first time.", + "B": "This is your third time.", + "C": "This is your second time going to the laundry room, and the laundry room is to your front-left.", + "D": "This is your second time going to the laundry room, but it’s actually to your back-right.", + "E": "This is your fourth time going to the laundry room, and it’s still to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + } +] diff --git a/QAs/3d7ba3f7-f393-41ec-88a2-c33b82ec8c47-mcq.json b/QAs/3d7ba3f7-f393-41ec-88a2-c33b82ec8c47-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e76403092b2b658bc113adbe0f76737182a6ba0f --- /dev/null +++ b/QAs/3d7ba3f7-f393-41ec-88a2-c33b82ec8c47-mcq.json @@ -0,0 +1,647 @@ +[ + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_3", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the cell phone just now?", + "answer": "The cell phone was moved from on the counter to below the counter, and then brought back onto the counter; during this process, it moved from the upper center of the table to the left side of the table.", + "question_chinese": "刚刚手机的运动轨迹是怎么样的", + "answer_chinese": "手机从台上被放到了台下，又拿回台上，这个过程中从桌子中上方到了桌子左侧。", + "evidence": [ + { + "description": "The cell phone was moved from on the counter to below the counter, and then brought back onto the counter.", + "timestamps": { + "start": "00:10:40", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:11:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It moved from center to the right.", + "B": "It stayed on the counter the whole time.", + "C": "The cell phone was moved from on the counter to below the counter, and then brought back onto the counter; during this process, it moved from the upper center of the table to the left side of the table.", + "D": "The cell phone was moved from below the counter onto the counter, then shifted from the left side to the upper center.", + "E": "The cell phone went from on the counter to below it, then returned, moving from the upper center to the right side." + }, + "answer_label": "C", + "answer_text": "The cell phone was moved from on the counter to below the counter, and then brought back onto the counter; during this process, it moved from the upper center of the table to the left side of the table.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_3", + "correct_answer": "The cell phone was moved from on the counter to below the counter, and then brought back onto the counter; during this process, it moved from the upper center of the table to the left side of the table.", + "original_options": { + "A": "It moved from center to the right.", + "B": "It stayed on the counter the whole time.", + "C": "The cell phone was moved from on the counter to below the counter, and then brought back onto the counter; during this process, it moved from the upper center of the table to the left side of the table.", + "D": "The cell phone was moved from below the counter onto the counter, then shifted from the left side to the upper center.", + "E": "The cell phone went from on the counter to below it, then returned, moving from the upper center to the right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 22, + 22 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_4", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the cell phone just now?", + "answer": "The cell phone was taken into hand from the left side of the counter, was above the center of the table, and then put back.", + "question_chinese": "刚刚手机的运动轨迹是怎么样的", + "answer_chinese": "手机从台面左侧被拿到手中，处于桌面中央上空，又被放回去。", + "evidence": [ + { + "description": "The cell phone was taken into hand and then put back.", + "timestamps": { + "start": "00:12:30", + "end": "00:12:50" + } + } + ], + "question_timestamps": "00:13:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You picked it up from the right side.", + "B": "The cell phone was taken into hand from the left side of the counter, was above the center of the table, and then put back.", + "C": "It stayed on the counter.", + "D": "The cell phone was taken into hand from the right side, hovered above the table’s center, then put back.", + "E": "The cell phone was taken into hand from the left side, stayed below the table’s center, then put back." + }, + "answer_label": "B", + "answer_text": "The cell phone was taken into hand from the left side of the counter, was above the center of the table, and then put back.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_4", + "correct_answer": "The cell phone was taken into hand from the left side of the counter, was above the center of the table, and then put back.", + "original_options": { + "A": "You picked it up from the right side.", + "B": "The cell phone was taken into hand from the left side of the counter, was above the center of the table, and then put back.", + "C": "It stayed on the counter.", + "D": "The cell phone was taken into hand from the right side, hovered above the table’s center, then put back.", + "E": "The cell phone was taken into hand from the left side, stayed below the table’s center, then put back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 5, + 20, + 20 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_7", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the pot just now?", + "answer": "The pot was taken out of the cabinet and placed on the stove.", + "question_chinese": "刚刚锅的运动轨迹是怎么样的？", + "answer_chinese": "锅被从柜子里拿出来，拿到灶台上。", + "evidence": [ + { + "description": "The pot is taken out of the cabinet and placed on the stove.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The pot was taken out of the cabinet and placed on the stove.", + "B": "The pot was placed on the countertop.", + "C": "The pot was set in the sink.", + "D": "The pot was taken out of the cabinet and placed next to the stove rather than on top of it.", + "E": "The pot was taken out of the fridge and then set onto the stove as if you were about to cook." + }, + "answer_label": "A", + "answer_text": "The pot was taken out of the cabinet and placed on the stove.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_7", + "correct_answer": "The pot was taken out of the cabinet and placed on the stove.", + "original_options": { + "A": "The pot was taken out of the cabinet and placed on the stove.", + "B": "The pot was placed on the countertop.", + "C": "The pot was set in the sink.", + "D": "The pot was taken out of the cabinet and placed next to the stove rather than on top of it.", + "E": "The pot was taken out of the fridge and then set onto the stove as if you were about to cook." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 21 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_8", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the pot just now?", + "answer": "The pot was carried to the sink, filled with water, and then carried back to the stove.", + "question_chinese": "刚刚锅的运动轨迹是怎么样的？", + "answer_chinese": "锅被端到水槽，装满水再被端回灶台。", + "evidence": [ + { + "description": "The pot is carried to the sink, filled with water, and then carried back to the stove.", + "timestamps": { + "start": "00:01:25", + "end": "00:01:40" + } + } + ], + "question_timestamps": "00:03:36", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It went to the stove first.", + "B": "You left it at the sink.", + "C": "The pot was carried to the sink, filled with water, and then carried back to the stove.", + "D": "The pot was carried to the stove, filled with water there, and then carried back to the sink.", + "E": "The pot was carried to the sink, filled with water, and then carried back to the counter instead of the stove." + }, + "answer_label": "C", + "answer_text": "The pot was carried to the sink, filled with water, and then carried back to the stove.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_8", + "correct_answer": "The pot was carried to the sink, filled with water, and then carried back to the stove.", + "original_options": { + "A": "It went to the stove first.", + "B": "You left it at the sink.", + "C": "The pot was carried to the sink, filled with water, and then carried back to the stove.", + "D": "The pot was carried to the stove, filled with water there, and then carried back to the sink.", + "E": "The pot was carried to the sink, filled with water, and then carried back to the counter instead of the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 21 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_9", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the pot just now?", + "answer": "The pot was carried to the sink, then carried over the glass bowl to be poured out, and then carried back to the stove.", + "question_chinese": "刚刚锅的运动轨迹是怎么样的？", + "answer_chinese": "锅被端到水槽，接着端到玻璃碗上方倾倒，再被端回灶台。", + "evidence": [ + { + "description": "The pot is carried to the sink, then carried over the glass bowl to be poured out, and then carried back to the stove.", + "timestamps": { + "start": "00:14:00", + "end": "00:14:30" + } + } + ], + "question_timestamps": "00:15:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It went to the counter, then back.", + "B": "It went to the sink and stayed there.", + "C": "The pot was carried to the sink, then set beside the glass bowl without pouring, and then brought back to the stove.", + "D": "The pot was taken from the stove to the sink, then carried over a metal bowl to pour, then returned.", + "E": "The pot was carried to the sink, then carried over the glass bowl to be poured out, and then carried back to the stove." + }, + "answer_label": "E", + "answer_text": "The pot was carried to the sink, then carried over the glass bowl to be poured out, and then carried back to the stove.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_9", + "correct_answer": "The pot was carried to the sink, then carried over the glass bowl to be poured out, and then carried back to the stove.", + "original_options": { + "A": "It went to the counter, then back.", + "B": "It went to the sink and stayed there.", + "C": "The pot was carried to the sink, then set beside the glass bowl without pouring, and then brought back to the stove.", + "D": "The pot was taken from the stove to the sink, then carried over a metal bowl to pour, then returned.", + "E": "The pot was carried to the sink, then carried over the glass bowl to be poured out, and then carried back to the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 22, + 20 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_2_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, is there still anything left in my pot?", + "answer": "There’s still something left in the pot in front of you.", + "question_chinese": "此时我的锅里还有东西吗？", + "answer_chinese": "你面前的锅里还有东西", + "evidence": [ + { + "description": "The pot is carried to the sink, filled with water, and then carried back to the stove.", + "timestamps": { + "start": "00:01:25", + "end": "00:01:40" + } + } + ], + "question_timestamps": "00:10:58", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—what was in the pot in front of you was just poured out by you.", + "B": "There’s still something left in the pot in front of you." + }, + "answer_label": "B", + "answer_text": "There’s still something left in the pot in front of you.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_2_T2", + "correct_answer": "There’s still something left in the pot in front of you.", + "original_options": { + "A": "No—what was in the pot in front of you was just poured out by you.", + "B": "There’s still something left in the pot in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_2_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, is there still anything left in my pot?", + "answer": "No—what was in the pot in front of you was just poured out by you.", + "question_chinese": "此时我的锅里还有东西吗？", + "answer_chinese": "没有，你面前锅里的东西刚刚被你倒掉了。", + "evidence": [ + { + "description": "The pot is carried to the sink, filled with water, and then carried back to the stove.", + "timestamps": { + "start": "00:01:25", + "end": "00:01:40" + } + }, + { + "description": "Pour the contents of the pot into the glass bowl.", + "timestamps": { + "start": "00:14:10", + "end": "00:14:20" + } + } + ], + "question_timestamps": "00:14:58", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—what was in the pot in front of you was just poured out by you.", + "B": "There’s still something left in the pot in front of you." + }, + "answer_label": "A", + "answer_text": "No—what was in the pot in front of you was just poured out by you.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_2_T1", + "correct_answer": "No—what was in the pot in front of you was just poured out by you.", + "original_options": { + "A": "No—what was in the pot in front of you was just poured out by you.", + "B": "There’s still something left in the pot in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_10", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there anything in the glass bowl right now?", + "answer": "No, the glass bowl was just emptied.", + "question_chinese": "此时玻璃碗里有东西吗？", + "answer_chinese": "没有，玻璃碗刚刚被倒空。", + "evidence": [ + { + "description": "The water in the glass bowl is poured out, leaving it empty.", + "timestamps": { + "start": "00:13:30", + "end": "00:13:40" + } + } + ], + "question_timestamps": "00:13:52", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes—the contents were just poured into the glass bowl by you.", + "B": "No, the glass bowl was just emptied." + }, + "answer_label": "B", + "answer_text": "No, the glass bowl was just emptied.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_10", + "correct_answer": "No, the glass bowl was just emptied.", + "original_options": { + "A": "Yes—the contents were just poured into the glass bowl by you.", + "B": "No, the glass bowl was just emptied." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_11", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there anything in the glass bowl right now?", + "answer": "Yes. The contents of the pot were just poured into the glass bowl by you.", + "question_chinese": "此时玻璃碗里有东西吗？", + "answer_chinese": "有。刚刚锅里的东西被你倒入了玻璃碗。", + "evidence": [ + { + "description": "The water in the glass bowl is poured out, leaving it empty.", + "timestamps": { + "start": "00:13:30", + "end": "00:13:40" + } + }, + { + "description": "The pot is carried to the sink, then carried over the glass bowl to be poured out, and then carried back to the stove.", + "timestamps": { + "start": "00:14:00", + "end": "00:14:30" + } + } + ], + "question_timestamps": "00:18:03", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes. The contents of the pot were just poured into the glass bowl by you.", + "B": "No, the glass bowl was just emptied." + }, + "answer_label": "A", + "answer_text": "Yes. The contents of the pot were just poured into the glass bowl by you.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_11", + "correct_answer": "Yes. The contents of the pot were just poured into the glass bowl by you.", + "original_options": { + "A": "Yes. The contents of the pot were just poured into the glass bowl by you.", + "B": "No, the glass bowl was just emptied." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_12", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there anything in the glass bowl right now?", + "answer": "No, it was just poured into the pot by you.", + "question_chinese": "此时玻璃碗里有东西吗？", + "answer_chinese": "没有，刚被你倒入锅里了。", + "evidence": [ + { + "description": "The water in the glass bowl is poured out, leaving it empty.", + "timestamps": { + "start": "00:13:30", + "end": "00:13:40" + } + }, + { + "description": "The pot is carried to the sink, then carried over the glass bowl to be poured out, and then carried back to the stove.", + "timestamps": { + "start": "00:14:00", + "end": "00:14:30" + } + }, + { + "description": "Pour into the pot.", + "timestamps": { + "start": "00:32:29", + "end": "00:32:39" + } + } + ], + "question_timestamps": "00:33:08", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Yes—the contents were just poured into the glass bowl by you.", + "B": "No, it was just poured into the pot by you." + }, + "answer_label": "B", + "answer_text": "No, it was just poured into the pot by you.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_12", + "correct_answer": "No, it was just poured into the pot by you.", + "original_options": { + "A": "Yes—the contents were just poured into the glass bowl by you.", + "B": "No, it was just poured into the pot by you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_13", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am standing now, can I reach the refrigerator?", + "answer": "No, you cannot reach it.", + "question_chinese": "站在我现在的位置，我能够碰到冰箱吗？", + "answer_chinese": "不行，你碰不到。", + "evidence": [ + { + "description": "The refrigerator is within the range of a right arm's length.", + "timestamps": { + "start": "00:08:25", + "end": "00:08:28" + } + } + ], + "question_timestamps": "00:09:55", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you cannot reach it.", + "B": "Yes, you can reach it." + }, + "answer_label": "A", + "answer_text": "No, you cannot reach it.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_13", + "correct_answer": "No, you cannot reach it.", + "original_options": { + "A": "No, you cannot reach it.", + "B": "Yes, you can reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_14", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am standing now, can I reach the refrigerator?", + "answer": "Yes, the refrigerator is right on your right side.", + "question_chinese": "站在我现在的位置，我能够碰到冰箱吗？", + "answer_chinese": "可以，冰箱就在你右手边。", + "evidence": [ + { + "description": "The refrigerator is within the range of a right arm's length.", + "timestamps": { + "start": "00:08:25", + "end": "00:08:28" + } + } + ], + "question_timestamps": "00:29:14", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you cannot reach it.", + "B": "Yes, the refrigerator is right on your right side." + }, + "answer_label": "B", + "answer_text": "Yes, the refrigerator is right on your right side.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_14", + "correct_answer": "Yes, the refrigerator is right on your right side.", + "original_options": { + "A": "No, you cannot reach it.", + "B": "Yes, the refrigerator is right on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47", + "video_length": "00:36:03", + "qid": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_15", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am standing now, can I reach the refrigerator?", + "answer": "Yes, the refrigerator is on your left side.", + "question_chinese": "站在我现在的位置，我能够碰到冰箱吗？", + "answer_chinese": "可以，冰箱在你左手边。", + "evidence": [ + { + "description": "The refrigerator is on the left.", + "timestamps": { + "start": "00:30:00", + "end": "00:30:03" + } + } + ], + "question_timestamps": "00:30:26", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you cannot reach it.", + "B": "Yes, the refrigerator is on your left side." + }, + "answer_label": "B", + "answer_text": "Yes, the refrigerator is on your left side.", + "q_id": "3d7ba3f7-f393-41ec-88a2-c33b82ec8c47_15", + "correct_answer": "Yes, the refrigerator is on your left side.", + "original_options": { + "A": "No, you cannot reach it.", + "B": "Yes, the refrigerator is on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + } +] diff --git a/QAs/3d93c92c-8c12-4972-9f29-c8bef5c321ba-mcq.json b/QAs/3d93c92c-8c12-4972-9f29-c8bef5c321ba-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..04b0a938b1c40086e8f0bc2df238e0577cc32e93 --- /dev/null +++ b/QAs/3d93c92c-8c12-4972-9f29-c8bef5c321ba-mcq.json @@ -0,0 +1,392 @@ +[ + { + "video_uid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba", + "video_length": "00:30:00", + "qid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_9", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How will I get to the telephone booth from where I am now?", + "answer": "You will walk straight to the entrance of the small amusement park, exit the small amusement park, turn left, walk straight to the end of the road, turn right, and walk straight to the telephone booth.", + "question_chinese": "我从现在所处位置走到电话亭，我将会怎么走？", + "answer_chinese": "你将会直走到小游乐园门口，出小游乐园，左转，直走到路尽头，右转，直走到电话亭旁。", + "evidence": [ + { + "description": "I turned left from the telephone booth and entered the lane, then turned right and walked straight to the second tree, turned left again, and walked straight to the entrance of the small amusement park.", + "timestamps": { + "start": "00:11:54", + "end": "00:12:51" + } + }, + { + "description": "After entering the small amusement park, I walked straight for a few steps, then turned right and walked straight until I reached the seesaw, which is where I am now.", + "timestamps": { + "start": "00:13:17", + "end": "00:14:03" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right after exiting the park.", + "B": "You will walk straight to the entrance of the small amusement park, exit the small amusement park, turn left, walk straight to the end of the road, turn right, and walk straight to the telephone booth.", + "C": "Exit the park and go straight ahead.", + "D": "Walk to the park entrance, exit, turn right, then continue straight until you reach the telephone booth.", + "E": "Go straight to the amusement park entrance, stay inside, turn left, and follow the path to the telephone booth." + }, + "answer_label": "B", + "answer_text": "You will walk straight to the entrance of the small amusement park, exit the small amusement park, turn left, walk straight to the end of the road, turn right, and walk straight to the telephone booth.", + "q_id": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_9", + "correct_answer": "You will walk straight to the entrance of the small amusement park, exit the small amusement park, turn left, walk straight to the end of the road, turn right, and walk straight to the telephone booth.", + "original_options": { + "A": "Turn right after exiting the park.", + "B": "You will walk straight to the entrance of the small amusement park, exit the small amusement park, turn left, walk straight to the end of the road, turn right, and walk straight to the telephone booth.", + "C": "Exit the park and go straight ahead.", + "D": "Walk to the park entrance, exit, turn right, then continue straight until you reach the telephone booth.", + "E": "Go straight to the amusement park entrance, stay inside, turn left, and follow the path to the telephone booth." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 19 + ] + }, + { + "video_uid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba", + "video_length": "00:30:00", + "qid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_10", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How will I get to the telephone booth from where I am now?", + "answer": "You will slide down the slide, then walk to the iron gate in the lower left corner of the small amusement park, exit the small amusement park, turn left, walk straight to the end of the road, turn right, and walk straight to the telephone booth.", + "question_chinese": "我从现在所处位置走到电话亭，我将会怎么走？", + "answer_chinese": "你将从滑梯滑下，然后走到左后方小游乐园铁门处，出小游乐园，左转，直走到路的尽头，右转，直走到电话亭。", + "evidence": [ + { + "description": "From the telephone booth, walk straight to the second tree, turn left, and walk straight to the entrance of the small amusement park.", + "timestamps": { + "start": "00:11:53", + "end": "00:12:49" + } + }, + { + "description": "The climbing arch is in front of the right side of the small amusement park's iron gate.", + "timestamps": { + "start": "00:13:05", + "end": "00:13:50" + } + }, + { + "description": "The climbing arch is to the left front of your current position.", + "timestamps": { + "start": "00:29:11", + "end": "00:29:18" + } + } + ], + "question_timestamps": "00:29:18", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You will slide down the slide, then walk to the iron gate in the lower left corner of the small amusement park, exit the small amusement park, turn left, walk straight to the end of the road, turn right, and walk straight to the telephone booth.", + "B": "Go to the lower right iron gate.", + "C": "Turn left at the road’s end.", + "D": "Slide down, reach the lower left iron gate, exit, then turn right and walk to the telephone booth.", + "E": "Slide down the slide, head to the lower left gate, exit, then walk straight without turning to the telephone booth." + }, + "answer_label": "A", + "answer_text": "You will slide down the slide, then walk to the iron gate in the lower left corner of the small amusement park, exit the small amusement park, turn left, walk straight to the end of the road, turn right, and walk straight to the telephone booth.", + "q_id": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_10", + "correct_answer": "You will slide down the slide, then walk to the iron gate in the lower left corner of the small amusement park, exit the small amusement park, turn left, walk straight to the end of the road, turn right, and walk straight to the telephone booth.", + "original_options": { + "A": "You will slide down the slide, then walk to the iron gate in the lower left corner of the small amusement park, exit the small amusement park, turn left, walk straight to the end of the road, turn right, and walk straight to the telephone booth.", + "B": "Go to the lower right iron gate.", + "C": "Turn left at the road’s end.", + "D": "Slide down, reach the lower left iron gate, exit, then turn right and walk to the telephone booth.", + "E": "Slide down the slide, head to the lower left gate, exit, then walk straight without turning to the telephone booth." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 20 + ] + }, + { + "video_uid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba", + "video_length": "00:30:00", + "qid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_14", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to walk to the iron gate of the small amusement park. How should I go?", + "answer": "You will walk forward, go around the seesaw and the trash can, and then reach the iron gate.", + "question_chinese": "我要走到小游乐园的铁门口，我将要怎么走？", + "answer_chinese": "你将会向前走并绕过跷跷板和垃圾桶，即可到达铁门。", + "evidence": [ + { + "description": "The iron gate is in front of me, separated from me by a seesaw and a trash can.", + "timestamps": { + "start": "00:14:03", + "end": "00:14:52" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You will walk forward, go around the seesaw and the trash can, and then reach the iron gate.", + "B": "Turn left and walk to the gate.", + "C": "Go straight to the park entrance.", + "D": "Walk forward, pass the seesaw directly, then head right to reach the iron gate.", + "E": "Slide down the slide first, then walk across the grass to the iron gate near the corner." + }, + "answer_label": "A", + "answer_text": "You will walk forward, go around the seesaw and the trash can, and then reach the iron gate.", + "q_id": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_14", + "correct_answer": "You will walk forward, go around the seesaw and the trash can, and then reach the iron gate.", + "original_options": { + "A": "You will walk forward, go around the seesaw and the trash can, and then reach the iron gate.", + "B": "Turn left and walk to the gate.", + "C": "Go straight to the park entrance.", + "D": "Walk forward, pass the seesaw directly, then head right to reach the iron gate.", + "E": "Slide down the slide first, then walk across the grass to the iron gate near the corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 17 + ] + }, + { + "video_uid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba", + "video_length": "00:30:00", + "qid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_13", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to walk to the iron gate of the small amusement park. How should I go?", + "answer": "You will slide down the slide, turn left to walk onto the grass, then turn left again and walk straight to the iron gate.", + "question_chinese": "我要走到小游乐园的铁门口，我将要怎么走？", + "answer_chinese": "你将要从滑梯上滑下，向左转走到草地上，向左转直走到铁门。", + "evidence": [ + { + "description": "The slide facility is in front of the right side of the iron gate.", + "timestamps": { + "start": "00:12:58", + "end": "00:13:06" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Slide down, then turn right on grass.", + "B": "Slide down, then go straight to the gate.", + "C": "Slide down the slide, turn left onto the grass, then turn right and walk straight to the iron gate.", + "D": "Slide down the slide, turn right onto the grass, then turn right again and walk straight to the iron gate.", + "E": "You will slide down the slide, turn left to walk onto the grass, then turn left again and walk straight to the iron gate." + }, + "answer_label": "E", + "answer_text": "You will slide down the slide, turn left to walk onto the grass, then turn left again and walk straight to the iron gate.", + "q_id": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_13", + "correct_answer": "You will slide down the slide, turn left to walk onto the grass, then turn left again and walk straight to the iron gate.", + "original_options": { + "A": "Slide down, then turn right on grass.", + "B": "Slide down, then go straight to the gate.", + "C": "Slide down the slide, turn left onto the grass, then turn right and walk straight to the iron gate.", + "D": "Slide down the slide, turn right onto the grass, then turn right again and walk straight to the iron gate.", + "E": "You will slide down the slide, turn left to walk onto the grass, then turn left again and walk straight to the iron gate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 20 + ] + }, + { + "video_uid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba", + "video_length": "00:30:00", + "qid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_9_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where are the two kids relative to me, respectively?", + "answer": "They’re on the seesaw near the iron gate in the small playground.", + "question_chinese": "现在两个小孩分别在我的哪里？", + "answer_chinese": "他们在小游乐园里的靠近铁门的跷跷板上。", + "evidence": [ + { + "description": "They are on the seesaw near the iron gate in the small amusement park.", + "timestamps": { + "start": "00:13:42", + "end": "00:14:21" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They’re by the slide.", + "B": "They’re on the seesaw near the iron gate in the small playground.", + "C": "They’re on the swings away.", + "D": "They’re on the seesaw farther from the iron gate, deeper inside the small playground.", + "E": "They’re near the iron gate, standing beside the seesaw instead of sitting on it." + }, + "answer_label": "B", + "answer_text": "They’re on the seesaw near the iron gate in the small playground.", + "q_id": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_9_T1", + "correct_answer": "They’re on the seesaw near the iron gate in the small playground.", + "original_options": { + "A": "They’re by the slide.", + "B": "They’re on the seesaw near the iron gate in the small playground.", + "C": "They’re on the swings away.", + "D": "They’re on the seesaw farther from the iron gate, deeper inside the small playground.", + "E": "They’re near the iron gate, standing beside the seesaw instead of sitting on it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba", + "video_length": "00:30:00", + "qid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_9_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where are the two kids relative to me, respectively?", + "answer": "They’re on the seesaw near the iron gate in the small playground, both in front of you and behind you.", + "question_chinese": "现在两个小孩分别在我的哪里？", + "answer_chinese": "他们在小游乐园里的靠近铁门的跷跷板上，都在我的前面和后面。", + "evidence": [ + { + "description": "They are on the seesaw near the iron gate in the small amusement park.", + "timestamps": { + "start": "00:13:42", + "end": "00:14:21" + } + } + ], + "question_timestamps": "00:15:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Both kids are in front of you.", + "B": "Both kids are behind you.", + "C": "They’re on the seesaw near the iron gate in the small playground, both in front of you and behind you.", + "D": "They’re on the seesaw near the iron gate, with one to your left and one to your right.", + "E": "They’re on the seesaw far from the iron gate, with one in front of you and one behind you." + }, + "answer_label": "C", + "answer_text": "They’re on the seesaw near the iron gate in the small playground, both in front of you and behind you.", + "q_id": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_9_T2", + "correct_answer": "They’re on the seesaw near the iron gate in the small playground, both in front of you and behind you.", + "original_options": { + "A": "Both kids are in front of you.", + "B": "Both kids are behind you.", + "C": "They’re on the seesaw near the iron gate in the small playground, both in front of you and behind you.", + "D": "They’re on the seesaw near the iron gate, with one to your left and one to your right.", + "E": "They’re on the seesaw far from the iron gate, with one in front of you and one behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 19, + 20 + ] + }, + { + "video_uid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba", + "video_length": "00:30:00", + "qid": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_6_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me—the seesaw or the tree?", + "answer": "The seesaw is closer to you.", + "question_chinese": "现在跷跷板和树哪个离我更近？", + "answer_chinese": "跷跷板离我更近.", + "evidence": [ + { + "description": "The seesaw is within one step directly in front of where I am now.", + "timestamps": { + "start": "00:14:39", + "end": "00:14:43" + } + }, + { + "description": "The tree is at least five steps away from my current position.", + "timestamps": { + "start": "00:13:48", + "end": "00:14:53" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The tree is closer to you.", + "B": "The seesaw is closer to you." + }, + "answer_label": "B", + "answer_text": "The seesaw is closer to you.", + "q_id": "3d93c92c-8c12-4972-9f29-c8bef5c321ba_6_T1", + "correct_answer": "The seesaw is closer to you.", + "original_options": { + "A": "The tree is closer to you.", + "B": "The seesaw is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/3e7e162a-1cb3-4bdc-b6c2-e14845298360-mcq.json b/QAs/3e7e162a-1cb3-4bdc-b6c2-e14845298360-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b5be3ae320c68f9816fcbf553d5f285a18c46899 --- /dev/null +++ b/QAs/3e7e162a-1cb3-4bdc-b6c2-e14845298360-mcq.json @@ -0,0 +1,1312 @@ +[ + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_15", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans did you buy?", + "answer": "You bought six cans.", + "question_chinese": "你买了几个罐头？", + "answer_chinese": "你买了六个罐头。", + "evidence": [ + { + "description": "Four cans", + "timestamps": { + "start": "00:21:34", + "end": "00:21:49" + } + }, + { + "description": "Two cans", + "timestamps": { + "start": "00:18:20", + "end": "00:18:38" + } + } + ], + "question_timestamps": "00:28:18", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You bought five cans.", + "B": "You bought seven cans.", + "C": "You bought four cans in total during this shopping trip at the store.", + "D": "You bought six cans.", + "E": "You bought eight cans altogether after picking up two extra cans from the shelf." + }, + "answer_label": "D", + "answer_text": "You bought six cans.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_15", + "correct_answer": "You bought six cans.", + "original_options": { + "A": "You bought five cans.", + "B": "You bought seven cans.", + "C": "You bought four cans in total during this shopping trip at the store.", + "D": "You bought six cans.", + "E": "You bought eight cans altogether after picking up two extra cans from the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_16", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans did you buy?", + "answer": "You bought six cans.", + "question_chinese": "你买了几个罐头？", + "answer_chinese": "你买了六个罐头。", + "evidence": [ + { + "description": "Four cans", + "timestamps": { + "start": "00:21:34", + "end": "00:21:49" + } + }, + { + "description": "Two cans", + "timestamps": { + "start": "00:18:20", + "end": "00:18:38" + } + } + ], + "question_timestamps": "00:56:17", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You bought six cans.", + "B": "You bought five cans.", + "C": "You bought seven cans.", + "D": "You bought eight cans total after grabbing two extra cans from the shelf.", + "E": "You bought nine cans altogether by the time you checked out at the register." + }, + "answer_label": "A", + "answer_text": "You bought six cans.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_16", + "correct_answer": "You bought six cans.", + "original_options": { + "A": "You bought six cans.", + "B": "You bought five cans.", + "C": "You bought seven cans.", + "D": "You bought eight cans total after grabbing two extra cans from the shelf.", + "E": "You bought nine cans altogether by the time you checked out at the register." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_14", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of milk did you buy?", + "answer": "You bought four boxes of milk.", + "question_chinese": "你买了几盒牛奶？", + "answer_chinese": "你买了四盒牛奶。", + "evidence": [ + { + "description": "You bought four boxes of milk.", + "timestamps": { + "start": "00:17:20", + "end": "00:17:25" + } + } + ], + "question_timestamps": "00:28:18", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You bought three boxes of milk.", + "B": "You bought five boxes of milk.", + "C": "You bought four boxes of milk.", + "D": "You bought two boxes of milk in total during this shopping trip at the store.", + "E": "You bought seven boxes of milk altogether after adding extra cartons before you checked out." + }, + "answer_label": "C", + "answer_text": "You bought four boxes of milk.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_14", + "correct_answer": "You bought four boxes of milk.", + "original_options": { + "A": "You bought three boxes of milk.", + "B": "You bought five boxes of milk.", + "C": "You bought four boxes of milk.", + "D": "You bought two boxes of milk in total during this shopping trip at the store.", + "E": "You bought seven boxes of milk altogether after adding extra cartons before you checked out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_13", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of milk did you buy?", + "answer": "You bought four boxes of milk.", + "question_chinese": "你买了几盒牛奶？", + "answer_chinese": "你买了四盒牛奶。", + "evidence": [ + { + "description": "You bought four boxes of milk.", + "timestamps": { + "start": "00:17:20", + "end": "00:17:25" + } + } + ], + "question_timestamps": "00:56:18", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You bought three boxes of milk.", + "B": "You bought four boxes of milk.", + "C": "You bought five boxes of milk.", + "D": "You bought six boxes of milk in total after picking up two more from the shelf.", + "E": "You bought eight boxes of milk altogether by checkout after adding more cartons at the end." + }, + "answer_label": "B", + "answer_text": "You bought four boxes of milk.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_13", + "correct_answer": "You bought four boxes of milk.", + "original_options": { + "A": "You bought three boxes of milk.", + "B": "You bought four boxes of milk.", + "C": "You bought five boxes of milk.", + "D": "You bought six boxes of milk in total after picking up two more from the shelf.", + "E": "You bought eight boxes of milk altogether by checkout after adding more cartons at the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_2_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, what kinds of items have I put in the shopping cart?", + "answer": "You put eggs and meat into the shopping cart.", + "question_chinese": "到目前为止，我在购物车里放了哪几种东西？", + "answer_chinese": "你在购物车里放了鸡蛋、肉。", + "evidence": [ + { + "description": "Egg", + "timestamps": { + "start": "00:08:58", + "end": "00:09:05" + } + }, + { + "description": "Meat", + "timestamps": { + "start": "00:10:02", + "end": "00:10:08" + } + } + ], + "question_timestamps": "00:10:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put only eggs in the cart.", + "B": "You put eggs and meat into the shopping cart.", + "C": "You put only meat in the cart.", + "D": "You put eggs, vegetables, and yogurt into the shopping cart during this trip so far.", + "E": "You put meat, vegetables, and yogurt into the shopping cart so far, but no eggs." + }, + "answer_label": "B", + "answer_text": "You put eggs and meat into the shopping cart.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_2_T2", + "correct_answer": "You put eggs and meat into the shopping cart.", + "original_options": { + "A": "You put only eggs in the cart.", + "B": "You put eggs and meat into the shopping cart.", + "C": "You put only meat in the cart.", + "D": "You put eggs, vegetables, and yogurt into the shopping cart during this trip so far.", + "E": "You put meat, vegetables, and yogurt into the shopping cart so far, but no eggs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_2_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, what kinds of items have I put in the shopping cart?", + "answer": "You put eggs, meat, vegetables, and yogurt into the shopping cart.", + "question_chinese": "到目前为止，我在购物车里放了哪几种东西？", + "answer_chinese": "你在购物车里放了鸡蛋、肉、蔬菜、酸奶。", + "evidence": [ + { + "description": "Egg", + "timestamps": { + "start": "00:08:58", + "end": "00:09:05" + } + }, + { + "description": "Meat", + "timestamps": { + "start": "00:10:02", + "end": "00:10:08" + } + }, + { + "description": "Vegetables", + "timestamps": { + "start": "00:10:53", + "end": "00:11:28" + } + }, + { + "description": "Yogurt", + "timestamps": { + "start": "00:12:30", + "end": "00:12:59" + } + } + ], + "question_timestamps": "00:14:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put eggs and meat in.", + "B": "You put eggs, meat, vegetables, and yogurt into the shopping cart.", + "C": "You put only vegetables in.", + "D": "You put eggs, meat, and vegetables into the shopping cart, but you skipped yogurt entirely.", + "E": "You put meat, vegetables, and yogurt into the shopping cart, but you did not add eggs." + }, + "answer_label": "B", + "answer_text": "You put eggs, meat, vegetables, and yogurt into the shopping cart.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_2_T1", + "correct_answer": "You put eggs, meat, vegetables, and yogurt into the shopping cart.", + "original_options": { + "A": "You put eggs and meat in.", + "B": "You put eggs, meat, vegetables, and yogurt into the shopping cart.", + "C": "You put only vegetables in.", + "D": "You put eggs, meat, and vegetables into the shopping cart, but you skipped yogurt entirely.", + "E": "You put meat, vegetables, and yogurt into the shopping cart, but you did not add eggs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 16 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_5_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to return the shopping cart to where it was originally parked, how should I go?", + "answer": "You should turn around, walk straight to the supermarket entrance, then turn left and walk straight to where the shopping carts are parked.", + "question_chinese": "如果我现在想送还购物车到最开始停放的地方，我该怎么走？", + "answer_chinese": "你应该向后转，直走到超市门口，然后左转直走到停放购物车的地方。", + "evidence": [ + { + "description": "When exiting through the supermarket entrance, the area where shopping carts are parked is on the left side of the entrance.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:35" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right at the entrance.", + "B": "You should turn around, walk straight to the supermarket entrance, then turn left and walk straight to where the shopping carts are parked.", + "C": "Walk straight; don’t turn left.", + "D": "Turn around, go to the entrance, then turn right and continue straight to the cart area.", + "E": "Keep walking forward to the entrance, then turn left and backtrack slightly to reach the carts." + }, + "answer_label": "B", + "answer_text": "You should turn around, walk straight to the supermarket entrance, then turn left and walk straight to where the shopping carts are parked.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_5_T1", + "correct_answer": "You should turn around, walk straight to the supermarket entrance, then turn left and walk straight to where the shopping carts are parked.", + "original_options": { + "A": "Turn right at the entrance.", + "B": "You should turn around, walk straight to the supermarket entrance, then turn left and walk straight to where the shopping carts are parked.", + "C": "Walk straight; don’t turn left.", + "D": "Turn around, go to the entrance, then turn right and continue straight to the cart area.", + "E": "Keep walking forward to the entrance, then turn left and backtrack slightly to reach the carts." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_5_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to return the shopping cart to where it was originally parked, how should I go?", + "answer": "You should walk straight out to the supermarket entrance, then turn left and walk straight to where the shopping carts are parked.", + "question_chinese": "如果我现在想送还购物车到最开始停放的地方，我该怎么走？", + "answer_chinese": "你应该在直走出超市门口后向左转，然后直走到停放购物车的地方。", + "evidence": [ + { + "description": "When exiting through the supermarket entrance, the area where shopping carts are parked is on the left side of the entrance.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:35" + } + } + ], + "question_timestamps": "00:48:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right at the entrance.", + "B": "You should walk straight out to the supermarket entrance, then turn left and walk straight to where the shopping carts are parked.", + "C": "Head back inside, then turn left.", + "D": "Walk straight to the entrance, then turn right and continue straight to the cart parking area.", + "E": "Turn around first, walk to the entrance, then turn left and stop at the checkout instead." + }, + "answer_label": "B", + "answer_text": "You should walk straight out to the supermarket entrance, then turn left and walk straight to where the shopping carts are parked.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_5_T2", + "correct_answer": "You should walk straight out to the supermarket entrance, then turn left and walk straight to where the shopping carts are parked.", + "original_options": { + "A": "Turn right at the entrance.", + "B": "You should walk straight out to the supermarket entrance, then turn left and walk straight to where the shopping carts are parked.", + "C": "Head back inside, then turn left.", + "D": "Walk straight to the entrance, then turn right and continue straight to the cart parking area.", + "E": "Turn around first, walk to the entrance, then turn left and stop at the checkout instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_10", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I exit through the supermarket entrance, if I want to return the shopping cart to where it was originally parked, how should I go?", + "answer": "You should turn left and walk straight to where the shopping carts are parked.", + "question_chinese": "当我从超市大门出来时，如果我想送还购物车到最开始停放的地方，我该怎么走？", + "answer_chinese": "你应该向左转，直走到停放购物车的地方。", + "evidence": [ + { + "description": "When exiting through the supermarket entrance, the area where shopping carts are parked is on the left side of the entrance.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:35" + } + } + ], + "question_timestamps": "00:14:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right and walk straight.", + "B": "Go straight ahead instead.", + "C": "Turn left, then immediately turn back and walk toward the checkout counters rather than the cart area.", + "D": "Turn around and walk straight back through the doors, then continue until you reach the cart bay.", + "E": "You should turn left and walk straight to where the shopping carts are parked." + }, + "answer_label": "E", + "answer_text": "You should turn left and walk straight to where the shopping carts are parked.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_10", + "correct_answer": "You should turn left and walk straight to where the shopping carts are parked.", + "original_options": { + "A": "Turn right and walk straight.", + "B": "Go straight ahead instead.", + "C": "Turn left, then immediately turn back and walk toward the checkout counters rather than the cart area.", + "D": "Turn around and walk straight back through the doors, then continue until you reach the cart bay.", + "E": "You should turn left and walk straight to where the shopping carts are parked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 17 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_9", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I exit through the supermarket entrance, if I want to return the shopping cart to where it was originally parked, how should I go?", + "answer": "You should turn left and walk straight to where the shopping carts are parked.", + "question_chinese": "当我从超市大门出来时，如果我想送还购物车到最开始停放的地方，我该怎么走？", + "answer_chinese": "你应该向左转，直走到停放购物车的地方。", + "evidence": [ + { + "description": "When exiting through the supermarket entrance, the area where shopping carts are parked is on the left side of the entrance.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:35" + } + } + ], + "question_timestamps": "00:42:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should turn left and walk straight to where the shopping carts are parked.", + "B": "Turn right and walk straight.", + "C": "Walk straight without turning.", + "D": "Turn left and walk straight, but head toward the bag storage area instead of the cart return.", + "E": "Turn around and walk straight toward the checkout counters, not toward the carts outside." + }, + "answer_label": "A", + "answer_text": "You should turn left and walk straight to where the shopping carts are parked.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_9", + "correct_answer": "You should turn left and walk straight to where the shopping carts are parked.", + "original_options": { + "A": "You should turn left and walk straight to where the shopping carts are parked.", + "B": "Turn right and walk straight.", + "C": "Walk straight without turning.", + "D": "Turn left and walk straight, but head toward the bag storage area instead of the cart return.", + "E": "Turn around and walk straight toward the checkout counters, not toward the carts outside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 14 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_3_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "After I got out of the car, was the rear door closed?", + "answer": "The car’s rear door was open.", + "question_chinese": "我下车后，车后门是不是关着的？", + "answer_chinese": "车后门是开着的。", + "evidence": [ + { + "description": "The rear door of the car is opened.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:46" + } + } + ], + "question_timestamps": "00:04:57", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The car’s rear door was closed.", + "B": "The car’s rear door was open." + }, + "answer_label": "B", + "answer_text": "The car’s rear door was open.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_3_T2", + "correct_answer": "The car’s rear door was open.", + "original_options": { + "A": "The car’s rear door was closed.", + "B": "The car’s rear door was open." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_3_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "After I got out of the car, was the rear door closed?", + "answer": "The car’s rear door was closed.", + "question_chinese": "我下车后，车后门是不是关着的？", + "answer_chinese": "车后门是关着的。", + "evidence": [ + { + "description": "The rear door of the car is closed.", + "timestamps": { + "start": "00:04:30", + "end": "00:05:02" + } + } + ], + "question_timestamps": "00:42:54", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The car’s rear door was closed.", + "B": "The car’s rear door was open." + }, + "answer_label": "A", + "answer_text": "The car’s rear door was closed.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_3_T1", + "correct_answer": "The car’s rear door was closed.", + "original_options": { + "A": "The car’s rear door was closed.", + "B": "The car’s rear door was open." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_8_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans did I buy?", + "answer": "You bought two cans.", + "question_chinese": "我买了几个罐头？", + "answer_chinese": "你买了两个罐头。", + "evidence": [ + { + "description": "Two cans", + "timestamps": { + "start": "00:18:20", + "end": "00:18:38" + } + } + ], + "question_timestamps": "00:19:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You bought one can.", + "B": "You bought three cans.", + "C": "You bought four cans in total, not just a couple you grabbed at the last minute.", + "D": "You bought two cans.", + "E": "You didn’t buy any cans; you left without adding them to your purchase at checkout." + }, + "answer_label": "D", + "answer_text": "You bought two cans.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_8_T1", + "correct_answer": "You bought two cans.", + "original_options": { + "A": "You bought one can.", + "B": "You bought three cans.", + "C": "You bought four cans in total, not just a couple you grabbed at the last minute.", + "D": "You bought two cans.", + "E": "You didn’t buy any cans; you left without adding them to your purchase at checkout." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_8_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans did I buy?", + "answer": "You bought six cans.", + "question_chinese": "我买了几个罐头？", + "answer_chinese": "你买了六个罐头。", + "evidence": [ + { + "description": "Four cans", + "timestamps": { + "start": "00:21:34", + "end": "00:21:49" + } + }, + { + "description": "Two cans", + "timestamps": { + "start": "00:18:20", + "end": "00:18:38" + } + } + ], + "question_timestamps": "00:56:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You bought five cans.", + "B": "You bought seven cans.", + "C": "You bought four cans, not a half-dozen, so you didn’t reach six in the end.", + "D": "You bought eight cans altogether, more than you originally planned to get.", + "E": "You bought six cans." + }, + "answer_label": "E", + "answer_text": "You bought six cans.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_8_T2", + "correct_answer": "You bought six cans.", + "original_options": { + "A": "You bought five cans.", + "B": "You bought seven cans.", + "C": "You bought four cans, not a half-dozen, so you didn’t reach six in the end.", + "D": "You bought eight cans altogether, more than you originally planned to get.", + "E": "You bought six cans." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 12 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_7_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kinds of fruit did I buy?", + "answer": "You bought one type of fruit.", + "question_chinese": "我买了几种水果？", + "answer_chinese": "你买了一种水果。", + "evidence": [ + { + "description": "You bought a kind of fruit.", + "timestamps": { + "start": "00:25:30", + "end": "00:26:10" + } + } + ], + "question_timestamps": "00:26:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You bought zero types of fruit.", + "B": "You bought two types of fruit.", + "C": "You bought one type of fruit.", + "D": "You bought three different kinds of fruit during the trip, not just a single variety.", + "E": "You bought four types of fruit, choosing several varieties rather than sticking to one." + }, + "answer_label": "C", + "answer_text": "You bought one type of fruit.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_7_T1", + "correct_answer": "You bought one type of fruit.", + "original_options": { + "A": "You bought zero types of fruit.", + "B": "You bought two types of fruit.", + "C": "You bought one type of fruit.", + "D": "You bought three different kinds of fruit during the trip, not just a single variety.", + "E": "You bought four types of fruit, choosing several varieties rather than sticking to one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 14 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_7_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kinds of fruit did I buy?", + "answer": "You bought two types of fruit.", + "question_chinese": "我买了几种水果？", + "answer_chinese": "你买了两种水果。", + "evidence": [ + { + "description": "You bought oranges.", + "timestamps": { + "start": "00:25:30", + "end": "00:26:10" + } + }, + { + "description": "You bought apples.", + "timestamps": { + "start": "00:26:15", + "end": "00:26:32" + } + } + ], + "question_timestamps": "00:56:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You bought one type of fruit.", + "B": "You bought three types of fruit.", + "C": "You bought four different kinds of fruit during your shopping trip today at the store.", + "D": "You bought two types of fruit.", + "E": "You ended up buying no fruit types at all from the store that day." + }, + "answer_label": "D", + "answer_text": "You bought two types of fruit.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_7_T2", + "correct_answer": "You bought two types of fruit.", + "original_options": { + "A": "You bought one type of fruit.", + "B": "You bought three types of fruit.", + "C": "You bought four different kinds of fruit during your shopping trip today at the store.", + "D": "You bought two types of fruit.", + "E": "You ended up buying no fruit types at all from the store that day." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 14 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_6_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of meat did I buy?", + "answer": "You bought two packs of meat.", + "question_chinese": "我买了多少盒肉？", + "answer_chinese": "你买了两盒肉。", + "evidence": [ + { + "description": "You bought two boxes of meat.", + "timestamps": { + "start": "00:09:20", + "end": "00:09:29" + } + } + ], + "question_timestamps": "00:09:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You bought one pack of meat.", + "B": "You bought two packs of meat.", + "C": "You bought three packs of meat.", + "D": "You picked up four packs of meat before heading to checkout at the store.", + "E": "You didn’t buy any packs of meat during the trip to the store." + }, + "answer_label": "B", + "answer_text": "You bought two packs of meat.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_6_T1", + "correct_answer": "You bought two packs of meat.", + "original_options": { + "A": "You bought one pack of meat.", + "B": "You bought two packs of meat.", + "C": "You bought three packs of meat.", + "D": "You picked up four packs of meat before heading to checkout at the store.", + "E": "You didn’t buy any packs of meat during the trip to the store." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_11", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of meat did I buy?", + "answer": "You bought three boxes of meat.", + "question_chinese": "我买了多少盒肉？", + "answer_chinese": "你买了三盒肉。", + "evidence": [ + { + "description": "You bought three boxes of meat.", + "timestamps": { + "start": "00:09:20", + "end": "00:09:49" + } + } + ], + "question_timestamps": "00:14:54", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You bought three boxes of meat.", + "B": "You bought two boxes of meat.", + "C": "You bought four boxes of meat.", + "D": "You ended up buying five boxes of meat in total during the trip.", + "E": "You only purchased one box of meat before leaving the store that day." + }, + "answer_label": "A", + "answer_text": "You bought three boxes of meat.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_11", + "correct_answer": "You bought three boxes of meat.", + "original_options": { + "A": "You bought three boxes of meat.", + "B": "You bought two boxes of meat.", + "C": "You bought four boxes of meat.", + "D": "You ended up buying five boxes of meat in total during the trip.", + "E": "You only purchased one box of meat before leaving the store that day." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 13 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_6_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of meat did I buy?", + "answer": "You bought three packs of meat.", + "question_chinese": "我买了多少盒肉？", + "answer_chinese": "你买了三盒肉。", + "evidence": [ + { + "description": "You bought three boxes of meat.", + "timestamps": { + "start": "00:09:20", + "end": "00:09:49" + } + } + ], + "question_timestamps": "00:56:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You bought two packs of meat.", + "B": "You bought four packs of meat.", + "C": "You bought three packs of meat.", + "D": "You purchased five packs of meat by the time you checked out at the store.", + "E": "You only bought one pack of meat during the shopping trip that day." + }, + "answer_label": "C", + "answer_text": "You bought three packs of meat.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_6_T2", + "correct_answer": "You bought three packs of meat.", + "original_options": { + "A": "You bought two packs of meat.", + "B": "You bought four packs of meat.", + "C": "You bought three packs of meat.", + "D": "You purchased five packs of meat by the time you checked out at the store.", + "E": "You only bought one pack of meat during the shopping trip that day." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 13 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_1_T1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of items did I take down from the car?", + "answer": "You took three bags off the car.", + "question_chinese": "我从车上拿下来了哪几种东西？", + "answer_chinese": "你从车上拿下来了三个袋子。", + "evidence": [ + { + "description": "You took down three bags from the car.", + "timestamps": { + "start": "00:04:35", + "end": "00:04:39" + } + } + ], + "question_timestamps": "00:04:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took two bags off the car.", + "B": "You took three bags off the car.", + "C": "You took four bags off the car.", + "D": "You carried five bags down from the car in one trip to the house.", + "E": "You only removed one bag from the car before stopping for a break." + }, + "answer_label": "B", + "answer_text": "You took three bags off the car.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_1_T1", + "correct_answer": "You took three bags off the car.", + "original_options": { + "A": "You took two bags off the car.", + "B": "You took three bags off the car.", + "C": "You took four bags off the car.", + "D": "You carried five bags down from the car in one trip to the house.", + "E": "You only removed one bag from the car before stopping for a break." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 13 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of things did I take off the car?", + "answer": "You took down three buckets of water and three bags from the car.", + "question_chinese": "我从车上拿下来了哪几种东西？", + "answer_chinese": "你从车上拿下来了三桶水和三个袋子。", + "evidence": [ + { + "description": "You took down three buckets of water and three bags from the car.", + "timestamps": { + "start": "00:04:35", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:42:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four buckets and four bags.", + "B": "Five buckets and three bags.", + "C": "You took down three buckets of water and three bags from the car.", + "D": "You took down three buckets of water and five bags from the car.", + "E": "You took down four buckets of water and three bags from the car." + }, + "answer_label": "C", + "answer_text": "You took down three buckets of water and three bags from the car.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_2", + "correct_answer": "You took down three buckets of water and three bags from the car.", + "original_options": { + "A": "Four buckets and four bags.", + "B": "Five buckets and three bags.", + "C": "You took down three buckets of water and three bags from the car.", + "D": "You took down three buckets of water and five bags from the car.", + "E": "You took down four buckets of water and three bags from the car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_1_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of items did I take down from the car?", + "answer": "You took three bags and three buckets off the car.", + "question_chinese": "我从车上拿下来了哪几种东西？", + "answer_chinese": "你从车上拿下来了三个袋子和三个水桶。", + "evidence": [ + { + "description": "You took down three buckets of water and three bags from the car.", + "timestamps": { + "start": "00:04:35", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:56:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two bags and three buckets.", + "B": "Four bags and three buckets.", + "C": "You took three bags and three buckets off the car.", + "D": "You took five bags and three buckets off the car in total.", + "E": "You removed one bag and three buckets from the car altogether that day." + }, + "answer_label": "C", + "answer_text": "You took three bags and three buckets off the car.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_1_T2", + "correct_answer": "You took three bags and three buckets off the car.", + "original_options": { + "A": "Two bags and three buckets.", + "B": "Four bags and three buckets.", + "C": "You took three bags and three buckets off the car.", + "D": "You took five bags and three buckets off the car in total.", + "E": "You removed one bag and three buckets from the car altogether that day." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_4_T1", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many onions did I take?", + "answer": "You took three onions.", + "question_chinese": "我拿了多少个洋葱？", + "answer_chinese": "你拿了三个洋葱。", + "evidence": [ + { + "description": "You have three onions.", + "timestamps": { + "start": "00:30:22", + "end": "00:30:37" + } + } + ], + "question_timestamps": "00:30:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took two onions.", + "B": "You took four onions.", + "C": "You grabbed five onions to take with you from the pile at the stand.", + "D": "You only took one onion before moving on to the next item at the stall.", + "E": "You took three onions." + }, + "answer_label": "E", + "answer_text": "You took three onions.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_4_T1", + "correct_answer": "You took three onions.", + "original_options": { + "A": "You took two onions.", + "B": "You took four onions.", + "C": "You grabbed five onions to take with you from the pile at the stand.", + "D": "You only took one onion before moving on to the next item at the stall.", + "E": "You took three onions." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360", + "video_length": "00:56:18", + "qid": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_4_T2", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many onions did I take?", + "answer": "You took four onions.", + "question_chinese": "我拿了多少个洋葱？", + "answer_chinese": "你拿了四个洋葱。", + "evidence": [ + { + "description": "You have four onions.", + "timestamps": { + "start": "00:30:22", + "end": "00:30:49" + } + } + ], + "question_timestamps": "00:42:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took four onions.", + "B": "You took two onions.", + "C": "You took six onions.", + "D": "You took three onions in total, picking them up one after another before moving on.", + "E": "You took five onions altogether, gathering an extra one at the end before stopping." + }, + "answer_label": "A", + "answer_text": "You took four onions.", + "q_id": "3e7e162a-1cb3-4bdc-b6c2-e14845298360_4_T2", + "correct_answer": "You took four onions.", + "original_options": { + "A": "You took four onions.", + "B": "You took two onions.", + "C": "You took six onions.", + "D": "You took three onions in total, picking them up one after another before moving on.", + "E": "You took five onions altogether, gathering an extra one at the end before stopping." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + } +] diff --git a/QAs/3ec3eab7-842d-409d-8866-42ddcbd24cd9-mcq.json b/QAs/3ec3eab7-842d-409d-8866-42ddcbd24cd9-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5db9257b2b60fb04739f9dc78d2c169a09f36d62 --- /dev/null +++ b/QAs/3ec3eab7-842d-409d-8866-42ddcbd24cd9-mcq.json @@ -0,0 +1,1096 @@ +[ + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_8", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates of raw doughnuts have I prepared so far?", + "answer": "So far, you have prepared a total of 4 plates of raw doughnuts.", + "question_chinese": "到现在为止，我一共装了几盘生面圈？", + "answer_chinese": "到现在为止，你一共装了4盘生面圈。", + "evidence": [ + { + "description": "There were two plates of raw doughnuts at the beginning.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:26" + } + }, + { + "description": "Two plates of raw doughnuts were prepared.", + "timestamps": { + "start": "00:05:37", + "end": "00:06:25" + } + } + ], + "question_timestamps": "00:07:17", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You prepared two plates.", + "B": "So far, you have prepared a total of 4 plates of raw doughnuts.", + "C": "You prepared six plates.", + "D": "So far, you have prepared a total of 3 plates of raw doughnuts, stopping after the third plate.", + "E": "So far, you have prepared a total of 5 plates of raw doughnuts, adding one more plate after the fourth." + }, + "answer_label": "B", + "answer_text": "So far, you have prepared a total of 4 plates of raw doughnuts.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_8", + "correct_answer": "So far, you have prepared a total of 4 plates of raw doughnuts.", + "original_options": { + "A": "You prepared two plates.", + "B": "So far, you have prepared a total of 4 plates of raw doughnuts.", + "C": "You prepared six plates.", + "D": "So far, you have prepared a total of 3 plates of raw doughnuts, stopping after the third plate.", + "E": "So far, you have prepared a total of 5 plates of raw doughnuts, adding one more plate after the fourth." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 20 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_7", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates of raw doughnuts have I prepared so far?", + "answer": "So far, you have prepared a total of 4 plates of raw doughnuts.", + "question_chinese": "到现在为止，我一共装了几盘生面圈？", + "answer_chinese": "到现在为止，你一共装了4盘生面圈。", + "evidence": [ + { + "description": "There were two plates of raw doughnuts at the beginning.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:26" + } + }, + { + "description": "Two plates of raw doughnuts were prepared.", + "timestamps": { + "start": "00:05:37", + "end": "00:06:25" + } + } + ], + "question_timestamps": "00:12:01", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "So far, you have prepared a total of 4 plates of raw doughnuts.", + "B": "You prepared three plates.", + "C": "You prepared five plates.", + "D": "So far, you have prepared a total of 2 plates of raw doughnuts, finishing after the second plate.", + "E": "So far, you have prepared a total of 6 plates of raw doughnuts, continuing beyond the fifth plate." + }, + "answer_label": "A", + "answer_text": "So far, you have prepared a total of 4 plates of raw doughnuts.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_7", + "correct_answer": "So far, you have prepared a total of 4 plates of raw doughnuts.", + "original_options": { + "A": "So far, you have prepared a total of 4 plates of raw doughnuts.", + "B": "You prepared three plates.", + "C": "You prepared five plates.", + "D": "So far, you have prepared a total of 2 plates of raw doughnuts, finishing after the second plate.", + "E": "So far, you have prepared a total of 6 plates of raw doughnuts, continuing beyond the fifth plate." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 18 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many batches of doughnuts have I fried so far?", + "answer": "So far, three batches have been fried.", + "question_chinese": "到目前为止，我一共炸了几锅面圈？", + "answer_chinese": "目前一共炸好了三锅。", + "evidence": [ + { + "description": "The first batch of doughnuts is fried.", + "timestamps": { + "start": "00:00:00", + "end": "00:05:02" + } + }, + { + "description": "The second batch of doughnuts is fried.", + "timestamps": { + "start": "00:05:29", + "end": "00:11:27" + } + }, + { + "description": "The third batch of doughnuts is fried.", + "timestamps": { + "start": "00:11:51", + "end": "00:17:47" + } + } + ], + "question_timestamps": "00:17:53", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two batches.", + "B": "Five batches.", + "C": "So far, three batches have been fried.", + "D": "So far, four batches have been fried, with one additional batch finished after the third.", + "E": "So far, six batches have been fried, continuing to fry multiple rounds beyond the fourth batch." + }, + "answer_label": "C", + "answer_text": "So far, three batches have been fried.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_1", + "correct_answer": "So far, three batches have been fried.", + "original_options": { + "A": "Two batches.", + "B": "Five batches.", + "C": "So far, three batches have been fried.", + "D": "So far, four batches have been fried, with one additional batch finished after the third.", + "E": "So far, six batches have been fried, continuing to fry multiple rounds beyond the fourth batch." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many batches of doughnuts have I fried so far?", + "answer": "So far, three batches have been fried.", + "question_chinese": "到目前为止，我一共炸了几锅面圈？", + "answer_chinese": "目前一共炸好了三锅。", + "evidence": [ + { + "description": "The first batch of doughnuts is fried.", + "timestamps": { + "start": "00:00:00", + "end": "00:05:02" + } + }, + { + "description": "The second batch of doughnuts is fried.", + "timestamps": { + "start": "00:05:29", + "end": "00:11:27" + } + }, + { + "description": "The third batch of doughnuts is fried.", + "timestamps": { + "start": "00:11:51", + "end": "00:17:47" + } + } + ], + "question_timestamps": "00:18:57", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One batch.", + "B": "Seven batches.", + "C": "So far, four batches have been fried, ending right after completing the fourth batch.", + "D": "So far, three batches have been fried.", + "E": "So far, five batches have been fried, with another batch cooked after the fourth was done." + }, + "answer_label": "D", + "answer_text": "So far, three batches have been fried.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_2", + "correct_answer": "So far, three batches have been fried.", + "original_options": { + "A": "One batch.", + "B": "Seven batches.", + "C": "So far, four batches have been fried, ending right after completing the fourth batch.", + "D": "So far, three batches have been fried.", + "E": "So far, five batches have been fried, with another batch cooked after the fourth was done." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_4_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates of raw dough rings did I plate in total?", + "answer": "You filled two plates of raw dough rings in total. The raw dough rings are to your front-right.", + "question_chinese": "我一共装了几盘生面圈？", + "answer_chinese": "你一共装了两盘生面圈。生面圈在你的右前方。", + "evidence": [ + { + "description": "There were two plates of raw doughnuts at the beginning.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:26" + } + } + ], + "question_timestamps": "00:05:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You filled two plates of raw dough rings in total. The raw dough rings are to your front-right.", + "B": "Zero plates total.", + "C": "Four plates total.", + "D": "You filled one plate of raw dough rings in total. The raw dough rings are to your front-right.", + "E": "You filled three plates of raw dough rings in total. The raw dough rings are to your front-right." + }, + "answer_label": "A", + "answer_text": "You filled two plates of raw dough rings in total. The raw dough rings are to your front-right.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_4_T1", + "correct_answer": "You filled two plates of raw dough rings in total. The raw dough rings are to your front-right.", + "original_options": { + "A": "You filled two plates of raw dough rings in total. The raw dough rings are to your front-right.", + "B": "Zero plates total.", + "C": "Four plates total.", + "D": "You filled one plate of raw dough rings in total. The raw dough rings are to your front-right.", + "E": "You filled three plates of raw dough rings in total. The raw dough rings are to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 18 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_4_T3", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates of raw dough rings did I plate in total?", + "answer": "You filled one plate of raw dough rings in total. The raw dough rings are in front of you.", + "question_chinese": "我一共装了几盘生面圈？", + "answer_chinese": "你一共装了一盘生面圈。生面圈在你的前面。", + "evidence": [ + { + "description": "There was a plate of raw doughnut at the beginning.", + "timestamps": { + "start": "00:05:12", + "end": "00:05:28" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero plates total.", + "B": "Two plates total.", + "C": "You filled one plate of raw dough rings in total. The raw dough rings are in front of you.", + "D": "You filled three plates of raw dough rings in total. The raw dough rings are in front of you.", + "E": "You filled four plates of raw dough rings in total. The raw dough rings are in front of you." + }, + "answer_label": "C", + "answer_text": "You filled one plate of raw dough rings in total. The raw dough rings are in front of you.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_4_T3", + "correct_answer": "You filled one plate of raw dough rings in total. The raw dough rings are in front of you.", + "original_options": { + "A": "Zero plates total.", + "B": "Two plates total.", + "C": "You filled one plate of raw dough rings in total. The raw dough rings are in front of you.", + "D": "You filled three plates of raw dough rings in total. The raw dough rings are in front of you.", + "E": "You filled four plates of raw dough rings in total. The raw dough rings are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 19, + 19 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_4_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates of raw dough rings did I plate in total?", + "answer": "You filled four plates of raw dough rings in total. The raw dough rings are to your right.", + "question_chinese": "我一共装了几盘生面圈？", + "answer_chinese": "你一共装了四盘生面圈。生面圈在你的右边。", + "evidence": [ + { + "description": "There were two plates of raw doughnuts at the beginning.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:26" + } + }, + { + "description": "Two plates of raw doughnuts were prepared.", + "timestamps": { + "start": "00:05:37", + "end": "00:06:25" + } + } + ], + "question_timestamps": "00:06:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two plates total.", + "B": "You filled four plates of raw dough rings in total. The raw dough rings are to your right.", + "C": "Six plates total.", + "D": "You filled three plates of raw dough rings in total. The raw dough rings are to your right.", + "E": "You filled five plates of raw dough rings in total. The raw dough rings are to your right." + }, + "answer_label": "B", + "answer_text": "You filled four plates of raw dough rings in total. The raw dough rings are to your right.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_4_T2", + "correct_answer": "You filled four plates of raw dough rings in total. The raw dough rings are to your right.", + "original_options": { + "A": "Two plates total.", + "B": "You filled four plates of raw dough rings in total. The raw dough rings are to your right.", + "C": "Six plates total.", + "D": "You filled three plates of raw dough rings in total. The raw dough rings are to your right.", + "E": "You filled five plates of raw dough rings in total. The raw dough rings are to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 18 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_5", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates of raw doughnuts have I prepared in total?", + "answer": "You have prepared a total of 8 plates of raw doughnuts.", + "question_chinese": "我一共装了几盘生面圈？", + "answer_chinese": "你一共装了8盘生面圈。", + "evidence": [ + { + "description": "There were two plates of raw doughnuts at the beginning.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:26" + } + }, + { + "description": "Two plates of raw doughnuts were prepared.", + "timestamps": { + "start": "00:05:37", + "end": "00:06:25" + } + }, + { + "description": "Another two plates of raw doughnuts were prepared.", + "timestamps": { + "start": "00:12:01", + "end": "00:12:41" + } + }, + { + "description": "Two more plates of raw doughnuts were prepared again.", + "timestamps": { + "start": "00:18:09", + "end": "00:18:51" + } + } + ], + "question_timestamps": "00:18:53", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have prepared a total of 8 plates of raw doughnuts.", + "B": "You prepared 7 plates total.", + "C": "You prepared 9 plates total.", + "D": "You have prepared a total of 6 plates of raw doughnuts overall, counting each plate you set out.", + "E": "You have prepared a total of 10 plates of raw doughnuts, including the last plate you finished arranging." + }, + "answer_label": "A", + "answer_text": "You have prepared a total of 8 plates of raw doughnuts.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_5", + "correct_answer": "You have prepared a total of 8 plates of raw doughnuts.", + "original_options": { + "A": "You have prepared a total of 8 plates of raw doughnuts.", + "B": "You prepared 7 plates total.", + "C": "You prepared 9 plates total.", + "D": "You have prepared a total of 6 plates of raw doughnuts overall, counting each plate you set out.", + "E": "You have prepared a total of 10 plates of raw doughnuts, including the last plate you finished arranging." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_6", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates of raw doughnuts have I prepared in total?", + "answer": "You have prepared a total of 8 plates of raw doughnuts.", + "question_chinese": "我一共装了几盘生面圈？", + "answer_chinese": "你一共装了8盘生面圈。", + "evidence": [ + { + "description": "There were two plates of raw doughnuts at the beginning.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:26" + } + }, + { + "description": "Two plates of raw doughnuts were prepared.", + "timestamps": { + "start": "00:05:37", + "end": "00:06:25" + } + }, + { + "description": "Another two plates of raw doughnuts were prepared.", + "timestamps": { + "start": "00:12:01", + "end": "00:12:41" + } + }, + { + "description": "Two more plates of raw doughnuts were prepared again.", + "timestamps": { + "start": "00:18:09", + "end": "00:18:51" + } + } + ], + "question_timestamps": "00:21:47", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have prepared a total of 8 plates of raw doughnuts.", + "B": "You prepared 7 plates total.", + "C": "You prepared 9 plates total.", + "D": "You have prepared a total of 6 plates of raw doughnuts overall, counting each plate you set out.", + "E": "You have prepared a total of 10 plates of raw doughnuts, including the last plate you finished arranging." + }, + "answer_label": "A", + "answer_text": "You have prepared a total of 8 plates of raw doughnuts.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_6", + "correct_answer": "You have prepared a total of 8 plates of raw doughnuts.", + "original_options": { + "A": "You have prepared a total of 8 plates of raw doughnuts.", + "B": "You prepared 7 plates total.", + "C": "You prepared 9 plates total.", + "D": "You have prepared a total of 6 plates of raw doughnuts overall, counting each plate you set out.", + "E": "You have prepared a total of 10 plates of raw doughnuts, including the last plate you finished arranging." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_8_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was plating the raw dough rings, how many different kinds of plates did I use?", + "answer": "When you were plating the raw dough rings, you used one kind of plate. The plate is in front of you.", + "question_chinese": "我在装生面圈的时候用了几种不同的碟子？", + "answer_chinese": "你在装生面圈的时候用了一种碟子。碟子在你的前面。", + "evidence": [ + { + "description": "One type of plate was used to hold the raw doughnuts.", + "timestamps": { + "start": "00:18:10", + "end": "00:18:14" + } + } + ], + "question_timestamps": "00:18:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two kinds of plates.", + "B": "Three kinds of plates.", + "C": "When you were plating the raw dough rings, you used two kinds of plates, both placed in front of you.", + "D": "When you were plating the raw dough rings, you used three kinds of plates, with the plate set in front of you.", + "E": "When you were plating the raw dough rings, you used one kind of plate. The plate is in front of you." + }, + "answer_label": "E", + "answer_text": "When you were plating the raw dough rings, you used one kind of plate. The plate is in front of you.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_8_T1", + "correct_answer": "When you were plating the raw dough rings, you used one kind of plate. The plate is in front of you.", + "original_options": { + "A": "Two kinds of plates.", + "B": "Three kinds of plates.", + "C": "When you were plating the raw dough rings, you used two kinds of plates, both placed in front of you.", + "D": "When you were plating the raw dough rings, you used three kinds of plates, with the plate set in front of you.", + "E": "When you were plating the raw dough rings, you used one kind of plate. The plate is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 22 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_8_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was plating the raw dough rings, how many different kinds of plates did I use?", + "answer": "When you were packing the raw noodle rings, you used two different kinds of plates. The plates were to your right.", + "question_chinese": "我在装生面圈的时候用了几种不同的碟子？", + "answer_chinese": "你在装生面圈的时候用了两种碟子。碟子在你的右边。", + "evidence": [ + { + "description": "One type of plate was used to hold the raw doughnuts, and another smaller plate was used to pick up the raw doughnuts from the blanket.", + "timestamps": { + "start": "00:18:10", + "end": "00:18:51" + } + } + ], + "question_timestamps": "00:18:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "When you were packing the raw noodle rings, you used two different kinds of plates. The plates were to your right.", + "B": "One kind of plate.", + "C": "Three kinds of plates.", + "D": "When you were packing the raw noodle rings, you used four different kinds of plates, and they were to your right.", + "E": "When you were packing the raw noodle rings, you used one kind of plate, with the plates positioned to your right." + }, + "answer_label": "A", + "answer_text": "When you were packing the raw noodle rings, you used two different kinds of plates. The plates were to your right.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_8_T2", + "correct_answer": "When you were packing the raw noodle rings, you used two different kinds of plates. The plates were to your right.", + "original_options": { + "A": "When you were packing the raw noodle rings, you used two different kinds of plates. The plates were to your right.", + "B": "One kind of plate.", + "C": "Three kinds of plates.", + "D": "When you were packing the raw noodle rings, you used four different kinds of plates, and they were to your right.", + "E": "When you were packing the raw noodle rings, you used one kind of plate, with the plates positioned to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 21, + 21 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_16", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different kinds of plates did I use when preparing the raw doughnuts?", + "answer": "Two.", + "question_chinese": "我在装生面圈的时候用了几种不同的碟子？", + "answer_chinese": "两种。", + "evidence": [ + { + "description": "One type of plate was used to hold the raw doughnuts, and another smaller plate was used to pick up the raw doughnuts from the blanket.", + "timestamps": { + "start": "00:18:10", + "end": "00:18:51" + } + } + ], + "question_timestamps": "00:19:53", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You used four different kinds of plates when preparing the raw doughnuts, switching between them as you worked.", + "D": "You used five different kinds of plates when preparing the raw doughnuts, alternating plate types during prep.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_16", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You used four different kinds of plates when preparing the raw doughnuts, switching between them as you worked.", + "D": "You used five different kinds of plates when preparing the raw doughnuts, alternating plate types during prep.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 17 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_13", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the gas cooker now?", + "answer": "No.", + "question_chinese": "我现在可以够到煤气炉吗？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The gas cooker is about two meters away on your left.", + "timestamps": { + "start": "00:17:55", + "end": "00:18:14" + } + } + ], + "question_timestamps": "00:18:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_13", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_14", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the gas cooker now?", + "answer": "No.", + "question_chinese": "我现在可以够到煤气炉吗？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The gas cooker is about two meters away on your left.", + "timestamps": { + "start": "00:17:55", + "end": "00:18:14" + } + } + ], + "question_timestamps": "00:18:47", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_14", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_11", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the stove from me?", + "answer": "The stove is on your left.", + "question_chinese": "灶台在我的哪个方位？", + "answer_chinese": "灶台在你的左边。", + "evidence": [ + { + "description": "The stove is on the left side of the blanket.", + "timestamps": { + "start": "00:11:57", + "end": "00:12:01" + } + } + ], + "question_timestamps": "00:12:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The stove is on your right.", + "B": "The stove is in front of you.", + "C": "The stove is behind you, so you’d need to turn around to face it.", + "D": "The stove is to your right side, positioned off to the side rather than directly ahead.", + "E": "The stove is on your left." + }, + "answer_label": "E", + "answer_text": "The stove is on your left.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_11", + "correct_answer": "The stove is on your left.", + "original_options": { + "A": "The stove is on your right.", + "B": "The stove is in front of you.", + "C": "The stove is behind you, so you’d need to turn around to face it.", + "D": "The stove is to your right side, positioned off to the side rather than directly ahead.", + "E": "The stove is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 16 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_12", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the stove from me?", + "answer": "The stove is on your left.", + "question_chinese": "灶台在我的哪个方位？", + "answer_chinese": "灶台在你的左边。", + "evidence": [ + { + "description": "The stove is on the left side of the blanket.", + "timestamps": { + "start": "00:11:57", + "end": "00:12:01" + } + } + ], + "question_timestamps": "00:12:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The stove is in front of you.", + "B": "The stove is on your left.", + "C": "The stove is on your right.", + "D": "The stove is directly behind you, not to your left side at all.", + "E": "The stove is slightly ahead of you, more toward your right side." + }, + "answer_label": "B", + "answer_text": "The stove is on your left.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_12", + "correct_answer": "The stove is on your left.", + "original_options": { + "A": "The stove is in front of you.", + "B": "The stove is on your left.", + "C": "The stove is on your right.", + "D": "The stove is directly behind you, not to your left side at all.", + "E": "The stove is slightly ahead of you, more toward your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 12 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_2_T3", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many batches of dough rings have I fried in total?", + "answer": "So far, you have fried one batch of dough rings in total. The dough rings are to your back-left.", + "question_chinese": "现在我一共炸了几锅面圈？", + "answer_chinese": "现在你一共炸了一锅面圈。面圈在你的左后方。", + "evidence": [ + { + "description": "The first batch of doughnuts is fried.", + "timestamps": { + "start": "00:00:00", + "end": "00:05:02" + } + } + ], + "question_timestamps": "00:05:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, you have fried one batch of dough rings in total. The dough rings are to your back-left.", + "B": "Zero batches so far.", + "C": "Two batches so far.", + "D": "So far, you have fried three batches of dough rings in total, and the dough rings are to your back-left.", + "E": "So far, you have fried four batches of dough rings in total, with the dough rings sitting to your back-left." + }, + "answer_label": "A", + "answer_text": "So far, you have fried one batch of dough rings in total. The dough rings are to your back-left.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_2_T3", + "correct_answer": "So far, you have fried one batch of dough rings in total. The dough rings are to your back-left.", + "original_options": { + "A": "So far, you have fried one batch of dough rings in total. The dough rings are to your back-left.", + "B": "Zero batches so far.", + "C": "Two batches so far.", + "D": "So far, you have fried three batches of dough rings in total, and the dough rings are to your back-left.", + "E": "So far, you have fried four batches of dough rings in total, with the dough rings sitting to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 20 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_3", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many batches of doughnuts have I fried now?", + "answer": "Two batches have been fried now.", + "question_chinese": "现在我一共炸了几锅面圈？", + "answer_chinese": "现在一共炸了两锅。", + "evidence": [ + { + "description": "The first batch of doughnuts is fried.", + "timestamps": { + "start": "00:00:00", + "end": "00:05:02" + } + }, + { + "description": "The second batch of doughnuts is fried.", + "timestamps": { + "start": "00:05:29", + "end": "00:11:27" + } + } + ], + "question_timestamps": "00:11:57", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One batch has been fried.", + "B": "Three batches have been fried.", + "C": "Two batches have been fried now.", + "D": "Four batches have been fried now, based on what you’ve cooked so far.", + "E": "Five batches have been fried now, meaning you completed several more batches already." + }, + "answer_label": "C", + "answer_text": "Two batches have been fried now.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_3", + "correct_answer": "Two batches have been fried now.", + "original_options": { + "A": "One batch has been fried.", + "B": "Three batches have been fried.", + "C": "Two batches have been fried now.", + "D": "Four batches have been fried now, based on what you’ve cooked so far.", + "E": "Five batches have been fried now, meaning you completed several more batches already." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9", + "video_length": "00:22:14", + "qid": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_4", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many batches of doughnuts have I fried now?", + "answer": "Two batches have been fried now.", + "question_chinese": "现在我一共炸了几锅面圈？", + "answer_chinese": "现在一共炸了两锅。", + "evidence": [ + { + "description": "The first batch of doughnuts is fried.", + "timestamps": { + "start": "00:00:00", + "end": "00:05:02" + } + }, + { + "description": "The second batch of doughnuts is fried.", + "timestamps": { + "start": "00:05:29", + "end": "00:11:27" + } + } + ], + "question_timestamps": "00:12:47", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two batches have been fried now.", + "B": "One batch has been fried.", + "C": "Three batches have been fried.", + "D": "Four batches have been fried now, counting the batches you finished up to this moment.", + "E": "Five batches have been fried now, indicating you’ve fried three additional batches already." + }, + "answer_label": "A", + "answer_text": "Two batches have been fried now.", + "q_id": "3ec3eab7-842d-409d-8866-42ddcbd24cd9_4", + "correct_answer": "Two batches have been fried now.", + "original_options": { + "A": "Two batches have been fried now.", + "B": "One batch has been fried.", + "C": "Three batches have been fried.", + "D": "Four batches have been fried now, counting the batches you finished up to this moment.", + "E": "Five batches have been fried now, indicating you’ve fried three additional batches already." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + } +] diff --git a/QAs/3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d-mcq.json b/QAs/3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..00146aa3001f6e3433ad92639efb6fbf867b501a --- /dev/null +++ b/QAs/3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d-mcq.json @@ -0,0 +1,915 @@ +[ + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_5", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the cardboard box?", + "answer": "You just put the cardboard box under the table.", + "question_chinese": "我刚刚把纸箱放在什么地方？", + "answer_chinese": "你刚刚把纸箱放在桌下了。", + "evidence": [ + { + "description": "You just put the cardboard box under the table.", + "timestamps": { + "start": "00:08:40", + "end": "00:08:44" + } + } + ], + "question_timestamps": "00:08:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just put the cardboard box under the table.", + "B": "You put it on the table.", + "C": "You put it beside the table.", + "D": "You just put the cardboard box on the shelf instead of under the table.", + "E": "You placed the cardboard box near the chair, rather than underneath the table." + }, + "answer_label": "A", + "answer_text": "You just put the cardboard box under the table.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_5", + "correct_answer": "You just put the cardboard box under the table.", + "original_options": { + "A": "You just put the cardboard box under the table.", + "B": "You put it on the table.", + "C": "You put it beside the table.", + "D": "You just put the cardboard box on the shelf instead of under the table.", + "E": "You placed the cardboard box near the chair, rather than underneath the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_9", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many petri dishes are left after I just got them?", + "answer": "There are still five left now.", + "question_chinese": "我拿了培养皿后现在还剩下几个？", + "answer_chinese": "现在还剩下五个。", + "evidence": [ + { + "description": "There are still five left now.", + "timestamps": { + "start": "00:16:00", + "end": "00:16:17" + } + } + ], + "question_timestamps": "00:16:19", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are still four left now.", + "B": "There are still six left now.", + "C": "There are still seven left now after you took some, leaving more than five behind.", + "D": "There are still five left now.", + "E": "There are still three left now because you removed additional petri dishes right after getting them." + }, + "answer_label": "D", + "answer_text": "There are still five left now.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_9", + "correct_answer": "There are still five left now.", + "original_options": { + "A": "There are still four left now.", + "B": "There are still six left now.", + "C": "There are still seven left now after you took some, leaving more than five behind.", + "D": "There are still five left now.", + "E": "There are still three left now because you removed additional petri dishes right after getting them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_10", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many petri dishes are left after I just got them?", + "answer": "There are still zero left.", + "question_chinese": "我拿了培养皿后现在还剩下几个？", + "answer_chinese": "还剩下0个。", + "evidence": [ + { + "description": "There are still zero left.", + "timestamps": { + "start": "00:16:26", + "end": "00:16:39" + } + } + ], + "question_timestamps": "00:21:14", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is still one left.", + "B": "There are still two left.", + "C": "There are still three left after you got them, so you did not run out completely.", + "D": "There are still four left now because you didn’t take every petri dish you just got.", + "E": "There are still zero left." + }, + "answer_label": "E", + "answer_text": "There are still zero left.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_10", + "correct_answer": "There are still zero left.", + "original_options": { + "A": "There is still one left.", + "B": "There are still two left.", + "C": "There are still three left after you got them, so you did not run out completely.", + "D": "There are still four left now because you didn’t take every petri dish you just got.", + "E": "There are still zero left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_1_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Right now, can I directly touch the trash can to my right or the one in front of me?", + "answer": "Yes, you can reach the trash can to your right.", + "question_chinese": "我现在可以直接触碰到右边或面前的垃圾桶吗？", + "answer_chinese": "是，你可以碰到右边的垃圾桶", + "evidence": [ + { + "description": "You can make direct contact now.", + "timestamps": { + "start": "00:02:27", + "end": "00:02:28" + } + } + ], + "question_timestamps": "00:03:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t directly reach the trash can to your right.", + "B": "Yes, you can reach the trash can to your right." + }, + "answer_label": "B", + "answer_text": "Yes, you can reach the trash can to your right.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_1_T2", + "correct_answer": "Yes, you can reach the trash can to your right.", + "original_options": { + "A": "No, you can’t directly reach the trash can to your right.", + "B": "Yes, you can reach the trash can to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_1_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Right now, can I directly touch the trash can to my right or the one in front of me?", + "answer": "No, you can’t directly reach the trash can in front of you.", + "question_chinese": "我现在可以直接触碰到右边或面前的垃圾桶吗？", + "answer_chinese": "不，你不可以直接碰到面前的垃圾桶", + "evidence": [ + { + "description": "You can't make direct contact now.", + "timestamps": { + "start": "00:04:36", + "end": "00:04:39" + } + } + ], + "question_timestamps": "00:04:41", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t directly reach the trash can in front of you.", + "B": "Yes, you can reach the trash can in front of you." + }, + "answer_label": "A", + "answer_text": "No, you can’t directly reach the trash can in front of you.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_1_T1", + "correct_answer": "No, you can’t directly reach the trash can in front of you.", + "original_options": { + "A": "No, you can’t directly reach the trash can in front of you.", + "B": "Yes, you can reach the trash can in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the trash can directly now?", + "answer": "You can't make direct contact now.", + "question_chinese": "我现在可以直接触碰到垃圾桶吗？", + "answer_chinese": "你现在无法直接触碰到。", + "evidence": [ + { + "description": "You can't make direct contact now.", + "timestamps": { + "start": "00:04:36", + "end": "00:04:39" + } + } + ], + "question_timestamps": "00:04:40", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't make direct contact now.", + "B": "You can make direct contact now." + }, + "answer_label": "A", + "answer_text": "You can't make direct contact now.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_1", + "correct_answer": "You can't make direct contact now.", + "original_options": { + "A": "You can't make direct contact now.", + "B": "You can make direct contact now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the trash can directly now?", + "answer": "You can't make direct contact now.", + "question_chinese": "我现在可以直接触碰到垃圾桶吗？", + "answer_chinese": "你现在无法直接触碰到。", + "evidence": [ + { + "description": "You can't make direct contact now.", + "timestamps": { + "start": "00:13:23", + "end": "00:13:25" + } + } + ], + "question_timestamps": "00:13:27", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can make direct contact now.", + "B": "You can't make direct contact now." + }, + "answer_label": "B", + "answer_text": "You can't make direct contact now.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_2", + "correct_answer": "You can't make direct contact now.", + "original_options": { + "A": "You can make direct contact now.", + "B": "You can't make direct contact now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_3", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the faucet directly now?", + "answer": "You can now make direct contact.", + "question_chinese": "我现在可以直接触碰到水龙头吗？", + "answer_chinese": "你现在可以直接触碰到。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:07:08", + "end": "00:07:11" + } + } + ], + "question_timestamps": "00:07:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now make direct contact.", + "B": "You cannot touch the faucet directly." + }, + "answer_label": "A", + "answer_text": "You can now make direct contact.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_3", + "correct_answer": "You can now make direct contact.", + "original_options": { + "A": "You can now make direct contact.", + "B": "You cannot touch the faucet directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_4", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the faucet directly now?", + "answer": "You can now make direct contact.", + "question_chinese": "我现在可以直接触碰到水龙头吗？", + "answer_chinese": "你现在可以直接触碰到。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:07:08", + "end": "00:07:11" + } + }, + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:23:12", + "end": "00:23:14" + } + } + ], + "question_timestamps": "00:23:15", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can now make direct contact.", + "B": "You cannot touch the faucet directly now." + }, + "answer_label": "A", + "answer_text": "You can now make direct contact.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_4", + "correct_answer": "You can now make direct contact.", + "original_options": { + "A": "You can now make direct contact.", + "B": "You cannot touch the faucet directly now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_2_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Right now, can I directly touch the faucet in front of me?", + "answer": "No, you can’t directly reach the faucet in front of you.", + "question_chinese": "我现在可以直接触碰到面前的水龙头吗？", + "answer_chinese": "不，你不可以直接碰到面前的水龙头", + "evidence": [ + { + "description": "You can't now make direct contact.", + "timestamps": { + "start": "00:08:51", + "end": "00:08:52" + } + } + ], + "question_timestamps": "00:08:54", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can directly reach the faucet in front of you.", + "B": "No, you can’t directly reach the faucet in front of you." + }, + "answer_label": "B", + "answer_text": "No, you can’t directly reach the faucet in front of you.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_2_T1", + "correct_answer": "No, you can’t directly reach the faucet in front of you.", + "original_options": { + "A": "Yes, you can directly reach the faucet in front of you.", + "B": "No, you can’t directly reach the faucet in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_2_T2", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Right now, can I directly touch the faucet in front of me?", + "answer": "Yes, you can directly reach the faucet in front of you.", + "question_chinese": "我现在可以直接触碰到面前的水龙头吗？", + "answer_chinese": "是，你可以直接碰到面前的水龙头", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:07:08", + "end": "00:07:11" + } + }, + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:23:12", + "end": "00:23:14" + } + } + ], + "question_timestamps": "00:23:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can directly reach the faucet in front of you.", + "B": "No, you can’t directly reach the faucet in front of you." + }, + "answer_label": "A", + "answer_text": "Yes, you can directly reach the faucet in front of you.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_2_T2", + "correct_answer": "Yes, you can directly reach the faucet in front of you.", + "original_options": { + "A": "Yes, you can directly reach the faucet in front of you.", + "B": "No, you can’t directly reach the faucet in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_11", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the trolley now?", + "answer": "You can touch it now.", + "question_chinese": "我现在可以触碰到手推车吗？", + "answer_chinese": "你现在可以触碰到。", + "evidence": [ + { + "description": "You can touch it now.", + "timestamps": { + "start": "00:25:27", + "end": "00:25:29" + } + } + ], + "question_timestamps": "00:25:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot touch it now.", + "B": "You can touch it now." + }, + "answer_label": "B", + "answer_text": "You can touch it now.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_11", + "correct_answer": "You can touch it now.", + "original_options": { + "A": "You cannot touch it now.", + "B": "You can touch it now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_6_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the cart right now?", + "answer": "No.", + "question_chinese": "我现在可以触碰到手推车吗？", + "answer_chinese": "不", + "evidence": [ + { + "description": "You can't touch it now.", + "timestamps": { + "start": "00:25:27", + "end": "00:25:29" + } + } + ], + "question_timestamps": "00:25:31", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_6_T1", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_12", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the trolley now?", + "answer": "You can touch it now.", + "question_chinese": "我现在可以触碰到手推车吗？", + "answer_chinese": "你现在可以触碰到。", + "evidence": [ + { + "description": "You can touch it now.", + "timestamps": { + "start": "00:29:40", + "end": "00:29:45" + } + } + ], + "question_timestamps": "00:29:46", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch it now.", + "B": "You cannot touch it now." + }, + "answer_label": "A", + "answer_text": "You can touch it now.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_12", + "correct_answer": "You can touch it now.", + "original_options": { + "A": "You can touch it now.", + "B": "You cannot touch it now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_6_T2", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the cart right now?", + "answer": "Yes.", + "question_chinese": "我现在可以触碰到手推车吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "You can touch it now.", + "timestamps": { + "start": "00:29:40", + "end": "00:29:45" + } + } + ], + "question_timestamps": "00:29:47", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_6_T2", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_13", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What position does the computer hold relative to me now?", + "answer": "The computer is in front of you.", + "question_chinese": "电脑现在相对我是什么位置？", + "answer_chinese": "电脑在你的前方。", + "evidence": [ + { + "description": "The computer is in front of you.", + "timestamps": { + "start": "00:31:10", + "end": "00:31:15" + } + } + ], + "question_timestamps": "00:31:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The computer is behind you.", + "B": "The computer is to your left.", + "C": "The computer is positioned off to your right side, not directly in front of you.", + "D": "The computer is in front of you.", + "E": "The computer sits behind you relative to your viewpoint, rather than facing you from the front." + }, + "answer_label": "D", + "answer_text": "The computer is in front of you.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_13", + "correct_answer": "The computer is in front of you.", + "original_options": { + "A": "The computer is behind you.", + "B": "The computer is to your left.", + "C": "The computer is positioned off to your right side, not directly in front of you.", + "D": "The computer is in front of you.", + "E": "The computer sits behind you relative to your viewpoint, rather than facing you from the front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_14", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What position does the computer hold relative to me now?", + "answer": "The computer is behind you.", + "question_chinese": "电脑现在相对我是什么位置？", + "answer_chinese": "电脑在你的后方。", + "evidence": [ + { + "description": "The computer is behind you.", + "timestamps": { + "start": "00:33:04", + "end": "00:33:21" + } + } + ], + "question_timestamps": "00:33:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The computer is in front of you.", + "B": "The computer is behind you.", + "C": "The computer is to your left.", + "D": "The computer is on your right side, slightly off to the side as you look forward.", + "E": "The computer is above you, positioned higher up rather than behind your current position." + }, + "answer_label": "B", + "answer_text": "The computer is behind you.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_14", + "correct_answer": "The computer is behind you.", + "original_options": { + "A": "The computer is in front of you.", + "B": "The computer is behind you.", + "C": "The computer is to your left.", + "D": "The computer is on your right side, slightly off to the side as you look forward.", + "E": "The computer is above you, positioned higher up rather than behind your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 14 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_7", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the position of the whiteboard relative to me now?", + "answer": "The whiteboard is now on your left.", + "question_chinese": "白板现在相对我是什么位置？", + "answer_chinese": "白板现在在你的左侧。", + "evidence": [ + { + "description": "The whiteboard is now on your left.", + "timestamps": { + "start": "00:10:16", + "end": "00:10:18" + } + } + ], + "question_timestamps": "00:10:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The whiteboard is in front of you.", + "B": "The whiteboard is on your right.", + "C": "The whiteboard is now on your left.", + "D": "The whiteboard is directly behind you now, not on your left side anymore.", + "E": "The whiteboard is positioned above you at the moment, rather than being off to your left." + }, + "answer_label": "C", + "answer_text": "The whiteboard is now on your left.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_7", + "correct_answer": "The whiteboard is now on your left.", + "original_options": { + "A": "The whiteboard is in front of you.", + "B": "The whiteboard is on your right.", + "C": "The whiteboard is now on your left.", + "D": "The whiteboard is directly behind you now, not on your left side anymore.", + "E": "The whiteboard is positioned above you at the moment, rather than being off to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 16 + ] + }, + { + "video_uid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d", + "video_length": "00:40:03", + "qid": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_8", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the position of the whiteboard relative to me now?", + "answer": "The whiteboard is on your right.", + "question_chinese": "白板现在相对我是什么位置？", + "answer_chinese": "白板在你的右侧。", + "evidence": [ + { + "description": "The whiteboard is on your right.", + "timestamps": { + "start": "00:23:34", + "end": "00:23:36" + } + } + ], + "question_timestamps": "00:23:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The whiteboard is in front of you.", + "B": "The whiteboard is on your right.", + "C": "The whiteboard is on your left.", + "D": "The whiteboard is behind you now, not to your right where it would be visible.", + "E": "The whiteboard is across from you in front, rather than positioned on your right side." + }, + "answer_label": "B", + "answer_text": "The whiteboard is on your right.", + "q_id": "3f8dbb4f-ad5e-4d54-bf95-3ddbe968a85d_8", + "correct_answer": "The whiteboard is on your right.", + "original_options": { + "A": "The whiteboard is in front of you.", + "B": "The whiteboard is on your right.", + "C": "The whiteboard is on your left.", + "D": "The whiteboard is behind you now, not to your right where it would be visible.", + "E": "The whiteboard is across from you in front, rather than positioned on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 15 + ] + } +] diff --git a/QAs/3fffb452-1c17-43fa-b473-c9e19ac21433-mcq.json b/QAs/3fffb452-1c17-43fa-b473-c9e19ac21433-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ea2a4fb0a20ed55ed9ec8457fac04a95b31b7056 --- /dev/null +++ b/QAs/3fffb452-1c17-43fa-b473-c9e19ac21433-mcq.json @@ -0,0 +1,1424 @@ +[ + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_6_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many beige long strips of cloth have I collected in total?", + "answer": "You collected two cloths.", + "question_chinese": "到现在为止我一共收走了几块米色的长条布？", + "answer_chinese": "你收走了两条布。", + "evidence": [ + { + "description": "You took away the first piece of cloth.", + "timestamps": { + "start": "00:18:52", + "end": "00:18:59" + } + }, + { + "description": "You took away the second piece of cloth.", + "timestamps": { + "start": "00:19:05", + "end": "00:19:12" + } + } + ], + "question_timestamps": "00:19:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You collected one cloth.", + "B": "You collected three cloths.", + "C": "You collected zero cloths in total, meaning you never picked up any beige long strips.", + "D": "You collected four cloths altogether, adding up to four beige long strips by this point.", + "E": "You collected two cloths." + }, + "answer_label": "E", + "answer_text": "You collected two cloths.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_6_T1", + "correct_answer": "You collected two cloths.", + "original_options": { + "A": "You collected one cloth.", + "B": "You collected three cloths.", + "C": "You collected zero cloths in total, meaning you never picked up any beige long strips.", + "D": "You collected four cloths altogether, adding up to four beige long strips by this point.", + "E": "You collected two cloths." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_10", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of beige strip cloth have I collected so far?", + "answer": "You took away four pieces of cloth.", + "question_chinese": "到现在为止我一共收走了几块米色的长条布？", + "answer_chinese": "你收走了四条布。", + "evidence": [ + { + "description": "You took away the first piece of cloth.", + "timestamps": { + "start": "00:18:52", + "end": "00:18:59" + } + }, + { + "description": "You took away the second piece of cloth.", + "timestamps": { + "start": "00:19:05", + "end": "00:19:12" + } + }, + { + "description": "You took away the third piece of cloth.", + "timestamps": { + "start": "00:19:24", + "end": "00:19:29" + } + }, + { + "description": "You took away the fourth piece of cloth.", + "timestamps": { + "start": "00:19:33", + "end": "00:19:38" + } + } + ], + "question_timestamps": "00:19:38", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You took away four pieces of cloth.", + "B": "You took away three pieces.", + "C": "You took away five pieces.", + "D": "You took away two pieces of cloth so far, for a total of two beige strip pieces collected.", + "E": "You took away six pieces of cloth in total up to now, meaning you removed six separate strips." + }, + "answer_label": "A", + "answer_text": "You took away four pieces of cloth.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_10", + "correct_answer": "You took away four pieces of cloth.", + "original_options": { + "A": "You took away four pieces of cloth.", + "B": "You took away three pieces.", + "C": "You took away five pieces.", + "D": "You took away two pieces of cloth so far, for a total of two beige strip pieces collected.", + "E": "You took away six pieces of cloth in total up to now, meaning you removed six separate strips." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_6_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many beige long strips of cloth have I collected in total?", + "answer": "You collected four cloths.", + "question_chinese": "到现在为止我一共收走了几块米色的长条布？", + "answer_chinese": "你收走了四条布。", + "evidence": [ + { + "description": "You took away the first piece of cloth.", + "timestamps": { + "start": "00:18:52", + "end": "00:18:59" + } + }, + { + "description": "You took away the second piece of cloth.", + "timestamps": { + "start": "00:19:05", + "end": "00:19:12" + } + }, + { + "description": "You took away the third piece of cloth.", + "timestamps": { + "start": "00:19:24", + "end": "00:19:29" + } + }, + { + "description": "You took away the fourth piece of cloth.", + "timestamps": { + "start": "00:19:33", + "end": "00:19:38" + } + } + ], + "question_timestamps": "00:35:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You collected four cloths.", + "B": "You collected three cloths.", + "C": "You collected five cloths.", + "D": "You collected two cloths in total so far, which would mean only two beige long strips collected.", + "E": "You collected six cloths altogether up to now, totaling six beige long strips of cloth." + }, + "answer_label": "A", + "answer_text": "You collected four cloths.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_6_T2", + "correct_answer": "You collected four cloths.", + "original_options": { + "A": "You collected four cloths.", + "B": "You collected three cloths.", + "C": "You collected five cloths.", + "D": "You collected two cloths in total so far, which would mean only two beige long strips collected.", + "E": "You collected six cloths altogether up to now, totaling six beige long strips of cloth." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 15 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_14", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the timer so far?", + "answer": "You used the timer twice.", + "question_chinese": "到现在为止，我用了几次计时器？", + "answer_chinese": "你用了两次计时器。", + "evidence": [ + { + "description": "You are using a timer for the first time.", + "timestamps": { + "start": "00:00:53", + "end": "00:01:03" + } + }, + { + "description": "You used the timer for the second time.", + "timestamps": { + "start": "00:17:57", + "end": "00:18:00" + } + }, + { + "description": "You are using the timer for the third time.", + "timestamps": { + "start": "00:36:09", + "end": "00:36:11" + } + }, + { + "description": "This is the fourth time you've used the timer.", + "timestamps": { + "start": "00:39:55", + "end": "00:40:01" + } + }, + { + "description": "You are using the timer for the fifth time.", + "timestamps": { + "start": "01:00:37", + "end": "01:00:41" + } + } + ], + "question_timestamps": "01:00:42", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used the timer once.", + "B": "You used the timer twice.", + "C": "You used the timer three times.", + "D": "You used the timer four times so far, meaning you started or checked it on four separate occasions.", + "E": "You used the timer five times up to now, counting five distinct timer uses during the activity." + }, + "answer_label": "B", + "answer_text": "You used the timer twice.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_14", + "correct_answer": "You used the timer twice.", + "original_options": { + "A": "You used the timer once.", + "B": "You used the timer twice.", + "C": "You used the timer three times.", + "D": "You used the timer four times so far, meaning you started or checked it on four separate occasions.", + "E": "You used the timer five times up to now, counting five distinct timer uses during the activity." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 17 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_15", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the timer so far?", + "answer": "You used the timer five times.", + "question_chinese": "到现在为止，我用了几次计时器？", + "answer_chinese": "你用了五次计时器。", + "evidence": [ + { + "description": "You are using a timer for the first time.", + "timestamps": { + "start": "00:00:53", + "end": "00:01:03" + } + }, + { + "description": "You used the timer for the second time.", + "timestamps": { + "start": "00:17:57", + "end": "00:18:00" + } + }, + { + "description": "You are using the timer for the third time.", + "timestamps": { + "start": "00:36:09", + "end": "00:36:11" + } + }, + { + "description": "This is the fourth time you've used the timer.", + "timestamps": { + "start": "00:39:55", + "end": "00:40:01" + } + }, + { + "description": "You are using the timer for the fifth time.", + "timestamps": { + "start": "01:00:37", + "end": "01:00:41" + } + } + ], + "question_timestamps": "01:00:45", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used the timer four times.", + "B": "You used the timer six times.", + "C": "You used the timer five times.", + "D": "You used the timer eight times so far, meaning you interacted with it on eight separate occasions.", + "E": "You used the timer nine times up to now, counting nine distinct uses during the session." + }, + "answer_label": "C", + "answer_text": "You used the timer five times.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_15", + "correct_answer": "You used the timer five times.", + "original_options": { + "A": "You used the timer four times.", + "B": "You used the timer six times.", + "C": "You used the timer five times.", + "D": "You used the timer eight times so far, meaning you interacted with it on eight separate occasions.", + "E": "You used the timer nine times up to now, counting nine distinct uses during the session." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_12", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the white long bread rack so far?", + "answer": "You used the white long bread rack three times.", + "question_chinese": "到目前为止，我用了几次白色的长面包架？", + "answer_chinese": "你用了三次白色长面包架。", + "evidence": [ + { + "description": "You are using the long bread rack for the first time.", + "timestamps": { + "start": "00:11:57", + "end": "00:12:10" + } + }, + { + "description": "You are using the long bread rack for the second time.", + "timestamps": { + "start": "00:16:58", + "end": "00:17:08" + } + }, + { + "description": "You are using the long bread rack for the third time.", + "timestamps": { + "start": "00:24:34", + "end": "00:24:46" + } + }, + { + "description": "You are using the long bread rack for the fourth time.", + "timestamps": { + "start": "00:28:12", + "end": "00:28:23" + } + }, + { + "description": "You are using the long bread rack for the fifth time.", + "timestamps": { + "start": "00:32:16", + "end": "00:32:28" + } + }, + { + "description": "You are using the long bread rack for the sixth time.", + "timestamps": { + "start": "00:35:56", + "end": "00:36:06" + } + } + ], + "question_timestamps": "00:36:06", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used it twice.", + "B": "You used it four times.", + "C": "You used the white long bread rack only once so far, not the three times mentioned.", + "D": "You used the white long bread rack five times in total up to now, counting each separate use.", + "E": "You used the white long bread rack three times." + }, + "answer_label": "E", + "answer_text": "You used the white long bread rack three times.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_12", + "correct_answer": "You used the white long bread rack three times.", + "original_options": { + "A": "You used it twice.", + "B": "You used it four times.", + "C": "You used the white long bread rack only once so far, not the three times mentioned.", + "D": "You used the white long bread rack five times in total up to now, counting each separate use.", + "E": "You used the white long bread rack three times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 18 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_13", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the white long bread rack so far?", + "answer": "You used the white long bread rack six times.", + "question_chinese": "到目前为止，我用了几次白色的长面包架？", + "answer_chinese": "你用了六次白色长面包架。", + "evidence": [ + { + "description": "You are using the long bread rack for the first time.", + "timestamps": { + "start": "00:11:57", + "end": "00:12:10" + } + }, + { + "description": "You are using the long bread rack for the second time.", + "timestamps": { + "start": "00:16:58", + "end": "00:17:08" + } + }, + { + "description": "You are using the long bread rack for the third time.", + "timestamps": { + "start": "00:24:34", + "end": "00:24:46" + } + }, + { + "description": "You are using the long bread rack for the fourth time.", + "timestamps": { + "start": "00:28:12", + "end": "00:28:23" + } + }, + { + "description": "You are using the long bread rack for the fifth time.", + "timestamps": { + "start": "00:32:16", + "end": "00:32:28" + } + }, + { + "description": "You are using the long bread rack for the sixth time.", + "timestamps": { + "start": "00:35:56", + "end": "00:36:06" + } + } + ], + "question_timestamps": "00:39:35", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used it five times.", + "B": "You used it seven times.", + "C": "You used the white long bread rack six times.", + "D": "You used the white long bread rack four times so far, which is fewer than six.", + "E": "You used the white long bread rack eight times in total up to this point, not six." + }, + "answer_label": "C", + "answer_text": "You used the white long bread rack six times.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_13", + "correct_answer": "You used the white long bread rack six times.", + "original_options": { + "A": "You used it five times.", + "B": "You used it seven times.", + "C": "You used the white long bread rack six times.", + "D": "You used the white long bread rack four times so far, which is fewer than six.", + "E": "You used the white long bread rack eight times in total up to this point, not six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_4", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to place the brush and shovel closer together, how far should I move them?", + "answer": "You can move the brush about 20cm to the lower right.", + "question_chinese": "如果我想让刷子和铲子放得近一些,应该移动多远?", + "answer_chinese": "你可以把刷子向右下方方移动20cm左右。", + "evidence": [ + { + "description": "They are currently about 25 centimeters apart, and you can move the brush about 20 centimeters to the right and downwards.", + "timestamps": { + "start": "00:00:56", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Move the brush 10cm right.", + "B": "Move the brush 30cm down.", + "C": "You can move the brush about 20cm to the lower left to bring it nearer to the shovel.", + "D": "You can move the brush about 15cm to the lower right so the brush and shovel end up closer together.", + "E": "You can move the brush about 20cm to the lower right." + }, + "answer_label": "E", + "answer_text": "You can move the brush about 20cm to the lower right.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_4", + "correct_answer": "You can move the brush about 20cm to the lower right.", + "original_options": { + "A": "Move the brush 10cm right.", + "B": "Move the brush 30cm down.", + "C": "You can move the brush about 20cm to the lower left to bring it nearer to the shovel.", + "D": "You can move the brush about 15cm to the lower right so the brush and shovel end up closer together.", + "E": "You can move the brush about 20cm to the lower right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 20 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_3", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to place the brush and shovel closer together, how far should I move them?", + "answer": "You can move the brush about 10cm to the left.", + "question_chinese": "如果我想让刷子和铲子放得近一些,应该移动多远?", + "answer_chinese": "你可以把刷子向左移动10cm左右。", + "evidence": [ + { + "description": "They are currently about 25 centimeters apart, and you can move the brush about 20 centimeters to the right and downwards.", + "timestamps": { + "start": "00:00:56", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:36:33", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Move the brush 5cm left.", + "B": "You can move the brush about 10cm to the left.", + "C": "Move the brush 15cm left.", + "D": "You can move the brush about 10cm to the right to reduce the gap between it and the shovel.", + "E": "You can move the shovel about 10cm to the left instead, bringing it closer to the brush." + }, + "answer_label": "B", + "answer_text": "You can move the brush about 10cm to the left.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_3", + "correct_answer": "You can move the brush about 10cm to the left.", + "original_options": { + "A": "Move the brush 5cm left.", + "B": "You can move the brush about 10cm to the left.", + "C": "Move the brush 15cm left.", + "D": "You can move the brush about 10cm to the right to reduce the gap between it and the shovel.", + "E": "You can move the shovel about 10cm to the left instead, bringing it closer to the brush." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 17 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_6", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I finish scraping the flour, on which side should I place the scraper?", + "answer": "The scraper should be placed on the operating platform behind you.", + "question_chinese": "当我刮完面粉后，刮子应该放在我的哪一侧？", + "answer_chinese": "刮子应该放在你后方的操作台上。", + "evidence": [ + { + "description": "There are several brushes placed on the workbench behind the one where the dough is placed.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:04:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The scraper should be placed on the operating platform behind you.", + "B": "Place it on the counter in front.", + "C": "Place it on the platform to your left.", + "D": "Put the scraper on the operating platform in front of you so it stays within easy reach after scraping.", + "E": "Set the scraper on the operating platform to your right, not behind you, once you finish scraping the flour." + }, + "answer_label": "A", + "answer_text": "The scraper should be placed on the operating platform behind you.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_6", + "correct_answer": "The scraper should be placed on the operating platform behind you.", + "original_options": { + "A": "The scraper should be placed on the operating platform behind you.", + "B": "Place it on the counter in front.", + "C": "Place it on the platform to your left.", + "D": "Put the scraper on the operating platform in front of you so it stays within easy reach after scraping.", + "E": "Set the scraper on the operating platform to your right, not behind you, once you finish scraping the flour." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 19 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_4_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I finish scraping the flour, which side of me should I put the scraper on?", + "answer": "The scraper is on the workbench in front of you.", + "question_chinese": "当我刮完面粉后，刮子应该放在我的哪一侧？", + "answer_chinese": "刮子就放在你面前的操作台上。", + "evidence": [ + { + "description": "The scraper is placed on the operation table in front of you.", + "timestamps": { + "start": "00:16:53", + "end": "00:16:55" + } + } + ], + "question_timestamps": "00:20:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the workbench behind you.", + "B": "On the workbench to your left.", + "C": "The scraper is on the workbench in front of you.", + "D": "You should put the scraper on the workbench to your right, rather than placing it in front of you.", + "E": "After scraping the flour, leave the scraper on the workbench behind you so it’s out of the way." + }, + "answer_label": "C", + "answer_text": "The scraper is on the workbench in front of you.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_4_T1", + "correct_answer": "The scraper is on the workbench in front of you.", + "original_options": { + "A": "On the workbench behind you.", + "B": "On the workbench to your left.", + "C": "The scraper is on the workbench in front of you.", + "D": "You should put the scraper on the workbench to your right, rather than placing it in front of you.", + "E": "After scraping the flour, leave the scraper on the workbench behind you so it’s out of the way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 19 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_4_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I finish scraping the flour, which side of me should I put the scraper on?", + "answer": "The scraper should be placed on the workbench behind you.", + "question_chinese": "当我刮完面粉后，刮子应该放在我的哪一侧？", + "answer_chinese": "刮子应该放在你后方的操作台上。", + "evidence": [ + { + "description": "There are several brushes placed on the workbench behind the one where the dough is placed.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:32:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Place it on the workbench in front.", + "B": "Place it on the workbench to your right.", + "C": "The scraper should be placed on the workbench behind you.", + "D": "You should place the scraper on the workbench to your left instead of putting it behind you.", + "E": "After scraping the flour, put the scraper on the workbench in front of you, not on the bench behind you." + }, + "answer_label": "C", + "answer_text": "The scraper should be placed on the workbench behind you.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_4_T2", + "correct_answer": "The scraper should be placed on the workbench behind you.", + "original_options": { + "A": "Place it on the workbench in front.", + "B": "Place it on the workbench to your right.", + "C": "The scraper should be placed on the workbench behind you.", + "D": "You should place the scraper on the workbench to your left instead of putting it behind you.", + "E": "After scraping the flour, put the scraper on the workbench in front of you, not on the bench behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 20 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_1_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I am facing the door, which side of me is the bread rack on?", + "answer": "The bread rack is on the left side of you.", + "question_chinese": "当我面对门时，面包架在我的哪一侧？", + "answer_chinese": "面包架在你的左侧。", + "evidence": [ + { + "description": "When you face the door, the bread rack is on your left.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bread rack is on your right.", + "B": "The bread rack is on the left side of you.", + "C": "The bread rack is behind you.", + "D": "When you face the door, the bread rack is positioned in front of you rather than to your left.", + "E": "Facing the door, you’ll find the bread rack on your right side, not on the left." + }, + "answer_label": "B", + "answer_text": "The bread rack is on the left side of you.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_1_T1", + "correct_answer": "The bread rack is on the left side of you.", + "original_options": { + "A": "The bread rack is on your right.", + "B": "The bread rack is on the left side of you.", + "C": "The bread rack is behind you.", + "D": "When you face the door, the bread rack is positioned in front of you rather than to your left.", + "E": "Facing the door, you’ll find the bread rack on your right side, not on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 17 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_1_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I am facing the door, which side of me is the bread rack on?", + "answer": "When you’re facing the door, the bread rack is on your left-hand side.", + "question_chinese": "当我面对门时，面包架在我的哪一侧？", + "answer_chinese": "当你面对门时，面包架在你的左手边。", + "evidence": [ + { + "description": "When you face the door, the bread rack is on your left.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:33:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on your right-hand side.", + "B": "When you’re facing the door, the bread rack is on your left-hand side.", + "C": "It’s directly in front of you.", + "D": "When you’re facing the door, the bread rack is behind you, not on the side you’re facing.", + "E": "When you’re facing the door, the bread rack is on your right, opposite the side you’d expect." + }, + "answer_label": "B", + "answer_text": "When you’re facing the door, the bread rack is on your left-hand side.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_1_T2", + "correct_answer": "When you’re facing the door, the bread rack is on your left-hand side.", + "original_options": { + "A": "It’s on your right-hand side.", + "B": "When you’re facing the door, the bread rack is on your left-hand side.", + "C": "It’s directly in front of you.", + "D": "When you’re facing the door, the bread rack is behind you, not on the side you’re facing.", + "E": "When you’re facing the door, the bread rack is on your right, opposite the side you’d expect." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 19 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_9", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I turned around just now, which side of the door was it on?", + "answer": "The door appears behind you.", + "question_chinese": "我刚刚转身的时候，门出现在我的哪一侧？", + "answer_chinese": "门出现你的后方。", + "evidence": [ + { + "description": "The door is on the right side of the bread rack. After you turn around, the bread rack will be on your right. Turning right twice means going backwards.", + "timestamps": { + "start": "00:36:01", + "end": "00:36:03" + } + } + ], + "question_timestamps": "00:36:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door appears in front of you.", + "B": "The door appears to your right.", + "C": "After you turn around, the door is now off to your left side.", + "D": "The door appears behind you.", + "E": "When you turn around, the door ends up in front of you rather than behind you." + }, + "answer_label": "D", + "answer_text": "The door appears behind you.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_9", + "correct_answer": "The door appears behind you.", + "original_options": { + "A": "The door appears in front of you.", + "B": "The door appears to your right.", + "C": "After you turn around, the door is now off to your left side.", + "D": "The door appears behind you.", + "E": "When you turn around, the door ends up in front of you rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 16 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_8", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I turned around just now, which side of the door was it on?", + "answer": "The door is on your left.", + "question_chinese": "我刚刚转身的时候，门出现在我的哪一侧？", + "answer_chinese": "门出现在你的左侧。", + "evidence": [ + { + "description": "The door is on the right side of the bread rack. After you turn around, the bread rack will be on your right. Turning right twice means going backwards.", + "timestamps": { + "start": "00:36:01", + "end": "00:36:03" + } + } + ], + "question_timestamps": "00:36:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is on your right.", + "B": "The door is in front of you.", + "C": "The door is on your left.", + "D": "After you turned around, the door ended up behind you instead of on your left.", + "E": "When you turned around, the door shifted to your right side rather than staying on your left." + }, + "answer_label": "C", + "answer_text": "The door is on your left.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_8", + "correct_answer": "The door is on your left.", + "original_options": { + "A": "The door is on your right.", + "B": "The door is in front of you.", + "C": "The door is on your left.", + "D": "After you turned around, the door ended up behind you instead of on your left.", + "E": "When you turned around, the door shifted to your right side rather than staying on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 17 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_17", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route am I taking?", + "answer": "You are turning around from the oven, walking straight down the aisle, and moving towards the operating table.", + "question_chinese": "我正在走什么路线？", + "answer_chinese": "你正从烤炉转身，在过道直行，朝着操作台的方向移动。", + "evidence": [ + { + "description": "You turn around from the oven and walk straight along the aisle to the workstation.", + "timestamps": { + "start": "00:21:40", + "end": "00:21:51" + } + } + ], + "question_timestamps": "00:21:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You walk toward the refrigerator.", + "B": "You are turning around from the oven, walking straight down the aisle, and moving towards the operating table.", + "C": "You head back toward the sink.", + "D": "You turn away from the oven, walk straight down the aisle, and move toward the refrigerator.", + "E": "You turn around from the oven, walk down the aisle, and then veer toward the pantry area." + }, + "answer_label": "B", + "answer_text": "You are turning around from the oven, walking straight down the aisle, and moving towards the operating table.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_17", + "correct_answer": "You are turning around from the oven, walking straight down the aisle, and moving towards the operating table.", + "original_options": { + "A": "You walk toward the refrigerator.", + "B": "You are turning around from the oven, walking straight down the aisle, and moving towards the operating table.", + "C": "You head back toward the sink.", + "D": "You turn away from the oven, walk straight down the aisle, and move toward the refrigerator.", + "E": "You turn around from the oven, walk down the aisle, and then veer toward the pantry area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_16", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route am I taking?", + "answer": "You are turning around from the aisle, walking straight, and moving towards the direction of the warehouse.", + "question_chinese": "我正在走什么路线？", + "answer_chinese": "你正从过道转身，直行，朝着仓库的方向移动。", + "evidence": [ + { + "description": "You turn around from the oven and walk straight along the aisle to the workstation.", + "timestamps": { + "start": "00:21:40", + "end": "00:21:51" + } + } + ], + "question_timestamps": "00:44:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You’re heading toward the entrance.", + "B": "You turn right toward the corner.", + "C": "You turn around from the aisle, walk straight, and move toward the entrance of the store.", + "D": "You are turning around from the aisle, walking straight, and moving towards the direction of the warehouse.", + "E": "You turn around from the aisle, walk forward, and head toward the garden area outside instead." + }, + "answer_label": "D", + "answer_text": "You are turning around from the aisle, walking straight, and moving towards the direction of the warehouse.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_16", + "correct_answer": "You are turning around from the aisle, walking straight, and moving towards the direction of the warehouse.", + "original_options": { + "A": "You’re heading toward the entrance.", + "B": "You turn right toward the corner.", + "C": "You turn around from the aisle, walk straight, and move toward the entrance of the store.", + "D": "You are turning around from the aisle, walking straight, and moving towards the direction of the warehouse.", + "E": "You turn around from the aisle, walk forward, and head toward the garden area outside instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_19", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I've finished using the spatula, where should I put it?", + "answer": "The grill spatula should be placed next to the wall on your left.", + "question_chinese": "我用完烤铲了，它应该放到哪里？", + "answer_chinese": "烤铲应该放到你左侧的墙旁边。", + "evidence": [ + { + "description": "You had placed the spatula beside the wall earlier.", + "timestamps": { + "start": "00:39:43", + "end": "00:39:46" + } + } + ], + "question_timestamps": "00:39:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put it on the counter behind you.", + "B": "Put it in the drawer to your right.", + "C": "The grill spatula should be placed next to the wall on your left.", + "D": "Place the grill spatula on the counter behind you rather than next to the left wall.", + "E": "Store the grill spatula on a shelf above you instead of placing it by the wall." + }, + "answer_label": "C", + "answer_text": "The grill spatula should be placed next to the wall on your left.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_19", + "correct_answer": "The grill spatula should be placed next to the wall on your left.", + "original_options": { + "A": "Put it on the counter behind you.", + "B": "Put it in the drawer to your right.", + "C": "The grill spatula should be placed next to the wall on your left.", + "D": "Place the grill spatula on the counter behind you rather than next to the left wall.", + "E": "Store the grill spatula on a shelf above you instead of placing it by the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 16 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_10_T2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I’m done using the spatula—where should I put it?", + "answer": "The spatula should be placed next to the wall on your left.", + "question_chinese": "我用完烤铲了，它应该放到哪里？", + "answer_chinese": "烤铲应该放到你左侧的墙旁边。", + "evidence": [ + { + "description": "You had placed the spatula beside the wall earlier.", + "timestamps": { + "start": "00:39:43", + "end": "00:39:46" + } + } + ], + "question_timestamps": "00:52:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it next to the wall ahead.", + "B": "Put it next to the wall behind.", + "C": "Place the spatula next to the wall behind you, not along the left side.", + "D": "The spatula should be placed next to the wall on your left.", + "E": "Set the spatula beside the wall in front of you, rather than next to the wall on your left." + }, + "answer_label": "D", + "answer_text": "The spatula should be placed next to the wall on your left.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_10_T2", + "correct_answer": "The spatula should be placed next to the wall on your left.", + "original_options": { + "A": "Put it next to the wall ahead.", + "B": "Put it next to the wall behind.", + "C": "Place the spatula next to the wall behind you, not along the left side.", + "D": "The spatula should be placed next to the wall on your left.", + "E": "Set the spatula beside the wall in front of you, rather than next to the wall on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 19 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_10_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I’m done using the spatula—where should I put it?", + "answer": "The spatula should be placed on the wall behind you.", + "question_chinese": "我用完烤铲了，它应该放到哪里？", + "answer_chinese": "烤铲应该放到你后方的墙上。", + "evidence": [ + { + "description": "The baking spatula should be placed against the wall behind you.", + "timestamps": { + "start": "00:39:43", + "end": "00:39:46" + } + } + ], + "question_timestamps": "01:01:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the wall in front of you.", + "B": "On the wall to your left.", + "C": "Put the spatula on the wall to your right instead of placing it behind you.", + "D": "The spatula should be placed on the wall behind you.", + "E": "Hang the spatula on the wall above you rather than on the wall behind you." + }, + "answer_label": "D", + "answer_text": "The spatula should be placed on the wall behind you.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_10_T1", + "correct_answer": "The spatula should be placed on the wall behind you.", + "original_options": { + "A": "On the wall in front of you.", + "B": "On the wall to your left.", + "C": "Put the spatula on the wall to your right instead of placing it behind you.", + "D": "The spatula should be placed on the wall behind you.", + "E": "Hang the spatula on the wall above you rather than on the wall behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 15 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_5", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I filled this yellow container with flour?", + "answer": "You used it to store flour twice.", + "question_chinese": "我用这个黄色的容器装了几次面粉？", + "answer_chinese": "你用它装了两次面粉。", + "evidence": [ + { + "description": "You pick up the yellow container filled with flour.", + "timestamps": { + "start": "00:02:27", + "end": "00:02:30" + } + }, + { + "description": "You used it to pack flour again.", + "timestamps": { + "start": "00:02:42", + "end": "00:02:50" + } + }, + { + "description": "You used a yellow container to store flour for the third time.", + "timestamps": { + "start": "00:44:58", + "end": "00:45:10" + } + } + ], + "question_timestamps": "00:45:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used it to store flour once.", + "B": "You used it to store flour twice.", + "C": "You used it to store flour three times.", + "D": "You used it to store flour four times during the earlier pours into the yellow container.", + "E": "You used it to store flour five times, refilling the yellow container on five separate occasions." + }, + "answer_label": "B", + "answer_text": "You used it to store flour twice.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_5", + "correct_answer": "You used it to store flour twice.", + "original_options": { + "A": "You used it to store flour once.", + "B": "You used it to store flour twice.", + "C": "You used it to store flour three times.", + "D": "You used it to store flour four times during the earlier pours into the yellow container.", + "E": "You used it to store flour five times, refilling the yellow container on five separate occasions." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 16 + ] + }, + { + "video_uid": "3fffb452-1c17-43fa-b473-c9e19ac21433", + "video_length": "01:02:11", + "qid": "3fffb452-1c17-43fa-b473-c9e19ac21433_18", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I filled this yellow container with flour?", + "answer": "You used it to store flour three times.", + "question_chinese": "我用这个黄色的容器装了几次面粉？", + "answer_chinese": "你用它装了三次面粉。", + "evidence": [ + { + "description": "You pick up the yellow container filled with flour.", + "timestamps": { + "start": "00:02:27", + "end": "00:02:30" + } + }, + { + "description": "You used it to pack flour again.", + "timestamps": { + "start": "00:02:42", + "end": "00:02:50" + } + }, + { + "description": "You used a yellow container to store flour for the third time.", + "timestamps": { + "start": "00:44:58", + "end": "00:45:10" + } + } + ], + "question_timestamps": "00:45:14", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used it to store flour three times.", + "B": "You used it to store flour once.", + "C": "You used it to store flour twice.", + "D": "You used it to store flour four times, refilling the yellow container on four occasions.", + "E": "You used it to store flour six times across the earlier refills of the yellow container." + }, + "answer_label": "A", + "answer_text": "You used it to store flour three times.", + "q_id": "3fffb452-1c17-43fa-b473-c9e19ac21433_18", + "correct_answer": "You used it to store flour three times.", + "original_options": { + "A": "You used it to store flour three times.", + "B": "You used it to store flour once.", + "C": "You used it to store flour twice.", + "D": "You used it to store flour four times, refilling the yellow container on four occasions.", + "E": "You used it to store flour six times across the earlier refills of the yellow container." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + } +] diff --git a/QAs/40a83cc4-0e1e-4de4-8a7b-d028ecd0c214-mcq.json b/QAs/40a83cc4-0e1e-4de4-8a7b-d028ecd0c214-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..169afe02d18425ad0ea11989430385a22e1e9edd --- /dev/null +++ b/QAs/40a83cc4-0e1e-4de4-8a7b-d028ecd0c214-mcq.json @@ -0,0 +1,1089 @@ +[ + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_1_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the toolbox relative to me?", + "answer": "The toolbox was to your right.", + "question_chinese": "工具箱在我的哪里？", + "answer_chinese": "工具箱在你的右边", + "evidence": [ + { + "description": "The toolbox is on your right.", + "timestamps": { + "start": "00:10:09", + "end": "00:10:15" + } + } + ], + "question_timestamps": "00:10:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The toolbox was to your left.", + "B": "The toolbox was behind you.", + "C": "The toolbox was in front of you, positioned directly ahead from your viewpoint.", + "D": "The toolbox was above you, higher up than your eye level as you looked around.", + "E": "The toolbox was to your right." + }, + "answer_label": "E", + "answer_text": "The toolbox was to your right.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_1_T1", + "correct_answer": "The toolbox was to your right.", + "original_options": { + "A": "The toolbox was to your left.", + "B": "The toolbox was behind you.", + "C": "The toolbox was in front of you, positioned directly ahead from your viewpoint.", + "D": "The toolbox was above you, higher up than your eye level as you looked around.", + "E": "The toolbox was to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 15 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_1_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the toolbox relative to me?", + "answer": "The toolbox was in front of you.", + "question_chinese": "工具箱在我的哪里？", + "answer_chinese": "工具箱在你的前面", + "evidence": [ + { + "description": "The toolbox is in front of you.", + "timestamps": { + "start": "00:11:59", + "end": "00:12:04" + } + } + ], + "question_timestamps": "00:12:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The toolbox was behind you.", + "B": "The toolbox was to your left.", + "C": "The toolbox was to your right, off to the side as you faced forward.", + "D": "The toolbox was in front of you.", + "E": "The toolbox was above you, positioned higher than you rather than directly in front." + }, + "answer_label": "D", + "answer_text": "The toolbox was in front of you.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_1_T2", + "correct_answer": "The toolbox was in front of you.", + "original_options": { + "A": "The toolbox was behind you.", + "B": "The toolbox was to your left.", + "C": "The toolbox was to your right, off to the side as you faced forward.", + "D": "The toolbox was in front of you.", + "E": "The toolbox was above you, positioned higher than you rather than directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 14 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_3_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I fix the earlier electrical wire?", + "answer": "No. You were fixing the earlier electrical wire, and the wire was in front of you.", + "question_chinese": "我修好之前的电线了吗？", + "answer_chinese": "没有，你正在修之前的电线，电线在你前面。", + "evidence": [ + { + "description": "You haven’t finished repairing the previous wire.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:31" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. You were fixing the earlier electrical wire, and the wire was in front of you.", + "B": "Yes. You had already fixed the earlier electrical wire, and the wire was above you." + }, + "answer_label": "A", + "answer_text": "No. You were fixing the earlier electrical wire, and the wire was in front of you.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_3_T1", + "correct_answer": "No. You were fixing the earlier electrical wire, and the wire was in front of you.", + "original_options": { + "A": "No. You were fixing the earlier electrical wire, and the wire was in front of you.", + "B": "Yes. You had already fixed the earlier electrical wire, and the wire was above you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_3_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I fix the earlier electrical wire?", + "answer": "You had already fixed the earlier electrical wire, and the wire was above you.", + "question_chinese": "我修好之前的电线了吗？", + "answer_chinese": "你已经修好了之前的电线，电线在你上面。", + "evidence": [ + { + "description": "You have finished repairing the previous wire.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:42" + } + } + ], + "question_timestamps": "00:11:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. You were fixing the earlier electrical wire, and the wire was in front of you.", + "B": "You had already fixed the earlier electrical wire, and the wire was above you." + }, + "answer_label": "B", + "answer_text": "You had already fixed the earlier electrical wire, and the wire was above you.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_3_T2", + "correct_answer": "You had already fixed the earlier electrical wire, and the wire was above you.", + "original_options": { + "A": "No. You were fixing the earlier electrical wire, and the wire was in front of you.", + "B": "You had already fixed the earlier electrical wire, and the wire was above you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_5", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Have I finished repairing the previous wire?", + "answer": "You have finished repairing the previous wire.", + "question_chinese": "我修好之前的电线了吗？", + "answer_chinese": "你已经修好了之前的电线。", + "evidence": [ + { + "description": "You have finished repairing the previous wire.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:47" + } + } + ], + "question_timestamps": "00:14:41", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You have finished repairing the previous wire.", + "B": "You have not finished repairing the previous wire." + }, + "answer_label": "A", + "answer_text": "You have finished repairing the previous wire.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_5", + "correct_answer": "You have finished repairing the previous wire.", + "original_options": { + "A": "You have finished repairing the previous wire.", + "B": "You have not finished repairing the previous wire." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_6", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Have I finished repairing the previous wire?", + "answer": "You have finished repairing the previous wire.", + "question_chinese": "我修好之前的电线了吗？", + "answer_chinese": "你已经修好了之前的电线。", + "evidence": [ + { + "description": "You have finished repairing the previous wire.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:47" + } + } + ], + "question_timestamps": "00:19:35", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You have finished repairing the previous wire.", + "B": "You have not finished repairing the previous wire." + }, + "answer_label": "A", + "answer_text": "You have finished repairing the previous wire.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_6", + "correct_answer": "You have finished repairing the previous wire.", + "original_options": { + "A": "You have finished repairing the previous wire.", + "B": "You have not finished repairing the previous wire." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_8", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many light bulbs did I just fix?", + "answer": "You just fixed two light bulbs.", + "question_chinese": "我刚刚修好了多少个灯泡？", + "answer_chinese": "你刚刚修好了2个灯泡", + "evidence": [ + { + "description": "You fixed the first light bulb.", + "timestamps": { + "start": "00:18:12", + "end": "00:18:35" + } + }, + { + "description": "You fixed the second light bulb.", + "timestamps": { + "start": "00:18:46", + "end": "00:19:09" + } + } + ], + "question_timestamps": "00:19:16", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You just fixed one light bulb.", + "B": "You just fixed two light bulbs.", + "C": "You just fixed three light bulbs.", + "D": "You just fixed four light bulbs after replacing the bad ones and checking they turned on.", + "E": "You just fixed five light bulbs by swapping them out and confirming each one worked afterward." + }, + "answer_label": "B", + "answer_text": "You just fixed two light bulbs.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_8", + "correct_answer": "You just fixed two light bulbs.", + "original_options": { + "A": "You just fixed one light bulb.", + "B": "You just fixed two light bulbs.", + "C": "You just fixed three light bulbs.", + "D": "You just fixed four light bulbs after replacing the bad ones and checking they turned on.", + "E": "You just fixed five light bulbs by swapping them out and confirming each one worked afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_7", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many light bulbs did I just fix?", + "answer": "You just fixed four light bulbs.", + "question_chinese": "我刚刚修好了多少个灯泡？", + "answer_chinese": "你刚刚修好了4个灯泡", + "evidence": [ + { + "description": "You fixed the first light bulb.", + "timestamps": { + "start": "00:18:12", + "end": "00:18:35" + } + }, + { + "description": "You fixed the second light bulb.", + "timestamps": { + "start": "00:18:46", + "end": "00:19:09" + } + }, + { + "description": "When I first climbed the ladder, what was in my pocket?", + "timestamps": { + "start": "00:19:57", + "end": "00:20:23" + } + } + ], + "question_timestamps": "00:22:23", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You just fixed three light bulbs.", + "B": "You just fixed five light bulbs.", + "C": "You just fixed two light bulbs by replacing them and making sure they lit up properly.", + "D": "You just fixed four light bulbs.", + "E": "You just fixed seven light bulbs after changing them out and testing each one to confirm it worked." + }, + "answer_label": "D", + "answer_text": "You just fixed four light bulbs.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_7", + "correct_answer": "You just fixed four light bulbs.", + "original_options": { + "A": "You just fixed three light bulbs.", + "B": "You just fixed five light bulbs.", + "C": "You just fixed two light bulbs by replacing them and making sure they lit up properly.", + "D": "You just fixed four light bulbs.", + "E": "You just fixed seven light bulbs after changing them out and testing each one to confirm it worked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just packed up the items on the ground with a black bag?", + "answer": "After you picked up the black bag next to the cardboard box, you set off towards the ladder to put things away, and then returned to the cardboard box.", + "question_chinese": "我刚才用黑色袋子把地上的东西打包好时，走的是什么路线？", + "answer_chinese": "你拿起纸箱旁边的黑色袋子后，就朝着梯子走去放东西，然后又回到了纸箱旁。", + "evidence": [ + { + "description": "Pick up the black bag and start from beside the carton.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:22" + } + }, + { + "description": "Walk to the ladder and pack up.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:50" + } + }, + { + "description": "Finally, I walked to the cardboard box to pack things up.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:05:40", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You went from the box to the ladder.", + "B": "You walked from the ladder to the box.", + "C": "After picking up the black bag by the cardboard box, you walked past the ladder and returned to the box.", + "D": "You picked up the black bag near the cardboard box, went to the ladder area, then continued onward instead of returning.", + "E": "After you picked up the black bag next to the cardboard box, you set off towards the ladder to put things away, and then returned to the cardboard box." + }, + "answer_label": "E", + "answer_text": "After you picked up the black bag next to the cardboard box, you set off towards the ladder to put things away, and then returned to the cardboard box.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_2", + "correct_answer": "After you picked up the black bag next to the cardboard box, you set off towards the ladder to put things away, and then returned to the cardboard box.", + "original_options": { + "A": "You went from the box to the ladder.", + "B": "You walked from the ladder to the box.", + "C": "After picking up the black bag by the cardboard box, you walked past the ladder and returned to the box.", + "D": "You picked up the black bag near the cardboard box, went to the ladder area, then continued onward instead of returning.", + "E": "After you picked up the black bag next to the cardboard box, you set off towards the ladder to put things away, and then returned to the cardboard box." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 20, + 21 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_4", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just packed up the items on the ground with a black bag?", + "answer": "After you picked up the black bag next to the cardboard box, you set off towards the ladder to put things away, and then returned to the cardboard box.", + "question_chinese": "我刚才用黑色袋子把地上的东西打包好时，走的是什么路线？", + "answer_chinese": "你拿起纸箱旁边的黑色袋子后，就朝着梯子走去放东西，然后又回到了纸箱旁。", + "evidence": [ + { + "description": "Pick up the black bag and start from beside the carton.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:22" + } + }, + { + "description": "Walk to the ladder and pack up.", + "timestamps": { + "start": "00:09:57", + "end": "00:10:03" + } + }, + { + "description": "To get back to your work position, you need to turn around first and then walk straight towards the ladder.", + "timestamps": { + "start": "00:10:03", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:14:41", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "After you picked up the black bag next to the cardboard box, you set off towards the ladder to put things away, and then returned to the cardboard box.", + "B": "You walked from the ladder to the box.", + "C": "You went to the door.", + "D": "After you picked up the black bag by the cardboard box, you went to the ladder and then walked away.", + "E": "You grabbed the black bag next to the cardboard box, headed to the ladder, and returned to a different spot." + }, + "answer_label": "A", + "answer_text": "After you picked up the black bag next to the cardboard box, you set off towards the ladder to put things away, and then returned to the cardboard box.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_4", + "correct_answer": "After you picked up the black bag next to the cardboard box, you set off towards the ladder to put things away, and then returned to the cardboard box.", + "original_options": { + "A": "After you picked up the black bag next to the cardboard box, you set off towards the ladder to put things away, and then returned to the cardboard box.", + "B": "You walked from the ladder to the box.", + "C": "You went to the door.", + "D": "After you picked up the black bag by the cardboard box, you went to the ladder and then walked away.", + "E": "You grabbed the black bag next to the cardboard box, headed to the ladder, and returned to a different spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 5, + 20, + 20 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_9", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly get the tools on the windowsill?", + "answer": "You can't reach the tools on the windowsill; you need to move a certain distance.", + "question_chinese": "我可以直接拿窗台上的工具吗？", + "answer_chinese": "你够不到窗台上的工具，你得移动一段距离。", + "evidence": [ + { + "description": "You can't touch the tools on the window.", + "timestamps": { + "start": "00:16:07", + "end": "00:16:09" + } + }, + { + "description": "You need to move a bit to get the tool on the window.", + "timestamps": { + "start": "00:15:44", + "end": "00:15:53" + } + } + ], + "question_timestamps": "00:19:35", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Yes, you can reach them directly.", + "B": "You can't reach the tools on the windowsill; you need to move a certain distance.", + "C": "No, not from here.", + "D": "You can directly grab the tools on the windowsill without moving your feet from where you are.", + "E": "You can reach the tools on the windowsill by leaning forward slightly, without needing to change position." + }, + "answer_label": "B", + "answer_text": "You can't reach the tools on the windowsill; you need to move a certain distance.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_9", + "correct_answer": "You can't reach the tools on the windowsill; you need to move a certain distance.", + "original_options": { + "A": "Yes, you can reach them directly.", + "B": "You can't reach the tools on the windowsill; you need to move a certain distance.", + "C": "No, not from here.", + "D": "You can directly grab the tools on the windowsill without moving your feet from where you are.", + "E": "You can reach the tools on the windowsill by leaning forward slightly, without needing to change position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 17, + 17 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_10", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly get the tools on the windowsill?", + "answer": "You can't reach the tools on the windowsill; you have to move a certain distance.", + "question_chinese": "我可以直接拿窗台上的工具吗？", + "answer_chinese": "你够不到窗台上的工具，你得移动一段距离。", + "evidence": [ + { + "description": "You can't reach the windowsill while standing on the ladder.", + "timestamps": { + "start": "00:20:28", + "end": "00:20:30" + } + } + ], + "question_timestamps": "00:20:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You can't reach the tools on the windowsill; you have to move a certain distance.", + "B": "Yes, you can reach them.", + "C": "Yes, they’re within arm’s reach.", + "D": "Yes, you can grab the tools on the windowsill without moving closer or changing where you’re standing.", + "E": "You can directly reach the tools on the windowsill; they’re close enough to take immediately with your hand." + }, + "answer_label": "A", + "answer_text": "You can't reach the tools on the windowsill; you have to move a certain distance.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_10", + "correct_answer": "You can't reach the tools on the windowsill; you have to move a certain distance.", + "original_options": { + "A": "You can't reach the tools on the windowsill; you have to move a certain distance.", + "B": "Yes, you can reach them.", + "C": "Yes, they’re within arm’s reach.", + "D": "Yes, you can grab the tools on the windowsill without moving closer or changing where you’re standing.", + "E": "You can directly reach the tools on the windowsill; they’re close enough to take immediately with your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 18, + 19 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_11", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get the wire tool I just had?", + "answer": "You have to get down the ladder first, then walk straight towards the box to retrieve the tools from the ground.", + "question_chinese": "我怎么将刚刚的铁丝工具拿过来？", + "answer_chinese": "你得先下梯子，然后直走走向箱子将工具从地上拿回", + "evidence": [ + { + "description": "You need to get down the ladder first.", + "timestamps": { + "start": "00:11:56", + "end": "00:11:59" + } + }, + { + "description": "Then you should go straight towards the box and get it back on the ground.", + "timestamps": { + "start": "00:11:59", + "end": "00:12:04" + } + } + ], + "question_timestamps": "00:15:06", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Climb the ladder, then go box.", + "B": "Go to the box first.", + "C": "You have to get down the ladder first, then walk straight towards the box to retrieve the tools from the ground.", + "D": "Get down the ladder first, then walk straight toward the shelf to pick the tools up from the ground.", + "E": "Get down the ladder, then walk around the box before grabbing the tools from the ground beside it." + }, + "answer_label": "C", + "answer_text": "You have to get down the ladder first, then walk straight towards the box to retrieve the tools from the ground.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_11", + "correct_answer": "You have to get down the ladder first, then walk straight towards the box to retrieve the tools from the ground.", + "original_options": { + "A": "Climb the ladder, then go box.", + "B": "Go to the box first.", + "C": "You have to get down the ladder first, then walk straight towards the box to retrieve the tools from the ground.", + "D": "Get down the ladder first, then walk straight toward the shelf to pick the tools up from the ground.", + "E": "Get down the ladder, then walk around the box before grabbing the tools from the ground beside it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 18 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_12", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get the wire tool I just had?", + "answer": "Go straight to the cardboard box and pick it up from the ground.", + "question_chinese": "我怎么将刚刚的铁丝工具拿过来？", + "answer_chinese": "直走到纸箱旁从地上拿起", + "evidence": [ + { + "description": "You should go straight towards the box and get it back on the ground.", + "timestamps": { + "start": "00:11:59", + "end": "00:12:04" + } + } + ], + "question_timestamps": "00:19:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go to the cardboard box and grab it.", + "B": "Pick it up from the box’s top.", + "C": "Walk straight to the cardboard box and take the wire tool from the top instead of the ground.", + "D": "Go to the cardboard box, but retrieve the wire tool from inside it rather than from the ground.", + "E": "Go straight to the cardboard box and pick it up from the ground." + }, + "answer_label": "E", + "answer_text": "Go straight to the cardboard box and pick it up from the ground.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_12", + "correct_answer": "Go straight to the cardboard box and pick it up from the ground.", + "original_options": { + "A": "Go to the cardboard box and grab it.", + "B": "Pick it up from the box’s top.", + "C": "Walk straight to the cardboard box and take the wire tool from the top instead of the ground.", + "D": "Go to the cardboard box, but retrieve the wire tool from inside it rather than from the ground.", + "E": "Go straight to the cardboard box and pick it up from the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_5_T2", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the ladder relative to me?", + "answer": "The ladder was in front of you.", + "question_chinese": "梯子在我的哪里？", + "answer_chinese": "梯子在你的前面", + "evidence": [ + { + "description": "You can't reach the windowsill while standing on the ladder.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:12" + } + } + ], + "question_timestamps": "00:05:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The ladder was behind you.", + "B": "The ladder was to your left.", + "C": "The ladder was to your right as you looked toward it, not directly in front of you.", + "D": "The ladder was above you, positioned higher than your eye level rather than straight ahead.", + "E": "The ladder was in front of you." + }, + "answer_label": "E", + "answer_text": "The ladder was in front of you.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_5_T2", + "correct_answer": "The ladder was in front of you.", + "original_options": { + "A": "The ladder was behind you.", + "B": "The ladder was to your left.", + "C": "The ladder was to your right as you looked toward it, not directly in front of you.", + "D": "The ladder was above you, positioned higher than your eye level rather than straight ahead.", + "E": "The ladder was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 15 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_5_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the ladder relative to me?", + "answer": "The ladder was below you.", + "question_chinese": "梯子在我的哪里？", + "answer_chinese": "梯子在你的下面", + "evidence": [ + { + "description": "The ladder is below you.", + "timestamps": { + "start": "00:06:27", + "end": "00:06:32" + } + } + ], + "question_timestamps": "00:06:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The ladder was in front of you.", + "B": "The ladder was behind you.", + "C": "The ladder was below you.", + "D": "The ladder was to your left side, not located below you near the floor.", + "E": "The ladder was to your right, positioned beside you rather than underneath your viewpoint." + }, + "answer_label": "C", + "answer_text": "The ladder was below you.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_5_T1", + "correct_answer": "The ladder was below you.", + "original_options": { + "A": "The ladder was in front of you.", + "B": "The ladder was behind you.", + "C": "The ladder was below you.", + "D": "The ladder was to your left side, not located below you near the floor.", + "E": "The ladder was to your right, positioned beside you rather than underneath your viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 14, + 14 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_2_T2", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the window relative to me?", + "answer": "The window was to your right.", + "question_chinese": "窗户在我的哪里？", + "answer_chinese": "窗户在你的右边。", + "evidence": [ + { + "description": "The window is on your right.", + "timestamps": { + "start": "00:12:04", + "end": "00:12:09" + } + } + ], + "question_timestamps": "00:12:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window was to your left.", + "B": "The window was in front of you.", + "C": "The window was behind you, so you would have needed to turn around to face it.", + "D": "The window was to your right.", + "E": "The window was above you, higher up rather than positioned on your right side." + }, + "answer_label": "D", + "answer_text": "The window was to your right.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_2_T2", + "correct_answer": "The window was to your right.", + "original_options": { + "A": "The window was to your left.", + "B": "The window was in front of you.", + "C": "The window was behind you, so you would have needed to turn around to face it.", + "D": "The window was to your right.", + "E": "The window was above you, higher up rather than positioned on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 14 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_2_T1", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the window relative to me?", + "answer": "The window was behind you.", + "question_chinese": "窗户在我的哪里？", + "answer_chinese": "窗户在你的后面", + "evidence": [ + { + "description": "The window is behind you.", + "timestamps": { + "start": "00:12:05", + "end": "00:12:14" + } + } + ], + "question_timestamps": "00:12:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window was in front of you.", + "B": "The window was to your right.", + "C": "The window was behind you.", + "D": "The window was to your left, positioned beside you rather than behind your back.", + "E": "The window was above you, located higher up instead of behind you in the room." + }, + "answer_label": "C", + "answer_text": "The window was behind you.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_2_T1", + "correct_answer": "The window was behind you.", + "original_options": { + "A": "The window was in front of you.", + "B": "The window was to your right.", + "C": "The window was behind you.", + "D": "The window was to your left, positioned beside you rather than behind your back.", + "E": "The window was above you, located higher up instead of behind you in the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 15 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_7_T2", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the box relative to me?", + "answer": "The box was in front of you.", + "question_chinese": "箱子在我的哪里？", + "answer_chinese": "箱子在你的前面", + "evidence": [ + { + "description": "The box is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The box was behind you.", + "B": "The box was to your left.", + "C": "The box was to your right side rather than positioned directly in front of you.", + "D": "The box was above you, placed higher up instead of sitting in front of your viewpoint.", + "E": "The box was in front of you." + }, + "answer_label": "E", + "answer_text": "The box was in front of you.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_7_T2", + "correct_answer": "The box was in front of you.", + "original_options": { + "A": "The box was behind you.", + "B": "The box was to your left.", + "C": "The box was to your right side rather than positioned directly in front of you.", + "D": "The box was above you, placed higher up instead of sitting in front of your viewpoint.", + "E": "The box was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214", + "video_length": "00:22:42", + "qid": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_7_T1", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the box relative to me?", + "answer": "The box was behind you.", + "question_chinese": "箱子在我的哪里？", + "answer_chinese": "箱子在你的后面", + "evidence": [ + { + "description": "To get back to your work position, you need to turn around first and then walk straight towards the ladder.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The box was in front of you.", + "B": "The box was to your left.", + "C": "The box was behind you.", + "D": "The box was to your right, positioned beside you instead of behind you as you moved.", + "E": "The box was above you, located higher up rather than behind you in the space." + }, + "answer_label": "C", + "answer_text": "The box was behind you.", + "q_id": "40a83cc4-0e1e-4de4-8a7b-d028ecd0c214_7_T1", + "correct_answer": "The box was behind you.", + "original_options": { + "A": "The box was in front of you.", + "B": "The box was to your left.", + "C": "The box was behind you.", + "D": "The box was to your right, positioned beside you instead of behind you as you moved.", + "E": "The box was above you, located higher up rather than behind you in the space." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 15 + ] + } +] diff --git a/QAs/45a964e8-4eb3-4b41-b9d3-72b949294e27-mcq.json b/QAs/45a964e8-4eb3-4b41-b9d3-72b949294e27-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e2d3ea6b4c75e0e13f2fbef3064825fbd47d136f --- /dev/null +++ b/QAs/45a964e8-4eb3-4b41-b9d3-72b949294e27-mcq.json @@ -0,0 +1,831 @@ +[ + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_12", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items were placed on the table when I first entered the room?", + "answer": "When you first entered the room, there were three objects placed on the table.", + "question_chinese": "在我刚进房间时桌子上摆有几个物件呢？", + "answer_chinese": "在我刚进房间时桌子上摆有几个物件呢？", + "evidence": [ + { + "description": "There were three objects on the table when you first entered the door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:20:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There were two objects on the table.", + "B": "There were four objects on the table.", + "C": "When you first entered the room, there were three objects placed on the table.", + "D": "When you first entered the room, there was one object placed on the table.", + "E": "When you first entered the room, there were five objects placed on the table." + }, + "answer_label": "C", + "answer_text": "When you first entered the room, there were three objects placed on the table.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_12", + "correct_answer": "When you first entered the room, there were three objects placed on the table.", + "original_options": { + "A": "There were two objects on the table.", + "B": "There were four objects on the table.", + "C": "When you first entered the room, there were three objects placed on the table.", + "D": "When you first entered the room, there was one object placed on the table.", + "E": "When you first entered the room, there were five objects placed on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 14 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_11", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items were placed on the table when I first entered the room?", + "answer": "When you first entered the room, there were three objects placed on the table.", + "question_chinese": "在我刚进房间时桌子上摆有几个物件呢？", + "answer_chinese": "在我刚进房间时桌子上摆有几个物件呢？", + "evidence": [ + { + "description": "There were three objects on the table when you first entered the door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:36:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There were two objects on the table.", + "B": "There were four objects on the table.", + "C": "When you first entered the room, there were six objects placed on the table.", + "D": "When you first entered the room, there were seven objects placed on the table.", + "E": "When you first entered the room, there were three objects placed on the table." + }, + "answer_label": "E", + "answer_text": "When you first entered the room, there were three objects placed on the table.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_11", + "correct_answer": "When you first entered the room, there were three objects placed on the table.", + "original_options": { + "A": "There were two objects on the table.", + "B": "There were four objects on the table.", + "C": "When you first entered the room, there were six objects placed on the table.", + "D": "When you first entered the room, there were seven objects placed on the table.", + "E": "When you first entered the room, there were three objects placed on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 14 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I take out of my bag before starting to draw?", + "answer": "You took out a pencil and a pencil sharpener from your bag before starting to draw.", + "question_chinese": "开始画图前我从包里拿了什么？", + "answer_chinese": "你在开始画图前从包里拿出了一支铅笔和削笔刀", + "evidence": [ + { + "description": "You take a pencil out of your bag.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:36" + } + }, + { + "description": "You took out a pencil sharpener from your bag.", + "timestamps": { + "start": "00:01:36", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:09:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took out a sketchbook and markers.", + "B": "You took out a mechanical pencil and eraser.", + "C": "You took out a pencil and a pencil sharpener from your bag before starting to draw.", + "D": "You pulled out a pencil and a small sketchpad from your bag before you started drawing.", + "E": "You took out a charcoal stick and a blending stump from your bag, then began to draw." + }, + "answer_label": "C", + "answer_text": "You took out a pencil and a pencil sharpener from your bag before starting to draw.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_1", + "correct_answer": "You took out a pencil and a pencil sharpener from your bag before starting to draw.", + "original_options": { + "A": "You took out a sketchbook and markers.", + "B": "You took out a mechanical pencil and eraser.", + "C": "You took out a pencil and a pencil sharpener from your bag before starting to draw.", + "D": "You pulled out a pencil and a small sketchpad from your bag before you started drawing.", + "E": "You took out a charcoal stick and a blending stump from your bag, then began to draw." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 17 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I take out of my bag before starting to draw?", + "answer": "You took out a pencil and a pencil sharpener from your bag before starting to draw.", + "question_chinese": "开始画图前我从包里拿了什么？", + "answer_chinese": "你在开始画图前从包里拿出了一支铅笔和削笔刀", + "evidence": [ + { + "description": "You take a pencil out of your bag.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:36" + } + }, + { + "description": "You took out a pencil sharpener from your bag.", + "timestamps": { + "start": "00:01:36", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:18:39", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took out a charcoal stick and stump.", + "B": "You took out a mechanical pencil and eraser.", + "C": "You took out a graphite pencil and a kneaded eraser from your bag before starting to draw.", + "D": "You took out a pencil and a pencil sharpener from your bag before starting to draw.", + "E": "You pulled out an art pad and a brush from your bag, then got ready to draw." + }, + "answer_label": "D", + "answer_text": "You took out a pencil and a pencil sharpener from your bag before starting to draw.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_2", + "correct_answer": "You took out a pencil and a pencil sharpener from your bag before starting to draw.", + "original_options": { + "A": "You took out a charcoal stick and stump.", + "B": "You took out a mechanical pencil and eraser.", + "C": "You took out a graphite pencil and a kneaded eraser from your bag before starting to draw.", + "D": "You took out a pencil and a pencil sharpener from your bag before starting to draw.", + "E": "You pulled out an art pad and a brush from your bag, then got ready to draw." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 17 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many line drawing pens did I just use?", + "answer": "You just used two liner pens.", + "question_chinese": "我刚刚用过多少只勾线笔？", + "answer_chinese": "你刚刚用过两只勾线笔", + "evidence": [ + { + "description": "You have used up a liner pen.", + "timestamps": { + "start": "00:22:03", + "end": "00:26:48" + } + }, + { + "description": "You have used the second liner pen.", + "timestamps": { + "start": "00:26:48", + "end": "00:29:55" + } + } + ], + "question_timestamps": "00:29:55", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You used one liner pen.", + "B": "You used three liner pens.", + "C": "You just used four liner pens to do the line drawing work.", + "D": "You just used two liner pens.", + "E": "You just used five liner pens while outlining the drawing you were making." + }, + "answer_label": "D", + "answer_text": "You just used two liner pens.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_9", + "correct_answer": "You just used two liner pens.", + "original_options": { + "A": "You used one liner pen.", + "B": "You used three liner pens.", + "C": "You just used four liner pens to do the line drawing work.", + "D": "You just used two liner pens.", + "E": "You just used five liner pens while outlining the drawing you were making." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_10", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many line drawing pens did I just use?", + "answer": "You just used three liner pens.", + "question_chinese": "我刚刚用过多少只勾线笔？", + "answer_chinese": "你刚刚用过三只勾线笔", + "evidence": [ + { + "description": "You have used the thrid liner pen.", + "timestamps": { + "start": "00:29:55", + "end": "00:30:55" + } + }, + { + "description": "You have used up a liner pen.", + "timestamps": { + "start": "00:22:03", + "end": "00:26:48" + } + }, + { + "description": "You have used the second liner \npen.", + "timestamps": { + "start": "00:26:48", + "end": "00:29:55" + } + } + ], + "question_timestamps": "00:30:57", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used one liner pen.", + "B": "You just used three liner pens.", + "C": "You used four liner pens.", + "D": "You just used five liner pens to complete the line drawing portion.", + "E": "You just used six liner pens while making the outlines for your drawing." + }, + "answer_label": "B", + "answer_text": "You just used three liner pens.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_10", + "correct_answer": "You just used three liner pens.", + "original_options": { + "A": "You used one liner pen.", + "B": "You just used three liner pens.", + "C": "You used four liner pens.", + "D": "You just used five liner pens to complete the line drawing portion.", + "E": "You just used six liner pens while making the outlines for your drawing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_7", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many pens did I just take out of the toolkit?", + "answer": "You just took out three pens from the toolkit.", + "question_chinese": "我刚才从工具包里拿出了多少支钢笔？", + "answer_chinese": "你刚从工具包里拿出了三支钢笔。", + "evidence": [ + { + "description": "You take out 3 pens from the toolkit.", + "timestamps": { + "start": "00:21:44", + "end": "00:22:06" + } + } + ], + "question_timestamps": "00:29:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You took out two pens.", + "B": "You just took out three pens from the toolkit.", + "C": "You took out four pens.", + "D": "You just took out one pen from the toolkit before continuing your work.", + "E": "You just took out five pens from the toolkit and set them aside to use." + }, + "answer_label": "B", + "answer_text": "You just took out three pens from the toolkit.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_7", + "correct_answer": "You just took out three pens from the toolkit.", + "original_options": { + "A": "You took out two pens.", + "B": "You just took out three pens from the toolkit.", + "C": "You took out four pens.", + "D": "You just took out one pen from the toolkit before continuing your work.", + "E": "You just took out five pens from the toolkit and set them aside to use." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 15 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_8", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many pens did I just take out of the toolkit?", + "answer": "You just took out three pens from the toolkit.", + "question_chinese": "我刚才从工具包里拿出了多少支钢笔？", + "answer_chinese": "你刚从工具包里拿出了三支钢笔。", + "evidence": [ + { + "description": "You take out 3 pens from the toolkit.", + "timestamps": { + "start": "00:21:44", + "end": "00:22:06" + } + } + ], + "question_timestamps": "00:36:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You took out two pens.", + "B": "You took out five pens.", + "C": "You just took out three pens from the toolkit.", + "D": "You just took out one pen from the toolkit and kept the rest inside.", + "E": "You just took out four pens from the toolkit and placed them together to use." + }, + "answer_label": "C", + "answer_text": "You just took out three pens from the toolkit.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_8", + "correct_answer": "You just took out three pens from the toolkit.", + "original_options": { + "A": "You took out two pens.", + "B": "You took out five pens.", + "C": "You just took out three pens from the toolkit.", + "D": "You just took out one pen from the toolkit and kept the rest inside.", + "E": "You just took out four pens from the toolkit and placed them together to use." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_6", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach out and get the painting kit?", + "answer": "You can directly reach out and get the painting kit.", + "question_chinese": "我可以直接获取绘画工具包吗？", + "answer_chinese": "你可以直接去拿绘画工具包。", + "evidence": [ + { + "description": "You can directly reach out and get the painting kit.", + "timestamps": { + "start": "00:01:53", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:09:28", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can directly reach out and get the painting kit.", + "B": "You cannot directly reach out and get the painting kit." + }, + "answer_label": "A", + "answer_text": "You can directly reach out and get the painting kit.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_6", + "correct_answer": "You can directly reach out and get the painting kit.", + "original_options": { + "A": "You can directly reach out and get the painting kit.", + "B": "You cannot directly reach out and get the painting kit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_5", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach out and get the painting kit?", + "answer": "You can directly reach out and get the painting kit.", + "question_chinese": "我可以直接获取绘画工具包吗？", + "answer_chinese": "你可以直接去拿绘画工具包。", + "evidence": [ + { + "description": "You can directly reach out and get the painting kit.", + "timestamps": { + "start": "00:01:53", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:18:22", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can directly reach out and get the painting kit.", + "B": "You cannot directly reach out to get the painting kit." + }, + "answer_label": "A", + "answer_text": "You can directly reach out and get the painting kit.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_5", + "correct_answer": "You can directly reach out and get the painting kit.", + "original_options": { + "A": "You can directly reach out and get the painting kit.", + "B": "You cannot directly reach out to get the painting kit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_14", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the window?", + "answer": "You need to turn around first, then walk straight to the window.", + "question_chinese": "我该怎么走到窗户边？", + "answer_chinese": "你需要先向后转，然后直走到窗户边", + "evidence": [ + { + "description": "The position of the window.", + "timestamps": { + "start": "00:35:10", + "end": "00:35:12" + } + } + ], + "question_timestamps": "00:36:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left, then walk to the window.", + "B": "Turn right, then walk to the window.", + "C": "Turn around first, then walk straight past the window toward the far wall.", + "D": "Walk straight first, then turn around near the window to reach it.", + "E": "You need to turn around first, then walk straight to the window." + }, + "answer_label": "E", + "answer_text": "You need to turn around first, then walk straight to the window.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_14", + "correct_answer": "You need to turn around first, then walk straight to the window.", + "original_options": { + "A": "Turn left, then walk to the window.", + "B": "Turn right, then walk to the window.", + "C": "Turn around first, then walk straight past the window toward the far wall.", + "D": "Walk straight first, then turn around near the window to reach it.", + "E": "You need to turn around first, then walk straight to the window." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 12 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_13", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the window?", + "answer": "You need to turn around first, then walk straight to the window.", + "question_chinese": "我该怎么走到窗户边？", + "answer_chinese": "你需要先向后转，然后直走到窗户边", + "evidence": [ + { + "description": "The position of the window.", + "timestamps": { + "start": "00:35:10", + "end": "00:35:12" + } + } + ], + "question_timestamps": "00:36:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left, then walk to the window.", + "B": "Turn right, then walk to the window.", + "C": "Turn around first, then walk straight to the door instead of the window.", + "D": "You need to turn around first, then walk straight to the window.", + "E": "Turn around and keep walking straight until you reach the corner, not the window." + }, + "answer_label": "D", + "answer_text": "You need to turn around first, then walk straight to the window.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_13", + "correct_answer": "You need to turn around first, then walk straight to the window.", + "original_options": { + "A": "Turn left, then walk to the window.", + "B": "Turn right, then walk to the window.", + "C": "Turn around first, then walk straight to the door instead of the window.", + "D": "You need to turn around first, then walk straight to the window.", + "E": "Turn around and keep walking straight until you reach the corner, not the window." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 14 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_3", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I erased the handwriting so far?", + "answer": "You have erased the handwriting six times so far.", + "question_chinese": "目前为止，我擦除了几次笔迹？", + "answer_chinese": "到目前为止你擦除了六次笔迹", + "evidence": [ + { + "description": "You erased the handwriting for the first time.", + "timestamps": { + "start": "00:11:30", + "end": "00:11:39" + } + }, + { + "description": "You erased the second handwriting.", + "timestamps": { + "start": "00:13:39", + "end": "00:13:44" + } + }, + { + "description": "You erased the third handwriting.", + "timestamps": { + "start": "00:14:58", + "end": "00:15:04" + } + }, + { + "description": "You erase the fourth handwriting.", + "timestamps": { + "start": "00:16:01", + "end": "00:16:08" + } + }, + { + "description": "You erased the fifth and sixth handwriting.", + "timestamps": { + "start": "00:16:50", + "end": "00:17:25" + } + } + ], + "question_timestamps": "00:18:02", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five times so far.", + "B": "Seven times so far.", + "C": "I have erased the handwriting four times up to now during this writing session.", + "D": "I have erased the handwriting nine times up to now during this writing session.", + "E": "You have erased the handwriting six times so far." + }, + "answer_label": "E", + "answer_text": "You have erased the handwriting six times so far.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_3", + "correct_answer": "You have erased the handwriting six times so far.", + "original_options": { + "A": "Five times so far.", + "B": "Seven times so far.", + "C": "I have erased the handwriting four times up to now during this writing session.", + "D": "I have erased the handwriting nine times up to now during this writing session.", + "E": "You have erased the handwriting six times so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "45a964e8-4eb3-4b41-b9d3-72b949294e27", + "video_length": "00:36:15", + "qid": "45a964e8-4eb3-4b41-b9d3-72b949294e27_4", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I erased the handwriting so far?", + "answer": "You have erased the handwriting eight times so far.", + "question_chinese": "目前为止，我擦除了几次笔迹？", + "answer_chinese": "到目前为止你擦除了八次笔迹", + "evidence": [ + { + "description": "The seventh time.", + "timestamps": { + "start": "00:19:19", + "end": "00:19:21" + } + }, + { + "description": "The eighth time.", + "timestamps": { + "start": "00:20:24", + "end": "00:20:27" + } + }, + { + "description": "You erased the fifth and sixth handwriting.", + "timestamps": { + "start": "00:16:50", + "end": "00:17:25" + } + }, + { + "description": "You erase the fourth handwriting.", + "timestamps": { + "start": "00:16:01", + "end": "00:16:08" + } + }, + { + "description": "You erased the third handwriting.", + "timestamps": { + "start": "00:14:58", + "end": "00:15:04" + } + }, + { + "description": "You erased the second handwriting.", + "timestamps": { + "start": "00:13:39", + "end": "00:13:44" + } + }, + { + "description": "You erased the handwriting for the first time.", + "timestamps": { + "start": "00:11:30", + "end": "00:11:39" + } + } + ], + "question_timestamps": "00:20:32", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have erased the handwriting eight times so far.", + "B": "Seven times so far.", + "C": "Nine times so far.", + "D": "I have erased the handwriting six times up to now during this writing session.", + "E": "I have erased the handwriting ten times up to now during this writing session." + }, + "answer_label": "A", + "answer_text": "You have erased the handwriting eight times so far.", + "q_id": "45a964e8-4eb3-4b41-b9d3-72b949294e27_4", + "correct_answer": "You have erased the handwriting eight times so far.", + "original_options": { + "A": "You have erased the handwriting eight times so far.", + "B": "Seven times so far.", + "C": "Nine times so far.", + "D": "I have erased the handwriting six times up to now during this writing session.", + "E": "I have erased the handwriting ten times up to now during this writing session." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + } +] diff --git a/QAs/48d07017-2af5-41c3-83a7-814721012f94-mcq.json b/QAs/48d07017-2af5-41c3-83a7-814721012f94-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1a32a0609714f284b1f9e5725a0451b6f35826f5 --- /dev/null +++ b/QAs/48d07017-2af5-41c3-83a7-814721012f94-mcq.json @@ -0,0 +1,1522 @@ +[ + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_4_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How do I walk to the TV?", + "answer": "You can go straight, then turn left; the TV is behind you.", + "question_chinese": "我如何走到电视机旁？", + "answer_chinese": "你可以直走，再左转，电视机在你后面。", + "evidence": [ + { + "description": "You can go straight and turn left.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:34" + } + } + ], + "question_timestamps": "00:08:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can go straight, then turn left; the TV is behind you.", + "B": "Go straight, then turn right.", + "C": "Go straight; the TV is ahead.", + "D": "Go straight, then turn left; the TV will be in front of you.", + "E": "Go straight, then turn left; the TV should end up on your right side." + }, + "answer_label": "A", + "answer_text": "You can go straight, then turn left; the TV is behind you.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_4_T1", + "correct_answer": "You can go straight, then turn left; the TV is behind you.", + "original_options": { + "A": "You can go straight, then turn left; the TV is behind you.", + "B": "Go straight, then turn right.", + "C": "Go straight; the TV is ahead.", + "D": "Go straight, then turn left; the TV will be in front of you.", + "E": "Go straight, then turn left; the TV should end up on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 14 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_7", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How do I get to the TV?", + "answer": "You can go straight.", + "question_chinese": "我如何走到电视机旁", + "answer_chinese": "你可以直走", + "evidence": [ + { + "description": "You can go straight.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:34" + } + } + ], + "question_timestamps": "00:09:36", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left to reach it.", + "B": "Back up first.", + "C": "Walk past the chair, then angle left slightly to get in front of the TV.", + "D": "You can go straight.", + "E": "Head backward a bit, then turn left and continue until you are at the TV." + }, + "answer_label": "D", + "answer_text": "You can go straight.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_7", + "correct_answer": "You can go straight.", + "original_options": { + "A": "Turn left to reach it.", + "B": "Back up first.", + "C": "Walk past the chair, then angle left slightly to get in front of the TV.", + "D": "You can go straight.", + "E": "Head backward a bit, then turn left and continue until you are at the TV." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 15, + 15 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_4_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How do I walk to the TV?", + "answer": "You can go straight; the TV is in front of you.", + "question_chinese": "我如何走到电视机旁？", + "answer_chinese": "你可以直走，电视机在你前面。", + "evidence": [ + { + "description": "You can go straight.", + "timestamps": { + "start": "00:12:10", + "end": "00:12:14" + } + } + ], + "question_timestamps": "00:12:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go straight; it's behind you.", + "B": "Go straight; it's to your left.", + "C": "Go straight; the TV will end up on your right side as you approach it.", + "D": "You can go straight; the TV is in front of you.", + "E": "Go straight; the TV is behind you, so you need to turn around after walking." + }, + "answer_label": "D", + "answer_text": "You can go straight; the TV is in front of you.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_4_T2", + "correct_answer": "You can go straight; the TV is in front of you.", + "original_options": { + "A": "Go straight; it's behind you.", + "B": "Go straight; it's to your left.", + "C": "Go straight; the TV will end up on your right side as you approach it.", + "D": "You can go straight; the TV is in front of you.", + "E": "Go straight; the TV is behind you, so you need to turn around after walking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_8", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How do I get to the TV?", + "answer": "You can go straight.", + "question_chinese": "我如何走到电视机旁", + "answer_chinese": "你可以直走", + "evidence": [ + { + "description": "You can go straight.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:34" + } + }, + { + "description": "You can go straight.", + "timestamps": { + "start": "00:12:10", + "end": "00:12:14" + } + } + ], + "question_timestamps": "00:19:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left.", + "B": "You can go straight.", + "C": "Turn right.", + "D": "Go back a few steps, then turn left to reach the TV.", + "E": "Walk past the TV first, then turn around and come back to it." + }, + "answer_label": "B", + "answer_text": "You can go straight.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_8", + "correct_answer": "You can go straight.", + "original_options": { + "A": "Turn left.", + "B": "You can go straight.", + "C": "Turn right.", + "D": "Go back a few steps, then turn left to reach the TV.", + "E": "Walk past the TV first, then turn around and come back to it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_3_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I do to the coffee?", + "answer": "You kept stirring it, and the coffee is in front of you.", + "question_chinese": "我对咖啡做了什么？", + "answer_chinese": "你持续搅拌它，咖啡在你前面。", + "evidence": [ + { + "description": "You keep stirring it.", + "timestamps": { + "start": "00:04:50", + "end": "00:05:45" + } + } + ], + "question_timestamps": "00:05:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You kept stirring it behind you.", + "B": "You kept stirring it to your left.", + "C": "You kept stirring it, then left the coffee to your right instead of in front of you.", + "D": "You kept stirring it continuously, but the coffee ended up behind you rather than in front.", + "E": "You kept stirring it, and the coffee is in front of you." + }, + "answer_label": "E", + "answer_text": "You kept stirring it, and the coffee is in front of you.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_3_T1", + "correct_answer": "You kept stirring it, and the coffee is in front of you.", + "original_options": { + "A": "You kept stirring it behind you.", + "B": "You kept stirring it to your left.", + "C": "You kept stirring it, then left the coffee to your right instead of in front of you.", + "D": "You kept stirring it continuously, but the coffee ended up behind you rather than in front.", + "E": "You kept stirring it, and the coffee is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 16 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_3_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I do to the coffee?", + "answer": "You put it into the refrigerator, and the refrigerator is behind you.", + "question_chinese": "我对咖啡做了什么？", + "answer_chinese": "你把它放入了冰箱，冰箱在你后面。", + "evidence": [ + { + "description": "You put it in the refrigerator.", + "timestamps": { + "start": "00:08:18", + "end": "00:08:28" + } + } + ], + "question_timestamps": "00:08:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put it on the counter.", + "B": "You poured it into a mug.", + "C": "You put it in the fridge to your left, not behind you at all.", + "D": "You put it into the refrigerator, but the refrigerator is to your right, not behind you.", + "E": "You put it into the refrigerator, and the refrigerator is behind you." + }, + "answer_label": "E", + "answer_text": "You put it into the refrigerator, and the refrigerator is behind you.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_3_T2", + "correct_answer": "You put it into the refrigerator, and the refrigerator is behind you.", + "original_options": { + "A": "You put it on the counter.", + "B": "You poured it into a mug.", + "C": "You put it in the fridge to your left, not behind you at all.", + "D": "You put it into the refrigerator, but the refrigerator is to your right, not behind you.", + "E": "You put it into the refrigerator, and the refrigerator is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 16 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_5", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I do to the coffee?", + "answer": "You kept stirring it and finally put it in the refrigerator.", + "question_chinese": "我对咖啡做了什么", + "answer_chinese": "你持续搅拌它最终放入了冰箱", + "evidence": [ + { + "description": "You keep stirring it.", + "timestamps": { + "start": "00:04:50", + "end": "00:08:05" + } + }, + { + "description": "You put it in the refrigerator.", + "timestamps": { + "start": "00:08:15", + "end": "00:08:30" + } + } + ], + "question_timestamps": "00:09:36", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You stopped stirring and set it down.", + "B": "You kept stirring and spilled it.", + "C": "You kept stirring it and finally put it in the refrigerator.", + "D": "You kept stirring it for a while, then poured it out into the sink instead of refrigerating it.", + "E": "You stirred it briefly and ended by leaving the coffee on the counter rather than in the refrigerator." + }, + "answer_label": "C", + "answer_text": "You kept stirring it and finally put it in the refrigerator.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_5", + "correct_answer": "You kept stirring it and finally put it in the refrigerator.", + "original_options": { + "A": "You stopped stirring and set it down.", + "B": "You kept stirring and spilled it.", + "C": "You kept stirring it and finally put it in the refrigerator.", + "D": "You kept stirring it for a while, then poured it out into the sink instead of refrigerating it.", + "E": "You stirred it briefly and ended by leaving the coffee on the counter rather than in the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 18 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I do to the coffee?", + "answer": "You kept stirring it and finally put it in the refrigerator.", + "question_chinese": "我对咖啡做了什么", + "answer_chinese": "你持续搅拌它最终放入了冰箱", + "evidence": [ + { + "description": "You keep stirring it.", + "timestamps": { + "start": "00:04:50", + "end": "00:08:05" + } + }, + { + "description": "You put it in the refrigerator.", + "timestamps": { + "start": "00:08:15", + "end": "00:08:30" + } + } + ], + "question_timestamps": "00:19:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You poured it down the sink.", + "B": "You left it on the counter.", + "C": "You kept stirring it, then set the coffee aside on the counter instead of putting it in the refrigerator.", + "D": "You poured the coffee into a mug after stirring, and then left it out rather than refrigerating it.", + "E": "You kept stirring it and finally put it in the refrigerator." + }, + "answer_label": "E", + "answer_text": "You kept stirring it and finally put it in the refrigerator.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_6", + "correct_answer": "You kept stirring it and finally put it in the refrigerator.", + "original_options": { + "A": "You poured it down the sink.", + "B": "You left it on the counter.", + "C": "You kept stirring it, then set the coffee aside on the counter instead of putting it in the refrigerator.", + "D": "You poured the coffee into a mug after stirring, and then left it out rather than refrigerating it.", + "E": "You kept stirring it and finally put it in the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_7_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I do to the vegetable leaves?", + "answer": "You put it on the lid, and the lid is in front of you.", + "question_chinese": "我对菜叶做了什么？", + "answer_chinese": "你把它放在盖子上了，盖子在你前面。", + "evidence": [ + { + "description": "you put it on the lid.", + "timestamps": { + "start": "00:19:20", + "end": "00:20:25" + } + } + ], + "question_timestamps": "00:20:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put it on the lid behind you.", + "B": "You put it on the lid left.", + "C": "You placed the vegetable leaves on the lid, but the lid is to your right, not in front.", + "D": "You set the vegetable leaves on the lid, and the lid ended up behind you rather than in front.", + "E": "You put it on the lid, and the lid is in front of you." + }, + "answer_label": "E", + "answer_text": "You put it on the lid, and the lid is in front of you.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_7_T1", + "correct_answer": "You put it on the lid, and the lid is in front of you.", + "original_options": { + "A": "You put it on the lid behind you.", + "B": "You put it on the lid left.", + "C": "You placed the vegetable leaves on the lid, but the lid is to your right, not in front.", + "D": "You set the vegetable leaves on the lid, and the lid ended up behind you rather than in front.", + "E": "You put it on the lid, and the lid is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 19 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_7_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I do to the vegetable leaves?", + "answer": "You wrapped it up and put it into the refrigerator, and the refrigerator is behind you.", + "question_chinese": "我对菜叶做了什么？", + "answer_chinese": "你把它包装起来放入了冰箱，冰箱在你后面。", + "evidence": [ + { + "description": "You put it in the refrigerator.", + "timestamps": { + "start": "00:23:35", + "end": "00:23:43" + } + } + ], + "question_timestamps": "00:23:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You wrapped it and put it in front.", + "B": "You wrapped it and put it left.", + "C": "You wrapped it up and put it into the refrigerator, and the refrigerator is behind you.", + "D": "You wrapped it up and put it into the refrigerator, but the refrigerator is to your right, not behind you.", + "E": "You wrapped it up and stored it in the refrigerator, and the refrigerator is in front of you rather than behind." + }, + "answer_label": "C", + "answer_text": "You wrapped it up and put it into the refrigerator, and the refrigerator is behind you.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_7_T2", + "correct_answer": "You wrapped it up and put it into the refrigerator, and the refrigerator is behind you.", + "original_options": { + "A": "You wrapped it and put it in front.", + "B": "You wrapped it and put it left.", + "C": "You wrapped it up and put it into the refrigerator, and the refrigerator is behind you.", + "D": "You wrapped it up and put it into the refrigerator, but the refrigerator is to your right, not behind you.", + "E": "You wrapped it up and stored it in the refrigerator, and the refrigerator is in front of you rather than behind." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 20, + 21 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_14", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I do to the vegetable leaves?", + "answer": "You wrapped it up and put it in the refrigerator.", + "question_chinese": "我对菜叶做了什么", + "answer_chinese": "你把它包装起来放入了冰箱", + "evidence": [ + { + "description": "You wrapped it up.", + "timestamps": { + "start": "00:19:12", + "end": "00:23:30" + } + }, + { + "description": "You put it in the refrigerator.", + "timestamps": { + "start": "00:23:35", + "end": "00:23:43" + } + } + ], + "question_timestamps": "00:28:48", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You wrapped it and left it out.", + "B": "You rinsed it and set it aside.", + "C": "You wrapped it up, but then left the vegetable leaves on the counter instead of refrigerating them.", + "D": "You bundled the vegetable leaves together and stored them in the pantry rather than putting them in the refrigerator.", + "E": "You wrapped it up and put it in the refrigerator." + }, + "answer_label": "E", + "answer_text": "You wrapped it up and put it in the refrigerator.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_14", + "correct_answer": "You wrapped it up and put it in the refrigerator.", + "original_options": { + "A": "You wrapped it and left it out.", + "B": "You rinsed it and set it aside.", + "C": "You wrapped it up, but then left the vegetable leaves on the counter instead of refrigerating them.", + "D": "You bundled the vegetable leaves together and stored them in the pantry rather than putting them in the refrigerator.", + "E": "You wrapped it up and put it in the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 19 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_13", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I do to the vegetable leaves?", + "answer": "You wrapped it up and put it in the refrigerator.", + "question_chinese": "我对菜叶做了什么", + "answer_chinese": "你把它包装起来放入了冰箱", + "evidence": [ + { + "description": "You wrapped it up.", + "timestamps": { + "start": "00:19:12", + "end": "00:23:30" + } + }, + { + "description": "You put it in the refrigerator.", + "timestamps": { + "start": "00:23:35", + "end": "00:23:43" + } + } + ], + "question_timestamps": "00:38:24", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You wrapped it up and put it in the refrigerator.", + "B": "You wrapped it and froze it.", + "C": "You left it on the counter.", + "D": "You wrapped the vegetable leaves up, but put them into the freezer instead of the refrigerator.", + "E": "You rinsed the vegetable leaves and left them on a drying rack, rather than wrapping them and refrigerating them." + }, + "answer_label": "A", + "answer_text": "You wrapped it up and put it in the refrigerator.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_13", + "correct_answer": "You wrapped it up and put it in the refrigerator.", + "original_options": { + "A": "You wrapped it up and put it in the refrigerator.", + "B": "You wrapped it and froze it.", + "C": "You left it on the counter.", + "D": "You wrapped the vegetable leaves up, but put them into the freezer instead of the refrigerator.", + "E": "You rinsed the vegetable leaves and left them on a drying rack, rather than wrapping them and refrigerating them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 19 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_4", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the cabinet?", + "answer": "You opened it three times.", + "question_chinese": "我打开了多少次柜子", + "answer_chinese": "你打开了三次", + "evidence": [ + { + "description": "You open it for the second time.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:28" + } + }, + { + "description": "You open it for the first time.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:20" + } + }, + { + "description": "You open it for the third time.", + "timestamps": { + "start": "00:01:30", + "end": "00:04:35" + } + } + ], + "question_timestamps": "00:09:36", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You opened it twice.", + "B": "You opened it four times.", + "C": "You opened the cabinet two times during the scene, closing it again after each use.", + "D": "You opened the cabinet four separate times overall, returning to it after doing something else.", + "E": "You opened it three times." + }, + "answer_label": "E", + "answer_text": "You opened it three times.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_4", + "correct_answer": "You opened it three times.", + "original_options": { + "A": "You opened it twice.", + "B": "You opened it four times.", + "C": "You opened the cabinet two times during the scene, closing it again after each use.", + "D": "You opened the cabinet four separate times overall, returning to it after doing something else.", + "E": "You opened it three times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 15 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_3", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the cabinet?", + "answer": "You opened the cabinet four times.", + "question_chinese": "我打开了多少次柜子", + "answer_chinese": "你打开了四次柜子", + "evidence": [ + { + "description": "You open it for the second time.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:28" + } + }, + { + "description": "You open it for the first time.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:20" + } + }, + { + "description": "You open it for the third time.", + "timestamps": { + "start": "00:01:30", + "end": "00:04:35" + } + }, + { + "description": "You open it for the fourth time.", + "timestamps": { + "start": "00:12:15", + "end": "00:12:22" + } + } + ], + "question_timestamps": "00:19:12", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You opened it three times.", + "B": "You opened the cabinet four times.", + "C": "You opened it five times.", + "D": "You opened the cabinet three times total, each time pulling it open and then shutting it.", + "E": "You opened the cabinet five times during the video, repeatedly going back to it for items." + }, + "answer_label": "B", + "answer_text": "You opened the cabinet four times.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_3", + "correct_answer": "You opened the cabinet four times.", + "original_options": { + "A": "You opened it three times.", + "B": "You opened the cabinet four times.", + "C": "You opened it five times.", + "D": "You opened the cabinet three times total, each time pulling it open and then shutting it.", + "E": "You opened the cabinet five times during the video, repeatedly going back to it for items." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_15", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do with my phone?", + "answer": "You didn't pick up the phone.", + "question_chinese": "我用手机干了什么", + "answer_chinese": "你没有拿起手机", + "evidence": [ + { + "description": "You didn't pick up the phone.", + "timestamps": { + "start": "00:19:12", + "end": "00:28:47" + } + } + ], + "question_timestamps": "00:28:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You picked up the phone.", + "B": "You didn't pick up the phone.", + "C": "You answered a call.", + "D": "You picked up the phone briefly, then set it back down without using it.", + "E": "You looked at your phone screen briefly, then put it away without picking it up." + }, + "answer_label": "B", + "answer_text": "You didn't pick up the phone.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_15", + "correct_answer": "You didn't pick up the phone.", + "original_options": { + "A": "You picked up the phone.", + "B": "You didn't pick up the phone.", + "C": "You answered a call.", + "D": "You picked up the phone briefly, then set it back down without using it.", + "E": "You looked at your phone screen briefly, then put it away without picking it up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 15 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_16", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do with my phone?", + "answer": "You checked the message on your phone, showed it to the man, and finally set an alarm.", + "question_chinese": "我用手机干了什么", + "answer_chinese": "你用手机看了消息并展示给了男人最后定了闹钟", + "evidence": [ + { + "description": "You didn't pick up the phone.", + "timestamps": { + "start": "00:19:12", + "end": "00:28:47" + } + }, + { + "description": "You checked the message on your phone and showed it to the man.", + "timestamps": { + "start": "00:29:25", + "end": "00:29:50" + } + }, + { + "description": "You set an alarm on your phone.", + "timestamps": { + "start": "00:37:40", + "end": "00:38:00" + } + } + ], + "question_timestamps": "00:38:24", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You checked the message on your phone, showed it to the man, and finally set an alarm.", + "B": "You made a call instead.", + "C": "You sent a text message.", + "D": "You checked your phone briefly, showed the screen to the man, and then turned the screen off.", + "E": "You read a notification on your phone, showed it to the man, and then made a call afterward." + }, + "answer_label": "A", + "answer_text": "You checked the message on your phone, showed it to the man, and finally set an alarm.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_16", + "correct_answer": "You checked the message on your phone, showed it to the man, and finally set an alarm.", + "original_options": { + "A": "You checked the message on your phone, showed it to the man, and finally set an alarm.", + "B": "You made a call instead.", + "C": "You sent a text message.", + "D": "You checked your phone briefly, showed the screen to the man, and then turned the screen off.", + "E": "You read a notification on your phone, showed it to the man, and then made a call afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_11", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the dishcloth?", + "answer": "You can.", + "question_chinese": "我能够到洗碗巾吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:23:50", + "end": "00:24:10" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:25:20", + "end": "00:25:45" + } + } + ], + "question_timestamps": "00:28:48", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you can’t—the dishwasher is to your right.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_11", + "correct_answer": "You can.", + "original_options": { + "A": "No, you can’t—the dishwasher is to your right.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_12", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the dishcloth?", + "answer": "You can.", + "question_chinese": "我能够到洗碗巾吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:23:50", + "end": "00:24:10" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:25:20", + "end": "00:25:45" + } + } + ], + "question_timestamps": "00:38:24", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "No, you can’t—the dishwasher is to your right." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_12", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "No, you can’t—the dishwasher is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_5_T1", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the kettle?", + "answer": "No, you can’t—the kettle is to your left.", + "question_chinese": "我能够到烧水壶吗？", + "answer_chinese": "你不可以，水壶在你左边。", + "evidence": [ + { + "description": "You can not.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:55" + } + } + ], + "question_timestamps": "00:06:57", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t—the kettle is to your left.", + "B": "Yes, you can—the kettle is in front of you." + }, + "answer_label": "A", + "answer_text": "No, you can’t—the kettle is to your left.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_5_T1", + "correct_answer": "No, you can’t—the kettle is to your left.", + "original_options": { + "A": "No, you can’t—the kettle is to your left.", + "B": "Yes, you can—the kettle is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_10", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the kettle?", + "answer": "You can.", + "question_chinese": "我能够到烧水壶吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:09:36", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "No, you can’t—the kettle is to your left." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_10", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "No, you can’t—the kettle is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_5_T2", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the kettle?", + "answer": "Yes, you can—the kettle is in front of you.", + "question_chinese": "我能够到烧水壶吗？", + "answer_chinese": "你可以，水壶在你前面。", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:14:02", + "end": "00:14:10" + } + } + ], + "question_timestamps": "00:14:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t—the kettle is behind you.", + "B": "Yes, you can—the kettle is in front of you." + }, + "answer_label": "B", + "answer_text": "Yes, you can—the kettle is in front of you.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_5_T2", + "correct_answer": "Yes, you can—the kettle is in front of you.", + "original_options": { + "A": "No, you can’t—the kettle is behind you.", + "B": "Yes, you can—the kettle is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_9", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the kettle?", + "answer": "You can.", + "question_chinese": "我能够到烧水壶吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:50" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:12:35", + "end": "00:12:45" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:14:02", + "end": "00:14:10" + } + } + ], + "question_timestamps": "00:19:12", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "No, you can’t—the kettle is out of reach." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_9", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "No, you can’t—the kettle is out of reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_1_T2", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the computer?", + "answer": "Yes, you can—the computer is in front of you.", + "question_chinese": "我能够到电脑吗？", + "answer_chinese": "你可以，电脑在你前面。", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t—the computer is to your right.", + "B": "Yes, you can—the computer is in front of you." + }, + "answer_label": "B", + "answer_text": "Yes, you can—the computer is in front of you.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_1_T2", + "correct_answer": "Yes, you can—the computer is in front of you.", + "original_options": { + "A": "No, you can’t—the computer is to your right.", + "B": "Yes, you can—the computer is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_1_T1", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the computer?", + "answer": "No, you can’t—the computer is behind you.", + "question_chinese": "我能够到电脑吗？", + "answer_chinese": "你不可以，电脑在你后面。", + "evidence": [ + { + "description": "You can not.", + "timestamps": { + "start": "00:01:09", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t—the computer is behind you.", + "B": "Yes, you can—the computer is right in front of you." + }, + "answer_label": "A", + "answer_text": "No, you can’t—the computer is behind you.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_1_T1", + "correct_answer": "No, you can’t—the computer is behind you.", + "original_options": { + "A": "No, you can’t—the computer is behind you.", + "B": "Yes, you can—the computer is right in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_1", + "qid_num": 27, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the computer?", + "answer": "You can.", + "question_chinese": "我能够到电脑吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:09:36", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You can’t." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_1", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You can’t." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_2", + "qid_num": 28, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the computer?", + "answer": "You can.", + "question_chinese": "我能够到电脑吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:12:14", + "end": "00:12:16" + } + } + ], + "question_timestamps": "00:19:12", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can’t.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_2", + "correct_answer": "You can.", + "original_options": { + "A": "You can’t.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_17", + "qid_num": 29, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I ask the man to do?", + "answer": "You ask him to pass the plate over.", + "question_chinese": "我让男人做了什么", + "answer_chinese": "你让他把盘子给过来", + "evidence": [ + { + "description": "You ask him to pass the plate over.", + "timestamps": { + "start": "00:25:10", + "end": "00:25:20" + } + } + ], + "question_timestamps": "00:28:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You ask him to pass the bowl over.", + "B": "You ask him to pass the cup back.", + "C": "You ask him to pass the plate over.", + "D": "You ask him to hand the notebook over instead of passing anything from the table.", + "E": "You ask him to bring the glass over to you rather than passing a plate across." + }, + "answer_label": "C", + "answer_text": "You ask him to pass the plate over.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_17", + "correct_answer": "You ask him to pass the plate over.", + "original_options": { + "A": "You ask him to pass the bowl over.", + "B": "You ask him to pass the cup back.", + "C": "You ask him to pass the plate over.", + "D": "You ask him to hand the notebook over instead of passing anything from the table.", + "E": "You ask him to bring the glass over to you rather than passing a plate across." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 16 + ] + }, + { + "video_uid": "48d07017-2af5-41c3-83a7-814721012f94", + "video_length": "00:38:24", + "qid": "48d07017-2af5-41c3-83a7-814721012f94_18", + "qid_num": 30, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I ask the man to do?", + "answer": "You ask him to look at the message on the phone.", + "question_chinese": "我让男人做了什么", + "answer_chinese": "你让他看手机上的消息", + "evidence": [ + { + "description": "You ask him to pass the plate over.", + "timestamps": { + "start": "00:25:10", + "end": "00:25:20" + } + }, + { + "description": "You ask him to look at the message on the phone.", + "timestamps": { + "start": "00:29:45", + "end": "00:29:51" + } + } + ], + "question_timestamps": "00:38:24", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You ask him to read the email.", + "B": "You ask him to check the battery.", + "C": "You ask him to glance at the document on the screen, not at anything on the phone.", + "D": "You ask him to focus on the document in the folder instead of checking a message.", + "E": "You ask him to look at the message on the phone." + }, + "answer_label": "E", + "answer_text": "You ask him to look at the message on the phone.", + "q_id": "48d07017-2af5-41c3-83a7-814721012f94_18", + "correct_answer": "You ask him to look at the message on the phone.", + "original_options": { + "A": "You ask him to read the email.", + "B": "You ask him to check the battery.", + "C": "You ask him to glance at the document on the screen, not at anything on the phone.", + "D": "You ask him to focus on the document in the folder instead of checking a message.", + "E": "You ask him to look at the message on the phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + } +] diff --git a/QAs/4bbc5b0a-5eb4-4973-a57a-45feceee9f39-mcq.json b/QAs/4bbc5b0a-5eb4-4973-a57a-45feceee9f39-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b8a60d2a63fff082c65cd353b692ad7c2716345e --- /dev/null +++ b/QAs/4bbc5b0a-5eb4-4973-a57a-45feceee9f39-mcq.json @@ -0,0 +1,941 @@ +[ + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_8_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I just walked past the stairs, where was the woodworking table relative to me?", + "answer": "When you went upstairs, the woodworking table was to your back-right.", + "question_chinese": "刚刚我经过楼梯的时候，木工桌在我的什么方位？", + "answer_chinese": "从楼梯上去的时候，木工桌在你的右后方。", + "evidence": [ + { + "description": "The woodworking table was to your left rear.", + "timestamps": { + "start": "00:21:02", + "end": "00:21:53" + } + } + ], + "question_timestamps": "00:21:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was to your back-left.", + "B": "When you went upstairs, the woodworking table was to your back-right.", + "C": "It was to your front-right.", + "D": "As you went upstairs past the stairs, the woodworking table was positioned off to your front-left.", + "E": "When you went downstairs past the stairs, the woodworking table ended up to your back-right behind you." + }, + "answer_label": "B", + "answer_text": "When you went upstairs, the woodworking table was to your back-right.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_8_T2", + "correct_answer": "When you went upstairs, the woodworking table was to your back-right.", + "original_options": { + "A": "It was to your back-left.", + "B": "When you went upstairs, the woodworking table was to your back-right.", + "C": "It was to your front-right.", + "D": "As you went upstairs past the stairs, the woodworking table was positioned off to your front-left.", + "E": "When you went downstairs past the stairs, the woodworking table ended up to your back-right behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_8_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I just walked past the stairs, where was the woodworking table relative to me?", + "answer": "When you just went upstairs, the woodworking table was to your back-left.", + "question_chinese": "刚刚我经过楼梯的时候，木工桌在我的什么方位？", + "answer_chinese": "你刚刚从楼梯上去的时候，木工桌在你的左后方。", + "evidence": [ + { + "description": "The woodworking table was to your left rear.", + "timestamps": { + "start": "00:21:02", + "end": "00:21:53" + } + } + ], + "question_timestamps": "00:32:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "When you just went upstairs, the woodworking table was to your back-left.", + "B": "It was to your back-right.", + "C": "It was to your front-left.", + "D": "When you just went upstairs, the woodworking table was positioned to your front-right in your field of view.", + "E": "When you went downstairs past the stairs, the woodworking table was to your back-left rather than behind you upstairs." + }, + "answer_label": "A", + "answer_text": "When you just went upstairs, the woodworking table was to your back-left.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_8_T1", + "correct_answer": "When you just went upstairs, the woodworking table was to your back-left.", + "original_options": { + "A": "When you just went upstairs, the woodworking table was to your back-left.", + "B": "It was to your back-right.", + "C": "It was to your front-left.", + "D": "When you just went upstairs, the woodworking table was positioned to your front-right in your field of view.", + "E": "When you went downstairs past the stairs, the woodworking table was to your back-left rather than behind you upstairs." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 19 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_7_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wooden boards did I carry to the woodworking table in total?", + "answer": "In total, you carried four wooden boards to the woodworking table in front of you.", + "question_chinese": "我一共拿了多少木板到木工桌上？", + "answer_chinese": "你一共拿了四块木板到你前方的木工桌上。", + "evidence": [ + { + "description": "You took two pieces of wood to the carpenter's workbench.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:29" + } + }, + { + "description": "You took two pieces of wood to the carpenter's workbench.", + "timestamps": { + "start": "00:21:45", + "end": "00:22:14" + } + } + ], + "question_timestamps": "00:26:38", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In total, you carried four wooden boards to the woodworking table in front of you.", + "B": "In total, you carried two boards.", + "C": "In total, you carried six boards.", + "D": "In total, you carried three wooden boards to the woodworking table in front of you.", + "E": "In total, you carried five wooden boards to the woodworking table in front of you." + }, + "answer_label": "A", + "answer_text": "In total, you carried four wooden boards to the woodworking table in front of you.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_7_T2", + "correct_answer": "In total, you carried four wooden boards to the woodworking table in front of you.", + "original_options": { + "A": "In total, you carried four wooden boards to the woodworking table in front of you.", + "B": "In total, you carried two boards.", + "C": "In total, you carried six boards.", + "D": "In total, you carried three wooden boards to the woodworking table in front of you.", + "E": "In total, you carried five wooden boards to the woodworking table in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_7_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wooden boards did I carry to the woodworking table in total?", + "answer": "In total, you carried eight wooden boards to the woodworking table to your back-left.", + "question_chinese": "我一共拿了多少木板到木工桌上？", + "answer_chinese": "你一共拿了八块木板到你左后方的木工桌上。", + "evidence": [ + { + "description": "You took two pieces of wood to the carpenter's workbench.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:29" + } + }, + { + "description": "You took two pieces of wood to the carpenter's workbench.", + "timestamps": { + "start": "00:21:45", + "end": "00:22:14" + } + }, + { + "description": "You took two pieces of wood to the carpenter's workbench.", + "timestamps": { + "start": "00:32:55", + "end": "00:33:17" + } + }, + { + "description": "You took two pieces of wood to the carpenter's workbench.", + "timestamps": { + "start": "00:46:10", + "end": "00:46:52" + } + } + ], + "question_timestamps": "00:52:41", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In total, you carried six boards.", + "B": "In total, you carried ten boards.", + "C": "In total, you carried seven wooden boards to the woodworking table to your back-left.", + "D": "In total, you carried nine wooden boards to the woodworking table to your back-left.", + "E": "In total, you carried eight wooden boards to the woodworking table to your back-left." + }, + "answer_label": "E", + "answer_text": "In total, you carried eight wooden boards to the woodworking table to your back-left.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_7_T1", + "correct_answer": "In total, you carried eight wooden boards to the woodworking table to your back-left.", + "original_options": { + "A": "In total, you carried six boards.", + "B": "In total, you carried ten boards.", + "C": "In total, you carried seven wooden boards to the woodworking table to your back-left.", + "D": "In total, you carried nine wooden boards to the woodworking table to your back-left.", + "E": "In total, you carried eight wooden boards to the woodworking table to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_5_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tape measures do I have in total?", + "answer": "You have three tape measure in total: two are on the saw table to your back-left, and one is on the upstairs platform to your back-right.", + "question_chinese": "我一共有几个卷尺？", + "answer_chinese": "你一共有三个卷尺，两个在你的左后方的锯子台上，一个在你右后方楼上的台子上。", + "evidence": [ + { + "description": "There are two tape measures beside the electric saw.", + "timestamps": { + "start": "00:44:24", + "end": "00:44:27" + } + }, + { + "description": "There is a tape measure on the carpenter's workbench.", + "timestamps": { + "start": "00:43:34", + "end": "00:43:40" + } + } + ], + "question_timestamps": "00:46:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have two tape measures.", + "B": "You have five tape measures.", + "C": "You have three tape measure in total: two are on the saw table to your back-left, and one is on the upstairs platform to your back-right.", + "D": "You have four tape measure: three on the saw table, and one on the upstairs platform.", + "E": "You have one tape measure: it is on the upstairs platform to your back-right." + }, + "answer_label": "C", + "answer_text": "You have three tape measure in total: two are on the saw table to your back-left, and one is on the upstairs platform to your back-right.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_5_T1", + "correct_answer": "You have three tape measure in total: two are on the saw table to your back-left, and one is on the upstairs platform to your back-right.", + "original_options": { + "A": "You have two tape measures.", + "B": "You have five tape measures.", + "C": "You have three tape measure in total: two are on the saw table to your back-left, and one is on the upstairs platform to your back-right.", + "D": "You have four tape measure: three on the saw table, and one on the upstairs platform.", + "E": "You have one tape measure: it is on the upstairs platform to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 14 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_5_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tape measures do I have in total?", + "answer": "You have three tape measure in total: one is on the platform to your right, and two are on the saw table to your back-left downstairs.", + "question_chinese": "我一共有几个卷尺？", + "answer_chinese": "你一共有三个卷尺，一个在你右边的台子上，两个在楼下的左后方的锯台上。", + "evidence": [ + { + "description": "There are two tape measures beside the electric saw.", + "timestamps": { + "start": "00:44:24", + "end": "00:44:27" + } + }, + { + "description": "There is a tape measure on the carpenter's workbench.", + "timestamps": { + "start": "00:43:34", + "end": "00:43:40" + } + } + ], + "question_timestamps": "00:47:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have two tape measures.", + "B": "You have five tape measures.", + "C": "You have four tape measure: two on the platform, and two on the saw table downstairs.", + "D": "You have three tape measure in total: one is on the platform to your right, and two are on the saw table to your back-left downstairs.", + "E": "You have one tape measure: it is on the platform to your right." + }, + "answer_label": "D", + "answer_text": "You have three tape measure in total: one is on the platform to your right, and two are on the saw table to your back-left downstairs.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_5_T2", + "correct_answer": "You have three tape measure in total: one is on the platform to your right, and two are on the saw table to your back-left downstairs.", + "original_options": { + "A": "You have two tape measures.", + "B": "You have five tape measures.", + "C": "You have four tape measure: two on the platform, and two on the saw table downstairs.", + "D": "You have three tape measure in total: one is on the platform to your right, and two are on the saw table to your back-left downstairs.", + "E": "You have one tape measure: it is on the platform to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 13 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_4_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the set square from here?", + "answer": "Yes, the set square is behind you.", + "question_chinese": "我在这里能拿到三角板吗？", + "answer_chinese": "可以，三角板在你后面。", + "evidence": [ + { + "description": "The triangular ruler is on the woodworking table.", + "timestamps": { + "start": "00:38:50", + "end": "00:38:55" + } + } + ], + "question_timestamps": "00:39:38", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, the set square is behind you.", + "B": "No, the set square is in front of you." + }, + "answer_label": "A", + "answer_text": "Yes, the set square is behind you.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_4_T1", + "correct_answer": "Yes, the set square is behind you.", + "original_options": { + "A": "Yes, the set square is behind you.", + "B": "No, the set square is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_4_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the set square from here?", + "answer": "No, the set square is on the table to your back-right.", + "question_chinese": "我在这里能拿到三角板吗？", + "answer_chinese": "不可以，三角板在你右后方的桌子上。", + "evidence": [ + { + "description": "The triangular ruler is on the woodworking table.", + "timestamps": { + "start": "00:40:55", + "end": "00:40:57" + } + } + ], + "question_timestamps": "00:41:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, the set square is behind you.", + "B": "No, the set square is on the table to your back-right." + }, + "answer_label": "B", + "answer_text": "No, the set square is on the table to your back-right.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_4_T2", + "correct_answer": "No, the set square is on the table to your back-right.", + "original_options": { + "A": "Yes, the set square is behind you.", + "B": "No, the set square is on the table to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_3_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many small wooden blocks were on the table in front of me?", + "answer": "There are six small wooden blocks on the table, all on your left hand side.", + "question_chinese": "我面前的桌子上的小木块有多少个？", + "answer_chinese": "桌子上的小木块有六块，都在你的左手边。", + "evidence": [ + { + "description": "There are six small wooden blocks on the table.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:46" + } + } + ], + "question_timestamps": "00:04:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are six small wooden blocks on the table, all on your left hand side.", + "B": "Seven blocks.", + "C": "Five blocks.", + "D": "There are seven small wooden blocks on the table, all on your left hand side.", + "E": "There are four small wooden blocks on the table, all on your left hand side." + }, + "answer_label": "A", + "answer_text": "There are six small wooden blocks on the table, all on your left hand side.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_3_T1", + "correct_answer": "There are six small wooden blocks on the table, all on your left hand side.", + "original_options": { + "A": "There are six small wooden blocks on the table, all on your left hand side.", + "B": "Seven blocks.", + "C": "Five blocks.", + "D": "There are seven small wooden blocks on the table, all on your left hand side.", + "E": "There are four small wooden blocks on the table, all on your left hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_3_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many small wooden blocks were on the table in front of me?", + "answer": "There are six small wooden blocks on the table, all behind you.", + "question_chinese": "我面前的桌子上的小木块有多少个？", + "answer_chinese": "桌子上的小木块有六块，都在你的后面。", + "evidence": [ + { + "description": "There are six small wooden blocks on the table.", + "timestamps": { + "start": "00:13:15", + "end": "00:13:21" + } + } + ], + "question_timestamps": "00:13:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Seven blocks.", + "B": "There are six small wooden blocks on the table, all behind you.", + "C": "Five blocks.", + "D": "There are eight small wooden blocks on the table, all behind you.", + "E": "There are four small wooden blocks on the table, all behind you." + }, + "answer_label": "B", + "answer_text": "There are six small wooden blocks on the table, all behind you.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_3_T2", + "correct_answer": "There are six small wooden blocks on the table, all behind you.", + "original_options": { + "A": "Seven blocks.", + "B": "There are six small wooden blocks on the table, all behind you.", + "C": "Five blocks.", + "D": "There are eight small wooden blocks on the table, all behind you.", + "E": "There are four small wooden blocks on the table, all behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did these two long wooden boards on the table get here?", + "answer": "These two long planks were moved here from the right side of the carpenter's bench by you.", + "question_chinese": "桌子上这两块长木板是如何移动到这里的？", + "answer_chinese": "这两块长木板从木工台右边被你搬到了这里。", + "evidence": [ + { + "description": "These two long planks were moved here from the right side of the carpenter's bench by you.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:02:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the left side of the bench.", + "B": "From the front of the cabinet.", + "C": "You carried the two long planks over from the left side of the carpenter's bench and set them on the table.", + "D": "These two long planks were moved here from the right side of the carpenter's bench by you.", + "E": "You brought the two long boards in from the storage shed behind you, then placed them down on the table." + }, + "answer_label": "D", + "answer_text": "These two long planks were moved here from the right side of the carpenter's bench by you.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_1", + "correct_answer": "These two long planks were moved here from the right side of the carpenter's bench by you.", + "original_options": { + "A": "From the left side of the bench.", + "B": "From the front of the cabinet.", + "C": "You carried the two long planks over from the left side of the carpenter's bench and set them on the table.", + "D": "These two long planks were moved here from the right side of the carpenter's bench by you.", + "E": "You brought the two long boards in from the storage shed behind you, then placed them down on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 21, + 20 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did these two long wooden boards on the table get here?", + "answer": "You picked up a plank of wood from the grassy area downstairs and took it to the saw for processing. Then you carried it up to the carpenter's workbench.", + "question_chinese": "桌子上这两块长木板是如何移动到这里的？", + "answer_chinese": "你在楼下的草地上拿了一根木板，拿到了锯子处加工。然后从锯子处搬上了木工台。", + "evidence": [ + { + "description": "You picked up a plank of wood from the grassy area downstairs and took it to the saw for processing.", + "timestamps": { + "start": "00:18:42", + "end": "00:18:55" + } + }, + { + "description": "You carried it up to the carpenter's workbench.", + "timestamps": { + "start": "00:21:44", + "end": "00:22:11" + } + } + ], + "question_timestamps": "00:22:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You picked up a plank of wood from the grassy area downstairs and took it to the saw for processing. Then you carried it up to the carpenter's workbench.", + "B": "You carried them from the saw area.", + "C": "You brought them from the workbench.", + "D": "You picked up the plank upstairs by the carpenter's bench, took it down to the saw, then returned it to the table.", + "E": "You grabbed the plank near the woodworking table first, then went to the saw afterward before leaving it on the bench." + }, + "answer_label": "A", + "answer_text": "You picked up a plank of wood from the grassy area downstairs and took it to the saw for processing. Then you carried it up to the carpenter's workbench.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_2", + "correct_answer": "You picked up a plank of wood from the grassy area downstairs and took it to the saw for processing. Then you carried it up to the carpenter's workbench.", + "original_options": { + "A": "You picked up a plank of wood from the grassy area downstairs and took it to the saw for processing. Then you carried it up to the carpenter's workbench.", + "B": "You carried them from the saw area.", + "C": "You brought them from the workbench.", + "D": "You picked up the plank upstairs by the carpenter's bench, took it down to the saw, then returned it to the table.", + "E": "You grabbed the plank near the woodworking table first, then went to the saw afterward before leaving it on the bench." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 22, + 21 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_3", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did these two long wooden boards on the table get here?", + "answer": "You took the long plank from beside the saw. After being processed by the saw, the long plank was then brought to the woodworking table.", + "question_chinese": "桌子上这两块长木板是如何移动到这里的？", + "answer_chinese": "你从锯子旁拿出了长木板。长木板被锯子加工之后被拿到了木工桌。", + "evidence": [ + { + "description": "You took the long plank from beside the saw.", + "timestamps": { + "start": "00:30:50", + "end": "00:30:55" + } + }, + { + "description": "After being processed by the saw, the long plank was then brought to the woodworking table.", + "timestamps": { + "start": "00:32:55", + "end": "00:33:15" + } + } + ], + "question_timestamps": "00:33:16", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took the long plank from beside the saw. After being processed by the saw, the long plank was then brought to the woodworking table.", + "B": "You took it from behind the saw.", + "C": "You used a short board instead.", + "D": "You grabbed a wide board right of the saw, processed it, then carried it to the woodworking table.", + "E": "You picked up a large plank in front of the saw, processed it, and brought it to the woodworking table." + }, + "answer_label": "A", + "answer_text": "You took the long plank from beside the saw. After being processed by the saw, the long plank was then brought to the woodworking table.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_3", + "correct_answer": "You took the long plank from beside the saw. After being processed by the saw, the long plank was then brought to the woodworking table.", + "original_options": { + "A": "You took the long plank from beside the saw. After being processed by the saw, the long plank was then brought to the woodworking table.", + "B": "You took it from behind the saw.", + "C": "You used a short board instead.", + "D": "You grabbed a wide board right of the saw, processed it, then carried it to the woodworking table.", + "E": "You picked up a large plank in front of the saw, processed it, and brought it to the woodworking table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 20 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_6_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I passed the bottom of the stairs, which direction was the chainsaw from me?", + "answer": "The chainsaw is to your right.", + "question_chinese": "经过楼梯底时，电锯在我的哪个方向？", + "answer_chinese": "电锯在你的右边。", + "evidence": [ + { + "description": "The elevator was on your right.", + "timestamps": { + "start": "00:32:15", + "end": "00:33:00" + } + } + ], + "question_timestamps": "00:33:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "Directly in front of you.", + "C": "The chainsaw was behind you as you passed the bottom of the stairs, on the same path you came from.", + "D": "The chainsaw was to your back-right when you reached the bottom of the stairs, slightly behind your shoulder.", + "E": "The chainsaw is to your right." + }, + "answer_label": "E", + "answer_text": "The chainsaw is to your right.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_6_T2", + "correct_answer": "The chainsaw is to your right.", + "original_options": { + "A": "To your left.", + "B": "Directly in front of you.", + "C": "The chainsaw was behind you as you passed the bottom of the stairs, on the same path you came from.", + "D": "The chainsaw was to your back-right when you reached the bottom of the stairs, slightly behind your shoulder.", + "E": "The chainsaw is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 5, + 20, + 18 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_6_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I passed the bottom of the stairs, which direction was the chainsaw from me?", + "answer": "When you just went downstairs, the elevator was to your left.", + "question_chinese": "经过楼梯底时，电锯在我的哪个方向？", + "answer_chinese": "刚刚从楼梯下去时，电梯在你的左边。", + "evidence": [ + { + "description": "The elevator was on your left.", + "timestamps": { + "start": "00:30:45", + "end": "00:30:50" + } + } + ], + "question_timestamps": "00:33:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was to your right.", + "B": "It was in front of you.", + "C": "After you went downstairs, the elevator was behind you at the bottom of the stairs.", + "D": "When you just went downstairs, the elevator was to your left.", + "E": "When you had just gone upstairs, the elevator was to your left as you reached the landing." + }, + "answer_label": "D", + "answer_text": "When you just went downstairs, the elevator was to your left.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_6_T1", + "correct_answer": "When you just went downstairs, the elevator was to your left.", + "original_options": { + "A": "It was to your right.", + "B": "It was in front of you.", + "C": "After you went downstairs, the elevator was behind you at the bottom of the stairs.", + "D": "When you just went downstairs, the elevator was to your left.", + "E": "When you had just gone upstairs, the elevator was to your left as you reached the landing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 17 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_2_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the stack of cardboard boxes holding sticks relative to me?", + "answer": "The cardboard boxes filled with sticks are stacked below you to your right, between the houses.", + "question_chinese": "装着棍子的纸盒子堆相对于我的位置是什么？", + "answer_chinese": "装着棍子的纸盒子堆在我当前下方的右边，在房子之间。", + "evidence": [ + { + "description": "Go down the stairs and walk towards the middle of the truck and the cabin. These boxes are placed against the wall.", + "timestamps": { + "start": "00:14:38", + "end": "00:15:02" + } + } + ], + "question_timestamps": "00:15:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The cardboard boxes filled with sticks are stacked below you to your right, between the houses.", + "B": "Stacked below you to your left.", + "C": "Stacked above you to your right.", + "D": "The cardboard boxes filled with sticks are stacked below you to your back-right, between the houses.", + "E": "The cardboard boxes filled with sticks are stacked below you to your right, but positioned beside the houses." + }, + "answer_label": "A", + "answer_text": "The cardboard boxes filled with sticks are stacked below you to your right, between the houses.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_2_T2", + "correct_answer": "The cardboard boxes filled with sticks are stacked below you to your right, between the houses.", + "original_options": { + "A": "The cardboard boxes filled with sticks are stacked below you to your right, between the houses.", + "B": "Stacked below you to your left.", + "C": "Stacked above you to your right.", + "D": "The cardboard boxes filled with sticks are stacked below you to your back-right, between the houses.", + "E": "The cardboard boxes filled with sticks are stacked below you to your right, but positioned beside the houses." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39", + "video_length": "00:55:13", + "qid": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_2_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the stack of cardboard boxes holding sticks relative to me?", + "answer": "These boxes are on the ground below you, to your back-left, placed up against the wall.", + "question_chinese": "装着棍子的纸盒子堆相对于我的位置是什么？", + "answer_chinese": "这些盒子在我下方地面的左后方，被放在靠墙的位置。", + "evidence": [ + { + "description": "Go down the stairs and walk towards the middle of the truck and the cabin. These boxes are placed against the wall.", + "timestamps": { + "start": "00:14:38", + "end": "00:15:02" + } + } + ], + "question_timestamps": "00:41:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Below you at your back-right.", + "B": "Below you at your front-left.", + "C": "These boxes are on the ground below you, to your back-left, but set a bit away from the wall.", + "D": "These boxes are on the ground above you, to your back-left, pressed up against the wall.", + "E": "These boxes are on the ground below you, to your back-left, placed up against the wall." + }, + "answer_label": "E", + "answer_text": "These boxes are on the ground below you, to your back-left, placed up against the wall.", + "q_id": "4bbc5b0a-5eb4-4973-a57a-45feceee9f39_2_T1", + "correct_answer": "These boxes are on the ground below you, to your back-left, placed up against the wall.", + "original_options": { + "A": "Below you at your back-right.", + "B": "Below you at your front-left.", + "C": "These boxes are on the ground below you, to your back-left, but set a bit away from the wall.", + "D": "These boxes are on the ground above you, to your back-left, pressed up against the wall.", + "E": "These boxes are on the ground below you, to your back-left, placed up against the wall." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 16 + ] + } +] diff --git a/QAs/4bc033f3-9bfe-4b00-8595-09148707bb02-mcq.json b/QAs/4bc033f3-9bfe-4b00-8595-09148707bb02-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..965a236192dcf80be7d385b3390b5a64a8921a84 --- /dev/null +++ b/QAs/4bc033f3-9bfe-4b00-8595-09148707bb02-mcq.json @@ -0,0 +1,1376 @@ +[ + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_9", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many cloves of garlic have I used in total?", + "answer": "You used two cloves of garlic.", + "question_chinese": "到目前为止，我一共用了几瓣蒜？", + "answer_chinese": "你用了两瓣蒜。", + "evidence": [ + { + "description": "You chopped two cloves of garlic.", + "timestamps": { + "start": "00:09:03", + "end": "00:09:15" + } + } + ], + "question_timestamps": "00:09:42", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used one clove of garlic.", + "B": "You used two cloves of garlic.", + "C": "You used three cloves of garlic.", + "D": "You used four cloves of garlic in total so far, counting the ones you already prepared earlier.", + "E": "You used five cloves of garlic overall up to now, including the cloves you chopped before cooking." + }, + "answer_label": "B", + "answer_text": "You used two cloves of garlic.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_9", + "correct_answer": "You used two cloves of garlic.", + "original_options": { + "A": "You used one clove of garlic.", + "B": "You used two cloves of garlic.", + "C": "You used three cloves of garlic.", + "D": "You used four cloves of garlic in total so far, counting the ones you already prepared earlier.", + "E": "You used five cloves of garlic overall up to now, including the cloves you chopped before cooking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_10", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many cloves of garlic have I used in total?", + "answer": "You used two cloves of garlic.", + "question_chinese": "到目前为止，我一共用了几瓣蒜？", + "answer_chinese": "你用了两瓣蒜。", + "evidence": [ + { + "description": "You chopped two cloves of garlic.", + "timestamps": { + "start": "00:09:03", + "end": "00:09:15" + } + } + ], + "question_timestamps": "00:20:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used one clove of garlic.", + "B": "You used three cloves of garlic.", + "C": "You used four cloves of garlic so far in total, including the ones you already peeled and chopped.", + "D": "You used two cloves of garlic.", + "E": "You used five cloves of garlic altogether up to this point, counting everything you added while prepping." + }, + "answer_label": "D", + "answer_text": "You used two cloves of garlic.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_10", + "correct_answer": "You used two cloves of garlic.", + "original_options": { + "A": "You used one clove of garlic.", + "B": "You used three cloves of garlic.", + "C": "You used four cloves of garlic so far in total, including the ones you already peeled and chopped.", + "D": "You used two cloves of garlic.", + "E": "You used five cloves of garlic altogether up to this point, counting everything you added while prepping." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_5_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Up to now, what have I cut on the cutting board?", + "answer": "You chopped two cloves of garlic on the cutting board in front of you.", + "question_chinese": "到目前为止，我在案板上切过什么？", + "answer_chinese": "你在面前的案板上切了两瓣蒜。", + "evidence": [ + { + "description": "You chopped two cloves of garlic.", + "timestamps": { + "start": "00:09:03", + "end": "00:09:15" + } + } + ], + "question_timestamps": "00:09:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You chopped garlic behind you.", + "B": "You chopped garlic to your left.", + "C": "You chopped two cloves of garlic on the cutting board to your right, not the one in front.", + "D": "You chopped two cloves of garlic on the cutting board above you, rather than on the board in front.", + "E": "You chopped two cloves of garlic on the cutting board in front of you." + }, + "answer_label": "E", + "answer_text": "You chopped two cloves of garlic on the cutting board in front of you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_5_T1", + "correct_answer": "You chopped two cloves of garlic on the cutting board in front of you.", + "original_options": { + "A": "You chopped garlic behind you.", + "B": "You chopped garlic to your left.", + "C": "You chopped two cloves of garlic on the cutting board to your right, not the one in front.", + "D": "You chopped two cloves of garlic on the cutting board above you, rather than on the board in front.", + "E": "You chopped two cloves of garlic on the cutting board in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 19 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_5_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Up to now, what have I cut on the cutting board?", + "answer": "You chopped garlic and ginger on the cutting board in front of you.", + "question_chinese": "到目前为止，我在案板上切过什么？", + "answer_chinese": "你在面前的案板上切了蒜和姜。", + "evidence": [ + { + "description": "You chopped two cloves of garlic.", + "timestamps": { + "start": "00:09:03", + "end": "00:09:15" + } + }, + { + "description": "You cut the ginger.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:11" + } + } + ], + "question_timestamps": "00:10:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You chopped only garlic.", + "B": "You chopped only ginger.", + "C": "You chopped garlic and ginger on the cutting board in front of you.", + "D": "You chopped garlic and ginger on the cutting board to your left instead of on the board in front.", + "E": "You chopped garlic and ginger on the cutting board behind you, not on the cutting board in front of you." + }, + "answer_label": "C", + "answer_text": "You chopped garlic and ginger on the cutting board in front of you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_5_T2", + "correct_answer": "You chopped garlic and ginger on the cutting board in front of you.", + "original_options": { + "A": "You chopped only garlic.", + "B": "You chopped only ginger.", + "C": "You chopped garlic and ginger on the cutting board in front of you.", + "D": "You chopped garlic and ginger on the cutting board to your left instead of on the board in front.", + "E": "You chopped garlic and ginger on the cutting board behind you, not on the cutting board in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 20 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_10_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many spoons have I used?", + "answer": "You used one spoon in front of you.", + "question_chinese": "到目前为止，我用了几个勺子？", + "answer_chinese": "你用了面前的一个勺子", + "evidence": [ + { + "description": "You use the first spoon to scoop the contents from the can into another container.", + "timestamps": { + "start": "00:23:46", + "end": "00:23:50" + } + } + ], + "question_timestamps": "00:23:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You used zero spoons in front of you.", + "B": "You used two spoons in front of you.", + "C": "You used three spoons in front of you, switching to a new one after the first.", + "D": "You used four spoons in front of you overall, using different spoons for separate steps.", + "E": "You used one spoon in front of you." + }, + "answer_label": "E", + "answer_text": "You used one spoon in front of you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_10_T1", + "correct_answer": "You used one spoon in front of you.", + "original_options": { + "A": "You used zero spoons in front of you.", + "B": "You used two spoons in front of you.", + "C": "You used three spoons in front of you, switching to a new one after the first.", + "D": "You used four spoons in front of you overall, using different spoons for separate steps.", + "E": "You used one spoon in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 15 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_20", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoons have I used so far?", + "answer": "You used two spoons.", + "question_chinese": "到目前为止，我用了几个勺子？", + "answer_chinese": "你用了两个勺子。", + "evidence": [ + { + "description": "You use the spoon placed next to the shovel to dig.", + "timestamps": { + "start": "00:25:28", + "end": "00:25:33" + } + } + ], + "question_timestamps": "00:25:35", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used two spoons.", + "B": "You used one spoon.", + "C": "You used three spoons.", + "D": "You have used four spoons so far during the task, not just a couple.", + "E": "Up to this point, you used five spoons in total while handling the food." + }, + "answer_label": "A", + "answer_text": "You used two spoons.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_20", + "correct_answer": "You used two spoons.", + "original_options": { + "A": "You used two spoons.", + "B": "You used one spoon.", + "C": "You used three spoons.", + "D": "You have used four spoons so far during the task, not just a couple.", + "E": "Up to this point, you used five spoons in total while handling the food." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_10_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many spoons have I used?", + "answer": "You used two spoons in front of you.", + "question_chinese": "到目前为止，我用了几个勺子？", + "answer_chinese": "你用了面前的两个勺子。", + "evidence": [ + { + "description": "You have taken a new spoon to scoop something.", + "timestamps": { + "start": "00:23:49", + "end": "00:23:56" + } + }, + { + "description": "You use the spoon placed next to the shovel to dig.", + "timestamps": { + "start": "00:25:28", + "end": "00:25:33" + } + } + ], + "question_timestamps": "00:39:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You used two spoons in front of you.", + "B": "You used no spoons yet.", + "C": "You used one spoon in front of you.", + "D": "So far, you used three spoons in front of you while continuing the activity.", + "E": "Up to now, you used four spoons in front of you, not just two." + }, + "answer_label": "A", + "answer_text": "You used two spoons in front of you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_10_T2", + "correct_answer": "You used two spoons in front of you.", + "original_options": { + "A": "You used two spoons in front of you.", + "B": "You used no spoons yet.", + "C": "You used one spoon in front of you.", + "D": "So far, you used three spoons in front of you while continuing the activity.", + "E": "Up to now, you used four spoons in front of you, not just two." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 14, + 14 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_17", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to get more spoons, where can I get them?", + "answer": "You can go to the first drawer on the right side of the baby chair to get it.", + "question_chinese": "如果我想拿更多的勺子，去哪拿？", + "answer_chinese": "你可以去婴儿椅右侧的第一个抽屉里拿。", + "evidence": [ + { + "description": "There are many spoons in the first drawer on the right side of the baby chair.", + "timestamps": { + "start": "00:23:22", + "end": "00:23:31" + } + } + ], + "question_timestamps": "00:23:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go to the left drawer.", + "B": "Check the drawer in front.", + "C": "You can get more spoons from the first drawer on the left side of the baby chair.", + "D": "You should open the first drawer behind the baby chair to find additional spoons.", + "E": "You can go to the first drawer on the right side of the baby chair to get it." + }, + "answer_label": "E", + "answer_text": "You can go to the first drawer on the right side of the baby chair to get it.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_17", + "correct_answer": "You can go to the first drawer on the right side of the baby chair to get it.", + "original_options": { + "A": "Go to the left drawer.", + "B": "Check the drawer in front.", + "C": "You can get more spoons from the first drawer on the left side of the baby chair.", + "D": "You should open the first drawer behind the baby chair to find additional spoons.", + "E": "You can go to the first drawer on the right side of the baby chair to get it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 14 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_9_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to get more spoons, where should I go to get them?", + "answer": "You can get it from the first drawer behind you.", + "question_chinese": "如果我想拿更多的勺子，去哪拿？", + "answer_chinese": "你可以去你身后的第一个抽屉里拿。", + "evidence": [ + { + "description": "There are many spoons in the first drawer behind you.", + "timestamps": { + "start": "00:23:22", + "end": "00:23:31" + } + } + ], + "question_timestamps": "00:23:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the drawer in front.", + "B": "From the drawer to your left.", + "C": "You can get it from the first drawer behind you.", + "D": "You should take it from the first drawer to your right, not from behind you.", + "E": "To get more spoons, go to the first drawer above you instead of the one behind you." + }, + "answer_label": "C", + "answer_text": "You can get it from the first drawer behind you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_9_T1", + "correct_answer": "You can get it from the first drawer behind you.", + "original_options": { + "A": "From the drawer in front.", + "B": "From the drawer to your left.", + "C": "You can get it from the first drawer behind you.", + "D": "You should take it from the first drawer to your right, not from behind you.", + "E": "To get more spoons, go to the first drawer above you instead of the one behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 17 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_9_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to get more spoons, where should I go to get them?", + "answer": "You can get it from the first drawer on the right side of the baby chair.", + "question_chinese": "如果我想拿更多的勺子，去哪拿？", + "answer_chinese": "你可以去婴儿椅右侧的第一个抽屉里拿。", + "evidence": [ + { + "description": "There are many spoons in the first drawer on the right side of the baby chair.", + "timestamps": { + "start": "00:23:22", + "end": "00:23:31" + } + } + ], + "question_timestamps": "00:34:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Get it from the left drawer.", + "B": "Get it from the front drawer.", + "C": "You can get it from the first drawer behind the baby chair instead of on its right side.", + "D": "To find more spoons, open the first drawer to the left of the baby chair, not the right.", + "E": "You can get it from the first drawer on the right side of the baby chair." + }, + "answer_label": "E", + "answer_text": "You can get it from the first drawer on the right side of the baby chair.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_9_T2", + "correct_answer": "You can get it from the first drawer on the right side of the baby chair.", + "original_options": { + "A": "Get it from the left drawer.", + "B": "Get it from the front drawer.", + "C": "You can get it from the first drawer behind the baby chair instead of on its right side.", + "D": "To find more spoons, open the first drawer to the left of the baby chair, not the right.", + "E": "You can get it from the first drawer on the right side of the baby chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_3", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to find more pots, where can I go?", + "answer": "There are three pots on the stove top.", + "question_chinese": "如果我想要找更多的锅，去哪里找？", + "answer_chinese": "灶台台面上有三个锅。", + "evidence": [ + { + "description": "There is one pot on the right side of the stove, and there are two pots on the stove.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There are two pots on the stove top.", + "B": "There is one pot on the stove top.", + "C": "You can find four pots on the stove top, not three, when you look there.", + "D": "There are five pots sitting on the stove top, so you can grab one from there.", + "E": "There are three pots on the stove top." + }, + "answer_label": "E", + "answer_text": "There are three pots on the stove top.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_3", + "correct_answer": "There are three pots on the stove top.", + "original_options": { + "A": "There are two pots on the stove top.", + "B": "There is one pot on the stove top.", + "C": "You can find four pots on the stove top, not three, when you look there.", + "D": "There are five pots sitting on the stove top, so you can grab one from there.", + "E": "There are three pots on the stove top." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 16 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_4", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to find more pots, where can I go?", + "answer": "Look for it on the stove top and in the third drawer below the stove.", + "question_chinese": "如果我想要找更多的锅，去哪里找？", + "answer_chinese": "你去灶台台面和灶台下面的第三个抽屉里找。", + "evidence": [ + { + "description": "There is a pot beside the sink.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:47" + } + }, + { + "description": "There is a pile of pots in the third drawer under the stove.", + "timestamps": { + "start": "00:00:52", + "end": "00:01:00" + } + }, + { + "description": "There are two pots on the stove.", + "timestamps": { + "start": "00:18:44", + "end": "00:18:49" + } + } + ], + "question_timestamps": "00:18:50", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Look on the counter.", + "B": "Check the pantry.", + "C": "Look for it on the stove top and in the third drawer below the stove.", + "D": "Look for it on the stove top and in the second drawer below the stove, not the third.", + "E": "Search on the stove top and in the top drawer beneath the stove instead of the third drawer." + }, + "answer_label": "C", + "answer_text": "Look for it on the stove top and in the third drawer below the stove.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_4", + "correct_answer": "Look for it on the stove top and in the third drawer below the stove.", + "original_options": { + "A": "Look on the counter.", + "B": "Check the pantry.", + "C": "Look for it on the stove top and in the third drawer below the stove.", + "D": "Look for it on the stove top and in the second drawer below the stove, not the third.", + "E": "Search on the stove top and in the top drawer beneath the stove instead of the third drawer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 18, + 18 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_6_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was cleaning up the food peels, which was closer to me in front of me, the knife or the grater?", + "answer": "The knife in front of you is closer to you.", + "question_chinese": "当我在收拾食物外皮的时候,面前的刀和磨碎器哪个离我更近?", + "answer_chinese": "面前的刀离你更近", + "evidence": [ + { + "description": "The grater is closer to you, about 5cm away, and the knife is above and to your right, about 25cm away.Then you hold the kitchen waste and move towards the knife.", + "timestamps": { + "start": "00:20:47", + "end": "00:20:58" + } + } + ], + "question_timestamps": "00:21:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The grater in front of you is closer.", + "B": "The knife in front of you is closer to you.", + "C": "The knife in front of you is farther.", + "D": "The grater in front of you was closer to you than the knife when you cleaned the peels.", + "E": "In front of you, the knife was farther away, so the grater was the nearer item." + }, + "answer_label": "B", + "answer_text": "The knife in front of you is closer to you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_6_T1", + "correct_answer": "The knife in front of you is closer to you.", + "original_options": { + "A": "The grater in front of you is closer.", + "B": "The knife in front of you is closer to you.", + "C": "The knife in front of you is farther.", + "D": "The grater in front of you was closer to you than the knife when you cleaned the peels.", + "E": "In front of you, the knife was farther away, so the grater was the nearer item." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 16 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_6_T2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was cleaning up the food peels, which was closer to me in front of me, the knife or the grater?", + "answer": "The grinder in front of you is closer to you.", + "question_chinese": "当我在收拾食物外皮的时候,面前的刀和磨碎器哪个离我更近?", + "answer_chinese": "面前的磨碎器离你更近。", + "evidence": [ + { + "description": "The grater is closer to you, about 5cm away, and the knife is above and to your right, about 25cm away.", + "timestamps": { + "start": "00:20:47", + "end": "00:21:01" + } + } + ], + "question_timestamps": "00:23:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The knife was closer.", + "B": "The grater was closer.", + "C": "The grinder in front of you is closer to you.", + "D": "While cleaning up the food peels, the knife in front of you was closer than the grinder.", + "E": "While cleaning up the food peels, the grater in front of you was closer than the grinder." + }, + "answer_label": "C", + "answer_text": "The grinder in front of you is closer to you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_6_T2", + "correct_answer": "The grinder in front of you is closer to you.", + "original_options": { + "A": "The knife was closer.", + "B": "The grater was closer.", + "C": "The grinder in front of you is closer to you.", + "D": "While cleaning up the food peels, the knife in front of you was closer than the grinder.", + "E": "While cleaning up the food peels, the grater in front of you was closer than the grinder." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_1_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I put my phone down for the first time, which direction was it from me?", + "answer": "The phone is on the table behind you.", + "question_chinese": "当我第一次放下手机后，它在我的哪个方向？", + "answer_chinese": "手机在你后方的桌上。", + "evidence": [ + { + "description": "You put your phone on the right table and walk forward.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The phone is on the table to your right, not behind you where you were standing.", + "D": "The phone is on the table behind you.", + "E": "The phone is set on the table in front of you, within your view as you put it down." + }, + "answer_label": "D", + "answer_text": "The phone is on the table behind you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_1_T1", + "correct_answer": "The phone is on the table behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The phone is on the table to your right, not behind you where you were standing.", + "D": "The phone is on the table behind you.", + "E": "The phone is set on the table in front of you, within your view as you put it down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 19 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_1_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I put my phone down for the first time, which direction was it from me?", + "answer": "The phone is on the table on the right side of you.", + "question_chinese": "当我第一次放下手机后，它在我的哪个方向？", + "answer_chinese": "手机在你的右侧的桌子上。", + "evidence": [ + { + "description": "You put down your phone and turned to the left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "The phone is on the table on the right side of you.", + "C": "To your left.", + "D": "The phone is on the table behind you, not on the right side.", + "E": "The phone was placed on the table straight ahead of you rather than off to your right." + }, + "answer_label": "B", + "answer_text": "The phone is on the table on the right side of you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_1_T2", + "correct_answer": "The phone is on the table on the right side of you.", + "original_options": { + "A": "In front of you.", + "B": "The phone is on the table on the right side of you.", + "C": "To your left.", + "D": "The phone is on the table behind you, not on the right side.", + "E": "The phone was placed on the table straight ahead of you rather than off to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 13, + 17 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_15", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I taking?", + "answer": "You are turning around from the workstation, taking two steps, and moving towards the stove.", + "question_chinese": "我刚刚走什么路线？", + "answer_chinese": "你正从操作台转身，走两步，朝着灶台方向移动。", + "evidence": [ + { + "description": "You turned around from the workbench with the mortar and walked over to the stove.", + "timestamps": { + "start": "00:23:11", + "end": "00:23:15" + } + } + ], + "question_timestamps": "00:23:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You move toward the sink.", + "B": "You are turning around from the workstation, taking two steps, and moving towards the stove.", + "C": "You head to the countertop.", + "D": "You turn around from the workstation, take two steps, and move toward the countertop instead of the stove.", + "E": "You turn slightly at the workstation, take two steps, and walk toward the sink rather than approaching the stove." + }, + "answer_label": "B", + "answer_text": "You are turning around from the workstation, taking two steps, and moving towards the stove.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_15", + "correct_answer": "You are turning around from the workstation, taking two steps, and moving towards the stove.", + "original_options": { + "A": "You move toward the sink.", + "B": "You are turning around from the workstation, taking two steps, and moving towards the stove.", + "C": "You head to the countertop.", + "D": "You turn around from the workstation, take two steps, and move toward the countertop instead of the stove.", + "E": "You turn slightly at the workstation, take two steps, and walk toward the sink rather than approaching the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 19 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_16", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I taking?", + "answer": "You are moving from the stove, bypassing the baby chair, towards the cupboar.d.", + "question_chinese": "我刚刚走什么路线？", + "answer_chinese": "你正从灶台绕过婴儿椅，朝着橱柜方向移动。", + "evidence": [ + { + "description": "You bypass the baby chair by the stove and take eight steps straight to the cabinet.", + "timestamps": { + "start": "00:29:57", + "end": "00:30:04" + } + } + ], + "question_timestamps": "00:30:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are moving from the stove, bypassing the baby chair, towards the cupboar.d.", + "B": "You go from the stove to the sink.", + "C": "You walk toward the refrigerator.", + "D": "You move from the stove, pass by the baby chair, and head toward the refrigerator instead of the cupboard.", + "E": "You leave the stove, go around the table area, and continue toward the pantry rather than the cupboard." + }, + "answer_label": "A", + "answer_text": "You are moving from the stove, bypassing the baby chair, towards the cupboar.d.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_16", + "correct_answer": "You are moving from the stove, bypassing the baby chair, towards the cupboar.d.", + "original_options": { + "A": "You are moving from the stove, bypassing the baby chair, towards the cupboar.d.", + "B": "You go from the stove to the sink.", + "C": "You walk toward the refrigerator.", + "D": "You move from the stove, pass by the baby chair, and head toward the refrigerator instead of the cupboard.", + "E": "You leave the stove, go around the table area, and continue toward the pantry rather than the cupboard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 5, + 19, + 18 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_7_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans did I take out from the drawer in front of me?", + "answer": "You took two cans out of the drawer in front of you.", + "question_chinese": "我在面前的抽屉拿出了几个罐头？", + "answer_chinese": "你在面前的抽屉拿出了两个罐头", + "evidence": [ + { + "description": "You took out a can.", + "timestamps": { + "start": "00:21:44", + "end": "00:21:45" + } + }, + { + "description": "You took out the second can.", + "timestamps": { + "start": "00:21:51", + "end": "00:21:52" + } + } + ], + "question_timestamps": "00:21:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took one can out.", + "B": "You took three cans out.", + "C": "You took four cans out of the drawer in front of you, not just two.", + "D": "You took two cans out of the drawer in front of you.", + "E": "You took zero cans out of the drawer in front of you and left everything inside." + }, + "answer_label": "D", + "answer_text": "You took two cans out of the drawer in front of you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_7_T2", + "correct_answer": "You took two cans out of the drawer in front of you.", + "original_options": { + "A": "You took one can out.", + "B": "You took three cans out.", + "C": "You took four cans out of the drawer in front of you, not just two.", + "D": "You took two cans out of the drawer in front of you.", + "E": "You took zero cans out of the drawer in front of you and left everything inside." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_7_T1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans did I take out from the drawer in front of me?", + "answer": "You took three cans out of the drawer in front of you.", + "question_chinese": "我在面前的抽屉拿出了几个罐头？", + "answer_chinese": "你在面前的抽屉拿出了三个罐头", + "evidence": [ + { + "description": "You took out a can.", + "timestamps": { + "start": "00:21:43", + "end": "00:21:44" + } + }, + { + "description": "You took out the second can.", + "timestamps": { + "start": "00:21:52", + "end": "00:21:53" + } + }, + { + "description": "You took out the third can.", + "timestamps": { + "start": "00:22:02", + "end": "00:22:04" + } + } + ], + "question_timestamps": "00:22:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took two cans out.", + "B": "You took four cans out.", + "C": "You took three cans out of the drawer in front of you.", + "D": "You took one can out of the drawer in front of you, not three.", + "E": "You took five cans out of the drawer in front of you, more than the amount asked." + }, + "answer_label": "C", + "answer_text": "You took three cans out of the drawer in front of you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_7_T1", + "correct_answer": "You took three cans out of the drawer in front of you.", + "original_options": { + "A": "You took two cans out.", + "B": "You took four cans out.", + "C": "You took three cans out of the drawer in front of you.", + "D": "You took one can out of the drawer in front of you, not three.", + "E": "You took five cans out of the drawer in front of you, more than the amount asked." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 17 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_5", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You was holding a cup and a spoon in your hand just now.", + "question_chinese": "我手里刚刚拿着什么?", + "answer_chinese": "你手里现在拿着杯子和铲子。", + "evidence": [ + { + "description": "You are holding a cup in your left hand and a spatula in your right hand.", + "timestamps": { + "start": "00:04:49", + "end": "00:04:54" + } + } + ], + "question_timestamps": "00:04:59", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You were holding a glass and a fork.", + "B": "You was holding a cup and a spoon in your hand just now.", + "C": "You were holding a bowl and a knife.", + "D": "You were holding a mug and a whisk in your hand just now, not a cup and spoon.", + "E": "You were holding a ladle and a fork in your hand just now rather than a cup and spoon." + }, + "answer_label": "B", + "answer_text": "You was holding a cup and a spoon in your hand just now.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_5", + "correct_answer": "You was holding a cup and a spoon in your hand just now.", + "original_options": { + "A": "You were holding a glass and a fork.", + "B": "You was holding a cup and a spoon in your hand just now.", + "C": "You were holding a bowl and a knife.", + "D": "You were holding a mug and a whisk in your hand just now, not a cup and spoon.", + "E": "You were holding a ladle and a fork in your hand just now rather than a cup and spoon." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 19 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_6", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You was holding a spoon and a hammer in your hand just now.", + "question_chinese": "我手里刚刚拿着什么?", + "answer_chinese": "你现在手里拿着勺子和锤子。", + "evidence": [ + { + "description": "You are holding a spoon in your right hand and a hammer in your left hand.", + "timestamps": { + "start": "00:23:50", + "end": "00:23:52" + } + } + ], + "question_timestamps": "00:23:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You were holding a spoon and a knife.", + "B": "You was holding a spoon and a hammer in your hand just now.", + "C": "You were holding a hammer and a fork.", + "D": "You were holding a spoon with one hand and a wrench in the other just now.", + "E": "You were holding a fork and a hammer in your hand just now, not a spoon." + }, + "answer_label": "B", + "answer_text": "You was holding a spoon and a hammer in your hand just now.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_6", + "correct_answer": "You was holding a spoon and a hammer in your hand just now.", + "original_options": { + "A": "You were holding a spoon and a knife.", + "B": "You was holding a spoon and a hammer in your hand just now.", + "C": "You were holding a hammer and a fork.", + "D": "You were holding a spoon with one hand and a wrench in the other just now.", + "E": "You were holding a fork and a hammer in your hand just now, not a spoon." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 16 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_4_T1", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Along what path did I bring the knife in my hand here?", + "answer": "You took the knife down from above you over the stove and then turned around to face behind you.", + "question_chinese": "我手里这把刀是沿着什么路径拿来的?", + "answer_chinese": "你从灶台上方取下刀,转身向后", + "evidence": [ + { + "description": "You took the knife from above the stove, turned around, and took three steps backward", + "timestamps": { + "start": "00:08:33", + "end": "00:08:34" + } + } + ], + "question_timestamps": "00:08:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned to face in front of you.", + "B": "You turned to face to your left.", + "C": "You lifted the knife down from above the stove, then turned to face to your right instead.", + "D": "You took the knife down from above you over the stove and stayed facing forward without turning around.", + "E": "You took the knife down from above you over the stove and then turned around to face behind you." + }, + "answer_label": "E", + "answer_text": "You took the knife down from above you over the stove and then turned around to face behind you.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_4_T1", + "correct_answer": "You took the knife down from above you over the stove and then turned around to face behind you.", + "original_options": { + "A": "You turned to face in front of you.", + "B": "You turned to face to your left.", + "C": "You lifted the knife down from above the stove, then turned to face to your right instead.", + "D": "You took the knife down from above you over the stove and stayed facing forward without turning around.", + "E": "You took the knife down from above you over the stove and then turned around to face behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 18 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_7", + "qid_num": 24, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I take to get this knife in my hand?", + "answer": "You took the knife down from above the stove, turned around, and walked to your current position.", + "question_chinese": "我手里这把刀是沿着什么路径拿来的?", + "answer_chinese": "你从灶台上方取下刀,转身向后,走到了现在的位置。", + "evidence": [ + { + "description": "You took the knife from above the stove, turned around, and took three steps backward, arriving in front of the current cutting board.", + "timestamps": { + "start": "00:08:33", + "end": "00:08:36" + } + } + ], + "question_timestamps": "00:08:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You grabbed it from the counter.", + "B": "You took it from the table.", + "C": "You picked the knife up from beside the sink, turned around, and walked to where you are now.", + "D": "You took the knife from a shelf above, turned left, and then walked to your current position.", + "E": "You took the knife down from above the stove, turned around, and walked to your current position." + }, + "answer_label": "E", + "answer_text": "You took the knife down from above the stove, turned around, and walked to your current position.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_7", + "correct_answer": "You took the knife down from above the stove, turned around, and walked to your current position.", + "original_options": { + "A": "You grabbed it from the counter.", + "B": "You took it from the table.", + "C": "You picked the knife up from beside the sink, turned around, and walked to where you are now.", + "D": "You took the knife from a shelf above, turned left, and then walked to your current position.", + "E": "You took the knife down from above the stove, turned around, and walked to your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "4bc033f3-9bfe-4b00-8595-09148707bb02", + "video_length": "00:41:07", + "qid": "4bc033f3-9bfe-4b00-8595-09148707bb02_4_T2", + "qid_num": 25, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Along what path did I bring the knife in my hand here?", + "answer": "You took the knife down from above you over the stove, turned around to face behind you, and walked to the cutting board.", + "question_chinese": "我手里这把刀是沿着什么路径拿来的?", + "answer_chinese": "你从灶台上方取下刀,转身向后,走到了案板前", + "evidence": [ + { + "description": "You took the knife from above the stove, turned around, and took three steps backward, arriving in front of the current cutting board.", + "timestamps": { + "start": "00:08:33", + "end": "00:08:36" + } + } + ], + "question_timestamps": "00:25:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned to face in front of you.", + "B": "You turned to face to your right.", + "C": "You took the knife down from above the stove, turned left, and walked to the cutting board.", + "D": "You took the knife down from above you over the stove, turned around to face behind you, and walked to the cutting board.", + "E": "You took the knife down from above the stove, then walked to the cutting board without turning around." + }, + "answer_label": "D", + "answer_text": "You took the knife down from above you over the stove, turned around to face behind you, and walked to the cutting board.", + "q_id": "4bc033f3-9bfe-4b00-8595-09148707bb02_4_T2", + "correct_answer": "You took the knife down from above you over the stove, turned around to face behind you, and walked to the cutting board.", + "original_options": { + "A": "You turned to face in front of you.", + "B": "You turned to face to your right.", + "C": "You took the knife down from above the stove, turned left, and walked to the cutting board.", + "D": "You took the knife down from above you over the stove, turned around to face behind you, and walked to the cutting board.", + "E": "You took the knife down from above the stove, then walked to the cutting board without turning around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 18 + ] + } +] diff --git a/QAs/4bf46557-3c5a-4a89-840d-158403590c9f-mcq.json b/QAs/4bf46557-3c5a-4a89-840d-158403590c9f-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..362fdb20d25d902788cc30f55a088bf4c549717a --- /dev/null +++ b/QAs/4bf46557-3c5a-4a89-840d-158403590c9f-mcq.json @@ -0,0 +1,892 @@ +[ + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_5_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I came back from outside just now, how many people were standing in the lobby?", + "answer": "In the hall where you are now, there are three people in front of you.", + "question_chinese": "刚刚我从外面回来的时候大厅站着几个人？", + "answer_chinese": "你现在所在的大厅你的面前你一共有三个人。", + "evidence": [ + { + "description": "There are three people in total.", + "timestamps": { + "start": "01:02:45", + "end": "01:02:55" + } + } + ], + "question_timestamps": "01:03:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There was one person in front of you.", + "B": "There were two people in front of you.", + "C": "In the hall where you are now, there are four people standing in front of you.", + "D": "In the hall where you are now, there are five people in front of you when you came back inside.", + "E": "In the hall where you are now, there are three people in front of you." + }, + "answer_label": "E", + "answer_text": "In the hall where you are now, there are three people in front of you.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_5_T2", + "correct_answer": "In the hall where you are now, there are three people in front of you.", + "original_options": { + "A": "There was one person in front of you.", + "B": "There were two people in front of you.", + "C": "In the hall where you are now, there are four people standing in front of you.", + "D": "In the hall where you are now, there are five people in front of you when you came back inside.", + "E": "In the hall where you are now, there are three people in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 20 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_5_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I came back from outside just now, how many people were standing in the lobby?", + "answer": "In the lobby where you are now, there are five people around you in total.", + "question_chinese": "刚刚我从外面回来的时候大厅站着几个人？", + "answer_chinese": "你现在所在的大厅你的周围一共有五个人。", + "evidence": [ + { + "description": "There are three people in total.", + "timestamps": { + "start": "01:02:45", + "end": "01:02:55" + } + } + ], + "question_timestamps": "01:03:29", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There were three people around you total.", + "B": "There were four people around you total.", + "C": "In the lobby where you are now, there are six people around you in total.", + "D": "In the lobby where you are now, there are five people around you in total.", + "E": "In the lobby where you are now, there are seven people around you in total after you came back." + }, + "answer_label": "D", + "answer_text": "In the lobby where you are now, there are five people around you in total.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_5_T1", + "correct_answer": "In the lobby where you are now, there are five people around you in total.", + "original_options": { + "A": "There were three people around you total.", + "B": "There were four people around you total.", + "C": "In the lobby where you are now, there are six people around you in total.", + "D": "In the lobby where you are now, there are five people around you in total.", + "E": "In the lobby where you are now, there are seven people around you in total after you came back." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 19 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I came back from outside just now, how many people were standing in the hall?", + "answer": "There are three people in total.", + "question_chinese": "刚刚我从外面回来的时候大厅站着几个人？", + "answer_chinese": "一共有三个人。", + "evidence": [ + { + "description": "There are three people in total.", + "timestamps": { + "start": "01:02:45", + "end": "01:02:55" + } + } + ], + "question_timestamps": "01:06:29", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are no people in total.", + "B": "There is one person in total.", + "C": "There are four people in total standing in the hall right now.", + "D": "There are five people in total in the hall when you came back from outside.", + "E": "There are three people in total." + }, + "answer_label": "E", + "answer_text": "There are three people in total.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_7", + "correct_answer": "There are three people in total.", + "original_options": { + "A": "There are no people in total.", + "B": "There is one person in total.", + "C": "There are four people in total standing in the hall right now.", + "D": "There are five people in total in the hall when you came back from outside.", + "E": "There are three people in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 15 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "As far as I can see, how many vehicles have been just released?", + "answer": "If we count the car you drove, a total of three vehicles left the factory.", + "question_chinese": "就我所看到的，刚刚一共推出去了几辆车？", + "answer_chinese": "如果算上你推的那辆车，从工厂内出去了三辆车。", + "evidence": [ + { + "description": "You pulled out a car.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:50" + } + }, + { + "description": "Two workers pushed out two cars.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:45" + } + } + ], + "question_timestamps": "00:06:09", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "If we count the car you drove, a total of three vehicles left the factory.", + "B": "A total of four vehicles left.", + "C": "A total of five vehicles left.", + "D": "If we count the car you drove, a total of six vehicles left the factory.", + "E": "If we count the car you drove, a total of seven vehicles were released from the factory." + }, + "answer_label": "A", + "answer_text": "If we count the car you drove, a total of three vehicles left the factory.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_2", + "correct_answer": "If we count the car you drove, a total of three vehicles left the factory.", + "original_options": { + "A": "If we count the car you drove, a total of three vehicles left the factory.", + "B": "A total of four vehicles left.", + "C": "A total of five vehicles left.", + "D": "If we count the car you drove, a total of six vehicles left the factory.", + "E": "If we count the car you drove, a total of seven vehicles were released from the factory." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "As far as I can see, how many vehicles have been just released?", + "answer": "If we count the car you drove, a total of four vehicles left the factory.", + "question_chinese": "就我所看到的，刚刚一共推出去了几辆车？", + "answer_chinese": "如果算上你推的那辆车，从工厂内出去了四辆车。", + "evidence": [ + { + "description": "You pushed out one car.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:50" + } + }, + { + "description": "Two workers pushed out two cars.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:45" + } + }, + { + "description": "One worker pushed out one car.", + "timestamps": { + "start": "00:06:55", + "end": "00:07:00" + } + } + ], + "question_timestamps": "00:07:13", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "A total of three vehicles left the factory.", + "B": "If we count the car you drove, a total of four vehicles left the factory.", + "C": "A total of five vehicles left the factory.", + "D": "Including the car you drove, only three vehicles were released from the factory just now.", + "E": "If you include the car you drove, five vehicles in total just left the factory gates." + }, + "answer_label": "B", + "answer_text": "If we count the car you drove, a total of four vehicles left the factory.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_3", + "correct_answer": "If we count the car you drove, a total of four vehicles left the factory.", + "original_options": { + "A": "A total of three vehicles left the factory.", + "B": "If we count the car you drove, a total of four vehicles left the factory.", + "C": "A total of five vehicles left the factory.", + "D": "Including the car you drove, only three vehicles were released from the factory just now.", + "E": "If you include the car you drove, five vehicles in total just left the factory gates." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 16 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_4_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "In the hallway outside the room I just passed through, besides the goods, what other things were placed there?", + "answer": "There’s blue padding under the goods in front of you.", + "question_chinese": "我刚刚经过的房间外面的走廊里除了货物还放着其他东西，请问是什么？", + "answer_chinese": "你面前的货物下面放着蓝色的垫料。", + "evidence": [ + { + "description": "Under the goods, there is a layer of blue padding.", + "timestamps": { + "start": "00:58:53", + "end": "00:58:54" + } + } + ], + "question_timestamps": "00:58:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There’s blue padding under the goods in front of you.", + "B": "Blue padding behind the goods.", + "C": "Blue padding to your left.", + "D": "There’s blue padding placed under the goods to your right in the hallway outside the room.", + "E": "In the hallway outside that room, blue padding sits beneath the goods, but it’s behind you." + }, + "answer_label": "A", + "answer_text": "There’s blue padding under the goods in front of you.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_4_T2", + "correct_answer": "There’s blue padding under the goods in front of you.", + "original_options": { + "A": "There’s blue padding under the goods in front of you.", + "B": "Blue padding behind the goods.", + "C": "Blue padding to your left.", + "D": "There’s blue padding placed under the goods to your right in the hallway outside the room.", + "E": "In the hallway outside that room, blue padding sits beneath the goods, but it’s behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_4_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "In the hallway outside the room I just passed through, besides the goods, what other things were placed there?", + "answer": "Along the corridor behind you, there’s a green bottle placed at about eye level.", + "question_chinese": "我刚刚经过的房间外面的走廊里除了货物还放着其他东西，请问是什么？", + "answer_chinese": "你身后的走廊边放着一个绿色的瓶子，和眼睛一样高。", + "evidence": [ + { + "description": "On the side of the corridor, there is a green bottle, at the same height as the eyes.", + "timestamps": { + "start": "00:58:50", + "end": "00:58:58" + } + } + ], + "question_timestamps": "01:00:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A green bottle is in front of you.", + "B": "A green bottle is to your left.", + "C": "Along the corridor to your right, there’s a green bottle positioned roughly at your eye level.", + "D": "Along the corridor behind you, there’s a green bottle placed at about eye level.", + "E": "Further down the corridor in front of you, a green bottle sits placed around eye height." + }, + "answer_label": "D", + "answer_text": "Along the corridor behind you, there’s a green bottle placed at about eye level.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_4_T1", + "correct_answer": "Along the corridor behind you, there’s a green bottle placed at about eye level.", + "original_options": { + "A": "A green bottle is in front of you.", + "B": "A green bottle is to your left.", + "C": "Along the corridor to your right, there’s a green bottle positioned roughly at your eye level.", + "D": "Along the corridor behind you, there’s a green bottle placed at about eye level.", + "E": "Further down the corridor in front of you, a green bottle sits placed around eye height." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 16 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_8_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, what phone-related item is in the office?", + "answer": "On the desk in the office in front of you, there was a phone, a phone case, and a charger.", + "question_chinese": "我在现在办公室放着什么和手机有关的东西？", + "answer_chinese": "你前方办公室的桌子上放着一个手机一个手机壳和一个充电器。", + "evidence": [ + { + "description": "There is a mobile phone and a phone case placed on the table.", + "timestamps": { + "start": "00:58:25", + "end": "00:58:30" + } + } + ], + "question_timestamps": "00:58:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There was only a phone.", + "B": "There was only a charger.", + "C": "On the desk in the office in front of you, there was a phone, a phone case, and a charger.", + "D": "On the desk in the office in front of you, there was a phone and a charger, but no phone case.", + "E": "On the desk in the office in front of you, there was a phone case, but no phone or charger." + }, + "answer_label": "C", + "answer_text": "On the desk in the office in front of you, there was a phone, a phone case, and a charger.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_8_T2", + "correct_answer": "On the desk in the office in front of you, there was a phone, a phone case, and a charger.", + "original_options": { + "A": "There was only a phone.", + "B": "There was only a charger.", + "C": "On the desk in the office in front of you, there was a phone, a phone case, and a charger.", + "D": "On the desk in the office in front of you, there was a phone and a charger, but no phone case.", + "E": "On the desk in the office in front of you, there was a phone case, but no phone or charger." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 20 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_8_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, what phone-related item is in the office?", + "answer": "On the desk in the office in front of you, there was a phone case and a charger.", + "question_chinese": "我在现在办公室放着什么和手机有关的东西？", + "answer_chinese": "你前方办公室的桌子上放着一个手机壳和一个充电器。", + "evidence": [ + { + "description": "There is a mobile phone and a phone case placed on the table.", + "timestamps": { + "start": "00:58:25", + "end": "00:58:30" + } + }, + { + "description": "You took the mobile phone.", + "timestamps": { + "start": "01:01:54", + "end": "01:01:56" + } + } + ], + "question_timestamps": "01:06:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There was only a phone case.", + "B": "On the desk in the office in front of you, there was a phone case and a charger.", + "C": "There was only a charger.", + "D": "On the desk in the office in front of you, there was a phone and a phone case, but no charger.", + "E": "On the desk in the office in front of you, there was a phone, a phone case, and a charger all together." + }, + "answer_label": "B", + "answer_text": "On the desk in the office in front of you, there was a phone case and a charger.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_8_T1", + "correct_answer": "On the desk in the office in front of you, there was a phone case and a charger.", + "original_options": { + "A": "There was only a phone case.", + "B": "On the desk in the office in front of you, there was a phone case and a charger.", + "C": "There was only a charger.", + "D": "On the desk in the office in front of you, there was a phone and a phone case, but no charger.", + "E": "On the desk in the office in front of you, there was a phone, a phone case, and a charger all together." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 22 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_3_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How do I get back to the room I was just in?", + "answer": "Turn around now, then go through the narrow corridor to return to the room.", + "question_chinese": "我如何回到刚刚的房间？", + "answer_chinese": "你现在向后转，经过狭窄的走廊可以回到房间。", + "evidence": [ + { + "description": "You come out of the room.", + "timestamps": { + "start": "00:58:50", + "end": "00:59:00" + } + }, + { + "description": "You are standing at the door, with a narrow corridor behind you.", + "timestamps": { + "start": "00:59:02", + "end": "00:59:03" + } + } + ], + "question_timestamps": "00:59:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left to get back.", + "B": "Go straight down the corridor.", + "C": "Turn right first, then walk through the narrow corridor to reach the room you were just in.", + "D": "Turn around now, then go through the narrow corridor to return to the room.", + "E": "Keep moving forward, and when you reach the corridor, go through it to return to the previous room." + }, + "answer_label": "D", + "answer_text": "Turn around now, then go through the narrow corridor to return to the room.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_3_T2", + "correct_answer": "Turn around now, then go through the narrow corridor to return to the room.", + "original_options": { + "A": "Turn left to get back.", + "B": "Go straight down the corridor.", + "C": "Turn right first, then walk through the narrow corridor to reach the room you were just in.", + "D": "Turn around now, then go through the narrow corridor to return to the room.", + "E": "Keep moving forward, and when you reach the corridor, go through it to return to the previous room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_3_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How do I get back to the room I was just in?", + "answer": "Turn right, then turn left when you reach the main gate; go through the narrow corridor to get back to the room from before.", + "question_chinese": "我如何回到刚刚的房间？", + "answer_chinese": "向右转，到大门时左转，经过狭窄的走廊能到达刚才的房间。", + "evidence": [ + { + "description": "You come out of the room.", + "timestamps": { + "start": "00:58:50", + "end": "00:59:00" + } + }, + { + "description": "Turn right and you'll reach the lobby.", + "timestamps": { + "start": "00:59:10", + "end": "00:59:20" + } + } + ], + "question_timestamps": "00:59:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right, then turn right.", + "B": "Turn right, then turn left when you reach the main gate; go through the narrow corridor to get back to the room from before.", + "C": "Turn left, then turn left.", + "D": "Turn left first, then turn right at the main gate; go through the narrow corridor to return to the room.", + "E": "Turn right at the start, then go straight past the main gate without turning, and enter the narrow corridor." + }, + "answer_label": "B", + "answer_text": "Turn right, then turn left when you reach the main gate; go through the narrow corridor to get back to the room from before.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_3_T1", + "correct_answer": "Turn right, then turn left when you reach the main gate; go through the narrow corridor to get back to the room from before.", + "original_options": { + "A": "Turn right, then turn right.", + "B": "Turn right, then turn left when you reach the main gate; go through the narrow corridor to get back to the room from before.", + "C": "Turn left, then turn left.", + "D": "Turn left first, then turn right at the main gate; go through the narrow corridor to return to the room.", + "E": "Turn right at the start, then go straight past the main gate without turning, and enter the narrow corridor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 19 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_4", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I'm now facing the center of the square. What's behind me?", + "answer": "Behind you is the yellow land.", + "question_chinese": "我现在面对广场中央，我的后面是什么？", + "answer_chinese": "你的身后是黄色的土地。", + "evidence": [ + { + "description": "The direction you face when you turn around is this one.", + "timestamps": { + "start": "00:18:40", + "end": "00:18:50" + } + } + ], + "question_timestamps": "00:29:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind you is the red garden.", + "B": "Behind you is the brown hill.", + "C": "Behind you is the white field, not the yellow land area you were near.", + "D": "Behind you is the yellow land.", + "E": "Behind you is the orange path, stretching along the edge of the square." + }, + "answer_label": "D", + "answer_text": "Behind you is the yellow land.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_4", + "correct_answer": "Behind you is the yellow land.", + "original_options": { + "A": "Behind you is the red garden.", + "B": "Behind you is the brown hill.", + "C": "Behind you is the white field, not the yellow land area you were near.", + "D": "Behind you is the yellow land.", + "E": "Behind you is the orange path, stretching along the edge of the square." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_2_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I’m facing the center of the plaza right now—what is behind me?", + "answer": "Behind you is yellow ground.", + "question_chinese": "我现在面对广场中央，我的后面是什么？", + "answer_chinese": "你的身后是黄色的土地。", + "evidence": [ + { + "description": "The direction you face when you turn around is this one.", + "timestamps": { + "start": "00:18:40", + "end": "00:18:50" + } + } + ], + "question_timestamps": "00:37:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you is yellow ground.", + "B": "To your left is yellow ground.", + "C": "To your right is yellow ground, and behind you is a different colored surface.", + "D": "Above you is yellow ground, while behind you is not the yellow area you saw.", + "E": "Behind you is yellow ground." + }, + "answer_label": "E", + "answer_text": "Behind you is yellow ground.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_2_T1", + "correct_answer": "Behind you is yellow ground.", + "original_options": { + "A": "In front of you is yellow ground.", + "B": "To your left is yellow ground.", + "C": "To your right is yellow ground, and behind you is a different colored surface.", + "D": "Above you is yellow ground, while behind you is not the yellow area you saw.", + "E": "Behind you is yellow ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 15 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_2_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I’m facing the center of the plaza right now—what is behind me?", + "answer": "Behind you is a warehouse with a green roof.", + "question_chinese": "我现在面对广场中央，我的后面是什么？", + "answer_chinese": "你的身后是有绿色房顶的仓库。", + "evidence": [ + { + "description": "The direction you face when you turn around is this one.", + "timestamps": { + "start": "00:37:50", + "end": "00:37:54" + } + } + ], + "question_timestamps": "00:38:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you is that warehouse.", + "B": "Behind you is a warehouse with a green roof.", + "C": "To your left is the warehouse.", + "D": "To your right is a warehouse with a green roof, not directly behind you.", + "E": "Above you is a warehouse with a green roof, while behind you is something else entirely." + }, + "answer_label": "B", + "answer_text": "Behind you is a warehouse with a green roof.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_2_T2", + "correct_answer": "Behind you is a warehouse with a green roof.", + "original_options": { + "A": "In front of you is that warehouse.", + "B": "Behind you is a warehouse with a green roof.", + "C": "To your left is the warehouse.", + "D": "To your right is a warehouse with a green roof, not directly behind you.", + "E": "Above you is a warehouse with a green roof, while behind you is something else entirely." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 16 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_7_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the end of the video, what time did I see on the clock?", + "answer": "At that time, when you were standing in the middle of the area and looked at the time on your phone, it was about 8:37.", + "question_chinese": "视频结束的时候我看时钟显示为几点？", + "answer_chinese": "当时你手机上站在场地中央的时间大概是八点三十七分左右。", + "evidence": [ + { + "description": "The time was approximately 8:37 at that moment.", + "timestamps": { + "start": "00:59:29", + "end": "00:59:31" + } + } + ], + "question_timestamps": "00:59:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was about 9:37.", + "B": "At that time, when you were standing in the middle of the area and looked at the time on your phone, it was about 8:37.", + "C": "It was about 7:37.", + "D": "When you checked your phone in the middle of the area, it read about 10:37.", + "E": "Standing in the center and looking at your phone, you saw the time was about 6:37." + }, + "answer_label": "B", + "answer_text": "At that time, when you were standing in the middle of the area and looked at the time on your phone, it was about 8:37.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_7_T2", + "correct_answer": "At that time, when you were standing in the middle of the area and looked at the time on your phone, it was about 8:37.", + "original_options": { + "A": "It was about 9:37.", + "B": "At that time, when you were standing in the middle of the area and looked at the time on your phone, it was about 8:37.", + "C": "It was about 7:37.", + "D": "When you checked your phone in the middle of the area, it read about 10:37.", + "E": "Standing in the center and looking at your phone, you saw the time was about 6:37." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "4bf46557-3c5a-4a89-840d-158403590c9f", + "video_length": "01:08:46", + "qid": "4bf46557-3c5a-4a89-840d-158403590c9f_7_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the end of the video, what time did I see on the clock?", + "answer": "At that time, you were standing in the middle of the area, and when you checked the time on your phone it was about 8:45.", + "question_chinese": "视频结束的时候我看时钟显示为几点？", + "answer_chinese": "当时你站在场地中央看手机上的时间大概是八点四十五分左右。", + "evidence": [ + { + "description": "The time was approximately eight forty-five at that moment.", + "timestamps": { + "start": "01:08:35", + "end": "01:08:40" + } + } + ], + "question_timestamps": "01:08:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "At that time, you were standing in the middle of the area, and when you checked the time on your phone it was about 8:45.", + "B": "It was about 9:45.", + "C": "It was about 7:45.", + "D": "While standing in the middle of the area, you checked your phone and it showed about 10:45.", + "E": "At the end, in the center of the area, your phone displayed the time as about 6:45." + }, + "answer_label": "A", + "answer_text": "At that time, you were standing in the middle of the area, and when you checked the time on your phone it was about 8:45.", + "q_id": "4bf46557-3c5a-4a89-840d-158403590c9f_7_T1", + "correct_answer": "At that time, you were standing in the middle of the area, and when you checked the time on your phone it was about 8:45.", + "original_options": { + "A": "At that time, you were standing in the middle of the area, and when you checked the time on your phone it was about 8:45.", + "B": "It was about 9:45.", + "C": "It was about 7:45.", + "D": "While standing in the middle of the area, you checked your phone and it showed about 10:45.", + "E": "At the end, in the center of the area, your phone displayed the time as about 6:45." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + } +] diff --git a/QAs/4c45d8f4-e026-4ca3-a1e4-6d32da434469-mcq.json b/QAs/4c45d8f4-e026-4ca3-a1e4-6d32da434469-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..824f1ad7008c7b0ce15c2b594c158da911b1d288 --- /dev/null +++ b/QAs/4c45d8f4-e026-4ca3-a1e4-6d32da434469-mcq.json @@ -0,0 +1,1456 @@ +[ + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_4_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The fridge is in front of you.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱在你的前方。", + "evidence": [ + { + "description": "The refrigerator is on front of you.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:24" + } + } + ], + "question_timestamps": "00:02:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is behind you.", + "B": "The fridge is to your left.", + "C": "The fridge is to your right, not directly in front of where you are standing.", + "D": "The fridge is above you, rather than positioned straight ahead in your line of sight.", + "E": "The fridge is in front of you." + }, + "answer_label": "E", + "answer_text": "The fridge is in front of you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_4_T2", + "correct_answer": "The fridge is in front of you.", + "original_options": { + "A": "The fridge is behind you.", + "B": "The fridge is to your left.", + "C": "The fridge is to your right, not directly in front of where you are standing.", + "D": "The fridge is above you, rather than positioned straight ahead in your line of sight.", + "E": "The fridge is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_4_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The fridge is behind you.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱在你的后方。", + "evidence": [ + { + "description": "The refrigerator is behind you.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:24" + } + } + ], + "question_timestamps": "00:08:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is in front of you.", + "B": "The fridge is to your left.", + "C": "The fridge is behind you.", + "D": "The fridge is to your right, so it isn’t located directly behind you right now.", + "E": "The fridge is above you, instead of being behind you at your current position." + }, + "answer_label": "C", + "answer_text": "The fridge is behind you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_4_T1", + "correct_answer": "The fridge is behind you.", + "original_options": { + "A": "The fridge is in front of you.", + "B": "The fridge is to your left.", + "C": "The fridge is behind you.", + "D": "The fridge is to your right, so it isn’t located directly behind you right now.", + "E": "The fridge is above you, instead of being behind you at your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 14 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_6_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the microwave relative to me right now?", + "answer": "The microwave is in front of you.", + "question_chinese": "微波炉现在相对我是什么位置？", + "answer_chinese": "微波炉在你的前方。", + "evidence": [ + { + "description": "The microwave oven is on your right.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:17" + } + }, + { + "description": "The microwave oven is on front of you.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The microwave is behind you.", + "B": "The microwave is to your left.", + "C": "The microwave is to your right, just off to the side as you face forward.", + "D": "The microwave is above you, positioned higher up than your current eye level.", + "E": "The microwave is in front of you." + }, + "answer_label": "E", + "answer_text": "The microwave is in front of you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_6_T2", + "correct_answer": "The microwave is in front of you.", + "original_options": { + "A": "The microwave is behind you.", + "B": "The microwave is to your left.", + "C": "The microwave is to your right, just off to the side as you face forward.", + "D": "The microwave is above you, positioned higher up than your current eye level.", + "E": "The microwave is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 13 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_6_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the microwave relative to me right now?", + "answer": "The microwave is on the left side.", + "question_chinese": "微波炉现在相对我是什么位置？", + "answer_chinese": "微波炉在你的左侧。", + "evidence": [ + { + "description": "The microwave oven is on your left.", + "timestamps": { + "start": "00:15:32", + "end": "00:15:35" + } + } + ], + "question_timestamps": "00:15:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The microwave is in front of you.", + "B": "The microwave is behind you.", + "C": "The microwave is on the left side.", + "D": "The microwave is to your right, not on the left side where you’re standing.", + "E": "The microwave is directly ahead of you, lined up with your body rather than off to the side." + }, + "answer_label": "C", + "answer_text": "The microwave is on the left side.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_6_T1", + "correct_answer": "The microwave is on the left side.", + "original_options": { + "A": "The microwave is in front of you.", + "B": "The microwave is behind you.", + "C": "The microwave is on the left side.", + "D": "The microwave is to your right, not on the left side where you’re standing.", + "E": "The microwave is directly ahead of you, lined up with your body rather than off to the side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 18 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I just opened the refrigerator?", + "answer": "You just opened it twice.", + "question_chinese": "我刚刚打开了冰箱几次？", + "answer_chinese": "你刚刚打开了两次。", + "evidence": [ + { + "description": "You opened the refrigerator for the second time.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:56" + } + }, + { + "description": "You opened the refrigerator for the first time.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:12:14", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You just opened it once.", + "B": "You just opened it three times.", + "C": "You just opened it four times, opening and closing the door each time in quick succession.", + "D": "You just opened it twice.", + "E": "You just opened it five times, repeatedly checking inside before shutting it again." + }, + "answer_label": "D", + "answer_text": "You just opened it twice.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_9", + "correct_answer": "You just opened it twice.", + "original_options": { + "A": "You just opened it once.", + "B": "You just opened it three times.", + "C": "You just opened it four times, opening and closing the door each time in quick succession.", + "D": "You just opened it twice.", + "E": "You just opened it five times, repeatedly checking inside before shutting it again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 13 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_5_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the refrigerator just now?", + "answer": "You just opened it twice, and the fridge is behind you.", + "question_chinese": "我刚刚打开了冰箱几次？", + "answer_chinese": "你刚刚打开了两次，冰箱在你后面。", + "evidence": [ + { + "description": "You opened the refrigerator for the second time.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:56" + } + }, + { + "description": "You opened the refrigerator for the first time.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:27:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just opened it once.", + "B": "You just opened it three times.", + "C": "You just opened it four times, and the fridge is behind you as you finish up.", + "D": "You just opened it five times, and the fridge is behind you when you turn away.", + "E": "You just opened it twice, and the fridge is behind you." + }, + "answer_label": "E", + "answer_text": "You just opened it twice, and the fridge is behind you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_5_T2", + "correct_answer": "You just opened it twice, and the fridge is behind you.", + "original_options": { + "A": "You just opened it once.", + "B": "You just opened it three times.", + "C": "You just opened it four times, and the fridge is behind you as you finish up.", + "D": "You just opened it five times, and the fridge is behind you when you turn away.", + "E": "You just opened it twice, and the fridge is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_5_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the refrigerator just now?", + "answer": "You just opened it four times, and the fridge is behind you.", + "question_chinese": "我刚刚打开了冰箱几次？", + "answer_chinese": "你刚刚打开了四次，冰箱在你后面。", + "evidence": [ + { + "description": "You opened the refrigerator for the first time.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:56" + } + }, + { + "description": "You opened the refrigerator for the first time.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:24" + } + }, + { + "description": "You opened the refrigerator for the third time.", + "timestamps": { + "start": "00:34:10", + "end": "00:34:18" + } + }, + { + "description": "You opened the refrigerator for the four time.", + "timestamps": { + "start": "00:38:08", + "end": "00:38:12" + } + } + ], + "question_timestamps": "00:38:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just opened it twice.", + "B": "You just opened it three times.", + "C": "You just opened it five times, and the fridge is behind you after you’re done.", + "D": "You just opened it four times, and the fridge is behind you.", + "E": "You just opened it six times, and the fridge is behind you as you step away." + }, + "answer_label": "D", + "answer_text": "You just opened it four times, and the fridge is behind you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_5_T1", + "correct_answer": "You just opened it four times, and the fridge is behind you.", + "original_options": { + "A": "You just opened it twice.", + "B": "You just opened it three times.", + "C": "You just opened it five times, and the fridge is behind you after you’re done.", + "D": "You just opened it four times, and the fridge is behind you.", + "E": "You just opened it six times, and the fridge is behind you as you step away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put my mobile phone?", + "answer": "You just put your mobile phone on the table.", + "question_chinese": "我刚刚把手机放在哪里了？", + "answer_chinese": "你刚刚把手机放在桌子上。", + "evidence": [ + { + "description": "You just put your mobile phone on the table.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it in your backpack.", + "B": "You put it on the chair.", + "C": "You just put your mobile phone on the sofa, leaving it on the cushion instead of a table surface.", + "D": "You just put your mobile phone on the kitchen counter, setting it down away from the table.", + "E": "You just put your mobile phone on the table." + }, + "answer_label": "E", + "answer_text": "You just put your mobile phone on the table.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_1", + "correct_answer": "You just put your mobile phone on the table.", + "original_options": { + "A": "You put it in your backpack.", + "B": "You put it on the chair.", + "C": "You just put your mobile phone on the sofa, leaving it on the cushion instead of a table surface.", + "D": "You just put your mobile phone on the kitchen counter, setting it down away from the table.", + "E": "You just put your mobile phone on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 17 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put my mobile phone?", + "answer": "You just put your mobile phone on the table.", + "question_chinese": "我刚刚把手机放在哪里了？", + "answer_chinese": "你刚刚把手机放在桌上。", + "evidence": [ + { + "description": "You just put your mobile phone on the table.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:21:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just put your mobile phone on the table.", + "B": "You put it on the chair.", + "C": "You put it in your backpack.", + "D": "You just put your mobile phone on the sofa, placing it on a cushion rather than the table.", + "E": "You just put your mobile phone on the kitchen counter, setting it down off the table surface." + }, + "answer_label": "A", + "answer_text": "You just put your mobile phone on the table.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_2", + "correct_answer": "You just put your mobile phone on the table.", + "original_options": { + "A": "You just put your mobile phone on the table.", + "B": "You put it on the chair.", + "C": "You put it in your backpack.", + "D": "You just put your mobile phone on the sofa, placing it on a cushion rather than the table.", + "E": "You just put your mobile phone on the kitchen counter, setting it down off the table surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_17", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the chopping board after washing it?", + "answer": "You put the chopping board in the sink.", + "question_chinese": "我刚刚把砧板洗完后放在什么地方了？", + "answer_chinese": "你把砧板放在水池里了。", + "evidence": [ + { + "description": "You put the chopping board in the sink.", + "timestamps": { + "start": "00:38:03", + "end": "00:38:08" + } + } + ], + "question_timestamps": "00:38:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it on the counter.", + "B": "You put it in the drawer.", + "C": "You put the chopping board in the sink.", + "D": "You put the chopping board under the faucet, leaving it positioned beneath the running water.", + "E": "You put the chopping board next to the stove, setting it down beside the cooktop instead." + }, + "answer_label": "C", + "answer_text": "You put the chopping board in the sink.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_17", + "correct_answer": "You put the chopping board in the sink.", + "original_options": { + "A": "You put it on the counter.", + "B": "You put it in the drawer.", + "C": "You put the chopping board in the sink.", + "D": "You put the chopping board under the faucet, leaving it positioned beneath the running water.", + "E": "You put the chopping board next to the stove, setting it down beside the cooktop instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_18", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the chopping board after washing it?", + "answer": "You put the chopping board on the table.", + "question_chinese": "我刚刚把砧板洗完后放在什么地方了？", + "answer_chinese": "你把砧板放在桌子上了。", + "evidence": [ + { + "description": "You put the chopping board on the table.", + "timestamps": { + "start": "00:38:14", + "end": "00:38:16" + } + } + ], + "question_timestamps": "00:41:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the counter.", + "B": "In the cabinet.", + "C": "After washing it, you placed the chopping board on the countertop beside the sink to dry.", + "D": "You put the chopping board away in a cabinet rather than leaving it out on any surface.", + "E": "You put the chopping board on the table." + }, + "answer_label": "E", + "answer_text": "You put the chopping board on the table.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_18", + "correct_answer": "You put the chopping board on the table.", + "original_options": { + "A": "On the counter.", + "B": "In the cabinet.", + "C": "After washing it, you placed the chopping board on the countertop beside the sink to dry.", + "D": "You put the chopping board away in a cabinet rather than leaving it out on any surface.", + "E": "You put the chopping board on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_3_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator directly right now?", + "answer": "Right now you can directly touch the refrigerator; the refrigerator is to your left.", + "question_chinese": "我现在可以直接触碰到冰箱吗？", + "answer_chinese": "你现在可以直接触碰到冰箱,冰箱在你左边。", + "evidence": [ + { + "description": "You can now touch the refrigerator directly.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can’t directly touch the refrigerator; the refrigerator is to your left.", + "B": "Right now you can directly touch the refrigerator; the refrigerator is to your left." + }, + "answer_label": "B", + "answer_text": "Right now you can directly touch the refrigerator; the refrigerator is to your left.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_3_T1", + "correct_answer": "Right now you can directly touch the refrigerator; the refrigerator is to your left.", + "original_options": { + "A": "Right now you can’t directly touch the refrigerator; the refrigerator is to your left.", + "B": "Right now you can directly touch the refrigerator; the refrigerator is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_3_T2", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator directly right now?", + "answer": "Right now you can’t directly touch the refrigerator; the refrigerator is to your left.", + "question_chinese": "我现在可以直接触碰到冰箱吗？", + "answer_chinese": "你现在不可以直接触碰到冰箱，冰箱在你左边。", + "evidence": [ + { + "description": "You can not now touch the refrigerator directly.", + "timestamps": { + "start": "00:02:25", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can’t directly touch the refrigerator; the refrigerator is to your left.", + "B": "Right now you can directly touch the refrigerator; the refrigerator is to your left." + }, + "answer_label": "A", + "answer_text": "Right now you can’t directly touch the refrigerator; the refrigerator is to your left.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_3_T2", + "correct_answer": "Right now you can’t directly touch the refrigerator; the refrigerator is to your left.", + "original_options": { + "A": "Right now you can’t directly touch the refrigerator; the refrigerator is to your left.", + "B": "Right now you can directly touch the refrigerator; the refrigerator is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_5", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator directly now?", + "answer": "You can now touch the refrigerator directly.", + "question_chinese": "我现在可以直接触碰到冰箱吗？", + "answer_chinese": "你现在可以直接触碰到冰箱。", + "evidence": [ + { + "description": "You can now touch the refrigerator directly.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:24" + } + } + ], + "question_timestamps": "00:05:21", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now touch the refrigerator directly.", + "B": "You cannot touch the refrigerator directly now." + }, + "answer_label": "A", + "answer_text": "You can now touch the refrigerator directly.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_5", + "correct_answer": "You can now touch the refrigerator directly.", + "original_options": { + "A": "You can now touch the refrigerator directly.", + "B": "You cannot touch the refrigerator directly now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_6", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator directly now?", + "answer": "You can now touch the refrigerator directly.", + "question_chinese": "我现在可以直接触碰到冰箱吗？", + "answer_chinese": "你现在可以直接触碰到冰箱。", + "evidence": [ + { + "description": "You can now touch the refrigerator directly.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:24" + } + } + ], + "question_timestamps": "00:17:21", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now touch the refrigerator directly.", + "B": "You cannot touch the refrigerator directly now." + }, + "answer_label": "A", + "answer_text": "You can now touch the refrigerator directly.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_6", + "correct_answer": "You can now touch the refrigerator directly.", + "original_options": { + "A": "You can now touch the refrigerator directly.", + "B": "You cannot touch the refrigerator directly now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_7_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the stovetop directly right now?", + "answer": "Right now you can directly touch the stove; the stove is to your front-right.", + "question_chinese": "我现在可以直接触碰到灶台吗？", + "answer_chinese": "你现在可以直接触碰到灶台，灶台在你右前方。", + "evidence": [ + { + "description": "You can now touch the stove directly.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:16" + } + } + ], + "question_timestamps": "00:02:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can’t directly touch the stove; the stove is to your front-right.", + "B": "Right now you can directly touch the stove; the stove is to your front-right." + }, + "answer_label": "B", + "answer_text": "Right now you can directly touch the stove; the stove is to your front-right.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_7_T1", + "correct_answer": "Right now you can directly touch the stove; the stove is to your front-right.", + "original_options": { + "A": "Right now you can’t directly touch the stove; the stove is to your front-right.", + "B": "Right now you can directly touch the stove; the stove is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_7_T2", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the stovetop directly right now?", + "answer": "Right now you can’t directly touch the stove, and the stove is in front of you.", + "question_chinese": "我现在可以直接触碰到灶台吗？", + "answer_chinese": "你现在不可以直接触碰到灶台，灶台在你前方·。", + "evidence": [ + { + "description": "You can now touch the stove directly.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:16" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can’t directly touch the stove, and the stove is in front of you.", + "B": "Right now you can directly touch the stove, and the stove is in front of you." + }, + "answer_label": "A", + "answer_text": "Right now you can’t directly touch the stove, and the stove is in front of you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_7_T2", + "correct_answer": "Right now you can’t directly touch the stove, and the stove is in front of you.", + "original_options": { + "A": "Right now you can’t directly touch the stove, and the stove is in front of you.", + "B": "Right now you can directly touch the stove, and the stove is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_15", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the stove directly now?", + "answer": "You can now touch the stove directly.", + "question_chinese": "我现在可以直接触碰到灶台吗？", + "answer_chinese": "你现在可以直接触碰到灶台。", + "evidence": [ + { + "description": "You can now touch the stove directly.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:16" + } + } + ], + "question_timestamps": "00:19:54", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now touch the stove directly.", + "B": "You cannot touch the stove directly now." + }, + "answer_label": "A", + "answer_text": "You can now touch the stove directly.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_15", + "correct_answer": "You can now touch the stove directly.", + "original_options": { + "A": "You can now touch the stove directly.", + "B": "You cannot touch the stove directly now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_16", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the stove directly now?", + "answer": "You can now touch the stove directly.", + "question_chinese": "我现在可以直接触碰到灶台吗？", + "answer_chinese": "你现在可以直接触碰到灶台。", + "evidence": [ + { + "description": "You can now touch the stove directly.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:16" + } + } + ], + "question_timestamps": "00:31:54", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now touch the stove directly.", + "B": "You cannot touch the stove directly now." + }, + "answer_label": "A", + "answer_text": "You can now touch the stove directly.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_16", + "correct_answer": "You can now touch the stove directly.", + "original_options": { + "A": "You can now touch the stove directly.", + "B": "You cannot touch the stove directly now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_1_T4", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where is my phone?", + "answer": "Right now, your phone is in your hand, and it is in front of you.", + "question_chinese": "此时，我的手机在哪里？", + "answer_chinese": "此时，你的手机在你手上，手机在你前面。", + "evidence": [ + { + "description": "Your phone is on your hand.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s behind you in your hand.", + "B": "It’s to your left in your hand.", + "C": "Right now, your phone is in your hand, and it is in front of you.", + "D": "Your phone is in your hand right now, but it’s positioned behind you rather than in front.", + "E": "At the moment, you’re holding your phone, and it sits off to your right instead of straight ahead." + }, + "answer_label": "C", + "answer_text": "Right now, your phone is in your hand, and it is in front of you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_1_T4", + "correct_answer": "Right now, your phone is in your hand, and it is in front of you.", + "original_options": { + "A": "It’s behind you in your hand.", + "B": "It’s to your left in your hand.", + "C": "Right now, your phone is in your hand, and it is in front of you.", + "D": "Your phone is in your hand right now, but it’s positioned behind you rather than in front.", + "E": "At the moment, you’re holding your phone, and it sits off to your right instead of straight ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 19 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_1_T3", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where is my phone?", + "answer": "At this moment, your phone is on the stove, and the stove is in front of you.", + "question_chinese": "此时，我的手机在哪里？", + "answer_chinese": "此时，你的手机在灶台上，灶台在你前面。", + "evidence": [ + { + "description": "Your phone is on desk.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on the stove behind you.", + "B": "It’s on the stove to your left.", + "C": "At this moment, your phone is on the stove, and the stove is in front of you.", + "D": "Your phone is on the stove right now, but the stove is located behind you, not in front.", + "E": "At this moment, the phone rests on the stove, and the stove is positioned to your right rather than ahead." + }, + "answer_label": "C", + "answer_text": "At this moment, your phone is on the stove, and the stove is in front of you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_1_T3", + "correct_answer": "At this moment, your phone is on the stove, and the stove is in front of you.", + "original_options": { + "A": "It’s on the stove behind you.", + "B": "It’s on the stove to your left.", + "C": "At this moment, your phone is on the stove, and the stove is in front of you.", + "D": "Your phone is on the stove right now, but the stove is located behind you, not in front.", + "E": "At this moment, the phone rests on the stove, and the stove is positioned to your right rather than ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 20 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_1_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where is my phone?", + "answer": "You just put your phone on the table, and the phone is to your right.", + "question_chinese": "此时，我的手机在哪里？", + "answer_chinese": "你刚刚把手机放在桌上，手机在你右边。", + "evidence": [ + { + "description": "You just put your mobile phone on the table.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You just put your phone on the table, and the phone is to your right.", + "B": "It’s on the table to your left.", + "C": "It’s on the table behind you.", + "D": "You just set your phone on the table, but it’s positioned behind you rather than to your right.", + "E": "Right now, your phone is in your hand, and it’s held out in front of you instead of on the table." + }, + "answer_label": "A", + "answer_text": "You just put your phone on the table, and the phone is to your right.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_1_T2", + "correct_answer": "You just put your phone on the table, and the phone is to your right.", + "original_options": { + "A": "You just put your phone on the table, and the phone is to your right.", + "B": "It’s on the table to your left.", + "C": "It’s on the table behind you.", + "D": "You just set your phone on the table, but it’s positioned behind you rather than to your right.", + "E": "Right now, your phone is in your hand, and it’s held out in front of you instead of on the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 22 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_3", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What position is the sofa relative to me now?", + "answer": "The sofa is now in front of you.", + "question_chinese": "沙发现在相对我是什么位置？", + "answer_chinese": "沙发现在在你的前方。", + "evidence": [ + { + "description": "The sofa is now in front of you.", + "timestamps": { + "start": "00:03:16", + "end": "00:03:19" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The sofa is now behind you.", + "B": "The sofa is now to your left.", + "C": "The sofa is now in front of you.", + "D": "The sofa is now to your right, rather than being positioned directly in front of you.", + "E": "The sofa is now off to one side behind you, not straight ahead in your current view." + }, + "answer_label": "C", + "answer_text": "The sofa is now in front of you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_3", + "correct_answer": "The sofa is now in front of you.", + "original_options": { + "A": "The sofa is now behind you.", + "B": "The sofa is now to your left.", + "C": "The sofa is now in front of you.", + "D": "The sofa is now to your right, rather than being positioned directly in front of you.", + "E": "The sofa is now off to one side behind you, not straight ahead in your current view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_2_T1", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the sofa relative to me right now?", + "answer": "The sofa is now in front of you.", + "question_chinese": "沙发现在相对我是什么位置？", + "answer_chinese": "沙发现在在你的前方。", + "evidence": [ + { + "description": "The sofa is now in front of you.", + "timestamps": { + "start": "00:03:16", + "end": "00:03:19" + } + } + ], + "question_timestamps": "00:3:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sofa is now behind you.", + "B": "The sofa is now to your left.", + "C": "The sofa is now to your right, instead of being positioned directly in front of you.", + "D": "The sofa is now in front of you.", + "E": "The sofa is now above you, rather than sitting in front of you at your current position." + }, + "answer_label": "D", + "answer_text": "The sofa is now in front of you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_2_T1", + "correct_answer": "The sofa is now in front of you.", + "original_options": { + "A": "The sofa is now behind you.", + "B": "The sofa is now to your left.", + "C": "The sofa is now to your right, instead of being positioned directly in front of you.", + "D": "The sofa is now in front of you.", + "E": "The sofa is now above you, rather than sitting in front of you at your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_4", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What position is the sofa relative to me now?", + "answer": "The sofa is now in front of you.", + "question_chinese": "沙发现在相对我是什么位置？", + "answer_chinese": "沙发现在在你的前方。", + "evidence": [ + { + "description": "The sofa is now in front of you.", + "timestamps": { + "start": "00:05:56", + "end": "00:05:59" + } + } + ], + "question_timestamps": "00:16:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The sofa is now to your right.", + "B": "The sofa is now in front of you.", + "C": "The sofa is now behind you.", + "D": "The sofa is now beside you, rather than being positioned directly in front of you.", + "E": "The sofa is now across the room from you, not sitting in front of your current viewpoint." + }, + "answer_label": "B", + "answer_text": "The sofa is now in front of you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_4", + "correct_answer": "The sofa is now in front of you.", + "original_options": { + "A": "The sofa is now to your right.", + "B": "The sofa is now in front of you.", + "C": "The sofa is now behind you.", + "D": "The sofa is now beside you, rather than being positioned directly in front of you.", + "E": "The sofa is now across the room from you, not sitting in front of your current viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 17 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_2_T2", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the sofa relative to me right now?", + "answer": "The sofa is now behind you.", + "question_chinese": "沙发现在相对我是什么位置？", + "answer_chinese": "沙发现在在你的后方。", + "evidence": [ + { + "description": "The sofa is now behind you.", + "timestamps": { + "start": "00:34:10", + "end": "00:34:17" + } + } + ], + "question_timestamps": "00:34:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sofa is now in front of you.", + "B": "The sofa is now to your left.", + "C": "The sofa is now to your right, instead of being positioned behind you at the moment.", + "D": "The sofa is now above you, rather than located behind you relative to your current position.", + "E": "The sofa is now behind you." + }, + "answer_label": "E", + "answer_text": "The sofa is now behind you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_2_T2", + "correct_answer": "The sofa is now behind you.", + "original_options": { + "A": "The sofa is now in front of you.", + "B": "The sofa is now to your left.", + "C": "The sofa is now to your right, instead of being positioned behind you at the moment.", + "D": "The sofa is now above you, rather than located behind you relative to your current position.", + "E": "The sofa is now behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 16 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_9_T1", + "qid_num": 27, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the cutting board?", + "answer": "The cutting board is to your right.", + "question_chinese": "菜板在我的什么方向？", + "answer_chinese": "菜板在你的右边。", + "evidence": [ + { + "description": "The cutting board is on your right.", + "timestamps": { + "start": "00:30:11", + "end": "00:30:26" + } + } + ], + "question_timestamps": "00:30:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your left.", + "B": "It’s in front of you.", + "C": "The cutting board is behind you, not on your right side where you’d reach for it.", + "D": "The cutting board is above you, so you’d need to look upward rather than to the right.", + "E": "The cutting board is to your right." + }, + "answer_label": "E", + "answer_text": "The cutting board is to your right.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_9_T1", + "correct_answer": "The cutting board is to your right.", + "original_options": { + "A": "It’s to your left.", + "B": "It’s in front of you.", + "C": "The cutting board is behind you, not on your right side where you’d reach for it.", + "D": "The cutting board is above you, so you’d need to look upward rather than to the right.", + "E": "The cutting board is to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 18 + ] + }, + { + "video_uid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469", + "video_length": "00:42:47", + "qid": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_9_T2", + "qid_num": 28, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the cutting board?", + "answer": "The cutting board is behind you.", + "question_chinese": "菜板在我的什么方向？", + "answer_chinese": "菜板在你的后面。", + "evidence": [ + { + "description": "The cutting board is behind you.", + "timestamps": { + "start": "00:37:47", + "end": "00:37:50" + } + } + ], + "question_timestamps": "00:37:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s to your right.", + "C": "The cutting board is to your left, not behind you where it would be out of your view.", + "D": "The cutting board is above you, so you’d look up instead of turning around behind you.", + "E": "The cutting board is behind you." + }, + "answer_label": "E", + "answer_text": "The cutting board is behind you.", + "q_id": "4c45d8f4-e026-4ca3-a1e4-6d32da434469_9_T2", + "correct_answer": "The cutting board is behind you.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s to your right.", + "C": "The cutting board is to your left, not behind you where it would be out of your view.", + "D": "The cutting board is above you, so you’d look up instead of turning around behind you.", + "E": "The cutting board is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 17 + ] + } +] diff --git a/QAs/4d76c2c7-c6a2-46a6-bc94-0033c629f040-mcq.json b/QAs/4d76c2c7-c6a2-46a6-bc94-0033c629f040-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..63a49aa970dbbda6608e6970d855e62912f521fa --- /dev/null +++ b/QAs/4d76c2c7-c6a2-46a6-bc94-0033c629f040-mcq.json @@ -0,0 +1,221 @@ +[ + { + "video_uid": "4d76c2c7-c6a2-46a6-bc94-0033c629f040", + "video_length": "01:00:16", + "qid": "4d76c2c7-c6a2-46a6-bc94-0033c629f040_10", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What item have I put between my shoes and boots?", + "answer": "There is a blue plastic shoe cover placed between your shoes and boots.", + "question_chinese": "我的鞋和靴子中间垫了什么物品？", + "answer_chinese": "你的鞋和靴子中间垫了一个蓝色塑料鞋套。", + "evidence": [ + { + "description": "There is a blue plastic shoe cover placed between your shoes and boots.", + "timestamps": { + "start": "00:01:06", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:02:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A blue cloth shoe bag.", + "B": "A blue nylon shoelace.", + "C": "A red rubber mat is placed between your shoes and boots, not a blue plastic cover.", + "D": "There is a blue plastic shoe cover placed between your shoes and boots.", + "E": "A green cotton towel sits between your shoes and boots instead of the blue plastic shoe cover." + }, + "answer_label": "D", + "answer_text": "There is a blue plastic shoe cover placed between your shoes and boots.", + "q_id": "4d76c2c7-c6a2-46a6-bc94-0033c629f040_10", + "correct_answer": "There is a blue plastic shoe cover placed between your shoes and boots.", + "original_options": { + "A": "A blue cloth shoe bag.", + "B": "A blue nylon shoelace.", + "C": "A red rubber mat is placed between your shoes and boots, not a blue plastic cover.", + "D": "There is a blue plastic shoe cover placed between your shoes and boots.", + "E": "A green cotton towel sits between your shoes and boots instead of the blue plastic shoe cover." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 17 + ] + }, + { + "video_uid": "4d76c2c7-c6a2-46a6-bc94-0033c629f040", + "video_length": "01:00:16", + "qid": "4d76c2c7-c6a2-46a6-bc94-0033c629f040_11", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What item have I put between my shoes and boots?", + "answer": "You aren't wearing boots.", + "question_chinese": "我的鞋和靴子中间垫了什么物品？", + "answer_chinese": "你没有穿靴子。", + "evidence": [ + { + "description": "There is a blue plastic shoe cover placed between your shoes and boots.", + "timestamps": { + "start": "00:01:06", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:30:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You placed a hat there.", + "B": "You aren't wearing boots.", + "C": "You left sandals by your boots.", + "D": "You put a jacket beside your shoes and boots, so something is actually between them.", + "E": "You are wearing gloves instead of boots, implying boots are present even though that isn’t the case." + }, + "answer_label": "B", + "answer_text": "You aren't wearing boots.", + "q_id": "4d76c2c7-c6a2-46a6-bc94-0033c629f040_11", + "correct_answer": "You aren't wearing boots.", + "original_options": { + "A": "You placed a hat there.", + "B": "You aren't wearing boots.", + "C": "You left sandals by your boots.", + "D": "You put a jacket beside your shoes and boots, so something is actually between them.", + "E": "You are wearing gloves instead of boots, implying boots are present even though that isn’t the case." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 18 + ] + }, + { + "video_uid": "4d76c2c7-c6a2-46a6-bc94-0033c629f040", + "video_length": "01:00:16", + "qid": "4d76c2c7-c6a2-46a6-bc94-0033c629f040_6_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many lines did I draw on the concrete floor?", + "answer": "You drew 1 line on the cement. The line is to your right.", + "question_chinese": "我给水泥地画了几条线？", + "answer_chinese": "你在水泥上画了1条线。线在你的右边。", + "evidence": [ + { + "description": "You drew a line on the cement.", + "timestamps": { + "start": "00:55:50", + "end": "00:56:50" + } + } + ], + "question_timestamps": "00:57:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You drew 0 lines.", + "B": "You drew 2 lines.", + "C": "You drew 3 lines on the cement, and the lines are to your right side.", + "D": "You drew 1 line on the cement. The line is to your right.", + "E": "You drew 4 lines on the cement, with all of the lines located to your right." + }, + "answer_label": "D", + "answer_text": "You drew 1 line on the cement. The line is to your right.", + "q_id": "4d76c2c7-c6a2-46a6-bc94-0033c629f040_6_T2", + "correct_answer": "You drew 1 line on the cement. The line is to your right.", + "original_options": { + "A": "You drew 0 lines.", + "B": "You drew 2 lines.", + "C": "You drew 3 lines on the cement, and the lines are to your right side.", + "D": "You drew 1 line on the cement. The line is to your right.", + "E": "You drew 4 lines on the cement, with all of the lines located to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "4d76c2c7-c6a2-46a6-bc94-0033c629f040", + "video_length": "01:00:16", + "qid": "4d76c2c7-c6a2-46a6-bc94-0033c629f040_6_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many lines did I draw on the concrete floor?", + "answer": "You drew 2 lines on the cement. The lines were in front of you.", + "question_chinese": "我给水泥地画了几条线？", + "answer_chinese": "你在水泥上画了2条线。线在你的前面。", + "evidence": [ + { + "description": "You drew a line on the cement.", + "timestamps": { + "start": "00:55:50", + "end": "00:56:50" + } + }, + { + "description": "You drew a line on the cement.", + "timestamps": { + "start": "00:58:10", + "end": "00:59:20" + } + } + ], + "question_timestamps": "00:59:22", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You drew 0 lines.", + "B": "You drew 1 line.", + "C": "You drew 2 lines on the cement. The lines were in front of you.", + "D": "You drew 3 lines on the cement, and all of the lines were in front of you.", + "E": "You drew 4 lines on the cement, with the lines positioned in front of you." + }, + "answer_label": "C", + "answer_text": "You drew 2 lines on the cement. The lines were in front of you.", + "q_id": "4d76c2c7-c6a2-46a6-bc94-0033c629f040_6_T1", + "correct_answer": "You drew 2 lines on the cement. The lines were in front of you.", + "original_options": { + "A": "You drew 0 lines.", + "B": "You drew 1 line.", + "C": "You drew 2 lines on the cement. The lines were in front of you.", + "D": "You drew 3 lines on the cement, and all of the lines were in front of you.", + "E": "You drew 4 lines on the cement, with the lines positioned in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 15 + ] + } +] diff --git a/QAs/4d836cd7-d7fb-411e-bd50-79b92f5e5bfd-mcq.json b/QAs/4d836cd7-d7fb-411e-bd50-79b92f5e5bfd-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f97d1fa918e58d3c2fef9f976a495fa6726f54d8 --- /dev/null +++ b/QAs/4d836cd7-d7fb-411e-bd50-79b92f5e5bfd-mcq.json @@ -0,0 +1,1591 @@ +[ + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_18", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the mini Christmas tree from me when I stand in front of the sink?", + "answer": "When you stand in front of the sink, the mini Christmas tree is three meters away from you.", + "question_chinese": "当我站在水槽前面，迷你圣诞树离我有多远？", + "answer_chinese": "当你站在水槽前面，迷你圣诞树离你三米远。", + "evidence": [ + { + "description": "The mini Christmas tree is next to the sofa.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:01:44", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "It’s one meter away.", + "B": "It’s five meters away.", + "C": "When you stand in front of the sink, the mini Christmas tree is two meters away from you.", + "D": "When you stand in front of the sink, the mini Christmas tree is three meters away from you.", + "E": "When you stand in front of the sink, the mini Christmas tree is four meters away from you." + }, + "answer_label": "D", + "answer_text": "When you stand in front of the sink, the mini Christmas tree is three meters away from you.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_18", + "correct_answer": "When you stand in front of the sink, the mini Christmas tree is three meters away from you.", + "original_options": { + "A": "It’s one meter away.", + "B": "It’s five meters away.", + "C": "When you stand in front of the sink, the mini Christmas tree is two meters away from you.", + "D": "When you stand in front of the sink, the mini Christmas tree is three meters away from you.", + "E": "When you stand in front of the sink, the mini Christmas tree is four meters away from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_19", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the mini Christmas tree from me when I stand in front of the sink?", + "answer": "When you stand in front of the sink, the mini Christmas tree is three meters away from you.", + "question_chinese": "当我站在水槽前面，迷你圣诞树离我有多远？", + "answer_chinese": "当你站在水槽前面，迷你圣诞树离你三米远。", + "evidence": [ + { + "description": "The mini Christmas tree is next to the sofa.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:14:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "When you stand in front of the sink, the mini Christmas tree is three meters away from you.", + "B": "It’s one meter away.", + "C": "It’s five meters away.", + "D": "When you stand in front of the sink, the mini Christmas tree is on your left, about four meters away.", + "E": "When you stand in front of the sink, the mini Christmas tree is on your right, about six meters away." + }, + "answer_label": "A", + "answer_text": "When you stand in front of the sink, the mini Christmas tree is three meters away from you.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_19", + "correct_answer": "When you stand in front of the sink, the mini Christmas tree is three meters away from you.", + "original_options": { + "A": "When you stand in front of the sink, the mini Christmas tree is three meters away from you.", + "B": "It’s one meter away.", + "C": "It’s five meters away.", + "D": "When you stand in front of the sink, the mini Christmas tree is on your left, about four meters away.", + "E": "When you stand in front of the sink, the mini Christmas tree is on your right, about six meters away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 20 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_9", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I face the puzzle, can I reach the purple blanket?", + "answer": "When you face the puzzle, you can reach the purple blanket.", + "question_chinese": "当我面向拼图时，我能够到紫色毛毯吗？", + "answer_chinese": "当你面向拼图时，你能够到紫色毛毯。", + "evidence": [ + { + "description": "A purple blanket is placed on the sofa to the left of the jigsaw puzzle.", + "timestamps": { + "start": "00:10:44", + "end": "00:10:47" + } + } + ], + "question_timestamps": "00:12:37", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you face the puzzle, you can reach the purple blanket.", + "B": "When you face the puzzle, you cannot reach the purple blanket." + }, + "answer_label": "A", + "answer_text": "When you face the puzzle, you can reach the purple blanket.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_9", + "correct_answer": "When you face the puzzle, you can reach the purple blanket.", + "original_options": { + "A": "When you face the puzzle, you can reach the purple blanket.", + "B": "When you face the puzzle, you cannot reach the purple blanket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_10", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I face the puzzle, can I reach the purple blanket?", + "answer": "When you face the puzzle, you can reach the purple blanket.", + "question_chinese": "当我面向拼图时，我能够到紫色毛毯吗？", + "answer_chinese": "当你面向拼图时，你能够到紫色毛毯。", + "evidence": [ + { + "description": "A purple blanket is placed on the sofa to the left of the jigsaw puzzle.", + "timestamps": { + "start": "00:10:44", + "end": "00:10:47" + } + } + ], + "question_timestamps": "00:29:23", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you face the puzzle, you cannot reach the purple blanket.", + "B": "When you face the puzzle, you can reach the purple blanket." + }, + "answer_label": "B", + "answer_text": "When you face the puzzle, you can reach the purple blanket.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_10", + "correct_answer": "When you face the puzzle, you can reach the purple blanket.", + "original_options": { + "A": "When you face the puzzle, you cannot reach the purple blanket.", + "B": "When you face the puzzle, you can reach the purple blanket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_13", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I face the sink, am I far from/near the sofa?", + "answer": "When you face the sink, you are far from the sofa.", + "question_chinese": "当我面对水槽时，我离沙发远/近？", + "answer_chinese": "当你面对水槽时，你离沙发远。", + "evidence": [ + { + "description": "The sofa is four meters away from the sink.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:02:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "When you face the sink, you are far from the sofa.", + "B": "You are near the sofa.", + "C": "You are beside the sofa.", + "D": "When you face the sink, you are close enough to the sofa to reach it easily.", + "E": "When you face the sink, the sofa is nearby, only a few steps away from you." + }, + "answer_label": "A", + "answer_text": "When you face the sink, you are far from the sofa.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_13", + "correct_answer": "When you face the sink, you are far from the sofa.", + "original_options": { + "A": "When you face the sink, you are far from the sofa.", + "B": "You are near the sofa.", + "C": "You are beside the sofa.", + "D": "When you face the sink, you are close enough to the sofa to reach it easily.", + "E": "When you face the sink, the sofa is nearby, only a few steps away from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_14", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I face the sink, am I far from/near the sofa?", + "answer": "When you face the sink, you are far from the sofa.", + "question_chinese": "当我面对水槽时，我离沙发远/近？", + "answer_chinese": "当你面对水槽时，你离沙发远。", + "evidence": [ + { + "description": "The sofa is four meters away from the sink.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:20:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are near the sofa.", + "B": "You are beside the sofa.", + "C": "When you face the sink, you are close enough to the sofa that it feels within easy reach.", + "D": "When you face the sink, you are far from the sofa.", + "E": "When you face the sink, the sofa is nearby, just a short distance away from you." + }, + "answer_label": "D", + "answer_text": "When you face the sink, you are far from the sofa.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_14", + "correct_answer": "When you face the sink, you are far from the sofa.", + "original_options": { + "A": "You are near the sofa.", + "B": "You are beside the sofa.", + "C": "When you face the sink, you are close enough to the sofa that it feels within easy reach.", + "D": "When you face the sink, you are far from the sofa.", + "E": "When you face the sink, the sofa is nearby, just a short distance away from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 16 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_3_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want more forks, where should I go to get them?", + "answer": "If you want more forks, go get them from the storage rack next to the sink.", + "question_chinese": "我想要更多的叉子应该去哪里？", + "answer_chinese": "你想要更多的叉子应该去水槽旁边的收纳架那里拿。", + "evidence": [ + { + "description": "If you want more forks, you should go to the storage rack next to the sink and get them.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:45" + } + } + ], + "question_timestamps": "00:05:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go to the drawer behind you.", + "B": "Go to the rack under the sink.", + "C": "If you want more forks, go get them from the drawer in front of you instead.", + "D": "If you want more forks, go get them from the storage rack next to the sink.", + "E": "If you want more forks, go to the storage rack above the sink and take them from there." + }, + "answer_label": "D", + "answer_text": "If you want more forks, go get them from the storage rack next to the sink.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_3_T1", + "correct_answer": "If you want more forks, go get them from the storage rack next to the sink.", + "original_options": { + "A": "Go to the drawer behind you.", + "B": "Go to the rack under the sink.", + "C": "If you want more forks, go get them from the drawer in front of you instead.", + "D": "If you want more forks, go get them from the storage rack next to the sink.", + "E": "If you want more forks, go to the storage rack above the sink and take them from there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 18 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_3_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want more forks, where should I go to get them?", + "answer": "If you want more forks, go get them from the drawer behind you.", + "question_chinese": "我想要更多的叉子应该去哪里？", + "answer_chinese": "你想要更多的叉子应该去你身后的抽屉那里拿。", + "evidence": [ + { + "description": "You want more forks, you should get them from the drawer behind you.", + "timestamps": { + "start": "00:05:30", + "end": "00:06:21" + } + } + ], + "question_timestamps": "00:06:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go to the drawer in front of you.", + "B": "If you want more forks, go get them from the drawer behind you.", + "C": "Go to the drawer to your left.", + "D": "If you want more forks, go get them from the drawer to your right instead of behind you.", + "E": "If you want more forks, go get them from the drawer above you rather than the one behind you." + }, + "answer_label": "B", + "answer_text": "If you want more forks, go get them from the drawer behind you.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_3_T2", + "correct_answer": "If you want more forks, go get them from the drawer behind you.", + "original_options": { + "A": "Go to the drawer in front of you.", + "B": "If you want more forks, go get them from the drawer behind you.", + "C": "Go to the drawer to your left.", + "D": "If you want more forks, go get them from the drawer to your right instead of behind you.", + "E": "If you want more forks, go get them from the drawer above you rather than the one behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 19 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_5", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I go if I want more forks?", + "answer": "You should go to the storage rack under the sink to get more forks.", + "question_chinese": "我想要更多的叉子应该去哪里？", + "answer_chinese": "你想要更多的叉子应该去水槽下面的收纳架那里拿。", + "evidence": [ + { + "description": "The fork is placed on the storage rack under the sink.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:48" + } + } + ], + "question_timestamps": "00:08:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go to the cupboard above the sink.", + "B": "You should go to the storage rack under the sink to get more forks.", + "C": "Go to the drawer beside the stove.", + "D": "You should go to the shelf next to the refrigerator to get more forks instead.", + "E": "You should go to the storage rack inside the pantry to get more forks, not under the sink." + }, + "answer_label": "B", + "answer_text": "You should go to the storage rack under the sink to get more forks.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_5", + "correct_answer": "You should go to the storage rack under the sink to get more forks.", + "original_options": { + "A": "Go to the cupboard above the sink.", + "B": "You should go to the storage rack under the sink to get more forks.", + "C": "Go to the drawer beside the stove.", + "D": "You should go to the shelf next to the refrigerator to get more forks instead.", + "E": "You should go to the storage rack inside the pantry to get more forks, not under the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 18 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_6", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I go if I want more forks?", + "answer": "You should go to the storage rack under the sink to get more forks.", + "question_chinese": "我想要更多的叉子应该去哪里？", + "answer_chinese": "你想要更多的叉子应该去水槽下面的收纳架那里拿。", + "evidence": [ + { + "description": "The fork is placed on the storage rack under the sink.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:48" + } + } + ], + "question_timestamps": "00:27:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go to the drawer above the sink.", + "B": "Go to the storage rack above the sink.", + "C": "You should go to the pantry beside the sink to get more forks instead of under the sink.", + "D": "You should go to the storage rack under the sink to get more forks.", + "E": "You should go to the shelf next to the sink to get more forks, rather than the rack under it." + }, + "answer_label": "D", + "answer_text": "You should go to the storage rack under the sink to get more forks.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_6", + "correct_answer": "You should go to the storage rack under the sink to get more forks.", + "original_options": { + "A": "Go to the drawer above the sink.", + "B": "Go to the storage rack above the sink.", + "C": "You should go to the pantry beside the sink to get more forks instead of under the sink.", + "D": "You should go to the storage rack under the sink to get more forks.", + "E": "You should go to the shelf next to the sink to get more forks, rather than the rack under it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 20 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_16", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I get more seasoning?", + "answer": "You can get more seasonings from the left side of the stove.", + "question_chinese": "我想要更多的调味料要去哪里拿？", + "answer_chinese": "你想要更多的调味料要去灶台的左边拿。", + "evidence": [ + { + "description": "The seasoning on the left side of the stove.", + "timestamps": { + "start": "00:12:50", + "end": "00:12:55" + } + } + ], + "question_timestamps": "00:15:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the right side of the stove.", + "B": "From the back side of the stove.", + "C": "You can get more seasonings from the left side of the stove.", + "D": "You can get more seasonings from the front side of the stove, near the burners closest to you.", + "E": "Look on the top shelf above the stove to find the extra seasonings stored up there." + }, + "answer_label": "C", + "answer_text": "You can get more seasonings from the left side of the stove.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_16", + "correct_answer": "You can get more seasonings from the left side of the stove.", + "original_options": { + "A": "From the right side of the stove.", + "B": "From the back side of the stove.", + "C": "You can get more seasonings from the left side of the stove.", + "D": "You can get more seasonings from the front side of the stove, near the burners closest to you.", + "E": "Look on the top shelf above the stove to find the extra seasonings stored up there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 16 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_17", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I get more seasoning?", + "answer": "You can get more seasonings from the left side of the stove.", + "question_chinese": "我想要更多的调味料要去哪里拿？", + "answer_chinese": "你想要更多的调味料要去灶台的左边拿。", + "evidence": [ + { + "description": "The seasoning is on the left side of the stove.", + "timestamps": { + "start": "00:12:50", + "end": "00:12:55" + } + } + ], + "question_timestamps": "00:22:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the right side of the stove.", + "B": "You can get more seasonings from the left side of the stove.", + "C": "From the back side of the stove.", + "D": "You can get more seasonings from the top shelf above the stove, where they’re lined up together.", + "E": "Check the cabinet below the stove to grab more seasonings stored on the lower shelf." + }, + "answer_label": "B", + "answer_text": "You can get more seasonings from the left side of the stove.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_17", + "correct_answer": "You can get more seasonings from the left side of the stove.", + "original_options": { + "A": "From the right side of the stove.", + "B": "You can get more seasonings from the left side of the stove.", + "C": "From the back side of the stove.", + "D": "You can get more seasonings from the top shelf above the stove, where they’re lined up together.", + "E": "Check the cabinet below the stove to grab more seasonings stored on the lower shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 15 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_5_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What furniture is in my living room?", + "answer": "There are two sofas in your living room: one is in front of you, and the other is to your left.", + "question_chinese": "我的客厅都有什么家具？", + "answer_chinese": "你的客厅有两张沙发，一张在你的正前方，另一张在你的左边。", + "evidence": [ + { + "description": "There are two sofas in your living room.", + "timestamps": { + "start": "00:10:44", + "end": "00:10:47" + } + } + ], + "question_timestamps": "00:10:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There are two sofas in your living room: one is in front of you, and the other is to your left.", + "B": "There are two tables in your living room: one is in front of you, and the other is to your right." + }, + "answer_label": "A", + "answer_text": "There are two sofas in your living room: one is in front of you, and the other is to your left.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_5_T1", + "correct_answer": "There are two sofas in your living room: one is in front of you, and the other is to your left.", + "original_options": { + "A": "There are two sofas in your living room: one is in front of you, and the other is to your left.", + "B": "There are two tables in your living room: one is in front of you, and the other is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 21 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_5_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What furniture is in my living room?", + "answer": "There are two tables in your living room: one is in front of you, and the other is to your right.", + "question_chinese": "我的客厅都有什么家具？", + "answer_chinese": "你的客厅有两张桌子。一张在你的正前方，另一张在你的右边。", + "evidence": [ + { + "description": "There are two tables in your living room.", + "timestamps": { + "start": "00:10:44", + "end": "00:10:47" + } + } + ], + "question_timestamps": "00:10:57", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There are two sofas in your living room: one is in front of you, and the other is to your left.", + "B": "There are two tables in your living room: one is in front of you, and the other is to your right." + }, + "answer_label": "B", + "answer_text": "There are two tables in your living room: one is in front of you, and the other is to your right.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_5_T2", + "correct_answer": "There are two tables in your living room: one is in front of you, and the other is to your right.", + "original_options": { + "A": "There are two sofas in your living room: one is in front of you, and the other is to your left.", + "B": "There are two tables in your living room: one is in front of you, and the other is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 21 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_10_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my microwave?", + "answer": "Your microwave is in front of you.", + "question_chinese": "我的微波炉在哪里？", + "answer_chinese": "你的微波炉在你的前面。", + "evidence": [ + { + "description": "Your microwave oven is in front of you", + "timestamps": { + "start": "00:00:32", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your microwave is in front of you.", + "B": "Your microwave is behind you.", + "C": "Your microwave is to your left.", + "D": "Your microwave is to your right, positioned off to the side rather than directly ahead.", + "E": "Your microwave is above you, mounted higher up instead of sitting at your eye level." + }, + "answer_label": "A", + "answer_text": "Your microwave is in front of you.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_10_T2", + "correct_answer": "Your microwave is in front of you.", + "original_options": { + "A": "Your microwave is in front of you.", + "B": "Your microwave is behind you.", + "C": "Your microwave is to your left.", + "D": "Your microwave is to your right, positioned off to the side rather than directly ahead.", + "E": "Your microwave is above you, mounted higher up instead of sitting at your eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_10_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my microwave?", + "answer": "Your microwave is behind you.", + "question_chinese": "我的微波炉在哪里？", + "answer_chinese": "你的微波炉在你的后面。", + "evidence": [ + { + "description": "Your microwave oven is behind you.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:13:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your microwave is in front of you.", + "B": "Your microwave is behind you.", + "C": "Your microwave is to your left.", + "D": "Your microwave is to your right, not behind you, so you’d turn to reach it.", + "E": "Your microwave is above you, placed higher up rather than located behind your position." + }, + "answer_label": "B", + "answer_text": "Your microwave is behind you.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_10_T1", + "correct_answer": "Your microwave is behind you.", + "original_options": { + "A": "Your microwave is in front of you.", + "B": "Your microwave is behind you.", + "C": "Your microwave is to your left.", + "D": "Your microwave is to your right, not behind you, so you’d turn to reach it.", + "E": "Your microwave is above you, placed higher up rather than located behind your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 14 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_12", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I stood next to the dishwasher. How many doors do I need to pass through to get to the bedroom?", + "answer": "You stand next to the dishwasher, and it takes going through seven doors to get to the bedroom.", + "question_chinese": "我站在洗碗机旁边，去卧室需要经过几道门？", + "answer_chinese": "你站在洗碗机旁边，去卧室需要经过七道门。", + "evidence": [ + { + "description": "There are two doors at the corner of the corridor.", + "timestamps": { + "start": "00:08:08", + "end": "00:08:11" + } + }, + { + "description": "There are five doors on both sides of the corridor.", + "timestamps": { + "start": "00:08:22", + "end": "00:08:30" + } + } + ], + "question_timestamps": "00:10:22", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Through five doors.", + "B": "You stand next to the dishwasher, and it takes going through seven doors to get to the bedroom.", + "C": "Through nine doors.", + "D": "Starting by the dishwasher, you only need to pass through six doors to reach the bedroom.", + "E": "From where you're standing by the dishwasher, it takes eight doors before you enter the bedroom." + }, + "answer_label": "B", + "answer_text": "You stand next to the dishwasher, and it takes going through seven doors to get to the bedroom.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_12", + "correct_answer": "You stand next to the dishwasher, and it takes going through seven doors to get to the bedroom.", + "original_options": { + "A": "Through five doors.", + "B": "You stand next to the dishwasher, and it takes going through seven doors to get to the bedroom.", + "C": "Through nine doors.", + "D": "Starting by the dishwasher, you only need to pass through six doors to reach the bedroom.", + "E": "From where you're standing by the dishwasher, it takes eight doors before you enter the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_11", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I stood next to the dishwasher. How many doors do I need to pass through to get to the bedroom?", + "answer": "You stand next to the dishwasher, and it takes going through seven doors to get to the bedroom.", + "question_chinese": "我站在洗碗机旁边，去卧室需要经过几道门？", + "answer_chinese": "你站在洗碗机旁边，去卧室需要经过七道门。", + "evidence": [ + { + "description": "There are two doors at the corner of the corridor.", + "timestamps": { + "start": "00:08:08", + "end": "00:08:11" + } + }, + { + "description": "There are five doors on both sides of the corridor.", + "timestamps": { + "start": "00:08:22", + "end": "00:08:30" + } + } + ], + "question_timestamps": "00:22:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Through two doors.", + "B": "Through ten doors.", + "C": "From beside the dishwasher, you pass through five doors before arriving at the bedroom.", + "D": "Standing next to the dishwasher, you need to go through nine doors to reach the bedroom.", + "E": "You stand next to the dishwasher, and it takes going through seven doors to get to the bedroom." + }, + "answer_label": "E", + "answer_text": "You stand next to the dishwasher, and it takes going through seven doors to get to the bedroom.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_11", + "correct_answer": "You stand next to the dishwasher, and it takes going through seven doors to get to the bedroom.", + "original_options": { + "A": "Through two doors.", + "B": "Through ten doors.", + "C": "From beside the dishwasher, you pass through five doors before arriving at the bedroom.", + "D": "Standing next to the dishwasher, you need to go through nine doors to reach the bedroom.", + "E": "You stand next to the dishwasher, and it takes going through seven doors to get to the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 16 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_3", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls are there in the cabinet?", + "answer": "There are twelve bowls in the cabinet.", + "question_chinese": "柜子里有几个碗？", + "answer_chinese": "柜子里有十二个碗。", + "evidence": [ + { + "description": "I took two bowls from the dishwasher and put them in the cabinet.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:26" + } + }, + { + "description": "There are twelve bowls in the cabinet.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:05:50", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are ten bowls.", + "B": "There are eight bowls.", + "C": "In the cabinet, you can count nine bowls in total from the ones stacked inside.", + "D": "The cabinet holds fourteen bowls when you count them all on the shelf.", + "E": "There are twelve bowls in the cabinet." + }, + "answer_label": "E", + "answer_text": "There are twelve bowls in the cabinet.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_3", + "correct_answer": "There are twelve bowls in the cabinet.", + "original_options": { + "A": "There are ten bowls.", + "B": "There are eight bowls.", + "C": "In the cabinet, you can count nine bowls in total from the ones stacked inside.", + "D": "The cabinet holds fourteen bowls when you count them all on the shelf.", + "E": "There are twelve bowls in the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 13 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_4", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls are there in the cabinet?", + "answer": "There are fourteen bowls in the cabinet.", + "question_chinese": "柜子里有几个碗？", + "answer_chinese": "柜子里有十四个碗。", + "evidence": [ + { + "description": "There are twelve bowls in the cabinet.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:25" + } + }, + { + "description": "I took two bowls from the dishwasher and put them in the cabinet.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:26" + } + } + ], + "question_timestamps": "00:19:49", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are eleven bowls.", + "B": "There are eight bowls.", + "C": "In the cabinet, there are twelve bowls altogether when you count each one carefully.", + "D": "There are fourteen bowls in the cabinet.", + "E": "You can count twenty bowls inside the cabinet when you look over every shelf." + }, + "answer_label": "D", + "answer_text": "There are fourteen bowls in the cabinet.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_4", + "correct_answer": "There are fourteen bowls in the cabinet.", + "original_options": { + "A": "There are eleven bowls.", + "B": "There are eight bowls.", + "C": "In the cabinet, there are twelve bowls altogether when you count each one carefully.", + "D": "There are fourteen bowls in the cabinet.", + "E": "You can count twenty bowls inside the cabinet when you look over every shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_1", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates are in the cupboard?", + "answer": "There are five plates in the cupboard.", + "question_chinese": "柜子里有几个碟子？", + "answer_chinese": "柜子里有五个碟子。", + "evidence": [ + { + "description": "There are five plates in the cupboard.", + "timestamps": { + "start": "00:01:14", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are five plates in the cupboard.", + "B": "There are two plates.", + "C": "There are seven plates.", + "D": "The cupboard contains four plates when you count them one by one inside.", + "E": "Inside the cupboard, there are eight plates stacked together in a single pile." + }, + "answer_label": "A", + "answer_text": "There are five plates in the cupboard.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_1", + "correct_answer": "There are five plates in the cupboard.", + "original_options": { + "A": "There are five plates in the cupboard.", + "B": "There are two plates.", + "C": "There are seven plates.", + "D": "The cupboard contains four plates when you count them one by one inside.", + "E": "Inside the cupboard, there are eight plates stacked together in a single pile." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 13 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_2", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates are in the cupboard?", + "answer": "There are fifteen plates in the cupboard.", + "question_chinese": "柜子里有几个碟子？", + "answer_chinese": "柜子里有十五个碟子。", + "evidence": [ + { + "description": "There are five plates in the cupboard.", + "timestamps": { + "start": "00:01:14", + "end": "00:01:17" + } + }, + { + "description": "I took a total of ten plates from the dishwasher and put them in the cabinet.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:16:18", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are ten plates.", + "B": "There are twenty plates.", + "C": "There are fifteen plates in the cupboard.", + "D": "In the cupboard, you can count thirteen plates total when you check each stack inside.", + "E": "The cupboard has sixteen plates when you count everything inside, including the plates at the back." + }, + "answer_label": "C", + "answer_text": "There are fifteen plates in the cupboard.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_2", + "correct_answer": "There are fifteen plates in the cupboard.", + "original_options": { + "A": "There are ten plates.", + "B": "There are twenty plates.", + "C": "There are fifteen plates in the cupboard.", + "D": "In the cupboard, you can count thirteen plates total when you check each stack inside.", + "E": "The cupboard has sixteen plates when you count everything inside, including the plates at the back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_6_T1", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the refrigerator relative to me right now?", + "answer": "Right now, the fridge is in front of you.", + "question_chinese": "现在，冰箱在我的哪里？", + "answer_chinese": "现在，冰箱在你的前面。", + "evidence": [ + { + "description": "Now the refrigerator is in front of you.", + "timestamps": { + "start": "00:01:53", + "end": "00:01:54" + } + } + ], + "question_timestamps": "00:01:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "Right now, the fridge is in front of you.", + "D": "Right now, the fridge is positioned to your right side, not directly ahead of you.", + "E": "At the moment, the fridge is above you rather than in front." + }, + "answer_label": "C", + "answer_text": "Right now, the fridge is in front of you.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_6_T1", + "correct_answer": "Right now, the fridge is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "Right now, the fridge is in front of you.", + "D": "Right now, the fridge is positioned to your right side, not directly ahead of you.", + "E": "At the moment, the fridge is above you rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 12 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_6_T2", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the refrigerator relative to me right now?", + "answer": "Right now, the fridge is behind you.", + "question_chinese": "现在，冰箱在我的哪里？", + "answer_chinese": "现在，冰箱在你的后面。", + "evidence": [ + { + "description": "Now the refrigerator is behind you.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:02:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In front of you.", + "B": "To your right.", + "C": "Right now, the fridge is behind you.", + "D": "Right now, the fridge is positioned to your left side, not directly behind you.", + "E": "At the moment, the fridge is above you rather than behind, near the upper cabinets." + }, + "answer_label": "C", + "answer_text": "Right now, the fridge is behind you.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_6_T2", + "correct_answer": "Right now, the fridge is behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your right.", + "C": "Right now, the fridge is behind you.", + "D": "Right now, the fridge is positioned to your left side, not directly behind you.", + "E": "At the moment, the fridge is above you rather than behind, near the upper cabinets." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 15 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_4_T2", + "qid_num": 27, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can relative to me right now?", + "answer": "Right now, the trash can is to your left.", + "question_chinese": "现在，垃圾桶在我的哪里？", + "answer_chinese": "现在，垃圾桶在你的左边。", + "evidence": [ + { + "description": "The trash can is to the left of the stove.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:04" + } + } + ], + "question_timestamps": "00:05:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right now, the trash can is to your left.", + "B": "The trash can is in front.", + "C": "The trash can is behind you.", + "D": "Right now, the trash can is to your right, just off your shoulder as you face forward.", + "E": "Right now, the trash can is above you, positioned higher than eye level rather than beside you." + }, + "answer_label": "A", + "answer_text": "Right now, the trash can is to your left.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_4_T2", + "correct_answer": "Right now, the trash can is to your left.", + "original_options": { + "A": "Right now, the trash can is to your left.", + "B": "The trash can is in front.", + "C": "The trash can is behind you.", + "D": "Right now, the trash can is to your right, just off your shoulder as you face forward.", + "E": "Right now, the trash can is above you, positioned higher than eye level rather than beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_4_T1", + "qid_num": 28, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can relative to me right now?", + "answer": "Right now, the trash can is behind you.", + "question_chinese": "现在，垃圾桶在我的哪里？", + "answer_chinese": "现在，垃圾桶在你的后面。", + "evidence": [ + { + "description": "The trash can is behind you.", + "timestamps": { + "start": "00:05:15", + "end": "00:05:16" + } + } + ], + "question_timestamps": "00:05:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can is in front.", + "B": "The trash can is to your left.", + "C": "Right now, the trash can is behind you.", + "D": "Right now, the trash can is to your right, slightly off to the side as you look ahead.", + "E": "Right now, the trash can is above you, not at your back where you’d expect it." + }, + "answer_label": "C", + "answer_text": "Right now, the trash can is behind you.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_4_T1", + "correct_answer": "Right now, the trash can is behind you.", + "original_options": { + "A": "The trash can is in front.", + "B": "The trash can is to your left.", + "C": "Right now, the trash can is behind you.", + "D": "Right now, the trash can is to your right, slightly off to the side as you look ahead.", + "E": "Right now, the trash can is above you, not at your back where you’d expect it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 17 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_7_T1", + "qid_num": 29, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, am I close to the sofa or far from it?", + "answer": "Right now, you’re far from the sofa; the sofa is to your front-right.", + "question_chinese": "现在，我离沙发远/近？", + "answer_chinese": "现在，你离沙发远，沙发在你的右前方。", + "evidence": [ + { + "description": "The sofa is four meters away from the sink.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’re close to the sofa.", + "B": "Right now, you’re far from the sofa; the sofa is to your front-right.", + "C": "The sofa is to your front-left.", + "D": "Right now, you’re far from the sofa; the sofa sits behind you on your left side.", + "E": "Right now, you’re close to the sofa; it’s positioned to your front-right within easy reaching distance." + }, + "answer_label": "B", + "answer_text": "Right now, you’re far from the sofa; the sofa is to your front-right.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_7_T1", + "correct_answer": "Right now, you’re far from the sofa; the sofa is to your front-right.", + "original_options": { + "A": "You’re close to the sofa.", + "B": "Right now, you’re far from the sofa; the sofa is to your front-right.", + "C": "The sofa is to your front-left.", + "D": "Right now, you’re far from the sofa; the sofa sits behind you on your left side.", + "E": "Right now, you’re close to the sofa; it’s positioned to your front-right within easy reaching distance." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_7_T2", + "qid_num": 30, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, am I close to the sofa or far from it?", + "answer": "Right now, you’re close to the sofas; there are sofas on both your left and right.", + "question_chinese": "现在，我离沙发远/近？", + "answer_chinese": "现在，你离沙发近，沙发在你的左右两边都有。", + "evidence": [ + { + "description": "Now you are near the sofa.", + "timestamps": { + "start": "00:10:43", + "end": "00:10:44" + } + } + ], + "question_timestamps": "00:10:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’re far from the sofas.", + "B": "The sofas are in front of you.", + "C": "Right now, you’re close to the sofas; there are sofas on both your left and right.", + "D": "Right now, you’re close to the sofas; there are sofas in front of you and behind you.", + "E": "Right now, you’re close to the sofas; both sofas are on your right side rather than split left and right." + }, + "answer_label": "C", + "answer_text": "Right now, you’re close to the sofas; there are sofas on both your left and right.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_7_T2", + "correct_answer": "Right now, you’re close to the sofas; there are sofas on both your left and right.", + "original_options": { + "A": "You’re far from the sofas.", + "B": "The sofas are in front of you.", + "C": "Right now, you’re close to the sofas; there are sofas on both your left and right.", + "D": "Right now, you’re close to the sofas; there are sofas in front of you and behind you.", + "E": "Right now, you’re close to the sofas; both sofas are on your right side rather than split left and right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 21 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_9_T2", + "qid_num": 31, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, is the mini Christmas tree close to me or far from me?", + "answer": "Right now, the mini Christmas tree is far from you, and it’s to your right.", + "question_chinese": "现在，迷你圣诞树离我远/近？", + "answer_chinese": "现在，迷你圣诞树离你远，迷你圣诞树在你的右边。", + "evidence": [ + { + "description": "Now the mini Christmas tree is far away from you.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It’s close to you.", + "B": "It’s to your left.", + "C": "Right now, the mini Christmas tree is far from you, and it’s to your left.", + "D": "Right now, the mini Christmas tree is close to you, and it’s to your right within arm’s reach.", + "E": "Right now, the mini Christmas tree is far from you, and it’s to your right." + }, + "answer_label": "E", + "answer_text": "Right now, the mini Christmas tree is far from you, and it’s to your right.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_9_T2", + "correct_answer": "Right now, the mini Christmas tree is far from you, and it’s to your right.", + "original_options": { + "A": "It’s close to you.", + "B": "It’s to your left.", + "C": "Right now, the mini Christmas tree is far from you, and it’s to your left.", + "D": "Right now, the mini Christmas tree is close to you, and it’s to your right within arm’s reach.", + "E": "Right now, the mini Christmas tree is far from you, and it’s to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 20 + ] + }, + { + "video_uid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd", + "video_length": "00:30:00", + "qid": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_9_T1", + "qid_num": 32, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, is the mini Christmas tree close to me or far from me?", + "answer": "Right now, the mini Christmas tree is close to you, and it’s to your left.", + "question_chinese": "现在，迷你圣诞树离我远/近？", + "answer_chinese": "现在，迷你圣诞树离你近，迷你圣诞树在你的左边。", + "evidence": [ + { + "description": "Now the mini Christmas tree is away from you", + "timestamps": { + "start": "00:10:27", + "end": "00:10:28" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It’s far from you.", + "B": "It’s to your right.", + "C": "Right now, the mini Christmas tree is close to you, and it’s to your right.", + "D": "Right now, the mini Christmas tree is close to you, and it’s to your left.", + "E": "Right now, the mini Christmas tree is far from you, and it’s to your left across the room." + }, + "answer_label": "D", + "answer_text": "Right now, the mini Christmas tree is close to you, and it’s to your left.", + "q_id": "4d836cd7-d7fb-411e-bd50-79b92f5e5bfd_9_T1", + "correct_answer": "Right now, the mini Christmas tree is close to you, and it’s to your left.", + "original_options": { + "A": "It’s far from you.", + "B": "It’s to your right.", + "C": "Right now, the mini Christmas tree is close to you, and it’s to your right.", + "D": "Right now, the mini Christmas tree is close to you, and it’s to your left.", + "E": "Right now, the mini Christmas tree is far from you, and it’s to your left across the room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 19 + ] + } +] diff --git a/QAs/4e1b12cd-710a-4979-8d91-a15424cb1fd6-mcq.json b/QAs/4e1b12cd-710a-4979-8d91-a15424cb1fd6-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a6d1d15d783211e87a5acccbece6e3e1e6b7a324 --- /dev/null +++ b/QAs/4e1b12cd-710a-4979-8d91-a15424cb1fd6-mcq.json @@ -0,0 +1,779 @@ +[ + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_5", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many paintings have I drawn in total?", + "answer": "You have drawn two paintings.", + "question_chinese": "我一共画了多少幅画呢?", + "answer_chinese": "你画了两幅画。", + "evidence": [ + { + "description": "You draw the first picture.", + "timestamps": { + "start": "00:00:48", + "end": "00:02:38" + } + }, + { + "description": "You drew the second picture.", + "timestamps": { + "start": "00:03:31", + "end": "00:06:02" + } + } + ], + "question_timestamps": "00:06:17", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have drawn one painting.", + "B": "You have drawn three paintings.", + "C": "You have drawn four paintings in total, not two as you might be thinking.", + "D": "You have drawn five paintings overall, counting everything you completed from start to finish.", + "E": "You have drawn two paintings." + }, + "answer_label": "E", + "answer_text": "You have drawn two paintings.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_5", + "correct_answer": "You have drawn two paintings.", + "original_options": { + "A": "You have drawn one painting.", + "B": "You have drawn three paintings.", + "C": "You have drawn four paintings in total, not two as you might be thinking.", + "D": "You have drawn five paintings overall, counting everything you completed from start to finish.", + "E": "You have drawn two paintings." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_6", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many paintings have I drawn in total?", + "answer": "You drew four pictures.", + "question_chinese": "我一共画了多少幅画呢?", + "answer_chinese": "你画了四幅画", + "evidence": [ + { + "description": "You draw the first picture.", + "timestamps": { + "start": "00:00:48", + "end": "00:02:38" + } + }, + { + "description": "You drew the second picture.", + "timestamps": { + "start": "00:03:31", + "end": "00:06:02" + } + }, + { + "description": "You draw the third picture.", + "timestamps": { + "start": "00:13:51", + "end": "00:16:13" + } + }, + { + "description": "You drew the fourth picture.", + "timestamps": { + "start": "00:15:15", + "end": "00:15:51" + } + } + ], + "question_timestamps": "00:31:02", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You drew three pictures.", + "B": "You drew four pictures.", + "C": "You drew five pictures.", + "D": "You drew one picture, not four, based on what you completed in the session.", + "E": "You drew seven pictures in total, including all the ones you finished." + }, + "answer_label": "B", + "answer_text": "You drew four pictures.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_6", + "correct_answer": "You drew four pictures.", + "original_options": { + "A": "You drew three pictures.", + "B": "You drew four pictures.", + "C": "You drew five pictures.", + "D": "You drew one picture, not four, based on what you completed in the session.", + "E": "You drew seven pictures in total, including all the ones you finished." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 12 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_4_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "For the cards in front of me: where is the blue card with a cup, relative to the yellow card with a cactus?", + "answer": "The blue card with a cup on it in front of you is to the left of the yellow card with a cactus on it.", + "question_chinese": "我面前的画有杯子的蓝色卡片在相对于画有仙人掌的黄色卡片的什么位置?", + "answer_chinese": "你面前的画有杯子的蓝色卡片在画有仙人掌的黄色卡片的左边。", + "evidence": [ + { + "description": "The blue card with a cup is on the left of the yellow card with a cactus.", + "timestamps": { + "start": "00:01:53", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:02:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to the right.", + "B": "The blue card with a cup on it in front of you is to the left of the yellow card with a cactus on it.", + "C": "It is directly above.", + "D": "The blue cup card is to the right of the yellow cactus card, not on its left side.", + "E": "The blue cup card is positioned above the yellow cactus card rather than left of it in front of you." + }, + "answer_label": "B", + "answer_text": "The blue card with a cup on it in front of you is to the left of the yellow card with a cactus on it.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_4_T1", + "correct_answer": "The blue card with a cup on it in front of you is to the left of the yellow card with a cactus on it.", + "original_options": { + "A": "It is to the right.", + "B": "The blue card with a cup on it in front of you is to the left of the yellow card with a cactus on it.", + "C": "It is directly above.", + "D": "The blue cup card is to the right of the yellow cactus card, not on its left side.", + "E": "The blue cup card is positioned above the yellow cactus card rather than left of it in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 20 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_4_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "For the cards in front of me: where is the blue card with a cup, relative to the yellow card with a cactus?", + "answer": "The blue card with a cup on it in front of you is below the yellow card with a cactus on it.", + "question_chinese": "我面前的画有杯子的蓝色卡片在相对于画有仙人掌的黄色卡片的什么位置?", + "answer_chinese": "你面前的画有杯子的蓝色卡片在画有仙人掌的黄色卡片的下面。", + "evidence": [ + { + "description": "The laptop is in front of your desk.", + "timestamps": { + "start": "00:12:35", + "end": "00:12:39" + } + } + ], + "question_timestamps": "00:14:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is above it.", + "B": "It is to the left.", + "C": "The blue card with a cup on it in front of you is below the yellow card with a cactus on it.", + "D": "The blue cup card is placed above the yellow cactus card, not underneath it in front of you.", + "E": "The blue cup card sits to the right of the yellow cactus card rather than below it among the cards." + }, + "answer_label": "C", + "answer_text": "The blue card with a cup on it in front of you is below the yellow card with a cactus on it.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_4_T2", + "correct_answer": "The blue card with a cup on it in front of you is below the yellow card with a cactus on it.", + "original_options": { + "A": "It is above it.", + "B": "It is to the left.", + "C": "The blue card with a cup on it in front of you is below the yellow card with a cactus on it.", + "D": "The blue cup card is placed above the yellow cactus card, not underneath it in front of you.", + "E": "The blue cup card sits to the right of the yellow cactus card rather than below it among the cards." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 20 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where are the watercolor pens relative to me?", + "answer": "The watercolor pens are in front of your right hand.", + "question_chinese": "相对于我,水彩笔的位置在哪呢?", + "answer_chinese": "水彩笔的位置在你的右手前方。", + "evidence": [ + { + "description": "The watercolor pens are in front of your right hand.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:02:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They are behind your right hand.", + "B": "They are in front of your left hand.", + "C": "The watercolor pens are in front of your right hand.", + "D": "The watercolor pens are positioned beside your right hand rather than directly in front of it.", + "E": "The watercolor pens sit behind your left hand instead of being in front of your right hand." + }, + "answer_label": "C", + "answer_text": "The watercolor pens are in front of your right hand.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_1", + "correct_answer": "The watercolor pens are in front of your right hand.", + "original_options": { + "A": "They are behind your right hand.", + "B": "They are in front of your left hand.", + "C": "The watercolor pens are in front of your right hand.", + "D": "The watercolor pens are positioned beside your right hand rather than directly in front of it.", + "E": "The watercolor pens sit behind your left hand instead of being in front of your right hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 16, + 17 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where are the watercolor pens relative to me?", + "answer": "The watercolor pen is on the tabletop in front of you to the left.", + "question_chinese": "相对于我,水彩笔的位置在哪呢?", + "answer_chinese": "水彩笔在你左前方的桌面上。", + "evidence": [ + { + "description": "You moved the watercolor pen from my right hand to your left hand.", + "timestamps": { + "start": "00:43:19", + "end": "00:43:33" + } + } + ], + "question_timestamps": "00:50:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the tabletop to your right.", + "B": "On the tabletop behind you.", + "C": "The watercolor pen is on the tabletop in front of you to the right, not to the left.", + "D": "The watercolor pen is on the tabletop off to your right side rather than in front-left.", + "E": "The watercolor pen is on the tabletop in front of you to the left." + }, + "answer_label": "E", + "answer_text": "The watercolor pen is on the tabletop in front of you to the left.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_4", + "correct_answer": "The watercolor pen is on the tabletop in front of you to the left.", + "original_options": { + "A": "On the tabletop to your right.", + "B": "On the tabletop behind you.", + "C": "The watercolor pen is on the tabletop in front of you to the right, not to the left.", + "D": "The watercolor pen is on the tabletop off to your right side rather than in front-left.", + "E": "The watercolor pen is on the tabletop in front of you to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 16 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_3", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the black outlining pen relative to me?", + "answer": "The black outlining pen is on your right-hand side.", + "question_chinese": "相对于我,黑色勾线笔的位置在哪里呢?", + "answer_chinese": "黑色勾线笔的位置在你的右手边。", + "evidence": [ + { + "description": "You moved this black outline pen to your right hand.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:04:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your left-hand side.", + "B": "The black outlining pen is on your right-hand side.", + "C": "It is behind you.", + "D": "The black outlining pen is behind you on a shelf instead of on your right-hand side.", + "E": "The black outlining pen is under the desk area, not positioned on your right side." + }, + "answer_label": "B", + "answer_text": "The black outlining pen is on your right-hand side.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_3", + "correct_answer": "The black outlining pen is on your right-hand side.", + "original_options": { + "A": "It is on your left-hand side.", + "B": "The black outlining pen is on your right-hand side.", + "C": "It is behind you.", + "D": "The black outlining pen is behind you on a shelf instead of on your right-hand side.", + "E": "The black outlining pen is under the desk area, not positioned on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 16, + 15 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the black outlining pen relative to me?", + "answer": "The black marker is located in the pencil holder to your.", + "question_chinese": "相对于我,黑色勾线笔的位置在哪里呢?", + "answer_chinese": "黑色勾线笔的位置在你右前方的笔筒中。", + "evidence": [ + { + "description": "You moved the black outline pen from your right hand to the pencil holder in front of you.", + "timestamps": { + "start": "00:43:17", + "end": "00:43:27" + } + } + ], + "question_timestamps": "00:51:33", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The black marker is located in the pencil holder to your.", + "B": "It is on the desk beside you.", + "C": "It is inside the desk drawer.", + "D": "The black outlining pen is placed on the shelf next to your books rather than in a pencil holder.", + "E": "The black outlining pen is under the chair in front of you instead of sitting in the pencil holder." + }, + "answer_label": "A", + "answer_text": "The black marker is located in the pencil holder to your.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_2", + "correct_answer": "The black marker is located in the pencil holder to your.", + "original_options": { + "A": "The black marker is located in the pencil holder to your.", + "B": "It is on the desk beside you.", + "C": "It is inside the desk drawer.", + "D": "The black outlining pen is placed on the shelf next to your books rather than in a pencil holder.", + "E": "The black outlining pen is under the chair in front of you instead of sitting in the pencil holder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 19 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_7", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the notebook?", + "answer": "The notebook is in front of you.", + "question_chinese": "笔记本的位置在哪呢?", + "answer_chinese": "笔记本的位置在你的前面。", + "evidence": [ + { + "description": "The laptop is in front of your desk.", + "timestamps": { + "start": "00:04:53", + "end": "00:05:05" + } + } + ], + "question_timestamps": "00:38:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is to your left.", + "B": "It is under your chair.", + "C": "The notebook is behind you rather than positioned in front of you on the workspace.", + "D": "The notebook is beside your right hand on the table instead of directly in front of you.", + "E": "The notebook is in front of you." + }, + "answer_label": "E", + "answer_text": "The notebook is in front of you.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_7", + "correct_answer": "The notebook is in front of you.", + "original_options": { + "A": "It is to your left.", + "B": "It is under your chair.", + "C": "The notebook is behind you rather than positioned in front of you on the workspace.", + "D": "The notebook is beside your right hand on the table instead of directly in front of you.", + "E": "The notebook is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_8", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the notebook?", + "answer": "The notebook is in front of you.", + "question_chinese": "笔记本的位置在哪呢?", + "answer_chinese": "笔记本的位置在你的前面。", + "evidence": [ + { + "description": "The laptop is in front of your desk.", + "timestamps": { + "start": "00:47:35", + "end": "00:49:00" + } + } + ], + "question_timestamps": "00:50:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your lap.", + "B": "It is near your right hand.", + "C": "The notebook is in front of you.", + "D": "The notebook is tucked behind a stack of papers rather than being located in front of you.", + "E": "The notebook is positioned off to your left side on the table, not directly in front of you." + }, + "answer_label": "C", + "answer_text": "The notebook is in front of you.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_8", + "correct_answer": "The notebook is in front of you.", + "original_options": { + "A": "It is on your lap.", + "B": "It is near your right hand.", + "C": "The notebook is in front of you.", + "D": "The notebook is tucked behind a stack of papers rather than being located in front of you.", + "E": "The notebook is positioned off to your left side on the table, not directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 18 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_11", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the pencil?", + "answer": "The pencil is on your right-hand side?", + "question_chinese": "铅笔的位置在哪呢?", + "answer_chinese": "铅笔在你右手边?", + "evidence": [ + { + "description": "You pick up the pencil and start drawing.", + "timestamps": { + "start": "00:25:02", + "end": "00:26:17" + } + }, + { + "description": "Pencil is on your right.", + "timestamps": { + "start": "00:27:20", + "end": "00:27:44" + } + } + ], + "question_timestamps": "00:27:49", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The pencil is on your right-hand side?", + "B": "On your left-hand side.", + "C": "Underneath your notebook.", + "D": "The pencil is tucked under your chair, not beside you on either side.", + "E": "The pencil is next to your laptop, positioned closer to the center than your right side." + }, + "answer_label": "A", + "answer_text": "The pencil is on your right-hand side?", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_11", + "correct_answer": "The pencil is on your right-hand side?", + "original_options": { + "A": "The pencil is on your right-hand side?", + "B": "On your left-hand side.", + "C": "Underneath your notebook.", + "D": "The pencil is tucked under your chair, not beside you on either side.", + "E": "The pencil is next to your laptop, positioned closer to the center than your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 13, + 16 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_12", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the pencil?", + "answer": "The pencil is at the far right side of the table.", + "question_chinese": "铅笔的位置在哪呢?", + "answer_chinese": "铅笔在桌子的最右边。", + "evidence": [ + { + "description": "You place the pencil from your right hand to the far right side of the table.", + "timestamps": { + "start": "00:44:55", + "end": "00:45:19" + } + } + ], + "question_timestamps": "00:45:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "At the far left side.", + "B": "In the center of the table.", + "C": "The pencil is at the far right side of the table.", + "D": "The pencil is near the middle of the table, not pushed all the way to the right edge.", + "E": "The pencil is at the near right side of the table, not all the way at the far edge." + }, + "answer_label": "C", + "answer_text": "The pencil is at the far right side of the table.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_12", + "correct_answer": "The pencil is at the far right side of the table.", + "original_options": { + "A": "At the far left side.", + "B": "In the center of the table.", + "C": "The pencil is at the far right side of the table.", + "D": "The pencil is near the middle of the table, not pushed all the way to the right edge.", + "E": "The pencil is at the near right side of the table, not all the way at the far edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 19 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_9", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where are the marker box relative to you?", + "answer": "The marker box are on your right-hand side.", + "question_chinese": "马克笔盒相对于你的位置在哪呢?", + "answer_chinese": "马克笔盒的位置在你的右手边。", + "evidence": [ + { + "description": "You take a pen out of the marker box and then place the marker box to your right.", + "timestamps": { + "start": "00:10:29", + "end": "00:10:44" + } + } + ], + "question_timestamps": "00:10:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The marker box are on your right-hand side.", + "B": "In front of you.", + "C": "On your left-hand side.", + "D": "The marker box is behind you, so it isn’t positioned on your right-hand side.", + "E": "The marker box sits directly in front of you on the tabletop, not off to your right." + }, + "answer_label": "A", + "answer_text": "The marker box are on your right-hand side.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_9", + "correct_answer": "The marker box are on your right-hand side.", + "original_options": { + "A": "The marker box are on your right-hand side.", + "B": "In front of you.", + "C": "On your left-hand side.", + "D": "The marker box is behind you, so it isn’t positioned on your right-hand side.", + "E": "The marker box sits directly in front of you on the tabletop, not off to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6", + "video_length": "00:52:00", + "qid": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_10", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where are the marker box relative to you?", + "answer": "The markers are on your right-hand side.", + "question_chinese": "马克笔盒相对于你的位置在哪呢?", + "answer_chinese": "马克笔的位置在你的右手边", + "evidence": [ + { + "description": "You take a pen out of the marker box and then place the marker box to your right.", + "timestamps": { + "start": "00:10:29", + "end": "00:10:44" + } + } + ], + "question_timestamps": "00:44:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "The markers are on your right-hand side.", + "C": "On your left-hand side.", + "D": "The markers are behind you on the shelf, not sitting to the right of your position.", + "E": "The markers are missing from the desk, so they aren’t located on your right-hand side." + }, + "answer_label": "B", + "answer_text": "The markers are on your right-hand side.", + "q_id": "4e1b12cd-710a-4979-8d91-a15424cb1fd6_10", + "correct_answer": "The markers are on your right-hand side.", + "original_options": { + "A": "In front of you.", + "B": "The markers are on your right-hand side.", + "C": "On your left-hand side.", + "D": "The markers are behind you on the shelf, not sitting to the right of your position.", + "E": "The markers are missing from the desk, so they aren’t located on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + } +] diff --git a/QAs/4efd8c65-82c6-481b-8cf2-9038b0b7bc3d-mcq.json b/QAs/4efd8c65-82c6-481b-8cf2-9038b0b7bc3d-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3fe06ee416861bbc7e7ca38c3dc572153ae5ea59 --- /dev/null +++ b/QAs/4efd8c65-82c6-481b-8cf2-9038b0b7bc3d-mcq.json @@ -0,0 +1,1784 @@ +[ + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_8_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the tape measure relative to me?", + "answer": "The tape measure is to your front-right.", + "question_chinese": "卷尺在我的哪个方位？", + "answer_chinese": "卷尺在你的右前方。", + "evidence": [ + { + "description": "The black and yellow tape measure is to your front right.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "The tape measure is directly in front of you, not offset diagonally to the right.", + "D": "The tape measure is behind and to your left, rather than sitting in the front-right area.", + "E": "The tape measure is to your front-right." + }, + "answer_label": "E", + "answer_text": "The tape measure is to your front-right.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_8_T1", + "correct_answer": "The tape measure is to your front-right.", + "original_options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "The tape measure is directly in front of you, not offset diagonally to the right.", + "D": "The tape measure is behind and to your left, rather than sitting in the front-right area.", + "E": "The tape measure is to your front-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_16", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the tape measure from me?", + "answer": "To your front right.", + "question_chinese": "卷尺在我的哪个方位？", + "answer_chinese": "在你的右前方。", + "evidence": [ + { + "description": "The black and yellow tape measure is to your front right.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To your front center.", + "B": "To your front left.", + "C": "To your front right.", + "D": "To your left rear, meaning it’s behind you and offset toward your left side.", + "E": "To your back right, placed behind you and slightly to the right rather than in front." + }, + "answer_label": "C", + "answer_text": "To your front right.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_16", + "correct_answer": "To your front right.", + "original_options": { + "A": "To your front center.", + "B": "To your front left.", + "C": "To your front right.", + "D": "To your left rear, meaning it’s behind you and offset toward your left side.", + "E": "To your back right, placed behind you and slightly to the right rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_8_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the tape measure relative to me?", + "answer": "The tape measure is in front of you.", + "question_chinese": "卷尺在我的哪个方位？", + "answer_chinese": "卷尺在你的前方。", + "evidence": [ + { + "description": "The black and yellow tape measure is to your front .", + "timestamps": { + "start": "00:00:44", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The tape measure is in front of you.", + "D": "The tape measure is to your right side, not positioned in front of your viewpoint.", + "E": "The tape measure is behind you and slightly off to the left, rather than directly ahead." + }, + "answer_label": "C", + "answer_text": "The tape measure is in front of you.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_8_T2", + "correct_answer": "The tape measure is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The tape measure is in front of you.", + "D": "The tape measure is to your right side, not positioned in front of your viewpoint.", + "E": "The tape measure is behind you and slightly off to the left, rather than directly ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 16 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_7", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many screws did I just screw into this long wooden board?", + "answer": "Four.", + "question_chinese": "我刚刚给这块长木板拧进去了几颗螺丝钉？", + "answer_chinese": "4颗。", + "evidence": [ + { + "description": "Screwed 3 screws into the side of the board.", + "timestamps": { + "start": "00:21:46", + "end": "00:21:55" + } + }, + { + "description": "Screwed 3 screws into the side of the board.", + "timestamps": { + "start": "00:22:38", + "end": "00:22:47" + } + } + ], + "question_timestamps": "00:22:51", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two screws.", + "B": "Three screws.", + "C": "Four.", + "D": "You screwed five screws into the long wooden board, not four as you just did.", + "E": "You put in six screws total along the long wooden board, which is more than four." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_7", + "correct_answer": "Four.", + "original_options": { + "A": "Two screws.", + "B": "Three screws.", + "C": "Four.", + "D": "You screwed five screws into the long wooden board, not four as you just did.", + "E": "You put in six screws total along the long wooden board, which is more than four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_8", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many screws did I just screw into this long wooden board?", + "answer": "Four.", + "question_chinese": "我刚刚给这块长木板拧进去了几颗螺丝钉？", + "answer_chinese": "4颗。", + "evidence": [ + { + "description": "Screwed 3 screws into the side of the board.", + "timestamps": { + "start": "00:21:46", + "end": "00:21:55" + } + }, + { + "description": "Screwed 3 screws into the side of the board.", + "timestamps": { + "start": "00:22:38", + "end": "00:22:47" + } + } + ], + "question_timestamps": "00:23:08", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "You screwed only three screws into the long wooden board, then stopped.", + "E": "You screwed five screws into the long wooden board before stopping and pulling the drill back." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_8", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "You screwed only three screws into the long wooden board, then stopped.", + "E": "You screwed five screws into the long wooden board before stopping and pulling the drill back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 16 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_4_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many screws did I just screw into this long wooden board?", + "answer": "You just screwed 4 screws into this long wooden board, and the wooden board is in front of you.", + "question_chinese": "我刚刚给这块长木板拧进去了几颗螺丝钉？", + "answer_chinese": "你刚刚给这块长木板拧进去了4颗螺丝钉，木板在你前面。", + "evidence": [ + { + "description": "Screwed the first screws into the side of the board.", + "timestamps": { + "start": "00:21:46", + "end": "00:21:55" + } + }, + { + "description": "Screwed the second screws into the side of the board.", + "timestamps": { + "start": "00:22:38", + "end": "00:22:47" + } + }, + { + "description": "Screwed the third screws into the side of the board.", + "timestamps": { + "start": "00:23:04", + "end": "00:23:13" + } + }, + { + "description": "Screwed the four screws into the side of the board.", + "timestamps": { + "start": "00:23:14", + "end": "00:23:16" + } + } + ], + "question_timestamps": "00:23:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three screws.", + "B": "Five screws.", + "C": "You just screwed 4 screws into this long wooden board, and the wooden board is in front of you.", + "D": "You just screwed 2 screws into this long wooden board, and the wooden board is in front of you.", + "E": "You just screwed 6 screws into this long wooden board, and the wooden board is in front of you." + }, + "answer_label": "C", + "answer_text": "You just screwed 4 screws into this long wooden board, and the wooden board is in front of you.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_4_T1", + "correct_answer": "You just screwed 4 screws into this long wooden board, and the wooden board is in front of you.", + "original_options": { + "A": "Three screws.", + "B": "Five screws.", + "C": "You just screwed 4 screws into this long wooden board, and the wooden board is in front of you.", + "D": "You just screwed 2 screws into this long wooden board, and the wooden board is in front of you.", + "E": "You just screwed 6 screws into this long wooden board, and the wooden board is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 19, + 19 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_4_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many screws did I just screw into this long wooden board?", + "answer": "You just screwed 0 screws into this long wooden board, and the wooden board is in front of you.", + "question_chinese": "我刚刚给这块长木板拧进去了几颗螺丝钉？", + "answer_chinese": "你刚刚给这块长木板拧进去了0颗螺丝钉，木板在你前面。", + "evidence": [ + { + "description": "Screwed 0 screws into the side of the board.", + "timestamps": { + "start": "00:23:18", + "end": "00:23:20" + } + } + ], + "question_timestamps": "00:23:22", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One screw.", + "B": "Two screws.", + "C": "You just screwed 3 screws into this long wooden board, and the wooden board is in front of you.", + "D": "You just screwed 0 screws into this long wooden board, and the wooden board is in front of you.", + "E": "You just screwed 4 screws into this long wooden board, and the wooden board is in front of you." + }, + "answer_label": "D", + "answer_text": "You just screwed 0 screws into this long wooden board, and the wooden board is in front of you.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_4_T2", + "correct_answer": "You just screwed 0 screws into this long wooden board, and the wooden board is in front of you.", + "original_options": { + "A": "One screw.", + "B": "Two screws.", + "C": "You just screwed 3 screws into this long wooden board, and the wooden board is in front of you.", + "D": "You just screwed 0 screws into this long wooden board, and the wooden board is in front of you.", + "E": "You just screwed 4 screws into this long wooden board, and the wooden board is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 19, + 19 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many extra pieces of wood did I just saw off?", + "answer": "three pieces.", + "question_chinese": "我刚刚锯掉了几块多余的木料？", + "answer_chinese": "3块。", + "evidence": [ + { + "description": "Sawed off the first piece of wood.", + "timestamps": { + "start": "00:09:26", + "end": "00:09:40" + } + }, + { + "description": "Sawed off the second piece of wood.", + "timestamps": { + "start": "00:09:41", + "end": "00:09:50" + } + }, + { + "description": "Sawed off the third piece of wood.", + "timestamps": { + "start": "00:09:55", + "end": "00:10:03" + } + } + ], + "question_timestamps": "00:10:07", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two pieces.", + "B": "Four pieces.", + "C": "You just sawed off one extra piece of wood from the main board.", + "D": "You just sawed off five extra pieces of wood and set them aside after cutting.", + "E": "three pieces." + }, + "answer_label": "E", + "answer_text": "three pieces.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_1", + "correct_answer": "three pieces.", + "original_options": { + "A": "Two pieces.", + "B": "Four pieces.", + "C": "You just sawed off one extra piece of wood from the main board.", + "D": "You just sawed off five extra pieces of wood and set them aside after cutting.", + "E": "three pieces." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many extra pieces of wood did I just saw off?", + "answer": "four pieces.", + "question_chinese": "我刚刚锯掉了几块多余的木料？", + "answer_chinese": "4块。", + "evidence": [ + { + "description": "Sawed off the first piece of wood.", + "timestamps": { + "start": "00:09:26", + "end": "00:09:40" + } + }, + { + "description": "Sawed off the second piece of wood.", + "timestamps": { + "start": "00:09:41", + "end": "00:09:50" + } + }, + { + "description": "Sawed off the third piece of wood.", + "timestamps": { + "start": "00:09:55", + "end": "00:10:03" + } + }, + { + "description": "Sawed off the fourth piece of wood.", + "timestamps": { + "start": "00:10:06", + "end": "00:10:17" + } + } + ], + "question_timestamps": "00:10:23", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three pieces.", + "B": "Five pieces.", + "C": "four pieces.", + "D": "You just sawed off two extra pieces of wood and moved them away from the cut.", + "E": "You just sawed off six extra pieces of wood, leaving a small pile next to you." + }, + "answer_label": "C", + "answer_text": "four pieces.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_2", + "correct_answer": "four pieces.", + "original_options": { + "A": "Three pieces.", + "B": "Five pieces.", + "C": "four pieces.", + "D": "You just sawed off two extra pieces of wood and moved them away from the cut.", + "E": "You just sawed off six extra pieces of wood, leaving a small pile next to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_3", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from getting off the ladder against the wall to getting the red ruler?", + "answer": "You first got off the ladder against the wall, then turned to the back left and walked straight to the two long wooden boards, then walked straight to the front left to another ladder, and finally got the red ruler on it.", + "question_chinese": "我刚才从走下靠墙的梯子到我拿到红色直尺的过程中，经过了什么路径？", + "answer_chinese": "你首先走下靠墙的梯子，然后转向左后方直走到两块长木板前，然后向左前方直走到另一个梯子前，最后在上面拿到红色的直尺。", + "evidence": [ + { + "description": "Got off the ladder against the wall.", + "timestamps": { + "start": "00:12:18", + "end": "00:12:22" + } + }, + { + "description": "After getting off the ladder, turned to the back left and walked straight to the two long wooden boards.", + "timestamps": { + "start": "00:12:22", + "end": "00:12:31" + } + }, + { + "description": "Walked straight to the front left to another ladder, and then got the red ruler on it.", + "timestamps": { + "start": "00:13:04", + "end": "00:13:23" + } + } + ], + "question_timestamps": "00:13:25", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You first got off the ladder against the wall, then turned to the back left and walked straight to the two long wooden boards, then walked straight to the front left to another ladder, and finally got the red ruler on it.", + "B": "You turned back right to the boards.", + "C": "You walked straight ahead to the ladder.", + "D": "You got off the ladder, turned back right, walked to the two long wooden boards, then went to the next ladder.", + "E": "You stepped down, turned to the back left, walked to the boards, then headed to the front right ladder to grab it." + }, + "answer_label": "A", + "answer_text": "You first got off the ladder against the wall, then turned to the back left and walked straight to the two long wooden boards, then walked straight to the front left to another ladder, and finally got the red ruler on it.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_3", + "correct_answer": "You first got off the ladder against the wall, then turned to the back left and walked straight to the two long wooden boards, then walked straight to the front left to another ladder, and finally got the red ruler on it.", + "original_options": { + "A": "You first got off the ladder against the wall, then turned to the back left and walked straight to the two long wooden boards, then walked straight to the front left to another ladder, and finally got the red ruler on it.", + "B": "You turned back right to the boards.", + "C": "You walked straight ahead to the ladder.", + "D": "You got off the ladder, turned back right, walked to the two long wooden boards, then went to the next ladder.", + "E": "You stepped down, turned to the back left, walked to the boards, then headed to the front right ladder to grab it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 22 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_4", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from getting off the ladder against the wall to getting the red ruler?", + "answer": "You first got off the ladder against the wall, then turned to the back left and walked straight to the two long wooden boards, then walked straight to the front left to another ladder, and finally got the red ruler on it.", + "question_chinese": "我刚才从走下靠墙的梯子到我拿到红色直尺的过程中，经过了什么路径？", + "answer_chinese": "你首先走下靠墙的梯子，然后转向左后方直走到两块长木板前，然后向左前方直走到另一个梯子前，最后在上面拿到红色的直尺。", + "evidence": [ + { + "description": "Got off the ladder against the wall.", + "timestamps": { + "start": "00:12:18", + "end": "00:12:22" + } + }, + { + "description": "After getting off the ladder, turned to the back left and walked straight to the two long wooden boards.", + "timestamps": { + "start": "00:12:22", + "end": "00:12:31" + } + }, + { + "description": "Walked straight to the front left to another ladder, and then got the red ruler on it.", + "timestamps": { + "start": "00:13:04", + "end": "00:13:23" + } + } + ], + "question_timestamps": "00:16:01", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turned to the front right first.", + "B": "You went to the workbench first.", + "C": "You got off the ladder, turned back right toward the boards, then walked to a second ladder and took the red ruler.", + "D": "You first got off the ladder against the wall, then turned to the back left and walked straight to the two long wooden boards, then walked straight to the front left to another ladder, and finally got the red ruler on it.", + "E": "You stepped down, turned to the back left, went past the boards to a shelf, then returned to the ladder." + }, + "answer_label": "D", + "answer_text": "You first got off the ladder against the wall, then turned to the back left and walked straight to the two long wooden boards, then walked straight to the front left to another ladder, and finally got the red ruler on it.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_4", + "correct_answer": "You first got off the ladder against the wall, then turned to the back left and walked straight to the two long wooden boards, then walked straight to the front left to another ladder, and finally got the red ruler on it.", + "original_options": { + "A": "You turned to the front right first.", + "B": "You went to the workbench first.", + "C": "You got off the ladder, turned back right toward the boards, then walked to a second ladder and took the red ruler.", + "D": "You first got off the ladder against the wall, then turned to the back left and walked straight to the two long wooden boards, then walked straight to the front left to another ladder, and finally got the red ruler on it.", + "E": "You stepped down, turned to the back left, went past the boards to a shelf, then returned to the ladder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 22, + 20 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_5", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I set up the ladder, what path did I take to get the long wooden board in my hand?", + "answer": "After you set up the ladder, you turned to the back left and walked straight to the two long wooden boards, then picked up a long wooden board, turned around, and walked straight back to the ladder.", + "question_chinese": "我刚才放好梯子，去拿手上这块长木板的时候，经过了什么路径？", + "answer_chinese": "你首先放好梯子后，转向左后方直走到两块长木板前，然后拿起一块长木板，然后转身直走回到梯子前。", + "evidence": [ + { + "description": "After setting up the ladder, turned to the back left and walked straight to the two long wooden boards.", + "timestamps": { + "start": "00:19:48", + "end": "00:20:00" + } + }, + { + "description": "Picked up a long wooden board, then turned around and walked straight back to the ladder.", + "timestamps": { + "start": "00:20:09", + "end": "00:20:23" + } + } + ], + "question_timestamps": "00:20:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turned back right to the boards.", + "B": "You walked forward left to the boards.", + "C": "After you set up the ladder, you turned to the back left and walked straight to the two long wooden boards, then picked up a long wooden board, turned around, and walked straight back to the ladder.", + "D": "After setting up the ladder, you went back left, picked up a board, then walked to the front right ladder.", + "E": "After setting up the ladder, you headed to the back right, grabbed a long board, then returned to the ladder." + }, + "answer_label": "C", + "answer_text": "After you set up the ladder, you turned to the back left and walked straight to the two long wooden boards, then picked up a long wooden board, turned around, and walked straight back to the ladder.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_5", + "correct_answer": "After you set up the ladder, you turned to the back left and walked straight to the two long wooden boards, then picked up a long wooden board, turned around, and walked straight back to the ladder.", + "original_options": { + "A": "You turned back right to the boards.", + "B": "You walked forward left to the boards.", + "C": "After you set up the ladder, you turned to the back left and walked straight to the two long wooden boards, then picked up a long wooden board, turned around, and walked straight back to the ladder.", + "D": "After setting up the ladder, you went back left, picked up a board, then walked to the front right ladder.", + "E": "After setting up the ladder, you headed to the back right, grabbed a long board, then returned to the ladder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 20 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_6", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I set up the ladder, what path did I take to get the long wooden board in my hand?", + "answer": "After you set up the ladder, you turned to the back left and walked straight to the two long wooden boards, then picked up a long wooden board, turned around, and walked straight back to the ladder.", + "question_chinese": "我刚才放好梯子，去拿手上这块长木板的时候，经过了什么路径？", + "answer_chinese": "你首先放好梯子后，转向左后方直走到两块长木板前，然后拿起一块长木板，然后转身直走回到梯子前。", + "evidence": [ + { + "description": "After setting up the ladder, turned to the back left and walked straight to the two long wooden boards.", + "timestamps": { + "start": "00:19:48", + "end": "00:20:00" + } + }, + { + "description": "Picked up a long wooden board, then turned around and walked straight back to the ladder.", + "timestamps": { + "start": "00:20:09", + "end": "00:20:23" + } + } + ], + "question_timestamps": "00:20:56", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turned back right to the boards.", + "B": "You walked forward to the boards.", + "C": "After setting up the ladder, you turned to the back right, grabbed a long board, then walked straight back.", + "D": "After you set up the ladder, you turned to the back left and walked straight to the two long wooden boards, then picked up a long wooden board, turned around, and walked straight back to the ladder.", + "E": "After setting up the ladder, you went back left to the boards, picked one up, and returned by a different route." + }, + "answer_label": "D", + "answer_text": "After you set up the ladder, you turned to the back left and walked straight to the two long wooden boards, then picked up a long wooden board, turned around, and walked straight back to the ladder.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_6", + "correct_answer": "After you set up the ladder, you turned to the back left and walked straight to the two long wooden boards, then picked up a long wooden board, turned around, and walked straight back to the ladder.", + "original_options": { + "A": "You turned back right to the boards.", + "B": "You walked forward to the boards.", + "C": "After setting up the ladder, you turned to the back right, grabbed a long board, then walked straight back.", + "D": "After you set up the ladder, you turned to the back left and walked straight to the two long wooden boards, then picked up a long wooden board, turned around, and walked straight back to the ladder.", + "E": "After setting up the ladder, you went back left to the boards, picked one up, and returned by a different route." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 21 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_18", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "After I finished sharpening the pencil with the knife, where did I put the knife?", + "answer": "In your left pants pocket.", + "question_chinese": "我刚才用小刀削完铅笔后，把小刀放在哪里了？", + "answer_chinese": "在你左边的裤兜里。", + "evidence": [ + { + "description": "After sharpening the pencil, you put the knife in your left pants pocket.", + "timestamps": { + "start": "00:04:08", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:04:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In your left pants pocket.", + "B": "In your right pants pocket.", + "C": "In your shirt pocket.", + "D": "You set the knife down on the table in front of you after sharpening the pencil.", + "E": "You tucked the knife into your backpack near the door once you finished sharpening." + }, + "answer_label": "A", + "answer_text": "In your left pants pocket.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_18", + "correct_answer": "In your left pants pocket.", + "original_options": { + "A": "In your left pants pocket.", + "B": "In your right pants pocket.", + "C": "In your shirt pocket.", + "D": "You set the knife down on the table in front of you after sharpening the pencil.", + "E": "You tucked the knife into your backpack near the door once you finished sharpening." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 14 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_19", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "After I finished sharpening the pencil with the knife, where did I put the knife?", + "answer": "In your left pants pocket.", + "question_chinese": "我刚才用小刀削完铅笔后，把小刀放在哪里了？", + "answer_chinese": "在你左边的裤兜里。", + "evidence": [ + { + "description": "After sharpening the pencil, you put the knife in your left pants pocket.", + "timestamps": { + "start": "00:04:08", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:05:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In your right pants pocket.", + "B": "In your left pants pocket.", + "C": "On the kitchen table.", + "D": "You placed the knife on the counter beside the sink after finishing the sharpening.", + "E": "You put the knife away in your backpack once the pencil was sharpened." + }, + "answer_label": "B", + "answer_text": "In your left pants pocket.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_19", + "correct_answer": "In your left pants pocket.", + "original_options": { + "A": "In your right pants pocket.", + "B": "In your left pants pocket.", + "C": "On the kitchen table.", + "D": "You placed the knife on the counter beside the sink after finishing the sharpening.", + "E": "You put the knife away in your backpack once the pencil was sharpened." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 13 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_5_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many small pieces did I cut off from that wooden board just now?", + "answer": "You just cut off 0 small pieces from that wooden board, and the wooden board is in front of you.", + "question_chinese": "我刚才给那块木板切割下来了几个小块？", + "answer_chinese": "我刚才给那块木板切割下来了0个小块，木板在你前面。", + "evidence": [ + { + "description": "Cut off a small piece from the board.", + "timestamps": { + "start": "00:38:00", + "end": "00:38:13" + } + } + ], + "question_timestamps": "00:38:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "1 small piece.", + "B": "2 small pieces.", + "C": "You just cut off 3 small pieces from that wooden board, and the wooden board is in front of you.", + "D": "You just cut off 4 small pieces from that wooden board, and the wooden board is in front of you.", + "E": "You just cut off 0 small pieces from that wooden board, and the wooden board is in front of you." + }, + "answer_label": "E", + "answer_text": "You just cut off 0 small pieces from that wooden board, and the wooden board is in front of you.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_5_T2", + "correct_answer": "You just cut off 0 small pieces from that wooden board, and the wooden board is in front of you.", + "original_options": { + "A": "1 small piece.", + "B": "2 small pieces.", + "C": "You just cut off 3 small pieces from that wooden board, and the wooden board is in front of you.", + "D": "You just cut off 4 small pieces from that wooden board, and the wooden board is in front of you.", + "E": "You just cut off 0 small pieces from that wooden board, and the wooden board is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 20, + 20 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_5_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many small pieces did I cut off from that wooden board just now?", + "answer": "You just cut off 1 small piece from that wooden board, and the wooden board is behind you.", + "question_chinese": "我刚才给那块木板切割下来了几个小块？", + "answer_chinese": "你刚才给那块木板切割下来了1个小块，木板在你后面。", + "evidence": [ + { + "description": "Cut off another small piece from the board.", + "timestamps": { + "start": "00:38:55", + "end": "00:39:08" + } + } + ], + "question_timestamps": "00:39:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "0 small pieces.", + "B": "2 small pieces.", + "C": "You just cut off 3 small pieces from that wooden board, and the wooden board is behind you.", + "D": "You just cut off 1 small piece from that wooden board, and the wooden board is behind you.", + "E": "You just cut off 4 small pieces from that wooden board, and the wooden board is behind you." + }, + "answer_label": "D", + "answer_text": "You just cut off 1 small piece from that wooden board, and the wooden board is behind you.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_5_T1", + "correct_answer": "You just cut off 1 small piece from that wooden board, and the wooden board is behind you.", + "original_options": { + "A": "0 small pieces.", + "B": "2 small pieces.", + "C": "You just cut off 3 small pieces from that wooden board, and the wooden board is behind you.", + "D": "You just cut off 1 small piece from that wooden board, and the wooden board is behind you.", + "E": "You just cut off 4 small pieces from that wooden board, and the wooden board is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 18 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_9", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many small pieces did I just cut from that wooden board?", + "answer": "Twice.", + "question_chinese": "我刚才给那块木板切割下来了几个小块？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "Cut off a small piece from the board.", + "timestamps": { + "start": "00:38:26", + "end": "00:38:38" + } + }, + { + "description": "Cut off another small piece from the board.", + "timestamps": { + "start": "00:38:55", + "end": "00:39:08" + } + } + ], + "question_timestamps": "00:39:19", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "You just cut from that wooden board four times in total during the last step.", + "D": "You cut small pieces from that wooden board five separate times while trimming the board.", + "E": "Twice." + }, + "answer_label": "E", + "answer_text": "Twice.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_9", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "You just cut from that wooden board four times in total during the last step.", + "D": "You cut small pieces from that wooden board five separate times while trimming the board.", + "E": "Twice." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 15 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_10", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many small pieces did I just cut from that wooden board?", + "answer": "Twice.", + "question_chinese": "我刚才给那块木板切割下来了几个小块？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "Cut off a small piece from the board.", + "timestamps": { + "start": "00:38:26", + "end": "00:38:38" + } + }, + { + "description": "Cut off another small piece from the board.", + "timestamps": { + "start": "00:38:55", + "end": "00:39:08" + } + } + ], + "question_timestamps": "00:39:39", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "You cut small pieces from that wooden board four times total during that single cutting sequence.", + "D": "Twice.", + "E": "You ended up cutting small pieces from that board five times while trimming it down." + }, + "answer_label": "D", + "answer_text": "Twice.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_10", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "You cut small pieces from that wooden board four times total during that single cutting sequence.", + "D": "Twice.", + "E": "You ended up cutting small pieces from that board five times while trimming it down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 16, + 15 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_6_T1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I cut off the extra wood from that wooden board just now, how many times did I use the red cutting machine?", + "answer": "When you were cutting off the extra wood from that wooden board just now, you used the red cutting machine 2 times, and the cutting machine was to your right.", + "question_chinese": "我刚才给那块木板切割多余木料的时候，使用了几次红色切割机？", + "answer_chinese": "你刚才给那块木板切割多余木料的时候，使用了2次红色切割机，切割机在你右边。", + "evidence": [ + { + "description": "Used the red cutting machine once.", + "timestamps": { + "start": "00:40:43", + "end": "00:40:56" + } + }, + { + "description": "Used the red cutting machine again.", + "timestamps": { + "start": "00:41:14", + "end": "00:41:28" + } + } + ], + "question_timestamps": "00:41:31", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "0 times.", + "B": "1 time.", + "C": "When cutting off the extra wood, you used the red cutting machine 3 times, and it was to your right.", + "D": "When cutting off the extra wood, you used the red cutting machine 4 times, and the cutting machine was to your right.", + "E": "When you were cutting off the extra wood from that wooden board just now, you used the red cutting machine 2 times, and the cutting machine was to your right." + }, + "answer_label": "E", + "answer_text": "When you were cutting off the extra wood from that wooden board just now, you used the red cutting machine 2 times, and the cutting machine was to your right.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_6_T1", + "correct_answer": "When you were cutting off the extra wood from that wooden board just now, you used the red cutting machine 2 times, and the cutting machine was to your right.", + "original_options": { + "A": "0 times.", + "B": "1 time.", + "C": "When cutting off the extra wood, you used the red cutting machine 3 times, and it was to your right.", + "D": "When cutting off the extra wood, you used the red cutting machine 4 times, and the cutting machine was to your right.", + "E": "When you were cutting off the extra wood from that wooden board just now, you used the red cutting machine 2 times, and the cutting machine was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 20, + 22 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_6_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I cut off the extra wood from that wooden board just now, how many times did I use the red cutting machine?", + "answer": "When you were cutting off the extra wood from that wooden board just now, you used the red cutting machine 0 times, and the cutting machine was to your right.", + "question_chinese": "我刚才给那块木板切割多余木料的时候，使用了几次红色切割机？", + "answer_chinese": "你刚才给那块木板切割多余木料的时候，使用了0次红色切割机，切割机在你右边。", + "evidence": [ + { + "description": "Used the red cutting machine .", + "timestamps": { + "start": "00:40:51", + "end": "00:40:55" + } + } + ], + "question_timestamps": "00:43:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "When you were cutting off the extra wood from that wooden board just now, you used the red cutting machine 0 times, and the cutting machine was to your right.", + "B": "Used it once.", + "C": "Used it twice.", + "D": "You used the red cutting machine 1 time while trimming the board; it was positioned to your left.", + "E": "You used the red cutting machine 2 times to cut extra wood from the board, with the machine in front." + }, + "answer_label": "A", + "answer_text": "When you were cutting off the extra wood from that wooden board just now, you used the red cutting machine 0 times, and the cutting machine was to your right.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_6_T2", + "correct_answer": "When you were cutting off the extra wood from that wooden board just now, you used the red cutting machine 0 times, and the cutting machine was to your right.", + "original_options": { + "A": "When you were cutting off the extra wood from that wooden board just now, you used the red cutting machine 0 times, and the cutting machine was to your right.", + "B": "Used it once.", + "C": "Used it twice.", + "D": "You used the red cutting machine 1 time while trimming the board; it was positioned to your left.", + "E": "You used the red cutting machine 2 times to cut extra wood from the board, with the machine in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 20 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_12", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I use the red cutting machine when I was cutting the excess wood from that board?", + "answer": "Twice.", + "question_chinese": "我刚才给那块木板切割多余木料的时候，使用了几次红色切割机？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "Used the red cutting machine once.", + "timestamps": { + "start": "00:40:43", + "end": "00:40:56" + } + }, + { + "description": "Used the red cutting machine again.", + "timestamps": { + "start": "00:41:14", + "end": "00:41:28" + } + } + ], + "question_timestamps": "00:43:16", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "You used the red cutting machine four times while cutting away the excess wood from the board.", + "E": "You used the red cutting machine five times during the trimming, making several repeated passes to remove extra wood." + }, + "answer_label": "A", + "answer_text": "Twice.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_12", + "correct_answer": "Twice.", + "original_options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "You used the red cutting machine four times while cutting away the excess wood from the board.", + "E": "You used the red cutting machine five times during the trimming, making several repeated passes to remove extra wood." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 17, + 19 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_3_T2", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many black boxes are in front of me?", + "answer": "There are four boxes in front of you.", + "question_chinese": "我前方有多少个黑箱子", + "answer_chinese": "你前方有四个箱子。", + "evidence": [ + { + "description": "There are four boxs.", + "timestamps": { + "start": "00:19:51", + "end": "00:19:54" + } + } + ], + "question_timestamps": "00:19:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Five boxes.", + "B": "Three boxes.", + "C": "There are four boxes in front of you.", + "D": "There are two boxes in front of you, lined up close together.", + "E": "There are six boxes in front of you, stacked and spread across the area." + }, + "answer_label": "C", + "answer_text": "There are four boxes in front of you.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_3_T2", + "correct_answer": "There are four boxes in front of you.", + "original_options": { + "A": "Five boxes.", + "B": "Three boxes.", + "C": "There are four boxes in front of you.", + "D": "There are two boxes in front of you, lined up close together.", + "E": "There are six boxes in front of you, stacked and spread across the area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_3_T3", + "qid_num": 24, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many black boxes are in front of me?", + "answer": "There are zero boxes in front of you.", + "question_chinese": "我前方有多少个黑箱子", + "answer_chinese": "你前方有零个箱子。", + "evidence": [ + { + "description": "There are zero boxs.", + "timestamps": { + "start": "00:20:16", + "end": "00:20:20" + } + } + ], + "question_timestamps": "00:20:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One box.", + "B": "Two boxes.", + "C": "There are three boxes in front of you, arranged side by side.", + "D": "There are zero boxes in front of you.", + "E": "There are four boxes in front of you, with one slightly offset from the others." + }, + "answer_label": "D", + "answer_text": "There are zero boxes in front of you.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_3_T3", + "correct_answer": "There are zero boxes in front of you.", + "original_options": { + "A": "One box.", + "B": "Two boxes.", + "C": "There are three boxes in front of you, arranged side by side.", + "D": "There are zero boxes in front of you.", + "E": "There are four boxes in front of you, with one slightly offset from the others." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 15 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_10_T1", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the black-and-yellow tape measure right now?", + "answer": "You can’t reach the black-and-yellow tape measure.", + "question_chinese": "我现在可以够得到黑黄色的卷尺吗？", + "answer_chinese": "你不可以够得到黑黄色的卷尺。", + "evidence": [ + { + "description": "The black and yellow tape measure is behind you, less than half a meter away.", + "timestamps": { + "start": "00:06:03", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can reach the black-and-yellow tape measure.", + "B": "You can’t reach the black-and-yellow tape measure." + }, + "answer_label": "B", + "answer_text": "You can’t reach the black-and-yellow tape measure.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_10_T1", + "correct_answer": "You can’t reach the black-and-yellow tape measure.", + "original_options": { + "A": "Right now you can reach the black-and-yellow tape measure.", + "B": "You can’t reach the black-and-yellow tape measure." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_10_T2", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the black-and-yellow tape measure right now?", + "answer": "Right now you can reach the black-and-yellow tape measure.", + "question_chinese": "我现在可以够得到黑黄色的卷尺吗？", + "answer_chinese": "你现在可以够得到黑黄色的卷尺。", + "evidence": [ + { + "description": "The black and yellow tape measure is behind you, less than half a meter away.", + "timestamps": { + "start": "00:06:08", + "end": "00:06:18" + } + } + ], + "question_timestamps": "00:06:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can reach the black-and-yellow tape measure.", + "B": "You can’t reach the black-and-yellow tape measure." + }, + "answer_label": "A", + "answer_text": "Right now you can reach the black-and-yellow tape measure.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_10_T2", + "correct_answer": "Right now you can reach the black-and-yellow tape measure.", + "original_options": { + "A": "Right now you can reach the black-and-yellow tape measure.", + "B": "You can’t reach the black-and-yellow tape measure." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_20", + "qid_num": 27, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the black and yellow tape measure now?", + "answer": "Yes, you can.", + "question_chinese": "我现在可以够得到黑黄色的卷尺吗？", + "answer_chinese": "可以的。", + "evidence": [ + { + "description": "The black and yellow tape measure is behind you, less than half a meter away.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:23" + } + } + ], + "question_timestamps": "00:06:44", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No, you cannot." + }, + "answer_label": "A", + "answer_text": "Yes, you can.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_20", + "correct_answer": "Yes, you can.", + "original_options": { + "A": "Yes, you can.", + "B": "No, you cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_7_T2", + "qid_num": 28, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach that red electric drill above right now?", + "answer": "Right now you can’t reach the red electric drill up above, and the electric drill is behind you.", + "question_chinese": "我现在能够到上面那个红色的电钻吗？", + "answer_chinese": "你现在不能够到上面那个红色的电钻，电钻在你后面。", + "evidence": [ + { + "description": "The red electric drill is on the wooden beam, more than an arm's length away from you.", + "timestamps": { + "start": "00:41:55", + "end": "00:41:57" + } + } + ], + "question_timestamps": "00:41:59", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can’t reach the red electric drill up above, and the electric drill is behind you.", + "B": "Right now you can reach the red electric drill up above, and the electric drill is in front of you." + }, + "answer_label": "A", + "answer_text": "Right now you can’t reach the red electric drill up above, and the electric drill is behind you.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_7_T2", + "correct_answer": "Right now you can’t reach the red electric drill up above, and the electric drill is behind you.", + "original_options": { + "A": "Right now you can’t reach the red electric drill up above, and the electric drill is behind you.", + "B": "Right now you can reach the red electric drill up above, and the electric drill is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_7_T1", + "qid_num": 29, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach that red electric drill above right now?", + "answer": "Right now you can reach the red electric drill up above, and the electric drill is in front of you.", + "question_chinese": "我现在能够到上面那个红色的电钻吗？", + "answer_chinese": "你现在能够到上面那个红色的电钻，电钻在你前面。", + "evidence": [ + { + "description": "The red electric drill is on the wooden beam, less than an arm's length away from you.", + "timestamps": { + "start": "00:46:22", + "end": "00:46:35" + } + } + ], + "question_timestamps": "00:46:37", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can’t reach the red electric drill up above, and the electric drill is in front of you.", + "B": "Right now you can reach the red electric drill up above, and the electric drill is in front of you." + }, + "answer_label": "B", + "answer_text": "Right now you can reach the red electric drill up above, and the electric drill is in front of you.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_7_T1", + "correct_answer": "Right now you can reach the red electric drill up above, and the electric drill is in front of you.", + "original_options": { + "A": "Right now you can’t reach the red electric drill up above, and the electric drill is in front of you.", + "B": "Right now you can reach the red electric drill up above, and the electric drill is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 21 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_14", + "qid_num": 30, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the red electric drill above me now?", + "answer": "Yes.", + "question_chinese": "我现在能够到上面那个红色的电钻吗？", + "answer_chinese": "能。", + "evidence": [ + { + "description": "The red electric drill is on the wooden beam, less than an arm's length away from you.", + "timestamps": { + "start": "00:46:22", + "end": "00:46:35" + } + } + ], + "question_timestamps": "00:47:11", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_14", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_2_T1", + "qid_num": 33, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "In which direction from me is the wooden board?", + "answer": "The wooden board is in front of you.", + "question_chinese": "木板在我的什么方向？", + "answer_chinese": "木板在你的前方。", + "evidence": [ + { + "description": "The wooden board in front of you.", + "timestamps": { + "start": "00:12:25", + "end": "00:12:39" + } + } + ], + "question_timestamps": "00:12:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The wooden board is behind you.", + "B": "The wooden board is in front of you.", + "C": "The wooden board is to your left.", + "D": "The wooden board is to your right, not straight ahead from where you’re standing.", + "E": "The wooden board is above you, so it isn’t positioned out in front of you." + }, + "answer_label": "B", + "answer_text": "The wooden board is in front of you.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_2_T1", + "correct_answer": "The wooden board is in front of you.", + "original_options": { + "A": "The wooden board is behind you.", + "B": "The wooden board is in front of you.", + "C": "The wooden board is to your left.", + "D": "The wooden board is to your right, not straight ahead from where you’re standing.", + "E": "The wooden board is above you, so it isn’t positioned out in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 16 + ] + }, + { + "video_uid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d", + "video_length": "01:02:01", + "qid": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_2_T2", + "qid_num": 34, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "In which direction from me is the wooden board?", + "answer": "The wooden board is to your right.", + "question_chinese": "木板在我的什么方向？", + "answer_chinese": "木板在你的右方。", + "evidence": [ + { + "description": "The wooden board is on your right.", + "timestamps": { + "start": "00:12:41", + "end": "00:12:55" + } + } + ], + "question_timestamps": "00:12:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The wooden board is in front of you.", + "B": "The wooden board is behind you.", + "C": "The wooden board is to your left rather than on your right side.", + "D": "The wooden board is to your right.", + "E": "The wooden board is above you, not positioned to your right at that moment." + }, + "answer_label": "D", + "answer_text": "The wooden board is to your right.", + "q_id": "4efd8c65-82c6-481b-8cf2-9038b0b7bc3d_2_T2", + "correct_answer": "The wooden board is to your right.", + "original_options": { + "A": "The wooden board is in front of you.", + "B": "The wooden board is behind you.", + "C": "The wooden board is to your left rather than on your right side.", + "D": "The wooden board is to your right.", + "E": "The wooden board is above you, not positioned to your right at that moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 13, + 14 + ] + } +] diff --git a/QAs/4f11d6b8-1f78-4d99-9b34-7b89bc007b25-mcq.json b/QAs/4f11d6b8-1f78-4d99-9b34-7b89bc007b25-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8a2e684d810175a884e37154d9d3d3c312b88ace --- /dev/null +++ b/QAs/4f11d6b8-1f78-4d99-9b34-7b89bc007b25-mcq.json @@ -0,0 +1,750 @@ +[ + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just as I was crossing the bridge, where were those stones in relation to me?", + "answer": "Just as you were crossing the bridge, those stones were right in front of you.", + "question_chinese": "我刚刚上桥时，那些石头在我的哪个方向？", + "answer_chinese": "你刚刚上桥时，那些石头在你的正前方。", + "evidence": [ + { + "description": "Just as you were crossing the bridge, those stones were right in front of you.", + "timestamps": { + "start": "00:22:50", + "end": "00:22:52" + } + } + ], + "question_timestamps": "00:22:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They were to your left.", + "B": "Just as you were crossing the bridge, those stones were right in front of you.", + "C": "They were behind you.", + "D": "Just as you were crossing the bridge, those stones were near your feet on the bridge path.", + "E": "Just as you were crossing the bridge, those stones were beside you along the railing rather than ahead." + }, + "answer_label": "B", + "answer_text": "Just as you were crossing the bridge, those stones were right in front of you.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_5", + "correct_answer": "Just as you were crossing the bridge, those stones were right in front of you.", + "original_options": { + "A": "They were to your left.", + "B": "Just as you were crossing the bridge, those stones were right in front of you.", + "C": "They were behind you.", + "D": "Just as you were crossing the bridge, those stones were near your feet on the bridge path.", + "E": "Just as you were crossing the bridge, those stones were beside you along the railing rather than ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 18 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just as I was crossing the bridge, where were those stones in relation to me?", + "answer": "Just as you were crossing the bridge, the stone was lying on the ground to your left.", + "question_chinese": "我刚刚上桥时，那些石头在我的哪个方向？", + "answer_chinese": "你刚刚上桥时，石头被在你的左边的地上。", + "evidence": [ + { + "description": "Just as you were crossing the bridge, the stone was lying on the ground to your left.", + "timestamps": { + "start": "00:22:00", + "end": "00:22:05" + } + } + ], + "question_timestamps": "00:22:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They were to your right.", + "B": "They were in front of you.", + "C": "Just as you were crossing the bridge, the stone was lying on the ground to your left.", + "D": "Just as you were crossing the bridge, the stone was lying on the ground behind you, not to your left.", + "E": "Just as you were crossing the bridge, the stone was on the other side of the railing rather than left of you." + }, + "answer_label": "C", + "answer_text": "Just as you were crossing the bridge, the stone was lying on the ground to your left.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_4", + "correct_answer": "Just as you were crossing the bridge, the stone was lying on the ground to your left.", + "original_options": { + "A": "They were to your right.", + "B": "They were in front of you.", + "C": "Just as you were crossing the bridge, the stone was lying on the ground to your left.", + "D": "Just as you were crossing the bridge, the stone was lying on the ground behind you, not to your left.", + "E": "Just as you were crossing the bridge, the stone was on the other side of the railing rather than left of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 22 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_4_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to use the saw—where did I put it?", + "answer": "The saw was placed on the ground to your left of the bricks behind you.", + "question_chinese": "我想使用锯子，请问我把它放在哪里了？", + "answer_chinese": "锯子被放在了你身后砖块的左边地面上。", + "evidence": [ + { + "description": "The saw was placed on the ground.", + "timestamps": { + "start": "00:05:45", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:05:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the ground in front of the bricks.", + "B": "On the ground right of the bricks.", + "C": "You left the saw on the ground to your right, just beside the bricks behind you.", + "D": "You put the saw on the ground to your left of the bricks in front of you, near your feet.", + "E": "The saw was placed on the ground to your left of the bricks behind you." + }, + "answer_label": "E", + "answer_text": "The saw was placed on the ground to your left of the bricks behind you.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_4_T2", + "correct_answer": "The saw was placed on the ground to your left of the bricks behind you.", + "original_options": { + "A": "On the ground in front of the bricks.", + "B": "On the ground right of the bricks.", + "C": "You left the saw on the ground to your right, just beside the bricks behind you.", + "D": "You put the saw on the ground to your left of the bricks in front of you, near your feet.", + "E": "The saw was placed on the ground to your left of the bricks behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 20 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_4_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to use the saw—where did I put it?", + "answer": "The saw was placed on the other side of the bridge.", + "question_chinese": "我想使用锯子，请问我把它放在哪里了？", + "answer_chinese": "锯子被放在了桥的另一边。", + "evidence": [ + { + "description": "The saw was placed on the ground.", + "timestamps": { + "start": "00:05:45", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:34:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the near side of the bridge.", + "B": "The saw was placed on the other side of the bridge.", + "C": "Right before the bridge entrance.", + "D": "You left the saw on the bridge itself, not past it on the far side.", + "E": "The saw was set down just off to the side before you crossed the bridge to the other side." + }, + "answer_label": "B", + "answer_text": "The saw was placed on the other side of the bridge.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_4_T1", + "correct_answer": "The saw was placed on the other side of the bridge.", + "original_options": { + "A": "On the near side of the bridge.", + "B": "The saw was placed on the other side of the bridge.", + "C": "Right before the bridge entrance.", + "D": "You left the saw on the bridge itself, not past it on the far side.", + "E": "The saw was set down just off to the side before you crossed the bridge to the other side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 19 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_5_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the hammer?", + "answer": "No. One hammer was on the truck, and the other was behind you.", + "question_chinese": "我能够得着锤子吗？", + "answer_chinese": "不能，有一个锤子放在卡车上，另一个在你的后面。", + "evidence": [ + { + "description": "There is a hammer placed on the truck.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:08" + } + }, + { + "description": "There is a hammer lying on the ground.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. One hammer was on the truck, and the other was behind you.", + "B": "No. One hammer was on the truck, and the other was behind you." + }, + "answer_label": "B", + "answer_text": "No. One hammer was on the truck, and the other was behind you.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_5_T1", + "correct_answer": "No. One hammer was on the truck, and the other was behind you.", + "original_options": { + "A": "Yes. One hammer was on the truck, and the other was behind you.", + "B": "No. One hammer was on the truck, and the other was behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_5_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the hammer?", + "answer": "You can reach the hammer—the red hammer is on your right hand side.", + "question_chinese": "我能够得着锤子吗？", + "answer_chinese": "你够得到锤子，红色的锤子在你的右手边。", + "evidence": [ + { + "description": "There is a hammer placed on the truck.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:08" + } + }, + { + "description": "There is a hammer lying on the ground.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:02:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the hammer—the red hammer is on your right hand side.", + "B": "No. One hammer was on the truck, and the other was behind you." + }, + "answer_label": "A", + "answer_text": "You can reach the hammer—the red hammer is on your right hand side.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_5_T2", + "correct_answer": "You can reach the hammer—the red hammer is on your right hand side.", + "original_options": { + "A": "You can reach the hammer—the red hammer is on your right hand side.", + "B": "No. One hammer was on the truck, and the other was behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_6_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the bridge from me?", + "answer": "The bridge is on your right-hand side.", + "question_chinese": "桥在我的哪个方向？", + "answer_chinese": "桥在你的右手边。", + "evidence": [ + { + "description": "When sitting in the truck, the bridge is on your right rear side.", + "timestamps": { + "start": "00:13:43", + "end": "00:13:55" + } + } + ], + "question_timestamps": "00:14:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bridge is in front of you.", + "B": "The bridge is behind you.", + "C": "The bridge is on your left-hand side, not on your right-hand side.", + "D": "The bridge is directly ahead of you, rather than off to your right-hand side.", + "E": "The bridge is on your right-hand side." + }, + "answer_label": "E", + "answer_text": "The bridge is on your right-hand side.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_6_T1", + "correct_answer": "The bridge is on your right-hand side.", + "original_options": { + "A": "The bridge is in front of you.", + "B": "The bridge is behind you.", + "C": "The bridge is on your left-hand side, not on your right-hand side.", + "D": "The bridge is directly ahead of you, rather than off to your right-hand side.", + "E": "The bridge is on your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 12, + 14 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_6_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the bridge from me?", + "answer": "The bridge is directly in front of you.", + "question_chinese": "桥在我的哪个方向？", + "answer_chinese": "桥在你的正前方。", + "evidence": [ + { + "description": "When sitting in the truck, the bridge is on your right rear side.", + "timestamps": { + "start": "00:13:43", + "end": "00:13:55" + } + }, + { + "description": "The bridge is right in front of you.", + "timestamps": { + "start": "00:18:09", + "end": "00:18:23" + } + } + ], + "question_timestamps": "00:18:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bridge is directly in front of you.", + "B": "The bridge is directly behind you.", + "C": "The bridge is directly to your left.", + "D": "The bridge is directly to your right, not straight ahead in front of you.", + "E": "The bridge is behind you, so you would need to turn around to face it." + }, + "answer_label": "A", + "answer_text": "The bridge is directly in front of you.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_6_T2", + "correct_answer": "The bridge is directly in front of you.", + "original_options": { + "A": "The bridge is directly in front of you.", + "B": "The bridge is directly behind you.", + "C": "The bridge is directly to your left.", + "D": "The bridge is directly to your right, not straight ahead in front of you.", + "E": "The bridge is behind you, so you would need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 15 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_6", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How was the level gauge moved to its current position?", + "answer": "The level gauge was placed on the stone, then moved from the stone onto the vehicle, then returned to the stone, and finally placed on the bridge.", + "question_chinese": "水平仪是如何被移动到现在的位置的？", + "answer_chinese": "水平仪被从石头上放到了车上，又从车上被放回了石头上，然后被放到了桥上。", + "evidence": [ + { + "description": "The level gauge was placed on the vehicle from the stone.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + }, + { + "description": "The level gauge was placed back on the stone from the vehicle.", + "timestamps": { + "start": "00:13:06", + "end": "00:13:15" + } + }, + { + "description": "The level gauge was placed from the stone onto the handrail of the bridge.", + "timestamps": { + "start": "00:20:16", + "end": "00:20:23" + } + } + ], + "question_timestamps": "00:22:14", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Stone, bridge, vehicle, bridge.", + "B": "The level gauge was placed on the stone, then moved from the stone onto the vehicle, then returned to the stone, and finally placed on the bridge.", + "C": "Vehicle, stone, bridge, vehicle.", + "D": "The level gauge was placed on the vehicle first, then moved to the stone, then put on the bridge.", + "E": "The level gauge started on the stone, moved onto the bridge, then onto the vehicle, and stayed there afterward." + }, + "answer_label": "B", + "answer_text": "The level gauge was placed on the stone, then moved from the stone onto the vehicle, then returned to the stone, and finally placed on the bridge.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_6", + "correct_answer": "The level gauge was placed on the stone, then moved from the stone onto the vehicle, then returned to the stone, and finally placed on the bridge.", + "original_options": { + "A": "Stone, bridge, vehicle, bridge.", + "B": "The level gauge was placed on the stone, then moved from the stone onto the vehicle, then returned to the stone, and finally placed on the bridge.", + "C": "Vehicle, stone, bridge, vehicle.", + "D": "The level gauge was placed on the vehicle first, then moved to the stone, then put on the bridge.", + "E": "The level gauge started on the stone, moved onto the bridge, then onto the vehicle, and stayed there afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 19 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_7", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How was the level gauge moved to its current position?", + "answer": "The level gauge was placed on the stone, then taken off the vehicle and put back on the stone, then placed on the bridge, and finally placed at the bridgehead.", + "question_chinese": "水平仪是如何被移动到现在的位置的？", + "answer_chinese": "水平仪被从石头上放到了车上，又从车上被放回了石头上，然后被放到了桥上，最后被放在了桥头。", + "evidence": [ + { + "description": "The level gauge was placed on the vehicle from the stone.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + }, + { + "description": "The level gauge was placed back on the stone from the vehicle.", + "timestamps": { + "start": "00:13:06", + "end": "00:13:15" + } + }, + { + "description": "The level gauge was placed from the stone onto the handrail of the bridge.", + "timestamps": { + "start": "00:20:16", + "end": "00:20:23" + } + }, + { + "description": "The level gauge was placed at the bridgehead.", + "timestamps": { + "start": "00:23:27", + "end": "00:23:44" + } + } + ], + "question_timestamps": "00:24:34", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Stone, vehicle, bridgehead, bridge.", + "B": "Vehicle, stone, bridge, bridgehead.", + "C": "The level gauge was placed on the stone, moved onto the vehicle, then set on the bridge at the end.", + "D": "The level gauge was placed on the stone, then taken off the vehicle and put back on the stone, then placed on the bridge, and finally placed at the bridgehead.", + "E": "The level gauge was placed on the stone, moved to the bridge, returned to the vehicle, and ended on the stone." + }, + "answer_label": "D", + "answer_text": "The level gauge was placed on the stone, then taken off the vehicle and put back on the stone, then placed on the bridge, and finally placed at the bridgehead.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_7", + "correct_answer": "The level gauge was placed on the stone, then taken off the vehicle and put back on the stone, then placed on the bridge, and finally placed at the bridgehead.", + "original_options": { + "A": "Stone, vehicle, bridgehead, bridge.", + "B": "Vehicle, stone, bridge, bridgehead.", + "C": "The level gauge was placed on the stone, moved onto the vehicle, then set on the bridge at the end.", + "D": "The level gauge was placed on the stone, then taken off the vehicle and put back on the stone, then placed on the bridge, and finally placed at the bridgehead.", + "E": "The level gauge was placed on the stone, moved to the bridge, returned to the vehicle, and ended on the stone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 21 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stones are there in total here?", + "answer": "There are a total of seven stones here.", + "question_chinese": "这里一共有多少块石头？", + "answer_chinese": "这里一共有七块石头。", + "evidence": [ + { + "description": "There are a total of seven stones here.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:01", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are a total of seven stones here.", + "B": "There are six stones here.", + "C": "There are eight stones here.", + "D": "There are a total of nine stones visible here right now, spread out on the surface.", + "E": "There are a total of five stones here altogether at the moment, not counting any missing ones." + }, + "answer_label": "A", + "answer_text": "There are a total of seven stones here.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_1", + "correct_answer": "There are a total of seven stones here.", + "original_options": { + "A": "There are a total of seven stones here.", + "B": "There are six stones here.", + "C": "There are eight stones here.", + "D": "There are a total of nine stones visible here right now, spread out on the surface.", + "E": "There are a total of five stones here altogether at the moment, not counting any missing ones." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stones are there in total here?", + "answer": "There are ten stones here. One of them was sawn in half by you.", + "question_chinese": "这里一共有多少块石头？", + "answer_chinese": "这里有十块石头，其中一块被你锯成了两半。", + "evidence": [ + { + "description": "There are a total of ten stones here.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:12" + } + }, + { + "description": "A stone was sawn into two halves by you.", + "timestamps": { + "start": "00:04:39", + "end": "00:05:34" + } + } + ], + "question_timestamps": "00:14:44", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are nine stones here.", + "B": "There are ten stones here. One of them was sawn in half by you.", + "C": "There are eleven stones here.", + "D": "There are twelve stones here, and one of them was sawn in half by you.", + "E": "There are eight stones here, with one of them having been sawn in half by you." + }, + "answer_label": "B", + "answer_text": "There are ten stones here. One of them was sawn in half by you.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_2", + "correct_answer": "There are ten stones here. One of them was sawn in half by you.", + "original_options": { + "A": "There are nine stones here.", + "B": "There are ten stones here. One of them was sawn in half by you.", + "C": "There are eleven stones here.", + "D": "There are twelve stones here, and one of them was sawn in half by you.", + "E": "There are eight stones here, with one of them having been sawn in half by you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25", + "video_length": "00:35:03", + "qid": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_3", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stones are there in total here?", + "answer": "There are twenty stones in total here, and one of them has been cut into two halves.", + "question_chinese": "这里一共有多少块石头？", + "answer_chinese": "这里一共有二十块石头，其中一块被锯成了两半。", + "evidence": [ + { + "description": "A stone was sawn into two halves by you.", + "timestamps": { + "start": "00:04:39", + "end": "00:05:34" + } + }, + { + "description": "There are ten stones here.", + "timestamps": { + "start": "00:21:45", + "end": "00:22:00" + } + }, + { + "description": "There are of ten stones here.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:22:07", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are nineteen stones here.", + "B": "There are twenty-one stones here.", + "C": "There are eighteen stones in total here, and one of them has been cut into two halves.", + "D": "There are twenty stones in total here, and one of them has been cut into two halves.", + "E": "There are twenty-two stones in total here, and one of them has been cut into two halves." + }, + "answer_label": "D", + "answer_text": "There are twenty stones in total here, and one of them has been cut into two halves.", + "q_id": "4f11d6b8-1f78-4d99-9b34-7b89bc007b25_3", + "correct_answer": "There are twenty stones in total here, and one of them has been cut into two halves.", + "original_options": { + "A": "There are nineteen stones here.", + "B": "There are twenty-one stones here.", + "C": "There are eighteen stones in total here, and one of them has been cut into two halves.", + "D": "There are twenty stones in total here, and one of them has been cut into two halves.", + "E": "There are twenty-two stones in total here, and one of them has been cut into two halves." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + } +] diff --git a/QAs/4f7fb690-4ee7-4555-87bd-0fe6171b4239-mcq.json b/QAs/4f7fb690-4ee7-4555-87bd-0fe6171b4239-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4412be1a083ca0493ab91d1d23db33eb0ed20707 --- /dev/null +++ b/QAs/4f7fb690-4ee7-4555-87bd-0fe6171b4239-mcq.json @@ -0,0 +1,1037 @@ +[ + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_9", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What position is the stool relative to me now?", + "answer": "The stool is on your left.", + "question_chinese": "凳子现在相对我是什么位置？", + "answer_chinese": "凳子在你的左侧。", + "evidence": [ + { + "description": "The stool is on your left.", + "timestamps": { + "start": "00:48:38", + "end": "00:48:52" + } + } + ], + "question_timestamps": "00:51:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The stool is on your left.", + "B": "The stool is on your right.", + "C": "The stool is in front of you.", + "D": "The stool is behind you now, slightly off to one side as you face forward.", + "E": "The stool is across from you now, positioned ahead rather than to your left." + }, + "answer_label": "A", + "answer_text": "The stool is on your left.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_9", + "correct_answer": "The stool is on your left.", + "original_options": { + "A": "The stool is on your left.", + "B": "The stool is on your right.", + "C": "The stool is in front of you.", + "D": "The stool is behind you now, slightly off to one side as you face forward.", + "E": "The stool is across from you now, positioned ahead rather than to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 14 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_10", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What position is the stool relative to me now?", + "answer": "The stool is on your left.", + "question_chinese": "凳子现在相对我是什么位置？", + "answer_chinese": "凳子在你的左侧。", + "evidence": [ + { + "description": "The stool is on your left.", + "timestamps": { + "start": "00:59:18", + "end": "00:59:27" + } + } + ], + "question_timestamps": "00:59:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The stool is on your right.", + "B": "The stool is behind you.", + "C": "The stool is directly in front of you now, not positioned to your left side.", + "D": "The stool is on your left.", + "E": "The stool is under your chair at the moment, rather than sitting off to your left." + }, + "answer_label": "D", + "answer_text": "The stool is on your left.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_10", + "correct_answer": "The stool is on your left.", + "original_options": { + "A": "The stool is on your right.", + "B": "The stool is behind you.", + "C": "The stool is directly in front of you now, not positioned to your left side.", + "D": "The stool is on your left.", + "E": "The stool is under your chair at the moment, rather than sitting off to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 16 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_7", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the water cup?", + "answer": "You just put the water cup on the table.", + "question_chinese": "我刚刚把水杯放在什么位置了？", + "answer_chinese": "你刚刚把水杯放在桌上了。", + "evidence": [ + { + "description": "You just put the water cup on the table.", + "timestamps": { + "start": "00:37:08", + "end": "00:37:11" + } + } + ], + "question_timestamps": "00:37:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it on the counter.", + "B": "You put it in the sink.", + "C": "You just put the water cup on the table.", + "D": "You just put the water cup behind the chair, not on the table surface.", + "E": "You just put the water cup near the window area, rather than setting it on the table." + }, + "answer_label": "C", + "answer_text": "You just put the water cup on the table.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_7", + "correct_answer": "You just put the water cup on the table.", + "original_options": { + "A": "You put it on the counter.", + "B": "You put it in the sink.", + "C": "You just put the water cup on the table.", + "D": "You just put the water cup behind the chair, not on the table surface.", + "E": "You just put the water cup near the window area, rather than setting it on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 17 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_8", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the water cup?", + "answer": "You just put the water cup on the table.", + "question_chinese": "我刚刚把水杯放在什么位置了？", + "answer_chinese": "你刚刚把水杯放在桌上了。", + "evidence": [ + { + "description": "You just put the water cup on the table.", + "timestamps": { + "start": "00:37:08", + "end": "00:37:11" + } + } + ], + "question_timestamps": "00:52:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it on the countertop.", + "B": "You put it on the cabinet.", + "C": "You just put the water cup on the windowsill instead of placing it down on the table.", + "D": "You just put the water cup in the drawer, rather than setting it on the table.", + "E": "You just put the water cup on the table." + }, + "answer_label": "E", + "answer_text": "You just put the water cup on the table.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_8", + "correct_answer": "You just put the water cup on the table.", + "original_options": { + "A": "You put it on the countertop.", + "B": "You put it on the cabinet.", + "C": "You just put the water cup on the windowsill instead of placing it down on the table.", + "D": "You just put the water cup in the drawer, rather than setting it on the table.", + "E": "You just put the water cup on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_3_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I rinse the rice just now?", + "answer": "You just washed it 1 time. The rice is in your hand.", + "question_chinese": "我刚刚洗了几遍大米？", + "answer_chinese": "你刚刚洗了1遍。大米在你的手里。", + "evidence": [ + { + "description": "You washed the rice once.", + "timestamps": { + "start": "00:11:05", + "end": "00:11:28" + } + } + ], + "question_timestamps": "00:11:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed it 0 times.", + "B": "You washed it 2 times.", + "C": "You just washed it 1 time. The rice is in your hand.", + "D": "You just washed it 3 times, and the rice is still in your hand right now.", + "E": "You just washed it 4 times, with the rice remaining in your hand as you finished." + }, + "answer_label": "C", + "answer_text": "You just washed it 1 time. The rice is in your hand.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_3_T1", + "correct_answer": "You just washed it 1 time. The rice is in your hand.", + "original_options": { + "A": "You washed it 0 times.", + "B": "You washed it 2 times.", + "C": "You just washed it 1 time. The rice is in your hand.", + "D": "You just washed it 3 times, and the rice is still in your hand right now.", + "E": "You just washed it 4 times, with the rice remaining in your hand as you finished." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_3_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I rinse the rice just now?", + "answer": "You just washed it 2 times. The rice is on the ground by your left hand.", + "question_chinese": "我刚刚洗了几遍大米？", + "answer_chinese": "你刚刚洗了2遍。大米在你左手边的地上。", + "evidence": [ + { + "description": "You washed the rice once.", + "timestamps": { + "start": "00:11:05", + "end": "00:11:28" + } + }, + { + "description": "You washed the rice for the second time.", + "timestamps": { + "start": "00:11:32", + "end": "00:11:50" + } + } + ], + "question_timestamps": "00:12:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just washed it 1 time.", + "B": "You just washed it 3 times.", + "C": "You just washed it 1 time, and the rice is still on the ground by your left hand.", + "D": "You just washed it 2 times. The rice is on the ground by your left hand.", + "E": "You just washed it 3 times, with the rice remaining on the ground by your left hand." + }, + "answer_label": "D", + "answer_text": "You just washed it 2 times. The rice is on the ground by your left hand.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_3_T2", + "correct_answer": "You just washed it 2 times. The rice is on the ground by your left hand.", + "original_options": { + "A": "You just washed it 1 time.", + "B": "You just washed it 3 times.", + "C": "You just washed it 1 time, and the rice is still on the ground by your left hand.", + "D": "You just washed it 2 times. The rice is on the ground by your left hand.", + "E": "You just washed it 3 times, with the rice remaining on the ground by your left hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I just washed the rice?", + "answer": "You just washed it twice.", + "question_chinese": "我刚刚洗了几遍大米？", + "answer_chinese": "你刚刚洗了两遍。", + "evidence": [ + { + "description": "You washed the rice once.", + "timestamps": { + "start": "00:11:05", + "end": "00:11:28" + } + }, + { + "description": "You washed the rice for the second time.", + "timestamps": { + "start": "00:11:32", + "end": "00:11:50" + } + } + ], + "question_timestamps": "00:12:39", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You just washed it once.", + "B": "You just washed it three times.", + "C": "You just washed it twice.", + "D": "You just washed it four times, going through an extra rinse before stopping.", + "E": "You just washed it five times, repeating the rinse cycle several times in a row." + }, + "answer_label": "C", + "answer_text": "You just washed it twice.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_5", + "correct_answer": "You just washed it twice.", + "original_options": { + "A": "You just washed it once.", + "B": "You just washed it three times.", + "C": "You just washed it twice.", + "D": "You just washed it four times, going through an extra rinse before stopping.", + "E": "You just washed it five times, repeating the rinse cycle several times in a row." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 15 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I just washed the rice?", + "answer": "You just washed it twice.", + "question_chinese": "我刚刚洗了几遍大米？", + "answer_chinese": "你刚刚洗了两遍。", + "evidence": [ + { + "description": "You washed the rice once.", + "timestamps": { + "start": "00:11:05", + "end": "00:11:28" + } + }, + { + "description": "You washed the rice for the second time.", + "timestamps": { + "start": "00:11:32", + "end": "00:11:50" + } + } + ], + "question_timestamps": "00:33:55", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You just washed it once.", + "B": "You just washed it twice.", + "C": "You just washed it three times.", + "D": "You just washed it four times, doing two extra rinses beyond what you did.", + "E": "You just washed it five times, continuing to rinse repeatedly before finishing." + }, + "answer_label": "B", + "answer_text": "You just washed it twice.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_6", + "correct_answer": "You just washed it twice.", + "original_options": { + "A": "You just washed it once.", + "B": "You just washed it twice.", + "C": "You just washed it three times.", + "D": "You just washed it four times, doing two extra rinses beyond what you did.", + "E": "You just washed it five times, continuing to rinse repeatedly before finishing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 12 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_1_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoonfuls of rice did I scoop just now?", + "answer": "You just scooped 2 spoonfuls of rice. The rice is to your front-right.", + "question_chinese": "我刚刚装了几勺大米？", + "answer_chinese": "你刚刚装了2勺大米。大米在你的右前方。", + "evidence": [ + { + "description": "You have packed a spoonful of rice.", + "timestamps": { + "start": "00:04:27", + "end": "00:04:31" + } + }, + { + "description": "You have filled the second spoonful of rice.", + "timestamps": { + "start": "00:04:31", + "end": "00:04:36" + } + } + ], + "question_timestamps": "00:04:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just scooped 2 spoonfuls of rice. The rice is to your front-right.", + "B": "You just scooped 1 spoonful of rice.", + "C": "You just scooped 3 spoonfuls of rice.", + "D": "You just scooped 1 spoonful of rice, and the rice is still to your front-right.", + "E": "You just scooped 4 spoonfuls of rice, with the rice remaining to your front-right." + }, + "answer_label": "A", + "answer_text": "You just scooped 2 spoonfuls of rice. The rice is to your front-right.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_1_T1", + "correct_answer": "You just scooped 2 spoonfuls of rice. The rice is to your front-right.", + "original_options": { + "A": "You just scooped 2 spoonfuls of rice. The rice is to your front-right.", + "B": "You just scooped 1 spoonful of rice.", + "C": "You just scooped 3 spoonfuls of rice.", + "D": "You just scooped 1 spoonful of rice, and the rice is still to your front-right.", + "E": "You just scooped 4 spoonfuls of rice, with the rice remaining to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 14 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_1_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoonfuls of rice did I scoop just now?", + "answer": "You just scooped 4 spoonfuls of rice. The rice is to your front-left.", + "question_chinese": "我刚刚装了几勺大米？", + "answer_chinese": "你刚刚装了4勺大米。大米在你的左前方。", + "evidence": [ + { + "description": "You have packed a spoonful of rice.", + "timestamps": { + "start": "00:04:27", + "end": "00:04:31" + } + }, + { + "description": "You have filled the second spoonful of rice.", + "timestamps": { + "start": "00:04:31", + "end": "00:04:36" + } + }, + { + "description": "You have packed the third spoonful of rice.", + "timestamps": { + "start": "00:04:36", + "end": "00:04:39" + } + }, + { + "description": "You have filled the fourth spoonful of rice.", + "timestamps": { + "start": "00:04:39", + "end": "00:04:43" + } + } + ], + "question_timestamps": "00:04:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just scooped 3 spoonfuls of rice.", + "B": "You just scooped 4 spoonfuls of rice. The rice is to your front-left.", + "C": "You just scooped 5 spoonfuls of rice.", + "D": "You just scooped 2 spoonfuls of rice, and the rice is positioned to your front-left.", + "E": "You just scooped 6 spoonfuls of rice, with the rice still to your front-left." + }, + "answer_label": "B", + "answer_text": "You just scooped 4 spoonfuls of rice. The rice is to your front-left.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_1_T2", + "correct_answer": "You just scooped 4 spoonfuls of rice. The rice is to your front-left.", + "original_options": { + "A": "You just scooped 3 spoonfuls of rice.", + "B": "You just scooped 4 spoonfuls of rice. The rice is to your front-left.", + "C": "You just scooped 5 spoonfuls of rice.", + "D": "You just scooped 2 spoonfuls of rice, and the rice is positioned to your front-left.", + "E": "You just scooped 6 spoonfuls of rice, with the rice still to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 14 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoons of rice did I just fill?", + "answer": "You just filled four spoons of rice.", + "question_chinese": "我刚刚装了几勺大米？", + "answer_chinese": "你刚刚装了四勺大米。", + "evidence": [ + { + "description": "You have packed a spoonful of rice.", + "timestamps": { + "start": "00:04:27", + "end": "00:04:31" + } + }, + { + "description": "You have filled the second spoonful of rice.", + "timestamps": { + "start": "00:04:31", + "end": "00:04:36" + } + }, + { + "description": "You have packed the third spoonful of rice.", + "timestamps": { + "start": "00:04:31", + "end": "00:04:39" + } + }, + { + "description": "You have filled the fourth spoonful of rice.", + "timestamps": { + "start": "00:04:39", + "end": "00:04:43" + } + } + ], + "question_timestamps": "00:05:08", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You just filled three spoons.", + "B": "You just filled five spoons.", + "C": "You just filled six spoons of rice, adding a couple more than you actually did.", + "D": "You just filled four spoons of rice.", + "E": "You just filled seven spoons of rice, continuing past the usual amount before stopping." + }, + "answer_label": "D", + "answer_text": "You just filled four spoons of rice.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_1", + "correct_answer": "You just filled four spoons of rice.", + "original_options": { + "A": "You just filled three spoons.", + "B": "You just filled five spoons.", + "C": "You just filled six spoons of rice, adding a couple more than you actually did.", + "D": "You just filled four spoons of rice.", + "E": "You just filled seven spoons of rice, continuing past the usual amount before stopping." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoons of rice did I just fill?", + "answer": "You just filled four spoons of rice.", + "question_chinese": "我刚刚装了几勺大米？", + "answer_chinese": "你刚刚装了四勺大米。", + "evidence": [ + { + "description": "You have packed a spoonful of rice.", + "timestamps": { + "start": "00:04:27", + "end": "00:04:31" + } + }, + { + "description": "You have filled the second spoonful of rice.", + "timestamps": { + "start": "00:04:31", + "end": "00:04:36" + } + }, + { + "description": "You have packed the third spoonful of rice.", + "timestamps": { + "start": "00:04:31", + "end": "00:04:39" + } + }, + { + "description": "You have filled the fourth spoonful of rice.", + "timestamps": { + "start": "00:04:39", + "end": "00:04:43" + } + } + ], + "question_timestamps": "00:28:47", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You just filled four spoons of rice.", + "B": "You just filled one spoon.", + "C": "You just filled three spoons.", + "D": "You just filled five spoons of rice, going one spoon beyond the amount you filled.", + "E": "You just filled six spoons of rice, keeping on filling spoons for longer than you did." + }, + "answer_label": "A", + "answer_text": "You just filled four spoons of rice.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_2", + "correct_answer": "You just filled four spoons of rice.", + "original_options": { + "A": "You just filled four spoons of rice.", + "B": "You just filled one spoon.", + "C": "You just filled three spoons.", + "D": "You just filled five spoons of rice, going one spoon beyond the amount you filled.", + "E": "You just filled six spoons of rice, keeping on filling spoons for longer than you did." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_3", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the faucet directly now?", + "answer": "You can touch it directly.", + "question_chinese": "我现在可以直接触碰到水龙头吗？", + "answer_chinese": "你可以直接触碰到。", + "evidence": [ + { + "description": "You can touch it directly.", + "timestamps": { + "start": "00:11:05", + "end": "00:11:14" + } + } + ], + "question_timestamps": "00:11:18", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch it directly.", + "B": "You cannot touch it directly." + }, + "answer_label": "A", + "answer_text": "You can touch it directly.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_3", + "correct_answer": "You can touch it directly.", + "original_options": { + "A": "You can touch it directly.", + "B": "You cannot touch it directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_4", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the faucet directly now?", + "answer": "You can't touch it directly.", + "question_chinese": "我现在可以直接触碰到水龙头吗？", + "answer_chinese": "你无法直接触碰到。", + "evidence": [ + { + "description": "You can't touch it directly.", + "timestamps": { + "start": "00:35:18", + "end": "00:35:23" + } + } + ], + "question_timestamps": "00:35:24", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't touch it directly.", + "B": "You can touch it directly." + }, + "answer_label": "A", + "answer_text": "You can't touch it directly.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_4", + "correct_answer": "You can't touch it directly.", + "original_options": { + "A": "You can't touch it directly.", + "B": "You can touch it directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_11", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the stove directly now?", + "answer": "You can now make direct contact.", + "question_chinese": "我现在可以直接触碰到灶台吗？", + "answer_chinese": "你现在可以直接触碰到。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "01:21:27", + "end": "01:21:32" + } + } + ], + "question_timestamps": "01:21:33", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now make direct contact.", + "B": "You cannot now make direct contact." + }, + "answer_label": "A", + "answer_text": "You can now make direct contact.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_11", + "correct_answer": "You can now make direct contact.", + "original_options": { + "A": "You can now make direct contact.", + "B": "You cannot now make direct contact." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_6_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the stovetop directly right now?", + "answer": "You can’t directly touch it right now. The stove is in the room behind you.", + "question_chinese": "我现在可以直接触碰到灶台吗？", + "answer_chinese": "你现在不可以直接触碰到。灶台在你身后的屋子里。", + "evidence": [ + { + "description": "You can't now make direct contact.", + "timestamps": { + "start": "01:22:35", + "end": "01:22:41" + } + } + ], + "question_timestamps": "01:22:43", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t directly touch it right now. The stove is in the room behind you.", + "B": "You can directly touch it right now. The stove is next to you." + }, + "answer_label": "A", + "answer_text": "You can’t directly touch it right now. The stove is in the room behind you.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_6_T1", + "correct_answer": "You can’t directly touch it right now. The stove is in the room behind you.", + "original_options": { + "A": "You can’t directly touch it right now. The stove is in the room behind you.", + "B": "You can directly touch it right now. The stove is next to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_6_T2", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the stovetop directly right now?", + "answer": "You can directly touch it right now. The stove is next to you.", + "question_chinese": "我现在可以直接触碰到灶台吗？", + "answer_chinese": "你现在可以直接触碰到。灶台在你旁边。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "01:30:10", + "end": "01:31:14" + } + } + ], + "question_timestamps": "01:31:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t directly touch it right now. The stove is in the room behind you.", + "B": "You can directly touch it right now. The stove is next to you." + }, + "answer_label": "B", + "answer_text": "You can directly touch it right now. The stove is next to you.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_6_T2", + "correct_answer": "You can directly touch it right now. The stove is next to you.", + "original_options": { + "A": "You can’t directly touch it right now. The stove is in the room behind you.", + "B": "You can directly touch it right now. The stove is next to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_7_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Do I need to go past the stairs to get back to the kitchen now?", + "answer": "Right now you need to go past the stairs. The stairs are behind you, and you just came down the stairs.", + "question_chinese": "我现在要回到厨房需要经过楼梯吗？", + "answer_chinese": "你现在需要经过楼梯。楼梯在你的身后，你刚刚从楼梯上下来。", + "evidence": [ + { + "description": "You went up the stairs to the roof.", + "timestamps": { + "start": "01:26:50", + "end": "01:27:07" + } + } + ], + "question_timestamps": "01:27:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you need to go past the stairs. The stairs are behind you, and you just came down the stairs.", + "B": "Right now, to get back to the kitchen you don’t need to go past the stairs. The stairs are on the other side of the house." + }, + "answer_label": "A", + "answer_text": "Right now you need to go past the stairs. The stairs are behind you, and you just came down the stairs.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_7_T2", + "correct_answer": "Right now you need to go past the stairs. The stairs are behind you, and you just came down the stairs.", + "original_options": { + "A": "Right now you need to go past the stairs. The stairs are behind you, and you just came down the stairs.", + "B": "Right now, to get back to the kitchen you don’t need to go past the stairs. The stairs are on the other side of the house." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 27 + ] + }, + { + "video_uid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239", + "video_length": "01:37:43", + "qid": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_7_T1", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Do I need to go past the stairs to get back to the kitchen now?", + "answer": "Right now, to get back to the kitchen you don’t need to go past the stairs. The stairs are on the other side of the house.", + "question_chinese": "我现在要回到厨房需要经过楼梯吗？", + "answer_chinese": "你现在回到厨房不需要经过楼梯。楼梯在房子另一边。", + "evidence": [ + { + "description": "You can go back to the kitchen without going through the stairs.", + "timestamps": { + "start": "01:27:08", + "end": "01:27:16" + } + } + ], + "question_timestamps": "01:27:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you need to go past the stairs. The stairs are behind you, and you just came down the stairs.", + "B": "Right now, to get back to the kitchen you don’t need to go past the stairs. The stairs are on the other side of the house." + }, + "answer_label": "B", + "answer_text": "Right now, to get back to the kitchen you don’t need to go past the stairs. The stairs are on the other side of the house.", + "q_id": "4f7fb690-4ee7-4555-87bd-0fe6171b4239_7_T1", + "correct_answer": "Right now, to get back to the kitchen you don’t need to go past the stairs. The stairs are on the other side of the house.", + "original_options": { + "A": "Right now you need to go past the stairs. The stairs are behind you, and you just came down the stairs.", + "B": "Right now, to get back to the kitchen you don’t need to go past the stairs. The stairs are on the other side of the house." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 21 + ] + } +] diff --git a/QAs/4f8112b1-0d67-44e3-aa48-117a13ffbc60-mcq.json b/QAs/4f8112b1-0d67-44e3-aa48-117a13ffbc60-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..cc0b1b1aca628d3ef195a890626d81bac21918de --- /dev/null +++ b/QAs/4f8112b1-0d67-44e3-aa48-117a13ffbc60-mcq.json @@ -0,0 +1,1909 @@ +[ + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many people have I seen in this house in total?", + "answer": "three people.", + "question_chinese": "到目前为止，我在这个房子里一共看到了几个人？", + "answer_chinese": "3个人。", + "evidence": [ + { + "description": "Saw a woman in a pink top.", + "timestamps": { + "start": "00:51:40", + "end": "00:51:59" + } + }, + { + "description": "Saw a woman in a white top with a tattoo on her arm, a man in a white top, and yourself.", + "timestamps": { + "start": "00:52:15", + "end": "00:52:29" + } + } + ], + "question_timestamps": "00:53:02", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two people.", + "B": "three people.", + "C": "Five people.", + "D": "Seven people in total, counting everyone you’ve encountered so far while moving through the house.", + "E": "Nine people altogether, including everyone you saw earlier in the house and those you noticed recently." + }, + "answer_label": "B", + "answer_text": "three people.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_7", + "correct_answer": "three people.", + "original_options": { + "A": "Two people.", + "B": "three people.", + "C": "Five people.", + "D": "Seven people in total, counting everyone you’ve encountered so far while moving through the house.", + "E": "Nine people altogether, including everyone you saw earlier in the house and those you noticed recently." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many people have I seen in this house in total?", + "answer": "Three people.", + "question_chinese": "到目前为止，我在这个房子里一共看到了几个人？", + "answer_chinese": "3个人。", + "evidence": [ + { + "description": "Saw a woman in a pink top.", + "timestamps": { + "start": "00:51:40", + "end": "00:51:59" + } + }, + { + "description": "Saw a woman in a white top with a tattoo on her arm, a man in a white top, and yourself.", + "timestamps": { + "start": "00:52:15", + "end": "00:52:29" + } + } + ], + "question_timestamps": "00:54:03", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two people.", + "B": "Three people.", + "C": "Five people.", + "D": "Seven people in total, counting everyone you’ve encountered so far while moving through the house.", + "E": "Nine people altogether, including everyone you saw earlier in the house and those you noticed recently." + }, + "answer_label": "B", + "answer_text": "Three people.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_8", + "correct_answer": "Three people.", + "original_options": { + "A": "Two people.", + "B": "Three people.", + "C": "Five people.", + "D": "Seven people in total, counting everyone you’ve encountered so far while moving through the house.", + "E": "Nine people altogether, including everyone you saw earlier in the house and those you noticed recently." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_9_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In the room, how many bathtubs did I see?", + "answer": "You saw a bathtub; it was in front of you.", + "question_chinese": "在房间里，我看到多少个浴缸？", + "answer_chinese": "你看到一个浴缸，它在你前面。", + "evidence": [ + { + "description": "The bathtub is in the room.", + "timestamps": { + "start": "00:00:38", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw no bathtubs.", + "B": "You saw two bathtubs.", + "C": "You saw a bathtub, but it was behind you when you looked around the room.", + "D": "You saw a bathtub; it was in front of you.", + "E": "You saw three bathtubs, spaced out along the left side rather than directly in front of you." + }, + "answer_label": "D", + "answer_text": "You saw a bathtub; it was in front of you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_9_T1", + "correct_answer": "You saw a bathtub; it was in front of you.", + "original_options": { + "A": "You saw no bathtubs.", + "B": "You saw two bathtubs.", + "C": "You saw a bathtub, but it was behind you when you looked around the room.", + "D": "You saw a bathtub; it was in front of you.", + "E": "You saw three bathtubs, spaced out along the left side rather than directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_9_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In the room, how many bathtubs did I see?", + "answer": "You didn’t see the bathtub. The bathtub was on the left side of the room in front of you, but it was blocked so you couldn’t see it.", + "question_chinese": "在房间里，我看到多少个浴缸？", + "answer_chinese": "你没有看到浴缸，浴缸在你前面房间的左边。被挡住了你看不到。", + "evidence": [ + { + "description": "Zero. You don't see a bathtub.", + "timestamps": { + "start": "00:17:14", + "end": "00:17:19" + } + } + ], + "question_timestamps": "00:17:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw a bathtub.", + "B": "You saw two bathtubs.", + "C": "You didn’t see the bathtub; it was on the right side in front of you and blocked from view.", + "D": "You didn’t see the bathtub; it was on the left side behind you, blocked so you couldn’t see it.", + "E": "You didn’t see the bathtub. The bathtub was on the left side of the room in front of you, but it was blocked so you couldn’t see it." + }, + "answer_label": "E", + "answer_text": "You didn’t see the bathtub. The bathtub was on the left side of the room in front of you, but it was blocked so you couldn’t see it.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_9_T2", + "correct_answer": "You didn’t see the bathtub. The bathtub was on the left side of the room in front of you, but it was blocked so you couldn’t see it.", + "original_options": { + "A": "You saw a bathtub.", + "B": "You saw two bathtubs.", + "C": "You didn’t see the bathtub; it was on the right side in front of you and blocked from view.", + "D": "You didn’t see the bathtub; it was on the left side behind you, blocked so you couldn’t see it.", + "E": "You didn’t see the bathtub. The bathtub was on the left side of the room in front of you, but it was blocked so you couldn’t see it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 21 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_10_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many times did I cut the wooden board?", + "answer": "The wooden board was in your right hand, and you didn’t cut the wooden board.", + "question_chinese": "我切割了多少次木板？", + "answer_chinese": "木板在你右手上，你没有切割木板。", + "evidence": [ + { + "description": "You didn't cut the wooden board.", + "timestamps": { + "start": "00:05:54", + "end": "00:05:59" + } + } + ], + "question_timestamps": "00:06:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You cut it once.", + "B": "You cut it twice.", + "C": "The wooden board was in your left hand, and you didn’t cut the wooden board at all.", + "D": "The wooden board was in your right hand, and you didn’t cut the wooden board.", + "E": "You cut the wooden board three times while holding it in your right hand, making three separate cuts." + }, + "answer_label": "D", + "answer_text": "The wooden board was in your right hand, and you didn’t cut the wooden board.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_10_T1", + "correct_answer": "The wooden board was in your right hand, and you didn’t cut the wooden board.", + "original_options": { + "A": "You cut it once.", + "B": "You cut it twice.", + "C": "The wooden board was in your left hand, and you didn’t cut the wooden board at all.", + "D": "The wooden board was in your right hand, and you didn’t cut the wooden board.", + "E": "You cut the wooden board three times while holding it in your right hand, making three separate cuts." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 18 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_10_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many times did I cut the wooden board?", + "answer": "You cut it once, and the wooden board was in front of you.", + "question_chinese": "我切割了多少次木板？", + "answer_chinese": "你切割了一次，木板在你前面", + "evidence": [ + { + "description": "You cut it twice.", + "timestamps": { + "start": "00:17:10", + "end": "00:17:15" + } + } + ], + "question_timestamps": "00:17:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You cut it twice.", + "B": "You didn’t cut it.", + "C": "You cut it three times, with the wooden board staying in front of you the whole time.", + "D": "You cut it once, but the wooden board was off to your left instead of directly in front.", + "E": "You cut it once, and the wooden board was in front of you." + }, + "answer_label": "E", + "answer_text": "You cut it once, and the wooden board was in front of you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_10_T2", + "correct_answer": "You cut it once, and the wooden board was in front of you.", + "original_options": { + "A": "You cut it twice.", + "B": "You didn’t cut it.", + "C": "You cut it three times, with the wooden board staying in front of you the whole time.", + "D": "You cut it once, but the wooden board was off to your left instead of directly in front.", + "E": "You cut it once, and the wooden board was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 18 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_3", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What places did I pass when I went out to get things and came back into the house?", + "answer": "After you went out, you first went up the steps, walked to the trash can in the front left, then walked straight from the trash can to the front left to the side of the car. After taking things from the trunk, you returned to the house along the original path to the front left.", + "question_chinese": "我刚才出去拿东西到进屋，经过了哪些地方？", + "answer_chinese": "你出门后，先走上台阶，走到左前方的垃圾箱，然后从垃圾箱向左前方直走，来到车旁边，在后备箱拿完东西后，向左前方原路返回进屋。", + "evidence": [ + { + "description": "After going out, walk up the steps to the trash can in the front left.", + "timestamps": { + "start": "00:14:32", + "end": "00:14:49" + } + }, + { + "description": "Walk straight from the trash can to the front left to the side of the car.", + "timestamps": { + "start": "00:14:54", + "end": "00:15:08" + } + }, + { + "description": "After taking things from the trunk, return to the house along the original path to the front left.", + "timestamps": { + "start": "00:15:29", + "end": "00:15:46" + } + } + ], + "question_timestamps": "00:16:26", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You went down the steps.", + "B": "You went to the front right.", + "C": "After you went out, you went up the steps and walked to a trash can on the front right.", + "D": "After you went out, you first went up the steps, walked to the trash can in the front left, then walked straight from the trash can to the front left to the side of the car. After taking things from the trunk, you returned to the house along the original path to the front left.", + "E": "You went up the steps, went straight to the car first, then walked back past the trash can to the house." + }, + "answer_label": "D", + "answer_text": "After you went out, you first went up the steps, walked to the trash can in the front left, then walked straight from the trash can to the front left to the side of the car. After taking things from the trunk, you returned to the house along the original path to the front left.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_3", + "correct_answer": "After you went out, you first went up the steps, walked to the trash can in the front left, then walked straight from the trash can to the front left to the side of the car. After taking things from the trunk, you returned to the house along the original path to the front left.", + "original_options": { + "A": "You went down the steps.", + "B": "You went to the front right.", + "C": "After you went out, you went up the steps and walked to a trash can on the front right.", + "D": "After you went out, you first went up the steps, walked to the trash can in the front left, then walked straight from the trash can to the front left to the side of the car. After taking things from the trunk, you returned to the house along the original path to the front left.", + "E": "You went up the steps, went straight to the car first, then walked back past the trash can to the house." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 21 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_4", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What places did I pass when I went out to get things and came back into the house?", + "answer": "After you went out, you first went up the steps, walked to the trash can in the front left, then walked straight from the trash can to the front left to the side of the car. After taking things from the trunk, you returned to the house along the original path to the front left.", + "question_chinese": "我刚才出去拿东西到进屋，经过了哪些地方？", + "answer_chinese": "你出门后，先走上台阶，走到左前方的垃圾箱，然后从垃圾箱向左前方直走，来到车旁边，在后备箱拿完东西后，向左前方原路返回进屋。", + "evidence": [ + { + "description": "After going out, walk up the steps to the trash can in the front left.", + "timestamps": { + "start": "00:14:32", + "end": "00:14:49" + } + }, + { + "description": "Walk straight from the trash can to the front left to the side of the car.", + "timestamps": { + "start": "00:14:54", + "end": "00:15:08" + } + }, + { + "description": "After taking things from the trunk, return to the house along the original path to the front left.", + "timestamps": { + "start": "00:15:29", + "end": "00:15:46" + } + } + ], + "question_timestamps": "00:19:54", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You went to the front right.", + "B": "You went to the back right.", + "C": "After you went out, you went up the steps to a recycling bin in the back left, then to the car.", + "D": "You went up the steps to a compost bin on the back right, then to the driver’s side before returning.", + "E": "After you went out, you first went up the steps, walked to the trash can in the front left, then walked straight from the trash can to the front left to the side of the car. After taking things from the trunk, you returned to the house along the original path to the front left." + }, + "answer_label": "E", + "answer_text": "After you went out, you first went up the steps, walked to the trash can in the front left, then walked straight from the trash can to the front left to the side of the car. After taking things from the trunk, you returned to the house along the original path to the front left.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_4", + "correct_answer": "After you went out, you first went up the steps, walked to the trash can in the front left, then walked straight from the trash can to the front left to the side of the car. After taking things from the trunk, you returned to the house along the original path to the front left.", + "original_options": { + "A": "You went to the front right.", + "B": "You went to the back right.", + "C": "After you went out, you went up the steps to a recycling bin in the back left, then to the car.", + "D": "You went up the steps to a compost bin on the back right, then to the driver’s side before returning.", + "E": "After you went out, you first went up the steps, walked to the trash can in the front left, then walked straight from the trash can to the front left to the side of the car. After taking things from the trunk, you returned to the house along the original path to the front left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 21 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_5", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What positions on the car did I check in order when I was looking for something just now?", + "answer": "You walked into the main compartment and rummaged around, then walked out of the main compartment, came to the left side of the car and opened the four small compartments on the left to check in turn, then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "question_chinese": "我刚才在找东西的时候，依次查看了车上的哪些位置？", + "answer_chinese": "你走进主车厢翻找了一圈，然后走出主车厢，来到车的左侧并依次打开了左边的4个小隔间查看，然后走到车的右侧，依次打开了右边的4个小隔间查看，最后离开了。", + "evidence": [ + { + "description": "Walked into the main compartment and rummaged around.", + "timestamps": { + "start": "00:48:48", + "end": "00:49:50" + } + }, + { + "description": "Walked out of the main compartment, came to the left side of the car and opened the 4 small compartments on the left to check in turn.", + "timestamps": { + "start": "00:49:51", + "end": "00:50:39" + } + }, + { + "description": "Then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "timestamps": { + "start": "00:50:37", + "end": "00:51:14" + } + } + ], + "question_timestamps": "00:51:25", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You checked the right side first.", + "B": "You walked into the main compartment and rummaged around, then walked out of the main compartment, came to the left side of the car and opened the four small compartments on the left to check in turn, then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "C": "You only checked the left side.", + "D": "You rummaged in the main compartment, then opened four small compartments on the right before checking the left side.", + "E": "You left the main compartment and checked only three small compartments on each side before leaving the car." + }, + "answer_label": "B", + "answer_text": "You walked into the main compartment and rummaged around, then walked out of the main compartment, came to the left side of the car and opened the four small compartments on the left to check in turn, then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_5", + "correct_answer": "You walked into the main compartment and rummaged around, then walked out of the main compartment, came to the left side of the car and opened the four small compartments on the left to check in turn, then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "original_options": { + "A": "You checked the right side first.", + "B": "You walked into the main compartment and rummaged around, then walked out of the main compartment, came to the left side of the car and opened the four small compartments on the left to check in turn, then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "C": "You only checked the left side.", + "D": "You rummaged in the main compartment, then opened four small compartments on the right before checking the left side.", + "E": "You left the main compartment and checked only three small compartments on each side before leaving the car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_6", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What positions on the car did I check in order when I was looking for something just now?", + "answer": "You walked into the main compartment and rummaged around, then walked out of the main compartment, came to the left side of the car and opened the four small compartments on the left to check in turn, then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "question_chinese": "我刚才在找东西的时候，依次查看了车上的哪些位置？", + "answer_chinese": "你走进主车厢翻找了一圈，然后走出主车厢，来到车的左侧并依次打开了左边的4个小隔间查看，然后走到车的右侧，依次打开了右边的4个小隔间查看，最后离开了。", + "evidence": [ + { + "description": "Walked into the main compartment and rummaged around.", + "timestamps": { + "start": "00:48:48", + "end": "00:49:50" + } + }, + { + "description": "Walked out of the main compartment, came to the left side of the car and opened the 4 small compartments on the left to check in turn.", + "timestamps": { + "start": "00:49:51", + "end": "00:50:39" + } + }, + { + "description": "Then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "timestamps": { + "start": "00:50:37", + "end": "00:51:14" + } + } + ], + "question_timestamps": "00:53:32", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You checked the right side twice.", + "B": "You checked five compartments on the left.", + "C": "You walked into the main compartment and rummaged around, then walked out of the main compartment, came to the left side of the car and opened the four small compartments on the left to check in turn, then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "D": "You rummaged in the main compartment, then checked four small compartments on the right and never checked the left.", + "E": "You checked four large crates on the left, then only three small boxes on the right before leaving." + }, + "answer_label": "C", + "answer_text": "You walked into the main compartment and rummaged around, then walked out of the main compartment, came to the left side of the car and opened the four small compartments on the left to check in turn, then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_6", + "correct_answer": "You walked into the main compartment and rummaged around, then walked out of the main compartment, came to the left side of the car and opened the four small compartments on the left to check in turn, then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "original_options": { + "A": "You checked the right side twice.", + "B": "You checked five compartments on the left.", + "C": "You walked into the main compartment and rummaged around, then walked out of the main compartment, came to the left side of the car and opened the four small compartments on the left to check in turn, then walked to the right side of the car, opened the 4 small compartments on the right to check in turn, and finally left.", + "D": "You rummaged in the main compartment, then checked four small compartments on the right and never checked the left.", + "E": "You checked four large crates on the left, then only three small boxes on the right before leaving." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 18 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_11", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I cut the water pipe while installing it just now?", + "answer": "Cut the water pipe two times in total.", + "question_chinese": "我刚才在装水管的过程中，剪了几次水管？", + "answer_chinese": "一共剪了2次水管。", + "evidence": [ + { + "description": "Cut a section of water pipe from a coil of water pipe.", + "timestamps": { + "start": "00:23:59", + "end": "00:24:20" + } + }, + { + "description": "Cut a small section of excess length from the installed water pipe.", + "timestamps": { + "start": "00:26:52", + "end": "00:27:02" + } + } + ], + "question_timestamps": "00:28:26", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Cut the water pipe once total.", + "B": "Cut the water pipe three times total.", + "C": "Cut the water pipe four times in total while installing it just now.", + "D": "Cut the water pipe five times in total during the installation you just did.", + "E": "Cut the water pipe two times in total." + }, + "answer_label": "E", + "answer_text": "Cut the water pipe two times in total.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_11", + "correct_answer": "Cut the water pipe two times in total.", + "original_options": { + "A": "Cut the water pipe once total.", + "B": "Cut the water pipe three times total.", + "C": "Cut the water pipe four times in total while installing it just now.", + "D": "Cut the water pipe five times in total during the installation you just did.", + "E": "Cut the water pipe two times in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 14 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_12", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I cut the water pipe while installing it just now?", + "answer": "Cut the water pipe two times in total.", + "question_chinese": "我刚才在装水管的过程中，剪了几次水管？", + "answer_chinese": "一共剪了2次水管。", + "evidence": [ + { + "description": "Cut a section of water pipe from a coil of water pipe.", + "timestamps": { + "start": "00:23:59", + "end": "00:24:20" + } + }, + { + "description": "Cut a small section of excess length from the installed water pipe.", + "timestamps": { + "start": "00:26:52", + "end": "00:27:02" + } + } + ], + "question_timestamps": "00:33:57", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Cut the water pipe once total.", + "B": "Cut the water pipe four times total.", + "C": "Cut the water pipe three times in total while installing it just now.", + "D": "Cut the water pipe five times in total during the installation you just did.", + "E": "Cut the water pipe two times in total." + }, + "answer_label": "E", + "answer_text": "Cut the water pipe two times in total.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_12", + "correct_answer": "Cut the water pipe two times in total.", + "original_options": { + "A": "Cut the water pipe once total.", + "B": "Cut the water pipe four times total.", + "C": "Cut the water pipe three times in total while installing it just now.", + "D": "Cut the water pipe five times in total during the installation you just did.", + "E": "Cut the water pipe two times in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 14 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_9", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many fasteners did I just install on the yellow board on the wall?", + "answer": "A total of 10 fasteners were installed.", + "question_chinese": "我刚才给墙上那块黄色的板材安装了几个固定器？", + "answer_chinese": "一共安装了10个固定器。", + "evidence": [ + { + "description": "Installed 5 fasteners in sequence on the right side of the board.", + "timestamps": { + "start": "00:53:11", + "end": "00:53:59" + } + }, + { + "description": "Installed 5 fasteners in sequence on the left side of the board.", + "timestamps": { + "start": "00:54:03", + "end": "00:55:16" + } + } + ], + "question_timestamps": "00:55:21", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "A total of 9 fasteners.", + "B": "A total of 11 fasteners.", + "C": "A total of 8 fasteners were installed on the yellow board on the wall.", + "D": "A total of 10 fasteners were installed.", + "E": "A total of 12 fasteners were installed on the yellow board mounted on the wall." + }, + "answer_label": "D", + "answer_text": "A total of 10 fasteners were installed.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_9", + "correct_answer": "A total of 10 fasteners were installed.", + "original_options": { + "A": "A total of 9 fasteners.", + "B": "A total of 11 fasteners.", + "C": "A total of 8 fasteners were installed on the yellow board on the wall.", + "D": "A total of 10 fasteners were installed.", + "E": "A total of 12 fasteners were installed on the yellow board mounted on the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_10", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many fasteners did I just install on the yellow board on the wall?", + "answer": "A total of 10 fasteners were installed.", + "question_chinese": "我刚才给墙上那块黄色的板材安装了几个固定器？", + "answer_chinese": "一共安装了10个固定器。", + "evidence": [ + { + "description": "Installed 5 fasteners in sequence on the right side of the board.", + "timestamps": { + "start": "00:53:11", + "end": "00:53:59" + } + }, + { + "description": "Installed 5 fasteners in sequence on the left side of the board.", + "timestamps": { + "start": "00:54:03", + "end": "00:55:16" + } + } + ], + "question_timestamps": "00:56:17", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "A total of 8 fasteners.", + "B": "A total of 12 fasteners.", + "C": "A total of 9 fasteners were installed on the yellow board on the wall.", + "D": "A total of 10 fasteners were installed.", + "E": "A total of 11 fasteners were installed on the yellow board mounted on the wall." + }, + "answer_label": "D", + "answer_text": "A total of 10 fasteners were installed.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_10", + "correct_answer": "A total of 10 fasteners were installed.", + "original_options": { + "A": "A total of 8 fasteners.", + "B": "A total of 12 fasteners.", + "C": "A total of 9 fasteners were installed on the yellow board on the wall.", + "D": "A total of 10 fasteners were installed.", + "E": "A total of 11 fasteners were installed on the yellow board mounted on the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_6_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I cut the water pipe?", + "answer": "You cut the water pipe zero times; the water pipe was in front of you.", + "question_chinese": "我剪了多少次水管？", + "answer_chinese": "你剪了零次水管，水管在你前面。", + "evidence": [ + { + "description": "You have cut the water pipe zero times.", + "timestamps": { + "start": "00:25:56", + "end": "00:26:04" + } + } + ], + "question_timestamps": "00:26:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You cut the water pipe once.", + "B": "You cut the water pipe zero times; the water pipe was in front of you.", + "C": "You cut the water pipe twice.", + "D": "You cut the water pipe three times; the water pipe was in front of you.", + "E": "You cut the water pipe four times; the water pipe stayed in front of you." + }, + "answer_label": "B", + "answer_text": "You cut the water pipe zero times; the water pipe was in front of you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_6_T2", + "correct_answer": "You cut the water pipe zero times; the water pipe was in front of you.", + "original_options": { + "A": "You cut the water pipe once.", + "B": "You cut the water pipe zero times; the water pipe was in front of you.", + "C": "You cut the water pipe twice.", + "D": "You cut the water pipe three times; the water pipe was in front of you.", + "E": "You cut the water pipe four times; the water pipe stayed in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_6_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I cut the water pipe?", + "answer": "You cut the water pipe twice; it was in front of you.", + "question_chinese": "我剪了多少次水管？", + "answer_chinese": "你剪了2次水管，它在你前面。", + "evidence": [ + { + "description": "Cut a section of water pipe from a coil of water pipe.", + "timestamps": { + "start": "00:23:59", + "end": "00:24:20" + } + }, + { + "description": "Cut a small section of excess length from the installed water pipe.", + "timestamps": { + "start": "00:26:52", + "end": "00:27:02" + } + } + ], + "question_timestamps": "00:27:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You cut the water pipe twice; it was in front of you.", + "B": "You cut the water pipe zero times.", + "C": "You cut the water pipe once.", + "D": "You cut the water pipe three times; it was in front of you the whole time.", + "E": "You cut the water pipe four times; it was in front of you as you worked." + }, + "answer_label": "A", + "answer_text": "You cut the water pipe twice; it was in front of you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_6_T1", + "correct_answer": "You cut the water pipe twice; it was in front of you.", + "original_options": { + "A": "You cut the water pipe twice; it was in front of you.", + "B": "You cut the water pipe zero times.", + "C": "You cut the water pipe once.", + "D": "You cut the water pipe three times; it was in front of you the whole time.", + "E": "You cut the water pipe four times; it was in front of you as you worked." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 16 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_3_T2", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I was outside, how many red sedans did I see?", + "answer": "You saw a red sedan. You had just passed it, and it was behind you.", + "question_chinese": "我在外面时，我看到多少辆红色的小轿车。", + "answer_chinese": "你看到一辆红色的小轿车，你刚刚经过它，它在你后面。", + "evidence": [ + { + "description": "You see a red car.", + "timestamps": { + "start": "00:48:43", + "end": "00:48:47" + } + } + ], + "question_timestamps": "00:48:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Zero red sedans.", + "B": "Two red sedans.", + "C": "You saw one red sedan outside, and it stayed in front of you as you approached.", + "D": "You saw a red sedan. You had just passed it, and it was behind you.", + "E": "You saw three red sedans outside, parked along the road as you walked past them." + }, + "answer_label": "D", + "answer_text": "You saw a red sedan. You had just passed it, and it was behind you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_3_T2", + "correct_answer": "You saw a red sedan. You had just passed it, and it was behind you.", + "original_options": { + "A": "Zero red sedans.", + "B": "Two red sedans.", + "C": "You saw one red sedan outside, and it stayed in front of you as you approached.", + "D": "You saw a red sedan. You had just passed it, and it was behind you.", + "E": "You saw three red sedans outside, parked along the road as you walked past them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 15 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_3_T1", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I was outside, how many red sedans did I see?", + "answer": "Zero. You didn’t see a red sedan. You were inside a white car, and the red sedan was to the left of the white car.", + "question_chinese": "我在外面时，我看到多少辆红色的小轿车。", + "answer_chinese": "零，你没有看到红色的小轿车，你在白色的车内，红色小轿车在白色车的左边。", + "evidence": [ + { + "description": "Zero. You don't see a red car.", + "timestamps": { + "start": "00:48:48", + "end": "00:48:51" + } + } + ], + "question_timestamps": "00:48:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One red sedan.", + "B": "Two red sedans.", + "C": "Zero. You didn’t see a red sedan. You were inside a white car, and the red sedan was to the left of the white car.", + "D": "Three red sedans while you were outside, near the white car you were in.", + "E": "Four red sedans outside, with at least one parked beside the white car." + }, + "answer_label": "C", + "answer_text": "Zero. You didn’t see a red sedan. You were inside a white car, and the red sedan was to the left of the white car.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_3_T1", + "correct_answer": "Zero. You didn’t see a red sedan. You were inside a white car, and the red sedan was to the left of the white car.", + "original_options": { + "A": "One red sedan.", + "B": "Two red sedans.", + "C": "Zero. You didn’t see a red sedan. You were inside a white car, and the red sedan was to the left of the white car.", + "D": "Three red sedans while you were outside, near the white car you were in.", + "E": "Four red sedans outside, with at least one parked beside the white car." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 13 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_1_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I am in the bathtub—who is closer to me?", + "answer": "The girl wearing a green headscarf was closer to you; she was to your right.", + "question_chinese": "我在浴缸里面，谁离我更近？", + "answer_chinese": "戴绿色头巾的女生离你更近，她在你右边", + "evidence": [ + { + "description": "The girl with a green headscarf is closer to you on your right front.", + "timestamps": { + "start": "00:02:48", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The girl wearing a green headscarf was closer to you; she was to your right.", + "B": "The boy in white was closer.", + "C": "Neither was closer to you.", + "D": "The girl wearing a green headscarf was closer to you, but she was to your left.", + "E": "The boy in white was closer to you; he was standing to your left beside the tub." + }, + "answer_label": "A", + "answer_text": "The girl wearing a green headscarf was closer to you; she was to your right.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_1_T1", + "correct_answer": "The girl wearing a green headscarf was closer to you; she was to your right.", + "original_options": { + "A": "The girl wearing a green headscarf was closer to you; she was to your right.", + "B": "The boy in white was closer.", + "C": "Neither was closer to you.", + "D": "The girl wearing a green headscarf was closer to you, but she was to your left.", + "E": "The boy in white was closer to you; he was standing to your left beside the tub." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_1_T2", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I am in the bathtub—who is closer to me?", + "answer": "The boy in white was closer to you; he was to your left.", + "question_chinese": "我在浴缸里面，谁离我更近？", + "answer_chinese": "穿白衣服的男生离你更近，他在你左边。", + "evidence": [ + { + "description": "The boy in white is closer to you.", + "timestamps": { + "start": "00:09:15", + "end": "00:09:18" + } + } + ], + "question_timestamps": "00:09:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The girl was closer to you.", + "B": "Neither was closer to you.", + "C": "The boy in white was closer to you, but he was to your right instead of your left.", + "D": "The boy in white was closer to you; he was to your left.", + "E": "The boy in white was closer to you; he was positioned behind you near the back of the tub." + }, + "answer_label": "D", + "answer_text": "The boy in white was closer to you; he was to your left.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_1_T2", + "correct_answer": "The boy in white was closer to you; he was to your left.", + "original_options": { + "A": "The girl was closer to you.", + "B": "Neither was closer to you.", + "C": "The boy in white was closer to you, but he was to your right instead of your left.", + "D": "The boy in white was closer to you; he was to your left.", + "E": "The boy in white was closer to you; he was positioned behind you near the back of the tub." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 19 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_8_T1", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there a ladder in the room I’m in?", + "answer": "There was no ladder in the room you were in. The ladder was next to the doorway of another room, and the room with the ladder was behind you.", + "question_chinese": "我在的房间里有梯子吗？", + "answer_chinese": "你在的房间里没有梯子，梯子在另一个房间门口旁边，放梯子的房间在你的后面。", + "evidence": [ + { + "description": "There is no ladder in the room you're in.", + "timestamps": { + "start": "00:45:58", + "end": "00:46:01" + } + } + ], + "question_timestamps": "00:46:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Yes, there was one ladder.", + "B": "No ladder anywhere nearby.", + "C": "There was no ladder in the room you were in. The ladder was next to the doorway of another room, and the room with the ladder was behind you.", + "D": "There was one ladder in your room, positioned beside the doorway and visible in front of you.", + "E": "There was one ladder, but it was in the room ahead of you, leaning next to the doorway." + }, + "answer_label": "C", + "answer_text": "There was no ladder in the room you were in. The ladder was next to the doorway of another room, and the room with the ladder was behind you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_8_T1", + "correct_answer": "There was no ladder in the room you were in. The ladder was next to the doorway of another room, and the room with the ladder was behind you.", + "original_options": { + "A": "Yes, there was one ladder.", + "B": "No ladder anywhere nearby.", + "C": "There was no ladder in the room you were in. The ladder was next to the doorway of another room, and the room with the ladder was behind you.", + "D": "There was one ladder in your room, positioned beside the doorway and visible in front of you.", + "E": "There was one ladder, but it was in the room ahead of you, leaning next to the doorway." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 18 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_8_T2", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there a ladder in the room I’m in?", + "answer": "There was a ladder in the room you were in; it was behind you.", + "question_chinese": "我在的房间里有梯子吗？", + "answer_chinese": "你在的房间里有梯子，它在你身后。", + "evidence": [ + { + "description": "There is ladder in the room you're in.", + "timestamps": { + "start": "00:48:09", + "end": "00:48:13" + } + } + ], + "question_timestamps": "00:48:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There was no ladder in the room.", + "B": "The ladder was in front of you.", + "C": "There was a ladder in the room you were in, but it was to your left rather than behind you.", + "D": "There was a ladder in the room you were in; it was behind you.", + "E": "There was no ladder in the room you were in; the ladder was next to another room’s doorway behind you." + }, + "answer_label": "D", + "answer_text": "There was a ladder in the room you were in; it was behind you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_8_T2", + "correct_answer": "There was a ladder in the room you were in; it was behind you.", + "original_options": { + "A": "There was no ladder in the room.", + "B": "The ladder was in front of you.", + "C": "There was a ladder in the room you were in, but it was to your left rather than behind you.", + "D": "There was a ladder in the room you were in; it was behind you.", + "E": "There was no ladder in the room you were in; the ladder was next to another room’s doorway behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 21 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_19", + "qid_num": 25, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get this white box in my hand?", + "answer": "From the place where the tiles are placed at the door.", + "question_chinese": "我手上这个白色的盒子是从哪里拿的？", + "answer_chinese": "从门口放瓷砖的地方拿的。", + "evidence": [ + { + "description": "Picked up a white box from the place where the tiles are placed at the door.", + "timestamps": { + "start": "00:45:54", + "end": "00:45:59" + } + } + ], + "question_timestamps": "00:46:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the place where the tiles are placed at the door.", + "B": "From the shelf by the window.", + "C": "From the cabinet under the sink.", + "D": "From the desk where the papers are piled, after you reached over and picked it up.", + "E": "From the basket beside the couch in the living room, where you grabbed it before leaving." + }, + "answer_label": "A", + "answer_text": "From the place where the tiles are placed at the door.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_19", + "correct_answer": "From the place where the tiles are placed at the door.", + "original_options": { + "A": "From the place where the tiles are placed at the door.", + "B": "From the shelf by the window.", + "C": "From the cabinet under the sink.", + "D": "From the desk where the papers are piled, after you reached over and picked it up.", + "E": "From the basket beside the couch in the living room, where you grabbed it before leaving." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_20", + "qid_num": 26, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get this white box in my hand?", + "answer": "From the place where the tiles are placed at the door.", + "question_chinese": "我手上这个白色的盒子是从哪里拿的？", + "answer_chinese": "从门口放瓷砖的地方拿的。", + "evidence": [ + { + "description": "Picked up a white box from the place where the tiles are placed at the door.", + "timestamps": { + "start": "00:45:54", + "end": "00:45:59" + } + } + ], + "question_timestamps": "00:46:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the desk in the office.", + "B": "From the place where the tiles are placed at the door.", + "C": "From the basket beside the couch.", + "D": "From the cabinet under the sink in the kitchen, where you opened the door and took it out.", + "E": "From the shelf where the books are stacked near the window, picking it up from the top." + }, + "answer_label": "B", + "answer_text": "From the place where the tiles are placed at the door.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_20", + "correct_answer": "From the place where the tiles are placed at the door.", + "original_options": { + "A": "From the desk in the office.", + "B": "From the place where the tiles are placed at the door.", + "C": "From the basket beside the couch.", + "D": "From the cabinet under the sink in the kitchen, where you opened the door and took it out.", + "E": "From the shelf where the books are stacked near the window, picking it up from the top." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_7_T1", + "qid_num": 27, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "After I finished using the pen, where did I put it?", + "answer": "You put it next to the table, and the table was to your right.", + "question_chinese": "我用完笔后，把它放在哪里了？", + "answer_chinese": "你把它放在桌子的旁边，桌子在你右边。", + "evidence": [ + { + "description": "You put it beside the table.", + "timestamps": { + "start": "00:43:38", + "end": "00:43:43" + } + } + ], + "question_timestamps": "00:43:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Next to the table ahead.", + "B": "Next to the table left.", + "C": "You put it next to the table, and the table was to your right.", + "D": "You placed it beside the table, and the table was positioned directly in front of you.", + "E": "You set it next to the table, with the table located slightly behind you as you stood." + }, + "answer_label": "C", + "answer_text": "You put it next to the table, and the table was to your right.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_7_T1", + "correct_answer": "You put it next to the table, and the table was to your right.", + "original_options": { + "A": "Next to the table ahead.", + "B": "Next to the table left.", + "C": "You put it next to the table, and the table was to your right.", + "D": "You placed it beside the table, and the table was positioned directly in front of you.", + "E": "You set it next to the table, with the table located slightly behind you as you stood." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_7_T2", + "qid_num": 28, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "After I finished using the pen, where did I put it?", + "answer": "You put it on the cabinet next to you, and the cabinet was right in front of you.", + "question_chinese": "我用完笔后，把它放在哪里了？", + "answer_chinese": "你把它放在旁边的柜子上，柜子就在你前面。", + "evidence": [ + { + "description": "You put it on the cabinet next to you.", + "timestamps": { + "start": "00:46:36", + "end": "00:46:39" + } + } + ], + "question_timestamps": "00:46:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "On the cabinet behind you.", + "B": "On the cabinet left.", + "C": "You put it on the cabinet next to you, and the cabinet was right in front of you.", + "D": "You placed it on the nearby cabinet, and the cabinet was positioned off to your right side.", + "E": "You set it on the cabinet next to you, with the cabinet located directly behind you." + }, + "answer_label": "C", + "answer_text": "You put it on the cabinet next to you, and the cabinet was right in front of you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_7_T2", + "correct_answer": "You put it on the cabinet next to you, and the cabinet was right in front of you.", + "original_options": { + "A": "On the cabinet behind you.", + "B": "On the cabinet left.", + "C": "You put it on the cabinet next to you, and the cabinet was right in front of you.", + "D": "You placed it on the nearby cabinet, and the cabinet was positioned off to your right side.", + "E": "You set it on the cabinet next to you, with the cabinet located directly behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 16 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_4_T1", + "qid_num": 29, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see?", + "answer": "You saw a girl in pink; she was in front of you.", + "question_chinese": "我看到了几个人？", + "answer_chinese": "你看到了一个穿粉色衣服的女生，她在你前面。", + "evidence": [ + { + "description": "Saw a woman in a pink top.", + "timestamps": { + "start": "00:51:40", + "end": "00:51:45" + } + } + ], + "question_timestamps": "00:51:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You didn’t see anyone.", + "B": "You saw two people.", + "C": "You saw three people, and they were standing together in front of you.", + "D": "You saw a girl in pink; she was in front of you.", + "E": "You saw one person, but she was positioned to your front-right instead of straight ahead." + }, + "answer_label": "D", + "answer_text": "You saw a girl in pink; she was in front of you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_4_T1", + "correct_answer": "You saw a girl in pink; she was in front of you.", + "original_options": { + "A": "You didn’t see anyone.", + "B": "You saw two people.", + "C": "You saw three people, and they were standing together in front of you.", + "D": "You saw a girl in pink; she was in front of you.", + "E": "You saw one person, but she was positioned to your front-right instead of straight ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 15 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_4_T2", + "qid_num": 30, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see?", + "answer": "You saw one girl; she was to your front-right.", + "question_chinese": "我看到了几个人？", + "answer_chinese": "你看到了一个女生，她在我右前方。", + "evidence": [ + { + "description": "Saw a woman in a white top with a tattoo on her arm.", + "timestamps": { + "start": "00:52:15", + "end": "00:52:19" + } + } + ], + "question_timestamps": "00:52:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw no one.", + "B": "You saw two girls.", + "C": "You saw three girls, and they were all to your front-right as you looked over.", + "D": "You saw four girls gathered to your front-right, rather than just a single person.", + "E": "You saw one girl; she was to your front-right." + }, + "answer_label": "E", + "answer_text": "You saw one girl; she was to your front-right.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_4_T2", + "correct_answer": "You saw one girl; she was to your front-right.", + "original_options": { + "A": "You saw no one.", + "B": "You saw two girls.", + "C": "You saw three girls, and they were all to your front-right as you looked over.", + "D": "You saw four girls gathered to your front-right, rather than just a single person.", + "E": "You saw one girl; she was to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_5_T1", + "qid_num": 31, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I see?", + "answer": "The trash can was outside. You were indoors, so you couldn’t see the trash can.", + "question_chinese": "我看到多少个垃圾桶？", + "answer_chinese": "垃圾桶在屋外，你在屋内，你看到不到垃圾桶。", + "evidence": [ + { + "description": "Zero. You don't see any trash can.", + "timestamps": { + "start": "00:05:32", + "end": "00:05:34" + } + } + ], + "question_timestamps": "00:05:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one trash can.", + "B": "You saw two trash cans.", + "C": "You saw one trash can outside, visible through the doorway from where you stood indoors.", + "D": "You saw three trash cans in front of you, lined up within your view as you looked ahead.", + "E": "The trash can was outside. You were indoors, so you couldn’t see the trash can." + }, + "answer_label": "E", + "answer_text": "The trash can was outside. You were indoors, so you couldn’t see the trash can.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_5_T1", + "correct_answer": "The trash can was outside. You were indoors, so you couldn’t see the trash can.", + "original_options": { + "A": "You saw one trash can.", + "B": "You saw two trash cans.", + "C": "You saw one trash can outside, visible through the doorway from where you stood indoors.", + "D": "You saw three trash cans in front of you, lined up within your view as you looked ahead.", + "E": "The trash can was outside. You were indoors, so you couldn’t see the trash can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_5_T2", + "qid_num": 32, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I see?", + "answer": "You saw a green trash can; the trash can was in front of you.", + "question_chinese": "我看到多少个垃圾桶？", + "answer_chinese": "你看到一个绿色的垃圾桶，垃圾桶在你前面。", + "evidence": [ + { + "description": "You see a green trash can", + "timestamps": { + "start": "00:14:37", + "end": "00:14:41" + } + } + ], + "question_timestamps": "00:14:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw a green trash can; the trash can was in front of you.", + "B": "You didn’t see any trash cans.", + "C": "You saw one trash can behind you.", + "D": "You saw two trash cans, with one in front of you and another positioned off to the side.", + "E": "You saw one trash can, but it was to your left instead of directly in front of you." + }, + "answer_label": "A", + "answer_text": "You saw a green trash can; the trash can was in front of you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_5_T2", + "correct_answer": "You saw a green trash can; the trash can was in front of you.", + "original_options": { + "A": "You saw a green trash can; the trash can was in front of you.", + "B": "You didn’t see any trash cans.", + "C": "You saw one trash can behind you.", + "D": "You saw two trash cans, with one in front of you and another positioned off to the side.", + "E": "You saw one trash can, but it was to your left instead of directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_15", + "qid_num": 33, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow boards are now on the bathroom wall in total?", + "answer": "one piece.", + "question_chinese": "现在浴室的墙上一共贴了几块黄色的板材？", + "answer_chinese": "1块。", + "evidence": [ + { + "description": "There is only 1 yellow board on the wall next to the bathtub.", + "timestamps": { + "start": "00:44:47", + "end": "00:45:24" + } + } + ], + "question_timestamps": "00:45:58", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two pieces.", + "B": "one piece.", + "C": "Three pieces.", + "D": "Four pieces are now on the bathroom wall in total, not just a single board.", + "E": "Five pieces are on the bathroom wall altogether after the change, exceeding one piece." + }, + "answer_label": "B", + "answer_text": "one piece.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_15", + "correct_answer": "one piece.", + "original_options": { + "A": "Two pieces.", + "B": "one piece.", + "C": "Three pieces.", + "D": "Four pieces are now on the bathroom wall in total, not just a single board.", + "E": "Five pieces are on the bathroom wall altogether after the change, exceeding one piece." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_16", + "qid_num": 34, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow boards are now on the bathroom wall in total?", + "answer": "one piece.", + "question_chinese": "现在浴室的墙上一共贴了几块黄色的板材？", + "answer_chinese": "1块。", + "evidence": [ + { + "description": "There is only 1 yellow board on the wall next to the bathtub.", + "timestamps": { + "start": "00:44:47", + "end": "00:45:24" + } + } + ], + "question_timestamps": "00:46:46", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "No pieces.", + "B": "Two pieces.", + "C": "Three pieces are now on the bathroom wall in total, rather than a single piece.", + "D": "one piece.", + "E": "Four pieces are on the bathroom wall altogether, not one piece as stated." + }, + "answer_label": "D", + "answer_text": "one piece.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_16", + "correct_answer": "one piece.", + "original_options": { + "A": "No pieces.", + "B": "Two pieces.", + "C": "Three pieces are now on the bathroom wall in total, rather than a single piece.", + "D": "one piece.", + "E": "Four pieces are on the bathroom wall altogether, not one piece as stated." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 13 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_13", + "qid_num": 35, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is that coil of white water pipe from me now?", + "answer": "The white water pipe is about seven meters away from you.", + "question_chinese": "那圈白色的水管现在离我多远？", + "answer_chinese": "白色的水管离你大约7米远。", + "evidence": [ + { + "description": "The white water pipe is about 7 meters away from the bathtub.", + "timestamps": { + "start": "00:34:14", + "end": "00:34:34" + } + } + ], + "question_timestamps": "00:34:37", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "The white water pipe is about seven meters away from you.", + "B": "About five meters away.", + "C": "About ten meters away.", + "D": "The white water pipe is about three meters away from you right now, not much farther.", + "E": "The white water pipe is about twelve meters away from you now, farther down the area." + }, + "answer_label": "A", + "answer_text": "The white water pipe is about seven meters away from you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_13", + "correct_answer": "The white water pipe is about seven meters away from you.", + "original_options": { + "A": "The white water pipe is about seven meters away from you.", + "B": "About five meters away.", + "C": "About ten meters away.", + "D": "The white water pipe is about three meters away from you right now, not much farther.", + "E": "The white water pipe is about twelve meters away from you now, farther down the area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60", + "video_length": "01:02:43", + "qid": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_14", + "qid_num": 36, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is that coil of white water pipe from me now?", + "answer": "The white water pipe is about seven meters away from you.", + "question_chinese": "那圈白色的水管现在离我多远？", + "answer_chinese": "白色的水管离你大约7米远。", + "evidence": [ + { + "description": "The white water pipe is about 7 meters away from the bathtub.", + "timestamps": { + "start": "00:34:14", + "end": "00:34:34" + } + } + ], + "question_timestamps": "00:39:17", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "The white water pipe is about seven meters away from you.", + "B": "About four meters away.", + "C": "About one meter away.", + "D": "The white water pipe is about ten meters away from you now, a bit farther than before.", + "E": "The white water pipe is about three meters away from you now, close enough to reach quickly." + }, + "answer_label": "A", + "answer_text": "The white water pipe is about seven meters away from you.", + "q_id": "4f8112b1-0d67-44e3-aa48-117a13ffbc60_14", + "correct_answer": "The white water pipe is about seven meters away from you.", + "original_options": { + "A": "The white water pipe is about seven meters away from you.", + "B": "About four meters away.", + "C": "About one meter away.", + "D": "The white water pipe is about ten meters away from you now, a bit farther than before.", + "E": "The white water pipe is about three meters away from you now, close enough to reach quickly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + } +] diff --git a/QAs/51eb2565-5911-45c0-9472-23c5b3400840-mcq.json b/QAs/51eb2565-5911-45c0-9472-23c5b3400840-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..638835290b8cbf6d1fb3e498a28249c580a22eb9 --- /dev/null +++ b/QAs/51eb2565-5911-45c0-9472-23c5b3400840-mcq.json @@ -0,0 +1,1174 @@ +[ + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_5_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Since I started cooking in the kitchen, how many pieces of cheese have I grated?", + "answer": "One piece, on the counter to your right-hand side.", + "question_chinese": "从我在厨房开始做饭开始，我将多少块芝士擦丝了？", + "answer_chinese": "一块，在你的右手边的操作台上", + "evidence": [ + { + "description": "One piece of cheese was grated, on the counter to your right-hand side.", + "timestamps": { + "start": "00:46:23", + "end": "00:46:24" + } + } + ], + "question_timestamps": "00:47:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero pieces.", + "B": "Two pieces.", + "C": "One piece, on the counter to your right-hand side.", + "D": "Three pieces, with them sitting on the counter to your right-hand side.", + "E": "Four pieces, and they are still on the counter to your right-hand side." + }, + "answer_label": "C", + "answer_text": "One piece, on the counter to your right-hand side.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_5_T1", + "correct_answer": "One piece, on the counter to your right-hand side.", + "original_options": { + "A": "Zero pieces.", + "B": "Two pieces.", + "C": "One piece, on the counter to your right-hand side.", + "D": "Three pieces, with them sitting on the counter to your right-hand side.", + "E": "Four pieces, and they are still on the counter to your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_5_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Since I started cooking in the kitchen, how many pieces of cheese have I grated?", + "answer": "Two pieces, on the counter to your right-hand side.", + "question_chinese": "从我在厨房开始做饭开始，我将多少块芝士擦丝了？", + "answer_chinese": "两块，在你的右手边的操作台上", + "evidence": [ + { + "description": "One piece of cheese was grated, on the counter to your right-hand side.", + "timestamps": { + "start": "00:46:23", + "end": "00:46:24" + } + }, + { + "description": "Another piece of cheese was grated, on the counter to your right-hand side.", + "timestamps": { + "start": "00:48:52", + "end": "00:48:53" + } + } + ], + "question_timestamps": "00:48:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One piece.", + "B": "Four pieces.", + "C": "Zero pieces, and there is nothing grated on the counter to your right-hand side.", + "D": "Two pieces, on the counter to your right-hand side.", + "E": "Three pieces, with the grated pieces on the counter to your right-hand side." + }, + "answer_label": "D", + "answer_text": "Two pieces, on the counter to your right-hand side.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_5_T2", + "correct_answer": "Two pieces, on the counter to your right-hand side.", + "original_options": { + "A": "One piece.", + "B": "Four pieces.", + "C": "Zero pieces, and there is nothing grated on the counter to your right-hand side.", + "D": "Two pieces, on the counter to your right-hand side.", + "E": "Three pieces, with the grated pieces on the counter to your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_2_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was cooking in the kitchen, how many bottles of seasonings were on the counter?", + "answer": "Seven bottles, on the counter to your front-right.", + "question_chinese": "当我在厨房做饭时，操作台上有几瓶调味品。", + "answer_chinese": "七瓶，在你右前方的操作台上", + "evidence": [ + { + "description": "There were seven bottles of seasonings on the counter.", + "timestamps": { + "start": "00:43:37", + "end": "00:43:39" + } + } + ], + "question_timestamps": "00:44:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five bottles.", + "B": "Nine bottles.", + "C": "Six bottles, arranged on the counter to your front-right while you were cooking.", + "D": "Eight bottles, sitting together on the counter to your front-right during cooking.", + "E": "Seven bottles, on the counter to your front-right." + }, + "answer_label": "E", + "answer_text": "Seven bottles, on the counter to your front-right.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_2_T1", + "correct_answer": "Seven bottles, on the counter to your front-right.", + "original_options": { + "A": "Five bottles.", + "B": "Nine bottles.", + "C": "Six bottles, arranged on the counter to your front-right while you were cooking.", + "D": "Eight bottles, sitting together on the counter to your front-right during cooking.", + "E": "Seven bottles, on the counter to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_2_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was cooking in the kitchen, how many bottles of seasonings were on the counter?", + "answer": "Three bottles, on the counter on the left side of you.", + "question_chinese": "当我在厨房做饭时，操作台上有几瓶调味品。", + "answer_chinese": "三瓶，在你左侧的操作台上。", + "evidence": [ + { + "description": "There were seven bottles of seasonings on the counter.", + "timestamps": { + "start": "00:43:37", + "end": "00:43:39" + } + }, + { + "description": "You put four bottles back into the cabinet.", + "timestamps": { + "start": "00:44:30", + "end": "00:44:46" + } + } + ], + "question_timestamps": "00:44:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three bottles, on the counter on the left side of you.", + "B": "One bottle.", + "C": "Five bottles.", + "D": "Two bottles, placed on the counter on the left side of you while you were cooking.", + "E": "Four bottles, lined up on the counter on the left side of you during cooking." + }, + "answer_label": "A", + "answer_text": "Three bottles, on the counter on the left side of you.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_2_T2", + "correct_answer": "Three bottles, on the counter on the left side of you.", + "original_options": { + "A": "Three bottles, on the counter on the left side of you.", + "B": "One bottle.", + "C": "Five bottles.", + "D": "Two bottles, placed on the counter on the left side of you while you were cooking.", + "E": "Four bottles, lined up on the counter on the left side of you during cooking." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 15 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_7_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many different kinds of vegetables did I chop?", + "answer": "Two kinds: carrot and onion, on the plate on the left side of you.", + "question_chinese": "我切了多少种蔬菜？", + "answer_chinese": "两种，胡萝卜和洋葱在你左侧的盘子上", + "evidence": [ + { + "description": "You chopped carrots, on the plate on the left side of you.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:35" + } + }, + { + "description": "You chopped onions, on the plate on the left side of you.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One kind.", + "B": "Three kinds.", + "C": "Four kinds: carrot, onion, tomato, and pepper, on the plate on the left side of you.", + "D": "One kind: onion only, with the chopped pieces on the plate on the left side of you.", + "E": "Two kinds: carrot and onion, on the plate on the left side of you." + }, + "answer_label": "E", + "answer_text": "Two kinds: carrot and onion, on the plate on the left side of you.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_7_T1", + "correct_answer": "Two kinds: carrot and onion, on the plate on the left side of you.", + "original_options": { + "A": "One kind.", + "B": "Three kinds.", + "C": "Four kinds: carrot, onion, tomato, and pepper, on the plate on the left side of you.", + "D": "One kind: onion only, with the chopped pieces on the plate on the left side of you.", + "E": "Two kinds: carrot and onion, on the plate on the left side of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_17", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kinds of vegetables did I cut?", + "answer": "Three.", + "question_chinese": "我切了多少种蔬菜？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:34", + "end": "00:00:35" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:09:46", + "end": "00:09:47" + } + } + ], + "question_timestamps": "00:09:48", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Two kinds of vegetables, cut into pieces and set aside together on the counter while cooking.", + "D": "Three.", + "E": "Four kinds of vegetables, all cut up and gathered together on the counter during the cooking." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_17", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Two kinds of vegetables, cut into pieces and set aside together on the counter while cooking.", + "D": "Three.", + "E": "Four kinds of vegetables, all cut up and gathered together on the counter during the cooking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_7_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many different kinds of vegetables did I chop?", + "answer": "Three kinds: carrot and onion on the plate on the left side of you, and tomato on the cutting board in front of you.", + "question_chinese": "我切了多少种蔬菜？", + "answer_chinese": "三种，胡萝卜、洋葱在你左侧盘子上，西红柿在你面前的砧板上", + "evidence": [ + { + "description": "You chopped carrots, on the plate on the left side of you.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:35" + } + }, + { + "description": "You chopped onions, on the plate on the left side of you.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + }, + { + "description": "You chopped tomatoes, on the cutting board in front of you.", + "timestamps": { + "start": "00:09:46", + "end": "00:09:47" + } + } + ], + "question_timestamps": "00:09:55", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three kinds: carrot and onion on the plate on the left side of you, and tomato on the cutting board in front of you.", + "B": "Two kinds.", + "C": "Four kinds.", + "D": "Two kinds: carrot on the plate left of you, and tomato on the cutting board in front of you.", + "E": "Four kinds: carrot and onion on the left plate, plus tomato and cucumber on the cutting board in front." + }, + "answer_label": "A", + "answer_text": "Three kinds: carrot and onion on the plate on the left side of you, and tomato on the cutting board in front of you.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_7_T2", + "correct_answer": "Three kinds: carrot and onion on the plate on the left side of you, and tomato on the cutting board in front of you.", + "original_options": { + "A": "Three kinds: carrot and onion on the plate on the left side of you, and tomato on the cutting board in front of you.", + "B": "Two kinds.", + "C": "Four kinds.", + "D": "Two kinds: carrot on the plate left of you, and tomato on the cutting board in front of you.", + "E": "Four kinds: carrot and onion on the left plate, plus tomato and cucumber on the cutting board in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 19, + 19 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_13", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of cheese did I shred?", + "answer": "Two.", + "question_chinese": "我将多少块芝士擦丝了？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:46:23", + "end": "00:46:24" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:48:52", + "end": "00:48:43" + } + } + ], + "question_timestamps": "00:48:44", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four pieces of cheese were shredded in total during the cooking process, not just one or two.", + "D": "Five pieces of cheese were shredded altogether, making a larger pile than the amount actually shown.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_13", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four pieces of cheese were shredded in total during the cooking process, not just one or two.", + "D": "Five pieces of cheese were shredded altogether, making a larger pile than the amount actually shown.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 16 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_14", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of cheese did I shred?", + "answer": "two.", + "question_chinese": "我将多少块芝士擦丝了？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:46:23", + "end": "00:46:24" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:48:52", + "end": "00:48:53" + } + } + ], + "question_timestamps": "00:48:54", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One piece.", + "B": "Three pieces.", + "C": "two.", + "D": "Four pieces of cheese were shredded in total during the cooking process, not just two pieces.", + "E": "Five pieces of cheese were shredded altogether, making a larger pile than what is actually shown." + }, + "answer_label": "C", + "answer_text": "two.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_14", + "correct_answer": "two.", + "original_options": { + "A": "One piece.", + "B": "Three pieces.", + "C": "two.", + "D": "Four pieces of cheese were shredded in total during the cooking process, not just two pieces.", + "E": "Five pieces of cheese were shredded altogether, making a larger pile than what is actually shown." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_7", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I put the seasoning back?", + "answer": "In the cabinet.", + "question_chinese": "我要把调料放回哪里？", + "answer_chinese": "橱柜里", + "evidence": [ + { + "description": "In the cabinet", + "timestamps": { + "start": "00:44:38", + "end": "00:44:39" + } + } + ], + "question_timestamps": "00:44:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the cabinet.", + "B": "In the spice drawer.", + "C": "Inside the pantry.", + "D": "Put it back on the kitchen shelf where you grabbed it earlier.", + "E": "Set it on the countertop next to the stove so it stays within reach." + }, + "answer_label": "A", + "answer_text": "In the cabinet.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_7", + "correct_answer": "In the cabinet.", + "original_options": { + "A": "In the cabinet.", + "B": "In the spice drawer.", + "C": "Inside the pantry.", + "D": "Put it back on the kitchen shelf where you grabbed it earlier.", + "E": "Set it on the countertop next to the stove so it stays within reach." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 14 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_8", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I put the seasoning back?", + "answer": "In the cabinet.", + "question_chinese": "我要把调料放回哪里？", + "answer_chinese": "橱柜里", + "evidence": [ + { + "description": "In the cabinet", + "timestamps": { + "start": "00:44:38", + "end": "00:44:39" + } + } + ], + "question_timestamps": "00:44:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the spice rack.", + "B": "Inside the pantry.", + "C": "In the cabinet.", + "D": "Leave it on the kitchen island instead of putting it away in the cabinet.", + "E": "Put it on the shelf next to the stove so it’s easy to grab later." + }, + "answer_label": "C", + "answer_text": "In the cabinet.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_8", + "correct_answer": "In the cabinet.", + "original_options": { + "A": "On the spice rack.", + "B": "Inside the pantry.", + "C": "In the cabinet.", + "D": "Leave it on the kitchen island instead of putting it away in the cabinet.", + "E": "Put it on the shelf next to the stove so it’s easy to grab later." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 16 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_9", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power outlets are there?", + "answer": "four.", + "question_chinese": "有几个插座？", + "answer_chinese": "4", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "another three outlets", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Six power outlets are visible on the wall, which is more than the correct count of four.", + "D": "four.", + "E": "Seven power outlets are shown in total, including the ones grouped together on the same strip." + }, + "answer_label": "D", + "answer_text": "four.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_9", + "correct_answer": "four.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Six power outlets are visible on the wall, which is more than the correct count of four.", + "D": "four.", + "E": "Seven power outlets are shown in total, including the ones grouped together on the same strip." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 16 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_10", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power outlets are there?", + "answer": "five.", + "question_chinese": "有几个插座？", + "answer_chinese": "5", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "another three outlets", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "the fifth one", + "timestamps": { + "start": "00:01:15", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:01:17", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "five.", + "C": "Four.", + "D": "Six power outlets are visible on the wall, which is more than the correct count of five.", + "E": "Eight power outlets are shown in total, counting every socket in the frame, not just one side." + }, + "answer_label": "B", + "answer_text": "five.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_10", + "correct_answer": "five.", + "original_options": { + "A": "Three.", + "B": "five.", + "C": "Four.", + "D": "Six power outlets are visible on the wall, which is more than the correct count of five.", + "E": "Eight power outlets are shown in total, counting every socket in the frame, not just one side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 17 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_5", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When cooking, where should I hold my hand?", + "answer": "Hold the pot with your left hand and hold the spatula with your right hand.", + "question_chinese": "炒菜时，我的手应该握哪里？", + "answer_chinese": "左手扶锅，右手拿锅铲", + "evidence": [ + { + "description": "Hold the pot with your left hand and hold the spatula with your right hand.", + "timestamps": { + "start": "00:30:03", + "end": "00:30:04" + } + } + ], + "question_timestamps": "00:30:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Hold the pot with your left hand and hold the spatula with your right hand.", + "B": "Hold the pot with your right hand.", + "C": "Hold the spatula with your left hand.", + "D": "Hold the pot with your right hand and hold the spatula with your left hand.", + "E": "Hold the pot with your left hand and hold the spatula with both hands." + }, + "answer_label": "A", + "answer_text": "Hold the pot with your left hand and hold the spatula with your right hand.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_5", + "correct_answer": "Hold the pot with your left hand and hold the spatula with your right hand.", + "original_options": { + "A": "Hold the pot with your left hand and hold the spatula with your right hand.", + "B": "Hold the pot with your right hand.", + "C": "Hold the spatula with your left hand.", + "D": "Hold the pot with your right hand and hold the spatula with your left hand.", + "E": "Hold the pot with your left hand and hold the spatula with both hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 14 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_6", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When cooking, where should I hold my hand?", + "answer": "Hold the pot with your left hand and hold the spatula with your right hand.", + "question_chinese": "炒菜时，我的手应该握哪里？", + "answer_chinese": "左手扶锅，右手拿锅铲", + "evidence": [ + { + "description": "Hold the pot with your left hand and hold the spatula with your right hand.", + "timestamps": { + "start": "00:30:03", + "end": "00:30:04" + } + } + ], + "question_timestamps": "00:30:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Hold the pot with your right hand.", + "B": "Hold the spatula with your left hand.", + "C": "Hold the pot with your right hand and use your left hand to stir with the spatula.", + "D": "Grip the pot with both hands for stability, and keep the spatula resting in the pan.", + "E": "Hold the pot with your left hand and hold the spatula with your right hand." + }, + "answer_label": "E", + "answer_text": "Hold the pot with your left hand and hold the spatula with your right hand.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_6", + "correct_answer": "Hold the pot with your left hand and hold the spatula with your right hand.", + "original_options": { + "A": "Hold the pot with your right hand.", + "B": "Hold the spatula with your left hand.", + "C": "Hold the pot with your right hand and use your left hand to stir with the spatula.", + "D": "Grip the pot with both hands for stability, and keep the spatula resting in the pan.", + "E": "Hold the pot with your left hand and hold the spatula with your right hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_6_T1", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I finished stir-frying the dish, what did I do?", + "answer": "You took seasonings from the upper cabinet at the stove’s back-right, and used them to season the food.", + "question_chinese": "翻炒完菜后我干了什么？", + "answer_chinese": "从灶台右后侧的上侧橱柜拿出调味品进行调味", + "evidence": [ + { + "description": "You took seasonings from the upper cabinet at the stove’s back-right, and used them to season the food.", + "timestamps": { + "start": "00:27:49", + "end": "00:27:50" + } + } + ], + "question_timestamps": "00:32:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You grabbed seasonings from the back-left cabinet.", + "B": "You took seasonings from the front-right cabinet.", + "C": "You took seasonings from the upper cabinet at the stove’s back-right, and used them to season the food.", + "D": "You took seasonings from the upper cabinet at the stove’s back-left and sprinkled them over the food.", + "E": "You grabbed the seasonings from the upper cabinet near the stove’s front-right and then continued cooking without adding them." + }, + "answer_label": "C", + "answer_text": "You took seasonings from the upper cabinet at the stove’s back-right, and used them to season the food.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_6_T1", + "correct_answer": "You took seasonings from the upper cabinet at the stove’s back-right, and used them to season the food.", + "original_options": { + "A": "You grabbed seasonings from the back-left cabinet.", + "B": "You took seasonings from the front-right cabinet.", + "C": "You took seasonings from the upper cabinet at the stove’s back-right, and used them to season the food.", + "D": "You took seasonings from the upper cabinet at the stove’s back-left and sprinkled them over the food.", + "E": "You grabbed the seasonings from the upper cabinet near the stove’s front-right and then continued cooking without adding them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 20 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_15", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after stir-frying the vegetables?", + "answer": "Season and cover the pot.", + "question_chinese": "翻炒完菜后我干了什么？", + "answer_chinese": "调味和盖锅盖", + "evidence": [ + { + "description": "Season", + "timestamps": { + "start": "00:27:49", + "end": "00:27:50" + } + }, + { + "description": "cover the pot", + "timestamps": { + "start": "00:43:21", + "end": "00:43:22" + } + } + ], + "question_timestamps": "00:43:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Season and cover the pot.", + "B": "Add sauce and simmer.", + "C": "Keep stir-frying longer.", + "D": "You sprinkled herbs over the vegetables and left the pan uncovered while they finished cooking.", + "E": "You added a little broth to the vegetables, then partially covered the pot to let them simmer." + }, + "answer_label": "A", + "answer_text": "Season and cover the pot.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_15", + "correct_answer": "Season and cover the pot.", + "original_options": { + "A": "Season and cover the pot.", + "B": "Add sauce and simmer.", + "C": "Keep stir-frying longer.", + "D": "You sprinkled herbs over the vegetables and left the pan uncovered while they finished cooking.", + "E": "You added a little broth to the vegetables, then partially covered the pot to let them simmer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 17 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_6_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I finished stir-frying the dish, what did I do?", + "answer": "You took seasonings from the upper cabinet at the stove’s back-right to season the food, and you also took a pot lid from the lower cabinet at the stove’s back-right and put it on.", + "question_chinese": "翻炒完菜后我干了什么？", + "answer_chinese": "从灶台右后侧的上侧橱柜拿出调味品进行调味，并且从灶台右后侧下部橱柜中拿出锅盖并盖上", + "evidence": [ + { + "description": "You took seasonings from the upper cabinet at the stove’s back-right, and used them to season the food.", + "timestamps": { + "start": "00:27:49", + "end": "00:27:50" + } + }, + { + "description": "You took a pot lid from the lower cabinet at the stove’s back-right and put it on.", + "timestamps": { + "start": "00:43:21", + "end": "00:43:22" + } + } + ], + "question_timestamps": "00:43:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You only seasoned the food.", + "B": "You took seasonings from the upper cabinet at the stove’s back-right to season the food, and you also took a pot lid from the lower cabinet at the stove’s back-right and put it on.", + "C": "You took seasonings from the back-left cabinet.", + "D": "You took seasonings from the upper cabinet at the stove’s front-right to season the food, then put a lid on.", + "E": "You grabbed seasonings from the upper cabinet at the stove’s back-left and covered the pot with a lid from below." + }, + "answer_label": "B", + "answer_text": "You took seasonings from the upper cabinet at the stove’s back-right to season the food, and you also took a pot lid from the lower cabinet at the stove’s back-right and put it on.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_6_T2", + "correct_answer": "You took seasonings from the upper cabinet at the stove’s back-right to season the food, and you also took a pot lid from the lower cabinet at the stove’s back-right and put it on.", + "original_options": { + "A": "You only seasoned the food.", + "B": "You took seasonings from the upper cabinet at the stove’s back-right to season the food, and you also took a pot lid from the lower cabinet at the stove’s back-right and put it on.", + "C": "You took seasonings from the back-left cabinet.", + "D": "You took seasonings from the upper cabinet at the stove’s front-right to season the food, then put a lid on.", + "E": "You grabbed seasonings from the upper cabinet at the stove’s back-left and covered the pot with a lid from below." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 21, + 21 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_1_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the pot relative to me right now?", + "answer": "To your front-left.", + "question_chinese": "锅目前在我的什么方位？", + "answer_chinese": "在我的左前侧", + "evidence": [ + { + "description": "You walked from the counter to the sink on the right side to wash your hands.", + "timestamps": { + "start": "00:12:20", + "end": "00:12:24" + } + }, + { + "description": "You picked up the olive oil from the cabinet at your front-left.", + "timestamps": { + "start": "00:12:24", + "end": "00:12:28" + } + } + ], + "question_timestamps": "00:12:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "To your front-left.", + "D": "Directly in front of you, slightly to the right of your body.", + "E": "Behind you on the left side, near where you were standing a moment ago." + }, + "answer_label": "C", + "answer_text": "To your front-left.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_1_T2", + "correct_answer": "To your front-left.", + "original_options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "To your front-left.", + "D": "Directly in front of you, slightly to the right of your body.", + "E": "Behind you on the left side, near where you were standing a moment ago." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 14 + ] + }, + { + "video_uid": "51eb2565-5911-45c0-9472-23c5b3400840", + "video_length": "00:54:31", + "qid": "51eb2565-5911-45c0-9472-23c5b3400840_1_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the pot relative to me right now?", + "answer": "On your left-hand side.", + "question_chinese": "锅目前在我的什么方位？", + "answer_chinese": "在我的手边", + "evidence": [ + { + "description": "The pot appeared in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "I turned right and then faced the counter.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:30:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your right-hand side.", + "B": "On your left-hand side.", + "C": "In front of you.", + "D": "Behind you and slightly to the right, not on your left-hand side.", + "E": "Off to your front-right, closer to your right hand than your left." + }, + "answer_label": "B", + "answer_text": "On your left-hand side.", + "q_id": "51eb2565-5911-45c0-9472-23c5b3400840_1_T1", + "correct_answer": "On your left-hand side.", + "original_options": { + "A": "On your right-hand side.", + "B": "On your left-hand side.", + "C": "In front of you.", + "D": "Behind you and slightly to the right, not on your left-hand side.", + "E": "Off to your front-right, closer to your right hand than your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + } +] diff --git a/QAs/546e4806-a7f5-46bd-bf65-7737db2c5594-001-mcq.json b/QAs/546e4806-a7f5-46bd-bf65-7737db2c5594-001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4d6fcdf78b240d01db0de77fff6cd4607ce1e2b1 --- /dev/null +++ b/QAs/546e4806-a7f5-46bd-bf65-7737db2c5594-001-mcq.json @@ -0,0 +1,951 @@ +[ + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_11", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position after going up the steps?", + "answer": "Walk forward 50 meters.", + "question_chinese": "上了台阶后，我是怎么走到当前位置的？", + "answer_chinese": "向前走五十米", + "evidence": [ + { + "description": "Walk forward 50 meters", + "timestamps": { + "start": "00:12:00", + "end": "00:12:01" + } + } + ], + "question_timestamps": "00:12:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk backward 50 meters.", + "B": "Walk forward 50 meters.", + "C": "Walk forward 25 meters.", + "D": "Walk forward about 50 meters, then turn right to reach your current position.", + "E": "Walk left for roughly 50 meters after the steps instead of continuing straight ahead." + }, + "answer_label": "B", + "answer_text": "Walk forward 50 meters.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_11", + "correct_answer": "Walk forward 50 meters.", + "original_options": { + "A": "Walk backward 50 meters.", + "B": "Walk forward 50 meters.", + "C": "Walk forward 25 meters.", + "D": "Walk forward about 50 meters, then turn right to reach your current position.", + "E": "Walk left for roughly 50 meters after the steps instead of continuing straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 2, + "C": 1, + "D": 2, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_12", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position after going up the steps?", + "answer": "Walk forward 50 meters and take the elevator on the right.", + "question_chinese": "上了台阶后，我是怎么走到当前位置的？", + "answer_chinese": "向前走五十米，坐右手边的电梯", + "evidence": [ + { + "description": "Walk forward 50 meters", + "timestamps": { + "start": "00:12:00", + "end": "00:12:01" + } + }, + { + "description": "take the elevator on the right.", + "timestamps": { + "start": "00:12:52", + "end": "00:12:53" + } + } + ], + "question_timestamps": "00:20:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Take the elevator on the left.", + "B": "Walk forward 25 meters, then elevator.", + "C": "Walk forward 50 meters and take the elevator on the left rather than the one on the right.", + "D": "Walk forward 50 meters, then take the stairs on the right instead of taking the elevator.", + "E": "Walk forward 50 meters and take the elevator on the right." + }, + "answer_label": "E", + "answer_text": "Walk forward 50 meters and take the elevator on the right.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_12", + "correct_answer": "Walk forward 50 meters and take the elevator on the right.", + "original_options": { + "A": "Take the elevator on the left.", + "B": "Walk forward 25 meters, then elevator.", + "C": "Walk forward 50 meters and take the elevator on the left rather than the one on the right.", + "D": "Walk forward 50 meters, then take the stairs on the right instead of taking the elevator.", + "E": "Walk forward 50 meters and take the elevator on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 2, + "C": 1, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 16 + ] + }, + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_3", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the red pavilion at the bus stop?", + "answer": "It’s in front of you.", + "question_chinese": "巴士站有红色的亭子在哪个方向？", + "answer_chinese": "在我的前面", + "evidence": [ + { + "description": "it’s in front of me.", + "timestamps": { + "start": "00:02:38", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:02:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s behind you.", + "B": "It’s in front of you.", + "C": "It’s to your left.", + "D": "It’s off to your right as you wait, not directly ahead at the stop.", + "E": "It’s behind you near the curb, so you’d need to turn around to see it." + }, + "answer_label": "B", + "answer_text": "It’s in front of you.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_3", + "correct_answer": "It’s in front of you.", + "original_options": { + "A": "It’s behind you.", + "B": "It’s in front of you.", + "C": "It’s to your left.", + "D": "It’s off to your right as you wait, not directly ahead at the stop.", + "E": "It’s behind you near the curb, so you’d need to turn around to see it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 17 + ] + }, + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_4", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the red pavilion at the bus stop?", + "answer": "It’s on your left.", + "question_chinese": "巴士站有红色的亭子在哪个方向？", + "answer_chinese": "在我的左边", + "evidence": [ + { + "description": "it’s on my left.", + "timestamps": { + "start": "00:02:58", + "end": "00:02:59" + } + } + ], + "question_timestamps": "00:04:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s on your right.", + "B": "It’s behind you.", + "C": "It’s straight ahead of you at the bus stop, not off to your left side.", + "D": "It’s behind you near the stop, so it wouldn’t be on your left when facing forward.", + "E": "It’s on your left." + }, + "answer_label": "E", + "answer_text": "It’s on your left.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_4", + "correct_answer": "It’s on your left.", + "original_options": { + "A": "It’s on your right.", + "B": "It’s behind you.", + "C": "It’s straight ahead of you at the bus stop, not off to your left side.", + "D": "It’s behind you near the stop, so it wouldn’t be on your left when facing forward.", + "E": "It’s on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 18 + ] + }, + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_3_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I?", + "answer": "On the bus.", + "question_chinese": "我在哪？", + "answer_chinese": "在巴士上", + "evidence": [ + { + "description": "on the bus", + "timestamps": { + "start": "00:10:30", + "end": "00:10:31" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the bus.", + "B": "Inside the office building.", + "C": "Outside the office building.", + "D": "You’re standing outside by the entrance, not riding inside the bus at all.", + "E": "You’re on a train or tram, not on the bus right now." + }, + "answer_label": "A", + "answer_text": "On the bus.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_3_T1", + "correct_answer": "On the bus.", + "original_options": { + "A": "On the bus.", + "B": "Inside the office building.", + "C": "Outside the office building.", + "D": "You’re standing outside by the entrance, not riding inside the bus at all.", + "E": "You’re on a train or tram, not on the bus right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 13 + ] + }, + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_3_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I?", + "answer": "Inside the office building.", + "question_chinese": "我在哪？", + "answer_chinese": "写字楼里", + "evidence": [ + { + "description": "In the office building", + "timestamps": { + "start": "00:12:15", + "end": "00:12:16" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Inside the office building.", + "B": "Outside the office building.", + "C": "On the bus.", + "D": "You’re outside near the building entrance, rather than being indoors in the office.", + "E": "You’re riding on the bus, so you aren’t inside the office building at this moment." + }, + "answer_label": "A", + "answer_text": "Inside the office building.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_3_T2", + "correct_answer": "Inside the office building.", + "original_options": { + "A": "Inside the office building.", + "B": "Outside the office building.", + "C": "On the bus.", + "D": "You’re outside near the building entrance, rather than being indoors in the office.", + "E": "You’re riding on the bus, so you aren’t inside the office building at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 17 + ] + }, + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get from indoors to my current position?", + "answer": "Push the door open, walk forward twenty meters, turn left and walk ten meters.", + "question_chinese": "我是怎么从室内走到当前位置的？", + "answer_chinese": "推门向前走20米，左转走10米", + "evidence": [ + { + "description": "Push the door open, walk forward 20 meters", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "turn left and walk 10 meters", + "timestamps": { + "start": "00:00:26", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Open the door, walk ten meters left.", + "B": "Open the door, turn right and walk ten.", + "C": "Push the door open, walk forward fifteen meters, turn left, then walk ten meters to reach the spot.", + "D": "Push the door open, walk forward twenty meters, turn left and walk ten meters.", + "E": "Push the door open, walk forward twenty meters, turn right, and continue walking ten meters instead of turning left." + }, + "answer_label": "D", + "answer_text": "Push the door open, walk forward twenty meters, turn left and walk ten meters.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_1", + "correct_answer": "Push the door open, walk forward twenty meters, turn left and walk ten meters.", + "original_options": { + "A": "Open the door, walk ten meters left.", + "B": "Open the door, turn right and walk ten.", + "C": "Push the door open, walk forward fifteen meters, turn left, then walk ten meters to reach the spot.", + "D": "Push the door open, walk forward twenty meters, turn left and walk ten meters.", + "E": "Push the door open, walk forward twenty meters, turn right, and continue walking ten meters instead of turning left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 0, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 19 + ] + }, + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get from indoors to my current position?", + "answer": "Push the door open, walk forward twenty meters, turn left and walk ten meters, then turn right and walk ten meters.", + "question_chinese": "我是怎么从室内走到当前位置的？", + "answer_chinese": "推门向前走20米，左转走10米，右转走10米", + "evidence": [ + { + "description": "Push the door open, walk forward 20 meters", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "turn left and walk 10 meters", + "timestamps": { + "start": "00:00:26", + "end": "00:00:27" + } + }, + { + "description": "turn right and walk 10 meters.", + "timestamps": { + "start": "00:00:52", + "end": "00:00:53" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Walk forward twenty, then turn right.", + "B": "Walk backward twenty, then turn left.", + "C": "Push the door open, walk forward twenty meters, turn left and walk ten meters, then turn right and walk ten meters.", + "D": "Push the door open, walk forward twenty meters, turn left ten meters, then turn right and walk five meters.", + "E": "Push the door open, walk forward twenty-five meters, turn left five meters, then turn right and walk fifteen meters." + }, + "answer_label": "C", + "answer_text": "Push the door open, walk forward twenty meters, turn left and walk ten meters, then turn right and walk ten meters.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_2", + "correct_answer": "Push the door open, walk forward twenty meters, turn left and walk ten meters, then turn right and walk ten meters.", + "original_options": { + "A": "Walk forward twenty, then turn right.", + "B": "Walk backward twenty, then turn left.", + "C": "Push the door open, walk forward twenty meters, turn left and walk ten meters, then turn right and walk ten meters.", + "D": "Push the door open, walk forward twenty meters, turn left ten meters, then turn right and walk five meters.", + "E": "Push the door open, walk forward twenty-five meters, turn left five meters, then turn right and walk fifteen meters." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_15", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position after the elevator door opened?", + "answer": "After getting out of the elevator, walk forward ten meters, walk left three meters.", + "question_chinese": "电梯开门后，我怎么走到当前位置的？", + "answer_chinese": "出电梯后向前走10米，向左走3米", + "evidence": [ + { + "description": "After getting out of the elevator, walk forward 10 meters", + "timestamps": { + "start": "00:13:23", + "end": "00:13:24" + } + }, + { + "description": "walk left 3 meters", + "timestamps": { + "start": "00:13:26", + "end": "00:13:27" + } + } + ], + "question_timestamps": "00:13:35", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "After getting out of the elevator, walk forward ten meters, walk left three meters.", + "B": "Walk forward ten, then right three.", + "C": "Walk forward eight, then left three.", + "D": "After getting out of the elevator, walk forward ten meters, then walk left five meters instead of three.", + "E": "After exiting the elevator, walk backward ten meters and then walk right three meters to end up there." + }, + "answer_label": "A", + "answer_text": "After getting out of the elevator, walk forward ten meters, walk left three meters.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_15", + "correct_answer": "After getting out of the elevator, walk forward ten meters, walk left three meters.", + "original_options": { + "A": "After getting out of the elevator, walk forward ten meters, walk left three meters.", + "B": "Walk forward ten, then right three.", + "C": "Walk forward eight, then left three.", + "D": "After getting out of the elevator, walk forward ten meters, then walk left five meters instead of three.", + "E": "After exiting the elevator, walk backward ten meters and then walk right three meters to end up there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 1, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_16", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position after the elevator door opened?", + "answer": "After getting out of the elevator, walk forward ten meters, walk left three meters, then walk forward 25 meters and turn left.", + "question_chinese": "电梯开门后，我怎么走到当前位置的？", + "answer_chinese": "出电梯后向前走10米，向左走3米，向前走25米后左转", + "evidence": [ + { + "description": "After getting out of the elevator, walk forward 10 meters", + "timestamps": { + "start": "00:13:23", + "end": "00:13:24" + } + }, + { + "description": "walk left 3 meters", + "timestamps": { + "start": "00:13:26", + "end": "00:13:27" + } + }, + { + "description": "walk forward 25 meters and turn left", + "timestamps": { + "start": "00:13:38", + "end": "00:13:39" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Then walk forward twenty and turn right.", + "B": "Then walk backward twenty-five and turn left.", + "C": "After getting out of the elevator, walk forward ten meters, walk left three meters, then walk forward 25 meters and turn left.", + "D": "After getting out, walk forward ten meters, left three meters, then forward twenty meters and turn right at the end.", + "E": "After getting out, walk forward fifteen meters, left five meters, then forward twenty meters and turn right instead of turning left." + }, + "answer_label": "C", + "answer_text": "After getting out of the elevator, walk forward ten meters, walk left three meters, then walk forward 25 meters and turn left.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_16", + "correct_answer": "After getting out of the elevator, walk forward ten meters, walk left three meters, then walk forward 25 meters and turn left.", + "original_options": { + "A": "Then walk forward twenty and turn right.", + "B": "Then walk backward twenty-five and turn left.", + "C": "After getting out of the elevator, walk forward ten meters, walk left three meters, then walk forward 25 meters and turn left.", + "D": "After getting out, walk forward ten meters, left three meters, then forward twenty meters and turn right at the end.", + "E": "After getting out, walk forward fifteen meters, left five meters, then forward twenty meters and turn right instead of turning left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 21 + ] + }, + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_9", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is there anyone at the workstation in front of me?", + "answer": "Yes.", + "question_chinese": "面前的工位有人吗？", + "answer_chinese": "有", + "evidence": [ + { + "description": "There is someone at the workstation in front of me.", + "timestamps": { + "start": "00:13:38", + "end": "00:13:39" + } + } + ], + "question_timestamps": "00:14:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_9", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001", + "video_length": "00:36:08", + "qid": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_10", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is there anyone at the workstation in front of me?", + "answer": "No.", + "question_chinese": "面前的工位有人吗？", + "answer_chinese": "没有", + "evidence": [ + { + "description": "No one is at the workstation in front of me.", + "timestamps": { + "start": "00:25:00", + "end": "00:25:01" + } + } + ], + "question_timestamps": "00:26:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "546e4806-a7f5-46bd-bf65-7737db2c5594-001_10", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/55cf338b-c86b-4f6e-85fd-dc4b722d1967-mcq.json b/QAs/55cf338b-c86b-4f6e-85fd-dc4b722d1967-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7cad23a8708e6113243eff959d45b9d30cf53132 --- /dev/null +++ b/QAs/55cf338b-c86b-4f6e-85fd-dc4b722d1967-mcq.json @@ -0,0 +1,654 @@ +[ + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_3_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I take out of the refrigerator?", + "answer": "Four items, and the refrigerator is behind you.", + "question_chinese": "我从冰箱里拿出了几样东西？", + "answer_chinese": "四样，冰箱在你的后方。", + "evidence": [ + { + "description": "1", + "timestamps": { + "start": "00:01:49", + "end": "00:01:56" + } + }, + { + "description": "2", + "timestamps": { + "start": "00:02:03", + "end": "00:02:24" + } + }, + { + "description": "3", + "timestamps": { + "start": "00:05:23", + "end": "00:05:35" + } + }, + { + "description": "4", + "timestamps": { + "start": "00:06:36", + "end": "00:06:55" + } + } + ], + "question_timestamps": "00:12:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two items.", + "B": "Five items.", + "C": "Three items, and the refrigerator is behind you, so you grabbed fewer things than the correct count.", + "D": "Four items, and the refrigerator is behind you.", + "E": "Six items, and the refrigerator is behind you, meaning you took more items out than stated in the answer." + }, + "answer_label": "D", + "answer_text": "Four items, and the refrigerator is behind you.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_3_T1", + "correct_answer": "Four items, and the refrigerator is behind you.", + "original_options": { + "A": "Two items.", + "B": "Five items.", + "C": "Three items, and the refrigerator is behind you, so you grabbed fewer things than the correct count.", + "D": "Four items, and the refrigerator is behind you.", + "E": "Six items, and the refrigerator is behind you, meaning you took more items out than stated in the answer." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 19 + ] + }, + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_3_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I take out of the refrigerator?", + "answer": "Four items, and the refrigerator is on your right.", + "question_chinese": "我从冰箱里拿出了几样东西？", + "answer_chinese": "四样，冰箱在你的右方。", + "evidence": [ + { + "description": "1", + "timestamps": { + "start": "00:01:49", + "end": "00:01:56" + } + }, + { + "description": "2", + "timestamps": { + "start": "00:02:03", + "end": "00:02:24" + } + }, + { + "description": "3", + "timestamps": { + "start": "00:05:23", + "end": "00:05:35" + } + }, + { + "description": "4", + "timestamps": { + "start": "00:06:36", + "end": "00:06:55" + } + } + ], + "question_timestamps": "00:13:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two items.", + "B": "Five items.", + "C": "Three items, and the refrigerator is on your right, so the number taken is less than stated.", + "D": "Six items, and the refrigerator is on your right, implying you removed more items than the answer says.", + "E": "Four items, and the refrigerator is on your right." + }, + "answer_label": "E", + "answer_text": "Four items, and the refrigerator is on your right.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_3_T2", + "correct_answer": "Four items, and the refrigerator is on your right.", + "original_options": { + "A": "Two items.", + "B": "Five items.", + "C": "Three items, and the refrigerator is on your right, so the number taken is less than stated.", + "D": "Six items, and the refrigerator is on your right, implying you removed more items than the answer says.", + "E": "Four items, and the refrigerator is on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the sink?", + "answer": "Turn left and then walk straight for a certain distance.", + "question_chinese": "我如何走到洗手池？", + "answer_chinese": "向左转后直走一段距离。", + "evidence": [ + { + "description": "The location of the sink.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:55" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:06:04", + "end": "00:06:04" + } + } + ], + "question_timestamps": "00:06:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left and then walk straight for a certain distance.", + "B": "Turn right, then go straight.", + "C": "Go straight ahead to the sink.", + "D": "Turn left first, but instead of continuing straight, veer right after a few steps.", + "E": "Turn right at the start and keep walking straight until you reach the sink area." + }, + "answer_label": "A", + "answer_text": "Turn left and then walk straight for a certain distance.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_2", + "correct_answer": "Turn left and then walk straight for a certain distance.", + "original_options": { + "A": "Turn left and then walk straight for a certain distance.", + "B": "Turn right, then go straight.", + "C": "Go straight ahead to the sink.", + "D": "Turn left first, but instead of continuing straight, veer right after a few steps.", + "E": "Turn right at the start and keep walking straight until you reach the sink area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 15 + ] + }, + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_4", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the sink?", + "answer": "It's right in front of you.", + "question_chinese": "我如何走到洗手池？", + "answer_chinese": "它就在你的正前方。", + "evidence": [ + { + "description": "The location of the sink.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:55" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:12:38", + "end": "00:12:38" + } + } + ], + "question_timestamps": "00:12:39", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It's behind you.", + "B": "It's to your left.", + "C": "It's on your right side, along the counter, not directly in front of you.", + "D": "Walk forward a bit; the sink is off to the left rather than straight ahead.", + "E": "It's right in front of you." + }, + "answer_label": "E", + "answer_text": "It's right in front of you.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_4", + "correct_answer": "It's right in front of you.", + "original_options": { + "A": "It's behind you.", + "B": "It's to your left.", + "C": "It's on your right side, along the counter, not directly in front of you.", + "D": "Walk forward a bit; the sink is off to the left rather than straight ahead.", + "E": "It's right in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 15 + ] + }, + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_1_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands?", + "answer": "Yes—you can. The faucet is on your left.", + "question_chinese": "我是否可以洗到手？", + "answer_chinese": "可以，水龙头在你的左方。", + "evidence": [ + { + "description": "The location of the sink.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:55" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:04" + } + } + ], + "question_timestamps": "00:06:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t. The faucet is on your left.", + "B": "Yes—you can. The faucet is on your left." + }, + "answer_label": "B", + "answer_text": "Yes—you can. The faucet is on your left.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_1_T1", + "correct_answer": "Yes—you can. The faucet is on your left.", + "original_options": { + "A": "No—you can’t. The faucet is on your left.", + "B": "Yes—you can. The faucet is on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_1_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands?", + "answer": "No—you can’t. The faucet is behind you.", + "question_chinese": "我是否可以洗到手？", + "answer_chinese": "不可以，水龙头在你的后方。", + "evidence": [ + { + "description": "The location of the sink.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:55" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:04" + } + } + ], + "question_timestamps": "00:06:46", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t. The faucet is behind you.", + "B": "Yes—you can. The faucet is behind you." + }, + "answer_label": "A", + "answer_text": "No—you can’t. The faucet is behind you.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_1_T2", + "correct_answer": "No—you can’t. The faucet is behind you.", + "original_options": { + "A": "No—you can’t. The faucet is behind you.", + "B": "Yes—you can. The faucet is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_4_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I ever put a cup into the cabinet?", + "answer": "Yes, it’s behind you.", + "question_chinese": "我是否往橱柜里面放过杯子？", + "answer_chinese": "是的，在你的后方。", + "evidence": [ + { + "description": "Put the cup.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:06:22", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, it’s behind you.", + "B": "No, you never put a cup into the cabinet." + }, + "answer_label": "A", + "answer_text": "Yes, it’s behind you.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_4_T1", + "correct_answer": "Yes, it’s behind you.", + "original_options": { + "A": "Yes, it’s behind you.", + "B": "No, you never put a cup into the cabinet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_4_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I ever put a cup into the cabinet?", + "answer": "Yes, it’s above you to your right.", + "question_chinese": "我是否往橱柜里面放过杯子？", + "answer_chinese": "是的，在你的右上方。", + "evidence": [ + { + "description": "Put the cup.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:12:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you never put a cup into the cabinet.", + "B": "Yes, it’s above you to your right." + }, + "answer_label": "B", + "answer_text": "Yes, it’s above you to your right.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_4_T2", + "correct_answer": "Yes, it’s above you to your right.", + "original_options": { + "A": "No, you never put a cup into the cabinet.", + "B": "Yes, it’s above you to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_6_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I look in the mirror?", + "answer": "Yes, the mirror is to your back-right.", + "question_chinese": "我是否照过镜子？", + "answer_chinese": "是的，镜子在你的右后。", + "evidence": [ + { + "description": "Look in the mirror.", + "timestamps": { + "start": "00:17:59", + "end": "00:18:06" + } + } + ], + "question_timestamps": "00:19:31", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, the mirror is to your back-right.", + "B": "No, you didn’t look in the mirror." + }, + "answer_label": "A", + "answer_text": "Yes, the mirror is to your back-right.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_6_T1", + "correct_answer": "Yes, the mirror is to your back-right.", + "original_options": { + "A": "Yes, the mirror is to your back-right.", + "B": "No, you didn’t look in the mirror." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_6_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I look in the mirror?", + "answer": "Yes, the mirror is behind you.", + "question_chinese": "我是否照过镜子？", + "answer_chinese": "是的，镜子在你的后方。", + "evidence": [ + { + "description": "Look in the mirror.", + "timestamps": { + "start": "00:17:59", + "end": "00:18:06" + } + } + ], + "question_timestamps": "00:25:32", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you didn’t look in the mirror.", + "B": "Yes, the mirror is behind you." + }, + "answer_label": "B", + "answer_text": "Yes, the mirror is behind you.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_6_T2", + "correct_answer": "Yes, the mirror is behind you.", + "original_options": { + "A": "No, you didn’t look in the mirror.", + "B": "Yes, the mirror is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_5_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stools were around my dining table?", + "answer": "Five, and they’re all behind you.", + "question_chinese": "我的餐桌周围有几个凳子？", + "answer_chinese": "五个，都在我的身后", + "evidence": [ + { + "description": "5", + "timestamps": { + "start": "00:13:01", + "end": "00:13:02" + } + } + ], + "question_timestamps": "00:20:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five, and they’re all behind you.", + "B": "Four, all behind you.", + "C": "Six, all behind you.", + "D": "Three stools, and they’re all behind you around the dining table.", + "E": "Seven stools, and they’re all behind you around the dining table." + }, + "answer_label": "A", + "answer_text": "Five, and they’re all behind you.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_5_T1", + "correct_answer": "Five, and they’re all behind you.", + "original_options": { + "A": "Five, and they’re all behind you.", + "B": "Four, all behind you.", + "C": "Six, all behind you.", + "D": "Three stools, and they’re all behind you around the dining table.", + "E": "Seven stools, and they’re all behind you around the dining table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967", + "video_length": "00:25:32", + "qid": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_5_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stools were around my dining table?", + "answer": "Five, and they’re all in front of you.", + "question_chinese": "我的餐桌周围有几个凳子？", + "answer_chinese": "五个，都在我的前方。", + "evidence": [ + { + "description": "5", + "timestamps": { + "start": "00:13:01", + "end": "00:13:02" + } + } + ], + "question_timestamps": "00:25:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four, all in front of you.", + "B": "Five, and they’re all in front of you.", + "C": "Six, all in front of you.", + "D": "Three stools, and they’re all in front of you around the dining table.", + "E": "Seven stools, and they’re all in front of you around the dining table." + }, + "answer_label": "B", + "answer_text": "Five, and they’re all in front of you.", + "q_id": "55cf338b-c86b-4f6e-85fd-dc4b722d1967_5_T2", + "correct_answer": "Five, and they’re all in front of you.", + "original_options": { + "A": "Four, all in front of you.", + "B": "Five, and they’re all in front of you.", + "C": "Six, all in front of you.", + "D": "Three stools, and they’re all in front of you around the dining table.", + "E": "Seven stools, and they’re all in front of you around the dining table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + } +] diff --git a/QAs/56bfc9ac-00f4-41ae-8112-3851c538d9b9-mcq.json b/QAs/56bfc9ac-00f4-41ae-8112-3851c538d9b9-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..207be8364a9d37b8c3bb98440c510c76f4ecc016 --- /dev/null +++ b/QAs/56bfc9ac-00f4-41ae-8112-3851c538d9b9-mcq.json @@ -0,0 +1,1202 @@ +[ + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_5_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do before cooking?", + "answer": "You washed and dried your hands at the sink to your left.", + "question_chinese": "做饭前我做了什么？", + "answer_chinese": "在你左边的洗手台上洗手和擦手", + "evidence": [ + { + "description": "Washing hands", + "timestamps": { + "start": "00:18:00", + "end": "00:18:01" + } + }, + { + "description": "drying hands", + "timestamps": { + "start": "00:18:17", + "end": "00:18:18" + } + } + ], + "question_timestamps": "00:18:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Hands washed to your right.", + "B": "Hands washed back-right.", + "C": "Before cooking, you washed and dried your hands at the sink on your right side.", + "D": "You washed and dried your hands at the sink to your left.", + "E": "Before cooking, you washed and dried your hands at the sink behind you and slightly to the right." + }, + "answer_label": "D", + "answer_text": "You washed and dried your hands at the sink to your left.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_5_T1", + "correct_answer": "You washed and dried your hands at the sink to your left.", + "original_options": { + "A": "Hands washed to your right.", + "B": "Hands washed back-right.", + "C": "Before cooking, you washed and dried your hands at the sink on your right side.", + "D": "You washed and dried your hands at the sink to your left.", + "E": "Before cooking, you washed and dried your hands at the sink behind you and slightly to the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 15, + 18 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_5_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do before cooking?", + "answer": "You washed and dried your hands at the sink to your back-right.", + "question_chinese": "做饭前我做了什么？", + "answer_chinese": "在你右后边的洗手台上洗手和擦手", + "evidence": [ + { + "description": "Washing hands", + "timestamps": { + "start": "00:18:00", + "end": "00:18:01" + } + }, + { + "description": "drying hands", + "timestamps": { + "start": "00:18:17", + "end": "00:18:18" + } + } + ], + "question_timestamps": "00:43:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed and dried your hands at back-left.", + "B": "You washed and dried your hands front-right.", + "C": "Before cooking, you washed and dried your hands at the sink behind you and to the left.", + "D": "Before cooking, you washed and dried your hands at the sink in front of you and slightly to the right.", + "E": "You washed and dried your hands at the sink to your back-right." + }, + "answer_label": "E", + "answer_text": "You washed and dried your hands at the sink to your back-right.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_5_T2", + "correct_answer": "You washed and dried your hands at the sink to your back-right.", + "original_options": { + "A": "You washed and dried your hands at back-left.", + "B": "You washed and dried your hands front-right.", + "C": "Before cooking, you washed and dried your hands at the sink behind you and to the left.", + "D": "Before cooking, you washed and dried your hands at the sink in front of you and slightly to the right.", + "E": "You washed and dried your hands at the sink to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 20 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_1_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the nightstand relative to me?", + "answer": "On both sides of the bed, to your back-left.", + "question_chinese": "床头柜在我的哪里？", + "answer_chinese": "床两边，在我的左后方。", + "evidence": [ + { + "description": "Both sides of the bed", + "timestamps": { + "start": "00:02:10", + "end": "00:02:11" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On both sides of the bed, to your back-left.", + "B": "To your back-right.", + "C": "To your front-left.", + "D": "On both sides of the bed, positioned to your back-right instead of behind your left shoulder.", + "E": "On both sides of the bed, but they were in front of you, closer to your front-right side." + }, + "answer_label": "A", + "answer_text": "On both sides of the bed, to your back-left.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_1_T2", + "correct_answer": "On both sides of the bed, to your back-left.", + "original_options": { + "A": "On both sides of the bed, to your back-left.", + "B": "To your back-right.", + "C": "To your front-left.", + "D": "On both sides of the bed, positioned to your back-right instead of behind your left shoulder.", + "E": "On both sides of the bed, but they were in front of you, closer to your front-right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 18 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_1_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the nightstand relative to me?", + "answer": "On both sides of the bed—above you and above-left of you.", + "question_chinese": "床头柜在我的哪里？", + "answer_chinese": "床两边，你的上方和左上方", + "evidence": [ + { + "description": "Both sides of the bed", + "timestamps": { + "start": "00:02:10", + "end": "00:02:11" + } + } + ], + "question_timestamps": "00:02:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above you and above-right.", + "B": "Below you and above-left.", + "C": "On both sides of the bed—above you, but shifted toward your above-right rather than your above-left.", + "D": "On both sides of the bed—still above you, though it sat closer to your lower-left area than above-left.", + "E": "On both sides of the bed—above you and above-left of you." + }, + "answer_label": "E", + "answer_text": "On both sides of the bed—above you and above-left of you.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_1_T1", + "correct_answer": "On both sides of the bed—above you and above-left of you.", + "original_options": { + "A": "Above you and above-right.", + "B": "Below you and above-left.", + "C": "On both sides of the bed—above you, but shifted toward your above-right rather than your above-left.", + "D": "On both sides of the bed—still above you, though it sat closer to your lower-left area than above-left.", + "E": "On both sides of the bed—above you and above-left of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 19 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_4_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I just make?", + "answer": "Two: you first turned right, then turned left.", + "question_chinese": "我刚才拐了多少个弯？", + "answer_chinese": "2个，先是右拐，然后左拐。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:17:29", + "end": "00:17:30" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:17:41", + "end": "00:17:42" + } + } + ], + "question_timestamps": "00:18:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One: you turned right once.", + "B": "Zero: you didn’t make any turns.", + "C": "Three: you turned right, then left, then turned right again before continuing forward.", + "D": "Two: you first turned right, then turned left.", + "E": "Two: you first turned left, then turned right, making two turns in the opposite order." + }, + "answer_label": "D", + "answer_text": "Two: you first turned right, then turned left.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_4_T2", + "correct_answer": "Two: you first turned right, then turned left.", + "original_options": { + "A": "One: you turned right once.", + "B": "Zero: you didn’t make any turns.", + "C": "Three: you turned right, then left, then turned right again before continuing forward.", + "D": "Two: you first turned right, then turned left.", + "E": "Two: you first turned left, then turned right, making two turns in the opposite order." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 13, + 15 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_4_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I just make?", + "answer": "Four: you walked from the living room behind you to the kitchen in front of you.", + "question_chinese": "我刚才拐了多少个弯？", + "answer_chinese": "4个，从后方的客厅走到前方的厨房。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:17:29", + "end": "00:17:30" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:17:41", + "end": "00:17:42" + } + } + ], + "question_timestamps": "00:18:29", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three turns.", + "B": "Five turns.", + "C": "Four: you walked from the living room behind you to the kitchen in front of you.", + "D": "Two: you walked from the living room behind you to the kitchen in front of you with only two turns.", + "E": "Six: you walked from the living room behind you to the kitchen in front of you, making six turns overall." + }, + "answer_label": "C", + "answer_text": "Four: you walked from the living room behind you to the kitchen in front of you.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_4_T1", + "correct_answer": "Four: you walked from the living room behind you to the kitchen in front of you.", + "original_options": { + "A": "Three turns.", + "B": "Five turns.", + "C": "Four: you walked from the living room behind you to the kitchen in front of you.", + "D": "Two: you walked from the living room behind you to the kitchen in front of you with only two turns.", + "E": "Six: you walked from the living room behind you to the kitchen in front of you, making six turns overall." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 20, + 20 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_3_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What should I do if I want to look up some information?", + "answer": "Walk forward two meters, then turn on the computer on your right-hand side.", + "question_chinese": "我想查资料该怎么办？", + "answer_chinese": "往前走两米，打开你右手边的电脑", + "evidence": [ + { + "description": "Walk forward two meters", + "timestamps": { + "start": "00:03:31", + "end": "00:03:32" + } + }, + { + "description": "turn on the computer.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:03:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk forward one meter.", + "B": "Walk forward two meters, then turn on the computer on your right-hand side.", + "C": "Walk forward three meters.", + "D": "Walk forward four meters, then turn on the computer on your right-hand side to look things up.", + "E": "Walk forward zero meters, then turn on the computer on your right-hand side without moving first." + }, + "answer_label": "B", + "answer_text": "Walk forward two meters, then turn on the computer on your right-hand side.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_3_T1", + "correct_answer": "Walk forward two meters, then turn on the computer on your right-hand side.", + "original_options": { + "A": "Walk forward one meter.", + "B": "Walk forward two meters, then turn on the computer on your right-hand side.", + "C": "Walk forward three meters.", + "D": "Walk forward four meters, then turn on the computer on your right-hand side to look things up.", + "E": "Walk forward zero meters, then turn on the computer on your right-hand side without moving first." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_3_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What should I do if I want to look up some information?", + "answer": "Turn on the computer to your back-right.", + "question_chinese": "我想查资料该怎么办？", + "answer_chinese": "打开你右后方的电脑", + "evidence": [ + { + "description": "Walk forward two meters", + "timestamps": { + "start": "00:03:31", + "end": "00:03:32" + } + }, + { + "description": "turn on the computer.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:18:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn on the computer to your back-left.", + "B": "Turn on the computer to your front-right.", + "C": "Turn on the computer to your back-right.", + "D": "Turn on the computer to your front-left, then use it to look up the information you need.", + "E": "Walk forward two meters, then turn on the computer on your right-hand side to search for information." + }, + "answer_label": "C", + "answer_text": "Turn on the computer to your back-right.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_3_T2", + "correct_answer": "Turn on the computer to your back-right.", + "original_options": { + "A": "Turn on the computer to your back-left.", + "B": "Turn on the computer to your front-right.", + "C": "Turn on the computer to your back-right.", + "D": "Turn on the computer to your front-left, then use it to look up the information you need.", + "E": "Walk forward two meters, then turn on the computer on your right-hand side to search for information." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_8_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates did I take?", + "answer": "Two, and they’re both to your back-left.", + "question_chinese": "我拿了几个盘子？", + "answer_chinese": "2，都在你的左后方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:20:18", + "end": "00:20:19" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:20:30", + "end": "00:20:31" + } + } + ], + "question_timestamps": "00:20:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, and it’s to your back-left.", + "B": "Two, and they’re both to your back-left.", + "C": "Zero—you didn’t take any plates.", + "D": "Three, and they’re all to your back-left, stacked together behind your left shoulder.", + "E": "Four, and they’re all to your back-left, grouped in the same area behind you on the left." + }, + "answer_label": "B", + "answer_text": "Two, and they’re both to your back-left.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_8_T1", + "correct_answer": "Two, and they’re both to your back-left.", + "original_options": { + "A": "One, and it’s to your back-left.", + "B": "Two, and they’re both to your back-left.", + "C": "Zero—you didn’t take any plates.", + "D": "Three, and they’re all to your back-left, stacked together behind your left shoulder.", + "E": "Four, and they’re all to your back-left, grouped in the same area behind you on the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 18 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_8_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates did I take?", + "answer": "Two, and they’re both to your back-right.", + "question_chinese": "我拿了几个盘子？", + "answer_chinese": "2个，都在你的右后方。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:20:18", + "end": "00:20:19" + } + }, + { + "description": "the first one", + "timestamps": { + "start": "00:20:30", + "end": "00:20:31" + } + } + ], + "question_timestamps": "00:29:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, and it’s to your back-right.", + "B": "Zero—you didn’t take any plates.", + "C": "Two, and they’re both to your back-right.", + "D": "Three, and they’re all to your back-right, collected together behind your right shoulder.", + "E": "Four, and they’re all to your back-right, kept in the same spot behind you on the right." + }, + "answer_label": "C", + "answer_text": "Two, and they’re both to your back-right.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_8_T2", + "correct_answer": "Two, and they’re both to your back-right.", + "original_options": { + "A": "One, and it’s to your back-right.", + "B": "Zero—you didn’t take any plates.", + "C": "Two, and they’re both to your back-right.", + "D": "Three, and they’re all to your back-right, collected together behind your right shoulder.", + "E": "Four, and they’re all to your back-right, kept in the same spot behind you on the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 18 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_6_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get the eggs?", + "answer": "Walk forward one meter, open the fridge, then take the egg.", + "question_chinese": "我要怎么才能拿到鸡蛋？", + "answer_chinese": "往前走一米，开冰箱，然后拿鸡蛋", + "evidence": [ + { + "description": "Walk forward one meter", + "timestamps": { + "start": "00:19:16", + "end": "00:19:37" + } + }, + { + "description": "open the fridge", + "timestamps": { + "start": "00:19:41", + "end": "00:19:42" + } + }, + { + "description": "take the eggs", + "timestamps": { + "start": "00:20:11", + "end": "00:20:12" + } + } + ], + "question_timestamps": "00:20:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk forward one meter, open the fridge, then take the egg.", + "B": "Walk forward two meters, then take the egg.", + "C": "Walk backward one meter first.", + "D": "Walk forward three meters, open the fridge, and then take the egg from inside.", + "E": "Walk forward one meter, open the freezer, then take the egg from inside." + }, + "answer_label": "A", + "answer_text": "Walk forward one meter, open the fridge, then take the egg.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_6_T2", + "correct_answer": "Walk forward one meter, open the fridge, then take the egg.", + "original_options": { + "A": "Walk forward one meter, open the fridge, then take the egg.", + "B": "Walk forward two meters, then take the egg.", + "C": "Walk backward one meter first.", + "D": "Walk forward three meters, open the fridge, and then take the egg from inside.", + "E": "Walk forward one meter, open the freezer, then take the egg from inside." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 5, + 14, + 13 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_6_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get the eggs?", + "answer": "Go to the fridge to your back-right, then take the egg.", + "question_chinese": "我要怎么才能拿到鸡蛋？", + "answer_chinese": "前往你右后方的冰箱，然后拿取鸡蛋", + "evidence": [ + { + "description": "Walk forward one meter", + "timestamps": { + "start": "00:19:16", + "end": "00:19:37" + } + }, + { + "description": "open the fridge", + "timestamps": { + "start": "00:19:41", + "end": "00:19:42" + } + }, + { + "description": "take the eggs", + "timestamps": { + "start": "00:20:11", + "end": "00:20:12" + } + } + ], + "question_timestamps": "00:20:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go to the fridge on your back-left.", + "B": "Go to the fridge in front-right.", + "C": "Go to the fridge to your front-left, then take the egg once you reach it.", + "D": "Walk forward one meter to the fridge, open it, and then take the egg inside.", + "E": "Go to the fridge to your back-right, then take the egg." + }, + "answer_label": "E", + "answer_text": "Go to the fridge to your back-right, then take the egg.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_6_T1", + "correct_answer": "Go to the fridge to your back-right, then take the egg.", + "original_options": { + "A": "Go to the fridge on your back-left.", + "B": "Go to the fridge in front-right.", + "C": "Go to the fridge to your front-left, then take the egg once you reach it.", + "D": "Walk forward one meter to the fridge, open it, and then take the egg inside.", + "E": "Go to the fridge to your back-right, then take the egg." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 15 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_7_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I take the eggs?", + "answer": "Walk forward one meter, open the fridge on the right side, then take the egg.", + "question_chinese": "我要怎么拿鸡蛋？", + "answer_chinese": "往前走一米，开右侧的冰箱，然后拿取鸡蛋。", + "evidence": [ + { + "description": "Walk forward one meter", + "timestamps": { + "start": "00:19:36", + "end": "00:19:37" + } + }, + { + "description": "open the fridge", + "timestamps": { + "start": "00:19:41", + "end": "00:19:42" + } + }, + { + "description": "take the eggs", + "timestamps": { + "start": "00:20:11", + "end": "00:20:12" + } + } + ], + "question_timestamps": "00:20:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Open the fridge on the left.", + "B": "Walk forward one meter, open the fridge on the right side, then take the egg.", + "C": "Walk forward two meters, then take the egg.", + "D": "Walk forward one meter, open the fridge on the left side, then take the egg.", + "E": "Walk forward three meters, open the fridge on the right side, then take the egg out." + }, + "answer_label": "B", + "answer_text": "Walk forward one meter, open the fridge on the right side, then take the egg.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_7_T2", + "correct_answer": "Walk forward one meter, open the fridge on the right side, then take the egg.", + "original_options": { + "A": "Open the fridge on the left.", + "B": "Walk forward one meter, open the fridge on the right side, then take the egg.", + "C": "Walk forward two meters, then take the egg.", + "D": "Walk forward one meter, open the fridge on the left side, then take the egg.", + "E": "Walk forward three meters, open the fridge on the right side, then take the egg out." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 15, + 16 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_7_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I take the eggs?", + "answer": "Walk forward one meter, open the fridge, then take the egg.", + "question_chinese": "我要怎么拿鸡蛋？", + "answer_chinese": "往前走一米，开冰箱，然后拿鸡蛋", + "evidence": [ + { + "description": "Walk forward one meter", + "timestamps": { + "start": "00:19:36", + "end": "00:19:37" + } + }, + { + "description": "open the fridge", + "timestamps": { + "start": "00:19:41", + "end": "00:19:42" + } + }, + { + "description": "take the eggs", + "timestamps": { + "start": "00:20:11", + "end": "00:20:12" + } + } + ], + "question_timestamps": "00:20:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk forward one meter, open the fridge, then take the egg.", + "B": "Walk forward two meters, then take the egg.", + "C": "Walk backward one meter first.", + "D": "Walk forward three meters, open the fridge, and then take the egg from inside.", + "E": "Walk forward one meter, open the fridge, then take the eggs from the bottom shelf." + }, + "answer_label": "A", + "answer_text": "Walk forward one meter, open the fridge, then take the egg.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_7_T1", + "correct_answer": "Walk forward one meter, open the fridge, then take the egg.", + "original_options": { + "A": "Walk forward one meter, open the fridge, then take the egg.", + "B": "Walk forward two meters, then take the egg.", + "C": "Walk backward one meter first.", + "D": "Walk forward three meters, open the fridge, and then take the egg from inside.", + "E": "Walk forward one meter, open the fridge, then take the eggs from the bottom shelf." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 5, + 14, + 15 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_9_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I fry the bread?", + "answer": "You mixed the egg mixture, coated the bread fully in it, and then fried it. The bread is to your back-right.", + "question_chinese": "我要怎么煎面包？", + "answer_chinese": "调蛋液，将面包沾满蛋液，然后煎，面包在你的右后方。。", + "evidence": [ + { + "description": "Beat the eggs", + "timestamps": { + "start": "00:25:18", + "end": "00:25:19" + } + }, + { + "description": "dip the bread completely in the egg mixture", + "timestamps": { + "start": "00:28:25", + "end": "00:28:26" + } + }, + { + "description": "fry", + "timestamps": { + "start": "00:28:56", + "end": "00:28:57" + } + } + ], + "question_timestamps": "00:29:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bread is behind you.", + "B": "The bread is to your front-left.", + "C": "You mixed the egg mixture, coated the bread fully in it, and then fried it. The bread is to your back-right.", + "D": "You mixed the egg mixture, coated the bread, and fried it; the bread is to your back-left.", + "E": "You mixed the egg mixture and fried the bread after coating it; the bread is in front of you." + }, + "answer_label": "C", + "answer_text": "You mixed the egg mixture, coated the bread fully in it, and then fried it. The bread is to your back-right.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_9_T1", + "correct_answer": "You mixed the egg mixture, coated the bread fully in it, and then fried it. The bread is to your back-right.", + "original_options": { + "A": "The bread is behind you.", + "B": "The bread is to your front-left.", + "C": "You mixed the egg mixture, coated the bread fully in it, and then fried it. The bread is to your back-right.", + "D": "You mixed the egg mixture, coated the bread, and fried it; the bread is to your back-left.", + "E": "You mixed the egg mixture and fried the bread after coating it; the bread is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 19 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_9_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I fry the bread?", + "answer": "You mixed the egg mixture, coated the bread fully in it, and then fried it. The bread is in front of you.", + "question_chinese": "我要怎么煎面包？", + "answer_chinese": "调蛋液，将面包沾满蛋液，然后煎。面包在你的前方。", + "evidence": [ + { + "description": "Beat the eggs", + "timestamps": { + "start": "00:25:18", + "end": "00:25:19" + } + }, + { + "description": "dip the bread completely in the egg mixture", + "timestamps": { + "start": "00:28:25", + "end": "00:28:26" + } + }, + { + "description": "fry", + "timestamps": { + "start": "00:28:56", + "end": "00:28:57" + } + } + ], + "question_timestamps": "00:29:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bread is behind you.", + "B": "The bread is to your back-right.", + "C": "You mixed the egg mixture, coated the bread, and fried it; the bread is to your back-left.", + "D": "You mixed the egg mixture, coated the bread fully in it, and then fried it. The bread is in front of you.", + "E": "You mixed the egg mixture and fried the bread after coating it; the bread is to your front-right." + }, + "answer_label": "D", + "answer_text": "You mixed the egg mixture, coated the bread fully in it, and then fried it. The bread is in front of you.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_9_T2", + "correct_answer": "You mixed the egg mixture, coated the bread fully in it, and then fried it. The bread is in front of you.", + "original_options": { + "A": "The bread is behind you.", + "B": "The bread is to your back-right.", + "C": "You mixed the egg mixture, coated the bread, and fried it; the bread is to your back-left.", + "D": "You mixed the egg mixture, coated the bread fully in it, and then fried it. The bread is in front of you.", + "E": "You mixed the egg mixture and fried the bread after coating it; the bread is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 18 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_5", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the sofa?", + "answer": "White.", + "question_chinese": "沙发是什么颜色的？", + "answer_chinese": "白色", + "evidence": [ + { + "description": "White", + "timestamps": { + "start": "00:42:36", + "end": "00:42:37" + } + } + ], + "question_timestamps": "00:42:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Red.", + "B": "Black.", + "C": "The sofa is gray, not white, matching the room’s neutral tones.", + "D": "The sofa is green rather than white, giving it a more colorful look.", + "E": "White." + }, + "answer_label": "E", + "answer_text": "White.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_5", + "correct_answer": "White.", + "original_options": { + "A": "Red.", + "B": "Black.", + "C": "The sofa is gray, not white, matching the room’s neutral tones.", + "D": "The sofa is green rather than white, giving it a more colorful look.", + "E": "White." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_6", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the sofa?", + "answer": "White.", + "question_chinese": "沙发是什么颜色的？", + "answer_chinese": "白的", + "evidence": [ + { + "description": "White", + "timestamps": { + "start": "00:42:37", + "end": "00:42:38" + } + } + ], + "question_timestamps": "00:42:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Red.", + "B": "Black.", + "C": "The sofa is gray, not white, matching the room’s neutral tones.", + "D": "White.", + "E": "The sofa is green rather than white, giving it a more colorful look." + }, + "answer_label": "D", + "answer_text": "White.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_6", + "correct_answer": "White.", + "original_options": { + "A": "Red.", + "B": "Black.", + "C": "The sofa is gray, not white, matching the room’s neutral tones.", + "D": "White.", + "E": "The sofa is green rather than white, giving it a more colorful look." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_2_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the blue chair relative to me?", + "answer": "To your left.", + "question_chinese": "蓝色的椅子在我的哪里？", + "answer_chinese": "我的左边", + "evidence": [ + { + "description": "My left side", + "timestamps": { + "start": "00:49:48", + "end": "00:49:49" + } + } + ], + "question_timestamps": "00:49:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your right.", + "B": "To your left.", + "C": "In front of you.", + "D": "It was positioned slightly behind you on the right side, not on your left.", + "E": "You had it off to your back-left area, rather than directly to your left side." + }, + "answer_label": "B", + "answer_text": "To your left.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_2_T2", + "correct_answer": "To your left.", + "original_options": { + "A": "To your right.", + "B": "To your left.", + "C": "In front of you.", + "D": "It was positioned slightly behind you on the right side, not on your left.", + "E": "You had it off to your back-left area, rather than directly to your left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 15 + ] + }, + { + "video_uid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9", + "video_length": "00:55:40", + "qid": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_2_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the blue chair relative to me?", + "answer": "To your back-right.", + "question_chinese": "蓝色的椅子在我的哪里？", + "answer_chinese": "在我的右后方。", + "evidence": [ + { + "description": "My left side", + "timestamps": { + "start": "00:49:48", + "end": "00:49:49" + } + } + ], + "question_timestamps": "00:53:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-right.", + "B": "To your back-left.", + "C": "To your front-right.", + "D": "It was located to your right side beside you, not behind you on the right.", + "E": "You would have seen it ahead on your left, rather than behind you to the right." + }, + "answer_label": "A", + "answer_text": "To your back-right.", + "q_id": "56bfc9ac-00f4-41ae-8112-3851c538d9b9_2_T1", + "correct_answer": "To your back-right.", + "original_options": { + "A": "To your back-right.", + "B": "To your back-left.", + "C": "To your front-right.", + "D": "It was located to your right side beside you, not behind you on the right.", + "E": "You would have seen it ahead on your left, rather than behind you to the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + } +] diff --git a/QAs/5b35682f-c4e4-4467-9718-8c2febcc1bce-mcq.json b/QAs/5b35682f-c4e4-4467-9718-8c2febcc1bce-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..59eb750539480e171e9a8e987356388a8c078181 --- /dev/null +++ b/QAs/5b35682f-c4e4-4467-9718-8c2febcc1bce-mcq.json @@ -0,0 +1,687 @@ +[ + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_9", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I sharpened my pencil in total?", + "answer": "You sharpened the pencil twice in total.", + "question_chinese": "我一共削了几次铅笔？", + "answer_chinese": "你一共削了两次铅笔。", + "evidence": [ + { + "description": "You sharpened the pencil for the first time.", + "timestamps": { + "start": "00:18:16", + "end": "00:18:27" + } + }, + { + "description": "You sharpened the pencil for the second time.", + "timestamps": { + "start": "00:20:15", + "end": "00:20:27" + } + } + ], + "question_timestamps": "00:20:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You sharpened the pencil twice in total.", + "B": "You sharpened it once.", + "C": "You sharpened it three times.", + "D": "You sharpened the pencil four times in total throughout the whole sequence.", + "E": "You sharpened the pencil five times in total, counting every sharpening you did." + }, + "answer_label": "A", + "answer_text": "You sharpened the pencil twice in total.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_9", + "correct_answer": "You sharpened the pencil twice in total.", + "original_options": { + "A": "You sharpened the pencil twice in total.", + "B": "You sharpened it once.", + "C": "You sharpened it three times.", + "D": "You sharpened the pencil four times in total throughout the whole sequence.", + "E": "You sharpened the pencil five times in total, counting every sharpening you did." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 12, + 13 + ] + }, + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_10", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I sharpened my pencil in total?", + "answer": "You sharpened the pencil three times in total.", + "question_chinese": "我一共削了几次铅笔？", + "answer_chinese": "你一共削了三次铅笔。", + "evidence": [ + { + "description": "You sharpened the pencil for the first time.", + "timestamps": { + "start": "00:18:16", + "end": "00:18:27" + } + }, + { + "description": "You sharpened the pencil for the second time.", + "timestamps": { + "start": "00:20:15", + "end": "00:20:27" + } + }, + { + "description": "You sharpened the pencil for the third time.", + "timestamps": { + "start": "00:20:29", + "end": "00:20:39" + } + } + ], + "question_timestamps": "00:29:39", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You sharpened it twice.", + "B": "You sharpened it four times.", + "C": "You sharpened the pencil once in total, with only a single sharpening overall.", + "D": "You sharpened the pencil three times in total.", + "E": "You sharpened the pencil five times in total, adding up each time you sharpened it." + }, + "answer_label": "D", + "answer_text": "You sharpened the pencil three times in total.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_10", + "correct_answer": "You sharpened the pencil three times in total.", + "original_options": { + "A": "You sharpened it twice.", + "B": "You sharpened it four times.", + "C": "You sharpened the pencil once in total, with only a single sharpening overall.", + "D": "You sharpened the pencil three times in total.", + "E": "You sharpened the pencil five times in total, adding up each time you sharpened it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 15 + ] + }, + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow ladders did I move in total?", + "answer": "You moved a yellow ladder altogether.", + "question_chinese": "我一共搬了几个黄色的梯子？", + "answer_chinese": "你一共搬了一个黄色的梯子。", + "evidence": [ + { + "description": "You moved the first yellow ladder.", + "timestamps": { + "start": "00:13:22", + "end": "00:13:33" + } + } + ], + "question_timestamps": "00:13:35", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You moved none.", + "B": "You moved a yellow ladder altogether.", + "C": "You moved two ladders.", + "D": "You moved three yellow ladders altogether during the entire set of actions.", + "E": "You moved four yellow ladders altogether, counting each ladder you relocated there." + }, + "answer_label": "B", + "answer_text": "You moved a yellow ladder altogether.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_7", + "correct_answer": "You moved a yellow ladder altogether.", + "original_options": { + "A": "You moved none.", + "B": "You moved a yellow ladder altogether.", + "C": "You moved two ladders.", + "D": "You moved three yellow ladders altogether during the entire set of actions.", + "E": "You moved four yellow ladders altogether, counting each ladder you relocated there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 12 + ] + }, + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_8", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow ladders did I move in total?", + "answer": "You moved a total of two yellow ladders.", + "question_chinese": "我一共搬了几个黄色的梯子？", + "answer_chinese": "你一共搬了两个黄色的梯子。", + "evidence": [ + { + "description": "You moved the first yellow ladder.", + "timestamps": { + "start": "00:13:22", + "end": "00:13:33" + } + }, + { + "description": "You moved the second yellow ladder.", + "timestamps": { + "start": "00:13:41", + "end": "00:13:58" + } + } + ], + "question_timestamps": "00:24:02", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You moved one ladder.", + "B": "You moved three ladders.", + "C": "You moved a total of four yellow ladders across all the times you moved them.", + "D": "You moved a total of five yellow ladders, counting every ladder you repositioned.", + "E": "You moved a total of two yellow ladders." + }, + "answer_label": "E", + "answer_text": "You moved a total of two yellow ladders.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_8", + "correct_answer": "You moved a total of two yellow ladders.", + "original_options": { + "A": "You moved one ladder.", + "B": "You moved three ladders.", + "C": "You moved a total of four yellow ladders across all the times you moved them.", + "D": "You moved a total of five yellow ladders, counting every ladder you repositioned.", + "E": "You moved a total of two yellow ladders." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 13 + ] + }, + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_5", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many electric drills did I use in total?", + "answer": "You used an electric drill.", + "question_chinese": "我一共用了几种电钻？", + "answer_chinese": "你用了一种电钻。", + "evidence": [ + { + "description": "You used a long electric drill.", + "timestamps": { + "start": "00:01:59", + "end": "00:02:14" + } + } + ], + "question_timestamps": "00:02:16", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used no electric drills.", + "B": "You used two electric drills.", + "C": "You used an electric drill.", + "D": "You used three electric drills in total during the task, not just one.", + "E": "You used four electric drills overall, counting each drill you picked up to use." + }, + "answer_label": "C", + "answer_text": "You used an electric drill.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_5", + "correct_answer": "You used an electric drill.", + "original_options": { + "A": "You used no electric drills.", + "B": "You used two electric drills.", + "C": "You used an electric drill.", + "D": "You used three electric drills in total during the task, not just one.", + "E": "You used four electric drills overall, counting each drill you picked up to use." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_6", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many electric drills did I use in total?", + "answer": "You used two types of electric drills.", + "question_chinese": "我一共用了几种电钻？", + "answer_chinese": "你用了两种电钻。", + "evidence": [ + { + "description": "You used a long electric drill.", + "timestamps": { + "start": "00:01:59", + "end": "00:02:14" + } + }, + { + "description": "You used a short electric drill.", + "timestamps": { + "start": "00:02:28", + "end": "00:02:34" + } + } + ], + "question_timestamps": "00:22:34", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You used one type.", + "B": "You used three types.", + "C": "You used two types of electric drills.", + "D": "You used four types of electric drills across the task, not only two types.", + "E": "You used five types of electric drills in total, counting each distinct drill type used." + }, + "answer_label": "C", + "answer_text": "You used two types of electric drills.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_6", + "correct_answer": "You used two types of electric drills.", + "original_options": { + "A": "You used one type.", + "B": "You used three types.", + "C": "You used two types of electric drills.", + "D": "You used four types of electric drills across the task, not only two types.", + "E": "You used five types of electric drills in total, counting each distinct drill type used." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You was holding a wooden board and an electric circular saw.", + "question_chinese": "我现在手里拿着什么？", + "answer_chinese": "你拿着木板和电动圆锯。", + "evidence": [ + { + "description": "You are holding the wooden board with your left hand and the electric circular saw with your right hand.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You was holding a wooden board and an electric circular saw.", + "B": "A metal rod and a drill.", + "C": "A wooden plank and a handsaw.", + "D": "You were holding a plywood board while using an electric jigsaw in your hand.", + "E": "You were holding a wooden board and a cordless drill, not an electric circular saw." + }, + "answer_label": "A", + "answer_text": "You was holding a wooden board and an electric circular saw.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_1", + "correct_answer": "You was holding a wooden board and an electric circular saw.", + "original_options": { + "A": "You was holding a wooden board and an electric circular saw.", + "B": "A metal rod and a drill.", + "C": "A wooden plank and a handsaw.", + "D": "You were holding a plywood board while using an electric jigsaw in your hand.", + "E": "You were holding a wooden board and a cordless drill, not an electric circular saw." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You are holding a tape measure and a pen.", + "question_chinese": "我现在手里拿着什么？", + "answer_chinese": "你拿着卷尺和笔。", + "evidence": [ + { + "description": "You are holding a tape measure in your left hand and a pen in your right hand.", + "timestamps": { + "start": "00:16:14", + "end": "00:16:19" + } + } + ], + "question_timestamps": "00:16:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are holding a ruler and a marker.", + "B": "You are holding a notebook and a pencil.", + "C": "You are holding a tape measure and a pencil, as if marking measurements.", + "D": "You are holding a ruler with a pen, rather than a tape measure.", + "E": "You are holding a tape measure and a pen." + }, + "answer_label": "E", + "answer_text": "You are holding a tape measure and a pen.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_2", + "correct_answer": "You are holding a tape measure and a pen.", + "original_options": { + "A": "You are holding a ruler and a marker.", + "B": "You are holding a notebook and a pencil.", + "C": "You are holding a tape measure and a pencil, as if marking measurements.", + "D": "You are holding a ruler with a pen, rather than a tape measure.", + "E": "You are holding a tape measure and a pen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 13, + 13 + ] + }, + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_11", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the gas gun take from the wooden board to its current position?", + "answer": "The gas gun was picked up from the wooden board, moved to the right, and is now placed in front of the stone pillar.", + "question_chinese": "瓦斯枪从木板上到现在的位置经过了什么路径?", + "answer_chinese": "瓦斯枪从木板上拿起,向右移动,现在放置在石柱前。", + "evidence": [ + { + "description": "You pick up the gas gun and move it to the right, placing it in front of the stone pillar.", + "timestamps": { + "start": "00:20:47", + "end": "00:20:55" + } + } + ], + "question_timestamps": "00:21:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The gas gun was picked up from the wooden board, moved to the right, and is now placed in front of the stone pillar.", + "B": "It went left to the stone pillar.", + "C": "It was moved straight ahead to the pillar.", + "D": "The gas gun was picked up from the wooden board, moved behind you, and placed in front of the stone pillar.", + "E": "The gas gun was lifted off the wooden board, shifted to your left, then set down in front of the stone pillar." + }, + "answer_label": "A", + "answer_text": "The gas gun was picked up from the wooden board, moved to the right, and is now placed in front of the stone pillar.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_11", + "correct_answer": "The gas gun was picked up from the wooden board, moved to the right, and is now placed in front of the stone pillar.", + "original_options": { + "A": "The gas gun was picked up from the wooden board, moved to the right, and is now placed in front of the stone pillar.", + "B": "It went left to the stone pillar.", + "C": "It was moved straight ahead to the pillar.", + "D": "The gas gun was picked up from the wooden board, moved behind you, and placed in front of the stone pillar.", + "E": "The gas gun was lifted off the wooden board, shifted to your left, then set down in front of the stone pillar." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 21, + 22 + ] + }, + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_12", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the gas gun take from the wooden board to its current position?", + "answer": "The gas gun was taken off the wooden board, moved to the right, and now placed on the ground.", + "question_chinese": "瓦斯枪从木板上到现在的位置经过了什么路径?", + "answer_chinese": "瓦斯枪从木板上拿开，向右移动，现在放置在地上。", + "evidence": [ + { + "description": "You take the gas gun and move it to the right to place it on the ground.", + "timestamps": { + "start": "00:28:33", + "end": "00:28:44" + } + } + ], + "question_timestamps": "00:28:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It went left and ended on the ground.", + "B": "It was moved forward and set down.", + "C": "The gas gun was taken off the wooden board, moved behind you, and then placed on the ground.", + "D": "The gas gun was removed from the wooden board, carried to your left, and set down on the ground.", + "E": "The gas gun was taken off the wooden board, moved to the right, and now placed on the ground." + }, + "answer_label": "E", + "answer_text": "The gas gun was taken off the wooden board, moved to the right, and now placed on the ground.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_12", + "correct_answer": "The gas gun was taken off the wooden board, moved to the right, and now placed on the ground.", + "original_options": { + "A": "It went left and ended on the ground.", + "B": "It was moved forward and set down.", + "C": "The gas gun was taken off the wooden board, moved behind you, and then placed on the ground.", + "D": "The gas gun was removed from the wooden board, carried to your left, and set down on the ground.", + "E": "The gas gun was taken off the wooden board, moved to the right, and now placed on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 19 + ] + }, + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_3", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Compared to my current position, where is the hammer?", + "answer": "A hammer is on your right, a hammer is in front of you, and a hammer is on your left.", + "question_chinese": "相对我现在的位置，锤子在哪里？", + "answer_chinese": "一个锤子在你的右边，一个锤子在你的前边，一个锤子在你的左边。", + "evidence": [ + { + "description": "There is a hammer on your right and another one in front of you.", + "timestamps": { + "start": "00:01:36", + "end": "00:01:44" + } + }, + { + "description": "There is a hammer on your left.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:02:01", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Hammers are right, left, and behind you.", + "B": "Hammers are right, front, and behind you.", + "C": "A hammer is on your right, a hammer is behind you, and a hammer is on your left.", + "D": "A hammer is on your right, another is to your right, and the third hammer is on your left.", + "E": "A hammer is on your right, a hammer is in front of you, and a hammer is on your left." + }, + "answer_label": "E", + "answer_text": "A hammer is on your right, a hammer is in front of you, and a hammer is on your left.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_3", + "correct_answer": "A hammer is on your right, a hammer is in front of you, and a hammer is on your left.", + "original_options": { + "A": "Hammers are right, left, and behind you.", + "B": "Hammers are right, front, and behind you.", + "C": "A hammer is on your right, a hammer is behind you, and a hammer is on your left.", + "D": "A hammer is on your right, another is to your right, and the third hammer is on your left.", + "E": "A hammer is on your right, a hammer is in front of you, and a hammer is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 19 + ] + }, + { + "video_uid": "5b35682f-c4e4-4467-9718-8c2febcc1bce", + "video_length": "00:30:50", + "qid": "5b35682f-c4e4-4467-9718-8c2febcc1bce_4", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Compared to my current position, where is the hammer?", + "answer": "One hammer is in front of you, and two hammers are on your right.", + "question_chinese": "相对我现在的位置，锤子在哪里？", + "answer_chinese": "一个锤子在你前边，两个锤子在你右边。", + "evidence": [ + { + "description": "There is a hammer in front of you.", + "timestamps": { + "start": "00:10:22", + "end": "00:10:27" + } + }, + { + "description": "There are two hammers on your right.", + "timestamps": { + "start": "00:10:43", + "end": "00:10:44" + } + } + ], + "question_timestamps": "00:10:49", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "One is behind you; two are left.", + "B": "One hammer is in front of you, and two hammers are on your right.", + "C": "Two are in front; one is right.", + "D": "One hammer is in front of you, and two hammers are on your left.", + "E": "One hammer is on your left, and two hammers are behind you instead of on your right." + }, + "answer_label": "B", + "answer_text": "One hammer is in front of you, and two hammers are on your right.", + "q_id": "5b35682f-c4e4-4467-9718-8c2febcc1bce_4", + "correct_answer": "One hammer is in front of you, and two hammers are on your right.", + "original_options": { + "A": "One is behind you; two are left.", + "B": "One hammer is in front of you, and two hammers are on your right.", + "C": "Two are in front; one is right.", + "D": "One hammer is in front of you, and two hammers are on your left.", + "E": "One hammer is on your left, and two hammers are behind you instead of on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 17 + ] + } +] diff --git a/QAs/5d01d5e3-ea8b-4d41-9d80-024c3c429095-mcq.json b/QAs/5d01d5e3-ea8b-4d41-9d80-024c3c429095-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4167c78128709ef5b86fe6d65c773824b24eb26c --- /dev/null +++ b/QAs/5d01d5e3-ea8b-4d41-9d80-024c3c429095-mcq.json @@ -0,0 +1,810 @@ +[ + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_3", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put these condiments back in their place, where should I put them?", + "answer": "The original position of these condiments is above the oven.", + "question_chinese": "如果我想将这些调料归位，请问我该把他们放到哪里呢？", + "answer_chinese": "这些调料原来的位置是烤箱上方。", + "evidence": [ + { + "description": "You took out the condiments from the cabinet above the oven.", + "timestamps": { + "start": "00:10:35", + "end": "00:11:23" + } + } + ], + "question_timestamps": "00:12:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put them below the sink.", + "B": "Put them in the fridge side drawer.", + "C": "The original position of these condiments is above the oven.", + "D": "The original position of these condiments is next to the toaster on the counter.", + "E": "The original position of these condiments is behind the coffee maker on the shelf." + }, + "answer_label": "C", + "answer_text": "The original position of these condiments is above the oven.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_3", + "correct_answer": "The original position of these condiments is above the oven.", + "original_options": { + "A": "Put them below the sink.", + "B": "Put them in the fridge side drawer.", + "C": "The original position of these condiments is above the oven.", + "D": "The original position of these condiments is next to the toaster on the counter.", + "E": "The original position of these condiments is behind the coffee maker on the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 14, + 14 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put these condiments back in their place, where should I put them?", + "answer": "You should return the condiments to the cabinet on top of the oven and return the cheese to the fridge.", + "question_chinese": "如果我想将这些调料归位，请问我该把他们放到哪里呢？", + "answer_chinese": "你应该将调料放回烤箱上的柜子，将芝士放回冰箱里。", + "evidence": [ + { + "description": "You took out the condiments from the cabinet above the oven.", + "timestamps": { + "start": "00:10:35", + "end": "00:11:23" + } + }, + { + "description": "You took cheese from the refrigerator.", + "timestamps": { + "start": "00:14:47", + "end": "00:15:38" + } + } + ], + "question_timestamps": "00:15:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Put condiments above the sink; cheese in fridge.", + "B": "You should return the condiments to the cabinet on top of the oven and return the cheese to the fridge.", + "C": "Put condiments by the stove; cheese in fridge.", + "D": "Store the condiments in the cabinet above the refrigerator and keep the cheese in the fridge.", + "E": "Place the condiments on the shelf above the microwave, then return the cheese to the fridge." + }, + "answer_label": "B", + "answer_text": "You should return the condiments to the cabinet on top of the oven and return the cheese to the fridge.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_4", + "correct_answer": "You should return the condiments to the cabinet on top of the oven and return the cheese to the fridge.", + "original_options": { + "A": "Put condiments above the sink; cheese in fridge.", + "B": "You should return the condiments to the cabinet on top of the oven and return the cheese to the fridge.", + "C": "Put condiments by the stove; cheese in fridge.", + "D": "Store the condiments in the cabinet above the refrigerator and keep the cheese in the fridge.", + "E": "Place the condiments on the shelf above the microwave, then return the cheese to the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 16 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_5", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to return the fork to its place, what route should I take?", + "answer": "You should turn around backwards and open the first cabinet drawer beneath the sink.", + "question_chinese": "如果我想放回叉子，我的移动路径应该是怎么样的呢？", + "answer_chinese": "你应该转身向后，打开洗手台下的第一层柜子。", + "evidence": [ + { + "description": "You took out a fork from the cabinet under the sink.", + "timestamps": { + "start": "00:17:29", + "end": "00:17:40" + } + } + ], + "question_timestamps": "00:17:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn around and open the second drawer.", + "B": "Turn left and open the first drawer.", + "C": "Turn around and open the top cabinet above the sink, not the drawer beneath it.", + "D": "You should turn around backwards and open the first cabinet drawer beneath the sink.", + "E": "Turn around backward, but open the second cabinet drawer beneath the sink instead of the first." + }, + "answer_label": "D", + "answer_text": "You should turn around backwards and open the first cabinet drawer beneath the sink.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_5", + "correct_answer": "You should turn around backwards and open the first cabinet drawer beneath the sink.", + "original_options": { + "A": "Turn around and open the second drawer.", + "B": "Turn left and open the first drawer.", + "C": "Turn around and open the top cabinet above the sink, not the drawer beneath it.", + "D": "You should turn around backwards and open the first cabinet drawer beneath the sink.", + "E": "Turn around backward, but open the second cabinet drawer beneath the sink instead of the first." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_6", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to return the fork to its place, what route should I take?", + "answer": "You should turn backwards toward the sink to grab the fork, then pick up the fork, and finally turn around to place the fork in the drawer.", + "question_chinese": "如果我想放回叉子，我的移动路径应该是怎么样的呢？", + "answer_chinese": "你应该向后转去洗手台中拿叉子，然后拿起叉子，最后转身放到叉子到抽屉里。", + "evidence": [ + { + "description": "You took out a fork from the cabinet under the sink.", + "timestamps": { + "start": "00:17:29", + "end": "00:17:40" + } + }, + { + "description": "You placed the fork in the sink.", + "timestamps": { + "start": "00:18:50", + "end": "00:19:00" + } + } + ], + "question_timestamps": "00:21:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You should turn backwards toward the sink to grab the fork, then pick up the fork, and finally turn around to place the fork in the drawer.", + "B": "Turn forward to the sink, then place it.", + "C": "Grab it from the counter, then drawer.", + "D": "Turn backward toward the sink, grab the fork, then place it in the cabinet instead of the drawer.", + "E": "Turn backward to the sink to pick up the fork, then turn left to put it away in a different drawer." + }, + "answer_label": "A", + "answer_text": "You should turn backwards toward the sink to grab the fork, then pick up the fork, and finally turn around to place the fork in the drawer.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_6", + "correct_answer": "You should turn backwards toward the sink to grab the fork, then pick up the fork, and finally turn around to place the fork in the drawer.", + "original_options": { + "A": "You should turn backwards toward the sink to grab the fork, then pick up the fork, and finally turn around to place the fork in the drawer.", + "B": "Turn forward to the sink, then place it.", + "C": "Grab it from the counter, then drawer.", + "D": "Turn backward toward the sink, grab the fork, then place it in the cabinet instead of the drawer.", + "E": "Turn backward to the sink to pick up the fork, then turn left to put it away in a different drawer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 21 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many types of toys did I play with my pet in total?", + "answer": "You only played with one baseball with your pet.", + "question_chinese": "我一共和宠物玩了几种玩具呢？", + "answer_chinese": "玩了一种。", + "evidence": [ + { + "description": "You are playing fetch with the dog.", + "timestamps": { + "start": "00:00:26", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two toy types.", + "B": "You only played with one baseball with your pet.", + "C": "Three toy types.", + "D": "You played with two toy types total: a baseball and a frisbee.", + "E": "You played with three toy types overall: a baseball, a frisbee, and a rope toy." + }, + "answer_label": "B", + "answer_text": "You only played with one baseball with your pet.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_9", + "correct_answer": "You only played with one baseball with your pet.", + "original_options": { + "A": "Two toy types.", + "B": "You only played with one baseball with your pet.", + "C": "Three toy types.", + "D": "You played with two toy types total: a baseball and a frisbee.", + "E": "You played with three toy types overall: a baseball, a frisbee, and a rope toy." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 15 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_10", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many types of toys did I play with my pet in total?", + "answer": "Two types: baseball and frisbee.", + "question_chinese": "我一共和宠物玩了几种玩具呢？", + "answer_chinese": "两种：棒球和飞盘。", + "evidence": [ + { + "description": "You are playing fetch with the dog.", + "timestamps": { + "start": "00:00:26", + "end": "00:01:29" + } + }, + { + "description": "You are playing frisbee with the dog.", + "timestamps": { + "start": "00:03:30", + "end": "00:04:04" + } + } + ], + "question_timestamps": "00:05:09", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two types: baseball and frisbee.", + "B": "One type: frisbee only.", + "C": "Four types total.", + "D": "You played with one toy type overall: a baseball, and no frisbee was used.", + "E": "You played with three toy types overall: baseball, frisbee, and a rope toy." + }, + "answer_label": "A", + "answer_text": "Two types: baseball and frisbee.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_10", + "correct_answer": "Two types: baseball and frisbee.", + "original_options": { + "A": "Two types: baseball and frisbee.", + "B": "One type: frisbee only.", + "C": "Four types total.", + "D": "You played with one toy type overall: a baseball, and no frisbee was used.", + "E": "You played with three toy types overall: baseball, frisbee, and a rope toy." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 13 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_8", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Can I reach the frisbee that I play with my dog?", + "answer": "No, you cannot reach it.", + "question_chinese": "我和狗狗玩的飞盘我能够伸手拿到吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The frisbee was thrown by you to the edge of the coffee table in the living room.", + "timestamps": { + "start": "00:03:38", + "end": "00:03:44" + } + } + ], + "question_timestamps": "00:03:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you cannot reach it.", + "B": "Yes, you can reach it." + }, + "answer_label": "A", + "answer_text": "No, you cannot reach it.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_8", + "correct_answer": "No, you cannot reach it.", + "original_options": { + "A": "No, you cannot reach it.", + "B": "Yes, you can reach it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_7", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Can I reach the frisbee that I play with my dog?", + "answer": "Yes, you can reach it.", + "question_chinese": "我和狗狗玩的飞盘我能够伸手拿到吗？", + "answer_chinese": "你可以伸手拿到。", + "evidence": [ + { + "description": "The red frisbee is under the coffee table.", + "timestamps": { + "start": "00:26:00", + "end": "00:26:07" + } + }, + { + "description": "You have also moved near the coffee table.", + "timestamps": { + "start": "00:26:08", + "end": "00:26:09" + } + } + ], + "question_timestamps": "00:26:15", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you cannot reach it.", + "B": "Yes, you can reach it." + }, + "answer_label": "B", + "answer_text": "Yes, you can reach it.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_7", + "correct_answer": "Yes, you can reach it.", + "original_options": { + "A": "No, you cannot reach it.", + "B": "Yes, you can reach it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_14", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I go to find my seasoned potatoes?", + "answer": "You can find the seasoned potatoes above the oven.", + "question_chinese": "我能去哪里找到我调料后的土豆？", + "answer_chinese": "你能在烤箱上方找到调料后的土豆。", + "evidence": [ + { + "description": "The seasoned potatoes are all above the oven.", + "timestamps": { + "start": "00:17:53", + "end": "00:18:48" + } + } + ], + "question_timestamps": "00:18:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Below the oven.", + "B": "You can find the seasoned potatoes above the oven.", + "C": "Beside the oven.", + "D": "You can find the seasoned potatoes inside the oven rather than stored above it.", + "E": "You can find the seasoned potatoes in a lower drawer under the counter, not above the oven." + }, + "answer_label": "B", + "answer_text": "You can find the seasoned potatoes above the oven.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_14", + "correct_answer": "You can find the seasoned potatoes above the oven.", + "original_options": { + "A": "Below the oven.", + "B": "You can find the seasoned potatoes above the oven.", + "C": "Beside the oven.", + "D": "You can find the seasoned potatoes inside the oven rather than stored above it.", + "E": "You can find the seasoned potatoes in a lower drawer under the counter, not above the oven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 17 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_13", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I go to find my seasoned potatoes?", + "answer": "The seasoned potatoes are in the oven.", + "question_chinese": "我能去哪里找到我调料后的土豆？", + "answer_chinese": "调料之后的土豆在烤箱里。", + "evidence": [ + { + "description": "The seasoned potatoes were placed in the oven.", + "timestamps": { + "start": "00:21:56", + "end": "00:22:07" + } + } + ], + "question_timestamps": "00:24:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the refrigerator.", + "B": "On the countertop.", + "C": "The seasoned potatoes are under the counter in a cabinet, not inside the oven.", + "D": "The seasoned potatoes are in the oven.", + "E": "The seasoned potatoes are behind the microwave on the counter rather than being cooked in the oven." + }, + "answer_label": "D", + "answer_text": "The seasoned potatoes are in the oven.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_13", + "correct_answer": "The seasoned potatoes are in the oven.", + "original_options": { + "A": "In the refrigerator.", + "B": "On the countertop.", + "C": "The seasoned potatoes are under the counter in a cabinet, not inside the oven.", + "D": "The seasoned potatoes are in the oven.", + "E": "The seasoned potatoes are behind the microwave on the counter rather than being cooked in the oven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 17 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_11", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of seasoning have I used so far?", + "answer": "You have now used three bottles of seasoning.", + "question_chinese": "目前我都用了多少瓶调料呢？", + "answer_chinese": "你现在用了三瓶调料。", + "evidence": [ + { + "description": "You used the first bottle of seasoning.", + "timestamps": { + "start": "00:11:55", + "end": "00:12:06" + } + }, + { + "description": "You used the second bottle of seasoning.", + "timestamps": { + "start": "00:12:26", + "end": "00:12:29" + } + }, + { + "description": "You used the third bottle of seasoning.", + "timestamps": { + "start": "00:12:32", + "end": "00:12:35" + } + } + ], + "question_timestamps": "00:12:42", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have used two bottles.", + "B": "You have used four bottles.", + "C": "You have used five bottles of seasoning so far, counting each bottle you already opened.", + "D": "You have used one bottle of seasoning so far, since you only opened a single container.", + "E": "You have now used three bottles of seasoning." + }, + "answer_label": "E", + "answer_text": "You have now used three bottles of seasoning.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_11", + "correct_answer": "You have now used three bottles of seasoning.", + "original_options": { + "A": "You have used two bottles.", + "B": "You have used four bottles.", + "C": "You have used five bottles of seasoning so far, counting each bottle you already opened.", + "D": "You have used one bottle of seasoning so far, since you only opened a single container.", + "E": "You have now used three bottles of seasoning." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_12", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of seasoning have I used so far?", + "answer": "You have used six bottles of seasoning.", + "question_chinese": "目前我都用了多少瓶调料呢？", + "answer_chinese": "你用了六瓶调料。", + "evidence": [ + { + "description": "You used the first bottle of seasoning.", + "timestamps": { + "start": "00:11:55", + "end": "00:12:06" + } + }, + { + "description": "You used the third bottle of seasoning.", + "timestamps": { + "start": "00:12:32", + "end": "00:12:35" + } + }, + { + "description": "You used the second bottle of seasoning.", + "timestamps": { + "start": "00:12:26", + "end": "00:12:29" + } + }, + { + "description": "You used the sixth bottle of seasoning.", + "timestamps": { + "start": "00:13:11", + "end": "00:13:21" + } + }, + { + "description": "You used the fourth bottle of seasoning.", + "timestamps": { + "start": "00:12:44", + "end": "00:12:48" + } + }, + { + "description": "You used the fifth bottle of seasoning.", + "timestamps": { + "start": "00:12:53", + "end": "00:13:11" + } + } + ], + "question_timestamps": "00:14:02", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have used five bottles.", + "B": "You have used seven bottles.", + "C": "You have used four bottles of seasoning so far, based on the containers you have already handled.", + "D": "You have used six bottles of seasoning.", + "E": "You have used eight bottles of seasoning so far, since you have gone through two more than six." + }, + "answer_label": "D", + "answer_text": "You have used six bottles of seasoning.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_12", + "correct_answer": "You have used six bottles of seasoning.", + "original_options": { + "A": "You have used five bottles.", + "B": "You have used seven bottles.", + "C": "You have used four bottles of seasoning so far, based on the containers you have already handled.", + "D": "You have used six bottles of seasoning.", + "E": "You have used eight bottles of seasoning so far, since you have gone through two more than six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are the foods kept in the house?", + "answer": "In the storage room that you enter by turning left, going straight, and turning left again.", + "question_chinese": "食物都放在屋子的哪里呢？", + "answer_chinese": "在你左转，直走再左转进入的储藏室里。", + "evidence": [ + { + "description": "You took out potatoes from the storage room.", + "timestamps": { + "start": "00:04:49", + "end": "00:05:24" + } + } + ], + "question_timestamps": "00:05:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the storage room after two rights.", + "B": "In the storage room after one left.", + "C": "In the storage room you reach by turning left, going straight, then turning right at the next corner.", + "D": "In the storage room you enter by turning right first, walking straight ahead, and then turning left again.", + "E": "In the storage room that you enter by turning left, going straight, and turning left again." + }, + "answer_label": "E", + "answer_text": "In the storage room that you enter by turning left, going straight, and turning left again.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_1", + "correct_answer": "In the storage room that you enter by turning left, going straight, and turning left again.", + "original_options": { + "A": "In the storage room after two rights.", + "B": "In the storage room after one left.", + "C": "In the storage room you reach by turning left, going straight, then turning right at the next corner.", + "D": "In the storage room you enter by turning right first, walking straight ahead, and then turning left again.", + "E": "In the storage room that you enter by turning left, going straight, and turning left again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095", + "video_length": "00:30:00", + "qid": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are the foods kept in the house?", + "answer": "The food is kept in the pantry and refrigerator.", + "question_chinese": "食物都放在屋子的哪里呢？", + "answer_chinese": "食物放在储藏室和冰箱里。", + "evidence": [ + { + "description": "You took out potatoes from the storage room.", + "timestamps": { + "start": "00:04:49", + "end": "00:05:24" + } + }, + { + "description": "There is also food in the refrigerator.", + "timestamps": { + "start": "00:14:12", + "end": "00:14:31" + } + } + ], + "question_timestamps": "00:18:17", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The food is kept in the pantry and refrigerator.", + "B": "The food is kept in the pantry.", + "C": "The food is kept in the refrigerator.", + "D": "The food is kept in the pantry and in the cupboard, rather than being stored in the refrigerator.", + "E": "The food is kept in the refrigerator and on the counter, with extra items left out instead of in the pantry." + }, + "answer_label": "A", + "answer_text": "The food is kept in the pantry and refrigerator.", + "q_id": "5d01d5e3-ea8b-4d41-9d80-024c3c429095_2", + "correct_answer": "The food is kept in the pantry and refrigerator.", + "original_options": { + "A": "The food is kept in the pantry and refrigerator.", + "B": "The food is kept in the pantry.", + "C": "The food is kept in the refrigerator.", + "D": "The food is kept in the pantry and in the cupboard, rather than being stored in the refrigerator.", + "E": "The food is kept in the refrigerator and on the counter, with extra items left out instead of in the pantry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 21 + ] + } +] diff --git a/QAs/5d7eb356-9b7d-4c9b-b963-4604f92953d1-mcq.json b/QAs/5d7eb356-9b7d-4c9b-b963-4604f92953d1-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d36237b44004e9da70ca25148dda5812b101950e --- /dev/null +++ b/QAs/5d7eb356-9b7d-4c9b-b963-4604f92953d1-mcq.json @@ -0,0 +1,1180 @@ +[ + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_7", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Just now, while I was using the press, could I reach the ceramic plant on the shelf?", + "answer": "You can get it; the ceramic potted plant is on the shelf to the right of the press, within reach.", + "question_chinese": "刚刚我在使用压力机时，我能拿到架子上的陶瓷盆栽吗？", + "answer_chinese": "你能拿到，陶瓷盆栽就在压力机右边架子上，伸手就能够到。", + "evidence": [ + { + "description": "The ceramic potted plant is on the shelf to the right of the press.", + "timestamps": { + "start": "00:22:55", + "end": "00:23:06" + } + }, + { + "description": "The ceramic potted plant is on the shelf to the right of the press.", + "timestamps": { + "start": "00:40:34", + "end": "00:40:42" + } + }, + { + "description": "The ceramic potted plant is on the shelf to the right of the press.", + "timestamps": { + "start": "00:50:06", + "end": "00:50:26" + } + } + ], + "question_timestamps": "00:52:59", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "You can get it; the ceramic potted plant is on the shelf to the right of the press, within reach.", + "B": "No, you can't reach it; the ceramic plant is too far away on the shelf while you're using the press." + }, + "answer_label": "A", + "answer_text": "You can get it; the ceramic potted plant is on the shelf to the right of the press, within reach.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_7", + "correct_answer": "You can get it; the ceramic potted plant is on the shelf to the right of the press, within reach.", + "original_options": { + "A": "You can get it; the ceramic potted plant is on the shelf to the right of the press, within reach.", + "B": "No, you can't reach it; the ceramic plant is too far away on the shelf while you're using the press." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_8", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Just now, while I was using the press, could I reach the ceramic plant on the shelf?", + "answer": "You can get it; the ceramic potted plant is on the shelf to the right of the press, within reach.", + "question_chinese": "刚刚我在使用压力机时，我能拿到架子上的陶瓷盆栽吗？", + "answer_chinese": "你能拿到，陶瓷盆栽就在压力机右边架子上，伸手就能够到。", + "evidence": [ + { + "description": "The ceramic potted plant is on the shelf to the right of the press.", + "timestamps": { + "start": "00:50:06", + "end": "00:50:26" + } + }, + { + "description": "The ceramic potted plant is on the shelf to the right of the press.", + "timestamps": { + "start": "00:40:34", + "end": "00:40:43" + } + }, + { + "description": "The ceramic potted plant is on the shelf to the right of the press.", + "timestamps": { + "start": "00:22:55", + "end": "00:23:06" + } + } + ], + "question_timestamps": "00:54:36", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "No, you can't reach it; the ceramic potted plant is too far from the press to grab while using it.", + "B": "You can get it; the ceramic potted plant is on the shelf to the right of the press, within reach." + }, + "answer_label": "B", + "answer_text": "You can get it; the ceramic potted plant is on the shelf to the right of the press, within reach.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_8", + "correct_answer": "You can get it; the ceramic potted plant is on the shelf to the right of the press, within reach.", + "original_options": { + "A": "No, you can't reach it; the ceramic potted plant is too far from the press to grab while using it.", + "B": "You can get it; the ceramic potted plant is on the shelf to the right of the press, within reach." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_6", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the scraper when I was just organizing the shapes of the small items made with clay?", + "answer": "You can grab it; the scraper is on the table, to your right.", + "question_chinese": "刚刚我在整理用陶泥制作的小物件的形状时，我能拿到刮刀吗？", + "answer_chinese": "你能拿到，挂刀就在桌上，你的右手边。", + "evidence": [ + { + "description": "The scraper is on the table, to my right.", + "timestamps": { + "start": "00:33:50", + "end": "00:34:50" + } + }, + { + "description": "The scraper is on the table, to my right.", + "timestamps": { + "start": "00:37:35", + "end": "00:39:15" + } + } + ], + "question_timestamps": "00:44:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You can grab it on the left.", + "B": "You can grab it; the scraper is on the table, to your right.", + "C": "You can grab it from the shelf.", + "D": "You can grab it; the scraper is on the table, but it sits to your left instead of your right.", + "E": "You can grab it; the scraper is on the shelf above the workspace, not on the table beside you." + }, + "answer_label": "B", + "answer_text": "You can grab it; the scraper is on the table, to your right.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_6", + "correct_answer": "You can grab it; the scraper is on the table, to your right.", + "original_options": { + "A": "You can grab it on the left.", + "B": "You can grab it; the scraper is on the table, to your right.", + "C": "You can grab it from the shelf.", + "D": "You can grab it; the scraper is on the table, but it sits to your left instead of your right.", + "E": "You can grab it; the scraper is on the shelf above the workspace, not on the table beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 19 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_5", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the scraper when I was just organizing the shapes of the small items made with clay?", + "answer": "You can grab it; the scraper is on the table, to your right.", + "question_chinese": "刚刚我在整理用陶泥制作的小物件的形状时，我能拿到刮刀吗？", + "answer_chinese": "你能拿到，刮刀就在桌上，你的右手边。", + "evidence": [ + { + "description": "The scraper is on the table, to my right.", + "timestamps": { + "start": "00:33:50", + "end": "00:34:50" + } + }, + { + "description": "The scraper is on the table, to my right.", + "timestamps": { + "start": "00:37:35", + "end": "00:39:15" + } + } + ], + "question_timestamps": "00:50:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You can grab it from the drawer.", + "B": "You can grab it on the floor.", + "C": "You can grab it; the scraper is on the table, to your right.", + "D": "You can grab it; the scraper is in the drawer to your left, not sitting on the table to your right.", + "E": "You can grab it; the scraper is on the floor to your left, rather than on the tabletop at your right." + }, + "answer_label": "C", + "answer_text": "You can grab it; the scraper is on the table, to your right.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_5", + "correct_answer": "You can grab it; the scraper is on the table, to your right.", + "original_options": { + "A": "You can grab it from the drawer.", + "B": "You can grab it on the floor.", + "C": "You can grab it; the scraper is on the table, to your right.", + "D": "You can grab it; the scraper is in the drawer to your left, not sitting on the table to your right.", + "E": "You can grab it; the scraper is on the floor to your left, rather than on the tabletop at your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 21 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_7_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many dogs did I just shut into the pen?", + "answer": "You just put 1 dog into the fenced enclosure.", + "question_chinese": "我刚刚关了几条狗进围栏？", + "answer_chinese": "你刚刚关了1条狗进围栏。", + "evidence": [ + { + "description": "I just put two dogs into the pen.", + "timestamps": { + "start": "00:31:05", + "end": "00:31:10" + } + } + ], + "question_timestamps": "00:31:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just put 1 dog into the fenced enclosure.", + "B": "You put 0 dogs in the pen.", + "C": "You put 2 dogs in the pen.", + "D": "You just shut 3 dogs into the fenced enclosure before stepping back from the gate.", + "E": "You just put 4 dogs into the fenced enclosure, closing the gate after the last one entered." + }, + "answer_label": "A", + "answer_text": "You just put 1 dog into the fenced enclosure.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_7_T2", + "correct_answer": "You just put 1 dog into the fenced enclosure.", + "original_options": { + "A": "You just put 1 dog into the fenced enclosure.", + "B": "You put 0 dogs in the pen.", + "C": "You put 2 dogs in the pen.", + "D": "You just shut 3 dogs into the fenced enclosure before stepping back from the gate.", + "E": "You just put 4 dogs into the fenced enclosure, closing the gate after the last one entered." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 17 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_7_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many dogs did I just shut into the pen?", + "answer": "You just put 2 dogs into the fenced enclosure.", + "question_chinese": "我刚刚关了几条狗进围栏？", + "answer_chinese": "你刚刚关了2条狗进围栏。", + "evidence": [ + { + "description": "I just put two dogs into the pen.", + "timestamps": { + "start": "00:31:05", + "end": "00:31:13" + } + } + ], + "question_timestamps": "00:49:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You put 1 dog in the pen.", + "B": "You put 3 dogs in the pen.", + "C": "You just put 0 dogs into the fenced enclosure, leaving the pen empty after shutting it.", + "D": "You just shut 4 dogs into the fenced enclosure, then latched the gate once they were inside.", + "E": "You just put 2 dogs into the fenced enclosure." + }, + "answer_label": "E", + "answer_text": "You just put 2 dogs into the fenced enclosure.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_7_T1", + "correct_answer": "You just put 2 dogs into the fenced enclosure.", + "original_options": { + "A": "You put 1 dog in the pen.", + "B": "You put 3 dogs in the pen.", + "C": "You just put 0 dogs into the fenced enclosure, leaving the pen empty after shutting it.", + "D": "You just shut 4 dogs into the fenced enclosure, then latched the gate once they were inside.", + "E": "You just put 2 dogs into the fenced enclosure." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_3", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the utility knife I just saw?", + "answer": "The utility knife you just saw is on the shelf to the right of the press.", + "question_chinese": "我刚刚看到的美工刀在哪里？", + "answer_chinese": "你刚刚看到的美工刀在压力机右边的架子上。", + "evidence": [ + { + "description": "The utility knife is on the shelf to the right of the press.", + "timestamps": { + "start": "00:03:35", + "end": "00:04:08" + } + }, + { + "description": "The utility knife is on the shelf to the right of the press.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:41" + } + }, + { + "description": "The utility knife you just saw is on the shelf to the right of the press.", + "timestamps": { + "start": "00:04:50", + "end": "00:05:15" + } + } + ], + "question_timestamps": "00:17:41", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "On the shelf left of the press.", + "B": "On the workbench surface.", + "C": "The utility knife you just saw is on the shelf below the press, near the front edge.", + "D": "The utility knife you just saw is on the shelf above the press, slightly toward the back.", + "E": "The utility knife you just saw is on the shelf to the right of the press." + }, + "answer_label": "E", + "answer_text": "The utility knife you just saw is on the shelf to the right of the press.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_3", + "correct_answer": "The utility knife you just saw is on the shelf to the right of the press.", + "original_options": { + "A": "On the shelf left of the press.", + "B": "On the workbench surface.", + "C": "The utility knife you just saw is on the shelf below the press, near the front edge.", + "D": "The utility knife you just saw is on the shelf above the press, slightly toward the back.", + "E": "The utility knife you just saw is on the shelf to the right of the press." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 4, + 17, + 17 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_4", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the utility knife I just saw?", + "answer": "The utility knife you just saw is on the shelf to the right of the press.", + "question_chinese": "我刚刚看到的美工刀在哪里？", + "answer_chinese": "你刚刚看到的美工刀在压力机右边的架子上。", + "evidence": [ + { + "description": "The utility knife you just saw is on the shelf to the right of the press.", + "timestamps": { + "start": "00:03:35", + "end": "00:04:08" + } + }, + { + "description": "The utility knife you just saw is on the shelf to the right of the press.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:41" + } + }, + { + "description": "The utility knife you just saw is on the shelf to the right of the press.", + "timestamps": { + "start": "00:04:50", + "end": "00:05:15" + } + } + ], + "question_timestamps": "00:31:11", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "On the shelf left of the press.", + "B": "On the shelf above the press.", + "C": "The utility knife you just saw is on the shelf below the press, near the front corner.", + "D": "The utility knife you just saw is on the shelf to the right of the press.", + "E": "The utility knife you just saw is on the shelf to the left of the press, but on the upper tier." + }, + "answer_label": "D", + "answer_text": "The utility knife you just saw is on the shelf to the right of the press.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_4", + "correct_answer": "The utility knife you just saw is on the shelf to the right of the press.", + "original_options": { + "A": "On the shelf left of the press.", + "B": "On the shelf above the press.", + "C": "The utility knife you just saw is on the shelf below the press, near the front corner.", + "D": "The utility knife you just saw is on the shelf to the right of the press.", + "E": "The utility knife you just saw is on the shelf to the left of the press, but on the upper tier." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 21 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_16", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many scrapers were there on the table while I was tidying the table up?", + "answer": "There were two scrapers on the table while you were tidying the table up.", + "question_chinese": "我在收拾桌子的时候，桌上有几把刮刀？", + "answer_chinese": "你在收拾桌子的时候，桌上有2把刮刀。", + "evidence": [ + { + "description": "There were two scrapers on the table.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:35" + } + }, + { + "description": "There were two scrapers on the table.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:13:35", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There was 1 scraper on the table.", + "B": "There were 3 scrapers on the table.", + "C": "There were four scrapers on the table while you were tidying it up, with multiple ones grouped together.", + "D": "There were five scrapers on the table while you were tidying the table up, spread across different spots.", + "E": "There were two scrapers on the table while you were tidying the table up." + }, + "answer_label": "E", + "answer_text": "There were two scrapers on the table while you were tidying the table up.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_16", + "correct_answer": "There were two scrapers on the table while you were tidying the table up.", + "original_options": { + "A": "There was 1 scraper on the table.", + "B": "There were 3 scrapers on the table.", + "C": "There were four scrapers on the table while you were tidying it up, with multiple ones grouped together.", + "D": "There were five scrapers on the table while you were tidying the table up, spread across different spots.", + "E": "There were two scrapers on the table while you were tidying the table up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_15", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many scrapers were there on the table while I was tidying the table up?", + "answer": "There were two scrapers on the table while you're tidying the table up.", + "question_chinese": "我在收拾桌子的时候，桌上有几把刮刀？", + "answer_chinese": "我在收拾桌子的时候，桌上有2把刮刀。", + "evidence": [ + { + "description": "There were two scrapers on the table.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:15" + } + }, + { + "description": "There were two scrapers on the table.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:27:45", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There was 1 scraper on the table.", + "B": "There were 3 scrapers on the table.", + "C": "There were four scrapers on the table while you're tidying the table up, not just a pair.", + "D": "There were two scrapers on the table while you're tidying the table up.", + "E": "There were five scrapers on the table while you're tidying the table up, with extras left out." + }, + "answer_label": "D", + "answer_text": "There were two scrapers on the table while you're tidying the table up.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_15", + "correct_answer": "There were two scrapers on the table while you're tidying the table up.", + "original_options": { + "A": "There was 1 scraper on the table.", + "B": "There were 3 scrapers on the table.", + "C": "There were four scrapers on the table while you're tidying the table up, not just a pair.", + "D": "There were two scrapers on the table while you're tidying the table up.", + "E": "There were five scrapers on the table while you're tidying the table up, with extras left out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_2_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the utility knife I saw?", + "answer": "The utility knife you saw is on the left side of the third shelf to your right.", + "question_chinese": "我看到的美工刀在哪里？", + "answer_chinese": "你看到的美工刀在你的右边第三层架子上的左侧。", + "evidence": [ + { + "description": "The utility knife is on the shelf to the right of the press.", + "timestamps": { + "start": "00:03:35", + "end": "00:04:08" + } + }, + { + "description": "The utility knife is on the shelf to the right of the press.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:41" + } + }, + { + "description": "The utility knife you just saw is on the shelf to the right of the press.", + "timestamps": { + "start": "00:04:50", + "end": "00:05:15" + } + } + ], + "question_timestamps": "00:17:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the third shelf to your left.", + "B": "The utility knife you saw is on the left side of the third shelf to your right.", + "C": "On the right side of that shelf.", + "D": "The utility knife you saw is on the left side of the second shelf to your right, not the third.", + "E": "The utility knife you saw is on the left side of the third shelf behind you, not to your right." + }, + "answer_label": "B", + "answer_text": "The utility knife you saw is on the left side of the third shelf to your right.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_2_T1", + "correct_answer": "The utility knife you saw is on the left side of the third shelf to your right.", + "original_options": { + "A": "On the third shelf to your left.", + "B": "The utility knife you saw is on the left side of the third shelf to your right.", + "C": "On the right side of that shelf.", + "D": "The utility knife you saw is on the left side of the second shelf to your right, not the third.", + "E": "The utility knife you saw is on the left side of the third shelf behind you, not to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 20 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_2_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the utility knife I saw?", + "answer": "The utility knife you saw is on the left side of the third shelf to your left.", + "question_chinese": "我看到的美工刀在哪里？", + "answer_chinese": "你看到的美工刀在你的左边第三层架子上的左侧。", + "evidence": [ + { + "description": "The utility knife you just saw is on the shelf to the right of the press.", + "timestamps": { + "start": "00:03:35", + "end": "00:04:08" + } + }, + { + "description": "The utility knife you just saw is on the shelf to the right of the press.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:41" + } + }, + { + "description": "The utility knife you just saw is on the shelf to the right of the press.", + "timestamps": { + "start": "00:04:50", + "end": "00:05:15" + } + }, + { + "description": "The shelf is on your left.", + "timestamps": { + "start": "00:38:25", + "end": "00:38:47" + } + } + ], + "question_timestamps": "00:39:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the third shelf to your right.", + "B": "The utility knife you saw is on the left side of the third shelf to your left.", + "C": "On the right side of that shelf.", + "D": "The utility knife you saw is on the left side of the second shelf to your left, not the third.", + "E": "The utility knife you saw is on the left side of the third shelf behind you rather than to your left." + }, + "answer_label": "B", + "answer_text": "The utility knife you saw is on the left side of the third shelf to your left.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_2_T2", + "correct_answer": "The utility knife you saw is on the left side of the third shelf to your left.", + "original_options": { + "A": "On the third shelf to your right.", + "B": "The utility knife you saw is on the left side of the third shelf to your left.", + "C": "On the right side of that shelf.", + "D": "The utility knife you saw is on the left side of the second shelf to your left, not the third.", + "E": "The utility knife you saw is on the left side of the third shelf behind you rather than to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 21 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_1_T2", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "The first time I saw the bucket move from the table to the floor, what path did it take?", + "answer": "The bucket was moved from on top of the table, passed below the table, and then set down on the ground to your right; now the table is behind you.", + "question_chinese": "我第一次看到桶从桌子到地上，经历了什么路径？", + "answer_chinese": "桶从桌子上经过桌子下方被放到右边的地上，现在桌子在你后面。", + "evidence": [ + { + "description": "The bucket was moved from on top of the table to the floor on the right, passing under the table.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:26" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Now the table is in front of you.", + "B": "Now the table is to your left.", + "C": "The bucket went from the table, under it, to the floor on your right; the table ended in front.", + "D": "The bucket was moved from on top of the table, passed below the table, and then set down on the ground to your right; now the table is behind you.", + "E": "It went from the tabletop, below the table, to the floor on your right; afterward the table was on your left." + }, + "answer_label": "D", + "answer_text": "The bucket was moved from on top of the table, passed below the table, and then set down on the ground to your right; now the table is behind you.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_1_T2", + "correct_answer": "The bucket was moved from on top of the table, passed below the table, and then set down on the ground to your right; now the table is behind you.", + "original_options": { + "A": "Now the table is in front of you.", + "B": "Now the table is to your left.", + "C": "The bucket went from the table, under it, to the floor on your right; the table ended in front.", + "D": "The bucket was moved from on top of the table, passed below the table, and then set down on the ground to your right; now the table is behind you.", + "E": "It went from the tabletop, below the table, to the floor on your right; afterward the table was on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 21 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_1_T1", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "The first time I saw the bucket move from the table to the floor, what path did it take?", + "answer": "The bucket was moved from on top of the table, passed below the table, and then set down on the ground to your right; now the table is in front of you.", + "question_chinese": "我第一次看到桶从桌子到地上，经历了什么路径？", + "answer_chinese": "桶从桌子上经过桌子下方被放到右边的地上，现在桌子在你前面。", + "evidence": [ + { + "description": "The bucket was moved from on top of the table to the floor on the right, passing under the table.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:26" + } + } + ], + "question_timestamps": "00:29:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Now the table is behind you.", + "B": "Now the table is to your right.", + "C": "The bucket moved from the tabletop, passed beneath the table, landed to your right, and the table ended up behind you.", + "D": "From on the table it went below it to the ground on your right, leaving the table to your left.", + "E": "The bucket was moved from on top of the table, passed below the table, and then set down on the ground to your right; now the table is in front of you." + }, + "answer_label": "E", + "answer_text": "The bucket was moved from on top of the table, passed below the table, and then set down on the ground to your right; now the table is in front of you.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_1_T1", + "correct_answer": "The bucket was moved from on top of the table, passed below the table, and then set down on the ground to your right; now the table is in front of you.", + "original_options": { + "A": "Now the table is behind you.", + "B": "Now the table is to your right.", + "C": "The bucket moved from the tabletop, passed beneath the table, landed to your right, and the table ended up behind you.", + "D": "From on the table it went below it to the ground on your right, leaving the table to your left.", + "E": "The bucket was moved from on top of the table, passed below the table, and then set down on the ground to your right; now the table is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 20 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_4_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the ceramic potted plant on the shelf?", + "answer": "You can’t reach the ceramic potted plant on the shelf.", + "question_chinese": "我能拿到架子上的陶瓷盆栽吗？", + "answer_chinese": "你不能拿到架子上的陶瓷盆栽。", + "evidence": [ + { + "description": "The ceramic potted plant is on the shelf to the right of the press.", + "timestamps": { + "start": "00:22:55", + "end": "00:23:06" + } + }, + { + "description": "The ceramic potted plant is on the shelf to the right of the press.The shelf is three steps away from you.", + "timestamps": { + "start": "00:40:34", + "end": "00:40:45" + } + } + ], + "question_timestamps": "00:46:47", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can reach it. The ceramic potted plant is on the shelf to the right of the press; it’s within arm’s reach.", + "B": "You can’t reach the ceramic potted plant on the shelf." + }, + "answer_label": "B", + "answer_text": "You can’t reach the ceramic potted plant on the shelf.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_4_T1", + "correct_answer": "You can’t reach the ceramic potted plant on the shelf.", + "original_options": { + "A": "Yes—you can reach it. The ceramic potted plant is on the shelf to the right of the press; it’s within arm’s reach.", + "B": "You can’t reach the ceramic potted plant on the shelf." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 25 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_4_T2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the ceramic potted plant on the shelf?", + "answer": "Yes—you can reach it. The ceramic potted plant is on the shelf to the right of the press; it’s within arm’s reach.", + "question_chinese": "我能拿到架子上的陶瓷盆栽吗？", + "answer_chinese": "你能拿到，陶瓷盆栽就在压力机右边架子上，伸手就能够到。", + "evidence": [ + { + "description": "The ceramic potted plant is on the shelf to the right of the press.", + "timestamps": { + "start": "00:50:06", + "end": "00:50:26" + } + }, + { + "description": "The ceramic potted plant is on the shelf to the right of the press.", + "timestamps": { + "start": "00:40:34", + "end": "00:40:43" + } + }, + { + "description": "The ceramic potted plant is on the shelf to the right of the press.", + "timestamps": { + "start": "00:22:55", + "end": "00:23:06" + } + } + ], + "question_timestamps": "00:54:36", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can reach it. The ceramic potted plant is on the shelf to the right of the press; it’s within arm’s reach.", + "B": "You can’t reach the ceramic potted plant on the shelf." + }, + "answer_label": "A", + "answer_text": "Yes—you can reach it. The ceramic potted plant is on the shelf to the right of the press; it’s within arm’s reach.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_4_T2", + "correct_answer": "Yes—you can reach it. The ceramic potted plant is on the shelf to the right of the press; it’s within arm’s reach.", + "original_options": { + "A": "Yes—you can reach it. The ceramic potted plant is on the shelf to the right of the press; it’s within arm’s reach.", + "B": "You can’t reach the ceramic potted plant on the shelf." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_6_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I passed by the doorway, how many vehicles did I see there?", + "answer": "You just saw 4 cars by the doorway.", + "question_chinese": "我路过门口的时候，看到门口有几台车？", + "answer_chinese": "你刚刚看到了门口有4台车。", + "evidence": [ + { + "description": "There are three cars at the door.", + "timestamps": { + "start": "00:31:00", + "end": "00:31:03" + } + }, + { + "description": "There are three cars at the door.", + "timestamps": { + "start": "00:31:15", + "end": "00:31:19" + } + } + ], + "question_timestamps": "00:31:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just saw 3 cars there.", + "B": "You just saw 5 cars there.", + "C": "You just saw 2 cars by the doorway as you passed it.", + "D": "You just saw 6 cars gathered near the doorway when you walked by.", + "E": "You just saw 4 cars by the doorway." + }, + "answer_label": "E", + "answer_text": "You just saw 4 cars by the doorway.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_6_T2", + "correct_answer": "You just saw 4 cars by the doorway.", + "original_options": { + "A": "You just saw 3 cars there.", + "B": "You just saw 5 cars there.", + "C": "You just saw 2 cars by the doorway as you passed it.", + "D": "You just saw 6 cars gathered near the doorway when you walked by.", + "E": "You just saw 4 cars by the doorway." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 13 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_6_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I passed by the doorway, how many vehicles did I see there?", + "answer": "You just saw 3 cars by the doorway.", + "question_chinese": "我路过门口的时候，看到门口有几台车？", + "answer_chinese": "你刚刚看到了门口有3台车。", + "evidence": [ + { + "description": "There are three cars at the door.", + "timestamps": { + "start": "00:31:00", + "end": "00:31:03" + } + }, + { + "description": "There are three cars at the door.", + "timestamps": { + "start": "00:31:15", + "end": "00:31:19" + } + } + ], + "question_timestamps": "00:43:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just saw 2 cars there.", + "B": "You just saw 4 cars there.", + "C": "You just saw 1 car by the doorway when you passed by.", + "D": "You just saw 3 cars by the doorway.", + "E": "You just saw 5 cars parked by the doorway as you walked past." + }, + "answer_label": "D", + "answer_text": "You just saw 3 cars by the doorway.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_6_T1", + "correct_answer": "You just saw 3 cars by the doorway.", + "original_options": { + "A": "You just saw 2 cars there.", + "B": "You just saw 4 cars there.", + "C": "You just saw 1 car by the doorway when you passed by.", + "D": "You just saw 3 cars by the doorway.", + "E": "You just saw 5 cars parked by the doorway as you walked past." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 13 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_9", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the painting?", + "answer": "The painting was in the room with pink wallpaper, to the right of the clock.", + "question_chinese": "挂画在哪个位置？", + "answer_chinese": "挂画在贴有粉色墙纸的房间，时钟的右边。", + "evidence": [ + { + "description": "The painting in the room with pink wallpaper, to the right of the clock.", + "timestamps": { + "start": "00:30:51", + "end": "00:30:58" + } + } + ], + "question_timestamps": "00:37:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left of the clock.", + "B": "Near the door.", + "C": "The painting was in the room with pink wallpaper, positioned to the left of the clock.", + "D": "The painting was in the room with pink wallpaper, hanging above the clock on the same wall.", + "E": "The painting was in the room with pink wallpaper, to the right of the clock." + }, + "answer_label": "E", + "answer_text": "The painting was in the room with pink wallpaper, to the right of the clock.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_9", + "correct_answer": "The painting was in the room with pink wallpaper, to the right of the clock.", + "original_options": { + "A": "Left of the clock.", + "B": "Near the door.", + "C": "The painting was in the room with pink wallpaper, positioned to the left of the clock.", + "D": "The painting was in the room with pink wallpaper, hanging above the clock on the same wall.", + "E": "The painting was in the room with pink wallpaper, to the right of the clock." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 17 + ] + }, + { + "video_uid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1", + "video_length": "00:56:15", + "qid": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_10", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the painting?", + "answer": "The painting was in the room with pink wallpaper, to the right of the clock.", + "question_chinese": "挂画在哪个位置？", + "answer_chinese": "挂画在贴有粉色墙纸的房间，时钟的右边。", + "evidence": [ + { + "description": "The painting was in the room with pink wallpaper, to the right of the clock.", + "timestamps": { + "start": "00:30:51", + "end": "00:30:58" + } + } + ], + "question_timestamps": "00:44:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left of the clock.", + "B": "In the blue wallpaper room.", + "C": "The painting was in the room with pink wallpaper, to the right of the clock.", + "D": "The painting was in the room with pink wallpaper, but it was to the left of the clock.", + "E": "The painting was in the room with green wallpaper, placed near the window rather than by the clock." + }, + "answer_label": "C", + "answer_text": "The painting was in the room with pink wallpaper, to the right of the clock.", + "q_id": "5d7eb356-9b7d-4c9b-b963-4604f92953d1_10", + "correct_answer": "The painting was in the room with pink wallpaper, to the right of the clock.", + "original_options": { + "A": "Left of the clock.", + "B": "In the blue wallpaper room.", + "C": "The painting was in the room with pink wallpaper, to the right of the clock.", + "D": "The painting was in the room with pink wallpaper, but it was to the left of the clock.", + "E": "The painting was in the room with green wallpaper, placed near the window rather than by the clock." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 18 + ] + } +] diff --git a/QAs/5d9261fb-7d80-4cb0-b601-38c17fbf6f75-mcq.json b/QAs/5d9261fb-7d80-4cb0-b601-38c17fbf6f75-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..93bc1f4e3d4ca40df038f105dd109e87d3feeead --- /dev/null +++ b/QAs/5d9261fb-7d80-4cb0-b601-38c17fbf6f75-mcq.json @@ -0,0 +1,962 @@ +[ + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_3_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the child standing relative to me?", + "answer": "In front of you.", + "question_chinese": "小孩站在我的哪里？", + "answer_chinese": "你的前方。", + "evidence": [ + { + "description": "My left side", + "timestamps": { + "start": "00:33:22", + "end": "00:33:23" + } + } + ], + "question_timestamps": "00:33:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "In front of you.", + "D": "The child was standing slightly behind you, not in front of you.", + "E": "The child was positioned to your right side when you looked over." + }, + "answer_label": "C", + "answer_text": "In front of you.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_3_T2", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "In front of you.", + "D": "The child was standing slightly behind you, not in front of you.", + "E": "The child was positioned to your right side when you looked over." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 12, + 12 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_3_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the child standing relative to me?", + "answer": "Directly to your left.", + "question_chinese": "小孩站在我的哪里？", + "answer_chinese": "我的正左边", + "evidence": [ + { + "description": "My left side", + "timestamps": { + "start": "00:33:22", + "end": "00:33:23" + } + } + ], + "question_timestamps": "00:33:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly to your right.", + "B": "Directly behind you.", + "C": "The child was directly in front of you rather than standing to your left.", + "D": "Directly to your left.", + "E": "The child was slightly behind you on the right side, not on your left." + }, + "answer_label": "D", + "answer_text": "Directly to your left.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_3_T1", + "correct_answer": "Directly to your left.", + "original_options": { + "A": "Directly to your right.", + "B": "Directly behind you.", + "C": "The child was directly in front of you rather than standing to your left.", + "D": "Directly to your left.", + "E": "The child was slightly behind you on the right side, not on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 14 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_8_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many kinds of ingredients did I prepare?", + "answer": "Three types: melon, spices, and chili peppers. They’re all in front of you.", + "question_chinese": "我处理了多少种食材？", + "answer_chinese": "三种，分别是瓜，香料和辣椒，他们都在你的正前方。", + "evidence": [ + { + "description": "melons", + "timestamps": { + "start": "00:06:37", + "end": "00:06:38" + } + }, + { + "description": "spices", + "timestamps": { + "start": "00:15:28", + "end": "00:15:29" + } + }, + { + "description": "chilies", + "timestamps": { + "start": "00:48:16", + "end": "00:48:17" + } + } + ], + "question_timestamps": "00:48:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two types, in front of you.", + "B": "Four types, in front of you.", + "C": "Two types: melon and spices; they’re both placed in front of you.", + "D": "Four types: melon, spices, chili peppers, and another ingredient, all laid out in front of you.", + "E": "Three types: melon, spices, and chili peppers. They’re all in front of you." + }, + "answer_label": "E", + "answer_text": "Three types: melon, spices, and chili peppers. They’re all in front of you.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_8_T2", + "correct_answer": "Three types: melon, spices, and chili peppers. They’re all in front of you.", + "original_options": { + "A": "Two types, in front of you.", + "B": "Four types, in front of you.", + "C": "Two types: melon and spices; they’re both placed in front of you.", + "D": "Four types: melon, spices, chili peppers, and another ingredient, all laid out in front of you.", + "E": "Three types: melon, spices, and chili peppers. They’re all in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 16 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_8_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many kinds of ingredients did I prepare?", + "answer": "Three types: melon, spices, and chili peppers, behind you.", + "question_chinese": "我处理了多少种食材？", + "answer_chinese": "三种，分别是瓜，香料和辣椒，在你的后方", + "evidence": [ + { + "description": "melons", + "timestamps": { + "start": "00:06:37", + "end": "00:06:38" + } + }, + { + "description": "spices", + "timestamps": { + "start": "00:15:28", + "end": "00:15:29" + } + }, + { + "description": "chilies", + "timestamps": { + "start": "00:48:16", + "end": "00:48:17" + } + } + ], + "question_timestamps": "00:49:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two types, behind you.", + "B": "Four types, behind you.", + "C": "Two types: melon and chili peppers, and they’re both behind you.", + "D": "Three types: melon, spices, and chili peppers, behind you.", + "E": "Five types: melon, spices, chili peppers, plus two more ingredients, all located behind you." + }, + "answer_label": "D", + "answer_text": "Three types: melon, spices, and chili peppers, behind you.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_8_T1", + "correct_answer": "Three types: melon, spices, and chili peppers, behind you.", + "original_options": { + "A": "Two types, behind you.", + "B": "Four types, behind you.", + "C": "Two types: melon and chili peppers, and they’re both behind you.", + "D": "Three types: melon, spices, and chili peppers, behind you.", + "E": "Five types: melon, spices, chili peppers, plus two more ingredients, all located behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_6_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I add firewood?", + "answer": "Three times; you put it into the stove to your right.", + "question_chinese": "我放过多少次柴火？", + "answer_chinese": "3次，放在了你右边的炉子里。", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:49:16", + "end": "00:49:17" + } + }, + { + "description": "the second time", + "timestamps": { + "start": "00:49:22", + "end": "00:49:23" + } + }, + { + "description": "the third time", + "timestamps": { + "start": "00:49:29", + "end": "00:49:30" + } + } + ], + "question_timestamps": "00:49:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once; into the stove on your right.", + "B": "Twice; into the stove on your right.", + "C": "Three times; you put it into the stove to your right.", + "D": "Four times; each time you placed the firewood into the stove to your right.", + "E": "Five times; you kept feeding firewood into the stove positioned to your right." + }, + "answer_label": "C", + "answer_text": "Three times; you put it into the stove to your right.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_6_T2", + "correct_answer": "Three times; you put it into the stove to your right.", + "original_options": { + "A": "Once; into the stove on your right.", + "B": "Twice; into the stove on your right.", + "C": "Three times; you put it into the stove to your right.", + "D": "Four times; each time you placed the firewood into the stove to your right.", + "E": "Five times; you kept feeding firewood into the stove positioned to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 13 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_6_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I add firewood?", + "answer": "Three times, into the stove to your back-right.", + "question_chinese": "我放过多少次柴火？", + "answer_chinese": "3次，在你右后方的炉子里。", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:49:16", + "end": "00:49:17" + } + }, + { + "description": "the second time", + "timestamps": { + "start": "00:49:22", + "end": "00:49:23" + } + }, + { + "description": "the third time", + "timestamps": { + "start": "00:49:29", + "end": "00:49:30" + } + } + ], + "question_timestamps": "00:50:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once, into the stove back-right.", + "B": "Three times, into the stove to your back-right.", + "C": "Twice, into the stove back-right.", + "D": "Four times, you added firewood repeatedly into the stove to your back-right.", + "E": "Five times, you kept putting firewood into the stove located to your back-right." + }, + "answer_label": "B", + "answer_text": "Three times, into the stove to your back-right.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_6_T1", + "correct_answer": "Three times, into the stove to your back-right.", + "original_options": { + "A": "Once, into the stove back-right.", + "B": "Three times, into the stove to your back-right.", + "C": "Twice, into the stove back-right.", + "D": "Four times, you added firewood repeatedly into the stove to your back-right.", + "E": "Five times, you kept putting firewood into the stove located to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_7_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I make the winter melon dish?", + "answer": "You chopped first, and then stir-fried in the pot to your front-right.", + "question_chinese": "我是怎么做冬瓜这道菜的？", + "answer_chinese": "先切，然后在你右前方的锅里炒", + "evidence": [ + { + "description": "cut", + "timestamps": { + "start": "00:06:46", + "end": "00:06:47" + } + }, + { + "description": "stir-fry", + "timestamps": { + "start": "01:03:07", + "end": "01:03:08" + } + } + ], + "question_timestamps": "01:03:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You stir-fried first, then chopped.", + "B": "You chopped, then boiled it.", + "C": "You chopped first, and then stir-fried in the pot to your front-right.", + "D": "You chopped first, then stir-fried in the pot to your front-left instead of front-right.", + "E": "You chopped first, then stir-fried, but you used the pot behind you rather than the one front-right." + }, + "answer_label": "C", + "answer_text": "You chopped first, and then stir-fried in the pot to your front-right.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_7_T1", + "correct_answer": "You chopped first, and then stir-fried in the pot to your front-right.", + "original_options": { + "A": "You stir-fried first, then chopped.", + "B": "You chopped, then boiled it.", + "C": "You chopped first, and then stir-fried in the pot to your front-right.", + "D": "You chopped first, then stir-fried in the pot to your front-left instead of front-right.", + "E": "You chopped first, then stir-fried, but you used the pot behind you rather than the one front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 17 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_7_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I make the winter melon dish?", + "answer": "In the pot in front of you, you chopped first and then stir-fried.", + "question_chinese": "我是怎么做冬瓜这道菜的？", + "answer_chinese": "在你前面的锅里，先切，然后炒。", + "evidence": [ + { + "description": "cut", + "timestamps": { + "start": "00:06:46", + "end": "00:06:47" + } + }, + { + "description": "stir-fry", + "timestamps": { + "start": "01:03:07", + "end": "01:03:08" + } + } + ], + "question_timestamps": "01:03:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You stir-fried first, then chopped.", + "B": "You sliced first, then stir-fried.", + "C": "In the pot behind you, you chopped first and then stir-fried instead of using the pot in front.", + "D": "In the pot in front of you, you chopped first and then stir-fried.", + "E": "In the pot in front of you, you chopped after stir-frying, reversing the order of the steps." + }, + "answer_label": "D", + "answer_text": "In the pot in front of you, you chopped first and then stir-fried.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_7_T2", + "correct_answer": "In the pot in front of you, you chopped first and then stir-fried.", + "original_options": { + "A": "You stir-fried first, then chopped.", + "B": "You sliced first, then stir-fried.", + "C": "In the pot behind you, you chopped first and then stir-fried instead of using the pot in front.", + "D": "In the pot in front of you, you chopped first and then stir-fried.", + "E": "In the pot in front of you, you chopped after stir-frying, reversing the order of the steps." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_9_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I rinse the rice?", + "answer": "First you picked up the rice jar, then you poured water into the bucket to your right.", + "question_chinese": "我是怎么淘米的？", + "answer_chinese": "先拿米缸，后倒水到我右边的水桶", + "evidence": [ + { + "description": "get the rice container", + "timestamps": { + "start": "00:56:49", + "end": "00:56:50" + } + }, + { + "description": "pour water to rinse the rice", + "timestamps": { + "start": "00:57:02", + "end": "00:57:03" + } + } + ], + "question_timestamps": "00:57:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First you picked up the rice jar, then you poured water into the bucket to your right.", + "B": "You poured water into the left bucket.", + "C": "You poured water into the front bucket.", + "D": "First you picked up the rice jar, then you poured water into the bucket in front of you.", + "E": "First you put down the rice jar, then you poured water into the bucket to your right." + }, + "answer_label": "A", + "answer_text": "First you picked up the rice jar, then you poured water into the bucket to your right.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_9_T2", + "correct_answer": "First you picked up the rice jar, then you poured water into the bucket to your right.", + "original_options": { + "A": "First you picked up the rice jar, then you poured water into the bucket to your right.", + "B": "You poured water into the left bucket.", + "C": "You poured water into the front bucket.", + "D": "First you picked up the rice jar, then you poured water into the bucket in front of you.", + "E": "First you put down the rice jar, then you poured water into the bucket to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 17 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_9_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I rinse the rice?", + "answer": "First you picked up the rice jar, then you poured water to rinse the rice. The rice is to your right.", + "question_chinese": "我是怎么淘米的？", + "answer_chinese": "先拿米缸，后倒水淘米。米在你的右边。", + "evidence": [ + { + "description": "get the rice container", + "timestamps": { + "start": "00:56:49", + "end": "00:56:50" + } + }, + { + "description": "pour water to rinse the rice", + "timestamps": { + "start": "00:57:02", + "end": "00:57:03" + } + } + ], + "question_timestamps": "00:58:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The rice is to your left.", + "B": "The rice is in front of you.", + "C": "First you set down the rice jar, then you poured water to rinse the rice, with the rice to your right.", + "D": "First you picked up the rice jar, then you poured water into the bucket to your right instead of rinsing directly.", + "E": "First you picked up the rice jar, then you poured water to rinse the rice. The rice is to your right." + }, + "answer_label": "E", + "answer_text": "First you picked up the rice jar, then you poured water to rinse the rice. The rice is to your right.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_9_T1", + "correct_answer": "First you picked up the rice jar, then you poured water to rinse the rice. The rice is to your right.", + "original_options": { + "A": "The rice is to your left.", + "B": "The rice is in front of you.", + "C": "First you set down the rice jar, then you poured water to rinse the rice, with the rice to your right.", + "D": "First you picked up the rice jar, then you poured water into the bucket to your right instead of rinsing directly.", + "E": "First you picked up the rice jar, then you poured water to rinse the rice. The rice is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 21 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_4_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pairs of slippers did I see?", + "answer": "Two, to your left.", + "question_chinese": "我看到有多少双拖鞋？", + "answer_chinese": "2，在你的左边", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:10:16", + "end": "00:10:17" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:19:47", + "end": "00:19:48" + } + } + ], + "question_timestamps": "00:24:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, to your left.", + "B": "One, to your left.", + "C": "Three, to your left.", + "D": "Four pairs of slippers were on your left side near where you were standing.", + "E": "You saw no slippers at all; nothing was on your left side." + }, + "answer_label": "A", + "answer_text": "Two, to your left.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_4_T2", + "correct_answer": "Two, to your left.", + "original_options": { + "A": "Two, to your left.", + "B": "One, to your left.", + "C": "Three, to your left.", + "D": "Four pairs of slippers were on your left side near where you were standing.", + "E": "You saw no slippers at all; nothing was on your left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 12 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_4_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pairs of slippers did I see?", + "answer": "Two, on your right-hand side.", + "question_chinese": "我看到有多少双拖鞋？", + "answer_chinese": "2，在你的右手边", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:10:16", + "end": "00:10:17" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:19:47", + "end": "00:19:48" + } + } + ], + "question_timestamps": "00:32:02", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, on your right-hand side.", + "B": "Three, on your right-hand side.", + "C": "Four pairs of slippers were visible on your right-hand side next to you.", + "D": "Zero pairs of slippers were there; you didn’t see any on your right-hand side.", + "E": "Two, on your right-hand side." + }, + "answer_label": "E", + "answer_text": "Two, on your right-hand side.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_4_T1", + "correct_answer": "Two, on your right-hand side.", + "original_options": { + "A": "One, on your right-hand side.", + "B": "Three, on your right-hand side.", + "C": "Four pairs of slippers were visible on your right-hand side next to you.", + "D": "Zero pairs of slippers were there; you didn’t see any on your right-hand side.", + "E": "Two, on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 15 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_5_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see stored in the bucket?", + "answer": "Water and cilantro, to your left.", + "question_chinese": "我看到桶装过什么？", + "answer_chinese": "水和香菜，在你的左边。", + "evidence": [ + { + "description": "Water", + "timestamps": { + "start": "00:56:53", + "end": "00:56:54" + } + }, + { + "description": "cilantro", + "timestamps": { + "start": "01:01:45", + "end": "01:01:46" + } + } + ], + "question_timestamps": "01:02:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Water and cilantro, in front.", + "B": "Water and cilantro, to your right.", + "C": "Water and cilantro, to your left.", + "D": "You saw water and cilantro stored in the bucket positioned directly in front of you.", + "E": "You saw water and cilantro in the bucket located behind you rather than to your left." + }, + "answer_label": "C", + "answer_text": "Water and cilantro, to your left.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_5_T2", + "correct_answer": "Water and cilantro, to your left.", + "original_options": { + "A": "Water and cilantro, in front.", + "B": "Water and cilantro, to your right.", + "C": "Water and cilantro, to your left.", + "D": "You saw water and cilantro stored in the bucket positioned directly in front of you.", + "E": "You saw water and cilantro in the bucket located behind you rather than to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_5_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see stored in the bucket?", + "answer": "Water and cilantro, to your back-left.", + "question_chinese": "我看到桶装过什么？", + "answer_chinese": "水和香菜，在你的左后方。", + "evidence": [ + { + "description": "Water", + "timestamps": { + "start": "00:56:53", + "end": "00:56:54" + } + }, + { + "description": "cilantro", + "timestamps": { + "start": "01:01:45", + "end": "01:01:46" + } + } + ], + "question_timestamps": "01:02:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Water and cilantro, to your back-left.", + "B": "Water and cilantro, to your back-right.", + "C": "Water and cilantro, to your front-left.", + "D": "The bucket held water and cilantro and was positioned at your front-right side.", + "E": "You saw water and cilantro in the bucket on your left side, not at your back-left." + }, + "answer_label": "A", + "answer_text": "Water and cilantro, to your back-left.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_5_T1", + "correct_answer": "Water and cilantro, to your back-left.", + "original_options": { + "A": "Water and cilantro, to your back-left.", + "B": "Water and cilantro, to your back-right.", + "C": "Water and cilantro, to your front-left.", + "D": "The bucket held water and cilantro and was positioned at your front-right side.", + "E": "You saw water and cilantro in the bucket on your left side, not at your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 16 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_1_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where were the blue slippers relative to me?", + "answer": "It is to the left of the bowl, to your right.", + "question_chinese": "蓝色拖鞋在我的哪里？", + "answer_chinese": "在盆的左边，你的右边", + "evidence": [ + { + "description": "To the left of the basin", + "timestamps": { + "start": "00:10:16", + "end": "00:10:17" + } + } + ], + "question_timestamps": "00:10:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To the right of the bowl.", + "B": "To your left of the bowl.", + "C": "It is to the left of the bowl, to your right.", + "D": "It is to the right of the bowl, and the slippers are positioned to your right.", + "E": "It is to the left of the bowl, but the slippers are located to your left instead of your right." + }, + "answer_label": "C", + "answer_text": "It is to the left of the bowl, to your right.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_1_T2", + "correct_answer": "It is to the left of the bowl, to your right.", + "original_options": { + "A": "To the right of the bowl.", + "B": "To your left of the bowl.", + "C": "It is to the left of the bowl, to your right.", + "D": "It is to the right of the bowl, and the slippers are positioned to your right.", + "E": "It is to the left of the bowl, but the slippers are located to your left instead of your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 20 + ] + }, + { + "video_uid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75", + "video_length": "01:14:57", + "qid": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_1_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where were the blue slippers relative to me?", + "answer": "It is to the left of the bowl, to your back-left.", + "question_chinese": "蓝色拖鞋在我的哪里？", + "answer_chinese": "在盆的左边，你的左后方。", + "evidence": [ + { + "description": "To the left of the basin", + "timestamps": { + "start": "00:10:16", + "end": "00:10:17" + } + } + ], + "question_timestamps": "00:11:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-right.", + "B": "It is to the left of the bowl, to your back-left.", + "C": "To your front-left.", + "D": "It is to the right of the bowl, and the slippers are positioned to your back-left.", + "E": "It is to the left of the bowl, but the slippers are located to your right rather than back-left." + }, + "answer_label": "B", + "answer_text": "It is to the left of the bowl, to your back-left.", + "q_id": "5d9261fb-7d80-4cb0-b601-38c17fbf6f75_1_T1", + "correct_answer": "It is to the left of the bowl, to your back-left.", + "original_options": { + "A": "To your back-right.", + "B": "It is to the left of the bowl, to your back-left.", + "C": "To your front-left.", + "D": "It is to the right of the bowl, and the slippers are positioned to your back-left.", + "E": "It is to the left of the bowl, but the slippers are located to your right rather than back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 19 + ] + } +] diff --git a/QAs/5dab6f94-5ca7-4985-bcd0-67f4fc04687a-mcq.json b/QAs/5dab6f94-5ca7-4985-bcd0-67f4fc04687a-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d1d27cf2a4b1a207053318cedebd7d4c780eb10d --- /dev/null +++ b/QAs/5dab6f94-5ca7-4985-bcd0-67f4fc04687a-mcq.json @@ -0,0 +1,1661 @@ +[ + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_12", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Was I cooking with two pots at the same time just now?", + "answer": "Yes.", + "question_chinese": "我刚才是同时料理两个锅吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:43:38", + "end": "00:43:39" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:44:23", + "end": "00:44:24" + } + } + ], + "question_timestamps": "00:44:25", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_12", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_11", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Was I cooking with two pots at the same time just now?", + "answer": "Yes.", + "question_chinese": "我刚才是同时料理两个锅吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:43:38", + "end": "00:43:39" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:44:23", + "end": "00:43:24" + } + } + ], + "question_timestamps": "00:44:55", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_11", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_7_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many different ingredients did I add into the pot?", + "answer": "You added one ingredient into the pot, and the pot was behind you.", + "question_chinese": "我加了多少种食材进锅？", + "answer_chinese": "你加了一种食材进锅，锅在你身后。", + "evidence": [ + { + "description": "You add an ingredient into the pot.", + "timestamps": { + "start": "00:44:21", + "end": "00:44:25" + } + } + ], + "question_timestamps": "00:44:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero ingredients.", + "B": "Two ingredients.", + "C": "You added three different ingredients into the pot, and the pot was behind you the whole time.", + "D": "You added four different ingredients into the pot, and the pot was behind you as you added them.", + "E": "You added one ingredient into the pot, and the pot was behind you." + }, + "answer_label": "E", + "answer_text": "You added one ingredient into the pot, and the pot was behind you.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_7_T1", + "correct_answer": "You added one ingredient into the pot, and the pot was behind you.", + "original_options": { + "A": "Zero ingredients.", + "B": "Two ingredients.", + "C": "You added three different ingredients into the pot, and the pot was behind you the whole time.", + "D": "You added four different ingredients into the pot, and the pot was behind you as you added them.", + "E": "You added one ingredient into the pot, and the pot was behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_16", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many ingredients did I add to the pot?", + "answer": "two.", + "question_chinese": "我加了多少种食材进锅？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:44:23", + "end": "00:44:24" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:45:01", + "end": "00:45:02" + } + } + ], + "question_timestamps": "00:45:03", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "two.", + "B": "One.", + "C": "Three.", + "D": "You added four ingredients to the pot in total during that step.", + "E": "You added six ingredients to the pot altogether, not just a couple." + }, + "answer_label": "A", + "answer_text": "two.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_16", + "correct_answer": "two.", + "original_options": { + "A": "two.", + "B": "One.", + "C": "Three.", + "D": "You added four ingredients to the pot in total during that step.", + "E": "You added six ingredients to the pot altogether, not just a couple." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_7_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many different ingredients did I add into the pot?", + "answer": "You added two ingredients into the pot, and the pot was by your right hand.", + "question_chinese": "我加了多少种食材进锅？", + "answer_chinese": "你加了两种食材进锅，锅在你右手边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:44:23", + "end": "00:44:26" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:45:01", + "end": "00:45:03" + } + } + ], + "question_timestamps": "00:45:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You added two ingredients into the pot, and the pot was by your right hand.", + "B": "Zero ingredients.", + "C": "One ingredient.", + "D": "You added three ingredients into the pot, and the pot was by your right hand throughout.", + "E": "You added four different ingredients into the pot, and the pot stayed by your right hand." + }, + "answer_label": "A", + "answer_text": "You added two ingredients into the pot, and the pot was by your right hand.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_7_T2", + "correct_answer": "You added two ingredients into the pot, and the pot was by your right hand.", + "original_options": { + "A": "You added two ingredients into the pot, and the pot was by your right hand.", + "B": "Zero ingredients.", + "C": "One ingredient.", + "D": "You added three ingredients into the pot, and the pot was by your right hand throughout.", + "E": "You added four different ingredients into the pot, and the pot stayed by your right hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_15", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many ingredients did I add to the pot?", + "answer": "Two.", + "question_chinese": "我加了多少种食材进锅？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:44:23", + "end": "00:44:24" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:45:01", + "end": "00:45:02" + } + } + ], + "question_timestamps": "00:45:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You added four ingredients to the pot, more than just a couple.", + "E": "You added five ingredients to the pot in total during that moment." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_15", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You added four ingredients to the pot, more than just a couple.", + "E": "You added five ingredients to the pot in total during that moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_2_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What am I doing?", + "answer": "You were washing rice, and the basin for washing rice was in front of you.", + "question_chinese": "我在干什么？", + "answer_chinese": "你在洗米，洗米的盆在你前面", + "evidence": [ + { + "description": "Rice", + "timestamps": { + "start": "00:00:22", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Washing rice, basin behind.", + "B": "Washing rice, basin left.", + "C": "You were washing rice, and the basin for washing rice was to your right as you worked.", + "D": "You were washing rice, and the basin for washing rice was in front of you.", + "E": "You were washing rice, and the basin for washing rice was above you rather than in front." + }, + "answer_label": "D", + "answer_text": "You were washing rice, and the basin for washing rice was in front of you.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_2_T2", + "correct_answer": "You were washing rice, and the basin for washing rice was in front of you.", + "original_options": { + "A": "Washing rice, basin behind.", + "B": "Washing rice, basin left.", + "C": "You were washing rice, and the basin for washing rice was to your right as you worked.", + "D": "You were washing rice, and the basin for washing rice was in front of you.", + "E": "You were washing rice, and the basin for washing rice was above you rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_2_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What am I doing?", + "answer": "You were sweeping the floor, and the broom was in your hand.", + "question_chinese": "我在干什么？", + "answer_chinese": "你在扫地，扫把在你手上。", + "evidence": [ + { + "description": "Sweeping the floor", + "timestamps": { + "start": "00:28:11", + "end": "00:29:10" + } + } + ], + "question_timestamps": "00:29:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were washing rice.", + "B": "You were mopping the floor.", + "C": "You were sweeping the floor, and the broom was in your hand.", + "D": "You were sweeping the floor, but the broom was leaning nearby instead of being in your hand.", + "E": "You were sweeping the floor, and you had set the broom down rather than holding it." + }, + "answer_label": "C", + "answer_text": "You were sweeping the floor, and the broom was in your hand.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_2_T1", + "correct_answer": "You were sweeping the floor, and the broom was in your hand.", + "original_options": { + "A": "You were washing rice.", + "B": "You were mopping the floor.", + "C": "You were sweeping the floor, and the broom was in your hand.", + "D": "You were sweeping the floor, but the broom was leaning nearby instead of being in your hand.", + "E": "You were sweeping the floor, and you had set the broom down rather than holding it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 16 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_4", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What am I doing?", + "answer": "Sweeping the floor.", + "question_chinese": "我在干什么？", + "answer_chinese": "扫地", + "evidence": [ + { + "description": "Sweeping the floor", + "timestamps": { + "start": "00:29:11", + "end": "00:29:12" + } + } + ], + "question_timestamps": "00:29:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Washing the car.", + "B": "Sweeping the floor.", + "C": "Cleaning the windows.", + "D": "Sweeping the stairs instead of sweeping the floor in that moment you are shown doing.", + "E": "Sweeping the garage rather than focusing on the floor area shown in the video." + }, + "answer_label": "B", + "answer_text": "Sweeping the floor.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_4", + "correct_answer": "Sweeping the floor.", + "original_options": { + "A": "Washing the car.", + "B": "Sweeping the floor.", + "C": "Cleaning the windows.", + "D": "Sweeping the stairs instead of sweeping the floor in that moment you are shown doing.", + "E": "Sweeping the garage rather than focusing on the floor area shown in the video." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 14 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I washing?", + "answer": "Rice.", + "question_chinese": "我在洗什么？", + "answer_chinese": "米", + "evidence": [ + { + "description": "Rice", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Rice.", + "B": "Quinoa.", + "C": "Lentils.", + "D": "You were washing pasta instead of washing rice at the sink in that moment.", + "E": "You were washing bowls rather than rinsing rice in the basin at the sink." + }, + "answer_label": "A", + "answer_text": "Rice.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_2", + "correct_answer": "Rice.", + "original_options": { + "A": "Rice.", + "B": "Quinoa.", + "C": "Lentils.", + "D": "You were washing pasta instead of washing rice at the sink in that moment.", + "E": "You were washing bowls rather than rinsing rice in the basin at the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_5", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I washing?", + "answer": "Bowls.", + "question_chinese": "我在洗什么？", + "answer_chinese": "碗", + "evidence": [ + { + "description": "Bowls", + "timestamps": { + "start": "00:09:41", + "end": "00:09:42" + } + } + ], + "question_timestamps": "00:09:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Plates.", + "B": "Glasses.", + "C": "You are washing a couple of spoons in the sink basin before setting them on the rack.", + "D": "Bowls.", + "E": "You are washing a set of trays, rinsing them off before stacking them to dry." + }, + "answer_label": "D", + "answer_text": "Bowls.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_5", + "correct_answer": "Bowls.", + "original_options": { + "A": "Plates.", + "B": "Glasses.", + "C": "You are washing a couple of spoons in the sink basin before setting them on the rack.", + "D": "Bowls.", + "E": "You are washing a set of trays, rinsing them off before stacking them to dry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 15 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_6", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I washing?", + "answer": "Bowls.", + "question_chinese": "我在洗什么？", + "answer_chinese": "碗", + "evidence": [ + { + "description": "Bowls", + "timestamps": { + "start": "00:09:41", + "end": "00:09:42" + } + } + ], + "question_timestamps": "00:09:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Bowls.", + "B": "Plates.", + "C": "Glasses.", + "D": "You are washing bottles, rinsing them out one by one under running water at the sink.", + "E": "You are washing jugs, scrubbing the inside before setting them aside to dry." + }, + "answer_label": "A", + "answer_text": "Bowls.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_6", + "correct_answer": "Bowls.", + "original_options": { + "A": "Bowls.", + "B": "Plates.", + "C": "Glasses.", + "D": "You are washing bottles, rinsing them out one by one under running water at the sink.", + "E": "You are washing jugs, scrubbing the inside before setting them aside to dry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 13 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_19", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What should I do if I want to eat coconut?", + "answer": "First crack open the coconut, pour out the coconut water, and then process the flesh.", + "question_chinese": "我想吃椰子怎么办？", + "answer_chinese": "先敲椰子，倒出椰子水，然后处理果肉", + "evidence": [ + { + "description": "crack open the coconut", + "timestamps": { + "start": "00:20:50", + "end": "00:20:51" + } + }, + { + "description": "pour out the coconut water", + "timestamps": { + "start": "00:20:56", + "end": "00:20:57" + } + }, + { + "description": "process the flesh.", + "timestamps": { + "start": "00:22:01", + "end": "00:22:02" + } + } + ], + "question_timestamps": "00:22:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Just drink the coconut water.", + "B": "Peel it and eat the shell.", + "C": "First crack open the coconut, pour out the coconut water, and then process the flesh.", + "D": "First slice the coconut open, keep the water inside, and then chew on the husk.", + "E": "First crack it open, pour the water into a bowl, then throw away the flesh afterward." + }, + "answer_label": "C", + "answer_text": "First crack open the coconut, pour out the coconut water, and then process the flesh.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_19", + "correct_answer": "First crack open the coconut, pour out the coconut water, and then process the flesh.", + "original_options": { + "A": "Just drink the coconut water.", + "B": "Peel it and eat the shell.", + "C": "First crack open the coconut, pour out the coconut water, and then process the flesh.", + "D": "First slice the coconut open, keep the water inside, and then chew on the husk.", + "E": "First crack it open, pour the water into a bowl, then throw away the flesh afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_20", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What should I do if I want to eat coconut?", + "answer": "First, crack open the coconut, pour out the coconut water, and then handle the flesh.", + "question_chinese": "我想吃椰子怎么办？", + "answer_chinese": "First crack open the coconut, pour out the coconut water, and then process the flesh.", + "evidence": [ + { + "description": "crack open the coconut", + "timestamps": { + "start": "00:20:50", + "end": "00:20:51" + } + }, + { + "description": "pour out the coconut water", + "timestamps": { + "start": "00:20:56", + "end": "00:20:57" + } + }, + { + "description": "process the flesh.", + "timestamps": { + "start": "00:22:01", + "end": "00:22:02" + } + } + ], + "question_timestamps": "00:22:45", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Discard the flesh afterward.", + "B": "Eat only the shell.", + "C": "First crack open the coconut, pour out the coconut water, and then throw the flesh away.", + "D": "First crack open the coconut, pour out the coconut water, and then toss the shell instead.", + "E": "First, crack open the coconut, pour out the coconut water, and then handle the flesh." + }, + "answer_label": "E", + "answer_text": "First, crack open the coconut, pour out the coconut water, and then handle the flesh.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_20", + "correct_answer": "First, crack open the coconut, pour out the coconut water, and then handle the flesh.", + "original_options": { + "A": "Discard the flesh afterward.", + "B": "Eat only the shell.", + "C": "First crack open the coconut, pour out the coconut water, and then throw the flesh away.", + "D": "First crack open the coconut, pour out the coconut water, and then toss the shell instead.", + "E": "First, crack open the coconut, pour out the coconut water, and then handle the flesh." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_6_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What do I have in my hand?", + "answer": "You had a jar in your hand, and the jar was in your left hand.", + "question_chinese": "我手里有什么？", + "answer_chinese": "你手里有一个罐子，罐子在你左手上", + "evidence": [ + { + "description": "You have a jar in your hand.", + "timestamps": { + "start": "00:11:13", + "end": "00:11:17" + } + } + ], + "question_timestamps": "00:11:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A bucket.", + "B": "A jar (right hand).", + "C": "You had a jar in your hand, and the jar was in your left hand.", + "D": "You had a jar in your hand, but it was held in your right hand instead.", + "E": "You had a bucket in your hand, and the bucket was positioned to your front-right." + }, + "answer_label": "C", + "answer_text": "You had a jar in your hand, and the jar was in your left hand.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_6_T1", + "correct_answer": "You had a jar in your hand, and the jar was in your left hand.", + "original_options": { + "A": "A bucket.", + "B": "A jar (right hand).", + "C": "You had a jar in your hand, and the jar was in your left hand.", + "D": "You had a jar in your hand, but it was held in your right hand instead.", + "E": "You had a bucket in your hand, and the bucket was positioned to your front-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 16, + 15 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_6_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What do I have in my hand?", + "answer": "You had a bucket in your hand, and the bucket was to your front-right.", + "question_chinese": "我手里有什么？", + "answer_chinese": "你手里有一个桶，桶在你右前方。", + "evidence": [ + { + "description": "You have a bucket in your hand.", + "timestamps": { + "start": "00:18:20", + "end": "00:18:24" + } + } + ], + "question_timestamps": "00:18:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You had a bucket in your hand, and the bucket was to your front-right.", + "B": "You had a jar in your hand.", + "C": "The bucket was to your front-left.", + "D": "You had a bucket in your hand, and the bucket was positioned to your back-right.", + "E": "You had a jar in your hand, and the jar was held in your left hand." + }, + "answer_label": "A", + "answer_text": "You had a bucket in your hand, and the bucket was to your front-right.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_6_T2", + "correct_answer": "You had a bucket in your hand, and the bucket was to your front-right.", + "original_options": { + "A": "You had a bucket in your hand, and the bucket was to your front-right.", + "B": "You had a jar in your hand.", + "C": "The bucket was to your front-left.", + "D": "You had a bucket in your hand, and the bucket was positioned to your back-right.", + "E": "You had a jar in your hand, and the jar was held in your left hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 16 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_17", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How do I make rice milk?", + "answer": "First rinse the rice, put the rice into the blender, and then start it.", + "question_chinese": "我是怎么做米浆的？", + "answer_chinese": "先淘米，把米放进破壁机，然后启动", + "evidence": [ + { + "description": "rinse the rice", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + }, + { + "description": "put the rice into the blender", + "timestamps": { + "start": "00:03:46", + "end": "00:03:47" + } + }, + { + "description": "start", + "timestamps": { + "start": "00:04:17", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:04:19", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "First rinse the rice, put the rice into the blender, and then start it.", + "B": "Soak the rice first.", + "C": "Use a food processor.", + "D": "First rinse the rice, put the rice into a pot, and then boil it on the stove.", + "E": "First soak the rice, put the rice into the blender, and then blend it until smooth." + }, + "answer_label": "A", + "answer_text": "First rinse the rice, put the rice into the blender, and then start it.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_17", + "correct_answer": "First rinse the rice, put the rice into the blender, and then start it.", + "original_options": { + "A": "First rinse the rice, put the rice into the blender, and then start it.", + "B": "Soak the rice first.", + "C": "Use a food processor.", + "D": "First rinse the rice, put the rice into a pot, and then boil it on the stove.", + "E": "First soak the rice, put the rice into the blender, and then blend it until smooth." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_18", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How do I make rice milk?", + "answer": "First rinse the rice, put the rice into the blender, and then start it.", + "question_chinese": "我是怎么做米浆的？", + "answer_chinese": "先淘米，把米放进破壁机，然后启动", + "evidence": [ + { + "description": "rinse the rice", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + }, + { + "description": "put the rice into the blender", + "timestamps": { + "start": "00:03:46", + "end": "00:03:47" + } + }, + { + "description": "start", + "timestamps": { + "start": "00:04:17", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:04:30", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Boil the rice in a pot.", + "B": "Serve the rice cold in bowls.", + "C": "First blend the rice, strain the mixture, and then refrigerate it for about an hour.", + "D": "First rinse the rice, put the rice into the blender, and then start it.", + "E": "First cook the rice, add spices and salt, and then let it cool down before serving." + }, + "answer_label": "D", + "answer_text": "First rinse the rice, put the rice into the blender, and then start it.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_18", + "correct_answer": "First rinse the rice, put the rice into the blender, and then start it.", + "original_options": { + "A": "Boil the rice in a pot.", + "B": "Serve the rice cold in bowls.", + "C": "First blend the rice, strain the mixture, and then refrigerate it for about an hour.", + "D": "First rinse the rice, put the rice into the blender, and then start it.", + "E": "First cook the rice, add spices and salt, and then let it cool down before serving." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_8_T2", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did I get back to the dining table?", + "answer": "Turn around, then turn left until you’re facing the doorway, and walk straight. The dining table is to your front-left.", + "question_chinese": "我是怎么回到餐桌？", + "answer_chinese": "你向后转弯左转面对门口，然后直走，餐桌在你的左前方。", + "evidence": [ + { + "description": "Turn around, turn left to face the door.", + "timestamps": { + "start": "00:01:38", + "end": "00:01:41" + } + }, + { + "description": "Then go straight and you'll reach the dining table", + "timestamps": { + "start": "00:01:41", + "end": "00:01:44" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around, then turn right and walk straight.", + "B": "Turn around, then turn left and walk straight.", + "C": "Turn around, then turn left until you’re facing the doorway, and walk straight. The dining table is to your front-left.", + "D": "Turn around, turn left until you face the doorway, then walk straight; the dining table will be front-right.", + "E": "Turn around, turn right until you’re facing the doorway, and walk straight; the dining table is to your front-left." + }, + "answer_label": "C", + "answer_text": "Turn around, then turn left until you’re facing the doorway, and walk straight. The dining table is to your front-left.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_8_T2", + "correct_answer": "Turn around, then turn left until you’re facing the doorway, and walk straight. The dining table is to your front-left.", + "original_options": { + "A": "Turn around, then turn right and walk straight.", + "B": "Turn around, then turn left and walk straight.", + "C": "Turn around, then turn left until you’re facing the doorway, and walk straight. The dining table is to your front-left.", + "D": "Turn around, turn left until you face the doorway, then walk straight; the dining table will be front-right.", + "E": "Turn around, turn right until you’re facing the doorway, and walk straight; the dining table is to your front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 20 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_8_T1", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did I get back to the dining table?", + "answer": "Just turn around and walk straight; the dining table is behind you.", + "question_chinese": "我是怎么回到餐桌？", + "answer_chinese": "你向后转直走即可，餐桌在你身后。", + "evidence": [ + { + "description": "Turn around and go straight.", + "timestamps": { + "start": "00:47:26", + "end": "00:47:28" + } + } + ], + "question_timestamps": "00:47:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around; it’s in front.", + "B": "Turn around; it’s on your left.", + "C": "Just turn around and walk straight; the dining table is behind you.", + "D": "Turn around and walk straight; the dining table will be on your right as you move.", + "E": "Keep walking forward without turning; the dining table is behind you so you shouldn’t turn around." + }, + "answer_label": "C", + "answer_text": "Just turn around and walk straight; the dining table is behind you.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_8_T1", + "correct_answer": "Just turn around and walk straight; the dining table is behind you.", + "original_options": { + "A": "Turn around; it’s in front.", + "B": "Turn around; it’s on your left.", + "C": "Just turn around and walk straight; the dining table is behind you.", + "D": "Turn around and walk straight; the dining table will be on your right as you move.", + "E": "Keep walking forward without turning; the dining table is behind you so you shouldn’t turn around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_5_T2", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many times did I wash the rice?", + "answer": "You washed the rice once, and the basin for washing rice was in front of you.", + "question_chinese": "我洗了多少次米？", + "answer_chinese": "你洗了一次米，洗米的盆在你前面", + "evidence": [ + { + "description": "The first", + "timestamps": { + "start": "00:00:46", + "end": "00:00:53" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You washed the rice twice, and the basin for washing rice was in front of you.", + "B": "You washed the rice once, and the basin for washing rice was in front of you." + }, + "answer_label": "B", + "answer_text": "You washed the rice once, and the basin for washing rice was in front of you.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_5_T2", + "correct_answer": "You washed the rice once, and the basin for washing rice was in front of you.", + "original_options": { + "A": "You washed the rice twice, and the basin for washing rice was in front of you.", + "B": "You washed the rice once, and the basin for washing rice was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_5_T1", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many times did I wash the rice?", + "answer": "You washed the rice twice, and the basin for washing rice was in front of you.", + "question_chinese": "我洗了多少次米？", + "answer_chinese": "你洗了两次米，洗米的盆在你前面", + "evidence": [ + { + "description": "the first", + "timestamps": { + "start": "00:00:45", + "end": "00:00:50" + } + }, + { + "description": "the second", + "timestamps": { + "start": "00:02:23", + "end": "00:02:41" + } + } + ], + "question_timestamps": "00:02:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You washed the rice twice, and the basin for washing rice was in front of you.", + "B": "You washed the rice once, and the basin for washing rice was in front of you." + }, + "answer_label": "A", + "answer_text": "You washed the rice twice, and the basin for washing rice was in front of you.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_5_T1", + "correct_answer": "You washed the rice twice, and the basin for washing rice was in front of you.", + "original_options": { + "A": "You washed the rice twice, and the basin for washing rice was in front of you.", + "B": "You washed the rice once, and the basin for washing rice was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_4_T1", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Am I washing dishes right now?", + "answer": "No. You were washing rice right now, and the basin for washing rice was in front of you.", + "question_chinese": "我现在在洗碗吗？", + "answer_chinese": "不，你现在在洗米，洗米的盆在你前面。", + "evidence": [ + { + "description": "No, you are washing rice now.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:03:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes, you’re washing dishes right now.", + "B": "No. You were washing rice right now, and the basin for washing rice was in front of you.", + "C": "No; the basin is behind you.", + "D": "No. You were washing rice right now, and the basin for washing rice was to your left.", + "E": "No. You were washing rice right now, and the basin for washing rice was to your right, not in front." + }, + "answer_label": "B", + "answer_text": "No. You were washing rice right now, and the basin for washing rice was in front of you.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_4_T1", + "correct_answer": "No. You were washing rice right now, and the basin for washing rice was in front of you.", + "original_options": { + "A": "Yes, you’re washing dishes right now.", + "B": "No. You were washing rice right now, and the basin for washing rice was in front of you.", + "C": "No; the basin is behind you.", + "D": "No. You were washing rice right now, and the basin for washing rice was to your left.", + "E": "No. You were washing rice right now, and the basin for washing rice was to your right, not in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 20 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_4_T2", + "qid_num": 24, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Am I washing dishes right now?", + "answer": "Yes. You were washing dishes right now, and the dishes were in front of you.", + "question_chinese": "我现在在洗碗吗？", + "answer_chinese": "是，你现在在洗碗，碗在你前面", + "evidence": [ + { + "description": "Yes, you are washing dishes now.", + "timestamps": { + "start": "00:11:56", + "end": "00:12:06" + } + } + ], + "question_timestamps": "00:12:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No, you’re not washing dishes right now.", + "B": "Yes. You were washing dishes right now, and the dishes were in front of you.", + "C": "Yes; the dishes are behind you.", + "D": "Yes. You were washing dishes right now, and the dishes were to your left instead of in front of you.", + "E": "Yes. You were washing dishes right now, but the dishes were to your right as you washed them." + }, + "answer_label": "B", + "answer_text": "Yes. You were washing dishes right now, and the dishes were in front of you.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_4_T2", + "correct_answer": "Yes. You were washing dishes right now, and the dishes were in front of you.", + "original_options": { + "A": "No, you’re not washing dishes right now.", + "B": "Yes. You were washing dishes right now, and the dishes were in front of you.", + "C": "Yes; the dishes are behind you.", + "D": "Yes. You were washing dishes right now, and the dishes were to your left instead of in front of you.", + "E": "Yes. You were washing dishes right now, but the dishes were to your right as you washed them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 20, + 18 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_7", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cups are there?", + "answer": "six.", + "question_chinese": "有多少个杯子？", + "answer_chinese": "6", + "evidence": [ + { + "description": "There are six cups", + "timestamps": { + "start": "00:02:47", + "end": "00:02:48" + } + } + ], + "question_timestamps": "00:02:49", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Seven.", + "C": "There were five cups on the table during the scene, spread out in front of you as you looked down.", + "D": "Only one cup was visible, sitting alone near the edge of the counter in front of you.", + "E": "six." + }, + "answer_label": "E", + "answer_text": "six.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_7", + "correct_answer": "six.", + "original_options": { + "A": "Four.", + "B": "Seven.", + "C": "There were five cups on the table during the scene, spread out in front of you as you looked down.", + "D": "Only one cup was visible, sitting alone near the edge of the counter in front of you.", + "E": "six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 20, + 17 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_8", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cups are there?", + "answer": "six.", + "question_chinese": "有多少个杯子？", + "answer_chinese": "6", + "evidence": [ + { + "description": "There were six cups", + "timestamps": { + "start": "00:02:47", + "end": "00:02:48" + } + } + ], + "question_timestamps": "00:02:55", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "There are five cups in total on the counter now, grouped together in one area.", + "D": "You can count seven cups altogether on the counter, spread across two separate spots nearby.", + "E": "six." + }, + "answer_label": "E", + "answer_text": "six.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_8", + "correct_answer": "six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "There are five cups in total on the counter now, grouped together in one area.", + "D": "You can count seven cups altogether on the counter, spread across two separate spots nearby.", + "E": "six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_1_T4", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which side of me is the bucket on?", + "answer": "The bucket was in front of you.", + "question_chinese": "桶在我的哪边？", + "answer_chinese": "桶在你的前面", + "evidence": [ + { + "description": "The bucket is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bucket was behind you.", + "B": "The bucket was in front of you.", + "C": "The bucket was to your left.", + "D": "The bucket was off to your right side, not directly ahead of you.", + "E": "The bucket was positioned above you rather than on the floor in front of you." + }, + "answer_label": "B", + "answer_text": "The bucket was in front of you.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_1_T4", + "correct_answer": "The bucket was in front of you.", + "original_options": { + "A": "The bucket was behind you.", + "B": "The bucket was in front of you.", + "C": "The bucket was to your left.", + "D": "The bucket was off to your right side, not directly ahead of you.", + "E": "The bucket was positioned above you rather than on the floor in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 15 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_1_T3", + "qid_num": 30, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which side of me is the bucket on?", + "answer": "The bucket was to your right.", + "question_chinese": "桶在我的哪边？", + "answer_chinese": "桶在你右边", + "evidence": [ + { + "description": "The bucket is on your right.", + "timestamps": { + "start": "01:27:16", + "end": "01:27:22" + } + } + ], + "question_timestamps": "01:27:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bucket was to your right.", + "B": "The bucket was to your left.", + "C": "The bucket was behind you.", + "D": "The bucket was in front of you instead of being on your right side.", + "E": "The bucket was above you, not located to the right of you." + }, + "answer_label": "A", + "answer_text": "The bucket was to your right.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_1_T3", + "correct_answer": "The bucket was to your right.", + "original_options": { + "A": "The bucket was to your right.", + "B": "The bucket was to your left.", + "C": "The bucket was behind you.", + "D": "The bucket was in front of you instead of being on your right side.", + "E": "The bucket was above you, not located to the right of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 12 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_3_T2", + "qid_num": 31, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Relative to me, where is the cup?", + "answer": "The cup was straight in front of you.", + "question_chinese": "相对于我的位置，杯子在哪里？", + "answer_chinese": "杯子在你的正前方。", + "evidence": [ + { + "description": "The cup is directly in front of you.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The cup was behind you.", + "B": "The cup was straight in front of you.", + "C": "The cup was to your left.", + "D": "The cup was to your right side rather than straight ahead of you.", + "E": "The cup was above you instead of being positioned directly in front." + }, + "answer_label": "B", + "answer_text": "The cup was straight in front of you.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_3_T2", + "correct_answer": "The cup was straight in front of you.", + "original_options": { + "A": "The cup was behind you.", + "B": "The cup was straight in front of you.", + "C": "The cup was to your left.", + "D": "The cup was to your right side rather than straight ahead of you.", + "E": "The cup was above you instead of being positioned directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 12 + ] + }, + { + "video_uid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a", + "video_length": "01:35:56", + "qid": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_3_T1", + "qid_num": 32, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Relative to me, where is the cup?", + "answer": "The cup was behind you.", + "question_chinese": "相对于我的位置，杯子在哪里？", + "answer_chinese": "杯子在你的身后", + "evidence": [ + { + "description": "The cup is behind you.", + "timestamps": { + "start": "00:03:15", + "end": "00:03:17" + } + } + ], + "question_timestamps": "00:03:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The cup was in front of you.", + "B": "The cup was to your left.", + "C": "The cup was to your right side rather than being directly behind you where you stood.", + "D": "The cup was behind you.", + "E": "The cup was above you, not positioned behind your body when you turned around." + }, + "answer_label": "D", + "answer_text": "The cup was behind you.", + "q_id": "5dab6f94-5ca7-4985-bcd0-67f4fc04687a_3_T1", + "correct_answer": "The cup was behind you.", + "original_options": { + "A": "The cup was in front of you.", + "B": "The cup was to your left.", + "C": "The cup was to your right side rather than being directly behind you where you stood.", + "D": "The cup was behind you.", + "E": "The cup was above you, not positioned behind your body when you turned around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 14 + ] + } +] diff --git a/QAs/5e64c89d-34ce-49a3-949a-dc90385cdbcc-mcq.json b/QAs/5e64c89d-34ce-49a3-949a-dc90385cdbcc-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..aa8235ec8c04ac86fed58ea47d34974b716d0b4f --- /dev/null +++ b/QAs/5e64c89d-34ce-49a3-949a-dc90385cdbcc-mcq.json @@ -0,0 +1,1191 @@ +[ + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_1_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to move the blue pen and the white pen closer together, how far should I move them?", + "answer": "You can move the blue pen about 48 cm to the right.", + "question_chinese": "如果我想蓝色的笔和白色的笔放得近一些,应该移动多远?", + "answer_chinese": "你可以把蓝色的笔向右移动48cm左右。", + "evidence": [ + { + "description": "The blue pen is on your left, and the white pen is on your right. They are currently about 50cm apart.", + "timestamps": { + "start": "00:07:54", + "end": "00:08:07" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Move the blue pen 46 cm right.", + "B": "You can move the blue pen about 48 cm to the right.", + "C": "Move the blue pen 47 cm right.", + "D": "You can move the blue pen about 49 cm to the right to bring it nearer the white pen.", + "E": "You can move the blue pen about 50 cm to the right so the two pens end up closer together." + }, + "answer_label": "B", + "answer_text": "You can move the blue pen about 48 cm to the right.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_1_T1", + "correct_answer": "You can move the blue pen about 48 cm to the right.", + "original_options": { + "A": "Move the blue pen 46 cm right.", + "B": "You can move the blue pen about 48 cm to the right.", + "C": "Move the blue pen 47 cm right.", + "D": "You can move the blue pen about 49 cm to the right to bring it nearer the white pen.", + "E": "You can move the blue pen about 50 cm to the right so the two pens end up closer together." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 20 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_1_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to move the blue pen and the white pen closer together, how far should I move them?", + "answer": "You can move the white pen about 48 cm to the left.", + "question_chinese": "如果我想蓝色的笔和白色的笔放得近一些,应该移动多远?", + "answer_chinese": "你可以把白色的笔向左移动48cm左右。", + "evidence": [ + { + "description": "The blue pen is on your left, and the white pen is on your right. They are currently about 50cm apart.", + "timestamps": { + "start": "00:07:54", + "end": "00:08:07" + } + } + ], + "question_timestamps": "00:28:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Move the white pen 46 cm left.", + "B": "Move the white pen 47 cm left.", + "C": "You can move the white pen about 48 cm to the left.", + "D": "You can move the white pen about 49 cm to the left to reduce the gap to the blue pen.", + "E": "You can move the white pen about 50 cm to the left so it ends up closer to the blue pen." + }, + "answer_label": "C", + "answer_text": "You can move the white pen about 48 cm to the left.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_1_T2", + "correct_answer": "You can move the white pen about 48 cm to the left.", + "original_options": { + "A": "Move the white pen 46 cm left.", + "B": "Move the white pen 47 cm left.", + "C": "You can move the white pen about 48 cm to the left.", + "D": "You can move the white pen about 49 cm to the left to reduce the gap to the blue pen.", + "E": "You can move the white pen about 50 cm to the left so it ends up closer to the blue pen." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 21 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_7_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I got back to the room and sat at the desk, which was closer to me, the computer or the phone?", + "answer": "The computer is closer to you.", + "question_chinese": "当我回到房间坐在桌前的时候,电脑和手机哪个离我更近?", + "answer_chinese": "电脑离你更近。", + "evidence": [ + { + "description": "The computer is closer to you, about 10cm away.", + "timestamps": { + "start": "00:45:53", + "end": "00:46:02" + } + }, + { + "description": "The phone is on your right side, about 30cm away.", + "timestamps": { + "start": "00:46:47", + "end": "00:46:52" + } + } + ], + "question_timestamps": "00:46:57", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The phone is closer to you.", + "B": "The computer is closer to you." + }, + "answer_label": "B", + "answer_text": "The computer is closer to you.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_7_T1", + "correct_answer": "The computer is closer to you.", + "original_options": { + "A": "The phone is closer to you.", + "B": "The computer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_7_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I got back to the room and sat at the desk, which was closer to me, the computer or the phone?", + "answer": "The phone is closer to you.", + "question_chinese": "当我回到房间坐在桌前的时候,电脑和手机哪个离我更近?", + "answer_chinese": "手机离你更近。", + "evidence": [ + { + "description": "The computer is closer to you, about 10cm away.", + "timestamps": { + "start": "00:45:53", + "end": "00:46:02" + } + }, + { + "description": "The phone is on your right side, about 30cm away.", + "timestamps": { + "start": "00:46:47", + "end": "00:46:52" + } + } + ], + "question_timestamps": "00:54:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The phone is closer to you.", + "B": "The computer is closer to you." + }, + "answer_label": "A", + "answer_text": "The phone is closer to you.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_7_T2", + "correct_answer": "The phone is closer to you.", + "original_options": { + "A": "The phone is closer to you.", + "B": "The computer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_5", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following just now?", + "answer": "You was exiting the room, heading downstairs, and moving towards the laundry room.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你正从房间出来，下楼，朝着洗衣房方向移动。", + "evidence": [ + { + "description": "You come out of the room.", + "timestamps": { + "start": "00:14:38", + "end": "00:14:49" + } + }, + { + "description": "You are going down the stairs.", + "timestamps": { + "start": "00:14:57", + "end": "00:15:17" + } + }, + { + "description": "You walked into the laundry room.", + "timestamps": { + "start": "00:15:33", + "end": "00:15:37" + } + } + ], + "question_timestamps": "00:15:42", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You were heading upstairs to the bathroom.", + "B": "You were going straight to the living room.", + "C": "You was exiting the room, heading downstairs, and moving towards the laundry room.", + "D": "You were leaving the room, going upstairs, and moving toward the laundry room instead of downstairs.", + "E": "You were exiting the room, heading downstairs, but turning toward the garage rather than the laundry room." + }, + "answer_label": "C", + "answer_text": "You was exiting the room, heading downstairs, and moving towards the laundry room.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_5", + "correct_answer": "You was exiting the room, heading downstairs, and moving towards the laundry room.", + "original_options": { + "A": "You were heading upstairs to the bathroom.", + "B": "You were going straight to the living room.", + "C": "You was exiting the room, heading downstairs, and moving towards the laundry room.", + "D": "You were leaving the room, going upstairs, and moving toward the laundry room instead of downstairs.", + "E": "You were exiting the room, heading downstairs, but turning toward the garage rather than the laundry room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 17 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_6", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following just now?", + "answer": "You was coming out of the kitchen, heading upstairs, and moving towards the room.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你正从厨房出来，上楼，朝着房间方向移动。", + "evidence": [ + { + "description": "You come out from the kitchen.", + "timestamps": { + "start": "00:43:46", + "end": "00:43:54" + } + }, + { + "description": "You are climbing the stairs.", + "timestamps": { + "start": "00:44:01", + "end": "00:44:20" + } + }, + { + "description": "You walked into the room.", + "timestamps": { + "start": "00:44:25", + "end": "00:44:28" + } + } + ], + "question_timestamps": "00:44:31", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You were leaving the living room downstairs.", + "B": "You was coming out of the kitchen, heading upstairs, and moving towards the room.", + "C": "You were heading downstairs toward the kitchen.", + "D": "You were coming out of the kitchen, heading downstairs, and moving toward the room instead of upstairs.", + "E": "You were coming out of the hallway, heading upstairs, and moving toward the kitchen rather than the room." + }, + "answer_label": "B", + "answer_text": "You was coming out of the kitchen, heading upstairs, and moving towards the room.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_6", + "correct_answer": "You was coming out of the kitchen, heading upstairs, and moving towards the room.", + "original_options": { + "A": "You were leaving the living room downstairs.", + "B": "You was coming out of the kitchen, heading upstairs, and moving towards the room.", + "C": "You were heading downstairs toward the kitchen.", + "D": "You were coming out of the kitchen, heading downstairs, and moving toward the room instead of upstairs.", + "E": "You were coming out of the hallway, heading upstairs, and moving toward the kitchen rather than the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_9_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk from the bedside to the wardrobe just now?", + "answer": "You walked straight down the corridor to the end, turned around, then turned right at the intersection to enter the cloakroom.", + "question_chinese": "我刚才从床边到衣柜是怎么走的?", + "answer_chinese": "你沿过道直行走到尽头，向后转,在路口右转进入衣帽间。", + "evidence": [ + { + "description": "You walk straight along the aisle.", + "timestamps": { + "start": "00:55:33", + "end": "00:55:35" + } + }, + { + "description": "Turn left at the intersection and proceed to the cloakroom.", + "timestamps": { + "start": "00:55:36", + "end": "00:55:45" + } + }, + { + "description": "You opened the wardrobe on the right.", + "timestamps": { + "start": "00:55:46", + "end": "00:55:50" + } + } + ], + "question_timestamps": "00:55:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left into the cloakroom.", + "B": "You walked straight down the corridor to the end, turned around, then turned right at the intersection to enter the cloakroom.", + "C": "You entered without turning around.", + "D": "You walked straight to the end of the corridor, turned around, then turned left at the intersection to enter the cloakroom.", + "E": "You went down the corridor, turned right at the intersection, and entered the cloakroom without reaching the very end." + }, + "answer_label": "B", + "answer_text": "You walked straight down the corridor to the end, turned around, then turned right at the intersection to enter the cloakroom.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_9_T1", + "correct_answer": "You walked straight down the corridor to the end, turned around, then turned right at the intersection to enter the cloakroom.", + "original_options": { + "A": "You turned left into the cloakroom.", + "B": "You walked straight down the corridor to the end, turned around, then turned right at the intersection to enter the cloakroom.", + "C": "You entered without turning around.", + "D": "You walked straight to the end of the corridor, turned around, then turned left at the intersection to enter the cloakroom.", + "E": "You went down the corridor, turned right at the intersection, and entered the cloakroom without reaching the very end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 19 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_9_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk from the bedside to the wardrobe just now?", + "answer": "You went straight down the corridor, then turned left at the intersection to enter the cloakroom.", + "question_chinese": "我刚才从床边到衣柜是怎么走的?", + "answer_chinese": "你沿过道直行,在路口左转进入衣帽间。", + "evidence": [ + { + "description": "You walk straight along the aisle.", + "timestamps": { + "start": "00:55:33", + "end": "00:55:35" + } + }, + { + "description": "Turn left at the intersection and proceed to the cloakroom.", + "timestamps": { + "start": "00:55:36", + "end": "00:55:45" + } + }, + { + "description": "You opened the wardrobe on the right.", + "timestamps": { + "start": "00:55:46", + "end": "00:55:50" + } + } + ], + "question_timestamps": "01:04:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went straight down the corridor, then turned left at the intersection to enter the cloakroom.", + "B": "You turned right into the cloakroom.", + "C": "You turned around before entering.", + "D": "You walked straight down the corridor to the end, turned around, then turned right at the intersection to enter the cloakroom.", + "E": "You went down the corridor, passed the intersection, then doubled back and turned left into the cloakroom." + }, + "answer_label": "A", + "answer_text": "You went straight down the corridor, then turned left at the intersection to enter the cloakroom.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_9_T2", + "correct_answer": "You went straight down the corridor, then turned left at the intersection to enter the cloakroom.", + "original_options": { + "A": "You went straight down the corridor, then turned left at the intersection to enter the cloakroom.", + "B": "You turned right into the cloakroom.", + "C": "You turned around before entering.", + "D": "You walked straight down the corridor to the end, turned around, then turned right at the intersection to enter the cloakroom.", + "E": "You went down the corridor, passed the intersection, then doubled back and turned left into the cloakroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 17 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_4_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I first entered the laundry room, how many detergent buckets were next to the washing machine?", + "answer": "There are two jugs of laundry detergent next to the washing machine.", + "question_chinese": "我刚进洗衣房的时候，洗衣机旁边有多少桶洗衣液？", + "answer_chinese": "洗衣机旁有两桶洗衣液。", + "evidence": [ + { + "description": "There is a bucket of pink laundry detergent and a bucket of blue laundry detergent on a washing machine.", + "timestamps": { + "start": "00:15:37", + "end": "00:15:41" + } + } + ], + "question_timestamps": "00:20:41", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are two jugs of laundry detergent next to the washing machine.", + "B": "One jug next to the washing machine.", + "C": "Three jugs next to the washing machine.", + "D": "There is one jug of laundry detergent next to the washing machine when you first enter the laundry room.", + "E": "There are three jugs of laundry detergent lined up next to the washing machine when you first enter the laundry room." + }, + "answer_label": "A", + "answer_text": "There are two jugs of laundry detergent next to the washing machine.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_4_T2", + "correct_answer": "There are two jugs of laundry detergent next to the washing machine.", + "original_options": { + "A": "There are two jugs of laundry detergent next to the washing machine.", + "B": "One jug next to the washing machine.", + "C": "Three jugs next to the washing machine.", + "D": "There is one jug of laundry detergent next to the washing machine when you first enter the laundry room.", + "E": "There are three jugs of laundry detergent lined up next to the washing machine when you first enter the laundry room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 21 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_4_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I first entered the laundry room, how many detergent buckets were next to the washing machine?", + "answer": "There are two jugs of laundry detergent next to the washing machine.", + "question_chinese": "我刚进洗衣房的时候，洗衣机旁边有多少桶洗衣液？", + "answer_chinese": "洗衣机旁边有两桶洗衣液。", + "evidence": [ + { + "description": "There is a bucket of pink laundry detergent and a bucket of blue laundry detergent on a washing machine.", + "timestamps": { + "start": "00:15:37", + "end": "00:15:41" + } + } + ], + "question_timestamps": "00:45:41", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero jugs next to the washing machine.", + "B": "Four jugs next to the washing machine.", + "C": "There are zero jugs of laundry detergent next to the washing machine when you first enter the laundry room.", + "D": "There are four jugs of laundry detergent next to the washing machine when you first enter the laundry room.", + "E": "There are two jugs of laundry detergent next to the washing machine." + }, + "answer_label": "E", + "answer_text": "There are two jugs of laundry detergent next to the washing machine.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_4_T1", + "correct_answer": "There are two jugs of laundry detergent next to the washing machine.", + "original_options": { + "A": "Zero jugs next to the washing machine.", + "B": "Four jugs next to the washing machine.", + "C": "There are zero jugs of laundry detergent next to the washing machine when you first enter the laundry room.", + "D": "There are four jugs of laundry detergent next to the washing machine when you first enter the laundry room.", + "E": "There are two jugs of laundry detergent next to the washing machine." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_8", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to drink orange juice, so where can I find it?", + "answer": "There is orange juice in the refrigerator.", + "question_chinese": "我想喝橙汁，我要去哪里找？", + "answer_chinese": "冰箱里有橙汁。", + "evidence": [ + { + "description": "When you just took the orange juice from the refrigerator, there was still some left.", + "timestamps": { + "start": "00:17:01", + "end": "00:17:12" + } + } + ], + "question_timestamps": "00:17:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s in the freezer.", + "B": "It’s in the kitchen cabinet.", + "C": "You can find the orange juice sitting on the dining table, not stored in the fridge.", + "D": "Look for the orange juice on the kitchen counter instead of inside any cold storage.", + "E": "There is orange juice in the refrigerator." + }, + "answer_label": "E", + "answer_text": "There is orange juice in the refrigerator.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_8", + "correct_answer": "There is orange juice in the refrigerator.", + "original_options": { + "A": "It’s in the freezer.", + "B": "It’s in the kitchen cabinet.", + "C": "You can find the orange juice sitting on the dining table, not stored in the fridge.", + "D": "Look for the orange juice on the kitchen counter instead of inside any cold storage.", + "E": "There is orange juice in the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_3_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to drink orange juice—where should I go to find it?", + "answer": "There is orange juice on the counter across from the fridge.", + "question_chinese": "我想喝橙汁，我要去哪里找？", + "answer_chinese": "冰箱对面的台子上面有橙汁。", + "evidence": [ + { + "description": "When you just took the orange juice from the refrigerator, there was still some left.", + "timestamps": { + "start": "00:17:01", + "end": "00:17:19" + } + } + ], + "question_timestamps": "00:17:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s inside the fridge.", + "B": "It’s on the table edge.", + "C": "Check the counter next to the fridge, rather than the counter directly across from it.", + "D": "There is orange juice on the counter across from the fridge.", + "E": "The orange juice is on the inner part of the table, not on a counter opposite the fridge." + }, + "answer_label": "D", + "answer_text": "There is orange juice on the counter across from the fridge.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_3_T2", + "correct_answer": "There is orange juice on the counter across from the fridge.", + "original_options": { + "A": "It’s inside the fridge.", + "B": "It’s on the table edge.", + "C": "Check the counter next to the fridge, rather than the counter directly across from it.", + "D": "There is orange juice on the counter across from the fridge.", + "E": "The orange juice is on the inner part of the table, not on a counter opposite the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 18 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_9", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to drink orange juice, so where can I find it?", + "answer": "There is orange juice in the refrigerator.", + "question_chinese": "我想喝橙汁，我要去哪里找？", + "answer_chinese": "冰箱里有橙汁。", + "evidence": [ + { + "description": "When you just took the orange juice from the refrigerator, there was still some left.", + "timestamps": { + "start": "00:17:01", + "end": "00:17:12" + } + } + ], + "question_timestamps": "00:35:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There is orange juice in the refrigerator.", + "B": "It’s in the freezer.", + "C": "It’s on the kitchen table.", + "D": "You can find the orange juice in a wine cooler, not inside the regular refrigerator.", + "E": "The juice you’re looking for is in the mini fridge, but it’s apple juice rather than orange." + }, + "answer_label": "A", + "answer_text": "There is orange juice in the refrigerator.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_9", + "correct_answer": "There is orange juice in the refrigerator.", + "original_options": { + "A": "There is orange juice in the refrigerator.", + "B": "It’s in the freezer.", + "C": "It’s on the kitchen table.", + "D": "You can find the orange juice in a wine cooler, not inside the regular refrigerator.", + "E": "The juice you’re looking for is in the mini fridge, but it’s apple juice rather than orange." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 19 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_3_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to drink orange juice—where should I go to find it?", + "answer": "There is orange juice inside the fridge.", + "question_chinese": "我想喝橙汁，我要去哪里找？", + "answer_chinese": "冰箱里面有橙汁。", + "evidence": [ + { + "description": "When you just took the orange juice from the refrigerator, there was still some left.", + "timestamps": { + "start": "00:17:01", + "end": "00:17:12" + } + } + ], + "question_timestamps": "00:35:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s outside the fridge.", + "B": "It’s on the table edge.", + "C": "There is orange juice on the counter across from the fridge, rather than inside the fridge itself.", + "D": "You should look in the refrigerator door area, but the orange juice isn’t inside the main fridge compartment.", + "E": "There is orange juice inside the fridge." + }, + "answer_label": "E", + "answer_text": "There is orange juice inside the fridge.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_3_T1", + "correct_answer": "There is orange juice inside the fridge.", + "original_options": { + "A": "It’s outside the fridge.", + "B": "It’s on the table edge.", + "C": "There is orange juice on the counter across from the fridge, rather than inside the fridge itself.", + "D": "You should look in the refrigerator door area, but the orange juice isn’t inside the main fridge compartment.", + "E": "There is orange juice inside the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 19 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_8_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many photos did I take of the building blocks with my phone?", + "answer": "You took one photo of the blocks.", + "question_chinese": "我拿手机给积木拍了几张照片？", + "answer_chinese": "你给积木拍了一张照片。", + "evidence": [ + { + "description": "You took the first photo of the building blocks.", + "timestamps": { + "start": "00:54:23", + "end": "00:54:26" + } + } + ], + "question_timestamps": "00:54:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took zero photos.", + "B": "You took two photos.", + "C": "You took three photos of the blocks, capturing them from different angles on your phone.", + "D": "You took one photo of the blocks.", + "E": "You took four photos of the blocks, taking multiple shots in a row with your phone." + }, + "answer_label": "D", + "answer_text": "You took one photo of the blocks.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_8_T1", + "correct_answer": "You took one photo of the blocks.", + "original_options": { + "A": "You took zero photos.", + "B": "You took two photos.", + "C": "You took three photos of the blocks, capturing them from different angles on your phone.", + "D": "You took one photo of the blocks.", + "E": "You took four photos of the blocks, taking multiple shots in a row with your phone." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_17", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many photos did I take of the building blocks with my phone?", + "answer": "You took three photos of the building blocks.", + "question_chinese": "我拿手机给积木拍了几张照片？", + "answer_chinese": "你给积木拍了三张照片。", + "evidence": [ + { + "description": "You took the first photo of the building blocks.", + "timestamps": { + "start": "00:54:23", + "end": "00:54:26" + } + }, + { + "description": "You took the second photo of the building blocks.", + "timestamps": { + "start": "00:54:32", + "end": "00:54:39" + } + }, + { + "description": "You took the third photo of the building blocks.", + "timestamps": { + "start": "00:54:49", + "end": "00:54:56" + } + } + ], + "question_timestamps": "00:55:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You took four photos.", + "B": "You took five photos.", + "C": "You took three photos of the building blocks.", + "D": "You took six photos of the building blocks, snapping several extra pictures beyond the initial set.", + "E": "You took eight photos of the building blocks, repeatedly photographing them with your phone." + }, + "answer_label": "C", + "answer_text": "You took three photos of the building blocks.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_17", + "correct_answer": "You took three photos of the building blocks.", + "original_options": { + "A": "You took four photos.", + "B": "You took five photos.", + "C": "You took three photos of the building blocks.", + "D": "You took six photos of the building blocks, snapping several extra pictures beyond the initial set.", + "E": "You took eight photos of the building blocks, repeatedly photographing them with your phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 14 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_8_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many photos did I take of the building blocks with my phone?", + "answer": "You took three photos of the blocks.", + "question_chinese": "我拿手机给积木拍了几张照片？", + "answer_chinese": "你给积木拍了三张照片。", + "evidence": [ + { + "description": "You took the first photo of the building blocks.", + "timestamps": { + "start": "00:54:23", + "end": "00:54:26" + } + }, + { + "description": "You took the second photo of the building blocks.", + "timestamps": { + "start": "00:54:32", + "end": "00:54:39" + } + }, + { + "description": "You took the third photo of the building blocks.", + "timestamps": { + "start": "00:54:49", + "end": "00:54:56" + } + } + ], + "question_timestamps": "01:01:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took one photo.", + "B": "You took two photos.", + "C": "You took four photos of the blocks, adding an extra shot beyond a small set of pictures.", + "D": "You took five photos of the blocks, taking several more pictures with your phone in succession.", + "E": "You took three photos of the blocks." + }, + "answer_label": "E", + "answer_text": "You took three photos of the blocks.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_8_T2", + "correct_answer": "You took three photos of the blocks.", + "original_options": { + "A": "You took one photo.", + "B": "You took two photos.", + "C": "You took four photos of the blocks, adding an extra shot beyond a small set of pictures.", + "D": "You took five photos of the blocks, taking several more pictures with your phone in succession.", + "E": "You took three photos of the blocks." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_13", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the sponge from my current position?", + "answer": "You can reach the sponge.", + "question_chinese": "我现在的位置,能够到海绵吗?", + "answer_chinese": "你能够到海绵。", + "evidence": [ + { + "description": "The sponge is about 30cm away from you, located above and to your right, and you can reach it with your hand.", + "timestamps": { + "start": "00:42:45", + "end": "00:42:48" + } + } + ], + "question_timestamps": "00:42:53", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't reach the sponge.", + "B": "You can reach the sponge." + }, + "answer_label": "B", + "answer_text": "You can reach the sponge.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_13", + "correct_answer": "You can reach the sponge.", + "original_options": { + "A": "You can't reach the sponge.", + "B": "You can reach the sponge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_14", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the sponge from my current position?", + "answer": "You can't reach the sponge.", + "question_chinese": "我现在的位置,能够到海绵吗?", + "answer_chinese": "你不能够到海绵。", + "evidence": [ + { + "description": "The sponge is in the kitchen.", + "timestamps": { + "start": "00:42:50", + "end": "00:42:55" + } + }, + { + "description": "The vertical distance between you and the sponge is about two floors, so you can't reach it.", + "timestamps": { + "start": "00:44:05", + "end": "00:44:22" + } + }, + { + "description": "You are in the room now.", + "timestamps": { + "start": "00:44:29", + "end": "00:44:35" + } + } + ], + "question_timestamps": "00:44:38", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "You can't reach the sponge.", + "B": "You can reach the sponge." + }, + "answer_label": "A", + "answer_text": "You can't reach the sponge.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_14", + "correct_answer": "You can't reach the sponge.", + "original_options": { + "A": "You can't reach the sponge.", + "B": "You can reach the sponge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_5_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "The bowls are washed—where should I put them?", + "answer": "You should put the bowl back on the dish rack to your left.", + "question_chinese": "碗洗好了，我应该把它放在哪里？", + "answer_chinese": "你应该把碗放回左手边的碗架上。", + "evidence": [ + { + "description": "There are many tableware items on the shelf to the left of the sink.", + "timestamps": { + "start": "00:38:23", + "end": "00:38:28" + } + } + ], + "question_timestamps": "00:38:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You should put the bowl back on the dish rack to your left.", + "B": "Put them on the rack in front.", + "C": "Put them on the rack to your right.", + "D": "Place the washed bowl back onto the dish rack directly in front of you, not off to the side.", + "E": "Set the bowl on the dish rack behind you, where you were placing items earlier." + }, + "answer_label": "A", + "answer_text": "You should put the bowl back on the dish rack to your left.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_5_T1", + "correct_answer": "You should put the bowl back on the dish rack to your left.", + "original_options": { + "A": "You should put the bowl back on the dish rack to your left.", + "B": "Put them on the rack in front.", + "C": "Put them on the rack to your right.", + "D": "Place the washed bowl back onto the dish rack directly in front of you, not off to the side.", + "E": "Set the bowl on the dish rack behind you, where you were placing items earlier." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 15 + ] + }, + { + "video_uid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc", + "video_length": "01:04:10", + "qid": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_5_T2", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "The bowls are washed—where should I put them?", + "answer": "You should put the bowl back on the dish rack to your left.", + "question_chinese": "碗洗好了，我应该把它放在哪里？", + "answer_chinese": "应该把碗放回你的左手边的碗架上。", + "evidence": [ + { + "description": "There are many tableware items on the shelf to the left of the sink.", + "timestamps": { + "start": "00:38:23", + "end": "00:38:28" + } + } + ], + "question_timestamps": "00:55:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put them on the rack in front.", + "B": "Put them on the rack behind you.", + "C": "Place the washed bowl back onto the dish rack to your right, next to the other dishes.", + "D": "You should put the bowl back on the dish rack to your left.", + "E": "Set the bowl on the dish rack above you, rather than on the counter-level rack." + }, + "answer_label": "D", + "answer_text": "You should put the bowl back on the dish rack to your left.", + "q_id": "5e64c89d-34ce-49a3-949a-dc90385cdbcc_5_T2", + "correct_answer": "You should put the bowl back on the dish rack to your left.", + "original_options": { + "A": "Put them on the rack in front.", + "B": "Put them on the rack behind you.", + "C": "Place the washed bowl back onto the dish rack to your right, next to the other dishes.", + "D": "You should put the bowl back on the dish rack to your left.", + "E": "Set the bowl on the dish rack above you, rather than on the counter-level rack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 15 + ] + } +] diff --git a/QAs/5e7dbc32-08e0-4266-9bd3-27c412d1af7c-mcq.json b/QAs/5e7dbc32-08e0-4266-9bd3-27c412d1af7c-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..561aaee0ef3d6dcba39e5f1ef07f4abf660572f1 --- /dev/null +++ b/QAs/5e7dbc32-08e0-4266-9bd3-27c412d1af7c-mcq.json @@ -0,0 +1,1390 @@ +[ + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_7_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to find more clothes, where should I go look?", + "answer": "Go check by the wardrobe; the wardrobe is in front of you.", + "question_chinese": "如果我想找更多的衣服，我要去哪找？", + "answer_chinese": "你去衣柜那里找，衣柜在你的正前方。", + "evidence": [ + { + "description": "There are many clothes hanging in the closet.", + "timestamps": { + "start": "00:23:21", + "end": "00:23:26" + } + } + ], + "question_timestamps": "00:23:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go check by the wardrobe; the wardrobe is in front of you.", + "B": "Go check behind you.", + "C": "Go check to your left.", + "D": "Go check by the wardrobe; the wardrobe is to your right, not straight ahead.", + "E": "Go check by the wardrobe; the wardrobe is behind you, so turn around to look." + }, + "answer_label": "A", + "answer_text": "Go check by the wardrobe; the wardrobe is in front of you.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_7_T1", + "correct_answer": "Go check by the wardrobe; the wardrobe is in front of you.", + "original_options": { + "A": "Go check by the wardrobe; the wardrobe is in front of you.", + "B": "Go check behind you.", + "C": "Go check to your left.", + "D": "Go check by the wardrobe; the wardrobe is to your right, not straight ahead.", + "E": "Go check by the wardrobe; the wardrobe is behind you, so turn around to look." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 15 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_7_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to find more clothes, where should I go look?", + "answer": "Go check by the sofa; the sofa is in front of you.", + "question_chinese": "如果我想找更多的衣服，我要去哪找？", + "answer_chinese": "你去沙发那里找，沙发在你的正前方。", + "evidence": [ + { + "description": "There are many clothes on the sofa and on the ground nearby.", + "timestamps": { + "start": "00:23:29", + "end": "00:23:32" + } + } + ], + "question_timestamps": "00:23:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go check behind you.", + "B": "Go check to your left.", + "C": "Go check by the sofa; the sofa is in front of you.", + "D": "Go check by the sofa; the sofa is to your right, not directly in front of you.", + "E": "Go check by the sofa; the sofa is behind you, so you need to turn around." + }, + "answer_label": "C", + "answer_text": "Go check by the sofa; the sofa is in front of you.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_7_T2", + "correct_answer": "Go check by the sofa; the sofa is in front of you.", + "original_options": { + "A": "Go check behind you.", + "B": "Go check to your left.", + "C": "Go check by the sofa; the sofa is in front of you.", + "D": "Go check by the sofa; the sofa is to your right, not directly in front of you.", + "E": "Go check by the sofa; the sofa is behind you, so you need to turn around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 16 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_16", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to find more clothes, where should I go?", + "answer": "You can look in the closet and on the sofa.", + "question_chinese": "如果我想找更多的衣服，我要去哪找？", + "answer_chinese": "你去衣柜和沙发那里找。", + "evidence": [ + { + "description": "There are many clothes hanging in the closet.", + "timestamps": { + "start": "00:23:21", + "end": "00:23:26" + } + }, + { + "description": "There are many clothes on the sofa and on the ground nearby.", + "timestamps": { + "start": "00:23:29", + "end": "00:23:36" + } + } + ], + "question_timestamps": "00:23:39", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Look in the dresser and under the bed.", + "B": "Look in the laundry basket and chair.", + "C": "You can look in the dresser and under the bed to find more clothes around the room.", + "D": "You can look in the closet and on the sofa.", + "E": "You can look in the laundry basket and on the chair for any extra clothing items." + }, + "answer_label": "D", + "answer_text": "You can look in the closet and on the sofa.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_16", + "correct_answer": "You can look in the closet and on the sofa.", + "original_options": { + "A": "Look in the dresser and under the bed.", + "B": "Look in the laundry basket and chair.", + "C": "You can look in the dresser and under the bed to find more clothes around the room.", + "D": "You can look in the closet and on the sofa.", + "E": "You can look in the laundry basket and on the chair for any extra clothing items." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 16 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_15", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to find more clothes, where should I go?", + "answer": "You can look in the closet and on the sofa.", + "question_chinese": "如果我想找更多的衣服，我要去哪找？", + "answer_chinese": "你去衣柜和沙发那里找。", + "evidence": [ + { + "description": "There are many clothes hanging in the closet.", + "timestamps": { + "start": "00:23:21", + "end": "00:23:26" + } + }, + { + "description": "There are many clothes on the sofa and on the ground nearby.", + "timestamps": { + "start": "00:23:29", + "end": "00:23:36" + } + } + ], + "question_timestamps": "00:23:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Look under the bed and in the cupboard.", + "B": "Look in the drawer and on the bench.", + "C": "You can look under the bed and in the cupboard to find more clothes stored away.", + "D": "You can look in the closet and on the sofa.", + "E": "You can look in the drawer and on the bench for additional clothes you might have left out." + }, + "answer_label": "D", + "answer_text": "You can look in the closet and on the sofa.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_15", + "correct_answer": "You can look in the closet and on the sofa.", + "original_options": { + "A": "Look under the bed and in the cupboard.", + "B": "Look in the drawer and on the bench.", + "C": "You can look under the bed and in the cupboard to find more clothes stored away.", + "D": "You can look in the closet and on the sofa.", + "E": "You can look in the drawer and on the bench for additional clothes you might have left out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 18 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_13", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I move to the car in total?", + "answer": "You moved a total of 3 chairs to the car.", + "question_chinese": "我一共搬了多少把椅子到车上？", + "answer_chinese": "你一共搬了3把椅子到车子。", + "evidence": [ + { + "description": "You initially moved two chairs from the restaurant to the car.", + "timestamps": { + "start": "00:12:38", + "end": "00:12:45" + } + }, + { + "description": "Later, a chair was moved from the bedroom to the car.", + "timestamps": { + "start": "00:14:35", + "end": "00:14:48" + } + } + ], + "question_timestamps": "00:15:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "2 chairs total.", + "B": "4 chairs total.", + "C": "You moved a total of 1 chair to the car, making just a single trip with one chair.", + "D": "You moved a total of 3 chairs to the car.", + "E": "You moved a total of 5 chairs to the car, bringing over several chairs across multiple trips." + }, + "answer_label": "D", + "answer_text": "You moved a total of 3 chairs to the car.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_13", + "correct_answer": "You moved a total of 3 chairs to the car.", + "original_options": { + "A": "2 chairs total.", + "B": "4 chairs total.", + "C": "You moved a total of 1 chair to the car, making just a single trip with one chair.", + "D": "You moved a total of 3 chairs to the car.", + "E": "You moved a total of 5 chairs to the car, bringing over several chairs across multiple trips." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 17 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_14", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I move to the car in total?", + "answer": "You moved a total of 3 chairs to the car.", + "question_chinese": "我一共搬了多少把椅子到车上？", + "answer_chinese": "你一共搬了3把椅子到车子。", + "evidence": [ + { + "description": "You initially moved two chairs from the restaurant to the car.", + "timestamps": { + "start": "00:12:38", + "end": "00:12:45" + } + }, + { + "description": "Later, a chair was moved from the bedroom to the car.", + "timestamps": { + "start": "00:14:35", + "end": "00:14:48" + } + } + ], + "question_timestamps": "00:25:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "1 chair.", + "B": "5 chairs.", + "C": "You moved a total of 2 chairs to the car overall in total.", + "D": "You moved a total of 4 chairs from the house to the car in total.", + "E": "You moved a total of 3 chairs to the car." + }, + "answer_label": "E", + "answer_text": "You moved a total of 3 chairs to the car.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_14", + "correct_answer": "You moved a total of 3 chairs to the car.", + "original_options": { + "A": "1 chair.", + "B": "5 chairs.", + "C": "You moved a total of 2 chairs to the car overall in total.", + "D": "You moved a total of 4 chairs from the house to the car in total.", + "E": "You moved a total of 3 chairs to the car." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_1_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I just walking along?", + "answer": "You’re walking straight ahead.", + "question_chinese": "我刚刚正在走什么路线?", + "answer_chinese": "你正在直走。", + "evidence": [ + { + "description": "You are going straight.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’re walking straight ahead.", + "B": "You’re turning.", + "C": "You’re walking backward.", + "D": "You’re veering left instead of continuing straight, shifting your path to one side as you move forward.", + "E": "You’re heading around a corner, changing direction rather than continuing straight down the same line." + }, + "answer_label": "A", + "answer_text": "You’re walking straight ahead.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_1_T1", + "correct_answer": "You’re walking straight ahead.", + "original_options": { + "A": "You’re walking straight ahead.", + "B": "You’re turning.", + "C": "You’re walking backward.", + "D": "You’re veering left instead of continuing straight, shifting your path to one side as you move forward.", + "E": "You’re heading around a corner, changing direction rather than continuing straight down the same line." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 18, + 16 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_1_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I just walking along?", + "answer": "You’re turning.", + "question_chinese": "我刚刚正在走什么路线?", + "answer_chinese": "你正在拐弯。", + "evidence": [ + { + "description": "You are turning the corner.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’re walking straight ahead.", + "B": "You’re walking backward.", + "C": "You’re continuing forward without changing direction, keeping your body aligned with the path in front of you.", + "D": "You’re moving along a straight line, not pivoting or rotating as you proceed through the space.", + "E": "You’re turning." + }, + "answer_label": "E", + "answer_text": "You’re turning.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_1_T2", + "correct_answer": "You’re turning.", + "original_options": { + "A": "You’re walking straight ahead.", + "B": "You’re walking backward.", + "C": "You’re continuing forward without changing direction, keeping your body aligned with the path in front of you.", + "D": "You’re moving along a straight line, not pivoting or rotating as you proceed through the space.", + "E": "You’re turning." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 17 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following just now?", + "answer": "You was moving from the kitchen, straight through the hallway, towards the dining table.", + "question_chinese": "我刚刚正在走什么路线?", + "answer_chinese": "你正从厨房出发,直行穿过过道,朝着餐桌方向移动。", + "evidence": [ + { + "description": "You start from the kitchen, walk straight through the hallway, and arrive beside the dining table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From kitchen to living room.", + "B": "From dining table to kitchen.", + "C": "You was moving from the kitchen, straight through the hallway, towards the dining table.", + "D": "You was moving from the kitchen, turning right through the hallway, towards the living room sofa.", + "E": "You was moving from the dining table, straight through the hallway, towards the kitchen island." + }, + "answer_label": "C", + "answer_text": "You was moving from the kitchen, straight through the hallway, towards the dining table.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_1", + "correct_answer": "You was moving from the kitchen, straight through the hallway, towards the dining table.", + "original_options": { + "A": "From kitchen to living room.", + "B": "From dining table to kitchen.", + "C": "You was moving from the kitchen, straight through the hallway, towards the dining table.", + "D": "You was moving from the kitchen, turning right through the hallway, towards the living room sofa.", + "E": "You was moving from the dining table, straight through the hallway, towards the kitchen island." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_11", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I just took two chairs from the dining room. How many chairs are left in the dining room now?", + "answer": "You took two just now, and there is one left.", + "question_chinese": "我刚才从餐厅拿了两把椅子,现在餐厅里还剩几把?", + "answer_chinese": "你刚才拿了两把,现在还剩1把。", + "evidence": [ + { + "description": "The restaurant originally had three chairs.", + "timestamps": { + "start": "00:12:07", + "end": "00:12:12" + } + }, + { + "description": "You took two, and now there is one left.", + "timestamps": { + "start": "00:12:38", + "end": "00:12:44" + } + } + ], + "question_timestamps": "00:12:47", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You took two just now, and there is one left.", + "B": "Zero left.", + "C": "Three left.", + "D": "After taking two chairs, there are four chairs left in the dining room now.", + "E": "After taking two chairs, there are five chairs still left in the dining room." + }, + "answer_label": "A", + "answer_text": "You took two just now, and there is one left.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_11", + "correct_answer": "You took two just now, and there is one left.", + "original_options": { + "A": "You took two just now, and there is one left.", + "B": "Zero left.", + "C": "Three left.", + "D": "After taking two chairs, there are four chairs left in the dining room now.", + "E": "After taking two chairs, there are five chairs still left in the dining room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_12", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I just took two chairs from the dining room. How many chairs are left in the dining room now?", + "answer": "There are currently zero left.", + "question_chinese": "我刚才从餐厅拿了两把椅子,现在餐厅里还剩几把?", + "answer_chinese": "现在还剩0把。", + "evidence": [ + { + "description": "The restaurant originally had three chairs.", + "timestamps": { + "start": "00:12:07", + "end": "00:12:12" + } + }, + { + "description": "You took two, and now there is one left.", + "timestamps": { + "start": "00:12:38", + "end": "00:12:44" + } + }, + { + "description": "The girl took 1 bunch, and now there are 0 bunches left.", + "timestamps": { + "start": "00:13:29", + "end": "00:13:30" + } + } + ], + "question_timestamps": "00:13:33", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One left.", + "B": "Two left.", + "C": "There are currently zero left.", + "D": "There are currently three chairs left in the dining room at this moment.", + "E": "There are currently four chairs left in the dining room right now." + }, + "answer_label": "C", + "answer_text": "There are currently zero left.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_12", + "correct_answer": "There are currently zero left.", + "original_options": { + "A": "One left.", + "B": "Two left.", + "C": "There are currently zero left.", + "D": "There are currently three chairs left in the dining room at this moment.", + "E": "There are currently four chairs left in the dining room right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_5", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I just opened the refrigerator, how many bottles of mineral water were there?", + "answer": "There is a bottle of mineral water.", + "question_chinese": "我刚打开冰箱的时候,冰箱里有几瓶矿泉水?", + "answer_chinese": "有一瓶矿泉水。", + "evidence": [ + { + "description": "When you just opened the refrigerator, there was a bottle of mineral water in the first layer.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:19" + } + } + ], + "question_timestamps": "00:06:34", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is a bottle of mineral water.", + "B": "There are zero bottles of mineral water.", + "C": "There are two bottles of mineral water.", + "D": "There are three bottles of mineral water in the refrigerator when you opened it just now.", + "E": "There are four bottles of mineral water visible inside the refrigerator when you opened the door." + }, + "answer_label": "A", + "answer_text": "There is a bottle of mineral water.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_5", + "correct_answer": "There is a bottle of mineral water.", + "original_options": { + "A": "There is a bottle of mineral water.", + "B": "There are zero bottles of mineral water.", + "C": "There are two bottles of mineral water.", + "D": "There are three bottles of mineral water in the refrigerator when you opened it just now.", + "E": "There are four bottles of mineral water visible inside the refrigerator when you opened the door." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_6", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I just opened the refrigerator, how many bottles of mineral water were there?", + "answer": "There is a bottle of mineral water.", + "question_chinese": "我刚打开冰箱的时候,冰箱里有几瓶矿泉水?", + "answer_chinese": "有一瓶矿泉水。", + "evidence": [ + { + "description": "When you just opened the refrigerator, there was a bottle of mineral water in the first layer.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:19" + } + } + ], + "question_timestamps": "00:18:34", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are zero bottles of mineral water.", + "B": "There are two bottles of mineral water.", + "C": "There are three bottles of mineral water in the refrigerator when you opened it just now.", + "D": "There are four bottles of mineral water visible inside the refrigerator when you opened the door.", + "E": "There is a bottle of mineral water." + }, + "answer_label": "E", + "answer_text": "There is a bottle of mineral water.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_6", + "correct_answer": "There is a bottle of mineral water.", + "original_options": { + "A": "There are zero bottles of mineral water.", + "B": "There are two bottles of mineral water.", + "C": "There are three bottles of mineral water in the refrigerator when you opened it just now.", + "D": "There are four bottles of mineral water visible inside the refrigerator when you opened the door.", + "E": "There is a bottle of mineral water." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_7", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I first entered the garbage room, how many trash bins were there?", + "answer": "There are three trash bins in the garbage room.", + "question_chinese": "我刚进垃圾房时，垃圾房里有几个垃圾桶？", + "answer_chinese": "垃圾房里有三个垃圾桶。", + "evidence": [ + { + "description": "When you first entered the garbage room, there were two trash cans on your left and one trash can on your right.", + "timestamps": { + "start": "00:08:14", + "end": "00:08:19" + } + } + ], + "question_timestamps": "00:08:21", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two trash bins.", + "B": "There are three trash bins in the garbage room.", + "C": "There are four trash bins.", + "D": "There is one trash bin in the garbage room when you first enter.", + "E": "There are five trash bins in the garbage room as you walk in the first time." + }, + "answer_label": "B", + "answer_text": "There are three trash bins in the garbage room.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_7", + "correct_answer": "There are three trash bins in the garbage room.", + "original_options": { + "A": "There are two trash bins.", + "B": "There are three trash bins in the garbage room.", + "C": "There are four trash bins.", + "D": "There is one trash bin in the garbage room when you first enter.", + "E": "There are five trash bins in the garbage room as you walk in the first time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 16 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_8", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I first entered the garbage room, how many trash bins were there?", + "answer": "There are three trash bins in the garbage room.", + "question_chinese": "我刚进垃圾房时，垃圾房里有几个垃圾桶？", + "answer_chinese": "垃圾房里有三个垃圾桶。", + "evidence": [ + { + "description": "When you first entered the garbage room, there were two trash cans on your left and one trash can on your right.", + "timestamps": { + "start": "00:08:14", + "end": "00:08:19" + } + } + ], + "question_timestamps": "00:20:21", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one trash bin.", + "B": "There are three trash bins in the garbage room.", + "C": "There are two trash bins.", + "D": "There are four trash bins in the garbage room when you first step inside.", + "E": "There are five trash bins in the garbage room at the moment you first entered." + }, + "answer_label": "B", + "answer_text": "There are three trash bins in the garbage room.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_8", + "correct_answer": "There are three trash bins in the garbage room.", + "original_options": { + "A": "There is one trash bin.", + "B": "There are three trash bins in the garbage room.", + "C": "There are two trash bins.", + "D": "There are four trash bins in the garbage room when you first step inside.", + "E": "There are five trash bins in the garbage room at the moment you first entered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_4_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first walked in the door, where did the knife appear relative to me?", + "answer": "The knife appeared in front of you.", + "question_chinese": "我刚进门时，刀出现在我的哪里？", + "answer_chinese": "刀出现在你的前面。", + "evidence": [ + { + "description": "The knife appears in front of you.", + "timestamps": { + "start": "00:10:23", + "end": "00:10:24" + } + } + ], + "question_timestamps": "00:10:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The knife appeared behind you.", + "B": "The knife appeared to your left.", + "C": "The knife appeared to your right as you walked in through the door.", + "D": "The knife appeared in front of you.", + "E": "The knife appeared above you when you first stepped inside the doorway." + }, + "answer_label": "D", + "answer_text": "The knife appeared in front of you.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_4_T2", + "correct_answer": "The knife appeared in front of you.", + "original_options": { + "A": "The knife appeared behind you.", + "B": "The knife appeared to your left.", + "C": "The knife appeared to your right as you walked in through the door.", + "D": "The knife appeared in front of you.", + "E": "The knife appeared above you when you first stepped inside the doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 12 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_4_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first walked in the door, where did the knife appear relative to me?", + "answer": "The knife appeared on the right side.", + "question_chinese": "我刚进门时，刀出现在我的哪里？", + "answer_chinese": "刀出现在你的右侧。", + "evidence": [ + { + "description": "There is a knife on the cutting board on the right side of the console.", + "timestamps": { + "start": "00:10:23", + "end": "00:10:26" + } + } + ], + "question_timestamps": "00:10:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The knife appeared on the left side.", + "B": "The knife appeared behind you.", + "C": "The knife appeared on the right side.", + "D": "The knife appeared in front of you when you first walked in the door.", + "E": "The knife appeared above you as you entered through the doorway the first time." + }, + "answer_label": "C", + "answer_text": "The knife appeared on the right side.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_4_T1", + "correct_answer": "The knife appeared on the right side.", + "original_options": { + "A": "The knife appeared on the left side.", + "B": "The knife appeared behind you.", + "C": "The knife appeared on the right side.", + "D": "The knife appeared in front of you when you first walked in the door.", + "E": "The knife appeared above you as you entered through the doorway the first time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 14, + 14 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_6_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I carry things onto the car?", + "answer": "This is the first time you’ve moved something onto the car, and the car is in front of you.", + "question_chinese": "我搬了多少次东西到车上？", + "answer_chinese": "这是你第一次搬东西到车上，车在你的正前方。", + "evidence": [ + { + "description": "first time", + "timestamps": { + "start": "00:12:38", + "end": "00:12:45" + } + } + ], + "question_timestamps": "00:13:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You moved items zero times.", + "B": "This is the first time you’ve moved something onto the car, and the car is in front of you.", + "C": "You moved items two times.", + "D": "You moved items three times, and the car is in front of you.", + "E": "You moved items four times, and the car is still positioned in front of you." + }, + "answer_label": "B", + "answer_text": "This is the first time you’ve moved something onto the car, and the car is in front of you.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_6_T2", + "correct_answer": "This is the first time you’ve moved something onto the car, and the car is in front of you.", + "original_options": { + "A": "You moved items zero times.", + "B": "This is the first time you’ve moved something onto the car, and the car is in front of you.", + "C": "You moved items two times.", + "D": "You moved items three times, and the car is in front of you.", + "E": "You moved items four times, and the car is still positioned in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 15 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_6_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I carry things onto the car?", + "answer": "You moved items two times, and the car is in front of you.", + "question_chinese": "我搬了多少次东西到车上？", + "answer_chinese": "你搬了两次，车在你的正前方。", + "evidence": [ + { + "description": "first time", + "timestamps": { + "start": "00:12:38", + "end": "00:12:45" + } + }, + { + "description": "second time", + "timestamps": { + "start": "00:14:35", + "end": "00:14:48" + } + } + ], + "question_timestamps": "00:15:01", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You moved items two times, and the car is in front of you.", + "B": "You moved items one time.", + "C": "You moved items three times.", + "D": "You moved items zero times, and the car is in front of you.", + "E": "You moved items four times, and the car is in front of you the whole time." + }, + "answer_label": "A", + "answer_text": "You moved items two times, and the car is in front of you.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_6_T1", + "correct_answer": "You moved items two times, and the car is in front of you.", + "original_options": { + "A": "You moved items two times, and the car is in front of you.", + "B": "You moved items one time.", + "C": "You moved items three times.", + "D": "You moved items zero times, and the car is in front of you.", + "E": "You moved items four times, and the car is in front of you the whole time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 16 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_9", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I'm trying to recall now, when I first entered, on which side of me was the knife?", + "answer": "The knife appears on your right side.", + "question_chinese": "我现在想回想一下，我刚进门时，刀出现在我的哪一侧？", + "answer_chinese": "刀出现在你的右侧。", + "evidence": [ + { + "description": "There is a knife on the cutting board on the right side of the console.", + "timestamps": { + "start": "00:10:23", + "end": "00:10:26" + } + } + ], + "question_timestamps": "00:10:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The knife appears on your left side.", + "B": "The knife appears on your right side.", + "C": "The knife appears behind you.", + "D": "The knife appears directly in front of you when you first entered.", + "E": "The knife appears above you relative to your position as you walked in." + }, + "answer_label": "B", + "answer_text": "The knife appears on your right side.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_9", + "correct_answer": "The knife appears on your right side.", + "original_options": { + "A": "The knife appears on your left side.", + "B": "The knife appears on your right side.", + "C": "The knife appears behind you.", + "D": "The knife appears directly in front of you when you first entered.", + "E": "The knife appears above you relative to your position as you walked in." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 12, + 13 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_10", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I'm trying to recall now, when I first entered, on which side of me was the knife?", + "answer": "The knife appears on your right side.", + "question_chinese": "我现在想回想一下，我刚进门时，刀出现在我的哪一侧？", + "answer_chinese": "刀出现在你的右侧。", + "evidence": [ + { + "description": "There is a knife on the cutting board on the right side of the console.", + "timestamps": { + "start": "00:10:23", + "end": "00:10:26" + } + } + ], + "question_timestamps": "00:21:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The knife appears on your left side.", + "B": "The knife appears in front of you.", + "C": "The knife appears behind you when you first entered the space, not beside you.", + "D": "The knife appears above you relative to where you were standing at entry.", + "E": "The knife appears on your right side." + }, + "answer_label": "E", + "answer_text": "The knife appears on your right side.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_10", + "correct_answer": "The knife appears on your right side.", + "original_options": { + "A": "The knife appears on your left side.", + "B": "The knife appears in front of you.", + "C": "The knife appears behind you when you first entered the space, not beside you.", + "D": "The knife appears above you relative to where you were standing at entry.", + "E": "The knife appears on your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 13 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_3_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is my refrigerator?", + "answer": "Your fridge is in front of you.", + "question_chinese": "我的冰箱在哪里？", + "answer_chinese": "你的冰箱在你的前面。", + "evidence": [ + { + "description": "Your refrigerator is in front of you.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:23" + } + } + ], + "question_timestamps": "00:06:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Your fridge is in front of you.", + "B": "Your fridge is behind you.", + "C": "Your fridge is to your left.", + "D": "Your fridge is to your right, just off to the side as you face forward.", + "E": "Your fridge is above you, positioned higher than your eye level in the scene." + }, + "answer_label": "A", + "answer_text": "Your fridge is in front of you.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_3_T1", + "correct_answer": "Your fridge is in front of you.", + "original_options": { + "A": "Your fridge is in front of you.", + "B": "Your fridge is behind you.", + "C": "Your fridge is to your left.", + "D": "Your fridge is to your right, just off to the side as you face forward.", + "E": "Your fridge is above you, positioned higher than your eye level in the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 14 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_3_T2", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is my refrigerator?", + "answer": "Your fridge is behind you.", + "question_chinese": "我的冰箱在哪里？", + "answer_chinese": "你的冰箱在你的后面。", + "evidence": [ + { + "description": "Your refrigerator is behind you.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:36" + } + } + ], + "question_timestamps": "00:06:39", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Your fridge is behind you.", + "B": "Your fridge is in front of you.", + "C": "Your fridge is to your right.", + "D": "Your fridge is to your left, positioned beside you rather than behind you.", + "E": "Your fridge is above you, not behind you, and would be seen higher up." + }, + "answer_label": "A", + "answer_text": "Your fridge is behind you.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_3_T2", + "correct_answer": "Your fridge is behind you.", + "original_options": { + "A": "Your fridge is behind you.", + "B": "Your fridge is in front of you.", + "C": "Your fridge is to your right.", + "D": "Your fridge is to your left, positioned beside you rather than behind you.", + "E": "Your fridge is above you, not behind you, and would be seen higher up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 14 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_2_T2", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times has my refrigerator been opened?", + "answer": "This is the first time your fridge has been opened, and it’s in front of you.", + "question_chinese": "我的冰箱被打开了什么次?", + "answer_chinese": "这是你的冰箱第一次被打开，冰箱在你的正前方。", + "evidence": [ + { + "description": "This is the first time that your refrigerator has been opened.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:19" + } + } + ], + "question_timestamps": "00:06:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Your fridge has been opened zero times.", + "B": "Your fridge has been opened twice.", + "C": "This is the first time your fridge has been opened, and it’s in front of you.", + "D": "Your fridge has been opened three times, and it’s in front of you as you face it.", + "E": "Your fridge has been opened four times, and it’s in front of you in the current view." + }, + "answer_label": "C", + "answer_text": "This is the first time your fridge has been opened, and it’s in front of you.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_2_T2", + "correct_answer": "This is the first time your fridge has been opened, and it’s in front of you.", + "original_options": { + "A": "Your fridge has been opened zero times.", + "B": "Your fridge has been opened twice.", + "C": "This is the first time your fridge has been opened, and it’s in front of you.", + "D": "Your fridge has been opened three times, and it’s in front of you as you face it.", + "E": "Your fridge has been opened four times, and it’s in front of you in the current view." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 18 + ] + }, + { + "video_uid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c", + "video_length": "00:30:00", + "qid": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_2_T1", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times has my refrigerator been opened?", + "answer": "Your fridge has been opened twice, and it’s to your left.", + "question_chinese": "我的冰箱被打开了什么次?", + "answer_chinese": "你的冰箱被打开了两次，冰箱在你的左边。", + "evidence": [ + { + "description": "first time", + "timestamps": { + "start": "00:06:15", + "end": "00:06:19" + } + }, + { + "description": "second time", + "timestamps": { + "start": "00:06:45", + "end": "00:06:46" + } + } + ], + "question_timestamps": "00:06:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Your fridge has been opened once.", + "B": "Your fridge has been opened twice, and it’s to your left.", + "C": "Your fridge has been opened three times.", + "D": "Your fridge has been opened twice, and it’s to your right.", + "E": "Your fridge has been opened four times, and it’s to your left in the scene." + }, + "answer_label": "B", + "answer_text": "Your fridge has been opened twice, and it’s to your left.", + "q_id": "5e7dbc32-08e0-4266-9bd3-27c412d1af7c_2_T1", + "correct_answer": "Your fridge has been opened twice, and it’s to your left.", + "original_options": { + "A": "Your fridge has been opened once.", + "B": "Your fridge has been opened twice, and it’s to your left.", + "C": "Your fridge has been opened three times.", + "D": "Your fridge has been opened twice, and it’s to your right.", + "E": "Your fridge has been opened four times, and it’s to your left in the scene." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 12, + 16 + ] + } +] diff --git a/QAs/5e97e33c-b3ca-48b2-9e7e-069fe39199fb-mcq.json b/QAs/5e97e33c-b3ca-48b2-9e7e-069fe39199fb-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8901c45a1007ca828a62d8d142e6ba25547f698f --- /dev/null +++ b/QAs/5e97e33c-b3ca-48b2-9e7e-069fe39199fb-mcq.json @@ -0,0 +1,736 @@ +[ + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_9", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to drink water, how should I move?", + "answer": "You should turn back, then walk straight for five steps, then turn left, and you will see the water bottle on the stone platform.", + "question_chinese": "如果我想喝水，我应该怎么移动呢？", + "answer_chinese": "你应该向后转，然后直走五步，再左转，就能看到石台上的水瓶。", + "evidence": [ + { + "description": "You took a sip of water, then put it back on the stone platform.", + "timestamps": { + "start": "00:05:11", + "end": "00:05:57" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn back and go four steps.", + "B": "Turn back, then turn right.", + "C": "Turn back, walk straight for five steps, then turn right to find the water bottle.", + "D": "You should turn back, then walk straight for five steps, then turn left, and you will see the water bottle on the stone platform.", + "E": "Turn back, walk straight for six steps, then turn left and look for the bottle on the platform." + }, + "answer_label": "D", + "answer_text": "You should turn back, then walk straight for five steps, then turn left, and you will see the water bottle on the stone platform.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_9", + "correct_answer": "You should turn back, then walk straight for five steps, then turn left, and you will see the water bottle on the stone platform.", + "original_options": { + "A": "Turn back and go four steps.", + "B": "Turn back, then turn right.", + "C": "Turn back, walk straight for five steps, then turn right to find the water bottle.", + "D": "You should turn back, then walk straight for five steps, then turn left, and you will see the water bottle on the stone platform.", + "E": "Turn back, walk straight for six steps, then turn left and look for the bottle on the platform." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 18 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_10", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to drink water, how should I move?", + "answer": "You turn directly right and look down.", + "question_chinese": "如果我想喝水，我应该怎么移动呢？", + "answer_chinese": "你直接右转，低头。", + "evidence": [ + { + "description": "You took a sip of water, then put it back on the stone platform.", + "timestamps": { + "start": "00:05:11", + "end": "00:05:57" + } + }, + { + "description": "The water bottle is on the stone platform on the right side.", + "timestamps": { + "start": "00:29:04", + "end": "00:29:07" + } + } + ], + "question_timestamps": "00:29:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turn directly right and look down.", + "B": "You turn slightly right and look down.", + "C": "You turn directly right and look ahead.", + "D": "You turn directly right, but keep your gaze slightly up instead of looking down.", + "E": "You turn slightly left and look down, rather than turning directly to the right." + }, + "answer_label": "A", + "answer_text": "You turn directly right and look down.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_10", + "correct_answer": "You turn directly right and look down.", + "original_options": { + "A": "You turn directly right and look down.", + "B": "You turn slightly right and look down.", + "C": "You turn directly right and look ahead.", + "D": "You turn directly right, but keep your gaze slightly up instead of looking down.", + "E": "You turn slightly left and look down, rather than turning directly to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 14 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I rode the Half-pipe?", + "answer": "One time.", + "question_chinese": "我刷了多少次半U型池?", + "answer_chinese": "一次。", + "evidence": [ + { + "description": "You came down from the half-pipe.", + "timestamps": { + "start": "00:05:15", + "end": "00:05:29" + } + } + ], + "question_timestamps": "00:05:33", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero times.", + "B": "Two times.", + "C": "Three times overall, counting each time you went down the Half-pipe during the session.", + "D": "Seven times overall, including repeated rides each time you returned to the Half-pipe.", + "E": "One time." + }, + "answer_label": "E", + "answer_text": "One time.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_3", + "correct_answer": "One time.", + "original_options": { + "A": "Zero times.", + "B": "Two times.", + "C": "Three times overall, counting each time you went down the Half-pipe during the session.", + "D": "Seven times overall, including repeated rides each time you returned to the Half-pipe.", + "E": "One time." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I rode the Half-pipe?", + "answer": "Two times.", + "question_chinese": "我刷了多少次半U型池?", + "answer_chinese": "两次", + "evidence": [ + { + "description": "You came down from the half-pipe.", + "timestamps": { + "start": "00:05:15", + "end": "00:05:29" + } + }, + { + "description": "You rode the half-pipe again.", + "timestamps": { + "start": "00:06:09", + "end": "00:06:16" + } + } + ], + "question_timestamps": "00:25:41", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three times.", + "B": "Four times.", + "C": "Five times total, counting every ride you completed on the Half-pipe in this session.", + "D": "Two times.", + "E": "Six times overall, including every ride you took on the Half-pipe across the session." + }, + "answer_label": "D", + "answer_text": "Two times.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_4", + "correct_answer": "Two times.", + "original_options": { + "A": "Three times.", + "B": "Four times.", + "C": "Five times total, counting every ride you completed on the Half-pipe in this session.", + "D": "Two times.", + "E": "Six times overall, including every ride you took on the Half-pipe across the session." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I rode the pump track?", + "answer": "One time.", + "question_chinese": "我刷了多少次泵道？", + "answer_chinese": "一次。", + "evidence": [ + { + "description": "You rode a pump track", + "timestamps": { + "start": "00:02:11", + "end": "00:02:29" + } + } + ], + "question_timestamps": "00:03:06", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Three times.", + "C": "One time.", + "D": "You rode the pump track twice during your session, completing two separate laps around the course.", + "E": "You rode the pump track three times, repeating the loop after returning to the start each time." + }, + "answer_label": "C", + "answer_text": "One time.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_1", + "correct_answer": "One time.", + "original_options": { + "A": "Two times.", + "B": "Three times.", + "C": "One time.", + "D": "You rode the pump track twice during your session, completing two separate laps around the course.", + "E": "You rode the pump track three times, repeating the loop after returning to the start each time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I rode the pump track?", + "answer": "Two times.", + "question_chinese": "我刷了多少次泵道？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "You rode a pump track", + "timestamps": { + "start": "00:02:11", + "end": "00:02:29" + } + }, + { + "description": "You rode the pump track again.", + "timestamps": { + "start": "00:04:27", + "end": "00:04:40" + } + } + ], + "question_timestamps": "00:04:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One time.", + "B": "Two times.", + "C": "Three times.", + "D": "You rode the pump track once, making a single pass before moving on to something else.", + "E": "You rode the pump track four times, circling the track repeatedly and finishing four distinct runs." + }, + "answer_label": "B", + "answer_text": "Two times.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_2", + "correct_answer": "Two times.", + "original_options": { + "A": "One time.", + "B": "Two times.", + "C": "Three times.", + "D": "You rode the pump track once, making a single pass before moving on to something else.", + "E": "You rode the pump track four times, circling the track repeatedly and finishing four distinct runs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_4_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I drink water?", + "answer": "You drank water once. The water was to your right.", + "question_chinese": "我喝了几次水？", + "answer_chinese": "你喝一次水。水在你的右边。", + "evidence": [ + { + "description": "You picked up the canned water and took a sip.", + "timestamps": { + "start": "00:05:36", + "end": "00:05:49" + } + } + ], + "question_timestamps": "00:05:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You drank water twice.", + "B": "You drank water once. The water was to your right.", + "C": "You drank water zero times.", + "D": "You drank water three times, and each time the water bottle stayed to your right.", + "E": "You drank water four times, reaching for the water to your right on every drink." + }, + "answer_label": "B", + "answer_text": "You drank water once. The water was to your right.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_4_T1", + "correct_answer": "You drank water once. The water was to your right.", + "original_options": { + "A": "You drank water twice.", + "B": "You drank water once. The water was to your right.", + "C": "You drank water zero times.", + "D": "You drank water three times, and each time the water bottle stayed to your right.", + "E": "You drank water four times, reaching for the water to your right on every drink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 15 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_4_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I drink water?", + "answer": "You drank water twice. The water was in front of you.", + "question_chinese": "我喝了几次水？", + "answer_chinese": "你喝了两次水。水在你的前面。", + "evidence": [ + { + "description": "You picked up the canned water and took a sip.", + "timestamps": { + "start": "00:05:36", + "end": "00:05:49" + } + }, + { + "description": "Drink water for the second time.", + "timestamps": { + "start": "00:31:02", + "end": "00:31:17" + } + } + ], + "question_timestamps": "00:31:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You drank water once.", + "B": "You drank water three times.", + "C": "You drank water twice. The water was in front of you.", + "D": "You drank water zero times, even though the water remained in front of you the whole time.", + "E": "You drank water four times, with the water positioned directly in front of you each time." + }, + "answer_label": "C", + "answer_text": "You drank water twice. The water was in front of you.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_4_T2", + "correct_answer": "You drank water twice. The water was in front of you.", + "original_options": { + "A": "You drank water once.", + "B": "You drank water three times.", + "C": "You drank water twice. The water was in front of you.", + "D": "You drank water zero times, even though the water remained in front of you the whole time.", + "E": "You drank water four times, with the water positioned directly in front of you each time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 16 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_13", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Who has my skateboard now？", + "answer": "The skateboard is under my feet now.", + "question_chinese": "我的滑板现在在谁哪里？", + "answer_chinese": "滑板现在在我脚下。", + "evidence": [ + { + "description": "You are riding this skateboard now.", + "timestamps": { + "start": "00:14:29", + "end": "00:14:45" + } + } + ], + "question_timestamps": "00:14:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The skateboard is beside you now.", + "B": "The skateboard is in your bag now.", + "C": "The skateboard is leaning next to you right now, not directly under your feet.", + "D": "The skateboard is on the floor nearby now, rather than being positioned beneath your feet.", + "E": "The skateboard is under my feet now." + }, + "answer_label": "E", + "answer_text": "The skateboard is under my feet now.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_13", + "correct_answer": "The skateboard is under my feet now.", + "original_options": { + "A": "The skateboard is beside you now.", + "B": "The skateboard is in your bag now.", + "C": "The skateboard is leaning next to you right now, not directly under your feet.", + "D": "The skateboard is on the floor nearby now, rather than being positioned beneath your feet.", + "E": "The skateboard is under my feet now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 15 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_14", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Who has my skateboard now？", + "answer": "Your skateboard is in your friend's hands.", + "question_chinese": "我的滑板现在在谁哪里？", + "answer_chinese": "你的朋友的手上。", + "evidence": [ + { + "description": "Your friend is riding this skateboard.", + "timestamps": { + "start": "00:21:29", + "end": "00:21:33" + } + } + ], + "question_timestamps": "00:22:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Your skateboard is in your friend's hands.", + "B": "Your skateboard is in your brother's hands.", + "C": "Your skateboard is in your friend's backpack.", + "D": "Your skateboard is being held by your sister right now, not by your friend.", + "E": "Your skateboard is in your neighbor's hands at the moment, instead of your friend's hands." + }, + "answer_label": "A", + "answer_text": "Your skateboard is in your friend's hands.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_14", + "correct_answer": "Your skateboard is in your friend's hands.", + "original_options": { + "A": "Your skateboard is in your friend's hands.", + "B": "Your skateboard is in your brother's hands.", + "C": "Your skateboard is in your friend's backpack.", + "D": "Your skateboard is being held by your sister right now, not by your friend.", + "E": "Your skateboard is in your neighbor's hands at the moment, instead of your friend's hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 15 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_6_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach out and pick up my phone?", + "answer": "You couldn’t reach out and grab the phone. The phone was to your front-right.", + "question_chinese": "我能伸手拿到我的手机吗？", + "answer_chinese": "你不可以伸手拿到手机。手机在你的右前方。", + "evidence": [ + { + "description": "The phone is on the low stone step.", + "timestamps": { + "start": "00:12:30", + "end": "00:12:34" + } + } + ], + "question_timestamps": "00:12:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You could reach out and grab your phone. The phone was to your front-right.", + "B": "You couldn’t reach out and grab the phone. The phone was to your front-right." + }, + "answer_label": "B", + "answer_text": "You couldn’t reach out and grab the phone. The phone was to your front-right.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_6_T1", + "correct_answer": "You couldn’t reach out and grab the phone. The phone was to your front-right.", + "original_options": { + "A": "You could reach out and grab your phone. The phone was to your front-right.", + "B": "You couldn’t reach out and grab the phone. The phone was to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_11", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach my phone?", + "answer": "You can reach for your phone.", + "question_chinese": "我能伸手拿到我的手机吗？", + "answer_chinese": "你可以伸手拿到手机。", + "evidence": [ + { + "description": "The phone is on the low stone step.", + "timestamps": { + "start": "00:12:30", + "end": "00:12:37" + } + } + ], + "question_timestamps": "00:12:54", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can reach for your phone.", + "B": "You cannot reach for your phone." + }, + "answer_label": "A", + "answer_text": "You can reach for your phone.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_11", + "correct_answer": "You can reach for your phone.", + "original_options": { + "A": "You can reach for your phone.", + "B": "You cannot reach for your phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_6_T2", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach out and pick up my phone?", + "answer": "You could reach out and grab your phone. The phone was in front of you.", + "question_chinese": "我能伸手拿到我的手机吗？", + "answer_chinese": "你可以伸手拿到你的手机。手机在你的前面。", + "evidence": [ + { + "description": "The phone was put on your body.", + "timestamps": { + "start": "00:18:20", + "end": "00:18:50" + } + } + ], + "question_timestamps": "00:18:52", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You could reach out and grab your phone. The phone was in front of you.", + "B": "You couldn’t reach out and grab your phone. It was too far in front of you." + }, + "answer_label": "A", + "answer_text": "You could reach out and grab your phone. The phone was in front of you.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_6_T2", + "correct_answer": "You could reach out and grab your phone. The phone was in front of you.", + "original_options": { + "A": "You could reach out and grab your phone. The phone was in front of you.", + "B": "You couldn’t reach out and grab your phone. It was too far in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb", + "video_length": "00:38:06", + "qid": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_12", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach my phone?", + "answer": "You can reach for your phone.", + "question_chinese": "我能伸手拿到我的手机吗？", + "answer_chinese": "可以的。", + "evidence": [ + { + "description": "The phone was put on your body.", + "timestamps": { + "start": "00:18:20", + "end": "00:18:56" + } + } + ], + "question_timestamps": "00:33:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can reach for your phone.", + "B": "You cannot reach for your phone." + }, + "answer_label": "A", + "answer_text": "You can reach for your phone.", + "q_id": "5e97e33c-b3ca-48b2-9e7e-069fe39199fb_12", + "correct_answer": "You can reach for your phone.", + "original_options": { + "A": "You can reach for your phone.", + "B": "You cannot reach for your phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/5ed59c41-af9e-4368-a0ce-30d536d34895-mcq.json b/QAs/5ed59c41-af9e-4368-a0ce-30d536d34895-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d7d17037378ebd631d6ffe5a5f88d5979a1636b3 --- /dev/null +++ b/QAs/5ed59c41-af9e-4368-a0ce-30d536d34895-mcq.json @@ -0,0 +1,923 @@ +[ + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_9", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cuts did I make after taking out the bread from the oven?", + "answer": "You made three cuts.", + "question_chinese": "从烤箱拿出来面包后，我切了几刀？", + "answer_chinese": "你切了三刀。", + "evidence": [ + { + "description": "You made the first cut.", + "timestamps": { + "start": "00:05:47", + "end": "00:05:53" + } + }, + { + "description": "You made the second cut.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:04" + } + }, + { + "description": "You made the third cut.", + "timestamps": { + "start": "00:06:17", + "end": "00:06:21" + } + } + ], + "question_timestamps": "00:06:25", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You made three cuts.", + "B": "You made two cuts.", + "C": "You made four cuts.", + "D": "You made five cuts in total after pulling the bread out of the oven.", + "E": "After taking the bread out of the oven, you ended up making six cuts." + }, + "answer_label": "A", + "answer_text": "You made three cuts.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_9", + "correct_answer": "You made three cuts.", + "original_options": { + "A": "You made three cuts.", + "B": "You made two cuts.", + "C": "You made four cuts.", + "D": "You made five cuts in total after pulling the bread out of the oven.", + "E": "After taking the bread out of the oven, you ended up making six cuts." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_10", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cuts did I make after taking out the bread from the oven?", + "answer": "You made seven cuts.", + "question_chinese": "从烤箱拿出来面包后，我切了几刀？", + "answer_chinese": "你切了七刀。", + "evidence": [ + { + "description": "You made the first cut.", + "timestamps": { + "start": "00:05:47", + "end": "00:05:53" + } + }, + { + "description": "You made the second cut.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:04" + } + }, + { + "description": "You made the third cut.", + "timestamps": { + "start": "00:06:17", + "end": "00:06:21" + } + }, + { + "description": "You made the fourth cut.", + "timestamps": { + "start": "00:06:26", + "end": "00:06:32" + } + }, + { + "description": "You made the fifth cut.", + "timestamps": { + "start": "00:06:44", + "end": "00:06:49" + } + }, + { + "description": "You made the sixth cut.", + "timestamps": { + "start": "00:07:03", + "end": "00:07:11" + } + }, + { + "description": "You made the seventh cut.", + "timestamps": { + "start": "00:07:29", + "end": "00:08:02" + } + } + ], + "question_timestamps": "00:35:22", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You made six cuts.", + "B": "You made eight cuts.", + "C": "After taking the bread out of the oven, you made nine cuts in total.", + "D": "You made seven cuts.", + "E": "You ended up making ten cuts after you pulled the bread out of the oven." + }, + "answer_label": "D", + "answer_text": "You made seven cuts.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_10", + "correct_answer": "You made seven cuts.", + "original_options": { + "A": "You made six cuts.", + "B": "You made eight cuts.", + "C": "After taking the bread out of the oven, you made nine cuts in total.", + "D": "You made seven cuts.", + "E": "You ended up making ten cuts after you pulled the bread out of the oven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_1_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Was the refrigerator to my left?", + "answer": "No—the refrigerator is behind you.", + "question_chinese": "冰箱是在我的左边吗？", + "answer_chinese": "不是的，冰箱在你的后方", + "evidence": [ + { + "description": "The refrigerator is on your left.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, the refrigerator is to your left.", + "B": "No—the refrigerator is behind you." + }, + "answer_label": "B", + "answer_text": "No—the refrigerator is behind you.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_1_T1", + "correct_answer": "No—the refrigerator is behind you.", + "original_options": { + "A": "Yes, the refrigerator is to your left.", + "B": "No—the refrigerator is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_1_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Was the refrigerator to my left?", + "answer": "Yes, the refrigerator is to your left.", + "question_chinese": "冰箱是在我的左边吗？", + "answer_chinese": "是的，冰箱在你的左边。", + "evidence": [ + { + "description": "The refrigerator is on your left.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:01:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, the refrigerator is to your left.", + "B": "No—the refrigerator is behind you." + }, + "answer_label": "A", + "answer_text": "Yes, the refrigerator is to your left.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_1_T2", + "correct_answer": "Yes, the refrigerator is to your left.", + "original_options": { + "A": "Yes, the refrigerator is to your left.", + "B": "No—the refrigerator is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_15", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I taken out butter from the fridge so far?", + "answer": "You took out butter from the fridge once.", + "question_chinese": "到现在为止，我一共从冰箱拿出了几次黄油？", + "answer_chinese": "你从冰箱拿出了一次黄油。", + "evidence": [ + { + "description": "The first time you take out butter from the fridge.", + "timestamps": { + "start": "00:09:50", + "end": "00:09:52" + } + } + ], + "question_timestamps": "00:09:57", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You took it out twice.", + "B": "You took out butter from the fridge once.", + "C": "You took it out three times.", + "D": "You took out butter from the fridge four times up to this point in the video.", + "E": "So far, you have taken butter out of the fridge six times during the video." + }, + "answer_label": "B", + "answer_text": "You took out butter from the fridge once.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_15", + "correct_answer": "You took out butter from the fridge once.", + "original_options": { + "A": "You took it out twice.", + "B": "You took out butter from the fridge once.", + "C": "You took it out three times.", + "D": "You took out butter from the fridge four times up to this point in the video.", + "E": "So far, you have taken butter out of the fridge six times during the video." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_16", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I taken out butter from the fridge so far?", + "answer": "You took out butter from the fridge twice.", + "question_chinese": "到现在为止，我一共从冰箱拿出了几次黄油？", + "answer_chinese": "你从冰箱拿出了两次黄油。", + "evidence": [ + { + "description": "The first time you take out butter from the fridge.", + "timestamps": { + "start": "00:09:50", + "end": "00:09:52" + } + }, + { + "description": "You take out butter from the fridge for the second time.", + "timestamps": { + "start": "00:18:15", + "end": "00:18:20" + } + } + ], + "question_timestamps": "00:22:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You took it out once.", + "B": "You took it out three times.", + "C": "You took out butter from the fridge four times up to this point in the video.", + "D": "So far, you have taken butter out of the fridge six times during the video.", + "E": "You took out butter from the fridge twice." + }, + "answer_label": "E", + "answer_text": "You took out butter from the fridge twice.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_16", + "correct_answer": "You took out butter from the fridge twice.", + "original_options": { + "A": "You took it out once.", + "B": "You took it out three times.", + "C": "You took out butter from the fridge four times up to this point in the video.", + "D": "So far, you have taken butter out of the fridge six times during the video.", + "E": "You took out butter from the fridge twice." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_3", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I just took two steps forward, which thing is closest to me now?", + "answer": "A spoon is closest to you.", + "question_chinese": "我刚向前走了两步，现在哪个东西离我最近？", + "answer_chinese": "一个勺子离你最近。", + "evidence": [ + { + "description": "You take the instant noodles forward, they are behind the spoon.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A fork is closest to you.", + "B": "A spoon is closest to you.", + "C": "A cup is closest to you.", + "D": "A knife is closest to you after you step forward two paces.", + "E": "A napkin ends up being the closest item once you move ahead two steps." + }, + "answer_label": "B", + "answer_text": "A spoon is closest to you.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_3", + "correct_answer": "A spoon is closest to you.", + "original_options": { + "A": "A fork is closest to you.", + "B": "A spoon is closest to you.", + "C": "A cup is closest to you.", + "D": "A knife is closest to you after you step forward two paces.", + "E": "A napkin ends up being the closest item once you move ahead two steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 14 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_4", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I just took two steps forward, which thing is closest to me now?", + "answer": "The refrigerator is closest to you.", + "question_chinese": "我刚向前走了两步，现在哪个东西离我最近？", + "answer_chinese": "冰箱离你最近。", + "evidence": [ + { + "description": "You took two steps forward and arrived at the refrigerator.", + "timestamps": { + "start": "00:21:11", + "end": "00:21:13" + } + } + ], + "question_timestamps": "00:21:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The table is closest to you.", + "B": "The chair is closest to you.", + "C": "After taking two steps forward, the cupboard is the nearest thing to where you are standing.", + "D": "Once you move forward two steps, the washing machine becomes the closest object to you.", + "E": "The refrigerator is closest to you." + }, + "answer_label": "E", + "answer_text": "The refrigerator is closest to you.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_4", + "correct_answer": "The refrigerator is closest to you.", + "original_options": { + "A": "The table is closest to you.", + "B": "The chair is closest to you.", + "C": "After taking two steps forward, the cupboard is the nearest thing to where you are standing.", + "D": "Once you move forward two steps, the washing machine becomes the closest object to you.", + "E": "The refrigerator is closest to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_9_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I went for a walk, how many turns did I make in total along the way?", + "answer": "You made a total of 2 turns, and the route is all to your back-left.", + "question_chinese": "我去散步，在路上，一共拐了几个弯？", + "answer_chinese": "你一共拐了2个弯，道路都在你的左后方。", + "evidence": [ + { + "description": "You took the first turn.", + "timestamps": { + "start": "00:28:56", + "end": "00:29:03" + } + }, + { + "description": "You took the second turn.", + "timestamps": { + "start": "00:29:49", + "end": "00:29:55" + } + } + ], + "question_timestamps": "00:30:01", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You made 0 turns total.", + "B": "You made a total of 2 turns, and the route is all to your back-left.", + "C": "You made 1 turn total.", + "D": "You made a total of 3 turns, and the route is all to your back-left.", + "E": "You made a total of 4 turns, and the route is all to your back-left." + }, + "answer_label": "B", + "answer_text": "You made a total of 2 turns, and the route is all to your back-left.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_9_T2", + "correct_answer": "You made a total of 2 turns, and the route is all to your back-left.", + "original_options": { + "A": "You made 0 turns total.", + "B": "You made a total of 2 turns, and the route is all to your back-left.", + "C": "You made 1 turn total.", + "D": "You made a total of 3 turns, and the route is all to your back-left.", + "E": "You made a total of 4 turns, and the route is all to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_9_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I went for a walk, how many turns did I make in total along the way?", + "answer": "You made a total of 4 turns.", + "question_chinese": "我去散步，在路上，一共拐了几个弯？", + "answer_chinese": "你一共拐了4个弯。", + "evidence": [ + { + "description": "You took the first turn.", + "timestamps": { + "start": "00:28:56", + "end": "00:29:03" + } + }, + { + "description": "You took the second turn.", + "timestamps": { + "start": "00:29:49", + "end": "00:29:55" + } + }, + { + "description": "You took the third turn.", + "timestamps": { + "start": "00:30:11", + "end": "00:30:16" + } + }, + { + "description": "You took the fourth turn.", + "timestamps": { + "start": "00:30:35", + "end": "00:30:40" + } + } + ], + "question_timestamps": "00:31:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You made a total of 4 turns.", + "B": "You made 2 turns total.", + "C": "You made 3 turns total.", + "D": "You made a total of 5 turns along the way during the walk.", + "E": "You made a total of 6 turns during the walk from start to finish." + }, + "answer_label": "A", + "answer_text": "You made a total of 4 turns.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_9_T1", + "correct_answer": "You made a total of 4 turns.", + "original_options": { + "A": "You made a total of 4 turns.", + "B": "You made 2 turns total.", + "C": "You made 3 turns total.", + "D": "You made a total of 5 turns along the way during the walk.", + "E": "You made a total of 6 turns during the walk from start to finish." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_13", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I put the plate after finishing the bread?", + "answer": "You placed the plate on the dining table to your left.", + "question_chinese": "我吃完面包把盘子放在了哪里？", + "answer_chinese": "你把盘子放在了你左侧的餐桌上。", + "evidence": [ + { + "description": "The orange plate is on the dining table.", + "timestamps": { + "start": "00:14:51", + "end": "00:14:54" + } + } + ], + "question_timestamps": "00:14:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it on the coffee table.", + "B": "You set it on the kitchen island.", + "C": "You left the plate on the sideboard to your left after you finished the bread.", + "D": "You placed the plate on the dining table to your left.", + "E": "You put the plate on the breakfast bar behind you once you were done eating." + }, + "answer_label": "D", + "answer_text": "You placed the plate on the dining table to your left.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_13", + "correct_answer": "You placed the plate on the dining table to your left.", + "original_options": { + "A": "You put it on the coffee table.", + "B": "You set it on the kitchen island.", + "C": "You left the plate on the sideboard to your left after you finished the bread.", + "D": "You placed the plate on the dining table to your left.", + "E": "You put the plate on the breakfast bar behind you once you were done eating." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_14", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I put the plate after finishing the bread?", + "answer": "You put the plate in the sink.", + "question_chinese": "我吃完面包把盘子放在了哪里？", + "answer_chinese": "你把盘子放在了水槽里。", + "evidence": [ + { + "description": "The orange plate is now in the sink on the left.", + "timestamps": { + "start": "00:17:33", + "end": "00:17:34" + } + } + ], + "question_timestamps": "00:28:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it on the counter.", + "B": "You placed it in the dishwasher.", + "C": "You left the plate on the kitchen table after finishing the bread.", + "D": "You put the plate down on the stove instead of washing it right away.", + "E": "You put the plate in the sink." + }, + "answer_label": "E", + "answer_text": "You put the plate in the sink.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_14", + "correct_answer": "You put the plate in the sink.", + "original_options": { + "A": "You put it on the counter.", + "B": "You placed it in the dishwasher.", + "C": "You left the plate on the kitchen table after finishing the bread.", + "D": "You put the plate down on the stove instead of washing it right away.", + "E": "You put the plate in the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 14 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_4_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the small knife in my hand take from the drawer to where it is now?", + "answer": "The small knife was taken out from the outer side of the drawer and moved upward, and it’s now placed on the workbench by your right hand.", + "question_chinese": "我手上的小刀从抽屉到现在的位置经过了什么路径?", + "answer_chinese": "小刀从抽屉靠外侧取出,向上移动,现在放置在你右手边的操作台上。", + "evidence": [ + { + "description": "The small knife moves upwards.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:17" + } + } + ], + "question_timestamps": "00:04:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Taken from inner side, moved upward.", + "B": "The small knife was taken out from the outer side of the drawer and moved upward, and it’s now placed on the workbench by your right hand.", + "C": "Taken from outer side, moved downward.", + "D": "Taken from the outer side, moved up, then placed on the workbench by your left hand.", + "E": "Taken from the outer side, moved up, and placed on the workbench to your front-left." + }, + "answer_label": "B", + "answer_text": "The small knife was taken out from the outer side of the drawer and moved upward, and it’s now placed on the workbench by your right hand.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_4_T2", + "correct_answer": "The small knife was taken out from the outer side of the drawer and moved upward, and it’s now placed on the workbench by your right hand.", + "original_options": { + "A": "Taken from inner side, moved upward.", + "B": "The small knife was taken out from the outer side of the drawer and moved upward, and it’s now placed on the workbench by your right hand.", + "C": "Taken from outer side, moved downward.", + "D": "Taken from the outer side, moved up, then placed on the workbench by your left hand.", + "E": "Taken from the outer side, moved up, and placed on the workbench to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_4_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the small knife in my hand take from the drawer to where it is now?", + "answer": "The small knife was taken out from the outer side of the drawer and moved upward, and it’s now placed on the workbench to your front-left.", + "question_chinese": "我手上的小刀从抽屉到现在的位置经过了什么路径?", + "answer_chinese": "小刀从抽屉靠外侧取出,向上移动,现在放置在你左前方的操作台上。", + "evidence": [ + { + "description": "The small knife moves upwards.The small knife is placed on the operating table. The small knife was taken out from the outer side of the drawer.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:17" + } + } + ], + "question_timestamps": "00:05:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The small knife was taken out from the outer side of the drawer and moved upward, and it’s now placed on the workbench to your front-left.", + "B": "Moved downward, placed front-left.", + "C": "Taken from inner side, moved upward.", + "D": "Taken from the outer side, moved up, and placed on the workbench to your front-right.", + "E": "Taken from the outer side, moved up, and placed on the workbench to your back-left." + }, + "answer_label": "A", + "answer_text": "The small knife was taken out from the outer side of the drawer and moved upward, and it’s now placed on the workbench to your front-left.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_4_T1", + "correct_answer": "The small knife was taken out from the outer side of the drawer and moved upward, and it’s now placed on the workbench to your front-left.", + "original_options": { + "A": "The small knife was taken out from the outer side of the drawer and moved upward, and it’s now placed on the workbench to your front-left.", + "B": "Moved downward, placed front-left.", + "C": "Taken from inner side, moved upward.", + "D": "Taken from the outer side, moved up, and placed on the workbench to your front-right.", + "E": "Taken from the outer side, moved up, and placed on the workbench to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 15 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_11", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You was holding butter, a knife, and a kettle in your hand.", + "question_chinese": "我手里刚刚拿着什么东西？", + "answer_chinese": "你手里拿着黄油、小刀和水壶。", + "evidence": [ + { + "description": "You hold a box of butter and a small knife in your left hand, and a kettle in your right hand.", + "timestamps": { + "start": "00:10:08", + "end": "00:10:14" + } + } + ], + "question_timestamps": "00:10:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You was holding butter, a knife, and a kettle in your hand.", + "B": "You was holding butter and a knife.", + "C": "You was holding a knife and a kettle.", + "D": "You was holding margarine, a knife, and a kettle in your hand just now.", + "E": "You was holding butter, a spoon, and a kettle in your hand a moment ago." + }, + "answer_label": "A", + "answer_text": "You was holding butter, a knife, and a kettle in your hand.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_11", + "correct_answer": "You was holding butter, a knife, and a kettle in your hand.", + "original_options": { + "A": "You was holding butter, a knife, and a kettle in your hand.", + "B": "You was holding butter and a knife.", + "C": "You was holding a knife and a kettle.", + "D": "You was holding margarine, a knife, and a kettle in your hand just now.", + "E": "You was holding butter, a spoon, and a kettle in your hand a moment ago." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 14, + 15 + ] + }, + { + "video_uid": "5ed59c41-af9e-4368-a0ce-30d536d34895", + "video_length": "00:57:26", + "qid": "5ed59c41-af9e-4368-a0ce-30d536d34895_12", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You was pushing a baby stroller in your hand.", + "question_chinese": "我手里刚刚拿着什么东西？", + "answer_chinese": "你手里推着婴儿车。", + "evidence": [ + { + "description": "You are holding the baby stroller with both hands.", + "timestamps": { + "start": "00:24:02", + "end": "00:24:07" + } + } + ], + "question_timestamps": "00:24:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You was holding a toy car.", + "B": "You was pushing a baby stroller in your hand.", + "C": "You was holding a coffee mug.", + "D": "You was holding a notebook and a pencil in your hand just now.", + "E": "You was holding a water bottle in your hand while walking forward." + }, + "answer_label": "B", + "answer_text": "You was pushing a baby stroller in your hand.", + "q_id": "5ed59c41-af9e-4368-a0ce-30d536d34895_12", + "correct_answer": "You was pushing a baby stroller in your hand.", + "original_options": { + "A": "You was holding a toy car.", + "B": "You was pushing a baby stroller in your hand.", + "C": "You was holding a coffee mug.", + "D": "You was holding a notebook and a pencil in your hand just now.", + "E": "You was holding a water bottle in your hand while walking forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 12 + ] + } +] diff --git a/QAs/6a5fa00b-64de-4851-98fa-7e96f3f441f6-mcq.json b/QAs/6a5fa00b-64de-4851-98fa-7e96f3f441f6-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4505cea3e303b40dc90b578af95140a388cb1a15 --- /dev/null +++ b/QAs/6a5fa00b-64de-4851-98fa-7e96f3f441f6-mcq.json @@ -0,0 +1,1267 @@ +[ + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_15", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I take while throwing out the trash?", + "answer": "three.", + "question_chinese": "我丢垃圾拐了多少个弯？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:15:57", + "end": "00:15:58" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:16:24", + "end": "00:16:25" + } + }, + { + "description": "tghe third one", + "timestamps": { + "start": "00:17:02", + "end": "00:17:03" + } + } + ], + "question_timestamps": "00:17:04", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "one.", + "B": "two.", + "C": "You took one turn while throwing out the trash in the hallway before reaching the bin.", + "D": "You took four turns while throwing out the trash, turning several times before you arrived outside.", + "E": "three." + }, + "answer_label": "E", + "answer_text": "three.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_15", + "correct_answer": "three.", + "original_options": { + "A": "one.", + "B": "two.", + "C": "You took one turn while throwing out the trash in the hallway before reaching the bin.", + "D": "You took four turns while throwing out the trash, turning several times before you arrived outside.", + "E": "three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_8_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I went to throw out the trash, how many turns did I make?", + "answer": "When you took out the trash, you made two turns.", + "question_chinese": "我丢垃圾拐了多少个弯？", + "answer_chinese": "你丢垃圾拐了2个弯", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:15:57", + "end": "00:15:58" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:16:24", + "end": "00:16:25" + } + } + ], + "question_timestamps": "00:17:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero turns.", + "B": "When you took out the trash, you made two turns.", + "C": "One turn.", + "D": "When you took out the trash, you made three turns in total before reaching the trash area.", + "E": "When you took out the trash, you made four turns before finishing and returning to the starting point." + }, + "answer_label": "B", + "answer_text": "When you took out the trash, you made two turns.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_8_T1", + "correct_answer": "When you took out the trash, you made two turns.", + "original_options": { + "A": "Zero turns.", + "B": "When you took out the trash, you made two turns.", + "C": "One turn.", + "D": "When you took out the trash, you made three turns in total before reaching the trash area.", + "E": "When you took out the trash, you made four turns before finishing and returning to the starting point." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_16", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I take while throwing out the trash?", + "answer": "three.", + "question_chinese": "我丢垃圾拐了多少个弯？", + "answer_chinese": "3", + "evidence": [ + { + "description": "tghe first one", + "timestamps": { + "start": "00:15:57", + "end": "00:17:58" + } + }, + { + "description": "tghe second one", + "timestamps": { + "start": "00:16:24", + "end": "00:16:25" + } + }, + { + "description": "tghe third one", + "timestamps": { + "start": "00:17:02", + "end": "00:17:03" + } + } + ], + "question_timestamps": "00:17:59", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "one.", + "B": "two.", + "C": "You took one turn while throwing out the trash, then walked straight to the bin afterward.", + "D": "You took four turns while throwing out the trash, changing direction multiple times before you arrived.", + "E": "three." + }, + "answer_label": "E", + "answer_text": "three.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_16", + "correct_answer": "three.", + "original_options": { + "A": "one.", + "B": "two.", + "C": "You took one turn while throwing out the trash, then walked straight to the bin afterward.", + "D": "You took four turns while throwing out the trash, changing direction multiple times before you arrived.", + "E": "three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_8_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I went to throw out the trash, how many turns did I make?", + "answer": "You made three turns.", + "question_chinese": "我丢垃圾拐了多少个弯？", + "answer_chinese": "你拐了3个弯", + "evidence": [ + { + "description": "tghe first one", + "timestamps": { + "start": "00:15:57", + "end": "00:17:58" + } + } + ], + "question_timestamps": "00:18:01", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You made one turn.", + "B": "You made two turns.", + "C": "You made three turns.", + "D": "You made four turns while going to throw out the trash, turning repeatedly before reaching the bin.", + "E": "You made five turns while going to throw out the trash, with extra turns before you arrived outside." + }, + "answer_label": "C", + "answer_text": "You made three turns.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_8_T2", + "correct_answer": "You made three turns.", + "original_options": { + "A": "You made one turn.", + "B": "You made two turns.", + "C": "You made three turns.", + "D": "You made four turns while going to throw out the trash, turning repeatedly before reaching the bin.", + "E": "You made five turns while going to throw out the trash, with extra turns before you arrived outside." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_5", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get a sweet potato?", + "answer": "Sure, the sweet potato is in front of me.", + "question_chinese": "我可以拿到红薯吗？", + "answer_chinese": "可以，红薯在我的面前", + "evidence": [ + { + "description": "the sweet potato is in front of me.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:27" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, the sweet potato is behind me.", + "B": "Sure, the sweet potato is in front of me." + }, + "answer_label": "B", + "answer_text": "Sure, the sweet potato is in front of me.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_5", + "correct_answer": "Sure, the sweet potato is in front of me.", + "original_options": { + "A": "No, the sweet potato is behind me.", + "B": "Sure, the sweet potato is in front of me." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_3_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the sweet potato?", + "answer": "Yes—you can. The sweet potato is in front of you.", + "question_chinese": "我可以拿到红薯吗？", + "answer_chinese": "可以，红薯在我的面前", + "evidence": [ + { + "description": "the sweet potato is in front of me", + "timestamps": { + "start": "00:01:26", + "end": "00:01:27" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. The sweet potato is in front of you.", + "B": "No, you can’t, because the sweet potato is behind you." + }, + "answer_label": "A", + "answer_text": "Yes—you can. The sweet potato is in front of you.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_3_T2", + "correct_answer": "Yes—you can. The sweet potato is in front of you.", + "original_options": { + "A": "Yes—you can. The sweet potato is in front of you.", + "B": "No, you can’t, because the sweet potato is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_3_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the sweet potato?", + "answer": "No, you can’t, because the sweet potato is behind you.", + "question_chinese": "我可以拿到红薯吗？", + "answer_chinese": "不可以，因为红薯在你的后面。", + "evidence": [ + { + "description": "Sweet potatoes are behind you.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:36", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can, because the sweet potato is in front of you.", + "B": "No, you can’t, because the sweet potato is behind you." + }, + "answer_label": "B", + "answer_text": "No, you can’t, because the sweet potato is behind you.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_3_T1", + "correct_answer": "No, you can’t, because the sweet potato is behind you.", + "original_options": { + "A": "Yes, you can, because the sweet potato is in front of you.", + "B": "No, you can’t, because the sweet potato is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_7_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I open the refrigerator for?", + "answer": "You picked up a red box.", + "question_chinese": "我开冰箱干什么？", + "answer_chinese": "你拿了一个红色的盒子。", + "evidence": [ + { + "description": "You picked up a red box.", + "timestamps": { + "start": "00:07:13", + "end": "00:07:15" + } + } + ], + "question_timestamps": "00:07:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To grab the butter.", + "B": "You picked up a red box.", + "C": "To put away the curry.", + "D": "You opened it to put the butter inside, then you closed the door without taking anything out.", + "E": "You opened the fridge to check what was inside, but you didn’t pick up the red box." + }, + "answer_label": "B", + "answer_text": "You picked up a red box.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_7_T1", + "correct_answer": "You picked up a red box.", + "original_options": { + "A": "To grab the butter.", + "B": "You picked up a red box.", + "C": "To put away the curry.", + "D": "You opened it to put the butter inside, then you closed the door without taking anything out.", + "E": "You opened the fridge to check what was inside, but you didn’t pick up the red box." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 18 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_13", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Why did I open the fridge?", + "answer": "Put in butter and get the curry.", + "question_chinese": "我开冰箱干什么？", + "answer_chinese": "放黄油，拿咖喱", + "evidence": [ + { + "description": "Put in butter", + "timestamps": { + "start": "00:14:20", + "end": "00:14:21" + } + }, + { + "description": "get the curry", + "timestamps": { + "start": "00:14:22", + "end": "00:14:23" + } + } + ], + "question_timestamps": "00:14:24", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Put in butter and get the curry.", + "B": "Put in cream and get curry.", + "C": "Put in butter and get stew.", + "D": "You opened the fridge to put in ghee, then you grabbed the curry afterward.", + "E": "You opened it to put butter away, but you took out dal instead of curry." + }, + "answer_label": "A", + "answer_text": "Put in butter and get the curry.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_13", + "correct_answer": "Put in butter and get the curry.", + "original_options": { + "A": "Put in butter and get the curry.", + "B": "Put in cream and get curry.", + "C": "Put in butter and get stew.", + "D": "You opened the fridge to put in ghee, then you grabbed the curry afterward.", + "E": "You opened it to put butter away, but you took out dal instead of curry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_7_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I open the refrigerator for?", + "answer": "Put down the butter, then pick up the curry.", + "question_chinese": "我开冰箱干什么？", + "answer_chinese": "放黄油，拿咖喱", + "evidence": [ + { + "description": "Put in butter", + "timestamps": { + "start": "00:14:20", + "end": "00:14:21" + } + }, + { + "description": "get the curry", + "timestamps": { + "start": "00:14:22", + "end": "00:14:23" + } + } + ], + "question_timestamps": "00:14:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked up a red box.", + "B": "Put in butter and get curry.", + "C": "Put down the butter, then pick up the curry.", + "D": "You opened it to put the curry back, then you picked up the butter afterward.", + "E": "You opened the refrigerator to set the butter inside, then you closed it without taking the curry." + }, + "answer_label": "C", + "answer_text": "Put down the butter, then pick up the curry.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_7_T2", + "correct_answer": "Put down the butter, then pick up the curry.", + "original_options": { + "A": "You picked up a red box.", + "B": "Put in butter and get curry.", + "C": "Put down the butter, then pick up the curry.", + "D": "You opened it to put the curry back, then you picked up the butter afterward.", + "E": "You opened the refrigerator to set the butter inside, then you closed it without taking the curry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_3", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the oil bottle?", + "answer": "In front of the microwave.", + "question_chinese": "我把油瓶放在了哪里？", + "answer_chinese": "微波炉前面", + "evidence": [ + { + "description": "In front of the microwave", + "timestamps": { + "start": "00:01:18", + "end": "00:01:19" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind the microwave.", + "B": "Next to the microwave.", + "C": "You put the oil bottle on top of the microwave instead of placing it in front.", + "D": "You left the oil bottle on the counter beside the microwave, not directly in front of it.", + "E": "In front of the microwave." + }, + "answer_label": "E", + "answer_text": "In front of the microwave.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_3", + "correct_answer": "In front of the microwave.", + "original_options": { + "A": "Behind the microwave.", + "B": "Next to the microwave.", + "C": "You put the oil bottle on top of the microwave instead of placing it in front.", + "D": "You left the oil bottle on the counter beside the microwave, not directly in front of it.", + "E": "In front of the microwave." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 17 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_2_T1", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the oil bottle?", + "answer": "You put the oil bottle in front of the microwave, and the microwave is behind you.", + "question_chinese": "我把油瓶放在了哪里？", + "answer_chinese": "你把油瓶微波炉前面，微波炉在你后面。", + "evidence": [ + { + "description": "In front of the microwave", + "timestamps": { + "start": "00:01:18", + "end": "00:01:19" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind the microwave.", + "B": "You put the oil bottle in front of the microwave, and the microwave is behind you.", + "C": "Next to the microwave.", + "D": "You put the oil bottle behind the microwave instead of placing it in front of it.", + "E": "You left the oil bottle beside the microwave on the counter, not directly in front of it." + }, + "answer_label": "B", + "answer_text": "You put the oil bottle in front of the microwave, and the microwave is behind you.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_2_T1", + "correct_answer": "You put the oil bottle in front of the microwave, and the microwave is behind you.", + "original_options": { + "A": "Behind the microwave.", + "B": "You put the oil bottle in front of the microwave, and the microwave is behind you.", + "C": "Next to the microwave.", + "D": "You put the oil bottle behind the microwave instead of placing it in front of it.", + "E": "You left the oil bottle beside the microwave on the counter, not directly in front of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 17 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_2_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the oil bottle?", + "answer": "In front of the microwave.", + "question_chinese": "我把油瓶放在了哪里？", + "answer_chinese": "微波炉前面", + "evidence": [ + { + "description": "In front of the microwave", + "timestamps": { + "start": "00:01:18", + "end": "00:01:19" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of the microwave.", + "B": "Behind the microwave.", + "C": "On top of the microwave.", + "D": "You put the oil bottle behind the microwave rather than setting it in front of it.", + "E": "You left the oil bottle beside the microwave on the counter, not positioned directly in front." + }, + "answer_label": "A", + "answer_text": "In front of the microwave.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_2_T2", + "correct_answer": "In front of the microwave.", + "original_options": { + "A": "In front of the microwave.", + "B": "Behind the microwave.", + "C": "On top of the microwave.", + "D": "You put the oil bottle behind the microwave rather than setting it in front of it.", + "E": "You left the oil bottle beside the microwave on the counter, not positioned directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 5, + 16, + 16 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_11", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after adding the rice to the pot?", + "answer": "Open the cabinet, put the rice inside, and then close the cabinet door.", + "question_chinese": "我把米加入锅后干了什么？", + "answer_chinese": "拉开柜子，把米放进去，然后关柜门", + "evidence": [ + { + "description": "Open the cabinet", + "timestamps": { + "start": "00:14:11", + "end": "00:14:12" + } + }, + { + "description": "put the rice inside", + "timestamps": { + "start": "00:14:13", + "end": "00:14:14" + } + }, + { + "description": "close the cabinet door", + "timestamps": { + "start": "00:14:14", + "end": "00:14:15" + } + } + ], + "question_timestamps": "00:14:16", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Open the fridge and close it.", + "B": "Open the cabinet, put the rice inside, and then close the cabinet door.", + "C": "Close the pot lid afterward.", + "D": "You opened the cabinet, took the rice back out, and then left the cabinet door open.", + "E": "You walked to the pantry, opened it, and set the rice on the counter instead of inside." + }, + "answer_label": "B", + "answer_text": "Open the cabinet, put the rice inside, and then close the cabinet door.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_11", + "correct_answer": "Open the cabinet, put the rice inside, and then close the cabinet door.", + "original_options": { + "A": "Open the fridge and close it.", + "B": "Open the cabinet, put the rice inside, and then close the cabinet door.", + "C": "Close the pot lid afterward.", + "D": "You opened the cabinet, took the rice back out, and then left the cabinet door open.", + "E": "You walked to the pantry, opened it, and set the rice on the counter instead of inside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_5_T2", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I added the rice to the pot, what did I do?", + "answer": "You pulled open the cabinet and put the rice inside.", + "question_chinese": "我把米加入锅后干了什么？", + "answer_chinese": "你拉开柜子，把米放了进去.", + "evidence": [ + { + "description": "Open the cabinet", + "timestamps": { + "start": "00:14:11", + "end": "00:14:12" + } + }, + { + "description": "put the rice inside", + "timestamps": { + "start": "00:14:13", + "end": "00:14:14" + } + } + ], + "question_timestamps": "00:14:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You closed the cabinet door.", + "B": "You pulled open the cabinet and put the rice inside.", + "C": "You left the cabinet open.", + "D": "You pulled open the cabinet and set the rice on the counter instead of putting it inside.", + "E": "You opened the cabinet, moved the rice to a different shelf, and walked away without closing it." + }, + "answer_label": "B", + "answer_text": "You pulled open the cabinet and put the rice inside.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_5_T2", + "correct_answer": "You pulled open the cabinet and put the rice inside.", + "original_options": { + "A": "You closed the cabinet door.", + "B": "You pulled open the cabinet and put the rice inside.", + "C": "You left the cabinet open.", + "D": "You pulled open the cabinet and set the rice on the counter instead of putting it inside.", + "E": "You opened the cabinet, moved the rice to a different shelf, and walked away without closing it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_5_T1", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I added the rice to the pot, what did I do?", + "answer": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "question_chinese": "我把米加入锅后干了什么？", + "answer_chinese": "拉开柜子，把米放进去，然后关柜门", + "evidence": [ + { + "description": "Open the cabinet", + "timestamps": { + "start": "00:14:11", + "end": "00:14:12" + } + }, + { + "description": "put the rice inside", + "timestamps": { + "start": "00:14:13", + "end": "00:14:14" + } + }, + { + "description": "close the cabinet door", + "timestamps": { + "start": "00:14:14", + "end": "00:14:15" + } + } + ], + "question_timestamps": "00:14:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "B": "You left the cabinet open.", + "C": "You put the rice outside.", + "D": "You pulled open the cabinet and put the rice inside, but you didn’t close the cabinet door.", + "E": "You opened the cabinet, set the rice down briefly, and then shut the door without placing it inside." + }, + "answer_label": "A", + "answer_text": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_5_T1", + "correct_answer": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "original_options": { + "A": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "B": "You left the cabinet open.", + "C": "You put the rice outside.", + "D": "You pulled open the cabinet and put the rice inside, but you didn’t close the cabinet door.", + "E": "You opened the cabinet, set the rice down briefly, and then shut the door without placing it inside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_6_T2", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I added the rice into the pot, what did I do?", + "answer": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "question_chinese": "我把米加进锅里后干了什么？", + "answer_chinese": "拉开柜子，把米放进去，然后关柜门", + "evidence": [ + { + "description": "Open the cabinet", + "timestamps": { + "start": "00:14:11", + "end": "00:14:12" + } + }, + { + "description": "put the rice inside", + "timestamps": { + "start": "00:14:13", + "end": "00:14:14" + } + }, + { + "description": "close the cabinet door", + "timestamps": { + "start": "00:14:14", + "end": "00:14:15" + } + } + ], + "question_timestamps": "00:14:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You closed the cabinet door.", + "B": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "C": "You put the rice outside.", + "D": "You pulled open the cabinet and put the rice inside, but you left the cabinet door open afterward.", + "E": "You opened the cabinet, placed the rice on the counter, and then closed the cabinet without storing it." + }, + "answer_label": "B", + "answer_text": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_6_T2", + "correct_answer": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "original_options": { + "A": "You closed the cabinet door.", + "B": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "C": "You put the rice outside.", + "D": "You pulled open the cabinet and put the rice inside, but you left the cabinet door open afterward.", + "E": "You opened the cabinet, placed the rice on the counter, and then closed the cabinet without storing it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_6_T1", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I added the rice into the pot, what did I do?", + "answer": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "question_chinese": "我把米加进锅里后干了什么？", + "answer_chinese": "拉开柜子，把米放进去，然后关柜门", + "evidence": [ + { + "description": "Open the cabinet", + "timestamps": { + "start": "00:14:11", + "end": "00:14:12" + } + }, + { + "description": "put the rice inside", + "timestamps": { + "start": "00:14:13", + "end": "00:14:14" + } + }, + { + "description": "close the cabinet door", + "timestamps": { + "start": "00:14:14", + "end": "00:14:15" + } + } + ], + "question_timestamps": "00:14:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You left the cabinet open.", + "B": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "C": "You put the rice outside.", + "D": "You pulled open the cabinet and put the rice inside, but you didn’t close the cabinet door afterward.", + "E": "You opened the cabinet, moved the rice to the counter, and then shut the cabinet without putting it away." + }, + "answer_label": "B", + "answer_text": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_6_T1", + "correct_answer": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "original_options": { + "A": "You left the cabinet open.", + "B": "You pulled open the cabinet, put the rice inside, and then closed the cabinet door.", + "C": "You put the rice outside.", + "D": "You pulled open the cabinet and put the rice inside, but you didn’t close the cabinet door afterward.", + "E": "You opened the cabinet, moved the rice to the counter, and then shut the cabinet without putting it away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_1_T1", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my guitar relative to me?", + "answer": "Your guitar is to the right of the TV in front of you.", + "question_chinese": "我的吉他在相对于我的什么位置？", + "answer_chinese": "你的吉他在你正前方的电视右边", + "evidence": [ + { + "description": "Your guitar is to the right of the TV right in front of you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your guitar is behind you.", + "B": "Your guitar is to the right of the TV in front of you.", + "C": "Your guitar is to your left.", + "D": "Your guitar is to the left of the TV in front of you, not on the right side.", + "E": "Your guitar is to the right of the TV but positioned behind you rather than in front." + }, + "answer_label": "B", + "answer_text": "Your guitar is to the right of the TV in front of you.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_1_T1", + "correct_answer": "Your guitar is to the right of the TV in front of you.", + "original_options": { + "A": "Your guitar is behind you.", + "B": "Your guitar is to the right of the TV in front of you.", + "C": "Your guitar is to your left.", + "D": "Your guitar is to the left of the TV in front of you, not on the right side.", + "E": "Your guitar is to the right of the TV but positioned behind you rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 17 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_1_T2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my guitar relative to me?", + "answer": "Your guitar is next to the TV, to your back-right.", + "question_chinese": "我的吉他在相对于我的什么位置？", + "answer_chinese": "你的吉他在你右后方电视的旁边", + "evidence": [ + { + "description": "Your guitar is right behind you to the right, next to the TV.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your back-left.", + "B": "It’s to your front-right.", + "C": "Your guitar is next to the TV, but it sits on your front-left rather than your back-right.", + "D": "Your guitar is next to the TV, to your back-right.", + "E": "Your guitar is beside the TV, but it’s positioned behind you on the left side, not the right." + }, + "answer_label": "D", + "answer_text": "Your guitar is next to the TV, to your back-right.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_1_T2", + "correct_answer": "Your guitar is next to the TV, to your back-right.", + "original_options": { + "A": "It’s to your back-left.", + "B": "It’s to your front-right.", + "C": "Your guitar is next to the TV, but it sits on your front-left rather than your back-right.", + "D": "Your guitar is next to the TV, to your back-right.", + "E": "Your guitar is beside the TV, but it’s positioned behind you on the left side, not the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_4_T1", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to get to the kitchen?", + "answer": "2.", + "question_chinese": "我要转多少个弯才能到厨房？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "1.", + "B": "2.", + "C": "3.", + "D": "You need to make 4 turns to reach the kitchen from where you are.", + "E": "You need to make 5 turns before you end up in the kitchen." + }, + "answer_label": "B", + "answer_text": "2.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_4_T1", + "correct_answer": "2.", + "original_options": { + "A": "1.", + "B": "2.", + "C": "3.", + "D": "You need to make 4 turns to reach the kitchen from where you are.", + "E": "You need to make 5 turns before you end up in the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6", + "video_length": "00:25:05", + "qid": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_4_T2", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to get to the kitchen?", + "answer": "You need to make one turn.", + "question_chinese": "我要转多少个弯才能到厨房？", + "answer_chinese": "你要拐1个弯", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero turns.", + "B": "Two turns.", + "C": "You need to make three turns before you reach the kitchen from where you are.", + "D": "You need to make one turn.", + "E": "You need to make four turns before you finally arrive in the kitchen area." + }, + "answer_label": "D", + "answer_text": "You need to make one turn.", + "q_id": "6a5fa00b-64de-4851-98fa-7e96f3f441f6_4_T2", + "correct_answer": "You need to make one turn.", + "original_options": { + "A": "Zero turns.", + "B": "Two turns.", + "C": "You need to make three turns before you reach the kitchen from where you are.", + "D": "You need to make one turn.", + "E": "You need to make four turns before you finally arrive in the kitchen area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + } +] diff --git a/QAs/6ab72273-4324-4cec-aad4-1a644b57356b-mcq.json b/QAs/6ab72273-4324-4cec-aad4-1a644b57356b-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7f5cd3af3e9be4444ed20dc21f3042a4a36c099e --- /dev/null +++ b/QAs/6ab72273-4324-4cec-aad4-1a644b57356b-mcq.json @@ -0,0 +1,1340 @@ +[ + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_7_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me?", + "answer": "The fridge was to your right.", + "question_chinese": "冰箱在我的哪里？", + "answer_chinese": "冰箱在你的右边。", + "evidence": [ + { + "description": "The fridge is on the left side of the door.", + "timestamps": { + "start": "00:09:25", + "end": "00:09:28" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge was to your left.", + "B": "The fridge was in front of you.", + "C": "The fridge was to your right.", + "D": "The fridge was behind you as you stood there, not on your right side.", + "E": "The fridge was above you in your view, rather than positioned to your right." + }, + "answer_label": "C", + "answer_text": "The fridge was to your right.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_7_T1", + "correct_answer": "The fridge was to your right.", + "original_options": { + "A": "The fridge was to your left.", + "B": "The fridge was in front of you.", + "C": "The fridge was to your right.", + "D": "The fridge was behind you as you stood there, not on your right side.", + "E": "The fridge was above you in your view, rather than positioned to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 14 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_7_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me?", + "answer": "The refrigerator is to your left.", + "question_chinese": "冰箱在我的哪里？", + "answer_chinese": "冰箱在你的左边。", + "evidence": [ + { + "description": "The fridge is on the left side of the door.", + "timestamps": { + "start": "00:09:17", + "end": "00:09:22" + } + } + ], + "question_timestamps": "00:19:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is to your right.", + "B": "The refrigerator is in front of you.", + "C": "The refrigerator is to your left.", + "D": "The refrigerator is behind you, not positioned on your left side where you are standing.", + "E": "The refrigerator is above you in your view, rather than being off to your left." + }, + "answer_label": "C", + "answer_text": "The refrigerator is to your left.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_7_T2", + "correct_answer": "The refrigerator is to your left.", + "original_options": { + "A": "The refrigerator is to your right.", + "B": "The refrigerator is in front of you.", + "C": "The refrigerator is to your left.", + "D": "The refrigerator is behind you, not positioned on your left side where you are standing.", + "E": "The refrigerator is above you in your view, rather than being off to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 15 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_15", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the spoon?", + "answer": "The spoon is in the empty space to the right of the microwave.", + "question_chinese": "勺子在哪里？", + "answer_chinese": "勺子在微波炉右边的空位上。", + "evidence": [ + { + "description": "The utensil rack with the spoon was moved by me to the right side of the microwave.", + "timestamps": { + "start": "00:06:12", + "end": "00:06:16" + } + } + ], + "question_timestamps": "00:07:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The spoon is left of the microwave.", + "B": "The spoon is behind the toaster.", + "C": "The spoon is in the empty space to the right of the microwave.", + "D": "The spoon is in the drawer to the left of the refrigerator, not in the open space by the microwave.", + "E": "The spoon is on the shelf above the sink, rather than in the empty area beside the microwave." + }, + "answer_label": "C", + "answer_text": "The spoon is in the empty space to the right of the microwave.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_15", + "correct_answer": "The spoon is in the empty space to the right of the microwave.", + "original_options": { + "A": "The spoon is left of the microwave.", + "B": "The spoon is behind the toaster.", + "C": "The spoon is in the empty space to the right of the microwave.", + "D": "The spoon is in the drawer to the left of the refrigerator, not in the open space by the microwave.", + "E": "The spoon is on the shelf above the sink, rather than in the empty area beside the microwave." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 18 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_16", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the spoon?", + "answer": "The spoon is on the draining rack to the left of the sink.", + "question_chinese": "勺子在哪里？", + "answer_chinese": "勺子在水槽左边的沥水架子上。", + "evidence": [ + { + "description": "The utensil rack with the spoon was moved by me to the right side of the microwave.", + "timestamps": { + "start": "00:06:12", + "end": "00:06:16" + } + }, + { + "description": "The utensil rack where the spoon was placed has been put back to the left of the sink by me.", + "timestamps": { + "start": "00:07:33", + "end": "00:07:39" + } + } + ], + "question_timestamps": "00:33:49", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The spoon is in the drawer.", + "B": "The spoon is on the counter.", + "C": "The spoon is near the sink on the right side, not on the draining rack to the left.", + "D": "You placed the spoon inside the cabinet above the stove instead of leaving it on the draining rack.", + "E": "The spoon is on the draining rack to the left of the sink." + }, + "answer_label": "E", + "answer_text": "The spoon is on the draining rack to the left of the sink.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_16", + "correct_answer": "The spoon is on the draining rack to the left of the sink.", + "original_options": { + "A": "The spoon is in the drawer.", + "B": "The spoon is on the counter.", + "C": "The spoon is near the sink on the right side, not on the draining rack to the left.", + "D": "You placed the spoon inside the cabinet above the stove instead of leaving it on the draining rack.", + "E": "The spoon is on the draining rack to the left of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "There are four chairs in the house.", + "question_chinese": "屋子里面有几张椅子？", + "answer_chinese": "屋子里面有四张椅子。", + "evidence": [ + { + "description": "There are four chairs around the table.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two chairs.", + "B": "There are five chairs.", + "C": "There are three chairs in the house, meaning one fewer than the stated total.", + "D": "There are six chairs in the house, which is two more than the correct count.", + "E": "There are four chairs in the house." + }, + "answer_label": "E", + "answer_text": "There are four chairs in the house.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_2", + "correct_answer": "There are four chairs in the house.", + "original_options": { + "A": "There are two chairs.", + "B": "There are five chairs.", + "C": "There are three chairs in the house, meaning one fewer than the stated total.", + "D": "There are six chairs in the house, which is two more than the correct count.", + "E": "There are four chairs in the house." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "There are four chairs in the house.", + "question_chinese": "屋子里面有几张椅子？", + "answer_chinese": "屋子里面有四张椅子。", + "evidence": [ + { + "description": "There are four chairs around the table.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:17:21", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three chairs.", + "B": "There are four chairs in the house.", + "C": "There are five chairs.", + "D": "There are six chairs in the house, which would be two more chairs than four.", + "E": "There are seven chairs in the house, indicating three additional chairs beyond the total of four." + }, + "answer_label": "B", + "answer_text": "There are four chairs in the house.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_1", + "correct_answer": "There are four chairs in the house.", + "original_options": { + "A": "There are three chairs.", + "B": "There are four chairs in the house.", + "C": "There are five chairs.", + "D": "There are six chairs in the house, which would be two more chairs than four.", + "E": "There are seven chairs in the house, indicating three additional chairs beyond the total of four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_11", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was washing things, which was closer to me, the red cutting board or the green one?", + "answer": "When you are washing things, the red cutting board is closer to you.", + "question_chinese": "当我在洗东西时，红色案板和绿色案板哪个离我更近？", + "answer_chinese": "当你在洗东西时，红色案板离你更近。", + "evidence": [ + { + "description": "The red cutting board is closer to me.", + "timestamps": { + "start": "00:24:08", + "end": "00:24:10" + } + } + ], + "question_timestamps": "00:25:13", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you are washing things, the red cutting board is closer to you.", + "B": "When you are washing things, the red cutting board is not closer to you." + }, + "answer_label": "A", + "answer_text": "When you are washing things, the red cutting board is closer to you.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_11", + "correct_answer": "When you are washing things, the red cutting board is closer to you.", + "original_options": { + "A": "When you are washing things, the red cutting board is closer to you.", + "B": "When you are washing things, the red cutting board is not closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_12", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was washing things, which was closer to me, the red cutting board or the green one?", + "answer": "When you are washing things, the red cutting board is closer to you.", + "question_chinese": "当我在洗东西时，红色案板和绿色案板哪个离我更近？", + "answer_chinese": "当你在洗东西时，红色案板离你更近。", + "evidence": [ + { + "description": "The red cutting board is closer to me.", + "timestamps": { + "start": "00:24:08", + "end": "00:24:10" + } + } + ], + "question_timestamps": "00:32:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you are washing things, the red cutting board is closer to you.", + "B": "When you are washing things, the red cutting board is not closer to you." + }, + "answer_label": "A", + "answer_text": "When you are washing things, the red cutting board is closer to you.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_12", + "correct_answer": "When you are washing things, the red cutting board is closer to you.", + "original_options": { + "A": "When you are washing things, the red cutting board is closer to you.", + "B": "When you are washing things, the red cutting board is not closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_14", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I brought the vacuum cleaner back to the living room, was the fridge on my left/right?", + "answer": "When you bring the vacuum cleaner back to the living room, the fridge is on your left.", + "question_chinese": "当我拿了吸尘器回到客厅时，冰箱在我的左/右？", + "answer_chinese": "当你拿了吸尘器回到客厅时，冰箱在你的左边。", + "evidence": [ + { + "description": "The fridge is on the left side of the door.", + "timestamps": { + "start": "00:09:17", + "end": "00:09:24" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The fridge is on your right.", + "B": "The fridge is in front of you.", + "C": "When you bring the vacuum cleaner back to the living room, the fridge ends up directly behind you.", + "D": "When you bring the vacuum cleaner back to the living room, the fridge is on your left.", + "E": "When you return the vacuum cleaner to the living room, the fridge is straight ahead rather than to either side." + }, + "answer_label": "D", + "answer_text": "When you bring the vacuum cleaner back to the living room, the fridge is on your left.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_14", + "correct_answer": "When you bring the vacuum cleaner back to the living room, the fridge is on your left.", + "original_options": { + "A": "The fridge is on your right.", + "B": "The fridge is in front of you.", + "C": "When you bring the vacuum cleaner back to the living room, the fridge ends up directly behind you.", + "D": "When you bring the vacuum cleaner back to the living room, the fridge is on your left.", + "E": "When you return the vacuum cleaner to the living room, the fridge is straight ahead rather than to either side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 20 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_13", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I brought the vacuum cleaner back to the living room, was the fridge on my left/right?", + "answer": "When you bring the vacuum cleaner back to the living room, the fridge is on your left.", + "question_chinese": "当我拿了吸尘器回到客厅时，冰箱在我的左/右？", + "answer_chinese": "当你拿了吸尘器回到客厅时，冰箱在你的左边。", + "evidence": [ + { + "description": "The fridge is on the left side of the door.", + "timestamps": { + "start": "00:09:17", + "end": "00:09:24" + } + } + ], + "question_timestamps": "00:19:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The fridge is on your right.", + "B": "When you bring the vacuum cleaner back to the living room, the fridge is on your left.", + "C": "The fridge is behind you.", + "D": "When you bring the vacuum cleaner back to the living room, the fridge is positioned directly in front of you.", + "E": "When you return the vacuum cleaner to the living room, the fridge is behind you instead of to your left." + }, + "answer_label": "B", + "answer_text": "When you bring the vacuum cleaner back to the living room, the fridge is on your left.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_13", + "correct_answer": "When you bring the vacuum cleaner back to the living room, the fridge is on your left.", + "original_options": { + "A": "The fridge is on your right.", + "B": "When you bring the vacuum cleaner back to the living room, the fridge is on your left.", + "C": "The fridge is behind you.", + "D": "When you bring the vacuum cleaner back to the living room, the fridge is positioned directly in front of you.", + "E": "When you return the vacuum cleaner to the living room, the fridge is behind you instead of to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 20 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_5_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What am I doing?", + "answer": "You were wiping the dining table, and the dining table was in front of you.", + "question_chinese": "我在干什么？", + "answer_chinese": "你在擦餐桌，餐桌在你前面", + "evidence": [ + { + "description": "You are cleaning the table.", + "timestamps": { + "start": "00:04:45", + "end": "00:04:50" + } + } + ], + "question_timestamps": "00:04:52", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were mopping the floor, and the mop was in front of you.", + "B": "You were wiping the dining table, and the dining table was in front of you." + }, + "answer_label": "B", + "answer_text": "You were wiping the dining table, and the dining table was in front of you.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_5_T2", + "correct_answer": "You were wiping the dining table, and the dining table was in front of you.", + "original_options": { + "A": "You were mopping the floor, and the mop was in front of you.", + "B": "You were wiping the dining table, and the dining table was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_5_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What am I doing?", + "answer": "You were mopping the floor, and the mop was in front of you.", + "question_chinese": "我在干什么？", + "answer_chinese": "你在拖地，拖把在你前面", + "evidence": [ + { + "description": "You are mopping the floor.", + "timestamps": { + "start": "00:14:45", + "end": "00:15:08" + } + } + ], + "question_timestamps": "00:15:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were mopping the floor, and the mop was in front of you.", + "B": "You were wiping the dining table, and the dining table was in front of you." + }, + "answer_label": "A", + "answer_text": "You were mopping the floor, and the mop was in front of you.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_5_T1", + "correct_answer": "You were mopping the floor, and the mop was in front of you.", + "original_options": { + "A": "You were mopping the floor, and the mop was in front of you.", + "B": "You were wiping the dining table, and the dining table was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_5", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to eat an apple, where can I get it?", + "answer": "You need to get the apples from the cupboard on the right side of the stove if you want to eat them.", + "question_chinese": "我想吃苹果要去哪里拿？", + "answer_chinese": "你想吃苹果要去灶台右边的柜台上拿。", + "evidence": [ + { + "description": "The apple is placed on the counter to the right of the stove.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:09" + } + } + ], + "question_timestamps": "00:04:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the cupboard left of the stove.", + "B": "From the drawer next to the oven.", + "C": "You need to get the apples from the cupboard on the right side of the stove if you want to eat them.", + "D": "You need to get the apples from the cabinet above the fridge if you want to eat them.", + "E": "You need to get the apples from the basket on the kitchen counter rather than from any cupboard by the stove." + }, + "answer_label": "C", + "answer_text": "You need to get the apples from the cupboard on the right side of the stove if you want to eat them.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_5", + "correct_answer": "You need to get the apples from the cupboard on the right side of the stove if you want to eat them.", + "original_options": { + "A": "From the cupboard left of the stove.", + "B": "From the drawer next to the oven.", + "C": "You need to get the apples from the cupboard on the right side of the stove if you want to eat them.", + "D": "You need to get the apples from the cabinet above the fridge if you want to eat them.", + "E": "You need to get the apples from the basket on the kitchen counter rather than from any cupboard by the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 21 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_6", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to eat an apple, where can I get it?", + "answer": "You need to get the apples from the cupboard on the right side of the stove if you want to eat them.", + "question_chinese": "我想吃苹果要去哪里拿？", + "answer_chinese": "你想吃苹果要去灶台右边的柜台上拿。", + "evidence": [ + { + "description": "The apple is placed on the counter to the right of the stove.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:09" + } + } + ], + "question_timestamps": "00:15:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the cupboard left of the stove.", + "B": "From the cupboard behind the stove.", + "C": "You need to get the apples from the cupboard on the right side of the stove if you want to eat them.", + "D": "You need to get the apples from the cupboard in front of the stove if you want to eat them.", + "E": "You need to get the apples from the cupboard behind you relative to the stove, not from the right side." + }, + "answer_label": "C", + "answer_text": "You need to get the apples from the cupboard on the right side of the stove if you want to eat them.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_6", + "correct_answer": "You need to get the apples from the cupboard on the right side of the stove if you want to eat them.", + "original_options": { + "A": "From the cupboard left of the stove.", + "B": "From the cupboard behind the stove.", + "C": "You need to get the apples from the cupboard on the right side of the stove if you want to eat them.", + "D": "You need to get the apples from the cupboard in front of the stove if you want to eat them.", + "E": "You need to get the apples from the cupboard behind you relative to the stove, not from the right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 20 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_1_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see?", + "answer": "You saw four chairs, and they were behind you.", + "question_chinese": "我看到几张椅子？", + "answer_chinese": "你看到四张椅子，它们在你身后。", + "evidence": [ + { + "description": "There are four chairs around the table.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw four chairs, and they were behind you.", + "B": "You saw two chairs behind you.", + "C": "You saw three chairs behind you.", + "D": "You saw five chairs, and they were behind you the whole time.", + "E": "You saw six chairs, and they were behind you the whole time." + }, + "answer_label": "A", + "answer_text": "You saw four chairs, and they were behind you.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_1_T1", + "correct_answer": "You saw four chairs, and they were behind you.", + "original_options": { + "A": "You saw four chairs, and they were behind you.", + "B": "You saw two chairs behind you.", + "C": "You saw three chairs behind you.", + "D": "You saw five chairs, and they were behind you the whole time.", + "E": "You saw six chairs, and they were behind you the whole time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 12 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_1_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see?", + "answer": "Then you saw one chair, and it was in front of you.", + "question_chinese": "我看到几张椅子？", + "answer_chinese": "那你看到一张椅子，它在你前面。", + "evidence": [ + { + "description": "There are one chairs around the table.", + "timestamps": { + "start": "00:04:49", + "end": "00:05:03" + } + } + ], + "question_timestamps": "00:05:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Then you saw zero chairs in front.", + "B": "Then you saw one chair, and it was in front of you.", + "C": "Then you saw two chairs in front.", + "D": "Then you saw three chairs, and they were in front of you.", + "E": "Then you saw four chairs, and they were in front of you." + }, + "answer_label": "B", + "answer_text": "Then you saw one chair, and it was in front of you.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_1_T2", + "correct_answer": "Then you saw one chair, and it was in front of you.", + "original_options": { + "A": "Then you saw zero chairs in front.", + "B": "Then you saw one chair, and it was in front of you.", + "C": "Then you saw two chairs in front.", + "D": "Then you saw three chairs, and they were in front of you.", + "E": "Then you saw four chairs, and they were in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 12 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_7", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I am standing by the table, how do I get to the room with two sofas?", + "answer": "Go straight ahead through a door, then enter the first doorway on the right side of the corridor.", + "question_chinese": "我站在桌子旁边，要去到有两个沙发的房间应该怎么走？", + "answer_chinese": "往前直走穿过一个门，然后走廊右手边第一个门口进去。", + "evidence": [ + { + "description": "The room with two sofas is the first room on the right side of the corridor.", + "timestamps": { + "start": "00:17:38", + "end": "00:17:45" + } + } + ], + "question_timestamps": "00:18:29", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Enter the second doorway on the right.", + "B": "Take the first doorway on the left.", + "C": "Go through the door ahead, then take the second doorway on the right side of the corridor.", + "D": "Go through the door ahead and continue down the corridor until you reach the last doorway on the right.", + "E": "Go straight ahead through a door, then enter the first doorway on the right side of the corridor." + }, + "answer_label": "E", + "answer_text": "Go straight ahead through a door, then enter the first doorway on the right side of the corridor.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_7", + "correct_answer": "Go straight ahead through a door, then enter the first doorway on the right side of the corridor.", + "original_options": { + "A": "Enter the second doorway on the right.", + "B": "Take the first doorway on the left.", + "C": "Go through the door ahead, then take the second doorway on the right side of the corridor.", + "D": "Go through the door ahead and continue down the corridor until you reach the last doorway on the right.", + "E": "Go straight ahead through a door, then enter the first doorway on the right side of the corridor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 19 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_8", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I am standing by the table, how do I get to the room with two sofas?", + "answer": "Go straight ahead through a door, then enter the first doorway on the right side of the corridor.", + "question_chinese": "我站在桌子旁边，要去到有两个沙发的房间应该怎么走？", + "answer_chinese": "往前直走穿过一个门，然后走廊右手边第一个门口进去。", + "evidence": [ + { + "description": "The room with two sofas is the first room on the right side of the corridor.", + "timestamps": { + "start": "00:17:38", + "end": "00:17:45" + } + } + ], + "question_timestamps": "00:33:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Take the first doorway on your left.", + "B": "Enter the second doorway on the right.", + "C": "Go straight through the door, then enter the first doorway on the left side of the corridor.", + "D": "Go straight ahead through a door, then enter the first doorway on the right side of the corridor.", + "E": "Go through the door ahead, then keep walking until you reach the second doorway on the right." + }, + "answer_label": "D", + "answer_text": "Go straight ahead through a door, then enter the first doorway on the right side of the corridor.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_8", + "correct_answer": "Go straight ahead through a door, then enter the first doorway on the right side of the corridor.", + "original_options": { + "A": "Take the first doorway on your left.", + "B": "Enter the second doorway on the right.", + "C": "Go straight through the door, then enter the first doorway on the left side of the corridor.", + "D": "Go straight ahead through a door, then enter the first doorway on the right side of the corridor.", + "E": "Go through the door ahead, then keep walking until you reach the second doorway on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_4", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are on the stove?", + "answer": "There is no pot on the stove.", + "question_chinese": "灶台上有几个锅？", + "answer_chinese": "灶台上没有锅。", + "evidence": [ + { + "description": "There is no pot on the stove.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:02:26", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one pot on the stove.", + "B": "There are two pots on the stove.", + "C": "There is no pot on the stove.", + "D": "There are three pots on the stove, spaced evenly across the burners.", + "E": "There are four pots on the stove, with one pot sitting on each burner." + }, + "answer_label": "C", + "answer_text": "There is no pot on the stove.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_4", + "correct_answer": "There is no pot on the stove.", + "original_options": { + "A": "There is one pot on the stove.", + "B": "There are two pots on the stove.", + "C": "There is no pot on the stove.", + "D": "There are three pots on the stove, spaced evenly across the burners.", + "E": "There are four pots on the stove, with one pot sitting on each burner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 14 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_3", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are on the stove?", + "answer": "There is one pot on the stove.", + "question_chinese": "灶台上有几个锅？", + "answer_chinese": "灶台上有一个锅。", + "evidence": [ + { + "description": "There is no pot on the stove.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:29" + } + }, + { + "description": "I put a pot on the stove.", + "timestamps": { + "start": "00:03:04", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:19:44", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There is one pot on the stove.", + "B": "There are no pots on the stove.", + "C": "There are two pots on the stove.", + "D": "There are three pots on the stove, with two on the back burners.", + "E": "There are four pots on the stove, covering most of the burners." + }, + "answer_label": "A", + "answer_text": "There is one pot on the stove.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_3", + "correct_answer": "There is one pot on the stove.", + "original_options": { + "A": "There is one pot on the stove.", + "B": "There are no pots on the stove.", + "C": "There are two pots on the stove.", + "D": "There are three pots on the stove, with two on the back burners.", + "E": "There are four pots on the stove, covering most of the burners." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 12 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_6_T2", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the red cutting board or the green cutting board?", + "answer": "The red cutting board was closer to you, and it was in front of you.", + "question_chinese": "红色案板和绿色案板哪个离我更近？", + "answer_chinese": "红色案板离你更近，它在你的前面。", + "evidence": [ + { + "description": "The red cutting board is closer to me.", + "timestamps": { + "start": "00:24:08", + "end": "00:24:12" + } + } + ], + "question_timestamps": "00:24:14", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The green cutting board was closer to you, and it was in front of you.", + "B": "The red cutting board was closer to you, and it was in front of you." + }, + "answer_label": "B", + "answer_text": "The red cutting board was closer to you, and it was in front of you.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_6_T2", + "correct_answer": "The red cutting board was closer to you, and it was in front of you.", + "original_options": { + "A": "The green cutting board was closer to you, and it was in front of you.", + "B": "The red cutting board was closer to you, and it was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_6_T1", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the red cutting board or the green cutting board?", + "answer": "The green cutting board was closer to you, and it was in front of you.", + "question_chinese": "红色案板和绿色案板哪个离我更近？", + "answer_chinese": "绿色案板离你更近，它在你的前面。", + "evidence": [ + { + "description": "The green cutting board is closer to me.", + "timestamps": { + "start": "00:24:27", + "end": "00:24:32" + } + } + ], + "question_timestamps": "00:24:34", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The green cutting board was closer to you, and it was in front of you.", + "B": "The red cutting board was closer to you, and it was in front of you." + }, + "answer_label": "A", + "answer_text": "The green cutting board was closer to you, and it was in front of you.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_6_T1", + "correct_answer": "The green cutting board was closer to you, and it was in front of you.", + "original_options": { + "A": "The green cutting board was closer to you, and it was in front of you.", + "B": "The red cutting board was closer to you, and it was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_3_T1", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the apple relative to me?", + "answer": "The apple is on the counter to your right.", + "question_chinese": "苹果在我的哪里？", + "answer_chinese": "苹果在你的右边柜台上", + "evidence": [ + { + "description": "The apple is placed on the counter to the right of the stove.", + "timestamps": { + "start": "00:03:06", + "end": "00:03:09" + } + } + ], + "question_timestamps": "00:03:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the counter ahead.", + "B": "The apple is on the counter to your right.", + "C": "On the counter left.", + "D": "The apple is on the counter behind you, out of your immediate view.", + "E": "The apple is on the counter in front of you, slightly offset toward your left side." + }, + "answer_label": "B", + "answer_text": "The apple is on the counter to your right.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_3_T1", + "correct_answer": "The apple is on the counter to your right.", + "original_options": { + "A": "On the counter ahead.", + "B": "The apple is on the counter to your right.", + "C": "On the counter left.", + "D": "The apple is on the counter behind you, out of your immediate view.", + "E": "The apple is on the counter in front of you, slightly offset toward your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 16 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_3_T2", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the apple relative to me?", + "answer": "The apple is behind you.", + "question_chinese": "苹果在我的哪里？", + "answer_chinese": "苹果在你的身后", + "evidence": [ + { + "description": "The apple is behind you.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The apple is in front of you.", + "B": "The apple is to your left.", + "C": "The apple is to your right, near your side as you face forward.", + "D": "The apple is behind you.", + "E": "The apple is in front of you, positioned slightly off to your right." + }, + "answer_label": "D", + "answer_text": "The apple is behind you.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_3_T2", + "correct_answer": "The apple is behind you.", + "original_options": { + "A": "The apple is in front of you.", + "B": "The apple is to your left.", + "C": "The apple is to your right, near your side as you face forward.", + "D": "The apple is behind you.", + "E": "The apple is in front of you, positioned slightly off to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 13 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_4_T1", + "qid_num": 25, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the dining table relative to me?", + "answer": "The dining table is in front of you.", + "question_chinese": "餐桌在我的哪里？", + "answer_chinese": "餐桌在你的前面", + "evidence": [ + { + "description": "The table is in front of you.", + "timestamps": { + "start": "00:05:37", + "end": "00:05:43" + } + } + ], + "question_timestamps": "00:05:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The dining table is behind you.", + "B": "The dining table is to your left.", + "C": "The dining table is to your right, just off your side as you face forward.", + "D": "The dining table is above you, positioned higher than your eye level in the room.", + "E": "The dining table is in front of you." + }, + "answer_label": "E", + "answer_text": "The dining table is in front of you.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_4_T1", + "correct_answer": "The dining table is in front of you.", + "original_options": { + "A": "The dining table is behind you.", + "B": "The dining table is to your left.", + "C": "The dining table is to your right, just off your side as you face forward.", + "D": "The dining table is above you, positioned higher than your eye level in the room.", + "E": "The dining table is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 15 + ] + }, + { + "video_uid": "6ab72273-4324-4cec-aad4-1a644b57356b", + "video_length": "00:34:39", + "qid": "6ab72273-4324-4cec-aad4-1a644b57356b_4_T2", + "qid_num": 26, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the dining table relative to me?", + "answer": "The dining table is to your right.", + "question_chinese": "餐桌在我的哪里？", + "answer_chinese": "餐桌在你的右边。", + "evidence": [ + { + "description": "The table is on your right.", + "timestamps": { + "start": "00:17:38", + "end": "00:17:41" + } + } + ], + "question_timestamps": "00:17:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The dining table is to your right.", + "B": "The dining table is to your left.", + "C": "The dining table is behind you.", + "D": "The dining table is in front of you, directly ahead as you look forward.", + "E": "The dining table is above you, higher up relative to where you are standing." + }, + "answer_label": "A", + "answer_text": "The dining table is to your right.", + "q_id": "6ab72273-4324-4cec-aad4-1a644b57356b_4_T2", + "correct_answer": "The dining table is to your right.", + "original_options": { + "A": "The dining table is to your right.", + "B": "The dining table is to your left.", + "C": "The dining table is behind you.", + "D": "The dining table is in front of you, directly ahead as you look forward.", + "E": "The dining table is above you, higher up relative to where you are standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 14 + ] + } +] diff --git a/QAs/6af04762-9ccf-41aa-bbfb-48a443c7cec3-mcq.json b/QAs/6af04762-9ccf-41aa-bbfb-48a443c7cec3-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..afd58d01203119adfabc7bb7decc125e31d4e8fc --- /dev/null +++ b/QAs/6af04762-9ccf-41aa-bbfb-48a443c7cec3-mcq.json @@ -0,0 +1,521 @@ +[ + { + "video_uid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3", + "video_length": "00:35:26", + "qid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_3", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was it before I picked up my phone for the first time?", + "answer": "The phone is on the ground on the first floor, to the right of the remote control.", + "question_chinese": "在我第一次拿起手机之前，它在哪里？", + "answer_chinese": "手机在一楼地上，遥控器的右侧。", + "evidence": [ + { + "description": "There is a mobile phone and a remote control on the ground, and the mobile phone is to the right of the remote control.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The phone is on the ground on the first floor, to the right of the remote control.", + "B": "On the ground, left of the remote.", + "C": "On the table on the first floor.", + "D": "The phone is on the ground on the first floor, positioned in front of the remote control.", + "E": "The phone is on the ground on the second floor, placed to the right of the remote control." + }, + "answer_label": "A", + "answer_text": "The phone is on the ground on the first floor, to the right of the remote control.", + "q_id": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_3", + "correct_answer": "The phone is on the ground on the first floor, to the right of the remote control.", + "original_options": { + "A": "The phone is on the ground on the first floor, to the right of the remote control.", + "B": "On the ground, left of the remote.", + "C": "On the table on the first floor.", + "D": "The phone is on the ground on the first floor, positioned in front of the remote control.", + "E": "The phone is on the ground on the second floor, placed to the right of the remote control." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3", + "video_length": "00:35:26", + "qid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was it before I picked up my phone for the first time?", + "answer": "The phone is on the ground on the first floor, to the right of the remote control.", + "question_chinese": "在我第一次拿起手机之前，它在哪里？", + "answer_chinese": "手机在一楼地上，遥控器的右侧。", + "evidence": [ + { + "description": "There is a mobile phone and a remote control on the ground, and the mobile phone is to the right of the remote control.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:25:29", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the couch on the first floor.", + "B": "On the shelf on the first floor.", + "C": "The phone is on the ground on the first floor, to the right of the remote control.", + "D": "The phone is on the table on the second floor, positioned to the left of the charger.", + "E": "The phone is on the floor on the third floor, placed to the left of the window." + }, + "answer_label": "C", + "answer_text": "The phone is on the ground on the first floor, to the right of the remote control.", + "q_id": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_4", + "correct_answer": "The phone is on the ground on the first floor, to the right of the remote control.", + "original_options": { + "A": "On the couch on the first floor.", + "B": "On the shelf on the first floor.", + "C": "The phone is on the ground on the first floor, to the right of the remote control.", + "D": "The phone is on the table on the second floor, positioned to the left of the charger.", + "E": "The phone is on the floor on the third floor, placed to the left of the window." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3", + "video_length": "00:35:26", + "qid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_11", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to get more plates, where should I go?", + "answer": "You should get it from the bowl cabinet at the lower right of the sink.", + "question_chinese": "如果我要拿更多的盘子，应该去哪里拿？", + "answer_chinese": "你应该去水槽右下方的碗柜里拿。", + "evidence": [ + { + "description": "There are many plates in the bowl cabinet at the lower right of the sink.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:01" + } + } + ], + "question_timestamps": "00:09:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go to the drawer by the fridge.", + "B": "Check the shelf above the sink.", + "C": "You should get it from the cabinet beside the oven at the upper right of the sink.", + "D": "You should go to the pantry across from the sink to get more plates.", + "E": "You should get it from the bowl cabinet at the lower right of the sink." + }, + "answer_label": "E", + "answer_text": "You should get it from the bowl cabinet at the lower right of the sink.", + "q_id": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_11", + "correct_answer": "You should get it from the bowl cabinet at the lower right of the sink.", + "original_options": { + "A": "Go to the drawer by the fridge.", + "B": "Check the shelf above the sink.", + "C": "You should get it from the cabinet beside the oven at the upper right of the sink.", + "D": "You should go to the pantry across from the sink to get more plates.", + "E": "You should get it from the bowl cabinet at the lower right of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 14 + ] + }, + { + "video_uid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3", + "video_length": "00:35:26", + "qid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_12", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to get more plates, where should I go?", + "answer": "You should get it from the bowl cabinet at the lower right of the sink.", + "question_chinese": "如果我要拿更多的盘子，应该去哪里拿？", + "answer_chinese": "你应该去水槽右下方的碗柜里拿。", + "evidence": [ + { + "description": "There are many plates in the bowl cabinet at the lower right of the sink.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:01" + } + } + ], + "question_timestamps": "00:24:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Use the drawer beside the stove.", + "B": "Go to the rack at lower left.", + "C": "You should get it from the bowl cabinet at the lower right of the sink.", + "D": "You should get it from the shelf near the refrigerator at the middle left of the sink.", + "E": "You should get it from the cupboard above the dishwasher at the upper right of the sink." + }, + "answer_label": "C", + "answer_text": "You should get it from the bowl cabinet at the lower right of the sink.", + "q_id": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_12", + "correct_answer": "You should get it from the bowl cabinet at the lower right of the sink.", + "original_options": { + "A": "Use the drawer beside the stove.", + "B": "Go to the rack at lower left.", + "C": "You should get it from the bowl cabinet at the lower right of the sink.", + "D": "You should get it from the shelf near the refrigerator at the middle left of the sink.", + "E": "You should get it from the cupboard above the dishwasher at the upper right of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 17 + ] + }, + { + "video_uid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3", + "video_length": "00:35:26", + "qid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_15", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was washing vegetables, which is closer to me, the soup pot in front of the sink or the pink kettle?", + "answer": "The pot is closer to you.", + "question_chinese": "当我在洗菜的时候,水槽前面的汤锅和粉色水壶哪个离我更近?", + "answer_chinese": "汤锅离你更近。", + "evidence": [ + { + "description": "The soup pot is closer to you, about 15cm away, and the pink kettle is in front of you to the right, about 40cm away.", + "timestamps": { + "start": "00:13:00", + "end": "00:13:09" + } + } + ], + "question_timestamps": "00:13:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The pink kettle is closer to you.", + "B": "The pot is closer to you." + }, + "answer_label": "B", + "answer_text": "The pot is closer to you.", + "q_id": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_15", + "correct_answer": "The pot is closer to you.", + "original_options": { + "A": "The pink kettle is closer to you.", + "B": "The pot is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3", + "video_length": "00:35:26", + "qid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_16", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was washing vegetables, which is closer to me, the soup pot in front of the sink or the pink kettle?", + "answer": "The pot is closer to you.", + "question_chinese": "当我在洗菜的时候,水槽前面的汤锅和粉色水壶哪个离我更近?", + "answer_chinese": "汤锅离你更近。", + "evidence": [ + { + "description": "The soup pot is closer to you, about 15cm away, and the pink kettle is in front of you to the right, about 40cm away.", + "timestamps": { + "start": "00:13:00", + "end": "00:13:09" + } + } + ], + "question_timestamps": "00:28:09", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The pink kettle is closer to you.", + "B": "The pot is closer to you." + }, + "answer_label": "B", + "answer_text": "The pot is closer to you.", + "q_id": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_16", + "correct_answer": "The pot is closer to you.", + "original_options": { + "A": "The pink kettle is closer to you.", + "B": "The pot is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3", + "video_length": "00:35:26", + "qid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_17", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cloths did I use to take the chicken out of the oven?", + "answer": "You used a cloth to take the chicken out of the oven.", + "question_chinese": "我用了几块抹布在拿烤箱里的鸡肉？", + "answer_chinese": "你用了一块抹布来拿烤箱里的鸡肉。", + "evidence": [ + { + "description": "You used a pink cloth.", + "timestamps": { + "start": "00:19:26", + "end": "00:19:43" + } + } + ], + "question_timestamps": "00:19:45", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used two cloths.", + "B": "You used a cloth to take the chicken out of the oven.", + "C": "You used three cloths.", + "D": "You used two cloths to take the chicken out of the oven, one in each hand.", + "E": "You used four cloths to take the chicken out of the oven, doubling up for extra heat protection." + }, + "answer_label": "B", + "answer_text": "You used a cloth to take the chicken out of the oven.", + "q_id": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_17", + "correct_answer": "You used a cloth to take the chicken out of the oven.", + "original_options": { + "A": "You used two cloths.", + "B": "You used a cloth to take the chicken out of the oven.", + "C": "You used three cloths.", + "D": "You used two cloths to take the chicken out of the oven, one in each hand.", + "E": "You used four cloths to take the chicken out of the oven, doubling up for extra heat protection." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 18 + ] + }, + { + "video_uid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3", + "video_length": "00:35:26", + "qid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_18", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cloths did I use to take the chicken out of the oven?", + "answer": "You used two cloths to take the chicken out of the oven.", + "question_chinese": "我用了几块抹布在拿烤箱里的鸡肉？", + "answer_chinese": "你用了两块抹布来拿烤箱里的鸡肉。", + "evidence": [ + { + "description": "You used a pink cloth.", + "timestamps": { + "start": "00:19:26", + "end": "00:19:43" + } + }, + { + "description": "You took another yellow rag from the cabinet to use.", + "timestamps": { + "start": "00:19:44", + "end": "00:19:54" + } + } + ], + "question_timestamps": "00:29:54", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You used one cloth.", + "B": "You used three cloths.", + "C": "You used four cloths to take the chicken out of the oven, stacking them together for insulation.", + "D": "You used five cloths to take the chicken out of the oven, switching cloths between lifting and setting down.", + "E": "You used two cloths to take the chicken out of the oven." + }, + "answer_label": "E", + "answer_text": "You used two cloths to take the chicken out of the oven.", + "q_id": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_18", + "correct_answer": "You used two cloths to take the chicken out of the oven.", + "original_options": { + "A": "You used one cloth.", + "B": "You used three cloths.", + "C": "You used four cloths to take the chicken out of the oven, stacking them together for insulation.", + "D": "You used five cloths to take the chicken out of the oven, switching cloths between lifting and setting down.", + "E": "You used two cloths to take the chicken out of the oven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 19 + ] + }, + { + "video_uid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3", + "video_length": "00:35:26", + "qid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_7", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The dish sponge is used up. Where should I put it?", + "answer": "You should put the dishwashing sponge into the box in front of the sink.", + "question_chinese": "洗碗海绵用完了，我应该把它放在哪里？", + "answer_chinese": "你应该把洗碗海绵放到水槽前方的盒子里。", + "evidence": [ + { + "description": "There is a dishwashing sponge in the box in front of the sink.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:57" + } + } + ], + "question_timestamps": "00:02:59", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should put the dishwashing sponge into the box in front of the sink.", + "B": "Put it in the drawer under the sink.", + "C": "Put it in the basket by the dishwasher.", + "D": "You should put the dishwashing sponge into the drawer beneath the oven so it’s out of the way.", + "E": "You should put the dishwashing sponge on the countertop next to the toaster until you throw it away." + }, + "answer_label": "A", + "answer_text": "You should put the dishwashing sponge into the box in front of the sink.", + "q_id": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_7", + "correct_answer": "You should put the dishwashing sponge into the box in front of the sink.", + "original_options": { + "A": "You should put the dishwashing sponge into the box in front of the sink.", + "B": "Put it in the drawer under the sink.", + "C": "Put it in the basket by the dishwasher.", + "D": "You should put the dishwashing sponge into the drawer beneath the oven so it’s out of the way.", + "E": "You should put the dishwashing sponge on the countertop next to the toaster until you throw it away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 18 + ] + }, + { + "video_uid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3", + "video_length": "00:35:26", + "qid": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_8", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The dish sponge is used up. Where should I put it?", + "answer": "You should put the dishwashing sponge into the box in front of the sink.", + "question_chinese": "洗碗海绵用完了，我应该把它放在哪里？", + "answer_chinese": "你应该把洗碗海绵放到水槽前方的盒子里。", + "evidence": [ + { + "description": "There is a dishwashing sponge in the box in front of the sink.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:57" + } + } + ], + "question_timestamps": "00:22:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put it in the left sink drawer.", + "B": "Put it in the basket by the sink.", + "C": "You should put the dishwashing sponge into the box in front of the sink.", + "D": "You should put the dishwashing sponge into the cupboard above the sink on the right for storage.", + "E": "You should put the dishwashing sponge into the container beside the refrigerator on the right to keep it contained." + }, + "answer_label": "C", + "answer_text": "You should put the dishwashing sponge into the box in front of the sink.", + "q_id": "6af04762-9ccf-41aa-bbfb-48a443c7cec3_8", + "correct_answer": "You should put the dishwashing sponge into the box in front of the sink.", + "original_options": { + "A": "Put it in the left sink drawer.", + "B": "Put it in the basket by the sink.", + "C": "You should put the dishwashing sponge into the box in front of the sink.", + "D": "You should put the dishwashing sponge into the cupboard above the sink on the right for storage.", + "E": "You should put the dishwashing sponge into the container beside the refrigerator on the right to keep it contained." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 19 + ] + } +] diff --git a/QAs/6b70f1de-16cd-4b62-ab46-bc988254f486-mcq.json b/QAs/6b70f1de-16cd-4b62-ab46-bc988254f486-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..84e402dfaec25ac0042427899ece064cf86c67a5 --- /dev/null +++ b/QAs/6b70f1de-16cd-4b62-ab46-bc988254f486-mcq.json @@ -0,0 +1,1539 @@ +[ + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From the room door to the elevator, how many doors did you pass by?", + "answer": "From the room door to the elevator, passed ten doors.", + "question_chinese": "从房间门口到电梯，经过了几扇门？", + "answer_chinese": "从房间门口到电梯，经过了十扇门。", + "evidence": [ + { + "description": "From the room door to the elevator, passed ten doors.", + "timestamps": { + "start": "00:05:22", + "end": "00:06:18" + } + } + ], + "question_timestamps": "00:07:32", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Passed nine doors.", + "B": "Passed eleven doors.", + "C": "From the room door to the elevator, passed seven doors along the hallway before reaching it.", + "D": "From the room door to the elevator, passed ten doors.", + "E": "From the room door to the elevator, passed thirteen doors in total on your way down the corridor." + }, + "answer_label": "D", + "answer_text": "From the room door to the elevator, passed ten doors.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_7", + "correct_answer": "From the room door to the elevator, passed ten doors.", + "original_options": { + "A": "Passed nine doors.", + "B": "Passed eleven doors.", + "C": "From the room door to the elevator, passed seven doors along the hallway before reaching it.", + "D": "From the room door to the elevator, passed ten doors.", + "E": "From the room door to the elevator, passed thirteen doors in total on your way down the corridor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 18 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From the room door to the elevator, how many doors did you pass by?", + "answer": "From the room door to the elevator, passed ten doors.", + "question_chinese": "从房间门口到电梯，经过了几扇门？", + "answer_chinese": "从房间门口到电梯，经过了十扇门。", + "evidence": [ + { + "description": "From the room door to the elevator, passed ten doors.", + "timestamps": { + "start": "00:05:22", + "end": "00:06:18" + } + } + ], + "question_timestamps": "00:16:57", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "From the room door to the elevator, passed ten doors.", + "B": "Passed nine doors.", + "C": "Passed eleven doors.", + "D": "From the room door to the elevator, passed seven doors before arriving at the elevator entrance.", + "E": "From the room door to the elevator, passed thirteen doors overall while walking straight to the elevator." + }, + "answer_label": "A", + "answer_text": "From the room door to the elevator, passed ten doors.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_8", + "correct_answer": "From the room door to the elevator, passed ten doors.", + "original_options": { + "A": "From the room door to the elevator, passed ten doors.", + "B": "Passed nine doors.", + "C": "Passed eleven doors.", + "D": "From the room door to the elevator, passed seven doors before arriving at the elevator entrance.", + "E": "From the room door to the elevator, passed thirteen doors overall while walking straight to the elevator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_17", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there?", + "answer": "There are six chairs in the scene.", + "question_chinese": "场景里有几把椅子？", + "answer_chinese": "场景里有六把椅子。", + "evidence": [ + { + "description": "There is a small stool and a slightly higher soft stool next to the cooking counter.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:14" + } + }, + { + "description": "Four chairs were placed around the table.", + "timestamps": { + "start": "00:13:06", + "end": "00:13:07" + } + } + ], + "question_timestamps": "00:14:26", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are five chairs.", + "B": "There are seven chairs.", + "C": "There are six chairs in the scene.", + "D": "There are four chairs in the scene, with two grouped together and two set apart.", + "E": "There are nine chairs in the scene, including several pushed back from the table." + }, + "answer_label": "C", + "answer_text": "There are six chairs in the scene.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_17", + "correct_answer": "There are six chairs in the scene.", + "original_options": { + "A": "There are five chairs.", + "B": "There are seven chairs.", + "C": "There are six chairs in the scene.", + "D": "There are four chairs in the scene, with two grouped together and two set apart.", + "E": "There are nine chairs in the scene, including several pushed back from the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_18", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there?", + "answer": "There are six chairs in the scene.", + "question_chinese": "场景里有几把椅子？", + "answer_chinese": "场景里有六把椅子。", + "evidence": [ + { + "description": "Four chairs were placed around the table.", + "timestamps": { + "start": "00:13:06", + "end": "00:13:07" + } + }, + { + "description": "There is a small stool and a slightly higher soft stool next to the cooking counter.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:14" + } + } + ], + "question_timestamps": "00:22:35", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are three chairs.", + "B": "There are six chairs in the scene.", + "C": "There are eight chairs.", + "D": "There are two chairs in the scene, positioned next to each other in the same area.", + "E": "There are nine chairs in the scene, spread out with a few clustered near the main surface." + }, + "answer_label": "B", + "answer_text": "There are six chairs in the scene.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_18", + "correct_answer": "There are six chairs in the scene.", + "original_options": { + "A": "There are three chairs.", + "B": "There are six chairs in the scene.", + "C": "There are eight chairs.", + "D": "There are two chairs in the scene, positioned next to each other in the same area.", + "E": "There are nine chairs in the scene, spread out with a few clustered near the main surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_7_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was sitting on the sofa, could I reach the picture frame?", + "answer": "When you’re sitting on the couch, you can’t reach the photo frame.", + "question_chinese": "当我坐在沙发上时，我能碰到相框吗？", + "answer_chinese": "当你坐在沙发上时，你碰不到相框", + "evidence": [ + { + "description": "There are ten bottles of water on the table.", + "timestamps": { + "start": "00:04:24", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:05:39", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you’re sitting on the couch, you can’t reach the photo frame.", + "B": "When you’re sitting on the couch, you can reach the photo frame." + }, + "answer_label": "A", + "answer_text": "When you’re sitting on the couch, you can’t reach the photo frame.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_7_T1", + "correct_answer": "When you’re sitting on the couch, you can’t reach the photo frame.", + "original_options": { + "A": "When you’re sitting on the couch, you can’t reach the photo frame.", + "B": "When you’re sitting on the couch, you can reach the photo frame." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_11", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I sit on the sofa, can I reach the photo frame?", + "answer": "When you sit on the sofa, you cannot touch the photo frame.", + "question_chinese": "当我坐在沙发上时，我能碰到相框吗？", + "answer_chinese": "当你坐在沙发上时，你不能碰到相框。", + "evidence": [ + { + "description": "The photo frame is placed on the shelf next to the TV.", + "timestamps": { + "start": "00:12:10", + "end": "00:12:18" + } + } + ], + "question_timestamps": "00:13:23", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you sit on the sofa, you cannot touch the photo frame.", + "B": "When you sit on the sofa, you can touch the photo frame." + }, + "answer_label": "A", + "answer_text": "When you sit on the sofa, you cannot touch the photo frame.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_11", + "correct_answer": "When you sit on the sofa, you cannot touch the photo frame.", + "original_options": { + "A": "When you sit on the sofa, you cannot touch the photo frame.", + "B": "When you sit on the sofa, you can touch the photo frame." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_7_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was sitting on the sofa, could I reach the picture frame?", + "answer": "There are ten bottles of water on the table.", + "question_chinese": "当我坐在沙发上时，我能碰到相框吗？", + "answer_chinese": "桌子上有十瓶水。", + "evidence": [ + { + "description": "There are ten bottles of water on the table.", + "timestamps": { + "start": "00:04:24", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:14:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There are ten bottles of water on the table.", + "B": "When you were sitting on the sofa, you could reach the picture frame." + }, + "answer_label": "A", + "answer_text": "There are ten bottles of water on the table.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_7_T2", + "correct_answer": "There are ten bottles of water on the table.", + "original_options": { + "A": "There are ten bottles of water on the table.", + "B": "When you were sitting on the sofa, you could reach the picture frame." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_4_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors did I just pass through?", + "answer": "You just passed through five doors.", + "question_chinese": "我刚刚经过了几扇门？", + "answer_chinese": "你刚刚经过了5扇门。", + "evidence": [ + { + "description": "passed five doors.", + "timestamps": { + "start": "00:05:22", + "end": "00:05:49" + } + } + ], + "question_timestamps": "00:07:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just passed through four doors.", + "B": "You just passed through six doors.", + "C": "You just passed through three doors while moving through the space, not counting the doorway you only looked through.", + "D": "You just passed through seven doors in total, including the last doorway you stepped through at the end.", + "E": "You just passed through five doors." + }, + "answer_label": "E", + "answer_text": "You just passed through five doors.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_4_T1", + "correct_answer": "You just passed through five doors.", + "original_options": { + "A": "You just passed through four doors.", + "B": "You just passed through six doors.", + "C": "You just passed through three doors while moving through the space, not counting the doorway you only looked through.", + "D": "You just passed through seven doors in total, including the last doorway you stepped through at the end.", + "E": "You just passed through five doors." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_4_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors did I just pass through?", + "answer": "You just passed through ten doors.", + "question_chinese": "我刚刚经过了几扇门？", + "answer_chinese": "你刚刚经过了10扇门。", + "evidence": [ + { + "description": "passed ten doors.", + "timestamps": { + "start": "00:05:22", + "end": "00:06:18" + } + } + ], + "question_timestamps": "00:16:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just passed through ten doors.", + "B": "You just passed through nine doors.", + "C": "You just passed through eleven doors.", + "D": "You just passed through eight doors, since one of the openings you approached wasn’t actually a door you went through.", + "E": "You just passed through twelve doors altogether, counting the final doorway you exited through right before stopping." + }, + "answer_label": "A", + "answer_text": "You just passed through ten doors.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_4_T2", + "correct_answer": "You just passed through ten doors.", + "original_options": { + "A": "You just passed through ten doors.", + "B": "You just passed through nine doors.", + "C": "You just passed through eleven doors.", + "D": "You just passed through eight doors, since one of the openings you approached wasn’t actually a door you went through.", + "E": "You just passed through twelve doors altogether, counting the final doorway you exited through right before stopping." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 17 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_2_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the cart in front of me empty, or is it full of things?", + "answer": "The cart in front of you is loaded with items.", + "question_chinese": "我前面的推车是空的还是放满东西的？", + "answer_chinese": "你面前的推车是放满东西的的。", + "evidence": [ + { + "description": "The cart is filled.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:40" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The cart in front of you is loaded with items.", + "B": "The cart in front of you is empty." + }, + "answer_label": "A", + "answer_text": "The cart in front of you is loaded with items.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_2_T1", + "correct_answer": "The cart in front of you is loaded with items.", + "original_options": { + "A": "The cart in front of you is loaded with items.", + "B": "The cart in front of you is empty." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_2_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the cart in front of me empty, or is it full of things?", + "answer": "The cart in front of you is empty.", + "question_chinese": "我前面的推车是空的还是放满东西的？", + "answer_chinese": "你面前的推车是空的。", + "evidence": [ + { + "description": "The cart is empty.", + "timestamps": { + "start": "00:06:13", + "end": "00:06:20" + } + } + ], + "question_timestamps": "00:06:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The cart in front of you is empty.", + "B": "The cart in front of you is full of things." + }, + "answer_label": "A", + "answer_text": "The cart in front of you is empty.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_2_T2", + "correct_answer": "The cart in front of you is empty.", + "original_options": { + "A": "The cart in front of you is empty.", + "B": "The cart in front of you is full of things." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_9_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs can I see in the scene?", + "answer": "You saw six chairs in the scene. The chairs are in the living room to your right.", + "question_chinese": "我可以看见场景里有几把椅子？", + "answer_chinese": "你看见场景里有六把椅子。椅子在你右边的客厅里。", + "evidence": [ + { + "description": "There is a small stool and a slightly higher soft stool next to the cooking counter.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:14" + } + }, + { + "description": "Four chairs were placed around the table.", + "timestamps": { + "start": "00:13:06", + "end": "00:13:07" + } + } + ], + "question_timestamps": "00:14:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw five chairs.", + "B": "You saw seven chairs.", + "C": "You saw four chairs in the scene, and they are in the living room to your right.", + "D": "You saw eight chairs in the scene, and the chairs are in the living room to your right.", + "E": "You saw six chairs in the scene. The chairs are in the living room to your right." + }, + "answer_label": "E", + "answer_text": "You saw six chairs in the scene. The chairs are in the living room to your right.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_9_T1", + "correct_answer": "You saw six chairs in the scene. The chairs are in the living room to your right.", + "original_options": { + "A": "You saw five chairs.", + "B": "You saw seven chairs.", + "C": "You saw four chairs in the scene, and they are in the living room to your right.", + "D": "You saw eight chairs in the scene, and the chairs are in the living room to your right.", + "E": "You saw six chairs in the scene. The chairs are in the living room to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_9_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs can I see in the scene?", + "answer": "There are six chairs in the scene.", + "question_chinese": "我可以看见场景里有几把椅子？", + "answer_chinese": "场景里有六把椅子。", + "evidence": [ + { + "description": "Four chairs were placed around the table.", + "timestamps": { + "start": "00:13:06", + "end": "00:13:07" + } + }, + { + "description": "There is a small stool and a slightly higher soft stool next to the cooking counter.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:14" + } + } + ], + "question_timestamps": "00:22:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are six chairs in the scene.", + "B": "There are five chairs.", + "C": "There are seven chairs.", + "D": "There are four chairs in the scene, arranged together near the living room area.", + "E": "There are eight chairs in the scene, visible around the living room and nearby space." + }, + "answer_label": "A", + "answer_text": "There are six chairs in the scene.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_9_T2", + "correct_answer": "There are six chairs in the scene.", + "original_options": { + "A": "There are six chairs in the scene.", + "B": "There are five chairs.", + "C": "There are seven chairs.", + "D": "There are four chairs in the scene, arranged together near the living room area.", + "E": "There are eight chairs in the scene, visible around the living room and nearby space." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_10_T1", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I'm sitting on the sofa directly facing the TV—how do I get to the balcony?", + "answer": "To get to the balcony, you need to stand up and then walk straight.", + "question_chinese": "我坐在电视正对面的沙发上，要怎么去阳台？", + "answer_chinese": "你要去阳台需要站起来然后直走。", + "evidence": [ + { + "description": "The balcony is on the left side of the living room.", + "timestamps": { + "start": "00:12:11", + "end": "00:12:14" + } + } + ], + "question_timestamps": "00:13:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Stand up and turn left.", + "B": "Stand up and turn right.", + "C": "To reach the balcony, stand up from the sofa, turn left toward the side, then walk forward.", + "D": "To get to the balcony, you need to stand up and then walk straight.", + "E": "To get to the balcony, stand up, turn around to face away from the TV, and walk straight ahead." + }, + "answer_label": "D", + "answer_text": "To get to the balcony, you need to stand up and then walk straight.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_10_T1", + "correct_answer": "To get to the balcony, you need to stand up and then walk straight.", + "original_options": { + "A": "Stand up and turn left.", + "B": "Stand up and turn right.", + "C": "To reach the balcony, stand up from the sofa, turn left toward the side, then walk forward.", + "D": "To get to the balcony, you need to stand up and then walk straight.", + "E": "To get to the balcony, stand up, turn around to face away from the TV, and walk straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_10_T2", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I'm sitting on the sofa directly facing the TV—how do I get to the balcony?", + "answer": "To get to the balcony, you need to stand up, turn left, and then walk straight.", + "question_chinese": "我坐在电视正对面的沙发上，要怎么去阳台？", + "answer_chinese": "去阳台需要站起来左转，然后直走。", + "evidence": [ + { + "description": "The balcony is on the left side of the living room.", + "timestamps": { + "start": "00:12:07", + "end": "00:12:12" + } + } + ], + "question_timestamps": "00:21:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Stand up and walk straight.", + "B": "Stand up and turn right.", + "C": "To get to the balcony, stand up from the sofa, turn right toward the other side, then walk straight.", + "D": "To reach the balcony, stand up, turn left once, then take two steps forward and stop.", + "E": "To get to the balcony, you need to stand up, turn left, and then walk straight." + }, + "answer_label": "E", + "answer_text": "To get to the balcony, you need to stand up, turn left, and then walk straight.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_10_T2", + "correct_answer": "To get to the balcony, you need to stand up, turn left, and then walk straight.", + "original_options": { + "A": "Stand up and walk straight.", + "B": "Stand up and turn right.", + "C": "To get to the balcony, stand up from the sofa, turn right toward the other side, then walk straight.", + "D": "To reach the balcony, stand up, turn left once, then take two steps forward and stop.", + "E": "To get to the balcony, you need to stand up, turn left, and then walk straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 16 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_3_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to eat a banana—where do I go to get bananas?", + "answer": "You need to get the bananas from the shelf to the right of the sink.", + "question_chinese": "我想吃香蕉，去哪里拿香蕉？", + "answer_chinese": "你要去洗手池右边的架子上拿香蕉。", + "evidence": [ + { + "description": "The bananas are on the shelf to the right of the sink.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:04:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the shelf left of the sink.", + "B": "From the shelf behind the sink.", + "C": "You need to get the bananas from the shelf to the left of the sink, near the counter area.", + "D": "You need to get the bananas from the shelf to the right of the sink.", + "E": "Go to the shelf in front of the sink to grab the bananas before returning to eat them." + }, + "answer_label": "D", + "answer_text": "You need to get the bananas from the shelf to the right of the sink.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_3_T1", + "correct_answer": "You need to get the bananas from the shelf to the right of the sink.", + "original_options": { + "A": "From the shelf left of the sink.", + "B": "From the shelf behind the sink.", + "C": "You need to get the bananas from the shelf to the left of the sink, near the counter area.", + "D": "You need to get the bananas from the shelf to the right of the sink.", + "E": "Go to the shelf in front of the sink to grab the bananas before returning to eat them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 18 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_6", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to eat bananas, where can I get bananas?", + "answer": "You need to get the bananas from the shelf next to the stove.", + "question_chinese": "我想吃香蕉，去哪里拿香蕉？", + "answer_chinese": "你要去灶台旁边的架子上拿香蕉。", + "evidence": [ + { + "description": "The bananas are placed on the shelf next to the stove.", + "timestamps": { + "start": "00:11:40", + "end": "00:11:43" + } + } + ], + "question_timestamps": "00:12:36", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the counter by the sink.", + "B": "In a drawer under the table.", + "C": "You need to get the bananas from the shelf next to the stove.", + "D": "You can get the bananas from the basket placed on the chair beside the dining area.", + "E": "Pick up the bananas from the shelf above the fridge, not from anywhere near the stove." + }, + "answer_label": "C", + "answer_text": "You need to get the bananas from the shelf next to the stove.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_6", + "correct_answer": "You need to get the bananas from the shelf next to the stove.", + "original_options": { + "A": "On the counter by the sink.", + "B": "In a drawer under the table.", + "C": "You need to get the bananas from the shelf next to the stove.", + "D": "You can get the bananas from the basket placed on the chair beside the dining area.", + "E": "Pick up the bananas from the shelf above the fridge, not from anywhere near the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_3_T2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to eat a banana—where do I go to get bananas?", + "answer": "You need to get the bananas from the shelf next to the stove.", + "question_chinese": "我想吃香蕉，去哪里拿香蕉？", + "answer_chinese": "你要去灶台旁边的架子上拿香蕉。", + "evidence": [ + { + "description": "The bananas are placed on the shelf next to the stove.", + "timestamps": { + "start": "00:11:40", + "end": "00:11:43" + } + } + ], + "question_timestamps": "00:20:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You need to get the bananas from the shelf next to the stove.", + "B": "From the shelf right of the sink.", + "C": "From a shelf above the fridge.", + "D": "You need to get the bananas from the counter next to the sink instead of the shelf by the stove.", + "E": "Go to the drawer beneath the table to find the bananas, then bring them back to eat." + }, + "answer_label": "A", + "answer_text": "You need to get the bananas from the shelf next to the stove.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_3_T2", + "correct_answer": "You need to get the bananas from the shelf next to the stove.", + "original_options": { + "A": "You need to get the bananas from the shelf next to the stove.", + "B": "From the shelf right of the sink.", + "C": "From a shelf above the fridge.", + "D": "You need to get the bananas from the counter next to the sink instead of the shelf by the stove.", + "E": "Go to the drawer beneath the table to find the bananas, then bring them back to eat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 17 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_1_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What foods did I see next to the stove?", + "answer": "You saw bananas next to the stove. The bananas are to your front-left.", + "question_chinese": "我看见灶台旁边有哪些食物？", + "answer_chinese": "你看见灶台旁边有香蕉。香蕉在你左前方。", + "evidence": [ + { + "description": "There are banans next to the stove.", + "timestamps": { + "start": "00:14:38", + "end": "00:14:40" + } + } + ], + "question_timestamps": "00:15:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw eggs next to the stove.", + "B": "You saw bananas to your front-right.", + "C": "You saw bananas next to the stove, but they were positioned to your front-right instead of front-left.", + "D": "You saw eggs next to the stove, and they were placed to your front-left near the counter edge.", + "E": "You saw bananas next to the stove. The bananas are to your front-left." + }, + "answer_label": "E", + "answer_text": "You saw bananas next to the stove. The bananas are to your front-left.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_1_T1", + "correct_answer": "You saw bananas next to the stove. The bananas are to your front-left.", + "original_options": { + "A": "You saw eggs next to the stove.", + "B": "You saw bananas to your front-right.", + "C": "You saw bananas next to the stove, but they were positioned to your front-right instead of front-left.", + "D": "You saw eggs next to the stove, and they were placed to your front-left near the counter edge.", + "E": "You saw bananas next to the stove. The bananas are to your front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 18 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_1_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What foods did I see next to the stove?", + "answer": "You saw eggs next to the stove, and the eggs are on your right-hand side.", + "question_chinese": "我看见灶台旁边有哪些食物？", + "answer_chinese": "你看见灶台旁边有鸡蛋，鸡蛋在你右手边", + "evidence": [ + { + "description": "There are 19 eggs next to the stove.", + "timestamps": { + "start": "00:14:36", + "end": "00:14:38" + } + } + ], + "question_timestamps": "00:19:52", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw eggs next to the stove, and the eggs are on your right-hand side.", + "B": "You saw eggs to your left.", + "C": "You saw eggs in front of you.", + "D": "You saw eggs next to the stove, but they were on your left-hand side rather than your right.", + "E": "You saw eggs next to the stove, and they were positioned behind you instead of on your right-hand side." + }, + "answer_label": "A", + "answer_text": "You saw eggs next to the stove, and the eggs are on your right-hand side.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_1_T2", + "correct_answer": "You saw eggs next to the stove, and the eggs are on your right-hand side.", + "original_options": { + "A": "You saw eggs next to the stove, and the eggs are on your right-hand side.", + "B": "You saw eggs to your left.", + "C": "You saw eggs in front of you.", + "D": "You saw eggs next to the stove, but they were on your left-hand side rather than your right.", + "E": "You saw eggs next to the stove, and they were positioned behind you instead of on your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_9", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I stand in front of the refrigerator, what is the straight-line distance to the stove?", + "answer": "You stand in front of the refrigerator, and the straight-line distance to the stove is five meters.", + "question_chinese": "我站在冰箱前，与灶台的直线距离是多少？", + "answer_chinese": "你站在冰箱前，与灶台的直线距离是五米。", + "evidence": [ + { + "description": "The stove is five meters to the right of the refrigerator.", + "timestamps": { + "start": "00:16:08", + "end": "00:16:12" + } + } + ], + "question_timestamps": "00:17:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three meters.", + "B": "Seven meters.", + "C": "You stand in front of the refrigerator, and the straight-line distance to the stove is five meters.", + "D": "You stand in front of the refrigerator, and the straight-line distance to the stove is three meters.", + "E": "You stand in front of the refrigerator, and the straight-line distance to the stove is eight meters." + }, + "answer_label": "C", + "answer_text": "You stand in front of the refrigerator, and the straight-line distance to the stove is five meters.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_9", + "correct_answer": "You stand in front of the refrigerator, and the straight-line distance to the stove is five meters.", + "original_options": { + "A": "Three meters.", + "B": "Seven meters.", + "C": "You stand in front of the refrigerator, and the straight-line distance to the stove is five meters.", + "D": "You stand in front of the refrigerator, and the straight-line distance to the stove is three meters.", + "E": "You stand in front of the refrigerator, and the straight-line distance to the stove is eight meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_10", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I stand in front of the refrigerator, what is the straight-line distance to the stove?", + "answer": "You stand in front of the refrigerator, and the straight-line distance to the stove is five meters.", + "question_chinese": "我站在冰箱前，与灶台的直线距离是多少？", + "answer_chinese": "你站在冰箱前，与灶台的直线距离是五米。", + "evidence": [ + { + "description": "The stove is five meters to the right of the refrigerator.", + "timestamps": { + "start": "00:16:08", + "end": "00:16:12" + } + } + ], + "question_timestamps": "00:22:44", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two meters.", + "B": "You stand in front of the refrigerator, and the straight-line distance to the stove is five meters.", + "C": "Eight meters.", + "D": "You stand in front of the refrigerator, and the straight-line distance to the stove is seven meters.", + "E": "You stand in front of the refrigerator, and the straight-line distance to the stove is three meters." + }, + "answer_label": "B", + "answer_text": "You stand in front of the refrigerator, and the straight-line distance to the stove is five meters.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_10", + "correct_answer": "You stand in front of the refrigerator, and the straight-line distance to the stove is five meters.", + "original_options": { + "A": "Two meters.", + "B": "You stand in front of the refrigerator, and the straight-line distance to the stove is five meters.", + "C": "Eight meters.", + "D": "You stand in front of the refrigerator, and the straight-line distance to the stove is seven meters.", + "E": "You stand in front of the refrigerator, and the straight-line distance to the stove is three meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_5_T1", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I am standing in front of the refrigerator, where is the stove relative to me?", + "answer": "You were standing in front of the fridge, and the stove was to your right.", + "question_chinese": "我站在冰箱前，灶台相对于我的位置是什么？", + "answer_chinese": "你站在冰箱前，灶台在你右边。", + "evidence": [ + { + "description": "The stove is on the right of the refrigerator.", + "timestamps": { + "start": "00:16:15", + "end": "00:16:17" + } + } + ], + "question_timestamps": "00:16:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The stove was to your left.", + "B": "The stove was behind you.", + "C": "You were standing in front of the fridge, and the stove was to your right.", + "D": "You were standing in front of the fridge, and the stove was directly in front of you, not to your right.", + "E": "You were standing in front of the fridge, and the stove was on your left-hand side instead of your right." + }, + "answer_label": "C", + "answer_text": "You were standing in front of the fridge, and the stove was to your right.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_5_T1", + "correct_answer": "You were standing in front of the fridge, and the stove was to your right.", + "original_options": { + "A": "The stove was to your left.", + "B": "The stove was behind you.", + "C": "You were standing in front of the fridge, and the stove was to your right.", + "D": "You were standing in front of the fridge, and the stove was directly in front of you, not to your right.", + "E": "You were standing in front of the fridge, and the stove was on your left-hand side instead of your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 20 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_5_T2", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I am standing in front of the refrigerator, where is the stove relative to me?", + "answer": "You were standing in front of the fridge, and the stove was directly in front of you.", + "question_chinese": "我站在冰箱前，灶台相对于我的位置是什么？", + "answer_chinese": "你站在冰箱前，灶台在你的正前方。", + "evidence": [ + { + "description": "The stove is right in front of you.", + "timestamps": { + "start": "00:16:54", + "end": "00:16:57" + } + } + ], + "question_timestamps": "00:16:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You were standing in front of the fridge, and the stove was directly in front of you.", + "B": "The stove was behind you.", + "C": "The stove was to your left.", + "D": "You were standing in front of the fridge, and the stove was to your right rather than directly in front.", + "E": "You were standing in front of the fridge, and the stove was directly behind you instead of directly in front of you." + }, + "answer_label": "A", + "answer_text": "You were standing in front of the fridge, and the stove was directly in front of you.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_5_T2", + "correct_answer": "You were standing in front of the fridge, and the stove was directly in front of you.", + "original_options": { + "A": "You were standing in front of the fridge, and the stove was directly in front of you.", + "B": "The stove was behind you.", + "C": "The stove was to your left.", + "D": "You were standing in front of the fridge, and the stove was to your right rather than directly in front.", + "E": "You were standing in front of the fridge, and the stove was directly behind you instead of directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 22 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_8_T2", + "qid_num": 27, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the sink on my left or my right?", + "answer": "The sink is to your left.", + "question_chinese": "水槽在我的左/右哪边？", + "answer_chinese": "水槽在你的左边。", + "evidence": [ + { + "description": "The sink is on your left.", + "timestamps": { + "start": "00:11:32", + "end": "00:11:36" + } + } + ], + "question_timestamps": "00:11:40", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sink is to your right.", + "B": "The sink is to your left." + }, + "answer_label": "B", + "answer_text": "The sink is to your left.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_8_T2", + "correct_answer": "The sink is to your left.", + "original_options": { + "A": "The sink is to your right.", + "B": "The sink is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_8_T1", + "qid_num": 28, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the sink on my left or my right?", + "answer": "The sink is to your right.", + "question_chinese": "水槽在我的左/右哪边？", + "answer_chinese": "水槽在你的右边。", + "evidence": [ + { + "description": "The sink is to the right of the fridge.", + "timestamps": { + "start": "00:11:25", + "end": "00:11:32" + } + } + ], + "question_timestamps": "00:11:52", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sink is to your right.", + "B": "The sink is to your left." + }, + "answer_label": "A", + "answer_text": "The sink is to your right.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_8_T1", + "correct_answer": "The sink is to your right.", + "original_options": { + "A": "The sink is to your right.", + "B": "The sink is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_1", + "qid_num": 29, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many eggs are there next to the stove?", + "answer": "There are nineteen eggs next to the stove.", + "question_chinese": "灶台旁边有几个鸡蛋？", + "answer_chinese": "灶台旁边有19个鸡蛋。", + "evidence": [ + { + "description": "There are 19 eggs next to the stove.", + "timestamps": { + "start": "00:14:36", + "end": "00:14:38" + } + } + ], + "question_timestamps": "00:15:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are eighteen eggs next to the stove.", + "B": "There are nineteen eggs next to the stove.", + "C": "There are twenty eggs next to the stove.", + "D": "There are sixteen eggs next to the stove, arranged in a couple of small clusters by the burner.", + "E": "There are twenty-one eggs next to the stove, spread out along the counter edge beside it." + }, + "answer_label": "B", + "answer_text": "There are nineteen eggs next to the stove.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_1", + "correct_answer": "There are nineteen eggs next to the stove.", + "original_options": { + "A": "There are eighteen eggs next to the stove.", + "B": "There are nineteen eggs next to the stove.", + "C": "There are twenty eggs next to the stove.", + "D": "There are sixteen eggs next to the stove, arranged in a couple of small clusters by the burner.", + "E": "There are twenty-one eggs next to the stove, spread out along the counter edge beside it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 16 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_2", + "qid_num": 30, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many eggs are there next to the stove?", + "answer": "There are nineteen eggs next to the stove.", + "question_chinese": "灶台旁边有几个鸡蛋？", + "answer_chinese": "灶台旁边有19个鸡蛋。", + "evidence": [ + { + "description": "There are 19 eggs next to the stove.", + "timestamps": { + "start": "00:14:36", + "end": "00:14:38" + } + } + ], + "question_timestamps": "00:19:52", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are seventeen eggs next to the stove.", + "B": "There are twenty-two eggs next to the stove.", + "C": "There are nineteen eggs next to the stove.", + "D": "There are fifteen eggs next to the stove, sitting together in a tight group beside the stovetop.", + "E": "There are twenty-four eggs next to the stove, lined up in several small groups along the counter." + }, + "answer_label": "C", + "answer_text": "There are nineteen eggs next to the stove.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_2", + "correct_answer": "There are nineteen eggs next to the stove.", + "original_options": { + "A": "There are seventeen eggs next to the stove.", + "B": "There are twenty-two eggs next to the stove.", + "C": "There are nineteen eggs next to the stove.", + "D": "There are fifteen eggs next to the stove, sitting together in a tight group beside the stovetop.", + "E": "There are twenty-four eggs next to the stove, lined up in several small groups along the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 17 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_15", + "qid_num": 31, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When facing the fridge, is the sink on my left/right?", + "answer": "When facing the fridge, the sink is on your right.", + "question_chinese": "面对冰箱时水槽在我的左/右？", + "answer_chinese": "面对冰箱时水槽在你的右边。", + "evidence": [ + { + "description": "The sink is to the right of the fridge.", + "timestamps": { + "start": "00:11:25", + "end": "00:11:32" + } + } + ], + "question_timestamps": "00:11:52", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When facing the fridge, the sink is on your right.", + "B": "When facing the fridge, the sink is on your left." + }, + "answer_label": "A", + "answer_text": "When facing the fridge, the sink is on your right.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_15", + "correct_answer": "When facing the fridge, the sink is on your right.", + "original_options": { + "A": "When facing the fridge, the sink is on your right.", + "B": "When facing the fridge, the sink is on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "6b70f1de-16cd-4b62-ab46-bc988254f486", + "video_length": "00:23:54", + "qid": "6b70f1de-16cd-4b62-ab46-bc988254f486_16", + "qid_num": 32, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When facing the fridge, is the sink on my left/right?", + "answer": "When facing the fridge, the sink is on your right.", + "question_chinese": "面对冰箱时水槽在我的左/右？", + "answer_chinese": "面对冰箱时水槽在你的右边。", + "evidence": [ + { + "description": "The sink is to the right of the fridge.", + "timestamps": { + "start": "00:11:25", + "end": "00:11:32" + } + } + ], + "question_timestamps": "00:21:11", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When facing the fridge, the sink is on your left.", + "B": "When facing the fridge, the sink is on your right." + }, + "answer_label": "B", + "answer_text": "When facing the fridge, the sink is on your right.", + "q_id": "6b70f1de-16cd-4b62-ab46-bc988254f486_16", + "correct_answer": "When facing the fridge, the sink is on your right.", + "original_options": { + "A": "When facing the fridge, the sink is on your left.", + "B": "When facing the fridge, the sink is on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + } +] diff --git a/QAs/6b7fc6db-e979-43a0-b61f-7dced89d423d-mcq.json b/QAs/6b7fc6db-e979-43a0-b61f-7dced89d423d-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6f700f6dd660f4214feb7643541e9cf796df7818 --- /dev/null +++ b/QAs/6b7fc6db-e979-43a0-b61f-7dced89d423d-mcq.json @@ -0,0 +1,1153 @@ +[ + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the scene?", + "answer": "Four trash cans appeared in the scene.", + "question_chinese": "场景里出现了几个垃圾桶？", + "answer_chinese": "场景里出现了四个垃圾桶。", + "evidence": [ + { + "description": "There are three trash cans on the left side of the aisle and one on the right.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:02:35", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three trash cans appeared in the scene.", + "B": "Four trash cans appeared in the scene.", + "C": "Five trash cans appeared in the scene.", + "D": "Two trash cans appeared in the scene, one near the counter and one closer to the doorway.", + "E": "Six trash cans appeared in the scene, visible in different spots around the room as you look around." + }, + "answer_label": "B", + "answer_text": "Four trash cans appeared in the scene.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_7", + "correct_answer": "Four trash cans appeared in the scene.", + "original_options": { + "A": "Three trash cans appeared in the scene.", + "B": "Four trash cans appeared in the scene.", + "C": "Five trash cans appeared in the scene.", + "D": "Two trash cans appeared in the scene, one near the counter and one closer to the doorway.", + "E": "Six trash cans appeared in the scene, visible in different spots around the room as you look around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the scene?", + "answer": "Four trash cans appeared in the scene.", + "question_chinese": "场景里出现了几个垃圾桶？", + "answer_chinese": "场景里出现了四个垃圾桶。", + "evidence": [ + { + "description": "There are three trash cans on the left side of the aisle and one on the right.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:15:17", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three trash cans appeared in the scene.", + "B": "Two trash cans appeared in the scene.", + "C": "Four trash cans appeared in the scene.", + "D": "Five trash cans appeared in the scene, with multiple bins visible along the wall and near the workstation.", + "E": "Seven trash cans appeared in the scene, spread around the area and noticeable at different angles." + }, + "answer_label": "C", + "answer_text": "Four trash cans appeared in the scene.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_8", + "correct_answer": "Four trash cans appeared in the scene.", + "original_options": { + "A": "Three trash cans appeared in the scene.", + "B": "Two trash cans appeared in the scene.", + "C": "Four trash cans appeared in the scene.", + "D": "Five trash cans appeared in the scene, with multiple bins visible along the wall and near the workstation.", + "E": "Seven trash cans appeared in the scene, spread around the area and noticeable at different angles." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 16 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_5", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I face the magazine shelf, is the ATM on my left or right?", + "answer": "When you're facing the magazine shelf, the ATM is on your right.", + "question_chinese": "当我面对着杂志书架，取款机在我的左/右？", + "answer_chinese": "当你面对着杂志书架，取款机在你的右边。", + "evidence": [ + { + "description": "The ATM is on my right.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:02:40", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you're facing the magazine shelf, the ATM is on your left.", + "B": "When you're facing the magazine shelf, the ATM is on your right." + }, + "answer_label": "B", + "answer_text": "When you're facing the magazine shelf, the ATM is on your right.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_5", + "correct_answer": "When you're facing the magazine shelf, the ATM is on your right.", + "original_options": { + "A": "When you're facing the magazine shelf, the ATM is on your left.", + "B": "When you're facing the magazine shelf, the ATM is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_6", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I face the magazine shelf, is the ATM on my left or right?", + "answer": "When you're facing the magazine shelf, the ATM is on your right.", + "question_chinese": "当我面对着杂志书架，取款机在我的左/右？", + "answer_chinese": "当你面对着杂志书架，取款机在你的右边。", + "evidence": [ + { + "description": "The ATM is on my right.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:19:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you're facing the magazine shelf, the ATM is on your left.", + "B": "When you're facing the magazine shelf, the ATM is on your right." + }, + "answer_label": "B", + "answer_text": "When you're facing the magazine shelf, the ATM is on your right.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_6", + "correct_answer": "When you're facing the magazine shelf, the ATM is on your right.", + "original_options": { + "A": "When you're facing the magazine shelf, the ATM is on your left.", + "B": "When you're facing the magazine shelf, the ATM is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_13", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was picking fruits, did I take any pears?", + "answer": "When you picked the fruit, you didn't take the pears.", + "question_chinese": "我挑选水果的时候有拿雪梨吗？", + "answer_chinese": "你挑选水果的时候没有拿雪梨。", + "evidence": [ + { + "description": "I took grapes, persimmons, mangoes, apples, and limes.", + "timestamps": { + "start": "00:14:00", + "end": "00:14:39" + } + } + ], + "question_timestamps": "00:15:14", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you picked the fruit, you did take the pears.", + "B": "When you picked the fruit, you didn't take the pears." + }, + "answer_label": "B", + "answer_text": "When you picked the fruit, you didn't take the pears.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_13", + "correct_answer": "When you picked the fruit, you didn't take the pears.", + "original_options": { + "A": "When you picked the fruit, you did take the pears.", + "B": "When you picked the fruit, you didn't take the pears." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_14", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was picking fruits, did I take any pears?", + "answer": "When you picked the fruit, you didn't take the pears.", + "question_chinese": "我挑选水果的时候有拿雪梨吗？", + "answer_chinese": "你挑选水果的时候没有拿雪梨。", + "evidence": [ + { + "description": "I took grapes, persimmons, mangoes, apples, and limes.", + "timestamps": { + "start": "00:14:00", + "end": "00:14:39" + } + } + ], + "question_timestamps": "00:22:29", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you picked the fruit, you did take the pears.", + "B": "When you picked the fruit, you didn't take the pears." + }, + "answer_label": "B", + "answer_text": "When you picked the fruit, you didn't take the pears.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_14", + "correct_answer": "When you picked the fruit, you didn't take the pears.", + "original_options": { + "A": "When you picked the fruit, you did take the pears.", + "B": "When you picked the fruit, you didn't take the pears." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_4_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I see appear in the scene?", + "answer": "You saw four trash cans appear in the scene. The trash cans were behind you.", + "question_chinese": "我看见场景里出现了几个垃圾桶？", + "answer_chinese": "你看见场景里出现了四个垃圾桶。垃圾桶在你的后边。", + "evidence": [ + { + "description": "There are three trash cans on the left side of the aisle and one on the right.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three trash cans appeared behind you.", + "B": "Five trash cans appeared behind you.", + "C": "You saw two trash cans appear in the scene, and they were positioned behind you.", + "D": "You saw four trash cans appear in the scene. The trash cans were behind you.", + "E": "You saw six trash cans appear in the scene, and all of the trash cans were behind you." + }, + "answer_label": "D", + "answer_text": "You saw four trash cans appear in the scene. The trash cans were behind you.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_4_T2", + "correct_answer": "You saw four trash cans appear in the scene. The trash cans were behind you.", + "original_options": { + "A": "Three trash cans appeared behind you.", + "B": "Five trash cans appeared behind you.", + "C": "You saw two trash cans appear in the scene, and they were positioned behind you.", + "D": "You saw four trash cans appear in the scene. The trash cans were behind you.", + "E": "You saw six trash cans appear in the scene, and all of the trash cans were behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 18 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_4_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I see appear in the scene?", + "answer": "You saw three trash cans appear in the scene. The trash cans were to your left.", + "question_chinese": "我看见场景里出现了几个垃圾桶？", + "answer_chinese": "你看见场景里出现了三个垃圾桶。垃圾桶在你的左边。", + "evidence": [ + { + "description": "There are three trash cans on the left side of the aisle and one on the right.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:02:29", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two trash cans appeared to your left.", + "B": "Four trash cans appeared to your left.", + "C": "You saw three trash cans appear in the scene. The trash cans were to your left.", + "D": "You saw one trash can appear in the scene, and it was located to your left.", + "E": "You saw five trash cans appear in the scene, and the trash cans were to your left." + }, + "answer_label": "C", + "answer_text": "You saw three trash cans appear in the scene. The trash cans were to your left.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_4_T1", + "correct_answer": "You saw three trash cans appear in the scene. The trash cans were to your left.", + "original_options": { + "A": "Two trash cans appeared to your left.", + "B": "Four trash cans appeared to your left.", + "C": "You saw three trash cans appear in the scene. The trash cans were to your left.", + "D": "You saw one trash can appear in the scene, and it was located to your left.", + "E": "You saw five trash cans appear in the scene, and the trash cans were to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_11", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I'm standing inside the convenience store, how do I get to the clothing store?", + "answer": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn left.", + "question_chinese": "我站在便利店里面，要怎么去衣服店？", + "answer_chinese": "你站在便利店里面，去衣服店需要出门口左拐。", + "evidence": [ + { + "description": "The clothing store is to the left of the convenience store.", + "timestamps": { + "start": "00:07:40", + "end": "00:07:57" + } + } + ], + "question_timestamps": "00:08:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Exit and turn right.", + "B": "Stay inside and turn left.", + "C": "From inside the convenience store, exit the door and then turn right to reach the clothing store.", + "D": "While inside the convenience store, walk straight to the exit, go outside, and turn right toward the clothing store.", + "E": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn left." + }, + "answer_label": "E", + "answer_text": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn left.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_11", + "correct_answer": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn left.", + "original_options": { + "A": "Exit and turn right.", + "B": "Stay inside and turn left.", + "C": "From inside the convenience store, exit the door and then turn right to reach the clothing store.", + "D": "While inside the convenience store, walk straight to the exit, go outside, and turn right toward the clothing store.", + "E": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 19 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_12", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I'm standing inside the convenience store, how do I get to the clothing store?", + "answer": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn left.", + "question_chinese": "我站在便利店里面，要怎么去衣服店？", + "answer_chinese": "你站在便利店里面，去衣服店需要出门口左拐。", + "evidence": [ + { + "description": "The clothing store is to the left of the convenience store.", + "timestamps": { + "start": "00:07:40", + "end": "00:07:57" + } + } + ], + "question_timestamps": "00:21:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn left.", + "B": "Exit and turn right.", + "C": "Exit and cross the street.", + "D": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn right.", + "E": "You are inside the convenience store, and to get to the clothing store, you need to exit and go up the stairs." + }, + "answer_label": "A", + "answer_text": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn left.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_12", + "correct_answer": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn left.", + "original_options": { + "A": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn left.", + "B": "Exit and turn right.", + "C": "Exit and cross the street.", + "D": "You are inside the convenience store, and to get to the clothing store, you need to exit and turn right.", + "E": "You are inside the convenience store, and to get to the clothing store, you need to exit and go up the stairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 20, + 22 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_3", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I stood at the cash register, how far is the straight-line distance to the fridge？", + "answer": "You stand at the cash register, with a straight-line distance of six meters to the fridge.", + "question_chinese": "我站在收银台处，距离冰箱的直线距离有多远？", + "answer_chinese": "你站在收银台处，距离冰箱的直线距离有六米。", + "evidence": [ + { + "description": "The fridge is diagonally across from me.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:01:56", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five meters to the fridge.", + "B": "Seven meters to the fridge.", + "C": "You stand at the cash register, with a straight-line distance of eight meters to the fridge.", + "D": "You stand at the cash register, with a straight-line distance of six meters to the fridge.", + "E": "You stand at the cash register, with a straight-line distance of nine meters to the fridge." + }, + "answer_label": "D", + "answer_text": "You stand at the cash register, with a straight-line distance of six meters to the fridge.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_3", + "correct_answer": "You stand at the cash register, with a straight-line distance of six meters to the fridge.", + "original_options": { + "A": "Five meters to the fridge.", + "B": "Seven meters to the fridge.", + "C": "You stand at the cash register, with a straight-line distance of eight meters to the fridge.", + "D": "You stand at the cash register, with a straight-line distance of six meters to the fridge.", + "E": "You stand at the cash register, with a straight-line distance of nine meters to the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_4", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I stood at the cash register, how far is the straight-line distance to the fridge？", + "answer": "You stand at the cash register, with a straight-line distance of six meters to the fridge.", + "question_chinese": "我站在收银台处，距离冰箱的直线距离有多远？", + "answer_chinese": "你站在收银台处，距离冰箱的直线距离有六米。", + "evidence": [ + { + "description": "The fridge is diagonally across from me.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:17:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You stand at the cash register, with a straight-line distance of six meters to the fridge.", + "B": "Five meters to the fridge.", + "C": "Seven meters to the fridge.", + "D": "You stand at the cash register, with a straight-line distance of eight meters to the fridge.", + "E": "You stand at the cash register, with a straight-line distance of nine meters to the fridge." + }, + "answer_label": "A", + "answer_text": "You stand at the cash register, with a straight-line distance of six meters to the fridge.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_4", + "correct_answer": "You stand at the cash register, with a straight-line distance of six meters to the fridge.", + "original_options": { + "A": "You stand at the cash register, with a straight-line distance of six meters to the fridge.", + "B": "Five meters to the fridge.", + "C": "Seven meters to the fridge.", + "D": "You stand at the cash register, with a straight-line distance of eight meters to the fridge.", + "E": "You stand at the cash register, with a straight-line distance of nine meters to the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_8_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I’m standing in front of the apple—can I reach the eggplant?", + "answer": "You were standing in front of the apples, and you couldn’t touch the eggplant. The eggplant was in front of you.", + "question_chinese": "我站在苹果前面，我能摸得到茄子吗？", + "answer_chinese": "你站在苹果前面，你不能摸得到茄子。茄子在你的前面。", + "evidence": [ + { + "description": "The eggplants are on another shelf to the right.", + "timestamps": { + "start": "00:14:36", + "end": "00:14:42" + } + } + ], + "question_timestamps": "00:14:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes, you could touch the eggplant.", + "B": "You were standing in front of the apples, and you couldn’t touch the eggplant. The eggplant was in front of you.", + "C": "No, the eggplant was behind you.", + "D": "You were behind the apples, and you still couldn’t reach the eggplant even though it was in front of you.", + "E": "You were in front of the apples, and you could touch the eggplant because it was close in front of you." + }, + "answer_label": "B", + "answer_text": "You were standing in front of the apples, and you couldn’t touch the eggplant. The eggplant was in front of you.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_8_T1", + "correct_answer": "You were standing in front of the apples, and you couldn’t touch the eggplant. The eggplant was in front of you.", + "original_options": { + "A": "Yes, you could touch the eggplant.", + "B": "You were standing in front of the apples, and you couldn’t touch the eggplant. The eggplant was in front of you.", + "C": "No, the eggplant was behind you.", + "D": "You were behind the apples, and you still couldn’t reach the eggplant even though it was in front of you.", + "E": "You were in front of the apples, and you could touch the eggplant because it was close in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 21 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_8_T2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I’m standing in front of the apple—can I reach the eggplant?", + "answer": "You were standing in front of the apples, and you could touch the eggplant. The eggplant was behind you.", + "question_chinese": "我站在苹果前面，我能摸得到茄子吗？", + "answer_chinese": "你站在苹果前面，你能摸得到茄子。茄子在你的后面。", + "evidence": [ + { + "description": "You stand in front of the apples and can touch the eggplants, which are in the basket beneath the apples.", + "timestamps": { + "start": "00:14:42", + "end": "00:15:18" + } + } + ], + "question_timestamps": "00:15:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No, you couldn’t touch it.", + "B": "Yes, it was in front of you.", + "C": "You were standing in front of the apples, and you could touch the eggplant. The eggplant was behind you.", + "D": "You were standing in front of the apples, and you couldn’t touch the eggplant even though it was behind you.", + "E": "You were standing behind the apples, and you could touch the eggplant with it positioned behind you." + }, + "answer_label": "C", + "answer_text": "You were standing in front of the apples, and you could touch the eggplant. The eggplant was behind you.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_8_T2", + "correct_answer": "You were standing in front of the apples, and you could touch the eggplant. The eggplant was behind you.", + "original_options": { + "A": "No, you couldn’t touch it.", + "B": "Yes, it was in front of you.", + "C": "You were standing in front of the apples, and you could touch the eggplant. The eggplant was behind you.", + "D": "You were standing in front of the apples, and you couldn’t touch the eggplant even though it was behind you.", + "E": "You were standing behind the apples, and you could touch the eggplant with it positioned behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 17 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_15", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the eggplant when I stand in front of the apple?", + "answer": "You stand in front of an apple but can't reach the eggplant.", + "question_chinese": "我站在苹果前面，我能摸得到茄子吗？", + "answer_chinese": "你站在苹果前面，不能摸得到茄子。", + "evidence": [ + { + "description": "The eggplants are on another shelf to the right.", + "timestamps": { + "start": "00:14:36", + "end": "00:14:42" + } + } + ], + "question_timestamps": "00:15:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Yes, you can touch the eggplant.", + "B": "No, it’s behind you.", + "C": "You stand in front of the apple and can reach the eggplant because it’s right within arm’s length.", + "D": "You stand in front of an apple but can't reach the eggplant.", + "E": "You stand in front of the apple but the eggplant is behind you, so you can still touch it." + }, + "answer_label": "D", + "answer_text": "You stand in front of an apple but can't reach the eggplant.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_15", + "correct_answer": "You stand in front of an apple but can't reach the eggplant.", + "original_options": { + "A": "Yes, you can touch the eggplant.", + "B": "No, it’s behind you.", + "C": "You stand in front of the apple and can reach the eggplant because it’s right within arm’s length.", + "D": "You stand in front of an apple but can't reach the eggplant.", + "E": "You stand in front of the apple but the eggplant is behind you, so you can still touch it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 19 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_16", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the eggplant when I stand in front of the apple?", + "answer": "You stand in front of an apple but can't reach the eggplant.", + "question_chinese": "我站在苹果前面，我能摸得到茄子吗？", + "answer_chinese": "你站在苹果前面，不能摸得到茄子。", + "evidence": [ + { + "description": "The eggplants are on another shelf to the right.", + "timestamps": { + "start": "00:14:36", + "end": "00:14:42" + } + } + ], + "question_timestamps": "00:22:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You stand in front of an apple but can't reach the eggplant.", + "B": "Yes, you can touch it.", + "C": "No, it’s behind you.", + "D": "You stand in front of the apple and can reach the eggplant because it’s placed close in front of you.", + "E": "You stand behind the apple and can touch the eggplant, since it’s positioned within easy reach." + }, + "answer_label": "A", + "answer_text": "You stand in front of an apple but can't reach the eggplant.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_16", + "correct_answer": "You stand in front of an apple but can't reach the eggplant.", + "original_options": { + "A": "You stand in front of an apple but can't reach the eggplant.", + "B": "Yes, you can touch it.", + "C": "No, it’s behind you.", + "D": "You stand in front of the apple and can reach the eggplant because it’s placed close in front of you.", + "E": "You stand behind the apple and can touch the eggplant, since it’s positioned within easy reach." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 17 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_1_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stacks of cash can I see on the counter?", + "answer": "You could see six stacks of cash on the counter. The cash was in front of you.", + "question_chinese": "我能看见柜台上放着多少沓现金？", + "answer_chinese": "你能看见柜台上放着六沓现金。现金在你的前面。", + "evidence": [ + { + "description": "Six stacks of cash are placed on the counter.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four stacks of cash.", + "B": "Eight stacks of cash.", + "C": "You could see five stacks of cash on the counter, and the cash was in front of you.", + "D": "You could see seven stacks of cash on the counter, with the stacks positioned in front of you.", + "E": "You could see six stacks of cash on the counter. The cash was in front of you." + }, + "answer_label": "E", + "answer_text": "You could see six stacks of cash on the counter. The cash was in front of you.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_1_T1", + "correct_answer": "You could see six stacks of cash on the counter. The cash was in front of you.", + "original_options": { + "A": "Four stacks of cash.", + "B": "Eight stacks of cash.", + "C": "You could see five stacks of cash on the counter, and the cash was in front of you.", + "D": "You could see seven stacks of cash on the counter, with the stacks positioned in front of you.", + "E": "You could see six stacks of cash on the counter. The cash was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 18 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_1_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stacks of cash can I see on the counter?", + "answer": "You could see three stacks of cash on the counter. The cash was to your left.", + "question_chinese": "我能看见柜台上放着多少沓现金？", + "answer_chinese": "你能看见柜台上放着三沓现金。现金在你的左边。", + "evidence": [ + { + "description": "Three stacks of cash are placed on the counter.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You could see three stacks of cash on the counter. The cash was to your left.", + "B": "Two stacks of cash.", + "C": "Five stacks of cash.", + "D": "You could see one stack of cash on the counter, and the cash was to your left.", + "E": "You could see four stacks of cash on the counter, with the cash located to your left." + }, + "answer_label": "A", + "answer_text": "You could see three stacks of cash on the counter. The cash was to your left.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_1_T2", + "correct_answer": "You could see three stacks of cash on the counter. The cash was to your left.", + "original_options": { + "A": "You could see three stacks of cash on the counter. The cash was to your left.", + "B": "Two stacks of cash.", + "C": "Five stacks of cash.", + "D": "You could see one stack of cash on the counter, and the cash was to your left.", + "E": "You could see four stacks of cash on the counter, with the cash located to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_9", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many green magazines are there on the bookshelf for magazines?", + "answer": "There are four green magazines on the bookshelf for magazines.", + "question_chinese": "放杂志的书架上有几本绿色的杂志？", + "answer_chinese": "放杂志的书架上有四本绿色的杂志。", + "evidence": [ + { + "description": "There are four green magazines on the bookshelf for magazines.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:38" + } + } + ], + "question_timestamps": "00:01:41", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two green magazines.", + "B": "Six green magazines.", + "C": "There are four green magazines on the bookshelf for magazines.", + "D": "There are five green magazines on the bookshelf for magazines, not four as expected.", + "E": "There are seven green magazines on the bookshelf for magazines, filling more slots than you recall." + }, + "answer_label": "C", + "answer_text": "There are four green magazines on the bookshelf for magazines.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_9", + "correct_answer": "There are four green magazines on the bookshelf for magazines.", + "original_options": { + "A": "Two green magazines.", + "B": "Six green magazines.", + "C": "There are four green magazines on the bookshelf for magazines.", + "D": "There are five green magazines on the bookshelf for magazines, not four as expected.", + "E": "There are seven green magazines on the bookshelf for magazines, filling more slots than you recall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 16 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_10", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many green magazines are there on the bookshelf for magazines?", + "answer": "There is a green magazine on the shelf for magazines.", + "question_chinese": "放杂志的书架上有几本绿色的杂志？", + "answer_chinese": "放杂志的书架上有一本绿色的杂志。", + "evidence": [ + { + "description": "There are four green magazines on the bookshelf for magazines.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:38" + } + }, + { + "description": "I took a green magazine.", + "timestamps": { + "start": "00:01:44", + "end": "00:01:47" + } + }, + { + "description": "I took a green magazine.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:19" + } + }, + { + "description": "Another girl took a green magazine.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There is a green magazine on the shelf for magazines.", + "B": "There are two green magazines.", + "C": "There are no green magazines.", + "D": "There are three green magazines on the shelf for magazines, rather than just a single one.", + "E": "There are four green magazines on the shelf for magazines, so you see more than one." + }, + "answer_label": "A", + "answer_text": "There is a green magazine on the shelf for magazines.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_10", + "correct_answer": "There is a green magazine on the shelf for magazines.", + "original_options": { + "A": "There is a green magazine on the shelf for magazines.", + "B": "There are two green magazines.", + "C": "There are no green magazines.", + "D": "There are three green magazines on the shelf for magazines, rather than just a single one.", + "E": "There are four green magazines on the shelf for magazines, so you see more than one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stacks of cash are on the counter?", + "answer": "Six stacks of cash are placed on the counter.", + "question_chinese": "柜台上放着多少沓现金？", + "answer_chinese": "柜台上放着六沓现金。", + "evidence": [ + { + "description": "Six stacks of cash are placed on the counter.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five stacks of cash.", + "B": "Seven stacks of cash.", + "C": "There are four stacks of cash placed on the counter in front of you.", + "D": "Eight stacks of cash are lined up on the counter when you look over.", + "E": "Six stacks of cash are placed on the counter." + }, + "answer_label": "E", + "answer_text": "Six stacks of cash are placed on the counter.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_2", + "correct_answer": "Six stacks of cash are placed on the counter.", + "original_options": { + "A": "Five stacks of cash.", + "B": "Seven stacks of cash.", + "C": "There are four stacks of cash placed on the counter in front of you.", + "D": "Eight stacks of cash are lined up on the counter when you look over.", + "E": "Six stacks of cash are placed on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "6b7fc6db-e979-43a0-b61f-7dced89d423d", + "video_length": "00:22:38", + "qid": "6b7fc6db-e979-43a0-b61f-7dced89d423d_1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stacks of cash are on the counter?", + "answer": "Six stacks of cash are placed on the counter.", + "question_chinese": "柜台上放着多少沓现金？", + "answer_chinese": "柜台上放着六沓现金。", + "evidence": [ + { + "description": "Six stacks of cash are placed on the counter.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:10:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five stacks of cash.", + "B": "Seven stacks of cash.", + "C": "Only two stacks of cash are placed on the counter in front of you.", + "D": "Six stacks of cash are placed on the counter.", + "E": "Eight stacks of cash are sitting on the counter when you check again." + }, + "answer_label": "D", + "answer_text": "Six stacks of cash are placed on the counter.", + "q_id": "6b7fc6db-e979-43a0-b61f-7dced89d423d_1", + "correct_answer": "Six stacks of cash are placed on the counter.", + "original_options": { + "A": "Five stacks of cash.", + "B": "Seven stacks of cash.", + "C": "Only two stacks of cash are placed on the counter in front of you.", + "D": "Six stacks of cash are placed on the counter.", + "E": "Eight stacks of cash are sitting on the counter when you check again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 13 + ] + } +] diff --git a/QAs/6baa673a-0cc6-4da2-a481-877de0d2faa5-mcq.json b/QAs/6baa673a-0cc6-4da2-a481-877de0d2faa5-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..cb69a017b220a779de129bd6a1cfd4d17a0824e8 --- /dev/null +++ b/QAs/6baa673a-0cc6-4da2-a481-877de0d2faa5-mcq.json @@ -0,0 +1,602 @@ +[ + { + "video_uid": "6baa673a-0cc6-4da2-a481-877de0d2faa5", + "video_length": "00:20:19", + "qid": "6baa673a-0cc6-4da2-a481-877de0d2faa5_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was sitting on the ground, where did I put the scissors relative to me?", + "answer": "Ahead of you.", + "question_chinese": "当我坐在地上时，我把剪刀放在相对我什么位置了？", + "answer_chinese": "在你前方。", + "evidence": [ + { + "description": "There is a pair of scissors on the ground in front of you.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your right.", + "C": "Ahead of you.", + "D": "You put the scissors on the ground to your left side while sitting.", + "E": "You placed the scissors slightly behind you on the ground when you sat down." + }, + "answer_label": "C", + "answer_text": "Ahead of you.", + "q_id": "6baa673a-0cc6-4da2-a481-877de0d2faa5_1", + "correct_answer": "Ahead of you.", + "original_options": { + "A": "Behind you.", + "B": "To your right.", + "C": "Ahead of you.", + "D": "You put the scissors on the ground to your left side while sitting.", + "E": "You placed the scissors slightly behind you on the ground when you sat down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 14 + ] + }, + { + "video_uid": "6baa673a-0cc6-4da2-a481-877de0d2faa5", + "video_length": "00:20:19", + "qid": "6baa673a-0cc6-4da2-a481-877de0d2faa5_2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was sitting on the ground, where did I put the scissors relative to me?", + "answer": "To your left front.", + "question_chinese": "当我坐在地上时，我把剪刀放在相对我什么位置了？", + "answer_chinese": "在你左前方。", + "evidence": [ + { + "description": "You just moved the scissors to your left front.", + "timestamps": { + "start": "00:05:49", + "end": "00:05:51" + } + } + ], + "question_timestamps": "00:07:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To your right front.", + "B": "To your left front.", + "C": "Behind you.", + "D": "You put the scissors behind your left side while you were sitting on the ground.", + "E": "You placed the scissors to your right rear, slightly behind you as you sat." + }, + "answer_label": "B", + "answer_text": "To your left front.", + "q_id": "6baa673a-0cc6-4da2-a481-877de0d2faa5_2", + "correct_answer": "To your left front.", + "original_options": { + "A": "To your right front.", + "B": "To your left front.", + "C": "Behind you.", + "D": "You put the scissors behind your left side while you were sitting on the ground.", + "E": "You placed the scissors to your right rear, slightly behind you as you sat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 15, + 14 + ] + }, + { + "video_uid": "6baa673a-0cc6-4da2-a481-877de0d2faa5", + "video_length": "00:20:19", + "qid": "6baa673a-0cc6-4da2-a481-877de0d2faa5_6", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I adjusted the first leg of the pants, how many pins did I use in total?", + "answer": "You used a total of 4 paper clips.", + "question_chinese": "当我调整第一个裤腿时，我一共使用了多少个大头针？", + "answer_chinese": "你一共使用了4个大头针。", + "evidence": [ + { + "description": "You used the first pin.", + "timestamps": { + "start": "00:02:32", + "end": "00:02:36" + } + }, + { + "description": "You used the second pin.", + "timestamps": { + "start": "00:02:58", + "end": "00:03:04" + } + }, + { + "description": "You used the third pin.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:41" + } + }, + { + "description": "You used the fourth pin.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:04:18", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three paper clips.", + "B": "Five paper clips.", + "C": "You used a total of two paper clips when adjusting the first leg of the pants.", + "D": "You used a total of six paper clips in all while adjusting the first pant leg.", + "E": "You used a total of 4 paper clips." + }, + "answer_label": "E", + "answer_text": "You used a total of 4 paper clips.", + "q_id": "6baa673a-0cc6-4da2-a481-877de0d2faa5_6", + "correct_answer": "You used a total of 4 paper clips.", + "original_options": { + "A": "Three paper clips.", + "B": "Five paper clips.", + "C": "You used a total of two paper clips when adjusting the first leg of the pants.", + "D": "You used a total of six paper clips in all while adjusting the first pant leg.", + "E": "You used a total of 4 paper clips." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "6baa673a-0cc6-4da2-a481-877de0d2faa5", + "video_length": "00:20:19", + "qid": "6baa673a-0cc6-4da2-a481-877de0d2faa5_5", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I adjusted the first leg of the pants, how many pins did I use in total?", + "answer": "You used a total of 8 paper clips.", + "question_chinese": "当我调整第一个裤腿时，我一共使用了多少个大头针？", + "answer_chinese": "你一共使用了8个大头针。", + "evidence": [ + { + "description": "You used the first pin.", + "timestamps": { + "start": "00:02:32", + "end": "00:02:36" + } + }, + { + "description": "You used the second pin.", + "timestamps": { + "start": "00:02:58", + "end": "00:03:04" + } + }, + { + "description": "You used the third pin.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:41" + } + }, + { + "description": "You used the fourth pin.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:10" + } + }, + { + "description": "You used the fifth pin.", + "timestamps": { + "start": "00:04:39", + "end": "00:04:42" + } + }, + { + "description": "You used the sixth pin.", + "timestamps": { + "start": "00:04:48", + "end": "00:04:51" + } + }, + { + "description": "You used the seventh pin.", + "timestamps": { + "start": "00:05:13", + "end": "00:05:16" + } + }, + { + "description": "You used the eighth pin.", + "timestamps": { + "start": "00:05:29", + "end": "00:05:32" + } + } + ], + "question_timestamps": "00:15:57", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Seven paper clips.", + "B": "Nine paper clips.", + "C": "You used a total of six paper clips when adjusting the first leg of the pants.", + "D": "You used a total of 8 paper clips.", + "E": "You used a total of ten paper clips in all while adjusting the first pant leg." + }, + "answer_label": "D", + "answer_text": "You used a total of 8 paper clips.", + "q_id": "6baa673a-0cc6-4da2-a481-877de0d2faa5_5", + "correct_answer": "You used a total of 8 paper clips.", + "original_options": { + "A": "Seven paper clips.", + "B": "Nine paper clips.", + "C": "You used a total of six paper clips when adjusting the first leg of the pants.", + "D": "You used a total of 8 paper clips.", + "E": "You used a total of ten paper clips in all while adjusting the first pant leg." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "6baa673a-0cc6-4da2-a481-877de0d2faa5", + "video_length": "00:20:19", + "qid": "6baa673a-0cc6-4da2-a481-877de0d2faa5_5_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take just now to go use the sewing machine?", + "answer": "You got up from the floor, then went straight two steps to the left to the sewing machine.", + "question_chinese": "我刚才去用缝纫机走了什么路线?", + "answer_chinese": "你从地上起身,向左直走两步来到缝纫机前。", + "evidence": [ + { + "description": "You just got up from the ground and took two steps to the left to reach the sewing machine.", + "timestamps": { + "start": "00:10:39", + "end": "00:10:44" + } + } + ], + "question_timestamps": "00:10:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went one step left.", + "B": "You got up from the floor, then went straight two steps to the left to the sewing machine.", + "C": "You went three steps left.", + "D": "You got up from the floor, then moved four steps to the left to reach the sewing machine.", + "E": "You got up from the floor and didn’t step left before arriving at the sewing machine." + }, + "answer_label": "B", + "answer_text": "You got up from the floor, then went straight two steps to the left to the sewing machine.", + "q_id": "6baa673a-0cc6-4da2-a481-877de0d2faa5_5_T1", + "correct_answer": "You got up from the floor, then went straight two steps to the left to the sewing machine.", + "original_options": { + "A": "You went one step left.", + "B": "You got up from the floor, then went straight two steps to the left to the sewing machine.", + "C": "You went three steps left.", + "D": "You got up from the floor, then moved four steps to the left to reach the sewing machine.", + "E": "You got up from the floor and didn’t step left before arriving at the sewing machine." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "6baa673a-0cc6-4da2-a481-877de0d2faa5", + "video_length": "00:20:19", + "qid": "6baa673a-0cc6-4da2-a481-877de0d2faa5_5_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take just now to go use the sewing machine?", + "answer": "You stood up from the floor, walked straight two steps to the left, and sat down in front of the sewing machine.", + "question_chinese": "我刚才去用缝纫机走了什么路线?", + "answer_chinese": "你从地上起身,向左直走两步来到缝纫机前坐下。", + "evidence": [ + { + "description": "You just got up from the ground and took two steps to the left to reach the sewing machine.", + "timestamps": { + "start": "00:10:39", + "end": "00:10:44" + } + } + ], + "question_timestamps": "00:17:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked one step left.", + "B": "You walked two steps right.", + "C": "You stood up from the floor, walked straight two steps to the left, and sat down in front of the sewing machine.", + "D": "You stood up from the floor, walked three steps to the left, and sat down in front of the sewing machine.", + "E": "You stood up from the floor, walked two steps to the left, and stayed standing in front of the sewing machine." + }, + "answer_label": "C", + "answer_text": "You stood up from the floor, walked straight two steps to the left, and sat down in front of the sewing machine.", + "q_id": "6baa673a-0cc6-4da2-a481-877de0d2faa5_5_T2", + "correct_answer": "You stood up from the floor, walked straight two steps to the left, and sat down in front of the sewing machine.", + "original_options": { + "A": "You walked one step left.", + "B": "You walked two steps right.", + "C": "You stood up from the floor, walked straight two steps to the left, and sat down in front of the sewing machine.", + "D": "You stood up from the floor, walked three steps to the left, and sat down in front of the sewing machine.", + "E": "You stood up from the floor, walked two steps to the left, and stayed standing in front of the sewing machine." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 21 + ] + }, + { + "video_uid": "6baa673a-0cc6-4da2-a481-877de0d2faa5", + "video_length": "00:20:19", + "qid": "6baa673a-0cc6-4da2-a481-877de0d2faa5_6_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I’m done using the foot pedal—where should I put it?", + "answer": "The foot pedal should be put back into the gap of the sewing machine to your left.", + "question_chinese": "我用完脚踏板了，它应该放在哪里？", + "answer_chinese": "脚踏板应该放回你左边的缝纫机的空隙里。", + "evidence": [ + { + "description": "You just took out the pedal from the gap of the sewing machine.", + "timestamps": { + "start": "00:10:56", + "end": "00:10:58" + } + } + ], + "question_timestamps": "00:11:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it in the gap right.", + "B": "Put it in the gap ahead.", + "C": "The foot pedal should be put back into the gap of the sewing machine to your left.", + "D": "Put the foot pedal into the sewing machine gap behind you, not the one on your left.", + "E": "Slide the foot pedal into the sewing machine gap on your right, rather than putting it back on the left." + }, + "answer_label": "C", + "answer_text": "The foot pedal should be put back into the gap of the sewing machine to your left.", + "q_id": "6baa673a-0cc6-4da2-a481-877de0d2faa5_6_T1", + "correct_answer": "The foot pedal should be put back into the gap of the sewing machine to your left.", + "original_options": { + "A": "Put it in the gap right.", + "B": "Put it in the gap ahead.", + "C": "The foot pedal should be put back into the gap of the sewing machine to your left.", + "D": "Put the foot pedal into the sewing machine gap behind you, not the one on your left.", + "E": "Slide the foot pedal into the sewing machine gap on your right, rather than putting it back on the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 20 + ] + }, + { + "video_uid": "6baa673a-0cc6-4da2-a481-877de0d2faa5", + "video_length": "00:20:19", + "qid": "6baa673a-0cc6-4da2-a481-877de0d2faa5_6_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I’m done using the foot pedal—where should I put it?", + "answer": "The foot pedal should be put back into the gap of the sewing machine in front of you.", + "question_chinese": "我用完脚踏板了，它应该放在哪里？", + "answer_chinese": "脚踏板应该放回你前方的缝纫机的空隙里。", + "evidence": [ + { + "description": "You just took out the pedal from the gap of the sewing machine.", + "timestamps": { + "start": "00:10:56", + "end": "00:10:58" + } + } + ], + "question_timestamps": "00:15:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it in the gap left.", + "B": "Put it in the gap behind.", + "C": "Return the foot pedal into the sewing machine gap to your right, not the one in front.", + "D": "The foot pedal should be put back into the gap of the sewing machine in front of you.", + "E": "Place the foot pedal back into the gap on the sewing machine behind you instead of the front one." + }, + "answer_label": "D", + "answer_text": "The foot pedal should be put back into the gap of the sewing machine in front of you.", + "q_id": "6baa673a-0cc6-4da2-a481-877de0d2faa5_6_T2", + "correct_answer": "The foot pedal should be put back into the gap of the sewing machine in front of you.", + "original_options": { + "A": "Put it in the gap left.", + "B": "Put it in the gap behind.", + "C": "Return the foot pedal into the sewing machine gap to your right, not the one in front.", + "D": "The foot pedal should be put back into the gap of the sewing machine in front of you.", + "E": "Place the foot pedal back into the gap on the sewing machine behind you instead of the front one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "6baa673a-0cc6-4da2-a481-877de0d2faa5", + "video_length": "00:20:19", + "qid": "6baa673a-0cc6-4da2-a481-877de0d2faa5_2_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the sofa relative to me?", + "answer": "The sofa is in front of you, toward your front-right.", + "question_chinese": "沙发相对于我的方位是？", + "answer_chinese": "沙发在你的前方和右前方。", + "evidence": [ + { + "description": "After you turn around, there is a sofa in front of you and another one to your right.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:13" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sofa is in front-left of you.", + "B": "The sofa is directly behind you.", + "C": "The sofa sits behind you toward your back-right, rather than being ahead on your right side.", + "D": "The sofa is in front of you, toward your front-right.", + "E": "The sofa is positioned in front of you but closer to your front-left, not toward your front-right." + }, + "answer_label": "D", + "answer_text": "The sofa is in front of you, toward your front-right.", + "q_id": "6baa673a-0cc6-4da2-a481-877de0d2faa5_2_T1", + "correct_answer": "The sofa is in front of you, toward your front-right.", + "original_options": { + "A": "The sofa is in front-left of you.", + "B": "The sofa is directly behind you.", + "C": "The sofa sits behind you toward your back-right, rather than being ahead on your right side.", + "D": "The sofa is in front of you, toward your front-right.", + "E": "The sofa is positioned in front of you but closer to your front-left, not toward your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 17 + ] + }, + { + "video_uid": "6baa673a-0cc6-4da2-a481-877de0d2faa5", + "video_length": "00:20:19", + "qid": "6baa673a-0cc6-4da2-a481-877de0d2faa5_2_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the sofa relative to me?", + "answer": "The sofa is behind you?", + "question_chinese": "沙发相对于我的方位是？", + "answer_chinese": "沙发在你的后面？", + "evidence": [ + { + "description": "After you turn around, there is a sofa in front of you and another one to your right.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:13" + } + } + ], + "question_timestamps": "00:11:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sofa is in front of you.", + "B": "The sofa is to your left.", + "C": "The sofa is in front of you, slightly toward your front-left side.", + "D": "The sofa is behind you, but closer to your back-right rather than directly behind.", + "E": "The sofa is behind you?" + }, + "answer_label": "E", + "answer_text": "The sofa is behind you?", + "q_id": "6baa673a-0cc6-4da2-a481-877de0d2faa5_2_T2", + "correct_answer": "The sofa is behind you?", + "original_options": { + "A": "The sofa is in front of you.", + "B": "The sofa is to your left.", + "C": "The sofa is in front of you, slightly toward your front-left side.", + "D": "The sofa is behind you, but closer to your back-right rather than directly behind.", + "E": "The sofa is behind you?" + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 14 + ] + } +] diff --git a/QAs/6bc12612-2c12-4938-a394-c65de370a159-mcq.json b/QAs/6bc12612-2c12-4938-a394-c65de370a159-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..22d7569bde6b6b69b8b1401471e6af0bc50aed80 --- /dev/null +++ b/QAs/6bc12612-2c12-4938-a394-c65de370a159-mcq.json @@ -0,0 +1,850 @@ +[ + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_14", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tables are there in the scene?", + "answer": "There are two chairs in the scene.", + "question_chinese": "场景中有多少个桌子？", + "answer_chinese": "场景中有两把椅子。", + "evidence": [ + { + "description": "There are two tables in the scene.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two chairs in the scene.", + "B": "There is one chair in the scene.", + "C": "There are three chairs in the scene.", + "D": "There are four chairs in the scene, arranged as two pairs rather than just a couple of seats.", + "E": "There are five chairs in the scene, so you can count more than two seating spots in total." + }, + "answer_label": "A", + "answer_text": "There are two chairs in the scene.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_14", + "correct_answer": "There are two chairs in the scene.", + "original_options": { + "A": "There are two chairs in the scene.", + "B": "There is one chair in the scene.", + "C": "There are three chairs in the scene.", + "D": "There are four chairs in the scene, arranged as two pairs rather than just a couple of seats.", + "E": "There are five chairs in the scene, so you can count more than two seating spots in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_15", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tables are there in the scene?", + "answer": "There are three tables in the scene.", + "question_chinese": "场景中有多少个桌子？", + "answer_chinese": "场景中有三个桌子。", + "evidence": [ + { + "description": "There are two tables in the scene.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "There is also a white coffee table in the scene.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:28:29", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There is one table in the scene.", + "B": "There are three tables in the scene.", + "C": "There are four tables in the scene.", + "D": "There are five tables in the scene, meaning the total count is higher than three.", + "E": "There are six tables in the scene, so you should be able to count six separate table surfaces." + }, + "answer_label": "B", + "answer_text": "There are three tables in the scene.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_15", + "correct_answer": "There are three tables in the scene.", + "original_options": { + "A": "There is one table in the scene.", + "B": "There are three tables in the scene.", + "C": "There are four tables in the scene.", + "D": "There are five tables in the scene, meaning the total count is higher than three.", + "E": "There are six tables in the scene, so you should be able to count six separate table surfaces." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 18 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_10", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get some water, where should I go?", + "answer": "You should turn around and walk two steps straight to the yellow wooden table.", + "question_chinese": "如果我想去喝水，我应该走去哪里呢？", + "answer_chinese": "你应该向后转身，然后直走两部到黄色木桌前。", + "evidence": [ + { + "description": "There is a glass of water on the table.", + "timestamps": { + "start": "00:09:56", + "end": "00:09:58" + } + } + ], + "question_timestamps": "00:14:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn around and walk three steps straight.", + "B": "Turn around and go two steps right.", + "C": "Turn around, then walk one step straight to the yellow wooden table instead of taking two steps.", + "D": "Turn around and walk two steps straight, but stop at the table to your right rather than the yellow wooden one.", + "E": "You should turn around and walk two steps straight to the yellow wooden table." + }, + "answer_label": "E", + "answer_text": "You should turn around and walk two steps straight to the yellow wooden table.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_10", + "correct_answer": "You should turn around and walk two steps straight to the yellow wooden table.", + "original_options": { + "A": "Turn around and walk three steps straight.", + "B": "Turn around and go two steps right.", + "C": "Turn around, then walk one step straight to the yellow wooden table instead of taking two steps.", + "D": "Turn around and walk two steps straight, but stop at the table to your right rather than the yellow wooden one.", + "E": "You should turn around and walk two steps straight to the yellow wooden table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 21 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_11", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get some water, where should I go?", + "answer": "You should turn right 30 degrees and then walk two steps straight to the yellow wooden table.", + "question_chinese": "如果我想去喝水，我应该走去哪里呢？", + "answer_chinese": "你应该向右转30度，然后直走两步到黄色木桌前。", + "evidence": [ + { + "description": "There is a glass of water on the table.", + "timestamps": { + "start": "00:09:56", + "end": "00:09:58" + } + } + ], + "question_timestamps": "00:38:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left 30 degrees, then walk two steps.", + "B": "You should turn right 30 degrees and then walk two steps straight to the yellow wooden table.", + "C": "Turn right 90 degrees, then walk one step.", + "D": "Turn right 60 degrees and then walk two steps straight to the yellow wooden table instead of turning only 30 degrees.", + "E": "Turn right 30 degrees, then walk three steps straight to the yellow wooden table rather than walking only two steps." + }, + "answer_label": "B", + "answer_text": "You should turn right 30 degrees and then walk two steps straight to the yellow wooden table.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_11", + "correct_answer": "You should turn right 30 degrees and then walk two steps straight to the yellow wooden table.", + "original_options": { + "A": "Turn left 30 degrees, then walk two steps.", + "B": "You should turn right 30 degrees and then walk two steps straight to the yellow wooden table.", + "C": "Turn right 90 degrees, then walk one step.", + "D": "Turn right 60 degrees and then walk two steps straight to the yellow wooden table instead of turning only 30 degrees.", + "E": "Turn right 30 degrees, then walk three steps straight to the yellow wooden table rather than walking only two steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 21, + 20 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_5", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I drunk water?", + "answer": "You drank water five times.", + "question_chinese": "我喝水喝了多少次呢？", + "answer_chinese": "你喝了五次水。", + "evidence": [ + { + "description": "You drank the first glass of water.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:55" + } + }, + { + "description": "You drank the second glass of water.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:08" + } + }, + { + "description": "You drank the third glass of water.", + "timestamps": { + "start": "00:13:20", + "end": "00:13:23" + } + }, + { + "description": "You drank the fourth glass of water.", + "timestamps": { + "start": "00:16:36", + "end": "00:16:42" + } + }, + { + "description": "You drank the fifth glass of water.", + "timestamps": { + "start": "00:20:51", + "end": "00:20:57" + } + } + ], + "question_timestamps": "00:21:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You drank water four times.", + "B": "You drank water six times.", + "C": "You drank water five times.", + "D": "You drank water eight times in total throughout the video, with multiple separate sips counted.", + "E": "You drank water three times overall, only counting the distinct moments you raised the bottle to drink." + }, + "answer_label": "C", + "answer_text": "You drank water five times.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_5", + "correct_answer": "You drank water five times.", + "original_options": { + "A": "You drank water four times.", + "B": "You drank water six times.", + "C": "You drank water five times.", + "D": "You drank water eight times in total throughout the video, with multiple separate sips counted.", + "E": "You drank water three times overall, only counting the distinct moments you raised the bottle to drink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_6", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I drunk water?", + "answer": "You drank water six times.", + "question_chinese": "我喝水喝了多少次呢？", + "answer_chinese": "你喝了六次水。", + "evidence": [ + { + "description": "You drank the first glass of water.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:55" + } + }, + { + "description": "You drank the second glass of water.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:08" + } + }, + { + "description": "You drank the third glass of water.", + "timestamps": { + "start": "00:13:20", + "end": "00:13:23" + } + }, + { + "description": "You drank the fourth glass of water.", + "timestamps": { + "start": "00:16:36", + "end": "00:16:42" + } + }, + { + "description": "You drank the fifth glass of water.", + "timestamps": { + "start": "00:20:51", + "end": "00:20:57" + } + }, + { + "description": "You drank water for the sixth time.", + "timestamps": { + "start": "00:27:19", + "end": "00:27:30" + } + } + ], + "question_timestamps": "00:27:35", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You drank water five times.", + "B": "You drank water seven times.", + "C": "You drank water eight times altogether, counting each separate instance you took a drink.", + "D": "You drank water six times.", + "E": "You drank water four times total, with only four distinct drinking moments across the scene." + }, + "answer_label": "D", + "answer_text": "You drank water six times.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_6", + "correct_answer": "You drank water six times.", + "original_options": { + "A": "You drank water five times.", + "B": "You drank water seven times.", + "C": "You drank water eight times altogether, counting each separate instance you took a drink.", + "D": "You drank water six times.", + "E": "You drank water four times total, with only four distinct drinking moments across the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_7", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I drunk water?", + "answer": "You drank water seven times.", + "question_chinese": "我喝水喝了多少次呢？", + "answer_chinese": "你喝了七次水。", + "evidence": [ + { + "description": "You drank the first glass of water.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:55" + } + }, + { + "description": "You drank the second glass of water.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:08" + } + }, + { + "description": "You drank the third glass of water.", + "timestamps": { + "start": "00:13:20", + "end": "00:13:23" + } + }, + { + "description": "You drank the fourth glass of water.", + "timestamps": { + "start": "00:16:36", + "end": "00:16:42" + } + }, + { + "description": "You drank the fifth glass of water.", + "timestamps": { + "start": "00:20:51", + "end": "00:20:57" + } + }, + { + "description": "You drank water for the sixth time.", + "timestamps": { + "start": "00:27:19", + "end": "00:27:30" + } + }, + { + "description": "You drank water for the seventh time.", + "timestamps": { + "start": "00:30:46", + "end": "00:30:52" + } + } + ], + "question_timestamps": "00:30:55", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You drank water six times.", + "B": "You drank water eight times.", + "C": "You drank water seven times.", + "D": "You drank water five times overall, counting only the clear moments you actually took a sip.", + "E": "You drank water nine times in total, including each separate time you paused to drink." + }, + "answer_label": "C", + "answer_text": "You drank water seven times.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_7", + "correct_answer": "You drank water seven times.", + "original_options": { + "A": "You drank water six times.", + "B": "You drank water eight times.", + "C": "You drank water seven times.", + "D": "You drank water five times overall, counting only the clear moments you actually took a sip.", + "E": "You drank water nine times in total, including each separate time you paused to drink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many glasses of water have I drunk so far?", + "answer": "You drank two glasses of water.", + "question_chinese": "我目前为止喝了多少杯水呢？", + "answer_chinese": "你喝了两杯水", + "evidence": [ + { + "description": "You drank the first glass of water.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:55" + } + }, + { + "description": "You drank the second glass of water.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:08" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You drank one glass of water.", + "B": "You drank three glasses of water.", + "C": "You drank four glasses of water so far, counting each time you finished a full glass.", + "D": "You drank two glasses of water.", + "E": "You drank five glasses of water in total up to now, including every separate glass you picked up to drink." + }, + "answer_label": "D", + "answer_text": "You drank two glasses of water.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_2", + "correct_answer": "You drank two glasses of water.", + "original_options": { + "A": "You drank one glass of water.", + "B": "You drank three glasses of water.", + "C": "You drank four glasses of water so far, counting each time you finished a full glass.", + "D": "You drank two glasses of water.", + "E": "You drank five glasses of water in total up to now, including every separate glass you picked up to drink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 20 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_3", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many glasses of water have I drunk so far?", + "answer": "You drank three glasses of water.", + "question_chinese": "我目前为止喝了多少杯水呢？", + "answer_chinese": "你喝了三杯水。", + "evidence": [ + { + "description": "You drank the first glass of water.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:55" + } + }, + { + "description": "You drank the second glass of water.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:08" + } + }, + { + "description": "You drank the third glass of water.", + "timestamps": { + "start": "00:13:20", + "end": "00:13:23" + } + } + ], + "question_timestamps": "00:13:25", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You drank two glasses of water.", + "B": "You drank three glasses of water.", + "C": "You drank four glasses of water.", + "D": "You drank one glass of water so far, with only a single completed glass counted.", + "E": "You drank five glasses of water up to this point, counting each distinct glass you drank from." + }, + "answer_label": "B", + "answer_text": "You drank three glasses of water.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_3", + "correct_answer": "You drank three glasses of water.", + "original_options": { + "A": "You drank two glasses of water.", + "B": "You drank three glasses of water.", + "C": "You drank four glasses of water.", + "D": "You drank one glass of water so far, with only a single completed glass counted.", + "E": "You drank five glasses of water up to this point, counting each distinct glass you drank from." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_4", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many glasses of water have I drunk so far?", + "answer": "You drank four glasses of water.", + "question_chinese": "我目前为止喝了多少杯水呢？", + "answer_chinese": "你喝了第四杯水。", + "evidence": [ + { + "description": "You drank the first glass of water.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:55" + } + }, + { + "description": "You drank the second glass of water.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:08" + } + }, + { + "description": "You drank the third glass of water.", + "timestamps": { + "start": "00:13:20", + "end": "00:13:23" + } + }, + { + "description": "You drank the fourth glass of water.", + "timestamps": { + "start": "00:16:36", + "end": "00:16:42" + } + } + ], + "question_timestamps": "00:16:46", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You drank three glasses of water.", + "B": "You drank five glasses of water.", + "C": "You drank six glasses of water overall, counting each time you went back for another glass.", + "D": "You drank two glasses of water so far, with only two distinct glasses consumed during the video.", + "E": "You drank four glasses of water." + }, + "answer_label": "E", + "answer_text": "You drank four glasses of water.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_4", + "correct_answer": "You drank four glasses of water.", + "original_options": { + "A": "You drank three glasses of water.", + "B": "You drank five glasses of water.", + "C": "You drank six glasses of water overall, counting each time you went back for another glass.", + "D": "You drank two glasses of water so far, with only two distinct glasses consumed during the video.", + "E": "You drank four glasses of water." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_13", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the white table or the yellow table?", + "answer": "You are closer to the yellow table.", + "question_chinese": "我离白色的桌子更近还是离黄色的桌子更近呢？", + "answer_chinese": "你离黄色的桌子更近。", + "evidence": [ + { + "description": "You are four steps away from the white table.", + "timestamps": { + "start": "00:38:22", + "end": "00:38:30" + } + }, + { + "description": "You are two steps away from the yellow table.", + "timestamps": { + "start": "00:38:30", + "end": "00:38:34" + } + } + ], + "question_timestamps": "00:38:36", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You are closer to the white table.", + "B": "You are closer to the yellow table." + }, + "answer_label": "B", + "answer_text": "You are closer to the yellow table.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_13", + "correct_answer": "You are closer to the yellow table.", + "original_options": { + "A": "You are closer to the white table.", + "B": "You are closer to the yellow table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_8", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What's on the desktop where the video is playing?", + "answer": "There is a computer, a mobile phone and a cup on the desk.", + "question_chinese": "播放视频的桌面上放了什么？", + "answer_chinese": "桌面面上有一个电脑，一部手机，一个杯子。", + "evidence": [ + { + "description": "The items on the desk include a computer, a mobile phone, and a cup.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:52" + } + } + ], + "question_timestamps": "00:03:59", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There is a computer, a mobile phone and a cup on the desk.", + "B": "A computer and a cup.", + "C": "A computer and a mobile phone.", + "D": "There is a computer, a mobile phone, and a backpack on the desk.", + "E": "On the desk, you have a computer, a mobile phone, and some snacks set beside it." + }, + "answer_label": "A", + "answer_text": "There is a computer, a mobile phone and a cup on the desk.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_8", + "correct_answer": "There is a computer, a mobile phone and a cup on the desk.", + "original_options": { + "A": "There is a computer, a mobile phone and a cup on the desk.", + "B": "A computer and a cup.", + "C": "A computer and a mobile phone.", + "D": "There is a computer, a mobile phone, and a backpack on the desk.", + "E": "On the desk, you have a computer, a mobile phone, and some snacks set beside it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 16 + ] + }, + { + "video_uid": "6bc12612-2c12-4938-a394-c65de370a159", + "video_length": "00:38:37", + "qid": "6bc12612-2c12-4938-a394-c65de370a159_9", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What's on the desktop where the video is playing?", + "answer": "The items on the desk include a computer, a mobile phone, a cup, a note and a vase of flowers.", + "question_chinese": "播放视频的桌面上放了什么？", + "answer_chinese": "桌面上的物品有一个电脑，一部手机，一个杯子，一张便签和一瓶鲜花。", + "evidence": [ + { + "description": "There is also a note and a vase of flowers on the desk.", + "timestamps": { + "start": "00:27:18", + "end": "00:27:26" + } + }, + { + "description": "The items on the desk include a computer, a mobile phone, and a cup.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:52" + } + } + ], + "question_timestamps": "00:27:37", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The items on the desk include a computer, a mobile phone, a cup, a note and a vase of flowers.", + "B": "A computer and a cup.", + "C": "A mobile phone and a note.", + "D": "The desk has a computer, a mobile phone, a cup, and a note, but no flowers.", + "E": "The items include a computer, a cup, a note, and a vase of flowers, without any phone." + }, + "answer_label": "A", + "answer_text": "The items on the desk include a computer, a mobile phone, a cup, a note and a vase of flowers.", + "q_id": "6bc12612-2c12-4938-a394-c65de370a159_9", + "correct_answer": "The items on the desk include a computer, a mobile phone, a cup, a note and a vase of flowers.", + "original_options": { + "A": "The items on the desk include a computer, a mobile phone, a cup, a note and a vase of flowers.", + "B": "A computer and a cup.", + "C": "A mobile phone and a note.", + "D": "The desk has a computer, a mobile phone, a cup, and a note, but no flowers.", + "E": "The items include a computer, a cup, a note, and a vase of flowers, without any phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + } +] diff --git a/QAs/6c1123bb-1aab-46e9-a69f-54e459e4bf50-mcq.json b/QAs/6c1123bb-1aab-46e9-a69f-54e459e4bf50-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3a571191849b158de9c329808ff3f3032e3bbd2f --- /dev/null +++ b/QAs/6c1123bb-1aab-46e9-a69f-54e459e4bf50-mcq.json @@ -0,0 +1,983 @@ +[ + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_5_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I peel?", + "answer": "4.", + "question_chinese": "我削了多少个土豆？", + "answer_chinese": "4", + "evidence": [ + { + "description": "the first one and the second one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "the fourth one", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "3.", + "B": "5.", + "C": "4.", + "D": "You peeled only 2 potatoes during the prep step, then stopped peeling.", + "E": "You peeled 6 potatoes altogether before moving on to cooking the next ingredients." + }, + "answer_label": "C", + "answer_text": "4.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_5_T2", + "correct_answer": "4.", + "original_options": { + "A": "3.", + "B": "5.", + "C": "4.", + "D": "You peeled only 2 potatoes during the prep step, then stopped peeling.", + "E": "You peeled 6 potatoes altogether before moving on to cooking the next ingredients." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_5_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I peel?", + "answer": "Four, all placed to your left.", + "question_chinese": "我削了多少个土豆？", + "answer_chinese": "4，都放在我的左边。", + "evidence": [ + { + "description": "the first one and the second one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "the fourth one", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three, to your left.", + "B": "Four, all placed to your left.", + "C": "Five, to your left.", + "D": "Two potatoes, all placed to your left after you finished peeling them completely.", + "E": "Six potatoes, all placed to your left once you had peeled every single one." + }, + "answer_label": "B", + "answer_text": "Four, all placed to your left.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_5_T1", + "correct_answer": "Four, all placed to your left.", + "original_options": { + "A": "Three, to your left.", + "B": "Four, all placed to your left.", + "C": "Five, to your left.", + "D": "Two potatoes, all placed to your left after you finished peeling them completely.", + "E": "Six potatoes, all placed to your left once you had peeled every single one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_9_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots can I cook at the same time?", + "answer": "Two.", + "question_chinese": "我同时料理多少口锅？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:54", + "end": "00:01:55" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:41:33", + "end": "00:41:34" + } + } + ], + "question_timestamps": "00:41:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You can cook 4 pots at the same time on that stove setup.", + "E": "You can cook 5 pots simultaneously on the stove without overcrowding any burner." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_9_T1", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You can cook 4 pots at the same time on that stove setup.", + "E": "You can cook 5 pots simultaneously on the stove without overcrowding any burner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_9_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots can I cook at the same time?", + "answer": "2.", + "question_chinese": "我同时料理多少口锅？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:54", + "end": "00:01:55" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:41:33", + "end": "00:41:34" + } + } + ], + "question_timestamps": "00:41:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "1.", + "B": "3.", + "C": "2.", + "D": "You can cook 4 pots at the same time on the stove as shown.", + "E": "You can cook 5 pots at the same time using all available burners." + }, + "answer_label": "C", + "answer_text": "2.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_9_T2", + "correct_answer": "2.", + "original_options": { + "A": "1.", + "B": "3.", + "C": "2.", + "D": "You can cook 4 pots at the same time on the stove as shown.", + "E": "You can cook 5 pots at the same time using all available burners." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_2_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I cutting?", + "answer": "You’re chopping onions on the floor to your right.", + "question_chinese": "我在切什么？", + "answer_chinese": "你在你右边的地上切洋葱。", + "evidence": [ + { + "description": "I am cutting onions", + "timestamps": { + "start": "00:07:17", + "end": "00:07:18" + } + } + ], + "question_timestamps": "00:07:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’re cutting potatoes.", + "B": "You’re chopping onions to your left.", + "C": "You’re chopping onions on the counter in front of you, not on the floor to your right.", + "D": "You’re chopping onions on the floor to your right.", + "E": "You’re chopping onions on the floor behind you rather than on the floor to your right." + }, + "answer_label": "D", + "answer_text": "You’re chopping onions on the floor to your right.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_2_T1", + "correct_answer": "You’re chopping onions on the floor to your right.", + "original_options": { + "A": "You’re cutting potatoes.", + "B": "You’re chopping onions to your left.", + "C": "You’re chopping onions on the counter in front of you, not on the floor to your right.", + "D": "You’re chopping onions on the floor to your right.", + "E": "You’re chopping onions on the floor behind you rather than on the floor to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 18, + 17 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_2_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I cutting?", + "answer": "The onion behind you.", + "question_chinese": "我在切什么？", + "answer_chinese": "你身后的洋葱", + "evidence": [ + { + "description": "I am cutting onions", + "timestamps": { + "start": "00:07:17", + "end": "00:07:18" + } + } + ], + "question_timestamps": "00:07:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The onion in front of you.", + "B": "The onion to your left.", + "C": "The onion on the cutting board in front of you, not the one positioned behind you.", + "D": "The onion to your right side, the one you reached for while facing the counter.", + "E": "The onion behind you." + }, + "answer_label": "E", + "answer_text": "The onion behind you.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_2_T2", + "correct_answer": "The onion behind you.", + "original_options": { + "A": "The onion in front of you.", + "B": "The onion to your left.", + "C": "The onion on the cutting board in front of you, not the one positioned behind you.", + "D": "The onion to your right side, the one you reached for while facing the counter.", + "E": "The onion behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 15 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_7_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I handle the potatoes?", + "answer": "Peel it first on the cutting board behind you, then cut it into chunks.", + "question_chinese": "我是怎么处理土豆的？", + "answer_chinese": "在你身后的菜板上先削皮，然后切块", + "evidence": [ + { + "description": "Peel", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "cut into pieces", + "timestamps": { + "start": "00:06:16", + "end": "00:06:17" + } + } + ], + "question_timestamps": "00:06:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Cut it into chunks first.", + "B": "Slice it thin instead.", + "C": "Peel it on the cutting board in front of you, then cut it into chunks afterward.", + "D": "Peel it first on the cutting board to your left, then cut it into chunks.", + "E": "Peel it first on the cutting board behind you, then cut it into chunks." + }, + "answer_label": "E", + "answer_text": "Peel it first on the cutting board behind you, then cut it into chunks.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_7_T2", + "correct_answer": "Peel it first on the cutting board behind you, then cut it into chunks.", + "original_options": { + "A": "Cut it into chunks first.", + "B": "Slice it thin instead.", + "C": "Peel it on the cutting board in front of you, then cut it into chunks afterward.", + "D": "Peel it first on the cutting board to your left, then cut it into chunks.", + "E": "Peel it first on the cutting board behind you, then cut it into chunks." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 15 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_7_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I handle the potatoes?", + "answer": "Peel it first, then cut it into chunks. Right now, all the potatoes have been placed behind you.", + "question_chinese": "我是怎么处理土豆的？", + "answer_chinese": "先削皮，然后切块.目前土豆都放置到身后了。", + "evidence": [ + { + "description": "Peel", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "cut into pieces", + "timestamps": { + "start": "00:06:16", + "end": "00:06:17" + } + } + ], + "question_timestamps": "00:06:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They are in front of you.", + "B": "They are to your left.", + "C": "Peel it first on the cutting board behind you, then cut it into chunks.", + "D": "Peel it first, then cut it into chunks. Right now, all the potatoes have been placed behind you.", + "E": "Peel it first, then cut it into chunks; the potatoes are currently placed to your right." + }, + "answer_label": "D", + "answer_text": "Peel it first, then cut it into chunks. Right now, all the potatoes have been placed behind you.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_7_T1", + "correct_answer": "Peel it first, then cut it into chunks. Right now, all the potatoes have been placed behind you.", + "original_options": { + "A": "They are in front of you.", + "B": "They are to your left.", + "C": "Peel it first on the cutting board behind you, then cut it into chunks.", + "D": "Peel it first, then cut it into chunks. Right now, all the potatoes have been placed behind you.", + "E": "Peel it first, then cut it into chunks; the potatoes are currently placed to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 16 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_8_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I grind the spices?", + "answer": "Put the spices on the grinding board behind you, add water, then grind them.", + "question_chinese": "我是怎么磨香料的？", + "answer_chinese": "把香料放在你身后的磨板上，加水，然后磨。", + "evidence": [ + { + "description": "Place the spices on the grinding board", + "timestamps": { + "start": "00:08:19", + "end": "00:08:20" + } + }, + { + "description": "add water", + "timestamps": { + "start": "00:08:38", + "end": "00:08:39" + } + }, + { + "description": "grind", + "timestamps": { + "start": "00:09:14", + "end": "00:09:15" + } + } + ], + "question_timestamps": "00:10:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put the spices on the grinding board behind you, add water, then grind them.", + "B": "Grind them without water.", + "C": "Use the board in front.", + "D": "Put the spices on the grinding board to your right, add water, then grind them.", + "E": "Place the spices on the grinding board behind you, then grind them without adding any water." + }, + "answer_label": "A", + "answer_text": "Put the spices on the grinding board behind you, add water, then grind them.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_8_T2", + "correct_answer": "Put the spices on the grinding board behind you, add water, then grind them.", + "original_options": { + "A": "Put the spices on the grinding board behind you, add water, then grind them.", + "B": "Grind them without water.", + "C": "Use the board in front.", + "D": "Put the spices on the grinding board to your right, add water, then grind them.", + "E": "Place the spices on the grinding board behind you, then grind them without adding any water." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 16 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_8_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I grind the spices?", + "answer": "Put the spices on the grinding board to your back-left, add water, then grind them.", + "question_chinese": "我是怎么磨香料的？", + "answer_chinese": "把香料放在你左后方的磨板上，加水，然后磨", + "evidence": [ + { + "description": "Place the spices on the grinding board", + "timestamps": { + "start": "00:08:19", + "end": "00:08:20" + } + }, + { + "description": "add water", + "timestamps": { + "start": "00:08:38", + "end": "00:08:39" + } + }, + { + "description": "grind", + "timestamps": { + "start": "00:09:14", + "end": "00:9:15" + } + } + ], + "question_timestamps": "00:10:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use the back-right board.", + "B": "Use the front-left board.", + "C": "Put the spices on the grinding board behind you, add water, then grind them.", + "D": "Put the spices on the grinding board to your back-left, then grind them without adding water.", + "E": "Put the spices on the grinding board to your back-left, add water, then grind them." + }, + "answer_label": "E", + "answer_text": "Put the spices on the grinding board to your back-left, add water, then grind them.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_8_T1", + "correct_answer": "Put the spices on the grinding board to your back-left, add water, then grind them.", + "original_options": { + "A": "Use the back-right board.", + "B": "Use the front-left board.", + "C": "Put the spices on the grinding board behind you, add water, then grind them.", + "D": "Put the spices on the grinding board to your back-left, then grind them without adding water.", + "E": "Put the spices on the grinding board to your back-left, add water, then grind them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 16 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_4_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the rolling pin?", + "answer": "On the bowl in front of you.", + "question_chinese": "我看到擀面杖在哪里？", + "answer_chinese": "在我前方的盆上。", + "evidence": [ + { + "description": "On the basin", + "timestamps": { + "start": "01:26:43", + "end": "01:26:44" + } + } + ], + "question_timestamps": "01:26:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the bowl in front of you.", + "B": "On the bowl behind you.", + "C": "On the bowl to your left.", + "D": "On the bowl to your right, resting across the top rather than in front of you.", + "E": "On the edge of the table in front of you, not sitting on top of any bowl." + }, + "answer_label": "A", + "answer_text": "On the bowl in front of you.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_4_T1", + "correct_answer": "On the bowl in front of you.", + "original_options": { + "A": "On the bowl in front of you.", + "B": "On the bowl behind you.", + "C": "On the bowl to your left.", + "D": "On the bowl to your right, resting across the top rather than in front of you.", + "E": "On the edge of the table in front of you, not sitting on top of any bowl." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_4_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the rolling pin?", + "answer": "On the bowl to your left.", + "question_chinese": "我看到擀面杖在哪里？", + "answer_chinese": "我左边的盆上", + "evidence": [ + { + "description": "On the basin", + "timestamps": { + "start": "01:26:43", + "end": "01:26:44" + } + } + ], + "question_timestamps": "01:28:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the bowl to your right.", + "B": "On the bowl to your left.", + "C": "On the table edge.", + "D": "On the bowl directly in front of you, not the one positioned to your left side.", + "E": "On the inner part of the table surface, away from the bowl on your left." + }, + "answer_label": "B", + "answer_text": "On the bowl to your left.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_4_T2", + "correct_answer": "On the bowl to your left.", + "original_options": { + "A": "On the bowl to your right.", + "B": "On the bowl to your left.", + "C": "On the table edge.", + "D": "On the bowl directly in front of you, not the one positioned to your left side.", + "E": "On the inner part of the table surface, away from the bowl on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 16, + 15 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_6_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I see dough wrapped around a filling?", + "answer": "Some do and some don’t; all the dough pieces are placed to your right.", + "question_chinese": "我看到面团包着馅料吗？", + "answer_chinese": "有的有，有的没有，面团都放置在我的右方。", + "evidence": [ + { + "description": "filled with", + "timestamps": { + "start": "01:27:13", + "end": "01:27:14" + } + }, + { + "description": "not filled with", + "timestamps": { + "start": "01:28:08", + "end": "01:28:09" + } + } + ], + "question_timestamps": "01:28:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Some do and some don’t; all the dough pieces are placed to your right.", + "B": "No; none are wrapped around a filling, and they’re all placed to your right." + }, + "answer_label": "A", + "answer_text": "Some do and some don’t; all the dough pieces are placed to your right.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_6_T1", + "correct_answer": "Some do and some don’t; all the dough pieces are placed to your right.", + "original_options": { + "A": "Some do and some don’t; all the dough pieces are placed to your right.", + "B": "No; none are wrapped around a filling, and they’re all placed to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_6_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I see dough wrapped around a filling?", + "answer": "Some do and some don’t; all the dough pieces are placed to your back-right.", + "question_chinese": "我看到面团包着馅料吗？", + "answer_chinese": "有的有，有的没有，面团都放置在我的右后方。", + "evidence": [ + { + "description": "filled with", + "timestamps": { + "start": "01:27:13", + "end": "01:27:14" + } + }, + { + "description": "not filled with", + "timestamps": { + "start": "01:28:08", + "end": "01:28:09" + } + } + ], + "question_timestamps": "01:31:42", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No; none are wrapped around a filling, and they’re all placed to your back-right.", + "B": "Some do and some don’t; all the dough pieces are placed to your back-right." + }, + "answer_label": "B", + "answer_text": "Some do and some don’t; all the dough pieces are placed to your back-right.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_6_T2", + "correct_answer": "Some do and some don’t; all the dough pieces are placed to your back-right.", + "original_options": { + "A": "No; none are wrapped around a filling, and they’re all placed to your back-right.", + "B": "Some do and some don’t; all the dough pieces are placed to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_1_T1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "The porridge in the pot is overflowing. What should I do?", + "answer": "Turn down the heat.", + "question_chinese": "锅里的粥溢出了我该怎么办？", + "answer_chinese": "把火调小", + "evidence": [ + { + "description": "Turn down the heat", + "timestamps": { + "start": "00:01:54", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:01:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Stir it gently.", + "B": "Cover the pot loosely.", + "C": "Increase the heat slightly so it cooks faster, then let it settle back down before it overflows again.", + "D": "Turn down the heat.", + "E": "Move the pot to a cooler surface and keep it there until the bubbling calms and the spill stops." + }, + "answer_label": "D", + "answer_text": "Turn down the heat.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_1_T1", + "correct_answer": "Turn down the heat.", + "original_options": { + "A": "Stir it gently.", + "B": "Cover the pot loosely.", + "C": "Increase the heat slightly so it cooks faster, then let it settle back down before it overflows again.", + "D": "Turn down the heat.", + "E": "Move the pot to a cooler surface and keep it there until the bubbling calms and the spill stops." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 18, + 19 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_1_T3", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "The porridge in the pot is boiling over—what should I do?", + "answer": "Turn down the flame on the right.", + "question_chinese": "锅里的粥溢出了我该怎么办？", + "answer_chinese": "把右边的火调小", + "evidence": [ + { + "description": "Turn down the heat", + "timestamps": { + "start": "00:01:54", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:02:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn down the left flame.", + "B": "Turn off the right flame.", + "C": "Turn down the flame in front of you instead of the one on the right that’s causing the boil-over.", + "D": "Lower the flame on the stove by your left hand, not the right-side burner under the pot.", + "E": "Turn down the flame on the right." + }, + "answer_label": "E", + "answer_text": "Turn down the flame on the right.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_1_T3", + "correct_answer": "Turn down the flame on the right.", + "original_options": { + "A": "Turn down the left flame.", + "B": "Turn off the right flame.", + "C": "Turn down the flame in front of you instead of the one on the right that’s causing the boil-over.", + "D": "Lower the flame on the stove by your left hand, not the right-side burner under the pot.", + "E": "Turn down the flame on the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 17 + ] + }, + { + "video_uid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50", + "video_length": "01:34:55", + "qid": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_1_T2", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "The porridge in the pot is boiling over—what should I do?", + "answer": "Turn down the flame on the stove by your right hand.", + "question_chinese": "锅里的粥溢出了我该怎么办？", + "answer_chinese": "把右手边的炉子的火调小", + "evidence": [ + { + "description": "Turn down the heat", + "timestamps": { + "start": "00:01:54", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:02:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn down the left flame.", + "B": "Turn off the right flame.", + "C": "Turn down the flame in front of you rather than the burner by your right hand that the pot is on.", + "D": "Turn down the flame on the stove by your right hand.", + "E": "Turn down the flame on the stove by your left hand, since that’s the burner you should adjust to stop it." + }, + "answer_label": "D", + "answer_text": "Turn down the flame on the stove by your right hand.", + "q_id": "6c1123bb-1aab-46e9-a69f-54e459e4bf50_1_T2", + "correct_answer": "Turn down the flame on the stove by your right hand.", + "original_options": { + "A": "Turn down the left flame.", + "B": "Turn off the right flame.", + "C": "Turn down the flame in front of you rather than the burner by your right hand that the pot is on.", + "D": "Turn down the flame on the stove by your right hand.", + "E": "Turn down the flame on the stove by your left hand, since that’s the burner you should adjust to stop it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 22 + ] + } +] diff --git a/QAs/6c3e3490-9f33-4fec-9ce1-042e4d409a70-mcq.json b/QAs/6c3e3490-9f33-4fec-9ce1-042e4d409a70-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..0512774ce482a33bf20b0c7be9339de090f21012 --- /dev/null +++ b/QAs/6c3e3490-9f33-4fec-9ce1-042e4d409a70-mcq.json @@ -0,0 +1,1092 @@ +[ + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_18", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the blue round ruler just now?", + "answer": "Starting from the pottery bowl, walk straight along the table for three steps, turn left around the table, and take the ruler down from the cabinet.", + "question_chinese": "刚刚蓝色圆尺在运动时的运动轨迹是？", + "answer_chinese": "从陶碗出发，沿着桌子直走三步，向左转绕过桌子，从柜子上拿下尺子", + "evidence": [ + { + "description": "Starting from the pottery bowl, walk straight along the table for three steps, turn left around the table, and take the ruler down from the cabinet.", + "timestamps": { + "start": "00:42:51", + "end": "00:42:58" + } + } + ], + "question_timestamps": "00:44:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right around the table.", + "B": "Starting from the pottery bowl, walk straight along the table for three steps, turn left around the table, and take the ruler down from the cabinet.", + "C": "Walk two steps, then stop.", + "D": "From the pottery bowl, walk three steps, turn right around the table, and take the ruler from the cabinet.", + "E": "Starting at the pottery bowl, go straight, turn left around the table, then place the ruler onto the tabletop." + }, + "answer_label": "B", + "answer_text": "Starting from the pottery bowl, walk straight along the table for three steps, turn left around the table, and take the ruler down from the cabinet.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_18", + "correct_answer": "Starting from the pottery bowl, walk straight along the table for three steps, turn left around the table, and take the ruler down from the cabinet.", + "original_options": { + "A": "Turn right around the table.", + "B": "Starting from the pottery bowl, walk straight along the table for three steps, turn left around the table, and take the ruler down from the cabinet.", + "C": "Walk two steps, then stop.", + "D": "From the pottery bowl, walk three steps, turn right around the table, and take the ruler from the cabinet.", + "E": "Starting at the pottery bowl, go straight, turn left around the table, then place the ruler onto the tabletop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_19", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the blue round ruler just now?", + "answer": "The ruler is taken out of the pottery bowl and placed on the table.", + "question_chinese": "刚刚蓝色圆尺在运动时的运动轨迹是？", + "answer_chinese": "尺子被从陶碗中拿出来，放到桌上。", + "evidence": [ + { + "description": "Starting from the pottery bowl, walk straight along the table for three steps, turn left around the table, and take the ruler down from the cabinet.", + "timestamps": { + "start": "00:42:51", + "end": "00:42:58" + } + }, + { + "description": "0Take the ruler out of the pottery bowl and put it on the table.", + "timestamps": { + "start": "00:45:07", + "end": "00:45:13" + } + } + ], + "question_timestamps": "00:48:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Put it into the pottery bowl.", + "B": "The ruler is taken out of the pottery bowl and placed on the table.", + "C": "Took it off the shelf.", + "D": "The ruler is taken from the table and placed back inside the pottery bowl instead of being set down on the table.", + "E": "The ruler is removed from the shelf area and put into the pottery bowl, rather than being placed on the table." + }, + "answer_label": "B", + "answer_text": "The ruler is taken out of the pottery bowl and placed on the table.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_19", + "correct_answer": "The ruler is taken out of the pottery bowl and placed on the table.", + "original_options": { + "A": "Put it into the pottery bowl.", + "B": "The ruler is taken out of the pottery bowl and placed on the table.", + "C": "Took it off the shelf.", + "D": "The ruler is taken from the table and placed back inside the pottery bowl instead of being set down on the table.", + "E": "The ruler is removed from the shelf area and put into the pottery bowl, rather than being placed on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 22, + 21 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_20", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the blue round ruler just now?", + "answer": "The ruler is picked up from the table and put into the pottery bowl.", + "question_chinese": "刚刚蓝色圆尺在运动时的运动轨迹是？", + "answer_chinese": "尺子被从桌上拿起来，放入陶碗中。", + "evidence": [ + { + "description": "Starting from the pottery bowl, walk straight along the table for three steps, turn left around the table, and take the ruler down from the cabinet.", + "timestamps": { + "start": "00:42:51", + "end": "00:42:58" + } + }, + { + "description": "0Take the ruler out of the pottery bowl and put it on the table.", + "timestamps": { + "start": "00:45:07", + "end": "00:45:13" + } + }, + { + "description": "Put the ruler back into the pottery bowl from the table.", + "timestamps": { + "start": "00:48:04", + "end": "00:48:19" + } + } + ], + "question_timestamps": "00:48:47", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The ruler is picked up from the table and put into the pottery bowl.", + "B": "You moved it from the bowl to table.", + "C": "You slid it into the drawer nearby.", + "D": "You picked the ruler up from the bowl and set it back down on the table beside it.", + "E": "You took the ruler off the table and placed it on the desk, leaving the pottery bowl empty." + }, + "answer_label": "A", + "answer_text": "The ruler is picked up from the table and put into the pottery bowl.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_20", + "correct_answer": "The ruler is picked up from the table and put into the pottery bowl.", + "original_options": { + "A": "The ruler is picked up from the table and put into the pottery bowl.", + "B": "You moved it from the bowl to table.", + "C": "You slid it into the drawer nearby.", + "D": "You picked the ruler up from the bowl and set it back down on the table beside it.", + "E": "You took the ruler off the table and placed it on the desk, leaving the pottery bowl empty." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 18 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_11", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to walk from here to the blue table, how should I walk?", + "answer": "Walk two steps to the right and you're there.", + "question_chinese": "如果我要从这里走到蓝色桌子处，我该怎么走？", + "answer_chinese": "向右走两步就到了。", + "evidence": [ + { + "description": "Walk from the tripod stand to the blue table.", + "timestamps": { + "start": "00:07:52", + "end": "00:07:57" + } + } + ], + "question_timestamps": "00:08:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk two steps to the left.", + "B": "Walk two steps to the right and you're there.", + "C": "Walk one step to the right.", + "D": "Walk three steps to the right and stop when you reach the blue table.", + "E": "Walk two steps forward instead of right, then turn toward the blue table." + }, + "answer_label": "B", + "answer_text": "Walk two steps to the right and you're there.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_11", + "correct_answer": "Walk two steps to the right and you're there.", + "original_options": { + "A": "Walk two steps to the left.", + "B": "Walk two steps to the right and you're there.", + "C": "Walk one step to the right.", + "D": "Walk three steps to the right and stop when you reach the blue table.", + "E": "Walk two steps forward instead of right, then turn toward the blue table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_12", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to walk from here to the blue table, how should I walk?", + "answer": "Walk two steps behind you.", + "question_chinese": "如果我要从这里走到蓝色桌子处，我该怎么走？", + "answer_chinese": "向你的后面走两步。", + "evidence": [ + { + "description": "Walk from the tripod stand to the blue table.", + "timestamps": { + "start": "00:07:52", + "end": "00:07:57" + } + } + ], + "question_timestamps": "00:32:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk two steps forward.", + "B": "Walk one step behind you.", + "C": "Walk two steps to your left, then stop once you line up with the blue table.", + "D": "Walk two steps in front of you toward the table, rather than moving behind you.", + "E": "Walk two steps behind you." + }, + "answer_label": "E", + "answer_text": "Walk two steps behind you.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_12", + "correct_answer": "Walk two steps behind you.", + "original_options": { + "A": "Walk two steps forward.", + "B": "Walk one step behind you.", + "C": "Walk two steps to your left, then stop once you line up with the blue table.", + "D": "Walk two steps in front of you toward the table, rather than moving behind you.", + "E": "Walk two steps behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 15 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_13", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to walk from here to the blue table, how should I walk?", + "answer": "Walk two steps to the right, then two steps forward, and you're there.", + "question_chinese": "如果我要从这里走到蓝色桌子处，我该怎么走？", + "answer_chinese": "向右走两步，再向前走两步就到了。", + "evidence": [ + { + "description": "Walk from the tripod stand to the blue table.", + "timestamps": { + "start": "00:07:52", + "end": "00:07:57" + } + } + ], + "question_timestamps": "00:40:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk two steps left, then forward.", + "B": "Walk two steps right, then backward.", + "C": "Walk two steps to the right, then two steps forward, and you're there.", + "D": "Walk one step to the right and then three steps forward to reach the blue table.", + "E": "Walk two steps to the right, then take one step forward and one step to the left." + }, + "answer_label": "C", + "answer_text": "Walk two steps to the right, then two steps forward, and you're there.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_13", + "correct_answer": "Walk two steps to the right, then two steps forward, and you're there.", + "original_options": { + "A": "Walk two steps left, then forward.", + "B": "Walk two steps right, then backward.", + "C": "Walk two steps to the right, then two steps forward, and you're there.", + "D": "Walk one step to the right and then three steps forward to reach the blue table.", + "E": "Walk two steps to the right, then take one step forward and one step to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_7", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the portrait craft item?", + "answer": "It was placed on the blue table opposite you.", + "question_chinese": "我把人像工艺品放在哪了？", + "answer_chinese": "放在你对面的蓝色桌子上了。", + "evidence": [ + { + "description": "Put the portrait craft item from the tripod stand onto the blue table.", + "timestamps": { + "start": "00:07:52", + "end": "00:08:00" + } + } + ], + "question_timestamps": "00:17:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the wooden shelf beside you.", + "B": "It was placed on the blue table opposite you.", + "C": "In the drawer below you.", + "D": "It was set on the green table behind you rather than on the blue table opposite you.", + "E": "You placed it on the round table in front of you, not on the blue table across from you." + }, + "answer_label": "B", + "answer_text": "It was placed on the blue table opposite you.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_7", + "correct_answer": "It was placed on the blue table opposite you.", + "original_options": { + "A": "On the wooden shelf beside you.", + "B": "It was placed on the blue table opposite you.", + "C": "In the drawer below you.", + "D": "It was set on the green table behind you rather than on the blue table opposite you.", + "E": "You placed it on the round table in front of you, not on the blue table across from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 19 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_8", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the portrait craft item?", + "answer": "It is right on the tripod stand next to you on your right.", + "question_chinese": "我把人像工艺品放在哪了？", + "answer_chinese": "就在你右边身旁的三脚台上。", + "evidence": [ + { + "description": "Put the portrait craft item from the tripod stand onto the blue table.", + "timestamps": { + "start": "00:07:52", + "end": "00:08:00" + } + }, + { + "description": "Put the portrait craft item back from the blue table onto the tripod stand.", + "timestamps": { + "start": "00:17:26", + "end": "00:17:45" + } + } + ], + "question_timestamps": "00:19:49", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the shelf behind you.", + "B": "On the table in front of you.", + "C": "It is placed on the chair next to you, on the left side rather than on your right.", + "D": "It is hanging on the wall opposite the door, not sitting on the tripod stand beside you.", + "E": "It is right on the tripod stand next to you on your right." + }, + "answer_label": "E", + "answer_text": "It is right on the tripod stand next to you on your right.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_8", + "correct_answer": "It is right on the tripod stand next to you on your right.", + "original_options": { + "A": "On the shelf behind you.", + "B": "On the table in front of you.", + "C": "It is placed on the chair next to you, on the left side rather than on your right.", + "D": "It is hanging on the wall opposite the door, not sitting on the tripod stand beside you.", + "E": "It is right on the tripod stand next to you on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 18, + 17 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_9", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the portrait craft item?", + "answer": "It is on the blue table opposite the table to your left.", + "question_chinese": "我把人像工艺品放在哪了？", + "answer_chinese": "在你左边的桌子对面的蓝色桌子上。", + "evidence": [ + { + "description": "Put the portrait craft item from the tripod stand onto the blue table.", + "timestamps": { + "start": "00:07:52", + "end": "00:08:00" + } + }, + { + "description": "Put the portrait craft item back from the blue table onto the tripod stand.", + "timestamps": { + "start": "00:17:26", + "end": "00:17:45" + } + }, + { + "description": "Put the portrait craft item from the tripod stand onto the blue table.", + "timestamps": { + "start": "00:32:09", + "end": "00:32:15" + } + } + ], + "question_timestamps": "00:33:59", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "On the desk near the door.", + "B": "On the green table to your left.", + "C": "It is on the blue table opposite the table to your left.", + "D": "You left it on the floor by the doorway in front of you, not on any blue table.", + "E": "It is on a blue table directly in front of you, instead of the one opposite the table to your left." + }, + "answer_label": "C", + "answer_text": "It is on the blue table opposite the table to your left.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_9", + "correct_answer": "It is on the blue table opposite the table to your left.", + "original_options": { + "A": "On the desk near the door.", + "B": "On the green table to your left.", + "C": "It is on the blue table opposite the table to your left.", + "D": "You left it on the floor by the doorway in front of you, not on any blue table.", + "E": "It is on a blue table directly in front of you, instead of the one opposite the table to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 21 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_10", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the portrait craft item?", + "answer": "It is on the blue table three steps away to your right.", + "question_chinese": "我把人像工艺品放在哪了？", + "answer_chinese": "在你右边三步远的蓝色桌子上。", + "evidence": [ + { + "description": "Put the portrait craft item from the tripod stand onto the blue table.", + "timestamps": { + "start": "00:07:52", + "end": "00:08:00" + } + }, + { + "description": "Put the portrait craft item back from the blue table onto the tripod stand.", + "timestamps": { + "start": "00:17:26", + "end": "00:17:45" + } + }, + { + "description": "Put the portrait craft item from the tripod stand onto the blue table.", + "timestamps": { + "start": "00:32:09", + "end": "00:32:15" + } + } + ], + "question_timestamps": "00:34:41", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "It is on the blue table three steps away to your right.", + "B": "On the blue table to your left.", + "C": "On the blue table one step right.", + "D": "It is on the wooden desk four steps away to your left, not on the blue table to your right.", + "E": "You placed it on the top shelf beside the window instead of the blue table three steps away." + }, + "answer_label": "A", + "answer_text": "It is on the blue table three steps away to your right.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_10", + "correct_answer": "It is on the blue table three steps away to your right.", + "original_options": { + "A": "It is on the blue table three steps away to your right.", + "B": "On the blue table to your left.", + "C": "On the blue table one step right.", + "D": "It is on the wooden desk four steps away to your left, not on the blue table to your right.", + "E": "You placed it on the top shelf beside the window instead of the blue table three steps away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 18 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the white spray bottle move just now?", + "answer": "The white spray bottle was picked up from the green stool, sprayed into the bag for a while, and finally placed on the table.", + "question_chinese": "白色喷雾瓶刚刚是怎么运动的？", + "answer_chinese": "白色的喷雾瓶从绿色凳子上被拿起，向袋子里喷洒了一会，最终被放在桌子上。", + "evidence": [ + { + "description": "There is a white spray bottle on the green stool opposite.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:40" + } + }, + { + "description": "The white spray bottle is picked up from the green stool, sprayed into the bag for a while, and finally placed on the table.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:18" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was moved from stool to table.", + "B": "It stayed on the green stool.", + "C": "The white spray bottle was picked up from the green stool, sprayed into the bag for a while, and finally placed on the table.", + "D": "The white spray bottle was picked up from the green stool, sprayed into the sink, and placed on the table.", + "E": "The white spray bottle was taken off the table, sprayed into the bag for a bit, and returned to the green stool." + }, + "answer_label": "C", + "answer_text": "The white spray bottle was picked up from the green stool, sprayed into the bag for a while, and finally placed on the table.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_1", + "correct_answer": "The white spray bottle was picked up from the green stool, sprayed into the bag for a while, and finally placed on the table.", + "original_options": { + "A": "It was moved from stool to table.", + "B": "It stayed on the green stool.", + "C": "The white spray bottle was picked up from the green stool, sprayed into the bag for a while, and finally placed on the table.", + "D": "The white spray bottle was picked up from the green stool, sprayed into the sink, and placed on the table.", + "E": "The white spray bottle was taken off the table, sprayed into the bag for a bit, and returned to the green stool." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 22 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the white spray bottle move just now?", + "answer": "The white spray bottle was picked up from the tripod stand, sprayed onto the towel for a while, and finally placed on the innermost miscellaneous table piled with things.", + "question_chinese": "白色喷雾瓶刚刚是怎么运动的？", + "answer_chinese": "白色喷雾瓶从三脚台上被拿起，对着毛巾喷洒一会，最终被放在最里面堆满东西的杂物桌上。", + "evidence": [ + { + "description": "The white spray bottle is picked up from the green stool, sprayed into the bag for a while, and finally placed on the table.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:18" + } + }, + { + "description": "The white spray bottle is picked up from the tripod stand, sprayed onto the towel for a while, and finally placed on the innermost miscellaneous table piled with things.", + "timestamps": { + "start": "00:34:35", + "end": "00:34:51" + } + }, + { + "description": "The white spray bottle is picked up from the table, sprayed into the bag for a while, then sprayed onto the porcelain ware, and finally placed on the tripod stand.", + "timestamps": { + "start": "00:32:32", + "end": "00:32:45" + } + } + ], + "question_timestamps": "00:38:33", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "It never left the tripod stand.", + "B": "It ended up on the table.", + "C": "The white spray bottle was picked up from the tripod stand, sprayed into the bag, and placed on the inner table.", + "D": "The white spray bottle was picked up from the tripod stand, sprayed onto the towel for a while, and finally placed on the innermost miscellaneous table piled with things.", + "E": "The white spray bottle was taken from the green stool, sprayed onto the towel briefly, and set on the miscellaneous table." + }, + "answer_label": "D", + "answer_text": "The white spray bottle was picked up from the tripod stand, sprayed onto the towel for a while, and finally placed on the innermost miscellaneous table piled with things.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_2", + "correct_answer": "The white spray bottle was picked up from the tripod stand, sprayed onto the towel for a while, and finally placed on the innermost miscellaneous table piled with things.", + "original_options": { + "A": "It never left the tripod stand.", + "B": "It ended up on the table.", + "C": "The white spray bottle was picked up from the tripod stand, sprayed into the bag, and placed on the inner table.", + "D": "The white spray bottle was picked up from the tripod stand, sprayed onto the towel for a while, and finally placed on the innermost miscellaneous table piled with things.", + "E": "The white spray bottle was taken from the green stool, sprayed onto the towel briefly, and set on the miscellaneous table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 21 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_14", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Standing where I am, can I reach that pottery bowl?", + "answer": "Yes, the pottery bowl is right on your left hand side.", + "question_chinese": "站在我所处的这个地方，我能够到那个陶碗吗？", + "answer_chinese": "可以，陶碗就在你的左手边", + "evidence": [ + { + "description": "Take an unpatterned pottery bowl blank from the bag.", + "timestamps": { + "start": "00:39:57", + "end": "00:40:02" + } + } + ], + "question_timestamps": "00:40:06", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you can’t reach it; it’s across the table.", + "B": "Yes, the pottery bowl is right on your left hand side." + }, + "answer_label": "B", + "answer_text": "Yes, the pottery bowl is right on your left hand side.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_14", + "correct_answer": "Yes, the pottery bowl is right on your left hand side.", + "original_options": { + "A": "No, you can’t reach it; it’s across the table.", + "B": "Yes, the pottery bowl is right on your left hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_15", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Standing where I am, can I reach that pottery bowl?", + "answer": "No, you cannot reach it; the pottery bowl is separated from you by a table.", + "question_chinese": "站在我所处的这个地方，我能够到那个陶碗吗？", + "answer_chinese": "你够不到，陶碗和你隔着一张桌子。", + "evidence": [ + { + "description": "Take an unpatterned pottery bowl blank from the bag.", + "timestamps": { + "start": "00:39:57", + "end": "00:40:02" + } + }, + { + "description": "Walk from the pottery bowl to the side of the cabinet.", + "timestamps": { + "start": "00:42:53", + "end": "00:42:57" + } + } + ], + "question_timestamps": "00:42:58", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you cannot reach it; the pottery bowl is separated from you by a table.", + "B": "Yes, you can reach it; the pottery bowl is right by your left hand." + }, + "answer_label": "A", + "answer_text": "No, you cannot reach it; the pottery bowl is separated from you by a table.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_15", + "correct_answer": "No, you cannot reach it; the pottery bowl is separated from you by a table.", + "original_options": { + "A": "No, you cannot reach it; the pottery bowl is separated from you by a table.", + "B": "Yes, you can reach it; the pottery bowl is right by your left hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_5", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From my current orientation, which side is the white spray bottle on?", + "answer": "It is on your right.", + "question_chinese": "站在我现在的方位，白色喷雾瓶在我的哪一边？", + "answer_chinese": "在你右边。", + "evidence": [ + { + "description": "The white spray bottle is picked up from the green stool, sprayed into the bag for a while, and finally placed on the table.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:18" + } + } + ], + "question_timestamps": "00:32:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your left.", + "B": "It is in front of you.", + "C": "It is on your right.", + "D": "It is behind you, positioned slightly toward your back-left side near the edge of the counter.", + "E": "It is in front of you, slightly toward your front-left side, close to the other cleaning items." + }, + "answer_label": "C", + "answer_text": "It is on your right.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_5", + "correct_answer": "It is on your right.", + "original_options": { + "A": "It is on your left.", + "B": "It is in front of you.", + "C": "It is on your right.", + "D": "It is behind you, positioned slightly toward your back-left side near the edge of the counter.", + "E": "It is in front of you, slightly toward your front-left side, close to the other cleaning items." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_4", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From my current orientation, which side is the white spray bottle on?", + "answer": "It is in front of you.", + "question_chinese": "站在我现在的方位，白色喷雾瓶在我的哪一边？", + "answer_chinese": "在你前面。", + "evidence": [ + { + "description": "The white spray bottle is picked up from the green stool, sprayed into the bag for a while, and finally placed on the table.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:18" + } + }, + { + "description": "The white spray bottle is picked up from the table, sprayed into the bag for a while, then sprayed onto the porcelain ware, and finally placed on the tripod stand.", + "timestamps": { + "start": "00:32:32", + "end": "00:32:45" + } + } + ], + "question_timestamps": "00:33:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is behind you.", + "C": "It is on your right.", + "D": "It is off to your left side, rather than directly in front of you where you are facing.", + "E": "It is on your right-hand side from your current orientation, not straight ahead of you." + }, + "answer_label": "A", + "answer_text": "It is in front of you.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_4", + "correct_answer": "It is in front of you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is behind you.", + "C": "It is on your right.", + "D": "It is off to your left side, rather than directly in front of you where you are facing.", + "E": "It is on your right-hand side from your current orientation, not straight ahead of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 15 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_3", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From my current orientation, which side is the white spray bottle on?", + "answer": "It is behind you.", + "question_chinese": "站在我现在的方位，白色喷雾瓶在我的哪一边？", + "answer_chinese": "在你后面。", + "evidence": [ + { + "description": "The white spray bottle is picked up from the green stool, sprayed into the bag for a while, and finally placed on the table.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:18" + } + }, + { + "description": "The white spray bottle is picked up from the tripod stand, sprayed onto the towel for a while, and finally placed on the innermost miscellaneous table piled with things.", + "timestamps": { + "start": "00:34:35", + "end": "00:34:51" + } + } + ], + "question_timestamps": "00:35:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is behind you.", + "C": "It is on your right.", + "D": "It is to your left side from your current orientation, not positioned behind you.", + "E": "It is in front of you in the direction you’re facing, rather than behind your back." + }, + "answer_label": "B", + "answer_text": "It is behind you.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_3", + "correct_answer": "It is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is behind you.", + "C": "It is on your right.", + "D": "It is to your left side from your current orientation, not positioned behind you.", + "E": "It is in front of you in the direction you’re facing, rather than behind your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 17 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_16", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there a pattern in the pottery bowl?", + "answer": "No, there is no pattern inside the pottery bowl.", + "question_chinese": "陶碗里有花纹吗？", + "answer_chinese": "没有，陶碗里面没有花纹。", + "evidence": [ + { + "description": "Take an unpatterned pottery bowl blank from the bag.", + "timestamps": { + "start": "00:39:57", + "end": "00:40:02" + } + }, + { + "description": "Put an unpatterned pottery bowl blank on the tripod stand.", + "timestamps": { + "start": "00:40:20", + "end": "00:40:30" + } + } + ], + "question_timestamps": "00:42:58", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, there is no pattern inside the pottery bowl.", + "B": "Yes, there is a patterned design inside the pottery bowl." + }, + "answer_label": "A", + "answer_text": "No, there is no pattern inside the pottery bowl.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_16", + "correct_answer": "No, there is no pattern inside the pottery bowl.", + "original_options": { + "A": "No, there is no pattern inside the pottery bowl.", + "B": "Yes, there is a patterned design inside the pottery bowl." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70", + "video_length": "00:53:20", + "qid": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_17", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there a pattern in the pottery bowl?", + "answer": "Yes, there is a newly carved pattern inside the pottery bowl.", + "question_chinese": "陶碗里有花纹吗？", + "answer_chinese": "有，陶碗里有刚刻画上去的花纹", + "evidence": [ + { + "description": "Take an unpatterned pottery bowl blank from the bag.", + "timestamps": { + "start": "00:39:57", + "end": "00:40:02" + } + }, + { + "description": "Put an unpatterned pottery bowl blank on the tripod stand.", + "timestamps": { + "start": "00:40:20", + "end": "00:40:30" + } + }, + { + "description": "Draw a pattern on the ceramic blank.", + "timestamps": { + "start": "00:45:00", + "end": "00:51:45" + } + } + ], + "question_timestamps": "00:52:00", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "No, there is no pattern inside the pottery bowl.", + "B": "Yes, there is a newly carved pattern inside the pottery bowl." + }, + "answer_label": "B", + "answer_text": "Yes, there is a newly carved pattern inside the pottery bowl.", + "q_id": "6c3e3490-9f33-4fec-9ce1-042e4d409a70_17", + "correct_answer": "Yes, there is a newly carved pattern inside the pottery bowl.", + "original_options": { + "A": "No, there is no pattern inside the pottery bowl.", + "B": "Yes, there is a newly carved pattern inside the pottery bowl." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + } +] diff --git a/QAs/6caf467a-a949-4b5c-b88b-b2ac6cbd9f36-mcq.json b/QAs/6caf467a-a949-4b5c-b88b-b2ac6cbd9f36-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..0d4b5f9da06a4a01b26bc94706125b71695521d6 --- /dev/null +++ b/QAs/6caf467a-a949-4b5c-b88b-b2ac6cbd9f36-mcq.json @@ -0,0 +1,1841 @@ +[ + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_10", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pairs of shoes were on the floor before you just organized the shoe cabinet?", + "answer": "Before you just organized the shoe cabinet, there were three pairs of shoes on the floor.", + "question_chinese": "刚刚整理鞋柜之前，地面上有多少双鞋", + "answer_chinese": "在你刚刚整理鞋柜之前，地面上有3双鞋", + "evidence": [ + { + "description": "Three pairs of shoes", + "timestamps": { + "start": "00:30:35", + "end": "00:30:55" + } + } + ], + "question_timestamps": "00:31:14", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One pair of shoes.", + "B": "Before you just organized the shoe cabinet, there were three pairs of shoes on the floor.", + "C": "Four pairs of shoes.", + "D": "Before you organized the shoe cabinet, there were two pairs of shoes on the floor.", + "E": "Before you just organized the shoe cabinet, there were five pairs of shoes left on the floor." + }, + "answer_label": "B", + "answer_text": "Before you just organized the shoe cabinet, there were three pairs of shoes on the floor.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_10", + "correct_answer": "Before you just organized the shoe cabinet, there were three pairs of shoes on the floor.", + "original_options": { + "A": "One pair of shoes.", + "B": "Before you just organized the shoe cabinet, there were three pairs of shoes on the floor.", + "C": "Four pairs of shoes.", + "D": "Before you organized the shoe cabinet, there were two pairs of shoes on the floor.", + "E": "Before you just organized the shoe cabinet, there were five pairs of shoes left on the floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_21", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pairs of shoes were on the floor before you just organized the shoe cabinet?", + "answer": "Before you just organized the shoe cabinet, there were three pairs of shoes on the floor.", + "question_chinese": "刚刚整理鞋柜之前，地面上有多少双鞋", + "answer_chinese": "在你刚刚整理鞋柜之前，地面上有3双鞋", + "evidence": [ + { + "description": "Three pairs of shoes", + "timestamps": { + "start": "00:30:35", + "end": "00:30:55" + } + } + ], + "question_timestamps": "00:53:18", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two pairs of shoes.", + "B": "Five pairs of shoes.", + "C": "Before you just organized the shoe cabinet, there was one pair of shoes on the floor.", + "D": "Before you organized the shoe cabinet, there were four pairs of shoes scattered on the floor.", + "E": "Before you just organized the shoe cabinet, there were three pairs of shoes on the floor." + }, + "answer_label": "E", + "answer_text": "Before you just organized the shoe cabinet, there were three pairs of shoes on the floor.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_21", + "correct_answer": "Before you just organized the shoe cabinet, there were three pairs of shoes on the floor.", + "original_options": { + "A": "Two pairs of shoes.", + "B": "Five pairs of shoes.", + "C": "Before you just organized the shoe cabinet, there was one pair of shoes on the floor.", + "D": "Before you organized the shoe cabinet, there were four pairs of shoes scattered on the floor.", + "E": "Before you just organized the shoe cabinet, there were three pairs of shoes on the floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_5", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the vacuum cleaner take from the utility room to the living room?", + "answer": "You retrieved the vacuum cleaner from the utility room, exited the utility room door, passed a football, then went through the living room door and into the living room.", + "question_chinese": "吸尘器从杂物间到客厅经过了什么路径", + "answer_chinese": "从杂物间取到了吸尘器，并出杂物间的门，经过一个足球，然后再经过了客厅的门，带入到了客厅", + "evidence": [ + { + "description": "Take the vacuum cleaner out of the utility room.", + "timestamps": { + "start": "00:20:16", + "end": "00:20:38" + } + }, + { + "description": "Walk past the football and through the living room door.", + "timestamps": { + "start": "00:20:40", + "end": "00:20:48" + } + }, + { + "description": "Put down the vacuum cleaner.", + "timestamps": { + "start": "00:20:49", + "end": "00:21:06" + } + } + ], + "question_timestamps": "00:21:08", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You went straight into the living room.", + "B": "You retrieved the vacuum cleaner from the utility room, exited the utility room door, passed a football, then went through the living room door and into the living room.", + "C": "You entered the living room first.", + "D": "You retrieved the vacuum, passed the football later, and then went into the living room through the door.", + "E": "You left the utility room, entered the living room door, and never passed the football on the way." + }, + "answer_label": "B", + "answer_text": "You retrieved the vacuum cleaner from the utility room, exited the utility room door, passed a football, then went through the living room door and into the living room.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_5", + "correct_answer": "You retrieved the vacuum cleaner from the utility room, exited the utility room door, passed a football, then went through the living room door and into the living room.", + "original_options": { + "A": "You went straight into the living room.", + "B": "You retrieved the vacuum cleaner from the utility room, exited the utility room door, passed a football, then went through the living room door and into the living room.", + "C": "You entered the living room first.", + "D": "You retrieved the vacuum, passed the football later, and then went into the living room through the door.", + "E": "You left the utility room, entered the living room door, and never passed the football on the way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 18 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_16", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the vacuum cleaner take from the utility room to the living room?", + "answer": "You retrieved the vacuum cleaner from the utility room, exited the utility room door, passed a football, then went through the living room door and into the living room.", + "question_chinese": "吸尘器从杂物间到客厅经过了什么路径", + "answer_chinese": "从杂物间取到了吸尘器，并出杂物间的门，经过一个足球，然后再经过了客厅的门，带入到了客厅", + "evidence": [ + { + "description": "Take the vacuum cleaner out of the utility room.", + "timestamps": { + "start": "00:20:16", + "end": "00:20:38" + } + }, + { + "description": "Crossing the football past the living room door", + "timestamps": { + "start": "00:20:40", + "end": "00:20:48" + } + }, + { + "description": "Put away the vacuum cleaner.", + "timestamps": { + "start": "00:20:49", + "end": "00:21:06" + } + } + ], + "question_timestamps": "00:53:21", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You went straight to the living room.", + "B": "You passed the football first.", + "C": "You retrieved the vacuum cleaner, passed a football, then entered the living room through a different doorway.", + "D": "You retrieved the vacuum cleaner, exited the utility room, and entered the living room without passing the football.", + "E": "You retrieved the vacuum cleaner from the utility room, exited the utility room door, passed a football, then went through the living room door and into the living room." + }, + "answer_label": "E", + "answer_text": "You retrieved the vacuum cleaner from the utility room, exited the utility room door, passed a football, then went through the living room door and into the living room.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_16", + "correct_answer": "You retrieved the vacuum cleaner from the utility room, exited the utility room door, passed a football, then went through the living room door and into the living room.", + "original_options": { + "A": "You went straight to the living room.", + "B": "You passed the football first.", + "C": "You retrieved the vacuum cleaner, passed a football, then entered the living room through a different doorway.", + "D": "You retrieved the vacuum cleaner, exited the utility room, and entered the living room without passing the football.", + "E": "You retrieved the vacuum cleaner from the utility room, exited the utility room door, passed a football, then went through the living room door and into the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 17, + 18 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_9", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "In the restroom, how far is the sink from the toilet?", + "answer": "The sink is on the left and the toilet is on the right, with the sink approximately two meters horizontally from the toilet.", + "question_chinese": "在洗手间中，洗手盆距离马桶有多远", + "answer_chinese": "洗手盆在左边，马桶在右边，洗手盆距离马桶约2m，水平距离", + "evidence": [ + { + "description": "Toilet", + "timestamps": { + "start": "00:24:48", + "end": "00:24:49" + } + }, + { + "description": "Overall", + "timestamps": { + "start": "00:24:47", + "end": "00:24:50" + } + } + ], + "question_timestamps": "00:24:59", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "About one meter apart.", + "B": "About four meters apart.", + "C": "The sink is on the right and the toilet is on the left, about two meters apart horizontally.", + "D": "The sink is on the left and the toilet is on the right, with the sink approximately two meters horizontally from the toilet.", + "E": "The sink is on the left and the toilet is on the right, separated by only about one meter horizontally." + }, + "answer_label": "D", + "answer_text": "The sink is on the left and the toilet is on the right, with the sink approximately two meters horizontally from the toilet.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_9", + "correct_answer": "The sink is on the left and the toilet is on the right, with the sink approximately two meters horizontally from the toilet.", + "original_options": { + "A": "About one meter apart.", + "B": "About four meters apart.", + "C": "The sink is on the right and the toilet is on the left, about two meters apart horizontally.", + "D": "The sink is on the left and the toilet is on the right, with the sink approximately two meters horizontally from the toilet.", + "E": "The sink is on the left and the toilet is on the right, separated by only about one meter horizontally." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 20 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_20", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "In the restroom, how far is the sink from the toilet?", + "answer": "The sink is on the left and the toilet is on the right, with the sink approximately two meters horizontally from the toilet.", + "question_chinese": "在洗手间中，洗手盆距离马桶有多远", + "answer_chinese": "洗手盆在左边，马桶在右边，洗手盆距离马桶约2m，水平距离", + "evidence": [ + { + "description": "Toilet", + "timestamps": { + "start": "00:24:48", + "end": "00:24:49" + } + }, + { + "description": "Overall", + "timestamps": { + "start": "00:24:47", + "end": "00:24:50" + } + } + ], + "question_timestamps": "00:30:49", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "About half a meter away.", + "B": "About three meters away.", + "C": "The sink is on the right and the toilet is on the left, with about two meters between them.", + "D": "The sink is on the left and the toilet is on the right, with the sink roughly four meters from the toilet.", + "E": "The sink is on the left and the toilet is on the right, with the sink approximately two meters horizontally from the toilet." + }, + "answer_label": "E", + "answer_text": "The sink is on the left and the toilet is on the right, with the sink approximately two meters horizontally from the toilet.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_20", + "correct_answer": "The sink is on the left and the toilet is on the right, with the sink approximately two meters horizontally from the toilet.", + "original_options": { + "A": "About half a meter away.", + "B": "About three meters away.", + "C": "The sink is on the right and the toilet is on the left, with about two meters between them.", + "D": "The sink is on the left and the toilet is on the right, with the sink roughly four meters from the toilet.", + "E": "The sink is on the left and the toilet is on the right, with the sink approximately two meters horizontally from the toilet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 19, + 22 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_7_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I left the bathroom, which direction was the soccer ball relative to me?", + "answer": "The soccer ball was directly in front of you.", + "question_chinese": "当我离开洗手间时，足球在相对于我什么方向的位置？", + "answer_chinese": "足球在你的正前方。", + "evidence": [ + { + "description": "Seeing football", + "timestamps": { + "start": "00:23:30", + "end": "00:23:31" + } + } + ], + "question_timestamps": "00:23:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly behind you.", + "B": "The soccer ball was directly in front of you.", + "C": "Directly to your left.", + "D": "The soccer ball was slightly behind you and to your right as you left the bathroom.", + "E": "The soccer ball was off to your left side, not directly in front when you exited." + }, + "answer_label": "B", + "answer_text": "The soccer ball was directly in front of you.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_7_T1", + "correct_answer": "The soccer ball was directly in front of you.", + "original_options": { + "A": "Directly behind you.", + "B": "The soccer ball was directly in front of you.", + "C": "Directly to your left.", + "D": "The soccer ball was slightly behind you and to your right as you left the bathroom.", + "E": "The soccer ball was off to your left side, not directly in front when you exited." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 16 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_7", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When you leave the restroom and start heading back towards the living room along the same path, the football is in front of you, but beyond the living room door. It's located on the path between the living room door and the utility room door.", + "answer": "The football is to my left and behind me.", + "question_chinese": "当我离开洗手间时，足球在相对于我什么方向的位置", + "answer_chinese": "足球在相对于我的左后方", + "evidence": [ + { + "description": "Seeing football", + "timestamps": { + "start": "00:23:34", + "end": "00:23:36" + } + } + ], + "question_timestamps": "00:23:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The football is left-front.", + "B": "The football is right-back.", + "C": "The football is to my left and behind me.", + "D": "The football is directly ahead, just past the living room door as you walk back toward it.", + "E": "The football is to your right in front of you while heading from the restroom toward the living room." + }, + "answer_label": "C", + "answer_text": "The football is to my left and behind me.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_7", + "correct_answer": "The football is to my left and behind me.", + "original_options": { + "A": "The football is left-front.", + "B": "The football is right-back.", + "C": "The football is to my left and behind me.", + "D": "The football is directly ahead, just past the living room door as you walk back toward it.", + "E": "The football is to your right in front of you while heading from the restroom toward the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 19 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_7_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I left the bathroom, which direction was the soccer ball relative to me?", + "answer": "The soccer ball was to your back-left relative to you.", + "question_chinese": "当我离开洗手间时，足球在相对于我什么方向的位置？", + "answer_chinese": "足球在相对于我的左后方", + "evidence": [ + { + "description": "See football", + "timestamps": { + "start": "00:23:34", + "end": "00:23:36" + } + } + ], + "question_timestamps": "00:31:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was to your back-right.", + "B": "It was to your front-left.", + "C": "The soccer ball was to your back-left relative to you.", + "D": "The soccer ball was directly in front of you as you stepped out of the bathroom doorway.", + "E": "The soccer ball was to your front-right relative to you when you first left the bathroom." + }, + "answer_label": "C", + "answer_text": "The soccer ball was to your back-left relative to you.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_7_T2", + "correct_answer": "The soccer ball was to your back-left relative to you.", + "original_options": { + "A": "It was to your back-right.", + "B": "It was to your front-left.", + "C": "The soccer ball was to your back-left relative to you.", + "D": "The soccer ball was directly in front of you as you stepped out of the bathroom doorway.", + "E": "The soccer ball was to your front-right relative to you when you first left the bathroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_8_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in the living room did I put the vacuum cleaner I took out earlier?", + "answer": "You were holding the vacuum cleaner in your hand.", + "question_chinese": "我之前拿出来的吸尘器放在客厅的哪里了？", + "answer_chinese": "吸尘器正被你拿在了手上。", + "evidence": [ + { + "description": "You took out the vacuum cleaner.", + "timestamps": { + "start": "00:20:21", + "end": "00:20:23" + } + } + ], + "question_timestamps": "00:23:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You left it by the sofa.", + "B": "You set it near the outlet.", + "C": "You placed the vacuum cleaner on the floor between the outlet by the door and the sofa.", + "D": "You were holding the vacuum cleaner in your hand.", + "E": "You leaned the vacuum cleaner against the wall beside the door, near the outlet in the living room." + }, + "answer_label": "D", + "answer_text": "You were holding the vacuum cleaner in your hand.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_8_T1", + "correct_answer": "You were holding the vacuum cleaner in your hand.", + "original_options": { + "A": "You left it by the sofa.", + "B": "You set it near the outlet.", + "C": "You placed the vacuum cleaner on the floor between the outlet by the door and the sofa.", + "D": "You were holding the vacuum cleaner in your hand.", + "E": "You leaned the vacuum cleaner against the wall beside the door, near the outlet in the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_8_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in the living room did I put the vacuum cleaner I took out earlier?", + "answer": "It was between the outlet next to the door and the sofa.", + "question_chinese": "我之前拿出来的吸尘器放在客厅的哪里了？", + "answer_chinese": "在门的旁边的插座和沙发中间", + "evidence": [ + { + "description": "Put away the vacuum cleaner.", + "timestamps": { + "start": "00:23:15", + "end": "00:23:22" + } + } + ], + "question_timestamps": "00:38:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were holding it in your hand.", + "B": "It was near the living room door.", + "C": "You put the vacuum cleaner on the floor in front of the sofa, not between the outlet and sofa.", + "D": "It was between the outlet next to the door and the sofa.", + "E": "You left the vacuum cleaner next to the outlet by the door, but not between it and the sofa." + }, + "answer_label": "D", + "answer_text": "It was between the outlet next to the door and the sofa.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_8_T2", + "correct_answer": "It was between the outlet next to the door and the sofa.", + "original_options": { + "A": "You were holding it in your hand.", + "B": "It was near the living room door.", + "C": "You put the vacuum cleaner on the floor in front of the sofa, not between the outlet and sofa.", + "D": "It was between the outlet next to the door and the sofa.", + "E": "You left the vacuum cleaner next to the outlet by the door, but not between it and the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_2_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the scissors I just used to trim the branches?", + "answer": "The scissors are in your hand now.", + "question_chinese": "我刚刚修剪树枝的剪刀放到哪里了？", + "answer_chinese": "剪刀现在在你的手里。", + "evidence": [ + { + "description": "You are pruning the branches.", + "timestamps": { + "start": "00:02:46", + "end": "00:02:47" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The scissors are in your hand now.", + "B": "You put them on the table.", + "C": "You dropped them on the ground.", + "D": "You set the scissors down on the table next to the tree in front of you.", + "E": "You placed the scissors on the ground beside the tree after trimming the branches." + }, + "answer_label": "A", + "answer_text": "The scissors are in your hand now.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_2_T1", + "correct_answer": "The scissors are in your hand now.", + "original_options": { + "A": "The scissors are in your hand now.", + "B": "You put them on the table.", + "C": "You dropped them on the ground.", + "D": "You set the scissors down on the table next to the tree in front of you.", + "E": "You placed the scissors on the ground beside the tree after trimming the branches." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 14 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the pruning shears I just used?", + "answer": "The shears are on the table next to the tree.", + "question_chinese": "我刚刚修剪树枝的剪刀放到哪里了？", + "answer_chinese": "剪刀在树旁边的桌子上", + "evidence": [ + { + "description": "Put down the scissors.", + "timestamps": { + "start": "00:02:54", + "end": "00:02:57" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They are on the windowsill.", + "B": "They are on the patio.", + "C": "The shears are on the table next to the tree.", + "D": "You left the shears on the garden bench beside the tree instead of on the table.", + "E": "The shears are resting on the fence post near the tree, not on the table." + }, + "answer_label": "C", + "answer_text": "The shears are on the table next to the tree.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_2", + "correct_answer": "The shears are on the table next to the tree.", + "original_options": { + "A": "They are on the windowsill.", + "B": "They are on the patio.", + "C": "The shears are on the table next to the tree.", + "D": "You left the shears on the garden bench beside the tree instead of on the table.", + "E": "The shears are resting on the fence post near the tree, not on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_2_T2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the scissors I just used to trim the branches?", + "answer": "The scissors were on the table next to the tree to your left.", + "question_chinese": "我刚刚修剪树枝的剪刀放到哪里了？", + "answer_chinese": "剪刀在你左边的树旁边的桌子上", + "evidence": [ + { + "description": "Put down the scissors.", + "timestamps": { + "start": "00:02:54", + "end": "00:02:57" + } + } + ], + "question_timestamps": "00:29:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They were in front of you.", + "B": "They were to your right.", + "C": "The scissors were on the table next to the tree to your left.", + "D": "The scissors were on the table next to the tree in front of you, not to your left.", + "E": "You left the scissors on the table next to the tree behind you after finishing the trimming." + }, + "answer_label": "C", + "answer_text": "The scissors were on the table next to the tree to your left.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_2_T2", + "correct_answer": "The scissors were on the table next to the tree to your left.", + "original_options": { + "A": "They were in front of you.", + "B": "They were to your right.", + "C": "The scissors were on the table next to the tree to your left.", + "D": "The scissors were on the table next to the tree in front of you, not to your left.", + "E": "You left the scissors on the table next to the tree behind you after finishing the trimming." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 17 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_3_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just went to throw out the trash?", + "answer": "You turned right all the way to the end and then opened the door.", + "question_chinese": "我刚刚去扔垃圾走了什么路径？", + "answer_chinese": "你向右转到尽头然后打开了门。", + "evidence": [ + { + "description": "Start beside the tree.", + "timestamps": { + "start": "00:02:58", + "end": "00:03:00" + } + }, + { + "description": "Turn right and walk to the end.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:09" + } + }, + { + "description": "Turn right after opening the door.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:03:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right all the way to the end and then opened the door.", + "B": "You turned left to the end.", + "C": "You went straight to the door.", + "D": "You turned right, walked only a couple of steps, and opened the door immediately.", + "E": "You turned right to the end, opened the door, then continued past it and turned right again." + }, + "answer_label": "A", + "answer_text": "You turned right all the way to the end and then opened the door.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_3_T1", + "correct_answer": "You turned right all the way to the end and then opened the door.", + "original_options": { + "A": "You turned right all the way to the end and then opened the door.", + "B": "You turned left to the end.", + "C": "You went straight to the door.", + "D": "You turned right, walked only a couple of steps, and opened the door immediately.", + "E": "You turned right to the end, opened the door, then continued past it and turned right again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 17 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_3_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just went to throw out the trash?", + "answer": "You started next to the tree, turned right and went all the way to the end, opened the door, and then turned right again.", + "question_chinese": "我刚刚去扔垃圾走了什么路径？", + "answer_chinese": "你从树旁边开始，向右转到尽头，开门后再向右转", + "evidence": [ + { + "description": "Start beside the tree.", + "timestamps": { + "start": "00:02:58", + "end": "00:03:00" + } + }, + { + "description": "Turn right and walk to the end.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:09" + } + }, + { + "description": "Turn right after opening the door.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:53:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left at the tree.", + "B": "You started next to the tree, turned right and went all the way to the end, opened the door, and then turned right again.", + "C": "You opened the door, then turned left.", + "D": "You started by the tree, turned left to the end, opened the door, then turned right.", + "E": "You began next to the tree, turned right to the end, opened the door, then turned left." + }, + "answer_label": "B", + "answer_text": "You started next to the tree, turned right and went all the way to the end, opened the door, and then turned right again.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_3_T2", + "correct_answer": "You started next to the tree, turned right and went all the way to the end, opened the door, and then turned right again.", + "original_options": { + "A": "You turned left at the tree.", + "B": "You started next to the tree, turned right and went all the way to the end, opened the door, and then turned right again.", + "C": "You opened the door, then turned left.", + "D": "You started by the tree, turned left to the end, opened the door, then turned right.", + "E": "You began next to the tree, turned right to the end, opened the door, then turned left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_1_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was just looking for the mirror, how many candles were on the tabletop?", + "answer": "There were four candles on the table to your front-left.", + "question_chinese": "我刚刚找镜子时，桌面上有几个蜡烛？", + "answer_chinese": "你左前方的桌子上有四根蜡烛", + "evidence": [ + { + "description": "Evidence of four Candles", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three candles.", + "B": "Five candles.", + "C": "There were two candles on the table to your front-left when you were looking for the mirror.", + "D": "There were six candles on the table to your front-left during the mirror search.", + "E": "There were four candles on the table to your front-left." + }, + "answer_label": "E", + "answer_text": "There were four candles on the table to your front-left.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_1_T1", + "correct_answer": "There were four candles on the table to your front-left.", + "original_options": { + "A": "Three candles.", + "B": "Five candles.", + "C": "There were two candles on the table to your front-left when you were looking for the mirror.", + "D": "There were six candles on the table to your front-left during the mirror search.", + "E": "There were four candles on the table to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 14 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was just looking for the mirror, how many candles were on the desk?", + "answer": "There are six candles on the table.", + "question_chinese": "我刚刚找镜子时，桌面上有几个蜡烛？", + "answer_chinese": "桌子上有6根蜡烛", + "evidence": [ + { + "description": "Evidence of Six Candles", + "timestamps": { + "start": "00:00:32", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five candles.", + "B": "There are six candles on the table.", + "C": "Three candles.", + "D": "There are four candles on the table when you were looking for the mirror.", + "E": "There is one candle on the table when you were looking for the mirror." + }, + "answer_label": "B", + "answer_text": "There are six candles on the table.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_1", + "correct_answer": "There are six candles on the table.", + "original_options": { + "A": "Five candles.", + "B": "There are six candles on the table.", + "C": "Three candles.", + "D": "There are four candles on the table when you were looking for the mirror.", + "E": "There is one candle on the table when you were looking for the mirror." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_1_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was just looking for the mirror, how many candles were on the tabletop?", + "answer": "There are 6 candles on the table in front of you.", + "question_chinese": "我刚刚找镜子时，桌面上有几个蜡烛？", + "answer_chinese": "你你的正前方的桌子上有6根蜡烛", + "evidence": [ + { + "description": "Six candle evidence", + "timestamps": { + "start": "00:00:32", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:28:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 6 candles on the table in front of you.", + "B": "Five candles.", + "C": "Seven candles.", + "D": "There are four candles on the table in front of you during the mirror search.", + "E": "There are eight candles on the table in front of you when you were looking for the mirror." + }, + "answer_label": "A", + "answer_text": "There are 6 candles on the table in front of you.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_1_T2", + "correct_answer": "There are 6 candles on the table in front of you.", + "original_options": { + "A": "There are 6 candles on the table in front of you.", + "B": "Five candles.", + "C": "Seven candles.", + "D": "There are four candles on the table in front of you during the mirror search.", + "E": "There are eight candles on the table in front of you when you were looking for the mirror." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 18 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_10_T1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Before I just organized the shoe cabinet, how many pairs of shoes were on the floor?", + "answer": "The shoe cabinet is to your front-right, and there are 2 pairs of shoes on the floor.", + "question_chinese": "我刚刚整理鞋柜之前，地面上有多少双鞋？", + "answer_chinese": "鞋柜在你的右前方，地面上有2双鞋", + "evidence": [ + { + "description": "You are about to organize the shoe cabinet.", + "timestamps": { + "start": "00:29:35", + "end": "00:29:55" + } + } + ], + "question_timestamps": "00:30:31", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One pair of shoes.", + "B": "Three pairs of shoes.", + "C": "The shoe cabinet is to your front-right, and there are 2 pairs of shoes on the floor.", + "D": "The shoe cabinet is to your front-right, and there are zero pairs of shoes on the floor.", + "E": "The shoe cabinet is to your front-right, and there are four pairs of shoes on the floor." + }, + "answer_label": "C", + "answer_text": "The shoe cabinet is to your front-right, and there are 2 pairs of shoes on the floor.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_10_T1", + "correct_answer": "The shoe cabinet is to your front-right, and there are 2 pairs of shoes on the floor.", + "original_options": { + "A": "One pair of shoes.", + "B": "Three pairs of shoes.", + "C": "The shoe cabinet is to your front-right, and there are 2 pairs of shoes on the floor.", + "D": "The shoe cabinet is to your front-right, and there are zero pairs of shoes on the floor.", + "E": "The shoe cabinet is to your front-right, and there are four pairs of shoes on the floor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_10_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Before I just organized the shoe cabinet, how many pairs of shoes were on the floor?", + "answer": "The shoe cabinet is in front of you, and there are 3 pairs of shoes on the floor.", + "question_chinese": "我刚刚整理鞋柜之前，地面上有多少双鞋？", + "answer_chinese": "鞋柜在你的正前方，地面上有3双鞋", + "evidence": [ + { + "description": "Three pairs of shoes", + "timestamps": { + "start": "00:30:35", + "end": "00:30:55" + } + } + ], + "question_timestamps": "00:53:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two pairs of shoes.", + "B": "Four pairs of shoes.", + "C": "The shoe cabinet is in front of you, and there is one pair of shoes on the floor.", + "D": "The shoe cabinet is in front of you, and there are 3 pairs of shoes on the floor.", + "E": "The shoe cabinet is in front of you, and there are five pairs of shoes on the floor." + }, + "answer_label": "D", + "answer_text": "The shoe cabinet is in front of you, and there are 3 pairs of shoes on the floor.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_10_T2", + "correct_answer": "The shoe cabinet is in front of you, and there are 3 pairs of shoes on the floor.", + "original_options": { + "A": "Two pairs of shoes.", + "B": "Four pairs of shoes.", + "C": "The shoe cabinet is in front of you, and there is one pair of shoes on the floor.", + "D": "The shoe cabinet is in front of you, and there are 3 pairs of shoes on the floor.", + "E": "The shoe cabinet is in front of you, and there are five pairs of shoes on the floor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 18 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_4_T1", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was sweeping just now, could I reach the hose directly?", + "answer": "You could reach it—the hose was under your feet.", + "question_chinese": "我刚才在扫地的时候，能直接够到水管吗？", + "answer_chinese": "能够到，水管在你的脚下。", + "evidence": [ + { + "description": "The water pipe is beneath your feet.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:14" + } + } + ], + "question_timestamps": "00:10:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn’t reach it because you had just moved the hose behind the wall.", + "B": "You could reach it—the hose was under your feet." + }, + "answer_label": "B", + "answer_text": "You could reach it—the hose was under your feet.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_4_T1", + "correct_answer": "You could reach it—the hose was under your feet.", + "original_options": { + "A": "You couldn’t reach it because you had just moved the hose behind the wall.", + "B": "You could reach it—the hose was under your feet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_4", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was sweeping just now, could I directly reach the hose?", + "answer": "You couldn't reach it because you just moved the hose behind the wall.", + "question_chinese": "我刚才在扫地的时候，能直接够到水管吗？", + "answer_chinese": "不能够到，因为你刚刚把水管移动到了墙后面", + "evidence": [ + { + "description": "Mobile water pipe", + "timestamps": { + "start": "00:10:00", + "end": "00:10:34" + } + } + ], + "question_timestamps": "00:12:57", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You could reach it—the hose was under your feet.", + "B": "You couldn't reach it because you just moved the hose behind the wall." + }, + "answer_label": "B", + "answer_text": "You couldn't reach it because you just moved the hose behind the wall.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_4", + "correct_answer": "You couldn't reach it because you just moved the hose behind the wall.", + "original_options": { + "A": "You could reach it—the hose was under your feet.", + "B": "You couldn't reach it because you just moved the hose behind the wall." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_4_T2", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was sweeping just now, could I reach the hose directly?", + "answer": "You couldn’t reach it, because you had just moved the hose behind the wall.", + "question_chinese": "我刚才在扫地的时候，能直接够到水管吗？", + "answer_chinese": "不能够到，因为你刚刚把水管移动到了墙后面", + "evidence": [ + { + "description": "Mobile water pipe", + "timestamps": { + "start": "00:10:00", + "end": "00:10:34" + } + } + ], + "question_timestamps": "00:33:43", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn’t reach it, because you had just moved the hose behind the wall.", + "B": "You could reach it, since the hose was right beside you." + }, + "answer_label": "A", + "answer_text": "You couldn’t reach it, because you had just moved the hose behind the wall.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_4_T2", + "correct_answer": "You couldn’t reach it, because you had just moved the hose behind the wall.", + "original_options": { + "A": "You couldn’t reach it, because you had just moved the hose behind the wall.", + "B": "You could reach it, since the hose was right beside you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_6_T1", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was using the vacuum cleaner, could I directly reach the water bottle on the table?", + "answer": "You couldn’t reach it. The water bottle was on the table by your left hand, and you were mopping the floor.", + "question_chinese": "我在使用吸尘器的时候，能直接够到桌子上的水瓶吗？", + "answer_chinese": "不能够到，水瓶就在你左手边的桌子上，你现在正在拖地。", + "evidence": [ + { + "description": "Seeing the water bottle.", + "timestamps": { + "start": "00:21:42", + "end": "00:22:10" + } + } + ], + "question_timestamps": "00:22:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You could reach it. The water bottle was on the table by your left hand while you were vacuuming.", + "B": "You couldn’t reach it. The water bottle was on the table by your left hand, and you were mopping the floor." + }, + "answer_label": "B", + "answer_text": "You couldn’t reach it. The water bottle was on the table by your left hand, and you were mopping the floor.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_6_T1", + "correct_answer": "You couldn’t reach it. The water bottle was on the table by your left hand, and you were mopping the floor.", + "original_options": { + "A": "You could reach it. The water bottle was on the table by your left hand while you were vacuuming.", + "B": "You couldn’t reach it. The water bottle was on the table by your left hand, and you were mopping the floor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 19 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_6", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was using the vacuum cleaner, could I directly reach the water bottle on the table?", + "answer": "You could reach it. The water bottle was on the table to your left, within arm's reach.", + "question_chinese": "我在使用吸尘器的时候，能直接够到桌子上的水瓶吗", + "answer_chinese": "能够到，水瓶就在你左手边的桌子上，伸手就能拿到", + "evidence": [ + { + "description": "Seeing the water bottle.", + "timestamps": { + "start": "00:21:42", + "end": "00:22:10" + } + } + ], + "question_timestamps": "00:22:29", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You couldn’t reach it. The water bottle was on the table to your left, but it was out of arm’s reach while you were vacuuming.", + "B": "You could reach it. The water bottle was on the table to your left, within arm's reach." + }, + "answer_label": "B", + "answer_text": "You could reach it. The water bottle was on the table to your left, within arm's reach.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_6", + "correct_answer": "You could reach it. The water bottle was on the table to your left, within arm's reach.", + "original_options": { + "A": "You couldn’t reach it. The water bottle was on the table to your left, but it was out of arm’s reach while you were vacuuming.", + "B": "You could reach it. The water bottle was on the table to your left, within arm's reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 27 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_6_T2", + "qid_num": 27, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was using the vacuum cleaner, could I directly reach the water bottle on the table?", + "answer": "You could reach it—the water bottle was on the table by your left hand, so you could grab it just by reaching out.", + "question_chinese": "我在使用吸尘器的时候，能直接够到桌子上的水瓶吗？", + "answer_chinese": "能够到，水瓶就在你左手边的桌子上，伸手就能拿到", + "evidence": [ + { + "description": "Seeing the water bottle.", + "timestamps": { + "start": "00:21:42", + "end": "00:22:10" + } + } + ], + "question_timestamps": "00:37:53", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You could reach it—the water bottle was on the table by your left hand, so you could grab it just by reaching out.", + "B": "You couldn’t reach it—the water bottle was on the table by your left hand, but it was still too far to grab directly while you were vacuuming." + }, + "answer_label": "A", + "answer_text": "You could reach it—the water bottle was on the table by your left hand, so you could grab it just by reaching out.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_6_T2", + "correct_answer": "You could reach it—the water bottle was on the table by your left hand, so you could grab it just by reaching out.", + "original_options": { + "A": "You could reach it—the water bottle was on the table by your left hand, so you could grab it just by reaching out.", + "B": "You couldn’t reach it—the water bottle was on the table by your left hand, but it was still too far to grab directly while you were vacuuming." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 29 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_11_T1", + "qid_num": 28, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I picked up the tablet, how many types of electronic devices were still on the table?", + "answer": "There were also four kinds of electronic devices on the table in front of you: a laptop, a mouse, headphones, and a tablet.", + "question_chinese": "我在拿平板的时候，桌上还有几种电子设备？", + "answer_chinese": "你面前桌子上还有四种电子设备，分别是笔记本，鼠标，耳机，平板", + "evidence": [ + { + "description": "Electronic devices", + "timestamps": { + "start": "00:34:39", + "end": "00:34:40" + } + } + ], + "question_timestamps": "00:34:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three kinds: laptop, mouse, headphones.", + "B": "Two kinds: laptop and mouse.", + "C": "Five kinds were still on the table: a laptop, a mouse, headphones, a tablet, and a charger.", + "D": "There were also four kinds of electronic devices on the table in front of you: a laptop, a mouse, headphones, and a tablet.", + "E": "Only one kind remained on the table in front of you: just the laptop." + }, + "answer_label": "D", + "answer_text": "There were also four kinds of electronic devices on the table in front of you: a laptop, a mouse, headphones, and a tablet.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_11_T1", + "correct_answer": "There were also four kinds of electronic devices on the table in front of you: a laptop, a mouse, headphones, and a tablet.", + "original_options": { + "A": "Three kinds: laptop, mouse, headphones.", + "B": "Two kinds: laptop and mouse.", + "C": "Five kinds were still on the table: a laptop, a mouse, headphones, a tablet, and a charger.", + "D": "There were also four kinds of electronic devices on the table in front of you: a laptop, a mouse, headphones, and a tablet.", + "E": "Only one kind remained on the table in front of you: just the laptop." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 14 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_11", + "qid_num": 29, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was picking up the tablet, how many other electronic devices were on the table?", + "answer": "There were three other electronic devices on the table: a laptop, a mouse, and headphones.", + "question_chinese": "我在拿平板的时候，桌上还有几种电子设备", + "answer_chinese": "桌子上还有三种电子设备，分别是笔记本，鼠标，耳机", + "evidence": [ + { + "description": "Electronic devices", + "timestamps": { + "start": "00:34:42", + "end": "00:34:47" + } + } + ], + "question_timestamps": "00:34:50", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two other devices.", + "B": "Four other devices.", + "C": "There were three other electronic devices on the table: a laptop, a mouse, and headphones.", + "D": "There was only one other electronic device on the table: a mouse, and nothing else besides the tablet.", + "E": "There were no other electronic devices on the table; only the tablet was there when you picked it up." + }, + "answer_label": "C", + "answer_text": "There were three other electronic devices on the table: a laptop, a mouse, and headphones.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_11", + "correct_answer": "There were three other electronic devices on the table: a laptop, a mouse, and headphones.", + "original_options": { + "A": "Two other devices.", + "B": "Four other devices.", + "C": "There were three other electronic devices on the table: a laptop, a mouse, and headphones.", + "D": "There was only one other electronic device on the table: a mouse, and nothing else besides the tablet.", + "E": "There were no other electronic devices on the table; only the tablet was there when you picked it up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 19 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_11_T2", + "qid_num": 30, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I picked up the tablet, how many types of electronic devices were still on the table?", + "answer": "There were also three kinds of electronic devices on the table in front of you: a laptop, a mouse, and headphones.", + "question_chinese": "我在拿平板的时候，桌上还有几种电子设备？", + "answer_chinese": "你面前桌子上还有三种电子设备，分别是笔记本，鼠标，耳机", + "evidence": [ + { + "description": "Electronic devices", + "timestamps": { + "start": "00:34:42", + "end": "00:34:47" + } + } + ], + "question_timestamps": "00:43:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four kinds: including the tablet.", + "B": "Two kinds: laptop and mouse.", + "C": "Only one kind of electronic device was on the table in front of you: just the laptop.", + "D": "Five kinds were still on the table: a laptop, a mouse, headphones, a tablet, and a phone.", + "E": "There were also three kinds of electronic devices on the table in front of you: a laptop, a mouse, and headphones." + }, + "answer_label": "E", + "answer_text": "There were also three kinds of electronic devices on the table in front of you: a laptop, a mouse, and headphones.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_11_T2", + "correct_answer": "There were also three kinds of electronic devices on the table in front of you: a laptop, a mouse, and headphones.", + "original_options": { + "A": "Four kinds: including the tablet.", + "B": "Two kinds: laptop and mouse.", + "C": "Only one kind of electronic device was on the table in front of you: just the laptop.", + "D": "Five kinds were still on the table: a laptop, a mouse, headphones, a tablet, and a phone.", + "E": "There were also three kinds of electronic devices on the table in front of you: a laptop, a mouse, and headphones." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_5_T1", + "qid_num": 31, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I take just now while I was holding the vacuum cleaner?", + "answer": "You took the vacuum cleaner from the storage room, walked out through the storage-room door, and passed by a soccer ball.", + "question_chinese": "我拿着吸尘器刚才经过了什么路径？", + "answer_chinese": "你从杂物间取到了吸尘器，并出杂物间的门，经过一个了足球。", + "evidence": [ + { + "description": "Take the vacuum cleaner out of the utility room.", + "timestamps": { + "start": "00:20:16", + "end": "00:20:38" + } + }, + { + "description": "Walk past the football and through the living room door.", + "timestamps": { + "start": "00:20:40", + "end": "00:20:48" + } + } + ], + "question_timestamps": "00:20:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You stayed in the storage room.", + "B": "You entered the storage room.", + "C": "You took the vacuum from the storage room and walked out, but you did not pass any soccer ball.", + "D": "You took the vacuum cleaner from the storage room, walked out through the storage-room door, and passed by a soccer ball.", + "E": "You took the vacuum from the storage room, walked out the door, and turned toward the kitchen instead." + }, + "answer_label": "D", + "answer_text": "You took the vacuum cleaner from the storage room, walked out through the storage-room door, and passed by a soccer ball.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_5_T1", + "correct_answer": "You took the vacuum cleaner from the storage room, walked out through the storage-room door, and passed by a soccer ball.", + "original_options": { + "A": "You stayed in the storage room.", + "B": "You entered the storage room.", + "C": "You took the vacuum from the storage room and walked out, but you did not pass any soccer ball.", + "D": "You took the vacuum cleaner from the storage room, walked out through the storage-room door, and passed by a soccer ball.", + "E": "You took the vacuum from the storage room, walked out the door, and turned toward the kitchen instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 18 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_5_T2", + "qid_num": 32, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I take just now while I was holding the vacuum cleaner?", + "answer": "You took the vacuum cleaner from the storage room, went out the storage-room door, passed by a soccer ball, then passed through the living-room door and brought it into the living room.", + "question_chinese": "我拿着吸尘器刚才经过了什么路径？", + "answer_chinese": "从杂物间取到了吸尘器，并出杂物间的门，经过一个足球，然后再经过了客厅的门，带入到了客厅", + "evidence": [ + { + "description": "Take the vacuum cleaner out of the utility room.", + "timestamps": { + "start": "00:20:16", + "end": "00:20:38" + } + }, + { + "description": "Crossing the football past the living room door", + "timestamps": { + "start": "00:20:40", + "end": "00:20:48" + } + }, + { + "description": "Put away the vacuum cleaner.", + "timestamps": { + "start": "00:20:49", + "end": "00:21:06" + } + } + ], + "question_timestamps": "00:53:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went straight to the living room.", + "B": "You went back into storage.", + "C": "You took the vacuum out, passed the soccer ball, then turned back before the living-room door.", + "D": "You took the vacuum cleaner from the storage room, went out the storage-room door, passed by a soccer ball, then passed through the living-room door and brought it into the living room.", + "E": "You carried the vacuum out, passed the soccer ball, and stopped in the hallway instead of entering." + }, + "answer_label": "D", + "answer_text": "You took the vacuum cleaner from the storage room, went out the storage-room door, passed by a soccer ball, then passed through the living-room door and brought it into the living room.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_5_T2", + "correct_answer": "You took the vacuum cleaner from the storage room, went out the storage-room door, passed by a soccer ball, then passed through the living-room door and brought it into the living room.", + "original_options": { + "A": "You went straight to the living room.", + "B": "You went back into storage.", + "C": "You took the vacuum out, passed the soccer ball, then turned back before the living-room door.", + "D": "You took the vacuum cleaner from the storage room, went out the storage-room door, passed by a soccer ball, then passed through the living-room door and brought it into the living room.", + "E": "You carried the vacuum out, passed the soccer ball, and stopped in the hallway instead of entering." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 17 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_9_T1", + "qid_num": 33, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far am I from the toilet right now?", + "answer": "The bathroom was behind you, and the toilet was about one or two steps away to your back-left.", + "question_chinese": "我现在距离马桶有多远？", + "answer_chinese": "洗手间在你的身后，马桶就在你的左后方一两步远。", + "evidence": [ + { + "description": "Toilet", + "timestamps": { + "start": "00:22:30", + "end": "00:22:32" + } + } + ], + "question_timestamps": "00:22:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s several steps in front of you.", + "B": "It’s right beside you on the right.", + "C": "The bathroom was behind you, and the toilet was about one or two steps away to your back-left.", + "D": "The bathroom is in front of you, and the toilet is about one or two steps away to your back-left.", + "E": "The bathroom is behind you, but the toilet is about four or five steps away directly behind you." + }, + "answer_label": "C", + "answer_text": "The bathroom was behind you, and the toilet was about one or two steps away to your back-left.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_9_T1", + "correct_answer": "The bathroom was behind you, and the toilet was about one or two steps away to your back-left.", + "original_options": { + "A": "It’s several steps in front of you.", + "B": "It’s right beside you on the right.", + "C": "The bathroom was behind you, and the toilet was about one or two steps away to your back-left.", + "D": "The bathroom is in front of you, and the toilet is about one or two steps away to your back-left.", + "E": "The bathroom is behind you, but the toilet is about four or five steps away directly behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 20, + 18 + ] + }, + { + "video_uid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36", + "video_length": "00:53:25", + "qid": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_9_T2", + "qid_num": 34, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far am I from the toilet right now?", + "answer": "You were in the bathroom, and the toilet was right behind you.", + "question_chinese": "我现在距离马桶有多远？", + "answer_chinese": "你正在洗手间中，马桶就在你的身后。", + "evidence": [ + { + "description": "Toilet", + "timestamps": { + "start": "00:23:27", + "end": "00:23:28" + } + } + ], + "question_timestamps": "00:23:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were in the bathroom, and the toilet was right behind you.", + "B": "It’s right in front of you.", + "C": "It’s right to your left.", + "D": "You were in the bathroom, and the toilet was right to your right when you turned.", + "E": "You were in the bathroom, and the toilet was a few steps ahead of you, not directly behind you." + }, + "answer_label": "A", + "answer_text": "You were in the bathroom, and the toilet was right behind you.", + "q_id": "6caf467a-a949-4b5c-b88b-b2ac6cbd9f36_9_T2", + "correct_answer": "You were in the bathroom, and the toilet was right behind you.", + "original_options": { + "A": "You were in the bathroom, and the toilet was right behind you.", + "B": "It’s right in front of you.", + "C": "It’s right to your left.", + "D": "You were in the bathroom, and the toilet was right to your right when you turned.", + "E": "You were in the bathroom, and the toilet was a few steps ahead of you, not directly behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 19 + ] + } +] diff --git a/QAs/6d0530ba-eb49-4ebb-9feb-2315c8315344-mcq.json b/QAs/6d0530ba-eb49-4ebb-9feb-2315c8315344-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..07f67e854187ef92ca5a7b7fbdbe70476283be1e --- /dev/null +++ b/QAs/6d0530ba-eb49-4ebb-9feb-2315c8315344-mcq.json @@ -0,0 +1,334 @@ +[ + { + "video_uid": "6d0530ba-eb49-4ebb-9feb-2315c8315344", + "video_length": "00:47:07", + "qid": "6d0530ba-eb49-4ebb-9feb-2315c8315344_2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many wooden sticks did I tie to the pipe in total?", + "answer": "You tied a wooden stick onto the pipe altogether.", + "question_chinese": "我在管子上一共绑了几根木棍？", + "answer_chinese": "你在管子上一共绑了一根木棍。", + "evidence": [ + { + "description": "You tied a stick to the pipe.", + "timestamps": { + "start": "00:04:02", + "end": "00:05:40" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You tied zero wooden sticks.", + "B": "You tied a wooden stick onto the pipe altogether.", + "C": "You tied two wooden sticks.", + "D": "You tied three wooden sticks onto the pipe altogether, one after another.", + "E": "You tied four wooden sticks onto the pipe altogether, completing the set before stopping." + }, + "answer_label": "B", + "answer_text": "You tied a wooden stick onto the pipe altogether.", + "q_id": "6d0530ba-eb49-4ebb-9feb-2315c8315344_2", + "correct_answer": "You tied a wooden stick onto the pipe altogether.", + "original_options": { + "A": "You tied zero wooden sticks.", + "B": "You tied a wooden stick onto the pipe altogether.", + "C": "You tied two wooden sticks.", + "D": "You tied three wooden sticks onto the pipe altogether, one after another.", + "E": "You tied four wooden sticks onto the pipe altogether, completing the set before stopping." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 14 + ] + }, + { + "video_uid": "6d0530ba-eb49-4ebb-9feb-2315c8315344", + "video_length": "00:47:07", + "qid": "6d0530ba-eb49-4ebb-9feb-2315c8315344_1_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many wooden sticks did I tie onto the pipe in total?", + "answer": "You tied a total of two wooden sticks onto the pipe, to your left.", + "question_chinese": "我在管子上一共绑了几根木棍？", + "answer_chinese": "你在管子上一共绑了两根木棍，在你的左边。", + "evidence": [ + { + "description": "You tied a stick to the pipe.", + "timestamps": { + "start": "00:04:42", + "end": "00:05:15" + } + }, + { + "description": "You tied a stick to the pipe.", + "timestamps": { + "start": "00:06:35", + "end": "00:07:17" + } + } + ], + "question_timestamps": "00:07:31", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You tied one wooden stick.", + "B": "You tied three wooden sticks.", + "C": "You tied a total of two wooden sticks onto the pipe, to your left.", + "D": "You tied a total of zero wooden sticks onto the pipe to your left.", + "E": "You tied a total of four wooden sticks onto the pipe to your left, adding them one by one." + }, + "answer_label": "C", + "answer_text": "You tied a total of two wooden sticks onto the pipe, to your left.", + "q_id": "6d0530ba-eb49-4ebb-9feb-2315c8315344_1_T1", + "correct_answer": "You tied a total of two wooden sticks onto the pipe, to your left.", + "original_options": { + "A": "You tied one wooden stick.", + "B": "You tied three wooden sticks.", + "C": "You tied a total of two wooden sticks onto the pipe, to your left.", + "D": "You tied a total of zero wooden sticks onto the pipe to your left.", + "E": "You tied a total of four wooden sticks onto the pipe to your left, adding them one by one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 19 + ] + }, + { + "video_uid": "6d0530ba-eb49-4ebb-9feb-2315c8315344", + "video_length": "00:47:07", + "qid": "6d0530ba-eb49-4ebb-9feb-2315c8315344_3", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How can I get the blue bucket?", + "answer": "Walk to the left and you will see the blue bucket.", + "question_chinese": "我该怎么拿到蓝色的桶？", + "answer_chinese": "向左走，你会看见蓝色的桶。", + "evidence": [ + { + "description": "You walked from the blue bucket to here.", + "timestamps": { + "start": "00:03:53", + "end": "00:04:03" + } + } + ], + "question_timestamps": "00:04:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk to the right.", + "B": "Walk straight ahead.", + "C": "Walk a little to the right, and you will see the blue bucket near the same area.", + "D": "Go forward first and then turn right; you will see the blue bucket from that direction.", + "E": "Walk to the left and you will see the blue bucket." + }, + "answer_label": "E", + "answer_text": "Walk to the left and you will see the blue bucket.", + "q_id": "6d0530ba-eb49-4ebb-9feb-2315c8315344_3", + "correct_answer": "Walk to the left and you will see the blue bucket.", + "original_options": { + "A": "Walk to the right.", + "B": "Walk straight ahead.", + "C": "Walk a little to the right, and you will see the blue bucket near the same area.", + "D": "Go forward first and then turn right; you will see the blue bucket from that direction.", + "E": "Walk to the left and you will see the blue bucket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 16 + ] + }, + { + "video_uid": "6d0530ba-eb49-4ebb-9feb-2315c8315344", + "video_length": "00:47:07", + "qid": "6d0530ba-eb49-4ebb-9feb-2315c8315344_4", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How can I get the blue bucket?", + "answer": "Walk along the pipe and you will see the blue barrels.", + "question_chinese": "我该怎么拿到蓝色的桶？", + "answer_chinese": "沿着管子一直走，你能看见蓝色的桶。", + "evidence": [ + { + "description": "You put down the blue bucket and then walked along the pipe to get here.", + "timestamps": { + "start": "00:39:49", + "end": "00:42:57" + } + } + ], + "question_timestamps": "00:42:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk around the corner.", + "B": "Walk along the pipe and you will see the blue barrels.", + "C": "Walk through the garden.", + "D": "Walk along the fence line and you will see the blue barrels near the far side.", + "E": "Follow the path past the pipe, then turn away from it to find the blue barrels." + }, + "answer_label": "B", + "answer_text": "Walk along the pipe and you will see the blue barrels.", + "q_id": "6d0530ba-eb49-4ebb-9feb-2315c8315344_4", + "correct_answer": "Walk along the pipe and you will see the blue barrels.", + "original_options": { + "A": "Walk around the corner.", + "B": "Walk along the pipe and you will see the blue barrels.", + "C": "Walk through the garden.", + "D": "Walk along the fence line and you will see the blue barrels near the far side.", + "E": "Follow the path past the pipe, then turn away from it to find the blue barrels." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "6d0530ba-eb49-4ebb-9feb-2315c8315344", + "video_length": "00:47:07", + "qid": "6d0530ba-eb49-4ebb-9feb-2315c8315344_7", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Which paths did the blue bucket take to arrive here?", + "answer": "The blue bucket was carried by you from one end of this field to here.", + "question_chinese": "蓝色的桶经过了哪些路径到达这里?", + "answer_chinese": "蓝色的桶从这块地的一头被你拿到了这里.", + "evidence": [ + { + "description": "The blue bucket was carried by you from one end of this field to here.", + "timestamps": { + "start": "00:04:20", + "end": "00:15:55" + } + } + ], + "question_timestamps": "00:15:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The blue bucket was carried by you from one end of this field to here.", + "B": "You carried it from mid-field.", + "C": "Someone else carried it here.", + "D": "The blue bucket was carried by you from the near edge of this field to here.", + "E": "The blue bucket was carried by you from one end, but you looped back before arriving here." + }, + "answer_label": "A", + "answer_text": "The blue bucket was carried by you from one end of this field to here.", + "q_id": "6d0530ba-eb49-4ebb-9feb-2315c8315344_7", + "correct_answer": "The blue bucket was carried by you from one end of this field to here.", + "original_options": { + "A": "The blue bucket was carried by you from one end of this field to here.", + "B": "You carried it from mid-field.", + "C": "Someone else carried it here.", + "D": "The blue bucket was carried by you from the near edge of this field to here.", + "E": "The blue bucket was carried by you from one end, but you looped back before arriving here." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "6d0530ba-eb49-4ebb-9feb-2315c8315344", + "video_length": "00:47:07", + "qid": "6d0530ba-eb49-4ebb-9feb-2315c8315344_8", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Which paths did the blue bucket take to arrive here?", + "answer": "This bucket was carried from one end of the field to the other end. After passing through one field, it was placed here.", + "question_chinese": "蓝色的桶经过了哪些路径到达这里?", + "answer_chinese": "这个桶从地的那头被拿到了这头。经过了一块地之后放到了这里。", + "evidence": [ + { + "description": "After passing through one field, it was placed here.", + "timestamps": { + "start": "00:39:34", + "end": "00:39:50" + } + }, + { + "description": "The blue bucket was carried by you from one end of this field to here.", + "timestamps": { + "start": "00:04:20", + "end": "00:39:33" + } + } + ], + "question_timestamps": "00:39:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was carried across just this field.", + "B": "It was moved only within the same field.", + "C": "You carried the bucket to the far end of the field, then brought it back and set it down here.", + "D": "This bucket was carried from one end of the field to the other end. After passing through one field, it was placed here.", + "E": "It was carried from the edge of the field to the middle, then placed here without passing through another field." + }, + "answer_label": "D", + "answer_text": "This bucket was carried from one end of the field to the other end. After passing through one field, it was placed here.", + "q_id": "6d0530ba-eb49-4ebb-9feb-2315c8315344_8", + "correct_answer": "This bucket was carried from one end of the field to the other end. After passing through one field, it was placed here.", + "original_options": { + "A": "It was carried across just this field.", + "B": "It was moved only within the same field.", + "C": "You carried the bucket to the far end of the field, then brought it back and set it down here.", + "D": "This bucket was carried from one end of the field to the other end. After passing through one field, it was placed here.", + "E": "It was carried from the edge of the field to the middle, then placed here without passing through another field." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 20, + 20 + ] + } +] diff --git a/QAs/6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3-mcq.json b/QAs/6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fe02815545f190f79c3028993a09df36de383caa --- /dev/null +++ b/QAs/6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3-mcq.json @@ -0,0 +1,725 @@ +[ + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have you measured the steel bar with a ruler?", + "answer": "You measured with a ruler six times.", + "question_chinese": "你用尺子量了钢条多少次？", + "answer_chinese": "你用尺子量了六次。", + "evidence": [ + { + "description": "Measured with a ruler five times.", + "timestamps": { + "start": "00:06:32", + "end": "00:07:42" + } + }, + { + "description": "Measured with a ruler one time.", + "timestamps": { + "start": "00:07:56", + "end": "00:08:04" + } + } + ], + "question_timestamps": "00:08:44", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You measured with a ruler four times.", + "B": "You measured with a ruler five times.", + "C": "You measured with a ruler seven times while checking the steel bar’s length at several different points.", + "D": "You measured with a ruler eight times, repeating the measurement after repositioning the ruler along the bar.", + "E": "You measured with a ruler six times." + }, + "answer_label": "E", + "answer_text": "You measured with a ruler six times.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_3", + "correct_answer": "You measured with a ruler six times.", + "original_options": { + "A": "You measured with a ruler four times.", + "B": "You measured with a ruler five times.", + "C": "You measured with a ruler seven times while checking the steel bar’s length at several different points.", + "D": "You measured with a ruler eight times, repeating the measurement after repositioning the ruler along the bar.", + "E": "You measured with a ruler six times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 17 + ] + }, + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have you measured the steel bar with a ruler?", + "answer": "You measured it seven times with a ruler.", + "question_chinese": "你用尺子量了钢条多少次？", + "answer_chinese": "你用尺子量了七次。", + "evidence": [ + { + "description": "Measured with a ruler five times.", + "timestamps": { + "start": "00:06:32", + "end": "00:07:42" + } + }, + { + "description": "Measured with a ruler one time.", + "timestamps": { + "start": "00:07:56", + "end": "00:08:04" + } + }, + { + "description": "Measured with a ruler one time.", + "timestamps": { + "start": "00:09:24", + "end": "00:09:58" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You measured it seven times with a ruler.", + "B": "You measured it five times.", + "C": "You measured it eight times.", + "D": "You measured it six times with a ruler, stopping once between checks to adjust the bar’s position.", + "E": "You measured it nine times with a ruler, repeating the process after sliding the ruler along the steel bar." + }, + "answer_label": "A", + "answer_text": "You measured it seven times with a ruler.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_4", + "correct_answer": "You measured it seven times with a ruler.", + "original_options": { + "A": "You measured it seven times with a ruler.", + "B": "You measured it five times.", + "C": "You measured it eight times.", + "D": "You measured it six times with a ruler, stopping once between checks to adjust the bar’s position.", + "E": "You measured it nine times with a ruler, repeating the process after sliding the ruler along the steel bar." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 19 + ] + }, + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_7", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I move to get the angle grinder?", + "answer": "You can turn around, walk back, take four steps, turn right, and reach to the top of the cabinet.", + "question_chinese": "我怎么样移动去拿到角磨机?", + "answer_chinese": "你可以转身向后，走四步再向右转，伸手到柜子的顶部。", + "evidence": [ + { + "description": "The angle grinder is on the low cabinet behind you.", + "timestamps": { + "start": "00:25:15", + "end": "00:25:23" + } + } + ], + "question_timestamps": "00:25:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn around, walk back, then turn left.", + "B": "Turn around, walk forward four steps, right.", + "C": "You can turn around, walk back, take four steps, turn right, and reach to the top of the cabinet.", + "D": "You can turn around, walk back, take three steps, turn right, and reach to the top of the cabinet.", + "E": "You can turn around, walk back, take four steps, turn left, and reach up to the top of the cabinet." + }, + "answer_label": "C", + "answer_text": "You can turn around, walk back, take four steps, turn right, and reach to the top of the cabinet.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_7", + "correct_answer": "You can turn around, walk back, take four steps, turn right, and reach to the top of the cabinet.", + "original_options": { + "A": "Turn around, walk back, then turn left.", + "B": "Turn around, walk forward four steps, right.", + "C": "You can turn around, walk back, take four steps, turn right, and reach to the top of the cabinet.", + "D": "You can turn around, walk back, take three steps, turn right, and reach to the top of the cabinet.", + "E": "You can turn around, walk back, take four steps, turn left, and reach up to the top of the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 20 + ] + }, + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_8", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I move to get the angle grinder?", + "answer": "You should turn left, then walk two steps straight, then turn right and reach over the work cabinet.", + "question_chinese": "我怎么样移动去拿到角磨机?", + "answer_chinese": "你应该左转，然后直走两步，再右转，伸手到工作柜的上方。", + "evidence": [ + { + "description": "The angle grinder is on the work cabinet.", + "timestamps": { + "start": "00:26:41", + "end": "00:27:24" + } + } + ], + "question_timestamps": "00:27:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should turn left, then walk two steps straight, then turn right and reach over the work cabinet.", + "B": "Turn right, walk two steps, then reach.", + "C": "Turn left, walk three steps, then right.", + "D": "You should turn left, then walk two steps straight, then turn left and reach over the work cabinet.", + "E": "You should turn left, then walk two steps straight, then turn right and reach under the work cabinet." + }, + "answer_label": "A", + "answer_text": "You should turn left, then walk two steps straight, then turn right and reach over the work cabinet.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_8", + "correct_answer": "You should turn left, then walk two steps straight, then turn right and reach over the work cabinet.", + "original_options": { + "A": "You should turn left, then walk two steps straight, then turn right and reach over the work cabinet.", + "B": "Turn right, walk two steps, then reach.", + "C": "Turn left, walk three steps, then right.", + "D": "You should turn left, then walk two steps straight, then turn left and reach over the work cabinet.", + "E": "You should turn left, then walk two steps straight, then turn right and reach under the work cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_11", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How do I move to turn off the welding tool?", + "answer": "Turn right to the gas cylinder in the corner.", + "question_chinese": "我怎么移动去关闭电焊工具？", + "answer_chinese": "向右转身到角落的气瓶处。", + "evidence": [ + { + "description": "You turned on the gas cylinder and welding tool in the corner.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:58" + } + }, + { + "description": "You started the welding tool.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:23" + } + } + ], + "question_timestamps": "00:02:24", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right to the gas cylinder in the corner.", + "B": "Turn left to the shelf.", + "C": "Go straight to the table.", + "D": "Turn back toward the outlet area and move to the nearby equipment to shut it off.", + "E": "Turn right, then head toward the window side and stop at the machine next to it." + }, + "answer_label": "A", + "answer_text": "Turn right to the gas cylinder in the corner.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_11", + "correct_answer": "Turn right to the gas cylinder in the corner.", + "original_options": { + "A": "Turn right to the gas cylinder in the corner.", + "B": "Turn left to the shelf.", + "C": "Go straight to the table.", + "D": "Turn back toward the outlet area and move to the nearby equipment to shut it off.", + "E": "Turn right, then head toward the window side and stop at the machine next to it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_12", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How do I move to turn off the welding tool?", + "answer": "You should turn right and go straight to the corner.", + "question_chinese": "我怎么移动去关闭电焊工具？", + "answer_chinese": "你应该右转直走到角落处。", + "evidence": [ + { + "description": "You turned on the gas cylinder and welding tool in the corner.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:58" + } + }, + { + "description": "You started the welding tool.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:23" + } + } + ], + "question_timestamps": "00:27:21", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left and go to the corner.", + "B": "Turn right and go to the steps.", + "C": "You should turn right and go straight to the platform instead of continuing all the way into the corner.", + "D": "You should turn right and go straight to the corner.", + "E": "You should turn left, go straight ahead, and stop at the equipment rather than the corner." + }, + "answer_label": "D", + "answer_text": "You should turn right and go straight to the corner.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_12", + "correct_answer": "You should turn right and go straight to the corner.", + "original_options": { + "A": "Turn left and go to the corner.", + "B": "Turn right and go to the steps.", + "C": "You should turn right and go straight to the platform instead of continuing all the way into the corner.", + "D": "You should turn right and go straight to the corner.", + "E": "You should turn left, go straight ahead, and stop at the equipment rather than the corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 16 + ] + }, + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I welded?", + "answer": "You welded nine times.", + "question_chinese": "我焊接了多少次呢？", + "answer_chinese": "你焊接了九次。", + "evidence": [ + { + "description": "First welding.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:21" + } + }, + { + "description": "Welded three times.", + "timestamps": { + "start": "00:02:23", + "end": "00:03:12" + } + }, + { + "description": "Welded five times.", + "timestamps": { + "start": "00:03:35", + "end": "00:04:33" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You welded seven times.", + "B": "You welded nine times.", + "C": "You welded eleven times.", + "D": "You welded eight times, completing several separate welds before stopping and repositioning the piece.", + "E": "You welded fourteen times, repeating the weld after moving along the seam in multiple short passes." + }, + "answer_label": "B", + "answer_text": "You welded nine times.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_1", + "correct_answer": "You welded nine times.", + "original_options": { + "A": "You welded seven times.", + "B": "You welded nine times.", + "C": "You welded eleven times.", + "D": "You welded eight times, completing several separate welds before stopping and repositioning the piece.", + "E": "You welded fourteen times, repeating the weld after moving along the seam in multiple short passes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 16 + ] + }, + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I welded?", + "answer": "You welded thirty-four times.", + "question_chinese": "我焊接了多少次呢？", + "answer_chinese": "你焊接了三十四次。", + "evidence": [ + { + "description": "First welding.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:21" + } + }, + { + "description": "Welded three times.", + "timestamps": { + "start": "00:02:23", + "end": "00:03:12" + } + }, + { + "description": "Welded five times.", + "timestamps": { + "start": "00:03:35", + "end": "00:04:33" + } + }, + { + "description": "Welded one time.", + "timestamps": { + "start": "00:10:13", + "end": "00:10:21" + } + }, + { + "description": "Welded one time.", + "timestamps": { + "start": "00:11:43", + "end": "00:11:47" + } + }, + { + "description": "Welded five times.", + "timestamps": { + "start": "00:13:26", + "end": "00:15:05" + } + }, + { + "description": "Welded four times.", + "timestamps": { + "start": "00:15:59", + "end": "00:17:04" + } + }, + { + "description": "Welded four times.", + "timestamps": { + "start": "00:17:15", + "end": "00:18:54" + } + }, + { + "description": "Welded ten times.", + "timestamps": { + "start": "00:19:58", + "end": "00:23:13" + } + } + ], + "question_timestamps": "00:23:17", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You welded thirty-three times.", + "B": "You welded thirty-five times.", + "C": "You welded thirty-four times.", + "D": "You welded thirty-two times in total, counting each separate bead you laid down during the task.", + "E": "You welded thirty-six times overall, including the brief touch-up welds you made near the end." + }, + "answer_label": "C", + "answer_text": "You welded thirty-four times.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_2", + "correct_answer": "You welded thirty-four times.", + "original_options": { + "A": "You welded thirty-three times.", + "B": "You welded thirty-five times.", + "C": "You welded thirty-four times.", + "D": "You welded thirty-two times in total, counting each separate bead you laid down during the task.", + "E": "You welded thirty-six times overall, including the brief touch-up welds you made near the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 15 + ] + }, + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_13", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I reach out and touch the set square?", + "answer": "You can touch the set square.", + "question_chinese": "我能伸手摸到三角尺吗？", + "answer_chinese": "你可以摸到三角尺。", + "evidence": [ + { + "description": "The set square is placed on the table top.", + "timestamps": { + "start": "00:08:33", + "end": "00:08:43" + } + } + ], + "question_timestamps": "00:09:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't touch the set square.", + "B": "You can touch the set square." + }, + "answer_label": "B", + "answer_text": "You can touch the set square.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_13", + "correct_answer": "You can touch the set square.", + "original_options": { + "A": "You can't touch the set square.", + "B": "You can touch the set square." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_14", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I reach out and touch the set square?", + "answer": "You can't touch the set square.", + "question_chinese": "我能伸手摸到三角尺吗？", + "answer_chinese": "你不能摸到三角尺。", + "evidence": [ + { + "description": "The set square is placed on the table top.", + "timestamps": { + "start": "00:08:33", + "end": "00:08:43" + } + } + ], + "question_timestamps": "00:26:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't touch the set square.", + "B": "You can touch the set square." + }, + "answer_label": "A", + "answer_text": "You can't touch the set square.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_14", + "correct_answer": "You can't touch the set square.", + "original_options": { + "A": "You can't touch the set square.", + "B": "You can touch the set square." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_9", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many workpieces are welded on the workbench?", + "answer": "There are two steel bars on the table top.", + "question_chinese": "桌面上有多少根钢条？", + "answer_chinese": "桌面上有两根钢条。", + "evidence": [ + { + "description": "There are two steel bars on the table top.", + "timestamps": { + "start": "00:01:42", + "end": "00:02:03" + } + } + ], + "question_timestamps": "00:02:24", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One steel bar.", + "B": "Three steel bars.", + "C": "Four steel bars are on the table top, welded and lined up next to each other on the workbench.", + "D": "There are two steel bars on the table top.", + "E": "Five steel bars are on the table top, with multiple pieces grouped together in the same work area." + }, + "answer_label": "D", + "answer_text": "There are two steel bars on the table top.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_9", + "correct_answer": "There are two steel bars on the table top.", + "original_options": { + "A": "One steel bar.", + "B": "Three steel bars.", + "C": "Four steel bars are on the table top, welded and lined up next to each other on the workbench.", + "D": "There are two steel bars on the table top.", + "E": "Five steel bars are on the table top, with multiple pieces grouped together in the same work area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 19, + 18 + ] + }, + { + "video_uid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3", + "video_length": "00:28:19", + "qid": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_10", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many workpieces are welded on the workbench?", + "answer": "There are four steel bars on the table.", + "question_chinese": "桌面上有多少根钢条？", + "answer_chinese": "桌面上有四根钢条。", + "evidence": [ + { + "description": "There are four steel bars on the table top.", + "timestamps": { + "start": "00:06:18", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:06:52", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three steel bars on the table.", + "B": "There are five steel bars on the table.", + "C": "There are two steel bars on the table, positioned as a small welded set on the work surface.", + "D": "There are four steel bars on the table.", + "E": "There are six steel bars on the table, making a larger welded group spread across the tabletop." + }, + "answer_label": "D", + "answer_text": "There are four steel bars on the table.", + "q_id": "6d0b8b1c-f6d5-4abc-baad-a3fee8466ae3_10", + "correct_answer": "There are four steel bars on the table.", + "original_options": { + "A": "There are three steel bars on the table.", + "B": "There are five steel bars on the table.", + "C": "There are two steel bars on the table, positioned as a small welded set on the work surface.", + "D": "There are four steel bars on the table.", + "E": "There are six steel bars on the table, making a larger welded group spread across the tabletop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 17 + ] + } +] diff --git a/QAs/6f7f3410-1e20-4eb2-b7b0-088c1500054b-mcq.json b/QAs/6f7f3410-1e20-4eb2-b7b0-088c1500054b-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4d4d42b5db458786a008c259ee0714a67e008b88 --- /dev/null +++ b/QAs/6f7f3410-1e20-4eb2-b7b0-088c1500054b-mcq.json @@ -0,0 +1,528 @@ +[ + { + "video_uid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b", + "video_length": "00:30:00", + "qid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of paper are posted on the wall?", + "answer": "six.", + "question_chinese": "墙上贴了多少张纸？", + "answer_chinese": "6", + "evidence": [ + { + "description": "There are six paper", + "timestamps": { + "start": "00:06:07", + "end": "00:06:08" + } + } + ], + "question_timestamps": "00:06:09", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "Seven pieces of paper are posted on the wall, spread across the same section above the work area.", + "D": "Eight pieces of paper are posted on the wall, with several sheets clustered together in one row.", + "E": "six." + }, + "answer_label": "E", + "answer_text": "six.", + "q_id": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_3", + "correct_answer": "six.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "Seven pieces of paper are posted on the wall, spread across the same section above the work area.", + "D": "Eight pieces of paper are posted on the wall, with several sheets clustered together in one row.", + "E": "six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 17 + ] + }, + { + "video_uid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b", + "video_length": "00:30:00", + "qid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of paper are posted on the wall?", + "answer": "six.", + "question_chinese": "墙上贴了多少张纸？", + "answer_chinese": "6", + "evidence": [ + { + "description": "There were six paper", + "timestamps": { + "start": "00:06:07", + "end": "00:06:08" + } + } + ], + "question_timestamps": "00:06:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "six.", + "C": "Eight.", + "D": "Four pieces of paper are posted on the wall, arranged together on the same visible wall section.", + "E": "Ten pieces of paper are posted on the wall, covering a larger portion of the wall space." + }, + "answer_label": "B", + "answer_text": "six.", + "q_id": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_4", + "correct_answer": "six.", + "original_options": { + "A": "Five.", + "B": "six.", + "C": "Eight.", + "D": "Four pieces of paper are posted on the wall, arranged together on the same visible wall section.", + "E": "Ten pieces of paper are posted on the wall, covering a larger portion of the wall space." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 17 + ] + }, + { + "video_uid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b", + "video_length": "00:30:00", + "qid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_9", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns have I made?", + "answer": "Two.", + "question_chinese": "我拐了多少个弯？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:12:51", + "end": "00:12:52" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:15:10", + "end": "00:15:11" + } + } + ], + "question_timestamps": "00:15:12", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four turns in total, counting each time you changed direction while moving through the area.", + "E": "Five turns altogether, including the smaller direction changes you made while adjusting your path." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_9", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four turns in total, counting each time you changed direction while moving through the area.", + "E": "Five turns altogether, including the smaller direction changes you made while adjusting your path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b", + "video_length": "00:30:00", + "qid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_10", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns have I made?", + "answer": "two.", + "question_chinese": "我拐了多少个弯？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:12:51", + "end": "00:12:52" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:15:10", + "end": "00:15:11" + } + } + ], + "question_timestamps": "00:15:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "one.", + "B": "two.", + "C": "three.", + "D": "You only made one turn during the route, changing direction a single time before continuing straight.", + "E": "You made three turns overall, with an extra direction change after the second corner before reaching the end." + }, + "answer_label": "B", + "answer_text": "two.", + "q_id": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_10", + "correct_answer": "two.", + "original_options": { + "A": "one.", + "B": "two.", + "C": "three.", + "D": "You only made one turn during the route, changing direction a single time before continuing straight.", + "E": "You made three turns overall, with an extra direction change after the second corner before reaching the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 18 + ] + }, + { + "video_uid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b", + "video_length": "00:30:00", + "qid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_9_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mailboxes did I pass by?", + "answer": "You passed two mailboxes. When you first saw the first mailbox, it was in front of you; when you first saw the second mailbox, it was on your right-hand side.", + "question_chinese": "我经过了多少个信箱？", + "answer_chinese": "你经过了两个信箱。第一个信箱你看见时在你的正前方，第二个信箱你看见时在你的右手边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:08:14", + "end": "00:08:15" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:12:41", + "end": "00:12:42" + } + } + ], + "question_timestamps": "00:13:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You passed one mailbox.", + "B": "You passed three mailboxes.", + "C": "You passed one mailbox, and when you first noticed it, it was directly in front of you.", + "D": "You passed three mailboxes: the first was ahead, the second appeared on your right, and the third was ahead again.", + "E": "You passed two mailboxes. When you first saw the first mailbox, it was in front of you; when you first saw the second mailbox, it was on your right-hand side." + }, + "answer_label": "E", + "answer_text": "You passed two mailboxes. When you first saw the first mailbox, it was in front of you; when you first saw the second mailbox, it was on your right-hand side.", + "q_id": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_9_T1", + "correct_answer": "You passed two mailboxes. When you first saw the first mailbox, it was in front of you; when you first saw the second mailbox, it was on your right-hand side.", + "original_options": { + "A": "You passed one mailbox.", + "B": "You passed three mailboxes.", + "C": "You passed one mailbox, and when you first noticed it, it was directly in front of you.", + "D": "You passed three mailboxes: the first was ahead, the second appeared on your right, and the third was ahead again.", + "E": "You passed two mailboxes. When you first saw the first mailbox, it was in front of you; when you first saw the second mailbox, it was on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 20 + ] + }, + { + "video_uid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b", + "video_length": "00:30:00", + "qid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_15", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mailboxes have I passed?", + "answer": "three.", + "question_chinese": "我经过了多少个信箱？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:08:14", + "end": "00:08:15" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:12:41", + "end": "00:12:42" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:13:56", + "end": "00:13:57" + } + } + ], + "question_timestamps": "00:13:58", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "two.", + "B": "four.", + "C": "You passed two mailboxes in total along the way, not three as you might think from the route.", + "D": "You passed four mailboxes altogether, encountering an additional one after the third before you moved on.", + "E": "three." + }, + "answer_label": "E", + "answer_text": "three.", + "q_id": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_15", + "correct_answer": "three.", + "original_options": { + "A": "two.", + "B": "four.", + "C": "You passed two mailboxes in total along the way, not three as you might think from the route.", + "D": "You passed four mailboxes altogether, encountering an additional one after the third before you moved on.", + "E": "three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 16 + ] + }, + { + "video_uid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b", + "video_length": "00:30:00", + "qid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_9_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mailboxes did I pass by?", + "answer": "You passed three mailboxes. When you first saw the first mailbox, it was in front of you; when you first saw the second mailbox, it was on your right-hand side; when you first saw the third mailbox, it was also on your right-hand side.", + "question_chinese": "我经过了多少个信箱？", + "answer_chinese": "你经过了三个信箱。第一个信箱你看见时在你的正前方，第二个信箱你看见时在你的右手边，第三个信箱你看见时在你的右手边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:08:14", + "end": "00:08:15" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:12:41", + "end": "00:12:42" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:13:56", + "end": "00:13:57" + } + } + ], + "question_timestamps": "00:14:58", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You passed three mailboxes. When you first saw the first mailbox, it was in front of you; when you first saw the second mailbox, it was on your right-hand side; when you first saw the third mailbox, it was also on your right-hand side.", + "B": "You passed two mailboxes.", + "C": "You passed four mailboxes.", + "D": "You passed two mailboxes: the first was in front of you, and the second was on your right-hand side.", + "E": "You passed four mailboxes: first ahead, then two on your right, and a final one on your left-hand side." + }, + "answer_label": "A", + "answer_text": "You passed three mailboxes. When you first saw the first mailbox, it was in front of you; when you first saw the second mailbox, it was on your right-hand side; when you first saw the third mailbox, it was also on your right-hand side.", + "q_id": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_9_T2", + "correct_answer": "You passed three mailboxes. When you first saw the first mailbox, it was in front of you; when you first saw the second mailbox, it was on your right-hand side; when you first saw the third mailbox, it was also on your right-hand side.", + "original_options": { + "A": "You passed three mailboxes. When you first saw the first mailbox, it was in front of you; when you first saw the second mailbox, it was on your right-hand side; when you first saw the third mailbox, it was also on your right-hand side.", + "B": "You passed two mailboxes.", + "C": "You passed four mailboxes.", + "D": "You passed two mailboxes: the first was in front of you, and the second was on your right-hand side.", + "E": "You passed four mailboxes: first ahead, then two on your right, and a final one on your left-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 19 + ] + }, + { + "video_uid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b", + "video_length": "00:30:00", + "qid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red backpack?", + "answer": "On the sofa.", + "question_chinese": "红色书包在哪里？", + "answer_chinese": "在沙发上", + "evidence": [ + { + "description": "On the sofa", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the chair.", + "B": "Under the table.", + "C": "On the sofa.", + "D": "It’s on the floor near the doorway, close to where you entered the room earlier.", + "E": "It’s by the window, resting next to the wall and slightly away from the sofa." + }, + "answer_label": "C", + "answer_text": "On the sofa.", + "q_id": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_1", + "correct_answer": "On the sofa.", + "original_options": { + "A": "On the chair.", + "B": "Under the table.", + "C": "On the sofa.", + "D": "It’s on the floor near the doorway, close to where you entered the room earlier.", + "E": "It’s by the window, resting next to the wall and slightly away from the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b", + "video_length": "00:30:00", + "qid": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red backpack?", + "answer": "On the sofa.", + "question_chinese": "红色书包在哪里？", + "answer_chinese": "在沙发上", + "evidence": [ + { + "description": "On the sofa", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the sofa.", + "B": "On the floor.", + "C": "Under the table.", + "D": "It’s behind the door, out of your direct view when you stand near the sofa.", + "E": "It’s in the closet, tucked away on a shelf rather than left in the open." + }, + "answer_label": "A", + "answer_text": "On the sofa.", + "q_id": "6f7f3410-1e20-4eb2-b7b0-088c1500054b_2", + "correct_answer": "On the sofa.", + "original_options": { + "A": "On the sofa.", + "B": "On the floor.", + "C": "Under the table.", + "D": "It’s behind the door, out of your direct view when you stand near the sofa.", + "E": "It’s in the closet, tucked away on a shelf rather than left in the open." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + } +] diff --git a/QAs/6fd90f8d-7a4d-425d-a812-3268db0b0342-mcq.json b/QAs/6fd90f8d-7a4d-425d-a812-3268db0b0342-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a2a370ee172231623de4e972af6fe91f9bb8f646 --- /dev/null +++ b/QAs/6fd90f8d-7a4d-425d-a812-3268db0b0342-mcq.json @@ -0,0 +1,1008 @@ +[ + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_7_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "While I was washing the teacup and the spoon, where was the pot relative to me?", + "answer": "The pot is to your back-right.", + "question_chinese": "在我洗茶杯和勺子的时候，锅在相对于我的什么方向的位置？", + "answer_chinese": "锅在相对于你的右后方", + "evidence": [ + { + "description": "put the pot", + "timestamps": { + "start": "00:32:09", + "end": "00:32:15" + } + } + ], + "question_timestamps": "00:33:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The pot is to your back-right.", + "B": "The pot is to your back-left.", + "C": "The pot is to your front-right.", + "D": "The pot is to your front-left, positioned ahead of you on the counter while you wash.", + "E": "The pot is directly behind you, centered in your rear view as you face the sink." + }, + "answer_label": "A", + "answer_text": "The pot is to your back-right.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_7_T2", + "correct_answer": "The pot is to your back-right.", + "original_options": { + "A": "The pot is to your back-right.", + "B": "The pot is to your back-left.", + "C": "The pot is to your front-right.", + "D": "The pot is to your front-left, positioned ahead of you on the counter while you wash.", + "E": "The pot is directly behind you, centered in your rear view as you face the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_7_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "While I was washing the teacup and the spoon, where was the pot relative to me?", + "answer": "The pot is now to your back-left.", + "question_chinese": "在我洗茶杯和勺子的时候，锅在相对于我的什么方向的位置？", + "answer_chinese": "锅现在相对于你的左后方。", + "evidence": [ + { + "description": "put the pot", + "timestamps": { + "start": "00:32:09", + "end": "00:32:15" + } + } + ], + "question_timestamps": "00:38:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The pot is now to your back-right.", + "B": "The pot is now to your front-left.", + "C": "The pot is now to your front-right, sitting ahead of you as you remain facing the sink.", + "D": "The pot is now directly behind you, aligned with your body rather than off to one side.", + "E": "The pot is now to your back-left." + }, + "answer_label": "E", + "answer_text": "The pot is now to your back-left.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_7_T1", + "correct_answer": "The pot is now to your back-left.", + "original_options": { + "A": "The pot is now to your back-right.", + "B": "The pot is now to your front-left.", + "C": "The pot is now to your front-right, sitting ahead of you as you remain facing the sink.", + "D": "The pot is now directly behind you, aligned with your body rather than off to one side.", + "E": "The pot is now to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_4_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces did I just cut the bread into for breakfast?", + "answer": "You just cut the bread into three pieces for breakfast, and the breakfast is now to your right.", + "question_chinese": "我刚刚把面包切成几块做早餐？", + "answer_chinese": "你刚刚把面包切成三块做早餐，现在早餐在你的右方。", + "evidence": [ + { + "description": "cut the bread into three pieces", + "timestamps": { + "start": "00:09:48", + "end": "00:10:07" + } + }, + { + "description": "put bread on the plate", + "timestamps": { + "start": "00:10:08", + "end": "00:10:10" + } + } + ], + "question_timestamps": "00:11:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two pieces.", + "B": "You just cut the bread into three pieces for breakfast, and the breakfast is now to your right.", + "C": "Four pieces.", + "D": "You cut the bread into two pieces for breakfast, and the breakfast is now to your right.", + "E": "You cut the bread into five pieces for breakfast, and the breakfast is now to your right." + }, + "answer_label": "B", + "answer_text": "You just cut the bread into three pieces for breakfast, and the breakfast is now to your right.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_4_T1", + "correct_answer": "You just cut the bread into three pieces for breakfast, and the breakfast is now to your right.", + "original_options": { + "A": "Two pieces.", + "B": "You just cut the bread into three pieces for breakfast, and the breakfast is now to your right.", + "C": "Four pieces.", + "D": "You cut the bread into two pieces for breakfast, and the breakfast is now to your right.", + "E": "You cut the bread into five pieces for breakfast, and the breakfast is now to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_4_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces did I just cut the bread into for breakfast?", + "answer": "You just cut the bread into three pieces for breakfast, and it’s now in front of you.", + "question_chinese": "我刚刚把面包切成几块做早餐？", + "answer_chinese": "你刚刚把面包切成三块做早餐，在你的前方。", + "evidence": [ + { + "description": "cut the bread into three pieces", + "timestamps": { + "start": "00:09:48", + "end": "00:10:07" + } + }, + { + "description": "put bread on the plate", + "timestamps": { + "start": "00:10:08", + "end": "00:10:10" + } + } + ], + "question_timestamps": "00:21:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two pieces.", + "B": "Five pieces.", + "C": "You just cut the bread into three pieces for breakfast, and it’s now in front of you.", + "D": "You cut the bread into four pieces for breakfast, and it’s now in front of you.", + "E": "You cut the bread into one piece for breakfast, and it’s now in front of you." + }, + "answer_label": "C", + "answer_text": "You just cut the bread into three pieces for breakfast, and it’s now in front of you.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_4_T2", + "correct_answer": "You just cut the bread into three pieces for breakfast, and it’s now in front of you.", + "original_options": { + "A": "Two pieces.", + "B": "Five pieces.", + "C": "You just cut the bread into three pieces for breakfast, and it’s now in front of you.", + "D": "You cut the bread into four pieces for breakfast, and it’s now in front of you.", + "E": "You cut the bread into one piece for breakfast, and it’s now in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_8_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in the living room did I put the puzzle I just sorted out?", + "answer": "The puzzle you just organized is in the compartment below the living-room TV.", + "question_chinese": "我刚刚整理的拼图放在客厅的哪里了？", + "answer_chinese": "你刚刚整理的拼图放在客厅电视机下方的柜格中", + "evidence": [ + { + "description": "put the puzzles", + "timestamps": { + "start": "00:39:31", + "end": "00:40:30" + } + } + ], + "question_timestamps": "00:40:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The puzzle you just organized is in the compartment below the living-room TV.", + "B": "In the compartment above the TV.", + "C": "On top of the TV stand.", + "D": "The puzzle you just organized is in the compartment next to the living-room TV, not underneath it.", + "E": "The puzzle you just organized is tucked into a side compartment of the TV stand, rather than the one below." + }, + "answer_label": "A", + "answer_text": "The puzzle you just organized is in the compartment below the living-room TV.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_8_T1", + "correct_answer": "The puzzle you just organized is in the compartment below the living-room TV.", + "original_options": { + "A": "The puzzle you just organized is in the compartment below the living-room TV.", + "B": "In the compartment above the TV.", + "C": "On top of the TV stand.", + "D": "The puzzle you just organized is in the compartment next to the living-room TV, not underneath it.", + "E": "The puzzle you just organized is tucked into a side compartment of the TV stand, rather than the one below." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 20 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_8_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in the living room did I put the puzzle I just sorted out?", + "answer": "The puzzle you just organized is in the compartment below the living-room TV, to your front-right.", + "question_chinese": "我刚刚整理的拼图放在客厅的哪里了？", + "answer_chinese": "你刚刚整理的拼图放在客厅电视机下方的柜格中，在你的右前方。", + "evidence": [ + { + "description": "put the puzzles", + "timestamps": { + "start": "00:39:31", + "end": "00:40:30" + } + } + ], + "question_timestamps": "00:55:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Below the TV, front-left.", + "B": "The puzzle you just organized is in the compartment below the living-room TV, to your front-right.", + "C": "Below the TV, back-right.", + "D": "The puzzle you just organized is in the compartment below the living-room TV, positioned to your back-left.", + "E": "The puzzle you just organized is stored in the compartment above the living-room TV, to your front-right." + }, + "answer_label": "B", + "answer_text": "The puzzle you just organized is in the compartment below the living-room TV, to your front-right.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_8_T2", + "correct_answer": "The puzzle you just organized is in the compartment below the living-room TV, to your front-right.", + "original_options": { + "A": "Below the TV, front-left.", + "B": "The puzzle you just organized is in the compartment below the living-room TV, to your front-right.", + "C": "Below the TV, back-right.", + "D": "The puzzle you just organized is in the compartment below the living-room TV, positioned to your back-left.", + "E": "The puzzle you just organized is stored in the compartment above the living-room TV, to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_6_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in the kitchen did I put the pot I just washed?", + "answer": "You put the pot you just finished washing into the cabinet, to your back-left.", + "question_chinese": "我刚刚洗完的锅放到了厨房的哪里了？", + "answer_chinese": "你把刚刚洗完的锅放到柜子里面了，在你的左后方。", + "evidence": [ + { + "description": "put the pot you just washed the cabinet", + "timestamps": { + "start": "00:32:09", + "end": "00:32:15" + } + } + ], + "question_timestamps": "00:49:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the cabinet, back-right.", + "B": "In the cabinet, front-left.", + "C": "You put the pot you just finished washing into the cabinet, to your front-right instead of behind you.", + "D": "You put the pot you just finished washing into the cabinet, to your back-left.", + "E": "You put the pot you just finished washing into the cabinet on your left side, not back-left." + }, + "answer_label": "D", + "answer_text": "You put the pot you just finished washing into the cabinet, to your back-left.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_6_T1", + "correct_answer": "You put the pot you just finished washing into the cabinet, to your back-left.", + "original_options": { + "A": "In the cabinet, back-right.", + "B": "In the cabinet, front-left.", + "C": "You put the pot you just finished washing into the cabinet, to your front-right instead of behind you.", + "D": "You put the pot you just finished washing into the cabinet, to your back-left.", + "E": "You put the pot you just finished washing into the cabinet on your left side, not back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 17 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_6_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where in the kitchen did I put the pot I just washed?", + "answer": "You put the pot you just finished washing into the cabinet, to your left.", + "question_chinese": "我刚刚洗完的锅放到了厨房的哪里了？", + "answer_chinese": "你把刚刚洗完的锅放到柜子里面了，在你的左方。", + "evidence": [ + { + "description": "put the pot you just washed the cabinet", + "timestamps": { + "start": "00:32:09", + "end": "00:32:15" + } + } + ], + "question_timestamps": "00:49:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the cabinet, to your right.", + "B": "Into the cabinet, behind you.", + "C": "You put the pot you just finished washing into the cabinet, to your front-left rather than directly left.", + "D": "You put the pot you just finished washing into the cabinet, to your back-right instead of on your left.", + "E": "You put the pot you just finished washing into the cabinet, to your left." + }, + "answer_label": "E", + "answer_text": "You put the pot you just finished washing into the cabinet, to your left.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_6_T2", + "correct_answer": "You put the pot you just finished washing into the cabinet, to your left.", + "original_options": { + "A": "Into the cabinet, to your right.", + "B": "Into the cabinet, behind you.", + "C": "You put the pot you just finished washing into the cabinet, to your front-left rather than directly left.", + "D": "You put the pot you just finished washing into the cabinet, to your back-right instead of on your left.", + "E": "You put the pot you just finished washing into the cabinet, to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 19 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_2_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take while carrying the breakfast?", + "answer": "After you picked up the ingredients, you walked straight three steps, turned left to exit through the kitchen door into the hallway, then turned left again into the living room. The breakfast is to your right now.", + "question_chinese": "我刚才拿着早餐走了什么路线？", + "answer_chinese": "你拿起食材后，直走三步，左转离开厨房门，进入过道，随后向左转进入客厅。早餐就在你现在的右边。", + "evidence": [ + { + "description": "pick up breakfast and walk straight three steps", + "timestamps": { + "start": "00:11:00", + "end": "00:11:03" + } + }, + { + "description": "turn left to exit the kitchen door and enter the hallway", + "timestamps": { + "start": "00:11:03", + "end": "00:11:06" + } + }, + { + "description": "turn left to exit the kitchen door and enter the hallway", + "timestamps": { + "start": "00:11:06", + "end": "00:11:09" + } + } + ], + "question_timestamps": "00:11:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked straight two steps.", + "B": "You walked straight four steps.", + "C": "You walked straight three steps, turned right into the hallway, then turned right into the living room.", + "D": "You walked straight three steps, turned left into the hallway, then turned right into the living room.", + "E": "After you picked up the ingredients, you walked straight three steps, turned left to exit through the kitchen door into the hallway, then turned left again into the living room. The breakfast is to your right now." + }, + "answer_label": "E", + "answer_text": "After you picked up the ingredients, you walked straight three steps, turned left to exit through the kitchen door into the hallway, then turned left again into the living room. The breakfast is to your right now.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_2_T1", + "correct_answer": "After you picked up the ingredients, you walked straight three steps, turned left to exit through the kitchen door into the hallway, then turned left again into the living room. The breakfast is to your right now.", + "original_options": { + "A": "You walked straight two steps.", + "B": "You walked straight four steps.", + "C": "You walked straight three steps, turned right into the hallway, then turned right into the living room.", + "D": "You walked straight three steps, turned left into the hallway, then turned right into the living room.", + "E": "After you picked up the ingredients, you walked straight three steps, turned left to exit through the kitchen door into the hallway, then turned left again into the living room. The breakfast is to your right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_2_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take while carrying the breakfast?", + "answer": "After you picked up the ingredients, you walked straight three steps, turned left to exit through the kitchen door into the hallway, then turned left again into the living room.", + "question_chinese": "我刚才拿着早餐走了什么路线？", + "answer_chinese": "你拿起食材后，直走三步，左转离开厨房门，进入过道，随后向左转进入客厅", + "evidence": [ + { + "description": "pick up breakfast and walk straight three steps", + "timestamps": { + "start": "00:11:00", + "end": "00:11:03" + } + }, + { + "description": "turn left to exit the kitchen door and enter the hallway", + "timestamps": { + "start": "00:11:03", + "end": "00:11:06" + } + }, + { + "description": "turn left to exit the kitchen door and enter the hallway", + "timestamps": { + "start": "00:11:06", + "end": "00:11:09" + } + } + ], + "question_timestamps": "00:36:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "After you picked up the ingredients, you walked straight three steps, turned left to exit through the kitchen door into the hallway, then turned left again into the living room.", + "B": "You walked straight one step.", + "C": "You walked straight four steps.", + "D": "You walked straight three steps, turned right into the hallway, then turned right again into the living room.", + "E": "You walked straight three steps, turned left into the hallway, then turned right into the living room." + }, + "answer_label": "A", + "answer_text": "After you picked up the ingredients, you walked straight three steps, turned left to exit through the kitchen door into the hallway, then turned left again into the living room.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_2_T2", + "correct_answer": "After you picked up the ingredients, you walked straight three steps, turned left to exit through the kitchen door into the hallway, then turned left again into the living room.", + "original_options": { + "A": "After you picked up the ingredients, you walked straight three steps, turned left to exit through the kitchen door into the hallway, then turned left again into the living room.", + "B": "You walked straight one step.", + "C": "You walked straight four steps.", + "D": "You walked straight three steps, turned right into the hallway, then turned right again into the living room.", + "E": "You walked straight three steps, turned left into the hallway, then turned right into the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_1_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many eggs did I just use to make breakfast?", + "answer": "I used four eggs to make breakfast, and the eggs are all to your back-right.", + "question_chinese": "我刚才用了几个鸡蛋做早餐？", + "answer_chinese": "我用了四个鸡蛋做早餐，鸡蛋都在你的右后方。", + "evidence": [ + { + "description": "make breakfast with four eggs", + "timestamps": { + "start": "00:06:22", + "end": "00:06:49" + } + } + ], + "question_timestamps": "00:06:52", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "I used three eggs.", + "B": "I used five eggs.", + "C": "I used two eggs to make breakfast, and the eggs are all to your back-right.", + "D": "I used six eggs to make breakfast, and the eggs are all to your back-right.", + "E": "I used four eggs to make breakfast, and the eggs are all to your back-right." + }, + "answer_label": "E", + "answer_text": "I used four eggs to make breakfast, and the eggs are all to your back-right.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_1_T2", + "correct_answer": "I used four eggs to make breakfast, and the eggs are all to your back-right.", + "original_options": { + "A": "I used three eggs.", + "B": "I used five eggs.", + "C": "I used two eggs to make breakfast, and the eggs are all to your back-right.", + "D": "I used six eggs to make breakfast, and the eggs are all to your back-right.", + "E": "I used four eggs to make breakfast, and the eggs are all to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_1_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many eggs did I just use to make breakfast?", + "answer": "I used four eggs to make breakfast, and the eggs are all on your right-hand side.", + "question_chinese": "我刚才用了几个鸡蛋做早餐？", + "answer_chinese": "我用了四个鸡蛋做早餐，鸡蛋都在你的右手边。", + "evidence": [ + { + "description": "make breakfast with four eggs", + "timestamps": { + "start": "00:06:22", + "end": "00:06:49" + } + } + ], + "question_timestamps": "00:11:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "I used three eggs.", + "B": "I used five eggs.", + "C": "I used two eggs to make breakfast, and the eggs are all on your right-hand side.", + "D": "I used four eggs to make breakfast, and the eggs are all on your right-hand side.", + "E": "I used six eggs to make breakfast, and the eggs are all on your right-hand side." + }, + "answer_label": "D", + "answer_text": "I used four eggs to make breakfast, and the eggs are all on your right-hand side.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_1_T1", + "correct_answer": "I used four eggs to make breakfast, and the eggs are all on your right-hand side.", + "original_options": { + "A": "I used three eggs.", + "B": "I used five eggs.", + "C": "I used two eggs to make breakfast, and the eggs are all on your right-hand side.", + "D": "I used four eggs to make breakfast, and the eggs are all on your right-hand side.", + "E": "I used six eggs to make breakfast, and the eggs are all on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_3", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the two glass cups in the kitchen?", + "answer": "Yes, they are in the middle between the kitchen and the living room.", + "question_chinese": "我在厨房中是否能够到两个玻璃杯？", + "answer_chinese": "可以，它就在厨房与客厅中间", + "evidence": [ + { + "description": "saw two glass cups", + "timestamps": { + "start": "00:10:32", + "end": "00:10:43" + } + } + ], + "question_timestamps": "00:10:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, they are in the middle between the kitchen and the living room.", + "B": "No, you can't, because you just took the glass cups to the dining table in the living room." + }, + "answer_label": "A", + "answer_text": "Yes, they are in the middle between the kitchen and the living room.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_3", + "correct_answer": "Yes, they are in the middle between the kitchen and the living room.", + "original_options": { + "A": "Yes, they are in the middle between the kitchen and the living room.", + "B": "No, you can't, because you just took the glass cups to the dining table in the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_4", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the two glass cups in the kitchen?", + "answer": "No, you can't, because you just took the glass cups to the dining table in the living room.", + "question_chinese": "我在厨房中是否能够到两个玻璃杯?", + "answer_chinese": "不可以，因为你刚刚把玻璃杯拿到客厅的餐桌上了", + "evidence": [ + { + "description": "took the glass cups to the dining table in the living room", + "timestamps": { + "start": "00:11:43", + "end": "00:11:46" + } + } + ], + "question_timestamps": "00:20:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, they are in the middle between the kitchen and the living room.", + "B": "No, you can't, because you just took the glass cups to the dining table in the living room." + }, + "answer_label": "B", + "answer_text": "No, you can't, because you just took the glass cups to the dining table in the living room.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_4", + "correct_answer": "No, you can't, because you just took the glass cups to the dining table in the living room.", + "original_options": { + "A": "Yes, they are in the middle between the kitchen and the living room.", + "B": "No, you can't, because you just took the glass cups to the dining table in the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_9_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "In the kitchen, I saw a pink cloth and a blue cloth both on the sink. How far apart were they?", + "answer": "The pink cloth and the blue cloth are 5 cm apart, to your back-left.", + "question_chinese": "我看到在厨房中，粉色布和蓝色布都在洗手盘上，他们之间距离有多远？", + "answer_chinese": "粉色布和蓝色布相距5厘米，在我的左后方。", + "evidence": [ + { + "description": "see two cloths", + "timestamps": { + "start": "00:45:33", + "end": "00:45:37" + } + } + ], + "question_timestamps": "00:45:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "They are 4 cm apart.", + "B": "The pink cloth and the blue cloth are 5 cm apart, to your back-left.", + "C": "They are 6 cm apart.", + "D": "The pink cloth and the blue cloth are 3 cm apart, to your back-left.", + "E": "The pink cloth and the blue cloth are 7 cm apart, to your back-left." + }, + "answer_label": "B", + "answer_text": "The pink cloth and the blue cloth are 5 cm apart, to your back-left.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_9_T1", + "correct_answer": "The pink cloth and the blue cloth are 5 cm apart, to your back-left.", + "original_options": { + "A": "They are 4 cm apart.", + "B": "The pink cloth and the blue cloth are 5 cm apart, to your back-left.", + "C": "They are 6 cm apart.", + "D": "The pink cloth and the blue cloth are 3 cm apart, to your back-left.", + "E": "The pink cloth and the blue cloth are 7 cm apart, to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_9_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "In the kitchen, I saw a pink cloth and a blue cloth both on the sink. How far apart were they?", + "answer": "The pink cloth and the blue cloth are 5 cm apart, to your back-right.", + "question_chinese": "我看到在厨房中，粉色布和蓝色布都在洗手盘上，他们之间距离有多远？", + "answer_chinese": "粉色布和蓝色布相距5厘米，在我的右后方。", + "evidence": [ + { + "description": "see two cloths", + "timestamps": { + "start": "00:45:33", + "end": "00:45:37" + } + } + ], + "question_timestamps": "00:46:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "They are 4 cm apart.", + "B": "They are 6 cm apart.", + "C": "The pink cloth and the blue cloth are 5 cm apart, to your back-right.", + "D": "The pink cloth and the blue cloth are 3 cm apart, to your back-right.", + "E": "The pink cloth and the blue cloth are 7 cm apart, to your back-right." + }, + "answer_label": "C", + "answer_text": "The pink cloth and the blue cloth are 5 cm apart, to your back-right.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_9_T2", + "correct_answer": "The pink cloth and the blue cloth are 5 cm apart, to your back-right.", + "original_options": { + "A": "They are 4 cm apart.", + "B": "They are 6 cm apart.", + "C": "The pink cloth and the blue cloth are 5 cm apart, to your back-right.", + "D": "The pink cloth and the blue cloth are 3 cm apart, to your back-right.", + "E": "The pink cloth and the blue cloth are 7 cm apart, to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_5_T1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I see the helmet take from the living room to the garage?", + "answer": "You picked up the helmet in the living room, left the living room, went through a door, walked to the end of the hallway, opened the garage door, walked to the bicycle in front of you, and put the helmet into the basket.", + "question_chinese": "我看到头盔从客厅到车库经过了什么路径？", + "answer_chinese": "从客厅中拿起头盔，并离开客厅，经过一扇门，走到过道尽头，打开车库门，并走到自行车面前，把头盔放到篮子里", + "evidence": [ + { + "description": "pick up the helmet from the living room and leave the living room", + "timestamps": { + "start": "00:12:25", + "end": "00:12:29" + } + }, + { + "description": "go through a door and walk to the end of the hallway", + "timestamps": { + "start": "00:12:30", + "end": "00:12:35" + } + }, + { + "description": "open the garage door, walk in front of the bicycle, and put the helmet in the basket", + "timestamps": { + "start": "00:12:36", + "end": "00:12:43" + } + } + ], + "question_timestamps": "00:12:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put it in the front basket.", + "B": "You walked to the bicycle behind you.", + "C": "You picked up the helmet in the living room, left the living room, went through a door, walked to the end of the hallway, opened the garage door, walked to the bicycle in front of you, and put the helmet into the basket.", + "D": "You picked it up, opened the garage door first, then walked down the hallway and placed it in the basket.", + "E": "You left the living room, reached the hallway end, opened the garage door, then put it into the basket behind you." + }, + "answer_label": "C", + "answer_text": "You picked up the helmet in the living room, left the living room, went through a door, walked to the end of the hallway, opened the garage door, walked to the bicycle in front of you, and put the helmet into the basket.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_5_T1", + "correct_answer": "You picked up the helmet in the living room, left the living room, went through a door, walked to the end of the hallway, opened the garage door, walked to the bicycle in front of you, and put the helmet into the basket.", + "original_options": { + "A": "You put it in the front basket.", + "B": "You walked to the bicycle behind you.", + "C": "You picked up the helmet in the living room, left the living room, went through a door, walked to the end of the hallway, opened the garage door, walked to the bicycle in front of you, and put the helmet into the basket.", + "D": "You picked it up, opened the garage door first, then walked down the hallway and placed it in the basket.", + "E": "You left the living room, reached the hallway end, opened the garage door, then put it into the basket behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 21 + ] + }, + { + "video_uid": "6fd90f8d-7a4d-425d-a812-3268db0b0342", + "video_length": "00:59:58", + "qid": "6fd90f8d-7a4d-425d-a812-3268db0b0342_5_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I see the helmet take from the living room to the garage?", + "answer": "You picked up the helmet in the living room, left the living room, went through a door, walked to the end of the hallway, opened the garage door, walked to the bicycle in front of you, and put the helmet into the basket behind you.", + "question_chinese": "我看到头盔从客厅到车库经过了什么路径？", + "answer_chinese": "从客厅中拿起头盔，并离开客厅，经过一扇门，走到过道尽头，打开车库门，并走到自行车面前，把头盔放到身后的篮子里。", + "evidence": [ + { + "description": "pick up the helmet from the living room and leave the living room", + "timestamps": { + "start": "00:12:25", + "end": "00:12:29" + } + }, + { + "description": "go through a door and walk to the end of the hallway", + "timestamps": { + "start": "00:12:30", + "end": "00:12:35" + } + }, + { + "description": "open the garage door, walk in front of the bicycle, and put the helmet in the basket", + "timestamps": { + "start": "00:12:36", + "end": "00:12:43" + } + } + ], + "question_timestamps": "00:12:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put the helmet into the basket.", + "B": "You walked to a bicycle behind you.", + "C": "You opened the garage door first, then walked down the hallway and placed the helmet into the basket behind you.", + "D": "You picked up the helmet in the living room, left the living room, went through a door, walked to the end of the hallway, opened the garage door, walked to the bicycle in front of you, and put the helmet into the basket behind you.", + "E": "You went through the door, reached the end of the hallway, then put the helmet into a basket in front of you." + }, + "answer_label": "D", + "answer_text": "You picked up the helmet in the living room, left the living room, went through a door, walked to the end of the hallway, opened the garage door, walked to the bicycle in front of you, and put the helmet into the basket behind you.", + "q_id": "6fd90f8d-7a4d-425d-a812-3268db0b0342_5_T2", + "correct_answer": "You picked up the helmet in the living room, left the living room, went through a door, walked to the end of the hallway, opened the garage door, walked to the bicycle in front of you, and put the helmet into the basket behind you.", + "original_options": { + "A": "You put the helmet into the basket.", + "B": "You walked to a bicycle behind you.", + "C": "You opened the garage door first, then walked down the hallway and placed the helmet into the basket behind you.", + "D": "You picked up the helmet in the living room, left the living room, went through a door, walked to the end of the hallway, opened the garage door, walked to the bicycle in front of you, and put the helmet into the basket behind you.", + "E": "You went through the door, reached the end of the hallway, then put the helmet into a basket in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 22 + ] + } +] diff --git a/QAs/70f2a750-f403-41b8-aabb-480eb3ab4ed4-mcq.json b/QAs/70f2a750-f403-41b8-aabb-480eb3ab4ed4-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c03a5e6fd1927d9222844e761e782ecd316be17e --- /dev/null +++ b/QAs/70f2a750-f403-41b8-aabb-480eb3ab4ed4-mcq.json @@ -0,0 +1,1322 @@ +[ + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_1_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pour the rice into?", + "answer": "You poured the rice into the jar, and the jar is in front of you.", + "question_chinese": "我把米倒进了哪里？", + "answer_chinese": "你把米倒进了罐子里，罐子在你前面。", + "evidence": [ + { + "description": "I poured the rice into jar", + "timestamps": { + "start": "00:14:15", + "end": "00:14:16" + } + } + ], + "question_timestamps": "00:14:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You poured the rice into the pot, and the pot is in front of you.", + "B": "You poured the rice into the jar, and the jar is in front of you." + }, + "answer_label": "B", + "answer_text": "You poured the rice into the jar, and the jar is in front of you.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_1_T2", + "correct_answer": "You poured the rice into the jar, and the jar is in front of you.", + "original_options": { + "A": "You poured the rice into the pot, and the pot is in front of you.", + "B": "You poured the rice into the jar, and the jar is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_1_T3", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pour the rice into?", + "answer": "You poured the rice into the pot, and the pot is in front of you.", + "question_chinese": "我把米倒进了哪里？", + "answer_chinese": "你把米倒进了锅里，锅在你前面。", + "evidence": [ + { + "description": "you poured the rice into po.", + "timestamps": { + "start": "00:14:50", + "end": "00:14:57" + } + } + ], + "question_timestamps": "00:14:59", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You poured the rice into the pot, and the pot is in front of you.", + "B": "You poured the rice into the jar, and the jar is in front of you." + }, + "answer_label": "A", + "answer_text": "You poured the rice into the pot, and the pot is in front of you.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_1_T3", + "correct_answer": "You poured the rice into the pot, and the pot is in front of you.", + "original_options": { + "A": "You poured the rice into the pot, and the pot is in front of you.", + "B": "You poured the rice into the jar, and the jar is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Did I pour the rice into the jar?", + "answer": "Yes.", + "question_chinese": "我把米倒进了罐子里吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "I poured the riceinto the jar", + "timestamps": { + "start": "00:14:15", + "end": "00:14:16" + } + } + ], + "question_timestamps": "00:14:17", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Did I pour the rice into the jar?", + "answer": "Yes.", + "question_chinese": "我把米倒进了罐子里吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "I poured the rice into jar", + "timestamps": { + "start": "00:14:15", + "end": "00:14:16" + } + } + ], + "question_timestamps": "00:14:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_2", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_4_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are next to me?", + "answer": "There is 1 pot next to you, and the pot is to your right.", + "question_chinese": "我旁边有多少个锅？", + "answer_chinese": "你旁边有1个锅，锅在你右边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "0 pots next to you.", + "B": "2 pots next to you.", + "C": "There is 1 pot next to you, and the pot is to your right.", + "D": "There are 3 pots next to you: two are to your right and one is to your left.", + "E": "There are 4 pots next to you: two are to your left and two are to your right." + }, + "answer_label": "C", + "answer_text": "There is 1 pot next to you, and the pot is to your right.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_4_T1", + "correct_answer": "There is 1 pot next to you, and the pot is to your right.", + "original_options": { + "A": "0 pots next to you.", + "B": "2 pots next to you.", + "C": "There is 1 pot next to you, and the pot is to your right.", + "D": "There are 3 pots next to you: two are to your right and one is to your left.", + "E": "There are 4 pots next to you: two are to your left and two are to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_4_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are next to me?", + "answer": "There are 2 pots next to you: one is to your left, and one is to your right.", + "question_chinese": "我旁边有多少个锅？", + "answer_chinese": "你旁边有2个锅，一个在你左边，一个在你右边。", + "evidence": [ + { + "description": "the second one", + "timestamps": { + "start": "00:24:34", + "end": "00:24:35" + } + } + ], + "question_timestamps": "00:24:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "0 pots next to you.", + "B": "1 pot next to you.", + "C": "There are 3 pots next to you: one is to your left and two are to your right.", + "D": "There are 2 pots next to you: one is to your left, and one is to your right.", + "E": "There are 4 pots next to you: two are to your left and two are to your right." + }, + "answer_label": "D", + "answer_text": "There are 2 pots next to you: one is to your left, and one is to your right.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_4_T2", + "correct_answer": "There are 2 pots next to you: one is to your left, and one is to your right.", + "original_options": { + "A": "0 pots next to you.", + "B": "1 pot next to you.", + "C": "There are 3 pots next to you: one is to your left and two are to your right.", + "D": "There are 2 pots next to you: one is to your left, and one is to your right.", + "E": "There are 4 pots next to you: two are to your left and two are to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_9", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I peel garlic?", + "answer": "Peel the garlic, then use a trash can to collect kitchen waste.", + "question_chinese": "我是怎么剥蒜的？", + "answer_chinese": "剥蒜，然后用垃圾桶接厨余垃圾", + "evidence": [ + { + "description": "Peel the garlic", + "timestamps": { + "start": "00:03:37", + "end": "00:03:38" + } + }, + { + "description": "use a trash can to collect kitchen waste", + "timestamps": { + "start": "00:03:39", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:03:41", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Peel garlic, then store it.", + "B": "Peel garlic, then chop it.", + "C": "Peel the garlic, then place the peeled cloves in a bowl and leave the skins on the counter.", + "D": "Peel the garlic, then use a trash can to collect kitchen waste.", + "E": "Peel the garlic, then use a plate to gather the skins before moving them aside for later cleanup." + }, + "answer_label": "D", + "answer_text": "Peel the garlic, then use a trash can to collect kitchen waste.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_9", + "correct_answer": "Peel the garlic, then use a trash can to collect kitchen waste.", + "original_options": { + "A": "Peel garlic, then store it.", + "B": "Peel garlic, then chop it.", + "C": "Peel the garlic, then place the peeled cloves in a bowl and leave the skins on the counter.", + "D": "Peel the garlic, then use a trash can to collect kitchen waste.", + "E": "Peel the garlic, then use a plate to gather the skins before moving them aside for later cleanup." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_10", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I peel garlic?", + "answer": "Peel the garlic, then use a trash can to collect kitchen waste.", + "question_chinese": "我是怎么剥蒜的？", + "answer_chinese": "剥蒜，然后用垃圾桶接厨余垃圾", + "evidence": [ + { + "description": "Peel the garlic", + "timestamps": { + "start": "00:03:37", + "end": "00:03:38" + } + }, + { + "description": "use a trash can to collect kitchen waste.", + "timestamps": { + "start": "00:03:39", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:03:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Use a compost bin.", + "B": "Use a paper bag.", + "C": "Peel the garlic, then put the skins into a small garbage bag and tie it closed afterward.", + "D": "Peel the garlic, then collect the peelings in a kitchen caddy and set it aside when finished.", + "E": "Peel the garlic, then use a trash can to collect kitchen waste." + }, + "answer_label": "E", + "answer_text": "Peel the garlic, then use a trash can to collect kitchen waste.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_10", + "correct_answer": "Peel the garlic, then use a trash can to collect kitchen waste.", + "original_options": { + "A": "Use a compost bin.", + "B": "Use a paper bag.", + "C": "Peel the garlic, then put the skins into a small garbage bag and tie it closed afterward.", + "D": "Peel the garlic, then collect the peelings in a kitchen caddy and set it aside when finished.", + "E": "Peel the garlic, then use a trash can to collect kitchen waste." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_7_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many king oyster mushrooms did I wash?", + "answer": "You washed 0 button mushrooms.", + "question_chinese": "我洗了多少个口蘑？", + "answer_chinese": "你洗了0个口蘑。", + "evidence": [ + { + "description": "zero", + "timestamps": { + "start": "00:06:50", + "end": "00:06:54" + } + } + ], + "question_timestamps": "00:06:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed 0 button mushrooms.", + "B": "You washed 1 button mushroom.", + "C": "You washed 2 button mushrooms.", + "D": "You washed 3 button mushrooms, rinsing each one under running water before setting them aside.", + "E": "You washed 4 button mushrooms, cleaning them one by one and placing them back together afterward." + }, + "answer_label": "A", + "answer_text": "You washed 0 button mushrooms.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_7_T1", + "correct_answer": "You washed 0 button mushrooms.", + "original_options": { + "A": "You washed 0 button mushrooms.", + "B": "You washed 1 button mushroom.", + "C": "You washed 2 button mushrooms.", + "D": "You washed 3 button mushrooms, rinsing each one under running water before setting them aside.", + "E": "You washed 4 button mushrooms, cleaning them one by one and placing them back together afterward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_15", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many button mushrooms did I wash?", + "answer": "ten.", + "question_chinese": "我洗了多少个口蘑？", + "answer_chinese": "10", + "evidence": [ + { + "description": "I washed two button mushrooms", + "timestamps": { + "start": "00:07:39", + "end": "00:07:40" + } + }, + { + "description": "I washed four button mushrooms", + "timestamps": { + "start": "00:07:48", + "end": "00:07:49" + } + }, + { + "description": "I washed ten button mushrooms", + "timestamps": { + "start": "00:08:08", + "end": "00:08:09" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Eight.", + "B": "Twelve.", + "C": "Nine, after rinsing them one by one and placing them together to drain.", + "D": "Eleven, since you washed an extra mushroom along with the rest before setting them down.", + "E": "ten." + }, + "answer_label": "E", + "answer_text": "ten.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_15", + "correct_answer": "ten.", + "original_options": { + "A": "Eight.", + "B": "Twelve.", + "C": "Nine, after rinsing them one by one and placing them together to drain.", + "D": "Eleven, since you washed an extra mushroom along with the rest before setting them down.", + "E": "ten." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_7_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many king oyster mushrooms did I wash?", + "answer": "You washed 10 button mushrooms.", + "question_chinese": "我洗了多少个口蘑？", + "answer_chinese": "你洗了10个口蘑。", + "evidence": [ + { + "description": "I washed ten button mushrooms", + "timestamps": { + "start": "00:08:08", + "end": "00:08:09" + } + } + ], + "question_timestamps": "00:08:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed 8 button mushrooms.", + "B": "You washed 10 button mushrooms.", + "C": "You washed 12 button mushrooms.", + "D": "You washed 9 button mushrooms after rinsing them individually under running water in the sink.", + "E": "You washed 11 button mushrooms, then set them aside together to drain before continuing cooking." + }, + "answer_label": "B", + "answer_text": "You washed 10 button mushrooms.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_7_T2", + "correct_answer": "You washed 10 button mushrooms.", + "original_options": { + "A": "You washed 8 button mushrooms.", + "B": "You washed 10 button mushrooms.", + "C": "You washed 12 button mushrooms.", + "D": "You washed 9 button mushrooms after rinsing them individually under running water in the sink.", + "E": "You washed 11 button mushrooms, then set them aside together to drain before continuing cooking." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_16", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many button mushrooms did I wash?", + "answer": "ten.", + "question_chinese": "我洗了多少个口蘑？", + "answer_chinese": "10", + "evidence": [ + { + "description": "I washed two button mushrooms", + "timestamps": { + "start": "00:07:39", + "end": "00:07:40" + } + }, + { + "description": "I washed four button mushrooms", + "timestamps": { + "start": "00:07:48", + "end": "00:07:49" + } + }, + { + "description": "I washed ten button mushrooms", + "timestamps": { + "start": "00:08:08", + "end": "00:08:09" + } + } + ], + "question_timestamps": "00:08:55", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Eight.", + "B": "ten.", + "C": "Twelve.", + "D": "Nine, washed individually and then collected together after rinsing in the sink.", + "E": "Eleven, because you rinsed one more mushroom along with the rest before stopping." + }, + "answer_label": "B", + "answer_text": "ten.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_16", + "correct_answer": "ten.", + "original_options": { + "A": "Eight.", + "B": "ten.", + "C": "Twelve.", + "D": "Nine, washed individually and then collected together after rinsing in the sink.", + "E": "Eleven, because you rinsed one more mushroom along with the rest before stopping." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_5_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I touch the knife?", + "answer": "You can’t reach the knife, and the knife is to your right.", + "question_chinese": "我能接触到刀吗？", + "answer_chinese": "你不能接触到刀，刀在你右边。", + "evidence": [ + { + "description": "You can not touch the knife.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:39" + } + } + ], + "question_timestamps": "00:03:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can reach the knife.", + "B": "The knife is to your left.", + "C": "You can reach the knife, and the knife is to your right within arm’s length.", + "D": "You can’t reach the knife, and the knife is to your right.", + "E": "You can’t reach the knife, and the knife is in front of you rather than to your right." + }, + "answer_label": "D", + "answer_text": "You can’t reach the knife, and the knife is to your right.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_5_T1", + "correct_answer": "You can’t reach the knife, and the knife is to your right.", + "original_options": { + "A": "You can reach the knife.", + "B": "The knife is to your left.", + "C": "You can reach the knife, and the knife is to your right within arm’s length.", + "D": "You can’t reach the knife, and the knife is to your right.", + "E": "You can’t reach the knife, and the knife is in front of you rather than to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 19 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_5_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I touch the knife?", + "answer": "You can reach the knife, and the knife is to your left.", + "question_chinese": "我能接触到刀吗？", + "answer_chinese": "你能接触到刀，刀在你左边。", + "evidence": [ + { + "description": "You can touch the kinfe.", + "timestamps": { + "start": "00:03:41", + "end": "00:03:44" + } + } + ], + "question_timestamps": "00:03:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can’t reach the knife.", + "B": "The knife is to your right.", + "C": "You can reach the knife, and the knife is to your right instead of your left.", + "D": "You can’t reach the knife, and the knife is to your left, just out of your reach.", + "E": "You can reach the knife, and the knife is to your left." + }, + "answer_label": "E", + "answer_text": "You can reach the knife, and the knife is to your left.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_5_T2", + "correct_answer": "You can reach the knife, and the knife is to your left.", + "original_options": { + "A": "You can’t reach the knife.", + "B": "The knife is to your right.", + "C": "You can reach the knife, and the knife is to your right instead of your left.", + "D": "You can’t reach the knife, and the knife is to your left, just out of your reach.", + "E": "You can reach the knife, and the knife is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_2_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Can I touch the faucet?", + "answer": "You can reach the faucet, and the faucet is in front of you.", + "question_chinese": "我能接触到水龙头吗？", + "answer_chinese": "你能接触到水龙头，水龙头在你前面。", + "evidence": [ + { + "description": "You can touch the faucet directly.", + "timestamps": { + "start": "00:17:04", + "end": "00:17:38" + } + } + ], + "question_timestamps": "00:17:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can reach the faucet, and the faucet is in front of you.", + "B": "You can reach it, behind you.", + "C": "You can't reach it, in front of you.", + "D": "You can reach the faucet, but it is positioned behind you rather than directly in front.", + "E": "You can't reach the faucet, even though it is in front of you within your view." + }, + "answer_label": "A", + "answer_text": "You can reach the faucet, and the faucet is in front of you.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_2_T1", + "correct_answer": "You can reach the faucet, and the faucet is in front of you.", + "original_options": { + "A": "You can reach the faucet, and the faucet is in front of you.", + "B": "You can reach it, behind you.", + "C": "You can't reach it, in front of you.", + "D": "You can reach the faucet, but it is positioned behind you rather than directly in front.", + "E": "You can't reach the faucet, even though it is in front of you within your view." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 16, + 16 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_2_T2", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Can I touch the faucet?", + "answer": "You can’t reach the faucet, and the faucet is behind you.", + "question_chinese": "我能接触到水龙头吗？", + "answer_chinese": "你不能接触到水龙头，水龙头在你后面。", + "evidence": [ + { + "description": "You can not touch the faucet directly.", + "timestamps": { + "start": "00:17:41", + "end": "00:17:50" + } + } + ], + "question_timestamps": "00:17:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can't reach it, in front of you.", + "B": "You can’t reach the faucet, and the faucet is behind you.", + "C": "You can reach it, behind you.", + "D": "You can reach the faucet, and it is behind you even though you can touch it.", + "E": "You can't reach the faucet, and it is located to your left instead of behind you." + }, + "answer_label": "B", + "answer_text": "You can’t reach the faucet, and the faucet is behind you.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_2_T2", + "correct_answer": "You can’t reach the faucet, and the faucet is behind you.", + "original_options": { + "A": "You can't reach it, in front of you.", + "B": "You can’t reach the faucet, and the faucet is behind you.", + "C": "You can reach it, behind you.", + "D": "You can reach the faucet, and it is behind you even though you can touch it.", + "E": "You can't reach the faucet, and it is located to your left instead of behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 16, + 16 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_6_T2", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I season it?", + "answer": "You seasoned it zero times, and the seasoning is in front of you.", + "question_chinese": "我调了几次味？", + "answer_chinese": "你调了零次味，调味平在你前面。", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:06:05", + "end": "00:06:13" + } + } + ], + "question_timestamps": "00:06:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You seasoned it zero times, and the seasoning is in front of you.", + "B": "You seasoned it once, in front.", + "C": "You seasoned it twice, in front.", + "D": "You seasoned it three times, and the seasoning container is still in front of you.", + "E": "You seasoned it four times, and the seasoning is positioned in front of you the whole time." + }, + "answer_label": "A", + "answer_text": "You seasoned it zero times, and the seasoning is in front of you.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_6_T2", + "correct_answer": "You seasoned it zero times, and the seasoning is in front of you.", + "original_options": { + "A": "You seasoned it zero times, and the seasoning is in front of you.", + "B": "You seasoned it once, in front.", + "C": "You seasoned it twice, in front.", + "D": "You seasoned it three times, and the seasoning container is still in front of you.", + "E": "You seasoned it four times, and the seasoning is positioned in front of you the whole time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_11", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I adjust the seasoning?", + "answer": "two.", + "question_chinese": "我调了几次味？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:10:05", + "end": "00:10:06" + } + }, + { + "description": "the second time", + "timestamps": { + "start": "00:12:21", + "end": "00:12:22" + } + } + ], + "question_timestamps": "00:12:23", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "two.", + "B": "One.", + "C": "Three.", + "D": "You adjusted the seasoning once during the process, not twice as expected.", + "E": "You adjusted the seasoning five times overall, with repeated changes throughout the preparation." + }, + "answer_label": "A", + "answer_text": "two.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_11", + "correct_answer": "two.", + "original_options": { + "A": "two.", + "B": "One.", + "C": "Three.", + "D": "You adjusted the seasoning once during the process, not twice as expected.", + "E": "You adjusted the seasoning five times overall, with repeated changes throughout the preparation." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_12", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I adjust the seasoning?", + "answer": "two.", + "question_chinese": "我调了几次味？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:10:05", + "end": "00:10:06" + } + }, + { + "description": "the second time", + "timestamps": { + "start": "00:12:21", + "end": "00:12:22" + } + } + ], + "question_timestamps": "00:12:45", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "You adjusted the seasoning seven times in total, making several changes beyond just a couple.", + "D": "You adjusted the seasoning eight times overall, with many repeated tweaks during the task.", + "E": "two." + }, + "answer_label": "E", + "answer_text": "two.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_12", + "correct_answer": "two.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "You adjusted the seasoning seven times in total, making several changes beyond just a couple.", + "D": "You adjusted the seasoning eight times overall, with many repeated tweaks during the task.", + "E": "two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_6_T1", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I season it?", + "answer": "You seasoned it 4 times, and the seasoning is in front of you.", + "question_chinese": "我调了几次味？", + "answer_chinese": "你调了4次味，调味品在你前面。", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:10:05", + "end": "00:11:07" + } + }, + { + "description": "the second time", + "timestamps": { + "start": "00:12:21", + "end": "00:12:53" + } + }, + { + "description": "the third time.", + "timestamps": { + "start": "00:12:54", + "end": "00:13:27" + } + } + ], + "question_timestamps": "00:13:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You seasoned it three times.", + "B": "You seasoned it five times.", + "C": "You seasoned it two times, and the seasoning is in front of you while you do it.", + "D": "You seasoned it six times, and the seasoning remains in front of you throughout the actions.", + "E": "You seasoned it 4 times, and the seasoning is in front of you." + }, + "answer_label": "E", + "answer_text": "You seasoned it 4 times, and the seasoning is in front of you.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_6_T1", + "correct_answer": "You seasoned it 4 times, and the seasoning is in front of you.", + "original_options": { + "A": "You seasoned it three times.", + "B": "You seasoned it five times.", + "C": "You seasoned it two times, and the seasoning is in front of you while you do it.", + "D": "You seasoned it six times, and the seasoning remains in front of you throughout the actions.", + "E": "You seasoned it 4 times, and the seasoning is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_7", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are there?", + "answer": "Two.", + "question_chinese": "有多少个锅？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:24:34", + "end": "00:24:35" + } + } + ], + "question_timestamps": "00:24:36", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "There are four pots total in the scene, which is more than just a couple.", + "D": "Two.", + "E": "There are five pots present overall, which is more than the number you expected." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_7", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "There are four pots total in the scene, which is more than just a couple.", + "D": "Two.", + "E": "There are five pots present overall, which is more than the number you expected." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_8", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are there?", + "answer": "two.", + "question_chinese": "有多少个锅？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:24:34", + "end": "00:24:35" + } + } + ], + "question_timestamps": "00:24:50", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One pot.", + "B": "Three pots.", + "C": "two.", + "D": "There are four pots in total, not the smaller number you might think.", + "E": "There are five pots altogether, which is more than just a few." + }, + "answer_label": "C", + "answer_text": "two.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_8", + "correct_answer": "two.", + "original_options": { + "A": "One pot.", + "B": "Three pots.", + "C": "two.", + "D": "There are four pots in total, not the smaller number you might think.", + "E": "There are five pots altogether, which is more than just a few." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_3", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are the shrimp?", + "answer": "On the plate.", + "question_chinese": "虾仁在哪里？", + "answer_chinese": "在盘子上", + "evidence": [ + { + "description": "On the plate", + "timestamps": { + "start": "00:17:04", + "end": "00:17:05" + } + } + ], + "question_timestamps": "00:17:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the bowl.", + "B": "On the counter.", + "C": "In the pot beside the stove, not served out onto anything yet.", + "D": "Under the table, dropped there instead of being placed up on the plate.", + "E": "On the plate." + }, + "answer_label": "E", + "answer_text": "On the plate.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_3", + "correct_answer": "On the plate.", + "original_options": { + "A": "In the bowl.", + "B": "On the counter.", + "C": "In the pot beside the stove, not served out onto anything yet.", + "D": "Under the table, dropped there instead of being placed up on the plate.", + "E": "On the plate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 13 + ] + }, + { + "video_uid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4", + "video_length": "00:27:55", + "qid": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_4", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are the shrimp?", + "answer": "On the plate.", + "question_chinese": "虾仁在哪里？", + "answer_chinese": "在盘子上", + "evidence": [ + { + "description": "On the plate", + "timestamps": { + "start": "00:17:04", + "end": "00:17:05" + } + } + ], + "question_timestamps": "00:17:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the plate.", + "B": "On the counter.", + "C": "In the bowl.", + "D": "On the cutting board, set out there while you prepare the rest of the food.", + "E": "Inside the basket, not on any dish, as if you just brought them over." + }, + "answer_label": "A", + "answer_text": "On the plate.", + "q_id": "70f2a750-f403-41b8-aabb-480eb3ab4ed4_4", + "correct_answer": "On the plate.", + "original_options": { + "A": "On the plate.", + "B": "On the counter.", + "C": "In the bowl.", + "D": "On the cutting board, set out there while you prepare the rest of the food.", + "E": "Inside the basket, not on any dish, as if you just brought them over." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 14 + ] + } +] diff --git a/QAs/7633b170-63e1-4f4a-839e-152cb232ea3b-001-mcq.json b/QAs/7633b170-63e1-4f4a-839e-152cb232ea3b-001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c2d52a391d09c69d952fa3650ceae2bbbd76d840 --- /dev/null +++ b/QAs/7633b170-63e1-4f4a-839e-152cb232ea3b-001-mcq.json @@ -0,0 +1,850 @@ +[ + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001", + "video_length": "00:20:03", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I touch after getting out of the car?", + "answer": "The trunk, dirt.", + "question_chinese": "我下车后碰了什么？", + "answer_chinese": "车后厢，泥土", + "evidence": [ + { + "description": "The trunk", + "timestamps": { + "start": "00:04:11", + "end": "00:04:12" + } + }, + { + "description": "dirt", + "timestamps": { + "start": "00:04:26", + "end": "00:04:27" + } + } + ], + "question_timestamps": "00:04:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The hood, grass.", + "B": "The mirror, sand.", + "C": "You touched the bumper first, then the concrete right after stepping away from the car.", + "D": "The trunk, dirt.", + "E": "After exiting, you brushed the seat and then stepped onto the asphalt outside." + }, + "answer_label": "D", + "answer_text": "The trunk, dirt.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_3", + "correct_answer": "The trunk, dirt.", + "original_options": { + "A": "The hood, grass.", + "B": "The mirror, sand.", + "C": "You touched the bumper first, then the concrete right after stepping away from the car.", + "D": "The trunk, dirt.", + "E": "After exiting, you brushed the seat and then stepped onto the asphalt outside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 13 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001", + "video_length": "00:20:03", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I touch after getting out of the car?", + "answer": "The trunk, dirt, and potted plants.", + "question_chinese": "我下车后碰了什么？", + "answer_chinese": "车后厢，泥土，盆栽", + "evidence": [ + { + "description": "The trunk", + "timestamps": { + "start": "00:04:11", + "end": "00:04:12" + } + }, + { + "description": "dirt", + "timestamps": { + "start": "00:04:26", + "end": "00:04:27" + } + }, + { + "description": "potted plants", + "timestamps": { + "start": "00:04:43", + "end": "00:04:44" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The car door handle.", + "B": "The trunk, dirt, and potted plants.", + "C": "The front bumper.", + "D": "You touched the front seats, then the pavement, and then some outdoor tools near the car.", + "E": "You brushed the windows, stepped onto the sidewalk, and then touched some leafy ferns nearby." + }, + "answer_label": "B", + "answer_text": "The trunk, dirt, and potted plants.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_4", + "correct_answer": "The trunk, dirt, and potted plants.", + "original_options": { + "A": "The car door handle.", + "B": "The trunk, dirt, and potted plants.", + "C": "The front bumper.", + "D": "You touched the front seats, then the pavement, and then some outdoor tools near the car.", + "E": "You brushed the windows, stepped onto the sidewalk, and then touched some leafy ferns nearby." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 15 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001", + "video_length": "00:20:03", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many things did I carry off the bus?", + "answer": "Three.", + "question_chinese": "我搬了几样东西下车？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "the second and third one", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You carried off four things in total, juggling an extra item as you stepped down from the bus.", + "E": "You carried off five items, with multiple belongings in hand rather than only three." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_1", + "correct_answer": "Three.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You carried off four things in total, juggling an extra item as you stepped down from the bus.", + "E": "You carried off five items, with multiple belongings in hand rather than only three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 1, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 14 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001", + "video_length": "00:20:03", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many things did I carry off the bus?", + "answer": "Four.", + "question_chinese": "我搬了几样东西下车？", + "answer_chinese": "4", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "the second and third one", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "the fourth one", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "You carried off one thing only, leaving the rest behind instead of taking four items.", + "E": "You carried off seven things, far more than four, as you exited the bus." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_2", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "You carried off one thing only, leaving the rest behind instead of taking four items.", + "E": "You carried off seven things, far more than four, as you exited the bus." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 1, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001", + "video_length": "00:20:03", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_9", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Am I touching the sapling with one hand or both hands now?", + "answer": "One hand.", + "question_chinese": "我现在是单手还是双手触碰树苗？", + "answer_chinese": "单手", + "evidence": [ + { + "description": "one hand", + "timestamps": { + "start": "00:15:41", + "end": "00:15:42" + } + } + ], + "question_timestamps": "00:15:57", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "One hand.", + "B": "Both hands." + }, + "answer_label": "A", + "answer_text": "One hand.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_9", + "correct_answer": "One hand.", + "original_options": { + "A": "One hand.", + "B": "Both hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001", + "video_length": "00:20:03", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_10", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Am I touching the sapling with one hand or both hands now?", + "answer": "Both hands.", + "question_chinese": "我现在是单手还是双手触碰树苗？", + "answer_chinese": "双手", + "evidence": [ + { + "description": "both hands", + "timestamps": { + "start": "00:15:47", + "end": "00:15:48" + } + } + ], + "question_timestamps": "00:16:25", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "One hand.", + "B": "Both hands." + }, + "answer_label": "B", + "answer_text": "Both hands.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_10", + "correct_answer": "Both hands.", + "original_options": { + "A": "One hand.", + "B": "Both hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001", + "video_length": "00:20:03", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_5", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position after putting down the potted plant?", + "answer": "Turn right and walk ten meters.", + "question_chinese": "放下盆栽后我怎么走到当前位置？", + "answer_chinese": "右转走10米", + "evidence": [ + { + "description": "Turn right and walk 10 meters", + "timestamps": { + "start": "00:05:32", + "end": "00:05:33" + } + } + ], + "question_timestamps": "00:05:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left and walk ten meters.", + "B": "Turn right and walk five meters.", + "C": "Turn right and walk about fifteen meters before stopping at your current spot.", + "D": "Walk straight for ten meters, then turn right to reach where you are now.", + "E": "Turn right and walk ten meters." + }, + "answer_label": "E", + "answer_text": "Turn right and walk ten meters.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_5", + "correct_answer": "Turn right and walk ten meters.", + "original_options": { + "A": "Turn left and walk ten meters.", + "B": "Turn right and walk five meters.", + "C": "Turn right and walk about fifteen meters before stopping at your current spot.", + "D": "Walk straight for ten meters, then turn right to reach where you are now.", + "E": "Turn right and walk ten meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 14 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001", + "video_length": "00:20:03", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position after putting down the potted plant?", + "answer": "Turn right and walk ten meters, then turn left and go upstairs.", + "question_chinese": "放下盆栽后我怎么走到当前位置？", + "answer_chinese": "右转走10米，然后左转上楼梯", + "evidence": [ + { + "description": "Turn right and walk 10 meters", + "timestamps": { + "start": "00:05:32", + "end": "00:05:33" + } + }, + { + "description": "turn left and go upstairs", + "timestamps": { + "start": "00:05:57", + "end": "00:05:58" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right and walk ten meters, then turn left and go upstairs.", + "B": "Turn right, go downstairs.", + "C": "Turn left, go upstairs.", + "D": "Turn right and walk fifteen meters, then turn left and go upstairs to your current position.", + "E": "Turn right and walk ten meters, then turn right and go upstairs to end up here." + }, + "answer_label": "A", + "answer_text": "Turn right and walk ten meters, then turn left and go upstairs.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_6", + "correct_answer": "Turn right and walk ten meters, then turn left and go upstairs.", + "original_options": { + "A": "Turn right and walk ten meters, then turn left and go upstairs.", + "B": "Turn right, go downstairs.", + "C": "Turn left, go upstairs.", + "D": "Turn right and walk fifteen meters, then turn left and go upstairs to your current position.", + "E": "Turn right and walk ten meters, then turn right and go upstairs to end up here." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001", + "video_length": "00:20:03", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_7", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after reading the drawings?", + "answer": "Checked your phone.", + "question_chinese": "看完图纸后我干了什么？", + "answer_chinese": "看手机", + "evidence": [ + { + "description": "Checked my phone", + "timestamps": { + "start": "00:10:09", + "end": "00:10:10" + } + } + ], + "question_timestamps": "00:10:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Checked your watch.", + "B": "Checked your laptop.", + "C": "Checked your phone.", + "D": "Checked your notebook behind the drawings after you finished reading them carefully.", + "E": "Checked your tablet on the table once you were done looking at the drawings." + }, + "answer_label": "C", + "answer_text": "Checked your phone.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_7", + "correct_answer": "Checked your phone.", + "original_options": { + "A": "Checked your watch.", + "B": "Checked your laptop.", + "C": "Checked your phone.", + "D": "Checked your notebook behind the drawings after you finished reading them carefully.", + "E": "Checked your tablet on the table once you were done looking at the drawings." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 2, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 14 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001", + "video_length": "00:20:03", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_8", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after reading the drawings?", + "answer": "Checked your phone, then went downstairs and walked left.", + "question_chinese": "看完图纸后我干了什么？", + "answer_chinese": "看手机然后下楼向左走", + "evidence": [ + { + "description": "Checked my phone", + "timestamps": { + "start": "00:10:09", + "end": "00:10:10" + } + }, + { + "description": "went downstairs and walked left.", + "timestamps": { + "start": "00:10:36", + "end": "00:10:37" + } + } + ], + "question_timestamps": "00:13:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Checked your phone, then went upstairs.", + "B": "Checked your phone, then went downstairs and walked left.", + "C": "Checked your phone, then walked right downstairs.", + "D": "Checked your phone, then went downstairs and walked straight ahead instead of turning left.", + "E": "Checked your phone, then went upstairs and walked left to reach your next position." + }, + "answer_label": "B", + "answer_text": "Checked your phone, then went downstairs and walked left.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-001_8", + "correct_answer": "Checked your phone, then went downstairs and walked left.", + "original_options": { + "A": "Checked your phone, then went upstairs.", + "B": "Checked your phone, then went downstairs and walked left.", + "C": "Checked your phone, then walked right downstairs.", + "D": "Checked your phone, then went downstairs and walked straight ahead instead of turning left.", + "E": "Checked your phone, then went upstairs and walked left to reach your next position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 14 + ] + } +] diff --git a/QAs/7633b170-63e1-4f4a-839e-152cb232ea3b-002-mcq.json b/QAs/7633b170-63e1-4f4a-839e-152cb232ea3b-002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..0de8c92fddbc78e96f5f9918c5dec647d443c36b --- /dev/null +++ b/QAs/7633b170-63e1-4f4a-839e-152cb232ea3b-002-mcq.json @@ -0,0 +1,866 @@ +[ + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002", + "video_length": "00:33:12", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_5", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I look at the drawings when I was on the stairs just now?", + "answer": "Two.", + "question_chinese": "刚才我在楼梯上看了多少次图纸？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:07:45", + "end": "00:07:46" + } + }, + { + "description": "the second time", + "timestamps": { + "start": "00:10:09", + "end": "00:10:10" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One time.", + "B": "Three times.", + "C": "You looked at the drawings once while you were standing on the stairs.", + "D": "Two.", + "E": "You looked at the drawings four times while you were on the stairs just now." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_5", + "correct_answer": "Two.", + "original_options": { + "A": "One time.", + "B": "Three times.", + "C": "You looked at the drawings once while you were standing on the stairs.", + "D": "Two.", + "E": "You looked at the drawings four times while you were on the stairs just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 0, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002", + "video_length": "00:33:12", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_6", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I look at the drawings when I was on the stairs just now?", + "answer": "Three.", + "question_chinese": "刚才我在楼梯上看了多少次图纸？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:07:45", + "end": "00:07:46" + } + }, + { + "description": "the second time", + "timestamps": { + "start": "00:10:09", + "end": "00:10:10" + } + }, + { + "description": "the tihid time", + "timestamps": { + "start": "00:17:02", + "end": "00:17:03" + } + } + ], + "question_timestamps": "00:25:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Four times.", + "C": "You only looked at the drawings one time while you were on the stairs.", + "D": "You looked at the drawings five times during your brief stop on the stairs.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_6", + "correct_answer": "Three.", + "original_options": { + "A": "Two times.", + "B": "Four times.", + "C": "You only looked at the drawings one time while you were on the stairs.", + "D": "You looked at the drawings five times during your brief stop on the stairs.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 0, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002", + "video_length": "00:33:12", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_3", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after picking up the rake?", + "answer": "Walked forward 30 meters, then turned left.", + "question_chinese": "拿完钉耙后我干了什么？", + "answer_chinese": "向前走了30米，然后左转上楼梯", + "evidence": [ + { + "description": "Walked forward 30 meters", + "timestamps": { + "start": "00:07:17", + "end": "00:07:18" + } + }, + { + "description": "turned left and went upstairs", + "timestamps": { + "start": "00:07:37", + "end": "00:07:38" + } + } + ], + "question_timestamps": "00:07:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walked forward 20 meters, then turned left.", + "B": "Walked forward 30 meters, then turned left.", + "C": "Walked backward 30 meters, then turned left.", + "D": "Walked forward 30 meters, then turned right instead of turning left afterward.", + "E": "Walked forward about forty meters first, then turned left to continue from there." + }, + "answer_label": "B", + "answer_text": "Walked forward 30 meters, then turned left.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_3", + "correct_answer": "Walked forward 30 meters, then turned left.", + "original_options": { + "A": "Walked forward 20 meters, then turned left.", + "B": "Walked forward 30 meters, then turned left.", + "C": "Walked backward 30 meters, then turned left.", + "D": "Walked forward 30 meters, then turned right instead of turning left afterward.", + "E": "Walked forward about forty meters first, then turned left to continue from there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 0, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 13 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002", + "video_length": "00:33:12", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_4", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after picking up the rake?", + "answer": "Walked forward 30 meters, then turned left and went upstairs to look at the drawings.", + "question_chinese": "拿完钉耙后我干了什么？", + "answer_chinese": "向前走了30米，然后左转上楼梯，看图纸", + "evidence": [ + { + "description": "Walked forward 30 meters", + "timestamps": { + "start": "00:07:17", + "end": "00:07:18" + } + }, + { + "description": "turned left and went upstairs", + "timestamps": { + "start": "00:07:37", + "end": "00:07:38" + } + }, + { + "description": "look at the drawings.", + "timestamps": { + "start": "00:07:45", + "end": "00:07:46" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Walked forward 30 meters, then turned left and went upstairs to look at the drawings.", + "B": "Turned right and went upstairs to the drawings.", + "C": "Went downstairs to look at the drawings.", + "D": "Walked forward 30 meters, then turned right and went upstairs to look at the drawings.", + "E": "Walked forward 40 meters, then turned left and went upstairs to look at the drawings." + }, + "answer_label": "A", + "answer_text": "Walked forward 30 meters, then turned left and went upstairs to look at the drawings.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_4", + "correct_answer": "Walked forward 30 meters, then turned left and went upstairs to look at the drawings.", + "original_options": { + "A": "Walked forward 30 meters, then turned left and went upstairs to look at the drawings.", + "B": "Turned right and went upstairs to the drawings.", + "C": "Went downstairs to look at the drawings.", + "D": "Walked forward 30 meters, then turned right and went upstairs to look at the drawings.", + "E": "Walked forward 40 meters, then turned left and went upstairs to look at the drawings." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 0, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 15 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002", + "video_length": "00:33:12", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I see the mound of dirt on the ground now that I'm standing in the back of the car?", + "answer": "Yes.", + "question_chinese": "现在我站在车后厢可以看到地上的土堆吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "I can see the soil pile from the trunk.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002", + "video_length": "00:33:12", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I see the mound of dirt on the ground now that I'm standing in the back of the car?", + "answer": "No.", + "question_chinese": "现在我站在车后厢可以看到地上的土堆吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "I cannot see the soil pile from the trunk.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_2", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002", + "video_length": "00:33:12", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_10", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position after looking at the drawings?", + "answer": "Turn around and go downstairs, walk forward for twenty meters.", + "question_chinese": "看完图纸后，我是怎么走到当前位置的？", + "answer_chinese": "向后转并下楼，向前走二十米", + "evidence": [ + { + "description": "Turn around and go downstairs", + "timestamps": { + "start": "00:07:48", + "end": "00:07:49" + } + }, + { + "description": "walk forward for twenty meters", + "timestamps": { + "start": "00:07:50", + "end": "00:07:51" + } + } + ], + "question_timestamps": "00:08:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn around and go downstairs, walk forward for twenty meters.", + "B": "Turn left and go downstairs.", + "C": "Turn around and go upstairs.", + "D": "Turn around, go downstairs, then walk forward for ten meters instead of twenty.", + "E": "Turn around and go downstairs, walk forward for twenty-five meters before stopping at your position." + }, + "answer_label": "A", + "answer_text": "Turn around and go downstairs, walk forward for twenty meters.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_10", + "correct_answer": "Turn around and go downstairs, walk forward for twenty meters.", + "original_options": { + "A": "Turn around and go downstairs, walk forward for twenty meters.", + "B": "Turn left and go downstairs.", + "C": "Turn around and go upstairs.", + "D": "Turn around, go downstairs, then walk forward for ten meters instead of twenty.", + "E": "Turn around and go downstairs, walk forward for twenty-five meters before stopping at your position." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 2, + "D": 1, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 15 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002", + "video_length": "00:33:12", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_9", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position after looking at the drawings?", + "answer": "Turn around and go downstairs, walk forward for twenty meters, turn left and walk forward for two meters.", + "question_chinese": "看完图纸后，我是怎么走到当前位置的？", + "answer_chinese": "向后转并下楼，向前走二十米，左转并向前走两米", + "evidence": [ + { + "description": "Turn around and go downstairs", + "timestamps": { + "start": "00:07:48", + "end": "00:07:49" + } + }, + { + "description": "walk forward for twenty meters", + "timestamps": { + "start": "00:07:50", + "end": "00:07:51" + } + }, + { + "description": "turn left and walk forward for two meters.", + "timestamps": { + "start": "00:08:37", + "end": "00:08:38" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn around and go upstairs.", + "B": "Turn around, then turn right.", + "C": "Turn around and go downstairs, walk forward for twenty meters, turn left and walk forward for two meters.", + "D": "Turn around and go downstairs, walk forward for twenty meters, then turn right and walk two meters.", + "E": "Turn around and go downstairs, walk forward for thirty meters, turn left, then walk forward for two meters." + }, + "answer_label": "C", + "answer_text": "Turn around and go downstairs, walk forward for twenty meters, turn left and walk forward for two meters.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_9", + "correct_answer": "Turn around and go downstairs, walk forward for twenty meters, turn left and walk forward for two meters.", + "original_options": { + "A": "Turn around and go upstairs.", + "B": "Turn around, then turn right.", + "C": "Turn around and go downstairs, walk forward for twenty meters, turn left and walk forward for two meters.", + "D": "Turn around and go downstairs, walk forward for twenty meters, then turn right and walk two meters.", + "E": "Turn around and go downstairs, walk forward for thirty meters, turn left, then walk forward for two meters." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 2, + "D": 1, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002", + "video_length": "00:33:12", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_7", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I go back and forth to get the potted plants after raking the soil?", + "answer": "One.", + "question_chinese": "耙完土后我往返拿了多少次盆栽？", + "answer_chinese": "1", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:15:03", + "end": "00:15:04" + } + } + ], + "question_timestamps": "00:15:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Three times.", + "C": "One.", + "D": "You went back and forth two times to bring the potted plants over after raking the soil.", + "E": "You made three back-and-forth trips to collect all the potted plants once the soil was raked." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_7", + "correct_answer": "One.", + "original_options": { + "A": "Two times.", + "B": "Three times.", + "C": "One.", + "D": "You went back and forth two times to bring the potted plants over after raking the soil.", + "E": "You made three back-and-forth trips to collect all the potted plants once the soil was raked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002", + "video_length": "00:33:12", + "qid": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_8", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I go back and forth to get the potted plants after raking the soil?", + "answer": "Two.", + "question_chinese": "耙完土后我往返拿了多少次盆栽？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:15:03", + "end": "00:15:04" + } + }, + { + "description": "the second time", + "timestamps": { + "start": "00:15:53", + "end": "00:15:54" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One time.", + "B": "Three times.", + "C": "You went back and forth one time to get the potted plants after you finished raking the soil.", + "D": "You made three back-and-forth trips to retrieve the potted plants once the raking was done.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "7633b170-63e1-4f4a-839e-152cb232ea3b-002_8", + "correct_answer": "Two.", + "original_options": { + "A": "One time.", + "B": "Three times.", + "C": "You went back and forth one time to get the potted plants after you finished raking the soil.", + "D": "You made three back-and-forth trips to retrieve the potted plants once the raking was done.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 16 + ] + } +] diff --git a/QAs/79b9f7e3-f1ce-4838-9887-1deeba08e600-mcq.json b/QAs/79b9f7e3-f1ce-4838-9887-1deeba08e600-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..243027f235ae4928393547f872595e2d4538d51e --- /dev/null +++ b/QAs/79b9f7e3-f1ce-4838-9887-1deeba08e600-mcq.json @@ -0,0 +1,1347 @@ +[ + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_3_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I’m thirsty—what should I do?", + "answer": "Pick up the cup on your right-hand side and drink water.", + "question_chinese": "我口渴了该怎么办？", + "answer_chinese": "拿起右手边的水杯喝水", + "evidence": [ + { + "description": "water cup on the right", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Pick up the cup on your right-hand side and drink water.", + "B": "Pick up the cup on your left.", + "C": "Grab the cup in front of you.", + "D": "Pick up the cup on your left-hand side and drink some water from it.", + "E": "Reach for the cup in front of you, lift it up, and take a drink of water." + }, + "answer_label": "A", + "answer_text": "Pick up the cup on your right-hand side and drink water.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_3_T1", + "correct_answer": "Pick up the cup on your right-hand side and drink water.", + "original_options": { + "A": "Pick up the cup on your right-hand side and drink water.", + "B": "Pick up the cup on your left.", + "C": "Grab the cup in front of you.", + "D": "Pick up the cup on your left-hand side and drink some water from it.", + "E": "Reach for the cup in front of you, lift it up, and take a drink of water." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 17 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_3_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I’m thirsty—what should I do?", + "answer": "Pick up the cup on the table in front of you and drink water.", + "question_chinese": "我口渴了该怎么办？", + "answer_chinese": "拿起你面前桌子伤的的水杯喝水", + "evidence": [ + { + "description": "water cup on the right", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Pick up the cup behind you.", + "B": "Pick up the cup on your left.", + "C": "Pick up the cup on the table to your right and drink water from it.", + "D": "Pick up the cup on the table in front of you and drink water.", + "E": "Pick up the cup on the table behind you, then take a drink of water." + }, + "answer_label": "D", + "answer_text": "Pick up the cup on the table in front of you and drink water.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_3_T2", + "correct_answer": "Pick up the cup on the table in front of you and drink water.", + "original_options": { + "A": "Pick up the cup behind you.", + "B": "Pick up the cup on your left.", + "C": "Pick up the cup on the table to your right and drink water from it.", + "D": "Pick up the cup on the table in front of you and drink water.", + "E": "Pick up the cup on the table behind you, then take a drink of water." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 15 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_8_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I wash the paintbrush?", + "answer": "First dip water from the box in front of you, then swipe some paint, and use a paper towel to absorb the extra water.", + "question_chinese": "我怎么洗画笔的？", + "answer_chinese": "先在面前的盒子里蘸水，然后抹一下颜料，纸巾吸收多余的水分", + "evidence": [ + { + "description": "wipe dry with a tissue", + "timestamps": { + "start": "00:31:49", + "end": "00:31:50" + } + }, + { + "description": "dab with water", + "timestamps": { + "start": "00:31:46", + "end": "00:31:47" + } + }, + { + "description": "Wipe some paint.", + "timestamps": { + "start": "00:31:47", + "end": "00:31:48" + } + } + ], + "question_timestamps": "00:31:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Dip water behind you, then swipe paint.", + "B": "First dip water from the box in front of you, then swipe some paint, and use a paper towel to absorb the extra water.", + "C": "Dip water to your left, then swipe paint.", + "D": "First dip water from the box to your right, then swipe paint and blot the brush on a paper towel.", + "E": "First dip water from the box behind you, then swipe some paint, and squeeze out excess water with a paper towel." + }, + "answer_label": "B", + "answer_text": "First dip water from the box in front of you, then swipe some paint, and use a paper towel to absorb the extra water.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_8_T2", + "correct_answer": "First dip water from the box in front of you, then swipe some paint, and use a paper towel to absorb the extra water.", + "original_options": { + "A": "Dip water behind you, then swipe paint.", + "B": "First dip water from the box in front of you, then swipe some paint, and use a paper towel to absorb the extra water.", + "C": "Dip water to your left, then swipe paint.", + "D": "First dip water from the box to your right, then swipe paint and blot the brush on a paper towel.", + "E": "First dip water from the box behind you, then swipe some paint, and squeeze out excess water with a paper towel." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 20, + 21 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_19", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I wash paintbrushes?", + "answer": "First, wipe dry with a tissue, then dab with water.", + "question_chinese": "我怎么洗画笔的？", + "answer_chinese": "先纸巾擦干，然后蘸水", + "evidence": [ + { + "description": "wipe dry with a tissue", + "timestamps": { + "start": "00:31:49", + "end": "00:31:50" + } + }, + { + "description": "dab with water", + "timestamps": { + "start": "00:58:40", + "end": "00:58:41" + } + } + ], + "question_timestamps": "00:58:42", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Soak in water, then wipe with tissue.", + "B": "First, wipe dry with a tissue, then dab with water.", + "C": "Rinse first, then dry with tissue.", + "D": "First dab the brush with water, then wipe it dry using a tissue until it stops dripping.", + "E": "First scrub the brush with a tissue, then rinse it in water and let it air dry." + }, + "answer_label": "B", + "answer_text": "First, wipe dry with a tissue, then dab with water.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_19", + "correct_answer": "First, wipe dry with a tissue, then dab with water.", + "original_options": { + "A": "Soak in water, then wipe with tissue.", + "B": "First, wipe dry with a tissue, then dab with water.", + "C": "Rinse first, then dry with tissue.", + "D": "First dab the brush with water, then wipe it dry using a tissue until it stops dripping.", + "E": "First scrub the brush with a tissue, then rinse it in water and let it air dry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 17 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_8_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I wash the paintbrush?", + "answer": "First dry it off with the paper towel in front of you, then dip it in water.", + "question_chinese": "我怎么洗画笔的？", + "answer_chinese": "先用面前的纸巾擦干，然后蘸水", + "evidence": [ + { + "description": "wipe dry with a tissue", + "timestamps": { + "start": "00:31:49", + "end": "00:31:50" + } + }, + { + "description": "dab with water", + "timestamps": { + "start": "00:58:40", + "end": "00:58:41" + } + } + ], + "question_timestamps": "00:58:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First dry it off with the paper towel in front of you, then dip it in water.", + "B": "Dip it in water, then dry it.", + "C": "Dry it with the towel behind you.", + "D": "First dry it off with the paper towel to your left, then dip it in water to rinse it.", + "E": "First dry it off with the paper towel to your right, then dip it in water and swirl it gently." + }, + "answer_label": "A", + "answer_text": "First dry it off with the paper towel in front of you, then dip it in water.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_8_T1", + "correct_answer": "First dry it off with the paper towel in front of you, then dip it in water.", + "original_options": { + "A": "First dry it off with the paper towel in front of you, then dip it in water.", + "B": "Dip it in water, then dry it.", + "C": "Dry it with the towel behind you.", + "D": "First dry it off with the paper towel to your left, then dip it in water to rinse it.", + "E": "First dry it off with the paper towel to your right, then dip it in water and swirl it gently." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 20 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_5_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I draw the house?", + "answer": "You first drew a wall on the paper in front of you.", + "question_chinese": "我是怎么画房子的？", + "answer_chinese": "你先在面前的纸上画了墙壁", + "evidence": [ + { + "description": "paint the walls", + "timestamps": { + "start": "00:04:12", + "end": "00:04:13" + } + } + ], + "question_timestamps": "00:04:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You first drew the roof.", + "B": "You first drew a wall behind you.", + "C": "You first drew a wall on the paper to your left before adding any other parts of the house.", + "D": "You first drew a wall on the paper in front of you.", + "E": "You first drew a wall on the paper to your right, then moved on to the next section of the house." + }, + "answer_label": "D", + "answer_text": "You first drew a wall on the paper in front of you.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_5_T1", + "correct_answer": "You first drew a wall on the paper in front of you.", + "original_options": { + "A": "You first drew the roof.", + "B": "You first drew a wall behind you.", + "C": "You first drew a wall on the paper to your left before adding any other parts of the house.", + "D": "You first drew a wall on the paper in front of you.", + "E": "You first drew a wall on the paper to your right, then moved on to the next section of the house." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 19, + 21 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_13", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I draw a house?", + "answer": "First paint the walls, then paint the roof.", + "question_chinese": "我是怎么画房子的？", + "answer_chinese": "先画墙壁，再画屋顶", + "evidence": [ + { + "description": "paint the walls", + "timestamps": { + "start": "00:04:12", + "end": "00:04:13" + } + }, + { + "description": "paint the roof", + "timestamps": { + "start": "00:09:20", + "end": "00:09:21" + } + } + ], + "question_timestamps": "00:09:22", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "First paint the roof, then walls.", + "B": "First paint the windows, then door.", + "C": "First paint the roof, then paint the walls afterward to fill in the main shape of the house.", + "D": "First paint the walls, then paint the roof.", + "E": "First paint the door and windows, then add the roof once the details are already in place." + }, + "answer_label": "D", + "answer_text": "First paint the walls, then paint the roof.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_13", + "correct_answer": "First paint the walls, then paint the roof.", + "original_options": { + "A": "First paint the roof, then walls.", + "B": "First paint the windows, then door.", + "C": "First paint the roof, then paint the walls afterward to fill in the main shape of the house.", + "D": "First paint the walls, then paint the roof.", + "E": "First paint the door and windows, then add the roof once the details are already in place." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_5_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I draw the house?", + "answer": "You first drew a wall on the paper in front of you, and then drew the roof.", + "question_chinese": "我是怎么画房子的？", + "answer_chinese": "你先在面前的纸上画了墙壁，再画屋顶", + "evidence": [ + { + "description": "paint the walls", + "timestamps": { + "start": "00:04:12", + "end": "00:04:13" + } + }, + { + "description": "paint the roof", + "timestamps": { + "start": "00:09:20", + "end": "00:09:21" + } + } + ], + "question_timestamps": "00:09:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You first drew a wall on the paper in front of you, and then drew the roof.", + "B": "You drew the roof first.", + "C": "You drew the wall behind you first.", + "D": "You first drew a wall on the paper to your left, and then drew the roof on top of it.", + "E": "You first drew a wall on the paper to your right, and then drew the roof to complete the outline." + }, + "answer_label": "A", + "answer_text": "You first drew a wall on the paper in front of you, and then drew the roof.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_5_T2", + "correct_answer": "You first drew a wall on the paper in front of you, and then drew the roof.", + "original_options": { + "A": "You first drew a wall on the paper in front of you, and then drew the roof.", + "B": "You drew the roof first.", + "C": "You drew the wall behind you first.", + "D": "You first drew a wall on the paper to your left, and then drew the roof on top of it.", + "E": "You first drew a wall on the paper to your right, and then drew the roof to complete the outline." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 20, + 20 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_6_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I mix the colors?", + "answer": "You first dipped into the white paint in front of you.", + "question_chinese": "我是怎么调色的？", + "answer_chinese": "你先蘸取了面前的白色", + "evidence": [ + { + "description": "dab with white", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:03:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You dipped into white paint behind you.", + "B": "You dipped into white paint to your left.", + "C": "You first dipped into the white paint in front of you.", + "D": "You first dipped into the white paint to your right before mixing any other color into it.", + "E": "You first dipped into the white paint above you, then continued mixing after loading the brush." + }, + "answer_label": "C", + "answer_text": "You first dipped into the white paint in front of you.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_6_T1", + "correct_answer": "You first dipped into the white paint in front of you.", + "original_options": { + "A": "You dipped into white paint behind you.", + "B": "You dipped into white paint to your left.", + "C": "You first dipped into the white paint in front of you.", + "D": "You first dipped into the white paint to your right before mixing any other color into it.", + "E": "You first dipped into the white paint above you, then continued mixing after loading the brush." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 16 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_15", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I color grade?", + "answer": "First dab with white, then mix with other colors.", + "question_chinese": "我是怎么调色的？", + "answer_chinese": "先蘸取白色，再混合其他颜色", + "evidence": [ + { + "description": "dab with white", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + }, + { + "description": "mix with other colors", + "timestamps": { + "start": "00:03:53", + "end": "00:03:54" + } + } + ], + "question_timestamps": "00:03:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "First dab with white, then mix with other colors.", + "B": "First dab with black, then mix.", + "C": "First mix colors, then add white.", + "D": "First dab with a darker color, then mix in white and other colors to lighten the tone gradually.", + "E": "First blend other colors together, then dab in white at the end to soften and brighten the mix." + }, + "answer_label": "A", + "answer_text": "First dab with white, then mix with other colors.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_15", + "correct_answer": "First dab with white, then mix with other colors.", + "original_options": { + "A": "First dab with white, then mix with other colors.", + "B": "First dab with black, then mix.", + "C": "First mix colors, then add white.", + "D": "First dab with a darker color, then mix in white and other colors to lighten the tone gradually.", + "E": "First blend other colors together, then dab in white at the end to soften and brighten the mix." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_6_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I mix the colors?", + "answer": "You first dipped into the white paint in front of you, and then mixed in other colors.", + "question_chinese": "我是怎么调色的？", + "answer_chinese": "你先蘸取了面前的白色，再混合其他颜色", + "evidence": [ + { + "description": "dab with white", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + }, + { + "description": "mix with other colors.", + "timestamps": { + "start": "00:03:53", + "end": "00:03:54" + } + } + ], + "question_timestamps": "00:03:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You started with paint behind you.", + "B": "You first dipped into the white paint in front of you, and then mixed in other colors.", + "C": "You began with the paint on your left.", + "D": "You dipped into the white paint to your right first, then blended in the other colors afterward.", + "E": "You dipped into a different paint well first, then added white later while mixing the remaining colors." + }, + "answer_label": "B", + "answer_text": "You first dipped into the white paint in front of you, and then mixed in other colors.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_6_T2", + "correct_answer": "You first dipped into the white paint in front of you, and then mixed in other colors.", + "original_options": { + "A": "You started with paint behind you.", + "B": "You first dipped into the white paint in front of you, and then mixed in other colors.", + "C": "You began with the paint on your left.", + "D": "You dipped into the white paint to your right first, then blended in the other colors afterward.", + "E": "You dipped into a different paint well first, then added white later while mixing the remaining colors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 17, + 17 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_3", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my phone?", + "answer": "On my right.", + "question_chinese": "我的手机在哪里？", + "answer_chinese": "我的右手边", + "evidence": [ + { + "description": "on my right", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On my left.", + "B": "On my right.", + "C": "In front of me.", + "D": "Behind your back, not at your side where you can easily reach it.", + "E": "On the table in front of you, rather than being positioned on your right side." + }, + "answer_label": "B", + "answer_text": "On my right.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_3", + "correct_answer": "On my right.", + "original_options": { + "A": "On my left.", + "B": "On my right.", + "C": "In front of me.", + "D": "Behind your back, not at your side where you can easily reach it.", + "E": "On the table in front of you, rather than being positioned on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 15 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_1_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my phone?", + "answer": "On your right-hand side.", + "question_chinese": "我的手机在哪里？", + "answer_chinese": "你的右手边", + "evidence": [ + { + "description": "on my right", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left-hand side.", + "B": "In front of you.", + "C": "On your right-hand side.", + "D": "Behind you on your left-hand side, not positioned on your right-hand side.", + "E": "Straight ahead of you rather than to either side, so it isn’t on your right-hand side." + }, + "answer_label": "C", + "answer_text": "On your right-hand side.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_1_T1", + "correct_answer": "On your right-hand side.", + "original_options": { + "A": "To your left-hand side.", + "B": "In front of you.", + "C": "On your right-hand side.", + "D": "Behind you on your left-hand side, not positioned on your right-hand side.", + "E": "Straight ahead of you rather than to either side, so it isn’t on your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 17 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_1_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my phone?", + "answer": "On the table in front of you.", + "question_chinese": "我的手机在哪里？", + "answer_chinese": "在你面前的桌子上", + "evidence": [ + { + "description": "on my right", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the table behind you.", + "B": "On the table to your left.", + "C": "On the table in front of you.", + "D": "On the table to your right, not on the surface directly in front of you.", + "E": "On a table behind you rather than the one in front, so you’d need to turn around to see it." + }, + "answer_label": "C", + "answer_text": "On the table in front of you.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_1_T2", + "correct_answer": "On the table in front of you.", + "original_options": { + "A": "On the table behind you.", + "B": "On the table to your left.", + "C": "On the table in front of you.", + "D": "On the table to your right, not on the surface directly in front of you.", + "E": "On a table behind you rather than the one in front, so you’d need to turn around to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 21 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color did I dip into?", + "answer": "White.", + "question_chinese": "我蘸取了什么颜色？", + "answer_chinese": "白色", + "evidence": [ + { + "description": "white", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Red.", + "B": "Blue.", + "C": "You dipped into yellow paint, not white, before continuing with the palette.", + "D": "White.", + "E": "You dipped into green paint rather than white when you touched the brush to the palette." + }, + "answer_label": "D", + "answer_text": "White.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_1", + "correct_answer": "White.", + "original_options": { + "A": "Red.", + "B": "Blue.", + "C": "You dipped into yellow paint, not white, before continuing with the palette.", + "D": "White.", + "E": "You dipped into green paint rather than white when you touched the brush to the palette." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 16 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color did I dip into?", + "answer": "White.", + "question_chinese": "我蘸取了什么颜色？", + "answer_chinese": "白色", + "evidence": [ + { + "description": "White", + "timestamps": { + "start": "00:00:21", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Blue.", + "B": "Yellow.", + "C": "White.", + "D": "You dipped into red paint rather than white when you first touched the brush to the palette.", + "E": "You dipped into green paint instead of white, choosing a darker color from the palette." + }, + "answer_label": "C", + "answer_text": "White.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_2", + "correct_answer": "White.", + "original_options": { + "A": "Blue.", + "B": "Yellow.", + "C": "White.", + "D": "You dipped into red paint rather than white when you first touched the brush to the palette.", + "E": "You dipped into green paint instead of white, choosing a darker color from the palette." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 15 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_17", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many colors did I dip into?", + "answer": "Three.", + "question_chinese": "我蘸取了多少种颜色？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:07:46", + "end": "00:07:47" + } + } + ], + "question_timestamps": "00:07:48", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "You dipped into one color total, using the same paint without switching to any other colors.", + "E": "You dipped into five colors, moving the brush through multiple wells before finishing the mix." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_17", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "You dipped into one color total, using the same paint without switching to any other colors.", + "E": "You dipped into five colors, moving the brush through multiple wells before finishing the mix." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_4_T1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pens are in the water box on the table in front of me?", + "answer": "There were three pens in the water cup at the upper-left area in front of you.", + "question_chinese": "我面前的桌上的水盒里面有多少支笔？", + "answer_chinese": "你面前的左上的水盒里有三支笔", + "evidence": [ + { + "description": "there are two pens", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "there are three pens", + "timestamps": { + "start": "00:52:33", + "end": "00:52:34" + } + } + ], + "question_timestamps": "00:52:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There were two pens.", + "B": "There were four pens.", + "C": "There were three pens in the water cup at the upper-left area in front of you.", + "D": "There was one pen in the water cup at the upper-left area in front of you.", + "E": "There were five pens in the water cup at the upper-left area in front of you." + }, + "answer_label": "C", + "answer_text": "There were three pens in the water cup at the upper-left area in front of you.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_4_T1", + "correct_answer": "There were three pens in the water cup at the upper-left area in front of you.", + "original_options": { + "A": "There were two pens.", + "B": "There were four pens.", + "C": "There were three pens in the water cup at the upper-left area in front of you.", + "D": "There was one pen in the water cup at the upper-left area in front of you.", + "E": "There were five pens in the water cup at the upper-left area in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_4_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pens are in the water box on the table in front of me?", + "answer": "There were four pens in the water cup on the table in front of you.", + "question_chinese": "我面前的桌上的水盒里面有多少支笔？", + "answer_chinese": "你面前的桌上的水盒里面有四支笔", + "evidence": [ + { + "description": "there are two pens", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "there are three pens", + "timestamps": { + "start": "00:52:33", + "end": "00:52:34" + } + }, + { + "description": "there are four pens", + "timestamps": { + "start": "01:05:15", + "end": "01:05:16" + } + } + ], + "question_timestamps": "01:05:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There were three pens.", + "B": "There were five pens.", + "C": "There were two pens in the water cup on the table in front of you.", + "D": "There were four pens in the water cup on the table in front of you.", + "E": "There were six pens in the water cup on the table in front of you." + }, + "answer_label": "D", + "answer_text": "There were four pens in the water cup on the table in front of you.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_4_T2", + "correct_answer": "There were four pens in the water cup on the table in front of you.", + "original_options": { + "A": "There were three pens.", + "B": "There were five pens.", + "C": "There were two pens in the water cup on the table in front of you.", + "D": "There were four pens in the water cup on the table in front of you.", + "E": "There were six pens in the water cup on the table in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_2_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many paintings are on the tabletop in front of me?", + "answer": "There was one painting on the tabletop in front of you.", + "question_chinese": "有多少幅画在我面前的的桌面上？", + "answer_chinese": "有一幅画在你面前的桌面上", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:04", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There was one painting on the tabletop in front of you.", + "B": "There were zero paintings.", + "C": "There were two paintings.", + "D": "There were three paintings laid out on the tabletop in front of you, not just one.", + "E": "There were four paintings spread across the tabletop in front of you, taking up most of the space." + }, + "answer_label": "A", + "answer_text": "There was one painting on the tabletop in front of you.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_2_T1", + "correct_answer": "There was one painting on the tabletop in front of you.", + "original_options": { + "A": "There was one painting on the tabletop in front of you.", + "B": "There were zero paintings.", + "C": "There were two paintings.", + "D": "There were three paintings laid out on the tabletop in front of you, not just one.", + "E": "There were four paintings spread across the tabletop in front of you, taking up most of the space." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 18 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_2_T2", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many paintings are on the tabletop in front of me?", + "answer": "There were three paintings on the tabletop in front of you.", + "question_chinese": "有多少幅画在我面前的的桌面上？", + "answer_chinese": "有三幅画在你面前的桌面上", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:04", + "end": "00:01:05" + } + }, + { + "description": "another two paintings", + "timestamps": { + "start": "00:04:46", + "end": "00:04:47" + } + } + ], + "question_timestamps": "00:04:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There were two paintings.", + "B": "There were three paintings on the tabletop in front of you.", + "C": "There were four paintings.", + "D": "There was one painting on the tabletop in front of you, not three.", + "E": "There were five paintings on the tabletop in front of you, covering much more of the table." + }, + "answer_label": "B", + "answer_text": "There were three paintings on the tabletop in front of you.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_2_T2", + "correct_answer": "There were three paintings on the tabletop in front of you.", + "original_options": { + "A": "There were two paintings.", + "B": "There were three paintings on the tabletop in front of you.", + "C": "There were four paintings.", + "D": "There was one painting on the tabletop in front of you, not three.", + "E": "There were five paintings on the tabletop in front of you, covering much more of the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 17 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_11", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pens are in the water box?", + "answer": "Four.", + "question_chinese": "水盒里面有多少支笔？", + "answer_chinese": "4", + "evidence": [ + { + "description": "there are two pens", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "there are three pens", + "timestamps": { + "start": "00:52:33", + "end": "00:52:34" + } + }, + { + "description": "there are four pens", + "timestamps": { + "start": "01:05:14", + "end": "01:05:15" + } + } + ], + "question_timestamps": "01:05:16", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "Two pens were in the water box instead of four, based on what you saw.", + "E": "Six pens were in the water box, filling it more than expected." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_11", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "Two pens were in the water box instead of four, based on what you saw.", + "E": "Six pens were in the water box, filling it more than expected." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 12 + ] + }, + { + "video_uid": "79b9f7e3-f1ce-4838-9887-1deeba08e600", + "video_length": "01:07:02", + "qid": "79b9f7e3-f1ce-4838-9887-1deeba08e600_12", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pens are in the water box?", + "answer": "Four.", + "question_chinese": "水盒里面有多少支笔？", + "answer_chinese": "4", + "evidence": [ + { + "description": "there are two pens", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "there are three pens", + "timestamps": { + "start": "00:52:33", + "end": "00:52:34" + } + }, + { + "description": "there are four pens", + "timestamps": { + "start": "01:05:15", + "end": "01:05:16" + } + } + ], + "question_timestamps": "01:05:45", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "Two pens were in the water box instead of four, based on what you saw.", + "E": "Six pens were in the water box, filling it more than expected." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "79b9f7e3-f1ce-4838-9887-1deeba08e600_12", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "Two pens were in the water box instead of four, based on what you saw.", + "E": "Six pens were in the water box, filling it more than expected." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 12 + ] + } +] diff --git a/QAs/7bc2ea44-9168-4379-a17e-5558d4e2fc24-mcq.json b/QAs/7bc2ea44-9168-4379-a17e-5558d4e2fc24-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a85a0fd3674610bbbb856125bf5d6c4b720f3df4 --- /dev/null +++ b/QAs/7bc2ea44-9168-4379-a17e-5558d4e2fc24-mcq.json @@ -0,0 +1,669 @@ +[ + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_4", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I previously heated the hot water kettle. How far is the faucet from the heated hot water kettle?", + "answer": "The hot water kettle is being heated and is about one meter horizontally from the faucet.", + "question_chinese": "我之前把热水壶进行加热，水龙头距离加热的热水壶有多远", + "answer_chinese": "热水壶正在进行加热，离水龙头约1米，水平距离", + "evidence": [ + { + "description": "heat the hot water kettle", + "timestamps": { + "start": "00:17:52", + "end": "00:17:56" + } + } + ], + "question_timestamps": "00:18:14", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "The hot water kettle is being heated and is about one meter horizontally from the faucet.", + "B": "About half a meter away.", + "C": "About two meters away.", + "D": "The heated hot water kettle sits roughly three meters from the faucet, farther than an arm’s reach.", + "E": "The heated hot water kettle is about one and a half meters from the faucet, not just one meter." + }, + "answer_label": "A", + "answer_text": "The hot water kettle is being heated and is about one meter horizontally from the faucet.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_4", + "correct_answer": "The hot water kettle is being heated and is about one meter horizontally from the faucet.", + "original_options": { + "A": "The hot water kettle is being heated and is about one meter horizontally from the faucet.", + "B": "About half a meter away.", + "C": "About two meters away.", + "D": "The heated hot water kettle sits roughly three meters from the faucet, farther than an arm’s reach.", + "E": "The heated hot water kettle is about one and a half meters from the faucet, not just one meter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 19 + ] + }, + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_14", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I previously heated the hot water kettle. How far is the faucet from the heated hot water kettle?", + "answer": "The hot water kettle is being heated and is about one meter horizontally from the faucet.", + "question_chinese": "我之前把热水壶进行加热，水龙头距离加热的热水壶有多远", + "answer_chinese": "热水壶正在进行加热，离水龙头约1米，水平距离", + "evidence": [ + { + "description": "heat the hot water kettle", + "timestamps": { + "start": "00:17:52", + "end": "00:17:56" + } + } + ], + "question_timestamps": "00:44:27", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "The hot water kettle is being heated and is about one meter horizontally from the faucet.", + "B": "About two meters from the faucet.", + "C": "About half a meter from the faucet.", + "D": "The kettle is being heated and sits about two meters horizontally away from the faucet.", + "E": "The kettle is being heated and is roughly one meter away, but diagonally rather than horizontally from the faucet." + }, + "answer_label": "A", + "answer_text": "The hot water kettle is being heated and is about one meter horizontally from the faucet.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_14", + "correct_answer": "The hot water kettle is being heated and is about one meter horizontally from the faucet.", + "original_options": { + "A": "The hot water kettle is being heated and is about one meter horizontally from the faucet.", + "B": "About two meters from the faucet.", + "C": "About half a meter from the faucet.", + "D": "The kettle is being heated and sits about two meters horizontally away from the faucet.", + "E": "The kettle is being heated and is roughly one meter away, but diagonally rather than horizontally from the faucet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 19 + ] + }, + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_7_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was mopping the floor just now, could I directly reach the leather chair?", + "answer": "You can. The leather chair is next to you.", + "question_chinese": "我刚刚拖地的时候，可以直接接触到皮质椅子吗？", + "answer_chinese": "你可以。皮质椅子在你的旁边。", + "evidence": [ + { + "description": "you are close to the leather chair.", + "timestamps": { + "start": "00:26:22", + "end": "00:26:45" + } + } + ], + "question_timestamps": "00:26:47", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t. The leather chair is behind you, several meters away.", + "B": "You can. The leather chair is next to you." + }, + "answer_label": "B", + "answer_text": "You can. The leather chair is next to you.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_7_T2", + "correct_answer": "You can. The leather chair is next to you.", + "original_options": { + "A": "You can’t. The leather chair is behind you, several meters away.", + "B": "You can. The leather chair is next to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_7_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was mopping the floor just now, could I directly reach the leather chair?", + "answer": "You can’t. The leather chair is behind you, several meters away.", + "question_chinese": "我刚刚拖地的时候，可以直接接触到皮质椅子吗？", + "answer_chinese": "你不可以。皮质椅子在你的后面，距离你好几米。", + "evidence": [ + { + "description": "You are far form the leather chair.", + "timestamps": { + "start": "00:27:04", + "end": "00:27:16" + } + } + ], + "question_timestamps": "00:27:25", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t. The leather chair is behind you, several meters away.", + "B": "You can. The leather chair is next to you." + }, + "answer_label": "A", + "answer_text": "You can’t. The leather chair is behind you, several meters away.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_7_T1", + "correct_answer": "You can’t. The leather chair is behind you, several meters away.", + "original_options": { + "A": "You can’t. The leather chair is behind you, several meters away.", + "B": "You can. The leather chair is next to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was mopping just now, how many leather chairs did I see?", + "answer": "You see three leather stools.", + "question_chinese": "我刚刚拖地的时候，看到了几个皮质椅子", + "answer_chinese": "我看到了三个皮质凳子。", + "evidence": [ + { + "description": "saw three leather stools", + "timestamps": { + "start": "00:26:51", + "end": "00:27:00" + } + } + ], + "question_timestamps": "00:27:34", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You see two leather stools.", + "B": "You see four leather stools.", + "C": "You see one leather stool when you’re mopping in that area.", + "D": "You see three leather stools.", + "E": "You see five leather stools grouped together as you mop the floor." + }, + "answer_label": "D", + "answer_text": "You see three leather stools.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_9", + "correct_answer": "You see three leather stools.", + "original_options": { + "A": "You see two leather stools.", + "B": "You see four leather stools.", + "C": "You see one leather stool when you’re mopping in that area.", + "D": "You see three leather stools.", + "E": "You see five leather stools grouped together as you mop the floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_19", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was mopping just now, how many leather chairs did I see?", + "answer": "You see three leather stools.", + "question_chinese": "我刚刚拖地的时候，看到了几个皮质椅子", + "answer_chinese": "我看到了三个皮质凳子。", + "evidence": [ + { + "description": "saw three leather stools", + "timestamps": { + "start": "00:26:51", + "end": "00:27:00" + } + } + ], + "question_timestamps": "00:32:01", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You see two leather stools.", + "B": "You see four leather stools.", + "C": "You see five leather stools while you’re mopping nearby on the floor.", + "D": "You see three leather stools.", + "E": "You see six leather stools lined up in the area you were mopping." + }, + "answer_label": "D", + "answer_text": "You see three leather stools.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_19", + "correct_answer": "You see three leather stools.", + "original_options": { + "A": "You see two leather stools.", + "B": "You see four leather stools.", + "C": "You see five leather stools while you’re mopping nearby on the floor.", + "D": "You see three leather stools.", + "E": "You see six leather stools lined up in the area you were mopping." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many slices of lemon juice did I squeeze into the hot water kettle?", + "answer": "You just squeezed the juice of two lemon slices into the hot water kettle.", + "question_chinese": "我挤了几块柠檬的柠檬汁到热水壶中", + "answer_chinese": "你刚刚挤了两块柠檬的柠檬汁到热水壶中", + "evidence": [ + { + "description": "squeeze lemon juice into the hot water kettle", + "timestamps": { + "start": "00:16:52", + "end": "00:17:03" + } + }, + { + "description": "cut two slices of lemon", + "timestamps": { + "start": "00:16:23", + "end": "00:16:31" + } + } + ], + "question_timestamps": "00:17:26", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You just squeezed the juice of two lemon slices into the hot water kettle.", + "B": "You squeezed one lemon slice.", + "C": "You squeezed three lemon slices.", + "D": "You just squeezed the juice of four lemon slices into the hot water kettle.", + "E": "You just squeezed the juice of five lemon slices into the hot water kettle before setting it down." + }, + "answer_label": "A", + "answer_text": "You just squeezed the juice of two lemon slices into the hot water kettle.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_5", + "correct_answer": "You just squeezed the juice of two lemon slices into the hot water kettle.", + "original_options": { + "A": "You just squeezed the juice of two lemon slices into the hot water kettle.", + "B": "You squeezed one lemon slice.", + "C": "You squeezed three lemon slices.", + "D": "You just squeezed the juice of four lemon slices into the hot water kettle.", + "E": "You just squeezed the juice of five lemon slices into the hot water kettle before setting it down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 18 + ] + }, + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_15", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many slices of lemon juice did I squeeze into the hot water kettle?", + "answer": "You just squeezed the juice of two lemon slices into the hot water kettle.", + "question_chinese": "我挤了几块柠檬的柠檬汁到热水壶中", + "answer_chinese": "你刚刚挤了两块柠檬的柠檬汁到热水壶中", + "evidence": [ + { + "description": "squeeze lemon juice into the hot water kettle", + "timestamps": { + "start": "00:16:52", + "end": "00:17:03" + } + }, + { + "description": "cut two slices of lemon", + "timestamps": { + "start": "00:16:23", + "end": "00:16:31" + } + } + ], + "question_timestamps": "00:43:55", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You just squeezed the juice of two lemon slices into the hot water kettle.", + "B": "You squeezed one lemon slice.", + "C": "You squeezed three lemon slices.", + "D": "You just squeezed the juice of four lemon slices into the hot water kettle.", + "E": "You just squeezed the juice of five lemon slices into the hot water kettle right after cutting them." + }, + "answer_label": "A", + "answer_text": "You just squeezed the juice of two lemon slices into the hot water kettle.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_15", + "correct_answer": "You just squeezed the juice of two lemon slices into the hot water kettle.", + "original_options": { + "A": "You just squeezed the juice of two lemon slices into the hot water kettle.", + "B": "You squeezed one lemon slice.", + "C": "You squeezed three lemon slices.", + "D": "You just squeezed the juice of four lemon slices into the hot water kettle.", + "E": "You just squeezed the juice of five lemon slices into the hot water kettle right after cutting them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 18 + ] + }, + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_7", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the lemon take from the refrigerator to the kitchen table?", + "answer": "The lemon's path from the refrigerator to the kitchen table was: you took the lemon out of the refrigerator, passed through the middle of the room, and placed it above the bowl.", + "question_chinese": "柠檬从冰箱到厨房桌经过了什么路径", + "answer_chinese": "柠檬从冰箱到厨房桌的路径是：你把柠檬从冰箱里拿出来，穿过房间中央，然后把它放在碗上方", + "evidence": [ + { + "description": "take the lemon out of the refrigerator", + "timestamps": { + "start": "00:16:12", + "end": "00:16:16" + } + }, + { + "description": "pass through the middle of the room", + "timestamps": { + "start": "00:16:16", + "end": "00:16:18" + } + }, + { + "description": "place it above the bowl", + "timestamps": { + "start": "00:16:18", + "end": "00:16:23" + } + } + ], + "question_timestamps": "00:20:52", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The lemon's path from the refrigerator to the kitchen table was: you took the lemon out of the refrigerator, passed through the middle of the room, and placed it above the bowl.", + "B": "You carried it along the wall.", + "C": "You set it beside the cutting board.", + "D": "You took the lemon from the refrigerator, walked past the counter, and placed it next to a plate.", + "E": "You removed the lemon from the refrigerator, went through the doorway, and set it near the fruit bowl." + }, + "answer_label": "A", + "answer_text": "The lemon's path from the refrigerator to the kitchen table was: you took the lemon out of the refrigerator, passed through the middle of the room, and placed it above the bowl.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_7", + "correct_answer": "The lemon's path from the refrigerator to the kitchen table was: you took the lemon out of the refrigerator, passed through the middle of the room, and placed it above the bowl.", + "original_options": { + "A": "The lemon's path from the refrigerator to the kitchen table was: you took the lemon out of the refrigerator, passed through the middle of the room, and placed it above the bowl.", + "B": "You carried it along the wall.", + "C": "You set it beside the cutting board.", + "D": "You took the lemon from the refrigerator, walked past the counter, and placed it next to a plate.", + "E": "You removed the lemon from the refrigerator, went through the doorway, and set it near the fruit bowl." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + }, + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_17", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the lemon take from the refrigerator to the kitchen table?", + "answer": "The lemon's path from the refrigerator to the kitchen table was: you took the lemon out of the refrigerator, passed through the middle of the room, and placed it above the bowl.", + "question_chinese": "柠檬从冰箱到厨房桌经过了什么路径", + "answer_chinese": "柠檬从冰箱到厨房桌的路径是：你把柠檬从冰箱里拿出来，穿过房间中央，然后把它放在碗上方", + "evidence": [ + { + "description": "take the lemon out of the refrigerator", + "timestamps": { + "start": "00:16:12", + "end": "00:16:16" + } + }, + { + "description": "pass through the middle of the room", + "timestamps": { + "start": "00:16:16", + "end": "00:16:18" + } + }, + { + "description": "place it above the bowl", + "timestamps": { + "start": "00:16:18", + "end": "00:16:23" + } + } + ], + "question_timestamps": "00:33:14", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You walked along the room’s edge.", + "B": "You went around the corner first.", + "C": "You took the lemon from the refrigerator, crossed near the window, and set it beside the bowl.", + "D": "You removed the lemon from the refrigerator, cut through the room’s corner, and placed it on the table near the bowl.", + "E": "The lemon's path from the refrigerator to the kitchen table was: you took the lemon out of the refrigerator, passed through the middle of the room, and placed it above the bowl." + }, + "answer_label": "E", + "answer_text": "The lemon's path from the refrigerator to the kitchen table was: you took the lemon out of the refrigerator, passed through the middle of the room, and placed it above the bowl.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_17", + "correct_answer": "The lemon's path from the refrigerator to the kitchen table was: you took the lemon out of the refrigerator, passed through the middle of the room, and placed it above the bowl.", + "original_options": { + "A": "You walked along the room’s edge.", + "B": "You went around the corner first.", + "C": "You took the lemon from the refrigerator, crossed near the window, and set it beside the bowl.", + "D": "You removed the lemon from the refrigerator, cut through the room’s corner, and placed it on the table near the bowl.", + "E": "The lemon's path from the refrigerator to the kitchen table was: you took the lemon out of the refrigerator, passed through the middle of the room, and placed it above the bowl." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 22 + ] + }, + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_11", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the bucket placed in the room?", + "answer": "The bucket was placed next to the stairs in the living room.", + "question_chinese": "水桶放在了房间的哪里了", + "answer_chinese": "水桶放在了客厅的楼梯旁边。", + "evidence": [ + { + "description": "bucket is next to the stairs in the living room", + "timestamps": { + "start": "00:25:48", + "end": "00:25:53" + } + } + ], + "question_timestamps": "00:26:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was behind the sofa.", + "B": "It was across from the fireplace.", + "C": "The bucket was placed in the living room, under the window, away from the stairs.", + "D": "You left the bucket in the living room near the doorway, not beside the stairs.", + "E": "The bucket was placed next to the stairs in the living room." + }, + "answer_label": "E", + "answer_text": "The bucket was placed next to the stairs in the living room.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_11", + "correct_answer": "The bucket was placed next to the stairs in the living room.", + "original_options": { + "A": "It was behind the sofa.", + "B": "It was across from the fireplace.", + "C": "The bucket was placed in the living room, under the window, away from the stairs.", + "D": "You left the bucket in the living room near the doorway, not beside the stairs.", + "E": "The bucket was placed next to the stairs in the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24", + "video_length": "00:44:27", + "qid": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_12", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the bucket placed in the room?", + "answer": "The bucket was placed between the cabinet and the kitchen table in the living room.", + "question_chinese": "水桶放在了房间的哪里了", + "answer_chinese": "水桶放在了客厅的柜子和厨桌中间", + "evidence": [ + { + "description": "move the bucket", + "timestamps": { + "start": "00:29:41", + "end": "00:29:46" + } + }, + { + "description": "saw the bucket", + "timestamps": { + "start": "00:30:00", + "end": "00:30:02" + } + } + ], + "question_timestamps": "00:30:21", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was behind the sofa.", + "B": "It was near the window.", + "C": "The bucket was placed between the cabinet and the kitchen table in the living room.", + "D": "The bucket was placed in the living room between the sofa and the coffee table, not by the cabinet.", + "E": "You put the bucket in the living room by the shelf and side table, rather than between the cabinet and table." + }, + "answer_label": "C", + "answer_text": "The bucket was placed between the cabinet and the kitchen table in the living room.", + "q_id": "7bc2ea44-9168-4379-a17e-5558d4e2fc24_12", + "correct_answer": "The bucket was placed between the cabinet and the kitchen table in the living room.", + "original_options": { + "A": "It was behind the sofa.", + "B": "It was near the window.", + "C": "The bucket was placed between the cabinet and the kitchen table in the living room.", + "D": "The bucket was placed in the living room between the sofa and the coffee table, not by the cabinet.", + "E": "You put the bucket in the living room by the shelf and side table, rather than between the cabinet and table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 21 + ] + } +] diff --git a/QAs/7c0c0057-3bbc-40ef-a91e-4d6e1a637642-mcq.json b/QAs/7c0c0057-3bbc-40ef-a91e-4d6e1a637642-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ae7fde75014ffbcf42b4a1e9fe93e8828526185f --- /dev/null +++ b/QAs/7c0c0057-3bbc-40ef-a91e-4d6e1a637642-mcq.json @@ -0,0 +1,796 @@ +[ + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_4_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I just used the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, where on the table did I put the scissors?", + "answer": "After you finished using the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, you placed the scissors on the second of the three stacked cardboard boxes on the table, behind you.", + "question_chinese": "我刚刚用完剪刀开完装有木板的纸盒的塑料膜后，我把剪刀放在桌子上哪里？", + "answer_chinese": "你用完剪刀开完装有木板的纸盒的塑料膜后，你把剪刀放在了桌子上叠放的三个纸盒的第二个纸盒上，在你的后边。", + "evidence": [ + { + "description": "After using the scissors to open the plastic film of the cardboard box containing the wooden board, I placed the scissors on the second of the three stacked cardboard boxes on the table.", + "timestamps": { + "start": "00:03:52", + "end": "00:03:54" + } + } + ], + "question_timestamps": "00:04:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the top box behind you.", + "B": "On the bottom box behind you.", + "C": "You set the scissors on the first of three stacked cardboard boxes on the table behind you.", + "D": "After you finished using the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, you placed the scissors on the second of the three stacked cardboard boxes on the table, behind you.", + "E": "After cutting the plastic wrap, you placed the scissors on the third of three stacked boxes behind you." + }, + "answer_label": "D", + "answer_text": "After you finished using the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, you placed the scissors on the second of the three stacked cardboard boxes on the table, behind you.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_4_T2", + "correct_answer": "After you finished using the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, you placed the scissors on the second of the three stacked cardboard boxes on the table, behind you.", + "original_options": { + "A": "On the top box behind you.", + "B": "On the bottom box behind you.", + "C": "You set the scissors on the first of three stacked cardboard boxes on the table behind you.", + "D": "After you finished using the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, you placed the scissors on the second of the three stacked cardboard boxes on the table, behind you.", + "E": "After cutting the plastic wrap, you placed the scissors on the third of three stacked boxes behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_4_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I just used the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, where on the table did I put the scissors?", + "answer": "After you finished using the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, you placed the scissors on the second of the three stacked cardboard boxes on the table, to your back-right.", + "question_chinese": "我刚刚用完剪刀开完装有木板的纸盒的塑料膜后，我把剪刀放在桌子上哪里？", + "answer_chinese": "你用完剪刀开完装有木板的纸盒的塑料膜后，你把剪刀放在了桌子上叠放的三个纸盒的第二个纸盒上，在你的右后方。", + "evidence": [ + { + "description": "After using the scissors to open the plastic film of the cardboard box containing the wooden board, I placed the scissors on the second of the three stacked cardboard boxes on the table.", + "timestamps": { + "start": "00:03:52", + "end": "00:03:54" + } + } + ], + "question_timestamps": "00:14:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the top box, back-right.", + "B": "On the bottom box, back-right.", + "C": "After you finished using the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, you placed the scissors on the second of the three stacked cardboard boxes on the table, to your back-right.", + "D": "You placed the scissors on the second of the three stacked cardboard boxes on the table, to your back-left.", + "E": "After cutting the plastic wrap, you set the scissors on the second box of the stack, to your front-right." + }, + "answer_label": "C", + "answer_text": "After you finished using the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, you placed the scissors on the second of the three stacked cardboard boxes on the table, to your back-right.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_4_T1", + "correct_answer": "After you finished using the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, you placed the scissors on the second of the three stacked cardboard boxes on the table, to your back-right.", + "original_options": { + "A": "On the top box, back-right.", + "B": "On the bottom box, back-right.", + "C": "After you finished using the scissors to cut open the plastic wrap on the cardboard box containing wooden boards, you placed the scissors on the second of the three stacked cardboard boxes on the table, to your back-right.", + "D": "You placed the scissors on the second of the three stacked cardboard boxes on the table, to your back-left.", + "E": "After cutting the plastic wrap, you set the scissors on the second box of the stack, to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_6_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, before I put down the wooden board in my hand and helped my partner hold down the wooden board on the ground, how many items did I pick up from the ground?", + "answer": "Earlier, before you set down the wooden board in your hands and helped your teammate hold down the board on the ground, you picked up four items from the floor.", + "question_chinese": "我刚才在放下手中的木板，帮同伴按住地上的木板前，我从地上捡起了几样东西？", + "answer_chinese": "你刚才在放下手中的木板，帮同伴按住地上的木板前，你从地上捡起了四样东西。", + "evidence": [ + { + "description": "I picked up three things from the ground.", + "timestamps": { + "start": "00:07:32", + "end": "00:07:42" + } + }, + { + "description": "I picked up one thing from the ground.", + "timestamps": { + "start": "00:07:42", + "end": "00:07:49" + } + } + ], + "question_timestamps": "00:07:52", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You picked up two items.", + "B": "You picked up six items.", + "C": "Earlier you picked up three items from the floor before helping your teammate hold the board down.", + "D": "Earlier you picked up five items from the ground before setting the board down and helping hold it.", + "E": "Earlier, before you set down the wooden board in your hands and helped your teammate hold down the board on the ground, you picked up four items from the floor." + }, + "answer_label": "E", + "answer_text": "Earlier, before you set down the wooden board in your hands and helped your teammate hold down the board on the ground, you picked up four items from the floor.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_6_T1", + "correct_answer": "Earlier, before you set down the wooden board in your hands and helped your teammate hold down the board on the ground, you picked up four items from the floor.", + "original_options": { + "A": "You picked up two items.", + "B": "You picked up six items.", + "C": "Earlier you picked up three items from the floor before helping your teammate hold the board down.", + "D": "Earlier you picked up five items from the ground before setting the board down and helping hold it.", + "E": "Earlier, before you set down the wooden board in your hands and helped your teammate hold down the board on the ground, you picked up four items from the floor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_6_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, before I put down the wooden board in my hand and helped my partner hold down the wooden board on the ground, how many items did I pick up from the ground?", + "answer": "Earlier, before you set down the wooden board in your hands and helped your teammate hold down the board on the ground, you picked up four items from the floor. Those items are all behind you.", + "question_chinese": "我刚才在放下手中的木板，帮同伴按住地上的木板前，我从地上捡起了几样东西？", + "answer_chinese": "你刚才在放下手中的木板，帮同伴按住地上的木板前，你从地上捡起了四样东西。东西都在你的后面。", + "evidence": [ + { + "description": "I picked up three things from the ground.", + "timestamps": { + "start": "00:07:32", + "end": "00:07:42" + } + }, + { + "description": "I picked up one thing from the ground.", + "timestamps": { + "start": "00:07:42", + "end": "00:07:49" + } + } + ], + "question_timestamps": "00:16:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Earlier, before you set down the wooden board in your hands and helped your teammate hold down the board on the ground, you picked up four items from the floor. Those items are all behind you.", + "B": "You picked up two items.", + "C": "You picked up six items.", + "D": "Earlier you picked up three items from the floor, and they were all behind you.", + "E": "Earlier you picked up five items from the floor, and they were all behind you." + }, + "answer_label": "A", + "answer_text": "Earlier, before you set down the wooden board in your hands and helped your teammate hold down the board on the ground, you picked up four items from the floor. Those items are all behind you.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_6_T2", + "correct_answer": "Earlier, before you set down the wooden board in your hands and helped your teammate hold down the board on the ground, you picked up four items from the floor. Those items are all behind you.", + "original_options": { + "A": "Earlier, before you set down the wooden board in your hands and helped your teammate hold down the board on the ground, you picked up four items from the floor. Those items are all behind you.", + "B": "You picked up two items.", + "C": "You picked up six items.", + "D": "Earlier you picked up three items from the floor, and they were all behind you.", + "E": "Earlier you picked up five items from the floor, and they were all behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_5_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I just used the saw to cut the wooden board, where did I put it?", + "answer": "You put the saw on the ground next to the bucket holding wooden boards, about 60 cm from the vacuum cleaner. They are both to your back-right.", + "question_chinese": "我刚才用完锯子切割完木板后，把它放到了哪里？", + "answer_chinese": "你把锯子放在了装木板的桶旁边的地上，离吸尘器大概60cm的地方。他们都在你的右后方。", + "evidence": [ + { + "description": "The saw was placed on the ground next to the wooden bucket, about 60cm away from the vacuum cleaner.", + "timestamps": { + "start": "00:32:53", + "end": "00:32:57" + } + } + ], + "question_timestamps": "00:33:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You set the saw beside the vacuum cleaner.", + "B": "You left the saw by the bucket.", + "C": "You put the saw on the ground beside the bucket, but it was about 30 cm from the vacuum cleaner.", + "D": "You put the saw on the ground next to the bucket holding wooden boards, about 60 cm from the vacuum cleaner. They are both to your back-right.", + "E": "You placed the saw on the ground next to the bucket of boards, and both items were to your back-left." + }, + "answer_label": "D", + "answer_text": "You put the saw on the ground next to the bucket holding wooden boards, about 60 cm from the vacuum cleaner. They are both to your back-right.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_5_T1", + "correct_answer": "You put the saw on the ground next to the bucket holding wooden boards, about 60 cm from the vacuum cleaner. They are both to your back-right.", + "original_options": { + "A": "You set the saw beside the vacuum cleaner.", + "B": "You left the saw by the bucket.", + "C": "You put the saw on the ground beside the bucket, but it was about 30 cm from the vacuum cleaner.", + "D": "You put the saw on the ground next to the bucket holding wooden boards, about 60 cm from the vacuum cleaner. They are both to your back-right.", + "E": "You placed the saw on the ground next to the bucket of boards, and both items were to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 20, + 20 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_5_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I just used the saw to cut the wooden board, where did I put it?", + "answer": "You put the saw on the ground next to the bucket holding wooden boards, about 60 cm from the vacuum cleaner. They are both to your back-left.", + "question_chinese": "我刚才用完锯子切割完木板后，把它放到了哪里？", + "answer_chinese": "你把锯子放在了装木板的桶旁边的地上，离吸尘器大概60cm的地方。他们都在你的左后方。", + "evidence": [ + { + "description": "The saw was placed on the ground next to the wooden bucket, about 60cm away from the vacuum cleaner.", + "timestamps": { + "start": "00:32:53", + "end": "00:32:57" + } + } + ], + "question_timestamps": "00:37:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the ground beside the vacuum cleaner.", + "B": "On the floor near the bucket.", + "C": "You set the saw down on the ground by the bucket, but it was closer to your back-right side.", + "D": "You placed the saw on the floor next to the bucket of boards, roughly a meter from the vacuum cleaner behind you.", + "E": "You put the saw on the ground next to the bucket holding wooden boards, about 60 cm from the vacuum cleaner. They are both to your back-left." + }, + "answer_label": "E", + "answer_text": "You put the saw on the ground next to the bucket holding wooden boards, about 60 cm from the vacuum cleaner. They are both to your back-left.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_5_T2", + "correct_answer": "You put the saw on the ground next to the bucket holding wooden boards, about 60 cm from the vacuum cleaner. They are both to your back-left.", + "original_options": { + "A": "On the ground beside the vacuum cleaner.", + "B": "On the floor near the bucket.", + "C": "You set the saw down on the ground by the bucket, but it was closer to your back-right side.", + "D": "You placed the saw on the floor next to the bucket of boards, roughly a meter from the vacuum cleaner behind you.", + "E": "You put the saw on the ground next to the bucket holding wooden boards, about 60 cm from the vacuum cleaner. They are both to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 22 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_8_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I picked up the tape measure from the ground, where did I put it?", + "answer": "You picked up the tape measure from the floor and put it on the cabinet to your right, next to the gray box.", + "question_chinese": "我把卷尺从地上捡起来放到了哪里？", + "answer_chinese": "你把卷尺从地上捡起来放到了你右边的柜子上，灰色盒子旁边。", + "evidence": [ + { + "description": "I picked up the tape measure from the ground and placed it on the cabinet to your right, next to the gray box.", + "timestamps": { + "start": "00:23:11", + "end": "00:23:17" + } + } + ], + "question_timestamps": "00:34:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You left it on the floor.", + "B": "You picked up the tape measure from the floor and put it on the cabinet to your right, next to the gray box.", + "C": "You put it on the left cabinet.", + "D": "You placed the tape measure on the cabinet in front of you, next to the gray box.", + "E": "You set the tape measure on the cabinet to your right, but it was positioned away from the gray box." + }, + "answer_label": "B", + "answer_text": "You picked up the tape measure from the floor and put it on the cabinet to your right, next to the gray box.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_8_T1", + "correct_answer": "You picked up the tape measure from the floor and put it on the cabinet to your right, next to the gray box.", + "original_options": { + "A": "You left it on the floor.", + "B": "You picked up the tape measure from the floor and put it on the cabinet to your right, next to the gray box.", + "C": "You put it on the left cabinet.", + "D": "You placed the tape measure on the cabinet in front of you, next to the gray box.", + "E": "You set the tape measure on the cabinet to your right, but it was positioned away from the gray box." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 20 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_8_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After I picked up the tape measure from the ground, where did I put it?", + "answer": "You picked up the tape measure from the floor and put it on the cabinet to your right, next to the gray box. The tape measure is currently to your left.", + "question_chinese": "我把卷尺从地上捡起来放到了哪里？", + "answer_chinese": "你把卷尺从地上捡起来放到了你右边的柜子上，灰色盒子旁边。目前卷尺在你的左边", + "evidence": [ + { + "description": "I picked up the tape measure from the ground and placed it on the cabinet to your right, next to the gray box.", + "timestamps": { + "start": "00:23:11", + "end": "00:23:17" + } + } + ], + "question_timestamps": "00:44:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the cabinet to your left.", + "B": "On the floor by you.", + "C": "You picked up the tape measure from the floor and put it on the cabinet to your right, next to the gray box. The tape measure is currently to your left.", + "D": "You placed it on the cabinet to your right, but it ended up to your right afterward.", + "E": "You set it on the cabinet to your left next to the gray box, leaving it in front of you." + }, + "answer_label": "C", + "answer_text": "You picked up the tape measure from the floor and put it on the cabinet to your right, next to the gray box. The tape measure is currently to your left.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_8_T2", + "correct_answer": "You picked up the tape measure from the floor and put it on the cabinet to your right, next to the gray box. The tape measure is currently to your left.", + "original_options": { + "A": "On the cabinet to your left.", + "B": "On the floor by you.", + "C": "You picked up the tape measure from the floor and put it on the cabinet to your right, next to the gray box. The tape measure is currently to your left.", + "D": "You placed it on the cabinet to your right, but it ended up to your right afterward.", + "E": "You set it on the cabinet to your left next to the gray box, leaving it in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 20 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_7_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I first used the tape measure and a pen to mark the wooden board, what route did I take to walk to the cutting machine?", + "answer": "You first walked 4 steps to the right, then 3 steps forward, and used the cutting machine to cut the wooden board.", + "question_chinese": "我第一次使用卷尺和笔对木板做了记号后，我走到切割机的路线是什么？", + "answer_chinese": "你先向右走了4步，然后向前走了3步，用切割机对木板进行切割。", + "evidence": [ + { + "description": "I took 4 steps to the right first.", + "timestamps": { + "start": "00:05:51", + "end": "00:06:00" + } + }, + { + "description": "Then I took three steps forward and used a cutting machine to cut the wooden board.", + "timestamps": { + "start": "00:06:11", + "end": "00:06:22" + } + } + ], + "question_timestamps": "00:16:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You first walked 4 steps to the right, then 3 steps forward, and used the cutting machine to cut the wooden board.", + "B": "You walked 3 right, then 3 forward.", + "C": "You walked 5 right, then 3 forward.", + "D": "You first walked 4 steps to the right, then 2 steps forward, and used the cutting machine to cut the board.", + "E": "You first walked 4 steps to the left, then 3 steps forward, and used the cutting machine to cut the wooden board." + }, + "answer_label": "A", + "answer_text": "You first walked 4 steps to the right, then 3 steps forward, and used the cutting machine to cut the wooden board.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_7_T1", + "correct_answer": "You first walked 4 steps to the right, then 3 steps forward, and used the cutting machine to cut the wooden board.", + "original_options": { + "A": "You first walked 4 steps to the right, then 3 steps forward, and used the cutting machine to cut the wooden board.", + "B": "You walked 3 right, then 3 forward.", + "C": "You walked 5 right, then 3 forward.", + "D": "You first walked 4 steps to the right, then 2 steps forward, and used the cutting machine to cut the board.", + "E": "You first walked 4 steps to the left, then 3 steps forward, and used the cutting machine to cut the wooden board." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 22 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_7_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I first used the tape measure and a pen to mark the wooden board, what route did I take to walk to the cutting machine?", + "answer": "You first walked 4 steps to the right, then 3 steps forward, and used the cutting machine in front of you to cut the wooden board.", + "question_chinese": "我第一次使用卷尺和笔对木板做了记号后，我走到切割机的路线是什么？", + "answer_chinese": "你先向右走了4步，然后向前走了3步，用前方的切割机对木板进行切割。", + "evidence": [ + { + "description": "I took 4 steps to the right first.", + "timestamps": { + "start": "00:05:51", + "end": "00:06:00" + } + }, + { + "description": "Then I took three steps forward and used a cutting machine to cut the wooden board.", + "timestamps": { + "start": "00:06:11", + "end": "00:06:22" + } + } + ], + "question_timestamps": "00:22:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked 3 right, then 3 forward.", + "B": "You first walked 4 steps to the right, then 3 steps forward, and used the cutting machine in front of you to cut the wooden board.", + "C": "You walked 4 right, then 2 forward.", + "D": "You first walked 5 steps to the right, then 3 steps forward, and used the cutting machine in front of you.", + "E": "You first walked 4 steps to the left, then 3 steps forward, and used the cutting machine in front of you." + }, + "answer_label": "B", + "answer_text": "You first walked 4 steps to the right, then 3 steps forward, and used the cutting machine in front of you to cut the wooden board.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_7_T2", + "correct_answer": "You first walked 4 steps to the right, then 3 steps forward, and used the cutting machine in front of you to cut the wooden board.", + "original_options": { + "A": "You walked 3 right, then 3 forward.", + "B": "You first walked 4 steps to the right, then 3 steps forward, and used the cutting machine in front of you to cut the wooden board.", + "C": "You walked 4 right, then 2 forward.", + "D": "You first walked 5 steps to the right, then 3 steps forward, and used the cutting machine in front of you.", + "E": "You first walked 4 steps to the left, then 3 steps forward, and used the cutting machine in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 21 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_3_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, is the red-handled tool in front of me, to my left, or to my right?", + "answer": "The tool with the red handle is now in front of you.", + "question_chinese": "现在红色手柄的工具在我的前面，左边还是右边？", + "answer_chinese": "现在红色手柄的工具在你的前面。", + "evidence": [ + { + "description": "he tool with the red handle is now in front of me.", + "timestamps": { + "start": "00:21:47", + "end": "00:22:15" + } + } + ], + "question_timestamps": "00:22:23", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The tool with the red handle is now to your left.", + "B": "The tool with the red handle is now in front of you." + }, + "answer_label": "B", + "answer_text": "The tool with the red handle is now in front of you.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_3_T1", + "correct_answer": "The tool with the red handle is now in front of you.", + "original_options": { + "A": "The tool with the red handle is now to your left.", + "B": "The tool with the red handle is now in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_3_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, is the red-handled tool in front of me, to my left, or to my right?", + "answer": "The tool with the red handle is now to your left.", + "question_chinese": "现在红色手柄的工具在我的前面，左边还是右边？", + "answer_chinese": "现在红色手柄的工具在你的左边。", + "evidence": [ + { + "description": "he tool with the red handle is now in front of me.", + "timestamps": { + "start": "00:21:47", + "end": "00:22:15" + } + } + ], + "question_timestamps": "00:23:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The tool with the red handle is now to your left.", + "B": "The tool with the red handle is now to your right." + }, + "answer_label": "A", + "answer_text": "The tool with the red handle is now to your left.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_3_T2", + "correct_answer": "The tool with the red handle is now to your left.", + "original_options": { + "A": "The tool with the red handle is now to your left.", + "B": "The tool with the red handle is now to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_2_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I exhaled smoke?", + "answer": "So far, I have exhaled smoke 4 times.", + "question_chinese": "目前为止，我吐出了几次烟雾？", + "answer_chinese": "目前为止，我吐出了4次烟雾。", + "evidence": [ + { + "description": "This is the first time I have exhaled smoke.", + "timestamps": { + "start": "00:15:41", + "end": "00:15:45" + } + }, + { + "description": "This is the second time I have exhaled smoke.", + "timestamps": { + "start": "00:15:45", + "end": "00:15:53" + } + }, + { + "description": "This is the third time I have exhaled smoke.", + "timestamps": { + "start": "00:15:53", + "end": "00:16:03" + } + }, + { + "description": "This is the fourth time I have exhaled smoke.", + "timestamps": { + "start": "00:16:03", + "end": "00:16:10" + } + } + ], + "question_timestamps": "00:16:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, I have exhaled smoke 4 times.", + "B": "I have exhaled smoke three times.", + "C": "I have exhaled smoke five times.", + "D": "So far, I have exhaled smoke two times in total during the video.", + "E": "Up to now, I have exhaled smoke six times overall, counting each visible puff." + }, + "answer_label": "A", + "answer_text": "So far, I have exhaled smoke 4 times.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_2_T1", + "correct_answer": "So far, I have exhaled smoke 4 times.", + "original_options": { + "A": "So far, I have exhaled smoke 4 times.", + "B": "I have exhaled smoke three times.", + "C": "I have exhaled smoke five times.", + "D": "So far, I have exhaled smoke two times in total during the video.", + "E": "Up to now, I have exhaled smoke six times overall, counting each visible puff." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 14 + ] + }, + { + "video_uid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642", + "video_length": "00:44:02", + "qid": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_2_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I exhaled smoke?", + "answer": "So far, I have exhaled smoke 6 times.", + "question_chinese": "目前为止，我吐出了几次烟雾？", + "answer_chinese": "目前为止，我吐出了6次烟雾。", + "evidence": [ + { + "description": "This is the first time I have exhaled smoke.", + "timestamps": { + "start": "00:15:41", + "end": "00:15:45" + } + }, + { + "description": "This is the second time I have exhaled smoke.", + "timestamps": { + "start": "00:15:45", + "end": "00:15:53" + } + }, + { + "description": "This is the third time I have exhaled smoke.", + "timestamps": { + "start": "00:15:53", + "end": "00:16:03" + } + }, + { + "description": "This is the fourth time I have exhaled smoke.", + "timestamps": { + "start": "00:16:03", + "end": "00:16:10" + } + } + ], + "question_timestamps": "00:16:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "I have exhaled smoke five times.", + "B": "So far, I have exhaled smoke 6 times.", + "C": "I have exhaled smoke seven times.", + "D": "So far, I have exhaled smoke four times total during the sequence.", + "E": "Up to now, I have exhaled smoke eight times altogether, including the most recent breath." + }, + "answer_label": "B", + "answer_text": "So far, I have exhaled smoke 6 times.", + "q_id": "7c0c0057-3bbc-40ef-a91e-4d6e1a637642_2_T2", + "correct_answer": "So far, I have exhaled smoke 6 times.", + "original_options": { + "A": "I have exhaled smoke five times.", + "B": "So far, I have exhaled smoke 6 times.", + "C": "I have exhaled smoke seven times.", + "D": "So far, I have exhaled smoke four times total during the sequence.", + "E": "Up to now, I have exhaled smoke eight times altogether, including the most recent breath." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 15 + ] + } +] diff --git a/QAs/7c420d19-88b4-4092-bef1-a9bb99bbe308-mcq.json b/QAs/7c420d19-88b4-4092-bef1-a9bb99bbe308-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c024d597734ca3c71653ccc0fce3ba004d89f4f4 --- /dev/null +++ b/QAs/7c420d19-88b4-4092-bef1-a9bb99bbe308-mcq.json @@ -0,0 +1,1852 @@ +[ + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_19", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many white stools are placed in the rest area?", + "answer": "Five white stools were placed.", + "question_chinese": "休息区摆了几条白色凳子？", + "answer_chinese": "摆了五条白色凳子。", + "evidence": [ + { + "description": "Five white stools were placed.", + "timestamps": { + "start": "00:01:37", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:08:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four white stools were placed.", + "B": "Six white stools were placed.", + "C": "Seven white stools were placed in the rest area in total, arranged along the wall.", + "D": "Eight white stools were placed in the rest area, all grouped together near the seating area.", + "E": "Five white stools were placed." + }, + "answer_label": "E", + "answer_text": "Five white stools were placed.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_19", + "correct_answer": "Five white stools were placed.", + "original_options": { + "A": "Four white stools were placed.", + "B": "Six white stools were placed.", + "C": "Seven white stools were placed in the rest area in total, arranged along the wall.", + "D": "Eight white stools were placed in the rest area, all grouped together near the seating area.", + "E": "Five white stools were placed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_20", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many white stools are placed in the rest area?", + "answer": "Five white stools were placed.", + "question_chinese": "休息区摆了几条白色凳子？", + "answer_chinese": "摆了五条白色凳子。", + "evidence": [ + { + "description": "Five white stools were placed.", + "timestamps": { + "start": "00:01:37", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:18:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four white stools were placed.", + "B": "Five white stools were placed.", + "C": "Six white stools were placed.", + "D": "Seven white stools were placed in the rest area in total, arranged along the wall.", + "E": "Eight white stools were placed in the rest area, all grouped together near the seating area." + }, + "answer_label": "B", + "answer_text": "Five white stools were placed.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_20", + "correct_answer": "Five white stools were placed.", + "original_options": { + "A": "Four white stools were placed.", + "B": "Five white stools were placed.", + "C": "Six white stools were placed.", + "D": "Seven white stools were placed in the rest area in total, arranged along the wall.", + "E": "Eight white stools were placed in the rest area, all grouped together near the seating area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_13", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the safety exit?", + "answer": "There is a safety exit on the right side of the table where you put the green potted plants.", + "question_chinese": "安全出口在哪里？", + "answer_chinese": "在你放绿色盆栽的桌子的右边有安全出口。", + "evidence": [ + { + "description": "There is a safety exit on the right side of the table where you put the green potted plants.", + "timestamps": { + "start": "00:46:17", + "end": "00:46:21" + } + } + ], + "question_timestamps": "00:49:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of the table.", + "B": "To the left of the table.", + "C": "There is a safety exit on the right side of the table where you put the green potted plants.", + "D": "The safety exit is behind the table where you placed the green potted plants.", + "E": "The safety exit is in front of you, positioned near the table with the green potted plants." + }, + "answer_label": "C", + "answer_text": "There is a safety exit on the right side of the table where you put the green potted plants.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_13", + "correct_answer": "There is a safety exit on the right side of the table where you put the green potted plants.", + "original_options": { + "A": "In front of the table.", + "B": "To the left of the table.", + "C": "There is a safety exit on the right side of the table where you put the green potted plants.", + "D": "The safety exit is behind the table where you placed the green potted plants.", + "E": "The safety exit is in front of you, positioned near the table with the green potted plants." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 17 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_14", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the safety exit?", + "answer": "There is a safety exit on the right side of the table where you put the green potted plants.", + "question_chinese": "安全出口在哪里？", + "answer_chinese": "在你放绿色盆栽的桌子的右边有安全出口。", + "evidence": [ + { + "description": "There is a safety exit on the right side of the table where you put the green potted plants.", + "timestamps": { + "start": "00:46:17", + "end": "00:46:21" + } + } + ], + "question_timestamps": "00:52:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind the table.", + "B": "There is a safety exit on the right side of the table where you put the green potted plants.", + "C": "To the left of the table.", + "D": "The safety exit is in front of the table where you put the green potted plants.", + "E": "The safety exit is behind you, aligned with the table that holds the green potted plants." + }, + "answer_label": "B", + "answer_text": "There is a safety exit on the right side of the table where you put the green potted plants.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_14", + "correct_answer": "There is a safety exit on the right side of the table where you put the green potted plants.", + "original_options": { + "A": "Behind the table.", + "B": "There is a safety exit on the right side of the table where you put the green potted plants.", + "C": "To the left of the table.", + "D": "The safety exit is in front of the table where you put the green potted plants.", + "E": "The safety exit is behind you, aligned with the table that holds the green potted plants." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 6, + 16, + 16 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_3_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trays of small potted plants did I take in total and put onto the cart?", + "answer": "You took one small potted plant and placed it on the small cart in front of you.", + "question_chinese": "我一共取了几盘小盆栽放在小推车上？", + "answer_chinese": "你取了一盆小盆栽放在你面前的小推车上。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:08" + } + } + ], + "question_timestamps": "00:06:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took zero trays.", + "B": "You took two trays.", + "C": "You took three small potted plants and placed them onto the small cart in front of you.", + "D": "In total, you took four small potted plants and set them on the small cart in front of you.", + "E": "You took one small potted plant and placed it on the small cart in front of you." + }, + "answer_label": "E", + "answer_text": "You took one small potted plant and placed it on the small cart in front of you.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_3_T1", + "correct_answer": "You took one small potted plant and placed it on the small cart in front of you.", + "original_options": { + "A": "You took zero trays.", + "B": "You took two trays.", + "C": "You took three small potted plants and placed them onto the small cart in front of you.", + "D": "In total, you took four small potted plants and set them on the small cart in front of you.", + "E": "You took one small potted plant and placed it on the small cart in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 19 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_5", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many small potted plants did I take in total and put on the trolley?", + "answer": "You took a total of two small potted plants.", + "question_chinese": "我一共取了几盘小盆栽放在小推车上？", + "answer_chinese": "你一共取了两盆小盆栽。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:12" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:06:21", + "end": "00:06:27" + } + } + ], + "question_timestamps": "00:011:33", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You took one small potted plant.", + "B": "You took three small potted plants.", + "C": "You took a total of four small potted plants and put all of them onto the trolley.", + "D": "You took a total of two small potted plants.", + "E": "You took a total of five small potted plants and placed each one on the trolley." + }, + "answer_label": "D", + "answer_text": "You took a total of two small potted plants.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_5", + "correct_answer": "You took a total of two small potted plants.", + "original_options": { + "A": "You took one small potted plant.", + "B": "You took three small potted plants.", + "C": "You took a total of four small potted plants and put all of them onto the trolley.", + "D": "You took a total of two small potted plants.", + "E": "You took a total of five small potted plants and placed each one on the trolley." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_3_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trays of small potted plants did I take in total and put onto the cart?", + "answer": "In total, you took two small potted plants and placed them on the small cart in front of you.", + "question_chinese": "我一共取了几盘小盆栽放在小推车上？", + "answer_chinese": "你一共取了两盆小盆栽放在你面前的小推车上。。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:12" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:06:21", + "end": "00:06:27" + } + } + ], + "question_timestamps": "00:021:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In total, you took two small potted plants and placed them on the small cart in front of you.", + "B": "In total, you took one tray.", + "C": "In total, you took three trays.", + "D": "In total, you took four trays of small potted plants and put them onto the cart in front of you.", + "E": "In total, you took zero trays of small potted plants, so nothing was placed onto the cart." + }, + "answer_label": "A", + "answer_text": "In total, you took two small potted plants and placed them on the small cart in front of you.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_3_T2", + "correct_answer": "In total, you took two small potted plants and placed them on the small cart in front of you.", + "original_options": { + "A": "In total, you took two small potted plants and placed them on the small cart in front of you.", + "B": "In total, you took one tray.", + "C": "In total, you took three trays.", + "D": "In total, you took four trays of small potted plants and put them onto the cart in front of you.", + "E": "In total, you took zero trays of small potted plants, so nothing was placed onto the cart." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 17 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_11", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many grinding dishes did I put on the table?", + "answer": "You put four in total.", + "question_chinese": "我在桌子上放了几个研磨皿？", + "answer_chinese": "你一共放了四个。", + "evidence": [ + { + "description": "You put four in total.", + "timestamps": { + "start": "00:41:17", + "end": "00:41:18" + } + } + ], + "question_timestamps": "00:48:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You put four in total.", + "B": "You put three in total.", + "C": "You put five in total.", + "D": "You put two grinding dishes in total onto the table before moving on to the next step.", + "E": "You put six grinding dishes in total on the table, placing them down one after another." + }, + "answer_label": "A", + "answer_text": "You put four in total.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_11", + "correct_answer": "You put four in total.", + "original_options": { + "A": "You put four in total.", + "B": "You put three in total.", + "C": "You put five in total.", + "D": "You put two grinding dishes in total onto the table before moving on to the next step.", + "E": "You put six grinding dishes in total on the table, placing them down one after another." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_3", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my route to push the small cart to get the potted plant?", + "answer": "You turned right out of the laboratory, then walked straight and took the elevator. You reached the underground laboratory and got the potted plant.", + "question_chinese": "我推小推车去拿盆栽的行动路线是什么？", + "answer_chinese": "你向右拐出了实验室，然后直走乘坐了电梯，走到了地下实验室拿到了盆栽。", + "evidence": [ + { + "description": "You turned right out of the laboratory and then walked straight.", + "timestamps": { + "start": "00:01:02", + "end": "00:02:35" + } + }, + { + "description": "You took the elevator.", + "timestamps": { + "start": "00:02:46", + "end": "00:03:08" + } + }, + { + "description": "You walked into the underground laboratory.", + "timestamps": { + "start": "00:03:13", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:10:15", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turned left and took the elevator.", + "B": "You went straight and took the stairs.", + "C": "You turned right out of the laboratory, walked straight down the corridor, then took the stairs to get the potted plant.", + "D": "You turned right out of the laboratory, walked straight, entered a nearby room, and got the potted plant without using the elevator.", + "E": "You turned right out of the laboratory, then walked straight and took the elevator. You reached the underground laboratory and got the potted plant." + }, + "answer_label": "E", + "answer_text": "You turned right out of the laboratory, then walked straight and took the elevator. You reached the underground laboratory and got the potted plant.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_3", + "correct_answer": "You turned right out of the laboratory, then walked straight and took the elevator. You reached the underground laboratory and got the potted plant.", + "original_options": { + "A": "You turned left and took the elevator.", + "B": "You went straight and took the stairs.", + "C": "You turned right out of the laboratory, walked straight down the corridor, then took the stairs to get the potted plant.", + "D": "You turned right out of the laboratory, walked straight, entered a nearby room, and got the potted plant without using the elevator.", + "E": "You turned right out of the laboratory, then walked straight and took the elevator. You reached the underground laboratory and got the potted plant." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 22 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_2_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my route when I pushed the cart to go get the potted plants?", + "answer": "You went straight, took the elevator to the underground lab, and got the potted plant.", + "question_chinese": "我推小推车去拿盆栽的行动路线是什么？", + "answer_chinese": "你直走乘坐了电梯到了地下实验室拿到了盆栽。", + "evidence": [ + { + "description": "You took the elevator.", + "timestamps": { + "start": "00:02:46", + "end": "00:03:08" + } + }, + { + "description": "You walked into the underground laboratory.", + "timestamps": { + "start": "00:03:13", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:12:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right, then took the stairs.", + "B": "You took the elevator up instead.", + "C": "You went straight down the hallway, skipped the elevator, and got the potted plant on the same floor.", + "D": "You turned right out of the lab, walked straight, took the stairs, and reached the underground lab to get the potted plant.", + "E": "You went straight, took the elevator to the underground lab, and got the potted plant." + }, + "answer_label": "E", + "answer_text": "You went straight, took the elevator to the underground lab, and got the potted plant.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_2_T1", + "correct_answer": "You went straight, took the elevator to the underground lab, and got the potted plant.", + "original_options": { + "A": "You turned right, then took the stairs.", + "B": "You took the elevator up instead.", + "C": "You went straight down the hallway, skipped the elevator, and got the potted plant on the same floor.", + "D": "You turned right out of the lab, walked straight, took the stairs, and reached the underground lab to get the potted plant.", + "E": "You went straight, took the elevator to the underground lab, and got the potted plant." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 22 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_2_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my route when I pushed the cart to go get the potted plants?", + "answer": "You turned right out of the lab, went straight to take the elevator, then walked to the underground lab and got the potted plant.", + "question_chinese": "我推小推车去拿盆栽的行动路线是什么？", + "answer_chinese": "你向右拐出了实验室，然后直走乘坐了电梯，走到了地下实验室拿到了盆栽。", + "evidence": [ + { + "description": "You turned right out of the laboratory and then walked straight.", + "timestamps": { + "start": "00:01:02", + "end": "00:02:35" + } + }, + { + "description": "You took the elevator.", + "timestamps": { + "start": "00:02:46", + "end": "00:03:08" + } + }, + { + "description": "You walked into the underground laboratory.", + "timestamps": { + "start": "00:03:13", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:15:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left out of the lab.", + "B": "You took the stairs, not the elevator.", + "C": "You turned right out of the lab, went straight to take the elevator, then walked to the underground lab and got the potted plant.", + "D": "You turned right out of the lab, went straight past the elevator, and got the potted plant without going underground.", + "E": "You turned right out of the lab, took the elevator, but ended up on a different floor, not underground." + }, + "answer_label": "C", + "answer_text": "You turned right out of the lab, went straight to take the elevator, then walked to the underground lab and got the potted plant.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_2_T2", + "correct_answer": "You turned right out of the lab, went straight to take the elevator, then walked to the underground lab and got the potted plant.", + "original_options": { + "A": "You turned left out of the lab.", + "B": "You took the stairs, not the elevator.", + "C": "You turned right out of the lab, went straight to take the elevator, then walked to the underground lab and got the potted plant.", + "D": "You turned right out of the lab, went straight past the elevator, and got the potted plant without going underground.", + "E": "You turned right out of the lab, took the elevator, but ended up on a different floor, not underground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 19 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_17", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What color is the cap of the test tube I'm using?", + "answer": "The cap is purple and white.", + "question_chinese": "我用的试管的盖帽是什么颜色的？", + "answer_chinese": "盖帽是紫色和白色的。", + "evidence": [ + { + "description": "The cap is purple.", + "timestamps": { + "start": "00:48:00", + "end": "00:48:03" + } + }, + { + "description": "The cap is white.", + "timestamps": { + "start": "00:53:02", + "end": "00:53:14" + } + } + ], + "question_timestamps": "00:53:16", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The cap is orange and white.", + "B": "The cap is pink and grey.", + "C": "The cap is purple and white.", + "D": "The cap is blue and green, with two distinct colored sections on the top and side.", + "E": "The cap is yellow and black, split into two bands that wrap around the test tube." + }, + "answer_label": "C", + "answer_text": "The cap is purple and white.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_17", + "correct_answer": "The cap is purple and white.", + "original_options": { + "A": "The cap is orange and white.", + "B": "The cap is pink and grey.", + "C": "The cap is purple and white.", + "D": "The cap is blue and green, with two distinct colored sections on the top and side.", + "E": "The cap is yellow and black, split into two bands that wrap around the test tube." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_18", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What color is the cap of the test tube I'm using?", + "answer": "The cap is purple and white.", + "question_chinese": "我用的试管的盖帽是什么颜色的？", + "answer_chinese": "盖帽是紫色和白色的。", + "evidence": [ + { + "description": "The cap is purple.", + "timestamps": { + "start": "00:48:00", + "end": "00:48:03" + } + }, + { + "description": "The cap is white.", + "timestamps": { + "start": "00:53:02", + "end": "00:53:14" + } + } + ], + "question_timestamps": "00:53:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The cap is purple and white.", + "B": "The cap is purple and pink.", + "C": "The cap is white and blue.", + "D": "The cap is purple and gray, with the darker shade around the rim and a lighter top.", + "E": "The cap is white and beige, with a pale cream tint covering most of the cap." + }, + "answer_label": "A", + "answer_text": "The cap is purple and white.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_18", + "correct_answer": "The cap is purple and white.", + "original_options": { + "A": "The cap is purple and white.", + "B": "The cap is purple and pink.", + "C": "The cap is white and blue.", + "D": "The cap is purple and gray, with the darker shade around the rim and a lighter top.", + "E": "The cap is white and beige, with a pale cream tint covering most of the cap." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_9_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many white stools did I see placed in the lounge area?", + "answer": "You saw four white stools lined up.", + "question_chinese": "我看见休息区摆了几条白色凳子？", + "answer_chinese": "你看见摆了四条白色凳子。", + "evidence": [ + { + "description": "Five white stools were placed.", + "timestamps": { + "start": "00:01:37", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:01:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw four white stools lined up.", + "B": "You saw three white stools.", + "C": "You saw five white stools.", + "D": "You saw two white stools lined up next to each other in the lounge area.", + "E": "You saw six white stools lined up, extending across the lounge area in a row." + }, + "answer_label": "A", + "answer_text": "You saw four white stools lined up.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_9_T1", + "correct_answer": "You saw four white stools lined up.", + "original_options": { + "A": "You saw four white stools lined up.", + "B": "You saw three white stools.", + "C": "You saw five white stools.", + "D": "You saw two white stools lined up next to each other in the lounge area.", + "E": "You saw six white stools lined up, extending across the lounge area in a row." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_9_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many white stools did I see placed in the lounge area?", + "answer": "You saw five white stools lined up.", + "question_chinese": "我看见休息区摆了几条白色凳子？", + "answer_chinese": "你看见摆了五条白色凳子。", + "evidence": [ + { + "description": "Five white stools were placed.", + "timestamps": { + "start": "00:01:37", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:18:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw four white stools.", + "B": "You saw five white stools lined up.", + "C": "You saw six white stools.", + "D": "You saw three white stools lined up in a row in the lounge area.", + "E": "You saw seven white stools lined up, forming a longer row along the lounge area." + }, + "answer_label": "B", + "answer_text": "You saw five white stools lined up.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_9_T2", + "correct_answer": "You saw five white stools lined up.", + "original_options": { + "A": "You saw four white stools.", + "B": "You saw five white stools lined up.", + "C": "You saw six white stools.", + "D": "You saw three white stools lined up in a row in the lounge area.", + "E": "You saw seven white stools lined up, forming a longer row along the lounge area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_7_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the emergency exit?", + "answer": "The emergency exit is behind the brown door.", + "question_chinese": "我看见安全出口在哪里？", + "answer_chinese": "棕色的门后面是安全出口。", + "evidence": [ + { + "description": "There is a safety exit on the right side of the table where you put the green potted plants.", + "timestamps": { + "start": "00:46:17", + "end": "00:46:21" + } + } + ], + "question_timestamps": "00:50:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of the door.", + "B": "Beside the door.", + "C": "The emergency exit is behind the brown door.", + "D": "The emergency exit is to your left of the table where you placed the green potted plant.", + "E": "The emergency exit is on the right side of the table where you put the green potted plant." + }, + "answer_label": "C", + "answer_text": "The emergency exit is behind the brown door.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_7_T1", + "correct_answer": "The emergency exit is behind the brown door.", + "original_options": { + "A": "In front of the door.", + "B": "Beside the door.", + "C": "The emergency exit is behind the brown door.", + "D": "The emergency exit is to your left of the table where you placed the green potted plant.", + "E": "The emergency exit is on the right side of the table where you put the green potted plant." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 17, + 18 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_7_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the emergency exit?", + "answer": "The emergency exit is to your right of the table where you placed the green potted plant.", + "question_chinese": "我看见安全出口在哪里？", + "answer_chinese": "在你放绿色盆栽的桌子的右边有安全出口。", + "evidence": [ + { + "description": "There is a safety exit on the right side of the table where you put the green potted plants.", + "timestamps": { + "start": "00:46:17", + "end": "00:46:21" + } + } + ], + "question_timestamps": "00:52:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The emergency exit is to your right of the table where you placed the green potted plant.", + "B": "The emergency exit is to your left.", + "C": "The emergency exit is behind you.", + "D": "The emergency exit is in front of the table where you placed the green potted plant.", + "E": "The emergency exit is above the table where you placed the green potted plant." + }, + "answer_label": "A", + "answer_text": "The emergency exit is to your right of the table where you placed the green potted plant.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_7_T2", + "correct_answer": "The emergency exit is to your right of the table where you placed the green potted plant.", + "original_options": { + "A": "The emergency exit is to your right of the table where you placed the green potted plant.", + "B": "The emergency exit is to your left.", + "C": "The emergency exit is behind you.", + "D": "The emergency exit is in front of the table where you placed the green potted plant.", + "E": "The emergency exit is above the table where you placed the green potted plant." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 14 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_1_T1", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I see the tissues? Where are the tissues?", + "answer": "The tissues were on the side wall of the cabinet directly in front of you.", + "question_chinese": "我看见抽纸在哪里？抽纸在哪里？", + "answer_chinese": "抽纸在你正前面的柜子侧壁上。", + "evidence": [ + { + "description": "Tissues are hung on the side of the table rack.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They were behind you on the cabinet.", + "B": "They were to your left on the cabinet.", + "C": "The tissues were on the side wall of the cabinet directly to your right as you faced it.", + "D": "The tissues were on the side wall of the cabinet directly in front of you.", + "E": "The tissues were on the side wall of the cabinet above your eye level rather than directly in front." + }, + "answer_label": "D", + "answer_text": "The tissues were on the side wall of the cabinet directly in front of you.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_1_T1", + "correct_answer": "The tissues were on the side wall of the cabinet directly in front of you.", + "original_options": { + "A": "They were behind you on the cabinet.", + "B": "They were to your left on the cabinet.", + "C": "The tissues were on the side wall of the cabinet directly to your right as you faced it.", + "D": "The tissues were on the side wall of the cabinet directly in front of you.", + "E": "The tissues were on the side wall of the cabinet above your eye level rather than directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 19 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_1_T2", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I see the tissues? Where are the tissues?", + "answer": "The tissues were hanging on the side of the shelf on the desk.", + "question_chinese": "我看见抽纸在哪里？抽纸在哪里？", + "answer_chinese": "抽纸在桌子架子上的侧面挂着。", + "evidence": [ + { + "description": "Tissues are hung on the side of the table rack.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:17:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They were on the cabinet wall.", + "B": "They were on the side of the rack.", + "C": "The tissues were stuck on the side wall of the cabinet in front of you instead of the desk shelf.", + "D": "The tissues were hanging on the side of the shelf on the desk.", + "E": "The tissues were hanging from the table rack rather than from the side of the shelf on the desk." + }, + "answer_label": "D", + "answer_text": "The tissues were hanging on the side of the shelf on the desk.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_1_T2", + "correct_answer": "The tissues were hanging on the side of the shelf on the desk.", + "original_options": { + "A": "They were on the cabinet wall.", + "B": "They were on the side of the rack.", + "C": "The tissues were stuck on the side wall of the cabinet in front of you instead of the desk shelf.", + "D": "The tissues were hanging on the side of the shelf on the desk.", + "E": "The tissues were hanging from the table rack rather than from the side of the shelf on the desk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 20, + 19 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_5_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What items are on the shelf of the cabinet in front of me, the one with the stone pestle on it?", + "answer": "On the shelf in front of you were petri dishes.", + "question_chinese": "我面前放石杵那个柜子的架子上放的是什么东西？", + "answer_chinese": "你面前的架子上放的是培养皿。", + "evidence": [ + { + "description": "The cabinet for storing stone pestles.", + "timestamps": { + "start": "00:16:02", + "end": "00:16:04" + } + }, + { + "description": "The shelves in front of you are filled with petri dishes.", + "timestamps": { + "start": "00:16:06", + "end": "00:16:24" + } + } + ], + "question_timestamps": "00:16:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the shelf were beakers.", + "B": "On the shelf were test tubes.", + "C": "On the shelf in front of you were a few beakers set beside the stone pestle.", + "D": "On the shelf in front of you were petri dishes.", + "E": "On the shelf in front of you were several flasks arranged together inside the cabinet." + }, + "answer_label": "D", + "answer_text": "On the shelf in front of you were petri dishes.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_5_T1", + "correct_answer": "On the shelf in front of you were petri dishes.", + "original_options": { + "A": "On the shelf were beakers.", + "B": "On the shelf were test tubes.", + "C": "On the shelf in front of you were a few beakers set beside the stone pestle.", + "D": "On the shelf in front of you were petri dishes.", + "E": "On the shelf in front of you were several flasks arranged together inside the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_5_T2", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What items are on the shelf of the cabinet in front of me, the one with the stone pestle on it?", + "answer": "In the cabinet in front of you were beakers.", + "question_chinese": "我面前放石杵那个柜子的架子上放的是什么东西？", + "answer_chinese": "你面前的柜子里放的是烧杯。", + "evidence": [ + { + "description": "The cabinet for storing stone pestles.", + "timestamps": { + "start": "00:16:02", + "end": "00:16:04" + } + }, + { + "description": "The second compartment holds a beaker.", + "timestamps": { + "start": "00:16:06", + "end": "00:16:24" + } + } + ], + "question_timestamps": "00:28:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the cabinet were petri dishes.", + "B": "In the cabinet were test tubes.", + "C": "In the cabinet in front of you were petri dishes stored on the shelf beneath the pestle.", + "D": "In the cabinet in front of you were beakers.", + "E": "In the cabinet in front of you were flasks lined up together on the shelf." + }, + "answer_label": "D", + "answer_text": "In the cabinet in front of you were beakers.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_5_T2", + "correct_answer": "In the cabinet in front of you were beakers.", + "original_options": { + "A": "In the cabinet were petri dishes.", + "B": "In the cabinet were test tubes.", + "C": "In the cabinet in front of you were petri dishes stored on the shelf beneath the pestle.", + "D": "In the cabinet in front of you were beakers.", + "E": "In the cabinet in front of you were flasks lined up together on the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 15 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_4_T1", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are in that tray of large-leaf small potted plants in front of me? How many pots are in that tray of large-leaf small potted plants?", + "answer": "First, you took out four potted plants.", + "question_chinese": "我面前的那盘大叶片小盆栽一共有多少盆？那盘大叶片小盆栽一共有多少盆？", + "answer_chinese": "你先拿出来了四盆盆栽。", + "evidence": [ + { + "description": "You took out four basins first.", + "timestamps": { + "start": "00:13:11", + "end": "00:13:23" + } + } + ], + "question_timestamps": "00:20:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "First, you took out four potted plants.", + "B": "First, you took out three potted plants.", + "C": "First, you took out five potted plants.", + "D": "First, you removed two potted plants from the tray of large-leaf plants.", + "E": "First, you pulled out six potted plants from the tray before doing anything else." + }, + "answer_label": "A", + "answer_text": "First, you took out four potted plants.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_4_T1", + "correct_answer": "First, you took out four potted plants.", + "original_options": { + "A": "First, you took out four potted plants.", + "B": "First, you took out three potted plants.", + "C": "First, you took out five potted plants.", + "D": "First, you removed two potted plants from the tray of large-leaf plants.", + "E": "First, you pulled out six potted plants from the tray before doing anything else." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 14 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_4_T2", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are in that tray of large-leaf small potted plants in front of me? How many pots are in that tray of large-leaf small potted plants?", + "answer": "First, you took out four pots. There were still nine left in the container, for a total of 13.", + "question_chinese": "我面前的那盘大叶片小盆栽一共有多少盆？那盘大叶片小盆栽一共有多少盆？", + "answer_chinese": "你先拿出来了四盆，盆子里还剩下九盆，一共有13盆。", + "evidence": [ + { + "description": "You took out four basins first.", + "timestamps": { + "start": "00:13:11", + "end": "00:13:23" + } + }, + { + "description": "There are still nine basins left in the basin.", + "timestamps": { + "start": "00:13:44", + "end": "00:13:45" + } + } + ], + "question_timestamps": "00:33:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three pots total.", + "B": "First, you took out four pots. There were still nine left in the container, for a total of 13.", + "C": "Fourteen pots total.", + "D": "You took out two pots first; nine remained in the container, making 11 in total.", + "E": "You took out six pots first; nine remained in the container, making 15 in total." + }, + "answer_label": "B", + "answer_text": "First, you took out four pots. There were still nine left in the container, for a total of 13.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_4_T2", + "correct_answer": "First, you took out four pots. There were still nine left in the container, for a total of 13.", + "original_options": { + "A": "Three pots total.", + "B": "First, you took out four pots. There were still nine left in the container, for a total of 13.", + "C": "Fourteen pots total.", + "D": "You took out two pots first; nine remained in the container, making 11 in total.", + "E": "You took out six pots first; nine remained in the container, making 15 in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_1", + "qid_num": 26, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where are the tissues?", + "answer": "Tissues are hung on the side of the table rack.", + "question_chinese": "抽纸在哪里？", + "answer_chinese": "抽纸在桌子架子上的侧面挂着。", + "evidence": [ + { + "description": "Tissues are hung on the side of the table rack.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:07:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On top of the rack.", + "B": "On the floor nearby.", + "C": "Tissues are hung on the side of the table rack.", + "D": "Tissues are tucked behind the chair next to the table rack, not hanging on it.", + "E": "Tissues are loaded inside the cabinet across from the table rack rather than on the rack." + }, + "answer_label": "C", + "answer_text": "Tissues are hung on the side of the table rack.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_1", + "correct_answer": "Tissues are hung on the side of the table rack.", + "original_options": { + "A": "On top of the rack.", + "B": "On the floor nearby.", + "C": "Tissues are hung on the side of the table rack.", + "D": "Tissues are tucked behind the chair next to the table rack, not hanging on it.", + "E": "Tissues are loaded inside the cabinet across from the table rack rather than on the rack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 16 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_2", + "qid_num": 27, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where are the tissues?", + "answer": "Tissues are hung on the side of the table rack.", + "question_chinese": "抽纸在哪里？", + "answer_chinese": "抽纸在桌子架子上的侧面挂着。", + "evidence": [ + { + "description": "Tissues are hung on the side of the table rack.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:17:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In a box on the table.", + "B": "On the shelf above.", + "C": "Tissues are placed underneath the table near the wall instead of hanging from the rack.", + "D": "Tissues are hung on the side of the table rack.", + "E": "Tissues are piled on the floor next to the chair rather than on the side of the rack." + }, + "answer_label": "D", + "answer_text": "Tissues are hung on the side of the table rack.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_2", + "correct_answer": "Tissues are hung on the side of the table rack.", + "original_options": { + "A": "In a box on the table.", + "B": "On the shelf above.", + "C": "Tissues are placed underneath the table near the wall instead of hanging from the rack.", + "D": "Tissues are hung on the side of the table rack.", + "E": "Tissues are piled on the floor next to the chair rather than on the side of the rack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 15, + 18 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_9", + "qid_num": 28, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is on the second shelf of the cabinet where the stone pestle is placed?", + "answer": "The second compartment holds a beaker.", + "question_chinese": "放石杵那个柜子的第二格架子上放的是什么东西？", + "answer_chinese": "第二格放的是烧杯。", + "evidence": [ + { + "description": "The cabinet for storing stone pestles.", + "timestamps": { + "start": "00:16:02", + "end": "00:16:04" + } + }, + { + "description": "The second compartment holds a beaker.", + "timestamps": { + "start": "00:016:06", + "end": "00:16:24" + } + } + ], + "question_timestamps": "00:24:44", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The second compartment holds a beaker.", + "B": "The second compartment holds a flask.", + "C": "The second compartment holds a jar.", + "D": "The second compartment holds a bowl sitting alone on that shelf inside the cabinet.", + "E": "The second compartment holds a vase placed toward the back of the shelf." + }, + "answer_label": "A", + "answer_text": "The second compartment holds a beaker.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_9", + "correct_answer": "The second compartment holds a beaker.", + "original_options": { + "A": "The second compartment holds a beaker.", + "B": "The second compartment holds a flask.", + "C": "The second compartment holds a jar.", + "D": "The second compartment holds a bowl sitting alone on that shelf inside the cabinet.", + "E": "The second compartment holds a vase placed toward the back of the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_10", + "qid_num": 29, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is on the second shelf of the cabinet where the stone pestle is placed?", + "answer": "The second compartment holds a beaker.", + "question_chinese": "放石杵那个柜子的第二格架子上放的是什么东西？", + "answer_chinese": "第二格放的是烧杯。", + "evidence": [ + { + "description": "The cabinet for storing stone pestles.", + "timestamps": { + "start": "00:16:02", + "end": "00:16:04" + } + }, + { + "description": "The second compartment holds a beaker.", + "timestamps": { + "start": "00:016:06", + "end": "00:16:24" + } + } + ], + "question_timestamps": "00:28:44", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The second compartment holds a flask.", + "B": "The second compartment holds a cup.", + "C": "The second compartment holds a strainer placed on the shelf where the beaker would be.", + "D": "The second compartment holds a beaker.", + "E": "The second compartment holds a spoon resting inside the compartment instead of lab glassware." + }, + "answer_label": "D", + "answer_text": "The second compartment holds a beaker.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_10", + "correct_answer": "The second compartment holds a beaker.", + "original_options": { + "A": "The second compartment holds a flask.", + "B": "The second compartment holds a cup.", + "C": "The second compartment holds a strainer placed on the shelf where the beaker would be.", + "D": "The second compartment holds a beaker.", + "E": "The second compartment holds a spoon resting inside the compartment instead of lab glassware." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 14 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_8_T1", + "qid_num": 30, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I finally get the serum test tube?", + "answer": "Outside the door there was a row of low cabinets, and you took a test tube from the second compartment of the first cabinet.", + "question_chinese": "最后我在哪里拿到了血清试管？", + "answer_chinese": "门外有一排矮柜，你在第一个柜子的第二格拿到了试管。", + "evidence": [ + { + "description": "You walked out of the safety door on the right side of your workstation", + "timestamps": { + "start": "00:46:21", + "end": "00:46:30" + } + }, + { + "description": "There is a row of low cabinets over there.", + "timestamps": { + "start": "00:46:30", + "end": "00:46:33" + } + }, + { + "description": "You got the test tube in the second compartment of the first cabinet.", + "timestamps": { + "start": "00:46:35", + "end": "00:46:53" + } + } + ], + "question_timestamps": "00:48:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the first compartment of the first cabinet.", + "B": "From the second compartment of the second cabinet.", + "C": "Outside the door there was a row of low cabinets, and you took a test tube from the second compartment of the first cabinet.", + "D": "Outside the door, you grabbed the test tube from the first compartment of the first low cabinet.", + "E": "Outside the door, you took the test tube from the second compartment of the second low cabinet." + }, + "answer_label": "C", + "answer_text": "Outside the door there was a row of low cabinets, and you took a test tube from the second compartment of the first cabinet.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_8_T1", + "correct_answer": "Outside the door there was a row of low cabinets, and you took a test tube from the second compartment of the first cabinet.", + "original_options": { + "A": "From the first compartment of the first cabinet.", + "B": "From the second compartment of the second cabinet.", + "C": "Outside the door there was a row of low cabinets, and you took a test tube from the second compartment of the first cabinet.", + "D": "Outside the door, you grabbed the test tube from the first compartment of the first low cabinet.", + "E": "Outside the door, you took the test tube from the second compartment of the second low cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 17 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_15", + "qid_num": 31, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I finally get the serum test tube?", + "answer": "After stepping out of the safety door on the right side of your workstation, there is a row of low cabinets on the left. You got the test tube in the second compartment of the first cabinet.", + "question_chinese": "最后我在哪里拿到了血清试管？", + "answer_chinese": "走出了你工位右边的安全门后，左边有一排矮柜，你在第一个柜子的第二格拿到了试管。", + "evidence": [ + { + "description": "You walked out of the safety door on the right side of your workstation", + "timestamps": { + "start": "00:46:21", + "end": "00:46:30" + } + }, + { + "description": "There is a row of low cabinets over there.", + "timestamps": { + "start": "00:46:29", + "end": "00:46:33" + } + }, + { + "description": "You got the test tube in the second compartment of the first cabinet.", + "timestamps": { + "start": "00:46:35", + "end": "00:46:53" + } + } + ], + "question_timestamps": "00:50:22", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "From the first cabinet on the right.", + "B": "From the third compartment of the first cabinet.", + "C": "After stepping out of the safety door on the right side of your workstation, there is a row of low cabinets on the left. You got the test tube in the second compartment of the first cabinet.", + "D": "After stepping out of the safety door, you took the tube from the first compartment of the first cabinet.", + "E": "After stepping out of the safety door on your right, you took the tube from the second compartment of the second cabinet." + }, + "answer_label": "C", + "answer_text": "After stepping out of the safety door on the right side of your workstation, there is a row of low cabinets on the left. You got the test tube in the second compartment of the first cabinet.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_15", + "correct_answer": "After stepping out of the safety door on the right side of your workstation, there is a row of low cabinets on the left. You got the test tube in the second compartment of the first cabinet.", + "original_options": { + "A": "From the first cabinet on the right.", + "B": "From the third compartment of the first cabinet.", + "C": "After stepping out of the safety door on the right side of your workstation, there is a row of low cabinets on the left. You got the test tube in the second compartment of the first cabinet.", + "D": "After stepping out of the safety door, you took the tube from the first compartment of the first cabinet.", + "E": "After stepping out of the safety door on your right, you took the tube from the second compartment of the second cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 22 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_8_T2", + "qid_num": 32, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I finally get the serum test tube?", + "answer": "After you walked out through the emergency door to your right by your workstation, there was a row of low cabinets on your left, and you took a test tube from the second compartment of the first cabinet.", + "question_chinese": "最后我在哪里拿到了血清试管？", + "answer_chinese": "走出了你工位右边的安全门后，左边有一排矮柜，你在第一个柜子的第二格拿到了试管。", + "evidence": [ + { + "description": "You walked out of the safety door on the right side of your workstation", + "timestamps": { + "start": "00:46:21", + "end": "00:46:30" + } + }, + { + "description": "There is a row of low cabinets over there.", + "timestamps": { + "start": "00:46:29", + "end": "00:46:33" + } + }, + { + "description": "You got the test tube in the second compartment of the first cabinet.", + "timestamps": { + "start": "00:46:35", + "end": "00:46:53" + } + } + ], + "question_timestamps": "00:53:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Through the door on your left.", + "B": "After you walked out through the emergency door to your right by your workstation, there was a row of low cabinets on your left, and you took a test tube from the second compartment of the first cabinet.", + "C": "Cabinets were on your right outside.", + "D": "After walking out the emergency door to your right, you took the test tube from the first compartment of the first cabinet.", + "E": "After walking out the emergency door to your right, you took the test tube from the second compartment of the second cabinet." + }, + "answer_label": "B", + "answer_text": "After you walked out through the emergency door to your right by your workstation, there was a row of low cabinets on your left, and you took a test tube from the second compartment of the first cabinet.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_8_T2", + "correct_answer": "After you walked out through the emergency door to your right by your workstation, there was a row of low cabinets on your left, and you took a test tube from the second compartment of the first cabinet.", + "original_options": { + "A": "Through the door on your left.", + "B": "After you walked out through the emergency door to your right by your workstation, there was a row of low cabinets on your left, and you took a test tube from the second compartment of the first cabinet.", + "C": "Cabinets were on your right outside.", + "D": "After walking out the emergency door to your right, you took the test tube from the first compartment of the first cabinet.", + "E": "After walking out the emergency door to your right, you took the test tube from the second compartment of the second cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 22 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_7", + "qid_num": 33, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many small potted plants with large leaves are there in that plate in total?", + "answer": "You took out four basins first. There are still nine basins left in the basins, making a total of thirteen.", + "question_chinese": "那盘大叶片小盆栽一共有多少盆？", + "answer_chinese": "你先拿出来了四盆，盆子里还剩下九盆，一共有13盆。", + "evidence": [ + { + "description": "You took out four basins first.", + "timestamps": { + "start": "00:13:11", + "end": "00:13:23" + } + }, + { + "description": "There are still nine basins left in the basin.", + "timestamps": { + "start": "00:13:44", + "end": "00:13:45" + } + } + ], + "question_timestamps": "00:20:21", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Twelve in total.", + "B": "Fourteen in total.", + "C": "You took out four basins first. There are still nine basins left in the basins, making a total of thirteen.", + "D": "You took out three basins first, and ten basins remained, making a total of thirteen.", + "E": "You took out five basins first, and nine basins remained, making a total of fourteen." + }, + "answer_label": "C", + "answer_text": "You took out four basins first. There are still nine basins left in the basins, making a total of thirteen.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_7", + "correct_answer": "You took out four basins first. There are still nine basins left in the basins, making a total of thirteen.", + "original_options": { + "A": "Twelve in total.", + "B": "Fourteen in total.", + "C": "You took out four basins first. There are still nine basins left in the basins, making a total of thirteen.", + "D": "You took out three basins first, and ten basins remained, making a total of thirteen.", + "E": "You took out five basins first, and nine basins remained, making a total of fourteen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "7c420d19-88b4-4092-bef1-a9bb99bbe308", + "video_length": "00:54:59", + "qid": "7c420d19-88b4-4092-bef1-a9bb99bbe308_8", + "qid_num": 34, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many small potted plants with large leaves are there in that plate in total?", + "answer": "You took out four basins first. There are still nine basins left in the basins, making a total of thirteen.", + "question_chinese": "那盘大叶片小盆栽一共有多少盆？", + "answer_chinese": "你先拿出来了四盆，盆子里还剩下九盆，一共有13盆。", + "evidence": [ + { + "description": "You took out four basins first.", + "timestamps": { + "start": "00:13:11", + "end": "00:33:23" + } + }, + { + "description": "There are still nine basins left in the basin.", + "timestamps": { + "start": "00:13:44", + "end": "00:13:45" + } + } + ], + "question_timestamps": "00:33:23", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Fifteen in total.", + "B": "You took out four basins first. There are still nine basins left in the basins, making a total of thirteen.", + "C": "Seventeen in total.", + "D": "You took out five basins first, and nine basins were left, making a total of fourteen.", + "E": "You took out four basins first, and ten basins were left, making a total of fourteen." + }, + "answer_label": "B", + "answer_text": "You took out four basins first. There are still nine basins left in the basins, making a total of thirteen.", + "q_id": "7c420d19-88b4-4092-bef1-a9bb99bbe308_8", + "correct_answer": "You took out four basins first. There are still nine basins left in the basins, making a total of thirteen.", + "original_options": { + "A": "Fifteen in total.", + "B": "You took out four basins first. There are still nine basins left in the basins, making a total of thirteen.", + "C": "Seventeen in total.", + "D": "You took out five basins first, and nine basins were left, making a total of fourteen.", + "E": "You took out four basins first, and ten basins were left, making a total of fourteen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + } +] diff --git a/QAs/7d0e0392-28b8-4cb8-955d-f5f1dfb16158-mcq.json b/QAs/7d0e0392-28b8-4cb8-955d-f5f1dfb16158-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..013799dce15cca08c54e95036ab69ec3181374b6 --- /dev/null +++ b/QAs/7d0e0392-28b8-4cb8-955d-f5f1dfb16158-mcq.json @@ -0,0 +1,1350 @@ +[ + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_10_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the workbench cabinet relative to me?", + "answer": "The workbench cabinet is to your back-right relative to you.", + "question_chinese": "工作台柜子在相对于我的什么方位？", + "answer_chinese": "工作台柜子相对于我的右后方", + "evidence": [ + { + "description": "After getting the pliers, close the workbench cabinet and go to repair the car.", + "timestamps": { + "start": "00:19:36", + "end": "00:19:41" + } + } + ], + "question_timestamps": "00:20:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The workbench cabinet is to your back-right relative to you.", + "B": "It was to your front-left.", + "C": "It was directly behind you.", + "D": "The workbench cabinet was positioned to your front-right relative to where you were standing.", + "E": "The workbench cabinet sat on your back-left side relative to your position in the room." + }, + "answer_label": "A", + "answer_text": "The workbench cabinet is to your back-right relative to you.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_10_T1", + "correct_answer": "The workbench cabinet is to your back-right relative to you.", + "original_options": { + "A": "The workbench cabinet is to your back-right relative to you.", + "B": "It was to your front-left.", + "C": "It was directly behind you.", + "D": "The workbench cabinet was positioned to your front-right relative to where you were standing.", + "E": "The workbench cabinet sat on your back-left side relative to your position in the room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_10_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the workbench cabinet relative to me?", + "answer": "The workbench cabinet is to your back-left relative to you.", + "question_chinese": "工作台柜子在相对于我的什么方位？", + "answer_chinese": "工作台柜子相对于我的左后方", + "evidence": [ + { + "description": "After getting the pliers, close the workbench cabinet and go to repair the car.", + "timestamps": { + "start": "00:19:36", + "end": "00:19:41" + } + } + ], + "question_timestamps": "00:31:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was to your front-right.", + "B": "The workbench cabinet is to your back-left relative to you.", + "C": "It was directly in front of you.", + "D": "The workbench cabinet was positioned to your back-right relative to where you were standing.", + "E": "The workbench cabinet sat on your front-left side relative to your position in the room." + }, + "answer_label": "B", + "answer_text": "The workbench cabinet is to your back-left relative to you.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_10_T2", + "correct_answer": "The workbench cabinet is to your back-left relative to you.", + "original_options": { + "A": "It was to your front-right.", + "B": "The workbench cabinet is to your back-left relative to you.", + "C": "It was directly in front of you.", + "D": "The workbench cabinet was positioned to your back-right relative to where you were standing.", + "E": "The workbench cabinet sat on your front-left side relative to your position in the room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 14, + 15 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_6_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many race cars did I tie a small green rope onto?", + "answer": "Two, and right now they’re both behind you.", + "question_chinese": "我一共给几辆赛车栓上了小绿绳？", + "answer_chinese": "两辆，目前都在你的后方。", + "evidence": [ + { + "description": "Saw the first race car tied with the small green rope", + "timestamps": { + "start": "00:07:31", + "end": "00:07:37" + } + }, + { + "description": "Tied the second one with the small green rope", + "timestamps": { + "start": "00:08:56", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:09:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, and it’s behind you.", + "B": "Three, and they’re all behind you.", + "C": "Two, and right now they’re both behind you.", + "D": "Four, and right now all of them are behind you in the same area.", + "E": "Zero, so there aren’t any race cars behind you with a green rope tied on." + }, + "answer_label": "C", + "answer_text": "Two, and right now they’re both behind you.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_6_T1", + "correct_answer": "Two, and right now they’re both behind you.", + "original_options": { + "A": "One, and it’s behind you.", + "B": "Three, and they’re all behind you.", + "C": "Two, and right now they’re both behind you.", + "D": "Four, and right now all of them are behind you in the same area.", + "E": "Zero, so there aren’t any race cars behind you with a green rope tied on." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 16 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_6_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many race cars did I tie a small green rope onto?", + "answer": "Two, and right now they’re both in front of you.", + "question_chinese": "我一共给几辆赛车栓上了小绿绳？", + "answer_chinese": "两辆，目前都在你的前方。", + "evidence": [ + { + "description": "Saw the first race car tied with the small green rope", + "timestamps": { + "start": "00:07:31", + "end": "00:07:37" + } + }, + { + "description": "Tied the second one with the small green rope", + "timestamps": { + "start": "00:08:56", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:10:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One race car.", + "B": "Three race cars.", + "C": "Four race cars, and right now they’re all lined up in front of you.", + "D": "Two, and right now they’re both in front of you.", + "E": "Zero race cars, so you didn’t tie the small green rope onto any of them." + }, + "answer_label": "D", + "answer_text": "Two, and right now they’re both in front of you.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_6_T2", + "correct_answer": "Two, and right now they’re both in front of you.", + "original_options": { + "A": "One race car.", + "B": "Three race cars.", + "C": "Four race cars, and right now they’re all lined up in front of you.", + "D": "Two, and right now they’re both in front of you.", + "E": "Zero race cars, so you didn’t tie the small green rope onto any of them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_9_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I went to get the pliers just now, how many red scissors did I see?", + "answer": "You just saw eight red scissors, all to your back-right.", + "question_chinese": "我刚刚去拿钳子的时候吗，看到了几把红色剪刀？", + "answer_chinese": "你刚刚看到了八把红色剪刀，都在你的右后方。", + "evidence": [ + { + "description": "aw eight red scissors", + "timestamps": { + "start": "00:19:28", + "end": "00:19:36" + } + } + ], + "question_timestamps": "00:19:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just saw eight red scissors, all to your back-right.", + "B": "You saw six red scissors.", + "C": "You saw ten red scissors.", + "D": "You just saw seven red scissors, all grouped together to your back-right.", + "E": "You just saw nine red scissors, and they were all positioned off to your back-right." + }, + "answer_label": "A", + "answer_text": "You just saw eight red scissors, all to your back-right.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_9_T1", + "correct_answer": "You just saw eight red scissors, all to your back-right.", + "original_options": { + "A": "You just saw eight red scissors, all to your back-right.", + "B": "You saw six red scissors.", + "C": "You saw ten red scissors.", + "D": "You just saw seven red scissors, all grouped together to your back-right.", + "E": "You just saw nine red scissors, and they were all positioned off to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 15 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_9_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I went to get the pliers just now, how many red scissors did I see?", + "answer": "You just saw eight red scissors, to your right.", + "question_chinese": "我刚刚去拿钳子的时候吗，看到了几把红色剪刀？", + "answer_chinese": "我刚刚看到了八把红色剪刀，在你的右方。", + "evidence": [ + { + "description": "aw eight red scissors", + "timestamps": { + "start": "00:19:28", + "end": "00:19:36" + } + } + ], + "question_timestamps": "00:42:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw six red scissors.", + "B": "You just saw eight red scissors, to your right.", + "C": "You saw ten red scissors.", + "D": "You just saw seven red scissors, all sitting to your right as you went for the pliers.", + "E": "You just saw nine red scissors, and they were all located to your right at that moment." + }, + "answer_label": "B", + "answer_text": "You just saw eight red scissors, to your right.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_9_T2", + "correct_answer": "You just saw eight red scissors, to your right.", + "original_options": { + "A": "You saw six red scissors.", + "B": "You just saw eight red scissors, to your right.", + "C": "You saw ten red scissors.", + "D": "You just saw seven red scissors, all sitting to your right as you went for the pliers.", + "E": "You just saw nine red scissors, and they were all located to your right at that moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_2_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different kinds of tools did I just take for repairing the race car?", + "answer": "You just took three types of tools for repairing the race car, and they’re all on the ground behind you.", + "question_chinese": "我刚刚拿了几种修理赛车的工具？", + "answer_chinese": "你刚刚拿了维修赛车的三种工具，都在你后面的地面上。", + "evidence": [ + { + "description": "saw three types of tools", + "timestamps": { + "start": "00:00:15", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:02:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One type of tool.", + "B": "Two types of tools.", + "C": "You just took four types of tools for repairing the race car, and they’re all on the ground behind you.", + "D": "You just took three types of tools for repairing the race car, and they’re all on the ground behind you.", + "E": "You just took five types of tools for repairing the race car, and all of them are on the ground behind you." + }, + "answer_label": "D", + "answer_text": "You just took three types of tools for repairing the race car, and they’re all on the ground behind you.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_2_T1", + "correct_answer": "You just took three types of tools for repairing the race car, and they’re all on the ground behind you.", + "original_options": { + "A": "One type of tool.", + "B": "Two types of tools.", + "C": "You just took four types of tools for repairing the race car, and they’re all on the ground behind you.", + "D": "You just took three types of tools for repairing the race car, and they’re all on the ground behind you.", + "E": "You just took five types of tools for repairing the race car, and all of them are on the ground behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 21, + 22 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_2_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different kinds of tools did I just take for repairing the race car?", + "answer": "You just took three types of tools for repairing the race car.", + "question_chinese": "我刚刚拿了几种修理赛车的工具？", + "answer_chinese": "你刚刚拿了维修赛车的三种工具", + "evidence": [ + { + "description": "saw three types of tools", + "timestamps": { + "start": "00:00:15", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:32:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One type of tool.", + "B": "Two types of tools.", + "C": "You just took four types of tools for repairing the race car.", + "D": "You just took five types of tools for repairing the race car just now.", + "E": "You just took three types of tools for repairing the race car." + }, + "answer_label": "E", + "answer_text": "You just took three types of tools for repairing the race car.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_2_T2", + "correct_answer": "You just took three types of tools for repairing the race car.", + "original_options": { + "A": "One type of tool.", + "B": "Two types of tools.", + "C": "You just took four types of tools for repairing the race car.", + "D": "You just took five types of tools for repairing the race car just now.", + "E": "You just took three types of tools for repairing the race car." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_4_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where on the workbench did I put the inspection report I was just holding?", + "answer": "It was put into the drawer of the workbench, behind you.", + "question_chinese": "我刚刚拿的检验单放到了工作台的哪里？", + "answer_chinese": "放到了工作台的抽屉中，在你的后方。", + "evidence": [ + { + "description": "Was placed in the workbench drawer. Closed the drawer.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:35" + } + } + ], + "question_timestamps": "00:03:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was put into the drawer of the workbench, behind you.", + "B": "On the workbench surface.", + "C": "In the drawer in front of you.", + "D": "It was placed on the surface of the workbench in front of you, not tucked into any drawer.", + "E": "It was put into the workbench drawer to your right, rather than the drawer behind you." + }, + "answer_label": "A", + "answer_text": "It was put into the drawer of the workbench, behind you.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_4_T1", + "correct_answer": "It was put into the drawer of the workbench, behind you.", + "original_options": { + "A": "It was put into the drawer of the workbench, behind you.", + "B": "On the workbench surface.", + "C": "In the drawer in front of you.", + "D": "It was placed on the surface of the workbench in front of you, not tucked into any drawer.", + "E": "It was put into the workbench drawer to your right, rather than the drawer behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 18, + 16 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_4_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where on the workbench did I put the inspection report I was just holding?", + "answer": "It was placed on the surface of the workbench, in front of you.", + "question_chinese": "我刚刚拿的检验单放到了工作台的哪里？", + "answer_chinese": "放到了工作台的表面上，在你的前方。", + "evidence": [ + { + "description": "Was placed in the workbench drawer. Closed the drawer.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:35" + } + } + ], + "question_timestamps": "00:04:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the workbench drawer.", + "B": "It was placed on the surface of the workbench, in front of you.", + "C": "On the surface behind you.", + "D": "It was put into the drawer of the workbench in front of you, instead of left out on top.", + "E": "It was placed on the surface of the workbench to your right, not directly in front of you." + }, + "answer_label": "B", + "answer_text": "It was placed on the surface of the workbench, in front of you.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_4_T2", + "correct_answer": "It was placed on the surface of the workbench, in front of you.", + "original_options": { + "A": "Into the workbench drawer.", + "B": "It was placed on the surface of the workbench, in front of you.", + "C": "On the surface behind you.", + "D": "It was put into the drawer of the workbench in front of you, instead of left out on top.", + "E": "It was placed on the surface of the workbench to your right, not directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 19, + 18 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_5_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I pushed the cart with the small green rope tied on it to go outside just now?", + "answer": "You tied the cart with the small green rope and then pushed it straight forward.", + "question_chinese": "我刚刚推拴上小绿绳的小车到户外走了什么路径？", + "answer_chinese": "你把小车拴上小绿绳，并直接把车向前推。", + "evidence": [ + { + "description": "Attached the small green rope", + "timestamps": { + "start": "00:08:56", + "end": "00:09:00" + } + }, + { + "description": "Pushed the car straight forward", + "timestamps": { + "start": "00:09:01", + "end": "00:09:10" + } + } + ], + "question_timestamps": "00:09:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You pushed it straight backward.", + "B": "You tied the cart with the small green rope and then pushed it straight forward.", + "C": "You turned left, then pushed forward.", + "D": "You tied the small green rope to the cart and then turned right before pushing it forward.", + "E": "You tied the cart with the small green rope and pushed it forward while weaving in a slight zigzag." + }, + "answer_label": "B", + "answer_text": "You tied the cart with the small green rope and then pushed it straight forward.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_5_T1", + "correct_answer": "You tied the cart with the small green rope and then pushed it straight forward.", + "original_options": { + "A": "You pushed it straight backward.", + "B": "You tied the cart with the small green rope and then pushed it straight forward.", + "C": "You turned left, then pushed forward.", + "D": "You tied the small green rope to the cart and then turned right before pushing it forward.", + "E": "You tied the cart with the small green rope and pushed it forward while weaving in a slight zigzag." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 19 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_5_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I pushed the cart with the small green rope tied on it to go outside just now?", + "answer": "You tied the cart with the small green rope and then pushed it straight forward. Right now, the cart is to your back-right.", + "question_chinese": "我刚刚推拴上小绿绳的小车到户外走了什么路径？", + "answer_chinese": "你把小车拴上小绿绳，并直接把车向前推。目前车在你的右后方。", + "evidence": [ + { + "description": "Attached the small green rope", + "timestamps": { + "start": "00:08:56", + "end": "00:09:00" + } + }, + { + "description": "Pushed the car straight forward", + "timestamps": { + "start": "00:09:01", + "end": "00:09:10" + } + } + ], + "question_timestamps": "00:16:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You pushed it straight; it’s back-left.", + "B": "You pushed it straight; it’s front-right.", + "C": "You tied the cart with the small green rope and then pushed it straight forward. Right now, the cart is to your back-right.", + "D": "You tied the small green rope, pushed the cart straight ahead, and now it sits to your front-left.", + "E": "After tying the cart with the green rope, you pushed forward, and it ended up directly behind you." + }, + "answer_label": "C", + "answer_text": "You tied the cart with the small green rope and then pushed it straight forward. Right now, the cart is to your back-right.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_5_T2", + "correct_answer": "You tied the cart with the small green rope and then pushed it straight forward. Right now, the cart is to your back-right.", + "original_options": { + "A": "You pushed it straight; it’s back-left.", + "B": "You pushed it straight; it’s front-right.", + "C": "You tied the cart with the small green rope and then pushed it straight forward. Right now, the cart is to your back-right.", + "D": "You tied the small green rope, pushed the cart straight ahead, and now it sits to your front-left.", + "E": "After tying the cart with the green rope, you pushed forward, and it ended up directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_12_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many fire extinguishers did I see just now?", + "answer": "You saw two fire extinguishers behind you.", + "question_chinese": "我刚刚看到了几个灭火器？", + "answer_chinese": "你看到了两个灭火器，在你的后方。", + "evidence": [ + { + "description": "saw two fire extinguishers", + "timestamps": { + "start": "00:00:53", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one behind you.", + "B": "You saw three behind you.", + "C": "You saw four fire extinguishers behind you, grouped together in the area you just passed.", + "D": "You saw two fire extinguishers behind you.", + "E": "You saw zero fire extinguishers behind you, with none visible when you looked back." + }, + "answer_label": "D", + "answer_text": "You saw two fire extinguishers behind you.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_12_T2", + "correct_answer": "You saw two fire extinguishers behind you.", + "original_options": { + "A": "You saw one behind you.", + "B": "You saw three behind you.", + "C": "You saw four fire extinguishers behind you, grouped together in the area you just passed.", + "D": "You saw two fire extinguishers behind you.", + "E": "You saw zero fire extinguishers behind you, with none visible when you looked back." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_12_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many fire extinguishers did I see just now?", + "answer": "You saw two fire extinguishers, to your back-right.", + "question_chinese": "我刚刚看到了几个灭火器？", + "answer_chinese": "你看到了两个灭火器，在你的右后方。", + "evidence": [ + { + "description": "saw two fire extinguishers", + "timestamps": { + "start": "00:00:53", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:31:55", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one, back-right.", + "B": "You saw three, back-right.", + "C": "You saw two fire extinguishers, to your back-right.", + "D": "You saw four fire extinguishers to your back-right, all visible in the same spot behind you.", + "E": "You saw zero fire extinguishers to your back-right, with nothing there when you checked that side." + }, + "answer_label": "C", + "answer_text": "You saw two fire extinguishers, to your back-right.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_12_T1", + "correct_answer": "You saw two fire extinguishers, to your back-right.", + "original_options": { + "A": "You saw one, back-right.", + "B": "You saw three, back-right.", + "C": "You saw two fire extinguishers, to your back-right.", + "D": "You saw four fire extinguishers to your back-right, all visible in the same spot behind you.", + "E": "You saw zero fire extinguishers to your back-right, with nothing there when you checked that side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_11_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the red plastic bucket?", + "answer": "Yes—you can reach it directly. It’s to your right, so you can grab it just by extending your hand.", + "question_chinese": "我可以直接够到红色塑料桶吗？", + "answer_chinese": "可以直接够到，他就在你的右方，伸手就能拿到。", + "evidence": [ + { + "description": "saw the red plastic bucket", + "timestamps": { + "start": "00:39:10", + "end": "00:39:14" + } + } + ], + "question_timestamps": "00:39:36", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can reach it directly. It’s to your right, so you can grab it just by extending your hand.", + "B": "No—you can’t reach it directly. It’s to your right, so you’d need to step closer first." + }, + "answer_label": "A", + "answer_text": "Yes—you can reach it directly. It’s to your right, so you can grab it just by extending your hand.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_11_T1", + "correct_answer": "Yes—you can reach it directly. It’s to your right, so you can grab it just by extending your hand.", + "original_options": { + "A": "Yes—you can reach it directly. It’s to your right, so you can grab it just by extending your hand.", + "B": "No—you can’t reach it directly. It’s to your right, so you’d need to step closer first." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_11_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the red plastic bucket?", + "answer": "No—you can’t reach it directly. It’s directly behind you.", + "question_chinese": "我可以直接够到红色塑料桶吗？", + "answer_chinese": "不可以直接够到，他就在你的正后方。", + "evidence": [ + { + "description": "saw the red plastic bucket", + "timestamps": { + "start": "00:39:10", + "end": "00:39:14" + } + } + ], + "question_timestamps": "00:42:01", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can reach it directly. It’s directly behind you, so you can grab it just by reaching back.", + "B": "No—you can’t reach it directly. It’s directly behind you." + }, + "answer_label": "B", + "answer_text": "No—you can’t reach it directly. It’s directly behind you.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_11_T2", + "correct_answer": "No—you can’t reach it directly. It’s directly behind you.", + "original_options": { + "A": "Yes—you can reach it directly. It’s directly behind you, so you can grab it just by reaching back.", + "B": "No—you can’t reach it directly. It’s directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_8_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where outdoors did I put the yellow race car?", + "answer": "It was left next to the unloading rack, to your back-right.", + "question_chinese": "我把黄色赛车放在户外的哪里？", + "answer_chinese": "放在卸货架旁，右后方。", + "evidence": [ + { + "description": "Went straight, left the main gate", + "timestamps": { + "start": "00:15:21", + "end": "00:15:35" + } + }, + { + "description": "Turned right and parked next to the unloading rack", + "timestamps": { + "start": "00:15:35", + "end": "00:15:40" + } + } + ], + "question_timestamps": "00:18:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Left next to the unloading rack, back-left.", + "B": "Left next to the unloading rack, front-right.", + "C": "It was placed next to the unloading rack, but it was in front of you rather than back-right outdoors.", + "D": "You left it next to the unloading rack on your left side, not in the back-right area outdoors.", + "E": "It was left next to the unloading rack, to your back-right." + }, + "answer_label": "E", + "answer_text": "It was left next to the unloading rack, to your back-right.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_8_T1", + "correct_answer": "It was left next to the unloading rack, to your back-right.", + "original_options": { + "A": "Left next to the unloading rack, back-left.", + "B": "Left next to the unloading rack, front-right.", + "C": "It was placed next to the unloading rack, but it was in front of you rather than back-right outdoors.", + "D": "You left it next to the unloading rack on your left side, not in the back-right area outdoors.", + "E": "It was left next to the unloading rack, to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 18 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_8_T2", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where outdoors did I put the yellow race car?", + "answer": "It was left next to the unloading rack, to your right.", + "question_chinese": "我把黄色赛车放在户外的哪里？", + "answer_chinese": "放在卸货架旁,在你的右边。", + "evidence": [ + { + "description": "Went straight, left the main gate", + "timestamps": { + "start": "00:15:21", + "end": "00:15:35" + } + }, + { + "description": "Turned right and parked next to the unloading rack", + "timestamps": { + "start": "00:15:35", + "end": "00:15:40" + } + } + ], + "question_timestamps": "00:18:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was left next to the unloading rack, to your right.", + "B": "Left next to the unloading rack, left.", + "C": "Left next to the unloading rack, back-right.", + "D": "It was left next to the unloading rack, directly in front of you instead of off to your right.", + "E": "You put it right next to the unloading rack on your left, rather than leaving it on your right side." + }, + "answer_label": "A", + "answer_text": "It was left next to the unloading rack, to your right.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_8_T2", + "correct_answer": "It was left next to the unloading rack, to your right.", + "original_options": { + "A": "It was left next to the unloading rack, to your right.", + "B": "Left next to the unloading rack, left.", + "C": "Left next to the unloading rack, back-right.", + "D": "It was left next to the unloading rack, directly in front of you instead of off to your right.", + "E": "You put it right next to the unloading rack on your left, rather than leaving it on your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 20 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_1_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the fire extinguisher right now?", + "answer": "Yes. The two fire extinguishers are right behind you.", + "question_chinese": "我现在能直接碰到灭火器吗？", + "answer_chinese": "可以，两个灭火器就在你的背后", + "evidence": [ + { + "description": "saw two fire extinguishers", + "timestamps": { + "start": "00:00:53", + "end": "00:01:00" + } + }, + { + "description": "Turned around and saw the race car", + "timestamps": { + "start": "00:01:00", + "end": "00:01:03" + } + }, + { + "description": "Repaired the race car", + "timestamps": { + "start": "00:01:08", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:02:51", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. The two fire extinguishers are right behind you.", + "B": "Yes. The two fire extinguishers are right behind you." + }, + "answer_label": "B", + "answer_text": "Yes. The two fire extinguishers are right behind you.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_1_T1", + "correct_answer": "Yes. The two fire extinguishers are right behind you.", + "original_options": { + "A": "No. The two fire extinguishers are right behind you.", + "B": "Yes. The two fire extinguishers are right behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_1_T2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the fire extinguisher right now?", + "answer": "No, the two fire extinguishers are right to your left.", + "question_chinese": "我现在能直接碰到灭火器吗？", + "answer_chinese": "不可以，两个灭火器就在你的左边。", + "evidence": [ + { + "description": "saw two fire extinguishers", + "timestamps": { + "start": "00:00:53", + "end": "00:01:00" + } + }, + { + "description": "Turned around and saw the race car", + "timestamps": { + "start": "00:01:00", + "end": "00:01:03" + } + }, + { + "description": "Repaired the race car", + "timestamps": { + "start": "00:01:08", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:04:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, the two fire extinguishers are right to your left.", + "B": "Yes, the two fire extinguishers are right to your left." + }, + "answer_label": "A", + "answer_text": "No, the two fire extinguishers are right to your left.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_1_T2", + "correct_answer": "No, the two fire extinguishers are right to your left.", + "original_options": { + "A": "No, the two fire extinguishers are right to your left.", + "B": "Yes, the two fire extinguishers are right to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_7_T2", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did I drive the yellow race car along from the garage to next to the unloading rack?", + "answer": "Go straight, exit through the main door, turn right, and park next to the unloading rack. It’s now parked to your back-left.", + "question_chinese": "我驾驶黄色赛车从车库到卸货架旁经过了什么路径？", + "answer_chinese": "直走，离开大门，向右转并在卸货架旁停车。现在停在你的左后方。", + "evidence": [ + { + "description": "Went straight, left the main gate", + "timestamps": { + "start": "00:15:21", + "end": "00:15:35" + } + }, + { + "description": "Turned right and parked next to the unloading rack", + "timestamps": { + "start": "00:15:35", + "end": "00:15:40" + } + } + ], + "question_timestamps": "00:15:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Exit the side door and turn right.", + "B": "Turn left after the main door.", + "C": "Go straight, exit through the main door, turn right, then park in front of the unloading rack.", + "D": "Go straight out the main door, turn right, and park by the unloading rack; it ends up to your back-right.", + "E": "Go straight, exit through the main door, turn right, and park next to the unloading rack. It’s now parked to your back-left." + }, + "answer_label": "E", + "answer_text": "Go straight, exit through the main door, turn right, and park next to the unloading rack. It’s now parked to your back-left.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_7_T2", + "correct_answer": "Go straight, exit through the main door, turn right, and park next to the unloading rack. It’s now parked to your back-left.", + "original_options": { + "A": "Exit the side door and turn right.", + "B": "Turn left after the main door.", + "C": "Go straight, exit through the main door, turn right, then park in front of the unloading rack.", + "D": "Go straight out the main door, turn right, and park by the unloading rack; it ends up to your back-right.", + "E": "Go straight, exit through the main door, turn right, and park next to the unloading rack. It’s now parked to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 20 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_7_T1", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did I drive the yellow race car along from the garage to next to the unloading rack?", + "answer": "Go straight, exit through the main door, turn right, and park next to the unloading rack.", + "question_chinese": "我驾驶黄色赛车从车库到卸货架旁经过了什么路径？", + "answer_chinese": "直走，离开大门，向右转并在卸货架旁停车。", + "evidence": [ + { + "description": "Went straight, left the main gate", + "timestamps": { + "start": "00:15:21", + "end": "00:15:35" + } + }, + { + "description": "Turned right and parked next to the unloading rack", + "timestamps": { + "start": "00:15:35", + "end": "00:15:40" + } + } + ], + "question_timestamps": "00:16:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Exit the side door, then turn right.", + "B": "Turn left after the main door.", + "C": "Go straight, exit through the main door, turn right, and park in front of the unloading rack.", + "D": "Go straight, exit through the main door, turn right, and park next to the unloading rack.", + "E": "Go straight, exit through the main door, turn right twice, and then park next to the unloading rack." + }, + "answer_label": "D", + "answer_text": "Go straight, exit through the main door, turn right, and park next to the unloading rack.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_7_T1", + "correct_answer": "Go straight, exit through the main door, turn right, and park next to the unloading rack.", + "original_options": { + "A": "Exit the side door, then turn right.", + "B": "Turn left after the main door.", + "C": "Go straight, exit through the main door, turn right, and park in front of the unloading rack.", + "D": "Go straight, exit through the main door, turn right, and park next to the unloading rack.", + "E": "Go straight, exit through the main door, turn right twice, and then park next to the unloading rack." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 18 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_3_T1", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I take to move the race car inspection form from the race car to the workbench?", + "answer": "The inspection sheet was taken down from the race car, carried through the aisle between the race cars to the space between the yellow car and the workbench, and then placed into the drawer of the workbench.", + "question_chinese": "赛车检验单从赛车上拿下到工作台我经过了什么路径？", + "answer_chinese": "从赛车上去下检验单，经过赛车间的过道，到黄色车和工作台中间，被放到工作台的抽屉中。", + "evidence": [ + { + "description": "Took off the inspection sheet from the race car", + "timestamps": { + "start": "00:02:26", + "end": "00:02:28" + } + }, + { + "description": "Passed through the aisle between the race cars, to the middle of the yellow car and the workbench", + "timestamps": { + "start": "00:02:29", + "end": "00:02:31" + } + }, + { + "description": "Was placed in the workbench drawer", + "timestamps": { + "start": "00:02:31", + "end": "00:02:32" + } + } + ], + "question_timestamps": "00:03:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You set it on the workbench.", + "B": "You carried it around the cars.", + "C": "You took the sheet down, walked straight to the workbench without the aisle, and placed it into the drawer.", + "D": "You removed the sheet, went through the aisle between cars, then left it on top of the workbench surface.", + "E": "The inspection sheet was taken down from the race car, carried through the aisle between the race cars to the space between the yellow car and the workbench, and then placed into the drawer of the workbench." + }, + "answer_label": "E", + "answer_text": "The inspection sheet was taken down from the race car, carried through the aisle between the race cars to the space between the yellow car and the workbench, and then placed into the drawer of the workbench.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_3_T1", + "correct_answer": "The inspection sheet was taken down from the race car, carried through the aisle between the race cars to the space between the yellow car and the workbench, and then placed into the drawer of the workbench.", + "original_options": { + "A": "You set it on the workbench.", + "B": "You carried it around the cars.", + "C": "You took the sheet down, walked straight to the workbench without the aisle, and placed it into the drawer.", + "D": "You removed the sheet, went through the aisle between cars, then left it on top of the workbench surface.", + "E": "The inspection sheet was taken down from the race car, carried through the aisle between the race cars to the space between the yellow car and the workbench, and then placed into the drawer of the workbench." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158", + "video_length": "00:42:48", + "qid": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_3_T2", + "qid_num": 24, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I take to move the race car inspection form from the race car to the workbench?", + "answer": "The inspection sheet was taken down from the race car, carried through the aisle between the race cars to the space between the yellow car and the workbench, and then placed into the drawer of the workbench in front of you.", + "question_chinese": "赛车检验单从赛车上拿下到工作台我经过了什么路径？", + "answer_chinese": "从赛车上去下检验单，经过赛车间的过道，到黄色车和工作台中间，被放到你前方工作台的抽屉中。", + "evidence": [ + { + "description": "Took off the inspection sheet from the race car", + "timestamps": { + "start": "00:02:26", + "end": "00:02:28" + } + }, + { + "description": "Passed through the aisle between the race cars, to the middle of the yellow car and the workbench", + "timestamps": { + "start": "00:02:29", + "end": "00:02:31" + } + }, + { + "description": "Was placed in the workbench drawer", + "timestamps": { + "start": "00:02:31", + "end": "00:02:32" + } + } + ], + "question_timestamps": "00:40:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The inspection sheet was taken down from the race car, carried through the aisle between the race cars to the space between the yellow car and the workbench, and then placed into the drawer of the workbench in front of you.", + "B": "You put it in the drawer behind you.", + "C": "You left it on the race car.", + "D": "You took it down, went through the aisle, and set it on top of the workbench in front of you.", + "E": "You removed it from the race car, walked around the race cars, then placed it into the workbench drawer in front." + }, + "answer_label": "A", + "answer_text": "The inspection sheet was taken down from the race car, carried through the aisle between the race cars to the space between the yellow car and the workbench, and then placed into the drawer of the workbench in front of you.", + "q_id": "7d0e0392-28b8-4cb8-955d-f5f1dfb16158_3_T2", + "correct_answer": "The inspection sheet was taken down from the race car, carried through the aisle between the race cars to the space between the yellow car and the workbench, and then placed into the drawer of the workbench in front of you.", + "original_options": { + "A": "The inspection sheet was taken down from the race car, carried through the aisle between the race cars to the space between the yellow car and the workbench, and then placed into the drawer of the workbench in front of you.", + "B": "You put it in the drawer behind you.", + "C": "You left it on the race car.", + "D": "You took it down, went through the aisle, and set it on top of the workbench in front of you.", + "E": "You removed it from the race car, walked around the race cars, then placed it into the workbench drawer in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 20, + 21 + ] + } +] diff --git a/QAs/7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5-mcq.json b/QAs/7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1a18316c713f7fa29873bd31606bc976f59d32c9 --- /dev/null +++ b/QAs/7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5-mcq.json @@ -0,0 +1,697 @@ +[ + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_13", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Are taco shells and oven mitts close to each other now? How far approximately?", + "answer": "Rather far, the taco shells is on the right of oven mitts, across the stove, about thirty centimeters apart.", + "question_chinese": "塔可饼皮和隔热手套现在离得近吗?大概多远?", + "answer_chinese": "较远，塔可饼皮在隔热手套的右边，跨过灶台，相距约30厘米。", + "evidence": [ + { + "description": "The taco shells is on the right of oven mitts, across the stove, about 30 centimeters apart.", + "timestamps": { + "start": "00:16:30", + "end": "00:17:00" + } + } + ], + "question_timestamps": "00:17:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Rather far, the taco shells is on the right of oven mitts, across the stove, about thirty centimeters apart.", + "B": "Very close, about ten centimeters.", + "C": "Moderately close, about fifteen centimeters.", + "D": "Rather far: the taco shells are left of the oven mitts, across the stove, about thirty centimeters apart.", + "E": "Rather far: the taco shells are right of the oven mitts across the stove, but closer to sixty centimeters apart." + }, + "answer_label": "A", + "answer_text": "Rather far, the taco shells is on the right of oven mitts, across the stove, about thirty centimeters apart.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_13", + "correct_answer": "Rather far, the taco shells is on the right of oven mitts, across the stove, about thirty centimeters apart.", + "original_options": { + "A": "Rather far, the taco shells is on the right of oven mitts, across the stove, about thirty centimeters apart.", + "B": "Very close, about ten centimeters.", + "C": "Moderately close, about fifteen centimeters.", + "D": "Rather far: the taco shells are left of the oven mitts, across the stove, about thirty centimeters apart.", + "E": "Rather far: the taco shells are right of the oven mitts across the stove, but closer to sixty centimeters apart." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 20 + ] + }, + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_14", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Are taco shells and oven mitts close to each other now? How far approximately?", + "answer": "Very close, the taco shells are next to the oven mitt, about ten centimeters away.", + "question_chinese": "塔可饼皮和隔热手套现在离得近吗?大概多远?", + "answer_chinese": "很近，塔可饼皮在隔热手套的旁边，大概10厘米。", + "evidence": [ + { + "description": "The taco shells were picked up and placed on the left side of the stove, about 10 cm from the oven mitt.", + "timestamps": { + "start": "00:17:14", + "end": "00:17:38" + } + } + ], + "question_timestamps": "00:19:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Very close, about eight centimeters away.", + "B": "Very close, the taco shells are next to the oven mitt, about ten centimeters away.", + "C": "Very close, about twelve centimeters away.", + "D": "Moderately close: the taco shells sit near the oven mitt, roughly twenty centimeters away, not right next to it.", + "E": "Rather far: the taco shells are across the stove from the oven mitt, around thirty centimeters apart." + }, + "answer_label": "B", + "answer_text": "Very close, the taco shells are next to the oven mitt, about ten centimeters away.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_14", + "correct_answer": "Very close, the taco shells are next to the oven mitt, about ten centimeters away.", + "original_options": { + "A": "Very close, about eight centimeters away.", + "B": "Very close, the taco shells are next to the oven mitt, about ten centimeters away.", + "C": "Very close, about twelve centimeters away.", + "D": "Moderately close: the taco shells sit near the oven mitt, roughly twenty centimeters away, not right next to it.", + "E": "Rather far: the taco shells are across the stove from the oven mitt, around thirty centimeters apart." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 17 + ] + }, + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_1_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the garlic move to its current position from the refrigerator?", + "answer": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop.", + "question_chinese": "大蒜从冰箱到现在的位置经过了什么路径?", + "answer_chinese": "大蒜从冰箱的第二格被取下，向右移动经过灶台旁，现在放置在操作台上。", + "evidence": [ + { + "description": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop.", + "timestamps": { + "start": "00:03:44", + "end": "00:03:48" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Taken from the top shelf, moved left.", + "B": "Removed from the door compartment, placed down.", + "C": "The garlic was taken from a lower compartment, carried left past the stove, and placed on the stovetop.", + "D": "The garlic was pulled from the vegetable drawer, moved forward past the microwave, and set down on the dining table.", + "E": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop." + }, + "answer_label": "E", + "answer_text": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_1_T1", + "correct_answer": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop.", + "original_options": { + "A": "Taken from the top shelf, moved left.", + "B": "Removed from the door compartment, placed down.", + "C": "The garlic was taken from a lower compartment, carried left past the stove, and placed on the stovetop.", + "D": "The garlic was pulled from the vegetable drawer, moved forward past the microwave, and set down on the dining table.", + "E": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 20 + ] + }, + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_1_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the garlic move to its current position from the refrigerator?", + "answer": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop.", + "question_chinese": "大蒜从冰箱到现在的位置经过了什么路径?", + "answer_chinese": "大蒜从冰箱的第二格被取下，向右移动经过灶台旁，现在放置在操作台上。", + "evidence": [ + { + "description": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop.", + "timestamps": { + "start": "00:03:44", + "end": "00:03:48" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop.", + "B": "It was set down beside the stove.", + "C": "You moved it left past the stove.", + "D": "You removed the garlic from the top compartment, carried it right past the stove, then set it on the counter.", + "E": "The garlic came from the fridge door bin, was carried right past the stove, and placed on the countertop." + }, + "answer_label": "A", + "answer_text": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_1_T2", + "correct_answer": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop.", + "original_options": { + "A": "The garlic was taken from the second compartment of the refrigerator, moved to the right past the stove, and is now placed on the countertop.", + "B": "It was set down beside the stove.", + "C": "You moved it left past the stove.", + "D": "You removed the garlic from the top compartment, carried it right past the stove, then set it on the counter.", + "E": "The garlic came from the fridge door bin, was carried right past the stove, and placed on the countertop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 19 + ] + }, + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_11", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands now?", + "answer": "Yes, you can wash your hands.", + "question_chinese": "我现在可以洗手吗？", + "answer_chinese": "是的你可以洗手？", + "evidence": [ + { + "description": "saw the tap.", + "timestamps": { + "start": "00:15:12", + "end": "00:15:18" + } + } + ], + "question_timestamps": "00:15:19", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can wash your hands.", + "B": "No, you cannot wash your hands right now." + }, + "answer_label": "A", + "answer_text": "Yes, you can wash your hands.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_11", + "correct_answer": "Yes, you can wash your hands.", + "original_options": { + "A": "Yes, you can wash your hands.", + "B": "No, you cannot wash your hands right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_12", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands now?", + "answer": "No, you cannot touch the faucet to wash your hands right now.", + "question_chinese": "我现在可以洗手吗？", + "answer_chinese": "不，你现在不能接触到水龙头洗手。", + "evidence": [ + { + "description": "saw the tap.", + "timestamps": { + "start": "00:15:12", + "end": "00:15:18" + } + }, + { + "description": "The sink is to the right rear of this person.", + "timestamps": { + "start": "00:16:50", + "end": "00:17:00" + } + } + ], + "question_timestamps": "00:17:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you can touch the faucet to wash your hands right now.", + "B": "No, you cannot touch the faucet to wash your hands right now." + }, + "answer_label": "B", + "answer_text": "No, you cannot touch the faucet to wash your hands right now.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_12", + "correct_answer": "No, you cannot touch the faucet to wash your hands right now.", + "original_options": { + "A": "Yes, you can touch the faucet to wash your hands right now.", + "B": "No, you cannot touch the faucet to wash your hands right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_8", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I'm done using this peeler. Where should I put it?", + "answer": "The peeler should be in the cabinet to your lower right.", + "question_chinese": "我用完这个削皮刀了，他应该放在哪里？", + "answer_chinese": "削皮刀应该放在你右下方的柜子里。", + "evidence": [ + { + "description": "The person retrieved a peeler from the cabinet below and to his right.", + "timestamps": { + "start": "00:14:25", + "end": "00:14:33" + } + }, + { + "description": "The cabinet is to the right and below this person.", + "timestamps": { + "start": "00:15:20", + "end": "00:15:40" + } + }, + { + "description": "The peeler was placed on the counter after use.", + "timestamps": { + "start": "00:15:01", + "end": "00:15:04" + } + } + ], + "question_timestamps": "00:15:55", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The peeler should be in the cabinet to your lower right.", + "B": "Put it in the drawer left.", + "C": "Put it on the counter.", + "D": "Place the peeler in the cabinet to your lower left, next to where the other utensils are stored.", + "E": "Put the peeler into the cabinet at your upper right, rather than leaving it out on the counter." + }, + "answer_label": "A", + "answer_text": "The peeler should be in the cabinet to your lower right.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_8", + "correct_answer": "The peeler should be in the cabinet to your lower right.", + "original_options": { + "A": "The peeler should be in the cabinet to your lower right.", + "B": "Put it in the drawer left.", + "C": "Put it on the counter.", + "D": "Place the peeler in the cabinet to your lower left, next to where the other utensils are stored.", + "E": "Put the peeler into the cabinet at your upper right, rather than leaving it out on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 18 + ] + }, + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_7", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I'm done using this peeler. Where should I put it?", + "answer": "The peeler should be in the cabinet behind your left side.", + "question_chinese": "我用完这个削皮刀了，他应该放在哪里？", + "answer_chinese": "削皮刀应该放在你左后方的柜子里。", + "evidence": [ + { + "description": "The person retrieved a peeler from the cabinet.", + "timestamps": { + "start": "00:14:25", + "end": "00:14:33" + } + }, + { + "description": "The cabinet is currently behind you to your left.", + "timestamps": { + "start": "00:17:59", + "end": "00:18:02" + } + }, + { + "description": "The peeler was placed on the counter after use.", + "timestamps": { + "start": "00:15:01", + "end": "00:15:04" + } + } + ], + "question_timestamps": "00:18:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Put it on the counter to your right.", + "B": "Put it in the drawer under your right.", + "C": "Place the peeler in the cabinet behind your right side, not the one behind your left.", + "D": "The peeler should be in the cabinet behind your left side.", + "E": "Set the peeler on the shelf directly in front of you instead of putting it into a cabinet." + }, + "answer_label": "D", + "answer_text": "The peeler should be in the cabinet behind your left side.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_7", + "correct_answer": "The peeler should be in the cabinet behind your left side.", + "original_options": { + "A": "Put it on the counter to your right.", + "B": "Put it in the drawer under your right.", + "C": "Place the peeler in the cabinet behind your right side, not the one behind your left.", + "D": "The peeler should be in the cabinet behind your left side.", + "E": "Set the peeler on the shelf directly in front of you instead of putting it into a cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 18 + ] + }, + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_5", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the stove?", + "answer": "The stove is to your right rear.", + "question_chinese": "根据我站的位置，灶台在我哪个方位？", + "answer_chinese": "灶台在你的右后方。", + "evidence": [ + { + "description": "The stove is to the person's right rear.", + "timestamps": { + "start": "00:03:16", + "end": "00:03:22" + } + } + ], + "question_timestamps": "00:03:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The stove is directly in front of you.", + "B": "The stove is to your left rear.", + "C": "The stove is to your right rear.", + "D": "The stove is on your right side, not behind you, so you would turn right to face it.", + "E": "The stove is behind you to the left, so you’d need to pivot around to see it." + }, + "answer_label": "C", + "answer_text": "The stove is to your right rear.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_5", + "correct_answer": "The stove is to your right rear.", + "original_options": { + "A": "The stove is directly in front of you.", + "B": "The stove is to your left rear.", + "C": "The stove is to your right rear.", + "D": "The stove is on your right side, not behind you, so you would turn right to face it.", + "E": "The stove is behind you to the left, so you’d need to pivot around to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 18 + ] + }, + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_6", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the stove?", + "answer": "The stove is on your left.", + "question_chinese": "根据我站的位置，灶台在我哪个方位？", + "answer_chinese": "灶台在你的左边", + "evidence": [ + { + "description": "The stove is to the left of this person.", + "timestamps": { + "start": "00:04:16", + "end": "00:04:22" + } + } + ], + "question_timestamps": "00:04:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The stove is on your right.", + "B": "The stove is straight ahead.", + "C": "The stove is behind you on the counter, so you’d need to turn around to reach it.", + "D": "The stove is to your right near the wall, positioned closer to the fridge than to you.", + "E": "The stove is on your left." + }, + "answer_label": "E", + "answer_text": "The stove is on your left.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_6", + "correct_answer": "The stove is on your left.", + "original_options": { + "A": "The stove is on your right.", + "B": "The stove is straight ahead.", + "C": "The stove is behind you on the counter, so you’d need to turn around to reach it.", + "D": "The stove is to your right near the wall, positioned closer to the fridge than to you.", + "E": "The stove is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 17 + ] + }, + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_4_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What's in the two pots now?", + "answer": "The pot on the left contains peppers and tomatoes, while the pot on the right has no ingredients added.", + "question_chinese": "现在两口锅里放了什么？", + "answer_chinese": "左边那口锅放了辣椒和西红柿，右边那口锅没有放入食材。", + "evidence": [ + { + "description": "Chopped and added a chili pepper in the pot on the left.", + "timestamps": { + "start": "00:05:18", + "end": "00:06:30" + } + }, + { + "description": "Placed the garlic in the pot on the right.", + "timestamps": { + "start": "00:09:36", + "end": "00:09:50" + } + }, + { + "description": "Placed four tomatoes in the pot on the left.", + "timestamps": { + "start": "00:06:40", + "end": "00:07:17" + } + }, + { + "description": "Placed the chicken breast in the pot on the right.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:51" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Both pots are still empty.", + "B": "Only the right pot has vegetables.", + "C": "The left pot has no ingredients added yet, while the right pot contains peppers and tomatoes.", + "D": "The pot on the left contains peppers and tomatoes, while the pot on the right has no ingredients added.", + "E": "The left pot contains peppers only, and the right pot contains tomatoes only, split between the two pots." + }, + "answer_label": "D", + "answer_text": "The pot on the left contains peppers and tomatoes, while the pot on the right has no ingredients added.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_4_T1", + "correct_answer": "The pot on the left contains peppers and tomatoes, while the pot on the right has no ingredients added.", + "original_options": { + "A": "Both pots are still empty.", + "B": "Only the right pot has vegetables.", + "C": "The left pot has no ingredients added yet, while the right pot contains peppers and tomatoes.", + "D": "The pot on the left contains peppers and tomatoes, while the pot on the right has no ingredients added.", + "E": "The left pot contains peppers only, and the right pot contains tomatoes only, split between the two pots." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 18 + ] + }, + { + "video_uid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5", + "video_length": "00:23:57", + "qid": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_4_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What's in the two pots now?", + "answer": "The pot on the left contains peppers and tomatoes, while the pot on the right has no ingredients added.", + "question_chinese": "现在两口锅里放了什么？", + "answer_chinese": "左边那口锅放了辣椒和西红柿，右边那口锅没有放入食材。", + "evidence": [ + { + "description": "Chopped and added a chili pepper in the pot on the left.", + "timestamps": { + "start": "00:05:18", + "end": "00:06:30" + } + }, + { + "description": "Placed the garlic in the pot on the right.", + "timestamps": { + "start": "00:09:36", + "end": "00:09:50" + } + }, + { + "description": "Placed four tomatoes in the pot on the left.", + "timestamps": { + "start": "00:06:40", + "end": "00:07:17" + } + }, + { + "description": "Placed the chicken breast in the pot on the right.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:51" + } + } + ], + "question_timestamps": "00:10:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Both pots have peppers and tomatoes.", + "B": "The pot on the left contains peppers and tomatoes, while the pot on the right has no ingredients added.", + "C": "Both pots have no ingredients added.", + "D": "The pot on the left has no ingredients, while the pot on the right contains peppers and tomatoes.", + "E": "The left pot contains only peppers, and the right pot contains only tomatoes, with nothing else added." + }, + "answer_label": "B", + "answer_text": "The pot on the left contains peppers and tomatoes, while the pot on the right has no ingredients added.", + "q_id": "7d2d4e8e-bc29-4cce-bbd5-ab1daf1b27b5_4_T2", + "correct_answer": "The pot on the left contains peppers and tomatoes, while the pot on the right has no ingredients added.", + "original_options": { + "A": "Both pots have peppers and tomatoes.", + "B": "The pot on the left contains peppers and tomatoes, while the pot on the right has no ingredients added.", + "C": "Both pots have no ingredients added.", + "D": "The pot on the left has no ingredients, while the pot on the right contains peppers and tomatoes.", + "E": "The left pot contains only peppers, and the right pot contains only tomatoes, with nothing else added." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + } +] diff --git a/QAs/7d6d6cae-0601-485d-b900-482b4254e9f8-mcq.json b/QAs/7d6d6cae-0601-485d-b900-482b4254e9f8-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..af29097feecf93beaaa3df519ae3ee975ac5b4f0 --- /dev/null +++ b/QAs/7d6d6cae-0601-485d-b900-482b4254e9f8-mcq.json @@ -0,0 +1,1449 @@ +[ + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I threw the foam tray into the trash just now, the broom was in front of me, left or right?", + "answer": "Just now, when you threw the foam tray into the trash can, the broom was in front of you.", + "question_chinese": "刚才我在将泡沫盘丢到垃圾桶的时候的时候，扫把在我的前面，左边还是右边？", + "answer_chinese": "刚才你在将泡沫盘丢到垃圾桶的时候的时候，扫把在你的前面。", + "evidence": [ + { + "description": "When I threw the foam tray into the trash can,the broom was in front of me.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:17:13", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Just now, when you threw the foam tray into the trash can, the broom was not in front of you.", + "B": "Just now, when you threw the foam tray into the trash can, the broom was in front of you." + }, + "answer_label": "B", + "answer_text": "Just now, when you threw the foam tray into the trash can, the broom was in front of you.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_2", + "correct_answer": "Just now, when you threw the foam tray into the trash can, the broom was in front of you.", + "original_options": { + "A": "Just now, when you threw the foam tray into the trash can, the broom was not in front of you.", + "B": "Just now, when you threw the foam tray into the trash can, the broom was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_14", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I threw the foam tray into the trash just now, the broom was in front of me, left or right?", + "answer": "Just now, when you threw the foam tray into the trash can, the broom was in front of you.", + "question_chinese": "刚才我在将泡沫盘丢到垃圾桶的时候的时候，扫把在我的前面，左边还是右边？", + "answer_chinese": "刚才你在将泡沫盘丢到垃圾桶的时候的时候，扫把在你的前面。", + "evidence": [ + { + "description": "When I threw the foam tray into the trash can,the broom was in front of me.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:21:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Just now, when you threw the foam tray into the trash can, the broom was in front of you.", + "B": "Just now, when you threw the foam tray into the trash can, the broom was not in front of you." + }, + "answer_label": "A", + "answer_text": "Just now, when you threw the foam tray into the trash can, the broom was in front of you.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_14", + "correct_answer": "Just now, when you threw the foam tray into the trash can, the broom was in front of you.", + "original_options": { + "A": "Just now, when you threw the foam tray into the trash can, the broom was in front of you.", + "B": "Just now, when you threw the foam tray into the trash can, the broom was not in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_8", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the white carpet of the toilet go through from the ground to the toilet seat?", + "answer": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times before being folded and placed on the toilet seat.", + "question_chinese": "厕所的白色地毯从地上到马桶盖上的经历了什么路径？", + "answer_chinese": "厕所的白色地毯从地上被拿到阳台上抖了几下，再被叠好放到马桶盖上。", + "evidence": [ + { + "description": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times.", + "timestamps": { + "start": "00:15:00", + "end": "00:15:10" + } + }, + { + "description": "The white carpet was folded and placed on the toilet seat", + "timestamps": { + "start": "00:15:10", + "end": "00:15:16" + } + } + ], + "question_timestamps": "00:15:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It went straight to the toilet seat.", + "B": "It was carried through the hallway.", + "C": "You picked it up, went to the balcony, shook it, and placed it on the toilet seat immediately.", + "D": "You carried it to the balcony, shook it a few times, then placed it on the toilet seat without folding.", + "E": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times before being folded and placed on the toilet seat." + }, + "answer_label": "E", + "answer_text": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times before being folded and placed on the toilet seat.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_8", + "correct_answer": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times before being folded and placed on the toilet seat.", + "original_options": { + "A": "It went straight to the toilet seat.", + "B": "It was carried through the hallway.", + "C": "You picked it up, went to the balcony, shook it, and placed it on the toilet seat immediately.", + "D": "You carried it to the balcony, shook it a few times, then placed it on the toilet seat without folding.", + "E": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times before being folded and placed on the toilet seat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 20 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_15", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the white carpet of the toilet go through from the ground to the toilet seat?", + "answer": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times before being folded and placed on the toilet seat.", + "question_chinese": "厕所的白色地毯从地上到马桶盖上的经历了什么路径？", + "answer_chinese": "厕所的白色地毯从地上被拿到阳台上抖了几下，再被叠好放到马桶盖上。", + "evidence": [ + { + "description": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times.", + "timestamps": { + "start": "00:15:00", + "end": "00:15:10" + } + }, + { + "description": "The white carpet was folded and placed on the toilet seat", + "timestamps": { + "start": "00:15:10", + "end": "00:15:16" + } + } + ], + "question_timestamps": "00:32:46", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was taken straight to the toilet.", + "B": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times before being folded and placed on the toilet seat.", + "C": "It went from the ground to the hallway.", + "D": "You picked it up, carried it to the balcony, shook it, and placed it on the toilet seat immediately.", + "E": "You lifted it from the ground, shook it indoors, and placed it on the toilet seat without going to the balcony." + }, + "answer_label": "B", + "answer_text": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times before being folded and placed on the toilet seat.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_15", + "correct_answer": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times before being folded and placed on the toilet seat.", + "original_options": { + "A": "It was taken straight to the toilet.", + "B": "The white carpet of the toilet was taken from the ground to the balcony and shook a few times before being folded and placed on the toilet seat.", + "C": "It went from the ground to the hallway.", + "D": "You picked it up, carried it to the balcony, shook it, and placed it on the toilet seat immediately.", + "E": "You lifted it from the ground, shook it indoors, and placed it on the toilet seat without going to the balcony." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 21 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How many remote controls were there on the table when I was cleaning it just now?", + "answer": "When you were cleaning the table just now, there were two remote controls on it.", + "question_chinese": "在我刚才擦桌子的时候，桌上有几个遥控器？", + "answer_chinese": "在你刚才擦桌子的时候，桌上有2个遥控器。", + "evidence": [ + { + "description": "There were two remote controls on the Table.", + "timestamps": { + "start": "00:00:38", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:9:17", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One remote control.", + "B": "Three remote controls.", + "C": "When you were cleaning the table just now, there was one remote control on it the whole time.", + "D": "When you were cleaning the table just now, there were three remote controls spread out on the table.", + "E": "When you were cleaning the table just now, there were two remote controls on it." + }, + "answer_label": "E", + "answer_text": "When you were cleaning the table just now, there were two remote controls on it.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_1", + "correct_answer": "When you were cleaning the table just now, there were two remote controls on it.", + "original_options": { + "A": "One remote control.", + "B": "Three remote controls.", + "C": "When you were cleaning the table just now, there was one remote control on it the whole time.", + "D": "When you were cleaning the table just now, there were three remote controls spread out on the table.", + "E": "When you were cleaning the table just now, there were two remote controls on it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 18 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_13", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How many remote controls were there on the table when I was cleaning it just now?", + "answer": "When you were cleaning the table just now, there were two remote controls on it.", + "question_chinese": "在我刚才擦桌子的时候，桌上有几个遥控器？", + "answer_chinese": "在你刚才擦桌子的时候，桌上有2个遥控器。", + "evidence": [ + { + "description": "There were two remote controls on the Table.", + "timestamps": { + "start": "00:00:38", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:27:55", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "When you were cleaning the table just now, there were two remote controls on it.", + "B": "One remote control.", + "C": "Four remote controls.", + "D": "When you were cleaning the table just now, there was one remote control sitting on the table.", + "E": "When you were cleaning the table just now, there were four remote controls lying on the table." + }, + "answer_label": "A", + "answer_text": "When you were cleaning the table just now, there were two remote controls on it.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_13", + "correct_answer": "When you were cleaning the table just now, there were two remote controls on it.", + "original_options": { + "A": "When you were cleaning the table just now, there were two remote controls on it.", + "B": "One remote control.", + "C": "Four remote controls.", + "D": "When you were cleaning the table just now, there was one remote control sitting on the table.", + "E": "When you were cleaning the table just now, there were four remote controls lying on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_1_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I wiped the table just now, how many pieces of trash on the table needed to be thrown away?", + "answer": "There were two pieces of trash on the table in front of you that needed to be thrown away.", + "question_chinese": "在我刚才擦桌子的时候，桌上有几个需要丢的垃圾？", + "answer_chinese": "你面前的桌上有两个需要丢的垃圾", + "evidence": [ + { + "description": "There are two pieces of paper to throw away on the table in front of you.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One piece of trash.", + "B": "Three pieces of trash.", + "C": "There were two pieces of trash on the table in front of you that needed to be thrown away.", + "D": "There were zero pieces of trash on the table in front of you that needed to be thrown away.", + "E": "There were four pieces of trash on the table in front of you that needed to be thrown away." + }, + "answer_label": "C", + "answer_text": "There were two pieces of trash on the table in front of you that needed to be thrown away.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_1_T1", + "correct_answer": "There were two pieces of trash on the table in front of you that needed to be thrown away.", + "original_options": { + "A": "One piece of trash.", + "B": "Three pieces of trash.", + "C": "There were two pieces of trash on the table in front of you that needed to be thrown away.", + "D": "There were zero pieces of trash on the table in front of you that needed to be thrown away.", + "E": "There were four pieces of trash on the table in front of you that needed to be thrown away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 19 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_1_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I wiped the table just now, how many pieces of trash on the table needed to be thrown away?", + "answer": "There was no trash on the table in front of you that needed to be thrown away.", + "question_chinese": "在我刚才擦桌子的时候，桌上有几个需要丢的垃圾？", + "answer_chinese": "你面前的桌上没有需要丢的垃圾", + "evidence": [ + { + "description": "You have no trash to throw away.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One piece of trash.", + "B": "Two pieces of trash.", + "C": "There was one piece of trash on the table in front of you that needed to be thrown away.", + "D": "There was no trash on the table in front of you that needed to be thrown away.", + "E": "There were three pieces of trash on the table in front of you that needed to be thrown away." + }, + "answer_label": "D", + "answer_text": "There was no trash on the table in front of you that needed to be thrown away.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_1_T2", + "correct_answer": "There was no trash on the table in front of you that needed to be thrown away.", + "original_options": { + "A": "One piece of trash.", + "B": "Two pieces of trash.", + "C": "There was one piece of trash on the table in front of you that needed to be thrown away.", + "D": "There was no trash on the table in front of you that needed to be thrown away.", + "E": "There were three pieces of trash on the table in front of you that needed to be thrown away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 19 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_7_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the doll back where it originally was, what should I do?", + "answer": "You should put it on the white chair with the green cushion, on the left armrest.", + "question_chinese": "如果我要把娃娃放回原位，我应该怎么做", + "answer_chinese": "你应该放到绿色靠垫的白色椅子上靠左边扶手上。", + "evidence": [ + { + "description": "The doll was originally leaning on the left armrest of a white chair with a green cushion.", + "timestamps": { + "start": "00:25:00", + "end": "00:25:02" + } + } + ], + "question_timestamps": "00:25:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You should put it on the white chair with the green cushion, on the left armrest.", + "B": "Put it on the right armrest.", + "C": "Put it on the seat cushion.", + "D": "Place the doll on the white chair with the green cushion, leaning it against the right armrest instead.", + "E": "Put the doll back on the white chair with the green cushion, but set it against the backrest, not the armrest." + }, + "answer_label": "A", + "answer_text": "You should put it on the white chair with the green cushion, on the left armrest.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_7_T1", + "correct_answer": "You should put it on the white chair with the green cushion, on the left armrest.", + "original_options": { + "A": "You should put it on the white chair with the green cushion, on the left armrest.", + "B": "Put it on the right armrest.", + "C": "Put it on the seat cushion.", + "D": "Place the doll on the white chair with the green cushion, leaning it against the right armrest instead.", + "E": "Put the doll back on the white chair with the green cushion, but set it against the backrest, not the armrest." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 21 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_7_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the doll back where it originally was, what should I do?", + "answer": "You should walk right toward the sofa and put the doll on the white chair with the green cushion, on the left armrest.", + "question_chinese": "如果我要把娃娃放回原位，我应该怎么做", + "answer_chinese": "你应该向右走向沙发，把娃娃放到绿色靠垫的白色椅子上靠左边扶手上。", + "evidence": [ + { + "description": "The doll was originally leaning on the left armrest of a white chair with a green cushion.", + "timestamps": { + "start": "00:25:00", + "end": "00:25:02" + } + }, + { + "description": "After storing some items, you walked to the left.", + "timestamps": { + "start": "00:25:25", + "end": "00:25:26" + } + } + ], + "question_timestamps": "00:25:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk left toward the sofa.", + "B": "Put it on the right armrest.", + "C": "You should walk right toward the sofa and put the doll on the white chair with the green cushion, on the left armrest.", + "D": "Walk right toward the sofa and place the doll on the white chair with the green cushion, on the right armrest.", + "E": "Walk left toward the sofa and put the doll on the white chair with the green cushion, on the left armrest." + }, + "answer_label": "C", + "answer_text": "You should walk right toward the sofa and put the doll on the white chair with the green cushion, on the left armrest.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_7_T2", + "correct_answer": "You should walk right toward the sofa and put the doll on the white chair with the green cushion, on the left armrest.", + "original_options": { + "A": "Walk left toward the sofa.", + "B": "Put it on the right armrest.", + "C": "You should walk right toward the sofa and put the doll on the white chair with the green cushion, on the left armrest.", + "D": "Walk right toward the sofa and place the doll on the white chair with the green cushion, on the right armrest.", + "E": "Walk left toward the sofa and put the doll on the white chair with the green cushion, on the left armrest." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 21, + 21 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_12", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the doll back in place, where should I put it?", + "answer": "You should place it on the white chair with a green cushion and lean it against the left armrest.", + "question_chinese": "如果我要把娃娃放回原位，我应该放到哪里？", + "answer_chinese": "你应该放到绿色靠垫的白色椅子上靠左边扶手上。", + "evidence": [ + { + "description": "The doll was originally leaning on the left armrest of a white chair with a green cushion.", + "timestamps": { + "start": "00:25:00", + "end": "00:25:02" + } + } + ], + "question_timestamps": "00:25:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Lean it against the right armrest.", + "B": "Put it on the seat cushion.", + "C": "Place it on the white chair with the green cushion, leaning it against the backrest instead of the armrest.", + "D": "You should place it on the white chair with a green cushion and lean it against the left armrest.", + "E": "Place the doll on the floor beside the white chair with the green cushion, not leaning on any armrest." + }, + "answer_label": "D", + "answer_text": "You should place it on the white chair with a green cushion and lean it against the left armrest.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_12", + "correct_answer": "You should place it on the white chair with a green cushion and lean it against the left armrest.", + "original_options": { + "A": "Lean it against the right armrest.", + "B": "Put it on the seat cushion.", + "C": "Place it on the white chair with the green cushion, leaning it against the backrest instead of the armrest.", + "D": "You should place it on the white chair with a green cushion and lean it against the left armrest.", + "E": "Place the doll on the floor beside the white chair with the green cushion, not leaning on any armrest." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_16", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the doll back in place, where should I put it?", + "answer": "You should place it on the white chair with a green cushion and lean it against the left armrest.", + "question_chinese": "如果我要把娃娃放回原位，我应该放到哪里？", + "answer_chinese": "你应该放到绿色靠垫的白色椅子上靠左边扶手上。", + "evidence": [ + { + "description": "The doll was originally leaning on the left armrest of a white chair with a green cushion.", + "timestamps": { + "start": "00:25:00", + "end": "00:25:02" + } + } + ], + "question_timestamps": "00:33:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Lean it against the right armrest.", + "B": "You should place it on the white chair with a green cushion and lean it against the left armrest.", + "C": "Lean it against the backrest.", + "D": "Place it on the white chair with a blue cushion and lean it against the left armrest.", + "E": "Put it on the white chair with the green cushion, but position it against the backrest instead of the left armrest." + }, + "answer_label": "B", + "answer_text": "You should place it on the white chair with a green cushion and lean it against the left armrest.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_16", + "correct_answer": "You should place it on the white chair with a green cushion and lean it against the left armrest.", + "original_options": { + "A": "Lean it against the right armrest.", + "B": "You should place it on the white chair with a green cushion and lean it against the left armrest.", + "C": "Lean it against the backrest.", + "D": "Place it on the white chair with a blue cushion and lean it against the left armrest.", + "E": "Put it on the white chair with the green cushion, but position it against the backrest instead of the left armrest." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 21 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_5_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I picked up the white carpet, what path did I take?", + "answer": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, and shook the rug.", + "question_chinese": "我将白色地毯拿起后，经过了什么路径", + "answer_chinese": "你在厕所地板上拿起了白色的地毯，转身向阳台走去，抖了抖地毯", + "evidence": [ + { + "description": "You pick up the white carpet on the bathroom floor.", + "timestamps": { + "start": "00:15:00", + "end": "00:15:01" + } + }, + { + "description": "Turn around and walk towards the balcony.", + "timestamps": { + "start": "00:15:02", + "end": "00:15:03" + } + }, + { + "description": "Shook the carpet.", + "timestamps": { + "start": "00:15:06", + "end": "00:15:07" + } + } + ], + "question_timestamps": "00:15:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, and shook the rug.", + "B": "You walked back into the bathroom.", + "C": "You went toward the balcony door.", + "D": "You picked up the white rug from the bathroom floor, turned toward the balcony, and placed it down without shaking it.", + "E": "You picked up the white rug, walked toward the balcony, shook it, and then set it on top of the blue rug." + }, + "answer_label": "A", + "answer_text": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, and shook the rug.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_5_T2", + "correct_answer": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, and shook the rug.", + "original_options": { + "A": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, and shook the rug.", + "B": "You walked back into the bathroom.", + "C": "You went toward the balcony door.", + "D": "You picked up the white rug from the bathroom floor, turned toward the balcony, and placed it down without shaking it.", + "E": "You picked up the white rug, walked toward the balcony, shook it, and then set it on top of the blue rug." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 22 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_5_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I picked up the white carpet, what path did I take?", + "answer": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, shook the rug, and then placed it on top of the blue rug.", + "question_chinese": "我将白色地毯拿起后，经过了什么路径", + "answer_chinese": "你在厕所地板上拿起了白色的地毯，转身向阳台走去，抖了抖地毯，然后把地毯放在了蓝色地毯上", + "evidence": [ + { + "description": "You pick up the white carpet on the bathroom floor.", + "timestamps": { + "start": "00:15:00", + "end": "00:15:01" + } + }, + { + "description": "Turn around and walk towards the balcony.", + "timestamps": { + "start": "00:15:02", + "end": "00:15:03" + } + }, + { + "description": "Shook the carpet.", + "timestamps": { + "start": "00:15:07", + "end": "00:15:08" + } + }, + { + "description": "You placed the carpet on the blue one.", + "timestamps": { + "start": "00:15:14", + "end": "00:15:15" + } + } + ], + "question_timestamps": "00:15:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You stopped after shaking it.", + "B": "You put it back on the bathroom floor.", + "C": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, shook the rug, and then placed it on top of the blue rug.", + "D": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, and shook the rug.", + "E": "You picked up the white rug, walked toward the balcony, and placed it on the blue rug without shaking it first." + }, + "answer_label": "C", + "answer_text": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, shook the rug, and then placed it on top of the blue rug.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_5_T1", + "correct_answer": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, shook the rug, and then placed it on top of the blue rug.", + "original_options": { + "A": "You stopped after shaking it.", + "B": "You put it back on the bathroom floor.", + "C": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, shook the rug, and then placed it on top of the blue rug.", + "D": "You picked up the white rug from the bathroom floor, turned and walked toward the balcony, and shook the rug.", + "E": "You picked up the white rug, walked toward the balcony, and placed it on the blue rug without shaking it first." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 20, + 21 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_2_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the broom in front of me, behind me, to my left, or to my right?", + "answer": "In front of you.", + "question_chinese": "扫把在我的前面，后面，左边还是右边？", + "answer_chinese": "在你的前面", + "evidence": [ + { + "description": "When I threw the foam tray into the trash can,the broom was in front of you", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The broom is positioned to your right, not in front of you at all right now.", + "D": "In front of you.", + "E": "The broom is positioned behind you rather than in front, near the area you just passed." + }, + "answer_label": "D", + "answer_text": "In front of you.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_2_T1", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The broom is positioned to your right, not in front of you at all right now.", + "D": "In front of you.", + "E": "The broom is positioned behind you rather than in front, near the area you just passed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 16, + 16 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_2_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the broom in front of me, behind me, to my left, or to my right?", + "answer": "The broom is behind you.", + "question_chinese": "扫把在我的前面，后面，左边还是右边？", + "answer_chinese": "扫把在你的后面。", + "evidence": [ + { + "description": "When I threw the foam tray into the trash can,the broom was in front of me. After throwing away the trash, you turn around and walk indoors.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:01:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The broom is to your right side, not behind you where you are standing now.", + "D": "The broom is positioned in front of you, not behind where you are standing.", + "E": "The broom is behind you." + }, + "answer_label": "E", + "answer_text": "The broom is behind you.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_2_T2", + "correct_answer": "The broom is behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The broom is to your right side, not behind you where you are standing now.", + "D": "The broom is positioned in front of you, not behind where you are standing.", + "E": "The broom is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 14 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_5", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I thrown away garbage so far?", + "answer": "So far, you have thrown away garbage once.", + "question_chinese": "目前为止，我丢了几次垃圾？", + "answer_chinese": "目前为止，你丢了一次垃圾。", + "evidence": [ + { + "description": "This is my first time throwing garbage.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero times.", + "B": "Twice.", + "C": "Up to now, you have thrown away garbage two times in total.", + "D": "So far, you have thrown away garbage once.", + "E": "Up to now, you have thrown away garbage three times in total." + }, + "answer_label": "D", + "answer_text": "So far, you have thrown away garbage once.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_5", + "correct_answer": "So far, you have thrown away garbage once.", + "original_options": { + "A": "Zero times.", + "B": "Twice.", + "C": "Up to now, you have thrown away garbage two times in total.", + "D": "So far, you have thrown away garbage once.", + "E": "Up to now, you have thrown away garbage three times in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 1, + 12, + 12 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_6", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I thrown away garbage so far?", + "answer": "So far, you have thrown away garbage twice.", + "question_chinese": "目前为止，我丢了几次垃圾？", + "answer_chinese": "目前为止，你丢了2次垃圾。", + "evidence": [ + { + "description": "This is my first time throwing garbage.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:12" + } + }, + { + "description": "This is my second time throwing garbage.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:26", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "So far, you have thrown away garbage twice.", + "C": "Three times.", + "D": "Up to now, you have thrown away garbage four times in total.", + "E": "Up to now, you have thrown away garbage five times in total." + }, + "answer_label": "B", + "answer_text": "So far, you have thrown away garbage twice.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_6", + "correct_answer": "So far, you have thrown away garbage twice.", + "original_options": { + "A": "Once.", + "B": "So far, you have thrown away garbage twice.", + "C": "Three times.", + "D": "Up to now, you have thrown away garbage four times in total.", + "E": "Up to now, you have thrown away garbage five times in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 12 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_7", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I thrown away garbage so far?", + "answer": "So far, you have thrown away garbage three times.", + "question_chinese": "目前为止，我丢了几次垃圾？", + "answer_chinese": "目前为止，你丢了3次垃圾。", + "evidence": [ + { + "description": "This is your third time throwing garbage.", + "timestamps": { + "start": "00:05:41", + "end": "00:05:47" + } + }, + { + "description": "This is your first time throwing garbage.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:12" + } + }, + { + "description": "This is your second time throwing garbage.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:05:50", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Four times.", + "C": "Up to now, you have thrown away garbage five times in total.", + "D": "So far, you have thrown away garbage three times.", + "E": "Up to now, you have thrown away garbage six times in total." + }, + "answer_label": "D", + "answer_text": "So far, you have thrown away garbage three times.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_7", + "correct_answer": "So far, you have thrown away garbage three times.", + "original_options": { + "A": "Twice.", + "B": "Four times.", + "C": "Up to now, you have thrown away garbage five times in total.", + "D": "So far, you have thrown away garbage three times.", + "E": "Up to now, you have thrown away garbage six times in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 12 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_3", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times have I washed the cloth so far?", + "answer": "So far, you have washed the cloth twice.", + "question_chinese": "目前为止，我洗了几次抹布？", + "answer_chinese": "目前为止，你洗了两次抹布。", + "evidence": [ + { + "description": "This is my second time washing the cloth.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:29" + } + }, + { + "description": "This is my first time washing the cloth.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:01:29", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "So far, you have washed the cloth twice.", + "D": "Up to now, you have washed the cloth four times in total.", + "E": "Up to now, you have washed the cloth six times in total." + }, + "answer_label": "C", + "answer_text": "So far, you have washed the cloth twice.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_3", + "correct_answer": "So far, you have washed the cloth twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "So far, you have washed the cloth twice.", + "D": "Up to now, you have washed the cloth four times in total.", + "E": "Up to now, you have washed the cloth six times in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 12 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_3_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many times have I washed the rag?", + "answer": "So far, you have washed the rag two times.", + "question_chinese": "目前为止，我洗了几次抹布？", + "answer_chinese": "目前为止，你洗了两次抹布。", + "evidence": [ + { + "description": "This is my second time washing the cloth.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:29" + } + }, + { + "description": "This is my first time washing the cloth.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:01:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Zero times.", + "B": "One time.", + "C": "Up to now, you have washed the rag three times in total.", + "D": "Up to now, you have washed the rag four times in total.", + "E": "So far, you have washed the rag two times." + }, + "answer_label": "E", + "answer_text": "So far, you have washed the rag two times.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_3_T1", + "correct_answer": "So far, you have washed the rag two times.", + "original_options": { + "A": "Zero times.", + "B": "One time.", + "C": "Up to now, you have washed the rag three times in total.", + "D": "Up to now, you have washed the rag four times in total.", + "E": "So far, you have washed the rag two times." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_3_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many times have I washed the rag?", + "answer": "So far, you have washed the rag three times.", + "question_chinese": "目前为止，我洗了几次抹布？", + "answer_chinese": "目前为止，你洗了三次抹布。", + "evidence": [ + { + "description": "This is my second time washing the cloth.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:29" + } + }, + { + "description": "This is my first time washing the cloth.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:36" + } + }, + { + "description": "This is my third time washing the cloth.", + "timestamps": { + "start": "00:07:53", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "So far, you have washed the rag three times.", + "B": "Two times.", + "C": "Four times.", + "D": "Up to now, you have washed the rag one time in total.", + "E": "Up to now, you have washed the rag five times in total." + }, + "answer_label": "A", + "answer_text": "So far, you have washed the rag three times.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_3_T2", + "correct_answer": "So far, you have washed the rag three times.", + "original_options": { + "A": "So far, you have washed the rag three times.", + "B": "Two times.", + "C": "Four times.", + "D": "Up to now, you have washed the rag one time in total.", + "E": "Up to now, you have washed the rag five times in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_11", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many things have I washed with a sponge?", + "answer": "So far, you have washed 2 things with a sponge.", + "question_chinese": "目前为止，我用海绵洗了几样东西？", + "answer_chinese": "目前为止，你用海绵洗了2样东西。", + "evidence": [ + { + "description": "I washed one thing with a sponge.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:15" + } + }, + { + "description": "I washed the second thing with a sponge.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:37" + } + } + ], + "question_timestamps": "00:06:39", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One thing.", + "B": "Three things.", + "C": "Up to now, you have washed 0 things with a sponge in total.", + "D": "So far, you have washed 2 things with a sponge.", + "E": "Up to now, you have washed 4 things with a sponge in total." + }, + "answer_label": "D", + "answer_text": "So far, you have washed 2 things with a sponge.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_11", + "correct_answer": "So far, you have washed 2 things with a sponge.", + "original_options": { + "A": "One thing.", + "B": "Three things.", + "C": "Up to now, you have washed 0 things with a sponge in total.", + "D": "So far, you have washed 2 things with a sponge.", + "E": "Up to now, you have washed 4 things with a sponge in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_10", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many things have I washed with a sponge?", + "answer": "So far, you have washed three things with a sponge.", + "question_chinese": "目前为止，我用海绵洗了几样东西？", + "answer_chinese": "目前为止，你用海绵洗了3样东西。", + "evidence": [ + { + "description": "This is the first thing I have washed with a sponge.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:15" + } + }, + { + "description": "This is the second thing I have washed with a sponge.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:37" + } + }, + { + "description": "This is the third thing I have washed with a sponge.", + "timestamps": { + "start": "00:06:37", + "end": "00:06:44" + } + } + ], + "question_timestamps": "00:06:48", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have washed two things.", + "B": "You have washed four things.", + "C": "So far, you have washed three things with a sponge.", + "D": "So far, you have washed five different things using a sponge during the cleaning.", + "E": "Up to now, you have washed six items with a sponge, counting each separate object once." + }, + "answer_label": "C", + "answer_text": "So far, you have washed three things with a sponge.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_10", + "correct_answer": "So far, you have washed three things with a sponge.", + "original_options": { + "A": "You have washed two things.", + "B": "You have washed four things.", + "C": "So far, you have washed three things with a sponge.", + "D": "So far, you have washed five different things using a sponge during the cleaning.", + "E": "Up to now, you have washed six items with a sponge, counting each separate object once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "7d6d6cae-0601-485d-b900-482b4254e9f8", + "video_length": "00:33:35", + "qid": "7d6d6cae-0601-485d-b900-482b4254e9f8_9", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many things have I washed with a sponge?", + "answer": "So far, you have washed 4 things with a sponge.", + "question_chinese": "目前为止，我用海绵洗了几样东西？", + "answer_chinese": "目前为止，你用海绵洗了4样东西。", + "evidence": [ + { + "description": "This is the first thing I have washed with a sponge.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:15" + } + }, + { + "description": "This is the second thing I have washed with a sponge.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:37" + } + }, + { + "description": "This is the fourth thing I have washed with a sponge.", + "timestamps": { + "start": "00:06:44", + "end": "00:06:52" + } + } + ], + "question_timestamps": "00:06:55", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "So far, you have washed 4 things with a sponge.", + "B": "You have washed three things.", + "C": "You have washed five things.", + "D": "So far, you have washed six different things with a sponge during the cleaning.", + "E": "Up to now, you have washed seven items with a sponge, counting each separate object once." + }, + "answer_label": "A", + "answer_text": "So far, you have washed 4 things with a sponge.", + "q_id": "7d6d6cae-0601-485d-b900-482b4254e9f8_9", + "correct_answer": "So far, you have washed 4 things with a sponge.", + "original_options": { + "A": "So far, you have washed 4 things with a sponge.", + "B": "You have washed three things.", + "C": "You have washed five things.", + "D": "So far, you have washed six different things with a sponge during the cleaning.", + "E": "Up to now, you have washed seven items with a sponge, counting each separate object once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + } +] diff --git a/QAs/7dc6494c-7d62-46f9-81d1-b28ada612764-mcq.json b/QAs/7dc6494c-7d62-46f9-81d1-b28ada612764-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3119bacbb671f0ed3fe97c7ba3df09e6636be68b --- /dev/null +++ b/QAs/7dc6494c-7d62-46f9-81d1-b28ada612764-mcq.json @@ -0,0 +1,948 @@ +[ + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_8_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I just throw the garbage I picked up?", + "answer": "You threw the garbage into the black garbage bag.", + "question_chinese": "我刚才将捡起来的垃圾扔进了哪里？", + "answer_chinese": "你将垃圾扔进了黑色垃圾袋里。", + "evidence": [ + { + "description": "You threw the garbage into the black garbage bag.", + "timestamps": { + "start": "00:07:14", + "end": "00:07:15" + } + }, + { + "description": "You are picking up garbage.", + "timestamps": { + "start": "00:07:09", + "end": "00:07:13" + } + } + ], + "question_timestamps": "00:016:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the gray garbage bag.", + "B": "Into the blue recycling bin.", + "C": "You threw the garbage into the yellow waste bin instead of placing it in a bag.", + "D": "You tossed the garbage into the red compost pile rather than into a garbage bag.", + "E": "You threw the garbage into the black garbage bag." + }, + "answer_label": "E", + "answer_text": "You threw the garbage into the black garbage bag.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_8_T1", + "correct_answer": "You threw the garbage into the black garbage bag.", + "original_options": { + "A": "Into the gray garbage bag.", + "B": "Into the blue recycling bin.", + "C": "You threw the garbage into the yellow waste bin instead of placing it in a bag.", + "D": "You tossed the garbage into the red compost pile rather than into a garbage bag.", + "E": "You threw the garbage into the black garbage bag." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_8_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I just throw the garbage I picked up?", + "answer": "You threw the garbage into the black garbage bag.", + "question_chinese": "我刚才将捡起来的垃圾扔进了哪里？", + "answer_chinese": "你将垃圾扔进了黑色垃圾袋里。", + "evidence": [ + { + "description": "You threw the garbage into the black garbage bag.", + "timestamps": { + "start": "00:07:14", + "end": "00:07:15" + } + }, + { + "description": "You are picking up garbage.", + "timestamps": { + "start": "00:07:09", + "end": "00:07:13" + } + } + ], + "question_timestamps": "00:16:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the gray garbage bag.", + "B": "Into the blue recycling bin.", + "C": "You threw the garbage into the yellow waste bin instead of placing it in a bag.", + "D": "You tossed the garbage into the red compost pile rather than into a garbage bag.", + "E": "You threw the garbage into the black garbage bag." + }, + "answer_label": "E", + "answer_text": "You threw the garbage into the black garbage bag.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_8_T2", + "correct_answer": "You threw the garbage into the black garbage bag.", + "original_options": { + "A": "Into the gray garbage bag.", + "B": "Into the blue recycling bin.", + "C": "You threw the garbage into the yellow waste bin instead of placing it in a bag.", + "D": "You tossed the garbage into the red compost pile rather than into a garbage bag.", + "E": "You threw the garbage into the black garbage bag." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_2_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What am I climbing up the triangular ladder for?", + "answer": "You climbed up the triangular ladder to install the light bulb.", + "question_chinese": "我爬上三角梯是为了干什么？", + "answer_chinese": "你爬上了三角梯安装灯泡。", + "evidence": [ + { + "description": "You climbed up the triangular ladder.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:16" + } + }, + { + "description": "You climbed up the triangular ladder to install the light bulb.", + "timestamps": { + "start": "00:00:19", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:04:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To fix the smoke detector.", + "B": "To adjust the ceiling fan.", + "C": "You climbed the triangular ladder to replace the ceiling light fixture, not just a bulb.", + "D": "You climbed up the triangular ladder to install the light bulb.", + "E": "You went up the triangular ladder to clean the ceiling air vents above you." + }, + "answer_label": "D", + "answer_text": "You climbed up the triangular ladder to install the light bulb.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_2_T1", + "correct_answer": "You climbed up the triangular ladder to install the light bulb.", + "original_options": { + "A": "To fix the smoke detector.", + "B": "To adjust the ceiling fan.", + "C": "You climbed the triangular ladder to replace the ceiling light fixture, not just a bulb.", + "D": "You climbed up the triangular ladder to install the light bulb.", + "E": "You went up the triangular ladder to clean the ceiling air vents above you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_2_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What am I climbing up the triangular ladder for?", + "answer": "You climbed up the triangular ladder to install the light bulb.", + "question_chinese": "我爬上三角梯是为了干什么？", + "answer_chinese": "你爬上了三角梯安装灯泡。", + "evidence": [ + { + "description": "You climbed up the triangular ladder.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:16" + } + }, + { + "description": "You climbed up the triangular ladder to install the light bulb.", + "timestamps": { + "start": "00:00:19", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:07:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To paint the left wall.", + "B": "You climbed up the triangular ladder to install the light bulb.", + "C": "To replace the ceiling tiles.", + "D": "You climbed the triangular ladder to adjust the ceiling fan, not to work on a light.", + "E": "You went up the triangular ladder to repair the broken window rather than installing anything overhead." + }, + "answer_label": "B", + "answer_text": "You climbed up the triangular ladder to install the light bulb.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_2_T2", + "correct_answer": "You climbed up the triangular ladder to install the light bulb.", + "original_options": { + "A": "To paint the left wall.", + "B": "You climbed up the triangular ladder to install the light bulb.", + "C": "To replace the ceiling tiles.", + "D": "You climbed the triangular ladder to adjust the ceiling fan, not to work on a light.", + "E": "You went up the triangular ladder to repair the broken window rather than installing anything overhead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_6_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the environment like around my house?", + "answer": "There are many trees and several detached buildings around your house.", + "question_chinese": "我的房子周围的环境是怎样的？", + "answer_chinese": "你的房子周围有很多树和几栋独立楼房。", + "evidence": [ + { + "description": "There are many trees and several detached buildings around your house.", + "timestamps": { + "start": "00:12:23", + "end": "00:12:24" + } + }, + { + "description": "There are many trees and several detached buildings around your house.", + "timestamps": { + "start": "00:12:31", + "end": "00:12:33" + } + } + ], + "question_timestamps": "00:23:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Few trees and big structures.", + "B": "Many trees and one building.", + "C": "There are many trees and several detached buildings around your house.", + "D": "There are many trees and several attached buildings around your house, not detached ones.", + "E": "There are many flowers and several detached buildings around your house instead of trees." + }, + "answer_label": "C", + "answer_text": "There are many trees and several detached buildings around your house.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_6_T1", + "correct_answer": "There are many trees and several detached buildings around your house.", + "original_options": { + "A": "Few trees and big structures.", + "B": "Many trees and one building.", + "C": "There are many trees and several detached buildings around your house.", + "D": "There are many trees and several attached buildings around your house, not detached ones.", + "E": "There are many flowers and several detached buildings around your house instead of trees." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_6_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the environment like around my house?", + "answer": "There are many trees and several detached buildings around your house.", + "question_chinese": "我的房子周围的环境是怎样的？", + "answer_chinese": "你的房子周围有很多树和几栋独立楼房。", + "evidence": [ + { + "description": "There are many trees and several detached buildings around your house.", + "timestamps": { + "start": "00:12:23", + "end": "00:12:24" + } + }, + { + "description": "There are many trees and several detached buildings around your house.", + "timestamps": { + "start": "00:12:31", + "end": "00:12:33" + } + } + ], + "question_timestamps": "00:31:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Many flowers and fences.", + "B": "Many rocks and ponds.", + "C": "There are many shrubs and several large sculptures behind your house, not detached buildings.", + "D": "There are many trees and several detached buildings around your house.", + "E": "There are few rocks and several tall fences to the right of your house instead of many trees." + }, + "answer_label": "D", + "answer_text": "There are many trees and several detached buildings around your house.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_6_T2", + "correct_answer": "There are many trees and several detached buildings around your house.", + "original_options": { + "A": "Many flowers and fences.", + "B": "Many rocks and ponds.", + "C": "There are many shrubs and several large sculptures behind your house, not detached buildings.", + "D": "There are many trees and several detached buildings around your house.", + "E": "There are few rocks and several tall fences to the right of your house instead of many trees." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 18 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_7_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put my broom?", + "answer": "Your broom is beside a pile of paper boxes.", + "question_chinese": "我的扫帚放在了哪里？", + "answer_chinese": "你的扫帚在一堆纸盒子的旁边。", + "evidence": [ + { + "description": "Your broom is beside a pile of paper boxes.", + "timestamps": { + "start": "00:15:17", + "end": "00:15:19" + } + } + ], + "question_timestamps": "00:27:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your broom is beside folded blankets.", + "B": "Your broom is beside a pile of paper boxes.", + "C": "Your broom is beside dry leaves.", + "D": "Your broom is leaning beside a large cardboard crate, not next to any stacked paper boxes.", + "E": "You left your broom beside a group of empty plastic containers rather than near a pile of boxes." + }, + "answer_label": "B", + "answer_text": "Your broom is beside a pile of paper boxes.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_7_T1", + "correct_answer": "Your broom is beside a pile of paper boxes.", + "original_options": { + "A": "Your broom is beside folded blankets.", + "B": "Your broom is beside a pile of paper boxes.", + "C": "Your broom is beside dry leaves.", + "D": "Your broom is leaning beside a large cardboard crate, not next to any stacked paper boxes.", + "E": "You left your broom beside a group of empty plastic containers rather than near a pile of boxes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_7_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put my broom?", + "answer": "Your broom is beside a pile of paper boxes.", + "question_chinese": "我的扫帚放在了哪里？", + "answer_chinese": "你的扫帚在一堆纸盒子的旁边。", + "evidence": [ + { + "description": "Your broom is beside a pile of paper boxes.", + "timestamps": { + "start": "00:15:17", + "end": "00:15:19" + } + } + ], + "question_timestamps": "00:27:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your broom is beside folded blankets.", + "B": "Your broom is beside a pile of paper boxes.", + "C": "Your broom is beside a cardboard crate.", + "D": "You placed your broom beside a bundle of dry leaves, away from any pile of paper boxes.", + "E": "Your broom is next to a group of empty plastic containers instead of beside stacked paper boxes." + }, + "answer_label": "B", + "answer_text": "Your broom is beside a pile of paper boxes.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_7_T2", + "correct_answer": "Your broom is beside a pile of paper boxes.", + "original_options": { + "A": "Your broom is beside folded blankets.", + "B": "Your broom is beside a pile of paper boxes.", + "C": "Your broom is beside a cardboard crate.", + "D": "You placed your broom beside a bundle of dry leaves, away from any pile of paper boxes.", + "E": "Your broom is next to a group of empty plastic containers instead of beside stacked paper boxes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 17 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_1_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many rattan chairs are there in my garden?", + "answer": "There are two rattan chairs in your garden.", + "question_chinese": "我的花园里有几条藤编椅子？", + "answer_chinese": "你的花园里有两条藤编椅子。", + "evidence": [ + { + "description": "There are two rattan chairs in your garden.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There is one rattan chair.", + "B": "There are three rattan chairs.", + "C": "There are four rattan chairs in your garden, arranged together in the seating area.", + "D": "There are two rattan chairs in your garden.", + "E": "There are five rattan chairs in your garden, with a couple grouped near the rest area." + }, + "answer_label": "D", + "answer_text": "There are two rattan chairs in your garden.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_1_T1", + "correct_answer": "There are two rattan chairs in your garden.", + "original_options": { + "A": "There is one rattan chair.", + "B": "There are three rattan chairs.", + "C": "There are four rattan chairs in your garden, arranged together in the seating area.", + "D": "There are two rattan chairs in your garden.", + "E": "There are five rattan chairs in your garden, with a couple grouped near the rest area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_1_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many rattan chairs are there in my garden?", + "answer": "There are two rattan chairs in your garden.", + "question_chinese": "我的花园里有几条藤编椅子？", + "answer_chinese": "你的花园里有两条藤编椅子。", + "evidence": [ + { + "description": "There are two rattan chairs in your garden.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:03:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There is one rattan chair.", + "B": "There are two rattan chairs in your garden.", + "C": "There are three rattan chairs.", + "D": "There are four rattan chairs in your garden, lined up along the edge of the yard.", + "E": "There are five rattan chairs in your garden, spread out rather than just two together." + }, + "answer_label": "B", + "answer_text": "There are two rattan chairs in your garden.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_1_T2", + "correct_answer": "There are two rattan chairs in your garden.", + "original_options": { + "A": "There is one rattan chair.", + "B": "There are two rattan chairs in your garden.", + "C": "There are three rattan chairs.", + "D": "There are four rattan chairs in your garden, lined up along the edge of the yard.", + "E": "There are five rattan chairs in your garden, spread out rather than just two together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_3_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What are the main components of my yard?", + "answer": "Your yard is mainly composed of three parts: the work area, the rest area and the swimming pool.", + "question_chinese": "我的院子里里主要由哪几个部分组成？", + "answer_chinese": "你的院子里主要由工作区，休息区和泳池三个部分组成。", + "evidence": [ + { + "description": "The rest area in your yard.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:28" + } + }, + { + "description": "The work area in your yard.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:42" + } + }, + { + "description": "The swimming pool in your yard.", + "timestamps": { + "start": "00:04:21", + "end": "00:04:23" + } + } + ], + "question_timestamps": "00:07:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Work area, rest area, and garden shed.", + "B": "Driveway, flower bed, and picnic area.", + "C": "Your yard is mainly composed of three parts: the vegetable patch, the seating area, and the fire pit.", + "D": "Your yard is mainly composed of three parts: the work area, the rest area and the swimming pool.", + "E": "Your yard is mainly composed of three parts: the garage, the fence line, and the storage shed." + }, + "answer_label": "D", + "answer_text": "Your yard is mainly composed of three parts: the work area, the rest area and the swimming pool.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_3_T1", + "correct_answer": "Your yard is mainly composed of three parts: the work area, the rest area and the swimming pool.", + "original_options": { + "A": "Work area, rest area, and garden shed.", + "B": "Driveway, flower bed, and picnic area.", + "C": "Your yard is mainly composed of three parts: the vegetable patch, the seating area, and the fire pit.", + "D": "Your yard is mainly composed of three parts: the work area, the rest area and the swimming pool.", + "E": "Your yard is mainly composed of three parts: the garage, the fence line, and the storage shed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_3_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What are the main components of my yard?", + "answer": "Your yard is mainly composed of three parts: the work area, the rest area and the swimming pool.", + "question_chinese": "我的院子里里主要由哪几个部分组成？", + "answer_chinese": "你的院子里主要由工作区，休息区和泳池三个部分组成。", + "evidence": [ + { + "description": "The rest area in your yard.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:28" + } + }, + { + "description": "The work area in your yard.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:42" + } + }, + { + "description": "The swimming pool in your yard.", + "timestamps": { + "start": "00:04:21", + "end": "00:04:23" + } + } + ], + "question_timestamps": "00:09:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your yard is mainly composed of three parts: the work area, the rest area and the swimming pool.", + "B": "Garden zone, lounge area, and fountain.", + "C": "Play corner, seating area, and pond.", + "D": "Your yard is mainly composed of three parts: the vegetable patch, the patio and the fire pit.", + "E": "Your yard is mainly composed of three parts: the driveway, the relaxation spot and the barbecue area." + }, + "answer_label": "A", + "answer_text": "Your yard is mainly composed of three parts: the work area, the rest area and the swimming pool.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_3_T2", + "correct_answer": "Your yard is mainly composed of three parts: the work area, the rest area and the swimming pool.", + "original_options": { + "A": "Your yard is mainly composed of three parts: the work area, the rest area and the swimming pool.", + "B": "Garden zone, lounge area, and fountain.", + "C": "Play corner, seating area, and pond.", + "D": "Your yard is mainly composed of three parts: the vegetable patch, the patio and the fire pit.", + "E": "Your yard is mainly composed of three parts: the driveway, the relaxation spot and the barbecue area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_5_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tripods are there in the yard in total?", + "answer": "Three tripods were placed in total.", + "question_chinese": "院子里一共放了几个三脚架？", + "answer_chinese": "一共放了三个三脚架。", + "evidence": [ + { + "description": "The first tripod.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "The second tripod.", + "timestamps": { + "start": "00:12:22", + "end": "00:12:24" + } + }, + { + "description": "The third tripod.", + "timestamps": { + "start": "00:14:52", + "end": "00:14:54" + } + } + ], + "question_timestamps": "00:22:34", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two tripods were placed in total.", + "B": "Four tripods were placed in total.", + "C": "Five tripods were placed in total, set up across the yard in different spots.", + "D": "Six tripods were placed in total, with multiple stands spread around the yard.", + "E": "Three tripods were placed in total." + }, + "answer_label": "E", + "answer_text": "Three tripods were placed in total.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_5_T1", + "correct_answer": "Three tripods were placed in total.", + "original_options": { + "A": "Two tripods were placed in total.", + "B": "Four tripods were placed in total.", + "C": "Five tripods were placed in total, set up across the yard in different spots.", + "D": "Six tripods were placed in total, with multiple stands spread around the yard.", + "E": "Three tripods were placed in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_5_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tripods are there in the yard in total?", + "answer": "Three tripods were placed in total.", + "question_chinese": "院子里一共放了几个三脚架？", + "answer_chinese": "一共放了三个三脚架。", + "evidence": [ + { + "description": "The first tripod.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "The second tripod.", + "timestamps": { + "start": "00:12:22", + "end": "00:12:24" + } + }, + { + "description": "The third tripod.", + "timestamps": { + "start": "00:14:52", + "end": "00:14:54" + } + } + ], + "question_timestamps": "00:27:34", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four tripods were placed in total.", + "B": "Five tripods were placed in total.", + "C": "Three tripods were placed in total.", + "D": "Six tripods were placed in total, positioned around the yard rather than just three.", + "E": "Seven tripods were placed in total, with several stands set up across the yard." + }, + "answer_label": "C", + "answer_text": "Three tripods were placed in total.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_5_T2", + "correct_answer": "Three tripods were placed in total.", + "original_options": { + "A": "Four tripods were placed in total.", + "B": "Five tripods were placed in total.", + "C": "Three tripods were placed in total.", + "D": "Six tripods were placed in total, positioned around the yard rather than just three.", + "E": "Seven tripods were placed in total, with several stands set up across the yard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_4_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many large black bags are placed against the wall?", + "answer": "Two large black bags were placed.", + "question_chinese": "靠墙的地方放了几个黑色大袋子？", + "answer_chinese": "放了两个黑色大袋子。", + "evidence": [ + { + "description": "Two large black bags were placed.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:04" + } + } + ], + "question_timestamps": "00:11:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One large black bag.", + "B": "Three large black bags.", + "C": "Four large black bags were leaning against the wall near the floor.", + "D": "Five large black bags were placed together against the wall in a single row.", + "E": "Two large black bags were placed." + }, + "answer_label": "E", + "answer_text": "Two large black bags were placed.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_4_T1", + "correct_answer": "Two large black bags were placed.", + "original_options": { + "A": "One large black bag.", + "B": "Three large black bags.", + "C": "Four large black bags were leaning against the wall near the floor.", + "D": "Five large black bags were placed together against the wall in a single row.", + "E": "Two large black bags were placed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + }, + { + "video_uid": "7dc6494c-7d62-46f9-81d1-b28ada612764", + "video_length": "00:40:48", + "qid": "7dc6494c-7d62-46f9-81d1-b28ada612764_4_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many large black bags are placed against the wall?", + "answer": "Two large black bags were placed.", + "question_chinese": "靠墙的地方放了几个黑色大袋子？", + "answer_chinese": "放了两个黑色大袋子。", + "evidence": [ + { + "description": "Two large black bags were placed.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:04" + } + } + ], + "question_timestamps": "00:17:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One large black bag.", + "B": "Two large black bags were placed.", + "C": "Three large black bags.", + "D": "Four large black bags were leaning against the wall near the floor.", + "E": "Five large black bags were placed together against the wall in a single row." + }, + "answer_label": "B", + "answer_text": "Two large black bags were placed.", + "q_id": "7dc6494c-7d62-46f9-81d1-b28ada612764_4_T2", + "correct_answer": "Two large black bags were placed.", + "original_options": { + "A": "One large black bag.", + "B": "Two large black bags were placed.", + "C": "Three large black bags.", + "D": "Four large black bags were leaning against the wall near the floor.", + "E": "Five large black bags were placed together against the wall in a single row." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + } +] diff --git a/QAs/7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b-mcq.json b/QAs/7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..34ff0b728a1999bb7dd40573ee29d0d4f845f684 --- /dev/null +++ b/QAs/7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b-mcq.json @@ -0,0 +1,1040 @@ +[ + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What vegetables on the table did you take to wash?", + "answer": "You took the loofah on the table to wash.", + "question_chinese": "你把桌子上的什么蔬菜拿去清洗了？", + "answer_chinese": "你把桌子上的丝瓜拿去清洗了。", + "evidence": [ + { + "description": "Loofah on the table.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + }, + { + "description": "You took the loofah on the table to wash.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:02:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took the cucumbers to wash.", + "B": "You took the carrots to wash.", + "C": "You took the radishes from the table and washed them in the sink.", + "D": "You picked up the peppers on the table and took them to be washed.", + "E": "You took the loofah on the table to wash." + }, + "answer_label": "E", + "answer_text": "You took the loofah on the table to wash.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_3", + "correct_answer": "You took the loofah on the table to wash.", + "original_options": { + "A": "You took the cucumbers to wash.", + "B": "You took the carrots to wash.", + "C": "You took the radishes from the table and washed them in the sink.", + "D": "You picked up the peppers on the table and took them to be washed.", + "E": "You took the loofah on the table to wash." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 14 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What vegetables on the table did you take to wash?", + "answer": "You took the loofah on the table to wash.", + "question_chinese": "你把桌子上的什么蔬菜拿去清洗了？", + "answer_chinese": "你把桌子上的丝瓜拿去清洗了。", + "evidence": [ + { + "description": "Loofah on the table.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + }, + { + "description": "You took the loofah on the table to wash.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:12:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took the onions to wash.", + "B": "You took the zucchini to wash.", + "C": "You took the loofah on the table to wash.", + "D": "You grabbed the carrots from the table and took them to the sink to wash.", + "E": "You took the peppers from the table and washed them before cooking." + }, + "answer_label": "C", + "answer_text": "You took the loofah on the table to wash.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_4", + "correct_answer": "You took the loofah on the table to wash.", + "original_options": { + "A": "You took the onions to wash.", + "B": "You took the zucchini to wash.", + "C": "You took the loofah on the table to wash.", + "D": "You grabbed the carrots from the table and took them to the sink to wash.", + "E": "You took the peppers from the table and washed them before cooking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 12 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_13", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What preparations did I make for the loofah before I was ready to cook it?", + "answer": "You washed the loofah, peeled it and removed the seeds, and finally cut it into pieces.", + "question_chinese": "在我准备烹饪丝瓜之前，我对丝瓜做了什么准备工作？", + "answer_chinese": "你清洗了丝瓜，将丝瓜剥皮并去瓤，最后将丝瓜切成了块。", + "evidence": [ + { + "description": "You washed the loofah.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:27" + } + }, + { + "description": "You peeled it and removed the seeds.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:57" + } + }, + { + "description": "You cut the loofah into pieces.", + "timestamps": { + "start": "00:02:56", + "end": "00:05:42" + } + } + ], + "question_timestamps": "00:08:56", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You only washed it first.", + "B": "You peeled it and sliced it.", + "C": "You rinsed the loofah, left the seeds in, and then cut it into thick pieces.", + "D": "You washed the loofah, kept the peel on, and chopped it up without removing the seeds.", + "E": "You washed the loofah, peeled it and removed the seeds, and finally cut it into pieces." + }, + "answer_label": "E", + "answer_text": "You washed the loofah, peeled it and removed the seeds, and finally cut it into pieces.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_13", + "correct_answer": "You washed the loofah, peeled it and removed the seeds, and finally cut it into pieces.", + "original_options": { + "A": "You only washed it first.", + "B": "You peeled it and sliced it.", + "C": "You rinsed the loofah, left the seeds in, and then cut it into thick pieces.", + "D": "You washed the loofah, kept the peel on, and chopped it up without removing the seeds.", + "E": "You washed the loofah, peeled it and removed the seeds, and finally cut it into pieces." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_14", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What preparations did I make for the loofah before I was ready to cook it?", + "answer": "You washed the loofah, peeled it and removed the seeds, and finally cut it into pieces.", + "question_chinese": "在我准备烹饪丝瓜之前，我对丝瓜做了什么准备工作？", + "answer_chinese": "你清洗了丝瓜，将丝瓜剥皮并去瓤，最后将丝瓜切成了块。", + "evidence": [ + { + "description": "You washed the loofah.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:27" + } + }, + { + "description": "You peeled it and removed the seeds.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:57" + } + }, + { + "description": "You cut the loofah into pieces.", + "timestamps": { + "start": "00:02:56", + "end": "00:05:42" + } + } + ], + "question_timestamps": "00:18:56", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You only rinsed it.", + "B": "You left the skin on.", + "C": "You washed the loofah, peeled it and removed the seeds, and finally cut it into pieces.", + "D": "You washed the loofah, cut it into rings, and kept the seeds inside.", + "E": "You peeled the loofah and sliced it up, but you didn’t remove the seeds first." + }, + "answer_label": "C", + "answer_text": "You washed the loofah, peeled it and removed the seeds, and finally cut it into pieces.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_14", + "correct_answer": "You washed the loofah, peeled it and removed the seeds, and finally cut it into pieces.", + "original_options": { + "A": "You only rinsed it.", + "B": "You left the skin on.", + "C": "You washed the loofah, peeled it and removed the seeds, and finally cut it into pieces.", + "D": "You washed the loofah, cut it into rings, and kept the seeds inside.", + "E": "You peeled the loofah and sliced it up, but you didn’t remove the seeds first." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 16 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_9", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the iron plate when I was cooking?", + "answer": "You took the iron plate from the shelf on the wall to the table.", + "question_chinese": "我做饭的时候把铁盘子从哪里拿到了哪里？", + "answer_chinese": "你把铁盘子从墙上的架子上拿到了桌子上。", + "evidence": [ + { + "description": "The iron plate is on the shelf on the wall.", + "timestamps": { + "start": "00:08:24", + "end": "00:08:26" + } + }, + { + "description": "You put the iron plate on the table.", + "timestamps": { + "start": "00:08:29", + "end": "00:08:32" + } + } + ], + "question_timestamps": "00:14:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You got it from the counter.", + "B": "You got it from a drawer.", + "C": "You took the iron plate from the shelf on the wall to the table.", + "D": "You took the iron plate from the cabinet above the counter and moved it to the stove.", + "E": "You grabbed the iron plate from the fridge area and carried it over to the countertop." + }, + "answer_label": "C", + "answer_text": "You took the iron plate from the shelf on the wall to the table.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_9", + "correct_answer": "You took the iron plate from the shelf on the wall to the table.", + "original_options": { + "A": "You got it from the counter.", + "B": "You got it from a drawer.", + "C": "You took the iron plate from the shelf on the wall to the table.", + "D": "You took the iron plate from the cabinet above the counter and moved it to the stove.", + "E": "You grabbed the iron plate from the fridge area and carried it over to the countertop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_10", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the iron plate when I was cooking?", + "answer": "You took the iron plate from the shelf on the wall to the table.", + "question_chinese": "我做饭的时候把铁盘子从哪里拿到了哪里？", + "answer_chinese": "你把铁盘子从墙上的架子上拿到了桌子上。", + "evidence": [ + { + "description": "The iron plate is on the shelf on the wall.", + "timestamps": { + "start": "00:08:24", + "end": "00:08:26" + } + }, + { + "description": "You put the iron plate on the table.", + "timestamps": { + "start": "00:08:29", + "end": "00:08:32" + } + } + ], + "question_timestamps": "00:23:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You got it from the stove.", + "B": "You took the iron plate from the shelf on the wall to the table.", + "C": "You got it from the cabinet.", + "D": "You pulled the iron plate from the drawer under the counter and placed it onto the countertop.", + "E": "You retrieved the iron plate from the counter beside the microwave and carried it over to the sink." + }, + "answer_label": "B", + "answer_text": "You took the iron plate from the shelf on the wall to the table.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_10", + "correct_answer": "You took the iron plate from the shelf on the wall to the table.", + "original_options": { + "A": "You got it from the stove.", + "B": "You took the iron plate from the shelf on the wall to the table.", + "C": "You got it from the cabinet.", + "D": "You pulled the iron plate from the drawer under the counter and placed it onto the countertop.", + "E": "You retrieved the iron plate from the counter beside the microwave and carried it over to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What powder did I scrape on the table just now?", + "answer": "You are scraping coconut powder.", + "question_chinese": "我刚才在桌子上刮什么粉末？", + "answer_chinese": "你在刮椰子粉末。", + "evidence": [ + { + "description": "You are scraping coconut powder.", + "timestamps": { + "start": "00:13:12", + "end": "00:13:14" + } + } + ], + "question_timestamps": "00:16:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are scraping chocolate powder.", + "B": "You are scraping almond powder.", + "C": "You are scraping cinnamon powder across the tabletop, not coconut powder at all.", + "D": "You are scraping coconut powder.", + "E": "You are scraping flour powder on the table, rather than coconut powder just now." + }, + "answer_label": "D", + "answer_text": "You are scraping coconut powder.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_11", + "correct_answer": "You are scraping coconut powder.", + "original_options": { + "A": "You are scraping chocolate powder.", + "B": "You are scraping almond powder.", + "C": "You are scraping cinnamon powder across the tabletop, not coconut powder at all.", + "D": "You are scraping coconut powder.", + "E": "You are scraping flour powder on the table, rather than coconut powder just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_12", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What powder did I scrape on the table just now?", + "answer": "You are scraping coconut powder.", + "question_chinese": "我刚才在桌子上刮什么粉末？", + "answer_chinese": "你在刮椰子粉末。", + "evidence": [ + { + "description": "You are scraping coconut powder.", + "timestamps": { + "start": "00:13:12", + "end": "00:13:14" + } + } + ], + "question_timestamps": "00:22:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are scraping coconut powder.", + "B": "You are scraping chocolate powder.", + "C": "You are scraping flour powder.", + "D": "You are scraping cinnamon powder on the table, not coconut powder this time.", + "E": "You are scraping garlic powder onto the table instead of coconut powder just now." + }, + "answer_label": "A", + "answer_text": "You are scraping coconut powder.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_12", + "correct_answer": "You are scraping coconut powder.", + "original_options": { + "A": "You are scraping coconut powder.", + "B": "You are scraping chocolate powder.", + "C": "You are scraping flour powder.", + "D": "You are scraping cinnamon powder on the table, not coconut powder this time.", + "E": "You are scraping garlic powder onto the table instead of coconut powder just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_15", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stove holes are there on the stove in my kitchen?", + "answer": "There are four stove holes.", + "question_chinese": "我厨房的灶台有几个灶孔？", + "answer_chinese": "有四个灶孔。", + "evidence": [ + { + "description": "There are four stove holes.", + "timestamps": { + "start": "00:06:53", + "end": "00:06:55" + } + } + ], + "question_timestamps": "00:10:22", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two stove holes.", + "B": "There are four stove holes.", + "C": "There are five stove holes.", + "D": "There are six stove holes on the stove in your kitchen right now.", + "E": "There are seven stove holes on the stove in your kitchen right now." + }, + "answer_label": "B", + "answer_text": "There are four stove holes.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_15", + "correct_answer": "There are four stove holes.", + "original_options": { + "A": "There are two stove holes.", + "B": "There are four stove holes.", + "C": "There are five stove holes.", + "D": "There are six stove holes on the stove in your kitchen right now.", + "E": "There are seven stove holes on the stove in your kitchen right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_16", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stove holes are there on the stove in my kitchen?", + "answer": "There are four stove holes.", + "question_chinese": "我厨房的灶台有几个灶孔？", + "answer_chinese": "有四个灶孔。", + "evidence": [ + { + "description": "There are four stove holes.", + "timestamps": { + "start": "00:06:53", + "end": "00:06:55" + } + } + ], + "question_timestamps": "00:14:22", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two stove holes.", + "B": "There are six stove holes.", + "C": "There are four stove holes.", + "D": "There are eight stove holes on the stove in your kitchen right now.", + "E": "There are ten stove holes on the stove in your kitchen right now." + }, + "answer_label": "C", + "answer_text": "There are four stove holes.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_16", + "correct_answer": "There are four stove holes.", + "original_options": { + "A": "There are two stove holes.", + "B": "There are six stove holes.", + "C": "There are four stove holes.", + "D": "There are eight stove holes on the stove in your kitchen right now.", + "E": "There are ten stove holes on the stove in your kitchen right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I was sitting on a bench in the room. What was my route to the sink?", + "answer": "You go straight out of the room first. The sink is on your right.", + "question_chinese": "我坐在屋子里的板凳上，我去到洗水池的路线是什么？", + "answer_chinese": "你先直走出房门，洗水槽在你的右边。", + "evidence": [ + { + "description": "You go straight out of the room first.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "The sink is on your right.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:07:14", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Exit the room, sink on your left.", + "B": "You go straight out of the room first. The sink is on your right.", + "C": "Turn around first, sink ahead.", + "D": "You walk out of the room, then the sink is directly in front of you.", + "E": "You leave the room first, then the sink is behind you instead of on your right." + }, + "answer_label": "B", + "answer_text": "You go straight out of the room first. The sink is on your right.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_1", + "correct_answer": "You go straight out of the room first. The sink is on your right.", + "original_options": { + "A": "Exit the room, sink on your left.", + "B": "You go straight out of the room first. The sink is on your right.", + "C": "Turn around first, sink ahead.", + "D": "You walk out of the room, then the sink is directly in front of you.", + "E": "You leave the room first, then the sink is behind you instead of on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 16 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I was sitting on a bench in the room. What was my route to the sink?", + "answer": "You go straight out of the room first. The sink is on your right.", + "question_chinese": "我坐在屋子里的板凳上，我去到洗水池的路线是什么？", + "answer_chinese": "你先直走出房门，洗水槽在你的右边。", + "evidence": [ + { + "description": "You go straight out of the room first.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "The sink is on your right.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:17:14", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left out; sink ahead.", + "B": "Go out, sink on your left.", + "C": "You turn left as you exit the room, and the sink ends up in front of you.", + "D": "You go straight out of the room first. The sink is on your right.", + "E": "You step out of the room first, then the sink is behind you rather than on your right." + }, + "answer_label": "D", + "answer_text": "You go straight out of the room first. The sink is on your right.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_2", + "correct_answer": "You go straight out of the room first. The sink is on your right.", + "original_options": { + "A": "Turn left out; sink ahead.", + "B": "Go out, sink on your left.", + "C": "You turn left as you exit the room, and the sink ends up in front of you.", + "D": "You go straight out of the room first. The sink is on your right.", + "E": "You step out of the room first, then the sink is behind you rather than on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 18 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_17", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to use chicken essence. Where can I find it?", + "answer": "You can find it on the cabinet door of the refrigerator.", + "question_chinese": "我想用鸡精，我该从哪里去找？", + "answer_chinese": "在冰箱的柜门上你可以找到。", + "evidence": [ + { + "description": "The refrigerator.", + "timestamps": { + "start": "00:07:07", + "end": "00:07:08" + } + }, + { + "description": "You can find chicken essence on the cabinet door of the refrigerator.", + "timestamps": { + "start": "00:07:08", + "end": "00:07:12" + } + } + ], + "question_timestamps": "00:13:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In the drawer under the sink.", + "B": "On the countertop by the stove.", + "C": "You can find it on the cabinet door of the refrigerator.", + "D": "You can find it in the pantry next to the spices, not on the refrigerator door.", + "E": "You can find it on the middle shelf of the cupboard instead of the cabinet door of the refrigerator." + }, + "answer_label": "C", + "answer_text": "You can find it on the cabinet door of the refrigerator.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_17", + "correct_answer": "You can find it on the cabinet door of the refrigerator.", + "original_options": { + "A": "In the drawer under the sink.", + "B": "On the countertop by the stove.", + "C": "You can find it on the cabinet door of the refrigerator.", + "D": "You can find it in the pantry next to the spices, not on the refrigerator door.", + "E": "You can find it on the middle shelf of the cupboard instead of the cabinet door of the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 19 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_18", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to use chicken essence. Where can I find it?", + "answer": "You can find it on the cabinet door of the refrigerator.", + "question_chinese": "我想用鸡精，我该从哪里去找？", + "answer_chinese": "在冰箱的柜门上你可以找到。", + "evidence": [ + { + "description": "The refrigerator.", + "timestamps": { + "start": "00:07:07", + "end": "00:07:08" + } + }, + { + "description": "You can find chicken essence on the cabinet door of the refrigerator.", + "timestamps": { + "start": "00:07:08", + "end": "00:07:12" + } + } + ], + "question_timestamps": "00:32:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You can find it on the cabinet door of the refrigerator.", + "B": "In the drawer under the sink.", + "C": "In the cupboard above the microwave.", + "D": "You can find it inside the basket on the kitchen counter, not on the refrigerator door.", + "E": "You can find it on the dining table next to the salad instead of the cabinet door of the refrigerator." + }, + "answer_label": "A", + "answer_text": "You can find it on the cabinet door of the refrigerator.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_18", + "correct_answer": "You can find it on the cabinet door of the refrigerator.", + "original_options": { + "A": "You can find it on the cabinet door of the refrigerator.", + "B": "In the drawer under the sink.", + "C": "In the cupboard above the microwave.", + "D": "You can find it inside the basket on the kitchen counter, not on the refrigerator door.", + "E": "You can find it on the dining table next to the salad instead of the cabinet door of the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 20 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_5", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the red gas cylinder placed?", + "answer": "The red gas pipe is placed beside the staircase.", + "question_chinese": "红色燃气瓶放在了哪里？", + "answer_chinese": "红色燃气管放在楼梯旁边。", + "evidence": [ + { + "description": "The red gas pipe is placed beside the staircase.", + "timestamps": { + "start": "00:07:50", + "end": "00:07:52" + } + } + ], + "question_timestamps": "00:09:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The red gas pipe is placed beside the staircase.", + "B": "Across from the staircase.", + "C": "At the bottom of the staircase.", + "D": "The red gas cylinder is positioned behind the staircase, partially tucked out of the walking path.", + "E": "The red gas cylinder sits near the top of the staircase, close to the upper landing area." + }, + "answer_label": "A", + "answer_text": "The red gas pipe is placed beside the staircase.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_5", + "correct_answer": "The red gas pipe is placed beside the staircase.", + "original_options": { + "A": "The red gas pipe is placed beside the staircase.", + "B": "Across from the staircase.", + "C": "At the bottom of the staircase.", + "D": "The red gas cylinder is positioned behind the staircase, partially tucked out of the walking path.", + "E": "The red gas cylinder sits near the top of the staircase, close to the upper landing area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 16, + 17 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_6", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the red gas cylinder placed?", + "answer": "The red gas pipe is placed beside the staircase.", + "question_chinese": "红色燃气瓶放在了哪里？", + "answer_chinese": "红色燃气管放在楼梯旁边。", + "evidence": [ + { + "description": "The red gas pipe is placed beside the staircase.", + "timestamps": { + "start": "00:07:50", + "end": "00:07:52" + } + } + ], + "question_timestamps": "00:13:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Under the dining table.", + "B": "Near the front door.", + "C": "The red gas cylinder is placed beside the kitchen island, close to the edge facing you.", + "D": "The red gas cylinder is kept inside the storage closet, out of view when standing in the kitchen.", + "E": "The red gas pipe is placed beside the staircase." + }, + "answer_label": "E", + "answer_text": "The red gas pipe is placed beside the staircase.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_6", + "correct_answer": "The red gas pipe is placed beside the staircase.", + "original_options": { + "A": "Under the dining table.", + "B": "Near the front door.", + "C": "The red gas cylinder is placed beside the kitchen island, close to the edge facing you.", + "D": "The red gas cylinder is kept inside the storage closet, out of view when standing in the kitchen.", + "E": "The red gas pipe is placed beside the staircase." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 18 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_19", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are placed beside the dining table?", + "answer": "Three chairs were placed beside the dining table.", + "question_chinese": "餐桌旁摆了几条椅子？", + "answer_chinese": "餐桌旁摆了三条椅子。", + "evidence": [ + { + "description": "Three chairs were placed beside the dining table.", + "timestamps": { + "start": "00:11:16", + "end": "00:11:17" + } + } + ], + "question_timestamps": "00:22:34", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two chairs.", + "B": "Three chairs were placed beside the dining table.", + "C": "Four chairs.", + "D": "Five chairs were placed beside the dining table, lined up along one side.", + "E": "One chair was placed beside the dining table, set alone near the table’s corner." + }, + "answer_label": "B", + "answer_text": "Three chairs were placed beside the dining table.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_19", + "correct_answer": "Three chairs were placed beside the dining table.", + "original_options": { + "A": "Two chairs.", + "B": "Three chairs were placed beside the dining table.", + "C": "Four chairs.", + "D": "Five chairs were placed beside the dining table, lined up along one side.", + "E": "One chair was placed beside the dining table, set alone near the table’s corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b", + "video_length": "01:34:05", + "qid": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_20", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are placed beside the dining table?", + "answer": "Three chairs were placed beside the dining table.", + "question_chinese": "餐桌旁摆了几条椅子？", + "answer_chinese": "餐桌旁摆了三条椅子。", + "evidence": [ + { + "description": "Three chairs were placed beside the dining table.", + "timestamps": { + "start": "00:11:16", + "end": "00:11:17" + } + } + ], + "question_timestamps": "00:33:34", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two chairs.", + "B": "Four chairs.", + "C": "Five chairs were placed beside the dining table, positioned closely together along the side.", + "D": "Three chairs were placed beside the dining table.", + "E": "Seven chairs were placed beside the dining table, extending from one end to the other." + }, + "answer_label": "D", + "answer_text": "Three chairs were placed beside the dining table.", + "q_id": "7ddbf8a2-5b3e-44cd-b9dc-db17ec06831b_20", + "correct_answer": "Three chairs were placed beside the dining table.", + "original_options": { + "A": "Two chairs.", + "B": "Four chairs.", + "C": "Five chairs were placed beside the dining table, positioned closely together along the side.", + "D": "Three chairs were placed beside the dining table.", + "E": "Seven chairs were placed beside the dining table, extending from one end to the other." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + } +] diff --git a/QAs/7df161bf-0068-45aa-9a29-a57c3092433c-mcq.json b/QAs/7df161bf-0068-45aa-9a29-a57c3092433c-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2ec8838cb679491ba44fde546ee33b8d9a7396fd --- /dev/null +++ b/QAs/7df161bf-0068-45aa-9a29-a57c3092433c-mcq.json @@ -0,0 +1,1693 @@ +[ + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_8_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the sink, what fruit did I wash?", + "answer": "You washed the lemon.", + "question_chinese": "在洗水槽我清洗了什么水果？", + "answer_chinese": "你洗了柠檬。", + "evidence": [ + { + "description": "You washed lemons.", + "timestamps": { + "start": "00:08:56", + "end": "00:08:57" + } + } + ], + "question_timestamps": "00:08:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed the lemon.", + "B": "You washed the banana.", + "C": "You washed the lime.", + "D": "You washed the banana under running water at the sink, then placed it aside on the counter.", + "E": "You washed an apple at the sink, rinsing it briefly before setting it aside to dry." + }, + "answer_label": "A", + "answer_text": "You washed the lemon.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_8_T1", + "correct_answer": "You washed the lemon.", + "original_options": { + "A": "You washed the lemon.", + "B": "You washed the banana.", + "C": "You washed the lime.", + "D": "You washed the banana under running water at the sink, then placed it aside on the counter.", + "E": "You washed an apple at the sink, rinsing it briefly before setting it aside to dry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_15", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What fruits did I wash in the sink?", + "answer": "You washed the bananas and lemons.", + "question_chinese": "在洗水槽我清洗了什么水果？", + "answer_chinese": "你洗了香蕉和柠檬。", + "evidence": [ + { + "description": "You washed lemons.", + "timestamps": { + "start": "00:08:56", + "end": "00:08:57" + } + }, + { + "description": "You washed the bananas.", + "timestamps": { + "start": "00:08:50", + "end": "00:08:52" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You washed apples and oranges.", + "B": "You washed grapes and limes.", + "C": "You washed the bananas and lemons.", + "D": "You washed cherries and pears in the sink, rinsing them thoroughly before moving on.", + "E": "You washed strawberries and kiwis at the sink, then placed them aside to dry." + }, + "answer_label": "C", + "answer_text": "You washed the bananas and lemons.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_15", + "correct_answer": "You washed the bananas and lemons.", + "original_options": { + "A": "You washed apples and oranges.", + "B": "You washed grapes and limes.", + "C": "You washed the bananas and lemons.", + "D": "You washed cherries and pears in the sink, rinsing them thoroughly before moving on.", + "E": "You washed strawberries and kiwis at the sink, then placed them aside to dry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_8_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the sink, what fruit did I wash?", + "answer": "You washed the bananas and the lemon.", + "question_chinese": "在洗水槽我清洗了什么水果？", + "answer_chinese": "你洗了香蕉和柠檬。", + "evidence": [ + { + "description": "You washed lemons.", + "timestamps": { + "start": "00:08:56", + "end": "00:08:57" + } + }, + { + "description": "You washed the bananas.", + "timestamps": { + "start": "00:08:50", + "end": "00:08:52" + } + } + ], + "question_timestamps": "00:13:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed the lemon.", + "B": "You washed the bananas.", + "C": "You washed the bananas and the lemon.", + "D": "You washed the bananas and lemons together, rinsing multiple lemons in the sink.", + "E": "You washed only one banana at the sink, then set it aside without washing any citrus." + }, + "answer_label": "C", + "answer_text": "You washed the bananas and the lemon.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_8_T2", + "correct_answer": "You washed the bananas and the lemon.", + "original_options": { + "A": "You washed the lemon.", + "B": "You washed the bananas.", + "C": "You washed the bananas and the lemon.", + "D": "You washed the bananas and lemons together, rinsing multiple lemons in the sink.", + "E": "You washed only one banana at the sink, then set it aside without washing any citrus." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 16 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_3_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I need seasonings for cooking—where are the seasonings?", + "answer": "The seasoning is in the cabinet to the left of the stovetop, and the stovetop is behind you.", + "question_chinese": "我做饭需要调料，调料放在哪里？", + "answer_chinese": "调料放在灶台左边的柜子里面，灶台在我的后面。", + "evidence": [ + { + "description": "The seasonings are placed in the cabinet on the left side of the stove.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:03:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The seasonings are on the counter.", + "B": "The seasoning is in the cabinet to the left of the stovetop, and the stovetop is behind you.", + "C": "The seasonings are in a drawer.", + "D": "The seasoning is in the cabinet to the right of the stovetop, and the stovetop is in front of you.", + "E": "The seasoning is in the cabinet above the stovetop, and the stovetop is directly in front of you." + }, + "answer_label": "B", + "answer_text": "The seasoning is in the cabinet to the left of the stovetop, and the stovetop is behind you.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_3_T1", + "correct_answer": "The seasoning is in the cabinet to the left of the stovetop, and the stovetop is behind you.", + "original_options": { + "A": "The seasonings are on the counter.", + "B": "The seasoning is in the cabinet to the left of the stovetop, and the stovetop is behind you.", + "C": "The seasonings are in a drawer.", + "D": "The seasoning is in the cabinet to the right of the stovetop, and the stovetop is in front of you.", + "E": "The seasoning is in the cabinet above the stovetop, and the stovetop is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 18 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_3_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I need seasonings for cooking—where are the seasonings?", + "answer": "The seasoning is in the cabinet to the left of the stovetop, and the stovetop is in front of you.", + "question_chinese": "我做饭需要调料，调料放在哪里？", + "answer_chinese": "调料放在灶台左边的柜子里面，灶台在我面前。", + "evidence": [ + { + "description": "The seasonings are placed in the cabinet on the left side of the stove.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:07:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the cabinet right of the stovetop.", + "B": "In the drawer beneath the stovetop.", + "C": "The seasoning is in the cabinet to the right of the stovetop, with the stovetop directly in front of you.", + "D": "The seasoning is in a cabinet above the stovetop, and the stovetop is positioned in front of you.", + "E": "The seasoning is in the cabinet to the left of the stovetop, and the stovetop is in front of you." + }, + "answer_label": "E", + "answer_text": "The seasoning is in the cabinet to the left of the stovetop, and the stovetop is in front of you.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_3_T2", + "correct_answer": "The seasoning is in the cabinet to the left of the stovetop, and the stovetop is in front of you.", + "original_options": { + "A": "In the cabinet right of the stovetop.", + "B": "In the drawer beneath the stovetop.", + "C": "The seasoning is in the cabinet to the right of the stovetop, with the stovetop directly in front of you.", + "D": "The seasoning is in a cabinet above the stovetop, and the stovetop is positioned in front of you.", + "E": "The seasoning is in the cabinet to the left of the stovetop, and the stovetop is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 18 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_6", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I need seasonings when cooking. Where should I put the seasonings?", + "answer": "The seasonings are placed in the cabinet on the left side of the stove.", + "question_chinese": "我做饭需要调料，调料放在哪里？", + "answer_chinese": "调料放在灶台左边的柜子里面。", + "evidence": [ + { + "description": "The seasonings are placed in the cabinet on the left side of the stove.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:13:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The seasonings are placed in the cabinet on the left side of the stove.", + "B": "Put them in the right cabinet.", + "C": "Store them in the drawer below.", + "D": "Place the seasonings on the counter next to the stove so they’re easy to grab while cooking.", + "E": "Keep the seasonings in the cabinet above the stove so you can reach them from the stovetop." + }, + "answer_label": "A", + "answer_text": "The seasonings are placed in the cabinet on the left side of the stove.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_6", + "correct_answer": "The seasonings are placed in the cabinet on the left side of the stove.", + "original_options": { + "A": "The seasonings are placed in the cabinet on the left side of the stove.", + "B": "Put them in the right cabinet.", + "C": "Store them in the drawer below.", + "D": "Place the seasonings on the counter next to the stove so they’re easy to grab while cooking.", + "E": "Keep the seasonings in the cabinet above the stove so you can reach them from the stovetop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_6_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What ingredients did I add into the glass bowl?", + "answer": "You added eggs and milk into the glass bowl.", + "question_chinese": "我在玻璃碗加入了哪些食材？", + "answer_chinese": "你在玻璃碗里加入了鸡蛋和牛奶。", + "evidence": [ + { + "description": "You added eggs to the glass bowl.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:42" + } + }, + { + "description": "You added milk to the glass bowl.", + "timestamps": { + "start": "00:04:21", + "end": "00:04:24" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You added only eggs.", + "B": "You added only milk.", + "C": "You added eggs, milk, and flour into the glass bowl before mixing everything together.", + "D": "You added eggs and milk into the glass bowl.", + "E": "You added eggs and flour into the glass bowl, then reached for milk separately afterward." + }, + "answer_label": "D", + "answer_text": "You added eggs and milk into the glass bowl.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_6_T1", + "correct_answer": "You added eggs and milk into the glass bowl.", + "original_options": { + "A": "You added only eggs.", + "B": "You added only milk.", + "C": "You added eggs, milk, and flour into the glass bowl before mixing everything together.", + "D": "You added eggs and milk into the glass bowl.", + "E": "You added eggs and flour into the glass bowl, then reached for milk separately afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_11", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What ingredients did I add to the glass bowl?", + "answer": "You added eggs, milk and flour to the glass bowl.v.", + "question_chinese": "我在玻璃碗加入了哪些食材？", + "answer_chinese": "你在玻璃碗里加入了鸡蛋，牛奶和面粉。", + "evidence": [ + { + "description": "You added eggs to the glass bowl.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:42" + } + }, + { + "description": "You added milk to the glass bowl.", + "timestamps": { + "start": "00:04:21", + "end": "00:04:24" + } + }, + { + "description": "You added flour to the glass bowl.", + "timestamps": { + "start": "00:05:58", + "end": "00:06:00" + } + } + ], + "question_timestamps": "00:12:23", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You added eggs and milk.", + "B": "You added eggs, milk and flour to the glass bowl.v.", + "C": "You added milk and flour.", + "D": "You added eggs, milk, and cornstarch to the glass bowl instead of using flour.", + "E": "You added eggs, cream, and sugar to the glass bowl before combining the mixture." + }, + "answer_label": "B", + "answer_text": "You added eggs, milk and flour to the glass bowl.v.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_11", + "correct_answer": "You added eggs, milk and flour to the glass bowl.v.", + "original_options": { + "A": "You added eggs and milk.", + "B": "You added eggs, milk and flour to the glass bowl.v.", + "C": "You added milk and flour.", + "D": "You added eggs, milk, and cornstarch to the glass bowl instead of using flour.", + "E": "You added eggs, cream, and sugar to the glass bowl before combining the mixture." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_6_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What ingredients did I add into the glass bowl?", + "answer": "You added eggs, milk, and flour into the glass bowl.", + "question_chinese": "我在玻璃碗加入了哪些食材？", + "answer_chinese": "你在玻璃碗里加入了鸡蛋，牛奶和面粉。", + "evidence": [ + { + "description": "You added eggs to the glass bowl.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:42" + } + }, + { + "description": "You added milk to the glass bowl.", + "timestamps": { + "start": "00:04:21", + "end": "00:04:24" + } + }, + { + "description": "You added flour to the glass bowl.", + "timestamps": { + "start": "00:05:58", + "end": "00:06:00" + } + } + ], + "question_timestamps": "00:17:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You added eggs, milk, and flour into the glass bowl.", + "B": "You added only eggs.", + "C": "You added only milk.", + "D": "You added eggs and milk into the glass bowl, leaving the flour out for later.", + "E": "You added eggs, milk, and sugar into the glass bowl before stirring the mixture together." + }, + "answer_label": "A", + "answer_text": "You added eggs, milk, and flour into the glass bowl.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_6_T2", + "correct_answer": "You added eggs, milk, and flour into the glass bowl.", + "original_options": { + "A": "You added eggs, milk, and flour into the glass bowl.", + "B": "You added only eggs.", + "C": "You added only milk.", + "D": "You added eggs and milk into the glass bowl, leaving the flour out for later.", + "E": "You added eggs, milk, and sugar into the glass bowl before stirring the mixture together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_1_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where did I move the kettle from, and where did I move it to?", + "answer": "You moved the kettle from the stovetop to the floor.", + "question_chinese": "我将水壶从哪里拿到了哪里？", + "answer_chinese": "你将水壶从灶台上放到了地上。", + "evidence": [ + { + "description": "The kettle is on the stove.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "You put the kettle on the ground.", + "timestamps": { + "start": "00:00:42", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the floor to the stovetop.", + "B": "From the counter to the floor.", + "C": "You moved the kettle from the stovetop onto the counter beside it to clear space for cooking.", + "D": "You moved the kettle from the floor back up to the stovetop after setting it down briefly.", + "E": "You moved the kettle from the stovetop to the floor." + }, + "answer_label": "E", + "answer_text": "You moved the kettle from the stovetop to the floor.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_1_T2", + "correct_answer": "You moved the kettle from the stovetop to the floor.", + "original_options": { + "A": "From the floor to the stovetop.", + "B": "From the counter to the floor.", + "C": "You moved the kettle from the stovetop onto the counter beside it to clear space for cooking.", + "D": "You moved the kettle from the floor back up to the stovetop after setting it down briefly.", + "E": "You moved the kettle from the stovetop to the floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_1_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where did I move the kettle from, and where did I move it to?", + "answer": "You moved the kettle from the stovetop to the floor, and then put it back on the stovetop.", + "question_chinese": "我将水壶从哪里拿到了哪里？", + "answer_chinese": "你将水壶从灶台上放到了地上又放回了灶台上。", + "evidence": [ + { + "description": "The kettle is on the stove.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "You put the kettle on the ground.", + "timestamps": { + "start": "00:00:42", + "end": "00:01:08" + } + }, + { + "description": "You put the kettle on the stove.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:04:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the floor back to the stovetop.", + "B": "You moved the kettle from the stovetop to the floor, and then put it back on the stovetop.", + "C": "From the stovetop to the counter.", + "D": "You moved the kettle from the stovetop to the floor and left it there without putting it back.", + "E": "You moved the kettle off the stovetop onto the counter, then returned it to the stovetop afterward." + }, + "answer_label": "B", + "answer_text": "You moved the kettle from the stovetop to the floor, and then put it back on the stovetop.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_1_T1", + "correct_answer": "You moved the kettle from the stovetop to the floor, and then put it back on the stovetop.", + "original_options": { + "A": "From the floor back to the stovetop.", + "B": "You moved the kettle from the stovetop to the floor, and then put it back on the stovetop.", + "C": "From the stovetop to the counter.", + "D": "You moved the kettle from the stovetop to the floor and left it there without putting it back.", + "E": "You moved the kettle off the stovetop onto the counter, then returned it to the stovetop afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_3", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I put the kraft paper bag?", + "answer": "You put the kraft paper bag into the cabinet above the stove.", + "question_chinese": "我将牛皮纸袋放在了哪里？", + "answer_chinese": "你将牛皮纸袋放进了灶台上面的柜子里面。", + "evidence": [ + { + "description": "You put the kraft paper bag into the cabinet above the stove.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:24" + } + }, + { + "description": "The cabinet on the upper right of the stove.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:21" + } + } + ], + "question_timestamps": "00:03:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the shelf beside the refrigerator.", + "B": "In the drawer next to the oven.", + "C": "You put the kraft paper bag into the cabinet above the stove.", + "D": "You placed the kraft paper bag under the table near the window after you were done with it.", + "E": "You stored the kraft paper bag on the floor beside the sink instead of putting it away in a cabinet." + }, + "answer_label": "C", + "answer_text": "You put the kraft paper bag into the cabinet above the stove.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_3", + "correct_answer": "You put the kraft paper bag into the cabinet above the stove.", + "original_options": { + "A": "On the shelf beside the refrigerator.", + "B": "In the drawer next to the oven.", + "C": "You put the kraft paper bag into the cabinet above the stove.", + "D": "You placed the kraft paper bag under the table near the window after you were done with it.", + "E": "You stored the kraft paper bag on the floor beside the sink instead of putting it away in a cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 20 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_4", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I put the kraft paper bag?", + "answer": "You put the kraft paper bag into the cabinet above the stove.", + "question_chinese": "我将牛皮纸袋放在了哪里？", + "answer_chinese": "你将牛皮纸袋放进了灶台上面的柜子里面。", + "evidence": [ + { + "description": "You put the kraft paper bag into the cabinet above the stove.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:24" + } + }, + { + "description": "The cabinet on the upper right of the stove.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:21" + } + } + ], + "question_timestamps": "00:07:28", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You put the kraft paper bag into the cabinet above the stove.", + "B": "Into the cabinet above the refrigerator.", + "C": "Into the cabinet next to the sink.", + "D": "You put the kraft paper bag into the cabinet below the countertop, not in an upper cabinet.", + "E": "You placed the kraft paper bag inside the pantry cabinet beside the oven rather than above the stove." + }, + "answer_label": "A", + "answer_text": "You put the kraft paper bag into the cabinet above the stove.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_4", + "correct_answer": "You put the kraft paper bag into the cabinet above the stove.", + "original_options": { + "A": "You put the kraft paper bag into the cabinet above the stove.", + "B": "Into the cabinet above the refrigerator.", + "C": "Into the cabinet next to the sink.", + "D": "You put the kraft paper bag into the cabinet below the countertop, not in an upper cabinet.", + "E": "You placed the kraft paper bag inside the pantry cabinet beside the oven rather than above the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 18 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_13", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where did I put the milk after I finished using it?", + "answer": "You put the milk from the stove into the refrigerator.", + "question_chinese": "我将用完牛奶后将牛奶从哪里放到了哪里？", + "answer_chinese": "你将牛奶从灶台上放进了冰箱里。", + "evidence": [ + { + "description": "The milk is on the stove.", + "timestamps": { + "start": "00:08:04", + "end": "00:08:09" + } + }, + { + "description": "The milk was put in the fridge.", + "timestamps": { + "start": "00:08:21", + "end": "00:08:23" + } + } + ], + "question_timestamps": "00:15:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From the stove onto the table.", + "B": "From the counter onto the shelf.", + "C": "You put the milk from the freezer into the refrigerator after you finished using it.", + "D": "You moved the milk from the cupboard and set it down in the sink instead of refrigerating it.", + "E": "You put the milk from the stove into the refrigerator." + }, + "answer_label": "E", + "answer_text": "You put the milk from the stove into the refrigerator.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_13", + "correct_answer": "You put the milk from the stove into the refrigerator.", + "original_options": { + "A": "From the stove onto the table.", + "B": "From the counter onto the shelf.", + "C": "You put the milk from the freezer into the refrigerator after you finished using it.", + "D": "You moved the milk from the cupboard and set it down in the sink instead of refrigerating it.", + "E": "You put the milk from the stove into the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 18 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_14", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where did I put the milk after I finished using it?", + "answer": "You put the milk from the stove into the refrigerator.", + "question_chinese": "我将用完牛奶后将牛奶从哪里放到了哪里？", + "answer_chinese": "你将牛奶从灶台上放进了冰箱里。", + "evidence": [ + { + "description": "The milk is on the stove.", + "timestamps": { + "start": "00:08:04", + "end": "00:08:09" + } + }, + { + "description": "The milk was put in the fridge.", + "timestamps": { + "start": "00:08:21", + "end": "00:08:23" + } + } + ], + "question_timestamps": "00:22:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You put the milk from the stove into the refrigerator.", + "B": "From the counter into the refrigerator.", + "C": "From the stove into the freezer.", + "D": "You put the milk from the stove into the kitchen cabinet after you finished using it.", + "E": "You moved the milk from the table into the refrigerator rather than taking it from the stove." + }, + "answer_label": "A", + "answer_text": "You put the milk from the stove into the refrigerator.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_14", + "correct_answer": "You put the milk from the stove into the refrigerator.", + "original_options": { + "A": "You put the milk from the stove into the refrigerator.", + "B": "From the counter into the refrigerator.", + "C": "From the stove into the freezer.", + "D": "You put the milk from the stove into the kitchen cabinet after you finished using it.", + "E": "You moved the milk from the table into the refrigerator rather than taking it from the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_7_T1", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "After I finished using the milk, where did I move it from and to, and where was it relative to me?", + "answer": "You put the milk from the stovetop into the fridge, and the fridge is behind you.", + "question_chinese": "我将用完牛奶后将牛奶从哪里放到了哪里，在相对于我的什么位置？", + "answer_chinese": "你将牛奶从灶台上放进了冰箱里，冰箱在你的后面。", + "evidence": [ + { + "description": "The milk is on the stove.", + "timestamps": { + "start": "00:08:04", + "end": "00:08:09" + } + }, + { + "description": "The milk was put in the fridge.", + "timestamps": { + "start": "00:08:21", + "end": "00:08:23" + } + } + ], + "question_timestamps": "00:08:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the stovetop to the fridge, in front.", + "B": "From the stovetop to the fridge, left.", + "C": "You put the milk from the stovetop into the fridge, and the fridge was positioned to your right.", + "D": "You moved the milk from the stovetop into the fridge, with the fridge located directly in front of you.", + "E": "You put the milk from the stovetop into the fridge, and the fridge is behind you." + }, + "answer_label": "E", + "answer_text": "You put the milk from the stovetop into the fridge, and the fridge is behind you.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_7_T1", + "correct_answer": "You put the milk from the stovetop into the fridge, and the fridge is behind you.", + "original_options": { + "A": "From the stovetop to the fridge, in front.", + "B": "From the stovetop to the fridge, left.", + "C": "You put the milk from the stovetop into the fridge, and the fridge was positioned to your right.", + "D": "You moved the milk from the stovetop into the fridge, with the fridge located directly in front of you.", + "E": "You put the milk from the stovetop into the fridge, and the fridge is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 19 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_7_T2", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "After I finished using the milk, where did I move it from and to, and where was it relative to me?", + "answer": "You put the milk from the stovetop into the fridge, and the fridge is to your left.", + "question_chinese": "我将用完牛奶后将牛奶从哪里放到了哪里，在相对于我的什么位置？", + "answer_chinese": "你将牛奶从灶台上放进了冰箱里，冰箱在我的左边。", + "evidence": [ + { + "description": "The milk is on the stove.", + "timestamps": { + "start": "00:08:04", + "end": "00:08:09" + } + }, + { + "description": "The milk was put in the fridge.", + "timestamps": { + "start": "00:08:21", + "end": "00:08:23" + } + } + ], + "question_timestamps": "00:22:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the stovetop to the fridge, behind you.", + "B": "You put the milk from the stovetop into the fridge, and the fridge is to your left.", + "C": "From the stovetop to the fridge, right.", + "D": "You put the milk from the stovetop into the fridge, and the fridge was in front of you.", + "E": "You moved the milk from the stovetop into the fridge, with the fridge positioned to your right." + }, + "answer_label": "B", + "answer_text": "You put the milk from the stovetop into the fridge, and the fridge is to your left.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_7_T2", + "correct_answer": "You put the milk from the stovetop into the fridge, and the fridge is to your left.", + "original_options": { + "A": "From the stovetop to the fridge, behind you.", + "B": "You put the milk from the stovetop into the fridge, and the fridge is to your left.", + "C": "From the stovetop to the fridge, right.", + "D": "You put the milk from the stovetop into the fridge, and the fridge was in front of you.", + "E": "You moved the milk from the stovetop into the fridge, with the fridge positioned to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 17 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_5_T1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many green placemats did I see on the dining table?", + "answer": "You saw four placemats laid out.", + "question_chinese": "我看到餐桌上放了几块绿色的餐垫？", + "answer_chinese": "你看到放了四块餐垫。", + "evidence": [ + { + "description": "Four placemats were placed.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:11" + } + } + ], + "question_timestamps": "00:14:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two placemats.", + "B": "You saw three placemats.", + "C": "You saw five green placemats laid out on the dining table that day.", + "D": "You saw six green placemats laid out across the dining table during the meal.", + "E": "You saw four placemats laid out." + }, + "answer_label": "E", + "answer_text": "You saw four placemats laid out.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_5_T1", + "correct_answer": "You saw four placemats laid out.", + "original_options": { + "A": "You saw two placemats.", + "B": "You saw three placemats.", + "C": "You saw five green placemats laid out on the dining table that day.", + "D": "You saw six green placemats laid out across the dining table during the meal.", + "E": "You saw four placemats laid out." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_5_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many green placemats did I see on the dining table?", + "answer": "You previously saw four placemats laid out.", + "question_chinese": "我看到餐桌上放了几块绿色的餐垫？", + "answer_chinese": "你之前看到放了四块餐垫。", + "evidence": [ + { + "description": "Four placemats were placed.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:11" + } + } + ], + "question_timestamps": "00:22:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You previously saw four placemats laid out.", + "B": "You previously saw three placemats laid out.", + "C": "You previously saw five placemats laid out.", + "D": "You previously saw two placemats laid out on the dining table when you looked over.", + "E": "You previously saw six placemats laid out across the dining table, filling most of the place settings." + }, + "answer_label": "A", + "answer_text": "You previously saw four placemats laid out.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_5_T2", + "correct_answer": "You previously saw four placemats laid out.", + "original_options": { + "A": "You previously saw four placemats laid out.", + "B": "You previously saw three placemats laid out.", + "C": "You previously saw five placemats laid out.", + "D": "You previously saw two placemats laid out on the dining table when you looked over.", + "E": "You previously saw six placemats laid out across the dining table, filling most of the place settings." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 17 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_4_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many pots did I see stored in the cabinet under the sink?", + "answer": "You can see that there are two pots set out.", + "question_chinese": "我看见洗水槽下面的柜子里放有几个锅？", + "answer_chinese": "你看见放有两个锅。", + "evidence": [ + { + "description": "You took out two pots.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:33" + } + }, + { + "description": "There are still two pots left.", + "timestamps": { + "start": "00:02:37", + "end": "00:02:38" + } + } + ], + "question_timestamps": "00:05:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can see one pot set out.", + "B": "You can see three pots set out.", + "C": "You can see that there are four pots set out under the sink cabinet when you check inside.", + "D": "You can see that there are two pots set out.", + "E": "You can see that there are zero pots set out in the cabinet, so nothing is stored there." + }, + "answer_label": "D", + "answer_text": "You can see that there are two pots set out.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_4_T1", + "correct_answer": "You can see that there are two pots set out.", + "original_options": { + "A": "You can see one pot set out.", + "B": "You can see three pots set out.", + "C": "You can see that there are four pots set out under the sink cabinet when you check inside.", + "D": "You can see that there are two pots set out.", + "E": "You can see that there are zero pots set out in the cabinet, so nothing is stored there." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_4_T2", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many pots did I see stored in the cabinet under the sink?", + "answer": "You previously saw four pots, and now there are two—you took two away.", + "question_chinese": "我看见洗水槽下面的柜子里放有几个锅？", + "answer_chinese": "你之前看见放有四个锅，现在有两个，你拿走了两个。", + "evidence": [ + { + "description": "You took out two pots.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:33" + } + }, + { + "description": "There are still two pots left.", + "timestamps": { + "start": "00:02:37", + "end": "00:02:38" + } + } + ], + "question_timestamps": "00:15:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two pots now.", + "B": "Three pots now.", + "C": "You previously saw five pots, and now there are two—you took three away.", + "D": "You previously saw six pots, and now there are two—you took four away and left two behind.", + "E": "You previously saw four pots, and now there are two—you took two away." + }, + "answer_label": "E", + "answer_text": "You previously saw four pots, and now there are two—you took two away.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_4_T2", + "correct_answer": "You previously saw four pots, and now there are two—you took two away.", + "original_options": { + "A": "Two pots now.", + "B": "Three pots now.", + "C": "You previously saw five pots, and now there are two—you took three away.", + "D": "You previously saw six pots, and now there are two—you took four away and left two behind.", + "E": "You previously saw four pots, and now there are two—you took two away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 18 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_7", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many POTS are there in the cabinet under the sink?", + "answer": "There are four pots in it.", + "question_chinese": "洗水槽下面的柜子里放有几个锅？", + "answer_chinese": "放有四个锅。", + "evidence": [ + { + "description": "You took out two pots.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:33" + } + }, + { + "description": "There are still two pots left.", + "timestamps": { + "start": "00:02:37", + "end": "00:02:38" + } + } + ], + "question_timestamps": "00:05:53", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are five pots in it.", + "B": "There are six pots in it.", + "C": "There are seven pots in it, filling the cabinet space under the sink.", + "D": "There are eight pots in it, stacked together inside the cabinet under the sink.", + "E": "There are four pots in it." + }, + "answer_label": "E", + "answer_text": "There are four pots in it.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_7", + "correct_answer": "There are four pots in it.", + "original_options": { + "A": "There are five pots in it.", + "B": "There are six pots in it.", + "C": "There are seven pots in it, filling the cabinet space under the sink.", + "D": "There are eight pots in it, stacked together inside the cabinet under the sink.", + "E": "There are four pots in it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 14 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_8", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many POTS are there in the cabinet under the sink?", + "answer": "There are four pots in it.", + "question_chinese": "洗水槽下面的柜子里放有几个锅？", + "answer_chinese": "放有四个锅。", + "evidence": [ + { + "description": "You took out two pots.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:33" + } + }, + { + "description": "There are still two pots left.", + "timestamps": { + "start": "00:02:37", + "end": "00:02:38" + } + } + ], + "question_timestamps": "00:15:53", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are three pots in it.", + "B": "There are four pots in it.", + "C": "There are five pots in it.", + "D": "There are six pots in it, lined up inside the cabinet under the sink.", + "E": "There are eight pots in it, with several stacked and several sitting side by side." + }, + "answer_label": "B", + "answer_text": "There are four pots in it.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_8", + "correct_answer": "There are four pots in it.", + "original_options": { + "A": "There are three pots in it.", + "B": "There are four pots in it.", + "C": "There are five pots in it.", + "D": "There are six pots in it, lined up inside the cabinet under the sink.", + "E": "There are eight pots in it, with several stacked and several sitting side by side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_9", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many green placemats are there on the dining table?", + "answer": "Four placemats were placed.", + "question_chinese": "餐桌上放了几块绿色的餐垫？", + "answer_chinese": "放了四块餐垫。", + "evidence": [ + { + "description": "Four placemats were placed.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:11" + } + } + ], + "question_timestamps": "00:14:17", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three placemats were placed.", + "B": "Five placemats were placed.", + "C": "Six placemats were placed on the dining table when you last checked the setup.", + "D": "Seven placemats were placed, covering nearly the entire dining table surface area.", + "E": "Four placemats were placed." + }, + "answer_label": "E", + "answer_text": "Four placemats were placed.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_9", + "correct_answer": "Four placemats were placed.", + "original_options": { + "A": "Three placemats were placed.", + "B": "Five placemats were placed.", + "C": "Six placemats were placed on the dining table when you last checked the setup.", + "D": "Seven placemats were placed, covering nearly the entire dining table surface area.", + "E": "Four placemats were placed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 12 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_10", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many green placemats are there on the dining table?", + "answer": "Four placemats were placed.", + "question_chinese": "餐桌上放了几块绿色的餐垫？", + "answer_chinese": "放了四块餐垫。", + "evidence": [ + { + "description": "Four placemats were placed.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:11" + } + } + ], + "question_timestamps": "00:22:17", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three placemats were placed.", + "B": "Five placemats were placed.", + "C": "Four placemats were placed.", + "D": "Six placemats were placed on the dining table, spaced evenly at each seat.", + "E": "Seven placemats were placed, leaving little empty space between the settings on the table." + }, + "answer_label": "C", + "answer_text": "Four placemats were placed.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_10", + "correct_answer": "Four placemats were placed.", + "original_options": { + "A": "Three placemats were placed.", + "B": "Five placemats were placed.", + "C": "Four placemats were placed.", + "D": "Six placemats were placed on the dining table, spaced evenly at each seat.", + "E": "Seven placemats were placed, leaving little empty space between the settings on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_17", + "qid_num": 28, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What variety of potted plants are there on the dining table?", + "answer": "It's the turtle-back bamboo.", + "question_chinese": "餐桌上的盆栽是什么品种？", + "answer_chinese": "是龟背竹。", + "evidence": [ + { + "description": "It's the turtle-back bamboo.", + "timestamps": { + "start": "00:28:55", + "end": "00:28:57" + } + } + ], + "question_timestamps": "00:39:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's the jade plant.", + "B": "It's the turtle-back bamboo.", + "C": "It's the spider plant.", + "D": "It's the rubber plant, the one with thicker, glossy leaves in the pot.", + "E": "It's the peace lily, with broad leaves and a distinctive white bloom near the center." + }, + "answer_label": "B", + "answer_text": "It's the turtle-back bamboo.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_17", + "correct_answer": "It's the turtle-back bamboo.", + "original_options": { + "A": "It's the jade plant.", + "B": "It's the turtle-back bamboo.", + "C": "It's the spider plant.", + "D": "It's the rubber plant, the one with thicker, glossy leaves in the pot.", + "E": "It's the peace lily, with broad leaves and a distinctive white bloom near the center." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 15 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_18", + "qid_num": 29, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What variety of potted plants are there on the dining table?", + "answer": "It's the turtle-back bamboo.", + "question_chinese": "餐桌上的盆栽是什么品种？", + "answer_chinese": "是龟背竹。", + "evidence": [ + { + "description": "It's the turtle-back bamboo.", + "timestamps": { + "start": "00:28:55", + "end": "00:28:57" + } + } + ], + "question_timestamps": "00:43:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's the turtle-back bamboo.", + "B": "It's lucky bamboo.", + "C": "It's a jade plant.", + "D": "It's a spider plant in a small pot on the dining table.", + "E": "It's a pothos plant sitting in a planter on the dining table." + }, + "answer_label": "A", + "answer_text": "It's the turtle-back bamboo.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_18", + "correct_answer": "It's the turtle-back bamboo.", + "original_options": { + "A": "It's the turtle-back bamboo.", + "B": "It's lucky bamboo.", + "C": "It's a jade plant.", + "D": "It's a spider plant in a small pot on the dining table.", + "E": "It's a pothos plant sitting in a planter on the dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 12 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_9_T1", + "qid_num": 30, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kind of plant is the potted plant on the dining table, and where is it relative to me?", + "answer": "It was a Monstera, and now it’s behind you.", + "question_chinese": "餐桌上的盆栽是什么品种，在相对于我的什么位置？", + "answer_chinese": "是龟背竹，现在在我的后面。", + "evidence": [ + { + "description": "It's the turtle-back bamboo.", + "timestamps": { + "start": "00:28:55", + "end": "00:28:57" + } + }, + { + "description": "The dining table is behind me.", + "timestamps": { + "start": "00:28:58", + "end": "00:29:08" + } + } + ], + "question_timestamps": "00:29:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was a Monstera, and now it’s behind you.", + "B": "It was a Monstera, in front of you.", + "C": "It was a Monstera, to your left.", + "D": "It was a Monstera, and now it’s to your right after you turned around.", + "E": "It was a Monstera, and now it’s above you, positioned higher than your eye level." + }, + "answer_label": "A", + "answer_text": "It was a Monstera, and now it’s behind you.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_9_T1", + "correct_answer": "It was a Monstera, and now it’s behind you.", + "original_options": { + "A": "It was a Monstera, and now it’s behind you.", + "B": "It was a Monstera, in front of you.", + "C": "It was a Monstera, to your left.", + "D": "It was a Monstera, and now it’s to your right after you turned around.", + "E": "It was a Monstera, and now it’s above you, positioned higher than your eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 16 + ] + }, + { + "video_uid": "7df161bf-0068-45aa-9a29-a57c3092433c", + "video_length": "00:53:09", + "qid": "7df161bf-0068-45aa-9a29-a57c3092433c_9_T2", + "qid_num": 31, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kind of plant is the potted plant on the dining table, and where is it relative to me?", + "answer": "It was a Monstera, and now it’s in front of you.", + "question_chinese": "餐桌上的盆栽是什么品种，在相对于我的什么位置？", + "answer_chinese": "是龟背竹，现在在我的前面。", + "evidence": [ + { + "description": "It's the turtle-back bamboo.", + "timestamps": { + "start": "00:28:55", + "end": "00:28:57" + } + }, + { + "description": "The dining table is in front of me.", + "timestamps": { + "start": "00:29:30", + "end": "00:29:34" + } + } + ], + "question_timestamps": "00:29:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was a Monstera, behind you.", + "B": "It was a Monstera, to your right.", + "C": "It was a Monstera, and now it’s in front of you.", + "D": "It was a Monstera, and now it’s to your left relative to where you’re facing.", + "E": "It was a Monstera, and now it’s above you, not at table height anymore." + }, + "answer_label": "C", + "answer_text": "It was a Monstera, and now it’s in front of you.", + "q_id": "7df161bf-0068-45aa-9a29-a57c3092433c_9_T2", + "correct_answer": "It was a Monstera, and now it’s in front of you.", + "original_options": { + "A": "It was a Monstera, behind you.", + "B": "It was a Monstera, to your right.", + "C": "It was a Monstera, and now it’s in front of you.", + "D": "It was a Monstera, and now it’s to your left relative to where you’re facing.", + "E": "It was a Monstera, and now it’s above you, not at table height anymore." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 15 + ] + } +] diff --git a/QAs/7df753b4-28b2-42ac-b448-4c7501a69871-mcq.json b/QAs/7df753b4-28b2-42ac-b448-4c7501a69871-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3ff430850d0ec3088f286ca22dbb2fd5110f064b --- /dev/null +++ b/QAs/7df753b4-28b2-42ac-b448-4c7501a69871-mcq.json @@ -0,0 +1,330 @@ +[ + { + "video_uid": "7df753b4-28b2-42ac-b448-4c7501a69871", + "video_length": "00:55:53", + "qid": "7df753b4-28b2-42ac-b448-4c7501a69871_1_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I climb up the ladder?", + "answer": "You climbed the ladder three times, and the ladder is to your back-right.", + "question_chinese": "我爬上了几次爬梯？", + "answer_chinese": "你爬过三次爬梯，梯子在你的右后方。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:00:27", + "end": "00:00:38" + } + }, + { + "description": "Second time", + "timestamps": { + "start": "00:06:35", + "end": "00:06:44" + } + }, + { + "description": "Third time", + "timestamps": { + "start": "00:12:30", + "end": "00:12:57" + } + } + ], + "question_timestamps": "00:14:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You climbed it twice.", + "B": "You climbed it four times.", + "C": "You climbed the ladder three times, and the ladder is to your back-right.", + "D": "You climbed the ladder two times, and the ladder is to your back-right.", + "E": "You climbed the ladder five times, and the ladder is to your back-right throughout." + }, + "answer_label": "C", + "answer_text": "You climbed the ladder three times, and the ladder is to your back-right.", + "q_id": "7df753b4-28b2-42ac-b448-4c7501a69871_1_T1", + "correct_answer": "You climbed the ladder three times, and the ladder is to your back-right.", + "original_options": { + "A": "You climbed it twice.", + "B": "You climbed it four times.", + "C": "You climbed the ladder three times, and the ladder is to your back-right.", + "D": "You climbed the ladder two times, and the ladder is to your back-right.", + "E": "You climbed the ladder five times, and the ladder is to your back-right throughout." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 14 + ] + }, + { + "video_uid": "7df753b4-28b2-42ac-b448-4c7501a69871", + "video_length": "00:55:53", + "qid": "7df753b4-28b2-42ac-b448-4c7501a69871_1_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I climb up the ladder?", + "answer": "You climbed the ladder three times, and the ladder is to your back-left.", + "question_chinese": "我爬上了几次爬梯？", + "answer_chinese": "你爬过三次爬梯，梯子在你的左后方。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:00:27", + "end": "00:00:38" + } + }, + { + "description": "Second time", + "timestamps": { + "start": "00:06:35", + "end": "00:06:44" + } + }, + { + "description": "Third time", + "timestamps": { + "start": "00:12:30", + "end": "00:12:57" + } + } + ], + "question_timestamps": "00:28:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You climbed it twice.", + "B": "You climbed it four times.", + "C": "You climbed the ladder one time, and the ladder is to your back-left.", + "D": "You climbed the ladder three times, and the ladder is to your back-left.", + "E": "You climbed the ladder five times, and the ladder is to your back-left the whole time." + }, + "answer_label": "D", + "answer_text": "You climbed the ladder three times, and the ladder is to your back-left.", + "q_id": "7df753b4-28b2-42ac-b448-4c7501a69871_1_T2", + "correct_answer": "You climbed the ladder three times, and the ladder is to your back-left.", + "original_options": { + "A": "You climbed it twice.", + "B": "You climbed it four times.", + "C": "You climbed the ladder one time, and the ladder is to your back-left.", + "D": "You climbed the ladder three times, and the ladder is to your back-left.", + "E": "You climbed the ladder five times, and the ladder is to your back-left the whole time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 16 + ] + }, + { + "video_uid": "7df753b4-28b2-42ac-b448-4c7501a69871", + "video_length": "00:55:53", + "qid": "7df753b4-28b2-42ac-b448-4c7501a69871_5", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I now reach the high-pressure water gun without moving my feet?", + "answer": "You can't get the water gun.", + "question_chinese": "我现在能在不移动脚步的情况下拿到高压水枪吗？", + "answer_chinese": "你不能拿到水枪。", + "evidence": [ + { + "description": "I put the water gun on the lower layer of the wooden board.", + "timestamps": { + "start": "00:14:00", + "end": "00:14:31" + } + } + ], + "question_timestamps": "00:15:28", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't get the water gun.", + "B": "Yes, you can reach the high-pressure water gun without moving your feet." + }, + "answer_label": "A", + "answer_text": "You can't get the water gun.", + "q_id": "7df753b4-28b2-42ac-b448-4c7501a69871_5", + "correct_answer": "You can't get the water gun.", + "original_options": { + "A": "You can't get the water gun.", + "B": "Yes, you can reach the high-pressure water gun without moving your feet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "7df753b4-28b2-42ac-b448-4c7501a69871", + "video_length": "00:55:53", + "qid": "7df753b4-28b2-42ac-b448-4c7501a69871_6", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I now reach the high-pressure water gun without moving my feet?", + "answer": "You cannot get a high-pressure water gun.", + "question_chinese": "我现在能在不移动脚步的情况下拿到高压水枪吗？", + "answer_chinese": "你不能拿到高压水枪。", + "evidence": [ + { + "description": "You put the water gun on the second floor.", + "timestamps": { + "start": "00:36:43", + "end": "00:36:58" + } + } + ], + "question_timestamps": "00:41:53", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can get the high-pressure water gun without moving your feet.", + "B": "You cannot get a high-pressure water gun." + }, + "answer_label": "B", + "answer_text": "You cannot get a high-pressure water gun.", + "q_id": "7df753b4-28b2-42ac-b448-4c7501a69871_6", + "correct_answer": "You cannot get a high-pressure water gun.", + "original_options": { + "A": "Yes, you can get the high-pressure water gun without moving your feet.", + "B": "You cannot get a high-pressure water gun." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "7df753b4-28b2-42ac-b448-4c7501a69871", + "video_length": "00:55:53", + "qid": "7df753b4-28b2-42ac-b448-4c7501a69871_2_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my pickaxe?", + "answer": "The pickaxe is to your back-right.", + "question_chinese": "我的镐子在哪里？", + "answer_chinese": "镐子在你的右后方。", + "evidence": [ + { + "description": "The pickaxe is to my left rear.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:03:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The pickaxe is to your back-left.", + "B": "The pickaxe is to your front-right.", + "C": "The pickaxe is to your front-left, a little off to that side.", + "D": "The pickaxe is directly behind you, not angled toward either side at all.", + "E": "The pickaxe is to your back-right." + }, + "answer_label": "E", + "answer_text": "The pickaxe is to your back-right.", + "q_id": "7df753b4-28b2-42ac-b448-4c7501a69871_2_T2", + "correct_answer": "The pickaxe is to your back-right.", + "original_options": { + "A": "The pickaxe is to your back-left.", + "B": "The pickaxe is to your front-right.", + "C": "The pickaxe is to your front-left, a little off to that side.", + "D": "The pickaxe is directly behind you, not angled toward either side at all.", + "E": "The pickaxe is to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 13 + ] + }, + { + "video_uid": "7df753b4-28b2-42ac-b448-4c7501a69871", + "video_length": "00:55:53", + "qid": "7df753b4-28b2-42ac-b448-4c7501a69871_2_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my pickaxe?", + "answer": "The pickaxe is to your back-left.", + "question_chinese": "我的镐子在哪里？", + "answer_chinese": "镐子在你的左后方。", + "evidence": [ + { + "description": "The pickaxe is to my left rear.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:14:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The pickaxe is to your front-left.", + "B": "The pickaxe is to your back-right.", + "C": "The pickaxe is off to your front-right, a little ahead of where you’re facing.", + "D": "The pickaxe is to your back-left.", + "E": "The pickaxe is directly behind you, centered rather than off to the left side." + }, + "answer_label": "D", + "answer_text": "The pickaxe is to your back-left.", + "q_id": "7df753b4-28b2-42ac-b448-4c7501a69871_2_T1", + "correct_answer": "The pickaxe is to your back-left.", + "original_options": { + "A": "The pickaxe is to your front-left.", + "B": "The pickaxe is to your back-right.", + "C": "The pickaxe is off to your front-right, a little ahead of where you’re facing.", + "D": "The pickaxe is to your back-left.", + "E": "The pickaxe is directly behind you, centered rather than off to the left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 14 + ] + } +] diff --git a/QAs/7e3b7dff-bf38-4352-8adc-43799beb556f-mcq.json b/QAs/7e3b7dff-bf38-4352-8adc-43799beb556f-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d53bd1d7b595a8ecabb4b2c642f2075702ffb42b --- /dev/null +++ b/QAs/7e3b7dff-bf38-4352-8adc-43799beb556f-mcq.json @@ -0,0 +1,1398 @@ +[ + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_6", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "The trash can is placed upright, or is it tilted and lying on the ground?", + "answer": "The trash can was lying on the ground at a slant.", + "question_chinese": "垃圾桶是正着放的，还是歪着倒在地上的？", + "answer_chinese": "垃圾桶是歪着倒在地上的。", + "evidence": [ + { + "description": "The trash can was lying on the ground at a slant.", + "timestamps": { + "start": "00:02:43", + "end": "00:02:46" + } + } + ], + "question_timestamps": "00:04:25", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The trash can was placed upright.", + "B": "The trash can was lying on the ground at a slant." + }, + "answer_label": "B", + "answer_text": "The trash can was lying on the ground at a slant.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_6", + "correct_answer": "The trash can was lying on the ground at a slant.", + "original_options": { + "A": "The trash can was placed upright.", + "B": "The trash can was lying on the ground at a slant." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_7", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "The trash can is placed upright, or is it tilted and lying on the ground?", + "answer": "The trash can is placed upright.", + "question_chinese": "垃圾桶是正着放的，还是歪着倒在地上的？", + "answer_chinese": "垃圾桶是正着放的。", + "evidence": [ + { + "description": "The trash can was lying on the ground at a slant.", + "timestamps": { + "start": "00:02:43", + "end": "00:02:46" + } + } + ], + "question_timestamps": "00:49:27", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The trash can is placed upright.", + "B": "The trash can was tilted and lying on the ground." + }, + "answer_label": "A", + "answer_text": "The trash can is placed upright.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_7", + "correct_answer": "The trash can is placed upright.", + "original_options": { + "A": "The trash can is placed upright.", + "B": "The trash can was tilted and lying on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_12", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many stools do you need to pass through to get from the living room to the sun-drying room?", + "answer": "The living room has to pass three stools to get to the room for drying things.", + "question_chinese": "客厅去到晒东西的房间要经过几张凳子？", + "answer_chinese": "客厅去到晒东西的房间要经过三张凳子。", + "evidence": [ + { + "description": "The living room has to pass three stools to get to the room for drying things.", + "timestamps": { + "start": "00:34:22", + "end": "00:34:28" + } + } + ], + "question_timestamps": "00:37:49", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "The living room has to pass three stools to get to the room for drying things.", + "B": "You pass two stools.", + "C": "You pass four stools.", + "D": "The living room has to pass one stool to get to the room for drying things.", + "E": "The living room has to pass five stools to get to the room for drying things." + }, + "answer_label": "A", + "answer_text": "The living room has to pass three stools to get to the room for drying things.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_12", + "correct_answer": "The living room has to pass three stools to get to the room for drying things.", + "original_options": { + "A": "The living room has to pass three stools to get to the room for drying things.", + "B": "You pass two stools.", + "C": "You pass four stools.", + "D": "The living room has to pass one stool to get to the room for drying things.", + "E": "The living room has to pass five stools to get to the room for drying things." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_11", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many stools do you need to pass through to get from the living room to the sun-drying room?", + "answer": "The living room has to pass three stools to get to the room for drying things.", + "question_chinese": "客厅去到晒东西的房间要经过几张凳子？", + "answer_chinese": "客厅去到晒东西的房间要经过三张凳子。", + "evidence": [ + { + "description": "The living room has to pass three stools to get to the room for drying things.", + "timestamps": { + "start": "00:34:22", + "end": "00:34:28" + } + } + ], + "question_timestamps": "00:50:58", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You pass four stools.", + "B": "You pass six stools.", + "C": "The living room has to pass five stools to get to the room for drying things.", + "D": "The living room has to pass seven stools to get to the room for drying things.", + "E": "The living room has to pass three stools to get to the room for drying things." + }, + "answer_label": "E", + "answer_text": "The living room has to pass three stools to get to the room for drying things.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_11", + "correct_answer": "The living room has to pass three stools to get to the room for drying things.", + "original_options": { + "A": "You pass four stools.", + "B": "You pass six stools.", + "C": "The living room has to pass five stools to get to the room for drying things.", + "D": "The living room has to pass seven stools to get to the room for drying things.", + "E": "The living room has to pass three stools to get to the room for drying things." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the living room?", + "answer": "There are five chairs in the living room.", + "question_chinese": "客厅里有几把椅子？", + "answer_chinese": "客厅里有五把椅子。", + "evidence": [ + { + "description": "There are five chairs in the living room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:01:59", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three chairs.", + "B": "There are six chairs.", + "C": "There are five chairs in the living room.", + "D": "There are seven chairs in the living room, not five as you might recall.", + "E": "There are eight chairs in the living room, spread around the seating area." + }, + "answer_label": "C", + "answer_text": "There are five chairs in the living room.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_1", + "correct_answer": "There are five chairs in the living room.", + "original_options": { + "A": "There are three chairs.", + "B": "There are six chairs.", + "C": "There are five chairs in the living room.", + "D": "There are seven chairs in the living room, not five as you might recall.", + "E": "There are eight chairs in the living room, spread around the seating area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 13 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the living room?", + "answer": "There are five chairs in the living room.", + "question_chinese": "客厅里有几把椅子？", + "answer_chinese": "客厅里有五把椅子。", + "evidence": [ + { + "description": "There are five chairs in the living room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:33:39", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are five chairs in the living room.", + "B": "There are four chairs.", + "C": "There are six chairs.", + "D": "There are three chairs in the living room, fewer than what you expected to see.", + "E": "There are seven chairs in the living room, making the space look more crowded." + }, + "answer_label": "A", + "answer_text": "There are five chairs in the living room.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_2", + "correct_answer": "There are five chairs in the living room.", + "original_options": { + "A": "There are five chairs in the living room.", + "B": "There are four chairs.", + "C": "There are six chairs.", + "D": "There are three chairs in the living room, fewer than what you expected to see.", + "E": "There are seven chairs in the living room, making the space look more crowded." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_17", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I stand in front of the sink, can I reach the trash can?", + "answer": "When you stand in front of the sink, you can't reach the trash can.", + "question_chinese": "当我站在水槽前，我能够到垃圾桶吗？", + "answer_chinese": "当你站在水槽前，你不能够到垃圾桶。", + "evidence": [ + { + "description": "The trash can is on the floor behind me.", + "timestamps": { + "start": "00:10:33", + "end": "00:10:43" + } + } + ], + "question_timestamps": "00:12:19", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you stand in front of the sink, you can reach the trash can.", + "B": "When you stand in front of the sink, you can't reach the trash can." + }, + "answer_label": "B", + "answer_text": "When you stand in front of the sink, you can't reach the trash can.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_17", + "correct_answer": "When you stand in front of the sink, you can't reach the trash can.", + "original_options": { + "A": "When you stand in front of the sink, you can reach the trash can.", + "B": "When you stand in front of the sink, you can't reach the trash can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_9_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I’m standing in front of the sink, can I reach the trash can?", + "answer": "When you’re standing in front of the sink, you can’t reach the trash can. The trash can is outside the door.", + "question_chinese": "当我站在水槽前，我能够到垃圾桶吗？", + "answer_chinese": "当你站在水槽前，你够不到垃圾桶，垃圾桶在门外。", + "evidence": [ + { + "description": "The trash can is on the floor behind me.", + "timestamps": { + "start": "00:10:33", + "end": "00:10:43" + } + }, + { + "description": "The trash can is placed on the platform by the door.", + "timestamps": { + "start": "00:10:45", + "end": "00:11:00" + } + } + ], + "question_timestamps": "00:12:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you’re standing in front of the sink, you can reach the trash can. It’s just inside the doorway.", + "B": "When you’re standing in front of the sink, you can’t reach the trash can. The trash can is outside the door." + }, + "answer_label": "B", + "answer_text": "When you’re standing in front of the sink, you can’t reach the trash can. The trash can is outside the door.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_9_T1", + "correct_answer": "When you’re standing in front of the sink, you can’t reach the trash can. The trash can is outside the door.", + "original_options": { + "A": "When you’re standing in front of the sink, you can reach the trash can. It’s just inside the doorway.", + "B": "When you’re standing in front of the sink, you can’t reach the trash can. The trash can is outside the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 21 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_9_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I’m standing in front of the sink, can I reach the trash can?", + "answer": "When you’re standing in front of the sink, you can’t reach the trash can. The trash can is outside the door.", + "question_chinese": "当我站在水槽前，我能够到垃圾桶吗？", + "answer_chinese": "当你站在水槽前，你不能够到垃圾桶。垃圾桶在门外。", + "evidence": [ + { + "description": "The trash can is on the floor behind me.", + "timestamps": { + "start": "00:10:33", + "end": "00:10:43" + } + } + ], + "question_timestamps": "00:41:46", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you’re standing in front of the sink, you can reach the trash can. It’s right next to the sink.", + "B": "When you’re standing in front of the sink, you can’t reach the trash can. The trash can is outside the door." + }, + "answer_label": "B", + "answer_text": "When you’re standing in front of the sink, you can’t reach the trash can. The trash can is outside the door.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_9_T2", + "correct_answer": "When you’re standing in front of the sink, you can’t reach the trash can. The trash can is outside the door.", + "original_options": { + "A": "When you’re standing in front of the sink, you can reach the trash can. It’s right next to the sink.", + "B": "When you’re standing in front of the sink, you can’t reach the trash can. The trash can is outside the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 22 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_6_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "On my way from the living room to the room where things are being dried, how many stools do I pass?", + "answer": "From the living room to the room where things are being dried, you pass by three stools.", + "question_chinese": "我从客厅去到晒东西的房间要经过几张凳子？", + "answer_chinese": "你从客厅去到晒东西的房间要经过三张凳子。", + "evidence": [ + { + "description": "The living room has to pass three stools to get to the room for drying things.", + "timestamps": { + "start": "00:34:22", + "end": "00:34:28" + } + } + ], + "question_timestamps": "00:37:49", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You pass one stool.", + "B": "You pass two stools.", + "C": "From the living room to the room where things are being dried, you pass by three stools.", + "D": "From the living room to the drying room, you pass by two stools along the way.", + "E": "From the living room to the room where things are being dried, you pass by four stools in total." + }, + "answer_label": "C", + "answer_text": "From the living room to the room where things are being dried, you pass by three stools.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_6_T1", + "correct_answer": "From the living room to the room where things are being dried, you pass by three stools.", + "original_options": { + "A": "You pass one stool.", + "B": "You pass two stools.", + "C": "From the living room to the room where things are being dried, you pass by three stools.", + "D": "From the living room to the drying room, you pass by two stools along the way.", + "E": "From the living room to the room where things are being dried, you pass by four stools in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 19 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_6_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "On my way from the living room to the room where things are being dried, how many stools do I pass?", + "answer": "From the living room to the room where things are being dried, you pass by three stools.", + "question_chinese": "我从客厅去到晒东西的房间要经过几张凳子？", + "answer_chinese": "客厅去到晒东西的房间要经过三张凳子。", + "evidence": [ + { + "description": "The living room has to pass three stools to get to the room for drying things.", + "timestamps": { + "start": "00:34:22", + "end": "00:34:28" + } + } + ], + "question_timestamps": "00:50:58", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You pass one stool.", + "B": "You pass four stools.", + "C": "From the living room to the drying room, you pass by two stools before you arrive.", + "D": "From the living room to the room where things are being dried, you pass by three stools.", + "E": "From the living room to the room where things are being dried, you pass by five stools along the route." + }, + "answer_label": "D", + "answer_text": "From the living room to the room where things are being dried, you pass by three stools.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_6_T2", + "correct_answer": "From the living room to the room where things are being dried, you pass by three stools.", + "original_options": { + "A": "You pass one stool.", + "B": "You pass four stools.", + "C": "From the living room to the drying room, you pass by two stools before you arrive.", + "D": "From the living room to the room where things are being dried, you pass by three stools.", + "E": "From the living room to the room where things are being dried, you pass by five stools along the route." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 20 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_8", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many onions did I chop?", + "answer": "You chopped four onions.", + "question_chinese": "我切了多少个洋葱？", + "answer_chinese": "你切了四个洋葱。", + "evidence": [ + { + "description": "I chopped four onions.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:54" + } + } + ], + "question_timestamps": "00:06:09", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You chopped two onions.", + "B": "You chopped four onions.", + "C": "You chopped three onions.", + "D": "You chopped five onions, cutting each one before moving to the next.", + "E": "You chopped six onions in total before setting the pieces aside to use later." + }, + "answer_label": "B", + "answer_text": "You chopped four onions.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_8", + "correct_answer": "You chopped four onions.", + "original_options": { + "A": "You chopped two onions.", + "B": "You chopped four onions.", + "C": "You chopped three onions.", + "D": "You chopped five onions, cutting each one before moving to the next.", + "E": "You chopped six onions in total before setting the pieces aside to use later." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_9", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many onions did I chop?", + "answer": "You chopped five onions.", + "question_chinese": "我切了多少个洋葱？", + "answer_chinese": "你切了五个洋葱。", + "evidence": [ + { + "description": "I chopped four onions.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:54" + } + }, + { + "description": "I chopped a onion.", + "timestamps": { + "start": "00:31:20", + "end": "00:31:32" + } + } + ], + "question_timestamps": "00:53:52", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You chopped six onions.", + "B": "You chopped seven onions.", + "C": "You chopped eight onions, finishing all of them before you cleaned up.", + "D": "You chopped five onions.", + "E": "You chopped nine onions in total, slicing them one after another on the board." + }, + "answer_label": "D", + "answer_text": "You chopped five onions.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_9", + "correct_answer": "You chopped five onions.", + "original_options": { + "A": "You chopped six onions.", + "B": "You chopped seven onions.", + "C": "You chopped eight onions, finishing all of them before you cleaned up.", + "D": "You chopped five onions.", + "E": "You chopped nine onions in total, slicing them one after another on the board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_2_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just pick up the garlic?", + "answer": "You just took the garlic from the bottom drawer of the fridge. The fridge is behind you.", + "question_chinese": "我刚刚在哪里拿的蒜头？", + "answer_chinese": "你刚刚在冰箱最下面的抽屉里拿的蒜头。冰箱在你后面。", + "evidence": [ + { + "description": "The garlic is placed in the bottom drawer of the refrigerator.", + "timestamps": { + "start": "00:02:02", + "end": "00:02:12" + } + } + ], + "question_timestamps": "00:02:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the top drawer.", + "B": "You just took the garlic from the bottom drawer of the fridge. The fridge is behind you.", + "C": "From the fridge door shelf.", + "D": "You just took the garlic from the middle drawer of the fridge, with the fridge behind you.", + "E": "You just took the garlic from the bottom drawer of the fridge, but the fridge was in front of you." + }, + "answer_label": "B", + "answer_text": "You just took the garlic from the bottom drawer of the fridge. The fridge is behind you.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_2_T2", + "correct_answer": "You just took the garlic from the bottom drawer of the fridge. The fridge is behind you.", + "original_options": { + "A": "From the top drawer.", + "B": "You just took the garlic from the bottom drawer of the fridge. The fridge is behind you.", + "C": "From the fridge door shelf.", + "D": "You just took the garlic from the middle drawer of the fridge, with the fridge behind you.", + "E": "You just took the garlic from the bottom drawer of the fridge, but the fridge was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 20 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_10", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just get the garlic?", + "answer": "The garlic you just took from the bottom drawer of the fridge.", + "question_chinese": "我刚刚在哪里拿的蒜头？", + "answer_chinese": "你刚刚在冰箱最下面的抽屉里拿的蒜头。", + "evidence": [ + { + "description": "The garlic is placed in the bottom drawer of the refrigerator.", + "timestamps": { + "start": "00:02:02", + "end": "00:02:12" + } + } + ], + "question_timestamps": "00:04:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the cupboard top shelf.", + "B": "From the stove-side counter.", + "C": "The garlic you just took from the bottom drawer of the fridge.", + "D": "The garlic you just took from the middle shelf of the fridge, not from any drawer.", + "E": "The garlic you just took from the drawer beside the sink, rather than from the fridge." + }, + "answer_label": "C", + "answer_text": "The garlic you just took from the bottom drawer of the fridge.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_10", + "correct_answer": "The garlic you just took from the bottom drawer of the fridge.", + "original_options": { + "A": "From the cupboard top shelf.", + "B": "From the stove-side counter.", + "C": "The garlic you just took from the bottom drawer of the fridge.", + "D": "The garlic you just took from the middle shelf of the fridge, not from any drawer.", + "E": "The garlic you just took from the drawer beside the sink, rather than from the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_2_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just pick up the garlic?", + "answer": "You just took the garlic from the bottom drawer of the fridge. The fridge is to your back-left.", + "question_chinese": "我刚刚在哪里拿的蒜头？", + "answer_chinese": "你刚刚在冰箱最下面的抽屉里拿的蒜头。冰箱在你的左后方。", + "evidence": [ + { + "description": "The garlic is placed in the bottom drawer of the refrigerator.", + "timestamps": { + "start": "00:02:02", + "end": "00:02:12" + } + }, + { + "description": "The fridge is on your right.", + "timestamps": { + "start": "00:07:23", + "end": "00:07:27" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the fridge’s top shelf.", + "B": "From the fridge’s bottom drawer, back-right.", + "C": "You picked up the garlic from the middle drawer of the fridge, with the fridge behind you.", + "D": "You grabbed the garlic from the bottom drawer of the fridge, but it was positioned to your front-left.", + "E": "You just took the garlic from the bottom drawer of the fridge. The fridge is to your back-left." + }, + "answer_label": "E", + "answer_text": "You just took the garlic from the bottom drawer of the fridge. The fridge is to your back-left.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_2_T1", + "correct_answer": "You just took the garlic from the bottom drawer of the fridge. The fridge is to your back-left.", + "original_options": { + "A": "From the fridge’s top shelf.", + "B": "From the fridge’s bottom drawer, back-right.", + "C": "You picked up the garlic from the middle drawer of the fridge, with the fridge behind you.", + "D": "You grabbed the garlic from the bottom drawer of the fridge, but it was positioned to your front-left.", + "E": "You just took the garlic from the bottom drawer of the fridge. The fridge is to your back-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 18 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_3", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just get the garlic?", + "answer": "The garlic you just took from the bottom drawer of the fridge.", + "question_chinese": "我刚刚在哪里拿的蒜头？", + "answer_chinese": "你刚刚在冰箱最下面的抽屉里拿的蒜头。", + "evidence": [ + { + "description": "The garlic is placed in the bottom drawer of the refrigerator.", + "timestamps": { + "start": "00:02:02", + "end": "00:02:12" + } + } + ], + "question_timestamps": "00:45:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the middle shelf of the fridge.", + "B": "From the top shelf of the pantry.", + "C": "You got the garlic off the countertop beside the stove, right where you were preparing ingredients.", + "D": "You took the garlic from a side drawer of the table, pulling it out from underneath the tabletop.", + "E": "The garlic you just took from the bottom drawer of the fridge." + }, + "answer_label": "E", + "answer_text": "The garlic you just took from the bottom drawer of the fridge.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_3", + "correct_answer": "The garlic you just took from the bottom drawer of the fridge.", + "original_options": { + "A": "From the middle shelf of the fridge.", + "B": "From the top shelf of the pantry.", + "C": "You got the garlic off the countertop beside the stove, right where you were preparing ingredients.", + "D": "You took the garlic from a side drawer of the table, pulling it out from underneath the tabletop.", + "E": "The garlic you just took from the bottom drawer of the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_4", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I find bananas?", + "answer": "You can find bananas on the stool in the room and in the refrigerator.", + "question_chinese": "我可以在哪里找到香蕉？", + "answer_chinese": "你可以在房间的凳子上和冰箱里找到香蕉。", + "evidence": [ + { + "description": "There are bananas on the stool in the room.", + "timestamps": { + "start": "00:20:37", + "end": "00:20:45" + } + }, + { + "description": "There are bananas in the fridge.", + "timestamps": { + "start": "00:21:58", + "end": "00:22:04" + } + } + ], + "question_timestamps": "00:24:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the kitchen table and counter.", + "B": "On a shelf near the microwave.", + "C": "You can find bananas on the stool in the room and in the refrigerator.", + "D": "You can find bananas in the pantry, and also on the counter near where you were working.", + "E": "You can find bananas in a basket by the front door, and another bunch inside the cupboard." + }, + "answer_label": "C", + "answer_text": "You can find bananas on the stool in the room and in the refrigerator.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_4", + "correct_answer": "You can find bananas on the stool in the room and in the refrigerator.", + "original_options": { + "A": "On the kitchen table and counter.", + "B": "On a shelf near the microwave.", + "C": "You can find bananas on the stool in the room and in the refrigerator.", + "D": "You can find bananas in the pantry, and also on the counter near where you were working.", + "E": "You can find bananas in a basket by the front door, and another bunch inside the cupboard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_3_T1", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I find bananas?", + "answer": "You can find bananas in a basket on the floor of the room to the right as you enter, and in the bottom bin of the fridge next to the table. The fridge is to your right.", + "question_chinese": "我可以在哪里找到香蕉？", + "answer_chinese": "你可以在进门右边的房间地上的筐子里和桌子旁边冰箱最底下的框里找到香蕉。冰箱在你右边。", + "evidence": [ + { + "description": "There are bananas on the stool in the room.", + "timestamps": { + "start": "00:20:42", + "end": "00:20:45" + } + }, + { + "description": "There are bananas in the fridge.", + "timestamps": { + "start": "00:21:58", + "end": "00:22:04" + } + }, + { + "description": "The bananas are in the basket on the room floor.", + "timestamps": { + "start": "00:20:35", + "end": "00:20:41" + } + }, + { + "description": "There are bananas in the bottom drawer of the refrigerator.", + "timestamps": { + "start": "00:22:25", + "end": "00:22:30" + } + } + ], + "question_timestamps": "00:27:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can find bananas in a basket on the floor of the room to the right as you enter, and in the bottom bin of the fridge next to the table. The fridge is to your right.", + "B": "Only in the floor basket.", + "C": "Only in the fridge bottom bin.", + "D": "You can find bananas in a basket on the floor to the left as you enter, and in the fridge.", + "E": "You can find bananas in the right room basket and in the fridge by the table, but the fridge is behind you." + }, + "answer_label": "A", + "answer_text": "You can find bananas in a basket on the floor of the room to the right as you enter, and in the bottom bin of the fridge next to the table. The fridge is to your right.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_3_T1", + "correct_answer": "You can find bananas in a basket on the floor of the room to the right as you enter, and in the bottom bin of the fridge next to the table. The fridge is to your right.", + "original_options": { + "A": "You can find bananas in a basket on the floor of the room to the right as you enter, and in the bottom bin of the fridge next to the table. The fridge is to your right.", + "B": "Only in the floor basket.", + "C": "Only in the fridge bottom bin.", + "D": "You can find bananas in a basket on the floor to the left as you enter, and in the fridge.", + "E": "You can find bananas in the right room basket and in the fridge by the table, but the fridge is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 22 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_3_T2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I find bananas?", + "answer": "You can find bananas on the stool in the room and inside the fridge.", + "question_chinese": "我可以在哪里找到香蕉？", + "answer_chinese": "你可以在房间的凳子上和冰箱里找到香蕉。", + "evidence": [ + { + "description": "There are bananas on the stool in the room.", + "timestamps": { + "start": "00:20:37", + "end": "00:20:45" + } + }, + { + "description": "There are bananas in the fridge.", + "timestamps": { + "start": "00:21:58", + "end": "00:22:04" + } + } + ], + "question_timestamps": "00:55:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the stool, not in the fridge.", + "B": "In the fridge only.", + "C": "You can find bananas in a basket on the floor of the room, and also in the bottom bin of the fridge.", + "D": "You can find bananas on the stool in the room and inside the fridge.", + "E": "You can find bananas on the counter in the kitchen area, and another bunch on a nearby table." + }, + "answer_label": "D", + "answer_text": "You can find bananas on the stool in the room and inside the fridge.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_3_T2", + "correct_answer": "You can find bananas on the stool in the room and inside the fridge.", + "original_options": { + "A": "On the stool, not in the fridge.", + "B": "In the fridge only.", + "C": "You can find bananas in a basket on the floor of the room, and also in the bottom bin of the fridge.", + "D": "You can find bananas on the stool in the room and inside the fridge.", + "E": "You can find bananas on the counter in the kitchen area, and another bunch on a nearby table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 4, + 22, + 18 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_1_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs can I see in the living room?", + "answer": "You saw five chairs in the living room. The chairs are around the table to your right.", + "question_chinese": "我可以看见客厅里有几把椅子？", + "answer_chinese": "你看到客厅里有五把椅子。椅子在你右边桌子的周围。", + "evidence": [ + { + "description": "There are five chairs in the living room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:01:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four chairs.", + "B": "Six chairs.", + "C": "You saw five chairs in the living room. The chairs are around the table to your right.", + "D": "You saw three chairs in the living room, and they were clustered around the table to your right.", + "E": "You saw seven chairs in the living room, positioned around the table to your right as you looked over." + }, + "answer_label": "C", + "answer_text": "You saw five chairs in the living room. The chairs are around the table to your right.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_1_T1", + "correct_answer": "You saw five chairs in the living room. The chairs are around the table to your right.", + "original_options": { + "A": "Four chairs.", + "B": "Six chairs.", + "C": "You saw five chairs in the living room. The chairs are around the table to your right.", + "D": "You saw three chairs in the living room, and they were clustered around the table to your right.", + "E": "You saw seven chairs in the living room, positioned around the table to your right as you looked over." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 19 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_1_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs can I see in the living room?", + "answer": "There are five chairs in the living room.", + "question_chinese": "我可以看见客厅里有几把椅子？", + "answer_chinese": "客厅里有五把椅子。", + "evidence": [ + { + "description": "There are five chairs in the living room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:33:39", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four chairs.", + "B": "Six chairs.", + "C": "There are three chairs in the living room, arranged together near the main table area.", + "D": "There are five chairs in the living room.", + "E": "There are seven chairs in the living room, spread around the seating area and table." + }, + "answer_label": "D", + "answer_text": "There are five chairs in the living room.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_1_T2", + "correct_answer": "There are five chairs in the living room.", + "original_options": { + "A": "Four chairs.", + "B": "Six chairs.", + "C": "There are three chairs in the living room, arranged together near the main table area.", + "D": "There are five chairs in the living room.", + "E": "There are seven chairs in the living room, spread around the seating area and table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_15", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are on the stove?", + "answer": "There are three pots on the stove.", + "question_chinese": "灶台上有几个锅？", + "answer_chinese": "灶台上有三个锅。", + "evidence": [ + { + "description": "There are three pots on the stove.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:07:31", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four pots.", + "B": "Two pots.", + "C": "There are five pots on the stove, with multiple burners occupied at the same time.", + "D": "There are six pots on the stove, lined up across the burners as you face the cooktop.", + "E": "There are three pots on the stove." + }, + "answer_label": "E", + "answer_text": "There are three pots on the stove.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_15", + "correct_answer": "There are three pots on the stove.", + "original_options": { + "A": "Four pots.", + "B": "Two pots.", + "C": "There are five pots on the stove, with multiple burners occupied at the same time.", + "D": "There are six pots on the stove, lined up across the burners as you face the cooktop.", + "E": "There are three pots on the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 17 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_16", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are on the stove?", + "answer": "There are four pots on the stove.", + "question_chinese": "灶台上有几个锅？", + "answer_chinese": "灶台上有四个锅。", + "evidence": [ + { + "description": "There are four pots on the stove.", + "timestamps": { + "start": "00:42:35", + "end": "00:42:39" + } + } + ], + "question_timestamps": "00:42:44", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three pots on the stove.", + "B": "There are five pots on the stove.", + "C": "There are four pots on the stove.", + "D": "There are two pots on the stove, one on the front burner and one on the back burner.", + "E": "There are six pots on the stove, spread across the front and back burners." + }, + "answer_label": "C", + "answer_text": "There are four pots on the stove.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_16", + "correct_answer": "There are four pots on the stove.", + "original_options": { + "A": "There are three pots on the stove.", + "B": "There are five pots on the stove.", + "C": "There are four pots on the stove.", + "D": "There are two pots on the stove, one on the front burner and one on the back burner.", + "E": "There are six pots on the stove, spread across the front and back burners." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 14 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_7_T1", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the doorway to my left or to my right?", + "answer": "The doorway is to your right.", + "question_chinese": "门口在我的左/右？", + "answer_chinese": "门口在你的右边。", + "evidence": [ + { + "description": "The door is to the right of the me.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:02:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The doorway is to your left.", + "B": "The doorway is straight ahead.", + "C": "The doorway is directly behind you, so you would need to turn around to face it.", + "D": "The doorway is to your right.", + "E": "The doorway is in front of you but slightly off to the left as you look forward." + }, + "answer_label": "D", + "answer_text": "The doorway is to your right.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_7_T1", + "correct_answer": "The doorway is to your right.", + "original_options": { + "A": "The doorway is to your left.", + "B": "The doorway is straight ahead.", + "C": "The doorway is directly behind you, so you would need to turn around to face it.", + "D": "The doorway is to your right.", + "E": "The doorway is in front of you but slightly off to the left as you look forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "7e3b7dff-bf38-4352-8adc-43799beb556f", + "video_length": "01:03:17", + "qid": "7e3b7dff-bf38-4352-8adc-43799beb556f_7_T2", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is the doorway to my left or to my right?", + "answer": "The doorway is to your left.", + "question_chinese": "门口在我的左/右？", + "answer_chinese": "门口在你的左边。", + "evidence": [ + { + "description": "The door is to the left of the fridge.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:23" + } + }, + { + "description": "The door is on your left.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:38" + } + } + ], + "question_timestamps": "00:38:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The doorway is to your right.", + "B": "The doorway is straight ahead.", + "C": "The doorway is directly behind you, so it isn’t on either side right now.", + "D": "The doorway is in front of you but slightly off to the right from your viewpoint.", + "E": "The doorway is to your left." + }, + "answer_label": "E", + "answer_text": "The doorway is to your left.", + "q_id": "7e3b7dff-bf38-4352-8adc-43799beb556f_7_T2", + "correct_answer": "The doorway is to your left.", + "original_options": { + "A": "The doorway is to your right.", + "B": "The doorway is straight ahead.", + "C": "The doorway is directly behind you, so it isn’t on either side right now.", + "D": "The doorway is in front of you but slightly off to the right from your viewpoint.", + "E": "The doorway is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 16 + ] + } +] diff --git a/QAs/7e512589-aa97-41e8-83d3-af2e83e4fd06-mcq.json b/QAs/7e512589-aa97-41e8-83d3-af2e83e4fd06-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f92bce8c5a06e2b498c81f89a43831782a12e15b --- /dev/null +++ b/QAs/7e512589-aa97-41e8-83d3-af2e83e4fd06-mcq.json @@ -0,0 +1,588 @@ +[ + { + "video_uid": "7e512589-aa97-41e8-83d3-af2e83e4fd06", + "video_length": "00:58:27", + "qid": "7e512589-aa97-41e8-83d3-af2e83e4fd06_14", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the tape measure initially placed?", + "answer": "The tape measure was placed on the stone floor in front of the woman's initial position.", + "question_chinese": "卷尺最开始被放在哪里了？", + "answer_chinese": "卷尺被放在了女人初始位置的左前方的石质地板上。", + "evidence": [ + { + "description": "The tape measure was placed on the stone floor in front of the woman's initial position.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:14:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the stone floor behind her.", + "B": "On the stone floor to her left.", + "C": "The tape measure was placed on the stone floor behind the woman's initial position, not in front of her.", + "D": "The tape measure was set on the stone floor beside the woman's initial position, slightly off to the right.", + "E": "The tape measure was placed on the stone floor in front of the woman's initial position." + }, + "answer_label": "E", + "answer_text": "The tape measure was placed on the stone floor in front of the woman's initial position.", + "q_id": "7e512589-aa97-41e8-83d3-af2e83e4fd06_14", + "correct_answer": "The tape measure was placed on the stone floor in front of the woman's initial position.", + "original_options": { + "A": "On the stone floor behind her.", + "B": "On the stone floor to her left.", + "C": "The tape measure was placed on the stone floor behind the woman's initial position, not in front of her.", + "D": "The tape measure was set on the stone floor beside the woman's initial position, slightly off to the right.", + "E": "The tape measure was placed on the stone floor in front of the woman's initial position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 19 + ] + }, + { + "video_uid": "7e512589-aa97-41e8-83d3-af2e83e4fd06", + "video_length": "00:58:27", + "qid": "7e512589-aa97-41e8-83d3-af2e83e4fd06_13", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the tape measure initially placed?", + "answer": "The tape measure was placed on the stone floor in front of the woman's initial position.", + "question_chinese": "卷尺最开始被放在哪里了？", + "answer_chinese": "卷尺被放在了女人初始位置的左前方的石质地板上。", + "evidence": [ + { + "description": "The tape measure was placed on the stone floor in front of the woman's initial position.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:43:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The tape measure was placed on the stone floor in front of the woman's initial position.", + "B": "On the stone floor behind her.", + "C": "On the stone floor beside her.", + "D": "The tape measure was placed on the stone floor behind the woman's initial position rather than in front of it.", + "E": "The tape measure was placed on the stone floor just to the right of the woman's initial position, not ahead." + }, + "answer_label": "A", + "answer_text": "The tape measure was placed on the stone floor in front of the woman's initial position.", + "q_id": "7e512589-aa97-41e8-83d3-af2e83e4fd06_13", + "correct_answer": "The tape measure was placed on the stone floor in front of the woman's initial position.", + "original_options": { + "A": "The tape measure was placed on the stone floor in front of the woman's initial position.", + "B": "On the stone floor behind her.", + "C": "On the stone floor beside her.", + "D": "The tape measure was placed on the stone floor behind the woman's initial position rather than in front of it.", + "E": "The tape measure was placed on the stone floor just to the right of the woman's initial position, not ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 20 + ] + }, + { + "video_uid": "7e512589-aa97-41e8-83d3-af2e83e4fd06", + "video_length": "00:58:27", + "qid": "7e512589-aa97-41e8-83d3-af2e83e4fd06_8", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the woman now?", + "answer": "A woman is in front of me to the left.", + "question_chinese": "女人现在在我的什么方向？", + "answer_chinese": "女人在我的左前方。", + "evidence": [ + { + "description": "A woman is in front of me to the left.", + "timestamps": { + "start": "00:14:00", + "end": "00:14:17" + } + } + ], + "question_timestamps": "00:14:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front to the right.", + "B": "Behind to the left.", + "C": "A woman is directly in front of me, centered, not offset to the left side.", + "D": "A woman is behind me and slightly to the right, so she is not in my forward-left view.", + "E": "A woman is in front of me to the left." + }, + "answer_label": "E", + "answer_text": "A woman is in front of me to the left.", + "q_id": "7e512589-aa97-41e8-83d3-af2e83e4fd06_8", + "correct_answer": "A woman is in front of me to the left.", + "original_options": { + "A": "In front to the right.", + "B": "Behind to the left.", + "C": "A woman is directly in front of me, centered, not offset to the left side.", + "D": "A woman is behind me and slightly to the right, so she is not in my forward-left view.", + "E": "A woman is in front of me to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 18 + ] + }, + { + "video_uid": "7e512589-aa97-41e8-83d3-af2e83e4fd06", + "video_length": "00:58:27", + "qid": "7e512589-aa97-41e8-83d3-af2e83e4fd06_7", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the woman now?", + "answer": "A woman is in front of me to the left.", + "question_chinese": "女人现在在我的什么方向？", + "answer_chinese": "女人在我的左前方。", + "evidence": [ + { + "description": "A woman is in front of me to the left.", + "timestamps": { + "start": "00:26:56", + "end": "00:27:02" + } + } + ], + "question_timestamps": "00:27:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind me to the left.", + "B": "A woman is in front of me to the left.", + "C": "In front to the right.", + "D": "A woman is directly behind me, so she isn’t in front of my field of view at all.", + "E": "A woman is beside me on my right, not positioned in front of me to the left." + }, + "answer_label": "B", + "answer_text": "A woman is in front of me to the left.", + "q_id": "7e512589-aa97-41e8-83d3-af2e83e4fd06_7", + "correct_answer": "A woman is in front of me to the left.", + "original_options": { + "A": "Behind me to the left.", + "B": "A woman is in front of me to the left.", + "C": "In front to the right.", + "D": "A woman is directly behind me, so she isn’t in front of my field of view at all.", + "E": "A woman is beside me on my right, not positioned in front of me to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 17 + ] + }, + { + "video_uid": "7e512589-aa97-41e8-83d3-af2e83e4fd06", + "video_length": "00:58:27", + "qid": "7e512589-aa97-41e8-83d3-af2e83e4fd06_15", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times have I picked up the water cup?", + "answer": "So far, you have picked up a water cup once.", + "question_chinese": "我拿起过几次水杯？", + "answer_chinese": "到目前为止，你拿起过一次水杯。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:00:09", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:14:21", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "So far, you have picked up a water cup once.", + "C": "Three times.", + "D": "So far, you have picked up a water cup four times, with multiple separate pick-ups throughout the scene.", + "E": "So far, you have picked up a water cup six times, repeatedly lifting it and setting it back down." + }, + "answer_label": "B", + "answer_text": "So far, you have picked up a water cup once.", + "q_id": "7e512589-aa97-41e8-83d3-af2e83e4fd06_15", + "correct_answer": "So far, you have picked up a water cup once.", + "original_options": { + "A": "Twice.", + "B": "So far, you have picked up a water cup once.", + "C": "Three times.", + "D": "So far, you have picked up a water cup four times, with multiple separate pick-ups throughout the scene.", + "E": "So far, you have picked up a water cup six times, repeatedly lifting it and setting it back down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 18, + 19 + ] + }, + { + "video_uid": "7e512589-aa97-41e8-83d3-af2e83e4fd06", + "video_length": "00:58:27", + "qid": "7e512589-aa97-41e8-83d3-af2e83e4fd06_16", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times have I picked up the water cup?", + "answer": "So far, you have picked up the water cup three times.", + "question_chinese": "我拿起过几次水杯？", + "answer_chinese": "到目前为止，你拿起过三次水杯。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:00:00", + "end": "00:00:36" + } + }, + { + "description": "Second time", + "timestamps": { + "start": "00:21:20", + "end": "00:21:37" + } + }, + { + "description": "Third time", + "timestamps": { + "start": "00:26:33", + "end": "00:26:47" + } + } + ], + "question_timestamps": "00:28:21", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You picked it up twice.", + "B": "You picked it up four times.", + "C": "Up to this point, you have picked up the water cup only two times in total.", + "D": "So far, you have picked up the water cup three times.", + "E": "So far, you have picked up the water cup four times during the video sequence." + }, + "answer_label": "D", + "answer_text": "So far, you have picked up the water cup three times.", + "q_id": "7e512589-aa97-41e8-83d3-af2e83e4fd06_16", + "correct_answer": "So far, you have picked up the water cup three times.", + "original_options": { + "A": "You picked it up twice.", + "B": "You picked it up four times.", + "C": "Up to this point, you have picked up the water cup only two times in total.", + "D": "So far, you have picked up the water cup three times.", + "E": "So far, you have picked up the water cup four times during the video sequence." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "7e512589-aa97-41e8-83d3-af2e83e4fd06", + "video_length": "00:58:27", + "qid": "7e512589-aa97-41e8-83d3-af2e83e4fd06_3", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are the handcarts?", + "answer": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "question_chinese": "手推车在哪里？", + "answer_chinese": "一个手推车在皮卡车上，一个在皮卡车后面的拖车上，还有一个在拖车后面的空地上。", + "evidence": [ + { + "description": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "timestamps": { + "start": "00:23:40", + "end": "00:24:14" + } + } + ], + "question_timestamps": "00:29:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "All three are in the trailer.", + "B": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "C": "One is beside the pickup truck.", + "D": "A shopping cart is on the pickup truck, another is in front of the trailer, and the last is beside it.", + "E": "One cart sits on the trailer, one is on the ground beside the trailer, and one is behind the pickup." + }, + "answer_label": "B", + "answer_text": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "q_id": "7e512589-aa97-41e8-83d3-af2e83e4fd06_3", + "correct_answer": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "original_options": { + "A": "All three are in the trailer.", + "B": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "C": "One is beside the pickup truck.", + "D": "A shopping cart is on the pickup truck, another is in front of the trailer, and the last is beside it.", + "E": "One cart sits on the trailer, one is on the ground beside the trailer, and one is behind the pickup." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 20 + ] + }, + { + "video_uid": "7e512589-aa97-41e8-83d3-af2e83e4fd06", + "video_length": "00:58:27", + "qid": "7e512589-aa97-41e8-83d3-af2e83e4fd06_4", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are the handcarts?", + "answer": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "question_chinese": "手推车在哪里？", + "answer_chinese": "一个手推车在皮卡车上，一个在皮卡车后面的拖车上，还有一个在拖车后面的空地上。", + "evidence": [ + { + "description": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "timestamps": { + "start": "00:23:40", + "end": "00:24:14" + } + } + ], + "question_timestamps": "00:43:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "B": "All three are on the pickup truck.", + "C": "Two are on the ground.", + "D": "One cart is on the pickup truck, another is in the trailer, and the third is on the ground beside the trailer.", + "E": "A shopping cart is in the pickup bed, another is behind the truck, and the last is inside the trailer." + }, + "answer_label": "A", + "answer_text": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "q_id": "7e512589-aa97-41e8-83d3-af2e83e4fd06_4", + "correct_answer": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "original_options": { + "A": "A shopping cart is on the pickup truck, one is in the trailer behind the pickup truck, and another is on the empty ground behind the trailer.", + "B": "All three are on the pickup truck.", + "C": "Two are on the ground.", + "D": "One cart is on the pickup truck, another is in the trailer, and the third is on the ground beside the trailer.", + "E": "A shopping cart is in the pickup bed, another is behind the truck, and the last is inside the trailer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 22, + 20 + ] + }, + { + "video_uid": "7e512589-aa97-41e8-83d3-af2e83e4fd06", + "video_length": "00:58:27", + "qid": "7e512589-aa97-41e8-83d3-af2e83e4fd06_5", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What containers are used for white powder?", + "answer": "There are white buckets, red buckets, black buckets, and a hand truck.", + "question_chinese": "装着白色粉末的容器有哪些？", + "answer_chinese": "有白桶、红桶、黑桶和手推车。", + "evidence": [ + { + "description": "Black bucket", + "timestamps": { + "start": "00:00:10", + "end": "00:00:18" + } + }, + { + "description": "Red bucket", + "timestamps": { + "start": "00:00:50", + "end": "00:01:11" + } + }, + { + "description": "Handcart", + "timestamps": { + "start": "00:23:44", + "end": "00:23:57" + } + }, + { + "description": "White bucket", + "timestamps": { + "start": "00:21:44", + "end": "00:21:47" + } + } + ], + "question_timestamps": "00:43:10", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Only white buckets.", + "B": "Red and black buckets.", + "C": "The white powder is kept in white, red, and black buckets, along with a flatbed cart nearby.", + "D": "You use white, red, and black buckets for the powder, plus a platform cart instead of a hand truck.", + "E": "There are white buckets, red buckets, black buckets, and a hand truck." + }, + "answer_label": "E", + "answer_text": "There are white buckets, red buckets, black buckets, and a hand truck.", + "q_id": "7e512589-aa97-41e8-83d3-af2e83e4fd06_5", + "correct_answer": "There are white buckets, red buckets, black buckets, and a hand truck.", + "original_options": { + "A": "Only white buckets.", + "B": "Red and black buckets.", + "C": "The white powder is kept in white, red, and black buckets, along with a flatbed cart nearby.", + "D": "You use white, red, and black buckets for the powder, plus a platform cart instead of a hand truck.", + "E": "There are white buckets, red buckets, black buckets, and a hand truck." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 17, + 19 + ] + }, + { + "video_uid": "7e512589-aa97-41e8-83d3-af2e83e4fd06", + "video_length": "00:58:27", + "qid": "7e512589-aa97-41e8-83d3-af2e83e4fd06_6", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What containers are used for white powder?", + "answer": "There are white buckets, red buckets, black buckets, and a hand truck.", + "question_chinese": "装着白色粉末的容器有哪些？", + "answer_chinese": "有白桶、红桶、黑桶和手推车。", + "evidence": [ + { + "description": "Black bucket", + "timestamps": { + "start": "00:00:10", + "end": "00:00:18" + } + }, + { + "description": "Red bucket", + "timestamps": { + "start": "00:00:50", + "end": "00:01:11" + } + }, + { + "description": "Handcart", + "timestamps": { + "start": "00:23:44", + "end": "00:23:57" + } + }, + { + "description": "White bucket", + "timestamps": { + "start": "00:21:44", + "end": "00:21:47" + } + } + ], + "question_timestamps": "00:58:27", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "There are white buckets, red buckets, black buckets, and a hand truck.", + "B": "White, red, and black bins.", + "C": "White bags and a trolley.", + "D": "The powder is stored in white, red, and black bags, and you move them with a hand trolley.", + "E": "You have white, red, and black buckets for the powder, plus a flatbed cart rather than a hand truck." + }, + "answer_label": "A", + "answer_text": "There are white buckets, red buckets, black buckets, and a hand truck.", + "q_id": "7e512589-aa97-41e8-83d3-af2e83e4fd06_6", + "correct_answer": "There are white buckets, red buckets, black buckets, and a hand truck.", + "original_options": { + "A": "There are white buckets, red buckets, black buckets, and a hand truck.", + "B": "White, red, and black bins.", + "C": "White bags and a trolley.", + "D": "The powder is stored in white, red, and black bags, and you move them with a hand trolley.", + "E": "You have white, red, and black buckets for the powder, plus a flatbed cart rather than a hand truck." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 19 + ] + } +] diff --git a/QAs/7e80931a-2067-419d-8f9e-2aff41dd330c-mcq.json b/QAs/7e80931a-2067-419d-8f9e-2aff41dd330c-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1637fa85882c1be05fad4d900f6425608622b071 --- /dev/null +++ b/QAs/7e80931a-2067-419d-8f9e-2aff41dd330c-mcq.json @@ -0,0 +1,796 @@ +[ + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "01:07:58", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the microwave oven?", + "answer": "The microwave oven is above the oven.", + "question_chinese": "微波炉在哪里？", + "answer_chinese": "微波炉在烤箱上面。", + "evidence": [ + { + "description": "The microwave oven is above the oven.", + "timestamps": { + "start": "00:34:43", + "end": "00:34:45" + } + } + ], + "question_timestamps": "00:37:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The microwave is below the oven.", + "B": "The microwave is beside the refrigerator.", + "C": "The microwave oven is placed inside a cabinet to the left of the oven, not directly above it.", + "D": "The microwave oven is above the oven.", + "E": "The microwave oven is on the countertop next to the oven, rather than mounted over the oven." + }, + "answer_label": "D", + "answer_text": "The microwave oven is above the oven.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_1", + "correct_answer": "The microwave oven is above the oven.", + "original_options": { + "A": "The microwave is below the oven.", + "B": "The microwave is beside the refrigerator.", + "C": "The microwave oven is placed inside a cabinet to the left of the oven, not directly above it.", + "D": "The microwave oven is above the oven.", + "E": "The microwave oven is on the countertop next to the oven, rather than mounted over the oven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "01:07:58", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the microwave oven?", + "answer": "The microwave oven is above the oven.", + "question_chinese": "微波炉在哪里？", + "answer_chinese": "微波炉在烤箱上面。", + "evidence": [ + { + "description": "The microwave oven is above the oven.", + "timestamps": { + "start": "00:34:43", + "end": "00:34:45" + } + } + ], + "question_timestamps": "00:41:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The microwave is below the oven.", + "B": "The microwave is beside the refrigerator.", + "C": "The microwave oven is mounted to the side of the oven, not positioned directly above it.", + "D": "The microwave oven is located on the counter near the toaster, rather than installed above the oven.", + "E": "The microwave oven is above the oven." + }, + "answer_label": "E", + "answer_text": "The microwave oven is above the oven.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_2", + "correct_answer": "The microwave oven is above the oven.", + "original_options": { + "A": "The microwave is below the oven.", + "B": "The microwave is beside the refrigerator.", + "C": "The microwave oven is mounted to the side of the oven, not positioned directly above it.", + "D": "The microwave oven is located on the counter near the toaster, rather than installed above the oven.", + "E": "The microwave oven is above the oven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "00:40:48", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_13", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I need carrots for cooking. Where can I find them?", + "answer": "There are carrots in the third compartment of the refrigerator.", + "question_chinese": "我做饭需要胡罗卜，我该在哪里找到？", + "answer_chinese": "在冰箱的第三格有胡萝卜。", + "evidence": [ + { + "description": "The refrigerator.", + "timestamps": { + "start": "00:12:45", + "end": "00:12:46" + } + }, + { + "description": "There are carrots in the third compartment of the refrigerator.", + "timestamps": { + "start": "00:12:52", + "end": "00:12:54" + } + } + ], + "question_timestamps": "00:16:56", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In the second refrigerator compartment.", + "B": "In the bottom refrigerator drawer.", + "C": "You can find the carrots in the top compartment of the refrigerator, not the third one.", + "D": "The carrots are stored in the middle refrigerator compartment, rather than the third compartment.", + "E": "There are carrots in the third compartment of the refrigerator." + }, + "answer_label": "E", + "answer_text": "There are carrots in the third compartment of the refrigerator.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_13", + "correct_answer": "There are carrots in the third compartment of the refrigerator.", + "original_options": { + "A": "In the second refrigerator compartment.", + "B": "In the bottom refrigerator drawer.", + "C": "You can find the carrots in the top compartment of the refrigerator, not the third one.", + "D": "The carrots are stored in the middle refrigerator compartment, rather than the third compartment.", + "E": "There are carrots in the third compartment of the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 14 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "00:40:48", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_14", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I need carrots for cooking. Where can I find them?", + "answer": "There are carrots in the third compartment of the refrigerator.", + "question_chinese": "我做饭需要胡罗卜，我该在哪里找到？", + "answer_chinese": "在冰箱的第三格有胡萝卜。", + "evidence": [ + { + "description": "The refrigerator.", + "timestamps": { + "start": "00:12:45", + "end": "00:12:46" + } + }, + { + "description": "There are carrots in the third compartment of the refrigerator.", + "timestamps": { + "start": "00:12:52", + "end": "00:12:54" + } + } + ], + "question_timestamps": "00:18:56", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "There are carrots in the third compartment of the refrigerator.", + "B": "In the vegetable drawer.", + "C": "On the refrigerator door shelf.", + "D": "You can find the carrots in the crisper drawer at the bottom of the refrigerator.", + "E": "Look for the carrots on the middle shelf of the refrigerator, near the other produce." + }, + "answer_label": "A", + "answer_text": "There are carrots in the third compartment of the refrigerator.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_14", + "correct_answer": "There are carrots in the third compartment of the refrigerator.", + "original_options": { + "A": "There are carrots in the third compartment of the refrigerator.", + "B": "In the vegetable drawer.", + "C": "On the refrigerator door shelf.", + "D": "You can find the carrots in the crisper drawer at the bottom of the refrigerator.", + "E": "Look for the carrots on the middle shelf of the refrigerator, near the other produce." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 15 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "01:07:58", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_2_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Just now, was I closer to the bed or closer to the wardrobe?", + "answer": "You are closer to the wardrobe, and the wardrobe is next to you.", + "question_chinese": "我刚才离床更近还是离衣柜更近？", + "answer_chinese": "你离衣柜更近，衣柜在你的旁边。", + "evidence": [ + { + "description": "You take the clothes to the washroom and walk past the wardrobe.", + "timestamps": { + "start": "00:58:50", + "end": "00:58:57" + } + } + ], + "question_timestamps": "00:58:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are closer to the wardrobe, and the wardrobe is next to you.", + "B": "Closer to the bed.", + "C": "About the same distance.", + "D": "You are closer to the bed, and the bed is right next to you.", + "E": "You are closer to the wardrobe, but it is across the room from you." + }, + "answer_label": "A", + "answer_text": "You are closer to the wardrobe, and the wardrobe is next to you.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_2_T2", + "correct_answer": "You are closer to the wardrobe, and the wardrobe is next to you.", + "original_options": { + "A": "You are closer to the wardrobe, and the wardrobe is next to you.", + "B": "Closer to the bed.", + "C": "About the same distance.", + "D": "You are closer to the bed, and the bed is right next to you.", + "E": "You are closer to the wardrobe, but it is across the room from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "01:07:58", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_2_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Just now, was I closer to the bed or closer to the wardrobe?", + "answer": "You are closer to the bed, and the bed is in front of you.", + "question_chinese": "我刚才离床更近还是离衣柜更近？", + "answer_chinese": "你离床更近，床在你的前面。", + "evidence": [ + { + "description": "You are folding clothes beside the bed.", + "timestamps": { + "start": "00:59:00", + "end": "00:59:07" + } + } + ], + "question_timestamps": "00:59:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Closer to the wardrobe.", + "B": "About the same distance.", + "C": "You are closer to the bed, but the bed is behind you rather than in front.", + "D": "You are closer to the wardrobe, and the wardrobe is in front of you, not the bed.", + "E": "You are closer to the bed, and the bed is in front of you." + }, + "answer_label": "E", + "answer_text": "You are closer to the bed, and the bed is in front of you.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_2_T1", + "correct_answer": "You are closer to the bed, and the bed is in front of you.", + "original_options": { + "A": "Closer to the wardrobe.", + "B": "About the same distance.", + "C": "You are closer to the bed, but the bed is behind you rather than in front.", + "D": "You are closer to the wardrobe, and the wardrobe is in front of you, not the bed.", + "E": "You are closer to the bed, and the bed is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "00:40:48", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_15", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many things did I add to the bowl in total?", + "answer": "You added a total of three things to the bowl.", + "question_chinese": "我在碗里一共加了几种东西？", + "answer_chinese": "你在碗里一共加了三种东西。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:14:07", + "end": "00:14:17" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:14:24", + "end": "00:14:26" + } + }, + { + "description": "The third one.", + "timestamps": { + "start": "00:16:12", + "end": "00:16:19" + } + } + ], + "question_timestamps": "00:19:05", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two things.", + "B": "Four things.", + "C": "You added a total of five things to the bowl over the whole sequence.", + "D": "You added a total of three things to the bowl.", + "E": "You added a total of six things to the bowl altogether by the end." + }, + "answer_label": "D", + "answer_text": "You added a total of three things to the bowl.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_15", + "correct_answer": "You added a total of three things to the bowl.", + "original_options": { + "A": "Two things.", + "B": "Four things.", + "C": "You added a total of five things to the bowl over the whole sequence.", + "D": "You added a total of three things to the bowl.", + "E": "You added a total of six things to the bowl altogether by the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "00:40:48", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_16", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many things did I add to the bowl in total?", + "answer": "You added a total of three things to the bowl.", + "question_chinese": "我在碗里一共加了几种东西？", + "answer_chinese": "你在碗里一共加了三种东西。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:14:07", + "end": "00:14:17" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:14:24", + "end": "00:14:26" + } + }, + { + "description": "The third one.", + "timestamps": { + "start": "00:16:12", + "end": "00:16:19" + } + } + ], + "question_timestamps": "00:20:55", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four things.", + "B": "Five things.", + "C": "You added a total of three things to the bowl.", + "D": "You added a total of six things to the bowl across all the additions you made.", + "E": "You added a total of seven things to the bowl altogether by the end of the task." + }, + "answer_label": "C", + "answer_text": "You added a total of three things to the bowl.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_16", + "correct_answer": "You added a total of three things to the bowl.", + "original_options": { + "A": "Four things.", + "B": "Five things.", + "C": "You added a total of three things to the bowl.", + "D": "You added a total of six things to the bowl across all the additions you made.", + "E": "You added a total of seven things to the bowl altogether by the end of the task." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "00:40:48", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_11", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I add to the juicer?", + "answer": "You added the vegetables from the pot and soda crackers to the juicer.", + "question_chinese": "我往榨汁机里加入了什么？", + "answer_chinese": "你在榨汁机里加入了锅里的菜和苏打饼干。", + "evidence": [ + { + "description": "You added the vegetables from the pot.", + "timestamps": { + "start": "00:10:57", + "end": "00:11:10" + } + }, + { + "description": "You added the soda crackers.", + "timestamps": { + "start": "00:11:46", + "end": "00:11:50" + } + } + ], + "question_timestamps": "00:16:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Fruits and tortilla chips.", + "B": "Herbs and breadsticks.", + "C": "You added the nuts from the jar along with rice cakes to the juicer.", + "D": "You added the vegetables from the pot and soda crackers to the juicer.", + "E": "You added the grains from the cupboard and pretzels to the juicer before turning it on." + }, + "answer_label": "D", + "answer_text": "You added the vegetables from the pot and soda crackers to the juicer.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_11", + "correct_answer": "You added the vegetables from the pot and soda crackers to the juicer.", + "original_options": { + "A": "Fruits and tortilla chips.", + "B": "Herbs and breadsticks.", + "C": "You added the nuts from the jar along with rice cakes to the juicer.", + "D": "You added the vegetables from the pot and soda crackers to the juicer.", + "E": "You added the grains from the cupboard and pretzels to the juicer before turning it on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 16 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "00:40:48", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_12", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I add to the juicer?", + "answer": "You added the vegetables from the pot and soda crackers to the juicer.", + "question_chinese": "我往榨汁机里加入了什么？", + "answer_chinese": "你在榨汁机里加入了锅里的菜和苏打饼干。", + "evidence": [ + { + "description": "You added the vegetables from the pot.", + "timestamps": { + "start": "00:10:57", + "end": "00:11:10" + } + }, + { + "description": "You added the soda crackers.", + "timestamps": { + "start": "00:11:46", + "end": "00:11:50" + } + } + ], + "question_timestamps": "00:23:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Fruits and potato chips.", + "B": "You added the vegetables from the pot and soda crackers to the juicer.", + "C": "Greens and rice cakes.", + "D": "You added the vegetables from the bowl and salsa to the juicer.", + "E": "You added the veggies from the counter and breadsticks to the juicer instead." + }, + "answer_label": "B", + "answer_text": "You added the vegetables from the pot and soda crackers to the juicer.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_12", + "correct_answer": "You added the vegetables from the pot and soda crackers to the juicer.", + "original_options": { + "A": "Fruits and potato chips.", + "B": "You added the vegetables from the pot and soda crackers to the juicer.", + "C": "Greens and rice cakes.", + "D": "You added the vegetables from the bowl and salsa to the juicer.", + "E": "You added the veggies from the counter and breadsticks to the juicer instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "00:40:48", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_10_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Can I directly reach the faucet?", + "answer": "You can. The faucet is in front of you, very close to you.", + "question_chinese": "我能直接接触到水龙头吗？", + "answer_chinese": "你可以，水龙头在你的前面，距离你很近。", + "evidence": [ + { + "description": "You are cooking.", + "timestamps": { + "start": "00:12:35", + "end": "00:16:18" + } + } + ], + "question_timestamps": "00:40:49", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can. The faucet is in front of you, very close to you.", + "B": "You can’t. The faucet is out of reach, farther away from you." + }, + "answer_label": "A", + "answer_text": "You can. The faucet is in front of you, very close to you.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_10_T1", + "correct_answer": "You can. The faucet is in front of you, very close to you.", + "original_options": { + "A": "You can. The faucet is in front of you, very close to you.", + "B": "You can’t. The faucet is out of reach, farther away from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "00:40:48", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_10_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Can I directly reach the faucet?", + "answer": "You can’t. The faucet is behind you.", + "question_chinese": "我能直接接触到水龙头吗？", + "answer_chinese": "你不能，水龙头在你的后面。", + "evidence": [ + { + "description": "The water tap is behind you, some distance away.", + "timestamps": { + "start": "00:30:22", + "end": "00:30:28" + } + } + ], + "question_timestamps": "00:40:53", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can. The faucet is in front of you.", + "B": "You can’t. The faucet is behind you." + }, + "answer_label": "B", + "answer_text": "You can’t. The faucet is behind you.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_10_T2", + "correct_answer": "You can’t. The faucet is behind you.", + "original_options": { + "A": "You can. The faucet is in front of you.", + "B": "You can’t. The faucet is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "01:07:58", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_7", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the cooked beans in the pot?", + "answer": "You put them in porcelain bowls and plastic boxes.", + "question_chinese": "锅里的煮的豆子被我装到了哪里？", + "answer_chinese": "被你装进了瓷碗和塑料盒里。", + "evidence": [ + { + "description": "You put them in plastic boxes.", + "timestamps": { + "start": "00:17:14", + "end": "00:17:40" + } + }, + { + "description": "You put them in porcelain bowls.", + "timestamps": { + "start": "00:16:05", + "end": "00:16:28" + } + } + ], + "question_timestamps": "00:22:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In metal containers and glass jars.", + "B": "You put them in porcelain bowls and plastic boxes.", + "C": "In glass bowls and metal trays.", + "D": "You put them into glass jars first, then transferred the rest into metal containers to store them.", + "E": "You divided the beans between glass bowls and a couple of metal trays, spreading them out to cool." + }, + "answer_label": "B", + "answer_text": "You put them in porcelain bowls and plastic boxes.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_7", + "correct_answer": "You put them in porcelain bowls and plastic boxes.", + "original_options": { + "A": "In metal containers and glass jars.", + "B": "You put them in porcelain bowls and plastic boxes.", + "C": "In glass bowls and metal trays.", + "D": "You put them into glass jars first, then transferred the rest into metal containers to store them.", + "E": "You divided the beans between glass bowls and a couple of metal trays, spreading them out to cool." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "7e80931a-2067-419d-8f9e-2aff41dd330c", + "video_length": "01:07:58", + "qid": "7e80931a-2067-419d-8f9e-2aff41dd330c_8", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the cooked beans in the pot?", + "answer": "You put them in porcelain bowls and plastic boxes.", + "question_chinese": "锅里的煮的豆子被我装到了哪里？", + "answer_chinese": "被你装进了瓷碗和塑料盒里。", + "evidence": [ + { + "description": "You put them in plastic boxes.", + "timestamps": { + "start": "00:17:14", + "end": "00:17:40" + } + }, + { + "description": "You put them in porcelain bowls.", + "timestamps": { + "start": "00:16:05", + "end": "00:16:28" + } + } + ], + "question_timestamps": "00:41:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In glass jars and metal containers.", + "B": "In paper bags and ceramic dishes.", + "C": "You poured the beans into plastic cups and then packed the remaining beans into wooden boxes for storage.", + "D": "You put them in porcelain bowls and plastic boxes.", + "E": "You spread the beans across aluminum trays and a few glass bowls, separating portions to cool down." + }, + "answer_label": "D", + "answer_text": "You put them in porcelain bowls and plastic boxes.", + "q_id": "7e80931a-2067-419d-8f9e-2aff41dd330c_8", + "correct_answer": "You put them in porcelain bowls and plastic boxes.", + "original_options": { + "A": "In glass jars and metal containers.", + "B": "In paper bags and ceramic dishes.", + "C": "You poured the beans into plastic cups and then packed the remaining beans into wooden boxes for storage.", + "D": "You put them in porcelain bowls and plastic boxes.", + "E": "You spread the beans across aluminum trays and a few glass bowls, separating portions to cool down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + } +] diff --git a/QAs/7ea94f18-3b89-4c8e-aa91-685a76fc4e01-mcq.json b/QAs/7ea94f18-3b89-4c8e-aa91-685a76fc4e01-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..59be9cfa9e401e2d938cb9db07f679a580cee2c8 --- /dev/null +++ b/QAs/7ea94f18-3b89-4c8e-aa91-685a76fc4e01-mcq.json @@ -0,0 +1,1572 @@ +[ + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_3_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the refrigerator from me right now?", + "answer": "The refrigerator is on the left side of you.", + "question_chinese": "冰箱现在我的哪个方向？", + "answer_chinese": "冰箱在你的左侧。", + "evidence": [ + { + "description": "You face the stove and turn left to open the refrigerator.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is on the left side of you.", + "B": "On your right.", + "C": "In front of you.", + "D": "The refrigerator is behind you, so you’d need to turn around to face it.", + "E": "The refrigerator is to your right side, roughly beside you rather than directly ahead." + }, + "answer_label": "A", + "answer_text": "The refrigerator is on the left side of you.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_3_T1", + "correct_answer": "The refrigerator is on the left side of you.", + "original_options": { + "A": "The refrigerator is on the left side of you.", + "B": "On your right.", + "C": "In front of you.", + "D": "The refrigerator is behind you, so you’d need to turn around to face it.", + "E": "The refrigerator is to your right side, roughly beside you rather than directly ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 14 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_3_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the refrigerator from me right now?", + "answer": "The refrigerator is behind you.", + "question_chinese": "冰箱现在我的哪个方向？", + "answer_chinese": "冰箱在你的身后。", + "evidence": [ + { + "description": "You face the stove and turn left to open the refrigerator.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "The refrigerator is to your right, so you’d have to look right to see it.", + "E": "The refrigerator is in front of you, close enough that you can walk straight to it." + }, + "answer_label": "A", + "answer_text": "The refrigerator is behind you.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_3_T2", + "correct_answer": "The refrigerator is behind you.", + "original_options": { + "A": "The refrigerator is behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "The refrigerator is to your right, so you’d have to look right to see it.", + "E": "The refrigerator is in front of you, close enough that you can walk straight to it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 16 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_7", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to use more plates, where should I go to get them?", + "answer": "You should go to the cupboard located on the lower right side of the sink to get it.", + "question_chinese": "如果我想用更多的盘子，我应该去哪里拿？", + "answer_chinese": "你应该去水槽右下方的碗柜拿。", + "evidence": [ + { + "description": "There are many plates in the cupboard located on the lower right side of the sink.", + "timestamps": { + "start": "00:02:11", + "end": "00:02:13" + } + } + ], + "question_timestamps": "00:02:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go to the lower-left cupboard by the sink.", + "B": "Check the upper-right cabinet above the sink.", + "C": "You should open the drawer on the lower right side of the sink to get more plates.", + "D": "You should go to the cupboard located on the lower right side of the sink to get it.", + "E": "You should go to the shelf on the upper left side of the sink and look there for plates." + }, + "answer_label": "D", + "answer_text": "You should go to the cupboard located on the lower right side of the sink to get it.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_7", + "correct_answer": "You should go to the cupboard located on the lower right side of the sink to get it.", + "original_options": { + "A": "Go to the lower-left cupboard by the sink.", + "B": "Check the upper-right cabinet above the sink.", + "C": "You should open the drawer on the lower right side of the sink to get more plates.", + "D": "You should go to the cupboard located on the lower right side of the sink to get it.", + "E": "You should go to the shelf on the upper left side of the sink and look there for plates." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 19 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_8", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to use more plates, where should I go to get them?", + "answer": "You should go to the cupboard located on the lower right side of the sink to get it.", + "question_chinese": "如果我想用更多的盘子，我应该去哪里拿？", + "answer_chinese": "你应该去水槽右下方的碗柜拿。", + "evidence": [ + { + "description": "There are many plates in the cupboard located on the lower right side of the sink.", + "timestamps": { + "start": "00:02:11", + "end": "00:02:13" + } + } + ], + "question_timestamps": "00:21:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Use the shelf above the sink.", + "B": "Go to the pantry on the lower-left.", + "C": "You should go to the cupboard located on the lower right side of the sink to get it.", + "D": "You should open the drawer located on the upper left side of the sink to get more plates.", + "E": "You should go to the cabinet on the left side of the refrigerator and get the plates from there." + }, + "answer_label": "C", + "answer_text": "You should go to the cupboard located on the lower right side of the sink to get it.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_8", + "correct_answer": "You should go to the cupboard located on the lower right side of the sink to get it.", + "original_options": { + "A": "Use the shelf above the sink.", + "B": "Go to the pantry on the lower-left.", + "C": "You should go to the cupboard located on the lower right side of the sink to get it.", + "D": "You should open the drawer located on the upper left side of the sink to get more plates.", + "E": "You should go to the cabinet on the left side of the refrigerator and get the plates from there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_4_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to use more plates, how should I go get them?", + "answer": "You should turn around and get it from the bowl cabinet at the lower-right of the sink behind you.", + "question_chinese": "如果我想用更多的盘子，我应该怎样去拿？", + "answer_chinese": "你应该向后转在你后面的去水槽右下方的碗柜拿。", + "evidence": [ + { + "description": "There are many plates in the cupboard located on the lower right side of the sink.", + "timestamps": { + "start": "00:02:11", + "end": "00:02:13" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around; it’s above you.", + "B": "You should turn around and get it from the bowl cabinet at the lower-right of the sink behind you.", + "C": "Don’t turn; it’s to your left.", + "D": "Turn around and get them from the upper-right cabinet by the sink in front of you.", + "E": "Turn around and get them from the lower-left cupboard by the sink in front of you." + }, + "answer_label": "B", + "answer_text": "You should turn around and get it from the bowl cabinet at the lower-right of the sink behind you.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_4_T1", + "correct_answer": "You should turn around and get it from the bowl cabinet at the lower-right of the sink behind you.", + "original_options": { + "A": "Turn around; it’s above you.", + "B": "You should turn around and get it from the bowl cabinet at the lower-right of the sink behind you.", + "C": "Don’t turn; it’s to your left.", + "D": "Turn around and get them from the upper-right cabinet by the sink in front of you.", + "E": "Turn around and get them from the lower-left cupboard by the sink in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 16, + 16 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_4_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to use more plates, how should I go get them?", + "answer": "You should squat down and get it from the bowl cabinet at the lower-right of the sink below you.", + "question_chinese": "如果我想用更多的盘子，我应该怎样去拿？", + "answer_chinese": "你应该蹲下，在你下面的水槽右下方的碗柜拿。", + "evidence": [ + { + "description": "There are many plates in the cupboard located on the lower right side of the sink.", + "timestamps": { + "start": "00:02:11", + "end": "00:02:13" + } + } + ], + "question_timestamps": "00:7:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Reach up to the upper cabinet.", + "B": "You should squat down and get it from the bowl cabinet at the lower-right of the sink below you.", + "C": "Open the cabinet left of the sink.", + "D": "Bend down and take plates from the lower-left cabinet beside the sink, not the one on the right.", + "E": "Crouch and grab the plates from the cabinet under the sink in front of you, rather than the lower-right bowl cabinet." + }, + "answer_label": "B", + "answer_text": "You should squat down and get it from the bowl cabinet at the lower-right of the sink below you.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_4_T2", + "correct_answer": "You should squat down and get it from the bowl cabinet at the lower-right of the sink below you.", + "original_options": { + "A": "Reach up to the upper cabinet.", + "B": "You should squat down and get it from the bowl cabinet at the lower-right of the sink below you.", + "C": "Open the cabinet left of the sink.", + "D": "Bend down and take plates from the lower-left cabinet beside the sink, not the one on the right.", + "E": "Crouch and grab the plates from the cabinet under the sink in front of you, rather than the lower-right bowl cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 21 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many boxes of apricots were there in the front pocket of the bag before you started digging through it, and how many boxes are left now?", + "answer": "Initially, there were two boxes of apricots, and now there are no boxes left.", + "question_chinese": "开始掏书包前袋子有多少盒杏子,现在还有几盒?", + "answer_chinese": "开始时有两盒杏子,现在还剩零盒。", + "evidence": [ + { + "description": "You took out a box of apricots.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:21" + } + }, + { + "description": "You took out the second box of apricots.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:28" + } + }, + { + "description": "You took out an empty bag.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One box before, one box now.", + "B": "Initially, there were two boxes of apricots, and now there are no boxes left.", + "C": "Two boxes before, one box now.", + "D": "Initially there were three boxes of apricots in the front pocket, and after digging there was one left.", + "E": "At the start you had one box of apricots in the front pocket, and after digging through it none remained." + }, + "answer_label": "B", + "answer_text": "Initially, there were two boxes of apricots, and now there are no boxes left.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_3", + "correct_answer": "Initially, there were two boxes of apricots, and now there are no boxes left.", + "original_options": { + "A": "One box before, one box now.", + "B": "Initially, there were two boxes of apricots, and now there are no boxes left.", + "C": "Two boxes before, one box now.", + "D": "Initially there were three boxes of apricots in the front pocket, and after digging there was one left.", + "E": "At the start you had one box of apricots in the front pocket, and after digging through it none remained." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 20 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many boxes of apricots were there in the front pocket of the bag before you started digging through it, and how many boxes are left now?", + "answer": "Initially, there were two boxes of apricots, and now there are no boxes left.", + "question_chinese": "开始掏书包前袋子有多少盒杏子,现在还有几盒?", + "answer_chinese": "开始时有两盒杏子,现在还剩零盒。", + "evidence": [ + { + "description": "You took out a box of apricots.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:21" + } + }, + { + "description": "You took out the second box of apricots.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:28" + } + }, + { + "description": "You took out an empty bag.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:12:35", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Initially, there were two boxes of apricots, and now there are no boxes left.", + "B": "One box before, one box now.", + "C": "Two boxes before, one box now.", + "D": "Initially there were three boxes of apricots in the front pocket, and after digging there was one left.", + "E": "At the start you had one box of apricots in the front pocket, and after digging through it none remained." + }, + "answer_label": "A", + "answer_text": "Initially, there were two boxes of apricots, and now there are no boxes left.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_4", + "correct_answer": "Initially, there were two boxes of apricots, and now there are no boxes left.", + "original_options": { + "A": "Initially, there were two boxes of apricots, and now there are no boxes left.", + "B": "One box before, one box now.", + "C": "Two boxes before, one box now.", + "D": "Initially there were three boxes of apricots in the front pocket, and after digging there was one left.", + "E": "At the start you had one box of apricots in the front pocket, and after digging through it none remained." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 20 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand at the edge of the table, which is closer to me, the computer or the phone?", + "answer": "The computer is closer to you.", + "question_chinese": "当我站在桌子边缘的时候,电脑和手机哪个离我更近?", + "answer_chinese": "电脑离你更近。", + "evidence": [ + { + "description": "The computer is closer to you, about 10cm away, and the phone is to your right front, about 30cm away.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The phone is closer to you.", + "B": "The computer is closer to you." + }, + "answer_label": "B", + "answer_text": "The computer is closer to you.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_1", + "correct_answer": "The computer is closer to you.", + "original_options": { + "A": "The phone is closer to you.", + "B": "The computer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand at the edge of the table, which is closer to me, the computer or the phone?", + "answer": "The phone is closer to you.", + "question_chinese": "当我站在桌子边缘的时候,电脑和手机哪个离我更近?", + "answer_chinese": "手机离你更近。", + "evidence": [ + { + "description": "The phone is closer to you, about 30cm away, while the computer is on your right, about 3m away.", + "timestamps": { + "start": "00:42:36", + "end": "00:42:43" + } + } + ], + "question_timestamps": "00:42:46", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The phone is closer to you.", + "B": "The computer is closer to you." + }, + "answer_label": "A", + "answer_text": "The phone is closer to you.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_2", + "correct_answer": "The phone is closer to you.", + "original_options": { + "A": "The phone is closer to you.", + "B": "The computer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_5_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I add leafy greens into the pot in total?", + "answer": "In total, you put leafy greens into the pot in front of you two times.", + "question_chinese": "我一共放了几次青菜在锅里？", + "answer_chinese": "你一共放了两次青菜在你面前的锅里。", + "evidence": [ + { + "description": "You put green vegetables in the pot for the first time.", + "timestamps": { + "start": "00:06:07", + "end": "00:06:09" + } + }, + { + "description": "You put green vegetables in the pot for the second time.", + "timestamps": { + "start": "00:06:41", + "end": "00:06:43" + } + } + ], + "question_timestamps": "00:7:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero times.", + "B": "In total, you put leafy greens into the pot in front of you two times.", + "C": "One time.", + "D": "In total, you put leafy greens into the pot in front of you three times during the cooking process.", + "E": "In total, you put leafy greens into the pot in front of you four separate times, adding them repeatedly." + }, + "answer_label": "B", + "answer_text": "In total, you put leafy greens into the pot in front of you two times.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_5_T2", + "correct_answer": "In total, you put leafy greens into the pot in front of you two times.", + "original_options": { + "A": "Zero times.", + "B": "In total, you put leafy greens into the pot in front of you two times.", + "C": "One time.", + "D": "In total, you put leafy greens into the pot in front of you three times during the cooking process.", + "E": "In total, you put leafy greens into the pot in front of you four separate times, adding them repeatedly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 19, + 19 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_5_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I add leafy greens into the pot in total?", + "answer": "In total, you put leafy greens into the pot in front of you three times.", + "question_chinese": "我一共放了几次青菜在锅里？", + "answer_chinese": "你一共放了三次青菜在你面前的锅里。", + "evidence": [ + { + "description": "You put green vegetables in the pot for the first time.", + "timestamps": { + "start": "00:06:07", + "end": "00:06:09" + } + }, + { + "description": "You put green vegetables in the pot for the second time.", + "timestamps": { + "start": "00:06:41", + "end": "00:06:43" + } + }, + { + "description": "You put green vegetables in the pot for the third time.", + "timestamps": { + "start": "00:07:30", + "end": "00:07:33" + } + } + ], + "question_timestamps": "00:27:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In total, you put leafy greens into the pot in front of you three times.", + "B": "One time.", + "C": "Two times.", + "D": "In total, you put leafy greens into the pot in front of you four times across the whole sequence.", + "E": "In total, you put leafy greens into the pot in front of you five times as you kept adding more." + }, + "answer_label": "A", + "answer_text": "In total, you put leafy greens into the pot in front of you three times.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_5_T1", + "correct_answer": "In total, you put leafy greens into the pot in front of you three times.", + "original_options": { + "A": "In total, you put leafy greens into the pot in front of you three times.", + "B": "One time.", + "C": "Two times.", + "D": "In total, you put leafy greens into the pot in front of you four times across the whole sequence.", + "E": "In total, you put leafy greens into the pot in front of you five times as you kept adding more." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 19, + 20 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_17", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many different seasonings did I just put in this meat dish?", + "answer": "You added five kinds of seasonings.", + "question_chinese": "我刚刚做这个肉菜，下了多少种调料？", + "answer_chinese": "你下了五种调料。", + "evidence": [ + { + "description": "You added the first seasoning.", + "timestamps": { + "start": "00:22:20", + "end": "00:22:22" + } + }, + { + "description": "You added the second seasoning.", + "timestamps": { + "start": "00:22:29", + "end": "00:22:31" + } + }, + { + "description": "You added the third seasoning.", + "timestamps": { + "start": "00:22:35", + "end": "00:22:43" + } + }, + { + "description": "You added the fourth seasoning.", + "timestamps": { + "start": "00:22:53", + "end": "00:22:55" + } + }, + { + "description": "You added the fifth seasoning.", + "timestamps": { + "start": "00:23:00", + "end": "00:23:01" + } + } + ], + "question_timestamps": "00:23:05", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You added five kinds of seasonings.", + "B": "Four kinds.", + "C": "Six kinds.", + "D": "You added seven kinds of seasonings to the meat dish, counting each distinct seasoning you sprinkled in.", + "E": "You added eight kinds of seasonings to the meat dish, with multiple different seasonings mixed in." + }, + "answer_label": "A", + "answer_text": "You added five kinds of seasonings.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_17", + "correct_answer": "You added five kinds of seasonings.", + "original_options": { + "A": "You added five kinds of seasonings.", + "B": "Four kinds.", + "C": "Six kinds.", + "D": "You added seven kinds of seasonings to the meat dish, counting each distinct seasoning you sprinkled in.", + "E": "You added eight kinds of seasonings to the meat dish, with multiple different seasonings mixed in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 16 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_18", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many different seasonings did I just put in this meat dish?", + "answer": "You added nine kinds of seasonings.", + "question_chinese": "我刚刚做这个肉菜，下了多少种调料？", + "answer_chinese": "你下了九种调料。", + "evidence": [ + { + "description": "You added the first seasoning.", + "timestamps": { + "start": "00:22:20", + "end": "00:22:22" + } + }, + { + "description": "You added the second seasoning.", + "timestamps": { + "start": "00:22:29", + "end": "00:22:31" + } + }, + { + "description": "You added the third seasoning.", + "timestamps": { + "start": "00:22:35", + "end": "00:22:43" + } + }, + { + "description": "You added the fourth seasoning.", + "timestamps": { + "start": "00:22:53", + "end": "00:22:55" + } + }, + { + "description": "You added the fifth seasoning.", + "timestamps": { + "start": "00:23:00", + "end": "00:23:01" + } + }, + { + "description": "You added the sixth seasoning.", + "timestamps": { + "start": "00:23:09", + "end": "00:23:10" + } + }, + { + "description": "You added the seventh seasoning.", + "timestamps": { + "start": "00:23:16", + "end": "00:23:19" + } + }, + { + "description": "You added the eighth seasoning.", + "timestamps": { + "start": "00:23:29", + "end": "00:23:34" + } + }, + { + "description": "You added the ninth seasoning.", + "timestamps": { + "start": "00:23:41", + "end": "00:23:46" + } + } + ], + "question_timestamps": "00:42:36", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You added seven kinds of seasonings.", + "B": "You added nine kinds of seasonings.", + "C": "You added eleven kinds of seasonings.", + "D": "You added eight kinds of seasonings to the meat dish, then stopped seasoning it.", + "E": "You added ten kinds of seasonings overall, counting each different spice you sprinkled in." + }, + "answer_label": "B", + "answer_text": "You added nine kinds of seasonings.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_18", + "correct_answer": "You added nine kinds of seasonings.", + "original_options": { + "A": "You added seven kinds of seasonings.", + "B": "You added nine kinds of seasonings.", + "C": "You added eleven kinds of seasonings.", + "D": "You added eight kinds of seasonings to the meat dish, then stopped seasoning it.", + "E": "You added ten kinds of seasonings overall, counting each different spice you sprinkled in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_19", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following just now?", + "answer": "You was moving from the stove towards the sink.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你正从灶台边出发，朝着水槽方向移动。", + "evidence": [ + { + "description": "You turn right from the stove and take two steps to reach the sink.", + "timestamps": { + "start": "00:24:05", + "end": "00:24:07" + } + } + ], + "question_timestamps": "00:24:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You was moving from the stove towards the sink.", + "B": "From sink to stove.", + "C": "From stove to counter.", + "D": "You walked away from the stove, then turned and headed back toward where you started.", + "E": "You moved from the sink past the counter, heading in the opposite direction from the stove." + }, + "answer_label": "A", + "answer_text": "You was moving from the stove towards the sink.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_19", + "correct_answer": "You was moving from the stove towards the sink.", + "original_options": { + "A": "You was moving from the stove towards the sink.", + "B": "From sink to stove.", + "C": "From stove to counter.", + "D": "You walked away from the stove, then turned and headed back toward where you started.", + "E": "You moved from the sink past the counter, heading in the opposite direction from the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_20", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following just now?", + "answer": "You was leaving the dining table, turning left, and moving towards the refrigerator.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你正从餐桌出发，左转身，朝着冰箱方向移动。", + "evidence": [ + { + "description": "You pick up the milk from the dining table, turn left, and walk straight to the refrigerator.", + "timestamps": { + "start": "00:36:43", + "end": "00:36:45" + } + } + ], + "question_timestamps": "00:36:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Leaving the dining table, turning right.", + "B": "You was leaving the dining table, turning left, and moving towards the refrigerator.", + "C": "From dining table to counter.", + "D": "You left the dining table, turned left, and moved toward the sink instead of the refrigerator.", + "E": "You left the dining table, turned around, and headed back toward the dining area rather than the refrigerator." + }, + "answer_label": "B", + "answer_text": "You was leaving the dining table, turning left, and moving towards the refrigerator.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_20", + "correct_answer": "You was leaving the dining table, turning left, and moving towards the refrigerator.", + "original_options": { + "A": "Leaving the dining table, turning right.", + "B": "You was leaving the dining table, turning left, and moving towards the refrigerator.", + "C": "From dining table to counter.", + "D": "You left the dining table, turned left, and moved toward the sink instead of the refrigerator.", + "E": "You left the dining table, turned around, and headed back toward the dining area rather than the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 18 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_15", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cucumbers did I just cut?", + "answer": "You sliced a cucumber.", + "question_chinese": "我刚才切了几根黄瓜?", + "answer_chinese": "你切了一根黄瓜。", + "evidence": [ + { + "description": "You are slicing the first cucumber.", + "timestamps": { + "start": "00:17:56", + "end": "00:18:08" + } + } + ], + "question_timestamps": "00:18:11", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You sliced zero cucumbers.", + "B": "You sliced two cucumbers.", + "C": "You sliced three cucumbers, cutting each one into pieces before setting them aside.", + "D": "You sliced five cucumbers in total, finishing one before moving on to the next.", + "E": "You sliced a cucumber." + }, + "answer_label": "E", + "answer_text": "You sliced a cucumber.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_15", + "correct_answer": "You sliced a cucumber.", + "original_options": { + "A": "You sliced zero cucumbers.", + "B": "You sliced two cucumbers.", + "C": "You sliced three cucumbers, cutting each one into pieces before setting them aside.", + "D": "You sliced five cucumbers in total, finishing one before moving on to the next.", + "E": "You sliced a cucumber." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_8_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cucumbers did I cut just now?", + "answer": "You sliced the two cucumbers in front of you.", + "question_chinese": "我刚才切了几根黄瓜?", + "answer_chinese": "你切了你面前的两根黄瓜。", + "evidence": [ + { + "description": "You are slicing the first cucumber.", + "timestamps": { + "start": "00:17:56", + "end": "00:18:08" + } + }, + { + "description": "You are slicing the second cucumber.", + "timestamps": { + "start": "00:18:20", + "end": "00:18:34" + } + } + ], + "question_timestamps": "00:18:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You sliced one cucumber in front of you.", + "B": "You sliced three cucumbers in front of you.", + "C": "You sliced four cucumbers in front of you, working through them one by one.", + "D": "You sliced zero cucumbers in front of you, leaving them uncut where they were.", + "E": "You sliced the two cucumbers in front of you." + }, + "answer_label": "E", + "answer_text": "You sliced the two cucumbers in front of you.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_8_T2", + "correct_answer": "You sliced the two cucumbers in front of you.", + "original_options": { + "A": "You sliced one cucumber in front of you.", + "B": "You sliced three cucumbers in front of you.", + "C": "You sliced four cucumbers in front of you, working through them one by one.", + "D": "You sliced zero cucumbers in front of you, leaving them uncut where they were.", + "E": "You sliced the two cucumbers in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 14, + 14 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_8_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cucumbers did I cut just now?", + "answer": "You sliced the three cucumbers in front of you.", + "question_chinese": "我刚才切了几根黄瓜?", + "answer_chinese": "你切了你面前的三根黄瓜。", + "evidence": [ + { + "description": "You are slicing the first cucumber.", + "timestamps": { + "start": "00:17:56", + "end": "00:18:08" + } + }, + { + "description": "You are slicing the second cucumber.", + "timestamps": { + "start": "00:18:20", + "end": "00:18:34" + } + }, + { + "description": "You are slicing the third cucumber.", + "timestamps": { + "start": "00:18:42", + "end": "00:18:50" + } + } + ], + "question_timestamps": "00:20:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You sliced one cucumber in front of you.", + "B": "You sliced two cucumbers in front of you.", + "C": "You sliced four cucumbers in front of you, cutting all of them into pieces before stopping.", + "D": "You sliced the three cucumbers in front of you.", + "E": "You sliced five cucumbers in front of you, finishing the last one just before you paused." + }, + "answer_label": "D", + "answer_text": "You sliced the three cucumbers in front of you.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_8_T1", + "correct_answer": "You sliced the three cucumbers in front of you.", + "original_options": { + "A": "You sliced one cucumber in front of you.", + "B": "You sliced two cucumbers in front of you.", + "C": "You sliced four cucumbers in front of you, cutting all of them into pieces before stopping.", + "D": "You sliced the three cucumbers in front of you.", + "E": "You sliced five cucumbers in front of you, finishing the last one just before you paused." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 16 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_11", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Did I put the knife on my right side or on my left side?", + "answer": "You placed the knife on the cutting board on the right.", + "question_chinese": "我到底把刀放在我右边了，还是放在左边了？", + "answer_chinese": "你把刀放在了右边的砧板上。", + "evidence": [ + { + "description": "The knife is on the cutting board to your right.", + "timestamps": { + "start": "00:10:34", + "end": "00:10:41" + } + } + ], + "question_timestamps": "00:10:43", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You placed the knife on the cutting board on the right.", + "B": "You placed the knife on the knife rack on the left." + }, + "answer_label": "A", + "answer_text": "You placed the knife on the cutting board on the right.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_11", + "correct_answer": "You placed the knife on the cutting board on the right.", + "original_options": { + "A": "You placed the knife on the cutting board on the right.", + "B": "You placed the knife on the knife rack on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_12", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Did I put the knife on my right side or on my left side?", + "answer": "You placed the knife on the knife rack on the left.", + "question_chinese": "我到底把刀放在我右边了，还是放在左边了？", + "answer_chinese": "你把刀放在了左边的刀架上。", + "evidence": [ + { + "description": "The knife is on the knife rack to your left.", + "timestamps": { + "start": "00:20:31", + "end": "00:20:38" + } + } + ], + "question_timestamps": "00:20:43", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You placed the knife on the cutting board on the right.", + "B": "You placed the knife on the knife rack on the left." + }, + "answer_label": "B", + "answer_text": "You placed the knife on the knife rack on the left.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_12", + "correct_answer": "You placed the knife on the knife rack on the left.", + "original_options": { + "A": "You placed the knife on the cutting board on the right.", + "B": "You placed the knife on the knife rack on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_2_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Before I started emptying the front pocket of my backpack, how many boxes of apricots were in it, and how many boxes are left now?", + "answer": "The backpack in front of you still has two boxes of apricots right now.", + "question_chinese": "我开始掏书包前袋子有多少盒杏子,现在还有几盒?", + "answer_chinese": "你面前的书包现在还有两盒杏子。", + "evidence": [ + { + "description": "You took out a box of apricots.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:21" + } + }, + { + "description": "You took out the second box of apricots.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It still has one box left.", + "B": "It still has three boxes left.", + "C": "The backpack in front of you still has two boxes of apricots right now.", + "D": "The backpack in front of you still has four boxes of apricots inside right now.", + "E": "The backpack in front of you has zero boxes of apricots left in it now." + }, + "answer_label": "C", + "answer_text": "The backpack in front of you still has two boxes of apricots right now.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_2_T1", + "correct_answer": "The backpack in front of you still has two boxes of apricots right now.", + "original_options": { + "A": "It still has one box left.", + "B": "It still has three boxes left.", + "C": "The backpack in front of you still has two boxes of apricots right now.", + "D": "The backpack in front of you still has four boxes of apricots inside right now.", + "E": "The backpack in front of you has zero boxes of apricots left in it now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_2_T2", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Before I started emptying the front pocket of my backpack, how many boxes of apricots were in it, and how many boxes are left now?", + "answer": "At the beginning, the backpack in front of you had two boxes of apricots, and now there are zero boxes left.", + "question_chinese": "我开始掏书包前袋子有多少盒杏子,现在还有几盒?", + "answer_chinese": "你面前的书包开始时有两盒杏子,现在还剩零盒。", + "evidence": [ + { + "description": "You took out a box of apricots.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:21" + } + }, + { + "description": "You took out the second box of apricots.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:28" + } + }, + { + "description": "You took out an empty bag.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:12:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It started with one, now one left.", + "B": "It started with two, now one left.", + "C": "At the beginning, the backpack in front of you had three boxes of apricots, and now there is one box left.", + "D": "At the beginning, the backpack in front of you had two boxes of apricots, and now there are zero boxes left.", + "E": "At the beginning, the backpack in front of you had two boxes of apricots, and now there are two boxes left." + }, + "answer_label": "D", + "answer_text": "At the beginning, the backpack in front of you had two boxes of apricots, and now there are zero boxes left.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_2_T2", + "correct_answer": "At the beginning, the backpack in front of you had two boxes of apricots, and now there are zero boxes left.", + "original_options": { + "A": "It started with one, now one left.", + "B": "It started with two, now one left.", + "C": "At the beginning, the backpack in front of you had three boxes of apricots, and now there is one box left.", + "D": "At the beginning, the backpack in front of you had two boxes of apricots, and now there are zero boxes left.", + "E": "At the beginning, the backpack in front of you had two boxes of apricots, and now there are two boxes left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 21 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_13", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I want to use tomatoes to cook. How many tomatoes are still usable now?", + "answer": "There are still two tomatoes that are still good to use.", + "question_chinese": "我想用番茄做菜，现在还能用的番茄有多少个？", + "answer_chinese": "现在还能用的番茄还有两个。", + "evidence": [ + { + "description": "You took out the only two tomatoes left in the fridge.", + "timestamps": { + "start": "00:14:30", + "end": "00:14:34" + } + } + ], + "question_timestamps": "00:14:39", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One tomato is still usable.", + "B": "Three tomatoes are still usable.", + "C": "There are still five tomatoes that are still good to use now.", + "D": "There are still seven tomatoes that are still good to use now.", + "E": "There are still two tomatoes that are still good to use." + }, + "answer_label": "E", + "answer_text": "There are still two tomatoes that are still good to use.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_13", + "correct_answer": "There are still two tomatoes that are still good to use.", + "original_options": { + "A": "One tomato is still usable.", + "B": "Three tomatoes are still usable.", + "C": "There are still five tomatoes that are still good to use now.", + "D": "There are still seven tomatoes that are still good to use now.", + "E": "There are still two tomatoes that are still good to use." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_14", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I want to use tomatoes to cook. How many tomatoes are still usable now?", + "answer": "There are no usable tomatoes left now.", + "question_chinese": "我想用番茄做菜，现在还能用的番茄有多少个？", + "answer_chinese": "现在没有还能用的番茄了。", + "evidence": [ + { + "description": "You just used up the last two Pomodoros.", + "timestamps": { + "start": "00:19:38", + "end": "00:19:55" + } + } + ], + "question_timestamps": "00:29:55", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One usable tomato is left.", + "B": "Three usable tomatoes are left.", + "C": "There are no usable tomatoes left now.", + "D": "There are five usable tomatoes left that are still good for cooking right now.", + "E": "There are seven usable tomatoes left that are still good for cooking right now." + }, + "answer_label": "C", + "answer_text": "There are no usable tomatoes left now.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_14", + "correct_answer": "There are no usable tomatoes left now.", + "original_options": { + "A": "One usable tomato is left.", + "B": "Three usable tomatoes are left.", + "C": "There are no usable tomatoes left now.", + "D": "There are five usable tomatoes left that are still good for cooking right now.", + "E": "There are seven usable tomatoes left that are still good for cooking right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_5", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I am facing the stove now. In which direction is the refrigerator from me?", + "answer": "The refrigerator is on your left.", + "question_chinese": "我现在面向灶台，冰箱在我的哪个方向？", + "answer_chinese": "冰箱在你的左侧。", + "evidence": [ + { + "description": "You face the stove and turn left to open the refrigerator.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is on your right.", + "B": "The refrigerator is behind you.", + "C": "The refrigerator is directly in front of you when you are facing the stove.", + "D": "The refrigerator is behind your right shoulder as you face the stove.", + "E": "The refrigerator is on your left." + }, + "answer_label": "E", + "answer_text": "The refrigerator is on your left.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_5", + "correct_answer": "The refrigerator is on your left.", + "original_options": { + "A": "The refrigerator is on your right.", + "B": "The refrigerator is behind you.", + "C": "The refrigerator is directly in front of you when you are facing the stove.", + "D": "The refrigerator is behind your right shoulder as you face the stove.", + "E": "The refrigerator is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 12 + ] + }, + { + "video_uid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01", + "video_length": "00:59:21", + "qid": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_6", + "qid_num": 27, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I am facing the stove now. In which direction is the refrigerator from me?", + "answer": "The refrigerator is on your left.", + "question_chinese": "我现在面向灶台，冰箱在我的哪个方向？", + "answer_chinese": "冰箱在你的左侧。", + "evidence": [ + { + "description": "You face the stove and turn left to open the refrigerator.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is on your right.", + "B": "The refrigerator is on your left.", + "C": "The refrigerator is in front of you.", + "D": "The refrigerator is directly behind you when you are facing the stove.", + "E": "The refrigerator is slightly to your right as you look toward the stove." + }, + "answer_label": "B", + "answer_text": "The refrigerator is on your left.", + "q_id": "7ea94f18-3b89-4c8e-aa91-685a76fc4e01_6", + "correct_answer": "The refrigerator is on your left.", + "original_options": { + "A": "The refrigerator is on your right.", + "B": "The refrigerator is on your left.", + "C": "The refrigerator is in front of you.", + "D": "The refrigerator is directly behind you when you are facing the stove.", + "E": "The refrigerator is slightly to your right as you look toward the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 12, + 13 + ] + } +] diff --git a/QAs/7f09822a-87b9-4eac-bb34-3f1059c704d1-mcq.json b/QAs/7f09822a-87b9-4eac-bb34-3f1059c704d1-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..15e6c6dbc5d4f6f4dbbe1a09a1e5190fbcc16b2f --- /dev/null +++ b/QAs/7f09822a-87b9-4eac-bb34-3f1059c704d1-mcq.json @@ -0,0 +1,1502 @@ +[ + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_3_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of lumber did I take from the garage?", + "answer": "You took one piece from the garage in front of you.", + "question_chinese": "我在车库里拿了多少块木材？", + "answer_chinese": "你在前面的车库拿了一块。", + "evidence": [ + { + "description": "One.", + "timestamps": { + "start": "00:16:51", + "end": "00:17:02" + } + } + ], + "question_timestamps": "00:17:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took zero pieces from the garage.", + "B": "You took two pieces from the garage.", + "C": "You took one piece from the garage in front of you.", + "D": "You took three pieces of lumber from the garage in front of you.", + "E": "You took four pieces of lumber from the garage in front of you." + }, + "answer_label": "C", + "answer_text": "You took one piece from the garage in front of you.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_3_T2", + "correct_answer": "You took one piece from the garage in front of you.", + "original_options": { + "A": "You took zero pieces from the garage.", + "B": "You took two pieces from the garage.", + "C": "You took one piece from the garage in front of you.", + "D": "You took three pieces of lumber from the garage in front of you.", + "E": "You took four pieces of lumber from the garage in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 13 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_3_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of lumber did I take from the garage?", + "answer": "You took two pieces from the garage in front of you.", + "question_chinese": "我在车库里拿了多少块木材？", + "answer_chinese": "你在前面的车库拿了两块。", + "evidence": [ + { + "description": "One.", + "timestamps": { + "start": "00:16:51", + "end": "00:17:02" + } + } + ], + "question_timestamps": "00:17:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took one piece from the garage.", + "B": "You took two pieces from the garage in front of you.", + "C": "You took three pieces from the garage.", + "D": "You took zero pieces of lumber from the garage in front of you.", + "E": "You took four pieces of lumber from the garage in front of you." + }, + "answer_label": "B", + "answer_text": "You took two pieces from the garage in front of you.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_3_T1", + "correct_answer": "You took two pieces from the garage in front of you.", + "original_options": { + "A": "You took one piece from the garage.", + "B": "You took two pieces from the garage in front of you.", + "C": "You took three pieces from the garage.", + "D": "You took zero pieces of lumber from the garage in front of you.", + "E": "You took four pieces of lumber from the garage in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 13 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of wood did I take in the garage?", + "answer": "Two.", + "question_chinese": "我在车库里拿了多少块木材？", + "answer_chinese": "两块。", + "evidence": [ + { + "description": "One.", + "timestamps": { + "start": "00:16:51", + "end": "00:17:02" + } + }, + { + "description": "Two.", + "timestamps": { + "start": "00:17:11", + "end": "00:17:26" + } + } + ], + "question_timestamps": "00:20:38", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You took only one piece of wood while you were in the garage.", + "D": "You grabbed three separate pieces of wood while you were in the garage.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_7", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You took only one piece of wood while you were in the garage.", + "D": "You grabbed three separate pieces of wood while you were in the garage.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_8", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of wood did I take in the garage?", + "answer": "Two.", + "question_chinese": "我在车库里拿了多少块木材？", + "answer_chinese": "两块。", + "evidence": [ + { + "description": "One.", + "timestamps": { + "start": "00:16:51", + "end": "00:17:02" + } + }, + { + "description": "Two.", + "timestamps": { + "start": "00:17:11", + "end": "00:17:26" + } + } + ], + "question_timestamps": "00:36:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You took one piece of wood while you were in the garage.", + "E": "You picked up three pieces of wood from the garage before leaving." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_8", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You took one piece of wood while you were in the garage.", + "E": "You picked up three pieces of wood from the garage before leaving." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_7_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I put the wood stick I chopped down?", + "answer": "You put it next to the bicycle in the garage in front of you.", + "question_chinese": "我将砍下来的木棍放在了哪里？", + "answer_chinese": "你放在了你面前的车库里自行车的旁边。", + "evidence": [ + { + "description": "You put it in the blue box in the garage.", + "timestamps": { + "start": "00:40:02", + "end": "00:40:21" + } + } + ], + "question_timestamps": "00:55:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind the bicycle.", + "B": "By the bicycle’s left side.", + "C": "You placed it next to the bicycle in the garage, but it was behind you.", + "D": "You set it next to the bicycle in the garage, off to your right side.", + "E": "You put it next to the bicycle in the garage in front of you." + }, + "answer_label": "E", + "answer_text": "You put it next to the bicycle in the garage in front of you.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_7_T1", + "correct_answer": "You put it next to the bicycle in the garage in front of you.", + "original_options": { + "A": "Behind the bicycle.", + "B": "By the bicycle’s left side.", + "C": "You placed it next to the bicycle in the garage, but it was behind you.", + "D": "You set it next to the bicycle in the garage, off to your right side.", + "E": "You put it next to the bicycle in the garage in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 6, + 15, + 15 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_15", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I put the cut sticks?", + "answer": "You put it in the blue box in the garage.", + "question_chinese": "我将砍下来的木棍放在了哪里？", + "answer_chinese": "你放在了车库的蓝箱子里", + "evidence": [ + { + "description": "You put it in the blue box in the garage.", + "timestamps": { + "start": "00:40:02", + "end": "00:40:21" + } + } + ], + "question_timestamps": "00:59:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the red crate.", + "B": "You put it in the blue box in the garage.", + "C": "In the green box.", + "D": "You put the cut sticks into a red crate in the garage.", + "E": "You placed the cut sticks in the blue box, but it was on the porch." + }, + "answer_label": "B", + "answer_text": "You put it in the blue box in the garage.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_15", + "correct_answer": "You put it in the blue box in the garage.", + "original_options": { + "A": "In the red crate.", + "B": "You put it in the blue box in the garage.", + "C": "In the green box.", + "D": "You put the cut sticks into a red crate in the garage.", + "E": "You placed the cut sticks in the blue box, but it was on the porch." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 15 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_7_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I put the wood stick I chopped down?", + "answer": "You put it into the blue box in the garage in front of you.", + "question_chinese": "我将砍下来的木棍放在了哪里？", + "answer_chinese": "你把它放在你面前的车库里的蓝色盒子里了。", + "evidence": [ + { + "description": "You put it in the blue box in the garage.", + "timestamps": { + "start": "00:40:02", + "end": "00:40:21" + } + } + ], + "question_timestamps": "01:10:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "You put it into the blue box in the garage in front of you.", + "C": "To your left.", + "D": "You put it into the blue box in the garage, but it was to your right.", + "E": "You dropped it into the blue box in the garage, positioned behind you." + }, + "answer_label": "B", + "answer_text": "You put it into the blue box in the garage in front of you.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_7_T2", + "correct_answer": "You put it into the blue box in the garage in front of you.", + "original_options": { + "A": "Behind you.", + "B": "You put it into the blue box in the garage in front of you.", + "C": "To your left.", + "D": "You put it into the blue box in the garage, but it was to your right.", + "E": "You dropped it into the blue box in the garage, positioned behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 16, + 13 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_6_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk to the workshop’s small door?", + "answer": "The small door is directly across from you; just go around the table and you’ll reach it.", + "question_chinese": "我怎么走到工作室的小门？", + "answer_chinese": "小门就在你的对面，你绕桌子过去就会到。", + "evidence": [ + { + "description": "The position of the small door.", + "timestamps": { + "start": "00:41:14", + "end": "00:41:16" + } + } + ], + "question_timestamps": "00:51:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go straight.", + "B": "The small door is directly across from you; just go around the table and you’ll reach it.", + "C": "Turn right around the table.", + "D": "The small door is off to your left; walk straight without going around the table.", + "E": "The small door is behind you, so turn around and pass the table to reach it." + }, + "answer_label": "B", + "answer_text": "The small door is directly across from you; just go around the table and you’ll reach it.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_6_T1", + "correct_answer": "The small door is directly across from you; just go around the table and you’ll reach it.", + "original_options": { + "A": "Turn left, then go straight.", + "B": "The small door is directly across from you; just go around the table and you’ll reach it.", + "C": "Turn right around the table.", + "D": "The small door is off to your left; walk straight without going around the table.", + "E": "The small door is behind you, so turn around and pass the table to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_13", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the small door of the studio?", + "answer": "Turn left and then go straight.", + "question_chinese": "我怎么走到工作室的小门", + "answer_chinese": "左转后直走。", + "evidence": [ + { + "description": "The position of the small door.", + "timestamps": { + "start": "00:41:14", + "end": "00:41:16" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:57:54", + "end": "00:57:54" + } + } + ], + "question_timestamps": "00:57:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right and go straight.", + "B": "Turn left, then stop.", + "C": "Turn around first, then go straight to the small door of the studio.", + "D": "Go straight ahead, then turn left to reach the small door of the studio.", + "E": "Turn left and then go straight." + }, + "answer_label": "E", + "answer_text": "Turn left and then go straight.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_13", + "correct_answer": "Turn left and then go straight.", + "original_options": { + "A": "Turn right and go straight.", + "B": "Turn left, then stop.", + "C": "Turn around first, then go straight to the small door of the studio.", + "D": "Go straight ahead, then turn left to reach the small door of the studio.", + "E": "Turn left and then go straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 14 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_6_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk to the workshop’s small door?", + "answer": "Turn left, then go straight.", + "question_chinese": "我怎么走到工作室的小门？", + "answer_chinese": "左转后直走。", + "evidence": [ + { + "description": "The position of the small door.", + "timestamps": { + "start": "00:41:14", + "end": "00:41:16" + } + } + ], + "question_timestamps": "01:09:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight.", + "B": "Go straight without turning.", + "C": "The small door is directly across from you, so go around the table to reach it.", + "D": "Turn left, then go straight.", + "E": "Head to your left side first, then circle the table before going straight to the door." + }, + "answer_label": "D", + "answer_text": "Turn left, then go straight.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_6_T2", + "correct_answer": "Turn left, then go straight.", + "original_options": { + "A": "Turn right, then go straight.", + "B": "Go straight without turning.", + "C": "The small door is directly across from you, so go around the table to reach it.", + "D": "Turn left, then go straight.", + "E": "Head to your left side first, then circle the table before going straight to the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_8_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I sweep up the wood shavings on the floor?", + "answer": "You didn’t sweep.", + "question_chinese": "我是否将地上的木屑扫了？", + "answer_chinese": "你并没有扫", + "evidence": [ + { + "description": "You didn't sweep.", + "timestamps": { + "start": "00:36:12", + "end": "00:36:13" + } + } + ], + "question_timestamps": "00:36:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you swept up the wood shavings.", + "B": "You didn’t sweep." + }, + "answer_label": "B", + "answer_text": "You didn’t sweep.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_8_T1", + "correct_answer": "You didn’t sweep.", + "original_options": { + "A": "Yes—you swept up the wood shavings.", + "B": "You didn’t sweep." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_17", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I sweep the wood chips on the ground?", + "answer": "Yes.", + "question_chinese": "我是否将地上的木屑扫了？", + "answer_chinese": "是的", + "evidence": [ + { + "description": "Clean the floor.", + "timestamps": { + "start": "00:37:24", + "end": "00:37:40" + } + } + ], + "question_timestamps": "00:56:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_17", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_8_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I sweep up the wood shavings on the floor?", + "answer": "Yes.", + "question_chinese": "我是否将地上的木屑扫了？", + "answer_chinese": "是的。", + "evidence": [ + { + "description": "Clean the floor.", + "timestamps": { + "start": "00:37:24", + "end": "00:37:40" + } + } + ], + "question_timestamps": "01:10:34", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_8_T2", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_4_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I ever use my phone to calculate a length?", + "answer": "Yes—you used it five minutes ago.", + "question_chinese": "我是否用手机计算过长度？", + "answer_chinese": "是，你五分钟之前使用过。", + "evidence": [ + { + "description": "Use a calculator.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:02" + } + } + ], + "question_timestamps": "00:16:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No—not recently.", + "B": "No—never.", + "C": "No—you didn’t use your phone to calculate any length at any point in this segment.", + "D": "Yes—you used it five minutes ago.", + "E": "No—you used a tape measure instead and never opened a phone calculator for measurements." + }, + "answer_label": "D", + "answer_text": "Yes—you used it five minutes ago.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_4_T2", + "correct_answer": "Yes—you used it five minutes ago.", + "original_options": { + "A": "No—not recently.", + "B": "No—never.", + "C": "No—you didn’t use your phone to calculate any length at any point in this segment.", + "D": "Yes—you used it five minutes ago.", + "E": "No—you used a tape measure instead and never opened a phone calculator for measurements." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 17, + 15 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_4_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I ever use my phone to calculate a length?", + "answer": "No—you haven’t used it at all within the last ten minutes.", + "question_chinese": "我是否用手机计算过长度？", + "answer_chinese": "否，你十分钟之内都没有使用过。", + "evidence": [ + { + "description": "Use a calculator.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:02" + } + } + ], + "question_timestamps": "00:20:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes—just now.", + "B": "No—within eight minutes.", + "C": "No—you haven’t used it at all within the last ten minutes.", + "D": "No—you haven’t used it at all within the last eight minutes.", + "E": "No—you haven’t used it at all within the last twelve minutes." + }, + "answer_label": "C", + "answer_text": "No—you haven’t used it at all within the last ten minutes.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_4_T1", + "correct_answer": "No—you haven’t used it at all within the last ten minutes.", + "original_options": { + "A": "Yes—just now.", + "B": "No—within eight minutes.", + "C": "No—you haven’t used it at all within the last ten minutes.", + "D": "No—you haven’t used it at all within the last eight minutes.", + "E": "No—you haven’t used it at all within the last twelve minutes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 13 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_10", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I ever used a mobile phone to calculate length?", + "answer": "Yes.", + "question_chinese": "我是否用手机计算过长度？", + "answer_chinese": "用过。", + "evidence": [ + { + "description": "Use a calculator.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:02" + } + }, + { + "description": "Use a calculator.", + "timestamps": { + "start": "00:33:37", + "end": "00:33:44" + } + } + ], + "question_timestamps": "00:36:15", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_10", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_1_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow cutting machines do I have?", + "answer": "You have one.", + "question_chinese": "我有几个黄色切割机？", + "answer_chinese": "你有一个。", + "evidence": [ + { + "description": "one.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have one.", + "B": "You have none.", + "C": "You have two.", + "D": "You have three yellow cutting machines in total right now, all visible in your workspace.", + "E": "You have four yellow cutting machines altogether at the moment, lined up together on the table." + }, + "answer_label": "A", + "answer_text": "You have one.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_1_T2", + "correct_answer": "You have one.", + "original_options": { + "A": "You have one.", + "B": "You have none.", + "C": "You have two.", + "D": "You have three yellow cutting machines in total right now, all visible in your workspace.", + "E": "You have four yellow cutting machines altogether at the moment, lined up together on the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_1_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow cutting machines do I have?", + "answer": "You have two.", + "question_chinese": "我有几个黄色切割机？", + "answer_chinese": "你有两个。", + "evidence": [ + { + "description": "Two.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:03" + } + } + ], + "question_timestamps": "00:19:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have one.", + "B": "You have three.", + "C": "You have zero yellow cutting machines right now in your workspace; none are visible on the table.", + "D": "You have four yellow cutting machines in total at this time, placed together in your workspace.", + "E": "You have two." + }, + "answer_label": "E", + "answer_text": "You have two.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_1_T1", + "correct_answer": "You have two.", + "original_options": { + "A": "You have one.", + "B": "You have three.", + "C": "You have zero yellow cutting machines right now in your workspace; none are visible on the table.", + "D": "You have four yellow cutting machines in total at this time, placed together in your workspace.", + "E": "You have two." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 16 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow cutting machines do I have?", + "answer": "Two.", + "question_chinese": "我有几个黄色切割机？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "Two.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:02" + } + } + ], + "question_timestamps": "00:40:37", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four yellow cutting machines are visible on the table in front of you.", + "E": "Five yellow cutting machines are lined up together on the workbench in front of you." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_2", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four yellow cutting machines are visible on the table in front of you.", + "E": "Five yellow cutting machines are lined up together on the workbench in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_19", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the wood-shaving machine directly now?", + "answer": "Yes.", + "question_chinese": "我现在是否可以直接拿到削木头的机器", + "answer_chinese": "可以", + "evidence": [ + { + "description": "Machine position.", + "timestamps": { + "start": "00:57:08", + "end": "00:57:10" + } + }, + { + "description": "Your location.", + "timestamps": { + "start": "00:57:49", + "end": "00:57:49" + } + } + ], + "question_timestamps": "00:57:50", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_19", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_20", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the wood-shaving machine directly now?", + "answer": "No.", + "question_chinese": "我现在是否可以直接拿到削木头的机器", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The machine is put into the toolbox.", + "timestamps": { + "start": "00:59:43", + "end": "00:59:46" + } + } + ], + "question_timestamps": "01:09:51", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_20", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_2_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "The first time I used the yellow cutting machine, how many sides of the wood did I cut?", + "answer": "You cut two faces.", + "question_chinese": "我第一次使用黄色切割机时，切割了木头的几个面？", + "answer_chinese": "你切割了两个面。", + "evidence": [ + { + "description": "One.", + "timestamps": { + "start": "00:00:06", + "end": "00:01:19" + } + }, + { + "description": "Two.", + "timestamps": { + "start": "00:01:23", + "end": "00:03:01" + } + } + ], + "question_timestamps": "00:03:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You cut one face.", + "B": "You cut two faces.", + "C": "You cut three faces.", + "D": "You only cut a single face of the wood on that first pass.", + "E": "You cut all four faces of the wood the first time you used it." + }, + "answer_label": "B", + "answer_text": "You cut two faces.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_2_T2", + "correct_answer": "You cut two faces.", + "original_options": { + "A": "You cut one face.", + "B": "You cut two faces.", + "C": "You cut three faces.", + "D": "You only cut a single face of the wood on that first pass.", + "E": "You cut all four faces of the wood the first time you used it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_2_T1", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "The first time I used the yellow cutting machine, how many sides of the wood did I cut?", + "answer": "You cut three faces.", + "question_chinese": "我第一次使用黄色切割机时，切割了木头的几个面？", + "answer_chinese": "你切割了三个面。", + "evidence": [ + { + "description": "One.", + "timestamps": { + "start": "00:00:06", + "end": "00:01:19" + } + }, + { + "description": "Two.", + "timestamps": { + "start": "00:01:23", + "end": "00:03:01" + } + }, + { + "description": "Tree.", + "timestamps": { + "start": "00:03:01", + "end": "00:04:19" + } + } + ], + "question_timestamps": "00:20:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You cut three faces.", + "B": "You cut two faces.", + "C": "You cut four faces.", + "D": "You cut only one face of the wood the first time you used the machine.", + "E": "You cut five faces of the wood during that first use of the yellow cutting machine." + }, + "answer_label": "A", + "answer_text": "You cut three faces.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_2_T1", + "correct_answer": "You cut three faces.", + "original_options": { + "A": "You cut three faces.", + "B": "You cut two faces.", + "C": "You cut four faces.", + "D": "You cut only one face of the wood the first time you used the machine.", + "E": "You cut five faces of the wood during that first use of the yellow cutting machine." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_4", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sides of the wood did I cut when I used the yellow cutting machine for the first time?", + "answer": "Three.", + "question_chinese": "我第一次使用黄色切割机时，切割了木头的几个面？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "One.", + "timestamps": { + "start": "00:00:06", + "end": "00:01:19" + } + }, + { + "description": "Two.", + "timestamps": { + "start": "00:01:23", + "end": "00:03:01" + } + }, + { + "description": "Tree.", + "timestamps": { + "start": "00:03:01", + "end": "00:04:19" + } + } + ], + "question_timestamps": "00:40:28", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "You cut only one side of the wood the first time you used the yellow cutting machine.", + "E": "You cut five sides of the wood when you used the yellow cutting machine for the first time." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_4", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "You cut only one side of the wood the first time you used the yellow cutting machine.", + "E": "You cut five sides of the wood when you used the yellow cutting machine for the first time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 18 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_5_T1", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "After I entered the small door, what did I take out?", + "answer": "You picked up some long, strip-shaped tools from in front of you.", + "question_chinese": "我进小门后拿出来了什么？", + "answer_chinese": "你在面前的拿了一些长条形的工具。", + "evidence": [ + { + "description": "You took some long-shaped tools.", + "timestamps": { + "start": "00:40:47", + "end": "00:40:51" + } + } + ], + "question_timestamps": "00:59:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Tools from behind you.", + "B": "Tools from your left.", + "C": "You picked up some long, strip-shaped tools from to your right after going through the small door.", + "D": "You took out long, strip-shaped tools from above you, reaching upward just after entering the small door.", + "E": "You picked up some long, strip-shaped tools from in front of you." + }, + "answer_label": "E", + "answer_text": "You picked up some long, strip-shaped tools from in front of you.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_5_T1", + "correct_answer": "You picked up some long, strip-shaped tools from in front of you.", + "original_options": { + "A": "Tools from behind you.", + "B": "Tools from your left.", + "C": "You picked up some long, strip-shaped tools from to your right after going through the small door.", + "D": "You took out long, strip-shaped tools from above you, reaching upward just after entering the small door.", + "E": "You picked up some long, strip-shaped tools from in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_5_T2", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "After I entered the small door, what did I take out?", + "answer": "You took the toolbox out from the small door in front of you.", + "question_chinese": "我进小门后拿出来了什么？", + "answer_chinese": "你从你面前的小门拿出了工具箱。", + "evidence": [ + { + "description": "Toolbox.", + "timestamps": { + "start": "00:41:15", + "end": "00:41:24" + } + } + ], + "question_timestamps": "01:10:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the door behind you.", + "B": "From the door to your left.", + "C": "You took the toolbox out from the small door to your right after you went inside.", + "D": "You took the toolbox out from the small door above you, lifting it down as you opened it.", + "E": "You took the toolbox out from the small door in front of you." + }, + "answer_label": "E", + "answer_text": "You took the toolbox out from the small door in front of you.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_5_T2", + "correct_answer": "You took the toolbox out from the small door in front of you.", + "original_options": { + "A": "From the door behind you.", + "B": "From the door to your left.", + "C": "You took the toolbox out from the small door to your right after you went inside.", + "D": "You took the toolbox out from the small door above you, lifting it down as you opened it.", + "E": "You took the toolbox out from the small door in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 18 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_5", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the car in the garage", + "answer": "White.", + "question_chinese": "车库里的车是什么颜色？", + "answer_chinese": "白色。", + "evidence": [ + { + "description": "White.", + "timestamps": { + "start": "00:09:47", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:19:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Red.", + "B": "Blue.", + "C": "White.", + "D": "It is a dark gray car parked inside the garage, not the white one.", + "E": "The car is painted bright yellow, standing out against the dim garage interior." + }, + "answer_label": "C", + "answer_text": "White.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_5", + "correct_answer": "White.", + "original_options": { + "A": "Red.", + "B": "Blue.", + "C": "White.", + "D": "It is a dark gray car parked inside the garage, not the white one.", + "E": "The car is painted bright yellow, standing out against the dim garage interior." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "7f09822a-87b9-4eac-bb34-3f1059c704d1", + "video_length": "01:10:34", + "qid": "7f09822a-87b9-4eac-bb34-3f1059c704d1_6", + "qid_num": 28, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the car in the garage", + "answer": "White.", + "question_chinese": "车库里的车是什么颜色？", + "answer_chinese": "白色。", + "evidence": [ + { + "description": "White.", + "timestamps": { + "start": "00:09:47", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:37:39", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Yellow.", + "B": "White.", + "C": "Red.", + "D": "It is a solid green car in the garage, not a white one.", + "E": "The car is a pearly silver color, not a plain painted finish." + }, + "answer_label": "B", + "answer_text": "White.", + "q_id": "7f09822a-87b9-4eac-bb34-3f1059c704d1_6", + "correct_answer": "White.", + "original_options": { + "A": "Yellow.", + "B": "White.", + "C": "Red.", + "D": "It is a solid green car in the garage, not a white one.", + "E": "The car is a pearly silver color, not a plain painted finish." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + } +] diff --git a/QAs/7f9f75fd-a660-4635-8890-239c6ad82023-mcq.json b/QAs/7f9f75fd-a660-4635-8890-239c6ad82023-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..09c19ab0a8f353c135d6cdeea2684905ac8f2b35 --- /dev/null +++ b/QAs/7f9f75fd-a660-4635-8890-239c6ad82023-mcq.json @@ -0,0 +1,1641 @@ +[ + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_2_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the bowl back where it originally was, where should I put it?", + "answer": "You should put the bowl into the cabinet directly in front of you.", + "question_chinese": "如果我要将碗放回原位，应该放在哪里？", + "answer_chinese": "你应该将碗放在你正面前的橱柜里。", + "evidence": [ + { + "description": "The bowl was originally located in the second drawer from top to bottom directly below the cutting board.", + "timestamps": { + "start": "00:04:54", + "end": "00:04:56" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the cabinet behind you.", + "B": "In the cabinet to your left.", + "C": "Put the bowl into the cabinet directly to your right, the one you can reach without stepping forward.", + "D": "Place the bowl into the cabinet above you, not the one straight ahead at eye level.", + "E": "You should put the bowl into the cabinet directly in front of you." + }, + "answer_label": "E", + "answer_text": "You should put the bowl into the cabinet directly in front of you.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_2_T1", + "correct_answer": "You should put the bowl into the cabinet directly in front of you.", + "original_options": { + "A": "In the cabinet behind you.", + "B": "In the cabinet to your left.", + "C": "Put the bowl into the cabinet directly to your right, the one you can reach without stepping forward.", + "D": "Place the bowl into the cabinet above you, not the one straight ahead at eye level.", + "E": "You should put the bowl into the cabinet directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 16 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the bowl back in place, where should I put it?", + "answer": "You should place the bowl directly under the cutting board and count it from top to bottom in the second drawer.", + "question_chinese": "如果我要将碗放回原位，应该放在哪里？", + "answer_chinese": "你应该将碗放在菜板正下方从上往下数第二个的抽屉里。", + "evidence": [ + { + "description": "The bowl was originally located in the second drawer from top to bottom directly below the cutting board.", + "timestamps": { + "start": "00:04:54", + "end": "00:04:56" + } + } + ], + "question_timestamps": "00:14:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put it in the cabinet ahead.", + "B": "Use the bottom drawer instead.", + "C": "Place the bowl under the cutting board, but put it in the top drawer rather than the second one.", + "D": "Put the bowl in the drawer below the cutting board, counting from the bottom so it ends up in the third drawer.", + "E": "You should place the bowl directly under the cutting board and count it from top to bottom in the second drawer." + }, + "answer_label": "E", + "answer_text": "You should place the bowl directly under the cutting board and count it from top to bottom in the second drawer.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_2", + "correct_answer": "You should place the bowl directly under the cutting board and count it from top to bottom in the second drawer.", + "original_options": { + "A": "Put it in the cabinet ahead.", + "B": "Use the bottom drawer instead.", + "C": "Place the bowl under the cutting board, but put it in the top drawer rather than the second one.", + "D": "Put the bowl in the drawer below the cutting board, counting from the bottom so it ends up in the third drawer.", + "E": "You should place the bowl directly under the cutting board and count it from top to bottom in the second drawer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 22 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_2_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the bowl back where it originally was, where should I put it?", + "answer": "You should put the bowl into the drawer that’s directly below the cutting board in front of you—the second drawer down from the top.", + "question_chinese": "如果我要将碗放回原位，应该放在哪里？", + "answer_chinese": "你应该将碗放在你面前的菜板正下方从上往下数第二个的抽屉里。", + "evidence": [ + { + "description": "The bowl was originally located in the second drawer from top to bottom directly below the cutting board.", + "timestamps": { + "start": "00:04:54", + "end": "00:04:56" + } + } + ], + "question_timestamps": "00:27:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you, second drawer down.", + "B": "You should put the bowl into the drawer that’s directly below the cutting board in front of you—the second drawer down from the top.", + "C": "Left side, second drawer down.", + "D": "Put the bowl in the second drawer down, but on the cabinet to your right instead of in front.", + "E": "Place the bowl into the drawer below the cutting board above you, choosing the second drawer from the top." + }, + "answer_label": "B", + "answer_text": "You should put the bowl into the drawer that’s directly below the cutting board in front of you—the second drawer down from the top.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_2_T2", + "correct_answer": "You should put the bowl into the drawer that’s directly below the cutting board in front of you—the second drawer down from the top.", + "original_options": { + "A": "Behind you, second drawer down.", + "B": "You should put the bowl into the drawer that’s directly below the cutting board in front of you—the second drawer down from the top.", + "C": "Left side, second drawer down.", + "D": "Put the bowl in the second drawer down, but on the cabinet to your right instead of in front.", + "E": "Place the bowl into the drawer below the cutting board above you, choosing the second drawer from the top." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_5_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the basket back to its original starting position, how should I place it?", + "answer": "You should squat down and put the basket into the bottom drawer directly below the cutting board.", + "question_chinese": "如果我要把篮子放回最开始的位置，我应该怎么放？", + "answer_chinese": "你应该蹲下，把篮子放到在菜板正下方最下面的抽屉里。", + "evidence": [ + { + "description": "The basket was in the bottom drawer directly below the cutting board.", + "timestamps": { + "start": "00:08:29", + "end": "00:08:31" + } + } + ], + "question_timestamps": "00:13:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it in the drawer above.", + "B": "Place it in the cabinet ahead.", + "C": "Turn right first, then squat and place the basket into the bottom drawer of the cabinet to your right.", + "D": "You should squat down and put the basket into the bottom drawer directly below the cutting board.", + "E": "Squat down and put the basket into the second drawer from the bottom, not the very bottom one." + }, + "answer_label": "D", + "answer_text": "You should squat down and put the basket into the bottom drawer directly below the cutting board.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_5_T1", + "correct_answer": "You should squat down and put the basket into the bottom drawer directly below the cutting board.", + "original_options": { + "A": "Put it in the drawer above.", + "B": "Place it in the cabinet ahead.", + "C": "Turn right first, then squat and place the basket into the bottom drawer of the cabinet to your right.", + "D": "You should squat down and put the basket into the bottom drawer directly below the cutting board.", + "E": "Squat down and put the basket into the second drawer from the bottom, not the very bottom one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_5_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the basket back to its original starting position, how should I place it?", + "answer": "You should turn right, then squat down and put the basket into the bottom drawer of the cabinet to your right.", + "question_chinese": "如果我要把篮子放回最开始的位置，我应该怎么放？", + "answer_chinese": "你应该向右转，然后蹲下，把篮子放到你右边橱柜最下面的抽屉里。", + "evidence": [ + { + "description": "The basket was in the bottom drawer directly below the cutting board.", + "timestamps": { + "start": "00:08:29", + "end": "00:08:31" + } + } + ], + "question_timestamps": "00:39:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use the cabinet in front of you.", + "B": "Use the cabinet behind you.", + "C": "Turn right, then squat down and place the basket into the bottom drawer of the cabinet to your left.", + "D": "You should turn right, then squat down and put the basket into the bottom drawer of the cabinet to your right.", + "E": "After turning right, squat and put the basket into the bottom drawer of the cabinet directly in front of you." + }, + "answer_label": "D", + "answer_text": "You should turn right, then squat down and put the basket into the bottom drawer of the cabinet to your right.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_5_T2", + "correct_answer": "You should turn right, then squat down and put the basket into the bottom drawer of the cabinet to your right.", + "original_options": { + "A": "Use the cabinet in front of you.", + "B": "Use the cabinet behind you.", + "C": "Turn right, then squat down and place the basket into the bottom drawer of the cabinet to your left.", + "D": "You should turn right, then squat down and put the basket into the bottom drawer of the cabinet to your right.", + "E": "After turning right, squat and put the basket into the bottom drawer of the cabinet directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 19, + 20 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_6", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the basket back in its original position, where should I put it?", + "answer": "You should put the basket in the bottom drawer directly below the cutting board.", + "question_chinese": "如果我要把篮子放回最开始的位置，我应该放到哪里？", + "answer_chinese": "你应该把篮子放到在菜板正下方最下面的抽屉里。", + "evidence": [ + { + "description": "The basket was in the bottom drawer directly below the cutting board.", + "timestamps": { + "start": "00:08:29", + "end": "00:08:31" + } + } + ], + "question_timestamps": "00:25:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should put the basket in the bottom drawer directly below the cutting board.", + "B": "In the top cupboard above it.", + "C": "On the counter left of the fridge.", + "D": "Put the basket into the right-hand drawer next to the microwave instead of the bottom drawer by the cutting board.", + "E": "Place the basket on the shelf in front of the window above the sink, not in any drawer." + }, + "answer_label": "A", + "answer_text": "You should put the basket in the bottom drawer directly below the cutting board.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_6", + "correct_answer": "You should put the basket in the bottom drawer directly below the cutting board.", + "original_options": { + "A": "You should put the basket in the bottom drawer directly below the cutting board.", + "B": "In the top cupboard above it.", + "C": "On the counter left of the fridge.", + "D": "Put the basket into the right-hand drawer next to the microwave instead of the bottom drawer by the cutting board.", + "E": "Place the basket on the shelf in front of the window above the sink, not in any drawer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 18 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_16", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to put the basket back in its original position, where should I put it?", + "answer": "You should put the basket in the bottom drawer directly below the cutting board.", + "question_chinese": "如果我要把篮子放回最开始的位置，我应该放到哪里？", + "answer_chinese": "你应该把篮子放到在菜板正下方最下面的抽屉里。", + "evidence": [ + { + "description": "The basket was in the bottom drawer directly below the cutting board.", + "timestamps": { + "start": "00:08:29", + "end": "00:08:31" + } + } + ], + "question_timestamps": "00:39:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should put the basket in the bottom drawer directly below the cutting board.", + "B": "Put it in the middle drawer.", + "C": "Put it in the drawer above.", + "D": "Put the basket in the bottom drawer, but on the opposite side from the cutting board.", + "E": "Return the basket to the lower cabinet under the cutting board area instead of inside a drawer." + }, + "answer_label": "A", + "answer_text": "You should put the basket in the bottom drawer directly below the cutting board.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_16", + "correct_answer": "You should put the basket in the bottom drawer directly below the cutting board.", + "original_options": { + "A": "You should put the basket in the bottom drawer directly below the cutting board.", + "B": "Put it in the middle drawer.", + "C": "Put it in the drawer above.", + "D": "Put the basket in the bottom drawer, but on the opposite side from the cutting board.", + "E": "Return the basket to the lower cabinet under the cutting board area instead of inside a drawer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_4_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just went to get the cabbage?", + "answer": "You spun around in place twice, then opened the refrigerator and took out the cabbage.", + "question_chinese": "我刚刚去拿卷心菜的时候经过了什么路径？", + "answer_chinese": "你在原地转了两圈打开冰箱拿出卷心菜。", + "evidence": [ + { + "description": "I turned back from the sink and took two steps to the fridge, then opened it and took out cabbage.", + "timestamps": { + "start": "00:11:56", + "end": "00:12:03" + } + } + ], + "question_timestamps": "00:17:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked straight to the refrigerator.", + "B": "You spun around in place twice, then opened the refrigerator and took out the cabbage.", + "C": "You turned left and opened the fridge.", + "D": "You turned away from the sink, took a couple steps, then opened the fridge and pulled out the cabbage.", + "E": "You made a quick half-turn, walked over to the refrigerator, opened the door, and grabbed the cabbage." + }, + "answer_label": "B", + "answer_text": "You spun around in place twice, then opened the refrigerator and took out the cabbage.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_4_T1", + "correct_answer": "You spun around in place twice, then opened the refrigerator and took out the cabbage.", + "original_options": { + "A": "You walked straight to the refrigerator.", + "B": "You spun around in place twice, then opened the refrigerator and took out the cabbage.", + "C": "You turned left and opened the fridge.", + "D": "You turned away from the sink, took a couple steps, then opened the fridge and pulled out the cabbage.", + "E": "You made a quick half-turn, walked over to the refrigerator, opened the door, and grabbed the cabbage." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 17 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_5", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take when I just went to get cabbage?", + "answer": "You turned back from the sink and took two steps to the fridge, then opened it and took out cabbage.", + "question_chinese": "我刚刚去拿卷心菜的时候经过了什么路径？", + "answer_chinese": "你从洗手台向后转身走了两步到冰箱旁边，打开冰箱拿出卷心菜。", + "evidence": [ + { + "description": "I turned back from the sink and took two steps to the fridge, then opened it and took out cabbage.", + "timestamps": { + "start": "00:11:56", + "end": "00:12:03" + } + } + ], + "question_timestamps": "00:27:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You went to the pantry for it.", + "B": "You grabbed it from the counter.", + "C": "You turned from the sink toward the table, then reached down and took the cabbage from a basket.", + "D": "You turned back from the sink and took two steps to the fridge, then opened it and took out cabbage.", + "E": "You backed away from the sink, walked toward the island, and picked up the cabbage from the top." + }, + "answer_label": "D", + "answer_text": "You turned back from the sink and took two steps to the fridge, then opened it and took out cabbage.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_5", + "correct_answer": "You turned back from the sink and took two steps to the fridge, then opened it and took out cabbage.", + "original_options": { + "A": "You went to the pantry for it.", + "B": "You grabbed it from the counter.", + "C": "You turned from the sink toward the table, then reached down and took the cabbage from a basket.", + "D": "You turned back from the sink and took two steps to the fridge, then opened it and took out cabbage.", + "E": "You backed away from the sink, walked toward the island, and picked up the cabbage from the top." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 18 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_4_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just went to get the cabbage?", + "answer": "From the sink, you turned around and walked two steps to next to the refrigerator, opened it, and took out the cabbage.", + "question_chinese": "我刚刚去拿卷心菜的时候经过了什么路径？", + "answer_chinese": "你从洗手台向后转身走了两步到冰箱旁边，打开冰箱拿出卷心菜。", + "evidence": [ + { + "description": "I turned back from the sink and took two steps to the fridge, then opened it and took out cabbage.", + "timestamps": { + "start": "00:11:56", + "end": "00:12:03" + } + } + ], + "question_timestamps": "00:34:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You spun in place twice.", + "B": "You walked right to the fridge.", + "C": "From the sink, you turned around and took three steps, then opened the refrigerator and took out cabbage.", + "D": "From the sink, you turned around and walked two steps to next to the refrigerator, opened it, and took out the cabbage.", + "E": "From the sink, you pivoted left and walked a few steps before opening the refrigerator to take out the cabbage." + }, + "answer_label": "D", + "answer_text": "From the sink, you turned around and walked two steps to next to the refrigerator, opened it, and took out the cabbage.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_4_T2", + "correct_answer": "From the sink, you turned around and walked two steps to next to the refrigerator, opened it, and took out the cabbage.", + "original_options": { + "A": "You spun in place twice.", + "B": "You walked right to the fridge.", + "C": "From the sink, you turned around and took three steps, then opened the refrigerator and took out cabbage.", + "D": "From the sink, you turned around and walked two steps to next to the refrigerator, opened it, and took out the cabbage.", + "E": "From the sink, you pivoted left and walked a few steps before opening the refrigerator to take out the cabbage." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 20 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_6_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was just scooping the vegetables out of the pot, could I reach the paper next to it?", + "answer": "You can’t reach the paper next to you; the paper is directly behind you.", + "question_chinese": "我刚刚把菜从锅里盛出来时，能拿到旁边的纸吗？", + "answer_chinese": "你不能拿到旁边的纸，纸在你的正后方。", + "evidence": [ + { + "description": "The paper was next to the pot, on your left.", + "timestamps": { + "start": "00:35:46", + "end": "00:35:48" + } + } + ], + "question_timestamps": "00:39:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes, it was within reach.", + "B": "No, it was in front of you.", + "C": "No, you couldn’t reach it because the paper was off to your left side.", + "D": "You can’t reach the paper next to you; the paper is directly behind you.", + "E": "No, you couldn’t reach it since the paper was to your right, away from where you were scooping." + }, + "answer_label": "D", + "answer_text": "You can’t reach the paper next to you; the paper is directly behind you.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_6_T1", + "correct_answer": "You can’t reach the paper next to you; the paper is directly behind you.", + "original_options": { + "A": "Yes, it was within reach.", + "B": "No, it was in front of you.", + "C": "No, you couldn’t reach it because the paper was off to your left side.", + "D": "You can’t reach the paper next to you; the paper is directly behind you.", + "E": "No, you couldn’t reach it since the paper was to your right, away from where you were scooping." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 15, + 19 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_7", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Can I get the paper next to it when I just took the dishes out of the pot?", + "answer": "You can get the paper next to it, it's next to the pot, on your left.", + "question_chinese": "我刚刚把菜从锅里盛出来时，能拿到旁边的纸吗？", + "answer_chinese": "你能拿到旁边的纸，纸就在锅旁边，你的左手边。", + "evidence": [ + { + "description": "The paper was next to the pot, on your left.", + "timestamps": { + "start": "00:35:46", + "end": "00:35:48" + } + }, + { + "description": "The paper was next to the pot, on your left.", + "timestamps": { + "start": "00:40:46", + "end": "00:41:00" + } + } + ], + "question_timestamps": "00:41:28", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You can get it; on your right.", + "B": "You can get it; it’s behind you.", + "C": "You can get the paper next to it, it's next to the pot, on your left.", + "D": "You can get the paper next to it; it’s next to the pot, but positioned on your right side.", + "E": "You can get the paper next to it; it’s on the counter above the pot rather than beside it." + }, + "answer_label": "C", + "answer_text": "You can get the paper next to it, it's next to the pot, on your left.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_7", + "correct_answer": "You can get the paper next to it, it's next to the pot, on your left.", + "original_options": { + "A": "You can get it; on your right.", + "B": "You can get it; it’s behind you.", + "C": "You can get the paper next to it, it's next to the pot, on your left.", + "D": "You can get the paper next to it; it’s next to the pot, but positioned on your right side.", + "E": "You can get the paper next to it; it’s on the counter above the pot rather than beside it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 20, + 20 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_6_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was just scooping the vegetables out of the pot, could I reach the paper next to it?", + "answer": "You can reach the paper next to you—the paper is next to the pot, by your left hand.", + "question_chinese": "我刚刚把菜从锅里盛出来时，能拿到旁边的纸吗？", + "answer_chinese": "你能拿到旁边的纸，纸就在锅旁边，你的左手边。", + "evidence": [ + { + "description": "The paper was next to the pot, on your left.", + "timestamps": { + "start": "00:35:46", + "end": "00:35:48" + } + }, + { + "description": "The paper was next to the pot, on your left.", + "timestamps": { + "start": "00:40:46", + "end": "00:41:00" + } + } + ], + "question_timestamps": "00:42:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can reach it, right side.", + "B": "You can’t reach it, behind you.", + "C": "You can reach the paper next to you—the paper is next to the pot, by your left hand.", + "D": "You can reach the paper next to you, but it sits beside the pot closer to your right hand.", + "E": "You can reach the paper next to you, but it is on the counter in front of the pot." + }, + "answer_label": "C", + "answer_text": "You can reach the paper next to you—the paper is next to the pot, by your left hand.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_6_T2", + "correct_answer": "You can reach the paper next to you—the paper is next to the pot, by your left hand.", + "original_options": { + "A": "You can reach it, right side.", + "B": "You can’t reach it, behind you.", + "C": "You can reach the paper next to you—the paper is next to the pot, by your left hand.", + "D": "You can reach the paper next to you, but it sits beside the pot closer to your right hand.", + "E": "You can reach the paper next to you, but it is on the counter in front of the pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 19 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_1", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the oil when I just chopped the vegetables?", + "answer": "When you were cutting vegetables just now, you could get the oil, which is on your left side.", + "question_chinese": "我刚才切菜的时候能拿到油吗？", + "answer_chinese": "你刚才切菜的时候能拿到油，油在你的左手边。", + "evidence": [ + { + "description": "The oil is about 15cm to the right of the cutting board.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:14:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "No, you couldn’t get the oil.", + "B": "When you were cutting vegetables just now, you could get the oil, which is on your left side.", + "C": "Yes, it was on your right.", + "D": "When you were cutting vegetables just now, you could get the oil, which was in front of you on the counter.", + "E": "When you were cutting vegetables just now, you could get the oil, but it was on your right side." + }, + "answer_label": "B", + "answer_text": "When you were cutting vegetables just now, you could get the oil, which is on your left side.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_1", + "correct_answer": "When you were cutting vegetables just now, you could get the oil, which is on your left side.", + "original_options": { + "A": "No, you couldn’t get the oil.", + "B": "When you were cutting vegetables just now, you could get the oil, which is on your left side.", + "C": "Yes, it was on your right.", + "D": "When you were cutting vegetables just now, you could get the oil, which was in front of you on the counter.", + "E": "When you were cutting vegetables just now, you could get the oil, but it was on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 21, + 19 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_13", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the oil when I just chopped the vegetables?", + "answer": "When you were cutting vegetables just now, you could get the oil, which is on your left side.", + "question_chinese": "我刚才切菜的时候能拿到油吗？", + "answer_chinese": "你刚才切菜的时候能拿到油，油在你的左手边。", + "evidence": [ + { + "description": "The oil is about 15cm to the right of the cutting board.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:30:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "No, the oil was out of reach.", + "B": "Yes, the oil was on your right.", + "C": "When you were cutting vegetables just now, you could get the oil, but it was behind you on the counter.", + "D": "When you were cutting vegetables just now, you couldn’t get the oil because it was tucked away in the cabinet.", + "E": "When you were cutting vegetables just now, you could get the oil, which is on your left side." + }, + "answer_label": "E", + "answer_text": "When you were cutting vegetables just now, you could get the oil, which is on your left side.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_13", + "correct_answer": "When you were cutting vegetables just now, you could get the oil, which is on your left side.", + "original_options": { + "A": "No, the oil was out of reach.", + "B": "Yes, the oil was on your right.", + "C": "When you were cutting vegetables just now, you could get the oil, but it was behind you on the counter.", + "D": "When you were cutting vegetables just now, you couldn’t get the oil because it was tucked away in the cabinet.", + "E": "When you were cutting vegetables just now, you could get the oil, which is on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 21 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_8", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many bowls were there next to me when I was frying cabbage just now?", + "answer": "When you were frying cabbage just now, there were two bowls next to it.", + "question_chinese": "我刚才炒卷心菜的时候，旁边有几个碗？", + "answer_chinese": "你刚才炒卷心菜的时候，旁边有2个碗。", + "evidence": [ + { + "description": "When I was frying cabbage just now, there were two bowls next to it.", + "timestamps": { + "start": "00:32:30", + "end": "00:36:01" + } + } + ], + "question_timestamps": "00:40:11", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "When you were frying cabbage just now, there were two bowls next to it.", + "B": "One bowl.", + "C": "Three bowls.", + "D": "When you were frying cabbage just now, there were four bowls next to it.", + "E": "When you were frying cabbage just now, there were no bowls next to it at all." + }, + "answer_label": "A", + "answer_text": "When you were frying cabbage just now, there were two bowls next to it.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_8", + "correct_answer": "When you were frying cabbage just now, there were two bowls next to it.", + "original_options": { + "A": "When you were frying cabbage just now, there were two bowls next to it.", + "B": "One bowl.", + "C": "Three bowls.", + "D": "When you were frying cabbage just now, there were four bowls next to it.", + "E": "When you were frying cabbage just now, there were no bowls next to it at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_18", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many bowls were there next to me when I was frying cabbage just now?", + "answer": "When you were frying cabbage just now, there were two bowls next to it.", + "question_chinese": "我刚才炒卷心菜的时候，旁边有几个碗？", + "answer_chinese": "你刚才炒卷心菜的时候，旁边有2个碗。", + "evidence": [ + { + "description": "When I was frying cabbage just now, there were two bowls next to it.", + "timestamps": { + "start": "00:32:30", + "end": "00:36:01" + } + } + ], + "question_timestamps": "00:42:51", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One bowl.", + "B": "When you were frying cabbage just now, there were two bowls next to it.", + "C": "Four bowls.", + "D": "When you were frying cabbage just now, there were three bowls next to it.", + "E": "When you were frying cabbage just now, there were no bowls next to it." + }, + "answer_label": "B", + "answer_text": "When you were frying cabbage just now, there were two bowls next to it.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_18", + "correct_answer": "When you were frying cabbage just now, there were two bowls next to it.", + "original_options": { + "A": "One bowl.", + "B": "When you were frying cabbage just now, there were two bowls next to it.", + "C": "Four bowls.", + "D": "When you were frying cabbage just now, there were three bowls next to it.", + "E": "When you were frying cabbage just now, there were no bowls next to it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_7_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was cooking at the stove, how many bowls were on the stovetop?", + "answer": "There are five bowls on the stovetop in front of you.", + "question_chinese": "我在灶台做饭时，灶台上有几个碗？", + "answer_chinese": "你面前的灶台上有五个碗。", + "evidence": [ + { + "description": "There are four bowls on the stove in front of you.", + "timestamps": { + "start": "00:26:13", + "end": "00:26:14" + } + } + ], + "question_timestamps": "00:26:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three bowls.", + "B": "Six bowls.", + "C": "There are five bowls on the stovetop in front of you.", + "D": "There are four bowls on the stovetop directly in front of you during cooking.", + "E": "There are seven bowls on the stovetop directly in front of you while you are cooking." + }, + "answer_label": "C", + "answer_text": "There are five bowls on the stovetop in front of you.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_7_T1", + "correct_answer": "There are five bowls on the stovetop in front of you.", + "original_options": { + "A": "Three bowls.", + "B": "Six bowls.", + "C": "There are five bowls on the stovetop in front of you.", + "D": "There are four bowls on the stovetop directly in front of you during cooking.", + "E": "There are seven bowls on the stovetop directly in front of you while you are cooking." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_7_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was cooking at the stove, how many bowls were on the stovetop?", + "answer": "There are five bowls on the stovetop in front of you.", + "question_chinese": "我在灶台做饭时，灶台上有几个碗？", + "answer_chinese": "你面前的灶台有五个碗。", + "evidence": [ + { + "description": "You took out a bowl from your hand.", + "timestamps": { + "start": "00:26:18", + "end": "00:26:19" + } + }, + { + "description": "There are five bowls on the stove in front of you.", + "timestamps": { + "start": "00:26:16", + "end": "00:26:17" + } + } + ], + "question_timestamps": "00:26:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four bowls.", + "B": "Seven bowls.", + "C": "There are three bowls on the stovetop directly in front of you during cooking.", + "D": "There are five bowls on the stovetop in front of you.", + "E": "There are six bowls on the stovetop directly in front of you while you are cooking." + }, + "answer_label": "D", + "answer_text": "There are five bowls on the stovetop in front of you.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_7_T2", + "correct_answer": "There are five bowls on the stovetop in front of you.", + "original_options": { + "A": "Four bowls.", + "B": "Seven bowls.", + "C": "There are three bowls on the stovetop directly in front of you during cooking.", + "D": "There are five bowls on the stovetop in front of you.", + "E": "There are six bowls on the stovetop directly in front of you while you are cooking." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_8_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I went to get the cabbage, could I reach the blue kettle?", + "answer": "You couldn’t reach the blue kettle; it was on the top shelf inside the deep refrigerator in front of you.", + "question_chinese": "我拿卷心菜的时候可以拿到蓝色水壶吗？", + "answer_chinese": "你不可以拿到蓝色水壶，蓝色水壶在你深厚的冰箱里最上面一格。", + "evidence": [ + { + "description": "The blue kettle was at the top of the fridge.", + "timestamps": { + "start": "00:11:58", + "end": "00:12:01" + } + } + ], + "question_timestamps": "00:12:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You could reach it.", + "B": "It was on the bottom shelf.", + "C": "You couldn’t reach the blue kettle; it was on the bottom shelf inside the deep refrigerator in front of you.", + "D": "You couldn’t reach the blue kettle; it was on the top shelf inside the deep refrigerator in front of you.", + "E": "You couldn’t reach the blue kettle; it was on the top shelf inside the deep refrigerator to your left." + }, + "answer_label": "D", + "answer_text": "You couldn’t reach the blue kettle; it was on the top shelf inside the deep refrigerator in front of you.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_8_T1", + "correct_answer": "You couldn’t reach the blue kettle; it was on the top shelf inside the deep refrigerator in front of you.", + "original_options": { + "A": "You could reach it.", + "B": "It was on the bottom shelf.", + "C": "You couldn’t reach the blue kettle; it was on the bottom shelf inside the deep refrigerator in front of you.", + "D": "You couldn’t reach the blue kettle; it was on the top shelf inside the deep refrigerator in front of you.", + "E": "You couldn’t reach the blue kettle; it was on the top shelf inside the deep refrigerator to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 21, + 20 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_9", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Can I bring the blue kettle when I take the cabbage?", + "answer": "When you take cabbage, you can take the blue kettle, which is at the top of the fridge.", + "question_chinese": "我拿卷心菜的时候可以拿到蓝色水壶吗？", + "answer_chinese": "你拿卷心菜的时候可以拿到蓝色水壶，蓝色水壶在冰箱里最上面一格。", + "evidence": [ + { + "description": "The blue kettle was at the top of the fridge.", + "timestamps": { + "start": "00:11:58", + "end": "00:12:01" + } + } + ], + "question_timestamps": "00:19:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "No, you can’t take it.", + "B": "It’s at the bottom of the fridge.", + "C": "When you take cabbage, you can take the blue kettle, which is on the counter next to you.", + "D": "When you take cabbage, you can take the blue kettle, which is inside a kitchen drawer by your side.", + "E": "When you take cabbage, you can take the blue kettle, which is at the top of the fridge." + }, + "answer_label": "E", + "answer_text": "When you take cabbage, you can take the blue kettle, which is at the top of the fridge.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_9", + "correct_answer": "When you take cabbage, you can take the blue kettle, which is at the top of the fridge.", + "original_options": { + "A": "No, you can’t take it.", + "B": "It’s at the bottom of the fridge.", + "C": "When you take cabbage, you can take the blue kettle, which is on the counter next to you.", + "D": "When you take cabbage, you can take the blue kettle, which is inside a kitchen drawer by your side.", + "E": "When you take cabbage, you can take the blue kettle, which is at the top of the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 18, + 19 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_8_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I went to get the cabbage, could I reach the blue kettle?", + "answer": "When you grabbed the cabbage, you could also reach the blue kettle; the blue kettle was on the top shelf inside the refrigerator.", + "question_chinese": "我拿卷心菜的时候可以拿到蓝色水壶吗？", + "answer_chinese": "你拿卷心菜的时候可以拿到蓝色水壶，蓝色水壶在冰箱里最上面一格。", + "evidence": [ + { + "description": "The blue kettle was at the top of the fridge.", + "timestamps": { + "start": "00:11:58", + "end": "00:12:01" + } + } + ], + "question_timestamps": "00:43:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You couldn’t reach it.", + "B": "When you grabbed the cabbage, you could also reach the blue kettle; the blue kettle was on the top shelf inside the refrigerator.", + "C": "It was outside the refrigerator.", + "D": "When you grabbed the cabbage, you couldn’t reach the blue kettle; it was on the top shelf inside the refrigerator.", + "E": "When you grabbed the cabbage, you could reach the blue kettle; it was on the bottom shelf inside the refrigerator." + }, + "answer_label": "B", + "answer_text": "When you grabbed the cabbage, you could also reach the blue kettle; the blue kettle was on the top shelf inside the refrigerator.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_8_T2", + "correct_answer": "When you grabbed the cabbage, you could also reach the blue kettle; the blue kettle was on the top shelf inside the refrigerator.", + "original_options": { + "A": "You couldn’t reach it.", + "B": "When you grabbed the cabbage, you could also reach the blue kettle; the blue kettle was on the top shelf inside the refrigerator.", + "C": "It was outside the refrigerator.", + "D": "When you grabbed the cabbage, you couldn’t reach the blue kettle; it was on the top shelf inside the refrigerator.", + "E": "When you grabbed the cabbage, you could reach the blue kettle; it was on the bottom shelf inside the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 20 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_1_T1", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the oil right now?", + "answer": "You couldn’t reach the oil—the oil was on the far left of the stovetop, and you were on the far right.", + "question_chinese": "我现在能拿到油吗？", + "answer_chinese": "你不能拿到油，油在灶台的上最左侧，你在最右侧。", + "evidence": [ + { + "description": "The oil is about 15cm to the right of the cutting board.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:14:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes, you could reach the oil.", + "B": "You couldn’t reach the oil—the oil was on the far left of the stovetop, and you were on the far right.", + "C": "It was right beside you.", + "D": "No, you couldn’t reach it because the oil was on the far right, and you were on the far left.", + "E": "No, you couldn’t reach it; the oil was behind you on the counter, out of reach." + }, + "answer_label": "B", + "answer_text": "You couldn’t reach the oil—the oil was on the far left of the stovetop, and you were on the far right.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_1_T1", + "correct_answer": "You couldn’t reach the oil—the oil was on the far left of the stovetop, and you were on the far right.", + "original_options": { + "A": "Yes, you could reach the oil.", + "B": "You couldn’t reach the oil—the oil was on the far left of the stovetop, and you were on the far right.", + "C": "It was right beside you.", + "D": "No, you couldn’t reach it because the oil was on the far right, and you were on the far left.", + "E": "No, you couldn’t reach it; the oil was behind you on the counter, out of reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 17 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_1_T2", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the oil right now?", + "answer": "When you were chopping vegetables just now, you could reach the oil; it was by your left hand.", + "question_chinese": "我现在能拿到油吗？", + "answer_chinese": "你刚才切菜的时候能拿到油，油在你的左手边。", + "evidence": [ + { + "description": "The oil is about 15cm to the right of the cutting board.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:30:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No, you couldn’t reach it.", + "B": "Yes, it was by your right hand.", + "C": "When you were chopping vegetables just now, you could reach the oil; it was by your left hand.", + "D": "When you were chopping vegetables just now, the oil was across the stovetop from you, out of reach.", + "E": "When you were chopping vegetables just now, you could reach the oil because it was near your right side." + }, + "answer_label": "C", + "answer_text": "When you were chopping vegetables just now, you could reach the oil; it was by your left hand.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_1_T2", + "correct_answer": "When you were chopping vegetables just now, you could reach the oil; it was by your left hand.", + "original_options": { + "A": "No, you couldn’t reach it.", + "B": "Yes, it was by your right hand.", + "C": "When you were chopping vegetables just now, you could reach the oil; it was by your left hand.", + "D": "When you were chopping vegetables just now, the oil was across the stovetop from you, out of reach.", + "E": "When you were chopping vegetables just now, you could reach the oil because it was near your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_10_T1", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After using the apron, I want to put it back where it was—where should I put it back?", + "answer": "You should put the apron back on the shelf next to the door.", + "question_chinese": "用完围裙，我想放回原位，我应该放回哪里？", + "answer_chinese": "你应该把围裙放回门旁边的架子上。", + "evidence": [ + { + "description": "The apron is hung on the wall near the side of the cabinet near the refrigerator.", + "timestamps": { + "start": "00:28:17", + "end": "00:28:19" + } + } + ], + "question_timestamps": "00:36:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it back on the cabinet wall.", + "B": "Hang it on the cabinet side.", + "C": "You should put the apron back on the shelf next to the door.", + "D": "You should put the apron back on the wall on the side of the cabinet near the refrigerator.", + "E": "You should put the apron back on the shelf near the refrigerator, not on the shelf next to the door." + }, + "answer_label": "C", + "answer_text": "You should put the apron back on the shelf next to the door.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_10_T1", + "correct_answer": "You should put the apron back on the shelf next to the door.", + "original_options": { + "A": "Put it back on the cabinet wall.", + "B": "Hang it on the cabinet side.", + "C": "You should put the apron back on the shelf next to the door.", + "D": "You should put the apron back on the wall on the side of the cabinet near the refrigerator.", + "E": "You should put the apron back on the shelf near the refrigerator, not on the shelf next to the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 20 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_10_T2", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After using the apron, I want to put it back where it was—where should I put it back?", + "answer": "You should put the apron back on the wall on the side of the cabinet near the refrigerator.", + "question_chinese": "用完围裙，我想放回原位，我应该放回哪里？", + "answer_chinese": "你应该把围裙放回靠近冰箱的柜子侧面的墙上。", + "evidence": [ + { + "description": "The apron is hung on the wall near the side of the cabinet near the refrigerator.", + "timestamps": { + "start": "00:28:17", + "end": "00:28:19" + } + } + ], + "question_timestamps": "00:43:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it back on the door shelf.", + "B": "Hang it on the cabinet front.", + "C": "You should put the apron back on the shelf next to the door, where you originally grabbed it from.", + "D": "You should put the apron back on the wall on the side of the cabinet far from the refrigerator.", + "E": "You should put the apron back on the wall on the side of the cabinet near the refrigerator." + }, + "answer_label": "E", + "answer_text": "You should put the apron back on the wall on the side of the cabinet near the refrigerator.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_10_T2", + "correct_answer": "You should put the apron back on the wall on the side of the cabinet near the refrigerator.", + "original_options": { + "A": "Put it back on the door shelf.", + "B": "Hang it on the cabinet front.", + "C": "You should put the apron back on the shelf next to the door, where you originally grabbed it from.", + "D": "You should put the apron back on the wall on the side of the cabinet far from the refrigerator.", + "E": "You should put the apron back on the wall on the side of the cabinet near the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 19 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_3", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chili stems have I cut so far?", + "answer": "So far, you have cut four chili stems.", + "question_chinese": "目前为止，我切了几个辣椒的蒂？", + "answer_chinese": "目前为止，你切了4个辣椒的蒂。", + "evidence": [ + { + "description": "I cut the stems of four chili peppers.", + "timestamps": { + "start": "00:07:54", + "end": "00:07:57" + } + } + ], + "question_timestamps": "00:08:01", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have cut three stems.", + "B": "You have cut five stems.", + "C": "So far, you have cut six chili stems in total at this point.", + "D": "So far, you have cut four chili stems.", + "E": "So far, you have cut seven chili stems up to this point." + }, + "answer_label": "D", + "answer_text": "So far, you have cut four chili stems.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_3", + "correct_answer": "So far, you have cut four chili stems.", + "original_options": { + "A": "You have cut three stems.", + "B": "You have cut five stems.", + "C": "So far, you have cut six chili stems in total at this point.", + "D": "So far, you have cut four chili stems.", + "E": "So far, you have cut seven chili stems up to this point." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 12 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_4", + "qid_num": 28, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chili stems have I cut so far?", + "answer": "So far, you have cut the stems of nine chili peppers.", + "question_chinese": "目前为止，我切了几个辣椒的蒂？", + "answer_chinese": "目前为止，你切了9个辣椒的蒂。", + "evidence": [ + { + "description": "I cut the stems of four chili peppers.", + "timestamps": { + "start": "00:07:55", + "end": "00:07:58" + } + }, + { + "description": "I cut the stems of three chili peppers.", + "timestamps": { + "start": "00:07:59", + "end": "00:08:04" + } + } + ], + "question_timestamps": "00:08:14", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "So far, you have cut the stems of nine chili peppers.", + "B": "You have cut five stems.", + "C": "You have cut twelve stems.", + "D": "So far, you have cut the stems of three chili peppers in total.", + "E": "So far, you have cut the stems of six chili peppers in total." + }, + "answer_label": "A", + "answer_text": "So far, you have cut the stems of nine chili peppers.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_4", + "correct_answer": "So far, you have cut the stems of nine chili peppers.", + "original_options": { + "A": "So far, you have cut the stems of nine chili peppers.", + "B": "You have cut five stems.", + "C": "You have cut twelve stems.", + "D": "So far, you have cut the stems of three chili peppers in total.", + "E": "So far, you have cut the stems of six chili peppers in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_11", + "qid_num": 29, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls have I taken so far?", + "answer": "So far, you have taken two bowls.", + "question_chinese": "目前为止，我拿了几个碗？", + "answer_chinese": "目前为止，你拿了两个碗。", + "evidence": [ + { + "description": "I took the first bowl.", + "timestamps": { + "start": "00:04:56", + "end": "00:04:58" + } + }, + { + "description": "I took the second bowl.", + "timestamps": { + "start": "00:11:01", + "end": "00:11:03" + } + } + ], + "question_timestamps": "00:11:05", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You took one bowl.", + "B": "You took three bowls.", + "C": "So far, you have taken four bowls in total up to this point.", + "D": "So far, you have taken five bowls up to this point overall.", + "E": "So far, you have taken two bowls." + }, + "answer_label": "E", + "answer_text": "So far, you have taken two bowls.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_11", + "correct_answer": "So far, you have taken two bowls.", + "original_options": { + "A": "You took one bowl.", + "B": "You took three bowls.", + "C": "So far, you have taken four bowls in total up to this point.", + "D": "So far, you have taken five bowls up to this point overall.", + "E": "So far, you have taken two bowls." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 12 + ] + }, + { + "video_uid": "7f9f75fd-a660-4635-8890-239c6ad82023", + "video_length": "00:43:38", + "qid": "7f9f75fd-a660-4635-8890-239c6ad82023_10", + "qid_num": 30, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls have I taken so far?", + "answer": "So far, you have taken three bowls.", + "question_chinese": "目前为止，我拿了几个碗？", + "answer_chinese": "目前为止，你拿了三个碗。", + "evidence": [ + { + "description": "I took the first bowl.", + "timestamps": { + "start": "00:04:56", + "end": "00:04:58" + } + }, + { + "description": "I took the second bowl.", + "timestamps": { + "start": "00:11:01", + "end": "00:11:08" + } + }, + { + "description": "I took the third bowl.", + "timestamps": { + "start": "00:12:38", + "end": "00:12:40" + } + } + ], + "question_timestamps": "00:12:43", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You took four bowls.", + "B": "You took five bowls.", + "C": "So far, you have taken three bowls.", + "D": "So far, you have taken six bowls in total up to this point.", + "E": "So far, you have taken seven bowls up to this point overall." + }, + "answer_label": "C", + "answer_text": "So far, you have taken three bowls.", + "q_id": "7f9f75fd-a660-4635-8890-239c6ad82023_10", + "correct_answer": "So far, you have taken three bowls.", + "original_options": { + "A": "You took four bowls.", + "B": "You took five bowls.", + "C": "So far, you have taken three bowls.", + "D": "So far, you have taken six bowls in total up to this point.", + "E": "So far, you have taken seven bowls up to this point overall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 12 + ] + } +] diff --git a/QAs/82efae54-9bea-4ee4-a3c9-4a6e034eb67d-mcq.json b/QAs/82efae54-9bea-4ee4-a3c9-4a6e034eb67d-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..99e82391701bb5f73614a4a07163eb67e57a9a87 --- /dev/null +++ b/QAs/82efae54-9bea-4ee4-a3c9-4a6e034eb67d-mcq.json @@ -0,0 +1,1377 @@ +[ + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_11", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I been ironing clothes?", + "answer": "No, I charged my phone.", + "question_chinese": "我一直在熨衣服吗？", + "answer_chinese": "不，我有给手机充电", + "evidence": [ + { + "description": "I am ironing clothes", + "timestamps": { + "start": "00:01:14", + "end": "00:01:15" + } + }, + { + "description": "I charged my phone", + "timestamps": { + "start": "00:01:52", + "end": "00:01:53" + } + } + ], + "question_timestamps": "00:01:54", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, I’ve been ironing clothes.", + "B": "No, I charged my phone." + }, + "answer_label": "B", + "answer_text": "No, I charged my phone.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_11", + "correct_answer": "No, I charged my phone.", + "original_options": { + "A": "Yes, I’ve been ironing clothes.", + "B": "No, I charged my phone." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_5_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I been ironing clothes the whole time?", + "answer": "No. Partway through, you charged your phone, and the phone was to your right.", + "question_chinese": "我一直在熨衣服吗？", + "answer_chinese": "不，中途你有给手机充电，手机在你的右边。", + "evidence": [ + { + "description": "I am ironing clothes", + "timestamps": { + "start": "00:01:14", + "end": "00:01:15" + } + }, + { + "description": "I charged my phone", + "timestamps": { + "start": "00:01:52", + "end": "00:01:53" + } + } + ], + "question_timestamps": "00:01:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. Partway through, you charged your phone, and the phone was to your right.", + "B": "Yes. You have been ironing clothes the whole time, with the iron on your right side." + }, + "answer_label": "A", + "answer_text": "No. Partway through, you charged your phone, and the phone was to your right.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_5_T1", + "correct_answer": "No. Partway through, you charged your phone, and the phone was to your right.", + "original_options": { + "A": "No. Partway through, you charged your phone, and the phone was to your right.", + "B": "Yes. You have been ironing clothes the whole time, with the iron on your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_12", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I been ironing clothes?", + "answer": "No, I charged my phone.", + "question_chinese": "我一直在熨衣服吗？", + "answer_chinese": "不，我有给手机充电", + "evidence": [ + { + "description": "I am ironing clothes", + "timestamps": { + "start": "00:01:14", + "end": "00:01:15" + } + }, + { + "description": "I charged my phone", + "timestamps": { + "start": "00:01:52", + "end": "00:01:53" + } + } + ], + "question_timestamps": "00:02:54", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, I charged my phone.", + "B": "Yes, you have been ironing clothes." + }, + "answer_label": "A", + "answer_text": "No, I charged my phone.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_12", + "correct_answer": "No, I charged my phone.", + "original_options": { + "A": "No, I charged my phone.", + "B": "Yes, you have been ironing clothes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_5_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I been ironing clothes the whole time?", + "answer": "No. Partway through, you were folding clothes, and the clothes were in front of you.", + "question_chinese": "我一直在熨衣服吗？", + "answer_chinese": "不，中途你有在折衣服，衣服在你的正前方。", + "evidence": [ + { + "description": "You were folding clothes halfway through.", + "timestamps": { + "start": "00:04:45", + "end": "00:04:49" + } + } + ], + "question_timestamps": "00:04:52", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. You have been ironing clothes the whole time, and the iron stayed in front of you.", + "B": "No. Partway through, you were folding clothes, and the clothes were in front of you." + }, + "answer_label": "B", + "answer_text": "No. Partway through, you were folding clothes, and the clothes were in front of you.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_5_T2", + "correct_answer": "No. Partway through, you were folding clothes, and the clothes were in front of you.", + "original_options": { + "A": "Yes. You have been ironing clothes the whole time, and the iron stayed in front of you.", + "B": "No. Partway through, you were folding clothes, and the clothes were in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_7", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Have I always kept the iron on the right-hand side?", + "answer": "No, it was in the front at first, then on the right-hand side.", + "question_chinese": "我一直把熨斗放在右手边吗？", + "answer_chinese": "不是，它一开始在前面，后来在右手边", + "evidence": [ + { + "description": "in the front", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + }, + { + "description": "on the right-hand side", + "timestamps": { + "start": "00:02:21", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:02:23", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, it was in the front at first, then on the right-hand side.", + "B": "Yes, you kept the iron on the right-hand side the whole time." + }, + "answer_label": "A", + "answer_text": "No, it was in the front at first, then on the right-hand side.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_7", + "correct_answer": "No, it was in the front at first, then on the right-hand side.", + "original_options": { + "A": "No, it was in the front at first, then on the right-hand side.", + "B": "Yes, you kept the iron on the right-hand side the whole time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_8", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Have I always kept the iron on my right-hand side?", + "answer": "No, it was in the front at first, then on the right-hand side.", + "question_chinese": "我一直把熨斗放在右手边吗？", + "answer_chinese": "不是，它一开始在前面，后来在右手边", + "evidence": [ + { + "description": "in the front", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + }, + { + "description": "on the right-hand side", + "timestamps": { + "start": "00:02:21", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, it stayed on your right-hand side the entire time.", + "B": "No, it was in the front at first, then on the right-hand side." + }, + "answer_label": "B", + "answer_text": "No, it was in the front at first, then on the right-hand side.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_8", + "correct_answer": "No, it was in the front at first, then on the right-hand side.", + "original_options": { + "A": "Yes, it stayed on your right-hand side the entire time.", + "B": "No, it was in the front at first, then on the right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_7_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothing have I folded?", + "answer": "Right now you have folded two items, and the clothes are to your left.", + "question_chinese": "我折了多少件衣服？", + "answer_chinese": "现在你折了两件，衣服在你的左边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:58", + "end": "00:01:59" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:07:19", + "end": "00:07:20" + } + } + ], + "question_timestamps": "00:07:22", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You folded one item.", + "B": "You folded three items.", + "C": "Right now you have folded two items, and the clothes are to your left.", + "D": "Right now you have folded four items, and the clothes are to your left.", + "E": "Right now you have folded zero items, and the clothes are to your left." + }, + "answer_label": "C", + "answer_text": "Right now you have folded two items, and the clothes are to your left.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_7_T2", + "correct_answer": "Right now you have folded two items, and the clothes are to your left.", + "original_options": { + "A": "You folded one item.", + "B": "You folded three items.", + "C": "Right now you have folded two items, and the clothes are to your left.", + "D": "Right now you have folded four items, and the clothes are to your left.", + "E": "Right now you have folded zero items, and the clothes are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_15", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothing did I fold?", + "answer": "nine.", + "question_chinese": "我折了多少件衣服？", + "answer_chinese": "9", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:58", + "end": "00:01:59" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + }, + { + "description": "the nineth one", + "timestamps": { + "start": "00:22:55", + "end": "00:22:56" + } + } + ], + "question_timestamps": "00:22:57", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "nine.", + "B": "eight.", + "C": "ten.", + "D": "You folded twelve pieces of clothing in total by the end of the sequence.", + "E": "You folded six items by the end of the sequence, not nine in total." + }, + "answer_label": "A", + "answer_text": "nine.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_15", + "correct_answer": "nine.", + "original_options": { + "A": "nine.", + "B": "eight.", + "C": "ten.", + "D": "You folded twelve pieces of clothing in total by the end of the sequence.", + "E": "You folded six items by the end of the sequence, not nine in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_7_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothing have I folded?", + "answer": "You folded nine pieces of clothing in total, and the nine items were to your left.", + "question_chinese": "我折了多少件衣服？", + "answer_chinese": "你一共折了九件衣服，九件衣服在你的左边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:58", + "end": "00:01:59" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + }, + { + "description": "the nineth one", + "timestamps": { + "start": "00:22:55", + "end": "00:22:56" + } + } + ], + "question_timestamps": "00:22:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You folded eight pieces total.", + "B": "You folded nine pieces of clothing in total, and the nine items were to your left.", + "C": "You folded ten pieces total.", + "D": "You folded seven pieces of clothing in total, and the seven items were to your left.", + "E": "You folded eleven pieces of clothing in total, and the eleven items were to your left." + }, + "answer_label": "B", + "answer_text": "You folded nine pieces of clothing in total, and the nine items were to your left.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_7_T1", + "correct_answer": "You folded nine pieces of clothing in total, and the nine items were to your left.", + "original_options": { + "A": "You folded eight pieces total.", + "B": "You folded nine pieces of clothing in total, and the nine items were to your left.", + "C": "You folded ten pieces total.", + "D": "You folded seven pieces of clothing in total, and the seven items were to your left.", + "E": "You folded eleven pieces of clothing in total, and the eleven items were to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_16", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothing did I fold?", + "answer": "Nine.", + "question_chinese": "我折了多少件衣服？", + "answer_chinese": "9", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:01:58", + "end": "00:01:59" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:07:19", + "end": "00:07:20" + } + }, + { + "description": "the nineth one", + "timestamps": { + "start": "00:22:55", + "end": "00:22:56" + } + } + ], + "question_timestamps": "00:23:15", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Eight.", + "B": "Nine.", + "C": "Ten.", + "D": "You folded seven pieces of clothing overall, not nine, by the time you finished.", + "E": "You folded eleven pieces of clothing in total, which is more than the number asked about." + }, + "answer_label": "B", + "answer_text": "Nine.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_16", + "correct_answer": "Nine.", + "original_options": { + "A": "Eight.", + "B": "Nine.", + "C": "Ten.", + "D": "You folded seven pieces of clothing overall, not nine, by the time you finished.", + "E": "You folded eleven pieces of clothing in total, which is more than the number asked about." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_6_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothing have I ironed?", + "answer": "Right now, you have ironed eight pieces of clothing, and the eight items are to your left.", + "question_chinese": "我熨了多少件衣服？", + "answer_chinese": "现在，你熨了八件衣服，八件衣服在你的左边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:52", + "end": "00:00:53" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:05:15", + "end": "00:05:16" + } + }, + { + "description": "the eighth one", + "timestamps": { + "start": "00:20:35", + "end": "00:20:59" + } + } + ], + "question_timestamps": "00:21:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You ironed seven pieces.", + "B": "Right now, you have ironed eight pieces of clothing, and the eight items are to your left.", + "C": "You ironed nine pieces.", + "D": "Right now, you have ironed six pieces of clothing, and the six items are to your left.", + "E": "Right now, you have ironed ten pieces of clothing, and the ten items are to your left." + }, + "answer_label": "B", + "answer_text": "Right now, you have ironed eight pieces of clothing, and the eight items are to your left.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_6_T2", + "correct_answer": "Right now, you have ironed eight pieces of clothing, and the eight items are to your left.", + "original_options": { + "A": "You ironed seven pieces.", + "B": "Right now, you have ironed eight pieces of clothing, and the eight items are to your left.", + "C": "You ironed nine pieces.", + "D": "Right now, you have ironed six pieces of clothing, and the six items are to your left.", + "E": "Right now, you have ironed ten pieces of clothing, and the ten items are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_13", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many clothes have I ironed?", + "answer": "nine.", + "question_chinese": "我熨了多少件衣服？", + "answer_chinese": "9", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:52", + "end": "00:00:53" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:05:15", + "end": "00:05:16" + } + }, + { + "description": "the nineth one", + "timestamps": { + "start": "00:22:58", + "end": "00:22:59" + } + } + ], + "question_timestamps": "00:23:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Ten.", + "B": "Five.", + "C": "nine.", + "D": "You have ironed three clothing items during this session, not nine in total.", + "E": "You pressed fifteen outfits in total, which is more than nine pieces overall today." + }, + "answer_label": "C", + "answer_text": "nine.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_13", + "correct_answer": "nine.", + "original_options": { + "A": "Ten.", + "B": "Five.", + "C": "nine.", + "D": "You have ironed three clothing items during this session, not nine in total.", + "E": "You pressed fifteen outfits in total, which is more than nine pieces overall today." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_6_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothing have I ironed?", + "answer": "You ironed nine pieces of clothing in total, and the nine items were to your left.", + "question_chinese": "我熨了多少件衣服？", + "answer_chinese": "你一共熨了九件衣服，九件衣服在你的左边。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:52", + "end": "00:00:53" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:05:15", + "end": "00:05:16" + } + }, + { + "description": "the nineth one", + "timestamps": { + "start": "00:22:58", + "end": "00:22:59" + } + } + ], + "question_timestamps": "00:23:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You ironed nine pieces of clothing in total, and the nine items were to your left.", + "B": "You ironed eight pieces total.", + "C": "You ironed ten pieces total.", + "D": "You ironed seven pieces of clothing in total, and the seven items were to your left.", + "E": "You ironed eleven pieces of clothing in total, and the eleven items were to your left." + }, + "answer_label": "A", + "answer_text": "You ironed nine pieces of clothing in total, and the nine items were to your left.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_6_T1", + "correct_answer": "You ironed nine pieces of clothing in total, and the nine items were to your left.", + "original_options": { + "A": "You ironed nine pieces of clothing in total, and the nine items were to your left.", + "B": "You ironed eight pieces total.", + "C": "You ironed ten pieces total.", + "D": "You ironed seven pieces of clothing in total, and the seven items were to your left.", + "E": "You ironed eleven pieces of clothing in total, and the eleven items were to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_14", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many clothes have I ironed?", + "answer": "nine.", + "question_chinese": "我熨了多少件衣服？", + "answer_chinese": "9", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:52", + "end": "00:00:53" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:05:15", + "end": "00:05:16" + } + }, + { + "description": "the nineth one", + "timestamps": { + "start": "00:22:58", + "end": "00:22:59" + } + } + ], + "question_timestamps": "00:23:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Five pieces.", + "C": "You ironed four garments in total, which is less than nine items overall.", + "D": "nine.", + "E": "You ironed twelve shirts altogether, which is more than nine items overall." + }, + "answer_label": "D", + "answer_text": "nine.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_14", + "correct_answer": "nine.", + "original_options": { + "A": "Two.", + "B": "Five pieces.", + "C": "You ironed four garments in total, which is less than nine items overall.", + "D": "nine.", + "E": "You ironed twelve shirts altogether, which is more than nine items overall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 12 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_2_T2", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is my iron?", + "answer": "Your iron is in front of you.", + "question_chinese": "我的熨斗在哪里？", + "answer_chinese": "你的熨斗在你的正前方。", + "evidence": [ + { + "description": "Your iron is in front of you.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Your iron is in front of you.", + "B": "Your iron is to your left." + }, + "answer_label": "A", + "answer_text": "Your iron is in front of you.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_2_T2", + "correct_answer": "Your iron is in front of you.", + "original_options": { + "A": "Your iron is in front of you.", + "B": "Your iron is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_2_T1", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is my iron?", + "answer": "Your iron is to your right.", + "question_chinese": "我的熨斗在哪里？", + "answer_chinese": "你的熨斗在你的右边。", + "evidence": [ + { + "description": "Your iron is on your right.", + "timestamps": { + "start": "00:02:21", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:02:24", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Your iron is in front of you.", + "B": "Your iron is to your right." + }, + "answer_label": "B", + "answer_text": "Your iron is to your right.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_2_T1", + "correct_answer": "Your iron is to your right.", + "original_options": { + "A": "Your iron is in front of you.", + "B": "Your iron is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_4_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my blue water bottle?", + "answer": "It is to your front-right.", + "question_chinese": "我的蓝色水瓶在哪里？", + "answer_chinese": "在你的右前方", + "evidence": [ + { + "description": "My front right", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to your front-right.", + "B": "It is to your front-left.", + "C": "It is to your back-right.", + "D": "It is directly in front of you, not angled to the right side.", + "E": "It is behind you on the right, closer to your back than your front." + }, + "answer_label": "A", + "answer_text": "It is to your front-right.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_4_T1", + "correct_answer": "It is to your front-right.", + "original_options": { + "A": "It is to your front-right.", + "B": "It is to your front-left.", + "C": "It is to your back-right.", + "D": "It is directly in front of you, not angled to the right side.", + "E": "It is behind you on the right, closer to your back than your front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_4_T2", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my blue water bottle?", + "answer": "It is in front of you.", + "question_chinese": "我的蓝色水瓶在哪里？", + "answer_chinese": "在你的前面。", + "evidence": [ + { + "description": "In front of you.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is in front of you.", + "C": "It is to your left.", + "D": "It is off to your right side, not straight ahead in your view.", + "E": "It is above you rather than positioned out in front at your eye level." + }, + "answer_label": "B", + "answer_text": "It is in front of you.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_4_T2", + "correct_answer": "It is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is in front of you.", + "C": "It is to your left.", + "D": "It is off to your right side, not straight ahead in your view.", + "E": "It is above you rather than positioned out in front at your eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 14 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_1_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is in front of me?", + "answer": "There’s an iron in front of you.", + "question_chinese": "我面前的是什么？", + "answer_chinese": "你的面前是一个熨斗", + "evidence": [ + { + "description": "Iron", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There’s an iron behind you.", + "B": "There’s an iron to your left.", + "C": "There’s an iron to your right, not positioned directly in front of you.", + "D": "There’s an iron in front of you.", + "E": "There’s an iron above you, rather than sitting out in front on the surface." + }, + "answer_label": "D", + "answer_text": "There’s an iron in front of you.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_1_T2", + "correct_answer": "There’s an iron in front of you.", + "original_options": { + "A": "There’s an iron behind you.", + "B": "There’s an iron to your left.", + "C": "There’s an iron to your right, not positioned directly in front of you.", + "D": "There’s an iron in front of you.", + "E": "There’s an iron above you, rather than sitting out in front on the surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 15 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_2", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is in front of me?", + "answer": "Iron.", + "question_chinese": "我面前的是什么？", + "answer_chinese": "熨斗", + "evidence": [ + { + "description": "Iron", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Cane.", + "B": "Iron.", + "C": "A wooden stick.", + "D": "A piece of clothing is in front of you, not a metal tool.", + "E": "You placed a green rod in front of you instead of an iron." + }, + "answer_label": "B", + "answer_text": "Iron.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_2", + "correct_answer": "Iron.", + "original_options": { + "A": "Cane.", + "B": "Iron.", + "C": "A wooden stick.", + "D": "A piece of clothing is in front of you, not a metal tool.", + "E": "You placed a green rod in front of you instead of an iron." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 3, + 13, + 13 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_1_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is in front of me?", + "answer": "There’s a black dress in front of you.", + "question_chinese": "我面前的是什么？", + "answer_chinese": "你的面前是一件黑色的裙子。", + "evidence": [ + { + "description": "In front of you is a black skirt.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There’s a black dress behind you.", + "B": "There’s a black dress to your left.", + "C": "There’s a black dress in front of you.", + "D": "There’s a black dress to your right, not directly in front of you.", + "E": "There’s a black dress above you rather than placed in front at your level." + }, + "answer_label": "C", + "answer_text": "There’s a black dress in front of you.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_1_T1", + "correct_answer": "There’s a black dress in front of you.", + "original_options": { + "A": "There’s a black dress behind you.", + "B": "There’s a black dress to your left.", + "C": "There’s a black dress in front of you.", + "D": "There’s a black dress to your right, not directly in front of you.", + "E": "There’s a black dress above you rather than placed in front at your level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 14, + 15 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_3_T1", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What color clothes am I ironing right now?", + "answer": "You’re ironing black clothes, and the black clothes are in front of you.", + "question_chinese": "现在，我在熨什么颜色的衣服？", + "answer_chinese": "你在熨黑色的衣服，黑色衣服在你的正前方。", + "evidence": [ + { + "description": "You are ironing black clothes.", + "timestamps": { + "start": "00:02:29", + "end": "00:02:32" + } + } + ], + "question_timestamps": "00:02:38", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You’re ironing black clothes, and the black clothes are in front of you.", + "B": "You are ironing a dark green piece of clothing, and it is in front of you." + }, + "answer_label": "A", + "answer_text": "You’re ironing black clothes, and the black clothes are in front of you.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_3_T1", + "correct_answer": "You’re ironing black clothes, and the black clothes are in front of you.", + "original_options": { + "A": "You’re ironing black clothes, and the black clothes are in front of you.", + "B": "You are ironing a dark green piece of clothing, and it is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_3_T2", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What color clothes am I ironing right now?", + "answer": "You are ironing a dark green piece of clothing, and it is in front of you.", + "question_chinese": "现在，我在熨什么颜色的衣服？", + "answer_chinese": "你在熨暗绿色的衣服，暗绿色的衣服在你的正前方。", + "evidence": [ + { + "description": "You are wearing dark green clothes.", + "timestamps": { + "start": "00:05:33", + "end": "00:05:35" + } + } + ], + "question_timestamps": "00:05:38", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You’re ironing black clothes, and the black clothes are in front of you.", + "B": "You are ironing a dark green piece of clothing, and it is in front of you." + }, + "answer_label": "B", + "answer_text": "You are ironing a dark green piece of clothing, and it is in front of you.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_3_T2", + "correct_answer": "You are ironing a dark green piece of clothing, and it is in front of you.", + "original_options": { + "A": "You’re ironing black clothes, and the black clothes are in front of you.", + "B": "You are ironing a dark green piece of clothing, and it is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_9", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the blue water bottle?", + "answer": "My front right.", + "question_chinese": "蓝色水瓶在哪里？", + "answer_chinese": "我的右前方", + "evidence": [ + { + "description": "My front right", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "My front right.", + "B": "My front left.", + "C": "My back right.", + "D": "It’s straight in front of me, not shifted toward my right side.", + "E": "It’s behind me on the left, not up ahead on my front-right." + }, + "answer_label": "A", + "answer_text": "My front right.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_9", + "correct_answer": "My front right.", + "original_options": { + "A": "My front right.", + "B": "My front left.", + "C": "My back right.", + "D": "It’s straight in front of me, not shifted toward my right side.", + "E": "It’s behind me on the left, not up ahead on my front-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 13 + ] + }, + { + "video_uid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d", + "video_length": "00:23:24", + "qid": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_10", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the blue water bottle?", + "answer": "My front right.", + "question_chinese": "蓝色水瓶在哪里？", + "answer_chinese": "我的右前方", + "evidence": [ + { + "description": "My front right", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "My front left.", + "B": "My back right.", + "C": "It is behind you on the right side, slightly off to the back.", + "D": "It is in front of you but closer to your left than your right.", + "E": "My front right." + }, + "answer_label": "E", + "answer_text": "My front right.", + "q_id": "82efae54-9bea-4ee4-a3c9-4a6e034eb67d_10", + "correct_answer": "My front right.", + "original_options": { + "A": "My front left.", + "B": "My back right.", + "C": "It is behind you on the right side, slightly off to the back.", + "D": "It is in front of you but closer to your left than your right.", + "E": "My front right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 14 + ] + } +] diff --git a/QAs/863011c8-ebf7-4900-87d1-3c7930f95dab-001-mcq.json b/QAs/863011c8-ebf7-4900-87d1-3c7930f95dab-001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..13024bdea508ab759c7bbfe4798505006ed10362 --- /dev/null +++ b/QAs/863011c8-ebf7-4900-87d1-3c7930f95dab-001-mcq.json @@ -0,0 +1,818 @@ +[ + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001", + "video_length": "00:20:02", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_3", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I walked from the concrete ground to the soil ground, did I keep walking on the soil ground all the time?", + "answer": "Yes.", + "question_chinese": "我从水泥地走到土地上，就一直走在土地上吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "Concrete ground", + "timestamps": { + "start": "00:02:24", + "end": "00:02:25" + } + }, + { + "description": "Soil ground", + "timestamps": { + "start": "00:02:28", + "end": "00:02:29" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_3", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 0, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001", + "video_length": "00:20:02", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_4", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I walked from the concrete ground to the soil ground, did I keep walking on the soil ground all the time?", + "answer": "No.", + "question_chinese": "我从水泥地走到土地上，就一直走在土地上吗？", + "answer_chinese": "不是", + "evidence": [ + { + "description": "Concrete ground", + "timestamps": { + "start": "00:02:24", + "end": "00:02:25" + } + }, + { + "description": "Soil ground", + "timestamps": { + "start": "00:02:28", + "end": "00:02:29" + } + }, + { + "description": "Concrete ground", + "timestamps": { + "start": "00:12:00", + "end": "00:12:01" + } + } + ], + "question_timestamps": "00:18:00", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_4", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001", + "video_length": "00:20:02", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I make when I walked along the road just now?", + "answer": "Two.", + "question_chinese": "我刚才沿着路走，转了多少个弯？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "You made one turn in total as you followed the road forward.", + "E": "You made three turns altogether while walking along the road just now." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_1", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "You made one turn in total as you followed the road forward.", + "E": "You made three turns altogether while walking along the road just now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 0, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001", + "video_length": "00:20:02", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I make when I walked along the road just now?", + "answer": "Three.", + "question_chinese": "我刚才沿着路走，转了多少个弯？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:01:58", + "end": "00:01:59" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "You made two turns total while walking along the road just now.", + "D": "Three.", + "E": "You made four turns altogether as you followed the road to your current position." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_2", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "You made two turns total while walking along the road just now.", + "D": "Three.", + "E": "You made four turns altogether as you followed the road to your current position." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 0, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001", + "video_length": "00:20:02", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_5", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I check my phone when I was walking along the road just now?", + "answer": "One.", + "question_chinese": "我刚才沿着路边走时，看了多少次手机？", + "answer_chinese": "1", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Three times.", + "C": "You checked your phone two times while you were walking along the road.", + "D": "You checked your phone three times during the walk along the road just now.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_5", + "correct_answer": "One.", + "original_options": { + "A": "Two times.", + "B": "Three times.", + "C": "You checked your phone two times while you were walking along the road.", + "D": "You checked your phone three times during the walk along the road just now.", + "E": "One." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 0, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001", + "video_length": "00:20:02", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_6", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I check my phone when I was walking along the road just now?", + "answer": "Two.", + "question_chinese": "我刚才沿着路边走时，看了多少次手机？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first time", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "the second time", + "timestamps": { + "start": "00:02:58", + "end": "00:02:59" + } + } + ], + "question_timestamps": "00:13:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You checked your phone one time while walking along the road just now.", + "E": "You checked your phone three times during your walk along the road." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_6", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You checked your phone one time while walking along the road just now.", + "E": "You checked your phone three times during your walk along the road." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001", + "video_length": "00:20:02", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_10", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position from the path paved with red and white tiles?", + "answer": "Turn right, walk along the path for twenty meters, turn right again and walk 100 meters.", + "question_chinese": "我是怎么从红白瓦砖小路走到当前位置的？", + "answer_chinese": "右转，沿着小路走20米，右转走100米", + "evidence": [ + { + "description": "Turn right", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "walk along the path for 20 meters", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "turn right again and walk 100 meters", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, then walk 100 meters.", + "B": "Turn right, then turn left.", + "C": "Turn right, walk along the path for twenty meters, turn right again and walk 100 meters.", + "D": "Turn right, walk along the path for twenty meters, turn left, then continue walking for 100 meters.", + "E": "Turn right, walk along the path for thirty meters, turn right again, and walk another 100 meters." + }, + "answer_label": "C", + "answer_text": "Turn right, walk along the path for twenty meters, turn right again and walk 100 meters.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_10", + "correct_answer": "Turn right, walk along the path for twenty meters, turn right again and walk 100 meters.", + "original_options": { + "A": "Turn left, then walk 100 meters.", + "B": "Turn right, then turn left.", + "C": "Turn right, walk along the path for twenty meters, turn right again and walk 100 meters.", + "D": "Turn right, walk along the path for twenty meters, turn left, then continue walking for 100 meters.", + "E": "Turn right, walk along the path for thirty meters, turn right again, and walk another 100 meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 2, + "D": 2, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 17 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001", + "video_length": "00:20:02", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_9", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position from the path paved with red and white tiles?", + "answer": "Turn right, walk along the path for twenty meters, turn right again and walk 100 meters, then turn right once more and walk five meters.", + "question_chinese": "我是怎么从红白瓦砖小路走到当前位置的？", + "answer_chinese": "右转，沿着小路走20米，右转走100米，再右转走5米", + "evidence": [ + { + "description": "Turn right", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "walk along the path for 20 meters", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "turn right again and walk 100 meters", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + }, + { + "description": "turn right once more and walk 5 meters.", + "timestamps": { + "start": "00:01:57", + "end": "00:01:58" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, then walk straight.", + "B": "Go straight, then turn right.", + "C": "Turn right, walk twenty meters, turn left and walk 100 meters, then turn right and walk five meters.", + "D": "Turn right, walk along the path for twenty meters, turn right again and walk 100 meters, then turn right once more and walk five meters.", + "E": "Turn right, walk twenty meters, turn right again for eighty meters, then turn right once more and walk five meters." + }, + "answer_label": "D", + "answer_text": "Turn right, walk along the path for twenty meters, turn right again and walk 100 meters, then turn right once more and walk five meters.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_9", + "correct_answer": "Turn right, walk along the path for twenty meters, turn right again and walk 100 meters, then turn right once more and walk five meters.", + "original_options": { + "A": "Turn left, then walk straight.", + "B": "Go straight, then turn right.", + "C": "Turn right, walk twenty meters, turn left and walk 100 meters, then turn right and walk five meters.", + "D": "Turn right, walk along the path for twenty meters, turn right again and walk 100 meters, then turn right once more and walk five meters.", + "E": "Turn right, walk twenty meters, turn right again for eighty meters, then turn right once more and walk five meters." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 1, + "D": 2, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 20 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001", + "video_length": "00:20:02", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_12", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Did any vehicle pass by me while I was walking along the road?", + "answer": "No.", + "question_chinese": "我沿着路走时，有车经过我吗？", + "answer_chinese": "没有", + "evidence": [ + { + "description": "No car passed by me.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_12", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 2, + "B": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001", + "video_length": "00:20:02", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_11", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Did any vehicle pass by me while I was walking along the road?", + "answer": "Yes.", + "question_chinese": "我沿着路走时，有车经过我吗？", + "answer_chinese": "有", + "evidence": [ + { + "description": "A car passed by me.", + "timestamps": { + "start": "00:01:58", + "end": "00:01:59" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-001_11", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 2, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/863011c8-ebf7-4900-87d1-3c7930f95dab-002-mcq.json b/QAs/863011c8-ebf7-4900-87d1-3c7930f95dab-002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b3a98275f7611eca6ac881b3c982b5a635fa1a14 --- /dev/null +++ b/QAs/863011c8-ebf7-4900-87d1-3c7930f95dab-002-mcq.json @@ -0,0 +1,906 @@ +[ + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_3", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Did a white car pass by me just now?", + "answer": "No.", + "question_chinese": "刚才有一辆白色的车经过我了吗？", + "answer_chinese": "不是", + "evidence": [ + { + "description": "No white car passed by me.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_3", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_4", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Did a white car pass by me just now?", + "answer": "Yes.", + "question_chinese": "刚才有一辆白色的车经过我了吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "A white car passed by me just now.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_4", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_15", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many people passed by me when I was walking on the road just now?", + "answer": "One.", + "question_chinese": "刚才走在路上有多少个人经过了我？", + "answer_chinese": "1", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:07:34", + "end": "00:07:35" + } + } + ], + "question_timestamps": "00:08:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "Four people passed by you while you were walking along the road.", + "E": "Five people passed by you on the road during that walk just now." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_15", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "Four people passed by you while you were walking along the road.", + "E": "Five people passed by you on the road during that walk just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_16", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many people passed by me when I was walking on the road just now?", + "answer": "Two.", + "question_chinese": "刚才走在路上有多少个人经过了我？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:07:34", + "end": "00:07:35" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:08:46", + "end": "00:08:47" + } + } + ], + "question_timestamps": "00:13:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four people passed by you while you were walking on the road.", + "E": "Five people passed by you during that stretch of walking on the road." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_16", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four people passed by you while you were walking on the road.", + "E": "Five people passed by you during that stretch of walking on the road." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_13", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I make when I walked along the road just now?", + "answer": "Two.", + "question_chinese": "我刚才沿着路走，转了多少个弯？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:06:45", + "end": "00:06:46" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:11:51", + "end": "00:11:52" + } + } + ], + "question_timestamps": "00:17:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four turns in total while you were walking along the road just now.", + "D": "Five turns altogether while you were walking along the road just now.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_13", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four turns in total while you were walking along the road just now.", + "D": "Five turns altogether while you were walking along the road just now.", + "E": "Two." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 0, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_14", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I make when I walked along the road just now?", + "answer": "Three.", + "question_chinese": "我刚才沿着路走，转了多少个弯？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:06:45", + "end": "00:06:46" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:11:51", + "end": "00:11:52" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:17:25", + "end": "00:17:26" + } + } + ], + "question_timestamps": "00:20:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "One.", + "C": "Two.", + "D": "Four turns in total along the road you just walked, with an extra bend near the end.", + "E": "Five turns overall as you followed the road, including two quick turns close together midway." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_14", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "One.", + "C": "Two.", + "D": "Four turns in total along the road you just walked, with an extra bend near the end.", + "E": "Five turns overall as you followed the road, including two quick turns close together midway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 0, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 15 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_11", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Has the section I just passed through been under construction all the time?", + "answer": "No.", + "question_chinese": "我刚才经过的路段一直在施工吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "Under construction", + "timestamps": { + "start": "00:01:09", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_11", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 2, + "B": 3 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_12", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Has the section I just passed through been under construction all the time?", + "answer": "No.", + "question_chinese": "我刚才经过的路段一直在施工吗？", + "answer_chinese": "不是", + "evidence": [ + { + "description": "Under construction", + "timestamps": { + "start": "00:01:09", + "end": "00:01:10" + } + }, + { + "description": "Not under construction", + "timestamps": { + "start": "00:09:00", + "end": "00:09:01" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_12", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 3, + "B": 3 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_9", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Has the area next to the asphalt road always been a sand and soil road?", + "answer": "Yes.", + "question_chinese": "沥青路旁边一直是沙土铺的路吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "A road paved with sand and soil", + "timestamps": { + "start": "00:05:09", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:06:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_9", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 2, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_10", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Has the area next to the asphalt road always been a sand and soil road?", + "answer": "No.", + "question_chinese": "沥青路旁边一直是沙土铺的路吗？", + "answer_chinese": "不是", + "evidence": [ + { + "description": "A road paved with sand and soil", + "timestamps": { + "start": "00:05:09", + "end": "00:05:10" + } + }, + { + "description": "A road paved with stones", + "timestamps": { + "start": "00:06:45", + "end": "00:06:46" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_10", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 2, + "B": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_5", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What’s on the side of the road?", + "answer": "Street lamps,bushes, trees.", + "question_chinese": "路边有什么？", + "answer_chinese": "路灯，灌木丛和树", + "evidence": [ + { + "description": "street lamps.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "bushes, trees", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Traffic signs and bushes.", + "B": "Sidewalk benches and trees.", + "C": "Traffic signals line the road, with patches of grass and a few low barriers beside them.", + "D": "Street lamps,bushes, trees.", + "E": "Orange traffic barrels sit along the roadside, with shrubs and some climbing vines nearby." + }, + "answer_label": "D", + "answer_text": "Street lamps,bushes, trees.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_5", + "correct_answer": "Street lamps,bushes, trees.", + "original_options": { + "A": "Traffic signs and bushes.", + "B": "Sidewalk benches and trees.", + "C": "Traffic signals line the road, with patches of grass and a few low barriers beside them.", + "D": "Street lamps,bushes, trees.", + "E": "Orange traffic barrels sit along the roadside, with shrubs and some climbing vines nearby." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 0, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 14 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002", + "video_length": "00:20:00", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_6", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What’s on the side of the road?", + "answer": "Street lamps, bushes, trees, and roadblocks.", + "question_chinese": "路边有什么？", + "answer_chinese": "路灯，灌木丛和树，路障 ", + "evidence": [ + { + "description": "street lamps.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "bushes, trees", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "Roadblocks", + "timestamps": { + "start": "00:01:21", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Parking meters and bushes.", + "B": "Street lamps, bushes, trees, and roadblocks.", + "C": "Public benches and trees.", + "D": "Street lamps line the road, but you mainly see bushes and trees with no roadblocks present.", + "E": "You pass bushes and trees beside the road, with cones and a few mailboxes instead of roadblocks." + }, + "answer_label": "B", + "answer_text": "Street lamps, bushes, trees, and roadblocks.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-002_6", + "correct_answer": "Street lamps, bushes, trees, and roadblocks.", + "original_options": { + "A": "Parking meters and bushes.", + "B": "Street lamps, bushes, trees, and roadblocks.", + "C": "Public benches and trees.", + "D": "Street lamps line the road, but you mainly see bushes and trees with no roadblocks present.", + "E": "You pass bushes and trees beside the road, with cones and a few mailboxes instead of roadblocks." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "bad", + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 0, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + } +] diff --git a/QAs/863011c8-ebf7-4900-87d1-3c7930f95dab-003-mcq.json b/QAs/863011c8-ebf7-4900-87d1-3c7930f95dab-003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a251813cca8a30153ce9f090df0ba22d4d368cdb --- /dev/null +++ b/QAs/863011c8-ebf7-4900-87d1-3c7930f95dab-003-mcq.json @@ -0,0 +1,1533 @@ +[ + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_11", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get the scooter when I come out of the gate?", + "answer": "Come out of the gate, walk forward two meters, turn left and walk five meters.", + "question_chinese": "从大门出来后，我做了什么？", + "answer_chinese": "出门向前走2米，左转向前走5米", + "evidence": [ + { + "description": "Come out of the gate, walk forward 2 meters", + "timestamps": { + "start": "00:16:20", + "end": "00:16:21" + } + }, + { + "description": "turn left and walk 5 meters", + "timestamps": { + "start": "00:17:13", + "end": "00:17:14" + } + } + ], + "question_timestamps": "00:17:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Come out of the gate, walk forward two meters, turn left and walk five meters.", + "B": "Turn right after two meters.", + "C": "Go straight about five meters.", + "D": "Come out of the gate, walk forward five meters, then turn left and walk two meters to the scooter.", + "E": "Exit the gate, walk forward two meters, turn right, and continue five meters until you reach the scooter." + }, + "answer_label": "A", + "answer_text": "Come out of the gate, walk forward two meters, turn left and walk five meters.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_11", + "correct_answer": "Come out of the gate, walk forward two meters, turn left and walk five meters.", + "original_options": { + "A": "Come out of the gate, walk forward two meters, turn left and walk five meters.", + "B": "Turn right after two meters.", + "C": "Go straight about five meters.", + "D": "Come out of the gate, walk forward five meters, then turn left and walk two meters to the scooter.", + "E": "Exit the gate, walk forward two meters, turn right, and continue five meters until you reach the scooter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 1, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 18 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_12", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get the scooter when I come out of the gate?", + "answer": "Come out of the gate, walk forward two meters, turn left and walk five meters, open the black door to get the scooter.", + "question_chinese": "从大门出来后，我做了什么？", + "answer_chinese": "出门向前走2米，左转向前走5米，打开黑色的门拿滑板车", + "evidence": [ + { + "description": "Come out of the gate, walk forward 2 meters", + "timestamps": { + "start": "00:16:20", + "end": "00:16:21" + } + }, + { + "description": "turn left and walk 5 meters", + "timestamps": { + "start": "00:17:13", + "end": "00:17:14" + } + }, + { + "description": "open the black door to get the scooter.", + "timestamps": { + "start": "00:17:27", + "end": "00:17:28" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Open the white door instead.", + "B": "Turn right after two meters.", + "C": "Come out of the gate, walk forward two meters, turn left and walk five meters, open the black door to get the scooter.", + "D": "Come out of the gate, walk forward two meters, turn left three meters, then open the green door for the scooter.", + "E": "Exit the gate, walk forward five meters, turn left five meters, and open a red door to get the scooter." + }, + "answer_label": "C", + "answer_text": "Come out of the gate, walk forward two meters, turn left and walk five meters, open the black door to get the scooter.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_12", + "correct_answer": "Come out of the gate, walk forward two meters, turn left and walk five meters, open the black door to get the scooter.", + "original_options": { + "A": "Open the white door instead.", + "B": "Turn right after two meters.", + "C": "Come out of the gate, walk forward two meters, turn left and walk five meters, open the black door to get the scooter.", + "D": "Come out of the gate, walk forward two meters, turn left three meters, then open the green door for the scooter.", + "E": "Exit the gate, walk forward five meters, turn left five meters, and open a red door to get the scooter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 20 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_5", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What were the people on the road doing just now?", + "answer": "Running and cycling.", + "question_chinese": "刚才马路上的人在干什么？", + "answer_chinese": "跑步", + "evidence": [ + { + "description": "Running", + "timestamps": { + "start": "00:07:15", + "end": "00:07:16" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walking and skating.", + "B": "Jogging and walking.", + "C": "They were walking along the road while another person skated past you in the same direction.", + "D": "Running and cycling.", + "E": "One person was jogging slowly, and another was just walking beside the road rather than cycling." + }, + "answer_label": "D", + "answer_text": "Running and cycling.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_5", + "correct_answer": "Running and cycling.", + "original_options": { + "A": "Walking and skating.", + "B": "Jogging and walking.", + "C": "They were walking along the road while another person skated past you in the same direction.", + "D": "Running and cycling.", + "E": "One person was jogging slowly, and another was just walking beside the road rather than cycling." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 0, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_6", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What were the people on the road doing just now?", + "answer": "Running and cycling.", + "question_chinese": "刚才马路上的人在干什么？", + "answer_chinese": "跑步和骑行", + "evidence": [ + { + "description": "Running", + "timestamps": { + "start": "00:07:15", + "end": "00:07:16" + } + }, + { + "description": "cycling", + "timestamps": { + "start": "00:09:07", + "end": "00:09:08" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Running and cycling.", + "B": "Jogging and walking.", + "C": "Running and skating.", + "D": "You saw people biking, but the others were walking casually rather than running on the road.", + "E": "Someone was skating while another person rode a bike, so it wasn’t running and cycling together." + }, + "answer_label": "A", + "answer_text": "Running and cycling.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_6", + "correct_answer": "Running and cycling.", + "original_options": { + "A": "Running and cycling.", + "B": "Jogging and walking.", + "C": "Running and skating.", + "D": "You saw people biking, but the others were walking casually rather than running on the road.", + "E": "Someone was skating while another person rode a bike, so it wasn’t running and cycling together." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 0, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_15", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the little boy playing with his scooter on the road?", + "answer": "No.", + "question_chinese": "小男孩在路上玩滑板车吗？", + "answer_chinese": "不是", + "evidence": [ + { + "description": "walking", + "timestamps": { + "start": "00:15:20", + "end": "00:15:21" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, he is playing with it on the road.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_15", + "correct_answer": "No.", + "original_options": { + "A": "Yes, he is playing with it on the road.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_16", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the little boy playing with his scooter on the road?", + "answer": "Yes.", + "question_chinese": "小男孩在路上玩滑板车吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "A scooter", + "timestamps": { + "start": "00:22:00", + "end": "00:22:01" + } + } + ], + "question_timestamps": "00:28:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No, he isn't playing with his scooter on the road." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_16", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No, he isn't playing with his scooter on the road." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 2, + "B": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_7", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position when I was standing in front of the door?", + "answer": "Open the door, walk forward five meters, turn right and walk ten meters.", + "question_chinese": "当我站在门前，怎么走到当前位置？", + "answer_chinese": "开门向前走5米，右转走10米", + "evidence": [ + { + "description": "Open the door, walk forward 5 meters", + "timestamps": { + "start": "00:15:45", + "end": "00:15:46" + } + }, + { + "description": "turn right and walk 10 meters", + "timestamps": { + "start": "00:15:48", + "end": "00:15:49" + } + } + ], + "question_timestamps": "00:15:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left and walk ten meters.", + "B": "Walk forward three meters first.", + "C": "Open the door, walk forward five meters, turn right and walk ten meters.", + "D": "Open the door, walk forward five meters, turn left, and continue walking ten meters to reach your spot.", + "E": "Open the door, walk forward five meters, turn right, then walk fifteen meters instead of ten to your position." + }, + "answer_label": "C", + "answer_text": "Open the door, walk forward five meters, turn right and walk ten meters.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_7", + "correct_answer": "Open the door, walk forward five meters, turn right and walk ten meters.", + "original_options": { + "A": "Turn left and walk ten meters.", + "B": "Walk forward three meters first.", + "C": "Open the door, walk forward five meters, turn right and walk ten meters.", + "D": "Open the door, walk forward five meters, turn left, and continue walking ten meters to reach your spot.", + "E": "Open the door, walk forward five meters, turn right, then walk fifteen meters instead of ten to your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 19 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_8", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position when I was standing in front of the door?", + "answer": "Open the door, walk forward five meters, turn right and walk ten meters, turn left and walk three meters.", + "question_chinese": "当我站在门前，怎么走到当前位置？", + "answer_chinese": "开门向前走5米，右转走10米，左转走3米", + "evidence": [ + { + "description": "Open the door, walk forward 5 meters", + "timestamps": { + "start": "00:15:45", + "end": "00:15:46" + } + }, + { + "description": "turn right and walk 10 meters", + "timestamps": { + "start": "00:15:49", + "end": "00:15:50" + } + }, + { + "description": "turn left and walk 3 meters", + "timestamps": { + "start": "00:15:54", + "end": "00:15:55" + } + } + ], + "question_timestamps": "00:25:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Open the door and walk straight ahead.", + "B": "Open the door, walk forward five meters, turn right and walk ten meters, turn left and walk three meters.", + "C": "Open the door, turn left, then walk forward.", + "D": "Open the door, walk forward five meters, turn left and walk ten meters, then turn right and walk three meters.", + "E": "Open the door, walk forward three meters, turn right and walk ten meters, then turn left and walk three meters." + }, + "answer_label": "B", + "answer_text": "Open the door, walk forward five meters, turn right and walk ten meters, turn left and walk three meters.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_8", + "correct_answer": "Open the door, walk forward five meters, turn right and walk ten meters, turn left and walk three meters.", + "original_options": { + "A": "Open the door and walk straight ahead.", + "B": "Open the door, walk forward five meters, turn right and walk ten meters, turn left and walk three meters.", + "C": "Open the door, turn left, then walk forward.", + "D": "Open the door, walk forward five meters, turn left and walk ten meters, then turn right and walk three meters.", + "E": "Open the door, walk forward three meters, turn right and walk ten meters, then turn left and walk three meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 2, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 20, + 20 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_3", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I been running on the asphalt road all the time?", + "answer": "Yes.", + "question_chinese": "我一直在沥青路上跑步吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "Running", + "timestamps": { + "start": "00:08:19", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:08:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_3", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 0, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_4", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I been running on the asphalt road all the time?", + "answer": "No.", + "question_chinese": "我一直在沥青路上跑步吗？", + "answer_chinese": "不是", + "evidence": [ + { + "description": "Walking", + "timestamps": { + "start": "00:21:00", + "end": "00:21:01" + } + } + ], + "question_timestamps": "00:23:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_4", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What am I doing by the road now?", + "answer": "Running.", + "question_chinese": "我现在在路边干什么？", + "answer_chinese": "跑步", + "evidence": [ + { + "description": "Running", + "timestamps": { + "start": "00:02:00", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walking.", + "B": "Standing still.", + "C": "You are riding along the road instead of running right now, moving forward at a steady pace.", + "D": "You are sitting by the roadside rather than running, staying in place and not moving down the road.", + "E": "Running." + }, + "answer_label": "E", + "answer_text": "Running.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_1", + "correct_answer": "Running.", + "original_options": { + "A": "Walking.", + "B": "Standing still.", + "C": "You are riding along the road instead of running right now, moving forward at a steady pace.", + "D": "You are sitting by the roadside rather than running, staying in place and not moving down the road.", + "E": "Running." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 0, + "D": 0, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 17, + 18 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What am I doing by the road now?", + "answer": "Walking.", + "question_chinese": "我现在在路边干什么？", + "answer_chinese": "走路", + "evidence": [ + { + "description": "Walking", + "timestamps": { + "start": "00:12:30", + "end": "00:12:31" + } + } + ], + "question_timestamps": "00:14:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Standing still.", + "B": "Walking.", + "C": "Running.", + "D": "You are resting by the road right now instead of walking, remaining in place without moving forward.", + "E": "You are dancing by the roadside rather than walking, moving your body in place instead of traveling." + }, + "answer_label": "B", + "answer_text": "Walking.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_2", + "correct_answer": "Walking.", + "original_options": { + "A": "Standing still.", + "B": "Walking.", + "C": "Running.", + "D": "You are resting by the road right now instead of walking, remaining in place without moving forward.", + "E": "You are dancing by the roadside rather than walking, moving your body in place instead of traveling." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "bad", + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 0, + "D": 0, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 1, + 17, + 17 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_5_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see next to the cat?", + "answer": "You saw the potted plant up ahead.", + "question_chinese": "我看见猫的身旁是什么？", + "answer_chinese": "你看见了前面的盆栽", + "evidence": [ + { + "description": "A potted plant", + "timestamps": { + "start": "00:16:24", + "end": "00:16:25" + } + } + ], + "question_timestamps": "00:17:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw a bicycle nearby.", + "B": "You saw a bowl on the floor.", + "C": "You saw the potted plant up ahead.", + "D": "You saw a bicycle to your left, positioned close to where the cat was sitting.", + "E": "You saw a pan and a bowl in your hands, rather than anything placed next to the cat." + }, + "answer_label": "C", + "answer_text": "You saw the potted plant up ahead.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_5_T1", + "correct_answer": "You saw the potted plant up ahead.", + "original_options": { + "A": "You saw a bicycle nearby.", + "B": "You saw a bowl on the floor.", + "C": "You saw the potted plant up ahead.", + "D": "You saw a bicycle to your left, positioned close to where the cat was sitting.", + "E": "You saw a pan and a bowl in your hands, rather than anything placed next to the cat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 15, + 18 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_5_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see next to the cat?", + "answer": "You saw the potted plant up ahead and the bike to your left.", + "question_chinese": "我看见猫的身旁是什么？", + "answer_chinese": "你看见了前面的盆栽和左边的单车", + "evidence": [ + { + "description": "A potted plant", + "timestamps": { + "start": "00:16:24", + "end": "00:16:25" + } + }, + { + "description": "bicycle", + "timestamps": { + "start": "00:17:19", + "end": "00:17:20" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw the potted plant up ahead and the bike to your left.", + "B": "A potted plant only.", + "C": "A bike to your right.", + "D": "You saw the potted plant up ahead and the bike in front of you next to the cat.", + "E": "You saw the potted plant up ahead and the bike behind you rather than to your left." + }, + "answer_label": "A", + "answer_text": "You saw the potted plant up ahead and the bike to your left.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_5_T2", + "correct_answer": "You saw the potted plant up ahead and the bike to your left.", + "original_options": { + "A": "You saw the potted plant up ahead and the bike to your left.", + "B": "A potted plant only.", + "C": "A bike to your right.", + "D": "You saw the potted plant up ahead and the bike in front of you next to the cat.", + "E": "You saw the potted plant up ahead and the bike behind you rather than to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 17 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_3_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I see the people on the road doing?", + "answer": "You saw a person in front of you running.", + "question_chinese": "我看见马路上的人在干什么？", + "answer_chinese": "你看见面前的人在跑步", + "evidence": [ + { + "description": "Running", + "timestamps": { + "start": "00:07:15", + "end": "00:07:16" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A person behind you running.", + "B": "A person to your left running.", + "C": "You saw a person in front of you running.", + "D": "You saw a person to your right running on the road as you looked toward the side.", + "E": "You saw a person ahead of you walking down the road instead of running in front of you." + }, + "answer_label": "C", + "answer_text": "You saw a person in front of you running.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_3_T1", + "correct_answer": "You saw a person in front of you running.", + "original_options": { + "A": "A person behind you running.", + "B": "A person to your left running.", + "C": "You saw a person in front of you running.", + "D": "You saw a person to your right running on the road as you looked toward the side.", + "E": "You saw a person ahead of you walking down the road instead of running in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 18 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_3_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I see the people on the road doing?", + "answer": "You saw the person in front of you running and cycling.", + "question_chinese": "我看见马路上的人在干什么？", + "answer_chinese": "你看见面前的人跑步和骑行", + "evidence": [ + { + "description": "Running", + "timestamps": { + "start": "00:07:15", + "end": "00:07:16" + } + }, + { + "description": "cycling", + "timestamps": { + "start": "00:09:07", + "end": "00:09:08" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw them walking together.", + "B": "You saw them standing still.", + "C": "You saw the person in front of you running and cycling.", + "D": "You saw the person ahead of you walking while another person cycled past on the road.", + "E": "You saw the person in front of you cycling only, with no one running at that moment." + }, + "answer_label": "C", + "answer_text": "You saw the person in front of you running and cycling.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_3_T2", + "correct_answer": "You saw the person in front of you running and cycling.", + "original_options": { + "A": "You saw them walking together.", + "B": "You saw them standing still.", + "C": "You saw the person in front of you running and cycling.", + "D": "You saw the person ahead of you walking while another person cycled past on the road.", + "E": "You saw the person in front of you cycling only, with no one running at that moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_9", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is beside the cat?", + "answer": "A potted plant.", + "question_chinese": "猫的身旁是什么？", + "answer_chinese": "盆栽", + "evidence": [ + { + "description": "A potted plant", + "timestamps": { + "start": "00:16:24", + "end": "00:16:25" + } + } + ], + "question_timestamps": "00:17:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A blue chair.", + "B": "A small toy mouse.", + "C": "A cozy blanket is beside the cat, spread out like a mat next to where it sits.", + "D": "A bookshelf with a few green books is placed beside the cat along the wall.", + "E": "A potted plant." + }, + "answer_label": "E", + "answer_text": "A potted plant.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_9", + "correct_answer": "A potted plant.", + "original_options": { + "A": "A blue chair.", + "B": "A small toy mouse.", + "C": "A cozy blanket is beside the cat, spread out like a mat next to where it sits.", + "D": "A bookshelf with a few green books is placed beside the cat along the wall.", + "E": "A potted plant." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 1, + "B": 2, + "C": 1, + "D": 1, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 17, + 15 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_10", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is beside the cat?", + "answer": "A potted plant.", + "question_chinese": "猫的身旁是什么？", + "answer_chinese": "盆栽和单车", + "evidence": [ + { + "description": "A potted plant", + "timestamps": { + "start": "00:16:24", + "end": "00:16:25" + } + }, + { + "description": "bicycle", + "timestamps": { + "start": "00:17:19", + "end": "00:17:20" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A wooden chair.", + "B": "A ceramic frog figurine.", + "C": "A small wooden stool with a folded blanket sits beside the cat, close to the floor.", + "D": "A potted plant.", + "E": "An empty cardboard box with scratch marks is beside the cat, like it has been used." + }, + "answer_label": "D", + "answer_text": "A potted plant.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_10", + "correct_answer": "A potted plant.", + "original_options": { + "A": "A wooden chair.", + "B": "A ceramic frog figurine.", + "C": "A small wooden stool with a folded blanket sits beside the cat, close to the floor.", + "D": "A potted plant.", + "E": "An empty cardboard box with scratch marks is beside the cat, like it has been used." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 1, + "B": 2, + "C": 1, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 16 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_13", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many people are walking side by side on the road now?", + "answer": "Two.", + "question_chinese": "现在路上有多少个人并排走？", + "answer_chinese": "2", + "evidence": [ + { + "description": "Two people are walking side by side.", + "timestamps": { + "start": "00:22:12", + "end": "00:22:13" + } + } + ], + "question_timestamps": "00:23:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four people are walking side by side on the road right now, spreading across the lane together.", + "D": "Five people are walking side by side on the road now, spread across the lane together.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_13", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four people are walking side by side on the road right now, spreading across the lane together.", + "D": "Five people are walking side by side on the road now, spread across the lane together.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 2, + "E": 3 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 16 + ] + }, + { + "video_uid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003", + "video_length": "00:30:56", + "qid": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_14", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many people are walking side by side on the road now?", + "answer": "Three.", + "question_chinese": "现在路上有多少个人并排走？", + "answer_chinese": "3", + "evidence": [ + { + "description": "three people are walking side by side.", + "timestamps": { + "start": "00:23:02", + "end": "00:23:03" + } + } + ], + "question_timestamps": "00:24:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "Five people are walking side by side on the road now, taking up most of the width.", + "E": "Seven people are walking side by side on the road now, forming a wide line across it." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "863011c8-ebf7-4900-87d1-3c7930f95dab-003_14", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "Five people are walking side by side on the road now, taking up most of the width.", + "E": "Seven people are walking side by side on the road now, forming a wide line across it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 3, + "C": 2, + "D": 2, + "E": 3 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 17 + ] + } +] diff --git a/QAs/8b72b54f-f87c-4fdb-84dc-97e42ab111ac-mcq.json b/QAs/8b72b54f-f87c-4fdb-84dc-97e42ab111ac-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..216795480fc2b341b122576062eeb36792686aaf --- /dev/null +++ b/QAs/8b72b54f-f87c-4fdb-84dc-97e42ab111ac-mcq.json @@ -0,0 +1,810 @@ +[ + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_2_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, how many radishes did I peel?", + "answer": "You just peeled three radishes and put them in front of you.", + "question_chinese": "我刚刚削了几个萝卜？", + "answer_chinese": "你刚刚削了三个萝卜，放在了你的正前方。", + "evidence": [ + { + "description": "You peeled a radish.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:50" + } + }, + { + "description": "You peeled the second radish.", + "timestamps": { + "start": "00:01:50", + "end": "00:02:14" + } + }, + { + "description": "You peeled the third radish.", + "timestamps": { + "start": "00:02:21", + "end": "00:02:42" + } + } + ], + "question_timestamps": "00:02:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You peeled one radish.", + "B": "You just peeled three radishes and put them in front of you.", + "C": "You peeled two radishes.", + "D": "You just peeled four radishes and put them in front of you, lined up near the edge.", + "E": "You just peeled five radishes and put them in front of you, grouped together on the surface." + }, + "answer_label": "B", + "answer_text": "You just peeled three radishes and put them in front of you.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_2_T1", + "correct_answer": "You just peeled three radishes and put them in front of you.", + "original_options": { + "A": "You peeled one radish.", + "B": "You just peeled three radishes and put them in front of you.", + "C": "You peeled two radishes.", + "D": "You just peeled four radishes and put them in front of you, lined up near the edge.", + "E": "You just peeled five radishes and put them in front of you, grouped together on the surface." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_2_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, how many radishes did I peel?", + "answer": "You just peeled four radishes, and they’re all currently on your left-hand side.", + "question_chinese": "我刚刚削了几个萝卜？", + "answer_chinese": "你刚刚削了四个萝卜，目前都放在你的左手边。", + "evidence": [ + { + "description": "You peeled a radish.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:50" + } + }, + { + "description": "You peeled the second radish.", + "timestamps": { + "start": "00:01:50", + "end": "00:02:14" + } + }, + { + "description": "You peeled the third radish.", + "timestamps": { + "start": "00:02:21", + "end": "00:02:43" + } + }, + { + "description": "You peeled the fourth radish.", + "timestamps": { + "start": "00:02:53", + "end": "00:03:17" + } + } + ], + "question_timestamps": "00:09:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You peeled two radishes.", + "B": "You peeled three radishes.", + "C": "You just peeled four radishes, and they’re all currently on your left-hand side.", + "D": "You just peeled five radishes, and they’re all currently on your left-hand side in a small pile.", + "E": "You just peeled six radishes, and they’re all currently on your left-hand side, arranged together." + }, + "answer_label": "C", + "answer_text": "You just peeled four radishes, and they’re all currently on your left-hand side.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_2_T2", + "correct_answer": "You just peeled four radishes, and they’re all currently on your left-hand side.", + "original_options": { + "A": "You peeled two radishes.", + "B": "You peeled three radishes.", + "C": "You just peeled four radishes, and they’re all currently on your left-hand side.", + "D": "You just peeled five radishes, and they’re all currently on your left-hand side in a small pile.", + "E": "You just peeled six radishes, and they’re all currently on your left-hand side, arranged together." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 16 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put my mobile phone?", + "answer": "You just put your mobile phone on the table.", + "question_chinese": "我刚刚把手机放在什么位置？", + "answer_chinese": "你刚刚把手机放在桌上。", + "evidence": [ + { + "description": "You put your mobile phone on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:59", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In your bag.", + "B": "In the drawer.", + "C": "You just put your mobile phone on the kitchen counter, leaving it near the edge.", + "D": "You just put your mobile phone on the table.", + "E": "You just put your mobile phone on a shelf, setting it down above the table level." + }, + "answer_label": "D", + "answer_text": "You just put your mobile phone on the table.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_1", + "correct_answer": "You just put your mobile phone on the table.", + "original_options": { + "A": "In your bag.", + "B": "In the drawer.", + "C": "You just put your mobile phone on the kitchen counter, leaving it near the edge.", + "D": "You just put your mobile phone on the table.", + "E": "You just put your mobile phone on a shelf, setting it down above the table level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put my mobile phone?", + "answer": "You put your mobile phone on the table.", + "question_chinese": "我刚刚把手机放在什么位置？", + "answer_chinese": "你把手机放在了桌子上。", + "evidence": [ + { + "description": "You put your mobile phone on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:13:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it in your backpack.", + "B": "You put your mobile phone on the table.", + "C": "You put it in the drawer.", + "D": "You set your mobile phone down on the kitchen island instead of leaving it on the table.", + "E": "You slipped your mobile phone into your jacket pocket, not onto any surface like the table." + }, + "answer_label": "B", + "answer_text": "You put your mobile phone on the table.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_2", + "correct_answer": "You put your mobile phone on the table.", + "original_options": { + "A": "You put it in your backpack.", + "B": "You put your mobile phone on the table.", + "C": "You put it in the drawer.", + "D": "You set your mobile phone down on the kitchen island instead of leaving it on the table.", + "E": "You slipped your mobile phone into your jacket pocket, not onto any surface like the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_4_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just now, where did I pour the vegetables from the cutting board?", + "answer": "You just dumped the vegetables from the cutting board into the trash can.", + "question_chinese": "我刚刚把砧板上的菜倒到什么地方了？", + "answer_chinese": "你刚刚把砧板上的菜倒垃圾桶里了。", + "evidence": [ + { + "description": "You just poured the vegetables on the cutting board into the trash can.", + "timestamps": { + "start": "00:06:28", + "end": "00:06:35" + } + } + ], + "question_timestamps": "00:06:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You poured them into the sink.", + "B": "You poured them into a bowl.", + "C": "You tipped the vegetables off the cutting board into a pot, not into the trash can.", + "D": "You just dumped the vegetables from the cutting board into the trash can.", + "E": "You slid the vegetables from the cutting board into a pan rather than dumping them into the trash." + }, + "answer_label": "D", + "answer_text": "You just dumped the vegetables from the cutting board into the trash can.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_4_T1", + "correct_answer": "You just dumped the vegetables from the cutting board into the trash can.", + "original_options": { + "A": "You poured them into the sink.", + "B": "You poured them into a bowl.", + "C": "You tipped the vegetables off the cutting board into a pot, not into the trash can.", + "D": "You just dumped the vegetables from the cutting board into the trash can.", + "E": "You slid the vegetables from the cutting board into a pan rather than dumping them into the trash." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_4_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just now, where did I pour the vegetables from the cutting board?", + "answer": "You just dumped the vegetables from the cutting board into the trash can.", + "question_chinese": "我刚刚把砧板上的菜倒到什么地方了？", + "answer_chinese": "你刚刚把砧板上的菜倒到垃圾桶里了。", + "evidence": [ + { + "description": "You just poured the vegetables on the cutting board into the trash can.", + "timestamps": { + "start": "00:06:28", + "end": "00:06:35" + } + } + ], + "question_timestamps": "00:11:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You poured them into the sink.", + "B": "You poured them into a pot.", + "C": "You emptied the vegetables off the cutting board into a bowl instead of throwing them in the trash.", + "D": "You pushed the vegetables from the cutting board into a pan rather than dumping them into the trash can.", + "E": "You just dumped the vegetables from the cutting board into the trash can." + }, + "answer_label": "E", + "answer_text": "You just dumped the vegetables from the cutting board into the trash can.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_4_T2", + "correct_answer": "You just dumped the vegetables from the cutting board into the trash can.", + "original_options": { + "A": "You poured them into the sink.", + "B": "You poured them into a pot.", + "C": "You emptied the vegetables off the cutting board into a bowl instead of throwing them in the trash.", + "D": "You pushed the vegetables from the cutting board into a pan rather than dumping them into the trash can.", + "E": "You just dumped the vegetables from the cutting board into the trash can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_7_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, how many mushrooms did I wash?", + "answer": "You washed three mushrooms; they’re in front of you.", + "question_chinese": "我刚刚洗了多少个蘑菇？", + "answer_chinese": "你洗了三个蘑菇，在你的前方。", + "evidence": [ + { + "description": "You washed a mushroom.", + "timestamps": { + "start": "00:12:58", + "end": "00:13:08" + } + }, + { + "description": "You have washed the second mushroom.", + "timestamps": { + "start": "00:13:08", + "end": "00:13:15" + } + }, + { + "description": "You have washed the second mushroom.", + "timestamps": { + "start": "00:13:16", + "end": "00:13:23" + } + } + ], + "question_timestamps": "00:13:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed two mushrooms.", + "B": "You washed three mushrooms; they’re in front of you.", + "C": "You washed four mushrooms.", + "D": "You washed five mushrooms; they’re in front of you, grouped together after you rinsed them.", + "E": "You washed one mushroom; it’s in front of you, sitting by itself after you finished washing it." + }, + "answer_label": "B", + "answer_text": "You washed three mushrooms; they’re in front of you.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_7_T1", + "correct_answer": "You washed three mushrooms; they’re in front of you.", + "original_options": { + "A": "You washed two mushrooms.", + "B": "You washed three mushrooms; they’re in front of you.", + "C": "You washed four mushrooms.", + "D": "You washed five mushrooms; they’re in front of you, grouped together after you rinsed them.", + "E": "You washed one mushroom; it’s in front of you, sitting by itself after you finished washing it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 18 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_7_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, how many mushrooms did I wash?", + "answer": "You washed four mushrooms and put them by your left hand.", + "question_chinese": "我刚刚洗了多少个蘑菇？", + "answer_chinese": "你洗了四个蘑菇，放在你的左手边。", + "evidence": [ + { + "description": "You washed a mushroom.", + "timestamps": { + "start": "00:12:58", + "end": "00:13:08" + } + }, + { + "description": "You have washed the second mushroom.", + "timestamps": { + "start": "00:13:08", + "end": "00:13:15" + } + }, + { + "description": "You have washed the second mushroom.", + "timestamps": { + "start": "00:13:16", + "end": "00:13:23" + } + }, + { + "description": "You have washed the fourth mushroom.", + "timestamps": { + "start": "00:13:24", + "end": "00:13:31" + } + } + ], + "question_timestamps": "00:13:39", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed three mushrooms.", + "B": "You washed five mushrooms.", + "C": "You washed four mushrooms and put them by your left hand.", + "D": "You washed two mushrooms and put them by your left hand after rinsing them off.", + "E": "You washed six mushrooms and placed them by your left hand, all together in one small cluster." + }, + "answer_label": "C", + "answer_text": "You washed four mushrooms and put them by your left hand.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_7_T2", + "correct_answer": "You washed four mushrooms and put them by your left hand.", + "original_options": { + "A": "You washed three mushrooms.", + "B": "You washed five mushrooms.", + "C": "You washed four mushrooms and put them by your left hand.", + "D": "You washed two mushrooms and put them by your left hand after rinsing them off.", + "E": "You washed six mushrooms and placed them by your left hand, all together in one small cluster." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_5_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Before I put the cucumber into the bowl, how many kinds of vegetables were there?", + "answer": "Before you put the cucumber into the bowl, there were two kinds of vegetables—also on your left-hand side.", + "question_chinese": "我把黄瓜放进碗里之前有几种菜？", + "answer_chinese": "你把黄瓜放进碗里之前有两种菜，也是在你的左手边。", + "evidence": [ + { + "description": "You put the green peppers into the bowl.", + "timestamps": { + "start": "00:07:47", + "end": "00:07:52" + } + }, + { + "description": "You put the carrots into the bowl.", + "timestamps": { + "start": "00:09:39", + "end": "00:09:45" + } + } + ], + "question_timestamps": "00:10:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There was one kind of vegetable.", + "B": "There were three kinds of vegetables.", + "C": "Before you added the cucumber, there were four kinds of vegetables, still positioned on your left-hand side.", + "D": "Before you put the cucumber into the bowl, there were zero kinds of vegetables, with nothing sitting on your left-hand side.", + "E": "Before you put the cucumber into the bowl, there were two kinds of vegetables—also on your left-hand side." + }, + "answer_label": "E", + "answer_text": "Before you put the cucumber into the bowl, there were two kinds of vegetables—also on your left-hand side.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_5_T1", + "correct_answer": "Before you put the cucumber into the bowl, there were two kinds of vegetables—also on your left-hand side.", + "original_options": { + "A": "There was one kind of vegetable.", + "B": "There were three kinds of vegetables.", + "C": "Before you added the cucumber, there were four kinds of vegetables, still positioned on your left-hand side.", + "D": "Before you put the cucumber into the bowl, there were zero kinds of vegetables, with nothing sitting on your left-hand side.", + "E": "Before you put the cucumber into the bowl, there were two kinds of vegetables—also on your left-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 21 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_5_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Before I put the cucumber into the bowl, how many kinds of vegetables were there?", + "answer": "Before you put the cucumber into the bowl, there were two kinds of vegetables on your left-hand side.", + "question_chinese": "我把黄瓜放进碗里之前有几种菜？", + "answer_chinese": "你把黄瓜放进碗里之前有两种菜，在你的左手边。", + "evidence": [ + { + "description": "You put the green peppers into the bowl.", + "timestamps": { + "start": "00:07:47", + "end": "00:07:52" + } + }, + { + "description": "You put the carrots into the bowl.", + "timestamps": { + "start": "00:09:39", + "end": "00:09:45" + } + } + ], + "question_timestamps": "00:18:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Before you put the cucumber into the bowl, there were two kinds of vegetables on your left-hand side.", + "B": "There was one kind of vegetable.", + "C": "There were three kinds of vegetables.", + "D": "Before you put the cucumber into the bowl, there were four kinds of vegetables on your left-hand side.", + "E": "Before adding the cucumber, there were zero kinds of vegetables on your left-hand side at that moment." + }, + "answer_label": "A", + "answer_text": "Before you put the cucumber into the bowl, there were two kinds of vegetables on your left-hand side.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_5_T2", + "correct_answer": "Before you put the cucumber into the bowl, there were two kinds of vegetables on your left-hand side.", + "original_options": { + "A": "Before you put the cucumber into the bowl, there were two kinds of vegetables on your left-hand side.", + "B": "There was one kind of vegetable.", + "C": "There were three kinds of vegetables.", + "D": "Before you put the cucumber into the bowl, there were four kinds of vegetables on your left-hand side.", + "E": "Before adding the cucumber, there were zero kinds of vegetables on your left-hand side at that moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_6_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the kettle right now?", + "answer": "You can reach it directly; it’s on your right-hand side.", + "question_chinese": "我现在可以直接触碰到烧水壶吗？", + "answer_chinese": "你可以直接触碰到，在你的右手边。", + "evidence": [ + { + "description": "You are at a certain distance from the kettle.", + "timestamps": { + "start": "00:19:35", + "end": "00:19:43" + } + } + ], + "question_timestamps": "00:19:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach it directly; it’s on your right-hand side.", + "B": "You can reach it directly; it’s on your right-hand side." + }, + "answer_label": "B", + "answer_text": "You can reach it directly; it’s on your right-hand side.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_6_T1", + "correct_answer": "You can reach it directly; it’s on your right-hand side.", + "original_options": { + "A": "You can’t reach it directly; it’s on your right-hand side.", + "B": "You can reach it directly; it’s on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_6_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the kettle right now?", + "answer": "You can’t touch it directly; it’s to your right.", + "question_chinese": "我现在可以直接触碰到烧水壶吗？", + "answer_chinese": "你无法直接触碰到，它在你的右方。", + "evidence": [ + { + "description": "You are at a certain distance from the kettle.", + "timestamps": { + "start": "00:19:35", + "end": "00:19:48" + } + } + ], + "question_timestamps": "00:23:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t touch it directly; it’s to your right.", + "B": "You can touch it directly; it’s to your right." + }, + "answer_label": "A", + "answer_text": "You can’t touch it directly; it’s to your right.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_6_T2", + "correct_answer": "You can’t touch it directly; it’s to your right.", + "original_options": { + "A": "You can’t touch it directly; it’s to your right.", + "B": "You can touch it directly; it’s to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_3_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the faucet relative to me right now?", + "answer": "The faucet is now on the left side.", + "question_chinese": "水龙头现在相对我是什么位置？", + "answer_chinese": "水龙头现在在你的左侧。", + "evidence": [ + { + "description": "The faucet is now on your left.", + "timestamps": { + "start": "00:04:04", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:04:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on your right.", + "B": "It’s straight ahead.", + "C": "The faucet is now on the left side.", + "D": "The faucet is on your right side, next to the sink, not on the left.", + "E": "The faucet sits directly in front of you at the sink, rather than off to your left." + }, + "answer_label": "C", + "answer_text": "The faucet is now on the left side.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_3_T1", + "correct_answer": "The faucet is now on the left side.", + "original_options": { + "A": "It’s on your right.", + "B": "It’s straight ahead.", + "C": "The faucet is now on the left side.", + "D": "The faucet is on your right side, next to the sink, not on the left.", + "E": "The faucet sits directly in front of you at the sink, rather than off to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 17 + ] + }, + { + "video_uid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac", + "video_length": "00:25:25", + "qid": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_3_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the faucet relative to me right now?", + "answer": "The faucet is on the left side.", + "question_chinese": "水龙头现在相对我是什么位置？", + "answer_chinese": "水龙头在你的左侧。", + "evidence": [ + { + "description": "The faucet is now on your left.", + "timestamps": { + "start": "00:06:40", + "end": "00:06:54" + } + } + ], + "question_timestamps": "00:14:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on your right.", + "B": "It’s behind you.", + "C": "The faucet is directly in front of you at the sink, not positioned on your left.", + "D": "The faucet is on the left side.", + "E": "The faucet is on your right-hand side right now, rather than on the left side." + }, + "answer_label": "D", + "answer_text": "The faucet is on the left side.", + "q_id": "8b72b54f-f87c-4fdb-84dc-97e42ab111ac_3_T2", + "correct_answer": "The faucet is on the left side.", + "original_options": { + "A": "It’s on your right.", + "B": "It’s behind you.", + "C": "The faucet is directly in front of you at the sink, not positioned on your left.", + "D": "The faucet is on the left side.", + "E": "The faucet is on your right-hand side right now, rather than on the left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 15 + ] + } +] diff --git a/QAs/8d3337ba-61a1-4c23-abb3-2728ba21adad-mcq.json b/QAs/8d3337ba-61a1-4c23-abb3-2728ba21adad-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fce0f75c2a9876a30bbd3222d8b0f58bca2a6292 --- /dev/null +++ b/QAs/8d3337ba-61a1-4c23-abb3-2728ba21adad-mcq.json @@ -0,0 +1,1061 @@ +[ + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first went out, was the bicycle on my left?", + "answer": "Yes, the bicycle is by the wall on your left.", + "question_chinese": "当我刚出门时，自行车是在我的左边吗？", + "answer_chinese": "是的，自行车在你左边的墙边。", + "evidence": [ + { + "description": "There is a bicycle by the wall on the left.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:52" + } + } + ], + "question_timestamps": "00:01:55", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, the bicycle is by the wall on your right.", + "B": "Yes, the bicycle is by the wall on your left." + }, + "answer_label": "B", + "answer_text": "Yes, the bicycle is by the wall on your left.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_5", + "correct_answer": "Yes, the bicycle is by the wall on your left.", + "original_options": { + "A": "No, the bicycle is by the wall on your right.", + "B": "Yes, the bicycle is by the wall on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_6", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first went out, was the bicycle on my left?", + "answer": "Yes, the bicycle is by the wall on your left.", + "question_chinese": "当我刚出门时，自行车是在我的左边吗？", + "answer_chinese": "是的，自行车在你左边的墙边。", + "evidence": [ + { + "description": "There is a bicycle by the wall on the left.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:52" + } + } + ], + "question_timestamps": "00:13:24", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, the bicycle is by the wall on your left.", + "B": "No, the bicycle is by the wall on your right." + }, + "answer_label": "A", + "answer_text": "Yes, the bicycle is by the wall on your left.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_6", + "correct_answer": "Yes, the bicycle is by the wall on your left.", + "original_options": { + "A": "Yes, the bicycle is by the wall on your left.", + "B": "No, the bicycle is by the wall on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_3_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Was the bicycle I saw when I went out to my left?", + "answer": "Yes—the bicycle is by the wall to your left.", + "question_chinese": "我出门看到的自行车是在我的左边吗？", + "answer_chinese": "是的，自行车在你左边的墙边。", + "evidence": [ + { + "description": "There is a bicycle by the wall on the left.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:52" + } + } + ], + "question_timestamps": "00:01:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—the bicycle is by the wall to your right.", + "B": "Yes—the bicycle is by the wall to your left." + }, + "answer_label": "B", + "answer_text": "Yes—the bicycle is by the wall to your left.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_3_T1", + "correct_answer": "Yes—the bicycle is by the wall to your left.", + "original_options": { + "A": "No—the bicycle is by the wall to your right.", + "B": "Yes—the bicycle is by the wall to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_3_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Was the bicycle I saw when I went out to my left?", + "answer": "No—the bicycle is by the wall to your right now.", + "question_chinese": "我出门看到的自行车是在我的左边吗？", + "answer_chinese": "不，自行车现在在你右边的墙边。", + "evidence": [ + { + "description": "There is a bicycle by the wall on the left.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:52" + } + }, + { + "description": "The bike is now by the right wall.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:12" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—the bicycle is by the wall to your right now.", + "B": "Yes—the bicycle is by the wall to your left." + }, + "answer_label": "A", + "answer_text": "No—the bicycle is by the wall to your right now.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_3_T2", + "correct_answer": "No—the bicycle is by the wall to your right now.", + "original_options": { + "A": "No—the bicycle is by the wall to your right now.", + "B": "Yes—the bicycle is by the wall to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I just folded a few pieces of clothing?", + "answer": "You just folded a piece of clothing.", + "question_chinese": "我刚刚叠了几件衣服？", + "answer_chinese": "你刚刚叠了一件衣服。", + "evidence": [ + { + "description": "You folded a white piece of clothing.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just folded a few pieces of clothing.", + "B": "You just folded a piece of clothing.", + "C": "You just folded two pieces of clothing.", + "D": "You just folded three pieces of clothing, not just a single piece.", + "E": "You just folded several pieces of clothing in a row, rather than only one item." + }, + "answer_label": "B", + "answer_text": "You just folded a piece of clothing.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_1", + "correct_answer": "You just folded a piece of clothing.", + "original_options": { + "A": "You just folded a few pieces of clothing.", + "B": "You just folded a piece of clothing.", + "C": "You just folded two pieces of clothing.", + "D": "You just folded three pieces of clothing, not just a single piece.", + "E": "You just folded several pieces of clothing in a row, rather than only one item." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 12, + 15 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I just folded a few pieces of clothing?", + "answer": "You just folded two pieces of clothing.", + "question_chinese": "我刚刚叠了几件衣服？", + "answer_chinese": "你刚刚叠了两件衣服。", + "evidence": [ + { + "description": "You folded a white piece of clothing.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:33" + } + }, + { + "description": "You folded another blue piece of clothing.", + "timestamps": { + "start": "00:00:39", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You just folded one piece of clothing.", + "B": "You just folded three pieces of clothing.", + "C": "You just folded four pieces of clothing, finishing them one after another before setting them aside.", + "D": "You just folded five pieces of clothing, stacking them neatly together once you were done.", + "E": "You just folded two pieces of clothing." + }, + "answer_label": "E", + "answer_text": "You just folded two pieces of clothing.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_2", + "correct_answer": "You just folded two pieces of clothing.", + "original_options": { + "A": "You just folded one piece of clothing.", + "B": "You just folded three pieces of clothing.", + "C": "You just folded four pieces of clothing, finishing them one after another before setting them aside.", + "D": "You just folded five pieces of clothing, stacking them neatly together once you were done.", + "E": "You just folded two pieces of clothing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 15 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_9", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I just wash the cup?", + "answer": "You just washed the cup in the sink on your right.", + "question_chinese": "我刚刚在哪里洗的杯子？", + "answer_chinese": "你刚刚在你右侧的水槽洗的杯子。", + "evidence": [ + { + "description": "You just washed two cups in the sink on the right.", + "timestamps": { + "start": "00:04:09", + "end": "00:04:24" + } + } + ], + "question_timestamps": "00:04:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just washed the cup in the sink on your right.", + "B": "In the left sink.", + "C": "In the front sink.", + "D": "You just washed the cup in the sink on your left, reaching across the counter to rinse it.", + "E": "You just washed the cup in the sink in front of you, scrubbing it and rinsing it there." + }, + "answer_label": "A", + "answer_text": "You just washed the cup in the sink on your right.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_9", + "correct_answer": "You just washed the cup in the sink on your right.", + "original_options": { + "A": "You just washed the cup in the sink on your right.", + "B": "In the left sink.", + "C": "In the front sink.", + "D": "You just washed the cup in the sink on your left, reaching across the counter to rinse it.", + "E": "You just washed the cup in the sink in front of you, scrubbing it and rinsing it there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 18 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_10", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I just wash the cup?", + "answer": "You just washed the cup in the sink on your right.", + "question_chinese": "我刚刚在哪里洗的杯子？", + "answer_chinese": "你刚刚在你右侧的水槽洗的杯子。", + "evidence": [ + { + "description": "You just washed two cups in the sink on the right.", + "timestamps": { + "start": "00:17:25", + "end": "00:17:31" + } + } + ], + "question_timestamps": "00:17:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the left sink.", + "B": "In the front sink.", + "C": "You just washed the cup in the sink on your left, rinsing it under the faucet there.", + "D": "You just washed the cup in the sink on your right.", + "E": "You just washed the cup in the sink in front of you, cleaning it there before setting it down." + }, + "answer_label": "D", + "answer_text": "You just washed the cup in the sink on your right.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_10", + "correct_answer": "You just washed the cup in the sink on your right.", + "original_options": { + "A": "In the left sink.", + "B": "In the front sink.", + "C": "You just washed the cup in the sink on your left, rinsing it under the faucet there.", + "D": "You just washed the cup in the sink on your right.", + "E": "You just washed the cup in the sink in front of you, cleaning it there before setting it down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 19 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_5_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I just wash the cups, and how many cups were there?", + "answer": "You just washed the cups in the sink on the right side; there were three cups.", + "question_chinese": "我刚刚在哪里洗的杯子，有几个杯子？", + "answer_chinese": "你刚刚在右侧的水槽洗的杯子,有三个杯子。", + "evidence": [ + { + "description": "You just washed two cups in the sink on the right and the sink is front of you.", + "timestamps": { + "start": "00:04:09", + "end": "00:04:24" + } + } + ], + "question_timestamps": "00:05:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the left sink; two cups.", + "B": "In the front sink; three cups.", + "C": "You just washed the cups in the sink on the left side; there were three cups in total.", + "D": "You just washed the cups in the sink on the right side; there were two cups that you rinsed.", + "E": "You just washed the cups in the sink on the right side; there were three cups." + }, + "answer_label": "E", + "answer_text": "You just washed the cups in the sink on the right side; there were three cups.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_5_T1", + "correct_answer": "You just washed the cups in the sink on the right side; there were three cups.", + "original_options": { + "A": "In the left sink; two cups.", + "B": "In the front sink; three cups.", + "C": "You just washed the cups in the sink on the left side; there were three cups in total.", + "D": "You just washed the cups in the sink on the right side; there were two cups that you rinsed.", + "E": "You just washed the cups in the sink on the right side; there were three cups." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_5_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I just wash the cups, and how many cups were there?", + "answer": "You just washed the cups in the sink on your right side; you washed two cups.", + "question_chinese": "我刚刚在哪里洗的杯子，有几个杯子？", + "answer_chinese": "你刚刚在你右侧的水槽洗的杯子,洗了两个杯子。", + "evidence": [ + { + "description": "You just washed two cups in the sink on the right.", + "timestamps": { + "start": "00:17:25", + "end": "00:17:31" + } + } + ], + "question_timestamps": "00:17:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You just washed the cups in the sink on your right side; you washed two cups.", + "B": "In the left sink; one cup.", + "C": "In the right sink; three cups.", + "D": "You just washed the cups in the sink on your left side; you washed two cups there.", + "E": "You just washed the cups in the sink on your right side; you washed four cups in total." + }, + "answer_label": "A", + "answer_text": "You just washed the cups in the sink on your right side; you washed two cups.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_5_T2", + "correct_answer": "You just washed the cups in the sink on your right side; you washed two cups.", + "original_options": { + "A": "You just washed the cups in the sink on your right side; you washed two cups.", + "B": "In the left sink; one cup.", + "C": "In the right sink; three cups.", + "D": "You just washed the cups in the sink on your left side; you washed two cups there.", + "E": "You just washed the cups in the sink on your right side; you washed four cups in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_4_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just walk?", + "answer": "You’re heading out from the kitchen, turning right, then turning right again.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你正从厨房出发，右转，再右转。", + "evidence": [ + { + "description": "You start from the kitchen, turn right, turn right again, walk straight along the hallway, and arrive at the living room.", + "timestamps": { + "start": "00:11:53", + "end": "00:11:56" + } + } + ], + "question_timestamps": "00:11:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You left the kitchen, turning left twice.", + "B": "You left the kitchen, turning left then right.", + "C": "You’re heading out from the kitchen, turning right first, then continuing straight without making a second turn.", + "D": "You’re heading out from the kitchen, turning right, then turning right again.", + "E": "You’re heading out from the kitchen, turning left, then turning left again before walking down the hallway." + }, + "answer_label": "D", + "answer_text": "You’re heading out from the kitchen, turning right, then turning right again.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_4_T2", + "correct_answer": "You’re heading out from the kitchen, turning right, then turning right again.", + "original_options": { + "A": "You left the kitchen, turning left twice.", + "B": "You left the kitchen, turning left then right.", + "C": "You’re heading out from the kitchen, turning right first, then continuing straight without making a second turn.", + "D": "You’re heading out from the kitchen, turning right, then turning right again.", + "E": "You’re heading out from the kitchen, turning left, then turning left again before walking down the hallway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 18 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_4_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just walk?", + "answer": "You’re heading out from the kitchen, turning right, then turning right again, walking straight down the hallway, and moving toward the living room.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你正从厨房出发，右转，再右转，沿着走廊直走，朝着客厅的方向移动。", + "evidence": [ + { + "description": "You start from the kitchen, turn right, turn right again, walk straight along the hallway, and arrive at the living room.", + "timestamps": { + "start": "00:11:53", + "end": "00:12:00" + } + } + ], + "question_timestamps": "00:12:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’re heading out from the kitchen, turning right, then turning right again, walking straight down the hallway, and moving toward the living room.", + "B": "You left the kitchen, turning left twice.", + "C": "You left the kitchen, turning right then left.", + "D": "You’re heading out from the kitchen, turning right, then turning right again, but stopping before the hallway.", + "E": "You’re heading out from the kitchen, turning left, then turning right again, walking straight down the hallway toward the living room." + }, + "answer_label": "A", + "answer_text": "You’re heading out from the kitchen, turning right, then turning right again, walking straight down the hallway, and moving toward the living room.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_4_T1", + "correct_answer": "You’re heading out from the kitchen, turning right, then turning right again, walking straight down the hallway, and moving toward the living room.", + "original_options": { + "A": "You’re heading out from the kitchen, turning right, then turning right again, walking straight down the hallway, and moving toward the living room.", + "B": "You left the kitchen, turning left twice.", + "C": "You left the kitchen, turning right then left.", + "D": "You’re heading out from the kitchen, turning right, then turning right again, but stopping before the hallway.", + "E": "You’re heading out from the kitchen, turning left, then turning right again, walking straight down the hallway toward the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 22 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_3", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is on the ground when I just arrived at the door?", + "answer": "When you just arrived at the door, there were shoes, chairs, and bottles on the ground.", + "question_chinese": "我刚到门口的时候,地上有什么东西?", + "answer_chinese": "你刚到门口时,地上有鞋子、椅子和瓶子。", + "evidence": [ + { + "description": "On the ground to the right of the door, there are many shoes, a chair, and a pile of bottles.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:46" + } + } + ], + "question_timestamps": "00:01:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There were shoes and bottles on the ground.", + "B": "There were chairs and bottles on the ground.", + "C": "When you just arrived at the door, there were shoes, chairs, and bottles on the ground.", + "D": "When you just arrived at the door, there were shoes, bottles, and a bag on the ground.", + "E": "When you just arrived at the door, there were chairs, shoes, and boxes scattered on the ground." + }, + "answer_label": "C", + "answer_text": "When you just arrived at the door, there were shoes, chairs, and bottles on the ground.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_3", + "correct_answer": "When you just arrived at the door, there were shoes, chairs, and bottles on the ground.", + "original_options": { + "A": "There were shoes and bottles on the ground.", + "B": "There were chairs and bottles on the ground.", + "C": "When you just arrived at the door, there were shoes, chairs, and bottles on the ground.", + "D": "When you just arrived at the door, there were shoes, bottles, and a bag on the ground.", + "E": "When you just arrived at the door, there were chairs, shoes, and boxes scattered on the ground." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 17 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_4", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is on the ground when I just arrived at the door?", + "answer": "When you just arrived at the door, there were shoes, chairs, and bottles on the ground.", + "question_chinese": "我刚到门口的时候,地上有什么东西?", + "answer_chinese": "你刚到门口时,地上有鞋子、椅子和瓶子。", + "evidence": [ + { + "description": "On the ground to the right of the door, there are many shoes, a chair, and a pile of bottles.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:46" + } + } + ], + "question_timestamps": "00:12:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "When you just arrived at the door, there were shoes, chairs, and bottles on the ground.", + "B": "Shoes and chairs.", + "C": "Shoes and bottles.", + "D": "When you just arrived at the door, there were shoes, chairs, and cups on the ground.", + "E": "When you just arrived at the door, there were shoes, a chair, and several bottles on the ground." + }, + "answer_label": "A", + "answer_text": "When you just arrived at the door, there were shoes, chairs, and bottles on the ground.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_4", + "correct_answer": "When you just arrived at the door, there were shoes, chairs, and bottles on the ground.", + "original_options": { + "A": "When you just arrived at the door, there were shoes, chairs, and bottles on the ground.", + "B": "Shoes and chairs.", + "C": "Shoes and bottles.", + "D": "When you just arrived at the door, there were shoes, chairs, and cups on the ground.", + "E": "When you just arrived at the door, there were shoes, a chair, and several bottles on the ground." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 18 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_2_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I had just arrived at the doorway, what was on the floor, and where was it relative to me?", + "answer": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor in front of you.", + "question_chinese": "我刚到门口的时候,地上有什么东西，在相对于我的什么位置?", + "answer_chinese": "你刚到门口时,地上有鞋子、椅子和瓶子，在你的前面。", + "evidence": [ + { + "description": "On the ground to the right of the door, there are many shoes, a chair, and a pile of bottles.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:44" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you on the floor.", + "B": "To your left on the floor.", + "C": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor in front of you.", + "D": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor behind you.", + "E": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor to your right." + }, + "answer_label": "C", + "answer_text": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor in front of you.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_2_T1", + "correct_answer": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor in front of you.", + "original_options": { + "A": "Behind you on the floor.", + "B": "To your left on the floor.", + "C": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor in front of you.", + "D": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor behind you.", + "E": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 21 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_2_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I had just arrived at the doorway, what was on the floor, and where was it relative to me?", + "answer": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor; now they’re behind you.", + "question_chinese": "我刚到门口的时候,地上有什么东西，在相对于我的什么位置?", + "answer_chinese": "你刚到门口时,地上有鞋子、椅子和瓶子，现在在我的后面。", + "evidence": [ + { + "description": "On the ground to the right of the door, there are many shoes, a chair, and a pile of bottles.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:46" + } + } + ], + "question_timestamps": "00:01:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Now they’re in front of you.", + "B": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor; now they’re behind you.", + "C": "Now they’re to your left.", + "D": "When you first arrived at the doorway, they were on the floor; now they’re to your right.", + "E": "When you first arrived at the doorway, the items were on the floor; now they’re in front of you." + }, + "answer_label": "B", + "answer_text": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor; now they’re behind you.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_2_T2", + "correct_answer": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor; now they’re behind you.", + "original_options": { + "A": "Now they’re in front of you.", + "B": "When you first arrived at the doorway, there were shoes, a chair, and a bottle on the floor; now they’re behind you.", + "C": "Now they’re to your left.", + "D": "When you first arrived at the doorway, they were on the floor; now they’re to your right.", + "E": "When you first arrived at the doorway, the items were on the floor; now they’re in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 20 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_11", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to get more cups, where should I go to get them?", + "answer": "You should get it from the cabinet to the right of the sink and above the microwave.", + "question_chinese": "我如果想拿更多的杯子，我应该去哪里拿？", + "answer_chinese": "你应该去水槽右边和微波炉上方的柜子里拿。", + "evidence": [ + { + "description": "There are two cups to the right of the sink.", + "timestamps": { + "start": "00:06:37", + "end": "00:06:51" + } + }, + { + "description": "There are many cups in the cabinet above the microwave.", + "timestamps": { + "start": "00:07:39", + "end": "00:07:44" + } + } + ], + "question_timestamps": "00:07:46", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From the cabinet left of the sink.", + "B": "From the cabinet below the microwave.", + "C": "You should get it from the cabinet to the left of the sink and above the microwave.", + "D": "You should get it from the cabinet to the right of the sink and below the microwave.", + "E": "You should get it from the cabinet to the right of the sink and above the microwave." + }, + "answer_label": "E", + "answer_text": "You should get it from the cabinet to the right of the sink and above the microwave.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_11", + "correct_answer": "You should get it from the cabinet to the right of the sink and above the microwave.", + "original_options": { + "A": "From the cabinet left of the sink.", + "B": "From the cabinet below the microwave.", + "C": "You should get it from the cabinet to the left of the sink and above the microwave.", + "D": "You should get it from the cabinet to the right of the sink and below the microwave.", + "E": "You should get it from the cabinet to the right of the sink and above the microwave." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 17 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_12", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to get more cups, where should I go to get them?", + "answer": "You should get it from the cabinet above the microwave, the counter opposite the fridge, or the cupboard to the right of the door.", + "question_chinese": "我如果想拿更多的杯子，我应该去哪里拿？", + "answer_chinese": "你应该去微波炉上方的柜子、冰箱对面的吧台、门右边的碗柜里拿。", + "evidence": [ + { + "description": "There are many cups in the cabinet above the microwave.", + "timestamps": { + "start": "00:07:39", + "end": "00:07:44" + } + }, + { + "description": "There are some cups on the bar opposite the refrigerator.", + "timestamps": { + "start": "00:07:58", + "end": "00:08:03" + } + }, + { + "description": "You put the two cups on the right side of the sink into the cupboard.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:17" + } + } + ], + "question_timestamps": "00:18:17", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "From the cabinet below the microwave.", + "B": "From the counter beside the fridge.", + "C": "You should get it from the cabinet above the microwave, the counter opposite the fridge, or the cupboard to the right of the door.", + "D": "You should get them from the cabinet below the microwave, the counter opposite the fridge, or the cupboard right of the door.", + "E": "You should get them from the cabinet above the microwave, the counter near the sink, or the cupboard to the right." + }, + "answer_label": "C", + "answer_text": "You should get it from the cabinet above the microwave, the counter opposite the fridge, or the cupboard to the right of the door.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_12", + "correct_answer": "You should get it from the cabinet above the microwave, the counter opposite the fridge, or the cupboard to the right of the door.", + "original_options": { + "A": "From the cabinet below the microwave.", + "B": "From the counter beside the fridge.", + "C": "You should get it from the cabinet above the microwave, the counter opposite the fridge, or the cupboard to the right of the door.", + "D": "You should get them from the cabinet below the microwave, the counter opposite the fridge, or the cupboard right of the door.", + "E": "You should get them from the cabinet above the microwave, the counter near the sink, or the cupboard to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 21 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_13", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Are the plate and the cup close to each other now? Approximately how far apart are they?", + "answer": "Very close, the cup is above and to the right of the plate, about ten cm apart.", + "question_chinese": "盘子和杯子现在离得近吗?大概多远?", + "answer_chinese": "很近,杯子在盘子的右上方,相距约10cm。", + "evidence": [ + { + "description": "The plate and the cup are about 10 cm apart.", + "timestamps": { + "start": "00:13:29", + "end": "00:13:38" + } + } + ], + "question_timestamps": "00:13:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Not very close, about 50 cm.", + "B": "Very close, about five cm.", + "C": "Somewhat close, the cup is below and to the right of the plate, about twenty cm apart.", + "D": "Very close, the cup is above and to the right of the plate, about ten cm apart.", + "E": "Quite far, the cup is above and to the left of the plate, about thirty cm apart." + }, + "answer_label": "D", + "answer_text": "Very close, the cup is above and to the right of the plate, about ten cm apart.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_13", + "correct_answer": "Very close, the cup is above and to the right of the plate, about ten cm apart.", + "original_options": { + "A": "Not very close, about 50 cm.", + "B": "Very close, about five cm.", + "C": "Somewhat close, the cup is below and to the right of the plate, about twenty cm apart.", + "D": "Very close, the cup is above and to the right of the plate, about ten cm apart.", + "E": "Quite far, the cup is above and to the left of the plate, about thirty cm apart." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 17 + ] + }, + { + "video_uid": "8d3337ba-61a1-4c23-abb3-2728ba21adad", + "video_length": "00:20:29", + "qid": "8d3337ba-61a1-4c23-abb3-2728ba21adad_14", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Are the plate and the cup close to each other now? Approximately how far apart are they?", + "answer": "Not very close, the cup is to the left of the plate, about fifty cm apart.", + "question_chinese": "盘子和杯子现在离得近吗?大概多远?", + "answer_chinese": "不太近，杯子在盘子的左边，相距约25cm。", + "evidence": [ + { + "description": "The cup is on the left side of the sink, and the plate is on the right side of the sink, about 50 cm apart.", + "timestamps": { + "start": "00:17:39", + "end": "00:17:40" + } + } + ], + "question_timestamps": "00:17:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Not very close, about 20 cm.", + "B": "Not very close, about one meter.", + "C": "Not very close, the cup is to the right of the plate, about fifty cm apart.", + "D": "Very close, the cup is to the left of the plate, about ten cm apart.", + "E": "Not very close, the cup is to the left of the plate, about fifty cm apart." + }, + "answer_label": "E", + "answer_text": "Not very close, the cup is to the left of the plate, about fifty cm apart.", + "q_id": "8d3337ba-61a1-4c23-abb3-2728ba21adad_14", + "correct_answer": "Not very close, the cup is to the left of the plate, about fifty cm apart.", + "original_options": { + "A": "Not very close, about 20 cm.", + "B": "Not very close, about one meter.", + "C": "Not very close, the cup is to the right of the plate, about fifty cm apart.", + "D": "Very close, the cup is to the left of the plate, about ten cm apart.", + "E": "Not very close, the cup is to the left of the plate, about fifty cm apart." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + } +] diff --git a/QAs/8d4f2bbd-c1ee-4f75-8404-37baba706b92-mcq.json b/QAs/8d4f2bbd-c1ee-4f75-8404-37baba706b92-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6bc576b2ba7cae3fe5c463bedfcf6cdb2cbcc79b --- /dev/null +++ b/QAs/8d4f2bbd-c1ee-4f75-8404-37baba706b92-mcq.json @@ -0,0 +1,1068 @@ +[ + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_11", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many screws have I screwed in so far?", + "answer": "You have screwed in a total of one screw.", + "question_chinese": "到目前为止,我一共钉了几个螺丝钉?", + "answer_chinese": "你一共钉了一个螺丝钉。", + "evidence": [ + { + "description": "You tightened the first screw.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two screws.", + "B": "You have screwed in a total of one screw.", + "C": "Three screws.", + "D": "You have screwed in a total of four screws so far in the project.", + "E": "You have screwed in a total of six screws up to now in total." + }, + "answer_label": "B", + "answer_text": "You have screwed in a total of one screw.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_11", + "correct_answer": "You have screwed in a total of one screw.", + "original_options": { + "A": "Two screws.", + "B": "You have screwed in a total of one screw.", + "C": "Three screws.", + "D": "You have screwed in a total of four screws so far in the project.", + "E": "You have screwed in a total of six screws up to now in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_12", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many screws have I screwed in so far?", + "answer": "You have screwed in a total of five screws.", + "question_chinese": "到目前为止,我一共钉了几个螺丝钉?", + "answer_chinese": "你一共钉了五个螺丝钉。", + "evidence": [ + { + "description": "You tightened the first screw.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + }, + { + "description": "You have tightened the second screw.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:18" + } + }, + { + "description": "You have tightened the third screw.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:45" + } + }, + { + "description": "You have tightened the fourth screw.", + "timestamps": { + "start": "00:00:52", + "end": "00:00:55" + } + }, + { + "description": "You have fastened the fifth screw.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:18:03", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have screwed in a total of five screws.", + "B": "You have screwed in three screws.", + "C": "You have screwed in six screws.", + "D": "You have screwed in a total of four screws so far, not including any screws you only positioned.", + "E": "You have screwed in a total of seven screws so far, counting every screw you fully tightened into place." + }, + "answer_label": "A", + "answer_text": "You have screwed in a total of five screws.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_12", + "correct_answer": "You have screwed in a total of five screws.", + "original_options": { + "A": "You have screwed in a total of five screws.", + "B": "You have screwed in three screws.", + "C": "You have screwed in six screws.", + "D": "You have screwed in a total of four screws so far, not including any screws you only positioned.", + "E": "You have screwed in a total of seven screws so far, counting every screw you fully tightened into place." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_4_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Now that I’m done using the tape measure, where should I put it?", + "answer": "You should put the tape measure back on the counter with the three-column drawers on your right-hand side.", + "question_chinese": "卷尺用好了，我应该把它放在哪里？", + "answer_chinese": "你应该把卷尺放回右手边的三列抽屉的台子上。", + "evidence": [ + { + "description": "You placed the tape measure on the counter with three drawers on your right rear", + "timestamps": { + "start": "00:10:54", + "end": "00:10:56" + } + } + ], + "question_timestamps": "00:10:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it on the left counter.", + "B": "Set it on the right-hand counter.", + "C": "You should put the tape measure back on the counter with the two-column drawers on your right-hand side.", + "D": "You should put the tape measure back on the counter with the four-column drawers on your right-hand side.", + "E": "You should put the tape measure back on the counter with the three-column drawers on your right-hand side." + }, + "answer_label": "E", + "answer_text": "You should put the tape measure back on the counter with the three-column drawers on your right-hand side.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_4_T1", + "correct_answer": "You should put the tape measure back on the counter with the three-column drawers on your right-hand side.", + "original_options": { + "A": "Put it on the left counter.", + "B": "Set it on the right-hand counter.", + "C": "You should put the tape measure back on the counter with the two-column drawers on your right-hand side.", + "D": "You should put the tape measure back on the counter with the four-column drawers on your right-hand side.", + "E": "You should put the tape measure back on the counter with the three-column drawers on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_4_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Now that I’m done using the tape measure, where should I put it?", + "answer": "You should put the tape measure back on the counter with the three-column drawers to your front-left.", + "question_chinese": "卷尺用好了，我应该把它放在哪里？", + "answer_chinese": "你应该把卷尺放回左前方的三列抽屉的台子上。", + "evidence": [ + { + "description": "You placed the tape measure on the counter with three drawers on your right rear", + "timestamps": { + "start": "00:10:54", + "end": "00:10:56" + } + } + ], + "question_timestamps": "00:11:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You should put the tape measure back on the counter with the three-column drawers to your front-left.", + "B": "Put it on the counter to your left.", + "C": "Put it on the counter behind you.", + "D": "You should put the tape measure back on the counter with the four-column drawers to your front-left.", + "E": "You should put the tape measure back on the counter with the three-column drawers to your front-right." + }, + "answer_label": "A", + "answer_text": "You should put the tape measure back on the counter with the three-column drawers to your front-left.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_4_T2", + "correct_answer": "You should put the tape measure back on the counter with the three-column drawers to your front-left.", + "original_options": { + "A": "You should put the tape measure back on the counter with the three-column drawers to your front-left.", + "B": "Put it on the counter to your left.", + "C": "Put it on the counter behind you.", + "D": "You should put the tape measure back on the counter with the four-column drawers to your front-left.", + "E": "You should put the tape measure back on the counter with the three-column drawers to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 17 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_1_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to get more screws, where should I look?", + "answer": "You should look for it in the plastic box in the storage room; it’s to your back-left.", + "question_chinese": "如果我想获取更多的螺丝应该去哪找呢？", + "answer_chinese": "你应该去储藏间里的塑料盒里找，它在你的左后方。", + "evidence": [ + { + "description": "There are many screws in the plastic box.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + }, + { + "description": "You put the plastic box in the storage room.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:02:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Look in the plastic box to your back-right.", + "B": "You should look for it in the plastic box in the storage room; it’s to your back-left.", + "C": "Look in the plastic box to your front-left.", + "D": "You should look for it in the plastic box in the storage room; it’s to your front-right.", + "E": "You should look for it in the plastic box in the storage room; it’s directly behind you." + }, + "answer_label": "B", + "answer_text": "You should look for it in the plastic box in the storage room; it’s to your back-left.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_1_T1", + "correct_answer": "You should look for it in the plastic box in the storage room; it’s to your back-left.", + "original_options": { + "A": "Look in the plastic box to your back-right.", + "B": "You should look for it in the plastic box in the storage room; it’s to your back-left.", + "C": "Look in the plastic box to your front-left.", + "D": "You should look for it in the plastic box in the storage room; it’s to your front-right.", + "E": "You should look for it in the plastic box in the storage room; it’s directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_1_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to get more screws, where should I look?", + "answer": "You should look for it in the plastic box in the storage room; it’s to your back-left.", + "question_chinese": "如果我想获取更多的螺丝应该去哪找呢？", + "answer_chinese": "你应该去储藏间里的塑料盒里找，它在你的左后方。", + "evidence": [ + { + "description": "There are many screws in the plastic box.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + }, + { + "description": "You put the plastic box in the storage room.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:17:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Look in the plastic box to your back-right.", + "B": "Look in the plastic box to your front-left.", + "C": "You should look for it in the plastic box in the storage room; it’s to your back-left.", + "D": "You should look for it in the plastic box in the storage room; it’s to your front-right.", + "E": "You should look for it in the plastic box in the storage room; it’s directly behind you." + }, + "answer_label": "C", + "answer_text": "You should look for it in the plastic box in the storage room; it’s to your back-left.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_1_T2", + "correct_answer": "You should look for it in the plastic box in the storage room; it’s to your back-left.", + "original_options": { + "A": "Look in the plastic box to your back-right.", + "B": "Look in the plastic box to your front-left.", + "C": "You should look for it in the plastic box in the storage room; it’s to your back-left.", + "D": "You should look for it in the plastic box in the storage room; it’s to your front-right.", + "E": "You should look for it in the plastic box in the storage room; it’s directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_5_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was standing in front of the workbench, how far was I from the pencil?", + "answer": "You were about 80 cm from the pencil; it was on the drawer to your front-left.", + "question_chinese": "当我站在操作台前时,离铅笔有多远?", + "answer_chinese": "你离铅笔大约80cm,在你左前方的抽屉上。", + "evidence": [ + { + "description": "You placed the pencil on the drawer.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:35" + } + }, + { + "description": "You stand in front of the console, approximately 80cm away from the pencil.", + "timestamps": { + "start": "00:06:37", + "end": "00:06:39" + } + } + ], + "question_timestamps": "00:06:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You were about 80 cm from the pencil; it was on the drawer to your front-left.", + "B": "You were about 78 cm away.", + "C": "You were about 82 cm away.", + "D": "You were about 81 cm from the pencil; it was on the drawer to your front-left.", + "E": "You were about 79 cm from the pencil; it was on the drawer to your front-left." + }, + "answer_label": "A", + "answer_text": "You were about 80 cm from the pencil; it was on the drawer to your front-left.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_5_T1", + "correct_answer": "You were about 80 cm from the pencil; it was on the drawer to your front-left.", + "original_options": { + "A": "You were about 80 cm from the pencil; it was on the drawer to your front-left.", + "B": "You were about 78 cm away.", + "C": "You were about 82 cm away.", + "D": "You were about 81 cm from the pencil; it was on the drawer to your front-left.", + "E": "You were about 79 cm from the pencil; it was on the drawer to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_5_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was standing in front of the workbench, how far was I from the pencil?", + "answer": "You were about 80 cm from the pencil; it was on the drawer to your back-left.", + "question_chinese": "当我站在操作台前时,离铅笔有多远?", + "answer_chinese": "你离铅笔大约80cm,在你左后方的抽屉上。", + "evidence": [ + { + "description": "You placed the pencil on the drawer.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:35" + } + }, + { + "description": "You stand in front of the console, approximately 80cm away from the pencil.", + "timestamps": { + "start": "00:06:37", + "end": "00:06:39" + } + } + ], + "question_timestamps": "00:07:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You were about 78 cm away.", + "B": "You were about 80 cm from the pencil; it was on the drawer to your back-left.", + "C": "You were about 82 cm away.", + "D": "You were about 81 cm from the pencil; it was on the drawer to your back-left.", + "E": "You were about 79 cm from the pencil; it was on the drawer to your back-left." + }, + "answer_label": "B", + "answer_text": "You were about 80 cm from the pencil; it was on the drawer to your back-left.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_5_T2", + "correct_answer": "You were about 80 cm from the pencil; it was on the drawer to your back-left.", + "original_options": { + "A": "You were about 78 cm away.", + "B": "You were about 80 cm from the pencil; it was on the drawer to your back-left.", + "C": "You were about 82 cm away.", + "D": "You were about 81 cm from the pencil; it was on the drawer to your back-left.", + "E": "You were about 79 cm from the pencil; it was on the drawer to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_15", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I place the plastic box on the wooden board in total?", + "answer": "Twice.", + "question_chinese": "我一共将塑料盒摆在木板上多少次?", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "You placed the plastic box on the wooden board for the first time.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "You placed the plastic box on the wooden board for the second time.", + "timestamps": { + "start": "00:14:02", + "end": "00:14:04" + } + } + ], + "question_timestamps": "00:14:56", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times in total, counting every time the plastic box rested on the wooden board.", + "D": "Twice.", + "E": "Five times in total, including every separate placement of the plastic box onto the wooden board." + }, + "answer_label": "D", + "answer_text": "Twice.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_15", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times in total, counting every time the plastic box rested on the wooden board.", + "D": "Twice.", + "E": "Five times in total, including every separate placement of the plastic box onto the wooden board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 16 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_16", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I place the plastic box on the wooden board in total?", + "answer": "Three times.", + "question_chinese": "我一共将塑料盒摆在木板上多少次?", + "answer_chinese": "三次。", + "evidence": [ + { + "description": "You placed the plastic box on the wooden board for the first time.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "You placed the plastic box on the wooden board for the second time.", + "timestamps": { + "start": "00:14:02", + "end": "00:14:04" + } + }, + { + "description": "You placed the plastic box on the wooden board for the third time.", + "timestamps": { + "start": "00:39:54", + "end": "00:39:56" + } + } + ], + "question_timestamps": "00:39:58", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Four times.", + "C": "You placed the plastic box on the wooden board two times in total during the task.", + "D": "You ended up setting the plastic box onto the wooden board four separate times overall.", + "E": "Three times." + }, + "answer_label": "E", + "answer_text": "Three times.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_16", + "correct_answer": "Three times.", + "original_options": { + "A": "Two times.", + "B": "Four times.", + "C": "You placed the plastic box on the wooden board two times in total during the task.", + "D": "You ended up setting the plastic box onto the wooden board four separate times overall.", + "E": "Three times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 15 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_19", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I use a table saw to split wide boards?", + "answer": "You cut the wide plank once in total.", + "question_chinese": "我一共用台锯分割了几次宽木板?", + "answer_chinese": "你一共分割了一次宽木板。", + "evidence": [ + { + "description": "You are splitting a wide plank of wood for the first time.", + "timestamps": { + "start": "00:11:57", + "end": "00:12:02" + } + } + ], + "question_timestamps": "00:12:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Three times.", + "C": "You cut the wide plank once in total.", + "D": "You used the table saw to split the wide boards two times in total.", + "E": "You ran the wide plank through the table saw three separate times overall." + }, + "answer_label": "C", + "answer_text": "You cut the wide plank once in total.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_19", + "correct_answer": "You cut the wide plank once in total.", + "original_options": { + "A": "Twice.", + "B": "Three times.", + "C": "You cut the wide plank once in total.", + "D": "You used the table saw to split the wide boards two times in total.", + "E": "You ran the wide plank through the table saw three separate times overall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 14, + 13 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_20", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I use a table saw to split wide boards?", + "answer": "You have cut the wide plank three times in total.", + "question_chinese": "我一共用台锯分割了几次宽木板?", + "answer_chinese": "你一共分割了三次宽木板。", + "evidence": [ + { + "description": "You are splitting a wide plank of wood for the first time.", + "timestamps": { + "start": "00:11:57", + "end": "00:12:02" + } + }, + { + "description": "You are cutting the wide plank for the second time.", + "timestamps": { + "start": "00:12:07", + "end": "00:12:13" + } + }, + { + "description": "You are splitting the wide plank for the third time.", + "timestamps": { + "start": "00:12:18", + "end": "00:12:23" + } + } + ], + "question_timestamps": "00:40:23", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have cut the wide plank three times in total.", + "B": "Once.", + "C": "Four times.", + "D": "You used the table saw to split the wide boards one time in total.", + "E": "You used the table saw to split the wide boards four times altogether." + }, + "answer_label": "A", + "answer_text": "You have cut the wide plank three times in total.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_20", + "correct_answer": "You have cut the wide plank three times in total.", + "original_options": { + "A": "You have cut the wide plank three times in total.", + "B": "Once.", + "C": "Four times.", + "D": "You used the table saw to split the wide boards one time in total.", + "E": "You used the table saw to split the wide boards four times altogether." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 14, + 13 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_2_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You’re holding an electric screwdriver and screws in your hand right now.", + "question_chinese": "我手里刚刚拿着什么东西？", + "answer_chinese": "你手里现在拿着电动螺丝刀和螺丝钉。", + "evidence": [ + { + "description": "You are holding a screw in your left hand and an electric screwdriver in your right hand.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Only the electric screwdriver.", + "B": "Only the screws.", + "C": "You’re holding an electric screwdriver and screws in your hand right now.", + "D": "You’re holding the electric screwdriver and the glue in your hand right now.", + "E": "You’re holding a glue bottle in your hand right now, with nothing else." + }, + "answer_label": "C", + "answer_text": "You’re holding an electric screwdriver and screws in your hand right now.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_2_T1", + "correct_answer": "You’re holding an electric screwdriver and screws in your hand right now.", + "original_options": { + "A": "Only the electric screwdriver.", + "B": "Only the screws.", + "C": "You’re holding an electric screwdriver and screws in your hand right now.", + "D": "You’re holding the electric screwdriver and the glue in your hand right now.", + "E": "You’re holding a glue bottle in your hand right now, with nothing else." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 14 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_2_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You’re holding the glue in your hand right now; it’s right in front of you.", + "question_chinese": "我手里刚刚拿着什么东西？", + "answer_chinese": "你手里现在拿着胶水，就放在你的前方。", + "evidence": [ + { + "description": "You are holding a screw in your left hand and an electric screwdriver in your right hand.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:18:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A screwdriver and screws.", + "B": "A paintbrush.", + "C": "You’re holding the glue in your hand, but it’s off to your left, not in front of you.", + "D": "You’re holding the glue in your hand right now; it’s right in front of you.", + "E": "You’re holding an electric screwdriver and screws right now instead of glue in your hand." + }, + "answer_label": "D", + "answer_text": "You’re holding the glue in your hand right now; it’s right in front of you.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_2_T2", + "correct_answer": "You’re holding the glue in your hand right now; it’s right in front of you.", + "original_options": { + "A": "A screwdriver and screws.", + "B": "A paintbrush.", + "C": "You’re holding the glue in your hand, but it’s off to your left, not in front of you.", + "D": "You’re holding the glue in your hand right now; it’s right in front of you.", + "E": "You’re holding an electric screwdriver and screws right now instead of glue in your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 20, + 16 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_7_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Before I picked up my phone to reply to messages, what tools were on the workbench?", + "answer": "Before you picked up your phone to reply to a message, the workbench had an electric screwdriver, a glue bottle, and a plastic box of screws on it; they were all to your left.", + "question_chinese": "我拿起手机回消息之前,操作台上放着哪些工具?", + "answer_chinese": "你拿起手机回消息之前,操作台上放着电动螺丝刀,胶瓶,螺丝塑料盒，都放在你的左方。", + "evidence": [ + { + "description": "The electric screwdriver is placed on the side closer to you.", + "timestamps": { + "start": "00:20:53", + "end": "00:20:55" + } + }, + { + "description": "On the side farther away from you, there are plastic bottles and plastic boxes for screws.", + "timestamps": { + "start": "00:20:25", + "end": "00:20:29" + } + } + ], + "question_timestamps": "00:25:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They were all to your right.", + "B": "They were all in front of you.", + "C": "Before you picked up your phone to reply to a message, the workbench had an electric screwdriver, a glue bottle, and a plastic box of screws on it; they were all to your left.", + "D": "Before replying, the workbench had an electric screwdriver and a glue bottle on it, both to your left.", + "E": "Before replying, the workbench had only a plastic box of screws on it, sitting to your left." + }, + "answer_label": "C", + "answer_text": "Before you picked up your phone to reply to a message, the workbench had an electric screwdriver, a glue bottle, and a plastic box of screws on it; they were all to your left.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_7_T1", + "correct_answer": "Before you picked up your phone to reply to a message, the workbench had an electric screwdriver, a glue bottle, and a plastic box of screws on it; they were all to your left.", + "original_options": { + "A": "They were all to your right.", + "B": "They were all in front of you.", + "C": "Before you picked up your phone to reply to a message, the workbench had an electric screwdriver, a glue bottle, and a plastic box of screws on it; they were all to your left.", + "D": "Before replying, the workbench had an electric screwdriver and a glue bottle on it, both to your left.", + "E": "Before replying, the workbench had only a plastic box of screws on it, sitting to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 17 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_7_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Before I picked up my phone to reply to messages, what tools were on the workbench?", + "answer": "Before you picked up your phone to reply to a message, the workbench had an electric screwdriver, a glue bottle, and a plastic box of screws on it; they’re all in front of you.", + "question_chinese": "我拿起手机回消息之前,操作台上放着哪些工具?", + "answer_chinese": "你拿起手机回消息之前,操作台上放着电动螺丝刀,胶瓶,螺丝塑料盒，目前都在你的前方。", + "evidence": [ + { + "description": "The electric screwdriver is placed on the side closer to you.", + "timestamps": { + "start": "00:20:53", + "end": "00:20:55" + } + }, + { + "description": "On the side farther away from you, there are plastic bottles and plastic boxes for screws.", + "timestamps": { + "start": "00:20:25", + "end": "00:20:29" + } + } + ], + "question_timestamps": "00:30:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They were all to your left.", + "B": "They were all to your right.", + "C": "Before you replied, the electric screwdriver, glue bottle, and screw box were all behind you on the workbench.", + "D": "Before you picked up your phone to reply to a message, the workbench had an electric screwdriver, a glue bottle, and a plastic box of screws on it; they’re all in front of you.", + "E": "Before you replied, the workbench had a glue bottle and a plastic box of screws in front of you, but no screwdriver." + }, + "answer_label": "D", + "answer_text": "Before you picked up your phone to reply to a message, the workbench had an electric screwdriver, a glue bottle, and a plastic box of screws on it; they’re all in front of you.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_7_T2", + "correct_answer": "Before you picked up your phone to reply to a message, the workbench had an electric screwdriver, a glue bottle, and a plastic box of screws on it; they’re all in front of you.", + "original_options": { + "A": "They were all to your left.", + "B": "They were all to your right.", + "C": "Before you replied, the electric screwdriver, glue bottle, and screw box were all behind you on the workbench.", + "D": "Before you picked up your phone to reply to a message, the workbench had an electric screwdriver, a glue bottle, and a plastic box of screws on it; they’re all in front of you.", + "E": "Before you replied, the workbench had a glue bottle and a plastic box of screws in front of you, but no screwdriver." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 22 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_17", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Are the electric drill and the glue bottle close to each other now? Roughly how far are they?", + "answer": "Very close, the electric drill is on the left side of the plastic bottle, about 20cm away.", + "question_chinese": "电钻和胶瓶现在离得近吗?大概多远?", + "answer_chinese": "很近,电钻在胶瓶的左边,相距约20cm。", + "evidence": [ + { + "description": "The electric drill is located on the left side of the plastic bottle, approximately 20cm away.", + "timestamps": { + "start": "00:14:34", + "end": "00:14:38" + } + } + ], + "question_timestamps": "00:14:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They’re far apart, about a meter.", + "B": "Very close, about 5cm apart.", + "C": "They aren’t very close; the drill sits left of the bottle at roughly half a meter away.", + "D": "Very close, the electric drill is on the left side of the plastic bottle, about 20cm away.", + "E": "Very close, but the drill is on the right side of the plastic bottle, around 20cm away." + }, + "answer_label": "D", + "answer_text": "Very close, the electric drill is on the left side of the plastic bottle, about 20cm away.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_17", + "correct_answer": "Very close, the electric drill is on the left side of the plastic bottle, about 20cm away.", + "original_options": { + "A": "They’re far apart, about a meter.", + "B": "Very close, about 5cm apart.", + "C": "They aren’t very close; the drill sits left of the bottle at roughly half a meter away.", + "D": "Very close, the electric drill is on the left side of the plastic bottle, about 20cm away.", + "E": "Very close, but the drill is on the right side of the plastic bottle, around 20cm away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 17 + ] + }, + { + "video_uid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92", + "video_length": "00:41:03", + "qid": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_18", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Are the electric drill and the glue bottle close to each other now? Roughly how far are they?", + "answer": "It is relatively far away, with the electric drill located on the right side of the plastic bottle, approximately 1m away.", + "question_chinese": "电钻和胶瓶现在离得近吗?大概多远?", + "answer_chinese": "较远,电钻在胶瓶的右边,相距约1m.", + "evidence": [ + { + "description": "The electric drill is on the bench to your right and behind you.", + "timestamps": { + "start": "00:16:55", + "end": "00:16:56" + } + }, + { + "description": "The plastic bottle is on the workbench.", + "timestamps": { + "start": "00:17:04", + "end": "00:17:06" + } + } + ], + "question_timestamps": "00:18:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "They’re very close, about 20cm.", + "B": "The drill is left of the bottle.", + "C": "It is relatively far away, with the electric drill located on the right side of the plastic bottle, approximately 50cm away.", + "D": "It is relatively far away, with the electric drill located on the right side of the plastic bottle, approximately 2m away.", + "E": "It is relatively far away, with the electric drill located on the right side of the plastic bottle, approximately 1m away." + }, + "answer_label": "E", + "answer_text": "It is relatively far away, with the electric drill located on the right side of the plastic bottle, approximately 1m away.", + "q_id": "8d4f2bbd-c1ee-4f75-8404-37baba706b92_18", + "correct_answer": "It is relatively far away, with the electric drill located on the right side of the plastic bottle, approximately 1m away.", + "original_options": { + "A": "They’re very close, about 20cm.", + "B": "The drill is left of the bottle.", + "C": "It is relatively far away, with the electric drill located on the right side of the plastic bottle, approximately 50cm away.", + "D": "It is relatively far away, with the electric drill located on the right side of the plastic bottle, approximately 2m away.", + "E": "It is relatively far away, with the electric drill located on the right side of the plastic bottle, approximately 1m away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 21 + ] + } +] diff --git a/QAs/8e6753c7-1d9c-469d-b1a5-6e6424955494-mcq.json b/QAs/8e6753c7-1d9c-469d-b1a5-6e6424955494-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e6c1f8a419e2808bcb481d36591ca8a66fb07652 --- /dev/null +++ b/QAs/8e6753c7-1d9c-469d-b1a5-6e6424955494-mcq.json @@ -0,0 +1,1287 @@ +[ + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_13", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Did anyone come to my side during my assembly process?", + "answer": "Someone comes next to you.", + "question_chinese": "在我组装过程中有人来我旁边吗", + "answer_chinese": "有人来你旁边", + "evidence": [ + { + "description": "Someone comes next to you.", + "timestamps": { + "start": "00:13:55", + "end": "00:14:04" + } + } + ], + "question_timestamps": "00:18:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, nobody came next to you.", + "B": "Someone comes next to you." + }, + "answer_label": "B", + "answer_text": "Someone comes next to you.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_13", + "correct_answer": "Someone comes next to you.", + "original_options": { + "A": "No, nobody came next to you.", + "B": "Someone comes next to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_14", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Did anyone come to my side during my assembly process?", + "answer": "Someone comes next to you.", + "question_chinese": "在我组装过程中有人来我旁边吗", + "answer_chinese": "有人来你旁边", + "evidence": [ + { + "description": "Someone comes next to you.", + "timestamps": { + "start": "00:13:55", + "end": "00:14:04" + } + } + ], + "question_timestamps": "00:25:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Someone comes next to you.", + "B": "No, nobody came next to you." + }, + "answer_label": "A", + "answer_text": "Someone comes next to you.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_14", + "correct_answer": "Someone comes next to you.", + "original_options": { + "A": "Someone comes next to you.", + "B": "No, nobody came next to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_15", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which finger should I use when referring to the instruction manual?", + "answer": "You use your middle finger.", + "question_chinese": "我指说明书时用哪个手指头", + "answer_chinese": "你用中指", + "evidence": [ + { + "description": "You use your middle finger.", + "timestamps": { + "start": "00:13:50", + "end": "00:13:57" + } + }, + { + "description": "You use your middle finger.", + "timestamps": { + "start": "00:14:01", + "end": "00:14:06" + } + } + ], + "question_timestamps": "00:18:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You use your middle finger.", + "B": "You use your index finger.", + "C": "You use your ring finger.", + "D": "You use your thumb, not the middle finger, when pointing at the instruction manual.", + "E": "You use your little finger to indicate the instruction manual while keeping your other fingers curled." + }, + "answer_label": "A", + "answer_text": "You use your middle finger.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_15", + "correct_answer": "You use your middle finger.", + "original_options": { + "A": "You use your middle finger.", + "B": "You use your index finger.", + "C": "You use your ring finger.", + "D": "You use your thumb, not the middle finger, when pointing at the instruction manual.", + "E": "You use your little finger to indicate the instruction manual while keeping your other fingers curled." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_16", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which finger should I use when referring to the instruction manual?", + "answer": "You use your middle finger.", + "question_chinese": "我指说明书时用哪个手指头", + "answer_chinese": "你用中指", + "evidence": [ + { + "description": "You use your middle finger.", + "timestamps": { + "start": "00:13:50", + "end": "00:13:57" + } + }, + { + "description": "You use your middle finger.", + "timestamps": { + "start": "00:14:01", + "end": "00:14:06" + } + }, + { + "description": "You use your middle finger.", + "timestamps": { + "start": "00:22:38", + "end": "00:22:45" + } + } + ], + "question_timestamps": "00:25:01", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You use your index finger.", + "B": "You use your middle finger.", + "C": "You use your ring finger.", + "D": "You use your thumb finger to point to the manual during the instructions and steps.", + "E": "You use your little finger to indicate the manual while checking the directions in the booklet." + }, + "answer_label": "B", + "answer_text": "You use your middle finger.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_16", + "correct_answer": "You use your middle finger.", + "original_options": { + "A": "You use your index finger.", + "B": "You use your middle finger.", + "C": "You use your ring finger.", + "D": "You use your thumb finger to point to the manual during the instructions and steps.", + "E": "You use your little finger to indicate the manual while checking the directions in the booklet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_8", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the scissors?", + "answer": "You used it nine times.", + "question_chinese": "我用了多少次剪刀", + "answer_chinese": "你用了九次", + "evidence": [ + { + "description": "You used it nine times.", + "timestamps": { + "start": "00:00:00", + "end": "00:06:10" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used it seven times.", + "B": "You used it eleven times.", + "C": "You used the scissors five times overall during the task, across several separate steps.", + "D": "You used it nine times.", + "E": "You used the scissors twice in total across the steps, only at the very beginning." + }, + "answer_label": "D", + "answer_text": "You used it nine times.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_8", + "correct_answer": "You used it nine times.", + "original_options": { + "A": "You used it seven times.", + "B": "You used it eleven times.", + "C": "You used the scissors five times overall during the task, across several separate steps.", + "D": "You used it nine times.", + "E": "You used the scissors twice in total across the steps, only at the very beginning." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_7", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the scissors?", + "answer": "You used the scissors thirteen times in total.", + "question_chinese": "我用了多少次剪刀", + "answer_chinese": "你一共用了十三次剪刀", + "evidence": [ + { + "description": "You used it nine times.", + "timestamps": { + "start": "00:00:00", + "end": "00:06:10" + } + }, + { + "description": "You used the scissors thirteen times in total.", + "timestamps": { + "start": "00:06:30", + "end": "00:12:20" + } + } + ], + "question_timestamps": "00:12:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You used them eleven times.", + "B": "You used them twenty times.", + "C": "You used the scissors thirteen times in total.", + "D": "You used the scissors seven times in total during the task, with only a few cutting steps.", + "E": "You used the scissors five times in total across the steps, mostly near the end." + }, + "answer_label": "C", + "answer_text": "You used the scissors thirteen times in total.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_7", + "correct_answer": "You used the scissors thirteen times in total.", + "original_options": { + "A": "You used them eleven times.", + "B": "You used them twenty times.", + "C": "You used the scissors thirteen times in total.", + "D": "You used the scissors seven times in total during the task, with only a few cutting steps.", + "E": "You used the scissors five times in total across the steps, mostly near the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 15 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_8_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times did I point at the instruction manual with my finger?", + "answer": "You pointed at the instruction manual once with your finger, and it was in front of you.", + "question_chinese": "我用手指指了多少次说明书？", + "answer_chinese": "你用手指指了一次说明书，说明书在你前面", + "evidence": [ + { + "description": "You use your middle finger.", + "timestamps": { + "start": "00:13:50", + "end": "00:13:57" + } + } + ], + "question_timestamps": "00:13:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You pointed at the instruction manual once with your finger, and it was in front of you.", + "B": "You pointed at it twice.", + "C": "You pointed at it three times.", + "D": "You pointed at the instruction manual once with your finger, and it was to your left.", + "E": "You pointed at the instruction manual once with your finger, and it was to your right." + }, + "answer_label": "A", + "answer_text": "You pointed at the instruction manual once with your finger, and it was in front of you.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_8_T1", + "correct_answer": "You pointed at the instruction manual once with your finger, and it was in front of you.", + "original_options": { + "A": "You pointed at the instruction manual once with your finger, and it was in front of you.", + "B": "You pointed at it twice.", + "C": "You pointed at it three times.", + "D": "You pointed at the instruction manual once with your finger, and it was to your left.", + "E": "You pointed at the instruction manual once with your finger, and it was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_8_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times did I point at the instruction manual with my finger?", + "answer": "You pointed at the instruction manual twice with your finger, and it was in front of you.", + "question_chinese": "我用手指指了多少次说明书？", + "answer_chinese": "你用手指指了两次说明书，说明书在你前面", + "evidence": [ + { + "description": "You use your middle finger.", + "timestamps": { + "start": "00:13:50", + "end": "00:13:57" + } + }, + { + "description": "You use your middle finger.", + "timestamps": { + "start": "00:14:01", + "end": "00:14:06" + } + } + ], + "question_timestamps": "00:14:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You pointed at it once.", + "B": "You pointed at the instruction manual twice with your finger, and it was in front of you.", + "C": "You pointed at it three times.", + "D": "You pointed at the instruction manual twice with your finger, and it was to your left.", + "E": "You pointed at the instruction manual twice with your finger, and it was to your right." + }, + "answer_label": "B", + "answer_text": "You pointed at the instruction manual twice with your finger, and it was in front of you.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_8_T2", + "correct_answer": "You pointed at the instruction manual twice with your finger, and it was in front of you.", + "original_options": { + "A": "You pointed at it once.", + "B": "You pointed at the instruction manual twice with your finger, and it was in front of you.", + "C": "You pointed at it three times.", + "D": "You pointed at the instruction manual twice with your finger, and it was to your left.", + "E": "You pointed at the instruction manual twice with your finger, and it was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_6_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Did my parts fall down?", + "answer": "Your part didn’t fall; it’s in front of you.", + "question_chinese": "我的零件掉下去了吗？", + "answer_chinese": "你的零件没有掉下去，在你的前面", + "evidence": [ + { + "description": "Didn't fall down.", + "timestamps": { + "start": "00:12:30", + "end": "00:12:44" + } + } + ], + "question_timestamps": "00:12:46", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It fell, and it’s at your feet.", + "B": "Your part didn’t fall; it’s in front of you." + }, + "answer_label": "B", + "answer_text": "Your part didn’t fall; it’s in front of you.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_6_T1", + "correct_answer": "Your part didn’t fall; it’s in front of you.", + "original_options": { + "A": "It fell, and it’s at your feet.", + "B": "Your part didn’t fall; it’s in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_6_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Did my parts fall down?", + "answer": "It fell, and it’s at your feet.", + "question_chinese": "我的零件掉下去了吗？", + "answer_chinese": "掉下去了，在你的脚下", + "evidence": [ + { + "description": "Didn't fall down.", + "timestamps": { + "start": "00:12:30", + "end": "00:12:44" + } + }, + { + "description": "Fell down.", + "timestamps": { + "start": "00:24:28", + "end": "00:24:32" + } + } + ], + "question_timestamps": "00:24:34", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It fell, and it’s at your feet.", + "B": "Your part didn’t fall; it’s in front of you." + }, + "answer_label": "A", + "answer_text": "It fell, and it’s at your feet.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_6_T2", + "correct_answer": "It fell, and it’s at your feet.", + "original_options": { + "A": "It fell, and it’s at your feet.", + "B": "Your part didn’t fall; it’s in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_12", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Did my part fall down?", + "answer": "Fell down.", + "question_chinese": "我的零件掉下去了吗", + "answer_chinese": "掉下去了", + "evidence": [ + { + "description": "Didn't fall down.", + "timestamps": { + "start": "00:12:30", + "end": "00:18:44" + } + }, + { + "description": "Fell down.", + "timestamps": { + "start": "00:24:28", + "end": "00:24:36" + } + } + ], + "question_timestamps": "00:25:01", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Fell down.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Fell down.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_12", + "correct_answer": "Fell down.", + "original_options": { + "A": "Fell down.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_3_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different colors of assembly props did I see?", + "answer": "You saw assembly props in three colors—black, blue, and gray—and they were all in front of you.", + "question_chinese": "我看到组装道具几种颜色？", + "answer_chinese": "你看到组装道具三种颜色，黑色、蓝色、灰色，它们都在你前面", + "evidence": [ + { + "description": "Three kinds.", + "timestamps": { + "start": "00:02:34", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:02:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw assembly props in three colors—black, blue, and gray—and they were all in front of you.", + "B": "Two colors of assembly props.", + "C": "Four colors of assembly props.", + "D": "You saw assembly props in two colors—black and gray—and they were all in front of you.", + "E": "You saw assembly props in four colors—black, blue, gray, and white—and they were all in front of you." + }, + "answer_label": "A", + "answer_text": "You saw assembly props in three colors—black, blue, and gray—and they were all in front of you.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_3_T1", + "correct_answer": "You saw assembly props in three colors—black, blue, and gray—and they were all in front of you.", + "original_options": { + "A": "You saw assembly props in three colors—black, blue, and gray—and they were all in front of you.", + "B": "Two colors of assembly props.", + "C": "Four colors of assembly props.", + "D": "You saw assembly props in two colors—black and gray—and they were all in front of you.", + "E": "You saw assembly props in four colors—black, blue, gray, and white—and they were all in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 20 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_3_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different colors of assembly props did I see?", + "answer": "You saw assembly props in two colors—blue and gray—and they were in front of you.", + "question_chinese": "我看到组装道具几种颜色？", + "answer_chinese": "你看到组装道具两种颜色，蓝色和灰色，它们在你前面", + "evidence": [ + { + "description": "Three kinds.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:09:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One color of assembly props.", + "B": "You saw assembly props in two colors—blue and gray—and they were in front of you.", + "C": "Three colors of assembly props.", + "D": "You saw assembly props in one color—gray—and they were in front of you.", + "E": "You saw assembly props in three colors—blue, gray, and black—and they were in front of you." + }, + "answer_label": "B", + "answer_text": "You saw assembly props in two colors—blue and gray—and they were in front of you.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_3_T2", + "correct_answer": "You saw assembly props in two colors—blue and gray—and they were in front of you.", + "original_options": { + "A": "One color of assembly props.", + "B": "You saw assembly props in two colors—blue and gray—and they were in front of you.", + "C": "Three colors of assembly props.", + "D": "You saw assembly props in one color—gray—and they were in front of you.", + "E": "You saw assembly props in three colors—blue, gray, and black—and they were in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_7_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many times did I flip through the instruction manual?", + "answer": "You flipped through the instruction manual twice, and it was in front of you.", + "question_chinese": "我翻了多少次说明书？", + "answer_chinese": "你翻了两次说明书，它在你前面", + "evidence": [ + { + "description": "You flipped the manual twice.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You flipped through the instruction manual twice, and it was in front of you.", + "B": "You flipped through the instruction manual once, and it was in front of you." + }, + "answer_label": "A", + "answer_text": "You flipped through the instruction manual twice, and it was in front of you.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_7_T1", + "correct_answer": "You flipped through the instruction manual twice, and it was in front of you.", + "original_options": { + "A": "You flipped through the instruction manual twice, and it was in front of you.", + "B": "You flipped through the instruction manual once, and it was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_7_T2", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many times did I flip through the instruction manual?", + "answer": "You flipped through the instruction manual once, and it was in front of you.", + "question_chinese": "我翻了多少次说明书？", + "answer_chinese": "你翻了1次说明书，说明书在你的前面", + "evidence": [ + { + "description": "You flipped the manual once", + "timestamps": { + "start": "00:10:20", + "end": "00:10:26" + } + } + ], + "question_timestamps": "00:10:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You flipped through the instruction manual twice, and it was in front of you.", + "B": "You flipped through the instruction manual once, and it was in front of you." + }, + "answer_label": "B", + "answer_text": "You flipped through the instruction manual once, and it was in front of you.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_7_T2", + "correct_answer": "You flipped through the instruction manual once, and it was in front of you.", + "original_options": { + "A": "You flipped through the instruction manual twice, and it was in front of you.", + "B": "You flipped through the instruction manual once, and it was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_1", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the schoolbag on the ground?", + "answer": "You can.", + "question_chinese": "我能够到地上的书包吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:38", + "end": "00:01:00" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:32" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You cannot." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_1", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_2", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the schoolbag on the ground?", + "answer": "You can.", + "question_chinese": "我能够到地上的书包吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:38", + "end": "00:01:00" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:32" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:08:44", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:12:30", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "You cannot.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_2", + "correct_answer": "You can.", + "original_options": { + "A": "You cannot.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_3", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the blue bottle?", + "answer": "You can.", + "question_chinese": "我能够到蓝色的瓶子吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You cannot." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_3", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_4", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the blue bottle?", + "answer": "You can.", + "question_chinese": "我能够到蓝色的瓶子吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:18" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:11:22", + "end": "00:11:27" + } + } + ], + "question_timestamps": "00:12:30", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You cannot.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_4", + "correct_answer": "You can.", + "original_options": { + "A": "You cannot.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_6", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many colors are there for assembling props?", + "answer": "Three kinds.", + "question_chinese": "组装道具有几种颜色", + "answer_chinese": "三种", + "evidence": [ + { + "description": "Three kinds.", + "timestamps": { + "start": "00:02:34", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three kinds.", + "B": "Two kinds.", + "C": "Four kinds.", + "D": "Only two colors are available for assembling the props, so you choose between them.", + "E": "There are four different colors provided for assembling the props, giving you more options." + }, + "answer_label": "A", + "answer_text": "Three kinds.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_6", + "correct_answer": "Three kinds.", + "original_options": { + "A": "Three kinds.", + "B": "Two kinds.", + "C": "Four kinds.", + "D": "Only two colors are available for assembling the props, so you choose between them.", + "E": "There are four different colors provided for assembling the props, giving you more options." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_5", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many colors are there for assembling props?", + "answer": "Three kinds.", + "question_chinese": "组装道具有几种颜色", + "answer_chinese": "三种", + "evidence": [ + { + "description": "Three kinds.", + "timestamps": { + "start": "00:02:34", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:12:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One kind.", + "B": "Two kinds.", + "C": "There are four types of colors for assembling props, not just a smaller set.", + "D": "Three kinds.", + "E": "You have six color options for assembling the props, so the set is larger than three." + }, + "answer_label": "D", + "answer_text": "Three kinds.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_5", + "correct_answer": "Three kinds.", + "original_options": { + "A": "One kind.", + "B": "Two kinds.", + "C": "There are four types of colors for assembling props, not just a smaller set.", + "D": "Three kinds.", + "E": "You have six color options for assembling the props, so the set is larger than three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_2_T2", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where are the pliers relative to me?", + "answer": "The pliers are in front of you.", + "question_chinese": "钳子在我的哪里？", + "answer_chinese": "钳子在你的前面", + "evidence": [ + { + "description": "The pliers are in front of you.", + "timestamps": { + "start": "00:12:15", + "end": "00:12:18" + } + } + ], + "question_timestamps": "00:12:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The pliers are in front of you.", + "B": "The pliers are behind you." + }, + "answer_label": "A", + "answer_text": "The pliers are in front of you.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_2_T2", + "correct_answer": "The pliers are in front of you.", + "original_options": { + "A": "The pliers are in front of you.", + "B": "The pliers are behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_2_T1", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where are the pliers relative to me?", + "answer": "The pliers are to your right.", + "question_chinese": "钳子在我的哪里？", + "answer_chinese": "钳子在你的右边", + "evidence": [ + { + "description": "The pliers are on your right.", + "timestamps": { + "start": "00:12:18", + "end": "00:12:22" + } + } + ], + "question_timestamps": "00:12:24", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The pliers are in front of you.", + "B": "The pliers are to your right." + }, + "answer_label": "B", + "answer_text": "The pliers are to your right.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_2_T1", + "correct_answer": "The pliers are to your right.", + "original_options": { + "A": "The pliers are in front of you.", + "B": "The pliers are to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_9", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many icons are there on the yellow box?", + "answer": "Twelve icons.", + "question_chinese": "黄色盒子上有多少个图标", + "answer_chinese": "十二个图标", + "evidence": [ + { + "description": "Twelve icons.", + "timestamps": { + "start": "00:13:52", + "end": "00:14:30" + } + } + ], + "question_timestamps": "00:18:45", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Eleven icons.", + "B": "Fourteen icons.", + "C": "There are nine icons printed on the yellow box, fewer than the stated total.", + "D": "Twelve icons.", + "E": "You can count six icons on the yellow box, not the full set you expected." + }, + "answer_label": "D", + "answer_text": "Twelve icons.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_9", + "correct_answer": "Twelve icons.", + "original_options": { + "A": "Eleven icons.", + "B": "Fourteen icons.", + "C": "There are nine icons printed on the yellow box, fewer than the stated total.", + "D": "Twelve icons.", + "E": "You can count six icons on the yellow box, not the full set you expected." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "8e6753c7-1d9c-469d-b1a5-6e6424955494", + "video_length": "00:25:01", + "qid": "8e6753c7-1d9c-469d-b1a5-6e6424955494_10", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many icons are there on the yellow box?", + "answer": "Twelve icons.", + "question_chinese": "黄色盒子上有多少个图标", + "answer_chinese": "十二个图标", + "evidence": [ + { + "description": "Twelve icons.", + "timestamps": { + "start": "00:13:52", + "end": "00:14:30" + } + } + ], + "question_timestamps": "00:25:01", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Eleven icons.", + "B": "Thirteen icons.", + "C": "The yellow box shows seven icons in total when you count them one by one.", + "D": "You see fifteen icons on the yellow box, which is more than twelve.", + "E": "Twelve icons." + }, + "answer_label": "E", + "answer_text": "Twelve icons.", + "q_id": "8e6753c7-1d9c-469d-b1a5-6e6424955494_10", + "correct_answer": "Twelve icons.", + "original_options": { + "A": "Eleven icons.", + "B": "Thirteen icons.", + "C": "The yellow box shows seven icons in total when you count them one by one.", + "D": "You see fifteen icons on the yellow box, which is more than twelve.", + "E": "Twelve icons." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 13 + ] + } +] diff --git a/QAs/8fe2d076-aef6-4093-9292-cf8afdd715d1-mcq.json b/QAs/8fe2d076-aef6-4093-9292-cf8afdd715d1-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..34b27524e2aa739a8a6fad4808c442a42353f8ea --- /dev/null +++ b/QAs/8fe2d076-aef6-4093-9292-cf8afdd715d1-mcq.json @@ -0,0 +1,1273 @@ +[ + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_11", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tools have you placed together?", + "answer": "Three.", + "question_chinese": "你将多少个工具摆放在一起了", + "answer_chinese": "三个", + "evidence": [ + { + "description": "Three,", + "timestamps": { + "start": "00:32:35", + "end": "00:32:42" + } + } + ], + "question_timestamps": "00:46:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "You have placed five tools together in the same spot, forming a larger group than three.", + "D": "Three.", + "E": "There are six tools that you grouped together, so the combined set is bigger than three." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_11", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "You have placed five tools together in the same spot, forming a larger group than three.", + "D": "Three.", + "E": "There are six tools that you grouped together, so the combined set is bigger than three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_12", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tools have you placed together?", + "answer": "Three.", + "question_chinese": "你将多少个工具摆放在一起了", + "answer_chinese": "三个", + "evidence": [ + { + "description": "Three,", + "timestamps": { + "start": "00:32:35", + "end": "00:32:42" + } + } + ], + "question_timestamps": "01:02:02", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You have placed five tools together, so the group contains more items than three.", + "E": "You ended up gathering six tools into one place, rather than just three." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_12", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You have placed five tools together, so the group contains more items than three.", + "E": "You ended up gathering six tools into one place, rather than just three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_5_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At where I am right now, how many tools are currently placed together?", + "answer": "Two, on the lawn at your feet in front of you.", + "question_chinese": "在你所在的位置，目前多少个工具摆放在一起了？", + "answer_chinese": "两个，在你前方脚下的草坪上", + "evidence": [ + { + "description": "You put the shovel and the rake together on the lawn at your feet in front of you.", + "timestamps": { + "start": "00:32:35", + "end": "00:32:37" + } + } + ], + "question_timestamps": "00:32:38", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, on the lawn at your feet in front of you.", + "B": "One, on the lawn at your feet.", + "C": "Three, on the lawn at your feet.", + "D": "Four, grouped together on the lawn directly in front of your feet where you are standing.", + "E": "Zero tools are placed together on the lawn at your feet in front of you right now." + }, + "answer_label": "A", + "answer_text": "Two, on the lawn at your feet in front of you.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_5_T1", + "correct_answer": "Two, on the lawn at your feet in front of you.", + "original_options": { + "A": "Two, on the lawn at your feet in front of you.", + "B": "One, on the lawn at your feet.", + "C": "Three, on the lawn at your feet.", + "D": "Four, grouped together on the lawn directly in front of your feet where you are standing.", + "E": "Zero tools are placed together on the lawn at your feet in front of you right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_5_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At where I am right now, how many tools are currently placed together?", + "answer": "Three, on the lawn at your feet behind you.", + "question_chinese": "在你所在的位置，目前多少个工具摆放在一起了？", + "answer_chinese": "三个，在你身后脚下的草坪上", + "evidence": [ + { + "description": "You put the shovel and the rake together on the lawn at your feet in front of you.", + "timestamps": { + "start": "00:32:35", + "end": "00:32:42" + } + }, + { + "description": "The person to your right put down another rake on the lawn at your feet in front of you.", + "timestamps": { + "start": "00:32:39", + "end": "00:32:41" + } + }, + { + "description": "You turned around once.", + "timestamps": { + "start": "00:32:41", + "end": "00:32:42" + } + } + ], + "question_timestamps": "00:32:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, on the lawn behind you.", + "B": "Three, on the lawn at your feet behind you.", + "C": "Four, on the lawn behind you.", + "D": "Five, placed together on the lawn at your feet behind you where you are standing now.", + "E": "One tool is placed together on the lawn at your feet behind you right at this moment." + }, + "answer_label": "B", + "answer_text": "Three, on the lawn at your feet behind you.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_5_T2", + "correct_answer": "Three, on the lawn at your feet behind you.", + "original_options": { + "A": "Two, on the lawn behind you.", + "B": "Three, on the lawn at your feet behind you.", + "C": "Four, on the lawn behind you.", + "D": "Five, placed together on the lawn at your feet behind you where you are standing now.", + "E": "One tool is placed together on the lawn at your feet behind you right at this moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many shovels of soil did I shovel into the small cart in total?", + "answer": "You shoveled fifteen times.", + "question_chinese": "我一共向小推车里铲了多少铲土", + "answer_chinese": "你铲了十五铲", + "evidence": [ + { + "description": "You shoveled fifteen times.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:15:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You shoveled ten times.", + "B": "You shoveled twelve times.", + "C": "You shoveled five times in total when filling the small cart with soil.", + "D": "You shoveled fifteen times.", + "E": "You shoveled twenty times overall, repeatedly scooping soil and dumping it into the small cart." + }, + "answer_label": "D", + "answer_text": "You shoveled fifteen times.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_1", + "correct_answer": "You shoveled fifteen times.", + "original_options": { + "A": "You shoveled ten times.", + "B": "You shoveled twelve times.", + "C": "You shoveled five times in total when filling the small cart with soil.", + "D": "You shoveled fifteen times.", + "E": "You shoveled twenty times overall, repeatedly scooping soil and dumping it into the small cart." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 15 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many shovels of soil did I shovel into the small cart in total?", + "answer": "You shoveled fifteen shovels of soil in total.", + "question_chinese": "我一共向小推车里铲了多少铲土", + "answer_chinese": "你一共铲了十五铲土", + "evidence": [ + { + "description": "You shoveled fifteen times.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:31:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You shoveled eight shovels total.", + "B": "You shoveled fifteen shovels of soil in total.", + "C": "You shoveled twelve shovels total.", + "D": "You shoveled ten shovels of soil in total into the small cart.", + "E": "You shoveled twenty shovels of soil in total while filling the small cart completely." + }, + "answer_label": "B", + "answer_text": "You shoveled fifteen shovels of soil in total.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_2", + "correct_answer": "You shoveled fifteen shovels of soil in total.", + "original_options": { + "A": "You shoveled eight shovels total.", + "B": "You shoveled fifteen shovels of soil in total.", + "C": "You shoveled twelve shovels total.", + "D": "You shoveled ten shovels of soil in total into the small cart.", + "E": "You shoveled twenty shovels of soil in total while filling the small cart completely." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 14 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_19", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do when I got into the trunk of the car?", + "answer": "You put the sapling down together with others.", + "question_chinese": "我上到车的后备箱做了什么", + "answer_chinese": "你跟别人一起把树苗放下去了", + "evidence": [ + { + "description": "You put the sapling down together with others.", + "timestamps": { + "start": "00:41:14", + "end": "00:44:15" + } + } + ], + "question_timestamps": "00:46:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You picked the sapling back up.", + "B": "You left the sapling standing alone.", + "C": "You put the sapling down together with others.", + "D": "You set the sapling down near the garden tools instead of placing it together with the others.", + "E": "You arranged the sapling upright in the trunk by itself rather than grouping it with the other saplings." + }, + "answer_label": "C", + "answer_text": "You put the sapling down together with others.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_19", + "correct_answer": "You put the sapling down together with others.", + "original_options": { + "A": "You picked the sapling back up.", + "B": "You left the sapling standing alone.", + "C": "You put the sapling down together with others.", + "D": "You set the sapling down near the garden tools instead of placing it together with the others.", + "E": "You arranged the sapling upright in the trunk by itself rather than grouping it with the other saplings." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_20", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do when I got into the trunk of the car?", + "answer": "You put the sapling down together with others.", + "question_chinese": "我上到车的后备箱做了什么", + "answer_chinese": "你和别人一起把树苗放下去了", + "evidence": [ + { + "description": "You put the sapling down together with others.", + "timestamps": { + "start": "00:41:14", + "end": "00:44:15" + } + } + ], + "question_timestamps": "01:02:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You picked the sapling back up.", + "B": "You moved the sapling to the side.", + "C": "You set the sapling down separately in the trunk rather than placing it together with the others.", + "D": "You adjusted items in the trunk but kept the sapling apart instead of putting it down with the other saplings.", + "E": "You put the sapling down together with others." + }, + "answer_label": "E", + "answer_text": "You put the sapling down together with others.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_20", + "correct_answer": "You put the sapling down together with others.", + "original_options": { + "A": "You picked the sapling back up.", + "B": "You moved the sapling to the side.", + "C": "You set the sapling down separately in the trunk rather than placing it together with the others.", + "D": "You adjusted items in the trunk but kept the sapling apart instead of putting it down with the other saplings.", + "E": "You put the sapling down together with others." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 20 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_5", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did the light intensity change after I poured out the soil?", + "answer": "The light intensity increased and then returned to its original state.", + "question_chinese": "我倒完土后光照强度变了吗", + "answer_chinese": "光照强度变强了然后恢复原样", + "evidence": [ + { + "description": "The light intensity has increased.", + "timestamps": { + "start": "00:02:25", + "end": "00:03:10" + } + }, + { + "description": "Light intensity restored to normal.", + "timestamps": { + "start": "00:03:13", + "end": "00:03:26" + } + } + ], + "question_timestamps": "00:15:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The light intensity stayed the same.", + "B": "The light intensity decreased first.", + "C": "The light intensity decreased and then returned to its original state after you poured out the soil.", + "D": "The light intensity increased and stayed brighter afterward, without returning to its original level.", + "E": "The light intensity increased and then returned to its original state." + }, + "answer_label": "E", + "answer_text": "The light intensity increased and then returned to its original state.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_5", + "correct_answer": "The light intensity increased and then returned to its original state.", + "original_options": { + "A": "The light intensity stayed the same.", + "B": "The light intensity decreased first.", + "C": "The light intensity decreased and then returned to its original state after you poured out the soil.", + "D": "The light intensity increased and stayed brighter afterward, without returning to its original level.", + "E": "The light intensity increased and then returned to its original state." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 14 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_6", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did the light intensity change after I poured out the soil?", + "answer": "The light intensity remains strong continuously.", + "question_chinese": "我倒完土后光照强度变了吗", + "answer_chinese": "光强度持续保持强烈", + "evidence": [ + { + "description": "The light intensity has increased.", + "timestamps": { + "start": "00:02:25", + "end": "00:03:10" + } + }, + { + "description": "The light intensity returns to its original state.", + "timestamps": { + "start": "00:03:13", + "end": "00:03:26" + } + }, + { + "description": "The light intensity remains strong continuously.", + "timestamps": { + "start": "00:21:45", + "end": "00:23:30" + } + } + ], + "question_timestamps": "00:31:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The light intensity dims a little.", + "B": "The light intensity increases a lot.", + "C": "The light intensity changes erratically, fluctuating between brighter and dimmer instead of staying steady.", + "D": "The light intensity brightens slowly over time rather than remaining strong continuously throughout.", + "E": "The light intensity remains strong continuously." + }, + "answer_label": "E", + "answer_text": "The light intensity remains strong continuously.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_6", + "correct_answer": "The light intensity remains strong continuously.", + "original_options": { + "A": "The light intensity dims a little.", + "B": "The light intensity increases a lot.", + "C": "The light intensity changes erratically, fluctuating between brighter and dimmer instead of staying steady.", + "D": "The light intensity brightens slowly over time rather than remaining strong continuously throughout.", + "E": "The light intensity remains strong continuously." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_6_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I take a puff from the e-cigarette?", + "answer": "Twice: on the left side of the car and at the back-right side of the car.", + "question_chinese": "我吸了几次电子烟？", + "answer_chinese": "两次，在车的左侧和右后侧", + "evidence": [ + { + "description": "You vaped on the left side of the car.", + "timestamps": { + "start": "00:34:00", + "end": "00:35:25" + } + }, + { + "description": "You vaped at the back-right side of the car.", + "timestamps": { + "start": "00:39:35", + "end": "00:39:55" + } + } + ], + "question_timestamps": "00:46:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once, on the left side.", + "B": "Twice: on the left side of the car and at the back-right side of the car.", + "C": "Three times total.", + "D": "Once: you only took a puff on the left side of the car, then stopped.", + "E": "Three times: on the left side of the car, at the back-right side, and at the front-left side." + }, + "answer_label": "B", + "answer_text": "Twice: on the left side of the car and at the back-right side of the car.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_6_T1", + "correct_answer": "Twice: on the left side of the car and at the back-right side of the car.", + "original_options": { + "A": "Once, on the left side.", + "B": "Twice: on the left side of the car and at the back-right side of the car.", + "C": "Three times total.", + "D": "Once: you only took a puff on the left side of the car, then stopped.", + "E": "Three times: on the left side of the car, at the back-right side, and at the front-left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 15, + 18 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_6_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I take a puff from the e-cigarette?", + "answer": "Three times: on the left side of the car, at the back-right side of the car, and at the back-left side of the car.", + "question_chinese": "我吸了几次电子烟？", + "answer_chinese": "三次，在车的左侧、右后侧和左后侧", + "evidence": [ + { + "description": "You vaped on the left side of the car.", + "timestamps": { + "start": "00:34:00", + "end": "00:35:25" + } + }, + { + "description": "You vaped at the back-right side of the car.", + "timestamps": { + "start": "00:39:35", + "end": "00:39:55" + } + }, + { + "description": "You vaped once more at the back-left side of the car.", + "timestamps": { + "start": "00:47:15", + "end": "00:47:27" + } + } + ], + "question_timestamps": "01:02:02", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two times total.", + "B": "Four times total.", + "C": "Three times: on the left side of the car, at the back-right side of the car, and at the back-left side of the car.", + "D": "One time: you only took a puff at the back-right side of the car.", + "E": "Five times: on the left side, back-right, back-left, front-left, and front-right sides of the car." + }, + "answer_label": "C", + "answer_text": "Three times: on the left side of the car, at the back-right side of the car, and at the back-left side of the car.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_6_T2", + "correct_answer": "Three times: on the left side of the car, at the back-right side of the car, and at the back-left side of the car.", + "original_options": { + "A": "Two times total.", + "B": "Four times total.", + "C": "Three times: on the left side of the car, at the back-right side of the car, and at the back-left side of the car.", + "D": "One time: you only took a puff at the back-right side of the car.", + "E": "Five times: on the left side, back-right, back-left, front-left, and front-right sides of the car." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 15 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_7", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I put in the place where I dump the soil?", + "answer": "You placed some potted plants.", + "question_chinese": "我在倒土的地方放了什么", + "answer_chinese": "你放了一些盆栽", + "evidence": [ + { + "description": "You placed some potted plants.", + "timestamps": { + "start": "00:04:05", + "end": "00:08:33" + } + } + ], + "question_timestamps": "00:15:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Some decorative stones.", + "B": "A few garden statues.", + "C": "You put a bag of potting soil there instead of placing any plants.", + "D": "You placed some potted plants.", + "E": "You set down a couple of gardening tools there, not potted plants." + }, + "answer_label": "D", + "answer_text": "You placed some potted plants.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_7", + "correct_answer": "You placed some potted plants.", + "original_options": { + "A": "Some decorative stones.", + "B": "A few garden statues.", + "C": "You put a bag of potting soil there instead of placing any plants.", + "D": "You placed some potted plants.", + "E": "You set down a couple of gardening tools there, not potted plants." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 12 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_8", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I put in the place where I dump the soil?", + "answer": "You placed some potted plants.", + "question_chinese": "我在倒土的地方放了什么", + "answer_chinese": "你放置了一些盆栽", + "evidence": [ + { + "description": "You placed some potted plants.", + "timestamps": { + "start": "00:04:05", + "end": "00:08:33" + } + } + ], + "question_timestamps": "00:31:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A garden rake.", + "B": "A bag of fertilizer.", + "C": "You placed some potted plants.", + "D": "You placed some ceramic vases there instead of putting down any potted plants.", + "E": "You put some vegetable seeds in that spot, rather than setting any plants down." + }, + "answer_label": "C", + "answer_text": "You placed some potted plants.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_8", + "correct_answer": "You placed some potted plants.", + "original_options": { + "A": "A garden rake.", + "B": "A bag of fertilizer.", + "C": "You placed some potted plants.", + "D": "You placed some ceramic vases there instead of putting down any potted plants.", + "E": "You put some vegetable seeds in that spot, rather than setting any plants down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 14 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_3_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At the place where I was dumping soil, how many new potted plants did I set down?", + "answer": "Two, at your feet to your front-left.", + "question_chinese": "我在倒土的地方新放了几个盆栽？", + "answer_chinese": "2个，在你的左前方脚下", + "evidence": [ + { + "description": "You just placed two potted plants at your feet to your front-left.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:04:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One, front-left.", + "B": "Three, front-left.", + "C": "Zero: you didn’t set any new potted plants down at your feet to your front-left.", + "D": "Two, at your feet to your front-left.", + "E": "Four, at your feet to your front-left, in a small cluster near the dumping spot." + }, + "answer_label": "D", + "answer_text": "Two, at your feet to your front-left.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_3_T1", + "correct_answer": "Two, at your feet to your front-left.", + "original_options": { + "A": "One, front-left.", + "B": "Three, front-left.", + "C": "Zero: you didn’t set any new potted plants down at your feet to your front-left.", + "D": "Two, at your feet to your front-left.", + "E": "Four, at your feet to your front-left, in a small cluster near the dumping spot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 15 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_3_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At the place where I was dumping soil, how many new potted plants did I set down?", + "answer": "Five, at your feet to your front-left.", + "question_chinese": "我在倒土的地方新放了几个盆栽？", + "answer_chinese": "5个，在你的左前方脚下", + "evidence": [ + { + "description": "You just placed two potted plants at your feet to your front-left.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:12" + } + }, + { + "description": "You just placed one potted plant at your feet to your front-left.", + "timestamps": { + "start": "00:04:17", + "end": "00:04:20" + } + }, + { + "description": "You just placed two potted plants at your feet to your front-left.", + "timestamps": { + "start": "00:04:27", + "end": "00:04:29" + } + } + ], + "question_timestamps": "00:04:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four, front-left.", + "B": "Six, front-left.", + "C": "Three, at your feet to your front-left, set down one after another in the same spot.", + "D": "Seven, at your feet to your front-left, with several grouped close together by the soil pile.", + "E": "Five, at your feet to your front-left." + }, + "answer_label": "E", + "answer_text": "Five, at your feet to your front-left.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_3_T2", + "correct_answer": "Five, at your feet to your front-left.", + "original_options": { + "A": "Four, front-left.", + "B": "Six, front-left.", + "C": "Three, at your feet to your front-left, set down one after another in the same spot.", + "D": "Seven, at your feet to your front-left, with several grouped close together by the soil pile.", + "E": "Five, at your feet to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_8_T1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do at the car's trunk?", + "answer": "You and someone else put the saplings onto the lawn behind the trunk.", + "question_chinese": "我在车的后备箱的位置做了什么？", + "answer_chinese": "你跟别人一起把树苗放到了后备箱后面的草坪上", + "evidence": [ + { + "description": "You and someone else put the saplings onto the lawn behind you, behind the trunk.", + "timestamps": { + "start": "00:41:14", + "end": "00:44:15" + } + } + ], + "question_timestamps": "00:44:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You closed the trunk lid.", + "B": "You smoked at the trunk.", + "C": "You and someone else put the saplings onto the lawn in front of the trunk.", + "D": "You and someone else put the saplings onto the lawn behind the trunk.", + "E": "You and someone else picked the saplings up from the lawn and placed them back into the trunk." + }, + "answer_label": "D", + "answer_text": "You and someone else put the saplings onto the lawn behind the trunk.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_8_T1", + "correct_answer": "You and someone else put the saplings onto the lawn behind the trunk.", + "original_options": { + "A": "You closed the trunk lid.", + "B": "You smoked at the trunk.", + "C": "You and someone else put the saplings onto the lawn in front of the trunk.", + "D": "You and someone else put the saplings onto the lawn behind the trunk.", + "E": "You and someone else picked the saplings up from the lawn and placed them back into the trunk." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_8_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do at the car's trunk?", + "answer": "You and someone else put the saplings on the lawn behind you, behind the trunk, closed the trunk lid, and then walked toward the driver’s seat at the front of the car.", + "question_chinese": "我在车的后备箱的位置做了什么？", + "answer_chinese": "你跟别人一起把树苗放到了后备箱后面的草坪上，把车后备箱门关上了，然后往车前的驾驶位走去", + "evidence": [ + { + "description": "You and someone else put the saplings onto the lawn behind you, behind the trunk.", + "timestamps": { + "start": "00:41:14", + "end": "00:44:15" + } + }, + { + "description": "You and someone else closed the trunk lid, then walked toward the driver’s seat at the front of the car.", + "timestamps": { + "start": "00:44:57", + "end": "00:45:31" + } + } + ], + "question_timestamps": "01:02:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You left the trunk open.", + "B": "You walked to the passenger seat.", + "C": "You and someone else put the saplings behind the trunk, closed it, then walked to the passenger seat.", + "D": "You and someone else put the saplings behind the trunk, left it open, then walked toward the driver’s seat.", + "E": "You and someone else put the saplings on the lawn behind you, behind the trunk, closed the trunk lid, and then walked toward the driver’s seat at the front of the car." + }, + "answer_label": "E", + "answer_text": "You and someone else put the saplings on the lawn behind you, behind the trunk, closed the trunk lid, and then walked toward the driver’s seat at the front of the car.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_8_T2", + "correct_answer": "You and someone else put the saplings on the lawn behind you, behind the trunk, closed the trunk lid, and then walked toward the driver’s seat at the front of the car.", + "original_options": { + "A": "You left the trunk open.", + "B": "You walked to the passenger seat.", + "C": "You and someone else put the saplings behind the trunk, closed it, then walked to the passenger seat.", + "D": "You and someone else put the saplings behind the trunk, left it open, then walked toward the driver’s seat.", + "E": "You and someone else put the saplings on the lawn behind you, behind the trunk, closed the trunk lid, and then walked toward the driver’s seat at the front of the car." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 20 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_13", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I relax myself?", + "answer": "You smoke e-cigarettes.", + "question_chinese": "我如何放松自己", + "answer_chinese": "你吸电子烟", + "evidence": [ + { + "description": "You smoke e-cigarettes.", + "timestamps": { + "start": "00:34:00", + "end": "00:35:25" + } + }, + { + "description": "You smoke e-cigarettes.", + "timestamps": { + "start": "00:39:35", + "end": "00:39:55" + } + } + ], + "question_timestamps": "00:46:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You drink an energy drink.", + "B": "You read a novel quietly.", + "C": "You smoke e-cigarettes.", + "D": "You relax by playing video games for a while instead of smoking anything.", + "E": "You relax by going for a short jog outside and cooling down afterward." + }, + "answer_label": "C", + "answer_text": "You smoke e-cigarettes.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_13", + "correct_answer": "You smoke e-cigarettes.", + "original_options": { + "A": "You drink an energy drink.", + "B": "You read a novel quietly.", + "C": "You smoke e-cigarettes.", + "D": "You relax by playing video games for a while instead of smoking anything.", + "E": "You relax by going for a short jog outside and cooling down afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_14", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I relax myself?", + "answer": "You smoke e-cigarettes.", + "question_chinese": "我如何放松自己", + "answer_chinese": "你吸电子烟", + "evidence": [ + { + "description": "You smoke e-cigarettes.", + "timestamps": { + "start": "00:34:00", + "end": "00:35:25" + } + }, + { + "description": "You smoke e-cigarettes.", + "timestamps": { + "start": "00:39:35", + "end": "00:39:55" + } + }, + { + "description": "You smoke e-cigarettes.", + "timestamps": { + "start": "00:47:15", + "end": "00:47:27" + } + } + ], + "question_timestamps": "01:02:02", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You go for a run.", + "B": "You play video games.", + "C": "You relax by reading a mystery novel quietly rather than using any nicotine.", + "D": "You relax by drinking an energy drink and scrolling online for a bit.", + "E": "You smoke e-cigarettes." + }, + "answer_label": "E", + "answer_text": "You smoke e-cigarettes.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_14", + "correct_answer": "You smoke e-cigarettes.", + "original_options": { + "A": "You go for a run.", + "B": "You play video games.", + "C": "You relax by reading a mystery novel quietly rather than using any nicotine.", + "D": "You relax by drinking an energy drink and scrolling online for a bit.", + "E": "You smoke e-cigarettes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 13 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_9", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I take out my phone for?", + "answer": "You took out your phone and checked the time.", + "question_chinese": "我掏出手机做什么", + "answer_chinese": "你掏出手机看了一眼时间", + "evidence": [ + { + "description": "You took out your phone and checked the time.", + "timestamps": { + "start": "00:10:15", + "end": "00:10:21" + } + } + ], + "question_timestamps": "00:15:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took out your phone and checked the time.", + "B": "To check your battery.", + "C": "To open the calendar.", + "D": "You took out your phone to look at the weather forecast before continuing on.", + "E": "You pulled out your phone to review your notes and then put it away." + }, + "answer_label": "A", + "answer_text": "You took out your phone and checked the time.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_9", + "correct_answer": "You took out your phone and checked the time.", + "original_options": { + "A": "You took out your phone and checked the time.", + "B": "To check your battery.", + "C": "To open the calendar.", + "D": "You took out your phone to look at the weather forecast before continuing on.", + "E": "You pulled out your phone to review your notes and then put it away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "8fe2d076-aef6-4093-9292-cf8afdd715d1", + "video_length": "01:02:02", + "qid": "8fe2d076-aef6-4093-9292-cf8afdd715d1_10", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I take out my phone for?", + "answer": "You didn't take out your phone again.", + "question_chinese": "我掏出手机做什么", + "answer_chinese": "你没有再掏出手机", + "evidence": [ + { + "description": "You took out your phone and checked the time.", + "timestamps": { + "start": "00:10:15", + "end": "00:10:21" + } + }, + { + "description": "You didn't take out your phone again.", + "timestamps": { + "start": "00:15:30", + "end": "00:30:00" + } + } + ], + "question_timestamps": "00:31:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To call someone.", + "B": "You didn't take out your phone again.", + "C": "To send a text.", + "D": "You took out your phone again to browse the internet for a moment.", + "E": "You pulled out your phone to take a quick picture and then put it away." + }, + "answer_label": "B", + "answer_text": "You didn't take out your phone again.", + "q_id": "8fe2d076-aef6-4093-9292-cf8afdd715d1_10", + "correct_answer": "You didn't take out your phone again.", + "original_options": { + "A": "To call someone.", + "B": "You didn't take out your phone again.", + "C": "To send a text.", + "D": "You took out your phone again to browse the internet for a moment.", + "E": "You pulled out your phone to take a quick picture and then put it away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 15 + ] + } +] diff --git a/QAs/9a7a189e-3494-40c0-84ee-5be47af574f7-mcq.json b/QAs/9a7a189e-3494-40c0-84ee-5be47af574f7-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1648cef936bae0ff604f2725cf4c97838eb1b422 --- /dev/null +++ b/QAs/9a7a189e-3494-40c0-84ee-5be47af574f7-mcq.json @@ -0,0 +1,1552 @@ +[ + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_7_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from entering the building entrance door to entering the room?", + "answer": "After you enter the building door, turn left and take the elevator up to the second floor. After you exit the elevator, turn left and go straight.", + "question_chinese": "我从进楼栋门到进入房间经过了什么路径？", + "answer_chinese": "你进楼栋门后，左转坐电梯上二楼，出电梯后左转直走。", + "evidence": [ + { + "description": "You entered the building and turned left to take the elevator to the second floor.", + "timestamps": { + "start": "00:22:58", + "end": "00:23:28" + } + } + ], + "question_timestamps": "00:23:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "After you enter the building door, turn left and take the elevator up to the second floor. After you exit the elevator, turn left and go straight.", + "B": "Turn right, then take the elevator.", + "C": "Take the stairs to the second floor.", + "D": "After entering, turn left, go straight past the trash can, then turn right to reach the room.", + "E": "After you enter the building, turn left to the elevator, then exit and turn right before going straight." + }, + "answer_label": "A", + "answer_text": "After you enter the building door, turn left and take the elevator up to the second floor. After you exit the elevator, turn left and go straight.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_7_T1", + "correct_answer": "After you enter the building door, turn left and take the elevator up to the second floor. After you exit the elevator, turn left and go straight.", + "original_options": { + "A": "After you enter the building door, turn left and take the elevator up to the second floor. After you exit the elevator, turn left and go straight.", + "B": "Turn right, then take the elevator.", + "C": "Take the stairs to the second floor.", + "D": "After entering, turn left, go straight past the trash can, then turn right to reach the room.", + "E": "After you enter the building, turn left to the elevator, then exit and turn right before going straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 18 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_10", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from entering the building door to entering the room?", + "answer": "You entered the building and turned left to take the elevator to the second floor. After you got off, you turned left and walked straight. At the fork, you turned left again and continued straight until you arrived at Room 201.", + "question_chinese": "我从进楼栋门到进入房间经过了什么路径？", + "answer_chinese": "你进楼栋门后，左转坐电梯上二楼，出电梯后左转直走，在分岔口左转，直走,直到你到达201。", + "evidence": [ + { + "description": "You entered the building and turned left to take the elevator to the second floor.", + "timestamps": { + "start": "00:22:58", + "end": "00:23:29" + } + }, + { + "description": "After you got off, you turned left and walked straight.", + "timestamps": { + "start": "00:23:30", + "end": "00:23:35" + } + }, + { + "description": "At the fork, you turned left again and continued straight until you arrived at Room 201.", + "timestamps": { + "start": "00:23:30", + "end": "00:24:12" + } + } + ], + "question_timestamps": "00:24:16", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You entered the building and turned left to take the elevator to the second floor. After you got off, you turned left and walked straight. At the fork, you turned left again and continued straight until you arrived at Room 201.", + "B": "Turn right to the elevator.", + "C": "Take the stairs upstairs.", + "D": "You entered, turned left to the elevator, then exited and turned right, continuing straight to Room 201.", + "E": "You entered the building, turned left, took the elevator up, then skipped the fork and went straight to the room." + }, + "answer_label": "A", + "answer_text": "You entered the building and turned left to take the elevator to the second floor. After you got off, you turned left and walked straight. At the fork, you turned left again and continued straight until you arrived at Room 201.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_10", + "correct_answer": "You entered the building and turned left to take the elevator to the second floor. After you got off, you turned left and walked straight. At the fork, you turned left again and continued straight until you arrived at Room 201.", + "original_options": { + "A": "You entered the building and turned left to take the elevator to the second floor. After you got off, you turned left and walked straight. At the fork, you turned left again and continued straight until you arrived at Room 201.", + "B": "Turn right to the elevator.", + "C": "Take the stairs upstairs.", + "D": "You entered, turned left to the elevator, then exited and turned right, continuing straight to Room 201.", + "E": "You entered the building, turned left, took the elevator up, then skipped the fork and went straight to the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 20 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_7_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from entering the building entrance door to entering the room?", + "answer": "After you enter the building door, turn left and take the elevator up to the second floor. After you exit the elevator, turn left and go straight. At the fork, turn left and keep going straight until you reach 201.", + "question_chinese": "我从进楼栋门到进入房间经过了什么路径？", + "answer_chinese": "你进楼栋门后，左转坐电梯上二楼，出电梯后左转直走，在分岔口左转，直走,直到你到达201。", + "evidence": [ + { + "description": "You entered the building and turned left to take the elevator to the second floor.", + "timestamps": { + "start": "00:22:58", + "end": "00:23:29" + } + }, + { + "description": "After you got off, you turned left and walked straight.", + "timestamps": { + "start": "00:23:30", + "end": "00:23:35" + } + }, + { + "description": "At the fork, you turned left again and continued straight until you arrived at Room 201.", + "timestamps": { + "start": "00:23:30", + "end": "00:24:12" + } + } + ], + "question_timestamps": "00:24:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right at the entrance.", + "B": "Turn right at the fork.", + "C": "After you enter the building door, turn left and take the elevator up to the second floor. After you exit the elevator, turn left and go straight. At the fork, turn left and keep going straight until you reach 201.", + "D": "After entering, turn left to the elevator, then exit and turn right before going straight to 201.", + "E": "After you exit the elevator, turn left and go straight, then take the right branch at the fork to reach 201." + }, + "answer_label": "C", + "answer_text": "After you enter the building door, turn left and take the elevator up to the second floor. After you exit the elevator, turn left and go straight. At the fork, turn left and keep going straight until you reach 201.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_7_T2", + "correct_answer": "After you enter the building door, turn left and take the elevator up to the second floor. After you exit the elevator, turn left and go straight. At the fork, turn left and keep going straight until you reach 201.", + "original_options": { + "A": "Turn right at the entrance.", + "B": "Turn right at the fork.", + "C": "After you enter the building door, turn left and take the elevator up to the second floor. After you exit the elevator, turn left and go straight. At the fork, turn left and keep going straight until you reach 201.", + "D": "After entering, turn left to the elevator, then exit and turn right before going straight to 201.", + "E": "After you exit the elevator, turn left and go straight, then take the right branch at the fork to reach 201." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 21 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_4_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take from grabbing my things at my seat to leaving the restaurant?", + "answer": "First, walk forward to next to the trash can, turn left, and then go straight.", + "question_chinese": "我刚才从座位拿完东西到离开餐厅经历了什么路线？", + "answer_chinese": "你先向前走到垃圾桶旁边左转后直走。", + "evidence": [ + { + "description": "I walked forward to the trash can, turned left and walked straight.", + "timestamps": { + "start": "00:21:18", + "end": "00:22:24" + } + } + ], + "question_timestamps": "00:22:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First, walk forward to next to the trash can, turn left, and then go straight.", + "B": "Turn right at the trash can.", + "C": "Walk straight out without turning.", + "D": "You walked forward past the trash can, turned right, then went straight to leave the restaurant.", + "E": "You went forward to the trash can, turned left, then turned left again before leaving the restaurant." + }, + "answer_label": "A", + "answer_text": "First, walk forward to next to the trash can, turn left, and then go straight.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_4_T1", + "correct_answer": "First, walk forward to next to the trash can, turn left, and then go straight.", + "original_options": { + "A": "First, walk forward to next to the trash can, turn left, and then go straight.", + "B": "Turn right at the trash can.", + "C": "Walk straight out without turning.", + "D": "You walked forward past the trash can, turned right, then went straight to leave the restaurant.", + "E": "You went forward to the trash can, turned left, then turned left again before leaving the restaurant." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_7", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take from taking my seat to leaving the restaurant?", + "answer": "You walk forward to the trash can, turn left and walk straight, then leave through the door on your left.", + "question_chinese": "我刚才从座位拿完东西到离开餐厅经历了什么路线？", + "answer_chinese": "你向前走到垃圾桶旁边左转后直走，从左手边的门离开。", + "evidence": [ + { + "description": "I walked forward to the trash can, turned left and walked straight, then left through the door on my left.", + "timestamps": { + "start": "00:21:18", + "end": "00:22:26" + } + } + ], + "question_timestamps": "00:23:29", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You exit through the door in front.", + "B": "You exit through the door on your right.", + "C": "You walk forward to the trash can, turn left and walk straight, then leave through the door on your left.", + "D": "You walk forward to the trash can, turn right, go straight, and leave through the door on your left.", + "E": "You walk forward past the trash can, turn left, continue straight, then exit through the door ahead of you." + }, + "answer_label": "C", + "answer_text": "You walk forward to the trash can, turn left and walk straight, then leave through the door on your left.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_7", + "correct_answer": "You walk forward to the trash can, turn left and walk straight, then leave through the door on your left.", + "original_options": { + "A": "You exit through the door in front.", + "B": "You exit through the door on your right.", + "C": "You walk forward to the trash can, turn left and walk straight, then leave through the door on your left.", + "D": "You walk forward to the trash can, turn right, go straight, and leave through the door on your left.", + "E": "You walk forward past the trash can, turn left, continue straight, then exit through the door ahead of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 19 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_4_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take from grabbing my things at my seat to leaving the restaurant?", + "answer": "You walked forward until you were next to the trash can, turned left, then went straight and exited through the door on your left-hand side.", + "question_chinese": "我刚才从座位拿完东西到离开餐厅经历了什么路线？", + "answer_chinese": "你向前走到垃圾桶旁边左转后直走，从左手边的门离开。", + "evidence": [ + { + "description": "I walked forward to the trash can, turned left and walked straight, then left through the door on my left.", + "timestamps": { + "start": "00:21:18", + "end": "00:22:26" + } + } + ], + "question_timestamps": "00:24:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You exit through the door on your right.", + "B": "You exit through the door straight ahead.", + "C": "You walked forward until you were next to the trash can, turned left, then went straight and exited through the door on your left-hand side.", + "D": "You walked forward next to the trash can, turned right, went straight, and exited through the door on your left.", + "E": "You walked forward until past the trash can, turned left, went straight, and exited through the door in front of you." + }, + "answer_label": "C", + "answer_text": "You walked forward until you were next to the trash can, turned left, then went straight and exited through the door on your left-hand side.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_4_T2", + "correct_answer": "You walked forward until you were next to the trash can, turned left, then went straight and exited through the door on your left-hand side.", + "original_options": { + "A": "You exit through the door on your right.", + "B": "You exit through the door straight ahead.", + "C": "You walked forward until you were next to the trash can, turned left, then went straight and exited through the door on your left-hand side.", + "D": "You walked forward next to the trash can, turned right, went straight, and exited through the door on your left.", + "E": "You walked forward until past the trash can, turned left, went straight, and exited through the door in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 20, + 21 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_8", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many khaki chairs did I see on the second floor?", + "answer": "You saw a total of four khaki chairs on the second floor.", + "question_chinese": "我在二楼一共看见了几把卡其色的椅子？", + "answer_chinese": "你在二楼一共看见了四把卡其色的椅子。", + "evidence": [ + { + "description": "I saw two khaki chairs.", + "timestamps": { + "start": "00:23:27", + "end": "00:23:28" + } + }, + { + "description": "I saw another two khaki chairs.", + "timestamps": { + "start": "00:23:29", + "end": "00:23:36" + } + } + ], + "question_timestamps": "00:23:39", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You saw a total of four khaki chairs on the second floor.", + "B": "You saw three khaki chairs.", + "C": "You saw five khaki chairs.", + "D": "You saw a total of two khaki chairs on the second floor, not four.", + "E": "You saw a total of six khaki chairs on the second floor during your time upstairs." + }, + "answer_label": "A", + "answer_text": "You saw a total of four khaki chairs on the second floor.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_8", + "correct_answer": "You saw a total of four khaki chairs on the second floor.", + "original_options": { + "A": "You saw a total of four khaki chairs on the second floor.", + "B": "You saw three khaki chairs.", + "C": "You saw five khaki chairs.", + "D": "You saw a total of two khaki chairs on the second floor, not four.", + "E": "You saw a total of six khaki chairs on the second floor during your time upstairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_14", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many khaki chairs did I see on the second floor?", + "answer": "You saw a total of four khaki chairs on the second floor.", + "question_chinese": "我在二楼一共看见了几把卡其色的椅子？", + "answer_chinese": "你在二楼一共看见了四把卡其色的椅子。", + "evidence": [ + { + "description": "I saw two khaki chairs.", + "timestamps": { + "start": "00:23:27", + "end": "00:23:28" + } + }, + { + "description": "I saw another two khaki chairs.", + "timestamps": { + "start": "00:23:29", + "end": "00:23:36" + } + } + ], + "question_timestamps": "00:24:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You saw five khaki chairs.", + "B": "You saw seven khaki chairs.", + "C": "You saw a total of four khaki chairs on the second floor.", + "D": "You saw a total of six khaki chairs on the second floor during your visit.", + "E": "You saw a total of eight khaki chairs on the second floor, more than four." + }, + "answer_label": "C", + "answer_text": "You saw a total of four khaki chairs on the second floor.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_14", + "correct_answer": "You saw a total of four khaki chairs on the second floor.", + "original_options": { + "A": "You saw five khaki chairs.", + "B": "You saw seven khaki chairs.", + "C": "You saw a total of four khaki chairs on the second floor.", + "D": "You saw a total of six khaki chairs on the second floor during your visit.", + "E": "You saw a total of eight khaki chairs on the second floor, more than four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_5_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many khaki-colored chairs did I see on the second floor?", + "answer": "On the second floor, you saw two khaki-colored chairs.", + "question_chinese": "我在二楼看见了几把卡其色的椅子？", + "answer_chinese": "你在二楼看见了两把卡其色的椅子。", + "evidence": [ + { + "description": "I saw two khaki chairs.", + "timestamps": { + "start": "00:23:25", + "end": "00:23:27" + } + } + ], + "question_timestamps": "00:23:29", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one khaki-colored chair.", + "B": "You saw three khaki-colored chairs.", + "C": "On the second floor, you saw zero khaki-colored chairs during the entire time upstairs.", + "D": "On the second floor, you saw four khaki-colored chairs rather than just two.", + "E": "On the second floor, you saw two khaki-colored chairs." + }, + "answer_label": "E", + "answer_text": "On the second floor, you saw two khaki-colored chairs.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_5_T1", + "correct_answer": "On the second floor, you saw two khaki-colored chairs.", + "original_options": { + "A": "You saw one khaki-colored chair.", + "B": "You saw three khaki-colored chairs.", + "C": "On the second floor, you saw zero khaki-colored chairs during the entire time upstairs.", + "D": "On the second floor, you saw four khaki-colored chairs rather than just two.", + "E": "On the second floor, you saw two khaki-colored chairs." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_5_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many khaki-colored chairs did I see on the second floor?", + "answer": "On the second floor, you saw four khaki-colored chairs.", + "question_chinese": "我在二楼看见了几把卡其色的椅子？", + "answer_chinese": "你在二楼看见了四把卡其色的椅子。", + "evidence": [ + { + "description": "I saw two khaki chairs.", + "timestamps": { + "start": "00:23:27", + "end": "00:23:28" + } + }, + { + "description": "I saw another two khaki chairs.", + "timestamps": { + "start": "00:23:29", + "end": "00:23:36" + } + } + ], + "question_timestamps": "00:24:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "On the second floor, you saw four khaki-colored chairs.", + "B": "You saw three khaki-colored chairs.", + "C": "You saw five khaki-colored chairs.", + "D": "On the second floor, you saw two khaki-colored chairs, not four in total.", + "E": "On the second floor, you saw six khaki-colored chairs spread around the seating area." + }, + "answer_label": "A", + "answer_text": "On the second floor, you saw four khaki-colored chairs.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_5_T2", + "correct_answer": "On the second floor, you saw four khaki-colored chairs.", + "original_options": { + "A": "On the second floor, you saw four khaki-colored chairs.", + "B": "You saw three khaki-colored chairs.", + "C": "You saw five khaki-colored chairs.", + "D": "On the second floor, you saw two khaki-colored chairs, not four in total.", + "E": "On the second floor, you saw six khaki-colored chairs spread around the seating area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_3_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I threw the foil tray from the table into the trash can, what path did it take?", + "answer": "You first put the foil container into a plastic bag and left your seat.", + "question_chinese": "我把锡纸盘从桌子上扔到垃圾桶，经过了什么路径？", + "answer_chinese": "你先把锡纸盒装进塑料袋里，离开座位。", + "evidence": [ + { + "description": "I put the tin foil box into a plastic bag, left my seat.", + "timestamps": { + "start": "00:20:15", + "end": "00:20:52" + } + } + ], + "question_timestamps": "00:20:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You tossed it straight into the trash.", + "B": "You stayed seated while throwing it.", + "C": "You carried the foil tray in your hand, walked a few steps forward, and dropped it into the trash can.", + "D": "You wrapped the foil container up, stayed seated, then leaned over and tossed it into the trash can.", + "E": "You first put the foil container into a plastic bag and left your seat." + }, + "answer_label": "E", + "answer_text": "You first put the foil container into a plastic bag and left your seat.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_3_T1", + "correct_answer": "You first put the foil container into a plastic bag and left your seat.", + "original_options": { + "A": "You tossed it straight into the trash.", + "B": "You stayed seated while throwing it.", + "C": "You carried the foil tray in your hand, walked a few steps forward, and dropped it into the trash can.", + "D": "You wrapped the foil container up, stayed seated, then leaned over and tossed it into the trash can.", + "E": "You first put the foil container into a plastic bag and left your seat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 18 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_6", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take when I threw the tin foil tray from the table into the trash can?", + "answer": "You put the tin foil box into a plastic bag, left my seat, picked it up, took 5 steps forward, and threw it into the trash can on my left.", + "question_chinese": "我把锡纸盘从桌子上扔到垃圾桶，经过了什么路径？", + "answer_chinese": "你把锡纸盒装进塑料袋里，离开座位，拿起来向前走5步，丢进左手边垃圾桶。", + "evidence": [ + { + "description": "I put the tin foil box into a plastic bag, left my seat, picked it up, took 5 steps forward, and threw it into the trash can on my left.", + "timestamps": { + "start": "00:20:15", + "end": "00:20:57" + } + } + ], + "question_timestamps": "00:22:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You threw it on your right.", + "B": "You took three steps forward.", + "C": "You put the tin foil box into a plastic bag, left my seat, picked it up, took 5 steps forward, and threw it into the trash can on my left.", + "D": "You put the tin foil box in a plastic bag, stayed at your seat, and tossed it into the nearby trash can.", + "E": "You put the tin foil box into a plastic bag, walked about four steps forward, and threw it into the trash ahead." + }, + "answer_label": "C", + "answer_text": "You put the tin foil box into a plastic bag, left my seat, picked it up, took 5 steps forward, and threw it into the trash can on my left.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_6", + "correct_answer": "You put the tin foil box into a plastic bag, left my seat, picked it up, took 5 steps forward, and threw it into the trash can on my left.", + "original_options": { + "A": "You threw it on your right.", + "B": "You took three steps forward.", + "C": "You put the tin foil box into a plastic bag, left my seat, picked it up, took 5 steps forward, and threw it into the trash can on my left.", + "D": "You put the tin foil box in a plastic bag, stayed at your seat, and tossed it into the nearby trash can.", + "E": "You put the tin foil box into a plastic bag, walked about four steps forward, and threw it into the trash ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 22, + 22 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_3_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I threw the foil tray from the table into the trash can, what path did it take?", + "answer": "You put the foil container into a plastic bag, left your seat, picked it up, walked forward 5 steps, and threw it into the trash can on your left-hand side.", + "question_chinese": "我把锡纸盘从桌子上扔到垃圾桶，经过了什么路径？", + "answer_chinese": "你把锡纸盒装进塑料袋里，离开座位，拿起来向前走5步，丢进左手边垃圾桶。", + "evidence": [ + { + "description": "I put the tin foil box into a plastic bag, left my seat, picked it up, took 5 steps forward, and threw it into the trash can on my left.", + "timestamps": { + "start": "00:20:15", + "end": "00:20:57" + } + } + ], + "question_timestamps": "00:23:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the bin in front.", + "B": "You put the foil container into a plastic bag, left your seat, picked it up, walked forward 5 steps, and threw it into the trash can on your left-hand side.", + "C": "Into the trash on your right.", + "D": "You bagged it, walked three steps forward, then tossed it into the trash on your left.", + "E": "You picked up the tray, walked five steps forward, and threw it into the trash can in front." + }, + "answer_label": "B", + "answer_text": "You put the foil container into a plastic bag, left your seat, picked it up, walked forward 5 steps, and threw it into the trash can on your left-hand side.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_3_T2", + "correct_answer": "You put the foil container into a plastic bag, left your seat, picked it up, walked forward 5 steps, and threw it into the trash can on your left-hand side.", + "original_options": { + "A": "Into the bin in front.", + "B": "You put the foil container into a plastic bag, left your seat, picked it up, walked forward 5 steps, and threw it into the trash can on your left-hand side.", + "C": "Into the trash on your right.", + "D": "You bagged it, walked three steps forward, then tossed it into the trash on your left.", + "E": "You picked up the tray, walked five steps forward, and threw it into the trash can in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 18 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_8_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I see the plastic knife take from inside the plastic bag to the top-left corner of the foil tray?", + "answer": "The plastic knife was taken from the plastic bag to your left onto the food, and then passed above you over the plastic bag.", + "question_chinese": "我看见塑料刀从塑料袋里面到锡纸盘左上角经过了什么路径？", + "answer_chinese": "塑料刀从左边塑料袋被拿到食物上又经过塑料袋上方。", + "evidence": [ + { + "description": "The plastic knife is carried from the left plastic bag onto the food.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:04" + } + }, + { + "description": "And then the plastic knife passed over the plastic bag.", + "timestamps": { + "start": "00:04:04", + "end": "00:04:09" + } + } + ], + "question_timestamps": "00:04:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From a bag in front.", + "B": "From the bag to your right.", + "C": "The knife was taken from a front bag onto the food, then moved across the tray.", + "D": "The knife was taken from the bag on your right onto the food, then passed above you.", + "E": "The plastic knife was taken from the plastic bag to your left onto the food, and then passed above you over the plastic bag." + }, + "answer_label": "E", + "answer_text": "The plastic knife was taken from the plastic bag to your left onto the food, and then passed above you over the plastic bag.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_8_T1", + "correct_answer": "The plastic knife was taken from the plastic bag to your left onto the food, and then passed above you over the plastic bag.", + "original_options": { + "A": "From a bag in front.", + "B": "From the bag to your right.", + "C": "The knife was taken from a front bag onto the food, then moved across the tray.", + "D": "The knife was taken from the bag on your right onto the food, then passed above you.", + "E": "The plastic knife was taken from the plastic bag to your left onto the food, and then passed above you over the plastic bag." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_8_T2", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I see the plastic knife take from inside the plastic bag to the top-left corner of the foil tray?", + "answer": "You saw the plastic knife get taken from the plastic bag to your left onto the food, pass above you over the plastic bag, and then get placed in the top-left corner of the foil tray.", + "question_chinese": "我看见塑料刀从塑料袋里面到锡纸盘左上角经过了什么路径？", + "answer_chinese": "你看到塑料刀从左边塑料袋被拿到食物上又经过塑料袋上方，被放到锡纸盘左上角。", + "evidence": [ + { + "description": "The plastic knife is carried from the left plastic bag onto the food.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:04" + } + }, + { + "description": "And then the plastic knife passed over the plastic bag.", + "timestamps": { + "start": "00:04:04", + "end": "00:04:09" + } + }, + { + "description": "The plastic knife was being placed on the top left corner of the tin foil tray.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:14" + } + } + ], + "question_timestamps": "00:10:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It ended up top-right.", + "B": "It went back in the bag.", + "C": "You saw the plastic knife get taken from the plastic bag to your left onto the food, pass above you over the plastic bag, and then get placed in the top-left corner of the foil tray.", + "D": "You saw the knife taken from a front bag onto the food, then placed in the top-left corner.", + "E": "You saw the knife taken from the left bag onto the food, pass above you, then land in the bottom-left corner." + }, + "answer_label": "C", + "answer_text": "You saw the plastic knife get taken from the plastic bag to your left onto the food, pass above you over the plastic bag, and then get placed in the top-left corner of the foil tray.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_8_T2", + "correct_answer": "You saw the plastic knife get taken from the plastic bag to your left onto the food, pass above you over the plastic bag, and then get placed in the top-left corner of the foil tray.", + "original_options": { + "A": "It ended up top-right.", + "B": "It went back in the bag.", + "C": "You saw the plastic knife get taken from the plastic bag to your left onto the food, pass above you over the plastic bag, and then get placed in the top-left corner of the foil tray.", + "D": "You saw the knife taken from a front bag onto the food, then placed in the top-left corner.", + "E": "You saw the knife taken from the left bag onto the food, pass above you, then land in the bottom-left corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 18, + 21 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_1_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plastic bags with stuff inside did I see on the table?", + "answer": "You saw 4 plastic bags with items inside on the table.", + "question_chinese": "我看见桌上有几个装着东西的塑料袋？", + "answer_chinese": "你看见桌上有4个装着东西的塑料袋。", + "evidence": [ + { + "description": "There are four plastic bags containing things on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:19" + } + } + ], + "question_timestamps": "00:01:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 4 plastic bags with items inside on the table.", + "B": "You saw 3 bags.", + "C": "You saw 5 bags.", + "D": "You saw 2 plastic bags with items inside on the table near the foil tray.", + "E": "You saw 6 plastic bags with items inside on the table during the meal." + }, + "answer_label": "A", + "answer_text": "You saw 4 plastic bags with items inside on the table.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_1_T1", + "correct_answer": "You saw 4 plastic bags with items inside on the table.", + "original_options": { + "A": "You saw 4 plastic bags with items inside on the table.", + "B": "You saw 3 bags.", + "C": "You saw 5 bags.", + "D": "You saw 2 plastic bags with items inside on the table near the foil tray.", + "E": "You saw 6 plastic bags with items inside on the table during the meal." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_1_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plastic bags with stuff inside did I see on the table?", + "answer": "You saw 4 plastic bags with items inside on the table.", + "question_chinese": "我看见桌上有几个装着东西的塑料袋？", + "answer_chinese": "你看见桌上有4个装着东西的塑料袋。", + "evidence": [ + { + "description": "There are four plastic bags containing things on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:19" + } + }, + { + "description": "I am taking the cola out of the bag and put the plastic bag into the blue plastic bag.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:23" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 3 bags.", + "B": "You saw 4 plastic bags with items inside on the table.", + "C": "You saw 5 bags.", + "D": "You saw 2 plastic bags with items inside on the table near the foil tray.", + "E": "You saw 6 plastic bags with items inside on the table during the meal." + }, + "answer_label": "B", + "answer_text": "You saw 4 plastic bags with items inside on the table.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_1_T2", + "correct_answer": "You saw 4 plastic bags with items inside on the table.", + "original_options": { + "A": "You saw 3 bags.", + "B": "You saw 4 plastic bags with items inside on the table.", + "C": "You saw 5 bags.", + "D": "You saw 2 plastic bags with items inside on the table near the foil tray.", + "E": "You saw 6 plastic bags with items inside on the table during the meal." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_1_T3", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plastic bags with stuff inside did I see on the table?", + "answer": "You saw 2 plastic bags with items inside on the table.", + "question_chinese": "我看见桌上有几个装着东西的塑料袋？", + "answer_chinese": "你看见桌上有2个装着东西的塑料袋。", + "evidence": [ + { + "description": "I put the white plastic bag into the blue plastic bag.", + "timestamps": { + "start": "00:21:10", + "end": "00:21:14" + } + }, + { + "description": "I took the cola out of the bag and put the plastic bag into the blue plastic bag.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:23" + } + }, + { + "description": "There are four plastic bags on the table.", + "timestamps": { + "start": "00:00:08", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:21:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 1 bag.", + "B": "You saw 3 bags.", + "C": "You saw 2 plastic bags with items inside on the table.", + "D": "You saw 0 plastic bags with items inside on the table at that time.", + "E": "You saw 4 plastic bags with items inside on the table during the meal." + }, + "answer_label": "C", + "answer_text": "You saw 2 plastic bags with items inside on the table.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_1_T3", + "correct_answer": "You saw 2 plastic bags with items inside on the table.", + "original_options": { + "A": "You saw 1 bag.", + "B": "You saw 3 bags.", + "C": "You saw 2 plastic bags with items inside on the table.", + "D": "You saw 0 plastic bags with items inside on the table at that time.", + "E": "You saw 4 plastic bags with items inside on the table during the meal." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_6_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "After I entered through the building entrance door, how many bicycles were inside the building, to the left of the doorway?", + "answer": "After you entered through the building door, there were three bicycles inside, to your left at the entrance.", + "question_chinese": "我进到楼栋门后，楼栋里面在门口左边的有几辆自行车？", + "answer_chinese": "你进到楼栋门后，楼栋里面在门口的左边的有3辆自行车。", + "evidence": [ + { + "description": "There were three bicycles on the left side of the entrance inside the building.", + "timestamps": { + "start": "00:22:55", + "end": "00:22:58" + } + } + ], + "question_timestamps": "00:23:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "After you entered through the building door, there were three bicycles inside, to your left at the entrance.", + "B": "There were two bicycles.", + "C": "There were four bicycles.", + "D": "After you entered through the building door, there were five bicycles inside, to your left at the entrance.", + "E": "After you entered through the building door, there was one bicycle inside, to your left at the entrance." + }, + "answer_label": "A", + "answer_text": "After you entered through the building door, there were three bicycles inside, to your left at the entrance.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_6_T1", + "correct_answer": "After you entered through the building door, there were three bicycles inside, to your left at the entrance.", + "original_options": { + "A": "After you entered through the building door, there were three bicycles inside, to your left at the entrance.", + "B": "There were two bicycles.", + "C": "There were four bicycles.", + "D": "After you entered through the building door, there were five bicycles inside, to your left at the entrance.", + "E": "After you entered through the building door, there was one bicycle inside, to your left at the entrance." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 18 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_9", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many bicycles are there on the left side of the door when I enter the building?", + "answer": "After you entered the building, there were three bicycles on the left side of the door inside the building.", + "question_chinese": "我进到楼栋门后，楼栋里面在门口左边的有几辆自行车？", + "answer_chinese": "你进到楼栋门后，楼栋里面在门口左边的有3辆自行车。", + "evidence": [ + { + "description": "There were three bicycles on the left side of the entrance inside the building.", + "timestamps": { + "start": "00:22:55", + "end": "00:22:58" + } + } + ], + "question_timestamps": "00:23:45", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There were two bicycles on the left.", + "B": "There were four bicycles on the left.", + "C": "After you entered the building, there were three bicycles on the right side of the door inside the building.", + "D": "After you entered the building, there were three bicycles on the left side of the door inside the building.", + "E": "After you entered the building, there were three bicycles on the left side of the door outside the building." + }, + "answer_label": "D", + "answer_text": "After you entered the building, there were three bicycles on the left side of the door inside the building.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_9", + "correct_answer": "After you entered the building, there were three bicycles on the left side of the door inside the building.", + "original_options": { + "A": "There were two bicycles on the left.", + "B": "There were four bicycles on the left.", + "C": "After you entered the building, there were three bicycles on the right side of the door inside the building.", + "D": "After you entered the building, there were three bicycles on the left side of the door inside the building.", + "E": "After you entered the building, there were three bicycles on the left side of the door outside the building." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_6_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "After I entered through the building entrance door, how many bicycles were inside the building, to the left of the doorway?", + "answer": "Earlier, after you went in through the stairwell door, you saw three bicycles inside the building, to your left at the entrance.", + "question_chinese": "我进到楼栋门后，楼栋里面在门口左边的有几辆自行车？", + "answer_chinese": "你之前在进到楼梯门后，看见楼栋里面在门口左边的有3辆自行车。", + "evidence": [ + { + "description": "There were three bicycles on the left side of the entrance inside the building.", + "timestamps": { + "start": "00:22:55", + "end": "00:22:58" + } + } + ], + "question_timestamps": "00:23:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two bicycles to your left.", + "B": "Earlier, after you went in through the stairwell door, you saw three bicycles inside the building, to your left at the entrance.", + "C": "You saw four bicycles to your left.", + "D": "After entering through the stairwell door, you noticed one bicycle inside the building on your left.", + "E": "After you went in through the stairwell door, you saw five bicycles parked inside, to your left by the entrance." + }, + "answer_label": "B", + "answer_text": "Earlier, after you went in through the stairwell door, you saw three bicycles inside the building, to your left at the entrance.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_6_T2", + "correct_answer": "Earlier, after you went in through the stairwell door, you saw three bicycles inside the building, to your left at the entrance.", + "original_options": { + "A": "You saw two bicycles to your left.", + "B": "Earlier, after you went in through the stairwell door, you saw three bicycles inside the building, to your left at the entrance.", + "C": "You saw four bicycles to your left.", + "D": "After entering through the stairwell door, you noticed one bicycle inside the building on your left.", + "E": "After you went in through the stairwell door, you saw five bicycles parked inside, to your left by the entrance." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 20 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plastic bags are there on the table containing things?", + "answer": "There are four plastic bags containing things on the table.", + "question_chinese": "桌上有几个装着东西的塑料袋？", + "answer_chinese": "桌上有4个装着东西的塑料袋。", + "evidence": [ + { + "description": "There are four plastic bags containing things on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:19" + } + } + ], + "question_timestamps": "00:01:23", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three plastic bags.", + "B": "There are four plastic bags containing things on the table.", + "C": "There are five plastic bags.", + "D": "There are two plastic bags containing things on the table, not four as you might expect.", + "E": "There are six plastic bags containing things on the table, all grouped together among the items." + }, + "answer_label": "B", + "answer_text": "There are four plastic bags containing things on the table.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_2", + "correct_answer": "There are four plastic bags containing things on the table.", + "original_options": { + "A": "There are three plastic bags.", + "B": "There are four plastic bags containing things on the table.", + "C": "There are five plastic bags.", + "D": "There are two plastic bags containing things on the table, not four as you might expect.", + "E": "There are six plastic bags containing things on the table, all grouped together among the items." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_1", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plastic bags are there on the table containing things?", + "answer": "There are three plastic bags containing things on the table.", + "question_chinese": "桌上有几个装着东西的塑料袋？", + "answer_chinese": "桌上有4个装着东西的塑料袋。", + "evidence": [ + { + "description": "There are four plastic bags containing things on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:19" + } + }, + { + "description": "I am taking the cola out of the bag and put the plastic bag into the blue plastic bag.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:23" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are two plastic bags.", + "B": "There are four plastic bags.", + "C": "There are five plastic bags containing things on the table, not three as stated.", + "D": "There are six plastic bags containing things on the table, each holding items inside.", + "E": "There are three plastic bags containing things on the table." + }, + "answer_label": "E", + "answer_text": "There are three plastic bags containing things on the table.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_1", + "correct_answer": "There are three plastic bags containing things on the table.", + "original_options": { + "A": "There are two plastic bags.", + "B": "There are four plastic bags.", + "C": "There are five plastic bags containing things on the table, not three as stated.", + "D": "There are six plastic bags containing things on the table, each holding items inside.", + "E": "There are three plastic bags containing things on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_3", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plastic bags are there on the table containing things?", + "answer": "There are two plastic bags on the table containing things.", + "question_chinese": "桌上有几个装着东西的塑料袋？", + "answer_chinese": "桌上有2个装着东西的塑料袋。", + "evidence": [ + { + "description": "I put the white plastic bag into the blue plastic bag.", + "timestamps": { + "start": "00:21:10", + "end": "00:21:14" + } + }, + { + "description": "I took the cola out of the bag and put the plastic bag into the blue plastic bag.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:23" + } + }, + { + "description": "There are four plastic bags on the table.", + "timestamps": { + "start": "00:00:08", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:21:14", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are two plastic bags on the table containing things.", + "B": "There is one plastic bag.", + "C": "There are three plastic bags.", + "D": "There are four plastic bags on the table containing things, rather than only two.", + "E": "There are five plastic bags on the table containing things, spread among the other items." + }, + "answer_label": "A", + "answer_text": "There are two plastic bags on the table containing things.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_3", + "correct_answer": "There are two plastic bags on the table containing things.", + "original_options": { + "A": "There are two plastic bags on the table containing things.", + "B": "There is one plastic bag.", + "C": "There are three plastic bags.", + "D": "There are four plastic bags on the table containing things, rather than only two.", + "E": "There are five plastic bags on the table containing things, spread among the other items." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_5", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of sauce have I opened so far?", + "answer": "So far, you have opened a box of sauce.", + "question_chinese": "目前为止我拆开了几盒酱料？", + "answer_chinese": "目前为止，你拆开了一盒酱料。", + "evidence": [ + { + "description": "I opened the first box of sauce.", + "timestamps": { + "start": "00:02:29", + "end": "00:02:35" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have opened zero boxes.", + "B": "You have opened two boxes.", + "C": "So far, you have opened a box of sauce.", + "D": "So far, you have opened three boxes of sauce while preparing everything.", + "E": "So far, you have opened four boxes of sauce, not just a single one." + }, + "answer_label": "C", + "answer_text": "So far, you have opened a box of sauce.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_5", + "correct_answer": "So far, you have opened a box of sauce.", + "original_options": { + "A": "You have opened zero boxes.", + "B": "You have opened two boxes.", + "C": "So far, you have opened a box of sauce.", + "D": "So far, you have opened three boxes of sauce while preparing everything.", + "E": "So far, you have opened four boxes of sauce, not just a single one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 14 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_2_T2", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many boxes of sauce have I opened?", + "answer": "So far, you’ve opened one box of sauce.", + "question_chinese": "目前为止我拆开了几盒酱料？", + "answer_chinese": "目前为止，你拆开了一盒酱料。", + "evidence": [ + { + "description": "I opened the first box of sauce.", + "timestamps": { + "start": "00:02:29", + "end": "00:02:35" + } + } + ], + "question_timestamps": "00:02:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve opened zero boxes.", + "B": "You’ve opened two boxes.", + "C": "So far, you’ve opened one box of sauce.", + "D": "So far, you’ve opened three boxes of sauce in total up to this point.", + "E": "So far, you’ve opened four boxes of sauce already, more than just one." + }, + "answer_label": "C", + "answer_text": "So far, you’ve opened one box of sauce.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_2_T2", + "correct_answer": "So far, you’ve opened one box of sauce.", + "original_options": { + "A": "You’ve opened zero boxes.", + "B": "You’ve opened two boxes.", + "C": "So far, you’ve opened one box of sauce.", + "D": "So far, you’ve opened three boxes of sauce in total up to this point.", + "E": "So far, you’ve opened four boxes of sauce already, more than just one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "9a7a189e-3494-40c0-84ee-5be47af574f7", + "video_length": "00:24:21", + "qid": "9a7a189e-3494-40c0-84ee-5be47af574f7_2_T1", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many boxes of sauce have I opened?", + "answer": "So far, you’ve opened two boxes of sauce.", + "question_chinese": "目前为止我拆开了几盒酱料？", + "answer_chinese": "目前为止，你拆开了两盒酱料。", + "evidence": [ + { + "description": "I opened the first box of sauce.", + "timestamps": { + "start": "00:02:29", + "end": "00:02:35" + } + }, + { + "description": "I opened the second box of sauce.", + "timestamps": { + "start": "00:07:22", + "end": "00:07:25" + } + } + ], + "question_timestamps": "00:07:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve opened one box.", + "B": "So far, you’ve opened two boxes of sauce.", + "C": "You’ve opened three boxes.", + "D": "So far, you’ve opened zero boxes of sauce, meaning none have been opened yet.", + "E": "So far, you’ve opened four boxes of sauce in total, not only two." + }, + "answer_label": "B", + "answer_text": "So far, you’ve opened two boxes of sauce.", + "q_id": "9a7a189e-3494-40c0-84ee-5be47af574f7_2_T1", + "correct_answer": "So far, you’ve opened two boxes of sauce.", + "original_options": { + "A": "You’ve opened one box.", + "B": "So far, you’ve opened two boxes of sauce.", + "C": "You’ve opened three boxes.", + "D": "So far, you’ve opened zero boxes of sauce, meaning none have been opened yet.", + "E": "So far, you’ve opened four boxes of sauce in total, not only two." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + } +] diff --git a/QAs/9ad4f150-15fe-4e1b-b145-331475afbe0c-mcq.json b/QAs/9ad4f150-15fe-4e1b-b145-331475afbe0c-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..375dc6315a354149bd4fb3f93028abd0354fc9c0 --- /dev/null +++ b/QAs/9ad4f150-15fe-4e1b-b145-331475afbe0c-mcq.json @@ -0,0 +1,1301 @@ +[ + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_6_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The fridge is behind you.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱在你的后方。", + "evidence": [ + { + "description": "The refrigerator is behind you.", + "timestamps": { + "start": "00:29:20", + "end": "00:29:24" + } + } + ], + "question_timestamps": "00:29:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is in front of you.", + "B": "The fridge is to your left.", + "C": "The fridge is to your right, positioned beside you rather than behind you.", + "D": "The fridge is above you, higher up in your view instead of behind your position.", + "E": "The fridge is behind you." + }, + "answer_label": "E", + "answer_text": "The fridge is behind you.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_6_T1", + "correct_answer": "The fridge is behind you.", + "original_options": { + "A": "The fridge is in front of you.", + "B": "The fridge is to your left.", + "C": "The fridge is to your right, positioned beside you rather than behind you.", + "D": "The fridge is above you, higher up in your view instead of behind your position.", + "E": "The fridge is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 15 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_12", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator now relative to me?", + "answer": "The refrigerator is behind you.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱在你的后方。", + "evidence": [ + { + "description": "The refrigerator is behind you.", + "timestamps": { + "start": "00:41:01", + "end": "00:41:02" + } + } + ], + "question_timestamps": "00:41:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is in front of you.", + "B": "The refrigerator is to your right.", + "C": "The refrigerator is behind you.", + "D": "The refrigerator is on your left, just off to the side as you face forward.", + "E": "The refrigerator is below you, positioned lower than your current eye level." + }, + "answer_label": "C", + "answer_text": "The refrigerator is behind you.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_12", + "correct_answer": "The refrigerator is behind you.", + "original_options": { + "A": "The refrigerator is in front of you.", + "B": "The refrigerator is to your right.", + "C": "The refrigerator is behind you.", + "D": "The refrigerator is on your left, just off to the side as you face forward.", + "E": "The refrigerator is below you, positioned lower than your current eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 12 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_6_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator relative to me right now?", + "answer": "The fridge was to your right.", + "question_chinese": "冰箱现在相对我是什么位置？", + "answer_chinese": "冰箱在你的右边。", + "evidence": [ + { + "description": "The refrigerator is behind you.", + "timestamps": { + "start": "00:41:01", + "end": "00:41:02" + } + } + ], + "question_timestamps": "00:42:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge was in front of you.", + "B": "The fridge was to your left.", + "C": "The fridge was behind you, so you would have needed to turn around to see it.", + "D": "The fridge was to your right.", + "E": "The fridge was above you, positioned higher than your head as you looked around." + }, + "answer_label": "D", + "answer_text": "The fridge was to your right.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_6_T2", + "correct_answer": "The fridge was to your right.", + "original_options": { + "A": "The fridge was in front of you.", + "B": "The fridge was to your left.", + "C": "The fridge was behind you, so you would have needed to turn around to see it.", + "D": "The fridge was to your right.", + "E": "The fridge was above you, positioned higher than your head as you looked around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 14 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_9", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the kitchen now relative to me?", + "answer": "The kitchen is now on your right.", + "question_chinese": "厨房现在相对我是什么位置？", + "answer_chinese": "厨房现在在你的右侧。", + "evidence": [ + { + "description": "The kitchen is now on your right.", + "timestamps": { + "start": "00:23:01", + "end": "00:23:05" + } + } + ], + "question_timestamps": "00:23:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The kitchen is now on your left.", + "B": "The kitchen is now behind you.", + "C": "The kitchen is now on your right.", + "D": "The kitchen is now in front of you, directly ahead as you face forward.", + "E": "The kitchen is now downstairs from you, requiring you to go down to reach it." + }, + "answer_label": "C", + "answer_text": "The kitchen is now on your right.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_9", + "correct_answer": "The kitchen is now on your right.", + "original_options": { + "A": "The kitchen is now on your left.", + "B": "The kitchen is now behind you.", + "C": "The kitchen is now on your right.", + "D": "The kitchen is now in front of you, directly ahead as you face forward.", + "E": "The kitchen is now downstairs from you, requiring you to go down to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 15 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_10", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the kitchen now relative to me?", + "answer": "The kitchen is now on your left.", + "question_chinese": "厨房现在相对我是什么位置？", + "answer_chinese": "厨房现在你的左侧。", + "evidence": [ + { + "description": "The kitchen is now on your left.", + "timestamps": { + "start": "00:36:04", + "end": "00:36:11" + } + } + ], + "question_timestamps": "00:36:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The kitchen is now on your right.", + "B": "The kitchen is now on your left.", + "C": "The kitchen is now behind you.", + "D": "The kitchen is now in front of you, straight ahead from where you’re facing.", + "E": "The kitchen is now above you, located on a higher level than you are." + }, + "answer_label": "B", + "answer_text": "The kitchen is now on your left.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_10", + "correct_answer": "The kitchen is now on your left.", + "original_options": { + "A": "The kitchen is now on your right.", + "B": "The kitchen is now on your left.", + "C": "The kitchen is now behind you.", + "D": "The kitchen is now in front of you, straight ahead from where you’re facing.", + "E": "The kitchen is now above you, located on a higher level than you are." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 14 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_3_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bathroom right now, do I need to go up or down any stairs?", + "answer": "You need to pass by the stairs.", + "question_chinese": "如果我现在要去到浴室需要经过楼梯吗？", + "answer_chinese": "你需要经过楼梯。", + "evidence": [ + { + "description": "You need to go to the second floor.", + "timestamps": { + "start": "00:07:40", + "end": "00:07:53" + } + } + ], + "question_timestamps": "00:08:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You need to pass by the stairs.", + "B": "You don’t need to pass by the stairs—you’re already in the bathroom." + }, + "answer_label": "A", + "answer_text": "You need to pass by the stairs.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_3_T1", + "correct_answer": "You need to pass by the stairs.", + "original_options": { + "A": "You need to pass by the stairs.", + "B": "You don’t need to pass by the stairs—you’re already in the bathroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_3_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bathroom right now, do I need to go up or down any stairs?", + "answer": "You don’t need to pass by the stairs—you’re already in the bathroom.", + "question_chinese": "如果我现在要去到浴室需要经过楼梯吗？", + "answer_chinese": "你不需要经过楼梯，你现在就在浴室。", + "evidence": [ + { + "description": "You need to go to the second floor.", + "timestamps": { + "start": "00:07:40", + "end": "00:07:53" + } + }, + { + "description": "You are on the first floor now.", + "timestamps": { + "start": "00:09:59", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:10:06", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You need to pass by the stairs.", + "B": "You don’t need to pass by the stairs—you’re already in the bathroom." + }, + "answer_label": "B", + "answer_text": "You don’t need to pass by the stairs—you’re already in the bathroom.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_3_T2", + "correct_answer": "You don’t need to pass by the stairs—you’re already in the bathroom.", + "original_options": { + "A": "You need to pass by the stairs.", + "B": "You don’t need to pass by the stairs—you’re already in the bathroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_5", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Do I need to go up the stairs if I want to get to the bathroom now?", + "answer": "You need to go up the stairs.", + "question_chinese": "如果我现在要去到浴室需要经过楼梯吗？", + "answer_chinese": "你需要经过楼梯。", + "evidence": [ + { + "description": "You need to go to the second floor.", + "timestamps": { + "start": "00:07:40", + "end": "00:07:53" + } + }, + { + "description": "You are on the first floor now.", + "timestamps": { + "start": "00:09:59", + "end": "00:10:05" + } + } + ], + "question_timestamps": "00:14:35", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you don't need to go up the stairs.", + "B": "You need to go up the stairs." + }, + "answer_label": "B", + "answer_text": "You need to go up the stairs.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_5", + "correct_answer": "You need to go up the stairs.", + "original_options": { + "A": "No, you don't need to go up the stairs.", + "B": "You need to go up the stairs." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_6", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Do I need to go up the stairs if I want to get to the bathroom now?", + "answer": "You need to go up the stairs.", + "question_chinese": "如果我现在要去到浴室需要经过楼梯吗？", + "answer_chinese": "你需要经过楼梯。", + "evidence": [ + { + "description": "You need to go to the second floor.", + "timestamps": { + "start": "00:07:40", + "end": "00:07:53" + } + }, + { + "description": "You are on the first floor now.", + "timestamps": { + "start": "00:09:59", + "end": "00:10:05" + } + } + ], + "question_timestamps": "00:22:35", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You need to go up the stairs.", + "B": "No, you don't need to go up the stairs." + }, + "answer_label": "A", + "answer_text": "You need to go up the stairs.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_6", + "correct_answer": "You need to go up the stairs.", + "original_options": { + "A": "You need to go up the stairs.", + "B": "No, you don't need to go up the stairs." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_15", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the ladder?", + "answer": "You just placed the ladder against the wall.", + "question_chinese": "我刚刚把梯子放在什么位置了？", + "answer_chinese": "你刚刚把梯子靠着墙边放着。", + "evidence": [ + { + "description": "You just placed the ladder against the wall.", + "timestamps": { + "start": "00:46:42", + "end": "00:46:50" + } + } + ], + "question_timestamps": "00:46:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You placed the ladder by the window.", + "B": "You just placed the ladder against the wall.", + "C": "You leaned the ladder against the door.", + "D": "You just leaned the ladder against the cabinet, not against the wall where you were standing.", + "E": "You just set the ladder beside the staircase, leaving it off to the side instead of against the wall." + }, + "answer_label": "B", + "answer_text": "You just placed the ladder against the wall.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_15", + "correct_answer": "You just placed the ladder against the wall.", + "original_options": { + "A": "You placed the ladder by the window.", + "B": "You just placed the ladder against the wall.", + "C": "You leaned the ladder against the door.", + "D": "You just leaned the ladder against the cabinet, not against the wall where you were standing.", + "E": "You just set the ladder beside the staircase, leaving it off to the side instead of against the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 19 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_8_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just place the ladder?", + "answer": "You just leaned the ladder against the wall, and it’s now in front of you.", + "question_chinese": "我刚刚把梯子放在什么位置了？", + "answer_chinese": "你刚刚把梯子靠着墙边放着，现在在你的前方。", + "evidence": [ + { + "description": "You just placed the ladder against the wall.", + "timestamps": { + "start": "00:46:42", + "end": "00:46:50" + } + } + ], + "question_timestamps": "00:46:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You just leaned the ladder against the wall, and it’s now in front of you.", + "B": "It’s behind you now.", + "C": "It’s to your left now.", + "D": "You just leaned the ladder against the wall, but it’s positioned behind you rather than in front.", + "E": "You just leaned the ladder against the wall, and it’s now off to your right side instead of in front." + }, + "answer_label": "A", + "answer_text": "You just leaned the ladder against the wall, and it’s now in front of you.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_8_T2", + "correct_answer": "You just leaned the ladder against the wall, and it’s now in front of you.", + "original_options": { + "A": "You just leaned the ladder against the wall, and it’s now in front of you.", + "B": "It’s behind you now.", + "C": "It’s to your left now.", + "D": "You just leaned the ladder against the wall, but it’s positioned behind you rather than in front.", + "E": "You just leaned the ladder against the wall, and it’s now off to your right side instead of in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 21 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_8_T1", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just place the ladder?", + "answer": "You just leaned the ladder against the wall, and it’s now on the left side.", + "question_chinese": "我刚刚把梯子放在什么位置了？", + "answer_chinese": "你刚刚把梯子靠着墙边放着，现在在你的左侧。", + "evidence": [ + { + "description": "You just placed the ladder against the wall.", + "timestamps": { + "start": "00:46:42", + "end": "00:46:50" + } + } + ], + "question_timestamps": "00:47:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You just leaned the ladder against the wall, and it’s now on the left side.", + "B": "It’s in front of you now.", + "C": "It’s behind you now.", + "D": "You just leaned the ladder against the wall, but it ended up on your right side, not the left.", + "E": "You just leaned the ladder against the wall, and it’s now directly in front of you rather than on the left." + }, + "answer_label": "A", + "answer_text": "You just leaned the ladder against the wall, and it’s now on the left side.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_8_T1", + "correct_answer": "You just leaned the ladder against the wall, and it’s now on the left side.", + "original_options": { + "A": "You just leaned the ladder against the wall, and it’s now on the left side.", + "B": "It’s in front of you now.", + "C": "It’s behind you now.", + "D": "You just leaned the ladder against the wall, but it ended up on your right side, not the left.", + "E": "You just leaned the ladder against the wall, and it’s now directly in front of you rather than on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 19, + 22 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_13", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the ladder directly now?", + "answer": "You can touch it directly.", + "question_chinese": "我现在可以直接触碰到梯子吗？", + "answer_chinese": "你可以直接触碰到。", + "evidence": [ + { + "description": "You can touch it directly.", + "timestamps": { + "start": "00:35:20", + "end": "00:35:25" + } + } + ], + "question_timestamps": "00:35:28", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can touch it directly.", + "B": "You cannot touch it directly." + }, + "answer_label": "A", + "answer_text": "You can touch it directly.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_13", + "correct_answer": "You can touch it directly.", + "original_options": { + "A": "You can touch it directly.", + "B": "You cannot touch it directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_7_T1", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the ladder right now?", + "answer": "You can’t directly reach the ladder.", + "question_chinese": "我现在可以直接触碰到梯子吗？", + "answer_chinese": "你无法直接触碰到梯子。", + "evidence": [ + { + "description": "You can touch it directly.", + "timestamps": { + "start": "00:35:20", + "end": "00:35:25" + } + } + ], + "question_timestamps": "00:35:42", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can directly reach the ladder.", + "B": "You can’t directly reach the ladder." + }, + "answer_label": "B", + "answer_text": "You can’t directly reach the ladder.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_7_T1", + "correct_answer": "You can’t directly reach the ladder.", + "original_options": { + "A": "You can directly reach the ladder.", + "B": "You can’t directly reach the ladder." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_14", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the ladder directly now?", + "answer": "You can touch it directly.", + "question_chinese": "我现在可以直接触碰到梯子吗？", + "answer_chinese": "你可以直接触碰到。", + "evidence": [ + { + "description": "You can touch it directly.", + "timestamps": { + "start": "00:43:48", + "end": "00:43:51" + } + } + ], + "question_timestamps": "00:43:52", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot touch it directly.", + "B": "You can touch it directly." + }, + "answer_label": "B", + "answer_text": "You can touch it directly.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_14", + "correct_answer": "You can touch it directly.", + "original_options": { + "A": "You cannot touch it directly.", + "B": "You can touch it directly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_7_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the ladder right now?", + "answer": "You can directly reach the ladder.", + "question_chinese": "我现在可以直接触碰到梯子吗？", + "answer_chinese": "你可以直接触碰到梯子。", + "evidence": [ + { + "description": "You can touch it directly.", + "timestamps": { + "start": "00:43:48", + "end": "00:43:51" + } + } + ], + "question_timestamps": "00:43:53", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can directly reach the ladder.", + "B": "You can’t directly reach the ladder." + }, + "answer_label": "A", + "answer_text": "You can directly reach the ladder.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_7_T2", + "correct_answer": "You can directly reach the ladder.", + "original_options": { + "A": "You can directly reach the ladder.", + "B": "You can’t directly reach the ladder." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_3", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the red bucket directly now?", + "answer": "You can't make direct contact now.", + "question_chinese": "我现在可以直接触碰到红色的桶吗？", + "answer_chinese": "你现在无法直接触碰到。", + "evidence": [ + { + "description": "You can't make direct contact now.", + "timestamps": { + "start": "00:09:01", + "end": "00:09:07" + } + } + ], + "question_timestamps": "00:09:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now make direct contact.", + "B": "You can't make direct contact now." + }, + "answer_label": "B", + "answer_text": "You can't make direct contact now.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_3", + "correct_answer": "You can't make direct contact now.", + "original_options": { + "A": "You can now make direct contact.", + "B": "You can't make direct contact now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_4", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the red bucket directly now?", + "answer": "You can now make direct contact.", + "question_chinese": "我现在可以直接触碰到红色的桶吗？", + "answer_chinese": "你现在可以直接触碰到。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:20:28", + "end": "00:20:33" + } + } + ], + "question_timestamps": "00:20:34", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now make direct contact.", + "B": "You cannot make direct contact yet." + }, + "answer_label": "A", + "answer_text": "You can now make direct contact.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_4", + "correct_answer": "You can now make direct contact.", + "original_options": { + "A": "You can now make direct contact.", + "B": "You cannot make direct contact yet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_1_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the mirror right now?", + "answer": "You can directly reach the mirror right now.", + "question_chinese": "我现在可以直接触碰到镜子吗？", + "answer_chinese": "你现在可以直接触碰到镜子。", + "evidence": [ + { + "description": "You can touch the mirror directly now.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the mirror right now.", + "B": "You can directly reach the mirror right now." + }, + "answer_label": "B", + "answer_text": "You can directly reach the mirror right now.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_1_T1", + "correct_answer": "You can directly reach the mirror right now.", + "original_options": { + "A": "You can’t reach the mirror right now.", + "B": "You can directly reach the mirror right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_1", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the mirror directly now?", + "answer": "Can I touch the mirror directly now?", + "question_chinese": "我现在可以直接触碰到镜子吗？", + "answer_chinese": "你现在无法直接触碰到。", + "evidence": [ + { + "description": "Can I touch the mirror directly now?", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you cannot touch the mirror yet.", + "B": "Can I touch the mirror directly now?" + }, + "answer_label": "B", + "answer_text": "Can I touch the mirror directly now?", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_1", + "correct_answer": "Can I touch the mirror directly now?", + "original_options": { + "A": "No, you cannot touch the mirror yet.", + "B": "Can I touch the mirror directly now?" + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_1_T2", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly touch the mirror right now?", + "answer": "You can’t reach the mirror right now.", + "question_chinese": "我现在可以直接触碰到镜子吗？", + "answer_chinese": "你现在无法触碰到镜子。", + "evidence": [ + { + "description": "Can I touch the mirror directly now?", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:15:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the mirror right now.", + "B": "You can reach the mirror right now." + }, + "answer_label": "A", + "answer_text": "You can’t reach the mirror right now.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_1_T2", + "correct_answer": "You can’t reach the mirror right now.", + "original_options": { + "A": "You can’t reach the mirror right now.", + "B": "You can reach the mirror right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_19", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the door handle directly now?", + "answer": "You can now make direct contact.", + "question_chinese": "我现在可以直接触碰到门把手吗？", + "answer_chinese": "你现在可以直接触碰到。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:18:14", + "end": "00:18:18" + } + } + ], + "question_timestamps": "00:18:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now make direct contact.", + "B": "You can't make direct contact now." + }, + "answer_label": "A", + "answer_text": "You can now make direct contact.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_19", + "correct_answer": "You can now make direct contact.", + "original_options": { + "A": "You can now make direct contact.", + "B": "You can't make direct contact now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_20", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the door handle directly now?", + "answer": "You can't make direct contact now.", + "question_chinese": "我现在可以直接触碰到门把手吗？", + "answer_chinese": "你现在无法直接触碰到。", + "evidence": [ + { + "description": "You can't make direct contact now.", + "timestamps": { + "start": "00:28:19", + "end": "00:28:30" + } + } + ], + "question_timestamps": "00:28:40", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't make direct contact now.", + "B": "You can now make direct contact." + }, + "answer_label": "A", + "answer_text": "You can't make direct contact now.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_20", + "correct_answer": "You can't make direct contact now.", + "original_options": { + "A": "You can't make direct contact now.", + "B": "You can now make direct contact." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_17", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the yellow bucket directly now?", + "answer": "You can now make direct contact.", + "question_chinese": "我现在可以直接触碰到黄色的桶吗？", + "answer_chinese": "你现在可以直接触碰到。", + "evidence": [ + { + "description": "You can now make direct contact.", + "timestamps": { + "start": "00:35:30", + "end": "00:35:34" + } + } + ], + "question_timestamps": "00:35:35", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now make direct contact.", + "B": "You can't make direct contact now." + }, + "answer_label": "A", + "answer_text": "You can now make direct contact.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_17", + "correct_answer": "You can now make direct contact.", + "original_options": { + "A": "You can now make direct contact.", + "B": "You can't make direct contact now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_18", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the yellow bucket directly now?", + "answer": "You can't make direct contact now.", + "question_chinese": "我现在可以直接触碰到黄色的桶吗？", + "answer_chinese": "你现在无法直接触碰到。", + "evidence": [ + { + "description": "You can't make direct contact now.", + "timestamps": { + "start": "00:43:51", + "end": "00:43:58" + } + } + ], + "question_timestamps": "00:44:14", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can now make direct contact.", + "B": "You can't make direct contact now." + }, + "answer_label": "B", + "answer_text": "You can't make direct contact now.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_18", + "correct_answer": "You can't make direct contact now.", + "original_options": { + "A": "You can now make direct contact.", + "B": "You can't make direct contact now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_7", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the bathroom now relative to me?", + "answer": "The bathroom is on your right.", + "question_chinese": "浴室现在相对我是什么位置？", + "answer_chinese": "浴室在你的右侧。", + "evidence": [ + { + "description": "The bathroom is on your right.", + "timestamps": { + "start": "00:07:35", + "end": "00:07:43" + } + } + ], + "question_timestamps": "00:07:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The bathroom is on your right.", + "B": "The bathroom is on your left.", + "C": "The bathroom is behind you.", + "D": "The bathroom is straight ahead of you, past the doorway in front of you.", + "E": "The bathroom is to your left, just beyond the hall corner you are facing." + }, + "answer_label": "A", + "answer_text": "The bathroom is on your right.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_7", + "correct_answer": "The bathroom is on your right.", + "original_options": { + "A": "The bathroom is on your right.", + "B": "The bathroom is on your left.", + "C": "The bathroom is behind you.", + "D": "The bathroom is straight ahead of you, past the doorway in front of you.", + "E": "The bathroom is to your left, just beyond the hall corner you are facing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 14 + ] + }, + { + "video_uid": "9ad4f150-15fe-4e1b-b145-331475afbe0c", + "video_length": "00:50:25", + "qid": "9ad4f150-15fe-4e1b-b145-331475afbe0c_8", + "qid_num": 27, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the bathroom now relative to me?", + "answer": "The bathroom is now on your left.", + "question_chinese": "浴室现在相对我是什么位置？", + "answer_chinese": "浴室现在在你的左侧。", + "evidence": [ + { + "description": "The bathroom is now on your left.", + "timestamps": { + "start": "00:23:32", + "end": "00:23:46" + } + } + ], + "question_timestamps": "00:23:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The bathroom is in front of you.", + "B": "The bathroom is to your right.", + "C": "The bathroom is behind you, in the direction you just came from.", + "D": "The bathroom is now on your left.", + "E": "The bathroom is upstairs from you, at the top of the stairs ahead." + }, + "answer_label": "D", + "answer_text": "The bathroom is now on your left.", + "q_id": "9ad4f150-15fe-4e1b-b145-331475afbe0c_8", + "correct_answer": "The bathroom is now on your left.", + "original_options": { + "A": "The bathroom is in front of you.", + "B": "The bathroom is to your right.", + "C": "The bathroom is behind you, in the direction you just came from.", + "D": "The bathroom is now on your left.", + "E": "The bathroom is upstairs from you, at the top of the stairs ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 13 + ] + } +] diff --git a/QAs/9b221f93-19a9-4593-9187-049d9dedfcf0-mcq.json b/QAs/9b221f93-19a9-4593-9187-049d9dedfcf0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..10c2f14788b02a6d9740a7c302945a8e0fa28e47 --- /dev/null +++ b/QAs/9b221f93-19a9-4593-9187-049d9dedfcf0-mcq.json @@ -0,0 +1,1061 @@ +[ + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_15", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Are the two buckets close to each other now? How far apart are they approximately?", + "answer": "It is relatively far away, about three meters apart.", + "question_chinese": "两个桶现在离得近吗?大概多远?", + "answer_chinese": "较远，相距约3m。", + "evidence": [ + { + "description": "The bucket on the ground is located about 3 meters to the right front of the bucket on the vehicle.", + "timestamps": { + "start": "00:14:56", + "end": "00:14:59" + } + } + ], + "question_timestamps": "00:15:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is relatively far away, about three meters apart.", + "B": "Very close, about 10cm apart.", + "C": "About one meter apart.", + "D": "They are only slightly separated, roughly two meters apart from each other.", + "E": "They are quite far from each other, approximately five meters apart now." + }, + "answer_label": "A", + "answer_text": "It is relatively far away, about three meters apart.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_15", + "correct_answer": "It is relatively far away, about three meters apart.", + "original_options": { + "A": "It is relatively far away, about three meters apart.", + "B": "Very close, about 10cm apart.", + "C": "About one meter apart.", + "D": "They are only slightly separated, roughly two meters apart from each other.", + "E": "They are quite far from each other, approximately five meters apart now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 12, + 12 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_16", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Are the two buckets close to each other now? How far apart are they approximately?", + "answer": "Very close, about 10cm apart.", + "question_chinese": "两个桶现在离得近吗?大概多远?", + "answer_chinese": "很近,相距约10cm。", + "evidence": [ + { + "description": "Very close, the bucket on the car is on the left side of the bucket on the ground, about 10cm apart.", + "timestamps": { + "start": "00:41:56", + "end": "00:42:07" + } + } + ], + "question_timestamps": "00:42:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "About one meter apart.", + "B": "Very close, about 10cm apart.", + "C": "Roughly 30cm apart.", + "D": "They are separated by about half a meter, so they are not that close.", + "E": "They are far apart now, with roughly three meters of space between them." + }, + "answer_label": "B", + "answer_text": "Very close, about 10cm apart.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_16", + "correct_answer": "Very close, about 10cm apart.", + "original_options": { + "A": "About one meter apart.", + "B": "Very close, about 10cm apart.", + "C": "Roughly 30cm apart.", + "D": "They are separated by about half a meter, so they are not that close.", + "E": "They are far apart now, with roughly three meters of space between them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 13 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_3_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If the leaf blower runs out of gas, where do I go to refuel it?", + "answer": "You need to go refuel the car.", + "question_chinese": "如果鼓风机没油了，我要去哪里加油？", + "answer_chinese": "你要去车上加油。", + "evidence": [ + { + "description": "There is a red oil can on the car.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:01:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go refuel the leaf blower.", + "B": "You need to go refuel the car.", + "C": "Go refill the gas can.", + "D": "You should refuel the leaf blower directly at the pump instead of refueling the car first.", + "E": "Go to the gas station and refill a spare fuel can, then bring it back to the leaf blower." + }, + "answer_label": "B", + "answer_text": "You need to go refuel the car.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_3_T1", + "correct_answer": "You need to go refuel the car.", + "original_options": { + "A": "Go refuel the leaf blower.", + "B": "You need to go refuel the car.", + "C": "Go refill the gas can.", + "D": "You should refuel the leaf blower directly at the pump instead of refueling the car first.", + "E": "Go to the gas station and refill a spare fuel can, then bring it back to the leaf blower." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 19 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_3_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If the leaf blower runs out of gas, where do I go to refuel it?", + "answer": "You should go refuel the car.", + "question_chinese": "如果鼓风机没油了，我要去哪里加油？", + "answer_chinese": "你应该去车上加油。", + "evidence": [ + { + "description": "There is a red oil can on the car.", + "timestamps": { + "start": "00:01:52", + "end": "00:01:54" + } + } + ], + "question_timestamps": "00:22:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go refuel the leaf blower.", + "B": "Go refill the gas can.", + "C": "You should refuel the leaf blower itself rather than refueling the car when it runs out.", + "D": "Go to the nearest pump and fill a spare container, then use it to top up the leaf blower.", + "E": "You should go refuel the car." + }, + "answer_label": "E", + "answer_text": "You should go refuel the car.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_3_T2", + "correct_answer": "You should go refuel the car.", + "original_options": { + "A": "Go refuel the leaf blower.", + "B": "Go refill the gas can.", + "C": "You should refuel the leaf blower itself rather than refueling the car when it runs out.", + "D": "Go to the nearest pump and fill a spare container, then use it to top up the leaf blower.", + "E": "You should go refuel the car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 19 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_13", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I put the gardening scissors after I use them?", + "answer": "The gardening scissors should be placed on the wall to the left of the window.", + "question_chinese": "当我用完园艺剪刀后，它应该放在哪里？", + "answer_chinese": "园艺剪刀应该放在窗户左边的墙边。", + "evidence": [ + { + "description": "There is a pair of garden scissors by the wall on the left of the window.", + "timestamps": { + "start": "00:23:40", + "end": "00:23:43" + } + } + ], + "question_timestamps": "00:23:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Hang them on the right wall.", + "B": "Put them on the shelf above.", + "C": "Place the gardening scissors on the shelf opposite the window after you finish using them.", + "D": "Store the gardening scissors in a drawer under the counter so they are out of the way.", + "E": "The gardening scissors should be placed on the wall to the left of the window." + }, + "answer_label": "E", + "answer_text": "The gardening scissors should be placed on the wall to the left of the window.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_13", + "correct_answer": "The gardening scissors should be placed on the wall to the left of the window.", + "original_options": { + "A": "Hang them on the right wall.", + "B": "Put them on the shelf above.", + "C": "Place the gardening scissors on the shelf opposite the window after you finish using them.", + "D": "Store the gardening scissors in a drawer under the counter so they are out of the way.", + "E": "The gardening scissors should be placed on the wall to the left of the window." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_14", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I put the gardening scissors after I use them?", + "answer": "The gardening scissors should be placed on the shelf opposite the window.", + "question_chinese": "当我用完园艺剪刀后，它应该放在哪里？", + "answer_chinese": "园艺剪刀应该放在窗户对面的台子上。", + "evidence": [ + { + "description": "There is a pair of garden scissors on the table opposite the window.", + "timestamps": { + "start": "00:29:38", + "end": "00:29:52" + } + } + ], + "question_timestamps": "00:29:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Hang them left of the window.", + "B": "The gardening scissors should be placed on the shelf opposite the window.", + "C": "Put them on the counter.", + "D": "Store the gardening scissors in the drawer under the sink once you are done using them.", + "E": "Hang the gardening scissors on the wall beside the other garden tools after you use them." + }, + "answer_label": "B", + "answer_text": "The gardening scissors should be placed on the shelf opposite the window.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_14", + "correct_answer": "The gardening scissors should be placed on the shelf opposite the window.", + "original_options": { + "A": "Hang them left of the window.", + "B": "The gardening scissors should be placed on the shelf opposite the window.", + "C": "Put them on the counter.", + "D": "Store the gardening scissors in the drawer under the sink once you are done using them.", + "E": "Hang the gardening scissors on the wall beside the other garden tools after you use them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 16 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_6_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sips of water did I take in total?", + "answer": "In total, you took three sips of water.", + "question_chinese": "我一共喝了几口水？", + "answer_chinese": "你一共喝了三口水。", + "evidence": [ + { + "description": "You took your first sip of water.", + "timestamps": { + "start": "00:16:16", + "end": "00:16:19" + } + }, + { + "description": "You took your second sip of water.", + "timestamps": { + "start": "00:16:24", + "end": "00:16:26" + } + }, + { + "description": "You took your third sip of water.", + "timestamps": { + "start": "00:16:39", + "end": "00:16:43" + } + } + ], + "question_timestamps": "00:17:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In total, you took two sips.", + "B": "In total, you took four sips.", + "C": "Altogether, you only took one sip of water before you stopped drinking.", + "D": "In total, you took three sips of water.", + "E": "Altogether, you took five sips of water across the whole moment you were drinking." + }, + "answer_label": "D", + "answer_text": "In total, you took three sips of water.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_6_T2", + "correct_answer": "In total, you took three sips of water.", + "original_options": { + "A": "In total, you took two sips.", + "B": "In total, you took four sips.", + "C": "Altogether, you only took one sip of water before you stopped drinking.", + "D": "In total, you took three sips of water.", + "E": "Altogether, you took five sips of water across the whole moment you were drinking." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 14 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_6_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sips of water did I take in total?", + "answer": "In total, you took seven sips of water.", + "question_chinese": "我一共喝了几口水？", + "answer_chinese": "你一共喝了七口水。", + "evidence": [ + { + "description": "You took your first sip of water.", + "timestamps": { + "start": "00:16:16", + "end": "00:16:19" + } + }, + { + "description": "You took your second sip of water.", + "timestamps": { + "start": "00:16:24", + "end": "00:16:26" + } + }, + { + "description": "You took your third sip of water.", + "timestamps": { + "start": "00:16:39", + "end": "00:16:43" + } + }, + { + "description": "You took your fourth sip of water.", + "timestamps": { + "start": "00:16:54", + "end": "00:16:55" + } + }, + { + "description": "You have taken your fifth sip of water.", + "timestamps": { + "start": "00:17:04", + "end": "00:17:06" + } + }, + { + "description": "You have taken your sixth sip of water.", + "timestamps": { + "start": "00:17:12", + "end": "00:17:14" + } + }, + { + "description": "You have drunk the seventh mouthful of water.", + "timestamps": { + "start": "00:17:17", + "end": "00:17:20" + } + } + ], + "question_timestamps": "00:37:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In total, you took six sips.", + "B": "In total, you took eight sips.", + "C": "In total, you took seven sips of water.", + "D": "In total, you took five sips of water altogether during the entire time you drank.", + "E": "In total, you took nine sips of water, counting each distinct drink you made." + }, + "answer_label": "C", + "answer_text": "In total, you took seven sips of water.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_6_T1", + "correct_answer": "In total, you took seven sips of water.", + "original_options": { + "A": "In total, you took six sips.", + "B": "In total, you took eight sips.", + "C": "In total, you took seven sips of water.", + "D": "In total, you took five sips of water altogether during the entire time you drank.", + "E": "In total, you took nine sips of water, counting each distinct drink you made." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 14 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_9", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I rake the grass into the bucket?", + "answer": "You raked the grass twice in total.", + "question_chinese": "我一共耙了几次草进桶里？", + "answer_chinese": "你一共耙了两次草。", + "evidence": [ + { + "description": "You raked the grass for the first time.", + "timestamps": { + "start": "00:12:46", + "end": "00:13:06" + } + }, + { + "description": "You raked the grass for the second time.", + "timestamps": { + "start": "00:13:21", + "end": "00:13:26" + } + } + ], + "question_timestamps": "00:13:28", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You raked the grass once.", + "B": "You raked the grass three times.", + "C": "You raked the grass four times in total, pushing the clippings into the bucket each time.", + "D": "You raked the grass twice in total.", + "E": "You raked the grass five times overall, repeatedly pulling the grass into the bucket." + }, + "answer_label": "D", + "answer_text": "You raked the grass twice in total.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_9", + "correct_answer": "You raked the grass twice in total.", + "original_options": { + "A": "You raked the grass once.", + "B": "You raked the grass three times.", + "C": "You raked the grass four times in total, pushing the clippings into the bucket each time.", + "D": "You raked the grass twice in total.", + "E": "You raked the grass five times overall, repeatedly pulling the grass into the bucket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 14 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_10", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I rake the grass into the bucket?", + "answer": "You raked the grass three times in total.", + "question_chinese": "我一共耙了几次草进桶里？", + "answer_chinese": "你一共耙了三次草。", + "evidence": [ + { + "description": "You raked the grass for the first time.", + "timestamps": { + "start": "00:12:46", + "end": "00:13:06" + } + }, + { + "description": "You raked the grass for the second time.", + "timestamps": { + "start": "00:13:21", + "end": "00:13:26" + } + }, + { + "description": "You raked the grass for the third time.", + "timestamps": { + "start": "00:39:46", + "end": "00:39:58" + } + } + ], + "question_timestamps": "00:40:11", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You raked the grass twice.", + "B": "You raked the grass three times in total.", + "C": "You raked the grass four times.", + "D": "You raked the grass five times in total, scraping the grass into the bucket each pass.", + "E": "You raked the grass six times overall, gathering and dumping grass into the bucket repeatedly." + }, + "answer_label": "B", + "answer_text": "You raked the grass three times in total.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_10", + "correct_answer": "You raked the grass three times in total.", + "original_options": { + "A": "You raked the grass twice.", + "B": "You raked the grass three times in total.", + "C": "You raked the grass four times.", + "D": "You raked the grass five times in total, scraping the grass into the bucket each pass.", + "E": "You raked the grass six times overall, gathering and dumping grass into the bucket repeatedly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_2_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From where I picked up the leaf blower to the car, how many turns did I make in total?", + "answer": "You turned one corner.", + "question_chinese": "我从拿鼓风机的地方到车上，一共拐了几个弯？", + "answer_chinese": "你拐了一个弯。", + "evidence": [ + { + "description": "You only made one turn.", + "timestamps": { + "start": "00:00:43", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You turned zero corners.", + "B": "You turned two corners.", + "C": "You turned three corners in total while walking from the leaf blower pickup spot to the car.", + "D": "You turned one corner.", + "E": "You turned four corners altogether on your way from picking up the leaf blower to reaching the car." + }, + "answer_label": "D", + "answer_text": "You turned one corner.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_2_T1", + "correct_answer": "You turned one corner.", + "original_options": { + "A": "You turned zero corners.", + "B": "You turned two corners.", + "C": "You turned three corners in total while walking from the leaf blower pickup spot to the car.", + "D": "You turned one corner.", + "E": "You turned four corners altogether on your way from picking up the leaf blower to reaching the car." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_2_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From where I picked up the leaf blower to the car, how many turns did I make in total?", + "answer": "You made two turns.", + "question_chinese": "我从拿鼓风机的地方到车上，一共拐了几个弯？", + "answer_chinese": "你拐了两个弯。", + "evidence": [ + { + "description": "You only made one turn.", + "timestamps": { + "start": "00:00:43", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You made one turn.", + "B": "You made three turns.", + "C": "You made zero turns in total, walking straight from the leaf blower to the car.", + "D": "You made four turns overall while going from where you picked up the leaf blower to the car.", + "E": "You made two turns." + }, + "answer_label": "E", + "answer_text": "You made two turns.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_2_T2", + "correct_answer": "You made two turns.", + "original_options": { + "A": "You made one turn.", + "B": "You made three turns.", + "C": "You made zero turns in total, walking straight from the leaf blower to the car.", + "D": "You made four turns overall while going from where you picked up the leaf blower to the car.", + "E": "You made two turns." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 18 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_1_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You are holding a rake.", + "question_chinese": "我手里刚刚拿着什么东西？", + "answer_chinese": "你拿着一个耙子。", + "evidence": [ + { + "description": "You are holding a rake in your hands.", + "timestamps": { + "start": "00:11:30", + "end": "00:11:32" + } + } + ], + "question_timestamps": "00:11:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A brick.", + "B": "A leaf blower.", + "C": "You were holding a brick in one hand while moving it around just before this moment.", + "D": "You were holding a leaf blower just now, gripping it as you moved before setting it down.", + "E": "You are holding a rake." + }, + "answer_label": "E", + "answer_text": "You are holding a rake.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_1_T1", + "correct_answer": "You are holding a rake.", + "original_options": { + "A": "A brick.", + "B": "A leaf blower.", + "C": "You were holding a brick in one hand while moving it around just before this moment.", + "D": "You were holding a leaf blower just now, gripping it as you moved before setting it down.", + "E": "You are holding a rake." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 16, + 17 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_1_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You are holding a rake in your left hand and a brick in your right hand.", + "question_chinese": "我手里刚刚拿着什么东西？", + "answer_chinese": "你拿着左手一个耙子右手拿着一块砖块。", + "evidence": [ + { + "description": "You are holding a rake and a brick in your hands.", + "timestamps": { + "start": "00:11:30", + "end": "00:11:49" + } + } + ], + "question_timestamps": "00:11:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are holding only a rake.", + "B": "You are holding only a brick.", + "C": "You are holding a rake in your left hand and a brick in your right hand.", + "D": "You are holding a rake in your right hand and a brick in your left hand.", + "E": "You are holding a rake and a brick, but both items are in the same hand." + }, + "answer_label": "C", + "answer_text": "You are holding a rake in your left hand and a brick in your right hand.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_1_T2", + "correct_answer": "You are holding a rake in your left hand and a brick in your right hand.", + "original_options": { + "A": "You are holding only a rake.", + "B": "You are holding only a brick.", + "C": "You are holding a rake in your left hand and a brick in your right hand.", + "D": "You are holding a rake in your right hand and a brick in your left hand.", + "E": "You are holding a rake and a brick, but both items are in the same hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_7", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I manage to walk to the exact spot after picking up the bucket?", + "answer": "After picking up the bucket on the ground, you turned around, walked straight ahead, and stood in the grass.", + "question_chinese": "我拿起桶后是怎么走到刚刚的位置的?", + "answer_chinese": "你在地上拿起桶后,转身直行,站在草地里。", + "evidence": [ + { + "description": "You pick up the bucket from the ground and turn around.", + "timestamps": { + "start": "00:12:25", + "end": "00:12:27" + } + }, + { + "description": "You keep going straight.", + "timestamps": { + "start": "00:12:29", + "end": "00:12:34" + } + }, + { + "description": "You have arrived in the meadow.", + "timestamps": { + "start": "00:12:38", + "end": "00:12:41" + } + } + ], + "question_timestamps": "00:12:43", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right and stop on the path.", + "B": "After picking up the bucket on the ground, you turned around, walked straight ahead, and stood in the grass.", + "C": "Walk straight ahead without turning around.", + "D": "After grabbing the bucket, you turned left, walked forward a few steps, and stopped beside the grass.", + "E": "After picking it up, you backed up briefly, then turned around and walked straight until you reached the driveway." + }, + "answer_label": "B", + "answer_text": "After picking up the bucket on the ground, you turned around, walked straight ahead, and stood in the grass.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_7", + "correct_answer": "After picking up the bucket on the ground, you turned around, walked straight ahead, and stood in the grass.", + "original_options": { + "A": "Turn right and stop on the path.", + "B": "After picking up the bucket on the ground, you turned around, walked straight ahead, and stood in the grass.", + "C": "Walk straight ahead without turning around.", + "D": "After grabbing the bucket, you turned left, walked forward a few steps, and stopped beside the grass.", + "E": "After picking it up, you backed up briefly, then turned around and walked straight until you reached the driveway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 19 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_8", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I manage to walk to the exact spot after picking up the bucket?", + "answer": "After picking up the bucket on the ground, you turned around and walked straight ahead, then turned left. Continue walking straight and you will come to the car.", + "question_chinese": "我拿起桶后是怎么走到刚刚的位置的?", + "answer_chinese": "你在地上拿起桶后转身直行，左转。继续直行，来到了车旁。", + "evidence": [ + { + "description": "After picking up the bucket on the ground, turn around and walk straight ahead, then turn left.", + "timestamps": { + "start": "00:40:14", + "end": "00:40:24" + } + }, + { + "description": "You keep going straight.", + "timestamps": { + "start": "00:40:25", + "end": "00:40:38" + } + }, + { + "description": "You have arrived beside the car.", + "timestamps": { + "start": "00:40:55", + "end": "00:40:57" + } + } + ], + "question_timestamps": "00:41:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "After picking up the bucket on the ground, you turned around and walked straight ahead, then turned left. Continue walking straight and you will come to the car.", + "B": "Turn left immediately, then walk straight.", + "C": "Turn around and go right to the car.", + "D": "After picking up the bucket, you turned around, walked straight ahead, then turned right and continued forward.", + "E": "After you picked it up, you walked straight without turning left, stopping near the grass instead of the car." + }, + "answer_label": "A", + "answer_text": "After picking up the bucket on the ground, you turned around and walked straight ahead, then turned left. Continue walking straight and you will come to the car.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_8", + "correct_answer": "After picking up the bucket on the ground, you turned around and walked straight ahead, then turned left. Continue walking straight and you will come to the car.", + "original_options": { + "A": "After picking up the bucket on the ground, you turned around and walked straight ahead, then turned left. Continue walking straight and you will come to the car.", + "B": "Turn left immediately, then walk straight.", + "C": "Turn around and go right to the car.", + "D": "After picking up the bucket, you turned around, walked straight ahead, then turned right and continued forward.", + "E": "After you picked it up, you walked straight without turning left, stopping near the grass instead of the car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 17, + 19 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_17", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the rake that was blown down by the blower?", + "answer": "The rake is on the ground.", + "question_chinese": "被鼓风机吹倒的耙子现在在哪里？", + "answer_chinese": "耙子在地上。", + "evidence": [ + { + "description": "The blower blew the rake to the ground.", + "timestamps": { + "start": "00:04:32", + "end": "00:04:59" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The rake is leaning on the fence.", + "B": "The rake is by the flower pot.", + "C": "The rake is on the ground.", + "D": "The rake is hanging up on a tree branch, not lying down where it fell.", + "E": "The rake ended up stored in the shed area, rather than being left out after the blower." + }, + "answer_label": "C", + "answer_text": "The rake is on the ground.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_17", + "correct_answer": "The rake is on the ground.", + "original_options": { + "A": "The rake is leaning on the fence.", + "B": "The rake is by the flower pot.", + "C": "The rake is on the ground.", + "D": "The rake is hanging up on a tree branch, not lying down where it fell.", + "E": "The rake ended up stored in the shed area, rather than being left out after the blower." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 17 + ] + }, + { + "video_uid": "9b221f93-19a9-4593-9187-049d9dedfcf0", + "video_length": "00:44:30", + "qid": "9b221f93-19a9-4593-9187-049d9dedfcf0_18", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the rake that was blown down by the blower?", + "answer": "The rake leaned against the wall.", + "question_chinese": "被鼓风机吹倒的耙子现在在哪里？", + "answer_chinese": "耙子靠在墙边。", + "evidence": [ + { + "description": "You leaned the rake against the wall.", + "timestamps": { + "start": "00:14:15", + "end": "00:14:19" + } + } + ], + "question_timestamps": "00:34:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The rake rested on the table.", + "B": "The rake was placed in the bucket.", + "C": "The rake was trapped under the tarp, so it wasn’t left propped up against anything.", + "D": "The rake leaned against the wall.", + "E": "The rake was broken during the storm and left elsewhere, not leaning against the wall." + }, + "answer_label": "D", + "answer_text": "The rake leaned against the wall.", + "q_id": "9b221f93-19a9-4593-9187-049d9dedfcf0_18", + "correct_answer": "The rake leaned against the wall.", + "original_options": { + "A": "The rake rested on the table.", + "B": "The rake was placed in the bucket.", + "C": "The rake was trapped under the tarp, so it wasn’t left propped up against anything.", + "D": "The rake leaned against the wall.", + "E": "The rake was broken during the storm and left elsewhere, not leaning against the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 15 + ] + } +] diff --git a/QAs/9b49246f-30e9-476f-ab8c-56a1bcb1d936-mcq.json b/QAs/9b49246f-30e9-476f-ab8c-56a1bcb1d936-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4f412a8a15ae9df87d90d9cf17a089a0fe74933c --- /dev/null +++ b/QAs/9b49246f-30e9-476f-ab8c-56a1bcb1d936-mcq.json @@ -0,0 +1,1227 @@ +[ + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_4_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to get more tableware, where should I go look?", + "answer": "You should look in the cabinet on your right-hand side.", + "question_chinese": "如果我想获取更多的餐具应该去哪找呢？", + "answer_chinese": "你应该去右手边的碗柜里找。", + "evidence": [ + { + "description": "There are many tableware items in the cupboard located at the bottom right of the sink.", + "timestamps": { + "start": "00:01:31", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:01:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Look in the cabinet on your left.", + "B": "You should look in the cabinet on your right-hand side.", + "C": "Look in the cabinet behind you.", + "D": "You should check the cabinet directly in front of you for more tableware.", + "E": "Go to the cabinet on your left-hand side, not the one positioned on your right." + }, + "answer_label": "B", + "answer_text": "You should look in the cabinet on your right-hand side.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_4_T1", + "correct_answer": "You should look in the cabinet on your right-hand side.", + "original_options": { + "A": "Look in the cabinet on your left.", + "B": "You should look in the cabinet on your right-hand side.", + "C": "Look in the cabinet behind you.", + "D": "You should check the cabinet directly in front of you for more tableware.", + "E": "Go to the cabinet on your left-hand side, not the one positioned on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 15 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_4_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to get more tableware, where should I go look?", + "answer": "You should first go downstairs to the kitchen and look in the cabinet that is down and to the right of the sink.", + "question_chinese": "如果我想获取更多的餐具应该去哪找呢？", + "answer_chinese": "你应该先下楼到厨房里面水槽右下方的碗柜里找。", + "evidence": [ + { + "description": "There are many tableware items in the cupboard located at the bottom right of the sink.", + "timestamps": { + "start": "00:01:31", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:35:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go upstairs and check the cabinet.", + "B": "Look in the cabinet left of the sink.", + "C": "You should go downstairs to the kitchen and check the cabinet below the sink, but on the left side.", + "D": "You should first go downstairs to the kitchen and look in the cabinet that is down and to the right of the sink.", + "E": "First head downstairs to the kitchen, then look in the cabinet below the sink that is centered in front of you." + }, + "answer_label": "D", + "answer_text": "You should first go downstairs to the kitchen and look in the cabinet that is down and to the right of the sink.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_4_T2", + "correct_answer": "You should first go downstairs to the kitchen and look in the cabinet that is down and to the right of the sink.", + "original_options": { + "A": "Go upstairs and check the cabinet.", + "B": "Look in the cabinet left of the sink.", + "C": "You should go downstairs to the kitchen and check the cabinet below the sink, but on the left side.", + "D": "You should first go downstairs to the kitchen and look in the cabinet that is down and to the right of the sink.", + "E": "First head downstairs to the kitchen, then look in the cabinet below the sink that is centered in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 19, + 21 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_13", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand in front of the sink, how far is it from the green basket?", + "answer": "You are about 50cm away from the green basket, which is in front of you on your right.", + "question_chinese": "当我站在水槽前时,离绿色篮子有多远?", + "answer_chinese": "你离绿色篮子大约50cm,在你的右前方。", + "evidence": [ + { + "description": "The green basket is on the edge of the workbench, about 50cm away from you.", + "timestamps": { + "start": "00:11:32", + "end": "00:11:33" + } + } + ], + "question_timestamps": "00:11:38", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "About 30cm away, behind you on the left.", + "B": "About 90cm away, beside you on the right.", + "C": "You are about 70cm away from the green basket, and it’s in front of you on your left.", + "D": "You are about 50cm away from the green basket, which is in front of you on your right.", + "E": "You are about 25cm from the green basket, positioned to your left and slightly above your line of sight." + }, + "answer_label": "D", + "answer_text": "You are about 50cm away from the green basket, which is in front of you on your right.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_13", + "correct_answer": "You are about 50cm away from the green basket, which is in front of you on your right.", + "original_options": { + "A": "About 30cm away, behind you on the left.", + "B": "About 90cm away, beside you on the right.", + "C": "You are about 70cm away from the green basket, and it’s in front of you on your left.", + "D": "You are about 50cm away from the green basket, which is in front of you on your right.", + "E": "You are about 25cm from the green basket, positioned to your left and slightly above your line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 19 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_14", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand in front of the sink, how far is it from the green basket?", + "answer": "You are about two meters away from the green basket. It's behind you.", + "question_chinese": "当我站在水槽前时,离绿色篮子有多远?", + "answer_chinese": "你离绿色篮子大约2m。在你的后面。", + "evidence": [ + { + "description": "The green basket is in the refrigerator, about 2 meters away from you.", + "timestamps": { + "start": "00:14:14", + "end": "00:14:21" + } + } + ], + "question_timestamps": "00:30:21", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "About one meter away on your left.", + "B": "About three meters away in front of you.", + "C": "You are about two meters away from the green basket. It's behind you.", + "D": "You are about four meters away from the green basket, and it’s above you rather than behind you.", + "E": "You are about five meters away from the green basket, and it’s off to your right instead of behind you." + }, + "answer_label": "C", + "answer_text": "You are about two meters away from the green basket. It's behind you.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_14", + "correct_answer": "You are about two meters away from the green basket. It's behind you.", + "original_options": { + "A": "About one meter away on your left.", + "B": "About three meters away in front of you.", + "C": "You are about two meters away from the green basket. It's behind you.", + "D": "You are about four meters away from the green basket, and it’s above you rather than behind you.", + "E": "You are about five meters away from the green basket, and it’s off to your right instead of behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 21 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many apricots in plastic boxes have I processed in total?", + "answer": "You processed apricots from two plastic boxes in total.", + "question_chinese": "我一共处理了多少个塑料盒里的杏子？", + "answer_chinese": "你一共处理了两个塑料盒里的杏子。", + "evidence": [ + { + "description": "You processed the apricots in the first plastic box.", + "timestamps": { + "start": "00:01:37", + "end": "00:02:00" + } + }, + { + "description": "You processed the apricots in the second plastic box.", + "timestamps": { + "start": "00:02:09", + "end": "00:02:23" + } + } + ], + "question_timestamps": "00:02:27", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You processed apricots from one plastic box.", + "B": "You processed apricots from three plastic boxes.", + "C": "You processed apricots from two plastic boxes in total.", + "D": "You processed apricots from four plastic boxes in total, finishing each box before moving to the next.", + "E": "You processed apricots from five plastic boxes overall, emptying multiple containers during the full sorting process." + }, + "answer_label": "C", + "answer_text": "You processed apricots from two plastic boxes in total.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_9", + "correct_answer": "You processed apricots from two plastic boxes in total.", + "original_options": { + "A": "You processed apricots from one plastic box.", + "B": "You processed apricots from three plastic boxes.", + "C": "You processed apricots from two plastic boxes in total.", + "D": "You processed apricots from four plastic boxes in total, finishing each box before moving to the next.", + "E": "You processed apricots from five plastic boxes overall, emptying multiple containers during the full sorting process." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_10", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many apricots in plastic boxes have I processed in total?", + "answer": "You processed apricots from three plastic boxes in total.", + "question_chinese": "我一共处理了多少个塑料盒里的杏子？", + "answer_chinese": "你一共处理了三个塑料盒里的杏子。", + "evidence": [ + { + "description": "You processed the apricots in the first plastic box.", + "timestamps": { + "start": "00:01:37", + "end": "00:02:00" + } + }, + { + "description": "You processed the apricots in the second plastic box.", + "timestamps": { + "start": "00:02:09", + "end": "00:02:23" + } + }, + { + "description": "You processed the apricots in the third plastic box.", + "timestamps": { + "start": "00:10:38", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:40:50", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You processed apricots from two plastic boxes.", + "B": "You processed apricots from three plastic boxes in total.", + "C": "You processed apricots from four plastic boxes.", + "D": "You processed apricots from five plastic boxes in total, working through several containers one after another.", + "E": "You processed apricots from six plastic boxes overall, completing additional boxes beyond the initial few you started." + }, + "answer_label": "B", + "answer_text": "You processed apricots from three plastic boxes in total.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_10", + "correct_answer": "You processed apricots from three plastic boxes in total.", + "original_options": { + "A": "You processed apricots from two plastic boxes.", + "B": "You processed apricots from three plastic boxes in total.", + "C": "You processed apricots from four plastic boxes.", + "D": "You processed apricots from five plastic boxes in total, working through several containers one after another.", + "E": "You processed apricots from six plastic boxes overall, completing additional boxes beyond the initial few you started." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_19", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I placed the orange box on the edge of the table earlier. How far is it from the headphones now?", + "answer": "The orange box is approximately 1cm away from the earphones.", + "question_chinese": "我之前把橙色盒子放在桌子边缘了,它现在离耳机有多远?", + "answer_chinese": "橙色盒子离耳机大约1cm。", + "evidence": [ + { + "description": "The orange box is on the edge of the table, about 1cm away from the headphones.", + "timestamps": { + "start": "00:20:49", + "end": "00:20:53" + } + } + ], + "question_timestamps": "00:20:55", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "About 5cm away from the headphones.", + "B": "Roughly 10cm away from the headphones.", + "C": "The orange box is approximately 15cm away from the headphones, leaving a noticeable gap between them.", + "D": "The orange box is approximately 20cm away from the headphones, not nearly close enough to be almost touching.", + "E": "The orange box is approximately 1cm away from the earphones." + }, + "answer_label": "E", + "answer_text": "The orange box is approximately 1cm away from the earphones.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_19", + "correct_answer": "The orange box is approximately 1cm away from the earphones.", + "original_options": { + "A": "About 5cm away from the headphones.", + "B": "Roughly 10cm away from the headphones.", + "C": "The orange box is approximately 15cm away from the headphones, leaving a noticeable gap between them.", + "D": "The orange box is approximately 20cm away from the headphones, not nearly close enough to be almost touching.", + "E": "The orange box is approximately 1cm away from the earphones." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_20", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I placed the orange box on the edge of the table earlier. How far is it from the headphones now?", + "answer": "The orange box is approximately 25cm away from the earphones.", + "question_chinese": "我之前把橙色盒子放在桌子边缘了,它现在离耳机有多远?", + "answer_chinese": "橙子盒子离耳机大约25cm。", + "evidence": [ + { + "description": "The Orange Box is on the laptop, approximately 25cm away from the headphones.", + "timestamps": { + "start": "00:23:30", + "end": "00:23:33" + } + } + ], + "question_timestamps": "00:45:35", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "About 10cm away from the earphones.", + "B": "The orange box is approximately 25cm away from the earphones.", + "C": "About 30cm away from the earphones.", + "D": "The orange box is approximately 5cm away from the earphones, positioned much closer than a handspan.", + "E": "The orange box is approximately 50cm away from the earphones, roughly half a meter apart on the table." + }, + "answer_label": "B", + "answer_text": "The orange box is approximately 25cm away from the earphones.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_20", + "correct_answer": "The orange box is approximately 25cm away from the earphones.", + "original_options": { + "A": "About 10cm away from the earphones.", + "B": "The orange box is approximately 25cm away from the earphones.", + "C": "About 30cm away from the earphones.", + "D": "The orange box is approximately 5cm away from the earphones, positioned much closer than a handspan.", + "E": "The orange box is approximately 50cm away from the earphones, roughly half a meter apart on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_3", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following just now?", + "answer": "You was leaving from the bedroom, walking straight through the hallway, and heading downstairs towards the kitchen.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你正从卧室出发,直行穿过走廊,下楼朝着厨房方向移动。", + "evidence": [ + { + "description": "You start from the bedroom and proceed straight through the hallway.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:11" + } + }, + { + "description": "You are going downstairs.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:32" + } + }, + { + "description": "You have arrived in the kitchen.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:00:44", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "From bedroom through hallway to living room.", + "B": "From bedroom down the hall to bathroom.", + "C": "You was leaving from the bedroom, turning left in the hallway, and heading downstairs toward the living room.", + "D": "You was leaving from the bedroom, walking straight through the hallway, and heading downstairs towards the kitchen.", + "E": "You was leaving from the bedroom, walking through the hallway, then going upstairs instead of downstairs toward the kitchen." + }, + "answer_label": "D", + "answer_text": "You was leaving from the bedroom, walking straight through the hallway, and heading downstairs towards the kitchen.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_3", + "correct_answer": "You was leaving from the bedroom, walking straight through the hallway, and heading downstairs towards the kitchen.", + "original_options": { + "A": "From bedroom through hallway to living room.", + "B": "From bedroom down the hall to bathroom.", + "C": "You was leaving from the bedroom, turning left in the hallway, and heading downstairs toward the living room.", + "D": "You was leaving from the bedroom, walking straight through the hallway, and heading downstairs towards the kitchen.", + "E": "You was leaving from the bedroom, walking through the hallway, then going upstairs instead of downstairs toward the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 19 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_4", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route was I following just now?", + "answer": "You was leaving your home, turning left and walking straight through the hallway, heading towards the garbage room.", + "question_chinese": "我刚刚在走什么路线？", + "answer_chinese": "你正从家里出发，左转直行穿过走廊，朝着垃圾房方向移动。", + "evidence": [ + { + "description": "You open the door of your home.", + "timestamps": { + "start": "00:15:10", + "end": "00:15:16" + } + }, + { + "description": "Turn left and proceed straight along the hallway.", + "timestamps": { + "start": "00:15:17", + "end": "00:15:46" + } + }, + { + "description": "You came to the garbage room to throw away the garbage.", + "timestamps": { + "start": "00:15:49", + "end": "00:16:01" + } + } + ], + "question_timestamps": "00:16:06", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turned right toward the garbage room.", + "B": "You headed toward the laundry room.", + "C": "You was leaving your home, turning left and walking straight through the hallway, heading towards the garbage room.", + "D": "You was leaving your home, turning left and walking straight through the hallway, heading towards the storage room.", + "E": "You was leaving your home, turning right and walking straight through the hallway, heading towards the bathroom instead." + }, + "answer_label": "C", + "answer_text": "You was leaving your home, turning left and walking straight through the hallway, heading towards the garbage room.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_4", + "correct_answer": "You was leaving your home, turning left and walking straight through the hallway, heading towards the garbage room.", + "original_options": { + "A": "You turned right toward the garbage room.", + "B": "You headed toward the laundry room.", + "C": "You was leaving your home, turning left and walking straight through the hallway, heading towards the garbage room.", + "D": "You was leaving your home, turning left and walking straight through the hallway, heading towards the storage room.", + "E": "You was leaving your home, turning right and walking straight through the hallway, heading towards the bathroom instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 18 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_8_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of grapes did I just sort out?", + "answer": "You organized one box of grapes on the shelf inside the fridge, and the fridge is in front of you.", + "question_chinese": "我刚刚整理多少盒葡萄？", + "answer_chinese": "你整理了一盒放在冰箱里架子上的葡萄，冰箱在你的前面。", + "evidence": [ + { + "description": "You placed the first box of grapes from your front to your right.", + "timestamps": { + "start": "00:13:55", + "end": "00:14:01" + } + } + ], + "question_timestamps": "00:14:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You organized one box of grapes on the shelf inside the fridge, and the fridge is in front of you.", + "B": "You organized zero boxes of grapes.", + "C": "You organized two boxes of grapes.", + "D": "You organized three boxes of grapes on the shelf inside the fridge, and the fridge is in front of you.", + "E": "You organized four boxes of grapes on the shelf inside the fridge, and the fridge is in front of you." + }, + "answer_label": "A", + "answer_text": "You organized one box of grapes on the shelf inside the fridge, and the fridge is in front of you.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_8_T1", + "correct_answer": "You organized one box of grapes on the shelf inside the fridge, and the fridge is in front of you.", + "original_options": { + "A": "You organized one box of grapes on the shelf inside the fridge, and the fridge is in front of you.", + "B": "You organized zero boxes of grapes.", + "C": "You organized two boxes of grapes.", + "D": "You organized three boxes of grapes on the shelf inside the fridge, and the fridge is in front of you.", + "E": "You organized four boxes of grapes on the shelf inside the fridge, and the fridge is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 20 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_15", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of grapes have I just sorted out?", + "answer": "You sorted two boxes of grapes.", + "question_chinese": "我刚刚整理多少盒葡萄？", + "answer_chinese": "你整理了两盒葡萄。", + "evidence": [ + { + "description": "You placed the first box of grapes from your front to your right.", + "timestamps": { + "start": "00:13:55", + "end": "00:14:02" + } + }, + { + "description": "You placed the second box of grapes from your front to the right side.", + "timestamps": { + "start": "00:14:04", + "end": "00:14:06" + } + } + ], + "question_timestamps": "00:14:11", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You sorted one box of grapes.", + "B": "You sorted three boxes of grapes.", + "C": "You sorted four boxes of grapes, going beyond the two boxes you were supposed to organize.", + "D": "You sorted two boxes of grapes.", + "E": "You sorted five boxes of grapes, handling several more boxes than just a couple during the task." + }, + "answer_label": "D", + "answer_text": "You sorted two boxes of grapes.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_15", + "correct_answer": "You sorted two boxes of grapes.", + "original_options": { + "A": "You sorted one box of grapes.", + "B": "You sorted three boxes of grapes.", + "C": "You sorted four boxes of grapes, going beyond the two boxes you were supposed to organize.", + "D": "You sorted two boxes of grapes.", + "E": "You sorted five boxes of grapes, handling several more boxes than just a couple during the task." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_8_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of grapes did I just sort out?", + "answer": "You organized two boxes of grapes on the shelf inside the fridge, and the fridge is to your back-left.", + "question_chinese": "我刚刚整理多少盒葡萄？", + "answer_chinese": "你整理了两盒冰箱里架子上的葡萄，冰箱在你的左后方。", + "evidence": [ + { + "description": "You placed the first box of grapes from your front to your right.", + "timestamps": { + "start": "00:13:55", + "end": "00:14:02" + } + }, + { + "description": "You placed the second box of grapes from your front to the right side.", + "timestamps": { + "start": "00:14:04", + "end": "00:14:06" + } + } + ], + "question_timestamps": "00:17:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You organized one box of grapes.", + "B": "You organized two boxes of grapes on the shelf inside the fridge, and the fridge is to your back-left.", + "C": "You organized three boxes of grapes.", + "D": "You organized four boxes of grapes on the shelf inside the fridge, with the fridge behind you to the left.", + "E": "You organized zero boxes of grapes on the shelf inside the fridge, and the fridge was behind you on the left." + }, + "answer_label": "B", + "answer_text": "You organized two boxes of grapes on the shelf inside the fridge, and the fridge is to your back-left.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_8_T2", + "correct_answer": "You organized two boxes of grapes on the shelf inside the fridge, and the fridge is to your back-left.", + "original_options": { + "A": "You organized one box of grapes.", + "B": "You organized two boxes of grapes on the shelf inside the fridge, and the fridge is to your back-left.", + "C": "You organized three boxes of grapes.", + "D": "You organized four boxes of grapes on the shelf inside the fridge, with the fridge behind you to the left.", + "E": "You organized zero boxes of grapes on the shelf inside the fridge, and the fridge was behind you on the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 21 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_16", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes of grapes have I just sorted out?", + "answer": "You sorted two boxes of grapes.", + "question_chinese": "我刚刚整理多少盒葡萄？", + "answer_chinese": "你整理了两盒葡萄。", + "evidence": [ + { + "description": "You placed the first box of grapes from your front to your right.", + "timestamps": { + "start": "00:13:55", + "end": "00:14:02" + } + }, + { + "description": "You placed the second box of grapes from your front to the right side.", + "timestamps": { + "start": "00:14:04", + "end": "00:14:06" + } + } + ], + "question_timestamps": "00:55:55", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You sorted two boxes of grapes.", + "B": "You sorted one box of grapes.", + "C": "You sorted three boxes of grapes.", + "D": "You sorted four boxes of grapes after taking them out, then set them aside in front of you.", + "E": "You sorted five boxes of grapes, separating them into groups before putting them away." + }, + "answer_label": "A", + "answer_text": "You sorted two boxes of grapes.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_16", + "correct_answer": "You sorted two boxes of grapes.", + "original_options": { + "A": "You sorted two boxes of grapes.", + "B": "You sorted one box of grapes.", + "C": "You sorted three boxes of grapes.", + "D": "You sorted four boxes of grapes after taking them out, then set them aside in front of you.", + "E": "You sorted five boxes of grapes, separating them into groups before putting them away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 14 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_1_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was just at the computer desk, was the glass cup on the table to my left or to my right?", + "answer": "To your right.", + "question_chinese": "我刚在电脑桌前的时候,桌子上的玻璃杯子在你的左边还是右边?", + "answer_chinese": "右边。", + "evidence": [ + { + "description": "The glass on the table is on your right.", + "timestamps": { + "start": "00:19:45", + "end": "00:19:50" + } + } + ], + "question_timestamps": "00:19:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "To your right.", + "B": "To your left." + }, + "answer_label": "A", + "answer_text": "To your right.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_1_T1", + "correct_answer": "To your right.", + "original_options": { + "A": "To your right.", + "B": "To your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_1_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was just at the computer desk, was the glass cup on the table to my left or to my right?", + "answer": "The glass cup on the table is to your left.", + "question_chinese": "我刚在电脑桌前的时候,桌子上的玻璃杯子在你的左边还是右边?", + "answer_chinese": "桌子上的玻璃杯子在你的左边。", + "evidence": [ + { + "description": "When you were just at your computer desk, there was a glass cup on the desk.", + "timestamps": { + "start": "00:23:51", + "end": "00:23:53" + } + } + ], + "question_timestamps": "00:42:58", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The glass cup on the table is to your left.", + "B": "To your right." + }, + "answer_label": "A", + "answer_text": "The glass cup on the table is to your left.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_1_T2", + "correct_answer": "The glass cup on the table is to your left.", + "original_options": { + "A": "The glass cup on the table is to your left.", + "B": "To your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_3_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was in the kitchen, how many boxes of apricots did I take out of the refrigerator?", + "answer": "You took out one box of apricots.", + "question_chinese": "我在厨房的时候，从冰箱拿出了几盒杏子？", + "answer_chinese": "你拿出了一盒杏子。", + "evidence": [ + { + "description": "You took out a box of apricots from the refrigerator.", + "timestamps": { + "start": "00:00:59", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took out zero boxes of apricots.", + "B": "You took out one box of apricots.", + "C": "You took out two boxes of apricots.", + "D": "You took out three boxes of apricots from the refrigerator while you were in the kitchen.", + "E": "You took out four boxes of apricots from the refrigerator and held them briefly before setting them down." + }, + "answer_label": "B", + "answer_text": "You took out one box of apricots.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_3_T2", + "correct_answer": "You took out one box of apricots.", + "original_options": { + "A": "You took out zero boxes of apricots.", + "B": "You took out one box of apricots.", + "C": "You took out two boxes of apricots.", + "D": "You took out three boxes of apricots from the refrigerator while you were in the kitchen.", + "E": "You took out four boxes of apricots from the refrigerator and held them briefly before setting them down." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_3_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was in the kitchen, how many boxes of apricots did I take out of the refrigerator?", + "answer": "You took out two boxes of apricots.", + "question_chinese": "我在厨房的时候，从冰箱拿出了几盒杏子？", + "answer_chinese": "你拿出了两盒杏子。", + "evidence": [ + { + "description": "You took out a box of apricots from the refrigerator.", + "timestamps": { + "start": "00:00:59", + "end": "00:01:05" + } + }, + { + "description": "You took out the second box of apricots from the refrigerator.", + "timestamps": { + "start": "00:01:08", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:30:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took out two boxes of apricots.", + "B": "You took out one box of apricots.", + "C": "You took out three boxes of apricots.", + "D": "You took out zero boxes of apricots from the refrigerator while you were in the kitchen.", + "E": "You took out four boxes of apricots from the refrigerator and moved them onto the counter." + }, + "answer_label": "A", + "answer_text": "You took out two boxes of apricots.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_3_T1", + "correct_answer": "You took out two boxes of apricots.", + "original_options": { + "A": "You took out two boxes of apricots.", + "B": "You took out one box of apricots.", + "C": "You took out three boxes of apricots.", + "D": "You took out zero boxes of apricots from the refrigerator while you were in the kitchen.", + "E": "You took out four boxes of apricots from the refrigerator and moved them onto the counter." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_17", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You was holding a bag of clothes in your hand now.", + "question_chinese": "我手里刚刚拿着什么东西？", + "answer_chinese": "你手里现在拿着一袋衣服。", + "evidence": [ + { + "description": "You put the clothes from the washing machine into a bag and then held them in your hand.", + "timestamps": { + "start": "00:16:45", + "end": "00:17:01" + } + } + ], + "question_timestamps": "00:17:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You was holding a bag of clothes in your hand now.", + "B": "You was holding a cup of water.", + "C": "You was holding a jar of cookies.", + "D": "You was holding a book and a pen in your hand now, as if you were about to write something down.", + "E": "You was holding a phone and a charger in your hand now, carrying them together in one grip." + }, + "answer_label": "A", + "answer_text": "You was holding a bag of clothes in your hand now.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_17", + "correct_answer": "You was holding a bag of clothes in your hand now.", + "original_options": { + "A": "You was holding a bag of clothes in your hand now.", + "B": "You was holding a cup of water.", + "C": "You was holding a jar of cookies.", + "D": "You was holding a book and a pen in your hand now, as if you were about to write something down.", + "E": "You was holding a phone and a charger in your hand now, carrying them together in one grip." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 18 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_18", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You was holding a fork in your hand.", + "question_chinese": "我手里刚刚拿着什么东西？", + "answer_chinese": "你手里拿着一把叉子。", + "evidence": [ + { + "description": "You are holding a fork in your right hand.", + "timestamps": { + "start": "00:25:44", + "end": "00:25:45" + } + } + ], + "question_timestamps": "00:25:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You was holding a spoon in your hand.", + "B": "You was holding a napkin in your hand.", + "C": "You was holding a fork in your hand.", + "D": "You was holding a chopstick in your hand, pinched between your fingers as if ready to eat.", + "E": "You was holding a plate in your hand, balancing it with your palm while moving it." + }, + "answer_label": "C", + "answer_text": "You was holding a fork in your hand.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_18", + "correct_answer": "You was holding a fork in your hand.", + "original_options": { + "A": "You was holding a spoon in your hand.", + "B": "You was holding a napkin in your hand.", + "C": "You was holding a fork in your hand.", + "D": "You was holding a chopstick in your hand, pinched between your fingers as if ready to eat.", + "E": "You was holding a plate in your hand, balancing it with your palm while moving it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 16 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_6_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bunches of vegetables have I picked now?", + "answer": "So far, you have picked one bunch of vegetables.", + "question_chinese": "我现在摘了几把蔬菜？", + "answer_chinese": "你现在摘了一把蔬菜。", + "evidence": [ + { + "description": "I just picked a bunch of vegetables.", + "timestamps": { + "start": "00:11:40", + "end": "00:12:20" + } + } + ], + "question_timestamps": "00:12:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have picked zero bunches.", + "B": "You have picked two bunches.", + "C": "So far, you have picked three bunches of vegetables in total now.", + "D": "Up to now, you have picked four bunches of vegetables altogether today.", + "E": "So far, you have picked one bunch of vegetables." + }, + "answer_label": "E", + "answer_text": "So far, you have picked one bunch of vegetables.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_6_T2", + "correct_answer": "So far, you have picked one bunch of vegetables.", + "original_options": { + "A": "You have picked zero bunches.", + "B": "You have picked two bunches.", + "C": "So far, you have picked three bunches of vegetables in total now.", + "D": "Up to now, you have picked four bunches of vegetables altogether today.", + "E": "So far, you have picked one bunch of vegetables." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936", + "video_length": "00:59:13", + "qid": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_6_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bunches of vegetables have I picked now?", + "answer": "So far, you have picked two bunches of vegetables.", + "question_chinese": "我现在摘了几把蔬菜？", + "answer_chinese": "你现在摘了两把蔬菜。", + "evidence": [ + { + "description": "I just picked two bunches of vegetables.", + "timestamps": { + "start": "00:12:20", + "end": "00:13:23" + } + } + ], + "question_timestamps": "00:23:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have picked one bunch.", + "B": "You have picked three bunches.", + "C": "So far, you have picked zero bunches of vegetables in total yet.", + "D": "So far, you have picked two bunches of vegetables.", + "E": "Up to now, you have picked four bunches of vegetables altogether today." + }, + "answer_label": "D", + "answer_text": "So far, you have picked two bunches of vegetables.", + "q_id": "9b49246f-30e9-476f-ab8c-56a1bcb1d936_6_T1", + "correct_answer": "So far, you have picked two bunches of vegetables.", + "original_options": { + "A": "You have picked one bunch.", + "B": "You have picked three bunches.", + "C": "So far, you have picked zero bunches of vegetables in total yet.", + "D": "So far, you have picked two bunches of vegetables.", + "E": "Up to now, you have picked four bunches of vegetables altogether today." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + } +] diff --git a/QAs/9c07f3a2-5033-4f78-a377-a4cf5a7d7add-mcq.json b/QAs/9c07f3a2-5033-4f78-a377-a4cf5a7d7add-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e9cc9ddf004bea1f14d5fdf7865099e280ef72bc --- /dev/null +++ b/QAs/9c07f3a2-5033-4f78-a377-a4cf5a7d7add-mcq.json @@ -0,0 +1,757 @@ +[ + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_9", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many wooden boards have I fixed to the iron frame?", + "answer": "So far, you have fixed two pieces of wood.", + "question_chinese": "到目前为止，我固定了几块木板到铁框架上？", + "answer_chinese": "到目前为止，你固定了两块木板。", + "evidence": [ + { + "description": "The first and second boards", + "timestamps": { + "start": "00:17:43", + "end": "00:18:27" + } + } + ], + "question_timestamps": "00:22:42", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have fixed one board.", + "B": "You have fixed three boards.", + "C": "So far, you have fixed four pieces of wood to the iron frame.", + "D": "So far, you have fixed two pieces of wood.", + "E": "Up to now, you have fixed five wooden boards onto the iron frame." + }, + "answer_label": "D", + "answer_text": "So far, you have fixed two pieces of wood.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_9", + "correct_answer": "So far, you have fixed two pieces of wood.", + "original_options": { + "A": "You have fixed one board.", + "B": "You have fixed three boards.", + "C": "So far, you have fixed four pieces of wood to the iron frame.", + "D": "So far, you have fixed two pieces of wood.", + "E": "Up to now, you have fixed five wooden boards onto the iron frame." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_10", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many wooden boards have I fixed to the iron frame?", + "answer": "So far, you have fixed three planks.", + "question_chinese": "到目前为止，我固定了几块木板到铁框架上？", + "answer_chinese": "到目前为止，你已经固定了三个木板。", + "evidence": [ + { + "description": "The first and second boards", + "timestamps": { + "start": "00:17:43", + "end": "00:18:28" + } + } + ], + "question_timestamps": "00:33:42", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have fixed two planks.", + "B": "You have fixed four planks.", + "C": "So far, you have fixed five planks to the iron frame already.", + "D": "Up to now, you have fixed six wooden planks onto the iron frame.", + "E": "So far, you have fixed three planks." + }, + "answer_label": "E", + "answer_text": "So far, you have fixed three planks.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_10", + "correct_answer": "So far, you have fixed three planks.", + "original_options": { + "A": "You have fixed two planks.", + "B": "You have fixed four planks.", + "C": "So far, you have fixed five planks to the iron frame already.", + "D": "Up to now, you have fixed six wooden planks onto the iron frame.", + "E": "So far, you have fixed three planks." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_11", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the tape measure or the carpenter's clamp?", + "answer": "The tape measure is closer to you.", + "question_chinese": "卷尺和木工夹哪个离我更近？", + "answer_chinese": "卷尺离你更近。", + "evidence": [ + { + "description": "The tape measure is fixed on the metal frame.", + "timestamps": { + "start": "00:09:35", + "end": "00:09:53" + } + }, + { + "description": "The tape measure is on the cardboard on the ground next to the pile of wooden boards", + "timestamps": { + "start": "00:09:54", + "end": "00:10:02" + } + } + ], + "question_timestamps": "00:10:56", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The tape measure is closer to you.", + "B": "The carpenter's clamp is closer to you." + }, + "answer_label": "A", + "answer_text": "The tape measure is closer to you.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_11", + "correct_answer": "The tape measure is closer to you.", + "original_options": { + "A": "The tape measure is closer to you.", + "B": "The carpenter's clamp is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_12", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the tape measure or the carpenter's clamp?", + "answer": "The tape measure is closer to you.", + "question_chinese": "卷尺和木工夹哪个离我更近？", + "answer_chinese": "卷尺离你更近。", + "evidence": [ + { + "description": "The woodworking clamp is on the metal frame.", + "timestamps": { + "start": "00:18:50", + "end": "00:19:06" + } + }, + { + "description": "The tape measure is on the cardboard on the ground.", + "timestamps": { + "start": "00:21:42", + "end": "00:22:16" + } + } + ], + "question_timestamps": "00:23:04", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The tape measure is closer to you.", + "B": "The carpenter's clamp is closer to you." + }, + "answer_label": "A", + "answer_text": "The tape measure is closer to you.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_12", + "correct_answer": "The tape measure is closer to you.", + "original_options": { + "A": "The tape measure is closer to you.", + "B": "The carpenter's clamp is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_7", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of things did I take from the car?", + "answer": "You took a set square, a chainsaw, a power drill, a red plastic box, a metal corrugated hose, a woodworking clamp, a nail box, and a plastic bag.", + "question_chinese": "我从车上拿了哪几种东西？", + "answer_chinese": "你拿了三角板、电锯、电钻、红色塑料盒、金属波纹软管、木工夹、铁钉盒以及塑料袋。", + "evidence": [ + { + "description": "Set square and chainsaw", + "timestamps": { + "start": "00:04:39", + "end": "00:04:52" + } + }, + { + "description": "Metal corrugated hoses and carpentry clamps", + "timestamps": { + "start": "00:07:48", + "end": "00:08:10" + } + }, + { + "description": "Nail packaging box, red plastic box and electric drill", + "timestamps": { + "start": "00:10:46", + "end": "00:11:08" + } + } + ], + "question_timestamps": "00:11:58", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You took a tape measure and hand saw.", + "B": "You took a hammer and circular saw.", + "C": "You took a carpenter's square, an electric chainsaw, a power screwdriver, and a clear plastic bin with a box of nails.", + "D": "You took a set square, a chainsaw, a power drill, a red plastic box, a metal corrugated hose, a woodworking clamp, a nail box, and a plastic bag.", + "E": "You took a bevel square, a reciprocating saw, a drill driver, and a yellow plastic crate with a container of screws." + }, + "answer_label": "D", + "answer_text": "You took a set square, a chainsaw, a power drill, a red plastic box, a metal corrugated hose, a woodworking clamp, a nail box, and a plastic bag.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_7", + "correct_answer": "You took a set square, a chainsaw, a power drill, a red plastic box, a metal corrugated hose, a woodworking clamp, a nail box, and a plastic bag.", + "original_options": { + "A": "You took a tape measure and hand saw.", + "B": "You took a hammer and circular saw.", + "C": "You took a carpenter's square, an electric chainsaw, a power screwdriver, and a clear plastic bin with a box of nails.", + "D": "You took a set square, a chainsaw, a power drill, a red plastic box, a metal corrugated hose, a woodworking clamp, a nail box, and a plastic bag.", + "E": "You took a bevel square, a reciprocating saw, a drill driver, and a yellow plastic crate with a container of screws." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 21, + 21 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_8", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kinds of things did I take from the car?", + "answer": "You took a set square, chainsaw, power drill, red plastic box, metal corrugated hose, woodworking clamp, nail box, plastic bag, yellow plastic box, angle impact driver, and lithium-ion impact drill.", + "question_chinese": "我从车上拿了哪几种东西？", + "answer_chinese": "你拿了三角板、电锯、电钻、红色塑料盒、金属波纹软管、木工夹、铁钉盒、塑料袋、黄色塑料盒、角向冲击起子、锂电冲击钻\r。", + "evidence": [ + { + "description": "Set square and chainsaw", + "timestamps": { + "start": "00:04:39", + "end": "00:04:52" + } + }, + { + "description": "Metal corrugated hoses and woodworking clamp", + "timestamps": { + "start": "00:07:48", + "end": "00:08:10" + } + }, + { + "description": "Yellow plastic box, angle impact driver, lithium-ion impact drill", + "timestamps": { + "start": "00:21:07", + "end": "00:21:32" + } + }, + { + "description": "Electric drill, red plastic box, nail box, and plastic bag", + "timestamps": { + "start": "00:10:46", + "end": "00:11:08" + } + } + ], + "question_timestamps": "00:21:42", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You took a set square, chainsaw, power drill, red plastic box, metal corrugated hose, woodworking clamp, nail box, plastic bag, yellow plastic box, angle impact driver, and lithium-ion impact drill.", + "B": "You took a tape measure and circular saw.", + "C": "You took a jigsaw and impact wrench.", + "D": "You took a carpenter's square, reciprocating saw, cordless drill, and a yellow plastic crate with a red tool case.", + "E": "You retrieved a framing square, pruning saw, electric hammer drill, and a green plastic bin with a white tool box." + }, + "answer_label": "A", + "answer_text": "You took a set square, chainsaw, power drill, red plastic box, metal corrugated hose, woodworking clamp, nail box, plastic bag, yellow plastic box, angle impact driver, and lithium-ion impact drill.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_8", + "correct_answer": "You took a set square, chainsaw, power drill, red plastic box, metal corrugated hose, woodworking clamp, nail box, plastic bag, yellow plastic box, angle impact driver, and lithium-ion impact drill.", + "original_options": { + "A": "You took a set square, chainsaw, power drill, red plastic box, metal corrugated hose, woodworking clamp, nail box, plastic bag, yellow plastic box, angle impact driver, and lithium-ion impact drill.", + "B": "You took a tape measure and circular saw.", + "C": "You took a jigsaw and impact wrench.", + "D": "You took a carpenter's square, reciprocating saw, cordless drill, and a yellow plastic crate with a red tool case.", + "E": "You retrieved a framing square, pruning saw, electric hammer drill, and a green plastic bin with a white tool box." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 20 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_4", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the tape measure the last time I saw it?", + "answer": "On the cardboard on the ground to your current right rear.", + "question_chinese": "我最后一次看到卷尺时，卷尺在哪里？", + "answer_chinese": "在你现在右后方的地上的纸板上。", + "evidence": [ + { + "description": "It is on the cardboard on the ground.", + "timestamps": { + "start": "00:21:32", + "end": "00:21:46" + } + } + ], + "question_timestamps": "00:23:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the cardboard to your right front.", + "B": "On the cardboard to your left rear.", + "C": "On the cardboard on the ground to your current right rear.", + "D": "On the cardboard on the ground, just off your left side, slightly behind where you're standing.", + "E": "On the ground on top of the cardboard, but positioned to your current left front instead of right rear." + }, + "answer_label": "C", + "answer_text": "On the cardboard on the ground to your current right rear.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_4", + "correct_answer": "On the cardboard on the ground to your current right rear.", + "original_options": { + "A": "On the cardboard to your right front.", + "B": "On the cardboard to your left rear.", + "C": "On the cardboard on the ground to your current right rear.", + "D": "On the cardboard on the ground, just off your left side, slightly behind where you're standing.", + "E": "On the ground on top of the cardboard, but positioned to your current left front instead of right rear." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 19 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_3", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the tape measure the last time I saw it?", + "answer": "The tape measure is in the grid of the second column from right to left and the second row from bottom to top in the metal frame.", + "question_chinese": "我最后一次看到卷尺时，卷尺在哪里？", + "answer_chinese": "卷尺位于金属框架中从右往左数第二列、从下往上数第二行的网格内。", + "evidence": [ + { + "description": "The tape measure is in the grid of the second column from right to left and the second row from bottom to top in the metal frame.", + "timestamps": { + "start": "00:33:27", + "end": "00:33:53" + } + } + ], + "question_timestamps": "00:33:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "First column from the right, second row up.", + "B": "The tape measure is in the grid of the second column from right to left and the second row from bottom to top in the metal frame.", + "C": "Second column from the right, top row.", + "D": "It is in the grid of the third column from right to left and the second row from bottom to top.", + "E": "It is in the grid of the second column from right to left, but in the third row from bottom to top." + }, + "answer_label": "B", + "answer_text": "The tape measure is in the grid of the second column from right to left and the second row from bottom to top in the metal frame.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_3", + "correct_answer": "The tape measure is in the grid of the second column from right to left and the second row from bottom to top in the metal frame.", + "original_options": { + "A": "First column from the right, second row up.", + "B": "The tape measure is in the grid of the second column from right to left and the second row from bottom to top in the metal frame.", + "C": "Second column from the right, top row.", + "D": "It is in the grid of the third column from right to left and the second row from bottom to top.", + "E": "It is in the grid of the second column from right to left, but in the third row from bottom to top." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 21, + 22 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_5", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to go to the swimming pool now, how should I get there?", + "answer": "You will turn right, go around the stack of wooden boards, and then you can see the swimming pool, after which you can walk to the side of the pool.", + "question_chinese": "我现在想去游泳池那，我将会怎么走？", + "answer_chinese": "你将会向右转，绕过木板堆，就可以看到泳池，然后走到泳池旁。", + "evidence": [ + { + "description": "There is a pile of wooden boards between the swimming pool and you.", + "timestamps": { + "start": "00:20:00", + "end": "00:20:10" + } + } + ], + "question_timestamps": "00:23:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left around the wooden boards.", + "B": "You will turn right, go around the stack of wooden boards, and then you can see the swimming pool, after which you can walk to the side of the pool.", + "C": "Go straight past the wooden boards.", + "D": "Turn right, pass the stack of wooden boards on your left, then continue straight until you reach the pool.", + "E": "Walk straight to the stack, loop around it to the left, then approach the swimming pool from the far side." + }, + "answer_label": "B", + "answer_text": "You will turn right, go around the stack of wooden boards, and then you can see the swimming pool, after which you can walk to the side of the pool.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_5", + "correct_answer": "You will turn right, go around the stack of wooden boards, and then you can see the swimming pool, after which you can walk to the side of the pool.", + "original_options": { + "A": "Turn left around the wooden boards.", + "B": "You will turn right, go around the stack of wooden boards, and then you can see the swimming pool, after which you can walk to the side of the pool.", + "C": "Go straight past the wooden boards.", + "D": "Turn right, pass the stack of wooden boards on your left, then continue straight until you reach the pool.", + "E": "Walk straight to the stack, loop around it to the left, then approach the swimming pool from the far side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 20 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_6", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to go to the swimming pool now, how should I get there?", + "answer": "Now you are on a ladder, walk towards the left rear, go around the stack of wooden boards, and then walk straight to get to the pool.", + "question_chinese": "我现在想去游泳池那，我将会怎么走？", + "answer_chinese": "你现在下人字梯，向左后方走，并且绕过木板堆，再直走就可以到泳池了。", + "evidence": [ + { + "description": "The swimming pool is behind you.", + "timestamps": { + "start": "00:44:57", + "end": "00:45:25" + } + } + ], + "question_timestamps": "00:45:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk to the right rear from the ladder.", + "B": "Step off the ladder and go forward.", + "C": "Now you are on a ladder, walk towards the left rear, go around the stack of wooden boards, and then walk straight to get to the pool.", + "D": "Now you are on a ladder, walk toward the left front, go around the stack of wooden boards, then continue straight.", + "E": "From the ladder, head to the left rear but turn left around the boards and stop before the pool area." + }, + "answer_label": "C", + "answer_text": "Now you are on a ladder, walk towards the left rear, go around the stack of wooden boards, and then walk straight to get to the pool.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_6", + "correct_answer": "Now you are on a ladder, walk towards the left rear, go around the stack of wooden boards, and then walk straight to get to the pool.", + "original_options": { + "A": "Walk to the right rear from the ladder.", + "B": "Step off the ladder and go forward.", + "C": "Now you are on a ladder, walk towards the left rear, go around the stack of wooden boards, and then walk straight to get to the pool.", + "D": "Now you are on a ladder, walk toward the left front, go around the stack of wooden boards, then continue straight.", + "E": "From the ladder, head to the left rear but turn left around the boards and stop before the pool area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 21, + 20 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_13", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach my phone without moving my feet right now?", + "answer": "You cannot get your phone.", + "question_chinese": "我现在能否在不移动脚步的情况下拿到手机？", + "answer_chinese": "你不能拿到手机。", + "evidence": [ + { + "description": "The mobile phone is on the windowsill.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:11:27", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can get your phone.", + "B": "You cannot get your phone." + }, + "answer_label": "B", + "answer_text": "You cannot get your phone.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_13", + "correct_answer": "You cannot get your phone.", + "original_options": { + "A": "You can get your phone.", + "B": "You cannot get your phone." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_14", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach my phone without moving my feet right now?", + "answer": "You cannot get your phone.", + "question_chinese": "我现在能否在不移动脚步的情况下拿到手机？", + "answer_chinese": "你不能拿到手机。", + "evidence": [ + { + "description": "The mobile phone is on the windowsill.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:22:03", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot get your phone.", + "B": "You can get your phone." + }, + "answer_label": "A", + "answer_text": "You cannot get your phone.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_14", + "correct_answer": "You cannot get your phone.", + "original_options": { + "A": "You cannot get your phone.", + "B": "You can get your phone." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_15", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the red broom when I first saw it?", + "answer": "The broom is leaning against the wall where the window is.", + "question_chinese": "我第一次看到红色扫把时，扫把在哪里？", + "answer_chinese": "扫帚靠在窗户旁边的墙上。", + "evidence": [ + { + "description": "The broom is leaning against the wall where the window is.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:22:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The broom is leaning against the wall where the window is.", + "B": "Leaning against the wall near the door.", + "C": "Leaning against the counter by the sink.", + "D": "The broom is leaning against the wall by the closet area, just beside the doorway rather than the window.", + "E": "The broom is leaning along the hallway wall near the stairs, not on the wall that has the window." + }, + "answer_label": "A", + "answer_text": "The broom is leaning against the wall where the window is.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_15", + "correct_answer": "The broom is leaning against the wall where the window is.", + "original_options": { + "A": "The broom is leaning against the wall where the window is.", + "B": "Leaning against the wall near the door.", + "C": "Leaning against the counter by the sink.", + "D": "The broom is leaning against the wall by the closet area, just beside the doorway rather than the window.", + "E": "The broom is leaning along the hallway wall near the stairs, not on the wall that has the window." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add", + "video_length": "00:45:44", + "qid": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_16", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the red broom when I first saw it?", + "answer": "The broom is leaning against the wall next to the window.", + "question_chinese": "我第一次看到红色扫把时，扫把在哪里？", + "answer_chinese": "扫帚靠在窗户旁边的墙上。", + "evidence": [ + { + "description": "The broom is leaning against the wall where the window is.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:45:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Leaning against the wall near the door.", + "B": "The broom is leaning against the wall next to the window.", + "C": "Lying flat on the ground by the table.", + "D": "The broom is standing upright beside the cabinet near the window, rather than leaning against the wall.", + "E": "The broom is positioned against the wall next to the bookshelf, not against the wall beside the window." + }, + "answer_label": "B", + "answer_text": "The broom is leaning against the wall next to the window.", + "q_id": "9c07f3a2-5033-4f78-a377-a4cf5a7d7add_16", + "correct_answer": "The broom is leaning against the wall next to the window.", + "original_options": { + "A": "Leaning against the wall near the door.", + "B": "The broom is leaning against the wall next to the window.", + "C": "Lying flat on the ground by the table.", + "D": "The broom is standing upright beside the cabinet near the window, rather than leaning against the wall.", + "E": "The broom is positioned against the wall next to the bookshelf, not against the wall beside the window." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 18 + ] + } +] diff --git a/QAs/9c502ef6-41cd-4037-a6be-216587ee8c3e-mcq.json b/QAs/9c502ef6-41cd-4037-a6be-216587ee8c3e-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..dcab65daa13d565442660a2e0fd3722d2642d400 --- /dev/null +++ b/QAs/9c502ef6-41cd-4037-a6be-216587ee8c3e-mcq.json @@ -0,0 +1,1033 @@ +[ + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_13", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I go to get more cloths?", + "answer": "You should look for it in the cardboard boxes next to the shelves.", + "question_chinese": "如果我想获取更多的抹布应该去哪找呢？", + "answer_chinese": "你应该去货架旁边的纸箱找。", + "evidence": [ + { + "description": "There are many cloths in the cardboard box.", + "timestamps": { + "start": "00:13:40", + "end": "00:13:42" + } + } + ], + "question_timestamps": "00:13:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Check the shelves by the racks.", + "B": "Look in the boxes under the table.", + "C": "You should check the plastic bins beside the shelves, not the cardboard boxes next to them.", + "D": "You should look for it on the shelves near the racks, rather than in any boxes nearby.", + "E": "You should look for it in the cardboard boxes next to the shelves." + }, + "answer_label": "E", + "answer_text": "You should look for it in the cardboard boxes next to the shelves.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_13", + "correct_answer": "You should look for it in the cardboard boxes next to the shelves.", + "original_options": { + "A": "Check the shelves by the racks.", + "B": "Look in the boxes under the table.", + "C": "You should check the plastic bins beside the shelves, not the cardboard boxes next to them.", + "D": "You should look for it on the shelves near the racks, rather than in any boxes nearby.", + "E": "You should look for it in the cardboard boxes next to the shelves." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_14", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should I go to get more cloths?", + "answer": "You should look for it in the cardboard boxes and on the shelves next to the racks.", + "question_chinese": "如果我想获取更多的抹布应该去哪找呢？", + "answer_chinese": "你应该去货架旁边的纸箱和横杠上找。", + "evidence": [ + { + "description": "There are many cloths in the cardboard box.", + "timestamps": { + "start": "00:13:40", + "end": "00:13:42" + } + }, + { + "description": "There is a pile of cloths on the crossbar.", + "timestamps": { + "start": "00:30:29", + "end": "00:30:36" + } + } + ], + "question_timestamps": "00:30:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You should look for it in the cardboard boxes and on the shelves next to the racks.", + "B": "Look in the green bags across from racks.", + "C": "Check the metal bins near the doors.", + "D": "You should look in the plastic crates and on the shelves above the tables, not by the racks.", + "E": "You should search the fabric crates and the shelves beside the tables, rather than near the racks." + }, + "answer_label": "A", + "answer_text": "You should look for it in the cardboard boxes and on the shelves next to the racks.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_14", + "correct_answer": "You should look for it in the cardboard boxes and on the shelves next to the racks.", + "original_options": { + "A": "You should look for it in the cardboard boxes and on the shelves next to the racks.", + "B": "Look in the green bags across from racks.", + "C": "Check the metal bins near the doors.", + "D": "You should look in the plastic crates and on the shelves above the tables, not by the racks.", + "E": "You should search the fabric crates and the shelves beside the tables, rather than near the racks." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 17 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_8_T2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want the pink rag and the white rag to be closer together, how far should I move them?", + "answer": "You can move the white rag about 2 meters to the right. The white rag is to your front-left.", + "question_chinese": "如果我想让粉色抹布和白色抹布放得近一些,应该移动多远?", + "answer_chinese": "你可以把白色抹布向右移动2米左右。白色抹布在你的左前方。", + "evidence": [ + { + "description": "They are currently about 2 meters apart. You can move the pink cloth about 2 meters to the left.", + "timestamps": { + "start": "00:24:00", + "end": "00:24:05" + } + } + ], + "question_timestamps": "00:24:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Move it about 1 meter right.", + "B": "You can move the white rag about 2 meters to the right. The white rag is to your front-left.", + "C": "Move it about 3 meters right.", + "D": "You can move the white rag about 4 meters to the right, since it sits to your front-left.", + "E": "You can leave the white rag where it is; it’s already close enough from your front-left." + }, + "answer_label": "B", + "answer_text": "You can move the white rag about 2 meters to the right. The white rag is to your front-left.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_8_T2", + "correct_answer": "You can move the white rag about 2 meters to the right. The white rag is to your front-left.", + "original_options": { + "A": "Move it about 1 meter right.", + "B": "You can move the white rag about 2 meters to the right. The white rag is to your front-left.", + "C": "Move it about 3 meters right.", + "D": "You can move the white rag about 4 meters to the right, since it sits to your front-left.", + "E": "You can leave the white rag where it is; it’s already close enough from your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_8_T1", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want the pink rag and the white rag to be closer together, how far should I move them?", + "answer": "You can move the pink rag about 2 meters to the left. The pink rag is to your front-right.", + "question_chinese": "如果我想让粉色抹布和白色抹布放得近一些,应该移动多远?", + "answer_chinese": "你可以把粉色抹布向左移动2米左右。粉色抹布在你的右前方。", + "evidence": [ + { + "description": "They are currently about 2 meters apart. You can move the pink cloth about 2 meters to the left.", + "timestamps": { + "start": "00:24:00", + "end": "00:24:05" + } + } + ], + "question_timestamps": "00:24:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can move the pink rag about 2 meters to the left. The pink rag is to your front-right.", + "B": "Move it about 1 meter left.", + "C": "Move it about 3 meters left.", + "D": "You can move the pink rag about 4 meters to the left, because it is positioned to your front-right.", + "E": "You can move the pink rag about 0 meters to the left; it’s already close from your front-right." + }, + "answer_label": "A", + "answer_text": "You can move the pink rag about 2 meters to the left. The pink rag is to your front-right.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_8_T1", + "correct_answer": "You can move the pink rag about 2 meters to the left. The pink rag is to your front-right.", + "original_options": { + "A": "You can move the pink rag about 2 meters to the left. The pink rag is to your front-right.", + "B": "Move it about 1 meter left.", + "C": "Move it about 3 meters left.", + "D": "You can move the pink rag about 4 meters to the left, because it is positioned to your front-right.", + "E": "You can move the pink rag about 0 meters to the left; it’s already close from your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_16", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to place the pink cloth and the white cloth closer together, how far should I move them?", + "answer": "You can move the pink cloth about two meters to the left.", + "question_chinese": "如果我想让粉色抹布和白色抹布放得近一些,应该移动多远?", + "answer_chinese": "你可以把粉色抹布向左移动2m左右。", + "evidence": [ + { + "description": "They are currently about 2 meters apart. You can move the pink cloth about 2 meters to the left.", + "timestamps": { + "start": "00:24:00", + "end": "00:24:05" + } + } + ], + "question_timestamps": "00:31:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Move it one meter left.", + "B": "You can move the pink cloth about two meters to the left.", + "C": "Move it two meters right.", + "D": "You can move the pink cloth about four meters to the left to bring it closer to the white cloth.", + "E": "You can move the pink cloth about two meters to the right to reduce the gap." + }, + "answer_label": "B", + "answer_text": "You can move the pink cloth about two meters to the left.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_16", + "correct_answer": "You can move the pink cloth about two meters to the left.", + "original_options": { + "A": "Move it one meter left.", + "B": "You can move the pink cloth about two meters to the left.", + "C": "Move it two meters right.", + "D": "You can move the pink cloth about four meters to the left to bring it closer to the white cloth.", + "E": "You can move the pink cloth about two meters to the right to reduce the gap." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 16 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_17", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand in front of the table, which is closer to me, the pink cloth or the wrench?", + "answer": "The wrench is closer to you.", + "question_chinese": "当我站在桌子前的时候,粉色抹布和扳手哪个离我更近?", + "answer_chinese": "扳手离你更近。", + "evidence": [ + { + "description": "The wrench is closer to you, about 30cm away, and the pink cloth is on your left, about 80cm away.", + "timestamps": { + "start": "00:24:48", + "end": "00:24:56" + } + } + ], + "question_timestamps": "00:24:59", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The wrench is closer to you.", + "B": "The pink cloth is closer to you." + }, + "answer_label": "A", + "answer_text": "The wrench is closer to you.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_17", + "correct_answer": "The wrench is closer to you.", + "original_options": { + "A": "The wrench is closer to you.", + "B": "The pink cloth is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_18", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand in front of the table, which is closer to me, the pink cloth or the wrench?", + "answer": "The wrench is closer to you.", + "question_chinese": "当我站在桌子前的时候,粉色抹布和扳手哪个离我更近?", + "answer_chinese": "扳手离你更近。", + "evidence": [ + { + "description": "The wrench is closer to you, about 30cm away, and the pink cloth is on your left, about 80cm away.", + "timestamps": { + "start": "00:24:48", + "end": "00:24:56" + } + } + ], + "question_timestamps": "00:32:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The pink cloth is closer to you.", + "B": "The wrench is closer to you." + }, + "answer_label": "B", + "answer_text": "The wrench is closer to you.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_18", + "correct_answer": "The wrench is closer to you.", + "original_options": { + "A": "The pink cloth is closer to you.", + "B": "The wrench is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_5", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots of black oil have I poured in total?", + "answer": "You poured a whole basin of black oil.", + "question_chinese": "我一共倒了几盆黑油？", + "answer_chinese": "你一共倒了一盆黑油。", + "evidence": [ + { + "description": "You poured the first basin of black oil.", + "timestamps": { + "start": "00:00:38", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:51", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You poured one pot total.", + "B": "You poured two pots total.", + "C": "You poured three pots of black oil in total across all pours, not an entire basin.", + "D": "You poured a whole basin of black oil.", + "E": "You poured four pots of black oil in total, filling only part of the basin." + }, + "answer_label": "D", + "answer_text": "You poured a whole basin of black oil.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_5", + "correct_answer": "You poured a whole basin of black oil.", + "original_options": { + "A": "You poured one pot total.", + "B": "You poured two pots total.", + "C": "You poured three pots of black oil in total across all pours, not an entire basin.", + "D": "You poured a whole basin of black oil.", + "E": "You poured four pots of black oil in total, filling only part of the basin." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_6", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots of black oil have I poured in total?", + "answer": "You poured two pots of black oil in total.", + "question_chinese": "我一共倒了几盆黑油？", + "answer_chinese": "你一共倒了两盆黑油。", + "evidence": [ + { + "description": "You poured the first basin of black oil.", + "timestamps": { + "start": "00:00:38", + "end": "00:00:48" + } + }, + { + "description": "You poured the second basin of black oil.", + "timestamps": { + "start": "00:04:38", + "end": "00:04:45" + } + } + ], + "question_timestamps": "00:24:45", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You poured one pot total.", + "B": "You poured two pots of black oil in total.", + "C": "You poured three pots total.", + "D": "You poured four pots of black oil in total, finishing after the fourth pour.", + "E": "You poured five pots of black oil altogether, counting each separate pour into the container." + }, + "answer_label": "B", + "answer_text": "You poured two pots of black oil in total.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_6", + "correct_answer": "You poured two pots of black oil in total.", + "original_options": { + "A": "You poured one pot total.", + "B": "You poured two pots of black oil in total.", + "C": "You poured three pots total.", + "D": "You poured four pots of black oil in total, finishing after the fourth pour.", + "E": "You poured five pots of black oil altogether, counting each separate pour into the container." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_4", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many black tubes did I throw into the bucket on the ground in total?", + "answer": "You threw one.", + "question_chinese": "我一共扔了几个黑筒在地上的桶里？", + "answer_chinese": "你扔了一个。", + "evidence": [ + { + "description": "You threw the first black cylinder.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:00:58", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You threw one.", + "B": "You threw none.", + "C": "You threw two.", + "D": "You threw three black tubes into the bucket on the ground in total, one after another.", + "E": "You threw four black tubes into the bucket on the ground overall, dropping each tube separately." + }, + "answer_label": "A", + "answer_text": "You threw one.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_4", + "correct_answer": "You threw one.", + "original_options": { + "A": "You threw one.", + "B": "You threw none.", + "C": "You threw two.", + "D": "You threw three black tubes into the bucket on the ground in total, one after another.", + "E": "You threw four black tubes into the bucket on the ground overall, dropping each tube separately." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_3", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many black tubes did I throw into the bucket on the ground in total?", + "answer": "You threw two.", + "question_chinese": "我一共扔了几个黑筒在地上的桶里？", + "answer_chinese": "你扔了两个。", + "evidence": [ + { + "description": "You threw the first black cylinder.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:57" + } + }, + { + "description": "You threw the second black cylinder.", + "timestamps": { + "start": "00:03:11", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:13:13", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You threw one.", + "B": "You threw three.", + "C": "You threw two.", + "D": "You threw four black tubes into the bucket on the ground in total, not just two.", + "E": "You threw five black tubes into the bucket on the ground overall, counting every tube you tossed." + }, + "answer_label": "C", + "answer_text": "You threw two.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_3", + "correct_answer": "You threw two.", + "original_options": { + "A": "You threw one.", + "B": "You threw three.", + "C": "You threw two.", + "D": "You threw four black tubes into the bucket on the ground in total, not just two.", + "E": "You threw five black tubes into the bucket on the ground overall, counting every tube you tossed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_19", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many wrenches did I use in total?", + "answer": "You used one type of wrenches.", + "question_chinese": "我一共用了几种扳手？", + "answer_chinese": "你用了一种扳手。", + "evidence": [ + { + "description": "You used a socket wrench.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:45" + } + } + ], + "question_timestamps": "00:01:48", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used two types.", + "B": "You used three types.", + "C": "You used four types of wrenches in total, switching between different wrench styles during the task.", + "D": "You used five types of wrenches overall, alternating tools instead of sticking with a single type.", + "E": "You used one type of wrenches." + }, + "answer_label": "E", + "answer_text": "You used one type of wrenches.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_19", + "correct_answer": "You used one type of wrenches.", + "original_options": { + "A": "You used two types.", + "B": "You used three types.", + "C": "You used four types of wrenches in total, switching between different wrench styles during the task.", + "D": "You used five types of wrenches overall, alternating tools instead of sticking with a single type.", + "E": "You used one type of wrenches." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_20", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many wrenches did I use in total?", + "answer": "You used two types of wrenches.", + "question_chinese": "我一共用了几种扳手？", + "answer_chinese": "你用了两种扳手。", + "evidence": [ + { + "description": "You used a socket wrench.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:45" + } + }, + { + "description": "You used an impact wrench.", + "timestamps": { + "start": "00:27:16", + "end": "00:27:18" + } + } + ], + "question_timestamps": "00:27:23", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You used three types.", + "B": "You used four types.", + "C": "You used two types of wrenches.", + "D": "You used five types of wrenches in total, changing to a different wrench multiple times.", + "E": "You used six types of wrenches overall, cycling through many wrench types during the work." + }, + "answer_label": "C", + "answer_text": "You used two types of wrenches.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_20", + "correct_answer": "You used two types of wrenches.", + "original_options": { + "A": "You used three types.", + "B": "You used four types.", + "C": "You used two types of wrenches.", + "D": "You used five types of wrenches in total, changing to a different wrench multiple times.", + "E": "You used six types of wrenches overall, cycling through many wrench types during the work." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_9", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans of engine oil did I take out of the white box?", + "answer": "You took out four cans of engine oil.", + "question_chinese": "我从白箱子里拿出了几罐机油？", + "answer_chinese": "你拿出了四罐机油。", + "evidence": [ + { + "description": "You first took out two cans of engine oil.", + "timestamps": { + "start": "00:09:19", + "end": "00:09:21" + } + }, + { + "description": "You took out two more cans of engine oil.", + "timestamps": { + "start": "00:09:23", + "end": "00:09:24" + } + } + ], + "question_timestamps": "00:09:29", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You took out two cans.", + "B": "You took out five cans.", + "C": "You took out six cans of engine oil from the white box, not just four.", + "D": "You took out seven cans of engine oil from the white box in total, removing each can one by one.", + "E": "You took out four cans of engine oil." + }, + "answer_label": "E", + "answer_text": "You took out four cans of engine oil.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_9", + "correct_answer": "You took out four cans of engine oil.", + "original_options": { + "A": "You took out two cans.", + "B": "You took out five cans.", + "C": "You took out six cans of engine oil from the white box, not just four.", + "D": "You took out seven cans of engine oil from the white box in total, removing each can one by one.", + "E": "You took out four cans of engine oil." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 20 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_10", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans of engine oil did I take out of the white box?", + "answer": "You took out four cans of engine oil.", + "question_chinese": "我从白箱子里拿出了几罐机油？", + "answer_chinese": "你拿出了四罐机油。", + "evidence": [ + { + "description": "You first took out two cans of engine oil.", + "timestamps": { + "start": "00:09:19", + "end": "00:09:21" + } + }, + { + "description": "You took out two more cans of engine oil.", + "timestamps": { + "start": "00:09:23", + "end": "00:09:24" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You took out three cans.", + "B": "You took out four cans of engine oil.", + "C": "You took out five cans.", + "D": "You took out six cans of engine oil from the white box, exceeding four cans.", + "E": "You took out eight cans of engine oil from the white box in total, pulling out many more than four." + }, + "answer_label": "B", + "answer_text": "You took out four cans of engine oil.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_10", + "correct_answer": "You took out four cans of engine oil.", + "original_options": { + "A": "You took out three cans.", + "B": "You took out four cans of engine oil.", + "C": "You took out five cans.", + "D": "You took out six cans of engine oil from the white box, exceeding four cans.", + "E": "You took out eight cans of engine oil from the white box in total, pulling out many more than four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 20 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You was holding a basin in your hand.", + "question_chinese": "我手里刚刚拿着什么东西？", + "answer_chinese": "你手里拿着一个盆。", + "evidence": [ + { + "description": "You are holding a black basin with both hands.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You was holding a cup.", + "B": "You was holding a plate.", + "C": "You was holding a towel in your hand just now, not a basin.", + "D": "You was holding a basin in your hand.", + "E": "You was holding a book in your hand a moment ago, rather than carrying a basin." + }, + "answer_label": "D", + "answer_text": "You was holding a basin in your hand.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_1", + "correct_answer": "You was holding a basin in your hand.", + "original_options": { + "A": "You was holding a cup.", + "B": "You was holding a plate.", + "C": "You was holding a towel in your hand just now, not a basin.", + "D": "You was holding a basin in your hand.", + "E": "You was holding a book in your hand a moment ago, rather than carrying a basin." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 16 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You was holding a box in your hand.", + "question_chinese": "我手里刚刚拿着什么东西？", + "answer_chinese": "你手里拿着一个箱子。", + "evidence": [ + { + "description": "You are holding a white box in your hand.", + "timestamps": { + "start": "00:09:12", + "end": "00:09:14" + } + } + ], + "question_timestamps": "00:09:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You was holding a cup.", + "B": "You was holding a box in your hand.", + "C": "You was holding a notebook.", + "D": "You was holding a bag of books in your hand just now instead of a box.", + "E": "You was holding a pencil and an eraser in your hand, not a single box." + }, + "answer_label": "B", + "answer_text": "You was holding a box in your hand.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_2", + "correct_answer": "You was holding a box in your hand.", + "original_options": { + "A": "You was holding a cup.", + "B": "You was holding a box in your hand.", + "C": "You was holding a notebook.", + "D": "You was holding a bag of books in your hand just now instead of a box.", + "E": "You was holding a pencil and an eraser in your hand, not a single box." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_21", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Compared to my current position, where is the cloth?", + "answer": "The cloth is on your right.", + "question_chinese": "相对我现在的位置，抹布在哪里？", + "answer_chinese": "抹布在你的右边。", + "evidence": [ + { + "description": "There is a pile of cloths on the right horizontal bar.", + "timestamps": { + "start": "00:28:06", + "end": "00:28:10" + } + } + ], + "question_timestamps": "00:28:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The cloth is on your left.", + "B": "The cloth is on your right.", + "C": "The cloth is behind you.", + "D": "The cloth is in front of you, not positioned to your right side.", + "E": "The cloth is above you rather than on your right when you look forward." + }, + "answer_label": "B", + "answer_text": "The cloth is on your right.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_21", + "correct_answer": "The cloth is on your right.", + "original_options": { + "A": "The cloth is on your left.", + "B": "The cloth is on your right.", + "C": "The cloth is behind you.", + "D": "The cloth is in front of you, not positioned to your right side.", + "E": "The cloth is above you rather than on your right when you look forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 14 + ] + }, + { + "video_uid": "9c502ef6-41cd-4037-a6be-216587ee8c3e", + "video_length": "00:34:03", + "qid": "9c502ef6-41cd-4037-a6be-216587ee8c3e_22", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Compared to my current position, where is the cloth?", + "answer": "The cloth is in front of you.", + "question_chinese": "相对我现在的位置，抹布在哪里？", + "answer_chinese": "抹布在你的前边。", + "evidence": [ + { + "description": "There is a pile of cloths on the front horizontal bar.", + "timestamps": { + "start": "00:30:27", + "end": "00:30:28" + } + } + ], + "question_timestamps": "00:30:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The cloth is behind you.", + "B": "The cloth is to your left.", + "C": "The cloth is slightly to your right side instead of straight in front of you.", + "D": "The cloth is in front of you.", + "E": "The cloth is positioned above your head, not in front of your current viewpoint." + }, + "answer_label": "D", + "answer_text": "The cloth is in front of you.", + "q_id": "9c502ef6-41cd-4037-a6be-216587ee8c3e_22", + "correct_answer": "The cloth is in front of you.", + "original_options": { + "A": "The cloth is behind you.", + "B": "The cloth is to your left.", + "C": "The cloth is slightly to your right side instead of straight in front of you.", + "D": "The cloth is in front of you.", + "E": "The cloth is positioned above your head, not in front of your current viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 14 + ] + } +] diff --git a/QAs/9d26362a-4411-4d8b-b114-03f4f89928d7-mcq.json b/QAs/9d26362a-4411-4d8b-b114-03f4f89928d7-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ab530d6f00c45c903f4985962ff08c13a87313e2 --- /dev/null +++ b/QAs/9d26362a-4411-4d8b-b114-03f4f89928d7-mcq.json @@ -0,0 +1,902 @@ +[ + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_11", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the warehouse?", + "answer": "Three.", + "question_chinese": "仓库内有几个椅子？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:27" + } + }, + { + "description": "The second and the third.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:20:19", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Five.", + "C": "Seven chairs were in the warehouse, spaced around the area rather than only three.", + "D": "Eight chairs were present in the warehouse, not the smaller group you recall.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_11", + "correct_answer": "Three.", + "original_options": { + "A": "One.", + "B": "Five.", + "C": "Seven chairs were in the warehouse, spaced around the area rather than only three.", + "D": "Eight chairs were present in the warehouse, not the smaller group you recall.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_12", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the warehouse?", + "answer": "Three.", + "question_chinese": "仓库内有几个椅子？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:27" + } + }, + { + "description": "The second and the third.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:27:24", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Six chairs were in the warehouse, meaning there were more than the number you stated.", + "E": "Seven chairs were in the warehouse, not just a small set of three." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_12", + "correct_answer": "Three.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Six chairs were in the warehouse, meaning there were more than the number you stated.", + "E": "Seven chairs were in the warehouse, not just a small set of three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 13 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many tires were there on the ground at the beginning?", + "answer": "Three.", + "question_chinese": "刚开始时地上有几个轮胎？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "There are three tires on the ground.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:06:59", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Five.", + "C": "Seven tires were on the ground at the beginning, not the smaller count you mentioned.", + "D": "Three.", + "E": "Eight tires were on the ground initially, meaning there were far more than three." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_3", + "correct_answer": "Three.", + "original_options": { + "A": "One.", + "B": "Five.", + "C": "Seven tires were on the ground at the beginning, not the smaller count you mentioned.", + "D": "Three.", + "E": "Eight tires were on the ground initially, meaning there were far more than three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many tires were there on the ground at the beginning?", + "answer": "Three.", + "question_chinese": "刚开始时地上有几个轮胎？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "There are three tires on the ground.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:14:22", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Five.", + "D": "Seven tires were on the ground at the beginning instead of only three.", + "E": "Nine tires were on the ground at the start, not the smaller number you recall." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_4", + "correct_answer": "Three.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Five.", + "D": "Seven tires were on the ground at the beginning instead of only three.", + "E": "Nine tires were on the ground at the start, not the smaller number you recall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_9", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do before and after cutting the wood?", + "answer": "You measured with a tape measure.", + "question_chinese": "在切割木头前后我干了什么？", + "answer_chinese": "你用卷尺进行了测量", + "evidence": [ + { + "description": "Cut the wood.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:31" + } + }, + { + "description": "Measure the wood.", + "timestamps": { + "start": "00:10:38", + "end": "00:10:42" + } + }, + { + "description": "Measure the wood.", + "timestamps": { + "start": "00:06:22", + "end": "00:06:27" + } + } + ], + "question_timestamps": "00:13:55", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You painted the wood.", + "B": "You adjusted the saw.", + "C": "You cleaned the surface before cutting, then tidied up afterward rather than measuring anything.", + "D": "You glued the pieces together before cutting and checked the fit after, not using a tape measure.", + "E": "You measured with a tape measure." + }, + "answer_label": "E", + "answer_text": "You measured with a tape measure.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_9", + "correct_answer": "You measured with a tape measure.", + "original_options": { + "A": "You painted the wood.", + "B": "You adjusted the saw.", + "C": "You cleaned the surface before cutting, then tidied up afterward rather than measuring anything.", + "D": "You glued the pieces together before cutting and checked the fit after, not using a tape measure.", + "E": "You measured with a tape measure." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 17 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_10", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do before and after cutting the wood?", + "answer": "You measured with a tape measure.", + "question_chinese": "在切割木头前后我干了什么？", + "answer_chinese": "你用卷尺进行了测量。", + "evidence": [ + { + "description": "Cut the wood.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:31" + } + }, + { + "description": "Measure the wood.", + "timestamps": { + "start": "00:10:38", + "end": "00:10:42" + } + }, + { + "description": "Measure the wood.", + "timestamps": { + "start": "00:06:22", + "end": "00:06:27" + } + } + ], + "question_timestamps": "00:21:30", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You marked the wood with a pencil.", + "B": "You used a folding ruler.", + "C": "You checked the length with a laser measure before cutting, then set the tool aside afterward.", + "D": "You measured with a tape measure.", + "E": "You verified the measurement with a square, then adjusted the board position after the cut." + }, + "answer_label": "D", + "answer_text": "You measured with a tape measure.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_10", + "correct_answer": "You measured with a tape measure.", + "original_options": { + "A": "You marked the wood with a pencil.", + "B": "You used a folding ruler.", + "C": "You checked the length with a laser measure before cutting, then set the tool aside afterward.", + "D": "You measured with a tape measure.", + "E": "You verified the measurement with a square, then adjusted the board position after the cut." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 15 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_2_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do before cutting the wood?", + "answer": "You made a mark with a pen. The pen was in front of you.", + "question_chinese": "在切割木头前我干了什么？", + "answer_chinese": "你用笔做了标记。笔在你的前面。", + "evidence": [ + { + "description": "You measured with a tape measure and made marks with a pen.", + "timestamps": { + "start": "00:06:22", + "end": "00:06:27" + } + } + ], + "question_timestamps": "00:06:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You marked it with a pencil.", + "B": "You measured with a tape measure.", + "C": "You made a mark with a pen, but the pen was behind you rather than in front.", + "D": "You measured the wood with a tape measure, and the tape measure sat in front of you.", + "E": "You made a mark with a pen. The pen was in front of you." + }, + "answer_label": "E", + "answer_text": "You made a mark with a pen. The pen was in front of you.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_2_T2", + "correct_answer": "You made a mark with a pen. The pen was in front of you.", + "original_options": { + "A": "You marked it with a pencil.", + "B": "You measured with a tape measure.", + "C": "You made a mark with a pen, but the pen was behind you rather than in front.", + "D": "You measured the wood with a tape measure, and the tape measure sat in front of you.", + "E": "You made a mark with a pen. The pen was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_2_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do before cutting the wood?", + "answer": "You measured with a tape measure. The tape measure was to your front-right.", + "question_chinese": "在切割木头前我干了什么？", + "answer_chinese": "你用卷尺进行了测量。卷尺在你的右前方。", + "evidence": [ + { + "description": "Cut the wood.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:31" + } + }, + { + "description": "Measure the wood.", + "timestamps": { + "start": "00:06:22", + "end": "00:06:27" + } + } + ], + "question_timestamps": "00:10:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You made a pen mark.", + "B": "Tape measure was front-left.", + "C": "You measured with a tape measure, but the tape measure was positioned to your back-right.", + "D": "You measured with a tape measure. The tape measure was to your front-right.", + "E": "You measured with a tape measure, and the tape measure was to your back-left before cutting." + }, + "answer_label": "D", + "answer_text": "You measured with a tape measure. The tape measure was to your front-right.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_2_T1", + "correct_answer": "You measured with a tape measure. The tape measure was to your front-right.", + "original_options": { + "A": "You made a pen mark.", + "B": "Tape measure was front-left.", + "C": "You measured with a tape measure, but the tape measure was positioned to your back-right.", + "D": "You measured with a tape measure. The tape measure was to your front-right.", + "E": "You measured with a tape measure, and the tape measure was to your back-left before cutting." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 16 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color gloves am I wearing?", + "answer": "Purple-black and blue-yellow.", + "question_chinese": "我带的什么颜色的手套？", + "answer_chinese": "紫黑色和蓝黄色。", + "evidence": [ + { + "description": "Purple-black gloves.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:22" + } + }, + { + "description": "Blue and yellow gloves.", + "timestamps": { + "start": "00:06:22", + "end": "00:06:29" + } + } + ], + "question_timestamps": "00:06:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Purple-black and blue-yellow.", + "B": "Black and red.", + "C": "Blue and white.", + "D": "You were wearing red-blue gloves on one hand and gray-black gloves on the other.", + "E": "Your gloves were blue-green on one hand, with the other glove being yellow-orange." + }, + "answer_label": "A", + "answer_text": "Purple-black and blue-yellow.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_1", + "correct_answer": "Purple-black and blue-yellow.", + "original_options": { + "A": "Purple-black and blue-yellow.", + "B": "Black and red.", + "C": "Blue and white.", + "D": "You were wearing red-blue gloves on one hand and gray-black gloves on the other.", + "E": "Your gloves were blue-green on one hand, with the other glove being yellow-orange." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 13 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color gloves am I wearing?", + "answer": "Purple-black and blue-yellow.", + "question_chinese": "我带的什么颜色的手套？", + "answer_chinese": "紫黑色和蓝黄色。", + "evidence": [ + { + "description": "Purple-black gloves.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:22" + } + }, + { + "description": "Blue and yellow gloves.", + "timestamps": { + "start": "00:06:22", + "end": "00:06:29" + } + } + ], + "question_timestamps": "00:14:02", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Pink and gray.", + "B": "Blue and white.", + "C": "You had black-red gloves on one hand, while the other glove was yellow-green.", + "D": "You were wearing two solid-colored gloves, one pink and the other gray, without any mixed patterning.", + "E": "Purple-black and blue-yellow." + }, + "answer_label": "E", + "answer_text": "Purple-black and blue-yellow.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_2", + "correct_answer": "Purple-black and blue-yellow.", + "original_options": { + "A": "Pink and gray.", + "B": "Blue and white.", + "C": "You had black-red gloves on one hand, while the other glove was yellow-green.", + "D": "You were wearing two solid-colored gloves, one pink and the other gray, without any mixed patterning.", + "E": "Purple-black and blue-yellow." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 16 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_13", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I move the trash can?", + "answer": "Yes.", + "question_chinese": "我是否将垃圾桶移动了位置？", + "answer_chinese": "是的。", + "evidence": [ + { + "description": "Move the position of the trash can.", + "timestamps": { + "start": "00:14:15", + "end": "00:14:41" + } + } + ], + "question_timestamps": "00:19:52", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_13", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_14", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I move the trash can?", + "answer": "Yes.", + "question_chinese": "我是否将垃圾桶移动了位置？", + "answer_chinese": "是的。", + "evidence": [ + { + "description": "Move the position of the trash can.", + "timestamps": { + "start": "00:14:15", + "end": "00:14:41" + } + } + ], + "question_timestamps": "00:27:42", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_14", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_1_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the very beginning, how many tires did I see on the ground?", + "answer": "At the very beginning, you saw three tires on the ground. The tires were to your back-left.", + "question_chinese": "我看见刚开始时地上有几个轮胎？", + "answer_chinese": "你看见刚开始时地上有三个轮胎。轮胎在你的左后方。", + "evidence": [ + { + "description": "There are three tires on the ground.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One tire, back-left.", + "B": "Two tires, back-left.", + "C": "At the very beginning, you saw four tires on the ground, and they were to your back-left.", + "D": "At the very beginning, you saw three tires on the ground. The tires were to your back-left.", + "E": "At the very beginning, you saw five tires on the ground, with the tires located to your back-left." + }, + "answer_label": "D", + "answer_text": "At the very beginning, you saw three tires on the ground. The tires were to your back-left.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_1_T2", + "correct_answer": "At the very beginning, you saw three tires on the ground. The tires were to your back-left.", + "original_options": { + "A": "One tire, back-left.", + "B": "Two tires, back-left.", + "C": "At the very beginning, you saw four tires on the ground, and they were to your back-left.", + "D": "At the very beginning, you saw three tires on the ground. The tires were to your back-left.", + "E": "At the very beginning, you saw five tires on the ground, with the tires located to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 18 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_1_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the very beginning, how many tires did I see on the ground?", + "answer": "At the very beginning, you saw two tires on the ground. The tires were to your back-right.", + "question_chinese": "我看见刚开始时地上有几个轮胎？", + "answer_chinese": "你看见刚开始时地上有两个轮胎。轮胎在你的右后方。", + "evidence": [ + { + "description": "There are two tires on the ground.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw no tires on the ground.", + "B": "You saw one tire on the ground.", + "C": "At the very beginning, you saw two tires on the ground. The tires were to your back-right.", + "D": "At the very beginning, you saw three tires on the ground, lying to your back-right.", + "E": "At the very beginning, you saw four tires on the ground, stacked together to your back-right." + }, + "answer_label": "C", + "answer_text": "At the very beginning, you saw two tires on the ground. The tires were to your back-right.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_1_T1", + "correct_answer": "At the very beginning, you saw two tires on the ground. The tires were to your back-right.", + "original_options": { + "A": "You saw no tires on the ground.", + "B": "You saw one tire on the ground.", + "C": "At the very beginning, you saw two tires on the ground. The tires were to your back-right.", + "D": "At the very beginning, you saw three tires on the ground, lying to your back-right.", + "E": "At the very beginning, you saw four tires on the ground, stacked together to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_3_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs can I see inside the warehouse?", + "answer": "You can see one chair inside the warehouse. The chair is to your left.", + "question_chinese": "我能看到仓库内有几个椅子？", + "answer_chinese": "你能看见仓库里有一个椅子。椅子在你的左边。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can see one chair inside the warehouse. The chair is to your left.", + "B": "You can see zero chairs inside.", + "C": "You can see two chairs inside.", + "D": "You can see three chairs inside the warehouse, and they’re grouped together to your left.", + "E": "You can see four chairs inside the warehouse, lined up on the left side as you look in." + }, + "answer_label": "A", + "answer_text": "You can see one chair inside the warehouse. The chair is to your left.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_3_T2", + "correct_answer": "You can see one chair inside the warehouse. The chair is to your left.", + "original_options": { + "A": "You can see one chair inside the warehouse. The chair is to your left.", + "B": "You can see zero chairs inside.", + "C": "You can see two chairs inside.", + "D": "You can see three chairs inside the warehouse, and they’re grouped together to your left.", + "E": "You can see four chairs inside the warehouse, lined up on the left side as you look in." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "9d26362a-4411-4d8b-b114-03f4f89928d7", + "video_length": "00:27:42", + "qid": "9d26362a-4411-4d8b-b114-03f4f89928d7_3_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs can I see inside the warehouse?", + "answer": "You can see three chairs inside the warehouse. The chairs are below you.", + "question_chinese": "我能看到仓库内有几个椅子？", + "answer_chinese": "你能看到的仓库内有三个椅子。椅子在你的下面。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:29" + } + }, + { + "description": "The second and the third.", + "timestamps": { + "start": "00:00:54", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:01:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can see one chair below you.", + "B": "You can see two chairs below you.", + "C": "You can see four chairs inside the warehouse, all positioned below you near the floor area.", + "D": "You can see five chairs inside the warehouse, clustered below you in the lower part of the scene.", + "E": "You can see three chairs inside the warehouse. The chairs are below you." + }, + "answer_label": "E", + "answer_text": "You can see three chairs inside the warehouse. The chairs are below you.", + "q_id": "9d26362a-4411-4d8b-b114-03f4f89928d7_3_T1", + "correct_answer": "You can see three chairs inside the warehouse. The chairs are below you.", + "original_options": { + "A": "You can see one chair below you.", + "B": "You can see two chairs below you.", + "C": "You can see four chairs inside the warehouse, all positioned below you near the floor area.", + "D": "You can see five chairs inside the warehouse, clustered below you in the lower part of the scene.", + "E": "You can see three chairs inside the warehouse. The chairs are below you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + } +] diff --git a/QAs/9d72fb6b-ba44-45fc-ab70-1588a572ebfe-mcq.json b/QAs/9d72fb6b-ba44-45fc-ab70-1588a572ebfe-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..95bce8ddb3863f191fbbdf8b7a98841810fda5e3 --- /dev/null +++ b/QAs/9d72fb6b-ba44-45fc-ab70-1588a572ebfe-mcq.json @@ -0,0 +1,1141 @@ +[ + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_3", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What will I do before cutting?", + "answer": "First, measure with a caliper and then mark with a pen.", + "question_chinese": "在切割之前我会做什么", + "answer_chinese": "你先用卡尺测量再用笔做记号", + "evidence": [ + { + "description": "You first measure with a caliper.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:25" + } + }, + { + "description": "You make a mark with a pen.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:11:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Measure with a ruler, then mark.", + "B": "Mark with a pen, then measure.", + "C": "First, measure the piece with a ruler, and then make a quick mark where the cut will be.", + "D": "First, use a tape measure to check the length, then mark the cut line before cutting.", + "E": "First, measure with a caliper and then mark with a pen." + }, + "answer_label": "E", + "answer_text": "First, measure with a caliper and then mark with a pen.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_3", + "correct_answer": "First, measure with a caliper and then mark with a pen.", + "original_options": { + "A": "Measure with a ruler, then mark.", + "B": "Mark with a pen, then measure.", + "C": "First, measure the piece with a ruler, and then make a quick mark where the cut will be.", + "D": "First, use a tape measure to check the length, then mark the cut line before cutting.", + "E": "First, measure with a caliper and then mark with a pen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 16 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_4", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What will I do before cutting?", + "answer": "You mark the already cut wooden boards.", + "question_chinese": "在切割之前我会做什么", + "answer_chinese": "你用已经切割好的木板做记号", + "evidence": [ + { + "description": "You first measure with a caliper.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:25" + } + }, + { + "description": "You make a mark with a pen.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:30" + } + }, + { + "description": "You mark the already cut wooden boards.", + "timestamps": { + "start": "00:11:25", + "end": "00:11:50" + } + } + ], + "question_timestamps": "00:22:50", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You measure the wooden boards.", + "B": "You clean the wooden boards.", + "C": "You mark the already cut wooden boards.", + "D": "You align the uncut wooden boards on the bench, getting them straight before you cut anything.", + "E": "You mark the uncut wooden boards first, drawing the cut line before doing any cutting." + }, + "answer_label": "C", + "answer_text": "You mark the already cut wooden boards.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_4", + "correct_answer": "You mark the already cut wooden boards.", + "original_options": { + "A": "You measure the wooden boards.", + "B": "You clean the wooden boards.", + "C": "You mark the already cut wooden boards.", + "D": "You align the uncut wooden boards on the bench, getting them straight before you cut anything.", + "E": "You mark the uncut wooden boards first, drawing the cut line before doing any cutting." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_6_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many wooden boards did I stand upright?", + "answer": "You stood up a total of eight planks, behind you.", + "question_chinese": "我一共竖放了多少木板？", + "answer_chinese": "你一共竖放了八个木板，在你的后方。", + "evidence": [ + { + "description": "You placed a total of eight wooden boards vertically.", + "timestamps": { + "start": "00:27:10", + "end": "00:28:18" + } + } + ], + "question_timestamps": "00:34:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You stood up a total of eight planks, behind you.", + "B": "You stood up six planks.", + "C": "You stood up seven planks.", + "D": "You stood up a total of nine planks, behind you, after finishing the last adjustment.", + "E": "You stood up a total of ten planks, behind you, placing them upright in the same area." + }, + "answer_label": "A", + "answer_text": "You stood up a total of eight planks, behind you.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_6_T1", + "correct_answer": "You stood up a total of eight planks, behind you.", + "original_options": { + "A": "You stood up a total of eight planks, behind you.", + "B": "You stood up six planks.", + "C": "You stood up seven planks.", + "D": "You stood up a total of nine planks, behind you, after finishing the last adjustment.", + "E": "You stood up a total of ten planks, behind you, placing them upright in the same area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_6_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many wooden boards did I stand upright?", + "answer": "You stood up a total of eight planks, to your left.", + "question_chinese": "我一共竖放了多少木板？", + "answer_chinese": "你一共竖放了八个木板，在你的左方。", + "evidence": [ + { + "description": "You placed a total of eight wooden boards vertically.", + "timestamps": { + "start": "00:27:10", + "end": "00:28:18" + } + } + ], + "question_timestamps": "00:45:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You stood up six planks.", + "B": "You stood up a total of eight planks, to your left.", + "C": "You stood up seven planks.", + "D": "You stood up a total of nine planks, to your left, leaning them upright together.", + "E": "You stood up a total of ten planks, to your left, setting them upright along that side." + }, + "answer_label": "B", + "answer_text": "You stood up a total of eight planks, to your left.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_6_T2", + "correct_answer": "You stood up a total of eight planks, to your left.", + "original_options": { + "A": "You stood up six planks.", + "B": "You stood up a total of eight planks, to your left.", + "C": "You stood up seven planks.", + "D": "You stood up a total of nine planks, to your left, leaning them upright together.", + "E": "You stood up a total of ten planks, to your left, setting them upright along that side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_5_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I keep every piece of the wooden board that I cut off?", + "answer": "No, you didn’t do that—all the planks are in front of you.", + "question_chinese": "我保留了所有切割下来的每个部分的木板吗？", + "answer_chinese": "你没有那样做，木板都在你的前方。", + "evidence": [ + { + "description": "You didn't do that.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:52" + } + }, + { + "description": "You didn't do that.", + "timestamps": { + "start": "00:12:00", + "end": "00:12:08" + } + } + ], + "question_timestamps": "00:13:37", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you kept every piece of the wooden board that you cut off.", + "B": "No, you didn’t do that—all the planks are in front of you." + }, + "answer_label": "B", + "answer_text": "No, you didn’t do that—all the planks are in front of you.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_5_T2", + "correct_answer": "No, you didn’t do that—all the planks are in front of you.", + "original_options": { + "A": "Yes, you kept every piece of the wooden board that you cut off.", + "B": "No, you didn’t do that—all the planks are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_5_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I keep every piece of the wooden board that I cut off?", + "answer": "No, you didn’t do that—all the planks are to your left.", + "question_chinese": "我保留了所有切割下来的每个部分的木板吗？", + "answer_chinese": "你没有那样做，木板都在你的左方。", + "evidence": [ + { + "description": "You didn't do that.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:14:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you didn’t do that—all the planks are to your left.", + "B": "Yes, you did—you kept every piece you cut off." + }, + "answer_label": "A", + "answer_text": "No, you didn’t do that—all the planks are to your left.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_5_T1", + "correct_answer": "No, you didn’t do that—all the planks are to your left.", + "original_options": { + "A": "No, you didn’t do that—all the planks are to your left.", + "B": "Yes, you did—you kept every piece you cut off." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wooden boards have I cut?", + "answer": "You cut three wooden boards.", + "question_chinese": "我切割了多少木板", + "answer_chinese": "你切割了三个木板", + "evidence": [ + { + "description": "The first wooden plank.", + "timestamps": { + "start": "00:00:40", + "end": "00:04:00" + } + }, + { + "description": "The second wooden board.", + "timestamps": { + "start": "00:04:05", + "end": "00:07:10" + } + }, + { + "description": "The third plank.", + "timestamps": { + "start": "00:07:45", + "end": "00:10:45" + } + } + ], + "question_timestamps": "00:11:25", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You cut two wooden boards.", + "B": "You cut four wooden boards.", + "C": "You cut five wooden boards in total, after making two additional cuts beyond the first three.", + "D": "You cut six wooden boards overall, counting every board you trimmed down during the sequence.", + "E": "You cut three wooden boards." + }, + "answer_label": "E", + "answer_text": "You cut three wooden boards.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_5", + "correct_answer": "You cut three wooden boards.", + "original_options": { + "A": "You cut two wooden boards.", + "B": "You cut four wooden boards.", + "C": "You cut five wooden boards in total, after making two additional cuts beyond the first three.", + "D": "You cut six wooden boards overall, counting every board you trimmed down during the sequence.", + "E": "You cut three wooden boards." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wooden boards have I cut?", + "answer": "You cut four wooden boards.", + "question_chinese": "我切割了多少木板", + "answer_chinese": "你切割了四个木板", + "evidence": [ + { + "description": "The first wooden plank.", + "timestamps": { + "start": "00:00:40", + "end": "00:04:00" + } + }, + { + "description": "The second wooden board.", + "timestamps": { + "start": "00:04:05", + "end": "00:07:10" + } + }, + { + "description": "The third plank.", + "timestamps": { + "start": "00:07:45", + "end": "00:10:45" + } + }, + { + "description": "The fourth plank.", + "timestamps": { + "start": "00:12:00", + "end": "00:14:55" + } + } + ], + "question_timestamps": "00:22:50", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You cut four wooden boards.", + "B": "You cut three wooden boards.", + "C": "You cut five wooden boards.", + "D": "You cut two wooden boards in total, finishing after the second board was cut to size.", + "E": "You cut six wooden boards altogether, continuing to cut boards well past the fourth one." + }, + "answer_label": "A", + "answer_text": "You cut four wooden boards.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_6", + "correct_answer": "You cut four wooden boards.", + "original_options": { + "A": "You cut four wooden boards.", + "B": "You cut three wooden boards.", + "C": "You cut five wooden boards.", + "D": "You cut two wooden boards in total, finishing after the second board was cut to size.", + "E": "You cut six wooden boards altogether, continuing to cut boards well past the fourth one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_9_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What did I put at the doorway?", + "answer": "You placed a tripod at the doorway, to your right.", + "question_chinese": "我在门口放了什么？", + "answer_chinese": "你在门口放了三脚架，在你的右方。", + "evidence": [ + { + "description": "You placed a tripod at the door.", + "timestamps": { + "start": "00:37:48", + "end": "00:38:09" + } + } + ], + "question_timestamps": "00:39:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A tripod to your left.", + "B": "A tripod in front of you.", + "C": "You placed a tripod at the doorway, positioned slightly behind you rather than to your right.", + "D": "You put a tripod at the doorway, but set it on your left side instead of the right.", + "E": "You placed a tripod at the doorway, to your right." + }, + "answer_label": "E", + "answer_text": "You placed a tripod at the doorway, to your right.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_9_T2", + "correct_answer": "You placed a tripod at the doorway, to your right.", + "original_options": { + "A": "A tripod to your left.", + "B": "A tripod in front of you.", + "C": "You placed a tripod at the doorway, positioned slightly behind you rather than to your right.", + "D": "You put a tripod at the doorway, but set it on your left side instead of the right.", + "E": "You placed a tripod at the doorway, to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 18 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_9_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What did I put at the doorway?", + "answer": "You placed a tripod at the doorway, to your left.", + "question_chinese": "我在门口放了什么？", + "answer_chinese": "你在门口放了三脚架，在你的左方。", + "evidence": [ + { + "description": "You placed a tripod at the door.", + "timestamps": { + "start": "00:37:48", + "end": "00:38:09" + } + } + ], + "question_timestamps": "00:44:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A tripod to your right.", + "B": "A tripod in front of you.", + "C": "You placed a tripod at the doorway, but it ended up behind you instead of to your left.", + "D": "You placed a tripod at the doorway, to your left.", + "E": "You set the tripod at the doorway on your right side, not on your left as you moved past." + }, + "answer_label": "D", + "answer_text": "You placed a tripod at the doorway, to your left.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_9_T1", + "correct_answer": "You placed a tripod at the doorway, to your left.", + "original_options": { + "A": "A tripod to your right.", + "B": "A tripod in front of you.", + "C": "You placed a tripod at the doorway, but it ended up behind you instead of to your left.", + "D": "You placed a tripod at the doorway, to your left.", + "E": "You set the tripod at the doorway on your right side, not on your left as you moved past." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 19 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_10_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What did I do to the window?", + "answer": "You pulled down the blinds—they’re all behind you.", + "question_chinese": "我对窗户做了什么？", + "answer_chinese": "你把百叶窗拉下来了，都在你的后方。", + "evidence": [ + { + "description": "You pulled down the blinds.", + "timestamps": { + "start": "00:40:00", + "end": "00:40:30" + } + } + ], + "question_timestamps": "00:40:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You left the blinds up.", + "B": "You pulled the blinds up.", + "C": "You pulled down the blinds, but they’re all in front of you rather than behind you.", + "D": "You pulled down the blinds, though they ended up behind you and to your right instead of all directly behind.", + "E": "You pulled down the blinds—they’re all behind you." + }, + "answer_label": "E", + "answer_text": "You pulled down the blinds—they’re all behind you.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_10_T2", + "correct_answer": "You pulled down the blinds—they’re all behind you.", + "original_options": { + "A": "You left the blinds up.", + "B": "You pulled the blinds up.", + "C": "You pulled down the blinds, but they’re all in front of you rather than behind you.", + "D": "You pulled down the blinds, though they ended up behind you and to your right instead of all directly behind.", + "E": "You pulled down the blinds—they’re all behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 20 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_10_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What did I do to the window?", + "answer": "You pulled down the blinds; they’re behind you and to your right, respectively.", + "question_chinese": "我对窗户做了什么？", + "answer_chinese": "你把百叶窗拉下来了，分别在你的后方和右方。", + "evidence": [ + { + "description": "You pulled down the blinds.", + "timestamps": { + "start": "00:40:00", + "end": "00:40:30" + } + } + ], + "question_timestamps": "00:45:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They’re behind you and left.", + "B": "They’re in front of you and right.", + "C": "You pulled down the blinds, but they’re in front of you and to your left, respectively, not behind-right.", + "D": "You pulled down the blinds; they’re behind you and to your right, respectively.", + "E": "You pulled down the blinds, with them all behind you rather than split behind you and to your right, respectively." + }, + "answer_label": "D", + "answer_text": "You pulled down the blinds; they’re behind you and to your right, respectively.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_10_T1", + "correct_answer": "You pulled down the blinds; they’re behind you and to your right, respectively.", + "original_options": { + "A": "They’re behind you and left.", + "B": "They’re in front of you and right.", + "C": "You pulled down the blinds, but they’re in front of you and to your left, respectively, not behind-right.", + "D": "You pulled down the blinds; they’re behind you and to your right, respectively.", + "E": "You pulled down the blinds, with them all behind you rather than split behind you and to your right, respectively." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 19, + 20 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_13", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where did I put the cut wooden boards?", + "answer": "You put them on the window.", + "question_chinese": "我把切割好的木板放在哪里了", + "answer_chinese": "你把它们搭在了窗户上", + "evidence": [ + { + "description": "You put them on the window.", + "timestamps": { + "start": "00:15:30", + "end": "00:16:38" + } + } + ], + "question_timestamps": "00:34:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the porch railing.", + "B": "On the dining table.", + "C": "You put the cut wooden boards on the bookshelf instead of placing them on the window.", + "D": "You put them on the window.", + "E": "You set the cut wooden boards on the back patio rather than resting them on the window." + }, + "answer_label": "D", + "answer_text": "You put them on the window.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_13", + "correct_answer": "You put them on the window.", + "original_options": { + "A": "On the porch railing.", + "B": "On the dining table.", + "C": "You put the cut wooden boards on the bookshelf instead of placing them on the window.", + "D": "You put them on the window.", + "E": "You set the cut wooden boards on the back patio rather than resting them on the window." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_14", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where did I put the cut wooden boards?", + "answer": "You threw away some of it and kept some for use.", + "question_chinese": "我把切割好的木板放在哪里了", + "answer_chinese": "你把一部分扔了一部分留着用", + "evidence": [ + { + "description": "You put them on the window.", + "timestamps": { + "start": "00:15:30", + "end": "00:16:38" + } + }, + { + "description": "You threw away some of it and kept some for use.", + "timestamps": { + "start": "00:32:15", + "end": "00:34:00" + } + } + ], + "question_timestamps": "00:45:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You kept all of them for use.", + "B": "You threw away some of it and kept some for use.", + "C": "You threw all of them away.", + "D": "You set the boards aside to reuse later, without throwing any away during the cleanup.", + "E": "You got rid of the boards entirely and didn’t save any pieces to use afterward." + }, + "answer_label": "B", + "answer_text": "You threw away some of it and kept some for use.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_14", + "correct_answer": "You threw away some of it and kept some for use.", + "original_options": { + "A": "You kept all of them for use.", + "B": "You threw away some of it and kept some for use.", + "C": "You threw all of them away.", + "D": "You set the boards aside to reuse later, without throwing any away during the cleanup.", + "E": "You got rid of the boards entirely and didn’t save any pieces to use afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 16 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_1_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Could I reach the yellow line on the ground?", + "answer": "Yes, you can—the line is right to your right.", + "question_chinese": "我能够到地上黄色的线吗？", + "answer_chinese": "你可以，线就在你的右方。", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:30" + } + } + ], + "question_timestamps": "00:11:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can—the line is right to your right.", + "B": "No, you can’t—the yellow line is behind you." + }, + "answer_label": "A", + "answer_text": "Yes, you can—the line is right to your right.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_1_T1", + "correct_answer": "Yes, you can—the line is right to your right.", + "original_options": { + "A": "Yes, you can—the line is right to your right.", + "B": "No, you can’t—the yellow line is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_1_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Could I reach the yellow line on the ground?", + "answer": "You can’t right now—the yellow line is to your back-right.", + "question_chinese": "我能够到地上黄色的线吗？", + "answer_chinese": "你现在不可以，黄色的线在你的右后方。", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:30" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:12:10", + "end": "00:13:20" + } + } + ], + "question_timestamps": "00:22:54", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can—the line is right to your right.", + "B": "You can’t right now—the yellow line is to your back-right." + }, + "answer_label": "B", + "answer_text": "You can’t right now—the yellow line is to your back-right.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_1_T2", + "correct_answer": "You can’t right now—the yellow line is to your back-right.", + "original_options": { + "A": "Yes, you can—the line is right to your right.", + "B": "You can’t right now—the yellow line is to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_17", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How can I get out of the room?", + "answer": "You can go straight first, then turn left, and finally go straight.", + "question_chinese": "我该如何从房间出来", + "answer_chinese": "你可以先直走再左拐最后直走", + "evidence": [ + { + "description": "You can go straight first.", + "timestamps": { + "start": "00:28:20", + "end": "00:28:22" + } + }, + { + "description": "Then you turn left.", + "timestamps": { + "start": "00:28:22", + "end": "00:28:23" + } + }, + { + "description": "Finally, you go straight.", + "timestamps": { + "start": "00:28:23", + "end": "00:28:27" + } + } + ], + "question_timestamps": "00:34:15", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go straight.", + "B": "You can go straight first, then turn left, and finally go straight.", + "C": "Go straight, then turn right.", + "D": "You can turn right first, walk straight ahead, then make another right turn to leave the room.", + "E": "You can go straight first, then turn right, and finally go straight until you reach the exit." + }, + "answer_label": "B", + "answer_text": "You can go straight first, then turn left, and finally go straight.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_17", + "correct_answer": "You can go straight first, then turn left, and finally go straight.", + "original_options": { + "A": "Turn left, then go straight.", + "B": "You can go straight first, then turn left, and finally go straight.", + "C": "Go straight, then turn right.", + "D": "You can turn right first, walk straight ahead, then make another right turn to leave the room.", + "E": "You can go straight first, then turn right, and finally go straight until you reach the exit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_18", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How can I get out of the room?", + "answer": "You can step back, then turn left, and then step back again.", + "question_chinese": "我该如何从房间出来", + "answer_chinese": "你可以向后退然后向左转再向后退", + "evidence": [ + { + "description": "You can go straight first.", + "timestamps": { + "start": "00:28:20", + "end": "00:28:22" + } + }, + { + "description": "Then you turn left.", + "timestamps": { + "start": "00:28:22", + "end": "00:28:23" + } + }, + { + "description": "Finally, you go straight.", + "timestamps": { + "start": "00:28:23", + "end": "00:28:27" + } + }, + { + "description": "You can step back, then turn left, and then step back again.", + "timestamps": { + "start": "00:36:02", + "end": "00:36:07" + } + } + ], + "question_timestamps": "00:45:40", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Step forward, then turn left.", + "B": "Step back, then turn right.", + "C": "You can step back first, then turn left, and then step forward to move away from the opening.", + "D": "You can step forward, turn left, and then step back again to return to where you started.", + "E": "You can step back, then turn left, and then step back again." + }, + "answer_label": "E", + "answer_text": "You can step back, then turn left, and then step back again.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_18", + "correct_answer": "You can step back, then turn left, and then step back again.", + "original_options": { + "A": "Step forward, then turn left.", + "B": "Step back, then turn right.", + "C": "You can step back first, then turn left, and then step forward to move away from the opening.", + "D": "You can step forward, turn left, and then step back again to return to where you started.", + "E": "You can step back, then turn left, and then step back again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_4_T2", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I cut the wooden board for the fourth time, did I finish it in one continuous go?", + "answer": "No, you didn’t do that—all the planks are in front of you.", + "question_chinese": "第四次切割木板时我一气呵成的完成了吗？", + "answer_chinese": "你没有那样做， 木板都在你的前方。", + "evidence": [ + { + "description": "You didn't do that.", + "timestamps": { + "start": "00:11:00", + "end": "00:11:24" + } + }, + { + "description": "You didn't do that.", + "timestamps": { + "start": "00:11:26", + "end": "00:14:55" + } + } + ], + "question_timestamps": "00:15:08", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you didn’t do that—all the planks are in front of you.", + "B": "Yes, you did—you finished it in one continuous go." + }, + "answer_label": "A", + "answer_text": "No, you didn’t do that—all the planks are in front of you.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_4_T2", + "correct_answer": "No, you didn’t do that—all the planks are in front of you.", + "original_options": { + "A": "No, you didn’t do that—all the planks are in front of you.", + "B": "Yes, you did—you finished it in one continuous go." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe", + "video_length": "00:45:40", + "qid": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_4_T1", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I cut the wooden board for the fourth time, did I finish it in one continuous go?", + "answer": "No, you didn’t do that—the plank is to your back-right.", + "question_chinese": "第四次切割木板时我一气呵成的完成了吗？", + "answer_chinese": "你没有那样做，木板在你的右后方。", + "evidence": [ + { + "description": "You didn't do that.", + "timestamps": { + "start": "00:11:00", + "end": "00:11:24" + } + } + ], + "question_timestamps": "00:16:41", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you did—you finished it in one continuous go.", + "B": "No, you didn’t do that—the plank is to your back-right." + }, + "answer_label": "B", + "answer_text": "No, you didn’t do that—the plank is to your back-right.", + "q_id": "9d72fb6b-ba44-45fc-ab70-1588a572ebfe_4_T1", + "correct_answer": "No, you didn’t do that—the plank is to your back-right.", + "original_options": { + "A": "Yes, you did—you finished it in one continuous go.", + "B": "No, you didn’t do that—the plank is to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + } +] diff --git a/QAs/9db30b24-9088-408f-b34d-330d0fa57749-001-mcq.json b/QAs/9db30b24-9088-408f-b34d-330d0fa57749-001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c213491805d0715cdc4f7751c889e944ce44163e --- /dev/null +++ b/QAs/9db30b24-9088-408f-b34d-330d0fa57749-001-mcq.json @@ -0,0 +1,763 @@ +[ + { + "video_uid": "9db30b24-9088-408f-b34d-330d0fa57749-001", + "video_length": "00:30:00", + "qid": "9db30b24-9088-408f-b34d-330d0fa57749-001_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get from the open space to my current position?", + "answer": "Turn left and walk forward ten meters.", + "question_chinese": "我从空地处是怎么走到当前位置的？", + "answer_chinese": "先向左转往前走10米", + "evidence": [ + { + "description": "First turn left and walk forward 10 meters", + "timestamps": { + "start": "00:01:08", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:01:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right and walk forward ten meters.", + "B": "Go straight forward ten meters.", + "C": "Turn left and walk backward about ten meters until you reach the spot you’re standing at now.", + "D": "Turn right first, then walk forward around ten meters to arrive at your current position.", + "E": "Turn left and walk forward ten meters." + }, + "answer_label": "E", + "answer_text": "Turn left and walk forward ten meters.", + "q_id": "9db30b24-9088-408f-b34d-330d0fa57749-001_7", + "correct_answer": "Turn left and walk forward ten meters.", + "original_options": { + "A": "Turn right and walk forward ten meters.", + "B": "Go straight forward ten meters.", + "C": "Turn left and walk backward about ten meters until you reach the spot you’re standing at now.", + "D": "Turn right first, then walk forward around ten meters to arrive at your current position.", + "E": "Turn left and walk forward ten meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 1, + "D": 1, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 15 + ] + }, + { + "video_uid": "9db30b24-9088-408f-b34d-330d0fa57749-001", + "video_length": "00:30:00", + "qid": "9db30b24-9088-408f-b34d-330d0fa57749-001_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get from the open space to my current position?", + "answer": "First turn left and walk forward ten meters, then turn left into the forest path and walk to the end.", + "question_chinese": "我从空地处是怎么走到当前位置的？", + "answer_chinese": "先向左转往前走10米，然后左转进树林小路走到底", + "evidence": [ + { + "description": "First turn left and walk forward 10 meters", + "timestamps": { + "start": "00:01:08", + "end": "00:01:09" + } + }, + { + "description": "turn left into the forest path and walk to the end.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:48" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right, walk ten meters, then left.", + "B": "Go straight ten meters, then turn left.", + "C": "First turn left and walk forward ten meters, then turn left into the forest path and walk to the end.", + "D": "First turn left and walk forward fifteen meters, then turn left into the forest path to the end.", + "E": "First turn left and walk forward ten meters, then turn right into the forest path and walk to the end." + }, + "answer_label": "C", + "answer_text": "First turn left and walk forward ten meters, then turn left into the forest path and walk to the end.", + "q_id": "9db30b24-9088-408f-b34d-330d0fa57749-001_8", + "correct_answer": "First turn left and walk forward ten meters, then turn left into the forest path and walk to the end.", + "original_options": { + "A": "Turn right, walk ten meters, then left.", + "B": "Go straight ten meters, then turn left.", + "C": "First turn left and walk forward ten meters, then turn left into the forest path and walk to the end.", + "D": "First turn left and walk forward fifteen meters, then turn left into the forest path to the end.", + "E": "First turn left and walk forward ten meters, then turn right into the forest path and walk to the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 2, + "D": 1, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 20 + ] + }, + { + "video_uid": "9db30b24-9088-408f-b34d-330d0fa57749-001", + "video_length": "00:30:00", + "qid": "9db30b24-9088-408f-b34d-330d0fa57749-001_1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I?", + "answer": "By the highway.", + "question_chinese": "我现在在哪？", + "answer_chinese": "公路旁", + "evidence": [ + { + "description": "by the highway", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "By the highway.", + "B": "By the basketball court.", + "C": "By the library.", + "D": "Near the old gas station where you stopped last time to check your map.", + "E": "On the edge of the city, close to the bus terminal you passed earlier." + }, + "answer_label": "A", + "answer_text": "By the highway.", + "q_id": "9db30b24-9088-408f-b34d-330d0fa57749-001_1", + "correct_answer": "By the highway.", + "original_options": { + "A": "By the highway.", + "B": "By the basketball court.", + "C": "By the library.", + "D": "Near the old gas station where you stopped last time to check your map.", + "E": "On the edge of the city, close to the bus terminal you passed earlier." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "bad", + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 0, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 14 + ] + }, + { + "video_uid": "9db30b24-9088-408f-b34d-330d0fa57749-001", + "video_length": "00:30:00", + "qid": "9db30b24-9088-408f-b34d-330d0fa57749-001_2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I?", + "answer": "On the mountain road.", + "question_chinese": "我现在在哪？", + "answer_chinese": "山路上", + "evidence": [ + { + "description": "on the mountain road", + "timestamps": { + "start": "00:03:50", + "end": "00:03:51" + } + } + ], + "question_timestamps": "00:04:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the hillside path.", + "B": "On the mountain plateau.", + "C": "Along the winding track you thought would lead to the summit, not on the main road.", + "D": "On the mountain road.", + "E": "On the mountain cliff with shrubs, overlooking the drop rather than standing on a road." + }, + "answer_label": "D", + "answer_text": "On the mountain road.", + "q_id": "9db30b24-9088-408f-b34d-330d0fa57749-001_2", + "correct_answer": "On the mountain road.", + "original_options": { + "A": "On the hillside path.", + "B": "On the mountain plateau.", + "C": "Along the winding track you thought would lead to the summit, not on the main road.", + "D": "On the mountain road.", + "E": "On the mountain cliff with shrubs, overlooking the drop rather than standing on a road." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 0, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 15 + ] + }, + { + "video_uid": "9db30b24-9088-408f-b34d-330d0fa57749-001", + "video_length": "00:30:00", + "qid": "9db30b24-9088-408f-b34d-330d0fa57749-001_6_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see at the bottom of the cliff?", + "answer": "Low vegetation, to your right.", + "question_chinese": "我看到峭壁下的是什么？", + "answer_chinese": "低矮植被，在你的右方。", + "evidence": [ + { + "description": "low-lying vegetation", + "timestamps": { + "start": "00:19:19", + "end": "00:19:20" + } + } + ], + "question_timestamps": "00:22:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Low vegetation, to your left.", + "B": "Low vegetation, to your right.", + "C": "Trees, to your right.", + "D": "Trees and low vegetation, to your front-left at the bottom of the cliff.", + "E": "Trees and low vegetation, to your back-right at the bottom of the cliff." + }, + "answer_label": "B", + "answer_text": "Low vegetation, to your right.", + "q_id": "9db30b24-9088-408f-b34d-330d0fa57749-001_6_T1", + "correct_answer": "Low vegetation, to your right.", + "original_options": { + "A": "Low vegetation, to your left.", + "B": "Low vegetation, to your right.", + "C": "Trees, to your right.", + "D": "Trees and low vegetation, to your front-left at the bottom of the cliff.", + "E": "Trees and low vegetation, to your back-right at the bottom of the cliff." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 13 + ] + }, + { + "video_uid": "9db30b24-9088-408f-b34d-330d0fa57749-001", + "video_length": "00:30:00", + "qid": "9db30b24-9088-408f-b34d-330d0fa57749-001_6_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see at the bottom of the cliff?", + "answer": "Trees and low vegetation, to your front-left.", + "question_chinese": "我看到峭壁下的是什么？", + "answer_chinese": "树和低矮植被，在你的左前方。", + "evidence": [ + { + "description": "low-lying vegetation", + "timestamps": { + "start": "00:19:19", + "end": "00:19:20" + } + }, + { + "description": "Trees", + "timestamps": { + "start": "00:23:41", + "end": "00:23:42" + } + } + ], + "question_timestamps": "00:27:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Low vegetation, to your right.", + "B": "Trees, to your front-left.", + "C": "Trees and low vegetation, to your front-left.", + "D": "Trees and low vegetation, to your front-right at the bottom of the cliff.", + "E": "Trees and low vegetation, to your back-left when you looked down from the cliff edge." + }, + "answer_label": "C", + "answer_text": "Trees and low vegetation, to your front-left.", + "q_id": "9db30b24-9088-408f-b34d-330d0fa57749-001_6_T2", + "correct_answer": "Trees and low vegetation, to your front-left.", + "original_options": { + "A": "Low vegetation, to your right.", + "B": "Trees, to your front-left.", + "C": "Trees and low vegetation, to your front-left.", + "D": "Trees and low vegetation, to your front-right at the bottom of the cliff.", + "E": "Trees and low vegetation, to your back-left when you looked down from the cliff edge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 15 + ] + }, + { + "video_uid": "9db30b24-9088-408f-b34d-330d0fa57749-001", + "video_length": "00:30:00", + "qid": "9db30b24-9088-408f-b34d-330d0fa57749-001_2_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see when I stood on this open space?", + "answer": "Six.", + "question_chinese": "我站在这块空地上看见了多少个人？", + "answer_chinese": "6", + "evidence": [ + { + "description": "5 people", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "the sixth person", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Five.", + "C": "Four people were visible in the open space while you stood there, spread out across the clearing.", + "D": "Eight people were visible in the open space while you stood there, moving around in front of you.", + "E": "Six." + }, + "answer_label": "E", + "answer_text": "Six.", + "q_id": "9db30b24-9088-408f-b34d-330d0fa57749-001_2_T1", + "correct_answer": "Six.", + "original_options": { + "A": "Two.", + "B": "Five.", + "C": "Four people were visible in the open space while you stood there, spread out across the clearing.", + "D": "Eight people were visible in the open space while you stood there, moving around in front of you.", + "E": "Six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 18 + ] + }, + { + "video_uid": "9db30b24-9088-408f-b34d-330d0fa57749-001", + "video_length": "00:30:00", + "qid": "9db30b24-9088-408f-b34d-330d0fa57749-001_2_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see when I stood on this open space?", + "answer": "6.", + "question_chinese": "我站在这块空地上看见了多少个人？", + "answer_chinese": "6", + "evidence": [ + { + "description": "5 people", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "the sixth person", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "5.", + "B": "7.", + "C": "6.", + "D": "I saw four people standing in the open space near me there.", + "E": "I saw eight people spread out across the open space as I stood there." + }, + "answer_label": "C", + "answer_text": "6.", + "q_id": "9db30b24-9088-408f-b34d-330d0fa57749-001_2_T2", + "correct_answer": "6.", + "original_options": { + "A": "5.", + "B": "7.", + "C": "6.", + "D": "I saw four people standing in the open space near me there.", + "E": "I saw eight people spread out across the open space as I stood there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "9db30b24-9088-408f-b34d-330d0fa57749-001", + "video_length": "00:30:00", + "qid": "9db30b24-9088-408f-b34d-330d0fa57749-001_10", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see before walking into the forest?", + "answer": "Trees.", + "question_chinese": "走进树林前我看到了什么？", + "answer_chinese": "树", + "evidence": [ + { + "description": "trees", + "timestamps": { + "start": "00:01:08", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Bushes.", + "B": "Trees.", + "C": "Rocks.", + "D": "A small clearing right before stepping into the forest entrance from the path.", + "E": "A narrow path leading toward the forest entrance before I went under the treeline." + }, + "answer_label": "B", + "answer_text": "Trees.", + "q_id": "9db30b24-9088-408f-b34d-330d0fa57749-001_10", + "correct_answer": "Trees.", + "original_options": { + "A": "Bushes.", + "B": "Trees.", + "C": "Rocks.", + "D": "A small clearing right before stepping into the forest entrance from the path.", + "E": "A narrow path leading toward the forest entrance before I went under the treeline." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "9db30b24-9088-408f-b34d-330d0fa57749-001", + "video_length": "00:30:00", + "qid": "9db30b24-9088-408f-b34d-330d0fa57749-001_9", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see before walking into the forest?", + "answer": "Road sign and trees.", + "question_chinese": "走进树林前我看到了什么？", + "answer_chinese": "树和路牌", + "evidence": [ + { + "description": "trees", + "timestamps": { + "start": "00:01:08", + "end": "00:01:09" + } + }, + { + "description": "road sign", + "timestamps": { + "start": "00:01:16", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Stop sign and ferns.", + "B": "Yield sign and rocks.", + "C": "A trail map sign posted beside dense trees at the forest entrance area.", + "D": "Road sign and trees.", + "E": "A metal warning sign near a dirt path lined with trees by the entrance." + }, + "answer_label": "D", + "answer_text": "Road sign and trees.", + "q_id": "9db30b24-9088-408f-b34d-330d0fa57749-001_9", + "correct_answer": "Road sign and trees.", + "original_options": { + "A": "Stop sign and ferns.", + "B": "Yield sign and rocks.", + "C": "A trail map sign posted beside dense trees at the forest entrance area.", + "D": "Road sign and trees.", + "E": "A metal warning sign near a dirt path lined with trees by the entrance." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 2, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + } +] diff --git a/QAs/9e168f5a-cbbb-4065-850d-2d98bd74b7cf-mcq.json b/QAs/9e168f5a-cbbb-4065-850d-2d98bd74b7cf-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1872bb965767b571c0260446961433dd5782eb74 --- /dev/null +++ b/QAs/9e168f5a-cbbb-4065-850d-2d98bd74b7cf-mcq.json @@ -0,0 +1,1266 @@ +[ + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_7_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to put the green sanding block back into its box now, how would I walk?", + "answer": "You will move to your front-left, walk over to the stool, and put the green sanding block into the box. The box is to your front-left.", + "question_chinese": "如果我现在要把绿色打磨块放回装它的盒子里，我将会怎么走？", + "answer_chinese": "你将会向左前方移动，走到凳子旁，将绿色打磨块放进盒子里。盒子在你左前方。", + "evidence": [ + { + "description": "You put the box containing the abrasive blocks on the stool.", + "timestamps": { + "start": "00:02:50", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:10:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Move to your front-right to the stool.", + "B": "Move to your back-left to the stool.", + "C": "You will step forward-right, walk to the stool, and place the green sanding block into the box.", + "D": "You will move to your front-left, walk over to the stool, and put the green sanding block into the box. The box is to your front-left.", + "E": "You will move to your front-left, walk past the stool, and put the green sanding block into a box behind it." + }, + "answer_label": "D", + "answer_text": "You will move to your front-left, walk over to the stool, and put the green sanding block into the box. The box is to your front-left.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_7_T1", + "correct_answer": "You will move to your front-left, walk over to the stool, and put the green sanding block into the box. The box is to your front-left.", + "original_options": { + "A": "Move to your front-right to the stool.", + "B": "Move to your back-left to the stool.", + "C": "You will step forward-right, walk to the stool, and place the green sanding block into the box.", + "D": "You will move to your front-left, walk over to the stool, and put the green sanding block into the box. The box is to your front-left.", + "E": "You will move to your front-left, walk past the stool, and put the green sanding block into a box behind it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 21 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_7_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to put the green sanding block back into its box now, how would I walk?", + "answer": "You will move left, walk to the stool, and put the green sandpaper block into the box.", + "question_chinese": "如果我现在要把绿色打磨块放回装它的盒子里，我将会怎么走？", + "answer_chinese": "你将向左移动，走到凳子旁边，把绿色的砂纸块放进盒子里。", + "evidence": [ + { + "description": "You put the box containing the abrasive blocks on the stool.", + "timestamps": { + "start": "00:02:50", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:20:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You will move right to the stool.", + "B": "You will move forward-left to the stool.", + "C": "You will move left, walk to the stool, and put the green sandpaper block into the box.", + "D": "You will move left, walk to the stool, and set the green sandpaper block beside the box.", + "E": "You will walk left to the corner and take a sanding block from the box on the wooden shelf." + }, + "answer_label": "C", + "answer_text": "You will move left, walk to the stool, and put the green sandpaper block into the box.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_7_T2", + "correct_answer": "You will move left, walk to the stool, and put the green sandpaper block into the box.", + "original_options": { + "A": "You will move right to the stool.", + "B": "You will move forward-left to the stool.", + "C": "You will move left, walk to the stool, and put the green sandpaper block into the box.", + "D": "You will move left, walk to the stool, and set the green sandpaper block beside the box.", + "E": "You will walk left to the corner and take a sanding block from the box on the wooden shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 19 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_2_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I need more sanding blocks, how would I walk to get them?", + "answer": "You will walk left to the corner and take a sanding block from the box on the wooden shelf.", + "question_chinese": "如果我需要更多的打磨块，我将会怎么走？", + "answer_chinese": "你将往左走，走到角落，在木架上的盒子里拿到打磨块", + "evidence": [ + { + "description": "You walk to the right rear, go to the side of the bench, place the box containing the abrasive blocks on the bench, and then take out some of the abrasive blocks.", + "timestamps": { + "start": "00:02:45", + "end": "00:02:58" + } + }, + { + "description": "The metal block is in the box on the wooden shelf.", + "timestamps": { + "start": "00:02:59", + "end": "00:03:05" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk right to the corner shelf.", + "B": "Take a sanding block from the stool box.", + "C": "You will walk left to the corner and take a sanding block from the box on the stool.", + "D": "You will head to your front-left, go around the metal table, and grab sanding blocks from the stool.", + "E": "You will walk left to the corner and take a sanding block from the box on the wooden shelf." + }, + "answer_label": "E", + "answer_text": "You will walk left to the corner and take a sanding block from the box on the wooden shelf.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_2_T1", + "correct_answer": "You will walk left to the corner and take a sanding block from the box on the wooden shelf.", + "original_options": { + "A": "Walk right to the corner shelf.", + "B": "Take a sanding block from the stool box.", + "C": "You will walk left to the corner and take a sanding block from the box on the stool.", + "D": "You will head to your front-left, go around the metal table, and grab sanding blocks from the stool.", + "E": "You will walk left to the corner and take a sanding block from the box on the wooden shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 18, + 18 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_3", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I need more crafting tables, how will I go about it?", + "answer": "You will walk forward to the left, go around the metal table that is not yet polished, walk to the side of the stool, and take the polishing block from the box on the stool.", + "question_chinese": "如果我需要更多的打磨块，我将会怎么走？", + "answer_chinese": "你将会往左前方走，绕过尚未打磨完成的金属桌，走到凳子旁，从凳上的盒子里取出打磨块。", + "evidence": [ + { + "description": "You walk to the right rear, go to the side of the bench, place the box containing the abrasive blocks on the bench, and then take out some of the abrasive blocks.", + "timestamps": { + "start": "00:02:45", + "end": "00:03:09" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk forward to your right.", + "B": "You will walk forward to the left, go around the metal table that is not yet polished, walk to the side of the stool, and take the polishing block from the box on the stool.", + "C": "Walk straight forward past the table.", + "D": "You will walk forward to the right, go around the unpolished metal table, then take the polishing block from the stool box.", + "E": "You will walk forward-left, go around the metal table, and take the polishing block from a box on the wooden shelf." + }, + "answer_label": "B", + "answer_text": "You will walk forward to the left, go around the metal table that is not yet polished, walk to the side of the stool, and take the polishing block from the box on the stool.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_3", + "correct_answer": "You will walk forward to the left, go around the metal table that is not yet polished, walk to the side of the stool, and take the polishing block from the box on the stool.", + "original_options": { + "A": "Walk forward to your right.", + "B": "You will walk forward to the left, go around the metal table that is not yet polished, walk to the side of the stool, and take the polishing block from the box on the stool.", + "C": "Walk straight forward past the table.", + "D": "You will walk forward to the right, go around the unpolished metal table, then take the polishing block from the stool box.", + "E": "You will walk forward-left, go around the metal table, and take the polishing block from a box on the wooden shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 22, + 21 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_2_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I need more sanding blocks, how would I walk to get them?", + "answer": "You will walk to your front-left, go around the metal table that hasn’t been fully polished yet, walk over to the stool, and take a sanding block out of the box on the stool.", + "question_chinese": "如果我需要更多的打磨块，我将会怎么走？", + "answer_chinese": "你将会往左前方走，绕过尚未打磨完成的金属桌，走到凳子旁，从凳上的盒子里取出打磨块。", + "evidence": [ + { + "description": "You walk to the right rear, go to the side of the bench, place the box containing the abrasive blocks on the bench, and then take out some of the abrasive blocks.", + "timestamps": { + "start": "00:02:45", + "end": "00:03:09" + } + } + ], + "question_timestamps": "00:20:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk to your front-right to the stool.", + "B": "Walk to your back-left to the stool.", + "C": "You will walk to your front-left, go around the metal table, and take a sanding block from the wooden shelf.", + "D": "You will walk to your front-left, go around the metal table that hasn’t been fully polished yet, walk over to the stool, and take a sanding block out of the box on the stool.", + "E": "You will walk left to the corner, then take a sanding block from the box on the wooden shelf." + }, + "answer_label": "D", + "answer_text": "You will walk to your front-left, go around the metal table that hasn’t been fully polished yet, walk over to the stool, and take a sanding block out of the box on the stool.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_2_T2", + "correct_answer": "You will walk to your front-left, go around the metal table that hasn’t been fully polished yet, walk over to the stool, and take a sanding block out of the box on the stool.", + "original_options": { + "A": "Walk to your front-right to the stool.", + "B": "Walk to your back-left to the stool.", + "C": "You will walk to your front-left, go around the metal table, and take a sanding block from the wooden shelf.", + "D": "You will walk to your front-left, go around the metal table that hasn’t been fully polished yet, walk over to the stool, and take a sanding block out of the box on the stool.", + "E": "You will walk left to the corner, then take a sanding block from the box on the wooden shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 19 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_7", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the utility knife?", + "answer": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "question_chinese": "工具刀放在哪里了？", + "answer_chinese": "工具刀被放在了蓝色桶所在的工作台上，旁边有一个白桶。", + "evidence": [ + { + "description": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "timestamps": { + "start": "00:08:00", + "end": "00:08:10" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the shelf behind you.", + "B": "On the workbench by the white bucket.", + "C": "The utility knife is on a shelf behind the workbench, near the blue bucket with a white bucket alongside.", + "D": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "E": "The utility knife is under the workbench near the blue bucket, with the white bucket next to it." + }, + "answer_label": "D", + "answer_text": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_7", + "correct_answer": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "original_options": { + "A": "On the shelf behind you.", + "B": "On the workbench by the white bucket.", + "C": "The utility knife is on a shelf behind the workbench, near the blue bucket with a white bucket alongside.", + "D": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "E": "The utility knife is under the workbench near the blue bucket, with the white bucket next to it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 19, + 18 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_8", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the utility knife?", + "answer": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "question_chinese": "工具刀放在哪里了？", + "answer_chinese": "工具刀被放在了蓝色桶所在的工作台上，旁边有一个白桶。", + "evidence": [ + { + "description": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "timestamps": { + "start": "00:08:00", + "end": "00:08:10" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the workbench by the white bucket.", + "B": "On the shelf behind the workbench.", + "C": "The utility knife is on the workbench where the white bucket is located, with the blue bucket next to it.", + "D": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "E": "The utility knife sits on the workbench with the blue bucket, but the white bucket is on the other side." + }, + "answer_label": "D", + "answer_text": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_8", + "correct_answer": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "original_options": { + "A": "On the workbench by the white bucket.", + "B": "On the shelf behind the workbench.", + "C": "The utility knife is on the workbench where the white bucket is located, with the blue bucket next to it.", + "D": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "E": "The utility knife sits on the workbench with the blue bucket, but the white bucket is on the other side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 20 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_8_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What colors of sanding blocks did I take out of the box that holds the sanding blocks?", + "answer": "You took out the green sanding block. The box is on the stool in front of you.", + "question_chinese": "我从装打磨块的盒子里拿出了哪几种颜色的打磨块？", + "answer_chinese": "你拿出了绿色的打磨块。，盒子在你面前的凳子上。", + "evidence": [ + { + "description": "You took out green sanding block.", + "timestamps": { + "start": "00:02:54", + "end": "00:02:58" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took out the yellow sanding block.", + "B": "You took out the green sanding block. The box is on the stool in front of you.", + "C": "You took out the blue sanding block.", + "D": "You took out the green and yellow sanding blocks from the box on the stool in front of you.", + "E": "You took out the green sanding block, but the box was on the stool behind you, not in front." + }, + "answer_label": "B", + "answer_text": "You took out the green sanding block. The box is on the stool in front of you.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_8_T1", + "correct_answer": "You took out the green sanding block. The box is on the stool in front of you.", + "original_options": { + "A": "You took out the yellow sanding block.", + "B": "You took out the green sanding block. The box is on the stool in front of you.", + "C": "You took out the blue sanding block.", + "D": "You took out the green and yellow sanding blocks from the box on the stool in front of you.", + "E": "You took out the green sanding block, but the box was on the stool behind you, not in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_8_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What colors of sanding blocks did I take out of the box that holds the sanding blocks?", + "answer": "You took the green and the yellow sanding blocks out of the box that holds the sanding blocks. That box is on the stool in front of you.", + "question_chinese": "我从装打磨块的盒子里拿出了哪几种颜色的打磨块？", + "answer_chinese": "你从装打磨块的盒子里拿出了绿色的和黄色的打磨块。装打磨块的盒子在你面前的凳子上。", + "evidence": [ + { + "description": "You took out green and yellow sanding blocks.", + "timestamps": { + "start": "00:02:54", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took out only the green block.", + "B": "You took out only the yellow block.", + "C": "You took the green and the yellow sanding blocks out of the box that holds the sanding blocks. That box is on the stool in front of you.", + "D": "You took out the green and yellow sanding blocks, but the box was on the stool behind you.", + "E": "You took out the green and yellow sanding blocks, but the box was on the stool to your left." + }, + "answer_label": "C", + "answer_text": "You took the green and the yellow sanding blocks out of the box that holds the sanding blocks. That box is on the stool in front of you.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_8_T2", + "correct_answer": "You took the green and the yellow sanding blocks out of the box that holds the sanding blocks. That box is on the stool in front of you.", + "original_options": { + "A": "You took out only the green block.", + "B": "You took out only the yellow block.", + "C": "You took the green and the yellow sanding blocks out of the box that holds the sanding blocks. That box is on the stool in front of you.", + "D": "You took out the green and yellow sanding blocks, but the box was on the stool behind you.", + "E": "You took out the green and yellow sanding blocks, but the box was on the stool to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 19 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_16", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which colors of sanding blocks did I take out from the box containing sanding blocks?", + "answer": "You took out green and yellow sanding blocks.", + "question_chinese": "我从装打磨块的盒子里拿出了哪几种颜色的打磨块？", + "answer_chinese": "你拿出了绿色的和黄色的打磨块。", + "evidence": [ + { + "description": "You took out green and yellow sanding blocks.", + "timestamps": { + "start": "00:02:54", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:20:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took out blue and orange blocks.", + "B": "You took out red and purple blocks.", + "C": "You took out green and blue sanding blocks from the box containing the sanding blocks.", + "D": "You took out yellow and black sanding blocks from the box containing sanding blocks.", + "E": "You took out green and yellow sanding blocks." + }, + "answer_label": "E", + "answer_text": "You took out green and yellow sanding blocks.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_16", + "correct_answer": "You took out green and yellow sanding blocks.", + "original_options": { + "A": "You took out blue and orange blocks.", + "B": "You took out red and purple blocks.", + "C": "You took out green and blue sanding blocks from the box containing the sanding blocks.", + "D": "You took out yellow and black sanding blocks from the box containing sanding blocks.", + "E": "You took out green and yellow sanding blocks." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 14 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_4_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the utility knife?", + "answer": "You put the utility knife on the workbench where you tossed the gloves and glasses. The workbench is in front of you.", + "question_chinese": "我把工具刀放在哪里了？", + "answer_chinese": "你把工具刀放在了你丢手套和眼镜的工作台上。工作台在你面前。", + "evidence": [ + { + "description": "The utility knife is on the table in front of you.", + "timestamps": { + "start": "00:10:52", + "end": "00:10:58" + } + } + ], + "question_timestamps": "00:11:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put the utility knife on the workbench where you tossed the gloves and glasses. The workbench is in front of you.", + "B": "On the workbench behind you.", + "C": "On the workbench to your left.", + "D": "You put the utility knife on the workbench where you tossed the gloves and glasses, but it was to your right.", + "E": "You placed the utility knife on the workbench where the gloves and glasses were, with the workbench behind you." + }, + "answer_label": "A", + "answer_text": "You put the utility knife on the workbench where you tossed the gloves and glasses. The workbench is in front of you.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_4_T1", + "correct_answer": "You put the utility knife on the workbench where you tossed the gloves and glasses. The workbench is in front of you.", + "original_options": { + "A": "You put the utility knife on the workbench where you tossed the gloves and glasses. The workbench is in front of you.", + "B": "On the workbench behind you.", + "C": "On the workbench to your left.", + "D": "You put the utility knife on the workbench where you tossed the gloves and glasses, but it was to your right.", + "E": "You placed the utility knife on the workbench where the gloves and glasses were, with the workbench behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 21, + 19 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_4_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the utility knife?", + "answer": "The utility knife was placed on the workbench with the blue bucket, next to a white bucket. The workbench is to your front-left.", + "question_chinese": "我把工具刀放在哪里了？", + "answer_chinese": "工具刀被放在了蓝色桶所在的工作台上，旁边有一个白桶。工作台在你左前方。", + "evidence": [ + { + "description": "The utility knife is placed on the workbench where the blue bucket is located, with a white bucket next to it.", + "timestamps": { + "start": "00:08:00", + "end": "00:08:10" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the workbench to your front-right.", + "B": "On the workbench to your back-left.", + "C": "The utility knife was placed on the workbench with the blue bucket, next to a white bucket, to your back-right.", + "D": "The utility knife was placed on the workbench with the blue bucket, next to a white bucket. The workbench is to your front-left.", + "E": "The utility knife was placed on the workbench with the blue bucket, but the workbench was directly in front of you." + }, + "answer_label": "D", + "answer_text": "The utility knife was placed on the workbench with the blue bucket, next to a white bucket. The workbench is to your front-left.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_4_T2", + "correct_answer": "The utility knife was placed on the workbench with the blue bucket, next to a white bucket. The workbench is to your front-left.", + "original_options": { + "A": "On the workbench to your front-right.", + "B": "On the workbench to your back-left.", + "C": "The utility knife was placed on the workbench with the blue bucket, next to a white bucket, to your back-right.", + "D": "The utility knife was placed on the workbench with the blue bucket, next to a white bucket. The workbench is to your front-left.", + "E": "The utility knife was placed on the workbench with the blue bucket, but the workbench was directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 21 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_5", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pairs of gloves have I used?", + "answer": "You used a pair of gloves.", + "question_chinese": "我用过了多少双手套？", + "answer_chinese": "你用了一双手套。", + "evidence": [ + { + "description": "These are the first pair of gloves you used.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:31" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You used a pair of gloves.", + "B": "You used two pairs of gloves.", + "C": "You used three pairs of gloves.", + "D": "You used four pairs of gloves during the task, switching pairs several times as you worked.", + "E": "You used five pairs of gloves in total, going through multiple pairs over the course of the work." + }, + "answer_label": "A", + "answer_text": "You used a pair of gloves.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_5", + "correct_answer": "You used a pair of gloves.", + "original_options": { + "A": "You used a pair of gloves.", + "B": "You used two pairs of gloves.", + "C": "You used three pairs of gloves.", + "D": "You used four pairs of gloves during the task, switching pairs several times as you worked.", + "E": "You used five pairs of gloves in total, going through multiple pairs over the course of the work." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_6", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pairs of gloves have I used?", + "answer": "You have used two pairs of gloves.", + "question_chinese": "我用过了多少双手套？", + "answer_chinese": "你用过了两双手套。", + "evidence": [ + { + "description": "These are the second pair of gloves you have used.", + "timestamps": { + "start": "00:10:57", + "end": "00:11:38" + } + }, + { + "description": "These are the first pair of gloves you've used.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:59" + } + } + ], + "question_timestamps": "00:20:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have used one pair.", + "B": "You have used three pairs.", + "C": "You have used two pairs of gloves.", + "D": "You have used four pairs of gloves during the task, switching to a new pair multiple times.", + "E": "You have used no pairs of gloves at all, completing the work without putting any on." + }, + "answer_label": "C", + "answer_text": "You have used two pairs of gloves.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_6", + "correct_answer": "You have used two pairs of gloves.", + "original_options": { + "A": "You have used one pair.", + "B": "You have used three pairs.", + "C": "You have used two pairs of gloves.", + "D": "You have used four pairs of gloves during the task, switching to a new pair multiple times.", + "E": "You have used no pairs of gloves at all, completing the work without putting any on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_1_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I first see the electric grinder?", + "answer": "At the very beginning, the electric grinder was on top of the white bag labeled “Microtek,” which was behind the blue transport rack. The electric grinder is to your front-left.", + "question_chinese": "我看见电动打磨机最开始在哪里？", + "answer_chinese": "电动打磨机最开始在蓝色搬运架后面的印着Microtek的白袋子上面。电动打磨机在你左前方。", + "evidence": [ + { + "description": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "At the very beginning, the electric grinder was on top of the white bag labeled “Microtek,” which was behind the blue transport rack. The electric grinder is to your front-left.", + "B": "It was to your front-right.", + "C": "It was to your back-left.", + "D": "At the beginning, it sat on the Microtek white bag beside the blue transport rack, to your front-right.", + "E": "At the very beginning, it was on the floor beside the Microtek white bag behind the blue transport rack." + }, + "answer_label": "A", + "answer_text": "At the very beginning, the electric grinder was on top of the white bag labeled “Microtek,” which was behind the blue transport rack. The electric grinder is to your front-left.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_1_T1", + "correct_answer": "At the very beginning, the electric grinder was on top of the white bag labeled “Microtek,” which was behind the blue transport rack. The electric grinder is to your front-left.", + "original_options": { + "A": "At the very beginning, the electric grinder was on top of the white bag labeled “Microtek,” which was behind the blue transport rack. The electric grinder is to your front-left.", + "B": "It was to your front-right.", + "C": "It was to your back-left.", + "D": "At the beginning, it sat on the Microtek white bag beside the blue transport rack, to your front-right.", + "E": "At the very beginning, it was on the floor beside the Microtek white bag behind the blue transport rack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 19 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_1_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I first see the electric grinder?", + "answer": "At the start, the electric grinder was behind the white bag labeled “Microtek,” and that bag was behind the blue cart.", + "question_chinese": "我看见电动打磨机最开始在哪里？", + "answer_chinese": "电动研磨机最初位于印有“Microtek”的白色袋子后面，而该袋子在蓝色推车的后面。", + "evidence": [ + { + "description": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:39:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "At the start, the electric grinder was behind the white bag labeled “Microtek,” and that bag was behind the blue cart.", + "B": "It was in front of the Microtek bag.", + "C": "It was on top of the bag.", + "D": "At the start, the electric grinder was in front of the white “Microtek” bag, with the bag still behind the blue cart.", + "E": "Initially, the grinder was on top of the Microtek bag, and that bag was positioned behind the blue transport rack." + }, + "answer_label": "A", + "answer_text": "At the start, the electric grinder was behind the white bag labeled “Microtek,” and that bag was behind the blue cart.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_1_T2", + "correct_answer": "At the start, the electric grinder was behind the white bag labeled “Microtek,” and that bag was behind the blue cart.", + "original_options": { + "A": "At the start, the electric grinder was behind the white bag labeled “Microtek,” and that bag was behind the blue cart.", + "B": "It was in front of the Microtek bag.", + "C": "It was on top of the bag.", + "D": "At the start, the electric grinder was in front of the white “Microtek” bag, with the bag still behind the blue cart.", + "E": "Initially, the grinder was on top of the Microtek bag, and that bag was positioned behind the blue transport rack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 22, + 20 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_5_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the phone that I saw first?", + "answer": "It’s on the workbench with the blue bucket, which is to your front-left of the table you’re currently touching. That workbench is in front of you.", + "question_chinese": "我第一次看到的手机在哪里？", + "answer_chinese": "它在你当前触碰的桌子的左前方那个放着蓝色桶子的工作台上，工作台在你前面。", + "evidence": [ + { + "description": "The phone is on the table with the blue bucket.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on the workbench with the blue bucket, which is to your front-left of the table you’re currently touching. That workbench is in front of you.", + "B": "It was on the workbench behind you.", + "C": "It was on the workbench to your right.", + "D": "It’s on the workbench with the blue bucket, front-left of the table you’re touching, but behind you.", + "E": "It’s on the workbench with the blue bucket, front-left of the table you’re touching, but to your right." + }, + "answer_label": "A", + "answer_text": "It’s on the workbench with the blue bucket, which is to your front-left of the table you’re currently touching. That workbench is in front of you.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_5_T2", + "correct_answer": "It’s on the workbench with the blue bucket, which is to your front-left of the table you’re currently touching. That workbench is in front of you.", + "original_options": { + "A": "It’s on the workbench with the blue bucket, which is to your front-left of the table you’re currently touching. That workbench is in front of you.", + "B": "It was on the workbench behind you.", + "C": "It was on the workbench to your right.", + "D": "It’s on the workbench with the blue bucket, front-left of the table you’re touching, but behind you.", + "E": "It’s on the workbench with the blue bucket, front-left of the table you’re touching, but to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 20 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_5_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the phone that I saw first?", + "answer": "It’s on the workbench covered with items, right next to the blue bucket. The workbench is to your front-left.", + "question_chinese": "我第一次看到的手机在哪里？", + "answer_chinese": "它在摆满物品的工作台上，挨着蓝色的桶。工作台在你左前方。", + "evidence": [ + { + "description": "It is on the workbench filled with items, next to the blue bucket.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:20:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the front-right workbench.", + "B": "Far from the blue bucket.", + "C": "It’s on the workbench covered with items, right next to the blue bucket. The workbench is to your front-left.", + "D": "It’s on the workbench covered with items, but it’s on the opposite side of the blue bucket, not beside it.", + "E": "It’s on the cluttered workbench next to the blue bucket, but the workbench is positioned to your back-left." + }, + "answer_label": "C", + "answer_text": "It’s on the workbench covered with items, right next to the blue bucket. The workbench is to your front-left.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_5_T1", + "correct_answer": "It’s on the workbench covered with items, right next to the blue bucket. The workbench is to your front-left.", + "original_options": { + "A": "On the front-right workbench.", + "B": "Far from the blue bucket.", + "C": "It’s on the workbench covered with items, right next to the blue bucket. The workbench is to your front-left.", + "D": "It’s on the workbench covered with items, but it’s on the opposite side of the blue bucket, not beside it.", + "E": "It’s on the cluttered workbench next to the blue bucket, but the workbench is positioned to your back-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 22, + 19 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_1", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did the electric grinder originate from?", + "answer": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "question_chinese": "电动打磨机最开始在哪里？", + "answer_chinese": "电动打磨机最开始在蓝色搬运架后面的印着Microtek的白袋子上面。", + "evidence": [ + { + "description": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:30:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the Microtek bag in front.", + "B": "On a red Microtek bag behind it.", + "C": "The electric grinder started on the white “Microtek” bag, but the bag was in front of the blue trolley.", + "D": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "E": "The electric grinder originally came from a Microtek-labeled bag, but it was positioned beside the blue trolley, not behind." + }, + "answer_label": "D", + "answer_text": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_1", + "correct_answer": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "original_options": { + "A": "On the Microtek bag in front.", + "B": "On a red Microtek bag behind it.", + "C": "The electric grinder started on the white “Microtek” bag, but the bag was in front of the blue trolley.", + "D": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "E": "The electric grinder originally came from a Microtek-labeled bag, but it was positioned beside the blue trolley, not behind." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 19 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did the electric grinder originate from?", + "answer": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "question_chinese": "电动打磨机最开始在哪里？", + "answer_chinese": "电动研磨机最初位于印有“Microtek”的白色袋子后面，而该袋子在蓝色推车的后面。", + "evidence": [ + { + "description": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:39:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On a black Microtek bag.", + "B": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "C": "In front of the blue trolley.", + "D": "The electric grinder was first placed on the white “Microtek” bag, but it was beside a red trolley.", + "E": "The electric grinder began on a plain white bag without the “Microtek” print, located behind the blue trolley." + }, + "answer_label": "B", + "answer_text": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_2", + "correct_answer": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "original_options": { + "A": "On a black Microtek bag.", + "B": "The electric grinder was initially on the white bag printed with 'Microtek' behind the blue trolley.", + "C": "In front of the blue trolley.", + "D": "The electric grinder was first placed on the white “Microtek” bag, but it was beside a red trolley.", + "E": "The electric grinder began on a plain white bag without the “Microtek” print, located behind the blue trolley." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 18 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_6_T1", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the green sanding block or the phone?", + "answer": "The phone is closer to you. The phone is on the workbench to your right.", + "question_chinese": "绿色打磨块和手机哪个离我更近？", + "answer_chinese": "手机离你更近。手机在你右边的工作台上。", + "evidence": [ + { + "description": "The phone is on the table with the blue bucket.", + "timestamps": { + "start": "00:10:52", + "end": "00:10:55" + } + } + ], + "question_timestamps": "00:10:57", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The phone is closer to you. The phone is on the workbench to your right.", + "B": "The green sanding block is closer to you. The phone is farther away on the workbench to your right." + }, + "answer_label": "A", + "answer_text": "The phone is closer to you. The phone is on the workbench to your right.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_6_T1", + "correct_answer": "The phone is closer to you. The phone is on the workbench to your right.", + "original_options": { + "A": "The phone is closer to you. The phone is on the workbench to your right.", + "B": "The green sanding block is closer to you. The phone is farther away on the workbench to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 19 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_12", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the green abrasive block or the mobile phone?", + "answer": "The green grinding block is closer to you.", + "question_chinese": "绿色打磨块和手机哪个离我更近？", + "answer_chinese": "绿色打磨块离你更近。", + "evidence": [ + { + "description": "The mobile phone is on the workbench.", + "timestamps": { + "start": "00:08:05", + "end": "00:08:09" + } + }, + { + "description": "A green grinding block is on the metal plate.", + "timestamps": { + "start": "00:10:45", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:11:18", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The mobile phone is closer to you.", + "B": "The green grinding block is closer to you." + }, + "answer_label": "B", + "answer_text": "The green grinding block is closer to you.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_12", + "correct_answer": "The green grinding block is closer to you.", + "original_options": { + "A": "The mobile phone is closer to you.", + "B": "The green grinding block is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_11", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the green abrasive block or the mobile phone?", + "answer": "The green grinding block is closer to you.", + "question_chinese": "绿色打磨块和手机哪个离我更近？", + "answer_chinese": "绿色打磨块离你更近。", + "evidence": [ + { + "description": "The green abrasive block is on the metal product in front of you.", + "timestamps": { + "start": "00:19:44", + "end": "00:19:51" + } + }, + { + "description": "The mobile phone is on the workbench, next to the blue bucket.", + "timestamps": { + "start": "00:16:01", + "end": "00:16:05" + } + } + ], + "question_timestamps": "00:19:52", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The mobile phone is closer to you.", + "B": "The green grinding block is closer to you." + }, + "answer_label": "B", + "answer_text": "The green grinding block is closer to you.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_11", + "correct_answer": "The green grinding block is closer to you.", + "original_options": { + "A": "The mobile phone is closer to you.", + "B": "The green grinding block is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf", + "video_length": "00:39:27", + "qid": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_6_T2", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the green sanding block or the phone?", + "answer": "The green sanding block is closer to you. The green sanding block is on the right side of the workbench in front of you.", + "question_chinese": "绿色打磨块和手机哪个离我更近？", + "answer_chinese": "绿色打磨块离你更近。绿色打磨块在你面前工作台的右边。", + "evidence": [ + { + "description": "The green abrasive block is on the metal product in front of you.", + "timestamps": { + "start": "00:19:44", + "end": "00:19:51" + } + } + ], + "question_timestamps": "00:19:53", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The phone is closer to you. The phone is on the right side of the workbench in front of you.", + "B": "The green sanding block is closer to you. The green sanding block is on the right side of the workbench in front of you." + }, + "answer_label": "B", + "answer_text": "The green sanding block is closer to you. The green sanding block is on the right side of the workbench in front of you.", + "q_id": "9e168f5a-cbbb-4065-850d-2d98bd74b7cf_6_T2", + "correct_answer": "The green sanding block is closer to you. The green sanding block is on the right side of the workbench in front of you.", + "original_options": { + "A": "The phone is closer to you. The phone is on the right side of the workbench in front of you.", + "B": "The green sanding block is closer to you. The green sanding block is on the right side of the workbench in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + } +] diff --git a/QAs/9e76a664-9014-4630-94dd-26b3b2594576-mcq.json b/QAs/9e76a664-9014-4630-94dd-26b3b2594576-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..baf2d2a551d3ef741ae07d871f866b4d4f80ba39 --- /dev/null +++ b/QAs/9e76a664-9014-4630-94dd-26b3b2594576-mcq.json @@ -0,0 +1,567 @@ +[ + { + "video_uid": "9e76a664-9014-4630-94dd-26b3b2594576", + "video_length": "00:30:00", + "qid": "9e76a664-9014-4630-94dd-26b3b2594576_4_T2", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the knife on the table next to me?", + "answer": "You can reach the knife on the table next to you on your right.", + "question_chinese": "我可以拿到旁边桌子上的刀吗？", + "answer_chinese": "你可以拿到你右边的旁边桌子上的刀。", + "evidence": [ + { + "description": "The knife was on the table next to me.", + "timestamps": { + "start": "00:13:47", + "end": "00:13:53" + } + } + ], + "question_timestamps": "00:13:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can reach the knife on the table next to you on your right.", + "B": "No, you can’t reach it.", + "C": "Yes, on the left side.", + "D": "You can’t reach the knife on the table next to you; it’s just out of reach in front of you.", + "E": "You can reach the knife, but it’s positioned more to your front-left, not right beside you." + }, + "answer_label": "A", + "answer_text": "You can reach the knife on the table next to you on your right.", + "q_id": "9e76a664-9014-4630-94dd-26b3b2594576_4_T2", + "correct_answer": "You can reach the knife on the table next to you on your right.", + "original_options": { + "A": "You can reach the knife on the table next to you on your right.", + "B": "No, you can’t reach it.", + "C": "Yes, on the left side.", + "D": "You can’t reach the knife on the table next to you; it’s just out of reach in front of you.", + "E": "You can reach the knife, but it’s positioned more to your front-left, not right beside you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 22, + 17 + ] + }, + { + "video_uid": "9e76a664-9014-4630-94dd-26b3b2594576", + "video_length": "00:30:00", + "qid": "9e76a664-9014-4630-94dd-26b3b2594576_4_T1", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the knife on the table next to me?", + "answer": "You can’t reach the knife on the table next to it. Right now the knife is to your front-right.", + "question_chinese": "我可以拿到旁边桌子上的刀吗？", + "answer_chinese": "不可以拿到旁边桌子上的刀。目前刀在你的右前方。", + "evidence": [ + { + "description": "The knife was on the table next to me.", + "timestamps": { + "start": "00:14:44", + "end": "00:14:52" + } + }, + { + "description": "The knife was on the table next to me.", + "timestamps": { + "start": "00:13:47", + "end": "00:13:53" + } + } + ], + "question_timestamps": "00:19:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes, you can reach it.", + "B": "No, it’s to your front-left.", + "C": "You can’t reach the knife on the table next to it; right now the knife is positioned to your back-right.", + "D": "You can’t reach the knife on the table next to it; the knife is currently behind you on the left.", + "E": "You can’t reach the knife on the table next to it. Right now the knife is to your front-right." + }, + "answer_label": "E", + "answer_text": "You can’t reach the knife on the table next to it. Right now the knife is to your front-right.", + "q_id": "9e76a664-9014-4630-94dd-26b3b2594576_4_T1", + "correct_answer": "You can’t reach the knife on the table next to it. Right now the knife is to your front-right.", + "original_options": { + "A": "Yes, you can reach it.", + "B": "No, it’s to your front-left.", + "C": "You can’t reach the knife on the table next to it; right now the knife is positioned to your back-right.", + "D": "You can’t reach the knife on the table next to it; the knife is currently behind you on the left.", + "E": "You can’t reach the knife on the table next to it. Right now the knife is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 21, + 21 + ] + }, + { + "video_uid": "9e76a664-9014-4630-94dd-26b3b2594576", + "video_length": "00:30:00", + "qid": "9e76a664-9014-4630-94dd-26b3b2594576_3_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many levels did the cabinet above the washing machine have?", + "answer": "The cabinet above the washing machine has 3 shelves, and it’s to your back-right.", + "question_chinese": "我看到洗衣机上面的柜子有几层？", + "answer_chinese": "洗衣机上面的柜子有3层，在你的右后方。", + "evidence": [ + { + "description": "The cabinet above the washing machine has three layers.", + "timestamps": { + "start": "00:16:58", + "end": "00:17:07" + } + } + ], + "question_timestamps": "00:20:49", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four shelves.", + "B": "Two shelves.", + "C": "The cabinet above the washing machine has 1 shelf, and it’s positioned to your back-right.", + "D": "The cabinet above the washing machine has 5 shelves, and it’s located to your back-right.", + "E": "The cabinet above the washing machine has 3 shelves, and it’s to your back-right." + }, + "answer_label": "E", + "answer_text": "The cabinet above the washing machine has 3 shelves, and it’s to your back-right.", + "q_id": "9e76a664-9014-4630-94dd-26b3b2594576_3_T2", + "correct_answer": "The cabinet above the washing machine has 3 shelves, and it’s to your back-right.", + "original_options": { + "A": "Four shelves.", + "B": "Two shelves.", + "C": "The cabinet above the washing machine has 1 shelf, and it’s positioned to your back-right.", + "D": "The cabinet above the washing machine has 5 shelves, and it’s located to your back-right.", + "E": "The cabinet above the washing machine has 3 shelves, and it’s to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "9e76a664-9014-4630-94dd-26b3b2594576", + "video_length": "00:30:00", + "qid": "9e76a664-9014-4630-94dd-26b3b2594576_3_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many levels did the cabinet above the washing machine have?", + "answer": "The cabinet above the washing machine has 3 shelves, and it’s to your front-left.", + "question_chinese": "我看到洗衣机上面的柜子有几层？", + "answer_chinese": "洗衣机上面的柜子有3层，在你的左前方。", + "evidence": [ + { + "description": "The cabinet above the washing machine has three layers.", + "timestamps": { + "start": "00:16:58", + "end": "00:17:07" + } + } + ], + "question_timestamps": "00:25:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four shelves.", + "B": "Two shelves.", + "C": "The cabinet above the washing machine has 1 shelf, and it’s positioned to your front-left.", + "D": "The cabinet above the washing machine has 3 shelves, and it’s to your front-left.", + "E": "The cabinet above the washing machine has 5 shelves, and it’s located to your front-left." + }, + "answer_label": "D", + "answer_text": "The cabinet above the washing machine has 3 shelves, and it’s to your front-left.", + "q_id": "9e76a664-9014-4630-94dd-26b3b2594576_3_T1", + "correct_answer": "The cabinet above the washing machine has 3 shelves, and it’s to your front-left.", + "original_options": { + "A": "Four shelves.", + "B": "Two shelves.", + "C": "The cabinet above the washing machine has 1 shelf, and it’s positioned to your front-left.", + "D": "The cabinet above the washing machine has 3 shelves, and it’s to your front-left.", + "E": "The cabinet above the washing machine has 5 shelves, and it’s located to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "9e76a664-9014-4630-94dd-26b3b2594576", + "video_length": "00:30:00", + "qid": "9e76a664-9014-4630-94dd-26b3b2594576_5_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where on the table did I see the Mickey figurine?", + "answer": "The Mickey figurine is on the table next to the white basket, and it’s to your front-left.", + "question_chinese": "我看到米奇摆件在桌上哪个位置？", + "answer_chinese": "米奇摆件在桌上白色篮子旁边，在你的左前方。", + "evidence": [ + { + "description": "Mickey's ornament is on the table next to the white basket.", + "timestamps": { + "start": "00:21:27", + "end": "00:21:35" + } + }, + { + "description": "Mickey's ornament is on the table next to the white basket.", + "timestamps": { + "start": "00:21:43", + "end": "00:21:53" + } + } + ], + "question_timestamps": "00:21:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the table to your front-right.", + "B": "The Mickey figurine is on the table next to the white basket, and it’s to your front-left.", + "C": "On the table to your back-left.", + "D": "The Mickey figurine is on the table next to the white basket, and it’s positioned to your back-right.", + "E": "The Mickey figurine is on the table next to the white basket, but it’s to your left rather than front-left." + }, + "answer_label": "B", + "answer_text": "The Mickey figurine is on the table next to the white basket, and it’s to your front-left.", + "q_id": "9e76a664-9014-4630-94dd-26b3b2594576_5_T2", + "correct_answer": "The Mickey figurine is on the table next to the white basket, and it’s to your front-left.", + "original_options": { + "A": "On the table to your front-right.", + "B": "The Mickey figurine is on the table next to the white basket, and it’s to your front-left.", + "C": "On the table to your back-left.", + "D": "The Mickey figurine is on the table next to the white basket, and it’s positioned to your back-right.", + "E": "The Mickey figurine is on the table next to the white basket, but it’s to your left rather than front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 21 + ] + }, + { + "video_uid": "9e76a664-9014-4630-94dd-26b3b2594576", + "video_length": "00:30:00", + "qid": "9e76a664-9014-4630-94dd-26b3b2594576_5_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where on the table did I see the Mickey figurine?", + "answer": "The Mickey figurine is on the table next to the white basket, and it’s to your front-right.", + "question_chinese": "我看到米奇摆件在桌上哪个位置？", + "answer_chinese": "米奇摆件在桌上白色篮子旁边，在你的右前方。", + "evidence": [ + { + "description": "Mickey's ornament is on the table next to the white basket.", + "timestamps": { + "start": "00:21:27", + "end": "00:21:35" + } + }, + { + "description": "Mickey's ornament is on the table next to the white basket.", + "timestamps": { + "start": "00:21:43", + "end": "00:21:53" + } + } + ], + "question_timestamps": "00:26:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The Mickey figurine is on the table next to the white basket, and it’s to your front-right.", + "B": "It’s to your front-left.", + "C": "It’s to your back-right.", + "D": "The Mickey figurine is on the table by the white basket, but it sits to your front-left.", + "E": "The Mickey figurine is on the table next to the white basket, and you would find it behind you on the right." + }, + "answer_label": "A", + "answer_text": "The Mickey figurine is on the table next to the white basket, and it’s to your front-right.", + "q_id": "9e76a664-9014-4630-94dd-26b3b2594576_5_T1", + "correct_answer": "The Mickey figurine is on the table next to the white basket, and it’s to your front-right.", + "original_options": { + "A": "The Mickey figurine is on the table next to the white basket, and it’s to your front-right.", + "B": "It’s to your front-left.", + "C": "It’s to your back-right.", + "D": "The Mickey figurine is on the table by the white basket, but it sits to your front-left.", + "E": "The Mickey figurine is on the table next to the white basket, and you would find it behind you on the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 22 + ] + }, + { + "video_uid": "9e76a664-9014-4630-94dd-26b3b2594576", + "video_length": "00:30:00", + "qid": "9e76a664-9014-4630-94dd-26b3b2594576_4", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I taken things from the house to the car so far?", + "answer": "So far, you have taken something from the house to the car once.", + "question_chinese": "目前为止，我从屋子里拿了几次东西到车里？", + "answer_chinese": "目前为止，你从屋子里拿了1次东西到车里。", + "evidence": [ + { + "description": "This is the first time I have taken something from the house to the car.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:58" + } + } + ], + "question_timestamps": "00:03:59", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Three times.", + "C": "So far, you have taken something from the house to the car once.", + "D": "So far, you have taken something from the house to the car two separate times.", + "E": "So far, you have taken something from the house to the car three different times already." + }, + "answer_label": "C", + "answer_text": "So far, you have taken something from the house to the car once.", + "q_id": "9e76a664-9014-4630-94dd-26b3b2594576_4", + "correct_answer": "So far, you have taken something from the house to the car once.", + "original_options": { + "A": "Twice.", + "B": "Three times.", + "C": "So far, you have taken something from the house to the car once.", + "D": "So far, you have taken something from the house to the car two separate times.", + "E": "So far, you have taken something from the house to the car three different times already." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 16 + ] + }, + { + "video_uid": "9e76a664-9014-4630-94dd-26b3b2594576", + "video_length": "00:30:00", + "qid": "9e76a664-9014-4630-94dd-26b3b2594576_5", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I taken things from the house to the car so far?", + "answer": "So far, you have taken things from the house to the car twice.", + "question_chinese": "目前为止，我从屋子里拿了几次东西到车里？", + "answer_chinese": "目前为止，你从屋子里拿了2次东西到车里。", + "evidence": [ + { + "description": "This is the second time I have taken something from the house to the car.", + "timestamps": { + "start": "00:04:17", + "end": "00:04:49" + } + }, + { + "description": "This is the first time I have taken something from the house to the car.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:58" + } + } + ], + "question_timestamps": "00:04:50", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "So far, you have taken things from the house to the car twice.", + "C": "Three times.", + "D": "So far, you have taken things from the house to the car only one time.", + "E": "So far, you have taken things from the house to the car three separate times already." + }, + "answer_label": "B", + "answer_text": "So far, you have taken things from the house to the car twice.", + "q_id": "9e76a664-9014-4630-94dd-26b3b2594576_5", + "correct_answer": "So far, you have taken things from the house to the car twice.", + "original_options": { + "A": "Once.", + "B": "So far, you have taken things from the house to the car twice.", + "C": "Three times.", + "D": "So far, you have taken things from the house to the car only one time.", + "E": "So far, you have taken things from the house to the car three separate times already." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 16 + ] + }, + { + "video_uid": "9e76a664-9014-4630-94dd-26b3b2594576", + "video_length": "00:30:00", + "qid": "9e76a664-9014-4630-94dd-26b3b2594576_1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I taken things from the car to the house so far?", + "answer": "So far, you have taken something from the car to the house once.", + "question_chinese": "目前为止，我从车里拿了几次东西到屋子里？", + "answer_chinese": "目前为止，你从车里拿了1次东西到屋子里。", + "evidence": [ + { + "description": "This was my first time taking something from the car to the house.", + "timestamps": { + "start": "00:01:28", + "end": "00:01:45" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Three times.", + "C": "So far, you have taken something from the car to the house two different times.", + "D": "So far, you have taken something from the car to the house three separate times already.", + "E": "So far, you have taken something from the car to the house once." + }, + "answer_label": "E", + "answer_text": "So far, you have taken something from the car to the house once.", + "q_id": "9e76a664-9014-4630-94dd-26b3b2594576_1", + "correct_answer": "So far, you have taken something from the car to the house once.", + "original_options": { + "A": "Twice.", + "B": "Three times.", + "C": "So far, you have taken something from the car to the house two different times.", + "D": "So far, you have taken something from the car to the house three separate times already.", + "E": "So far, you have taken something from the car to the house once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 16 + ] + }, + { + "video_uid": "9e76a664-9014-4630-94dd-26b3b2594576", + "video_length": "00:30:00", + "qid": "9e76a664-9014-4630-94dd-26b3b2594576_2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I taken things from the car to the house so far?", + "answer": "So far, you have taken things from the car to the house twice.", + "question_chinese": "目前为止，我从车里拿了几次东西到屋子里？", + "answer_chinese": "目前为止，你从车里拿了2次东西到屋子里。", + "evidence": [ + { + "description": "This was the first time I have taken something from the car into the house.", + "timestamps": { + "start": "00:01:28", + "end": "00:01:45" + } + }, + { + "description": "This was the second time I have taken something from the car into the house.", + "timestamps": { + "start": "00:01:56", + "end": "00:02:19" + } + } + ], + "question_timestamps": "00:02:23", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three times.", + "B": "Four times.", + "C": "So far, you have taken things from the car to the house three separate times.", + "D": "So far, you have taken things from the car to the house twice.", + "E": "So far, you have taken things from the car to the house four different times already." + }, + "answer_label": "D", + "answer_text": "So far, you have taken things from the car to the house twice.", + "q_id": "9e76a664-9014-4630-94dd-26b3b2594576_2", + "correct_answer": "So far, you have taken things from the car to the house twice.", + "original_options": { + "A": "Three times.", + "B": "Four times.", + "C": "So far, you have taken things from the car to the house three separate times.", + "D": "So far, you have taken things from the car to the house twice.", + "E": "So far, you have taken things from the car to the house four different times already." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + } +] diff --git a/QAs/9e88251a-1468-45f8-b54e-669832057792-mcq.json b/QAs/9e88251a-1468-45f8-b54e-669832057792-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ae3202d75576ff4a5d814123961dfa31f7c35380 --- /dev/null +++ b/QAs/9e88251a-1468-45f8-b54e-669832057792-mcq.json @@ -0,0 +1,1062 @@ +[ + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_4_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the scissors or the roll of tape that wasn’t used up?", + "answer": "The tape is closer to you.", + "question_chinese": "剪刀和没用完的胶带卷哪个离我更近？", + "answer_chinese": "胶带离你更近。", + "evidence": [ + { + "description": "The tape is in your hand, and the scissors are on the table.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:01:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The scissors are closer to you.", + "B": "The tape is closer to you." + }, + "answer_label": "B", + "answer_text": "The tape is closer to you.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_4_T1", + "correct_answer": "The tape is closer to you.", + "original_options": { + "A": "The scissors are closer to you.", + "B": "The tape is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_8", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the scissors or the roll of tape that hasn't been used up?", + "answer": "The scissors are closer to you.", + "question_chinese": "剪刀和没用完的胶带卷哪个离我更近？", + "answer_chinese": "剪刀离你更近。", + "evidence": [ + { + "description": "The scissors are between me and the roll of tape that hasn't been used up.", + "timestamps": { + "start": "00:34:14", + "end": "00:34:21" + } + } + ], + "question_timestamps": "00:37:48", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The scissors are closer to you.", + "B": "The roll of tape is closer to you." + }, + "answer_label": "A", + "answer_text": "The scissors are closer to you.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_8", + "correct_answer": "The scissors are closer to you.", + "original_options": { + "A": "The scissors are closer to you.", + "B": "The roll of tape is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_4_T2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the scissors or the roll of tape that wasn’t used up?", + "answer": "The scissors are closer to you.", + "question_chinese": "剪刀和没用完的胶带卷哪个离我更近？", + "answer_chinese": "剪刀离你更近。", + "evidence": [ + { + "description": "The scissors are between me and the roll of tape that hasn't been used up.", + "timestamps": { + "start": "00:34:14", + "end": "00:34:21" + } + } + ], + "question_timestamps": "00:50:34", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The scissors are closer to you.", + "B": "The tape is closer to you." + }, + "answer_label": "A", + "answer_text": "The scissors are closer to you.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_4_T2", + "correct_answer": "The scissors are closer to you.", + "original_options": { + "A": "The scissors are closer to you.", + "B": "The tape is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_2_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the scissors?", + "answer": "You put the scissors on the table, behind the scissors.", + "question_chinese": "我把剪刀放在哪里了？", + "answer_chinese": "我把剪刀放在了桌子上，在剪刀的后方。", + "evidence": [ + { + "description": "I placed the scissors on the table, behind the scissors.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the table, in front.", + "B": "On the desk, behind.", + "C": "You put the scissors on the table, behind the scissors.", + "D": "You put the scissors on the table, in front of the scissors instead of behind them.", + "E": "You left the scissors on the desk, positioned behind the scissors rather than on the table." + }, + "answer_label": "C", + "answer_text": "You put the scissors on the table, behind the scissors.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_2_T1", + "correct_answer": "You put the scissors on the table, behind the scissors.", + "original_options": { + "A": "On the table, in front.", + "B": "On the desk, behind.", + "C": "You put the scissors on the table, behind the scissors.", + "D": "You put the scissors on the table, in front of the scissors instead of behind them.", + "E": "You left the scissors on the desk, positioned behind the scissors rather than on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_2_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the scissors?", + "answer": "You put the scissors on the table, behind the scissors.", + "question_chinese": "我把剪刀放在哪里了？", + "answer_chinese": "我把剪刀放在桌子上，剪刀的后面。", + "evidence": [ + { + "description": "I placed the scissors on the table, behind the scissors.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:25:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put the scissors on the table, behind the scissors.", + "B": "On the table, in front.", + "C": "On the desk, behind.", + "D": "You put the scissors on the table, in front of the scissors instead of behind them.", + "E": "You left the scissors on the desk, positioned behind the scissors rather than on the table." + }, + "answer_label": "A", + "answer_text": "You put the scissors on the table, behind the scissors.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_2_T2", + "correct_answer": "You put the scissors on the table, behind the scissors.", + "original_options": { + "A": "You put the scissors on the table, behind the scissors.", + "B": "On the table, in front.", + "C": "On the desk, behind.", + "D": "You put the scissors on the table, in front of the scissors instead of behind them.", + "E": "You left the scissors on the desk, positioned behind the scissors rather than on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the tape that I didn't use up?", + "answer": "I put the unused tape on the desk, to the right rear of the mouse.", + "question_chinese": "我把没用完的胶带放哪里了？", + "answer_chinese": "我把没用完的胶带放在了桌子上，在鼠标右后方。", + "evidence": [ + { + "description": "I put the unused tape on the desk, to the right rear of the mouse.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I put the unused tape on the desk, to the right rear of the mouse.", + "B": "On the desk, left rear.", + "C": "On the desk, right front.", + "D": "I put the unused tape on the desk, to the left rear of the mouse instead of the right rear.", + "E": "I left the unused tape on the desk, just to the right of the mouse rather than behind it." + }, + "answer_label": "A", + "answer_text": "I put the unused tape on the desk, to the right rear of the mouse.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_1", + "correct_answer": "I put the unused tape on the desk, to the right rear of the mouse.", + "original_options": { + "A": "I put the unused tape on the desk, to the right rear of the mouse.", + "B": "On the desk, left rear.", + "C": "On the desk, right front.", + "D": "I put the unused tape on the desk, to the left rear of the mouse instead of the right rear.", + "E": "I left the unused tape on the desk, just to the right of the mouse rather than behind it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 19 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the tape that I didn't use up?", + "answer": "I put the unused tape on the table, to the right rear of the mouse.", + "question_chinese": "我把没用完的胶带放哪里了？", + "answer_chinese": "我把没用完的胶带放在了桌子上，在鼠标右后方。", + "evidence": [ + { + "description": "I put the unused tape on the desk, to the right rear of the mouse.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:25:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the table, left rear.", + "B": "I put the unused tape on the table, to the right rear of the mouse.", + "C": "On the table, right front.", + "D": "I put the unused tape on the table, to the left rear of the mouse instead of the right rear.", + "E": "I left the unused tape on the desk to the right of the keyboard, not on the table near the mouse." + }, + "answer_label": "B", + "answer_text": "I put the unused tape on the table, to the right rear of the mouse.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_2", + "correct_answer": "I put the unused tape on the table, to the right rear of the mouse.", + "original_options": { + "A": "On the table, left rear.", + "B": "I put the unused tape on the table, to the right rear of the mouse.", + "C": "On the table, right front.", + "D": "I put the unused tape on the table, to the left rear of the mouse instead of the right rear.", + "E": "I left the unused tape on the desk to the right of the keyboard, not on the table near the mouse." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 21 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_9", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the scissors without moving my feet?", + "answer": "You can get the scissors.", + "question_chinese": "我现在在不移动脚步的情况下能拿到剪刀吗？", + "answer_chinese": "你可以拿到剪刀。", + "evidence": [ + { + "description": "I placed the scissors in a place with my hands without moving my feet.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:12:48", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot get the scissors.", + "B": "You can get the scissors." + }, + "answer_label": "B", + "answer_text": "You can get the scissors.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_9", + "correct_answer": "You can get the scissors.", + "original_options": { + "A": "You cannot get the scissors.", + "B": "You can get the scissors." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_10", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the scissors without moving my feet?", + "answer": "You can get the scissors.", + "question_chinese": "我现在在不移动脚步的情况下能拿到剪刀吗？", + "answer_chinese": "你可以拿到剪刀。", + "evidence": [ + { + "description": "I placed the scissors in a place with my hands without moving my feet.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:36:57", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot get the scissors.", + "B": "You can get the scissors." + }, + "answer_label": "B", + "answer_text": "You can get the scissors.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_10", + "correct_answer": "You can get the scissors.", + "original_options": { + "A": "You cannot get the scissors.", + "B": "You can get the scissors." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_5_T1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Right now, can I touch the scissors without moving my feet?", + "answer": "You can reach the scissors.", + "question_chinese": "我现在在不移动脚步的情况下能碰到剪刀吗？", + "answer_chinese": "你可以碰到剪刀。", + "evidence": [ + { + "description": "I placed the scissors in a place with my hands without moving my feet.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the scissors.", + "B": "You cannot reach the scissors." + }, + "answer_label": "A", + "answer_text": "You can reach the scissors.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_5_T1", + "correct_answer": "You can reach the scissors.", + "original_options": { + "A": "You can reach the scissors.", + "B": "You cannot reach the scissors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_5_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Right now, can I touch the scissors without moving my feet?", + "answer": "You can't reach the scissors.", + "question_chinese": "我现在在不移动脚步的情况下能碰到剪刀吗？", + "answer_chinese": "你不可以碰到剪刀。", + "evidence": [ + { + "description": "I placed the scissors in a place with my hands without moving my feet.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:36:57", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can't reach the scissors.", + "B": "You can reach the scissors." + }, + "answer_label": "A", + "answer_text": "You can't reach the scissors.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_5_T2", + "correct_answer": "You can't reach the scissors.", + "original_options": { + "A": "You can't reach the scissors.", + "B": "You can reach the scissors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_6_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I switched pens so far?", + "answer": "You changed pens once.", + "question_chinese": "我现在换了几次笔？", + "answer_chinese": "你换了一次笔。", + "evidence": [ + { + "description": "Firest.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:42" + } + } + ], + "question_timestamps": "00:06:34", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You did not change pens.", + "B": "You changed pens twice.", + "C": "You switched pens three times so far during the task, moving from one pen to another each time.", + "D": "You have already changed pens four times, each switch ending with a different pen in your hand.", + "E": "You changed pens once." + }, + "answer_label": "E", + "answer_text": "You changed pens once.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_6_T1", + "correct_answer": "You changed pens once.", + "original_options": { + "A": "You did not change pens.", + "B": "You changed pens twice.", + "C": "You switched pens three times so far during the task, moving from one pen to another each time.", + "D": "You have already changed pens four times, each switch ending with a different pen in your hand.", + "E": "You changed pens once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 17 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_6_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I switched pens so far?", + "answer": "You have changed pens twice so far.", + "question_chinese": "我现在换了几次笔？", + "answer_chinese": "你现在换了两次笔。", + "evidence": [ + { + "description": "First.", + "timestamps": { + "start": "00:05:37", + "end": "00:05:41" + } + }, + { + "description": "Second.", + "timestamps": { + "start": "00:07:15", + "end": "00:07:34" + } + } + ], + "question_timestamps": "00:08:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have changed pens twice so far.", + "B": "You did not change pens.", + "C": "You changed pens once.", + "D": "You have changed pens three times so far, switching away from the previous pen each time you resumed writing.", + "E": "You switched pens four times so far, ending up with a new pen after each separate change." + }, + "answer_label": "A", + "answer_text": "You have changed pens twice so far.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_6_T2", + "correct_answer": "You have changed pens twice so far.", + "original_options": { + "A": "You have changed pens twice so far.", + "B": "You did not change pens.", + "C": "You changed pens once.", + "D": "You have changed pens three times so far, switching away from the previous pen each time you resumed writing.", + "E": "You switched pens four times so far, ending up with a new pen after each separate change." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 19, + 17 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_7_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many outlining pens are placed to my right right now?", + "answer": "There are two fineliners on your right-hand side.", + "question_chinese": "我的右手边现在放了几支勾线笔？", + "answer_chinese": "你的右手边有两支勾线笔。", + "evidence": [ + { + "description": "First", + "timestamps": { + "start": "00:39:27", + "end": "00:39:29" + } + }, + { + "description": "Second", + "timestamps": { + "start": "00:39:58", + "end": "00:40:00" + } + } + ], + "question_timestamps": "00:40:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There is one fineliner on your right.", + "B": "There are two fineliners on your right-hand side.", + "C": "There are three fineliners on your right.", + "D": "There are zero fineliners on your right-hand side, with none placed beside you at the moment.", + "E": "There are four fineliners on your right-hand side, grouped together to the right of where you are working." + }, + "answer_label": "B", + "answer_text": "There are two fineliners on your right-hand side.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_7_T2", + "correct_answer": "There are two fineliners on your right-hand side.", + "original_options": { + "A": "There is one fineliner on your right.", + "B": "There are two fineliners on your right-hand side.", + "C": "There are three fineliners on your right.", + "D": "There are zero fineliners on your right-hand side, with none placed beside you at the moment.", + "E": "There are four fineliners on your right-hand side, grouped together to the right of where you are working." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_7_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many outlining pens are placed to my right right now?", + "answer": "There are three fineliners on your right-hand side.", + "question_chinese": "我的右手边现在放了几支勾线笔？", + "answer_chinese": "你的右手边有三支勾线笔。", + "evidence": [ + { + "description": "First", + "timestamps": { + "start": "00:39:27", + "end": "00:39:29" + } + }, + { + "description": "Second", + "timestamps": { + "start": "00:39:58", + "end": "00:40:00" + } + }, + { + "description": "Third", + "timestamps": { + "start": "00:41:19", + "end": "00:41:20" + } + } + ], + "question_timestamps": "00:41:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are three fineliners on your right-hand side.", + "B": "There are two fineliners on your right.", + "C": "There are four fineliners on your right.", + "D": "There is one fineliner on your right-hand side, with only a single outlining pen placed there.", + "E": "There are five fineliners on your right-hand side, with five outlining pens laid out to your right." + }, + "answer_label": "A", + "answer_text": "There are three fineliners on your right-hand side.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_7_T1", + "correct_answer": "There are three fineliners on your right-hand side.", + "original_options": { + "A": "There are three fineliners on your right-hand side.", + "B": "There are two fineliners on your right.", + "C": "There are four fineliners on your right.", + "D": "There is one fineliner on your right-hand side, with only a single outlining pen placed there.", + "E": "There are five fineliners on your right-hand side, with five outlining pens laid out to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_3_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the scissors?", + "answer": "The scissors are on your right-hand side.", + "question_chinese": "我看到剪刀在哪里呢？", + "answer_chinese": "剪刀在你的右手边。", + "evidence": [ + { + "description": "The scissors are on your right side.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The scissors are on your left-hand side.", + "B": "The scissors are on your right-hand side." + }, + "answer_label": "B", + "answer_text": "The scissors are on your right-hand side.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_3_T1", + "correct_answer": "The scissors are on your right-hand side.", + "original_options": { + "A": "The scissors are on your left-hand side.", + "B": "The scissors are on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_3_T2", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the scissors?", + "answer": "The scissors are next to the mouse.", + "question_chinese": "我看到剪刀在哪里呢？", + "answer_chinese": "剪刀在鼠标的旁边。", + "evidence": [ + { + "description": "The scissors are next to the mouse.", + "timestamps": { + "start": "00:30:45", + "end": "00:31:20" + } + } + ], + "question_timestamps": "00:34:23", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The scissors are next to the mouse.", + "B": "The scissors are on your right-hand side." + }, + "answer_label": "A", + "answer_text": "The scissors are next to the mouse.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_3_T2", + "correct_answer": "The scissors are next to the mouse.", + "original_options": { + "A": "The scissors are next to the mouse.", + "B": "The scissors are on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_1_T1", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the tape?", + "answer": "The tape is in your hand.", + "question_chinese": "我看到胶带在哪里了？", + "answer_chinese": "胶带在你的手中。", + "evidence": [ + { + "description": "The tape is in your hand.", + "timestamps": { + "start": "00:00:55", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The tape is on the table.", + "B": "The tape is by the mouse.", + "C": "You put the tape on the desk, behind and to the right of the mouse.", + "D": "The tape is in your hand.", + "E": "You left the tape on the table, just behind the mouse rather than holding it." + }, + "answer_label": "D", + "answer_text": "The tape is in your hand.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_1_T1", + "correct_answer": "The tape is in your hand.", + "original_options": { + "A": "The tape is on the table.", + "B": "The tape is by the mouse.", + "C": "You put the tape on the desk, behind and to the right of the mouse.", + "D": "The tape is in your hand.", + "E": "You left the tape on the table, just behind the mouse rather than holding it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_1_T2", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the tape?", + "answer": "You put the tape on the table, behind and to the right of the mouse.", + "question_chinese": "我看到胶带在哪里了？", + "answer_chinese": "你把胶带放在了桌子上，在鼠标右后方。", + "evidence": [ + { + "description": "I put the unused tape on the desk, to the right rear of the mouse.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:25:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the table, in front of the mouse.", + "B": "On the table, behind and to your left.", + "C": "You put the tape on the table, directly behind the mouse without it being to the right.", + "D": "You placed the tape on the table, in front and to the right of the mouse instead.", + "E": "You put the tape on the table, behind and to the right of the mouse." + }, + "answer_label": "E", + "answer_text": "You put the tape on the table, behind and to the right of the mouse.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_1_T2", + "correct_answer": "You put the tape on the table, behind and to the right of the mouse.", + "original_options": { + "A": "On the table, in front of the mouse.", + "B": "On the table, behind and to your left.", + "C": "You put the tape on the table, directly behind the mouse without it being to the right.", + "D": "You placed the tape on the table, in front and to the right of the mouse instead.", + "E": "You put the tape on the table, behind and to the right of the mouse." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 17 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_11", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the computer now?", + "answer": "The computer is now directly in front of me.", + "question_chinese": "电脑现在在我的什么方向？", + "answer_chinese": "电脑现在在我的正前方。", + "evidence": [ + { + "description": "The computer is now directly in front of me.", + "timestamps": { + "start": "00:34:37", + "end": "00:34:51" + } + } + ], + "question_timestamps": "00:36:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The computer is now directly behind me.", + "B": "The computer is now to my left.", + "C": "The computer is now slightly to my right, not directly in front of me anymore.", + "D": "The computer is now directly in front of me.", + "E": "The computer is now off to my right side rather than positioned straight ahead." + }, + "answer_label": "D", + "answer_text": "The computer is now directly in front of me.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_11", + "correct_answer": "The computer is now directly in front of me.", + "original_options": { + "A": "The computer is now directly behind me.", + "B": "The computer is now to my left.", + "C": "The computer is now slightly to my right, not directly in front of me anymore.", + "D": "The computer is now directly in front of me.", + "E": "The computer is now off to my right side rather than positioned straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 14 + ] + }, + { + "video_uid": "9e88251a-1468-45f8-b54e-669832057792", + "video_length": "00:50:35", + "qid": "9e88251a-1468-45f8-b54e-669832057792_12", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the computer now?", + "answer": "The computer is now directly in front of me.", + "question_chinese": "电脑现在在我的什么方向？", + "answer_chinese": "电脑现在在我的正前方。", + "evidence": [ + { + "description": "The computer is now directly in front of me.", + "timestamps": { + "start": "00:34:37", + "end": "00:34:51" + } + } + ], + "question_timestamps": "00:49:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The computer is now slightly to my left.", + "B": "The computer is now directly in front of me.", + "C": "The computer is now directly behind me.", + "D": "The computer is now positioned to my right side instead of being directly in front of me.", + "E": "The computer is now above me rather than straight ahead in my line of sight." + }, + "answer_label": "B", + "answer_text": "The computer is now directly in front of me.", + "q_id": "9e88251a-1468-45f8-b54e-669832057792_12", + "correct_answer": "The computer is now directly in front of me.", + "original_options": { + "A": "The computer is now slightly to my left.", + "B": "The computer is now directly in front of me.", + "C": "The computer is now directly behind me.", + "D": "The computer is now positioned to my right side instead of being directly in front of me.", + "E": "The computer is now above me rather than straight ahead in my line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 15 + ] + } +] diff --git a/QAs/9e9c5b05-c7d4-450d-9850-e6ae83caa9a5-mcq.json b/QAs/9e9c5b05-c7d4-450d-9850-e6ae83caa9a5-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ae26c6bb0ac373774d8f7453b37e8e45895ede68 --- /dev/null +++ b/QAs/9e9c5b05-c7d4-450d-9850-e6ae83caa9a5-mcq.json @@ -0,0 +1,1191 @@ +[ + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_3", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the small spoon just now?", + "answer": "In the bowl on the table.", + "question_chinese": "刚才我把小勺子放哪了？", + "answer_chinese": "桌上的碗里", + "evidence": [ + { + "description": "I put the spoon in the bowl on the table", + "timestamps": { + "start": "00:03:44", + "end": "00:03:49" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Next to the bowl on the table.", + "B": "Under the bowl on the table.", + "C": "You put the small spoon on the table beside the bowl, not inside it.", + "D": "In the bowl on the table.", + "E": "You placed the small spoon next to the cup on the table rather than in the bowl." + }, + "answer_label": "D", + "answer_text": "In the bowl on the table.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_3", + "correct_answer": "In the bowl on the table.", + "original_options": { + "A": "Next to the bowl on the table.", + "B": "Under the bowl on the table.", + "C": "You put the small spoon on the table beside the bowl, not inside it.", + "D": "In the bowl on the table.", + "E": "You placed the small spoon next to the cup on the table rather than in the bowl." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 17 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the small spoon just now?", + "answer": "On the cabinet on the left wall of the cooktop.", + "question_chinese": "刚才我把小勺子放哪了？", + "answer_chinese": "灶台的左边墙上的柜子上", + "evidence": [ + { + "description": "I hung the spoon on the cabinet on the left wall of the cooktop", + "timestamps": { + "start": "00:37:21", + "end": "00:37:27" + } + } + ], + "question_timestamps": "00:44:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the cabinet on the right wall.", + "B": "On the table beside the dining chairs.", + "C": "You put the small spoon on the cabinet on the left wall beside the sink, not by the cooktop.", + "D": "You left the small spoon on the counter near the refrigerator instead of on the left-wall cabinet.", + "E": "On the cabinet on the left wall of the cooktop." + }, + "answer_label": "E", + "answer_text": "On the cabinet on the left wall of the cooktop.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_4", + "correct_answer": "On the cabinet on the left wall of the cooktop.", + "original_options": { + "A": "On the cabinet on the right wall.", + "B": "On the table beside the dining chairs.", + "C": "You put the small spoon on the cabinet on the left wall beside the sink, not by the cooktop.", + "D": "You left the small spoon on the counter near the refrigerator instead of on the left-wall cabinet.", + "E": "On the cabinet on the left wall of the cooktop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 17 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after finishing making pancakes just now?", + "answer": "Cut loofah.", + "question_chinese": "刚才我烙完饼后去干嘛了？", + "answer_chinese": "切丝瓜", + "evidence": [ + { + "description": "I was making pancakes", + "timestamps": { + "start": "00:19:46", + "end": "00:20:00" + } + }, + { + "description": "I was cutting loofah", + "timestamps": { + "start": "00:20:06", + "end": "00:20:12" + } + } + ], + "question_timestamps": "00:27:35", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Cut cucumber.", + "B": "Cut zucchini.", + "C": "Cut loofah.", + "D": "Cut tomato after finishing the pancakes, using the knife on the cutting board.", + "E": "Cut radish next, slicing it up right after you finished making pancakes." + }, + "answer_label": "C", + "answer_text": "Cut loofah.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_1", + "correct_answer": "Cut loofah.", + "original_options": { + "A": "Cut cucumber.", + "B": "Cut zucchini.", + "C": "Cut loofah.", + "D": "Cut tomato after finishing the pancakes, using the knife on the cutting board.", + "E": "Cut radish next, slicing it up right after you finished making pancakes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after finishing making pancakes just now?", + "answer": "Applied essential oil to family member's hair.", + "question_chinese": "刚才我烙完饼后去干嘛了？", + "answer_chinese": "给家人头发摸精油", + "evidence": [ + { + "description": "I was making pancakes", + "timestamps": { + "start": "00:50:50", + "end": "00:51:43" + } + }, + { + "description": "Applied essential oil to family member's hair", + "timestamps": { + "start": "00:54:36", + "end": "00:54:44" + } + } + ], + "question_timestamps": "00:56:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Applied essential oil to family member's hair.", + "B": "Brushed their hair.", + "C": "Applied oil to their hands.", + "D": "Applied essential oil to the family member’s face after finishing making pancakes.", + "E": "Applied essential oil to the family member’s feet right after you finished the pancakes." + }, + "answer_label": "A", + "answer_text": "Applied essential oil to family member's hair.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_2", + "correct_answer": "Applied essential oil to family member's hair.", + "original_options": { + "A": "Applied essential oil to family member's hair.", + "B": "Brushed their hair.", + "C": "Applied oil to their hands.", + "D": "Applied essential oil to the family member’s face after finishing making pancakes.", + "E": "Applied essential oil to the family member’s feet right after you finished the pancakes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 5, + 13, + 15 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_7", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the refrigerator in for me now?", + "answer": "Left back side.", + "question_chinese": "现在冰箱在我的什么方位？", + "answer_chinese": "左后侧", + "evidence": [ + { + "description": "There is a refrigerator on the left side after entering the door, opposite the table", + "timestamps": { + "start": "00:05:04", + "end": "00:05:06" + } + } + ], + "question_timestamps": "00:05:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right back side.", + "B": "Left back side.", + "C": "Left front side.", + "D": "It is on your right front side from where you are standing now.", + "E": "It is on your left middle side, not behind you, relative to your current position." + }, + "answer_label": "B", + "answer_text": "Left back side.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_7", + "correct_answer": "Left back side.", + "original_options": { + "A": "Right back side.", + "B": "Left back side.", + "C": "Left front side.", + "D": "It is on your right front side from where you are standing now.", + "E": "It is on your left middle side, not behind you, relative to your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 15 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_8", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the refrigerator in for me now?", + "answer": "Back side.", + "question_chinese": "现在冰箱在我的什么方位？", + "answer_chinese": "后方", + "evidence": [ + { + "description": "There is a refrigerator on the left side after entering the door, opposite the table", + "timestamps": { + "start": "00:05:04", + "end": "00:05:06" + } + }, + { + "description": "I am now standing in front of the table", + "timestamps": { + "start": "00:31:39", + "end": "00:31:50" + } + } + ], + "question_timestamps": "00:32:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Back side.", + "B": "Front side.", + "C": "Right side.", + "D": "The refrigerator is on your left side relative to where you are now.", + "E": "The refrigerator is above you, positioned on the top side from your viewpoint." + }, + "answer_label": "A", + "answer_text": "Back side.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_8", + "correct_answer": "Back side.", + "original_options": { + "A": "Back side.", + "B": "Front side.", + "C": "Right side.", + "D": "The refrigerator is on your left side relative to where you are now.", + "E": "The refrigerator is above you, positioned on the top side from your viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_13", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I washed things so far?", + "answer": "Four.", + "question_chinese": "现在我已经洗了几次东西？", + "answer_chinese": "4次", + "evidence": [ + { + "description": "I was washing mixing utensils", + "timestamps": { + "start": "00:04:35", + "end": "00:04:40" + } + }, + { + "description": "I was washing spoons and plates", + "timestamps": { + "start": "00:08:41", + "end": "00:08:47" + } + }, + { + "description": "I was washing vegetables", + "timestamps": { + "start": "00:09:26", + "end": "00:09:33" + } + }, + { + "description": "I was washing the cutting board", + "timestamps": { + "start": "00:10:43", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Three times.", + "C": "Four.", + "D": "You have washed things five times in total so far today already.", + "E": "You have washed things six times so far, counting each separate washing action." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_13", + "correct_answer": "Four.", + "original_options": { + "A": "Two times.", + "B": "Three times.", + "C": "Four.", + "D": "You have washed things five times in total so far today already.", + "E": "You have washed things six times so far, counting each separate washing action." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_14", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I washed things so far?", + "answer": "seven.", + "question_chinese": "现在我已经洗了几次东西？", + "answer_chinese": "7次", + "evidence": [ + { + "description": "I was washing mixing utensils", + "timestamps": { + "start": "00:04:35", + "end": "00:04:40" + } + }, + { + "description": "I was washing spoons and plates", + "timestamps": { + "start": "00:08:41", + "end": "00:08:47" + } + }, + { + "description": "I was washing vegetables", + "timestamps": { + "start": "00:09:26", + "end": "00:09:33" + } + }, + { + "description": "I was washing the cutting board", + "timestamps": { + "start": "00:10:43", + "end": "00:10:50" + } + }, + { + "description": "I was washing the peeler", + "timestamps": { + "start": "00:11:48", + "end": "00:11:53" + } + }, + { + "description": "I was washing plates", + "timestamps": { + "start": "00:13:17", + "end": "00:13:23" + } + }, + { + "description": "I was washing the cutting board", + "timestamps": { + "start": "00:22:04", + "end": "00:22:08" + } + } + ], + "question_timestamps": "00:22:40", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five times.", + "B": "Eight times.", + "C": "You have washed things three times so far in total today already.", + "D": "seven.", + "E": "You have washed things ten times so far, counting every distinct wash you did." + }, + "answer_label": "D", + "answer_text": "seven.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_14", + "correct_answer": "seven.", + "original_options": { + "A": "Five times.", + "B": "Eight times.", + "C": "You have washed things three times so far in total today already.", + "D": "seven.", + "E": "You have washed things ten times so far, counting every distinct wash you did." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_5", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many choices do I have if I want to wash hands now?", + "answer": "Two.", + "question_chinese": "现在我想去洗手有几种选择？", + "answer_chinese": "2种", + "evidence": [ + { + "description": "There is a faucet in front of the washing area", + "timestamps": { + "start": "00:04:54", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:05:44", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One choice.", + "C": "Three choices.", + "D": "You have four choices available if you want to wash your hands now.", + "E": "You have five different choices right now for washing your hands easily." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_5", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One choice.", + "C": "Three choices.", + "D": "You have four choices available if you want to wash your hands now.", + "E": "You have five different choices right now for washing your hands easily." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_6", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many choices do I have if I want to wash hands now?", + "answer": "Four.", + "question_chinese": "现在我想去洗手有几种选择？", + "answer_chinese": "4种", + "evidence": [ + { + "description": "There is a faucet in front of the washing area", + "timestamps": { + "start": "00:04:54", + "end": "00:05:00" + } + }, + { + "description": "There is also a sink on the left side of the washing area", + "timestamps": { + "start": "00:11:38", + "end": "00:11:47" + } + }, + { + "description": "There is also a faucet inside the kitchen", + "timestamps": { + "start": "00:37:10", + "end": "00:37:17" + } + } + ], + "question_timestamps": "00:44:32", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two choices.", + "B": "Three choices.", + "C": "You have five choices available if you want to wash your hands right now.", + "D": "Four.", + "E": "You have six different choices at the moment for washing your hands." + }, + "answer_label": "D", + "answer_text": "Four.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_6", + "correct_answer": "Four.", + "original_options": { + "A": "Two choices.", + "B": "Three choices.", + "C": "You have five choices available if you want to wash your hands right now.", + "D": "Four.", + "E": "You have six different choices at the moment for washing your hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 12 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_9", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I do if I want to get some water to boil now?", + "answer": "Go to the left cooktop to get the water scoop, go to the sink outside the door to get water, then pour it into the left bottom pot, and light the fire.", + "question_chinese": "现在我想打点水烧应该怎么做？", + "answer_chinese": "去左侧的灶台上拿舀水盆去门外的洗手池接水再倒入左边下面的那个锅，点火烧", + "evidence": [ + { + "description": "There is a water scoop on the cooktop", + "timestamps": { + "start": "00:05:08", + "end": "00:05:12" + } + }, + { + "description": "There are 4 stoves on the cooktop, one is currently used, the right bottom one, other stoves have pots but not used", + "timestamps": { + "start": "00:05:12", + "end": "00:05:20" + } + } + ], + "question_timestamps": "00:05:44", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Use the right cooktop and pot.", + "B": "Fill water at the sink inside.", + "C": "Go to the left cooktop to get the water scoop, go to the sink outside the door to get water, then pour it into the left bottom pot, and light the fire.", + "D": "Grab the scoop from the left cooktop, fill it at the sink inside, then pour into the left bottom pot.", + "E": "Take the scoop from the left cooktop, get water outside the door, pour it into the left top pot." + }, + "answer_label": "C", + "answer_text": "Go to the left cooktop to get the water scoop, go to the sink outside the door to get water, then pour it into the left bottom pot, and light the fire.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_9", + "correct_answer": "Go to the left cooktop to get the water scoop, go to the sink outside the door to get water, then pour it into the left bottom pot, and light the fire.", + "original_options": { + "A": "Use the right cooktop and pot.", + "B": "Fill water at the sink inside.", + "C": "Go to the left cooktop to get the water scoop, go to the sink outside the door to get water, then pour it into the left bottom pot, and light the fire.", + "D": "Grab the scoop from the left cooktop, fill it at the sink inside, then pour into the left bottom pot.", + "E": "Take the scoop from the left cooktop, get water outside the door, pour it into the left top pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 19 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_10", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I do if I want to get some water to boil now?", + "answer": "Go to the left cooktop to get the water scoop, go to the sink outside the door to get water, then pour it into the left top pot, and light the fire.", + "question_chinese": "现在我想打点水烧应该怎么做？", + "answer_chinese": "去左侧的灶台上拿舀水锅去门外的洗手池接水再倒入左边上面的那个锅，点火烧", + "evidence": [ + { + "description": "There is a water scoop on the cooktop", + "timestamps": { + "start": "00:05:08", + "end": "00:05:12" + } + }, + { + "description": "There are 4 stoves on the cooktop, currently using the right bottom one, other stoves have pots, the left bottom and right top pots already have things in them", + "timestamps": { + "start": "00:43:44", + "end": "00:43:50" + } + } + ], + "question_timestamps": "00:44:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Pour it into the left bottom pot.", + "B": "Fill it at the sink inside.", + "C": "Go to the right cooktop for the scoop, get water outside the door, pour into the right top pot.", + "D": "Go to the left cooktop to get the water scoop, go to the sink outside the door to get water, then pour it into the left top pot, and light the fire.", + "E": "Take the scoop from the left cooktop, fetch water outside the door, pour into the left bottom pot." + }, + "answer_label": "D", + "answer_text": "Go to the left cooktop to get the water scoop, go to the sink outside the door to get water, then pour it into the left top pot, and light the fire.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_10", + "correct_answer": "Go to the left cooktop to get the water scoop, go to the sink outside the door to get water, then pour it into the left top pot, and light the fire.", + "original_options": { + "A": "Pour it into the left bottom pot.", + "B": "Fill it at the sink inside.", + "C": "Go to the right cooktop for the scoop, get water outside the door, pour into the right top pot.", + "D": "Go to the left cooktop to get the water scoop, go to the sink outside the door to get water, then pour it into the left top pot, and light the fire.", + "E": "Take the scoop from the left cooktop, fetch water outside the door, pour into the left bottom pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 18 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_11", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stoves can I use if I want to use the cooktop now?", + "answer": "three.", + "question_chinese": "现在我想用灶台有几个可以使用？", + "answer_chinese": "3个", + "evidence": [ + { + "description": "There are 4 stoves on the cooktop, currently using one, the right bottom one", + "timestamps": { + "start": "00:05:12", + "end": "00:05:20" + } + } + ], + "question_timestamps": "00:05:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "three.", + "D": "You can use two stoves on the cooktop right now, not three.", + "E": "There are four stoves available to use on the cooktop at the moment." + }, + "answer_label": "C", + "answer_text": "three.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_11", + "correct_answer": "three.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "three.", + "D": "You can use two stoves on the cooktop right now, not three.", + "E": "There are four stoves available to use on the cooktop at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_12", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stoves can I use if I want to use the cooktop now?", + "answer": "two.", + "question_chinese": "现在我想用灶台有几个可以使用？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "There are 4 stoves on the cooktop, currently using one, the right bottom one", + "timestamps": { + "start": "00:05:12", + "end": "00:05:20" + } + }, + { + "description": "Currently using the left bottom one and also using the right bottom one", + "timestamps": { + "start": "00:26:20", + "end": "00:26:55" + } + } + ], + "question_timestamps": "00:31:40", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You can use only one stove on the cooktop right now, not two.", + "D": "There are four stoves available to use on the cooktop at the moment.", + "E": "two." + }, + "answer_label": "E", + "answer_text": "two.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_12", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You can use only one stove on the cooktop right now, not two.", + "D": "There are four stoves available to use on the cooktop at the moment.", + "E": "two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_19", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move to pick more peppers now?", + "answer": "Turn left then right, walk straight forward about ten steps to see the peppers.", + "question_chinese": "现在我要再去摘几个辣椒应该怎么移动？", + "answer_chinese": "左转再右转往前直走个10步左右可以看到辣椒", + "evidence": [ + { + "description": "From the table, I exited the door, turned right and walked straight about 10 steps to pick peppers", + "timestamps": { + "start": "00:29:38", + "end": "00:29:45" + } + } + ], + "question_timestamps": "00:29:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right then left.", + "B": "Turn left then right, walk straight forward about ten steps to see the peppers.", + "C": "Turn left then left.", + "D": "Turn right, then right, and walk straight forward about ten steps to see the peppers.", + "E": "Turn left then right, but walk forward only a few steps until you reach the peppers." + }, + "answer_label": "B", + "answer_text": "Turn left then right, walk straight forward about ten steps to see the peppers.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_19", + "correct_answer": "Turn left then right, walk straight forward about ten steps to see the peppers.", + "original_options": { + "A": "Turn right then left.", + "B": "Turn left then right, walk straight forward about ten steps to see the peppers.", + "C": "Turn left then left.", + "D": "Turn right, then right, and walk straight forward about ten steps to see the peppers.", + "E": "Turn left then right, but walk forward only a few steps until you reach the peppers." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_20", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move to pick more peppers now?", + "answer": "First turn around and walk a few steps, turn right to exit the door, then turn right and walk straight forward about 10 steps to see the peppers.", + "question_chinese": "现在我要再去摘几个辣椒应该怎么移动？", + "answer_chinese": "先掉头走几步右转出门再右转往前直走个10步左右可以看到辣椒", + "evidence": [ + { + "description": "The left side of the table is the cooktop", + "timestamps": { + "start": "00:00:38", + "end": "00:00:43" + } + }, + { + "description": "I am now at the cooktop", + "timestamps": { + "start": "00:50:27", + "end": "00:50:32" + } + }, + { + "description": "From the table, I exited the door, turned right and walked straight about 10 steps to pick peppers", + "timestamps": { + "start": "00:29:38", + "end": "00:29:45" + } + } + ], + "question_timestamps": "00:50:35", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "First turn around and walk a few steps, turn right to exit the door, then turn right and walk straight forward about 10 steps to see the peppers.", + "B": "Exit left, then go straight.", + "C": "Take fifteen steps forward.", + "D": "First turn around, walk a few steps, turn left to exit, then turn right and walk about 10 steps.", + "E": "First turn around, walk a few steps, turn right to exit, then turn left and walk about 10 steps." + }, + "answer_label": "A", + "answer_text": "First turn around and walk a few steps, turn right to exit the door, then turn right and walk straight forward about 10 steps to see the peppers.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_20", + "correct_answer": "First turn around and walk a few steps, turn right to exit the door, then turn right and walk straight forward about 10 steps to see the peppers.", + "original_options": { + "A": "First turn around and walk a few steps, turn right to exit the door, then turn right and walk straight forward about 10 steps to see the peppers.", + "B": "Exit left, then go straight.", + "C": "Take fifteen steps forward.", + "D": "First turn around, walk a few steps, turn left to exit, then turn right and walk about 10 steps.", + "E": "First turn around, walk a few steps, turn right to exit, then turn left and walk about 10 steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 19, + 19 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_9_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move to get to the dining table now?", + "answer": "Turn around and walk straight to reach the dining table; the dining table is in front of you.", + "question_chinese": "现在我要到餐桌那应该怎么移动？", + "answer_chinese": "向后转直走就可以到餐桌那，餐桌在你前面。", + "evidence": [ + { + "description": "Turn around and go straight, you can get to the dining table.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around; it’s behind you.", + "B": "Turn around; it’s to your left.", + "C": "Turn around and walk straight to reach the dining table; the dining table is to your right.", + "D": "Turn around and walk straight to reach the dining table; the dining table is off to your left side.", + "E": "Turn around and walk straight to reach the dining table; the dining table is in front of you." + }, + "answer_label": "E", + "answer_text": "Turn around and walk straight to reach the dining table; the dining table is in front of you.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_9_T2", + "correct_answer": "Turn around and walk straight to reach the dining table; the dining table is in front of you.", + "original_options": { + "A": "Turn around; it’s behind you.", + "B": "Turn around; it’s to your left.", + "C": "Turn around and walk straight to reach the dining table; the dining table is to your right.", + "D": "Turn around and walk straight to reach the dining table; the dining table is off to your left side.", + "E": "Turn around and walk straight to reach the dining table; the dining table is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 19 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_9_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How should I move to get to the dining table now?", + "answer": "Turn around and then turn left to reach the doorway, then walk straight to the dining table; the dining table is in front of you.", + "question_chinese": "现在我要到餐桌那应该怎么移动？", + "answer_chinese": "向后左转到达门口，然年直走就可以到餐桌，餐桌在你前面。", + "evidence": [ + { + "description": "Turn around, then turn left to reach the door.", + "timestamps": { + "start": "00:13:17", + "end": "00:13:20" + } + }, + { + "description": "Then go straight to the dining table.", + "timestamps": { + "start": "00:13:20", + "end": "00:13:23" + } + } + ], + "question_timestamps": "00:13:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around, then turn right.", + "B": "Walk straight without turning left.", + "C": "Turn around and then turn left to reach the doorway, then walk straight to the dining table; the dining table is in front of you.", + "D": "Turn around and then turn left to reach the doorway, then walk straight; the dining table will be behind you.", + "E": "Turn around, turn left at the doorway, then walk straight; the dining table should be on your right." + }, + "answer_label": "C", + "answer_text": "Turn around and then turn left to reach the doorway, then walk straight to the dining table; the dining table is in front of you.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_9_T1", + "correct_answer": "Turn around and then turn left to reach the doorway, then walk straight to the dining table; the dining table is in front of you.", + "original_options": { + "A": "Turn around, then turn right.", + "B": "Walk straight without turning left.", + "C": "Turn around and then turn left to reach the doorway, then walk straight to the dining table; the dining table is in front of you.", + "D": "Turn around and then turn left to reach the doorway, then walk straight; the dining table will be behind you.", + "E": "Turn around, turn left at the doorway, then walk straight; the dining table should be on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 18 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_15", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the trash can within 10 steps if I want to throw garbage now?", + "answer": "Yes.", + "question_chinese": "现在我要去丢垃圾能10步走到吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The trash can is in front left of the washing area, about 10 steps from the table", + "timestamps": { + "start": "00:13:10", + "end": "00:13:18" + } + } + ], + "question_timestamps": "00:13:25", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_15", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5", + "video_length": "00:57:29", + "qid": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_16", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the trash can within 10 steps if I want to throw garbage now?", + "answer": "No.", + "question_chinese": "现在我要去丢垃圾能10步走到吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The trash can is in front left of the washing area, about 10 steps from the table", + "timestamps": { + "start": "00:13:10", + "end": "00:13:18" + } + }, + { + "description": "I am now at the cooktop, still a few steps away from the table", + "timestamps": { + "start": "00:46:20", + "end": "00:46:24" + } + } + ], + "question_timestamps": "00:46:28", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "9e9c5b05-c7d4-450d-9850-e6ae83caa9a5_16", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/9ed1345b-3b40-4179-9ed3-1071e5795723-mcq.json b/QAs/9ed1345b-3b40-4179-9ed3-1071e5795723-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..da5f9beb66c1a913974f0b1ee8658af9a63471d3 --- /dev/null +++ b/QAs/9ed1345b-3b40-4179-9ed3-1071e5795723-mcq.json @@ -0,0 +1,615 @@ +[ + { + "video_uid": "9ed1345b-3b40-4179-9ed3-1071e5795723", + "video_length": "00:49:03", + "qid": "9ed1345b-3b40-4179-9ed3-1071e5795723_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many planks have I fixed so far?", + "answer": "You fixed six wooden boards.", + "question_chinese": "到目前为止，我固定了多少根木板？", + "answer_chinese": "你固定了六根木板。", + "evidence": [ + { + "description": "The first board", + "timestamps": { + "start": "00:03:06", + "end": "00:03:26" + } + }, + { + "description": "The second board", + "timestamps": { + "start": "00:04:49", + "end": "00:05:04" + } + }, + { + "description": "The third board.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:38" + } + }, + { + "description": "The fourth board", + "timestamps": { + "start": "00:07:30", + "end": "00:07:49" + } + }, + { + "description": "The fifth board.", + "timestamps": { + "start": "00:08:32", + "end": "00:09:01" + } + }, + { + "description": "The sixth board.", + "timestamps": { + "start": "00:09:44", + "end": "00:10:14" + } + } + ], + "question_timestamps": "00:12:13", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You fixed five wooden boards.", + "B": "You fixed six wooden boards.", + "C": "You fixed seven wooden boards.", + "D": "You fixed four wooden boards in total, not six, based on what you repaired so far.", + "E": "You fixed eight wooden boards overall, which is more than the number you’ve completed up to now." + }, + "answer_label": "B", + "answer_text": "You fixed six wooden boards.", + "q_id": "9ed1345b-3b40-4179-9ed3-1071e5795723_1", + "correct_answer": "You fixed six wooden boards.", + "original_options": { + "A": "You fixed five wooden boards.", + "B": "You fixed six wooden boards.", + "C": "You fixed seven wooden boards.", + "D": "You fixed four wooden boards in total, not six, based on what you repaired so far.", + "E": "You fixed eight wooden boards overall, which is more than the number you’ve completed up to now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "9ed1345b-3b40-4179-9ed3-1071e5795723", + "video_length": "00:49:03", + "qid": "9ed1345b-3b40-4179-9ed3-1071e5795723_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many planks have I fixed so far?", + "answer": "You fixed eleven planks.", + "question_chinese": "到目前为止，我固定了多少根木板？", + "answer_chinese": "你固定了十一根木板。", + "evidence": [ + { + "description": "The first board", + "timestamps": { + "start": "00:03:06", + "end": "00:03:26" + } + }, + { + "description": "The second board", + "timestamps": { + "start": "00:04:49", + "end": "00:05:04" + } + }, + { + "description": "The third board.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:38" + } + }, + { + "description": "The fourth board", + "timestamps": { + "start": "00:07:30", + "end": "00:07:49" + } + }, + { + "description": "The fifth board.", + "timestamps": { + "start": "00:08:32", + "end": "00:09:01" + } + }, + { + "description": "The sixth board.", + "timestamps": { + "start": "00:09:44", + "end": "00:10:14" + } + }, + { + "description": "The seventh board", + "timestamps": { + "start": "00:12:43", + "end": "00:13:40" + } + }, + { + "description": "The eighth board", + "timestamps": { + "start": "00:14:32", + "end": "00:14:52" + } + }, + { + "description": "The ninth board", + "timestamps": { + "start": "00:15:29", + "end": "00:16:02" + } + }, + { + "description": "The tenth board", + "timestamps": { + "start": "00:16:49", + "end": "00:17:18" + } + }, + { + "description": "The Eleventh Board", + "timestamps": { + "start": "00:18:30", + "end": "00:19:22" + } + } + ], + "question_timestamps": "00:24:12", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You fixed nine planks.", + "B": "You fixed fourteen planks.", + "C": "You fixed seven planks in total so far, which is fewer than eleven.", + "D": "You fixed five planks altogether up to this point, not the full eleven.", + "E": "You fixed eleven planks." + }, + "answer_label": "E", + "answer_text": "You fixed eleven planks.", + "q_id": "9ed1345b-3b40-4179-9ed3-1071e5795723_2", + "correct_answer": "You fixed eleven planks.", + "original_options": { + "A": "You fixed nine planks.", + "B": "You fixed fourteen planks.", + "C": "You fixed seven planks in total so far, which is fewer than eleven.", + "D": "You fixed five planks altogether up to this point, not the full eleven.", + "E": "You fixed eleven planks." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 13 + ] + }, + { + "video_uid": "9ed1345b-3b40-4179-9ed3-1071e5795723", + "video_length": "00:49:03", + "qid": "9ed1345b-3b40-4179-9ed3-1071e5795723_9", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I climb the A-frame ladder?", + "answer": "You once climbed a ladder shaped like the Chinese character 'ren'.", + "question_chinese": "我爬上了几次人字梯？", + "answer_chinese": "你爬上过一次人字梯。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:19:45", + "end": "00:19:59" + } + } + ], + "question_timestamps": "00:23:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You climbed it twice.", + "B": "You once climbed a ladder shaped like the Chinese character 'ren'.", + "C": "You never climbed it.", + "D": "You climbed the A-frame ladder twice, not once, during your movement through the area.", + "E": "You went up the ladder three times, so it wasn’t a single climb like you described." + }, + "answer_label": "B", + "answer_text": "You once climbed a ladder shaped like the Chinese character 'ren'.", + "q_id": "9ed1345b-3b40-4179-9ed3-1071e5795723_9", + "correct_answer": "You once climbed a ladder shaped like the Chinese character 'ren'.", + "original_options": { + "A": "You climbed it twice.", + "B": "You once climbed a ladder shaped like the Chinese character 'ren'.", + "C": "You never climbed it.", + "D": "You climbed the A-frame ladder twice, not once, during your movement through the area.", + "E": "You went up the ladder three times, so it wasn’t a single climb like you described." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 17 + ] + }, + { + "video_uid": "9ed1345b-3b40-4179-9ed3-1071e5795723", + "video_length": "00:49:03", + "qid": "9ed1345b-3b40-4179-9ed3-1071e5795723_10", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I climb the A-frame ladder?", + "answer": "You have climbed the A-frame ladder three times.", + "question_chinese": "我爬上了几次人字梯？", + "answer_chinese": "你爬上过三次人字梯。", + "evidence": [ + { + "description": "First time", + "timestamps": { + "start": "00:19:40", + "end": "00:20:00" + } + }, + { + "description": "Second time", + "timestamps": { + "start": "00:23:32", + "end": "00:23:38" + } + }, + { + "description": "Third time", + "timestamps": { + "start": "00:36:50", + "end": "00:37:17" + } + } + ], + "question_timestamps": "00:49:03", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You climbed it twice.", + "B": "You climbed it four times.", + "C": "You have climbed the A-frame ladder three times.", + "D": "You have climbed the A-frame ladder five times, which is more than three total climbs.", + "E": "You have climbed the A-frame ladder once so far, not three times as stated." + }, + "answer_label": "C", + "answer_text": "You have climbed the A-frame ladder three times.", + "q_id": "9ed1345b-3b40-4179-9ed3-1071e5795723_10", + "correct_answer": "You have climbed the A-frame ladder three times.", + "original_options": { + "A": "You climbed it twice.", + "B": "You climbed it four times.", + "C": "You have climbed the A-frame ladder three times.", + "D": "You have climbed the A-frame ladder five times, which is more than three total climbs.", + "E": "You have climbed the A-frame ladder once so far, not three times as stated." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 14 + ] + }, + { + "video_uid": "9ed1345b-3b40-4179-9ed3-1071e5795723", + "video_length": "00:49:03", + "qid": "9ed1345b-3b40-4179-9ed3-1071e5795723_6_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the very beginning, did I see anyone on the stepladder?", + "answer": "There isn’t anyone on the A-frame ladder, and it’s to your front-right.", + "question_chinese": "我看到人字梯上一开始有没有人？", + "answer_chinese": "人字梯上没有人，在你的右前方。", + "evidence": [ + { + "description": "There is no one on the A-frame ladder.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:36:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There is someone on the A-frame ladder, and it’s to your front-right.", + "B": "There isn’t anyone on the A-frame ladder, and it’s to your front-right." + }, + "answer_label": "B", + "answer_text": "There isn’t anyone on the A-frame ladder, and it’s to your front-right.", + "q_id": "9ed1345b-3b40-4179-9ed3-1071e5795723_6_T1", + "correct_answer": "There isn’t anyone on the A-frame ladder, and it’s to your front-right.", + "original_options": { + "A": "There is someone on the A-frame ladder, and it’s to your front-right.", + "B": "There isn’t anyone on the A-frame ladder, and it’s to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "9ed1345b-3b40-4179-9ed3-1071e5795723", + "video_length": "00:49:03", + "qid": "9ed1345b-3b40-4179-9ed3-1071e5795723_6_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the very beginning, did I see anyone on the stepladder?", + "answer": "There isn’t anyone on the A-frame ladder, and it’s to your front-left.", + "question_chinese": "我看到人字梯上一开始有没有人？", + "answer_chinese": "人字梯上没有人，在你的左前方。", + "evidence": [ + { + "description": "There is no one on the A-frame ladder.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:49:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There isn’t anyone on the A-frame ladder, and it’s to your front-left.", + "B": "There is someone on the A-frame ladder, and it’s to your front-left." + }, + "answer_label": "A", + "answer_text": "There isn’t anyone on the A-frame ladder, and it’s to your front-left.", + "q_id": "9ed1345b-3b40-4179-9ed3-1071e5795723_6_T2", + "correct_answer": "There isn’t anyone on the A-frame ladder, and it’s to your front-left.", + "original_options": { + "A": "There isn’t anyone on the A-frame ladder, and it’s to your front-left.", + "B": "There is someone on the A-frame ladder, and it’s to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "9ed1345b-3b40-4179-9ed3-1071e5795723", + "video_length": "00:49:03", + "qid": "9ed1345b-3b40-4179-9ed3-1071e5795723_2_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I first see the tape measure?", + "answer": "The tape measure first appeared under the awning, to your back-left.", + "question_chinese": "我看到卷尺最早出现在哪个地方？", + "answer_chinese": "卷尺最早出现在遮阳棚下面，在你的左后方。", + "evidence": [ + { + "description": "The tape measure first appeared under the awning.", + "timestamps": { + "start": "00:19:17", + "end": "00:19:42" + } + } + ], + "question_timestamps": "00:24:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Under the awning, back-right.", + "B": "The tape measure first appeared under the awning, to your back-left.", + "C": "Under the awning, front-left.", + "D": "The tape measure first appeared under the awning, positioned to your front-right as you looked around.", + "E": "The first time you noticed the tape measure, it was under the awning, directly behind you." + }, + "answer_label": "B", + "answer_text": "The tape measure first appeared under the awning, to your back-left.", + "q_id": "9ed1345b-3b40-4179-9ed3-1071e5795723_2_T1", + "correct_answer": "The tape measure first appeared under the awning, to your back-left.", + "original_options": { + "A": "Under the awning, back-right.", + "B": "The tape measure first appeared under the awning, to your back-left.", + "C": "Under the awning, front-left.", + "D": "The tape measure first appeared under the awning, positioned to your front-right as you looked around.", + "E": "The first time you noticed the tape measure, it was under the awning, directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "9ed1345b-3b40-4179-9ed3-1071e5795723", + "video_length": "00:49:03", + "qid": "9ed1345b-3b40-4179-9ed3-1071e5795723_2_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I first see the tape measure?", + "answer": "The tape measure first appeared under the awning, to your front-right.", + "question_chinese": "我看到卷尺最早出现在哪个地方？", + "answer_chinese": "卷尺最早出现在遮阳棚下面，在你的右前方。", + "evidence": [ + { + "description": "The tape measure first appeared under the awning.", + "timestamps": { + "start": "00:19:17", + "end": "00:19:42" + } + } + ], + "question_timestamps": "00:49:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Under the awning, front-left.", + "B": "Under the awning, back-right.", + "C": "The tape measure first appeared under the awning, to your front-right.", + "D": "The tape measure first appeared under the awning, sitting off to your back-left side.", + "E": "You first spotted the tape measure under the awning, directly behind you as you moved past." + }, + "answer_label": "C", + "answer_text": "The tape measure first appeared under the awning, to your front-right.", + "q_id": "9ed1345b-3b40-4179-9ed3-1071e5795723_2_T2", + "correct_answer": "The tape measure first appeared under the awning, to your front-right.", + "original_options": { + "A": "Under the awning, front-left.", + "B": "Under the awning, back-right.", + "C": "The tape measure first appeared under the awning, to your front-right.", + "D": "The tape measure first appeared under the awning, sitting off to your back-left side.", + "E": "You first spotted the tape measure under the awning, directly behind you as you moved past." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 16 + ] + }, + { + "video_uid": "9ed1345b-3b40-4179-9ed3-1071e5795723", + "video_length": "00:49:03", + "qid": "9ed1345b-3b40-4179-9ed3-1071e5795723_3_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which was closer to me: the white canopy or the white plastic box on the second floor?", + "answer": "The white awning is closer to you, and it’s to your right.", + "question_chinese": "白色遮阳棚和在二楼的白色塑料盒哪个离我更近？", + "answer_chinese": "白色遮阳棚离你更近，在你的右边。", + "evidence": [ + { + "description": "A white plastic box is between me and the sunshade.", + "timestamps": { + "start": "00:11:55", + "end": "00:12:03" + } + } + ], + "question_timestamps": "00:12:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The white awning is closer to you, and it’s to your right.", + "B": "The white plastic box on the second floor is closer to you, and it’s to your front-right." + }, + "answer_label": "A", + "answer_text": "The white awning is closer to you, and it’s to your right.", + "q_id": "9ed1345b-3b40-4179-9ed3-1071e5795723_3_T1", + "correct_answer": "The white awning is closer to you, and it’s to your right.", + "original_options": { + "A": "The white awning is closer to you, and it’s to your right.", + "B": "The white plastic box on the second floor is closer to you, and it’s to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "9ed1345b-3b40-4179-9ed3-1071e5795723", + "video_length": "00:49:03", + "qid": "9ed1345b-3b40-4179-9ed3-1071e5795723_3_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which was closer to me: the white canopy or the white plastic box on the second floor?", + "answer": "The white plastic box is closer to you, and it’s to your front-right.", + "question_chinese": "白色遮阳棚和在二楼的白色塑料盒哪个离我更近？", + "answer_chinese": "白色塑料盒离你更近，在你的右前方。", + "evidence": [ + { + "description": "A white plastic box is between me and the sunshade.", + "timestamps": { + "start": "00:11:55", + "end": "00:12:03" + } + } + ], + "question_timestamps": "00:23:51", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The white canopy is closer to you, and it’s to your front-right.", + "B": "The white plastic box is closer to you, and it’s to your front-right." + }, + "answer_label": "B", + "answer_text": "The white plastic box is closer to you, and it’s to your front-right.", + "q_id": "9ed1345b-3b40-4179-9ed3-1071e5795723_3_T2", + "correct_answer": "The white plastic box is closer to you, and it’s to your front-right.", + "original_options": { + "A": "The white canopy is closer to you, and it’s to your front-right.", + "B": "The white plastic box is closer to you, and it’s to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + } +] diff --git a/QAs/9fc49180-2cd2-4a96-9fcd-0bd54eb969ef-mcq.json b/QAs/9fc49180-2cd2-4a96-9fcd-0bd54eb969ef-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2b73019a60c81740642cb1d129931d7f66e0adfa --- /dev/null +++ b/QAs/9fc49180-2cd2-4a96-9fcd-0bd54eb969ef-mcq.json @@ -0,0 +1,1309 @@ +[ + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_17", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my backpack when I park and stay still?", + "answer": "When your car is parked and stationary, your backpack is on the front passenger seat.", + "question_chinese": "当我停车不动时，我的书包在哪里？", + "answer_chinese": "当你停车不动时，你的书包在副驾驶座位上。", + "evidence": [ + { + "description": "I put the schoolbag in the front passenger seat.", + "timestamps": { + "start": "00:22:25", + "end": "00:22:35" + } + } + ], + "question_timestamps": "00:24:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the driver’s seat.", + "B": "In the trunk.", + "C": "When your car is parked and stationary, your backpack is on the floor near the driver’s seat.", + "D": "When your car is parked and stationary, your backpack is tucked in the glove compartment next to you.", + "E": "When your car is parked and stationary, your backpack is on the front passenger seat." + }, + "answer_label": "E", + "answer_text": "When your car is parked and stationary, your backpack is on the front passenger seat.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_17", + "correct_answer": "When your car is parked and stationary, your backpack is on the front passenger seat.", + "original_options": { + "A": "On the driver’s seat.", + "B": "In the trunk.", + "C": "When your car is parked and stationary, your backpack is on the floor near the driver’s seat.", + "D": "When your car is parked and stationary, your backpack is tucked in the glove compartment next to you.", + "E": "When your car is parked and stationary, your backpack is on the front passenger seat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 18, + 18 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_18", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my backpack when I park and stay still?", + "answer": "When your car is parked and stationary, your backpack is on the front passenger seat.", + "question_chinese": "当我停车不动时，我的书包在哪里？", + "answer_chinese": "当你停车不动时，你的书包在副驾驶座位上。", + "evidence": [ + { + "description": "I put the schoolbag in the front passenger seat.", + "timestamps": { + "start": "00:22:25", + "end": "00:22:35" + } + } + ], + "question_timestamps": "00:28:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "When your car is parked and stationary, your backpack is on the front passenger seat.", + "B": "On the driver’s seat.", + "C": "In the glove box.", + "D": "When your car is parked and stationary, your backpack is on the green floor mat in front of you.", + "E": "When your car is parked and stationary, your backpack is strapped to the roof rack above the car." + }, + "answer_label": "A", + "answer_text": "When your car is parked and stationary, your backpack is on the front passenger seat.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_18", + "correct_answer": "When your car is parked and stationary, your backpack is on the front passenger seat.", + "original_options": { + "A": "When your car is parked and stationary, your backpack is on the front passenger seat.", + "B": "On the driver’s seat.", + "C": "In the glove box.", + "D": "When your car is parked and stationary, your backpack is on the green floor mat in front of you.", + "E": "When your car is parked and stationary, your backpack is strapped to the roof rack above the car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 19, + 18 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_19", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I park the car and stop, what is the straight-line distance in meters to the black car parked in front?", + "answer": "When you stop the car, the straight-line distance to the black car parked in front is four meters.", + "question_chinese": "当我停车不动时，距离前面停着的黑色小车直线距离多少米？", + "answer_chinese": "当你停车不动时，距离前面停着的黑色小车直线距离四米。", + "evidence": [ + { + "description": "A black car parked in front of my car.", + "timestamps": { + "start": "00:15:00", + "end": "00:15:07" + } + } + ], + "question_timestamps": "00:16:55", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five meters.", + "B": "Six meters.", + "C": "When you stop the car, the straight-line distance to the black car parked in front is seven meters.", + "D": "When you stop the car, the straight-line distance to the black car parked in front is four meters.", + "E": "When you stop the car, the straight-line distance to the black car parked in front is eight meters." + }, + "answer_label": "D", + "answer_text": "When you stop the car, the straight-line distance to the black car parked in front is four meters.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_19", + "correct_answer": "When you stop the car, the straight-line distance to the black car parked in front is four meters.", + "original_options": { + "A": "Five meters.", + "B": "Six meters.", + "C": "When you stop the car, the straight-line distance to the black car parked in front is seven meters.", + "D": "When you stop the car, the straight-line distance to the black car parked in front is four meters.", + "E": "When you stop the car, the straight-line distance to the black car parked in front is eight meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 18 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_20", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I park the car and stop, what is the straight-line distance in meters to the black car parked in front?", + "answer": "When you stop the car, the straight-line distance to the black car parked in front is four meters.", + "question_chinese": "当我停车不动时，距离前面停着的黑色小车直线距离多少米？", + "answer_chinese": "当你停车不动时，距离前面停着的黑色小车直线距离四米。", + "evidence": [ + { + "description": "A black car parked in front of my car.", + "timestamps": { + "start": "00:15:00", + "end": "00:15:07" + } + } + ], + "question_timestamps": "00:25:26", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two meters.", + "B": "When you stop the car, the straight-line distance to the black car parked in front is four meters.", + "C": "Five meters.", + "D": "When you stop the car, the straight-line distance to the black car parked in front is six meters.", + "E": "When you stop the car, the straight-line distance to the black car parked in front is seven meters." + }, + "answer_label": "B", + "answer_text": "When you stop the car, the straight-line distance to the black car parked in front is four meters.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_20", + "correct_answer": "When you stop the car, the straight-line distance to the black car parked in front is four meters.", + "original_options": { + "A": "Two meters.", + "B": "When you stop the car, the straight-line distance to the black car parked in front is four meters.", + "C": "Five meters.", + "D": "When you stop the car, the straight-line distance to the black car parked in front is six meters.", + "E": "When you stop the car, the straight-line distance to the black car parked in front is seven meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 18 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_15", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I park the car and stay still, what is the straight-line distance between me and the green trash bin?", + "answer": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty meters.", + "question_chinese": "当我停车不动的时候，我和绿色垃圾桶的直线距离有多远？", + "answer_chinese": "当你停车不动的时候，你和绿色垃圾桶的直线距离是二十米。", + "evidence": [ + { + "description": "The green trash can is diagonally across the street.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:02:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "It is ten meters away.", + "B": "It is thirty meters away.", + "C": "When you park your car and stay still, the straight-line distance to the green trash bin is fifteen meters.", + "D": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty-five meters.", + "E": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty meters." + }, + "answer_label": "E", + "answer_text": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty meters.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_15", + "correct_answer": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty meters.", + "original_options": { + "A": "It is ten meters away.", + "B": "It is thirty meters away.", + "C": "When you park your car and stay still, the straight-line distance to the green trash bin is fifteen meters.", + "D": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty-five meters.", + "E": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 21 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_16", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I park the car and stay still, what is the straight-line distance between me and the green trash bin?", + "answer": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty meters.", + "question_chinese": "当我停车不动的时候，我和绿色垃圾桶的直线距离有多远？", + "answer_chinese": "当你停车不动的时候，你和绿色垃圾桶的直线距离是二十米。", + "evidence": [ + { + "description": "The green trash can is diagonally across the street.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:11:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "It is five meters away.", + "B": "It is thirty meters away.", + "C": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty meters.", + "D": "When you park your car and stay still, the straight-line distance between you and the green trash bin is fifteen meters.", + "E": "When you park your car and stay still, the straight-line distance between you and the green trash bin is ten meters." + }, + "answer_label": "C", + "answer_text": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty meters.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_16", + "correct_answer": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty meters.", + "original_options": { + "A": "It is five meters away.", + "B": "It is thirty meters away.", + "C": "When you park your car and stay still, the straight-line distance between you and the green trash bin is twenty meters.", + "D": "When you park your car and stay still, the straight-line distance between you and the green trash bin is fifteen meters.", + "E": "When you park your car and stay still, the straight-line distance between you and the green trash bin is ten meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 21 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_10", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I stopped the car, how many trash cans did I see?", + "answer": "When you park and stop, you see seven trash cans.", + "question_chinese": "当我停车不动的时候，我看到几个垃圾桶？", + "answer_chinese": "当你停车不动的时候，你看到七个垃圾桶。", + "evidence": [ + { + "description": "There are five trash cans directly in front of the car.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "There are two trash cans placed on the left side of the road by the car.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You see four trash cans.", + "B": "You see eight trash cans.", + "C": "When you park and stop, you see five trash cans lined up near where you stopped.", + "D": "When you park and stop, you see six trash cans in view as you look around.", + "E": "When you park and stop, you see seven trash cans." + }, + "answer_label": "E", + "answer_text": "When you park and stop, you see seven trash cans.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_10", + "correct_answer": "When you park and stop, you see seven trash cans.", + "original_options": { + "A": "You see four trash cans.", + "B": "You see eight trash cans.", + "C": "When you park and stop, you see five trash cans lined up near where you stopped.", + "D": "When you park and stop, you see six trash cans in view as you look around.", + "E": "When you park and stop, you see seven trash cans." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_11", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I stopped the car, how many trash cans did I see?", + "answer": "When you park and stop, you see seven trash cans.", + "question_chinese": "当我停车不动的时候，我看到几个垃圾桶？", + "answer_chinese": "当你停车不动的时候，你看到七个垃圾桶。", + "evidence": [ + { + "description": "There are five trash cans directly in front of the car.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "There are two trash cans placed on the left side of the road by the car.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:13:39", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You see five trash cans.", + "B": "You see nine trash cans.", + "C": "When you park and stop, you see four trash cans in total near the car.", + "D": "When you park and stop, you see seven trash cans.", + "E": "When you park and stop, you see six trash cans around you as you come to a stop." + }, + "answer_label": "D", + "answer_text": "When you park and stop, you see seven trash cans.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_11", + "correct_answer": "When you park and stop, you see seven trash cans.", + "original_options": { + "A": "You see five trash cans.", + "B": "You see nine trash cans.", + "C": "When you park and stop, you see four trash cans in total near the car.", + "D": "When you park and stop, you see seven trash cans.", + "E": "When you park and stop, you see six trash cans around you as you come to a stop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_6", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I sit in the driver's seat, how many gloves are there in the car?", + "answer": "When you sit in the driver's seat, there is a glove on the car.", + "question_chinese": "当我坐在驾驶座时，车上有几只手套？", + "answer_chinese": "当你坐在驾驶座时，车上有一只手套。", + "evidence": [ + { + "description": "There is a black glove on the passenger seat.", + "timestamps": { + "start": "00:16:14", + "end": "00:16:17" + } + } + ], + "question_timestamps": "00:17:18", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are no gloves.", + "B": "There are two gloves.", + "C": "When you sit in the driver's seat, there is a glove on the car.", + "D": "When you sit in the driver's seat, there are three gloves in the car.", + "E": "When you sit in the driver's seat, there are five gloves in the car." + }, + "answer_label": "C", + "answer_text": "When you sit in the driver's seat, there is a glove on the car.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_6", + "correct_answer": "When you sit in the driver's seat, there is a glove on the car.", + "original_options": { + "A": "There are no gloves.", + "B": "There are two gloves.", + "C": "When you sit in the driver's seat, there is a glove on the car.", + "D": "When you sit in the driver's seat, there are three gloves in the car.", + "E": "When you sit in the driver's seat, there are five gloves in the car." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_7", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I sit in the driver's seat, how many gloves are there in the car?", + "answer": "When you sit in the driver's seat, there are three gloves on the car.", + "question_chinese": "当我坐在驾驶座时，车上有几只手套？", + "answer_chinese": "当你坐在驾驶座时，车上有三只手套。", + "evidence": [ + { + "description": "There is a black glove on the passenger seat.", + "timestamps": { + "start": "00:16:14", + "end": "00:16:17" + } + }, + { + "description": "I found a black glove again.", + "timestamps": { + "start": "00:18:15", + "end": "00:18:18" + } + }, + { + "description": "I found a pink glove.", + "timestamps": { + "start": "00:18:29", + "end": "00:18:33" + } + } + ], + "question_timestamps": "00:27:06", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are two gloves.", + "B": "When you sit in the driver's seat, there are three gloves on the car.", + "C": "There are five gloves.", + "D": "When you sit in the driver's seat, there is one glove on the car.", + "E": "When you sit in the driver's seat, there are four gloves on the car." + }, + "answer_label": "B", + "answer_text": "When you sit in the driver's seat, there are three gloves on the car.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_7", + "correct_answer": "When you sit in the driver's seat, there are three gloves on the car.", + "original_options": { + "A": "There are two gloves.", + "B": "When you sit in the driver's seat, there are three gloves on the car.", + "C": "There are five gloves.", + "D": "When you sit in the driver's seat, there is one glove on the car.", + "E": "When you sit in the driver's seat, there are four gloves on the car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_3_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I’m sitting in the driver’s seat, which is closer to me: the black glove or the blue brush?", + "answer": "When you were sitting in the driver’s seat, the blue brush was closer to you. The blue brush was to your right.", + "question_chinese": "当我坐在驾驶座时，黑色手套和蓝色刷子，哪个离我更近？", + "answer_chinese": "当你坐在驾驶座时，蓝色刷子离你更近。蓝色刷子在你的右边。", + "evidence": [ + { + "description": "When you are sitting in the driver's seat, the blue brush is closer to you.", + "timestamps": { + "start": "00:16:44", + "end": "00:17:06" + } + } + ], + "question_timestamps": "00:17:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were sitting in the driver’s seat, the black glove was closer to you. The black glove was to your right.", + "B": "When you were sitting in the driver’s seat, the blue brush was closer to you. The blue brush was to your right." + }, + "answer_label": "B", + "answer_text": "When you were sitting in the driver’s seat, the blue brush was closer to you. The blue brush was to your right.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_3_T2", + "correct_answer": "When you were sitting in the driver’s seat, the blue brush was closer to you. The blue brush was to your right.", + "original_options": { + "A": "When you were sitting in the driver’s seat, the black glove was closer to you. The black glove was to your right.", + "B": "When you were sitting in the driver’s seat, the blue brush was closer to you. The blue brush was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 23 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_3_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I’m sitting in the driver’s seat, which is closer to me: the black glove or the blue brush?", + "answer": "When you were sitting in the driver’s seat, the black glove was closer to you. The black glove was to your right.", + "question_chinese": "当我坐在驾驶座时，黑色手套和蓝色刷子，哪个离我更近？", + "answer_chinese": "当你坐在驾驶座时，黑色手套离你更近。黑色手套在你的右边。", + "evidence": [ + { + "description": "Black gloves are on the front passenger seat, and the blue brush is on the floor of the front passenger side.", + "timestamps": { + "start": "00:16:15", + "end": "00:16:22" + } + } + ], + "question_timestamps": "00:17:34", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were sitting in the driver’s seat, the black glove was closer to you. The black glove was to your right.", + "B": "When you were sitting in the driver’s seat, the blue brush was closer to you. The blue brush was to your right." + }, + "answer_label": "A", + "answer_text": "When you were sitting in the driver’s seat, the black glove was closer to you. The black glove was to your right.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_3_T1", + "correct_answer": "When you were sitting in the driver’s seat, the black glove was closer to you. The black glove was to your right.", + "original_options": { + "A": "When you were sitting in the driver’s seat, the black glove was closer to you. The black glove was to your right.", + "B": "When you were sitting in the driver’s seat, the blue brush was closer to you. The blue brush was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 23 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_5", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I sit in the driver's seat, which is closer to me, the black gloves or the blue brush?", + "answer": "When you sit in the driver's seat, the black gloves are closer to you.", + "question_chinese": "当我坐在驾驶座时，黑色手套和蓝色刷子，哪个离我更近？", + "answer_chinese": "当你坐在驾驶座时，黑色手套离你更近。", + "evidence": [ + { + "description": "Black gloves are on the front passenger seat, and the blue brush is on the floor of the front passenger side.", + "timestamps": { + "start": "00:16:15", + "end": "00:16:22" + } + } + ], + "question_timestamps": "00:29:13", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "When you sit in the driver's seat, the blue brush is closer to you.", + "B": "When you sit in the driver's seat, the black gloves are closer to you." + }, + "answer_label": "B", + "answer_text": "When you sit in the driver's seat, the black gloves are closer to you.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_5", + "correct_answer": "When you sit in the driver's seat, the black gloves are closer to you.", + "original_options": { + "A": "When you sit in the driver's seat, the blue brush is closer to you.", + "B": "When you sit in the driver's seat, the black gloves are closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_8", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I find my driver's license?", + "answer": "I found my driver's license in the wallet inside my school bag.", + "question_chinese": "我在哪里找到的驾驶证？", + "answer_chinese": "我在放在书包的钱包里面，找到了驾驶证。", + "evidence": [ + { + "description": "The driver's license is in the wallet.", + "timestamps": { + "start": "00:22:34", + "end": "00:22:53" + } + } + ], + "question_timestamps": "00:23:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I found my driver's license in the wallet inside my school bag.", + "B": "In the wallet inside my work backpack.", + "C": "In the wallet inside my gym bag.", + "D": "I found my driver's license in the wallet tucked inside my purse after checking other bags.", + "E": "I found my driver's license in the wallet that was inside my coat pocket, not in any bag." + }, + "answer_label": "A", + "answer_text": "I found my driver's license in the wallet inside my school bag.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_8", + "correct_answer": "I found my driver's license in the wallet inside my school bag.", + "original_options": { + "A": "I found my driver's license in the wallet inside my school bag.", + "B": "In the wallet inside my work backpack.", + "C": "In the wallet inside my gym bag.", + "D": "I found my driver's license in the wallet tucked inside my purse after checking other bags.", + "E": "I found my driver's license in the wallet that was inside my coat pocket, not in any bag." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_9", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I find my driver's license?", + "answer": "I found my driver's license in the wallet inside my school bag.", + "question_chinese": "我在哪里找到的驾驶证？", + "answer_chinese": "我在放在书包的钱包里面，找到了驾驶证。", + "evidence": [ + { + "description": "The driver's license is in the wallet.", + "timestamps": { + "start": "00:22:34", + "end": "00:22:53" + } + } + ], + "question_timestamps": "00:29:59", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the drawer by the front door.", + "B": "In the glove box of my car.", + "C": "I found my driver's license in the pocket of my jacket after checking my backpack first.", + "D": "I found my driver's license in the wallet inside my school bag.", + "E": "I found my driver's license in the side compartment of my backpack, not inside the wallet." + }, + "answer_label": "D", + "answer_text": "I found my driver's license in the wallet inside my school bag.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_9", + "correct_answer": "I found my driver's license in the wallet inside my school bag.", + "original_options": { + "A": "In the drawer by the front door.", + "B": "In the glove box of my car.", + "C": "I found my driver's license in the pocket of my jacket after checking my backpack first.", + "D": "I found my driver's license in the wallet inside my school bag.", + "E": "I found my driver's license in the side compartment of my backpack, not inside the wallet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_1", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I'm sitting in the driver's seat, can I reach the black gloves?", + "answer": "You are in the driver's seat and can reach the black gloves.", + "question_chinese": "我坐在驾驶座上，能够得着黑色手套吗？", + "answer_chinese": "你坐在驾驶座上，能够得着黑色手套。", + "evidence": [ + { + "description": "The black gloves are on the passenger seat.", + "timestamps": { + "start": "00:16:13", + "end": "00:16:20" + } + } + ], + "question_timestamps": "00:17:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You are in the driver's seat and can reach the black gloves.", + "B": "You are in the driver's seat but cannot reach the black gloves." + }, + "answer_label": "A", + "answer_text": "You are in the driver's seat and can reach the black gloves.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_1", + "correct_answer": "You are in the driver's seat and can reach the black gloves.", + "original_options": { + "A": "You are in the driver's seat and can reach the black gloves.", + "B": "You are in the driver's seat but cannot reach the black gloves." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_14", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I'm sitting in the driver's seat, can I reach the black gloves?", + "answer": "You are in the driver's seat and can reach the black gloves.", + "question_chinese": "我坐在驾驶座上，能够得着黑色手套吗？", + "answer_chinese": "You are in the driver's seat and can reach the black gloves.", + "evidence": [ + { + "description": "The black gloves are on the passenger seat.", + "timestamps": { + "start": "00:16:13", + "end": "00:16:20" + } + } + ], + "question_timestamps": "00:28:39", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You are in the driver's seat and can reach the black gloves.", + "B": "You are in the driver's seat but cannot reach the black gloves." + }, + "answer_label": "A", + "answer_text": "You are in the driver's seat and can reach the black gloves.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_14", + "correct_answer": "You are in the driver's seat and can reach the black gloves.", + "original_options": { + "A": "You are in the driver's seat and can reach the black gloves.", + "B": "You are in the driver's seat but cannot reach the black gloves." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_7_T1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "While I was driving, how many traffic lights did I pass?", + "answer": "You passed three traffic lights while driving. The traffic lights were behind you.", + "question_chinese": "我开车的途中经过了几个红绿灯？", + "answer_chinese": "你开车的途中经过了三个红绿灯。红绿灯在你的后面。", + "evidence": [ + { + "description": "The second traffic light.", + "timestamps": { + "start": "00:02:51", + "end": "00:03:02" + } + } + ], + "question_timestamps": "00:03:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You passed one traffic light.", + "B": "You passed three traffic lights while driving. The traffic lights were behind you.", + "C": "You passed five traffic lights.", + "D": "You passed two traffic lights while driving, and the traffic lights ended up behind you afterward.", + "E": "You passed four traffic lights while driving, and each traffic light was behind you once you went through." + }, + "answer_label": "B", + "answer_text": "You passed three traffic lights while driving. The traffic lights were behind you.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_7_T1", + "correct_answer": "You passed three traffic lights while driving. The traffic lights were behind you.", + "original_options": { + "A": "You passed one traffic light.", + "B": "You passed three traffic lights while driving. The traffic lights were behind you.", + "C": "You passed five traffic lights.", + "D": "You passed two traffic lights while driving, and the traffic lights ended up behind you afterward.", + "E": "You passed four traffic lights while driving, and each traffic light was behind you once you went through." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_7_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "While I was driving, how many traffic lights did I pass?", + "answer": "You passed eighteen traffic lights while driving. The traffic lights were behind you.", + "question_chinese": "我开车的途中经过了几个红绿灯？", + "answer_chinese": "你开车的途中经过了十八个红绿灯。红绿灯在你的后面。", + "evidence": [ + { + "description": "The first traffic light.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:21" + } + }, + { + "description": "The second traffic light.", + "timestamps": { + "start": "00:02:51", + "end": "00:03:02" + } + }, + { + "description": "The fourth traffic light.", + "timestamps": { + "start": "00:09:16", + "end": "00:09:21" + } + }, + { + "description": "The fifth traffic light.", + "timestamps": { + "start": "00:10:57", + "end": "00:11:25" + } + }, + { + "description": "The sixth traffic light.", + "timestamps": { + "start": "00:12:07", + "end": "00:12:14" + } + }, + { + "description": "The third traffic light.", + "timestamps": { + "start": "00:03:31", + "end": "00:04:19" + } + } + ], + "question_timestamps": "00:12:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You passed seventeen traffic lights while driving.", + "B": "You passed twenty traffic lights while driving.", + "C": "You passed eighteen traffic lights while driving. The traffic lights were behind you.", + "D": "You passed sixteen traffic lights while driving, and by the time you noticed, the lights were already behind you.", + "E": "You passed nineteen traffic lights while driving, with each one ending up behind you as you continued forward." + }, + "answer_label": "C", + "answer_text": "You passed eighteen traffic lights while driving. The traffic lights were behind you.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_7_T2", + "correct_answer": "You passed eighteen traffic lights while driving. The traffic lights were behind you.", + "original_options": { + "A": "You passed seventeen traffic lights while driving.", + "B": "You passed twenty traffic lights while driving.", + "C": "You passed eighteen traffic lights while driving. The traffic lights were behind you.", + "D": "You passed sixteen traffic lights while driving, and by the time you noticed, the lights were already behind you.", + "E": "You passed nineteen traffic lights while driving, with each one ending up behind you as you continued forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 18 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_12", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many traffic lights did I pass while driving?", + "answer": "You passed six traffic lights on your way driving.", + "question_chinese": "我开车的途中经过了几个红绿灯？", + "answer_chinese": "你开车的途中经过了六个红绿灯。", + "evidence": [ + { + "description": "The first traffic light.", + "timestamps": { + "start": "00:04:07", + "end": "00:04:16" + } + }, + { + "description": "The second traffic light.", + "timestamps": { + "start": "00:02:51", + "end": "00:03:02" + } + }, + { + "description": "The third traffic light.", + "timestamps": { + "start": "00:04:07", + "end": "00:04:16" + } + }, + { + "description": "The fourth traffic light.", + "timestamps": { + "start": "00:09:16", + "end": "00:09:21" + } + }, + { + "description": "The fifth traffic light.", + "timestamps": { + "start": "00:10:57", + "end": "00:11:25" + } + }, + { + "description": "The sixth traffic light.", + "timestamps": { + "start": "00:12:07", + "end": "00:12:14" + } + } + ], + "question_timestamps": "00:16:42", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You passed six traffic lights on your way driving.", + "B": "You passed four traffic lights.", + "C": "You passed eight traffic lights.", + "D": "You passed five traffic lights on your way driving, stopping briefly at only a couple of intersections.", + "E": "You passed seven traffic lights on your way driving, continuing through most intersections without turning off." + }, + "answer_label": "A", + "answer_text": "You passed six traffic lights on your way driving.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_12", + "correct_answer": "You passed six traffic lights on your way driving.", + "original_options": { + "A": "You passed six traffic lights on your way driving.", + "B": "You passed four traffic lights.", + "C": "You passed eight traffic lights.", + "D": "You passed five traffic lights on your way driving, stopping briefly at only a couple of intersections.", + "E": "You passed seven traffic lights on your way driving, continuing through most intersections without turning off." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_13", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many traffic lights did I pass while driving?", + "answer": "You passed six traffic lights on your way driving.", + "question_chinese": "我开车的途中经过了几个红绿灯？", + "answer_chinese": "你开车的途中经过了六个红绿灯。", + "evidence": [ + { + "description": "The first traffic light.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:21" + } + }, + { + "description": "The second traffic light.", + "timestamps": { + "start": "00:02:51", + "end": "00:03:02" + } + }, + { + "description": "The third traffic light.", + "timestamps": { + "start": "00:04:07", + "end": "00:04:16" + } + }, + { + "description": "The fourth traffic light.", + "timestamps": { + "start": "00:09:16", + "end": "00:09:21" + } + }, + { + "description": "The fifth traffic light.", + "timestamps": { + "start": "00:10:57", + "end": "00:11:25" + } + }, + { + "description": "The sixth traffic light.", + "timestamps": { + "start": "00:12:07", + "end": "00:12:14" + } + } + ], + "question_timestamps": "00:28:56", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You passed six traffic lights on your way driving.", + "B": "You passed three traffic lights.", + "C": "You passed eight traffic lights.", + "D": "You passed four traffic lights on your way driving, encountering only a few controlled intersections.", + "E": "You passed seven traffic lights on your way driving, going through several intersections before reaching your destination." + }, + "answer_label": "A", + "answer_text": "You passed six traffic lights on your way driving.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_13", + "correct_answer": "You passed six traffic lights on your way driving.", + "original_options": { + "A": "You passed six traffic lights on your way driving.", + "B": "You passed three traffic lights.", + "C": "You passed eight traffic lights.", + "D": "You passed four traffic lights on your way driving, encountering only a few controlled intersections.", + "E": "You passed seven traffic lights on your way driving, going through several intersections before reaching your destination." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_2", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to use wet wipes, where should I get them?", + "answer": "You should go to the hidden compartment in front of the passenger seat to get the wet wipes you want.", + "question_chinese": "我想要用湿纸巾，应该去哪里拿？", + "answer_chinese": "你想要用湿纸巾应该去副驾驶前面的暗格那里拿。", + "evidence": [ + { + "description": "The wet wipes are in the hidden compartment in front of the passenger seat.", + "timestamps": { + "start": "00:17:05", + "end": "00:17:10" + } + } + ], + "question_timestamps": "00:18:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should go to the hidden compartment in front of the passenger seat to get the wet wipes you want.", + "B": "Get them behind the passenger seat.", + "C": "Check the hidden compartment by the driver.", + "D": "You should grab the wet wipes from the hidden compartment behind the passenger seat, not the area in front.", + "E": "You should look in the hidden compartment near the driver’s side, rather than in front of the passenger seat." + }, + "answer_label": "A", + "answer_text": "You should go to the hidden compartment in front of the passenger seat to get the wet wipes you want.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_2", + "correct_answer": "You should go to the hidden compartment in front of the passenger seat to get the wet wipes you want.", + "original_options": { + "A": "You should go to the hidden compartment in front of the passenger seat to get the wet wipes you want.", + "B": "Get them behind the passenger seat.", + "C": "Check the hidden compartment by the driver.", + "D": "You should grab the wet wipes from the hidden compartment behind the passenger seat, not the area in front.", + "E": "You should look in the hidden compartment near the driver’s side, rather than in front of the passenger seat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 20 + ] + }, + { + "video_uid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef", + "video_length": "00:30:00", + "qid": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_3", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to use wet wipes, where should I get them?", + "answer": "You should go to the hidden compartment in front of the passenger seat to get the wet wipes you want.", + "question_chinese": "我想要用湿纸巾，应该去哪里拿？", + "answer_chinese": "你想要用湿纸巾应该去副驾驶前面的暗格那里拿。", + "evidence": [ + { + "description": "The wet wipes are in the hidden compartment in front of the passenger seat.", + "timestamps": { + "start": "00:17:05", + "end": "00:17:10" + } + } + ], + "question_timestamps": "00:27:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should go to the hidden compartment in front of the passenger seat to get the wet wipes you want.", + "B": "Get them behind the passenger seat.", + "C": "Check the hidden compartment by the driver.", + "D": "You should grab the wet wipes from the hidden compartment behind the passenger seat, not the area in front.", + "E": "You should look in the hidden compartment near the driver’s side, rather than in front of the passenger seat." + }, + "answer_label": "A", + "answer_text": "You should go to the hidden compartment in front of the passenger seat to get the wet wipes you want.", + "q_id": "9fc49180-2cd2-4a96-9fcd-0bd54eb969ef_3", + "correct_answer": "You should go to the hidden compartment in front of the passenger seat to get the wet wipes you want.", + "original_options": { + "A": "You should go to the hidden compartment in front of the passenger seat to get the wet wipes you want.", + "B": "Get them behind the passenger seat.", + "C": "Check the hidden compartment by the driver.", + "D": "You should grab the wet wipes from the hidden compartment behind the passenger seat, not the area in front.", + "E": "You should look in the hidden compartment near the driver’s side, rather than in front of the passenger seat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 20 + ] + } +] diff --git a/QAs/9fd11936-e3e9-4043-b2ef-a29c3b3575ed-mcq.json b/QAs/9fd11936-e3e9-4043-b2ef-a29c3b3575ed-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f2eb501946cf9d0f072ce8b5e258043b558c4017 --- /dev/null +++ b/QAs/9fd11936-e3e9-4043-b2ef-a29c3b3575ed-mcq.json @@ -0,0 +1,1231 @@ +[ + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I look at my phone?", + "answer": "You used your phone twice.", + "question_chinese": "我使用了多少次手机呢？", + "answer_chinese": "你使用了两次手机。", + "evidence": [ + { + "description": "It's your first time using a mobile phone.", + "timestamps": { + "start": "00:09:31", + "end": "00:13:46" + } + }, + { + "description": "You are using your phone for the second time.", + "timestamps": { + "start": "00:15:30", + "end": "00:16:18" + } + } + ], + "question_timestamps": "00:16:33", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You used your phone once.", + "B": "You used your phone three times.", + "C": "You used your phone four times, checking it repeatedly during the drive.", + "D": "You used your phone five times, glancing down at it at several different moments.", + "E": "You used your phone twice." + }, + "answer_label": "E", + "answer_text": "You used your phone twice.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_7", + "correct_answer": "You used your phone twice.", + "original_options": { + "A": "You used your phone once.", + "B": "You used your phone three times.", + "C": "You used your phone four times, checking it repeatedly during the drive.", + "D": "You used your phone five times, glancing down at it at several different moments.", + "E": "You used your phone twice." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 14 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I look at my phone?", + "answer": "You used your phone five times.", + "question_chinese": "我使用了多少次手机呢？", + "answer_chinese": "你使用了五次手机。", + "evidence": [ + { + "description": "It's your first time using a mobile phone.", + "timestamps": { + "start": "00:09:31", + "end": "00:13:46" + } + }, + { + "description": "This is your second time using the phone.", + "timestamps": { + "start": "00:15:30", + "end": "00:16:18" + } + }, + { + "description": "This is your third time using your phone.", + "timestamps": { + "start": "00:19:14", + "end": "00:20:29" + } + }, + { + "description": "You are using your phone for the fourth time.", + "timestamps": { + "start": "00:21:23", + "end": "00:22:06" + } + }, + { + "description": "This is your fifth time using the phone.", + "timestamps": { + "start": "00:23:21", + "end": "00:23:30" + } + } + ], + "question_timestamps": "00:24:18", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used your phone four times.", + "B": "You used your phone five times.", + "C": "You used your phone six times.", + "D": "You used your phone seven times, looking at it multiple times throughout the period.", + "E": "You used your phone eight times, checking it frequently at separate moments." + }, + "answer_label": "B", + "answer_text": "You used your phone five times.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_8", + "correct_answer": "You used your phone five times.", + "original_options": { + "A": "You used your phone four times.", + "B": "You used your phone five times.", + "C": "You used your phone six times.", + "D": "You used your phone seven times, looking at it multiple times throughout the period.", + "E": "You used your phone eight times, checking it frequently at separate moments." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 12 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_6_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I go to get the wooden board?", + "answer": "You found the wooden board in the low cabinet to your right—the wooden board is inside the low cabinet to your right.", + "question_chinese": "我去哪里拿的木板？", + "answer_chinese": "你在右边的矮柜中找到的木板，木板在你右边的矮柜里。", + "evidence": [ + { + "description": "you find the wooden board in the low cabinet on your right.", + "timestamps": { + "start": "00:17:00", + "end": "00:17:06" + } + } + ], + "question_timestamps": "00:17:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the left cabinet.", + "B": "You found the wooden board in the low cabinet to your right—the wooden board is inside the low cabinet to your right.", + "C": "From the front low cabinet.", + "D": "You found the wooden board in the low cabinet behind you, not the low cabinet to your right.", + "E": "You found the wooden board in the low cabinet in front of you, rather than inside the low cabinet to your right." + }, + "answer_label": "B", + "answer_text": "You found the wooden board in the low cabinet to your right—the wooden board is inside the low cabinet to your right.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_6_T1", + "correct_answer": "You found the wooden board in the low cabinet to your right—the wooden board is inside the low cabinet to your right.", + "original_options": { + "A": "From the left cabinet.", + "B": "You found the wooden board in the low cabinet to your right—the wooden board is inside the low cabinet to your right.", + "C": "From the front low cabinet.", + "D": "You found the wooden board in the low cabinet behind you, not the low cabinet to your right.", + "E": "You found the wooden board in the low cabinet in front of you, rather than inside the low cabinet to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 22 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_6_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I go to get the wooden board?", + "answer": "You found the wooden board in the low cabinet in front of you—the wooden board is inside the low cabinet in front of you.", + "question_chinese": "我去哪里拿的木板？", + "answer_chinese": "你在前面的矮柜中找到的木板，木板在你前面的矮柜里。", + "evidence": [ + { + "description": "you find the wooden board in the low cabinet in front of you.", + "timestamps": { + "start": "00:17:07", + "end": "00:17:10" + } + } + ], + "question_timestamps": "00:17:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the low cabinet behind you.", + "B": "You found the wooden board in the low cabinet in front of you—the wooden board is inside the low cabinet in front of you.", + "C": "In the low cabinet to your left.", + "D": "You got the wooden board from a low cabinet to your right, not the one directly ahead.", + "E": "You retrieved the wooden board from a low cabinet behind you, rather than the low cabinet in front." + }, + "answer_label": "B", + "answer_text": "You found the wooden board in the low cabinet in front of you—the wooden board is inside the low cabinet in front of you.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_6_T2", + "correct_answer": "You found the wooden board in the low cabinet in front of you—the wooden board is inside the low cabinet in front of you.", + "original_options": { + "A": "In the low cabinet behind you.", + "B": "You found the wooden board in the low cabinet in front of you—the wooden board is inside the low cabinet in front of you.", + "C": "In the low cabinet to your left.", + "D": "You got the wooden board from a low cabinet to your right, not the one directly ahead.", + "E": "You retrieved the wooden board from a low cabinet behind you, rather than the low cabinet in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 18 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_10", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I get the wooden board from?", + "answer": "The wooden board you found in the low cabinet on the left side of the lab equipment.", + "question_chinese": "我去哪里拿的木板？", + "answer_chinese": "你在实验设备左边的矮柜中找到的木板。", + "evidence": [ + { + "description": "You walk from in front of the experimental equipment to the low cabinet on the left side of the experimental equipment.", + "timestamps": { + "start": "00:15:38", + "end": "00:17:14" + } + } + ], + "question_timestamps": "00:17:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the low cabinet on the right.", + "B": "From a top drawer by the desk.", + "C": "You took the wooden board from a tall cabinet beside the lab equipment, not from a low cabinet.", + "D": "The wooden board you found in the low cabinet on the left side of the lab equipment.", + "E": "You retrieved the wooden board from a low cabinet on the opposite side of the lab equipment, not the left." + }, + "answer_label": "D", + "answer_text": "The wooden board you found in the low cabinet on the left side of the lab equipment.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_10", + "correct_answer": "The wooden board you found in the low cabinet on the left side of the lab equipment.", + "original_options": { + "A": "From the low cabinet on the right.", + "B": "From a top drawer by the desk.", + "C": "You took the wooden board from a tall cabinet beside the lab equipment, not from a low cabinet.", + "D": "The wooden board you found in the low cabinet on the left side of the lab equipment.", + "E": "You retrieved the wooden board from a low cabinet on the opposite side of the lab equipment, not the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 20 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_12", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I get the wooden board from?", + "answer": "The wooden board you found in the low cabinet on the left side of the lab equipment.", + "question_chinese": "我去哪里拿的木板？", + "answer_chinese": "你在实验设备左边的矮柜中找到的木板。", + "evidence": [ + { + "description": "You walk from the experimental equipment to the low cabinet on the left side of the equipment.", + "timestamps": { + "start": "00:15:38", + "end": "00:17:14" + } + } + ], + "question_timestamps": "00:44:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the low cabinet on the right.", + "B": "From the high shelf above the sink.", + "C": "The wooden board you found in the low cabinet on the left side of the lab equipment.", + "D": "You got the wooden board from the tall cabinet on the right side of the lab equipment, not the left.", + "E": "You retrieved the wooden board from a drawer under the desk near the lab equipment rather than the low cabinet." + }, + "answer_label": "C", + "answer_text": "The wooden board you found in the low cabinet on the left side of the lab equipment.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_12", + "correct_answer": "The wooden board you found in the low cabinet on the left side of the lab equipment.", + "original_options": { + "A": "From the low cabinet on the right.", + "B": "From the high shelf above the sink.", + "C": "The wooden board you found in the low cabinet on the left side of the lab equipment.", + "D": "You got the wooden board from the tall cabinet on the right side of the lab equipment, not the left.", + "E": "You retrieved the wooden board from a drawer under the desk near the lab equipment rather than the low cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 20 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the trajectory for putting the wrench back into the red tool cabinet?", + "answer": "You should turn to the left, walk straight ahead for five steps, then turn right.", + "question_chinese": "我将扳手放回红色工具柜的行进轨迹？", + "answer_chinese": "你应该向左转身，然后直行五步，再右转。", + "evidence": [ + { + "description": "After taking the hex wrench from the red toolbox, I turned right, walked straight, and then turned left to reach the experimental equipment area.", + "timestamps": { + "start": "01:02:55", + "end": "01:03:13" + } + }, + { + "description": "I turned left from the experimental equipment and walked straight to the red tool cabinet.", + "timestamps": { + "start": "01:01:35", + "end": "01:02:27" + } + } + ], + "question_timestamps": "01:03:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right, walk five steps, then right.", + "B": "Turn left, walk four steps, then right.", + "C": "You should turn to the left, walk straight ahead for five steps, then turn right.", + "D": "You should turn to the left, walk straight ahead for six steps, then turn right to reach it.", + "E": "You should turn to the left, walk straight ahead for five steps, then turn left instead of turning right." + }, + "answer_label": "C", + "answer_text": "You should turn to the left, walk straight ahead for five steps, then turn right.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_1", + "correct_answer": "You should turn to the left, walk straight ahead for five steps, then turn right.", + "original_options": { + "A": "Turn right, walk five steps, then right.", + "B": "Turn left, walk four steps, then right.", + "C": "You should turn to the left, walk straight ahead for five steps, then turn right.", + "D": "You should turn to the left, walk straight ahead for six steps, then turn right to reach it.", + "E": "You should turn to the left, walk straight ahead for five steps, then turn left instead of turning right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 19 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the trajectory for putting the wrench back into the red tool cabinet?", + "answer": "You should turn to the left, walk straight ahead for five steps, then turn right.", + "question_chinese": "我将扳手放回红色工具柜的行进轨迹？", + "answer_chinese": "你应该向左转身，然后直行五步，再右转。", + "evidence": [ + { + "description": "After taking the hex wrench from the red toolbox, I turned right, walked straight, and then turned left to reach the experimental equipment area.", + "timestamps": { + "start": "01:02:55", + "end": "01:03:13" + } + }, + { + "description": "I turned left from the experimental equipment, walked straight to the red tool cabinet.", + "timestamps": { + "start": "01:01:35", + "end": "01:02:27" + } + } + ], + "question_timestamps": "01:06:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right, walk three steps, then left.", + "B": "Turn left, walk five steps, then left.", + "C": "You should turn to the right, walk straight ahead for seven steps, and then turn left at the end.", + "D": "You should turn to the left, walk straight ahead for five steps, then turn right.", + "E": "You should turn to the left, walk straight ahead for eight steps, then turn right to approach the cabinet." + }, + "answer_label": "D", + "answer_text": "You should turn to the left, walk straight ahead for five steps, then turn right.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_2", + "correct_answer": "You should turn to the left, walk straight ahead for five steps, then turn right.", + "original_options": { + "A": "Turn right, walk three steps, then left.", + "B": "Turn left, walk five steps, then left.", + "C": "You should turn to the right, walk straight ahead for seven steps, and then turn left at the end.", + "D": "You should turn to the left, walk straight ahead for five steps, then turn right.", + "E": "You should turn to the left, walk straight ahead for eight steps, then turn right to approach the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_5_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to find the yellow cabinet—how should I walk there?", + "answer": "Go straight until you reach the end; the cabinet is in front of you.", + "question_chinese": "我想去找黄色柜子，我应该怎么走呢？", + "answer_chinese": "你直行到尽头，柜子在你前面。", + "evidence": [ + { + "description": "You go stright.", + "timestamps": { + "start": "00:24:20", + "end": "00:24:30" + } + } + ], + "question_timestamps": "00:24:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go straight; the cabinet is behind you.", + "B": "Go straight until you reach the end; the cabinet is in front of you.", + "C": "Go straight; the cabinet is to your left.", + "D": "Go straight until you reach the end, then the cabinet will be on your right instead of ahead.", + "E": "Walk straight to the end of the path, but the cabinet will be behind you when you stop." + }, + "answer_label": "B", + "answer_text": "Go straight until you reach the end; the cabinet is in front of you.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_5_T2", + "correct_answer": "Go straight until you reach the end; the cabinet is in front of you.", + "original_options": { + "A": "Go straight; the cabinet is behind you.", + "B": "Go straight until you reach the end; the cabinet is in front of you.", + "C": "Go straight; the cabinet is to your left.", + "D": "Go straight until you reach the end, then the cabinet will be on your right instead of ahead.", + "E": "Walk straight to the end of the path, but the cabinet will be behind you when you stop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 18 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_5_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to find the yellow cabinet—how should I walk there?", + "answer": "First turn right, then go straight until you reach the end; the cabinet is in front of you.", + "question_chinese": "我想去找黄色柜子，我应该怎么走呢？", + "answer_chinese": "你先右转,然后直行到尽，柜子在你前面。", + "evidence": [ + { + "description": "you turn right first and then go stright.", + "timestamps": { + "start": "00:24:08", + "end": "00:24:30" + } + } + ], + "question_timestamps": "00:24:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right; the cabinet is behind you.", + "B": "Turn right; the cabinet is to your left.", + "C": "First turn right, then go straight to the end; the cabinet will be on your right, not ahead.", + "D": "First turn right, then go straight until you reach the end; the cabinet ends up behind you rather than in front.", + "E": "First turn right, then go straight until you reach the end; the cabinet is in front of you." + }, + "answer_label": "E", + "answer_text": "First turn right, then go straight until you reach the end; the cabinet is in front of you.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_5_T1", + "correct_answer": "First turn right, then go straight until you reach the end; the cabinet is in front of you.", + "original_options": { + "A": "Turn right; the cabinet is behind you.", + "B": "Turn right; the cabinet is to your left.", + "C": "First turn right, then go straight to the end; the cabinet will be on your right, not ahead.", + "D": "First turn right, then go straight until you reach the end; the cabinet ends up behind you rather than in front.", + "E": "First turn right, then go straight until you reach the end; the cabinet is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 21 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_9", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to find the yellow cabinet, how should I go?", + "answer": "You turn right, go straight to the end, and walk to the yellow cabinet.", + "question_chinese": "我想去找黄色柜子，我应该怎么走呢？", + "answer_chinese": "你右转,直行到尽头,走到黄色柜子。", + "evidence": [ + { + "description": "You face the experimental equipment, then turn right, walk straight to the end, and go to the yellow cabinet.", + "timestamps": { + "start": "00:24:08", + "end": "00:24:30" + } + } + ], + "question_timestamps": "00:25:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You turn left and go to the end.", + "B": "You turn right and walk backward first.", + "C": "You turn right, go straight ahead, and find the yellow cabinet on your left rather than straight ahead.", + "D": "You turn left, go straight to the end, and walk toward a different cabinet instead of the yellow one.", + "E": "You turn right, go straight to the end, and walk to the yellow cabinet." + }, + "answer_label": "E", + "answer_text": "You turn right, go straight to the end, and walk to the yellow cabinet.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_9", + "correct_answer": "You turn right, go straight to the end, and walk to the yellow cabinet.", + "original_options": { + "A": "You turn left and go to the end.", + "B": "You turn right and walk backward first.", + "C": "You turn right, go straight ahead, and find the yellow cabinet on your left rather than straight ahead.", + "D": "You turn left, go straight to the end, and walk toward a different cabinet instead of the yellow one.", + "E": "You turn right, go straight to the end, and walk to the yellow cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 19 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_11", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to find the yellow cabinet, how should I go?", + "answer": "You turn right, go straight to the end, and walk to the yellow cabinet.", + "question_chinese": "我想去找黄色柜子，我应该怎么走呢？", + "answer_chinese": "你右转,直行到尽头,走到黄色柜子。", + "evidence": [ + { + "description": "Face the experimental equipment directly, then turn right, walk straight to the end, and go to the yellow cabinet.", + "timestamps": { + "start": "00:24:08", + "end": "00:24:30" + } + } + ], + "question_timestamps": "00:55:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You turn right, go straight to the end, and walk to the yellow cabinet.", + "B": "Turn left and go straight.", + "C": "Turn right and stop halfway.", + "D": "Turn right, walk straight until the corner, then continue toward the cabinet you see ahead.", + "E": "Turn left, go straight to the end of the hallway, and head to the cabinet on that side." + }, + "answer_label": "A", + "answer_text": "You turn right, go straight to the end, and walk to the yellow cabinet.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_11", + "correct_answer": "You turn right, go straight to the end, and walk to the yellow cabinet.", + "original_options": { + "A": "You turn right, go straight to the end, and walk to the yellow cabinet.", + "B": "Turn left and go straight.", + "C": "Turn right and stop halfway.", + "D": "Turn right, walk straight until the corner, then continue toward the cabinet you see ahead.", + "E": "Turn left, go straight to the end of the hallway, and head to the cabinet on that side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_2_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Am I holding my phone in my hand?", + "answer": "You’re holding the phone in your hand, and the phone is in front of you.", + "question_chinese": "我手上拿着手机吗？", + "answer_chinese": "你手上拿着手机，手机在你前面\n。", + "evidence": [ + { + "description": "The parts of the two experimental devices have been placed on the base of the experimental equipment.", + "timestamps": { + "start": "01:04:54", + "end": "01:04:56" + } + } + ], + "question_timestamps": "01:04:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No, it's in your pocket.", + "B": "No, it's on the table.", + "C": "No, you aren’t holding it; the phone is in your pocket rather than in front of you.", + "D": "No, you don’t have it in hand; it’s been set down nearby instead of held.", + "E": "You’re holding the phone in your hand, and the phone is in front of you." + }, + "answer_label": "E", + "answer_text": "You’re holding the phone in your hand, and the phone is in front of you.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_2_T2", + "correct_answer": "You’re holding the phone in your hand, and the phone is in front of you.", + "original_options": { + "A": "No, it's in your pocket.", + "B": "No, it's on the table.", + "C": "No, you aren’t holding it; the phone is in your pocket rather than in front of you.", + "D": "No, you don’t have it in hand; it’s been set down nearby instead of held.", + "E": "You’re holding the phone in your hand, and the phone is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_2_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Am I holding my phone in my hand?", + "answer": "You’re not holding the phone in your hand; the phone is in your pocket.", + "question_chinese": "我手上拿着手机吗？", + "answer_chinese": "你手上没有拿着手机，手机在你口袋里。", + "evidence": [ + { + "description": "you are not holding a phone in your hand.", + "timestamps": { + "start": "01:05:07", + "end": "01:05:09" + } + } + ], + "question_timestamps": "01:05:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes, you're holding it in your hand.", + "B": "No, it's lying on the table.", + "C": "You're holding the phone in your hand while walking, keeping it in front of you as you move.", + "D": "You aren't holding the phone; you left it on the nearby counter instead of putting it away.", + "E": "You’re not holding the phone in your hand; the phone is in your pocket." + }, + "answer_label": "E", + "answer_text": "You’re not holding the phone in your hand; the phone is in your pocket.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_2_T1", + "correct_answer": "You’re not holding the phone in your hand; the phone is in your pocket.", + "original_options": { + "A": "Yes, you're holding it in your hand.", + "B": "No, it's lying on the table.", + "C": "You're holding the phone in your hand while walking, keeping it in front of you as you move.", + "D": "You aren't holding the phone; you left it on the nearby counter instead of putting it away.", + "E": "You’re not holding the phone in your hand; the phone is in your pocket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_3", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the parts I removed?", + "answer": "All the parts you took apart have been placed on the base of the experimental apparatus.", + "question_chinese": "我拆下来的零件放在哪里了？", + "answer_chinese": "你拆下来的零件都放在实验设备的底座上了。", + "evidence": [ + { + "description": "A part of the experimental equipment is placed on the base of the experimental device.", + "timestamps": { + "start": "00:54:25", + "end": "00:54:58" + } + }, + { + "description": "The parts of the two experimental devices are placed on the base of the experimental equipment.", + "timestamps": { + "start": "00:57:36", + "end": "00:57:54" + } + } + ], + "question_timestamps": "00:58:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the workbench surface.", + "B": "Inside the side compartment.", + "C": "You placed the removed parts on the workbench, spread out in front of the experimental apparatus.", + "D": "You stored all the parts in the storage unit’s top shelf instead of leaving them with the apparatus.", + "E": "All the parts you took apart have been placed on the base of the experimental apparatus." + }, + "answer_label": "E", + "answer_text": "All the parts you took apart have been placed on the base of the experimental apparatus.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_3", + "correct_answer": "All the parts you took apart have been placed on the base of the experimental apparatus.", + "original_options": { + "A": "On the workbench surface.", + "B": "Inside the side compartment.", + "C": "You placed the removed parts on the workbench, spread out in front of the experimental apparatus.", + "D": "You stored all the parts in the storage unit’s top shelf instead of leaving them with the apparatus.", + "E": "All the parts you took apart have been placed on the base of the experimental apparatus." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 19 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_4", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the parts I removed?", + "answer": "You placed all the parts you removed on the base of the laboratory equipment.", + "question_chinese": "我拆下来的零件放在哪里了？", + "answer_chinese": "你拆下来的零件都放在实验设备的底座上了。", + "evidence": [ + { + "description": "The parts of the two experimental devices have been placed on the base of the experimental equipment.", + "timestamps": { + "start": "01:05:17", + "end": "01:05:32" + } + } + ], + "question_timestamps": "01:06:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You placed all the parts you removed on the base of the laboratory equipment.", + "B": "On the shelf nearby.", + "C": "In the equipment drawer.", + "D": "You placed the parts on the stand next to the laboratory equipment, not directly on its base.", + "E": "You set the removed parts on the attached tray of the laboratory equipment rather than on the base." + }, + "answer_label": "A", + "answer_text": "You placed all the parts you removed on the base of the laboratory equipment.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_4", + "correct_answer": "You placed all the parts you removed on the base of the laboratory equipment.", + "original_options": { + "A": "You placed all the parts you removed on the base of the laboratory equipment.", + "B": "On the shelf nearby.", + "C": "In the equipment drawer.", + "D": "You placed the parts on the stand next to the laboratory equipment, not directly on its base.", + "E": "You set the removed parts on the attached tray of the laboratory equipment rather than on the base." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_1_T2", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, in which direction from me is the toolbox?", + "answer": "It’s behind you.", + "question_chinese": "此时工具箱在我的什么方向？", + "answer_chinese": "在你的后面。", + "evidence": [ + { + "description": "There is on behind of you.", + "timestamps": { + "start": "01:03:04", + "end": "01:03:07" + } + } + ], + "question_timestamps": "01:03:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s to your left.", + "C": "It’s to your right, positioned off to the side rather than directly behind you.", + "D": "It’s behind you.", + "E": "It’s above you, not behind you, relative to where you’re currently facing." + }, + "answer_label": "D", + "answer_text": "It’s behind you.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_1_T2", + "correct_answer": "It’s behind you.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s to your left.", + "C": "It’s to your right, positioned off to the side rather than directly behind you.", + "D": "It’s behind you.", + "E": "It’s above you, not behind you, relative to where you’re currently facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 14 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_1_T1", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, in which direction from me is the toolbox?", + "answer": "In front of you.", + "question_chinese": "此时工具箱在我的什么方向？", + "answer_chinese": "在你的前方。", + "evidence": [ + { + "description": "There is on front of you.", + "timestamps": { + "start": "01:02:53", + "end": "01:02:57" + } + } + ], + "question_timestamps": "01:03:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "To your right, offset from your current line of sight instead of directly ahead.", + "E": "It’s above you rather than in front of you from your current viewpoint." + }, + "answer_label": "B", + "answer_text": "In front of you.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_1_T1", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "To your right, offset from your current line of sight instead of directly ahead.", + "E": "It’s above you rather than in front of you from your current viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 14, + 14 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_3_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Relative to my current viewpoint, where is my computer?", + "answer": "The computer is to your right.", + "question_chinese": "相对我当前的视角，我的电脑在哪呢？", + "answer_chinese": "电脑在你的右边。", + "evidence": [ + { + "description": "You are facing the experimental equipment directly.", + "timestamps": { + "start": "00:06:32", + "end": "00:06:37" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer is to your left.", + "B": "The computer is in front of you.", + "C": "The computer is behind you, not to your right, relative to your current viewpoint.", + "D": "The computer is to your right.", + "E": "The computer is above you rather than on your right side from where you’re facing." + }, + "answer_label": "D", + "answer_text": "The computer is to your right.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_3_T2", + "correct_answer": "The computer is to your right.", + "original_options": { + "A": "The computer is to your left.", + "B": "The computer is in front of you.", + "C": "The computer is behind you, not to your right, relative to your current viewpoint.", + "D": "The computer is to your right.", + "E": "The computer is above you rather than on your right side from where you’re facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 16 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_3_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Relative to my current viewpoint, where is my computer?", + "answer": "The computer is in front of you.", + "question_chinese": "相对我当前的视角，我的电脑在哪呢？", + "answer_chinese": "电脑在你的前面。", + "evidence": [ + { + "description": "The computer is on front of the experimental equipment.", + "timestamps": { + "start": "00:06:48", + "end": "00:06:52" + } + } + ], + "question_timestamps": "00:06:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer is in front of you.", + "B": "The computer is behind you.", + "C": "The computer is to your left.", + "D": "The computer is off to your right side, not directly in front of your view.", + "E": "The computer is above your line of sight, positioned higher than where you are looking." + }, + "answer_label": "A", + "answer_text": "The computer is in front of you.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_3_T1", + "correct_answer": "The computer is in front of you.", + "original_options": { + "A": "The computer is in front of you.", + "B": "The computer is behind you.", + "C": "The computer is to your left.", + "D": "The computer is off to your right side, not directly in front of your view.", + "E": "The computer is above your line of sight, positioned higher than where you are looking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_5", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Relative to my current viewpoint, where is my computer?", + "answer": "The computer is to your right.", + "question_chinese": "相对我当前的视角，我的电脑在哪呢？", + "answer_chinese": "电脑在你的右边。", + "evidence": [ + { + "description": "The computer is on the right side of the experimental equipment.", + "timestamps": { + "start": "00:06:40", + "end": "00:06:52" + } + }, + { + "description": "You are facing the experimental equipment.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:07:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The computer is in front of you.", + "B": "The computer is to your right.", + "C": "The computer is to your left.", + "D": "The computer is behind you, rather than positioned on the right side of your body.", + "E": "The computer is below your desk, so it is not located to your right from here." + }, + "answer_label": "B", + "answer_text": "The computer is to your right.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_5", + "correct_answer": "The computer is to your right.", + "original_options": { + "A": "The computer is in front of you.", + "B": "The computer is to your right.", + "C": "The computer is to your left.", + "D": "The computer is behind you, rather than positioned on the right side of your body.", + "E": "The computer is below your desk, so it is not located to your right from here." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 16 + ] + }, + { + "video_uid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed", + "video_length": "01:07:39", + "qid": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_6", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Relative to my current viewpoint, where is my computer?", + "answer": "The computer is on your right.", + "question_chinese": "相对我当前的视角，我的电脑在哪呢？", + "answer_chinese": "电脑在你的右边。", + "evidence": [ + { + "description": "You are facing the experimental equipment directly.", + "timestamps": { + "start": "00:50:22", + "end": "00:50:40" + } + } + ], + "question_timestamps": "00:50:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The computer is in front of you.", + "B": "The computer is behind you.", + "C": "The computer is on your right.", + "D": "The computer is to your left side, not on your right from your current viewpoint.", + "E": "The computer is on the desk in front of you, rather than sitting to your right." + }, + "answer_label": "C", + "answer_text": "The computer is on your right.", + "q_id": "9fd11936-e3e9-4043-b2ef-a29c3b3575ed_6", + "correct_answer": "The computer is on your right.", + "original_options": { + "A": "The computer is in front of you.", + "B": "The computer is behind you.", + "C": "The computer is on your right.", + "D": "The computer is to your left side, not on your right from your current viewpoint.", + "E": "The computer is on the desk in front of you, rather than sitting to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 16 + ] + } +] diff --git a/QAs/D2-P2-am-reception-2_part001-mcq.json b/QAs/D2-P2-am-reception-2_part001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8fdb9ef06c98102ccc909f55049fab872739037d --- /dev/null +++ b/QAs/D2-P2-am-reception-2_part001-mcq.json @@ -0,0 +1,468 @@ +[ + { + "video_uid": "D2-P2-am-reception-2_part001", + "video_length": "00:20:00", + "qid": "D2-P2-am-reception-2_part001_12", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the chair?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接坐在椅子上吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The chair is right next to you.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:06" + } + }, + { + "description": "You can sit directly on it.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "D2-P2-am-reception-2_part001_12", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part001", + "video_length": "00:20:00", + "qid": "D2-P2-am-reception-2_part001_11", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the chair?", + "answer": "No,you can't.", + "question_chinese": "我可以直接坐在椅子上吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The chair is on your right.", + "timestamps": { + "start": "00:03:06", + "end": "00:03:09" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:03:06", + "end": "00:03:09" + } + }, + { + "description": "You cannot sit directly on it.", + "timestamps": { + "start": "00:03:06", + "end": "00:03:09 " + } + } + ], + "question_timestamps": "00:03:13", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "D2-P2-am-reception-2_part001_11", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part001", + "video_length": "00:20:00", + "qid": "D2-P2-am-reception-2_part001_3_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tables and how many chairs did I see in the room?", + "answer": "You saw 8 tables and 4 chairs in the room.", + "question_chinese": "我看见房间里有几张桌子几把椅子？", + "answer_chinese": "你看见房间里有8张桌子，4把椅子。", + "evidence": [ + { + "description": "There are 6 tables and 4 chairs.", + "timestamps": { + "start": "00:06:56", + "end": "00:06:57" + } + } + ], + "question_timestamps": "00:07:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 7 tables and 4 chairs.", + "B": "You saw 8 tables and 4 chairs in the room.", + "C": "You saw 9 tables and 4 chairs.", + "D": "You saw 8 tables and 3 chairs in the room while looking around the seating area.", + "E": "You saw 8 tables and 5 chairs in the room, with an extra chair near one table." + }, + "answer_label": "B", + "answer_text": "You saw 8 tables and 4 chairs in the room.", + "q_id": "D2-P2-am-reception-2_part001_3_T1", + "correct_answer": "You saw 8 tables and 4 chairs in the room.", + "original_options": { + "A": "You saw 7 tables and 4 chairs.", + "B": "You saw 8 tables and 4 chairs in the room.", + "C": "You saw 9 tables and 4 chairs.", + "D": "You saw 8 tables and 3 chairs in the room while looking around the seating area.", + "E": "You saw 8 tables and 5 chairs in the room, with an extra chair near one table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part001", + "video_length": "00:20:00", + "qid": "D2-P2-am-reception-2_part001_3_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tables and how many chairs did I see in the room?", + "answer": "There are 6 tables and 4 chairs in the room.", + "question_chinese": "我看见房间里有几张桌子几把椅子？", + "answer_chinese": "房间里有6张桌子，4把椅子。", + "evidence": [ + { + "description": "There are 6 tables.", + "timestamps": { + "start": "00:06:56", + "end": "00:06:57" + } + }, + { + "description": "There are 4 chairs in the room.", + "timestamps": { + "start": "00:06:56", + "end": "00:06:57" + } + } + ], + "question_timestamps": "00:07:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 7 tables and 4 chairs.", + "B": "There are 6 tables and 3 chairs.", + "C": "There are 6 tables and 4 chairs in the room.", + "D": "There are 5 tables and 4 chairs in the room, with the tables outnumbering the chairs.", + "E": "There are 6 tables and 5 chairs in the room, so you see one more chair than before." + }, + "answer_label": "C", + "answer_text": "There are 6 tables and 4 chairs in the room.", + "q_id": "D2-P2-am-reception-2_part001_3_T2", + "correct_answer": "There are 6 tables and 4 chairs in the room.", + "original_options": { + "A": "There are 7 tables and 4 chairs.", + "B": "There are 6 tables and 3 chairs.", + "C": "There are 6 tables and 4 chairs in the room.", + "D": "There are 5 tables and 4 chairs in the room, with the tables outnumbering the chairs.", + "E": "There are 6 tables and 5 chairs in the room, so you see one more chair than before." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part001", + "video_length": "00:20:00", + "qid": "D2-P2-am-reception-2_part001_9", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I use this facial recognition to unlock the door directly?", + "answer": "No, it's too far away from you.", + "question_chinese": "我能直接使用这个面部解锁进门吗？", + "answer_chinese": "不可以，它离你太远了。", + "evidence": [ + { + "description": "Facial unlocking is too far away from you.", + "timestamps": { + "start": "00:10:42", + "end": "00:10:43" + } + } + ], + "question_timestamps": "00:10:47", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, it's too far away from you.", + "B": "Yes, it's close enough to you." + }, + "answer_label": "A", + "answer_text": "No, it's too far away from you.", + "q_id": "D2-P2-am-reception-2_part001_9", + "correct_answer": "No, it's too far away from you.", + "original_options": { + "A": "No, it's too far away from you.", + "B": "Yes, it's close enough to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part001", + "video_length": "00:20:00", + "qid": "D2-P2-am-reception-2_part001_10", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I use this facial recognition to unlock the door directly?", + "answer": "Yes,you can . It's right in front of you.", + "question_chinese": "我能直接使用这个面部解锁进门吗？", + "answer_chinese": "可以，它就在你面前。", + "evidence": [ + { + "description": "It's right in front of you.", + "timestamps": { + "start": "00:10:54", + "end": "00:10:55" + } + } + ], + "question_timestamps": "00:10:59", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,you can . It's right in front of you.", + "B": "No, you can't. It's too far away from you." + }, + "answer_label": "A", + "answer_text": "Yes,you can . It's right in front of you.", + "q_id": "D2-P2-am-reception-2_part001_10", + "correct_answer": "Yes,you can . It's right in front of you.", + "original_options": { + "A": "Yes,you can . It's right in front of you.", + "B": "No, you can't. It's too far away from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part001", + "video_length": "00:20:00", + "qid": "D2-P2-am-reception-2_part001_15", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tables and chairs are there in the room?", + "answer": "There are six tables and four chairs in the room.", + "question_chinese": "房间里有几张桌子几把椅子？", + "answer_chinese": "房间里有6张桌子，4把椅子。", + "evidence": [ + { + "description": "There are 6 tables", + "timestamps": { + "start": "00:06:56", + "end": "00:06:57" + } + }, + { + "description": "There are and 4 chairs.", + "timestamps": { + "start": "00:06:56", + "end": "00:06:57" + } + } + ], + "question_timestamps": "00:07:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are 6 tables and 3 chairs.", + "B": "There are six tables and four chairs in the room.", + "C": "There are 5 tables and 4 chairs.", + "D": "There are six tables and five chairs in the room, with one extra chair compared to four.", + "E": "There are seven tables and four chairs in the room, so the tables count is higher than six." + }, + "answer_label": "B", + "answer_text": "There are six tables and four chairs in the room.", + "q_id": "D2-P2-am-reception-2_part001_15", + "correct_answer": "There are six tables and four chairs in the room.", + "original_options": { + "A": "There are 6 tables and 3 chairs.", + "B": "There are six tables and four chairs in the room.", + "C": "There are 5 tables and 4 chairs.", + "D": "There are six tables and five chairs in the room, with one extra chair compared to four.", + "E": "There are seven tables and four chairs in the room, so the tables count is higher than six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part001", + "video_length": "00:20:00", + "qid": "D2-P2-am-reception-2_part001_16", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many tables and chairs are there in the room?", + "answer": "There are six tables and four chairs in the room.", + "question_chinese": "房间里有几张桌子几把椅子？", + "answer_chinese": "房间里有6张桌子，4把椅子。", + "evidence": [ + { + "description": "There are 6 tables.", + "timestamps": { + "start": "00:06:56", + "end": "00:06:57" + } + }, + { + "description": "There are 4 chairs in the room.", + "timestamps": { + "start": "00:06:56", + "end": "00:06:57" + } + } + ], + "question_timestamps": "00:07:40", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are 6 tables and 2 chairs.", + "B": "There are six tables and four chairs in the room.", + "C": "There are 6 tables and 5 chairs.", + "D": "There are six tables and three chairs in the room, meaning you see one fewer chair than four.", + "E": "There are five tables and four chairs in the room, so you have one fewer table than six." + }, + "answer_label": "B", + "answer_text": "There are six tables and four chairs in the room.", + "q_id": "D2-P2-am-reception-2_part001_16", + "correct_answer": "There are six tables and four chairs in the room.", + "original_options": { + "A": "There are 6 tables and 2 chairs.", + "B": "There are six tables and four chairs in the room.", + "C": "There are 6 tables and 5 chairs.", + "D": "There are six tables and three chairs in the room, meaning you see one fewer chair than four.", + "E": "There are five tables and four chairs in the room, so you have one fewer table than six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + } +] diff --git a/QAs/D2-P2-am-reception-2_part002-mcq.json b/QAs/D2-P2-am-reception-2_part002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..148aaf23c89bb3c2c05c4dd09b269912075dab99 --- /dev/null +++ b/QAs/D2-P2-am-reception-2_part002-mcq.json @@ -0,0 +1,592 @@ +[ + { + "video_uid": "D2-P2-am-reception-2_part002", + "video_length": "00:20:01", + "qid": "D2-P2-am-reception-2_part002_3_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Can I see the fire extinguisher in front of me right now?", + "answer": "You could see the fire extinguisher in front of you.", + "question_chinese": "我现在在可以看到面前的灭火器吗？", + "answer_chinese": "你可以看到面前的灭火器", + "evidence": [ + { + "description": "There is a fire extinguisher.", + "timestamps": { + "start": "00:12:47", + "end": "00:12:48" + } + }, + { + "description": "There is a fire extinguisher.", + "timestamps": { + "start": "00:12:35", + "end": "00:12:37" + } + } + ], + "question_timestamps": "00:13:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No, you can't see it.", + "B": "Yes, it is behind you.", + "C": "You could see the fire extinguisher in front of you.", + "D": "You could see the fire extinguisher to your left, not directly in front of you.", + "E": "You could see the fire extinguisher off to your right side rather than straight ahead." + }, + "answer_label": "C", + "answer_text": "You could see the fire extinguisher in front of you.", + "q_id": "D2-P2-am-reception-2_part002_3_T1", + "correct_answer": "You could see the fire extinguisher in front of you.", + "original_options": { + "A": "No, you can't see it.", + "B": "Yes, it is behind you.", + "C": "You could see the fire extinguisher in front of you.", + "D": "You could see the fire extinguisher to your left, not directly in front of you.", + "E": "You could see the fire extinguisher off to your right side rather than straight ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part002", + "video_length": "00:20:01", + "qid": "D2-P2-am-reception-2_part002_3_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Can I see the fire extinguisher in front of me right now?", + "answer": "You couldn't see the fire extinguisher in front of you.", + "question_chinese": "我现在在可以看到面前的灭火器吗？", + "answer_chinese": "你不可以看到面前的灭火器", + "evidence": [ + { + "description": "There is a fire extinguisher.", + "timestamps": { + "start": "00:12:35", + "end": "00:12:37" + } + }, + { + "description": "There is a fire extinguisher.", + "timestamps": { + "start": "00:12:42", + "end": "00:12:48" + } + } + ], + "question_timestamps": "00:15:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn't see the fire extinguisher in front of you.", + "B": "You could see the fire extinguisher in front of you." + }, + "answer_label": "A", + "answer_text": "You couldn't see the fire extinguisher in front of you.", + "q_id": "D2-P2-am-reception-2_part002_3_T2", + "correct_answer": "You couldn't see the fire extinguisher in front of you.", + "original_options": { + "A": "You couldn't see the fire extinguisher in front of you.", + "B": "You could see the fire extinguisher in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part002", + "video_length": "00:20:01", + "qid": "D2-P2-am-reception-2_part002_2_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potted green plants did I see in the room?", + "answer": "You saw four pots of green plants in front of you.", + "question_chinese": "我看见房间里有几盆绿植？", + "answer_chinese": "你看见面前的4盆绿植。", + "evidence": [ + { + "description": "There are 2 pots of green plants in the room.", + "timestamps": { + "start": "00:07:48", + "end": "00:07:49" + } + }, + { + "description": "You see two potted plants.", + "timestamps": { + "start": "00:07:54", + "end": "00:07:55" + } + } + ], + "question_timestamps": "00:07:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw three pots of green plants.", + "B": "You saw five pots of green plants.", + "C": "You saw four pots of green plants in front of you.", + "D": "You saw two pots of green plants in front of you, not the larger group of four.", + "E": "You saw six pots of green plants in front of you, with more pots than four visible." + }, + "answer_label": "C", + "answer_text": "You saw four pots of green plants in front of you.", + "q_id": "D2-P2-am-reception-2_part002_2_T2", + "correct_answer": "You saw four pots of green plants in front of you.", + "original_options": { + "A": "You saw three pots of green plants.", + "B": "You saw five pots of green plants.", + "C": "You saw four pots of green plants in front of you.", + "D": "You saw two pots of green plants in front of you, not the larger group of four.", + "E": "You saw six pots of green plants in front of you, with more pots than four visible." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part002", + "video_length": "00:20:01", + "qid": "D2-P2-am-reception-2_part002_2_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potted green plants did I see in the room?", + "answer": "You saw five pots of green plants in front of you.", + "question_chinese": "我看见房间里有几盆绿植？", + "answer_chinese": "你看见面前的5盆绿植。", + "evidence": [ + { + "description": "You see two potted plants.", + "timestamps": { + "start": "00:07:48", + "end": "00:07:49" + } + }, + { + "description": "You spot two small potted plants on the ground.", + "timestamps": { + "start": "00:07:54", + "end": "00:07:55" + } + }, + { + "description": "You spot another small potted plant on the ground.", + "timestamps": { + "start": "00:07:59", + "end": "00:08:00" + } + } + ], + "question_timestamps": "00:08:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw four potted green plants.", + "B": "You saw five pots of green plants in front of you.", + "C": "You saw six potted green plants.", + "D": "You saw three pots of green plants placed in front of you, clustered together in the room.", + "E": "You saw seven pots of green plants in front of you, spread out across the room." + }, + "answer_label": "B", + "answer_text": "You saw five pots of green plants in front of you.", + "q_id": "D2-P2-am-reception-2_part002_2_T1", + "correct_answer": "You saw five pots of green plants in front of you.", + "original_options": { + "A": "You saw four potted green plants.", + "B": "You saw five pots of green plants in front of you.", + "C": "You saw six potted green plants.", + "D": "You saw three pots of green plants placed in front of you, clustered together in the room.", + "E": "You saw seven pots of green plants in front of you, spread out across the room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part002", + "video_length": "00:20:01", + "qid": "D2-P2-am-reception-2_part002_14", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potted plants are there in the room?", + "answer": "There are five pots of green plants in the room.", + "question_chinese": "房间里有几盆绿植？", + "answer_chinese": "房间里有5盆绿植。", + "evidence": [ + { + "description": "There are 2 pots of green plants in the room.", + "timestamps": { + "start": "00:07:56", + "end": "00:07:58" + } + }, + { + "description": "There are 3 pots of green plants in the room.", + "timestamps": { + "start": "00:07:56", + "end": "00:07:58" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are four potted plants.", + "B": "There are six potted plants.", + "C": "There are three pots of green plants in the room, arranged close together along the same area.", + "D": "There are seven pots of green plants in the room, with several spaced farther apart than the others.", + "E": "There are five pots of green plants in the room." + }, + "answer_label": "E", + "answer_text": "There are five pots of green plants in the room.", + "q_id": "D2-P2-am-reception-2_part002_14", + "correct_answer": "There are five pots of green plants in the room.", + "original_options": { + "A": "There are four potted plants.", + "B": "There are six potted plants.", + "C": "There are three pots of green plants in the room, arranged close together along the same area.", + "D": "There are seven pots of green plants in the room, with several spaced farther apart than the others.", + "E": "There are five pots of green plants in the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part002", + "video_length": "00:20:01", + "qid": "D2-P2-am-reception-2_part002_13", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potted plants are there in the room?", + "answer": "There are five pots of green plants in the room.", + "question_chinese": "房间里有几盆绿植？", + "answer_chinese": "房间里有5盆绿植。", + "evidence": [ + { + "description": "There are 2 pots of green plants in the room.", + "timestamps": { + "start": "00:07:55", + "end": "00:07:58" + } + }, + { + "description": "There are 3 pots of green plants in the room.", + "timestamps": { + "start": "00:07:55", + "end": "00:07:58" + } + } + ], + "question_timestamps": "00:08:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are four potted plants.", + "B": "There are six potted plants.", + "C": "There are two pots of green plants in the room, positioned near each other in the same section.", + "D": "There are five pots of green plants in the room.", + "E": "There are seven pots of green plants in the room, visible at once across the space." + }, + "answer_label": "D", + "answer_text": "There are five pots of green plants in the room.", + "q_id": "D2-P2-am-reception-2_part002_13", + "correct_answer": "There are five pots of green plants in the room.", + "original_options": { + "A": "There are four potted plants.", + "B": "There are six potted plants.", + "C": "There are two pots of green plants in the room, positioned near each other in the same section.", + "D": "There are five pots of green plants in the room.", + "E": "There are seven pots of green plants in the room, visible at once across the space." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 16 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part002", + "video_length": "00:20:01", + "qid": "D2-P2-am-reception-2_part002_15", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Do you have a fire extinguisher?", + "answer": "Yes,there is.", + "question_chinese": "有灭火器吗？", + "answer_chinese": "有。", + "evidence": [ + { + "description": "There is a fire extinguisher.", + "timestamps": { + "start": "00:12:47", + "end": "00:12:48" + } + }, + { + "description": "There is a fire extinguisher.", + "timestamps": { + "start": "00:12:35", + "end": "00:12:37" + } + } + ], + "question_timestamps": "00:13:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,there is.", + "B": "No, there isn't." + }, + "answer_label": "A", + "answer_text": "Yes,there is.", + "q_id": "D2-P2-am-reception-2_part002_15", + "correct_answer": "Yes,there is.", + "original_options": { + "A": "Yes,there is.", + "B": "No, there isn't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part002", + "video_length": "00:20:01", + "qid": "D2-P2-am-reception-2_part002_16", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Do you have a fire extinguisher?", + "answer": "Yes,there is.", + "question_chinese": "有灭火器吗？", + "answer_chinese": "有。", + "evidence": [ + { + "description": "There is a fire extinguisher.", + "timestamps": { + "start": "00:12:35", + "end": "00:12:37" + } + }, + { + "description": "There is a fire extinguisher.", + "timestamps": { + "start": "00:12:42", + "end": "00:12:48" + } + } + ], + "question_timestamps": "00:15:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, there isn't.", + "B": "Yes,there is." + }, + "answer_label": "B", + "answer_text": "Yes,there is.", + "q_id": "D2-P2-am-reception-2_part002_16", + "correct_answer": "Yes,there is.", + "original_options": { + "A": "No, there isn't.", + "B": "Yes,there is." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part002", + "video_length": "00:20:01", + "qid": "D2-P2-am-reception-2_part002_1_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the computer from me?", + "answer": "The computer is in front of you.", + "question_chinese": "电脑在我的哪个方向？", + "answer_chinese": "电脑在你正前方。", + "evidence": [ + { + "description": "The computer is in front of you.", + "timestamps": { + "start": "00:10:57", + "end": "00:10:58" + } + } + ], + "question_timestamps": "00:11:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer is behind you.", + "B": "The computer is to your left.", + "C": "The computer is positioned to your right side rather than directly in front of you.", + "D": "The computer is in front of you.", + "E": "The computer is above you, not sitting in the space directly ahead of your view." + }, + "answer_label": "D", + "answer_text": "The computer is in front of you.", + "q_id": "D2-P2-am-reception-2_part002_1_T1", + "correct_answer": "The computer is in front of you.", + "original_options": { + "A": "The computer is behind you.", + "B": "The computer is to your left.", + "C": "The computer is positioned to your right side rather than directly in front of you.", + "D": "The computer is in front of you.", + "E": "The computer is above you, not sitting in the space directly ahead of your view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "D2-P2-am-reception-2_part002", + "video_length": "00:20:01", + "qid": "D2-P2-am-reception-2_part002_1_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the computer from me?", + "answer": "The computer is to your front-right.", + "question_chinese": "电脑在我的哪个方向？", + "answer_chinese": "电脑在你右前方。", + "evidence": [ + { + "description": "The computer is in front of you.", + "timestamps": { + "start": "00:10:57", + "end": "00:10:58" + } + } + ], + "question_timestamps": "00:11:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer is to your front-left.", + "B": "The computer is to your front-right.", + "C": "The computer is to your back-right.", + "D": "The computer is directly in front of you rather than angled off toward your front-right side.", + "E": "The computer is behind you on the right side, not located in the forward-right area." + }, + "answer_label": "B", + "answer_text": "The computer is to your front-right.", + "q_id": "D2-P2-am-reception-2_part002_1_T2", + "correct_answer": "The computer is to your front-right.", + "original_options": { + "A": "The computer is to your front-left.", + "B": "The computer is to your front-right.", + "C": "The computer is to your back-right.", + "D": "The computer is directly in front of you rather than angled off toward your front-right side.", + "E": "The computer is behind you on the right side, not located in the forward-right area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + } +] diff --git a/QAs/D2-P3-am-shopping_part001-mcq.json b/QAs/D2-P3-am-shopping_part001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fe2efaadeaac9560c55e44850d83be406c74a038 --- /dev/null +++ b/QAs/D2-P3-am-shopping_part001-mcq.json @@ -0,0 +1,514 @@ +[ + { + "video_uid": "D2-P3-am-shopping_part001", + "video_length": "00:20:00", + "qid": "D2-P3-am-shopping_part001_11", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is Miniso located from me?", + "answer": "It's on your left.", + "question_chinese": "名创优品在我的哪个方向？", + "answer_chinese": "它在你左侧。", + "evidence": [ + { + "description": "It's on your left.", + "timestamps": { + "start": "00:07:16", + "end": "00:07:17" + } + } + ], + "question_timestamps": "00:07:22", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "It's on your right.", + "B": "It's on your left." + }, + "answer_label": "B", + "answer_text": "It's on your left.", + "q_id": "D2-P3-am-shopping_part001_11", + "correct_answer": "It's on your left.", + "original_options": { + "A": "It's on your right.", + "B": "It's on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part001", + "video_length": "00:20:00", + "qid": "D2-P3-am-shopping_part001_10", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is Miniso located from me?", + "answer": "It's behind you.", + "question_chinese": "名创优品在我的哪个方向？", + "answer_chinese": "它在你后方。", + "evidence": [ + { + "description": "Miniso is on your left.", + "timestamps": { + "start": "00:07:12", + "end": "00:07:13" + } + }, + { + "description": "Miniso is behind you.", + "timestamps": { + "start": "00:07:18", + "end": "00:07:25" + } + } + ], + "question_timestamps": "00:07:28", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It's in front of you.", + "B": "It's behind you.", + "C": "It's to your left.", + "D": "It's to your right, just off your side as you face forward.", + "E": "It's ahead of you, so you need to keep walking forward to reach it." + }, + "answer_label": "B", + "answer_text": "It's behind you.", + "q_id": "D2-P3-am-shopping_part001_10", + "correct_answer": "It's behind you.", + "original_options": { + "A": "It's in front of you.", + "B": "It's behind you.", + "C": "It's to your left.", + "D": "It's to your right, just off your side as you face forward.", + "E": "It's ahead of you, so you need to keep walking forward to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 12, + 14 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part001", + "video_length": "00:20:00", + "qid": "D2-P3-am-shopping_part001_12", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the chair?", + "answer": "No,you can't.", + "question_chinese": "我可以直接坐在椅子上吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The chair is in front of you.", + "timestamps": { + "start": "00:12:16", + "end": "00:12:17" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:12:16", + "end": "00:12:17" + } + }, + { + "description": "You cannot sit directly on it.", + "timestamps": { + "start": "00:12:16", + "end": "00:12:17" + } + } + ], + "question_timestamps": "00:12:22", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "D2-P3-am-shopping_part001_12", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part001", + "video_length": "00:20:00", + "qid": "D2-P3-am-shopping_part001_13", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "May I sit directly on the chair?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接坐在椅子上吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The chair is right next to you.", + "timestamps": { + "start": "00:12:23", + "end": "00:12:24" + } + }, + { + "description": "You can sit directly on it.", + "timestamps": { + "start": "00:12:23", + "end": "00:12:24" + } + } + ], + "question_timestamps": "00:12:28", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "D2-P3-am-shopping_part001_13", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part001", + "video_length": "00:20:00", + "qid": "D2-P3-am-shopping_part001_15", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Have I picked up and looked at a few energy bowls?", + "answer": "Two.", + "question_chinese": "我拿起来几个能量碗看过？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "You looked at an energy bowl.", + "timestamps": { + "start": "00:10:28", + "end": "00:10:29" + } + }, + { + "description": "You looked at an energy bowl.", + "timestamps": { + "start": "00:11:18", + "end": "00:11:19" + } + } + ], + "question_timestamps": "00:15:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You picked up and looked at just one energy bowl in the clip shown.", + "E": "You picked up and looked at three energy bowls before stopping and moving on." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "D2-P3-am-shopping_part001_15", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You picked up and looked at just one energy bowl in the clip shown.", + "E": "You picked up and looked at three energy bowls before stopping and moving on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part001", + "video_length": "00:20:00", + "qid": "D2-P3-am-shopping_part001_4_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What snacks did I pass by?", + "answer": "You passed the rice noodles, and the rice noodles are to your back-right.", + "question_chinese": "我经过了什么小吃？", + "answer_chinese": "你经过了拌粉，拌粉在你的右后方。", + "evidence": [ + { + "description": "You passed the rice noodle mixing area.", + "timestamps": { + "start": "00:12:18", + "end": "00:12:20" + } + } + ], + "question_timestamps": "00:12:27", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You passed the rice noodles, and the rice noodles are to your back-right.", + "B": "You walked past Hippo Stir-Fry. Hippo Stir-Fry was in front of you." + }, + "answer_label": "A", + "answer_text": "You passed the rice noodles, and the rice noodles are to your back-right.", + "q_id": "D2-P3-am-shopping_part001_4_T1", + "correct_answer": "You passed the rice noodles, and the rice noodles are to your back-right.", + "original_options": { + "A": "You passed the rice noodles, and the rice noodles are to your back-right.", + "B": "You walked past Hippo Stir-Fry. Hippo Stir-Fry was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part001", + "video_length": "00:20:00", + "qid": "D2-P3-am-shopping_part001_4_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What snacks did I pass by?", + "answer": "You walked past Hippo Stir-Fry. Hippo Stir-Fry was in front of you.", + "question_chinese": "我经过了什么小吃？", + "answer_chinese": "你经过了河马小炒。河马小炒在你的前面。", + "evidence": [ + { + "description": "You passed Hema Stir-Fry area.", + "timestamps": { + "start": "00:11:58", + "end": "00:12:04" + } + } + ], + "question_timestamps": "00:12:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You passed the rice noodles, and the rice noodles are to your back-right.", + "B": "You walked past Hippo Stir-Fry. Hippo Stir-Fry was in front of you." + }, + "answer_label": "B", + "answer_text": "You walked past Hippo Stir-Fry. Hippo Stir-Fry was in front of you.", + "q_id": "D2-P3-am-shopping_part001_4_T2", + "correct_answer": "You walked past Hippo Stir-Fry. Hippo Stir-Fry was in front of you.", + "original_options": { + "A": "You passed the rice noodles, and the rice noodles are to your back-right.", + "B": "You walked past Hippo Stir-Fry. Hippo Stir-Fry was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part001", + "video_length": "00:20:00", + "qid": "D2-P3-am-shopping_part001_6", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there anyone else in the elevator besides me?", + "answer": "No,there isn't.", + "question_chinese": "电梯里除了我有其他人吗？", + "answer_chinese": "没有。", + "evidence": [ + { + "description": "The elevator was empty before you went up.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + }, + { + "description": "Only you got into the elevator yourself.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:34", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, there is.", + "B": "No,there isn't." + }, + "answer_label": "B", + "answer_text": "No,there isn't.", + "q_id": "D2-P3-am-shopping_part001_6", + "correct_answer": "No,there isn't.", + "original_options": { + "A": "Yes, there is.", + "B": "No,there isn't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part001", + "video_length": "00:20:00", + "qid": "D2-P3-am-shopping_part001_7", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there anyone else in the elevator besides me?", + "answer": "Yes,there is.", + "question_chinese": "电梯里除了我有其他人吗？", + "answer_chinese": "有。", + "evidence": [ + { + "description": "The elevator was empty before you went up.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + }, + { + "description": "Only you got into the elevator yourself.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + }, + { + "description": "Another person has come up.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,there is.", + "B": "No, there isn't." + }, + "answer_label": "A", + "answer_text": "Yes,there is.", + "q_id": "D2-P3-am-shopping_part001_7", + "correct_answer": "Yes,there is.", + "original_options": { + "A": "Yes,there is.", + "B": "No, there isn't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + } +] diff --git a/QAs/D2-P3-am-shopping_part002-mcq.json b/QAs/D2-P3-am-shopping_part002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c82c2d67c121df3dd9f0d116ef0f1df7f53391da --- /dev/null +++ b/QAs/D2-P3-am-shopping_part002-mcq.json @@ -0,0 +1,578 @@ +[ + { + "video_uid": "D2-P3-am-shopping_part002", + "video_length": "00:20:01", + "qid": "D2-P3-am-shopping_part002_15", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the exit?", + "answer": "The exit is on your right.", + "question_chinese": "出口在我的哪个方向？", + "answer_chinese": "出口在你右边。", + "evidence": [ + { + "description": "The exit is on your right.", + "timestamps": { + "start": "00:18:58", + "end": "00:19:00" + } + } + ], + "question_timestamps": "00:19:04", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The exit is behind you.", + "B": "The exit is on your right." + }, + "answer_label": "B", + "answer_text": "The exit is on your right.", + "q_id": "D2-P3-am-shopping_part002_15", + "correct_answer": "The exit is on your right.", + "original_options": { + "A": "The exit is behind you.", + "B": "The exit is on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part002", + "video_length": "00:20:01", + "qid": "D2-P3-am-shopping_part002_16", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the exit?", + "answer": "The exit is behind you.", + "question_chinese": "出口在我的哪个方向？", + "answer_chinese": "出口在你后方。", + "evidence": [ + { + "description": "The exit is behind you.", + "timestamps": { + "start": "00:19:05", + "end": "00:19:06" + } + }, + { + "description": "The exit is behind you.", + "timestamps": { + "start": "00:19:00", + "end": "00:19:01" + } + } + ], + "question_timestamps": "00:19:09", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The exit is behind you.", + "B": "In front of you.", + "C": "On your right.", + "D": "The exit is above you, not at your back, so you need to look upward to spot it.", + "E": "The exit is below you near the floor area, so you would have to look down to see it." + }, + "answer_label": "A", + "answer_text": "The exit is behind you.", + "q_id": "D2-P3-am-shopping_part002_16", + "correct_answer": "The exit is behind you.", + "original_options": { + "A": "The exit is behind you.", + "B": "In front of you.", + "C": "On your right.", + "D": "The exit is above you, not at your back, so you need to look upward to spot it.", + "E": "The exit is below you near the floor area, so you would have to look down to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 18, + 19 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part002", + "video_length": "00:20:01", + "qid": "D2-P3-am-shopping_part002_2_T2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the coffee liquid or the whole milk?", + "answer": "The milk was closer to you.", + "question_chinese": "咖啡液和纯牛奶哪个离我更近？", + "answer_chinese": "牛奶离你更近。", + "evidence": [ + { + "description": "Milk is in front of you.", + "timestamps": { + "start": "00:19:33", + "end": "00:19:35" + } + } + ], + "question_timestamps": "00:19:39", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The coffee concentrate was closer to you.", + "B": "The milk was closer to you." + }, + "answer_label": "B", + "answer_text": "The milk was closer to you.", + "q_id": "D2-P3-am-shopping_part002_2_T2", + "correct_answer": "The milk was closer to you.", + "original_options": { + "A": "The coffee concentrate was closer to you.", + "B": "The milk was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part002", + "video_length": "00:20:01", + "qid": "D2-P3-am-shopping_part002_2_T1", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the coffee liquid or the whole milk?", + "answer": "The coffee concentrate was closer to you.", + "question_chinese": "咖啡液和纯牛奶哪个离我更近？", + "answer_chinese": "咖啡液离你更近。", + "evidence": [ + { + "description": "The coffee is right in front of you.", + "timestamps": { + "start": "00:19:45", + "end": "00:19:46" + } + }, + { + "description": "The milk is on your right.", + "timestamps": { + "start": "00:19:44", + "end": "00:19:45" + } + } + ], + "question_timestamps": "00:19:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The coffee concentrate was closer to you.", + "B": "The milk was closer to you." + }, + "answer_label": "A", + "answer_text": "The coffee concentrate was closer to you.", + "q_id": "D2-P3-am-shopping_part002_2_T1", + "correct_answer": "The coffee concentrate was closer to you.", + "original_options": { + "A": "The coffee concentrate was closer to you.", + "B": "The milk was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part002", + "video_length": "00:20:01", + "qid": "D2-P3-am-shopping_part002_3_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What book review am I reading right now?", + "answer": "You kept browsing the homepage and didn’t look at any book reviews.", + "question_chinese": "我现在在看什么书的书评？", + "answer_chinese": "你一直在首页浏览，没有看书评", + "evidence": [ + { + "description": "You have been just browsing on the homepage and haven't read the book reviews.", + "timestamps": { + "start": "00:15:06", + "end": "00:15:14" + } + } + ], + "question_timestamps": "00:15:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "“Not Forgiving Is Okay.”.", + "B": "A different book review.", + "C": "You kept browsing the homepage and didn’t look at any book reviews.", + "D": "You are reading a specific book review page instead of the homepage, focusing on the review content.", + "E": "You are currently reading the book review titled “Not Forgiving Is Okay.” rather than browsing the homepage." + }, + "answer_label": "C", + "answer_text": "You kept browsing the homepage and didn’t look at any book reviews.", + "q_id": "D2-P3-am-shopping_part002_3_T1", + "correct_answer": "You kept browsing the homepage and didn’t look at any book reviews.", + "original_options": { + "A": "“Not Forgiving Is Okay.”.", + "B": "A different book review.", + "C": "You kept browsing the homepage and didn’t look at any book reviews.", + "D": "You are reading a specific book review page instead of the homepage, focusing on the review content.", + "E": "You are currently reading the book review titled “Not Forgiving Is Okay.” rather than browsing the homepage." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part002", + "video_length": "00:20:01", + "qid": "D2-P3-am-shopping_part002_11", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What book review am I currently reading?", + "answer": "You are reading the book review of ‘不原谅也没关系’.", + "question_chinese": "我现在在看什么书的书评？", + "answer_chinese": "《不原谅也没关系》。", + "evidence": [ + { + "description": "You are reading the book review of \"It's Okay Not to Forgive.\"", + "timestamps": { + "start": "00:18:26", + "end": "00:18:27" + } + }, + { + "description": "You are reading the book review of \"It's Okay Not to Forgive.\"", + "timestamps": { + "start": "00:18:20", + "end": "00:18:24" + } + } + ], + "question_timestamps": "00:18:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "No book review.", + "B": "You are reading the book review of ‘不原谅也没关系’.", + "C": "A different book review.", + "D": "You kept browsing the homepage and didn’t open any review page, so you aren’t reading a book review now.", + "E": "You are reading a different book review in English rather than the one titled ‘不原谅也没关系’." + }, + "answer_label": "B", + "answer_text": "You are reading the book review of ‘不原谅也没关系’.", + "q_id": "D2-P3-am-shopping_part002_11", + "correct_answer": "You are reading the book review of ‘不原谅也没关系’.", + "original_options": { + "A": "No book review.", + "B": "You are reading the book review of ‘不原谅也没关系’.", + "C": "A different book review.", + "D": "You kept browsing the homepage and didn’t open any review page, so you aren’t reading a book review now.", + "E": "You are reading a different book review in English rather than the one titled ‘不原谅也没关系’." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 21, + 14 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part002", + "video_length": "00:20:01", + "qid": "D2-P3-am-shopping_part002_3_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What book review am I reading right now?", + "answer": "“Not Forgiving Is Okay.”.", + "question_chinese": "我现在在看什么书的书评？", + "answer_chinese": "《不原谅也没关系》。", + "evidence": [ + { + "description": "You are reading the book review of \"It's Okay Not to Forgive.\"", + "timestamps": { + "start": "00:18:26", + "end": "00:18:27" + } + }, + { + "description": "You are reading the book review of \"It's Okay Not to Forgive.\"", + "timestamps": { + "start": "00:18:20", + "end": "00:18:24" + } + } + ], + "question_timestamps": "00:19:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No book review.", + "B": "“Not Forgiving Is Okay.”.", + "C": "A different review.", + "D": "You are reading the book review of ‘不原谅也没关系’ rather than the one titled “Not Forgiving Is Okay.”.", + "E": "You kept browsing the homepage and did not open any book review page, so you aren’t reading a review now." + }, + "answer_label": "B", + "answer_text": "“Not Forgiving Is Okay.”.", + "q_id": "D2-P3-am-shopping_part002_3_T2", + "correct_answer": "“Not Forgiving Is Okay.”.", + "original_options": { + "A": "No book review.", + "B": "“Not Forgiving Is Okay.”.", + "C": "A different review.", + "D": "You are reading the book review of ‘不原谅也没关系’ rather than the one titled “Not Forgiving Is Okay.”.", + "E": "You kept browsing the homepage and did not open any book review page, so you aren’t reading a review now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 21 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part002", + "video_length": "00:20:01", + "qid": "D2-P3-am-shopping_part002_4_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many e-books have I opened?", + "answer": "You opened one e-book, and the e-book was on your phone in your hand.", + "question_chinese": "截止到现在我打开了几本电子书？", + "answer_chinese": "你打开了一本，电子书在你手中的手机上。", + "evidence": [ + { + "description": "You are reading the first book.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened zero e-books.", + "B": "You opened two e-books.", + "C": "You opened one e-book, and the e-book was on your phone in your hand.", + "D": "You opened three e-books, and they were on your phone in your hand throughout.", + "E": "You opened four e-books, and all of them were on the phone you were holding." + }, + "answer_label": "C", + "answer_text": "You opened one e-book, and the e-book was on your phone in your hand.", + "q_id": "D2-P3-am-shopping_part002_4_T1", + "correct_answer": "You opened one e-book, and the e-book was on your phone in your hand.", + "original_options": { + "A": "You opened zero e-books.", + "B": "You opened two e-books.", + "C": "You opened one e-book, and the e-book was on your phone in your hand.", + "D": "You opened three e-books, and they were on your phone in your hand throughout.", + "E": "You opened four e-books, and all of them were on the phone you were holding." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part002", + "video_length": "00:20:01", + "qid": "D2-P3-am-shopping_part002_14", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many e-books have I opened so far?", + "answer": "Two.", + "question_chinese": "截止到现在我打开了几本电子书？", + "answer_chinese": "两本。", + "evidence": [ + { + "description": "You are reading the first book.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You are reading the first book.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four e-books opened in total so far, as counted from the titles you viewed on the device.", + "E": "Five e-books opened up to this point, based on the number of separate book screens you entered." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "D2-P3-am-shopping_part002_14", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four e-books opened in total so far, as counted from the titles you viewed on the device.", + "E": "Five e-books opened up to this point, based on the number of separate book screens you entered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 17 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part002", + "video_length": "00:20:01", + "qid": "D2-P3-am-shopping_part002_4_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many e-books have I opened?", + "answer": "You opened two e-books, and they were on the phone in your hand.", + "question_chinese": "截止到现在我打开了几本电子书？", + "answer_chinese": "你打开了两本，电子书在你手中的手机上。", + "evidence": [ + { + "description": "You are reading the first book.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You are reading the second book.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + }, + { + "description": "You are reading the first book.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened one e-book.", + "B": "You opened three e-books.", + "C": "You opened zero e-books, and none were opened on the phone in your hand.", + "D": "You opened two e-books, and they were on the phone in your hand.", + "E": "You opened four e-books, and they were all on the phone you were holding." + }, + "answer_label": "D", + "answer_text": "You opened two e-books, and they were on the phone in your hand.", + "q_id": "D2-P3-am-shopping_part002_4_T2", + "correct_answer": "You opened two e-books, and they were on the phone in your hand.", + "original_options": { + "A": "You opened one e-book.", + "B": "You opened three e-books.", + "C": "You opened zero e-books, and none were opened on the phone in your hand.", + "D": "You opened two e-books, and they were on the phone in your hand.", + "E": "You opened four e-books, and they were all on the phone you were holding." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + } +] diff --git a/QAs/D2-P3-am-shopping_part003-mcq.json b/QAs/D2-P3-am-shopping_part003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2ce4e32115e17950b7591c89eddba346f696fe53 --- /dev/null +++ b/QAs/D2-P3-am-shopping_part003-mcq.json @@ -0,0 +1,803 @@ +[ + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_13", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get from the place that sells handkerchiefs to the place that sells milk?", + "answer": "Go straight to the left and then turn right.", + "question_chinese": "从卖手帕纸的地方怎么走到卖牛奶的地方？", + "answer_chinese": "先往左边直走再右转。", + "evidence": [ + { + "description": "Go straight to the left.", + "timestamps": { + "start": "00:01:28", + "end": "00:01:30" + } + }, + { + "description": "Turn right.", + "timestamps": { + "start": "00:01:33", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight to the left and then turn right.", + "B": "Go straight right, then turn left.", + "C": "Go straight left, then turn left.", + "D": "Go straight ahead for a bit, then turn right to reach the milk section.", + "E": "Head left briefly, then make a left turn and continue straight to milk." + }, + "answer_label": "A", + "answer_text": "Go straight to the left and then turn right.", + "q_id": "D2-P3-am-shopping_part003_13", + "correct_answer": "Go straight to the left and then turn right.", + "original_options": { + "A": "Go straight to the left and then turn right.", + "B": "Go straight right, then turn left.", + "C": "Go straight left, then turn left.", + "D": "Go straight ahead for a bit, then turn right to reach the milk section.", + "E": "Head left briefly, then make a left turn and continue straight to milk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_14", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get from the place that sells handkerchiefs to the place that sells milk?", + "answer": "Go straight to the left and then turn right.", + "question_chinese": "从卖手帕纸的地方怎么走到卖牛奶的地方？", + "answer_chinese": "先往左边直走再右转。", + "evidence": [ + { + "description": "Go straight to the left.", + "timestamps": { + "start": "00:01:28", + "end": "00:01:30" + } + }, + { + "description": "turn right.", + "timestamps": { + "start": "00:01:33", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight right, then turn left.", + "B": "Go straight ahead, then turn right.", + "C": "Go straight to the left and then turn left to reach the milk area.", + "D": "Go straight left for a short distance, then turn around and head back toward milk.", + "E": "Go straight to the left and then turn right." + }, + "answer_label": "E", + "answer_text": "Go straight to the left and then turn right.", + "q_id": "D2-P3-am-shopping_part003_14", + "correct_answer": "Go straight to the left and then turn right.", + "original_options": { + "A": "Go straight right, then turn left.", + "B": "Go straight ahead, then turn right.", + "C": "Go straight to the left and then turn left to reach the milk area.", + "D": "Go straight left for a short distance, then turn around and head back toward milk.", + "E": "Go straight to the left and then turn right." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_3_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I walk from the place that sells tissues to the place that sells milk?", + "answer": "Walk straight, then turn right.", + "question_chinese": "我从卖手帕纸的地方怎么走到卖牛奶的地方？", + "answer_chinese": "你直走再右转。", + "evidence": [ + { + "description": "you go stright then turn right.", + "timestamps": { + "start": "00:01:33", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:01:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk straight, then turn left.", + "B": "Turn right first, then walk straight.", + "C": "Walk straight for a moment, then turn left to reach the place that sells milk.", + "D": "Walk straight, then turn right.", + "E": "Walk straight, then make two left turns before continuing toward the milk section." + }, + "answer_label": "D", + "answer_text": "Walk straight, then turn right.", + "q_id": "D2-P3-am-shopping_part003_3_T2", + "correct_answer": "Walk straight, then turn right.", + "original_options": { + "A": "Walk straight, then turn left.", + "B": "Turn right first, then walk straight.", + "C": "Walk straight for a moment, then turn left to reach the place that sells milk.", + "D": "Walk straight, then turn right.", + "E": "Walk straight, then make two left turns before continuing toward the milk section." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 13 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_3_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I walk from the place that sells tissues to the place that sells milk?", + "answer": "First walk straight toward your left, then turn right.", + "question_chinese": "我从卖手帕纸的地方怎么走到卖牛奶的地方？", + "answer_chinese": "你先往左边直走再右转。", + "evidence": [ + { + "description": "Go straight to the left.", + "timestamps": { + "start": "00:01:28", + "end": "00:01:30" + } + }, + { + "description": "Turn right.", + "timestamps": { + "start": "00:01:33", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:01:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk straight to your right, then turn right.", + "B": "Walk straight toward your left, then turn left.", + "C": "First walk straight toward your right, then turn left at the next aisle to reach the milk.", + "D": "First walk straight ahead without veering left, then turn right to get from tissues to milk.", + "E": "First walk straight toward your left, then turn right." + }, + "answer_label": "E", + "answer_text": "First walk straight toward your left, then turn right.", + "q_id": "D2-P3-am-shopping_part003_3_T1", + "correct_answer": "First walk straight toward your left, then turn right.", + "original_options": { + "A": "Walk straight to your right, then turn right.", + "B": "Walk straight toward your left, then turn left.", + "C": "First walk straight toward your right, then turn left at the next aisle to reach the milk.", + "D": "First walk straight ahead without veering left, then turn right to get from tissues to milk.", + "E": "First walk straight toward your left, then turn right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 16 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_1_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I grab the blueberries directly?", + "answer": "You can't directly reach the blueberries; the blueberries are to your front-left.", + "question_chinese": "我可以直接拿到蓝莓吗？", + "answer_chinese": "你不可以直接拿到蓝莓，蓝莓在你左前方。", + "evidence": [ + { + "description": "You can not get the blueberries directly.", + "timestamps": { + "start": "00:05:43", + "end": "00:05:44" + } + } + ], + "question_timestamps": "00:05:47", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can directly reach the blueberries; the blueberries are to your front-left.", + "B": "You can't directly reach the blueberries; the blueberries are to your front-left." + }, + "answer_label": "B", + "answer_text": "You can't directly reach the blueberries; the blueberries are to your front-left.", + "q_id": "D2-P3-am-shopping_part003_1_T1", + "correct_answer": "You can't directly reach the blueberries; the blueberries are to your front-left.", + "original_options": { + "A": "You can directly reach the blueberries; the blueberries are to your front-left.", + "B": "You can't directly reach the blueberries; the blueberries are to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_1_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I grab the blueberries directly?", + "answer": "You can directly reach the blueberries; the blueberries are to your left.", + "question_chinese": "我可以直接拿到蓝莓吗？", + "answer_chinese": "你可以直接拿到蓝莓，蓝莓在你左边。", + "evidence": [ + { + "description": "You can get the blueberries directly.", + "timestamps": { + "start": "00:05:51", + "end": "00:05:53" + } + } + ], + "question_timestamps": "00:05:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can directly reach the blueberries; the blueberries are to your left.", + "B": "You can't directly reach the blueberries; the blueberries are to your left." + }, + "answer_label": "A", + "answer_text": "You can directly reach the blueberries; the blueberries are to your left.", + "q_id": "D2-P3-am-shopping_part003_1_T2", + "correct_answer": "You can directly reach the blueberries; the blueberries are to your left.", + "original_options": { + "A": "You can directly reach the blueberries; the blueberries are to your left.", + "B": "You can't directly reach the blueberries; the blueberries are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_4_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many bottles of milk do I have in my hand?", + "answer": "You have 1 bottle of milk in your hand, and the milk is in front of you.", + "question_chinese": "我手里有多少瓶牛奶？", + "answer_chinese": "你手里有1瓶牛奶，牛奶在你前面。", + "evidence": [ + { + "description": "Your hand a milk.", + "timestamps": { + "start": "00:02:03", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:02:08", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You have 1 bottle of milk in your hand, and the milk is in front of you.", + "B": "You have 2 bottles of milk in your hand, and the milk is in front of you." + }, + "answer_label": "A", + "answer_text": "You have 1 bottle of milk in your hand, and the milk is in front of you.", + "q_id": "D2-P3-am-shopping_part003_4_T1", + "correct_answer": "You have 1 bottle of milk in your hand, and the milk is in front of you.", + "original_options": { + "A": "You have 1 bottle of milk in your hand, and the milk is in front of you.", + "B": "You have 2 bottles of milk in your hand, and the milk is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_4_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many bottles of milk do I have in my hand?", + "answer": "You have 2 bottles of milk in your hand, and the milk is in front of you.", + "question_chinese": "我手里有多少瓶牛奶？", + "answer_chinese": "你手里有2瓶牛奶，牛奶在你前面。", + "evidence": [ + { + "description": "Your hand two milk.", + "timestamps": { + "start": "00:02:49", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:03:01", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You have 1 bottle of milk in your hand, and the milk is in front of you.", + "B": "You have 2 bottles of milk in your hand, and the milk is in front of you." + }, + "answer_label": "B", + "answer_text": "You have 2 bottles of milk in your hand, and the milk is in front of you.", + "q_id": "D2-P3-am-shopping_part003_4_T2", + "correct_answer": "You have 2 bottles of milk in your hand, and the milk is in front of you.", + "original_options": { + "A": "You have 1 bottle of milk in your hand, and the milk is in front of you.", + "B": "You have 2 bottles of milk in your hand, and the milk is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_16", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which of these two bottles of milk in my hand is more expensive?", + "answer": "The milk in the right hand is more expensive.", + "question_chinese": "我手里这两瓶牛奶哪个更贵？", + "answer_chinese": "右边手里的牛奶更贵。", + "evidence": [ + { + "description": "The milk on the right is 10.9 yuan.", + "timestamps": { + "start": "00:02:03", + "end": "00:02:05" + } + }, + { + "description": "The milk on the left is 9.9 yuan.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:13" + } + }, + { + "description": "The milk in the right hand is more expensive.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:02:26", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The milk in the left hand is more expensive.", + "B": "The milk in the right hand is more expensive." + }, + "answer_label": "B", + "answer_text": "The milk in the right hand is more expensive.", + "q_id": "D2-P3-am-shopping_part003_16", + "correct_answer": "The milk in the right hand is more expensive.", + "original_options": { + "A": "The milk in the left hand is more expensive.", + "B": "The milk in the right hand is more expensive." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_15", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which of these two bottles of milk in my hand is more expensive?", + "answer": "The milk in the right hand is more expensive.", + "question_chinese": "我手里这两瓶牛奶哪个更贵？", + "answer_chinese": "右边手里的牛奶更贵。", + "evidence": [ + { + "description": "The milk on the right is 9.9 yuan.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:13" + } + }, + { + "description": "The milk on the left is 7.9 yuan.", + "timestamps": { + "start": "00:02:45", + "end": "00:02:48" + } + }, + { + "description": "The milk in the right hand is more expensive.", + "timestamps": { + "start": "00:02:49", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:03:01", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The milk in the right hand is more expensive.", + "B": "The milk in the left hand is more expensive." + }, + "answer_label": "A", + "answer_text": "The milk in the right hand is more expensive.", + "q_id": "D2-P3-am-shopping_part003_15", + "correct_answer": "The milk in the right hand is more expensive.", + "original_options": { + "A": "The milk in the right hand is more expensive.", + "B": "The milk in the left hand is more expensive." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_11", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the tissue paper relative to the coffee?", + "answer": "The handkerchief paper is on the opposite side of the coffee.", + "question_chinese": "手帕纸在咖啡的哪个方向？", + "answer_chinese": "手帕纸在咖啡的对面。", + "evidence": [ + { + "description": "The handkerchief paper is on the opposite side of the coffee.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:09" + } + }, + { + "description": "The handkerchief paper is on the opposite side of the coffee.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:05" + } + }, + { + "description": "The handkerchief paper is on the opposite side of the coffee.", + "timestamps": { + "start": "00:01:11", + "end": "00:01:13" + } + } + ], + "question_timestamps": "00:02:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The handkerchief paper is behind the coffee.", + "B": "The handkerchief paper is in front.", + "C": "The handkerchief paper is positioned to the left of the coffee, not across from it on the shelf.", + "D": "The handkerchief paper sits to the right of the coffee rather than being on the opposite side.", + "E": "The handkerchief paper is on the opposite side of the coffee." + }, + "answer_label": "E", + "answer_text": "The handkerchief paper is on the opposite side of the coffee.", + "q_id": "D2-P3-am-shopping_part003_11", + "correct_answer": "The handkerchief paper is on the opposite side of the coffee.", + "original_options": { + "A": "The handkerchief paper is behind the coffee.", + "B": "The handkerchief paper is in front.", + "C": "The handkerchief paper is positioned to the left of the coffee, not across from it on the shelf.", + "D": "The handkerchief paper sits to the right of the coffee rather than being on the opposite side.", + "E": "The handkerchief paper is on the opposite side of the coffee." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_12", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the tissue paper relative to the coffee?", + "answer": "The handkerchief paper is on the opposite side of the coffee.", + "question_chinese": "手帕纸在咖啡的哪个方向？", + "answer_chinese": "手帕纸在咖啡的对面。", + "evidence": [ + { + "description": "The handkerchief paper is on the opposite side of the coffee.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:09" + } + }, + { + "description": "The handkerchief paper is on the opposite side of the coffee.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:05" + } + }, + { + "description": "The handkerchief paper is on the opposite side of the coffee.", + "timestamps": { + "start": "00:01:11", + "end": "00:01:13" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To the left of the coffee.", + "B": "In front of the coffee.", + "C": "The handkerchief paper is on the opposite side of the coffee.", + "D": "The handkerchief paper is positioned just behind the coffee, closer to the back edge of the surface.", + "E": "The handkerchief paper sits to the right of the coffee, slightly offset rather than directly aligned with it." + }, + "answer_label": "C", + "answer_text": "The handkerchief paper is on the opposite side of the coffee.", + "q_id": "D2-P3-am-shopping_part003_12", + "correct_answer": "The handkerchief paper is on the opposite side of the coffee.", + "original_options": { + "A": "To the left of the coffee.", + "B": "In front of the coffee.", + "C": "The handkerchief paper is on the opposite side of the coffee.", + "D": "The handkerchief paper is positioned just behind the coffee, closer to the back edge of the surface.", + "E": "The handkerchief paper sits to the right of the coffee, slightly offset rather than directly aligned with it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 18 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_2_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me are the tissues?", + "answer": "The pocket tissues are behind you.", + "question_chinese": "手帕纸在我的哪个方向？", + "answer_chinese": "手帕纸在你的后面。", + "evidence": [ + { + "description": "The handkerchief paper is on the opposite side of you.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The pocket tissues are to your right, roughly at your side rather than behind you.", + "D": "The pocket tissues are above you, positioned higher up instead of being located behind you.", + "E": "The pocket tissues are behind you." + }, + "answer_label": "E", + "answer_text": "The pocket tissues are behind you.", + "q_id": "D2-P3-am-shopping_part003_2_T1", + "correct_answer": "The pocket tissues are behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The pocket tissues are to your right, roughly at your side rather than behind you.", + "D": "The pocket tissues are above you, positioned higher up instead of being located behind you.", + "E": "The pocket tissues are behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 15 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part003", + "video_length": "00:20:04", + "qid": "D2-P3-am-shopping_part003_2_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me are the tissues?", + "answer": "The pocket tissues are in front of you.", + "question_chinese": "手帕纸在我的哪个方向？", + "answer_chinese": "手帕纸在你的前面。", + "evidence": [ + { + "description": "The handkerchief paper is on front of you.", + "timestamps": { + "start": "00:01:11", + "end": "00:01:13" + } + } + ], + "question_timestamps": "00:01:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your right.", + "C": "The pocket tissues are to your left, sitting off to the side instead of in front of you.", + "D": "The pocket tissues are above you, located higher up rather than directly in front of you.", + "E": "The pocket tissues are in front of you." + }, + "answer_label": "E", + "answer_text": "The pocket tissues are in front of you.", + "q_id": "D2-P3-am-shopping_part003_2_T2", + "correct_answer": "The pocket tissues are in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your right.", + "C": "The pocket tissues are to your left, sitting off to the side instead of in front of you.", + "D": "The pocket tissues are above you, located higher up rather than directly in front of you.", + "E": "The pocket tissues are in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 18, + 16 + ] + } +] diff --git a/QAs/D2-P3-am-shopping_part004-mcq.json b/QAs/D2-P3-am-shopping_part004-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..bd9196a658aef355444224ade3cbeb1777093b9a --- /dev/null +++ b/QAs/D2-P3-am-shopping_part004-mcq.json @@ -0,0 +1,478 @@ +[ + { + "video_uid": "D2-P3-am-shopping_part004", + "video_length": "00:20:02", + "qid": "D2-P3-am-shopping_part004_1_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I looking at on my phone?", + "answer": "You were reading a novel, and the phone was in your hand.", + "question_chinese": "我手机上在看什么？", + "answer_chinese": "你在看小说，手机在你的手上。", + "evidence": [ + { + "description": "You read novels on your mobile phone.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:05:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were checking the time.", + "B": "You were scrolling a menu.", + "C": "You were reading a novel, and the phone was in your hand.", + "D": "You were memorizing English vocabulary on your phone while holding it in your hand.", + "E": "You were reviewing a list of English words on the phone in your hand instead of reading a novel." + }, + "answer_label": "C", + "answer_text": "You were reading a novel, and the phone was in your hand.", + "q_id": "D2-P3-am-shopping_part004_1_T2", + "correct_answer": "You were reading a novel, and the phone was in your hand.", + "original_options": { + "A": "You were checking the time.", + "B": "You were scrolling a menu.", + "C": "You were reading a novel, and the phone was in your hand.", + "D": "You were memorizing English vocabulary on your phone while holding it in your hand.", + "E": "You were reviewing a list of English words on the phone in your hand instead of reading a novel." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 19 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part004", + "video_length": "00:20:02", + "qid": "D2-P3-am-shopping_part004_1_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I looking at on my phone?", + "answer": "You were memorizing English vocabulary, and the phone was in your hand.", + "question_chinese": "我手机上在看什么？", + "answer_chinese": "你在背英语单词 ，手机在你的手上。", + "evidence": [ + { + "description": "You are memorizing English words.", + "timestamps": { + "start": "00:08:27", + "end": "00:08:28" + } + } + ], + "question_timestamps": "00:09:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were reading a novel.", + "B": "You were memorizing English vocabulary, and the phone was in your hand.", + "C": "You were checking the time.", + "D": "You were reading a book on your phone, but it wasn’t for memorizing English vocabulary.", + "E": "You were watching a short video on your phone instead of studying English vocabulary." + }, + "answer_label": "B", + "answer_text": "You were memorizing English vocabulary, and the phone was in your hand.", + "q_id": "D2-P3-am-shopping_part004_1_T1", + "correct_answer": "You were memorizing English vocabulary, and the phone was in your hand.", + "original_options": { + "A": "You were reading a novel.", + "B": "You were memorizing English vocabulary, and the phone was in your hand.", + "C": "You were checking the time.", + "D": "You were reading a book on your phone, but it wasn’t for memorizing English vocabulary.", + "E": "You were watching a short video on your phone instead of studying English vocabulary." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 14 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part004", + "video_length": "00:20:02", + "qid": "D2-P3-am-shopping_part004_2_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many things did I do with my phone?", + "answer": "You did one thing: you used the phone in your hand to read a novel.", + "question_chinese": "我拿手机做了几件事情", + "answer_chinese": "做了一件事情，你在用你手上的手机上看小说。", + "evidence": [ + { + "description": "You read novels on your mobile phone.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:53" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You did zero things.", + "B": "You did two things.", + "C": "You did one thing: you used the phone in your hand to read a novel.", + "D": "You did three things: you read a novel, checked the time, and then opened a vocabulary list on your phone.", + "E": "You did four things: you read a novel, memorized words, sent a message, and then checked the time." + }, + "answer_label": "C", + "answer_text": "You did one thing: you used the phone in your hand to read a novel.", + "q_id": "D2-P3-am-shopping_part004_2_T1", + "correct_answer": "You did one thing: you used the phone in your hand to read a novel.", + "original_options": { + "A": "You did zero things.", + "B": "You did two things.", + "C": "You did one thing: you used the phone in your hand to read a novel.", + "D": "You did three things: you read a novel, checked the time, and then opened a vocabulary list on your phone.", + "E": "You did four things: you read a novel, memorized words, sent a message, and then checked the time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 18 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part004", + "video_length": "00:20:02", + "qid": "D2-P3-am-shopping_part004_2_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many things did I do with my phone?", + "answer": "You did two things: you used the phone in your hand to read a novel and memorize vocabulary.", + "question_chinese": "我拿手机做了几件事情", + "answer_chinese": "做了两件事情，你在用你手上的手机看小说和背单词", + "evidence": [ + { + "description": "You memorize vocabulary using \"Bu Bei Dan Ci.\"", + "timestamps": { + "start": "00:06:50", + "end": "00:06:53" + } + }, + { + "description": "You read novels on your mobile phone.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:14:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You did one thing.", + "B": "You did zero things.", + "C": "You did three things: you read a novel, memorized vocabulary, and then checked the time on your phone.", + "D": "You did two things: you used the phone in your hand to read a novel and memorize vocabulary.", + "E": "You did four things: you read a novel, memorized vocabulary, browsed photos, and then checked the time." + }, + "answer_label": "D", + "answer_text": "You did two things: you used the phone in your hand to read a novel and memorize vocabulary.", + "q_id": "D2-P3-am-shopping_part004_2_T2", + "correct_answer": "You did two things: you used the phone in your hand to read a novel and memorize vocabulary.", + "original_options": { + "A": "You did one thing.", + "B": "You did zero things.", + "C": "You did three things: you read a novel, memorized vocabulary, and then checked the time on your phone.", + "D": "You did two things: you used the phone in your hand to read a novel and memorize vocabulary.", + "E": "You did four things: you read a novel, memorized vocabulary, browsed photos, and then checked the time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 17 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part004", + "video_length": "00:20:02", + "qid": "D2-P3-am-shopping_part004_15", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many vague words do I have?", + "answer": "Two.", + "question_chinese": "我有几个模糊的单词？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "You have a vague word.", + "timestamps": { + "start": "00:07:10", + "end": "00:07:13" + } + }, + { + "description": "You have a vague word.", + "timestamps": { + "start": "00:07:22", + "end": "00:07:23" + } + } + ], + "question_timestamps": "00:07:35", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four, because you counted two extra vague words beyond the two that were actually written down.", + "E": "Five, since you included additional vague words from earlier speech rather than only the ones you have." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "D2-P3-am-shopping_part004_15", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four, because you counted two extra vague words beyond the two that were actually written down.", + "E": "Five, since you included additional vague words from earlier speech rather than only the ones you have." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 17 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part004", + "video_length": "00:20:02", + "qid": "D2-P3-am-shopping_part004_16", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many vague words do I have?", + "answer": "Three.", + "question_chinese": "我有几个模糊的单词？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "You have a vague word.", + "timestamps": { + "start": "00:07:10", + "end": "00:07:13" + } + }, + { + "description": "You have a vague word.", + "timestamps": { + "start": "00:07:22", + "end": "00:07:23" + } + }, + { + "description": "You have a vague word.", + "timestamps": { + "start": "00:07:29", + "end": "00:07:31" + } + } + ], + "question_timestamps": "00:07:42", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "You have two vague words written down in your notes right now.", + "D": "Three.", + "E": "You have four vague words recorded so far in total on the page." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "D2-P3-am-shopping_part004_16", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "You have two vague words written down in your notes right now.", + "D": "Three.", + "E": "You have four vague words recorded so far in total on the page." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part004", + "video_length": "00:20:02", + "qid": "D2-P3-am-shopping_part004_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What software should I use to memorize vocabulary?", + "answer": "不背单词。.", + "question_chinese": "我用什么软件背单词？", + "answer_chinese": "不背单词。", + "evidence": [ + { + "description": "You memorize vocabulary using \"Bu Bei Dan Ci.\"", + "timestamps": { + "start": "00:06:50", + "end": "00:06:53" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Anki.", + "B": "Quizlet.", + "C": "Use Anki with spaced repetition to memorize and review vocabulary every single day.", + "D": "Use Quizlet flashcards and short practice tests to drill vocabulary regularly each week.", + "E": "不背单词。." + }, + "answer_label": "E", + "answer_text": "不背单词。.", + "q_id": "D2-P3-am-shopping_part004_11", + "correct_answer": "不背单词。.", + "original_options": { + "A": "Anki.", + "B": "Quizlet.", + "C": "Use Anki with spaced repetition to memorize and review vocabulary every single day.", + "D": "Use Quizlet flashcards and short practice tests to drill vocabulary regularly each week.", + "E": "不背单词。." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part004", + "video_length": "00:20:02", + "qid": "D2-P3-am-shopping_part004_12", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What software should I use to memorize vocabulary?", + "answer": "不背单词。.", + "question_chinese": "我用什么软件背单词？", + "answer_chinese": "不背单词。", + "evidence": [ + { + "description": "You memorize vocabulary using \"Bu Bei Dan Ci.\"", + "timestamps": { + "start": "00:06:50", + "end": "00:06:53" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Anki.", + "B": "不背单词。.", + "C": "Quizlet.", + "D": "Use Anki with spaced repetition to memorize and review vocabulary every single day.", + "E": "Use Quizlet flashcards and short practice tests to drill vocabulary regularly each week." + }, + "answer_label": "B", + "answer_text": "不背单词。.", + "q_id": "D2-P3-am-shopping_part004_12", + "correct_answer": "不背单词。.", + "original_options": { + "A": "Anki.", + "B": "不背单词。.", + "C": "Quizlet.", + "D": "Use Anki with spaced repetition to memorize and review vocabulary every single day.", + "E": "Use Quizlet flashcards and short practice tests to drill vocabulary regularly each week." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + } +] diff --git a/QAs/D2-P3-am-shopping_part005-mcq.json b/QAs/D2-P3-am-shopping_part005-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..cb61e10d401b1ddc0fbe74879737bba223b3bb0f --- /dev/null +++ b/QAs/D2-P3-am-shopping_part005-mcq.json @@ -0,0 +1,505 @@ +[ + { + "video_uid": "D2-P3-am-shopping_part005", + "video_length": "00:13:40", + "qid": "D2-P3-am-shopping_part005_14", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement pattern after leaving the coffee shop?", + "answer": "After you come out, go straight to the right.", + "question_chinese": "从咖啡店出来后我的移动轨迹是怎样的？", + "answer_chinese": "出来后你往右边直行。", + "evidence": [ + { + "description": "After you come out, go straight to the right.", + "timestamps": { + "start": "00:09:55", + "end": "00:10:01" + } + } + ], + "question_timestamps": "00:10:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "After you come out, go straight left.", + "B": "After you come out, turn around.", + "C": "After you come out, walk straight ahead for a bit, then turn left at the next opening.", + "D": "After you come out, go straight to the right.", + "E": "After you come out, walk straight ahead and then turn left after you pass the storefront." + }, + "answer_label": "D", + "answer_text": "After you come out, go straight to the right.", + "q_id": "D2-P3-am-shopping_part005_14", + "correct_answer": "After you come out, go straight to the right.", + "original_options": { + "A": "After you come out, go straight left.", + "B": "After you come out, turn around.", + "C": "After you come out, walk straight ahead for a bit, then turn left at the next opening.", + "D": "After you come out, go straight to the right.", + "E": "After you come out, walk straight ahead and then turn left after you pass the storefront." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 16 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part005", + "video_length": "00:13:40", + "qid": "D2-P3-am-shopping_part005_13", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement pattern after leaving the coffee shop?", + "answer": "After you come out, go straight to the right and then turn left.", + "question_chinese": "从咖啡店出来后我的移动轨迹是怎样的？", + "answer_chinese": "出来后你先往右边直行，然后左转。", + "evidence": [ + { + "description": "After you come out, go straight to the right.", + "timestamps": { + "start": "00:09:55", + "end": "00:10:00" + } + }, + { + "description": "Then turn left.", + "timestamps": { + "start": "00:10:08", + "end": "00:10:10" + } + } + ], + "question_timestamps": "00:10:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go left, then right.", + "B": "Go straight, then left.", + "C": "After you come out, go straight to the right and then turn left.", + "D": "After you come out, go straight to the right and then turn right.", + "E": "After you come out, go straight to the left first, then turn right at the next corner." + }, + "answer_label": "C", + "answer_text": "After you come out, go straight to the right and then turn left.", + "q_id": "D2-P3-am-shopping_part005_13", + "correct_answer": "After you come out, go straight to the right and then turn left.", + "original_options": { + "A": "Go left, then right.", + "B": "Go straight, then left.", + "C": "After you come out, go straight to the right and then turn left.", + "D": "After you come out, go straight to the right and then turn right.", + "E": "After you come out, go straight to the left first, then turn right at the next corner." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 17 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part005", + "video_length": "00:13:40", + "qid": "D2-P3-am-shopping_part005_11", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How do I get to the trash can from where I'm sitting?", + "answer": "Go straight first, then turn right.", + "question_chinese": "从我刚刚坐的位置怎么去垃圾桶？", + "answer_chinese": "先直行，然后右转。", + "evidence": [ + { + "description": "Go straight first.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:32" + } + }, + { + "description": "Then turn right.", + "timestamps": { + "start": "00:09:32", + "end": "00:09:33" + } + } + ], + "question_timestamps": "00:09:38", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right first.", + "B": "Go straight, then left.", + "C": "Go straight first, then turn right.", + "D": "Go left first, then turn right after a moment, not straight ahead.", + "E": "Go straight first, then turn back instead of turning right at the corner." + }, + "answer_label": "C", + "answer_text": "Go straight first, then turn right.", + "q_id": "D2-P3-am-shopping_part005_11", + "correct_answer": "Go straight first, then turn right.", + "original_options": { + "A": "Turn right first.", + "B": "Go straight, then left.", + "C": "Go straight first, then turn right.", + "D": "Go left first, then turn right after a moment, not straight ahead.", + "E": "Go straight first, then turn back instead of turning right at the corner." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 13 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part005", + "video_length": "00:13:40", + "qid": "D2-P3-am-shopping_part005_12", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How do I get to the trash can from where I'm sitting?", + "answer": "Go straight first, then turn right.", + "question_chinese": "从我刚刚坐的位置怎么去垃圾桶？", + "answer_chinese": "先直行，然后右转。", + "evidence": [ + { + "description": "Go straight first", + "timestamps": { + "start": "00:09:30", + "end": "00:09:32" + } + }, + { + "description": "Then turn right.", + "timestamps": { + "start": "00:09:32", + "end": "00:09:33" + } + } + ], + "question_timestamps": "00:10:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right first.", + "B": "Go straight first, then turn right.", + "C": "Go straight, then left.", + "D": "Go left first, then turn right after a moment, not straight ahead.", + "E": "Go straight first, then turn back instead of turning right at the corner." + }, + "answer_label": "B", + "answer_text": "Go straight first, then turn right.", + "q_id": "D2-P3-am-shopping_part005_12", + "correct_answer": "Go straight first, then turn right.", + "original_options": { + "A": "Turn right first.", + "B": "Go straight first, then turn right.", + "C": "Go straight, then left.", + "D": "Go left first, then turn right after a moment, not straight ahead.", + "E": "Go straight first, then turn back instead of turning right at the corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 13 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part005", + "video_length": "00:13:40", + "qid": "D2-P3-am-shopping_part005_3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many traffic lights did I wait at today at the crosswalks?", + "answer": "One.", + "question_chinese": "我今天在几个斑马线处等了红绿灯？", + "answer_chinese": "一个。", + "evidence": [ + { + "description": "You are waiting for the traffic lights.", + "timestamps": { + "start": "00:11:09", + "end": "00:11:11" + } + } + ], + "question_timestamps": "00:11:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "You waited at four traffic lights at crosswalks during your trip today before crossing.", + "E": "You waited at five traffic lights today, stopping at multiple crosswalk signals before crossing." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "D2-P3-am-shopping_part005_3", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "You waited at four traffic lights at crosswalks during your trip today before crossing.", + "E": "You waited at five traffic lights today, stopping at multiple crosswalk signals before crossing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part005", + "video_length": "00:13:40", + "qid": "D2-P3-am-shopping_part005_4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many traffic lights did I wait at today at the crosswalks?", + "answer": "Six.", + "question_chinese": "我今天在几个斑马线处等了红绿灯？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "You are waiting for the traffic lights.", + "timestamps": { + "start": "00:11:09", + "end": "00:11:11" + } + }, + { + "description": "You are waiting for the traffic lights.", + "timestamps": { + "start": "00:12:47", + "end": "00:12:49" + } + } + ], + "question_timestamps": "00:13:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "Four.", + "C": "Eight.", + "D": "You waited at five traffic lights at the crosswalks today, not six like you said.", + "E": "You ended up waiting at seven traffic lights at the crosswalks today during your walk." + }, + "answer_label": "A", + "answer_text": "Six.", + "q_id": "D2-P3-am-shopping_part005_4", + "correct_answer": "Six.", + "original_options": { + "A": "Six.", + "B": "Four.", + "C": "Eight.", + "D": "You waited at five traffic lights at the crosswalks today, not six like you said.", + "E": "You ended up waiting at seven traffic lights at the crosswalks today during your walk." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part005", + "video_length": "00:13:40", + "qid": "D2-P3-am-shopping_part005_15", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just hop on a blue shared bike?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接骑上蓝色的共享单车吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The blue shared bike is right next to you, you can ride it directly.", + "timestamps": { + "start": "00:10:49", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:10:53", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "D2-P3-am-shopping_part005_15", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P3-am-shopping_part005", + "video_length": "00:13:40", + "qid": "D2-P3-am-shopping_part005_16", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just hop on a blue shared bike?", + "answer": "No,you can't.", + "question_chinese": "我可以直接骑上蓝色的共享单车吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The blue shared bike is in front of you.", + "timestamps": { + "start": "00:10:52", + "end": "00:10:53" + } + }, + { + "description": "The blue shared bike is behind you.", + "timestamps": { + "start": "00:10:52", + "end": "00:10:53" + } + }, + { + "description": "They are too far away from you.", + "timestamps": { + "start": "00:10:52", + "end": "00:10:53" + } + }, + { + "description": "You cannot ride directly.", + "timestamps": { + "start": "00:10:52", + "end": "00:10:53" + } + } + ], + "question_timestamps": "00:10:56", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "D2-P3-am-shopping_part005_16", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + } +] diff --git a/QAs/D2-P4-pm-1-fitness_video_part001-mcq.json b/QAs/D2-P4-pm-1-fitness_video_part001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6ad3b952dbddb2fcbf57b5f0301e2f3eb7723158 --- /dev/null +++ b/QAs/D2-P4-pm-1-fitness_video_part001-mcq.json @@ -0,0 +1,247 @@ +[ + { + "video_uid": "D2-P4-pm-1-fitness_video_part001", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part001_12", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many types of gym equipment did I use at the gym today?", + "answer": "One kind.", + "question_chinese": "今天在健身房我用了几种健身器材？", + "answer_chinese": "一种。", + "evidence": [ + { + "description": "You used the first type.", + "timestamps": { + "start": "00:09:40", + "end": "00:10:00" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One kind.", + "B": "Two kinds.", + "C": "Three kinds.", + "D": "You used four different kinds of gym equipment during your workout at the gym today.", + "E": "You used five types of equipment at the gym during your workout session today." + }, + "answer_label": "A", + "answer_text": "One kind.", + "q_id": "D2-P4-pm-1-fitness_video_part001_12", + "correct_answer": "One kind.", + "original_options": { + "A": "One kind.", + "B": "Two kinds.", + "C": "Three kinds.", + "D": "You used four different kinds of gym equipment during your workout at the gym today.", + "E": "You used five types of equipment at the gym during your workout session today." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part001", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part001_11", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many types of gym equipment did I use at the gym today?", + "answer": "Two kinds.", + "question_chinese": "今天在健身房我用了几种健身器材？", + "answer_chinese": "两种。", + "evidence": [ + { + "description": "You used the first one.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:20" + } + }, + { + "description": "You used the second type.", + "timestamps": { + "start": "00:10:50", + "end": "00:11:10" + } + } + ], + "question_timestamps": "00:12:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One kind.", + "B": "Three kinds.", + "C": "You used four different kinds of gym equipment during your workout at the gym today.", + "D": "Two kinds.", + "E": "You ended up using five types of equipment during your gym session today." + }, + "answer_label": "D", + "answer_text": "Two kinds.", + "q_id": "D2-P4-pm-1-fitness_video_part001_11", + "correct_answer": "Two kinds.", + "original_options": { + "A": "One kind.", + "B": "Three kinds.", + "C": "You used four different kinds of gym equipment during your workout at the gym today.", + "D": "Two kinds.", + "E": "You ended up using five types of equipment during your gym session today." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 13 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part001", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part001_1_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What floor am I on right now?", + "answer": "On the 18th floor, the elevator is in front of you.", + "question_chinese": "我现在在几楼？", + "answer_chinese": "18，电梯在我的前方。", + "evidence": [ + { + "description": "The elevator shows that you are heading to the 18th floor.", + "timestamps": { + "start": "00:02:04", + "end": "00:02:07" + } + }, + { + "description": "The house number shows that it is currently on the 18th floor", + "timestamps": { + "start": "00:02:47", + "end": "00:02:49" + } + } + ], + "question_timestamps": "00:03:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "17th floor, elevator ahead.", + "B": "On the 18th floor, the elevator is in front of you.", + "C": "19th floor, elevator ahead.", + "D": "You're on the 18th floor, but the elevator is behind you instead of in front.", + "E": "You're on the 18th floor, and the elevator is to your left rather than directly ahead." + }, + "answer_label": "B", + "answer_text": "On the 18th floor, the elevator is in front of you.", + "q_id": "D2-P4-pm-1-fitness_video_part001_1_T2", + "correct_answer": "On the 18th floor, the elevator is in front of you.", + "original_options": { + "A": "17th floor, elevator ahead.", + "B": "On the 18th floor, the elevator is in front of you.", + "C": "19th floor, elevator ahead.", + "D": "You're on the 18th floor, but the elevator is behind you instead of in front.", + "E": "You're on the 18th floor, and the elevator is to your left rather than directly ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part001", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part001_1_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What floor am I on right now?", + "answer": "On the 18th floor, the elevator is behind you.", + "question_chinese": "我现在在几楼？", + "answer_chinese": "18楼，电梯在我的身后。", + "evidence": [ + { + "description": "The house number shows that it is currently on the 18th floor.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:51" + } + }, + { + "description": "The elevator shows that you are heading to the 18th floor.", + "timestamps": { + "start": "00:02:04", + "end": "00:02:07" + } + } + ], + "question_timestamps": "00:04:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the 18th floor, the elevator is behind you.", + "B": "18th floor, elevator ahead.", + "C": "19th floor, elevator behind.", + "D": "You're on the 17th floor, and the elevator is behind you right now.", + "E": "You're on the 18th floor, but the elevator is to your right instead of behind you." + }, + "answer_label": "A", + "answer_text": "On the 18th floor, the elevator is behind you.", + "q_id": "D2-P4-pm-1-fitness_video_part001_1_T1", + "correct_answer": "On the 18th floor, the elevator is behind you.", + "original_options": { + "A": "On the 18th floor, the elevator is behind you.", + "B": "18th floor, elevator ahead.", + "C": "19th floor, elevator behind.", + "D": "You're on the 17th floor, and the elevator is behind you right now.", + "E": "You're on the 18th floor, but the elevator is to your right instead of behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 16 + ] + } +] diff --git a/QAs/D2-P4-pm-1-fitness_video_part002-mcq.json b/QAs/D2-P4-pm-1-fitness_video_part002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..418b0978104e8e3a0b0a2211b39021a97d204e81 --- /dev/null +++ b/QAs/D2-P4-pm-1-fitness_video_part002-mcq.json @@ -0,0 +1,382 @@ +[ + { + "video_uid": "D2-P4-pm-1-fitness_video_part002", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part002_11", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the bench press machine or the dumbbells?", + "answer": "The bench press is closer to you.", + "question_chinese": "卧推机和哑铃哪个离我更近？", + "answer_chinese": "卧推机离你更近。", + "evidence": [ + { + "description": "The bench press is right next to you.", + "timestamps": { + "start": "00:10:32", + "end": "00:10:33" + } + }, + { + "description": "Dumbbells are in front of you.", + "timestamps": { + "start": "00:10:32", + "end": "00:10:33" + } + }, + { + "description": "The bench press is closer to you.", + "timestamps": { + "start": "00:10:32", + "end": "00:10:33" + } + } + ], + "question_timestamps": "00:10:43", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The bench press is closer to you.", + "B": "The dumbbells are closer to you." + }, + "answer_label": "A", + "answer_text": "The bench press is closer to you.", + "q_id": "D2-P4-pm-1-fitness_video_part002_11", + "correct_answer": "The bench press is closer to you.", + "original_options": { + "A": "The bench press is closer to you.", + "B": "The dumbbells are closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part002", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part002_12", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the bench press machine or the dumbbells?", + "answer": "The dumbbell is right next to you.", + "question_chinese": "卧推机和哑铃哪个离我更近？", + "answer_chinese": "哑铃离你更近。", + "evidence": [ + { + "description": "The dumbbell is right next to you.", + "timestamps": { + "start": "00:10:41", + "end": "00:10:42" + } + }, + { + "description": "The bench press is on your right。", + "timestamps": { + "start": "00:10:41", + "end": "00:10:42" + } + }, + { + "description": "Dumbbells are closer to you.", + "timestamps": { + "start": "00:10:41", + "end": "00:10:42" + } + } + ], + "question_timestamps": "00:10:52", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The bench press machine is right next to you.", + "B": "The dumbbell is right next to you." + }, + "answer_label": "B", + "answer_text": "The dumbbell is right next to you.", + "q_id": "D2-P4-pm-1-fitness_video_part002_12", + "correct_answer": "The dumbbell is right next to you.", + "original_options": { + "A": "The bench press machine is right next to you.", + "B": "The dumbbell is right next to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part002", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part002_2_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many types of gym equipment did I teach the person in white to use?", + "answer": "Two types: one is to your front-right, and the other is at your feet.", + "question_chinese": "我教穿白色衣服的人几种健身器材的使用？", + "answer_chinese": "两种，一种在你右前侧，另一种在你脚下", + "evidence": [ + { + "description": "This is the first type.", + "timestamps": { + "start": "00:02:50", + "end": "00:03:00" + } + }, + { + "description": "This is the second type.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:33" + } + } + ], + "question_timestamps": "00:04:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One type: to your front-right.", + "B": "Three types: front-right, feet, front-left.", + "C": "Two types: one is to your front-right, and the other is at your feet.", + "D": "Four types: one is to your front-right, another at your feet, one to your front-left, and one behind you.", + "E": "Zero types: you didn’t teach them to use any equipment during this session." + }, + "answer_label": "C", + "answer_text": "Two types: one is to your front-right, and the other is at your feet.", + "q_id": "D2-P4-pm-1-fitness_video_part002_2_T1", + "correct_answer": "Two types: one is to your front-right, and the other is at your feet.", + "original_options": { + "A": "One type: to your front-right.", + "B": "Three types: front-right, feet, front-left.", + "C": "Two types: one is to your front-right, and the other is at your feet.", + "D": "Four types: one is to your front-right, another at your feet, one to your front-left, and one behind you.", + "E": "Zero types: you didn’t teach them to use any equipment during this session." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 14 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part002", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part002_3_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many types of gym equipment did I teach the person in white to use?", + "answer": "Three types: one is to your front-right, another is at your feet, and one more is to your front-left.", + "question_chinese": "我教穿白色衣服的人几种健身器材的使用？", + "answer_chinese": "三种，一种在你右前侧，另一种在你脚下，还有一种在你左前侧", + "evidence": [ + { + "description": "This is the first type.", + "timestamps": { + "start": "00:02:50", + "end": "00:03:00" + } + }, + { + "description": "This is the second type.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:40" + } + }, + { + "description": "This is the third type.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:10" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two types: front-right and feet.", + "B": "Four types: front-right, feet, front-left, behind.", + "C": "One type: you only taught the equipment at your feet, not the ones on either side.", + "D": "Three types: one is to your front-right, another is at your feet, and one more is to your front-left.", + "E": "Five types: front-right, feet, front-left, plus two additional pieces positioned behind you." + }, + "answer_label": "D", + "answer_text": "Three types: one is to your front-right, another is at your feet, and one more is to your front-left.", + "q_id": "D2-P4-pm-1-fitness_video_part002_3_T1", + "correct_answer": "Three types: one is to your front-right, another is at your feet, and one more is to your front-left.", + "original_options": { + "A": "Two types: front-right and feet.", + "B": "Four types: front-right, feet, front-left, behind.", + "C": "One type: you only taught the equipment at your feet, not the ones on either side.", + "D": "Three types: one is to your front-right, another is at your feet, and one more is to your front-left.", + "E": "Five types: front-right, feet, front-left, plus two additional pieces positioned behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 12 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part002", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part002_7", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is my drink?", + "answer": "On the fitness equipment in front of you.", + "question_chinese": "我的饮料在哪里？", + "answer_chinese": "在你面前的健身器材上。", + "evidence": [ + { + "description": "Your drink is on the fitness equipment in front of you.", + "timestamps": { + "start": "00:04:23", + "end": "00:04:24" + } + } + ], + "question_timestamps": "00:04:29", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the fitness equipment in front of you.", + "B": "On the equipment to your right.", + "C": "On the fitness equipment near the window.", + "D": "On the fitness equipment behind you, not on the one directly in front of your view.", + "E": "On a different machine off to your left, away from the equipment that’s directly in front of you." + }, + "answer_label": "A", + "answer_text": "On the fitness equipment in front of you.", + "q_id": "D2-P4-pm-1-fitness_video_part002_7", + "correct_answer": "On the fitness equipment in front of you.", + "original_options": { + "A": "On the fitness equipment in front of you.", + "B": "On the equipment to your right.", + "C": "On the fitness equipment near the window.", + "D": "On the fitness equipment behind you, not on the one directly in front of your view.", + "E": "On a different machine off to your left, away from the equipment that’s directly in front of you." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 19 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part002", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part002_8", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is my drink?", + "answer": "On the fitness equipment next to you.", + "question_chinese": "我的饮料在哪里？", + "answer_chinese": "在你旁边的健身器材上。", + "evidence": [ + { + "description": "Your drink is on the fitness equipment next to you.", + "timestamps": { + "start": "00:05:24", + "end": "00:05:25" + } + }, + { + "description": "Your drink is on the fitness equipment next to you.", + "timestamps": { + "start": "00:04:23", + "end": "00:04:25" + } + } + ], + "question_timestamps": "00:05:29", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the table beside you.", + "B": "On the fitness equipment behind you.", + "C": "On the fitness equipment next to you.", + "D": "On the couch in front of you, not on the fitness equipment close to your side.", + "E": "On the fitness equipment to your right, rather than the one positioned immediately next to you." + }, + "answer_label": "C", + "answer_text": "On the fitness equipment next to you.", + "q_id": "D2-P4-pm-1-fitness_video_part002_8", + "correct_answer": "On the fitness equipment next to you.", + "original_options": { + "A": "On the table beside you.", + "B": "On the fitness equipment behind you.", + "C": "On the fitness equipment next to you.", + "D": "On the couch in front of you, not on the fitness equipment close to your side.", + "E": "On the fitness equipment to your right, rather than the one positioned immediately next to you." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + } +] diff --git a/QAs/D2-P4-pm-1-fitness_video_part003-mcq.json b/QAs/D2-P4-pm-1-fitness_video_part003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c041af0e250d5d14821dfa8a0730b250685a06dd --- /dev/null +++ b/QAs/D2-P4-pm-1-fitness_video_part003-mcq.json @@ -0,0 +1,332 @@ +[ + { + "video_uid": "D2-P4-pm-1-fitness_video_part003", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part003_1_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the ping pong table or the treadmill?", + "answer": "The treadmill is closer to you.", + "question_chinese": "乒乓球桌和跑步机哪个离我更近？", + "answer_chinese": "乒乓球桌离你更近。", + "evidence": [ + { + "description": "The table tennis table is next to you.", + "timestamps": { + "start": "00:05:47", + "end": "00:05:48" + } + }, + { + "description": "The treadmill is on your right.", + "timestamps": { + "start": "00:05:47", + "end": "00:05:48" + } + } + ], + "question_timestamps": "00:05:58", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The ping pong table is closer to you.", + "B": "The treadmill is closer to you." + }, + "answer_label": "B", + "answer_text": "The treadmill is closer to you.", + "q_id": "D2-P4-pm-1-fitness_video_part003_1_T1", + "correct_answer": "The treadmill is closer to you.", + "original_options": { + "A": "The ping pong table is closer to you.", + "B": "The treadmill is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part003", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part003_1_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the ping pong table or the treadmill?", + "answer": "The treadmill is closer to you.", + "question_chinese": "乒乓球桌和跑步机哪个离我更近？", + "answer_chinese": "跑步机离你更近。", + "evidence": [ + { + "description": "The treadmill is in front of you.", + "timestamps": { + "start": "00:11:56", + "end": "00:11:57" + } + }, + { + "description": "The table tennis table is in front of you on your left.", + "timestamps": { + "start": "00:11:56", + "end": "00:11:57" + } + }, + { + "description": "The treadmill is closer to you.", + "timestamps": { + "start": "00:11:56", + "end": "00:11:57" + } + } + ], + "question_timestamps": "00:12:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The table tennis table is closer to you.", + "B": "The treadmill is closer to you." + }, + "answer_label": "B", + "answer_text": "The treadmill is closer to you.", + "q_id": "D2-P4-pm-1-fitness_video_part003_1_T2", + "correct_answer": "The treadmill is closer to you.", + "original_options": { + "A": "The table tennis table is closer to you.", + "B": "The treadmill is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part003", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part003_16", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get my water directly?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接拿到我的水吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The water is right in front of you.", + "timestamps": { + "start": "00:11:27", + "end": "00:11:28" + } + }, + { + "description": "You can get it directly.", + "timestamps": { + "start": "00:11:27", + "end": "00:11:28" + } + } + ], + "question_timestamps": "00:11:31", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you cannot.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "D2-P4-pm-1-fitness_video_part003_16", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No, you cannot.", + "B": "Yes,you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part003", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part003_15", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get my water directly?", + "answer": "No,you can't.", + "question_chinese": "我可以直接拿到我的水吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "Water is on the table.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:15" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:12:00", + "end": "00:12:05" + } + }, + { + "description": "You cannot directly obtain it.", + "timestamps": { + "start": "00:12:00", + "end": "00:12:05" + } + } + ], + "question_timestamps": "00:12:08", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "D2-P4-pm-1-fitness_video_part003_15", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part003", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part003_13", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there anyone on the treadmill?", + "answer": "Yes,there is.", + "question_chinese": "跑步机上有人吗？", + "answer_chinese": "有。", + "evidence": [ + { + "description": "There is a person on the treadmill.", + "timestamps": { + "start": "00:03:01", + "end": "00:03:02" + } + } + ], + "question_timestamps": "00:03:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,there is.", + "B": "No, there isn't." + }, + "answer_label": "A", + "answer_text": "Yes,there is.", + "q_id": "D2-P4-pm-1-fitness_video_part003_13", + "correct_answer": "Yes,there is.", + "original_options": { + "A": "Yes,there is.", + "B": "No, there isn't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part003", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part003_14", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there anyone on the treadmill?", + "answer": "No,there isn't.", + "question_chinese": "跑步机上有人吗？", + "answer_chinese": "没有。", + "evidence": [ + { + "description": "There is no one on the treadmill.", + "timestamps": { + "start": "00:03:13", + "end": "00:03:14" + } + } + ], + "question_timestamps": "00:03:27", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, there is.", + "B": "No,there isn't." + }, + "answer_label": "B", + "answer_text": "No,there isn't.", + "q_id": "D2-P4-pm-1-fitness_video_part003_14", + "correct_answer": "No,there isn't.", + "original_options": { + "A": "Yes, there is.", + "B": "No,there isn't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + } +] diff --git a/QAs/D2-P4-pm-1-fitness_video_part004-mcq.json b/QAs/D2-P4-pm-1-fitness_video_part004-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4e2317b14a5bae3e4d950ba8756ca8206765f971 --- /dev/null +++ b/QAs/D2-P4-pm-1-fitness_video_part004-mcq.json @@ -0,0 +1,274 @@ +[ + { + "video_uid": "D2-P4-pm-1-fitness_video_part004", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part004_2_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the ping-pong table?", + "answer": "The ping-pong table is to your left.", + "question_chinese": "乒乓球桌在我的哪个方向？", + "answer_chinese": "乒乓球桌在你左边。", + "evidence": [ + { + "description": "The table tennis table is on the left side of the treadmill.", + "timestamps": { + "start": "00:14:11", + "end": "00:14:12" + } + }, + { + "description": "You're next to the treadmill.", + "timestamps": { + "start": "00:14:11", + "end": "00:14:12" + } + }, + { + "description": "The table tennis table is on your left.", + "timestamps": { + "start": "00:14:11", + "end": "00:14:12" + } + } + ], + "question_timestamps": "00:14:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It's in front of you.", + "B": "It's to your right.", + "C": "The ping-pong table is to your left.", + "D": "The ping-pong table is behind you, so you would need to turn around to face it.", + "E": "The ping-pong table is off to your right side, so you'd look right to see it clearly." + }, + "answer_label": "C", + "answer_text": "The ping-pong table is to your left.", + "q_id": "D2-P4-pm-1-fitness_video_part004_2_T1", + "correct_answer": "The ping-pong table is to your left.", + "original_options": { + "A": "It's in front of you.", + "B": "It's to your right.", + "C": "The ping-pong table is to your left.", + "D": "The ping-pong table is behind you, so you would need to turn around to face it.", + "E": "The ping-pong table is off to your right side, so you'd look right to see it clearly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 17 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part004", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part004_2_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the ping-pong table?", + "answer": "The ping-pong table is in front of you.", + "question_chinese": "乒乓球桌在我的哪个方向？", + "answer_chinese": "乒乓球桌在你前面。", + "evidence": [ + { + "description": "The table tennis table is on the left side of the treadmill.You're next to the treadmill.The table tennis table is on your left.", + "timestamps": { + "start": "00:14:11", + "end": "00:14:12" + } + } + ], + "question_timestamps": "00:15:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The ping-pong table is in front of you.", + "B": "To your left.", + "C": "Behind you.", + "D": "The ping-pong table is to your right side rather than straight ahead.", + "E": "The ping-pong table is behind you, so you would need to turn around to see it." + }, + "answer_label": "A", + "answer_text": "The ping-pong table is in front of you.", + "q_id": "D2-P4-pm-1-fitness_video_part004_2_T2", + "correct_answer": "The ping-pong table is in front of you.", + "original_options": { + "A": "The ping-pong table is in front of you.", + "B": "To your left.", + "C": "Behind you.", + "D": "The ping-pong table is to your right side rather than straight ahead.", + "E": "The ping-pong table is behind you, so you would need to turn around to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 12, + 16 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part004", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part004_1_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the leg press machine or the kettlebell?", + "answer": "The leg press machine is closer to you.", + "question_chinese": "倒蹬机和壶铃哪个离我更近？", + "answer_chinese": "倒蹬机离你更近。", + "evidence": [ + { + "description": "The reverse kick machine is on your right.The kettle bell is in front of you.", + "timestamps": { + "start": "00:04:11", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:04:22", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The leg press machine is closer to you.", + "B": "The kettlebell is closer to you." + }, + "answer_label": "A", + "answer_text": "The leg press machine is closer to you.", + "q_id": "D2-P4-pm-1-fitness_video_part004_1_T1", + "correct_answer": "The leg press machine is closer to you.", + "original_options": { + "A": "The leg press machine is closer to you.", + "B": "The kettlebell is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part004", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part004_8", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the leg press machine or the kettlebell?", + "answer": "The kettlebell is closer to you.", + "question_chinese": "倒蹬机和壶铃哪个离我更近？", + "answer_chinese": "倒蹬机离你更近。", + "evidence": [ + { + "description": "The kettle bell is in front of you.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:26" + } + }, + { + "description": "The reverse kick machine is behind you.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:26" + } + }, + { + "description": "The kettlebell is closer to you.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:26" + } + } + ], + "question_timestamps": "00:04:36", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The kettlebell is closer to you.", + "B": "The leg press machine is closer to you." + }, + "answer_label": "A", + "answer_text": "The kettlebell is closer to you.", + "q_id": "D2-P4-pm-1-fitness_video_part004_8", + "correct_answer": "The kettlebell is closer to you.", + "original_options": { + "A": "The kettlebell is closer to you.", + "B": "The leg press machine is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part004", + "video_length": "00:20:00", + "qid": "D2-P4-pm-1-fitness_video_part004_1_T2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the leg press machine or the kettlebell?", + "answer": "The kettlebell is closer to you now.", + "question_chinese": "倒蹬机和壶铃哪个离我更近？", + "answer_chinese": "现在壶铃离你更近。", + "evidence": [ + { + "description": "The kettle bell is in front of you.The reverse kick machine is behind you.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:26" + } + } + ], + "question_timestamps": "00:04:38", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The leg press machine is closer to you.", + "B": "The kettlebell is closer to you now." + }, + "answer_label": "B", + "answer_text": "The kettlebell is closer to you now.", + "q_id": "D2-P4-pm-1-fitness_video_part004_1_T2", + "correct_answer": "The kettlebell is closer to you now.", + "original_options": { + "A": "The leg press machine is closer to you.", + "B": "The kettlebell is closer to you now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + } +] diff --git a/QAs/D2-P4-pm-1-fitness_video_part005-mcq.json b/QAs/D2-P4-pm-1-fitness_video_part005-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..173bd7c896c57c449af7d90aa18f483530339687 --- /dev/null +++ b/QAs/D2-P4-pm-1-fitness_video_part005-mcq.json @@ -0,0 +1,327 @@ +[ + { + "video_uid": "D2-P4-pm-1-fitness_video_part005", + "video_length": "00:06:23", + "qid": "D2-P4-pm-1-fitness_video_part005_6", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the dumbbells or the treadmill?", + "answer": "Dumbbells are closer to you.", + "question_chinese": "哑铃和跑步机哪个离我更近？", + "answer_chinese": "哑铃离你更近。", + "evidence": [ + { + "description": "The dumbbell is right in front of you.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:47" + } + }, + { + "description": "The treadmill is on your right back.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:47" + } + }, + { + "description": "Dumbbells are closer to you.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The treadmill is closer to you.", + "B": "Dumbbells are closer to you." + }, + "answer_label": "B", + "answer_text": "Dumbbells are closer to you.", + "q_id": "D2-P4-pm-1-fitness_video_part005_6", + "correct_answer": "Dumbbells are closer to you.", + "original_options": { + "A": "The treadmill is closer to you.", + "B": "Dumbbells are closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 75, + "reasons": [ + "placeholder" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part005", + "video_length": "00:06:23", + "qid": "D2-P4-pm-1-fitness_video_part005_5", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the dumbbells or the treadmill?", + "answer": "The treadmill is closer to you.", + "question_chinese": "哑铃和跑步机哪个离我更近？", + "answer_chinese": "跑步机离你更近。", + "evidence": [ + { + "description": "The treadmill is closer to you.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:47" + } + }, + { + "description": "The treadmill is right next to you.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:47" + } + }, + { + "description": "The dumbbell is on your left back.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:47" + } + }, + { + "description": "The treadmill is closer to you.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:47" + } + } + ], + "question_timestamps": "00:03:57", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The treadmill is closer to you.", + "B": "The dumbbells are closer to you." + }, + "answer_label": "A", + "answer_text": "The treadmill is closer to you.", + "q_id": "D2-P4-pm-1-fitness_video_part005_5", + "correct_answer": "The treadmill is closer to you.", + "original_options": { + "A": "The treadmill is closer to you.", + "B": "The dumbbells are closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part005", + "video_length": "00:06:23", + "qid": "D2-P4-pm-1-fitness_video_part005_14", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Have I done any aerobic exercise?", + "answer": "No,you didn't.", + "question_chinese": "我做有氧运动了吗？", + "answer_chinese": "没做。", + "evidence": [ + { + "description": "You didn't do aerobic exercise.", + "timestamps": { + "start": "00:03:18", + "end": "00:03:19" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,you didn't.", + "B": "Yes, you did." + }, + "answer_label": "A", + "answer_text": "No,you didn't.", + "q_id": "D2-P4-pm-1-fitness_video_part005_14", + "correct_answer": "No,you didn't.", + "original_options": { + "A": "No,you didn't.", + "B": "Yes, you did." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part005", + "video_length": "00:06:23", + "qid": "D2-P4-pm-1-fitness_video_part005_13", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Have I done any aerobic exercise?", + "answer": "Yes,you did.", + "question_chinese": "我做有氧运动了吗？", + "answer_chinese": "做了。", + "evidence": [ + { + "description": "You ran.", + "timestamps": { + "start": "00:04:12", + "end": "00:04:13" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you didn't.", + "B": "Yes,you did." + }, + "answer_label": "B", + "answer_text": "Yes,you did.", + "q_id": "D2-P4-pm-1-fitness_video_part005_13", + "correct_answer": "Yes,you did.", + "original_options": { + "A": "No, you didn't.", + "B": "Yes,you did." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part005", + "video_length": "00:06:23", + "qid": "D2-P4-pm-1-fitness_video_part005_2_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which was closer to me: the treadmill or the elliptical machine?", + "answer": "The elliptical machine is closer to you, and it’s to your left.", + "question_chinese": "跑步机和椭圆机哪个离我更近？", + "answer_chinese": "椭圆机离你更近，在你的左边。", + "evidence": [ + { + "description": "The elliptical machine is right in front of yo.u", + "timestamps": { + "start": "00:03:00", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:03:52", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The elliptical machine is closer to you, and it’s to your left.", + "B": "The treadmill is closer to you, and it’s in front of you." + }, + "answer_label": "A", + "answer_text": "The elliptical machine is closer to you, and it’s to your left.", + "q_id": "D2-P4-pm-1-fitness_video_part005_2_T1", + "correct_answer": "The elliptical machine is closer to you, and it’s to your left.", + "original_options": { + "A": "The elliptical machine is closer to you, and it’s to your left.", + "B": "The treadmill is closer to you, and it’s in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "D2-P4-pm-1-fitness_video_part005", + "video_length": "00:06:23", + "qid": "D2-P4-pm-1-fitness_video_part005_2_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which was closer to me: the treadmill or the elliptical machine?", + "answer": "The treadmill is closer to you, and it’s in front of you.", + "question_chinese": "跑步机和椭圆机哪个离我更近？", + "answer_chinese": "跑步机离你更近，在你的前方。", + "evidence": [ + { + "description": "The elliptical machine is behind you.", + "timestamps": { + "start": "00:03:48", + "end": "00:03:49" + } + } + ], + "question_timestamps": "00:04:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The elliptical machine is closer to you, and it’s to your left.", + "B": "The treadmill is closer to you, and it’s in front of you." + }, + "answer_label": "B", + "answer_text": "The treadmill is closer to you, and it’s in front of you.", + "q_id": "D2-P4-pm-1-fitness_video_part005_2_T2", + "correct_answer": "The treadmill is closer to you, and it’s in front of you.", + "original_options": { + "A": "The elliptical machine is closer to you, and it’s to your left.", + "B": "The treadmill is closer to you, and it’s in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + } +] diff --git a/QAs/D2-P5-am-walking_part001-mcq.json b/QAs/D2-P5-am-walking_part001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5ec3040c2e1e097053012b8b6fd5b4302817d7be --- /dev/null +++ b/QAs/D2-P5-am-walking_part001-mcq.json @@ -0,0 +1,613 @@ +[ + { + "video_uid": "D2-P5-am-walking_part001", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part001_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is Peet's Coffee?", + "answer": "It's on your right.", + "question_chinese": "Peet's Coffee在哪里？", + "answer_chinese": "它在你右边。", + "evidence": [ + { + "description": "It's on your right.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's on your left.", + "B": "It's straight ahead.", + "C": "It's behind you on the left side, not on your right when you’re facing forward.", + "D": "It's on your right.", + "E": "It's in front of you and slightly to the left, rather than being located on your right." + }, + "answer_label": "D", + "answer_text": "It's on your right.", + "q_id": "D2-P5-am-walking_part001_1", + "correct_answer": "It's on your right.", + "original_options": { + "A": "It's on your left.", + "B": "It's straight ahead.", + "C": "It's behind you on the left side, not on your right when you’re facing forward.", + "D": "It's on your right.", + "E": "It's in front of you and slightly to the left, rather than being located on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 17 + ] + }, + { + "video_uid": "D2-P5-am-walking_part001", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part001_2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is Peet's Coffee?", + "answer": "It's behind you to the right.", + "question_chinese": "Peet's Coffee在哪里？", + "answer_chinese": "它在你右后方。", + "evidence": [ + { + "description": "It's on your right.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:28" + } + }, + { + "description": "It's behind you to the right.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It's behind you to the right.", + "B": "It's behind you to the left.", + "C": "It's in front of you.", + "D": "It's in front of you to the left, not behind you to the right from where you’re standing.", + "E": "It's behind you but on the left side, so it isn’t positioned behind-right relative to you." + }, + "answer_label": "A", + "answer_text": "It's behind you to the right.", + "q_id": "D2-P5-am-walking_part001_2", + "correct_answer": "It's behind you to the right.", + "original_options": { + "A": "It's behind you to the right.", + "B": "It's behind you to the left.", + "C": "It's in front of you.", + "D": "It's in front of you to the left, not behind you to the right from where you’re standing.", + "E": "It's behind you but on the left side, so it isn’t positioned behind-right relative to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 17 + ] + }, + { + "video_uid": "D2-P5-am-walking_part001", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part001_11", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I ride the yellow shared bike directly?", + "answer": "No,you can't.", + "question_chinese": "我可以直接骑黄色的共享单车吗?", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The yellow shared bike is in front of you.", + "timestamps": { + "start": "00:04:49", + "end": "00:04:50" + } + }, + { + "description": "They are too far away from you.", + "timestamps": { + "start": "00:04:49", + "end": "00:04:50" + } + }, + { + "description": "You cannot ride it directly.", + "timestamps": { + "start": "00:04:49", + "end": "00:04:50" + } + } + ], + "question_timestamps": "00:04:53", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "D2-P5-am-walking_part001_11", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P5-am-walking_part001", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part001_12", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I ride the yellow shared bike directly?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接骑黄色的共享单车吗?", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The yellow shared bike is right next to you.", + "timestamps": { + "start": "00:05:10", + "end": "00:05:20" + } + }, + { + "description": "You can ride it directly.", + "timestamps": { + "start": "00:05:19", + "end": "00:05:20" + } + } + ], + "question_timestamps": "00:05:23", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "D2-P5-am-walking_part001_12", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P5-am-walking_part001", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part001_2_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many coffee shops did I pass by?", + "answer": "You walked past one coffee shop.", + "question_chinese": "我经过了几个咖啡店？", + "answer_chinese": "你经过了一个咖啡店。", + "evidence": [ + { + "description": "You passed Peet's Coffee.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You passed zero coffee shops.", + "B": "You passed two coffee shops.", + "C": "You walked past three coffee shops during your route, passing them at separate points along the way.", + "D": "You walked past one coffee shop.", + "E": "You walked past four coffee shops in total, with multiple storefronts appearing as you continued forward." + }, + "answer_label": "D", + "answer_text": "You walked past one coffee shop.", + "q_id": "D2-P5-am-walking_part001_2_T1", + "correct_answer": "You walked past one coffee shop.", + "original_options": { + "A": "You passed zero coffee shops.", + "B": "You passed two coffee shops.", + "C": "You walked past three coffee shops during your route, passing them at separate points along the way.", + "D": "You walked past one coffee shop.", + "E": "You walked past four coffee shops in total, with multiple storefronts appearing as you continued forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "D2-P5-am-walking_part001", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part001_2_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many coffee shops did I pass by?", + "answer": "You walked past two coffee shops.", + "question_chinese": "我经过了几个咖啡店？", + "answer_chinese": "你经过了两个咖啡店。", + "evidence": [ + { + "description": "You passed Peet's Coffee.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:35" + } + }, + { + "description": "You passed by Luckin Coffee.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:50" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You passed one coffee shop.", + "B": "You passed three coffee shops.", + "C": "You walked past zero coffee shops on the route, without seeing any coffee shop storefronts.", + "D": "You walked past four coffee shops in total, encountering several coffee shop signs as you continued.", + "E": "You walked past two coffee shops." + }, + "answer_label": "E", + "answer_text": "You walked past two coffee shops.", + "q_id": "D2-P5-am-walking_part001_2_T2", + "correct_answer": "You walked past two coffee shops.", + "original_options": { + "A": "You passed one coffee shop.", + "B": "You passed three coffee shops.", + "C": "You walked past zero coffee shops on the route, without seeing any coffee shop storefronts.", + "D": "You walked past four coffee shops in total, encountering several coffee shop signs as you continued.", + "E": "You walked past two coffee shops." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "D2-P5-am-walking_part001", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part001_4_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many traffic lights did I pass by?", + "answer": "Two.", + "question_chinese": "我经过了几个红绿灯？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "You passed the first one.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:20" + } + }, + { + "description": "You passed the second one.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:49" + } + } + ], + "question_timestamps": "00:06:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four traffic lights in total, spaced out along the route you traveled.", + "E": "Five traffic lights altogether, appearing at multiple intersections as you moved forward." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "D2-P5-am-walking_part001_4_T1", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four traffic lights in total, spaced out along the route you traveled.", + "E": "Five traffic lights altogether, appearing at multiple intersections as you moved forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "D2-P5-am-walking_part001", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part001_4_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many traffic lights did I pass by?", + "answer": "Three.", + "question_chinese": "我经过了几个红绿灯？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "You passed the first one.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:20" + } + }, + { + "description": "You passed the second one.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:49" + } + }, + { + "description": "You passed the third one.", + "timestamps": { + "start": "00:07:33", + "end": "00:07:40" + } + } + ], + "question_timestamps": "00:07:55", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "One traffic light in total, with only a single intersection showing a signal.", + "E": "Five traffic lights altogether, with signals appearing repeatedly at several different intersections." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "D2-P5-am-walking_part001_4_T2", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "One traffic light in total, with only a single intersection showing a signal.", + "E": "Five traffic lights altogether, with signals appearing repeatedly at several different intersections." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "D2-P5-am-walking_part001", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part001_15", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is Lawson?", + "answer": "Rosen is on your right.", + "question_chinese": "罗森在哪里？", + "answer_chinese": "罗森在你右边。", + "evidence": [ + { + "description": "Rosen is on your right.", + "timestamps": { + "start": "00:02:35", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:02:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Rosen is on your left.", + "B": "Rosen is behind you.", + "C": "Rosen is on your right.", + "D": "Rosen is in front of you, positioned straight ahead as you face forward.", + "E": "Rosen is slightly behind you on the left, not on the side you are facing." + }, + "answer_label": "C", + "answer_text": "Rosen is on your right.", + "q_id": "D2-P5-am-walking_part001_15", + "correct_answer": "Rosen is on your right.", + "original_options": { + "A": "Rosen is on your left.", + "B": "Rosen is behind you.", + "C": "Rosen is on your right.", + "D": "Rosen is in front of you, positioned straight ahead as you face forward.", + "E": "Rosen is slightly behind you on the left, not on the side you are facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 15 + ] + }, + { + "video_uid": "D2-P5-am-walking_part001", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part001_16", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is Lawson?", + "answer": "Rosen is behind of you.", + "question_chinese": "罗森在哪里？", + "answer_chinese": "罗森在你后方。", + "evidence": [ + { + "description": "Rosen is on your right.", + "timestamps": { + "start": "00:02:35", + "end": "00:02:37" + } + }, + { + "description": "Rosen is behind of you.", + "timestamps": { + "start": "00:02:37", + "end": "00:02:47" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Rosen is behind of you.", + "B": "Rosen is in front of you.", + "C": "Rosen is to your left.", + "D": "Rosen is on your right side, not positioned behind you as you look around.", + "E": "Rosen is slightly ahead of you on the right, rather than directly behind your position." + }, + "answer_label": "A", + "answer_text": "Rosen is behind of you.", + "q_id": "D2-P5-am-walking_part001_16", + "correct_answer": "Rosen is behind of you.", + "original_options": { + "A": "Rosen is behind of you.", + "B": "Rosen is in front of you.", + "C": "Rosen is to your left.", + "D": "Rosen is on your right side, not positioned behind you as you look around.", + "E": "Rosen is slightly ahead of you on the right, rather than directly behind your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 15 + ] + } +] diff --git a/QAs/D2-P5-am-walking_part002-mcq.json b/QAs/D2-P5-am-walking_part002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f99ebc348b601b15f73c515a12b1ca37383302a3 --- /dev/null +++ b/QAs/D2-P5-am-walking_part002-mcq.json @@ -0,0 +1,422 @@ +[ + { + "video_uid": "D2-P5-am-walking_part002", + "video_length": "00:20:01", + "qid": "D2-P5-am-walking_part002_14", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I buy anything at the first store?", + "answer": "Yes,you did.", + "question_chinese": "在第一家店我买东西了吗？", + "answer_chinese": "买了。", + "evidence": [ + { + "description": "You bought something.", + "timestamps": { + "start": "00:02:45", + "end": "00:02:50" + } + }, + { + "description": "You bought something.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you didn't.", + "B": "Yes,you did." + }, + "answer_label": "B", + "answer_text": "Yes,you did.", + "q_id": "D2-P5-am-walking_part002_14", + "correct_answer": "Yes,you did.", + "original_options": { + "A": "No, you didn't.", + "B": "Yes,you did." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P5-am-walking_part002", + "video_length": "00:20:01", + "qid": "D2-P5-am-walking_part002_7", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can?", + "answer": "The trash can is on your left.", + "question_chinese": "垃圾桶在哪里？", + "answer_chinese": "垃圾桶在你左边。", + "evidence": [ + { + "description": "The trash can is on your left.", + "timestamps": { + "start": "00:10:49", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:10:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The trash can is on your left.", + "B": "On your right.", + "C": "In front of you.", + "D": "The trash can is behind you, slightly off to the right as you face forward.", + "E": "The trash can is ahead of you near the wall, just past your right side." + }, + "answer_label": "A", + "answer_text": "The trash can is on your left.", + "q_id": "D2-P5-am-walking_part002_7", + "correct_answer": "The trash can is on your left.", + "original_options": { + "A": "The trash can is on your left.", + "B": "On your right.", + "C": "In front of you.", + "D": "The trash can is behind you, slightly off to the right as you face forward.", + "E": "The trash can is ahead of you near the wall, just past your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 15 + ] + }, + { + "video_uid": "D2-P5-am-walking_part002", + "video_length": "00:20:01", + "qid": "D2-P5-am-walking_part002_8", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can?", + "answer": "The trash can is behind you.", + "question_chinese": "垃圾桶在哪里？", + "answer_chinese": "垃圾桶在你后边。", + "evidence": [ + { + "description": "The trash can is behind you.", + "timestamps": { + "start": "00:11:52", + "end": "00:11:53" + } + }, + { + "description": "The trash can is behind you.", + "timestamps": { + "start": "00:10:49", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:11:56", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "On your left.", + "C": "The trash can is behind you.", + "D": "The trash can is slightly ahead of you, positioned to your right as you look forward.", + "E": "The trash can is in front of you near the doorway, a little to the left side." + }, + "answer_label": "C", + "answer_text": "The trash can is behind you.", + "q_id": "D2-P5-am-walking_part002_8", + "correct_answer": "The trash can is behind you.", + "original_options": { + "A": "In front of you.", + "B": "On your left.", + "C": "The trash can is behind you.", + "D": "The trash can is slightly ahead of you, positioned to your right as you look forward.", + "E": "The trash can is in front of you near the doorway, a little to the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 17 + ] + }, + { + "video_uid": "D2-P5-am-walking_part002", + "video_length": "00:20:01", + "qid": "D2-P5-am-walking_part002_5", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the checkout counter?", + "answer": "First go right, then proceed forward.", + "question_chinese": "我怎样走到收银台？", + "answer_chinese": "先往右走，再往前方走。", + "evidence": [ + { + "description": "The cashier is in front of you on the right.", + "timestamps": { + "start": "00:01:21", + "end": "00:01:22" + } + }, + { + "description": "You need to go to the front right.", + "timestamps": { + "start": "00:01:21", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go left, then forward.", + "B": "Go straight, then forward.", + "C": "First go right, then continue backward until you reach the checkout counter.", + "D": "First go left, then proceed forward down the aisle to reach the checkout counter.", + "E": "First go right, then proceed forward." + }, + "answer_label": "E", + "answer_text": "First go right, then proceed forward.", + "q_id": "D2-P5-am-walking_part002_5", + "correct_answer": "First go right, then proceed forward.", + "original_options": { + "A": "Go left, then forward.", + "B": "Go straight, then forward.", + "C": "First go right, then continue backward until you reach the checkout counter.", + "D": "First go left, then proceed forward down the aisle to reach the checkout counter.", + "E": "First go right, then proceed forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + }, + { + "video_uid": "D2-P5-am-walking_part002", + "video_length": "00:20:01", + "qid": "D2-P5-am-walking_part002_6", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the checkout counter?", + "answer": "Turn around first, then move forward.", + "question_chinese": "我怎样走到收银台？", + "answer_chinese": "先后转，然后往前方走。", + "evidence": [ + { + "description": "The cashier is behind you.", + "timestamps": { + "start": "00:01:54", + "end": "00:01:55" + } + }, + { + "description": "You need to turn back first and then move forward.", + "timestamps": { + "start": "00:01:54", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:01:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Step backward, then forward.", + "B": "Turn around first, then move forward.", + "C": "Turn left, then forward.", + "D": "First turn left, then move forward toward the checkout counter in one continuous path.", + "E": "First keep facing forward, then walk straight ahead until you arrive at the checkout counter." + }, + "answer_label": "B", + "answer_text": "Turn around first, then move forward.", + "q_id": "D2-P5-am-walking_part002_6", + "correct_answer": "Turn around first, then move forward.", + "original_options": { + "A": "Step backward, then forward.", + "B": "Turn around first, then move forward.", + "C": "Turn left, then forward.", + "D": "First turn left, then move forward toward the checkout counter in one continuous path.", + "E": "First keep facing forward, then walk straight ahead until you arrive at the checkout counter." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "D2-P5-am-walking_part002", + "video_length": "00:20:01", + "qid": "D2-P5-am-walking_part002_3_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What does this shop mainly sell that I saw?", + "answer": "You saw that this shop mainly sells cups. The cups were to your right.", + "question_chinese": "我看到这个店里主要卖什么？", + "answer_chinese": "你看到这个店里主要卖杯子。杯子在你的右边。", + "evidence": [ + { + "description": "This store mainly sells cups.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw that this shop mainly sells cups. The cups were to your right.", + "B": "The shop mainly sells pendants.", + "C": "The shop mainly sells plates.", + "D": "You saw that this shop mainly sells cups, but the cups were positioned to your left.", + "E": "You could see that this shop mainly sells pendants, and the pendants were to your right." + }, + "answer_label": "A", + "answer_text": "You saw that this shop mainly sells cups. The cups were to your right.", + "q_id": "D2-P5-am-walking_part002_3_T1", + "correct_answer": "You saw that this shop mainly sells cups. The cups were to your right.", + "original_options": { + "A": "You saw that this shop mainly sells cups. The cups were to your right.", + "B": "The shop mainly sells pendants.", + "C": "The shop mainly sells plates.", + "D": "You saw that this shop mainly sells cups, but the cups were positioned to your left.", + "E": "You could see that this shop mainly sells pendants, and the pendants were to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "D2-P5-am-walking_part002", + "video_length": "00:20:01", + "qid": "D2-P5-am-walking_part002_3_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What does this shop mainly sell that I saw?", + "answer": "You could see that this shop mainly sells pendants. The pendants were to your left.", + "question_chinese": "我看到这个店里主要卖什么？", + "answer_chinese": "你能看到这个店里主要卖挂坠。挂坠在你的左边。", + "evidence": [ + { + "description": "The shop mainly sells pendants.", + "timestamps": { + "start": "00:03:41", + "end": "00:04:15" + } + } + ], + "question_timestamps": "00:04:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The shop mainly sells cups.", + "B": "You could see that this shop mainly sells pendants. The pendants were to your left.", + "C": "The shop mainly sells bracelets.", + "D": "You could see that this shop mainly sells pendants, but the pendants were to your right.", + "E": "You saw that this shop mainly sells cups, and the cups were positioned to your left." + }, + "answer_label": "B", + "answer_text": "You could see that this shop mainly sells pendants. The pendants were to your left.", + "q_id": "D2-P5-am-walking_part002_3_T2", + "correct_answer": "You could see that this shop mainly sells pendants. The pendants were to your left.", + "original_options": { + "A": "The shop mainly sells cups.", + "B": "You could see that this shop mainly sells pendants. The pendants were to your left.", + "C": "The shop mainly sells bracelets.", + "D": "You could see that this shop mainly sells pendants, but the pendants were to your right.", + "E": "You saw that this shop mainly sells cups, and the cups were positioned to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + } +] diff --git a/QAs/D2-P5-am-walking_part003-mcq.json b/QAs/D2-P5-am-walking_part003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..90a2d8de6ede2397251e8a730bc948c36de54a2c --- /dev/null +++ b/QAs/D2-P5-am-walking_part003-mcq.json @@ -0,0 +1,863 @@ +[ + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_3", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is CITYBOX from me?", + "answer": "On your left.", + "question_chinese": "CITYBOX在我的哪个方向？", + "answer_chinese": "在你左边。", + "evidence": [ + { + "description": "On your left.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:18" + } + } + ], + "question_timestamps": "00:01:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your right.", + "B": "In front of you.", + "C": "It is behind you, slightly off to the right side, not on your left.", + "D": "It is straight ahead of you, a little to the right, rather than on your left.", + "E": "On your left." + }, + "answer_label": "E", + "answer_text": "On your left.", + "q_id": "D2-P5-am-walking_part003_3", + "correct_answer": "On your left.", + "original_options": { + "A": "On your right.", + "B": "In front of you.", + "C": "It is behind you, slightly off to the right side, not on your left.", + "D": "It is straight ahead of you, a little to the right, rather than on your left.", + "E": "On your left." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 16 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is CITYBOX from me?", + "answer": "To your behind.", + "question_chinese": "CITYBOX在我的哪个方向？", + "answer_chinese": "在你后方。", + "evidence": [ + { + "description": "To your behind.", + "timestamps": { + "start": "00:01:17", + "end": "00:01:18" + } + }, + { + "description": "To your behind.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:01:38", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "To your behind.", + "C": "In front of you.", + "D": "It is off to your right side, not directly behind you from your viewpoint.", + "E": "It is above you right now, rather than being positioned behind you at all." + }, + "answer_label": "B", + "answer_text": "To your behind.", + "q_id": "D2-P5-am-walking_part003_4", + "correct_answer": "To your behind.", + "original_options": { + "A": "To your left.", + "B": "To your behind.", + "C": "In front of you.", + "D": "It is off to your right side, not directly behind you from your viewpoint.", + "E": "It is above you right now, rather than being positioned behind you at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 14 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_6_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What color clothes is the person in front of me wearing?", + "answer": "The person ahead of you is wearing black clothes, and that person is in front of you.", + "question_chinese": "我前面的人穿什么颜色的衣服？", + "answer_chinese": "你前面的人穿黑色的衣服，穿黑色衣服的人在你的正前方。", + "evidence": [ + { + "description": "The person in front of you is dressed in black.", + "timestamps": { + "start": "00:14:43", + "end": "00:14:45" + } + } + ], + "question_timestamps": "00:14:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They’re wearing white clothes.", + "B": "They’re wearing gray clothes.", + "C": "The person ahead of you is wearing black clothes, but that person is behind you.", + "D": "The person ahead of you is wearing black clothes, but that person is off to your right side.", + "E": "The person ahead of you is wearing black clothes, and that person is in front of you." + }, + "answer_label": "E", + "answer_text": "The person ahead of you is wearing black clothes, and that person is in front of you.", + "q_id": "D2-P5-am-walking_part003_6_T1", + "correct_answer": "The person ahead of you is wearing black clothes, and that person is in front of you.", + "original_options": { + "A": "They’re wearing white clothes.", + "B": "They’re wearing gray clothes.", + "C": "The person ahead of you is wearing black clothes, but that person is behind you.", + "D": "The person ahead of you is wearing black clothes, but that person is off to your right side.", + "E": "The person ahead of you is wearing black clothes, and that person is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_6_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What color clothes is the person in front of me wearing?", + "answer": "The person ahead of you is wearing white clothes, and that person is in front of you.", + "question_chinese": "我前面的人穿什么颜色的衣服？", + "answer_chinese": "你前面的人穿白色的衣服，穿白色衣服的人在你的正前方。", + "evidence": [ + { + "description": "The person in front of you is dressed in white.", + "timestamps": { + "start": "00:18:10", + "end": "00:18:12" + } + } + ], + "question_timestamps": "00:18:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They’re wearing black clothes.", + "B": "They’re wearing gray clothes.", + "C": "The person ahead of you is wearing white clothes, but that person is behind you.", + "D": "The person ahead of you is wearing white clothes, but that person is off to your left side.", + "E": "The person ahead of you is wearing white clothes, and that person is in front of you." + }, + "answer_label": "E", + "answer_text": "The person ahead of you is wearing white clothes, and that person is in front of you.", + "q_id": "D2-P5-am-walking_part003_6_T2", + "correct_answer": "The person ahead of you is wearing white clothes, and that person is in front of you.", + "original_options": { + "A": "They’re wearing black clothes.", + "B": "They’re wearing gray clothes.", + "C": "The person ahead of you is wearing white clothes, but that person is behind you.", + "D": "The person ahead of you is wearing white clothes, but that person is off to your left side.", + "E": "The person ahead of you is wearing white clothes, and that person is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_11", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the black chair?", + "answer": "No,you can't.", + "question_chinese": "我可以直接坐在黑色椅子上吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The black chair is in front of you.", + "timestamps": { + "start": "00:14:41", + "end": "00:14:42" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:14:41", + "end": "00:14:42" + } + }, + { + "description": "You cannot sit directly on it.", + "timestamps": { + "start": "00:14:41", + "end": "00:14:42" + } + } + ], + "question_timestamps": "00:14:45", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "D2-P5-am-walking_part003_11", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_12", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "May I sit directly on the black chair?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接坐在黑色椅子上吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The black chair is next to you.", + "timestamps": { + "start": "00:14:45", + "end": "00:14:46" + } + }, + { + "description": "You can sit directly on it.", + "timestamps": { + "start": "00:14:45", + "end": "00:14:46" + } + } + ], + "question_timestamps": "00:14:49", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "D2-P5-am-walking_part003_12", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_3_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the black chair right now?", + "answer": "You can sit directly on the black chair; the black chair is to your left.", + "question_chinese": "我可以直接坐在黑色椅子上吗？", + "answer_chinese": "你可以直接坐在黑色椅子上，黑色椅子在你的左边。", + "evidence": [ + { + "description": "you can", + "timestamps": { + "start": "00:14:40", + "end": "00:14:51" + } + } + ], + "question_timestamps": "00:14:54", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can sit directly on the black chair; the black chair is to your left.", + "B": "You cannot sit directly on the black chair; the black chair is to your right." + }, + "answer_label": "A", + "answer_text": "You can sit directly on the black chair; the black chair is to your left.", + "q_id": "D2-P5-am-walking_part003_3_T1", + "correct_answer": "You can sit directly on the black chair; the black chair is to your left.", + "original_options": { + "A": "You can sit directly on the black chair; the black chair is to your left.", + "B": "You cannot sit directly on the black chair; the black chair is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_3_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the black chair right now?", + "answer": "You cannot sit directly on the black chair; the black chair is to your right.", + "question_chinese": "我可以直接坐在黑色椅子上吗？", + "answer_chinese": "你不可以直接坐在黑色椅子上，黑色椅子在你的右边。", + "evidence": [ + { + "description": "You can't.", + "timestamps": { + "start": "00:15:16", + "end": "00:15:19" + } + } + ], + "question_timestamps": "00:15:22", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can sit directly on the black chair; the black chair is to your left.", + "B": "You cannot sit directly on the black chair; the black chair is to your right." + }, + "answer_label": "B", + "answer_text": "You cannot sit directly on the black chair; the black chair is to your right.", + "q_id": "D2-P5-am-walking_part003_3_T2", + "correct_answer": "You cannot sit directly on the black chair; the black chair is to your right.", + "original_options": { + "A": "You can sit directly on the black chair; the black chair is to your left.", + "B": "You cannot sit directly on the black chair; the black chair is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_13", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get to Building 4?", + "answer": "Keep walking straight ahead.", + "question_chinese": "我想去4栋应该怎么走？", + "answer_chinese": "直着往前走。", + "evidence": [ + { + "description": "Four buildings are in front.", + "timestamps": { + "start": "00:16:48", + "end": "00:16:49" + } + }, + { + "description": "Four buildings are in front.", + "timestamps": { + "start": "00:16:48", + "end": "00:16:49" + } + }, + { + "description": "Four buildings are in front.", + "timestamps": { + "start": "00:16:47", + "end": "00:16:48" + } + } + ], + "question_timestamps": "00:16:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right at the entrance.", + "B": "Veer left as you walk.", + "C": "Keep walking straight ahead.", + "D": "Walk straight for a bit, then turn left at the stairs to reach Building 4.", + "E": "Go straight down the corridor, then make a right turn to head toward Building 4." + }, + "answer_label": "C", + "answer_text": "Keep walking straight ahead.", + "q_id": "D2-P5-am-walking_part003_13", + "correct_answer": "Keep walking straight ahead.", + "original_options": { + "A": "Turn right at the entrance.", + "B": "Veer left as you walk.", + "C": "Keep walking straight ahead.", + "D": "Walk straight for a bit, then turn left at the stairs to reach Building 4.", + "E": "Go straight down the corridor, then make a right turn to head toward Building 4." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_14", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to Building 4?", + "answer": "The fourth building is on your right.", + "question_chinese": "我想去4栋应该怎么走？", + "answer_chinese": "四栋就在你右边。", + "evidence": [ + { + "description": "The fourth building is on your right.", + "timestamps": { + "start": "00:17:04", + "end": "00:17:05" + } + }, + { + "description": "The fourth building is on your right.", + "timestamps": { + "start": "00:16:45", + "end": "00:17:00" + } + } + ], + "question_timestamps": "00:17:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The fourth building is on your left.", + "B": "The fourth building is on your right.", + "C": "The fourth building is behind you.", + "D": "The fourth building is in front of you on the left, just past the open area.", + "E": "The fourth building is behind you on the left, so turn around and walk back." + }, + "answer_label": "B", + "answer_text": "The fourth building is on your right.", + "q_id": "D2-P5-am-walking_part003_14", + "correct_answer": "The fourth building is on your right.", + "original_options": { + "A": "The fourth building is on your left.", + "B": "The fourth building is on your right.", + "C": "The fourth building is behind you.", + "D": "The fourth building is in front of you on the left, just past the open area.", + "E": "The fourth building is behind you on the left, so turn around and walk back." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 15 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_2_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is in front of me?", + "answer": "There are two parcel lockers in front of you.", + "question_chinese": "我的前面有什么？", + "answer_chinese": "你的前面有两个快递柜。", + "evidence": [ + { + "description": "There are two express cabinets in front of you.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There are two parcel lockers in front of you.", + "B": "There is a bridge in front of you." + }, + "answer_label": "A", + "answer_text": "There are two parcel lockers in front of you.", + "q_id": "D2-P5-am-walking_part003_2_T2", + "correct_answer": "There are two parcel lockers in front of you.", + "original_options": { + "A": "There are two parcel lockers in front of you.", + "B": "There is a bridge in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_2_T1", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is in front of me?", + "answer": "There is a bridge in front of you.", + "question_chinese": "我的前面有什么？", + "answer_chinese": "你的前面有一座桥。", + "evidence": [ + { + "description": "There is a bridge in front of you.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:02:14", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There are two parcel lockers in front of you.", + "B": "There is a bridge in front of you." + }, + "answer_label": "B", + "answer_text": "There is a bridge in front of you.", + "q_id": "D2-P5-am-walking_part003_2_T1", + "correct_answer": "There is a bridge in front of you.", + "original_options": { + "A": "There are two parcel lockers in front of you.", + "B": "There is a bridge in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_5_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the woman next to me relative to me right now?", + "answer": "The woman next to you is now to your left.", + "question_chinese": "我身边的女士现在在我的哪里？", + "answer_chinese": "你身边的女士现在在你的左边。", + "evidence": [ + { + "description": "On your left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "She is to your right.", + "B": "She is in front of you.", + "C": "The woman next to you is now to your left.", + "D": "The woman next to you is now behind you as you continue moving forward.", + "E": "The woman next to you is now in front of you, slightly ahead as you walk." + }, + "answer_label": "C", + "answer_text": "The woman next to you is now to your left.", + "q_id": "D2-P5-am-walking_part003_5_T2", + "correct_answer": "The woman next to you is now to your left.", + "original_options": { + "A": "She is to your right.", + "B": "She is in front of you.", + "C": "The woman next to you is now to your left.", + "D": "The woman next to you is now behind you as you continue moving forward.", + "E": "The woman next to you is now in front of you, slightly ahead as you walk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 16 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_5_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the woman next to me relative to me right now?", + "answer": "The woman next to you is now to your right.", + "question_chinese": "我身边的女士现在在我的哪里？", + "answer_chinese": "你身边的女士现在在你的右边。", + "evidence": [ + { + "description": "On your right.", + "timestamps": { + "start": "00:18:25", + "end": "00:18:30" + } + } + ], + "question_timestamps": "00:18:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "She is to your left.", + "B": "She is behind you.", + "C": "The woman next to you is now in front of you, walking slightly ahead of you.", + "D": "The woman next to you is now to your right.", + "E": "The woman next to you is now behind you, having fallen a step back from you." + }, + "answer_label": "D", + "answer_text": "The woman next to you is now to your right.", + "q_id": "D2-P5-am-walking_part003_5_T1", + "correct_answer": "The woman next to you is now to your right.", + "original_options": { + "A": "She is to your left.", + "B": "She is behind you.", + "C": "The woman next to you is now in front of you, walking slightly ahead of you.", + "D": "The woman next to you is now to your right.", + "E": "The woman next to you is now behind you, having fallen a step back from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_4_T2", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What color is the umbrella in front of me?", + "answer": "In front of you is a red umbrella.", + "question_chinese": "我面前是一顶什么颜色的伞？", + "answer_chinese": "你面前是一顶红色的伞。", + "evidence": [ + { + "description": "In front of you is a red umbrella.", + "timestamps": { + "start": "00:14:54", + "end": "00:14:55" + } + } + ], + "question_timestamps": "00:14:57", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "In front of you is a red umbrella.", + "B": "In front of you is a white umbrella." + }, + "answer_label": "A", + "answer_text": "In front of you is a red umbrella.", + "q_id": "D2-P5-am-walking_part003_4_T2", + "correct_answer": "In front of you is a red umbrella.", + "original_options": { + "A": "In front of you is a red umbrella.", + "B": "In front of you is a white umbrella." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "D2-P5-am-walking_part003", + "video_length": "00:20:00", + "qid": "D2-P5-am-walking_part003_4_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What color is the umbrella in front of me?", + "answer": "In front of you is a white umbrella.", + "question_chinese": "我面前是一顶什么颜色的伞？", + "answer_chinese": "你面前是一顶白色的伞。", + "evidence": [ + { + "description": "In front of you is a white umbrella.", + "timestamps": { + "start": "00:15:09", + "end": "00:15:10" + } + } + ], + "question_timestamps": "00:15:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "In front of you is a red umbrella.", + "B": "In front of you is a white umbrella." + }, + "answer_label": "B", + "answer_text": "In front of you is a white umbrella.", + "q_id": "D2-P5-am-walking_part003_4_T1", + "correct_answer": "In front of you is a white umbrella.", + "original_options": { + "A": "In front of you is a red umbrella.", + "B": "In front of you is a white umbrella." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + } +] diff --git a/QAs/D3-P3-am-museum_part001-mcq.json b/QAs/D3-P3-am-museum_part001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f0c234d0c534621b341cbc47a3758ef5e8da69d0 --- /dev/null +++ b/QAs/D3-P3-am-museum_part001-mcq.json @@ -0,0 +1,530 @@ +[ + { + "video_uid": "D3-P3-am-museum_part001", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part001_11", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the telescope directly?", + "answer": "No,you can't.", + "question_chinese": "我可以直接拿到望远镜吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The telescope closest to you is in front of you.", + "timestamps": { + "start": "00:07:35", + "end": "00:07:36" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:07:35", + "end": "00:07:36" + } + }, + { + "description": "You cannot directly obtain it.", + "timestamps": { + "start": "00:07:35", + "end": "00:07:36" + } + } + ], + "question_timestamps": "00:07:41", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "D3-P3-am-museum_part001_11", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D3-P3-am-museum_part001", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part001_12", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the telescope directly?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接拿到望远镜吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The telescope is right in front of you.", + "timestamps": { + "start": "00:06:42", + "end": "00:06:43" + } + }, + { + "description": "You can get it directly.", + "timestamps": { + "start": "00:06:42", + "end": "00:06:43" + } + } + ], + "question_timestamps": "00:07:46", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "D3-P3-am-museum_part001_12", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D3-P3-am-museum_part001", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part001_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my exercise trajectory like?", + "answer": "Go straight first, then turn right.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "先直走，然后右转。", + "evidence": [ + { + "description": "Go straight first.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:10" + } + }, + { + "description": "Then turn right.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:13" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go straight.", + "B": "Go straight, then turn left.", + "C": "Turn right first, then continue straight ahead for the rest of the exercise route.", + "D": "Go straight first, then turn right.", + "E": "Go straight at the start, then make a left turn instead of turning right." + }, + "answer_label": "D", + "answer_text": "Go straight first, then turn right.", + "q_id": "D3-P3-am-museum_part001_1", + "correct_answer": "Go straight first, then turn right.", + "original_options": { + "A": "Turn left, then go straight.", + "B": "Go straight, then turn left.", + "C": "Turn right first, then continue straight ahead for the rest of the exercise route.", + "D": "Go straight first, then turn right.", + "E": "Go straight at the start, then make a left turn instead of turning right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "D3-P3-am-museum_part001", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part001_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my exercise trajectory like?", + "answer": "Go straight, then turn right, and then go straight again.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "先直走，然后右转，然后再直走。", + "evidence": [ + { + "description": "Go straight first.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:10" + } + }, + { + "description": "Then turn right.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:20" + } + }, + { + "description": "Then go straight again.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:40" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn right, and then go straight again.", + "B": "Go straight, then turn left.", + "C": "Turn right, then go straight.", + "D": "Go straight, then turn right, then turn left instead of going straight again.", + "E": "Go straight, then turn left, and then go straight again to finish the route." + }, + "answer_label": "A", + "answer_text": "Go straight, then turn right, and then go straight again.", + "q_id": "D3-P3-am-museum_part001_2", + "correct_answer": "Go straight, then turn right, and then go straight again.", + "original_options": { + "A": "Go straight, then turn right, and then go straight again.", + "B": "Go straight, then turn left.", + "C": "Turn right, then go straight.", + "D": "Go straight, then turn right, then turn left instead of going straight again.", + "E": "Go straight, then turn left, and then go straight again to finish the route." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "D3-P3-am-museum_part001", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part001_5_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many supernova pictures did I look at?", + "answer": "Three, on the display screen to your front-left.", + "question_chinese": "我看了多少张超新星的图片？", + "answer_chinese": "3张，在你的左前侧的显示屏上", + "evidence": [ + { + "description": "You are looking at the first picture.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:15" + } + }, + { + "description": "You are looking at the second picture.", + "timestamps": { + "start": "00:04:17", + "end": "00:04:19" + } + }, + { + "description": "You are looking at the third picture.", + "timestamps": { + "start": "00:04:24", + "end": "00:04:26" + } + } + ], + "question_timestamps": "00:04:38", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, on the front-left screen.", + "B": "Two, on the front-left screen.", + "C": "Four, all shown on the display screen positioned to your front-left side during the viewing.", + "D": "Five, viewed on the display screen located slightly to your front-left side during the viewing.", + "E": "Three, on the display screen to your front-left." + }, + "answer_label": "E", + "answer_text": "Three, on the display screen to your front-left.", + "q_id": "D3-P3-am-museum_part001_5_T1", + "correct_answer": "Three, on the display screen to your front-left.", + "original_options": { + "A": "One, on the front-left screen.", + "B": "Two, on the front-left screen.", + "C": "Four, all shown on the display screen positioned to your front-left side during the viewing.", + "D": "Five, viewed on the display screen located slightly to your front-left side during the viewing.", + "E": "Three, on the display screen to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "D3-P3-am-museum_part001", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part001_5_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many supernova pictures did I look at?", + "answer": "Five, on the display screen to your front-left.", + "question_chinese": "我看了多少张超新星的图片？", + "answer_chinese": "5张。在你的左前侧的显示屏上", + "evidence": [ + { + "description": "You are looking at the first picture.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:15" + } + }, + { + "description": "You are looking at the second picture.", + "timestamps": { + "start": "00:04:17", + "end": "00:04:19" + } + }, + { + "description": "You are looking at the third picture.", + "timestamps": { + "start": "00:04:24", + "end": "00:04:26" + } + }, + { + "description": "You are looking at the fourth picture.", + "timestamps": { + "start": "00:04:31", + "end": "00:04:33" + } + }, + { + "description": "You are looking at the fifth picture.", + "timestamps": { + "start": "00:04:40", + "end": "00:04:43" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five, on the display screen to your front-left.", + "B": "Three, on the front-left screen.", + "C": "Four, on the front-left screen.", + "D": "Six, all displayed on the screen positioned to your front-left as you looked.", + "E": "Seven, on the display screen to your front-left when you checked the pictures." + }, + "answer_label": "A", + "answer_text": "Five, on the display screen to your front-left.", + "q_id": "D3-P3-am-museum_part001_5_T2", + "correct_answer": "Five, on the display screen to your front-left.", + "original_options": { + "A": "Five, on the display screen to your front-left.", + "B": "Three, on the front-left screen.", + "C": "Four, on the front-left screen.", + "D": "Six, all displayed on the screen positioned to your front-left as you looked.", + "E": "Seven, on the display screen to your front-left when you checked the pictures." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "D3-P3-am-museum_part001", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part001_2_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the fire extinguisher from me?", + "answer": "The fire extinguisher is on your left.", + "question_chinese": "灭火器在我的哪个方向？", + "answer_chinese": "灭火器在你左边。", + "evidence": [ + { + "description": "The fire extinguisher is on your left.", + "timestamps": { + "start": "00:06:30", + "end": "00:06:31" + } + }, + { + "description": "The fire extinguisher is on your left.", + "timestamps": { + "start": "00:06:23", + "end": "00:06:24" + } + } + ], + "question_timestamps": "00:06:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "The fire extinguisher is on your left.", + "C": "It is on your right.", + "D": "The fire extinguisher is behind you, not on your left side at all from here.", + "E": "The fire extinguisher is slightly in front of you rather than positioned to your left." + }, + "answer_label": "B", + "answer_text": "The fire extinguisher is on your left.", + "q_id": "D3-P3-am-museum_part001_2_T1", + "correct_answer": "The fire extinguisher is on your left.", + "original_options": { + "A": "It is in front of you.", + "B": "The fire extinguisher is on your left.", + "C": "It is on your right.", + "D": "The fire extinguisher is behind you, not on your left side at all from here.", + "E": "The fire extinguisher is slightly in front of you rather than positioned to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 15 + ] + }, + { + "video_uid": "D3-P3-am-museum_part001", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part001_3_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the fire extinguisher from me?", + "answer": "The fire extinguisher is behind you.", + "question_chinese": "灭火器在我的哪个方向？", + "answer_chinese": "灭火器在你后边。", + "evidence": [ + { + "description": "The fire extinguisher is behind you.", + "timestamps": { + "start": "00:06:23", + "end": "00:06:24" + } + }, + { + "description": "The fire extinguisher is behind you.", + "timestamps": { + "start": "00:07:04", + "end": "00:07:05" + } + } + ], + "question_timestamps": "00:07:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is on your left.", + "C": "The fire extinguisher is to your right, not behind you where you are standing.", + "D": "The fire extinguisher is in front of you rather than behind you from your viewpoint.", + "E": "The fire extinguisher is behind you." + }, + "answer_label": "E", + "answer_text": "The fire extinguisher is behind you.", + "q_id": "D3-P3-am-museum_part001_3_T1", + "correct_answer": "The fire extinguisher is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is on your left.", + "C": "The fire extinguisher is to your right, not behind you where you are standing.", + "D": "The fire extinguisher is in front of you rather than behind you from your viewpoint.", + "E": "The fire extinguisher is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 15 + ] + } +] diff --git a/QAs/D3-P3-am-museum_part002-mcq.json b/QAs/D3-P3-am-museum_part002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..159debfd6310d9dad8c8afba88e5965d22c3d5c4 --- /dev/null +++ b/QAs/D3-P3-am-museum_part002-mcq.json @@ -0,0 +1,357 @@ +[ + { + "video_uid": "D3-P3-am-museum_part002", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part002_1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which specimen, number 11 or number 14, is closer to me?", + "answer": "Specimen No. fourteen is closer to you.", + "question_chinese": "11号标本和14号标本哪个离我更近？", + "answer_chinese": "14号标本离你更近。", + "evidence": [ + { + "description": "Specimen No. 14 is right beside you.", + "timestamps": { + "start": "00:01:14", + "end": "00:01:15" + } + }, + { + "description": "Specimen No. 11 is on your left.", + "timestamps": { + "start": "00:01:14", + "end": "00:01:15" + } + }, + { + "description": "Specimen No. 14 is closer to you.", + "timestamps": { + "start": "00:01:14", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:25", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Specimen No. eleven is closer to you.", + "B": "Specimen No. fourteen is closer to you." + }, + "answer_label": "B", + "answer_text": "Specimen No. fourteen is closer to you.", + "q_id": "D3-P3-am-museum_part002_1", + "correct_answer": "Specimen No. fourteen is closer to you.", + "original_options": { + "A": "Specimen No. eleven is closer to you.", + "B": "Specimen No. fourteen is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "D3-P3-am-museum_part002", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part002_2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which specimen, number 11 or number 14, is closer to me?", + "answer": "Sample eleven is closer to you.", + "question_chinese": "11号标本和14号标本哪个离我更近？", + "answer_chinese": "11号标本离你更近。", + "evidence": [ + { + "description": "The specimen number 11 is right next to you.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:15" + } + }, + { + "description": "Sample 14 is on your right.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:15" + } + }, + { + "description": "Sample 11 is closer to you.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:02:25", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Sample fourteen is closer to you.", + "B": "Sample eleven is closer to you." + }, + "answer_label": "B", + "answer_text": "Sample eleven is closer to you.", + "q_id": "D3-P3-am-museum_part002_2", + "correct_answer": "Sample eleven is closer to you.", + "original_options": { + "A": "Sample fourteen is closer to you.", + "B": "Sample eleven is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "D3-P3-am-museum_part002", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part002_9", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I go down the stairs directly?", + "answer": "No,you can't.", + "question_chinese": "我可以直接走楼梯下楼吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The stairs are in front of you.", + "timestamps": { + "start": "00:09:37", + "end": "00:09:38" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:09:37", + "end": "00:09:38" + } + }, + { + "description": "You cannot go downstairs directly by taking the stairs.", + "timestamps": { + "start": "00:09:37", + "end": "00:09:38" + } + } + ], + "question_timestamps": "00:09:42", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "D3-P3-am-museum_part002_9", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D3-P3-am-museum_part002", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part002_10", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I go down the stairs directly?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接走楼梯下楼吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The stairs are right in front of you.", + "timestamps": { + "start": "00:09:50", + "end": "00:09:51" + } + }, + { + "description": "You can go downstairs directly by taking the stairs.", + "timestamps": { + "start": "00:09:50", + "end": "00:09:51" + } + } + ], + "question_timestamps": "00:09:55", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you cannot.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "D3-P3-am-museum_part002_10", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No, you cannot.", + "B": "Yes,you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D3-P3-am-museum_part002", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part002_3_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What animal is in front of me?", + "answer": "In front of you was a king crab.", + "question_chinese": "我面前是什么动物？", + "answer_chinese": "你面前的是帝王蟹", + "evidence": [ + { + "description": "The king crab is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you was a king crab.", + "B": "A cheetah.", + "C": "A lobster.", + "D": "In front of you was a cheetah near the exhibit display, not a crab at all.", + "E": "In front of you was a lobster positioned ahead of you, not the king crab you later mentioned." + }, + "answer_label": "A", + "answer_text": "In front of you was a king crab.", + "q_id": "D3-P3-am-museum_part002_3_T2", + "correct_answer": "In front of you was a king crab.", + "original_options": { + "A": "In front of you was a king crab.", + "B": "A cheetah.", + "C": "A lobster.", + "D": "In front of you was a cheetah near the exhibit display, not a crab at all.", + "E": "In front of you was a lobster positioned ahead of you, not the king crab you later mentioned." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 18 + ] + }, + { + "video_uid": "D3-P3-am-museum_part002", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part002_3_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What animal is in front of me?", + "answer": "In front of you was a cheetah.", + "question_chinese": "我面前是什么动物？", + "answer_chinese": "你面前的是猎豹。", + "evidence": [ + { + "description": "In front of you is a cheetah.", + "timestamps": { + "start": "00:18:47", + "end": "00:18:48" + } + } + ], + "question_timestamps": "00:19:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A king crab.", + "B": "A leopard.", + "C": "In front of you was a leopard, not the cheetah you expected.", + "D": "In front of you was a cheetah.", + "E": "In front of you was a king crab, placed ahead of you on the exhibit floor." + }, + "answer_label": "D", + "answer_text": "In front of you was a cheetah.", + "q_id": "D3-P3-am-museum_part002_3_T1", + "correct_answer": "In front of you was a cheetah.", + "original_options": { + "A": "A king crab.", + "B": "A leopard.", + "C": "In front of you was a leopard, not the cheetah you expected.", + "D": "In front of you was a cheetah.", + "E": "In front of you was a king crab, placed ahead of you on the exhibit floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 12, + 16 + ] + } +] diff --git a/QAs/D3-P3-am-museum_part003-mcq.json b/QAs/D3-P3-am-museum_part003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2b66748e0e9885fe79b82cc80cb1dc22ef156f9b --- /dev/null +++ b/QAs/D3-P3-am-museum_part003-mcq.json @@ -0,0 +1,422 @@ +[ + { + "video_uid": "D3-P3-am-museum_part003", + "video_length": "00:20:02", + "qid": "D3-P3-am-museum_part003_15", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement path after getting off the elevator?", + "answer": "Go to the right first, then go straight and look at the words on the blue background wall.", + "question_chinese": "下了电梯后我的移动轨迹是怎样的？", + "answer_chinese": "先往右边走，进去后直行看蓝色背景墙上的字。", + "evidence": [ + { + "description": "Go to the right first.", + "timestamps": { + "start": "00:09:42", + "end": "00:09:45" + } + }, + { + "description": "Then go straight and look at the words on the blue background wall.", + "timestamps": { + "start": "00:09:50", + "end": "00:09:55" + } + } + ], + "question_timestamps": "00:10:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go left, then straight.", + "B": "Go straight, then right.", + "C": "Go to the left first, then go straight and look at the words on the blue background wall.", + "D": "Go to the right first, then go straight and look at the images on the blue background wall.", + "E": "Go to the right first, then go straight and look at the words on the blue background wall." + }, + "answer_label": "E", + "answer_text": "Go to the right first, then go straight and look at the words on the blue background wall.", + "q_id": "D3-P3-am-museum_part003_15", + "correct_answer": "Go to the right first, then go straight and look at the words on the blue background wall.", + "original_options": { + "A": "Go left, then straight.", + "B": "Go straight, then right.", + "C": "Go to the left first, then go straight and look at the words on the blue background wall.", + "D": "Go to the right first, then go straight and look at the images on the blue background wall.", + "E": "Go to the right first, then go straight and look at the words on the blue background wall." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 18 + ] + }, + { + "video_uid": "D3-P3-am-museum_part003", + "video_length": "00:20:02", + "qid": "D3-P3-am-museum_part003_16", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement pattern after I got off the elevator?", + "answer": "First go to the right, then go straight and look at the words on the blue background wall, and then go to the right.", + "question_chinese": "下了电梯后我的移动轨迹是怎样的？", + "answer_chinese": "先往右边走，进去后直行看蓝色背景墙上的字，然后往右边走。", + "evidence": [ + { + "description": "First go to the right.", + "timestamps": { + "start": "00:09:42", + "end": "00:09:45" + } + }, + { + "description": "then go straight and look at the words on the blue background wall.", + "timestamps": { + "start": "00:09:50", + "end": "00:09:55" + } + }, + { + "description": "Then go to the right.", + "timestamps": { + "start": "00:10:13", + "end": "00:10:15" + } + } + ], + "question_timestamps": "00:10:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Left, straight, then left.", + "B": "Right, straight, then left.", + "C": "First go to the right, then go straight and look at the words on the blue background wall, then go left.", + "D": "First go to the right, then go straight and look at the words on the blue background wall, and then go to the right.", + "E": "First go straight, then turn right to view the words on the blue background wall, and then go right." + }, + "answer_label": "D", + "answer_text": "First go to the right, then go straight and look at the words on the blue background wall, and then go to the right.", + "q_id": "D3-P3-am-museum_part003_16", + "correct_answer": "First go to the right, then go straight and look at the words on the blue background wall, and then go to the right.", + "original_options": { + "A": "Left, straight, then left.", + "B": "Right, straight, then left.", + "C": "First go to the right, then go straight and look at the words on the blue background wall, then go left.", + "D": "First go to the right, then go straight and look at the words on the blue background wall, and then go to the right.", + "E": "First go straight, then turn right to view the words on the blue background wall, and then go right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 21, + 19 + ] + }, + { + "video_uid": "D3-P3-am-museum_part003", + "video_length": "00:20:02", + "qid": "D3-P3-am-museum_part003_9", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many types of animal specimens have I seen?", + "answer": "Three.", + "question_chinese": "我看了几种动物标本？", + "answer_chinese": "三种。", + "evidence": [ + { + "description": "You are looking at the first one.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:05" + } + }, + { + "description": "You are looking at the second type.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:15" + } + }, + { + "description": "You are looking at the third type.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You have seen two different types of animal specimens in total so far.", + "E": "You have seen four types of animal specimens altogether, not three in total." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "D3-P3-am-museum_part003_9", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You have seen two different types of animal specimens in total so far.", + "E": "You have seen four types of animal specimens altogether, not three in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "D3-P3-am-museum_part003", + "video_length": "00:20:02", + "qid": "D3-P3-am-museum_part003_10", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many types of animal specimens have I seen?", + "answer": "Three.", + "question_chinese": "我看了几种动物标本？", + "answer_chinese": "四种。", + "evidence": [ + { + "description": "You are looking at the first one.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:05" + } + }, + { + "description": "You are looking at the second type.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:15" + } + }, + { + "description": "You are looking at the third type.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:40" + } + }, + { + "description": "You are looking at the fourth type.", + "timestamps": { + "start": "00:04:50", + "end": "00:04:55" + } + } + ], + "question_timestamps": "00:05:05", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Five.", + "C": "You have seen two types of animal specimens in total so far.", + "D": "Three.", + "E": "You have seen five different types of animal specimens, counting each distinct one once." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "D3-P3-am-museum_part003_10", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Five.", + "C": "You have seen two types of animal specimens in total so far.", + "D": "Three.", + "E": "You have seen five different types of animal specimens, counting each distinct one once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "D3-P3-am-museum_part003", + "video_length": "00:20:02", + "qid": "D3-P3-am-museum_part003_11", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which bear is closer to me, the gray one or the black one?", + "answer": "The gray bear is closer to you.", + "question_chinese": "灰色的熊和黑色的熊哪个离我更近？", + "answer_chinese": "灰色的熊离你更近。", + "evidence": [ + { + "description": "The gray bear is right next to you.", + "timestamps": { + "start": "00:04:17", + "end": "00:04:20" + } + }, + { + "description": "The black bear is in front of you.", + "timestamps": { + "start": "00:04:17", + "end": "00:04:20" + } + }, + { + "description": "The gray bear is closer to you.", + "timestamps": { + "start": "00:04:17", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:04:30", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The gray bear is closer to you.", + "B": "The black bear is closer to you." + }, + "answer_label": "A", + "answer_text": "The gray bear is closer to you.", + "q_id": "D3-P3-am-museum_part003_11", + "correct_answer": "The gray bear is closer to you.", + "original_options": { + "A": "The gray bear is closer to you.", + "B": "The black bear is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "D3-P3-am-museum_part003", + "video_length": "00:20:02", + "qid": "D3-P3-am-museum_part003_12", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which bear is closer to me, the gray one or the black one?", + "answer": "The gray bear is closer to you.", + "question_chinese": "灰色的熊和黑色的熊哪个离我更近？", + "answer_chinese": "黑色的熊离你更近。", + "evidence": [ + { + "description": "The black bear is right in front of you.", + "timestamps": { + "start": "00:04:44", + "end": "00:04:45" + } + }, + { + "description": "The gray bear is on your left.", + "timestamps": { + "start": "00:04:44", + "end": "00:04:45" + } + }, + { + "description": "The black bear is closer to you.", + "timestamps": { + "start": "00:04:44", + "end": "00:04:45" + } + } + ], + "question_timestamps": "00:04:55", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The gray bear is closer to you.", + "B": "The black bear is closer to you." + }, + "answer_label": "A", + "answer_text": "The gray bear is closer to you.", + "q_id": "D3-P3-am-museum_part003_12", + "correct_answer": "The gray bear is closer to you.", + "original_options": { + "A": "The gray bear is closer to you.", + "B": "The black bear is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + } +] diff --git a/QAs/D3-P3-am-museum_part004-mcq.json b/QAs/D3-P3-am-museum_part004-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..031fc8dd52b67c188ddf0014b366c8ff146457eb --- /dev/null +++ b/QAs/D3-P3-am-museum_part004-mcq.json @@ -0,0 +1,1046 @@ +[ + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_13", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my path after buying ice cream?", + "answer": "Go straight to the left.", + "question_chinese": "买完冰激凌我的移动轨迹是怎样的？", + "answer_chinese": "往左边直走。", + "evidence": [ + { + "description": "Go straight to the left.", + "timestamps": { + "start": "00:07:50", + "end": "00:07:52" + } + } + ], + "question_timestamps": "00:08:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight to the left.", + "B": "Go straight to the right.", + "C": "Go straight ahead.", + "D": "Go straight to the right after you buy the ice cream there.", + "E": "Go straight ahead first, then turn right after you pass the ice cream shop." + }, + "answer_label": "A", + "answer_text": "Go straight to the left.", + "q_id": "D3-P3-am-museum_part004_13", + "correct_answer": "Go straight to the left.", + "original_options": { + "A": "Go straight to the left.", + "B": "Go straight to the right.", + "C": "Go straight ahead.", + "D": "Go straight to the right after you buy the ice cream there.", + "E": "Go straight ahead first, then turn right after you pass the ice cream shop." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 12, + 14 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_14", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my path after buying the ice cream?", + "answer": "Go left first, then turn left again.", + "question_chinese": "买完冰激凌我的移动轨迹是怎样的？", + "answer_chinese": "先往左边走，然后又左转。", + "evidence": [ + { + "description": "Go left first.", + "timestamps": { + "start": "00:07:50", + "end": "00:07:52" + } + }, + { + "description": "Then turn left again.", + "timestamps": { + "start": "00:08:06", + "end": "00:08:10" + } + } + ], + "question_timestamps": "00:08:19", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go right first, then turn left.", + "B": "Go left first, then turn left again.", + "C": "Go straight first, then turn right.", + "D": "Go left first, then turn right afterward instead of turning left again.", + "E": "Go straight first, then turn left afterward, rather than making two left turns." + }, + "answer_label": "B", + "answer_text": "Go left first, then turn left again.", + "q_id": "D3-P3-am-museum_part004_14", + "correct_answer": "Go left first, then turn left again.", + "original_options": { + "A": "Go right first, then turn left.", + "B": "Go left first, then turn left again.", + "C": "Go straight first, then turn right.", + "D": "Go left first, then turn right afterward instead of turning left again.", + "E": "Go straight first, then turn left afterward, rather than making two left turns." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 13 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_6_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which side of me is the ice cream shop on?", + "answer": "The shop is to your left.", + "question_chinese": "冰激凌店铺在我的哪边？", + "answer_chinese": "店铺在你的左边。", + "evidence": [ + { + "description": "The shop is on your left.", + "timestamps": { + "start": "00:05:12", + "end": "00:05:16" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The shop is to your left.", + "B": "The shop is to your right.", + "C": "The shop is in front of you.", + "D": "The shop is behind you, not on your left side at all.", + "E": "The shop is off to your right, rather than positioned to your left." + }, + "answer_label": "A", + "answer_text": "The shop is to your left.", + "q_id": "D3-P3-am-museum_part004_6_T2", + "correct_answer": "The shop is to your left.", + "original_options": { + "A": "The shop is to your left.", + "B": "The shop is to your right.", + "C": "The shop is in front of you.", + "D": "The shop is behind you, not on your left side at all.", + "E": "The shop is off to your right, rather than positioned to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 12, + 13 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_6_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which side of me is the ice cream shop on?", + "answer": "The shop is in front of you.", + "question_chinese": "冰激凌店铺在我的哪边？", + "answer_chinese": "店铺在你的前面。", + "evidence": [ + { + "description": "The shop is in front of you.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:25" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The shop is behind you.", + "B": "The shop is to your left.", + "C": "The shop is to your right instead of directly in front of you.", + "D": "The shop is behind you, meaning you would need to turn around to face it.", + "E": "The shop is in front of you." + }, + "answer_label": "E", + "answer_text": "The shop is in front of you.", + "q_id": "D3-P3-am-museum_part004_6_T1", + "correct_answer": "The shop is in front of you.", + "original_options": { + "A": "The shop is behind you.", + "B": "The shop is to your left.", + "C": "The shop is to your right instead of directly in front of you.", + "D": "The shop is behind you, meaning you would need to turn around to face it.", + "E": "The shop is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 15 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_5", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the trash can from me?", + "answer": "The trash can is in front of you.", + "question_chinese": "垃圾桶在我的哪个方向？", + "answer_chinese": "垃圾桶在你前方。", + "evidence": [ + { + "description": "The trash can is in front of you.", + "timestamps": { + "start": "00:03:50", + "end": "00:03:51" + } + } + ], + "question_timestamps": "00:03:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The trash can is behind you.", + "B": "The trash can is in front of you.", + "C": "The trash can is to your left.", + "D": "The trash can is to your right, just off to the side as you face forward.", + "E": "The trash can is above you, positioned higher than your eye level when you look around." + }, + "answer_label": "B", + "answer_text": "The trash can is in front of you.", + "q_id": "D3-P3-am-museum_part004_5", + "correct_answer": "The trash can is in front of you.", + "original_options": { + "A": "The trash can is behind you.", + "B": "The trash can is in front of you.", + "C": "The trash can is to your left.", + "D": "The trash can is to your right, just off to the side as you face forward.", + "E": "The trash can is above you, positioned higher than your eye level when you look around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 16 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_2_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the trash can from me?", + "answer": "The trash can is in front of you.", + "question_chinese": "垃圾桶在我的哪个方向？", + "answer_chinese": "垃圾桶在你前方。", + "evidence": [ + { + "description": "The trash can is in front of you.", + "timestamps": { + "start": "00:03:50", + "end": "00:03:54" + } + } + ], + "question_timestamps": "00:03:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can is behind you.", + "B": "The trash can is to your right.", + "C": "The trash can is to your left, slightly off to the side as you stand there.", + "D": "The trash can is above you, positioned higher than you and not at ground level.", + "E": "The trash can is in front of you." + }, + "answer_label": "E", + "answer_text": "The trash can is in front of you.", + "q_id": "D3-P3-am-museum_part004_2_T1", + "correct_answer": "The trash can is in front of you.", + "original_options": { + "A": "The trash can is behind you.", + "B": "The trash can is to your right.", + "C": "The trash can is to your left, slightly off to the side as you stand there.", + "D": "The trash can is above you, positioned higher than you and not at ground level.", + "E": "The trash can is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 15 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_2_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the trash can from me?", + "answer": "The trash can is behind you.", + "question_chinese": "垃圾桶在我的哪个方向？", + "answer_chinese": "垃圾桶在你后方。", + "evidence": [ + { + "description": "The trash can is behind of you.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:04" + } + } + ], + "question_timestamps": "00:04:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can is in front of you.", + "B": "The trash can is behind you.", + "C": "The trash can is to your left.", + "D": "The trash can is to your right, not directly in your line of sight as you face forward.", + "E": "The trash can is above you, sitting higher up rather than at the level you’re standing." + }, + "answer_label": "B", + "answer_text": "The trash can is behind you.", + "q_id": "D3-P3-am-museum_part004_2_T2", + "correct_answer": "The trash can is behind you.", + "original_options": { + "A": "The trash can is in front of you.", + "B": "The trash can is behind you.", + "C": "The trash can is to your left.", + "D": "The trash can is to your right, not directly in your line of sight as you face forward.", + "E": "The trash can is above you, sitting higher up rather than at the level you’re standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 17 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_6", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the trash can from me?", + "answer": "The trash can is in front of you.", + "question_chinese": "垃圾桶在我的哪个方向？", + "answer_chinese": "垃圾桶在你后方。", + "evidence": [ + { + "description": "The trash can is behind of you.", + "timestamps": { + "start": "00:04:06", + "end": "00:04:07" + } + }, + { + "description": "The trash can is behind of you.", + "timestamps": { + "start": "00:03:50", + "end": "00:03:51" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The trash can is behind you.", + "B": "The trash can is in front of you.", + "C": "The trash can is to your right.", + "D": "The trash can is to your left, off to the side rather than straight ahead of you.", + "E": "The trash can is below you, positioned lower than where you are standing at that moment." + }, + "answer_label": "B", + "answer_text": "The trash can is in front of you.", + "q_id": "D3-P3-am-museum_part004_6", + "correct_answer": "The trash can is in front of you.", + "original_options": { + "A": "The trash can is behind you.", + "B": "The trash can is in front of you.", + "C": "The trash can is to your right.", + "D": "The trash can is to your left, off to the side rather than straight ahead of you.", + "E": "The trash can is below you, positioned lower than where you are standing at that moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 16 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_4_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Am I holding anything in my hand?", + "answer": "You weren’t holding anything in your hand, and it was to your left.", + "question_chinese": "我手里有东西吗？", + "answer_chinese": "你手里没有东西，它在你左边", + "evidence": [ + { + "description": "You have nothing in your hand.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were holding something in your hand.", + "B": "You weren’t holding anything; in front.", + "C": "You weren’t holding anything in your hand, and it was in front of you instead of to your left.", + "D": "You weren’t holding anything in your hand, and it was to your right rather than on your left side.", + "E": "You weren’t holding anything in your hand, and it was to your left." + }, + "answer_label": "E", + "answer_text": "You weren’t holding anything in your hand, and it was to your left.", + "q_id": "D3-P3-am-museum_part004_4_T1", + "correct_answer": "You weren’t holding anything in your hand, and it was to your left.", + "original_options": { + "A": "You were holding something in your hand.", + "B": "You weren’t holding anything; in front.", + "C": "You weren’t holding anything in your hand, and it was in front of you instead of to your left.", + "D": "You weren’t holding anything in your hand, and it was to your right rather than on your left side.", + "E": "You weren’t holding anything in your hand, and it was to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 20 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_4_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Am I holding anything in my hand?", + "answer": "You were holding something in your hand, and it was in front of you.", + "question_chinese": "我手里有东西吗？", + "answer_chinese": "你手里有东西，它在你前面", + "evidence": [ + { + "description": "You have an ice cream in your hand.", + "timestamps": { + "start": "00:07:38", + "end": "00:07:45" + } + } + ], + "question_timestamps": "00:07:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You weren’t holding anything in your hand.", + "B": "You were holding it behind you.", + "C": "You were holding something in your hand, and it was in front of you.", + "D": "You were holding something in your hand, and it was to your left rather than in front of you.", + "E": "You were holding something in your hand, and it was to your right instead of straight ahead." + }, + "answer_label": "C", + "answer_text": "You were holding something in your hand, and it was in front of you.", + "q_id": "D3-P3-am-museum_part004_4_T2", + "correct_answer": "You were holding something in your hand, and it was in front of you.", + "original_options": { + "A": "You weren’t holding anything in your hand.", + "B": "You were holding it behind you.", + "C": "You were holding something in your hand, and it was in front of you.", + "D": "You were holding something in your hand, and it was to your left rather than in front of you.", + "E": "You were holding something in your hand, and it was to your right instead of straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 19, + 17 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_5_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I buy anything?", + "answer": "You didn’t buy anything; it was in front of you.", + "question_chinese": "我有买东西吗？", + "answer_chinese": "你没有买东西，它在你前面", + "evidence": [ + { + "description": "You didn't buy anything.", + "timestamps": { + "start": "00:04:46", + "end": "00:04:51" + } + } + ], + "question_timestamps": "00:04:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes—you bought something.", + "B": "You didn’t buy anything; behind you.", + "C": "You didn’t buy anything; it was in front of you.", + "D": "Yes—you bought an ice cream, and the ice cream was in your right hand as you held it.", + "E": "You didn’t buy anything; it was to your left, not in front of you where you were facing." + }, + "answer_label": "C", + "answer_text": "You didn’t buy anything; it was in front of you.", + "q_id": "D3-P3-am-museum_part004_5_T1", + "correct_answer": "You didn’t buy anything; it was in front of you.", + "original_options": { + "A": "Yes—you bought something.", + "B": "You didn’t buy anything; behind you.", + "C": "You didn’t buy anything; it was in front of you.", + "D": "Yes—you bought an ice cream, and the ice cream was in your right hand as you held it.", + "E": "You didn’t buy anything; it was to your left, not in front of you where you were facing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 19, + 19 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_5_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I buy anything?", + "answer": "Yes—you bought an ice cream, and the ice cream was in your left hand.", + "question_chinese": "我有买东西吗？", + "answer_chinese": "有，你买了一个冰激凌，冰激凌在你的左手上。", + "evidence": [ + { + "description": "You bought an ice cream.", + "timestamps": { + "start": "00:06:50", + "end": "00:07:40" + } + } + ], + "question_timestamps": "00:07:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You didn’t buy anything.", + "B": "You bought ice cream in your right hand.", + "C": "Yes—you bought something, and it was in your right hand rather than your left hand.", + "D": "Yes—you bought an ice cream, and the ice cream was in your left hand.", + "E": "You didn’t buy anything; you had nothing purchased with you while it stayed in front of you." + }, + "answer_label": "D", + "answer_text": "Yes—you bought an ice cream, and the ice cream was in your left hand.", + "q_id": "D3-P3-am-museum_part004_5_T2", + "correct_answer": "Yes—you bought an ice cream, and the ice cream was in your left hand.", + "original_options": { + "A": "You didn’t buy anything.", + "B": "You bought ice cream in your right hand.", + "C": "Yes—you bought something, and it was in your right hand rather than your left hand.", + "D": "Yes—you bought an ice cream, and the ice cream was in your left hand.", + "E": "You didn’t buy anything; you had nothing purchased with you while it stayed in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 16, + 18 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_1", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my exercise trajectory like?", + "answer": "Turn back first, then walk to the right.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "先往后转，然后往右边走。", + "evidence": [ + { + "description": "Turn back first.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:15" + } + }, + { + "description": "Then walk to the right.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn back, then walk left.", + "B": "Turn back, then walk straight.", + "C": "Turn back first, then head to the left side instead of moving right afterward.", + "D": "Turn back first, then walk to the right.", + "E": "Turn back first, then continue forward rather than shifting to the right as your next movement." + }, + "answer_label": "D", + "answer_text": "Turn back first, then walk to the right.", + "q_id": "D3-P3-am-museum_part004_1", + "correct_answer": "Turn back first, then walk to the right.", + "original_options": { + "A": "Turn back, then walk left.", + "B": "Turn back, then walk straight.", + "C": "Turn back first, then head to the left side instead of moving right afterward.", + "D": "Turn back first, then walk to the right.", + "E": "Turn back first, then continue forward rather than shifting to the right as your next movement." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my exercise trajectory like?", + "answer": "Turn back first, then to the right, and then back again.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "先往后转，然后往右边走，然后又回来。", + "evidence": [ + { + "description": "Turn back first.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:15" + } + }, + { + "description": "Then walk to the right.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:40" + } + }, + { + "description": "And then back again.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:00:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn back first, then to the right, and then back again.", + "B": "Turn forward, then right, then back.", + "C": "Turn back, then left, then back.", + "D": "Turn back first, then go straight ahead, and finally turn to the right at the end.", + "E": "Turn back first, then turn to the right, and then keep facing that way without turning back again." + }, + "answer_label": "A", + "answer_text": "Turn back first, then to the right, and then back again.", + "q_id": "D3-P3-am-museum_part004_2", + "correct_answer": "Turn back first, then to the right, and then back again.", + "original_options": { + "A": "Turn back first, then to the right, and then back again.", + "B": "Turn forward, then right, then back.", + "C": "Turn back, then left, then back.", + "D": "Turn back first, then go straight ahead, and finally turn to the right at the end.", + "E": "Turn back first, then turn to the right, and then keep facing that way without turning back again." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_3_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the vending machine from me?", + "answer": "The vending machine is in front of you.", + "question_chinese": "自动售卖机在我的哪个方向？", + "answer_chinese": "自动售卖机在你前面。", + "evidence": [ + { + "description": "The vending machine is in front of you.", + "timestamps": { + "start": "00:12:40", + "end": "00:12:44" + } + } + ], + "question_timestamps": "00:12:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The vending machine is to your right rather than directly ahead of you.", + "D": "The vending machine is below you, so you would look down instead of looking straight ahead.", + "E": "The vending machine is in front of you." + }, + "answer_label": "E", + "answer_text": "The vending machine is in front of you.", + "q_id": "D3-P3-am-museum_part004_3_T1", + "correct_answer": "The vending machine is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The vending machine is to your right rather than directly ahead of you.", + "D": "The vending machine is below you, so you would look down instead of looking straight ahead.", + "E": "The vending machine is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 16 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_9", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the vending machine from here?", + "answer": "The vending machine is behind you.", + "question_chinese": "自动售卖机在我的哪个方向？", + "answer_chinese": "自动售卖机在你后方。", + "evidence": [ + { + "description": "The vending machine is behind you.", + "timestamps": { + "start": "00:12:49", + "end": "00:12:50" + } + }, + { + "description": "The vending machine is behind you.", + "timestamps": { + "start": "00:12:45", + "end": "00:12:46" + } + } + ], + "question_timestamps": "00:12:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is to your right.", + "C": "The vending machine is on your left side, not directly behind your current position.", + "D": "The vending machine is above you, meaning you would look up instead of turning around to see it.", + "E": "The vending machine is behind you." + }, + "answer_label": "E", + "answer_text": "The vending machine is behind you.", + "q_id": "D3-P3-am-museum_part004_9", + "correct_answer": "The vending machine is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is to your right.", + "C": "The vending machine is on your left side, not directly behind your current position.", + "D": "The vending machine is above you, meaning you would look up instead of turning around to see it.", + "E": "The vending machine is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 18 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_3_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the vending machine from me?", + "answer": "The vending machine is behind you.", + "question_chinese": "自动售卖机在我的哪个方向？", + "answer_chinese": "自动售卖机在你后方。", + "evidence": [ + { + "description": "The vending machine is behind you.", + "timestamps": { + "start": "00:12:44", + "end": "00:12:50" + } + } + ], + "question_timestamps": "00:12:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is to your right.", + "C": "The vending machine is to your left side rather than positioned behind you.", + "D": "The vending machine is behind you.", + "E": "The vending machine is above you, so you would look upward instead of turning around." + }, + "answer_label": "D", + "answer_text": "The vending machine is behind you.", + "q_id": "D3-P3-am-museum_part004_3_T2", + "correct_answer": "The vending machine is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is to your right.", + "C": "The vending machine is to your left side rather than positioned behind you.", + "D": "The vending machine is behind you.", + "E": "The vending machine is above you, so you would look upward instead of turning around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 15 + ] + }, + { + "video_uid": "D3-P3-am-museum_part004", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part004_10", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the vending machine from me?", + "answer": "The vending machine is behind you.", + "question_chinese": "自动售卖机在我的哪个方向？", + "answer_chinese": "自动售卖机在你面前。", + "evidence": [ + { + "description": "The vending machine is in front of you.", + "timestamps": { + "start": "00:12:56", + "end": "00:12:57" + } + } + ], + "question_timestamps": "00:13:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "The vending machine is behind you.", + "C": "It is to your left.", + "D": "The vending machine is to your right rather than behind your current position.", + "E": "The vending machine is above you, meaning it is overhead instead of behind you." + }, + "answer_label": "B", + "answer_text": "The vending machine is behind you.", + "q_id": "D3-P3-am-museum_part004_10", + "correct_answer": "The vending machine is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "The vending machine is behind you.", + "C": "It is to your left.", + "D": "The vending machine is to your right rather than behind your current position.", + "E": "The vending machine is above you, meaning it is overhead instead of behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 14 + ] + } +] diff --git a/QAs/D3-P3-am-museum_part005-mcq.json b/QAs/D3-P3-am-museum_part005-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d2fb0a8a198098e4635c1c81ce80be83bf69d16b --- /dev/null +++ b/QAs/D3-P3-am-museum_part005-mcq.json @@ -0,0 +1,437 @@ +[ + { + "video_uid": "D3-P3-am-museum_part005", + "video_length": "00:20:05", + "qid": "D3-P3-am-museum_part005_3_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the penguin or the deer?", + "answer": "The deer is closer to you.", + "question_chinese": "企鹅和鹿哪个离我更近？", + "answer_chinese": "鹿离你更近。", + "evidence": [ + { + "description": "The deer is right next to you.Penguins are ahead of you.Deer are closer to you.", + "timestamps": { + "start": "00:15:05", + "end": "00:15:10" + } + } + ], + "question_timestamps": "00:15:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The penguin is closer to you.", + "B": "The deer is closer to you." + }, + "answer_label": "B", + "answer_text": "The deer is closer to you.", + "q_id": "D3-P3-am-museum_part005_3_T1", + "correct_answer": "The deer is closer to you.", + "original_options": { + "A": "The penguin is closer to you.", + "B": "The deer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "D3-P3-am-museum_part005", + "video_length": "00:20:05", + "qid": "D3-P3-am-museum_part005_3_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the penguin or the deer?", + "answer": "The penguin is closer to you.", + "question_chinese": "企鹅和鹿哪个离我更近？", + "answer_chinese": "企鹅离你更近。", + "evidence": [ + { + "description": "The deer is right next to you.Penguins are ahead of you.Deer are closer to you.", + "timestamps": { + "start": "00:15:05", + "end": "00:15:10" + } + } + ], + "question_timestamps": "00:16:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The penguin is closer to you.", + "B": "The deer is closer to you." + }, + "answer_label": "A", + "answer_text": "The penguin is closer to you.", + "q_id": "D3-P3-am-museum_part005_3_T2", + "correct_answer": "The penguin is closer to you.", + "original_options": { + "A": "The penguin is closer to you.", + "B": "The deer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "D3-P3-am-museum_part005", + "video_length": "00:20:05", + "qid": "D3-P3-am-museum_part005_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory like?", + "answer": "You go ahead first, buy a bottle of beverage at the vending machine, and then come back.", + "question_chinese": "我的移动轨迹是怎样的?", + "answer_chinese": "你先往前走，在自动售卖机买了一瓶饮料，然后又回来。", + "evidence": [ + { + "description": "You go ahead first.", + "timestamps": { + "start": "00:00:57", + "end": "00:00:59" + } + }, + { + "description": "You buy a bottle of beverage at the vending machine.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:10" + } + }, + { + "description": "Then come back.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:01:38", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You go ahead first, buy a bottle of beverage at the vending machine, and then come back.", + "B": "You go ahead and then turn left.", + "C": "You walk straight ahead and keep going.", + "D": "You head forward to the vending machine but continue past it instead of turning back.", + "E": "You walk out first, return briefly, and then head away again rather than coming back once." + }, + "answer_label": "A", + "answer_text": "You go ahead first, buy a bottle of beverage at the vending machine, and then come back.", + "q_id": "D3-P3-am-museum_part005_1", + "correct_answer": "You go ahead first, buy a bottle of beverage at the vending machine, and then come back.", + "original_options": { + "A": "You go ahead first, buy a bottle of beverage at the vending machine, and then come back.", + "B": "You go ahead and then turn left.", + "C": "You walk straight ahead and keep going.", + "D": "You head forward to the vending machine but continue past it instead of turning back.", + "E": "You walk out first, return briefly, and then head away again rather than coming back once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "D3-P3-am-museum_part005", + "video_length": "00:20:05", + "qid": "D3-P3-am-museum_part005_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory like?", + "answer": "You go ahead and buy a bottle of beverage from the vending machine, then come back and sit down.", + "question_chinese": "我的移动轨迹是怎样的?", + "answer_chinese": "你先往前走，在自动售卖机买了一瓶饮料，然后又回来，然后坐下。", + "evidence": [ + { + "description": "You go ahead", + "timestamps": { + "start": "00:00:57", + "end": "00:00:59" + } + }, + { + "description": "Buy a bottle of beverage from the vending machine.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:10" + } + }, + { + "description": "Then come back.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:35" + } + }, + { + "description": "Then sit down.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:01:52", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You go ahead, then stand up.", + "B": "You go ahead and buy a bottle of beverage from the vending machine, then come back and sit down.", + "C": "You go back, then sit down.", + "D": "You go to the vending machine for a drink, then come back and remain standing nearby.", + "E": "You walk to the vending machine, pick up a beverage, and then continue onward without returning to sit." + }, + "answer_label": "B", + "answer_text": "You go ahead and buy a bottle of beverage from the vending machine, then come back and sit down.", + "q_id": "D3-P3-am-museum_part005_2", + "correct_answer": "You go ahead and buy a bottle of beverage from the vending machine, then come back and sit down.", + "original_options": { + "A": "You go ahead, then stand up.", + "B": "You go ahead and buy a bottle of beverage from the vending machine, then come back and sit down.", + "C": "You go back, then sit down.", + "D": "You go to the vending machine for a drink, then come back and remain standing nearby.", + "E": "You walk to the vending machine, pick up a beverage, and then continue onward without returning to sit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "D3-P3-am-museum_part005", + "video_length": "00:20:05", + "qid": "D3-P3-am-museum_part005_2_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What animal is in front of me?", + "answer": "The animal in front of you is a polar bear.", + "question_chinese": "我面前是什么动物？", + "answer_chinese": "你面前的动物是北极熊。", + "evidence": [ + { + "description": "In front of you is a polar bear.", + "timestamps": { + "start": "00:10:56", + "end": "00:10:58" + } + } + ], + "question_timestamps": "00:11:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The animal is a reindeer.", + "B": "The animal is a brown bear.", + "C": "The animal in front of you is a penguin, not a bear.", + "D": "The animal in front of you is a polar bear.", + "E": "The animal in front of you appears to be a seal rather than a polar bear." + }, + "answer_label": "D", + "answer_text": "The animal in front of you is a polar bear.", + "q_id": "D3-P3-am-museum_part005_2_T1", + "correct_answer": "The animal in front of you is a polar bear.", + "original_options": { + "A": "The animal is a reindeer.", + "B": "The animal is a brown bear.", + "C": "The animal in front of you is a penguin, not a bear.", + "D": "The animal in front of you is a polar bear.", + "E": "The animal in front of you appears to be a seal rather than a polar bear." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 16 + ] + }, + { + "video_uid": "D3-P3-am-museum_part005", + "video_length": "00:20:05", + "qid": "D3-P3-am-museum_part005_2_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What animal is in front of me?", + "answer": "The animal in front of you is a reindeer.", + "question_chinese": "我面前是什么动物？", + "answer_chinese": "你面前的动物是驯鹿。", + "evidence": [ + { + "description": "The animal in front of you is a reindeer.", + "timestamps": { + "start": "00:12:15", + "end": "00:12:25" + } + } + ], + "question_timestamps": "00:12:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The animal is a polar bear.", + "B": "The animal in front of you is a reindeer.", + "C": "The animal is an elk.", + "D": "The animal in front of you is a moose, not a reindeer.", + "E": "The animal in front of you looks like a deer with no antlers, not a reindeer." + }, + "answer_label": "B", + "answer_text": "The animal in front of you is a reindeer.", + "q_id": "D3-P3-am-museum_part005_2_T2", + "correct_answer": "The animal in front of you is a reindeer.", + "original_options": { + "A": "The animal is a polar bear.", + "B": "The animal in front of you is a reindeer.", + "C": "The animal is an elk.", + "D": "The animal in front of you is a moose, not a reindeer.", + "E": "The animal in front of you looks like a deer with no antlers, not a reindeer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 12, + 16 + ] + }, + { + "video_uid": "D3-P3-am-museum_part005", + "video_length": "00:20:05", + "qid": "D3-P3-am-museum_part005_4_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the straight ladder relative to me?", + "answer": "The straight ladder is to your front-right.", + "question_chinese": "直梯在我的什么位置？", + "answer_chinese": "直梯在你的右前方。", + "evidence": [ + { + "description": "The elevator is in front of you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:22", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The straight ladder is to your front-right.", + "B": "The straight ladder is to your front-left." + }, + "answer_label": "A", + "answer_text": "The straight ladder is to your front-right.", + "q_id": "D3-P3-am-museum_part005_4_T2", + "correct_answer": "The straight ladder is to your front-right.", + "original_options": { + "A": "The straight ladder is to your front-right.", + "B": "The straight ladder is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "D3-P3-am-museum_part005", + "video_length": "00:20:05", + "qid": "D3-P3-am-museum_part005_4_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the straight ladder relative to me?", + "answer": "The straight ladder is to your front-left.", + "question_chinese": "直梯在我的什么位置？", + "answer_chinese": "直梯在你的左前方。", + "evidence": [ + { + "description": "You turned around, faced the elevator, and walked a few steps.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The straight ladder is to your front-right.", + "B": "The straight ladder is to your front-left." + }, + "answer_label": "B", + "answer_text": "The straight ladder is to your front-left.", + "q_id": "D3-P3-am-museum_part005_4_T1", + "correct_answer": "The straight ladder is to your front-left.", + "original_options": { + "A": "The straight ladder is to your front-right.", + "B": "The straight ladder is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + } +] diff --git a/QAs/D3-P3-am-museum_part006-mcq.json b/QAs/D3-P3-am-museum_part006-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..10a653fa1b730979ba4a2d8ec62571f6550f61d6 --- /dev/null +++ b/QAs/D3-P3-am-museum_part006-mcq.json @@ -0,0 +1,878 @@ +[ + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_9", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What will my movement trajectory be after I come out?", + "answer": "Go straight first, then turn right to enter the next exhibition hall.", + "question_chinese": "出来后我的移动轨迹是怎样的？", + "answer_chinese": "先直行，然后右转进入下一个展馆。", + "evidence": [ + { + "description": "Go straight first", + "timestamps": { + "start": "00:14:25", + "end": "00:14:30" + } + }, + { + "description": "Then turn right to enter the next exhibition hall.", + "timestamps": { + "start": "00:14:30", + "end": "00:14:36" + } + } + ], + "question_timestamps": "00:14:39", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight first, then turn right to enter the next exhibition hall.", + "B": "Turn right first, then go straight.", + "C": "Go straight, then turn left.", + "D": "Go straight first, then turn right to enter the previous exhibition hall instead of the next one.", + "E": "Go straight out, then turn left and continue until you reach the next exhibition hall entrance." + }, + "answer_label": "A", + "answer_text": "Go straight first, then turn right to enter the next exhibition hall.", + "q_id": "D3-P3-am-museum_part006_9", + "correct_answer": "Go straight first, then turn right to enter the next exhibition hall.", + "original_options": { + "A": "Go straight first, then turn right to enter the next exhibition hall.", + "B": "Turn right first, then go straight.", + "C": "Go straight, then turn left.", + "D": "Go straight first, then turn right to enter the previous exhibition hall instead of the next one.", + "E": "Go straight out, then turn left and continue until you reach the next exhibition hall entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 16 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_10", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What will my movement trajectory be after I come out?", + "answer": "Go straight first, then turn right to enter the next exhibition hall, and then move forward to see the dolls.", + "question_chinese": "出来后我的移动轨迹是怎样的？", + "answer_chinese": "先直行，然后右转进入下一个展馆，然后往前走看玩偶。", + "evidence": [ + { + "description": "Go straight first.", + "timestamps": { + "start": "00:14:25", + "end": "00:14:30" + } + }, + { + "description": "Then turn right to enter the next exhibition hall.", + "timestamps": { + "start": "00:14:30", + "end": "00:14:35" + } + } + ], + "question_timestamps": "00:14:47", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn left.", + "B": "Turn right, then go straight.", + "C": "Go straight first, then turn right into the next exhibition hall, and back up instead of moving forward to the dolls.", + "D": "Go straight, turn right into the next exhibition hall, then continue forward to see sculptures rather than the dolls.", + "E": "Go straight first, then turn right to enter the next exhibition hall, and then move forward to see the dolls." + }, + "answer_label": "E", + "answer_text": "Go straight first, then turn right to enter the next exhibition hall, and then move forward to see the dolls.", + "q_id": "D3-P3-am-museum_part006_10", + "correct_answer": "Go straight first, then turn right to enter the next exhibition hall, and then move forward to see the dolls.", + "original_options": { + "A": "Go straight, then turn left.", + "B": "Turn right, then go straight.", + "C": "Go straight first, then turn right into the next exhibition hall, and back up instead of moving forward to the dolls.", + "D": "Go straight, turn right into the next exhibition hall, then continue forward to see sculptures rather than the dolls.", + "E": "Go straight first, then turn right to enter the next exhibition hall, and then move forward to see the dolls." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 19 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_2_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "On this small electronic screen, how many kinds of animals did I see?", + "answer": "Three types.", + "question_chinese": "在这个小电子屏上我看了几种动物？", + "answer_chinese": "三种。", + "evidence": [ + { + "description": "You are looking at the first one.", + "timestamps": { + "start": "00:06:35", + "end": "00:06:38" + } + }, + { + "description": "You are looking at the second type.", + "timestamps": { + "start": "00:06:39", + "end": "00:06:41" + } + }, + { + "description": "You are looking at the third type.", + "timestamps": { + "start": "00:06:43", + "end": "00:06:46" + } + } + ], + "question_timestamps": "00:06:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two types.", + "B": "Three types.", + "C": "Four types.", + "D": "Only two kinds of animals appeared on the small electronic screen you viewed.", + "E": "You saw four different kinds of animals displayed on that small electronic screen." + }, + "answer_label": "B", + "answer_text": "Three types.", + "q_id": "D3-P3-am-museum_part006_2_T1", + "correct_answer": "Three types.", + "original_options": { + "A": "Two types.", + "B": "Three types.", + "C": "Four types.", + "D": "Only two kinds of animals appeared on the small electronic screen you viewed.", + "E": "You saw four different kinds of animals displayed on that small electronic screen." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_2_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "On this small electronic screen, how many kinds of animals did I see?", + "answer": "Six types.", + "question_chinese": "在这个小电子屏上我看了几种动物？", + "answer_chinese": "六种。", + "evidence": [ + { + "description": "You are looking at the first one.", + "timestamps": { + "start": "00:06:35", + "end": "00:06:38" + } + }, + { + "description": "You are looking at the second type.", + "timestamps": { + "start": "00:06:39", + "end": "00:06:41" + } + }, + { + "description": "You are looking at the third type.", + "timestamps": { + "start": "00:06:43", + "end": "00:06:46" + } + }, + { + "description": "You are looking at the fourth type.", + "timestamps": { + "start": "00:06:49", + "end": "00:06:50" + } + }, + { + "description": "You are looking at the fifth one.", + "timestamps": { + "start": "00:06:54", + "end": "00:06:56" + } + }, + { + "description": "You are looking at the sixth one.", + "timestamps": { + "start": "00:06:58", + "end": "00:07:00" + } + } + ], + "question_timestamps": "00:07:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five types.", + "B": "Seven types.", + "C": "Six types.", + "D": "You saw five different kinds of animals on the small electronic screen.", + "E": "There were seven animal types shown on the small electronic screen you looked at." + }, + "answer_label": "C", + "answer_text": "Six types.", + "q_id": "D3-P3-am-museum_part006_2_T2", + "correct_answer": "Six types.", + "original_options": { + "A": "Five types.", + "B": "Seven types.", + "C": "Six types.", + "D": "You saw five different kinds of animals on the small electronic screen.", + "E": "There were seven animal types shown on the small electronic screen you looked at." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the fire extinguisher directly?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接拿到灭火器吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The fire extinguisher is on your left.", + "timestamps": { + "start": "00:03:56", + "end": "00:03:59" + } + }, + { + "description": "You can get it directly.", + "timestamps": { + "start": "00:03:56", + "end": "00:03:59" + } + } + ], + "question_timestamps": "00:04:02", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you cannot.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "D3-P3-am-museum_part006_2", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No, you cannot.", + "B": "Yes,you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the fire extinguisher directly?", + "answer": "No,you can't.", + "question_chinese": "我可以直接拿到灭火器吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The fire extinguisher is behind your left.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:03" + } + }, + { + "description": "It is too far away from you.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:03" + } + }, + { + "description": "You cannot take it directly.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:03" + } + } + ], + "question_timestamps": "00:04:06", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "D3-P3-am-museum_part006_1", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_15", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the panda plush toy directly?", + "answer": "No,you can't.", + "question_chinese": "我可以直接拿到熊猫玩偶吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The panda doll is in front of you.", + "timestamps": { + "start": "00:19:51", + "end": "00:19:52" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:19:51", + "end": "00:19:52" + } + }, + { + "description": "You cannot directly obtain it.", + "timestamps": { + "start": "00:19:51", + "end": "00:19:52" + } + } + ], + "question_timestamps": "00:19:55", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "D3-P3-am-museum_part006_15", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_16", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the panda plush toy directly?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接拿到熊猫玩偶吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The panda doll is right next to you.", + "timestamps": { + "start": "00:19:54", + "end": "00:19:55" + } + }, + { + "description": "You can get it directly.", + "timestamps": { + "start": "00:19:54", + "end": "00:19:55" + } + } + ], + "question_timestamps": "00:19:58", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "D3-P3-am-museum_part006_16", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_5_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I pick up to look at?", + "answer": "Two.", + "question_chinese": "我拿起来几个东西看了？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "You are looking at the first one.", + "timestamps": { + "start": "00:14:37", + "end": "00:14:42" + } + }, + { + "description": "You're looking at the second one.", + "timestamps": { + "start": "00:15:07", + "end": "00:15:12" + } + } + ], + "question_timestamps": "00:15:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You picked up four items to look at closely before putting them back down.", + "E": "You did not pick up anything to look at; you only looked without touching any items." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "D3-P3-am-museum_part006_5_T1", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You picked up four items to look at closely before putting them back down.", + "E": "You did not pick up anything to look at; you only looked without touching any items." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_5_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I pick up to look at?", + "answer": "Three.", + "question_chinese": "我拿起来几个东西看了？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "You are looking at the first one.", + "timestamps": { + "start": "00:14:37", + "end": "00:14:42" + } + }, + { + "description": "You're looking at the second one.", + "timestamps": { + "start": "00:15:07", + "end": "00:15:12" + } + }, + { + "description": "You're looking at the third one.", + "timestamps": { + "start": "00:15:40", + "end": "00:15:45" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You picked up one item to look at briefly before setting it back down.", + "E": "You picked up five items to look at, one after another, before setting them down." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "D3-P3-am-museum_part006_5_T2", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You picked up one item to look at briefly before setting it back down.", + "E": "You picked up five items to look at, one after another, before setting them down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_6_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the butterfly specimen?", + "answer": "The butterfly specimen was to your right.", + "question_chinese": "蝴蝶标本在我的哪个方向？", + "answer_chinese": "蝴蝶标本在你的右边。", + "evidence": [ + { + "description": "The butterfly specimen is on your right.", + "timestamps": { + "start": "00:18:04", + "end": "00:18:05" + } + } + ], + "question_timestamps": "00:18:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The butterfly specimen was to your right.", + "B": "The butterfly specimen was to your left." + }, + "answer_label": "A", + "answer_text": "The butterfly specimen was to your right.", + "q_id": "D3-P3-am-museum_part006_6_T1", + "correct_answer": "The butterfly specimen was to your right.", + "original_options": { + "A": "The butterfly specimen was to your right.", + "B": "The butterfly specimen was to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_6_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the butterfly specimen?", + "answer": "The butterfly specimen was to your left.", + "question_chinese": "蝴蝶标本在我的哪个方向？", + "answer_chinese": "蝴蝶标本在你的左边。", + "evidence": [ + { + "description": "The butterfly specimen is on your left.", + "timestamps": { + "start": "00:18:15", + "end": "00:18:16" + } + }, + { + "description": "The butterfly specimen is on your left.", + "timestamps": { + "start": "00:18:05", + "end": "00:18:06" + } + } + ], + "question_timestamps": "00:18:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The butterfly specimen was to your right.", + "B": "The butterfly specimen was to your left." + }, + "answer_label": "B", + "answer_text": "The butterfly specimen was to your left.", + "q_id": "D3-P3-am-museum_part006_6_T2", + "correct_answer": "The butterfly specimen was to your left.", + "original_options": { + "A": "The butterfly specimen was to your right.", + "B": "The butterfly specimen was to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_7", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the entrance from where I am?", + "answer": "The door is on your left.", + "question_chinese": "门口在我的哪个方向？", + "answer_chinese": "门口在你的左边。", + "evidence": [ + { + "description": "The door is on your left.", + "timestamps": { + "start": "00:13:47", + "end": "00:13:48" + } + }, + { + "description": "The door is on your left.", + "timestamps": { + "start": "00:13:46", + "end": "00:13:47" + } + } + ], + "question_timestamps": "00:13:51", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The door is in front of you.", + "B": "The door is on your left.", + "C": "The door is behind you.", + "D": "The entrance door is directly ahead of you from where you're standing, not off to the left.", + "E": "From your position, the doorway is behind your back, so you'd need to turn around to face it." + }, + "answer_label": "B", + "answer_text": "The door is on your left.", + "q_id": "D3-P3-am-museum_part006_7", + "correct_answer": "The door is on your left.", + "original_options": { + "A": "The door is in front of you.", + "B": "The door is on your left.", + "C": "The door is behind you.", + "D": "The entrance door is directly ahead of you from where you're standing, not off to the left.", + "E": "From your position, the doorway is behind your back, so you'd need to turn around to face it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 17, + 18 + ] + }, + { + "video_uid": "D3-P3-am-museum_part006", + "video_length": "00:20:00", + "qid": "D3-P3-am-museum_part006_8", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the entrance from where I am?", + "answer": "The door is right in front of you.", + "question_chinese": "门口在我的哪个方向？", + "answer_chinese": "门口就在你面前。", + "evidence": [ + { + "description": "The door is right in front of you.", + "timestamps": { + "start": "00:13:53", + "end": "00:13:54" + } + } + ], + "question_timestamps": "00:13:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is to your left.", + "B": "The door is on your right.", + "C": "The doorway is positioned behind you, so you would have to turn around to see it.", + "D": "The entrance is slightly behind you on the left, not straight ahead from where you're standing.", + "E": "The door is right in front of you." + }, + "answer_label": "E", + "answer_text": "The door is right in front of you.", + "q_id": "D3-P3-am-museum_part006_8", + "correct_answer": "The door is right in front of you.", + "original_options": { + "A": "The door is to your left.", + "B": "The door is on your right.", + "C": "The doorway is positioned behind you, so you would have to turn around to see it.", + "D": "The entrance is slightly behind you on the left, not straight ahead from where you're standing.", + "E": "The door is right in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + } +] diff --git a/QAs/D3-P3-am-museum_part007-mcq.json b/QAs/D3-P3-am-museum_part007-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7bf8683a1d9db61be590cbcd6736d41f62d7c1b6 --- /dev/null +++ b/QAs/D3-P3-am-museum_part007-mcq.json @@ -0,0 +1,990 @@ +[ + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_3", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the umbrella in?", + "answer": "The umbrella is on your right.", + "question_chinese": "伞在我的哪个方向？", + "answer_chinese": "伞在你的右边。", + "evidence": [ + { + "description": "The umbrella is on your right.", + "timestamps": { + "start": "00:01:45", + "end": "00:01:47" + } + }, + { + "description": "The umbrella is on your right.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:51", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The umbrella is on your left.", + "B": "The umbrella is in front of you.", + "C": "The umbrella is behind you rather than on your right side as you face forward.", + "D": "The umbrella is on your right.", + "E": "The umbrella is above you, so you would need to look up instead of to your right." + }, + "answer_label": "D", + "answer_text": "The umbrella is on your right.", + "q_id": "D3-P3-am-museum_part007_3", + "correct_answer": "The umbrella is on your right.", + "original_options": { + "A": "The umbrella is on your left.", + "B": "The umbrella is in front of you.", + "C": "The umbrella is behind you rather than on your right side as you face forward.", + "D": "The umbrella is on your right.", + "E": "The umbrella is above you, so you would need to look up instead of to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 17 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the umbrella in?", + "answer": "The umbrella is behind you.", + "question_chinese": "伞在我的哪个方向？", + "answer_chinese": "伞在你的后边。", + "evidence": [ + { + "description": "The umbrella is behind you.", + "timestamps": { + "start": "00:02:27", + "end": "00:02:28" + } + }, + { + "description": "The umbrella is behind you.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:02:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The umbrella is behind you.", + "B": "The umbrella is in front of you.", + "C": "The umbrella is to your left.", + "D": "The umbrella is on your right side from where you're standing, not located behind you.", + "E": "The umbrella is above your head, so it's overhead rather than behind your back." + }, + "answer_label": "A", + "answer_text": "The umbrella is behind you.", + "q_id": "D3-P3-am-museum_part007_4", + "correct_answer": "The umbrella is behind you.", + "original_options": { + "A": "The umbrella is behind you.", + "B": "The umbrella is in front of you.", + "C": "The umbrella is to your left.", + "D": "The umbrella is on your right side from where you're standing, not located behind you.", + "E": "The umbrella is above your head, so it's overhead rather than behind your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 14 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_6_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I came out, what path did I take?", + "answer": "First go to your left, then throw away the trash.", + "question_chinese": "出来后我的移动轨迹是怎样的？", + "answer_chinese": "先往左边走，然后扔垃圾。", + "evidence": [ + { + "description": "Go left first.", + "timestamps": { + "start": "00:18:28", + "end": "00:18:36" + } + }, + { + "description": "Then throw away the garbage.", + "timestamps": { + "start": "00:18:28", + "end": "00:18:36" + } + } + ], + "question_timestamps": "00:18:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First go to your right.", + "B": "First go straight ahead.", + "C": "First go to your right, then throw away the trash once you reach the bin.", + "D": "First head straight ahead out of the area, and then throw away the trash afterward.", + "E": "First go to your left, then throw away the trash." + }, + "answer_label": "E", + "answer_text": "First go to your left, then throw away the trash.", + "q_id": "D3-P3-am-museum_part007_6_T1", + "correct_answer": "First go to your left, then throw away the trash.", + "original_options": { + "A": "First go to your right.", + "B": "First go straight ahead.", + "C": "First go to your right, then throw away the trash once you reach the bin.", + "D": "First head straight ahead out of the area, and then throw away the trash afterward.", + "E": "First go to your left, then throw away the trash." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 15 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_6_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I came out, what path did I take?", + "answer": "First go to your left to throw away the trash, then go straight to the right.", + "question_chinese": "出来后我的移动轨迹是怎样的？", + "answer_chinese": "先往左边走扔垃圾，然后向右直走。", + "evidence": [ + { + "description": "Go left first to throw away the rubbish,then go straight to the right.", + "timestamps": { + "start": "00:18:28", + "end": "00:18:36" + } + } + ], + "question_timestamps": "00:18:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First go right, then right.", + "B": "First go forward, then right.", + "C": "First go to your right to throw away the trash, then continue straight to the right afterward.", + "D": "First walk straight ahead to throw away the trash, then turn left instead of going straight to the right.", + "E": "First go to your left to throw away the trash, then go straight to the right." + }, + "answer_label": "E", + "answer_text": "First go to your left to throw away the trash, then go straight to the right.", + "q_id": "D3-P3-am-museum_part007_6_T2", + "correct_answer": "First go to your left to throw away the trash, then go straight to the right.", + "original_options": { + "A": "First go right, then right.", + "B": "First go forward, then right.", + "C": "First go to your right to throw away the trash, then continue straight to the right afterward.", + "D": "First walk straight ahead to throw away the trash, then turn left instead of going straight to the right.", + "E": "First go to your left to throw away the trash, then go straight to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_5_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I came out, what path did I take?", + "answer": "First turn left to throw away the trash, then walk back.", + "question_chinese": "出来后我的移动轨迹是怎样的？", + "answer_chinese": "先左转扔垃圾，然后往回走。", + "evidence": [ + { + "description": "Turn left to throw the garbage first.", + "timestamps": { + "start": "00:18:28", + "end": "00:18:36" + } + }, + { + "description": "Then walk back.", + "timestamps": { + "start": "00:18:38", + "end": "00:18:44" + } + } + ], + "question_timestamps": "00:19:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right, then walk back.", + "B": "First turn left to throw away the trash, then walk back.", + "C": "Turn left, then walk forward.", + "D": "First turn right to throw away the trash, then retrace your steps back the way you came.", + "E": "First turn left to throw away the trash, then continue straight ahead instead of returning back." + }, + "answer_label": "B", + "answer_text": "First turn left to throw away the trash, then walk back.", + "q_id": "D3-P3-am-museum_part007_5_T1", + "correct_answer": "First turn left to throw away the trash, then walk back.", + "original_options": { + "A": "Turn right, then walk back.", + "B": "First turn left to throw away the trash, then walk back.", + "C": "Turn left, then walk forward.", + "D": "First turn right to throw away the trash, then retrace your steps back the way you came.", + "E": "First turn left to throw away the trash, then continue straight ahead instead of returning back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_5_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I came out, what path did I take?", + "answer": "First turn left to throw away the trash, then go straight to the right, and then walk back.", + "question_chinese": "出来后我的移动轨迹是怎样的？", + "answer_chinese": "先左转扔垃圾，然后向右直走，然后往回走。", + "evidence": [ + { + "description": "Turn left to throw the garbage first.", + "timestamps": { + "start": "00:18:28", + "end": "00:18:36" + } + }, + { + "description": "Then walk back.", + "timestamps": { + "start": "00:19:55", + "end": "00:20:02" + } + }, + { + "description": "Go straight to the right.", + "timestamps": { + "start": "00:18:37", + "end": "00:18:45" + } + } + ], + "question_timestamps": "00:20:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go straight ahead, then walk back.", + "B": "Go straight to the left, then return.", + "C": "First turn left to throw away the trash, then go straight to the right, and then walk back.", + "D": "First turn left to throw away the trash, then go straight in front of you, and then walk back.", + "E": "First turn left to throw away the trash, then go straight behind you, and then walk back." + }, + "answer_label": "C", + "answer_text": "First turn left to throw away the trash, then go straight to the right, and then walk back.", + "q_id": "D3-P3-am-museum_part007_5_T2", + "correct_answer": "First turn left to throw away the trash, then go straight to the right, and then walk back.", + "original_options": { + "A": "Go straight ahead, then walk back.", + "B": "Go straight to the left, then return.", + "C": "First turn left to throw away the trash, then go straight to the right, and then walk back.", + "D": "First turn left to throw away the trash, then go straight in front of you, and then walk back.", + "E": "First turn left to throw away the trash, then go straight behind you, and then walk back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 17 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_5", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my trajectory after opening the first store?", + "answer": "After you come out, turn left and then go straight.", + "question_chinese": "出来第一家店后我的移动轨迹是怎样的？", + "answer_chinese": "出来后你左转，然后直走。", + "evidence": [ + { + "description": "Turn left.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:25" + } + }, + { + "description": "Then go straight.", + "timestamps": { + "start": "00:05:25", + "end": "00:05:30" + } + } + ], + "question_timestamps": "00:05:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn left.", + "B": "After you come out, turn left and then go straight.", + "C": "Turn right, then go straight.", + "D": "After you come out, turn left and then go back the way you came instead of continuing forward.", + "E": "After you come out, go straight first and then turn right, rather than turning left." + }, + "answer_label": "B", + "answer_text": "After you come out, turn left and then go straight.", + "q_id": "D3-P3-am-museum_part007_5", + "correct_answer": "After you come out, turn left and then go straight.", + "original_options": { + "A": "Go straight, then turn left.", + "B": "After you come out, turn left and then go straight.", + "C": "Turn right, then go straight.", + "D": "After you come out, turn left and then go back the way you came instead of continuing forward.", + "E": "After you come out, go straight first and then turn right, rather than turning left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 15 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my trajectory after opening the first store?", + "answer": "After you come out, turn left, then go straight, and enter the next exhibition hall.", + "question_chinese": "出来第一家店后我的移动轨迹是怎样的？", + "answer_chinese": "出来后你左转，然后直走，然后进入下一个展馆。", + "evidence": [ + { + "description": "Turn left.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:25" + } + }, + { + "description": "Then go straight.", + "timestamps": { + "start": "00:05:25", + "end": "00:05:30" + } + }, + { + "description": "Then enter the next exhibition hall.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:15" + } + } + ], + "question_timestamps": "00:06:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right into the next hall.", + "B": "Enter the first exhibition hall.", + "C": "After you come out, turn left, then go straight, and enter the next exhibition hall.", + "D": "After you come out, turn left, then go straight, and enter the previous exhibition hall instead.", + "E": "After you come out, turn right, then go straight, and enter the previous exhibition hall, not the next." + }, + "answer_label": "C", + "answer_text": "After you come out, turn left, then go straight, and enter the next exhibition hall.", + "q_id": "D3-P3-am-museum_part007_6", + "correct_answer": "After you come out, turn left, then go straight, and enter the next exhibition hall.", + "original_options": { + "A": "Turn right into the next hall.", + "B": "Enter the first exhibition hall.", + "C": "After you come out, turn left, then go straight, and enter the next exhibition hall.", + "D": "After you come out, turn left, then go straight, and enter the previous exhibition hall instead.", + "E": "After you come out, turn right, then go straight, and enter the previous exhibition hall, not the next." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 18 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_9", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the white stool?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接坐在白色的凳子上吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The white stool is right next to you.", + "timestamps": { + "start": "00:17:02", + "end": "00:17:04" + } + }, + { + "description": "You can sit directly on it.", + "timestamps": { + "start": "00:17:02", + "end": "00:17:04" + } + } + ], + "question_timestamps": "00:17:07", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "D3-P3-am-museum_part007_9", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_10", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the white stool?", + "answer": "No,you can't.", + "question_chinese": "我可以直接坐在白色的凳子上吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The white stool is behind you.", + "timestamps": { + "start": "00:17:05", + "end": "00:17:10" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:17:05", + "end": "00:17:10" + } + }, + { + "description": "You cannot sit directly on it.", + "timestamps": { + "start": "00:17:05", + "end": "00:17:10" + } + } + ], + "question_timestamps": "00:17:13", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "D3-P3-am-museum_part007_10", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_7_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I pick up and look at?", + "answer": "You picked up and looked at one thing: a brown panda.", + "question_chinese": "我拿起来看了几个东西？", + "answer_chinese": "你拿起来看了1个东西，一个棕色熊猫。", + "evidence": [ + { + "description": "You are looking at the first one.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero items.", + "B": "Two items total.", + "C": "You picked up and looked at two things: a brown panda and one other item.", + "D": "You picked up and looked at one thing: a brown panda.", + "E": "You picked up and looked at three things: a brown panda plus two other items." + }, + "answer_label": "D", + "answer_text": "You picked up and looked at one thing: a brown panda.", + "q_id": "D3-P3-am-museum_part007_7_T2", + "correct_answer": "You picked up and looked at one thing: a brown panda.", + "original_options": { + "A": "Zero items.", + "B": "Two items total.", + "C": "You picked up and looked at two things: a brown panda and one other item.", + "D": "You picked up and looked at one thing: a brown panda.", + "E": "You picked up and looked at three things: a brown panda plus two other items." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 15 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_7_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I pick up and look at?", + "answer": "You looked at two things: a brown panda and a blue doll.", + "question_chinese": "我拿起来看了几个东西？", + "answer_chinese": "你看了两个东西。一个棕色熊猫，一个蓝色玩偶。", + "evidence": [ + { + "description": "You are looking at the first one.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "You're looking at the second one.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero items.", + "B": "One item.", + "C": "You looked at two things: a brown panda and a blue doll.", + "D": "You looked at three things: a brown panda, a blue doll, and one more item.", + "E": "You looked at four things: a brown panda, a blue doll, and two additional items." + }, + "answer_label": "C", + "answer_text": "You looked at two things: a brown panda and a blue doll.", + "q_id": "D3-P3-am-museum_part007_7_T1", + "correct_answer": "You looked at two things: a brown panda and a blue doll.", + "original_options": { + "A": "Zero items.", + "B": "One item.", + "C": "You looked at two things: a brown panda and a blue doll.", + "D": "You looked at three things: a brown panda, a blue doll, and one more item.", + "E": "You looked at four things: a brown panda, a blue doll, and two additional items." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_8_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I pick up and look at?", + "answer": "Three.", + "question_chinese": "我拿起来看了几个东西？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "You are looking at the first one.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "You're looking at the second one.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:42" + } + }, + { + "description": "You're looking at the third one.", + "timestamps": { + "start": "00:03:16", + "end": "00:03:17" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "You picked up and looked at two items in total, then set them back down afterward.", + "D": "Three.", + "E": "You picked up and looked at four separate items before putting them back in place afterward." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "D3-P3-am-museum_part007_8_T1", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "You picked up and looked at two items in total, then set them back down afterward.", + "D": "Three.", + "E": "You picked up and looked at four separate items before putting them back in place afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_8_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I pick up and look at?", + "answer": "You picked up and looked at four things.", + "question_chinese": "我拿起来看了几个东西？", + "answer_chinese": "你拿起来看了4个东西。", + "evidence": [ + { + "description": "You are looking at the first one.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + }, + { + "description": "You're looking at the second one.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:42" + } + }, + { + "description": "You're looking at the third one.", + "timestamps": { + "start": "00:03:16", + "end": "00:03:17" + } + }, + { + "description": "You're looking at the forth one.", + "timestamps": { + "start": "00:03:53", + "end": "00:03:57" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three things.", + "B": "Five things.", + "C": "You picked up and looked at two things before moving on to something else afterward.", + "D": "You picked up and looked at six things during that moment before continuing onward.", + "E": "You picked up and looked at four things." + }, + "answer_label": "E", + "answer_text": "You picked up and looked at four things.", + "q_id": "D3-P3-am-museum_part007_8_T2", + "correct_answer": "You picked up and looked at four things.", + "original_options": { + "A": "Three things.", + "B": "Five things.", + "C": "You picked up and looked at two things before moving on to something else afterward.", + "D": "You picked up and looked at six things during that moment before continuing onward.", + "E": "You picked up and looked at four things." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_1_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the gray panda or the black panda?", + "answer": "The gray panda is closer to you. The brown panda is on your right-hand side.", + "question_chinese": "灰色的熊猫和黑色的熊猫哪个离我更近？", + "answer_chinese": "灰色的熊猫离你更近。你右手边是棕色的熊猫。", + "evidence": [ + { + "description": "The gray panda is closer to you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The gray panda is closer to you. The brown panda is on your right-hand side.", + "B": "The black panda is closer to you. The brown panda is on your right-hand side." + }, + "answer_label": "A", + "answer_text": "The gray panda is closer to you. The brown panda is on your right-hand side.", + "q_id": "D3-P3-am-museum_part007_1_T1", + "correct_answer": "The gray panda is closer to you. The brown panda is on your right-hand side.", + "original_options": { + "A": "The gray panda is closer to you. The brown panda is on your right-hand side.", + "B": "The black panda is closer to you. The brown panda is on your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "D3-P3-am-museum_part007", + "video_length": "00:20:04", + "qid": "D3-P3-am-museum_part007_1_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the gray panda or the black panda?", + "answer": "The black panda is closer to you.", + "question_chinese": "灰色的熊猫和黑色的熊猫哪个离我更近？", + "answer_chinese": "黑色的熊猫离你更近。", + "evidence": [ + { + "description": "The black panda is in front of you.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + }, + { + "description": "The gray panda is on your right.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + }, + { + "description": "The black panda is closer to you.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The black panda is closer to you.", + "B": "The gray panda is closer to you. The brown panda is on your right-hand side." + }, + "answer_label": "A", + "answer_text": "The black panda is closer to you.", + "q_id": "D3-P3-am-museum_part007_1_T2", + "correct_answer": "The black panda is closer to you.", + "original_options": { + "A": "The black panda is closer to you.", + "B": "The gray panda is closer to you. The brown panda is on your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + } +] diff --git a/QAs/P01_101-mcq.json b/QAs/P01_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..47b75e04eb1773c17ab9638b70361f6add624e03 --- /dev/null +++ b/QAs/P01_101-mcq.json @@ -0,0 +1,1935 @@ +[ + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_1_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the refrigerator relative to me?", + "answer": "The fridge is in front of you.", + "question_chinese": "冰箱在我的哪个方位？", + "answer_chinese": "冰箱在你的前面。", + "evidence": [ + { + "description": "The microwave is to the right of the refrigerator.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is behind you.", + "B": "The fridge is in front of you.", + "C": "The fridge is to your left.", + "D": "The fridge is off to your right side, not directly ahead of where you are facing.", + "E": "The fridge is beside you rather than in front, positioned to the side at about the same distance." + }, + "answer_label": "B", + "answer_text": "The fridge is in front of you.", + "q_id": "P01_101_1_T1", + "correct_answer": "The fridge is in front of you.", + "original_options": { + "A": "The fridge is behind you.", + "B": "The fridge is in front of you.", + "C": "The fridge is to your left.", + "D": "The fridge is off to your right side, not directly ahead of where you are facing.", + "E": "The fridge is beside you rather than in front, positioned to the side at about the same distance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 18 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_1_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the refrigerator relative to me?", + "answer": "The fridge is to your left.", + "question_chinese": "冰箱在我的哪个方位？", + "answer_chinese": "冰箱在你的左边。", + "evidence": [ + { + "description": "The microwave is to the right of the refrigerator.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is in front of you.", + "B": "The fridge is behind you.", + "C": "The fridge is to your left.", + "D": "The fridge is to your right side, opposite from where it would be if it were on your left.", + "E": "The fridge is positioned ahead of you rather than to the left, directly along your facing direction." + }, + "answer_label": "C", + "answer_text": "The fridge is to your left.", + "q_id": "P01_101_1_T2", + "correct_answer": "The fridge is to your left.", + "original_options": { + "A": "The fridge is in front of you.", + "B": "The fridge is behind you.", + "C": "The fridge is to your left.", + "D": "The fridge is to your right side, opposite from where it would be if it were on your left.", + "E": "The fridge is positioned ahead of you rather than to the left, directly along your facing direction." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 19, + 17 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_14", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of bread are there on the colored mat on the wooden table outside?", + "answer": "Two pieces.", + "question_chinese": "外面木桌的彩色垫巾上现在有几块面包？", + "answer_chinese": "两块。", + "evidence": [ + { + "description": "You took two pieces of bread from the cutting board.", + "timestamps": { + "start": "00:24:04", + "end": "00:24:09" + } + }, + { + "description": "You placed the bread on the colorful placemat on the wooden table outside.", + "timestamps": { + "start": "00:24:10", + "end": "00:24:12" + } + } + ], + "question_timestamps": "00:24:14", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One piece.", + "B": "Three pieces.", + "C": "Two pieces.", + "D": "Four pieces of bread are on the colored mat on the wooden table outside.", + "E": "Five pieces are laid out on the colored mat outside on the wooden table." + }, + "answer_label": "C", + "answer_text": "Two pieces.", + "q_id": "P01_101_14", + "correct_answer": "Two pieces.", + "original_options": { + "A": "One piece.", + "B": "Three pieces.", + "C": "Two pieces.", + "D": "Four pieces of bread are on the colored mat on the wooden table outside.", + "E": "Five pieces are laid out on the colored mat outside on the wooden table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_15", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of bread are there on the colored mat on the wooden table outside?", + "answer": "Two pieces.", + "question_chinese": "外面木桌的彩色垫巾上现在有几块面包？", + "answer_chinese": "两块。", + "evidence": [ + { + "description": "You took two pieces of bread from the cutting board.", + "timestamps": { + "start": "00:24:04", + "end": "00:24:09" + } + }, + { + "description": "You placed the bread on the colorful mat on the outdoor wooden table.", + "timestamps": { + "start": "00:24:10", + "end": "00:24:12" + } + } + ], + "question_timestamps": "00:27:12", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One piece.", + "B": "Three pieces.", + "C": "Four pieces of bread are laid out on the colored mat on the wooden table outside.", + "D": "Two pieces.", + "E": "Six pieces of bread are resting on the colored mat on the wooden table outside." + }, + "answer_label": "D", + "answer_text": "Two pieces.", + "q_id": "P01_101_15", + "correct_answer": "Two pieces.", + "original_options": { + "A": "One piece.", + "B": "Three pieces.", + "C": "Four pieces of bread are laid out on the colored mat on the wooden table outside.", + "D": "Two pieces.", + "E": "Six pieces of bread are resting on the colored mat on the wooden table outside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 15 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where does the microwave oven sit in the refrigerator?", + "answer": "The microwave oven is on the right side of the refrigerator.", + "question_chinese": "微波炉在冰箱的哪个方位？", + "answer_chinese": "微波炉在冰箱的右边。", + "evidence": [ + { + "description": "The microwave is to the right of the refrigerator.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:01:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The microwave oven is on the right side of the refrigerator.", + "B": "On the left side.", + "C": "In the middle.", + "D": "The microwave oven sits on the left side of the refrigerator, not on the right.", + "E": "The microwave oven sits below the countertop area of the refrigerator rather than on either side." + }, + "answer_label": "A", + "answer_text": "The microwave oven is on the right side of the refrigerator.", + "q_id": "P01_101_2", + "correct_answer": "The microwave oven is on the right side of the refrigerator.", + "original_options": { + "A": "The microwave oven is on the right side of the refrigerator.", + "B": "On the left side.", + "C": "In the middle.", + "D": "The microwave oven sits on the left side of the refrigerator, not on the right.", + "E": "The microwave oven sits below the countertop area of the refrigerator rather than on either side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 16 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where does the microwave oven sit in the refrigerator?", + "answer": "The microwave oven is on the right side of the refrigerator.", + "question_chinese": "微波炉在冰箱的哪个方位？", + "answer_chinese": "微波炉在冰箱的右边。", + "evidence": [ + { + "description": "The microwave is to the right of the refrigerator.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the left side.", + "B": "Inside the freezer.", + "C": "The microwave oven is on the right side of the refrigerator.", + "D": "The microwave oven sits beneath the countertop beside the refrigerator, not on the refrigerator’s right side.", + "E": "The microwave oven is on the left shelf of the cupboard rather than positioned on the refrigerator." + }, + "answer_label": "C", + "answer_text": "The microwave oven is on the right side of the refrigerator.", + "q_id": "P01_101_1", + "correct_answer": "The microwave oven is on the right side of the refrigerator.", + "original_options": { + "A": "On the left side.", + "B": "Inside the freezer.", + "C": "The microwave oven is on the right side of the refrigerator.", + "D": "The microwave oven sits beneath the countertop beside the refrigerator, not on the refrigerator’s right side.", + "E": "The microwave oven is on the left shelf of the cupboard rather than positioned on the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 17 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_9_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many portions of vegetables did I take out of the refrigerator?", + "answer": "You took 0 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "question_chinese": "我从冰箱里拿出了几份蔬菜？", + "answer_chinese": "你从冰箱里拿出了0份蔬菜，冰箱在你前面。", + "evidence": [ + { + "description": "You took zero servings of vegetables out of the bridge.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One portion.", + "B": "Two portions.", + "C": "You took 3 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "D": "You took 0 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "E": "You took 4 portions of vegetables out of the refrigerator, and the refrigerator is in front of you." + }, + "answer_label": "D", + "answer_text": "You took 0 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "q_id": "P01_101_9_T1", + "correct_answer": "You took 0 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "original_options": { + "A": "One portion.", + "B": "Two portions.", + "C": "You took 3 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "D": "You took 0 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "E": "You took 4 portions of vegetables out of the refrigerator, and the refrigerator is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 18 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_9_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many portions of vegetables did I take out of the refrigerator?", + "answer": "You took 4 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "question_chinese": "我从冰箱里拿出了几份蔬菜？", + "answer_chinese": "你从冰箱里拿出了4份蔬菜，冰箱在你前面。", + "evidence": [ + { + "description": "You took four servings of vegetables out of the fridge.", + "timestamps": { + "start": "00:00:57", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two portions.", + "B": "Three portions.", + "C": "You took 5 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "D": "You took 6 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "E": "You took 4 portions of vegetables out of the refrigerator, and the refrigerator is in front of you." + }, + "answer_label": "E", + "answer_text": "You took 4 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "q_id": "P01_101_9_T2", + "correct_answer": "You took 4 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "original_options": { + "A": "Two portions.", + "B": "Three portions.", + "C": "You took 5 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "D": "You took 6 portions of vegetables out of the refrigerator, and the refrigerator is in front of you.", + "E": "You took 4 portions of vegetables out of the refrigerator, and the refrigerator is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 18 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_18", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many veggies did I just pulled out from the fridge?", + "answer": "four parts.", + "question_chinese": "我刚刚从冰箱里拿出了几份蔬菜？", + "answer_chinese": "4份。", + "evidence": [ + { + "description": "Take an eggplant out of the refrigerator.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:09" + } + }, + { + "description": "Take a chili pepper out of the refrigerator.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:30" + } + }, + { + "description": "Take a carrot out of the refrigerator.", + "timestamps": { + "start": "00:00:57", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two parts.", + "B": "four parts.", + "C": "Three parts.", + "D": "Five parts, meaning you pulled out five veggie portions from the fridge just now.", + "E": "Six parts, so you removed six separate veggie portions from the fridge." + }, + "answer_label": "B", + "answer_text": "four parts.", + "q_id": "P01_101_18", + "correct_answer": "four parts.", + "original_options": { + "A": "Two parts.", + "B": "four parts.", + "C": "Three parts.", + "D": "Five parts, meaning you pulled out five veggie portions from the fridge just now.", + "E": "Six parts, so you removed six separate veggie portions from the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 12 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_19", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many veggies did I just pulled out from the fridge?", + "answer": "four parts.", + "question_chinese": "我刚刚从冰箱里拿出了几份蔬菜？", + "answer_chinese": "4份。", + "evidence": [ + { + "description": "Take an eggplant out of the refrigerator.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:09" + } + }, + { + "description": "Take a chili pepper out of the refrigerator.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:30" + } + }, + { + "description": "Take a carrot out of the refrigerator.", + "timestamps": { + "start": "00:00:57", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:04:42", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three parts.", + "B": "Five parts.", + "C": "Six parts, meaning you pulled out six veggie portions from the fridge just now.", + "D": "four parts.", + "E": "Seven parts, so you removed seven separate veggie portions from the fridge." + }, + "answer_label": "D", + "answer_text": "four parts.", + "q_id": "P01_101_19", + "correct_answer": "four parts.", + "original_options": { + "A": "Three parts.", + "B": "Five parts.", + "C": "Six parts, meaning you pulled out six veggie portions from the fridge just now.", + "D": "four parts.", + "E": "Seven parts, so you removed seven separate veggie portions from the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 12 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_3_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the white steamer basket with green vegetables?", + "answer": "The white steamer with green vegetables was placed next to the rice cooker, and the rice cooker is to your left.", + "question_chinese": "我刚刚把装有绿色蔬菜的白色蒸笼放在哪儿了？", + "answer_chinese": "装有绿色蔬菜的白色蒸笼被放在电饭煲旁边，电饭煲在你左边。", + "evidence": [ + { + "description": "A white steamer filled with green vegetables has been placed next to the rice cooker.", + "timestamps": { + "start": "00:12:35", + "end": "00:12:42" + } + } + ], + "question_timestamps": "00:12:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Next to the rice cooker in front.", + "B": "Next to the rice cooker behind you.", + "C": "The white steamer with green vegetables was placed next to the rice cooker, and the rice cooker is to your right.", + "D": "The white steamer with green vegetables was placed next to the rice cooker, and the rice cooker is to your left.", + "E": "The white steamer with green vegetables was placed next to the rice cooker, and the rice cooker is behind you." + }, + "answer_label": "D", + "answer_text": "The white steamer with green vegetables was placed next to the rice cooker, and the rice cooker is to your left.", + "q_id": "P01_101_3_T1", + "correct_answer": "The white steamer with green vegetables was placed next to the rice cooker, and the rice cooker is to your left.", + "original_options": { + "A": "Next to the rice cooker in front.", + "B": "Next to the rice cooker behind you.", + "C": "The white steamer with green vegetables was placed next to the rice cooker, and the rice cooker is to your right.", + "D": "The white steamer with green vegetables was placed next to the rice cooker, and the rice cooker is to your left.", + "E": "The white steamer with green vegetables was placed next to the rice cooker, and the rice cooker is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 20 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_6", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the white steamer with the green vegetables?", + "answer": "A white steamer with green vegetables is placed next to the rice cooker.", + "question_chinese": "我刚刚把装有绿色蔬菜的白色蒸笼放在哪儿了？", + "answer_chinese": "装有绿色蔬菜的白色蒸笼被放在电饭煲里面了。", + "evidence": [ + { + "description": "A white steamer filled with green vegetables is placed next to the rice cooker.", + "timestamps": { + "start": "00:12:35", + "end": "00:12:42" + } + }, + { + "description": "A white steamer filled with green vegetables is placed inside the rice cooker.", + "timestamps": { + "start": "00:12:53", + "end": "00:12:59" + } + } + ], + "question_timestamps": "00:13:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A white steamer with green vegetables is placed next to the rice cooker.", + "B": "On the counter by the toaster.", + "C": "On the stove near the pan.", + "D": "You left the white steamer with green vegetables on the kitchen island, beside the cutting board.", + "E": "You set the white steamer with green vegetables on the shelf above the microwave, not near the rice cooker." + }, + "answer_label": "A", + "answer_text": "A white steamer with green vegetables is placed next to the rice cooker.", + "q_id": "P01_101_6", + "correct_answer": "A white steamer with green vegetables is placed next to the rice cooker.", + "original_options": { + "A": "A white steamer with green vegetables is placed next to the rice cooker.", + "B": "On the counter by the toaster.", + "C": "On the stove near the pan.", + "D": "You left the white steamer with green vegetables on the kitchen island, beside the cutting board.", + "E": "You set the white steamer with green vegetables on the shelf above the microwave, not near the rice cooker." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 19 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_3_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the white steamer basket with green vegetables?", + "answer": "The white steamer with green vegetables was put inside the rice cooker, and the rice cooker is in front of you.", + "question_chinese": "我刚刚把装有绿色蔬菜的白色蒸笼放在哪儿了？", + "answer_chinese": "装有绿色蔬菜的白色蒸笼被放在电饭煲里面了，电饭煲在你前面。", + "evidence": [ + { + "description": "A white steamer filled with green vegetables is placed inside the rice cooker.", + "timestamps": { + "start": "00:12:53", + "end": "00:12:59" + } + } + ], + "question_timestamps": "00:13:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Inside the rice cooker behind you.", + "B": "Inside the rice cooker to your left.", + "C": "You put the white steamer with green vegetables inside the rice cooker, and the rice cooker is to your right.", + "D": "The white steamer with green vegetables was put inside the rice cooker, and the rice cooker is in front of you.", + "E": "The white steamer basket is inside the rice cooker, but the rice cooker is positioned above you." + }, + "answer_label": "D", + "answer_text": "The white steamer with green vegetables was put inside the rice cooker, and the rice cooker is in front of you.", + "q_id": "P01_101_3_T2", + "correct_answer": "The white steamer with green vegetables was put inside the rice cooker, and the rice cooker is in front of you.", + "original_options": { + "A": "Inside the rice cooker behind you.", + "B": "Inside the rice cooker to your left.", + "C": "You put the white steamer with green vegetables inside the rice cooker, and the rice cooker is to your right.", + "D": "The white steamer with green vegetables was put inside the rice cooker, and the rice cooker is in front of you.", + "E": "The white steamer basket is inside the rice cooker, but the rice cooker is positioned above you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 17 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_3", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the stove when I first entered the kitchen?", + "answer": "When you first enter the kitchen, the stove is on your right.", + "question_chinese": "我刚进厨房时，灶台在我的哪个方位？", + "answer_chinese": "刚进厨房时，灶台在你的右边。", + "evidence": [ + { + "description": "As soon as I entered the kitchen, the stove was on the right side of my view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:04:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The stove was on your left.", + "B": "When you first enter the kitchen, the stove is on your right.", + "C": "The stove was straight ahead.", + "D": "When you first entered the kitchen, the stove was behind you, so you had to turn around to see it.", + "E": "As you entered the kitchen, the stove was off to your back-left, not on your right side." + }, + "answer_label": "B", + "answer_text": "When you first enter the kitchen, the stove is on your right.", + "q_id": "P01_101_3", + "correct_answer": "When you first enter the kitchen, the stove is on your right.", + "original_options": { + "A": "The stove was on your left.", + "B": "When you first enter the kitchen, the stove is on your right.", + "C": "The stove was straight ahead.", + "D": "When you first entered the kitchen, the stove was behind you, so you had to turn around to see it.", + "E": "As you entered the kitchen, the stove was off to your back-left, not on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 17 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_4", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was the stove when I first entered the kitchen?", + "answer": "When you first enter the kitchen, the stove is on your right.", + "question_chinese": "我刚进厨房时，灶台在我的哪个方位？", + "answer_chinese": "刚进厨房时，灶台在你的右边。", + "evidence": [ + { + "description": "When I first entered the kitchen, the stove was on the right side of my sight.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:07:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The stove was straight ahead.", + "B": "The stove was on your left.", + "C": "When you first enter the kitchen, the stove is behind you on the right, not beside you.", + "D": "When you first entered the kitchen, the stove was to your left rear, so it wasn’t on your right.", + "E": "When you first enter the kitchen, the stove is on your right." + }, + "answer_label": "E", + "answer_text": "When you first enter the kitchen, the stove is on your right.", + "q_id": "P01_101_4", + "correct_answer": "When you first enter the kitchen, the stove is on your right.", + "original_options": { + "A": "The stove was straight ahead.", + "B": "The stove was on your left.", + "C": "When you first enter the kitchen, the stove is behind you on the right, not beside you.", + "D": "When you first entered the kitchen, the stove was to your left rear, so it wasn’t on your right.", + "E": "When you first enter the kitchen, the stove is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 20 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_4_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many blue lids am I holding in my hand?", + "answer": "You are holding 2 blue lids in your hand, and the lids are in front of you.", + "question_chinese": "我手上拿着几个蓝色盖子？", + "answer_chinese": "你手上拿着2个蓝色盖子，盖子在你前面。", + "evidence": [ + { + "description": "You are holding two blue lids in your hand.", + "timestamps": { + "start": "00:17:19", + "end": "00:17:21" + } + } + ], + "question_timestamps": "00:17:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You are holding 1 blue lid.", + "B": "You are holding 3 blue lids.", + "C": "You are holding 4 blue lids in your hand, and the lids are in front of you.", + "D": "You are holding 0 blue lids in your hand, and your hands are empty in front of you.", + "E": "You are holding 2 blue lids in your hand, and the lids are in front of you." + }, + "answer_label": "E", + "answer_text": "You are holding 2 blue lids in your hand, and the lids are in front of you.", + "q_id": "P01_101_4_T2", + "correct_answer": "You are holding 2 blue lids in your hand, and the lids are in front of you.", + "original_options": { + "A": "You are holding 1 blue lid.", + "B": "You are holding 3 blue lids.", + "C": "You are holding 4 blue lids in your hand, and the lids are in front of you.", + "D": "You are holding 0 blue lids in your hand, and your hands are empty in front of you.", + "E": "You are holding 2 blue lids in your hand, and the lids are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_4_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many blue lids am I holding in my hand?", + "answer": "You are holding 1 blue lid in your hand, and the lid is in front of you.", + "question_chinese": "我手上拿着几个蓝色盖子？", + "answer_chinese": "你手上拿着1个蓝色盖子,盖子在你前面。", + "evidence": [ + { + "description": "You are holding one blue lids in your hand.", + "timestamps": { + "start": "00:17:23", + "end": "00:17:25" + } + } + ], + "question_timestamps": "00:17:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You are holding 0 blue lids.", + "B": "You are holding 2 blue lids.", + "C": "You are holding 3 blue lids in your hand, and the lids are in front of you.", + "D": "You are holding 1 blue lid in your hand, and the lid is in front of you.", + "E": "You are holding 4 blue lids in your hand, and the lids are in front of you." + }, + "answer_label": "D", + "answer_text": "You are holding 1 blue lid in your hand, and the lid is in front of you.", + "q_id": "P01_101_4_T1", + "correct_answer": "You are holding 1 blue lid in your hand, and the lid is in front of you.", + "original_options": { + "A": "You are holding 0 blue lids.", + "B": "You are holding 2 blue lids.", + "C": "You are holding 3 blue lids in your hand, and the lids are in front of you.", + "D": "You are holding 1 blue lid in your hand, and the lid is in front of you.", + "E": "You are holding 4 blue lids in your hand, and the lids are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_20", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the microwave now?", + "answer": "Yes you can.", + "question_chinese": "我现在可以够得到微波炉吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "You are taking something out from under the microwave in the cabinet.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:03:44", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No you cannot.", + "B": "Yes you can." + }, + "answer_label": "B", + "answer_text": "Yes you can.", + "q_id": "P01_101_20", + "correct_answer": "Yes you can.", + "original_options": { + "A": "No you cannot.", + "B": "Yes you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_10_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the microwave right now?", + "answer": "You can reach the microwave right now; it is to your right.", + "question_chinese": "我现在可以够得到微波炉吗？", + "answer_chinese": "你现在可以够得到微波炉,微波炉在你右边。", + "evidence": [ + { + "description": "You are taking something out from under the microwave in the cabinet.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:03:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the microwave right now; it is to your right.", + "B": "You cannot reach the microwave right now; it is to your right." + }, + "answer_label": "A", + "answer_text": "You can reach the microwave right now; it is to your right.", + "q_id": "P01_101_10_T1", + "correct_answer": "You can reach the microwave right now; it is to your right.", + "original_options": { + "A": "You can reach the microwave right now; it is to your right.", + "B": "You cannot reach the microwave right now; it is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_10_T2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the microwave right now?", + "answer": "You cannot reach the microwave right now; it is to your left.", + "question_chinese": "我现在可以够得到微波炉吗？", + "answer_chinese": "你现在不可以够得到微波炉，微波炉在你左边。", + "evidence": [ + { + "description": "You are taking something out from under the microwave in the cabinet.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:03:46", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the microwave right now; it is in front of you.", + "B": "You cannot reach the microwave right now; it is to your left." + }, + "answer_label": "B", + "answer_text": "You cannot reach the microwave right now; it is to your left.", + "q_id": "P01_101_10_T2", + "correct_answer": "You cannot reach the microwave right now; it is to your left.", + "original_options": { + "A": "You can reach the microwave right now; it is in front of you.", + "B": "You cannot reach the microwave right now; it is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_6_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mats are on the shelf in front of me?", + "answer": "There are two mats on the shelf in front of you.", + "question_chinese": "我面前的架子上有几块垫子？", + "answer_chinese": "你面前的架子上有两块垫子。", + "evidence": [ + { + "description": "There are ywo mats on the shelf in front of you.", + "timestamps": { + "start": "00:28:20", + "end": "00:28:26" + } + } + ], + "question_timestamps": "00:28:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are two mats on the shelf in front of you.", + "B": "There is one mat.", + "C": "There are three mats.", + "D": "There are four mats on the shelf in front of you right now.", + "E": "There are zero mats on the shelf in front of you at the moment." + }, + "answer_label": "A", + "answer_text": "There are two mats on the shelf in front of you.", + "q_id": "P01_101_6_T1", + "correct_answer": "There are two mats on the shelf in front of you.", + "original_options": { + "A": "There are two mats on the shelf in front of you.", + "B": "There is one mat.", + "C": "There are three mats.", + "D": "There are four mats on the shelf in front of you right now.", + "E": "There are zero mats on the shelf in front of you at the moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_6_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mats are on the shelf in front of me?", + "answer": "There is a mat on the shelf in front of you.", + "question_chinese": "我面前的架子上有几块垫子？", + "answer_chinese": "你面前的架子上有一块垫子。", + "evidence": [ + { + "description": "There are two mats on the shelf in front of you.", + "timestamps": { + "start": "00:28:38", + "end": "00:28:40" + } + } + ], + "question_timestamps": "00:28:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are zero mats.", + "B": "There is a mat on the shelf in front of you.", + "C": "There are two mats.", + "D": "There are three mats on the shelf in front of you right now.", + "E": "There are four mats on the shelf in front of you at the moment." + }, + "answer_label": "B", + "answer_text": "There is a mat on the shelf in front of you.", + "q_id": "P01_101_6_T2", + "correct_answer": "There is a mat on the shelf in front of you.", + "original_options": { + "A": "There are zero mats.", + "B": "There is a mat on the shelf in front of you.", + "C": "There are two mats.", + "D": "There are three mats on the shelf in front of you right now.", + "E": "There are four mats on the shelf in front of you at the moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_8_T1", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the white plate on the tabletop relative to me right now?", + "answer": "The white plate on the tabletop is now to your front-right.", + "question_chinese": "桌面上的白色盘子现在在我的哪个方位？", + "answer_chinese": "桌面上的白色盘子现在在你右前方。", + "evidence": [ + { + "description": "The white plate on the table is on your right front.", + "timestamps": { + "start": "00:20:39", + "end": "00:21:00" + } + } + ], + "question_timestamps": "00:21:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "The white plate on the tabletop is now directly in front of you.", + "D": "The white plate on the tabletop is now to your front-right.", + "E": "The white plate on the tabletop is now off to your right side, not in front." + }, + "answer_label": "D", + "answer_text": "The white plate on the tabletop is now to your front-right.", + "q_id": "P01_101_8_T1", + "correct_answer": "The white plate on the tabletop is now to your front-right.", + "original_options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "The white plate on the tabletop is now directly in front of you.", + "D": "The white plate on the tabletop is now to your front-right.", + "E": "The white plate on the tabletop is now off to your right side, not in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 16 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_16", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the white plate on my desk now?", + "answer": "On the table right behind you.", + "question_chinese": "桌面上的白色盘子现在在我的哪个方位？", + "answer_chinese": "在你正后方的桌面上。", + "evidence": [ + { + "description": "The white plate on the table is across from the stove.", + "timestamps": { + "start": "00:20:39", + "end": "00:21:15" + } + } + ], + "question_timestamps": "00:21:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the desk to your right.", + "B": "On the table in front of you.", + "C": "On the table right behind you.", + "D": "On the shelf above your desk, positioned over your workspace area right now.", + "E": "Inside the drawer of your desk, not on the table behind you right now." + }, + "answer_label": "C", + "answer_text": "On the table right behind you.", + "q_id": "P01_101_16", + "correct_answer": "On the table right behind you.", + "original_options": { + "A": "On the desk to your right.", + "B": "On the table in front of you.", + "C": "On the table right behind you.", + "D": "On the shelf above your desk, positioned over your workspace area right now.", + "E": "Inside the drawer of your desk, not on the table behind you right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 14 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_8_T2", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the white plate on the tabletop relative to me right now?", + "answer": "The white plate on the tabletop is now behind you.", + "question_chinese": "桌面上的白色盘子现在在我的哪个方位？", + "answer_chinese": "桌面上的白色盘子现在在你后面。", + "evidence": [ + { + "description": "The white plate on the table is behind of you.", + "timestamps": { + "start": "00:21:14", + "end": "00:21:40" + } + } + ], + "question_timestamps": "00:21:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The white plate on the tabletop is now to your front-right side.", + "D": "The white plate on the tabletop is now slightly to your right, not behind you.", + "E": "The white plate on the tabletop is now behind you." + }, + "answer_label": "E", + "answer_text": "The white plate on the tabletop is now behind you.", + "q_id": "P01_101_8_T2", + "correct_answer": "The white plate on the tabletop is now behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The white plate on the tabletop is now to your front-right side.", + "D": "The white plate on the tabletop is now slightly to your right, not behind you.", + "E": "The white plate on the tabletop is now behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 15 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_17", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the white plate on my desk now?", + "answer": "On the table right behind you.", + "question_chinese": "桌面上的白色盘子现在在我的哪个方位？", + "answer_chinese": "在你正后方的桌面上。", + "evidence": [ + { + "description": "The white plate on the table is opposite the stove.", + "timestamps": { + "start": "00:20:39", + "end": "00:21:15" + } + } + ], + "question_timestamps": "00:22:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the chair in front of you.", + "B": "On the table right behind you.", + "C": "On the shelf above your laptop.", + "D": "On the counter beside the window, away from the table behind you.", + "E": "Under the desk, with the plate placed below your workspace instead of behind you." + }, + "answer_label": "B", + "answer_text": "On the table right behind you.", + "q_id": "P01_101_17", + "correct_answer": "On the table right behind you.", + "original_options": { + "A": "On the chair in front of you.", + "B": "On the table right behind you.", + "C": "On the shelf above your laptop.", + "D": "On the counter beside the window, away from the table behind you.", + "E": "Under the desk, with the plate placed below your workspace instead of behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 14 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_7_T1", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many slices of bread have I cut?", + "answer": "At this moment, you have cut 0 slices of bread, and the bread is in front of you.", + "question_chinese": "此时我切了几块面包？", + "answer_chinese": "此时你切了0块面包，面包在你前面。", + "evidence": [ + { + "description": "You have cut zero pieces of brand right now.", + "timestamps": { + "start": "00:23:10", + "end": "00:23:13" + } + } + ], + "question_timestamps": "00:23:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have cut 1 slice.", + "B": "At this moment, you have cut 0 slices of bread, and the bread is in front of you.", + "C": "You have cut 2 slices.", + "D": "At this moment, you have cut 3 slices of bread, and the bread is in front of you.", + "E": "At this moment, you have cut 4 slices of bread, and the bread is in front of you." + }, + "answer_label": "B", + "answer_text": "At this moment, you have cut 0 slices of bread, and the bread is in front of you.", + "q_id": "P01_101_7_T1", + "correct_answer": "At this moment, you have cut 0 slices of bread, and the bread is in front of you.", + "original_options": { + "A": "You have cut 1 slice.", + "B": "At this moment, you have cut 0 slices of bread, and the bread is in front of you.", + "C": "You have cut 2 slices.", + "D": "At this moment, you have cut 3 slices of bread, and the bread is in front of you.", + "E": "At this moment, you have cut 4 slices of bread, and the bread is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_7_T2", + "qid_num": 28, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many slices of bread have I cut?", + "answer": "At this moment, you have cut 2 slices of bread, and the bread is in front of you.", + "question_chinese": "此时我切了几块面包？", + "answer_chinese": "此时你切了2块面包，面包在你前面。", + "evidence": [ + { + "description": "You have cut two pieces of brand right now.", + "timestamps": { + "start": "00:23:21", + "end": "00:23:24" + } + } + ], + "question_timestamps": "00:23:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have cut 1 slice.", + "B": "You have cut 3 slices.", + "C": "At this moment, you have cut 2 slices of bread, and the bread is in front of you.", + "D": "At this moment, you have cut 4 slices of bread, and the bread is in front of you.", + "E": "At this moment, you have cut 0 slices of bread, and the bread is in front of you." + }, + "answer_label": "C", + "answer_text": "At this moment, you have cut 2 slices of bread, and the bread is in front of you.", + "q_id": "P01_101_7_T2", + "correct_answer": "At this moment, you have cut 2 slices of bread, and the bread is in front of you.", + "original_options": { + "A": "You have cut 1 slice.", + "B": "You have cut 3 slices.", + "C": "At this moment, you have cut 2 slices of bread, and the bread is in front of you.", + "D": "At this moment, you have cut 4 slices of bread, and the bread is in front of you.", + "E": "At this moment, you have cut 0 slices of bread, and the bread is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_2_T1", + "qid_num": 29, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the stove relative to me?", + "answer": "The stove is to your right.", + "question_chinese": "灶台在我的哪个方位？", + "answer_chinese": "灶台在你的右边。", + "evidence": [ + { + "description": "As soon as I entered the kitchen, the stove was on the right side of my view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stove is in front of you.", + "B": "The stove is behind you.", + "C": "The stove is to your right.", + "D": "The stove is to your left, not on your right side where you would normally reach for it.", + "E": "The stove is directly in front of you, so you would face it without turning to either side." + }, + "answer_label": "C", + "answer_text": "The stove is to your right.", + "q_id": "P01_101_2_T1", + "correct_answer": "The stove is to your right.", + "original_options": { + "A": "The stove is in front of you.", + "B": "The stove is behind you.", + "C": "The stove is to your right.", + "D": "The stove is to your left, not on your right side where you would normally reach for it.", + "E": "The stove is directly in front of you, so you would face it without turning to either side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 18 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_2_T2", + "qid_num": 30, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the stove relative to me?", + "answer": "The stove is behind you.", + "question_chinese": "灶台在我的哪个方位？", + "answer_chinese": "灶台在你的后面。", + "evidence": [ + { + "description": "the stove was behind of you.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stove is in front of you.", + "B": "The stove is to your left.", + "C": "The stove is behind you.", + "D": "The stove is to your right, beside you rather than directly behind your back.", + "E": "The stove is off to the side, so you would need to turn slightly to face it." + }, + "answer_label": "C", + "answer_text": "The stove is behind you.", + "q_id": "P01_101_2_T2", + "correct_answer": "The stove is behind you.", + "original_options": { + "A": "The stove is in front of you.", + "B": "The stove is to your left.", + "C": "The stove is behind you.", + "D": "The stove is to your right, beside you rather than directly behind your back.", + "E": "The stove is off to the side, so you would need to turn slightly to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 17 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_12", + "qid_num": 31, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cushions are left on the shelf of the electric rice cooker?", + "answer": "There's still one left.", + "question_chinese": "现在放电饭煲的架子上还剩几块垫巾？", + "answer_chinese": "还剩一块。", + "evidence": [ + { + "description": "There are still two mats left on the shelf of the rice cooker.", + "timestamps": { + "start": "00:28:20", + "end": "00:28:35" + } + }, + { + "description": "Took away a mat.", + "timestamps": { + "start": "00:28:36", + "end": "00:28:47" + } + } + ], + "question_timestamps": "00:28:49", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There's still one left.", + "B": "There are still two left.", + "C": "There are still three left.", + "D": "There are still four left on the shelf of the electric rice cooker.", + "E": "There are still five left on the shelf of the electric rice cooker." + }, + "answer_label": "A", + "answer_text": "There's still one left.", + "q_id": "P01_101_12", + "correct_answer": "There's still one left.", + "original_options": { + "A": "There's still one left.", + "B": "There are still two left.", + "C": "There are still three left.", + "D": "There are still four left on the shelf of the electric rice cooker.", + "E": "There are still five left on the shelf of the electric rice cooker." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_13", + "qid_num": 32, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cushions are left on the shelf of the electric rice cooker?", + "answer": "There's still one left.", + "question_chinese": "现在放电饭煲的架子上还剩几块垫巾？", + "answer_chinese": "还剩一块。", + "evidence": [ + { + "description": "There are still two pot holders left on the shelf of the rice cooker.", + "timestamps": { + "start": "00:28:20", + "end": "00:28:35" + } + }, + { + "description": "Took away a piece of mat.", + "timestamps": { + "start": "00:28:36", + "end": "00:28:47" + } + } + ], + "question_timestamps": "00:29:45", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are still two left.", + "B": "There are still three left.", + "C": "There are still four left on the shelf of the electric rice cooker.", + "D": "There are still five left on the shelf of the electric rice cooker.", + "E": "There's still one left." + }, + "answer_label": "E", + "answer_text": "There's still one left.", + "q_id": "P01_101_13", + "correct_answer": "There's still one left.", + "original_options": { + "A": "There are still two left.", + "B": "There are still three left.", + "C": "There are still four left on the shelf of the electric rice cooker.", + "D": "There are still five left on the shelf of the electric rice cooker.", + "E": "There's still one left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_9", + "qid_num": 33, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many knives are in the sink now?", + "answer": "There are three knives in the sink now.", + "question_chinese": "现在洗碗池里面有几把小刀？", + "answer_chinese": "现在洗碗池一共有3把小刀。", + "evidence": [ + { + "description": "Carrying the cutting board towards the sink, there are two small knives on the cutting board.", + "timestamps": { + "start": "00:24:57", + "end": "00:25:01" + } + }, + { + "description": "There is already a small knife in the sink.", + "timestamps": { + "start": "00:25:01", + "end": "00:25:03" + } + } + ], + "question_timestamps": "00:25:31", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are four knives.", + "B": "There are three knives in the sink now.", + "C": "There are five knives.", + "D": "There are six knives in the sink now, clustered together where you left them.", + "E": "There are seven knives in the sink now, more than the three you would expect to see." + }, + "answer_label": "B", + "answer_text": "There are three knives in the sink now.", + "q_id": "P01_101_9", + "correct_answer": "There are three knives in the sink now.", + "original_options": { + "A": "There are four knives.", + "B": "There are three knives in the sink now.", + "C": "There are five knives.", + "D": "There are six knives in the sink now, clustered together where you left them.", + "E": "There are seven knives in the sink now, more than the three you would expect to see." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 17 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_10", + "qid_num": 34, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many knives are in the sink now?", + "answer": "There are one knive in the sink now.", + "question_chinese": "现在洗碗池里面有几把小刀？", + "answer_chinese": "现在洗碗池一共有1把小刀。", + "evidence": [ + { + "description": "There are a total of 3 small knives in the sink.", + "timestamps": { + "start": "00:25:01", + "end": "00:25:37" + } + }, + { + "description": "A small knife was placed on the drying rack outside the sink.", + "timestamps": { + "start": "00:25:40", + "end": "00:25:51" + } + }, + { + "description": "Another small knife was placed on the drying rack outside the sink.", + "timestamps": { + "start": "00:25:53", + "end": "00:25:55" + } + } + ], + "question_timestamps": "00:25:56", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are no knives.", + "B": "There are two knives.", + "C": "There are one knive in the sink now.", + "D": "There are three knives in the sink now, not just a single one.", + "E": "There are four knives in the sink now, which is more than one knife." + }, + "answer_label": "C", + "answer_text": "There are one knive in the sink now.", + "q_id": "P01_101_10", + "correct_answer": "There are one knive in the sink now.", + "original_options": { + "A": "There are no knives.", + "B": "There are two knives.", + "C": "There are one knive in the sink now.", + "D": "There are three knives in the sink now, not just a single one.", + "E": "There are four knives in the sink now, which is more than one knife." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "P01_101", + "video_length": "00:30:31", + "qid": "P01_101_11", + "qid_num": 35, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many knives are in the sink now?", + "answer": "There are no knives in the sink now.", + "question_chinese": "现在洗碗池里面有几把小刀？", + "answer_chinese": "现在洗碗池一共有0把小刀。", + "evidence": [ + { + "description": "There are a total of 3 small knives in the sink.", + "timestamps": { + "start": "00:25:01", + "end": "00:25:37" + } + }, + { + "description": "A small knife was placed on the drying rack outside the sink.", + "timestamps": { + "start": "00:25:40", + "end": "00:25:51" + } + }, + { + "description": "Another small knife was also placed on the draining rack outside the sink.", + "timestamps": { + "start": "00:25:53", + "end": "00:25:55" + } + }, + { + "description": "The last small knife was also placed on the drying rack outside the sink.", + "timestamps": { + "start": "00:25:58", + "end": "00:26:00" + } + } + ], + "question_timestamps": "00:26:01", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There is one knife.", + "B": "There are two knives.", + "C": "There are three knives in the sink now, instead of the sink being completely empty of knives.", + "D": "There are no knives in the sink now.", + "E": "There are four knives in the sink now, so you can see multiple knives sitting there." + }, + "answer_label": "D", + "answer_text": "There are no knives in the sink now.", + "q_id": "P01_101_11", + "correct_answer": "There are no knives in the sink now.", + "original_options": { + "A": "There is one knife.", + "B": "There are two knives.", + "C": "There are three knives in the sink now, instead of the sink being completely empty of knives.", + "D": "There are no knives in the sink now.", + "E": "There are four knives in the sink now, so you can see multiple knives sitting there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + } +] diff --git a/QAs/P01_109-mcq.json b/QAs/P01_109-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..df3311fb9f3e88a4a15f8245e810aa56c38e0243 --- /dev/null +++ b/QAs/P01_109-mcq.json @@ -0,0 +1,2046 @@ +[ + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_5_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the microwave relative to me?", + "answer": "The microwave is to your left.", + "question_chinese": "微波炉在我的哪个方位？", + "answer_chinese": "微波炉在你的左边。", + "evidence": [ + { + "description": "The microwave oven is on your left.", + "timestamps": { + "start": "00:23:22", + "end": "00:23:24" + } + } + ], + "question_timestamps": "00:23:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The microwave is in front of you.", + "B": "The microwave is to your right.", + "C": "The microwave is behind you, so you would need to turn around to face it.", + "D": "The microwave is to your left.", + "E": "The microwave is above you, positioned higher than your eye level where you are standing." + }, + "answer_label": "D", + "answer_text": "The microwave is to your left.", + "q_id": "P01_109_5_T1", + "correct_answer": "The microwave is to your left.", + "original_options": { + "A": "The microwave is in front of you.", + "B": "The microwave is to your right.", + "C": "The microwave is behind you, so you would need to turn around to face it.", + "D": "The microwave is to your left.", + "E": "The microwave is above you, positioned higher than your eye level where you are standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 15 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_5_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the microwave relative to me?", + "answer": "The microwave is behind you.", + "question_chinese": "微波炉在我的哪个方位？", + "answer_chinese": "微波炉在你的后面。", + "evidence": [ + { + "description": "The microwave oven is behind you.", + "timestamps": { + "start": "00:24:04", + "end": "00:24:14" + } + } + ], + "question_timestamps": "00:24:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The microwave is behind you.", + "B": "The microwave is in front of you.", + "C": "The microwave is to your left.", + "D": "The microwave is to your right, off to the side as you look ahead.", + "E": "The microwave is above you, sitting higher up rather than at your back." + }, + "answer_label": "A", + "answer_text": "The microwave is behind you.", + "q_id": "P01_109_5_T2", + "correct_answer": "The microwave is behind you.", + "original_options": { + "A": "The microwave is behind you.", + "B": "The microwave is in front of you.", + "C": "The microwave is to your left.", + "D": "The microwave is to your right, off to the side as you look ahead.", + "E": "The microwave is above you, sitting higher up rather than at your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 13 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_11", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the microwave in relation to the stove?", + "answer": "The microwave is in front and to the left of the stove.", + "question_chinese": "微波炉在灶台的哪个方位？", + "answer_chinese": "微波炉在灶台的左前方。", + "evidence": [ + { + "description": "The microwave is to the left of the cutting board.", + "timestamps": { + "start": "00:24:44", + "end": "00:21:45" + } + }, + { + "description": "The cutting board is directly opposite the stove.", + "timestamps": { + "start": "00:26:45", + "end": "00:26:53" + } + } + ], + "question_timestamps": "00:28:02", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It is behind the stove.", + "B": "The microwave is in front and to the left of the stove.", + "C": "It is to the right of the stove.", + "D": "The microwave is in front and to the right of the stove, shifted toward its right side.", + "E": "The microwave is to the left and behind the stove, set back rather than forward of it." + }, + "answer_label": "B", + "answer_text": "The microwave is in front and to the left of the stove.", + "q_id": "P01_109_11", + "correct_answer": "The microwave is in front and to the left of the stove.", + "original_options": { + "A": "It is behind the stove.", + "B": "The microwave is in front and to the left of the stove.", + "C": "It is to the right of the stove.", + "D": "The microwave is in front and to the right of the stove, shifted toward its right side.", + "E": "The microwave is to the left and behind the stove, set back rather than forward of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 17, + 17 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_12", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the microwave in relation to the stove?", + "answer": "The microwave is in front and to the left of the stove.", + "question_chinese": "微波炉在灶台的哪个方位？", + "answer_chinese": "微波炉在灶台的左前方。", + "evidence": [ + { + "description": "The microwave is to the left of the cutting board.", + "timestamps": { + "start": "00:24:44", + "end": "00:21:45" + } + }, + { + "description": "The cutting board is directly opposite the stove.", + "timestamps": { + "start": "00:26:45", + "end": "00:26:53" + } + } + ], + "question_timestamps": "00:41:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The microwave is in front and to the left of the stove.", + "B": "It is behind the stove.", + "C": "It is to the right of the stove.", + "D": "The microwave is behind and to the left of the stove, positioned back from it rather than ahead.", + "E": "The microwave is in front and to the right of the stove, on the opposite side from the left." + }, + "answer_label": "A", + "answer_text": "The microwave is in front and to the left of the stove.", + "q_id": "P01_109_12", + "correct_answer": "The microwave is in front and to the left of the stove.", + "original_options": { + "A": "The microwave is in front and to the left of the stove.", + "B": "It is behind the stove.", + "C": "It is to the right of the stove.", + "D": "The microwave is behind and to the left of the stove, positioned back from it rather than ahead.", + "E": "The microwave is in front and to the right of the stove, on the opposite side from the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 18, + 19 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I cut in total?", + "answer": "Three.", + "question_chinese": "我一共切了几个土豆？", + "answer_chinese": "3个。", + "evidence": [ + { + "description": "Cut the 1st potato.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:59" + } + }, + { + "description": "Cut the 2nd potato.", + "timestamps": { + "start": "00:04:22", + "end": "00:06:16" + } + }, + { + "description": "Cut the 3rd potato.", + "timestamps": { + "start": "00:06:34", + "end": "00:07:40" + } + } + ], + "question_timestamps": "00:07:58", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "One, meaning you only cut a single potato altogether during the task.", + "D": "Five, indicating you cut five separate potatoes in total by the end.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "P01_109_3", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "One, meaning you only cut a single potato altogether during the task.", + "D": "Five, indicating you cut five separate potatoes in total by the end.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I cut in total?", + "answer": "Three.", + "question_chinese": "我一共切了几个土豆？", + "answer_chinese": "3个。", + "evidence": [ + { + "description": "Cut the 1st potato.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:59" + } + }, + { + "description": "Cut the 2nd potato.", + "timestamps": { + "start": "00:04:22", + "end": "00:06:16" + } + }, + { + "description": "Cut the 3rd potato.", + "timestamps": { + "start": "00:06:34", + "end": "00:07:40" + } + } + ], + "question_timestamps": "00:11:16", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Four, so you ended up cutting four potatoes altogether during the preparation.", + "D": "Three.", + "E": "Five, meaning you cut five potatoes in total by the time you finished." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "P01_109_4", + "correct_answer": "Three.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Four, so you ended up cutting four potatoes altogether during the preparation.", + "D": "Three.", + "E": "Five, meaning you cut five potatoes in total by the time you finished." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_8_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I take out of the cabinet?", + "answer": "You took a pot out of the cabinet, and the pot is to your right.", + "question_chinese": "我从柜子里拿出了什么？", + "answer_chinese": "你从柜子里拿出了一个锅，锅在你的右边。", + "evidence": [ + { + "description": "Take out a pot from the cupboard.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:27" + } + } + ], + "question_timestamps": "00:01:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took a pot, in front of you.", + "B": "You took a pot, to your left.", + "C": "You took a pot out of the cabinet, and the pot is to your right.", + "D": "You took a pot out of the cabinet, and the pot is behind you after you set it down.", + "E": "You took a pot out of the cabinet, and the pot is above you, placed higher than your position." + }, + "answer_label": "C", + "answer_text": "You took a pot out of the cabinet, and the pot is to your right.", + "q_id": "P01_109_8_T2", + "correct_answer": "You took a pot out of the cabinet, and the pot is to your right.", + "original_options": { + "A": "You took a pot, in front of you.", + "B": "You took a pot, to your left.", + "C": "You took a pot out of the cabinet, and the pot is to your right.", + "D": "You took a pot out of the cabinet, and the pot is behind you after you set it down.", + "E": "You took a pot out of the cabinet, and the pot is above you, placed higher than your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 19 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_8_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I take out of the cabinet?", + "answer": "You took a cutting board out of the cabinet, and the cutting board is in front of you.", + "question_chinese": "我从柜子里拿出了什么？", + "answer_chinese": "你从柜子里拿出一个案板，案板在你的正前方。", + "evidence": [ + { + "description": "Took out a chopping board from the cupboard.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took a cutting board out of the cabinet, and the cutting board is in front of you.", + "B": "You took a cutting board, behind you.", + "C": "You took a cutting board, to your left.", + "D": "You took a cutting board out of the cabinet, and the cutting board is to your right after you pulled it out.", + "E": "You took a cutting board out of the cabinet, and the cutting board is above you, stored higher up." + }, + "answer_label": "A", + "answer_text": "You took a cutting board out of the cabinet, and the cutting board is in front of you.", + "q_id": "P01_109_8_T1", + "correct_answer": "You took a cutting board out of the cabinet, and the cutting board is in front of you.", + "original_options": { + "A": "You took a cutting board out of the cabinet, and the cutting board is in front of you.", + "B": "You took a cutting board, behind you.", + "C": "You took a cutting board, to your left.", + "D": "You took a cutting board out of the cabinet, and the cutting board is to your right after you pulled it out.", + "E": "You took a cutting board out of the cabinet, and the cutting board is above you, stored higher up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 22, + 19 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_3_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I take from the shelf next to the table?", + "answer": "You have now taken two potatoes from the shelf next to the table, and the potatoes are to your right.", + "question_chinese": "我从桌子旁边的架子上拿了几个土豆？", + "answer_chinese": "你现在从桌子旁边的架子上拿了二个土豆，土豆在你的右边。", + "evidence": [ + { + "description": "Took two potatoes from the shelf.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:01:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have now taken two potatoes from the shelf next to the table, and the potatoes are to your right.", + "B": "You took one potato.", + "C": "You took three potatoes.", + "D": "You have now taken one potato from the shelf next to the table, and it is to your right.", + "E": "You have now taken four potatoes from the shelf next to the table, and the potatoes are to your right." + }, + "answer_label": "A", + "answer_text": "You have now taken two potatoes from the shelf next to the table, and the potatoes are to your right.", + "q_id": "P01_109_3_T1", + "correct_answer": "You have now taken two potatoes from the shelf next to the table, and the potatoes are to your right.", + "original_options": { + "A": "You have now taken two potatoes from the shelf next to the table, and the potatoes are to your right.", + "B": "You took one potato.", + "C": "You took three potatoes.", + "D": "You have now taken one potato from the shelf next to the table, and it is to your right.", + "E": "You have now taken four potatoes from the shelf next to the table, and the potatoes are to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 20 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_3_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I take from the shelf next to the table?", + "answer": "In total, you took three potatoes from the shelf next to the table, and the potatoes are in front of you.", + "question_chinese": "我从桌子旁边的架子上拿了几个土豆？", + "answer_chinese": "你从桌子旁边的架子上一共拿了三个土豆，土豆在你的正前方。", + "evidence": [ + { + "description": "Took two potatoes from the shelf.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:11" + } + }, + { + "description": "Took one more potato from the shelf.", + "timestamps": { + "start": "00:06:31", + "end": "00:06:37" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In total, you took two potatoes.", + "B": "In total, you took three potatoes from the shelf next to the table, and the potatoes are in front of you.", + "C": "In total, you took four potatoes.", + "D": "In total, you took one potato from the shelf next to the table, and the potato is in front of you.", + "E": "In total, you took five potatoes from the shelf next to the table, and the potatoes are in front of you." + }, + "answer_label": "B", + "answer_text": "In total, you took three potatoes from the shelf next to the table, and the potatoes are in front of you.", + "q_id": "P01_109_3_T2", + "correct_answer": "In total, you took three potatoes from the shelf next to the table, and the potatoes are in front of you.", + "original_options": { + "A": "In total, you took two potatoes.", + "B": "In total, you took three potatoes from the shelf next to the table, and the potatoes are in front of you.", + "C": "In total, you took four potatoes.", + "D": "In total, you took one potato from the shelf next to the table, and the potato is in front of you.", + "E": "In total, you took five potatoes from the shelf next to the table, and the potatoes are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 21 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_5", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I take from the shelf next to the table?", + "answer": "Three.", + "question_chinese": "我从桌子旁边的架子上拿了几个土豆？", + "answer_chinese": "3个", + "evidence": [ + { + "description": "Took two potatoes from the shelf.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:11" + } + }, + { + "description": "Took one more potato from the shelf.", + "timestamps": { + "start": "00:06:31", + "end": "00:06:37" + } + } + ], + "question_timestamps": "00:08:38", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Six.", + "C": "Three.", + "D": "You took eight potatoes from the shelf next to the table during this step, not three in total.", + "E": "You took twelve potatoes from the shelf next to the table overall, which is far more than three." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "P01_109_5", + "correct_answer": "Three.", + "original_options": { + "A": "Five.", + "B": "Six.", + "C": "Three.", + "D": "You took eight potatoes from the shelf next to the table during this step, not three in total.", + "E": "You took twelve potatoes from the shelf next to the table overall, which is far more than three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 18 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_6", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I take from the shelf next to the table?", + "answer": "Three.", + "question_chinese": "我从桌子旁边的架子上拿了几个土豆？", + "answer_chinese": "3个", + "evidence": [ + { + "description": "Took two potatoes from the shelf.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:11" + } + }, + { + "description": "Took one more potato from the shelf.", + "timestamps": { + "start": "00:06:31", + "end": "00:06:37" + } + } + ], + "question_timestamps": "00:14:01", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Six.", + "C": "You took seven potatoes from the shelf next to the table in total, rather than only three.", + "D": "Three.", + "E": "You took eight potatoes from the shelf next to the table overall, not three potatoes." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "P01_109_6", + "correct_answer": "Three.", + "original_options": { + "A": "Five.", + "B": "Six.", + "C": "You took seven potatoes from the shelf next to the table in total, rather than only three.", + "D": "Three.", + "E": "You took eight potatoes from the shelf next to the table overall, not three potatoes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 15 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_2_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I cut?", + "answer": "You have now cut one potato, and the potato is in front of you.", + "question_chinese": "我切了几个土豆？", + "answer_chinese": "你现在切了一个土豆，土豆在你的正前方。", + "evidence": [ + { + "description": "You cut a potato now.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:31" + } + } + ], + "question_timestamps": "00:03:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You cut zero potatoes.", + "B": "You cut two potatoes.", + "C": "You have now cut three potatoes, and the potatoes are in front of you.", + "D": "You have now cut four potatoes, and the potatoes are in front of you.", + "E": "You have now cut one potato, and the potato is in front of you." + }, + "answer_label": "E", + "answer_text": "You have now cut one potato, and the potato is in front of you.", + "q_id": "P01_109_2_T1", + "correct_answer": "You have now cut one potato, and the potato is in front of you.", + "original_options": { + "A": "You cut zero potatoes.", + "B": "You cut two potatoes.", + "C": "You have now cut three potatoes, and the potatoes are in front of you.", + "D": "You have now cut four potatoes, and the potatoes are in front of you.", + "E": "You have now cut one potato, and the potato is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_2_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potatoes did I cut?", + "answer": "In total, you cut three potatoes, and the potatoes are in front of you.", + "question_chinese": "我切了几个土豆？", + "answer_chinese": "你一共切三个土豆，土豆在你的正前方。", + "evidence": [ + { + "description": "Cut the 1st potato.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:59" + } + }, + { + "description": "Cut the 2nd potato.", + "timestamps": { + "start": "00:04:22", + "end": "00:06:16" + } + }, + { + "description": "Cut the 3rd potato.", + "timestamps": { + "start": "00:06:34", + "end": "00:07:40" + } + } + ], + "question_timestamps": "00:07:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In total, you cut three potatoes, and the potatoes are in front of you.", + "B": "In total, you cut two potatoes.", + "C": "In total, you cut four potatoes.", + "D": "In total, you cut one potato, and the potato is in front of you.", + "E": "In total, you cut five potatoes, and the potatoes are in front of you." + }, + "answer_label": "A", + "answer_text": "In total, you cut three potatoes, and the potatoes are in front of you.", + "q_id": "P01_109_2_T2", + "correct_answer": "In total, you cut three potatoes, and the potatoes are in front of you.", + "original_options": { + "A": "In total, you cut three potatoes, and the potatoes are in front of you.", + "B": "In total, you cut two potatoes.", + "C": "In total, you cut four potatoes.", + "D": "In total, you cut one potato, and the potato is in front of you.", + "E": "In total, you cut five potatoes, and the potatoes are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I just wash in total?", + "answer": "A total of three items were washed.", + "question_chinese": "我刚才一共洗了几件物品？", + "answer_chinese": "一共洗了三件物品。", + "evidence": [ + { + "description": "Washing the peeler.", + "timestamps": { + "start": "00:41:28", + "end": "00:41:45" + } + }, + { + "description": "Washing the small knife.", + "timestamps": { + "start": "00:41:45", + "end": "00:41:49" + } + }, + { + "description": "Washing the cutting board.", + "timestamps": { + "start": "00:41:49", + "end": "00:42:03" + } + } + ], + "question_timestamps": "00:42:08", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "A total of three items were washed.", + "B": "Four items.", + "C": "Five items.", + "D": "A total of six items were washed just now, not three items overall.", + "E": "A total of seven items were washed during this washing step, not three." + }, + "answer_label": "A", + "answer_text": "A total of three items were washed.", + "q_id": "P01_109_1", + "correct_answer": "A total of three items were washed.", + "original_options": { + "A": "A total of three items were washed.", + "B": "Four items.", + "C": "Five items.", + "D": "A total of six items were washed just now, not three items overall.", + "E": "A total of seven items were washed during this washing step, not three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I just wash in total?", + "answer": "A total of three items were washed.", + "question_chinese": "我刚才一共洗了几件物品？", + "answer_chinese": "一共洗了三件物品。", + "evidence": [ + { + "description": "Washing the peeler.", + "timestamps": { + "start": "00:41:28", + "end": "00:41:45" + } + }, + { + "description": "Washing the small knife.", + "timestamps": { + "start": "00:41:45", + "end": "00:41:49" + } + }, + { + "description": "Washing the cutting board.", + "timestamps": { + "start": "00:41:49", + "end": "00:42:03" + } + } + ], + "question_timestamps": "00:46:04", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four items.", + "B": "Five items.", + "C": "A total of three items were washed.", + "D": "A total of six items were washed just now, not three items overall.", + "E": "A total of seven items were washed during this washing step, not three." + }, + "answer_label": "C", + "answer_text": "A total of three items were washed.", + "q_id": "P01_109_2", + "correct_answer": "A total of three items were washed.", + "original_options": { + "A": "Four items.", + "B": "Five items.", + "C": "A total of three items were washed.", + "D": "A total of six items were washed just now, not three items overall.", + "E": "A total of seven items were washed during this washing step, not three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_13", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I first entered the kitchen, how many bottles of wine did I take out of the bag?", + "answer": "Three bottles.", + "question_chinese": "我刚进厨房时，从包里拿出来了几瓶酒？", + "answer_chinese": "3瓶。", + "evidence": [ + { + "description": "Took 3 bottles of wine out of the bag and put them in the refrigerator.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:30:01", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two bottles.", + "B": "Four bottles.", + "C": "You took out two bottles of wine from the bag when you first entered the kitchen.", + "D": "You removed four wine bottles from the bag as soon as you stepped into the kitchen.", + "E": "Three bottles." + }, + "answer_label": "E", + "answer_text": "Three bottles.", + "q_id": "P01_109_13", + "correct_answer": "Three bottles.", + "original_options": { + "A": "Two bottles.", + "B": "Four bottles.", + "C": "You took out two bottles of wine from the bag when you first entered the kitchen.", + "D": "You removed four wine bottles from the bag as soon as you stepped into the kitchen.", + "E": "Three bottles." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_14", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I first entered the kitchen, how many bottles of wine did I take out of the bag?", + "answer": "Three bottles.", + "question_chinese": "我刚进厨房时，从包里拿出来了几瓶酒？", + "answer_chinese": "3瓶。", + "evidence": [ + { + "description": "Took 3 bottles of wine out of the bag and put them in the refrigerator.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:43:34", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One bottle.", + "B": "Five bottles.", + "C": "You took out four bottles of wine from the bag when you first entered the kitchen.", + "D": "Three bottles.", + "E": "You pulled out two bottles of wine from the bag right after entering the kitchen." + }, + "answer_label": "D", + "answer_text": "Three bottles.", + "q_id": "P01_109_14", + "correct_answer": "Three bottles.", + "original_options": { + "A": "One bottle.", + "B": "Five bottles.", + "C": "You took out four bottles of wine from the bag when you first entered the kitchen.", + "D": "Three bottles.", + "E": "You pulled out two bottles of wine from the bag right after entering the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 15 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_9_T2", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I add to the potatoes in the pot?", + "answer": "You added olive oil to the potatoes in the pot, and the olive oil is in your right hand.", + "question_chinese": "我往锅里的土豆加了什么？", + "answer_chinese": "你往锅里的土豆加了橄榄油，橄榄油在你的右手里。", + "evidence": [ + { + "description": "I added olive oil to the potatoes in the pot.", + "timestamps": { + "start": "00:08:23", + "end": "00:08:28" + } + } + ], + "question_timestamps": "00:08:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You added salt to the potatoes.", + "B": "You added butter to the potatoes.", + "C": "You added salt to the potatoes in the pot, and the salt is to your left.", + "D": "You added olive oil to the potatoes in the pot, and the olive oil is in your right hand.", + "E": "You added olive oil to the potatoes in the pot, and the olive oil is in your left hand." + }, + "answer_label": "D", + "answer_text": "You added olive oil to the potatoes in the pot, and the olive oil is in your right hand.", + "q_id": "P01_109_9_T2", + "correct_answer": "You added olive oil to the potatoes in the pot, and the olive oil is in your right hand.", + "original_options": { + "A": "You added salt to the potatoes.", + "B": "You added butter to the potatoes.", + "C": "You added salt to the potatoes in the pot, and the salt is to your left.", + "D": "You added olive oil to the potatoes in the pot, and the olive oil is in your right hand.", + "E": "You added olive oil to the potatoes in the pot, and the olive oil is in your left hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 19 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_9_T1", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What did I add to the potatoes in the pot?", + "answer": "You added salt to the potatoes in the pot, and the salt is to your left.", + "question_chinese": "我往锅里的土豆加了什么？", + "answer_chinese": "你往锅里的土豆加了盐，盐在你的左边。", + "evidence": [ + { + "description": "You added salt to the potatoes in the pot.", + "timestamps": { + "start": "00:08:58", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:09:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You added olive oil to them.", + "B": "You added butter to them.", + "C": "You added salt to the potatoes in the pot, and the salt is to your left.", + "D": "You added salt to the potatoes in the pot, and the salt is in front of you.", + "E": "You added salt to the potatoes in the pot, and the salt is to your right." + }, + "answer_label": "C", + "answer_text": "You added salt to the potatoes in the pot, and the salt is to your left.", + "q_id": "P01_109_9_T1", + "correct_answer": "You added salt to the potatoes in the pot, and the salt is to your left.", + "original_options": { + "A": "You added olive oil to them.", + "B": "You added butter to them.", + "C": "You added salt to the potatoes in the pot, and the salt is to your left.", + "D": "You added salt to the potatoes in the pot, and the salt is in front of you.", + "E": "You added salt to the potatoes in the pot, and the salt is to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 16 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_19", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to reheat this plate of potatoes and eggs. How should I walk to the stove?", + "answer": "The stove is behind you. Turn around to walk to the stove.", + "question_chinese": "我想把这盘土豆鸡蛋再加热一下，我应该怎么走到灶台？", + "answer_chinese": "灶台在你的身后，向后转就能走到灶台。", + "evidence": [ + { + "description": "The cooked potatoes and eggs were placed on the counter opposite the stove.", + "timestamps": { + "start": "00:55:31", + "end": "00:55:39" + } + } + ], + "question_timestamps": "01:00:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left to reach the stove.", + "B": "Walk straight ahead to the stove.", + "C": "The stove is behind you. Turn around to walk to the stove.", + "D": "The stove is to your right, so turn right and walk a few steps to reach it.", + "E": "The stove is in front of you; keep walking forward until you arrive at the stove." + }, + "answer_label": "C", + "answer_text": "The stove is behind you. Turn around to walk to the stove.", + "q_id": "P01_109_19", + "correct_answer": "The stove is behind you. Turn around to walk to the stove.", + "original_options": { + "A": "Turn left to reach the stove.", + "B": "Walk straight ahead to the stove.", + "C": "The stove is behind you. Turn around to walk to the stove.", + "D": "The stove is to your right, so turn right and walk a few steps to reach it.", + "E": "The stove is in front of you; keep walking forward until you arrive at the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_20", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to reheat this plate of potatoes and eggs. How should I walk to the stove?", + "answer": "The stove is behind you. Turn around to walk to the stove.", + "question_chinese": "我想把这盘土豆鸡蛋再加热一下，我应该怎么走到灶台？", + "answer_chinese": "灶台在你的身后，向后转就能走到灶台。", + "evidence": [ + { + "description": "The cooked potatoes and eggs were placed on the counter opposite the stove.", + "timestamps": { + "start": "00:55:31", + "end": "00:55:39" + } + } + ], + "question_timestamps": "01:01:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right to reach the stove.", + "B": "The stove is behind you. Turn around to walk to the stove.", + "C": "Walk straight to the stove.", + "D": "The stove is to your left, so turn left and walk over to the stove.", + "E": "The stove is in front of you; walk straight ahead and stop when you reach it." + }, + "answer_label": "B", + "answer_text": "The stove is behind you. Turn around to walk to the stove.", + "q_id": "P01_109_20", + "correct_answer": "The stove is behind you. Turn around to walk to the stove.", + "original_options": { + "A": "Turn right to reach the stove.", + "B": "The stove is behind you. Turn around to walk to the stove.", + "C": "Walk straight to the stove.", + "D": "The stove is to your left, so turn left and walk over to the stove.", + "E": "The stove is in front of you; walk straight ahead and stop when you reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 16 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_1_T1", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I washing right now?", + "answer": "You’re washing a peeler right now, and the peeler is in front of you.", + "question_chinese": "我现在在洗什么东西？", + "answer_chinese": "你现在在洗削皮刀，削皮刀在你的正前方。", + "evidence": [ + { + "description": "Washing the peeler.", + "timestamps": { + "start": "00:41:28", + "end": "00:41:37" + } + } + ], + "question_timestamps": "00:41:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’re washing a knife.", + "B": "You’re washing a spoon.", + "C": "You’re washing a peeler right now, and the peeler is to your left.", + "D": "You’re washing a peeler right now, and the peeler is in front of you.", + "E": "You’re washing a cutting board right now, and the cutting board is in front of you." + }, + "answer_label": "D", + "answer_text": "You’re washing a peeler right now, and the peeler is in front of you.", + "q_id": "P01_109_1_T1", + "correct_answer": "You’re washing a peeler right now, and the peeler is in front of you.", + "original_options": { + "A": "You’re washing a knife.", + "B": "You’re washing a spoon.", + "C": "You’re washing a peeler right now, and the peeler is to your left.", + "D": "You’re washing a peeler right now, and the peeler is in front of you.", + "E": "You’re washing a cutting board right now, and the cutting board is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 17 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_1_T2", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I washing right now?", + "answer": "You’re washing a cutting board right now, and the cutting board is to your left.", + "question_chinese": "我现在在洗什么东西？", + "answer_chinese": "你现在在洗案板，案板在你的左边。", + "evidence": [ + { + "description": "Washing the cutting board.", + "timestamps": { + "start": "00:41:49", + "end": "00:42:00" + } + } + ], + "question_timestamps": "00:42:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’re washing a peeler.", + "B": "You’re washing a knife.", + "C": "You’re washing a cutting board right now, and the cutting board is in front of you.", + "D": "You’re washing a cutting board right now, and the cutting board is to your right.", + "E": "You’re washing a cutting board right now, and the cutting board is to your left." + }, + "answer_label": "E", + "answer_text": "You’re washing a cutting board right now, and the cutting board is to your left.", + "q_id": "P01_109_1_T2", + "correct_answer": "You’re washing a cutting board right now, and the cutting board is to your left.", + "original_options": { + "A": "You’re washing a peeler.", + "B": "You’re washing a knife.", + "C": "You’re washing a cutting board right now, and the cutting board is in front of you.", + "D": "You’re washing a cutting board right now, and the cutting board is to your right.", + "E": "You’re washing a cutting board right now, and the cutting board is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_15", + "qid_num": 25, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did this box of eggs in green packaging in front of me come from?", + "answer": "It was taken out of the bag.", + "question_chinese": "我面前的这盒绿色包装的鸡蛋是从哪里来的？", + "answer_chinese": "从包里拿出来的。", + "evidence": [ + { + "description": "Took a box of eggs in green packaging out of the bag and placed it on the counter in front of me.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:20:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was taken out of the backpack.", + "B": "It was taken out of the cooler.", + "C": "It was taken out of the bag.", + "D": "It was taken out of the refrigerator after you opened the door and reached inside for it.", + "E": "It was picked up from the basket on the counter and then brought in front of you." + }, + "answer_label": "C", + "answer_text": "It was taken out of the bag.", + "q_id": "P01_109_15", + "correct_answer": "It was taken out of the bag.", + "original_options": { + "A": "It was taken out of the backpack.", + "B": "It was taken out of the cooler.", + "C": "It was taken out of the bag.", + "D": "It was taken out of the refrigerator after you opened the door and reached inside for it.", + "E": "It was picked up from the basket on the counter and then brought in front of you." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_16", + "qid_num": 26, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did this box of eggs in green packaging in front of me come from?", + "answer": "It was taken out of the bag.", + "question_chinese": "我面前的这盒绿色包装的鸡蛋是从哪里来的？", + "answer_chinese": "从包里拿出来的。", + "evidence": [ + { + "description": "Took a box of eggs in green packaging out of the bag and placed it on the counter in front of me.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:21:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was taken out of the bag.", + "B": "It was taken out of the refrigerator.", + "C": "It was picked up from the cart.", + "D": "It was taken from the basket on the counter and then carried over to you.", + "E": "You placed the box on the kitchen table earlier, then picked it up again." + }, + "answer_label": "A", + "answer_text": "It was taken out of the bag.", + "q_id": "P01_109_16", + "correct_answer": "It was taken out of the bag.", + "original_options": { + "A": "It was taken out of the bag.", + "B": "It was taken out of the refrigerator.", + "C": "It was picked up from the cart.", + "D": "It was taken from the basket on the counter and then carried over to you.", + "E": "You placed the box on the kitchen table earlier, then picked it up again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 14 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_9", + "qid_num": 27, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the washing machine or the sink?", + "answer": "The washing machine is closer to you.", + "question_chinese": "洗衣机和洗碗池，哪个离我更近？", + "answer_chinese": "洗衣机离你更近。", + "evidence": [ + { + "description": "The stove is to the upper right of the washing machine, and the sink is to the upper left of the washing machine.", + "timestamps": { + "start": "00:08:01", + "end": "00:08:16" + } + }, + { + "description": "You are now standing in front of the stove.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:06" + } + } + ], + "question_timestamps": "00:09:07", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The sink is closer to you.", + "B": "The washing machine is closer to you." + }, + "answer_label": "B", + "answer_text": "The washing machine is closer to you.", + "q_id": "P01_109_9", + "correct_answer": "The washing machine is closer to you.", + "original_options": { + "A": "The sink is closer to you.", + "B": "The washing machine is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_10", + "qid_num": 28, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the washing machine or the sink?", + "answer": "The washing machine is closer to you.", + "question_chinese": "洗衣机和洗碗池，哪个离我更近？", + "answer_chinese": "洗衣机离你更近。", + "evidence": [ + { + "description": "The stove is to the upper right of the washing machine, and the sink is to the upper left of the washing machine.", + "timestamps": { + "start": "00:08:01", + "end": "00:08:16" + } + }, + { + "description": "You are now standing in front of the stove.", + "timestamps": { + "start": "00:11:02", + "end": "00:11:04" + } + } + ], + "question_timestamps": "00:11:12", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The sink is closer to you.", + "B": "The washing machine is closer to you." + }, + "answer_label": "B", + "answer_text": "The washing machine is closer to you.", + "q_id": "P01_109_10", + "correct_answer": "The washing machine is closer to you.", + "original_options": { + "A": "The sink is closer to you.", + "B": "The washing machine is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_7_T2", + "qid_num": 29, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the stove relative to me?", + "answer": "The stove is behind you.", + "question_chinese": "灶台在我的哪里？", + "answer_chinese": "灶台在你的后面。", + "evidence": [ + { + "description": "The stove is behind you.", + "timestamps": { + "start": "00:07:58", + "end": "00:07:59" + } + } + ], + "question_timestamps": "00:08:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stove is in front of you.", + "B": "The stove is behind you.", + "C": "The stove is to your left.", + "D": "The stove is to your right, near your side as you face forward.", + "E": "The stove is above you, positioned higher than your eye level in the scene." + }, + "answer_label": "B", + "answer_text": "The stove is behind you.", + "q_id": "P01_109_7_T2", + "correct_answer": "The stove is behind you.", + "original_options": { + "A": "The stove is in front of you.", + "B": "The stove is behind you.", + "C": "The stove is to your left.", + "D": "The stove is to your right, near your side as you face forward.", + "E": "The stove is above you, positioned higher than your eye level in the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 14 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_7_T3", + "qid_num": 30, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the stove relative to me?", + "answer": "The stove is to your left.", + "question_chinese": "灶台在我的哪里？", + "answer_chinese": "灶台在你的左边。", + "evidence": [ + { + "description": "The stove is on your left.", + "timestamps": { + "start": "00:08:12", + "end": "00:08:15" + } + } + ], + "question_timestamps": "00:08:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stove is in front of you.", + "B": "The stove is behind you.", + "C": "The stove is to your right, opposite from where you’re standing and looking.", + "D": "The stove is above you, not at counter height but higher up in view.", + "E": "The stove is to your left." + }, + "answer_label": "E", + "answer_text": "The stove is to your left.", + "q_id": "P01_109_7_T3", + "correct_answer": "The stove is to your left.", + "original_options": { + "A": "The stove is in front of you.", + "B": "The stove is behind you.", + "C": "The stove is to your right, opposite from where you’re standing and looking.", + "D": "The stove is above you, not at counter height but higher up in view.", + "E": "The stove is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 14, + 14 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_7_T1", + "qid_num": 31, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the stove relative to me?", + "answer": "The stove is in front of you.", + "question_chinese": "灶台在我的哪里？", + "answer_chinese": "灶台在你的正前方。", + "evidence": [ + { + "description": "The stove is in front of you.", + "timestamps": { + "start": "00:08:17", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:08:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stove is behind you.", + "B": "The stove is to your right.", + "C": "The stove is to your left, positioned off to the side rather than straight ahead.", + "D": "The stove is above you, located higher than you and not directly in your path.", + "E": "The stove is in front of you." + }, + "answer_label": "E", + "answer_text": "The stove is in front of you.", + "q_id": "P01_109_7_T1", + "correct_answer": "The stove is in front of you.", + "original_options": { + "A": "The stove is behind you.", + "B": "The stove is to your right.", + "C": "The stove is to your left, positioned off to the side rather than straight ahead.", + "D": "The stove is above you, located higher than you and not directly in your path.", + "E": "The stove is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_6_T1", + "qid_num": 32, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I take out of my backpack just now?", + "answer": "You have now taken out a carton of eggs from your backpack, and the backpack is at your feet.", + "question_chinese": "现在我从书包里拿出了什么？", + "answer_chinese": "现在你从书包里拿出了一盒鸡蛋，书包在你的脚下。", + "evidence": [ + { + "description": "Now you take out a box of eggs from your bag.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took out an onion and potatoes.", + "B": "You took out a water bottle.", + "C": "You have now taken out a carton of eggs from your backpack, and the backpack is in front of you.", + "D": "You have now taken out a carton of eggs from your backpack, and the backpack is at your feet.", + "E": "You have now taken out a carton of eggs from your backpack, and the backpack is to your left." + }, + "answer_label": "D", + "answer_text": "You have now taken out a carton of eggs from your backpack, and the backpack is at your feet.", + "q_id": "P01_109_6_T1", + "correct_answer": "You have now taken out a carton of eggs from your backpack, and the backpack is at your feet.", + "original_options": { + "A": "You took out an onion and potatoes.", + "B": "You took out a water bottle.", + "C": "You have now taken out a carton of eggs from your backpack, and the backpack is in front of you.", + "D": "You have now taken out a carton of eggs from your backpack, and the backpack is at your feet.", + "E": "You have now taken out a carton of eggs from your backpack, and the backpack is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 19 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_6_T2", + "qid_num": 33, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I take out of my backpack just now?", + "answer": "You have now taken out one onion and three potatoes from your backpack, and the onion and potatoes are in front of you.", + "question_chinese": "现在我从书包里拿出了什么？", + "answer_chinese": "现在你从书包里拿出了一个洋葱和三个土豆，洋葱和土豆在你的正前方。", + "evidence": [ + { + "description": "Now you take out an onion and three potatoes from your bag.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took out one onion and two potatoes.", + "B": "You took out two onions and three potatoes.", + "C": "You have now taken out one onion and four potatoes from your backpack, and they are in front of you.", + "D": "You have now taken out no onion and three potatoes from your backpack, and the potatoes are in front of you.", + "E": "You have now taken out one onion and three potatoes from your backpack, and the onion and potatoes are in front of you." + }, + "answer_label": "E", + "answer_text": "You have now taken out one onion and three potatoes from your backpack, and the onion and potatoes are in front of you.", + "q_id": "P01_109_6_T2", + "correct_answer": "You have now taken out one onion and three potatoes from your backpack, and the onion and potatoes are in front of you.", + "original_options": { + "A": "You took out one onion and two potatoes.", + "B": "You took out two onions and three potatoes.", + "C": "You have now taken out one onion and four potatoes from your backpack, and they are in front of you.", + "D": "You have now taken out no onion and three potatoes from your backpack, and the potatoes are in front of you.", + "E": "You have now taken out one onion and three potatoes from your backpack, and the onion and potatoes are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 20, + 21 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_4_T1", + "qid_num": 34, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me—the washing machine or the sink?", + "answer": "Right now, the sink is closer to you, and it is in front of you.", + "question_chinese": "现在洗衣机和洗碗池，哪个离我更近？", + "answer_chinese": "现在洗碗池离你更近，洗碗池在你的正前方。", + "evidence": [ + { + "description": "Now the sink is closer to you.", + "timestamps": { + "start": "00:08:01", + "end": "00:08:05" + } + } + ], + "question_timestamps": "00:08:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, the sink is closer to you, and it is in front of you.", + "B": "Right now, the washing machine is closer to you, and it is in front of you." + }, + "answer_label": "A", + "answer_text": "Right now, the sink is closer to you, and it is in front of you.", + "q_id": "P01_109_4_T1", + "correct_answer": "Right now, the sink is closer to you, and it is in front of you.", + "original_options": { + "A": "Right now, the sink is closer to you, and it is in front of you.", + "B": "Right now, the washing machine is closer to you, and it is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_4_T2", + "qid_num": 35, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me—the washing machine or the sink?", + "answer": "Right now, the washing machine is closer to you, and it is to your left.", + "question_chinese": "现在洗衣机和洗碗池，哪个离我更近？", + "answer_chinese": "现在洗衣机离你更近，洗衣机在你的左边。", + "evidence": [ + { + "description": "Now the washing machine is closer to you.", + "timestamps": { + "start": "00:08:18", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:08:22", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, the sink is closer to you, and it is to your right.", + "B": "Right now, the washing machine is closer to you, and it is to your left." + }, + "answer_label": "B", + "answer_text": "Right now, the washing machine is closer to you, and it is to your left.", + "q_id": "P01_109_4_T2", + "correct_answer": "Right now, the washing machine is closer to you, and it is to your left.", + "original_options": { + "A": "Right now, the sink is closer to you, and it is to your right.", + "B": "Right now, the washing machine is closer to you, and it is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_10_T1", + "qid_num": 36, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the red kettle relative to me?", + "answer": "The red kettle is to your left.", + "question_chinese": "红色烧水壶在我的哪里？", + "answer_chinese": "红色烧水壶在你的左边。", + "evidence": [ + { + "description": "The red kettle is on your left.", + "timestamps": { + "start": "00:10:03", + "end": "00:10:11" + } + } + ], + "question_timestamps": "00:10:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The red kettle is ahead of you.", + "B": "The red kettle is to your right.", + "C": "The red kettle is behind you, a little farther away than the counter in front of you.", + "D": "The red kettle is to your left.", + "E": "The red kettle is above you, positioned higher than eye level compared with the nearby countertop." + }, + "answer_label": "D", + "answer_text": "The red kettle is to your left.", + "q_id": "P01_109_10_T1", + "correct_answer": "The red kettle is to your left.", + "original_options": { + "A": "The red kettle is ahead of you.", + "B": "The red kettle is to your right.", + "C": "The red kettle is behind you, a little farther away than the counter in front of you.", + "D": "The red kettle is to your left.", + "E": "The red kettle is above you, positioned higher than eye level compared with the nearby countertop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "P01_109", + "video_length": "01:01:48", + "qid": "P01_109_10_T2", + "qid_num": 37, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the red kettle relative to me?", + "answer": "The red kettle is behind you.", + "question_chinese": "红色烧水壶在我的哪里？", + "answer_chinese": "红色烧水壶在你的后面。", + "evidence": [ + { + "description": "The red kettle is behind you.", + "timestamps": { + "start": "00:10:50", + "end": "00:11:05" + } + } + ], + "question_timestamps": "00:11:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The red kettle is in front of you.", + "B": "The red kettle is to your left.", + "C": "The red kettle is to your right, not directly behind you where you just were standing.", + "D": "The red kettle is above you, higher up than the surrounding items on the counter.", + "E": "The red kettle is behind you." + }, + "answer_label": "E", + "answer_text": "The red kettle is behind you.", + "q_id": "P01_109_10_T2", + "correct_answer": "The red kettle is behind you.", + "original_options": { + "A": "The red kettle is in front of you.", + "B": "The red kettle is to your left.", + "C": "The red kettle is to your right, not directly behind you where you just were standing.", + "D": "The red kettle is above you, higher up than the surrounding items on the counter.", + "E": "The red kettle is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 15 + ] + } +] diff --git a/QAs/P02_101-mcq.json b/QAs/P02_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..96b05af5531ae295b0655edee278a1dbd8fb77f8 --- /dev/null +++ b/QAs/P02_101-mcq.json @@ -0,0 +1,814 @@ +[ + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the refrigerator door so far?", + "answer": "Three times.", + "question_chinese": "到目前为止，我一共开了几次冰箱门？", + "answer_chinese": "三次", + "evidence": [ + { + "description": "Opened the refrigerator door for the first time", + "timestamps": { + "start": "00:00:18", + "end": "00:00:21" + } + }, + { + "description": "Opened the refrigerator door for the second time", + "timestamps": { + "start": "00:00:32", + "end": "00:00:36" + } + }, + { + "description": "Opened the refrigerator door for the third time", + "timestamps": { + "start": "00:01:26", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:02:35", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Twice.", + "C": "Three times.", + "D": "I have opened the refrigerator door four times so far, counting each time I pulled it open.", + "E": "I have opened the refrigerator door five times so far, including the most recent time I reached inside." + }, + "answer_label": "C", + "answer_text": "Three times.", + "q_id": "P02_101_2", + "correct_answer": "Three times.", + "original_options": { + "A": "Once.", + "B": "Twice.", + "C": "Three times.", + "D": "I have opened the refrigerator door four times so far, counting each time I pulled it open.", + "E": "I have opened the refrigerator door five times so far, including the most recent time I reached inside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 18 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the refrigerator door so far?", + "answer": "Four times.", + "question_chinese": "到目前为止，我一共开了几次冰箱门？", + "answer_chinese": "四次", + "evidence": [ + { + "description": "Opened the refrigerator door for the first time", + "timestamps": { + "start": "00:00:18", + "end": "00:00:21" + } + }, + { + "description": "Opened the refrigerator door for the second time", + "timestamps": { + "start": "00:00:32", + "end": "00:00:36" + } + }, + { + "description": "Opened the refrigerator door for the third time", + "timestamps": { + "start": "00:01:26", + "end": "00:01:31" + } + }, + { + "description": "Opened the refrigerator door for the fourth time", + "timestamps": { + "start": "00:03:57", + "end": "00:04:02" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four times.", + "B": "Twice.", + "C": "Three times.", + "D": "I have opened the refrigerator door five times so far, counting each separate time I swung it open.", + "E": "I have opened the refrigerator door six times so far, including the last time I closed it again." + }, + "answer_label": "A", + "answer_text": "Four times.", + "q_id": "P02_101_1", + "correct_answer": "Four times.", + "original_options": { + "A": "Four times.", + "B": "Twice.", + "C": "Three times.", + "D": "I have opened the refrigerator door five times so far, counting each separate time I swung it open.", + "E": "I have opened the refrigerator door six times so far, including the last time I closed it again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 18, + 18 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_9", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where should this butter on the countertop be put back?", + "answer": "It should be put back in the refrigerator.", + "question_chinese": "台面上这个黄油应该放回哪里？", + "answer_chinese": "应该放回冰箱里。", + "evidence": [ + { + "description": "Took the butter and jam out of the refrigerator and placed them on the countertop.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:01:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put it on the kitchen table.", + "B": "Put it back in the cupboard.", + "C": "It should be put back in the refrigerator.", + "D": "It should be put back on top of the refrigerator, not inside any appliance.", + "E": "It should be put back next to the refrigerator, leaving it out rather than returning it to cold storage." + }, + "answer_label": "C", + "answer_text": "It should be put back in the refrigerator.", + "q_id": "P02_101_9", + "correct_answer": "It should be put back in the refrigerator.", + "original_options": { + "A": "Put it on the kitchen table.", + "B": "Put it back in the cupboard.", + "C": "It should be put back in the refrigerator.", + "D": "It should be put back on top of the refrigerator, not inside any appliance.", + "E": "It should be put back next to the refrigerator, leaving it out rather than returning it to cold storage." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 19 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_10", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where should this butter on the countertop be put back?", + "answer": "It should be put back in the refrigerator.", + "question_chinese": "台面上这个黄油应该放回哪里？", + "answer_chinese": "应该放回冰箱里。", + "evidence": [ + { + "description": "Took the butter and jam out of the refrigerator and placed them on the countertop.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:03:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put it back in the pantry.", + "B": "Put it back in a cabinet.", + "C": "It should be put back on the pantry shelf where you stored it earlier, not in the refrigerator.", + "D": "It should be put back in the refrigerator.", + "E": "It should be put back into the cupboard on the counter side, rather than being chilled in the fridge." + }, + "answer_label": "D", + "answer_text": "It should be put back in the refrigerator.", + "q_id": "P02_101_10", + "correct_answer": "It should be put back in the refrigerator.", + "original_options": { + "A": "Put it back in the pantry.", + "B": "Put it back in a cabinet.", + "C": "It should be put back on the pantry shelf where you stored it earlier, not in the refrigerator.", + "D": "It should be put back in the refrigerator.", + "E": "It should be put back into the cupboard on the counter side, rather than being chilled in the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_6_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of bread did I take out of the bag?", + "answer": "You have taken out 1 piece now.", + "question_chinese": "我从袋子里取出了几块面包？", + "answer_chinese": "你现在取出1块。", + "evidence": [ + { + "description": "Opened the bag, took out 1 slice of bread, and threw it in the trash.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took out 0 pieces.", + "B": "You have taken out 1 piece now.", + "C": "You took out 2 pieces.", + "D": "You have taken out 3 pieces now, leaving the rest of the slices still inside the bag.", + "E": "You have taken out 4 pieces now, with several slices remaining in the bread bag." + }, + "answer_label": "B", + "answer_text": "You have taken out 1 piece now.", + "q_id": "P02_101_6_T1", + "correct_answer": "You have taken out 1 piece now.", + "original_options": { + "A": "You took out 0 pieces.", + "B": "You have taken out 1 piece now.", + "C": "You took out 2 pieces.", + "D": "You have taken out 3 pieces now, leaving the rest of the slices still inside the bag.", + "E": "You have taken out 4 pieces now, with several slices remaining in the bread bag." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 15 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_6_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of bread did I take out of the bag?", + "answer": "You have taken out 3 pieces now.", + "question_chinese": "我从袋子里取出了几块面包？", + "answer_chinese": "你现在取出3块。", + "evidence": [ + { + "description": "Opened the bag, took out 1 slice of bread, and threw it in the trash.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:55" + } + }, + { + "description": "Took two slices of bread from the bag and put them into the toaster.", + "timestamps": { + "start": "00:00:57", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:03:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took out 1 piece.", + "B": "You took out 2 pieces.", + "C": "You have taken out 3 pieces now.", + "D": "You have taken out 4 pieces now, so there are four slices outside the bag in total.", + "E": "You have taken out 5 pieces now, meaning five slices have been removed from the bag altogether." + }, + "answer_label": "C", + "answer_text": "You have taken out 3 pieces now.", + "q_id": "P02_101_6_T2", + "correct_answer": "You have taken out 3 pieces now.", + "original_options": { + "A": "You took out 1 piece.", + "B": "You took out 2 pieces.", + "C": "You have taken out 3 pieces now.", + "D": "You have taken out 4 pieces now, so there are four slices outside the bag in total.", + "E": "You have taken out 5 pieces now, meaning five slices have been removed from the bag altogether." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_5_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I just went to get milk, which places did I pass through?", + "answer": "You went along the left side of the countertop, passed the microwave, and walked to the refrigerator.", + "question_chinese": "我刚刚去拿牛奶的时候，经过了哪些地方？", + "answer_chinese": "你从台面左侧经过微波炉，走到冰箱。", + "evidence": [ + { + "description": "Went from the left side of the countertop, past the microwave, to the refrigerator.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went straight to the refrigerator.", + "B": "You went along the left side of the countertop, passed the microwave, and walked to the refrigerator.", + "C": "You passed the microwave on the right.", + "D": "You went along the right side of the countertop, passed the microwave, and then walked to the refrigerator.", + "E": "You walked to the microwave first and then went to the refrigerator without following the left side of the counter." + }, + "answer_label": "B", + "answer_text": "You went along the left side of the countertop, passed the microwave, and walked to the refrigerator.", + "q_id": "P02_101_5_T2", + "correct_answer": "You went along the left side of the countertop, passed the microwave, and walked to the refrigerator.", + "original_options": { + "A": "You went straight to the refrigerator.", + "B": "You went along the left side of the countertop, passed the microwave, and walked to the refrigerator.", + "C": "You passed the microwave on the right.", + "D": "You went along the right side of the countertop, passed the microwave, and then walked to the refrigerator.", + "E": "You walked to the microwave first and then went to the refrigerator without following the left side of the counter." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 20 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_5_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I just went to get milk, which places did I pass through?", + "answer": "You went along the left side of the countertop, passed the microwave, walked to the refrigerator, and then returned the same way.", + "question_chinese": "我刚刚去拿牛奶的时候，经过了哪些地方？", + "answer_chinese": "你从台面左侧经过微波炉，走到冰箱，然后又以同样的方式返回。", + "evidence": [ + { + "description": "Went from the left side of the countertop, past the microwave, to the refrigerator.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:19" + } + }, + { + "description": "Returned from the refrigerator, past the microwave, to the left side of the countertop.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went along the left side of the countertop, passed the microwave, walked to the refrigerator, and then returned the same way.", + "B": "You didn’t return the same way.", + "C": "You went along the right side.", + "D": "You went along the left side, passed the microwave, and walked to the refrigerator without returning.", + "E": "You went to the refrigerator first, then passed the microwave only on the way back along the left side." + }, + "answer_label": "A", + "answer_text": "You went along the left side of the countertop, passed the microwave, walked to the refrigerator, and then returned the same way.", + "q_id": "P02_101_5_T1", + "correct_answer": "You went along the left side of the countertop, passed the microwave, walked to the refrigerator, and then returned the same way.", + "original_options": { + "A": "You went along the left side of the countertop, passed the microwave, walked to the refrigerator, and then returned the same way.", + "B": "You didn’t return the same way.", + "C": "You went along the right side.", + "D": "You went along the left side, passed the microwave, and walked to the refrigerator without returning.", + "E": "You went to the refrigerator first, then passed the microwave only on the way back along the left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 19 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_4_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where should I put this butter back on the countertop?", + "answer": "It should be put back in the refrigerator, and the refrigerator is to your right.", + "question_chinese": "我台面上这个黄油应该放回哪里？", + "answer_chinese": "应该放回冰箱里，冰箱在你右边。", + "evidence": [ + { + "description": "Took the butter and jam out of the refrigerator and placed them on the countertop.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:01:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it back on your left.", + "B": "Put it back behind you.", + "C": "It should be put back in the pantry, and the pantry is in front of you near the counter edge.", + "D": "It should be put back in a cabinet, and the cabinet is straight ahead rather than on your right side.", + "E": "It should be put back in the refrigerator, and the refrigerator is to your right." + }, + "answer_label": "E", + "answer_text": "It should be put back in the refrigerator, and the refrigerator is to your right.", + "q_id": "P02_101_4_T1", + "correct_answer": "It should be put back in the refrigerator, and the refrigerator is to your right.", + "original_options": { + "A": "Put it back on your left.", + "B": "Put it back behind you.", + "C": "It should be put back in the pantry, and the pantry is in front of you near the counter edge.", + "D": "It should be put back in a cabinet, and the cabinet is straight ahead rather than on your right side.", + "E": "It should be put back in the refrigerator, and the refrigerator is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 20 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_4_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where should I put this butter back on the countertop?", + "answer": "It should be put back in the refrigerator, and the refrigerator is behind you.", + "question_chinese": "我台面上这个黄油应该放回哪里？", + "answer_chinese": "应该放回冰箱里，冰箱在你后边。", + "evidence": [ + { + "description": "Took the butter and jam out of the refrigerator and placed them on the countertop.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It should be put back in the refrigerator, and the refrigerator is behind you.", + "B": "Put it back in the pantry.", + "C": "Put it back in a cabinet.", + "D": "It should be put back in the pantry, and the pantry is to your left, not behind you.", + "E": "It should be put back in a cabinet, and the cabinet is straight ahead rather than located behind you." + }, + "answer_label": "A", + "answer_text": "It should be put back in the refrigerator, and the refrigerator is behind you.", + "q_id": "P02_101_4_T2", + "correct_answer": "It should be put back in the refrigerator, and the refrigerator is behind you.", + "original_options": { + "A": "It should be put back in the refrigerator, and the refrigerator is behind you.", + "B": "Put it back in the pantry.", + "C": "Put it back in a cabinet.", + "D": "It should be put back in the pantry, and the pantry is to your left, not behind you.", + "E": "It should be put back in a cabinet, and the cabinet is straight ahead rather than located behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_3_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "This bag of bread in my hand—along what path did it move to get here?", + "answer": "The bread was on the countertop to your left of the sink.", + "question_chinese": "我手里这袋面包是从什么路径移动过来的？", + "answer_chinese": "面包从洗碗池左边的台面。", + "evidence": [ + { + "description": "Picked up a bag of bread from the countertop to the left of the sink and walked past the sink and the washing machine to the countertop on the right.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was right of the sink.", + "B": "It was in front of the sink.", + "C": "The bread was on the countertop to your right of the sink, not on the left side.", + "D": "The bread was on the countertop to your left of the sink.", + "E": "The bread started on the countertop directly in front of the sink before you picked it up into your hand." + }, + "answer_label": "D", + "answer_text": "The bread was on the countertop to your left of the sink.", + "q_id": "P02_101_3_T1", + "correct_answer": "The bread was on the countertop to your left of the sink.", + "original_options": { + "A": "It was right of the sink.", + "B": "It was in front of the sink.", + "C": "The bread was on the countertop to your right of the sink, not on the left side.", + "D": "The bread was on the countertop to your left of the sink.", + "E": "The bread started on the countertop directly in front of the sink before you picked it up into your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 20 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_3_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "This bag of bread in my hand—along what path did it move to get here?", + "answer": "The bread moved from the countertop to your left of the sink, past the sink and the washing machine, to the countertop to your right.", + "question_chinese": "我手里这袋面包是从什么路径移动过来的？", + "answer_chinese": "面包从洗碗池左边的台面，经过洗碗池和洗衣机，到达右边的台面。", + "evidence": [ + { + "description": "Picked up a bag of bread from the countertop to the left of the sink and walked past the sink and the washing machine to the countertop on the right.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the sink area to the right countertop.", + "B": "From the right countertop to your hand.", + "C": "The bread moved from the countertop in front of the sink, past the washing machine, ending on the right countertop.", + "D": "The bread moved from the countertop to your left of the sink, past the sink and the washing machine, to the countertop to your right.", + "E": "The bread moved from the left countertop, around the washing machine first, then past the sink to the right countertop." + }, + "answer_label": "D", + "answer_text": "The bread moved from the countertop to your left of the sink, past the sink and the washing machine, to the countertop to your right.", + "q_id": "P02_101_3_T2", + "correct_answer": "The bread moved from the countertop to your left of the sink, past the sink and the washing machine, to the countertop to your right.", + "original_options": { + "A": "From the sink area to the right countertop.", + "B": "From the right countertop to your hand.", + "C": "The bread moved from the countertop in front of the sink, past the washing machine, ending on the right countertop.", + "D": "The bread moved from the countertop to your left of the sink, past the sink and the washing machine, to the countertop to your right.", + "E": "The bread moved from the left countertop, around the washing machine first, then past the sink to the right countertop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 20, + 20 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_2_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which one is closer to me, the washing machine or the refrigerator?", + "answer": "The washing machine.", + "question_chinese": "现在洗衣机和冰箱哪个离我更近？", + "answer_chinese": "洗衣机。", + "evidence": [ + { + "description": "The washing machine is about half a meter away at the bottom left of the countertop.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:16" + } + }, + { + "description": "The refrigerator is about two meters away on the right side of the countertop.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator.", + "B": "Neither; they’re equally close.", + "C": "The washing machine.", + "D": "The refrigerator is closer to you than the washing machine right now.", + "E": "They’re about the same distance from you, so neither one is noticeably closer." + }, + "answer_label": "C", + "answer_text": "The washing machine.", + "q_id": "P02_101_2_T1", + "correct_answer": "The washing machine.", + "original_options": { + "A": "The refrigerator.", + "B": "Neither; they’re equally close.", + "C": "The washing machine.", + "D": "The refrigerator is closer to you than the washing machine right now.", + "E": "They’re about the same distance from you, so neither one is noticeably closer." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 5, + 12, + 14 + ] + }, + { + "video_uid": "P02_101", + "video_length": "00:04:03", + "qid": "P02_101_2_T2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which one is closer to me, the washing machine or the refrigerator?", + "answer": "The refrigerator.", + "question_chinese": "现在洗衣机和冰箱哪个离我更近？", + "answer_chinese": "冰箱。", + "evidence": [ + { + "description": "The washing machine is about half a meter away at the bottom left of the countertop.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:16" + } + }, + { + "description": "The refrigerator is about two meters away on the right side of the countertop.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:23" + } + }, + { + "description": "You returned to the countertop from near the refrigerator.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:02:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The washing machine.", + "B": "Neither; they’re equally close.", + "C": "The washing machine is closer to you than the refrigerator right now.", + "D": "The refrigerator.", + "E": "They’re about the same distance from you, so neither one is noticeably closer." + }, + "answer_label": "D", + "answer_text": "The refrigerator.", + "q_id": "P02_101_2_T2", + "correct_answer": "The refrigerator.", + "original_options": { + "A": "The washing machine.", + "B": "Neither; they’re equally close.", + "C": "The washing machine is closer to you than the refrigerator right now.", + "D": "The refrigerator.", + "E": "They’re about the same distance from you, so neither one is noticeably closer." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 5, + 12, + 14 + ] + } +] diff --git a/QAs/P02_115-mcq.json b/QAs/P02_115-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..26d79d70666fbc41f0d37e6dcb616533d4feae36 --- /dev/null +++ b/QAs/P02_115-mcq.json @@ -0,0 +1,1163 @@ +[ + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_7_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I opened the oven?", + "answer": "You opened it once.", + "question_chinese": "到目前为止，我一共打开了几次烤箱？", + "answer_chinese": "你打开了一次。", + "evidence": [ + { + "description": "Opened the oven for the first time.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:0:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You never opened it.", + "B": "You opened it twice.", + "C": "You opened it once.", + "D": "You opened the oven three separate times up to this point in the video.", + "E": "You opened the oven four times in total, including a couple of quick checks." + }, + "answer_label": "C", + "answer_text": "You opened it once.", + "q_id": "P02_115_7_T1", + "correct_answer": "You opened it once.", + "original_options": { + "A": "You never opened it.", + "B": "You opened it twice.", + "C": "You opened it once.", + "D": "You opened the oven three separate times up to this point in the video.", + "E": "You opened the oven four times in total, including a couple of quick checks." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_16", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many times have I opened the oven in total?", + "answer": "Twice.", + "question_chinese": "到目前为止，我一共打开了几次烤箱？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "Opened the oven for the first time.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:39" + } + }, + { + "description": "Opened the oven for the second time.", + "timestamps": { + "start": "00:01:37", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Zero times.", + "C": "Three times.", + "D": "You opened the oven once in total so far in this scene.", + "E": "You opened the oven four times altogether up to now in the video." + }, + "answer_label": "A", + "answer_text": "Twice.", + "q_id": "P02_115_16", + "correct_answer": "Twice.", + "original_options": { + "A": "Twice.", + "B": "Zero times.", + "C": "Three times.", + "D": "You opened the oven once in total so far in this scene.", + "E": "You opened the oven four times altogether up to now in the video." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_7_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I opened the oven?", + "answer": "You opened it twice.", + "question_chinese": "到目前为止，我一共打开了几次烤箱？", + "answer_chinese": "你打开了两次。", + "evidence": [ + { + "description": "Opened the oven for the first time.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:39" + } + }, + { + "description": "Opened the oven for the second time.", + "timestamps": { + "start": "00:01:37", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You never opened it.", + "B": "You opened it once.", + "C": "You opened the oven three separate times up to this point in the video.", + "D": "You opened it twice.", + "E": "You opened the oven four times in total, including a couple of quick checks." + }, + "answer_label": "D", + "answer_text": "You opened it twice.", + "q_id": "P02_115_7_T2", + "correct_answer": "You opened it twice.", + "original_options": { + "A": "You never opened it.", + "B": "You opened it once.", + "C": "You opened the oven three separate times up to this point in the video.", + "D": "You opened it twice.", + "E": "You opened the oven four times in total, including a couple of quick checks." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_6_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I used the gloves?", + "answer": "You used it twice.", + "question_chinese": "到目前为止，我一共用了几次手套？", + "answer_chinese": "你用了2次。", + "evidence": [ + { + "description": "First time using the gloves.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:25" + } + }, + { + "description": "Second time using the gloves.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You didn’t use the gloves.", + "B": "You used it twice.", + "C": "You used it once.", + "D": "You used the gloves three different times so far during the task.", + "E": "You used the gloves four times in total up to now in the video." + }, + "answer_label": "B", + "answer_text": "You used it twice.", + "q_id": "P02_115_6_T1", + "correct_answer": "You used it twice.", + "original_options": { + "A": "You didn’t use the gloves.", + "B": "You used it twice.", + "C": "You used it once.", + "D": "You used the gloves three different times so far during the task.", + "E": "You used the gloves four times in total up to now in the video." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 12, + 14 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_13", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many times have I used the gloves in total?", + "answer": "three times.", + "question_chinese": "到目前为止，我一共用了几次手套？", + "answer_chinese": "3次。", + "evidence": [ + { + "description": "First time using the gloves.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:25" + } + }, + { + "description": "Second time using the gloves.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:31" + } + }, + { + "description": "Third time using the gloves.", + "timestamps": { + "start": "00:01:37", + "end": "00:01:43" + } + } + ], + "question_timestamps": "00:01:44", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four times.", + "B": "Five times.", + "C": "You used the gloves twice in total so far during the task.", + "D": "three times.", + "E": "You used the gloves six times altogether up to this point in the video." + }, + "answer_label": "D", + "answer_text": "three times.", + "q_id": "P02_115_13", + "correct_answer": "three times.", + "original_options": { + "A": "Four times.", + "B": "Five times.", + "C": "You used the gloves twice in total so far during the task.", + "D": "three times.", + "E": "You used the gloves six times altogether up to this point in the video." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_6_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I used the gloves?", + "answer": "You used it three times.", + "question_chinese": "到目前为止，我一共用了几次手套？", + "answer_chinese": "你用了3次。", + "evidence": [ + { + "description": "First time using the gloves.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:25" + } + }, + { + "description": "Second time using the gloves.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:31" + } + }, + { + "description": "Third time using the gloves.", + "timestamps": { + "start": "00:01:37", + "end": "00:01:43" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used it once.", + "B": "You used it twice.", + "C": "You used it three times.", + "D": "Up to now, you have used the gloves four times in total.", + "E": "So far, you have not used the gloves at any point yet." + }, + "answer_label": "C", + "answer_text": "You used it three times.", + "q_id": "P02_115_6_T2", + "correct_answer": "You used it three times.", + "original_options": { + "A": "You used it once.", + "B": "You used it twice.", + "C": "You used it three times.", + "D": "Up to now, you have used the gloves four times in total.", + "E": "So far, you have not used the gloves at any point yet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many loaves of bread have I taken out of the oven in total?", + "answer": "eleven.", + "question_chinese": "到目前为止，我从烤箱一共拿出了多少个面包？", + "answer_chinese": "11个。", + "evidence": [ + { + "description": "Took a tray of 3 loaves of bread from the oven.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:25" + } + }, + { + "description": "Took a tray of 8 loaves of bread from the oven.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Nine.", + "B": "Thirteen.", + "C": "In total so far, you have taken out fourteen loaves of bread from the oven.", + "D": "Up to this point, you have removed six loaves of bread from the oven in total.", + "E": "eleven." + }, + "answer_label": "E", + "answer_text": "eleven.", + "q_id": "P02_115_11", + "correct_answer": "eleven.", + "original_options": { + "A": "Nine.", + "B": "Thirteen.", + "C": "In total so far, you have taken out fourteen loaves of bread from the oven.", + "D": "Up to this point, you have removed six loaves of bread from the oven in total.", + "E": "eleven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_12", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many loaves of bread have I taken out of the oven in total?", + "answer": "seventeen.", + "question_chinese": "到目前为止，我从烤箱一共拿出了多少个面包？", + "answer_chinese": "17个。", + "evidence": [ + { + "description": "Took a tray of 3 loaves of bread from the oven.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:25" + } + }, + { + "description": "Took a tray of 8 loaves of bread from the oven.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:31" + } + }, + { + "description": "Took a tray of 6 loaves of bread from the oven.", + "timestamps": { + "start": "00:01:37", + "end": "00:01:43" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Fourteen.", + "B": "seventeen.", + "C": "Twenty-three.", + "D": "You’ve taken out eight loaves of bread in total so far.", + "E": "In total, you’ve removed three loaves of bread from the oven so far." + }, + "answer_label": "B", + "answer_text": "seventeen.", + "q_id": "P02_115_12", + "correct_answer": "seventeen.", + "original_options": { + "A": "Fourteen.", + "B": "seventeen.", + "C": "Twenty-three.", + "D": "You’ve taken out eight loaves of bread in total so far.", + "E": "In total, you’ve removed three loaves of bread from the oven so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_8", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to cook something in a pot now, where is the pot?", + "answer": "The pot is on the top left of the stove.", + "question_chinese": "如果我现在想用锅子煮点东西，锅在哪里？", + "answer_chinese": "锅在灶台的左上方。", + "evidence": [ + { + "description": "The pot is on the top left of the stove.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the top right of the stove.", + "B": "At the bottom right of the stove.", + "C": "The pot is on the center shelf of the cabinet, not sitting on the stovetop right now.", + "D": "The pot is on the top left of the stove.", + "E": "The pot is beside the frying pan on the counter, rather than on the stove at all." + }, + "answer_label": "D", + "answer_text": "The pot is on the top left of the stove.", + "q_id": "P02_115_8", + "correct_answer": "The pot is on the top left of the stove.", + "original_options": { + "A": "On the top right of the stove.", + "B": "At the bottom right of the stove.", + "C": "The pot is on the center shelf of the cabinet, not sitting on the stovetop right now.", + "D": "The pot is on the top left of the stove.", + "E": "The pot is beside the frying pan on the counter, rather than on the stove at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_7", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to cook something in a pot now, where is the pot?", + "answer": "The pot is on the top left of the stove.", + "question_chinese": "如果我现在想用锅子煮点东西，锅在哪里？", + "answer_chinese": "锅在灶台的左上方。", + "evidence": [ + { + "description": "The pot is on the top left of the stove.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:36" + } + } + ], + "question_timestamps": "00:01:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the bottom right of the stove.", + "B": "On the far left countertop.", + "C": "The pot is on the top left of the stove.", + "D": "The pot is on the middle shelf of the cabinet, not placed on the stove right now.", + "E": "The pot is underneath the frying pan in the cabinet, rather than on the stovetop." + }, + "answer_label": "C", + "answer_text": "The pot is on the top left of the stove.", + "q_id": "P02_115_7", + "correct_answer": "The pot is on the top left of the stove.", + "original_options": { + "A": "On the bottom right of the stove.", + "B": "On the far left countertop.", + "C": "The pot is on the top left of the stove.", + "D": "The pot is on the middle shelf of the cabinet, not placed on the stove right now.", + "E": "The pot is underneath the frying pan in the cabinet, rather than on the stovetop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 17, + 15 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_3_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to cook something in a pot right now, what should I do?", + "answer": "You should turn to your back-left; the pot is above you to the left of the stove in front of you.", + "question_chinese": "如果我现在想用锅煮点东西，该怎么办？", + "answer_chinese": "你应该向左后方转，锅在面前的灶台的左上方。", + "evidence": [ + { + "description": "The pot is on the top left of the stove.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn to your back-right.", + "B": "Turn to your front-left.", + "C": "You should turn to your back-left; the pot is above you to the left of the stove behind you.", + "D": "You should turn to your back-left; the pot is above you to the left of the stove to your right.", + "E": "You should turn to your back-left; the pot is above you to the left of the stove in front of you." + }, + "answer_label": "E", + "answer_text": "You should turn to your back-left; the pot is above you to the left of the stove in front of you.", + "q_id": "P02_115_3_T1", + "correct_answer": "You should turn to your back-left; the pot is above you to the left of the stove in front of you.", + "original_options": { + "A": "Turn to your back-right.", + "B": "Turn to your front-left.", + "C": "You should turn to your back-left; the pot is above you to the left of the stove behind you.", + "D": "You should turn to your back-left; the pot is above you to the left of the stove to your right.", + "E": "You should turn to your back-left; the pot is above you to the left of the stove in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 20 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_3_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to cook something in a pot right now, what should I do?", + "answer": "The pot is above you to the left of the stove in front of you.", + "question_chinese": "如果我现在想用锅煮点东西，该怎么办？", + "answer_chinese": "锅在你面前的灶台的左上方。", + "evidence": [ + { + "description": "The pot is on the top left of the stove.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:36" + } + } + ], + "question_timestamps": "00:01:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The pot is above you to the left of the stove in front of you.", + "B": "The pot is behind you.", + "C": "The pot is to your right.", + "D": "The pot is above you to the left of the stove behind you.", + "E": "The pot is above you to the left of the stove to your right." + }, + "answer_label": "A", + "answer_text": "The pot is above you to the left of the stove in front of you.", + "q_id": "P02_115_3_T2", + "correct_answer": "The pot is above you to the left of the stove in front of you.", + "original_options": { + "A": "The pot is above you to the left of the stove in front of you.", + "B": "The pot is behind you.", + "C": "The pot is to your right.", + "D": "The pot is above you to the left of the stove behind you.", + "E": "The pot is above you to the left of the stove to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 14 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_2_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates did I just take out of the cabinet?", + "answer": "You took out one.", + "question_chinese": "我刚刚从柜子里拿出了几个盘子？", + "answer_chinese": "你拿出了一个。", + "evidence": [ + { + "description": "Took the first plate out of the cupboard.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took out zero.", + "B": "You took out two.", + "C": "You took out one.", + "D": "You took out three plates from the cabinet just now, not only a single plate.", + "E": "You took out four plates from the cabinet just now, instead of taking out one plate." + }, + "answer_label": "C", + "answer_text": "You took out one.", + "q_id": "P02_115_2_T1", + "correct_answer": "You took out one.", + "original_options": { + "A": "You took out zero.", + "B": "You took out two.", + "C": "You took out one.", + "D": "You took out three plates from the cabinet just now, not only a single plate.", + "E": "You took out four plates from the cabinet just now, instead of taking out one plate." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_5", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates did I just take out of the cupboard?", + "answer": "Two.", + "question_chinese": "我刚刚从柜子里拿出了几个盘子？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "Took the first plate out of the cupboard.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:45" + } + }, + { + "description": "Took the second plate out of the cupboard.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:01:26", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You took out one plate from the cupboard, leaving the other plates still stacked inside.", + "D": "You took out three plates from the cupboard, lifting the stack out in one go.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "P02_115_5", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You took out one plate from the cupboard, leaving the other plates still stacked inside.", + "D": "You took out three plates from the cupboard, lifting the stack out in one go.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_2_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates did I just take out of the cabinet?", + "answer": "You took out two.", + "question_chinese": "我刚刚从柜子里拿出了几个盘子？", + "answer_chinese": "你拿出了两个。", + "evidence": [ + { + "description": "Took the first plate out of the cupboard.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:45" + } + }, + { + "description": "Took the second plate out of the cupboard.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:01:41", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took out one.", + "B": "You took out three.", + "C": "You took out zero plates from the cabinet and didn’t remove any dishes at all.", + "D": "You took out two.", + "E": "You took out four plates from the cabinet, pulling out a larger stack together." + }, + "answer_label": "D", + "answer_text": "You took out two.", + "q_id": "P02_115_2_T2", + "correct_answer": "You took out two.", + "original_options": { + "A": "You took out one.", + "B": "You took out three.", + "C": "You took out zero plates from the cabinet and didn’t remove any dishes at all.", + "D": "You took out two.", + "E": "You took out four plates from the cabinet, pulling out a larger stack together." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 14 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_10", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the white oven mitts now?", + "answer": "Yes.", + "question_chinese": "我现在可以够到白色的隔热手套吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The white oven mitts are placed on the left side of the stove, close to your position.", + "timestamps": { + "start": "00:00:38", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:14", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "P02_115_10", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_9", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the white oven mitts now?", + "answer": "Yes.", + "question_chinese": "我现在可以够到白色的隔热手套吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The white oven mitts are placed on the left side of the stove, close to your position.", + "timestamps": { + "start": "00:00:38", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:25", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "P02_115_9", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_4_T2", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the white insulated gloves on the table in front of me right now?", + "answer": "No, right now you can’t reach the white insulated glove on the table in front of you.", + "question_chinese": "我现在可以够到面前桌上的白色的隔热手套吗？", + "answer_chinese": "你现在不可以够到面前桌上的白色的隔热手套。", + "evidence": [ + { + "description": "The white gloves are in front of the oven, two steps away from the sink.", + "timestamps": { + "start": "00:00:38", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, right now you can’t reach the white insulated glove on the table in front of you.", + "B": "Yes, right now you can reach the white insulated glove on the table in front of you." + }, + "answer_label": "A", + "answer_text": "No, right now you can’t reach the white insulated glove on the table in front of you.", + "q_id": "P02_115_4_T2", + "correct_answer": "No, right now you can’t reach the white insulated glove on the table in front of you.", + "original_options": { + "A": "No, right now you can’t reach the white insulated glove on the table in front of you.", + "B": "Yes, right now you can reach the white insulated glove on the table in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_4_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the white insulated gloves on the table in front of me right now?", + "answer": "Yes, right now you can reach the white insulated glove on the table in front of you.", + "question_chinese": "我现在可以够到面前桌上的白色的隔热手套吗？", + "answer_chinese": "你现在可以够到面前桌上的白色的隔热手套。", + "evidence": [ + { + "description": "The white oven mitts are placed on the left side of the stove, close to your position.", + "timestamps": { + "start": "00:00:38", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, right now you can’t reach the white insulated glove on the table in front of you.", + "B": "Yes, right now you can reach the white insulated glove on the table in front of you." + }, + "answer_label": "B", + "answer_text": "Yes, right now you can reach the white insulated glove on the table in front of you.", + "q_id": "P02_115_4_T1", + "correct_answer": "Yes, right now you can reach the white insulated glove on the table in front of you.", + "original_options": { + "A": "No, right now you can’t reach the white insulated glove on the table in front of you.", + "B": "Yes, right now you can reach the white insulated glove on the table in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_1_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine relative to me?", + "answer": "The washing machine is on your right-hand side.", + "question_chinese": "洗衣机在我的哪个方位？", + "answer_chinese": "洗衣机在你的右手边。", + "evidence": [ + { + "description": "When you just entered the kitchen, the washing machine was in the upper left of the field of view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "The washing machine is on your right-hand side.", + "C": "On your left-hand side.", + "D": "The washing machine is directly behind you, not on either side as you’re standing there.", + "E": "The washing machine is slightly in front of you, closer to your left than your right." + }, + "answer_label": "B", + "answer_text": "The washing machine is on your right-hand side.", + "q_id": "P02_115_1_T1", + "correct_answer": "The washing machine is on your right-hand side.", + "original_options": { + "A": "In front of you.", + "B": "The washing machine is on your right-hand side.", + "C": "On your left-hand side.", + "D": "The washing machine is directly behind you, not on either side as you’re standing there.", + "E": "The washing machine is slightly in front of you, closer to your left than your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "P02_115", + "video_length": "00:01:46", + "qid": "P02_115_1_T2", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine relative to me?", + "answer": "The washing machine is to your back-right.", + "question_chinese": "洗衣机在我的哪个方位？", + "answer_chinese": "洗衣机在你的右后方。", + "evidence": [ + { + "description": "When you just entered the kitchen, the washing machine was in the upper left of the field of view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The washing machine is to your back-right.", + "B": "To your back-left.", + "C": "To your front-right.", + "D": "The washing machine is straight to your right, not diagonally behind you from where you are.", + "E": "The washing machine is to your front-left, positioned ahead of you rather than behind your shoulder." + }, + "answer_label": "A", + "answer_text": "The washing machine is to your back-right.", + "q_id": "P02_115_1_T2", + "correct_answer": "The washing machine is to your back-right.", + "original_options": { + "A": "The washing machine is to your back-right.", + "B": "To your back-left.", + "C": "To your front-right.", + "D": "The washing machine is straight to your right, not diagonally behind you from where you are.", + "E": "The washing machine is to your front-left, positioned ahead of you rather than behind your shoulder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + } +] diff --git a/QAs/P02_137-mcq.json b/QAs/P02_137-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2faba7c49672c139515a19ff8d886adca4d19df4 --- /dev/null +++ b/QAs/P02_137-mcq.json @@ -0,0 +1,1202 @@ +[ + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_8_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I opened the refrigerator in total?", + "answer": "Twice, and the refrigerator is in front of you.", + "question_chinese": "到现在为止，我一共打开了几次冰箱？", + "answer_chinese": "两次，冰箱在你面前。", + "evidence": [ + { + "description": "Open the refrigerator for the first time (to take out the fish).", + "timestamps": { + "start": "00:01:24", + "end": "00:01:30" + } + }, + { + "description": "Open the fridge for the second time (get the butter).", + "timestamps": { + "start": "00:01:56", + "end": "00:02:02" + } + } + ], + "question_timestamps": "00:02:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Once, in front of you.", + "B": "Three times, in front of you.", + "C": "Twice, and the refrigerator is in front of you.", + "D": "Twice, but the refrigerator is to your right when you look up from where you are standing.", + "E": "Twice, but the refrigerator is behind you, so you’d need to turn around to face it." + }, + "answer_label": "C", + "answer_text": "Twice, and the refrigerator is in front of you.", + "q_id": "P02_137_8_T1", + "correct_answer": "Twice, and the refrigerator is in front of you.", + "original_options": { + "A": "Once, in front of you.", + "B": "Three times, in front of you.", + "C": "Twice, and the refrigerator is in front of you.", + "D": "Twice, but the refrigerator is to your right when you look up from where you are standing.", + "E": "Twice, but the refrigerator is behind you, so you’d need to turn around to face it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 17 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_15", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the refrigerator so far?", + "answer": "Twice.", + "question_chinese": "到现在为止，我一共打开了几次冰箱？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "Open the refrigerator for the first time (to take out the fish).", + "timestamps": { + "start": "00:01:24", + "end": "00:01:30" + } + }, + { + "description": "Open the fridge for the second time (get the butter).", + "timestamps": { + "start": "00:01:56", + "end": "00:02:02" + } + }, + { + "description": "Open the refrigerator for the third time (take the lemon).", + "timestamps": { + "start": "00:02:17", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:02:23", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times, based on each time you returned and opened it again during the sequence so far.", + "D": "Twice.", + "E": "Five times, counting the initial open plus several later checks back at the refrigerator." + }, + "answer_label": "D", + "answer_text": "Twice.", + "q_id": "P02_137_15", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times, based on each time you returned and opened it again during the sequence so far.", + "D": "Twice.", + "E": "Five times, counting the initial open plus several later checks back at the refrigerator." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 17, + 14 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_8_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I opened the refrigerator in total?", + "answer": "Three times, and the refrigerator is to your right.", + "question_chinese": "到现在为止，我一共打开了几次冰箱？", + "answer_chinese": "三次，冰箱在你右边。", + "evidence": [ + { + "description": "Open the refrigerator for the first time (to take out the fish).", + "timestamps": { + "start": "00:01:24", + "end": "00:01:30" + } + }, + { + "description": "Open the fridge for the second time (get the butter).", + "timestamps": { + "start": "00:01:56", + "end": "00:02:02" + } + }, + { + "description": "Open the refrigerator for the third time (take the lemon).", + "timestamps": { + "start": "00:02:17", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:08:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Two times, on your right.", + "B": "Four times, on your right.", + "C": "Three times, but the refrigerator is in front of you rather than to your right at this moment.", + "D": "Three times, and the refrigerator is to your right.", + "E": "Three times, but the refrigerator is to your left, so you’d turn the opposite way to reach it." + }, + "answer_label": "D", + "answer_text": "Three times, and the refrigerator is to your right.", + "q_id": "P02_137_8_T2", + "correct_answer": "Three times, and the refrigerator is to your right.", + "original_options": { + "A": "Two times, on your right.", + "B": "Four times, on your right.", + "C": "Three times, but the refrigerator is in front of you rather than to your right at this moment.", + "D": "Three times, and the refrigerator is to your right.", + "E": "Three times, but the refrigerator is to your left, so you’d turn the opposite way to reach it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 19 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_13", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many trash cans have I seen?", + "answer": "One.", + "question_chinese": "到现在为止，我一共看到了几个垃圾桶？", + "answer_chinese": "1个。", + "evidence": [ + { + "description": "The first trash can appears.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:16" + } + } + ], + "question_timestamps": "00:02:45", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four, counting the one near you plus three others you passed earlier in the same area.", + "E": "Five, since you encountered multiple bins in different spots while moving around so far." + }, + "answer_label": "B", + "answer_text": "One.", + "q_id": "P02_137_13", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four, counting the one near you plus three others you passed earlier in the same area.", + "E": "Five, since you encountered multiple bins in different spots while moving around so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 14 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_14", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many trash cans have I seen?", + "answer": "Two.", + "question_chinese": "到现在为止，我一共看到了几个垃圾桶？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "The first trash can appears.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:16" + } + }, + { + "description": "The second trash can appears.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:12" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four, including the two you already saw plus two more along the way as you moved through the kitchen.", + "D": "Two.", + "E": "Five, because you passed several different trash cans in separate locations during the events so far." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "P02_137_14", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four, including the two you already saw plus two more along the way as you moved through the kitchen.", + "D": "Two.", + "E": "Five, because you passed several different trash cans in separate locations during the events so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 19, + 16 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_3_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take to go get the fish?", + "answer": "From where you were facing the stovetop, turn around and walk straight 6 steps.", + "question_chinese": "我刚才去拿鱼肉走了什么路线？", + "answer_chinese": "从面向灶台的位置向后转，直走6步。", + "evidence": [ + { + "description": "Turn around from the stove position and walk six steps straight. Take out the fish from the refrigerator on the left.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:11:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Turn around and walk 5 steps.", + "B": "Turn around and walk 7 steps.", + "C": "From where you were facing the stovetop, turn around and walk straight 6 steps.", + "D": "From where you were facing the stovetop, turn around and walk straight 4 steps to reach the fish.", + "E": "From where you were facing the stovetop, turn around and walk straight 8 steps before getting the fish." + }, + "answer_label": "C", + "answer_text": "From where you were facing the stovetop, turn around and walk straight 6 steps.", + "q_id": "P02_137_3_T1", + "correct_answer": "From where you were facing the stovetop, turn around and walk straight 6 steps.", + "original_options": { + "A": "Turn around and walk 5 steps.", + "B": "Turn around and walk 7 steps.", + "C": "From where you were facing the stovetop, turn around and walk straight 6 steps.", + "D": "From where you were facing the stovetop, turn around and walk straight 4 steps to reach the fish.", + "E": "From where you were facing the stovetop, turn around and walk straight 8 steps before getting the fish." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_3_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take to go get the fish?", + "answer": "From facing the stove, turn around and walk straight 6 steps, then take the fish from the refrigerator to your left.", + "question_chinese": "我刚才去拿鱼肉走了什么路线？", + "answer_chinese": "从面向灶台的位置向后转，直走6步，在左边的冰箱中取出鱼肉。", + "evidence": [ + { + "description": "Turn around from the stove position and walk six steps straight. Take out the fish from the refrigerator on the left.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:14:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Turn around, walk 5 steps, left fridge.", + "B": "Turn around, walk 7 steps, left fridge.", + "C": "From facing the stove, turn around and walk straight 4 steps, then take the fish from the refrigerator to your left.", + "D": "From facing the stove, turn around and walk straight 6 steps, then take the fish from the refrigerator to your left.", + "E": "From facing the stove, turn around and walk straight 8 steps, then take the fish from the refrigerator to your left." + }, + "answer_label": "D", + "answer_text": "From facing the stove, turn around and walk straight 6 steps, then take the fish from the refrigerator to your left.", + "q_id": "P02_137_3_T2", + "correct_answer": "From facing the stove, turn around and walk straight 6 steps, then take the fish from the refrigerator to your left.", + "original_options": { + "A": "Turn around, walk 5 steps, left fridge.", + "B": "Turn around, walk 7 steps, left fridge.", + "C": "From facing the stove, turn around and walk straight 4 steps, then take the fish from the refrigerator to your left.", + "D": "From facing the stove, turn around and walk straight 6 steps, then take the fish from the refrigerator to your left.", + "E": "From facing the stove, turn around and walk straight 8 steps, then take the fish from the refrigerator to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 21 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction was the stove when I first entered the kitchen?", + "answer": "When you first enter the kitchen, the stove is on your left.", + "question_chinese": "我刚进厨房时，灶台在我的哪个方向？", + "answer_chinese": "刚进厨房时，灶台在你的左侧。", + "evidence": [ + { + "description": "When you first entered the kitchen, the stove was on the left side of vision.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The stove is straight ahead.", + "B": "When you first enter the kitchen, the stove is on your left.", + "C": "The stove is on your right.", + "D": "When you first enter the kitchen, the stove is behind you, so you’d have to turn around to see it.", + "E": "When you first enter the kitchen, the stove is off to the front-right side rather than on your left." + }, + "answer_label": "B", + "answer_text": "When you first enter the kitchen, the stove is on your left.", + "q_id": "P02_137_1", + "correct_answer": "When you first enter the kitchen, the stove is on your left.", + "original_options": { + "A": "The stove is straight ahead.", + "B": "When you first enter the kitchen, the stove is on your left.", + "C": "The stove is on your right.", + "D": "When you first enter the kitchen, the stove is behind you, so you’d have to turn around to see it.", + "E": "When you first enter the kitchen, the stove is off to the front-right side rather than on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 21, + 19 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction was the stove when I first entered the kitchen?", + "answer": "When you first enter the kitchen, the stove is on your left.", + "question_chinese": "我刚进厨房时，灶台在我的哪个方向？", + "answer_chinese": "刚进厨房时，灶台在你的左侧。", + "evidence": [ + { + "description": "When you first entered the kitchen, the stove was on the left side of vision.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "When you first enter the kitchen, the stove is on your left.", + "B": "It was on your right.", + "C": "It was straight ahead of you.", + "D": "When you first stepped into the kitchen, the stove was behind you rather than to your left.", + "E": "As you entered the kitchen the first time, the stove sat directly in front of you, not on the left." + }, + "answer_label": "A", + "answer_text": "When you first enter the kitchen, the stove is on your left.", + "q_id": "P02_137_2", + "correct_answer": "When you first enter the kitchen, the stove is on your left.", + "original_options": { + "A": "When you first enter the kitchen, the stove is on your left.", + "B": "It was on your right.", + "C": "It was straight ahead of you.", + "D": "When you first stepped into the kitchen, the stove was behind you rather than to your left.", + "E": "As you entered the kitchen the first time, the stove sat directly in front of you, not on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 20 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_11", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was taking the fish out of the fridge, there were some fruits in front of the microwave?", + "answer": "Three.", + "question_chinese": "我在取冰箱里面的鱼肉时，微波炉前面摆了几个水果？", + "answer_chinese": "3个。", + "evidence": [ + { + "description": "Three fruits are placed in front of the microwave.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:25" + } + }, + { + "description": "Three fruits are placed in front of the microwave.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:57" + } + } + ], + "question_timestamps": "00:11:26", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "There were two pieces of fruit sitting in front of the microwave while you took the fish out.", + "D": "You had four fruits lined up in front of the microwave when you opened the fridge for the fish.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "P02_137_11", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "There were two pieces of fruit sitting in front of the microwave while you took the fish out.", + "D": "You had four fruits lined up in front of the microwave when you opened the fridge for the fish.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 19 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_12", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was taking the fish out of the fridge, there were some fruits in front of the microwave?", + "answer": "Three.", + "question_chinese": "我在取冰箱里面的鱼肉时，微波炉前面摆了几个水果？", + "answer_chinese": "3个。", + "evidence": [ + { + "description": "Three fruits are placed in front of the microwave.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:25" + } + }, + { + "description": "Three fruits are placed in front of the microwave.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:57" + } + } + ], + "question_timestamps": "00:18:02", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "One.", + "C": "Two.", + "D": "There were four fruits in front of the microwave as you removed the fish from the refrigerator.", + "E": "You saw six pieces of fruit placed in front of the microwave while taking the fish out." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "P02_137_12", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "One.", + "C": "Two.", + "D": "There were four fruits in front of the microwave as you removed the fish from the refrigerator.", + "E": "You saw six pieces of fruit placed in front of the microwave while taking the fish out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 17 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_6_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many pieces of fruit did I place in front of the microwave?", + "answer": "Three, and the microwave is to your left.", + "question_chinese": "我微波炉前面摆了几个水果？", + "answer_chinese": "3个，微波炉在你左边。", + "evidence": [ + { + "description": "Three fruits are placed in front of the microwave.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:25" + } + }, + { + "description": "Three fruits are placed in front of the microwave.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:57" + } + } + ], + "question_timestamps": "00:01:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, microwave on your left.", + "B": "Three, and the microwave is to your left.", + "C": "Four, microwave on your left.", + "D": "Five pieces of fruit, and the microwave is still positioned to your left side.", + "E": "One piece of fruit, and the microwave remains on your left when you placed it there." + }, + "answer_label": "B", + "answer_text": "Three, and the microwave is to your left.", + "q_id": "P02_137_6_T2", + "correct_answer": "Three, and the microwave is to your left.", + "original_options": { + "A": "Two, microwave on your left.", + "B": "Three, and the microwave is to your left.", + "C": "Four, microwave on your left.", + "D": "Five pieces of fruit, and the microwave is still positioned to your left side.", + "E": "One piece of fruit, and the microwave remains on your left when you placed it there." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_6_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many pieces of fruit did I place in front of the microwave?", + "answer": "Three, and the microwave is to your right.", + "question_chinese": "我微波炉前面摆了几个水果？", + "answer_chinese": "3个，微波炉在你右边。", + "evidence": [ + { + "description": "Three fruits are placed in front of the microwave.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:25" + } + }, + { + "description": "Three fruits are placed in front of the microwave.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:57" + } + } + ], + "question_timestamps": "00:11:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three, and the microwave is to your right.", + "B": "Two, microwave on your right.", + "C": "Four, microwave on your right.", + "D": "Five pieces of fruit, and the microwave is still located to your right as you set them down.", + "E": "One piece of fruit, and the microwave stays on your right side when you placed it there." + }, + "answer_label": "A", + "answer_text": "Three, and the microwave is to your right.", + "q_id": "P02_137_6_T1", + "correct_answer": "Three, and the microwave is to your right.", + "original_options": { + "A": "Three, and the microwave is to your right.", + "B": "Two, microwave on your right.", + "C": "Four, microwave on your right.", + "D": "Five pieces of fruit, and the microwave is still located to your right as you set them down.", + "E": "One piece of fruit, and the microwave stays on your right side when you placed it there." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_2_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where on the stovetop did I last put the white gloves?", + "answer": "The white glove is placed at the upper-left corner of the stove, and the stove is in front of you.", + "question_chinese": "我最后把白色手套放在灶台的哪个地方了？", + "answer_chinese": "白色手套放在灶台左上角，灶台在你正前方", + "evidence": [ + { + "description": "Finally, the white glove was placed in the upper left corner of the stove.", + "timestamps": { + "start": "00:18:27", + "end": "00:18:30" + } + } + ], + "question_timestamps": "00:18:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Upper-right corner, stove in front.", + "B": "The white glove is placed at the upper-left corner of the stove, and the stove is in front of you.", + "C": "Lower-left corner, stove in front.", + "D": "You left the white glove on the lower-right corner of the stove, with the stove directly in front of you.", + "E": "The white glove was set on the center of the stove, and the stove was in front of you." + }, + "answer_label": "B", + "answer_text": "The white glove is placed at the upper-left corner of the stove, and the stove is in front of you.", + "q_id": "P02_137_2_T1", + "correct_answer": "The white glove is placed at the upper-left corner of the stove, and the stove is in front of you.", + "original_options": { + "A": "Upper-right corner, stove in front.", + "B": "The white glove is placed at the upper-left corner of the stove, and the stove is in front of you.", + "C": "Lower-left corner, stove in front.", + "D": "You left the white glove on the lower-right corner of the stove, with the stove directly in front of you.", + "E": "The white glove was set on the center of the stove, and the stove was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 19 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_2_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where on the stovetop did I last put the white gloves?", + "answer": "The white glove is placed at the upper-left corner of the stovetop, and the stovetop is behind you.", + "question_chinese": "我最后把白色手套放在灶台的哪个地方了？", + "answer_chinese": "白色手套放在灶台左上角。，灶台在你后方", + "evidence": [ + { + "description": "Finally, the white glove was placed in the upper left corner of the stove.", + "timestamps": { + "start": "00:18:27", + "end": "00:18:30" + } + } + ], + "question_timestamps": "00:19:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Upper-right corner, stovetop behind you.", + "B": "Lower-left corner, stovetop behind you.", + "C": "The white glove is placed at the upper-left corner of the stovetop, and the stovetop is behind you.", + "D": "You last placed the white glove on the lower-right corner of the stovetop while the stovetop was behind you.", + "E": "The white glove was set on the center of the stovetop, and the stovetop was behind you." + }, + "answer_label": "C", + "answer_text": "The white glove is placed at the upper-left corner of the stovetop, and the stovetop is behind you.", + "q_id": "P02_137_2_T2", + "correct_answer": "The white glove is placed at the upper-left corner of the stovetop, and the stovetop is behind you.", + "original_options": { + "A": "Upper-right corner, stovetop behind you.", + "B": "Lower-left corner, stovetop behind you.", + "C": "The white glove is placed at the upper-left corner of the stovetop, and the stovetop is behind you.", + "D": "You last placed the white glove on the lower-right corner of the stovetop while the stovetop was behind you.", + "E": "The white glove was set on the center of the stovetop, and the stovetop was behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 17 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_5_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the refrigerator door directly from where I am?", + "answer": "No—you can’t. The refrigerator is to your right.", + "question_chinese": "我能直接够到冰箱门吗？", + "answer_chinese": "不能，冰箱在你右边。", + "evidence": [ + { + "description": "The path from the stove to the refrigerator passes through the sink, washing machine, window sill and microwave oven, with the refrigerator next to the microwave oven.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:30" + } + }, + { + "description": "To get from the sink to the refrigerator, you have to go through the washing machine, the windowsill and the microwave, which is next to the microwave.", + "timestamps": { + "start": "00:07:17", + "end": "00:07:25" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. The refrigerator is to your right.", + "B": "No—you can’t. The refrigerator is to your right." + }, + "answer_label": "B", + "answer_text": "No—you can’t. The refrigerator is to your right.", + "q_id": "P02_137_5_T2", + "correct_answer": "No—you can’t. The refrigerator is to your right.", + "original_options": { + "A": "Yes—you can. The refrigerator is to your right.", + "B": "No—you can’t. The refrigerator is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_5_T1", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the refrigerator door directly from where I am?", + "answer": "No—you can’t. The refrigerator is behind you.", + "question_chinese": "我能直接够到冰箱门吗？", + "answer_chinese": "不能，冰箱在你后边。", + "evidence": [ + { + "description": "The path from the stove to the refrigerator passes through the sink, washing machine, window sill and microwave oven, with the refrigerator next to the microwave oven.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:30" + } + }, + { + "description": "To get from the sink to the refrigerator, you have to go through the washing machine, the windowsill and the microwave, which is next to the microwave.", + "timestamps": { + "start": "00:07:17", + "end": "00:07:25" + } + } + ], + "question_timestamps": "00:12:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t. The refrigerator is behind you.", + "B": "Yes—you can. The refrigerator is right in front of you." + }, + "answer_label": "A", + "answer_text": "No—you can’t. The refrigerator is behind you.", + "q_id": "P02_137_5_T1", + "correct_answer": "No—you can’t. The refrigerator is behind you.", + "original_options": { + "A": "No—you can’t. The refrigerator is behind you.", + "B": "Yes—you can. The refrigerator is right in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_4_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I move the fish-packaging box along, from the refrigerator to the trash can?", + "answer": "The fish packaging tray was taken out from the refrigerator shelf and placed on the tabletop next to the stovetop.", + "question_chinese": "我鱼肉包装盒从冰箱到垃圾桶的运动轨迹是怎样的？", + "answer_chinese": "鱼肉包装盒从冰箱货架中取出，并放在灶台旁的桌面上。", + "evidence": [ + { + "description": "Take the fish out of the fridge and place it on a table by the stove.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:33" + } + }, + { + "description": "Tear open the fish packaging.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:10" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved it straight into the trash.", + "B": "You set it by the sink first.", + "C": "You pulled the fish tray from the refrigerator shelf and set it down on the countertop beside the sink.", + "D": "You took the fish packaging tray from the refrigerator door shelf and placed it on the tabletop next to the stovetop.", + "E": "The fish packaging tray was taken out from the refrigerator shelf and placed on the tabletop next to the stovetop." + }, + "answer_label": "E", + "answer_text": "The fish packaging tray was taken out from the refrigerator shelf and placed on the tabletop next to the stovetop.", + "q_id": "P02_137_4_T2", + "correct_answer": "The fish packaging tray was taken out from the refrigerator shelf and placed on the tabletop next to the stovetop.", + "original_options": { + "A": "You moved it straight into the trash.", + "B": "You set it by the sink first.", + "C": "You pulled the fish tray from the refrigerator shelf and set it down on the countertop beside the sink.", + "D": "You took the fish packaging tray from the refrigerator door shelf and placed it on the tabletop next to the stovetop.", + "E": "The fish packaging tray was taken out from the refrigerator shelf and placed on the tabletop next to the stovetop." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 21 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_4_T1", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did I move the fish-packaging box along, from the refrigerator to the trash can?", + "answer": "The fish packaging tray was taken out from the refrigerator shelf and placed on the tabletop next to the stovetop, and in the end it was thrown into the trash can.", + "question_chinese": "我鱼肉包装盒从冰箱到垃圾桶的运动轨迹是怎样的？", + "answer_chinese": "鱼肉包装盒从冰箱货架中取出，并放在灶台旁的桌面上，最后被扔进垃圾桶。", + "evidence": [ + { + "description": "Take the fish out of the fridge and place it on a table by the stove.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:33" + } + }, + { + "description": "Tear open the fish packaging.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:10" + } + }, + { + "description": "Throw the fish packaging into the trash can.", + "timestamps": { + "start": "00:10:54", + "end": "00:11:00" + } + } + ], + "question_timestamps": "00:19:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You tossed it directly into the trash.", + "B": "You set it by the sink first.", + "C": "You removed the fish tray from the refrigerator shelf, placed it on the countertop beside the sink, then threw it away.", + "D": "The fish packaging tray was taken out from the refrigerator shelf and placed on the tabletop next to the stovetop, and in the end it was thrown into the trash can.", + "E": "You took the fish packaging tray from the refrigerator door shelf, set it on the tabletop by the stovetop, then trashed it." + }, + "answer_label": "D", + "answer_text": "The fish packaging tray was taken out from the refrigerator shelf and placed on the tabletop next to the stovetop, and in the end it was thrown into the trash can.", + "q_id": "P02_137_4_T1", + "correct_answer": "The fish packaging tray was taken out from the refrigerator shelf and placed on the tabletop next to the stovetop, and in the end it was thrown into the trash can.", + "original_options": { + "A": "You tossed it directly into the trash.", + "B": "You set it by the sink first.", + "C": "You removed the fish tray from the refrigerator shelf, placed it on the countertop beside the sink, then threw it away.", + "D": "The fish packaging tray was taken out from the refrigerator shelf and placed on the tabletop next to the stovetop, and in the end it was thrown into the trash can.", + "E": "You took the fish packaging tray from the refrigerator door shelf, set it on the tabletop by the stovetop, then trashed it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 22 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_1_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the stovetop relative to me?", + "answer": "When you first entered the kitchen, the stove was on the left side of you.", + "question_chinese": "灶台在我的哪个方向？", + "answer_chinese": "刚进厨房时，灶台在你的左侧。", + "evidence": [ + { + "description": "When you first entered the kitchen, the stove was on the left side of vision.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "When you first entered the kitchen, the stove was on the left side of you.", + "B": "It was to your right.", + "C": "It was behind you.", + "D": "When you first entered the kitchen, the stove was directly in front of you.", + "E": "When you first entered the kitchen, the stove was on the right side of you." + }, + "answer_label": "A", + "answer_text": "When you first entered the kitchen, the stove was on the left side of you.", + "q_id": "P02_137_1_T1", + "correct_answer": "When you first entered the kitchen, the stove was on the left side of you.", + "original_options": { + "A": "When you first entered the kitchen, the stove was on the left side of you.", + "B": "It was to your right.", + "C": "It was behind you.", + "D": "When you first entered the kitchen, the stove was directly in front of you.", + "E": "When you first entered the kitchen, the stove was on the right side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 15 + ] + }, + { + "video_uid": "P02_137", + "video_length": "00:19:26", + "qid": "P02_137_1_T2", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the stovetop relative to me?", + "answer": "The stove is in front of you.", + "question_chinese": "灶台在我的哪个方向？", + "answer_chinese": "灶台在你的正前方。", + "evidence": [ + { + "description": "The stove is right in front of you.", + "timestamps": { + "start": "00:14:55", + "end": "00:15:00" + } + } + ], + "question_timestamps": "00:15:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stove is behind you.", + "B": "The stove is in front of you.", + "C": "The stove is to your left.", + "D": "The stove is to your right, not directly in front of you.", + "E": "The stove is behind you and slightly to your left side, not straight behind you." + }, + "answer_label": "B", + "answer_text": "The stove is in front of you.", + "q_id": "P02_137_1_T2", + "correct_answer": "The stove is in front of you.", + "original_options": { + "A": "The stove is behind you.", + "B": "The stove is in front of you.", + "C": "The stove is to your left.", + "D": "The stove is to your right, not directly in front of you.", + "E": "The stove is behind you and slightly to your left side, not straight behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 15 + ] + } +] diff --git a/QAs/P03_101-mcq.json b/QAs/P03_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c751d433b4317c0eeb989e0ed23564d8153701c6 --- /dev/null +++ b/QAs/P03_101-mcq.json @@ -0,0 +1,1114 @@ +[ + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_6_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was washing dishes just now, how many pots did I wash in total?", + "answer": "You washed one pot on the stove in front of you.", + "question_chinese": "刚才我洗碗时一共洗了几口锅？", + "answer_chinese": "你洗了你面前的灶台上的一口锅。", + "evidence": [ + { + "description": "Washing a pot.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed zero pots.", + "B": "You washed two pots.", + "C": "You washed one pot on the stove in front of you.", + "D": "You washed three pots on the stove in front of you today.", + "E": "You washed four pots on the stove in front of you today." + }, + "answer_label": "C", + "answer_text": "You washed one pot on the stove in front of you.", + "q_id": "P03_101_6_T1", + "correct_answer": "You washed one pot on the stove in front of you.", + "original_options": { + "A": "You washed zero pots.", + "B": "You washed two pots.", + "C": "You washed one pot on the stove in front of you.", + "D": "You washed three pots on the stove in front of you today.", + "E": "You washed four pots on the stove in front of you today." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_13", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I wash the pot when I was washing dishes?", + "answer": "Two.", + "question_chinese": "刚才我洗碗时一共洗了几口锅？", + "answer_chinese": "两口。", + "evidence": [ + { + "description": "Washing a pot.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "I washed the pot four times in total during the dishwashing, counting each scrub-and-rinse cycle.", + "E": "I washed the pot six times altogether, repeating the scrub-and-rinse cycle several times before finishing." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "P03_101_13", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "I washed the pot four times in total during the dishwashing, counting each scrub-and-rinse cycle.", + "E": "I washed the pot six times altogether, repeating the scrub-and-rinse cycle several times before finishing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_6_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was washing dishes just now, how many pots did I wash in total?", + "answer": "You washed two pots on the stove in front of you.", + "question_chinese": "刚才我洗碗时一共洗了几口锅？", + "answer_chinese": "你洗了你面前的灶台上的两口锅。", + "evidence": [ + { + "description": "Washing a pot.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:12:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed one pot.", + "B": "You washed three pots.", + "C": "You washed zero pots on the stove in front of you today.", + "D": "You washed two pots on the stove in front of you.", + "E": "You washed four pots on the stove in front of you today." + }, + "answer_label": "D", + "answer_text": "You washed two pots on the stove in front of you.", + "q_id": "P03_101_6_T2", + "correct_answer": "You washed two pots on the stove in front of you.", + "original_options": { + "A": "You washed one pot.", + "B": "You washed three pots.", + "C": "You washed zero pots on the stove in front of you today.", + "D": "You washed two pots on the stove in front of you.", + "E": "You washed four pots on the stove in front of you today." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_7_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the fridge in total?", + "answer": "You opened it twice, and the fridge was to your left.", + "question_chinese": "我一共打开了几次冰箱？", + "answer_chinese": "你打开了两次，冰箱在你的左边。", + "evidence": [ + { + "description": "The first time opening the fridge.", + "timestamps": { + "start": "00:09:04", + "end": "00:09:12" + } + }, + { + "description": "The second time opening the fridge.", + "timestamps": { + "start": "00:10:15", + "end": "00:10:19" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened it once.", + "B": "You opened it three times.", + "C": "You opened it four times in total, with the fridge positioned to your left the whole time.", + "D": "You opened it twice, and the fridge was to your left.", + "E": "You opened it once overall, and the fridge stayed on your left side as you did it." + }, + "answer_label": "D", + "answer_text": "You opened it twice, and the fridge was to your left.", + "q_id": "P03_101_7_T1", + "correct_answer": "You opened it twice, and the fridge was to your left.", + "original_options": { + "A": "You opened it once.", + "B": "You opened it three times.", + "C": "You opened it four times in total, with the fridge positioned to your left the whole time.", + "D": "You opened it twice, and the fridge was to your left.", + "E": "You opened it once overall, and the fridge stayed on your left side as you did it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_15", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the fridge?", + "answer": "Three times.", + "question_chinese": "我一共打开了几次冰箱？", + "answer_chinese": "3次。", + "evidence": [ + { + "description": "The first time opening the fridge.", + "timestamps": { + "start": "00:09:04", + "end": "00:09:12" + } + }, + { + "description": "The second time opening the fridge.", + "timestamps": { + "start": "00:10:15", + "end": "00:10:19" + } + }, + { + "description": "The third time opening the fridge.", + "timestamps": { + "start": "00:11:35", + "end": "00:11:38" + } + } + ], + "question_timestamps": "00:11:59", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Twice.", + "C": "You opened the fridge four times across the entire sequence, not three.", + "D": "Three times.", + "E": "You opened the fridge five times in total during the video, which is more than three." + }, + "answer_label": "D", + "answer_text": "Three times.", + "q_id": "P03_101_15", + "correct_answer": "Three times.", + "original_options": { + "A": "Once.", + "B": "Twice.", + "C": "You opened the fridge four times across the entire sequence, not three.", + "D": "Three times.", + "E": "You opened the fridge five times in total during the video, which is more than three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 16 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_7_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the fridge in total?", + "answer": "You opened it three times, and the fridge was in front of you.", + "question_chinese": "我一共打开了几次冰箱？", + "answer_chinese": "你打开了3次，冰箱在你的面前。", + "evidence": [ + { + "description": "The first time opening the fridge.", + "timestamps": { + "start": "00:09:04", + "end": "00:09:12" + } + }, + { + "description": "The second time opening the fridge.", + "timestamps": { + "start": "00:10:15", + "end": "00:10:19" + } + }, + { + "description": "The third time opening the fridge.", + "timestamps": { + "start": "00:11:35", + "end": "00:11:38" + } + } + ], + "question_timestamps": "00:14:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened it once.", + "B": "You opened it twice.", + "C": "You opened it four times total, and the fridge was still directly in front of you.", + "D": "You opened it five times in all, with the fridge remaining in front of you each time.", + "E": "You opened it three times, and the fridge was in front of you." + }, + "answer_label": "E", + "answer_text": "You opened it three times, and the fridge was in front of you.", + "q_id": "P03_101_7_T2", + "correct_answer": "You opened it three times, and the fridge was in front of you.", + "original_options": { + "A": "You opened it once.", + "B": "You opened it twice.", + "C": "You opened it four times total, and the fridge was still directly in front of you.", + "D": "You opened it five times in all, with the fridge remaining in front of you each time.", + "E": "You opened it three times, and the fridge was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I just wash the dishes?", + "answer": "You washed the dishes in the sink behind you.", + "question_chinese": "我刚刚在哪里洗的碗？", + "answer_chinese": "在你后面的洗碗池洗的碗。", + "evidence": [ + { + "description": "The table is opposite the sink, and the sink is above the washing machine.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "From the refrigerator to the table.", + "timestamps": { + "start": "00:09:11", + "end": "00:09:14" + } + }, + { + "description": "From the sink to the washing machine in the lower right.", + "timestamps": { + "start": "00:12:02", + "end": "00:12:06" + } + } + ], + "question_timestamps": "00:12:07", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "In the sink in front of you.", + "B": "In the sink to your left.", + "C": "You washed the dishes in the sink behind you.", + "D": "You washed the dishes at the sink to your right, not at the sink behind you.", + "E": "You washed the dishes in the sink beside you, rather than using the sink behind you." + }, + "answer_label": "C", + "answer_text": "You washed the dishes in the sink behind you.", + "q_id": "P03_101_1", + "correct_answer": "You washed the dishes in the sink behind you.", + "original_options": { + "A": "In the sink in front of you.", + "B": "In the sink to your left.", + "C": "You washed the dishes in the sink behind you.", + "D": "You washed the dishes at the sink to your right, not at the sink behind you.", + "E": "You washed the dishes in the sink beside you, rather than using the sink behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 16 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I just wash the dishes?", + "answer": "You washed the dishes in the sink to your upper left.", + "question_chinese": "我刚刚在哪里洗的碗？", + "answer_chinese": "在你左上方的洗碗池洗的碗。", + "evidence": [ + { + "description": "The table is opposite the sink, and the sink is above the washing machine.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "From the refrigerator to the table.", + "timestamps": { + "start": "00:09:11", + "end": "00:09:14" + } + }, + { + "description": "From the sink to the washing machine in the lower right.", + "timestamps": { + "start": "00:12:02", + "end": "00:12:06" + } + } + ], + "question_timestamps": "00:13:45", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "In the sink to your lower left.", + "B": "In the sink to your upper right.", + "C": "You washed the dishes in the sink to your lower right, not in the sink to your upper left.", + "D": "You washed the dishes in the sink to your upper left.", + "E": "You washed the dishes in the sink directly to your left, rather than up and to your left." + }, + "answer_label": "D", + "answer_text": "You washed the dishes in the sink to your upper left.", + "q_id": "P03_101_2", + "correct_answer": "You washed the dishes in the sink to your upper left.", + "original_options": { + "A": "In the sink to your lower left.", + "B": "In the sink to your upper right.", + "C": "You washed the dishes in the sink to your lower right, not in the sink to your upper left.", + "D": "You washed the dishes in the sink to your upper left.", + "E": "You washed the dishes in the sink directly to your left, rather than up and to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 18 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_2_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I first entered the kitchen, which direction was the sink from me?", + "answer": "When you first entered the kitchen, the sink was in front of you.", + "question_chinese": "我刚进厨房时，洗碗池在我的哪个方向？", + "answer_chinese": "刚进厨房时，洗碗池就在你的面前。", + "evidence": [ + { + "description": "When you first entered the kitchen, the sink was on the right side of vision.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "When you first entered the kitchen, the sink was in front of you.", + "B": "The sink was behind you.", + "C": "The sink was to your left.", + "D": "When you first entered the kitchen, the sink was to your right rather than directly in front of you.", + "E": "When you first entered the kitchen, the sink was above you, not directly in front of you as you walked in." + }, + "answer_label": "A", + "answer_text": "When you first entered the kitchen, the sink was in front of you.", + "q_id": "P03_101_2_T1", + "correct_answer": "When you first entered the kitchen, the sink was in front of you.", + "original_options": { + "A": "When you first entered the kitchen, the sink was in front of you.", + "B": "The sink was behind you.", + "C": "The sink was to your left.", + "D": "When you first entered the kitchen, the sink was to your right rather than directly in front of you.", + "E": "When you first entered the kitchen, the sink was above you, not directly in front of you as you walked in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 21 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_3", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction was the sink when I first entered the kitchen?", + "answer": "When you first enter the kitchen, the sink is on your right.", + "question_chinese": "我刚进厨房时，洗碗池在我的哪个方向？", + "answer_chinese": "刚进厨房时，洗碗池在你的右侧。", + "evidence": [ + { + "description": "When you first entered the kitchen, the sink was on the right side of vision.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:09:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The sink is on your left.", + "B": "The sink is in front of you.", + "C": "When you first enter the kitchen, the sink is behind you instead of on your right.", + "D": "When you first enter the kitchen, the sink is to your left side, not on your right.", + "E": "When you first enter the kitchen, the sink is on your right." + }, + "answer_label": "E", + "answer_text": "When you first enter the kitchen, the sink is on your right.", + "q_id": "P03_101_3", + "correct_answer": "When you first enter the kitchen, the sink is on your right.", + "original_options": { + "A": "The sink is on your left.", + "B": "The sink is in front of you.", + "C": "When you first enter the kitchen, the sink is behind you instead of on your right.", + "D": "When you first enter the kitchen, the sink is to your left side, not on your right.", + "E": "When you first enter the kitchen, the sink is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_2_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I first entered the kitchen, which direction was the sink from me?", + "answer": "When you first entered the kitchen, the sink was on your right side.", + "question_chinese": "我刚进厨房时，洗碗池在我的哪个方向？", + "answer_chinese": "刚进厨房时，洗碗池在你的右侧。", + "evidence": [ + { + "description": "When you first entered the kitchen, the sink was on the right side of vision.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:14:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sink was on your left side.", + "B": "The sink was behind you.", + "C": "When you first entered the kitchen, the sink was in front of you rather than on your right side.", + "D": "When you first entered the kitchen, the sink was behind you, not positioned on your right side.", + "E": "When you first entered the kitchen, the sink was on your right side." + }, + "answer_label": "E", + "answer_text": "When you first entered the kitchen, the sink was on your right side.", + "q_id": "P03_101_2_T2", + "correct_answer": "When you first entered the kitchen, the sink was on your right side.", + "original_options": { + "A": "The sink was on your left side.", + "B": "The sink was behind you.", + "C": "When you first entered the kitchen, the sink was in front of you rather than on your right side.", + "D": "When you first entered the kitchen, the sink was behind you, not positioned on your right side.", + "E": "When you first entered the kitchen, the sink was on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 19, + 17 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_5_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I first entered the kitchen, how many items were placed inside the sink?", + "answer": "After you took away the green spoon in front of you, there were only two objects left.", + "question_chinese": "我刚进厨房时，洗碗池里面放着几个物品？", + "answer_chinese": "你拿走了你面前的绿勺子后只剩下两个物体。", + "evidence": [ + { + "description": "When you first entered the kitchen, there was a white square basin, a green ladle and a fork in the sink.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:01:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One item.", + "B": "After you took away the green spoon in front of you, there were only two objects left.", + "C": "Three items.", + "D": "After removing the green spoon, four objects remained in the sink basin.", + "E": "After you took away the green spoon, there were no objects left inside the sink." + }, + "answer_label": "B", + "answer_text": "After you took away the green spoon in front of you, there were only two objects left.", + "q_id": "P03_101_5_T1", + "correct_answer": "After you took away the green spoon in front of you, there were only two objects left.", + "original_options": { + "A": "One item.", + "B": "After you took away the green spoon in front of you, there were only two objects left.", + "C": "Three items.", + "D": "After removing the green spoon, four objects remained in the sink basin.", + "E": "After you took away the green spoon, there were no objects left inside the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 15 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_11", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I first walked into the kitchen, how many items were in the sink?", + "answer": "Three.", + "question_chinese": "我刚进厨房时，洗碗池里面放着几个物品？", + "answer_chinese": "3个。", + "evidence": [ + { + "description": "When you first entered the kitchen, there was a white square basin, a green ladle and a fork in the sink.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:08:45", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Four.", + "C": "Three.", + "D": "Six items were in the sink when you first entered the kitchen.", + "E": "Eight items were sitting in the sink as you walked into the kitchen." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "P03_101_11", + "correct_answer": "Three.", + "original_options": { + "A": "One.", + "B": "Four.", + "C": "Three.", + "D": "Six items were in the sink when you first entered the kitchen.", + "E": "Eight items were sitting in the sink as you walked into the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_5_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I first entered the kitchen, how many items were placed inside the sink?", + "answer": "There were three objects inside the sink in front of you.", + "question_chinese": "我刚进厨房时，洗碗池里面放着几个物品？", + "answer_chinese": "你面前的洗碗池里面放着3个。", + "evidence": [ + { + "description": "When you first entered the kitchen, there was a white square basin, a green ladle and a fork in the sink.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:14:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There were two objects inside the sink.", + "B": "There were four objects inside the sink.", + "C": "There were three objects inside the sink in front of you.", + "D": "There was only one object inside the sink in front of you.", + "E": "There were five objects inside the sink in front of you when you entered." + }, + "answer_label": "C", + "answer_text": "There were three objects inside the sink in front of you.", + "q_id": "P03_101_5_T2", + "correct_answer": "There were three objects inside the sink in front of you.", + "original_options": { + "A": "There were two objects inside the sink.", + "B": "There were four objects inside the sink.", + "C": "There were three objects inside the sink in front of you.", + "D": "There was only one object inside the sink in front of you.", + "E": "There were five objects inside the sink in front of you when you entered." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 14 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_3_T2", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Before I started washing dishes, what was the movement trajectory of the green colander?", + "answer": "The green colander was moved from the white plate in the sink to the white plate on the left side of the sink, and then moved from the plate into the sink basin.", + "question_chinese": "我在开始洗碗前，绿色漏勺的运动轨迹是怎样的？", + "answer_chinese": "绿色漏勺从洗碗池中的白色盘子里被放在洗碗池左侧的白色盘子上，然后再从盘子里放进了洗碗槽。", + "evidence": [ + { + "description": "The green strainer sits on a white plate in the sink.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "The green funnel was placed on a white plate on the left side of the sink.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + }, + { + "description": "It was placed from the plate into the sink.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:01" + } + } + ], + "question_timestamps": "00:03:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The green colander was moved from the white plate in the sink to the white plate on the left side of the sink, and then moved from the plate into the sink basin.", + "B": "Sink plate to right plate.", + "C": "From sink basin to left plate.", + "D": "The green colander moved from the left-side plate back onto the plate in the sink.", + "E": "The green colander was moved from the white plate in the sink to the right-side plate, then back onto that plate." + }, + "answer_label": "A", + "answer_text": "The green colander was moved from the white plate in the sink to the white plate on the left side of the sink, and then moved from the plate into the sink basin.", + "q_id": "P03_101_3_T2", + "correct_answer": "The green colander was moved from the white plate in the sink to the white plate on the left side of the sink, and then moved from the plate into the sink basin.", + "original_options": { + "A": "The green colander was moved from the white plate in the sink to the white plate on the left side of the sink, and then moved from the plate into the sink basin.", + "B": "Sink plate to right plate.", + "C": "From sink basin to left plate.", + "D": "The green colander moved from the left-side plate back onto the plate in the sink.", + "E": "The green colander was moved from the white plate in the sink to the right-side plate, then back onto that plate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 21 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_3_T1", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Before I started washing dishes, what was the movement trajectory of the green colander?", + "answer": "The green colander was moved from the white plate in the sink to the white plate on the left side of the sink.", + "question_chinese": "我在开始洗碗前，绿色漏勺的运动轨迹是怎样的？", + "answer_chinese": "绿色漏勺从洗碗池中的白色盘子里被放在洗碗池左侧的白色盘子上。", + "evidence": [ + { + "description": "The green strainer sits on a white plate in the sink.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "The green funnel was placed on a white plate on the left side of the sink.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:09:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Moved to the right side.", + "B": "Moved to the back plate.", + "C": "The green colander was moved from the white plate in the sink to the white plate on the left side of the sink.", + "D": "The green colander was moved from the white plate in the sink to the white plate behind you.", + "E": "The green colander was moved from the white plate in the sink to a white plate on the front side." + }, + "answer_label": "C", + "answer_text": "The green colander was moved from the white plate in the sink to the white plate on the left side of the sink.", + "q_id": "P03_101_3_T1", + "correct_answer": "The green colander was moved from the white plate in the sink to the white plate on the left side of the sink.", + "original_options": { + "A": "Moved to the right side.", + "B": "Moved to the back plate.", + "C": "The green colander was moved from the white plate in the sink to the white plate on the left side of the sink.", + "D": "The green colander was moved from the white plate in the sink to the white plate behind you.", + "E": "The green colander was moved from the white plate in the sink to a white plate on the front side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 20 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_4_T1", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was washing dishes next to the sink, how far away was the stove from me?", + "answer": "The stove was about one step to your left.", + "question_chinese": "我在洗碗池旁边洗碗时，灶台离我有多远？", + "answer_chinese": "灶台在你的左边大约一步远。", + "evidence": [ + { + "description": "The stove is about two steps to the left of the sink.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "From the sink to the stove.", + "timestamps": { + "start": "00:05:42", + "end": "00:05:45" + } + } + ], + "question_timestamps": "00:8:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "About one step to your right.", + "B": "About one step in front of you.", + "C": "The stove was about two steps to your left from where you were washing dishes.", + "D": "The stove was several steps away, positioned behind you while you stood at the sink.", + "E": "The stove was about one step to your left." + }, + "answer_label": "E", + "answer_text": "The stove was about one step to your left.", + "q_id": "P03_101_4_T1", + "correct_answer": "The stove was about one step to your left.", + "original_options": { + "A": "About one step to your right.", + "B": "About one step in front of you.", + "C": "The stove was about two steps to your left from where you were washing dishes.", + "D": "The stove was several steps away, positioned behind you while you stood at the sink.", + "E": "The stove was about one step to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 15 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_9", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How far away was the stove when I was washing dishes by the sink?", + "answer": "The stove is about two steps to your left.", + "question_chinese": "我在洗碗池旁边洗碗时，灶台离我有多远？", + "answer_chinese": "灶台在你的左边大约两步远。", + "evidence": [ + { + "description": "The stove is about two steps to the left of the sink.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "From the sink to the stove.", + "timestamps": { + "start": "00:05:42", + "end": "00:05:45" + } + } + ], + "question_timestamps": "00:10:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "About one step to your left.", + "B": "About two steps to your right.", + "C": "The stove is about four steps in front of you while you are washing dishes by the sink.", + "D": "The stove is about three steps to your right from where you are standing at the sink.", + "E": "The stove is about two steps to your left." + }, + "answer_label": "E", + "answer_text": "The stove is about two steps to your left.", + "q_id": "P03_101_9", + "correct_answer": "The stove is about two steps to your left.", + "original_options": { + "A": "About one step to your left.", + "B": "About two steps to your right.", + "C": "The stove is about four steps in front of you while you are washing dishes by the sink.", + "D": "The stove is about three steps to your right from where you are standing at the sink.", + "E": "The stove is about two steps to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "P03_101", + "video_length": "00:14:12", + "qid": "P03_101_4_T2", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was washing dishes next to the sink, how far away was the stove from me?", + "answer": "The stove was about two steps to your left.", + "question_chinese": "我在洗碗池旁边洗碗时，灶台离我有多远？", + "answer_chinese": "灶台在你的左边大约两步远。", + "evidence": [ + { + "description": "The stove is about two steps to the left of the sink.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "From the sink to the stove.", + "timestamps": { + "start": "00:05:42", + "end": "00:05:45" + } + } + ], + "question_timestamps": "00:14:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stove was about two steps to your left.", + "B": "About two steps to your right.", + "C": "About two steps in front of you.", + "D": "The stove was about one step to your left while you were washing dishes next to the sink.", + "E": "The stove was about two steps behind you from where you were standing at the sink." + }, + "answer_label": "A", + "answer_text": "The stove was about two steps to your left.", + "q_id": "P03_101_4_T2", + "correct_answer": "The stove was about two steps to your left.", + "original_options": { + "A": "The stove was about two steps to your left.", + "B": "About two steps to your right.", + "C": "About two steps in front of you.", + "D": "The stove was about one step to your left while you were washing dishes next to the sink.", + "E": "The stove was about two steps behind you from where you were standing at the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 16 + ] + } +] diff --git a/QAs/P03_115-mcq.json b/QAs/P03_115-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2e0341399bd1280ea8f9799de4a958999e1ecfb9 --- /dev/null +++ b/QAs/P03_115-mcq.json @@ -0,0 +1,581 @@ +[ + { + "video_uid": "P03_115", + "video_length": "00:01:46", + "qid": "P03_115_3_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the fridge in total?", + "answer": "Once. The refrigerator is directly to your right.", + "question_chinese": "我一共打开了几次冰箱？", + "answer_chinese": "1次。冰箱在你正右方。", + "evidence": [ + { + "description": "Opened the fridge for the first time.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once. The refrigerator is directly to your right.", + "B": "Twice, to your right.", + "C": "Three times, to your right.", + "D": "Twice. You opened the refrigerator two separate times while it stayed directly to your right.", + "E": "Three times. You opened the refrigerator three separate times, and it was directly to your right." + }, + "answer_label": "A", + "answer_text": "Once. The refrigerator is directly to your right.", + "q_id": "P03_115_3_T2", + "correct_answer": "Once. The refrigerator is directly to your right.", + "original_options": { + "A": "Once. The refrigerator is directly to your right.", + "B": "Twice, to your right.", + "C": "Three times, to your right.", + "D": "Twice. You opened the refrigerator two separate times while it stayed directly to your right.", + "E": "Three times. You opened the refrigerator three separate times, and it was directly to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 16 + ] + }, + { + "video_uid": "P03_115", + "video_length": "00:01:46", + "qid": "P03_115_3_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the fridge in total?", + "answer": "Twice. The refrigerator is to your back-right.", + "question_chinese": "我一共打开了几次冰箱？", + "answer_chinese": "两次。冰箱在你右后方。", + "evidence": [ + { + "description": "Opened the fridge for the first time.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + }, + { + "description": "Opened the fridge for the second time.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once. The refrigerator is to your back-right.", + "B": "Three times. The refrigerator is to your back-right.", + "C": "Once. You only opened the refrigerator a single time, and it was positioned to your back-right.", + "D": "Three times. You opened the refrigerator three times overall, with it located to your back-right.", + "E": "Twice. The refrigerator is to your back-right." + }, + "answer_label": "E", + "answer_text": "Twice. The refrigerator is to your back-right.", + "q_id": "P03_115_3_T1", + "correct_answer": "Twice. The refrigerator is to your back-right.", + "original_options": { + "A": "Once. The refrigerator is to your back-right.", + "B": "Three times. The refrigerator is to your back-right.", + "C": "Once. You only opened the refrigerator a single time, and it was positioned to your back-right.", + "D": "Three times. You opened the refrigerator three times overall, with it located to your back-right.", + "E": "Twice. The refrigerator is to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 15 + ] + }, + { + "video_uid": "P03_115", + "video_length": "00:01:46", + "qid": "P03_115_4_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different kinds of items did I take out of the refrigerator?", + "answer": "One type. The refrigerator is directly to your right.", + "question_chinese": "我从冰箱里面取了几种物品？", + "answer_chinese": "一种。冰箱在你正右方。", + "evidence": [ + { + "description": "Opened the fridge and took out a bottle of Coke.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One type. The refrigerator is directly to your right.", + "B": "Zero types, to your right.", + "C": "Two types, to your right.", + "D": "Three types. You took out three different kinds of items from the refrigerator directly to your right.", + "E": "Four types. You removed four distinct kinds of items from the refrigerator directly to your right." + }, + "answer_label": "A", + "answer_text": "One type. The refrigerator is directly to your right.", + "q_id": "P03_115_4_T1", + "correct_answer": "One type. The refrigerator is directly to your right.", + "original_options": { + "A": "One type. The refrigerator is directly to your right.", + "B": "Zero types, to your right.", + "C": "Two types, to your right.", + "D": "Three types. You took out three different kinds of items from the refrigerator directly to your right.", + "E": "Four types. You removed four distinct kinds of items from the refrigerator directly to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "P03_115", + "video_length": "00:01:46", + "qid": "P03_115_4_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different kinds of items did I take out of the refrigerator?", + "answer": "One type. The refrigerator is to your back-right.", + "question_chinese": "我从冰箱里面取了几种物品？", + "answer_chinese": "一种。冰箱在你右后方。", + "evidence": [ + { + "description": "Opened the fridge and took out a bottle of Coke.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + }, + { + "description": "Opened the fridge, but only put the Coke back.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero types. The refrigerator is to your back-right.", + "B": "One type. The refrigerator is to your back-right.", + "C": "Two types. The refrigerator is to your back-right.", + "D": "Three types. You took out three different kinds of items from the refrigerator to your back-right.", + "E": "Four types. You removed four distinct kinds of items from the refrigerator positioned to your back-right." + }, + "answer_label": "B", + "answer_text": "One type. The refrigerator is to your back-right.", + "q_id": "P03_115_4_T2", + "correct_answer": "One type. The refrigerator is to your back-right.", + "original_options": { + "A": "Zero types. The refrigerator is to your back-right.", + "B": "One type. The refrigerator is to your back-right.", + "C": "Two types. The refrigerator is to your back-right.", + "D": "Three types. You took out three different kinds of items from the refrigerator to your back-right.", + "E": "Four types. You removed four distinct kinds of items from the refrigerator positioned to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 16 + ] + }, + { + "video_uid": "P03_115", + "video_length": "00:01:46", + "qid": "P03_115_5_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path have I walked from when I entered the door up to now?", + "answer": "You first walked nearby the refrigerator, then nearby the sink, then back nearby the refrigerator, and finally returned nearby the sink.", + "question_chinese": "我从进门到现在走过的路径是怎样的？", + "answer_chinese": "首先走到冰箱附近，然后走到洗碗池附近，再次走到冰箱附近，最后回到洗碗池附近。", + "evidence": [ + { + "description": "Walked to the vicinity of the fridge.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "Walked to the vicinity of the sink.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + }, + { + "description": "Walked to the vicinity of the fridge again.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + }, + { + "description": "Returned to the vicinity of the sink.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Sink, refrigerator, sink, refrigerator.", + "B": "You first walked nearby the refrigerator, then nearby the sink, then back nearby the refrigerator, and finally returned nearby the sink.", + "C": "Refrigerator, sink, sink, refrigerator.", + "D": "You first walked nearby the sink, then nearby the refrigerator, then back nearby the sink, and finally returned nearby the refrigerator.", + "E": "You first walked nearby the refrigerator, then back nearby the refrigerator, then nearby the sink, and finally returned nearby the sink." + }, + "answer_label": "B", + "answer_text": "You first walked nearby the refrigerator, then nearby the sink, then back nearby the refrigerator, and finally returned nearby the sink.", + "q_id": "P03_115_5_T1", + "correct_answer": "You first walked nearby the refrigerator, then nearby the sink, then back nearby the refrigerator, and finally returned nearby the sink.", + "original_options": { + "A": "Sink, refrigerator, sink, refrigerator.", + "B": "You first walked nearby the refrigerator, then nearby the sink, then back nearby the refrigerator, and finally returned nearby the sink.", + "C": "Refrigerator, sink, sink, refrigerator.", + "D": "You first walked nearby the sink, then nearby the refrigerator, then back nearby the sink, and finally returned nearby the refrigerator.", + "E": "You first walked nearby the refrigerator, then back nearby the refrigerator, then nearby the sink, and finally returned nearby the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 21, + 21 + ] + }, + { + "video_uid": "P03_115", + "video_length": "00:01:46", + "qid": "P03_115_5_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path have I walked from when I entered the door up to now?", + "answer": "You first walked nearby the refrigerator on your right, then nearby the sink, then back nearby the refrigerator, and finally returned nearby the sink.", + "question_chinese": "我从进门到现在走过的路径是怎样的？", + "answer_chinese": "首先走到右方的冰箱附近，然后走到洗碗池附近，再次走到冰箱附近，最后回到洗碗池附近。", + "evidence": [ + { + "description": "Walked to the vicinity of the fridge.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "Walked to the vicinity of the sink.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "Returned to the vicinity of the sink.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went sink, fridge, sink, fridge.", + "B": "You walked near the refrigerator on your left.", + "C": "You first walked nearby the refrigerator on your right, then nearby the sink, then back nearby the refrigerator, and finally returned nearby the sink.", + "D": "You first walked nearby the sink, then nearby the refrigerator on your right, then returned to the sink again.", + "E": "You first walked nearby the refrigerator on your right, then stayed by the sink, and never went back to the refrigerator." + }, + "answer_label": "C", + "answer_text": "You first walked nearby the refrigerator on your right, then nearby the sink, then back nearby the refrigerator, and finally returned nearby the sink.", + "q_id": "P03_115_5_T2", + "correct_answer": "You first walked nearby the refrigerator on your right, then nearby the sink, then back nearby the refrigerator, and finally returned nearby the sink.", + "original_options": { + "A": "You went sink, fridge, sink, fridge.", + "B": "You walked near the refrigerator on your left.", + "C": "You first walked nearby the refrigerator on your right, then nearby the sink, then back nearby the refrigerator, and finally returned nearby the sink.", + "D": "You first walked nearby the sink, then nearby the refrigerator on your right, then returned to the sink again.", + "E": "You first walked nearby the refrigerator on your right, then stayed by the sink, and never went back to the refrigerator." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 19, + 21 + ] + }, + { + "video_uid": "P03_115", + "video_length": "00:01:46", + "qid": "P03_115_2_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to cook with a frying pan—where is the frying pan?", + "answer": "The frying pan is inside the sink, to your front-left.", + "question_chinese": "我想用平底锅做饭，平底锅在哪里？", + "answer_chinese": "平底锅在洗碗池里面，在你的左前方。", + "evidence": [ + { + "description": "The frying pan is in the sink.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Inside the sink, front-right.", + "B": "Inside the sink, back-left.", + "C": "The frying pan is outside the sink, positioned to your front-left near the sink area.", + "D": "The frying pan is inside the sink, but it sits behind you rather than to your front-left.", + "E": "The frying pan is inside the sink, to your front-left." + }, + "answer_label": "E", + "answer_text": "The frying pan is inside the sink, to your front-left.", + "q_id": "P03_115_2_T2", + "correct_answer": "The frying pan is inside the sink, to your front-left.", + "original_options": { + "A": "Inside the sink, front-right.", + "B": "Inside the sink, back-left.", + "C": "The frying pan is outside the sink, positioned to your front-left near the sink area.", + "D": "The frying pan is inside the sink, but it sits behind you rather than to your front-left.", + "E": "The frying pan is inside the sink, to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "P03_115", + "video_length": "00:01:46", + "qid": "P03_115_2_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to cook with a frying pan—where is the frying pan?", + "answer": "The frying pan is inside the sink, in front of you.", + "question_chinese": "我想用平底锅做饭，平底锅在哪里？", + "answer_chinese": "平底锅在洗碗池里面，在你的正前方。", + "evidence": [ + { + "description": "The frying pan is in the sink.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Inside the sink, behind you.", + "B": "Outside the sink, in front.", + "C": "The frying pan is inside the sink, but it is positioned to your front-left instead of in front of you.", + "D": "The frying pan is inside the sink, in front of you.", + "E": "The frying pan is inside the sink, but it sits to your front-right rather than directly in front of you." + }, + "answer_label": "D", + "answer_text": "The frying pan is inside the sink, in front of you.", + "q_id": "P03_115_2_T1", + "correct_answer": "The frying pan is inside the sink, in front of you.", + "original_options": { + "A": "Inside the sink, behind you.", + "B": "Outside the sink, in front.", + "C": "The frying pan is inside the sink, but it is positioned to your front-left instead of in front of you.", + "D": "The frying pan is inside the sink, in front of you.", + "E": "The frying pan is inside the sink, but it sits to your front-right rather than directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 20 + ] + }, + { + "video_uid": "P03_115", + "video_length": "00:00:22", + "qid": "P03_115_1_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine relative to me?", + "answer": "The washing machine is to your front-right.", + "question_chinese": "洗衣机在我的哪个方位？", + "answer_chinese": "洗衣机在你的右前方。", + "evidence": [ + { + "description": "When I first entered the door, the washing machine was in the front right of my field of view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The washing machine is to your front-left.", + "B": "The washing machine is to your back-right.", + "C": "The washing machine is to your front-right.", + "D": "The washing machine is to your back-left, slightly behind your shoulder as you face forward.", + "E": "The washing machine is directly in front of you, not off to the right side." + }, + "answer_label": "C", + "answer_text": "The washing machine is to your front-right.", + "q_id": "P03_115_1_T1", + "correct_answer": "The washing machine is to your front-right.", + "original_options": { + "A": "The washing machine is to your front-left.", + "B": "The washing machine is to your back-right.", + "C": "The washing machine is to your front-right.", + "D": "The washing machine is to your back-left, slightly behind your shoulder as you face forward.", + "E": "The washing machine is directly in front of you, not off to the right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "P03_115", + "video_length": "00:00:22", + "qid": "P03_115_1_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine relative to me?", + "answer": "The washing machine is to your back-right.", + "question_chinese": "洗衣机在我的哪个方位？", + "answer_chinese": "洗衣机在你的右后方。", + "evidence": [ + { + "description": "When I first entered the door, the washing machine was in the front right of my field of view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The washing machine is to your back-right.", + "B": "The washing machine is to your front-right.", + "C": "The washing machine is to your back-left.", + "D": "The washing machine is directly behind you, not offset to the right side.", + "E": "The washing machine is on your front-left, across your body from your right shoulder." + }, + "answer_label": "A", + "answer_text": "The washing machine is to your back-right.", + "q_id": "P03_115_1_T2", + "correct_answer": "The washing machine is to your back-right.", + "original_options": { + "A": "The washing machine is to your back-right.", + "B": "The washing machine is to your front-right.", + "C": "The washing machine is to your back-left.", + "D": "The washing machine is directly behind you, not offset to the right side.", + "E": "The washing machine is on your front-left, across your body from your right shoulder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 14 + ] + } +] diff --git a/QAs/P03_117-mcq.json b/QAs/P03_117-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..17104755fb1408a6f7a52b1f6822c1942b3c1538 --- /dev/null +++ b/QAs/P03_117-mcq.json @@ -0,0 +1,930 @@ +[ + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_13", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can?", + "answer": "The trash can is opposite the washing machine.", + "question_chinese": "垃圾桶在哪个位置？", + "answer_chinese": "垃圾桶在洗衣机的对面。", + "evidence": [ + { + "description": "The trash can is opposite the washing machine.", + "timestamps": { + "start": "00:08:47", + "end": "00:08:55" + } + } + ], + "question_timestamps": "00:09:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The trash can is beside the refrigerator.", + "B": "The trash can is behind the door.", + "C": "The trash can is next to the washing machine, not positioned across from it.", + "D": "The trash can is in front of the washing machine, rather than on the opposite side.", + "E": "The trash can is opposite the washing machine." + }, + "answer_label": "E", + "answer_text": "The trash can is opposite the washing machine.", + "q_id": "P03_117_13", + "correct_answer": "The trash can is opposite the washing machine.", + "original_options": { + "A": "The trash can is beside the refrigerator.", + "B": "The trash can is behind the door.", + "C": "The trash can is next to the washing machine, not positioned across from it.", + "D": "The trash can is in front of the washing machine, rather than on the opposite side.", + "E": "The trash can is opposite the washing machine." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 16 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_14", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can?", + "answer": "The trash can is opposite the washing machine.", + "question_chinese": "垃圾桶在哪个位置？", + "answer_chinese": "垃圾桶在洗衣机的对面。", + "evidence": [ + { + "description": "The trash can is opposite the washing machine.", + "timestamps": { + "start": "00:08:47", + "end": "00:08:55" + } + } + ], + "question_timestamps": "00:11:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The trash can is next to the refrigerator.", + "B": "The trash can is beside the entryway.", + "C": "The trash can is opposite the washing machine.", + "D": "The trash can is positioned next to the washing machine rather than across from it.", + "E": "The trash can is located behind the couch, away from the washing machine’s line of sight." + }, + "answer_label": "C", + "answer_text": "The trash can is opposite the washing machine.", + "q_id": "P03_117_14", + "correct_answer": "The trash can is opposite the washing machine.", + "original_options": { + "A": "The trash can is next to the refrigerator.", + "B": "The trash can is beside the entryway.", + "C": "The trash can is opposite the washing machine.", + "D": "The trash can is positioned next to the washing machine rather than across from it.", + "E": "The trash can is located behind the couch, away from the washing machine’s line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 17 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the fridge in total?", + "answer": "three times.", + "question_chinese": "我一共打开了几次冰箱？", + "answer_chinese": "3次。", + "evidence": [ + { + "description": "Opened the fridge for the first time.", + "timestamps": { + "start": "00:09:20", + "end": "00:09:28" + } + }, + { + "description": "Opened the fridge for the second time.", + "timestamps": { + "start": "00:10:06", + "end": "00:10:18" + } + }, + { + "description": "Opened the fridge for the third time.", + "timestamps": { + "start": "00:11:50", + "end": "00:11:53" + } + } + ], + "question_timestamps": "00:11:56", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "three times.", + "C": "Twice.", + "D": "You opened the fridge four times total over the entire sequence of actions.", + "E": "You opened the fridge five times in total, including quick checks and longer openings." + }, + "answer_label": "B", + "answer_text": "three times.", + "q_id": "P03_117_1", + "correct_answer": "three times.", + "original_options": { + "A": "Once.", + "B": "three times.", + "C": "Twice.", + "D": "You opened the fridge four times total over the entire sequence of actions.", + "E": "You opened the fridge five times in total, including quick checks and longer openings." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the fridge in total?", + "answer": "three times.", + "question_chinese": "我一共打开了几次冰箱？", + "answer_chinese": "3次。", + "evidence": [ + { + "description": "Opened the fridge for the first time.", + "timestamps": { + "start": "00:09:20", + "end": "00:09:28" + } + }, + { + "description": "Opened the fridge for the second time.", + "timestamps": { + "start": "00:10:06", + "end": "00:10:18" + } + }, + { + "description": "Opened the fridge for the third time.", + "timestamps": { + "start": "00:11:50", + "end": "00:11:53" + } + } + ], + "question_timestamps": "00:12:01", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Twice.", + "C": "three times.", + "D": "You opened the fridge four times in total throughout the entire clip.", + "E": "You opened the fridge five times overall, counting every separate opening action." + }, + "answer_label": "C", + "answer_text": "three times.", + "q_id": "P03_117_2", + "correct_answer": "three times.", + "original_options": { + "A": "Once.", + "B": "Twice.", + "C": "three times.", + "D": "You opened the fridge four times in total throughout the entire clip.", + "E": "You opened the fridge five times overall, counting every separate opening action." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates did I just wash in total?", + "answer": "three.", + "question_chinese": "我刚才一共洗了几个盘子？", + "answer_chinese": "3个。", + "evidence": [ + { + "description": "Washed the first plate.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:15" + } + }, + { + "description": "Washed the second plate.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:49" + } + }, + { + "description": "Washed the third plate.", + "timestamps": { + "start": "00:04:30", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:05:05", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "You washed four plates in total, finishing the last one right before stopping.", + "D": "You washed five plates altogether, counting every plate you scrubbed and rinsed.", + "E": "three." + }, + "answer_label": "E", + "answer_text": "three.", + "q_id": "P03_117_3", + "correct_answer": "three.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "You washed four plates in total, finishing the last one right before stopping.", + "D": "You washed five plates altogether, counting every plate you scrubbed and rinsed.", + "E": "three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates did I just wash in total?", + "answer": "Three.", + "question_chinese": "我刚才一共洗了几个盘子？", + "answer_chinese": "3个。", + "evidence": [ + { + "description": "Washed the first plate.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:15" + } + }, + { + "description": "Washed the second plate.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:49" + } + }, + { + "description": "Washed the third plate.", + "timestamps": { + "start": "00:04:30", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:08:55", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "You washed four plates in total, including the final plate you rinsed at the end.", + "D": "Three.", + "E": "You washed five plates altogether, counting each separate plate you cleaned and set aside." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "P03_117_4", + "correct_answer": "Three.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "You washed four plates in total, including the final plate you rinsed at the end.", + "D": "Three.", + "E": "You washed five plates altogether, counting each separate plate you cleaned and set aside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_7", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls did I just wash in total?", + "answer": "Two.", + "question_chinese": "我刚才一共洗了几个碗？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "Washed the first bowl.", + "timestamps": { + "start": "00:01:31", + "end": "00:01:45" + } + }, + { + "description": "Washed the second bowl.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:44" + } + } + ], + "question_timestamps": "00:06:55", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One bowl.", + "B": "Two.", + "C": "Three bowls.", + "D": "You washed four bowls in total, counting everything you rinsed and scrubbed before setting them aside.", + "E": "In total, you ended up washing five bowls after finishing the last rinse and putting them on the rack." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "P03_117_7", + "correct_answer": "Two.", + "original_options": { + "A": "One bowl.", + "B": "Two.", + "C": "Three bowls.", + "D": "You washed four bowls in total, counting everything you rinsed and scrubbed before setting them aside.", + "E": "In total, you ended up washing five bowls after finishing the last rinse and putting them on the rack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 19 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_8", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls did I just wash in total?", + "answer": "Two.", + "question_chinese": "我刚才一共洗了几个碗？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "Washed the first bowl.", + "timestamps": { + "start": "00:01:31", + "end": "00:01:45" + } + }, + { + "description": "Washed the second bowl.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:44" + } + } + ], + "question_timestamps": "00:08:05", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One bowl.", + "B": "Two.", + "C": "Four bowls.", + "D": "You washed five bowls in total, including the ones you rinsed quickly before the final scrub.", + "E": "Altogether, you finished washing seven bowls after you cleaned each one and placed them to dry." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "P03_117_8", + "correct_answer": "Two.", + "original_options": { + "A": "One bowl.", + "B": "Two.", + "C": "Four bowls.", + "D": "You washed five bowls in total, including the ones you rinsed quickly before the final scrub.", + "E": "Altogether, you finished washing seven bowls after you cleaned each one and placed them to dry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_9", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots did I just wash in total?", + "answer": "Two pots were washed in total.", + "question_chinese": "我刚才一共洗了几口锅？", + "answer_chinese": "一共洗了两口锅。", + "evidence": [ + { + "description": "Washed the first pot.", + "timestamps": { + "start": "00:02:37", + "end": "00:02:59" + } + }, + { + "description": "Washed the second pot.", + "timestamps": { + "start": "00:03:50", + "end": "00:04:27" + } + } + ], + "question_timestamps": "00:04:53", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One pot.", + "B": "Three pots.", + "C": "In total, you washed four pots, including the ones you rinsed briefly before scrubbing them clean.", + "D": "You ended up washing five pots altogether after cleaning each pot and setting them aside to dry.", + "E": "Two pots were washed in total." + }, + "answer_label": "E", + "answer_text": "Two pots were washed in total.", + "q_id": "P03_117_9", + "correct_answer": "Two pots were washed in total.", + "original_options": { + "A": "One pot.", + "B": "Three pots.", + "C": "In total, you washed four pots, including the ones you rinsed briefly before scrubbing them clean.", + "D": "You ended up washing five pots altogether after cleaning each pot and setting them aside to dry.", + "E": "Two pots were washed in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_10", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots did I just wash in total?", + "answer": "Two pots were washed in total.", + "question_chinese": "我刚才一共洗了几口锅？", + "answer_chinese": "一共洗了两口锅。", + "evidence": [ + { + "description": "Washed the first pot.", + "timestamps": { + "start": "00:02:37", + "end": "00:02:59" + } + }, + { + "description": "Washed the second pot.", + "timestamps": { + "start": "00:03:50", + "end": "00:04:27" + } + } + ], + "question_timestamps": "00:08:12", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One pot.", + "B": "Three pots.", + "C": "In total, you washed four pots, counting the ones you rinsed and then scrubbed in the sink.", + "D": "Two pots were washed in total.", + "E": "You ended up washing five pots altogether after finishing the last rinse and leaving them to dry." + }, + "answer_label": "D", + "answer_text": "Two pots were washed in total.", + "q_id": "P03_117_10", + "correct_answer": "Two pots were washed in total.", + "original_options": { + "A": "One pot.", + "B": "Three pots.", + "C": "In total, you washed four pots, counting the ones you rinsed and then scrubbed in the sink.", + "D": "Two pots were washed in total.", + "E": "You ended up washing five pots altogether after finishing the last rinse and leaving them to dry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_16", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first entered the kitchen, where was the microwave in relation to me?", + "answer": "When you first entered the kitchen, the microwave was to your front left.", + "question_chinese": "我刚进厨房时，微波炉在我的哪个方位？", + "answer_chinese": "刚进厨房时，微波炉在你的偏左前方的位置。", + "evidence": [ + { + "description": "Upon entering the kitchen, the microwave is in the front left of the field of view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:06:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "When you first entered the kitchen, the microwave was to your front left.", + "B": "Directly in front of you.", + "C": "To your front right.", + "D": "When you entered, the microwave was behind you on the left side.", + "E": "When you first came in, the microwave was to your left but slightly behind you." + }, + "answer_label": "A", + "answer_text": "When you first entered the kitchen, the microwave was to your front left.", + "q_id": "P03_117_16", + "correct_answer": "When you first entered the kitchen, the microwave was to your front left.", + "original_options": { + "A": "When you first entered the kitchen, the microwave was to your front left.", + "B": "Directly in front of you.", + "C": "To your front right.", + "D": "When you entered, the microwave was behind you on the left side.", + "E": "When you first came in, the microwave was to your left but slightly behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 12, + 15 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_15", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first entered the kitchen, where was the microwave in relation to me?", + "answer": "When you first entered the kitchen, the microwave was to your front left.", + "question_chinese": "我刚进厨房时，微波炉在我的哪个方位？", + "answer_chinese": "刚进厨房时，微波炉在你的偏左前方的位置。", + "evidence": [ + { + "description": "Upon entering the kitchen, the microwave is in the front left of the field of view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:12:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Directly in front of you.", + "B": "When you first entered the kitchen, the microwave was to your front left.", + "C": "To your far left.", + "D": "When you entered, the microwave was to your front right relative to your position.", + "E": "When you first entered the kitchen, the microwave was behind you to your right." + }, + "answer_label": "B", + "answer_text": "When you first entered the kitchen, the microwave was to your front left.", + "q_id": "P03_117_15", + "correct_answer": "When you first entered the kitchen, the microwave was to your front left.", + "original_options": { + "A": "Directly in front of you.", + "B": "When you first entered the kitchen, the microwave was to your front left.", + "C": "To your far left.", + "D": "When you entered, the microwave was to your front right relative to your position.", + "E": "When you first entered the kitchen, the microwave was behind you to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 14 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_5", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the stove or the washing machine?", + "answer": "The washing machine is closer to you.", + "question_chinese": "灶台和洗衣机，哪个离我最近？", + "answer_chinese": "洗衣机离你最近。", + "evidence": [ + { + "description": "The stove and sink, and the washing machine are about two meters apart, with the sink to the upper left of the washing machine.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "You are now standing directly in front of the sink.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The washing machine is closer to you.", + "B": "The stove is closer to you." + }, + "answer_label": "A", + "answer_text": "The washing machine is closer to you.", + "q_id": "P03_117_5", + "correct_answer": "The washing machine is closer to you.", + "original_options": { + "A": "The washing machine is closer to you.", + "B": "The stove is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_6", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the stove or the washing machine?", + "answer": "The washing machine is closer to you.", + "question_chinese": "灶台和洗衣机，哪个离我最近？", + "answer_chinese": "洗衣机离你最近。", + "evidence": [ + { + "description": "The stove and sink, and the washing machine are about two meters apart, with the sink to the upper left of the washing machine.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "You are now standing directly in front of the sink.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:03:06", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The stove is closer to you.", + "B": "The washing machine is closer to you." + }, + "answer_label": "B", + "answer_text": "The washing machine is closer to you.", + "q_id": "P03_117_6", + "correct_answer": "The washing machine is closer to you.", + "original_options": { + "A": "The stove is closer to you.", + "B": "The washing machine is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_11", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did the bread on the plate come from?", + "answer": "It was taken from the fridge.", + "question_chinese": "盘子里的面包是从哪里拿的？", + "answer_chinese": "从冰箱里拿出来的。", + "evidence": [ + { + "description": "Took a bag of bread from the fridge and placed two slices on a plate.", + "timestamps": { + "start": "00:09:21", + "end": "00:09:58" + } + } + ], + "question_timestamps": "00:11:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was taken from the fridge.", + "B": "It was taken from the pantry.", + "C": "It was taken from the countertop.", + "D": "It was taken from the bread box beside the counter before being set down on the plate.", + "E": "It was taken from the oven shelf after heating, then placed onto the plate on the counter." + }, + "answer_label": "A", + "answer_text": "It was taken from the fridge.", + "q_id": "P03_117_11", + "correct_answer": "It was taken from the fridge.", + "original_options": { + "A": "It was taken from the fridge.", + "B": "It was taken from the pantry.", + "C": "It was taken from the countertop.", + "D": "It was taken from the bread box beside the counter before being set down on the plate.", + "E": "It was taken from the oven shelf after heating, then placed onto the plate on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "P03_117", + "video_length": "00:12:02", + "qid": "P03_117_12", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did the bread on the plate come from?", + "answer": "It was taken from the fridge.", + "question_chinese": "盘子里的面包是从哪里拿的？", + "answer_chinese": "从冰箱里拿出来的。", + "evidence": [ + { + "description": "Took a bag of bread from the fridge and placed two slices on a plate.", + "timestamps": { + "start": "00:09:21", + "end": "00:09:58" + } + } + ], + "question_timestamps": "00:11:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was taken from the oven.", + "B": "It was taken from the bread box.", + "C": "It was taken from the pantry cabinet, then brought over and put onto the plate.", + "D": "It was taken from the fridge.", + "E": "It was taken from the countertop where it was left out, and then placed on the plate." + }, + "answer_label": "D", + "answer_text": "It was taken from the fridge.", + "q_id": "P03_117_12", + "correct_answer": "It was taken from the fridge.", + "original_options": { + "A": "It was taken from the oven.", + "B": "It was taken from the bread box.", + "C": "It was taken from the pantry cabinet, then brought over and put onto the plate.", + "D": "It was taken from the fridge.", + "E": "It was taken from the countertop where it was left out, and then placed on the plate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 17 + ] + } +] diff --git a/QAs/P03_122-mcq.json b/QAs/P03_122-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..953f916a297ecca2b1282ac0c084adfad319c3ac --- /dev/null +++ b/QAs/P03_122-mcq.json @@ -0,0 +1,549 @@ +[ + { + "video_uid": "P03_122", + "video_length": "00:00:38", + "qid": "P03_122_3_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What has my movement path been from when I entered the door up to now?", + "answer": "Starting from the entrance, you walked straight to the front of the stove.", + "question_chinese": "我从进门到现在的运动轨迹是怎样的？", + "answer_chinese": "从入口处出发，你直走到炉灶前。", + "evidence": [ + { + "description": "Walked from the door to the front of the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked to the back of the stove.", + "B": "You turned right toward the stove.", + "C": "Starting from the entrance, you walked straight to the front of the stove.", + "D": "Starting from the entrance, you walked to the stove, then turned left to the counter and stayed there.", + "E": "Starting from the entrance, you walked forward, turned around near the stove, and went back to the entrance." + }, + "answer_label": "C", + "answer_text": "Starting from the entrance, you walked straight to the front of the stove.", + "q_id": "P03_122_3_T1", + "correct_answer": "Starting from the entrance, you walked straight to the front of the stove.", + "original_options": { + "A": "You walked to the back of the stove.", + "B": "You turned right toward the stove.", + "C": "Starting from the entrance, you walked straight to the front of the stove.", + "D": "Starting from the entrance, you walked to the stove, then turned left to the counter and stayed there.", + "E": "Starting from the entrance, you walked forward, turned around near the stove, and went back to the entrance." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 18, + 18 + ] + }, + { + "video_uid": "P03_122", + "video_length": "00:00:38", + "qid": "P03_122_3_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What has my movement path been from when I entered the door up to now?", + "answer": "Starting from the entrance, you walked to the front of the stove, then turned left to the counter with the cutting board, and finally turned back to the stove on your right.", + "question_chinese": "我从进门到现在的运动轨迹是怎样的？", + "answer_chinese": "从入口处出发，你走到炉灶前，然后转向左边放着切菜板的柜台，最后又转回右边的炉灶。", + "evidence": [ + { + "description": "Walked from the door to the front of the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "Turned from the stove to the counter with the cutting board on the left.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:34" + } + }, + { + "description": "Turned back from the front of the counter to the stove on the right.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right to the counter.", + "B": "You returned to the stove on your left.", + "C": "Starting from the entrance, you walked to the back of the stove, then turned left to the counter with the cutting board.", + "D": "Starting from the entrance, you walked to the front of the stove, then turned left to the counter with the cutting board, and finally turned back to the stove on your right.", + "E": "Starting from the entrance, you walked to the front of the stove, turned left to the counter, then continued straight ahead." + }, + "answer_label": "D", + "answer_text": "Starting from the entrance, you walked to the front of the stove, then turned left to the counter with the cutting board, and finally turned back to the stove on your right.", + "q_id": "P03_122_3_T2", + "correct_answer": "Starting from the entrance, you walked to the front of the stove, then turned left to the counter with the cutting board, and finally turned back to the stove on your right.", + "original_options": { + "A": "You turned right to the counter.", + "B": "You returned to the stove on your left.", + "C": "Starting from the entrance, you walked to the back of the stove, then turned left to the counter with the cutting board.", + "D": "Starting from the entrance, you walked to the front of the stove, then turned left to the counter with the cutting board, and finally turned back to the stove on your right.", + "E": "Starting from the entrance, you walked to the front of the stove, turned left to the counter, then continued straight ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 22, + 21 + ] + }, + { + "video_uid": "P03_122", + "video_length": "00:00:38", + "qid": "P03_122_2_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the spoon?", + "answer": "The spoon is above the stove, in front of you.", + "question_chinese": "我刚刚把勺子放在哪儿了？", + "answer_chinese": "勺子在灶台上方，在你的前面。", + "evidence": [ + { + "description": "spoon", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above the stove, behind you.", + "B": "The spoon is above the stove, in front of you.", + "C": "Below the stove, in front.", + "D": "The spoon is to the right of the stove, positioned on your right side near the burners.", + "E": "The spoon is to the left of the stove, sitting to your left rather than above it in front." + }, + "answer_label": "B", + "answer_text": "The spoon is above the stove, in front of you.", + "q_id": "P03_122_2_T1", + "correct_answer": "The spoon is above the stove, in front of you.", + "original_options": { + "A": "Above the stove, behind you.", + "B": "The spoon is above the stove, in front of you.", + "C": "Below the stove, in front.", + "D": "The spoon is to the right of the stove, positioned on your right side near the burners.", + "E": "The spoon is to the left of the stove, sitting to your left rather than above it in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "P03_122", + "video_length": "00:00:38", + "qid": "P03_122_2_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the spoon?", + "answer": "The spoon is to the right of the stove, to your right.", + "question_chinese": "我刚刚把勺子放在哪儿了？", + "answer_chinese": "勺子在灶台的右边，在你的右边。", + "evidence": [ + { + "description": "The spoon was placed on the right side of the stove.", + "timestamps": { + "start": "00:00:29", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Left of the stove.", + "B": "Above the stove, behind you.", + "C": "The spoon is to the right of the stove, to your right.", + "D": "The spoon is below the stove, in front of you, not over on the right side.", + "E": "The spoon is above the stove in front of you, rather than placed to the right of the stove." + }, + "answer_label": "C", + "answer_text": "The spoon is to the right of the stove, to your right.", + "q_id": "P03_122_2_T2", + "correct_answer": "The spoon is to the right of the stove, to your right.", + "original_options": { + "A": "Left of the stove.", + "B": "Above the stove, behind you.", + "C": "The spoon is to the right of the stove, to your right.", + "D": "The spoon is below the stove, in front of you, not over on the right side.", + "E": "The spoon is above the stove in front of you, rather than placed to the right of the stove." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 19 + ] + }, + { + "video_uid": "P03_122", + "video_length": "00:00:38", + "qid": "P03_122_5_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the sink right now?", + "answer": "Yes, you can reach it—it’s to your right.", + "question_chinese": "我现在可以够得到洗碗池吗？", + "answer_chinese": "可以的，它在你的右方。", + "evidence": [ + { + "description": "The sink is about an arm's length away from the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t reach it—it’s to your back-right.", + "B": "Yes, you can reach it—it’s to your right." + }, + "answer_label": "B", + "answer_text": "Yes, you can reach it—it’s to your right.", + "q_id": "P03_122_5_T1", + "correct_answer": "Yes, you can reach it—it’s to your right.", + "original_options": { + "A": "No, you can’t reach it—it’s to your back-right.", + "B": "Yes, you can reach it—it’s to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "P03_122", + "video_length": "00:00:38", + "qid": "P03_122_5_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the sink right now?", + "answer": "No, you can’t reach it—it’s to your back-right.", + "question_chinese": "我现在可以够得到洗碗池吗？", + "answer_chinese": "不可以，它在你右后方。", + "evidence": [ + { + "description": "The sink is about an arm's length away from the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "You are now standing in front of the stove.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t reach it—it’s to your back-right.", + "B": "Yes, you can reach it—it’s to your right." + }, + "answer_label": "A", + "answer_text": "No, you can’t reach it—it’s to your back-right.", + "q_id": "P03_122_5_T2", + "correct_answer": "No, you can’t reach it—it’s to your back-right.", + "original_options": { + "A": "No, you can’t reach it—it’s to your back-right.", + "B": "Yes, you can reach it—it’s to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "P03_122", + "video_length": "00:00:38", + "qid": "P03_122_4_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how far am I from the washing machine right now?", + "answer": "About 2 meters away, to your right.", + "question_chinese": "我现在离洗衣机大概有多远？", + "answer_chinese": "大概2米远，在你的右方。", + "evidence": [ + { + "description": "The stove is about 2 meters away from the washing machine.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "You are now standing in front of the stove.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "About 1 meter away, to your right.", + "B": "About 3 meters away, to your right.", + "C": "About 4 meters away, positioned off to your right side from where you're standing.", + "D": "About 2 meters away, to your right.", + "E": "Roughly half a meter away, still on your right, close enough to reach in one step." + }, + "answer_label": "D", + "answer_text": "About 2 meters away, to your right.", + "q_id": "P03_122_4_T1", + "correct_answer": "About 2 meters away, to your right.", + "original_options": { + "A": "About 1 meter away, to your right.", + "B": "About 3 meters away, to your right.", + "C": "About 4 meters away, positioned off to your right side from where you're standing.", + "D": "About 2 meters away, to your right.", + "E": "Roughly half a meter away, still on your right, close enough to reach in one step." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 16 + ] + }, + { + "video_uid": "P03_122", + "video_length": "00:00:38", + "qid": "P03_122_4_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how far am I from the washing machine right now?", + "answer": "About 2 meters away; it is to your back-right.", + "question_chinese": "我现在离洗衣机大概有多远？", + "answer_chinese": "大概2米远，它在你的右后方。", + "evidence": [ + { + "description": "The stove is about 2 meters away from the washing machine.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "You are now standing in front of the stove.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:38", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "About 1 meter away; to your back-right.", + "B": "About 3 meters away; to your back-right.", + "C": "About 4 meters away; it sits behind you and slightly to the right.", + "D": "Roughly half a meter away; it’s just behind your right shoulder, very close by.", + "E": "About 2 meters away; it is to your back-right." + }, + "answer_label": "E", + "answer_text": "About 2 meters away; it is to your back-right.", + "q_id": "P03_122_4_T2", + "correct_answer": "About 2 meters away; it is to your back-right.", + "original_options": { + "A": "About 1 meter away; to your back-right.", + "B": "About 3 meters away; to your back-right.", + "C": "About 4 meters away; it sits behind you and slightly to the right.", + "D": "Roughly half a meter away; it’s just behind your right shoulder, very close by.", + "E": "About 2 meters away; it is to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 15 + ] + }, + { + "video_uid": "P03_122", + "video_length": "00:00:38", + "qid": "P03_122_1_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the wooden cutting board relative to me?", + "answer": "The wooden cutting board is to your front-left.", + "question_chinese": "木质案板在我的哪个方向？", + "answer_chinese": "木质案板在你的左前方。", + "evidence": [ + { + "description": "When you first entered the door, the wooden cutting board was in the front left of the field of view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The wooden cutting board is to your front-left.", + "B": "It is to your front-right.", + "C": "It is to your back-left.", + "D": "The wooden cutting board is directly in front of you, centered ahead on the counter.", + "E": "The wooden cutting board sits to your left side, slightly behind your current position." + }, + "answer_label": "A", + "answer_text": "The wooden cutting board is to your front-left.", + "q_id": "P03_122_1_T1", + "correct_answer": "The wooden cutting board is to your front-left.", + "original_options": { + "A": "The wooden cutting board is to your front-left.", + "B": "It is to your front-right.", + "C": "It is to your back-left.", + "D": "The wooden cutting board is directly in front of you, centered ahead on the counter.", + "E": "The wooden cutting board sits to your left side, slightly behind your current position." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "P03_122", + "video_length": "00:00:38", + "qid": "P03_122_1_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the wooden cutting board relative to me?", + "answer": "The wooden cutting board is to your left.", + "question_chinese": "木质案板在我的哪个方向？", + "answer_chinese": "木质案板在你的左方。", + "evidence": [ + { + "description": "When you first entered the door, the wooden cutting board was in the front left of the field of view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The wooden cutting board is to your right.", + "B": "The wooden cutting board is to your left.", + "C": "The wooden cutting board is to your front-left.", + "D": "The wooden cutting board is directly in front of you, aligned straight ahead.", + "E": "The wooden cutting board is behind you on the left, closer to your back-left side." + }, + "answer_label": "B", + "answer_text": "The wooden cutting board is to your left.", + "q_id": "P03_122_1_T2", + "correct_answer": "The wooden cutting board is to your left.", + "original_options": { + "A": "The wooden cutting board is to your right.", + "B": "The wooden cutting board is to your left.", + "C": "The wooden cutting board is to your front-left.", + "D": "The wooden cutting board is directly in front of you, aligned straight ahead.", + "E": "The wooden cutting board is behind you on the left, closer to your back-left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 13, + 15 + ] + } +] diff --git a/QAs/P04_101-mcq.json b/QAs/P04_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9a26b80beca27bf994fb680b61903f65dbb81873 --- /dev/null +++ b/QAs/P04_101-mcq.json @@ -0,0 +1,824 @@ +[ + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the refrigerator in relation to the stove?", + "answer": "The refrigerator is to the right of the stove.", + "question_chinese": "冰箱在灶台的哪个方位？", + "answer_chinese": "冰箱在灶台的右边。", + "evidence": [ + { + "description": "The refrigerator is to the right of the stove.", + "timestamps": { + "start": "00:03:56", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:04:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is left of the stove.", + "B": "The refrigerator is to the right of the stove.", + "C": "The refrigerator is behind the stove.", + "D": "The refrigerator is directly in front of the stove, facing it across the aisle.", + "E": "The refrigerator is across from the stove, directly opposite it along the walkway." + }, + "answer_label": "B", + "answer_text": "The refrigerator is to the right of the stove.", + "q_id": "P04_101_1", + "correct_answer": "The refrigerator is to the right of the stove.", + "original_options": { + "A": "The refrigerator is left of the stove.", + "B": "The refrigerator is to the right of the stove.", + "C": "The refrigerator is behind the stove.", + "D": "The refrigerator is directly in front of the stove, facing it across the aisle.", + "E": "The refrigerator is across from the stove, directly opposite it along the walkway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 13 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the refrigerator in relation to the stove?", + "answer": "The refrigerator is to the right of the stove.", + "question_chinese": "冰箱在灶台的哪个方位？", + "answer_chinese": "冰箱在灶台的右边。", + "evidence": [ + { + "description": "The refrigerator is to the right of the stove.", + "timestamps": { + "start": "00:03:56", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:04:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is to the right of the stove.", + "B": "The refrigerator is left of the stove.", + "C": "The refrigerator is behind the stove.", + "D": "The refrigerator is directly in front of the stove, positioned opposite it across the space.", + "E": "The refrigerator is on the far side of the stove, across the walkway rather than beside it." + }, + "answer_label": "A", + "answer_text": "The refrigerator is to the right of the stove.", + "q_id": "P04_101_2", + "correct_answer": "The refrigerator is to the right of the stove.", + "original_options": { + "A": "The refrigerator is to the right of the stove.", + "B": "The refrigerator is left of the stove.", + "C": "The refrigerator is behind the stove.", + "D": "The refrigerator is directly in front of the stove, positioned opposite it across the space.", + "E": "The refrigerator is on the far side of the stove, across the walkway rather than beside it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 17 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_11", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the cabinet in total so far?", + "answer": "Twice.", + "question_chinese": "到目前为止，我总共打开了几次储物柜？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "Opened the cabinet for the first time.", + "timestamps": { + "start": "00:04:57", + "end": "00:05:03" + } + }, + { + "description": "Opened the cabinet for the second time.", + "timestamps": { + "start": "00:05:21", + "end": "00:05:35" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Twice.", + "D": "Four times in total, including a quick check inside earlier today again.", + "E": "Zero times so far, because the cabinet has stayed closed the whole time." + }, + "answer_label": "C", + "answer_text": "Twice.", + "q_id": "P04_101_11", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Twice.", + "D": "Four times in total, including a quick check inside earlier today again.", + "E": "Zero times so far, because the cabinet has stayed closed the whole time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 13 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_12", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the cabinet in total so far?", + "answer": "Twice.", + "question_chinese": "到目前为止，我总共打开了几次储物柜？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "Opened the cabinet for the first time.", + "timestamps": { + "start": "00:04:57", + "end": "00:05:03" + } + }, + { + "description": "Opened the cabinet for the second time.", + "timestamps": { + "start": "00:05:21", + "end": "00:05:35" + } + } + ], + "question_timestamps": "00:08:51", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times in total, including a quick check inside earlier today again.", + "D": "Zero times so far, because the cabinet has stayed closed the whole time.", + "E": "Twice." + }, + "answer_label": "E", + "answer_text": "Twice.", + "q_id": "P04_101_12", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times in total, including a quick check inside earlier today again.", + "D": "Zero times so far, because the cabinet has stayed closed the whole time.", + "E": "Twice." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 13 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_15", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement trajectory of this electronic scale on the countertop?", + "answer": "The electronic scale was taken out from the drawer diagonally opposite the sink, and then moved to the countertop between the stove and the sink.", + "question_chinese": "台面上这个电子秤的运动轨迹是怎样的？", + "answer_chinese": "电子秤从洗碗池斜对面的抽屉里拿出，然后运动到灶台和洗碗池之间的台面上。", + "evidence": [ + { + "description": "Took out the electronic scale from the drawer diagonally opposite the sink.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:40" + } + }, + { + "description": "Walked towards the stove with the electronic scale.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:43" + } + }, + { + "description": "Placed the electronic scale on the countertop between the stove and the sink.", + "timestamps": { + "start": "00:00:43", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The electronic scale was taken out from the drawer diagonally opposite the sink, and then moved to the countertop between the stove and the sink.", + "B": "You moved it from the sink to stove.", + "C": "You took it from a nearby drawer.", + "D": "You pulled the scale from the drawer beside the sink and set it on the counter by the stove.", + "E": "You removed the scale from the drawer opposite the stove, then placed it on the counter closer to the sink." + }, + "answer_label": "A", + "answer_text": "The electronic scale was taken out from the drawer diagonally opposite the sink, and then moved to the countertop between the stove and the sink.", + "q_id": "P04_101_15", + "correct_answer": "The electronic scale was taken out from the drawer diagonally opposite the sink, and then moved to the countertop between the stove and the sink.", + "original_options": { + "A": "The electronic scale was taken out from the drawer diagonally opposite the sink, and then moved to the countertop between the stove and the sink.", + "B": "You moved it from the sink to stove.", + "C": "You took it from a nearby drawer.", + "D": "You pulled the scale from the drawer beside the sink and set it on the counter by the stove.", + "E": "You removed the scale from the drawer opposite the stove, then placed it on the counter closer to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 20 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_16", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement trajectory of this electronic scale on the countertop?", + "answer": "The electronic scale was taken out from the drawer diagonally opposite the sink, and then moved to the countertop between the stove and the sink.", + "question_chinese": "台面上这个电子秤的运动轨迹是怎样的？", + "answer_chinese": "电子秤从洗碗池斜对面的抽屉里拿出，然后运动到灶台和洗碗池之间的台面上。", + "evidence": [ + { + "description": "Took out the electronic scale from the drawer diagonally opposite the sink.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:40" + } + }, + { + "description": "Walked towards the stove with the electronic scale.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:43" + } + }, + { + "description": "Placed the electronic scale on the countertop between the stove and the sink.", + "timestamps": { + "start": "00:00:43", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:03:43", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You took it from a cabinet.", + "B": "You moved it beside the coffee maker.", + "C": "You lifted the scale from a cabinet under the microwave and put it on the counter next to the blender.", + "D": "You grabbed the scale from a drawer next to the oven and set it down on the countertop by the toaster.", + "E": "The electronic scale was taken out from the drawer diagonally opposite the sink, and then moved to the countertop between the stove and the sink." + }, + "answer_label": "E", + "answer_text": "The electronic scale was taken out from the drawer diagonally opposite the sink, and then moved to the countertop between the stove and the sink.", + "q_id": "P04_101_16", + "correct_answer": "The electronic scale was taken out from the drawer diagonally opposite the sink, and then moved to the countertop between the stove and the sink.", + "original_options": { + "A": "You took it from a cabinet.", + "B": "You moved it beside the coffee maker.", + "C": "You lifted the scale from a cabinet under the microwave and put it on the counter next to the blender.", + "D": "You grabbed the scale from a drawer next to the oven and set it down on the countertop by the toaster.", + "E": "The electronic scale was taken out from the drawer diagonally opposite the sink, and then moved to the countertop between the stove and the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 21 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_3", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the cutlery drawer now?", + "answer": "Yes.", + "question_chinese": "我现在可以够到装餐具的抽屉吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The cutlery cabinet is to the left of the sink.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:14" + } + }, + { + "description": "You are standing in front of the sink.", + "timestamps": { + "start": "00:04:40", + "end": "00:04:43" + } + } + ], + "question_timestamps": "00:04:44", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "P04_101_3", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_4", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the cutlery drawer now?", + "answer": "Yes.", + "question_chinese": "我现在可以够到装餐具的抽屉吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The cutlery cabinet is to the left of the sink.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:14" + } + }, + { + "description": "You are standing in front of the sink.", + "timestamps": { + "start": "00:05:43", + "end": "00:05:46" + } + } + ], + "question_timestamps": "00:05:47", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "P04_101_4", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_13", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement trajectory of the green spoon in front of me?", + "answer": "The green spoon was taken out of the cabinet drawer, then moved from above the sink to the left of the cup.", + "question_chinese": "我面前的这柄绿色的勺子的运动轨迹是怎样的？", + "answer_chinese": "绿色的勺子从储物柜抽屉里拿出，然后从洗碗池的上方运动到杯子的左边。", + "evidence": [ + { + "description": "Took the green spoon out of the cabinet drawer to the left of the sink.", + "timestamps": { + "start": "00:01:42", + "end": "00:01:45" + } + }, + { + "description": "Used the green spoon to get water from the sink and pour it into the cup on the right.", + "timestamps": { + "start": "00:01:46", + "end": "00:02:02" + } + }, + { + "description": "Placed the green spoon to the left of the cup.", + "timestamps": { + "start": "00:02:03", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:02:53", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You placed it right of the cup.", + "B": "You moved it beside the bowl.", + "C": "You took the green spoon from the cabinet drawer, then shifted it from above the sink to the right of the cup.", + "D": "The green spoon was taken out of the cabinet drawer, then moved from above the sink to the left of the cup.", + "E": "You removed the green spoon from the cabinet drawer and moved it from above the sink to the left of the bowl." + }, + "answer_label": "D", + "answer_text": "The green spoon was taken out of the cabinet drawer, then moved from above the sink to the left of the cup.", + "q_id": "P04_101_13", + "correct_answer": "The green spoon was taken out of the cabinet drawer, then moved from above the sink to the left of the cup.", + "original_options": { + "A": "You placed it right of the cup.", + "B": "You moved it beside the bowl.", + "C": "You took the green spoon from the cabinet drawer, then shifted it from above the sink to the right of the cup.", + "D": "The green spoon was taken out of the cabinet drawer, then moved from above the sink to the left of the cup.", + "E": "You removed the green spoon from the cabinet drawer and moved it from above the sink to the left of the bowl." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 22, + 22 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_14", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement trajectory of the green spoon in front of me?", + "answer": "The green spoon was taken out of the cabinet drawer, then moved from above the sink to the left of the cup.", + "question_chinese": "我面前的这柄绿色的勺子的运动轨迹是怎样的？", + "answer_chinese": "绿色的勺子从储物柜抽屉里拿出，然后从洗碗池的上方运动到杯子的左边。", + "evidence": [ + { + "description": "Took the green spoon out of the cabinet drawer to the left of the sink.", + "timestamps": { + "start": "00:01:42", + "end": "00:01:45" + } + }, + { + "description": "Used the green spoon to get water from the sink and pour it into the cup on the right.", + "timestamps": { + "start": "00:01:46", + "end": "00:02:02" + } + }, + { + "description": "Placed the green spoon to the left of the cup, between the sink and the stove.", + "timestamps": { + "start": "00:02:03", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:06:05", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You moved it to the right.", + "B": "You took it from the dish rack.", + "C": "The green spoon was taken out of the cabinet drawer, then moved from above the sink to the left of the cup.", + "D": "You pulled it from the cabinet drawer, then moved it from above the counter to the right of the cup.", + "E": "You picked up the spoon from a utensil holder, then moved it from near the stove to the right of the bowl." + }, + "answer_label": "C", + "answer_text": "The green spoon was taken out of the cabinet drawer, then moved from above the sink to the left of the cup.", + "q_id": "P04_101_14", + "correct_answer": "The green spoon was taken out of the cabinet drawer, then moved from above the sink to the left of the cup.", + "original_options": { + "A": "You moved it to the right.", + "B": "You took it from the dish rack.", + "C": "The green spoon was taken out of the cabinet drawer, then moved from above the sink to the left of the cup.", + "D": "You pulled it from the cabinet drawer, then moved it from above the counter to the right of the cup.", + "E": "You picked up the spoon from a utensil holder, then moved it from near the stove to the right of the bowl." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 22 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_9", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the refrigerator in total so far?", + "answer": "Twice.", + "question_chinese": "现在我一共打开了几次冰箱？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "Opened the refrigerator for the first time.", + "timestamps": { + "start": "00:03:57", + "end": "00:04:05" + } + }, + { + "description": "Opened the refrigerator for the second time.", + "timestamps": { + "start": "00:07:19", + "end": "00:07:14" + } + } + ], + "question_timestamps": "00:07:57", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Twice.", + "C": "Three times.", + "D": "You have opened the refrigerator four times in total so far today.", + "E": "You have opened the refrigerator five times in total so far today." + }, + "answer_label": "B", + "answer_text": "Twice.", + "q_id": "P04_101_9", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Twice.", + "C": "Three times.", + "D": "You have opened the refrigerator four times in total so far today.", + "E": "You have opened the refrigerator five times in total so far today." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 12 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_10", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the refrigerator in total so far?", + "answer": "Twice.", + "question_chinese": "现在我一共打开了几次冰箱？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "Opened the refrigerator for the first time.", + "timestamps": { + "start": "00:03:57", + "end": "00:04:05" + } + }, + { + "description": "Opened the refrigerator for the second time.", + "timestamps": { + "start": "00:07:19", + "end": "00:07:14" + } + } + ], + "question_timestamps": "00:09:01", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Four times.", + "C": "You have opened the refrigerator five times in total so far today.", + "D": "Twice.", + "E": "You have opened the refrigerator six times in total so far today." + }, + "answer_label": "D", + "answer_text": "Twice.", + "q_id": "P04_101_10", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Four times.", + "C": "You have opened the refrigerator five times in total so far today.", + "D": "Twice.", + "E": "You have opened the refrigerator six times in total so far today." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 12 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_7", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are currently on the stove?", + "answer": "There is currently one pot on the stove.", + "question_chinese": "目前灶台上有几口锅？", + "answer_chinese": "目前灶台上有一口锅。", + "evidence": [ + { + "description": "There is one pot on the stove.", + "timestamps": { + "start": "00:07:41", + "end": "00:08:32" + } + } + ], + "question_timestamps": "00:08:50", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two pots.", + "B": "Three pots.", + "C": "There are currently two pots on the stove, sitting side by side on different burners.", + "D": "There are currently four pots on the stove, with multiple burners in use at the same time.", + "E": "There is currently one pot on the stove." + }, + "answer_label": "E", + "answer_text": "There is currently one pot on the stove.", + "q_id": "P04_101_7", + "correct_answer": "There is currently one pot on the stove.", + "original_options": { + "A": "Two pots.", + "B": "Three pots.", + "C": "There are currently two pots on the stove, sitting side by side on different burners.", + "D": "There are currently four pots on the stove, with multiple burners in use at the same time.", + "E": "There is currently one pot on the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 17 + ] + }, + { + "video_uid": "P04_101", + "video_length": "00:09:09", + "qid": "P04_101_8", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are currently on the stove?", + "answer": "There is currently one pot on the stove.", + "question_chinese": "目前灶台上有几口锅？", + "answer_chinese": "目前灶台上有一口锅。", + "evidence": [ + { + "description": "There is one pot on the stove.", + "timestamps": { + "start": "00:07:41", + "end": "00:08:32" + } + } + ], + "question_timestamps": "00:09:03", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is currently one pot on the stove.", + "B": "Two pots.", + "C": "Four pots.", + "D": "There are currently three pots on the stove, spread across the front and back burners.", + "E": "There are currently five pots on the stove, leaving little open space on the stovetop." + }, + "answer_label": "A", + "answer_text": "There is currently one pot on the stove.", + "q_id": "P04_101_8", + "correct_answer": "There is currently one pot on the stove.", + "original_options": { + "A": "There is currently one pot on the stove.", + "B": "Two pots.", + "C": "Four pots.", + "D": "There are currently three pots on the stove, spread across the front and back burners.", + "E": "There are currently five pots on the stove, leaving little open space on the stovetop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + } +] diff --git a/QAs/P04_104-mcq.json b/QAs/P04_104-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c0e6c94b089e7a3393d2881b429a1b6e71c49812 --- /dev/null +++ b/QAs/P04_104-mcq.json @@ -0,0 +1,1649 @@ +[ + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_15", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the refrigerator in total?", + "answer": "Once.", + "question_chinese": "我一共开了几次冰箱？", + "answer_chinese": "1次。", + "evidence": [ + { + "description": "Opened the refrigerator once.", + "timestamps": { + "start": "00:03:24", + "end": "00:03:28" + } + } + ], + "question_timestamps": "00:05:02", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Three times.", + "C": "You opened the refrigerator two times in total, closing it fully after the first opening.", + "D": "You opened the refrigerator four times in total, returning to it multiple times during the task.", + "E": "Once." + }, + "answer_label": "E", + "answer_text": "Once.", + "q_id": "P04_104_15", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Three times.", + "C": "You opened the refrigerator two times in total, closing it fully after the first opening.", + "D": "You opened the refrigerator four times in total, returning to it multiple times during the task.", + "E": "Once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 16 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_16", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the refrigerator in total?", + "answer": "Once.", + "question_chinese": "我一共开了几次冰箱？", + "answer_chinese": "1次。", + "evidence": [ + { + "description": "Opened the refrigerator once.", + "timestamps": { + "start": "00:03:24", + "end": "00:03:28" + } + } + ], + "question_timestamps": "00:06:47", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Five times.", + "C": "You opened the refrigerator three times in total, checking inside more than once before finishing.", + "D": "Once.", + "E": "You opened the refrigerator four times in total, with repeated opens and closes throughout the sequence." + }, + "answer_label": "D", + "answer_text": "Once.", + "q_id": "P04_104_16", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Five times.", + "C": "You opened the refrigerator three times in total, checking inside more than once before finishing.", + "D": "Once.", + "E": "You opened the refrigerator four times in total, with repeated opens and closes throughout the sequence." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 16 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many new batteries did I just take out, and how many are left after replacing the batteries in the electronic scale?", + "answer": "one battery.", + "question_chinese": "我刚刚取出了几块新电池，给电子秤上完电池后还剩几块新电池？", + "answer_chinese": "1块。", + "evidence": [ + { + "description": "Took out three new batteries from the drawer.", + "timestamps": { + "start": "00:04:52", + "end": "00:05:02" + } + }, + { + "description": "Took one new battery from the packaging and replaced it.", + "timestamps": { + "start": "00:05:03", + "end": "00:05:24" + } + }, + { + "description": "Took a second new battery from the packaging and replaced it.", + "timestamps": { + "start": "00:05:25", + "end": "00:05:56" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "one battery.", + "B": "Two batteries.", + "C": "Three batteries.", + "D": "You took out two new batteries, and two were left after replacing the batteries in the scale.", + "E": "You took out four new batteries, and one remained left after swapping the old ones in the scale." + }, + "answer_label": "A", + "answer_text": "one battery.", + "q_id": "P04_104_5", + "correct_answer": "one battery.", + "original_options": { + "A": "one battery.", + "B": "Two batteries.", + "C": "Three batteries.", + "D": "You took out two new batteries, and two were left after replacing the batteries in the scale.", + "E": "You took out four new batteries, and one remained left after swapping the old ones in the scale." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many new batteries did I just take out, and how many are left after replacing the batteries in the electronic scale?", + "answer": "one battery.", + "question_chinese": "我刚刚取出了几块新电池，给电子秤上完电池后还剩几块新电池？", + "answer_chinese": "1块。", + "evidence": [ + { + "description": "Took out three new batteries from the drawer.", + "timestamps": { + "start": "00:04:52", + "end": "00:05:02" + } + }, + { + "description": "Took one new battery from the packaging and replaced it.", + "timestamps": { + "start": "00:05:03", + "end": "00:05:24" + } + }, + { + "description": "Took a second new battery from the packaging and replaced it.", + "timestamps": { + "start": "00:05:25", + "end": "00:05:56" + } + } + ], + "question_timestamps": "00:07:43", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two batteries.", + "B": "Five batteries.", + "C": "You took out three new batteries, and one was left after replacing the batteries in the electronic scale.", + "D": "one battery.", + "E": "You took out four new batteries, and two remained after you finished replacing the scale batteries." + }, + "answer_label": "D", + "answer_text": "one battery.", + "q_id": "P04_104_6", + "correct_answer": "one battery.", + "original_options": { + "A": "Two batteries.", + "B": "Five batteries.", + "C": "You took out three new batteries, and one was left after replacing the batteries in the electronic scale.", + "D": "one battery.", + "E": "You took out four new batteries, and two remained after you finished replacing the scale batteries." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 16 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my trajectory when I went to get the green bag of flour just now?", + "answer": "You walked from the stove to the cutting board rack opposite, took out a green bag of flour from under the rack, then walked back to the stove and poured the flour out.", + "question_chinese": "我刚才去拿绿色的袋装面粉经过了什么轨迹？", + "answer_chinese": "", + "evidence": [ + { + "description": "Walked from the stove to the cutting board rack opposite, and then took out a green bag of flour from under the rack.", + "timestamps": { + "start": "00:03:34", + "end": "00:03:42" + } + }, + { + "description": "Walked from the cutting board rack to the front of the stove, and then poured flour out of the green bag.", + "timestamps": { + "start": "00:03:43", + "end": "00:03:51" + } + } + ], + "question_timestamps": "00:03:56", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You grabbed it from in front.", + "B": "You took a red bag of salt.", + "C": "You walked from the stove to the rack opposite, took the green flour from in front, then returned to the stove.", + "D": "You walked from the stove to the rack opposite, pulled out a yellow bag of rice from under it.", + "E": "You walked from the stove to the cutting board rack opposite, took out a green bag of flour from under the rack, then walked back to the stove and poured the flour out." + }, + "answer_label": "E", + "answer_text": "You walked from the stove to the cutting board rack opposite, took out a green bag of flour from under the rack, then walked back to the stove and poured the flour out.", + "q_id": "P04_104_1", + "correct_answer": "You walked from the stove to the cutting board rack opposite, took out a green bag of flour from under the rack, then walked back to the stove and poured the flour out.", + "original_options": { + "A": "You grabbed it from in front.", + "B": "You took a red bag of salt.", + "C": "You walked from the stove to the rack opposite, took the green flour from in front, then returned to the stove.", + "D": "You walked from the stove to the rack opposite, pulled out a yellow bag of rice from under it.", + "E": "You walked from the stove to the cutting board rack opposite, took out a green bag of flour from under the rack, then walked back to the stove and poured the flour out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 19 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my trajectory when I went to get the green bag of flour just now?", + "answer": "You walked from the stove to the cutting board rack opposite, took out a green bag of flour from under the rack, then walked back to the stove and poured the flour out.", + "question_chinese": "我刚才去拿绿色的袋装面粉经过了什么轨迹？", + "answer_chinese": "", + "evidence": [ + { + "description": "Walked from the stove to the cutting board rack opposite, and then took out a green bag of flour from under the rack.", + "timestamps": { + "start": "00:03:34", + "end": "00:03:42" + } + }, + { + "description": "Walked from the cutting board rack to the front of the stove, and then poured flour out of the green bag.", + "timestamps": { + "start": "00:03:43", + "end": "00:03:51" + } + } + ], + "question_timestamps": "00:05:06", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You walked to the rack and stopped.", + "B": "You walked from the stove to the cutting board rack opposite, took out a green bag of flour from under the rack, then walked back to the stove and poured the flour out.", + "C": "You returned to the stove without pouring.", + "D": "You walked from the stove to the cutting board rack opposite, took the green flour from above the rack, then came back.", + "E": "You walked from the stove to the rack opposite, grabbed a different bag from under it, and returned." + }, + "answer_label": "B", + "answer_text": "You walked from the stove to the cutting board rack opposite, took out a green bag of flour from under the rack, then walked back to the stove and poured the flour out.", + "q_id": "P04_104_2", + "correct_answer": "You walked from the stove to the cutting board rack opposite, took out a green bag of flour from under the rack, then walked back to the stove and poured the flour out.", + "original_options": { + "A": "You walked to the rack and stopped.", + "B": "You walked from the stove to the cutting board rack opposite, took out a green bag of flour from under the rack, then walked back to the stove and poured the flour out.", + "C": "You returned to the stove without pouring.", + "D": "You walked from the stove to the cutting board rack opposite, took the green flour from above the rack, then came back.", + "E": "You walked from the stove to the rack opposite, grabbed a different bag from under it, and returned." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 22, + 18 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_9", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I go with the electronic scale just now?", + "answer": "You took the electronic scale from the side of the stove to the storage cabinet drawer on the right, then walked to the cutting board rack next to it, and finally returned to the side of the stove.", + "question_chinese": "我刚才拿着电子秤去了哪里？", + "answer_chinese": "你拿着电子秤从灶台旁边走到了右侧的储物柜抽屉旁，然后走到了旁边的案板架前，最后回到了灶台旁边。", + "evidence": [ + { + "description": "Took the electronic scale from the side of the stove to the storage cabinet drawer on the right.", + "timestamps": { + "start": "00:04:16", + "end": "00:04:46" + } + }, + { + "description": "Took the electronic scale from the storage cabinet drawer to the cutting board rack next to it, and then returned to the side of the stove.", + "timestamps": { + "start": "00:04:47", + "end": "00:05:02" + } + } + ], + "question_timestamps": "00:05:22", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took the electronic scale from the side of the stove to the storage cabinet drawer on the right, then walked to the cutting board rack next to it, and finally returned to the side of the stove.", + "B": "To the left cabinet drawer.", + "C": "To the cutting board rack first.", + "D": "You carried the scale from the stove to the left storage drawer, then checked the rack, and went back.", + "E": "You moved the scale from the stove into the right drawer, walked to the sink area, then returned to the stove." + }, + "answer_label": "A", + "answer_text": "You took the electronic scale from the side of the stove to the storage cabinet drawer on the right, then walked to the cutting board rack next to it, and finally returned to the side of the stove.", + "q_id": "P04_104_9", + "correct_answer": "You took the electronic scale from the side of the stove to the storage cabinet drawer on the right, then walked to the cutting board rack next to it, and finally returned to the side of the stove.", + "original_options": { + "A": "You took the electronic scale from the side of the stove to the storage cabinet drawer on the right, then walked to the cutting board rack next to it, and finally returned to the side of the stove.", + "B": "To the left cabinet drawer.", + "C": "To the cutting board rack first.", + "D": "You carried the scale from the stove to the left storage drawer, then checked the rack, and went back.", + "E": "You moved the scale from the stove into the right drawer, walked to the sink area, then returned to the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 21 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_10", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I go with the electronic scale just now?", + "answer": "You took the electronic scale from the side of the stove to the storage cabinet drawer on the right, then walked to the cutting board rack next to it, and finally returned to the side of the stove.", + "question_chinese": "我刚才拿着电子秤去了哪里？", + "answer_chinese": "你拿着电子秤从灶台旁边走到了右侧的储物柜抽屉旁，然后走到了旁边的案板架前，最后回到了灶台旁边。", + "evidence": [ + { + "description": "Took the electronic scale from the side of the stove to the storage cabinet drawer on the right.", + "timestamps": { + "start": "00:04:16", + "end": "00:04:46" + } + }, + { + "description": "Took the electronic scale from the storage cabinet drawer to the cutting board rack next to it, and then returned to the side of the stove.", + "timestamps": { + "start": "00:04:47", + "end": "00:05:02" + } + } + ], + "question_timestamps": "00:06:01", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To the sink area nearby.", + "B": "To the upper shelf above.", + "C": "You took the electronic scale from the side of the stove to the storage cabinet drawer on the right, then walked to the cutting board rack next to it, and finally returned to the side of the stove.", + "D": "You took the scale from the stove to the right drawer, then walked over to the sink, and returned to the stove.", + "E": "You moved the scale from the stove to the right drawer, then went to the spice rack beside it, and came back." + }, + "answer_label": "C", + "answer_text": "You took the electronic scale from the side of the stove to the storage cabinet drawer on the right, then walked to the cutting board rack next to it, and finally returned to the side of the stove.", + "q_id": "P04_104_10", + "correct_answer": "You took the electronic scale from the side of the stove to the storage cabinet drawer on the right, then walked to the cutting board rack next to it, and finally returned to the side of the stove.", + "original_options": { + "A": "To the sink area nearby.", + "B": "To the upper shelf above.", + "C": "You took the electronic scale from the side of the stove to the storage cabinet drawer on the right, then walked to the cutting board rack next to it, and finally returned to the side of the stove.", + "D": "You took the scale from the stove to the right drawer, then walked over to the sink, and returned to the stove.", + "E": "You moved the scale from the stove to the right drawer, then went to the spice rack beside it, and came back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 22, + 22 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_3_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many new batteries did I take out?", + "answer": "You took out 0 new batteries, and the batteries are in front of you.", + "question_chinese": "我取出了几块新电池？", + "answer_chinese": "你取出了0块新电池，电池在你前面。", + "evidence": [ + { + "description": "Took out zero new batterie .", + "timestamps": { + "start": "00:04:53", + "end": "00:04:55" + } + } + ], + "question_timestamps": "00:04:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "1 new battery.", + "B": "2 new batteries.", + "C": "You took out 3 new batteries, and the batteries are in front of you.", + "D": "You took out 0 new batteries, and the batteries are in front of you.", + "E": "You took out 4 new batteries, and the batteries are in front of you." + }, + "answer_label": "D", + "answer_text": "You took out 0 new batteries, and the batteries are in front of you.", + "q_id": "P04_104_3_T1", + "correct_answer": "You took out 0 new batteries, and the batteries are in front of you.", + "original_options": { + "A": "1 new battery.", + "B": "2 new batteries.", + "C": "You took out 3 new batteries, and the batteries are in front of you.", + "D": "You took out 0 new batteries, and the batteries are in front of you.", + "E": "You took out 4 new batteries, and the batteries are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 14 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_3_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many new batteries did I take out?", + "answer": "You took out 1 new battery, and the battery is in front of you.", + "question_chinese": "我取出了几块新电池？", + "answer_chinese": "你取出了1块新电池,电池在你前面。", + "evidence": [ + { + "description": "Took out one new battery.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:11" + } + } + ], + "question_timestamps": "00:05:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "0 new batteries.", + "B": "2 new batteries.", + "C": "You took out 3 new batteries, and the batteries are in front of you.", + "D": "You took out 4 new batteries, and the batteries are in front of you.", + "E": "You took out 1 new battery, and the battery is in front of you." + }, + "answer_label": "E", + "answer_text": "You took out 1 new battery, and the battery is in front of you.", + "q_id": "P04_104_3_T2", + "correct_answer": "You took out 1 new battery, and the battery is in front of you.", + "original_options": { + "A": "0 new batteries.", + "B": "2 new batteries.", + "C": "You took out 3 new batteries, and the batteries are in front of you.", + "D": "You took out 4 new batteries, and the batteries are in front of you.", + "E": "You took out 1 new battery, and the battery is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 14 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_8_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoons am I holding in my hand?", + "answer": "You are holding 1 spoon in your hand, and the spoon is in front of you.", + "question_chinese": "我手上拿着几个勺子？", + "answer_chinese": "你手上拿着1个勺子，勺子在你前面。", + "evidence": [ + { + "description": "You are holding a spoon in your hand.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "0 spoons.", + "B": "2 spoons.", + "C": "You are holding 3 spoons in your hand, and the spoons are in front of you.", + "D": "You are holding 4 spoons in your hand, and the spoons are in front of you.", + "E": "You are holding 1 spoon in your hand, and the spoon is in front of you." + }, + "answer_label": "E", + "answer_text": "You are holding 1 spoon in your hand, and the spoon is in front of you.", + "q_id": "P04_104_8_T2", + "correct_answer": "You are holding 1 spoon in your hand, and the spoon is in front of you.", + "original_options": { + "A": "0 spoons.", + "B": "2 spoons.", + "C": "You are holding 3 spoons in your hand, and the spoons are in front of you.", + "D": "You are holding 4 spoons in your hand, and the spoons are in front of you.", + "E": "You are holding 1 spoon in your hand, and the spoon is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_8_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoons am I holding in my hand?", + "answer": "You are holding 2 spoons in your hand, and the spoons are in front of you.", + "question_chinese": "我手上拿着几个勺子？", + "answer_chinese": "你手上拿着2个勺子，勺子在你前面。", + "evidence": [ + { + "description": "You are holding two spoons in your hand.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "1 spoon.", + "B": "3 spoons.", + "C": "You are holding 0 spoons in your hand, and no spoons are visible in front of you.", + "D": "You are holding 2 spoons in your hand, and the spoons are in front of you.", + "E": "You are holding 4 spoons in your hand, and the spoons are in front of you." + }, + "answer_label": "D", + "answer_text": "You are holding 2 spoons in your hand, and the spoons are in front of you.", + "q_id": "P04_104_8_T1", + "correct_answer": "You are holding 2 spoons in your hand, and the spoons are in front of you.", + "original_options": { + "A": "1 spoon.", + "B": "3 spoons.", + "C": "You are holding 0 spoons in your hand, and no spoons are visible in front of you.", + "D": "You are holding 2 spoons in your hand, and the spoons are in front of you.", + "E": "You are holding 4 spoons in your hand, and the spoons are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 16 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_1_T1", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, which direction is the refrigerator relative to me?", + "answer": "At this moment, the refrigerator is to your front-right.", + "question_chinese": "此时，冰箱在我的哪个方向？", + "answer_chinese": "此时，冰箱在你的右前方。", + "evidence": [ + { + "description": "The brigde is front of your right side.", + "timestamps": { + "start": "00:03:13", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:03:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "At this moment, the refrigerator is behind you, slightly to the right as you face forward.", + "D": "At this moment, the refrigerator is to your front-right.", + "E": "At this moment, the refrigerator is in front of you, centered rather than off to the right." + }, + "answer_label": "D", + "answer_text": "At this moment, the refrigerator is to your front-right.", + "q_id": "P04_104_1_T1", + "correct_answer": "At this moment, the refrigerator is to your front-right.", + "original_options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "At this moment, the refrigerator is behind you, slightly to the right as you face forward.", + "D": "At this moment, the refrigerator is to your front-right.", + "E": "At this moment, the refrigerator is in front of you, centered rather than off to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_1_T2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, which direction is the refrigerator relative to me?", + "answer": "At this moment, the refrigerator is in front of you.", + "question_chinese": "此时，冰箱在我的哪个方向？", + "answer_chinese": "此时，冰箱在你前面。", + "evidence": [ + { + "description": "The brigde is o front of you.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:42" + } + } + ], + "question_timestamps": "00:03:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "At this moment, the refrigerator is to your right, not directly in front of your viewpoint.", + "D": "At this moment, the refrigerator is in front of you.", + "E": "At this moment, the refrigerator is to your front-right rather than straight ahead." + }, + "answer_label": "D", + "answer_text": "At this moment, the refrigerator is in front of you.", + "q_id": "P04_104_1_T2", + "correct_answer": "At this moment, the refrigerator is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "At this moment, the refrigerator is to your right, not directly in front of your viewpoint.", + "D": "At this moment, the refrigerator is in front of you.", + "E": "At this moment, the refrigerator is to your front-right rather than straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 16, + 13 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_6_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the kitchen sink relative to me?", + "answer": "The sink is to your left.", + "question_chinese": "洗碗池在我的哪个方位？", + "answer_chinese": "洗碗池在你左边。", + "evidence": [ + { + "description": "The sink is to the left of the stove.", + "timestamps": { + "start": "00:02:43", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sink is in front of you.", + "B": "The sink is to your right.", + "C": "The sink is slightly behind you on the right side, not on your left.", + "D": "The sink is to your left.", + "E": "The sink is behind you, so you would need to turn around to face it." + }, + "answer_label": "D", + "answer_text": "The sink is to your left.", + "q_id": "P04_104_6_T2", + "correct_answer": "The sink is to your left.", + "original_options": { + "A": "The sink is in front of you.", + "B": "The sink is to your right.", + "C": "The sink is slightly behind you on the right side, not on your left.", + "D": "The sink is to your left.", + "E": "The sink is behind you, so you would need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 15 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_6_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the kitchen sink relative to me?", + "answer": "The sink is in front of you.", + "question_chinese": "洗碗池在我的哪个方位？", + "answer_chinese": "洗碗池在你的前面。", + "evidence": [ + { + "description": "The sink is in front of you.", + "timestamps": { + "start": "00:02:49", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:02:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sink is to your left.", + "B": "The sink is in front of you.", + "C": "The sink is behind you.", + "D": "The sink is to your right, not straight ahead of you at the counter.", + "E": "The sink is off to your left side, not positioned in front of you." + }, + "answer_label": "B", + "answer_text": "The sink is in front of you.", + "q_id": "P04_104_6_T1", + "correct_answer": "The sink is in front of you.", + "original_options": { + "A": "The sink is to your left.", + "B": "The sink is in front of you.", + "C": "The sink is behind you.", + "D": "The sink is to your right, not straight ahead of you at the counter.", + "E": "The sink is off to your left side, not positioned in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 14 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_12", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the sink from me?", + "answer": "The sink is to your left.", + "question_chinese": "洗碗池在我的哪个方位？", + "answer_chinese": "洗碗池在你的左边。", + "evidence": [ + { + "description": "The sink is to the left of the stove.", + "timestamps": { + "start": "00:02:49", + "end": "00:03:05" + } + } + ], + "question_timestamps": "00:04:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The sink is to your right.", + "B": "The sink is behind you.", + "C": "The sink is in front of you, not off to either side.", + "D": "The sink is to your left.", + "E": "The sink sits to your right side, so you would turn right to face it." + }, + "answer_label": "D", + "answer_text": "The sink is to your left.", + "q_id": "P04_104_12", + "correct_answer": "The sink is to your left.", + "original_options": { + "A": "The sink is to your right.", + "B": "The sink is behind you.", + "C": "The sink is in front of you, not off to either side.", + "D": "The sink is to your left.", + "E": "The sink sits to your right side, so you would turn right to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 12, + 15 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_2_T2", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me, the microwave or the refrigerator?", + "answer": "The microwave is the closest to you; it is to your left.", + "question_chinese": "现在，微波炉和冰箱哪个离我最近？", + "answer_chinese": "微波炉离你最近，它在你左边。", + "evidence": [ + { + "description": "The stove is on your left.", + "timestamps": { + "start": "00:02:29", + "end": "00:02:31" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The refrigerator is the closest to you; it is in front of you.", + "B": "The microwave is the closest to you; it is to your left." + }, + "answer_label": "B", + "answer_text": "The microwave is the closest to you; it is to your left.", + "q_id": "P04_104_2_T2", + "correct_answer": "The microwave is the closest to you; it is to your left.", + "original_options": { + "A": "The refrigerator is the closest to you; it is in front of you.", + "B": "The microwave is the closest to you; it is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_2_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me, the microwave or the refrigerator?", + "answer": "The refrigerator is the closest to you; it is in front of you.", + "question_chinese": "现在，微波炉和冰箱哪个离我最近？", + "answer_chinese": "冰箱离你最近，它在你前面。", + "evidence": [ + { + "description": "The refrigerator is in front of you.", + "timestamps": { + "start": "00:03:22", + "end": "00:03:24" + } + } + ], + "question_timestamps": "00:03:26", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The microwave is the closest to you; it is to your left.", + "B": "The refrigerator is the closest to you; it is in front of you." + }, + "answer_label": "B", + "answer_text": "The refrigerator is the closest to you; it is in front of you.", + "q_id": "P04_104_2_T1", + "correct_answer": "The refrigerator is the closest to you; it is in front of you.", + "original_options": { + "A": "The microwave is the closest to you; it is to your left.", + "B": "The refrigerator is the closest to you; it is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_3", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me, the microwave or the refrigerator?", + "answer": "The microwave is closest to you.", + "question_chinese": "现在，微波炉和冰箱哪个离我最近？", + "answer_chinese": "微波炉离你最近。", + "evidence": [ + { + "description": "The refrigerator is to the right of the stove.", + "timestamps": { + "start": "00:03:22", + "end": "00:03:29" + } + }, + { + "description": "The microwave is to the left of the stove.", + "timestamps": { + "start": "00:02:37", + "end": "00:03:23" + } + }, + { + "description": "You are now standing to the left of the stove.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:06" + } + } + ], + "question_timestamps": "00:05:08", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The refrigerator is closest to you.", + "B": "The microwave is closest to you." + }, + "answer_label": "B", + "answer_text": "The microwave is closest to you.", + "q_id": "P04_104_3", + "correct_answer": "The microwave is closest to you.", + "original_options": { + "A": "The refrigerator is closest to you.", + "B": "The microwave is closest to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_4_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many pots can I see on the stove?", + "answer": "Right now you can see two pots; they are to your left.", + "question_chinese": "现在我能看到灶台上有几口锅？", + "answer_chinese": "现在你能看到两口锅，它们在你左边。", + "evidence": [ + { + "description": "There are two pots on the stove.", + "timestamps": { + "start": "00:08:42", + "end": "00:08:45" + } + } + ], + "question_timestamps": "00:08:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can see one pot.", + "B": "You can see three pots.", + "C": "Right now you can see four pots on the stove; they are positioned to your left.", + "D": "Right now you can see zero pots on the stove; the burners to your left are empty.", + "E": "Right now you can see two pots; they are to your left." + }, + "answer_label": "E", + "answer_text": "Right now you can see two pots; they are to your left.", + "q_id": "P04_104_4_T1", + "correct_answer": "Right now you can see two pots; they are to your left.", + "original_options": { + "A": "You can see one pot.", + "B": "You can see three pots.", + "C": "Right now you can see four pots on the stove; they are positioned to your left.", + "D": "Right now you can see zero pots on the stove; the burners to your left are empty.", + "E": "Right now you can see two pots; they are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_4_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many pots can I see on the stove?", + "answer": "Right now you can see 1 pot on the stove; it is to your right.", + "question_chinese": "现在我能看到灶台上有几口锅？", + "answer_chinese": "现在你能看到灶台上有1口锅，它在你右边。", + "evidence": [ + { + "description": "Took one pot from the stove and put it in the sink.", + "timestamps": { + "start": "00:08:57", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:09:02", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Right now you can see 1 pot on the stove; it is to your right.", + "B": "You can see two pots.", + "C": "You can see zero pots.", + "D": "Right now you can see three pots on the stove; they are all to your right.", + "E": "Right now you can see four pots on the stove; they are clustered together to your right." + }, + "answer_label": "A", + "answer_text": "Right now you can see 1 pot on the stove; it is to your right.", + "q_id": "P04_104_4_T2", + "correct_answer": "Right now you can see 1 pot on the stove; it is to your right.", + "original_options": { + "A": "Right now you can see 1 pot on the stove; it is to your right.", + "B": "You can see two pots.", + "C": "You can see zero pots.", + "D": "Right now you can see three pots on the stove; they are all to your right.", + "E": "Right now you can see four pots on the stove; they are clustered together to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_7", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are left on the stove now?", + "answer": "One.", + "question_chinese": "现在灶台上还剩几口锅？", + "answer_chinese": "一口。", + "evidence": [ + { + "description": "There are two pots on the stove.", + "timestamps": { + "start": "00:08:44", + "end": "00:08:58" + } + }, + { + "description": "Took one pot from the stove and put it in the sink.", + "timestamps": { + "start": "00:08:56", + "end": "00:09:01" + } + } + ], + "question_timestamps": "00:09:07", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "None.", + "B": "Two.", + "C": "Two pots are still on the stove, with both burners holding cookware side by side.", + "D": "Three pots remain on the stove now, spread across the back and front burners.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "P04_104_7", + "correct_answer": "One.", + "original_options": { + "A": "None.", + "B": "Two.", + "C": "Two pots are still on the stove, with both burners holding cookware side by side.", + "D": "Three pots remain on the stove now, spread across the back and front burners.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_8", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are left on the stove now?", + "answer": "One.", + "question_chinese": "现在灶台上还剩几口锅？", + "answer_chinese": "一口。", + "evidence": [ + { + "description": "There are two pots on the stove.", + "timestamps": { + "start": "00:08:44", + "end": "00:08:58" + } + }, + { + "description": "Took one pot from the stove and put it in the sink.", + "timestamps": { + "start": "00:08:56", + "end": "00:09:01" + } + } + ], + "question_timestamps": "00:09:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "None.", + "B": "One.", + "C": "Two.", + "D": "Three pots are left on the stove now, covering most of the available burners.", + "E": "Four pots remain on the stove, with multiple burners occupied at the same time." + }, + "answer_label": "B", + "answer_text": "One.", + "q_id": "P04_104_8", + "correct_answer": "One.", + "original_options": { + "A": "None.", + "B": "One.", + "C": "Two.", + "D": "Three pots are left on the stove now, covering most of the available burners.", + "E": "Four pots remain on the stove, with multiple burners occupied at the same time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_5_T1", + "qid_num": 25, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which direction is the cabinet with the batteries relative to me?", + "answer": "The cabinet for batteries is to your left.", + "question_chinese": "装电池的柜子在我哪个方向？", + "answer_chinese": "装电池的柜子在你左边", + "evidence": [ + { + "description": "The desk is on your left.", + "timestamps": { + "start": "00:04:40", + "end": "00:04:44" + } + } + ], + "question_timestamps": "00:04:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The cabinet for batteries is to your left.", + "B": "It’s in front of you.", + "C": "It’s to your right.", + "D": "The cabinet with the batteries is behind you, so you’d need to turn around to face it.", + "E": "The cabinet with the batteries is above you, positioned higher than eye level on the wall." + }, + "answer_label": "A", + "answer_text": "The cabinet for batteries is to your left.", + "q_id": "P04_104_5_T1", + "correct_answer": "The cabinet for batteries is to your left.", + "original_options": { + "A": "The cabinet for batteries is to your left.", + "B": "It’s in front of you.", + "C": "It’s to your right.", + "D": "The cabinet with the batteries is behind you, so you’d need to turn around to face it.", + "E": "The cabinet with the batteries is above you, positioned higher than eye level on the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 16 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_5_T2", + "qid_num": 26, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which direction is the cabinet with the batteries relative to me?", + "answer": "The cabinet for batteries is in front of you.", + "question_chinese": "装电池的柜子在我哪个方向？", + "answer_chinese": "装电池的柜子在你前面.", + "evidence": [ + { + "description": "The desk is in front of you.", + "timestamps": { + "start": "00:04:47", + "end": "00:04:52" + } + } + ], + "question_timestamps": "00:04:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s behind you.", + "B": "It’s to your left.", + "C": "The cabinet with the batteries is to your right, so you’d turn that way to reach it.", + "D": "The cabinet with the batteries is above you, sitting higher up than where you’re looking.", + "E": "The cabinet for batteries is in front of you." + }, + "answer_label": "E", + "answer_text": "The cabinet for batteries is in front of you.", + "q_id": "P04_104_5_T2", + "correct_answer": "The cabinet for batteries is in front of you.", + "original_options": { + "A": "It’s behind you.", + "B": "It’s to your left.", + "C": "The cabinet with the batteries is to your right, so you’d turn that way to reach it.", + "D": "The cabinet with the batteries is above you, sitting higher up than where you’re looking.", + "E": "The cabinet for batteries is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 16 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_7_T1", + "qid_num": 27, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the drawer for tableware relative to me?", + "answer": "The drawer for tableware is in front of you.", + "question_chinese": "装餐具的抽屉在我的哪个方位？", + "answer_chinese": "装餐具的抽屉在你的前面。", + "evidence": [ + { + "description": "The cutlery drawer is in front of you.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s behind you.", + "B": "It’s to your left.", + "C": "The drawer for tableware is to your right, so you’d reach over to that side to open it.", + "D": "The drawer for tableware is above you, positioned higher than your current line of sight.", + "E": "The drawer for tableware is in front of you." + }, + "answer_label": "E", + "answer_text": "The drawer for tableware is in front of you.", + "q_id": "P04_104_7_T1", + "correct_answer": "The drawer for tableware is in front of you.", + "original_options": { + "A": "It’s behind you.", + "B": "It’s to your left.", + "C": "The drawer for tableware is to your right, so you’d reach over to that side to open it.", + "D": "The drawer for tableware is above you, positioned higher than your current line of sight.", + "E": "The drawer for tableware is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 19, + 15 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_7_T2", + "qid_num": 28, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the drawer for tableware relative to me?", + "answer": "The drawer for tableware is behind you.", + "question_chinese": "装餐具的抽屉在我的哪个方位？", + "answer_chinese": "装餐具的抽屉在你的后面。", + "evidence": [ + { + "description": "The cutlery drawer is behind of you.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s to your right.", + "C": "The drawer for tableware is to your left, so you’d turn slightly that way to access it.", + "D": "The drawer for tableware is above you, located higher up rather than at counter level.", + "E": "The drawer for tableware is behind you." + }, + "answer_label": "E", + "answer_text": "The drawer for tableware is behind you.", + "q_id": "P04_104_7_T2", + "correct_answer": "The drawer for tableware is behind you.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s to your right.", + "C": "The drawer for tableware is to your left, so you’d turn slightly that way to access it.", + "D": "The drawer for tableware is above you, located higher up rather than at counter level.", + "E": "The drawer for tableware is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 15 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_13", + "qid_num": 29, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the cutlery drawer from the sink?", + "answer": "The cutlery drawer is to the left of the sink.", + "question_chinese": "装餐具的抽屉在洗碗池的哪个方位？", + "answer_chinese": "装餐具的抽屉在洗碗池的左边。", + "evidence": [ + { + "description": "The cutlery drawer is to the left of the sink.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The cutlery drawer is to the left of the sink.", + "B": "To the right of the sink.", + "C": "In front of the sink.", + "D": "The cutlery drawer is behind the sink, so it sits on the far side away from you.", + "E": "The cutlery drawer is diagonally opposite the sink, offset rather than directly beside it." + }, + "answer_label": "A", + "answer_text": "The cutlery drawer is to the left of the sink.", + "q_id": "P04_104_13", + "correct_answer": "The cutlery drawer is to the left of the sink.", + "original_options": { + "A": "The cutlery drawer is to the left of the sink.", + "B": "To the right of the sink.", + "C": "In front of the sink.", + "D": "The cutlery drawer is behind the sink, so it sits on the far side away from you.", + "E": "The cutlery drawer is diagonally opposite the sink, offset rather than directly beside it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 14 + ] + }, + { + "video_uid": "P04_104", + "video_length": "00:11:24", + "qid": "P04_104_14", + "qid_num": 30, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the cutlery drawer from the sink?", + "answer": "The cutlery drawer is to the left of the sink.", + "question_chinese": "装餐具的抽屉在洗碗池的哪个方位？", + "answer_chinese": "装餐具的抽屉在洗碗池的左边。", + "evidence": [ + { + "description": "The cutlery drawer is to the left of the sink.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:04:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To the right of the sink.", + "B": "The cutlery drawer is to the left of the sink.", + "C": "Directly in front of the sink.", + "D": "The cutlery drawer is behind the sink, positioned on the far side instead of beside it.", + "E": "The cutlery drawer is diagonally opposite the sink, offset across from it rather than adjacent." + }, + "answer_label": "B", + "answer_text": "The cutlery drawer is to the left of the sink.", + "q_id": "P04_104_14", + "correct_answer": "The cutlery drawer is to the left of the sink.", + "original_options": { + "A": "To the right of the sink.", + "B": "The cutlery drawer is to the left of the sink.", + "C": "Directly in front of the sink.", + "D": "The cutlery drawer is behind the sink, positioned on the far side instead of beside it.", + "E": "The cutlery drawer is diagonally opposite the sink, offset across from it rather than adjacent." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + } +] diff --git a/QAs/P04_112-mcq.json b/QAs/P04_112-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2d6e0e8b08924e4f8f7553d491f6a96a733c94c1 --- /dev/null +++ b/QAs/P04_112-mcq.json @@ -0,0 +1,909 @@ +[ + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_4_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I use plastic wrap in total just now?", + "answer": "Once. The plastic wrap is to your left.", + "question_chinese": "我刚才一共用了几次保鲜膜？", + "answer_chinese": "一次。保鲜膜在你的左方。", + "evidence": [ + { + "description": "Used plastic wrap for the first time.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:17" + } + } + ], + "question_timestamps": "00:07:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Twice, to your left.", + "B": "Three times, to your left.", + "C": "Twice. The plastic wrap is to your left when you reach for it again after the first use.", + "D": "Once. The plastic wrap is to your left.", + "E": "Three times. The plastic wrap stays to your left each time you pull out a new sheet." + }, + "answer_label": "D", + "answer_text": "Once. The plastic wrap is to your left.", + "q_id": "P04_112_4_T1", + "correct_answer": "Once. The plastic wrap is to your left.", + "original_options": { + "A": "Twice, to your left.", + "B": "Three times, to your left.", + "C": "Twice. The plastic wrap is to your left when you reach for it again after the first use.", + "D": "Once. The plastic wrap is to your left.", + "E": "Three times. The plastic wrap stays to your left each time you pull out a new sheet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 17 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_4_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I use plastic wrap in total just now?", + "answer": "Twice. The plastic wrap is to your back-left.", + "question_chinese": "我刚才一共用了几次保鲜膜？", + "answer_chinese": "两次。保鲜膜在你的左后方。", + "evidence": [ + { + "description": "Used plastic wrap for the first time.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:17" + } + }, + { + "description": "Used plastic wrap for the second time.", + "timestamps": { + "start": "00:07:29", + "end": "00:07:53" + } + } + ], + "question_timestamps": "00:07:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once, to your back-left.", + "B": "Three times, to your back-left.", + "C": "Once. The plastic wrap is to your back-left, and you only pulled and tore it a single time.", + "D": "Three times. The plastic wrap is to your back-left, and you returned to it for two additional wraps.", + "E": "Twice. The plastic wrap is to your back-left." + }, + "answer_label": "E", + "answer_text": "Twice. The plastic wrap is to your back-left.", + "q_id": "P04_112_4_T2", + "correct_answer": "Twice. The plastic wrap is to your back-left.", + "original_options": { + "A": "Once, to your back-left.", + "B": "Three times, to your back-left.", + "C": "Once. The plastic wrap is to your back-left, and you only pulled and tore it a single time.", + "D": "Three times. The plastic wrap is to your back-left, and you returned to it for two additional wraps.", + "E": "Twice. The plastic wrap is to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 18 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_3_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "A moment ago, what route did I take when I carried the plastic wrap to the stove?", + "answer": "You turned back from in front of the stove and walked to the storage cabinet, took plastic wrap out of the drawer, then walked right from the storage cabinet back to the stove.", + "question_chinese": "我刚才拿保鲜膜到灶台走了什么路径？", + "answer_chinese": "从灶台前向后转走到储物柜前，从抽屉里取出保鲜膜，然后从储物柜前向右走到灶台前。", + "evidence": [ + { + "description": "Turned around from the front of the stove, walked to the storage cabinet, and took out the plastic wrap from the drawer.", + "timestamps": { + "start": "00:05:55", + "end": "00:06:11" + } + }, + { + "description": "Walked from the front of the storage cabinet to the right to the front of the stove.", + "timestamps": { + "start": "00:06:11", + "end": "00:06:15" + } + } + ], + "question_timestamps": "00:06:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked left back to the stove.", + "B": "You started from behind the stove.", + "C": "You turned back from in front of the stove and walked to the storage cabinet, took plastic wrap out of the drawer, then walked right from the storage cabinet back to the stove.", + "D": "You turned back from in front of the stove, went to the storage cabinet, then walked left back to the stove.", + "E": "You turned back from in front of the stove, grabbed the plastic wrap, then detoured left before returning to the stove." + }, + "answer_label": "C", + "answer_text": "You turned back from in front of the stove and walked to the storage cabinet, took plastic wrap out of the drawer, then walked right from the storage cabinet back to the stove.", + "q_id": "P04_112_3_T1", + "correct_answer": "You turned back from in front of the stove and walked to the storage cabinet, took plastic wrap out of the drawer, then walked right from the storage cabinet back to the stove.", + "original_options": { + "A": "You walked left back to the stove.", + "B": "You started from behind the stove.", + "C": "You turned back from in front of the stove and walked to the storage cabinet, took plastic wrap out of the drawer, then walked right from the storage cabinet back to the stove.", + "D": "You turned back from in front of the stove, went to the storage cabinet, then walked left back to the stove.", + "E": "You turned back from in front of the stove, grabbed the plastic wrap, then detoured left before returning to the stove." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 21, + 21 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_3_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "A moment ago, what route did I take when I carried the plastic wrap to the stove?", + "answer": "You turned back from in front of the stove and walked to the storage cabinet.", + "question_chinese": "我刚才拿保鲜膜到灶台走了什么路径？", + "answer_chinese": "从灶台前向后转走到储物柜前。", + "evidence": [ + { + "description": "Turned around from the front of the stove, walked to the storage cabinet, and took out the plastic wrap from the drawer.", + "timestamps": { + "start": "00:05:55", + "end": "00:06:11" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned back from behind the stove.", + "B": "You turned forward from the stove.", + "C": "You turned back from in front of the stove and walked away from the storage cabinet instead of toward it.", + "D": "You turned back from in front of the stove and walked to the storage cabinet.", + "E": "You turned back from in front of the stove, went to the storage cabinet, then continued on to the stove." + }, + "answer_label": "D", + "answer_text": "You turned back from in front of the stove and walked to the storage cabinet.", + "q_id": "P04_112_3_T2", + "correct_answer": "You turned back from in front of the stove and walked to the storage cabinet.", + "original_options": { + "A": "You turned back from behind the stove.", + "B": "You turned forward from the stove.", + "C": "You turned back from in front of the stove and walked away from the storage cabinet instead of toward it.", + "D": "You turned back from in front of the stove and walked to the storage cabinet.", + "E": "You turned back from in front of the stove, went to the storage cabinet, then continued on to the stove." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 20 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_5_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many heated flatbreads are in the baking tray on the stove?", + "answer": "Three, to your back-right.", + "question_chinese": "我现在灶台上的烤盘里有几个加热好的饼？", + "answer_chinese": "三个，在你右后方。", + "evidence": [ + { + "description": "Took out three heated cakes from the electric hot pot and placed them in the baking pan on the stove.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:38" + } + }, + { + "description": "Took an unheated cake from the baking pan and placed it in the electric hot pot.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:01:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, to your back-right.", + "B": "Two, to your back-right.", + "C": "Four, to your back-right, filling most of the baking tray on the stove.", + "D": "Five, to your back-right, stacked closely together in the baking tray on the stove.", + "E": "Three, to your back-right." + }, + "answer_label": "E", + "answer_text": "Three, to your back-right.", + "q_id": "P04_112_5_T1", + "correct_answer": "Three, to your back-right.", + "original_options": { + "A": "One, to your back-right.", + "B": "Two, to your back-right.", + "C": "Four, to your back-right, filling most of the baking tray on the stove.", + "D": "Five, to your back-right, stacked closely together in the baking tray on the stove.", + "E": "Three, to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_5_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many heated flatbreads are in the baking tray on the stove?", + "answer": "Three, to your front-left.", + "question_chinese": "我现在灶台上的烤盘里有几个加热好的饼？", + "answer_chinese": "三个，在你的左前方。", + "evidence": [ + { + "description": "Took out three heated cakes from the electric hot pot and placed them in the baking pan on the stove.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:38" + } + }, + { + "description": "Took an unheated cake from the baking pan and placed it in the electric hot pot.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:01:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three, to your front-left.", + "B": "One, to your front-left.", + "C": "Two, to your front-left.", + "D": "Four, to your front-left, spread across the baking tray sitting on the stove.", + "E": "Five, to your front-left, arranged tightly together in the baking tray on the stove." + }, + "answer_label": "A", + "answer_text": "Three, to your front-left.", + "q_id": "P04_112_5_T2", + "correct_answer": "Three, to your front-left.", + "original_options": { + "A": "Three, to your front-left.", + "B": "One, to your front-left.", + "C": "Two, to your front-left.", + "D": "Four, to your front-left, spread across the baking tray sitting on the stove.", + "E": "Five, to your front-left, arranged tightly together in the baking tray on the stove." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_7_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the microwave relative to the sink?", + "answer": "The microwave is above and to the right of the sink, and it’s to your back-left.", + "question_chinese": "我看到微波炉在洗碗池的哪个方位？", + "answer_chinese": "微波炉在洗碗池的右上方，在你的左后方。", + "evidence": [ + { + "description": "The microwave is to the upper right of the sink.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:03" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above-left of sink, to your back-left.", + "B": "Above-right of sink, to your back-right.", + "C": "The microwave is above and to the right of the sink, and it’s to your back-left.", + "D": "The microwave is below and to the right of the sink, and it’s positioned to your back-left.", + "E": "The microwave is above and to the right of the sink, but you saw it from your front-left." + }, + "answer_label": "C", + "answer_text": "The microwave is above and to the right of the sink, and it’s to your back-left.", + "q_id": "P04_112_7_T2", + "correct_answer": "The microwave is above and to the right of the sink, and it’s to your back-left.", + "original_options": { + "A": "Above-left of sink, to your back-left.", + "B": "Above-right of sink, to your back-right.", + "C": "The microwave is above and to the right of the sink, and it’s to your back-left.", + "D": "The microwave is below and to the right of the sink, and it’s positioned to your back-left.", + "E": "The microwave is above and to the right of the sink, but you saw it from your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_7_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the microwave relative to the sink?", + "answer": "The microwave is above and to the right of the sink, and it’s to your left.", + "question_chinese": "我看到微波炉在洗碗池的哪个方位？", + "answer_chinese": "微波炉在洗碗池的右上方，在你的左方", + "evidence": [ + { + "description": "The microwave is to the upper right of the sink.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:03" + } + } + ], + "question_timestamps": "00:02:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above-right of sink, to your right.", + "B": "The microwave is above and to the right of the sink, and it’s to your left.", + "C": "Above-left of sink, to your left.", + "D": "The microwave is below and to the right of the sink, and it’s still to your left.", + "E": "The microwave is above and to the right of the sink, and you saw it over your back-left." + }, + "answer_label": "B", + "answer_text": "The microwave is above and to the right of the sink, and it’s to your left.", + "q_id": "P04_112_7_T1", + "correct_answer": "The microwave is above and to the right of the sink, and it’s to your left.", + "original_options": { + "A": "Above-right of sink, to your right.", + "B": "The microwave is above and to the right of the sink, and it’s to your left.", + "C": "Above-left of sink, to your left.", + "D": "The microwave is below and to the right of the sink, and it’s still to your left.", + "E": "The microwave is above and to the right of the sink, and you saw it over your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_8_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots in total did I see on the stove right now?", + "answer": "One pot, to your front-left.", + "question_chinese": "我看到现在灶台上一共有几口锅？", + "answer_chinese": "一口锅。在你的左前方", + "evidence": [ + { + "description": "There is only one frying pan on the stove.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero pots, to your front-left.", + "B": "Two pots, to your front-left.", + "C": "One pot, to your front-left.", + "D": "Three pots total, all sitting on the stove to your front-left right now.", + "E": "Four pots total, grouped together on the stove at your front-left right now." + }, + "answer_label": "C", + "answer_text": "One pot, to your front-left.", + "q_id": "P04_112_8_T1", + "correct_answer": "One pot, to your front-left.", + "original_options": { + "A": "Zero pots, to your front-left.", + "B": "Two pots, to your front-left.", + "C": "One pot, to your front-left.", + "D": "Three pots total, all sitting on the stove to your front-left right now.", + "E": "Four pots total, grouped together on the stove at your front-left right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_8_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots in total did I see on the stove right now?", + "answer": "One pot, to your back-right.", + "question_chinese": "我看到现在灶台上一共有几口锅？", + "answer_chinese": "一口锅，在你的右后方。", + "evidence": [ + { + "description": "There is only one frying pan on the stove.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:01:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero pots, to your back-right.", + "B": "Two pots, to your back-right.", + "C": "Three pots total, visible on the stove to your back-right right now clearly.", + "D": "One pot, to your back-right.", + "E": "Four pots in total, all on the stove behind you to the right at once." + }, + "answer_label": "D", + "answer_text": "One pot, to your back-right.", + "q_id": "P04_112_8_T2", + "correct_answer": "One pot, to your back-right.", + "original_options": { + "A": "Zero pots, to your back-right.", + "B": "Two pots, to your back-right.", + "C": "Three pots total, visible on the stove to your back-right right now clearly.", + "D": "One pot, to your back-right.", + "E": "Four pots in total, all on the stove behind you to the right at once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 15 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_6_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates are on my dining table?", + "answer": "Four. To your back-left.", + "question_chinese": "我餐桌上有几个碟子？", + "answer_chinese": "4个。在你的左后方。", + "evidence": [ + { + "description": "When you just entered the kitchen, there were 3 dishes on the dining table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four. To your back-left.", + "B": "Two. To your back-left.", + "C": "Three. To your back-left.", + "D": "Five. To your back-left, all laid out on the dining table in one row.", + "E": "Six. To your back-left, spread across the dining table in the same area." + }, + "answer_label": "A", + "answer_text": "Four. To your back-left.", + "q_id": "P04_112_6_T1", + "correct_answer": "Four. To your back-left.", + "original_options": { + "A": "Four. To your back-left.", + "B": "Two. To your back-left.", + "C": "Three. To your back-left.", + "D": "Five. To your back-left, all laid out on the dining table in one row.", + "E": "Six. To your back-left, spread across the dining table in the same area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 13 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_6_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates are on my dining table?", + "answer": "Three, in front of you.", + "question_chinese": "我餐桌上有几个碟子？", + "answer_chinese": "3个，在你的前方。", + "evidence": [ + { + "description": "When you just entered the kitchen, there were 3 dishes on the dining table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, in front of you.", + "B": "Three, in front of you.", + "C": "Two, in front of you.", + "D": "Four, in front of you, arranged together on the dining table in one cluster.", + "E": "Five, in front of you, all placed on the dining table in one cluster." + }, + "answer_label": "B", + "answer_text": "Three, in front of you.", + "q_id": "P04_112_6_T2", + "correct_answer": "Three, in front of you.", + "original_options": { + "A": "One, in front of you.", + "B": "Three, in front of you.", + "C": "Two, in front of you.", + "D": "Four, in front of you, arranged together on the dining table in one cluster.", + "E": "Five, in front of you, all placed on the dining table in one cluster." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the fridge now?", + "answer": "three times.", + "question_chinese": "现在我一共开了几次冰箱？", + "answer_chinese": "3次。", + "evidence": [ + { + "description": "Opened the fridge for the first time.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:54" + } + }, + { + "description": "Opened the fridge for the second time.", + "timestamps": { + "start": "00:03:14", + "end": "00:03:19" + } + }, + { + "description": "Opened the fridge for the third time.", + "timestamps": { + "start": "00:06:17", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:07:32", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "three times.", + "C": "Twice.", + "D": "Four times, counting each time you pulled the fridge door open during this sequence.", + "E": "Five times, including the last open-close cycle you did at the fridge." + }, + "answer_label": "B", + "answer_text": "three times.", + "q_id": "P04_112_1", + "correct_answer": "three times.", + "original_options": { + "A": "Once.", + "B": "three times.", + "C": "Twice.", + "D": "Four times, counting each time you pulled the fridge door open during this sequence.", + "E": "Five times, including the last open-close cycle you did at the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 12 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the fridge now?", + "answer": "Four times.", + "question_chinese": "现在我一共开了几次冰箱？", + "answer_chinese": "4次。", + "evidence": [ + { + "description": "Opened the fridge for the first time.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:54" + } + }, + { + "description": "Opened the fridge for the second time.", + "timestamps": { + "start": "00:03:14", + "end": "00:03:19" + } + }, + { + "description": "Opened the fridge for the third time.", + "timestamps": { + "start": "00:06:17", + "end": "00:07:10" + } + }, + { + "description": "Opened the fridge for the fourth time.", + "timestamps": { + "start": "00:07:55", + "end": "00:08:08" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five times.", + "B": "Six times.", + "C": "Four times.", + "D": "Seven times, based on the number of times you opened the fridge door overall.", + "E": "Eight times, counting every separate fridge opening you did up to this point." + }, + "answer_label": "C", + "answer_text": "Four times.", + "q_id": "P04_112_2", + "correct_answer": "Four times.", + "original_options": { + "A": "Five times.", + "B": "Six times.", + "C": "Four times.", + "D": "Seven times, based on the number of times you opened the fridge door overall.", + "E": "Eight times, counting every separate fridge opening you did up to this point." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_2_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me, the microwave or the refrigerator?", + "answer": "The microwave is closer, to your front-right.", + "question_chinese": "现在我距离微波炉和冰箱，哪个更近？", + "answer_chinese": "微波炉更近，在你的右前方。", + "evidence": [ + { + "description": "The dishwasher is about 1 meter to the left of the microwave.", + "timestamps": { + "start": "00:07:18", + "end": "00:07:25" + } + }, + { + "description": "You are now standing in front of the sink.", + "timestamps": { + "start": "00:07:26", + "end": "00:07:35" + } + } + ], + "question_timestamps": "00:07:40", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The microwave is closer, to your front-right.", + "B": "The refrigerator is closer, to your front-right." + }, + "answer_label": "A", + "answer_text": "The microwave is closer, to your front-right.", + "q_id": "P04_112_2_T1", + "correct_answer": "The microwave is closer, to your front-right.", + "original_options": { + "A": "The microwave is closer, to your front-right.", + "B": "The refrigerator is closer, to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P04_112", + "video_length": "00:08:11", + "qid": "P04_112_2_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me, the microwave or the refrigerator?", + "answer": "The microwave is closer, in front of you.", + "question_chinese": "现在我距离微波炉和冰箱，哪个更近？", + "answer_chinese": "微波炉更近，在你的前方。", + "evidence": [ + { + "description": "The fridge is about two meters to the right of the microwave.", + "timestamps": { + "start": "00:07:07", + "end": "00:07:19" + } + }, + { + "description": "You are now standing in front of the sink.", + "timestamps": { + "start": "00:07:26", + "end": "00:07:35" + } + } + ], + "question_timestamps": "00:07:51", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The refrigerator is closer, in front of you.", + "B": "The microwave is closer, in front of you." + }, + "answer_label": "B", + "answer_text": "The microwave is closer, in front of you.", + "q_id": "P04_112_2_T2", + "correct_answer": "The microwave is closer, in front of you.", + "original_options": { + "A": "The refrigerator is closer, in front of you.", + "B": "The microwave is closer, in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + } +] diff --git a/QAs/P04_119-mcq.json b/QAs/P04_119-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e0edd0f1b4d8ca03b50e7f6d12cb38ff81062f2b --- /dev/null +++ b/QAs/P04_119-mcq.json @@ -0,0 +1,1022 @@ +[ + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I'm cooking at the stove, where is the refrigerator in my position?", + "answer": "The fridge is on your right.", + "question_chinese": "当我面向灶台做饭时，冰箱在我的哪个方位？", + "answer_chinese": "冰箱在你的右侧。", + "evidence": [ + { + "description": "Turn right from the stove to the refrigerator.", + "timestamps": { + "start": "00:03:34", + "end": "00:03:37" + } + } + ], + "question_timestamps": "00:08:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The fridge is on your right.", + "B": "The fridge is on your left.", + "C": "The fridge is behind you.", + "D": "The refrigerator is directly in front of you while you cook at the stove, across from your position.", + "E": "The refrigerator is slightly behind you on the left side as you face the stove to cook." + }, + "answer_label": "A", + "answer_text": "The fridge is on your right.", + "q_id": "P04_119_1", + "correct_answer": "The fridge is on your right.", + "original_options": { + "A": "The fridge is on your right.", + "B": "The fridge is on your left.", + "C": "The fridge is behind you.", + "D": "The refrigerator is directly in front of you while you cook at the stove, across from your position.", + "E": "The refrigerator is slightly behind you on the left side as you face the stove to cook." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 17 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I'm cooking at the stove, where is the refrigerator in my position?", + "answer": "The fridge is on your right.", + "question_chinese": "当我面向灶台做饭时，冰箱在我的哪个方位？", + "answer_chinese": "冰箱在你的右侧。", + "evidence": [ + { + "description": "Turn right from the stove to the refrigerator.", + "timestamps": { + "start": "00:03:34", + "end": "00:03:37" + } + } + ], + "question_timestamps": "00:14:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The fridge is on your left.", + "B": "The fridge is in front of you.", + "C": "The refrigerator is behind you as you stand at the stove, so you’d need to turn around.", + "D": "The fridge is on your right.", + "E": "The fridge is positioned slightly ahead on your left while you’re facing the stove and cooking." + }, + "answer_label": "D", + "answer_text": "The fridge is on your right.", + "q_id": "P04_119_2", + "correct_answer": "The fridge is on your right.", + "original_options": { + "A": "The fridge is on your left.", + "B": "The fridge is in front of you.", + "C": "The refrigerator is behind you as you stand at the stove, so you’d need to turn around.", + "D": "The fridge is on your right.", + "E": "The fridge is positioned slightly ahead on your left while you’re facing the stove and cooking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 17 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_6_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of milk did I take out of the fridge?", + "answer": "You took one bottle. The refrigerator was in front of you.", + "question_chinese": "我从冰箱里拿出了几瓶牛奶？", + "answer_chinese": "你拿了一瓶。冰箱在你的前面。", + "evidence": [ + { + "description": "Take out the first bottle of milk.", + "timestamps": { + "start": "00:03:36", + "end": "00:03:38" + } + } + ], + "question_timestamps": "00:03:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took zero bottles.", + "B": "You took two bottles.", + "C": "You took one bottle. The refrigerator was in front of you.", + "D": "You took three bottles of milk. The refrigerator was in front of you when you removed them.", + "E": "You took four bottles of milk. The refrigerator was in front of you as you took them out." + }, + "answer_label": "C", + "answer_text": "You took one bottle. The refrigerator was in front of you.", + "q_id": "P04_119_6_T1", + "correct_answer": "You took one bottle. The refrigerator was in front of you.", + "original_options": { + "A": "You took zero bottles.", + "B": "You took two bottles.", + "C": "You took one bottle. The refrigerator was in front of you.", + "D": "You took three bottles of milk. The refrigerator was in front of you when you removed them.", + "E": "You took four bottles of milk. The refrigerator was in front of you as you took them out." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_6_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of milk did I take out of the fridge?", + "answer": "You took two bottles. The refrigerator was to your right.", + "question_chinese": "我从冰箱里拿出了几瓶牛奶？", + "answer_chinese": "你拿了两瓶。冰箱在你的右边。", + "evidence": [ + { + "description": "Take out the first bottle of milk.", + "timestamps": { + "start": "00:03:36", + "end": "00:03:39" + } + }, + { + "description": "Take out the second bottle of milk.", + "timestamps": { + "start": "00:03:39", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took zero bottles.", + "B": "You took one bottle.", + "C": "You took three bottles of milk. The refrigerator was to your right when you pulled them out.", + "D": "You took two bottles. The refrigerator was to your right.", + "E": "You took four bottles of milk. The refrigerator was to your right as you removed them." + }, + "answer_label": "D", + "answer_text": "You took two bottles. The refrigerator was to your right.", + "q_id": "P04_119_6_T2", + "correct_answer": "You took two bottles. The refrigerator was to your right.", + "original_options": { + "A": "You took zero bottles.", + "B": "You took one bottle.", + "C": "You took three bottles of milk. The refrigerator was to your right when you pulled them out.", + "D": "You took two bottles. The refrigerator was to your right.", + "E": "You took four bottles of milk. The refrigerator was to your right as you removed them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_11", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of milk did I take out from the fridge?", + "answer": "Two bottles.", + "question_chinese": "我从冰箱里拿出了几瓶牛奶？", + "answer_chinese": "两瓶。", + "evidence": [ + { + "description": "Take out the first bottle of milk.", + "timestamps": { + "start": "00:03:36", + "end": "00:03:39" + } + }, + { + "description": "Take out the second bottle of milk.", + "timestamps": { + "start": "00:03:39", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:10:36", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One bottle.", + "B": "Three bottles.", + "C": "You took four bottles of milk from the fridge before putting the container back inside.", + "D": "You took five bottles of milk from the fridge, then closed the refrigerator door afterward.", + "E": "Two bottles." + }, + "answer_label": "E", + "answer_text": "Two bottles.", + "q_id": "P04_119_11", + "correct_answer": "Two bottles.", + "original_options": { + "A": "One bottle.", + "B": "Three bottles.", + "C": "You took four bottles of milk from the fridge before putting the container back inside.", + "D": "You took five bottles of milk from the fridge, then closed the refrigerator door afterward.", + "E": "Two bottles." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_12", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of milk did I take out from the fridge?", + "answer": "Two bottles.", + "question_chinese": "我从冰箱里拿出了几瓶牛奶？", + "answer_chinese": "两瓶。", + "evidence": [ + { + "description": "Take out the first bottle of milk.", + "timestamps": { + "start": "00:03:36", + "end": "00:03:39" + } + }, + { + "description": "Take out the second bottle of milk.", + "timestamps": { + "start": "00:03:39", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:12:58", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One bottle.", + "B": "Three bottles.", + "C": "You took four bottles of milk from the fridge before putting the container back inside.", + "D": "Two bottles.", + "E": "You took five bottles of milk from the fridge, then closed the refrigerator door afterward." + }, + "answer_label": "D", + "answer_text": "Two bottles.", + "q_id": "P04_119_12", + "correct_answer": "Two bottles.", + "original_options": { + "A": "One bottle.", + "B": "Three bottles.", + "C": "You took four bottles of milk from the fridge before putting the container back inside.", + "D": "Two bottles.", + "E": "You took five bottles of milk from the fridge, then closed the refrigerator door afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_8_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pills did I take? How many are left now?", + "answer": "You took zero pills, and 9 were left. The pills were on the cutting board in front of you.", + "question_chinese": "我取了几片药？现在还剩几片？", + "answer_chinese": "你取了零片药，还剩9片。药在你前面的菜板上面。", + "evidence": [ + { + "description": "There were nine pills at first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took one pill, 8 left.", + "B": "You took two pills, 7 left.", + "C": "You took three pills, and 6 were left. The pills were on the cutting board in front of you.", + "D": "You took four pills, and 5 were left. The pills were on the cutting board in front of you.", + "E": "You took zero pills, and 9 were left. The pills were on the cutting board in front of you." + }, + "answer_label": "E", + "answer_text": "You took zero pills, and 9 were left. The pills were on the cutting board in front of you.", + "q_id": "P04_119_8_T1", + "correct_answer": "You took zero pills, and 9 were left. The pills were on the cutting board in front of you.", + "original_options": { + "A": "You took one pill, 8 left.", + "B": "You took two pills, 7 left.", + "C": "You took three pills, and 6 were left. The pills were on the cutting board in front of you.", + "D": "You took four pills, and 5 were left. The pills were on the cutting board in front of you.", + "E": "You took zero pills, and 9 were left. The pills were on the cutting board in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_15", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pills did I take? How many are left?", + "answer": "You took one pill and you have eight left.", + "question_chinese": "我取了几片药？现在还剩几片？", + "answer_chinese": "你取了一片药，还剩8片。", + "evidence": [ + { + "description": "There were nine pills at first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + }, + { + "description": "Take out a pill and put the rest in the box on the right.", + "timestamps": { + "start": "00:02:11", + "end": "00:02:22" + } + }, + { + "description": "Put the rest of the pills in the box on the right.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:34" + } + } + ], + "question_timestamps": "00:10:11", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You took one pill and you have eight left.", + "B": "You took two pills, seven left.", + "C": "You took three pills, six left.", + "D": "You took four pills and you have five left after taking them.", + "E": "You took five pills and you have four left after taking them." + }, + "answer_label": "A", + "answer_text": "You took one pill and you have eight left.", + "q_id": "P04_119_15", + "correct_answer": "You took one pill and you have eight left.", + "original_options": { + "A": "You took one pill and you have eight left.", + "B": "You took two pills, seven left.", + "C": "You took three pills, six left.", + "D": "You took four pills and you have five left after taking them.", + "E": "You took five pills and you have four left after taking them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 12 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_8_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pills did I take? How many are left now?", + "answer": "You took one pill, and 8 were left. You put the pill into the box to your left-hand side.", + "question_chinese": "我取了几片药？现在还剩几片？", + "answer_chinese": "你取了一片药，还剩8片。药被你放进了左手边的盒子里。", + "evidence": [ + { + "description": "There were nine pills at first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + }, + { + "description": "Take out a pill and put the rest in the box on the right.", + "timestamps": { + "start": "00:02:11", + "end": "00:02:22" + } + }, + { + "description": "Put the rest of the pills in the box on the right.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:34" + } + } + ], + "question_timestamps": "00:14:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took one pill, and 8 were left. You put the pill into the box to your left-hand side.", + "B": "You took zero pills; 9 were left.", + "C": "You took two pills; 7 were left.", + "D": "You took three pills, leaving 6 in the pack, and then set the pills into the box on your left.", + "E": "You took four pills, so 5 remained afterward, and you placed the pills into the box on your left-hand side." + }, + "answer_label": "A", + "answer_text": "You took one pill, and 8 were left. You put the pill into the box to your left-hand side.", + "q_id": "P04_119_8_T2", + "correct_answer": "You took one pill, and 8 were left. You put the pill into the box to your left-hand side.", + "original_options": { + "A": "You took one pill, and 8 were left. You put the pill into the box to your left-hand side.", + "B": "You took zero pills; 9 were left.", + "C": "You took two pills; 7 were left.", + "D": "You took three pills, leaving 6 in the pack, and then set the pills into the box on your left.", + "E": "You took four pills, so 5 remained afterward, and you placed the pills into the box on your left-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 20 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_10", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get those yellow rubber gloves now?", + "answer": "You put the yellow rubber gloves in the middle of the sink and stove.", + "question_chinese": "我现在可以够得到那双黄色橡胶手套吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "You put the yellow rubber gloves in the middle of the sink and stove.", + "timestamps": { + "start": "00:01:52", + "end": "00:02:39" + } + }, + { + "description": "You're now squatting in front of the cupboard between the sink and the stove.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:12" + } + } + ], + "question_timestamps": "00:06:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "They’re behind the sink.", + "B": "They’re next to the fridge.", + "C": "You put the yellow rubber gloves in the middle of the sink and stove.", + "D": "You left the yellow rubber gloves on the edge by the sink, closer to the counter than the stove.", + "E": "You placed the yellow rubber gloves between the stove and the refrigerator, not in the gap beside the sink." + }, + "answer_label": "C", + "answer_text": "You put the yellow rubber gloves in the middle of the sink and stove.", + "q_id": "P04_119_10", + "correct_answer": "You put the yellow rubber gloves in the middle of the sink and stove.", + "original_options": { + "A": "They’re behind the sink.", + "B": "They’re next to the fridge.", + "C": "You put the yellow rubber gloves in the middle of the sink and stove.", + "D": "You left the yellow rubber gloves on the edge by the sink, closer to the counter than the stove.", + "E": "You placed the yellow rubber gloves between the stove and the refrigerator, not in the gap beside the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 19 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_9", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get those yellow rubber gloves now?", + "answer": "You put the yellow rubber gloves in the middle of the sink and stove.", + "question_chinese": "我现在可以够得到那双黄色橡胶手套吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "You put the yellow rubber gloves in the middle of the sink and stove.", + "timestamps": { + "start": "00:01:52", + "end": "00:02:39" + } + }, + { + "description": "You are now standing in front of the stove.", + "timestamps": { + "start": "00:08:05", + "end": "00:08:39" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the sink’s edge.", + "B": "Between stove and fridge.", + "C": "You put the yellow rubber gloves next to the sink and dishwasher, rather than in the gap by the stove.", + "D": "You left the yellow rubber gloves near the stove by the kitchen table, not between the sink and stovetop.", + "E": "You put the yellow rubber gloves in the middle of the sink and stove." + }, + "answer_label": "E", + "answer_text": "You put the yellow rubber gloves in the middle of the sink and stove.", + "q_id": "P04_119_9", + "correct_answer": "You put the yellow rubber gloves in the middle of the sink and stove.", + "original_options": { + "A": "On the sink’s edge.", + "B": "Between stove and fridge.", + "C": "You put the yellow rubber gloves next to the sink and dishwasher, rather than in the gap by the stove.", + "D": "You left the yellow rubber gloves near the stove by the kitchen table, not between the sink and stovetop.", + "E": "You put the yellow rubber gloves in the middle of the sink and stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 20, + 19 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_4_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far was the cabinet with seasonings from the stovetop?", + "answer": "It was about 1 meter away. The spice cabinet was in front of you.", + "question_chinese": "我看到装调料的橱柜离灶台有多远？", + "answer_chinese": "你看到大约1米远。装调料的橱柜在你的前面。", + "evidence": [ + { + "description": "The cabinet for spices is on the left side of the sink, separated by the dishwasher in the middle.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + }, + { + "description": "On the left side of the sink is a cupboard and on the right side is the sink.", + "timestamps": { + "start": "00:09:18", + "end": "00:09:31" + } + }, + { + "description": "From the cabinet of spices to the stove.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:12" + } + } + ], + "question_timestamps": "00:12:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It was about 1 meter away. The spice cabinet was in front of you.", + "B": "About 0.5 meters away.", + "C": "About 2 meters away.", + "D": "It was about 3 meters away, and the spice cabinet was still positioned directly in front of you.", + "E": "It was about 4 meters away, with the spice cabinet straight ahead of you rather than off to the side." + }, + "answer_label": "A", + "answer_text": "It was about 1 meter away. The spice cabinet was in front of you.", + "q_id": "P04_119_4_T1", + "correct_answer": "It was about 1 meter away. The spice cabinet was in front of you.", + "original_options": { + "A": "It was about 1 meter away. The spice cabinet was in front of you.", + "B": "About 0.5 meters away.", + "C": "About 2 meters away.", + "D": "It was about 3 meters away, and the spice cabinet was still positioned directly in front of you.", + "E": "It was about 4 meters away, with the spice cabinet straight ahead of you rather than off to the side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 20 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_4_T2", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far was the cabinet with seasonings from the stovetop?", + "answer": "It was about 1 meter away. The spice cabinet was to your back-left.", + "question_chinese": "我看到装调料的橱柜离灶台有多远？", + "answer_chinese": "大约1米远左右。装调料的橱柜在你的左后方。", + "evidence": [ + { + "description": "The cabinet for spices is on the left side of the sink, separated by the dishwasher in the middle.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + }, + { + "description": "On the left side of the sink is a cupboard and on the right side is the sink.", + "timestamps": { + "start": "00:09:18", + "end": "00:09:31" + } + }, + { + "description": "From the cabinet of spices to the stove.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:12" + } + } + ], + "question_timestamps": "00:14:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "About 0.5 meters away.", + "B": "It was about 1 meter away. The spice cabinet was to your back-left.", + "C": "About 2 meters away.", + "D": "It was about 3 meters away, and the spice cabinet sat to your back-left from the stovetop.", + "E": "It was about 4 meters away, with the spice cabinet located behind you toward your back-left side." + }, + "answer_label": "B", + "answer_text": "It was about 1 meter away. The spice cabinet was to your back-left.", + "q_id": "P04_119_4_T2", + "correct_answer": "It was about 1 meter away. The spice cabinet was to your back-left.", + "original_options": { + "A": "About 0.5 meters away.", + "B": "It was about 1 meter away. The spice cabinet was to your back-left.", + "C": "About 2 meters away.", + "D": "It was about 3 meters away, and the spice cabinet sat to your back-left from the stovetop.", + "E": "It was about 4 meters away, with the spice cabinet located behind you toward your back-left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 17 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_5", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the wooden cutting board placed?", + "answer": "It's in the lower right corner of the sink.", + "question_chinese": "木质小案板放在了哪个位置？", + "answer_chinese": "在洗碗池的右下方。", + "evidence": [ + { + "description": "The wooden cutting board is placed in the lower right of the sink.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:05:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the upper left of the sink.", + "B": "On the counter beside the sink.", + "C": "It’s in the upper right corner of the sink, not down in the lower right corner.", + "D": "You placed the wooden cutting board in a drawer next to the sink instead of leaving it inside the basin.", + "E": "It's in the lower right corner of the sink." + }, + "answer_label": "E", + "answer_text": "It's in the lower right corner of the sink.", + "q_id": "P04_119_5", + "correct_answer": "It's in the lower right corner of the sink.", + "original_options": { + "A": "In the upper left of the sink.", + "B": "On the counter beside the sink.", + "C": "It’s in the upper right corner of the sink, not down in the lower right corner.", + "D": "You placed the wooden cutting board in a drawer next to the sink instead of leaving it inside the basin.", + "E": "It's in the lower right corner of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 20 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_6", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the wooden cutting board placed?", + "answer": "It's in the lower right corner of the sink.", + "question_chinese": "木质小案板放在了哪个位置？", + "answer_chinese": "在洗碗池的右下方。", + "evidence": [ + { + "description": "The wooden cutting board is placed in the lower right of the sink.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:10:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the upper left of the sink.", + "B": "On the counter by the sink.", + "C": "The wooden cutting board is in the upper right corner of the sink rather than in the lower right corner.", + "D": "It's in the lower right corner of the sink.", + "E": "You set the wooden cutting board beside the sink on the counter, not inside the sink corner." + }, + "answer_label": "D", + "answer_text": "It's in the lower right corner of the sink.", + "q_id": "P04_119_6", + "correct_answer": "It's in the lower right corner of the sink.", + "original_options": { + "A": "In the upper left of the sink.", + "B": "On the counter by the sink.", + "C": "The wooden cutting board is in the upper right corner of the sink rather than in the lower right corner.", + "D": "It's in the lower right corner of the sink.", + "E": "You set the wooden cutting board beside the sink on the counter, not inside the sink corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 17 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_7", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the spice cabinet from the stove?", + "answer": "About one meter away.", + "question_chinese": "装调料的橱柜离灶台有多远？", + "answer_chinese": "大约1米远。", + "evidence": [ + { + "description": "The cabinet for spices is on the left side of the sink, separated by the dishwasher in the middle.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + }, + { + "description": "On the left side of the sink is a cupboard and on the right side is the sink.", + "timestamps": { + "start": "00:09:18", + "end": "00:09:31" + } + }, + { + "description": "From the cabinet of spices to the stove.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:12" + } + } + ], + "question_timestamps": "00:12:59", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "About one meter away.", + "B": "About half a meter away.", + "C": "About two meters away.", + "D": "About three meters away from the stove, not roughly one meter as stated.", + "E": "About four meters away from the stove, much farther than a one-meter reach." + }, + "answer_label": "A", + "answer_text": "About one meter away.", + "q_id": "P04_119_7", + "correct_answer": "About one meter away.", + "original_options": { + "A": "About one meter away.", + "B": "About half a meter away.", + "C": "About two meters away.", + "D": "About three meters away from the stove, not roughly one meter as stated.", + "E": "About four meters away from the stove, much farther than a one-meter reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 13 + ] + }, + { + "video_uid": "P04_119", + "video_length": "00:14:11", + "qid": "P04_119_8", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the spice cabinet from the stove?", + "answer": "About one meter away.", + "question_chinese": "装调料的橱柜离灶台有多远？", + "answer_chinese": "大约1米远。", + "evidence": [ + { + "description": "The cabinet for spices is on the left side of the sink, separated by the dishwasher in the middle.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + }, + { + "description": "On the left side of the sink is a cupboard and on the right side is the sink.", + "timestamps": { + "start": "00:09:18", + "end": "00:09:31" + } + }, + { + "description": "From the cabinet of spices to the stove.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:12" + } + } + ], + "question_timestamps": "00:14:09", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "About one meter away.", + "B": "Right next to the stove.", + "C": "About three meters away.", + "D": "Roughly half a meter away, close enough to reach without taking a step.", + "E": "Around two meters away, requiring you to step sideways to get to the cabinet." + }, + "answer_label": "A", + "answer_text": "About one meter away.", + "q_id": "P04_119_8", + "correct_answer": "About one meter away.", + "original_options": { + "A": "About one meter away.", + "B": "Right next to the stove.", + "C": "About three meters away.", + "D": "Roughly half a meter away, close enough to reach without taking a step.", + "E": "Around two meters away, requiring you to step sideways to get to the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 14 + ] + } +] diff --git a/QAs/P04_123-mcq.json b/QAs/P04_123-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5089bdde768376d04abd2c66c14fedfd31d8632e --- /dev/null +++ b/QAs/P04_123-mcq.json @@ -0,0 +1,1451 @@ +[ + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_5_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can relative to me?", + "answer": "The trash can is directly in front of you.", + "question_chinese": "垃圾桶在我的的哪个位置？", + "answer_chinese": "垃圾桶在你的正前方。", + "evidence": [ + { + "description": "The trash can is directly in front of you.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can is behind you.", + "B": "The trash can is to your left.", + "C": "The trash can is slightly to your right, not directly in your line of sight.", + "D": "The trash can is off to one side of you rather than being straight ahead.", + "E": "The trash can is directly in front of you." + }, + "answer_label": "E", + "answer_text": "The trash can is directly in front of you.", + "q_id": "P04_123_5_T1", + "correct_answer": "The trash can is directly in front of you.", + "original_options": { + "A": "The trash can is behind you.", + "B": "The trash can is to your left.", + "C": "The trash can is slightly to your right, not directly in your line of sight.", + "D": "The trash can is off to one side of you rather than being straight ahead.", + "E": "The trash can is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 15 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_5_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can relative to me?", + "answer": "The trash can is to your left.", + "question_chinese": "垃圾桶在我的的哪个位置？", + "answer_chinese": "垃圾桶在你的左边。", + "evidence": [ + { + "description": "On your left.", + "timestamps": { + "start": "00:01:12", + "end": "00:01:13" + } + } + ], + "question_timestamps": "00:02:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can is in front of you.", + "B": "The trash can is to your left.", + "C": "The trash can is to your right.", + "D": "The trash can is behind you, positioned away from where you are facing.", + "E": "The trash can is above you, not at floor level beside you." + }, + "answer_label": "B", + "answer_text": "The trash can is to your left.", + "q_id": "P04_123_5_T2", + "correct_answer": "The trash can is to your left.", + "original_options": { + "A": "The trash can is in front of you.", + "B": "The trash can is to your left.", + "C": "The trash can is to your right.", + "D": "The trash can is behind you, positioned away from where you are facing.", + "E": "The trash can is above you, not at floor level beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 13, + 12 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_9", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can in relation to the sink?", + "answer": "The trash can is to the left front of the sink.", + "question_chinese": "垃圾桶在洗碗池的哪个位置？", + "answer_chinese": "垃圾桶在洗碗池的左前方。", + "evidence": [ + { + "description": "The trash can is to the left front of the sink.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To the right of the sink.", + "B": "The trash can is to the left front of the sink.", + "C": "Behind the sink.", + "D": "In front of the sink but shifted to the right side instead of the left.", + "E": "To the left of the sink, but further back near the wall rather than in front." + }, + "answer_label": "B", + "answer_text": "The trash can is to the left front of the sink.", + "q_id": "P04_123_9", + "correct_answer": "The trash can is to the left front of the sink.", + "original_options": { + "A": "To the right of the sink.", + "B": "The trash can is to the left front of the sink.", + "C": "Behind the sink.", + "D": "In front of the sink but shifted to the right side instead of the left.", + "E": "To the left of the sink, but further back near the wall rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 3, + 15, + 16 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_10", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can in relation to the sink?", + "answer": "The trash can is to the left front of the sink.", + "question_chinese": "垃圾桶在洗碗池的哪个位置？", + "answer_chinese": "垃圾桶在洗碗池的左前方。", + "evidence": [ + { + "description": "The trash can is to the left front of the sink.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:03:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind the sink.", + "B": "To the right of the sink.", + "C": "In front of the sink but positioned more toward the right than the left.", + "D": "Left of the sink, yet it sits further back near the wall instead of front-left.", + "E": "The trash can is to the left front of the sink." + }, + "answer_label": "E", + "answer_text": "The trash can is to the left front of the sink.", + "q_id": "P04_123_10", + "correct_answer": "The trash can is to the left front of the sink.", + "original_options": { + "A": "Behind the sink.", + "B": "To the right of the sink.", + "C": "In front of the sink but positioned more toward the right than the left.", + "D": "Left of the sink, yet it sits further back near the wall instead of front-left.", + "E": "The trash can is to the left front of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 6, + 14, + 15 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_15", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the fridge in total?", + "answer": "Once.", + "question_chinese": "我一共打开了几次冰箱？", + "answer_chinese": "一次。", + "evidence": [ + { + "description": "Opened the fridge once.", + "timestamps": { + "start": "00:19:55", + "end": "00:20:00" + } + } + ], + "question_timestamps": "00:20:49", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Three times.", + "C": "Once.", + "D": "You opened the fridge two times in total during the whole sequence.", + "E": "You opened the fridge three times overall, including returning to it after stepping away." + }, + "answer_label": "C", + "answer_text": "Once.", + "q_id": "P04_123_15", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Three times.", + "C": "Once.", + "D": "You opened the fridge two times in total during the whole sequence.", + "E": "You opened the fridge three times overall, including returning to it after stepping away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 14 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_16", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the fridge in total?", + "answer": "Once.", + "question_chinese": "我一共打开了几次冰箱？", + "answer_chinese": "一次。", + "evidence": [ + { + "description": "Opened the fridge once.", + "timestamps": { + "start": "00:19:55", + "end": "00:20:00" + } + } + ], + "question_timestamps": "00:21:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Twice.", + "C": "Three times.", + "D": "In total, you opened the fridge four times across the entire clip.", + "E": "Overall, you opened the fridge five times, going back to it repeatedly during the task." + }, + "answer_label": "A", + "answer_text": "Once.", + "q_id": "P04_123_16", + "correct_answer": "Once.", + "original_options": { + "A": "Once.", + "B": "Twice.", + "C": "Three times.", + "D": "In total, you opened the fridge four times across the entire clip.", + "E": "Overall, you opened the fridge five times, going back to it repeatedly during the task." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 15 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many slices of bread did I just cut in total?", + "answer": "Four slices.", + "question_chinese": "我刚才一共切了几片面包？", + "answer_chinese": "四片。", + "evidence": [ + { + "description": "Cut the first slice of bread and place it on the plate.", + "timestamps": { + "start": "00:01:15", + "end": "00:02:04" + } + }, + { + "description": "Cut the second slice of bread and place it on the plate.", + "timestamps": { + "start": "00:02:05", + "end": "00:02:41" + } + }, + { + "description": "Cut the third slice of bread and place it on the plate.", + "timestamps": { + "start": "00:02:45", + "end": "00:03:30" + } + }, + { + "description": "Cut the fourth slice of bread and place it on the plate.", + "timestamps": { + "start": "00:03:57", + "end": "00:04:15" + } + } + ], + "question_timestamps": "00:06:48", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two slices.", + "B": "Four slices.", + "C": "Five slices.", + "D": "You cut three slices of bread altogether, not the larger batch you might expect.", + "E": "You ended up cutting six slices in total, continuing past the initial few cuts." + }, + "answer_label": "B", + "answer_text": "Four slices.", + "q_id": "P04_123_1", + "correct_answer": "Four slices.", + "original_options": { + "A": "Two slices.", + "B": "Four slices.", + "C": "Five slices.", + "D": "You cut three slices of bread altogether, not the larger batch you might expect.", + "E": "You ended up cutting six slices in total, continuing past the initial few cuts." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many slices of bread did I just cut in total?", + "answer": "Four slices.", + "question_chinese": "我刚才一共切了几片面包？", + "answer_chinese": "四片。", + "evidence": [ + { + "description": "Cut the first slice of bread and place it on the plate.", + "timestamps": { + "start": "00:01:15", + "end": "00:02:04" + } + }, + { + "description": "Cut the second slice of bread and place it on the plate.", + "timestamps": { + "start": "00:02:05", + "end": "00:02:41" + } + }, + { + "description": "Cut the third slice of bread and place it on the plate.", + "timestamps": { + "start": "00:02:45", + "end": "00:03:30" + } + }, + { + "description": "Cut the fourth slice of bread and place it on the plate.", + "timestamps": { + "start": "00:03:57", + "end": "00:04:15" + } + } + ], + "question_timestamps": "00:09:51", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three slices.", + "B": "Five slices.", + "C": "You cut six slices of bread in total after finishing the last cut.", + "D": "Four slices.", + "E": "You cut three slices of bread in total, and the loaf remains on the cutting board." + }, + "answer_label": "D", + "answer_text": "Four slices.", + "q_id": "P04_123_2", + "correct_answer": "Four slices.", + "original_options": { + "A": "Three slices.", + "B": "Five slices.", + "C": "You cut six slices of bread in total after finishing the last cut.", + "D": "Four slices.", + "E": "You cut three slices of bread in total, and the loaf remains on the cutting board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 16 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_3", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take to get from the windowsill to the stove initially?", + "answer": "Turn back from the windowsill, walk straight past the TV and coffee table, the shelf on the left and the cabinet on the right, walk to the cabinet on the right of the stove, then turn left to the front of the stove.", + "question_chinese": "我最初从窗台走到灶台经过了什么路径？", + "answer_chinese": "从窗台向后转，直走依次经过电视和茶几、左边的杂物架和右边的杂物柜，一直走到灶台右边的柜子，然后左转走到灶台前。", + "evidence": [ + { + "description": "Turn back from the windowsill and pass the TV and coffee table.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:38" + } + }, + { + "description": "Go straight, passing the shelf on the left and the cabinet on the right in turn.", + "timestamps": { + "start": "00:00:38", + "end": "00:00:41" + } + }, + { + "description": "Walk to the cabinet to the right of the stove, then turn left to the front of the stove.", + "timestamps": { + "start": "00:00:41", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn back from the windowsill, walk straight past the TV and coffee table, the shelf on the left and the cabinet on the right, walk to the cabinet on the right of the stove, then turn left to the front of the stove.", + "B": "Walked straight to the stove.", + "C": "Turned right to the stove.", + "D": "Turn back, walk past the TV and coffee table, then turn right at the cabinet beside the stove.", + "E": "Turn back from the windowsill, pass the TV and coffee table, reach the left-side cabinet, then turn right to the stove." + }, + "answer_label": "A", + "answer_text": "Turn back from the windowsill, walk straight past the TV and coffee table, the shelf on the left and the cabinet on the right, walk to the cabinet on the right of the stove, then turn left to the front of the stove.", + "q_id": "P04_123_3", + "correct_answer": "Turn back from the windowsill, walk straight past the TV and coffee table, the shelf on the left and the cabinet on the right, walk to the cabinet on the right of the stove, then turn left to the front of the stove.", + "original_options": { + "A": "Turn back from the windowsill, walk straight past the TV and coffee table, the shelf on the left and the cabinet on the right, walk to the cabinet on the right of the stove, then turn left to the front of the stove.", + "B": "Walked straight to the stove.", + "C": "Turned right to the stove.", + "D": "Turn back, walk past the TV and coffee table, then turn right at the cabinet beside the stove.", + "E": "Turn back from the windowsill, pass the TV and coffee table, reach the left-side cabinet, then turn right to the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 21 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_4", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take to get from the windowsill to the stove initially?", + "answer": "Turn back from the windowsill, walk straight past the TV and coffee table, the shelf on the left and the cabinet on the right, walk to the cabinet on the right of the stove, then turn left to the front of the stove.", + "question_chinese": "我最初从窗台走到灶台经过了什么路径？", + "answer_chinese": "从窗台向后转，直走依次经过电视和茶几、左边的杂物架和右边的杂物柜，一直走到灶台右边的柜子，然后左转走到灶台前。", + "evidence": [ + { + "description": "Turn back from the windowsill and pass the TV and coffee table.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:38" + } + }, + { + "description": "Go straight, passing the shelf on the left and the cabinet on the right in turn.", + "timestamps": { + "start": "00:00:38", + "end": "00:00:41" + } + }, + { + "description": "Walk to the cabinet to the right of the stove, then turn left to the front of the stove.", + "timestamps": { + "start": "00:00:41", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:04:47", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Approached the stove from the left.", + "B": "Turned left at the stove.", + "C": "Turn back and walk past the TV and coffee table, then go to the left of the stove and stop there.", + "D": "Turn away from the windowsill, pass the TV and coffee table, head to the cabinet left of the stove, then face forward.", + "E": "Turn back from the windowsill, walk straight past the TV and coffee table, the shelf on the left and the cabinet on the right, walk to the cabinet on the right of the stove, then turn left to the front of the stove." + }, + "answer_label": "E", + "answer_text": "Turn back from the windowsill, walk straight past the TV and coffee table, the shelf on the left and the cabinet on the right, walk to the cabinet on the right of the stove, then turn left to the front of the stove.", + "q_id": "P04_123_4", + "correct_answer": "Turn back from the windowsill, walk straight past the TV and coffee table, the shelf on the left and the cabinet on the right, walk to the cabinet on the right of the stove, then turn left to the front of the stove.", + "original_options": { + "A": "Approached the stove from the left.", + "B": "Turned left at the stove.", + "C": "Turn back and walk past the TV and coffee table, then go to the left of the stove and stop there.", + "D": "Turn away from the windowsill, pass the TV and coffee table, head to the cabinet left of the stove, then face forward.", + "E": "Turn back from the windowsill, walk straight past the TV and coffee table, the shelf on the left and the cabinet on the right, walk to the cabinet on the right of the stove, then turn left to the front of the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 22 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_1_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many slices of bread have I cut right now?", + "answer": "You have now cut one slice of bread, and the bread is in front of you.", + "question_chinese": "我现在切了几片面包？", + "answer_chinese": "你现在切了一片面包，面包在你的正前方。", + "evidence": [ + { + "description": "You have cut a piece of bread now.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:52" + } + } + ], + "question_timestamps": "00:01:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero slices.", + "B": "Two slices.", + "C": "You have now cut one slice of bread, and the bread is in front of you.", + "D": "You have now cut two slices of bread, and the bread is in front of you.", + "E": "You have now cut three slices of bread, and the bread is in front of you." + }, + "answer_label": "C", + "answer_text": "You have now cut one slice of bread, and the bread is in front of you.", + "q_id": "P04_123_1_T1", + "correct_answer": "You have now cut one slice of bread, and the bread is in front of you.", + "original_options": { + "A": "Zero slices.", + "B": "Two slices.", + "C": "You have now cut one slice of bread, and the bread is in front of you.", + "D": "You have now cut two slices of bread, and the bread is in front of you.", + "E": "You have now cut three slices of bread, and the bread is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_1_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many slices of bread have I cut right now?", + "answer": "You have now cut four slices of bread in total, and the bread is to your right.", + "question_chinese": "我现在切了几片面包？", + "answer_chinese": "你现在一共切了四片面包，面包在你的右边。", + "evidence": [ + { + "description": "Cut the first slice of bread and place it on the plate.", + "timestamps": { + "start": "00:01:15", + "end": "00:02:04" + } + }, + { + "description": "Cut the second slice of bread and place it on the plate.", + "timestamps": { + "start": "00:02:05", + "end": "00:02:41" + } + }, + { + "description": "Cut the third slice of bread and place it on the plate.", + "timestamps": { + "start": "00:02:45", + "end": "00:03:30" + } + }, + { + "description": "Cut the fourth slice of bread and place it on the plate.", + "timestamps": { + "start": "00:03:57", + "end": "00:04:15" + } + } + ], + "question_timestamps": "00:04:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three slices total.", + "B": "Five slices total.", + "C": "You have now cut three slices of bread in total, and the bread is to your right.", + "D": "You have now cut four slices of bread in total, and the bread is to your right.", + "E": "You have now cut six slices of bread in total, and the bread is to your right." + }, + "answer_label": "D", + "answer_text": "You have now cut four slices of bread in total, and the bread is to your right.", + "q_id": "P04_123_1_T2", + "correct_answer": "You have now cut four slices of bread in total, and the bread is to your right.", + "original_options": { + "A": "Three slices total.", + "B": "Five slices total.", + "C": "You have now cut three slices of bread in total, and the bread is to your right.", + "D": "You have now cut four slices of bread in total, and the bread is to your right.", + "E": "You have now cut six slices of bread in total, and the bread is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_13", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get to the cabinet with the bowls now?", + "answer": "Yes.", + "question_chinese": "我现在能够得到装碗的柜子吗？", + "answer_chinese": "能。", + "evidence": [ + { + "description": "The cabinet with the bowls is about half a meter away to your upper left.", + "timestamps": { + "start": "00:13:57", + "end": "00:14:02" + } + } + ], + "question_timestamps": "00:14:03", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "P04_123_13", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_6_T1", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many slices of bread are in my plate right now?", + "answer": "There is one slice of bread on your plate right now, and the bread is to your right.", + "question_chinese": "我的盘子里现在有几片面包？", + "answer_chinese": "你的盘子里现在有一片面包，面包在你的右边。", + "evidence": [ + { + "description": "There is a piece of bread on the plate now.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:03" + } + } + ], + "question_timestamps": "00:02:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There are zero slices on your plate.", + "B": "There are two slices on your plate.", + "C": "There is one slice of bread on your plate right now, and the bread is to your right.", + "D": "There are three slices of bread on your plate right now, and the bread is to your right.", + "E": "There are four slices of bread on your plate right now, and the bread is to your right." + }, + "answer_label": "C", + "answer_text": "There is one slice of bread on your plate right now, and the bread is to your right.", + "q_id": "P04_123_6_T1", + "correct_answer": "There is one slice of bread on your plate right now, and the bread is to your right.", + "original_options": { + "A": "There are zero slices on your plate.", + "B": "There are two slices on your plate.", + "C": "There is one slice of bread on your plate right now, and the bread is to your right.", + "D": "There are three slices of bread on your plate right now, and the bread is to your right.", + "E": "There are four slices of bread on your plate right now, and the bread is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_6_T2", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many slices of bread are in my plate right now?", + "answer": "There are two slices of bread on your plate right now, and the bread is to your right.", + "question_chinese": "我的盘子里现在有几片面包？", + "answer_chinese": "你的盘子里现在有两片面包，面包在你的右边。", + "evidence": [ + { + "description": "There are two pieces of bread on your plate now.", + "timestamps": { + "start": "00:02:36", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:02:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There is one slice on your plate.", + "B": "There are three slices on your plate.", + "C": "There are zero slices of bread on your plate right now, and the bread is to your right.", + "D": "There are two slices of bread on your plate right now, and the bread is to your right.", + "E": "There are four slices of bread on your plate right now, and the bread is to your right." + }, + "answer_label": "D", + "answer_text": "There are two slices of bread on your plate right now, and the bread is to your right.", + "q_id": "P04_123_6_T2", + "correct_answer": "There are two slices of bread on your plate right now, and the bread is to your right.", + "original_options": { + "A": "There is one slice on your plate.", + "B": "There are three slices on your plate.", + "C": "There are zero slices of bread on your plate right now, and the bread is to your right.", + "D": "There are two slices of bread on your plate right now, and the bread is to your right.", + "E": "There are four slices of bread on your plate right now, and the bread is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_4_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the dishwasher relative to me?", + "answer": "The dishwasher is in front of you.", + "question_chinese": "洗碗机在相对于我的什么位置？", + "answer_chinese": "洗碗机在相对于你的正前方。", + "evidence": [ + { + "description": "The dishwasher is directly in front of you.", + "timestamps": { + "start": "00:13:50", + "end": "00:13:54" + } + } + ], + "question_timestamps": "00:13:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The dishwasher is behind you.", + "B": "The dishwasher is to your left.", + "C": "The dishwasher is in front of you.", + "D": "The dishwasher is to your right, not directly in front of you where you are facing.", + "E": "The dishwasher is above you, positioned higher than your eye level rather than straight ahead." + }, + "answer_label": "C", + "answer_text": "The dishwasher is in front of you.", + "q_id": "P04_123_4_T1", + "correct_answer": "The dishwasher is in front of you.", + "original_options": { + "A": "The dishwasher is behind you.", + "B": "The dishwasher is to your left.", + "C": "The dishwasher is in front of you.", + "D": "The dishwasher is to your right, not directly in front of you where you are facing.", + "E": "The dishwasher is above you, positioned higher than your eye level rather than straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_4_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the dishwasher relative to me?", + "answer": "The dishwasher is behind you.", + "question_chinese": "洗碗机在相对于我的什么位置？", + "answer_chinese": "洗碗机在相对于你的后面。", + "evidence": [ + { + "description": "The dishwasher is directly in front of you.", + "timestamps": { + "start": "00:13:58", + "end": "00:14:05" + } + } + ], + "question_timestamps": "00:14:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The dishwasher is in front of you.", + "B": "The dishwasher is to your right.", + "C": "The dishwasher is to your left, not positioned behind you as you stand there.", + "D": "The dishwasher is above you, located higher up instead of directly behind your position.", + "E": "The dishwasher is behind you." + }, + "answer_label": "E", + "answer_text": "The dishwasher is behind you.", + "q_id": "P04_123_4_T2", + "correct_answer": "The dishwasher is behind you.", + "original_options": { + "A": "The dishwasher is in front of you.", + "B": "The dishwasher is to your right.", + "C": "The dishwasher is to your left, not positioned behind you as you stand there.", + "D": "The dishwasher is above you, located higher up instead of directly behind your position.", + "E": "The dishwasher is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 14 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_7", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the dishwasher now in relation to me?", + "answer": "The dishwasher is to your right front.", + "question_chinese": "洗碗机现在在我的哪个方位？", + "answer_chinese": "洗碗机在你的右前方。", + "evidence": [ + { + "description": "The dishwasher is in front of the stove and to the right.", + "timestamps": { + "start": "00:13:51", + "end": "00:14:05" + } + }, + { + "description": "You are now standing in front of the stove, facing the stove.", + "timestamps": { + "start": "00:14:04", + "end": "00:14:26" + } + } + ], + "question_timestamps": "00:14:29", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The dishwasher is to your right front.", + "B": "The dishwasher is to your left front.", + "C": "The dishwasher is to your right rear.", + "D": "The dishwasher is directly in front of you, not angled to the right front from your viewpoint.", + "E": "The dishwasher is behind you on the left side, rather than located in your right-front area." + }, + "answer_label": "A", + "answer_text": "The dishwasher is to your right front.", + "q_id": "P04_123_7", + "correct_answer": "The dishwasher is to your right front.", + "original_options": { + "A": "The dishwasher is to your right front.", + "B": "The dishwasher is to your left front.", + "C": "The dishwasher is to your right rear.", + "D": "The dishwasher is directly in front of you, not angled to the right front from your viewpoint.", + "E": "The dishwasher is behind you on the left side, rather than located in your right-front area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_8", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the dishwasher now in relation to me?", + "answer": "The dishwasher is to your right front.", + "question_chinese": "洗碗机现在在我的哪个方位？", + "answer_chinese": "洗碗机在你的右前方。", + "evidence": [ + { + "description": "The dishwasher is in front of the stove and to the right.", + "timestamps": { + "start": "00:13:51", + "end": "00:14:05" + } + }, + { + "description": "You are now standing in front of the stove, facing the stove.", + "timestamps": { + "start": "00:14:04", + "end": "00:14:26" + } + } + ], + "question_timestamps": "00:15:31", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The dishwasher is to your left front.", + "B": "The dishwasher is to your right rear.", + "C": "The dishwasher is to your right front.", + "D": "The dishwasher is straight in front of you, instead of being positioned to your right front.", + "E": "The dishwasher is behind you on your left side, not in the right-front direction from you." + }, + "answer_label": "C", + "answer_text": "The dishwasher is to your right front.", + "q_id": "P04_123_8", + "correct_answer": "The dishwasher is to your right front.", + "original_options": { + "A": "The dishwasher is to your left front.", + "B": "The dishwasher is to your right rear.", + "C": "The dishwasher is to your right front.", + "D": "The dishwasher is straight in front of you, instead of being positioned to your right front.", + "E": "The dishwasher is behind you on your left side, not in the right-front direction from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_11", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did this white plate on the stove come from?", + "answer": "From the cupboard to the upper left of the stove.", + "question_chinese": "灶台上这个白色的盘子是从哪里来的？", + "answer_chinese": "从灶台左上方的橱柜里来的。", + "evidence": [ + { + "description": "Take the plate from the cabinet to the upper left of the stove and place the sliced bread on the plate.", + "timestamps": { + "start": "00:01:53", + "end": "00:04:18" + } + }, + { + "description": "Place the slices of bread from the plate on the oven rack.", + "timestamps": { + "start": "00:08:32", + "end": "00:08:50" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From the upper right cupboard.", + "B": "From the lower left cupboard.", + "C": "From the cupboard to the upper left of the stove.", + "D": "From the drawer to the upper left of the stove, not from the cupboard above it.", + "E": "From the cupboard to the upper right of the stove, instead of the cupboard on the upper left." + }, + "answer_label": "C", + "answer_text": "From the cupboard to the upper left of the stove.", + "q_id": "P04_123_11", + "correct_answer": "From the cupboard to the upper left of the stove.", + "original_options": { + "A": "From the upper right cupboard.", + "B": "From the lower left cupboard.", + "C": "From the cupboard to the upper left of the stove.", + "D": "From the drawer to the upper left of the stove, not from the cupboard above it.", + "E": "From the cupboard to the upper right of the stove, instead of the cupboard on the upper left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_12", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did this white plate on the stove come from?", + "answer": "From the cupboard to the upper left of the stove.", + "question_chinese": "灶台上这个白色的盘子是从哪里来的？", + "answer_chinese": "从灶台左上方的橱柜里来的。", + "evidence": [ + { + "description": "Take the plate from the cabinet to the upper left of the stove and place the sliced bread on the plate.", + "timestamps": { + "start": "00:01:53", + "end": "00:04:18" + } + }, + { + "description": "Place the slices of bread from the plate on the oven rack.", + "timestamps": { + "start": "00:08:32", + "end": "00:08:50" + } + } + ], + "question_timestamps": "00:10:11", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From the upper right cupboard.", + "B": "From the cupboard to the upper left of the stove.", + "C": "From the upper left drawer.", + "D": "From the shelf to the lower left of the stove, not from the cupboard above the stove.", + "E": "From the cabinet to the lower right of the stove, instead of the upper-left cupboard." + }, + "answer_label": "B", + "answer_text": "From the cupboard to the upper left of the stove.", + "q_id": "P04_123_12", + "correct_answer": "From the cupboard to the upper left of the stove.", + "original_options": { + "A": "From the upper right cupboard.", + "B": "From the cupboard to the upper left of the stove.", + "C": "From the upper left drawer.", + "D": "From the shelf to the lower left of the stove, not from the cupboard above the stove.", + "E": "From the cabinet to the lower right of the stove, instead of the upper-left cupboard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 15 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_5", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the oven in relation to me?", + "answer": "The oven is below you.", + "question_chinese": "烤箱在我的哪个方位？", + "answer_chinese": "烤箱在你的下面。", + "evidence": [ + { + "description": "The oven is below the stove.", + "timestamps": { + "start": "00:09:50", + "end": "00:10:02" + } + }, + { + "description": "You are now standing in front of the stove.", + "timestamps": { + "start": "00:10:02", + "end": "00:10:08" + } + } + ], + "question_timestamps": "00:10:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The oven is above you.", + "B": "The oven is below you.", + "C": "The oven is in front of you.", + "D": "The oven is beside you at your level, not underneath where you are standing.", + "E": "The oven is behind you, so you would need to turn around to face it." + }, + "answer_label": "B", + "answer_text": "The oven is below you.", + "q_id": "P04_123_5", + "correct_answer": "The oven is below you.", + "original_options": { + "A": "The oven is above you.", + "B": "The oven is below you.", + "C": "The oven is in front of you.", + "D": "The oven is beside you at your level, not underneath where you are standing.", + "E": "The oven is behind you, so you would need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 14, + 15 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_6", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the oven in relation to me?", + "answer": "The oven is below you.", + "question_chinese": "烤箱在我的哪个方位？", + "answer_chinese": "烤箱在你的下面。", + "evidence": [ + { + "description": "The oven is below the stove.", + "timestamps": { + "start": "00:09:50", + "end": "00:10:02" + } + }, + { + "description": "You are now standing in front of the stove.", + "timestamps": { + "start": "00:14:00", + "end": "00:14:08" + } + } + ], + "question_timestamps": "00:14:09", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The oven is above you.", + "B": "The oven is behind you.", + "C": "The oven is in front of you, directly within your forward line of sight.", + "D": "The oven is below you.", + "E": "The oven is beside you, positioned to one side rather than underneath you." + }, + "answer_label": "D", + "answer_text": "The oven is below you.", + "q_id": "P04_123_6", + "correct_answer": "The oven is below you.", + "original_options": { + "A": "The oven is above you.", + "B": "The oven is behind you.", + "C": "The oven is in front of you, directly within your forward line of sight.", + "D": "The oven is below you.", + "E": "The oven is beside you, positioned to one side rather than underneath you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_2_T1", + "qid_num": 30, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is that pot of pink flowers on the windowsill relative to me?", + "answer": "The pink flower pot on the windowsill is to your left.", + "question_chinese": "窗台上那盆粉色的花在相对于我的什么位置？", + "answer_chinese": "窗台上那盆粉色的花在相对于你的左边。", + "evidence": [ + { + "description": "The pink flower in the pot on the windowsill is on your left", + "timestamps": { + "start": "00:00:04", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to your right.", + "B": "It is in front of you.", + "C": "The pink flower pot on the windowsill is behind you, so it is not on your left side.", + "D": "The pink flower pot on the windowsill is above you, higher than your eye level near the window.", + "E": "The pink flower pot on the windowsill is to your left." + }, + "answer_label": "E", + "answer_text": "The pink flower pot on the windowsill is to your left.", + "q_id": "P04_123_2_T1", + "correct_answer": "The pink flower pot on the windowsill is to your left.", + "original_options": { + "A": "It is to your right.", + "B": "It is in front of you.", + "C": "The pink flower pot on the windowsill is behind you, so it is not on your left side.", + "D": "The pink flower pot on the windowsill is above you, higher than your eye level near the window.", + "E": "The pink flower pot on the windowsill is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 18 + ] + }, + { + "video_uid": "P04_123", + "video_length": "00:21:41", + "qid": "P04_123_2_T2", + "qid_num": 31, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is that pot of pink flowers on the windowsill relative to me?", + "answer": "The pink flower pot on the windowsill is behind you.", + "question_chinese": "窗台上那盆粉色的花在相对于我的什么位置？", + "answer_chinese": "窗台上那盆粉色的花在相对于你的后面。", + "evidence": [ + { + "description": "The pink flower pot on the windowsill is behind you.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "The pink flower pot on the windowsill is to your right, positioned on the opposite side from behind you.", + "D": "The pink flower pot on the windowsill is behind you.", + "E": "The pink flower pot on the windowsill is above you, so it is not located behind your body." + }, + "answer_label": "D", + "answer_text": "The pink flower pot on the windowsill is behind you.", + "q_id": "P04_123_2_T2", + "correct_answer": "The pink flower pot on the windowsill is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "The pink flower pot on the windowsill is to your right, positioned on the opposite side from behind you.", + "D": "The pink flower pot on the windowsill is behind you.", + "E": "The pink flower pot on the windowsill is above you, so it is not located behind your body." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 18 + ] + } +] diff --git a/QAs/P06_101-mcq.json b/QAs/P06_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..bd642acb514b59f34a69801c8a47ca76ed9c7fef --- /dev/null +++ b/QAs/P06_101-mcq.json @@ -0,0 +1,1791 @@ +[ + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_8_T2", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the refrigerator relative to me?", + "answer": "The refrigerator is to your left.", + "question_chinese": "冰箱在我的哪里？", + "answer_chinese": "冰箱在你的左边。", + "evidence": [ + { + "description": "The fridge is on your left.", + "timestamps": { + "start": "00:30:19", + "end": "00:30:21" + } + } + ], + "question_timestamps": "00:30:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is in front of you.", + "B": "The refrigerator is to your right.", + "C": "The refrigerator is behind you, just past your shoulder as you turn around to look.", + "D": "The refrigerator is to your left.", + "E": "The refrigerator is above you, positioned higher than your eye level rather than on your left." + }, + "answer_label": "D", + "answer_text": "The refrigerator is to your left.", + "q_id": "P06_101_8_T2", + "correct_answer": "The refrigerator is to your left.", + "original_options": { + "A": "The refrigerator is in front of you.", + "B": "The refrigerator is to your right.", + "C": "The refrigerator is behind you, just past your shoulder as you turn around to look.", + "D": "The refrigerator is to your left.", + "E": "The refrigerator is above you, positioned higher than your eye level rather than on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 16 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_8_T1", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the refrigerator relative to me?", + "answer": "The refrigerator is in front of you.", + "question_chinese": "冰箱在我的哪里？", + "answer_chinese": "冰箱在你的前面。", + "evidence": [ + { + "description": "The fridge is in front of you.", + "timestamps": { + "start": "00:30:22", + "end": "00:30:24" + } + } + ], + "question_timestamps": "00:30:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is behind you.", + "B": "The refrigerator is in front of you.", + "C": "The refrigerator is to your left.", + "D": "The refrigerator is to your right, not directly ahead in your line of sight.", + "E": "The refrigerator is above you, so you would need to look upward instead of forward." + }, + "answer_label": "B", + "answer_text": "The refrigerator is in front of you.", + "q_id": "P06_101_8_T1", + "correct_answer": "The refrigerator is in front of you.", + "original_options": { + "A": "The refrigerator is behind you.", + "B": "The refrigerator is in front of you.", + "C": "The refrigerator is to your left.", + "D": "The refrigerator is to your right, not directly ahead in your line of sight.", + "E": "The refrigerator is above you, so you would need to look upward instead of forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 15 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_15", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the refrigerator from me?", + "answer": "The refrigerator is about two meters to your left.", + "question_chinese": "冰箱离我有多远？", + "answer_chinese": "冰箱在你的左边约2米远处。", + "evidence": [ + { + "description": "The refrigerator is about 2 meters to the left of the sink.", + "timestamps": { + "start": "00:30:16", + "end": "00:30:28" + } + } + ], + "question_timestamps": "00:30:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "About one meter to your left.", + "B": "About two meters to your right.", + "C": "The refrigerator is about two meters in front of you, not positioned off to your left side.", + "D": "The refrigerator is about two meters to your left.", + "E": "The refrigerator is roughly three meters to your left, farther away than two meters from you." + }, + "answer_label": "D", + "answer_text": "The refrigerator is about two meters to your left.", + "q_id": "P06_101_15", + "correct_answer": "The refrigerator is about two meters to your left.", + "original_options": { + "A": "About one meter to your left.", + "B": "About two meters to your right.", + "C": "The refrigerator is about two meters in front of you, not positioned off to your left side.", + "D": "The refrigerator is about two meters to your left.", + "E": "The refrigerator is roughly three meters to your left, farther away than two meters from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_16", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the refrigerator from me?", + "answer": "The refrigerator is about two meters to your left.", + "question_chinese": "冰箱离我有多远？", + "answer_chinese": "冰箱在你的左边约2米远处。", + "evidence": [ + { + "description": "The refrigerator is about 2 meters to the left of the sink.", + "timestamps": { + "start": "00:30:16", + "end": "00:30:28" + } + } + ], + "question_timestamps": "00:30:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is about two meters to your left.", + "B": "About one meter to your left.", + "C": "About three meters to your left.", + "D": "The refrigerator is about two meters to your right, placing it on the opposite side from you.", + "E": "The refrigerator is around five meters behind you, much farther than two meters and not on your left." + }, + "answer_label": "A", + "answer_text": "The refrigerator is about two meters to your left.", + "q_id": "P06_101_16", + "correct_answer": "The refrigerator is about two meters to your left.", + "original_options": { + "A": "The refrigerator is about two meters to your left.", + "B": "About one meter to your left.", + "C": "About three meters to your left.", + "D": "The refrigerator is about two meters to your right, placing it on the opposite side from you.", + "E": "The refrigerator is around five meters behind you, much farther than two meters and not on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_2_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans did I open in total just now?", + "answer": "You opened a total of two, and they are all to your left.", + "question_chinese": "刚才我一共开了几个罐头？", + "answer_chinese": "你一共开了两个，它们都在你的左边。", + "evidence": [ + { + "description": "Opened the 1st can.", + "timestamps": { + "start": "00:13:28", + "end": "00:13:42" + } + }, + { + "description": "Opened the 2nd can.", + "timestamps": { + "start": "00:13:59", + "end": "00:14:11" + } + } + ], + "question_timestamps": "00:14:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened one can total.", + "B": "You opened a total of two, and they are all to your left.", + "C": "You opened three cans total.", + "D": "You opened a total of four, and they are all to your left, not just two cans.", + "E": "You opened a total of zero, and there aren’t any opened cans sitting to your left." + }, + "answer_label": "B", + "answer_text": "You opened a total of two, and they are all to your left.", + "q_id": "P06_101_2_T1", + "correct_answer": "You opened a total of two, and they are all to your left.", + "original_options": { + "A": "You opened one can total.", + "B": "You opened a total of two, and they are all to your left.", + "C": "You opened three cans total.", + "D": "You opened a total of four, and they are all to your left, not just two cans.", + "E": "You opened a total of zero, and there aren’t any opened cans sitting to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_3", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans did I just open in total?", + "answer": "four cans.", + "question_chinese": "刚才我一共开了几个罐头？", + "answer_chinese": "4个。", + "evidence": [ + { + "description": "Opened the 1st can.", + "timestamps": { + "start": "00:13:28", + "end": "00:13:42" + } + }, + { + "description": "Opened the 2nd can.", + "timestamps": { + "start": "00:13:59", + "end": "00:14:11" + } + }, + { + "description": "Opened the 3rd can.", + "timestamps": { + "start": "00:14:47", + "end": "00:14:58" + } + }, + { + "description": "Opened the 4th can.", + "timestamps": { + "start": "00:15:05", + "end": "00:15:20" + } + } + ], + "question_timestamps": "00:16:15", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "three cans.", + "B": "five cans.", + "C": "You opened a total of two cans just now, not as many as four.", + "D": "You opened a total of six cans in all, which is more than the number opened.", + "E": "four cans." + }, + "answer_label": "E", + "answer_text": "four cans.", + "q_id": "P06_101_3", + "correct_answer": "four cans.", + "original_options": { + "A": "three cans.", + "B": "five cans.", + "C": "You opened a total of two cans just now, not as many as four.", + "D": "You opened a total of six cans in all, which is more than the number opened.", + "E": "four cans." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_4", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans did I just open in total?", + "answer": "four cans.", + "question_chinese": "刚才我一共开了几个罐头？", + "answer_chinese": "4个。", + "evidence": [ + { + "description": "Opened the 1st can.", + "timestamps": { + "start": "00:13:28", + "end": "00:13:42" + } + }, + { + "description": "Opened the 2nd can.", + "timestamps": { + "start": "00:13:59", + "end": "00:14:11" + } + }, + { + "description": "Opened the 3rd can.", + "timestamps": { + "start": "00:14:47", + "end": "00:14:58" + } + }, + { + "description": "Opened the 4th can.", + "timestamps": { + "start": "00:15:05", + "end": "00:15:20" + } + } + ], + "question_timestamps": "00:17:57", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One can.", + "B": "four cans.", + "C": "Five cans.", + "D": "You opened a total of three cans, which is fewer than the actual amount.", + "E": "You opened a total of seven cans in total, far more than what you opened." + }, + "answer_label": "B", + "answer_text": "four cans.", + "q_id": "P06_101_4", + "correct_answer": "four cans.", + "original_options": { + "A": "One can.", + "B": "four cans.", + "C": "Five cans.", + "D": "You opened a total of three cans, which is fewer than the actual amount.", + "E": "You opened a total of seven cans in total, far more than what you opened." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_2_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans did I open in total just now?", + "answer": "You opened a total of four, and they are all to your left.", + "question_chinese": "刚才我一共开了几个罐头？", + "answer_chinese": "你一共开了4个，它们都在你的左边。", + "evidence": [ + { + "description": "Opened the 1st can.", + "timestamps": { + "start": "00:13:28", + "end": "00:13:42" + } + }, + { + "description": "Opened the 2nd can.", + "timestamps": { + "start": "00:13:59", + "end": "00:14:11" + } + }, + { + "description": "Opened the 3rd can.", + "timestamps": { + "start": "00:14:47", + "end": "00:14:58" + } + }, + { + "description": "Opened the 4th can.", + "timestamps": { + "start": "00:15:05", + "end": "00:15:20" + } + } + ], + "question_timestamps": "00:17:58", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened two cans total.", + "B": "You opened five cans total.", + "C": "You opened a total of four, and they are all to your left.", + "D": "You opened a total of three, and they are all to your left, not four cans.", + "E": "You opened a total of six, and they are all to your left, which is more than you opened." + }, + "answer_label": "C", + "answer_text": "You opened a total of four, and they are all to your left.", + "q_id": "P06_101_2_T2", + "correct_answer": "You opened a total of four, and they are all to your left.", + "original_options": { + "A": "You opened two cans total.", + "B": "You opened five cans total.", + "C": "You opened a total of four, and they are all to your left.", + "D": "You opened a total of three, and they are all to your left, not four cans.", + "E": "You opened a total of six, and they are all to your left, which is more than you opened." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 19 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_7_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is this small trash can on the countertop relative to me?", + "answer": "It is on the windowsill by the sink, to your front-left.", + "question_chinese": "台面上的这个小垃圾桶在我的哪里？", + "answer_chinese": "在你左前方洗碗池的窗台上。", + "evidence": [ + { + "description": "It's on the windowsill by the sink in your left front.", + "timestamps": { + "start": "00:29:07", + "end": "00:29:09" + } + } + ], + "question_timestamps": "00:29:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is on the windowsill by the sink, to your front-left.", + "B": "On the windowsill by the sink, front-right.", + "C": "On the windowsill by the sink, back-left.", + "D": "It is on the windowsill by the sink, positioned slightly to your front-right as you face the counter.", + "E": "It is on the windowsill by the sink, set behind you toward the left side rather than in front." + }, + "answer_label": "A", + "answer_text": "It is on the windowsill by the sink, to your front-left.", + "q_id": "P06_101_7_T2", + "correct_answer": "It is on the windowsill by the sink, to your front-left.", + "original_options": { + "A": "It is on the windowsill by the sink, to your front-left.", + "B": "On the windowsill by the sink, front-right.", + "C": "On the windowsill by the sink, back-left.", + "D": "It is on the windowsill by the sink, positioned slightly to your front-right as you face the counter.", + "E": "It is on the windowsill by the sink, set behind you toward the left side rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 19 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_7_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is this small trash can on the countertop relative to me?", + "answer": "It is in front of you.", + "question_chinese": "台面上的这个小垃圾桶在我的哪里？", + "answer_chinese": "在你的前面。", + "evidence": [ + { + "description": "It's in front of you.", + "timestamps": { + "start": "00:29:10", + "end": "00:29:14" + } + } + ], + "question_timestamps": "00:29:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is in front of you.", + "C": "It is to your left.", + "D": "It is to your right as you face the countertop, not directly in front of you.", + "E": "It is behind you rather than ahead, so you would need to turn around to see it." + }, + "answer_label": "B", + "answer_text": "It is in front of you.", + "q_id": "P06_101_7_T1", + "correct_answer": "It is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is in front of you.", + "C": "It is to your left.", + "D": "It is to your right as you face the countertop, not directly in front of you.", + "E": "It is behind you rather than ahead, so you would need to turn around to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 17 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_13", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should this small trash can on the counter be put back later?", + "answer": "It should be put back on the windowsill in front of the left of the sink.", + "question_chinese": "台面上的这个小垃圾桶等会应该放回哪里？", + "answer_chinese": "应该放回洗碗池左前方的窗台上。", + "evidence": [ + { + "description": "Took the small trash can from the windowsill in front of the left of the sink.", + "timestamps": { + "start": "00:29:07", + "end": "00:29:12" + } + } + ], + "question_timestamps": "00:29:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On countertop right of sink.", + "B": "On windowsill above sink.", + "C": "It should be put back on the windowsill in front of the left of the sink.", + "D": "It should be put back on the windowsill in front of the right side of the sink, not left.", + "E": "It should be put back on the countertop beside the left of the sink, not up on the windowsill." + }, + "answer_label": "C", + "answer_text": "It should be put back on the windowsill in front of the left of the sink.", + "q_id": "P06_101_13", + "correct_answer": "It should be put back on the windowsill in front of the left of the sink.", + "original_options": { + "A": "On countertop right of sink.", + "B": "On windowsill above sink.", + "C": "It should be put back on the windowsill in front of the left of the sink.", + "D": "It should be put back on the windowsill in front of the right side of the sink, not left.", + "E": "It should be put back on the countertop beside the left of the sink, not up on the windowsill." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 19, + 19 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_14", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where should this small trash can on the counter be put back later?", + "answer": "It should be put back on the windowsill in front of the left of the sink.", + "question_chinese": "台面上的这个小垃圾桶等会应该放回哪里？", + "answer_chinese": "应该放回洗碗池左前方的窗台上。", + "evidence": [ + { + "description": "Took the small trash can from the windowsill in front of the left of the sink.", + "timestamps": { + "start": "00:29:07", + "end": "00:29:12" + } + } + ], + "question_timestamps": "00:30:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put it on the countertop in front-left.", + "B": "It should be put back on the windowsill in front of the left of the sink.", + "C": "Put it on the table in front-left.", + "D": "It should be put back on the shelf in front of the right side of the sink instead of the left.", + "E": "It should be put back on the floor in front of the right of the sink, not on the windowsill." + }, + "answer_label": "B", + "answer_text": "It should be put back on the windowsill in front of the left of the sink.", + "q_id": "P06_101_14", + "correct_answer": "It should be put back on the windowsill in front of the left of the sink.", + "original_options": { + "A": "Put it on the countertop in front-left.", + "B": "It should be put back on the windowsill in front of the left of the sink.", + "C": "Put it on the table in front-left.", + "D": "It should be put back on the shelf in front of the right side of the sink instead of the left.", + "E": "It should be put back on the floor in front of the right of the sink, not on the windowsill." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 20 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_5_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sweet potatoes did I cut in total?", + "answer": "You cut one sweet potato, and it is on your left-hand side.", + "question_chinese": "我一共切了几个红薯？", + "answer_chinese": "你切了一个红薯，在你的左手边。", + "evidence": [ + { + "description": "Cutting the first sweet potato.", + "timestamps": { + "start": "00:19:31", + "end": "00:21:43" + } + } + ], + "question_timestamps": "00:21:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You cut zero sweet potatoes.", + "B": "You cut two sweet potatoes.", + "C": "You cut three sweet potatoes, and they are on your left-hand side.", + "D": "You cut four sweet potatoes, and they are on your left-hand side near where you were working.", + "E": "You cut one sweet potato, and it is on your left-hand side." + }, + "answer_label": "E", + "answer_text": "You cut one sweet potato, and it is on your left-hand side.", + "q_id": "P06_101_5_T1", + "correct_answer": "You cut one sweet potato, and it is on your left-hand side.", + "original_options": { + "A": "You cut zero sweet potatoes.", + "B": "You cut two sweet potatoes.", + "C": "You cut three sweet potatoes, and they are on your left-hand side.", + "D": "You cut four sweet potatoes, and they are on your left-hand side near where you were working.", + "E": "You cut one sweet potato, and it is on your left-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 17 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_5_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sweet potatoes did I cut in total?", + "answer": "You cut a total of two, and they are both in front of you.", + "question_chinese": "我一共切了几个红薯？", + "answer_chinese": "你一共切了两个，都在你的前面", + "evidence": [ + { + "description": "Cutting the first sweet potato.", + "timestamps": { + "start": "00:19:31", + "end": "00:21:43" + } + }, + { + "description": "Cutting the second sweet potato.", + "timestamps": { + "start": "00:21:44", + "end": "00:23:10" + } + } + ], + "question_timestamps": "00:23:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You cut a total of two, and they are both in front of you.", + "B": "You cut zero in total.", + "C": "You cut one in total.", + "D": "You cut a total of three, and they are all in front of you.", + "E": "You cut a total of four, and they are all in front of you on the cutting surface." + }, + "answer_label": "A", + "answer_text": "You cut a total of two, and they are both in front of you.", + "q_id": "P06_101_5_T2", + "correct_answer": "You cut a total of two, and they are both in front of you.", + "original_options": { + "A": "You cut a total of two, and they are both in front of you.", + "B": "You cut zero in total.", + "C": "You cut one in total.", + "D": "You cut a total of three, and they are all in front of you.", + "E": "You cut a total of four, and they are all in front of you on the cutting surface." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 18 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_9", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sweet potatoes have I cut in total?", + "answer": "Two.", + "question_chinese": "我一共切了几个红薯？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "Cutting the first sweet potato.", + "timestamps": { + "start": "00:19:31", + "end": "00:21:43" + } + }, + { + "description": "Cutting the second sweet potato.", + "timestamps": { + "start": "00:21:44", + "end": "00:23:10" + } + } + ], + "question_timestamps": "00:23:22", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You cut four sweet potatoes in total, which is more than the correct total of two.", + "D": "You cut five sweet potatoes altogether, so the total is higher than two.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "P06_101_9", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You cut four sweet potatoes in total, which is more than the correct total of two.", + "D": "You cut five sweet potatoes altogether, so the total is higher than two.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 13 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_10", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sweet potatoes have I cut in total?", + "answer": "Two.", + "question_chinese": "我一共切了几个红薯？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "Cutting the first sweet potato.", + "timestamps": { + "start": "00:19:31", + "end": "00:21:43" + } + }, + { + "description": "Cutting the second sweet potato.", + "timestamps": { + "start": "00:21:44", + "end": "00:23:10" + } + } + ], + "question_timestamps": "00:27:07", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You cut four sweet potatoes in total, which is more than the correct total of two.", + "D": "Two.", + "E": "You cut five sweet potatoes altogether, so the total is higher than two." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "P06_101_10", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You cut four sweet potatoes in total, which is more than the correct total of two.", + "D": "Two.", + "E": "You cut five sweet potatoes altogether, so the total is higher than two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 13 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_11", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I just take to put the can back in the refrigerator?", + "answer": "Walked from the right of the sink to the left to the refrigerator, put the can back in the refrigerator, then walked from the refrigerator to the right past the sink, and returned to the counter in front of the right of the sink.", + "question_chinese": "我刚才把罐头放回冰箱的过程中经过了什么路径？", + "answer_chinese": "从洗碗池右边向左走到冰箱，将罐头放回冰箱，然后从冰箱向右走经过洗碗池，回到洗碗池右边的台面前。", + "evidence": [ + { + "description": "Walked from the right of the sink to the left to the refrigerator and put the can back in the refrigerator.", + "timestamps": { + "start": "00:28:00", + "end": "00:28:06" + } + }, + { + "description": "Walked from the refrigerator to the right past the sink and returned to the counter in front of the right of the sink.", + "timestamps": { + "start": "00:28:07", + "end": "00:28:12" + } + } + ], + "question_timestamps": "00:28:17", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walked from the right of the sink to the left to the refrigerator, put the can back in the refrigerator, then walked from the refrigerator to the right past the sink, and returned to the counter in front of the right of the sink.", + "B": "Walked from the left of the sink.", + "C": "Went straight to the refrigerator and back.", + "D": "Walked from the right of the sink to the refrigerator, then returned left past the sink to the counter.", + "E": "Started at the counter by the sink, went to the refrigerator to store the can, then came back to the sink area." + }, + "answer_label": "A", + "answer_text": "Walked from the right of the sink to the left to the refrigerator, put the can back in the refrigerator, then walked from the refrigerator to the right past the sink, and returned to the counter in front of the right of the sink.", + "q_id": "P06_101_11", + "correct_answer": "Walked from the right of the sink to the left to the refrigerator, put the can back in the refrigerator, then walked from the refrigerator to the right past the sink, and returned to the counter in front of the right of the sink.", + "original_options": { + "A": "Walked from the right of the sink to the left to the refrigerator, put the can back in the refrigerator, then walked from the refrigerator to the right past the sink, and returned to the counter in front of the right of the sink.", + "B": "Walked from the left of the sink.", + "C": "Went straight to the refrigerator and back.", + "D": "Walked from the right of the sink to the refrigerator, then returned left past the sink to the counter.", + "E": "Started at the counter by the sink, went to the refrigerator to store the can, then came back to the sink area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 22 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_12", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I just take to put the can back in the refrigerator?", + "answer": "Walked from the right of the sink to the left to the refrigerator, put the can back in the refrigerator, then walked from the refrigerator to the right past the sink, and returned to the counter in front of the right of the sink.", + "question_chinese": "我刚才把罐头放回冰箱的过程中经过了什么路径？", + "answer_chinese": "从洗碗池右边向左走到冰箱，将罐头放回冰箱，然后从冰箱向右走经过洗碗池，回到洗碗池右边的台面前。", + "evidence": [ + { + "description": "Walked from the right of the sink to the left to the refrigerator and put the can back in the refrigerator.", + "timestamps": { + "start": "00:28:00", + "end": "00:28:06" + } + }, + { + "description": "Walked from the refrigerator to the right past the sink and returned to the counter in front of the right of the sink.", + "timestamps": { + "start": "00:28:07", + "end": "00:28:12" + } + } + ], + "question_timestamps": "00:28:52", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walked from the left of the sink.", + "B": "Came back past the stove, not the sink.", + "C": "Walked from the right of the sink to the refrigerator, put the can in, then returned left to the counter.", + "D": "Walked from the right of the stove to the refrigerator, then went right past the sink to the counter.", + "E": "Walked from the right of the sink to the left to the refrigerator, put the can back in the refrigerator, then walked from the refrigerator to the right past the sink, and returned to the counter in front of the right of the sink." + }, + "answer_label": "E", + "answer_text": "Walked from the right of the sink to the left to the refrigerator, put the can back in the refrigerator, then walked from the refrigerator to the right past the sink, and returned to the counter in front of the right of the sink.", + "q_id": "P06_101_12", + "correct_answer": "Walked from the right of the sink to the left to the refrigerator, put the can back in the refrigerator, then walked from the refrigerator to the right past the sink, and returned to the counter in front of the right of the sink.", + "original_options": { + "A": "Walked from the left of the sink.", + "B": "Came back past the stove, not the sink.", + "C": "Walked from the right of the sink to the refrigerator, put the can in, then returned left to the counter.", + "D": "Walked from the right of the stove to the refrigerator, then went right past the sink to the counter.", + "E": "Walked from the right of the sink to the left to the refrigerator, put the can back in the refrigerator, then walked from the refrigerator to the right past the sink, and returned to the counter in front of the right of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 20, + 19 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_9_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to eat a banana—where are the bananas?", + "answer": "The bananas are above the countertop to the left of the sink.", + "question_chinese": "我想吃香蕉，香蕉在哪里？", + "answer_chinese": "香蕉在洗碗池左边台面的上方。", + "evidence": [ + { + "description": "The bananas are on the counter above the left side of the sink.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right of the sink.", + "B": "In front of the sink.", + "C": "The bananas are above the countertop to the right of the sink, near the edge of the counter.", + "D": "The bananas are above the countertop to the left of the sink.", + "E": "The bananas are above the countertop behind the sink area, not on the left side." + }, + "answer_label": "D", + "answer_text": "The bananas are above the countertop to the left of the sink.", + "q_id": "P06_101_9_T2", + "correct_answer": "The bananas are above the countertop to the left of the sink.", + "original_options": { + "A": "Right of the sink.", + "B": "In front of the sink.", + "C": "The bananas are above the countertop to the right of the sink, near the edge of the counter.", + "D": "The bananas are above the countertop to the left of the sink.", + "E": "The bananas are above the countertop behind the sink area, not on the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 15 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_9_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to eat a banana—where are the bananas?", + "answer": "The bananas are to your front-left.", + "question_chinese": "我想吃香蕉，香蕉在哪里？", + "answer_chinese": "香蕉在你左边前方。", + "evidence": [ + { + "description": "The banana is above the sink counter on my left.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bananas are to your front-right.", + "B": "The bananas are to your back-left.", + "C": "The bananas are to your back-right, behind you rather than in front of you.", + "D": "The bananas are to your front-left.", + "E": "The bananas are directly in front of you, not off to your front-left side." + }, + "answer_label": "D", + "answer_text": "The bananas are to your front-left.", + "q_id": "P06_101_9_T1", + "correct_answer": "The bananas are to your front-left.", + "original_options": { + "A": "The bananas are to your front-right.", + "B": "The bananas are to your back-left.", + "C": "The bananas are to your back-right, behind you rather than in front of you.", + "D": "The bananas are to your front-left.", + "E": "The bananas are directly in front of you, not off to your front-left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_18", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to eat a banana, where are the bananas?", + "answer": "The bananas are on the counter above the left side of the sink.", + "question_chinese": "我想吃香蕉，香蕉在哪里？", + "answer_chinese": "香蕉在洗碗池左边台面的上方。", + "evidence": [ + { + "description": "The bananas are on the counter above the left side of the sink.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:01:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Above the sink’s right side.", + "B": "On the counter by the fridge.", + "C": "The bananas are on the counter above the right side of the sink, not the left side.", + "D": "The bananas are on the counter above the left side of the sink.", + "E": "The bananas are on a nearby counter section away from the sink area rather than above the left basin." + }, + "answer_label": "D", + "answer_text": "The bananas are on the counter above the left side of the sink.", + "q_id": "P06_101_18", + "correct_answer": "The bananas are on the counter above the left side of the sink.", + "original_options": { + "A": "Above the sink’s right side.", + "B": "On the counter by the fridge.", + "C": "The bananas are on the counter above the right side of the sink, not the left side.", + "D": "The bananas are on the counter above the left side of the sink.", + "E": "The bananas are on a nearby counter section away from the sink area rather than above the left basin." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_4_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "The plate of beans in my hand right now is about how many cans’ worth?", + "answer": "The plate of beans in your hand is the equivalent of one can, and the plate is in front of you.", + "question_chinese": "我手里的这盘豆子有几个罐头的量？", + "answer_chinese": "你手里的这盘豆子有一个罐头的量，盘子在你的前面", + "evidence": [ + { + "description": "Poured one can of beans onto the plate.", + "timestamps": { + "start": "00:14:49", + "end": "00:14:58" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero cans’ worth.", + "B": "Two cans’ worth.", + "C": "The plate of beans in your hand is about three cans’ worth, and the plate is in front of you.", + "D": "The plate of beans in your hand is the equivalent of one can, and the plate is in front of you.", + "E": "The plate of beans in your hand is roughly four cans’ worth, and the plate is in front of you." + }, + "answer_label": "D", + "answer_text": "The plate of beans in your hand is the equivalent of one can, and the plate is in front of you.", + "q_id": "P06_101_4_T1", + "correct_answer": "The plate of beans in your hand is the equivalent of one can, and the plate is in front of you.", + "original_options": { + "A": "Zero cans’ worth.", + "B": "Two cans’ worth.", + "C": "The plate of beans in your hand is about three cans’ worth, and the plate is in front of you.", + "D": "The plate of beans in your hand is the equivalent of one can, and the plate is in front of you.", + "E": "The plate of beans in your hand is roughly four cans’ worth, and the plate is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 20, + 20 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_4_T2", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "The plate of beans in my hand right now is about how many cans’ worth?", + "answer": "There is the equivalent of two cans on your plate, and the plate is on your right-hand side.", + "question_chinese": "我手里的这盘豆子有几个罐头的量？", + "answer_chinese": "你的盘子里有两个罐头的量，盘子在你的右手边。", + "evidence": [ + { + "description": "Poured one can of beans onto the plate.", + "timestamps": { + "start": "00:14:49", + "end": "00:14:58" + } + }, + { + "description": "Poured another can of beans onto the plate.", + "timestamps": { + "start": "00:15:07", + "end": "00:15:20" + } + } + ], + "question_timestamps": "00:15:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One can’s worth.", + "B": "Three cans’ worth.", + "C": "There is the equivalent of zero cans on your plate, and the plate is on your right-hand side.", + "D": "There is the equivalent of four cans on your plate, and the plate is on your right-hand side.", + "E": "There is the equivalent of two cans on your plate, and the plate is on your right-hand side." + }, + "answer_label": "E", + "answer_text": "There is the equivalent of two cans on your plate, and the plate is on your right-hand side.", + "q_id": "P06_101_4_T2", + "correct_answer": "There is the equivalent of two cans on your plate, and the plate is on your right-hand side.", + "original_options": { + "A": "One can’s worth.", + "B": "Three cans’ worth.", + "C": "There is the equivalent of zero cans on your plate, and the plate is on your right-hand side.", + "D": "There is the equivalent of four cans on your plate, and the plate is on your right-hand side.", + "E": "There is the equivalent of two cans on your plate, and the plate is on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 18, + 18 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_7", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans of beans are on this plate in my hand?", + "answer": "Two cans.", + "question_chinese": "我手里的这盘豆子有几个罐头的量？", + "answer_chinese": "两个罐头的量。", + "evidence": [ + { + "description": "Poured one can of beans onto the plate.", + "timestamps": { + "start": "00:14:49", + "end": "00:14:58" + } + }, + { + "description": "Poured another can of beans onto the plate.", + "timestamps": { + "start": "00:15:07", + "end": "00:15:20" + } + } + ], + "question_timestamps": "00:15:35", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two cans.", + "B": "One can.", + "C": "Three cans.", + "D": "There are about four cans on the plate in your hand, spread across the plate surface.", + "E": "Roughly five cans are on the plate in your hand, covering most of the plate area." + }, + "answer_label": "A", + "answer_text": "Two cans.", + "q_id": "P06_101_7", + "correct_answer": "Two cans.", + "original_options": { + "A": "Two cans.", + "B": "One can.", + "C": "Three cans.", + "D": "There are about four cans on the plate in your hand, spread across the plate surface.", + "E": "Roughly five cans are on the plate in your hand, covering most of the plate area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_8", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cans of beans are on this plate in my hand?", + "answer": "Two cans.", + "question_chinese": "我手里的这盘豆子有几个罐头的量？", + "answer_chinese": "两个罐头的量。", + "evidence": [ + { + "description": "Poured one can of beans onto the plate.", + "timestamps": { + "start": "00:14:49", + "end": "00:14:58" + } + }, + { + "description": "Poured another can of beans onto the plate.", + "timestamps": { + "start": "00:15:07", + "end": "00:15:20" + } + } + ], + "question_timestamps": "00:15:57", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One can.", + "B": "Three cans.", + "C": "Four cans of beans are on the plate you are holding in your hand.", + "D": "Two cans.", + "E": "Five cans of beans are stacked together on the plate you’re holding." + }, + "answer_label": "D", + "answer_text": "Two cans.", + "q_id": "P06_101_8", + "correct_answer": "Two cans.", + "original_options": { + "A": "One can.", + "B": "Three cans.", + "C": "Four cans of beans are on the plate you are holding in your hand.", + "D": "Two cans.", + "E": "Five cans of beans are stacked together on the plate you’re holding." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_5", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the microwave now?", + "answer": "No.", + "question_chinese": "我现在能够到到微波炉吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The microwave is about 1 meter to the right of the sink.", + "timestamps": { + "start": "00:14:32", + "end": "00:14:36" + } + }, + { + "description": "You are now standing in front of the sink.", + "timestamps": { + "start": "00:14:45", + "end": "00:14:47" + } + } + ], + "question_timestamps": "00:14:48", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "P06_101_5", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_3_T1", + "qid_num": 27, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the microwave right now?", + "answer": "You can’t reach the microwave right now because it is to your right.", + "question_chinese": "我现在能够到到微波炉吗？", + "answer_chinese": "你现在不能够到微波炉，因为它在你的右边。", + "evidence": [ + { + "description": "The microwave is about 1 meter to the right of the sink.", + "timestamps": { + "start": "00:14:32", + "end": "00:14:36" + } + }, + { + "description": "You are now standing in front of the sink.", + "timestamps": { + "start": "00:14:45", + "end": "00:14:47" + } + } + ], + "question_timestamps": "00:14:59", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the microwave right now; the microwave is to your right.", + "B": "You can’t reach the microwave right now because it is to your right." + }, + "answer_label": "B", + "answer_text": "You can’t reach the microwave right now because it is to your right.", + "q_id": "P06_101_3_T1", + "correct_answer": "You can’t reach the microwave right now because it is to your right.", + "original_options": { + "A": "You can reach the microwave right now; the microwave is to your right.", + "B": "You can’t reach the microwave right now because it is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_3_T2", + "qid_num": 28, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the microwave right now?", + "answer": "You can reach the microwave right now; the microwave is to your right.", + "question_chinese": "我现在能够到到微波炉吗？", + "answer_chinese": "你现在能够够到微波炉，微波炉在你的右边", + "evidence": [ + { + "description": "You can reach the microwave now. It's very close to you.", + "timestamps": { + "start": "00:14:58", + "end": "00:15:00" + } + } + ], + "question_timestamps": "00:15:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the microwave right now; the microwave is to your right.", + "B": "You can’t reach the microwave right now because it is to your right." + }, + "answer_label": "A", + "answer_text": "You can reach the microwave right now; the microwave is to your right.", + "q_id": "P06_101_3_T2", + "correct_answer": "You can reach the microwave right now; the microwave is to your right.", + "original_options": { + "A": "You can reach the microwave right now; the microwave is to your right.", + "B": "You can’t reach the microwave right now because it is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_6", + "qid_num": 29, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the microwave now?", + "answer": "No.", + "question_chinese": "我现在能够到到微波炉吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The microwave is about 1 meter to the right of the sink.", + "timestamps": { + "start": "00:14:32", + "end": "00:14:36" + } + }, + { + "description": "You are now standing in front of the sink.", + "timestamps": { + "start": "00:14:45", + "end": "00:14:47" + } + } + ], + "question_timestamps": "00:15:32", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "P06_101_6", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_1", + "qid_num": 30, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Now, how many times have I opened the refrigerator in total?", + "answer": "three times.", + "question_chinese": "现在，我一共打开了几次冰箱？", + "answer_chinese": "3次。", + "evidence": [ + { + "description": "Opened the refrigerator for the first time.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:38" + } + }, + { + "description": "Opened the refrigerator for the second time.", + "timestamps": { + "start": "00:06:26", + "end": "00:06:42" + } + }, + { + "description": "Opened the refrigerator for the third time.", + "timestamps": { + "start": "00:28:03", + "end": "00:28:08" + } + } + ], + "question_timestamps": "00:28:38", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "three times.", + "B": "Two times.", + "C": "Four times.", + "D": "You opened the refrigerator four times in total so far today already.", + "E": "You opened the refrigerator five times altogether up to this point already." + }, + "answer_label": "A", + "answer_text": "three times.", + "q_id": "P06_101_1", + "correct_answer": "three times.", + "original_options": { + "A": "three times.", + "B": "Two times.", + "C": "Four times.", + "D": "You opened the refrigerator four times in total so far today already.", + "E": "You opened the refrigerator five times altogether up to this point already." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "P06_101", + "video_length": "00:30:59", + "qid": "P06_101_2", + "qid_num": 31, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Now, how many times have I opened the refrigerator in total?", + "answer": "four times.", + "question_chinese": "现在，我一共打开了几次冰箱？", + "answer_chinese": "4次。", + "evidence": [ + { + "description": "Opened the refrigerator for the first time.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:38" + } + }, + { + "description": "Opened the refrigerator for the second time.", + "timestamps": { + "start": "00:06:26", + "end": "00:06:42" + } + }, + { + "description": "Opened the refrigerator for the third time.", + "timestamps": { + "start": "00:28:03", + "end": "00:28:08" + } + }, + { + "description": "Opened the refrigerator for the fourth time.", + "timestamps": { + "start": "00:30:21", + "end": "00:30:26" + } + } + ], + "question_timestamps": "00:30:58", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three times.", + "B": "Five times.", + "C": "You opened the refrigerator three times in total up to now already.", + "D": "four times.", + "E": "You opened the refrigerator six times altogether by this moment already, not just four." + }, + "answer_label": "D", + "answer_text": "four times.", + "q_id": "P06_101_2", + "correct_answer": "four times.", + "original_options": { + "A": "Three times.", + "B": "Five times.", + "C": "You opened the refrigerator three times in total up to now already.", + "D": "four times.", + "E": "You opened the refrigerator six times altogether by this moment already, not just four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + } +] diff --git a/QAs/P06_106-mcq.json b/QAs/P06_106-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..35b8ce92422922d85630bab7c79f43a5b5efedf9 --- /dev/null +++ b/QAs/P06_106-mcq.json @@ -0,0 +1,895 @@ +[ + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_5_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just put the cheese back into the refrigerator?", + "answer": "From facing the countertop, you turned left and walked straight to the fridge, and put the cheese back in the fridge.", + "question_chinese": "我刚才把奶酪放回冰箱经过了什么路径？", + "answer_chinese": "你从面向台面的位置左转直走到冰箱，将奶酪放回冰箱。", + "evidence": [ + { + "description": "Turn left from facing the counter, walk straight to the refrigerator, and put the cheese back.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:46" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right and went to the fridge.", + "B": "You walked straight to the fridge.", + "C": "From facing the countertop, you turned left, reached the fridge, and took the cheese out instead.", + "D": "From facing the countertop, you turned left to the fridge, then went left back to the countertop afterward.", + "E": "From facing the countertop, you turned left and walked straight to the fridge, and put the cheese back in the fridge." + }, + "answer_label": "E", + "answer_text": "From facing the countertop, you turned left and walked straight to the fridge, and put the cheese back in the fridge.", + "q_id": "P06_106_5_T1", + "correct_answer": "From facing the countertop, you turned left and walked straight to the fridge, and put the cheese back in the fridge.", + "original_options": { + "A": "You turned right and went to the fridge.", + "B": "You walked straight to the fridge.", + "C": "From facing the countertop, you turned left, reached the fridge, and took the cheese out instead.", + "D": "From facing the countertop, you turned left to the fridge, then went left back to the countertop afterward.", + "E": "From facing the countertop, you turned left and walked straight to the fridge, and put the cheese back in the fridge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 16, + 18 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_5_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just put the cheese back into the refrigerator?", + "answer": "From facing the countertop, you turned left and walked straight to the fridge, put the cheese back in the fridge, then from in front of the fridge you went straight to the right and arrived in front of the countertop.", + "question_chinese": "我刚才把奶酪放回冰箱经过了什么路径？", + "answer_chinese": "你从面向台面的位置左转直走到冰箱，将奶酪放回冰箱，然后从冰箱前的位置向右直走来到台面前。", + "evidence": [ + { + "description": "Turn left from facing the counter, walk straight to the refrigerator, and put the cheese back.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:46" + } + } + ], + "question_timestamps": "00:02:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From facing the countertop, you turned left and walked straight to the fridge, put the cheese back in the fridge, then from in front of the fridge you went straight to the right and arrived in front of the countertop.", + "B": "You turned right before heading to the fridge.", + "C": "You went left back to the countertop.", + "D": "From facing the countertop, you turned left to the fridge, then walked straight ahead to the countertop.", + "E": "From facing the countertop, you turned left to the fridge, then returned from the fridge’s left side to the countertop." + }, + "answer_label": "A", + "answer_text": "From facing the countertop, you turned left and walked straight to the fridge, put the cheese back in the fridge, then from in front of the fridge you went straight to the right and arrived in front of the countertop.", + "q_id": "P06_106_5_T2", + "correct_answer": "From facing the countertop, you turned left and walked straight to the fridge, put the cheese back in the fridge, then from in front of the fridge you went straight to the right and arrived in front of the countertop.", + "original_options": { + "A": "From facing the countertop, you turned left and walked straight to the fridge, put the cheese back in the fridge, then from in front of the fridge you went straight to the right and arrived in front of the countertop.", + "B": "You turned right before heading to the fridge.", + "C": "You went left back to the countertop.", + "D": "From facing the countertop, you turned left to the fridge, then walked straight ahead to the countertop.", + "E": "From facing the countertop, you turned left to the fridge, then returned from the fridge’s left side to the countertop." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 21 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_7_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pick up the spoon in my hand from?", + "answer": "You took it from the drawer to the left of the sink; it’s to your left.", + "question_chinese": "我手里的这个勺子是从哪里拿的？", + "answer_chinese": "从洗碗池左边的抽屉里拿的，在你的左方。", + "evidence": [ + { + "description": "Take the spoon out of the drawer to the left of the sink.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took it from the right drawer.", + "B": "You took it from the drawer to the left of the sink; it’s to your left.", + "C": "You took it from under the sink.", + "D": "You took it from the drawer right of the sink, positioned to your back-left.", + "E": "You took it from the drawer to the left of the sink, but it was on your right side." + }, + "answer_label": "B", + "answer_text": "You took it from the drawer to the left of the sink; it’s to your left.", + "q_id": "P06_106_7_T1", + "correct_answer": "You took it from the drawer to the left of the sink; it’s to your left.", + "original_options": { + "A": "You took it from the right drawer.", + "B": "You took it from the drawer to the left of the sink; it’s to your left.", + "C": "You took it from under the sink.", + "D": "You took it from the drawer right of the sink, positioned to your back-left.", + "E": "You took it from the drawer to the left of the sink, but it was on your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 19 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_7_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pick up the spoon in my hand from?", + "answer": "You took it from the drawer to the left of the sink; it’s to your back-left.", + "question_chinese": "我手里的这个勺子是从哪里拿的？", + "answer_chinese": "从洗碗池左边的抽屉里拿的，在你的左后方。", + "evidence": [ + { + "description": "Take the spoon out of the drawer to the left of the sink.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:02:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your back-right.", + "B": "It’s to your front-left.", + "C": "You took it from the drawer to the left of the sink; it’s to your back-left.", + "D": "You took it from the drawer to the right of the sink; it’s to your back-left.", + "E": "You took it from the drawer to the left of the sink, but it sits directly to your left." + }, + "answer_label": "C", + "answer_text": "You took it from the drawer to the left of the sink; it’s to your back-left.", + "q_id": "P06_106_7_T2", + "correct_answer": "You took it from the drawer to the left of the sink; it’s to your back-left.", + "original_options": { + "A": "It’s to your back-right.", + "B": "It’s to your front-left.", + "C": "You took it from the drawer to the left of the sink; it’s to your back-left.", + "D": "You took it from the drawer to the right of the sink; it’s to your back-left.", + "E": "You took it from the drawer to the left of the sink, but it sits directly to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_8_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to add a bit more rice right now—where is the rice?", + "answer": "The rice is in the sink, to your left.", + "question_chinese": "我现在想要加一点饭，饭在哪里？", + "answer_chinese": "饭在洗碗池里,在你的左方。", + "evidence": [ + { + "description": "Scoop some rice from the pot in the sink into the plate.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:45" + } + } + ], + "question_timestamps": "00:02:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the sink, to your right.", + "B": "The rice is behind you.", + "C": "The rice is in the sink, to your left.", + "D": "The rice is in the sink, positioned to your back-right rather than on your left.", + "E": "The rice is on the edge of the table, not in the sink beside you." + }, + "answer_label": "C", + "answer_text": "The rice is in the sink, to your left.", + "q_id": "P06_106_8_T1", + "correct_answer": "The rice is in the sink, to your left.", + "original_options": { + "A": "In the sink, to your right.", + "B": "The rice is behind you.", + "C": "The rice is in the sink, to your left.", + "D": "The rice is in the sink, positioned to your back-right rather than on your left.", + "E": "The rice is on the edge of the table, not in the sink beside you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 15 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_8_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to add a bit more rice right now—where is the rice?", + "answer": "The rice is in the sink, to your back-right.", + "question_chinese": "我现在想要加一点饭，饭在哪里？", + "answer_chinese": "饭在洗碗池里,在你的右后方。", + "evidence": [ + { + "description": "Scoop some rice from the pot in the sink into the plate.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:45" + } + } + ], + "question_timestamps": "00:02:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the sink, back-left.", + "B": "In the sink, front-right.", + "C": "The rice is in the sink, to your left, not behind you on the right.", + "D": "The rice is in the sink, to your back-right.", + "E": "The rice is on the edge of the table instead of being in the sink." + }, + "answer_label": "D", + "answer_text": "The rice is in the sink, to your back-right.", + "q_id": "P06_106_8_T2", + "correct_answer": "The rice is in the sink, to your back-right.", + "original_options": { + "A": "In the sink, back-left.", + "B": "In the sink, front-right.", + "C": "The rice is in the sink, to your left, not behind you on the right.", + "D": "The rice is in the sink, to your back-right.", + "E": "The rice is on the edge of the table instead of being in the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_3_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are left on the stove right now?", + "answer": "There is one pot left.", + "question_chinese": "我现在灶台上还剩几口锅？", + "answer_chinese": "还剩一口锅。", + "evidence": [ + { + "description": "There are two pots on the stove.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:23" + } + }, + { + "description": "One pot was taken away.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "No pots are left.", + "B": "Two pots are left.", + "C": "Three pots are left on the stove right now, not just a single one.", + "D": "There is one pot left.", + "E": "Four pots are left on the stove at the moment, meaning multiple pots remain." + }, + "answer_label": "D", + "answer_text": "There is one pot left.", + "q_id": "P06_106_3_T1", + "correct_answer": "There is one pot left.", + "original_options": { + "A": "No pots are left.", + "B": "Two pots are left.", + "C": "Three pots are left on the stove right now, not just a single one.", + "D": "There is one pot left.", + "E": "Four pots are left on the stove at the moment, meaning multiple pots remain." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_3_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are left on the stove right now?", + "answer": "There is one pot left, to your right.", + "question_chinese": "我现在灶台上还剩几口锅？", + "answer_chinese": "还剩一口锅，在你的右方。", + "evidence": [ + { + "description": "There are two pots on the stove.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:22" + } + }, + { + "description": "One pot was taken away.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are zero pots left.", + "B": "There are two pots left.", + "C": "There are three pots left on the stove, with the nearest one sitting to your right side.", + "D": "There is one pot left, to your right.", + "E": "There are four pots left on the stove, clustered together on the burner area to your right." + }, + "answer_label": "D", + "answer_text": "There is one pot left, to your right.", + "q_id": "P06_106_3_T2", + "correct_answer": "There is one pot left, to your right.", + "original_options": { + "A": "There are zero pots left.", + "B": "There are two pots left.", + "C": "There are three pots left on the stove, with the nearest one sitting to your right side.", + "D": "There is one pot left, to your right.", + "E": "There are four pots left on the stove, clustered together on the burner area to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from entering the kitchen to my current position?", + "answer": "After entering the kitchen, you walked straight to the stove, then turned left to the sink, and then turned right back to the stove.", + "question_chinese": "我进厨房到现在的位置经过了什么路径？", + "answer_chinese": "你进厨房后直走到灶台前，然后左转到洗碗池前，再右转回到灶台前。", + "evidence": [ + { + "description": "Enter the kitchen and walk straight to the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "Turn left from the stove and walk to the sink.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:13" + } + }, + { + "description": "Turn right from the sink and walk to the stove.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You went to the sink first.", + "B": "You went straight to the fridge.", + "C": "After entering, you walked straight to the sink, then turned right to the stove, then stopped there.", + "D": "After entering the kitchen, you walked to the stove, turned right to the counter, then turned left to the sink.", + "E": "After entering the kitchen, you walked straight to the stove, then turned left to the sink, and then turned right back to the stove." + }, + "answer_label": "E", + "answer_text": "After entering the kitchen, you walked straight to the stove, then turned left to the sink, and then turned right back to the stove.", + "q_id": "P06_106_1", + "correct_answer": "After entering the kitchen, you walked straight to the stove, then turned left to the sink, and then turned right back to the stove.", + "original_options": { + "A": "You went to the sink first.", + "B": "You went straight to the fridge.", + "C": "After entering, you walked straight to the sink, then turned right to the stove, then stopped there.", + "D": "After entering the kitchen, you walked to the stove, turned right to the counter, then turned left to the sink.", + "E": "After entering the kitchen, you walked straight to the stove, then turned left to the sink, and then turned right back to the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 20 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from entering the kitchen to my current position?", + "answer": "After entering the kitchen, you walked straight to the stove, then turned left to the sink, then turned right back to the stove, and finally turned left again to the sink.", + "question_chinese": "我进厨房到现在的位置经过了什么路径？", + "answer_chinese": "你进厨房后直走到灶台前，然后左转到洗碗池前，再右转回到灶台前，最后再次左转到洗碗池前。", + "evidence": [ + { + "description": "Enter the kitchen and walk straight to the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "Turn left from the stove and walk to the sink.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:13" + } + }, + { + "description": "Turn right from the sink and walk to the stove.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:18" + } + }, + { + "description": "Turn left again from the front of the stove to the front of the sink.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "After entering the kitchen, you walked straight to the stove, then turned left to the sink, then turned right back to the stove, and finally turned left again to the sink.", + "B": "You ended at the stove.", + "C": "You ended at the counter.", + "D": "After entering, you went straight to the stove, left to the sink, then stopped at the sink.", + "E": "After entering, you walked straight to the sink, turned right to the stove, then turned left to the counter." + }, + "answer_label": "A", + "answer_text": "After entering the kitchen, you walked straight to the stove, then turned left to the sink, then turned right back to the stove, and finally turned left again to the sink.", + "q_id": "P06_106_2", + "correct_answer": "After entering the kitchen, you walked straight to the stove, then turned left to the sink, then turned right back to the stove, and finally turned left again to the sink.", + "original_options": { + "A": "After entering the kitchen, you walked straight to the stove, then turned left to the sink, then turned right back to the stove, and finally turned left again to the sink.", + "B": "You ended at the stove.", + "C": "You ended at the counter.", + "D": "After entering, you went straight to the stove, left to the sink, then stopped at the sink.", + "E": "After entering, you walked straight to the sink, turned right to the stove, then turned left to the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_6_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine relative to me?", + "answer": "The washing machine is to your back-left.", + "question_chinese": "洗衣机在我的哪个方位？", + "answer_chinese": "洗衣机在你的左后边。", + "evidence": [ + { + "description": "Upon entering the kitchen, the washing machine was on the left side of the field of view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:02:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The washing machine is back-right.", + "B": "The washing machine is to your back-left.", + "C": "The washing machine is front-left.", + "D": "The washing machine is to your front-right, positioned ahead of you on the right-hand side.", + "E": "The washing machine is below you on your left, down and slightly off to your left side." + }, + "answer_label": "B", + "answer_text": "The washing machine is to your back-left.", + "q_id": "P06_106_6_T2", + "correct_answer": "The washing machine is to your back-left.", + "original_options": { + "A": "The washing machine is back-right.", + "B": "The washing machine is to your back-left.", + "C": "The washing machine is front-left.", + "D": "The washing machine is to your front-right, positioned ahead of you on the right-hand side.", + "E": "The washing machine is below you on your left, down and slightly off to your left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_6_T1", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine relative to me?", + "answer": "The washing machine is below you on your left.", + "question_chinese": "洗衣机在我的哪个方位？", + "answer_chinese": "洗衣机在你的左下方。", + "evidence": [ + { + "description": "Upon entering the kitchen, the washing machine was on the left side of the field of view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:02:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The washing machine is below you on your left.", + "B": "The washing machine is below-right.", + "C": "The washing machine is above-left.", + "D": "The washing machine is to your back-left, located behind you and a bit to the left.", + "E": "The washing machine is to your front-left, positioned ahead of you on the left side." + }, + "answer_label": "A", + "answer_text": "The washing machine is below you on your left.", + "q_id": "P06_106_6_T1", + "correct_answer": "The washing machine is below you on your left.", + "original_options": { + "A": "The washing machine is below you on your left.", + "B": "The washing machine is below-right.", + "C": "The washing machine is above-left.", + "D": "The washing machine is to your back-left, located behind you and a bit to the left.", + "E": "The washing machine is to your front-left, positioned ahead of you on the left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_2_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me, the microwave or the washing machine?", + "answer": "The microwave is closer to you, to your upper-left.", + "question_chinese": "现在，微波炉和洗衣机，哪个离我更近？", + "answer_chinese": "微波炉离你更近，在你的左上方。", + "evidence": [ + { + "description": "The washing machine is to the lower left of the sink.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The washing machine is closer to you, below you.", + "B": "The microwave is closer to you, to your upper-left." + }, + "answer_label": "B", + "answer_text": "The microwave is closer to you, to your upper-left.", + "q_id": "P06_106_2_T1", + "correct_answer": "The microwave is closer to you, to your upper-left.", + "original_options": { + "A": "The washing machine is closer to you, below you.", + "B": "The microwave is closer to you, to your upper-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_2_T2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me, the microwave or the washing machine?", + "answer": "The washing machine is closer to you, below you.", + "question_chinese": "现在，微波炉和洗衣机，哪个离我更近？", + "answer_chinese": "洗衣机离你更近，在你的下方。", + "evidence": [ + { + "description": "The washing machine is to the lower left of the sink.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "You are now standing in front of the sink.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The washing machine is closer to you, below you.", + "B": "The microwave is closer to you, to your upper-left." + }, + "answer_label": "A", + "answer_text": "The washing machine is closer to you, below you.", + "q_id": "P06_106_2_T2", + "correct_answer": "The washing machine is closer to you, below you.", + "original_options": { + "A": "The washing machine is closer to you, below you.", + "B": "The microwave is closer to you, to your upper-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_4_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the potted plant relative to me right now?", + "answer": "The potted plant is in front of and above the sink to your right.", + "question_chinese": "盆栽现在在我的哪个方位？", + "answer_chinese": "盆栽在你右边的洗碗池的前上方。", + "evidence": [ + { + "description": "The potted plant is above and in front of the sink.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:16" + } + }, + { + "description": "The sink is about 1.5 meters to the right of the refrigerator.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:02:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s above the sink to your left.", + "B": "It’s above the sink to your right.", + "C": "The potted plant is behind and above the sink to your right, sitting back from the sink area.", + "D": "The potted plant is in front of and above the sink to your right.", + "E": "The potted plant is in front of and below the sink to your right, positioned lower than the sink level." + }, + "answer_label": "D", + "answer_text": "The potted plant is in front of and above the sink to your right.", + "q_id": "P06_106_4_T1", + "correct_answer": "The potted plant is in front of and above the sink to your right.", + "original_options": { + "A": "It’s above the sink to your left.", + "B": "It’s above the sink to your right.", + "C": "The potted plant is behind and above the sink to your right, sitting back from the sink area.", + "D": "The potted plant is in front of and above the sink to your right.", + "E": "The potted plant is in front of and below the sink to your right, positioned lower than the sink level." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 20 + ] + }, + { + "video_uid": "P06_106", + "video_length": "00:02:52", + "qid": "P06_106_4_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the potted plant relative to me right now?", + "answer": "The potted plant is within arm’s reach to your right.", + "question_chinese": "盆栽现在在我的哪个方位？", + "answer_chinese": "盆栽在你右边的手边。", + "evidence": [ + { + "description": "The potted plant is above and in front of the sink.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:02:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Within arm’s reach to your left.", + "B": "Just out of reach to your right.", + "C": "The potted plant is behind you, slightly to the right, not within arm’s reach.", + "D": "The potted plant is in front of you to the right, close but not within arm’s reach.", + "E": "The potted plant is within arm’s reach to your right." + }, + "answer_label": "E", + "answer_text": "The potted plant is within arm’s reach to your right.", + "q_id": "P06_106_4_T2", + "correct_answer": "The potted plant is within arm’s reach to your right.", + "original_options": { + "A": "Within arm’s reach to your left.", + "B": "Just out of reach to your right.", + "C": "The potted plant is behind you, slightly to the right, not within arm’s reach.", + "D": "The potted plant is in front of you to the right, close but not within arm’s reach.", + "E": "The potted plant is within arm’s reach to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 18 + ] + } +] diff --git a/QAs/P06_108-mcq.json b/QAs/P06_108-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..984d56755aa245bc39a2921c621842591484d91f --- /dev/null +++ b/QAs/P06_108-mcq.json @@ -0,0 +1,1428 @@ +[ + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I just make the flatbread?", + "answer": "You made the flatbread on the countertop to the left of the sink.", + "question_chinese": "我刚刚在哪里做的面饼？", + "answer_chinese": "你在洗碗台左边的台面上做的面饼。", + "evidence": [ + { + "description": "To the left of the sink is a countertop with 4 pieces of dough on it.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:26" + } + }, + { + "description": "You moved the dough on the countertop and started making the flatbread.", + "timestamps": { + "start": "00:00:51", + "end": "00:02:44" + } + } + ], + "question_timestamps": "00:05:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the countertop right of the sink.", + "B": "On the stove by the sink.", + "C": "You made the flatbread on the countertop in front of the sink, rather than on its left side.", + "D": "You made the flatbread on a cutting board near the sink, not directly on the left countertop.", + "E": "You made the flatbread on the countertop to the left of the sink." + }, + "answer_label": "E", + "answer_text": "You made the flatbread on the countertop to the left of the sink.", + "q_id": "P06_108_1", + "correct_answer": "You made the flatbread on the countertop to the left of the sink.", + "original_options": { + "A": "On the countertop right of the sink.", + "B": "On the stove by the sink.", + "C": "You made the flatbread on the countertop in front of the sink, rather than on its left side.", + "D": "You made the flatbread on a cutting board near the sink, not directly on the left countertop.", + "E": "You made the flatbread on the countertop to the left of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_1_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was I when I just made the flatbread?", + "answer": "The flatbread you made on the countertop to your back-left.", + "question_chinese": "我刚刚在哪里做的面饼？", + "answer_chinese": "在你的左后方的台面上做的面饼", + "evidence": [ + { + "description": "To the left of the sink is a countertop with 4 pieces of dough on it.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:26" + } + }, + { + "description": "You moved the dough on the countertop and started making the flatbread.", + "timestamps": { + "start": "00:00:51", + "end": "00:02:44" + } + } + ], + "question_timestamps": "00:05:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the countertop to your back-right.", + "B": "On the countertop to your front-left.", + "C": "The flatbread you made was on the countertop directly in front of you, not to your back-left.", + "D": "The flatbread you made on the countertop to your back-left.", + "E": "The flatbread you made sat on the countertop to your right side, rather than behind you on the left." + }, + "answer_label": "D", + "answer_text": "The flatbread you made on the countertop to your back-left.", + "q_id": "P06_108_1_T1", + "correct_answer": "The flatbread you made on the countertop to your back-left.", + "original_options": { + "A": "On the countertop to your back-right.", + "B": "On the countertop to your front-left.", + "C": "The flatbread you made was on the countertop directly in front of you, not to your back-left.", + "D": "The flatbread you made on the countertop to your back-left.", + "E": "The flatbread you made sat on the countertop to your right side, rather than behind you on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_1_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was I when I just made the flatbread?", + "answer": "The flatbread you made on the countertop to the left of the sink.", + "question_chinese": "我刚刚在哪里做的面饼？", + "answer_chinese": "你在洗碗台左边的台面上做的面饼。", + "evidence": [ + { + "description": "To the left of the sink is a countertop with 4 pieces of dough on it.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:26" + } + }, + { + "description": "You moved the dough on the countertop and started making the flatbread.", + "timestamps": { + "start": "00:00:51", + "end": "00:02:44" + } + } + ], + "question_timestamps": "00:12:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The flatbread you made on the countertop to the left of the sink.", + "B": "To the right of the sink.", + "C": "In front of the sink.", + "D": "You were positioned behind the sink area, not at the left-side countertop.", + "E": "You were on the right side of the sink, rather than on the left-side countertop." + }, + "answer_label": "A", + "answer_text": "The flatbread you made on the countertop to the left of the sink.", + "q_id": "P06_108_1_T2", + "correct_answer": "The flatbread you made on the countertop to the left of the sink.", + "original_options": { + "A": "The flatbread you made on the countertop to the left of the sink.", + "B": "To the right of the sink.", + "C": "In front of the sink.", + "D": "You were positioned behind the sink area, not at the left-side countertop.", + "E": "You were on the right side of the sink, rather than on the left-side countertop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 12, + 15 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_15", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was making the flatbread, how many pieces of dough were still on the countertop?", + "answer": "five pieces.", + "question_chinese": "我在做面饼的时候，台面上还有几块面团？", + "answer_chinese": "5块。", + "evidence": [ + { + "description": "In the beginning, there were 4 pieces of dough on the countertop, and one piece of dough on each of the two plates.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:25" + } + }, + { + "description": "You used one piece of dough from a plate.", + "timestamps": { + "start": "00:00:57", + "end": "00:05:30" + } + } + ], + "question_timestamps": "00:08:21", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four pieces.", + "B": "Six pieces.", + "C": "five pieces.", + "D": "Three pieces of dough were still on the countertop while you were making the flatbread.", + "E": "Seven pieces of dough remained on the countertop at that time, not five." + }, + "answer_label": "C", + "answer_text": "five pieces.", + "q_id": "P06_108_15", + "correct_answer": "five pieces.", + "original_options": { + "A": "Four pieces.", + "B": "Six pieces.", + "C": "five pieces.", + "D": "Three pieces of dough were still on the countertop while you were making the flatbread.", + "E": "Seven pieces of dough remained on the countertop at that time, not five." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 13 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_16", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was making the flatbread, how many pieces of dough were still on the countertop?", + "answer": "five pieces.", + "question_chinese": "我在做面饼的时候，台面上还有几块面团？", + "answer_chinese": "5块。", + "evidence": [ + { + "description": "In the beginning, there were 4 pieces of dough on the countertop, and one piece of dough on each of the two plates.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:25" + } + }, + { + "description": "You used one piece of dough from a plate.", + "timestamps": { + "start": "00:00:57", + "end": "00:05:30" + } + } + ], + "question_timestamps": "00:11:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three pieces.", + "B": "Seven pieces.", + "C": "Four pieces of dough were still on the countertop when you were making the flatbread.", + "D": "five pieces.", + "E": "Six pieces of dough remained on the countertop during the flatbread-making step, not five." + }, + "answer_label": "D", + "answer_text": "five pieces.", + "q_id": "P06_108_16", + "correct_answer": "five pieces.", + "original_options": { + "A": "Three pieces.", + "B": "Seven pieces.", + "C": "Four pieces of dough were still on the countertop when you were making the flatbread.", + "D": "five pieces.", + "E": "Six pieces of dough remained on the countertop during the flatbread-making step, not five." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_5_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did the white bowl in my hand come from?", + "answer": "You brought it over from the countertop to your left.", + "question_chinese": "我手中白色的碗是从哪里拿过来的？", + "answer_chinese": "从你左边的台面上拿过来的。", + "evidence": [ + { + "description": "The bowl was taken from the countertop to the left of the sink.", + "timestamps": { + "start": "00:09:42", + "end": "00:09:44" + } + } + ], + "question_timestamps": "00:09:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the countertop to your right.", + "B": "From the countertop behind you.", + "C": "You brought it over from the countertop in front of you, not from the counter on your left.", + "D": "You brought it over from the countertop to your left.", + "E": "You carried it from the countertop behind you to your hands, rather than from the left side." + }, + "answer_label": "D", + "answer_text": "You brought it over from the countertop to your left.", + "q_id": "P06_108_5_T1", + "correct_answer": "You brought it over from the countertop to your left.", + "original_options": { + "A": "From the countertop to your right.", + "B": "From the countertop behind you.", + "C": "You brought it over from the countertop in front of you, not from the counter on your left.", + "D": "You brought it over from the countertop to your left.", + "E": "You carried it from the countertop behind you to your hands, rather than from the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 17 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_5_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did the white bowl in my hand come from?", + "answer": "You brought it over from the countertop to your back-left.", + "question_chinese": "我手中白色的碗是从哪里拿过来的？", + "answer_chinese": "从你左后方的台面上拿过来的。", + "evidence": [ + { + "description": "The bowl was taken from the countertop to the left of the sink.", + "timestamps": { + "start": "00:09:42", + "end": "00:09:44" + } + } + ], + "question_timestamps": "00:09:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You brought it over from the countertop to your back-left.", + "B": "From the countertop to your back-right.", + "C": "From the countertop to your front-left.", + "D": "You brought it over from the countertop to your right side, not from behind you on the left.", + "E": "You carried it from the countertop directly in front of you, rather than from the back-left area." + }, + "answer_label": "A", + "answer_text": "You brought it over from the countertop to your back-left.", + "q_id": "P06_108_5_T2", + "correct_answer": "You brought it over from the countertop to your back-left.", + "original_options": { + "A": "You brought it over from the countertop to your back-left.", + "B": "From the countertop to your back-right.", + "C": "From the countertop to your front-left.", + "D": "You brought it over from the countertop to your right side, not from behind you on the left.", + "E": "You carried it from the countertop directly in front of you, rather than from the back-left area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_10", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the white bowl in my hand from?", + "answer": "From the countertop to the left of the sink.", + "question_chinese": "我手中白色的碗是从哪里拿过来的？", + "answer_chinese": "从洗碗池左边的台面上拿过来的。", + "evidence": [ + { + "description": "The bowl was taken from the countertop to the left of the sink.", + "timestamps": { + "start": "00:09:42", + "end": "00:09:44" + } + } + ], + "question_timestamps": "00:10:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the right countertop.", + "B": "From a drawer near the sink.", + "C": "From the countertop to the left of the sink.", + "D": "From the table to the right of the sink, not from the left-side countertop.", + "E": "From the upper-right countertop corner by the sink, rather than the left-side countertop." + }, + "answer_label": "C", + "answer_text": "From the countertop to the left of the sink.", + "q_id": "P06_108_10", + "correct_answer": "From the countertop to the left of the sink.", + "original_options": { + "A": "From the right countertop.", + "B": "From a drawer near the sink.", + "C": "From the countertop to the left of the sink.", + "D": "From the table to the right of the sink, not from the left-side countertop.", + "E": "From the upper-right countertop corner by the sink, rather than the left-side countertop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 14, + 13 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_3_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "While I was heating the flatbread on the stovetop, how many times did I go to the sink?", + "answer": "You went to the sink once.", + "question_chinese": "我把面饼放在灶台加热的时候，我去了几次洗碗池？", + "answer_chinese": "你去了洗碗池一次。", + "evidence": [ + { + "description": "Placed the flatbread on the stove to heat.", + "timestamps": { + "start": "00:08:23", + "end": "00:08:35" + } + }, + { + "description": "Turned from the stove to the wooden cutting board next to it, then turned to the sink.", + "timestamps": { + "start": "00:08:51", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went to the sink twice.", + "B": "You didn't go to the sink.", + "C": "You went to the sink three times while heating the flatbread on the stovetop.", + "D": "You went to the sink four times, returning between checks on the flatbread as it heated.", + "E": "You went to the sink once." + }, + "answer_label": "E", + "answer_text": "You went to the sink once.", + "q_id": "P06_108_3_T1", + "correct_answer": "You went to the sink once.", + "original_options": { + "A": "You went to the sink twice.", + "B": "You didn't go to the sink.", + "C": "You went to the sink three times while heating the flatbread on the stovetop.", + "D": "You went to the sink four times, returning between checks on the flatbread as it heated.", + "E": "You went to the sink once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 16 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_5", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I was heating the flatbread on the stove, how many times did I go to the sink?", + "answer": "Twice.", + "question_chinese": "我把面饼放在灶台加热的时候，我去了几次洗碗池？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "Placed the flatbread on the stove to heat.", + "timestamps": { + "start": "00:08:23", + "end": "00:08:35" + } + }, + { + "description": "Turned from the stove to the wooden cutting board next to it, then turned to the sink.", + "timestamps": { + "start": "00:08:51", + "end": "00:09:34" + } + }, + { + "description": "Walked from the sink to the fridge to get something.", + "timestamps": { + "start": "00:09:35", + "end": "00:09:42" + } + }, + { + "description": "Returned from the fridge to the sink again.", + "timestamps": { + "start": "00:09:42", + "end": "00:10:42" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "Four times in total while the flatbread was heating on the stove.", + "E": "Five times, going back and forth between the stove and the sink during heating." + }, + "answer_label": "A", + "answer_text": "Twice.", + "q_id": "P06_108_5", + "correct_answer": "Twice.", + "original_options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "Four times in total while the flatbread was heating on the stove.", + "E": "Five times, going back and forth between the stove and the sink during heating." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 14 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_3_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "While I was heating the flatbread on the stovetop, how many times did I go to the sink?", + "answer": "You went to the sink twice.", + "question_chinese": "我把面饼放在灶台加热的时候，我去了几次洗碗池？", + "answer_chinese": "你去了洗碗池两次。", + "evidence": [ + { + "description": "Placed the flatbread on the stove to heat.", + "timestamps": { + "start": "00:08:23", + "end": "00:08:35" + } + }, + { + "description": "Turned from the stove to the wooden cutting board next to it, then turned to the sink.", + "timestamps": { + "start": "00:08:51", + "end": "00:09:34" + } + }, + { + "description": "Walked from the sink to the fridge to get something.", + "timestamps": { + "start": "00:09:35", + "end": "00:09:42" + } + }, + { + "description": "Returned from the fridge to the sink again.", + "timestamps": { + "start": "00:09:42", + "end": "00:10:42" + } + } + ], + "question_timestamps": "00:12:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went to the sink twice.", + "B": "You went to the sink once.", + "C": "You didn't go to the sink.", + "D": "You went to the sink three times while heating the flatbread on the stovetop.", + "E": "You went to the sink four times, repeatedly stepping over from the stove during heating." + }, + "answer_label": "A", + "answer_text": "You went to the sink twice.", + "q_id": "P06_108_3_T2", + "correct_answer": "You went to the sink twice.", + "original_options": { + "A": "You went to the sink twice.", + "B": "You went to the sink once.", + "C": "You didn't go to the sink.", + "D": "You went to the sink three times while heating the flatbread on the stovetop.", + "E": "You went to the sink four times, repeatedly stepping over from the stove during heating." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_7", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to go to the fridge to get something now, how should I go?", + "answer": "Turn around, walk straight past the sink and the countertop in order, and after going out the door, the fridge will be on your right.", + "question_chinese": "我现在想去冰箱拿东西，应该怎么走？", + "answer_chinese": "向后转，直走依次经过洗碗池和桌台，出门后冰箱在你的右侧。", + "evidence": [ + { + "description": "The fridge is on the left side near the door; after entering, the sink is on the left side of the view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + }, + { + "description": "Turn around from the stove, pass the sink and the countertop in order, you can reach the fridge on the right side after going out the door.", + "timestamps": { + "start": "00:08:54", + "end": "00:09:42" + } + } + ], + "question_timestamps": "00:10:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn around, walk straight past the sink and the countertop in order, and after going out the door, the fridge will be on your right.", + "B": "Turn right; the fridge is on your left.", + "C": "Turn around; pass the stove, then exit.", + "D": "Turn around, walk past the sink and the stove in order, and after the door the fridge is on your left.", + "E": "Turn around, walk straight past the sink and the countertop, and after leaving the door the fridge will be straight ahead." + }, + "answer_label": "A", + "answer_text": "Turn around, walk straight past the sink and the countertop in order, and after going out the door, the fridge will be on your right.", + "q_id": "P06_108_7", + "correct_answer": "Turn around, walk straight past the sink and the countertop in order, and after going out the door, the fridge will be on your right.", + "original_options": { + "A": "Turn around, walk straight past the sink and the countertop in order, and after going out the door, the fridge will be on your right.", + "B": "Turn right; the fridge is on your left.", + "C": "Turn around; pass the stove, then exit.", + "D": "Turn around, walk past the sink and the stove in order, and after the door the fridge is on your left.", + "E": "Turn around, walk straight past the sink and the countertop, and after leaving the door the fridge will be straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 21, + 21 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_8", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to go to the fridge to get something now, how should I go?", + "answer": "Turn left, walk straight out the door, and the fridge will be on your right.", + "question_chinese": "我现在想去冰箱拿东西，应该怎么走？", + "answer_chinese": "向左转，直走出门，冰箱在你的右侧。", + "evidence": [ + { + "description": "The fridge is on the left side near the door; after entering, the sink is on the left side of the view.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + }, + { + "description": "Turn around from the stove, pass the sink and the countertop in order, you can reach the fridge on the right side after going out the door.", + "timestamps": { + "start": "00:08:54", + "end": "00:09:42" + } + } + ], + "question_timestamps": "00:12:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right, walk out the door, fridge left.", + "B": "Turn left, go out the door, fridge ahead.", + "C": "Turn left, walk straight out the door, and the fridge will be on your left when you exit.", + "D": "Turn right first, then walk straight out the door, and the fridge will be on your right outside.", + "E": "Turn left, walk straight out the door, and the fridge will be on your right." + }, + "answer_label": "E", + "answer_text": "Turn left, walk straight out the door, and the fridge will be on your right.", + "q_id": "P06_108_8", + "correct_answer": "Turn left, walk straight out the door, and the fridge will be on your right.", + "original_options": { + "A": "Turn right, walk out the door, fridge left.", + "B": "Turn left, go out the door, fridge ahead.", + "C": "Turn left, walk straight out the door, and the fridge will be on your left when you exit.", + "D": "Turn right first, then walk straight out the door, and the fridge will be on your right outside.", + "E": "Turn left, walk straight out the door, and the fridge will be on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_11", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the stove now?", + "answer": "Yes.", + "question_chinese": "我现在能够得到灶台吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The small wooden cutting board is right next to the stove.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:05:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "P06_108_11", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_12", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the stove now?", + "answer": "No, there is at least a sink between you and the stove.", + "question_chinese": "我现在能够得到灶台吗？", + "answer_chinese": "不能，你和灶台之间至少隔了一个洗碗池。", + "evidence": [ + { + "description": "Turning around from the stove, you can reach the countertop with the dough by passing the sink.", + "timestamps": { + "start": "00:08:54", + "end": "00:09:37" + } + } + ], + "question_timestamps": "00:12:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, there is at least a sink between you and the stove.", + "B": "Yes, you can reach the stove from where you are now." + }, + "answer_label": "A", + "answer_text": "No, there is at least a sink between you and the stove.", + "q_id": "P06_108_12", + "correct_answer": "No, there is at least a sink between you and the stove.", + "original_options": { + "A": "No, there is at least a sink between you and the stove.", + "B": "Yes, you can reach the stove from where you are now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_8_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of unprocessed dough did I see on the countertop?", + "answer": "There were six pieces on the countertop in front of you.", + "question_chinese": "我看见台面上有几块未加工的面团？", + "answer_chinese": "你面前的台面上有6块。", + "evidence": [ + { + "description": "In the beginning, there were 4 pieces of dough on the countertop, and one piece of dough on each of the two plates.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There were five pieces.", + "B": "There were seven pieces.", + "C": "There were four pieces on the countertop directly in front of you.", + "D": "There were eight pieces laid out on the countertop in front of you.", + "E": "There were six pieces on the countertop in front of you." + }, + "answer_label": "E", + "answer_text": "There were six pieces on the countertop in front of you.", + "q_id": "P06_108_8_T1", + "correct_answer": "There were six pieces on the countertop in front of you.", + "original_options": { + "A": "There were five pieces.", + "B": "There were seven pieces.", + "C": "There were four pieces on the countertop directly in front of you.", + "D": "There were eight pieces laid out on the countertop in front of you.", + "E": "There were six pieces on the countertop in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_8_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of unprocessed dough did I see on the countertop?", + "answer": "You saw five pieces on the countertop.", + "question_chinese": "我看见台面上有几块未加工的面团？", + "answer_chinese": "你看见台面上有5块。", + "evidence": [ + { + "description": "In the beginning, there were 4 pieces of dough on the countertop, and one piece of dough on each of the two plates.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:25" + } + }, + { + "description": "You used one piece of dough from a plate.", + "timestamps": { + "start": "00:00:57", + "end": "00:05:30" + } + } + ], + "question_timestamps": "00:11:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw five pieces on the countertop.", + "B": "You saw four pieces.", + "C": "You saw six pieces.", + "D": "You saw three pieces of unprocessed dough on the countertop in front of you while working.", + "E": "You saw seven pieces sitting on the countertop where you were working." + }, + "answer_label": "A", + "answer_text": "You saw five pieces on the countertop.", + "q_id": "P06_108_8_T2", + "correct_answer": "You saw five pieces on the countertop.", + "original_options": { + "A": "You saw five pieces on the countertop.", + "B": "You saw four pieces.", + "C": "You saw six pieces.", + "D": "You saw three pieces of unprocessed dough on the countertop in front of you while working.", + "E": "You saw seven pieces sitting on the countertop where you were working." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 12 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_3", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the small wooden cutting board placed?", + "answer": "The small wooden cutting board was to the upper right of the sink.", + "question_chinese": "木质小案板放在了哪个位置？", + "answer_chinese": "木质小案板在洗碗台的右上方。", + "evidence": [ + { + "description": "The small wooden cutting board is to the upper right of the sink.", + "timestamps": { + "start": "00:11:41", + "end": "00:11:33" + } + } + ], + "question_timestamps": "00:11:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Upper left of the sink.", + "B": "Lower right of the sink.", + "C": "The small wooden cutting board was positioned directly in front of the sink on the counter.", + "D": "The small wooden cutting board was to the upper right of the sink.", + "E": "The small wooden cutting board was set to the lower left side of the sink area." + }, + "answer_label": "D", + "answer_text": "The small wooden cutting board was to the upper right of the sink.", + "q_id": "P06_108_3", + "correct_answer": "The small wooden cutting board was to the upper right of the sink.", + "original_options": { + "A": "Upper left of the sink.", + "B": "Lower right of the sink.", + "C": "The small wooden cutting board was positioned directly in front of the sink on the counter.", + "D": "The small wooden cutting board was to the upper right of the sink.", + "E": "The small wooden cutting board was set to the lower left side of the sink area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_4", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the small wooden cutting board placed?", + "answer": "The small wooden cutting board was to the upper right of the sink.", + "question_chinese": "木质小案板放在了哪个位置？", + "answer_chinese": "木质小案板在洗碗台的右上方。", + "evidence": [ + { + "description": "The small wooden cutting board is to the upper right of the sink.", + "timestamps": { + "start": "00:11:41", + "end": "00:11:33" + } + } + ], + "question_timestamps": "00:12:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Upper left of the sink.", + "B": "Lower right of the sink.", + "C": "The small wooden cutting board was to the upper right of the sink.", + "D": "The small wooden cutting board was placed directly behind the sink along the back edge of the counter.", + "E": "The small wooden cutting board was set on the kitchen island next to the sink rather than beside it." + }, + "answer_label": "C", + "answer_text": "The small wooden cutting board was to the upper right of the sink.", + "q_id": "P06_108_4", + "correct_answer": "The small wooden cutting board was to the upper right of the sink.", + "original_options": { + "A": "Upper left of the sink.", + "B": "Lower right of the sink.", + "C": "The small wooden cutting board was to the upper right of the sink.", + "D": "The small wooden cutting board was placed directly behind the sink along the back edge of the counter.", + "E": "The small wooden cutting board was set on the kitchen island next to the sink rather than beside it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 19 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_2_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I place the small wooden cutting board relative to me?", + "answer": "On the countertop on your left-hand side.", + "question_chinese": "木质小案板放在了我的哪个位置？", + "answer_chinese": "在你的左手边的台面上", + "evidence": [ + { + "description": "The small wooden cutting board is to the upper right of the sink.", + "timestamps": { + "start": "00:11:31", + "end": "00:11:33" + } + } + ], + "question_timestamps": "00:11:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the countertop to your right.", + "B": "On the countertop on your left-hand side.", + "C": "On the countertop behind you.", + "D": "On the countertop directly in front of you, within easy reach while you worked.", + "E": "On the countertop slightly behind and to your right-hand side while you were working." + }, + "answer_label": "B", + "answer_text": "On the countertop on your left-hand side.", + "q_id": "P06_108_2_T1", + "correct_answer": "On the countertop on your left-hand side.", + "original_options": { + "A": "On the countertop to your right.", + "B": "On the countertop on your left-hand side.", + "C": "On the countertop behind you.", + "D": "On the countertop directly in front of you, within easy reach while you worked.", + "E": "On the countertop slightly behind and to your right-hand side while you were working." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 14 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_2_T2", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I place the small wooden cutting board relative to me?", + "answer": "On the countertop in front of you.", + "question_chinese": "木质小案板放在了我的哪个位置？", + "answer_chinese": "在你面前的台面上", + "evidence": [ + { + "description": "The small wooden cutting board is to the upper right of the sink.", + "timestamps": { + "start": "00:11:31", + "end": "00:11:33" + } + } + ], + "question_timestamps": "00:12:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the countertop behind you.", + "B": "On the countertop to your left.", + "C": "On the countertop in front of you.", + "D": "On the countertop off to your right side, not positioned in front of you.", + "E": "On the countertop behind you, closer to the back edge rather than near your hands." + }, + "answer_label": "C", + "answer_text": "On the countertop in front of you.", + "q_id": "P06_108_2_T2", + "correct_answer": "On the countertop in front of you.", + "original_options": { + "A": "On the countertop behind you.", + "B": "On the countertop to your left.", + "C": "On the countertop in front of you.", + "D": "On the countertop off to your right side, not positioned in front of you.", + "E": "On the countertop behind you, closer to the back edge rather than near your hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 15 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_13", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many types of potted plants are on the windowsill in front of the sink?", + "answer": "four types.", + "question_chinese": "洗碗池前面的窗台上放了几种盆栽？", + "answer_chinese": "4种。", + "evidence": [ + { + "description": "There are 4 types of potted plants on the windowsill in front of the sink.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:21" + } + }, + { + "description": "There are 4 types of potted plants on the windowsill in front of the sink.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:18" + } + } + ], + "question_timestamps": "00:10:13", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three types.", + "B": "four types.", + "C": "Five types.", + "D": "There were two types of potted plants on the windowsill in front of the sink.", + "E": "There was one type of potted plant on the windowsill in front of the sink." + }, + "answer_label": "B", + "answer_text": "four types.", + "q_id": "P06_108_13", + "correct_answer": "four types.", + "original_options": { + "A": "Three types.", + "B": "four types.", + "C": "Five types.", + "D": "There were two types of potted plants on the windowsill in front of the sink.", + "E": "There was one type of potted plant on the windowsill in front of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_14", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many types of potted plants are on the windowsill in front of the sink?", + "answer": "Four types.", + "question_chinese": "洗碗池前面的窗台上放了几种盆栽？", + "answer_chinese": "4种。", + "evidence": [ + { + "description": "There are 4 types of potted plants on the windowsill in front of the sink.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:21" + } + }, + { + "description": "There are 4 types of potted plants on the windowsill in front of the sink.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:18" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three types.", + "B": "Four types.", + "C": "Two types.", + "D": "There were five types of potted plants on the windowsill in front of the sink.", + "E": "There was one type of potted plant on the windowsill in front of the sink." + }, + "answer_label": "B", + "answer_text": "Four types.", + "q_id": "P06_108_14", + "correct_answer": "Four types.", + "original_options": { + "A": "Three types.", + "B": "Four types.", + "C": "Two types.", + "D": "There were five types of potted plants on the windowsill in front of the sink.", + "E": "There was one type of potted plant on the windowsill in front of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_7_T1", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the potted plant relative to me?", + "answer": "The potted plant is above you over the sink on your left-hand side.", + "question_chinese": "盆栽在我的哪个方位？", + "answer_chinese": "盆栽在你左手边的洗碗池上方", + "evidence": [ + { + "description": "There are 4 types of potted plants on the windowsill in front of the sink.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above you on your right.", + "B": "Over the sink in front.", + "C": "The potted plant is below you, under the sink, on your left-hand side.", + "D": "The potted plant is above you over the sink, but it’s positioned behind you on the left.", + "E": "The potted plant is above you over the sink on your left-hand side." + }, + "answer_label": "E", + "answer_text": "The potted plant is above you over the sink on your left-hand side.", + "q_id": "P06_108_7_T1", + "correct_answer": "The potted plant is above you over the sink on your left-hand side.", + "original_options": { + "A": "Above you on your right.", + "B": "Over the sink in front.", + "C": "The potted plant is below you, under the sink, on your left-hand side.", + "D": "The potted plant is above you over the sink, but it’s positioned behind you on the left.", + "E": "The potted plant is above you over the sink on your left-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 18 + ] + }, + { + "video_uid": "P06_108", + "video_length": "00:12:08", + "qid": "P06_108_7_T2", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the potted plant relative to me?", + "answer": "On the windowsill in front of you.", + "question_chinese": "盆栽在我的哪个方位？", + "answer_chinese": "在你面前的窗台上", + "evidence": [ + { + "description": "There are 4 types of potted plants on the windowsill in front of the sink.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the windowsill behind you.", + "B": "On the windowsill to your left.", + "C": "The potted plant is on the windowsill to your right, not directly in front of you.", + "D": "The potted plant sits on the windowsill above you, rather than on the one in front.", + "E": "On the windowsill in front of you." + }, + "answer_label": "E", + "answer_text": "On the windowsill in front of you.", + "q_id": "P06_108_7_T2", + "correct_answer": "On the windowsill in front of you.", + "original_options": { + "A": "On the windowsill behind you.", + "B": "On the windowsill to your left.", + "C": "The potted plant is on the windowsill to your right, not directly in front of you.", + "D": "The potted plant sits on the windowsill above you, rather than on the one in front.", + "E": "On the windowsill in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + } +] diff --git a/QAs/P06_113-mcq.json b/QAs/P06_113-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b690972460db26024fb1ce4b5751204482d59b2e --- /dev/null +++ b/QAs/P06_113-mcq.json @@ -0,0 +1,934 @@ +[ + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_7_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the refrigerator from me?", + "answer": "The fridge is behind the door to your left hand.", + "question_chinese": "冰箱在我的哪个方位？", + "answer_chinese": "冰箱在你左手边的门后。", + "evidence": [ + { + "description": "The fridge is to the left of the counter.", + "timestamps": { + "start": "00:02:04", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind the right door.", + "B": "In front of the left door.", + "C": "The fridge is behind the door on your right hand, not on your left side.", + "D": "The fridge is in front of the door to your left hand, rather than tucked behind it.", + "E": "The fridge is behind the door to your left hand." + }, + "answer_label": "E", + "answer_text": "The fridge is behind the door to your left hand.", + "q_id": "P06_113_7_T1", + "correct_answer": "The fridge is behind the door to your left hand.", + "original_options": { + "A": "Behind the right door.", + "B": "In front of the left door.", + "C": "The fridge is behind the door on your right hand, not on your left side.", + "D": "The fridge is in front of the door to your left hand, rather than tucked behind it.", + "E": "The fridge is behind the door to your left hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 17 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_7_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the refrigerator from me?", + "answer": "The fridge is in front of you.", + "question_chinese": "冰箱在我的哪个方位？", + "answer_chinese": "冰箱在你的前方。", + "evidence": [ + { + "description": "The fridge is to the left of the counter.", + "timestamps": { + "start": "00:02:04", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:04:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is behind you.", + "B": "The fridge is in front of you.", + "C": "The fridge is to your left.", + "D": "The fridge is on your right side, so you would turn right to reach it.", + "E": "The fridge is behind you to the left, not directly in front of your viewpoint." + }, + "answer_label": "B", + "answer_text": "The fridge is in front of you.", + "q_id": "P06_113_7_T2", + "correct_answer": "The fridge is in front of you.", + "original_options": { + "A": "The fridge is behind you.", + "B": "The fridge is in front of you.", + "C": "The fridge is to your left.", + "D": "The fridge is on your right side, so you would turn right to reach it.", + "E": "The fridge is behind you to the left, not directly in front of your viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_2_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I just take out of the refrigerator?", + "answer": "Two items. They are both to your right hand right now.", + "question_chinese": "我刚刚从冰箱取出了几样东西？", + "answer_chinese": "两样物品。目前都在你的右手边。", + "evidence": [ + { + "description": "Took a pot and a food container out of the refrigerator.", + "timestamps": { + "start": "00:01:55", + "end": "00:02:04" + } + } + ], + "question_timestamps": "00:02:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two items. They are both to your right hand right now.", + "B": "One item to your right.", + "C": "Three items to your right.", + "D": "Zero items; you didn’t take anything out of the refrigerator just now.", + "E": "Four items, and they are all sitting to your right hand at the moment." + }, + "answer_label": "A", + "answer_text": "Two items. They are both to your right hand right now.", + "q_id": "P06_113_2_T2", + "correct_answer": "Two items. They are both to your right hand right now.", + "original_options": { + "A": "Two items. They are both to your right hand right now.", + "B": "One item to your right.", + "C": "Three items to your right.", + "D": "Zero items; you didn’t take anything out of the refrigerator just now.", + "E": "Four items, and they are all sitting to your right hand at the moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_2_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I just take out of the refrigerator?", + "answer": "Two items. They are both to your left hand right now.", + "question_chinese": "我刚刚从冰箱取出了几样东西？", + "answer_chinese": "两样。目前都在你的左手边。", + "evidence": [ + { + "description": "Took a pot and a food container out of the refrigerator.", + "timestamps": { + "start": "00:01:55", + "end": "00:02:04" + } + }, + { + "description": "Took a bag of food out of the refrigerator.", + "timestamps": { + "start": "00:02:04", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One item to your left.", + "B": "Three items to your left.", + "C": "Zero items; you didn’t remove anything from the refrigerator just now.", + "D": "Four items, and they are all positioned to your left hand right now.", + "E": "Two items. They are both to your left hand right now." + }, + "answer_label": "E", + "answer_text": "Two items. They are both to your left hand right now.", + "q_id": "P06_113_2_T1", + "correct_answer": "Two items. They are both to your left hand right now.", + "original_options": { + "A": "One item to your left.", + "B": "Three items to your left.", + "C": "Zero items; you didn’t remove anything from the refrigerator just now.", + "D": "Four items, and they are all positioned to your left hand right now.", + "E": "Two items. They are both to your left hand right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_1_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the frying pan from the sink shelf?", + "answer": "You put it in the lower cabinet.", + "question_chinese": "我刚刚把洗碗池架子上的平底锅放在哪里了？", + "answer_chinese": "放在了下面的柜子里。", + "evidence": [ + { + "description": "Put the frying pan from the sink rack into the cabinet below.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put it in the upper cabinet.", + "B": "You put it on the countertop.", + "C": "You set the frying pan on the table edge instead of placing it into a cabinet.", + "D": "You put it in the lower cabinet.", + "E": "You placed the frying pan back onto the sink shelf area rather than inside the lower cabinet." + }, + "answer_label": "D", + "answer_text": "You put it in the lower cabinet.", + "q_id": "P06_113_1_T1", + "correct_answer": "You put it in the lower cabinet.", + "original_options": { + "A": "You put it in the upper cabinet.", + "B": "You put it on the countertop.", + "C": "You set the frying pan on the table edge instead of placing it into a cabinet.", + "D": "You put it in the lower cabinet.", + "E": "You placed the frying pan back onto the sink shelf area rather than inside the lower cabinet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 17 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_1_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I just put the frying pan from the sink shelf?", + "answer": "It was put into the cabinet below to your left hand.", + "question_chinese": "我刚刚把洗碗池架子上的平底锅放在哪里了？", + "answer_chinese": "放在了你左手边的下面的柜子里。", + "evidence": [ + { + "description": "Put the frying pan from the sink rack into the cabinet below.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:01:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the cabinet below to your right.", + "B": "Into the cabinet above to your left.", + "C": "It was put into the cabinet below in front of you, not in the one to your left.", + "D": "You placed it into the cabinet above on your left side, rather than the lower cabinet.", + "E": "It was put into the cabinet below to your left hand." + }, + "answer_label": "E", + "answer_text": "It was put into the cabinet below to your left hand.", + "q_id": "P06_113_1_T2", + "correct_answer": "It was put into the cabinet below to your left hand.", + "original_options": { + "A": "Into the cabinet below to your right.", + "B": "Into the cabinet above to your left.", + "C": "It was put into the cabinet below in front of you, not in the one to your left.", + "D": "You placed it into the cabinet above on your left side, rather than the lower cabinet.", + "E": "It was put into the cabinet below to your left hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 16 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_5_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take while carrying this drink bottle?", + "answer": "You took a drink bottle out of the fridge, then walked to the countertop in front of you to your right, and finally walked to the sink in front of you to your right.", + "question_chinese": "我刚才拿着这个饮料瓶经过了什么路径？", + "answer_chinese": "你从冰箱取出饮料瓶，然后走到右边的台面前，最后走到右边的洗碗池前。", + "evidence": [ + { + "description": "Took the drink bottle from the fridge and walked to the counter on the right.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:22" + } + }, + { + "description": "Carried the drink bottle from the counter to the sink on the right.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:29" + } + } + ], + "question_timestamps": "00:04:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Fridge, sink, then countertop.", + "B": "Fridge, countertop left, sink left.", + "C": "You took a drink bottle out of the fridge, then walked to the countertop in front of you to your right, and finally walked to the sink in front of you to your right.", + "D": "You took the bottle from the fridge, walked to the countertop front-left, then went to the sink front-left.", + "E": "You took a drink bottle from the fridge, walked to a countertop behind-right, then went to the sink behind-right." + }, + "answer_label": "C", + "answer_text": "You took a drink bottle out of the fridge, then walked to the countertop in front of you to your right, and finally walked to the sink in front of you to your right.", + "q_id": "P06_113_5_T1", + "correct_answer": "You took a drink bottle out of the fridge, then walked to the countertop in front of you to your right, and finally walked to the sink in front of you to your right.", + "original_options": { + "A": "Fridge, sink, then countertop.", + "B": "Fridge, countertop left, sink left.", + "C": "You took a drink bottle out of the fridge, then walked to the countertop in front of you to your right, and finally walked to the sink in front of you to your right.", + "D": "You took the bottle from the fridge, walked to the countertop front-left, then went to the sink front-left.", + "E": "You took a drink bottle from the fridge, walked to a countertop behind-right, then went to the sink behind-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 19 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_5_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take while carrying this drink bottle?", + "answer": "You took a drink bottle out of the fridge, then walked to the countertop in front of you to your right, then to the sink in front of you to your right, and finally put it back into the fridge.", + "question_chinese": "我刚才拿着这个饮料瓶经过了什么路径？", + "answer_chinese": "你从冰箱取出饮料瓶，然后走到右边的台面前，最后走到右边的洗碗池前，最后将其放回冰箱里。", + "evidence": [ + { + "description": "Took the drink bottle from the fridge and walked to the counter on the right.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:22" + } + }, + { + "description": "Carried the drink bottle from the counter to the sink on the right.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:29" + } + } + ], + "question_timestamps": "00:04:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Fridge, sink, countertop, then fridge.", + "B": "Fridge, countertop, then back to fridge.", + "C": "You took the bottle from the fridge, went to the countertop front-right, returned to the fridge, then walked to the sink front-right.", + "D": "You took a drink bottle out of the fridge, then walked to the countertop in front of you to your right, then to the sink in front of you to your right, and finally put it back into the fridge.", + "E": "You took a drink bottle from the fridge, went sink front-right, then countertop front-right, and left it there." + }, + "answer_label": "D", + "answer_text": "You took a drink bottle out of the fridge, then walked to the countertop in front of you to your right, then to the sink in front of you to your right, and finally put it back into the fridge.", + "q_id": "P06_113_5_T2", + "correct_answer": "You took a drink bottle out of the fridge, then walked to the countertop in front of you to your right, then to the sink in front of you to your right, and finally put it back into the fridge.", + "original_options": { + "A": "Fridge, sink, countertop, then fridge.", + "B": "Fridge, countertop, then back to fridge.", + "C": "You took the bottle from the fridge, went to the countertop front-right, returned to the fridge, then walked to the sink front-right.", + "D": "You took a drink bottle out of the fridge, then walked to the countertop in front of you to your right, then to the sink in front of you to your right, and finally put it back into the fridge.", + "E": "You took a drink bottle from the fridge, went sink front-right, then countertop front-right, and left it there." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 22, + 18 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_4_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different foods did I put into the lunch box in total?", + "answer": "You put down two kinds of food in total, and they are both on the tabletop to your back-left now.", + "question_chinese": "我在饭盒里一共放了几样食物？", + "answer_chinese": "一共放了两样食物，目前都在你左后方的桌面上。", + "evidence": [ + { + "description": "Put some rice into the lunch box.", + "timestamps": { + "start": "00:02:17", + "end": "00:02:21" + } + }, + { + "description": "Put some salad dressing into the lunch box.", + "timestamps": { + "start": "00:03:01", + "end": "00:03:26" + } + } + ], + "question_timestamps": "00:03:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One kind of food.", + "B": "Three kinds of food.", + "C": "You put down two kinds of food in total, and they are both on the tabletop to your back-left now.", + "D": "You put down one kind of food in total, and it is on the tabletop to your back-left now.", + "E": "You put down four kinds of food in total, and they are all on the tabletop to your back-left now." + }, + "answer_label": "C", + "answer_text": "You put down two kinds of food in total, and they are both on the tabletop to your back-left now.", + "q_id": "P06_113_4_T2", + "correct_answer": "You put down two kinds of food in total, and they are both on the tabletop to your back-left now.", + "original_options": { + "A": "One kind of food.", + "B": "Three kinds of food.", + "C": "You put down two kinds of food in total, and they are both on the tabletop to your back-left now.", + "D": "You put down one kind of food in total, and it is on the tabletop to your back-left now.", + "E": "You put down four kinds of food in total, and they are all on the tabletop to your back-left now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 20 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_4_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different foods did I put into the lunch box in total?", + "answer": "You put down three kinds of food in total, all on the tabletop in the room to your right hand.", + "question_chinese": "我在饭盒里一共放了几样食物？", + "answer_chinese": "一共放了三样食物，都在你右手边房间里的桌面上。", + "evidence": [ + { + "description": "Put some rice into the lunch box.", + "timestamps": { + "start": "00:02:17", + "end": "00:02:21" + } + }, + { + "description": "Put some salad dressing into the lunch box.", + "timestamps": { + "start": "00:03:01", + "end": "00:03:26" + } + }, + { + "description": "Put a pancake into the lunch box and then closed the lid.", + "timestamps": { + "start": "00:03:30", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two kinds of food.", + "B": "You put down three kinds of food in total, all on the tabletop in the room to your right hand.", + "C": "Four kinds of food.", + "D": "You put down one kind of food in total, all on the tabletop in the room to your right hand.", + "E": "You put down five kinds of food in total, all on the tabletop in the room to your right hand." + }, + "answer_label": "B", + "answer_text": "You put down three kinds of food in total, all on the tabletop in the room to your right hand.", + "q_id": "P06_113_4_T1", + "correct_answer": "You put down three kinds of food in total, all on the tabletop in the room to your right hand.", + "original_options": { + "A": "Two kinds of food.", + "B": "You put down three kinds of food in total, all on the tabletop in the room to your right hand.", + "C": "Four kinds of food.", + "D": "You put down one kind of food in total, all on the tabletop in the room to your right hand.", + "E": "You put down five kinds of food in total, all on the tabletop in the room to your right hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 20 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_8_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potted plants did I see on the windowsill?", + "answer": "There are 4 potted plants on the windowsill, and they are to your left hand.", + "question_chinese": "我看到窗台上有几棵盆栽？", + "answer_chinese": "窗台上有4棵盆栽，它们在你的左手边。", + "evidence": [ + { + "description": "There are 4 potted plants on the windowsill.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 4 potted plants on the windowsill, and they are to your left hand.", + "B": "Five potted plants.", + "C": "Three potted plants.", + "D": "There are 2 potted plants on the windowsill, and they are to your left hand.", + "E": "There are 6 potted plants on the windowsill, and they are to your left hand." + }, + "answer_label": "A", + "answer_text": "There are 4 potted plants on the windowsill, and they are to your left hand.", + "q_id": "P06_113_8_T1", + "correct_answer": "There are 4 potted plants on the windowsill, and they are to your left hand.", + "original_options": { + "A": "There are 4 potted plants on the windowsill, and they are to your left hand.", + "B": "Five potted plants.", + "C": "Three potted plants.", + "D": "There are 2 potted plants on the windowsill, and they are to your left hand.", + "E": "There are 6 potted plants on the windowsill, and they are to your left hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_8_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potted plants did I see on the windowsill?", + "answer": "There are 4 potted plants on the windowsill; they are all to your front-right.", + "question_chinese": "我看到窗台上有几棵盆栽？", + "answer_chinese": "窗台上有4棵盆栽，他们都在你的右前方。", + "evidence": [ + { + "description": "There are 4 potted plants on the windowsill.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five potted plants.", + "B": "There are 4 potted plants on the windowsill; they are all to your front-right.", + "C": "Three potted plants.", + "D": "There are 2 potted plants on the windowsill; they are all to your front-right.", + "E": "There are 6 potted plants on the windowsill; they are all to your front-right." + }, + "answer_label": "B", + "answer_text": "There are 4 potted plants on the windowsill; they are all to your front-right.", + "q_id": "P06_113_8_T2", + "correct_answer": "There are 4 potted plants on the windowsill; they are all to your front-right.", + "original_options": { + "A": "Five potted plants.", + "B": "There are 4 potted plants on the windowsill; they are all to your front-right.", + "C": "Three potted plants.", + "D": "There are 2 potted plants on the windowsill; they are all to your front-right.", + "E": "There are 6 potted plants on the windowsill; they are all to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 14 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_6_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In the sink, how many spoons did I see?", + "answer": "There are 3 spoons in total, all to your right hand.", + "question_chinese": "洗碗池里，我看到有多少只勺子？", + "answer_chinese": "一共有3只勺子，都在你的右手边。", + "evidence": [ + { + "description": "Took 3 spoons from the cup on the dish rack and put them in the drawer on the left.", + "timestamps": { + "start": "00:00:53", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two spoons.", + "B": "Four spoons.", + "C": "There is 1 spoon in total, and it is on the right side of the sink.", + "D": "There are 3 spoons in total, all to your right hand.", + "E": "There are 5 spoons in total, and they are all on the right side of the sink." + }, + "answer_label": "D", + "answer_text": "There are 3 spoons in total, all to your right hand.", + "q_id": "P06_113_6_T1", + "correct_answer": "There are 3 spoons in total, all to your right hand.", + "original_options": { + "A": "Two spoons.", + "B": "Four spoons.", + "C": "There is 1 spoon in total, and it is on the right side of the sink.", + "D": "There are 3 spoons in total, all to your right hand.", + "E": "There are 5 spoons in total, and they are all on the right side of the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_6_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In the sink, how many spoons did I see?", + "answer": "There are 5 spoons in total, in the basket to your left hand.", + "question_chinese": "洗碗池里，我看到有多少只勺子？", + "answer_chinese": "一共有5只勺子，在你左手边的篮子里。", + "evidence": [ + { + "description": "Took 3 spoons from the cup on the dish rack and put them in the drawer on the left.", + "timestamps": { + "start": "00:00:53", + "end": "00:00:59" + } + }, + { + "description": "There are two spoons left.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:44" + } + } + ], + "question_timestamps": "00:02:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four spoons.", + "B": "Six spoons.", + "C": "There are 3 spoons in total, in the basket to your left hand.", + "D": "There are 7 spoons in total, in the basket to your left hand.", + "E": "There are 5 spoons in total, in the basket to your left hand." + }, + "answer_label": "E", + "answer_text": "There are 5 spoons in total, in the basket to your left hand.", + "q_id": "P06_113_6_T2", + "correct_answer": "There are 5 spoons in total, in the basket to your left hand.", + "original_options": { + "A": "Four spoons.", + "B": "Six spoons.", + "C": "There are 3 spoons in total, in the basket to your left hand.", + "D": "There are 7 spoons in total, in the basket to your left hand.", + "E": "There are 5 spoons in total, in the basket to your left hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_5", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the fridge now?", + "answer": "Three times.", + "question_chinese": "现在我一共打开了几次冰箱？", + "answer_chinese": "三次", + "evidence": [ + { + "description": "Opened the fridge for the first time.", + "timestamps": { + "start": "00:01:55", + "end": "00:02:13" + } + }, + { + "description": "Opened the fridge for the second time.", + "timestamps": { + "start": "00:03:50", + "end": "00:03:58" + } + }, + { + "description": "Opened the fridge for the third time.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:17" + } + } + ], + "question_timestamps": "00:04:48", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Twice.", + "C": "You opened the fridge two times up to now, counting each time you pulled the door open.", + "D": "You opened the fridge four times so far, including the most recent time you checked inside.", + "E": "Three times." + }, + "answer_label": "E", + "answer_text": "Three times.", + "q_id": "P06_113_5", + "correct_answer": "Three times.", + "original_options": { + "A": "Once.", + "B": "Twice.", + "C": "You opened the fridge two times up to now, counting each time you pulled the door open.", + "D": "You opened the fridge four times so far, including the most recent time you checked inside.", + "E": "Three times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 16 + ] + }, + { + "video_uid": "P06_113", + "video_length": "00:05:09", + "qid": "P06_113_6", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the fridge now?", + "answer": "Four times.", + "question_chinese": "现在我一共打开了几次冰箱？", + "answer_chinese": "四次。", + "evidence": [ + { + "description": "Opened the fridge for the first time.", + "timestamps": { + "start": "00:01:55", + "end": "00:02:13" + } + }, + { + "description": "Opened the fridge for the second time.", + "timestamps": { + "start": "00:03:50", + "end": "00:03:58" + } + }, + { + "description": "Opened the fridge for the third time.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:17" + } + }, + { + "description": "Opened the fridge for the fourth time.", + "timestamps": { + "start": "00:04:53", + "end": "00:04:57" + } + } + ], + "question_timestamps": "00:05:02", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three times.", + "B": "Five times.", + "C": "Four times.", + "D": "You opened the fridge two times in total so far, counting every completed open-and-close.", + "E": "You opened the fridge six times up to this point, including the last time you reached inside." + }, + "answer_label": "C", + "answer_text": "Four times.", + "q_id": "P06_113_6", + "correct_answer": "Four times.", + "original_options": { + "A": "Three times.", + "B": "Five times.", + "C": "Four times.", + "D": "You opened the fridge two times in total so far, counting every completed open-and-close.", + "E": "You opened the fridge six times up to this point, including the last time you reached inside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 17 + ] + } +] diff --git a/QAs/P07_104-mcq.json b/QAs/P07_104-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..df5920eb59872fda8b22066ec282db69cf0603b6 --- /dev/null +++ b/QAs/P07_104-mcq.json @@ -0,0 +1,1088 @@ +[ + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_4_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many plates did I take from the countertop above the dishwasher?", + "answer": "You picked up one plate.", + "question_chinese": "我从洗碗机上面的台面拿了几个盘子？", + "answer_chinese": "你拿了一个盘子。", + "evidence": [ + { + "description": "I took the first plate from the countertop above the dishwasher.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked up no plates.", + "B": "You picked up two plates.", + "C": "You picked up three plates from the countertop above the dishwasher before moving on.", + "D": "You picked up one plate.", + "E": "You picked up four plates from the countertop above the dishwasher and carried them away together." + }, + "answer_label": "D", + "answer_text": "You picked up one plate.", + "q_id": "P07_104_4_T2", + "correct_answer": "You picked up one plate.", + "original_options": { + "A": "You picked up no plates.", + "B": "You picked up two plates.", + "C": "You picked up three plates from the countertop above the dishwasher before moving on.", + "D": "You picked up one plate.", + "E": "You picked up four plates from the countertop above the dishwasher and carried them away together." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_4_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many plates did I take from the countertop above the dishwasher?", + "answer": "You picked up two plates.", + "question_chinese": "我从洗碗机上面的台面拿了几个盘子？", + "answer_chinese": "你拿了两个盘子。", + "evidence": [ + { + "description": "I took the first plate from the countertop above the dishwasher.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + }, + { + "description": "I took the second plate from the countertop above the dishwasher.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked up one plate.", + "B": "You picked up three plates.", + "C": "You picked up two plates.", + "D": "You picked up zero plates from the countertop above the dishwasher and left them there.", + "E": "You picked up four plates from the countertop above the dishwasher before placing anything elsewhere." + }, + "answer_label": "C", + "answer_text": "You picked up two plates.", + "q_id": "P07_104_4_T1", + "correct_answer": "You picked up two plates.", + "original_options": { + "A": "You picked up one plate.", + "B": "You picked up three plates.", + "C": "You picked up two plates.", + "D": "You picked up zero plates from the countertop above the dishwasher and left them there.", + "E": "You picked up four plates from the countertop above the dishwasher before placing anything elsewhere." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_5_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, how many plates did I put into the dishwasher in total?", + "answer": "You put one plate in, and the dishwasher was in front of you.", + "question_chinese": "我刚刚一共把几个盘子放进了洗碗机？", + "answer_chinese": "你放进去了一个盘子，洗碗机在你的前面", + "evidence": [ + { + "description": "Put the first plate into the dishwasher.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero plates; in front of you.", + "B": "Two plates; in front of you.", + "C": "You put three plates in, and the dishwasher was directly in front of you the whole time.", + "D": "You put four plates in, and the dishwasher stayed in front of you as you loaded them.", + "E": "You put one plate in, and the dishwasher was in front of you." + }, + "answer_label": "E", + "answer_text": "You put one plate in, and the dishwasher was in front of you.", + "q_id": "P07_104_5_T2", + "correct_answer": "You put one plate in, and the dishwasher was in front of you.", + "original_options": { + "A": "Zero plates; in front of you.", + "B": "Two plates; in front of you.", + "C": "You put three plates in, and the dishwasher was directly in front of you the whole time.", + "D": "You put four plates in, and the dishwasher stayed in front of you as you loaded them.", + "E": "You put one plate in, and the dishwasher was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_5_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, how many plates did I put into the dishwasher in total?", + "answer": "You put two plates in, and the dishwasher was to your front-right.", + "question_chinese": "我刚刚一共把几个盘子放进了洗碗机？", + "answer_chinese": "你放进去了两个盘子，洗碗机在你的右前方", + "evidence": [ + { + "description": "Put the first plate into the dishwasher.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:15" + } + }, + { + "description": "Put the second plate into the dishwasher.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One plate; front-right.", + "B": "Three plates; front-right.", + "C": "You put zero plates in, and the dishwasher was to your front-right when you approached it.", + "D": "You put two plates in, and the dishwasher was to your front-right.", + "E": "You put four plates in, and the dishwasher was positioned to your front-right as you loaded them." + }, + "answer_label": "D", + "answer_text": "You put two plates in, and the dishwasher was to your front-right.", + "q_id": "P07_104_5_T1", + "correct_answer": "You put two plates in, and the dishwasher was to your front-right.", + "original_options": { + "A": "One plate; front-right.", + "B": "Three plates; front-right.", + "C": "You put zero plates in, and the dishwasher was to your front-right when you approached it.", + "D": "You put two plates in, and the dishwasher was to your front-right.", + "E": "You put four plates in, and the dishwasher was positioned to your front-right as you loaded them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_2_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the stovetop right now?", + "answer": "You couldn't reach the stovetop.", + "question_chinese": "我现在可以够到灶台吗？", + "answer_chinese": "你不能够到灶台。", + "evidence": [ + { + "description": "The stove is about two meters away from the dishwasher.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "You are now standing in front of the dishwasher.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn't reach the stovetop.", + "B": "You could reach the stovetop." + }, + "answer_label": "A", + "answer_text": "You couldn't reach the stovetop.", + "q_id": "P07_104_2_T1", + "correct_answer": "You couldn't reach the stovetop.", + "original_options": { + "A": "You couldn't reach the stovetop.", + "B": "You could reach the stovetop." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_4", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the stove now?", + "answer": "No.", + "question_chinese": "我现在可以够到灶台吗？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The stove is about two meters away from the dishwasher.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "You are now standing in front of the dishwasher.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "P07_104_4", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_2_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the stovetop right now?", + "answer": "You could reach the stovetop.", + "question_chinese": "我现在可以够到灶台吗？", + "answer_chinese": "你可以够到灶台。", + "evidence": [ + { + "description": "The stove is about two meters away from the dishwasher.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "You are now standing in front of the dishwasher.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:35" + } + }, + { + "description": "I stood by the stove and could reach it by stretching out my hand.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn't reach the stovetop.", + "B": "You could reach the stovetop." + }, + "answer_label": "B", + "answer_text": "You could reach the stovetop.", + "q_id": "P07_104_2_T2", + "correct_answer": "You could reach the stovetop.", + "original_options": { + "A": "You couldn't reach the stovetop.", + "B": "You could reach the stovetop." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_6_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how far am I from the bananas on the table right now?", + "answer": "The horizontal distance was about half a meter.", + "question_chinese": "我现在的位置离桌上的香蕉大概有多远？", + "answer_chinese": "水平距离大概半米远。", + "evidence": [ + { + "description": "The bananas are on the shelf above the dishwasher.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "About one meter away.", + "B": "About two meters away.", + "C": "The bananas were roughly a meter away from you across the table.", + "D": "You were several steps back, putting the bananas at around two meters horizontally.", + "E": "The horizontal distance was about half a meter." + }, + "answer_label": "E", + "answer_text": "The horizontal distance was about half a meter.", + "q_id": "P07_104_6_T1", + "correct_answer": "The horizontal distance was about half a meter.", + "original_options": { + "A": "About one meter away.", + "B": "About two meters away.", + "C": "The bananas were roughly a meter away from you across the table.", + "D": "You were several steps back, putting the bananas at around two meters horizontally.", + "E": "The horizontal distance was about half a meter." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_14", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far am I from the bananas on the table now?", + "answer": "The horizontal distance is about half a meter.", + "question_chinese": "我现在的位置离桌上的香蕉大概有多远？", + "answer_chinese": "水平距离大概半米远。", + "evidence": [ + { + "description": "The bananas are on the shelf above the dishwasher.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "About one meter away.", + "B": "About three meters away.", + "C": "You are about a meter from the bananas, not close enough to touch them easily.", + "D": "You are a few steps away, making the horizontal distance to the bananas around three meters.", + "E": "The horizontal distance is about half a meter." + }, + "answer_label": "E", + "answer_text": "The horizontal distance is about half a meter.", + "q_id": "P07_104_14", + "correct_answer": "The horizontal distance is about half a meter.", + "original_options": { + "A": "About one meter away.", + "B": "About three meters away.", + "C": "You are about a meter from the bananas, not close enough to touch them easily.", + "D": "You are a few steps away, making the horizontal distance to the bananas around three meters.", + "E": "The horizontal distance is about half a meter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_6_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how far am I from the bananas on the table right now?", + "answer": "The horizontal distance was about two meters.", + "question_chinese": "我现在的位置离桌上的香蕉大概有多远？", + "answer_chinese": "水平距离大概两米远。", + "evidence": [ + { + "description": "The bananas are on the shelf above the dishwasher.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "I took three steps from the cabinet where the bananas were placed toward the oven.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The horizontal distance was about two meters.", + "B": "About one meter away.", + "C": "About three meters away.", + "D": "The bananas were only about a meter away from you horizontally at that moment.", + "E": "You were farther back, placing the bananas around three meters away across the table." + }, + "answer_label": "A", + "answer_text": "The horizontal distance was about two meters.", + "q_id": "P07_104_6_T2", + "correct_answer": "The horizontal distance was about two meters.", + "original_options": { + "A": "The horizontal distance was about two meters.", + "B": "About one meter away.", + "C": "About three meters away.", + "D": "The bananas were only about a meter away from you horizontally at that moment.", + "E": "You were farther back, placing the bananas around three meters away across the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_6", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory from entering the kitchen until now?", + "answer": "After entering the kitchen, you first walked to the front of the dishwasher, then to the cabinet on the left, then back to the dishwasher, and finally to the front of the stove.", + "question_chinese": "我进厨房到现在的运动轨迹是怎样的？", + "answer_chinese": "你进厨房后首先走到洗碗机前，然后走到左边的柜子，然后回到洗碗机，最后再走到灶台前。", + "evidence": [ + { + "description": "After entering the kitchen, walked to the front of the dishwasher and opened it.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:21" + } + }, + { + "description": "Walked to the cabinet to the left of the dishwasher and opened it.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:28" + } + }, + { + "description": "Returned to the front of the dishwasher.", + "timestamps": { + "start": "00:00:29", + "end": "00:01:30" + } + }, + { + "description": "Walked from the front of the dishwasher to the front of the stove.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:01:34", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "After entering the kitchen, you first walked to the front of the dishwasher, then to the cabinet on the left, then back to the dishwasher, and finally to the front of the stove.", + "B": "Dishwasher, stove, left cabinet, dishwasher.", + "C": "Left cabinet, dishwasher, left cabinet, stove.", + "D": "After entering, you went to the dishwasher, then to the stove, then to the left cabinet, ending near the dishwasher.", + "E": "You walked in, moved to the left cabinet first, then to the dishwasher, returned to the cabinet, and finished at the stove." + }, + "answer_label": "A", + "answer_text": "After entering the kitchen, you first walked to the front of the dishwasher, then to the cabinet on the left, then back to the dishwasher, and finally to the front of the stove.", + "q_id": "P07_104_6", + "correct_answer": "After entering the kitchen, you first walked to the front of the dishwasher, then to the cabinet on the left, then back to the dishwasher, and finally to the front of the stove.", + "original_options": { + "A": "After entering the kitchen, you first walked to the front of the dishwasher, then to the cabinet on the left, then back to the dishwasher, and finally to the front of the stove.", + "B": "Dishwasher, stove, left cabinet, dishwasher.", + "C": "Left cabinet, dishwasher, left cabinet, stove.", + "D": "After entering, you went to the dishwasher, then to the stove, then to the left cabinet, ending near the dishwasher.", + "E": "You walked in, moved to the left cabinet first, then to the dishwasher, returned to the cabinet, and finished at the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 22 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_5", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory from entering the kitchen until now?", + "answer": "After entering the kitchen, you first walk to the dishwasher, then to the cabinet on the left, then back to the dishwasher, then to the stove, and finally turn around and walk out of the kitchen.", + "question_chinese": "我进厨房到现在的运动轨迹是怎样的？", + "answer_chinese": "你进厨房后首先走到洗碗机前，然后走到左边的柜子，然后回到洗碗机，再走到灶台，最后转身走出厨房。", + "evidence": [ + { + "description": "After entering the kitchen, walked to the front of the dishwasher and opened it.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:21" + } + }, + { + "description": "Walked to the cabinet to the left of the dishwasher and opened it.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:28" + } + }, + { + "description": "Returned to the front of the dishwasher.", + "timestamps": { + "start": "00:00:29", + "end": "00:01:30" + } + }, + { + "description": "Walked from the front of the dishwasher to the front of the stove.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:34" + } + }, + { + "description": "Turned around and walked out of the kitchen.", + "timestamps": { + "start": "00:01:34", + "end": "00:01:36" + } + } + ], + "question_timestamps": "00:01:36", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You walk to the sink first.", + "B": "You go to the refrigerator first.", + "C": "After entering, you go to the sink, then the left cabinet, return to the sink, then to the stove, and exit.", + "D": "You walk to the dishwasher, then the right cabinet, back to the dishwasher, then to the stove, and walk out.", + "E": "After entering the kitchen, you first walk to the dishwasher, then to the cabinet on the left, then back to the dishwasher, then to the stove, and finally turn around and walk out of the kitchen." + }, + "answer_label": "E", + "answer_text": "After entering the kitchen, you first walk to the dishwasher, then to the cabinet on the left, then back to the dishwasher, then to the stove, and finally turn around and walk out of the kitchen.", + "q_id": "P07_104_5", + "correct_answer": "After entering the kitchen, you first walk to the dishwasher, then to the cabinet on the left, then back to the dishwasher, then to the stove, and finally turn around and walk out of the kitchen.", + "original_options": { + "A": "You walk to the sink first.", + "B": "You go to the refrigerator first.", + "C": "After entering, you go to the sink, then the left cabinet, return to the sink, then to the stove, and exit.", + "D": "You walk to the dishwasher, then the right cabinet, back to the dishwasher, then to the stove, and walk out.", + "E": "After entering the kitchen, you first walk to the dishwasher, then to the cabinet on the left, then back to the dishwasher, then to the stove, and finally turn around and walk out of the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 20 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red patterned towel?", + "answer": "The red patterned towel is on the handle of the oven.", + "question_chinese": "红色的印花毛巾放在哪里了？", + "answer_chinese": "红色印花毛巾放在烤箱的把手上。", + "evidence": [ + { + "description": "The red patterned towel is on the handle of the oven.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the countertop by the sink.", + "B": "Hanging from a cabinet handle.", + "C": "The red patterned towel is draped over the back of a chair near the table.", + "D": "The red patterned towel is on the handle of the oven.", + "E": "It’s hanging on a nearby cabinet door handle rather than on the oven." + }, + "answer_label": "D", + "answer_text": "The red patterned towel is on the handle of the oven.", + "q_id": "P07_104_1", + "correct_answer": "The red patterned towel is on the handle of the oven.", + "original_options": { + "A": "On the countertop by the sink.", + "B": "Hanging from a cabinet handle.", + "C": "The red patterned towel is draped over the back of a chair near the table.", + "D": "The red patterned towel is on the handle of the oven.", + "E": "It’s hanging on a nearby cabinet door handle rather than on the oven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 14 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red patterned towel?", + "answer": "The red patterned towel is on the handle of the oven.", + "question_chinese": "红色的印花毛巾放在哪里了？", + "answer_chinese": "红色印花毛巾放在烤箱的把手上。", + "evidence": [ + { + "description": "The red patterned towel is on the handle of the oven.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:01:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the kitchen counter.", + "B": "The red patterned towel is on the handle of the oven.", + "C": "On a cabinet door handle.", + "D": "The red patterned towel is hanging near the sink area, not on the oven.", + "E": "It’s draped over a chair by the table instead of being on the oven handle." + }, + "answer_label": "B", + "answer_text": "The red patterned towel is on the handle of the oven.", + "q_id": "P07_104_2", + "correct_answer": "The red patterned towel is on the handle of the oven.", + "original_options": { + "A": "On the kitchen counter.", + "B": "The red patterned towel is on the handle of the oven.", + "C": "On a cabinet door handle.", + "D": "The red patterned towel is hanging near the sink area, not on the oven.", + "E": "It’s draped over a chair by the table instead of being on the oven handle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 16 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_1_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red patterned towel relative to me?", + "answer": "The red printed towel was on the oven handle to your left.", + "question_chinese": "红色的印花毛巾放在相对于我的什么位置？", + "answer_chinese": "红色印花毛巾放在你左边的烤箱的把手上。", + "evidence": [ + { + "description": "The red patterned towel is on the handle of the oven.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the oven handle in front of you.", + "B": "The red printed towel was on the oven handle to your left.", + "C": "On the oven handle to your right.", + "D": "The red printed towel was on the oven handle behind you, not on the side.", + "E": "The red printed towel hung on the oven handle above your line of sight, rather than to your left." + }, + "answer_label": "B", + "answer_text": "The red printed towel was on the oven handle to your left.", + "q_id": "P07_104_1_T1", + "correct_answer": "The red printed towel was on the oven handle to your left.", + "original_options": { + "A": "On the oven handle in front of you.", + "B": "The red printed towel was on the oven handle to your left.", + "C": "On the oven handle to your right.", + "D": "The red printed towel was on the oven handle behind you, not on the side.", + "E": "The red printed towel hung on the oven handle above your line of sight, rather than to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 19 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_1_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red patterned towel relative to me?", + "answer": "The red printed towel was on the oven handle in front of you.", + "question_chinese": "红色的印花毛巾放在相对于我的什么位置？", + "answer_chinese": "红色印花毛巾放在你前面的烤箱的把手上。", + "evidence": [ + { + "description": "The red patterned towel is on the handle of the oven.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "The oven is in front of us, and a red patterned towel is hanging on the oven handle.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the oven handle to your left.", + "B": "On the oven handle behind you.", + "C": "The red printed towel was on the oven handle to your right, not directly in front.", + "D": "The red printed towel was on the oven handle in front of you.", + "E": "The red printed towel was on the oven handle above you, rather than in front of you." + }, + "answer_label": "D", + "answer_text": "The red printed towel was on the oven handle in front of you.", + "q_id": "P07_104_1_T2", + "correct_answer": "The red printed towel was on the oven handle in front of you.", + "original_options": { + "A": "On the oven handle to your left.", + "B": "On the oven handle behind you.", + "C": "The red printed towel was on the oven handle to your right, not directly in front.", + "D": "The red printed towel was on the oven handle in front of you.", + "E": "The red printed towel was on the oven handle above you, rather than in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 17 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_7_T1", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the bananas or the patterned towel?", + "answer": "The banana was closer to you.", + "question_chinese": "香蕉和印花毛巾哪个离我更近？", + "answer_chinese": "香蕉离你更近。", + "evidence": [ + { + "description": "The bananas are on the shelf above the dishwasher.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "The red patterned towel is on the handle of the oven, about two meters to the left of the sink.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:00:34", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The banana was closer to you.", + "B": "The printed towel was closer to you." + }, + "answer_label": "A", + "answer_text": "The banana was closer to you.", + "q_id": "P07_104_7_T1", + "correct_answer": "The banana was closer to you.", + "original_options": { + "A": "The banana was closer to you.", + "B": "The printed towel was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_16", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the banana or the patterned towel?", + "answer": "The banana is closer to you.", + "question_chinese": "香蕉和印花毛巾哪个离我更近？", + "answer_chinese": "香蕉离你更近。", + "evidence": [ + { + "description": "The bananas are on the shelf above the dishwasher.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "The red patterned towel is on the handle of the oven, about two meters to the left of the sink.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:01:01", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The banana is closer to you.", + "B": "The patterned towel is closer to you." + }, + "answer_label": "A", + "answer_text": "The banana is closer to you.", + "q_id": "P07_104_16", + "correct_answer": "The banana is closer to you.", + "original_options": { + "A": "The banana is closer to you.", + "B": "The patterned towel is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P07_104", + "video_length": "00:01:39", + "qid": "P07_104_7_T2", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the bananas or the patterned towel?", + "answer": "The printed towel was closer to you.", + "question_chinese": "香蕉和印花毛巾哪个离我更近？", + "answer_chinese": "印花毛巾离你更近。", + "evidence": [ + { + "description": "The bananas are on the shelf above the dishwasher.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "The red patterned towel is on the handle of the oven, about two meters to the left of the sink.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:27" + } + }, + { + "description": "You are next to the oven, with the patterned towel on the oven handle, and you are an arm's length away from the patterned towel. The banana is two meters away from you.", + "timestamps": { + "start": "00:01:25", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The printed towel was closer to you.", + "B": "The banana was closer to you." + }, + "answer_label": "A", + "answer_text": "The printed towel was closer to you.", + "q_id": "P07_104_7_T2", + "correct_answer": "The printed towel was closer to you.", + "original_options": { + "A": "The printed towel was closer to you.", + "B": "The banana was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/P07_108-mcq.json b/QAs/P07_108-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..dad91781e5321cd2753ef5bc9143ce1b3d7cabc6 --- /dev/null +++ b/QAs/P07_108-mcq.json @@ -0,0 +1,1424 @@ +[ + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_3_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I first entered the kitchen, what path did I take while carrying the frying pan?", + "answer": "You walked from the stove to the sink to your back-right.", + "question_chinese": "刚进厨房时，我拿着平底锅经过了什么路径？", + "answer_chinese": "你从灶台走向了右后方的洗碗池", + "evidence": [ + { + "description": "Picked up the frying pan from the stove and placed it in the sink to the back right.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went from the sink to the stove.", + "B": "You walked from stove to front-right sink.", + "C": "You carried the pan from the stove to the sink on your back-left side instead of back-right.", + "D": "You walked from the stove to the sink positioned front-left of you, rather than moving toward your back-right.", + "E": "You walked from the stove to the sink to your back-right." + }, + "answer_label": "E", + "answer_text": "You walked from the stove to the sink to your back-right.", + "q_id": "P07_108_3_T1", + "correct_answer": "You walked from the stove to the sink to your back-right.", + "original_options": { + "A": "You went from the sink to the stove.", + "B": "You walked from stove to front-right sink.", + "C": "You carried the pan from the stove to the sink on your back-left side instead of back-right.", + "D": "You walked from the stove to the sink positioned front-left of you, rather than moving toward your back-right.", + "E": "You walked from the stove to the sink to your back-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 18 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_5", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I first entered the kitchen, what path did I take while holding the frying pan?", + "answer": "From the stove to the sink in the back right, and then to the dishwasher in the bottom right.", + "question_chinese": "刚进厨房时，我拿着平底锅经过了什么路径？", + "answer_chinese": "从灶台走向了右后方的洗碗池，然后走到右下方的洗碗机。", + "evidence": [ + { + "description": "Picked up the frying pan from the stove and placed it in the sink to the back right.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:10" + } + }, + { + "description": "Put the frying pan from the sink into the dishwasher to the bottom right.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:02:35", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Stove to sink, back-left.", + "B": "Stove to sink, front-right.", + "C": "From the stove to the sink in the back right, then to the dishwasher on your left side.", + "D": "From the stove to the sink in the back right, and then you went to the counter instead of the dishwasher.", + "E": "From the stove to the sink in the back right, and then to the dishwasher in the bottom right." + }, + "answer_label": "E", + "answer_text": "From the stove to the sink in the back right, and then to the dishwasher in the bottom right.", + "q_id": "P07_108_5", + "correct_answer": "From the stove to the sink in the back right, and then to the dishwasher in the bottom right.", + "original_options": { + "A": "Stove to sink, back-left.", + "B": "Stove to sink, front-right.", + "C": "From the stove to the sink in the back right, then to the dishwasher on your left side.", + "D": "From the stove to the sink in the back right, and then you went to the counter instead of the dishwasher.", + "E": "From the stove to the sink in the back right, and then to the dishwasher in the bottom right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 21 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_3_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I first entered the kitchen, what path did I take while carrying the frying pan?", + "answer": "You walked from the stove to the sink to your back-right, then walked to the dishwasher below you on the right.", + "question_chinese": "刚进厨房时，我拿着平底锅经过了什么路径？", + "answer_chinese": "从灶台走向了右后方的洗碗池，然后走到右下方的洗碗机。", + "evidence": [ + { + "description": "Picked up the frying pan from the stove and placed it in the sink to the back right.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:10" + } + }, + { + "description": "Put the frying pan from the sink into the dishwasher to the bottom right.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:05:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Then walked to the dishwasher in front-right.", + "B": "Then walked to the dishwasher behind-right.", + "C": "You walked from the stove to the sink to your back-right, then walked to the dishwasher to your left.", + "D": "You walked from the stove to the sink to your back-right, then walked to the dishwasher higher up on the right.", + "E": "You walked from the stove to the sink to your back-right, then walked to the dishwasher below you on the right." + }, + "answer_label": "E", + "answer_text": "You walked from the stove to the sink to your back-right, then walked to the dishwasher below you on the right.", + "q_id": "P07_108_3_T2", + "correct_answer": "You walked from the stove to the sink to your back-right, then walked to the dishwasher below you on the right.", + "original_options": { + "A": "Then walked to the dishwasher in front-right.", + "B": "Then walked to the dishwasher behind-right.", + "C": "You walked from the stove to the sink to your back-right, then walked to the dishwasher to your left.", + "D": "You walked from the stove to the sink to your back-right, then walked to the dishwasher higher up on the right.", + "E": "You walked from the stove to the sink to your back-right, then walked to the dishwasher below you on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 21 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_2_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the small trash can on the countertop relative to me?", + "answer": "Behind you.", + "question_chinese": "台面上的小垃圾桶在我的哪个方向？", + "answer_chinese": "在你的后面", + "evidence": [ + { + "description": "The small trash can is to the left of the sink.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:02:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "It’s slightly to your right on the countertop, not behind you where you were standing.", + "D": "Behind you.", + "E": "It’s positioned in front of you near the sink area, rather than behind your body on the counter." + }, + "answer_label": "D", + "answer_text": "Behind you.", + "q_id": "P07_108_2_T1", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "It’s slightly to your right on the countertop, not behind you where you were standing.", + "D": "Behind you.", + "E": "It’s positioned in front of you near the sink area, rather than behind your body on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 19 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_2_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the small trash can on the countertop relative to me?", + "answer": "To your right.", + "question_chinese": "台面上的小垃圾桶在我的哪个方向？", + "answer_chinese": "在你的右边", + "evidence": [ + { + "description": "The small trash can is to the left of the sink.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:04:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "Behind you.", + "C": "It’s directly in front of you on the counter, not off to your right side.", + "D": "It’s located behind you near the back edge of the countertop, rather than to your right.", + "E": "To your right." + }, + "answer_label": "E", + "answer_text": "To your right.", + "q_id": "P07_108_2_T2", + "correct_answer": "To your right.", + "original_options": { + "A": "To your left.", + "B": "Behind you.", + "C": "It’s directly in front of you on the counter, not off to your right side.", + "D": "It’s located behind you near the back edge of the countertop, rather than to your right.", + "E": "To your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 16, + 17 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_3", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the small trash can on the countertop from the sink?", + "answer": "The small trash can is to the left of the sink.", + "question_chinese": "台面上的小垃圾桶在洗碗池的哪个方向？", + "answer_chinese": "小垃圾桶在洗碗池的左边。", + "evidence": [ + { + "description": "The small trash can is to the left of the sink.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The small trash can is to the right.", + "B": "The small trash can is to the left of the sink.", + "C": "The small trash can is in front.", + "D": "The small trash can is positioned to the right of the sink, on the opposite side from left.", + "E": "The small trash can sits behind the sink area on the countertop, not to the sink’s left." + }, + "answer_label": "B", + "answer_text": "The small trash can is to the left of the sink.", + "q_id": "P07_108_3", + "correct_answer": "The small trash can is to the left of the sink.", + "original_options": { + "A": "The small trash can is to the right.", + "B": "The small trash can is to the left of the sink.", + "C": "The small trash can is in front.", + "D": "The small trash can is positioned to the right of the sink, on the opposite side from left.", + "E": "The small trash can sits behind the sink area on the countertop, not to the sink’s left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 18 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_4", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the small trash can on the countertop from the sink?", + "answer": "The small trash can is to the left of the sink.", + "question_chinese": "台面上的小垃圾桶在洗碗池的哪个方向？", + "answer_chinese": "小垃圾桶在洗碗池的左边。", + "evidence": [ + { + "description": "The small trash can is to the left of the sink.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The small trash can is to the right.", + "B": "The small trash can is above the sink.", + "C": "The small trash can is in front of the sink on the counter, rather than on the left side.", + "D": "The small trash can is to the left of the sink.", + "E": "The small trash can is behind the sink area on the countertop, not positioned to the left." + }, + "answer_label": "D", + "answer_text": "The small trash can is to the left of the sink.", + "q_id": "P07_108_4", + "correct_answer": "The small trash can is to the left of the sink.", + "original_options": { + "A": "The small trash can is to the right.", + "B": "The small trash can is above the sink.", + "C": "The small trash can is in front of the sink on the counter, rather than on the left side.", + "D": "The small trash can is to the left of the sink.", + "E": "The small trash can is behind the sink area on the countertop, not positioned to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 17 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_9", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I am facing the stove and cooking, which is closer to me, the cabinet or the sink?", + "answer": "The cabinet.", + "question_chinese": "当我面向灶台做饭时，橱柜和洗碗池哪个离我更近？", + "answer_chinese": "橱柜。", + "evidence": [ + { + "description": "The sink is about one meter to the back right of the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + }, + { + "description": "The cabinet is about 50 cm to the top left of the stove.", + "timestamps": { + "start": "00:03:57", + "end": "00:04:02" + } + } + ], + "question_timestamps": "00:04:15", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The sink.", + "B": "The cabinet." + }, + "answer_label": "B", + "answer_text": "The cabinet.", + "q_id": "P07_108_9", + "correct_answer": "The cabinet.", + "original_options": { + "A": "The sink.", + "B": "The cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_10", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I am facing the stove and cooking, which is closer to me, the cabinet or the sink?", + "answer": "The cabinet.", + "question_chinese": "当我面向灶台做饭时，橱柜和洗碗池哪个离我更近？", + "answer_chinese": "橱柜。", + "evidence": [ + { + "description": "The sink is about one meter to the back right of the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + }, + { + "description": "The cabinet is about 50 cm to the top left of the stove.", + "timestamps": { + "start": "00:03:57", + "end": "00:04:02" + } + } + ], + "question_timestamps": "00:05:06", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The cabinet.", + "B": "The sink." + }, + "answer_label": "A", + "answer_text": "The cabinet.", + "q_id": "P07_108_10", + "correct_answer": "The cabinet.", + "original_options": { + "A": "The cabinet.", + "B": "The sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first entered the kitchen, in which direction was the stove?", + "answer": "When you first entered the kitchen, the stove was about three meters directly in front of you.", + "question_chinese": "我刚进厨房时，灶台在我的哪个方向？", + "answer_chinese": "刚进厨房时，灶台在视野正前方约3米远的位置。", + "evidence": [ + { + "description": "When first entering the kitchen, the stove is in the center of the field of view, about 3 meters away.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "When you first entered the kitchen, the stove was about three meters directly in front of you.", + "B": "Directly behind you.", + "C": "Directly to your left.", + "D": "When you first entered the kitchen, the stove was about three meters directly to your right instead of ahead.", + "E": "When you first entered the kitchen, the stove was about three meters directly behind you, not in front." + }, + "answer_label": "A", + "answer_text": "When you first entered the kitchen, the stove was about three meters directly in front of you.", + "q_id": "P07_108_1", + "correct_answer": "When you first entered the kitchen, the stove was about three meters directly in front of you.", + "original_options": { + "A": "When you first entered the kitchen, the stove was about three meters directly in front of you.", + "B": "Directly behind you.", + "C": "Directly to your left.", + "D": "When you first entered the kitchen, the stove was about three meters directly to your right instead of ahead.", + "E": "When you first entered the kitchen, the stove was about three meters directly behind you, not in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 19, + 18 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first entered the kitchen, in which direction was the stove?", + "answer": "When you first entered the kitchen, the stove was about three meters directly in front of you.", + "question_chinese": "我刚进厨房时，灶台在我的哪个方向？", + "answer_chinese": "刚进厨房时，灶台在视野正前方约3米远的位置。", + "evidence": [ + { + "description": "When first entering the kitchen, the stove is in the center of the field of view, about 3 meters away.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Directly behind you.", + "B": "Directly to your right.", + "C": "When you first entered the kitchen, the stove was about three meters directly to your left rather than straight ahead.", + "D": "When you first entered the kitchen, the stove was about three meters directly in front of you.", + "E": "When you first entered the kitchen, the stove was about three meters above you in that direction, not in front." + }, + "answer_label": "D", + "answer_text": "When you first entered the kitchen, the stove was about three meters directly in front of you.", + "q_id": "P07_108_2", + "correct_answer": "When you first entered the kitchen, the stove was about three meters directly in front of you.", + "original_options": { + "A": "Directly behind you.", + "B": "Directly to your right.", + "C": "When you first entered the kitchen, the stove was about three meters directly to your left rather than straight ahead.", + "D": "When you first entered the kitchen, the stove was about three meters directly in front of you.", + "E": "When you first entered the kitchen, the stove was about three meters above you in that direction, not in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 20, + 20 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_7_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "While I was cooking, how many different things did I put into the pot?", + "answer": "You put in two types.", + "question_chinese": "我在做饭的过程中，往锅里放了几样东西？", + "answer_chinese": "你放了两种。", + "evidence": [ + { + "description": "Poured oil into the pot.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:41" + } + }, + { + "description": "Poured beans into the pot.", + "timestamps": { + "start": "00:02:06", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You put in one type.", + "B": "You put in three types.", + "C": "You put in four different types while cooking, not just a couple of ingredients.", + "D": "You only put in one type overall, so there weren’t multiple different things added.", + "E": "You put in two types." + }, + "answer_label": "E", + "answer_text": "You put in two types.", + "q_id": "P07_108_7_T1", + "correct_answer": "You put in two types.", + "original_options": { + "A": "You put in one type.", + "B": "You put in three types.", + "C": "You put in four different types while cooking, not just a couple of ingredients.", + "D": "You only put in one type overall, so there weren’t multiple different things added.", + "E": "You put in two types." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_13", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many things did I put in the pot while I was cooking?", + "answer": "three types.", + "question_chinese": "我在做饭的过程中，往锅里放了几样东西？", + "answer_chinese": "3种。", + "evidence": [ + { + "description": "Poured oil into the pot.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:41" + } + }, + { + "description": "Poured beans into the pot.", + "timestamps": { + "start": "00:01:19", + "end": "00:02:20" + } + }, + { + "description": "Poured another kind of beans into the pot.", + "timestamps": { + "start": "00:03:31", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One type.", + "B": "three types.", + "C": "Four types.", + "D": "You put in two types total, not as many as three different things.", + "E": "You added five types into the pot while cooking, which is more than you actually used." + }, + "answer_label": "B", + "answer_text": "three types.", + "q_id": "P07_108_13", + "correct_answer": "three types.", + "original_options": { + "A": "One type.", + "B": "three types.", + "C": "Four types.", + "D": "You put in two types total, not as many as three different things.", + "E": "You added five types into the pot while cooking, which is more than you actually used." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 16 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_7_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "While I was cooking, how many different things did I put into the pot?", + "answer": "You put in three types.", + "question_chinese": "我在做饭的过程中，往锅里放了几样东西？", + "answer_chinese": "你放了3种。", + "evidence": [ + { + "description": "Poured oil into the pot.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:41" + } + }, + { + "description": "Poured beans into the pot.", + "timestamps": { + "start": "00:02:06", + "end": "00:02:20" + } + }, + { + "description": "Poured another kind of beans into the pot.", + "timestamps": { + "start": "00:03:21", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You put in three types.", + "B": "You put in two types.", + "C": "You put in four types.", + "D": "You put in one type overall, so there weren’t several different things added to the pot.", + "E": "You put in five different types while cooking, which is more than what actually went in." + }, + "answer_label": "A", + "answer_text": "You put in three types.", + "q_id": "P07_108_7_T2", + "correct_answer": "You put in three types.", + "original_options": { + "A": "You put in three types.", + "B": "You put in two types.", + "C": "You put in four types.", + "D": "You put in one type overall, so there weren’t several different things added to the pot.", + "E": "You put in five different types while cooking, which is more than what actually went in." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_11", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the stove from my current position?", + "answer": "No.", + "question_chinese": "我现在这个位置可以够得到灶台吗？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The stove is about one meter to the left of the sink.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:09" + } + }, + { + "description": "The dishwasher is to the bottom right of the sink.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "P07_108_11", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_6_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach the stovetop?", + "answer": "Yes, you can.", + "question_chinese": "我现在这个位置可以够得到灶台吗？", + "answer_chinese": "能。", + "evidence": [ + { + "description": "The stove is about one meter to the left of the sink.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:09" + } + }, + { + "description": "The dishwasher is to the bottom right of the sink.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "answer_label": "A", + "answer_text": "Yes, you can.", + "q_id": "P07_108_6_T1", + "correct_answer": "Yes, you can.", + "original_options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_6_T2", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach the stovetop?", + "answer": "No, you can't.", + "question_chinese": "我现在这个位置可以够得到灶台吗？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The stove is about one meter to the left of the sink.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:09" + } + }, + { + "description": "The dishwasher is to the bottom right of the sink.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:05:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "answer_label": "B", + "answer_text": "No, you can't.", + "q_id": "P07_108_6_T2", + "correct_answer": "No, you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_4_T1", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What should I do to put the pot on the stovetop that is closer to me back to its original place?", + "answer": "You put the pot into the cabinet below you to the right of the stove.", + "question_chinese": "我要把灶台上离我更近的锅放回原位置该怎么办？", + "answer_chinese": "把锅放到灶台右下方的柜子里", + "evidence": [ + { + "description": "Took the stainless steel frying pan out of the cabinet to the bottom right of the stove and placed it on the stove.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:02:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it in the cabinet in front.", + "B": "Put it in the cabinet behind you.", + "C": "Put the pot back into the cabinet to your left of the stove, not below to the right.", + "D": "You put the pot into the cabinet below you to the right of the stove.", + "E": "Return the pot to the cabinet on your right side near the stove, rather than below you." + }, + "answer_label": "D", + "answer_text": "You put the pot into the cabinet below you to the right of the stove.", + "q_id": "P07_108_4_T1", + "correct_answer": "You put the pot into the cabinet below you to the right of the stove.", + "original_options": { + "A": "Put it in the cabinet in front.", + "B": "Put it in the cabinet behind you.", + "C": "Put the pot back into the cabinet to your left of the stove, not below to the right.", + "D": "You put the pot into the cabinet below you to the right of the stove.", + "E": "Return the pot to the cabinet on your right side near the stove, rather than below you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 17 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_4_T2", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What should I do to put the pot on the stovetop that is closer to me back to its original place?", + "answer": "You put the pot into the cabinet directly below you.", + "question_chinese": "我要把灶台上离我更近的锅放回原位置该怎么办？", + "answer_chinese": "把锅放到正下方的柜子里", + "evidence": [ + { + "description": "Took the stainless steel frying pan out of the cabinet to the bottom right of the stove and placed it on the stove.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:04:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it in the cabinet in front.", + "B": "You put the pot into the cabinet directly below you.", + "C": "Put it in the cabinet behind you.", + "D": "Put the pot into the cabinet directly to your left, instead of the one positioned below you.", + "E": "Put the pot into the cabinet directly to your right, rather than returning it to the cabinet below." + }, + "answer_label": "B", + "answer_text": "You put the pot into the cabinet directly below you.", + "q_id": "P07_108_4_T2", + "correct_answer": "You put the pot into the cabinet directly below you.", + "original_options": { + "A": "Put it in the cabinet in front.", + "B": "You put the pot into the cabinet directly below you.", + "C": "Put it in the cabinet behind you.", + "D": "Put the pot into the cabinet directly to your left, instead of the one positioned below you.", + "E": "Put the pot into the cabinet directly to your right, rather than returning it to the cabinet below." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_7", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did this stainless steel frying pan on the stove that is closer to me come from?", + "answer": "It was taken from the cabinet to the bottom right of the stove.", + "question_chinese": "灶台上这个离我更近的不锈钢平底锅是从哪里来的？", + "answer_chinese": "从灶台右下方的柜子里拿出来的。", + "evidence": [ + { + "description": "Took the stainless steel frying pan out of the cabinet to the bottom right of the stove and placed it on the stove.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:02:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the upper-left drawer.", + "B": "From the top-left cabinet.", + "C": "It was taken from the counter to the middle right of the stove, then placed on the front burner.", + "D": "It was pulled off the shelf at the bottom left of the stove before being set down closer to you.", + "E": "It was taken from the cabinet to the bottom right of the stove." + }, + "answer_label": "E", + "answer_text": "It was taken from the cabinet to the bottom right of the stove.", + "q_id": "P07_108_7", + "correct_answer": "It was taken from the cabinet to the bottom right of the stove.", + "original_options": { + "A": "From the upper-left drawer.", + "B": "From the top-left cabinet.", + "C": "It was taken from the counter to the middle right of the stove, then placed on the front burner.", + "D": "It was pulled off the shelf at the bottom left of the stove before being set down closer to you.", + "E": "It was taken from the cabinet to the bottom right of the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 20 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_8", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did this stainless steel frying pan on the stove that is closer to me come from?", + "answer": "It was taken from the cabinet to the bottom right of the stove.", + "question_chinese": "灶台上这个离我更近的不锈钢平底锅是从哪里来的？", + "answer_chinese": "从灶台右下方的柜子里拿出来的。", + "evidence": [ + { + "description": "Took the stainless steel frying pan out of the cabinet to the bottom right of the stove and placed it on the stove.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:04:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the upper-left pantry.", + "B": "From the bottom-left drawer.", + "C": "It was taken from the cabinet to the bottom right of the stove.", + "D": "It was taken from the counter on the upper right of the stove and set down on the stovetop.", + "E": "It was pulled from the drawer at the top left of the stove, then moved onto the burner closest to you." + }, + "answer_label": "C", + "answer_text": "It was taken from the cabinet to the bottom right of the stove.", + "q_id": "P07_108_8", + "correct_answer": "It was taken from the cabinet to the bottom right of the stove.", + "original_options": { + "A": "From the upper-left pantry.", + "B": "From the bottom-left drawer.", + "C": "It was taken from the cabinet to the bottom right of the stove.", + "D": "It was taken from the counter on the upper right of the stove and set down on the stovetop.", + "E": "It was pulled from the drawer at the top left of the stove, then moved onto the burner closest to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 21 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_5_T1", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the stovetop or the sink?", + "answer": "The sink.", + "question_chinese": "灶台和洗碗池哪个离我更近？", + "answer_chinese": "洗手池。", + "evidence": [ + { + "description": "When I face the stove, the sink is on my right hand side.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sink.", + "B": "The stovetop." + }, + "answer_label": "A", + "answer_text": "The sink.", + "q_id": "P07_108_5_T1", + "correct_answer": "The sink.", + "original_options": { + "A": "The sink.", + "B": "The stovetop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_1_T1", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the stovetop relative to me?", + "answer": "On your right-hand side.", + "question_chinese": "灶台在我的哪个方向？", + "answer_chinese": "在你的右手边", + "evidence": [ + { + "description": "When first entering the kitchen, the stove is in the center of the field of view, about 3 meters away.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:03:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The stovetop is on your left-hand side rather than on the right.", + "D": "It sits directly ahead of you, not off to either side as you face forward.", + "E": "On your right-hand side." + }, + "answer_label": "E", + "answer_text": "On your right-hand side.", + "q_id": "P07_108_1_T1", + "correct_answer": "On your right-hand side.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The stovetop is on your left-hand side rather than on the right.", + "D": "It sits directly ahead of you, not off to either side as you face forward.", + "E": "On your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 12, + 15 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_1_T2", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the stovetop relative to me?", + "answer": "On your left-hand side.", + "question_chinese": "灶台在我的哪个方向？", + "answer_chinese": "在你的左手边", + "evidence": [ + { + "description": "When first entering the kitchen, the stove is in the center of the field of view, about 3 meters away.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:03:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "On your left-hand side.", + "D": "The stovetop is on your right-hand side instead of on the left.", + "E": "It is positioned directly in front of you, not to your left as you stand there." + }, + "answer_label": "C", + "answer_text": "On your left-hand side.", + "q_id": "P07_108_1_T2", + "correct_answer": "On your left-hand side.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "On your left-hand side.", + "D": "The stovetop is on your right-hand side instead of on the left.", + "E": "It is positioned directly in front of you, not to your left as you stand there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 12, + 16 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_15", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are left on the stove now?", + "answer": "One pot.", + "question_chinese": "现在灶台上还剩几口锅？", + "answer_chinese": "一口锅。", + "evidence": [ + { + "description": "At the beginning, there were two pots on the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "One pot was taken away.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One pot.", + "B": "Two pots.", + "C": "No pots.", + "D": "Three pots are still on the stove right now, even after moving some cookware away earlier.", + "E": "Four pots remain on the stovetop after the others were moved away." + }, + "answer_label": "A", + "answer_text": "One pot.", + "q_id": "P07_108_15", + "correct_answer": "One pot.", + "original_options": { + "A": "One pot.", + "B": "Two pots.", + "C": "No pots.", + "D": "Three pots are still on the stove right now, even after moving some cookware away earlier.", + "E": "Four pots remain on the stovetop after the others were moved away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 12 + ] + }, + { + "video_uid": "P07_108", + "video_length": "00:05:06", + "qid": "P07_108_16", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are left on the stove now?", + "answer": "Two pots.", + "question_chinese": "现在灶台上还剩几口锅？", + "answer_chinese": "两口锅。", + "evidence": [ + { + "description": "There is one pot on the stove.", + "timestamps": { + "start": "00:00:29", + "end": "00:00:30" + } + }, + { + "description": "A pot was taken out of the cabinet to the bottom right of the stove and placed on the stove.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One pot.", + "B": "Two pots.", + "C": "Three pots.", + "D": "No pots are left on the stove right now, because everything was removed from the burners.", + "E": "Five pots are still sitting on the stovetop at this moment, covering most of the available burners." + }, + "answer_label": "B", + "answer_text": "Two pots.", + "q_id": "P07_108_16", + "correct_answer": "Two pots.", + "original_options": { + "A": "One pot.", + "B": "Two pots.", + "C": "Three pots.", + "D": "No pots are left on the stove right now, because everything was removed from the burners.", + "E": "Five pots are still sitting on the stovetop at this moment, covering most of the available burners." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + } +] diff --git a/QAs/P07_111-mcq.json b/QAs/P07_111-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f29b60caf2e2c41aef8b7dd0418184a61965681d --- /dev/null +++ b/QAs/P07_111-mcq.json @@ -0,0 +1,853 @@ +[ + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_8_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "While I was washing the dishes, how many dishes did I scrub with the sponge?", + "answer": "You washed 2 items in total using the sponge.", + "question_chinese": "在我清洗餐具时，我用海绵清洗了几个餐具？", + "answer_chinese": "你用海绵一共洗了2个物品", + "evidence": [ + { + "description": "The cup.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:36" + } + }, + { + "description": "The knife.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed 1 item using the sponge.", + "B": "You washed 3 items using the sponge.", + "C": "You scrubbed 4 dishes in total with the sponge while washing up at the sink.", + "D": "You didn’t scrub any dishes with the sponge during the washing; you only rinsed them.", + "E": "You washed 2 items in total using the sponge." + }, + "answer_label": "E", + "answer_text": "You washed 2 items in total using the sponge.", + "q_id": "P07_111_8_T1", + "correct_answer": "You washed 2 items in total using the sponge.", + "original_options": { + "A": "You washed 1 item using the sponge.", + "B": "You washed 3 items using the sponge.", + "C": "You scrubbed 4 dishes in total with the sponge while washing up at the sink.", + "D": "You didn’t scrub any dishes with the sponge during the washing; you only rinsed them.", + "E": "You washed 2 items in total using the sponge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_8_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "While I was washing the dishes, how many dishes did I scrub with the sponge?", + "answer": "You washed 3 items in total using the sponge.", + "question_chinese": "在我清洗餐具时，我用海绵清洗了几个餐具？", + "answer_chinese": "你用海绵一共洗了三个物品", + "evidence": [ + { + "description": "The cup.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:38" + } + }, + { + "description": "The knife.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:40" + } + }, + { + "description": "The pot", + "timestamps": { + "start": "00:01:36", + "end": "00:01:45" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed 3 items in total using the sponge.", + "B": "You washed 2 items using the sponge.", + "C": "You washed 4 items using the sponge.", + "D": "You scrubbed 1 dish in total with the sponge during the dishwashing sequence at the sink.", + "E": "You scrubbed 5 dishes in total with the sponge before setting them aside to rinse." + }, + "answer_label": "A", + "answer_text": "You washed 3 items in total using the sponge.", + "q_id": "P07_111_8_T2", + "correct_answer": "You washed 3 items in total using the sponge.", + "original_options": { + "A": "You washed 3 items in total using the sponge.", + "B": "You washed 2 items using the sponge.", + "C": "You washed 4 items using the sponge.", + "D": "You scrubbed 1 dish in total with the sponge during the dishwashing sequence at the sink.", + "E": "You scrubbed 5 dishes in total with the sponge before setting them aside to rinse." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 15 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_3_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I washed the juicer, where did I put the juicer?", + "answer": "You put the juicer next to the sink, and the juicer is by your left hand.", + "question_chinese": "当我洗榨汁机时，我把榨汁机放在了哪里？", + "answer_chinese": "你把榨汁机放在水槽旁边，榨汁机在你左手边", + "evidence": [ + { + "description": "The juicer is on your left.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Next to the sink by your right.", + "B": "You put the juicer next to the sink, and the juicer is by your left hand.", + "C": "Inside the cabinet under the sink.", + "D": "You set the juicer next to the sink, but positioned it closer to your right hand than your left.", + "E": "After washing, you placed the juicer away from the sink area, tucking it into the cabinet beneath." + }, + "answer_label": "B", + "answer_text": "You put the juicer next to the sink, and the juicer is by your left hand.", + "q_id": "P07_111_3_T2", + "correct_answer": "You put the juicer next to the sink, and the juicer is by your left hand.", + "original_options": { + "A": "Next to the sink by your right.", + "B": "You put the juicer next to the sink, and the juicer is by your left hand.", + "C": "Inside the cabinet under the sink.", + "D": "You set the juicer next to the sink, but positioned it closer to your right hand than your left.", + "E": "After washing, you placed the juicer away from the sink area, tucking it into the cabinet beneath." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 17 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_3_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I washed the juicer, where did I put the juicer?", + "answer": "The juicer is inside the cabinet under the dishwashing sink.", + "question_chinese": "当我洗榨汁机时，我把榨汁机放在了哪里？", + "answer_chinese": "榨汁机在洗碗槽下面的柜子里面", + "evidence": [ + { + "description": "The cabinet.", + "timestamps": { + "start": "00:01:24", + "end": "00:02:02" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the counter next to the sink.", + "B": "Inside the cabinet above the sink.", + "C": "You left the juicer next to the sink on the counter instead of putting it away inside the cabinet.", + "D": "The juicer is inside the cabinet under the dishwashing sink.", + "E": "You stored the juicer in a different cabinet, not the one directly under the dishwashing sink." + }, + "answer_label": "D", + "answer_text": "The juicer is inside the cabinet under the dishwashing sink.", + "q_id": "P07_111_3_T1", + "correct_answer": "The juicer is inside the cabinet under the dishwashing sink.", + "original_options": { + "A": "On the counter next to the sink.", + "B": "Inside the cabinet above the sink.", + "C": "You left the juicer next to the sink on the counter instead of putting it away inside the cabinet.", + "D": "The juicer is inside the cabinet under the dishwashing sink.", + "E": "You stored the juicer in a different cabinet, not the one directly under the dishwashing sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 16 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_4_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I came in through the door, what was the first thing I washed?", + "answer": "The first thing you washed was a cup.", + "question_chinese": "当我进门时，我第一个清洗的东西是什么？", + "answer_chinese": "你第一个清洗的是一个杯子。", + "evidence": [ + { + "description": "You are cleaning a cup.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The first thing you washed was a bowl.", + "B": "The first thing you washed was a plate.", + "C": "When you came in, you started by washing a bowl before moving on to other dishes.", + "D": "The first thing you washed was a cup.", + "E": "Right after entering through the door, you washed a plate first, then continued with the rest." + }, + "answer_label": "D", + "answer_text": "The first thing you washed was a cup.", + "q_id": "P07_111_4_T1", + "correct_answer": "The first thing you washed was a cup.", + "original_options": { + "A": "The first thing you washed was a bowl.", + "B": "The first thing you washed was a plate.", + "C": "When you came in, you started by washing a bowl before moving on to other dishes.", + "D": "The first thing you washed was a cup.", + "E": "Right after entering through the door, you washed a plate first, then continued with the rest." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 16 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_4_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I came in through the door, what was the first thing I washed?", + "answer": "You washed a cup.", + "question_chinese": "当我进门时，我第一个清洗的东西是什么？", + "answer_chinese": "你清洗的是一个杯子", + "evidence": [ + { + "description": "You are cleaning a cup.", + "timestamps": { + "start": "00:01:20", + "end": "00:02:01" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed a bowl first.", + "B": "You washed a cup.", + "C": "You washed a plate first.", + "D": "When you came in through the door, the first item you washed was a bowl at the sink.", + "E": "After entering, you began by washing a plate before handling any other dishes." + }, + "answer_label": "B", + "answer_text": "You washed a cup.", + "q_id": "P07_111_4_T2", + "correct_answer": "You washed a cup.", + "original_options": { + "A": "You washed a bowl first.", + "B": "You washed a cup.", + "C": "You washed a plate first.", + "D": "When you came in through the door, the first item you washed was a bowl at the sink.", + "E": "After entering, you began by washing a plate before handling any other dishes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 13 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_13", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to throw theaway the kitchen waste,where is the trashcan?", + "answer": "The trash can is on your left when you enter.", + "question_chinese": "我想要扔掉厨余垃圾，垃圾桶在什么位置", + "answer_chinese": "垃圾桶在你进门的左手边", + "evidence": [ + { + "description": "The trash can is on the left side of the door.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The trash can is on your right.", + "B": "The trash can is on your left when you enter.", + "C": "The trash can is behind you.", + "D": "The trash can is beside the sink, not on your left when you enter.", + "E": "When you enter, the trash can is straight ahead near the counter, not on your left." + }, + "answer_label": "B", + "answer_text": "The trash can is on your left when you enter.", + "q_id": "P07_111_13", + "correct_answer": "The trash can is on your left when you enter.", + "original_options": { + "A": "The trash can is on your right.", + "B": "The trash can is on your left when you enter.", + "C": "The trash can is behind you.", + "D": "The trash can is beside the sink, not on your left when you enter.", + "E": "When you enter, the trash can is straight ahead near the counter, not on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 16 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_7_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to throw away the food waste—where is the trash can?", + "answer": "The trash can is now to your back-right.", + "question_chinese": "我想要扔掉厨余垃圾，垃圾桶在什么位置？", + "answer_chinese": "垃圾桶现在在你右后方。", + "evidence": [ + { + "description": "The trash can is on the left side of the door.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can is now to your back-left.", + "B": "The trash can is now to your front-right.", + "C": "The trash can is currently in front of you on the right, not behind you to the right.", + "D": "The trash can has shifted to your left side now, ending up behind you rather than back-right.", + "E": "The trash can is now to your back-right." + }, + "answer_label": "E", + "answer_text": "The trash can is now to your back-right.", + "q_id": "P07_111_7_T1", + "correct_answer": "The trash can is now to your back-right.", + "original_options": { + "A": "The trash can is now to your back-left.", + "B": "The trash can is now to your front-right.", + "C": "The trash can is currently in front of you on the right, not behind you to the right.", + "D": "The trash can has shifted to your left side now, ending up behind you rather than back-right.", + "E": "The trash can is now to your back-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 17 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_7_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to throw away the food waste—where is the trash can?", + "answer": "When you entered, the trash can was by your left hand.", + "question_chinese": "我想要扔掉厨余垃圾，垃圾桶在什么位置？", + "answer_chinese": "垃圾桶在你进门的左手边", + "evidence": [ + { + "description": "The trash can is on the left side of the door.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "By your right hand when you entered.", + "B": "Straight ahead as you entered.", + "C": "When you entered, the trash can was by your left hand.", + "D": "When you walked in, the trash can was on your right side near the doorway.", + "E": "As you entered the room, the trash can was behind you on the left, not beside your hand." + }, + "answer_label": "C", + "answer_text": "When you entered, the trash can was by your left hand.", + "q_id": "P07_111_7_T2", + "correct_answer": "When you entered, the trash can was by your left hand.", + "original_options": { + "A": "By your right hand when you entered.", + "B": "Straight ahead as you entered.", + "C": "When you entered, the trash can was by your left hand.", + "D": "When you walked in, the trash can was on your right side near the doorway.", + "E": "As you entered the room, the trash can was behind you on the left, not beside your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 18 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_5_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was washing the dishes, where did I put the milk?", + "answer": "The milk was placed on the dishwashing counter by your right hand.", + "question_chinese": "我洗碗餐具时，牛奶放在了哪里？", + "answer_chinese": "牛奶放在了你右手边的洗碗台上。", + "evidence": [ + { + "description": "The milk is placed on the sink on your right.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:57" + } + } + ], + "question_timestamps": "00:01:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the counter by your left hand.", + "B": "Under the radiator pipe.", + "C": "You left the milk on the dishwashing counter in front of you, not by your right hand.", + "D": "You placed the milk on the dishwashing counter closer to your left side while you washed dishes.", + "E": "The milk was placed on the dishwashing counter by your right hand." + }, + "answer_label": "E", + "answer_text": "The milk was placed on the dishwashing counter by your right hand.", + "q_id": "P07_111_5_T2", + "correct_answer": "The milk was placed on the dishwashing counter by your right hand.", + "original_options": { + "A": "On the counter by your left hand.", + "B": "Under the radiator pipe.", + "C": "You left the milk on the dishwashing counter in front of you, not by your right hand.", + "D": "You placed the milk on the dishwashing counter closer to your left side while you washed dishes.", + "E": "The milk was placed on the dishwashing counter by your right hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 4, + 17, + 17 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_5_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was washing the dishes, where did I put the milk?", + "answer": "The milk was placed under the radiator pipe.", + "question_chinese": "我洗碗餐具时，牛奶放在了哪里？", + "answer_chinese": "牛奶放在了暖气管下面", + "evidence": [ + { + "description": "The milk was placed under the radiator.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:03" + } + } + ], + "question_timestamps": "00:02:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the counter by your right hand.", + "B": "On the counter by your left hand.", + "C": "You set the milk on the dishwashing counter beside the sink while you were washing the dishes.", + "D": "You put the milk on the dishwashing counter near your right side, rather than under any pipe.", + "E": "The milk was placed under the radiator pipe." + }, + "answer_label": "E", + "answer_text": "The milk was placed under the radiator pipe.", + "q_id": "P07_111_5_T1", + "correct_answer": "The milk was placed under the radiator pipe.", + "original_options": { + "A": "On the counter by your right hand.", + "B": "On the counter by your left hand.", + "C": "You set the milk on the dishwashing counter beside the sink while you were washing the dishes.", + "D": "You put the milk on the dishwashing counter near your right side, rather than under any pipe.", + "E": "The milk was placed under the radiator pipe." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_12", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When i finished washing the dishes,where was the milk?", + "answer": "The milk was placed under the radiator.", + "question_chinese": "我洗碗餐具时，牛奶放在了哪里", + "answer_chinese": "牛奶放在了暖气管下面", + "evidence": [ + { + "description": "The milk was placed under the radiator.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:03" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The milk was placed under the radiator.", + "B": "It was stored in the refrigerator.", + "C": "It was left on the counter.", + "D": "After finishing the dishes, you placed the milk beside the sink instead of putting it under anything.", + "E": "When you were done, you put the milk inside the cupboard, not under the radiator." + }, + "answer_label": "A", + "answer_text": "The milk was placed under the radiator.", + "q_id": "P07_111_12", + "correct_answer": "The milk was placed under the radiator.", + "original_options": { + "A": "The milk was placed under the radiator.", + "B": "It was stored in the refrigerator.", + "C": "It was left on the counter.", + "D": "After finishing the dishes, you placed the milk beside the sink instead of putting it under anything.", + "E": "When you were done, you put the milk inside the cupboard, not under the radiator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 15 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_1_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I want to wash something later—which one is the detergent, the green bottle or the red bottle?", + "answer": "The red bottle is detergent. You just used the red bottle to your front-right.", + "question_chinese": "我等下想要清洗东西，绿色瓶子和红色瓶子那个是洗涤剂？", + "answer_chinese": "红色瓶子是洗涤剂，你刚刚使用了你右前方的红色瓶子。", + "evidence": [ + { + "description": "You used the detergent in the red bottle.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The red bottle is detergent. You just used the red bottle to your front-right.", + "B": "The green bottle is detergent. You just used the green bottle to your front-right." + }, + "answer_label": "A", + "answer_text": "The red bottle is detergent. You just used the red bottle to your front-right.", + "q_id": "P07_111_1_T1", + "correct_answer": "The red bottle is detergent. You just used the red bottle to your front-right.", + "original_options": { + "A": "The red bottle is detergent. You just used the red bottle to your front-right.", + "B": "The green bottle is detergent. You just used the green bottle to your front-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_1_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I want to wash something later—which one is the detergent, the green bottle or the red bottle?", + "answer": "The red bottle is detergent. You used the red bottle.", + "question_chinese": "我等下想要清洗东西，绿色瓶子和红色瓶子那个是洗涤剂？", + "answer_chinese": "红色瓶子是洗涤剂，你用了红色瓶子。", + "evidence": [ + { + "description": "You used the detergent in the red bottle.", + "timestamps": { + "start": "00:00:18", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The red bottle is detergent. You used the red bottle.", + "B": "The green bottle is detergent. You used the green bottle." + }, + "answer_label": "A", + "answer_text": "The red bottle is detergent. You used the red bottle.", + "q_id": "P07_111_1_T2", + "correct_answer": "The red bottle is detergent. You used the red bottle.", + "original_options": { + "A": "The red bottle is detergent. You used the red bottle.", + "B": "The green bottle is detergent. You used the green bottle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_2_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I entered the kitchen, what were the main things I did?", + "answer": "You were just in the kitchen washing the dishes.", + "question_chinese": "我进厨房主要做了些什么事情？", + "answer_chinese": "你刚在进厨房清洗了餐具。", + "evidence": [ + { + "description": "You washed the tableware.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were just in the kitchen washing the dishes.", + "B": "You were just tidying up the kitchen.", + "C": "You were cooking in the kitchen.", + "D": "After entering the kitchen, you mainly tidied up and put items away rather than washing dishes.", + "E": "Once you went into the kitchen, you focused on organizing and cleaning surfaces instead of doing the dishes." + }, + "answer_label": "A", + "answer_text": "You were just in the kitchen washing the dishes.", + "q_id": "P07_111_2_T2", + "correct_answer": "You were just in the kitchen washing the dishes.", + "original_options": { + "A": "You were just in the kitchen washing the dishes.", + "B": "You were just tidying up the kitchen.", + "C": "You were cooking in the kitchen.", + "D": "After entering the kitchen, you mainly tidied up and put items away rather than washing dishes.", + "E": "Once you went into the kitchen, you focused on organizing and cleaning surfaces instead of doing the dishes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 18 + ] + }, + { + "video_uid": "P07_111", + "video_length": "00:02:10", + "qid": "P07_111_2_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I entered the kitchen, what were the main things I did?", + "answer": "You were just in the kitchen washing the dishes and tidying up the kitchen.", + "question_chinese": "我进厨房主要做了些什么事情？", + "answer_chinese": "你刚在进厨房清洗了餐具和整理厨房。", + "evidence": [ + { + "description": "You washed the tableware.", + "timestamps": { + "start": "00:01:01", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were just in the kitchen washing dishes.", + "B": "You were just tidying up the kitchen.", + "C": "You were just in the kitchen washing the dishes and tidying up the kitchen.", + "D": "After you entered the kitchen, you only washed the dishes and did not tidy up anything else.", + "E": "In the kitchen, you mainly cleaned up and organized, without actually washing the dishes." + }, + "answer_label": "C", + "answer_text": "You were just in the kitchen washing the dishes and tidying up the kitchen.", + "q_id": "P07_111_2_T1", + "correct_answer": "You were just in the kitchen washing the dishes and tidying up the kitchen.", + "original_options": { + "A": "You were just in the kitchen washing dishes.", + "B": "You were just tidying up the kitchen.", + "C": "You were just in the kitchen washing the dishes and tidying up the kitchen.", + "D": "After you entered the kitchen, you only washed the dishes and did not tidy up anything else.", + "E": "In the kitchen, you mainly cleaned up and organized, without actually washing the dishes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 14 + ] + } +] diff --git a/QAs/P07_117-mcq.json b/QAs/P07_117-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7b5f31f15085e7cc5c72b5c9a48b216115b64a8c --- /dev/null +++ b/QAs/P07_117-mcq.json @@ -0,0 +1,711 @@ +[ + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_5", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there on the ground? What did I do with them?", + "answer": "Two, you threw the toilet paper in.", + "question_chinese": "地上有几个垃圾桶？我用它们干了什么？", + "answer_chinese": "2个，你把卫生纸扔进去了。", + "evidence": [ + { + "description": "Two trash cans.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:24" + } + }, + { + "description": "Threw away the toilet paper.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:02:25", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One, you threw it in.", + "B": "Three, you threw paper in.", + "C": "Four, you threw the toilet paper into one of them after setting the others aside.", + "D": "Two, you threw the toilet paper in.", + "E": "Five, you used them as bins and tossed paper into a couple before moving on." + }, + "answer_label": "D", + "answer_text": "Two, you threw the toilet paper in.", + "q_id": "P07_117_5", + "correct_answer": "Two, you threw the toilet paper in.", + "original_options": { + "A": "One, you threw it in.", + "B": "Three, you threw paper in.", + "C": "Four, you threw the toilet paper into one of them after setting the others aside.", + "D": "Two, you threw the toilet paper in.", + "E": "Five, you used them as bins and tossed paper into a couple before moving on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_6", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there on the ground? What did I do with them?", + "answer": "Two, you threw the toilet paper in.", + "question_chinese": "地上有几个垃圾桶？我用它们干了什么？", + "answer_chinese": "2个，你把卫生纸扔进去了。", + "evidence": [ + { + "description": "Two trash cans.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:24" + } + }, + { + "description": "You threw the toilet paper in", + "timestamps": { + "start": "00:00:22", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:05:08", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One, you threw it in.", + "B": "Three, you tossed wrappers in.", + "C": "Two, you threw the toilet paper in.", + "D": "Four, you discarded some paper into one of the cans while the others stayed on the floor.", + "E": "Five, you threw plastic bottles into a trash can and left the remaining cans on the ground." + }, + "answer_label": "C", + "answer_text": "Two, you threw the toilet paper in.", + "q_id": "P07_117_6", + "correct_answer": "Two, you threw the toilet paper in.", + "original_options": { + "A": "One, you threw it in.", + "B": "Three, you tossed wrappers in.", + "C": "Two, you threw the toilet paper in.", + "D": "Four, you discarded some paper into one of the cans while the others stayed on the floor.", + "E": "Five, you threw plastic bottles into a trash can and left the remaining cans on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I put in the sink as soon as I entered the door?", + "answer": "You put the knife and fork in the sink when you first entered the door.", + "question_chinese": "我刚进门往水池中放了什么东西", + "answer_chinese": "你刚进门时往水池里放了刀叉", + "evidence": [ + { + "description": "You just entered the door and put down the knife and fork.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:02:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put a spoon in.", + "B": "You put the knife and fork in the sink when you first entered the door.", + "C": "You put a cup in.", + "D": "You put the knife and plate in the sink when you first entered the door.", + "E": "You put the bowl and fork in the sink as soon as you came through the door." + }, + "answer_label": "B", + "answer_text": "You put the knife and fork in the sink when you first entered the door.", + "q_id": "P07_117_2", + "correct_answer": "You put the knife and fork in the sink when you first entered the door.", + "original_options": { + "A": "You put a spoon in.", + "B": "You put the knife and fork in the sink when you first entered the door.", + "C": "You put a cup in.", + "D": "You put the knife and plate in the sink when you first entered the door.", + "E": "You put the bowl and fork in the sink as soon as you came through the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I put in the sink as soon as I entered the door?", + "answer": "You put the knives and forks in the sink when you just entered the door.", + "question_chinese": "我刚进门往水池中放了什么东西", + "answer_chinese": "你刚进门时往水池里放了刀叉", + "evidence": [ + { + "description": "You just entered the door and put down the knife and fork.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:05:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put the knives and forks in the sink when you just entered the door.", + "B": "You put plates and cups.", + "C": "You put pots and pans.", + "D": "You put the glasses and mugs in the sink when you just entered the door.", + "E": "You put trays and cutting boards in the sink right after you came in the door." + }, + "answer_label": "A", + "answer_text": "You put the knives and forks in the sink when you just entered the door.", + "q_id": "P07_117_1", + "correct_answer": "You put the knives and forks in the sink when you just entered the door.", + "original_options": { + "A": "You put the knives and forks in the sink when you just entered the door.", + "B": "You put plates and cups.", + "C": "You put pots and pans.", + "D": "You put the glasses and mugs in the sink when you just entered the door.", + "E": "You put trays and cutting boards in the sink right after you came in the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_14", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the toaster directly?", + "answer": "You can.", + "question_chinese": "我可以直接拿到烤面包机吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "The toaster is on the table to the left of the kitchen.", + "timestamps": { + "start": "00:03:26", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:03:31", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "P07_117_14", + "correct_answer": "You can.", + "original_options": { + "A": "You can't.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_13", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the toaster directly?", + "answer": "You can't.", + "question_chinese": "我可以直接拿到烤面包机吗", + "answer_chinese": "你不能", + "evidence": [ + { + "description": "The toaster is on the table to the left of the kitchen.", + "timestamps": { + "start": "00:03:26", + "end": "00:03:27" + } + }, + { + "description": "Can't see the toaster.", + "timestamps": { + "start": "00:05:11", + "end": "00:05:13" + } + } + ], + "question_timestamps": "00:05:16", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't.", + "B": "You can." + }, + "answer_label": "A", + "answer_text": "You can't.", + "q_id": "P07_117_13", + "correct_answer": "You can't.", + "original_options": { + "A": "You can't.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_11", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How do I get out of the kitchen door?", + "answer": "Turn left and then go straight.", + "question_chinese": "我如何出厨房门", + "answer_chinese": "向左转然后直走", + "evidence": [ + { + "description": "Position at the kitchen door and inside the kitchen.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:01:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight.", + "B": "Go straight, then turn left.", + "C": "Turn left and then go straight.", + "D": "Turn left first, then turn right before walking straight out the kitchen door.", + "E": "Turn around and head straight toward the doorway until you exit the kitchen." + }, + "answer_label": "C", + "answer_text": "Turn left and then go straight.", + "q_id": "P07_117_11", + "correct_answer": "Turn left and then go straight.", + "original_options": { + "A": "Turn right, then go straight.", + "B": "Go straight, then turn left.", + "C": "Turn left and then go straight.", + "D": "Turn left first, then turn right before walking straight out the kitchen door.", + "E": "Turn around and head straight toward the doorway until you exit the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_12", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How do I get out of the kitchen door?", + "answer": "Turn around and then go straight.", + "question_chinese": "我如何出厨房门", + "answer_chinese": "向后转然后直走", + "evidence": [ + { + "description": "Location at the kitchen entrance and inside the kitchen.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:03:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go straight.", + "B": "Turn around and then go straight.", + "C": "Turn right, then go straight.", + "D": "Go straight first, then turn around to face the kitchen door and continue forward.", + "E": "Turn around, then turn left at the doorway before walking straight out." + }, + "answer_label": "B", + "answer_text": "Turn around and then go straight.", + "q_id": "P07_117_12", + "correct_answer": "Turn around and then go straight.", + "original_options": { + "A": "Turn left, then go straight.", + "B": "Turn around and then go straight.", + "C": "Turn right, then go straight.", + "D": "Go straight first, then turn around to face the kitchen door and continue forward.", + "E": "Turn around, then turn left at the doorway before walking straight out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 12 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_9", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I put the rice in the box?", + "answer": "You don't have it.", + "question_chinese": "我是否将米饭放入盒子了", + "answer_chinese": "你没有", + "evidence": [ + { + "description": "You didn't put the rice in the box.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:44" + } + } + ], + "question_timestamps": "00:01:44", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You put the rice in the box.", + "B": "You don't have it." + }, + "answer_label": "B", + "answer_text": "You don't have it.", + "q_id": "P07_117_9", + "correct_answer": "You don't have it.", + "original_options": { + "A": "You put the rice in the box.", + "B": "You don't have it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_10", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I put the rice in the box?", + "answer": "You put the rice in the box.", + "question_chinese": "我是否将米饭放入盒子了", + "answer_chinese": "你将米饭放入盒子里了", + "evidence": [ + { + "description": "You put the rice in the box.", + "timestamps": { + "start": "00:03:23", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:04:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You put the rice in the box.", + "B": "You did not put the rice in the box." + }, + "answer_label": "A", + "answer_text": "You put the rice in the box.", + "q_id": "P07_117_10", + "correct_answer": "You put the rice in the box.", + "original_options": { + "A": "You put the rice in the box.", + "B": "You did not put the rice in the box." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_3", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I directly touch the trash bin on the desktop now?", + "answer": "Yes,you can.", + "question_chinese": "我现在可以直接碰到桌面上的垃圾桶吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "The trash bin on the desktop was used directly at the sink.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:02:18", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,you cannot.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "P07_117_3", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No,you cannot.", + "B": "Yes,you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_4", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I directly touch the trash bin on the desktop now?", + "answer": "Yes,you can.", + "question_chinese": "我现在可以直接碰到桌面上的垃圾桶吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "The trash bin on the desktop was used directly at the sink.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No,you cannot." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "P07_117_4", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No,you cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_7", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are there on the stove?", + "answer": "There are two pots on the stove.", + "question_chinese": "灶台上有多少个锅", + "answer_chinese": "灶台上有两个锅", + "evidence": [ + { + "description": "There are two pots on the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:02:01", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one pot on the stove.", + "B": "There are two pots on the stove.", + "C": "There are three pots on the stove.", + "D": "There are four pots on the stove, spaced across the burners in front of you.", + "E": "There are five pots on the stove, with one sitting on each burner." + }, + "answer_label": "B", + "answer_text": "There are two pots on the stove.", + "q_id": "P07_117_7", + "correct_answer": "There are two pots on the stove.", + "original_options": { + "A": "There is one pot on the stove.", + "B": "There are two pots on the stove.", + "C": "There are three pots on the stove.", + "D": "There are four pots on the stove, spaced across the burners in front of you.", + "E": "There are five pots on the stove, with one sitting on each burner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 13 + ] + }, + { + "video_uid": "P07_117", + "video_length": "00:06:13", + "qid": "P07_117_8", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are there on the stove?", + "answer": "There is a pot on the stove.", + "question_chinese": "灶台上有多少个锅", + "answer_chinese": "灶台上有一个锅", + "evidence": [ + { + "description": "There is a pot on the stove.", + "timestamps": { + "start": "00:03:59", + "end": "00:04:13" + } + } + ], + "question_timestamps": "00:04:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two pots on the stove.", + "B": "There are three pots on the stove.", + "C": "There is a pot on the stove.", + "D": "There are four pots on the stove, with two on the front burners and two in back.", + "E": "There are five pots on the stove, covering most of the available burners." + }, + "answer_label": "C", + "answer_text": "There is a pot on the stove.", + "q_id": "P07_117_8", + "correct_answer": "There is a pot on the stove.", + "original_options": { + "A": "There are two pots on the stove.", + "B": "There are three pots on the stove.", + "C": "There is a pot on the stove.", + "D": "There are four pots on the stove, with two on the front burners and two in back.", + "E": "There are five pots on the stove, covering most of the available burners." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 13 + ] + } +] diff --git a/QAs/P09_101-mcq.json b/QAs/P09_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..96e8fd905f198fcd8b173a8b9dd7eb860374bc68 --- /dev/null +++ b/QAs/P09_101-mcq.json @@ -0,0 +1,1384 @@ +[ + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_6_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was chopping vegetables, I needed to use a rag to clean up—where is the rag?", + "answer": "The rag was originally on the stool.", + "question_chinese": "切菜时我需要使用抹布清理，抹布在哪里？", + "answer_chinese": "抹布原来在凳子上", + "evidence": [ + { + "description": "The dishcloth oas on the stool.", + "timestamps": { + "start": "00:00:54", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:07:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was on the counter.", + "B": "It was on the floor.", + "C": "The rag was originally in the first drawer, right beside where you were chopping.", + "D": "The rag was originally on the stool.", + "E": "The rag was originally on the sink edge, near your cutting area when you cleaned up." + }, + "answer_label": "D", + "answer_text": "The rag was originally on the stool.", + "q_id": "P09_101_6_T1", + "correct_answer": "The rag was originally on the stool.", + "original_options": { + "A": "It was on the counter.", + "B": "It was on the floor.", + "C": "The rag was originally in the first drawer, right beside where you were chopping.", + "D": "The rag was originally on the stool.", + "E": "The rag was originally on the sink edge, near your cutting area when you cleaned up." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_12", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When i need to clean while chopping vegetables, where is the dishcloth?", + "answer": "The dishcloth originally on the stool, but later fell down.", + "question_chinese": "切菜时我需要使用抹布清理，抹布在哪里", + "answer_chinese": "抹布原来在凳子上，后来掉到了地面上", + "evidence": [ + { + "description": "The dishcloth oas on the stool.", + "timestamps": { + "start": "00:00:54:", + "end": "00:07:29" + } + }, + { + "description": "The dishcloth oas fell down.", + "timestamps": { + "start": "00:07:33", + "end": "00:07:29" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was on the floor.", + "B": "The dishcloth originally on the stool, but later fell down.", + "C": "It was on the counter.", + "D": "The dishcloth started on the table and stayed there while you chopped the vegetables.", + "E": "The dishcloth began in the sink, then you moved it onto the counter before cleaning up." + }, + "answer_label": "B", + "answer_text": "The dishcloth originally on the stool, but later fell down.", + "q_id": "P09_101_12", + "correct_answer": "The dishcloth originally on the stool, but later fell down.", + "original_options": { + "A": "It was on the floor.", + "B": "The dishcloth originally on the stool, but later fell down.", + "C": "It was on the counter.", + "D": "The dishcloth started on the table and stayed there while you chopped the vegetables.", + "E": "The dishcloth began in the sink, then you moved it onto the counter before cleaning up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_6_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was chopping vegetables, I needed to use a rag to clean up—where is the rag?", + "answer": "It fell onto the floor.", + "question_chinese": "切菜时我需要使用抹布清理，抹布在哪里？", + "answer_chinese": "它掉到了地面上", + "evidence": [ + { + "description": "The dishcloth oas fell down.", + "timestamps": { + "start": "00:07:29", + "end": "00:07:33" + } + } + ], + "question_timestamps": "00:07:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on the stool.", + "B": "It’s in the first drawer.", + "C": "It’s in the second drawer to your left, where you can grab it without bending down.", + "D": "It stayed on the stool the whole time, so you didn’t have to pick it up from below.", + "E": "It fell onto the floor." + }, + "answer_label": "E", + "answer_text": "It fell onto the floor.", + "q_id": "P09_101_6_T2", + "correct_answer": "It fell onto the floor.", + "original_options": { + "A": "It’s on the stool.", + "B": "It’s in the first drawer.", + "C": "It’s in the second drawer to your left, where you can grab it without bending down.", + "D": "It stayed on the stool the whole time, so you didn’t have to pick it up from below.", + "E": "It fell onto the floor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 19 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_10", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mugs are there in the kitchen?", + "answer": "There are two in total.", + "question_chinese": "厨房里一共有几个马克杯", + "answer_chinese": "一共有两个", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:06:57:", + "end": "00:07:29" + } + }, + { + "description": " The second one.", + "timestamps": { + "start": "07:05:", + "end": "00:07:29" + } + } + ], + "question_timestamps": "00:07:29", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There is one mug.", + "B": "There are three mugs.", + "C": "There are five mugs in total, spread out around the kitchen counter area.", + "D": "There are seven mugs in total, with several grouped together near the sink.", + "E": "There are two in total." + }, + "answer_label": "E", + "answer_text": "There are two in total.", + "q_id": "P09_101_10", + "correct_answer": "There are two in total.", + "original_options": { + "A": "There is one mug.", + "B": "There are three mugs.", + "C": "There are five mugs in total, spread out around the kitchen counter area.", + "D": "There are seven mugs in total, with several grouped together near the sink.", + "E": "There are two in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 13 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mugs are there in the kitchen?", + "answer": "There are two in total.", + "question_chinese": "厨房里一共有几个马克杯", + "answer_chinese": "一共有两个", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:06:57:", + "end": "00:07:29" + } + }, + { + "description": " The second one.", + "timestamps": { + "start": "00:07:05:", + "end": "00:07:29" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are two in total.", + "B": "There is one mug.", + "C": "There are three mugs.", + "D": "There are four mugs in total, including one set aside near the counter edge.", + "E": "There are five mugs in total, with a few clustered together in the kitchen." + }, + "answer_label": "A", + "answer_text": "There are two in total.", + "q_id": "P09_101_9", + "correct_answer": "There are two in total.", + "original_options": { + "A": "There are two in total.", + "B": "There is one mug.", + "C": "There are three mugs.", + "D": "There are four mugs in total, including one set aside near the counter edge.", + "E": "There are five mugs in total, with a few clustered together in the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_3", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many carrots did I use in cooking?", + "answer": "You used two carrots.", + "question_chinese": "我做菜一共用了几根胡罗卜", + "answer_chinese": "你用了两根", + "evidence": [ + { + "description": "The first carrort.", + "timestamps": { + "start": "00:00:27", + "end": "00:07:29" + } + }, + { + "description": "The second carrort.", + "timestamps": { + "start": "00:00:59", + "end": "00:07:29" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You used one carrot.", + "B": "You used three carrots.", + "C": "You used four carrots, cutting them up along with the other vegetables you prepared.", + "D": "You used two carrots.", + "E": "You used five carrots, working through them one by one while cooking." + }, + "answer_label": "D", + "answer_text": "You used two carrots.", + "q_id": "P09_101_3", + "correct_answer": "You used two carrots.", + "original_options": { + "A": "You used one carrot.", + "B": "You used three carrots.", + "C": "You used four carrots, cutting them up along with the other vegetables you prepared.", + "D": "You used two carrots.", + "E": "You used five carrots, working through them one by one while cooking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 12 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_4", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many carrots did I use in cooking?", + "answer": "You used two carrorts.", + "question_chinese": "我做菜一共用了几根胡罗卜", + "answer_chinese": "你用了两根", + "evidence": [ + { + "description": "The first carrort.", + "timestamps": { + "start": "00:00:27", + "end": "00:07:29" + } + }, + { + "description": "The second carrort.", + "timestamps": { + "start": "00:00:59", + "end": "00:07:29" + } + } + ], + "question_timestamps": "00:07:31", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You used one carrot.", + "B": "You used three carrots.", + "C": "You used two carrorts.", + "D": "You used four carrots, chopping them down into pieces during the cooking.", + "E": "You used five carrots, finishing all of them before moving on to other ingredients." + }, + "answer_label": "C", + "answer_text": "You used two carrorts.", + "q_id": "P09_101_4", + "correct_answer": "You used two carrorts.", + "original_options": { + "A": "You used one carrot.", + "B": "You used three carrots.", + "C": "You used two carrorts.", + "D": "You used four carrots, chopping them down into pieces during the cooking.", + "E": "You used five carrots, finishing all of them before moving on to other ingredients." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_2_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many carrots have I used for cooking so far?", + "answer": "You used one.", + "question_chinese": "我做菜现在用了几根胡罗卜？", + "answer_chinese": "你用了一根", + "evidence": [ + { + "description": "The first carrort.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You used zero.", + "B": "You used two.", + "C": "You used three so far, continuing to add more as you prep the vegetables.", + "D": "You used four so far, counting each carrot you handled during the cooking.", + "E": "You used one." + }, + "answer_label": "E", + "answer_text": "You used one.", + "q_id": "P09_101_2_T1", + "correct_answer": "You used one.", + "original_options": { + "A": "You used zero.", + "B": "You used two.", + "C": "You used three so far, continuing to add more as you prep the vegetables.", + "D": "You used four so far, counting each carrot you handled during the cooking.", + "E": "You used one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 13 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_2_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many carrots have I used for cooking so far?", + "answer": "You used two.", + "question_chinese": "我做菜现在用了几根胡罗卜？", + "answer_chinese": "你用了两根", + "evidence": [ + { + "description": "The first carrort.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:29" + } + }, + { + "description": "The second carrort.", + "timestamps": { + "start": "00:00:59", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:07:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You used two.", + "B": "You used one.", + "C": "You used three.", + "D": "You used zero so far, since you haven’t started using any carrots yet.", + "E": "You used four so far, having gone through several carrots during preparation." + }, + "answer_label": "A", + "answer_text": "You used two.", + "q_id": "P09_101_2_T2", + "correct_answer": "You used two.", + "original_options": { + "A": "You used two.", + "B": "You used one.", + "C": "You used three.", + "D": "You used zero so far, since you haven’t started using any carrots yet.", + "E": "You used four so far, having gone through several carrots during preparation." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 12 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_5", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I stillneed some scallions for cooking , can I reach them?", + "answer": "Got it,on your left.", + "question_chinese": "我做饭还需要一些葱，我伸手拿的到吗", + "answer_chinese": "拿得到，在你左边", + "evidence": [ + { + "description": "on your left", + "timestamps": { + "start": "00:04:14", + "end": "00:04:21" + } + } + ], + "question_timestamps": "00:04:21", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Got it,on your left.", + "B": "Got it,on your right." + }, + "answer_label": "A", + "answer_text": "Got it,on your left.", + "q_id": "P09_101_5", + "correct_answer": "Got it,on your left.", + "original_options": { + "A": "Got it,on your left.", + "B": "Got it,on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_3_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I still need some scallions for cooking—can I reach them by stretching out my hand?", + "answer": "Yes, you can reach it—it’s to your left.", + "question_chinese": "我做饭还需要一些葱，我伸手拿的到吗？", + "answer_chinese": "拿得到，在你左边", + "evidence": [ + { + "description": "on your left", + "timestamps": { + "start": "00:04:14", + "end": "00:04:21" + } + } + ], + "question_timestamps": "00:04:26", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can reach it—it’s to your left.", + "B": "No, you can’t reach it—it’s to your left." + }, + "answer_label": "A", + "answer_text": "Yes, you can reach it—it’s to your left.", + "q_id": "P09_101_3_T1", + "correct_answer": "Yes, you can reach it—it’s to your left.", + "original_options": { + "A": "Yes, you can reach it—it’s to your left.", + "B": "No, you can’t reach it—it’s to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_3_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I still need some scallions for cooking—can I reach them by stretching out my hand?", + "answer": "You can reach it; it’s in front of you.", + "question_chinese": "我做饭还需要一些葱，我伸手拿的到吗？", + "answer_chinese": "拿得到，在你前边", + "evidence": [ + { + "description": "on your left", + "timestamps": { + "start": "00:04:14", + "end": "00:04:21" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t reach it; it’s in front of you.", + "B": "You can reach it; it’s in front of you." + }, + "answer_label": "B", + "answer_text": "You can reach it; it’s in front of you.", + "q_id": "P09_101_3_T2", + "correct_answer": "You can reach it; it’s in front of you.", + "original_options": { + "A": "No, you can’t reach it; it’s in front of you.", + "B": "You can reach it; it’s in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_8_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now during cooking, what processing steps did I do to the scallions?", + "answer": "You just peeled the scallion and removed the top.", + "question_chinese": "我刚在烹饪的过程中对葱做了什么处理步骤？", + "answer_chinese": "你刚才对葱进行了剥皮，去头的处理。", + "evidence": [ + { + "description": "You peeled scallions.", + "timestamps": { + "start": "00:04:34", + "end": "00:04:38" + } + }, + { + "description": "You removed the heads.", + "timestamps": { + "start": "00:04:47", + "end": "00:04:59" + } + } + ], + "question_timestamps": "00:07:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You only peeled the scallion.", + "B": "You only removed the top.", + "C": "You just peeled the scallion and removed the top.", + "D": "You peeled the scallion, then cut it into pieces, but you didn’t remove the top.", + "E": "You removed the top and sliced it up, but you didn’t peel the scallion first." + }, + "answer_label": "C", + "answer_text": "You just peeled the scallion and removed the top.", + "q_id": "P09_101_8_T1", + "correct_answer": "You just peeled the scallion and removed the top.", + "original_options": { + "A": "You only peeled the scallion.", + "B": "You only removed the top.", + "C": "You just peeled the scallion and removed the top.", + "D": "You peeled the scallion, then cut it into pieces, but you didn’t remove the top.", + "E": "You removed the top and sliced it up, but you didn’t peel the scallion first." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_8_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now during cooking, what processing steps did I do to the scallions?", + "answer": "You just peeled the scallion, removed the top, and cut it into pieces.", + "question_chinese": "我刚在烹饪的过程中对葱做了什么处理步骤？", + "answer_chinese": "你刚才对葱进行了剥皮，去头和切断的处理。", + "evidence": [ + { + "description": "You peeled scallions.", + "timestamps": { + "start": "00:04:34", + "end": "00:04:38" + } + }, + { + "description": "You removed the heads.", + "timestamps": { + "start": "00:04:47", + "end": "00:04:55" + } + }, + { + "description": "You cut the scallions.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:38" + } + } + ], + "question_timestamps": "00:07:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You just peeled the scallion.", + "B": "You only removed the top.", + "C": "You peeled the scallion and removed the top, but you didn’t cut it into pieces afterward.", + "D": "You removed the top and cut it into pieces, but you didn’t peel the scallion first.", + "E": "You just peeled the scallion, removed the top, and cut it into pieces." + }, + "answer_label": "E", + "answer_text": "You just peeled the scallion, removed the top, and cut it into pieces.", + "q_id": "P09_101_8_T2", + "correct_answer": "You just peeled the scallion, removed the top, and cut it into pieces.", + "original_options": { + "A": "You just peeled the scallion.", + "B": "You only removed the top.", + "C": "You peeled the scallion and removed the top, but you didn’t cut it into pieces afterward.", + "D": "You removed the top and cut it into pieces, but you didn’t peel the scallion first.", + "E": "You just peeled the scallion, removed the top, and cut it into pieces." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_5_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mugs are there in my kitchen in total?", + "answer": "You have two in total, and the mug is above you.", + "question_chinese": "我厨房里一共有几个马克杯？", + "answer_chinese": "你一共有两个，马克杯在你上边", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:06:57", + "end": "00:07:05" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:07:05", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:07:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You have one in total.", + "B": "You have three in total.", + "C": "You have two in total, and the mug is above you.", + "D": "You have four in total, and the mug is above you where you can see it.", + "E": "You have zero in total, and there isn’t any mug above you right now." + }, + "answer_label": "C", + "answer_text": "You have two in total, and the mug is above you.", + "q_id": "P09_101_5_T1", + "correct_answer": "You have two in total, and the mug is above you.", + "original_options": { + "A": "You have one in total.", + "B": "You have three in total.", + "C": "You have two in total, and the mug is above you.", + "D": "You have four in total, and the mug is above you where you can see it.", + "E": "You have zero in total, and there isn’t any mug above you right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_5_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mugs are there in my kitchen in total?", + "answer": "You have two in total; it’s to your front-left.", + "question_chinese": "我厨房里一共有几个马克杯？", + "answer_chinese": "你一共有两个，在你左前方", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:06:57", + "end": "00:07:05" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:07:05", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:07:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You have one in total.", + "B": "You have three in total.", + "C": "You have four in total; they’re positioned to your front-left near where you’re looking.", + "D": "You have two in total; it’s to your front-left.", + "E": "You have zero in total; there aren’t any mugs to your front-left at the moment." + }, + "answer_label": "D", + "answer_text": "You have two in total; it’s to your front-left.", + "q_id": "P09_101_5_T2", + "correct_answer": "You have two in total; it’s to your front-left.", + "original_options": { + "A": "You have one in total.", + "B": "You have three in total.", + "C": "You have four in total; they’re positioned to your front-left near where you’re looking.", + "D": "You have two in total; it’s to your front-left.", + "E": "You have zero in total; there aren’t any mugs to your front-left at the moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_7_T1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take as I moved around in the kitchen?", + "answer": "You first turned half a circle counterclockwise.", + "question_chinese": "我在厨房的运动轨迹是什么？", + "answer_chinese": "你先逆时针转了半圈", + "evidence": [ + { + "description": "You turned counteredclockwise for half a circle.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You first turned clockwise.", + "B": "You first turned a full circle.", + "C": "You first turned a quarter circle counterclockwise before doing anything else in the kitchen.", + "D": "You first turned half a circle clockwise, then continued moving forward afterward.", + "E": "You first turned half a circle counterclockwise." + }, + "answer_label": "E", + "answer_text": "You first turned half a circle counterclockwise.", + "q_id": "P09_101_7_T1", + "correct_answer": "You first turned half a circle counterclockwise.", + "original_options": { + "A": "You first turned clockwise.", + "B": "You first turned a full circle.", + "C": "You first turned a quarter circle counterclockwise before doing anything else in the kitchen.", + "D": "You first turned half a circle clockwise, then continued moving forward afterward.", + "E": "You first turned half a circle counterclockwise." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 14, + 12 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_7_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take as I moved around in the kitchen?", + "answer": "After you turned half a circle counterclockwise, you didn’t move anymore.", + "question_chinese": "我在厨房的运动轨迹是什么？", + "answer_chinese": "你先逆时针转了半圈后便一直未在运动", + "evidence": [ + { + "description": "You turned counteredclockwise for half a circle.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:06" + } + }, + { + "description": "You have been standing there without moving ", + "timestamps": { + "start": "00:00:06", + "end": "00:07:29" + } + } + ], + "question_timestamps": "00:07:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "After you turned half a circle counterclockwise, you didn’t move anymore.", + "B": "You kept walking forward.", + "C": "You turned clockwise instead.", + "D": "After you turned half a circle counterclockwise, you walked forward two steps before stopping.", + "E": "After you turned half a circle counterclockwise, you turned another quarter circle and then stopped." + }, + "answer_label": "A", + "answer_text": "After you turned half a circle counterclockwise, you didn’t move anymore.", + "q_id": "P09_101_7_T2", + "correct_answer": "After you turned half a circle counterclockwise, you didn’t move anymore.", + "original_options": { + "A": "After you turned half a circle counterclockwise, you didn’t move anymore.", + "B": "You kept walking forward.", + "C": "You turned clockwise instead.", + "D": "After you turned half a circle counterclockwise, you walked forward two steps before stopping.", + "E": "After you turned half a circle counterclockwise, you turned another quarter circle and then stopped." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_1_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to put the peeler back where it was—where did I take it from?", + "answer": "It was in the second drawer to your left.", + "question_chinese": "我想将削皮器放回原位，我是从哪里拿的？", + "answer_chinese": "在你左边的第二个抽屉里", + "evidence": [ + { + "description": "In the first drawer.", + "timestamps": { + "start": "00:00:01", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the first drawer to your left.", + "B": "In the second drawer to your right.", + "C": "It was in the second drawer to your left.", + "D": "You took it from the top drawer on your left, not the second one.", + "E": "It came from the second drawer on your right side, where you reached in earlier." + }, + "answer_label": "C", + "answer_text": "It was in the second drawer to your left.", + "q_id": "P09_101_1_T2", + "correct_answer": "It was in the second drawer to your left.", + "original_options": { + "A": "In the first drawer to your left.", + "B": "In the second drawer to your right.", + "C": "It was in the second drawer to your left.", + "D": "You took it from the top drawer on your left, not the second one.", + "E": "It came from the second drawer on your right side, where you reached in earlier." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 15 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to put the peeler back where I took it from?", + "answer": "In the first drawer.", + "question_chinese": "我想将削皮器放回原位，我是从哪里拿的", + "answer_chinese": "在第一格抽屉里", + "evidence": [ + { + "description": "In the first drawer.", + "timestamps": { + "start": "00:00:01", + "end": "00:07:29" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the second drawer.", + "B": "In the top drawer.", + "C": "In the first drawer.", + "D": "You took it from the bottom drawer, not the first drawer you opened.", + "E": "It was kept in the drawer beside the counter, not in the first drawer." + }, + "answer_label": "C", + "answer_text": "In the first drawer.", + "q_id": "P09_101_1", + "correct_answer": "In the first drawer.", + "original_options": { + "A": "In the second drawer.", + "B": "In the top drawer.", + "C": "In the first drawer.", + "D": "You took it from the bottom drawer, not the first drawer you opened.", + "E": "It was kept in the drawer beside the counter, not in the first drawer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_1_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to put the peeler back where it was—where did I take it from?", + "answer": "It was in the first drawer compartment.", + "question_chinese": "我想将削皮器放回原位，我是从哪里拿的？", + "answer_chinese": "你从第一格抽屉里", + "evidence": [ + { + "description": "In the first drawer.", + "timestamps": { + "start": "00:00:01", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:07:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was in the second drawer.", + "B": "It was in the main drawer space.", + "C": "You took it from the first drawer itself, not a separate compartment inside it.", + "D": "It was in the first drawer compartment.", + "E": "It came from the second drawer compartment, rather than the first drawer compartment you mentioned." + }, + "answer_label": "D", + "answer_text": "It was in the first drawer compartment.", + "q_id": "P09_101_1_T1", + "correct_answer": "It was in the first drawer compartment.", + "original_options": { + "A": "It was in the second drawer.", + "B": "It was in the main drawer space.", + "C": "You took it from the first drawer itself, not a separate compartment inside it.", + "D": "It was in the first drawer compartment.", + "E": "It came from the second drawer compartment, rather than the first drawer compartment you mentioned." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 15 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_4_T1", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the electric kettle relative to me?", + "answer": "On your left-hand side.", + "question_chinese": "我热水壶在哪个方位？", + "answer_chinese": "在你的左手边", + "evidence": [ + { + "description": "On the left", + "timestamps": { + "start": "00:04:15", + "end": "00:04:16" + } + } + ], + "question_timestamps": "00:04:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your left-hand side.", + "B": "On your right-hand side.", + "C": "Directly in front of you.", + "D": "Behind you, so you would need to turn around to see the electric kettle.", + "E": "Off to your front-right side, rather than sitting on your left-hand side." + }, + "answer_label": "A", + "answer_text": "On your left-hand side.", + "q_id": "P09_101_4_T1", + "correct_answer": "On your left-hand side.", + "original_options": { + "A": "On your left-hand side.", + "B": "On your right-hand side.", + "C": "Directly in front of you.", + "D": "Behind you, so you would need to turn around to see the electric kettle.", + "E": "Off to your front-right side, rather than sitting on your left-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 12 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_4_T2", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the electric kettle relative to me?", + "answer": "It’s above you.", + "question_chinese": "我热水壶在哪个方位？", + "answer_chinese": "在你的上边", + "evidence": [ + { + "description": "Above you", + "timestamps": { + "start": "00:07:25", + "end": "00:07:27" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s below you.", + "B": "It’s in front of you.", + "C": "It’s above you.", + "D": "It’s behind you at your level, rather than above you where you would look up.", + "E": "The kettle is off to your left at counter height, not above you where you would look up." + }, + "answer_label": "C", + "answer_text": "It’s above you.", + "q_id": "P09_101_4_T2", + "correct_answer": "It’s above you.", + "original_options": { + "A": "It’s below you.", + "B": "It’s in front of you.", + "C": "It’s above you.", + "D": "It’s behind you at your level, rather than above you where you would look up.", + "E": "The kettle is off to your left at counter height, not above you where you would look up." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 16, + 18 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_7", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The kettle is in which direction?", + "answer": "On you left hand side.", + "question_chinese": "热水壶在哪个方位", + "answer_chinese": "在你的左手边", + "evidence": [ + { + "description": "On the left", + "timestamps": { + "start": "00:04:15", + "end": "00:04:16" + } + } + ], + "question_timestamps": "00:04:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "On your right-hand side.", + "C": "The kettle is behind you at your level, not on your left-hand side.", + "D": "On you left hand side.", + "E": "It’s positioned directly above you, rather than on your left-hand side." + }, + "answer_label": "D", + "answer_text": "On you left hand side.", + "q_id": "P09_101_7", + "correct_answer": "On you left hand side.", + "original_options": { + "A": "In front of you.", + "B": "On your right-hand side.", + "C": "The kettle is behind you at your level, not on your left-hand side.", + "D": "On you left hand side.", + "E": "It’s positioned directly above you, rather than on your left-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 12 + ] + }, + { + "video_uid": "P09_101", + "video_length": "00:07:38", + "qid": "P09_101_8", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The kettle is in which direction?", + "answer": "On your lef hand side.", + "question_chinese": "热水壶在哪个方位", + "answer_chinese": "在你的左手边", + "evidence": [ + { + "description": "On the left", + "timestamps": { + "start": "00:04:15", + "end": "00:04:016" + } + } + ], + "question_timestamps": "00:06:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "On your right-hand side.", + "C": "On your lef hand side.", + "D": "The kettle is above you, not on your left-hand side where you’re standing.", + "E": "The kettle is behind you instead of on your left-hand side near the counter." + }, + "answer_label": "C", + "answer_text": "On your lef hand side.", + "q_id": "P09_101_8", + "correct_answer": "On your lef hand side.", + "original_options": { + "A": "In front of you.", + "B": "On your right-hand side.", + "C": "On your lef hand side.", + "D": "The kettle is above you, not on your left-hand side where you’re standing.", + "E": "The kettle is behind you instead of on your left-hand side near the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + } +] diff --git a/QAs/P09_104-mcq.json b/QAs/P09_104-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..32f2594c973f3ea71f53c255523754b6ca5bce36 --- /dev/null +++ b/QAs/P09_104-mcq.json @@ -0,0 +1,713 @@ +[ + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_5", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the olive oil directly?", + "answer": "You can.", + "question_chinese": "我可以直接拿到橄榄油吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "Put down the olive oil.", + "timestamps": { + "start": "00:04:21", + "end": "00:04:25" + } + }, + { + "description": "Take the olive oil.", + "timestamps": { + "start": "00:04:16", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:04:25", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "P09_104_5", + "correct_answer": "You can.", + "original_options": { + "A": "You can't.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_6", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the olive oil directly?", + "answer": "You can't.", + "question_chinese": "我可以直接拿到橄榄油吗", + "answer_chinese": "你不可以", + "evidence": [ + { + "description": "You turned to the other side of the kitchen.", + "timestamps": { + "start": "00:06:14", + "end": "00:06:19" + } + } + ], + "question_timestamps": "00:06:21", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't.", + "B": "You can." + }, + "answer_label": "A", + "answer_text": "You can't.", + "q_id": "P09_104_6", + "correct_answer": "You can't.", + "original_options": { + "A": "You can't.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_7", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the eggs directly?", + "answer": "You can get the eggs directly.", + "question_chinese": "我可以直接拿到鸡蛋吗", + "answer_chinese": "你可以直接拿到鸡蛋", + "evidence": [ + { + "description": "You can get the eggs directly.", + "timestamps": { + "start": "00:06:02", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:06:14", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can get the eggs directly.", + "B": "You cannot get the eggs directly." + }, + "answer_label": "A", + "answer_text": "You can get the eggs directly.", + "q_id": "P09_104_7", + "correct_answer": "You can get the eggs directly.", + "original_options": { + "A": "You can get the eggs directly.", + "B": "You cannot get the eggs directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_8", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the eggs directly?", + "answer": "You can get the eggs directly.", + "question_chinese": "我可以直接拿到鸡蛋吗", + "answer_chinese": "你可以直接拿到鸡蛋", + "evidence": [ + { + "description": "You can get the eggs directly.", + "timestamps": { + "start": "00:06:02", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can get the eggs directly.", + "B": "You cannot get the eggs directly." + }, + "answer_label": "A", + "answer_text": "You can get the eggs directly.", + "q_id": "P09_104_8", + "correct_answer": "You can get the eggs directly.", + "original_options": { + "A": "You can get the eggs directly.", + "B": "You cannot get the eggs directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I wash the pot?", + "answer": "You have washed the pot.", + "question_chinese": "我洗过锅了吗", + "answer_chinese": "你洗过锅了", + "evidence": [ + { + "description": "You have washed the pot.", + "timestamps": { + "start": "00:00:13", + "end": "00:01:18" + } + } + ], + "question_timestamps": "00:02:14", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You have washed the pot.", + "B": "You have not washed the pot." + }, + "answer_label": "A", + "answer_text": "You have washed the pot.", + "q_id": "P09_104_3", + "correct_answer": "You have washed the pot.", + "original_options": { + "A": "You have washed the pot.", + "B": "You have not washed the pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_4", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I wash the pot?", + "answer": "You have washed the pot.", + "question_chinese": "我洗过锅了吗", + "answer_chinese": "你洗过锅了", + "evidence": [ + { + "description": "You have washed the pot.", + "timestamps": { + "start": "00:00:13", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You have washed the pot.", + "B": "You have not washed the pot." + }, + "answer_label": "A", + "answer_text": "You have washed the pot.", + "q_id": "P09_104_4", + "correct_answer": "You have washed the pot.", + "original_options": { + "A": "You have washed the pot.", + "B": "You have not washed the pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_1_T3", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What did I take out of the refrigerator just now?", + "answer": "You have now taken a box of strawberries out of the fridge, and the fridge is behind you.", + "question_chinese": "我现在从冰箱里拿出了什么？", + "answer_chinese": "你现在从冰箱里拿出了一盒草莓，冰箱在你的后面。", + "evidence": [ + { + "description": "You have taken out a box of strawberries from the refrigerator now.", + "timestamps": { + "start": "00:01:36", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:01:49", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You have now taken out a half-moon-shaped sausage from the fridge, and the fridge is behind you.", + "B": "You have now taken a box of strawberries out of the fridge, and the fridge is behind you." + }, + "answer_label": "B", + "answer_text": "You have now taken a box of strawberries out of the fridge, and the fridge is behind you.", + "q_id": "P09_104_1_T3", + "correct_answer": "You have now taken a box of strawberries out of the fridge, and the fridge is behind you.", + "original_options": { + "A": "You have now taken out a half-moon-shaped sausage from the fridge, and the fridge is behind you.", + "B": "You have now taken a box of strawberries out of the fridge, and the fridge is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_1_T4", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What did I take out of the refrigerator just now?", + "answer": "You have now taken out a half-moon-shaped sausage from the fridge, and the fridge is behind you.", + "question_chinese": "我现在从冰箱里拿出了什么？", + "answer_chinese": "你现在从冰箱里拿出了半圆形的香肠，冰箱在你的后面。", + "evidence": [ + { + "description": "Now you have taken out a semicircle sausage from the refrigerator.", + "timestamps": { + "start": "00:02:17", + "end": "00:02:19" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You have now taken out a half-moon-shaped sausage from the fridge, and the fridge is behind you.", + "B": "You have now taken a box of strawberries out of the fridge, and the fridge is behind you." + }, + "answer_label": "A", + "answer_text": "You have now taken out a half-moon-shaped sausage from the fridge, and the fridge is behind you.", + "q_id": "P09_104_1_T4", + "correct_answer": "You have now taken out a half-moon-shaped sausage from the fridge, and the fridge is behind you.", + "original_options": { + "A": "You have now taken out a half-moon-shaped sausage from the fridge, and the fridge is behind you.", + "B": "You have now taken a box of strawberries out of the fridge, and the fridge is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_4_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the eggs right now?", + "answer": "Right now you can directly reach the eggs, and the eggs are in front of you.", + "question_chinese": "我现在可以直接拿到鸡蛋吗？", + "answer_chinese": "你现在可以直接拿到鸡蛋，鸡蛋在你的正前方。", + "evidence": [ + { + "description": "You can directly reach for the egg, which is right in front of you.", + "timestamps": { + "start": "00:06:02", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:06:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can directly reach the eggs, and the eggs are in front of you.", + "B": "Right now you cannot directly reach the eggs, and the eggs are behind you." + }, + "answer_label": "A", + "answer_text": "Right now you can directly reach the eggs, and the eggs are in front of you.", + "q_id": "P09_104_4_T1", + "correct_answer": "Right now you can directly reach the eggs, and the eggs are in front of you.", + "original_options": { + "A": "Right now you can directly reach the eggs, and the eggs are in front of you.", + "B": "Right now you cannot directly reach the eggs, and the eggs are behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_4_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the eggs right now?", + "answer": "Right now you cannot directly reach the eggs, and the eggs are behind you.", + "question_chinese": "我现在可以直接拿到鸡蛋吗？", + "answer_chinese": "你现在不可以直接拿到鸡蛋，鸡蛋在你的后面。", + "evidence": [ + { + "description": "You can't get the eggs directly. You turn around and come to the table. The egg is behind you.", + "timestamps": { + "start": "00:06:16", + "end": "00:06:27" + } + } + ], + "question_timestamps": "00:06:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you can directly reach the eggs, and the eggs are behind you.", + "B": "Right now you cannot directly reach the eggs, and the eggs are behind you." + }, + "answer_label": "B", + "answer_text": "Right now you cannot directly reach the eggs, and the eggs are behind you.", + "q_id": "P09_104_4_T2", + "correct_answer": "Right now you cannot directly reach the eggs, and the eggs are behind you.", + "original_options": { + "A": "Right now you can directly reach the eggs, and the eggs are behind you.", + "B": "Right now you cannot directly reach the eggs, and the eggs are behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_3_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Am I close to the stove right now, or far from it?", + "answer": "You are close to the stove right now, and the stove is in front of you.", + "question_chinese": "我现在离灶台远/近？", + "answer_chinese": "你现在离灶台近，灶台在你的正前方。", + "evidence": [ + { + "description": "You are now close to the stove, which is directly in front of you.", + "timestamps": { + "start": "00:04:21", + "end": "00:04:25" + } + } + ], + "question_timestamps": "00:04:27", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are close to the stove right now, and the stove is in front of you.", + "B": "You are far from the stove right now, and the stove is in front of you." + }, + "answer_label": "A", + "answer_text": "You are close to the stove right now, and the stove is in front of you.", + "q_id": "P09_104_3_T1", + "correct_answer": "You are close to the stove right now, and the stove is in front of you.", + "original_options": { + "A": "You are close to the stove right now, and the stove is in front of you.", + "B": "You are far from the stove right now, and the stove is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_3_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Am I close to the stove right now, or far from it?", + "answer": "You are far from the stove right now, and the stove is behind you.", + "question_chinese": "我现在离灶台远/近？", + "answer_chinese": "你现在离灶台远，灶台在你的后面。", + "evidence": [ + { + "description": "You are currently far from the stove, which is behind you.", + "timestamps": { + "start": "00:04:50", + "end": "00:04:53" + } + } + ], + "question_timestamps": "00:04:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are far from the stove right now, and the stove is behind you.", + "B": "You are close to the stove right now, and the stove is behind you." + }, + "answer_label": "A", + "answer_text": "You are far from the stove right now, and the stove is behind you.", + "q_id": "P09_104_3_T2", + "correct_answer": "You are far from the stove right now, and the stove is behind you.", + "original_options": { + "A": "You are far from the stove right now, and the stove is behind you.", + "B": "You are close to the stove right now, and the stove is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands now?", + "answer": "Can wash hands.", + "question_chinese": "我现在能不能洗到手？", + "answer_chinese": "可以洗到手", + "evidence": [ + { + "description": "You can wash your hands next to the pool.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:02:34", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Cannot wash hands.", + "B": "Can wash hands." + }, + "answer_label": "B", + "answer_text": "Can wash hands.", + "q_id": "P09_104_1", + "correct_answer": "Can wash hands.", + "original_options": { + "A": "Cannot wash hands.", + "B": "Can wash hands." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands now?", + "answer": "You can wash your hands.", + "question_chinese": "我现在能不能洗到手？", + "answer_chinese": "你可以洗到手。", + "evidence": [ + { + "description": "You can wash your hands next to the pool.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:04:04", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can wash your hands.", + "B": "You cannot wash your hands." + }, + "answer_label": "A", + "answer_text": "You can wash your hands.", + "q_id": "P09_104_2", + "correct_answer": "You can wash your hands.", + "original_options": { + "A": "You can wash your hands.", + "B": "You cannot wash your hands." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_2_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the stove relative to me?", + "answer": "The stove is in front of you.", + "question_chinese": "灶台在相对于我的什么位置、", + "answer_chinese": "灶台在相对于你的正前方。", + "evidence": [ + { + "description": "The stove is in front of you.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The stove is in front of you.", + "B": "The stove is behind you." + }, + "answer_label": "A", + "answer_text": "The stove is in front of you.", + "q_id": "P09_104_2_T1", + "correct_answer": "The stove is in front of you.", + "original_options": { + "A": "The stove is in front of you.", + "B": "The stove is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P09_104", + "video_length": "00:09:13", + "qid": "P09_104_2_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the stove relative to me?", + "answer": "The stove is to your left.", + "question_chinese": "灶台在相对于我的什么位置、", + "answer_chinese": "灶台在相对于你的左边。", + "evidence": [ + { + "description": "You turned around and the stove became on your left.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The stove is to your right.", + "B": "The stove is to your left." + }, + "answer_label": "B", + "answer_text": "The stove is to your left.", + "q_id": "P09_104_2_T2", + "correct_answer": "The stove is to your left.", + "original_options": { + "A": "The stove is to your right.", + "B": "The stove is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/P09_106-mcq.json b/QAs/P09_106-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d316b4631a5229e3fc4ef52c21e3f64f0d87b8b4 --- /dev/null +++ b/QAs/P09_106-mcq.json @@ -0,0 +1,799 @@ +[ + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_7_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the microwave relative to me?", + "answer": "The microwave is above your head.", + "question_chinese": "微波炉在我的什么位置呢？", + "answer_chinese": "微波炉在你头顶的上方。", + "evidence": [ + { + "description": "You have used the microwave oven.", + "timestamps": { + "start": "00:00:54", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The microwave is above your head.", + "B": "The microwave is below your head.", + "C": "The microwave is behind you.", + "D": "The microwave is in front of you at chest height, not positioned above your head.", + "E": "The microwave is off to your right side near eye level, rather than directly overhead." + }, + "answer_label": "A", + "answer_text": "The microwave is above your head.", + "q_id": "P09_106_7_T2", + "correct_answer": "The microwave is above your head.", + "original_options": { + "A": "The microwave is above your head.", + "B": "The microwave is below your head.", + "C": "The microwave is behind you.", + "D": "The microwave is in front of you at chest height, not positioned above your head.", + "E": "The microwave is off to your right side near eye level, rather than directly overhead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 15 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_7_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the microwave relative to me?", + "answer": "The microwave is to your back-right.", + "question_chinese": "微波炉在我的什么位置呢？", + "answer_chinese": "微波炉在你的右后方。", + "evidence": [ + { + "description": "You have used the microwave oven.", + "timestamps": { + "start": "00:00:54", + "end": "00:01:05" + } + }, + { + "description": "You have used the microwave oven.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:24" + } + } + ], + "question_timestamps": "00:02:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-left.", + "B": "To your front-right.", + "C": "The microwave is positioned to your front-left as you face forward now.", + "D": "The microwave is directly above you, centered over your head when you look up.", + "E": "The microwave is to your back-right." + }, + "answer_label": "E", + "answer_text": "The microwave is to your back-right.", + "q_id": "P09_106_7_T1", + "correct_answer": "The microwave is to your back-right.", + "original_options": { + "A": "To your back-left.", + "B": "To your front-right.", + "C": "The microwave is positioned to your front-left as you face forward now.", + "D": "The microwave is directly above you, centered over your head when you look up.", + "E": "The microwave is to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 14 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_9", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the pot lid directly?", + "answer": "You can directly get the pot lid.", + "question_chinese": "我可以直接拿到锅盖么", + "answer_chinese": "你可以直接拿到锅盖", + "evidence": [ + { + "description": "You can directly get the pot lid.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can directly get the pot lid.", + "B": "You cannot directly get the pot lid." + }, + "answer_label": "A", + "answer_text": "You can directly get the pot lid.", + "q_id": "P09_106_9", + "correct_answer": "You can directly get the pot lid.", + "original_options": { + "A": "You can directly get the pot lid.", + "B": "You cannot directly get the pot lid." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_10", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the pot lid directly?", + "answer": "You can't.", + "question_chinese": "我可以直接拿到锅盖么", + "answer_chinese": "你不能", + "evidence": [ + { + "description": "Position of the pot lid.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:17" + } + }, + { + "description": "You are very far from the pot lid where you are.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:23", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You can't." + }, + "answer_label": "B", + "answer_text": "You can't.", + "q_id": "P09_106_10", + "correct_answer": "You can't.", + "original_options": { + "A": "You can.", + "B": "You can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_3_T2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach the sink to wash my hands?", + "answer": "No, you can’t—the sink is pretty far away, to your back-right.", + "question_chinese": "我现在可以洗到手吗？", + "answer_chinese": "你不能，洗手池在你右后方比较远的地方。", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:01:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t—the sink is pretty far away, to your back-right.", + "B": "Yes, you can—the sink is right below you." + }, + "answer_label": "A", + "answer_text": "No, you can’t—the sink is pretty far away, to your back-right.", + "q_id": "P09_106_3_T2", + "correct_answer": "No, you can’t—the sink is pretty far away, to your back-right.", + "original_options": { + "A": "No, you can’t—the sink is pretty far away, to your back-right.", + "B": "Yes, you can—the sink is right below you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_3_T1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach the sink to wash my hands?", + "answer": "Yes, you can—the sink is right below you.", + "question_chinese": "我现在可以洗到手吗？", + "answer_chinese": "你可以的，洗手池就在你的下方。", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t—the sink is pretty far away, to your back-right.", + "B": "Yes, you can—the sink is right below you." + }, + "answer_label": "B", + "answer_text": "Yes, you can—the sink is right below you.", + "q_id": "P09_106_3_T1", + "correct_answer": "Yes, you can—the sink is right below you.", + "original_options": { + "A": "No, you can’t—the sink is pretty far away, to your back-right.", + "B": "Yes, you can—the sink is right below you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_2_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there anything in my microwave right now?", + "answer": "The microwave is directly above your head, and there is food inside it.", + "question_chinese": "我的微波炉里现在有什么东西吗？", + "answer_chinese": "微波炉在你的头顶正上方，有食物在里面。", + "evidence": [ + { + "description": "Put the food in the microwave.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There’s nothing inside.", + "B": "It’s empty right now.", + "C": "The microwave is directly above your head, but there isn’t any food inside it.", + "D": "The microwave is above you on your right, and it’s empty with nothing placed inside.", + "E": "The microwave is directly above your head, and there is food inside it." + }, + "answer_label": "E", + "answer_text": "The microwave is directly above your head, and there is food inside it.", + "q_id": "P09_106_2_T1", + "correct_answer": "The microwave is directly above your head, and there is food inside it.", + "original_options": { + "A": "There’s nothing inside.", + "B": "It’s empty right now.", + "C": "The microwave is directly above your head, but there isn’t any food inside it.", + "D": "The microwave is above you on your right, and it’s empty with nothing placed inside.", + "E": "The microwave is directly above your head, and there is food inside it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 16 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_2_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there anything in my microwave right now?", + "answer": "There’s nothing in the microwave now, and it is above you on your left.", + "question_chinese": "我的微波炉里现在有什么东西吗？", + "answer_chinese": "现在微波炉里已经没有东西了，它就在你的左上方。", + "evidence": [ + { + "description": "There is food in the microwave.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:06" + } + }, + { + "description": "The food is taken out.", + "timestamps": { + "start": "00:02:18", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:02:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There’s nothing in the microwave now, and it is above you on your left.", + "B": "There’s food inside.", + "C": "It’s above you on your right.", + "D": "There’s nothing in the microwave, but it’s positioned below you on your left side.", + "E": "There is food in the microwave right now, and it is located above you on your left." + }, + "answer_label": "A", + "answer_text": "There’s nothing in the microwave now, and it is above you on your left.", + "q_id": "P09_106_2_T2", + "correct_answer": "There’s nothing in the microwave now, and it is above you on your left.", + "original_options": { + "A": "There’s nothing in the microwave now, and it is above you on your left.", + "B": "There’s food inside.", + "C": "It’s above you on your right.", + "D": "There’s nothing in the microwave, but it’s positioned below you on your left side.", + "E": "There is food in the microwave right now, and it is located above you on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 16, + 17 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_11", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many scoops of rice did I take?", + "answer": "five times.", + "question_chinese": "我舀了多少勺米饭", + "answer_chinese": "5勺", + "evidence": [ + { + "description": "One.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:19" + } + }, + { + "description": "Two.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:24" + } + }, + { + "description": "Three.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:31" + } + }, + { + "description": "Four.", + "timestamps": { + "start": "00:00:31", + "end": "00:00:34" + } + }, + { + "description": "Five.", + "timestamps": { + "start": "00:00:38", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three times.", + "B": "Four times.", + "C": "You took rice exactly two times, making two separate scoops into the container.", + "D": "You scooped the rice six times in total, repeating the same motion for each scoop.", + "E": "five times." + }, + "answer_label": "E", + "answer_text": "five times.", + "q_id": "P09_106_11", + "correct_answer": "five times.", + "original_options": { + "A": "Three times.", + "B": "Four times.", + "C": "You took rice exactly two times, making two separate scoops into the container.", + "D": "You scooped the rice six times in total, repeating the same motion for each scoop.", + "E": "five times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_12", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many scoops of rice did I take?", + "answer": "seven times.", + "question_chinese": "我舀了多少勺米饭", + "answer_chinese": "7勺", + "evidence": [ + { + "description": "One.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:19" + } + }, + { + "description": "Two.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:24" + } + }, + { + "description": "Three.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:31" + } + }, + { + "description": "Four.", + "timestamps": { + "start": "00:00:31", + "end": "00:00:34" + } + }, + { + "description": "Five.", + "timestamps": { + "start": "00:00:38", + "end": "00:00:39" + } + }, + { + "description": "Six.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:42" + } + }, + { + "description": "Seven.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Six times.", + "B": "Eight times.", + "C": "You took five scoops of rice in total during the serving process at the counter.", + "D": "You took nine scoops overall before you stopped serving the rice into the container.", + "E": "seven times." + }, + "answer_label": "E", + "answer_text": "seven times.", + "q_id": "P09_106_12", + "correct_answer": "seven times.", + "original_options": { + "A": "Six times.", + "B": "Eight times.", + "C": "You took five scoops of rice in total during the serving process at the counter.", + "D": "You took nine scoops overall before you stopped serving the rice into the container.", + "E": "seven times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_1_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory to the refrigerator?", + "answer": "You just turned left, and then you reached the fridge.", + "question_chinese": "我走到冰箱的移动轨迹是？", + "answer_chinese": "你刚刚向左转身，然后就到达了冰箱了。", + "evidence": [ + { + "description": "Open the refrigerator door.", + "timestamps": { + "start": "00:01:06", + "end": "00:01:21" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right to reach it.", + "B": "You went straight to the fridge.", + "C": "You turned right, walked forward a bit, and then reached the refrigerator.", + "D": "You just turned left, and then you reached the fridge.", + "E": "You turned around first, then moved a couple steps before arriving at the fridge." + }, + "answer_label": "D", + "answer_text": "You just turned left, and then you reached the fridge.", + "q_id": "P09_106_1_T1", + "correct_answer": "You just turned left, and then you reached the fridge.", + "original_options": { + "A": "You turned right to reach it.", + "B": "You went straight to the fridge.", + "C": "You turned right, walked forward a bit, and then reached the refrigerator.", + "D": "You just turned left, and then you reached the fridge.", + "E": "You turned around first, then moved a couple steps before arriving at the fridge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 14 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_1_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory to the refrigerator?", + "answer": "You just turned left, then went straight and reached the fridge.", + "question_chinese": "我走到冰箱的移动轨迹是？", + "answer_chinese": "你刚刚向左转身，直走就到达了冰箱了。", + "evidence": [ + { + "description": "Open the refrigerator door.", + "timestamps": { + "start": "00:01:06", + "end": "00:01:21" + } + } + ], + "question_timestamps": "00:01:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You just turned right first.", + "B": "You reached it without going straight.", + "C": "You turned left and immediately reached the fridge without walking straight afterward.", + "D": "You turned right, went straight for a moment, and then reached the refrigerator.", + "E": "You just turned left, then went straight and reached the fridge." + }, + "answer_label": "E", + "answer_text": "You just turned left, then went straight and reached the fridge.", + "q_id": "P09_106_1_T2", + "correct_answer": "You just turned left, then went straight and reached the fridge.", + "original_options": { + "A": "You just turned right first.", + "B": "You reached it without going straight.", + "C": "You turned left and immediately reached the fridge without walking straight afterward.", + "D": "You turned right, went straight for a moment, and then reached the refrigerator.", + "E": "You just turned left, then went straight and reached the fridge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 13 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_4_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where are the bowls in the cabinet relative to me?", + "answer": "The bowl in the cabinet is above your left hand.", + "question_chinese": "柜子里的碗相对于我的方位是？", + "answer_chinese": "柜子里的碗在你的左手边的上方。", + "evidence": [ + { + "description": "Six bowls.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above your right hand.", + "B": "The bowl in the cabinet is above your left hand.", + "C": "Below your left hand.", + "D": "The bowl in the cabinet is below your right hand as you face it.", + "E": "The bowl in the cabinet is above you but positioned on your right side." + }, + "answer_label": "B", + "answer_text": "The bowl in the cabinet is above your left hand.", + "q_id": "P09_106_4_T1", + "correct_answer": "The bowl in the cabinet is above your left hand.", + "original_options": { + "A": "Above your right hand.", + "B": "The bowl in the cabinet is above your left hand.", + "C": "Below your left hand.", + "D": "The bowl in the cabinet is below your right hand as you face it.", + "E": "The bowl in the cabinet is above you but positioned on your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "P09_106", + "video_length": "00:02:29", + "qid": "P09_106_4_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where are the bowls in the cabinet relative to me?", + "answer": "The bowl in the cabinet is above you, on your left side.", + "question_chinese": "柜子里的碗相对于我的方位是？", + "answer_chinese": "柜子里的碗在你左侧的上方。", + "evidence": [ + { + "description": "Six bowls.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:01:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above you, on your right side.", + "B": "Below you, on your left side.", + "C": "The bowl in the cabinet is above you, on your left side.", + "D": "The bowl in the cabinet is above your left hand rather than above you on a side.", + "E": "The bowl in the cabinet is below your right hand, closer to your body than eye level." + }, + "answer_label": "C", + "answer_text": "The bowl in the cabinet is above you, on your left side.", + "q_id": "P09_106_4_T2", + "correct_answer": "The bowl in the cabinet is above you, on your left side.", + "original_options": { + "A": "Above you, on your right side.", + "B": "Below you, on your left side.", + "C": "The bowl in the cabinet is above you, on your left side.", + "D": "The bowl in the cabinet is above your left hand rather than above you on a side.", + "E": "The bowl in the cabinet is below your right hand, closer to your body than eye level." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + } +] diff --git a/QAs/P11_101-mcq.json b/QAs/P11_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..88596c8125889e70c76d9c6098a07b3d8f3ae81c --- /dev/null +++ b/QAs/P11_101-mcq.json @@ -0,0 +1,966 @@ +[ + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was the first thing I did when I just entered the room?", + "answer": "You closed the cabinet.", + "question_chinese": "我刚进入房间第一件事做了什么", + "answer_chinese": "你关闭了橱柜", + "evidence": [ + { + "description": "Close the cabinet.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You opened the drawer.", + "B": "You closed the cabinet.", + "C": "You checked your phone.", + "D": "You set down your bag first before doing anything else in the room.", + "E": "You rearranged the furniture as the first thing right after you entered." + }, + "answer_label": "B", + "answer_text": "You closed the cabinet.", + "q_id": "P11_101_7", + "correct_answer": "You closed the cabinet.", + "original_options": { + "A": "You opened the drawer.", + "B": "You closed the cabinet.", + "C": "You checked your phone.", + "D": "You set down your bag first before doing anything else in the room.", + "E": "You rearranged the furniture as the first thing right after you entered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 12 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was the first thing I did when I just entered the room?", + "answer": "You closed the cabinet.", + "question_chinese": "我刚进入房间第一件事做了什么", + "answer_chinese": "你关闭了橱柜", + "evidence": [ + { + "description": "Close the cabinet.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:02:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You opened the drawer.", + "B": "You adjusted the curtains.", + "C": "You placed your shoes by the door as the very first thing upon entering.", + "D": "You closed the cabinet.", + "E": "You rearranged the furniture first before doing anything else in the room." + }, + "answer_label": "D", + "answer_text": "You closed the cabinet.", + "q_id": "P11_101_8", + "correct_answer": "You closed the cabinet.", + "original_options": { + "A": "You opened the drawer.", + "B": "You adjusted the curtains.", + "C": "You placed your shoes by the door as the very first thing upon entering.", + "D": "You closed the cabinet.", + "E": "You rearranged the furniture first before doing anything else in the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 12 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_5", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash directly up to my hands?", + "answer": "You can.", + "question_chinese": "我可以直接洗到手吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "The pool is right in front of you.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:51", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "P11_101_5", + "correct_answer": "You can.", + "original_options": { + "A": "You can't.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_6", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash directly up to my hands?", + "answer": "You can't.", + "question_chinese": "我可以直接洗到手吗", + "answer_chinese": "你不能", + "evidence": [ + { + "description": "The location of the pool.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:48" + } + }, + { + "description": "Your current location.", + "timestamps": { + "start": "00:01:27", + "end": "00:01:28" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't.", + "B": "You can." + }, + "answer_label": "A", + "answer_text": "You can't.", + "q_id": "P11_101_6", + "correct_answer": "You can't.", + "original_options": { + "A": "You can't.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_7_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I take out of the cabinet?", + "answer": "You took out a pot; it is in front of you.", + "question_chinese": "我在柜子中拿出了什么？", + "answer_chinese": "你拿出了一个锅，它在你前面。", + "evidence": [ + { + "description": "You took out a pot.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took out a pot; it is in front of you.", + "B": "You took out a lid.", + "C": "You took out a pan.", + "D": "You took out a kettle and set it down behind you, not directly in front of you.", + "E": "You took out a saucepan and placed it to your right, rather than leaving it in front of you." + }, + "answer_label": "A", + "answer_text": "You took out a pot; it is in front of you.", + "q_id": "P11_101_7_T2", + "correct_answer": "You took out a pot; it is in front of you.", + "original_options": { + "A": "You took out a pot; it is in front of you.", + "B": "You took out a lid.", + "C": "You took out a pan.", + "D": "You took out a kettle and set it down behind you, not directly in front of you.", + "E": "You took out a saucepan and placed it to your right, rather than leaving it in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_7_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I take out of the cabinet?", + "answer": "You took out a heating rod; it is in front of you.", + "question_chinese": "我在柜子中拿出了什么？", + "answer_chinese": "你拿出了一个加热棒，它在你前面。", + "evidence": [ + { + "description": "You took out a heating rod.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took out a whisk.", + "B": "You took out a saucepan.", + "C": "You took out a kettle and placed it to your left instead of keeping it in front of you.", + "D": "You took out a pot and set it down behind you, not in front of you as stated.", + "E": "You took out a heating rod; it is in front of you." + }, + "answer_label": "E", + "answer_text": "You took out a heating rod; it is in front of you.", + "q_id": "P11_101_7_T1", + "correct_answer": "You took out a heating rod; it is in front of you.", + "original_options": { + "A": "You took out a whisk.", + "B": "You took out a saucepan.", + "C": "You took out a kettle and placed it to your left instead of keeping it in front of you.", + "D": "You took out a pot and set it down behind you, not in front of you as stated.", + "E": "You took out a heating rod; it is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 18 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I take out of the cabinet?", + "answer": "You took out a heating rod.", + "question_chinese": "我在柜子中拿出了什么", + "answer_chinese": "你拿出了一个加热棒", + "evidence": [ + { + "description": "You took out a heating rod.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took out a frying pan.", + "B": "You took out a wooden spoon.", + "C": "You took out a heating rod.", + "D": "You took out a baking sheet from the cabinet and held it briefly before putting it down.", + "E": "You took out a set of tools from the cabinet, not a single heating rod." + }, + "answer_label": "C", + "answer_text": "You took out a heating rod.", + "q_id": "P11_101_11", + "correct_answer": "You took out a heating rod.", + "original_options": { + "A": "You took out a frying pan.", + "B": "You took out a wooden spoon.", + "C": "You took out a heating rod.", + "D": "You took out a baking sheet from the cabinet and held it briefly before putting it down.", + "E": "You took out a set of tools from the cabinet, not a single heating rod." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 15 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_12", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I take out of the cabinet?", + "answer": "You took out a heating rod.", + "question_chinese": "我在柜子中拿出了什么", + "answer_chinese": "你拿出了一个加热棒", + "evidence": [ + { + "description": "You took out a heating rod.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:01:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took out a heating rod.", + "B": "You took out a cooking pot.", + "C": "You took out a cutting board.", + "D": "You took out a baking sheet from the cabinet, not the heating rod mentioned.", + "E": "You took out an electric mixer from the cabinet instead of removing a heating rod." + }, + "answer_label": "A", + "answer_text": "You took out a heating rod.", + "q_id": "P11_101_12", + "correct_answer": "You took out a heating rod.", + "original_options": { + "A": "You took out a heating rod.", + "B": "You took out a cooking pot.", + "C": "You took out a cutting board.", + "D": "You took out a baking sheet from the cabinet, not the heating rod mentioned.", + "E": "You took out an electric mixer from the cabinet instead of removing a heating rod." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_5_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I directly reach the faucet?", + "answer": "You can directly reach the faucet; it is in front of you.", + "question_chinese": "我能直接接触到水龙头吗？", + "answer_chinese": "你能直接接触到水龙头，它在你前面。", + "evidence": [ + { + "description": "You can reach the faucet directly.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can directly reach the faucet; it is in front of you.", + "B": "You can't reach the faucet.", + "C": "You can reach it on your left.", + "D": "You can directly reach the faucet, but it is to your right rather than in front of you.", + "E": "You cannot directly reach the faucet even though it sits in front of you at the sink." + }, + "answer_label": "A", + "answer_text": "You can directly reach the faucet; it is in front of you.", + "q_id": "P11_101_5_T1", + "correct_answer": "You can directly reach the faucet; it is in front of you.", + "original_options": { + "A": "You can directly reach the faucet; it is in front of you.", + "B": "You can't reach the faucet.", + "C": "You can reach it on your left.", + "D": "You can directly reach the faucet, but it is to your right rather than in front of you.", + "E": "You cannot directly reach the faucet even though it sits in front of you at the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 18, + 17 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_5_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I directly reach the faucet?", + "answer": "You cannot directly reach the faucet; it is to your left.", + "question_chinese": "我能直接接触到水龙头吗？", + "answer_chinese": "你不能直接接触到水龙头，它在你左边。", + "evidence": [ + { + "description": "You can not reach the faucet directly.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can reach the faucet.", + "B": "You cannot directly reach the faucet; it is to your left.", + "C": "It is to your right.", + "D": "You cannot directly reach the faucet because it is behind you, not positioned to your left.", + "E": "You can directly reach the faucet even though it is to your left, unlike the stated limitation." + }, + "answer_label": "B", + "answer_text": "You cannot directly reach the faucet; it is to your left.", + "q_id": "P11_101_5_T2", + "correct_answer": "You cannot directly reach the faucet; it is to your left.", + "original_options": { + "A": "You can reach the faucet.", + "B": "You cannot directly reach the faucet; it is to your left.", + "C": "It is to your right.", + "D": "You cannot directly reach the faucet because it is behind you, not positioned to your left.", + "E": "You can directly reach the faucet even though it is to your left, unlike the stated limitation." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_3", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I tested the waters?", + "answer": "Twice.", + "question_chinese": "我试了多少次水", + "answer_chinese": "两次", + "evidence": [ + { + "description": "Two.", + "timestamps": { + "start": "00:00:31", + "end": "00:00:36" + } + }, + { + "description": "One.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:49", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "Four times in total, counting each separate check of the water temperature.", + "E": "Five times overall, with each brief touch counted as a separate test." + }, + "answer_label": "A", + "answer_text": "Twice.", + "q_id": "P11_101_3", + "correct_answer": "Twice.", + "original_options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "Four times in total, counting each separate check of the water temperature.", + "E": "Five times overall, with each brief touch counted as a separate test." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 12 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_4", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I tested the waters?", + "answer": "Three times.", + "question_chinese": "我试了多少次水", + "answer_chinese": "三次", + "evidence": [ + { + "description": "Two.", + "timestamps": { + "start": "00:00:31", + "end": "00:00:36" + } + }, + { + "description": "One.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:19" + } + }, + { + "description": "Three.", + "timestamps": { + "start": "00:00:55", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Three times.", + "C": "Four times.", + "D": "You tested the waters twice before moving on to the next step afterward.", + "E": "You tested the waters four times, repeating the check after each small adjustment." + }, + "answer_label": "B", + "answer_text": "Three times.", + "q_id": "P11_101_4", + "correct_answer": "Three times.", + "original_options": { + "A": "Two times.", + "B": "Three times.", + "C": "Four times.", + "D": "You tested the waters twice before moving on to the next step afterward.", + "E": "You tested the waters four times, repeating the check after each small adjustment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_1_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls are in front of me?", + "answer": "There are 3 bowls in front of you; they are in front of you.", + "question_chinese": "我面前有几个碗？", + "answer_chinese": "你面前有3个碗，它们在你前面。", + "evidence": [ + { + "description": "Three bowls.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two bowls.", + "B": "Four bowls.", + "C": "There are 3 bowls in front of you; they are in front of you.", + "D": "There are 2 bowls in front of you; they are in front of you.", + "E": "There are 4 bowls in front of you; they are in front of you." + }, + "answer_label": "C", + "answer_text": "There are 3 bowls in front of you; they are in front of you.", + "q_id": "P11_101_1_T2", + "correct_answer": "There are 3 bowls in front of you; they are in front of you.", + "original_options": { + "A": "Two bowls.", + "B": "Four bowls.", + "C": "There are 3 bowls in front of you; they are in front of you.", + "D": "There are 2 bowls in front of you; they are in front of you.", + "E": "There are 4 bowls in front of you; they are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_1_T3", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls are in front of me?", + "answer": "There are 5 bowls in front of you; they are in front of you.", + "question_chinese": "我面前有几个碗？", + "answer_chinese": "你面前有5个碗，它们在你前面。", + "evidence": [ + { + "description": "Five bowls.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four bowls.", + "B": "Six bowls.", + "C": "There are 4 bowls in front of you; they are in front of you.", + "D": "There are 5 bowls in front of you; they are in front of you.", + "E": "There are 6 bowls in front of you; they are in front of you." + }, + "answer_label": "D", + "answer_text": "There are 5 bowls in front of you; they are in front of you.", + "q_id": "P11_101_1_T3", + "correct_answer": "There are 5 bowls in front of you; they are in front of you.", + "original_options": { + "A": "Four bowls.", + "B": "Six bowls.", + "C": "There are 4 bowls in front of you; they are in front of you.", + "D": "There are 5 bowls in front of you; they are in front of you.", + "E": "There are 6 bowls in front of you; they are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "There are several bowls in the drawer.", + "answer": "Three bowls.", + "question_chinese": "抽屉里有几个碗", + "answer_chinese": "三个碗", + "evidence": [ + { + "description": "There are three bowls in the drawer.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:49", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two bowls.", + "B": "Four bowls.", + "C": "There are 2 bowls in the drawer when you look inside and check again carefully.", + "D": "There are 4 bowls in the drawer, grouped together in the same compartment.", + "E": "Three bowls." + }, + "answer_label": "E", + "answer_text": "Three bowls.", + "q_id": "P11_101_1", + "correct_answer": "Three bowls.", + "original_options": { + "A": "Two bowls.", + "B": "Four bowls.", + "C": "There are 2 bowls in the drawer when you look inside and check again carefully.", + "D": "There are 4 bowls in the drawer, grouped together in the same compartment.", + "E": "Three bowls." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 13 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "There are several bowls in the drawer.", + "answer": "Three bowls.", + "question_chinese": "抽屉里有几个碗", + "answer_chinese": "三个碗", + "evidence": [ + { + "description": "Three bowls.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:01:19", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two bowls.", + "B": "Five bowls.", + "C": "There are 2 bowls in the drawer when you check it and look again carefully.", + "D": "Three bowls.", + "E": "There are 5 bowls in the drawer, stacked or nested together in one place." + }, + "answer_label": "D", + "answer_text": "Three bowls.", + "q_id": "P11_101_2", + "correct_answer": "Three bowls.", + "original_options": { + "A": "Two bowls.", + "B": "Five bowls.", + "C": "There are 2 bowls in the drawer when you check it and look again carefully.", + "D": "Three bowls.", + "E": "There are 5 bowls in the drawer, stacked or nested together in one place." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_10", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cups are there by the sink?", + "answer": "One.", + "question_chinese": "洗碗池边有几个杯子？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There is a cup beside the sink.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are 2 cups by the sink when you look over and count them.", + "E": "There are 3 cups by the sink, sitting together along the edge." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "P11_101_10", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are 2 cups by the sink when you look over and count them.", + "E": "There are 3 cups by the sink, sitting together along the edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 12 + ] + }, + { + "video_uid": "P11_101", + "video_length": "00:03:04", + "qid": "P11_101_9", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cups are there by the sink?", + "answer": "There is a cup beside the sink.", + "question_chinese": "洗碗池边有几个杯子？", + "answer_chinese": "水槽旁边有一个杯子。", + "evidence": [ + { + "description": "There is a cup beside the sink.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:02:21", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is a cup beside the sink.", + "B": "There are two cups beside the sink.", + "C": "There is no cup beside the sink.", + "D": "There are two cups beside the sink, both sitting right next to it.", + "E": "There is no cup beside the sink; the area next to it is empty." + }, + "answer_label": "A", + "answer_text": "There is a cup beside the sink.", + "q_id": "P11_101_9", + "correct_answer": "There is a cup beside the sink.", + "original_options": { + "A": "There is a cup beside the sink.", + "B": "There are two cups beside the sink.", + "C": "There is no cup beside the sink.", + "D": "There are two cups beside the sink, both sitting right next to it.", + "E": "There is no cup beside the sink; the area next to it is empty." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 14 + ] + } +] diff --git a/QAs/P11_105-mcq.json b/QAs/P11_105-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..acb1000362d3e91b38d2630c0877d0d3ea15bf70 --- /dev/null +++ b/QAs/P11_105-mcq.json @@ -0,0 +1,316 @@ +[ + { + "video_uid": "P11_105", + "video_length": "00:05:33", + "qid": "P11_105_5", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the refrigerator?", + "answer": "Once.", + "question_chinese": "我开了几次冰箱？", + "answer_chinese": "一次", + "evidence": [ + { + "description": "Opening the refrigerator for the first time.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:11" + } + }, + { + "description": "The second time.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:02:31", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "You opened the refrigerator twice during the sequence, then continued with other tasks.", + "E": "You opened the refrigerator three times, returning to it repeatedly across the sequence." + }, + "answer_label": "B", + "answer_text": "Once.", + "q_id": "P11_105_5", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "You opened the refrigerator twice during the sequence, then continued with other tasks.", + "E": "You opened the refrigerator three times, returning to it repeatedly across the sequence." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 13 + ] + }, + { + "video_uid": "P11_105", + "video_length": "00:05:33", + "qid": "P11_105_6", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the refrigerator?", + "answer": "Twice.", + "question_chinese": "我开了几次冰箱？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "Opening the refrigerator for the first time.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:11" + } + }, + { + "description": "The second time.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Twice.", + "D": "You opened the refrigerator three times during the task, checking inside on separate occasions.", + "E": "You opened the refrigerator only once, then kept it closed for the rest of the time." + }, + "answer_label": "C", + "answer_text": "Twice.", + "q_id": "P11_105_6", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Twice.", + "D": "You opened the refrigerator three times during the task, checking inside on separate occasions.", + "E": "You opened the refrigerator only once, then kept it closed for the rest of the time." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 14, + 16 + ] + }, + { + "video_uid": "P11_105", + "video_length": "00:05:33", + "qid": "P11_105_13", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the oven", + "answer": "Once.", + "question_chinese": "我打开了几次烤炉？", + "answer_chinese": "一次", + "evidence": [ + { + "description": "The first time.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:02:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Three times.", + "C": "You opened the oven twice, opening it again after the first check.", + "D": "Once.", + "E": "You opened the oven four times in total, returning repeatedly to look inside." + }, + "answer_label": "D", + "answer_text": "Once.", + "q_id": "P11_105_13", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Three times.", + "C": "You opened the oven twice, opening it again after the first check.", + "D": "Once.", + "E": "You opened the oven four times in total, returning repeatedly to look inside." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 13 + ] + }, + { + "video_uid": "P11_105", + "video_length": "00:05:33", + "qid": "P11_105_14", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the oven", + "answer": "Once.", + "question_chinese": "我打开了几次烤炉？", + "answer_chinese": "一次", + "evidence": [ + { + "description": "The first time.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:05:33", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Four times.", + "C": "You opened the oven three times, closing it between each separate opening.", + "D": "You opened the oven five times overall, repeatedly opening it to check on it.", + "E": "Once." + }, + "answer_label": "E", + "answer_text": "Once.", + "q_id": "P11_105_14", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Four times.", + "C": "You opened the oven three times, closing it between each separate opening.", + "D": "You opened the oven five times overall, repeatedly opening it to check on it.", + "E": "Once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 14 + ] + }, + { + "video_uid": "P11_105", + "video_length": "00:05:33", + "qid": "P11_105_11", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I sharpened the knife?", + "answer": "You have sharpened a knife.", + "question_chinese": "我是否磨过刀", + "answer_chinese": "你磨过刀", + "evidence": [ + { + "description": "Sharpen the knife.", + "timestamps": { + "start": "00:01:39", + "end": "00:01:48" + } + } + ], + "question_timestamps": "00:02:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You have sharpened a knife.", + "B": "You have not sharpened a knife." + }, + "answer_label": "A", + "answer_text": "You have sharpened a knife.", + "q_id": "P11_105_11", + "correct_answer": "You have sharpened a knife.", + "original_options": { + "A": "You have sharpened a knife.", + "B": "You have not sharpened a knife." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P11_105", + "video_length": "00:05:33", + "qid": "P11_105_12", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I sharpened the knife?", + "answer": "You have sharpened a knife.", + "question_chinese": "我是否磨过刀", + "answer_chinese": "你已经磨好了一把刀", + "evidence": [ + { + "description": "Sharpen the knife.", + "timestamps": { + "start": "00:01:39", + "end": "00:01:48" + } + } + ], + "question_timestamps": "00:05:32", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You have sharpened a knife.", + "B": "You have not sharpened a knife." + }, + "answer_label": "A", + "answer_text": "You have sharpened a knife.", + "q_id": "P11_105_12", + "correct_answer": "You have sharpened a knife.", + "original_options": { + "A": "You have sharpened a knife.", + "B": "You have not sharpened a knife." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/P11_109-mcq.json b/QAs/P11_109-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4e0bbab2300cf6c28cea37fd3e4068140e04694d --- /dev/null +++ b/QAs/P11_109-mcq.json @@ -0,0 +1,1115 @@ +[ + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_3_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cups did I take out of the drawer?", + "answer": "You took eight cups out of the drawer in front of you.", + "question_chinese": "我从抽屉中拿出了多少个杯子？", + "answer_chinese": "你从你面前的抽屉里拿出八个杯子", + "evidence": [ + { + "description": "Take out eight cups.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Seven cups.", + "B": "You took eight cups out of the drawer in front of you.", + "C": "Nine cups.", + "D": "You took six cups out of the drawer in front of you, removing them one after another.", + "E": "You took ten cups out of the drawer in front of you, emptying most of the drawer." + }, + "answer_label": "B", + "answer_text": "You took eight cups out of the drawer in front of you.", + "q_id": "P11_109_3_T1", + "correct_answer": "You took eight cups out of the drawer in front of you.", + "original_options": { + "A": "Seven cups.", + "B": "You took eight cups out of the drawer in front of you.", + "C": "Nine cups.", + "D": "You took six cups out of the drawer in front of you, removing them one after another.", + "E": "You took ten cups out of the drawer in front of you, emptying most of the drawer." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_3_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cups did I take out of the drawer?", + "answer": "You took eight cups out of the drawer in front of you.", + "question_chinese": "我从抽屉中拿出了多少个杯子？", + "answer_chinese": "你从你面前的抽屉里拿出八个杯子。", + "evidence": [ + { + "description": "Take out eight cups.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Seven cups.", + "B": "Ten cups.", + "C": "You took eight cups out of the drawer in front of you.", + "D": "You took nine cups out of the drawer in front of you, pulling out one additional cup.", + "E": "You took six cups out of the drawer in front of you, stopping after removing only a few." + }, + "answer_label": "C", + "answer_text": "You took eight cups out of the drawer in front of you.", + "q_id": "P11_109_3_T2", + "correct_answer": "You took eight cups out of the drawer in front of you.", + "original_options": { + "A": "Seven cups.", + "B": "Ten cups.", + "C": "You took eight cups out of the drawer in front of you.", + "D": "You took nine cups out of the drawer in front of you, pulling out one additional cup.", + "E": "You took six cups out of the drawer in front of you, stopping after removing only a few." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_1_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes did I take out of the drawer?", + "answer": "You took five boxes out of the drawer in front of you.", + "question_chinese": "我从抽屉中拿出了多少个盒子？", + "answer_chinese": "你从你面前的抽屉里拿出了五个盒子。", + "evidence": [ + { + "description": "The first box.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "The second and third boxes.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:14" + } + }, + { + "description": "The fourth and fifth boxes.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four boxes.", + "B": "Six boxes.", + "C": "You took three boxes out of the drawer in front of you, removing only a small set.", + "D": "You took seven boxes out of the drawer in front of you, pulling out several more than that.", + "E": "You took five boxes out of the drawer in front of you." + }, + "answer_label": "E", + "answer_text": "You took five boxes out of the drawer in front of you.", + "q_id": "P11_109_1_T1", + "correct_answer": "You took five boxes out of the drawer in front of you.", + "original_options": { + "A": "Four boxes.", + "B": "Six boxes.", + "C": "You took three boxes out of the drawer in front of you, removing only a small set.", + "D": "You took seven boxes out of the drawer in front of you, pulling out several more than that.", + "E": "You took five boxes out of the drawer in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes did I take out of the drawer?", + "answer": "Seven.", + "question_chinese": "我从抽屉中拿出了多少个盒子？", + "answer_chinese": "七个。", + "evidence": [ + { + "description": "The first box.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "The second and third boxes.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:14" + } + }, + { + "description": "The fourth and fifth boxes.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:23" + } + }, + { + "description": "The seventh box.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "Eight.", + "C": "You took five boxes out of the drawer next to you on the counter.", + "D": "Seven.", + "E": "You took nine boxes out of the drawer in front of you on the table." + }, + "answer_label": "D", + "answer_text": "Seven.", + "q_id": "P11_109_1", + "correct_answer": "Seven.", + "original_options": { + "A": "Six.", + "B": "Eight.", + "C": "You took five boxes out of the drawer next to you on the counter.", + "D": "Seven.", + "E": "You took nine boxes out of the drawer in front of you on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_1_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes did I take out of the drawer?", + "answer": "You took seven boxes out of the drawer in front of you.", + "question_chinese": "我从抽屉中拿出了多少个盒子？", + "answer_chinese": "你从你面前的抽屉里拿出了七个盒子。", + "evidence": [ + { + "description": "The first box.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "The second and third boxes.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:14" + } + }, + { + "description": "The fourth and fifth boxes.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:23" + } + }, + { + "description": "The seventh box.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:27" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took seven boxes out of the drawer in front of you.", + "B": "You took six boxes.", + "C": "You took eight boxes.", + "D": "You took five boxes out of the drawer in front of you.", + "E": "You took nine boxes out of the drawer in front of you." + }, + "answer_label": "A", + "answer_text": "You took seven boxes out of the drawer in front of you.", + "q_id": "P11_109_1_T2", + "correct_answer": "You took seven boxes out of the drawer in front of you.", + "original_options": { + "A": "You took seven boxes out of the drawer in front of you.", + "B": "You took six boxes.", + "C": "You took eight boxes.", + "D": "You took five boxes out of the drawer in front of you.", + "E": "You took nine boxes out of the drawer in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_9", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands at my current location?", + "answer": "You can't wash your hands.", + "question_chinese": "我在当前位置可以洗到手吗", + "answer_chinese": "你洗不到手", + "evidence": [ + { + "description": "The position of the sink.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:01:51", + "end": "00:01:51" + } + } + ], + "question_timestamps": "00:01:51", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can wash your hands.", + "B": "You can't wash your hands." + }, + "answer_label": "B", + "answer_text": "You can't wash your hands.", + "q_id": "P11_109_9", + "correct_answer": "You can't wash your hands.", + "original_options": { + "A": "You can wash your hands.", + "B": "You can't wash your hands." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_5_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I’m standing right now, can I wash my hands?", + "answer": "You couldn't wash your hands.", + "question_chinese": "我在当前位置可以洗到手吗？", + "answer_chinese": "你洗不到手", + "evidence": [ + { + "description": "The position of the sink.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:41" + } + } + ], + "question_timestamps": "00:02:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn't wash your hands.", + "B": "You could wash your hands." + }, + "answer_label": "A", + "answer_text": "You couldn't wash your hands.", + "q_id": "P11_109_5_T2", + "correct_answer": "You couldn't wash your hands.", + "original_options": { + "A": "You couldn't wash your hands.", + "B": "You could wash your hands." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_5_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I’m standing right now, can I wash my hands?", + "answer": "You could wash your hands.", + "question_chinese": "我在当前位置可以洗到手吗？", + "answer_chinese": "你可以洗到手。", + "evidence": [ + { + "description": "The position of the sink.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:02:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn't wash your hands.", + "B": "You could wash your hands." + }, + "answer_label": "B", + "answer_text": "You could wash your hands.", + "q_id": "P11_109_5_T1", + "correct_answer": "You could wash your hands.", + "original_options": { + "A": "You couldn't wash your hands.", + "B": "You could wash your hands." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_4_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I put the bowl?", + "answer": "You put the bowls into the cabinet above the electric kettle to your right.", + "question_chinese": "我将碗放在了什么地方？", + "answer_chinese": "你把碗放在了你右边的电水壶上面的橱柜里。", + "evidence": [ + { + "description": "You put the bowl in the first cabinet.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:53" + } + } + ], + "question_timestamps": "00:01:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In the cabinet in front of you.", + "B": "In the cabinet to your left.", + "C": "You put the bowls into the cabinet above the electric kettle in front of you.", + "D": "You put the bowls into the cabinet above the electric kettle behind you.", + "E": "You put the bowls into the cabinet above the electric kettle to your right." + }, + "answer_label": "E", + "answer_text": "You put the bowls into the cabinet above the electric kettle to your right.", + "q_id": "P11_109_4_T1", + "correct_answer": "You put the bowls into the cabinet above the electric kettle to your right.", + "original_options": { + "A": "In the cabinet in front of you.", + "B": "In the cabinet to your left.", + "C": "You put the bowls into the cabinet above the electric kettle in front of you.", + "D": "You put the bowls into the cabinet above the electric kettle behind you.", + "E": "You put the bowls into the cabinet above the electric kettle to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 13 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_7", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I put the bowl?", + "answer": "You put the bowl in two cabinets.", + "question_chinese": "我将碗放在了什么地方？", + "answer_chinese": "你把碗放在了两个橱柜中", + "evidence": [ + { + "description": "You put the bowl in the first cabinet.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:53" + } + }, + { + "description": "You put the bowl in the second cabinet.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:02" + } + } + ], + "question_timestamps": "00:02:26", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You put the bowl in one cabinet.", + "B": "You put the bowl in two cabinets.", + "C": "You put the bowl in three cabinets.", + "D": "You put the bowl in two drawers instead of cabinets, not in any cabinet.", + "E": "You put the bowl in one basket rather than any cabinet in the kitchen." + }, + "answer_label": "B", + "answer_text": "You put the bowl in two cabinets.", + "q_id": "P11_109_7", + "correct_answer": "You put the bowl in two cabinets.", + "original_options": { + "A": "You put the bowl in one cabinet.", + "B": "You put the bowl in two cabinets.", + "C": "You put the bowl in three cabinets.", + "D": "You put the bowl in two drawers instead of cabinets, not in any cabinet.", + "E": "You put the bowl in one basket rather than any cabinet in the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 14 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_4_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I put the bowl?", + "answer": "You put the bowls into two cabinets above you.", + "question_chinese": "我将碗放在了什么地方？", + "answer_chinese": "你把碗放在了你上方的两个橱柜中", + "evidence": [ + { + "description": "You put the bowl in the first cabinet.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:53" + } + }, + { + "description": "You put the bowl in the second cabinet.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:02" + } + } + ], + "question_timestamps": "00:03:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You put the bowls into one cabinet.", + "B": "You put the bowls into three cabinets.", + "C": "You put the bowls into four cabinets above you, not just two cabinets.", + "D": "You put the bowls into two cabinets above you.", + "E": "You didn't put the bowls into any cabinets above you at that moment." + }, + "answer_label": "D", + "answer_text": "You put the bowls into two cabinets above you.", + "q_id": "P11_109_4_T2", + "correct_answer": "You put the bowls into two cabinets above you.", + "original_options": { + "A": "You put the bowls into one cabinet.", + "B": "You put the bowls into three cabinets.", + "C": "You put the bowls into four cabinets above you, not just two cabinets.", + "D": "You put the bowls into two cabinets above you.", + "E": "You didn't put the bowls into any cabinets above you at that moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 13 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_6_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk to the storage bin?", + "answer": "You walked straight for a while.", + "question_chinese": "我怎么走到收纳箱处？", + "answer_chinese": "你直走一段距离", + "evidence": [ + { + "description": "Go straight for a certain distance", + "timestamps": { + "start": "00:00:39", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:01:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left first.", + "B": "You walked straight for a while.", + "C": "You turned right first.", + "D": "You turned left, then walked straight for a short distance before reaching the storage bin.", + "E": "You walked straight briefly, then veered right and continued forward to the storage bin." + }, + "answer_label": "B", + "answer_text": "You walked straight for a while.", + "q_id": "P11_109_6_T1", + "correct_answer": "You walked straight for a while.", + "original_options": { + "A": "You turned left first.", + "B": "You walked straight for a while.", + "C": "You turned right first.", + "D": "You turned left, then walked straight for a short distance before reaching the storage bin.", + "E": "You walked straight briefly, then veered right and continued forward to the storage bin." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_6_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk to the storage bin?", + "answer": "After turning left, you walked straight for a while.", + "question_chinese": "我怎么走到收纳箱处？", + "answer_chinese": "向左转后，直走一段距离", + "evidence": [ + { + "description": "Turn left", + "timestamps": { + "start": "00:00:38", + "end": "00:00:39" + } + }, + { + "description": "Go straight for a certain distance", + "timestamps": { + "start": "00:00:39", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right first.", + "B": "You just walked straight.", + "C": "You turned left, then immediately stopped without walking straight before reaching the storage bin.", + "D": "After turning left, you walked straight for a while.", + "E": "You turned right and walked straight for a while until you reached the storage bin." + }, + "answer_label": "D", + "answer_text": "After turning left, you walked straight for a while.", + "q_id": "P11_109_6_T2", + "correct_answer": "After turning left, you walked straight for a while.", + "original_options": { + "A": "You turned right first.", + "B": "You just walked straight.", + "C": "You turned left, then immediately stopped without walking straight before reaching the storage bin.", + "D": "After turning left, you walked straight for a while.", + "E": "You turned right and walked straight for a while until you reached the storage bin." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_7_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the drawer?", + "answer": "You opened the drawer in front of you three times.", + "question_chinese": "我打开过多少次抽屉？", + "answer_chinese": "你打开过你面前的三次抽屉", + "evidence": [ + { + "description": "First opening.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + }, + { + "description": "Open for the second time.", + "timestamps": { + "start": "00:01:38", + "end": "00:01:39" + } + }, + { + "description": "Open for the third time.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "You opened the drawer in front of you three times.", + "C": "Four times.", + "D": "You opened the drawer in front of you only one time during the whole sequence.", + "E": "You opened the drawer in front of you five separate times before finishing the task." + }, + "answer_label": "B", + "answer_text": "You opened the drawer in front of you three times.", + "q_id": "P11_109_7_T2", + "correct_answer": "You opened the drawer in front of you three times.", + "original_options": { + "A": "Two times.", + "B": "You opened the drawer in front of you three times.", + "C": "Four times.", + "D": "You opened the drawer in front of you only one time during the whole sequence.", + "E": "You opened the drawer in front of you five separate times before finishing the task." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_7_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the drawer?", + "answer": "You opened the drawer in front of you four times.", + "question_chinese": "我打开过多少次抽屉？", + "answer_chinese": "你打开过你面前的抽屉里四次", + "evidence": [ + { + "description": "First opening.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + }, + { + "description": "Open for the second time.", + "timestamps": { + "start": "00:01:38", + "end": "00:01:39" + } + }, + { + "description": "Open for the third time.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:25" + } + }, + { + "description": "The fourth time to open", + "timestamps": { + "start": "00:02:40", + "end": "00:02:42" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened the drawer in front of you four times.", + "B": "Three times.", + "C": "Five times.", + "D": "You opened the drawer in front of you only two times during the whole sequence.", + "E": "You opened the drawer in front of you six separate times before finishing the task." + }, + "answer_label": "A", + "answer_text": "You opened the drawer in front of you four times.", + "q_id": "P11_109_7_T1", + "correct_answer": "You opened the drawer in front of you four times.", + "original_options": { + "A": "You opened the drawer in front of you four times.", + "B": "Three times.", + "C": "Five times.", + "D": "You opened the drawer in front of you only two times during the whole sequence.", + "E": "You opened the drawer in front of you six separate times before finishing the task." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_2_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many glass cups did I put into the cabinet?", + "answer": "You put four glass cups into the cabinet in front of you.", + "question_chinese": "我放了几个玻璃杯进柜子中？", + "answer_chinese": "你在你面前的柜子里放了四个玻璃杯。", + "evidence": [ + { + "description": "There are three iron cups in the cabinet.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You put four glass cups into the cabinet in front of you.", + "B": "Two glass cups.", + "C": "Five glass cups.", + "D": "You put three glass cups into the cabinet in front of you.", + "E": "You put six glass cups into the cabinet in front of you." + }, + "answer_label": "A", + "answer_text": "You put four glass cups into the cabinet in front of you.", + "q_id": "P11_109_2_T1", + "correct_answer": "You put four glass cups into the cabinet in front of you.", + "original_options": { + "A": "You put four glass cups into the cabinet in front of you.", + "B": "Two glass cups.", + "C": "Five glass cups.", + "D": "You put three glass cups into the cabinet in front of you.", + "E": "You put six glass cups into the cabinet in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 12 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_2_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many glass cups did I put into the cabinet?", + "answer": "You put eight glass cups into the cabinet in front of you.", + "question_chinese": "我放了几个玻璃杯进柜子中？", + "answer_chinese": "你在你面前的柜子里放了八个玻璃杯", + "evidence": [ + { + "description": "You placed four glass cups.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:09" + } + }, + { + "description": "You placed four glass cups.", + "timestamps": { + "start": "00:01:09", + "end": "00:01:18" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Seven glass cups.", + "B": "You put eight glass cups into the cabinet in front of you.", + "C": "Nine glass cups.", + "D": "You put six glass cups into the cabinet in front of you.", + "E": "You put ten glass cups into the cabinet in front of you." + }, + "answer_label": "B", + "answer_text": "You put eight glass cups into the cabinet in front of you.", + "q_id": "P11_109_2_T2", + "correct_answer": "You put eight glass cups into the cabinet in front of you.", + "original_options": { + "A": "Seven glass cups.", + "B": "You put eight glass cups into the cabinet in front of you.", + "C": "Nine glass cups.", + "D": "You put six glass cups into the cabinet in front of you.", + "E": "You put ten glass cups into the cabinet in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 12 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_3", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many iron cups are there in the cabinet?", + "answer": "Three.", + "question_chinese": "柜子中有多少个铁杯子", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "There are three iron cups in the cabinet.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:01:39", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "Six.", + "D": "There are eight iron cups inside the cabinet during that moment in the video.", + "E": "There are nine iron cups in the cabinet when you finish looking inside." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "P11_109_3", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "Six.", + "D": "There are eight iron cups inside the cabinet during that moment in the video.", + "E": "There are nine iron cups in the cabinet when you finish looking inside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "P11_109", + "video_length": "00:03:33", + "qid": "P11_109_4", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many iron cups are there in the cabinet?", + "answer": "Three.", + "question_chinese": "柜子中有多少个铁杯子", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "There are three iron cups in the cabinet.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Four.", + "C": "Three.", + "D": "There are five iron cups inside the cabinet during that moment in the video.", + "E": "There are seven iron cups in the cabinet when you finish looking inside." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "P11_109_4", + "correct_answer": "Three.", + "original_options": { + "A": "One.", + "B": "Four.", + "C": "Three.", + "D": "There are five iron cups inside the cabinet during that moment in the video.", + "E": "There are seven iron cups in the cabinet when you finish looking inside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + } +] diff --git a/QAs/P12_103-mcq.json b/QAs/P12_103-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1d30b37c9cc8ab541b9143d04cfb6128e68debbd --- /dev/null +++ b/QAs/P12_103-mcq.json @@ -0,0 +1,786 @@ +[ + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_1_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was cooking, which side of me was the stove on?", + "answer": "The stove is on your left side.", + "question_chinese": "当我做饭时，炉子在我的哪一侧？", + "answer_chinese": "炉子在你的左侧", + "evidence": [ + { + "description": "The stove is on your left.", + "timestamps": { + "start": "00:00:11", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stove is on your right.", + "B": "The stove is behind you.", + "C": "The stove is to your front-right, positioned ahead of you rather than directly beside you.", + "D": "The stove is to your back-left, so you'd need to turn around and left to face it.", + "E": "The stove is on your left side." + }, + "answer_label": "E", + "answer_text": "The stove is on your left side.", + "q_id": "P12_103_1_T1", + "correct_answer": "The stove is on your left side.", + "original_options": { + "A": "The stove is on your right.", + "B": "The stove is behind you.", + "C": "The stove is to your front-right, positioned ahead of you rather than directly beside you.", + "D": "The stove is to your back-left, so you'd need to turn around and left to face it.", + "E": "The stove is on your left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 17 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_1_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was cooking, which side of me was the stove on?", + "answer": "The stove is to your back-right.", + "question_chinese": "当我做饭时，炉子在我的哪一侧？", + "answer_chinese": "炉子在你的右后方。", + "evidence": [ + { + "description": "The stove was on your left and you turn around.", + "timestamps": { + "start": "00:10:14", + "end": "00:10:37" + } + } + ], + "question_timestamps": "00:10:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stove is to your back-right.", + "B": "The stove is to your back-left.", + "C": "The stove is to your front-right.", + "D": "The stove is on your left side, roughly beside you instead of behind you.", + "E": "The stove is on your right side, more to the side than diagonally behind you." + }, + "answer_label": "A", + "answer_text": "The stove is to your back-right.", + "q_id": "P12_103_1_T2", + "correct_answer": "The stove is to your back-right.", + "original_options": { + "A": "The stove is to your back-right.", + "B": "The stove is to your back-left.", + "C": "The stove is to your front-right.", + "D": "The stove is on your left side, roughly beside you instead of behind you.", + "E": "The stove is on your right side, more to the side than diagonally behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_2_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I was cooking, what was my movement path?", + "answer": "You first turned a bit to the right, then moved to your right, and then moved left again.", + "question_chinese": "当我在做饭时，我的运动轨迹是什么？", + "answer_chinese": "你先向右边转了一下，然后移动到右边，然后又向左移动。", + "evidence": [ + { + "description": "You turned right..", + "timestamps": { + "start": "00:00:08", + "end": "00:00:14" + } + }, + { + "description": "You turn left and then move a little to the left.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You first turned a bit to the right, then moved to your right, and then moved left again.", + "B": "Turn left, move left, then right.", + "C": "Turn right, move left, then left.", + "D": "You first turned a bit to the left, then moved to your right, and finally moved right again.", + "E": "You first turned a bit to the right, then moved to your left, and then shifted back to your right." + }, + "answer_label": "A", + "answer_text": "You first turned a bit to the right, then moved to your right, and then moved left again.", + "q_id": "P12_103_2_T1", + "correct_answer": "You first turned a bit to the right, then moved to your right, and then moved left again.", + "original_options": { + "A": "You first turned a bit to the right, then moved to your right, and then moved left again.", + "B": "Turn left, move left, then right.", + "C": "Turn right, move left, then left.", + "D": "You first turned a bit to the left, then moved to your right, and finally moved right again.", + "E": "You first turned a bit to the right, then moved to your left, and then shifted back to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 20 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_2_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I was cooking, what was my movement path?", + "answer": "You moved right to the sink.", + "question_chinese": "当我在做饭时，我的运动轨迹是什么？", + "answer_chinese": "你向右移动到洗手池了。", + "evidence": [ + { + "description": "You move to the right toward the sink.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:10:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved left to the sink.", + "B": "You moved right to the sink.", + "C": "You walked away from the sink.", + "D": "You turned slightly left first, then moved right past the sink rather than stopping at it.", + "E": "You stayed near where you were and reached toward the sink without actually moving over to it." + }, + "answer_label": "B", + "answer_text": "You moved right to the sink.", + "q_id": "P12_103_2_T2", + "correct_answer": "You moved right to the sink.", + "original_options": { + "A": "You moved left to the sink.", + "B": "You moved right to the sink.", + "C": "You walked away from the sink.", + "D": "You turned slightly left first, then moved right past the sink rather than stopping at it.", + "E": "You stayed near where you were and reached toward the sink without actually moving over to it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_3_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cups did I see in the kitchen?", + "answer": "There are 3 cups total, on the table in front of you right now.", + "question_chinese": "我在厨房里看到多少个杯子？", + "answer_chinese": "一共有3个杯子，在你现在前面的桌子上。", + "evidence": [ + { + "description": "Ther are three mugs there.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 2 cups total.", + "B": "There are 4 cups total.", + "C": "There are 3 cups total, on the table in front of you right now.", + "D": "There are 1 cup total, on the table in front of you right now.", + "E": "There are 5 cups total, all grouped together on the table in front of you right now." + }, + "answer_label": "C", + "answer_text": "There are 3 cups total, on the table in front of you right now.", + "q_id": "P12_103_3_T2", + "correct_answer": "There are 3 cups total, on the table in front of you right now.", + "original_options": { + "A": "There are 2 cups total.", + "B": "There are 4 cups total.", + "C": "There are 3 cups total, on the table in front of you right now.", + "D": "There are 1 cup total, on the table in front of you right now.", + "E": "There are 5 cups total, all grouped together on the table in front of you right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 17 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_3_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cups did I see in the kitchen?", + "answer": "There are 9 cups total: 3 mugs are by your right hand, and 6 glass cups are on the tabletop to your back-right.", + "question_chinese": "我在厨房里看到多少个杯子？", + "answer_chinese": "一共有9个杯子，3个马克杯在你的右手边，6个玻璃杯在你的右后方的桌面上。", + "evidence": [ + { + "description": "There are three mugs.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:30" + } + }, + { + "description": "Ther are six glasses ther.e", + "timestamps": { + "start": "00:02:53", + "end": "00:02:58" + } + } + ], + "question_timestamps": "00:10:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 8 cups total.", + "B": "There are 9 cups total: 3 mugs are by your right hand, and 6 glass cups are on the tabletop to your back-right.", + "C": "There are 10 cups total.", + "D": "There are 7 cups total: 2 mugs by your right hand and 5 glass cups back-right.", + "E": "There are 11 cups total: 3 mugs by your right hand and 8 glass cups back-right." + }, + "answer_label": "B", + "answer_text": "There are 9 cups total: 3 mugs are by your right hand, and 6 glass cups are on the tabletop to your back-right.", + "q_id": "P12_103_3_T1", + "correct_answer": "There are 9 cups total: 3 mugs are by your right hand, and 6 glass cups are on the tabletop to your back-right.", + "original_options": { + "A": "There are 8 cups total.", + "B": "There are 9 cups total: 3 mugs are by your right hand, and 6 glass cups are on the tabletop to your back-right.", + "C": "There are 10 cups total.", + "D": "There are 7 cups total: 2 mugs by your right hand and 5 glass cups back-right.", + "E": "There are 11 cups total: 3 mugs by your right hand and 8 glass cups back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_5_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of tableware did I see myself wash?", + "answer": "You washed 1 dish in total, and you put it by your right hand.", + "question_chinese": "我看到自己洗过多少个餐具？", + "answer_chinese": "你一共洗了一个碟子，就放在你的右手边。", + "evidence": [ + { + "description": "the plate", + "timestamps": { + "start": "00:04:30", + "end": "00:04:47" + } + } + ], + "question_timestamps": "00:04:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed 0 dishes total.", + "B": "You washed 2 dishes total.", + "C": "You washed 3 dishes in total, and you set them down near your right hand afterward.", + "D": "You washed 1 dish in total, and you put it by your right hand.", + "E": "You washed 4 dishes in total, then placed all of them by your right hand to dry." + }, + "answer_label": "D", + "answer_text": "You washed 1 dish in total, and you put it by your right hand.", + "q_id": "P12_103_5_T1", + "correct_answer": "You washed 1 dish in total, and you put it by your right hand.", + "original_options": { + "A": "You washed 0 dishes total.", + "B": "You washed 2 dishes total.", + "C": "You washed 3 dishes in total, and you set them down near your right hand afterward.", + "D": "You washed 1 dish in total, and you put it by your right hand.", + "E": "You washed 4 dishes in total, then placed all of them by your right hand to dry." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_5_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of tableware did I see myself wash?", + "answer": "You washed 2 pieces of tableware total: a dish is to your right, and a pot is straight in front of you.", + "question_chinese": "我看到自己洗过多少个餐具？", + "answer_chinese": "你一共洗了两个餐具，一个碟子在你的右边，一个锅在你的正前面。", + "evidence": [ + { + "description": "The plate.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:47" + } + }, + { + "description": "The pot.", + "timestamps": { + "start": "00:06:11", + "end": "00:06:40" + } + } + ], + "question_timestamps": "00:07:01", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed 1 piece total.", + "B": "You washed 3 pieces total.", + "C": "You washed 0 pieces of tableware total; nothing ended up to your right or straight in front of you.", + "D": "You washed 4 pieces of tableware total, with some to your right and the rest straight in front of you.", + "E": "You washed 2 pieces of tableware total: a dish is to your right, and a pot is straight in front of you." + }, + "answer_label": "E", + "answer_text": "You washed 2 pieces of tableware total: a dish is to your right, and a pot is straight in front of you.", + "q_id": "P12_103_5_T2", + "correct_answer": "You washed 2 pieces of tableware total: a dish is to your right, and a pot is straight in front of you.", + "original_options": { + "A": "You washed 1 piece total.", + "B": "You washed 3 pieces total.", + "C": "You washed 0 pieces of tableware total; nothing ended up to your right or straight in front of you.", + "D": "You washed 4 pieces of tableware total, with some to your right and the rest straight in front of you.", + "E": "You washed 2 pieces of tableware total: a dish is to your right, and a pot is straight in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 20 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_6_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the cooked food take as it moved along with me?", + "answer": "You moved it from the pot directly in front of you into the bowl to your left hand.", + "question_chinese": "烹饪的食物跟随我的移动轨迹是什么？", + "answer_chinese": "你把它从正前面锅里放到了左手边的碗里。", + "evidence": [ + { + "description": "Food is in the pot.", + "timestamps": { + "start": "00:05:28", + "end": "00:05:40" + } + } + ], + "question_timestamps": "00:05:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved it from the pot directly in front of you into the bowl to your left hand.", + "B": "From the bowl to the pot in front.", + "C": "From the pot to your right-hand bowl.", + "D": "You moved it from the pot directly in front of you into the bowl held by your right hand.", + "E": "You moved it from a pot positioned behind you and then placed it into the bowl on your left side." + }, + "answer_label": "A", + "answer_text": "You moved it from the pot directly in front of you into the bowl to your left hand.", + "q_id": "P12_103_6_T2", + "correct_answer": "You moved it from the pot directly in front of you into the bowl to your left hand.", + "original_options": { + "A": "You moved it from the pot directly in front of you into the bowl to your left hand.", + "B": "From the bowl to the pot in front.", + "C": "From the pot to your right-hand bowl.", + "D": "You moved it from the pot directly in front of you into the bowl held by your right hand.", + "E": "You moved it from a pot positioned behind you and then placed it into the bowl on your left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 20 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_6_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the cooked food take as it moved along with me?", + "answer": "You moved it from the bowl on your right-hand side to the pot on your left-hand side.", + "question_chinese": "烹饪的食物跟随我的移动轨迹是什么？", + "answer_chinese": "你把它从你右手边的碗里移动到左手边的锅里。", + "evidence": [ + { + "description": "The food on the right-hand side was placed into the pot on the left-hand side.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:28" + } + } + ], + "question_timestamps": "00:07:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the right bowl to the right pot.", + "B": "From the left bowl to the left pot.", + "C": "You moved it from the pot on your left-hand side into the bowl on your right-hand side.", + "D": "You moved it from the bowl straight in front of you and then transferred it into the pot on your left side.", + "E": "You moved it from the bowl on your right-hand side to the pot on your left-hand side." + }, + "answer_label": "E", + "answer_text": "You moved it from the bowl on your right-hand side to the pot on your left-hand side.", + "q_id": "P12_103_6_T1", + "correct_answer": "You moved it from the bowl on your right-hand side to the pot on your left-hand side.", + "original_options": { + "A": "From the right bowl to the right pot.", + "B": "From the left bowl to the left pot.", + "C": "You moved it from the pot on your left-hand side into the bowl on your right-hand side.", + "D": "You moved it from the bowl straight in front of you and then transferred it into the pot on your left side.", + "E": "You moved it from the bowl on your right-hand side to the pot on your left-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 22 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_4_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the ketchup relative to me?", + "answer": "In the fridge to your right.", + "question_chinese": "番茄酱相对于我的位置是？", + "answer_chinese": "在你右边的冰箱里", + "evidence": [ + { + "description": "The ketchup was taken out of the refrigerator and then put back.", + "timestamps": { + "start": "00:02:55", + "end": "00:03:17" + } + } + ], + "question_timestamps": "00:03:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the fridge to your left.", + "B": "In the fridge in front of you.", + "C": "In the fridge to your right.", + "D": "In the fridge behind you on the left side, not on your right.", + "E": "In the fridge to your back-right side, rather than directly to your right." + }, + "answer_label": "C", + "answer_text": "In the fridge to your right.", + "q_id": "P12_103_4_T1", + "correct_answer": "In the fridge to your right.", + "original_options": { + "A": "In the fridge to your left.", + "B": "In the fridge in front of you.", + "C": "In the fridge to your right.", + "D": "In the fridge behind you on the left side, not on your right.", + "E": "In the fridge to your back-right side, rather than directly to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 13 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_4_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the ketchup relative to me?", + "answer": "The ketchup is in the fridge to your back-right.", + "question_chinese": "番茄酱相对于我的位置是？", + "answer_chinese": "番茄酱在你右后边的冰箱里", + "evidence": [ + { + "description": "The ketchup was taken out of the refrigerator and then put back.", + "timestamps": { + "start": "00:02:55", + "end": "00:03:17" + } + } + ], + "question_timestamps": "00:04:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the fridge to your back-left.", + "B": "In the fridge to your front-right.", + "C": "The ketchup is in the fridge to your front-left, not behind you on the right side.", + "D": "The ketchup is in the fridge to your back-right.", + "E": "The ketchup is in the fridge directly to your right, rather than positioned behind you to the right." + }, + "answer_label": "D", + "answer_text": "The ketchup is in the fridge to your back-right.", + "q_id": "P12_103_4_T2", + "correct_answer": "The ketchup is in the fridge to your back-right.", + "original_options": { + "A": "In the fridge to your back-left.", + "B": "In the fridge to your front-right.", + "C": "The ketchup is in the fridge to your front-left, not behind you on the right side.", + "D": "The ketchup is in the fridge to your back-right.", + "E": "The ketchup is in the fridge directly to your right, rather than positioned behind you to the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_8_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the suitcase relative to me?", + "answer": "The suitcase is to your back-right.", + "question_chinese": "行李箱在我的那个位置？", + "answer_chinese": "行李箱在你的右后方。", + "evidence": [ + { + "description": "The suitcaes", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "You are in the kitchen.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The suitcase is to your back-left.", + "B": "The suitcase is to your back-right.", + "C": "The suitcase is to your front-right.", + "D": "The suitcase is directly to your right, not behind you on the right-hand side.", + "E": "The suitcase is directly behind you, rather than sitting off to your back-right side." + }, + "answer_label": "B", + "answer_text": "The suitcase is to your back-right.", + "q_id": "P12_103_8_T1", + "correct_answer": "The suitcase is to your back-right.", + "original_options": { + "A": "The suitcase is to your back-left.", + "B": "The suitcase is to your back-right.", + "C": "The suitcase is to your front-right.", + "D": "The suitcase is directly to your right, not behind you on the right-hand side.", + "E": "The suitcase is directly behind you, rather than sitting off to your back-right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "P12_103", + "video_length": "00:10:40", + "qid": "P12_103_8_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the suitcase relative to me?", + "answer": "The suitcase is directly to your right.", + "question_chinese": "行李箱在我的那个位置？", + "answer_chinese": "行李箱在你的正右方。", + "evidence": [ + { + "description": "The suitcaes", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "You are in the kitchen.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:20" + } + }, + { + "description": "You moved to the right and turned right.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The suitcase is directly to your left.", + "B": "The suitcase is directly in front of you.", + "C": "The suitcase is directly behind you, not positioned at your right-hand side.", + "D": "The suitcase is directly to your right.", + "E": "The suitcase is off to your back-right side, rather than directly to your right." + }, + "answer_label": "D", + "answer_text": "The suitcase is directly to your right.", + "q_id": "P12_103_8_T2", + "correct_answer": "The suitcase is directly to your right.", + "original_options": { + "A": "The suitcase is directly to your left.", + "B": "The suitcase is directly in front of you.", + "C": "The suitcase is directly behind you, not positioned at your right-hand side.", + "D": "The suitcase is directly to your right.", + "E": "The suitcase is off to your back-right side, rather than directly to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 12, + 14 + ] + } +] diff --git a/QAs/P12_105-mcq.json b/QAs/P12_105-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..716db320c7d514ccea84469f30c3637551af9b2c --- /dev/null +++ b/QAs/P12_105-mcq.json @@ -0,0 +1,1266 @@ +[ + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_9_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the kitchen relative to me?", + "answer": "The kitchen is in front of you.", + "question_chinese": "厨房在我的哪个方位？", + "answer_chinese": "厨房在你的前方", + "evidence": [ + { + "description": "The kitchen is on the left side facing the sofa.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The kitchen is behind you.", + "B": "The kitchen is to your left.", + "C": "The kitchen is to your right, not directly in front of you.", + "D": "The kitchen is off to the side of you, rather than straight ahead.", + "E": "The kitchen is in front of you." + }, + "answer_label": "E", + "answer_text": "The kitchen is in front of you.", + "q_id": "P12_105_9_T1", + "correct_answer": "The kitchen is in front of you.", + "original_options": { + "A": "The kitchen is behind you.", + "B": "The kitchen is to your left.", + "C": "The kitchen is to your right, not directly in front of you.", + "D": "The kitchen is off to the side of you, rather than straight ahead.", + "E": "The kitchen is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 13 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_9_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the kitchen relative to me?", + "answer": "The kitchen is to your left.", + "question_chinese": "厨房在我的哪个方位？", + "answer_chinese": "厨房在你的左边", + "evidence": [ + { + "description": "The kitchen is on the left side facing the sofa.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The kitchen is in front of you.", + "B": "The kitchen is to your left.", + "C": "The kitchen is behind you.", + "D": "The kitchen is to your right, not on your left side where you are standing.", + "E": "The kitchen is straight ahead of you, so you would not need to turn left to see it." + }, + "answer_label": "B", + "answer_text": "The kitchen is to your left.", + "q_id": "P12_105_9_T2", + "correct_answer": "The kitchen is to your left.", + "original_options": { + "A": "The kitchen is in front of you.", + "B": "The kitchen is to your left.", + "C": "The kitchen is behind you.", + "D": "The kitchen is to your right, not on your left side where you are standing.", + "E": "The kitchen is straight ahead of you, so you would not need to turn left to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 18 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many POTS are there in the kitchen altogether?", + "answer": "There are two POTS in total.", + "question_chinese": "厨房里一共有几口锅？", + "answer_chinese": "一共有两口锅。", + "evidence": [ + { + "description": "There are two POTS in total.", + "timestamps": { + "start": "00:02:23", + "end": "00:02:23" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one POT in total.", + "B": "There are two POTS in total.", + "C": "There are three POTS in total.", + "D": "There are four POTS in total, counted across the kitchen area altogether.", + "E": "There are five POTS in total, including the ones placed together in the kitchen." + }, + "answer_label": "B", + "answer_text": "There are two POTS in total.", + "q_id": "P12_105_5", + "correct_answer": "There are two POTS in total.", + "original_options": { + "A": "There is one POT in total.", + "B": "There are two POTS in total.", + "C": "There are three POTS in total.", + "D": "There are four POTS in total, counted across the kitchen area altogether.", + "E": "There are five POTS in total, including the ones placed together in the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 14 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many POTS are there in the kitchen altogether?", + "answer": "There are two POTS in total.", + "question_chinese": "厨房里一共有几口锅？", + "answer_chinese": "一共有两口锅。", + "evidence": [ + { + "description": "There are two POTS in total.", + "timestamps": { + "start": "00:02:23", + "end": "00:05:19" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one POT in total.", + "B": "There are three POTS in total.", + "C": "There are four POTS in total, based on everything visible in the kitchen.", + "D": "There are two POTS in total.", + "E": "There are five POTS in total, counting all the pots present in the kitchen." + }, + "answer_label": "D", + "answer_text": "There are two POTS in total.", + "q_id": "P12_105_6", + "correct_answer": "There are two POTS in total.", + "original_options": { + "A": "There is one POT in total.", + "B": "There are three POTS in total.", + "C": "There are four POTS in total, based on everything visible in the kitchen.", + "D": "There are two POTS in total.", + "E": "There are five POTS in total, counting all the pots present in the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 14 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_5_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many food storage bags did I use in total?", + "answer": "You used three in total.", + "question_chinese": "我一共用了几个食品保鲜袋？", + "answer_chinese": "你一共用了三个。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:11" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:56" + } + }, + { + "description": "The third one", + "timestamps": { + "start": "00:02:20", + "end": "00:02:21" + } + } + ], + "question_timestamps": "00:02:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used two in total.", + "B": "You used three in total.", + "C": "You used four in total.", + "D": "You used one in total, meaning only a single food storage bag was used.", + "E": "You used five in total, so you ended up using five separate food storage bags." + }, + "answer_label": "B", + "answer_text": "You used three in total.", + "q_id": "P12_105_5_T2", + "correct_answer": "You used three in total.", + "original_options": { + "A": "You used two in total.", + "B": "You used three in total.", + "C": "You used four in total.", + "D": "You used one in total, meaning only a single food storage bag was used.", + "E": "You used five in total, so you ended up using five separate food storage bags." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_5_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many food storage bags did I use in total?", + "answer": "You used four in total.", + "question_chinese": "我一共用了几个食品保鲜袋？", + "answer_chinese": "你一共用了四个。", + "evidence": [ + { + "description": "The forth one.", + "timestamps": { + "start": "00:03:21", + "end": "00:03:22" + } + }, + { + "description": "The first one.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:11" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:56" + } + }, + { + "description": "The third one.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:21" + } + } + ], + "question_timestamps": "00:03:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used four in total.", + "B": "You used three in total.", + "C": "You used five in total.", + "D": "You used two in total, so only two food storage bags were used overall.", + "E": "You used six in total, meaning six separate food storage bags were used in all." + }, + "answer_label": "A", + "answer_text": "You used four in total.", + "q_id": "P12_105_5_T1", + "correct_answer": "You used four in total.", + "original_options": { + "A": "You used four in total.", + "B": "You used three in total.", + "C": "You used five in total.", + "D": "You used two in total, so only two food storage bags were used overall.", + "E": "You used six in total, meaning six separate food storage bags were used in all." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_9", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many food preservation bags did I share in total?", + "answer": "You used three in total.", + "question_chinese": "我一共用了几个食品保鲜袋？", + "answer_chinese": "你一共用了三个。", + "evidence": [ + { + "description": "The forth one.", + "timestamps": { + "start": "00:03:21", + "end": "00:05:19" + } + }, + { + "description": "The first one.", + "timestamps": { + "start": "00:01:10", + "end": "00:05:19" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:01:55", + "end": "00:05:19" + } + }, + { + "description": "The third one.", + "timestamps": { + "start": "00:02:20", + "end": "00:05:19" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used three in total.", + "B": "You used four in total.", + "C": "You used five in total.", + "D": "You used six in total, so you shared six food preservation bags altogether.", + "E": "You used seven in total, meaning you shared seven separate preservation bags in all." + }, + "answer_label": "A", + "answer_text": "You used three in total.", + "q_id": "P12_105_9", + "correct_answer": "You used three in total.", + "original_options": { + "A": "You used three in total.", + "B": "You used four in total.", + "C": "You used five in total.", + "D": "You used six in total, so you shared six food preservation bags altogether.", + "E": "You used seven in total, meaning you shared seven separate preservation bags in all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_4_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement path in the living room?", + "answer": "You walked forward a few steps.", + "question_chinese": "我在客厅的运动轨迹是什么？", + "answer_chinese": "你向前走了几步", + "evidence": [ + { + "description": "You walked forward a few steps.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked forward a few steps.", + "B": "You turned around in place.", + "C": "You walked backward a few steps.", + "D": "You walked forward a few steps, then shifted right and ended up facing the balcony.", + "E": "You walked forward, moved right to line up with the balcony, then turned left toward the prep counter." + }, + "answer_label": "A", + "answer_text": "You walked forward a few steps.", + "q_id": "P12_105_4_T2", + "correct_answer": "You walked forward a few steps.", + "original_options": { + "A": "You walked forward a few steps.", + "B": "You turned around in place.", + "C": "You walked backward a few steps.", + "D": "You walked forward a few steps, then shifted right and ended up facing the balcony.", + "E": "You walked forward, moved right to line up with the balcony, then turned left toward the prep counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 18 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_4_T3", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement path in the living room?", + "answer": "You walked forward a few steps, moved to the right, and faced the balcony.", + "question_chinese": "我在客厅的运动轨迹是什么？", + "answer_chinese": "你向前走了几步,向右走，面向阳台", + "evidence": [ + { + "description": "You walked forward a few steps.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "turn to the right and face the balcony.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved left and faced the balcony.", + "B": "You stepped back and faced the balcony.", + "C": "You walked forward a few steps, shifted to the left, and then faced the balcony doors.", + "D": "You walked forward a few steps, moved to the right, and faced the balcony.", + "E": "You walked forward several steps, moved to the right, and ended up facing the kitchen instead of the balcony." + }, + "answer_label": "D", + "answer_text": "You walked forward a few steps, moved to the right, and faced the balcony.", + "q_id": "P12_105_4_T3", + "correct_answer": "You walked forward a few steps, moved to the right, and faced the balcony.", + "original_options": { + "A": "You moved left and faced the balcony.", + "B": "You stepped back and faced the balcony.", + "C": "You walked forward a few steps, shifted to the left, and then faced the balcony doors.", + "D": "You walked forward a few steps, moved to the right, and faced the balcony.", + "E": "You walked forward several steps, moved to the right, and ended up facing the kitchen instead of the balcony." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 19 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_4_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement path in the living room?", + "answer": "You walked forward a few steps, moved to the right and faced the balcony, then turned left and walked toward the prep counter.", + "question_chinese": "我在客厅的运动轨迹是什么？", + "answer_chinese": "你向前走了几步,向右走，面向阳台,再左转走向料理台", + "evidence": [ + { + "description": "You walked forward a few steps.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "turn to the right and face the balcony.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "Turn left and walk towards the countertop.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked forward a few steps, moved to the right and faced the balcony, then turned left and walked toward the prep counter.", + "B": "You turned right toward the prep counter.", + "C": "You faced the balcony, then walked straight.", + "D": "You walked forward a few steps, moved to the left to face the balcony, then turned left toward the prep counter.", + "E": "You walked forward a few steps, moved right to face the balcony, then turned right and walked toward the prep counter." + }, + "answer_label": "A", + "answer_text": "You walked forward a few steps, moved to the right and faced the balcony, then turned left and walked toward the prep counter.", + "q_id": "P12_105_4_T1", + "correct_answer": "You walked forward a few steps, moved to the right and faced the balcony, then turned left and walked toward the prep counter.", + "original_options": { + "A": "You walked forward a few steps, moved to the right and faced the balcony, then turned left and walked toward the prep counter.", + "B": "You turned right toward the prep counter.", + "C": "You faced the balcony, then walked straight.", + "D": "You walked forward a few steps, moved to the left to face the balcony, then turned left toward the prep counter.", + "E": "You walked forward a few steps, moved right to face the balcony, then turned right and walked toward the prep counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 21 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_7", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory in the living room?", + "answer": "You turned clockwise once.", + "question_chinese": "我在客厅的运动轨迹是什么？", + "answer_chinese": "你顺时针转了一圈。", + "evidence": [ + { + "description": "You turned clockwise once.", + "timestamps": { + "start": "00:00:13", + "end": "00:05:19" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You turned counterclockwise once.", + "B": "You turned clockwise once.", + "C": "You turned left 180 degrees.", + "D": "You turned halfway to the right and stopped, rather than completing a full clockwise turn.", + "E": "You turned clockwise twice, making two full rotations before stopping in the living room." + }, + "answer_label": "B", + "answer_text": "You turned clockwise once.", + "q_id": "P12_105_7", + "correct_answer": "You turned clockwise once.", + "original_options": { + "A": "You turned counterclockwise once.", + "B": "You turned clockwise once.", + "C": "You turned left 180 degrees.", + "D": "You turned halfway to the right and stopped, rather than completing a full clockwise turn.", + "E": "You turned clockwise twice, making two full rotations before stopping in the living room." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 14 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_8_T1", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to watch TV—where is the remote control?", + "answer": "The remote control is on the coffee table in front of you.", + "question_chinese": "我想看电视，遥控器在哪里？", + "answer_chinese": "遥控器在你面前的茶几上", + "evidence": [ + { + "description": "The remote control is on the coffee table.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s behind you.", + "B": "It’s to your left.", + "C": "The remote control is on the coffee table to your right, next to where you were standing.", + "D": "The remote control is on the coffee table in front of you.", + "E": "The remote control is on the coffee table behind you, closer to the sofa than to the TV." + }, + "answer_label": "D", + "answer_text": "The remote control is on the coffee table in front of you.", + "q_id": "P12_105_8_T1", + "correct_answer": "The remote control is on the coffee table in front of you.", + "original_options": { + "A": "It’s behind you.", + "B": "It’s to your left.", + "C": "The remote control is on the coffee table to your right, next to where you were standing.", + "D": "The remote control is on the coffee table in front of you.", + "E": "The remote control is on the coffee table behind you, closer to the sofa than to the TV." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 18 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_8_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to watch TV—where is the remote control?", + "answer": "The remote control is on the coffee table behind you.", + "question_chinese": "我想看电视，遥控器在哪里？", + "answer_chinese": "遥控器在你身后的茶几上", + "evidence": [ + { + "description": "The remote control is on the coffee table.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The remote control is on the coffee table behind you.", + "B": "In front of you.", + "C": "To your right.", + "D": "The remote control is on the coffee table to your left, not behind you.", + "E": "The remote control is on the coffee table in front of you, positioned between you and the TV." + }, + "answer_label": "A", + "answer_text": "The remote control is on the coffee table behind you.", + "q_id": "P12_105_8_T2", + "correct_answer": "The remote control is on the coffee table behind you.", + "original_options": { + "A": "The remote control is on the coffee table behind you.", + "B": "In front of you.", + "C": "To your right.", + "D": "The remote control is on the coffee table to your left, not behind you.", + "E": "The remote control is on the coffee table in front of you, positioned between you and the TV." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 18 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_14", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want watch TV， where is the remote control?", + "answer": "The remote control in on the coffee table.", + "question_chinese": "我想看电视，遥控器在哪里", + "answer_chinese": "遥控器在茶几上", + "evidence": [ + { + "description": "The remote control is on the coffee table.", + "timestamps": { + "start": "00:00:15", + "end": "00:05:19" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The remote is on the sofa armrest.", + "B": "The remote is on the TV stand.", + "C": "The remote control in on the coffee table.", + "D": "The remote control is on the bookshelf near the TV, not on the coffee table.", + "E": "The remote control is on the dining table, so you’d need to grab it from there." + }, + "answer_label": "C", + "answer_text": "The remote control in on the coffee table.", + "q_id": "P12_105_14", + "correct_answer": "The remote control in on the coffee table.", + "original_options": { + "A": "The remote is on the sofa armrest.", + "B": "The remote is on the TV stand.", + "C": "The remote control in on the coffee table.", + "D": "The remote control is on the bookshelf near the TV, not on the coffee table.", + "E": "The remote control is on the dining table, so you’d need to grab it from there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 17 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_6_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to wash my hands—how should I go?", + "answer": "You need to go from the balcony into the living room, then from the living room into the kitchen—the sink is in the kitchen.", + "question_chinese": "我想要洗手，我应该怎么走？", + "answer_chinese": "你要从阳台进到客厅，再从客厅进到厨房，洗手池在厨房里。", + "evidence": [ + { + "description": "Walk from the sofa toward the kitchen, passing a sink along the way.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go from the kitchen to the living room.", + "B": "Go from the living room to the balcony.", + "C": "You need to go from the balcony into the kitchen directly, because the sink is by the balcony entrance.", + "D": "You need to go from the balcony into the living room, then from the living room into the kitchen—the sink is in the kitchen.", + "E": "Go from the balcony into the living room, then into the bedroom, where the sink is located." + }, + "answer_label": "D", + "answer_text": "You need to go from the balcony into the living room, then from the living room into the kitchen—the sink is in the kitchen.", + "q_id": "P12_105_6_T1", + "correct_answer": "You need to go from the balcony into the living room, then from the living room into the kitchen—the sink is in the kitchen.", + "original_options": { + "A": "Go from the kitchen to the living room.", + "B": "Go from the living room to the balcony.", + "C": "You need to go from the balcony into the kitchen directly, because the sink is by the balcony entrance.", + "D": "You need to go from the balcony into the living room, then from the living room into the kitchen—the sink is in the kitchen.", + "E": "Go from the balcony into the living room, then into the bedroom, where the sink is located." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 17 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_6_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to wash my hands—how should I go?", + "answer": "The sink is in front of you.", + "question_chinese": "我想要洗手，我应该怎么走？", + "answer_chinese": "洗手池在你的面前", + "evidence": [ + { + "description": "Walk from the sofa toward the kitchen, passing a sink along the way.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sink is in front of you.", + "B": "The sink is behind you.", + "C": "The sink is to your left.", + "D": "The sink is to your right, so you should turn slightly and walk that way.", + "E": "The sink is behind you near the counter, so you need to turn around to reach it." + }, + "answer_label": "A", + "answer_text": "The sink is in front of you.", + "q_id": "P12_105_6_T2", + "correct_answer": "The sink is in front of you.", + "original_options": { + "A": "The sink is in front of you.", + "B": "The sink is behind you.", + "C": "The sink is to your left.", + "D": "The sink is to your right, so you should turn slightly and walk that way.", + "E": "The sink is behind you near the counter, so you need to turn around to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 17 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_3_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, what container do I see the food in?", + "answer": "You saw the food in the pot.", + "question_chinese": "我现在看见料理在什么容器里？", + "answer_chinese": "你看见料理在锅里", + "evidence": [ + { + "description": "You see the food in the pot.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw the food in a bowl.", + "B": "You saw the food in a pan.", + "C": "You saw the food in a bowl on the counter, not inside a pot.", + "D": "You saw the food in the pot.", + "E": "You saw the food in a plastic bag on the table rather than in a pot." + }, + "answer_label": "D", + "answer_text": "You saw the food in the pot.", + "q_id": "P12_105_3_T1", + "correct_answer": "You saw the food in the pot.", + "original_options": { + "A": "You saw the food in a bowl.", + "B": "You saw the food in a pan.", + "C": "You saw the food in a bowl on the counter, not inside a pot.", + "D": "You saw the food in the pot.", + "E": "You saw the food in a plastic bag on the table rather than in a pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 16 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_3_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, what container do I see the food in?", + "answer": "You saw the food in the pot and in a plastic bag.", + "question_chinese": "我现在看见料理在什么容器里？", + "answer_chinese": "你看见料理在锅里和塑料袋里", + "evidence": [ + { + "description": "You see the food in the pot.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + }, + { + "description": "You put the food into a plastic bag.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw the food in the pot.", + "B": "You saw the food in a bowl.", + "C": "You saw the food in the pot and in a plastic bag.", + "D": "You saw the food only in a plastic bag, not in the pot at all.", + "E": "You saw the food in a pan on the stove, without any plastic bag present." + }, + "answer_label": "C", + "answer_text": "You saw the food in the pot and in a plastic bag.", + "q_id": "P12_105_3_T2", + "correct_answer": "You saw the food in the pot and in a plastic bag.", + "original_options": { + "A": "You saw the food in the pot.", + "B": "You saw the food in a bowl.", + "C": "You saw the food in the pot and in a plastic bag.", + "D": "You saw the food only in a plastic bag, not in the pot at all.", + "E": "You saw the food in a pan on the stove, without any plastic bag present." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_2_T1", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my scarf?", + "answer": "It was on the sofa to your left.", + "question_chinese": "我的围巾在哪里？", + "answer_chinese": "在你左边的沙发上", + "evidence": [ + { + "description": "It was placed on the sofa.You walk towards the sofa.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was on the sofa to your right.", + "B": "It was on the sofa to your left.", + "C": "It was on the sofa ahead of you.", + "D": "It was on the sofa behind you, not on the left side.", + "E": "It was on the sofa to your back-right, rather than on the left." + }, + "answer_label": "B", + "answer_text": "It was on the sofa to your left.", + "q_id": "P12_105_2_T1", + "correct_answer": "It was on the sofa to your left.", + "original_options": { + "A": "It was on the sofa to your right.", + "B": "It was on the sofa to your left.", + "C": "It was on the sofa ahead of you.", + "D": "It was on the sofa behind you, not on the left side.", + "E": "It was on the sofa to your back-right, rather than on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 12, + 13 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_2_T2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my scarf?", + "answer": "It was on the sofa to your back-left.", + "question_chinese": "我的围巾在哪里？", + "answer_chinese": "在你左后方的沙发上", + "evidence": [ + { + "description": "It was placed on the sofa.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "You walk toward the countertop, which is straight ahead of you.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:03:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was on the sofa to your back-right.", + "B": "It was on the sofa to your front-left.", + "C": "It was on the sofa directly to your left, not in the back-left area.", + "D": "It was on the sofa to your back-left.", + "E": "It was on the sofa behind you in the center, rather than toward your back-left." + }, + "answer_label": "D", + "answer_text": "It was on the sofa to your back-left.", + "q_id": "P12_105_2_T2", + "correct_answer": "It was on the sofa to your back-left.", + "original_options": { + "A": "It was on the sofa to your back-right.", + "B": "It was on the sofa to your front-left.", + "C": "It was on the sofa directly to your left, not in the back-left area.", + "D": "It was on the sofa to your back-left.", + "E": "It was on the sofa behind you in the center, rather than toward your back-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 14, + 15 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_1_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mugs did I see in front of me?", + "answer": "You saw one on the table in front of you.", + "question_chinese": "我看见了面前的几个马克杯？", + "answer_chinese": "你看见了面前桌子上的一个", + "evidence": [ + { + "description": "There is one mug on the desktop.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw zero in front of you.", + "B": "You saw two in front of you.", + "C": "You saw one on the table in front of you.", + "D": "You saw three mugs on the table directly in front of you.", + "E": "You saw four mugs arranged on the table directly in front of you." + }, + "answer_label": "C", + "answer_text": "You saw one on the table in front of you.", + "q_id": "P12_105_1_T2", + "correct_answer": "You saw one on the table in front of you.", + "original_options": { + "A": "You saw zero in front of you.", + "B": "You saw two in front of you.", + "C": "You saw one on the table in front of you.", + "D": "You saw three mugs on the table directly in front of you.", + "E": "You saw four mugs arranged on the table directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 13 + ] + }, + { + "video_uid": "P12_105", + "video_length": "00:05:19", + "qid": "P12_105_1_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mugs did I see in front of me?", + "answer": "You saw one on the table in front of you, and two at the sink.", + "question_chinese": "我看见了面前的几个马克杯？", + "answer_chinese": "你看见了面前桌子上的一个，洗手台的两个", + "evidence": [ + { + "description": "There are two mugs by the sink.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:38" + } + }, + { + "description": "There is one mug on the desktop.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero on table; two at sink.", + "B": "You saw one on the table in front of you, and two at the sink.", + "C": "Two on table; two at sink.", + "D": "You saw three mugs on the table in front of you, and two mugs at the sink.", + "E": "You saw four mugs on the table in front of you, and two mugs at the sink." + }, + "answer_label": "B", + "answer_text": "You saw one on the table in front of you, and two at the sink.", + "q_id": "P12_105_1_T1", + "correct_answer": "You saw one on the table in front of you, and two at the sink.", + "original_options": { + "A": "Zero on table; two at sink.", + "B": "You saw one on the table in front of you, and two at the sink.", + "C": "Two on table; two at sink.", + "D": "You saw three mugs on the table in front of you, and two mugs at the sink.", + "E": "You saw four mugs on the table in front of you, and two mugs at the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + } +] diff --git a/QAs/P22_101-mcq.json b/QAs/P22_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ac84d0e71b07239e62c409e33264f20a00bd5d97 --- /dev/null +++ b/QAs/P22_101-mcq.json @@ -0,0 +1,916 @@ +[ + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_5_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I walk from the kitchen to the dining table?", + "answer": "The dining table is behind the kitchen, behind you, so you should walk backward.", + "question_chinese": "我从厨房到餐桌应该怎么走？", + "answer_chinese": "餐桌在厨房的后面，在你的后方，你应该向后走。", + "evidence": [ + { + "description": "kitche.n", + "timestamps": { + "start": "00:03:06", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:06:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk forward to the dining table.", + "B": "Walk backward toward back-left.", + "C": "The dining table is behind the kitchen but slightly to your back-left, so step backward diagonally.", + "D": "The dining table is behind the kitchen, behind you, so you should walk backward.", + "E": "The dining table is behind the kitchen but to your back-right, so walk backward and angle right." + }, + "answer_label": "D", + "answer_text": "The dining table is behind the kitchen, behind you, so you should walk backward.", + "q_id": "P22_101_5_T2", + "correct_answer": "The dining table is behind the kitchen, behind you, so you should walk backward.", + "original_options": { + "A": "Walk forward to the dining table.", + "B": "Walk backward toward back-left.", + "C": "The dining table is behind the kitchen but slightly to your back-left, so step backward diagonally.", + "D": "The dining table is behind the kitchen, behind you, so you should walk backward.", + "E": "The dining table is behind the kitchen but to your back-right, so walk backward and angle right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 16, + 17 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_5_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I walk from the kitchen to the dining table?", + "answer": "The dining table is behind the kitchen, to your back-left, so you should walk backward.", + "question_chinese": "我从厨房到餐桌应该怎么走？", + "answer_chinese": "餐桌在厨房的后面，在你的左后方，你应该向后走。", + "evidence": [ + { + "description": "The dining table is behind the kitchen. backward.", + "timestamps": { + "start": "00:05:59", + "end": "00:06:01" + } + } + ], + "question_timestamps": "00:06:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk backward toward back-right.", + "B": "Walk backward straight behind you.", + "C": "The dining table is behind the kitchen, to your back-left, so you should walk backward.", + "D": "The dining table is behind the kitchen but to your front-left, so you should move forward and left.", + "E": "The dining table is in front of the kitchen to your back-left, so turn around and walk forward." + }, + "answer_label": "C", + "answer_text": "The dining table is behind the kitchen, to your back-left, so you should walk backward.", + "q_id": "P22_101_5_T1", + "correct_answer": "The dining table is behind the kitchen, to your back-left, so you should walk backward.", + "original_options": { + "A": "Walk backward toward back-right.", + "B": "Walk backward straight behind you.", + "C": "The dining table is behind the kitchen, to your back-left, so you should walk backward.", + "D": "The dining table is behind the kitchen but to your front-left, so you should move forward and left.", + "E": "The dining table is in front of the kitchen to your back-left, so turn around and walk forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 18 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_3_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many different kinds of dairy foods did I use while cooking?", + "answer": "You used two kinds: milk powder and yogurt. They are to your left and to your right.", + "question_chinese": "我做饭一共使用了几种奶制品食物？", + "answer_chinese": "你使用了两种，奶粉和酸奶。在你的左方和右方。", + "evidence": [ + { + "description": "The milk powdre", + "timestamps": { + "start": "00:00:57", + "end": "00:01:06" + } + }, + { + "description": "The yogutr", + "timestamps": { + "start": "00:03:01", + "end": "00:03:49" + } + } + ], + "question_timestamps": "00:04:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used two kinds: milk powder and yogurt. They are to your left and to your right.", + "B": "One kind of dairy food.", + "C": "Three kinds of dairy foods.", + "D": "You used one kind: just milk powder, and it was kept on your left side.", + "E": "You used three kinds: milk powder, yogurt, and another dairy item, spread on both your left and right." + }, + "answer_label": "A", + "answer_text": "You used two kinds: milk powder and yogurt. They are to your left and to your right.", + "q_id": "P22_101_3_T1", + "correct_answer": "You used two kinds: milk powder and yogurt. They are to your left and to your right.", + "original_options": { + "A": "You used two kinds: milk powder and yogurt. They are to your left and to your right.", + "B": "One kind of dairy food.", + "C": "Three kinds of dairy foods.", + "D": "You used one kind: just milk powder, and it was kept on your left side.", + "E": "You used three kinds: milk powder, yogurt, and another dairy item, spread on both your left and right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_3_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many different kinds of dairy foods did I use while cooking?", + "answer": "You used two kinds: milk powder and yogurt. They are to your left.", + "question_chinese": "我做饭一共使用了几种奶制品食物？", + "answer_chinese": "你使用了两种，奶粉和酸奶。在你的左方。", + "evidence": [ + { + "description": "The milk powdre", + "timestamps": { + "start": "00:00:57", + "end": "00:01:06" + } + }, + { + "description": "The yogutr", + "timestamps": { + "start": "00:03:01", + "end": "00:03:49" + } + } + ], + "question_timestamps": "00:04:49", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One kind of dairy food.", + "B": "You used two kinds: milk powder and yogurt. They are to your left.", + "C": "Three kinds of dairy foods.", + "D": "You used one kind: milk powder only, and it was located to your left while you cooked.", + "E": "You used three kinds: milk powder, yogurt, and butter, and all of them were positioned to your left." + }, + "answer_label": "B", + "answer_text": "You used two kinds: milk powder and yogurt. They are to your left.", + "q_id": "P22_101_3_T2", + "correct_answer": "You used two kinds: milk powder and yogurt. They are to your left.", + "original_options": { + "A": "One kind of dairy food.", + "B": "You used two kinds: milk powder and yogurt. They are to your left.", + "C": "Three kinds of dairy foods.", + "D": "You used one kind: milk powder only, and it was located to your left while you cooked.", + "E": "You used three kinds: milk powder, yogurt, and butter, and all of them were positioned to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_8_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory while I was cooking in the kitchen just now?", + "answer": "You were just alternating between counterclockwise and clockwise motions.", + "question_chinese": "我刚在才在厨房做饭的运动轨迹是什么？", + "answer_chinese": "你刚在做逆顺时针交替运动。", + "evidence": [ + { + "description": "You were just doing clockwise movement.", + "timestamps": { + "start": "00:02:34", + "end": "00:05:55" + } + } + ], + "question_timestamps": "00:05:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Mostly clockwise the whole time.", + "B": "You were just alternating between counterclockwise and clockwise motions.", + "C": "Mostly counterclockwise the whole time.", + "D": "You walked back and forth in a straight line between two nearby spots instead of circling around.", + "E": "You made one counterclockwise loop around the kitchen area and then stopped moving in circles." + }, + "answer_label": "B", + "answer_text": "You were just alternating between counterclockwise and clockwise motions.", + "q_id": "P22_101_8_T2", + "correct_answer": "You were just alternating between counterclockwise and clockwise motions.", + "original_options": { + "A": "Mostly clockwise the whole time.", + "B": "You were just alternating between counterclockwise and clockwise motions.", + "C": "Mostly counterclockwise the whole time.", + "D": "You walked back and forth in a straight line between two nearby spots instead of circling around.", + "E": "You made one counterclockwise loop around the kitchen area and then stopped moving in circles." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 15 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_8_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory while I was cooking in the kitchen just now?", + "answer": "You were just alternating between clockwise and counterclockwise motions.", + "question_chinese": "我刚在才在厨房做饭的运动轨迹是什么？", + "answer_chinese": "你刚在做顺逆时针交替运动。", + "evidence": [ + { + "description": "You were just doing clockwise movemen.t", + "timestamps": { + "start": "00:02:34", + "end": "00:02:37" + } + }, + { + "description": "You were just doing counterclockwise movemen.t", + "timestamps": { + "start": "00:02:16", + "end": "00:02:23" + } + }, + { + "description": "You were just doing clockwise movement.", + "timestamps": { + "start": "00:05:12", + "end": "00:05:14" + } + } + ], + "question_timestamps": "00:06:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were just alternating between clockwise and counterclockwise motions.", + "B": "Mostly clockwise the whole time.", + "C": "Mostly counterclockwise the whole time.", + "D": "You walked back and forth in a straight line between two spots rather than rotating around the space.", + "E": "You made a single clockwise loop around the kitchen area and then stopped instead of switching directions." + }, + "answer_label": "A", + "answer_text": "You were just alternating between clockwise and counterclockwise motions.", + "q_id": "P22_101_8_T1", + "correct_answer": "You were just alternating between clockwise and counterclockwise motions.", + "original_options": { + "A": "You were just alternating between clockwise and counterclockwise motions.", + "B": "Mostly clockwise the whole time.", + "C": "Mostly counterclockwise the whole time.", + "D": "You walked back and forth in a straight line between two spots rather than rotating around the space.", + "E": "You made a single clockwise loop around the kitchen area and then stopped instead of switching directions." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_2_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wall cabinets are there in my kitchen in total?", + "answer": "There are two in total: one is to your front-left and the other is to your back-right.", + "question_chinese": "我厨房里一共有几个壁柜？", + "answer_chinese": "一共有两个，在你的左前方和右后方。", + "evidence": [ + { + "description": "The second on.e", + "timestamps": { + "start": "00:04:07", + "end": "00:04:13" + } + }, + { + "description": "The first on.e", + "timestamps": { + "start": "00:00:01", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:04:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are two in total: one is to your front-left and the other is to your back-right.", + "B": "One wall cabinet total.", + "C": "Three wall cabinets total.", + "D": "There is one wall cabinet in total, and it is located to your front-left.", + "E": "There are three wall cabinets in total: two are to your front-left and one sits to your back-right." + }, + "answer_label": "A", + "answer_text": "There are two in total: one is to your front-left and the other is to your back-right.", + "q_id": "P22_101_2_T2", + "correct_answer": "There are two in total: one is to your front-left and the other is to your back-right.", + "original_options": { + "A": "There are two in total: one is to your front-left and the other is to your back-right.", + "B": "One wall cabinet total.", + "C": "Three wall cabinets total.", + "D": "There is one wall cabinet in total, and it is located to your front-left.", + "E": "There are three wall cabinets in total: two are to your front-left and one sits to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 18 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_2_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wall cabinets are there in my kitchen in total?", + "answer": "There are two in total, and both are behind you.", + "question_chinese": "我厨房里一共有几个壁柜？", + "answer_chinese": "一共有两个,都在你的后方。", + "evidence": [ + { + "description": "The secongd one.", + "timestamps": { + "start": "00:04:07", + "end": "00:04:16" + } + }, + { + "description": "The first one.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:06:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One wall cabinet behind you.", + "B": "Three wall cabinets behind you.", + "C": "There are four wall cabinets in total, and every one of them is positioned behind you.", + "D": "There are zero wall cabinets in total, so there aren’t any wall cabinets behind you at all.", + "E": "There are two in total, and both are behind you." + }, + "answer_label": "E", + "answer_text": "There are two in total, and both are behind you.", + "q_id": "P22_101_2_T1", + "correct_answer": "There are two in total, and both are behind you.", + "original_options": { + "A": "One wall cabinet behind you.", + "B": "Three wall cabinets behind you.", + "C": "There are four wall cabinets in total, and every one of them is positioned behind you.", + "D": "There are zero wall cabinets in total, so there aren’t any wall cabinets behind you at all.", + "E": "There are two in total, and both are behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_1_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to drink water—where is the glass cup?", + "answer": "The glass cup is in the cabinet above the kitchen sink, on your left-hand side.", + "question_chinese": "我想要喝水，玻璃杯在哪里？", + "answer_chinese": "玻璃杯在厨房洗手池上面的柜子里，在你的左手边。", + "evidence": [ + { + "description": "The sink in the kitchne", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + }, + { + "description": "Glass cpu", + "timestamps": { + "start": "00:00:06", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the cabinet above the sink, right side.", + "B": "In the cabinet below the sink, left side.", + "C": "The glass cup is stored in the cabinet above the kitchen sink, but it’s on the side behind you.", + "D": "The glass cup is in the cabinet above the kitchen sink, on your left-hand side.", + "E": "You’ll find the glass cup in the cabinet below the kitchen sink, positioned on your right-hand side." + }, + "answer_label": "D", + "answer_text": "The glass cup is in the cabinet above the kitchen sink, on your left-hand side.", + "q_id": "P22_101_1_T1", + "correct_answer": "The glass cup is in the cabinet above the kitchen sink, on your left-hand side.", + "original_options": { + "A": "In the cabinet above the sink, right side.", + "B": "In the cabinet below the sink, left side.", + "C": "The glass cup is stored in the cabinet above the kitchen sink, but it’s on the side behind you.", + "D": "The glass cup is in the cabinet above the kitchen sink, on your left-hand side.", + "E": "You’ll find the glass cup in the cabinet below the kitchen sink, positioned on your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 20, + 18 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_1_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to drink water—where is the glass cup?", + "answer": "The glass cup is in the cabinet above the kitchen sink, on the side behind you.", + "question_chinese": "我想要喝水，玻璃杯在哪里？", + "answer_chinese": "玻璃杯在厨房洗手池上面的柜子里，在你的后手边。", + "evidence": [ + { + "description": "The sink in the kitchne", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + }, + { + "description": "Glass cpu", + "timestamps": { + "start": "00:00:06", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:06:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the cabinet above the sink, front side.", + "B": "In the cabinet below the sink, behind you.", + "C": "The glass cup is in the cabinet above the kitchen sink, located on your left-hand side.", + "D": "The glass cup is in the cabinet above the kitchen sink, positioned on your right-hand side.", + "E": "The glass cup is in the cabinet above the kitchen sink, on the side behind you." + }, + "answer_label": "E", + "answer_text": "The glass cup is in the cabinet above the kitchen sink, on the side behind you.", + "q_id": "P22_101_1_T2", + "correct_answer": "The glass cup is in the cabinet above the kitchen sink, on the side behind you.", + "original_options": { + "A": "In the cabinet above the sink, front side.", + "B": "In the cabinet below the sink, behind you.", + "C": "The glass cup is in the cabinet above the kitchen sink, located on your left-hand side.", + "D": "The glass cup is in the cabinet above the kitchen sink, positioned on your right-hand side.", + "E": "The glass cup is in the cabinet above the kitchen sink, on the side behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 16 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_6_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to cook—where is the spice rack?", + "answer": "The spice rack is behind the microwave, above you to your left.", + "question_chinese": "我想要烹饪，调料架在哪里？", + "answer_chinese": "调料架在微波炉的后面，在你的左上方。", + "evidence": [ + { + "description": "The seasoning rack is behind the microwave oven.", + "timestamps": { + "start": "00:01:13", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above you to your right.", + "B": "In front of the microwave.", + "C": "The spice rack is behind the microwave, above you but on your right side instead of left.", + "D": "The spice rack is behind the microwave, but it sits below you on your left side.", + "E": "The spice rack is behind the microwave, above you to your left." + }, + "answer_label": "E", + "answer_text": "The spice rack is behind the microwave, above you to your left.", + "q_id": "P22_101_6_T2", + "correct_answer": "The spice rack is behind the microwave, above you to your left.", + "original_options": { + "A": "Above you to your right.", + "B": "In front of the microwave.", + "C": "The spice rack is behind the microwave, above you but on your right side instead of left.", + "D": "The spice rack is behind the microwave, but it sits below you on your left side.", + "E": "The spice rack is behind the microwave, above you to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_6_T1", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to cook—where is the spice rack?", + "answer": "The spice rack is behind the microwave, above you to your right.", + "question_chinese": "我想要烹饪，调料架在哪里？", + "answer_chinese": "调料架在微波炉的后面，在你的右上方。", + "evidence": [ + { + "description": "The seasoning rack is behind the microwave oven.", + "timestamps": { + "start": "00:01:13", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:02:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above you to your left.", + "B": "In front of the microwave.", + "C": "The spice rack is behind the microwave, above you but on your left side instead of right.", + "D": "The spice rack is behind the microwave, above you to your right.", + "E": "The spice rack is behind the microwave, but it’s located below you on the right side." + }, + "answer_label": "D", + "answer_text": "The spice rack is behind the microwave, above you to your right.", + "q_id": "P22_101_6_T1", + "correct_answer": "The spice rack is behind the microwave, above you to your right.", + "original_options": { + "A": "Above you to your left.", + "B": "In front of the microwave.", + "C": "The spice rack is behind the microwave, above you but on your left side instead of right.", + "D": "The spice rack is behind the microwave, above you to your right.", + "E": "The spice rack is behind the microwave, but it’s located below you on the right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_7_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many different ingredients did I use in total to make the oatmeal bowl right now?", + "answer": "You’ve used three ingredients for the oatmeal bowl—yogurt, oats, and cookies—and they’re to your front-left.", + "question_chinese": "我现在制作燕麦碗一共用了几种材料？", + "answer_chinese": "你现在制作燕麦碗用了酸奶，燕麦，饼干三种材料，在你的左前方。", + "evidence": [ + { + "description": "The yogurt.", + "timestamps": { + "start": "00:02:59", + "end": "00:03:06" + } + }, + { + "description": "The oats.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:24" + } + }, + { + "description": "The biscuits.", + "timestamps": { + "start": "00:04:35", + "end": "00:04:46" + } + } + ], + "question_timestamps": "00:04:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two ingredients total.", + "B": "Four ingredients total.", + "C": "You’ve used two ingredients for the oatmeal bowl—yogurt and oats—and they’re to your front-left.", + "D": "You’ve used four ingredients for the oatmeal bowl—yogurt, oats, cookies, and honey—and they’re to your front-left.", + "E": "You’ve used three ingredients for the oatmeal bowl—yogurt, oats, and cookies—and they’re to your front-left." + }, + "answer_label": "E", + "answer_text": "You’ve used three ingredients for the oatmeal bowl—yogurt, oats, and cookies—and they’re to your front-left.", + "q_id": "P22_101_7_T1", + "correct_answer": "You’ve used three ingredients for the oatmeal bowl—yogurt, oats, and cookies—and they’re to your front-left.", + "original_options": { + "A": "Two ingredients total.", + "B": "Four ingredients total.", + "C": "You’ve used two ingredients for the oatmeal bowl—yogurt and oats—and they’re to your front-left.", + "D": "You’ve used four ingredients for the oatmeal bowl—yogurt, oats, cookies, and honey—and they’re to your front-left.", + "E": "You’ve used three ingredients for the oatmeal bowl—yogurt, oats, and cookies—and they’re to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 20 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_7_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many different ingredients did I use in total to make the oatmeal bowl right now?", + "answer": "You’ve used three ingredients for the oatmeal bowl—yogurt, oats, and cookies—and they’re all to your left.", + "question_chinese": "我现在制作燕麦碗一共用了几种材料？", + "answer_chinese": "你现在制作燕麦碗用了酸奶，燕麦，饼干三种材料，都在你的左方。", + "evidence": [ + { + "description": "The yogurt.", + "timestamps": { + "start": "00:02:59", + "end": "00:03:06" + } + }, + { + "description": "The oats.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:30" + } + }, + { + "description": "The biscuits.", + "timestamps": { + "start": "00:04:35", + "end": "00:05:06" + } + } + ], + "question_timestamps": "00:06:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve used three ingredients for the oatmeal bowl—yogurt, oats, and cookies—and they’re all to your left.", + "B": "Two ingredients total.", + "C": "Four ingredients total.", + "D": "You’ve used one ingredient for the oatmeal bowl—oats—and it’s to your left.", + "E": "You’ve used five ingredients for the oatmeal bowl—yogurt, oats, cookies, honey, and cinnamon—and they’re all to your left." + }, + "answer_label": "A", + "answer_text": "You’ve used three ingredients for the oatmeal bowl—yogurt, oats, and cookies—and they’re all to your left.", + "q_id": "P22_101_7_T2", + "correct_answer": "You’ve used three ingredients for the oatmeal bowl—yogurt, oats, and cookies—and they’re all to your left.", + "original_options": { + "A": "You’ve used three ingredients for the oatmeal bowl—yogurt, oats, and cookies—and they’re all to your left.", + "B": "Two ingredients total.", + "C": "Four ingredients total.", + "D": "You’ve used one ingredient for the oatmeal bowl—oats—and it’s to your left.", + "E": "You’ve used five ingredients for the oatmeal bowl—yogurt, oats, cookies, honey, and cinnamon—and they’re all to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 22 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_4_T2", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I’m washing dishes right now—can I directly reach the refrigerator from here?", + "answer": "You can’t reach it right now; the fridge is to your left.", + "question_chinese": "我现在在洗碗，我能直接够到冰箱吗?", + "answer_chinese": "你现在不能够够到，冰箱在你的左边。", + "evidence": [ + { + "description": "The refrigerator.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:35" + } + } + ], + "question_timestamps": "00:02:38", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach it right now; the fridge is to your left.", + "B": "You can’t reach it right now; the fridge is to your left." + }, + "answer_label": "B", + "answer_text": "You can’t reach it right now; the fridge is to your left.", + "q_id": "P22_101_4_T2", + "correct_answer": "You can’t reach it right now; the fridge is to your left.", + "original_options": { + "A": "You can reach it right now; the fridge is to your left.", + "B": "You can’t reach it right now; the fridge is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "P22_101", + "video_length": "00:06:06", + "qid": "P22_101_4_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I’m washing dishes right now—can I directly reach the refrigerator from here?", + "answer": "You can reach it right now; the fridge is to your left.", + "question_chinese": "我现在在洗碗，我能直接够到冰箱吗?", + "answer_chinese": "你现在能够够到，冰箱在你的左边。", + "evidence": [ + { + "description": "The refrigerator.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:36" + } + } + ], + "question_timestamps": "00:02:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach it right now; the fridge is to your left.", + "B": "You can’t reach it right now; the fridge is to your left." + }, + "answer_label": "A", + "answer_text": "You can reach it right now; the fridge is to your left.", + "q_id": "P22_101_4_T1", + "correct_answer": "You can reach it right now; the fridge is to your left.", + "original_options": { + "A": "You can reach it right now; the fridge is to your left.", + "B": "You can’t reach it right now; the fridge is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + } +] diff --git a/QAs/P22_105-mcq.json b/QAs/P22_105-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..34326a156f3c6cdd72f505fc2eb090b400631935 --- /dev/null +++ b/QAs/P22_105-mcq.json @@ -0,0 +1,941 @@ +[ + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_5_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the microwave relative to me?", + "answer": "The microwave is to the right of the fridge, and it’s to your left.", + "question_chinese": "微波炉在我哪里?", + "answer_chinese": "微波炉在冰箱的右边，在你的左边。", + "evidence": [ + { + "description": "The microwave oven is on the right of the refrigerator.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your right.", + "B": "The microwave is to the right of the fridge, and it’s to your left.", + "C": "In front of you.", + "D": "The microwave is to the left of the fridge, and it’s positioned on your left side.", + "E": "The microwave is directly in front of the fridge, and it’s located on your left side." + }, + "answer_label": "B", + "answer_text": "The microwave is to the right of the fridge, and it’s to your left.", + "q_id": "P22_105_5_T1", + "correct_answer": "The microwave is to the right of the fridge, and it’s to your left.", + "original_options": { + "A": "On your right.", + "B": "The microwave is to the right of the fridge, and it’s to your left.", + "C": "In front of you.", + "D": "The microwave is to the left of the fridge, and it’s positioned on your left side.", + "E": "The microwave is directly in front of the fridge, and it’s located on your left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 17, + 17 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_5_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the microwave relative to me?", + "answer": "The microwave is to the right of the fridge, and it’s to your right.", + "question_chinese": "微波炉在我哪里?", + "answer_chinese": "微波炉在冰箱的右边 在你的右边。", + "evidence": [ + { + "description": "The microwave oven is on the right of the refrigerator.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your left.", + "B": "In front of you.", + "C": "The microwave is to the right of the fridge, and it’s to your right.", + "D": "The microwave is to the left of the fridge, and it’s positioned on your right side.", + "E": "The microwave is directly in front of the fridge, and it’s located on your right side." + }, + "answer_label": "C", + "answer_text": "The microwave is to the right of the fridge, and it’s to your right.", + "q_id": "P22_105_5_T2", + "correct_answer": "The microwave is to the right of the fridge, and it’s to your right.", + "original_options": { + "A": "On your left.", + "B": "In front of you.", + "C": "The microwave is to the right of the fridge, and it’s to your right.", + "D": "The microwave is to the left of the fridge, and it’s positioned on your right side.", + "E": "The microwave is directly in front of the fridge, and it’s located on your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 17, + 17 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_2_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the thing I just took out of the refrigerator?", + "answer": "You put it on top of the sink, to your back-right.", + "question_chinese": "我刚才从冰箱里拿出来的东西放在了哪里？", + "answer_chinese": "你把它放在了洗水槽上面，在你的右后方。", + "evidence": [ + { + "description": "You took it out of the fridge.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "You put it on the sink.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the sink, back-left.", + "B": "On the counter, back-right.", + "C": "You set it on the sink, but it ended up to your front-right instead of behind you.", + "D": "You placed it on the counter beside the sink, positioned to your back-right as you turned away.", + "E": "You put it on top of the sink, to your back-right." + }, + "answer_label": "E", + "answer_text": "You put it on top of the sink, to your back-right.", + "q_id": "P22_105_2_T2", + "correct_answer": "You put it on top of the sink, to your back-right.", + "original_options": { + "A": "On the sink, back-left.", + "B": "On the counter, back-right.", + "C": "You set it on the sink, but it ended up to your front-right instead of behind you.", + "D": "You placed it on the counter beside the sink, positioned to your back-right as you turned away.", + "E": "You put it on top of the sink, to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_2_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the thing I just took out of the refrigerator?", + "answer": "You put it on top of the sink, to your back-left.", + "question_chinese": "我刚才从冰箱里拿出来的东西放在了哪里？", + "answer_chinese": "你把它放在了洗水槽上面，在你的左后方。", + "evidence": [ + { + "description": "You took it out of the fridge.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "You put it on the sink.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the sink, back-right.", + "B": "On the sink, front-left.", + "C": "You left it on top of the sink, but it was to your front-right rather than back-left.", + "D": "You put it on top of the sink, to your back-left.", + "E": "You put it on the sink counter, sitting directly behind you instead of off to your back-left." + }, + "answer_label": "D", + "answer_text": "You put it on top of the sink, to your back-left.", + "q_id": "P22_105_2_T1", + "correct_answer": "You put it on top of the sink, to your back-left.", + "original_options": { + "A": "On the sink, back-right.", + "B": "On the sink, front-left.", + "C": "You left it on top of the sink, but it was to your front-right rather than back-left.", + "D": "You put it on top of the sink, to your back-left.", + "E": "You put it on the sink counter, sitting directly behind you instead of off to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_3_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was the thing I just took out of the refrigerator?", + "answer": "It’s a box of yogurt, and the fridge is to your left.", + "question_chinese": "我刚才从冰箱里拿出来的东西是什么？", + "answer_chinese": "是一盒酸奶，冰箱在你的左方。", + "evidence": [ + { + "description": "You took it out of the fridge.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "It's a box of yogurt.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A box of yogurt; fridge on your right.", + "B": "A box of yogurt; fridge behind you.", + "C": "It’s a box of yogurt, and the refrigerator is positioned to your front-left rather than to your left.", + "D": "It’s a box of yogurt, and you were facing so the fridge sat at your back-right, not on your left side.", + "E": "It’s a box of yogurt, and the fridge is to your left." + }, + "answer_label": "E", + "answer_text": "It’s a box of yogurt, and the fridge is to your left.", + "q_id": "P22_105_3_T1", + "correct_answer": "It’s a box of yogurt, and the fridge is to your left.", + "original_options": { + "A": "A box of yogurt; fridge on your right.", + "B": "A box of yogurt; fridge behind you.", + "C": "It’s a box of yogurt, and the refrigerator is positioned to your front-left rather than to your left.", + "D": "It’s a box of yogurt, and you were facing so the fridge sat at your back-right, not on your left side.", + "E": "It’s a box of yogurt, and the fridge is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 22 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_3_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was the thing I just took out of the refrigerator?", + "answer": "It’s a box of yogurt. The fridge is to your back-right.", + "question_chinese": "我刚才从冰箱里拿出来的东西是什么？", + "answer_chinese": "是一盒酸奶。冰箱在你的右后方。", + "evidence": [ + { + "description": "You took it out of the fridge.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "It's a box of yogurt.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s a box of yogurt. The fridge is to your back-right.", + "B": "A box of yogurt; fridge back-left.", + "C": "A box of yogurt; fridge in front.", + "D": "It’s a box of yogurt, and the refrigerator is to your left instead of to your back-right.", + "E": "It’s a box of yogurt, with the fridge to your front-right as you stand, not behind your right shoulder." + }, + "answer_label": "A", + "answer_text": "It’s a box of yogurt. The fridge is to your back-right.", + "q_id": "P22_105_3_T2", + "correct_answer": "It’s a box of yogurt. The fridge is to your back-right.", + "original_options": { + "A": "It’s a box of yogurt. The fridge is to your back-right.", + "B": "A box of yogurt; fridge back-left.", + "C": "A box of yogurt; fridge in front.", + "D": "It’s a box of yogurt, and the refrigerator is to your left instead of to your back-right.", + "E": "It’s a box of yogurt, with the fridge to your front-right as you stand, not behind your right shoulder." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 20 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_8_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What was I doing in the kitchen just now?", + "answer": "You prepared food in the kitchen, to your back-right.", + "question_chinese": "我刚才在厨房里做什么？", + "answer_chinese": "你在厨房里准备了食物，在你的右后方。", + "evidence": [ + { + "description": "You have prepared food in the kitchen.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:36" + } + }, + { + "description": "You have prepared food in the kitchen.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:56" + } + }, + { + "description": "You have prepared food in the kitchen.", + "timestamps": { + "start": "00:00:56", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You prepared food, back-left.", + "B": "You prepared food, front-right.", + "C": "You prepared food in the kitchen, but it was positioned to your front-left rather than back-right.", + "D": "You prepared food in the kitchen, and the food ended up in front of you, not off to your back-right.", + "E": "You prepared food in the kitchen, to your back-right." + }, + "answer_label": "E", + "answer_text": "You prepared food in the kitchen, to your back-right.", + "q_id": "P22_105_8_T1", + "correct_answer": "You prepared food in the kitchen, to your back-right.", + "original_options": { + "A": "You prepared food, back-left.", + "B": "You prepared food, front-right.", + "C": "You prepared food in the kitchen, but it was positioned to your front-left rather than back-right.", + "D": "You prepared food in the kitchen, and the food ended up in front of you, not off to your back-right.", + "E": "You prepared food in the kitchen, to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 20 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_8_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What was I doing in the kitchen just now?", + "answer": "You prepared food in the kitchen. The food is behind you.", + "question_chinese": "我刚才在厨房里做什么？", + "answer_chinese": "你在厨房里准备了食物。食物在你身后方。", + "evidence": [ + { + "description": "You have prepared food in the kitchen.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:36" + } + }, + { + "description": "You have prepared food in the kitchen.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:56" + } + }, + { + "description": "You have prepared food in the kitchen.", + "timestamps": { + "start": "00:00:56", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You prepared food in the kitchen. The food is behind you.", + "B": "The food was in front of you.", + "C": "The food was to your right.", + "D": "You prepared food in the kitchen, but the food was placed in front of you rather than behind you.", + "E": "You prepared food in the kitchen, with the food positioned to your left side instead of behind you." + }, + "answer_label": "A", + "answer_text": "You prepared food in the kitchen. The food is behind you.", + "q_id": "P22_105_8_T2", + "correct_answer": "You prepared food in the kitchen. The food is behind you.", + "original_options": { + "A": "You prepared food in the kitchen. The food is behind you.", + "B": "The food was in front of you.", + "C": "The food was to your right.", + "D": "You prepared food in the kitchen, but the food was placed in front of you rather than behind you.", + "E": "You prepared food in the kitchen, with the food positioned to your left side instead of behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 18 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_4_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my walking route in the kitchen?", + "answer": "You walked from the sink to the fridge, and then went back from the fridge to the sink.", + "question_chinese": "我在厨房里的行走路线是什么？", + "answer_chinese": "你从洗水槽走到冰箱旁，又从冰箱旁返回了洗水槽。", + "evidence": [ + { + "description": "You walk from the sink to the refrigerator.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "You returned to the sink from beside the refrigerator again.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Sink to table, then back.", + "B": "You walked from the sink to the fridge, and then went back from the fridge to the sink.", + "C": "Fridge to sink, then back.", + "D": "You walked from the sink to the table, then went back from the table to the sink.", + "E": "You started at the fridge, went to the sink, then continued to the table without returning to the fridge." + }, + "answer_label": "B", + "answer_text": "You walked from the sink to the fridge, and then went back from the fridge to the sink.", + "q_id": "P22_105_4_T2", + "correct_answer": "You walked from the sink to the fridge, and then went back from the fridge to the sink.", + "original_options": { + "A": "Sink to table, then back.", + "B": "You walked from the sink to the fridge, and then went back from the fridge to the sink.", + "C": "Fridge to sink, then back.", + "D": "You walked from the sink to the table, then went back from the table to the sink.", + "E": "You started at the fridge, went to the sink, then continued to the table without returning to the fridge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_4_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my walking route in the kitchen?", + "answer": "You walked from the sink to the fridge, then back to the sink; after that you went to the table and came back.", + "question_chinese": "我在厨房里的行走路线是什么？", + "answer_chinese": "你从洗水槽走到冰箱旁，又从冰箱旁返回了洗水槽，然后去到了桌子旁并返回。", + "evidence": [ + { + "description": "You walk from the sink to the refrigerator.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "You returned to the sink from beside the refrigerator again.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + }, + { + "description": "You then went to the table and returned.", + "timestamps": { + "start": "00:00:35", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked from the sink to the fridge, then back to the sink; after that you went to the table and came back.", + "B": "Sink to fridge, then stopped.", + "C": "Sink to table first.", + "D": "You walked from the sink to the fridge and back, but you never went to the table afterward.", + "E": "You went from the sink to the fridge, then to the table, and did not come back from the table." + }, + "answer_label": "A", + "answer_text": "You walked from the sink to the fridge, then back to the sink; after that you went to the table and came back.", + "q_id": "P22_105_4_T1", + "correct_answer": "You walked from the sink to the fridge, then back to the sink; after that you went to the table and came back.", + "original_options": { + "A": "You walked from the sink to the fridge, then back to the sink; after that you went to the table and came back.", + "B": "Sink to fridge, then stopped.", + "C": "Sink to table first.", + "D": "You walked from the sink to the fridge and back, but you never went to the table afterward.", + "E": "You went from the sink to the fridge, then to the table, and did not come back from the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 20 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_6_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoons are in the salad bowl on my table?", + "answer": "There are two spoons, to your front-right.", + "question_chinese": "我桌子上的沙拉碗中有几个勺子？", + "answer_chinese": "有两个勺子，在你的右前方。", + "evidence": [ + { + "description": "There are two spoons.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:49" + } + } + ], + "question_timestamps": "00:00:58", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There is one spoon, to your front-right.", + "B": "There are three spoons, to your front-right.", + "C": "There are four spoons in the bowl, sitting to your front-right on the table.", + "D": "There are two spoons, to your front-right.", + "E": "There are zero spoons in the salad bowl, even though it’s placed to your front-right." + }, + "answer_label": "D", + "answer_text": "There are two spoons, to your front-right.", + "q_id": "P22_105_6_T2", + "correct_answer": "There are two spoons, to your front-right.", + "original_options": { + "A": "There is one spoon, to your front-right.", + "B": "There are three spoons, to your front-right.", + "C": "There are four spoons in the bowl, sitting to your front-right on the table.", + "D": "There are two spoons, to your front-right.", + "E": "There are zero spoons in the salad bowl, even though it’s placed to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 16 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_6_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoons are in the salad bowl on my table?", + "answer": "There are two spoons, to your right.", + "question_chinese": "我桌子上的沙拉碗中有几个勺子？", + "answer_chinese": "有两个勺子，在你的右方。", + "evidence": [ + { + "description": "There are two spoons.", + "timestamps": { + "start": "00:00:57", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There is one spoon, to your right.", + "B": "There are three spoons, to your right.", + "C": "There are two spoons, to your right.", + "D": "There are four spoons in the salad bowl, positioned to your right on the table.", + "E": "There are zero spoons in the bowl, despite it being to your right." + }, + "answer_label": "C", + "answer_text": "There are two spoons, to your right.", + "q_id": "P22_105_6_T1", + "correct_answer": "There are two spoons, to your right.", + "original_options": { + "A": "There is one spoon, to your right.", + "B": "There are three spoons, to your right.", + "C": "There are two spoons, to your right.", + "D": "There are four spoons in the salad bowl, positioned to your right on the table.", + "E": "There are zero spoons in the bowl, despite it being to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 13 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_7_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many benches are under my table?", + "answer": "There are four benches, split between in front of you and behind you.", + "question_chinese": "我桌子下面有几条板凳？", + "answer_chinese": "有四条板凳，分别在你的前方和后方。", + "evidence": [ + { + "description": "There are four benches.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are three benches under the table.", + "B": "There are five benches under the table.", + "C": "There are two benches, with one in front of you and one behind you under the table.", + "D": "There are six benches, split between the space in front of you and the space behind you.", + "E": "There are four benches, split between in front of you and behind you." + }, + "answer_label": "E", + "answer_text": "There are four benches, split between in front of you and behind you.", + "q_id": "P22_105_7_T2", + "correct_answer": "There are four benches, split between in front of you and behind you.", + "original_options": { + "A": "There are three benches under the table.", + "B": "There are five benches under the table.", + "C": "There are two benches, with one in front of you and one behind you under the table.", + "D": "There are six benches, split between the space in front of you and the space behind you.", + "E": "There are four benches, split between in front of you and behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_7_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many benches are under my table?", + "answer": "There are four benches, all behind you.", + "question_chinese": "我桌子下面有几条板凳？", + "answer_chinese": "有四条板凳，都在你的后方。", + "evidence": [ + { + "description": "There are four benches.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are three benches, all behind you.", + "B": "There are five benches, all behind you.", + "C": "There are two benches under the table, and both of them are behind you.", + "D": "There are four benches, all behind you.", + "E": "There are six benches under the table, all grouped together behind you." + }, + "answer_label": "D", + "answer_text": "There are four benches, all behind you.", + "q_id": "P22_105_7_T1", + "correct_answer": "There are four benches, all behind you.", + "original_options": { + "A": "There are three benches, all behind you.", + "B": "There are five benches, all behind you.", + "C": "There are two benches under the table, and both of them are behind you.", + "D": "There are four benches, all behind you.", + "E": "There are six benches under the table, all grouped together behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 12 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_1_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I’m cutting vegetables right now. How should I get the vegetables from the refrigerator?", + "answer": "The fridge is to your right. Turn right and open the cabinet door, and you’ll see the fridge.", + "question_chinese": "我现在在切菜，我该怎么拿到冰箱里的蔬菜？", + "answer_chinese": "冰箱在你右边，你向右转打开柜门就是冰箱。", + "evidence": [ + { + "description": "The refrigerator cabinet door is on your left.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "Behind the cabinet door is the refrigerator.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is to your left.", + "B": "The fridge is behind you.", + "C": "Turn left and open the cabinet door; the fridge should be right there in front of you.", + "D": "The fridge is to your right. Turn right and open the cabinet door, and you’ll see the fridge.", + "E": "Turn around and open the cabinet door, and you’ll spot the fridge straight ahead." + }, + "answer_label": "D", + "answer_text": "The fridge is to your right. Turn right and open the cabinet door, and you’ll see the fridge.", + "q_id": "P22_105_1_T2", + "correct_answer": "The fridge is to your right. Turn right and open the cabinet door, and you’ll see the fridge.", + "original_options": { + "A": "The fridge is to your left.", + "B": "The fridge is behind you.", + "C": "Turn left and open the cabinet door; the fridge should be right there in front of you.", + "D": "The fridge is to your right. Turn right and open the cabinet door, and you’ll see the fridge.", + "E": "Turn around and open the cabinet door, and you’ll spot the fridge straight ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 15 + ] + }, + { + "video_uid": "P22_105", + "video_length": "00:01:08", + "qid": "P22_105_1_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I’m cutting vegetables right now. How should I get the vegetables from the refrigerator?", + "answer": "The fridge is to your left. Turn left and open the cabinet door, and you’ll see the fridge.", + "question_chinese": "我现在在切菜，我该怎么拿到冰箱里的蔬菜？", + "answer_chinese": "冰箱在你左边，你向左转打开柜门就是冰箱。", + "evidence": [ + { + "description": "The refrigerator cabinet door is on your left.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "Behind the cabinet door is the refrigerator.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge is to your right.", + "B": "The fridge is behind you.", + "C": "The fridge is to your left. Turn left and open the cabinet door, and you’ll see the fridge.", + "D": "Turn right and open the cabinet door, and you should see the fridge immediately.", + "E": "Turn around first, then open the cabinet door, and the fridge will be right in view." + }, + "answer_label": "C", + "answer_text": "The fridge is to your left. Turn left and open the cabinet door, and you’ll see the fridge.", + "q_id": "P22_105_1_T1", + "correct_answer": "The fridge is to your left. Turn left and open the cabinet door, and you’ll see the fridge.", + "original_options": { + "A": "The fridge is to your right.", + "B": "The fridge is behind you.", + "C": "The fridge is to your left. Turn left and open the cabinet door, and you’ll see the fridge.", + "D": "Turn right and open the cabinet door, and you should see the fridge immediately.", + "E": "Turn around first, then open the cabinet door, and the fridge will be right in view." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 16 + ] + } +] diff --git a/QAs/P22_113-mcq.json b/QAs/P22_113-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..26057e7cf67e4f20f44ab7a7e89686c2d1f68d58 --- /dev/null +++ b/QAs/P22_113-mcq.json @@ -0,0 +1,920 @@ +[ + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_3_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In the kitchen, where were the apples relative to me?", + "answer": "The apple was on the left side of you.", + "question_chinese": "厨房里苹果相对于我在什么位置？", + "answer_chinese": "苹果在你的左侧。", + "evidence": [ + { + "description": "The apple is on the left side of the stove.", + "timestamps": { + "start": "00:02:23", + "end": "00:02:24" + } + } + ], + "question_timestamps": "00:05:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The apple was in front of you.", + "B": "The apple was behind you.", + "C": "The apple was on the right side of you while you were standing in the kitchen.", + "D": "The apple was on the left side of you.", + "E": "In the kitchen, the apple was positioned behind you rather than off to either side." + }, + "answer_label": "D", + "answer_text": "The apple was on the left side of you.", + "q_id": "P22_113_3_T1", + "correct_answer": "The apple was on the left side of you.", + "original_options": { + "A": "The apple was in front of you.", + "B": "The apple was behind you.", + "C": "The apple was on the right side of you while you were standing in the kitchen.", + "D": "The apple was on the left side of you.", + "E": "In the kitchen, the apple was positioned behind you rather than off to either side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 15 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_3_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In the kitchen, where were the apples relative to me?", + "answer": "The apple was on the right side of you.", + "question_chinese": "厨房里苹果相对于我在什么位置？", + "answer_chinese": "苹果在你的右侧。", + "evidence": [ + { + "description": "The apple is on the left side of the stove.", + "timestamps": { + "start": "00:05:17", + "end": "00:05:22" + } + } + ], + "question_timestamps": "00:11:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The apple was in front of you.", + "B": "The apple was on the right side of you.", + "C": "The apple was behind you.", + "D": "The apple was on the left side of you while you were in the kitchen.", + "E": "In the kitchen, the apple was positioned in front of you, not off to your right." + }, + "answer_label": "B", + "answer_text": "The apple was on the right side of you.", + "q_id": "P22_113_3_T2", + "correct_answer": "The apple was on the right side of you.", + "original_options": { + "A": "The apple was in front of you.", + "B": "The apple was on the right side of you.", + "C": "The apple was behind you.", + "D": "The apple was on the left side of you while you were in the kitchen.", + "E": "In the kitchen, the apple was positioned in front of you, not off to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 16 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_8_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the microwave from the fridge? Without moving, can I directly take food out of the fridge and put it into the microwave to heat it?", + "answer": "The microwave was next to the fridge, about two meters on your left side, and you couldn't reach it.", + "question_chinese": "微波炉距离冰箱有多远，我能在不移动的情况下直接将冰箱里的食物拿到微波炉里加热吗？", + "answer_chinese": "微波炉就在冰箱的旁边，在你左侧两米远，你不能碰到。", + "evidence": [ + { + "description": "The microwave oven is right next to the refrigerator.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:11:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was within arm’s reach.", + "B": "It was on your right side.", + "C": "The microwave was next to the fridge, about two meters on your left side, and you couldn't reach it.", + "D": "The microwave was a few steps away from the fridge, so you could reach it without moving your feet.", + "E": "The microwave was across from the fridge in front of you, and you could transfer food directly into it." + }, + "answer_label": "C", + "answer_text": "The microwave was next to the fridge, about two meters on your left side, and you couldn't reach it.", + "q_id": "P22_113_8_T1", + "correct_answer": "The microwave was next to the fridge, about two meters on your left side, and you couldn't reach it.", + "original_options": { + "A": "It was within arm’s reach.", + "B": "It was on your right side.", + "C": "The microwave was next to the fridge, about two meters on your left side, and you couldn't reach it.", + "D": "The microwave was a few steps away from the fridge, so you could reach it without moving your feet.", + "E": "The microwave was across from the fridge in front of you, and you could transfer food directly into it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_8_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the microwave from the fridge? Without moving, can I directly take food out of the fridge and put it into the microwave to heat it?", + "answer": "The microwave was next to the fridge; you could do it with just an arm movement.", + "question_chinese": "微波炉距离冰箱有多远，我能在不移动的情况下直接将冰箱里的食物拿到微波炉里加热吗？", + "answer_chinese": "微波炉就在冰箱的旁边，你只需要手臂运动就可以完成。", + "evidence": [ + { + "description": "The microwave oven is right next to the refrigerator.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:17" + } + } + ], + "question_timestamps": "00:11:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was a few steps away.", + "B": "The microwave was next to the fridge; you could do it with just an arm movement.", + "C": "You couldn’t reach it.", + "D": "The microwave was next to the fridge, but you still needed to step closer to put food inside.", + "E": "The microwave was near the fridge on your left, yet it was far enough that you couldn’t reach it." + }, + "answer_label": "B", + "answer_text": "The microwave was next to the fridge; you could do it with just an arm movement.", + "q_id": "P22_113_8_T2", + "correct_answer": "The microwave was next to the fridge; you could do it with just an arm movement.", + "original_options": { + "A": "It was a few steps away.", + "B": "The microwave was next to the fridge; you could do it with just an arm movement.", + "C": "You couldn’t reach it.", + "D": "The microwave was next to the fridge, but you still needed to step closer to put food inside.", + "E": "The microwave was near the fridge on your left, yet it was far enough that you couldn’t reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 20 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_1_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just now, where did I move the soup spoon from, and where did I put it?", + "answer": "Just now, you were holding the soup spoon in your hand.", + "question_chinese": "我刚才将汤勺从哪里放在了哪里？", + "answer_chinese": "你刚才将汤勺拿在了手里。", + "evidence": [ + { + "description": "You just put the spoon from the basket into the cabinet below.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:28" + } + }, + { + "description": "You held the soup spoon in your hand just now.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved it into the cabinet.", + "B": "You set it down in the sink.", + "C": "You took the soup spoon from the basket and placed it into the cabinet below the counter.", + "D": "Just now, you were holding the soup spoon in your hand.", + "E": "You picked up the soup spoon from the cabinet and put it back into the basket afterward." + }, + "answer_label": "D", + "answer_text": "Just now, you were holding the soup spoon in your hand.", + "q_id": "P22_113_1_T1", + "correct_answer": "Just now, you were holding the soup spoon in your hand.", + "original_options": { + "A": "You moved it into the cabinet.", + "B": "You set it down in the sink.", + "C": "You took the soup spoon from the basket and placed it into the cabinet below the counter.", + "D": "Just now, you were holding the soup spoon in your hand.", + "E": "You picked up the soup spoon from the cabinet and put it back into the basket afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 17 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_1_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just now, where did I move the soup spoon from, and where did I put it?", + "answer": "Just now, you moved the soup spoon from the basket into the cabinet below.", + "question_chinese": "我刚才将汤勺从哪里放在了哪里？", + "answer_chinese": "你刚才将汤勺从篮子里放在了下面的柜子里。", + "evidence": [ + { + "description": "You just put the spoon from the basket into the cabinet below.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:28" + } + }, + { + "description": "You just put the spoon from the basket into the cabinet below.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:11:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved it into the cabinet above.", + "B": "You left it in your hand.", + "C": "You moved the soup spoon from the cabinet below and placed it back into the basket.", + "D": "You took the soup spoon out of the sink and put it into the cabinet below.", + "E": "Just now, you moved the soup spoon from the basket into the cabinet below." + }, + "answer_label": "E", + "answer_text": "Just now, you moved the soup spoon from the basket into the cabinet below.", + "q_id": "P22_113_1_T2", + "correct_answer": "Just now, you moved the soup spoon from the basket into the cabinet below.", + "original_options": { + "A": "You moved it into the cabinet above.", + "B": "You left it in your hand.", + "C": "You moved the soup spoon from the cabinet below and placed it back into the basket.", + "D": "You took the soup spoon out of the sink and put it into the cabinet below.", + "E": "Just now, you moved the soup spoon from the basket into the cabinet below." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 16 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_5_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just now, where did I put the glass cup that was on the table?", + "answer": "You held the glass cup in your hand.", + "question_chinese": "我刚才把桌子上的玻璃杯放到了哪里？", + "answer_chinese": "你把玻璃杯拿在了手里。", + "evidence": [ + { + "description": "The glass is on the table.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "You picked up the glass.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put it in the sink.", + "B": "You held the glass cup in your hand.", + "C": "You set it back on the table.", + "D": "You placed the glass cup into the sink in front of you after picking it up from the table.", + "E": "You moved the glass cup from the table into a cabinet below the counter and left it there." + }, + "answer_label": "B", + "answer_text": "You held the glass cup in your hand.", + "q_id": "P22_113_5_T1", + "correct_answer": "You held the glass cup in your hand.", + "original_options": { + "A": "You put it in the sink.", + "B": "You held the glass cup in your hand.", + "C": "You set it back on the table.", + "D": "You placed the glass cup into the sink in front of you after picking it up from the table.", + "E": "You moved the glass cup from the table into a cabinet below the counter and left it there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 18 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_5_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just now, where did I put the glass cup that was on the table?", + "answer": "You put the glass cup into the sink in front of you.", + "question_chinese": "我刚才把桌子上的玻璃杯放到了哪里？", + "answer_chinese": "你把玻璃杯放在了面前的洗手池里。", + "evidence": [ + { + "description": "The glass is in the table.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The glass is in the sink.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "You picked up the glass.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:11:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put the glass cup into the sink in front of you.", + "B": "You put it behind you.", + "C": "You put it on the counter.", + "D": "You placed the glass cup into the sink to your right, not the one directly in front of you.", + "E": "You set the glass cup down on the counter beside the sink instead of placing it into the sink." + }, + "answer_label": "A", + "answer_text": "You put the glass cup into the sink in front of you.", + "q_id": "P22_113_5_T2", + "correct_answer": "You put the glass cup into the sink in front of you.", + "original_options": { + "A": "You put the glass cup into the sink in front of you.", + "B": "You put it behind you.", + "C": "You put it on the counter.", + "D": "You placed the glass cup into the sink to your right, not the one directly in front of you.", + "E": "You set the glass cup down on the counter beside the sink instead of placing it into the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 19 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_2_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to toast bread—where is the toaster?", + "answer": "The toaster was to your left.", + "question_chinese": "我想烤面包，面包机在哪里？", + "answer_chinese": "面包机在你的左方。", + "evidence": [ + { + "description": "The bread machine is under the orange bag on the counter.", + "timestamps": { + "start": "00:01:44", + "end": "00:01:45" + } + }, + { + "description": "The bread machine is under the orange bag on the counter.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:48" + } + } + ], + "question_timestamps": "00:06:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The toaster was in front of you.", + "B": "The toaster was to your left.", + "C": "The toaster was behind you.", + "D": "The toaster was to your right on the counter, not on your left side.", + "E": "The toaster was on the counter ahead of you, slightly to the right of where you were standing." + }, + "answer_label": "B", + "answer_text": "The toaster was to your left.", + "q_id": "P22_113_2_T1", + "correct_answer": "The toaster was to your left.", + "original_options": { + "A": "The toaster was in front of you.", + "B": "The toaster was to your left.", + "C": "The toaster was behind you.", + "D": "The toaster was to your right on the counter, not on your left side.", + "E": "The toaster was on the counter ahead of you, slightly to the right of where you were standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 14, + 18 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_2_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to toast bread—where is the toaster?", + "answer": "The toaster was under the orange bag on the counter to your left.", + "question_chinese": "我想烤面包，面包机在哪里？", + "answer_chinese": "面包机在你左边柜台上橙色袋子的下面。", + "evidence": [ + { + "description": "The bread machine is under the orange bag on the counter.", + "timestamps": { + "start": "00:01:44", + "end": "00:01:45" + } + }, + { + "description": "The bread machine is under the orange bag on the counter.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:48" + } + } + ], + "question_timestamps": "00:11:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Under the orange bag in front.", + "B": "The toaster was under the orange bag on the counter to your left.", + "C": "Under the orange bag to your right.", + "D": "The toaster was under the orange bag on the counter behind you, not on the left side.", + "E": "The toaster sat under the orange bag on the counter in front of you rather than to your left." + }, + "answer_label": "B", + "answer_text": "The toaster was under the orange bag on the counter to your left.", + "q_id": "P22_113_2_T2", + "correct_answer": "The toaster was under the orange bag on the counter to your left.", + "original_options": { + "A": "Under the orange bag in front.", + "B": "The toaster was under the orange bag on the counter to your left.", + "C": "Under the orange bag to your right.", + "D": "The toaster was under the orange bag on the counter behind you, not on the left side.", + "E": "The toaster sat under the orange bag on the counter in front of you rather than to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 19 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_4_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I move the fried rice from, and where did I move it to?", + "answer": "You transferred the fried rice from the pot into a container.", + "question_chinese": "我把炒饭从哪里转移到了哪里？", + "answer_chinese": "你把炒饭从锅里转移到了盒子里。", + "evidence": [ + { + "description": "The fried rice is in the frying pan.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:36" + } + }, + { + "description": "The fried rice was transferred to box.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:40" + } + } + ], + "question_timestamps": "00:05:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the pan to a plate.", + "B": "From the bowl into the pot.", + "C": "You transferred the fried rice from the pot into a container.", + "D": "You moved the fried rice from the container back into the pot to keep it warm.", + "E": "You transferred the fried rice from the pot onto a plate, leaving the container unused on the counter." + }, + "answer_label": "C", + "answer_text": "You transferred the fried rice from the pot into a container.", + "q_id": "P22_113_4_T1", + "correct_answer": "You transferred the fried rice from the pot into a container.", + "original_options": { + "A": "From the pan to a plate.", + "B": "From the bowl into the pot.", + "C": "You transferred the fried rice from the pot into a container.", + "D": "You moved the fried rice from the container back into the pot to keep it warm.", + "E": "You transferred the fried rice from the pot onto a plate, leaving the container unused on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_4_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I move the fried rice from, and where did I move it to?", + "answer": "You transferred the fried rice from the pot into a container and a plate.", + "question_chinese": "我把炒饭从哪里转移到了哪里？", + "answer_chinese": "你把炒饭从锅里转移到了盒子里与盘子里。", + "evidence": [ + { + "description": "The fried rice is in the frying pan.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:36" + } + }, + { + "description": "The fried rice was transferred to box.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:39" + } + }, + { + "description": "The fried rice was transferred to plate.", + "timestamps": { + "start": "00:05:56", + "end": "00:05:57" + } + } + ], + "question_timestamps": "00:11:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the pot into a bowl.", + "B": "From the pan onto a plate.", + "C": "You moved the fried rice from the container into the pot, then stirred it again before serving.", + "D": "You transferred the fried rice from the pot into a container and a plate.", + "E": "You transferred the fried rice from the pot into just one container, not splitting any onto a plate." + }, + "answer_label": "D", + "answer_text": "You transferred the fried rice from the pot into a container and a plate.", + "q_id": "P22_113_4_T2", + "correct_answer": "You transferred the fried rice from the pot into a container and a plate.", + "original_options": { + "A": "From the pot into a bowl.", + "B": "From the pan onto a plate.", + "C": "You moved the fried rice from the container into the pot, then stirred it again before serving.", + "D": "You transferred the fried rice from the pot into a container and a plate.", + "E": "You transferred the fried rice from the pot into just one container, not splitting any onto a plate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_7_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the area where I put plates?", + "answer": "It was in the cabinet directly in front of you.", + "question_chinese": "我放盘子的区域在哪里？", + "answer_chinese": "在你的正前方的橱柜里。", + "evidence": [ + { + "description": "It's in the cupboard directly above the washbasin.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:05:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the cabinet behind you.", + "B": "It was in the cabinet directly in front of you.", + "C": "In the cabinet to your left.", + "D": "It was in the cabinet to your right, at about arm’s height next to you.", + "E": "It was in the cabinet above you, so you had to reach up to store the plates." + }, + "answer_label": "B", + "answer_text": "It was in the cabinet directly in front of you.", + "q_id": "P22_113_7_T1", + "correct_answer": "It was in the cabinet directly in front of you.", + "original_options": { + "A": "In the cabinet behind you.", + "B": "It was in the cabinet directly in front of you.", + "C": "In the cabinet to your left.", + "D": "It was in the cabinet to your right, at about arm’s height next to you.", + "E": "It was in the cabinet above you, so you had to reach up to store the plates." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_7_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the area where I put plates?", + "answer": "It was in the wall cabinet directly above the sink.", + "question_chinese": "我放盘子的区域在哪里？", + "answer_chinese": "在洗手池正上方的壁柜里。", + "evidence": [ + { + "description": "It's in the cupboard directly above the washbasin.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:07:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the base cabinet under the sink.", + "B": "In the wall cabinet beside the sink.", + "C": "It was in the wall cabinet directly above the stove, not the one above the sink.", + "D": "It was in the cabinet directly in front of you, rather than the wall cabinet above the sink.", + "E": "It was in the wall cabinet directly above the sink." + }, + "answer_label": "E", + "answer_text": "It was in the wall cabinet directly above the sink.", + "q_id": "P22_113_7_T2", + "correct_answer": "It was in the wall cabinet directly above the sink.", + "original_options": { + "A": "In the base cabinet under the sink.", + "B": "In the wall cabinet beside the sink.", + "C": "It was in the wall cabinet directly above the stove, not the one above the sink.", + "D": "It was in the cabinet directly in front of you, rather than the wall cabinet above the sink.", + "E": "It was in the wall cabinet directly above the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_6_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I want to host guests tomorrow—where are the dining table stools relative to me?", + "answer": "The stool was behind you.", + "question_chinese": "明天我想接待客人，餐桌的凳子在我的什么方位？", + "answer_chinese": "凳子在你的后方。", + "evidence": [ + { + "description": "The chair is in the living room.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:5:41", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The stool was in front of you.", + "B": "The stool was behind you.", + "C": "The stool was to your left.", + "D": "The stool was to your right, close enough that you could reach it without turning around.", + "E": "The stool was positioned in front of you near the dining table, not behind where you stood." + }, + "answer_label": "B", + "answer_text": "The stool was behind you.", + "q_id": "P22_113_6_T2", + "correct_answer": "The stool was behind you.", + "original_options": { + "A": "The stool was in front of you.", + "B": "The stool was behind you.", + "C": "The stool was to your left.", + "D": "The stool was to your right, close enough that you could reach it without turning around.", + "E": "The stool was positioned in front of you near the dining table, not behind where you stood." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 17 + ] + }, + { + "video_uid": "P22_113", + "video_length": "00:11:22", + "qid": "P22_113_6_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I want to host guests tomorrow—where are the dining table stools relative to me?", + "answer": "Your stool was on the left side of you.", + "question_chinese": "明天我想接待客人，餐桌的凳子在我的什么方位？", + "answer_chinese": "你的凳子在你的左侧。", + "evidence": [ + { + "description": "The chair is in the living room.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:08:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Your stool was in front of you.", + "B": "Your stool was behind you.", + "C": "Your stool was to your right, so you would sit down by turning slightly that way.", + "D": "Your stool was on the left side of you.", + "E": "Your stool was positioned in front of you by the table, not on the left side of you." + }, + "answer_label": "D", + "answer_text": "Your stool was on the left side of you.", + "q_id": "P22_113_6_T1", + "correct_answer": "Your stool was on the left side of you.", + "original_options": { + "A": "Your stool was in front of you.", + "B": "Your stool was behind you.", + "C": "Your stool was to your right, so you would sit down by turning slightly that way.", + "D": "Your stool was on the left side of you.", + "E": "Your stool was positioned in front of you by the table, not on the left side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 18 + ] + } +] diff --git a/QAs/P22_117-mcq.json b/QAs/P22_117-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..af733890ce84e32f4da89a2e45aa02071f0f9cf0 --- /dev/null +++ b/QAs/P22_117-mcq.json @@ -0,0 +1,993 @@ +[ + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_13", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What ingredients did I add to the pot?", + "answer": "You added cucumber and onion.", + "question_chinese": "我向锅中加入了什么食材", + "answer_chinese": "你加入了黄瓜和洋葱", + "evidence": [ + { + "description": "You added cucumber.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:50" + } + }, + { + "description": "You joined the onion.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:43" + } + } + ], + "question_timestamps": "00:12:09", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You added carrots and garlic.", + "B": "You added mushrooms and celery.", + "C": "You added zucchini and chives, dropping them into the pot together before stirring.", + "D": "You added cucumber and onion.", + "E": "You added lettuce and radishes into the pot, then mixed them in with the rest of the contents." + }, + "answer_label": "D", + "answer_text": "You added cucumber and onion.", + "q_id": "P22_117_13", + "correct_answer": "You added cucumber and onion.", + "original_options": { + "A": "You added carrots and garlic.", + "B": "You added mushrooms and celery.", + "C": "You added zucchini and chives, dropping them into the pot together before stirring.", + "D": "You added cucumber and onion.", + "E": "You added lettuce and radishes into the pot, then mixed them in with the rest of the contents." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 18 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_14", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What ingredients did I add to the pot?", + "answer": "You added cucumber and onion.", + "question_chinese": "我向锅中加入了什么食材", + "answer_chinese": "你加入了黄瓜和洋葱", + "evidence": [ + { + "description": "You added cucumber.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:50" + } + }, + { + "description": "You joined the onion.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:43" + } + } + ], + "question_timestamps": "00:16:11", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You added spinach and garlic.", + "B": "You added cucumber and onion.", + "C": "You added tomatoes and lettuce.", + "D": "You added carrots and celery, tossing both into the pot before continuing to cook.", + "E": "You added zucchini and chives to the pot, then stirred them through to combine with everything." + }, + "answer_label": "B", + "answer_text": "You added cucumber and onion.", + "q_id": "P22_117_14", + "correct_answer": "You added cucumber and onion.", + "original_options": { + "A": "You added spinach and garlic.", + "B": "You added cucumber and onion.", + "C": "You added tomatoes and lettuce.", + "D": "You added carrots and celery, tossing both into the pot before continuing to cook.", + "E": "You added zucchini and chives to the pot, then stirred them through to combine with everything." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_15", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I put on the table?", + "answer": "You put a cushion.", + "question_chinese": "我在桌子上放了什么", + "answer_chinese": "你放了一张垫子", + "evidence": [ + { + "description": "You put a mat.", + "timestamps": { + "start": "00:14:30", + "end": "00:14:33" + } + } + ], + "question_timestamps": "00:14:39", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put a blanket.", + "B": "You put a mat.", + "C": "You put a cushion.", + "D": "You put a tablecloth down neatly across the entire table surface before setting anything else on it.", + "E": "You put a chair next to the table instead of placing anything on the tabletop." + }, + "answer_label": "C", + "answer_text": "You put a cushion.", + "q_id": "P22_117_15", + "correct_answer": "You put a cushion.", + "original_options": { + "A": "You put a blanket.", + "B": "You put a mat.", + "C": "You put a cushion.", + "D": "You put a tablecloth down neatly across the entire table surface before setting anything else on it.", + "E": "You put a chair next to the table instead of placing anything on the tabletop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 15 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_16", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I put on the table?", + "answer": "You put a spoon, a plate of food, a cup, and a bottle of water.", + "question_chinese": "我在桌子上放了什么", + "answer_chinese": "你放了一个勺子一盘食物一个杯子和一瓶水", + "evidence": [ + { + "description": "You put a mat.", + "timestamps": { + "start": "00:14:30", + "end": "00:14:33" + } + }, + { + "description": "You put a spoon.", + "timestamps": { + "start": "00:14:40", + "end": "00:14:44" + } + }, + { + "description": "You put a plate of cooked food.", + "timestamps": { + "start": "00:15:20", + "end": "00:15:23" + } + }, + { + "description": "You placed a cup and a bottle of water.", + "timestamps": { + "start": "00:15:45", + "end": "00:15:55" + } + } + ], + "question_timestamps": "00:16:11", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You put a fork, plate, cup, and water.", + "B": "You put a spoon, bowl, cup, and water.", + "C": "You put a spoon, a plate of food, a mug, and a bottle of water.", + "D": "You put a spoon, a plate of food, a cup, and a bottle of juice.", + "E": "You put a spoon, a plate of food, a cup, and a bottle of water." + }, + "answer_label": "E", + "answer_text": "You put a spoon, a plate of food, a cup, and a bottle of water.", + "q_id": "P22_117_16", + "correct_answer": "You put a spoon, a plate of food, a cup, and a bottle of water.", + "original_options": { + "A": "You put a fork, plate, cup, and water.", + "B": "You put a spoon, bowl, cup, and water.", + "C": "You put a spoon, a plate of food, a mug, and a bottle of water.", + "D": "You put a spoon, a plate of food, a cup, and a bottle of juice.", + "E": "You put a spoon, a plate of food, a cup, and a bottle of water." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 15 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_10", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How do I get to the refrigerator?", + "answer": "You can turn left and go straight.", + "question_chinese": "我如何到达冰箱旁", + "answer_chinese": "你可以向左转直走", + "evidence": [ + { + "description": "You can turn left and go straight.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:25" + } + } + ], + "question_timestamps": "00:12:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You can turn left and go straight.", + "B": "Turn right and go straight.", + "C": "Turn left, then turn right.", + "D": "Turn left, then make another left at the next opening before continuing forward toward the refrigerator.", + "E": "Go straight first, then turn left at the next opening to reach the refrigerator." + }, + "answer_label": "A", + "answer_text": "You can turn left and go straight.", + "q_id": "P22_117_10", + "correct_answer": "You can turn left and go straight.", + "original_options": { + "A": "You can turn left and go straight.", + "B": "Turn right and go straight.", + "C": "Turn left, then turn right.", + "D": "Turn left, then make another left at the next opening before continuing forward toward the refrigerator.", + "E": "Go straight first, then turn left at the next opening to reach the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 14 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_9", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How do I get to the refrigerator?", + "answer": "You can turn right.", + "question_chinese": "我如何到达冰箱旁", + "answer_chinese": "你可以向右转", + "evidence": [ + { + "description": "You can turn left and go straight.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:25" + } + }, + { + "description": "You can turn right.", + "timestamps": { + "start": "00:14:32", + "end": "00:14:33" + } + } + ], + "question_timestamps": "00:16:11", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You can turn left.", + "B": "You can go straight ahead.", + "C": "You should turn around completely first, then walk forward until you reach the refrigerator area.", + "D": "You can turn right.", + "E": "You need to go straight toward the doorway and avoid turning right to reach the refrigerator." + }, + "answer_label": "D", + "answer_text": "You can turn right.", + "q_id": "P22_117_9", + "correct_answer": "You can turn right.", + "original_options": { + "A": "You can turn left.", + "B": "You can go straight ahead.", + "C": "You should turn around completely first, then walk forward until you reach the refrigerator area.", + "D": "You can turn right.", + "E": "You need to go straight toward the doorway and avoid turning right to reach the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 16 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the drawer?", + "answer": "You opened it five times.", + "question_chinese": "我打开了多少次抽屉", + "answer_chinese": "你打开了五次", + "evidence": [ + { + "description": "The first time.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + }, + { + "description": "The second time.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:28" + } + }, + { + "description": "The third time.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:55" + } + }, + { + "description": "The fourth time.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:18" + } + }, + { + "description": "The fifth time.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:51" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You opened it three times.", + "B": "You opened it five times.", + "C": "You opened it six times.", + "D": "You opened the drawer a total of seven separate times over the course of the video.", + "E": "You opened and closed the drawer eight times altogether while preparing the items on the counter." + }, + "answer_label": "B", + "answer_text": "You opened it five times.", + "q_id": "P22_117_3", + "correct_answer": "You opened it five times.", + "original_options": { + "A": "You opened it three times.", + "B": "You opened it five times.", + "C": "You opened it six times.", + "D": "You opened the drawer a total of seven separate times over the course of the video.", + "E": "You opened and closed the drawer eight times altogether while preparing the items on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the drawer?", + "answer": "You opened it eight times.", + "question_chinese": "我打开了多少次抽屉", + "answer_chinese": "你打开了八次", + "evidence": [ + { + "description": "The first time.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + }, + { + "description": "The second time.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:28" + } + }, + { + "description": "The third time.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:55" + } + }, + { + "description": "The fourth time.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:18" + } + }, + { + "description": "The fifth time.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:51" + } + }, + { + "description": "The sixth time.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:22" + } + }, + { + "description": "The seventh time.", + "timestamps": { + "start": "00:04:52", + "end": "00:05:05" + } + }, + { + "description": "The eighth time.", + "timestamps": { + "start": "00:07:59", + "end": "00:08:05" + } + } + ], + "question_timestamps": "00:08:06", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You opened it eight times.", + "B": "You opened it six times.", + "C": "You opened it seven times.", + "D": "You opened the drawer only two times in total throughout the entire sequence of actions shown.", + "E": "You opened and closed the drawer nine times altogether during the task, counting each open action." + }, + "answer_label": "A", + "answer_text": "You opened it eight times.", + "q_id": "P22_117_4", + "correct_answer": "You opened it eight times.", + "original_options": { + "A": "You opened it eight times.", + "B": "You opened it six times.", + "C": "You opened it seven times.", + "D": "You opened the drawer only two times in total throughout the entire sequence of actions shown.", + "E": "You opened and closed the drawer nine times altogether during the task, counting each open action." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_12", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I open the refrigerator to get?", + "answer": "You took a bottle of ingredients.", + "question_chinese": "我打开冰箱拿了什么", + "answer_chinese": "你拿了一瓶食材", + "evidence": [ + { + "description": "You took a bottle of ingredients.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took a bottle of juice.", + "B": "You took a bag of vegetables.", + "C": "You opened the refrigerator to grab a can of beans from inside before closing it again.", + "D": "You opened the refrigerator and took out a package of fruit from the shelf before shutting it.", + "E": "You took a bottle of ingredients." + }, + "answer_label": "E", + "answer_text": "You took a bottle of ingredients.", + "q_id": "P22_117_12", + "correct_answer": "You took a bottle of ingredients.", + "original_options": { + "A": "You took a bottle of juice.", + "B": "You took a bag of vegetables.", + "C": "You opened the refrigerator to grab a can of beans from inside before closing it again.", + "D": "You opened the refrigerator and took out a package of fruit from the shelf before shutting it.", + "E": "You took a bottle of ingredients." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_11", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I take when I opened the refrigerator?", + "answer": "You took a cucumber.", + "question_chinese": "我打开冰箱拿了什么", + "answer_chinese": "你拿了一根黄瓜", + "evidence": [ + { + "description": "You took a bottle of ingredients.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:11" + } + }, + { + "description": "You took a cucumber.", + "timestamps": { + "start": "00:04:27", + "end": "00:04:32" + } + } + ], + "question_timestamps": "00:08:06", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took a cucumber.", + "B": "You took a zucchini.", + "C": "You took a tomato.", + "D": "You took a carrot out of the crisper drawer in the fridge.", + "E": "You took a pepper that was sitting on a plate inside the refrigerator." + }, + "answer_label": "A", + "answer_text": "You took a cucumber.", + "q_id": "P22_117_11", + "correct_answer": "You took a cucumber.", + "original_options": { + "A": "You took a cucumber.", + "B": "You took a zucchini.", + "C": "You took a tomato.", + "D": "You took a carrot out of the crisper drawer in the fridge.", + "E": "You took a pepper that was sitting on a plate inside the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "P22_117 ", + "video_length": "00:16:11", + "qid": "P22_117_1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do when I opened the cabinet?", + "answer": "You put a plate, a small cup, and a bag.", + "question_chinese": "我打开柜子做了什么", + "answer_chinese": "你放了一个盘子和一个小杯子还有一个袋子", + "evidence": [ + { + "description": "You put a plate.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:15" + } + }, + { + "description": "You put a small cup.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:44" + } + }, + { + "description": "You put a bag.", + "timestamps": { + "start": "00:02:05", + "end": "00:02:12" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You put a plate, a small cup, and a bag.", + "B": "You put a bowl and a cup.", + "C": "You put a plate and a container.", + "D": "You put a bowl, a small cup, and a jar inside the cabinet after opening it.", + "E": "You placed a plate, a large cup, and a container into the cabinet once you opened it." + }, + "answer_label": "A", + "answer_text": "You put a plate, a small cup, and a bag.", + "q_id": "P22_117_1", + "correct_answer": "You put a plate, a small cup, and a bag.", + "original_options": { + "A": "You put a plate, a small cup, and a bag.", + "B": "You put a bowl and a cup.", + "C": "You put a plate and a container.", + "D": "You put a bowl, a small cup, and a jar inside the cabinet after opening it.", + "E": "You placed a plate, a large cup, and a container into the cabinet once you opened it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do when I opened the cabinet?", + "answer": "You put a plate, a small cup, and a bag.", + "question_chinese": "我打开柜子做了什么", + "answer_chinese": "你放了一个盘子和一个小杯子还有一个袋子", + "evidence": [ + { + "description": "You put a plate.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:15" + } + }, + { + "description": "You put a small cup.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:44" + } + }, + { + "description": "You put a bag.", + "timestamps": { + "start": "00:02:05", + "end": "00:02:12" + } + } + ], + "question_timestamps": "00:08:06", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You put a bowl and a jar.", + "B": "You put a plate, a small cup, and a bag.", + "C": "You put a glass and a box.", + "D": "You put a dinner plate, a large mug, and a container into the cabinet when you opened it.", + "E": "You placed a cup, a plate, and a jar in the cabinet right after opening the door." + }, + "answer_label": "B", + "answer_text": "You put a plate, a small cup, and a bag.", + "q_id": "P22_117_2", + "correct_answer": "You put a plate, a small cup, and a bag.", + "original_options": { + "A": "You put a bowl and a jar.", + "B": "You put a plate, a small cup, and a bag.", + "C": "You put a glass and a box.", + "D": "You put a dinner plate, a large mug, and a container into the cabinet when you opened it.", + "E": "You placed a cup, a plate, and a jar in the cabinet right after opening the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 17 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_5", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoons did I wash?", + "answer": "You washed two.", + "question_chinese": "我洗了多少个勺子", + "answer_chinese": "你洗了两个", + "evidence": [ + { + "description": "You washed two.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You washed one.", + "B": "You washed three.", + "C": "You washed four spoons, rinsing each one and setting them aside to dry afterward.", + "D": "You washed five spoons in total, scrubbing them and then leaving them to drain near the sink.", + "E": "You washed two." + }, + "answer_label": "E", + "answer_text": "You washed two.", + "q_id": "P22_117_5", + "correct_answer": "You washed two.", + "original_options": { + "A": "You washed one.", + "B": "You washed three.", + "C": "You washed four spoons, rinsing each one and setting them aside to dry afterward.", + "D": "You washed five spoons in total, scrubbing them and then leaving them to drain near the sink.", + "E": "You washed two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 17 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_6", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many spoons did I wash?", + "answer": "You washed two.", + "question_chinese": "我洗了多少个勺子", + "answer_chinese": "你洗了两个", + "evidence": [ + { + "description": "You washed two.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:08:06", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You washed one.", + "B": "You washed three.", + "C": "You washed four spoons, cleaning each one before putting them aside to dry.", + "D": "You washed two.", + "E": "You washed six spoons altogether, rinsing them thoroughly and then setting them down to drain." + }, + "answer_label": "D", + "answer_text": "You washed two.", + "q_id": "P22_117_6", + "correct_answer": "You washed two.", + "original_options": { + "A": "You washed one.", + "B": "You washed three.", + "C": "You washed four spoons, cleaning each one before putting them aside to dry.", + "D": "You washed two.", + "E": "You washed six spoons altogether, rinsing them thoroughly and then setting them down to drain." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 15 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_7", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the cooking oil?", + "answer": "You can.", + "question_chinese": "我能够到食用油吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:08:18", + "end": "00:08:30" + } + } + ], + "question_timestamps": "00:12:09", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You cannot." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "P22_117_7", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P22_117", + "video_length": "00:16:11", + "qid": "P22_117_8", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the cooking oil?", + "answer": "You can.", + "question_chinese": "我能够到食用油吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:08:18", + "end": "00:08:30" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:13:32", + "end": "00:13:40" + } + } + ], + "question_timestamps": "00:16:11", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You cannot.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "P22_117_8", + "correct_answer": "You can.", + "original_options": { + "A": "You cannot.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + } +] diff --git a/QAs/P23_102-mcq.json b/QAs/P23_102-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..443113b9f2ea8cd1be372ddeae41857eafc16274 --- /dev/null +++ b/QAs/P23_102-mcq.json @@ -0,0 +1,1001 @@ +[ + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_6_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the upside-down cup relative to me?", + "answer": "It was on your left-hand side, next to the sink.", + "question_chinese": "倒扣的杯子在我的哪个方位？", + "answer_chinese": "在你的左手边，洗碗池的旁边", + "evidence": [ + { + "description": "It's beside the sink on your left hand side.", + "timestamps": { + "start": "00:05:52", + "end": "00:05:53" + } + } + ], + "question_timestamps": "00:06:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was on your right-hand side.", + "B": "It was in front of you.", + "C": "It was on your left-hand side, next to the sink.", + "D": "It was behind you, positioned close to the sink rather than off to your left.", + "E": "It was on your right-hand side, next to the sink, instead of on your left." + }, + "answer_label": "C", + "answer_text": "It was on your left-hand side, next to the sink.", + "q_id": "P23_102_6_T1", + "correct_answer": "It was on your left-hand side, next to the sink.", + "original_options": { + "A": "It was on your right-hand side.", + "B": "It was in front of you.", + "C": "It was on your left-hand side, next to the sink.", + "D": "It was behind you, positioned close to the sink rather than off to your left.", + "E": "It was on your right-hand side, next to the sink, instead of on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_6_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the upside-down cup relative to me?", + "answer": "It was next to the sink behind you.", + "question_chinese": "倒扣的杯子在我的哪个方位？", + "answer_chinese": "在你身后的洗碗池的旁边", + "evidence": [ + { + "description": "It's beside the sink on your left hand side.", + "timestamps": { + "start": "00:05:52", + "end": "00:05:53" + } + } + ], + "question_timestamps": "00:08:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Next to the sink, in front.", + "B": "Next to the sink, left side.", + "C": "It was next to the sink to your right, not positioned behind you where you were standing.", + "D": "It was next to the sink behind you.", + "E": "It was next to the sink in front of you, closer to your view instead of behind your position." + }, + "answer_label": "D", + "answer_text": "It was next to the sink behind you.", + "q_id": "P23_102_6_T2", + "correct_answer": "It was next to the sink behind you.", + "original_options": { + "A": "Next to the sink, in front.", + "B": "Next to the sink, left side.", + "C": "It was next to the sink to your right, not positioned behind you where you were standing.", + "D": "It was next to the sink behind you.", + "E": "It was next to the sink in front of you, closer to your view instead of behind your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_4_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the kitchen window relative to me?", + "answer": "The window is in front of you.", + "question_chinese": "厨房的窗子在我的哪个方位？", + "answer_chinese": "窗子在你的面前", + "evidence": [ + { + "description": "The window is in front of you", + "timestamps": { + "start": "00:05:17", + "end": "00:05:18" + } + } + ], + "question_timestamps": "00:05:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is behind you.", + "C": "The window is to your left.", + "D": "The window is to your right, not directly in front of you where you’re facing.", + "E": "The window is above you, positioned higher than your eye level rather than straight ahead." + }, + "answer_label": "A", + "answer_text": "The window is in front of you.", + "q_id": "P23_102_4_T1", + "correct_answer": "The window is in front of you.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is behind you.", + "C": "The window is to your left.", + "D": "The window is to your right, not directly in front of you where you’re facing.", + "E": "The window is above you, positioned higher than your eye level rather than straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_4_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the kitchen window relative to me?", + "answer": "The window is to your right.", + "question_chinese": "厨房的窗子在我的哪个方位？", + "answer_chinese": "窗子在你的右边", + "evidence": [ + { + "description": "The window is in front of you", + "timestamps": { + "start": "00:05:18", + "end": "00:05:20" + } + }, + { + "description": "Walk left towards the stove.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:11" + } + } + ], + "question_timestamps": "00:06:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is behind you, not on your right side as you stand there.", + "D": "The window is to your right.", + "E": "The window is above you, higher up instead of being off to your right." + }, + "answer_label": "D", + "answer_text": "The window is to your right.", + "q_id": "P23_102_4_T2", + "correct_answer": "The window is to your right.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is behind you, not on your right side as you stand there.", + "D": "The window is to your right.", + "E": "The window is above you, higher up instead of being off to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 14 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_1_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many dishes did I cook in the kitchen?", + "answer": "You cooked two dishes in the kitchen in total.", + "question_chinese": "我在厨房里一共烹饪了几道菜品？", + "answer_chinese": "你在厨房里一共烹饪了两道菜品。", + "evidence": [ + { + "description": "You cooked the meal.", + "timestamps": { + "start": "00:03:43", + "end": "00:03:44" + } + }, + { + "description": "You cooked a vegetable paste.", + "timestamps": { + "start": "00:15:35", + "end": "00:15:36" + } + } + ], + "question_timestamps": "00:16:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You cooked two dishes in the kitchen in total.", + "B": "You cooked one dish.", + "C": "You cooked three dishes.", + "D": "You cooked four dishes in the kitchen in total, counting all separate items you made.", + "E": "You cooked zero dishes in the kitchen in total, since you didn’t actually make any complete dish." + }, + "answer_label": "A", + "answer_text": "You cooked two dishes in the kitchen in total.", + "q_id": "P23_102_1_T1", + "correct_answer": "You cooked two dishes in the kitchen in total.", + "original_options": { + "A": "You cooked two dishes in the kitchen in total.", + "B": "You cooked one dish.", + "C": "You cooked three dishes.", + "D": "You cooked four dishes in the kitchen in total, counting all separate items you made.", + "E": "You cooked zero dishes in the kitchen in total, since you didn’t actually make any complete dish." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 18 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_1_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many dishes did I cook in the kitchen?", + "answer": "You cooked three dishes in the kitchen in total.", + "question_chinese": "我在厨房里一共烹饪了几道菜品？", + "answer_chinese": "你在厨房里一共烹饪了三道菜品。", + "evidence": [ + { + "description": "You cooked the meal.", + "timestamps": { + "start": "00:03:43", + "end": "00:03:44" + } + }, + { + "description": "You cooked a vegetable paste.", + "timestamps": { + "start": "00:15:35", + "end": "00:15:36" + } + }, + { + "description": "You cooked the dishes.", + "timestamps": { + "start": "00:47:26", + "end": "00:47:27" + } + } + ], + "question_timestamps": "00:48:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You cooked two dishes.", + "B": "You cooked three dishes in the kitchen in total.", + "C": "You cooked four dishes.", + "D": "You cooked one dish in the kitchen in total, with everything else just being prep work.", + "E": "You cooked five dishes in the kitchen in total, including every distinct dish you finished." + }, + "answer_label": "B", + "answer_text": "You cooked three dishes in the kitchen in total.", + "q_id": "P23_102_1_T2", + "correct_answer": "You cooked three dishes in the kitchen in total.", + "original_options": { + "A": "You cooked two dishes.", + "B": "You cooked three dishes in the kitchen in total.", + "C": "You cooked four dishes.", + "D": "You cooked one dish in the kitchen in total, with everything else just being prep work.", + "E": "You cooked five dishes in the kitchen in total, including every distinct dish you finished." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 15 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_8_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was prepping ingredients at the counter in front of me, how many kinds of seasonings did I prepare?", + "answer": "You prepped two types on the prep counter in front of you: green chili peppers and garlic.", + "question_chinese": "我在面前的料理台备菜时准备了几种调料？", + "answer_chinese": "你在面前的里前的料理台上准备了两种，分别是青辣椒和蒜。", + "evidence": [ + { + "description": "The garlic.", + "timestamps": { + "start": "00:35:41", + "end": "00:35:45" + } + }, + { + "description": "The namely green peppers.", + "timestamps": { + "start": "00:13:05", + "end": "00:13:10" + } + } + ], + "question_timestamps": "00:40:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You prepped one type: garlic.", + "B": "You prepped zero types.", + "C": "You prepped three types on the prep counter in front of you: green chili peppers, garlic, and dried chili peppers.", + "D": "You prepped two types on the prep counter in front of you: green chili peppers and garlic.", + "E": "You prepped four types on the prep counter in front of you: green chili peppers, garlic, dried chili peppers, and black pepper." + }, + "answer_label": "D", + "answer_text": "You prepped two types on the prep counter in front of you: green chili peppers and garlic.", + "q_id": "P23_102_8_T2", + "correct_answer": "You prepped two types on the prep counter in front of you: green chili peppers and garlic.", + "original_options": { + "A": "You prepped one type: garlic.", + "B": "You prepped zero types.", + "C": "You prepped three types on the prep counter in front of you: green chili peppers, garlic, and dried chili peppers.", + "D": "You prepped two types on the prep counter in front of you: green chili peppers and garlic.", + "E": "You prepped four types on the prep counter in front of you: green chili peppers, garlic, dried chili peppers, and black pepper." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 20, + 22 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_8_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was prepping ingredients at the counter in front of me, how many kinds of seasonings did I prepare?", + "answer": "You prepped three types on the prep counter in front of you: green chili peppers, dried chili peppers, and garlic.", + "question_chinese": "我在面前的料理台备菜时准备了几种调料？", + "answer_chinese": "你在面前的里前的料理台上准备了三种，分别是青辣椒，干辣椒和蒜。", + "evidence": [ + { + "description": "The garlic.", + "timestamps": { + "start": "00:35:41", + "end": "00:35:45" + } + }, + { + "description": "The namely green peppers.", + "timestamps": { + "start": "00:13:05", + "end": "00:13:10" + } + }, + { + "description": "The dried peppers.", + "timestamps": { + "start": "00:42:50", + "end": "00:42:55" + } + } + ], + "question_timestamps": "00:45:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You prepped one type: garlic.", + "B": "You prepped two types: chili and garlic.", + "C": "You prepped three types on the prep counter in front of you: green chili peppers, dried chili peppers, and garlic.", + "D": "You prepped four types on the prep counter: green chili peppers, dried chili peppers, garlic, and black pepper.", + "E": "You prepped five types on the prep counter: green chili peppers, dried chili peppers, garlic, black pepper, and salt." + }, + "answer_label": "C", + "answer_text": "You prepped three types on the prep counter in front of you: green chili peppers, dried chili peppers, and garlic.", + "q_id": "P23_102_8_T1", + "correct_answer": "You prepped three types on the prep counter in front of you: green chili peppers, dried chili peppers, and garlic.", + "original_options": { + "A": "You prepped one type: garlic.", + "B": "You prepped two types: chili and garlic.", + "C": "You prepped three types on the prep counter in front of you: green chili peppers, dried chili peppers, and garlic.", + "D": "You prepped four types on the prep counter: green chili peppers, dried chili peppers, garlic, and black pepper.", + "E": "You prepped five types on the prep counter: green chili peppers, dried chili peppers, garlic, black pepper, and salt." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 18, + 19 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_5_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the diced carrots I used for cooking?", + "answer": "You put it on the dining table in front of you.", + "question_chinese": "我把做饭用的胡萝卜粒放在了哪里？", + "answer_chinese": "你把它放在了面前的餐桌上。", + "evidence": [ + { + "description": "You put the diced carrots on the dining table.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put it behind you.", + "B": "You put it to your left.", + "C": "You put it on the dining table in front of you.", + "D": "You put it on the dining table to your right, not on the section directly in front of you.", + "E": "You put it on the dining table behind you, placing it out of your forward view." + }, + "answer_label": "C", + "answer_text": "You put it on the dining table in front of you.", + "q_id": "P23_102_5_T1", + "correct_answer": "You put it on the dining table in front of you.", + "original_options": { + "A": "You put it behind you.", + "B": "You put it to your left.", + "C": "You put it on the dining table in front of you.", + "D": "You put it on the dining table to your right, not on the section directly in front of you.", + "E": "You put it on the dining table behind you, placing it out of your forward view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 16 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_5_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the diced carrots I used for cooking?", + "answer": "You put it on the dining table behind you.", + "question_chinese": "我把做饭用的胡萝卜粒放在了哪里？", + "answer_chinese": "你把它放在了身后的餐桌上。", + "evidence": [ + { + "description": "You put the diced carrotson the dining table.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:05:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put it in front of you.", + "B": "You put it to your right.", + "C": "You put it on the dining table to your left, rather than placing it behind you.", + "D": "You put it on the dining table in front of you, keeping it within your direct line of sight.", + "E": "You put it on the dining table behind you." + }, + "answer_label": "E", + "answer_text": "You put it on the dining table behind you.", + "q_id": "P23_102_5_T2", + "correct_answer": "You put it on the dining table behind you.", + "original_options": { + "A": "You put it in front of you.", + "B": "You put it to your right.", + "C": "You put it on the dining table to your left, rather than placing it behind you.", + "D": "You put it on the dining table in front of you, keeping it within your direct line of sight.", + "E": "You put it on the dining table behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 19 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_7_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I get the lid for the pot I used to cook?", + "answer": "You took it from the spot against the wall to your right.", + "question_chinese": "我煮饭的锅盖是从哪里拿的？", + "answer_chinese": "是从你右边靠墙的位置拿的。", + "evidence": [ + { + "description": "It was taken from the position on the far right of the kitchen against the wall.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:07" + } + }, + { + "description": "You put the lid on the pot for cooking rice.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:12" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the wall ahead.", + "B": "From the wall to your left.", + "C": "You grabbed it from the spot against the wall behind you, not from the right side.", + "D": "You took it from the spot against the wall to your right.", + "E": "You picked it up from the spot against the wall in front of you, rather than from your right." + }, + "answer_label": "D", + "answer_text": "You took it from the spot against the wall to your right.", + "q_id": "P23_102_7_T1", + "correct_answer": "You took it from the spot against the wall to your right.", + "original_options": { + "A": "From the wall ahead.", + "B": "From the wall to your left.", + "C": "You grabbed it from the spot against the wall behind you, not from the right side.", + "D": "You took it from the spot against the wall to your right.", + "E": "You picked it up from the spot against the wall in front of you, rather than from your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 16, + 19 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_7_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I get the lid for the pot I used to cook?", + "answer": "You took it from the far-right spot against the wall in the kitchen behind you.", + "question_chinese": "我煮饭的锅盖是从哪里拿的？", + "answer_chinese": "是从你身后，厨房最右边靠墙的位置拿的。", + "evidence": [ + { + "description": "It was taken from the position on the far right of the kitchen against the wall.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:07" + } + }, + { + "description": "You put the lid on the pot for cooking rice.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:12" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took it from the far-right spot against the wall in the kitchen behind you.", + "B": "From the far-right spot in front of you.", + "C": "From the far-right spot to your left.", + "D": "You took it from the far-right spot against the wall in the kitchen in front of you, not behind.", + "E": "You grabbed it from the far-right spot against the wall in the kitchen to your right, not behind you." + }, + "answer_label": "A", + "answer_text": "You took it from the far-right spot against the wall in the kitchen behind you.", + "q_id": "P23_102_7_T2", + "correct_answer": "You took it from the far-right spot against the wall in the kitchen behind you.", + "original_options": { + "A": "You took it from the far-right spot against the wall in the kitchen behind you.", + "B": "From the far-right spot in front of you.", + "C": "From the far-right spot to your left.", + "D": "You took it from the far-right spot against the wall in the kitchen in front of you, not behind.", + "E": "You grabbed it from the far-right spot against the wall in the kitchen to your right, not behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 19 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_13", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I get the pot lid for cooking?", + "answer": "It was taken from the position on the far right of the kitchen against the wall.", + "question_chinese": "我煮饭的锅盖是从哪里拿的？", + "answer_chinese": "是从厨房最右边靠墙的位置拿的。", + "evidence": [ + { + "description": "It was taken from the position on the far right of the kitchen against the wall.", + "timestamps": { + "start": "00:06:06", + "end": "00:59:29" + } + }, + { + "description": "You put the lid on the pot for cooking rice.", + "timestamps": { + "start": "00:06:12", + "end": "00:59:29" + } + } + ], + "question_timestamps": "00:59:29", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was on the far left wall.", + "B": "It was taken from the position on the far right of the kitchen against the wall.", + "C": "It was in the middle of the kitchen.", + "D": "It was taken from the position on the far left of the kitchen against the wall, not the far right.", + "E": "It was retrieved from a central spot in the kitchen against the wall, rather than the far-right position." + }, + "answer_label": "B", + "answer_text": "It was taken from the position on the far right of the kitchen against the wall.", + "q_id": "P23_102_13", + "correct_answer": "It was taken from the position on the far right of the kitchen against the wall.", + "original_options": { + "A": "It was on the far left wall.", + "B": "It was taken from the position on the far right of the kitchen against the wall.", + "C": "It was in the middle of the kitchen.", + "D": "It was taken from the position on the far left of the kitchen against the wall, not the far right.", + "E": "It was retrieved from a central spot in the kitchen against the wall, rather than the far-right position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 20, + 18 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_2_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I take my pot from?", + "answer": "You took your pot out of the cabinet below the stove in front of you.", + "question_chinese": "我的锅是从哪里拿出来的？", + "answer_chinese": "你的锅是从面前灶台下方的柜子拿出来的。", + "evidence": [ + { + "description": "Your pot was taken out of the stove.", + "timestamps": { + "start": "00:03:56", + "end": "00:04:14" + } + } + ], + "question_timestamps": "00:04:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the cabinet behind the stove.", + "B": "From the left cabinet below stove.", + "C": "You took your pot out of the cabinet below the stove in front of you.", + "D": "You took your pot out of the cabinet below the stove to your right, not from the one in front of you.", + "E": "You pulled your pot from the cabinet below the stove behind you, rather than the cabinet directly in front." + }, + "answer_label": "C", + "answer_text": "You took your pot out of the cabinet below the stove in front of you.", + "q_id": "P23_102_2_T1", + "correct_answer": "You took your pot out of the cabinet below the stove in front of you.", + "original_options": { + "A": "From the cabinet behind the stove.", + "B": "From the left cabinet below stove.", + "C": "You took your pot out of the cabinet below the stove in front of you.", + "D": "You took your pot out of the cabinet below the stove to your right, not from the one in front of you.", + "E": "You pulled your pot from the cabinet below the stove behind you, rather than the cabinet directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 19 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_2_T2", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I take my pot from?", + "answer": "You took your pot out of the cabinet below the stove behind you.", + "question_chinese": "我的锅是从哪里拿出来的？", + "answer_chinese": "你的锅是从身后的灶台下方柜子拿出来的。", + "evidence": [ + { + "description": "Your pot was taken out of the stove.", + "timestamps": { + "start": "00:03:56", + "end": "00:04:14" + } + }, + { + "description": "Turn around and walk over to the table.", + "timestamps": { + "start": "00:04:15", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:59:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the cabinet below ahead.", + "B": "You took your pot out of the cabinet below the stove behind you.", + "C": "From the cabinet below right.", + "D": "You took your pot out of the cabinet below the stove to your left, not from behind you.", + "E": "You pulled your pot from the cabinet below the stove in front of you, rather than from the cabinet behind you." + }, + "answer_label": "B", + "answer_text": "You took your pot out of the cabinet below the stove behind you.", + "q_id": "P23_102_2_T2", + "correct_answer": "You took your pot out of the cabinet below the stove behind you.", + "original_options": { + "A": "From the cabinet below ahead.", + "B": "You took your pot out of the cabinet below the stove behind you.", + "C": "From the cabinet below right.", + "D": "You took your pot out of the cabinet below the stove to your left, not from behind you.", + "E": "You pulled your pot from the cabinet below the stove in front of you, rather than from the cabinet behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 21 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_3_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many cups of diced carrots did I scoop into the pot in front of me?", + "answer": "You scooped two cups of diced carrots into the pot in front of you.", + "question_chinese": "我舀了几杯胡萝卜粒到面前的锅里？", + "answer_chinese": "你舀了两杯胡萝卜粒到面前的锅里", + "evidence": [ + { + "description": "the first", + "timestamps": { + "start": "00:04:38", + "end": "00:04:39" + } + }, + { + "description": "the second", + "timestamps": { + "start": "00:04:43", + "end": "00:04:44" + } + } + ], + "question_timestamps": "00:04:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You scooped one cup of diced carrots.", + "B": "You scooped three cups of diced carrots.", + "C": "You scooped two cups of diced carrots into the pot in front of you.", + "D": "You scooped zero cups of diced carrots into the pot in front of you, adding none at all.", + "E": "You scooped four cups of diced carrots into the pot in front of you, not just two cups." + }, + "answer_label": "C", + "answer_text": "You scooped two cups of diced carrots into the pot in front of you.", + "q_id": "P23_102_3_T1", + "correct_answer": "You scooped two cups of diced carrots into the pot in front of you.", + "original_options": { + "A": "You scooped one cup of diced carrots.", + "B": "You scooped three cups of diced carrots.", + "C": "You scooped two cups of diced carrots into the pot in front of you.", + "D": "You scooped zero cups of diced carrots into the pot in front of you, adding none at all.", + "E": "You scooped four cups of diced carrots into the pot in front of you, not just two cups." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "P23_102", + "video_length": "00:59:29", + "qid": "P23_102_3_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many cups of diced carrots did I scoop into the pot in front of me?", + "answer": "You scooped three cups of diced carrots into the pot in front of you.", + "question_chinese": "我舀了几杯胡萝卜粒到面前的锅里？", + "answer_chinese": "你舀了三杯胡萝卜粒到面前的锅里", + "evidence": [ + { + "description": "the first", + "timestamps": { + "start": "00:04:39", + "end": "00:04:40" + } + }, + { + "description": "the second", + "timestamps": { + "start": "00:04:43", + "end": "00:04:44" + } + }, + { + "description": "the third", + "timestamps": { + "start": "00:04:53", + "end": "00:04:54" + } + } + ], + "question_timestamps": "00:04:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You scooped two cups of diced carrots.", + "B": "You scooped four cups of diced carrots.", + "C": "You scooped one cup of diced carrots into the pot in front of you, not three cups total.", + "D": "You scooped three cups of diced carrots into the pot in front of you.", + "E": "You scooped five cups of diced carrots into the pot in front of you, instead of three cups." + }, + "answer_label": "D", + "answer_text": "You scooped three cups of diced carrots into the pot in front of you.", + "q_id": "P23_102_3_T2", + "correct_answer": "You scooped three cups of diced carrots into the pot in front of you.", + "original_options": { + "A": "You scooped two cups of diced carrots.", + "B": "You scooped four cups of diced carrots.", + "C": "You scooped one cup of diced carrots into the pot in front of you, not three cups total.", + "D": "You scooped three cups of diced carrots into the pot in front of you.", + "E": "You scooped five cups of diced carrots into the pot in front of you, instead of three cups." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + } +] diff --git a/QAs/P25_101-mcq.json b/QAs/P25_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..31616fca1d56eb58610feded1e11aad35c340704 --- /dev/null +++ b/QAs/P25_101-mcq.json @@ -0,0 +1,1064 @@ +[ + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_8_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Earlier, when I was cooking, what was my movement path in the kitchen?", + "answer": "You moved forward first, then stayed in place without moving.", + "question_chinese": "刚才我在做饭时在厨房的运动轨迹是什么？", + "answer_chinese": "你首先向前运动，后来原地没有移动。", + "evidence": [ + { + "description": "You just moved forward first.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + }, + { + "description": "You remained in place without moving", + "timestamps": { + "start": "00:00:05", + "end": "00:06:00" + } + } + ], + "question_timestamps": "00:06:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved forward, then stepped back.", + "B": "You stayed still the entire time.", + "C": "You moved forward first, then stayed in place without moving.", + "D": "You moved forward first, then turned and walked to the side instead of staying in place.", + "E": "You moved forward at first, then continued shifting left and right rather than remaining stationary." + }, + "answer_label": "C", + "answer_text": "You moved forward first, then stayed in place without moving.", + "q_id": "P25_101_8_T1", + "correct_answer": "You moved forward first, then stayed in place without moving.", + "original_options": { + "A": "You moved forward, then stepped back.", + "B": "You stayed still the entire time.", + "C": "You moved forward first, then stayed in place without moving.", + "D": "You moved forward first, then turned and walked to the side instead of staying in place.", + "E": "You moved forward at first, then continued shifting left and right rather than remaining stationary." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_8_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Earlier, when I was cooking, what was my movement path in the kitchen?", + "answer": "You moved forward first, then stayed in place without moving, and finally shifted left and right.", + "question_chinese": "刚才我在做饭时在厨房的运动轨迹是什么？", + "answer_chinese": "你首先向前运动，后来原地没有移动，最后左右水平移动。", + "evidence": [ + { + "description": "You just moved forward first.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + }, + { + "description": "You remained in place without moving", + "timestamps": { + "start": "00:00:05", + "end": "00:06:00" + } + }, + { + "description": "You moved horizontally left and right.", + "timestamps": { + "start": "00:06:03", + "end": "00:07:00" + } + } + ], + "question_timestamps": "00:07:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Forward, then still, then only left.", + "B": "Forward, then still, then only right.", + "C": "You stepped forward, paused in place, then shifted only to the right instead of moving side to side.", + "D": "You moved forward, stayed still for a moment, then backed up slightly rather than shifting left and right.", + "E": "You moved forward first, then stayed in place without moving, and finally shifted left and right." + }, + "answer_label": "E", + "answer_text": "You moved forward first, then stayed in place without moving, and finally shifted left and right.", + "q_id": "P25_101_8_T2", + "correct_answer": "You moved forward first, then stayed in place without moving, and finally shifted left and right.", + "original_options": { + "A": "Forward, then still, then only left.", + "B": "Forward, then still, then only right.", + "C": "You stepped forward, paused in place, then shifted only to the right instead of moving side to side.", + "D": "You moved forward, stayed still for a moment, then backed up slightly rather than shifting left and right.", + "E": "You moved forward first, then stayed in place without moving, and finally shifted left and right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_15", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory in the kitchen just now when I was cooking?", + "answer": "You just moved forward first, then remained in place without moving, and finally moved horizontally left and right.", + "question_chinese": "刚才我在做饭时在厨房的运动轨迹是什么。", + "answer_chinese": "你首先向前运动，后来原地没有移动，最后左右水平移动。", + "evidence": [ + { + "description": "You just moved forward first.", + "timestamps": { + "start": "00:00:01", + "end": "00:07:15" + } + }, + { + "description": "You remained in place without moving", + "timestamps": { + "start": "00:00:05", + "end": "00:07:15" + } + }, + { + "description": "You moved horizontally left and right.", + "timestamps": { + "start": "00:06:03", + "end": "00:07:15" + } + } + ], + "question_timestamps": "00:07:15", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Forward, then back up, then still.", + "B": "Forward, then still, then up and down.", + "C": "You just moved forward first, then remained in place without moving, and finally moved horizontally left and right.", + "D": "You moved forward first, then shifted backward a little, and ended by moving side to side briefly.", + "E": "You went forward, paused without moving, and then stepped diagonally left instead of moving left and right." + }, + "answer_label": "C", + "answer_text": "You just moved forward first, then remained in place without moving, and finally moved horizontally left and right.", + "q_id": "P25_101_15", + "correct_answer": "You just moved forward first, then remained in place without moving, and finally moved horizontally left and right.", + "original_options": { + "A": "Forward, then back up, then still.", + "B": "Forward, then still, then up and down.", + "C": "You just moved forward first, then remained in place without moving, and finally moved horizontally left and right.", + "D": "You moved forward first, then shifted backward a little, and ended by moving side to side briefly.", + "E": "You went forward, paused without moving, and then stepped diagonally left instead of moving left and right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 17 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_2_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "After I lit it, where did I put the lighter?", + "answer": "You put the lighter on the cabinet to your left.", + "question_chinese": "在我点完火后，打火机被我放到了哪里？", + "answer_chinese": "你把打火机放在了你左边的柜子上。", + "evidence": [ + { + "description": "You were lighting the fire.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:17" + } + }, + { + "description": "You put the lighter on the cabinet beside the stove.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the cabinet in front of you.", + "B": "You put the lighter on the cabinet to your left.", + "C": "On the cabinet to your right.", + "D": "You placed the lighter on the cabinet behind you, not on the left-hand cabinet.", + "E": "You set the lighter down on the counter by the stove rather than on the cabinet to your left." + }, + "answer_label": "B", + "answer_text": "You put the lighter on the cabinet to your left.", + "q_id": "P25_101_2_T2", + "correct_answer": "You put the lighter on the cabinet to your left.", + "original_options": { + "A": "On the cabinet in front of you.", + "B": "You put the lighter on the cabinet to your left.", + "C": "On the cabinet to your right.", + "D": "You placed the lighter on the cabinet behind you, not on the left-hand cabinet.", + "E": "You set the lighter down on the counter by the stove rather than on the cabinet to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 19 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_2_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "After I lit it, where did I put the lighter?", + "answer": "You put the lighter on the cabinet in front of you.", + "question_chinese": "在我点完火后，打火机被我放到了哪里？", + "answer_chinese": "你把打火机放在了你前面的柜子上。", + "evidence": [ + { + "description": "You were lighting the fire.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:17" + } + }, + { + "description": "You put the lighter on the cabinet beside the stove.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the cabinet to your left.", + "B": "On the cabinet behind you.", + "C": "You placed the lighter on the cabinet to your right instead of putting it on the cabinet in front.", + "D": "You set the lighter down on the cabinet behind you rather than on the cabinet directly in front of you.", + "E": "You put the lighter on the cabinet in front of you." + }, + "answer_label": "E", + "answer_text": "You put the lighter on the cabinet in front of you.", + "q_id": "P25_101_2_T1", + "correct_answer": "You put the lighter on the cabinet in front of you.", + "original_options": { + "A": "On the cabinet to your left.", + "B": "On the cabinet behind you.", + "C": "You placed the lighter on the cabinet to your right instead of putting it on the cabinet in front.", + "D": "You set the lighter down on the cabinet behind you rather than on the cabinet directly in front of you.", + "E": "You put the lighter on the cabinet in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 20 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_3", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the lighter after I lit it?", + "answer": "You put the lighter on the cabinet beside the stove.", + "question_chinese": "在我点完火后，打火机被我放到了哪里？", + "answer_chinese": "你把打火机放在了灶台旁边的柜子上。", + "evidence": [ + { + "description": "You were lighting the fire.", + "timestamps": { + "start": "00:00:08", + "end": "00:07:15" + } + }, + { + "description": "You put the lighter on the cabinet beside the stove.", + "timestamps": { + "start": "00:00:17", + "end": "00:07:15" + } + } + ], + "question_timestamps": "00:07:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the counter in front of the stove.", + "B": "In the drawer under the stove.", + "C": "You set the lighter on the shelf above the stove rather than on the cabinet beside the stove.", + "D": "You left the lighter on the floor next to the stove instead of placing it on the nearby cabinet.", + "E": "You put the lighter on the cabinet beside the stove." + }, + "answer_label": "E", + "answer_text": "You put the lighter on the cabinet beside the stove.", + "q_id": "P25_101_3", + "correct_answer": "You put the lighter on the cabinet beside the stove.", + "original_options": { + "A": "On the counter in front of the stove.", + "B": "In the drawer under the stove.", + "C": "You set the lighter on the shelf above the stove rather than on the cabinet beside the stove.", + "D": "You left the lighter on the floor next to the stove instead of placing it on the nearby cabinet.", + "E": "You put the lighter on the cabinet beside the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 18, + 19 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_7_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to eat—where did I put my knife and fork?", + "answer": "It was in the drawer in front of you.", + "question_chinese": "我想要吃饭，我的刀叉放在哪里的？", + "answer_chinese": "放在你前面的抽屉里。", + "evidence": [ + { + "description": "The knives and forks are in the drawer in front of you.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the drawer to your left.", + "B": "In the drawer behind you.", + "C": "You put the knife and fork in the drawer to your right, not the one directly in front.", + "D": "It was in the drawer in front of you.", + "E": "You stored the knife and fork in the drawer behind you rather than in the drawer in front of you." + }, + "answer_label": "D", + "answer_text": "It was in the drawer in front of you.", + "q_id": "P25_101_7_T1", + "correct_answer": "It was in the drawer in front of you.", + "original_options": { + "A": "In the drawer to your left.", + "B": "In the drawer behind you.", + "C": "You put the knife and fork in the drawer to your right, not the one directly in front.", + "D": "It was in the drawer in front of you.", + "E": "You stored the knife and fork in the drawer behind you rather than in the drawer in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 20 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_7_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to eat—where did I put my knife and fork?", + "answer": "It was in the drawer to your front-left.", + "question_chinese": "我想要吃饭，我的刀叉放在哪里的？", + "answer_chinese": "放在你左前方的抽屉里。", + "evidence": [ + { + "description": "The knife and fork are in the drawer on your left.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:12" + } + } + ], + "question_timestamps": "00:06:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the drawer to your front-right.", + "B": "In the drawer to your back-left.", + "C": "It was in the drawer directly in front of you, not in the drawer at your front-left.", + "D": "You put it in the drawer to your back-right rather than the drawer positioned at your front-left.", + "E": "It was in the drawer to your front-left." + }, + "answer_label": "E", + "answer_text": "It was in the drawer to your front-left.", + "q_id": "P25_101_7_T2", + "correct_answer": "It was in the drawer to your front-left.", + "original_options": { + "A": "In the drawer to your front-right.", + "B": "In the drawer to your back-left.", + "C": "It was in the drawer directly in front of you, not in the drawer at your front-left.", + "D": "You put it in the drawer to your back-right rather than the drawer positioned at your front-left.", + "E": "It was in the drawer to your front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_13", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to have a meal. Where did I put my knife and fork?", + "answer": "Put it in the drawer on the left side of the stove.", + "question_chinese": "我想要吃饭，我的刀叉放在哪里的？", + "answer_chinese": "放在灶台左边的抽屉里。", + "evidence": [ + { + "description": "Put it in the drawer on the left side of the stove.", + "timestamps": { + "start": "00:01:05", + "end": "00:07:15" + } + } + ], + "question_timestamps": "00:07:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put it in the drawer on the left side of the stove.", + "B": "Put it in the drawer below the stove.", + "C": "Put it in the drawer above the dishwasher.", + "D": "Put it in the drawer on the right side of the sink instead of the drawer on the stove’s left.", + "E": "Put it in the cabinet beside the stove rather than in the drawer located on the left side of the stove." + }, + "answer_label": "A", + "answer_text": "Put it in the drawer on the left side of the stove.", + "q_id": "P25_101_13", + "correct_answer": "Put it in the drawer on the left side of the stove.", + "original_options": { + "A": "Put it in the drawer on the left side of the stove.", + "B": "Put it in the drawer below the stove.", + "C": "Put it in the drawer above the dishwasher.", + "D": "Put it in the drawer on the right side of the sink instead of the drawer on the stove’s left.", + "E": "Put it in the cabinet beside the stove rather than in the drawer located on the left side of the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 21, + 21 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_1_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many lighters did I see in the kitchen?", + "answer": "You saw one.", + "question_chinese": "我看到厨房里有几个打火机？", + "answer_chinese": "你看到一个。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw zero.", + "B": "You saw two.", + "C": "You saw one.", + "D": "You saw three lighters in the kitchen area in total, not just a single one.", + "E": "You saw four lighters in the kitchen altogether, so it was more than one." + }, + "answer_label": "C", + "answer_text": "You saw one.", + "q_id": "P25_101_1_T2", + "correct_answer": "You saw one.", + "original_options": { + "A": "You saw zero.", + "B": "You saw two.", + "C": "You saw one.", + "D": "You saw three lighters in the kitchen area in total, not just a single one.", + "E": "You saw four lighters in the kitchen altogether, so it was more than one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 14 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_1_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many lighters did I see in the kitchen?", + "answer": "You saw two.", + "question_chinese": "我看到厨房里有几个打火机？", + "answer_chinese": "你看到两个。", + "evidence": [ + { + "description": "The first lighter is on the table next to the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "I took the second lighter from the drawer.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:06:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one.", + "B": "You saw two.", + "C": "You saw three.", + "D": "You saw zero lighters in the kitchen, so there weren’t any to count.", + "E": "You saw four lighters in the kitchen altogether, which is more than two." + }, + "answer_label": "B", + "answer_text": "You saw two.", + "q_id": "P25_101_1_T1", + "correct_answer": "You saw two.", + "original_options": { + "A": "You saw one.", + "B": "You saw two.", + "C": "You saw three.", + "D": "You saw zero lighters in the kitchen, so there weren’t any to count.", + "E": "You saw four lighters in the kitchen altogether, which is more than two." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 13 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_6_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, how many pancakes do I see in the pan?", + "answer": "You saw six pancakes in the pan.", + "question_chinese": "我看到现在锅里有几个煎饼？", + "answer_chinese": "你看到锅里有六个煎饼。", + "evidence": [ + { + "description": "I put six pancakes into the pan.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:58" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You saw six pancakes in the pan.", + "B": "You saw three pancakes in the pan." + }, + "answer_label": "A", + "answer_text": "You saw six pancakes in the pan.", + "q_id": "P25_101_6_T2", + "correct_answer": "You saw six pancakes in the pan.", + "original_options": { + "A": "You saw six pancakes in the pan.", + "B": "You saw three pancakes in the pan." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_6_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, how many pancakes do I see in the pan?", + "answer": "You saw three pancakes in the pan.", + "question_chinese": "我看到现在锅里有几个煎饼？", + "answer_chinese": "你看到锅里有三个煎饼。", + "evidence": [ + { + "description": "I put three pancakes into the pan.", + "timestamps": { + "start": "00:04:53", + "end": "00:05:01" + } + } + ], + "question_timestamps": "00:06:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You saw four pancakes in the pan.", + "B": "You saw three pancakes in the pan." + }, + "answer_label": "B", + "answer_text": "You saw three pancakes in the pan.", + "q_id": "P25_101_6_T1", + "correct_answer": "You saw three pancakes in the pan.", + "original_options": { + "A": "You saw four pancakes in the pan.", + "B": "You saw three pancakes in the pan." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_5_T2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I directly touch the food-waste trash bag?", + "answer": "You couldn't touch it.", + "question_chinese": "我能直接触碰到厨余垃圾袋吗？", + "answer_chinese": "你不可以触碰到。", + "evidence": [ + { + "description": "The kitchen waste bag on the right side of the washbasin.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:08" + } + }, + { + "description": "You took two steps to the left from the place where the kitchen waste bag hangs.", + "timestamps": { + "start": "00:06:08", + "end": "00:06:10" + } + } + ], + "question_timestamps": "00:06:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes, you could touch it.", + "B": "You couldn't touch it.", + "C": "You could reach it easily.", + "D": "You could touch it if you leaned forward slightly from where you were standing.", + "E": "Your hand could reach the food-waste trash bag without needing to move closer." + }, + "answer_label": "B", + "answer_text": "You couldn't touch it.", + "q_id": "P25_101_5_T2", + "correct_answer": "You couldn't touch it.", + "original_options": { + "A": "Yes, you could touch it.", + "B": "You couldn't touch it.", + "C": "You could reach it easily.", + "D": "You could touch it if you leaned forward slightly from where you were standing.", + "E": "Your hand could reach the food-waste trash bag without needing to move closer." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_5_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I directly touch the food-waste trash bag?", + "answer": "You could touch it directly.", + "question_chinese": "我能直接触碰到厨余垃圾袋吗？", + "answer_chinese": "你可以直接触碰到。", + "evidence": [ + { + "description": "The kitchen waste bag is on my right-hand side.", + "timestamps": { + "start": "00:07:05", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:07:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You could touch it directly.", + "B": "No, you couldn't touch it.", + "C": "It was out of reach.", + "D": "No, the food-waste trash bag was too far away for you to touch directly.", + "E": "You couldn’t reach the food-waste trash bag with your hand from where you were standing." + }, + "answer_label": "A", + "answer_text": "You could touch it directly.", + "q_id": "P25_101_5_T1", + "correct_answer": "You could touch it directly.", + "original_options": { + "A": "You could touch it directly.", + "B": "No, you couldn't touch it.", + "C": "It was out of reach.", + "D": "No, the food-waste trash bag was too far away for you to touch directly.", + "E": "You couldn’t reach the food-waste trash bag with your hand from where you were standing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_3_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the plate, and can I reach it with my hand?", + "answer": "The plate was in the cabinet above the stovetop, and you couldn't get it right now.", + "question_chinese": "盘子被在哪里，我伸手能够拿到吗？", + "answer_chinese": "盘子在灶台上方的柜子里，你现在拿不到。", + "evidence": [ + { + "description": "You opened the cabinet.There are plates in the cabinet.", + "timestamps": { + "start": "00:00:55", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The plate was in the cabinet above the stovetop, and you could get it just by raising your hand.", + "B": "The plate was in the cabinet above the stovetop, and you couldn't get it right now." + }, + "answer_label": "B", + "answer_text": "The plate was in the cabinet above the stovetop, and you couldn't get it right now.", + "q_id": "P25_101_3_T2", + "correct_answer": "The plate was in the cabinet above the stovetop, and you couldn't get it right now.", + "original_options": { + "A": "The plate was in the cabinet above the stovetop, and you could get it just by raising your hand.", + "B": "The plate was in the cabinet above the stovetop, and you couldn't get it right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 19 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_3_T1", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the plate, and can I reach it with my hand?", + "answer": "The plate was in the cabinet above the stovetop, and you could get it just by raising your hand.", + "question_chinese": "盘子被在哪里，我伸手能够拿到吗？", + "answer_chinese": "盘子在灶台上方的柜子里，你抬手就能拿到。", + "evidence": [ + { + "description": "You opened the cabinet.There are plates in the cabinet.", + "timestamps": { + "start": "00:00:55", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The plate was in the cabinet above the stovetop, and you could get it just by raising your hand.", + "B": "The plate was in the cabinet above the stovetop, and you couldn't get it right now." + }, + "answer_label": "A", + "answer_text": "The plate was in the cabinet above the stovetop, and you could get it just by raising your hand.", + "q_id": "P25_101_3_T1", + "correct_answer": "The plate was in the cabinet above the stovetop, and you could get it just by raising your hand.", + "original_options": { + "A": "The plate was in the cabinet above the stovetop, and you could get it just by raising your hand.", + "B": "The plate was in the cabinet above the stovetop, and you couldn't get it right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_4_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many pancakes have I pan-fried in total?", + "answer": "So far, you’ve pan-fried a total of six pancakes.", + "question_chinese": "目前我一共煎了多少个煎饼？", + "answer_chinese": "目前你一共煎了六个煎饼。", + "evidence": [ + { + "description": "You put pancakes into the pot", + "timestamps": { + "start": "00:01:27", + "end": "00:01:45" + } + }, + { + "description": "You put six pancakes for the first time.", + "timestamps": { + "start": "00:01:48", + "end": "00:04:57" + } + } + ], + "question_timestamps": "00:04:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve pan-fried five pancakes.", + "B": "You’ve pan-fried seven pancakes.", + "C": "So far, you’ve pan-fried a total of four pancakes, finishing each one before starting the next.", + "D": "So far, you’ve pan-fried a total of eight pancakes, counting every cooked pancake you’ve made up to now.", + "E": "So far, you’ve pan-fried a total of six pancakes." + }, + "answer_label": "E", + "answer_text": "So far, you’ve pan-fried a total of six pancakes.", + "q_id": "P25_101_4_T1", + "correct_answer": "So far, you’ve pan-fried a total of six pancakes.", + "original_options": { + "A": "You’ve pan-fried five pancakes.", + "B": "You’ve pan-fried seven pancakes.", + "C": "So far, you’ve pan-fried a total of four pancakes, finishing each one before starting the next.", + "D": "So far, you’ve pan-fried a total of eight pancakes, counting every cooked pancake you’ve made up to now.", + "E": "So far, you’ve pan-fried a total of six pancakes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 20 + ] + }, + { + "video_uid": "P25_101", + "video_length": "00:07:15", + "qid": "P25_101_4_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many pancakes have I pan-fried in total?", + "answer": "So far, you’ve pan-fried a total of nine pancakes.", + "question_chinese": "目前我一共煎了多少个煎饼？", + "answer_chinese": "目前你一共煎了九个煎饼。", + "evidence": [ + { + "description": "You put pancakes into the pot", + "timestamps": { + "start": "00:01:27", + "end": "00:01:48" + } + }, + { + "description": "You put six pancakes for the first time.", + "timestamps": { + "start": "00:01:48", + "end": "00:04:57" + } + }, + { + "description": "You put three pancakes for the second time.", + "timestamps": { + "start": "00:05:00", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:07:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, you’ve pan-fried a total of nine pancakes.", + "B": "You’ve pan-fried eight pancakes.", + "C": "You’ve pan-fried ten pancakes.", + "D": "So far, you’ve pan-fried a total of seven pancakes, counting only the ones fully cooked through.", + "E": "So far, you’ve pan-fried a total of eleven pancakes, including all the pancakes you’ve cooked up to this point." + }, + "answer_label": "A", + "answer_text": "So far, you’ve pan-fried a total of nine pancakes.", + "q_id": "P25_101_4_T2", + "correct_answer": "So far, you’ve pan-fried a total of nine pancakes.", + "original_options": { + "A": "So far, you’ve pan-fried a total of nine pancakes.", + "B": "You’ve pan-fried eight pancakes.", + "C": "You’ve pan-fried ten pancakes.", + "D": "So far, you’ve pan-fried a total of seven pancakes, counting only the ones fully cooked through.", + "E": "So far, you’ve pan-fried a total of eleven pancakes, including all the pancakes you’ve cooked up to this point." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 21 + ] + } +] diff --git a/QAs/P25_107-mcq.json b/QAs/P25_107-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8978783ab4fbe3225401760557c0778afc870efc --- /dev/null +++ b/QAs/P25_107-mcq.json @@ -0,0 +1,616 @@ +[ + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_15", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I wash my hands during the cooking process?", + "answer": "Hands have been washed.", + "question_chinese": "做饭过程中我洗过手么？", + "answer_chinese": "洗过手。", + "evidence": [ + { + "description": "Hands have been washed.", + "timestamps": { + "start": "00:22:10", + "end": "00:22:13" + } + } + ], + "question_timestamps": "00:27:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Hands have not been washed.", + "B": "Hands have been washed." + }, + "answer_label": "B", + "answer_text": "Hands have been washed.", + "q_id": "P25_107_15", + "correct_answer": "Hands have been washed.", + "original_options": { + "A": "Hands have not been washed.", + "B": "Hands have been washed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_16", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I wash my hands during the cooking process?", + "answer": "Hands have been washed.", + "question_chinese": "做饭过程中我洗过手么？", + "answer_chinese": "洗过手。", + "evidence": [ + { + "description": "Hands have been washed.", + "timestamps": { + "start": "00:22:10", + "end": "00:22:13" + } + } + ], + "question_timestamps": "00:36:31", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Hands have been washed.", + "B": "Hands have not been washed." + }, + "answer_label": "A", + "answer_text": "Hands have been washed.", + "q_id": "P25_107_16", + "correct_answer": "Hands have been washed.", + "original_options": { + "A": "Hands have been washed.", + "B": "Hands have not been washed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_1_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I first walked in, was there a pot on the stovetop?", + "answer": "You saw a pot. The pot was to your right.", + "question_chinese": "我刚进门时灶台上有锅么？", + "answer_chinese": "你看到有锅。锅在你的右边。", + "evidence": [ + { + "description": "There is a pot.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You did not see a pot on the stovetop.", + "B": "You saw a pot. The pot was to your right." + }, + "answer_label": "B", + "answer_text": "You saw a pot. The pot was to your right.", + "q_id": "P25_107_1_T2", + "correct_answer": "You saw a pot. The pot was to your right.", + "original_options": { + "A": "You did not see a pot on the stovetop.", + "B": "You saw a pot. The pot was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_1_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I first walked in, was there a pot on the stovetop?", + "answer": "You saw a pot. The pot was in front of you.", + "question_chinese": "我刚进门时灶台上有锅么？", + "answer_chinese": "你看到有锅。锅在你的前方。", + "evidence": [ + { + "description": "There is a pot.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You saw a pot. The pot was in front of you.", + "B": "You did not see a pot on the stovetop." + }, + "answer_label": "A", + "answer_text": "You saw a pot. The pot was in front of you.", + "q_id": "P25_107_1_T1", + "correct_answer": "You saw a pot. The pot was in front of you.", + "original_options": { + "A": "You saw a pot. The pot was in front of you.", + "B": "You did not see a pot on the stovetop." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Was there a pot on the stove when I just entered the door?", + "answer": "There is a pot.", + "question_chinese": "我刚进门时灶台上有锅么？", + "answer_chinese": "有", + "evidence": [ + { + "description": "There is a pot.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:18:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "There is not a pot.", + "B": "There is a pot." + }, + "answer_label": "B", + "answer_text": "There is a pot.", + "q_id": "P25_107_2", + "correct_answer": "There is a pot.", + "original_options": { + "A": "There is not a pot.", + "B": "There is a pot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_17", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I used paper towels?", + "answer": "You have used it.", + "question_chinese": "我是否使用过擦手纸？", + "answer_chinese": "你使用过。", + "evidence": [ + { + "description": "Use paper towels to dry hands.", + "timestamps": { + "start": "00:24:56", + "end": "00:25:00" + } + } + ], + "question_timestamps": "00:27:11", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you have not used paper towels.", + "B": "You have used it." + }, + "answer_label": "B", + "answer_text": "You have used it.", + "q_id": "P25_107_17", + "correct_answer": "You have used it.", + "original_options": { + "A": "No, you have not used paper towels.", + "B": "You have used it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_18", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I used paper towels?", + "answer": "You have used it.", + "question_chinese": "我是否使用过擦手纸？", + "answer_chinese": "你已经用过它了。", + "evidence": [ + { + "description": "Use paper towels to dry hands.", + "timestamps": { + "start": "00:24:56", + "end": "00:25:00" + } + } + ], + "question_timestamps": "00:36:26", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You have used it.", + "B": "No, you have not used paper towels." + }, + "answer_label": "A", + "answer_text": "You have used it.", + "q_id": "P25_107_18", + "correct_answer": "You have used it.", + "original_options": { + "A": "You have used it.", + "B": "No, you have not used paper towels." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_13", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands?", + "answer": "No.", + "question_chinese": "我是否可以洗到手？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:20:43", + "end": "00:20:47" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:27:28", + "end": "00:27:28" + } + } + ], + "question_timestamps": "00:27:28", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "P25_107_13", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_14", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands?", + "answer": "No.", + "question_chinese": "我是否可以洗到手？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:20:43", + "end": "00:20:47" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:36:25", + "end": "00:36:25" + } + } + ], + "question_timestamps": "00:36:25", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "P25_107_14", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_11", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I sprinkle seasoning on the eggplant?", + "answer": "Yes.", + "question_chinese": "我是否往茄子上洒过调料", + "answer_chinese": "是的", + "evidence": [ + { + "description": "Sprinkle the seasoning.", + "timestamps": { + "start": "00:18:08", + "end": "00:18:12" + } + }, + { + "description": "Sprinkle the seasoning.", + "timestamps": { + "start": "00:19:58", + "end": "00:20:06" + } + }, + { + "description": "Sprinkle the seasoning.", + "timestamps": { + "start": "00:25:13", + "end": "00:25:18" + } + } + ], + "question_timestamps": "00:27:09", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "P25_107_11", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_12", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I sprinkle seasoning on the eggplant?", + "answer": "Yes.", + "question_chinese": "我是否往茄子上洒过调料", + "answer_chinese": "是的", + "evidence": [ + { + "description": "Sprinkle the seasoning.", + "timestamps": { + "start": "00:18:08", + "end": "00:18:12" + } + }, + { + "description": "Sprinkle the seasoning.", + "timestamps": { + "start": "00:19:58", + "end": "00:20:06" + } + }, + { + "description": "Sprinkle the seasoning.", + "timestamps": { + "start": "00:25:13", + "end": "00:25:18" + } + } + ], + "question_timestamps": "00:36:45", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "P25_107_12", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_7", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I ignite it?", + "answer": "Yes.", + "question_chinese": "我是否点火了？", + "answer_chinese": "点火了。", + "evidence": [ + { + "description": "Ignition.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:09:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "P25_107_7", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P25_107", + "video_length": "00:36:52", + "qid": "P25_107_8", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I ignite it?", + "answer": "Yes.", + "question_chinese": "我是否点火了？", + "answer_chinese": "点火了。", + "evidence": [ + { + "description": "Ignition.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:18:34", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "P25_107_8", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/P28_101-mcq.json b/QAs/P28_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..948084316a09107fc39bb77985c4c80f366e5ac7 --- /dev/null +++ b/QAs/P28_101-mcq.json @@ -0,0 +1,1099 @@ +[ + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do before cutting the tomatoes?", + "answer": "You washed the tomatoes.", + "question_chinese": "我切西红柿之前做了什么", + "answer_chinese": "你洗了西红柿", + "evidence": [ + { + "description": "You washed the tomatoes.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You dried the tomatoes.", + "B": "You peeled the tomatoes.", + "C": "You washed the tomatoes.", + "D": "You inspected the tomatoes for bruises before you started slicing them on the cutting board.", + "E": "You arranged the tomatoes on the cutting board, then you began cutting them into pieces." + }, + "answer_label": "C", + "answer_text": "You washed the tomatoes.", + "q_id": "P28_101_7", + "correct_answer": "You washed the tomatoes.", + "original_options": { + "A": "You dried the tomatoes.", + "B": "You peeled the tomatoes.", + "C": "You washed the tomatoes.", + "D": "You inspected the tomatoes for bruises before you started slicing them on the cutting board.", + "E": "You arranged the tomatoes on the cutting board, then you began cutting them into pieces." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do before cutting the tomatoes?", + "answer": "You washed the tomatoes.", + "question_chinese": "我切西红柿之前做了什么", + "answer_chinese": "你洗了西红柿", + "evidence": [ + { + "description": "You washed the tomatoes.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You dried the tomatoes.", + "B": "You washed the tomatoes.", + "C": "You picked the tomatoes up.", + "D": "You inspected the tomatoes for bruises before you started cutting them on the board.", + "E": "You rinsed the cutting board, then you moved on to cutting the tomatoes." + }, + "answer_label": "B", + "answer_text": "You washed the tomatoes.", + "q_id": "P28_101_8", + "correct_answer": "You washed the tomatoes.", + "original_options": { + "A": "You dried the tomatoes.", + "B": "You washed the tomatoes.", + "C": "You picked the tomatoes up.", + "D": "You inspected the tomatoes for bruises before you started cutting them on the board.", + "E": "You rinsed the cutting board, then you moved on to cutting the tomatoes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 13 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_5_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the remaining ingredients?", + "answer": "You put it into the bag, and the bag is in front of you.", + "question_chinese": "我将剩下的食材放哪了？", + "answer_chinese": "你把它放进袋子里了，袋子在你前面。", + "evidence": [ + { + "description": "You wrapped it up", + "timestamps": { + "start": "00:06:20", + "end": "00:06:37" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the bag behind you.", + "B": "You put it into the bag, and the bag is in front of you.", + "C": "Into the bag to your left.", + "D": "You put it into the bag, and the bag is to your right side.", + "E": "You put it into the bag, and the bag is positioned above you rather than in front." + }, + "answer_label": "B", + "answer_text": "You put it into the bag, and the bag is in front of you.", + "q_id": "P28_101_5_T1", + "correct_answer": "You put it into the bag, and the bag is in front of you.", + "original_options": { + "A": "Into the bag behind you.", + "B": "You put it into the bag, and the bag is in front of you.", + "C": "Into the bag to your left.", + "D": "You put it into the bag, and the bag is to your right side.", + "E": "You put it into the bag, and the bag is positioned above you rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 17 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_5_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the remaining ingredients?", + "answer": "You put it into the refrigerator, and the refrigerator is in front of you.", + "question_chinese": "我将剩下的食材放哪了？", + "answer_chinese": "你把它放进冰箱里了，冰箱在你前面。", + "evidence": [ + { + "description": "You put it in the refrigerator.", + "timestamps": { + "start": "00:06:41", + "end": "00:06:42" + } + } + ], + "question_timestamps": "00:06:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the refrigerator behind you.", + "B": "In the refrigerator to your left.", + "C": "You put it into the refrigerator, and the refrigerator is to your right side.", + "D": "You put it into the refrigerator, and the refrigerator is above you rather than in front.", + "E": "You put it into the refrigerator, and the refrigerator is in front of you." + }, + "answer_label": "E", + "answer_text": "You put it into the refrigerator, and the refrigerator is in front of you.", + "q_id": "P28_101_5_T2", + "correct_answer": "You put it into the refrigerator, and the refrigerator is in front of you.", + "original_options": { + "A": "In the refrigerator behind you.", + "B": "In the refrigerator to your left.", + "C": "You put it into the refrigerator, and the refrigerator is to your right side.", + "D": "You put it into the refrigerator, and the refrigerator is above you rather than in front.", + "E": "You put it into the refrigerator, and the refrigerator is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 16 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_9", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the remaining ingredients?", + "answer": "You wrapped it up and put it in the refrigerator.", + "question_chinese": "我将剩下的食材放哪了", + "answer_chinese": "你把它包好放进了冰箱里", + "evidence": [ + { + "description": "You wrapped it up and put it in the refrigerator.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:45" + } + } + ], + "question_timestamps": "00:11:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You left it on the counter.", + "B": "You put it in the cabinet.", + "C": "You covered it and set it on the counter so it would be ready to use again soon.", + "D": "You wrapped it up and put it in the refrigerator.", + "E": "You stored it in a cabinet after wrapping it up, then closed the door and left it there." + }, + "answer_label": "D", + "answer_text": "You wrapped it up and put it in the refrigerator.", + "q_id": "P28_101_9", + "correct_answer": "You wrapped it up and put it in the refrigerator.", + "original_options": { + "A": "You left it on the counter.", + "B": "You put it in the cabinet.", + "C": "You covered it and set it on the counter so it would be ready to use again soon.", + "D": "You wrapped it up and put it in the refrigerator.", + "E": "You stored it in a cabinet after wrapping it up, then closed the door and left it there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the cabinet?", + "answer": "You opened it three times.", + "question_chinese": "我打开了多少次柜子", + "answer_chinese": "你打开了三次", + "evidence": [ + { + "description": "First opening.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + }, + { + "description": "Open for the second time.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:44" + } + }, + { + "description": "Open for the third time.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You opened it once.", + "B": "You opened it twice.", + "C": "You opened it four times, going back to the cabinet again after you had already closed it.", + "D": "You opened it three times.", + "E": "You opened it five times, repeatedly returning to the cabinet to grab items and then shutting it again." + }, + "answer_label": "D", + "answer_text": "You opened it three times.", + "q_id": "P28_101_1", + "correct_answer": "You opened it three times.", + "original_options": { + "A": "You opened it once.", + "B": "You opened it twice.", + "C": "You opened it four times, going back to the cabinet again after you had already closed it.", + "D": "You opened it three times.", + "E": "You opened it five times, repeatedly returning to the cabinet to grab items and then shutting it again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the cabinet?", + "answer": "You opened it four times.", + "question_chinese": "我打开了多少次柜子", + "answer_chinese": "你打开了四次", + "evidence": [ + { + "description": "First opening.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + }, + { + "description": "Open for the second time.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:44" + } + }, + { + "description": "Open for the third time.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:25" + } + }, + { + "description": "You opened it for the fourth time.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:45" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You opened it twice.", + "B": "You opened it three times.", + "C": "You opened it four times.", + "D": "You opened it five times, reopening the cabinet again after each time you closed it.", + "E": "You opened it six times, making several trips back to the cabinet while preparing things." + }, + "answer_label": "C", + "answer_text": "You opened it four times.", + "q_id": "P28_101_2", + "correct_answer": "You opened it four times.", + "original_options": { + "A": "You opened it twice.", + "B": "You opened it three times.", + "C": "You opened it four times.", + "D": "You opened it five times, reopening the cabinet again after each time you closed it.", + "E": "You opened it six times, making several trips back to the cabinet while preparing things." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 15 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_3_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Into how many pieces did I cut the tomato?", + "answer": "You cut the tomato into 2 pieces; it is in front of you.", + "question_chinese": "我把西红柿切成了几块？", + "answer_chinese": "你把西红柿切成了2块，它在你前面。", + "evidence": [ + { + "description": "You cut the tomato into two pieces.", + "timestamps": { + "start": "00:01:59", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:02:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You cut it into 1 piece.", + "B": "You cut it into 3 pieces.", + "C": "You cut the tomato into 2 pieces; it is in front of you.", + "D": "You cut the tomato into 4 pieces; it is in front of you on the work surface.", + "E": "You cut the tomato into 5 pieces; it is in front of you, arranged as several smaller sections." + }, + "answer_label": "C", + "answer_text": "You cut the tomato into 2 pieces; it is in front of you.", + "q_id": "P28_101_3_T2", + "correct_answer": "You cut the tomato into 2 pieces; it is in front of you.", + "original_options": { + "A": "You cut it into 1 piece.", + "B": "You cut it into 3 pieces.", + "C": "You cut the tomato into 2 pieces; it is in front of you.", + "D": "You cut the tomato into 4 pieces; it is in front of you on the work surface.", + "E": "You cut the tomato into 5 pieces; it is in front of you, arranged as several smaller sections." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_3_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Into how many pieces did I cut the tomato?", + "answer": "You cut the tomato into 3 pieces; it is in front of you.", + "question_chinese": "我把西红柿切成了几块？", + "answer_chinese": "我把西红柿切成了3块，它在你前面。", + "evidence": [ + { + "description": "You cut the tomato into three pieces.", + "timestamps": { + "start": "00:03:59", + "end": "00:04:01" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You cut it into 2 pieces.", + "B": "You cut the tomato into 3 pieces; it is in front of you.", + "C": "You cut it into 4 pieces.", + "D": "You cut the tomato into 1 piece; it is in front of you and still looks mostly whole.", + "E": "You cut the tomato into 5 pieces; it is in front of you with multiple small segments visible." + }, + "answer_label": "B", + "answer_text": "You cut the tomato into 3 pieces; it is in front of you.", + "q_id": "P28_101_3_T1", + "correct_answer": "You cut the tomato into 3 pieces; it is in front of you.", + "original_options": { + "A": "You cut it into 2 pieces.", + "B": "You cut the tomato into 3 pieces; it is in front of you.", + "C": "You cut it into 4 pieces.", + "D": "You cut the tomato into 1 piece; it is in front of you and still looks mostly whole.", + "E": "You cut the tomato into 5 pieces; it is in front of you with multiple small segments visible." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_4_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many plates did I take?", + "answer": "You took 0 plates, and the plate is in front of you.", + "question_chinese": "我拿了几个盘子？", + "answer_chinese": "你拿了0个盘子，盘子在你前面。", + "evidence": [ + { + "description": "I took two plates.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took 1 plate.", + "B": "You took 2 plates.", + "C": "You took 0 plates, and the plate is in front of you.", + "D": "You took 3 plates, and the plates are in front of you where you set them down.", + "E": "You took 4 plates, and the plates are in front of you after you brought them over." + }, + "answer_label": "C", + "answer_text": "You took 0 plates, and the plate is in front of you.", + "q_id": "P28_101_4_T1", + "correct_answer": "You took 0 plates, and the plate is in front of you.", + "original_options": { + "A": "You took 1 plate.", + "B": "You took 2 plates.", + "C": "You took 0 plates, and the plate is in front of you.", + "D": "You took 3 plates, and the plates are in front of you where you set them down.", + "E": "You took 4 plates, and the plates are in front of you after you brought them over." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_4_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many plates did I take?", + "answer": "You took 1 plate, and the plate is in front of you.", + "question_chinese": "我拿了几个盘子？", + "answer_chinese": "你拿了1个盘子，盘子在你前面。", + "evidence": [ + { + "description": "I took one plate.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:49" + } + } + ], + "question_timestamps": "00:00:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took 0 plates.", + "B": "You took 2 plates.", + "C": "You took 3 plates, and the plates are in front of you after you grabbed extras as well.", + "D": "You took 1 plate, and the plate is in front of you.", + "E": "You took 4 plates, and the plates are in front of you even though you carried several at once." + }, + "answer_label": "D", + "answer_text": "You took 1 plate, and the plate is in front of you.", + "q_id": "P28_101_4_T2", + "correct_answer": "You took 1 plate, and the plate is in front of you.", + "original_options": { + "A": "You took 0 plates.", + "B": "You took 2 plates.", + "C": "You took 3 plates, and the plates are in front of you after you grabbed extras as well.", + "D": "You took 1 plate, and the plate is in front of you.", + "E": "You took 4 plates, and the plates are in front of you even though you carried several at once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_13", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after finishing plating?", + "answer": "You washed your hands.", + "question_chinese": "我摆完盘之后做了什么", + "answer_chinese": "你洗了手", + "evidence": [ + { + "description": "You washed your hands.", + "timestamps": { + "start": "00:08:40", + "end": "00:08:53" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put away the ingredients.", + "B": "You washed your hands.", + "C": "You dried the dishes carefully.", + "D": "You swept the floor thoroughly after plating, cleaning up crumbs and any mess around your feet.", + "E": "You checked the oven settings after finishing plating, making sure the temperature and timer were correct." + }, + "answer_label": "B", + "answer_text": "You washed your hands.", + "q_id": "P28_101_13", + "correct_answer": "You washed your hands.", + "original_options": { + "A": "You put away the ingredients.", + "B": "You washed your hands.", + "C": "You dried the dishes carefully.", + "D": "You swept the floor thoroughly after plating, cleaning up crumbs and any mess around your feet.", + "E": "You checked the oven settings after finishing plating, making sure the temperature and timer were correct." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_14", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after finishing plating?", + "answer": "You added seasoning.", + "question_chinese": "我摆完盘之后做了什么", + "answer_chinese": "你加了调味料", + "evidence": [ + { + "description": "You washed your hands.", + "timestamps": { + "start": "00:08:40", + "end": "00:08:53" + } + }, + { + "description": "You added seasoning.", + "timestamps": { + "start": "00:09:05", + "end": "00:09:38" + } + } + ], + "question_timestamps": "00:11:57", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You served the dishes.", + "B": "You cleaned the countertop.", + "C": "You started washing up, clearing the plates and wiping the counter where you had been plating.", + "D": "You moved on to set the table, placing utensils and napkins near the plated food for serving.", + "E": "You added seasoning." + }, + "answer_label": "E", + "answer_text": "You added seasoning.", + "q_id": "P28_101_14", + "correct_answer": "You added seasoning.", + "original_options": { + "A": "You served the dishes.", + "B": "You cleaned the countertop.", + "C": "You started washing up, clearing the plates and wiping the counter where you had been plating.", + "D": "You moved on to set the table, placing utensils and napkins near the plated food for serving.", + "E": "You added seasoning." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_11", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the banana?", + "answer": "You can.", + "question_chinese": "我能够到香蕉吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:06:47", + "end": "00:07:30" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You cannot." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "P28_101_11", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_12", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the banana?", + "answer": "You can.", + "question_chinese": "我能够到香蕉吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:06:47", + "end": "00:07:30" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:09:15", + "end": "00:09:25" + } + } + ], + "question_timestamps": "00:11:57", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You cannot.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "P28_101_12", + "correct_answer": "You can.", + "original_options": { + "A": "You cannot.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_2_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the kettle?", + "answer": "You can reach the kettle; it is in front of you.", + "question_chinese": "我能够得到烧水壶吗？", + "answer_chinese": "你能够得到烧水壶，它在你前面。", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the kettle; it is in front of you.", + "B": "You cannot reach the kettle; it is in front of you." + }, + "answer_label": "A", + "answer_text": "You can reach the kettle; it is in front of you.", + "q_id": "P28_101_2_T1", + "correct_answer": "You can reach the kettle; it is in front of you.", + "original_options": { + "A": "You can reach the kettle; it is in front of you.", + "B": "You cannot reach the kettle; it is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_2_T2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the kettle?", + "answer": "You cannot reach the kettle; it is behind you.", + "question_chinese": "我能够得到烧水壶吗？", + "answer_chinese": "你不能够得到烧水壶，它在你后面·。", + "evidence": [ + { + "description": "You can.not.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:00:49", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the kettle; it is behind you.", + "B": "You cannot reach the kettle; it is behind you." + }, + "answer_label": "B", + "answer_text": "You cannot reach the kettle; it is behind you.", + "q_id": "P28_101_2_T2", + "correct_answer": "You cannot reach the kettle; it is behind you.", + "original_options": { + "A": "You can reach the kettle; it is behind you.", + "B": "You cannot reach the kettle; it is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_3", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get a kettle?", + "answer": "You can.", + "question_chinese": "我能够得到烧水壶吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "P28_101_3", + "correct_answer": "You can.", + "original_options": { + "A": "You cannot.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_4", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get a kettle?", + "answer": "You can.", + "question_chinese": "我能够得到烧水壶吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:25" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:03:38", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You cannot." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "P28_101_4", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_8_T2", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I touch the faucet?", + "answer": "You can touch the faucet; it is in front of you.", + "question_chinese": "我能碰到水龙头吗？", + "answer_chinese": "你能碰到水龙头，它在你前面。", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:09:47", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:09:52", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You cannot touch the faucet; it is in front of you.", + "B": "You can touch the faucet; it is in front of you." + }, + "answer_label": "B", + "answer_text": "You can touch the faucet; it is in front of you.", + "q_id": "P28_101_8_T2", + "correct_answer": "You can touch the faucet; it is in front of you.", + "original_options": { + "A": "You cannot touch the faucet; it is in front of you.", + "B": "You can touch the faucet; it is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "P28_101", + "video_length": "00:11:57", + "qid": "P28_101_8_T1", + "qid_num": 24, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I touch the faucet?", + "answer": "You cannot touch the faucet; it is behind you.", + "question_chinese": "我能碰到水龙头吗？", + "answer_chinese": "你不能碰到水龙头，它在你后面。", + "evidence": [ + { + "description": "You can not.", + "timestamps": { + "start": "00:11:45", + "end": "00:11:48" + } + } + ], + "question_timestamps": "00:11:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You cannot touch the faucet; it is behind you.", + "B": "You can touch the faucet; it is behind you." + }, + "answer_label": "A", + "answer_text": "You cannot touch the faucet; it is behind you.", + "q_id": "P28_101_8_T1", + "correct_answer": "You cannot touch the faucet; it is behind you.", + "original_options": { + "A": "You cannot touch the faucet; it is behind you.", + "B": "You can touch the faucet; it is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + } +] diff --git a/QAs/P28_108-mcq.json b/QAs/P28_108-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1c70eb1adc932d2d28a9bf8560a397782d85be93 --- /dev/null +++ b/QAs/P28_108-mcq.json @@ -0,0 +1,677 @@ +[ + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_6_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How did I get to the side of the fridge?", + "answer": "You can turn right first, then turn around. The refrigerator is to your back-right.", + "question_chinese": "我如何到达冰箱旁？", + "answer_chinese": "你可以先右转，然后再转身。冰箱在你的右后方。", + "evidence": [ + { + "description": "You can turn around.", + "timestamps": { + "start": "00:02:45", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:05:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then turn around.", + "B": "You can turn right first, then turn around. The refrigerator is to your back-right.", + "C": "Turn around; it’s directly in front.", + "D": "You can turn left first, then turn around; the refrigerator ends up to your back-right.", + "E": "You can turn right first, then turn around; the refrigerator will be positioned to your front-right." + }, + "answer_label": "B", + "answer_text": "You can turn right first, then turn around. The refrigerator is to your back-right.", + "q_id": "P28_108_6_T1", + "correct_answer": "You can turn right first, then turn around. The refrigerator is to your back-right.", + "original_options": { + "A": "Turn left, then turn around.", + "B": "You can turn right first, then turn around. The refrigerator is to your back-right.", + "C": "Turn around; it’s directly in front.", + "D": "You can turn left first, then turn around; the refrigerator ends up to your back-right.", + "E": "You can turn right first, then turn around; the refrigerator will be positioned to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 15, + 16 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_6_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How did I get to the side of the fridge?", + "answer": "You can turn around—the refrigerator is directly behind you.", + "question_chinese": "我如何到达冰箱旁？", + "answer_chinese": "你可以转身，冰箱在你的正后方。", + "evidence": [ + { + "description": "You can turn around.", + "timestamps": { + "start": "00:02:45", + "end": "00:02:50" + } + }, + { + "description": "You can turn around.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:40" + } + } + ], + "question_timestamps": "00:06:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around; it’s directly in front.", + "B": "Turn right first, then turn around.", + "C": "You can turn around—the refrigerator is directly behind you.", + "D": "You can turn left first, then turn around; the refrigerator will be to your back-left.", + "E": "You can turn right first, then turn around; the refrigerator ends up to your front-right." + }, + "answer_label": "C", + "answer_text": "You can turn around—the refrigerator is directly behind you.", + "q_id": "P28_108_6_T2", + "correct_answer": "You can turn around—the refrigerator is directly behind you.", + "original_options": { + "A": "Turn around; it’s directly in front.", + "B": "Turn right first, then turn around.", + "C": "You can turn around—the refrigerator is directly behind you.", + "D": "You can turn left first, then turn around; the refrigerator will be to your back-left.", + "E": "You can turn right first, then turn around; the refrigerator ends up to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 15 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_9", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do to the cooked food?", + "answer": "You pour it into the pot and stir-fry.", + "question_chinese": "我对烹饪过的食物做了什么", + "answer_chinese": "你将它倒入锅中翻炒", + "evidence": [ + { + "description": "You pour it into the pot and stir-fry.", + "timestamps": { + "start": "00:03:40", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:05:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You bake it on the counter.", + "B": "You chill it off the stove.", + "C": "You put it into a bowl first and then roast it until it is heated through.", + "D": "You toss it into the microwave and then boil it afterward instead of stir-frying.", + "E": "You pour it into the pot and stir-fry." + }, + "answer_label": "E", + "answer_text": "You pour it into the pot and stir-fry.", + "q_id": "P28_108_9", + "correct_answer": "You pour it into the pot and stir-fry.", + "original_options": { + "A": "You bake it on the counter.", + "B": "You chill it off the stove.", + "C": "You put it into a bowl first and then roast it until it is heated through.", + "D": "You toss it into the microwave and then boil it afterward instead of stir-frying.", + "E": "You pour it into the pot and stir-fry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 14 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_10", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do to the cooked food?", + "answer": "You put it on the other side.", + "question_chinese": "我对烹饪过的食物做了什么", + "answer_chinese": "你把它放到了另一边", + "evidence": [ + { + "description": "You pour it into the pot and stir-fry.", + "timestamps": { + "start": "00:03:40", + "end": "00:04:00" + } + }, + { + "description": "You put it on the other side.", + "timestamps": { + "start": "00:05:50", + "end": "00:05:56" + } + } + ], + "question_timestamps": "00:06:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You put it on the other side.", + "B": "You put it in front.", + "C": "You put it behind.", + "D": "You put it in front of the other side rather than moving it fully to the other side.", + "E": "You put it behind the other side and left it there instead of placing it on the other side." + }, + "answer_label": "A", + "answer_text": "You put it on the other side.", + "q_id": "P28_108_10", + "correct_answer": "You put it on the other side.", + "original_options": { + "A": "You put it on the other side.", + "B": "You put it in front.", + "C": "You put it behind.", + "D": "You put it in front of the other side rather than moving it fully to the other side.", + "E": "You put it behind the other side and left it there instead of placing it on the other side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 19 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_3_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I open that package?", + "answer": "You didn’t open it. It’s on the table to your back-right.", + "question_chinese": "我打开那个包装了吗？", + "answer_chinese": "你没有打开,在你右后方的桌子上。", + "evidence": [ + { + "description": "You didn't turn it on.", + "timestamps": { + "start": "00:00:20", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You didn’t open it. It’s on the table to your back-right.", + "B": "You opened it. It’s in your hand." + }, + "answer_label": "A", + "answer_text": "You didn’t open it. It’s on the table to your back-right.", + "q_id": "P28_108_3_T1", + "correct_answer": "You didn’t open it. It’s on the table to your back-right.", + "original_options": { + "A": "You didn’t open it. It’s on the table to your back-right.", + "B": "You opened it. It’s in your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_3_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I open that package?", + "answer": "You opened it. It’s in your hand.", + "question_chinese": "我打开那个包装了吗？", + "answer_chinese": "你打开了。在你的手里。", + "evidence": [ + { + "description": "You turn it on.", + "timestamps": { + "start": "00:03:07", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:04:43", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You didn’t open it. It’s still sealed.", + "B": "You opened it. It’s in your hand." + }, + "answer_label": "B", + "answer_text": "You opened it. It’s in your hand.", + "q_id": "P28_108_3_T2", + "correct_answer": "You opened it. It’s in your hand.", + "original_options": { + "A": "You didn’t open it. It’s still sealed.", + "B": "You opened it. It’s in your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used a fork?", + "answer": "You used it twice.", + "question_chinese": "我用了多少次叉子", + "answer_chinese": "你用了两次", + "evidence": [ + { + "description": "You used it for the first time.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:20" + } + }, + { + "description": "You used it the second time.", + "timestamps": { + "start": "00:01:08", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You used it once.", + "B": "You used it three times.", + "C": "You used the fork one time during the meal, then set it down and didn’t pick it up again.", + "D": "You used the fork three separate times, picking it up and putting it down between each use.", + "E": "You used it twice." + }, + "answer_label": "E", + "answer_text": "You used it twice.", + "q_id": "P28_108_1", + "correct_answer": "You used it twice.", + "original_options": { + "A": "You used it once.", + "B": "You used it three times.", + "C": "You used the fork one time during the meal, then set it down and didn’t pick it up again.", + "D": "You used the fork three separate times, picking it up and putting it down between each use.", + "E": "You used it twice." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 20, + 17 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_3", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used a fork?", + "answer": "You used it three times.", + "question_chinese": "我用了多少次叉子", + "answer_chinese": "你用了三次", + "evidence": [ + { + "description": "You used it for the first time.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:20" + } + }, + { + "description": "You used it the second time.", + "timestamps": { + "start": "00:01:08", + "end": "00:01:35" + } + }, + { + "description": "You used it the third time.", + "timestamps": { + "start": "00:02:35", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:03:26", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You used it twice.", + "B": "You used it four times.", + "C": "You used the fork two times total, and the rest you ate without picking it up again.", + "D": "You used it three times.", + "E": "You used the fork four separate times, pausing and setting it down between uses." + }, + "answer_label": "D", + "answer_text": "You used it three times.", + "q_id": "P28_108_3", + "correct_answer": "You used it three times.", + "original_options": { + "A": "You used it twice.", + "B": "You used it four times.", + "C": "You used the fork two times total, and the rest you ate without picking it up again.", + "D": "You used it three times.", + "E": "You used the fork four separate times, pausing and setting it down between uses." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 14 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_7_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Could I reach the onion?", + "answer": "No, you can’t. The onion is at the far right end of the table in front of you.", + "question_chinese": "我能够到洋葱吗？", + "answer_chinese": "你不可以。洋葱在你前面桌子的右边尽头。", + "evidence": [ + { + "description": "You can't reach it because you're on the other side.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:01:48", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t. The onion is at the far right end of the table in front of you.", + "B": "Yes, you can. The onion is to your front-right." + }, + "answer_label": "A", + "answer_text": "No, you can’t. The onion is at the far right end of the table in front of you.", + "q_id": "P28_108_7_T1", + "correct_answer": "No, you can’t. The onion is at the far right end of the table in front of you.", + "original_options": { + "A": "No, you can’t. The onion is at the far right end of the table in front of you.", + "B": "Yes, you can. The onion is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_13", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Can I reach the onion?", + "answer": "You can.", + "question_chinese": "我能够到洋葱吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:50" + } + } + ], + "question_timestamps": "00:05:09", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "P28_108_13", + "correct_answer": "You can.", + "original_options": { + "A": "You cannot.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_7_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Could I reach the onion?", + "answer": "Yes, you can. The onion is to your front-right.", + "question_chinese": "我能够到洋葱吗？", + "answer_chinese": "你可以，洋葱在你的右前方。", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:50" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:05:19", + "end": "00:05:35" + } + } + ], + "question_timestamps": "00:06:53", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t. The onion is at the far right end of the table in front of you.", + "B": "Yes, you can. The onion is to your front-right." + }, + "answer_label": "B", + "answer_text": "Yes, you can. The onion is to your front-right.", + "q_id": "P28_108_7_T2", + "correct_answer": "Yes, you can. The onion is to your front-right.", + "original_options": { + "A": "No, you can’t. The onion is at the far right end of the table in front of you.", + "B": "Yes, you can. The onion is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 19 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_2_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Could I reach the seasonings?", + "answer": "Yes, you can. The seasoning is in front of you.", + "question_chinese": "我能够到调料吗？", + "answer_chinese": "你可以，调料在你的前面。", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:28" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t. The seasoning is on the cabinet to your left.", + "B": "Yes, you can. The seasoning is in front of you." + }, + "answer_label": "B", + "answer_text": "Yes, you can. The seasoning is in front of you.", + "q_id": "P28_108_2_T1", + "correct_answer": "Yes, you can. The seasoning is in front of you.", + "original_options": { + "A": "No, you can’t. The seasoning is on the cabinet to your left.", + "B": "Yes, you can. The seasoning is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "P28_108", + "video_length": "00:06:53", + "qid": "P28_108_2_T2", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Could I reach the seasonings?", + "answer": "No, you can’t. The seasoning is on the cabinet to your left.", + "question_chinese": "我能够到调料吗？", + "answer_chinese": "你不可以。调料在你左手边的柜子上。", + "evidence": [ + { + "description": "You can't reach the seasonings because you went to the other stove to get something.", + "timestamps": { + "start": "00:01:36", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:03:27", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t. The seasoning is on the cabinet to your left.", + "B": "Yes, you can. The seasoning is in front of you." + }, + "answer_label": "A", + "answer_text": "No, you can’t. The seasoning is on the cabinet to your left.", + "q_id": "P28_108_2_T2", + "correct_answer": "No, you can’t. The seasoning is on the cabinet to your left.", + "original_options": { + "A": "No, you can’t. The seasoning is on the cabinet to your left.", + "B": "Yes, you can. The seasoning is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + } +] diff --git a/QAs/P28_117-mcq.json b/QAs/P28_117-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..0e6483b1388b9bf8e14f04072649a5743949a5a2 --- /dev/null +++ b/QAs/P28_117-mcq.json @@ -0,0 +1,694 @@ +[ + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many seasonings did I just put into the stir-fried dish in the pot?", + "answer": "You put in two kinds of seasonings in total.", + "question_chinese": "我刚才往锅里的炒菜放了几种调料？", + "answer_chinese": "你一共放了两种调料。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:45" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:02:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You put in one seasoning.", + "B": "You put in two kinds of seasonings in total.", + "C": "You put in three seasonings.", + "D": "You added four kinds of seasonings altogether, mixing them into the stir-fry in the pot.", + "E": "You added five different seasonings in total before stirring the dish around in the pot." + }, + "answer_label": "B", + "answer_text": "You put in two kinds of seasonings in total.", + "q_id": "P28_117_3", + "correct_answer": "You put in two kinds of seasonings in total.", + "original_options": { + "A": "You put in one seasoning.", + "B": "You put in two kinds of seasonings in total.", + "C": "You put in three seasonings.", + "D": "You added four kinds of seasonings altogether, mixing them into the stir-fry in the pot.", + "E": "You added five different seasonings in total before stirring the dish around in the pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many seasonings did I just put into the stir-fried dish in the pot?", + "answer": "You put in two kinds of seasonings in total.", + "question_chinese": "我刚才往锅里的炒菜放了几种调料？", + "answer_chinese": "你一共放了两种调料。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:45" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You put in one seasoning.", + "B": "You put in three seasonings.", + "C": "You put in two kinds of seasonings in total.", + "D": "You added four kinds of seasonings altogether, mixing them into the stir-fry in the pot.", + "E": "You added five different seasonings in total before stirring the dish around in the pot." + }, + "answer_label": "C", + "answer_text": "You put in two kinds of seasonings in total.", + "q_id": "P28_117_4", + "correct_answer": "You put in two kinds of seasonings in total.", + "original_options": { + "A": "You put in one seasoning.", + "B": "You put in three seasonings.", + "C": "You put in two kinds of seasonings in total.", + "D": "You added four kinds of seasonings altogether, mixing them into the stir-fry in the pot.", + "E": "You added five different seasonings in total before stirring the dish around in the pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_7", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the olive oil from?", + "answer": "You took the olive oil from above the stove to the cabinet.", + "question_chinese": "我将橄榄油从哪里拿到了哪里？", + "answer_chinese": "你将橄榄油从灶台上面拿到了柜子里。", + "evidence": [ + { + "description": "The olive oil is on the stove.", + "timestamps": { + "start": "00:04:01", + "end": "00:04:02" + } + }, + { + "description": "The olive oil was taken into the cabinet.", + "timestamps": { + "start": "00:04:02", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:06:06", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From inside the pantry.", + "B": "From under the sink.", + "C": "You took the olive oil from the cabinet and moved it over to the stove area.", + "D": "You grabbed the olive oil from beside the refrigerator and carried it to the table.", + "E": "You took the olive oil from above the stove to the cabinet." + }, + "answer_label": "E", + "answer_text": "You took the olive oil from above the stove to the cabinet.", + "q_id": "P28_117_7", + "correct_answer": "You took the olive oil from above the stove to the cabinet.", + "original_options": { + "A": "From inside the pantry.", + "B": "From under the sink.", + "C": "You took the olive oil from the cabinet and moved it over to the stove area.", + "D": "You grabbed the olive oil from beside the refrigerator and carried it to the table.", + "E": "You took the olive oil from above the stove to the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 15 + ] + }, + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_8", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the olive oil from?", + "answer": "You took the olive oil from above the stove to the cabinet.", + "question_chinese": "我将橄榄油从哪里拿到了哪里？", + "answer_chinese": "你将橄榄油从灶台上面拿到了柜子里。", + "evidence": [ + { + "description": "The olive oil is on the stove.", + "timestamps": { + "start": "00:04:01", + "end": "00:04:02" + } + }, + { + "description": "The olive oil was taken into the cabinet.", + "timestamps": { + "start": "00:04:02", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:08:06", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From inside the fridge.", + "B": "From below the stove.", + "C": "You took the olive oil from above the stove to the cabinet.", + "D": "You took the olive oil from under the sink and brought it over to the table.", + "E": "You picked up the olive oil from beside the toaster and moved it to the shelf." + }, + "answer_label": "C", + "answer_text": "You took the olive oil from above the stove to the cabinet.", + "q_id": "P28_117_8", + "correct_answer": "You took the olive oil from above the stove to the cabinet.", + "original_options": { + "A": "From inside the fridge.", + "B": "From below the stove.", + "C": "You took the olive oil from above the stove to the cabinet.", + "D": "You took the olive oil from under the sink and brought it over to the table.", + "E": "You picked up the olive oil from beside the toaster and moved it to the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_11", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I want to eat some fruit. What fruits are there in the fridge?", + "answer": "There are oranges in the fridge.", + "question_chinese": "我想吃水果，冰箱里有什么水果？", + "answer_chinese": "冰箱里有橙子。", + "evidence": [ + { + "description": "There are oranges in the fridge.", + "timestamps": { + "start": "00:06:14", + "end": "00:06:15" + } + } + ], + "question_timestamps": "00:06:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There are apples in the fridge.", + "B": "There are grapes in the fridge.", + "C": "There are oranges in the fridge.", + "D": "There are pears in the fridge, sitting in the crisper drawer at the bottom.", + "E": "There are strawberries in the fridge, stored in a container on a shelf." + }, + "answer_label": "C", + "answer_text": "There are oranges in the fridge.", + "q_id": "P28_117_11", + "correct_answer": "There are oranges in the fridge.", + "original_options": { + "A": "There are apples in the fridge.", + "B": "There are grapes in the fridge.", + "C": "There are oranges in the fridge.", + "D": "There are pears in the fridge, sitting in the crisper drawer at the bottom.", + "E": "There are strawberries in the fridge, stored in a container on a shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_12", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I want to eat some fruit. What fruits are there in the fridge?", + "answer": "There are oranges in the fridge.", + "question_chinese": "我想吃水果，冰箱里有什么水果？", + "answer_chinese": "冰箱里有橙子。", + "evidence": [ + { + "description": "There are oranges in the fridge.", + "timestamps": { + "start": "00:06:14", + "end": "00:06:15" + } + } + ], + "question_timestamps": "00:07:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There are oranges in the fridge.", + "B": "There are lemons in the fridge.", + "C": "There are apples in the fridge.", + "D": "There are cherries in the fridge, kept in a small bowl on a shelf.", + "E": "There are grapes in the fridge, stored in a bag in the lower drawer." + }, + "answer_label": "A", + "answer_text": "There are oranges in the fridge.", + "q_id": "P28_117_12", + "correct_answer": "There are oranges in the fridge.", + "original_options": { + "A": "There are oranges in the fridge.", + "B": "There are lemons in the fridge.", + "C": "There are apples in the fridge.", + "D": "There are cherries in the fridge, kept in a small bowl on a shelf.", + "E": "There are grapes in the fridge, stored in a bag in the lower drawer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_15", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to throw away the garbage. Where is the trash can?", + "answer": "The trash can is outside the door.", + "question_chinese": "我想要扔垃圾，垃圾桶在哪里？", + "answer_chinese": "垃圾桶在门的外面。", + "evidence": [ + { + "description": "The trash can is outside the door.", + "timestamps": { + "start": "00:03:59", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:06:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The trash can is in the garage.", + "B": "The trash can is beside the back door.", + "C": "The trash can is near the driveway, set a short distance away from the entrance.", + "D": "The trash can is close to the garden gate, outside but not right by the door.", + "E": "The trash can is outside the door." + }, + "answer_label": "E", + "answer_text": "The trash can is outside the door.", + "q_id": "P28_117_15", + "correct_answer": "The trash can is outside the door.", + "original_options": { + "A": "The trash can is in the garage.", + "B": "The trash can is beside the back door.", + "C": "The trash can is near the driveway, set a short distance away from the entrance.", + "D": "The trash can is close to the garden gate, outside but not right by the door.", + "E": "The trash can is outside the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 16 + ] + }, + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_16", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to throw away the garbage. Where is the trash can?", + "answer": "The trash can is outside the door.", + "question_chinese": "我想要扔垃圾，垃圾桶在哪里？", + "answer_chinese": "垃圾桶在门的外面。", + "evidence": [ + { + "description": "The trash can is outside the door.", + "timestamps": { + "start": "00:03:59", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:08:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The trash can is in the garage.", + "B": "The trash can is in the laundry room.", + "C": "The trash can is outside the door.", + "D": "The trash can is outside near the driveway, not directly in front of the door.", + "E": "The trash can is beside the back door, positioned off to the side of the entrance." + }, + "answer_label": "C", + "answer_text": "The trash can is outside the door.", + "q_id": "P28_117_16", + "correct_answer": "The trash can is outside the door.", + "original_options": { + "A": "The trash can is in the garage.", + "B": "The trash can is in the laundry room.", + "C": "The trash can is outside the door.", + "D": "The trash can is outside near the driveway, not directly in front of the door.", + "E": "The trash can is beside the back door, positioned off to the side of the entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 16 + ] + }, + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_13", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sponges are placed beside the washbasin?", + "answer": "There are three sponges placed.", + "question_chinese": "洗手池旁边摆有几块海绵？", + "answer_chinese": "摆有三块海绵。", + "evidence": [ + { + "description": "There are three sponges placed.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:27" + } + } + ], + "question_timestamps": "00:07:11", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two sponges placed.", + "B": "There are four sponges placed.", + "C": "There are five sponges placed beside the washbasin, arranged together on the same side.", + "D": "There is only one sponge placed beside the washbasin, with no other sponges nearby.", + "E": "There are three sponges placed." + }, + "answer_label": "E", + "answer_text": "There are three sponges placed.", + "q_id": "P28_117_13", + "correct_answer": "There are three sponges placed.", + "original_options": { + "A": "There are two sponges placed.", + "B": "There are four sponges placed.", + "C": "There are five sponges placed beside the washbasin, arranged together on the same side.", + "D": "There is only one sponge placed beside the washbasin, with no other sponges nearby.", + "E": "There are three sponges placed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_14", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sponges are placed beside the washbasin?", + "answer": "There are three sponges placed.", + "question_chinese": "洗手池旁边摆有几块海绵？", + "answer_chinese": "摆有三块海绵。", + "evidence": [ + { + "description": "There are three sponges placed.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:27" + } + } + ], + "question_timestamps": "00:08:11", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one sponge placed.", + "B": "There are two sponges placed.", + "C": "There are four sponges placed beside the washbasin, grouped together next to the sink.", + "D": "There are three sponges placed.", + "E": "There are five sponges placed beside the washbasin, lined up along the edge." + }, + "answer_label": "D", + "answer_text": "There are three sponges placed.", + "q_id": "P28_117_14", + "correct_answer": "There are three sponges placed.", + "original_options": { + "A": "There is one sponge placed.", + "B": "There are two sponges placed.", + "C": "There are four sponges placed beside the washbasin, grouped together next to the sink.", + "D": "There are three sponges placed.", + "E": "There are five sponges placed beside the washbasin, lined up along the edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement trajectory of the pink plate?", + "answer": "I took the top of the pot from the table and finally put it back in its original position.", + "question_chinese": "粉色盘子的运动轨迹是什么？", + "answer_chinese": "从台子上拿到了锅的上方最后又放回了原位。", + "evidence": [ + { + "description": "The pink plate is on the table.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:21" + } + }, + { + "description": "The pink plate was taken above the pot.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:27" + } + }, + { + "description": "The pink plate was put back on the cabinet.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:05:46", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "I moved it to the counter.", + "B": "I placed it in the sink.", + "C": "I took the top of the pot from the table and finally put it back in its original position.", + "D": "I took the top of the pot from the table and left it on the counter instead of returning it.", + "E": "I picked up the top of the pot from the table and put it onto the pot, not back on the table." + }, + "answer_label": "C", + "answer_text": "I took the top of the pot from the table and finally put it back in its original position.", + "q_id": "P28_117_1", + "correct_answer": "I took the top of the pot from the table and finally put it back in its original position.", + "original_options": { + "A": "I moved it to the counter.", + "B": "I placed it in the sink.", + "C": "I took the top of the pot from the table and finally put it back in its original position.", + "D": "I took the top of the pot from the table and left it on the counter instead of returning it.", + "E": "I picked up the top of the pot from the table and put it onto the pot, not back on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 22 + ] + }, + { + "video_uid": "P28_117", + "video_length": "00:08:53", + "qid": "P28_117_2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement trajectory of the pink plate?", + "answer": "I took the top of the pot from the table and finally put it back in its original position.", + "question_chinese": "粉色盘子的运动轨迹是什么？", + "answer_chinese": "从台子上拿到了锅的上方最后又放回了原位。", + "evidence": [ + { + "description": "The pink plate is on the table.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:21" + } + }, + { + "description": "The pink plate was taken above the pot.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:27" + } + }, + { + "description": "The pink plate was put back on the cabinet.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:06:43", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "I took the top of the pot from the table and finally put it back in its original position.", + "B": "I moved it to the counter.", + "C": "I placed it in the sink.", + "D": "I took the top of the pot from the table and set it down on the stove rather than returning it.", + "E": "I picked up the top of the pot from the table and put it into the cabinet instead." + }, + "answer_label": "A", + "answer_text": "I took the top of the pot from the table and finally put it back in its original position.", + "q_id": "P28_117_2", + "correct_answer": "I took the top of the pot from the table and finally put it back in its original position.", + "original_options": { + "A": "I took the top of the pot from the table and finally put it back in its original position.", + "B": "I moved it to the counter.", + "C": "I placed it in the sink.", + "D": "I took the top of the pot from the table and set it down on the stove rather than returning it.", + "E": "I picked up the top of the pot from the table and put it into the cabinet instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 18 + ] + } +] diff --git a/QAs/P30_101-mcq.json b/QAs/P30_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2047b0a3d9e0e82c2c36bf295cf894d94dfb6435 --- /dev/null +++ b/QAs/P30_101-mcq.json @@ -0,0 +1,1311 @@ +[ + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many knives were there on the table at the beginning?", + "answer": "Two knives.", + "question_chinese": "刚开始时桌上有几把刀？", + "answer_chinese": "两把刀。", + "evidence": [ + { + "description": "Two knives.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two knives.", + "B": "One knife.", + "C": "Three knives.", + "D": "There were four knives on the table at the beginning, placed near each other.", + "E": "There were five knives on the table at the beginning, all visible before anything was moved." + }, + "answer_label": "A", + "answer_text": "Two knives.", + "q_id": "P30_101_1", + "correct_answer": "Two knives.", + "original_options": { + "A": "Two knives.", + "B": "One knife.", + "C": "Three knives.", + "D": "There were four knives on the table at the beginning, placed near each other.", + "E": "There were five knives on the table at the beginning, all visible before anything was moved." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many knives were there on the table at the beginning?", + "answer": "Two knives.", + "question_chinese": "刚开始时桌上有几把刀？", + "answer_chinese": "两把刀。", + "evidence": [ + { + "description": "Two knives.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One knife.", + "B": "Three knives.", + "C": "There were four knives on the table at the beginning, laid out together before you started.", + "D": "There were five knives on the table at the beginning, visible on the tabletop from the start.", + "E": "Two knives." + }, + "answer_label": "E", + "answer_text": "Two knives.", + "q_id": "P30_101_2", + "correct_answer": "Two knives.", + "original_options": { + "A": "One knife.", + "B": "Three knives.", + "C": "There were four knives on the table at the beginning, laid out together before you started.", + "D": "There were five knives on the table at the beginning, visible on the tabletop from the start.", + "E": "Two knives." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I take out of the cabinet?", + "answer": "A piece of filter paper.", + "question_chinese": "我从柜子中拿出了什么", + "answer_chinese": "一张滤纸。", + "evidence": [ + { + "description": "A piece of filter paper.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A glass beaker.", + "B": "A roll of aluminum foil.", + "C": "You took a box of tissues out of the cabinet and held it in your hands.", + "D": "A piece of filter paper.", + "E": "You pulled a stack of notepads from the cabinet and set them down nearby." + }, + "answer_label": "D", + "answer_text": "A piece of filter paper.", + "q_id": "P30_101_3", + "correct_answer": "A piece of filter paper.", + "original_options": { + "A": "A glass beaker.", + "B": "A roll of aluminum foil.", + "C": "You took a box of tissues out of the cabinet and held it in your hands.", + "D": "A piece of filter paper.", + "E": "You pulled a stack of notepads from the cabinet and set them down nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 5, + 16, + 14 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I take out of the cabinet?", + "answer": "A piece of filter paper.", + "question_chinese": "我从柜子中拿出了什么", + "answer_chinese": "一张滤纸。", + "evidence": [ + { + "description": "A piece of filter paper.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A bottle of vinegar.", + "B": "A piece of filter paper.", + "C": "A roll of plastic wrap.", + "D": "You took a box of matches out of the cabinet and brought it out with you.", + "E": "You brought out a stack of notebooks from the cabinet and placed them on a surface." + }, + "answer_label": "B", + "answer_text": "A piece of filter paper.", + "q_id": "P30_101_4", + "correct_answer": "A piece of filter paper.", + "original_options": { + "A": "A bottle of vinegar.", + "B": "A piece of filter paper.", + "C": "A roll of plastic wrap.", + "D": "You took a box of matches out of the cabinet and brought it out with you.", + "E": "You brought out a stack of notebooks from the cabinet and placed them on a surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 16 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_2_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I take out of the cabinet in front of me?", + "answer": "You took a box out of the cabinet in front of you.", + "question_chinese": "我从面前的柜子中拿出了什么？", + "answer_chinese": "你在面前的柜子里拿出了一个盒子。", + "evidence": [ + { + "description": "A box", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took a jar out.", + "B": "You took a bag out.", + "C": "You took a container out of the cabinet in front of you instead of a box.", + "D": "You took a carton out of the cabinet in front of you, not the box mentioned.", + "E": "You took a box out of the cabinet in front of you." + }, + "answer_label": "E", + "answer_text": "You took a box out of the cabinet in front of you.", + "q_id": "P30_101_2_T1", + "correct_answer": "You took a box out of the cabinet in front of you.", + "original_options": { + "A": "You took a jar out.", + "B": "You took a bag out.", + "C": "You took a container out of the cabinet in front of you instead of a box.", + "D": "You took a carton out of the cabinet in front of you, not the box mentioned.", + "E": "You took a box out of the cabinet in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_2_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I take out of the cabinet in front of me?", + "answer": "You took a filter paper out of the cabinet in front of you.", + "question_chinese": "我从面前的柜子中拿出了什么？", + "answer_chinese": "你从面前的柜子里拿出了一张滤纸。", + "evidence": [ + { + "description": "A piece of filter paper.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took a napkin out.", + "B": "You took a paper towel out.", + "C": "You took a filter paper out of the cabinet in front of you.", + "D": "You took a sheet of tissue out of the cabinet in front of you rather than filter paper.", + "E": "You took a wax paper sheet out of the cabinet in front of you, not filter paper." + }, + "answer_label": "C", + "answer_text": "You took a filter paper out of the cabinet in front of you.", + "q_id": "P30_101_2_T2", + "correct_answer": "You took a filter paper out of the cabinet in front of you.", + "original_options": { + "A": "You took a napkin out.", + "B": "You took a paper towel out.", + "C": "You took a filter paper out of the cabinet in front of you.", + "D": "You took a sheet of tissue out of the cabinet in front of you rather than filter paper.", + "E": "You took a wax paper sheet out of the cabinet in front of you, not filter paper." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 17 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_7_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of pastry did I eat from the plate in front of me?", + "answer": "You ate two pieces from the plate in front of you.", + "question_chinese": "我吃了面前碟子里的几块饼？", + "answer_chinese": "你吃了面前碟子里的两块。", + "evidence": [ + { + "description": "The first piece of cake.", + "timestamps": { + "start": "00:06:11", + "end": "00:06:59" + } + }, + { + "description": "The second piece of cake.", + "timestamps": { + "start": "00:07:12", + "end": "00:07:35" + } + } + ], + "question_timestamps": "00:07:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You ate one piece.", + "B": "You ate three pieces.", + "C": "You ate four pieces of pastry from the plate in front of you during that moment.", + "D": "You ate zero pieces from the plate in front of you and left all the pastry untouched.", + "E": "You ate two pieces from the plate in front of you." + }, + "answer_label": "E", + "answer_text": "You ate two pieces from the plate in front of you.", + "q_id": "P30_101_7_T2", + "correct_answer": "You ate two pieces from the plate in front of you.", + "original_options": { + "A": "You ate one piece.", + "B": "You ate three pieces.", + "C": "You ate four pieces of pastry from the plate in front of you during that moment.", + "D": "You ate zero pieces from the plate in front of you and left all the pastry untouched.", + "E": "You ate two pieces from the plate in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_7_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of pastry did I eat from the plate in front of me?", + "answer": "You ate three pieces from the plate in front of you.", + "question_chinese": "我吃了面前碟子里的几块饼？", + "answer_chinese": "你吃了面前碟子里的三块。", + "evidence": [ + { + "description": "The first piece of cake.", + "timestamps": { + "start": "00:06:11", + "end": "00:06:59" + } + }, + { + "description": "The second piece of cake.", + "timestamps": { + "start": "00:07:12", + "end": "00:07:35" + } + }, + { + "description": "The third piece of cake.", + "timestamps": { + "start": "00:08:03", + "end": "00:08:45" + } + } + ], + "question_timestamps": "00:10:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You ate two pieces.", + "B": "You ate four pieces.", + "C": "You ate one piece of pastry from the plate in front of you and stopped after the first.", + "D": "You ate three pieces from the plate in front of you.", + "E": "You ate five pieces from the plate in front of you, finishing more than half of what was there." + }, + "answer_label": "D", + "answer_text": "You ate three pieces from the plate in front of you.", + "q_id": "P30_101_7_T1", + "correct_answer": "You ate three pieces from the plate in front of you.", + "original_options": { + "A": "You ate two pieces.", + "B": "You ate four pieces.", + "C": "You ate one piece of pastry from the plate in front of you and stopped after the first.", + "D": "You ate three pieces from the plate in front of you.", + "E": "You ate five pieces from the plate in front of you, finishing more than half of what was there." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_6_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I’m standing, can I wash my hands at the sink in front of me?", + "answer": "No, the sink is behind you.", + "question_chinese": "我在当前位置能否在面前的洗碗池洗到手？", + "answer_chinese": "不可以，洗手池在你的后面。", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:08:44", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, the sink is behind you.", + "B": "Yes, the sink is in front of you." + }, + "answer_label": "A", + "answer_text": "No, the sink is behind you.", + "q_id": "P30_101_6_T2", + "correct_answer": "No, the sink is behind you.", + "original_options": { + "A": "No, the sink is behind you.", + "B": "Yes, the sink is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_6_T1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I’m standing, can I wash my hands at the sink in front of me?", + "answer": "Yes, the sink is to your left.", + "question_chinese": "我在当前位置能否在面前的洗碗池洗到手？", + "answer_chinese": "可以，洗手池在你的左边。", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:10:25", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, the sink is behind you.", + "B": "Yes, the sink is to your left." + }, + "answer_label": "B", + "answer_text": "Yes, the sink is to your left.", + "q_id": "P30_101_6_T1", + "correct_answer": "Yes, the sink is to your left.", + "original_options": { + "A": "No, the sink is behind you.", + "B": "Yes, the sink is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_4_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sets of knives and forks did I wash in the sink in front of me?", + "answer": "You washed two knives.", + "question_chinese": "我在面前的洗碗池洗了几把刀叉？", + "answer_chinese": "你洗了两把刀", + "evidence": [ + { + "description": "Wash the first knife.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:52" + } + }, + { + "description": "Wash the second knife.", + "timestamps": { + "start": "00:01:53", + "end": "00:02:02" + } + } + ], + "question_timestamps": "00:02:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed two knives.", + "B": "You washed one knife.", + "C": "You washed three knives.", + "D": "You washed four knives in the sink in front of you instead of just two.", + "E": "You washed zero knives in the sink in front of you and didn’t clean any cutlery." + }, + "answer_label": "A", + "answer_text": "You washed two knives.", + "q_id": "P30_101_4_T1", + "correct_answer": "You washed two knives.", + "original_options": { + "A": "You washed two knives.", + "B": "You washed one knife.", + "C": "You washed three knives.", + "D": "You washed four knives in the sink in front of you instead of just two.", + "E": "You washed zero knives in the sink in front of you and didn’t clean any cutlery." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_4_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sets of knives and forks did I wash in the sink in front of me?", + "answer": "You washed two knives and one fork in the sink in front of you.", + "question_chinese": "我在面前的洗碗池洗了几把刀叉？", + "answer_chinese": "你在面前的洗碗池洗了两把刀，一把叉子", + "evidence": [ + { + "description": "Wash the first knife.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:52" + } + }, + { + "description": "Wash the second knife.", + "timestamps": { + "start": "00:01:53", + "end": "00:02:02" + } + }, + { + "description": "Wash the fork.", + "timestamps": { + "start": "00:02:32", + "end": "00:02:44" + } + } + ], + "question_timestamps": "00:02:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed one knife and one fork.", + "B": "You washed two knives and one fork in the sink in front of you.", + "C": "You washed three knives and one fork.", + "D": "You washed four knives and one fork in the sink in front of you, not two knives and one fork.", + "E": "You washed zero knives and one fork in the sink in front of you, leaving the knives unwashed." + }, + "answer_label": "B", + "answer_text": "You washed two knives and one fork in the sink in front of you.", + "q_id": "P30_101_4_T2", + "correct_answer": "You washed two knives and one fork in the sink in front of you.", + "original_options": { + "A": "You washed one knife and one fork.", + "B": "You washed two knives and one fork in the sink in front of you.", + "C": "You washed three knives and one fork.", + "D": "You washed four knives and one fork in the sink in front of you, not two knives and one fork.", + "E": "You washed zero knives and one fork in the sink in front of you, leaving the knives unwashed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 18 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_3_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates did I wash in the sink in front of me?", + "answer": "You washed one plate in the sink in front of you.", + "question_chinese": "我在面前的洗碗池洗了多少个盘子？", + "answer_chinese": "你在面前的洗碗池洗了一个盘子。", + "evidence": [ + { + "description": "The first plate.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:01:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed one plate in the sink in front of you.", + "B": "None.", + "C": "Two plates.", + "D": "You washed two plates in the sink in front of you after rinsing them briefly.", + "E": "You washed three plates in the sink in front of you before stopping to do something else." + }, + "answer_label": "A", + "answer_text": "You washed one plate in the sink in front of you.", + "q_id": "P30_101_3_T2", + "correct_answer": "You washed one plate in the sink in front of you.", + "original_options": { + "A": "You washed one plate in the sink in front of you.", + "B": "None.", + "C": "Two plates.", + "D": "You washed two plates in the sink in front of you after rinsing them briefly.", + "E": "You washed three plates in the sink in front of you before stopping to do something else." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 17 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_3_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates did I wash in the sink in front of me?", + "answer": "You washed two plates in the sink in front of you.", + "question_chinese": "我在面前的洗碗池洗了多少个盘子？", + "answer_chinese": "你在面前的洗碗池洗了两个盘子。", + "evidence": [ + { + "description": "The first plate.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:17" + } + }, + { + "description": "The second plate.", + "timestamps": { + "start": "00:01:43", + "end": "00:01:52" + } + } + ], + "question_timestamps": "00:01:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One plate.", + "B": "Three plates.", + "C": "You washed one plate in the sink in front of you, then stopped without washing another.", + "D": "You washed four plates in the sink in front of you in total, including a quick rinse at the end.", + "E": "You washed two plates in the sink in front of you." + }, + "answer_label": "E", + "answer_text": "You washed two plates in the sink in front of you.", + "q_id": "P30_101_3_T1", + "correct_answer": "You washed two plates in the sink in front of you.", + "original_options": { + "A": "One plate.", + "B": "Three plates.", + "C": "You washed one plate in the sink in front of you, then stopped without washing another.", + "D": "You washed four plates in the sink in front of you in total, including a quick rinse at the end.", + "E": "You washed two plates in the sink in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 20 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_7", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I wash the knives and forks?", + "answer": "You have washed it.", + "question_chinese": "我是否洗过刀叉了", + "answer_chinese": "你洗过了", + "evidence": [ + { + "description": "Wash the first knife.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:52" + } + }, + { + "description": "Wash the second knife.", + "timestamps": { + "start": "00:01:53", + "end": "00:02:02" + } + }, + { + "description": "Wash the fork.", + "timestamps": { + "start": "00:02:32", + "end": "00:02:44" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "You have washed it.", + "B": "No, you didn’t wash the knives and forks." + }, + "answer_label": "A", + "answer_text": "You have washed it.", + "q_id": "P30_101_7", + "correct_answer": "You have washed it.", + "original_options": { + "A": "You have washed it.", + "B": "No, you didn’t wash the knives and forks." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_8", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I wash the knives and forks?", + "answer": "You have washed it.", + "question_chinese": "我是否洗过刀叉了", + "answer_chinese": "你洗过了", + "evidence": [ + { + "description": "Wash the first knife.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:52" + } + }, + { + "description": "Wash the second knife.", + "timestamps": { + "start": "00:01:53", + "end": "00:02:02" + } + }, + { + "description": "Wash the fork.", + "timestamps": { + "start": "00:02:32", + "end": "00:02:44" + } + } + ], + "question_timestamps": "00:07:32", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "No, you didn’t wash the knives and forks.", + "B": "You have washed it." + }, + "answer_label": "B", + "answer_text": "You have washed it.", + "q_id": "P30_101_8", + "correct_answer": "You have washed it.", + "original_options": { + "A": "No, you didn’t wash the knives and forks.", + "B": "You have washed it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_9", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I turn on the water heater?", + "answer": "You opened it.", + "question_chinese": "我是否点开热水器了", + "answer_chinese": "你打开了。", + "evidence": [ + { + "description": "You turned on the water heater.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You opened it.", + "B": "No, you didn’t turn on the water heater." + }, + "answer_label": "A", + "answer_text": "You opened it.", + "q_id": "P30_101_9", + "correct_answer": "You opened it.", + "original_options": { + "A": "You opened it.", + "B": "No, you didn’t turn on the water heater." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_10", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I turn on the water heater?", + "answer": "You opened it.", + "question_chinese": "我是否点开热水器了", + "answer_chinese": "你打开了", + "evidence": [ + { + "description": "You turned on the water heater.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:10:26", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You opened it.", + "B": "No, you didn’t turn on the water heater." + }, + "answer_label": "A", + "answer_text": "You opened it.", + "q_id": "P30_101_10", + "correct_answer": "You opened it.", + "original_options": { + "A": "You opened it.", + "B": "No, you didn’t turn on the water heater." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_5", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates did I wash?", + "answer": "Two.", + "question_chinese": "我洗了多少个盘子", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "The first plate.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:17" + } + }, + { + "description": "The second plate.", + "timestamps": { + "start": "00:01:43", + "end": "00:01:52" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You washed four plates in total, including one that you rinsed only briefly in the sink.", + "E": "You washed five plates in total, with several stacked together before you started scrubbing them." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "P30_101_5", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You washed four plates in total, including one that you rinsed only briefly in the sink.", + "E": "You washed five plates in total, with several stacked together before you started scrubbing them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_6", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates did I wash?", + "answer": "Two.", + "question_chinese": "我洗了多少个盘子", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "The first plate.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:17" + } + }, + { + "description": "The second plate.", + "timestamps": { + "start": "00:01:43", + "end": "00:01:52" + } + } + ], + "question_timestamps": "00:05:08", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You washed four plates in total, including one that you only rinsed quickly before putting it aside.", + "D": "You washed seven plates in total, scrubbing each one before stacking them neatly to dry afterward.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "P30_101_6", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You washed four plates in total, including one that you only rinsed quickly before putting it aside.", + "D": "You washed seven plates in total, scrubbing each one before stacking them neatly to dry afterward.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 16 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_1_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the knife on the cutting board relative to me?", + "answer": "Directly in front of you.", + "question_chinese": "案板上的刀相对于我是什么位置？", + "answer_chinese": "你的正前方", + "evidence": [ + { + "description": "On the counter in front of me, there is a cutting board to the right of the sink with two knives on it.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly behind you.", + "B": "Directly to your left.", + "C": "Directly to your right, not in front of you on the cutting board.", + "D": "Off to your side, rather than directly in front of you at the cutting board.", + "E": "Directly in front of you." + }, + "answer_label": "E", + "answer_text": "Directly in front of you.", + "q_id": "P30_101_1_T1", + "correct_answer": "Directly in front of you.", + "original_options": { + "A": "Directly behind you.", + "B": "Directly to your left.", + "C": "Directly to your right, not in front of you on the cutting board.", + "D": "Off to your side, rather than directly in front of you at the cutting board.", + "E": "Directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 15 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_1_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the knife on the cutting board relative to me?", + "answer": "To your right.", + "question_chinese": "案板上的刀相对于我是什么位置？", + "answer_chinese": "在你的右边", + "evidence": [ + { + "description": "On the counter in front of me, there is a cutting board to the right of the sink with two knives on it.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your right.", + "D": "To your left, not on your right while standing at the cutting board.", + "E": "Above you instead of being positioned to your right on the cutting board." + }, + "answer_label": "C", + "answer_text": "To your right.", + "q_id": "P30_101_1_T2", + "correct_answer": "To your right.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your right.", + "D": "To your left, not on your right while standing at the cutting board.", + "E": "Above you instead of being positioned to your right on the cutting board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 13, + 13 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_5_T1", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the electric kettle relative to me?", + "answer": "On your left-hand side.", + "question_chinese": "烧水壶相对于我是什么方位？", + "answer_chinese": "在你的左手边", + "evidence": [ + { + "description": "There is a kettle on the left of the sink, and you press its switch.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your right-hand side.", + "B": "Directly in front of you.", + "C": "Behind you rather than on your left-hand side, near where you are standing.", + "D": "On your left-hand side.", + "E": "Slightly to your front-right, not positioned on your left-hand side at all." + }, + "answer_label": "D", + "answer_text": "On your left-hand side.", + "q_id": "P30_101_5_T1", + "correct_answer": "On your left-hand side.", + "original_options": { + "A": "On your right-hand side.", + "B": "Directly in front of you.", + "C": "Behind you rather than on your left-hand side, near where you are standing.", + "D": "On your left-hand side.", + "E": "Slightly to your front-right, not positioned on your left-hand side at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 12 + ] + }, + { + "video_uid": "P30_101", + "video_length": "00:10:28", + "qid": "P30_101_5_T2", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the electric kettle relative to me?", + "answer": "To your back-left, next to the sink.", + "question_chinese": "烧水壶相对于我是什么方位？", + "answer_chinese": "在你的左后方，洗碗池的旁边", + "evidence": [ + { + "description": "There is a kettle on the left of the sink, and you press its switch.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:08:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-right, next to the sink.", + "B": "To your front-left, next to the sink.", + "C": "Directly behind you beside the sink, not back-left from where you are standing.", + "D": "On your left near the sink, but not specifically behind you on that side.", + "E": "To your back-left, next to the sink." + }, + "answer_label": "E", + "answer_text": "To your back-left, next to the sink.", + "q_id": "P30_101_5_T2", + "correct_answer": "To your back-left, next to the sink.", + "original_options": { + "A": "To your back-right, next to the sink.", + "B": "To your front-left, next to the sink.", + "C": "Directly behind you beside the sink, not back-left from where you are standing.", + "D": "On your left near the sink, but not specifically behind you on that side.", + "E": "To your back-left, next to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 14 + ] + } +] diff --git a/QAs/P30_103-mcq.json b/QAs/P30_103-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2dc0945940f70f46632c22464e2f659a855b3a33 --- /dev/null +++ b/QAs/P30_103-mcq.json @@ -0,0 +1,856 @@ +[ + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_13", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I add water to the kettle?", + "answer": "Twice.", + "question_chinese": "我往热水壶中加了几次水？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "The first time.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:37" + } + }, + { + "description": "The second time.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "You added water only once while filling the kettle at the sink before turning it on.", + "E": "You added water three separate times to the kettle during the task before setting it back down." + }, + "answer_label": "A", + "answer_text": "Twice.", + "q_id": "P30_103_13", + "correct_answer": "Twice.", + "original_options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "You added water only once while filling the kettle at the sink before turning it on.", + "E": "You added water three separate times to the kettle during the task before setting it back down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 16, + 17 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_14", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I add water to the kettle?", + "answer": "Twice.", + "question_chinese": "我往热水壶中加了几次水？", + "answer_chinese": "两次。", + "evidence": [ + { + "description": "The first time.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:37" + } + }, + { + "description": "The second time.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:01:31", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "You added water four separate times to the kettle during the task.", + "D": "You refilled the kettle six times over the course of the task.", + "E": "Twice." + }, + "answer_label": "E", + "answer_text": "Twice.", + "q_id": "P30_103_14", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "You added water four separate times to the kettle during the task.", + "D": "You refilled the kettle six times over the course of the task.", + "E": "Twice." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 12 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the room I'm in?", + "answer": "Two.", + "question_chinese": "我所在房间里有多少个垃圾桶", + "answer_chinese": "两个", + "evidence": [ + { + "description": "First.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:13" + } + }, + { + "description": "Second.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:21" + } + } + ], + "question_timestamps": "00:01:39", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "There is one trash can in the room you are in right now.", + "D": "Two.", + "E": "There are three trash cans visible in the room you are in." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "P30_103_7", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "There is one trash can in the room you are in right now.", + "D": "Two.", + "E": "There are three trash cans visible in the room you are in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_8", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the room I'm in?", + "answer": "Two.", + "question_chinese": "我所在房间里有多少个垃圾桶", + "answer_chinese": "两个", + "evidence": [ + { + "description": "First.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:13" + } + }, + { + "description": "Second.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:21" + } + } + ], + "question_timestamps": "00:02:11", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "There are four trash cans in the room you are in right now.", + "E": "There are five trash cans visible in the room you are in." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "P30_103_8", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "There are four trash cans in the room you are in right now.", + "E": "There are five trash cans visible in the room you are in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_9", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I eaten vegetable salad?", + "answer": "Have never eaten vegetable salad.", + "question_chinese": "我是否吃过蔬菜沙拉", + "answer_chinese": "没有吃过蔬菜沙拉", + "evidence": [ + { + "description": "Have never eaten vegetable salad.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Have never eaten vegetable salad.", + "B": "You have eaten vegetable salad." + }, + "answer_label": "A", + "answer_text": "Have never eaten vegetable salad.", + "q_id": "P30_103_9", + "correct_answer": "Have never eaten vegetable salad.", + "original_options": { + "A": "Have never eaten vegetable salad.", + "B": "You have eaten vegetable salad." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_10", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I eaten vegetable salad?", + "answer": "You have eaten vegetable salad.", + "question_chinese": "我是否吃过蔬菜沙拉", + "answer_chinese": "你吃过蔬菜沙拉。", + "evidence": [ + { + "description": "You have eaten vegetable salad.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:03:04", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You have eaten vegetable salad.", + "B": "You have not eaten vegetable salad." + }, + "answer_label": "A", + "answer_text": "You have eaten vegetable salad.", + "q_id": "P30_103_10", + "correct_answer": "You have eaten vegetable salad.", + "original_options": { + "A": "You have eaten vegetable salad.", + "B": "You have not eaten vegetable salad." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_3_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I boil water?", + "answer": "Zero times—you didn’t boil any water. The kettle is to your front-left.", + "question_chinese": "我烧了多少次水？", + "answer_chinese": "零次，你没有烧水，烧水壶在你左前方。", + "evidence": [ + { + "description": "You haven't boiled the water.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One time.", + "B": "Two times.", + "C": "Zero times—you didn’t boil any water. The kettle is to your front-left.", + "D": "One time—you boiled water once. The kettle is to your front-left.", + "E": "Three times—you boiled water three times. The kettle is to your front-left." + }, + "answer_label": "C", + "answer_text": "Zero times—you didn’t boil any water. The kettle is to your front-left.", + "q_id": "P30_103_3_T2", + "correct_answer": "Zero times—you didn’t boil any water. The kettle is to your front-left.", + "original_options": { + "A": "One time.", + "B": "Two times.", + "C": "Zero times—you didn’t boil any water. The kettle is to your front-left.", + "D": "One time—you boiled water once. The kettle is to your front-left.", + "E": "Three times—you boiled water three times. The kettle is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_3_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I boil water?", + "answer": "You boiled water once, and the kettle is to your left.", + "question_chinese": "我烧了多少次水？", + "answer_chinese": "你烧了一次热水，烧水壶在你的左边。", + "evidence": [ + { + "description": "First.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero times.", + "B": "You boiled water once, and the kettle is to your left.", + "C": "Two times.", + "D": "Three times—you boiled water three times, and the kettle is to your left.", + "E": "Four times—you boiled water four times, and the kettle is to your left." + }, + "answer_label": "B", + "answer_text": "You boiled water once, and the kettle is to your left.", + "q_id": "P30_103_3_T1", + "correct_answer": "You boiled water once, and the kettle is to your left.", + "original_options": { + "A": "Zero times.", + "B": "You boiled water once, and the kettle is to your left.", + "C": "Two times.", + "D": "Three times—you boiled water three times, and the kettle is to your left.", + "E": "Four times—you boiled water four times, and the kettle is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_3", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands now?", + "answer": "Can't wash hands.", + "question_chinese": "我现在是否可以洗到手", + "answer_chinese": "洗不到手", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:22" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Can't wash hands.", + "B": "Can wash hands." + }, + "answer_label": "A", + "answer_text": "Can't wash hands.", + "q_id": "P30_103_3", + "correct_answer": "Can't wash hands.", + "original_options": { + "A": "Can't wash hands.", + "B": "Can wash hands." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_4", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands now?", + "answer": "You can't wash your hands right now.", + "question_chinese": "我现在是否可以洗到手", + "answer_chinese": "你目前洗不到手", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:22" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:01:37", + "end": "00:01:37" + } + } + ], + "question_timestamps": "00:01:37", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't wash your hands right now.", + "B": "You can wash your hands right now." + }, + "answer_label": "A", + "answer_text": "You can't wash your hands right now.", + "q_id": "P30_103_4", + "correct_answer": "You can't wash your hands right now.", + "original_options": { + "A": "You can't wash your hands right now.", + "B": "You can wash your hands right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_1_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where did I put my fork?", + "answer": "Your fork is on the table, and the table is in front of you.", + "question_chinese": "我的叉子放在哪里？", + "answer_chinese": "你的叉子放在桌子上，桌子在你前面。", + "evidence": [ + { + "description": "Your fork is on the table.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your fork is on the table, and the table is in front of you.", + "B": "On the table behind you.", + "C": "On the table to your left.", + "D": "Your fork is on the table, and the table is behind you.", + "E": "Your fork is on the table, and the table is to your right." + }, + "answer_label": "A", + "answer_text": "Your fork is on the table, and the table is in front of you.", + "q_id": "P30_103_1_T1", + "correct_answer": "Your fork is on the table, and the table is in front of you.", + "original_options": { + "A": "Your fork is on the table, and the table is in front of you.", + "B": "On the table behind you.", + "C": "On the table to your left.", + "D": "Your fork is on the table, and the table is behind you.", + "E": "Your fork is on the table, and the table is to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 13 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_1_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where did I put my fork?", + "answer": "Your fork is in the bowl, and the bowl is to your left.", + "question_chinese": "我的叉子放在哪里？", + "answer_chinese": "你的叉子放在碗里，碗在你的左边。", + "evidence": [ + { + "description": "Your fork is in the bowl.", + "timestamps": { + "start": "00:02:17", + "end": "00:02:19" + } + } + ], + "question_timestamps": "00:02:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your fork is in the bowl, and the bowl is to your left.", + "B": "In the bowl in front of you.", + "C": "In the bowl behind you.", + "D": "Your fork is in the bowl, and the bowl is to your right.", + "E": "Your fork is in the bowl, and the bowl is in front of you." + }, + "answer_label": "A", + "answer_text": "Your fork is in the bowl, and the bowl is to your left.", + "q_id": "P30_103_1_T2", + "correct_answer": "Your fork is in the bowl, and the bowl is to your left.", + "original_options": { + "A": "Your fork is in the bowl, and the bowl is to your left.", + "B": "In the bowl in front of you.", + "C": "In the bowl behind you.", + "D": "Your fork is in the bowl, and the bowl is to your right.", + "E": "Your fork is in the bowl, and the bowl is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 13, + 14 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_5_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which side of me is the kettle on?", + "answer": "The kettle is to your left.", + "question_chinese": "烧水壶在我的哪边？", + "answer_chinese": "烧水壶在你的左边。", + "evidence": [ + { + "description": "The kettle is on your left.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The kettle is to your right.", + "B": "The kettle is in front of you.", + "C": "The kettle is to your left.", + "D": "The kettle is behind you, not beside you, so you’d need to turn around to see it.", + "E": "The kettle is above you, placed higher up rather than sitting to either side." + }, + "answer_label": "C", + "answer_text": "The kettle is to your left.", + "q_id": "P30_103_5_T2", + "correct_answer": "The kettle is to your left.", + "original_options": { + "A": "The kettle is to your right.", + "B": "The kettle is in front of you.", + "C": "The kettle is to your left.", + "D": "The kettle is behind you, not beside you, so you’d need to turn around to see it.", + "E": "The kettle is above you, placed higher up rather than sitting to either side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 14 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_5_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which side of me is the kettle on?", + "answer": "The kettle is in front of you.", + "question_chinese": "烧水壶在我的哪边？", + "answer_chinese": "烧水壶在你前面。", + "evidence": [ + { + "description": "The kettle is in front of you.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The kettle is behind you.", + "B": "The kettle is to your left.", + "C": "The kettle is to your right, not centered ahead of you when you look forward.", + "D": "The kettle is in front of you.", + "E": "The kettle is above you, positioned higher rather than directly in your line of sight." + }, + "answer_label": "D", + "answer_text": "The kettle is in front of you.", + "q_id": "P30_103_5_T1", + "correct_answer": "The kettle is in front of you.", + "original_options": { + "A": "The kettle is behind you.", + "B": "The kettle is to your left.", + "C": "The kettle is to your right, not centered ahead of you when you look forward.", + "D": "The kettle is in front of you.", + "E": "The kettle is above you, positioned higher rather than directly in your line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_6_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Relative to me, where is the glass cup?", + "answer": "The glass is behind you.", + "question_chinese": "相对于我的位置，玻璃杯在哪里？", + "answer_chinese": "玻璃杯在你的背后。", + "evidence": [ + { + "description": "The glass is behind you.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The glass is in front of you.", + "B": "The glass is to your left.", + "C": "The glass is behind you.", + "D": "The glass is to your right, not behind you, so you don’t need to turn around.", + "E": "The glass is above you, placed up higher rather than sitting behind your position." + }, + "answer_label": "C", + "answer_text": "The glass is behind you.", + "q_id": "P30_103_6_T2", + "correct_answer": "The glass is behind you.", + "original_options": { + "A": "The glass is in front of you.", + "B": "The glass is to your left.", + "C": "The glass is behind you.", + "D": "The glass is to your right, not behind you, so you don’t need to turn around.", + "E": "The glass is above you, placed up higher rather than sitting behind your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 14 + ] + }, + { + "video_uid": "P30_103", + "video_length": "00:03:04", + "qid": "P30_103_6_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Relative to me, where is the glass cup?", + "answer": "The glass is on your right-hand side.", + "question_chinese": "相对于我的位置，玻璃杯在哪里？", + "answer_chinese": "玻璃杯在你的右手边。", + "evidence": [ + { + "description": "The glass is on your right hand side.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The glass is in front of you.", + "B": "The glass is on your right-hand side.", + "C": "The glass is on your left-hand side.", + "D": "The glass is behind you, so it isn’t on your right-hand side when you’re facing forward.", + "E": "The glass is above you, positioned higher up rather than to your right-hand side." + }, + "answer_label": "B", + "answer_text": "The glass is on your right-hand side.", + "q_id": "P30_103_6_T1", + "correct_answer": "The glass is on your right-hand side.", + "original_options": { + "A": "The glass is in front of you.", + "B": "The glass is on your right-hand side.", + "C": "The glass is on your left-hand side.", + "D": "The glass is behind you, so it isn’t on your right-hand side when you’re facing forward.", + "E": "The glass is above you, positioned higher up rather than to your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 14 + ] + } +] diff --git a/QAs/P30_108-mcq.json b/QAs/P30_108-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9aa88aae7f9d95e39675a994ebf910858d20a726 --- /dev/null +++ b/QAs/P30_108-mcq.json @@ -0,0 +1,1591 @@ +[ + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_16", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the spice arrangement area in the kitchen?", + "answer": "In the cabinet above the sink?", + "question_chinese": "Where is the seasoning display area in the kitchen?", + "answer_chinese": "In the cabinet above the washbasin?", + "evidence": [ + { + "description": "In the cabinet above the washbasin?", + "timestamps": { + "start": "00:08:24", + "end": "00:08:31" + } + } + ], + "question_timestamps": "00:08:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the cabinet by the refrigerator.", + "B": "In the drawer under the counter.", + "C": "On the shelf next to the cooktop, rather than inside the cabinet above the sink.", + "D": "Above the countertop near the window, not in the upper cabinet over the sink area.", + "E": "In the cabinet above the sink?" + }, + "answer_label": "E", + "answer_text": "In the cabinet above the sink?", + "q_id": "P30_108_16", + "correct_answer": "In the cabinet above the sink?", + "original_options": { + "A": "In the cabinet by the refrigerator.", + "B": "In the drawer under the counter.", + "C": "On the shelf next to the cooktop, rather than inside the cabinet above the sink.", + "D": "Above the countertop near the window, not in the upper cabinet over the sink area.", + "E": "In the cabinet above the sink?" + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_15", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the spice rack in the kitchen?", + "answer": "In the cabinet above the sink?", + "question_chinese": "Where is the seasoning display area in the kitchen?", + "answer_chinese": "In the cabinet above the washbasin?", + "evidence": [ + { + "description": "In the cabinet above the washbasin?", + "timestamps": { + "start": "00:08:24", + "end": "00:08:31" + } + } + ], + "question_timestamps": "00:08:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's on the counter by the stove.", + "B": "It's on the shelf above the fridge.", + "C": "In the cabinet above the sink?", + "D": "The drawer under the oven holds it, rather than being stored in the cabinet above the sink.", + "E": "It's behind the blender on the counter, not put away in the upper cabinet over the sink." + }, + "answer_label": "C", + "answer_text": "In the cabinet above the sink?", + "q_id": "P30_108_15", + "correct_answer": "In the cabinet above the sink?", + "original_options": { + "A": "It's on the counter by the stove.", + "B": "It's on the shelf above the fridge.", + "C": "In the cabinet above the sink?", + "D": "The drawer under the oven holds it, rather than being stored in the cabinet above the sink.", + "E": "It's behind the blender on the counter, not put away in the upper cabinet over the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_14", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mugs are there on the kitchen counter?", + "answer": "There are four mugs.", + "question_chinese": "厨房柜台上有几个马克杯？", + "answer_chinese": "有四个马克杯。", + "evidence": [ + { + "description": "There are four mugs.", + "timestamps": { + "start": "00:08:22", + "end": "00:08:44" + } + } + ], + "question_timestamps": "00:08:57", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three mugs.", + "B": "There are four mugs.", + "C": "There are five mugs.", + "D": "There are two mugs on the kitchen counter, not four mugs in total.", + "E": "There are six mugs on the kitchen counter, which is more than the stated amount." + }, + "answer_label": "B", + "answer_text": "There are four mugs.", + "q_id": "P30_108_14", + "correct_answer": "There are four mugs.", + "original_options": { + "A": "There are three mugs.", + "B": "There are four mugs.", + "C": "There are five mugs.", + "D": "There are two mugs on the kitchen counter, not four mugs in total.", + "E": "There are six mugs on the kitchen counter, which is more than the stated amount." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 15 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_13", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mugs are there on the kitchen counter?", + "answer": "There are four mugs.", + "question_chinese": "厨房柜台上有几个马克杯？", + "answer_chinese": "有四个马克杯。", + "evidence": [ + { + "description": "There are four mugs.", + "timestamps": { + "start": "00:08:22", + "end": "00:08:44" + } + } + ], + "question_timestamps": "00:09:32", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are five mugs.", + "B": "There are six mugs.", + "C": "There are seven mugs on the kitchen counter, not four mugs as stated.", + "D": "There are four mugs.", + "E": "There are eight mugs on the kitchen counter, which is double the number four." + }, + "answer_label": "D", + "answer_text": "There are four mugs.", + "q_id": "P30_108_13", + "correct_answer": "There are four mugs.", + "original_options": { + "A": "There are five mugs.", + "B": "There are six mugs.", + "C": "There are seven mugs on the kitchen counter, not four mugs as stated.", + "D": "There are four mugs.", + "E": "There are eight mugs on the kitchen counter, which is double the number four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls did I put into the closet in total?", + "answer": "You put in four bowls.", + "question_chinese": "我一共往壁柜里放入了几个碗？", + "answer_chinese": "你放了四个碗。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:48" + } + }, + { + "description": "The third one.", + "timestamps": { + "start": "00:00:53", + "end": "00:00:54" + } + }, + { + "description": "The forth one.", + "timestamps": { + "start": "00:00:56", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:02:12", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You put in three bowls.", + "B": "You put in five bowls.", + "C": "You put in four bowls.", + "D": "You put in two bowls in total across all trips to the closet.", + "E": "You put in six bowls altogether, counting each bowl you carried into the closet." + }, + "answer_label": "C", + "answer_text": "You put in four bowls.", + "q_id": "P30_108_1", + "correct_answer": "You put in four bowls.", + "original_options": { + "A": "You put in three bowls.", + "B": "You put in five bowls.", + "C": "You put in four bowls.", + "D": "You put in two bowls in total across all trips to the closet.", + "E": "You put in six bowls altogether, counting each bowl you carried into the closet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls did I put into the closet in total?", + "answer": "You put in four bowls.", + "question_chinese": "我一共往壁柜里放入了几个碗？", + "answer_chinese": "你放了四个碗。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:48" + } + }, + { + "description": "The third one.", + "timestamps": { + "start": "00:00:53", + "end": "00:00:54" + } + }, + { + "description": "The forth one.", + "timestamps": { + "start": "00:00:56", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You put in four bowls.", + "B": "You put in five bowls.", + "C": "You put in six bowls.", + "D": "You put in seven bowls total when you finished putting them into the closet.", + "E": "You put in eight bowls altogether, including every bowl you placed inside the closet." + }, + "answer_label": "A", + "answer_text": "You put in four bowls.", + "q_id": "P30_108_2", + "correct_answer": "You put in four bowls.", + "original_options": { + "A": "You put in four bowls.", + "B": "You put in five bowls.", + "C": "You put in six bowls.", + "D": "You put in seven bowls total when you finished putting them into the closet.", + "E": "You put in eight bowls altogether, including every bowl you placed inside the closet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_3", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I mainly do in the kitchen just now?", + "answer": "You mixed the drink.", + "question_chinese": "我刚才在厨房主要做了哪些事。", + "answer_chinese": "你刚才清洗了餐具并清洁了厨房，还冲兑了饮料。", + "evidence": [ + { + "description": "You just washed the tableware.", + "timestamps": { + "start": "00:01:09", + "end": "00:07:17" + } + }, + { + "description": "You cleaned the kitchen.", + "timestamps": { + "start": "00:07:27", + "end": "00:08:23" + } + } + ], + "question_timestamps": "00:08:44", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You chopped vegetables.", + "B": "You mixed the drink.", + "C": "You prepared the salad.", + "D": "You mainly arranged the dishes on the counter instead of mixing anything.", + "E": "You mostly prepared soup, focusing on heating and stirring rather than making a drink." + }, + "answer_label": "B", + "answer_text": "You mixed the drink.", + "q_id": "P30_108_3", + "correct_answer": "You mixed the drink.", + "original_options": { + "A": "You chopped vegetables.", + "B": "You mixed the drink.", + "C": "You prepared the salad.", + "D": "You mainly arranged the dishes on the counter instead of mixing anything.", + "E": "You mostly prepared soup, focusing on heating and stirring rather than making a drink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 14 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_4", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I mainly do in the kitchen just now?", + "answer": "You mixed the drink.", + "question_chinese": "我刚才在厨房主要做了哪些事。", + "answer_chinese": "你刚才清洗了餐具并清洁了厨房，还冲兑了饮料。", + "evidence": [ + { + "description": "You just washed the tableware.", + "timestamps": { + "start": "00:01:09", + "end": "00:07:17" + } + }, + { + "description": "You cleaned the kitchen.", + "timestamps": { + "start": "00:07:27", + "end": "00:08:23" + } + } + ], + "question_timestamps": "00:09:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You mixed the drink.", + "B": "You sliced the fruit.", + "C": "You wiped the dishes.", + "D": "You mainly stirred the soup on the stove rather than mixing a drink.", + "E": "You mostly shook the bottle for a moment, then set it down without mixing." + }, + "answer_label": "A", + "answer_text": "You mixed the drink.", + "q_id": "P30_108_4", + "correct_answer": "You mixed the drink.", + "original_options": { + "A": "You mixed the drink.", + "B": "You sliced the fruit.", + "C": "You wiped the dishes.", + "D": "You mainly stirred the soup on the stove rather than mixing a drink.", + "E": "You mostly shook the bottle for a moment, then set it down without mixing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_7", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I just put the iron basin?", + "answer": "You put the iron basin from beside the washbasin into the cabinet.", + "question_chinese": "我刚才将铁盆从哪里放在了哪里？", + "answer_chinese": "你将铁盆从洗手池旁放进了柜子里。", + "evidence": [ + { + "description": "You picked up the iron basin.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:04" + } + }, + { + "description": "You put the iron basin in the cabinet.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:03:24", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You put the iron basin from beside the washbasin into the cabinet.", + "B": "Into the drawer.", + "C": "Onto the countertop.", + "D": "You put the iron basin from in front of the washbasin onto the countertop.", + "E": "You moved the iron basin from beside the washbasin into a drawer instead of the cabinet." + }, + "answer_label": "A", + "answer_text": "You put the iron basin from beside the washbasin into the cabinet.", + "q_id": "P30_108_7", + "correct_answer": "You put the iron basin from beside the washbasin into the cabinet.", + "original_options": { + "A": "You put the iron basin from beside the washbasin into the cabinet.", + "B": "Into the drawer.", + "C": "Onto the countertop.", + "D": "You put the iron basin from in front of the washbasin onto the countertop.", + "E": "You moved the iron basin from beside the washbasin into a drawer instead of the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 16 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_8", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I just put the iron basin?", + "answer": "You put the iron basin from beside the washbasin into the cabinet.", + "question_chinese": "我刚才将铁盆从哪里放在了哪里？", + "answer_chinese": "你将铁盆从洗手池旁放进了柜子里。", + "evidence": [ + { + "description": "You picked up the iron basin.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:04" + } + }, + { + "description": "You put the iron basin in the cabinet.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:06:24", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Onto the countertop.", + "B": "Into the drawer.", + "C": "You put the iron basin from beside the washbasin into the cabinet.", + "D": "You put the iron basin from on the countertop into the cabinet, not from beside the washbasin.", + "E": "You moved the iron basin from beside the cabinet onto the shelf rather than into the cabinet." + }, + "answer_label": "C", + "answer_text": "You put the iron basin from beside the washbasin into the cabinet.", + "q_id": "P30_108_8", + "correct_answer": "You put the iron basin from beside the washbasin into the cabinet.", + "original_options": { + "A": "Onto the countertop.", + "B": "Into the drawer.", + "C": "You put the iron basin from beside the washbasin into the cabinet.", + "D": "You put the iron basin from on the countertop into the cabinet, not from beside the washbasin.", + "E": "You moved the iron basin from beside the cabinet onto the shelf rather than into the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_4_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pick up the pot from, and where did I move it to just now?", + "answer": "Just now, you moved the pot from next to the kettle to the side of the sink, and the pot is to your right.", + "question_chinese": "我刚才将锅从哪里拿到了哪里？", + "answer_chinese": "你刚才把锅从烧水壶旁边，拿到了洗碗池边上，锅在你的右边。", + "evidence": [ + { + "description": "You just took the pot from the kettle to the sink.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:01:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Just now, you moved the pot from next to the kettle to the side of the sink, and the pot is to your right.", + "B": "The pot is to your left.", + "C": "The pot is in front of you.", + "D": "You moved the pot from next to the kettle to the side of the sink, and it is behind you.", + "E": "You moved the pot from next to the kettle to the side of the sink, and it is above you." + }, + "answer_label": "A", + "answer_text": "Just now, you moved the pot from next to the kettle to the side of the sink, and the pot is to your right.", + "q_id": "P30_108_4_T1", + "correct_answer": "Just now, you moved the pot from next to the kettle to the side of the sink, and the pot is to your right.", + "original_options": { + "A": "Just now, you moved the pot from next to the kettle to the side of the sink, and the pot is to your right.", + "B": "The pot is to your left.", + "C": "The pot is in front of you.", + "D": "You moved the pot from next to the kettle to the side of the sink, and it is behind you.", + "E": "You moved the pot from next to the kettle to the side of the sink, and it is above you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 20 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_4_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pick up the pot from, and where did I move it to just now?", + "answer": "You just washed the pot next to the sink and then placed it on the drying rack; the pot is to your left.", + "question_chinese": "我刚才将锅从哪里拿到了哪里？", + "answer_chinese": "你刚才将锅从洗碗池旁清洗后放到了晾晒网上，锅在你的左边。", + "evidence": [ + { + "description": "You just put the pot on the drying net from the sink.", + "timestamps": { + "start": "00:05:24", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:06:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved the pot to your right.", + "B": "You just washed the pot next to the sink and then placed it on the drying rack; the pot is to your left.", + "C": "You moved the pot in front of you.", + "D": "You washed the pot next to the sink and placed it on the drying rack; the pot is behind you.", + "E": "You washed the pot next to the sink and placed it on the drying rack; the pot is above you." + }, + "answer_label": "B", + "answer_text": "You just washed the pot next to the sink and then placed it on the drying rack; the pot is to your left.", + "q_id": "P30_108_4_T2", + "correct_answer": "You just washed the pot next to the sink and then placed it on the drying rack; the pot is to your left.", + "original_options": { + "A": "You moved the pot to your right.", + "B": "You just washed the pot next to the sink and then placed it on the drying rack; the pot is to your left.", + "C": "You moved the pot in front of you.", + "D": "You washed the pot next to the sink and placed it on the drying rack; the pot is behind you.", + "E": "You washed the pot next to the sink and placed it on the drying rack; the pot is above you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 20, + 20 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_5", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I ate an apple. How many apples are left?", + "answer": "There are a total of 11 apples. After eating one, there are still six left.", + "question_chinese": "我吃掉了一个苹果，还剩下几个苹果？", + "answer_chinese": "一共有11个苹果，吃掉一个后还剩下六个。", + "evidence": [ + { + "description": "The first bag contains four apples.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:12" + } + }, + { + "description": "There are six apples in the second bag.", + "timestamps": { + "start": "00:09:14", + "end": "00:09:16" + } + } + ], + "question_timestamps": "00:09:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are a total of 11 apples. After eating one, there are still six left.", + "B": "Ten apples remain.", + "C": "Five apples remain.", + "D": "There are a total of 11 apples, so after eating one, there should be ten apples left.", + "E": "There are a total of 11 apples; after you eat one apple, you would still have five left." + }, + "answer_label": "A", + "answer_text": "There are a total of 11 apples. After eating one, there are still six left.", + "q_id": "P30_108_5", + "correct_answer": "There are a total of 11 apples. After eating one, there are still six left.", + "original_options": { + "A": "There are a total of 11 apples. After eating one, there are still six left.", + "B": "Ten apples remain.", + "C": "Five apples remain.", + "D": "There are a total of 11 apples, so after eating one, there should be ten apples left.", + "E": "There are a total of 11 apples; after you eat one apple, you would still have five left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 18 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_6", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I ate an apple. How many apples are left?", + "answer": "There are a total of 11 apples. After eating one, there are still six left.", + "question_chinese": "我吃掉了一个苹果，还剩下几个苹果？", + "answer_chinese": "一共有11个苹果，吃掉一个后还剩下六个。", + "evidence": [ + { + "description": "The first bag contains four apples.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:12" + } + }, + { + "description": "There are six apples in the second bag.", + "timestamps": { + "start": "00:09:14", + "end": "00:09:16" + } + } + ], + "question_timestamps": "00:09:35", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Nine apples are left.", + "B": "There are a total of 11 apples. After eating one, there are still six left.", + "C": "Four apples are left.", + "D": "There are a total of 11 apples, and after eating one, you would have nine apples left.", + "E": "There are a total of 11 apples, so eating one means you still have three apples left." + }, + "answer_label": "B", + "answer_text": "There are a total of 11 apples. After eating one, there are still six left.", + "q_id": "P30_108_6", + "correct_answer": "There are a total of 11 apples. After eating one, there are still six left.", + "original_options": { + "A": "Nine apples are left.", + "B": "There are a total of 11 apples. After eating one, there are still six left.", + "C": "Four apples are left.", + "D": "There are a total of 11 apples, and after eating one, you would have nine apples left.", + "E": "There are a total of 11 apples, so eating one means you still have three apples left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_5_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I washing right now?", + "answer": "You are washing a frying pan right now, and the sink is in front of you.", + "question_chinese": "我现在在洗什么东西？", + "answer_chinese": "你现在在洗一个平底锅，水槽在你的正前方。", + "evidence": [ + { + "description": "You are washing a pan now.", + "timestamps": { + "start": "00:04:07", + "end": "00:04:13" + } + } + ], + "question_timestamps": "00:04:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are washing a pot.", + "B": "You are washing a frying pan right now, and the sink is in front of you.", + "C": "You are washing a plate.", + "D": "You are washing a saucepan right now, and the sink is behind you.", + "E": "You are washing a lid at the moment, and the sink is in front of you as you scrub." + }, + "answer_label": "B", + "answer_text": "You are washing a frying pan right now, and the sink is in front of you.", + "q_id": "P30_108_5_T1", + "correct_answer": "You are washing a frying pan right now, and the sink is in front of you.", + "original_options": { + "A": "You are washing a pot.", + "B": "You are washing a frying pan right now, and the sink is in front of you.", + "C": "You are washing a plate.", + "D": "You are washing a saucepan right now, and the sink is behind you.", + "E": "You are washing a lid at the moment, and the sink is in front of you as you scrub." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 19 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_5_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I washing right now?", + "answer": "You are washing a plate right now, and the sink is in front of you.", + "question_chinese": "我现在在洗什么东西？", + "answer_chinese": "你现在在洗一个盘子，水槽在你的正前方。", + "evidence": [ + { + "description": "You are washing a dish in your hand.", + "timestamps": { + "start": "00:07:07", + "end": "00:07:12" + } + } + ], + "question_timestamps": "00:07:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are washing a bowl.", + "B": "You are washing a pan.", + "C": "You are washing a frying pan right now, and the sink is directly in front of you.", + "D": "You are washing a pot at the sink right now, with the basin positioned in front of you.", + "E": "You are washing a plate right now, and the sink is in front of you." + }, + "answer_label": "E", + "answer_text": "You are washing a plate right now, and the sink is in front of you.", + "q_id": "P30_108_5_T2", + "correct_answer": "You are washing a plate right now, and the sink is in front of you.", + "original_options": { + "A": "You are washing a bowl.", + "B": "You are washing a pan.", + "C": "You are washing a frying pan right now, and the sink is directly in front of you.", + "D": "You are washing a pot at the sink right now, with the basin positioned in front of you.", + "E": "You are washing a plate right now, and the sink is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_1_T3", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls did I put into the wall cabinet just now?", + "answer": "You have now put in one bowl, and the bowl is in front of you.", + "question_chinese": "我现在往壁柜里放入了几个碗？", + "answer_chinese": "你现在放了一个碗，碗在你的正前方。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two bowls.", + "B": "You have now put in one bowl, and the bowl is in front of you.", + "C": "Zero bowls.", + "D": "You have now put in two bowls, and the bowls are in front of you.", + "E": "You have now put in three bowls into the wall cabinet, and the bowls are in front of you." + }, + "answer_label": "B", + "answer_text": "You have now put in one bowl, and the bowl is in front of you.", + "q_id": "P30_108_1_T3", + "correct_answer": "You have now put in one bowl, and the bowl is in front of you.", + "original_options": { + "A": "Two bowls.", + "B": "You have now put in one bowl, and the bowl is in front of you.", + "C": "Zero bowls.", + "D": "You have now put in two bowls, and the bowls are in front of you.", + "E": "You have now put in three bowls into the wall cabinet, and the bowls are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 19 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_1_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls did I put into the wall cabinet just now?", + "answer": "You have now put a total of four bowls into the cupboard, and the bowls are in front of you.", + "question_chinese": "我现在往壁柜里放入了几个碗？", + "answer_chinese": "你现在往壁橱柜里一共放了四个碗，碗在你的正前方。", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:48" + } + }, + { + "description": "The third one.", + "timestamps": { + "start": "00:00:56", + "end": "00:00:59" + } + }, + { + "description": "The fourth one.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have now put a total of four bowls into the cupboard, and the bowls are in front of you.", + "B": "Five bowls.", + "C": "Two bowls.", + "D": "You have now put a total of three bowls into the cupboard, and the bowls are in front of you.", + "E": "You have now put a total of six bowls into the cupboard, and the bowls are in front of you." + }, + "answer_label": "A", + "answer_text": "You have now put a total of four bowls into the cupboard, and the bowls are in front of you.", + "q_id": "P30_108_1_T2", + "correct_answer": "You have now put a total of four bowls into the cupboard, and the bowls are in front of you.", + "original_options": { + "A": "You have now put a total of four bowls into the cupboard, and the bowls are in front of you.", + "B": "Five bowls.", + "C": "Two bowls.", + "D": "You have now put a total of three bowls into the cupboard, and the bowls are in front of you.", + "E": "You have now put a total of six bowls into the cupboard, and the bowls are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 20, + 20 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_3_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What items are on the drying rack next to my sink right now?", + "answer": "Right now, there is one bowl on the drying rack next to your sink, and the bowl is to your left.", + "question_chinese": "我的洗碗池旁的晾网上现在有什么东西？", + "answer_chinese": "你的洗碗池旁的晾网上现在有一个碗，碗在你的左边。", + "evidence": [ + { + "description": "Air drying, there is a bowl you just washed online.", + "timestamps": { + "start": "00:01:31", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two bowls.", + "B": "Right now, there is one bowl on the drying rack next to your sink, and the bowl is to your left.", + "C": "Zero bowls.", + "D": "Right now, there are three bowls on the drying rack next to your sink, and the bowls are to your left.", + "E": "Right now, there are four bowls on the drying rack next to your sink, and the bowls are to your left." + }, + "answer_label": "B", + "answer_text": "Right now, there is one bowl on the drying rack next to your sink, and the bowl is to your left.", + "q_id": "P30_108_3_T1", + "correct_answer": "Right now, there is one bowl on the drying rack next to your sink, and the bowl is to your left.", + "original_options": { + "A": "Two bowls.", + "B": "Right now, there is one bowl on the drying rack next to your sink, and the bowl is to your left.", + "C": "Zero bowls.", + "D": "Right now, there are three bowls on the drying rack next to your sink, and the bowls are to your left.", + "E": "Right now, there are four bowls on the drying rack next to your sink, and the bowls are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 21, + 21 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_3_T2", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What items are on the drying rack next to my sink right now?", + "answer": "Right now, there is one bowl, one spatula, and one lid on the drying rack next to your sink, and the bowl, spatula, and lid are to your left.", + "question_chinese": "我的洗碗池旁的晾网上现在有什么东西？", + "answer_chinese": "你的洗碗池旁的凉网上现在有一个碗、一个铲子和一个盖子,碗、铲子和盖子在你的左边。", + "evidence": [ + { + "description": "There is a bowl, a shovel and a lid on the cold net next to your sink.", + "timestamps": { + "start": "00:02:18", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:02:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two bowls, one spatula, one lid.", + "B": "One bowl, two spatulas, one lid.", + "C": "Right now, there is one bowl, one spatula, and one lid on the drying rack next to your sink, and the bowl, spatula, and lid are to your left.", + "D": "Right now, there is one bowl, one spatula, and two lids on the drying rack.", + "E": "Right now, there are zero bowls, one spatula, and one lid on the drying rack." + }, + "answer_label": "C", + "answer_text": "Right now, there is one bowl, one spatula, and one lid on the drying rack next to your sink, and the bowl, spatula, and lid are to your left.", + "q_id": "P30_108_3_T2", + "correct_answer": "Right now, there is one bowl, one spatula, and one lid on the drying rack next to your sink, and the bowl, spatula, and lid are to your left.", + "original_options": { + "A": "Two bowls, one spatula, one lid.", + "B": "One bowl, two spatulas, one lid.", + "C": "Right now, there is one bowl, one spatula, and one lid on the drying rack next to your sink, and the bowl, spatula, and lid are to your left.", + "D": "Right now, there is one bowl, one spatula, and two lids on the drying rack.", + "E": "Right now, there are zero bowls, one spatula, and one lid on the drying rack." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_6_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the hand towel relative to me?", + "answer": "The hand towel is in front of you.", + "question_chinese": "擦手巾在相对于我的哪里？", + "answer_chinese": "擦手巾在相对于你的正前方。", + "evidence": [ + { + "description": "Right in front of you.", + "timestamps": { + "start": "00:08:13", + "end": "00:08:15" + } + } + ], + "question_timestamps": "00:08:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The hand towel is positioned to your right, not directly in front of where you are standing.", + "D": "The hand towel is above you, hanging higher than your eye level rather than in front of you.", + "E": "The hand towel is in front of you." + }, + "answer_label": "E", + "answer_text": "The hand towel is in front of you.", + "q_id": "P30_108_6_T1", + "correct_answer": "The hand towel is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The hand towel is positioned to your right, not directly in front of where you are standing.", + "D": "The hand towel is above you, hanging higher than your eye level rather than in front of you.", + "E": "The hand towel is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 18 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_6_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the hand towel relative to me?", + "answer": "The hand towel is behind you.", + "question_chinese": "擦手巾在相对于我的哪里？", + "answer_chinese": "擦手巾在相对于你的后面。", + "evidence": [ + { + "description": "You turned around, so you became behind you.", + "timestamps": { + "start": "00:08:19", + "end": "00:08:24" + } + } + ], + "question_timestamps": "00:08:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is to your right.", + "C": "The hand towel is to your left side, not behind you as you face forward.", + "D": "The hand towel is above you, placed higher up rather than located behind your back.", + "E": "The hand towel is behind you." + }, + "answer_label": "E", + "answer_text": "The hand towel is behind you.", + "q_id": "P30_108_6_T2", + "correct_answer": "The hand towel is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is to your right.", + "C": "The hand towel is to your left side, not behind you as you face forward.", + "D": "The hand towel is above you, placed higher up rather than located behind your back.", + "E": "The hand towel is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 15 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_2_T1", + "qid_num": 25, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the toaster relative to me?", + "answer": "The toaster is in front of you.", + "question_chinese": "烤面包机在相对于我的哪里？", + "answer_chinese": "烤面包机在你的正前方。", + "evidence": [ + { + "description": "Right in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The toaster is to your right, not positioned straight ahead in your forward line of sight.", + "D": "The toaster is in front of you.", + "E": "The toaster is above you, located higher up rather than directly in front of you." + }, + "answer_label": "D", + "answer_text": "The toaster is in front of you.", + "q_id": "P30_108_2_T1", + "correct_answer": "The toaster is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The toaster is to your right, not positioned straight ahead in your forward line of sight.", + "D": "The toaster is in front of you.", + "E": "The toaster is above you, located higher up rather than directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 15 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_2_T2", + "qid_num": 26, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the toaster relative to me?", + "answer": "The toaster is behind you.", + "question_chinese": "烤面包机在相对于我的哪里？", + "answer_chinese": "烤面包机在你的后面。", + "evidence": [ + { + "description": "Behind you", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The toaster is behind you.", + "B": "It is in front of you.", + "C": "It is above you.", + "D": "The toaster is to your left side, not behind you relative to your current position.", + "E": "The toaster is to your right, positioned beside you rather than behind your back." + }, + "answer_label": "A", + "answer_text": "The toaster is behind you.", + "q_id": "P30_108_2_T2", + "correct_answer": "The toaster is behind you.", + "original_options": { + "A": "The toaster is behind you.", + "B": "It is in front of you.", + "C": "It is above you.", + "D": "The toaster is to your left side, not behind you relative to your current position.", + "E": "The toaster is to your right, positioned beside you rather than behind your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 15, + 14 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_8_T2", + "qid_num": 27, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the kettle relative to me?", + "answer": "The kettle is in front of you.", + "question_chinese": "烧水壶在相对于我的哪里？", + "answer_chinese": "烧水壶在相对于你的前面。", + "evidence": [ + { + "description": "The kettle is in front of you.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The kettle is in front of you.", + "B": "It is behind you.", + "C": "It is to your right.", + "D": "The kettle is to your left side, not directly in front of you as you look ahead.", + "E": "The kettle is above you, placed higher than your view rather than in front of your body." + }, + "answer_label": "A", + "answer_text": "The kettle is in front of you.", + "q_id": "P30_108_8_T2", + "correct_answer": "The kettle is in front of you.", + "original_options": { + "A": "The kettle is in front of you.", + "B": "It is behind you.", + "C": "It is to your right.", + "D": "The kettle is to your left side, not directly in front of you as you look ahead.", + "E": "The kettle is above you, placed higher than your view rather than in front of your body." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_8_T1", + "qid_num": 28, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the kettle relative to me?", + "answer": "The kettle is to your left.", + "question_chinese": "烧水壶在相对于我的哪里？", + "answer_chinese": "烧水壶在相对于你的左边。", + "evidence": [ + { + "description": "The kettle is on your left.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The kettle is to your left.", + "B": "It is in front of you.", + "C": "It is behind you.", + "D": "The kettle is to your right, positioned on the opposite side rather than on your left.", + "E": "The kettle is above you, set higher up instead of being located to your left side." + }, + "answer_label": "A", + "answer_text": "The kettle is to your left.", + "q_id": "P30_108_8_T1", + "correct_answer": "The kettle is to your left.", + "original_options": { + "A": "The kettle is to your left.", + "B": "It is in front of you.", + "C": "It is behind you.", + "D": "The kettle is to your right, positioned on the opposite side rather than on your left.", + "E": "The kettle is above you, set higher up instead of being located to your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 16, + 16 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_9_T1", + "qid_num": 29, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How many white cups are on the table in front of me right now?", + "answer": "There are two white cups on the table in front of you; the cups are in front of you.", + "question_chinese": "现在我面前的桌子上有多少个白色的杯子？", + "answer_chinese": "现在你面前的桌子上有两个白色的杯子，杯子在你的前面。", + "evidence": [ + { + "description": "Now there are two white cups on the table in front of you.", + "timestamps": { + "start": "00:08:31", + "end": "00:08:32" + } + } + ], + "question_timestamps": "00:08:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One white cup.", + "B": "Three white cups.", + "C": "There are two white cups on the table in front of you; the cups are in front of you.", + "D": "There are zero white cups on the table in front of you; there are no cups directly ahead.", + "E": "There are four white cups on the table in front of you; all four cups sit directly in front of you." + }, + "answer_label": "C", + "answer_text": "There are two white cups on the table in front of you; the cups are in front of you.", + "q_id": "P30_108_9_T1", + "correct_answer": "There are two white cups on the table in front of you; the cups are in front of you.", + "original_options": { + "A": "One white cup.", + "B": "Three white cups.", + "C": "There are two white cups on the table in front of you; the cups are in front of you.", + "D": "There are zero white cups on the table in front of you; there are no cups directly ahead.", + "E": "There are four white cups on the table in front of you; all four cups sit directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 21 + ] + }, + { + "video_uid": "P30_108", + "video_length": "00:09:35", + "qid": "P30_108_9_T2", + "qid_num": 30, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How many white cups are on the table in front of me right now?", + "answer": "There is one white cup on the table in front of you; the cup is in front of you.", + "question_chinese": "现在我面前的桌子上有多少个白色的杯子？", + "answer_chinese": "现在你面前的桌子上有一个白色的杯子，杯子在你的前面。", + "evidence": [ + { + "description": "Now there is a white cup on the table in front of you.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:04" + } + } + ], + "question_timestamps": "00:09:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Two white cups.", + "B": "Zero white cups.", + "C": "There are three white cups on the table in front of you; all three cups are positioned ahead of you.", + "D": "There is one white cup on the table in front of you; the cup is in front of you.", + "E": "There are four white cups on the table in front of you; each cup is placed in front of you." + }, + "answer_label": "D", + "answer_text": "There is one white cup on the table in front of you; the cup is in front of you.", + "q_id": "P30_108_9_T2", + "correct_answer": "There is one white cup on the table in front of you; the cup is in front of you.", + "original_options": { + "A": "Two white cups.", + "B": "Zero white cups.", + "C": "There are three white cups on the table in front of you; all three cups are positioned ahead of you.", + "D": "There is one white cup on the table in front of you; the cup is in front of you.", + "E": "There are four white cups on the table in front of you; each cup is placed in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 20, + 20 + ] + } +] diff --git a/QAs/P30_112-mcq.json b/QAs/P30_112-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..60deb62bdc21bcc5d33d958f27f844ba7b91391a --- /dev/null +++ b/QAs/P30_112-mcq.json @@ -0,0 +1,1272 @@ +[ + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_3_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "To handle this pizza, how many tools did I use just now?", + "answer": "In total, you used three tools: a baking tray, a cutting board, and a knife.", + "question_chinese": "为了处理这份披萨，我刚才用了几个道具？", + "answer_chinese": "你总共用了三个道具，烤盘，菜板和刀。", + "evidence": [ + { + "description": "You put the pizza on the baking tray.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:16" + } + }, + { + "description": "You used a knife and a cutting board.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used one tool: a knife.", + "B": "You used two tools: board and knife.", + "C": "In total, you used three tools: a baking tray, a cutting board, and a knife.", + "D": "In total, you used four tools: a baking tray, a cutting board, a knife, and a plate.", + "E": "In total, you used five tools: a baking tray, a cutting board, a knife, a plate, and a pizza cutter." + }, + "answer_label": "C", + "answer_text": "In total, you used three tools: a baking tray, a cutting board, and a knife.", + "q_id": "P30_112_3_T2", + "correct_answer": "In total, you used three tools: a baking tray, a cutting board, and a knife.", + "original_options": { + "A": "You used one tool: a knife.", + "B": "You used two tools: board and knife.", + "C": "In total, you used three tools: a baking tray, a cutting board, and a knife.", + "D": "In total, you used four tools: a baking tray, a cutting board, a knife, and a plate.", + "E": "In total, you used five tools: a baking tray, a cutting board, a knife, a plate, and a pizza cutter." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 20 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_3_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "To handle this pizza, how many tools did I use just now?", + "answer": "You used four tools in total: a baking tray, a cutting board, a knife, and a plate.", + "question_chinese": "为了处理这份披萨，我刚才用了几个道具？", + "answer_chinese": "你总共用了四个道具，烤盘，菜板，刀和盘子。", + "evidence": [ + { + "description": "You put the pizza on the baking tray.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:16" + } + }, + { + "description": "You used a knife and a cutting board.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:52" + } + }, + { + "description": "You put the pizza on the plate.", + "timestamps": { + "start": "00:00:57", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used two tools in total.", + "B": "You used four tools in total: a baking tray, a cutting board, a knife, and a plate.", + "C": "You used three tools in total.", + "D": "You used five tools in total: a baking tray, a cutting board, a knife, a plate, and a pizza cutter.", + "E": "You used six tools in total: a baking tray, a cutting board, a knife, a plate, a pizza cutter, and a spatula." + }, + "answer_label": "B", + "answer_text": "You used four tools in total: a baking tray, a cutting board, a knife, and a plate.", + "q_id": "P30_112_3_T1", + "correct_answer": "You used four tools in total: a baking tray, a cutting board, a knife, and a plate.", + "original_options": { + "A": "You used two tools in total.", + "B": "You used four tools in total: a baking tray, a cutting board, a knife, and a plate.", + "C": "You used three tools in total.", + "D": "You used five tools in total: a baking tray, a cutting board, a knife, a plate, and a pizza cutter.", + "E": "You used six tools in total: a baking tray, a cutting board, a knife, a plate, a pizza cutter, and a spatula." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 22 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_15", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the general layout of the kitchen?", + "answer": "There is a small corridor with sundries scattered on both sides. On the left is an oven, a sink and a refrigerator.", + "question_chinese": "厨房的大致布局是什么？", + "answer_chinese": "是一条小走廊左右分布着杂物，左边一次是烤箱洗水槽和冰箱。", + "evidence": [ + { + "description": "There is a small corridor with sundries scattered on both sides.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The oven.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:18" + } + }, + { + "description": "The sink.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:25" + } + }, + { + "description": "The refrigerator.", + "timestamps": { + "start": "00:01:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "There is a small corridor with sundries scattered on both sides. On the left is an oven, a sink and a refrigerator.", + "B": "Appliances are on your right.", + "C": "Corridor is completely clear.", + "D": "There is a small corridor with sundries scattered on both sides, and the left side has a blender, microwave, and countertop.", + "E": "There is a small corridor with sundries on both sides, and on the left you see a kettle, toaster, and food processor." + }, + "answer_label": "A", + "answer_text": "There is a small corridor with sundries scattered on both sides. On the left is an oven, a sink and a refrigerator.", + "q_id": "P30_112_15", + "correct_answer": "There is a small corridor with sundries scattered on both sides. On the left is an oven, a sink and a refrigerator.", + "original_options": { + "A": "There is a small corridor with sundries scattered on both sides. On the left is an oven, a sink and a refrigerator.", + "B": "Appliances are on your right.", + "C": "Corridor is completely clear.", + "D": "There is a small corridor with sundries scattered on both sides, and the left side has a blender, microwave, and countertop.", + "E": "There is a small corridor with sundries on both sides, and on the left you see a kettle, toaster, and food processor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 21, + 22 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_16", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the general layout of the kitchen?", + "answer": "There is a small corridor with sundries scattered on both sides. On the left is an oven, a sink and a refrigerator.", + "question_chinese": "厨房的大致布局是什么？", + "answer_chinese": "是一条小走廊左右分布着杂物，左边一次是烤箱洗水槽和冰箱。", + "evidence": [ + { + "description": "There is a small corridor with sundries scattered on both sides.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The oven.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:18" + } + }, + { + "description": "The sink.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:25" + } + }, + { + "description": "The refrigerator.", + "timestamps": { + "start": "00:01:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:01:19", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "There is a small corridor with sundries scattered on both sides. On the left is an oven, a sink and a refrigerator.", + "B": "Appliances are on the right.", + "C": "Left has blender, dishwasher.", + "D": "There is a small corridor with sundries scattered on both sides, and on the right is a grill and a counter.", + "E": "There is a small corridor with sundries scattered on both sides, and on the right is a freezer and a stove." + }, + "answer_label": "A", + "answer_text": "There is a small corridor with sundries scattered on both sides. On the left is an oven, a sink and a refrigerator.", + "q_id": "P30_112_16", + "correct_answer": "There is a small corridor with sundries scattered on both sides. On the left is an oven, a sink and a refrigerator.", + "original_options": { + "A": "There is a small corridor with sundries scattered on both sides. On the left is an oven, a sink and a refrigerator.", + "B": "Appliances are on the right.", + "C": "Left has blender, dishwasher.", + "D": "There is a small corridor with sundries scattered on both sides, and on the right is a grill and a counter.", + "E": "There is a small corridor with sundries scattered on both sides, and on the right is a freezer and a stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 21, + 21 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_5_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "About how far is the green plant on the kitchen window from me?", + "answer": "The green plant was about 1 meter away from you.", + "question_chinese": "厨房里的窗户上的绿植距离我大概多远？", + "answer_chinese": "绿植离你大概一米。", + "evidence": [ + { + "description": "The green plants on the window are about one meter away from me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The green plant was about 5 meters away from you.", + "B": "The green plant was about 1 meter away from you." + }, + "answer_label": "B", + "answer_text": "The green plant was about 1 meter away from you.", + "q_id": "P30_112_5_T2", + "correct_answer": "The green plant was about 1 meter away from you.", + "original_options": { + "A": "The green plant was about 5 meters away from you.", + "B": "The green plant was about 1 meter away from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_5_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "About how far is the green plant on the kitchen window from me?", + "answer": "The green plant was about 5 meters away from you.", + "question_chinese": "厨房里的窗户上的绿植距离我大概多远？", + "answer_chinese": "绿植离你大概五米远。", + "evidence": [ + { + "description": "The green plants on the window are about one meter away from me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "You walked four steps to the right", + "timestamps": { + "start": "00:00:22", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The green plant was about 5 meters away from you.", + "B": "The green plant was about 1 meter away from you." + }, + "answer_label": "A", + "answer_text": "The green plant was about 5 meters away from you.", + "q_id": "P30_112_5_T1", + "correct_answer": "The green plant was about 5 meters away from you.", + "original_options": { + "A": "The green plant was about 5 meters away from you.", + "B": "The green plant was about 1 meter away from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the condition of the window in the kitchen?", + "answer": "The window is closed.", + "question_chinese": "厨房里的窗户是什么状态？", + "answer_chinese": "窗户是关着的。", + "evidence": [ + { + "description": "The window is closed.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The window is open.", + "B": "The window is closed." + }, + "answer_label": "B", + "answer_text": "The window is closed.", + "q_id": "P30_112_11", + "correct_answer": "The window is closed.", + "original_options": { + "A": "The window is open.", + "B": "The window is closed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_12", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the condition of the window in the kitchen?", + "answer": "The window is closed.", + "question_chinese": "厨房里的窗户是什么状态？", + "answer_chinese": "窗户是关着的。", + "evidence": [ + { + "description": "The window is closed.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The window is closed.", + "B": "The window is open." + }, + "answer_label": "A", + "answer_text": "The window is closed.", + "q_id": "P30_112_12", + "correct_answer": "The window is closed.", + "original_options": { + "A": "The window is closed.", + "B": "The window is open." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_13", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the knife for cutting pizza?", + "answer": "It was taken from the plate rack.", + "question_chinese": "我切披萨的刀是从哪里拿的？", + "answer_chinese": "是从餐盘架上拿的。", + "evidence": [ + { + "description": "It was taken from the plate rack.", + "timestamps": { + "start": "00:00:026", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the sink drawer.", + "B": "From the knife block.", + "C": "It was taken from the utensil holder near the stove, not from the plate rack.", + "D": "It was taken from the plate rack.", + "E": "It was taken from the shelf above the cutting board rather than being pulled from the plate rack." + }, + "answer_label": "D", + "answer_text": "It was taken from the plate rack.", + "q_id": "P30_112_13", + "correct_answer": "It was taken from the plate rack.", + "original_options": { + "A": "From the sink drawer.", + "B": "From the knife block.", + "C": "It was taken from the utensil holder near the stove, not from the plate rack.", + "D": "It was taken from the plate rack.", + "E": "It was taken from the shelf above the cutting board rather than being pulled from the plate rack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 18 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_14", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the knife for cutting pizza?", + "answer": "It was taken from the plate rack.", + "question_chinese": "我切披萨的刀是从哪里拿的？", + "answer_chinese": "是从餐盘架上拿的。", + "evidence": [ + { + "description": "It was taken from the plate rack.", + "timestamps": { + "start": "00:00:026", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the sink drawer.", + "B": "From the cutting board.", + "C": "It was taken from the plate rack.", + "D": "It was taken from the utensil holder by the stove, instead of being retrieved from the plate rack.", + "E": "It was taken from the shelf above the kitchen sink rather than from the plate rack." + }, + "answer_label": "C", + "answer_text": "It was taken from the plate rack.", + "q_id": "P30_112_14", + "correct_answer": "It was taken from the plate rack.", + "original_options": { + "A": "From the sink drawer.", + "B": "From the cutting board.", + "C": "It was taken from the plate rack.", + "D": "It was taken from the utensil holder by the stove, instead of being retrieved from the plate rack.", + "E": "It was taken from the shelf above the kitchen sink rather than from the plate rack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 16 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_2_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, how many slices did I cut the pizza into?", + "answer": "You cut the pizza into two pieces.", + "question_chinese": "我刚才把披萨切成了几份？", + "answer_chinese": "你把披萨切成了两份。", + "evidence": [ + { + "description": "You cut the pizza.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:33" + } + }, + { + "description": "You cut the pizza into two pieces.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You cut the pizza into two pieces.", + "B": "One piece.", + "C": "Three pieces.", + "D": "You cut the pizza into four pieces by making two cuts across it.", + "E": "You didn't cut it at all, leaving it whole without making any slices." + }, + "answer_label": "A", + "answer_text": "You cut the pizza into two pieces.", + "q_id": "P30_112_2_T1", + "correct_answer": "You cut the pizza into two pieces.", + "original_options": { + "A": "You cut the pizza into two pieces.", + "B": "One piece.", + "C": "Three pieces.", + "D": "You cut the pizza into four pieces by making two cuts across it.", + "E": "You didn't cut it at all, leaving it whole without making any slices." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_5", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many portions did I just cut the pizza into?", + "answer": "You cut the pizza into six pieces.", + "question_chinese": "我刚才把披萨切成了几份？", + "answer_chinese": "你把披萨切成了六份。", + "evidence": [ + { + "description": "You cut the pizza.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:33" + } + }, + { + "description": "You cut the pizza into six pieces.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:00:59", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four pieces.", + "B": "Eight pieces.", + "C": "You cut the pizza into six pieces.", + "D": "You cut the pizza into five pieces, stopping one cut short of making six equal portions.", + "E": "You cut the pizza into seven pieces, making more portions than six by adding an extra slice." + }, + "answer_label": "C", + "answer_text": "You cut the pizza into six pieces.", + "q_id": "P30_112_5", + "correct_answer": "You cut the pizza into six pieces.", + "original_options": { + "A": "Four pieces.", + "B": "Eight pieces.", + "C": "You cut the pizza into six pieces.", + "D": "You cut the pizza into five pieces, stopping one cut short of making six equal portions.", + "E": "You cut the pizza into seven pieces, making more portions than six by adding an extra slice." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_2_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, how many slices did I cut the pizza into?", + "answer": "You cut the pizza into six slices.", + "question_chinese": "我刚才把披萨切成了几份？", + "answer_chinese": "你把披萨切成了六份。", + "evidence": [ + { + "description": "You cut the pizza.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:33" + } + }, + { + "description": "You cut the pizza into six pieces.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four slices.", + "B": "You cut the pizza into six slices.", + "C": "Five slices.", + "D": "You cut the pizza into seven slices, creating one more slice than a six-slice cut would.", + "E": "You cut the pizza into eight slices, dividing it further than six by adding two extra cuts." + }, + "answer_label": "B", + "answer_text": "You cut the pizza into six slices.", + "q_id": "P30_112_2_T2", + "correct_answer": "You cut the pizza into six slices.", + "original_options": { + "A": "Four slices.", + "B": "You cut the pizza into six slices.", + "C": "Five slices.", + "D": "You cut the pizza into seven slices, creating one more slice than a six-slice cut would.", + "E": "You cut the pizza into eight slices, dividing it further than six by adding two extra cuts." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_1_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my walking route in the kitchen?", + "answer": "You walked to the front of the oven first.", + "question_chinese": "我在厨房的行走路线是什么？", + "answer_chinese": "你先走到了烤箱前。", + "evidence": [ + { + "description": "You walked to the oven first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went to the sink first.", + "B": "You walked behind the oven first.", + "C": "You walked to the front of the oven first.", + "D": "You walked to the sink first, then headed over to the front of the oven afterward.", + "E": "You started by turning left toward the counter, then approached the oven rather than going to its front first." + }, + "answer_label": "C", + "answer_text": "You walked to the front of the oven first.", + "q_id": "P30_112_1_T1", + "correct_answer": "You walked to the front of the oven first.", + "original_options": { + "A": "You went to the sink first.", + "B": "You walked behind the oven first.", + "C": "You walked to the front of the oven first.", + "D": "You walked to the sink first, then headed over to the front of the oven afterward.", + "E": "You started by turning left toward the counter, then approached the oven rather than going to its front first." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 19 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_3", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What's my walking route in the kitchen?", + "answer": "You first walked to the oven and then went to the sink.", + "question_chinese": "我在厨房的行走路线是什么？", + "answer_chinese": "你先走到了烤箱前，然后去到了洗水池旁。", + "evidence": [ + { + "description": "You walked to the oven first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "You then went to the sink.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went to the sink, then the oven.", + "B": "You walked to the oven, then refrigerator.", + "C": "You first walked over to the dishwasher, and after that you headed to the sink.", + "D": "You started at the refrigerator and then crossed the kitchen to stop at the oven.", + "E": "You first walked to the oven and then went to the sink." + }, + "answer_label": "E", + "answer_text": "You first walked to the oven and then went to the sink.", + "q_id": "P30_112_3", + "correct_answer": "You first walked to the oven and then went to the sink.", + "original_options": { + "A": "You went to the sink, then the oven.", + "B": "You walked to the oven, then refrigerator.", + "C": "You first walked over to the dishwasher, and after that you headed to the sink.", + "D": "You started at the refrigerator and then crossed the kitchen to stop at the oven.", + "E": "You first walked to the oven and then went to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 15 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_1_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my walking route in the kitchen?", + "answer": "You walked to the front of the oven first, then went over to the sink.", + "question_chinese": "我在厨房的行走路线是什么？", + "answer_chinese": "你先走到了烤箱前，然后去到了洗水池旁。", + "evidence": [ + { + "description": "You walked to the oven first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "You then went to the sink.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went to the back of the oven.", + "B": "You walked to the sink first.", + "C": "You walked to the back of the oven first, then headed over toward the sink.", + "D": "You walked to the front of the oven first, then went over to the sink.", + "E": "You walked to the oven area and stopped there, without continuing on to the sink." + }, + "answer_label": "D", + "answer_text": "You walked to the front of the oven first, then went over to the sink.", + "q_id": "P30_112_1_T2", + "correct_answer": "You walked to the front of the oven first, then went over to the sink.", + "original_options": { + "A": "You went to the back of the oven.", + "B": "You walked to the sink first.", + "C": "You walked to the back of the oven first, then headed over toward the sink.", + "D": "You walked to the front of the oven first, then went over to the sink.", + "E": "You walked to the oven area and stopped there, without continuing on to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 15, + 15 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_6_T1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pick up the knife I used to cut the pizza?", + "answer": "From the dish rack in front of you.", + "question_chinese": "我用来切披萨的刀从哪里拿的？", + "answer_chinese": "从你前面的餐盘架。", + "evidence": [ + { + "description": "It was taken from the plate rack.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:28" + } + }, + { + "description": "You walked up to the tray rack.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the dish rack behind you.", + "B": "From the dish rack to your left.", + "C": "You picked it up from the dish rack to your right, not from the one in front of you.", + "D": "From the dish rack in front of you.", + "E": "You grabbed the knife from a dish rack positioned above you, before using it to cut the pizza." + }, + "answer_label": "D", + "answer_text": "From the dish rack in front of you.", + "q_id": "P30_112_6_T1", + "correct_answer": "From the dish rack in front of you.", + "original_options": { + "A": "From the dish rack behind you.", + "B": "From the dish rack to your left.", + "C": "You picked it up from the dish rack to your right, not from the one in front of you.", + "D": "From the dish rack in front of you.", + "E": "You grabbed the knife from a dish rack positioned above you, before using it to cut the pizza." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 18 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_6_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I pick up the knife I used to cut the pizza?", + "answer": "From the dish rack to your left.", + "question_chinese": "我用来切披萨的刀从哪里拿的？", + "answer_chinese": "从你左边的餐盘架。", + "evidence": [ + { + "description": "It was taken from the plate rack.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the dish rack in front of you.", + "B": "From the dish rack to your right.", + "C": "From the dish rack to your left.", + "D": "You picked it up from the dish rack behind you rather than the one to your left.", + "E": "You grabbed the knife from the dish rack above you, not from the rack on your left side." + }, + "answer_label": "C", + "answer_text": "From the dish rack to your left.", + "q_id": "P30_112_6_T2", + "correct_answer": "From the dish rack to your left.", + "original_options": { + "A": "From the dish rack in front of you.", + "B": "From the dish rack to your right.", + "C": "From the dish rack to your left.", + "D": "You picked it up from the dish rack behind you rather than the one to your left.", + "E": "You grabbed the knife from the dish rack above you, not from the rack on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 18 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_4_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the sponge on the kitchen shelf relative to me?", + "answer": "It is to your left.", + "question_chinese": "我看到厨房里架子上的海绵在相对于我的什么位置？", + "answer_chinese": "在你的左面。", + "evidence": [ + { + "description": "The shelf for sponges is in front of me", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It is to your left.", + "B": "It is to your right.", + "C": "It is behind you.", + "D": "It is in front of you, on the kitchen shelf rather than off to your left.", + "E": "It is above you on the shelf, not positioned on the left side relative to you." + }, + "answer_label": "A", + "answer_text": "It is to your left.", + "q_id": "P30_112_4_T2", + "correct_answer": "It is to your left.", + "original_options": { + "A": "It is to your left.", + "B": "It is to your right.", + "C": "It is behind you.", + "D": "It is in front of you, on the kitchen shelf rather than off to your left.", + "E": "It is above you on the shelf, not positioned on the left side relative to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_4_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the sponge on the kitchen shelf relative to me?", + "answer": "It is behind you.", + "question_chinese": "我看到厨房里架子上的海绵在相对于我的什么位置？", + "answer_chinese": "在你的后面。", + "evidence": [ + { + "description": "There are three in total.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The rack is opposite the oven, and I am facing the oven.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It is in front of you.", + "B": "It is behind you.", + "C": "It is to your left.", + "D": "It is to your right on the shelf, rather than being located behind you.", + "E": "It is above you on the kitchen shelf, not positioned behind you relative to where you stand." + }, + "answer_label": "B", + "answer_text": "It is behind you.", + "q_id": "P30_112_4_T1", + "correct_answer": "It is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is behind you.", + "C": "It is to your left.", + "D": "It is to your right on the shelf, rather than being located behind you.", + "E": "It is above you on the kitchen shelf, not positioned behind you relative to where you stand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 17 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_7_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, what is the rough layout of the kitchen that I can see?", + "answer": "You saw a short corridor with clutter on both the left and right.", + "question_chinese": "目前我看到厨房的大致布局是什么？", + "answer_chinese": "你看到是一条小走廊左右分布着杂物", + "evidence": [ + { + "description": "There is a small corridor with sundries scattered on both sides.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw a short corridor with clutter on both the left and right.", + "B": "You saw clutter only on the left.", + "C": "You saw clutter only on the right.", + "D": "You saw a short corridor with clutter on both sides, but the space opened up immediately at the end.", + "E": "You saw a corridor with clutter on both sides, and it looked longer and more open than a short passage." + }, + "answer_label": "A", + "answer_text": "You saw a short corridor with clutter on both the left and right.", + "q_id": "P30_112_7_T1", + "correct_answer": "You saw a short corridor with clutter on both the left and right.", + "original_options": { + "A": "You saw a short corridor with clutter on both the left and right.", + "B": "You saw clutter only on the left.", + "C": "You saw clutter only on the right.", + "D": "You saw a short corridor with clutter on both sides, but the space opened up immediately at the end.", + "E": "You saw a corridor with clutter on both sides, and it looked longer and more open than a short passage." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 20 + ] + }, + { + "video_uid": "P30_112", + "video_length": "00:01:20", + "qid": "P30_112_7_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, what is the rough layout of the kitchen that I can see?", + "answer": "You saw a short corridor with clutter on both the left and right; to your left in order were an oven, a sink, and a refrigerator.", + "question_chinese": "目前我看到厨房的大致布局是什么？", + "answer_chinese": "你看到是一条小走廊左右分布着杂物，左边一次是烤箱洗水槽和冰箱。", + "evidence": [ + { + "description": "There is a small corridor with sundries scattered on both sides.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The oven.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:17" + } + }, + { + "description": "The sink.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:25" + } + }, + { + "description": "The refrigerator.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:01:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The appliances were in front of you.", + "B": "The appliances were to your right.", + "C": "You saw a short corridor with clutter on both sides, but the oven, sink, and refrigerator were behind you in order.", + "D": "You saw a short corridor with clutter on both sides; to your left were a refrigerator, then a sink, then an oven.", + "E": "You saw a short corridor with clutter on both the left and right; to your left in order were an oven, a sink, and a refrigerator." + }, + "answer_label": "E", + "answer_text": "You saw a short corridor with clutter on both the left and right; to your left in order were an oven, a sink, and a refrigerator.", + "q_id": "P30_112_7_T2", + "correct_answer": "You saw a short corridor with clutter on both the left and right; to your left in order were an oven, a sink, and a refrigerator.", + "original_options": { + "A": "The appliances were in front of you.", + "B": "The appliances were to your right.", + "C": "You saw a short corridor with clutter on both sides, but the oven, sink, and refrigerator were behind you in order.", + "D": "You saw a short corridor with clutter on both sides; to your left were a refrigerator, then a sink, then an oven.", + "E": "You saw a short corridor with clutter on both the left and right; to your left in order were an oven, a sink, and a refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 21, + 22 + ] + } +] diff --git a/QAs/P33_101-mcq.json b/QAs/P33_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..67937a3d9f06c31171b9a1a00bcb5b48d45d691d --- /dev/null +++ b/QAs/P33_101-mcq.json @@ -0,0 +1,1025 @@ +[ + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What was the first thing I did when I just entered the door?", + "answer": "Turn on the light.", + "question_chinese": "我刚进门第一件事做了什么", + "answer_chinese": "开灯", + "evidence": [ + { + "description": "Turn on the light.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:03:33", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Take off your shoes.", + "B": "Close the window.", + "C": "You looked out the window right after stepping inside, before doing anything else.", + "D": "Turn on the light.", + "E": "You walked straight to the sofa first, then paused as if deciding what to do next." + }, + "answer_label": "D", + "answer_text": "Turn on the light.", + "q_id": "P33_101_1", + "correct_answer": "Turn on the light.", + "original_options": { + "A": "Take off your shoes.", + "B": "Close the window.", + "C": "You looked out the window right after stepping inside, before doing anything else.", + "D": "Turn on the light.", + "E": "You walked straight to the sofa first, then paused as if deciding what to do next." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 13, + 16 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What was the first thing I did when I just entered the door?", + "answer": "Turn on the light.", + "question_chinese": "我刚进门第一件事做了什么", + "answer_chinese": "开灯", + "evidence": [ + { + "description": "Turn on the light.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:06:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Remove your shoes.", + "B": "Look at your watch.", + "C": "Turn on the light.", + "D": "You placed your bag on the floor immediately after entering, before moving any further inside.", + "E": "You closed the window first, then turned your attention back toward the entryway area." + }, + "answer_label": "C", + "answer_text": "Turn on the light.", + "q_id": "P33_101_2", + "correct_answer": "Turn on the light.", + "original_options": { + "A": "Remove your shoes.", + "B": "Look at your watch.", + "C": "Turn on the light.", + "D": "You placed your bag on the floor immediately after entering, before moving any further inside.", + "E": "You closed the window first, then turned your attention back toward the entryway area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 14 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_9", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands at my current location?", + "answer": "Yes.", + "question_chinese": "我在当前位置可以洗到手吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "You are in front of the sink.", + "timestamps": { + "start": "00:03:15", + "end": "00:03:19" + } + } + ], + "question_timestamps": "00:03:19", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "P33_101_9", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_10", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands at my current location?", + "answer": "No.", + "question_chinese": "我在当前位置可以洗到手吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:03:15", + "end": "00:03:19" + } + }, + { + "description": "Your current location.", + "timestamps": { + "start": "00:07:04", + "end": "00:07:04" + } + } + ], + "question_timestamps": "00:07:04", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "P33_101_10", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_7_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I walk from where I am now to the refrigerator?", + "answer": "You turn right and then walk straight. The refrigerator is right next to you on your right.", + "question_chinese": "我如何从当前位置走到冰箱？", + "answer_chinese": "你向右转再直走。冰箱就在你的右边。", + "evidence": [ + { + "description": "Turn right again and then go straight.", + "timestamps": { + "start": "00:10:47", + "end": "00:10:49" + } + } + ], + "question_timestamps": "00:10:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then walk straight.", + "B": "Go straight without turning.", + "C": "You turn right and walk straight, but the refrigerator ends up directly in front of you.", + "D": "You turn right and then walk straight. The refrigerator is right next to you on your right.", + "E": "You turn right and walk straight, then the refrigerator is right beside you on your left." + }, + "answer_label": "D", + "answer_text": "You turn right and then walk straight. The refrigerator is right next to you on your right.", + "q_id": "P33_101_7_T2", + "correct_answer": "You turn right and then walk straight. The refrigerator is right next to you on your right.", + "original_options": { + "A": "Turn left, then walk straight.", + "B": "Go straight without turning.", + "C": "You turn right and walk straight, but the refrigerator ends up directly in front of you.", + "D": "You turn right and then walk straight. The refrigerator is right next to you on your right.", + "E": "You turn right and walk straight, then the refrigerator is right beside you on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_7_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I walk from where I am now to the refrigerator?", + "answer": "You first turn left and walk straight, then turn right and walk straight; the refrigerator is on your right-hand side.", + "question_chinese": "我如何从当前位置走到冰箱？", + "answer_chinese": "你先左转直走后再右转直走，冰箱在你的右手边。", + "evidence": [ + { + "description": "Turn left first and then go straight.", + "timestamps": { + "start": "00:10:44", + "end": "00:10:47" + } + }, + { + "description": "Turn right again and then go straight.", + "timestamps": { + "start": "00:10:48", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:10:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You first turn left and walk straight, then turn right and walk straight; the refrigerator is on your right-hand side.", + "B": "Turn right, then go straight.", + "C": "Turn left and stop.", + "D": "You first turn left and walk straight, then turn right and walk straight; the refrigerator is on your left-hand side.", + "E": "You first turn left and walk straight, then turn right and walk straight; the refrigerator ends up behind you." + }, + "answer_label": "A", + "answer_text": "You first turn left and walk straight, then turn right and walk straight; the refrigerator is on your right-hand side.", + "q_id": "P33_101_7_T1", + "correct_answer": "You first turn left and walk straight, then turn right and walk straight; the refrigerator is on your right-hand side.", + "original_options": { + "A": "You first turn left and walk straight, then turn right and walk straight; the refrigerator is on your right-hand side.", + "B": "Turn right, then go straight.", + "C": "Turn left and stop.", + "D": "You first turn left and walk straight, then turn right and walk straight; the refrigerator is on your left-hand side.", + "E": "You first turn left and walk straight, then turn right and walk straight; the refrigerator ends up behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 20, + 19 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_13", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get from my current location to the refrigerator?", + "answer": "Turn left, go straight, then turn right and go straight. The refrigerator is on your right.", + "question_chinese": "我如何从当前位置走到冰箱", + "answer_chinese": "左转直走后再右转直走，冰箱在你的右手边", + "evidence": [ + { + "description": "Turn left first and then go straight.", + "timestamps": { + "start": "00:10:48", + "end": "00:10:46" + } + }, + { + "description": "Turn right again and then go straight.", + "timestamps": { + "start": "00:10:46", + "end": "00:10:50" + } + }, + { + "description": "The refrigerator is on your right.", + "timestamps": { + "start": "00:10:50", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:11:23", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, go straight, then turn right and go straight. The refrigerator is on your right.", + "B": "Turn right, then go straight.", + "C": "Go straight the whole way.", + "D": "Turn left, go straight, then turn right and go straight; the refrigerator is directly in front of you.", + "E": "Turn left, go straight, then turn right and go straight; the refrigerator is on your left." + }, + "answer_label": "A", + "answer_text": "Turn left, go straight, then turn right and go straight. The refrigerator is on your right.", + "q_id": "P33_101_13", + "correct_answer": "Turn left, go straight, then turn right and go straight. The refrigerator is on your right.", + "original_options": { + "A": "Turn left, go straight, then turn right and go straight. The refrigerator is on your right.", + "B": "Turn right, then go straight.", + "C": "Go straight the whole way.", + "D": "Turn left, go straight, then turn right and go straight; the refrigerator is directly in front of you.", + "E": "Turn left, go straight, then turn right and go straight; the refrigerator is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 16 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_14", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get from my current location to the refrigerator?", + "answer": "Turn left, go straight, then turn right and go straight. The refrigerator is on your right.", + "question_chinese": "我如何从当前位置走到冰箱", + "answer_chinese": "向左转，直走，然后向右转再直走。冰箱就在你的右边。", + "evidence": [ + { + "description": "Turn left first and then go straight.", + "timestamps": { + "start": "00:10:48", + "end": "00:10:46" + } + }, + { + "description": "Turn right again and then go straight.", + "timestamps": { + "start": "00:10:46", + "end": "00:10:50" + } + }, + { + "description": "The refrigerator is on your right.", + "timestamps": { + "start": "00:10:50", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:14:09", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight.", + "B": "Turn left, go straight, then turn right and go straight. The refrigerator is on your right.", + "C": "Walk straight, then turn right.", + "D": "Walk straight ahead, turn right at the corner, and the refrigerator is on your left.", + "E": "Go straight, turn left at the dining table, then walk straight; the refrigerator ends up behind you." + }, + "answer_label": "B", + "answer_text": "Turn left, go straight, then turn right and go straight. The refrigerator is on your right.", + "q_id": "P33_101_14", + "correct_answer": "Turn left, go straight, then turn right and go straight. The refrigerator is on your right.", + "original_options": { + "A": "Turn right, then go straight.", + "B": "Turn left, go straight, then turn right and go straight. The refrigerator is on your right.", + "C": "Walk straight, then turn right.", + "D": "Walk straight ahead, turn right at the corner, and the refrigerator is on your left.", + "E": "Go straight, turn left at the dining table, then walk straight; the refrigerator ends up behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_8", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the trash can?", + "answer": "You can turn right, and the trash can is right in front of you.", + "question_chinese": "我如何走到垃圾桶旁边", + "answer_chinese": "你可以向右转，垃圾桶就在你的正前方", + "evidence": [ + { + "description": "You can turn right, and the trash can is right in front of you.", + "timestamps": { + "start": "00:09:55", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:10:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You can turn right, and the trash can is right in front of you.", + "B": "Turn left; it's in front of you.", + "C": "Go straight; it's on your right.", + "D": "Turn right, then walk forward a bit until the trash can appears on your left.", + "E": "Turn left at the corner and continue straight; the trash can will be behind you." + }, + "answer_label": "A", + "answer_text": "You can turn right, and the trash can is right in front of you.", + "q_id": "P33_101_8", + "correct_answer": "You can turn right, and the trash can is right in front of you.", + "original_options": { + "A": "You can turn right, and the trash can is right in front of you.", + "B": "Turn left; it's in front of you.", + "C": "Go straight; it's on your right.", + "D": "Turn right, then walk forward a bit until the trash can appears on your left.", + "E": "Turn left at the corner and continue straight; the trash can will be behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 15 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_7", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the trash can?", + "answer": "You can first turn left, then go straight, then turn right, then go straight again, and finally turn right.", + "question_chinese": "我如何走到垃圾桶旁边", + "answer_chinese": "你可以先左转在直走再右转再直走最后向右转", + "evidence": [ + { + "description": "You can turn left.", + "timestamps": { + "start": "00:11:30", + "end": "00:11:37" + } + }, + { + "description": "You can go straight.", + "timestamps": { + "start": "00:11:40", + "end": "00:11:49" + } + }, + { + "description": "Then you turn right.", + "timestamps": { + "start": "00:11:50", + "end": "00:11:51" + } + }, + { + "description": "Then you go straight ahead.", + "timestamps": { + "start": "00:11:52", + "end": "00:11:54" + } + }, + { + "description": "Finally, you turn right.", + "timestamps": { + "start": "00:12:23", + "end": "00:12:27" + } + } + ], + "question_timestamps": "00:12:27", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight, then left.", + "B": "Turn left, go straight, then left again.", + "C": "You can first turn left, then go straight, then turn right, then go straight again, and finally turn right.", + "D": "First turn left, go straight, turn right, then turn left, and finally go straight.", + "E": "First turn right, go straight, turn right, go straight again, and finish with a left turn." + }, + "answer_label": "C", + "answer_text": "You can first turn left, then go straight, then turn right, then go straight again, and finally turn right.", + "q_id": "P33_101_7", + "correct_answer": "You can first turn left, then go straight, then turn right, then go straight again, and finally turn right.", + "original_options": { + "A": "Turn right, then go straight, then left.", + "B": "Turn left, go straight, then left again.", + "C": "You can first turn left, then go straight, then turn right, then go straight again, and finally turn right.", + "D": "First turn left, go straight, turn right, then turn left, and finally go straight.", + "E": "First turn right, go straight, turn right, go straight again, and finish with a left turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 16 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_3", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the drawer?", + "answer": "Six times.", + "question_chinese": "我打开了多少次抽屉", + "answer_chinese": "六次", + "evidence": [ + { + "description": "Three times.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:25" + } + }, + { + "description": "Three times.", + "timestamps": { + "start": "00:04:36", + "end": "00:04:53" + } + } + ], + "question_timestamps": "00:07:06", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Six times.", + "B": "Five times.", + "C": "Seven times.", + "D": "You opened the drawer four times during the entire clip, based on each clear open-close cycle.", + "E": "You opened the drawer nine times across the video, counting every distinct pull that fully opened it." + }, + "answer_label": "A", + "answer_text": "Six times.", + "q_id": "P33_101_3", + "correct_answer": "Six times.", + "original_options": { + "A": "Six times.", + "B": "Five times.", + "C": "Seven times.", + "D": "You opened the drawer four times during the entire clip, based on each clear open-close cycle.", + "E": "You opened the drawer nine times across the video, counting every distinct pull that fully opened it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_4", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the drawer?", + "answer": "Seven times.", + "question_chinese": "我打开了多少次抽屉", + "answer_chinese": "七次", + "evidence": [ + { + "description": "Three times.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:25" + } + }, + { + "description": "Three times.", + "timestamps": { + "start": "00:04:36", + "end": "00:04:53" + } + }, + { + "description": "One time.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:18" + } + } + ], + "question_timestamps": "00:11:30", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Six times.", + "B": "Eight times.", + "C": "You opened the drawer four times during the entire clip, based on each clear open-close cycle.", + "D": "You opened the drawer nine times across the video, counting every distinct pull that fully opened it.", + "E": "Seven times." + }, + "answer_label": "E", + "answer_text": "Seven times.", + "q_id": "P33_101_4", + "correct_answer": "Seven times.", + "original_options": { + "A": "Six times.", + "B": "Eight times.", + "C": "You opened the drawer four times during the entire clip, based on each clear open-close cycle.", + "D": "You opened the drawer nine times across the video, counting every distinct pull that fully opened it.", + "E": "Seven times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_5", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I do when I opened the drawer?", + "answer": "You took a shovel.", + "question_chinese": "我打开抽屉做了什么", + "answer_chinese": "你拿了一把铲子", + "evidence": [ + { + "description": "You took a shovel.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:26" + } + } + ], + "question_timestamps": "00:11:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took a bucket.", + "B": "You grabbed a broom.", + "C": "When you opened the drawer, you pulled out a rake instead of taking a shovel.", + "D": "After opening the drawer, you picked up a mop and removed it from the drawer.", + "E": "You took a shovel." + }, + "answer_label": "E", + "answer_text": "You took a shovel.", + "q_id": "P33_101_5", + "correct_answer": "You took a shovel.", + "original_options": { + "A": "You took a bucket.", + "B": "You grabbed a broom.", + "C": "When you opened the drawer, you pulled out a rake instead of taking a shovel.", + "D": "After opening the drawer, you picked up a mop and removed it from the drawer.", + "E": "You took a shovel." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_6", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I do when I opened the drawer?", + "answer": "You took a shovel first and then a spoon.", + "question_chinese": "我打开抽屉做了什么", + "answer_chinese": "你先拿了铲子又拿了勺子", + "evidence": [ + { + "description": "You took a shovel.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:26" + } + }, + { + "description": "You took a spoon.", + "timestamps": { + "start": "00:12:40", + "end": "00:12:52" + } + } + ], + "question_timestamps": "00:12:52", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took a spoon first, then a shovel.", + "B": "You took a shovel, then a knife.", + "C": "You took a shovel first and then picked up a fork from the drawer.", + "D": "You took a shovel first and then a spoon.", + "E": "You grabbed a shovel first, and afterward you took a cup from the same drawer." + }, + "answer_label": "D", + "answer_text": "You took a shovel first and then a spoon.", + "q_id": "P33_101_6", + "correct_answer": "You took a shovel first and then a spoon.", + "original_options": { + "A": "You took a spoon first, then a shovel.", + "B": "You took a shovel, then a knife.", + "C": "You took a shovel first and then picked up a fork from the drawer.", + "D": "You took a shovel first and then a spoon.", + "E": "You grabbed a shovel first, and afterward you took a cup from the same drawer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 15 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_11", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I clean the table?", + "answer": "You haven't cleaned the table.", + "question_chinese": "我是否擦过桌子", + "answer_chinese": "你没有擦过桌子", + "evidence": [ + { + "description": "You haven't cleaned the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:03:18" + } + } + ], + "question_timestamps": "00:03:18", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You haven't cleaned the table.", + "B": "You have cleaned the table." + }, + "answer_label": "A", + "answer_text": "You haven't cleaned the table.", + "q_id": "P33_101_11", + "correct_answer": "You haven't cleaned the table.", + "original_options": { + "A": "You haven't cleaned the table.", + "B": "You have cleaned the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_12", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I clean the table?", + "answer": "You cleaned the table.", + "question_chinese": "我是否擦过桌子", + "answer_chinese": "你擦过桌子", + "evidence": [ + { + "description": "The first time.", + "timestamps": { + "start": "00:08:58", + "end": "00:09:13" + } + }, + { + "description": "The second time.", + "timestamps": { + "start": "00:09:19", + "end": "00:09:24" + } + } + ], + "question_timestamps": "00:10:31", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You cleaned the table.", + "B": "You did not clean the table." + }, + "answer_label": "A", + "answer_text": "You cleaned the table.", + "q_id": "P33_101_12", + "correct_answer": "You cleaned the table.", + "original_options": { + "A": "You cleaned the table.", + "B": "You did not clean the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_1_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Can I touch the faucet?", + "answer": "You cannot touch the faucet; it is in front of you.", + "question_chinese": "我能碰到水龙头吗？", + "answer_chinese": "你不能碰到水龙头，它在你前面。", + "evidence": [ + { + "description": "You can not.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You cannot touch the faucet; it is in front of you.", + "B": "You can touch the faucet.", + "C": "It’s behind you and reachable.", + "D": "You can touch the faucet, and it is positioned directly in front of you.", + "E": "You can’t touch the faucet because it’s off to your left, not in front." + }, + "answer_label": "A", + "answer_text": "You cannot touch the faucet; it is in front of you.", + "q_id": "P33_101_1_T1", + "correct_answer": "You cannot touch the faucet; it is in front of you.", + "original_options": { + "A": "You cannot touch the faucet; it is in front of you.", + "B": "You can touch the faucet.", + "C": "It’s behind you and reachable.", + "D": "You can touch the faucet, and it is positioned directly in front of you.", + "E": "You can’t touch the faucet because it’s off to your left, not in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 16 + ] + }, + { + "video_uid": "P33_101", + "video_length": "00:14:14", + "qid": "P33_101_1_T2", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Can I touch the faucet?", + "answer": "You can touch the faucet; it is in front of you.", + "question_chinese": "我能碰到水龙头吗？", + "answer_chinese": "你能碰到水龙头，它在你前面。", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You cannot touch the faucet.", + "B": "You can touch the faucet; it is in front of you.", + "C": "It’s behind you and unreachable.", + "D": "You can’t touch the faucet even though it is directly in front of you.", + "E": "You can touch the faucet, but it is to your right rather than in front of you." + }, + "answer_label": "B", + "answer_text": "You can touch the faucet; it is in front of you.", + "q_id": "P33_101_1_T2", + "correct_answer": "You can touch the faucet; it is in front of you.", + "original_options": { + "A": "You cannot touch the faucet.", + "B": "You can touch the faucet; it is in front of you.", + "C": "It’s behind you and unreachable.", + "D": "You can’t touch the faucet even though it is directly in front of you.", + "E": "You can touch the faucet, but it is to your right rather than in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 17 + ] + } +] diff --git a/QAs/P33_103-mcq.json b/QAs/P33_103-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..48085f46f77ac606d8fa41f221cacd1f660d6700 --- /dev/null +++ b/QAs/P33_103-mcq.json @@ -0,0 +1,1020 @@ +[ + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_5_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kinds of seasonings did I add?", + "answer": "You added three kinds.", + "question_chinese": "我加入了多少种调料？", + "answer_chinese": "你加了三种", + "evidence": [ + { + "description": "You added a second type.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:13" + } + }, + { + "description": "You added the first kind.", + "timestamps": { + "start": "00:03:03", + "end": "00:03:08" + } + }, + { + "description": "You added a third kind.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:50" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You added one kind.", + "B": "You added three kinds.", + "C": "You added two kinds.", + "D": "You added four kinds of seasonings in total during the entire preparation process.", + "E": "You added five kinds of seasonings altogether, counting each different seasoning you used." + }, + "answer_label": "B", + "answer_text": "You added three kinds.", + "q_id": "P33_103_5_T1", + "correct_answer": "You added three kinds.", + "original_options": { + "A": "You added one kind.", + "B": "You added three kinds.", + "C": "You added two kinds.", + "D": "You added four kinds of seasonings in total during the entire preparation process.", + "E": "You added five kinds of seasonings altogether, counting each different seasoning you used." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 13 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_5_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kinds of seasonings did I add?", + "answer": "You added three kinds.", + "question_chinese": "我加入了多少种调料？", + "answer_chinese": "你加入了三种", + "evidence": [ + { + "description": "You added a second type.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:13" + } + }, + { + "description": "You added the first kind.", + "timestamps": { + "start": "00:03:03", + "end": "00:03:08" + } + }, + { + "description": "You added a third kind.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:50" + } + } + ], + "question_timestamps": "00:05:34", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You added one kind.", + "B": "You added four kinds.", + "C": "You added three kinds.", + "D": "You added two kinds of seasonings overall during the full cooking process.", + "E": "You added five kinds of seasonings in total across the steps you completed." + }, + "answer_label": "C", + "answer_text": "You added three kinds.", + "q_id": "P33_103_5_T2", + "correct_answer": "You added three kinds.", + "original_options": { + "A": "You added one kind.", + "B": "You added four kinds.", + "C": "You added three kinds.", + "D": "You added two kinds of seasonings overall during the full cooking process.", + "E": "You added five kinds of seasonings in total across the steps you completed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times did I add water to the pot?", + "answer": "You added it zero times.", + "question_chinese": "我向锅中加了多少次水", + "answer_chinese": "你加了0次", + "evidence": [ + { + "description": "You added it 0 times.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You added it one time.", + "B": "You added it three times.", + "C": "You added water five times to the pot over the course of cooking.", + "D": "You added it zero times.", + "E": "You added water seven times, topping the pot off repeatedly during preparation." + }, + "answer_label": "D", + "answer_text": "You added it zero times.", + "q_id": "P33_103_5", + "correct_answer": "You added it zero times.", + "original_options": { + "A": "You added it one time.", + "B": "You added it three times.", + "C": "You added water five times to the pot over the course of cooking.", + "D": "You added it zero times.", + "E": "You added water seven times, topping the pot off repeatedly during preparation." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 12 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times did I add water to the pot?", + "answer": "You added water four times.", + "question_chinese": "我向锅中加了多少次水", + "answer_chinese": "你加了四次水", + "evidence": [ + { + "description": "You added it 0 times.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:20" + } + }, + { + "description": "You added water four times.", + "timestamps": { + "start": "00:01:40", + "end": "00:02:01" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You added water four times.", + "B": "You added water once.", + "C": "You added water three times.", + "D": "You added water five times while cooking, adding another pour after the earlier additions.", + "E": "You added water seven times in total, repeatedly refilling the pot throughout the process." + }, + "answer_label": "A", + "answer_text": "You added water four times.", + "q_id": "P33_103_6", + "correct_answer": "You added water four times.", + "original_options": { + "A": "You added water four times.", + "B": "You added water once.", + "C": "You added water three times.", + "D": "You added water five times while cooking, adding another pour after the earlier additions.", + "E": "You added water seven times in total, repeatedly refilling the pot throughout the process." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 14 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_15", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the trash can?", + "answer": "You can go left.", + "question_chinese": "我如何到达垃圾桶旁", + "answer_chinese": "你可以向左走", + "evidence": [ + { + "description": "You can go left.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You can go left.", + "B": "You can go right.", + "C": "You can go straight ahead.", + "D": "You can turn around and head back to reach the trash can.", + "E": "You can move diagonally forward-right to get to the trash can from where you are." + }, + "answer_label": "A", + "answer_text": "You can go left.", + "q_id": "P33_103_15", + "correct_answer": "You can go left.", + "original_options": { + "A": "You can go left.", + "B": "You can go right.", + "C": "You can go straight ahead.", + "D": "You can turn around and head back to reach the trash can.", + "E": "You can move diagonally forward-right to get to the trash can from where you are." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 12, + 15 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_16", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the trash can?", + "answer": "You can go straight to the right.", + "question_chinese": "我如何到达垃圾桶旁", + "answer_chinese": "你可以向右直走", + "evidence": [ + { + "description": "You can go left.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:36" + } + }, + { + "description": "You can go straight to the right.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:05:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You can go straight to the left.", + "B": "You can turn around and head back.", + "C": "You can go straight to the right.", + "D": "You can go straight ahead without turning to reach the trash can.", + "E": "You can take a longer route by going left first, then looping around to the trash can." + }, + "answer_label": "C", + "answer_text": "You can go straight to the right.", + "q_id": "P33_103_16", + "correct_answer": "You can go straight to the right.", + "original_options": { + "A": "You can go straight to the left.", + "B": "You can turn around and head back.", + "C": "You can go straight to the right.", + "D": "You can go straight ahead without turning to reach the trash can.", + "E": "You can take a longer route by going left first, then looping around to the trash can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 17 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_3", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the stove?", + "answer": "You can go straight first, then turn left, and then go straight.", + "question_chinese": "我如何走到灶台旁", + "answer_chinese": "你可以先直走再左拐然后直走", + "evidence": [ + { + "description": "Turn left first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + }, + { + "description": "You go straight ahead again.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:08" + } + }, + { + "description": "Finally, you go straight.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn right.", + "B": "Turn left first, then go straight.", + "C": "You can go straight first, then turn left, and then go straight.", + "D": "Go straight first, then turn right, and continue straight until you reach it.", + "E": "Turn left, walk straight for a bit, then make another left to get to the stove." + }, + "answer_label": "C", + "answer_text": "You can go straight first, then turn left, and then go straight.", + "q_id": "P33_103_3", + "correct_answer": "You can go straight first, then turn left, and then go straight.", + "original_options": { + "A": "Go straight, then turn right.", + "B": "Turn left first, then go straight.", + "C": "You can go straight first, then turn left, and then go straight.", + "D": "Go straight first, then turn right, and continue straight until you reach it.", + "E": "Turn left, walk straight for a bit, then make another left to get to the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 16 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_4", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the stove?", + "answer": "You can turn right, then go straight, and then turn left.", + "question_chinese": "我如何走到灶台旁", + "answer_chinese": "你可以向右转然后直走再左转", + "evidence": [ + { + "description": "You turn right first.", + "timestamps": { + "start": "00:02:32", + "end": "00:02:34" + } + }, + { + "description": "Then you go straight ahead.", + "timestamps": { + "start": "00:02:34", + "end": "00:02:35" + } + }, + { + "description": "Finally, you go straight.", + "timestamps": { + "start": "00:02:35", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go straight.", + "B": "Go straight, then turn right.", + "C": "Turn right, go straight for a short distance, then turn right again instead of left.", + "D": "Turn left first, continue straight, and then turn left at the end to reach the stove.", + "E": "You can turn right, then go straight, and then turn left." + }, + "answer_label": "E", + "answer_text": "You can turn right, then go straight, and then turn left.", + "q_id": "P33_103_4", + "correct_answer": "You can turn right, then go straight, and then turn left.", + "original_options": { + "A": "Turn left, then go straight.", + "B": "Go straight, then turn right.", + "C": "Turn right, go straight for a short distance, then turn right again instead of left.", + "D": "Turn left first, continue straight, and then turn left at the end to reach the stove.", + "E": "You can turn right, then go straight, and then turn left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_7", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I opened the cabinet?", + "answer": "You opened it once.", + "question_chinese": "我打开了多少次柜子", + "answer_chinese": "你打开了一次", + "evidence": [ + { + "description": "You opened it once.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You opened it twice.", + "B": "You never opened it.", + "C": "You opened it once.", + "D": "You opened the cabinet three times during the task, closing it in between each time.", + "E": "You opened the cabinet four separate times, each time returning to it after doing something else." + }, + "answer_label": "C", + "answer_text": "You opened it once.", + "q_id": "P33_103_7", + "correct_answer": "You opened it once.", + "original_options": { + "A": "You opened it twice.", + "B": "You never opened it.", + "C": "You opened it once.", + "D": "You opened the cabinet three times during the task, closing it in between each time.", + "E": "You opened the cabinet four separate times, each time returning to it after doing something else." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_4_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the cabinet?", + "answer": "You opened it once.", + "question_chinese": "我打开了多少次柜子？", + "answer_chinese": "你打开了一次", + "evidence": [ + { + "description": "You opened it once.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened it twice.", + "B": "You opened it once.", + "C": "You didn’t open it.", + "D": "You opened the cabinet three different times, closing it again after each opening.", + "E": "You opened the cabinet four times throughout the sequence, returning to it repeatedly." + }, + "answer_label": "B", + "answer_text": "You opened it once.", + "q_id": "P33_103_4_T2", + "correct_answer": "You opened it once.", + "original_options": { + "A": "You opened it twice.", + "B": "You opened it once.", + "C": "You didn’t open it.", + "D": "You opened the cabinet three different times, closing it again after each opening.", + "E": "You opened the cabinet four times throughout the sequence, returning to it repeatedly." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 13 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_4_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the cabinet?", + "answer": "You opened the cabinet twice.", + "question_chinese": "我打开了多少次柜子？", + "answer_chinese": "你打开了2次柜子。", + "evidence": [ + { + "description": "You opened it first.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:55" + } + }, + { + "description": "you opened it second.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:25" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You opened the cabinet twice.", + "B": "You opened it once.", + "C": "You didn’t open it.", + "D": "You opened the cabinet three times, with a close after each time you opened it.", + "E": "You opened the cabinet four separate times, revisiting it multiple times during the activity." + }, + "answer_label": "A", + "answer_text": "You opened the cabinet twice.", + "q_id": "P33_103_4_T1", + "correct_answer": "You opened the cabinet twice.", + "original_options": { + "A": "You opened the cabinet twice.", + "B": "You opened it once.", + "C": "You didn’t open it.", + "D": "You opened the cabinet three times, with a close after each time you opened it.", + "E": "You opened the cabinet four separate times, revisiting it multiple times during the activity." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 14 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_10", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do when I opened the cabinet?", + "answer": "You took a pot lid.", + "question_chinese": "我打开柜子做了什么", + "answer_chinese": "你拿了一个锅盖", + "evidence": [ + { + "description": "You took a pot lid.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:30" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took a frying pan.", + "B": "You took a pot lid.", + "C": "You took a cutting board.", + "D": "You opened the cabinet and pulled out a baking dish before closing it again.", + "E": "You opened the cabinet and removed a soup pot, then carried it away with you." + }, + "answer_label": "B", + "answer_text": "You took a pot lid.", + "q_id": "P33_103_10", + "correct_answer": "You took a pot lid.", + "original_options": { + "A": "You took a frying pan.", + "B": "You took a pot lid.", + "C": "You took a cutting board.", + "D": "You opened the cabinet and pulled out a baking dish before closing it again.", + "E": "You opened the cabinet and removed a soup pot, then carried it away with you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_12", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do when I opened the cabinet?", + "answer": "You took a bottle of seasoning, used it up, and then put it back.", + "question_chinese": "我打开柜子做了什么", + "answer_chinese": "你拿了一瓶调料用完后放了回去", + "evidence": [ + { + "description": "You took a bottle of seasoning.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:42" + } + }, + { + "description": "You put it back.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:22" + } + } + ], + "question_timestamps": "00:05:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You left the seasoning out.", + "B": "You took a bottle of seasoning, used it up, and then put it back.", + "C": "You put the seasoning away unused.", + "D": "You took a bottle of seasoning, used some, and then left it on the counter instead of returning it.", + "E": "You grabbed a bottle of seasoning, didn’t use it, and placed it back into the cabinet right away." + }, + "answer_label": "B", + "answer_text": "You took a bottle of seasoning, used it up, and then put it back.", + "q_id": "P33_103_12", + "correct_answer": "You took a bottle of seasoning, used it up, and then put it back.", + "original_options": { + "A": "You left the seasoning out.", + "B": "You took a bottle of seasoning, used it up, and then put it back.", + "C": "You put the seasoning away unused.", + "D": "You took a bottle of seasoning, used some, and then left it on the counter instead of returning it.", + "E": "You grabbed a bottle of seasoning, didn’t use it, and placed it back into the cabinet right away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 19 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_7_T1", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the dishwashing liquid?", + "answer": "You can’t reach the dishwashing liquid. It’s to your front-left, about 2 meters away from you.", + "question_chinese": "我能够到洗洁精吗？", + "answer_chinese": "你不可以够到洗洁精，洗洁精在你左前方，距离你大约2米的位置。", + "evidence": [ + { + "description": "You can not.", + "timestamps": { + "start": "00:03:41", + "end": "00:03:46" + } + } + ], + "question_timestamps": "00:03:48", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the dishwashing liquid. It’s to your front-left, about 2 meters away from you.", + "B": "You can’t reach the dishwashing liquid. It’s to your front-left, about 2 meters away from you." + }, + "answer_label": "B", + "answer_text": "You can’t reach the dishwashing liquid. It’s to your front-left, about 2 meters away from you.", + "q_id": "P33_103_7_T1", + "correct_answer": "You can’t reach the dishwashing liquid. It’s to your front-left, about 2 meters away from you.", + "original_options": { + "A": "You can reach the dishwashing liquid. It’s to your front-left, about 2 meters away from you.", + "B": "You can’t reach the dishwashing liquid. It’s to your front-left, about 2 meters away from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_13", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the detergent?", + "answer": "You can.", + "question_chinese": "我能够到洗洁精吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You cannot." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "P33_103_13", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_7_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the dishwashing liquid?", + "answer": "You can.", + "question_chinese": "我能够到洗洁精吗？", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:41" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:30" + } + } + ], + "question_timestamps": "00:05:34", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You cannot reach the dishwashing liquid; it's to your front-left, about 2 meters away from you." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "P33_103_7_T2", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You cannot reach the dishwashing liquid; it's to your front-left, about 2 meters away from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_1_T1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I entered through the door, what was the first thing I did?", + "answer": "You picked up a plastic bowl with food in it, and the plastic bowl was in front of you.", + "question_chinese": "我进门时第一件事做了什么？", + "answer_chinese": "你拿起了一个装着食物的塑料碗，塑料碗在你正前方。", + "evidence": [ + { + "description": "You picked up a pot of uncooked food.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked up a bowl behind you.", + "B": "You picked up a plastic bowl with food in it, and the plastic bowl was in front of you.", + "C": "You picked up a bowl to your left.", + "D": "You picked up a plastic bowl with food in it, but it was behind you as you entered.", + "E": "You picked up a plastic bowl with food in it, and it was positioned to your right." + }, + "answer_label": "B", + "answer_text": "You picked up a plastic bowl with food in it, and the plastic bowl was in front of you.", + "q_id": "P33_103_1_T1", + "correct_answer": "You picked up a plastic bowl with food in it, and the plastic bowl was in front of you.", + "original_options": { + "A": "You picked up a bowl behind you.", + "B": "You picked up a plastic bowl with food in it, and the plastic bowl was in front of you.", + "C": "You picked up a bowl to your left.", + "D": "You picked up a plastic bowl with food in it, but it was behind you as you entered.", + "E": "You picked up a plastic bowl with food in it, and it was positioned to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 17 + ] + }, + { + "video_uid": "P33_103", + "video_length": "00:05:34", + "qid": "P33_103_1_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I entered through the door, what was the first thing I did?", + "answer": "You picked up a basin of uncooked food.", + "question_chinese": "我进门时第一件事做了什么？", + "answer_chinese": "你拿起了一盆未烹饪的食物", + "evidence": [ + { + "description": "You picked up a pot of uncooked food.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked up a plastic bowl.", + "B": "You moved forward without picking anything up.", + "C": "You picked up a plastic bowl with food in it, not a basin of uncooked food.", + "D": "You started by walking straight ahead, then turning left, instead of picking up any food container.", + "E": "You picked up a basin of uncooked food." + }, + "answer_label": "E", + "answer_text": "You picked up a basin of uncooked food.", + "q_id": "P33_103_1_T2", + "correct_answer": "You picked up a basin of uncooked food.", + "original_options": { + "A": "You picked up a plastic bowl.", + "B": "You moved forward without picking anything up.", + "C": "You picked up a plastic bowl with food in it, not a basin of uncooked food.", + "D": "You started by walking straight ahead, then turning left, instead of picking up any food container.", + "E": "You picked up a basin of uncooked food." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 16 + ] + } +] diff --git a/QAs/P33_108-mcq.json b/QAs/P33_108-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..116780eb2b2f8fca60da168ad53209585e9a2c5d --- /dev/null +++ b/QAs/P33_108-mcq.json @@ -0,0 +1,736 @@ +[ + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_11", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "There are several black sofas in total.", + "answer": "There are two in total.", + "question_chinese": "一共有几个黑色沙发。", + "answer_chinese": "一共有两个。", + "evidence": [ + { + "description": "There are two in total.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:51" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two in total.", + "B": "There is one in total.", + "C": "There are three in total.", + "D": "There are four black sofas in total, positioned around the living room seating area.", + "E": "There are five black sofas in total, with several arranged together along the same seating area." + }, + "answer_label": "A", + "answer_text": "There are two in total.", + "q_id": "P33_108_11", + "correct_answer": "There are two in total.", + "original_options": { + "A": "There are two in total.", + "B": "There is one in total.", + "C": "There are three in total.", + "D": "There are four black sofas in total, positioned around the living room seating area.", + "E": "There are five black sofas in total, with several arranged together along the same seating area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_12", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "There are several black sofas in total.", + "answer": "There are two in total.", + "question_chinese": "一共有几个黑色沙发。", + "answer_chinese": "一共有两个。", + "evidence": [ + { + "description": "There are two in total.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:51" + } + } + ], + "question_timestamps": "00:03:03", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one in total.", + "B": "There are three in total.", + "C": "There are two in total.", + "D": "There are four black sofas in total, positioned around the living room seating area.", + "E": "There are five black sofas in total, with several arranged together along the same seating area." + }, + "answer_label": "C", + "answer_text": "There are two in total.", + "q_id": "P33_108_12", + "correct_answer": "There are two in total.", + "original_options": { + "A": "There is one in total.", + "B": "There are three in total.", + "C": "There are two in total.", + "D": "There are four black sofas in total, positioned around the living room seating area.", + "E": "There are five black sofas in total, with several arranged together along the same seating area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_4", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was my purpose in taking the cup just now?", + "answer": "You are trying to fill the cup with water.", + "question_chinese": "我刚才拿杯子的目的是干什么？", + "answer_chinese": "你是为了往杯子里面装水。", + "evidence": [ + { + "description": "You washed the cup.", + "timestamps": { + "start": "00:03:31", + "end": "00:03:36" + } + }, + { + "description": "You filled the cup with water.", + "timestamps": { + "start": "00:03:36", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You are trying to fill it with tea.", + "B": "You are trying to fill it with milk.", + "C": "You are trying to fill the cup with juice from the fridge instead of water.", + "D": "You are trying to fill the cup with water.", + "E": "You are trying to fill the cup with soda from the fridge, not with plain water." + }, + "answer_label": "D", + "answer_text": "You are trying to fill the cup with water.", + "q_id": "P33_108_4", + "correct_answer": "You are trying to fill the cup with water.", + "original_options": { + "A": "You are trying to fill it with tea.", + "B": "You are trying to fill it with milk.", + "C": "You are trying to fill the cup with juice from the fridge instead of water.", + "D": "You are trying to fill the cup with water.", + "E": "You are trying to fill the cup with soda from the fridge, not with plain water." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 16 + ] + }, + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_3", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was my purpose in taking the cup just now?", + "answer": "You are trying to fill the cup with water.", + "question_chinese": "我刚才拿杯子的目的是干什么？", + "answer_chinese": "你是为了往杯子里面装水。", + "evidence": [ + { + "description": "You washed the cup.", + "timestamps": { + "start": "00:03:31", + "end": "00:03:36" + } + }, + { + "description": "You filled the cup with water.", + "timestamps": { + "start": "00:03:36", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:03:54", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You are trying to fill it with coffee.", + "B": "You are moving the cup to the counter.", + "C": "You are trying to fill the cup with juice rather than filling it with water.", + "D": "You are trying to fill the cup with milk instead of using it for water.", + "E": "You are trying to fill the cup with water." + }, + "answer_label": "E", + "answer_text": "You are trying to fill the cup with water.", + "q_id": "P33_108_3", + "correct_answer": "You are trying to fill the cup with water.", + "original_options": { + "A": "You are trying to fill it with coffee.", + "B": "You are moving the cup to the counter.", + "C": "You are trying to fill the cup with juice rather than filling it with water.", + "D": "You are trying to fill the cup with milk instead of using it for water.", + "E": "You are trying to fill the cup with water." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 15 + ] + }, + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_15", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my exercise route in the kitchen?", + "answer": "You walked from the dining table to the sink, from the sink to the sofa and back, and finally you walked around the kitchen once.", + "question_chinese": "我在厨房的运动路线是什么？", + "answer_chinese": "你从餐桌走到了洗水槽，从洗水槽走到了沙发旁并返回，最后你绕厨房行走了一圈。", + "evidence": [ + { + "description": "You walked from the dining table to the sink.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:06" + } + }, + { + "description": "You walked from the sink to the sofa and then returned.", + "timestamps": { + "start": "00:01:44", + "end": "00:01:59" + } + }, + { + "description": "Finally, you walked around the kitchen.", + "timestamps": { + "start": "00:02:40", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:03:44", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You walked from the table to the fridge.", + "B": "You walked around the kitchen twice.", + "C": "You walked from the dining table to the sink, from the sink to the sofa and back, and finally you walked around the kitchen once.", + "D": "You walked from the dining table to the sink, then to the armchair and back, ending with one lap.", + "E": "You walked from the countertop to the sink, then to the sofa and back, finishing without looping the kitchen." + }, + "answer_label": "C", + "answer_text": "You walked from the dining table to the sink, from the sink to the sofa and back, and finally you walked around the kitchen once.", + "q_id": "P33_108_15", + "correct_answer": "You walked from the dining table to the sink, from the sink to the sofa and back, and finally you walked around the kitchen once.", + "original_options": { + "A": "You walked from the table to the fridge.", + "B": "You walked around the kitchen twice.", + "C": "You walked from the dining table to the sink, from the sink to the sofa and back, and finally you walked around the kitchen once.", + "D": "You walked from the dining table to the sink, then to the armchair and back, ending with one lap.", + "E": "You walked from the countertop to the sink, then to the sofa and back, finishing without looping the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 19, + 19 + ] + }, + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_16", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my exercise route in the kitchen?", + "answer": "You walked from the dining table to the sink, from the sink to the sofa and back, and finally you walked around the kitchen once.", + "question_chinese": "我在厨房的运动路线是什么？", + "answer_chinese": "你从餐桌走到了洗水槽，从洗水槽走到了沙发旁并返回，最后你绕厨房行走了一圈。", + "evidence": [ + { + "description": "You walked from the dining table to the sink.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:06" + } + }, + { + "description": "You walked from the sink to the sofa and then returned.", + "timestamps": { + "start": "00:01:44", + "end": "00:01:59" + } + }, + { + "description": "Finally, you walked around the kitchen.", + "timestamps": { + "start": "00:02:40", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:03:53", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You walked from the sink to the oven.", + "B": "You started at the refrigerator and circled twice.", + "C": "You walked from the dining table to the sink, from the sink to the sofa and back, and finally you walked around the kitchen once.", + "D": "You walked from the countertop to the sink, then went to the dining table and returned without circling the kitchen.", + "E": "You moved from the sofa to the countertop, then to the sink and back to the sofa, skipping any full loop." + }, + "answer_label": "C", + "answer_text": "You walked from the dining table to the sink, from the sink to the sofa and back, and finally you walked around the kitchen once.", + "q_id": "P33_108_16", + "correct_answer": "You walked from the dining table to the sink, from the sink to the sofa and back, and finally you walked around the kitchen once.", + "original_options": { + "A": "You walked from the sink to the oven.", + "B": "You started at the refrigerator and circled twice.", + "C": "You walked from the dining table to the sink, from the sink to the sofa and back, and finally you walked around the kitchen once.", + "D": "You walked from the countertop to the sink, then went to the dining table and returned without circling the kitchen.", + "E": "You moved from the sofa to the countertop, then to the sink and back to the sofa, skipping any full loop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 20, + 21 + ] + }, + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kinds of tableware did I wash in the sink in total?", + "answer": "You washed a total of six kinds of tableware.", + "question_chinese": "我在洗水槽一共洗了多少种餐具？", + "answer_chinese": "你一共洗了六种餐具。", + "evidence": [ + { + "description": "You washed the pot and the shovel.", + "timestamps": { + "start": "00:00:09", + "end": "00:01:03" + } + }, + { + "description": "You washed the cup.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:17" + } + }, + { + "description": "You washed the dishes and spoons.", + "timestamps": { + "start": "00:01:19", + "end": "00:02:21" + } + }, + { + "description": "You washed the basin and the spoon.", + "timestamps": { + "start": "00:01:04", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:02:55", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You washed four kinds.", + "B": "You washed a total of six kinds of tableware.", + "C": "You washed seven kinds.", + "D": "You washed a total of five kinds of tableware in the sink altogether.", + "E": "You washed a total of eight kinds of tableware in the sink in total." + }, + "answer_label": "B", + "answer_text": "You washed a total of six kinds of tableware.", + "q_id": "P33_108_1", + "correct_answer": "You washed a total of six kinds of tableware.", + "original_options": { + "A": "You washed four kinds.", + "B": "You washed a total of six kinds of tableware.", + "C": "You washed seven kinds.", + "D": "You washed a total of five kinds of tableware in the sink altogether.", + "E": "You washed a total of eight kinds of tableware in the sink in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kinds of tableware did I wash in the sink in total?", + "answer": "You washed a total of six kinds of tableware.", + "question_chinese": "我在洗水槽一共洗了多少种餐具？", + "answer_chinese": "你一共洗了六种餐具。", + "evidence": [ + { + "description": "You washed the pot and the shovel.", + "timestamps": { + "start": "00:00:09", + "end": "00:01:03" + } + }, + { + "description": "You washed the cup.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:17" + } + }, + { + "description": "You washed the dishes and spoons.", + "timestamps": { + "start": "00:01:19", + "end": "00:02:21" + } + }, + { + "description": "You washed the basin and the spoon.", + "timestamps": { + "start": "00:01:04", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:03:55", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You washed seven kinds.", + "B": "You washed ten kinds.", + "C": "You washed a total of eight kinds of tableware in the sink altogether.", + "D": "You washed a total of six kinds of tableware.", + "E": "You washed a total of nine kinds of tableware in the sink in total." + }, + "answer_label": "D", + "answer_text": "You washed a total of six kinds of tableware.", + "q_id": "P33_108_2", + "correct_answer": "You washed a total of six kinds of tableware.", + "original_options": { + "A": "You washed seven kinds.", + "B": "You washed ten kinds.", + "C": "You washed a total of eight kinds of tableware in the sink altogether.", + "D": "You washed a total of six kinds of tableware.", + "E": "You washed a total of nine kinds of tableware in the sink in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_10", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the small potted plant placed on my dining table?", + "answer": "It's a small potted chili plant.", + "question_chinese": "我的餐桌上摆放的小盆栽是什么？", + "answer_chinese": "是辣椒小盆栽。", + "evidence": [ + { + "description": "It's a small potted chili plant.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's a small potted basil plant.", + "B": "It's a small potted cactus plant.", + "C": "It's a small potted chili plant.", + "D": "It's a small potted succulent plant sitting in the middle of your dining table.", + "E": "It's a small potted fern plant placed in the center of the dining table." + }, + "answer_label": "C", + "answer_text": "It's a small potted chili plant.", + "q_id": "P33_108_10", + "correct_answer": "It's a small potted chili plant.", + "original_options": { + "A": "It's a small potted basil plant.", + "B": "It's a small potted cactus plant.", + "C": "It's a small potted chili plant.", + "D": "It's a small potted succulent plant sitting in the middle of your dining table.", + "E": "It's a small potted fern plant placed in the center of the dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_9", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the small potted plant placed on my dining table?", + "answer": "It's a small potted chili plant.", + "question_chinese": "我的餐桌上摆放的小盆栽是什么？", + "answer_chinese": "是辣椒小盆栽。", + "evidence": [ + { + "description": "It's a small potted chili plant.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:02:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's a small potted basil plant.", + "B": "It's a small potted cactus plant.", + "C": "It's a small potted fern plant placed on your dining table as a centerpiece.", + "D": "It's a small potted rose plant sitting on the dining table as decoration.", + "E": "It's a small potted chili plant." + }, + "answer_label": "E", + "answer_text": "It's a small potted chili plant.", + "q_id": "P33_108_9", + "correct_answer": "It's a small potted chili plant.", + "original_options": { + "A": "It's a small potted basil plant.", + "B": "It's a small potted cactus plant.", + "C": "It's a small potted fern plant placed on your dining table as a centerpiece.", + "D": "It's a small potted rose plant sitting on the dining table as decoration.", + "E": "It's a small potted chili plant." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_7", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which of the three bottles placed beside the sink is detergent?", + "answer": "You washed the tableware with a bottle containing yellow liquid.", + "question_chinese": "洗水槽旁边摆的三个瓶子哪个是洗涤剂？", + "answer_chinese": "你用装有黄色液体的瓶子洗了餐具。", + "evidence": [ + { + "description": "There are three bottles beside the sink.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:32" + } + }, + { + "description": "You wash the dishes with a bottle containing yellow liquid.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You used the blue liquid bottle.", + "B": "You washed the tableware with a bottle containing yellow liquid.", + "C": "You used the clear liquid bottle.", + "D": "You washed the tableware with a bottle containing orange liquid beside the sink.", + "E": "You washed the tableware with a bottle containing green liquid from the three bottles next to the sink." + }, + "answer_label": "B", + "answer_text": "You washed the tableware with a bottle containing yellow liquid.", + "q_id": "P33_108_7", + "correct_answer": "You washed the tableware with a bottle containing yellow liquid.", + "original_options": { + "A": "You used the blue liquid bottle.", + "B": "You washed the tableware with a bottle containing yellow liquid.", + "C": "You used the clear liquid bottle.", + "D": "You washed the tableware with a bottle containing orange liquid beside the sink.", + "E": "You washed the tableware with a bottle containing green liquid from the three bottles next to the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 18 + ] + }, + { + "video_uid": "P33_108", + "video_length": "00:03:55", + "qid": "P33_108_8", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which of the three bottles placed beside the sink is detergent?", + "answer": "You washed the tableware with a bottle containing yellow liquid.", + "question_chinese": "洗水槽旁边摆的三个瓶子哪个是洗涤剂？", + "answer_chinese": "你用装有黄色液体的瓶子洗了餐具。", + "evidence": [ + { + "description": "There are three bottles beside the sink.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:32" + } + }, + { + "description": "You wash the dishes with a bottle containing yellow liquid.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:03:22", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You used the blue liquid bottle.", + "B": "You used the clear liquid bottle.", + "C": "You scrubbed the tableware with a bottle containing green liquid beside the sink.", + "D": "You washed the tableware with a bottle containing yellow liquid.", + "E": "You cleaned the tableware using a bottle that dispensed yellow foam instead of yellow liquid." + }, + "answer_label": "D", + "answer_text": "You washed the tableware with a bottle containing yellow liquid.", + "q_id": "P33_108_8", + "correct_answer": "You washed the tableware with a bottle containing yellow liquid.", + "original_options": { + "A": "You used the blue liquid bottle.", + "B": "You used the clear liquid bottle.", + "C": "You scrubbed the tableware with a bottle containing green liquid beside the sink.", + "D": "You washed the tableware with a bottle containing yellow liquid.", + "E": "You cleaned the tableware using a bottle that dispensed yellow foam instead of yellow liquid." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 15 + ] + } +] diff --git a/QAs/P33_111-mcq.json b/QAs/P33_111-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c56a6f67fa637246093958f5ba67ecb688bb5fb6 --- /dev/null +++ b/QAs/P33_111-mcq.json @@ -0,0 +1,662 @@ +[ + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_15", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What items did I wash in the sink just now?", + "answer": "You washed the plastic lunch box, the shovel and the mug.", + "question_chinese": "我刚才在洗碗槽里洗了哪些物品？", + "answer_chinese": "你洗了塑料饭盒，铲子和马克杯。", + "evidence": [ + { + "description": "You washed the plastic lunch box.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:18" + } + }, + { + "description": "You washed two shovels.", + "timestamps": { + "start": "00:05:12", + "end": "00:05:30" + } + }, + { + "description": "You washed a mug.", + "timestamps": { + "start": "00:06:47", + "end": "00:06:49" + } + } + ], + "question_timestamps": "00:06:56", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You washed the lunch box and the mug.", + "B": "You washed the shovel and the mug.", + "C": "You washed the plastic lunch box, the rake, and the mug in the sink just now.", + "D": "You washed the plastic lunch box, the shovel and the mug.", + "E": "You washed the plastic lunch box, the shovel, and a glass instead of a mug at the sink." + }, + "answer_label": "D", + "answer_text": "You washed the plastic lunch box, the shovel and the mug.", + "q_id": "P33_111_15", + "correct_answer": "You washed the plastic lunch box, the shovel and the mug.", + "original_options": { + "A": "You washed the lunch box and the mug.", + "B": "You washed the shovel and the mug.", + "C": "You washed the plastic lunch box, the rake, and the mug in the sink just now.", + "D": "You washed the plastic lunch box, the shovel and the mug.", + "E": "You washed the plastic lunch box, the shovel, and a glass instead of a mug at the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 18 + ] + }, + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_16", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What items did I wash in the sink just now?", + "answer": "You washed the plastic lunch box, the shovel and the mug.", + "question_chinese": "我刚才在洗碗槽里洗了哪些物品？", + "answer_chinese": "你洗了塑料饭盒，铲子和马克杯。", + "evidence": [ + { + "description": "You washed the plastic lunch box.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:18" + } + }, + { + "description": "You washed two shovels.", + "timestamps": { + "start": "00:05:12", + "end": "00:05:30" + } + }, + { + "description": "You washed a mug.", + "timestamps": { + "start": "00:06:47", + "end": "00:06:49" + } + } + ], + "question_timestamps": "00:07:06", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You washed the lunch box and the mug.", + "B": "You washed the shovel and the mug.", + "C": "You washed the plastic lunch box, the trowel, and the mug in the sink just now.", + "D": "You washed the plastic lunch box, the shovel and the mug.", + "E": "You washed the plastic lunch box, the shovel, and a cup rather than a mug in the sink." + }, + "answer_label": "D", + "answer_text": "You washed the plastic lunch box, the shovel and the mug.", + "q_id": "P33_111_16", + "correct_answer": "You washed the plastic lunch box, the shovel and the mug.", + "original_options": { + "A": "You washed the lunch box and the mug.", + "B": "You washed the shovel and the mug.", + "C": "You washed the plastic lunch box, the trowel, and the mug in the sink just now.", + "D": "You washed the plastic lunch box, the shovel and the mug.", + "E": "You washed the plastic lunch box, the shovel, and a cup rather than a mug in the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 18 + ] + }, + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the garlic from and where did I get it?", + "answer": "You took the garlic from the table to the cabinet.", + "question_chinese": "我把大蒜从哪里拿到了哪里？", + "answer_chinese": "你把大蒜从桌子上拿到了柜子里。", + "evidence": [ + { + "description": "You picked up the garlic on the table.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:07" + } + }, + { + "description": "You took the garlic from the table to the cabinet.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:04:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From counter to cabinet.", + "B": "From table to drawer.", + "C": "You took the garlic from the table to the cabinet.", + "D": "You moved the garlic from the cabinet and set it down on the table afterward.", + "E": "You picked up the garlic from the sink area and placed it into the cabinet for storage." + }, + "answer_label": "C", + "answer_text": "You took the garlic from the table to the cabinet.", + "q_id": "P33_111_1", + "correct_answer": "You took the garlic from the table to the cabinet.", + "original_options": { + "A": "From counter to cabinet.", + "B": "From table to drawer.", + "C": "You took the garlic from the table to the cabinet.", + "D": "You moved the garlic from the cabinet and set it down on the table afterward.", + "E": "You picked up the garlic from the sink area and placed it into the cabinet for storage." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the garlic from and where did I get it?", + "answer": "You took the garlic from the table to the cabinet.", + "question_chinese": "我把大蒜从哪里拿到了哪里？", + "answer_chinese": "你把大蒜从桌子上拿到了柜子里。", + "evidence": [ + { + "description": "You picked up the garlic on the table.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:07" + } + }, + { + "description": "You took the garlic from the table to the cabinet.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:05:26", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From counter to cabinet.", + "B": "From table to drawer.", + "C": "You carried the garlic from the table and put it into the fridge instead of the cabinet.", + "D": "You took the garlic from the table to the cabinet.", + "E": "You grabbed the garlic from inside the cabinet and placed it on the table when you were done." + }, + "answer_label": "D", + "answer_text": "You took the garlic from the table to the cabinet.", + "q_id": "P33_111_2", + "correct_answer": "You took the garlic from the table to the cabinet.", + "original_options": { + "A": "From counter to cabinet.", + "B": "From table to drawer.", + "C": "You carried the garlic from the table and put it into the fridge instead of the cabinet.", + "D": "You took the garlic from the table to the cabinet.", + "E": "You grabbed the garlic from inside the cabinet and placed it on the table when you were done." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the rice in the pot?", + "answer": "You put the rice into porcelain bowls and lunch boxes.", + "question_chinese": "我把锅里的米饭装进了哪里？", + "answer_chinese": "你把米饭装进了瓷碗里和饭盒里。", + "evidence": [ + { + "description": "You put the rice into unch boxes.", + "timestamps": { + "start": "00:03:22", + "end": "00:03:24" + } + } + ], + "question_timestamps": "00:04:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put the rice into porcelain bowls and lunch boxes.", + "B": "You put the rice into plastic bowls.", + "C": "You put the rice into lunch boxes only.", + "D": "You put the rice into ceramic bowls and plastic containers instead of porcelain bowls and lunch boxes.", + "E": "You portioned the rice into glass bowls and takeout containers rather than porcelain bowls and lunch boxes." + }, + "answer_label": "A", + "answer_text": "You put the rice into porcelain bowls and lunch boxes.", + "q_id": "P33_111_3", + "correct_answer": "You put the rice into porcelain bowls and lunch boxes.", + "original_options": { + "A": "You put the rice into porcelain bowls and lunch boxes.", + "B": "You put the rice into plastic bowls.", + "C": "You put the rice into lunch boxes only.", + "D": "You put the rice into ceramic bowls and plastic containers instead of porcelain bowls and lunch boxes.", + "E": "You portioned the rice into glass bowls and takeout containers rather than porcelain bowls and lunch boxes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 17 + ] + }, + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_4", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the rice in the pot?", + "answer": "You put the rice into porcelain bowls and lunch boxes.", + "question_chinese": "我把锅里的米饭装进了哪里？", + "answer_chinese": "你把米饭装进了瓷碗里和饭盒里。", + "evidence": [ + { + "description": "You put the rice into unch boxes.", + "timestamps": { + "start": "00:03:22", + "end": "00:03:24" + } + } + ], + "question_timestamps": "00:06:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put the rice into plastic bowls.", + "B": "You put the rice into lunch boxes only.", + "C": "You divided the rice into metal bowls and storage containers rather than porcelain bowls and lunch boxes.", + "D": "You transferred the rice into paper containers and ceramic bowls instead of porcelain bowls and lunch boxes.", + "E": "You put the rice into porcelain bowls and lunch boxes." + }, + "answer_label": "E", + "answer_text": "You put the rice into porcelain bowls and lunch boxes.", + "q_id": "P33_111_4", + "correct_answer": "You put the rice into porcelain bowls and lunch boxes.", + "original_options": { + "A": "You put the rice into plastic bowls.", + "B": "You put the rice into lunch boxes only.", + "C": "You divided the rice into metal bowls and storage containers rather than porcelain bowls and lunch boxes.", + "D": "You transferred the rice into paper containers and ceramic bowls instead of porcelain bowls and lunch boxes.", + "E": "You put the rice into porcelain bowls and lunch boxes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 17 + ] + }, + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_12", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What variety of potted plant am I placing on the table?", + "answer": "It's a small potted plant of colorful peppers.", + "question_chinese": "我摆在桌子上的盆栽是什么品种？", + "answer_chinese": "是彩椒小盆栽。", + "evidence": [ + { + "description": "It's a small potted plant of colorful peppers.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:03:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's a small potted basil plant.", + "B": "It's a small potted fern plant.", + "C": "It's a small potted plant of colorful peppers.", + "D": "It's a small potted plant of leafy greens, not a pepper plant, that you're placing on the table.", + "E": "It's a small potted flowering plant with daisies rather than colorful peppers as you set it down." + }, + "answer_label": "C", + "answer_text": "It's a small potted plant of colorful peppers.", + "q_id": "P33_111_12", + "correct_answer": "It's a small potted plant of colorful peppers.", + "original_options": { + "A": "It's a small potted basil plant.", + "B": "It's a small potted fern plant.", + "C": "It's a small potted plant of colorful peppers.", + "D": "It's a small potted plant of leafy greens, not a pepper plant, that you're placing on the table.", + "E": "It's a small potted flowering plant with daisies rather than colorful peppers as you set it down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_11", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What variety of potted plant am I placing on the table?", + "answer": "It's a small potted plant of colorful peppers.", + "question_chinese": "我摆在桌子上的盆栽是什么品种？", + "answer_chinese": "是彩椒小盆栽。", + "evidence": [ + { + "description": "It's a small potted plant of colorful peppers.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:04:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's a small potted basil plant.", + "B": "It's a small potted plant of colorful peppers.", + "C": "It's a small potted fern plant.", + "D": "It's a small potted herb plant with basil leaves, not a pepper plant, sitting on the table.", + "E": "It's a small potted plant of bright sunflowers rather than colorful peppers that you're placing down." + }, + "answer_label": "B", + "answer_text": "It's a small potted plant of colorful peppers.", + "q_id": "P33_111_11", + "correct_answer": "It's a small potted plant of colorful peppers.", + "original_options": { + "A": "It's a small potted basil plant.", + "B": "It's a small potted plant of colorful peppers.", + "C": "It's a small potted fern plant.", + "D": "It's a small potted herb plant with basil leaves, not a pepper plant, sitting on the table.", + "E": "It's a small potted plant of bright sunflowers rather than colorful peppers that you're placing down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_9", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I'm currently washing dishes in the sink. I want to get a knife and fork. Can I get them directly?", + "answer": "The knife and fork are right below you. You can reach them with your hand.", + "question_chinese": "我现在在洗碗槽洗碗，我想要拿刀叉，能直接拿到吗？", + "answer_chinese": "刀叉就在你的右下方，你伸手就能拿到。", + "evidence": [ + { + "description": "The knife and fork are right below you. You can reach them with your hand.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:03:43", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The knife and fork are right below you. You can reach them with your hand.", + "B": "The knife and fork are right below you. You cannot reach them with your hand." + }, + "answer_label": "A", + "answer_text": "The knife and fork are right below you. You can reach them with your hand.", + "q_id": "P33_111_9", + "correct_answer": "The knife and fork are right below you. You can reach them with your hand.", + "original_options": { + "A": "The knife and fork are right below you. You can reach them with your hand.", + "B": "The knife and fork are right below you. You cannot reach them with your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_10", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I'm currently washing dishes in the sink. I want to get a knife and fork. Can I get them directly?", + "answer": "The knife and fork are right below you. You can reach them with your hand.", + "question_chinese": "我现在在洗碗槽洗碗，我想要拿刀叉，能直接拿到吗？", + "answer_chinese": "刀叉就在你的右下方，你伸手就能拿到。", + "evidence": [ + { + "description": "The knife and fork are right below you. You can reach them with your hand.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:05:13", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The knife and fork are right below you. You cannot reach them with your hand.", + "B": "The knife and fork are right below you. You can reach them with your hand." + }, + "answer_label": "B", + "answer_text": "The knife and fork are right below you. You can reach them with your hand.", + "q_id": "P33_111_10", + "correct_answer": "The knife and fork are right below you. You can reach them with your hand.", + "original_options": { + "A": "The knife and fork are right below you. You cannot reach them with your hand.", + "B": "The knife and fork are right below you. You can reach them with your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_13", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many coasters are placed on the table?", + "answer": "Three coasters were placed.", + "question_chinese": "桌子上摆了几个杯垫？", + "answer_chinese": "摆了三个杯垫。", + "evidence": [ + { + "description": "Three coasters were placed.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:03:22", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three coasters were placed.", + "B": "Two coasters were placed.", + "C": "Four coasters were placed.", + "D": "Five coasters were placed on the table in total, spread out near the center.", + "E": "One coaster was placed on the table, with no other coasters visible nearby." + }, + "answer_label": "A", + "answer_text": "Three coasters were placed.", + "q_id": "P33_111_13", + "correct_answer": "Three coasters were placed.", + "original_options": { + "A": "Three coasters were placed.", + "B": "Two coasters were placed.", + "C": "Four coasters were placed.", + "D": "Five coasters were placed on the table in total, spread out near the center.", + "E": "One coaster was placed on the table, with no other coasters visible nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 13 + ] + }, + { + "video_uid": "P33_111", + "video_length": "00:07:12", + "qid": "P33_111_14", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many coasters are placed on the table?", + "answer": "Three coasters were placed.", + "question_chinese": "桌子上摆了几个杯垫？", + "answer_chinese": "摆了三个杯垫。", + "evidence": [ + { + "description": "Three coasters were placed.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:06:42", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two coasters were placed.", + "B": "Four coasters were placed.", + "C": "Five coasters were placed on the table, arranged across the surface rather than grouped together.", + "D": "One coaster was placed on the table, and there weren’t any additional coasters beside it.", + "E": "Three coasters were placed." + }, + "answer_label": "E", + "answer_text": "Three coasters were placed.", + "q_id": "P33_111_14", + "correct_answer": "Three coasters were placed.", + "original_options": { + "A": "Two coasters were placed.", + "B": "Four coasters were placed.", + "C": "Five coasters were placed on the table, arranged across the surface rather than grouped together.", + "D": "One coaster was placed on the table, and there weren’t any additional coasters beside it.", + "E": "Three coasters were placed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + } +] diff --git a/QAs/P34_102-mcq.json b/QAs/P34_102-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..227e7074ea64571b5619fdba3f5725ce746fef93 --- /dev/null +++ b/QAs/P34_102-mcq.json @@ -0,0 +1,1475 @@ +[ + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_7_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the refrigerator relative to me?", + "answer": "The refrigerator was relatively to your front-left.", + "question_chinese": "冰箱相对在我的哪个方向？", + "answer_chinese": "冰箱相对在你左前方。", + "evidence": [ + { + "description": "The refrigerator is on your left front.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:35" + } + } + ], + "question_timestamps": "00:06:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Front-right.", + "B": "Back-left.", + "C": "The refrigerator was positioned to your front-right, slightly off to the side rather than straight ahead.", + "D": "The refrigerator was behind you on the left, so you would need to turn around to face it.", + "E": "The refrigerator was relatively to your front-left." + }, + "answer_label": "E", + "answer_text": "The refrigerator was relatively to your front-left.", + "q_id": "P34_102_7_T1", + "correct_answer": "The refrigerator was relatively to your front-left.", + "original_options": { + "A": "Front-right.", + "B": "Back-left.", + "C": "The refrigerator was positioned to your front-right, slightly off to the side rather than straight ahead.", + "D": "The refrigerator was behind you on the left, so you would need to turn around to face it.", + "E": "The refrigerator was relatively to your front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 18 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_7_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the refrigerator relative to me?", + "answer": "The refrigerator was directly in front of you.", + "question_chinese": "冰箱相对在我的哪个方向？", + "answer_chinese": "冰箱相对在你正前方。", + "evidence": [ + { + "description": "The refrigerator is in front of you.", + "timestamps": { + "start": "00:06:36", + "end": "00:06:37" + } + } + ], + "question_timestamps": "00:06:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly behind you.", + "B": "The refrigerator was directly in front of you.", + "C": "Directly to your left.", + "D": "The refrigerator was directly to your right, not centered in front of your view.", + "E": "The refrigerator was off to your back-left, so it wasn’t something you were facing head-on." + }, + "answer_label": "B", + "answer_text": "The refrigerator was directly in front of you.", + "q_id": "P34_102_7_T2", + "correct_answer": "The refrigerator was directly in front of you.", + "original_options": { + "A": "Directly behind you.", + "B": "The refrigerator was directly in front of you.", + "C": "Directly to your left.", + "D": "The refrigerator was directly to your right, not centered in front of your view.", + "E": "The refrigerator was off to your back-left, so it wasn’t something you were facing head-on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 16 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_6_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the garlic relative to me?", + "answer": "The garlic was relatively to your front-left.", + "question_chinese": "大蒜相对在我的哪个位置？", + "answer_chinese": "大蒜相对在你的左前方。", + "evidence": [ + { + "description": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Front-right.", + "B": "Back-left.", + "C": "The garlic was relatively to your right side, not on the front-left side of your view.", + "D": "The garlic was relatively to your front-left.", + "E": "The garlic was positioned behind you to the left, so you would have to turn to see it." + }, + "answer_label": "D", + "answer_text": "The garlic was relatively to your front-left.", + "q_id": "P34_102_6_T1", + "correct_answer": "The garlic was relatively to your front-left.", + "original_options": { + "A": "Front-right.", + "B": "Back-left.", + "C": "The garlic was relatively to your right side, not on the front-left side of your view.", + "D": "The garlic was relatively to your front-left.", + "E": "The garlic was positioned behind you to the left, so you would have to turn to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 18 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_6_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the garlic relative to me?", + "answer": "The garlic was relatively to your left.", + "question_chinese": "大蒜相对在我的哪个位置？", + "answer_chinese": "大蒜相对在你的左边。", + "evidence": [ + { + "description": "Garlic is on your left", + "timestamps": { + "start": "00:05:13", + "end": "00:05:16" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "The garlic was relatively to your left.", + "C": "To your right.", + "D": "The garlic was relatively behind you, not on your left side where you could easily see it.", + "E": "The garlic was positioned in front and slightly to your right, rather than directly to your left." + }, + "answer_label": "B", + "answer_text": "The garlic was relatively to your left.", + "q_id": "P34_102_6_T2", + "correct_answer": "The garlic was relatively to your left.", + "original_options": { + "A": "In front of you.", + "B": "The garlic was relatively to your left.", + "C": "To your right.", + "D": "The garlic was relatively behind you, not on your left side where you could easily see it.", + "E": "The garlic was positioned in front and slightly to your right, rather than directly to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 17 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_4_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of seasoning did I take out of my pocket in total?", + "answer": "In total, you took two bottles of seasoning; they were in front of you.", + "question_chinese": "我一共从口袋里拿了几瓶调料？", + "answer_chinese": "你一共拿了两瓶调料，它们在你前面。", + "evidence": [ + { + "description": "You took two bottles the first time.", + "timestamps": { + "start": "00:06:40", + "end": "00:06:45" + } + } + ], + "question_timestamps": "00:06:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In total, you took two bottles of seasoning; they were in front of you.", + "B": "You took one bottle total.", + "C": "You took three bottles total.", + "D": "In total, you took four bottles of seasoning, and they ended up set down in front of you.", + "E": "Altogether, you took zero bottles of seasoning from your pocket, so none were placed in front of you." + }, + "answer_label": "A", + "answer_text": "In total, you took two bottles of seasoning; they were in front of you.", + "q_id": "P34_102_4_T1", + "correct_answer": "In total, you took two bottles of seasoning; they were in front of you.", + "original_options": { + "A": "In total, you took two bottles of seasoning; they were in front of you.", + "B": "You took one bottle total.", + "C": "You took three bottles total.", + "D": "In total, you took four bottles of seasoning, and they ended up set down in front of you.", + "E": "Altogether, you took zero bottles of seasoning from your pocket, so none were placed in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_4_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of seasoning did I take out of my pocket in total?", + "answer": "In total, you took three bottles of seasoning; they were in front of you.", + "question_chinese": "我一共从口袋里拿了几瓶调料？", + "answer_chinese": "你一共拿了三瓶调料，它们在你前面。", + "evidence": [ + { + "description": "You took a bottle the second time.", + "timestamps": { + "start": "00:06:48", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:06:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You took two bottles total.", + "B": "In total, you took three bottles of seasoning; they were in front of you.", + "C": "You took four bottles total.", + "D": "In total, you took one bottle of seasoning, and it was placed in front of you afterward.", + "E": "Altogether, you took five bottles of seasoning out of your pocket, leaving them lined up in front of you." + }, + "answer_label": "B", + "answer_text": "In total, you took three bottles of seasoning; they were in front of you.", + "q_id": "P34_102_4_T2", + "correct_answer": "In total, you took three bottles of seasoning; they were in front of you.", + "original_options": { + "A": "You took two bottles total.", + "B": "In total, you took three bottles of seasoning; they were in front of you.", + "C": "You took four bottles total.", + "D": "In total, you took one bottle of seasoning, and it was placed in front of you afterward.", + "E": "Altogether, you took five bottles of seasoning out of your pocket, leaving them lined up in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_7", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of seasonings did I take out of my pocket in total?", + "answer": "You took a total of three bottles of seasonings.", + "question_chinese": "我一共从口袋里拿了几瓶调料？", + "answer_chinese": "你一共拿了三瓶调料。", + "evidence": [ + { + "description": "You took two bottles the first time.", + "timestamps": { + "start": "00:06:37", + "end": "00:06:46" + } + }, + { + "description": "You took a bottle the second time.", + "timestamps": { + "start": "00:06:48", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:07:22", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You took two bottles total.", + "B": "You took a total of three bottles of seasonings.", + "C": "You took four bottles total.", + "D": "You took a total of five bottles of seasonings over the whole sequence.", + "E": "You took a total of six bottles of seasonings altogether from your pocket." + }, + "answer_label": "B", + "answer_text": "You took a total of three bottles of seasonings.", + "q_id": "P34_102_7", + "correct_answer": "You took a total of three bottles of seasonings.", + "original_options": { + "A": "You took two bottles total.", + "B": "You took a total of three bottles of seasonings.", + "C": "You took four bottles total.", + "D": "You took a total of five bottles of seasonings over the whole sequence.", + "E": "You took a total of six bottles of seasonings altogether from your pocket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_8", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of seasonings did I take out of my pocket in total?", + "answer": "You took a total of three bottles of seasonings.", + "question_chinese": "我一共从口袋里拿了几瓶调料？", + "answer_chinese": "你一共拿了三瓶调料。", + "evidence": [ + { + "description": "You took two bottles the first time.", + "timestamps": { + "start": "00:06:37", + "end": "00:06:46" + } + }, + { + "description": "You took a bottle the second time.", + "timestamps": { + "start": "00:06:48", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:08:24", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You took four bottles total.", + "B": "You took five bottles total.", + "C": "You took a total of six bottles of seasonings across the entire moment.", + "D": "You took a total of seven bottles of seasonings altogether from your pocket.", + "E": "You took a total of three bottles of seasonings." + }, + "answer_label": "E", + "answer_text": "You took a total of three bottles of seasonings.", + "q_id": "P34_102_8", + "correct_answer": "You took a total of three bottles of seasonings.", + "original_options": { + "A": "You took four bottles total.", + "B": "You took five bottles total.", + "C": "You took a total of six bottles of seasonings across the entire moment.", + "D": "You took a total of seven bottles of seasonings altogether from your pocket.", + "E": "You took a total of three bottles of seasonings." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_15", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the state when I am washing the dishes and facing my window directly?", + "answer": "The window is closed.", + "question_chinese": "我在清洗餐具时，正对着我的窗子的状态是什么？", + "answer_chinese": "窗子是关着的。", + "evidence": [ + { + "description": "The window is closed.", + "timestamps": { + "start": "00:08:28", + "end": "00:08:29" + } + } + ], + "question_timestamps": "00:08:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is open.", + "B": "The window is broken.", + "C": "The window is closed.", + "D": "The window is slightly open while you wash dishes facing it directly.", + "E": "The window is fogged up as you stand at the sink looking straight at it." + }, + "answer_label": "C", + "answer_text": "The window is closed.", + "q_id": "P34_102_15", + "correct_answer": "The window is closed.", + "original_options": { + "A": "The window is open.", + "B": "The window is broken.", + "C": "The window is closed.", + "D": "The window is slightly open while you wash dishes facing it directly.", + "E": "The window is fogged up as you stand at the sink looking straight at it." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 15 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_16", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the state when I am washing the dishes and facing my window directly?", + "answer": "The window is closed.", + "question_chinese": "我在清洗餐具时，正对着我的窗子的状态是什么？", + "answer_chinese": "窗子是关着的。", + "evidence": [ + { + "description": "The window is closed.", + "timestamps": { + "start": "00:08:28", + "end": "00:08:29" + } + } + ], + "question_timestamps": "00:08:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is open.", + "B": "The window is fogged up.", + "C": "The window is slightly open while you are washing the dishes and facing it.", + "D": "The window is completely shattered as you wash dishes, looking directly toward it.", + "E": "The window is closed." + }, + "answer_label": "E", + "answer_text": "The window is closed.", + "q_id": "P34_102_16", + "correct_answer": "The window is closed.", + "original_options": { + "A": "The window is open.", + "B": "The window is fogged up.", + "C": "The window is slightly open while you are washing the dishes and facing it.", + "D": "The window is completely shattered as you wash dishes, looking directly toward it.", + "E": "The window is closed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 13 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_13", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to cook with the vegetables in the fridge. Where is the fridge?", + "answer": "The refrigerator is right opposite the kitchen door.", + "question_chinese": "我想要拿冰箱里的菜做饭，冰箱在哪里？", + "answer_chinese": "冰箱在正对着厨房门的地方。", + "evidence": [ + { + "description": "The refrigerator is right opposite the kitchen door.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:35" + } + } + ], + "question_timestamps": "00:07:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s next to the kitchen island.", + "B": "It’s beside the pantry door.", + "C": "The refrigerator is across from the dining table, not directly opposite the kitchen door.", + "D": "The refrigerator is right opposite the kitchen door.", + "E": "The refrigerator is at the back of the room, away from the kitchen door you entered through." + }, + "answer_label": "D", + "answer_text": "The refrigerator is right opposite the kitchen door.", + "q_id": "P34_102_13", + "correct_answer": "The refrigerator is right opposite the kitchen door.", + "original_options": { + "A": "It’s next to the kitchen island.", + "B": "It’s beside the pantry door.", + "C": "The refrigerator is across from the dining table, not directly opposite the kitchen door.", + "D": "The refrigerator is right opposite the kitchen door.", + "E": "The refrigerator is at the back of the room, away from the kitchen door you entered through." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 17 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_14", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to cook with the vegetables in the fridge. Where is the fridge?", + "answer": "The refrigerator is right opposite the kitchen door.", + "question_chinese": "我想要拿冰箱里的菜做饭，冰箱在哪里？", + "answer_chinese": "冰箱在正对着厨房门的地方。", + "evidence": [ + { + "description": "The refrigerator is right opposite the kitchen door.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:35" + } + } + ], + "question_timestamps": "00:07:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s next to the window.", + "B": "It’s behind the dining table.", + "C": "The refrigerator is across from the pantry door rather than opposite the kitchen door.", + "D": "The refrigerator is positioned above the kitchen countertop, not on the wall facing the door.", + "E": "The refrigerator is right opposite the kitchen door." + }, + "answer_label": "E", + "answer_text": "The refrigerator is right opposite the kitchen door.", + "q_id": "P34_102_14", + "correct_answer": "The refrigerator is right opposite the kitchen door.", + "original_options": { + "A": "It’s next to the window.", + "B": "It’s behind the dining table.", + "C": "The refrigerator is across from the pantry door rather than opposite the kitchen door.", + "D": "The refrigerator is positioned above the kitchen countertop, not on the wall facing the door.", + "E": "The refrigerator is right opposite the kitchen door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_9", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the box for the ribs from?", + "answer": "You threw the box from the table into the trash can.", + "question_chinese": "我把装肋骨的盒子从哪里拿到了哪里？", + "answer_chinese": "你把盒子从桌子上扔到了垃圾桶里。", + "evidence": [ + { + "description": "The box was on the table.", + "timestamps": { + "start": "00:07:38", + "end": "00:07:43" + } + }, + { + "description": "You threw the box into the trash can.", + "timestamps": { + "start": "00:07:45", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:08:07", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took it from the shelf.", + "B": "You grabbed it off the counter.", + "C": "You picked the box up from the floor near the doorway before tossing it away.", + "D": "You threw the box from the table into the trash can.", + "E": "You lifted the box from the chair beside the table and then threw it into the bin." + }, + "answer_label": "D", + "answer_text": "You threw the box from the table into the trash can.", + "q_id": "P34_102_9", + "correct_answer": "You threw the box from the table into the trash can.", + "original_options": { + "A": "You took it from the shelf.", + "B": "You grabbed it off the counter.", + "C": "You picked the box up from the floor near the doorway before tossing it away.", + "D": "You threw the box from the table into the trash can.", + "E": "You lifted the box from the chair beside the table and then threw it into the bin." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_10", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the box for the ribs from?", + "answer": "You threw the box from the table into the trash can.", + "question_chinese": "我把装肋骨的盒子从哪里拿到了哪里？", + "answer_chinese": "你把盒子从桌子上扔到了垃圾桶里。", + "evidence": [ + { + "description": "The box was on the table.", + "timestamps": { + "start": "00:07:38", + "end": "00:07:43" + } + }, + { + "description": "You threw the box into the trash can.", + "timestamps": { + "start": "00:07:45", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:08:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You threw the box from the table into the trash can.", + "B": "You took it from the counter.", + "C": "You got it from the shelf.", + "D": "You grabbed the box from the counter by the sink and then set it down nearby.", + "E": "You picked the box up from a shelf above the stove before moving it toward the trash." + }, + "answer_label": "A", + "answer_text": "You threw the box from the table into the trash can.", + "q_id": "P34_102_10", + "correct_answer": "You threw the box from the table into the trash can.", + "original_options": { + "A": "You threw the box from the table into the trash can.", + "B": "You took it from the counter.", + "C": "You got it from the shelf.", + "D": "You grabbed the box from the counter by the sink and then set it down nearby.", + "E": "You picked the box up from a shelf above the stove before moving it toward the trash." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_5_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the box with the ribs?", + "answer": "You put the box holding the ribs onto the table; it was in front of you.", + "question_chinese": "我把装肋骨的盒子放到了哪里？", + "answer_chinese": "你把装肋骨的盒子放到了桌子上，它在你前面。", + "evidence": [ + { + "description": "The box was on the table.", + "timestamps": { + "start": "00:07:34", + "end": "00:07:38" + } + } + ], + "question_timestamps": "00:07:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put the box holding the ribs onto the table; it was in front of you.", + "B": "You put it behind you.", + "C": "You put it to your left.", + "D": "You placed the ribs box onto the table, but it ended up on your right side.", + "E": "You set the box on the table and positioned it behind you rather than directly ahead." + }, + "answer_label": "A", + "answer_text": "You put the box holding the ribs onto the table; it was in front of you.", + "q_id": "P34_102_5_T1", + "correct_answer": "You put the box holding the ribs onto the table; it was in front of you.", + "original_options": { + "A": "You put the box holding the ribs onto the table; it was in front of you.", + "B": "You put it behind you.", + "C": "You put it to your left.", + "D": "You placed the ribs box onto the table, but it ended up on your right side.", + "E": "You set the box on the table and positioned it behind you rather than directly ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_5_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the box with the ribs?", + "answer": "You put the box holding the ribs into the trash can; it was to your right.", + "question_chinese": "我把装肋骨的盒子放到了哪里？", + "answer_chinese": "我把装肋骨的盒子放到了垃圾桶，它在你右边。", + "evidence": [ + { + "description": "You threw the box into the trash can.", + "timestamps": { + "start": "00:07:45", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "It was to your left.", + "C": "You put the box holding the ribs into the trash can; it was to your right.", + "D": "You put the ribs box into the trash can, but the can was positioned behind you.", + "E": "You placed the box into the trash can while it sat in front of you instead of to the right." + }, + "answer_label": "C", + "answer_text": "You put the box holding the ribs into the trash can; it was to your right.", + "q_id": "P34_102_5_T2", + "correct_answer": "You put the box holding the ribs into the trash can; it was to your right.", + "original_options": { + "A": "It was in front of you.", + "B": "It was to your left.", + "C": "You put the box holding the ribs into the trash can; it was to your right.", + "D": "You put the ribs box into the trash can, but the can was positioned behind you.", + "E": "You placed the box into the trash can while it sat in front of you instead of to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 20 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_2_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How did I wash the kitchen knife I used to cut the meat?", + "answer": "You washed it by hand; it is in front of you.", + "question_chinese": "我是怎么清洗的我切肉的菜刀的？", + "answer_chinese": "你用手清洗，它在你前面。", + "evidence": [ + { + "description": "First, roughly clean the kitchen knife by hand.", + "timestamps": { + "start": "00:07:59", + "end": "00:08:12" + } + } + ], + "question_timestamps": "00:08:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed it by hand; it is in front of you.", + "B": "You washed it in the dishwasher.", + "C": "You only rinsed it quickly.", + "D": "You washed it by hand, but the knife was positioned to your right instead of in front.", + "E": "You washed it by hand with the knife located behind you, not directly in front of you." + }, + "answer_label": "A", + "answer_text": "You washed it by hand; it is in front of you.", + "q_id": "P34_102_2_T1", + "correct_answer": "You washed it by hand; it is in front of you.", + "original_options": { + "A": "You washed it by hand; it is in front of you.", + "B": "You washed it in the dishwasher.", + "C": "You only rinsed it quickly.", + "D": "You washed it by hand, but the knife was positioned to your right instead of in front.", + "E": "You washed it by hand with the knife located behind you, not directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_2_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How did I wash the kitchen knife I used to cut the meat?", + "answer": "You washed the knife blade with a sponge; it was in front of you.", + "question_chinese": "我是怎么清洗的我切肉的菜刀的？", + "answer_chinese": "你用海绵清洗了刀身，它在你前面。", + "evidence": [ + { + "description": "You wiped the blade with a sponge.", + "timestamps": { + "start": "00:08:14", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:08:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed it with a brush.", + "B": "You rinsed it under water.", + "C": "You washed the knife blade with a sponge, but it was positioned to your right.", + "D": "You washed the knife blade with a sponge; it was in front of you.", + "E": "You scrubbed the knife blade using a sponge while it sat behind you rather than in front." + }, + "answer_label": "D", + "answer_text": "You washed the knife blade with a sponge; it was in front of you.", + "q_id": "P34_102_2_T2", + "correct_answer": "You washed the knife blade with a sponge; it was in front of you.", + "original_options": { + "A": "You washed it with a brush.", + "B": "You rinsed it under water.", + "C": "You washed the knife blade with a sponge, but it was positioned to your right.", + "D": "You washed the knife blade with a sponge; it was in front of you.", + "E": "You scrubbed the knife blade using a sponge while it sat behind you rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 17 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_4", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How do I clean the kitchen knife I use for cutting meat?", + "answer": "First, you roughly washed it by hand, and then wiped the blade with a sponge.", + "question_chinese": "我是怎么清洗的我切肉的菜刀的？", + "answer_chinese": "你先用手粗略的清洗，再用海绵擦拭了刀身。", + "evidence": [ + { + "description": "First, roughly clean the kitchen knife by hand.", + "timestamps": { + "start": "00:07:59", + "end": "00:08:12" + } + }, + { + "description": "You wiped the blade with a sponge.", + "timestamps": { + "start": "00:08:14", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:08:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Rinse it, then wipe the handle.", + "B": "Scrub in a bowl, then wipe with cloth.", + "C": "First, you carefully rinsed the knife under cold water, then wiped the handle with a towel.", + "D": "First, you roughly washed it by hand, and then wiped the blade with a sponge.", + "E": "First, you dried it on a rack, and then wiped the blade with a paper towel afterward." + }, + "answer_label": "D", + "answer_text": "First, you roughly washed it by hand, and then wiped the blade with a sponge.", + "q_id": "P34_102_4", + "correct_answer": "First, you roughly washed it by hand, and then wiped the blade with a sponge.", + "original_options": { + "A": "Rinse it, then wipe the handle.", + "B": "Scrub in a bowl, then wipe with cloth.", + "C": "First, you carefully rinsed the knife under cold water, then wiped the handle with a towel.", + "D": "First, you roughly washed it by hand, and then wiped the blade with a sponge.", + "E": "First, you dried it on a rack, and then wiped the blade with a paper towel afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 16, + 17 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_12", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I'm cutting meat on the cutting board. How can I get the garlic?", + "answer": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "question_chinese": "我正在菜板上切肉，我该怎样拿到大蒜？", + "answer_chinese": "大蒜菜板的正前面，你伸伸手能够拿到。", + "evidence": [ + { + "description": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:04:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "B": "To your right; reach slightly.", + "C": "Behind the board; reach back.", + "D": "Beside the fish cutting board, you can get the garlic by reaching across the table.", + "E": "On the left near the spice cutting board, you need to stand up to reach the garlic." + }, + "answer_label": "A", + "answer_text": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "q_id": "P34_102_12", + "correct_answer": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "original_options": { + "A": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "B": "To your right; reach slightly.", + "C": "Behind the board; reach back.", + "D": "Beside the fish cutting board, you can get the garlic by reaching across the table.", + "E": "On the left near the spice cutting board, you need to stand up to reach the garlic." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_11", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I'm cutting meat on the cutting board. How can I get the garlic?", + "answer": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "question_chinese": "我正在菜板上切肉，我该怎样拿到大蒜？", + "answer_chinese": "大蒜菜板的正前面，你伸伸手能够拿到。", + "evidence": [ + { + "description": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:06:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Just behind the cutting board.", + "B": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "C": "Farther left of the cutting board.", + "D": "It’s slightly behind and left of the cutting board, so you’d need to lean forward to grab it.", + "E": "It sits a bit lower than the cutting board in front, so you’d have to bend down to reach it." + }, + "answer_label": "B", + "answer_text": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "q_id": "P34_102_11", + "correct_answer": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "original_options": { + "A": "Just behind the cutting board.", + "B": "Right in front of the garlic cutting board, you can reach it by stretching out your hand.", + "C": "Farther left of the cutting board.", + "D": "It’s slightly behind and left of the cutting board, so you’d need to lean forward to grab it.", + "E": "It sits a bit lower than the cutting board in front, so you’d have to bend down to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 21 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_8_T3", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, how many pots did I take from the stovetop?", + "answer": "At this moment, you took 0 pots from the stove; it was to your left.", + "question_chinese": "此时，我从灶台上拿了几个锅？", + "answer_chinese": "此时，你从灶台上拿了0个锅，它在你左边。", + "evidence": [ + { + "description": "You did not take a pot.", + "timestamps": { + "start": "00:04:28", + "end": "00:04:29" + } + } + ], + "question_timestamps": "00:04:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took 1 pot.", + "B": "At this moment, you took 0 pots from the stove; it was to your left.", + "C": "You took 2 pots.", + "D": "At this moment, you took 3 pots from the stove, and they were positioned to your left.", + "E": "At this moment, you took 4 pots from the stove, and they were all on the left side." + }, + "answer_label": "B", + "answer_text": "At this moment, you took 0 pots from the stove; it was to your left.", + "q_id": "P34_102_8_T3", + "correct_answer": "At this moment, you took 0 pots from the stove; it was to your left.", + "original_options": { + "A": "You took 1 pot.", + "B": "At this moment, you took 0 pots from the stove; it was to your left.", + "C": "You took 2 pots.", + "D": "At this moment, you took 3 pots from the stove, and they were positioned to your left.", + "E": "At this moment, you took 4 pots from the stove, and they were all on the left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_8_T2", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, how many pots did I take from the stovetop?", + "answer": "At this moment, you took 1 pot from the stove; it was in front of you.", + "question_chinese": "此时，我从灶台上拿了几个锅？", + "answer_chinese": "此时，你从灶台上拿了1个锅，它在你前面。", + "evidence": [ + { + "description": "You took a pot.", + "timestamps": { + "start": "00:04:32", + "end": "00:04:34" + } + } + ], + "question_timestamps": "00:04:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "At this moment, you took 1 pot from the stove; it was in front of you.", + "B": "You took 0 pots.", + "C": "You took 2 pots.", + "D": "At this moment, you took 3 pots from the stove, and they were positioned in front of you.", + "E": "At this moment, you took 4 pots from the stove, and they were all located in front of you." + }, + "answer_label": "A", + "answer_text": "At this moment, you took 1 pot from the stove; it was in front of you.", + "q_id": "P34_102_8_T2", + "correct_answer": "At this moment, you took 1 pot from the stove; it was in front of you.", + "original_options": { + "A": "At this moment, you took 1 pot from the stove; it was in front of you.", + "B": "You took 0 pots.", + "C": "You took 2 pots.", + "D": "At this moment, you took 3 pots from the stove, and they were positioned in front of you.", + "E": "At this moment, you took 4 pots from the stove, and they were all located in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_1_T2", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, what did I add into the pot?", + "answer": "You added pork ribs into the pot; they are in front of you.", + "question_chinese": "此时，我在锅里加入了哪些东西？", + "answer_chinese": "你在锅里加入了排骨，它在你前面。", + "evidence": [ + { + "description": "You added the ribs to the pot.", + "timestamps": { + "start": "00:04:36", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:05:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You added chicken pieces into the pot.", + "B": "You added pork ribs into the pot; they are in front of you.", + "C": "You added beef chunks into the pot.", + "D": "You added pork ribs into the pot, but they ended up positioned to your right side.", + "E": "You added pork ribs into the pot, and they were located above you rather than directly in front." + }, + "answer_label": "B", + "answer_text": "You added pork ribs into the pot; they are in front of you.", + "q_id": "P34_102_1_T2", + "correct_answer": "You added pork ribs into the pot; they are in front of you.", + "original_options": { + "A": "You added chicken pieces into the pot.", + "B": "You added pork ribs into the pot; they are in front of you.", + "C": "You added beef chunks into the pot.", + "D": "You added pork ribs into the pot, but they ended up positioned to your right side.", + "E": "You added pork ribs into the pot, and they were located above you rather than directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_1_T1", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, what did I add into the pot?", + "answer": "You added ginger and cooking wine into the pot; they are in front of you.", + "question_chinese": "此时，我在锅里加入了哪些东西？", + "answer_chinese": "你在锅里加入了姜和料酒，它们在你前面。", + "evidence": [ + { + "description": "You added ginger to the pot.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:09" + } + }, + { + "description": "You added cooking wine to the pot.", + "timestamps": { + "start": "00:06:52", + "end": "00:06:58" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You added only ginger into the pot.", + "B": "You added ginger and cooking wine into the pot; they are in front of you.", + "C": "You added only cooking wine into the pot.", + "D": "You added ginger and cooking wine into the pot, but they were positioned to your right side.", + "E": "You added ginger and cooking wine into the pot, and they ended up above you instead of in front." + }, + "answer_label": "B", + "answer_text": "You added ginger and cooking wine into the pot; they are in front of you.", + "q_id": "P34_102_1_T1", + "correct_answer": "You added ginger and cooking wine into the pot; they are in front of you.", + "original_options": { + "A": "You added only ginger into the pot.", + "B": "You added ginger and cooking wine into the pot; they are in front of you.", + "C": "You added only cooking wine into the pot.", + "D": "You added ginger and cooking wine into the pot, but they were positioned to your right side.", + "E": "You added ginger and cooking wine into the pot, and they ended up above you instead of in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 19 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_3_T2", + "qid_num": 26, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Right now, where did I put the ribs?", + "answer": "At this moment, you had placed the ribs on the cutting board; they were in front of you.", + "question_chinese": "此时，肋骨被我放在哪里了？", + "answer_chinese": "此时，肋骨被你放在砧板上了，它在你前面。", + "evidence": [ + { + "description": "The ribs are in the cutting board。", + "timestamps": { + "start": "00:00:20", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:00:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the board behind you.", + "B": "On the board to your left.", + "C": "At this moment, you had placed the ribs on the cutting board, and they were positioned to your right.", + "D": "At this moment, you had placed the ribs on the cutting board, and they were located above you.", + "E": "At this moment, you had placed the ribs on the cutting board; they were in front of you." + }, + "answer_label": "E", + "answer_text": "At this moment, you had placed the ribs on the cutting board; they were in front of you.", + "q_id": "P34_102_3_T2", + "correct_answer": "At this moment, you had placed the ribs on the cutting board; they were in front of you.", + "original_options": { + "A": "On the board behind you.", + "B": "On the board to your left.", + "C": "At this moment, you had placed the ribs on the cutting board, and they were positioned to your right.", + "D": "At this moment, you had placed the ribs on the cutting board, and they were located above you.", + "E": "At this moment, you had placed the ribs on the cutting board; they were in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 18 + ] + }, + { + "video_uid": "P34_102", + "video_length": "00:08:32", + "qid": "P34_102_3_T1", + "qid_num": 27, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Right now, where did I put the ribs?", + "answer": "You put it into the pot; it was in front of you.", + "question_chinese": "此时，肋骨被我放在哪里了？", + "answer_chinese": "被你放进了锅里了，它在你前面。", + "evidence": [ + { + "description": "The ribs are put into the pot.", + "timestamps": { + "start": "00:04:34", + "end": "00:05:12" + } + } + ], + "question_timestamps": "00:05:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Into the pot behind you.", + "B": "Into the pot to your left.", + "C": "You put it into the pot; it was in front of you.", + "D": "You put it into the pot, and it was positioned to your right rather than in front of you.", + "E": "You put it into the pot, and it was located above you instead of directly in front." + }, + "answer_label": "C", + "answer_text": "You put it into the pot; it was in front of you.", + "q_id": "P34_102_3_T1", + "correct_answer": "You put it into the pot; it was in front of you.", + "original_options": { + "A": "Into the pot behind you.", + "B": "Into the pot to your left.", + "C": "You put it into the pot; it was in front of you.", + "D": "You put it into the pot, and it was positioned to your right rather than in front of you.", + "E": "You put it into the pot, and it was located above you instead of directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 17 + ] + } +] diff --git a/QAs/P34_104-mcq.json b/QAs/P34_104-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..246a5d65cfd222dbd4c406acd8795723ffbf3a88 --- /dev/null +++ b/QAs/P34_104-mcq.json @@ -0,0 +1,955 @@ +[ + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_15", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the approximate distance between the kitchen door and the refrigerator?", + "answer": "There is approximately a distance of four body positions.", + "question_chinese": "厨房门与冰箱的距离大致有多少？", + "answer_chinese": "大致有四个身位的距离。", + "evidence": [ + { + "description": "There is approximately a distance of two body positions.", + "timestamps": { + "start": "00:07:20", + "end": "00:07:21" + } + } + ], + "question_timestamps": "00:08:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "About two body positions.", + "B": "About three body positions.", + "C": "There is approximately a distance of five body positions between the kitchen door and the refrigerator.", + "D": "There is approximately a distance of four body positions.", + "E": "There is approximately a distance of six body positions between the kitchen door and the refrigerator." + }, + "answer_label": "D", + "answer_text": "There is approximately a distance of four body positions.", + "q_id": "P34_104_15", + "correct_answer": "There is approximately a distance of four body positions.", + "original_options": { + "A": "About two body positions.", + "B": "About three body positions.", + "C": "There is approximately a distance of five body positions between the kitchen door and the refrigerator.", + "D": "There is approximately a distance of four body positions.", + "E": "There is approximately a distance of six body positions between the kitchen door and the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_16", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the approximate distance between the kitchen door and the refrigerator?", + "answer": "There is approximately a distance of four body positions.", + "question_chinese": "厨房门与冰箱的距离大致有多少？", + "answer_chinese": "大致有四个身位的距离。", + "evidence": [ + { + "description": "There is approximately a distance of two body positions.", + "timestamps": { + "start": "00:07:20", + "end": "00:07:21" + } + } + ], + "question_timestamps": "00:11:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "About three body positions.", + "B": "About five body positions.", + "C": "It looks closer, roughly three body positions from the kitchen door to the refrigerator.", + "D": "There is approximately a distance of four body positions.", + "E": "It seems farther, around five body positions separating the kitchen door and the refrigerator." + }, + "answer_label": "D", + "answer_text": "There is approximately a distance of four body positions.", + "q_id": "P34_104_16", + "correct_answer": "There is approximately a distance of four body positions.", + "original_options": { + "A": "About three body positions.", + "B": "About five body positions.", + "C": "It looks closer, roughly three body positions from the kitchen door to the refrigerator.", + "D": "There is approximately a distance of four body positions.", + "E": "It seems farther, around five body positions separating the kitchen door and the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I need scallions for cooking. Where can I get them?", + "answer": "The scallions are in the fourth compartment of the refrigerator opposite the door.", + "question_chinese": "我做饭需要葱，葱从哪里可以拿到？", + "answer_chinese": "葱在门对面里的冰箱里第四格。", + "evidence": [ + { + "description": "The refrigerator is opposite the door.", + "timestamps": { + "start": "00:07:21", + "end": "00:07:22" + } + }, + { + "description": "The scallions are in the fourth compartment of the refrigerator.", + "timestamps": { + "start": "00:07:25", + "end": "00:07:26" + } + } + ], + "question_timestamps": "00:10:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In the third fridge compartment.", + "B": "In the bottom freezer drawer.", + "C": "You can find the scallions in the second compartment of the refrigerator, closer to the door.", + "D": "The scallions are in the fourth compartment of the refrigerator opposite the door.", + "E": "Check the first compartment in the refrigerator; the scallions are stored there beside other produce." + }, + "answer_label": "D", + "answer_text": "The scallions are in the fourth compartment of the refrigerator opposite the door.", + "q_id": "P34_104_7", + "correct_answer": "The scallions are in the fourth compartment of the refrigerator opposite the door.", + "original_options": { + "A": "In the third fridge compartment.", + "B": "In the bottom freezer drawer.", + "C": "You can find the scallions in the second compartment of the refrigerator, closer to the door.", + "D": "The scallions are in the fourth compartment of the refrigerator opposite the door.", + "E": "Check the first compartment in the refrigerator; the scallions are stored there beside other produce." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_8", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I need scallions for cooking. Where can I get them?", + "answer": "The scallions are in the fourth compartment of the refrigerator opposite the door.", + "question_chinese": "我做饭需要葱，葱从哪里可以拿到？", + "answer_chinese": "葱在门对面里的冰箱里第四格。", + "evidence": [ + { + "description": "The refrigerator is opposite the door.", + "timestamps": { + "start": "00:07:21", + "end": "00:07:22" + } + }, + { + "description": "The scallions are in the fourth compartment of the refrigerator.", + "timestamps": { + "start": "00:07:25", + "end": "00:07:26" + } + } + ], + "question_timestamps": "00:11:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In the third fridge compartment.", + "B": "In the first fridge compartment.", + "C": "The scallions are in the second compartment of the refrigerator, positioned near the front of the shelf.", + "D": "Look in the pantry instead; they’re kept in a compartment near where the spices are stored.", + "E": "The scallions are in the fourth compartment of the refrigerator opposite the door." + }, + "answer_label": "E", + "answer_text": "The scallions are in the fourth compartment of the refrigerator opposite the door.", + "q_id": "P34_104_8", + "correct_answer": "The scallions are in the fourth compartment of the refrigerator opposite the door.", + "original_options": { + "A": "In the third fridge compartment.", + "B": "In the first fridge compartment.", + "C": "The scallions are in the second compartment of the refrigerator, positioned near the front of the shelf.", + "D": "Look in the pantry instead; they’re kept in a compartment near where the spices are stored.", + "E": "The scallions are in the fourth compartment of the refrigerator opposite the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_13", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potted plants are there on the windowsill of my kitchen?", + "answer": "You placed seven potted plants.", + "question_chinese": "我厨房的窗台上放了几个盆栽？", + "answer_chinese": "你放了七个盆栽。", + "evidence": [ + { + "description": "You placed seven potted plants.", + "timestamps": { + "start": "00:14:07", + "end": "00:14:08" + } + } + ], + "question_timestamps": "00:14:35", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You placed five potted plants.", + "B": "You placed nine potted plants.", + "C": "You placed six potted plants on the kitchen windowsill, lined up next to each other.", + "D": "You placed seven potted plants.", + "E": "You placed eight potted plants on the windowsill, filling most of the available space." + }, + "answer_label": "D", + "answer_text": "You placed seven potted plants.", + "q_id": "P34_104_13", + "correct_answer": "You placed seven potted plants.", + "original_options": { + "A": "You placed five potted plants.", + "B": "You placed nine potted plants.", + "C": "You placed six potted plants on the kitchen windowsill, lined up next to each other.", + "D": "You placed seven potted plants.", + "E": "You placed eight potted plants on the windowsill, filling most of the available space." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_14", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many potted plants are there on the windowsill of my kitchen?", + "answer": "You placed seven potted plants.", + "question_chinese": "我厨房的窗台上放了几个盆栽？", + "answer_chinese": "你放了七个盆栽。", + "evidence": [ + { + "description": "You placed seven potted plants.", + "timestamps": { + "start": "00:14:07", + "end": "00:14:08" + } + } + ], + "question_timestamps": "00:14:55", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You placed seven potted plants.", + "B": "You placed six potted plants.", + "C": "You placed nine potted plants.", + "D": "You placed five potted plants on the kitchen windowsill, grouped together in one row.", + "E": "You placed eight potted plants along the windowsill, spread out across the length of it." + }, + "answer_label": "A", + "answer_text": "You placed seven potted plants.", + "q_id": "P34_104_14", + "correct_answer": "You placed seven potted plants.", + "original_options": { + "A": "You placed seven potted plants.", + "B": "You placed six potted plants.", + "C": "You placed nine potted plants.", + "D": "You placed five potted plants on the kitchen windowsill, grouped together in one row.", + "E": "You placed eight potted plants along the windowsill, spread out across the length of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_1_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was stir-frying the ribs, what did I add into the wok?", + "answer": "You added rock sugar, scallion, ginger, and garlic. The wok is in front of you.", + "question_chinese": "我在炒排骨的时候，我在炒锅里加入了什么东西？", + "answer_chinese": "你加入了冰糖，葱姜蒜。炒锅在你前面。", + "evidence": [ + { + "description": "You added rock sugar.", + "timestamps": { + "start": "00:05:28", + "end": "00:07:20" + } + }, + { + "description": "You added scallions, ginger, garlic.", + "timestamps": { + "start": "00:10:49", + "end": "00:10:54" + } + } + ], + "question_timestamps": "00:10:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You added rock sugar and scallion.", + "B": "You added ginger and garlic.", + "C": "You added rock sugar, ginger, and garlic into the wok, which is in front of you.", + "D": "You added rock sugar, scallion, ginger, garlic, and five-spice seasoning; the wok is in front of you.", + "E": "You added rock sugar, scallion, ginger, and garlic. The wok is in front of you." + }, + "answer_label": "E", + "answer_text": "You added rock sugar, scallion, ginger, and garlic. The wok is in front of you.", + "q_id": "P34_104_1_T1", + "correct_answer": "You added rock sugar, scallion, ginger, and garlic. The wok is in front of you.", + "original_options": { + "A": "You added rock sugar and scallion.", + "B": "You added ginger and garlic.", + "C": "You added rock sugar, ginger, and garlic into the wok, which is in front of you.", + "D": "You added rock sugar, scallion, ginger, garlic, and five-spice seasoning; the wok is in front of you.", + "E": "You added rock sugar, scallion, ginger, and garlic. The wok is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was stir-frying the pork ribs, what did I add to the wok?", + "answer": "You added rock sugar, scallions, ginger, garlic, and five-spice seasonings and condiments.", + "question_chinese": "我在炒排骨的时候，我在炒锅里加入了什么东西？", + "answer_chinese": "你加入了冰糖，葱姜蒜和五香调料和调味料。", + "evidence": [ + { + "description": "You added rock sugar.", + "timestamps": { + "start": "00:05:28", + "end": "00:07:20" + } + }, + { + "description": "You added scallions, ginger, garlic.", + "timestamps": { + "start": "00:10:49", + "end": "00:10:54" + } + }, + { + "description": "You added rock five-spice seasonings.", + "timestamps": { + "start": "00:11:18", + "end": "00:11:47" + } + }, + { + "description": "You added condiments.", + "timestamps": { + "start": "00:12:33", + "end": "00:13:01" + } + } + ], + "question_timestamps": "00:13:23", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You added brown sugar and leeks.", + "B": "You added rock sugar and soy sauce.", + "C": "You added white sugar, scallions, ginger, garlic, and five-spice powder into the wok while cooking.", + "D": "You added rock sugar, scallions, ginger, garlic, plus star anise and other spices to the wok.", + "E": "You added rock sugar, scallions, ginger, garlic, and five-spice seasonings and condiments." + }, + "answer_label": "E", + "answer_text": "You added rock sugar, scallions, ginger, garlic, and five-spice seasonings and condiments.", + "q_id": "P34_104_1", + "correct_answer": "You added rock sugar, scallions, ginger, garlic, and five-spice seasonings and condiments.", + "original_options": { + "A": "You added brown sugar and leeks.", + "B": "You added rock sugar and soy sauce.", + "C": "You added white sugar, scallions, ginger, garlic, and five-spice powder into the wok while cooking.", + "D": "You added rock sugar, scallions, ginger, garlic, plus star anise and other spices to the wok.", + "E": "You added rock sugar, scallions, ginger, garlic, and five-spice seasonings and condiments." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 16 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_1_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was stir-frying the ribs, what did I add into the wok?", + "answer": "You added rock sugar, scallion, ginger, garlic, five-spice seasoning, and seasonings. The wok is to your back-left.", + "question_chinese": "我在炒排骨的时候，我在炒锅里加入了什么东西？", + "answer_chinese": "你加入了冰糖，葱姜蒜和五香调料和调味料。炒锅在你左后方。", + "evidence": [ + { + "description": "You added rock sugar.", + "timestamps": { + "start": "00:05:28", + "end": "00:07:20" + } + }, + { + "description": "You added scallions, ginger, garlic.", + "timestamps": { + "start": "00:10:49", + "end": "00:10:54" + } + }, + { + "description": "You added rock five-spice seasonings.", + "timestamps": { + "start": "00:11:18", + "end": "00:11:47" + } + }, + { + "description": "You added condiments.", + "timestamps": { + "start": "00:12:33", + "end": "00:13:01" + } + } + ], + "question_timestamps": "00:13:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You added rock sugar, scallion, ginger, garlic, five-spice seasoning, and seasonings. The wok is to your back-left.", + "B": "You added scallion and ginger.", + "C": "You added garlic and seasonings.", + "D": "You added rock sugar, scallion, ginger, and garlic, but didn’t put in five-spice seasoning or extra seasonings.", + "E": "You added rock sugar, scallion, ginger, garlic, and seasonings, leaving out the five-spice seasoning entirely." + }, + "answer_label": "A", + "answer_text": "You added rock sugar, scallion, ginger, garlic, five-spice seasoning, and seasonings. The wok is to your back-left.", + "q_id": "P34_104_1_T2", + "correct_answer": "You added rock sugar, scallion, ginger, garlic, five-spice seasoning, and seasonings. The wok is to your back-left.", + "original_options": { + "A": "You added rock sugar, scallion, ginger, garlic, five-spice seasoning, and seasonings. The wok is to your back-left.", + "B": "You added scallion and ginger.", + "C": "You added garlic and seasonings.", + "D": "You added rock sugar, scallion, ginger, and garlic, but didn’t put in five-spice seasoning or extra seasonings.", + "E": "You added rock sugar, scallion, ginger, garlic, and seasonings, leaving out the five-spice seasoning entirely." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 15 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was stir-frying the pork ribs, what did I add to the wok?", + "answer": "You added rock sugar, scallions, ginger, garlic, and five-spice seasonings and condiments.", + "question_chinese": "我在炒排骨的时候，我在炒锅里加入了什么东西？", + "answer_chinese": "你加入了冰糖，葱姜蒜和五香调料和调味料。", + "evidence": [ + { + "description": "You added rock sugar.", + "timestamps": { + "start": "00:05:28", + "end": "00:07:20" + } + }, + { + "description": "You added scallions, ginger, garlic.", + "timestamps": { + "start": "00:10:49", + "end": "00:10:54" + } + }, + { + "description": "You added rock five-spice seasonings.", + "timestamps": { + "start": "00:11:18", + "end": "00:11:47" + } + }, + { + "description": "You added condiments.", + "timestamps": { + "start": "00:12:33", + "end": "00:13:01" + } + } + ], + "question_timestamps": "00:14:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You added soy sauce and scallions.", + "B": "You added honey and garlic.", + "C": "You added rock sugar, scallions, ginger, garlic, and five-spice seasonings and condiments.", + "D": "You added rock sugar, shallots, ginger, garlic, and star anise to season the ribs.", + "E": "You added brown sugar, scallions, ginger, garlic, and soy sauce, then stirred to coat everything evenly." + }, + "answer_label": "C", + "answer_text": "You added rock sugar, scallions, ginger, garlic, and five-spice seasonings and condiments.", + "q_id": "P34_104_2", + "correct_answer": "You added rock sugar, scallions, ginger, garlic, and five-spice seasonings and condiments.", + "original_options": { + "A": "You added soy sauce and scallions.", + "B": "You added honey and garlic.", + "C": "You added rock sugar, scallions, ginger, garlic, and five-spice seasonings and condiments.", + "D": "You added rock sugar, shallots, ginger, garlic, and star anise to season the ribs.", + "E": "You added brown sugar, scallions, ginger, garlic, and soy sauce, then stirred to coat everything evenly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 16 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_5", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "After I finished stir-frying the pork ribs, where did I get the kitchen knife?", + "answer": "You took the kitchen knife from the counter to the sink.", + "question_chinese": "我炒完排骨后，我将菜刀从哪里拿到了哪里？", + "answer_chinese": "你将菜刀从柜台上拿到了洗水槽旁。", + "evidence": [ + { + "description": "The kitchen knife is on the counter.", + "timestamps": { + "start": "00:14:19", + "end": "00:14:21" + } + }, + { + "description": "You took the kitchen knife to the sink.", + "timestamps": { + "start": "00:14:26", + "end": "00:14:39" + } + } + ], + "question_timestamps": "00:14:48", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You moved it from the sink to counter.", + "B": "You took the kitchen knife from the counter to the sink.", + "C": "You took it from the drawer to counter.", + "D": "You picked the kitchen knife up from the cabinet under the sink and brought it to the countertop.", + "E": "You got the kitchen knife from the rack beside the refrigerator and carried it over to the prep area." + }, + "answer_label": "B", + "answer_text": "You took the kitchen knife from the counter to the sink.", + "q_id": "P34_104_5", + "correct_answer": "You took the kitchen knife from the counter to the sink.", + "original_options": { + "A": "You moved it from the sink to counter.", + "B": "You took the kitchen knife from the counter to the sink.", + "C": "You took it from the drawer to counter.", + "D": "You picked the kitchen knife up from the cabinet under the sink and brought it to the countertop.", + "E": "You got the kitchen knife from the rack beside the refrigerator and carried it over to the prep area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 19 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_6", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "After I finished stir-frying the pork ribs, where did I get the kitchen knife?", + "answer": "You took the kitchen knife from the counter to the sink.", + "question_chinese": "我炒完排骨后，我将菜刀从哪里拿到了哪里？", + "answer_chinese": "你将菜刀从柜台上拿到了洗水槽旁。", + "evidence": [ + { + "description": "The kitchen knife is on the counter.", + "timestamps": { + "start": "00:14:19", + "end": "00:14:21" + } + }, + { + "description": "You took the kitchen knife to the sink.", + "timestamps": { + "start": "00:14:26", + "end": "00:14:39" + } + } + ], + "question_timestamps": "00:15:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took it from the table to fridge.", + "B": "You took the kitchen knife from the counter to the sink.", + "C": "You got it from the drawer by stove.", + "D": "You picked the kitchen knife up from the cutting board and put it away in the cabinet.", + "E": "You grabbed the kitchen knife from the shelf above the sink and brought it down to the counter." + }, + "answer_label": "B", + "answer_text": "You took the kitchen knife from the counter to the sink.", + "q_id": "P34_104_6", + "correct_answer": "You took the kitchen knife from the counter to the sink.", + "original_options": { + "A": "You took it from the table to fridge.", + "B": "You took the kitchen knife from the counter to the sink.", + "C": "You got it from the drawer by stove.", + "D": "You picked the kitchen knife up from the cutting board and put it away in the cabinet.", + "E": "You grabbed the kitchen knife from the shelf above the sink and brought it down to the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 18 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_11", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put my Minion mug?", + "answer": "Your Minion mug is placed in the draining basket.", + "question_chinese": "我的小黄人马克杯是放在哪里的？", + "answer_chinese": "你的小黄人马克杯放在沥水篮里。", + "evidence": [ + { + "description": "Your Minion mug is placed in the draining basket.", + "timestamps": { + "start": "00:13:53", + "end": "00:13:57" + } + } + ], + "question_timestamps": "00:14:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Your Minion mug is on the counter.", + "B": "Your Minion mug is placed in the draining basket.", + "C": "Your Minion mug is on the dining table.", + "D": "Your Minion mug is stored on the pantry shelf, away from the sink and drying area.", + "E": "Your Minion mug is sitting in the kitchen sink, not placed in the draining basket." + }, + "answer_label": "B", + "answer_text": "Your Minion mug is placed in the draining basket.", + "q_id": "P34_104_11", + "correct_answer": "Your Minion mug is placed in the draining basket.", + "original_options": { + "A": "Your Minion mug is on the counter.", + "B": "Your Minion mug is placed in the draining basket.", + "C": "Your Minion mug is on the dining table.", + "D": "Your Minion mug is stored on the pantry shelf, away from the sink and drying area.", + "E": "Your Minion mug is sitting in the kitchen sink, not placed in the draining basket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 15 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_12", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put my Minion mug?", + "answer": "Your Minion mug is placed in the draining basket.", + "question_chinese": "我的小黄人马克杯是放在哪里的？", + "answer_chinese": "你的小黄人马克杯放在沥水篮里。", + "evidence": [ + { + "description": "Your Minion mug is placed in the draining basket.", + "timestamps": { + "start": "00:13:53", + "end": "00:13:57" + } + } + ], + "question_timestamps": "00:15:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Your Minion mug is on the counter.", + "B": "Your Minion mug is in the sink.", + "C": "Your Minion mug is sitting on the windowsill above the kitchen sink instead of in the basket.", + "D": "Your Minion mug is placed on the drying rack next to the sink, not in the draining basket.", + "E": "Your Minion mug is placed in the draining basket." + }, + "answer_label": "E", + "answer_text": "Your Minion mug is placed in the draining basket.", + "q_id": "P34_104_12", + "correct_answer": "Your Minion mug is placed in the draining basket.", + "original_options": { + "A": "Your Minion mug is on the counter.", + "B": "Your Minion mug is in the sink.", + "C": "Your Minion mug is sitting on the windowsill above the kitchen sink instead of in the basket.", + "D": "Your Minion mug is placed on the drying rack next to the sink, not in the draining basket.", + "E": "Your Minion mug is placed in the draining basket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_5_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "After I finished using the seasoning packet, how many clips did I use to seal it?", + "answer": "You used zero clips. The seasoning packet is in front of you.", + "question_chinese": "用完调料袋后我用了几个夹子给调料袋封口？", + "answer_chinese": "你用了零个夹子。调料袋在你的前面。", + "evidence": [ + { + "description": "You are sealing the other seasoning bags.", + "timestamps": { + "start": "00:14:52", + "end": "00:15:05" + } + } + ], + "question_timestamps": "00:15:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used one clip.", + "B": "You used two clips.", + "C": "You used three clips to seal it, and the seasoning packet is still in front of you.", + "D": "You used zero clips. The seasoning packet is in front of you.", + "E": "You used four clips to seal the packet, and it remains positioned in front of you." + }, + "answer_label": "D", + "answer_text": "You used zero clips. The seasoning packet is in front of you.", + "q_id": "P34_104_5_T1", + "correct_answer": "You used zero clips. The seasoning packet is in front of you.", + "original_options": { + "A": "You used one clip.", + "B": "You used two clips.", + "C": "You used three clips to seal it, and the seasoning packet is still in front of you.", + "D": "You used zero clips. The seasoning packet is in front of you.", + "E": "You used four clips to seal the packet, and it remains positioned in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + }, + { + "video_uid": "P34_104", + "video_length": "00:15:12", + "qid": "P34_104_5_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "After I finished using the seasoning packet, how many clips did I use to seal it?", + "answer": "You only used one red clip. The seasoning packet is to your front-left.", + "question_chinese": "用完调料袋后我用了几个夹子给调料袋封口？", + "answer_chinese": "你只用了一个红色夹子。调料袋在你的左前方。", + "evidence": [ + { + "description": "You only used a red clip.", + "timestamps": { + "start": "00:14:46", + "end": "00:14:50" + } + }, + { + "description": "You are sealing the other seasoning bags.", + "timestamps": { + "start": "00:14:52", + "end": "00:15:05" + } + } + ], + "question_timestamps": "00:15:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used zero red clips.", + "B": "You used two red clips.", + "C": "You used three red clips to seal it, and the seasoning packet is still to your front-left.", + "D": "You used four red clips to seal it, with the seasoning packet positioned to your front-left.", + "E": "You only used one red clip. The seasoning packet is to your front-left." + }, + "answer_label": "E", + "answer_text": "You only used one red clip. The seasoning packet is to your front-left.", + "q_id": "P34_104_5_T2", + "correct_answer": "You only used one red clip. The seasoning packet is to your front-left.", + "original_options": { + "A": "You used zero red clips.", + "B": "You used two red clips.", + "C": "You used three red clips to seal it, and the seasoning packet is still to your front-left.", + "D": "You used four red clips to seal it, with the seasoning packet positioned to your front-left.", + "E": "You only used one red clip. The seasoning packet is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + } +] diff --git a/QAs/P34_108-mcq.json b/QAs/P34_108-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b665e1e8e9d2d1ee98d7556c361d3aa7b08e71ab --- /dev/null +++ b/QAs/P34_108-mcq.json @@ -0,0 +1,950 @@ +[ + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_3_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands?", + "answer": "Yes, you can—the sink is in front of you.", + "question_chinese": "我可以洗到手么？", + "answer_chinese": "你能，水池在你前面。", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes, you can—the sink is in front of you.", + "B": "No, you can’t.", + "C": "Yes, you can—sink left.", + "D": "Yes, you can wash your hands because the sink is to your right, within easy reach.", + "E": "Yes, you can wash your hands, but the sink is behind you rather than directly in front." + }, + "answer_label": "A", + "answer_text": "Yes, you can—the sink is in front of you.", + "q_id": "P34_108_3_T1", + "correct_answer": "Yes, you can—the sink is in front of you.", + "original_options": { + "A": "Yes, you can—the sink is in front of you.", + "B": "No, you can’t.", + "C": "Yes, you can—sink left.", + "D": "Yes, you can wash your hands because the sink is to your right, within easy reach.", + "E": "Yes, you can wash your hands, but the sink is behind you rather than directly in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 17 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_8", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands?", + "answer": "You can't.", + "question_chinese": "我可以洗到手么", + "answer_chinese": "你不能。", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:11" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:44", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "You can't.", + "q_id": "P34_108_8", + "correct_answer": "You can't.", + "original_options": { + "A": "You can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_3_T2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I wash my hands?", + "answer": "No, you can’t—the sink is to your back-right.", + "question_chinese": "我可以洗到手么？", + "answer_chinese": "你不能，水池在你右后方", + "evidence": [ + { + "description": "Location of the sink.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes, you can—the sink is nearby.", + "B": "No, you can’t—the sink is to your back-right.", + "C": "No, you can’t—the sink is back-left.", + "D": "Yes, you can—the sink is directly in front of you, so you can reach it easily from where you are.", + "E": "Yes, you can—the sink is to your front-right, so you can walk up and wash your hands." + }, + "answer_label": "B", + "answer_text": "No, you can’t—the sink is to your back-right.", + "q_id": "P34_108_3_T2", + "correct_answer": "No, you can’t—the sink is to your back-right.", + "original_options": { + "A": "Yes, you can—the sink is nearby.", + "B": "No, you can’t—the sink is to your back-right.", + "C": "No, you can’t—the sink is back-left.", + "D": "Yes, you can—the sink is directly in front of you, so you can reach it easily from where you are.", + "E": "Yes, you can—the sink is to your front-right, so you can walk up and wash your hands." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 21, + 18 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_4", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get Yangchun noodles?", + "answer": "Turn around and walk straight for a certain distance; it will be on your right - hand side.", + "question_chinese": "我如何拿到阳春面", + "answer_chinese": "向后转直走一段距离，它在你的右手边", + "evidence": [ + { + "description": "The location of Yangchun noodles.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Walking trajectory.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:06" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn around and walk straight for a certain distance; it will be on your right - hand side.", + "B": "Turn around; it’ll be on your left.", + "C": "Turn around; it’ll be ahead.", + "D": "Turn around and walk straight for a certain distance; it will be on your left-hand side.", + "E": "Turn right first, then walk straight for a certain distance; it will be on your right-hand side." + }, + "answer_label": "A", + "answer_text": "Turn around and walk straight for a certain distance; it will be on your right - hand side.", + "q_id": "P34_108_4", + "correct_answer": "Turn around and walk straight for a certain distance; it will be on your right - hand side.", + "original_options": { + "A": "Turn around and walk straight for a certain distance; it will be on your right - hand side.", + "B": "Turn around; it’ll be on your left.", + "C": "Turn around; it’ll be ahead.", + "D": "Turn around and walk straight for a certain distance; it will be on your left-hand side.", + "E": "Turn right first, then walk straight for a certain distance; it will be on your right-hand side." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 16, + 17 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get Yangchun noodles?", + "answer": "Turn around and walk straight for a certain distance; it will be on your right.", + "question_chinese": "我如何拿到阳春面", + "answer_chinese": "向后转直走一段距离，它在你的右边", + "evidence": [ + { + "description": "The location of Yangchun noodles.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Current location.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:25" + } + }, + { + "description": "Walking trajectory.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn around; it will be on your left.", + "B": "Turn around; it will be straight ahead.", + "C": "Turn around and walk straight for a certain distance; it will be on your left side.", + "D": "Turn around and walk straight for a certain distance; it will be behind you on the left.", + "E": "Turn around and walk straight for a certain distance; it will be on your right." + }, + "answer_label": "E", + "answer_text": "Turn around and walk straight for a certain distance; it will be on your right.", + "q_id": "P34_108_3", + "correct_answer": "Turn around and walk straight for a certain distance; it will be on your right.", + "original_options": { + "A": "Turn around; it will be on your left.", + "B": "Turn around; it will be straight ahead.", + "C": "Turn around and walk straight for a certain distance; it will be on your left side.", + "D": "Turn around and walk straight for a certain distance; it will be behind you on the left.", + "E": "Turn around and walk straight for a certain distance; it will be on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 17 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_13", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the door?", + "answer": "Turn around and go straight, then the door is on your right.", + "question_chinese": "我如何走到门口", + "answer_chinese": "向后转直走后，门在右手边", + "evidence": [ + { + "description": "The position of the door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Motion trajectory.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn around; the door is on your left.", + "B": "Turn around and go straight, then the door is on your right.", + "C": "Go straight; the door is ahead.", + "D": "Turn around and go straight, then the door is on your left-hand side instead of the right.", + "E": "Turn around first, then walk straight a short distance; the door will be directly in front of you." + }, + "answer_label": "B", + "answer_text": "Turn around and go straight, then the door is on your right.", + "q_id": "P34_108_13", + "correct_answer": "Turn around and go straight, then the door is on your right.", + "original_options": { + "A": "Turn around; the door is on your left.", + "B": "Turn around and go straight, then the door is on your right.", + "C": "Go straight; the door is ahead.", + "D": "Turn around and go straight, then the door is on your left-hand side instead of the right.", + "E": "Turn around first, then walk straight a short distance; the door will be directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 17, + 18 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_14", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the door?", + "answer": "Turn right and then go straight; the door will be on your right.", + "question_chinese": "我如何走到门口", + "answer_chinese": "右转后直走，门在你右方。", + "evidence": [ + { + "description": "The position of the door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Motion trajectory(1).", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + }, + { + "description": "Motion trajectory(2).", + "timestamps": { + "start": "00:00:32", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right; the door is on your left.", + "B": "Turn left; the door is on your right.", + "C": "Turn right and then go straight; the door will be on your left-hand side.", + "D": "Go straight first and then turn right; the door will be on your right-hand side.", + "E": "Turn right and then go straight; the door will be on your right." + }, + "answer_label": "E", + "answer_text": "Turn right and then go straight; the door will be on your right.", + "q_id": "P34_108_14", + "correct_answer": "Turn right and then go straight; the door will be on your right.", + "original_options": { + "A": "Turn right; the door is on your left.", + "B": "Turn left; the door is on your right.", + "C": "Turn right and then go straight; the door will be on your left-hand side.", + "D": "Go straight first and then turn right; the door will be on your right-hand side.", + "E": "Turn right and then go straight; the door will be on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 14, + 15 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_5_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates are next to the sink?", + "answer": "Two, and the sink is in front of you.", + "question_chinese": "我水池旁有几个盘子？", + "answer_chinese": "两个，水池在你正前方。", + "evidence": [ + { + "description": "There are two plates by the sink.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Zero plates.", + "B": "One plate.", + "C": "Two, and the sink is in front of you.", + "D": "Three plates are next to the sink, and the sink is in front of you.", + "E": "Four plates are next to the sink, and the sink is in front of you." + }, + "answer_label": "C", + "answer_text": "Two, and the sink is in front of you.", + "q_id": "P34_108_5_T1", + "correct_answer": "Two, and the sink is in front of you.", + "original_options": { + "A": "Zero plates.", + "B": "One plate.", + "C": "Two, and the sink is in front of you.", + "D": "Three plates are next to the sink, and the sink is in front of you.", + "E": "Four plates are next to the sink, and the sink is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_5_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates are next to the sink?", + "answer": "Two, and the sink is to your right.", + "question_chinese": "我水池旁有几个盘子？", + "answer_chinese": "两个，水池在你右边。", + "evidence": [ + { + "description": "There are two plates by the sink.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Zero, and the sink is to your right.", + "B": "One, and the sink is to your right.", + "C": "Three, and the sink is to your right, with the plates sitting together beside it.", + "D": "Two, and the sink is to your right.", + "E": "Four, and the sink is to your right, and all four plates are stacked next to it." + }, + "answer_label": "D", + "answer_text": "Two, and the sink is to your right.", + "q_id": "P34_108_5_T2", + "correct_answer": "Two, and the sink is to your right.", + "original_options": { + "A": "Zero, and the sink is to your right.", + "B": "One, and the sink is to your right.", + "C": "Three, and the sink is to your right, with the plates sitting together beside it.", + "D": "Two, and the sink is to your right.", + "E": "Four, and the sink is to your right, and all four plates are stacked next to it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 17 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_4_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was originally in my pot?", + "answer": "There is one pair of chopsticks, and the pot is in your hand.", + "question_chinese": "我锅里本来有什么？", + "answer_chinese": "有一双筷子，锅在你手上", + "evidence": [ + { + "description": "There is a pair of chopsticks.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No chopsticks were in the pot.", + "B": "There is one pair of chopsticks, and the pot is in your hand.", + "C": "Two pairs of chopsticks were inside.", + "D": "There were no chopsticks in the pot, and you were holding the pot in your hand.", + "E": "There were two pairs of chopsticks in the pot, and you kept the pot in your hand." + }, + "answer_label": "B", + "answer_text": "There is one pair of chopsticks, and the pot is in your hand.", + "q_id": "P34_108_4_T1", + "correct_answer": "There is one pair of chopsticks, and the pot is in your hand.", + "original_options": { + "A": "No chopsticks were in the pot.", + "B": "There is one pair of chopsticks, and the pot is in your hand.", + "C": "Two pairs of chopsticks were inside.", + "D": "There were no chopsticks in the pot, and you were holding the pot in your hand.", + "E": "There were two pairs of chopsticks in the pot, and you kept the pot in your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_4_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was originally in my pot?", + "answer": "There is one pair of chopsticks, and the pot is in front of you.", + "question_chinese": "我锅里本来有什么？", + "answer_chinese": "有一双筷子，锅在你正前方", + "evidence": [ + { + "description": "There is a pair of chopsticks.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No chopsticks were in the pot.", + "B": "Two pairs of chopsticks were inside.", + "C": "There is one pair of chopsticks, and the pot is in front of you.", + "D": "There were no chopsticks in the pot, and the pot sat in front of you.", + "E": "There were two pairs of chopsticks in the pot, and the pot was placed in front of you." + }, + "answer_label": "C", + "answer_text": "There is one pair of chopsticks, and the pot is in front of you.", + "q_id": "P34_108_4_T2", + "correct_answer": "There is one pair of chopsticks, and the pot is in front of you.", + "original_options": { + "A": "No chopsticks were in the pot.", + "B": "Two pairs of chopsticks were inside.", + "C": "There is one pair of chopsticks, and the pot is in front of you.", + "D": "There were no chopsticks in the pot, and the pot sat in front of you.", + "E": "There were two pairs of chopsticks in the pot, and the pot was placed in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 18 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_11", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates are there beside the sink?", + "answer": "Two.", + "question_chinese": "水池旁有几个盘子？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "There are two plates by the sink.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "One.", + "C": "Three plates were beside the sink on the counter, arranged in a small stack near the faucet.", + "D": "Two.", + "E": "Four plates were sitting beside the sink, lined up along the edge and not inside it." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "P34_108_11", + "correct_answer": "Two.", + "original_options": { + "A": "Zero.", + "B": "One.", + "C": "Three plates were beside the sink on the counter, arranged in a small stack near the faucet.", + "D": "Two.", + "E": "Four plates were sitting beside the sink, lined up along the edge and not inside it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 16 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_12", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates are there beside the sink?", + "answer": "Two.", + "question_chinese": "水池旁有几个盘子？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "There are two plates by the sink.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:53", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Zero.", + "C": "One.", + "D": "Three plates were beside the sink on the counter, arranged in a small stack near the faucet.", + "E": "Four plates were sitting beside the sink, lined up along the edge and not inside it." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "P34_108_12", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "Zero.", + "C": "One.", + "D": "Three plates were beside the sink on the counter, arranged in a small stack near the faucet.", + "E": "Four plates were sitting beside the sink, lined up along the edge and not inside it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 16 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_9", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was originally in the pot?", + "answer": "There is a pair of chopsticks.", + "question_chinese": "锅里本来有什么", + "answer_chinese": "有一双筷子", + "evidence": [ + { + "description": "There is a pair of chopsticks.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There is a bowl of rice.", + "B": "There is a pair of chopsticks.", + "C": "There is a serving tray.", + "D": "There is a lid placed inside the pot, with no chopsticks or other utensils in it at all.", + "E": "There are spoons inside the pot, rather than chopsticks, resting at the bottom of the pot." + }, + "answer_label": "B", + "answer_text": "There is a pair of chopsticks.", + "q_id": "P34_108_9", + "correct_answer": "There is a pair of chopsticks.", + "original_options": { + "A": "There is a bowl of rice.", + "B": "There is a pair of chopsticks.", + "C": "There is a serving tray.", + "D": "There is a lid placed inside the pot, with no chopsticks or other utensils in it at all.", + "E": "There are spoons inside the pot, rather than chopsticks, resting at the bottom of the pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 16 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_10", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was originally in the pot?", + "answer": "There is a pair of chopsticks.", + "question_chinese": "锅里本来有什么", + "answer_chinese": "有一双筷子", + "evidence": [ + { + "description": "There is a pair of chopsticks.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There is a pot of stew.", + "B": "There is a pot lid.", + "C": "You had a spoon and a plate beside the pot, not chopsticks inside it.", + "D": "You set a bowl of rice on the counter, and nothing like chopsticks was in the pot.", + "E": "There is a pair of chopsticks." + }, + "answer_label": "E", + "answer_text": "There is a pair of chopsticks.", + "q_id": "P34_108_10", + "correct_answer": "There is a pair of chopsticks.", + "original_options": { + "A": "There is a pot of stew.", + "B": "There is a pot lid.", + "C": "You had a spoon and a plate beside the pot, not chopsticks inside it.", + "D": "You set a bowl of rice on the counter, and nothing like chopsticks was in the pot.", + "E": "There is a pair of chopsticks." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 17 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_1_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where are the Yangchun noodles?", + "answer": "It was on the table when you first walked in.", + "question_chinese": "阳春面在什么地方？", + "answer_chinese": "在你刚进门时的桌子上", + "evidence": [ + { + "description": "The location of Yangchun noodles.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was on the counter.", + "B": "It was by the sink.", + "C": "You left it on the counter when you first walked in, not on the table.", + "D": "It was on the table when you first walked in.", + "E": "It wasn’t on the table at entry; you placed it there later after setting things down." + }, + "answer_label": "D", + "answer_text": "It was on the table when you first walked in.", + "q_id": "P34_108_1_T1", + "correct_answer": "It was on the table when you first walked in.", + "original_options": { + "A": "It was on the counter.", + "B": "It was by the sink.", + "C": "You left it on the counter when you first walked in, not on the table.", + "D": "It was on the table when you first walked in.", + "E": "It wasn’t on the table at entry; you placed it there later after setting things down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "P34_108", + "video_length": "00:00:53", + "qid": "P34_108_1_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where are the Yangchun noodles?", + "answer": "When you first came in, it was already on the table.", + "question_chinese": "阳春面在什么地方？", + "answer_chinese": "你刚进门时，它就在桌子上。", + "evidence": [ + { + "description": "The location of Yangchun noodles.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was on the counter.", + "B": "It was by the sink.", + "C": "When you first came in, it was on the counter rather than already on the table.", + "D": "It wasn’t already on the table at entry; you put it there later after moving items around.", + "E": "When you first came in, it was already on the table." + }, + "answer_label": "E", + "answer_text": "When you first came in, it was already on the table.", + "q_id": "P34_108_1_T2", + "correct_answer": "When you first came in, it was already on the table.", + "original_options": { + "A": "It was on the counter.", + "B": "It was by the sink.", + "C": "When you first came in, it was on the counter rather than already on the table.", + "D": "It wasn’t already on the table at entry; you put it there later after moving items around.", + "E": "When you first came in, it was already on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + } +] diff --git a/QAs/P34_113-mcq.json b/QAs/P34_113-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ce715c3489ecd37b3c52efd8df1fd798743f434d --- /dev/null +++ b/QAs/P34_113-mcq.json @@ -0,0 +1,764 @@ +[ + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_13", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the door?", + "answer": "Turn left first, then go straight, then turn left again, and then go straight.", + "question_chinese": "我如何走到门跟前", + "answer_chinese": "先左转再直走再左转然后直走", + "evidence": [ + { + "description": "Turn left first, then go straight, then turn left again, and then go straight.", + "timestamps": { + "start": "00:09:45", + "end": "00:09:52" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight.", + "B": "Go straight, then turn left.", + "C": "Turn left first, go straight, then turn right and continue straight to the door.", + "D": "Turn left first, then go straight, then turn left again, and then go straight.", + "E": "Go straight first, then turn left, then turn right, and go straight until you reach the door." + }, + "answer_label": "D", + "answer_text": "Turn left first, then go straight, then turn left again, and then go straight.", + "q_id": "P34_113_13", + "correct_answer": "Turn left first, then go straight, then turn left again, and then go straight.", + "original_options": { + "A": "Turn right, then go straight.", + "B": "Go straight, then turn left.", + "C": "Turn left first, go straight, then turn right and continue straight to the door.", + "D": "Turn left first, then go straight, then turn left again, and then go straight.", + "E": "Go straight first, then turn left, then turn right, and go straight until you reach the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 17 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_14", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the door?", + "answer": "Turn left first and then go straight.", + "question_chinese": "我如何走到门跟前", + "answer_chinese": "先左转再直走", + "evidence": [ + { + "description": "Turn left first, then go straight, then turn left again, and then go straight.", + "timestamps": { + "start": "00:09:45", + "end": "00:09:52" + } + }, + { + "description": "Turn left first and then go straight.", + "timestamps": { + "start": "00:10:59", + "end": "00:11:04" + } + } + ], + "question_timestamps": "00:12:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight.", + "B": "Turn left first and then go straight.", + "C": "Go straight, then turn left.", + "D": "Turn left first, then turn left again instead of going straight to the door.", + "E": "Turn left and then walk backward a few steps rather than continuing straight ahead." + }, + "answer_label": "B", + "answer_text": "Turn left first and then go straight.", + "q_id": "P34_113_14", + "correct_answer": "Turn left first and then go straight.", + "original_options": { + "A": "Turn right, then go straight.", + "B": "Turn left first and then go straight.", + "C": "Go straight, then turn left.", + "D": "Turn left first, then turn left again instead of going straight to the door.", + "E": "Turn left and then walk backward a few steps rather than continuing straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_7", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do to the kitchen waste bin?", + "answer": "You take out the trash.", + "question_chinese": "我对厨余垃圾桶做了什么", + "answer_chinese": "你倒垃圾", + "evidence": [ + { + "description": "You take out the trash.", + "timestamps": { + "start": "00:08:45", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:09:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You cleaned the bin.", + "B": "You take out the trash.", + "C": "You kept it in place.", + "D": "You tied the trash bag closed and left it in the kitchen waste bin.", + "E": "You opened the kitchen waste bin and pushed the garbage down to make more space." + }, + "answer_label": "B", + "answer_text": "You take out the trash.", + "q_id": "P34_113_7", + "correct_answer": "You take out the trash.", + "original_options": { + "A": "You cleaned the bin.", + "B": "You take out the trash.", + "C": "You kept it in place.", + "D": "You tied the trash bag closed and left it in the kitchen waste bin.", + "E": "You opened the kitchen waste bin and pushed the garbage down to make more space." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 15 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_8", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do to the kitchen waste bin?", + "answer": "You pack the garbage and then put a garbage bag over it.", + "question_chinese": "我对厨余垃圾桶做了什么", + "answer_chinese": "你打包垃圾然后套垃圾袋", + "evidence": [ + { + "description": "You take out the trash.", + "timestamps": { + "start": "00:08:45", + "end": "00:09:05" + } + }, + { + "description": "You pack the garbage.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:50" + } + }, + { + "description": "You put on the garbage bag.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:25" + } + } + ], + "question_timestamps": "00:12:27", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You removed the bag.", + "B": "You left it unbagged.", + "C": "You packed the garbage, but you tied the bag shut and set it aside instead of covering the bin.", + "D": "You gathered the garbage, then placed the bag inside the bin rather than putting a garbage bag over it.", + "E": "You pack the garbage and then put a garbage bag over it." + }, + "answer_label": "E", + "answer_text": "You pack the garbage and then put a garbage bag over it.", + "q_id": "P34_113_8", + "correct_answer": "You pack the garbage and then put a garbage bag over it.", + "original_options": { + "A": "You removed the bag.", + "B": "You left it unbagged.", + "C": "You packed the garbage, but you tied the bag shut and set it aside instead of covering the bin.", + "D": "You gathered the garbage, then placed the bag inside the bin rather than putting a garbage bag over it.", + "E": "You pack the garbage and then put a garbage bag over it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 19 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_4", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do when I opened the cabinet?", + "answer": "You took the fresh-keeping bag.", + "question_chinese": "我打开橱柜做了什么", + "answer_chinese": "你拿了保鲜袋", + "evidence": [ + { + "description": "You took the fresh-keeping bag.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:03:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took the jar.", + "B": "You grabbed the bowl.", + "C": "You opened the cabinet and took the empty storage box out from the shelf.", + "D": "You took the fresh-keeping bag.", + "E": "You opened the cabinet and pulled out a metal canister, leaving the fresh-keeping bag behind." + }, + "answer_label": "D", + "answer_text": "You took the fresh-keeping bag.", + "q_id": "P34_113_4", + "correct_answer": "You took the fresh-keeping bag.", + "original_options": { + "A": "You took the jar.", + "B": "You grabbed the bowl.", + "C": "You opened the cabinet and took the empty storage box out from the shelf.", + "D": "You took the fresh-keeping bag.", + "E": "You opened the cabinet and pulled out a metal canister, leaving the fresh-keeping bag behind." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_3", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do when I opened the cabinet?", + "answer": "You put the plastic wrap back.", + "question_chinese": "我打开橱柜做了什么", + "answer_chinese": "你放回了保鲜袋", + "evidence": [ + { + "description": "You took the fresh-keeping bag.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:55" + } + }, + { + "description": "You put the plastic wrap back.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:06:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took it out.", + "B": "You moved it aside.", + "C": "You put the plastic wrap back.", + "D": "You opened the cabinet and pulled the plastic wrap out to use it, instead of putting it back.", + "E": "You opened the cabinet and moved the paper towels onto the shelf, not the plastic wrap." + }, + "answer_label": "C", + "answer_text": "You put the plastic wrap back.", + "q_id": "P34_113_3", + "correct_answer": "You put the plastic wrap back.", + "original_options": { + "A": "You took it out.", + "B": "You moved it aside.", + "C": "You put the plastic wrap back.", + "D": "You opened the cabinet and pulled the plastic wrap out to use it, instead of putting it back.", + "E": "You opened the cabinet and moved the paper towels onto the shelf, not the plastic wrap." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 16 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_9", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after washing the dishes?", + "answer": "You cleaned the stove.", + "question_chinese": "我洗完碗干了什么", + "answer_chinese": "你擦了灶台", + "evidence": [ + { + "description": "You cleaned the stove.", + "timestamps": { + "start": "00:06:25", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:09:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You wiped the counter.", + "B": "You took out recycling.", + "C": "After washing the dishes, you dried the countertop and put items back instead of cleaning the stove.", + "D": "You cleaned the stove.", + "E": "Once the dishes were washed, you organized the pantry shelves rather than scrubbing the stovetop." + }, + "answer_label": "D", + "answer_text": "You cleaned the stove.", + "q_id": "P34_113_9", + "correct_answer": "You cleaned the stove.", + "original_options": { + "A": "You wiped the counter.", + "B": "You took out recycling.", + "C": "After washing the dishes, you dried the countertop and put items back instead of cleaning the stove.", + "D": "You cleaned the stove.", + "E": "Once the dishes were washed, you organized the pantry shelves rather than scrubbing the stovetop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 15 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_10", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after washing the dishes?", + "answer": "You sorted out the trash can.", + "question_chinese": "我洗完碗干了什么", + "answer_chinese": "你整理了垃圾桶", + "evidence": [ + { + "description": "You cleaned the stove.", + "timestamps": { + "start": "00:06:25", + "end": "00:06:50" + } + }, + { + "description": "You sorted out the trash can.", + "timestamps": { + "start": "00:09:30", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:12:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You sorted out the trash can.", + "B": "You put dishes away.", + "C": "You left dishes on counter.", + "D": "After washing the dishes, you carried the trash bag to the garage instead of sorting the trash can.", + "E": "Once you finished washing, you placed the pot lid in the sink and did not sort the trash." + }, + "answer_label": "A", + "answer_text": "You sorted out the trash can.", + "q_id": "P34_113_10", + "correct_answer": "You sorted out the trash can.", + "original_options": { + "A": "You sorted out the trash can.", + "B": "You put dishes away.", + "C": "You left dishes on counter.", + "D": "After washing the dishes, you carried the trash bag to the garage instead of sorting the trash can.", + "E": "Once you finished washing, you placed the pot lid in the sink and did not sort the trash." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 18 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_5", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the dishwashing liquid?", + "answer": "You can.", + "question_chinese": "我能够到洗碗液吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:03:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "P34_113_5", + "correct_answer": "You can.", + "original_options": { + "A": "You cannot.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_6", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the dishwashing liquid?", + "answer": "You can.", + "question_chinese": "我能够到洗碗液吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:55" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:03:43", + "end": "00:03:53" + } + } + ], + "question_timestamps": "00:06:13", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You cannot." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "P34_113_6", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_12", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get a kitchen knife?", + "answer": "You can.", + "question_chinese": "我能拿到菜刀吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:06:47", + "end": "00:06:49" + } + } + ], + "question_timestamps": "00:09:21", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "P34_113_12", + "correct_answer": "You can.", + "original_options": { + "A": "You cannot.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_11", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get a kitchen knife?", + "answer": "You can.", + "question_chinese": "我能拿到菜刀吗", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can.", + "timestamps": { + "start": "00:06:47", + "end": "00:06:49" + } + }, + { + "description": "You can.", + "timestamps": { + "start": "00:09:53", + "end": "00:09:54" + } + } + ], + "question_timestamps": "00:12:27", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can.", + "B": "You cannot." + }, + "answer_label": "A", + "answer_text": "You can.", + "q_id": "P34_113_11", + "correct_answer": "You can.", + "original_options": { + "A": "You can.", + "B": "You cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chopsticks and how many spoons are there in the sink?", + "answer": "Two chopsticks and a spoon.", + "question_chinese": "洗碗池里有多少根筷子多少个勺子", + "answer_chinese": "两根筷子和一个勺子", + "evidence": [ + { + "description": "Two chopsticks and a spoon.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:03:07", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two chopsticks and a spoon.", + "B": "One chopstick and one spoon.", + "C": "Two chopsticks and two spoons.", + "D": "There are three chopsticks and one spoon lying together at the bottom of the sink.", + "E": "You have one chopstick and two spoons sitting in the sink near the drain." + }, + "answer_label": "A", + "answer_text": "Two chopsticks and a spoon.", + "q_id": "P34_113_1", + "correct_answer": "Two chopsticks and a spoon.", + "original_options": { + "A": "Two chopsticks and a spoon.", + "B": "One chopstick and one spoon.", + "C": "Two chopsticks and two spoons.", + "D": "There are three chopsticks and one spoon lying together at the bottom of the sink.", + "E": "You have one chopstick and two spoons sitting in the sink near the drain." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "P34_113", + "video_length": "00:12:27", + "qid": "P34_113_2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chopsticks and how many spoons are there in the sink?", + "answer": "zero chopsticks and spoons.", + "question_chinese": "洗碗池里有多少根筷子多少个勺子", + "answer_chinese": "0个筷子和勺子", + "evidence": [ + { + "description": "Two chopsticks and a spoon.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "0 chopsticks and spoons.", + "timestamps": { + "start": "00:03:50", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:06:13", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One chopstick and no spoons.", + "B": "zero chopsticks and spoons.", + "C": "No chopsticks and one spoon.", + "D": "There are two chopsticks and one spoon resting together in the sink basin.", + "E": "You can see three chopsticks in the sink, with no spoons present at all." + }, + "answer_label": "B", + "answer_text": "zero chopsticks and spoons.", + "q_id": "P34_113_2", + "correct_answer": "zero chopsticks and spoons.", + "original_options": { + "A": "One chopstick and no spoons.", + "B": "zero chopsticks and spoons.", + "C": "No chopsticks and one spoon.", + "D": "There are two chopsticks and one spoon resting together in the sink basin.", + "E": "You can see three chopsticks in the sink, with no spoons present at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + } +] diff --git a/QAs/P35_101-mcq.json b/QAs/P35_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..210815a8956c00e4b27142e7037da582f0c76f92 --- /dev/null +++ b/QAs/P35_101-mcq.json @@ -0,0 +1,1654 @@ +[ + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_2_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the refrigerator relative to me?", + "answer": "The refrigerator is in front of you.", + "question_chinese": "冰箱在相对于我的哪里？", + "answer_chinese": "冰箱在相对于你的正前方。", + "evidence": [ + { + "description": "The refrigerator is right in front of you.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is behind you.", + "B": "The refrigerator is to your left.", + "C": "The refrigerator is in front of you.", + "D": "The refrigerator is to your right as you face forward in the kitchen area.", + "E": "The refrigerator is above you, higher than your eye level on the wall." + }, + "answer_label": "C", + "answer_text": "The refrigerator is in front of you.", + "q_id": "P35_101_2_T1", + "correct_answer": "The refrigerator is in front of you.", + "original_options": { + "A": "The refrigerator is behind you.", + "B": "The refrigerator is to your left.", + "C": "The refrigerator is in front of you.", + "D": "The refrigerator is to your right as you face forward in the kitchen area.", + "E": "The refrigerator is above you, higher than your eye level on the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 13 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_2_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the refrigerator relative to me?", + "answer": "The refrigerator is to your left.", + "question_chinese": "冰箱在相对于我的哪里？", + "answer_chinese": "冰箱在相对于你的左边。", + "evidence": [ + { + "description": "You turn around and come to the right, so the refrigerator is on your left.", + "timestamps": { + "start": "00:01:12", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is in front of you.", + "B": "The refrigerator is behind you.", + "C": "The refrigerator is to your right, not on your left side as you look forward.", + "D": "The refrigerator is to your left.", + "E": "The refrigerator is above you, positioned higher than where you are looking now." + }, + "answer_label": "D", + "answer_text": "The refrigerator is to your left.", + "q_id": "P35_101_2_T2", + "correct_answer": "The refrigerator is to your left.", + "original_options": { + "A": "The refrigerator is in front of you.", + "B": "The refrigerator is behind you.", + "C": "The refrigerator is to your right, not on your left side as you look forward.", + "D": "The refrigerator is to your left.", + "E": "The refrigerator is above you, positioned higher than where you are looking now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 13 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_4", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the relative position of the refrigerator?", + "answer": "The refrigerator is behind the sink and to the left of the microwave oven.", + "question_chinese": "冰箱的相对位置在哪里？", + "answer_chinese": "冰箱在洗水槽的后面和微波炉的左边。", + "evidence": [ + { + "description": "The refrigerator is behind the sink.", + "timestamps": { + "start": "00:00:57", + "end": "00:01:01" + } + }, + { + "description": "The refrigerator is to the left of the microwave oven.", + "timestamps": { + "start": "00:01:12", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:02:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of the sink.", + "B": "The refrigerator is behind the sink and to the left of the microwave oven.", + "C": "Right of the microwave.", + "D": "The refrigerator is in front of the sink and to the left of the microwave oven.", + "E": "The refrigerator is behind the sink but sits to the right of the microwave oven instead." + }, + "answer_label": "B", + "answer_text": "The refrigerator is behind the sink and to the left of the microwave oven.", + "q_id": "P35_101_4", + "correct_answer": "The refrigerator is behind the sink and to the left of the microwave oven.", + "original_options": { + "A": "In front of the sink.", + "B": "The refrigerator is behind the sink and to the left of the microwave oven.", + "C": "Right of the microwave.", + "D": "The refrigerator is in front of the sink and to the left of the microwave oven.", + "E": "The refrigerator is behind the sink but sits to the right of the microwave oven instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_3", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the relative position of the refrigerator?", + "answer": "The refrigerator is behind the sink and to the left of the microwave oven.", + "question_chinese": "冰箱的相对位置在哪里？", + "answer_chinese": "冰箱在洗水槽的后面和微波炉的左边。", + "evidence": [ + { + "description": "The refrigerator is behind the sink.", + "timestamps": { + "start": "00:00:57", + "end": "00:01:01" + } + }, + { + "description": "The refrigerator is to the left of the microwave oven.", + "timestamps": { + "start": "00:01:12", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:04:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of the sink.", + "B": "Right of the microwave.", + "C": "The refrigerator is behind the sink and positioned to the right of the microwave oven.", + "D": "The refrigerator is in front of the sink while also being to the left of the microwave oven.", + "E": "The refrigerator is behind the sink and to the left of the microwave oven." + }, + "answer_label": "E", + "answer_text": "The refrigerator is behind the sink and to the left of the microwave oven.", + "q_id": "P35_101_3", + "correct_answer": "The refrigerator is behind the sink and to the left of the microwave oven.", + "original_options": { + "A": "In front of the sink.", + "B": "Right of the microwave.", + "C": "The refrigerator is behind the sink and positioned to the right of the microwave oven.", + "D": "The refrigerator is in front of the sink while also being to the left of the microwave oven.", + "E": "The refrigerator is behind the sink and to the left of the microwave oven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 18 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_15", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pairs of knives and forks are there in the sink when I'm doing the dishes?", + "answer": "There is a pair of knives and forks.", + "question_chinese": "在我洗碗时，洗水槽里有几副刀叉？", + "answer_chinese": "有一副刀叉。", + "evidence": [ + { + "description": "There is a pair of knives and forks.", + "timestamps": { + "start": "00:06:11", + "end": "00:06:16" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two pairs.", + "B": "Three pairs.", + "C": "There are two pairs of knives and forks in the sink while you are doing the dishes.", + "D": "There is a pair of knives and forks.", + "E": "There are three pairs of knives and forks sitting together in the sink as you wash dishes." + }, + "answer_label": "D", + "answer_text": "There is a pair of knives and forks.", + "q_id": "P35_101_15", + "correct_answer": "There is a pair of knives and forks.", + "original_options": { + "A": "Two pairs.", + "B": "Three pairs.", + "C": "There are two pairs of knives and forks in the sink while you are doing the dishes.", + "D": "There is a pair of knives and forks.", + "E": "There are three pairs of knives and forks sitting together in the sink as you wash dishes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_16", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pairs of knives and forks are there in the sink when I'm doing the dishes?", + "answer": "There is a pair of knives and forks.", + "question_chinese": "在我洗碗时，洗水槽里有几副刀叉？", + "answer_chinese": "有一副刀叉。", + "evidence": [ + { + "description": "There is a pair of knives and forks.", + "timestamps": { + "start": "00:06:11", + "end": "00:06:16" + } + } + ], + "question_timestamps": "00:06:22", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero pairs.", + "B": "Two pairs.", + "C": "There is a pair of knives and forks.", + "D": "There are no pairs of knives and forks in the sink when you are doing the dishes.", + "E": "There are two pairs of knives and forks in the sink at the time you are washing dishes." + }, + "answer_label": "C", + "answer_text": "There is a pair of knives and forks.", + "q_id": "P35_101_16", + "correct_answer": "There is a pair of knives and forks.", + "original_options": { + "A": "Zero pairs.", + "B": "Two pairs.", + "C": "There is a pair of knives and forks.", + "D": "There are no pairs of knives and forks in the sink when you are doing the dishes.", + "E": "There are two pairs of knives and forks in the sink at the time you are washing dishes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_7_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can relative to me?", + "answer": "The trash can is in front of you.", + "question_chinese": "垃圾桶在相对于我的哪里？", + "answer_chinese": "垃圾桶在相对于你的正前方。", + "evidence": [ + { + "description": "The trash can is right in front of you.", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The trash can is in front of you.", + "D": "The trash can is to your right, not directly in front of where you are standing.", + "E": "The trash can is behind you, so you would need to turn around to reach it." + }, + "answer_label": "C", + "answer_text": "The trash can is in front of you.", + "q_id": "P35_101_7_T1", + "correct_answer": "The trash can is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The trash can is in front of you.", + "D": "The trash can is to your right, not directly in front of where you are standing.", + "E": "The trash can is behind you, so you would need to turn around to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 16, + 16 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_7_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can relative to me?", + "answer": "The trash can is to your right.", + "question_chinese": "垃圾桶在相对于我的哪里？", + "answer_chinese": "垃圾桶在相对于你的右边。", + "evidence": [ + { + "description": "The trash can is on your right.", + "timestamps": { + "start": "00:01:28", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The trash can is to your left rather than on your right side.", + "D": "The trash can is in front of you, directly ahead instead of to the right.", + "E": "The trash can is to your right." + }, + "answer_label": "E", + "answer_text": "The trash can is to your right.", + "q_id": "P35_101_7_T2", + "correct_answer": "The trash can is to your right.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The trash can is to your left rather than on your right side.", + "D": "The trash can is in front of you, directly ahead instead of to the right.", + "E": "The trash can is to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 13, + 15 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to use a pot. Where in the kitchen can I find a pot?", + "answer": "The pot can be found in the sink and in the first and third compartments of the cabinet to the lower left of the sink.", + "question_chinese": "我想用锅，厨房的哪些位置可以找到锅？", + "answer_chinese": "在洗水槽里以及洗水槽左下方的柜子里的第一格和第三格可以找到锅。", + "evidence": [ + { + "description": "The pot is in the sink.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "The pot is in the first compartment of the cabinet.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "The pot is in the second compartment of the cabinet.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Only in the sink.", + "B": "The pot can be found in the sink and in the first and third compartments of the cabinet to the lower left of the sink.", + "C": "Only in the cabinet.", + "D": "The pot is in the sink and in the first and second compartments of the lower-left cabinet.", + "E": "The pot can be found in the sink and in the second and third compartments of the cabinet below-left of the sink." + }, + "answer_label": "B", + "answer_text": "The pot can be found in the sink and in the first and third compartments of the cabinet to the lower left of the sink.", + "q_id": "P35_101_1", + "correct_answer": "The pot can be found in the sink and in the first and third compartments of the cabinet to the lower left of the sink.", + "original_options": { + "A": "Only in the sink.", + "B": "The pot can be found in the sink and in the first and third compartments of the cabinet to the lower left of the sink.", + "C": "Only in the cabinet.", + "D": "The pot is in the sink and in the first and second compartments of the lower-left cabinet.", + "E": "The pot can be found in the sink and in the second and third compartments of the cabinet below-left of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 22 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to use a pot. Where in the kitchen can I find a pot?", + "answer": "The pot can be found in the sink and in the first and third compartments of the cabinet to the lower left of the sink.", + "question_chinese": "我想用锅，厨房的哪些位置可以找到锅？", + "answer_chinese": "在洗水槽里以及洗水槽左下方的柜子里的第一格和第三格可以找到锅。", + "evidence": [ + { + "description": "The pot is in the sink.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "The pot is in the first compartment of the cabinet.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "The pot is in the second compartment of the cabinet.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:03:02", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Only in the sink.", + "B": "Only in the cabinet.", + "C": "The pot is in the sink and in the first and second compartments of the cabinet below-left of the sink.", + "D": "The pot can be found in the sink and in the first and third compartments of the cabinet to the lower left of the sink.", + "E": "The pot is in the sink and in the second and third compartments of the cabinet below-left of the sink." + }, + "answer_label": "D", + "answer_text": "The pot can be found in the sink and in the first and third compartments of the cabinet to the lower left of the sink.", + "q_id": "P35_101_2", + "correct_answer": "The pot can be found in the sink and in the first and third compartments of the cabinet to the lower left of the sink.", + "original_options": { + "A": "Only in the sink.", + "B": "Only in the cabinet.", + "C": "The pot is in the sink and in the first and second compartments of the cabinet below-left of the sink.", + "D": "The pot can be found in the sink and in the first and third compartments of the cabinet to the lower left of the sink.", + "E": "The pot is in the sink and in the second and third compartments of the cabinet below-left of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 20 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_7", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I'm currently washing dishes in front of the sink. Which is closer to me, the refrigerator or the microwave oven?", + "answer": "The refrigerator is closer to you.", + "question_chinese": "我现在在洗水槽前清洗餐具，冰箱和微波炉哪个离我更近一些？", + "answer_chinese": "冰箱离你更近。", + "evidence": [ + { + "description": "The refrigerator is behind you.", + "timestamps": { + "start": "00:00:56", + "end": "00:01:00" + } + }, + { + "description": "The microwave oven is diagonally behind you.", + "timestamps": { + "start": "00:01:13", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:02:25", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The microwave oven is closer to you.", + "B": "The refrigerator is closer to you." + }, + "answer_label": "B", + "answer_text": "The refrigerator is closer to you.", + "q_id": "P35_101_7", + "correct_answer": "The refrigerator is closer to you.", + "original_options": { + "A": "The microwave oven is closer to you.", + "B": "The refrigerator is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_8", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "I'm currently washing dishes in front of the sink. Which is closer to me, the refrigerator or the microwave oven?", + "answer": "The refrigerator is closer to you.", + "question_chinese": "我现在在洗水槽前清洗餐具，冰箱和微波炉哪个离我更近一些？", + "answer_chinese": "冰箱离你更近。", + "evidence": [ + { + "description": "The refrigerator is behind you.", + "timestamps": { + "start": "00:00:56", + "end": "00:01:00" + } + }, + { + "description": "The microwave oven is diagonally behind you.", + "timestamps": { + "start": "00:01:13", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:03:37", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The microwave oven is closer to you.", + "B": "The refrigerator is closer to you." + }, + "answer_label": "B", + "answer_text": "The refrigerator is closer to you.", + "q_id": "P35_101_8", + "correct_answer": "The refrigerator is closer to you.", + "original_options": { + "A": "The microwave oven is closer to you.", + "B": "The refrigerator is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_4_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the stovetop right now, or farther away?", + "answer": "You are closer to the stove right now, and the stove is in front of you.", + "question_chinese": "我现在离灶台远/近？", + "answer_chinese": "你现在离灶台近，灶台在你的正前方。", + "evidence": [ + { + "description": "You are standing in front of the stove now.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are closer to the stove right now, and the stove is in front of you.", + "B": "You are farther from the stove right now, and the stove is behind you." + }, + "answer_label": "A", + "answer_text": "You are closer to the stove right now, and the stove is in front of you.", + "q_id": "P35_101_4_T1", + "correct_answer": "You are closer to the stove right now, and the stove is in front of you.", + "original_options": { + "A": "You are closer to the stove right now, and the stove is in front of you.", + "B": "You are farther from the stove right now, and the stove is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_4_T2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the stovetop right now, or farther away?", + "answer": "You are farther from the stove right now, and the stove is behind you.", + "question_chinese": "我现在离灶台远/近？", + "answer_chinese": "你现在离灶台远，灶台在你的后面。", + "evidence": [ + { + "description": "Now you turn around and come to the refrigerator, so you are far away from the stove.", + "timestamps": { + "start": "00:00:56", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:01:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are closer to the stove right now, and the stove is in front of you.", + "B": "You are farther from the stove right now, and the stove is behind you." + }, + "answer_label": "B", + "answer_text": "You are farther from the stove right now, and the stove is behind you.", + "q_id": "P35_101_4_T2", + "correct_answer": "You are farther from the stove right now, and the stove is behind you.", + "original_options": { + "A": "You are closer to the stove right now, and the stove is in front of you.", + "B": "You are farther from the stove right now, and the stove is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_6", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I cut open with scissors?", + "answer": "You cut open a bag with scissors, and inside it was hollow powder.", + "question_chinese": "我用剪刀剪开了什么东西？", + "answer_chinese": "你用剪刀剪开了一个袋子，里面装的是空心粉。", + "evidence": [ + { + "description": "You cut open a bag with scissors.", + "timestamps": { + "start": "00:01:46", + "end": "00:01:56" + } + }, + { + "description": "There is hollow powder in the bag.", + "timestamps": { + "start": "00:02:32", + "end": "00:02:55" + } + } + ], + "question_timestamps": "00:03:57", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You cut open a bag with scissors, and inside it was hollow powder.", + "B": "You cut open a box.", + "C": "You cut open a pouch.", + "D": "You cut open a bag with scissors, but inside it was solid powder instead of hollow powder.", + "E": "You cut open a bag with scissors, and the contents were granulated sugar rather than hollow powder." + }, + "answer_label": "A", + "answer_text": "You cut open a bag with scissors, and inside it was hollow powder.", + "q_id": "P35_101_6", + "correct_answer": "You cut open a bag with scissors, and inside it was hollow powder.", + "original_options": { + "A": "You cut open a bag with scissors, and inside it was hollow powder.", + "B": "You cut open a box.", + "C": "You cut open a pouch.", + "D": "You cut open a bag with scissors, but inside it was solid powder instead of hollow powder.", + "E": "You cut open a bag with scissors, and the contents were granulated sugar rather than hollow powder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_5", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I cut open with scissors?", + "answer": "You cut open a bag with scissors, and inside it was hollow powder.", + "question_chinese": "我用剪刀剪开了什么东西？", + "answer_chinese": "你用剪刀剪开了一个袋子，里面装的是空心粉。", + "evidence": [ + { + "description": "You cut open a bag with scissors.", + "timestamps": { + "start": "00:01:46", + "end": "00:01:56" + } + }, + { + "description": "There is hollow powder in the bag.", + "timestamps": { + "start": "00:02:32", + "end": "00:02:55" + } + } + ], + "question_timestamps": "00:04:26", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You cut open a bag with scissors, and inside it was hollow powder.", + "B": "You cut open a box.", + "C": "You cut open a jar.", + "D": "You cut open a bag with scissors, and inside it was fine powder rather than hollow powder.", + "E": "You cut open a bag with scissors, but the contents were powdered sugar instead of hollow powder." + }, + "answer_label": "A", + "answer_text": "You cut open a bag with scissors, and inside it was hollow powder.", + "q_id": "P35_101_5", + "correct_answer": "You cut open a bag with scissors, and inside it was hollow powder.", + "original_options": { + "A": "You cut open a bag with scissors, and inside it was hollow powder.", + "B": "You cut open a box.", + "C": "You cut open a jar.", + "D": "You cut open a bag with scissors, and inside it was fine powder rather than hollow powder.", + "E": "You cut open a bag with scissors, but the contents were powdered sugar instead of hollow powder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_6_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many pots are on my stovetop right now?", + "answer": "There is one pot on your stove right now, and the pot is in front of you.", + "question_chinese": "我的灶台上现在有几个锅?", + "answer_chinese": "你的灶台上现在有一个锅，锅在你的正前方。", + "evidence": [ + { + "description": "So far, there is only one pot on your stove.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:00:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No pots.", + "B": "Two pots.", + "C": "There is one pot on your stove right now, and the pot is in front of you.", + "D": "There are three pots on your stove right now, and all three pots are in front of you.", + "E": "There are four pots on your stove right now, and the four pots are positioned in front of you." + }, + "answer_label": "C", + "answer_text": "There is one pot on your stove right now, and the pot is in front of you.", + "q_id": "P35_101_6_T1", + "correct_answer": "There is one pot on your stove right now, and the pot is in front of you.", + "original_options": { + "A": "No pots.", + "B": "Two pots.", + "C": "There is one pot on your stove right now, and the pot is in front of you.", + "D": "There are three pots on your stove right now, and all three pots are in front of you.", + "E": "There are four pots on your stove right now, and the four pots are positioned in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 19 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_6_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many pots are on my stovetop right now?", + "answer": "There are two pots on your stove right now, and the pots are in front of you.", + "question_chinese": "我的灶台上现在有几个锅?", + "answer_chinese": "你的灶台上现在有两个锅，锅在你的正前方。", + "evidence": [ + { + "description": "There used to be only one pot, but now you put a pot full of peas on it to cook. So now there are two pots.", + "timestamps": { + "start": "00:05:21", + "end": "00:05:24" + } + } + ], + "question_timestamps": "00:05:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One pot.", + "B": "No pots.", + "C": "There are three pots on your stove right now, and the three pots are in front of you.", + "D": "There are two pots on your stove right now, and the pots are in front of you.", + "E": "There are four pots on your stove right now, and all four pots are in front of you." + }, + "answer_label": "D", + "answer_text": "There are two pots on your stove right now, and the pots are in front of you.", + "q_id": "P35_101_6_T2", + "correct_answer": "There are two pots on your stove right now, and the pots are in front of you.", + "original_options": { + "A": "One pot.", + "B": "No pots.", + "C": "There are three pots on your stove right now, and the three pots are in front of you.", + "D": "There are two pots on your stove right now, and the pots are in front of you.", + "E": "There are four pots on your stove right now, and all four pots are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 18 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_8_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the roll of paper on the table relative to me?", + "answer": "The roll of paper on the table is in front of you.", + "question_chinese": "桌子上的卷纸在相对于我的哪里？", + "answer_chinese": "桌子上的卷纸在相对于你的正前方。", + "evidence": [ + { + "description": "The roll of paper on the table is directly in front of you.", + "timestamps": { + "start": "00:06:18", + "end": "00:06:19" + } + } + ], + "question_timestamps": "00:06:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It is behind you.", + "B": "The roll of paper on the table is in front of you.", + "C": "It is to your left.", + "D": "The roll of paper on the table is to your right, not directly in front of you.", + "E": "The roll of paper on the table is to your left, positioned away from your forward view." + }, + "answer_label": "B", + "answer_text": "The roll of paper on the table is in front of you.", + "q_id": "P35_101_8_T1", + "correct_answer": "The roll of paper on the table is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "The roll of paper on the table is in front of you.", + "C": "It is to your left.", + "D": "The roll of paper on the table is to your right, not directly in front of you.", + "E": "The roll of paper on the table is to your left, positioned away from your forward view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_8_T2", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the roll of paper on the table relative to me?", + "answer": "The roll of paper on the table is to your right.", + "question_chinese": "桌子上的卷纸在相对于我的哪里？", + "answer_chinese": "桌子上的卷纸在相对于你的右边。", + "evidence": [ + { + "description": "You went to the left, so the paper on the table became on your right again.", + "timestamps": { + "start": "00:06:25", + "end": "00:06:26" + } + } + ], + "question_timestamps": "00:06:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "The roll of paper on the table is behind you rather than positioned on your right side.", + "D": "The roll of paper on the table is in front of you, not off to the right.", + "E": "The roll of paper on the table is to your right." + }, + "answer_label": "E", + "answer_text": "The roll of paper on the table is to your right.", + "q_id": "P35_101_8_T2", + "correct_answer": "The roll of paper on the table is to your right.", + "original_options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "The roll of paper on the table is behind you rather than positioned on your right side.", + "D": "The roll of paper on the table is in front of you, not off to the right.", + "E": "The roll of paper on the table is to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 17 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_1_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the stovetop relative to me?", + "answer": "The stove is in front of you.", + "question_chinese": "灶台在相对于我的哪里？", + "answer_chinese": "灶台在相对于你的正前方。", + "evidence": [ + { + "description": "The stove is right in front of you.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:51" + } + } + ], + "question_timestamps": "00:00:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stove is behind you.", + "B": "The stove is to your left.", + "C": "The stove is in front of you.", + "D": "The stove is to your right, not directly in front of where you are facing.", + "E": "The stove is above you, mounted higher than your eye level on the wall." + }, + "answer_label": "C", + "answer_text": "The stove is in front of you.", + "q_id": "P35_101_1_T1", + "correct_answer": "The stove is in front of you.", + "original_options": { + "A": "The stove is behind you.", + "B": "The stove is to your left.", + "C": "The stove is in front of you.", + "D": "The stove is to your right, not directly in front of where you are facing.", + "E": "The stove is above you, mounted higher than your eye level on the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 14 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_1_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the stovetop relative to me?", + "answer": "The stove is behind you.", + "question_chinese": "灶台在相对于我的哪里？", + "answer_chinese": "灶台在相对于你的后面。", + "evidence": [ + { + "description": "The stove is behind you.", + "timestamps": { + "start": "00:00:56", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stove is in front of you.", + "B": "The stove is to your right.", + "C": "The stove is to your left rather than behind you from your current viewpoint.", + "D": "The stove is behind you.", + "E": "The stove is in front of you, positioned ahead instead of behind your body." + }, + "answer_label": "D", + "answer_text": "The stove is behind you.", + "q_id": "P35_101_1_T2", + "correct_answer": "The stove is behind you.", + "original_options": { + "A": "The stove is in front of you.", + "B": "The stove is to your right.", + "C": "The stove is to your left rather than behind you from your current viewpoint.", + "D": "The stove is behind you.", + "E": "The stove is in front of you, positioned ahead instead of behind your body." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 14 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_5_T2", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I move the macaroni from, and where did I put it now?", + "answer": "You just moved the macaroni from the cabinet onto the table, and the macaroni is behind you.", + "question_chinese": "现在我将空心粉从哪里拿到了哪里？", + "answer_chinese": "现在你将空心粉从柜子里拿到了桌子上，空心粉在你的后面。", + "evidence": [ + { + "description": "You took the hollow powder from the cupboard to the table.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:01:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You just moved the macaroni from the cabinet onto the table, and the macaroni is behind you.", + "B": "From cabinet to table, in front.", + "C": "From cabinet to table, to your left.", + "D": "You moved the macaroni from the cabinet onto the table, and the macaroni is now in front of you.", + "E": "You moved the macaroni from the cabinet onto the table, and it is now sitting to your right." + }, + "answer_label": "A", + "answer_text": "You just moved the macaroni from the cabinet onto the table, and the macaroni is behind you.", + "q_id": "P35_101_5_T2", + "correct_answer": "You just moved the macaroni from the cabinet onto the table, and the macaroni is behind you.", + "original_options": { + "A": "You just moved the macaroni from the cabinet onto the table, and the macaroni is behind you.", + "B": "From cabinet to table, in front.", + "C": "From cabinet to table, to your left.", + "D": "You moved the macaroni from the cabinet onto the table, and the macaroni is now in front of you.", + "E": "You moved the macaroni from the cabinet onto the table, and it is now sitting to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 18 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_5_T1", + "qid_num": 24, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I move the macaroni from, and where did I put it now?", + "answer": "You just moved the macaroni from the table into the cabinet, and the macaroni is behind you.", + "question_chinese": "现在我将空心粉从哪里拿到了哪里？", + "answer_chinese": "现在你将空心粉从桌子上拿到了柜子里，空心粉在你的后面。", + "evidence": [ + { + "description": "Now you have taken the hollow powder from the table to the cupboard.", + "timestamps": { + "start": "00:02:56", + "end": "00:03:21" + } + } + ], + "question_timestamps": "00:03:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From table to cabinet, in front.", + "B": "From table to cabinet, to your right.", + "C": "You just moved the macaroni from the table into the cabinet, and the macaroni is behind you.", + "D": "You moved the macaroni from the table into the cabinet, and the macaroni is now to your left.", + "E": "You moved the macaroni from the table into the cabinet, and the macaroni is now in front of you." + }, + "answer_label": "C", + "answer_text": "You just moved the macaroni from the table into the cabinet, and the macaroni is behind you.", + "q_id": "P35_101_5_T1", + "correct_answer": "You just moved the macaroni from the table into the cabinet, and the macaroni is behind you.", + "original_options": { + "A": "From table to cabinet, in front.", + "B": "From table to cabinet, to your right.", + "C": "You just moved the macaroni from the table into the cabinet, and the macaroni is behind you.", + "D": "You moved the macaroni from the table into the cabinet, and the macaroni is now to your left.", + "E": "You moved the macaroni from the table into the cabinet, and the macaroni is now in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_3_T2", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is on my purple mat right now?", + "answer": "There is a kitchen scale and a pair of scissors on your purple mat right now, and both the kitchen scale and the scissors are in front of you.", + "question_chinese": "现在我的紫色垫子上有什么东西？", + "answer_chinese": "现在你的紫色垫子上有一个食品称和一把剪刀，食品称和剪刀在你的正前方。", + "evidence": [ + { + "description": "Now there is a food scale and a pair of scissors on your purple mat.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:19" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Only the kitchen scale is on it.", + "B": "Only the scissors are on it.", + "C": "There is a kitchen scale and a pair of scissors on your purple mat right now, and both the kitchen scale and the scissors are in front of you.", + "D": "There is a kitchen scale and scissors on your purple mat right now, and both are behind you.", + "E": "There is a kitchen scale and a pair of scissors on your purple mat, and both items are to your left." + }, + "answer_label": "C", + "answer_text": "There is a kitchen scale and a pair of scissors on your purple mat right now, and both the kitchen scale and the scissors are in front of you.", + "q_id": "P35_101_3_T2", + "correct_answer": "There is a kitchen scale and a pair of scissors on your purple mat right now, and both the kitchen scale and the scissors are in front of you.", + "original_options": { + "A": "Only the kitchen scale is on it.", + "B": "Only the scissors are on it.", + "C": "There is a kitchen scale and a pair of scissors on your purple mat right now, and both the kitchen scale and the scissors are in front of you.", + "D": "There is a kitchen scale and scissors on your purple mat right now, and both are behind you.", + "E": "There is a kitchen scale and a pair of scissors on your purple mat, and both items are to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 21 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_3_T1", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is on my purple mat right now?", + "answer": "There is a small pot on your purple mat right now, and the small pot is to your left.", + "question_chinese": "现在我的紫色垫子上有什么东西？", + "answer_chinese": "现在你的紫色垫子上有一个小锅，小锅在你的左边。", + "evidence": [ + { + "description": "You put a small pot on the mat now.", + "timestamps": { + "start": "00:05:55", + "end": "00:05:58" + } + } + ], + "question_timestamps": "00:06:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There is a small pot on your purple mat right now, and the small pot is to your left.", + "B": "A small pot is in front of you.", + "C": "A small pot is behind you.", + "D": "There is a small pot on your purple mat right now, and the small pot is to your right.", + "E": "There is a small pot on your purple mat right now, and the small pot is positioned in front of you." + }, + "answer_label": "A", + "answer_text": "There is a small pot on your purple mat right now, and the small pot is to your left.", + "q_id": "P35_101_3_T1", + "correct_answer": "There is a small pot on your purple mat right now, and the small pot is to your left.", + "original_options": { + "A": "There is a small pot on your purple mat right now, and the small pot is to your left.", + "B": "A small pot is in front of you.", + "C": "A small pot is behind you.", + "D": "There is a small pot on your purple mat right now, and the small pot is to your right.", + "E": "There is a small pot on your purple mat right now, and the small pot is positioned in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 19, + 21 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_12", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the movement trajectory of hollow powder in the kitchen?", + "answer": "You took the hollow powder from the cabinet on the right side of the TV to the counter and finally put it in the cabinet on the left side of the refrigerator.", + "question_chinese": "空心粉在厨房的运动轨迹是什么？", + "answer_chinese": "空心粉被你从电视机右边的柜子里拿到了柜台上，最后又放进了冰箱左边的柜子里。", + "evidence": [ + { + "description": "The hollow powder is in the cabinet on the right side of the TV.", + "timestamps": { + "start": "00:01:13", + "end": "00:01:21" + } + }, + { + "description": "You put the hollow powder on the counter.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:31" + } + }, + { + "description": "You put the hollow powder in the cabinet on the left side of the refrigerator.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:19" + } + } + ], + "question_timestamps": "00:04:24", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You moved it from the left cabinet.", + "B": "You took the hollow powder from the cabinet on the right side of the TV to the counter and finally put it in the cabinet on the left side of the refrigerator.", + "C": "You moved it to the refrigerator.", + "D": "You took it from the cabinet left of the TV, set it on the counter, then stored it left of the refrigerator.", + "E": "You took it from the cabinet right of the TV, set it on the counter, then stored it above the refrigerator." + }, + "answer_label": "B", + "answer_text": "You took the hollow powder from the cabinet on the right side of the TV to the counter and finally put it in the cabinet on the left side of the refrigerator.", + "q_id": "P35_101_12", + "correct_answer": "You took the hollow powder from the cabinet on the right side of the TV to the counter and finally put it in the cabinet on the left side of the refrigerator.", + "original_options": { + "A": "You moved it from the left cabinet.", + "B": "You took the hollow powder from the cabinet on the right side of the TV to the counter and finally put it in the cabinet on the left side of the refrigerator.", + "C": "You moved it to the refrigerator.", + "D": "You took it from the cabinet left of the TV, set it on the counter, then stored it left of the refrigerator.", + "E": "You took it from the cabinet right of the TV, set it on the counter, then stored it above the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 22, + 21 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_11", + "qid_num": 28, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the movement trajectory of hollow powder in the kitchen?", + "answer": "You took the hollow powder from the cabinet on the right side of the TV to the counter and finally put it in the cabinet on the left side of the refrigerator.", + "question_chinese": "空心粉在厨房的运动轨迹是什么？", + "answer_chinese": "空心粉被你从电视机右边的柜子里拿到了柜台上，最后又放进了冰箱左边的柜子里。", + "evidence": [ + { + "description": "The hollow powder is in the cabinet on the right side of the TV.", + "timestamps": { + "start": "00:01:13", + "end": "00:01:21" + } + }, + { + "description": "You put the hollow powder on the counter.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:31" + } + }, + { + "description": "You put the hollow powder in the cabinet on the left side of the refrigerator.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:19" + } + } + ], + "question_timestamps": "00:04:47", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You put it in the drawer below.", + "B": "You stored it above the refrigerator.", + "C": "You took the hollow powder from the cabinet on the right side of the TV to the counter and finally put it in the cabinet on the left side of the refrigerator.", + "D": "You took it from the cabinet left of the TV, set it on the counter, then put it away.", + "E": "You grabbed it from the shelf next to the TV, set it on the counter, then stored it right of the refrigerator." + }, + "answer_label": "C", + "answer_text": "You took the hollow powder from the cabinet on the right side of the TV to the counter and finally put it in the cabinet on the left side of the refrigerator.", + "q_id": "P35_101_11", + "correct_answer": "You took the hollow powder from the cabinet on the right side of the TV to the counter and finally put it in the cabinet on the left side of the refrigerator.", + "original_options": { + "A": "You put it in the drawer below.", + "B": "You stored it above the refrigerator.", + "C": "You took the hollow powder from the cabinet on the right side of the TV to the counter and finally put it in the cabinet on the left side of the refrigerator.", + "D": "You took it from the cabinet left of the TV, set it on the counter, then put it away.", + "E": "You grabbed it from the shelf next to the TV, set it on the counter, then stored it right of the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 22 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_13", + "qid_num": 29, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are peas taken from?", + "answer": "The peas were taken out from the first compartment of the refrigerator.", + "question_chinese": "豌豆是从哪里拿出来的？", + "answer_chinese": "豌豆是从冰箱的第一格拿出来的。", + "evidence": [ + { + "description": "The peas were taken out from the first compartment of the refrigerator.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:05:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the second compartment.", + "B": "The peas were taken out from the first compartment of the refrigerator.", + "C": "From the door shelf.", + "D": "The peas were taken out from the vegetable drawer inside the refrigerator, not from the first compartment.", + "E": "The peas were taken out from the bottom shelf of the refrigerator rather than the first compartment." + }, + "answer_label": "B", + "answer_text": "The peas were taken out from the first compartment of the refrigerator.", + "q_id": "P35_101_13", + "correct_answer": "The peas were taken out from the first compartment of the refrigerator.", + "original_options": { + "A": "From the second compartment.", + "B": "The peas were taken out from the first compartment of the refrigerator.", + "C": "From the door shelf.", + "D": "The peas were taken out from the vegetable drawer inside the refrigerator, not from the first compartment.", + "E": "The peas were taken out from the bottom shelf of the refrigerator rather than the first compartment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "P35_101", + "video_length": "00:06:28", + "qid": "P35_101_14", + "qid_num": 30, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are peas taken from?", + "answer": "The peas were taken out from the first compartment of the refrigerator.", + "question_chinese": "豌豆是从哪里拿出来的？", + "answer_chinese": "豌豆是从冰箱的第一格拿出来的。", + "evidence": [ + { + "description": "The peas were taken out from the first compartment of the refrigerator.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:06:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the second compartment.", + "B": "From the bottom drawer.", + "C": "The peas were taken out from the door shelf of the refrigerator, not from the first compartment.", + "D": "The peas were taken out from the first compartment of the refrigerator.", + "E": "The peas were taken out from the vegetable crisper of the refrigerator rather than the first compartment." + }, + "answer_label": "D", + "answer_text": "The peas were taken out from the first compartment of the refrigerator.", + "q_id": "P35_101_14", + "correct_answer": "The peas were taken out from the first compartment of the refrigerator.", + "original_options": { + "A": "From the second compartment.", + "B": "From the bottom drawer.", + "C": "The peas were taken out from the door shelf of the refrigerator, not from the first compartment.", + "D": "The peas were taken out from the first compartment of the refrigerator.", + "E": "The peas were taken out from the vegetable crisper of the refrigerator rather than the first compartment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + } +] diff --git a/QAs/P35_103-mcq.json b/QAs/P35_103-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1f374799ba02da14823d5f556b895d81b5771718 --- /dev/null +++ b/QAs/P35_103-mcq.json @@ -0,0 +1,1141 @@ +[ + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I add the white powder?", + "answer": "You poured it three times.", + "question_chinese": "我加入了多少次白色粉末", + "answer_chinese": "你倒入了三次", + "evidence": [ + { + "description": "You poured it twice.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:59" + } + }, + { + "description": "You joined the third time.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You poured it once.", + "B": "You poured it twice.", + "C": "You poured the white powder four times during the process, not three times.", + "D": "You poured it three times.", + "E": "You poured the white powder five times in total, with several quick additions." + }, + "answer_label": "D", + "answer_text": "You poured it three times.", + "q_id": "P35_103_1", + "correct_answer": "You poured it three times.", + "original_options": { + "A": "You poured it once.", + "B": "You poured it twice.", + "C": "You poured the white powder four times during the process, not three times.", + "D": "You poured it three times.", + "E": "You poured the white powder five times in total, with several quick additions." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 13 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I add the white powder?", + "answer": "You poured it in three times.", + "question_chinese": "我加入了多少次白色粉末", + "answer_chinese": "你倒入了三次", + "evidence": [ + { + "description": "You poured it twice.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:59" + } + }, + { + "description": "You joined the third time.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:12:36", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You poured it in four times.", + "B": "You poured it in five times.", + "C": "You poured it in three times.", + "D": "You poured the white powder in six times over the course of the task, not three.", + "E": "You poured the white powder in seven times altogether, adding it repeatedly rather than just three." + }, + "answer_label": "C", + "answer_text": "You poured it in three times.", + "q_id": "P35_103_2", + "correct_answer": "You poured it in three times.", + "original_options": { + "A": "You poured it in four times.", + "B": "You poured it in five times.", + "C": "You poured it in three times.", + "D": "You poured the white powder in six times over the course of the task, not three.", + "E": "You poured the white powder in seven times altogether, adding it repeatedly rather than just three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_15", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Did I put the towel back in its original place?", + "answer": "You put it back.", + "question_chinese": "我把毛巾放回原位了吗", + "answer_chinese": "你放回了", + "evidence": [ + { + "description": "You put it back.", + "timestamps": { + "start": "00:21:05", + "end": "00:21:12" + } + } + ], + "question_timestamps": "00:21:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You put it back.", + "B": "You did not put it back." + }, + "answer_label": "A", + "answer_text": "You put it back.", + "q_id": "P35_103_15", + "correct_answer": "You put it back.", + "original_options": { + "A": "You put it back.", + "B": "You did not put it back." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_16", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Did I put the towel back in its original place?", + "answer": "You put it back.", + "question_chinese": "我把毛巾放回原位了吗", + "answer_chinese": "你放回了", + "evidence": [ + { + "description": "You put it back.", + "timestamps": { + "start": "00:21:05", + "end": "00:21:12" + } + }, + { + "description": "You put it back.", + "timestamps": { + "start": "00:22:50", + "end": "00:22:55" + } + } + ], + "question_timestamps": "00:25:12", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You didn't put it back.", + "B": "You put it back." + }, + "answer_label": "B", + "answer_text": "You put it back.", + "q_id": "P35_103_16", + "correct_answer": "You put it back.", + "original_options": { + "A": "You didn't put it back.", + "B": "You put it back." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_13", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the towel I use to dry my hands the same as the towel I use to dry bowls?", + "answer": "Yes.", + "question_chinese": "我擦手用的毛巾和擦碗用的毛巾是同一个吗", + "answer_chinese": "是", + "evidence": [ + { + "description": "Yes.", + "timestamps": { + "start": "00:18:30", + "end": "00:19:00" + } + } + ], + "question_timestamps": "00:19:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "P35_103_13", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_14", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the towel I use to dry my hands the same as the towel I use to dry bowls?", + "answer": "Yes.", + "question_chinese": "我擦手用的毛巾和擦碗用的毛巾是同一个吗", + "answer_chinese": "是", + "evidence": [ + { + "description": "Yes.", + "timestamps": { + "start": "00:18:30", + "end": "00:19:00" + } + }, + { + "description": "A towel for wiping bowls.", + "timestamps": { + "start": "00:20:00", + "end": "00:21:10" + } + }, + { + "description": "A towel for drying hands.", + "timestamps": { + "start": "00:22:45", + "end": "00:22:55" + } + } + ], + "question_timestamps": "00:25:12", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "P35_103_14", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_11", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I dry my hands after washing things?", + "answer": "Turn left first and then go straight.", + "question_chinese": "我洗完东西怎么样才能擦手", + "answer_chinese": "先向左转再直走", + "evidence": [ + { + "description": "Turn left first and then go straight.", + "timestamps": { + "start": "00:18:25", + "end": "00:18:40" + } + } + ], + "question_timestamps": "00:18:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left first and then go straight.", + "B": "Go straight, then turn left.", + "C": "Turn right, then go straight.", + "D": "Turn right first, then go straight to reach the towel area and dry your hands.", + "E": "Go straight first and keep going until you reach the area where you can dry your hands." + }, + "answer_label": "A", + "answer_text": "Turn left first and then go straight.", + "q_id": "P35_103_11", + "correct_answer": "Turn left first and then go straight.", + "original_options": { + "A": "Turn left first and then go straight.", + "B": "Go straight, then turn left.", + "C": "Turn right, then go straight.", + "D": "Turn right first, then go straight to reach the towel area and dry your hands.", + "E": "Go straight first and keep going until you reach the area where you can dry your hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_12", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I dry my hands after washing things?", + "answer": "Turn left first and then go straight.", + "question_chinese": "我洗完东西怎么样才能擦手", + "answer_chinese": "先左转再直走", + "evidence": [ + { + "description": "Turn left first and then go straight.", + "timestamps": { + "start": "00:18:25", + "end": "00:18:40" + } + }, + { + "description": "Turn left first and then go straight.", + "timestamps": { + "start": "00:22:45", + "end": "00:22:55" + } + } + ], + "question_timestamps": "00:25:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn left.", + "B": "Turn right, then go straight.", + "C": "Turn left first, then turn right to get to the place where you can dry your hands.", + "D": "Turn right first, then go straight to reach the spot where you can dry your hands.", + "E": "Turn left first and then go straight." + }, + "answer_label": "E", + "answer_text": "Turn left first and then go straight.", + "q_id": "P35_103_12", + "correct_answer": "Turn left first and then go straight.", + "original_options": { + "A": "Go straight, then turn left.", + "B": "Turn right, then go straight.", + "C": "Turn left first, then turn right to get to the place where you can dry your hands.", + "D": "Turn right first, then go straight to reach the spot where you can dry your hands.", + "E": "Turn left first and then go straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_2_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are on my stovetop right now?", + "answer": "There are two pots on your stove right now, and the stove is to your left.", + "question_chinese": "我的灶台上现在有几个锅？", + "answer_chinese": "你的灶台上现在有两个锅，灶台在你的左边。", + "evidence": [ + { + "description": "There are two pots on your stove now, and the stove is on your left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are two pots on your stove right now, and the stove is to your left.", + "B": "One pot.", + "C": "Three pots.", + "D": "There is one pot on your stove right now, and the stove is to your left.", + "E": "There are three pots on your stove right now, and the stove is to your left." + }, + "answer_label": "A", + "answer_text": "There are two pots on your stove right now, and the stove is to your left.", + "q_id": "P35_103_2_T1", + "correct_answer": "There are two pots on your stove right now, and the stove is to your left.", + "original_options": { + "A": "There are two pots on your stove right now, and the stove is to your left.", + "B": "One pot.", + "C": "Three pots.", + "D": "There is one pot on your stove right now, and the stove is to your left.", + "E": "There are three pots on your stove right now, and the stove is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_2_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pots are on my stovetop right now?", + "answer": "There are no pots on your stove right now, and the stove is to your front-left.", + "question_chinese": "我的灶台上现在有几个锅？", + "answer_chinese": "你的灶台上现在没有锅，灶台在你的左前方。", + "evidence": [ + { + "description": "There is no pot on your stove now. The stove is in front of you on the left.", + "timestamps": { + "start": "00:18:41", + "end": "00:18:43" + } + } + ], + "question_timestamps": "00:18:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One pot.", + "B": "There are no pots on your stove right now, and the stove is to your front-left.", + "C": "Two pots.", + "D": "There is one pot on your stove right now, and the stove is to your front-left.", + "E": "There are two pots on your stove right now, and the stove is to your front-left." + }, + "answer_label": "B", + "answer_text": "There are no pots on your stove right now, and the stove is to your front-left.", + "q_id": "P35_103_2_T2", + "correct_answer": "There are no pots on your stove right now, and the stove is to your front-left.", + "original_options": { + "A": "One pot.", + "B": "There are no pots on your stove right now, and the stove is to your front-left.", + "C": "Two pots.", + "D": "There is one pot on your stove right now, and the stove is to your front-left.", + "E": "There are two pots on your stove right now, and the stove is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_9", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I reach the green bottle?", + "answer": "You can reach the green bottle.", + "question_chinese": "我能不能够到绿色瓶子", + "answer_chinese": "你可以够到绿色瓶子", + "evidence": [ + { + "description": "You can reach the green bottle.", + "timestamps": { + "start": "00:11:30", + "end": "00:12:00" + } + } + ], + "question_timestamps": "00:18:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can reach the green bottle.", + "B": "You cannot reach the green bottle." + }, + "answer_label": "A", + "answer_text": "You can reach the green bottle.", + "q_id": "P35_103_9", + "correct_answer": "You can reach the green bottle.", + "original_options": { + "A": "You can reach the green bottle.", + "B": "You cannot reach the green bottle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_10", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I reach the green bottle?", + "answer": "You can.", + "question_chinese": "我能不能够到绿色瓶子", + "answer_chinese": "你可以", + "evidence": [ + { + "description": "You can reach the green bottle.", + "timestamps": { + "start": "00:11:30", + "end": "00:12:00" + } + }, + { + "description": "You can still reach it.", + "timestamps": { + "start": "00:19:12", + "end": "00:19:15" + } + } + ], + "question_timestamps": "00:25:12", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't.", + "B": "You can." + }, + "answer_label": "B", + "answer_text": "You can.", + "q_id": "P35_103_10", + "correct_answer": "You can.", + "original_options": { + "A": "You can't.", + "B": "You can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_5_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I walk to a spot in front of the refrigerator?", + "answer": "You walked straight, and the refrigerator is in front of you.", + "question_chinese": "我要怎样才能走到冰箱面前？", + "answer_chinese": "你直走，冰箱在你的正前方。", + "evidence": [ + { + "description": "Go straight, and the refrigerator is in front of you.", + "timestamps": { + "start": "00:07:10", + "end": "00:07:13" + } + } + ], + "question_timestamps": "00:07:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked straight; it’s behind you.", + "B": "You walked straight; it’s to your left.", + "C": "You walked straight, and the refrigerator is in front of you.", + "D": "You walked straight, but you overshot the spot and the refrigerator ends up behind you.", + "E": "You walked straight, then drifted slightly right, leaving the refrigerator off to your left side." + }, + "answer_label": "C", + "answer_text": "You walked straight, and the refrigerator is in front of you.", + "q_id": "P35_103_5_T2", + "correct_answer": "You walked straight, and the refrigerator is in front of you.", + "original_options": { + "A": "You walked straight; it’s behind you.", + "B": "You walked straight; it’s to your left.", + "C": "You walked straight, and the refrigerator is in front of you.", + "D": "You walked straight, but you overshot the spot and the refrigerator ends up behind you.", + "E": "You walked straight, then drifted slightly right, leaving the refrigerator off to your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 15 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_5_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I walk to a spot in front of the refrigerator?", + "answer": "You turned left first and then walked straight; the refrigerator is in front of you.", + "question_chinese": "我要怎样才能走到冰箱面前？", + "answer_chinese": "你先向左转再直走，冰箱在你的正前方。", + "evidence": [ + { + "description": "Turn left first and then go straight. The refrigerator is right in front of you.", + "timestamps": { + "start": "00:09:59", + "end": "00:10:02" + } + } + ], + "question_timestamps": "00:18:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then it’s behind you.", + "B": "Turn left, then it’s to your right.", + "C": "You turned left first and then walked straight, but the refrigerator ends up to your left.", + "D": "You turned left first and then walked straight; the refrigerator is in front of you.", + "E": "You turned left first and then walked straight, and you stop short so it remains behind you." + }, + "answer_label": "D", + "answer_text": "You turned left first and then walked straight; the refrigerator is in front of you.", + "q_id": "P35_103_5_T1", + "correct_answer": "You turned left first and then walked straight; the refrigerator is in front of you.", + "original_options": { + "A": "Turn left, then it’s behind you.", + "B": "Turn left, then it’s to your right.", + "C": "You turned left first and then walked straight, but the refrigerator ends up to your left.", + "D": "You turned left first and then walked straight; the refrigerator is in front of you.", + "E": "You turned left first and then walked straight, and you stop short so it remains behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 17 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_7_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the white roll of paper on the table relative to me?", + "answer": "The white roll of paper on the table is in front of you.", + "question_chinese": "桌子上的白色卷纸在相对于我的哪里？", + "answer_chinese": "桌子上的白色卷纸在相对于你的正前方。", + "evidence": [ + { + "description": "The white roll of paper on the table is right in front of you.", + "timestamps": { + "start": "00:20:00", + "end": "00:20:03" + } + } + ], + "question_timestamps": "00:20:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The white roll of paper on the table is behind you.", + "B": "The white roll of paper on the table is in front of you." + }, + "answer_label": "B", + "answer_text": "The white roll of paper on the table is in front of you.", + "q_id": "P35_103_7_T1", + "correct_answer": "The white roll of paper on the table is in front of you.", + "original_options": { + "A": "The white roll of paper on the table is behind you.", + "B": "The white roll of paper on the table is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_7_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the white roll of paper on the table relative to me?", + "answer": "The white roll of paper on the table is behind you.", + "question_chinese": "桌子上的白色卷纸在相对于我的哪里？", + "answer_chinese": "桌子上的白色卷纸在相对于你的后面。", + "evidence": [ + { + "description": "The white roll of paper on the table is behind you.", + "timestamps": { + "start": "00:20:04", + "end": "00:20:08" + } + } + ], + "question_timestamps": "00:20:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The white roll of paper on the table is behind you.", + "B": "The white roll of paper on the table is in front of you." + }, + "answer_label": "A", + "answer_text": "The white roll of paper on the table is behind you.", + "q_id": "P35_103_7_T2", + "correct_answer": "The white roll of paper on the table is behind you.", + "original_options": { + "A": "The white roll of paper on the table is behind you.", + "B": "The white roll of paper on the table is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_3_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the orange cabinet relative to me?", + "answer": "The orange cabinet is to your left.", + "question_chinese": "橙色的柜子在相对于我的哪里？", + "answer_chinese": "橙色的柜子在相对于你的左边。", + "evidence": [ + { + "description": "The orange cabinet is on your left.", + "timestamps": { + "start": "00:07:09", + "end": "00:07:11" + } + } + ], + "question_timestamps": "00:07:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The orange cabinet is in front of you.", + "B": "The orange cabinet is to your left.", + "C": "The orange cabinet is behind you.", + "D": "The orange cabinet is to your right, not on your left side.", + "E": "The orange cabinet is above you, rather than positioned to your left." + }, + "answer_label": "B", + "answer_text": "The orange cabinet is to your left.", + "q_id": "P35_103_3_T2", + "correct_answer": "The orange cabinet is to your left.", + "original_options": { + "A": "The orange cabinet is in front of you.", + "B": "The orange cabinet is to your left.", + "C": "The orange cabinet is behind you.", + "D": "The orange cabinet is to your right, not on your left side.", + "E": "The orange cabinet is above you, rather than positioned to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 12, + 12 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_3_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the orange cabinet relative to me?", + "answer": "The orange cabinet is behind you.", + "question_chinese": "橙色的柜子在相对于我的哪里？", + "answer_chinese": "橙色的柜子在相对于你的后面。", + "evidence": [ + { + "description": "The orange cabinet is behind you.", + "timestamps": { + "start": "00:07:19", + "end": "00:07:22" + } + } + ], + "question_timestamps": "00:07:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The orange cabinet is behind you.", + "B": "The orange cabinet is in front of you.", + "C": "The orange cabinet is to your left.", + "D": "The orange cabinet is to your right, instead of being behind you.", + "E": "The orange cabinet is above you, not located behind you where you’re facing away." + }, + "answer_label": "A", + "answer_text": "The orange cabinet is behind you.", + "q_id": "P35_103_3_T1", + "correct_answer": "The orange cabinet is behind you.", + "original_options": { + "A": "The orange cabinet is behind you.", + "B": "The orange cabinet is in front of you.", + "C": "The orange cabinet is to your left.", + "D": "The orange cabinet is to your right, instead of being behind you.", + "E": "The orange cabinet is above you, not located behind you where you’re facing away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 12, + 15 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_1_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I added white powder so far?", + "answer": "You have added the white powder twice by now, and the white powder is in front of you.", + "question_chinese": "现在我加了多少次白色粉末？", + "answer_chinese": "现在你加了两次白色粉末，白色粉末在你的正前方。", + "evidence": [ + { + "description": "You poured it twice.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:01", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You added it once so far.", + "B": "You added it three times so far.", + "C": "You have added the white powder four times by now, and the white powder is in front of you.", + "D": "You have added the white powder five times by now, and the white powder is in front of you.", + "E": "You have added the white powder twice by now, and the white powder is in front of you." + }, + "answer_label": "E", + "answer_text": "You have added the white powder twice by now, and the white powder is in front of you.", + "q_id": "P35_103_1_T1", + "correct_answer": "You have added the white powder twice by now, and the white powder is in front of you.", + "original_options": { + "A": "You added it once so far.", + "B": "You added it three times so far.", + "C": "You have added the white powder four times by now, and the white powder is in front of you.", + "D": "You have added the white powder five times by now, and the white powder is in front of you.", + "E": "You have added the white powder twice by now, and the white powder is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 19 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_1_T2", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I added white powder so far?", + "answer": "You have added the white powder three times by now, and the white powder is in front of you.", + "question_chinese": "现在我加了多少次白色粉末？", + "answer_chinese": "现在你加了三次白色粉末，白色粉末在你的正前方。", + "evidence": [ + { + "description": "You poured it twice.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:59" + } + }, + { + "description": "You joined the third time.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:01:34", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have added the white powder three times by now, and the white powder is in front of you.", + "B": "You added it once so far.", + "C": "You added it twice so far.", + "D": "You have added the white powder four times by now, and the white powder is in front of you.", + "E": "You have added the white powder five times by now, and the white powder is in front of you." + }, + "answer_label": "A", + "answer_text": "You have added the white powder three times by now, and the white powder is in front of you.", + "q_id": "P35_103_1_T2", + "correct_answer": "You have added the white powder three times by now, and the white powder is in front of you.", + "original_options": { + "A": "You have added the white powder three times by now, and the white powder is in front of you.", + "B": "You added it once so far.", + "C": "You added it twice so far.", + "D": "You have added the white powder four times by now, and the white powder is in front of you.", + "E": "You have added the white powder five times by now, and the white powder is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_6_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is in my sink right now?", + "answer": "There is a plate in your sink right now, and the sink is in front of you.", + "question_chinese": "现在我的水槽里有什么？", + "answer_chinese": "现在你的水槽里有一个盘子，水槽在你的正前方。", + "evidence": [ + { + "description": "Now there is a plate in your sink, and the sink is in front of you.", + "timestamps": { + "start": "00:14:25", + "end": "00:14:26" + } + } + ], + "question_timestamps": "00:14:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There is a plate in your sink right now, and the sink is in front of you.", + "B": "There is nothing in your sink right now, and the sink is in front of you." + }, + "answer_label": "A", + "answer_text": "There is a plate in your sink right now, and the sink is in front of you.", + "q_id": "P35_103_6_T1", + "correct_answer": "There is a plate in your sink right now, and the sink is in front of you.", + "original_options": { + "A": "There is a plate in your sink right now, and the sink is in front of you.", + "B": "There is nothing in your sink right now, and the sink is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "P35_103", + "video_length": "00:25:12", + "qid": "P35_103_6_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is in my sink right now?", + "answer": "There is a plate and a spatula in your sink right now, and the sink is in front of you.", + "question_chinese": "现在我的水槽里有什么？", + "answer_chinese": "现在你的水槽里有盘子和铲子，水槽在你的正前方。", + "evidence": [ + { + "description": "plate", + "timestamps": { + "start": "00:14:25", + "end": "00:14:27" + } + }, + { + "description": "shovel", + "timestamps": { + "start": "00:14:29", + "end": "00:14:32" + } + } + ], + "question_timestamps": "00:14:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There is only a plate in your sink right now, and the sink is in front of you.", + "B": "There is a plate and a spatula in your sink right now, and the sink is in front of you." + }, + "answer_label": "B", + "answer_text": "There is a plate and a spatula in your sink right now, and the sink is in front of you.", + "q_id": "P35_103_6_T2", + "correct_answer": "There is a plate and a spatula in your sink right now, and the sink is in front of you.", + "original_options": { + "A": "There is only a plate in your sink right now, and the sink is in front of you.", + "B": "There is a plate and a spatula in your sink right now, and the sink is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + } +] diff --git a/QAs/P36_101-mcq.json b/QAs/P36_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f97a8b17b8dfefaadd42a0f4e992325fa666814f --- /dev/null +++ b/QAs/P36_101-mcq.json @@ -0,0 +1,694 @@ +[ + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_11", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What is the condition of the kitchen window?", + "answer": "The window is open.", + "question_chinese": "厨房窗子的状态是什么？", + "answer_chinese": "窗子是打开的。", + "evidence": [ + { + "description": "The window is open.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:04:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is open.", + "B": "The window is closed.", + "C": "The window is locked.", + "D": "The kitchen window is shut tightly, with no gap indicating that it’s open at all.", + "E": "The window looks closed and secured, as if it has been latched to keep it from opening." + }, + "answer_label": "A", + "answer_text": "The window is open.", + "q_id": "P36_101_11", + "correct_answer": "The window is open.", + "original_options": { + "A": "The window is open.", + "B": "The window is closed.", + "C": "The window is locked.", + "D": "The kitchen window is shut tightly, with no gap indicating that it’s open at all.", + "E": "The window looks closed and secured, as if it has been latched to keep it from opening." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_12", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What is the condition of the kitchen window?", + "answer": "The window is open.", + "question_chinese": "厨房窗子的状态是什么？", + "answer_chinese": "窗子是打开的。", + "evidence": [ + { + "description": "The window is open.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:07:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is closed.", + "B": "The window is shattered.", + "C": "The kitchen window is fully closed, with both panes aligned and no opening visible.", + "D": "The window is open.", + "E": "The kitchen window looks shut, with the latch engaged and no gap along the frame." + }, + "answer_label": "D", + "answer_text": "The window is open.", + "q_id": "P36_101_12", + "correct_answer": "The window is open.", + "original_options": { + "A": "The window is closed.", + "B": "The window is shattered.", + "C": "The kitchen window is fully closed, with both panes aligned and no opening visible.", + "D": "The window is open.", + "E": "The kitchen window looks shut, with the latch engaged and no gap along the frame." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory when I went to the kitchen?", + "answer": "Go straight first and then turn right.", + "question_chinese": "去厨房时我的运动轨迹是什么？", + "answer_chinese": "你先直走然后向右转弯。", + "evidence": [ + { + "description": "You go straight first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "You turn right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:01:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn left.", + "B": "Go straight first and then turn right.", + "C": "Turn right, then go straight.", + "D": "Turn left first, then continue straight until you reach the kitchen entrance.", + "E": "Go left first and then turn right after you pass the hallway opening." + }, + "answer_label": "B", + "answer_text": "Go straight first and then turn right.", + "q_id": "P36_101_1", + "correct_answer": "Go straight first and then turn right.", + "original_options": { + "A": "Go straight, then turn left.", + "B": "Go straight first and then turn right.", + "C": "Turn right, then go straight.", + "D": "Turn left first, then continue straight until you reach the kitchen entrance.", + "E": "Go left first and then turn right after you pass the hallway opening." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory when I went to the kitchen?", + "answer": "Go straight first and then turn right.", + "question_chinese": "去厨房时我的运动轨迹是什么？", + "answer_chinese": "你先直走然后向右转弯。", + "evidence": [ + { + "description": "You go straight first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "You turn right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn left.", + "B": "Turn right, then go straight.", + "C": "Go straight first and then turn right.", + "D": "Turn left first, then continue straight down the hall toward where the kitchen is.", + "E": "Go straight first, then turn left at the corner before reaching the kitchen doorway." + }, + "answer_label": "C", + "answer_text": "Go straight first and then turn right.", + "q_id": "P36_101_2", + "correct_answer": "Go straight first and then turn right.", + "original_options": { + "A": "Go straight, then turn left.", + "B": "Turn right, then go straight.", + "C": "Go straight first and then turn right.", + "D": "Turn left first, then continue straight down the hall toward where the kitchen is.", + "E": "Go straight first, then turn left at the corner before reaching the kitchen doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stools are placed under the round table?", + "answer": "Four benches were placed.", + "question_chinese": "圆桌下面放了几条凳子？", + "answer_chinese": "放了四条板凳。", + "evidence": [ + { + "description": "At first, you saw three stools.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "At second, you saw four stools.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three benches were placed.", + "B": "Five benches were placed.", + "C": "Two benches were placed under the round table, leaving the other sides without seating.", + "D": "Six benches were placed under the round table, filling nearly all available space around it.", + "E": "Four benches were placed." + }, + "answer_label": "E", + "answer_text": "Four benches were placed.", + "q_id": "P36_101_9", + "correct_answer": "Four benches were placed.", + "original_options": { + "A": "Three benches were placed.", + "B": "Five benches were placed.", + "C": "Two benches were placed under the round table, leaving the other sides without seating.", + "D": "Six benches were placed under the round table, filling nearly all available space around it.", + "E": "Four benches were placed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_10", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stools are placed under the round table?", + "answer": "Four benches were placed.", + "question_chinese": "圆桌下面放了几条凳子？", + "answer_chinese": "放了四条板凳。", + "evidence": [ + { + "description": "At first, you saw three stools.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "At second, you saw four stools.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four benches were placed.", + "B": "Two benches were placed.", + "C": "Five benches were placed.", + "D": "Three benches were placed under the round table, with one side left without a bench.", + "E": "Six benches were placed under the round table, crowding the space around the table." + }, + "answer_label": "A", + "answer_text": "Four benches were placed.", + "q_id": "P36_101_10", + "correct_answer": "Four benches were placed.", + "original_options": { + "A": "Four benches were placed.", + "B": "Two benches were placed.", + "C": "Five benches were placed.", + "D": "Three benches were placed under the round table, with one side left without a bench.", + "E": "Six benches were placed under the round table, crowding the space around the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + }, + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which tableware did I wash in the sink?", + "answer": "You washed the dishes, the shovel and the pot.", + "question_chinese": "我在洗水槽洗了哪些餐具？", + "answer_chinese": "你洗了盘子，铲子和锅。", + "evidence": [ + { + "description": "You washed the dishes.", + "timestamps": { + "start": "00:09:33", + "end": "00:09:52" + } + }, + { + "description": "You washed the shovel.", + "timestamps": { + "start": "00:10:31", + "end": "00:10:49" + } + }, + { + "description": "You washed the pot.", + "timestamps": { + "start": "00:10:50", + "end": "00:11:30" + } + } + ], + "question_timestamps": "00:11:37", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You washed the cups and the pot.", + "B": "You washed the dishes and the pot.", + "C": "You washed the dishes, the spoon, and the pot in the sink before setting them aside to dry.", + "D": "You washed the plates, the shovel, and the pan in the sink, then rinsed them thoroughly.", + "E": "You washed the dishes, the shovel and the pot." + }, + "answer_label": "E", + "answer_text": "You washed the dishes, the shovel and the pot.", + "q_id": "P36_101_5", + "correct_answer": "You washed the dishes, the shovel and the pot.", + "original_options": { + "A": "You washed the cups and the pot.", + "B": "You washed the dishes and the pot.", + "C": "You washed the dishes, the spoon, and the pot in the sink before setting them aside to dry.", + "D": "You washed the plates, the shovel, and the pan in the sink, then rinsed them thoroughly.", + "E": "You washed the dishes, the shovel and the pot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 16 + ] + }, + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which tableware did I wash in the sink?", + "answer": "You washed the dishes, the shovel and the pot.", + "question_chinese": "我在洗水槽洗了哪些餐具？", + "answer_chinese": "你洗了盘子，铲子和锅。", + "evidence": [ + { + "description": "You washed the dishes.", + "timestamps": { + "start": "00:09:33", + "end": "00:09:52" + } + }, + { + "description": "You washed the shovel.", + "timestamps": { + "start": "00:10:31", + "end": "00:10:49" + } + }, + { + "description": "You washed the pot.", + "timestamps": { + "start": "00:10:50", + "end": "00:11:30" + } + } + ], + "question_timestamps": "00:11:41", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You washed the dishes, the shovel and the pot.", + "B": "You washed the bowls and the cup.", + "C": "You washed the plates and the tray.", + "D": "You washed the dishes, the ladle, and the pot in the sink, then drained the water.", + "E": "You washed the dishes, the shovel, and the pan in the sink and left them on the counter." + }, + "answer_label": "A", + "answer_text": "You washed the dishes, the shovel and the pot.", + "q_id": "P36_101_6", + "correct_answer": "You washed the dishes, the shovel and the pot.", + "original_options": { + "A": "You washed the dishes, the shovel and the pot.", + "B": "You washed the bowls and the cup.", + "C": "You washed the plates and the tray.", + "D": "You washed the dishes, the ladle, and the pot in the sink, then drained the water.", + "E": "You washed the dishes, the shovel, and the pan in the sink and left them on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_15", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles are placed in front of the sink?", + "answer": "Six bottles were placed.", + "question_chinese": "洗水槽前面摆了几个瓶子？", + "answer_chinese": "摆了六个瓶子。", + "evidence": [ + { + "description": "Six bottles were placed.", + "timestamps": { + "start": "00:09:43", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:10:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five bottles were placed.", + "B": "Six bottles were placed.", + "C": "Seven bottles were placed.", + "D": "Four bottles were placed in front of the sink, lined up together along the counter edge.", + "E": "Eight bottles were placed in front of the sink, grouped in a cluster rather than a single row." + }, + "answer_label": "B", + "answer_text": "Six bottles were placed.", + "q_id": "P36_101_15", + "correct_answer": "Six bottles were placed.", + "original_options": { + "A": "Five bottles were placed.", + "B": "Six bottles were placed.", + "C": "Seven bottles were placed.", + "D": "Four bottles were placed in front of the sink, lined up together along the counter edge.", + "E": "Eight bottles were placed in front of the sink, grouped in a cluster rather than a single row." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 18 + ] + }, + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_16", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles are placed in front of the sink?", + "answer": "Six bottles were placed.", + "question_chinese": "洗水槽前面摆了几个瓶子？", + "answer_chinese": "摆了六个瓶子。", + "evidence": [ + { + "description": "Six bottles were placed.", + "timestamps": { + "start": "00:09:43", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:11:22", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five bottles were placed.", + "B": "Eight bottles were placed.", + "C": "Six bottles were placed.", + "D": "Two bottles were placed in front of the sink, with space left between them on the counter.", + "E": "Nine bottles were placed in front of the sink, spread across the area in a loose row." + }, + "answer_label": "C", + "answer_text": "Six bottles were placed.", + "q_id": "P36_101_16", + "correct_answer": "Six bottles were placed.", + "original_options": { + "A": "Five bottles were placed.", + "B": "Eight bottles were placed.", + "C": "Six bottles were placed.", + "D": "Two bottles were placed in front of the sink, with space left between them on the counter.", + "E": "Nine bottles were placed in front of the sink, spread across the area in a loose row." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_13", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the sink on the cutting board?", + "answer": "The sink is on the left side of the cutting board.", + "question_chinese": "洗水槽在菜板的哪个位置？", + "answer_chinese": "洗水槽在菜板的左边。", + "evidence": [ + { + "description": "The sink is on the left side of the cutting board.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:32" + } + } + ], + "question_timestamps": "00:10:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s on the right side.", + "B": "It’s in front of the board.", + "C": "The sink is on the right side of the cutting board, not along the left edge.", + "D": "The sink sits behind the cutting board, so it isn’t positioned on the left side.", + "E": "The sink is on the left side of the cutting board." + }, + "answer_label": "E", + "answer_text": "The sink is on the left side of the cutting board.", + "q_id": "P36_101_13", + "correct_answer": "The sink is on the left side of the cutting board.", + "original_options": { + "A": "It’s on the right side.", + "B": "It’s in front of the board.", + "C": "The sink is on the right side of the cutting board, not along the left edge.", + "D": "The sink sits behind the cutting board, so it isn’t positioned on the left side.", + "E": "The sink is on the left side of the cutting board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 16 + ] + }, + { + "video_uid": "P36_101", + "video_length": "00:11:42", + "qid": "P36_101_14", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the sink on the cutting board?", + "answer": "The sink is on the left side of the cutting board.", + "question_chinese": "洗水槽在菜板的哪个位置？", + "answer_chinese": "洗水槽在菜板的左边。", + "evidence": [ + { + "description": "The sink is on the left side of the cutting board.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:32" + } + } + ], + "question_timestamps": "00:11:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s on the right side.", + "B": "It’s at the bottom edge.", + "C": "The sink is on the left side of the cutting board.", + "D": "The sink is on the right side of the cutting board, closer to the right corner.", + "E": "The sink is in front of the cutting board rather than on the left side of it." + }, + "answer_label": "C", + "answer_text": "The sink is on the left side of the cutting board.", + "q_id": "P36_101_14", + "correct_answer": "The sink is on the left side of the cutting board.", + "original_options": { + "A": "It’s on the right side.", + "B": "It’s at the bottom edge.", + "C": "The sink is on the left side of the cutting board.", + "D": "The sink is on the right side of the cutting board, closer to the right corner.", + "E": "The sink is in front of the cutting board rather than on the left side of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + } +] diff --git a/QAs/P37_101-mcq.json b/QAs/P37_101-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..eefe97fdc3b2eb42ddd5cd7d27bc7072dab806d9 --- /dev/null +++ b/QAs/P37_101-mcq.json @@ -0,0 +1,1400 @@ +[ + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_11", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pumpkins are placed beside the refrigerator?", + "answer": "Two pumpkins were put in.", + "question_chinese": "冰箱旁边放了几个南瓜？", + "answer_chinese": "放了两个南瓜。", + "evidence": [ + { + "description": "Two pumpkins were put in.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:07:11", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One pumpkin was put in.", + "B": "Three pumpkins were put in.", + "C": "Four pumpkins were placed beside the refrigerator, arranged together in the same spot.", + "D": "Two pumpkins were put in.", + "E": "Five pumpkins were set beside the refrigerator, grouped near the side rather than elsewhere." + }, + "answer_label": "D", + "answer_text": "Two pumpkins were put in.", + "q_id": "P37_101_11", + "correct_answer": "Two pumpkins were put in.", + "original_options": { + "A": "One pumpkin was put in.", + "B": "Three pumpkins were put in.", + "C": "Four pumpkins were placed beside the refrigerator, arranged together in the same spot.", + "D": "Two pumpkins were put in.", + "E": "Five pumpkins were set beside the refrigerator, grouped near the side rather than elsewhere." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_12", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pumpkins are placed beside the refrigerator?", + "answer": "Two pumpkins were put in.", + "question_chinese": "冰箱旁边放了几个南瓜？", + "answer_chinese": "放了两个南瓜。", + "evidence": [ + { + "description": "Two pumpkins were put in.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:13:32", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One pumpkin was put in.", + "B": "Three pumpkins were put in.", + "C": "Four pumpkins were placed beside the refrigerator, positioned together along the side.", + "D": "Six pumpkins were put in beside the refrigerator, forming a larger cluster than just two.", + "E": "Two pumpkins were put in." + }, + "answer_label": "E", + "answer_text": "Two pumpkins were put in.", + "q_id": "P37_101_12", + "correct_answer": "Two pumpkins were put in.", + "original_options": { + "A": "One pumpkin was put in.", + "B": "Three pumpkins were put in.", + "C": "Four pumpkins were placed beside the refrigerator, positioned together along the side.", + "D": "Six pumpkins were put in beside the refrigerator, forming a larger cluster than just two.", + "E": "Two pumpkins were put in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 15 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chicken legs did I cook in the pot in total?", + "answer": "You cooked a total of nine chicken legs.", + "question_chinese": "我一共在锅里煮了几个鸡腿？", + "answer_chinese": "你一共煮了九个鸡腿。", + "evidence": [ + { + "description": "There are nine chicken legs in the box.", + "timestamps": { + "start": "00:02:05", + "end": "00:02:07" + } + }, + { + "description": "You poured all the chicken legs in the bowl into the pot.", + "timestamps": { + "start": "00:02:54", + "end": "00:02:57" + } + }, + { + "description": "You put all nine chicken legs into the bowl.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:48" + } + } + ], + "question_timestamps": "00:05:32", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You cooked six chicken legs.", + "B": "You cooked ten chicken legs.", + "C": "You cooked a total of eight chicken legs in the pot altogether.", + "D": "You cooked a total of twelve chicken legs in the pot from start to finish.", + "E": "You cooked a total of nine chicken legs." + }, + "answer_label": "E", + "answer_text": "You cooked a total of nine chicken legs.", + "q_id": "P37_101_5", + "correct_answer": "You cooked a total of nine chicken legs.", + "original_options": { + "A": "You cooked six chicken legs.", + "B": "You cooked ten chicken legs.", + "C": "You cooked a total of eight chicken legs in the pot altogether.", + "D": "You cooked a total of twelve chicken legs in the pot from start to finish.", + "E": "You cooked a total of nine chicken legs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 15 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chicken legs did I cook in the pot in total?", + "answer": "You cooked a total of nine chicken legs.", + "question_chinese": "我一共在锅里煮了几个鸡腿？", + "answer_chinese": "你一共煮了九个鸡腿。", + "evidence": [ + { + "description": "There are nine chicken legs in the box.", + "timestamps": { + "start": "00:02:05", + "end": "00:02:07" + } + }, + { + "description": "You poured all the chicken legs in the bowl into the pot.", + "timestamps": { + "start": "00:02:54", + "end": "00:02:57" + } + }, + { + "description": "You put all nine chicken legs into the bowl.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:48" + } + } + ], + "question_timestamps": "00:16:32", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You cooked seven chicken legs.", + "B": "You cooked a total of nine chicken legs.", + "C": "You cooked eleven chicken legs.", + "D": "You cooked a total of five chicken legs in the pot overall.", + "E": "You cooked a total of thirteen chicken legs in the pot altogether." + }, + "answer_label": "B", + "answer_text": "You cooked a total of nine chicken legs.", + "q_id": "P37_101_6", + "correct_answer": "You cooked a total of nine chicken legs.", + "original_options": { + "A": "You cooked seven chicken legs.", + "B": "You cooked a total of nine chicken legs.", + "C": "You cooked eleven chicken legs.", + "D": "You cooked a total of five chicken legs in the pot overall.", + "E": "You cooked a total of thirteen chicken legs in the pot altogether." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_7_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where did I take the chicken legs out from?", + "answer": "You took the chicken drumsticks out of the refrigerator; the refrigerator was in front of you.", + "question_chinese": "我从哪里拿出了鸡腿？", + "answer_chinese": "你从冰箱拿出了鸡腿，冰箱在你前面。", + "evidence": [ + { + "description": "You took the chickend legs out of the fridge.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Out of the freezer.", + "B": "Off the counter.", + "C": "You took the chicken drumsticks out of the refrigerator; the refrigerator was in front of you.", + "D": "You took the chicken drumsticks out of the refrigerator, but the refrigerator was behind you.", + "E": "You took the chicken drumsticks out of the refrigerator, and it was positioned to your left." + }, + "answer_label": "C", + "answer_text": "You took the chicken drumsticks out of the refrigerator; the refrigerator was in front of you.", + "q_id": "P37_101_7_T1", + "correct_answer": "You took the chicken drumsticks out of the refrigerator; the refrigerator was in front of you.", + "original_options": { + "A": "Out of the freezer.", + "B": "Off the counter.", + "C": "You took the chicken drumsticks out of the refrigerator; the refrigerator was in front of you.", + "D": "You took the chicken drumsticks out of the refrigerator, but the refrigerator was behind you.", + "E": "You took the chicken drumsticks out of the refrigerator, and it was positioned to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 16 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_7_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where did I take the chicken legs out from?", + "answer": "You took the chicken drumsticks off the table; the table was in front of you.", + "question_chinese": "我从哪里拿出了鸡腿？", + "answer_chinese": "你从桌子拿出了鸡腿，桌子在你前面。", + "evidence": [ + { + "description": "You took the chocken legs out of the table.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:01:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Out of the fridge.", + "B": "Off the counter.", + "C": "You took the chicken drumsticks off the table, but the table was behind you.", + "D": "You took the chicken drumsticks off the table, and the table was to your right.", + "E": "You took the chicken drumsticks off the table; the table was in front of you." + }, + "answer_label": "E", + "answer_text": "You took the chicken drumsticks off the table; the table was in front of you.", + "q_id": "P37_101_7_T2", + "correct_answer": "You took the chicken drumsticks off the table; the table was in front of you.", + "original_options": { + "A": "Out of the fridge.", + "B": "Off the counter.", + "C": "You took the chicken drumsticks off the table, but the table was behind you.", + "D": "You took the chicken drumsticks off the table, and the table was to your right.", + "E": "You took the chicken drumsticks off the table; the table was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 15 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_6_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I stretch out my hand, can I reach the cutting board?", + "answer": "You could reach it; the cutting board was right in front of you.", + "question_chinese": "我伸手能够拿到菜板吗？", + "answer_chinese": "你能够拿到，菜板就在你的面前。", + "evidence": [ + { + "description": "You can get it. The cutting board is right in front of you.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You could reach it; the cutting board was right in front of you.", + "B": "You couldn’t reach it; the cutting board was right behind you." + }, + "answer_label": "A", + "answer_text": "You could reach it; the cutting board was right in front of you.", + "q_id": "P37_101_6_T1", + "correct_answer": "You could reach it; the cutting board was right in front of you.", + "original_options": { + "A": "You could reach it; the cutting board was right in front of you.", + "B": "You couldn’t reach it; the cutting board was right behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_6_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I stretch out my hand, can I reach the cutting board?", + "answer": "You couldn’t reach it; the cutting board was right behind you.", + "question_chinese": "我伸手能够拿到菜板吗？", + "answer_chinese": "你不能够拿到，菜板就在你的后面。", + "evidence": [ + { + "description": "You can not get it. The cutting board is right in behind you.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:28" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You could reach it; the cutting board was right in front of you.", + "B": "You couldn’t reach it; the cutting board was right behind you." + }, + "answer_label": "B", + "answer_text": "You couldn’t reach it; the cutting board was right behind you.", + "q_id": "P37_101_6_T2", + "correct_answer": "You couldn’t reach it; the cutting board was right behind you.", + "original_options": { + "A": "You could reach it; the cutting board was right in front of you.", + "B": "You couldn’t reach it; the cutting board was right behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_15", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I need scallions for cooking. Where can I get them?", + "answer": "You can get the scallions in the fridge.", + "question_chinese": "我做饭需要葱，我从哪里能够拿到？", + "answer_chinese": "你可以在冰箱里拿到葱。", + "evidence": [ + { + "description": "The scallions are in the refrigerator.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:13:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the vegetable drawer.", + "B": "You can get the scallions in the fridge.", + "C": "On the kitchen counter.", + "D": "You can get the scallions on the kitchen shelf near the other ingredients.", + "E": "You can get the scallions in the spice cabinet where you keep dried seasonings." + }, + "answer_label": "B", + "answer_text": "You can get the scallions in the fridge.", + "q_id": "P37_101_15", + "correct_answer": "You can get the scallions in the fridge.", + "original_options": { + "A": "In the vegetable drawer.", + "B": "You can get the scallions in the fridge.", + "C": "On the kitchen counter.", + "D": "You can get the scallions on the kitchen shelf near the other ingredients.", + "E": "You can get the scallions in the spice cabinet where you keep dried seasonings." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_16", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I need scallions for cooking. Where can I get them?", + "answer": "You can get the scallions in the fridge.", + "question_chinese": "我做饭需要葱，我从哪里能够拿到？", + "answer_chinese": "你可以在冰箱里拿到葱。", + "evidence": [ + { + "description": "The scallions are in the refrigerator.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:19:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the vegetable drawer.", + "B": "On the kitchen counter.", + "C": "You can get the scallions on the kitchen shelf near the other produce.", + "D": "You can get the scallions in the fridge.", + "E": "You can get the scallions in a pantry cabinet away from the refrigerator." + }, + "answer_label": "D", + "answer_text": "You can get the scallions in the fridge.", + "q_id": "P37_101_16", + "correct_answer": "You can get the scallions in the fridge.", + "original_options": { + "A": "In the vegetable drawer.", + "B": "On the kitchen counter.", + "C": "You can get the scallions on the kitchen shelf near the other produce.", + "D": "You can get the scallions in the fridge.", + "E": "You can get the scallions in a pantry cabinet away from the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 13 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I need chicken legs for cooking. Where can I find chicken legs?", + "answer": "The chicken leg you found in the second compartment of the refrigerator.", + "question_chinese": "我做饭需要鸡腿，我从哪里找到的鸡腿？", + "answer_chinese": "你在冰箱的第二格找到的鸡腿。", + "evidence": [ + { + "description": "The chicken leg you found in the second compartment of the refrigerator.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:09" + } + }, + { + "description": "The refrigerator.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:04:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In the first refrigerator compartment.", + "B": "In the freezer compartment.", + "C": "The chicken legs are in the top compartment of the refrigerator, not the second one you checked earlier.", + "D": "You can find the chicken legs in the refrigerator door shelf, rather than inside the second compartment.", + "E": "The chicken leg you found in the second compartment of the refrigerator." + }, + "answer_label": "E", + "answer_text": "The chicken leg you found in the second compartment of the refrigerator.", + "q_id": "P37_101_1", + "correct_answer": "The chicken leg you found in the second compartment of the refrigerator.", + "original_options": { + "A": "In the first refrigerator compartment.", + "B": "In the freezer compartment.", + "C": "The chicken legs are in the top compartment of the refrigerator, not the second one you checked earlier.", + "D": "You can find the chicken legs in the refrigerator door shelf, rather than inside the second compartment.", + "E": "The chicken leg you found in the second compartment of the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 17 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I need chicken legs for cooking. Where can I find chicken legs?", + "answer": "The chicken leg you found in the second compartment of the refrigerator.", + "question_chinese": "我做饭需要鸡腿，我从哪里找到的鸡腿？", + "answer_chinese": "你在冰箱的第二格找到的鸡腿。", + "evidence": [ + { + "description": "The chicken leg you found in the second compartment of the refrigerator.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:09" + } + }, + { + "description": "The refrigerator.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:08:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In the first refrigerator compartment.", + "B": "On the pantry shelf.", + "C": "The chicken legs are in the freezer compartment next to the ice trays, not in the refrigerator section.", + "D": "The chicken leg you found in the second compartment of the refrigerator.", + "E": "You left the chicken legs under the kitchen counter, instead of placing them in the second refrigerator compartment." + }, + "answer_label": "D", + "answer_text": "The chicken leg you found in the second compartment of the refrigerator.", + "q_id": "P37_101_2", + "correct_answer": "The chicken leg you found in the second compartment of the refrigerator.", + "original_options": { + "A": "In the first refrigerator compartment.", + "B": "On the pantry shelf.", + "C": "The chicken legs are in the freezer compartment next to the ice trays, not in the refrigerator section.", + "D": "The chicken leg you found in the second compartment of the refrigerator.", + "E": "You left the chicken legs under the kitchen counter, instead of placing them in the second refrigerator compartment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 18 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_9", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I move the black bench from and to?", + "answer": "You moved the black stool from the front of the cabinet to the door.", + "question_chinese": "我将黑板凳从哪里搬到了哪里？", + "answer_chinese": "你把黑板凳从柜子前面搬到了门前面。", + "evidence": [ + { + "description": "The black stool is in front of the cabinet.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + }, + { + "description": "You moved the black stool to the front of the door.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:03:44", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You moved the black stool from the front of the cabinet to the door.", + "B": "From the cabinet front to the chair.", + "C": "From the cabinet back to the floor.", + "D": "You moved the black stool from the left side of the cabinet over to the wall nearby.", + "E": "You moved the black stool from the top of the cabinet down to the desk, not toward the door." + }, + "answer_label": "A", + "answer_text": "You moved the black stool from the front of the cabinet to the door.", + "q_id": "P37_101_9", + "correct_answer": "You moved the black stool from the front of the cabinet to the door.", + "original_options": { + "A": "You moved the black stool from the front of the cabinet to the door.", + "B": "From the cabinet front to the chair.", + "C": "From the cabinet back to the floor.", + "D": "You moved the black stool from the left side of the cabinet over to the wall nearby.", + "E": "You moved the black stool from the top of the cabinet down to the desk, not toward the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 19 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_10", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I move the black bench from and to?", + "answer": "You moved the black stool from the front of the cabinet to the door.", + "question_chinese": "我将黑板凳从哪里搬到了哪里？", + "answer_chinese": "你把黑板凳从柜子前面搬到了门前面。", + "evidence": [ + { + "description": "The black stool is in front of the cabinet.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + }, + { + "description": "You moved the black stool to the front of the door.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:07:54", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From the cabinet left to the window.", + "B": "From behind the cabinet to the corner.", + "C": "You moved the black stool from the front of the cabinet to the door.", + "D": "You moved the black stool from the bottom of the cabinet to the table instead of taking it to the door.", + "E": "You moved the black stool from the middle area by the cabinet over to the shelf, not from the front." + }, + "answer_label": "C", + "answer_text": "You moved the black stool from the front of the cabinet to the door.", + "q_id": "P37_101_10", + "correct_answer": "You moved the black stool from the front of the cabinet to the door.", + "original_options": { + "A": "From the cabinet left to the window.", + "B": "From behind the cabinet to the corner.", + "C": "You moved the black stool from the front of the cabinet to the door.", + "D": "You moved the black stool from the bottom of the cabinet to the table instead of taking it to the door.", + "E": "You moved the black stool from the middle area by the cabinet over to the shelf, not from the front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 20 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_3_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chicken legs did I wash?", + "answer": "You washed 0 chicken drumsticks; they were to your left.", + "question_chinese": "我洗了几个鸡腿？", + "answer_chinese": "你洗了0个鸡腿，它在你左边。", + "evidence": [ + { + "description": "There are nine chicken legs in the box.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:03" + } + } + ], + "question_timestamps": "00:02:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed 1 drumstick.", + "B": "You washed 0 chicken drumsticks; they were to your left.", + "C": "You washed 2 drumsticks.", + "D": "You washed 3 chicken drumsticks, and they were positioned to your left while you rinsed them.", + "E": "You washed 4 chicken drumsticks; they stayed to your left as you washed them at the sink." + }, + "answer_label": "B", + "answer_text": "You washed 0 chicken drumsticks; they were to your left.", + "q_id": "P37_101_3_T1", + "correct_answer": "You washed 0 chicken drumsticks; they were to your left.", + "original_options": { + "A": "You washed 1 drumstick.", + "B": "You washed 0 chicken drumsticks; they were to your left.", + "C": "You washed 2 drumsticks.", + "D": "You washed 3 chicken drumsticks, and they were positioned to your left while you rinsed them.", + "E": "You washed 4 chicken drumsticks; they stayed to your left as you washed them at the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_3_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chicken legs did I wash?", + "answer": "You washed 9 chicken drumsticks; they were in front of you.", + "question_chinese": "我洗了几个鸡腿？", + "answer_chinese": "你洗了9个鸡腿，它在你前面。", + "evidence": [ + { + "description": "You washed nine chicken legs.", + "timestamps": { + "start": "00:02:07", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:02:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You washed 8 drumsticks.", + "B": "You washed 10 drumsticks.", + "C": "You washed 9 chicken drumsticks; they were in front of you.", + "D": "You washed 7 chicken drumsticks, and they were lined up in front of you while you cleaned them.", + "E": "You washed 11 chicken drumsticks; they were arranged directly in front of you during washing." + }, + "answer_label": "C", + "answer_text": "You washed 9 chicken drumsticks; they were in front of you.", + "q_id": "P37_101_3_T2", + "correct_answer": "You washed 9 chicken drumsticks; they were in front of you.", + "original_options": { + "A": "You washed 8 drumsticks.", + "B": "You washed 10 drumsticks.", + "C": "You washed 9 chicken drumsticks; they were in front of you.", + "D": "You washed 7 chicken drumsticks, and they were lined up in front of you while you cleaned them.", + "E": "You washed 11 chicken drumsticks; they were arranged directly in front of you during washing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 15 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_13", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I'm washing dishes in front of the sink now. Can I reach the cutting board with my hand?", + "answer": "You can get it. The cutting board is right in front of you.", + "question_chinese": "我现在在洗水槽前面洗碗，我伸手能够拿到菜板吗？", + "answer_chinese": "能够拿到，菜板就在你的面前。", + "evidence": [ + { + "description": "You can get it. The cutting board is right in front of you.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:08:31", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot get it. The cutting board is out of reach to your right.", + "B": "You can get it. The cutting board is right in front of you." + }, + "answer_label": "B", + "answer_text": "You can get it. The cutting board is right in front of you.", + "q_id": "P37_101_13", + "correct_answer": "You can get it. The cutting board is right in front of you.", + "original_options": { + "A": "You cannot get it. The cutting board is out of reach to your right.", + "B": "You can get it. The cutting board is right in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_14", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I'm washing dishes in front of the sink now. Can I reach the cutting board with my hand?", + "answer": "You can get it. The cutting board is right in front of you.", + "question_chinese": "我现在在洗水槽前面洗碗，我伸手能够拿到菜板吗？", + "answer_chinese": "能够拿到，菜板就在你的面前。", + "evidence": [ + { + "description": "You can get it. The cutting board is right in front of you.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:022:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot get it. The cutting board is out of reach to your right.", + "B": "You can get it. The cutting board is right in front of you." + }, + "answer_label": "B", + "answer_text": "You can get it. The cutting board is right in front of you.", + "q_id": "P37_101_14", + "correct_answer": "You can get it. The cutting board is right in front of you.", + "original_options": { + "A": "You cannot get it. The cutting board is out of reach to your right.", + "B": "You can get it. The cutting board is right in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_1_T1", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where is the refrigerator relative to me?", + "answer": "At this moment, the refrigerator was in front of you.", + "question_chinese": "此时，冰箱在我哪个方位？", + "answer_chinese": "此时，冰箱在你前面.", + "evidence": [ + { + "description": "The refrigerator is in front of you.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was behind you.", + "B": "It was to your left.", + "C": "At this moment, the refrigerator was in front of you.", + "D": "At this moment, the refrigerator was positioned to your right as you faced forward.", + "E": "At this moment, the refrigerator was above you, higher than your eye level." + }, + "answer_label": "C", + "answer_text": "At this moment, the refrigerator was in front of you.", + "q_id": "P37_101_1_T1", + "correct_answer": "At this moment, the refrigerator was in front of you.", + "original_options": { + "A": "It was behind you.", + "B": "It was to your left.", + "C": "At this moment, the refrigerator was in front of you.", + "D": "At this moment, the refrigerator was positioned to your right as you faced forward.", + "E": "At this moment, the refrigerator was above you, higher than your eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 13 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_1_T2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where is the refrigerator relative to me?", + "answer": "At this moment, the refrigerator was behind you.", + "question_chinese": "此时，冰箱在我哪个方位？", + "answer_chinese": "此时，冰箱在你后面", + "evidence": [ + { + "description": "The refrigerator is behind of you..", + "timestamps": { + "start": "00:00:13", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "At this moment, the refrigerator was behind you.", + "B": "It was in front of you.", + "C": "It was to your right.", + "D": "At this moment, the refrigerator was to your left, not behind you where you were standing.", + "E": "At this moment, the refrigerator was above you, rather than behind your back." + }, + "answer_label": "A", + "answer_text": "At this moment, the refrigerator was behind you.", + "q_id": "P37_101_1_T2", + "correct_answer": "At this moment, the refrigerator was behind you.", + "original_options": { + "A": "At this moment, the refrigerator was behind you.", + "B": "It was in front of you.", + "C": "It was to your right.", + "D": "At this moment, the refrigerator was to your left, not behind you where you were standing.", + "E": "At this moment, the refrigerator was above you, rather than behind your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 13 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_5_T1", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, where is the pumpkin relative to me?", + "answer": "At this moment, the pumpkin was to your left.", + "question_chinese": "此时，南瓜在我哪个方位？", + "answer_chinese": "此时，南瓜在你左边。", + "evidence": [ + { + "description": "Two pumpkins is in your left.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It was in front of you.", + "B": "At this moment, the pumpkin was to your left.", + "C": "It was behind you.", + "D": "At this moment, the pumpkin was to your right, not on your left side.", + "E": "At this moment, the pumpkin was above you, positioned higher up than your head." + }, + "answer_label": "B", + "answer_text": "At this moment, the pumpkin was to your left.", + "q_id": "P37_101_5_T1", + "correct_answer": "At this moment, the pumpkin was to your left.", + "original_options": { + "A": "It was in front of you.", + "B": "At this moment, the pumpkin was to your left.", + "C": "It was behind you.", + "D": "At this moment, the pumpkin was to your right, not on your left side.", + "E": "At this moment, the pumpkin was above you, positioned higher up than your head." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 14, + 14 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_5_T2", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, where is the pumpkin relative to me?", + "answer": "At this moment, the pumpkin was behind you.", + "question_chinese": "此时，南瓜在我哪个方位？", + "answer_chinese": "此时，南瓜在你后面。", + "evidence": [ + { + "description": "Two pumpkins is behind of you.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It was in front of you.", + "B": "At this moment, the pumpkin was behind you.", + "C": "It was to your left.", + "D": "At this moment, the pumpkin was to your right rather than directly behind you.", + "E": "At this moment, the pumpkin was above you, not located behind your body." + }, + "answer_label": "B", + "answer_text": "At this moment, the pumpkin was behind you.", + "q_id": "P37_101_5_T2", + "correct_answer": "At this moment, the pumpkin was behind you.", + "original_options": { + "A": "It was in front of you.", + "B": "At this moment, the pumpkin was behind you.", + "C": "It was to your left.", + "D": "At this moment, the pumpkin was to your right rather than directly behind you.", + "E": "At this moment, the pumpkin was above you, not located behind your body." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 13 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_2_T1", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, can I directly touch the faucet?", + "answer": "At this moment, you couldn’t directly reach the faucet; it was to your right.", + "question_chinese": "此时，我能直接接触水龙头吗？", + "answer_chinese": "此时，你不能直接接触水龙头，它在你右边。", + "evidence": [ + { + "description": "You can not touch the faucet.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You could reach it directly.", + "B": "It was in front of you.", + "C": "At this moment, you couldn’t directly reach the faucet; it was positioned to your left.", + "D": "At this moment, you couldn’t directly reach the faucet; it was behind you instead of to your right.", + "E": "At this moment, you couldn’t directly reach the faucet; it was to your right." + }, + "answer_label": "E", + "answer_text": "At this moment, you couldn’t directly reach the faucet; it was to your right.", + "q_id": "P37_101_2_T1", + "correct_answer": "At this moment, you couldn’t directly reach the faucet; it was to your right.", + "original_options": { + "A": "You could reach it directly.", + "B": "It was in front of you.", + "C": "At this moment, you couldn’t directly reach the faucet; it was positioned to your left.", + "D": "At this moment, you couldn’t directly reach the faucet; it was behind you instead of to your right.", + "E": "At this moment, you couldn’t directly reach the faucet; it was to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 19 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_2_T2", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, can I directly touch the faucet?", + "answer": "At this moment, you could directly reach the faucet; it was in front of you.", + "question_chinese": "此时，我能直接接触水龙头吗？", + "answer_chinese": "此时，你能直接接触水龙头，它在你前面。", + "evidence": [ + { + "description": "You can touch the faucet.", + "timestamps": { + "start": "00:00:51", + "end": "00:00:53" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You couldn’t reach it directly.", + "B": "It was to your left.", + "C": "At this moment, you could directly reach the faucet; it was to your right, not in front of you.", + "D": "At this moment, you could directly reach the faucet; it was in front of you.", + "E": "At this moment, you could directly reach the faucet; it was behind you rather than straight ahead." + }, + "answer_label": "D", + "answer_text": "At this moment, you could directly reach the faucet; it was in front of you.", + "q_id": "P37_101_2_T2", + "correct_answer": "At this moment, you could directly reach the faucet; it was in front of you.", + "original_options": { + "A": "You couldn’t reach it directly.", + "B": "It was to your left.", + "C": "At this moment, you could directly reach the faucet; it was to your right, not in front of you.", + "D": "At this moment, you could directly reach the faucet; it was in front of you.", + "E": "At this moment, you could directly reach the faucet; it was behind you rather than straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 17 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_4_T1", + "qid_num": 25, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the black stool relative to me?", + "answer": "The black stool was to your right.", + "question_chinese": "黑板凳在我哪个方位？", + "answer_chinese": "黑板凳在你右边。", + "evidence": [ + { + "description": "The black stool is on your right.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "The black stool was to your right.", + "C": "It was to your left.", + "D": "The black stool was behind you, not on your right side at that moment.", + "E": "The black stool was above you, positioned higher than where you were looking." + }, + "answer_label": "B", + "answer_text": "The black stool was to your right.", + "q_id": "P37_101_4_T1", + "correct_answer": "The black stool was to your right.", + "original_options": { + "A": "It was in front of you.", + "B": "The black stool was to your right.", + "C": "It was to your left.", + "D": "The black stool was behind you, not on your right side at that moment.", + "E": "The black stool was above you, positioned higher than where you were looking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 13 + ] + }, + { + "video_uid": "P37_101", + "video_length": "00:28:01", + "qid": "P37_101_4_T2", + "qid_num": 26, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the black stool relative to me?", + "answer": "The black stool was in front of you.", + "question_chinese": "黑板凳在我哪个方位？", + "answer_chinese": "黑板凳在你前面", + "evidence": [ + { + "description": "The black stool is in front of you.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was behind you.", + "B": "The black stool was in front of you.", + "C": "It was to your right.", + "D": "The black stool was to your left rather than directly in front of you.", + "E": "The black stool was above you, not in front at your standing height." + }, + "answer_label": "B", + "answer_text": "The black stool was in front of you.", + "q_id": "P37_101_4_T2", + "correct_answer": "The black stool was in front of you.", + "original_options": { + "A": "It was behind you.", + "B": "The black stool was in front of you.", + "C": "It was to your right.", + "D": "The black stool was to your left rather than directly in front of you.", + "E": "The black stool was above you, not in front at your standing height." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 13 + ] + } +] diff --git a/QAs/adf11624-aca0-4bad-a35f-51b9f330ee74-001-mcq.json b/QAs/adf11624-aca0-4bad-a35f-51b9f330ee74-001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c6ee80e2129d36bbb50b3fabb65841485919aa45 --- /dev/null +++ b/QAs/adf11624-aca0-4bad-a35f-51b9f330ee74-001-mcq.json @@ -0,0 +1,420 @@ +[ + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001", + "video_length": "00:20:01", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did anyone ride past me just now?", + "answer": "No.", + "question_chinese": "刚才有人骑车经过我了吗？", + "answer_chinese": "没有", + "evidence": [ + { + "description": "No one rode past me.", + "timestamps": { + "start": "00:07:30", + "end": "00:07:31" + } + } + ], + "question_timestamps": "00:07:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_7", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001", + "video_length": "00:20:01", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did anyone ride past me just now?", + "answer": "Yes.", + "question_chinese": "刚才有人骑车经过我了吗？", + "answer_chinese": "有", + "evidence": [ + { + "description": "Someone rode past me.", + "timestamps": { + "start": "00:07:55", + "end": "00:07:56" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_8", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001", + "video_length": "00:20:01", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_15", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I been walking on the asphalt road all the time?", + "answer": "Yes.", + "question_chinese": "我一直走在沥青路上吗？", + "answer_chinese": "是", + "evidence": [ + { + "description": "On the asphalt road", + "timestamps": { + "start": "00:11:00", + "end": "00:11:01" + } + } + ], + "question_timestamps": "00:11:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_15", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 2, + "B": 3 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001", + "video_length": "00:20:01", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_16", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I been walking on the asphalt road all the time?", + "answer": "No.", + "question_chinese": "我一直走在沥青路上吗？", + "answer_chinese": "不是", + "evidence": [ + { + "description": "On the asphalt road", + "timestamps": { + "start": "00:11:00", + "end": "00:11:01" + } + }, + { + "description": "On the flagstone road", + "timestamps": { + "start": "00:12:00", + "end": "00:12:01" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_16", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 3, + "B": 3 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001", + "video_length": "00:20:01", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I make when I walked along the road just now?", + "answer": "Two.", + "question_chinese": "我刚才沿着路走，转了多少个弯？", + "answer_chinese": "2", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:01:05", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four. You made four turns while walking along the road just now.", + "D": "Five. You turned five separate times as you continued walking down the road.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_1", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four. You made four turns while walking along the road just now.", + "D": "Five. You turned five separate times as you continued walking down the road.", + "E": "Two." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 0, + "D": 0, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001", + "video_length": "00:20:01", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns did I make when I walked along the road just now?", + "answer": "Three.", + "question_chinese": "我刚才沿着路走，转了多少个弯？", + "answer_chinese": "3", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:01:11", + "end": "00:01:12" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "You made four turns while walking along the road, changing direction at several corners.", + "D": "Three.", + "E": "You made five turns along the road, turning repeatedly as you followed the route." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-001_2", + "correct_answer": "Three.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "You made four turns while walking along the road, changing direction at several corners.", + "D": "Three.", + "E": "You made five turns along the road, turning repeatedly as you followed the route." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 0, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + } +] diff --git a/QAs/adf11624-aca0-4bad-a35f-51b9f330ee74-002-mcq.json b/QAs/adf11624-aca0-4bad-a35f-51b9f330ee74-002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..906fdf51c61f5392b3fcfcb3a3be13feaf3e5c0c --- /dev/null +++ b/QAs/adf11624-aca0-4bad-a35f-51b9f330ee74-002-mcq.json @@ -0,0 +1,1117 @@ +[ + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I make a turn just now?", + "answer": "Two turns.", + "question_chinese": "刚才我一共拐过了几次弯？", + "answer_chinese": "两次弯", + "evidence": [ + { + "description": "the first one.", + "timestamps": { + "start": "00:04:46", + "end": "00:04:48" + } + }, + { + "description": "the second one.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:40" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One turn.", + "B": "Two turns.", + "C": "Three turns.", + "D": "You made just one turn and then continued straight for the rest of the way.", + "E": "You made three turns in total, changing direction at each bend along the path." + }, + "answer_label": "B", + "answer_text": "Two turns.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_1", + "correct_answer": "Two turns.", + "original_options": { + "A": "One turn.", + "B": "Two turns.", + "C": "Three turns.", + "D": "You made just one turn and then continued straight for the rest of the way.", + "E": "You made three turns in total, changing direction at each bend along the path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 0, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I make a turn just now?", + "answer": "Three turns.", + "question_chinese": "刚才我一共拐过了几次弯？", + "answer_chinese": "三次弯.", + "evidence": [ + { + "description": "the first one.", + "timestamps": { + "start": "00:04:46", + "end": "00:04:48" + } + }, + { + "description": "the second one.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:40" + } + }, + { + "description": "the third one.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:21" + } + } + ], + "question_timestamps": "00:06:35", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two turns.", + "B": "Four turns.", + "C": "You only made one turn before heading straight again for the remainder of the walk.", + "D": "Three turns.", + "E": "You made five turns overall, turning at every corner you reached along the way." + }, + "answer_label": "D", + "answer_text": "Three turns.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_2", + "correct_answer": "Three turns.", + "original_options": { + "A": "Two turns.", + "B": "Four turns.", + "C": "You only made one turn before heading straight again for the remainder of the walk.", + "D": "Three turns.", + "E": "You made five turns overall, turning at every corner you reached along the way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 0, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_7", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was I doing when I stopped just now?", + "answer": "You took off your shoes and emptied the small stones in them.", + "question_chinese": "我刚才停下来是在干什么？", + "answer_chinese": "你脱了鞋在倒鞋里的小石子。", + "evidence": [ + { + "description": "You took off your shoes and emptied the small stones in them.", + "timestamps": { + "start": "00:26:44", + "end": "00:26:47" + } + } + ], + "question_timestamps": "00:27:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You were tying your shoes.", + "B": "You were adjusting your shoelaces.", + "C": "You took off your shoes and shook out dirt from inside them onto the ground.", + "D": "You took off your shoes and picked out leaves from inside them before putting them back on.", + "E": "You took off your shoes and emptied the small stones in them." + }, + "answer_label": "E", + "answer_text": "You took off your shoes and emptied the small stones in them.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_7", + "correct_answer": "You took off your shoes and emptied the small stones in them.", + "original_options": { + "A": "You were tying your shoes.", + "B": "You were adjusting your shoelaces.", + "C": "You took off your shoes and shook out dirt from inside them onto the ground.", + "D": "You took off your shoes and picked out leaves from inside them before putting them back on.", + "E": "You took off your shoes and emptied the small stones in them." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 1, + "D": 1, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_8", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was I doing when I stopped just now?", + "answer": "You are tying your shoelaces.", + "question_chinese": "我刚才停下来是在干什么？", + "answer_chinese": "你在系鞋带。", + "evidence": [ + { + "description": "You are tying your shoelaces.", + "timestamps": { + "start": "00:26:44", + "end": "00:26:47" + } + } + ], + "question_timestamps": "00:28:22", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are tying your belt.", + "B": "You are adjusting your cap.", + "C": "You are fastening your backpack straps after shifting the load on your shoulders.", + "D": "You are tying your shoelaces.", + "E": "You are tying a strap on your pack to secure it before walking again." + }, + "answer_label": "D", + "answer_text": "You are tying your shoelaces.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_8", + "correct_answer": "You are tying your shoelaces.", + "original_options": { + "A": "You are tying your belt.", + "B": "You are adjusting your cap.", + "C": "You are fastening your backpack straps after shifting the load on your shoulders.", + "D": "You are tying your shoelaces.", + "E": "You are tying a strap on your pack to secure it before walking again." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 1, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_5", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many red signs did I pass by just now?", + "answer": "Two signs.", + "question_chinese": "我刚才路过了几块红色的牌子？", + "answer_chinese": "两块牌子", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:31" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One sign.", + "B": "Three signs.", + "C": "You passed by four red signs during that stretch of the walk just now.", + "D": "You went past five red signs in total as you kept moving forward along the road.", + "E": "Two signs." + }, + "answer_label": "E", + "answer_text": "Two signs.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_5", + "correct_answer": "Two signs.", + "original_options": { + "A": "One sign.", + "B": "Three signs.", + "C": "You passed by four red signs during that stretch of the walk just now.", + "D": "You went past five red signs in total as you kept moving forward along the road.", + "E": "Two signs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_6", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many red signs did I pass by just now?", + "answer": "Three brands.", + "question_chinese": "我刚才路过了几块红色的牌子？", + "answer_chinese": "三块牌子", + "evidence": [ + { + "description": "The first one.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "The second one.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:31" + } + }, + { + "description": "The third one.", + "timestamps": { + "start": "00:07:22", + "end": "00:07:24" + } + } + ], + "question_timestamps": "00:08:30", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three brands.", + "B": "Two brands.", + "C": "Four brands.", + "D": "You passed by just one brand on the red signs along the route you took.", + "E": "You passed by five different brands on the red signs along the way just now." + }, + "answer_label": "A", + "answer_text": "Three brands.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_6", + "correct_answer": "Three brands.", + "original_options": { + "A": "Three brands.", + "B": "Two brands.", + "C": "Four brands.", + "D": "You passed by just one brand on the red signs along the route you took.", + "E": "You passed by five different brands on the red signs along the way just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kind of flower was the white one in front of me just now?", + "answer": "It's frangipani.", + "question_chinese": "我刚才面前白色的花是什么花？", + "answer_chinese": "是鸡蛋花", + "evidence": [ + { + "description": "It's frangipani", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's jasmine.", + "B": "It's magnolia.", + "C": "It's frangipani.", + "D": "It's a white tulip that you saw right in front of you just now.", + "E": "It's a white carnation, not the flower you first thought it was." + }, + "answer_label": "C", + "answer_text": "It's frangipani.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_3", + "correct_answer": "It's frangipani.", + "original_options": { + "A": "It's jasmine.", + "B": "It's magnolia.", + "C": "It's frangipani.", + "D": "It's a white tulip that you saw right in front of you just now.", + "E": "It's a white carnation, not the flower you first thought it was." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 12 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kind of flower is behind me now?", + "answer": "It's a red flower.", + "question_chinese": "我刚才面前白色的花是什么花？", + "answer_chinese": "是红色的花", + "evidence": [ + { + "description": "It's a red flower", + "timestamps": { + "start": "00:00:33", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:01:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's a red flower.", + "B": "It's a pink flower.", + "C": "It's a purple flower.", + "D": "Behind you now is an orange flower instead of a red one.", + "E": "The flower behind you right now is green rather than red in color." + }, + "answer_label": "A", + "answer_text": "It's a red flower.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_4", + "correct_answer": "It's a red flower.", + "original_options": { + "A": "It's a red flower.", + "B": "It's a pink flower.", + "C": "It's a purple flower.", + "D": "Behind you now is an orange flower instead of a red one.", + "E": "The flower behind you right now is green rather than red in color." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_2_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kind of flower was the one that was in front of me just now?", + "answer": "What was in front of you just now was plumeria, and the plumeria was to your right.", + "question_chinese": "我刚才面前的花是什么花？", + "answer_chinese": "你刚才面前的是鸡蛋花，鸡蛋花在你的右边。", + "evidence": [ + { + "description": "It's frangipani", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was behind you, plumeria.", + "B": "What was in front of you just now was plumeria, and the plumeria was to your right.", + "C": "It was plumeria on your left.", + "D": "What was in front of you just now was plumeria, and it was to your left.", + "E": "What was in front of you just now was plumeria, but the plumeria was behind you." + }, + "answer_label": "B", + "answer_text": "What was in front of you just now was plumeria, and the plumeria was to your right.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_2_T1", + "correct_answer": "What was in front of you just now was plumeria, and the plumeria was to your right.", + "original_options": { + "A": "It was behind you, plumeria.", + "B": "What was in front of you just now was plumeria, and the plumeria was to your right.", + "C": "It was plumeria on your left.", + "D": "What was in front of you just now was plumeria, and it was to your left.", + "E": "What was in front of you just now was plumeria, but the plumeria was behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_2_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What kind of flower was the one that was in front of me just now?", + "answer": "What was in front of you just now was a pomegranate blossom, and the pomegranate blossom was in front of you.", + "question_chinese": "我刚才面前的花是什么花？", + "answer_chinese": "你刚才面前的是石榴花，石榴花在你的正前方。", + "evidence": [ + { + "description": "You were looking at pomegranate flowers just now.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:02:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was behind you just now.", + "B": "It was to your left just now.", + "C": "What was in front of you just now was a pomegranate blossom, but it was to your right.", + "D": "What was in front of you just now was a pomegranate blossom, and the pomegranate blossom was in front of you.", + "E": "What was in front of you just now was a pomegranate blossom, and it was behind you." + }, + "answer_label": "D", + "answer_text": "What was in front of you just now was a pomegranate blossom, and the pomegranate blossom was in front of you.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_2_T2", + "correct_answer": "What was in front of you just now was a pomegranate blossom, and the pomegranate blossom was in front of you.", + "original_options": { + "A": "It was behind you just now.", + "B": "It was to your left just now.", + "C": "What was in front of you just now was a pomegranate blossom, but it was to your right.", + "D": "What was in front of you just now was a pomegranate blossom, and the pomegranate blossom was in front of you.", + "E": "What was in front of you just now was a pomegranate blossom, and it was behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 17 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_9", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What vehicles are parked around my home?", + "answer": "There is one type of parking, a small car.", + "question_chinese": "我的家周围停放有哪些载具？", + "answer_chinese": "停放有一种，一辆小轿车。", + "evidence": [ + { + "description": "There is one type of parking, a small car.", + "timestamps": { + "start": "00:34:43", + "end": "00:34:44" + } + } + ], + "question_timestamps": "00:35:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two types of vehicles.", + "B": "There is one type of parking, a small car.", + "C": "One type, a truck.", + "D": "There are two types of parking, including a small car and an SUV.", + "E": "There are three types of parking, such as a small car, an SUV, and a truck." + }, + "answer_label": "B", + "answer_text": "There is one type of parking, a small car.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_9", + "correct_answer": "There is one type of parking, a small car.", + "original_options": { + "A": "Two types of vehicles.", + "B": "There is one type of parking, a small car.", + "C": "One type, a truck.", + "D": "There are two types of parking, including a small car and an SUV.", + "E": "There are three types of parking, such as a small car, an SUV, and a truck." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 1, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 16 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_10", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What vehicles are parked around my home?", + "answer": "Two types.", + "question_chinese": "我的家周围停放有哪些载具？", + "answer_chinese": "停放有两种，一辆小轿车和一辆电瓶车。", + "evidence": [ + { + "description": "There is one type of parking, a small car.", + "timestamps": { + "start": "00:34:43", + "end": "00:34:44" + } + }, + { + "description": "There are two types of parking: one for a car and the other for an electric bike.", + "timestamps": { + "start": "00:35:38", + "end": "00:35:39" + } + } + ], + "question_timestamps": "00:35:41", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One type.", + "B": "Four types.", + "C": "Two types.", + "D": "There are five different types of vehicles parked around your home right now.", + "E": "Seven types of vehicles are parked around your home, not just two." + }, + "answer_label": "C", + "answer_text": "Two types.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_10", + "correct_answer": "Two types.", + "original_options": { + "A": "One type.", + "B": "Four types.", + "C": "Two types.", + "D": "There are five different types of vehicles parked around your home right now.", + "E": "Seven types of vehicles are parked around your home, not just two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_3_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the first item to my left right now?", + "answer": "The first item on your left was a fire hydrant, and the fire hydrant was to your left.", + "question_chinese": "现在我左边的第一个物品是什么？", + "answer_chinese": "你左边的第一个物品是消防栓，消防栓在你的左边。", + "evidence": [ + { + "description": "The first item on your left ght is a fire hydrant.", + "timestamps": { + "start": "00:31:40", + "end": "00:31:42" + } + } + ], + "question_timestamps": "00:31:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "It was to your right.", + "C": "The first item on your left was a fire hydrant, but it was actually behind you.", + "D": "The first item on your left was a fire hydrant, and the fire hydrant was to your left.", + "E": "The first item on your left was a fire hydrant, and the hydrant was in front of you." + }, + "answer_label": "D", + "answer_text": "The first item on your left was a fire hydrant, and the fire hydrant was to your left.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_3_T1", + "correct_answer": "The first item on your left was a fire hydrant, and the fire hydrant was to your left.", + "original_options": { + "A": "It was in front of you.", + "B": "It was to your right.", + "C": "The first item on your left was a fire hydrant, but it was actually behind you.", + "D": "The first item on your left was a fire hydrant, and the fire hydrant was to your left.", + "E": "The first item on your left was a fire hydrant, and the hydrant was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 18 + ] + }, + { + "video_uid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002", + "video_length": "00:35:41", + "qid": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_3_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the first item to my left right now?", + "answer": "Right now, the first item on your left is a black trash can, and the black trash can is to your left.", + "question_chinese": "现在我左边的第一个物品是什么？", + "answer_chinese": "现在你左边的第一个物品是一个黑色垃圾桶，黑色垃圾桶在你的左边。", + "evidence": [ + { + "description": "Now the first item on your left is a black trash can.", + "timestamps": { + "start": "00:31:44", + "end": "00:31:45" + } + } + ], + "question_timestamps": "00:31:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It's in front of you.", + "B": "It's to your right.", + "C": "Right now, the first item on your left is a black trash can, but it is behind you.", + "D": "Right now, the first item on your left is a black trash can, and the black trash can is to your left.", + "E": "Right now, the first item on your left is a black trash can, and the can is in front of you." + }, + "answer_label": "D", + "answer_text": "Right now, the first item on your left is a black trash can, and the black trash can is to your left.", + "q_id": "adf11624-aca0-4bad-a35f-51b9f330ee74-002_3_T2", + "correct_answer": "Right now, the first item on your left is a black trash can, and the black trash can is to your left.", + "original_options": { + "A": "It's in front of you.", + "B": "It's to your right.", + "C": "Right now, the first item on your left is a black trash can, but it is behind you.", + "D": "Right now, the first item on your left is a black trash can, and the black trash can is to your left.", + "E": "Right now, the first item on your left is a black trash can, and the can is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 21 + ] + } +] diff --git a/QAs/af4d1274-015f-4876-879d-ee32914c621f-001-mcq.json b/QAs/af4d1274-015f-4876-879d-ee32914c621f-001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b1bf6f124ffd9ede66a06e3e1bc95776ecfd9f5f --- /dev/null +++ b/QAs/af4d1274-015f-4876-879d-ee32914c621f-001-mcq.json @@ -0,0 +1,1102 @@ +[ + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow poles were planted in the ground on the intersection I just passed?", + "answer": "Four were inserted.", + "question_chinese": "刚才我路过的路口地里插了几根黄色的杆子？", + "answer_chinese": "插了四根。", + "evidence": [ + { + "description": "Four were inserted.", + "timestamps": { + "start": "00:07:05", + "end": "00:07:09" + } + } + ], + "question_timestamps": "00:07:22", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four were inserted.", + "B": "One was inserted.", + "C": "Two were inserted.", + "D": "Three yellow poles were inserted into the ground at the intersection you passed.", + "E": "Five yellow poles were inserted into the ground on the intersection you just passed." + }, + "answer_label": "A", + "answer_text": "Four were inserted.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_7", + "correct_answer": "Four were inserted.", + "original_options": { + "A": "Four were inserted.", + "B": "One was inserted.", + "C": "Two were inserted.", + "D": "Three yellow poles were inserted into the ground at the intersection you passed.", + "E": "Five yellow poles were inserted into the ground on the intersection you just passed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 14 + ] + }, + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow poles were planted in the ground on the intersection I just passed?", + "answer": "Eight were inserted.", + "question_chinese": "刚才我路过的路口地里插了几根黄色的杆子？", + "answer_chinese": "插了八根。", + "evidence": [ + { + "description": "Four were inserted.", + "timestamps": { + "start": "00:07:05", + "end": "00:07:09" + } + }, + { + "description": "Four are inserted on the right side", + "timestamps": { + "start": "00:07:15", + "end": "00:07:16" + } + } + ], + "question_timestamps": "00:07:51", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Seven were inserted.", + "B": "Nine were inserted.", + "C": "Six yellow poles were inserted into the ground at the intersection you just passed.", + "D": "Ten yellow poles were planted in the ground at that intersection as you went by.", + "E": "Eight were inserted." + }, + "answer_label": "E", + "answer_text": "Eight were inserted.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_8", + "correct_answer": "Eight were inserted.", + "original_options": { + "A": "Seven were inserted.", + "B": "Nine were inserted.", + "C": "Six yellow poles were inserted into the ground at the intersection you just passed.", + "D": "Ten yellow poles were planted in the ground at that intersection as you went by.", + "E": "Eight were inserted." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 1, + "D": 1, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 15 + ] + }, + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_5", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns have I made so far?", + "answer": "Two.", + "question_chinese": "到现在为止我一共拐了几个弯？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "First", + "timestamps": { + "start": "00:01:10", + "end": "00:01:12" + } + }, + { + "description": "second", + "timestamps": { + "start": "00:01:22", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One turn.", + "B": "Three turns.", + "C": "You have made four turns so far along your route up to this point.", + "D": "Two.", + "E": "You have already made five turns up to this point in your route so far." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_5", + "correct_answer": "Two.", + "original_options": { + "A": "One turn.", + "B": "Three turns.", + "C": "You have made four turns so far along your route up to this point.", + "D": "Two.", + "E": "You have already made five turns up to this point in your route so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_6", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns have I made so far?", + "answer": "Four.", + "question_chinese": "到现在为止我一共拐了几个弯？", + "answer_chinese": "四个", + "evidence": [ + { + "description": "First", + "timestamps": { + "start": "00:01:10", + "end": "00:01:12" + } + }, + { + "description": "second", + "timestamps": { + "start": "00:01:22", + "end": "00:01:24" + } + }, + { + "description": "third", + "timestamps": { + "start": "00:01:45", + "end": "00:01:47" + } + }, + { + "description": "forth", + "timestamps": { + "start": "00:02:06", + "end": "00:02:08" + } + } + ], + "question_timestamps": "00:02:15", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three turns.", + "B": "Four.", + "C": "Five turns.", + "D": "You have made two turns so far while moving through the space.", + "E": "You have only made one turn up to this point in the route." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_6", + "correct_answer": "Four.", + "original_options": { + "A": "Three turns.", + "B": "Four.", + "C": "Five turns.", + "D": "You have made two turns so far while moving through the space.", + "E": "You have only made one turn up to this point in the route." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 2, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_10", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I'm washing dishes in front of the sink now, can I reach that bottle of blue detergent?", + "answer": "You cann't make it.", + "question_chinese": "如果我现在正在洗碗池前洗碗，那瓶蓝色洗涤剂我能够到吗？", + "answer_chinese": "你不能够到。", + "evidence": [ + { + "description": "Blue detergent", + "timestamps": { + "start": "00:21:22", + "end": "00:21:23" + } + } + ], + "question_timestamps": "00:21:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Yes, you can reach it.", + "B": "Only if you step closer.", + "C": "You can reach the blue detergent without moving, since it's close enough to the sink area.", + "D": "You cann't make it.", + "E": "You can stretch your arm out and take the blue detergent bottle from where it sits beside you." + }, + "answer_label": "D", + "answer_text": "You cann't make it.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_10", + "correct_answer": "You cann't make it.", + "original_options": { + "A": "Yes, you can reach it.", + "B": "Only if you step closer.", + "C": "You can reach the blue detergent without moving, since it's close enough to the sink area.", + "D": "You cann't make it.", + "E": "You can stretch your arm out and take the blue detergent bottle from where it sits beside you." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_9", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I'm washing dishes in front of the sink now, can I reach that bottle of blue detergent?", + "answer": "You can make it.", + "question_chinese": "如果我现在正在洗碗池前洗碗，那瓶蓝色洗涤剂我能够到吗？", + "answer_chinese": "你能够到。", + "evidence": [ + { + "description": "Blue detergent", + "timestamps": { + "start": "00:21:23", + "end": "00:21:25" + } + } + ], + "question_timestamps": "00:23:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "No, you can't reach it.", + "B": "You can't make it.", + "C": "You can make it.", + "D": "From where you're washing dishes at the sink, the blue detergent bottle is out of reach.", + "E": "While standing in front of the sink, you wouldn't be able to reach the bottle of blue detergent." + }, + "answer_label": "C", + "answer_text": "You can make it.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_9", + "correct_answer": "You can make it.", + "original_options": { + "A": "No, you can't reach it.", + "B": "You can't make it.", + "C": "You can make it.", + "D": "From where you're washing dishes at the sink, the blue detergent bottle is out of reach.", + "E": "While standing in front of the sink, you wouldn't be able to reach the bottle of blue detergent." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 1, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 18 + ] + }, + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_3", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I'm working at my computer desk, how should I get to the bathroom?", + "answer": "Walk to the right and open the door to reach the bathroom.", + "question_chinese": "当我在电脑桌前工作时，想要去浴室应该怎么走？", + "answer_chinese": "向右走，打开门就是浴室。", + "evidence": [ + { + "description": "Computer desk", + "timestamps": { + "start": "00:00:19", + "end": "00:00:21" + } + }, + { + "description": "Bathroom", + "timestamps": { + "start": "00:00:27", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk left and open the door.", + "B": "Walk straight and open the door.", + "C": "Turn around, walk to the left, and open the door to get to the bathroom.", + "D": "Walk to the right and open the door to reach the bathroom.", + "E": "Walk to the right, then open the closet door to reach the bathroom." + }, + "answer_label": "D", + "answer_text": "Walk to the right and open the door to reach the bathroom.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_3", + "correct_answer": "Walk to the right and open the door to reach the bathroom.", + "original_options": { + "A": "Walk left and open the door.", + "B": "Walk straight and open the door.", + "C": "Turn around, walk to the left, and open the door to get to the bathroom.", + "D": "Walk to the right and open the door to reach the bathroom.", + "E": "Walk to the right, then open the closet door to reach the bathroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 13 + ] + }, + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_4", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "When I'm working at my computer desk, how should I get to the bathroom?", + "answer": "Turn around and walk to the right.", + "question_chinese": "当我在电脑桌前工作时，想要去浴室应该怎么走？", + "answer_chinese": "转身然后向右走", + "evidence": [ + { + "description": "Computer desk", + "timestamps": { + "start": "00:00:19", + "end": "00:00:21" + } + }, + { + "description": "Bathroom", + "timestamps": { + "start": "00:00:27", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn around and walk to the right.", + "B": "Turn around and walk left.", + "C": "Turn around and go straight.", + "D": "Turn around and walk to the back of the room instead of heading to the right.", + "E": "Turn around and walk into the closet area rather than moving to the right." + }, + "answer_label": "A", + "answer_text": "Turn around and walk to the right.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_4", + "correct_answer": "Turn around and walk to the right.", + "original_options": { + "A": "Turn around and walk to the right.", + "B": "Turn around and walk left.", + "C": "Turn around and go straight.", + "D": "Turn around and walk to the back of the room instead of heading to the right.", + "E": "Turn around and walk into the closet area rather than moving to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 14 + ] + }, + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_12", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to use the microwave oven. Where is it?", + "answer": "The microwave oven is opposite the sink.", + "question_chinese": "我想使用微波炉，微波炉在哪里？", + "answer_chinese": "微波炉在洗碗池的对面", + "evidence": [ + { + "description": "The top of the refrigerator", + "timestamps": { + "start": "00:19:01", + "end": "00:19:02" + } + }, + { + "description": "The microwave oven is opposite the sink", + "timestamps": { + "start": "00:19:04", + "end": "00:19:05" + } + } + ], + "question_timestamps": "00:19:48", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The microwave oven is opposite the sink.", + "B": "The microwave oven is left of the sink.", + "C": "The microwave oven is behind the sink.", + "D": "The microwave oven is in front of the sink, close to where you wash dishes.", + "E": "The microwave oven is located to the right of the sink, not opposite it." + }, + "answer_label": "A", + "answer_text": "The microwave oven is opposite the sink.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_12", + "correct_answer": "The microwave oven is opposite the sink.", + "original_options": { + "A": "The microwave oven is opposite the sink.", + "B": "The microwave oven is left of the sink.", + "C": "The microwave oven is behind the sink.", + "D": "The microwave oven is in front of the sink, close to where you wash dishes.", + "E": "The microwave oven is located to the right of the sink, not opposite it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 3, + "B": 2, + "C": 3, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 14 + ] + }, + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_11", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I want to use the microwave oven. Where is it?", + "answer": "The microwave oven is located right behind the sink and above the refrigerator.", + "question_chinese": "我想使用微波炉，微波炉在哪里？", + "answer_chinese": "微波炉洗碗池右后方冰箱的上面。", + "evidence": [ + { + "description": "The top of the refrigerator", + "timestamps": { + "start": "00:19:01", + "end": "00:19:02" + } + }, + { + "description": "The right rear of the sink", + "timestamps": { + "start": "00:19:04", + "end": "00:19:05" + } + } + ], + "question_timestamps": "00:20:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It’s above the sink.", + "B": "It’s beside the refrigerator.", + "C": "The microwave oven is located right behind the sink and above the refrigerator.", + "D": "The microwave oven is located to the left of the sink, positioned above the refrigerator door.", + "E": "You can find the microwave oven in front of the sink area, mounted above the countertop space." + }, + "answer_label": "C", + "answer_text": "The microwave oven is located right behind the sink and above the refrigerator.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_11", + "correct_answer": "The microwave oven is located right behind the sink and above the refrigerator.", + "original_options": { + "A": "It’s above the sink.", + "B": "It’s beside the refrigerator.", + "C": "The microwave oven is located right behind the sink and above the refrigerator.", + "D": "The microwave oven is located to the left of the sink, positioned above the refrigerator door.", + "E": "You can find the microwave oven in front of the sink area, mounted above the countertop space." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 3, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many paintings are hung by my bedside?", + "answer": "Five paintings were hung.", + "question_chinese": "我的床头挂了几幅画？", + "answer_chinese": "挂了五幅画。", + "evidence": [ + { + "description": "Five paintings were hung.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:20" + } + }, + { + "description": "Five paintings were hung.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four paintings.", + "B": "Six paintings.", + "C": "Five paintings were hung.", + "D": "Only three paintings were hung by your bedside, spaced out along the wall.", + "E": "Seven paintings were hung by your bedside, forming a longer row above the bed." + }, + "answer_label": "C", + "answer_text": "Five paintings were hung.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_1", + "correct_answer": "Five paintings were hung.", + "original_options": { + "A": "Four paintings.", + "B": "Six paintings.", + "C": "Five paintings were hung.", + "D": "Only three paintings were hung by your bedside, spaced out along the wall.", + "E": "Seven paintings were hung by your bedside, forming a longer row above the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 1, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "af4d1274-015f-4876-879d-ee32914c621f-001", + "video_length": "00:30:00", + "qid": "af4d1274-015f-4876-879d-ee32914c621f-001_2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many paintings are hung by my bedside?", + "answer": "Six paintings were hung.", + "question_chinese": "我的床头挂了几幅画？", + "answer_chinese": "挂了六幅画。", + "evidence": [ + { + "description": "A painting is hung on the left side of the bed", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + }, + { + "description": "Five paintings were hung.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five paintings.", + "B": "Seven paintings.", + "C": "Four paintings were hung by your bedside, arranged in a smaller grouping on the wall.", + "D": "Eight paintings were hung by your bedside, filling most of the space above the bed.", + "E": "Six paintings were hung." + }, + "answer_label": "E", + "answer_text": "Six paintings were hung.", + "q_id": "af4d1274-015f-4876-879d-ee32914c621f-001_2", + "correct_answer": "Six paintings were hung.", + "original_options": { + "A": "Five paintings.", + "B": "Seven paintings.", + "C": "Four paintings were hung by your bedside, arranged in a smaller grouping on the wall.", + "D": "Eight paintings were hung by your bedside, filling most of the space above the bed.", + "E": "Six paintings were hung." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 1, + "D": 0, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + } +] diff --git a/QAs/bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001-mcq.json b/QAs/bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..cbfb2a8ffe12b89798ba798bd40f9506e1638b09 --- /dev/null +++ b/QAs/bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001-mcq.json @@ -0,0 +1,937 @@ +[ + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many turns have I made since I left home?", + "answer": "Two.", + "question_chinese": "从出门到现在我一共拐了几个弯？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "First", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "second", + "timestamps": { + "start": "00:00:45", + "end": "00:00:46" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "You’ve made four turns since leaving home, changing direction multiple times along the route.", + "E": "You’ve made five turns since leaving home, with several corners before reaching here." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_7", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "You’ve made four turns since leaving home, changing direction multiple times along the route.", + "E": "You’ve made five turns since leaving home, with several corners before reaching here." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 2, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many turns have I made since I left home?", + "answer": "Three.", + "question_chinese": "从出门到现在我一共拐了几个弯？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "First", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "second", + "timestamps": { + "start": "00:00:45", + "end": "00:00:46" + } + }, + { + "description": "third", + "timestamps": { + "start": "00:03:52", + "end": "00:03:54" + } + } + ], + "question_timestamps": "00:04:18", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You’ve made five turns since you left home, with repeated direction changes along the way.", + "E": "You’ve made six turns since leaving home, turning at nearly every major corner." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_8", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You’ve made five turns since you left home, with repeated direction changes along the way.", + "E": "You’ve made six turns since leaving home, turning at nearly every major corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 2, + "D": 2, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 14 + ] + }, + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From the time I saw the setting sun until the roadside of the street I'm walking on now, how many colors of flowers have I seen in total?", + "answer": "Two kinds.", + "question_chinese": "从我看到夕阳时到我现在走的这条街的路边我一共看到了几种颜色的花？", + "answer_chinese": "两种", + "evidence": [ + { + "description": "White flowers", + "timestamps": { + "start": "00:11:58", + "end": "00:12:00" + } + }, + { + "description": "Red flowers", + "timestamps": { + "start": "00:13:17", + "end": "00:13:19" + } + } + ], + "question_timestamps": "00:13:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One kind.", + "B": "Two kinds.", + "C": "Three kinds.", + "D": "You’ve seen zero kinds of flowers in total since the setting sun until this roadside.", + "E": "You’ve seen four kinds of flower colors in total between the setting sun and this street." + }, + "answer_label": "B", + "answer_text": "Two kinds.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_5", + "correct_answer": "Two kinds.", + "original_options": { + "A": "One kind.", + "B": "Two kinds.", + "C": "Three kinds.", + "D": "You’ve seen zero kinds of flowers in total since the setting sun until this roadside.", + "E": "You’ve seen four kinds of flower colors in total between the setting sun and this street." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From the time I saw the setting sun until the roadside of the street I'm walking on now, how many colors of flowers have I seen in total?", + "answer": "Three types.", + "question_chinese": "从我看到夕阳时到我现在走的这条街的路边我一共看到了几种颜色的花？", + "answer_chinese": "三种", + "evidence": [ + { + "description": "White flowers", + "timestamps": { + "start": "00:11:58", + "end": "00:12:00" + } + }, + { + "description": "Red flowers", + "timestamps": { + "start": "00:13:17", + "end": "00:13:19" + } + }, + { + "description": "Pink flowers", + "timestamps": { + "start": "00:13:46", + "end": "00:13:47" + } + } + ], + "question_timestamps": "00:14:11", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two types.", + "B": "Four types.", + "C": "Three types.", + "D": "You’ve seen five types of flower colors in total from the setting sun until this roadside.", + "E": "You’ve seen seven types of flower colors overall between the setting sun and where you are now." + }, + "answer_label": "C", + "answer_text": "Three types.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_6", + "correct_answer": "Three types.", + "original_options": { + "A": "Two types.", + "B": "Four types.", + "C": "Three types.", + "D": "You’ve seen five types of flower colors in total from the setting sun until this roadside.", + "E": "You’ve seen seven types of flower colors overall between the setting sun and where you are now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 2, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_1_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the trash can relative to me?", + "answer": "The trash can is to your right.", + "question_chinese": "垃圾桶在我的哪里？", + "answer_chinese": "垃圾桶在你的右边", + "evidence": [ + { + "description": "The trash can is on your right.", + "timestamps": { + "start": "00:14:57", + "end": "00:15:06" + } + } + ], + "question_timestamps": "00:15:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The trash can is to your left.", + "B": "The trash can is to your right.", + "C": "The trash can is in front of you.", + "D": "The trash can is behind you, slightly off to one side as you stand there.", + "E": "The trash can is at your front-left, angled ahead of you rather than on your right." + }, + "answer_label": "B", + "answer_text": "The trash can is to your right.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_1_T1", + "correct_answer": "The trash can is to your right.", + "original_options": { + "A": "The trash can is to your left.", + "B": "The trash can is to your right.", + "C": "The trash can is in front of you.", + "D": "The trash can is behind you, slightly off to one side as you stand there.", + "E": "The trash can is at your front-left, angled ahead of you rather than on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 16 + ] + }, + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_1_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the trash can relative to me?", + "answer": "The trash can is to your left.", + "question_chinese": "垃圾桶在我的哪里？", + "answer_chinese": "垃圾桶在你的左边", + "evidence": [ + { + "description": "The trash can is on your left.", + "timestamps": { + "start": "00:20:57", + "end": "00:21:06" + } + } + ], + "question_timestamps": "00:21:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The trash can is to your right.", + "B": "The trash can is in front of you.", + "C": "The trash can is to your left.", + "D": "The trash can is directly behind you as you stand still, not on either side.", + "E": "The trash can sits off your right-rear side, slightly behind your shoulder rather than to your left." + }, + "answer_label": "C", + "answer_text": "The trash can is to your left.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_1_T2", + "correct_answer": "The trash can is to your left.", + "original_options": { + "A": "The trash can is to your right.", + "B": "The trash can is in front of you.", + "C": "The trash can is to your left.", + "D": "The trash can is directly behind you as you stand still, not on either side.", + "E": "The trash can sits off your right-rear side, slightly behind your shoulder rather than to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 17 + ] + }, + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many passers-by did I meet on the cement road with yellow soil on both sides?", + "answer": "One.", + "question_chinese": "我在两边都是黄土的水泥路上碰到了几个路人？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "The first", + "timestamps": { + "start": "00:07:28", + "end": "00:07:30" + } + } + ], + "question_timestamps": "00:07:45", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "Four passers-by, spaced out along the road, as you walked between the yellow soil banks.", + "E": "Five passers-by in total, encountered one after another on the cement road with yellow soil on both sides." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_1", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "Four passers-by, spaced out along the road, as you walked between the yellow soil banks.", + "E": "Five passers-by in total, encountered one after another on the cement road with yellow soil on both sides." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 1, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 18 + ] + }, + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many passers-by did I meet on the cement road with yellow soil on both sides?", + "answer": "One.", + "question_chinese": "我在两边都是黄土的水泥路上碰到了几个路人？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "The first", + "timestamps": { + "start": "00:07:28", + "end": "00:07:30" + } + } + ], + "question_timestamps": "00:08:19", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Four.", + "D": "Three passers-by, met at different points while walking down the cement road bordered by yellow soil.", + "E": "Five passers-by altogether, all encountered on the cement road with yellow soil lining both sides." + }, + "answer_label": "A", + "answer_text": "One.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_2", + "correct_answer": "One.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Four.", + "D": "Three passers-by, met at different points while walking down the cement road bordered by yellow soil.", + "E": "Five passers-by altogether, all encountered on the cement road with yellow soil lining both sides." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 1, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_3", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At what time of the day do I usually walk?", + "answer": "It should be dusk.", + "question_chinese": "我在走路的时间点大概是一天中的什么时候？", + "answer_chinese": "应该是黄昏时分。", + "evidence": [ + { + "description": "There is a setting sun at the intersection.", + "timestamps": { + "start": "00:11:23", + "end": "00:11:30" + } + } + ], + "question_timestamps": "00:12:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It should be morning.", + "B": "It should be afternoon.", + "C": "It should be around midday, with brighter light suggesting the sun is still high.", + "D": "It should be dusk.", + "E": "It should be late at night, when the surroundings would look darker than at dusk." + }, + "answer_label": "D", + "answer_text": "It should be dusk.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_3", + "correct_answer": "It should be dusk.", + "original_options": { + "A": "It should be morning.", + "B": "It should be afternoon.", + "C": "It should be around midday, with brighter light suggesting the sun is still high.", + "D": "It should be dusk.", + "E": "It should be late at night, when the surroundings would look darker than at dusk." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_4", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At what time of the day do I usually walk?", + "answer": "It should be in the evening.", + "question_chinese": "我在走路的时间点大概是一天中的什么时候？", + "answer_chinese": "应该是傍晚时分。", + "evidence": [ + { + "description": "It was already getting dark", + "timestamps": { + "start": "00:24:08", + "end": "00:24:10" + } + } + ], + "question_timestamps": "00:25:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It should be late morning.", + "B": "It should be at midnight.", + "C": "It should be in the early afternoon, before the light starts fading toward evening.", + "D": "It should be in the late afternoon, when it is still bright but not yet evening.", + "E": "It should be in the evening." + }, + "answer_label": "E", + "answer_text": "It should be in the evening.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_4", + "correct_answer": "It should be in the evening.", + "original_options": { + "A": "It should be late morning.", + "B": "It should be at midnight.", + "C": "It should be in the early afternoon, before the light starts fading toward evening.", + "D": "It should be in the late afternoon, when it is still bright but not yet evening.", + "E": "It should be in the evening." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_5_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Are there any cars nearby me?", + "answer": "There is a car nearby you; it's on the left side of the garden.", + "question_chinese": "我附近有车吗？", + "answer_chinese": "你附近有车，它在花园的左侧。", + "evidence": [ + { + "description": "On the left side of the garden", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There are no cars nearby.", + "B": "There is a car nearby you; it's on the left side of the garden.", + "C": "There is a car on the right.", + "D": "There is a car nearby you; it's straight ahead in the garden rather than on your left.", + "E": "There is a car nearby you; it's at the back of the garden, not on the left side." + }, + "answer_label": "B", + "answer_text": "There is a car nearby you; it's on the left side of the garden.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_5_T2", + "correct_answer": "There is a car nearby you; it's on the left side of the garden.", + "original_options": { + "A": "There are no cars nearby.", + "B": "There is a car nearby you; it's on the left side of the garden.", + "C": "There is a car on the right.", + "D": "There is a car nearby you; it's straight ahead in the garden rather than on your left.", + "E": "There is a car nearby you; it's at the back of the garden, not on the left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 17, + 18 + ] + }, + { + "video_uid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001", + "video_length": "00:37:56", + "qid": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_5_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Are there any cars nearby me?", + "answer": "There are no cars nearby you.", + "question_chinese": "我附近有车吗？", + "answer_chinese": "你附近没有车。", + "evidence": [ + { + "description": "There are no cars on the road.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There are no cars nearby you.", + "B": "A car is nearby on the left.", + "C": "A car is nearby on the right.", + "D": "There is a car nearby you; it's straight ahead in the garden, so you are not alone.", + "E": "There is a car nearby you; it's at the back of the garden behind you, not absent." + }, + "answer_label": "A", + "answer_text": "There are no cars nearby you.", + "q_id": "bf295f25-d05d-4aa0-a6b9-4e97334e1ee9-001_5_T1", + "correct_answer": "There are no cars nearby you.", + "original_options": { + "A": "There are no cars nearby you.", + "B": "A car is nearby on the left.", + "C": "A car is nearby on the right.", + "D": "There is a car nearby you; it's straight ahead in the garden, so you are not alone.", + "E": "There is a car nearby you; it's at the back of the garden behind you, not absent." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + } +] diff --git a/QAs/c1fca588-dd91-4f40-9b60-01c55e9e03c8-001-mcq.json b/QAs/c1fca588-dd91-4f40-9b60-01c55e9e03c8-001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c6f920c8d720860280d901522510ed19e0326777 --- /dev/null +++ b/QAs/c1fca588-dd91-4f40-9b60-01c55e9e03c8-001-mcq.json @@ -0,0 +1,746 @@ +[ + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001", + "video_length": "00:20:02", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_11", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "At the fork just now, should I turn left or right?", + "answer": "You choose to turn left.", + "question_chinese": "刚才的分岔路口我选择向左拐还是右拐？", + "answer_chinese": "你选择向左拐。", + "evidence": [ + { + "description": "You choose to turn left.", + "timestamps": { + "start": "00:15:19", + "end": "00:15:20" + } + } + ], + "question_timestamps": "00:15:33", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You choose to turn right.", + "B": "You choose to turn left." + }, + "answer_label": "B", + "answer_text": "You choose to turn left.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_11", + "correct_answer": "You choose to turn left.", + "original_options": { + "A": "You choose to turn right.", + "B": "You choose to turn left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001", + "video_length": "00:20:02", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_12", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "At the fork just now, should I turn left or right?", + "answer": "You choose to turn right.", + "question_chinese": "刚才的分岔路口我选择向左拐还是右拐？", + "answer_chinese": "你选择向右拐。", + "evidence": [ + { + "description": "You choose to turn right.", + "timestamps": { + "start": "00:17:54", + "end": "00:17:55" + } + } + ], + "question_timestamps": "00:18:34", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You choose to turn right.", + "B": "You choose to turn left." + }, + "answer_label": "A", + "answer_text": "You choose to turn right.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_12", + "correct_answer": "You choose to turn right.", + "original_options": { + "A": "You choose to turn right.", + "B": "You choose to turn left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 2, + "B": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001", + "video_length": "00:20:02", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many turns did I make just now?", + "answer": "One.", + "question_chinese": "我刚才一共拐了多少个弯?", + "answer_chinese": "一个", + "evidence": [ + { + "description": "The first", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:01:58", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two turns.", + "B": "Zero turns.", + "C": "You made two turns overall during that short stretch just now, before heading straight.", + "D": "One.", + "E": "You didn’t make any turns at all in that segment just now, just went straight." + }, + "answer_label": "D", + "answer_text": "One.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_1", + "correct_answer": "One.", + "original_options": { + "A": "Two turns.", + "B": "Zero turns.", + "C": "You made two turns overall during that short stretch just now, before heading straight.", + "D": "One.", + "E": "You didn’t make any turns at all in that segment just now, just went straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 0, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001", + "video_length": "00:20:02", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many turns did I make just now?", + "answer": "Three.", + "question_chinese": "我刚才一共拐了多少个弯?", + "answer_chinese": "三个", + "evidence": [ + { + "description": "The first", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + }, + { + "description": "the second", + "timestamps": { + "start": "00:02:15", + "end": "00:02:16" + } + }, + { + "description": "the third", + "timestamps": { + "start": "00:02:33", + "end": "00:02:34" + } + } + ], + "question_timestamps": "00:02:43", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "You made two turns in that sequence just a moment ago, not three.", + "D": "You ended up making four turns before continuing straight ahead afterward, not three.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_2", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "You made two turns in that sequence just a moment ago, not three.", + "D": "You ended up making four turns before continuing straight ahead afterward, not three.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 0, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001", + "video_length": "00:20:02", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many red cars were parked in the yard I just passed by?", + "answer": "Two cars were parked.", + "question_chinese": "我刚才路过的院子里停了几辆红色的车？", + "answer_chinese": "停了两辆。", + "evidence": [ + { + "description": "Two cars were parked.", + "timestamps": { + "start": "00:14:34", + "end": "00:14:35" + } + } + ], + "question_timestamps": "00:15:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One car was parked.", + "B": "Three cars were parked.", + "C": "Four red cars were parked in the yard you just passed by moments ago.", + "D": "Two cars were parked.", + "E": "Five red cars were parked there in the yard as you went by just now." + }, + "answer_label": "D", + "answer_text": "Two cars were parked.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_9", + "correct_answer": "Two cars were parked.", + "original_options": { + "A": "One car was parked.", + "B": "Three cars were parked.", + "C": "Four red cars were parked in the yard you just passed by moments ago.", + "D": "Two cars were parked.", + "E": "Five red cars were parked there in the yard as you went by just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 2, + "C": 1, + "D": 2, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001", + "video_length": "00:20:02", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_10", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many red cars were parked in the yard I just passed by?", + "answer": "One cars were parked.", + "question_chinese": "我刚才路过的院子里停了几辆红色的车？", + "answer_chinese": "停了一辆。", + "evidence": [ + { + "description": "One cars were parked.", + "timestamps": { + "start": "00:16:40", + "end": "00:16:41" + } + } + ], + "question_timestamps": "00:16:52", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two cars were parked.", + "B": "Three cars were parked.", + "C": "Four red cars were parked in the yard you just passed by moments ago.", + "D": "Five red cars were parked there in the yard as you went by just now.", + "E": "One cars were parked." + }, + "answer_label": "E", + "answer_text": "One cars were parked.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_10", + "correct_answer": "One cars were parked.", + "original_options": { + "A": "Two cars were parked.", + "B": "Three cars were parked.", + "C": "Four red cars were parked in the yard you just passed by moments ago.", + "D": "Five red cars were parked there in the yard as you went by just now.", + "E": "One cars were parked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 2, + "C": 1, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001", + "video_length": "00:20:02", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many red roadblocks are there on the road I'm taking now?", + "answer": "Three.", + "question_chinese": "我现在走的这条路有多少个红色路障？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "Three roadblocks can be seen now", + "timestamps": { + "start": "00:05:10", + "end": "00:05:11" + } + } + ], + "question_timestamps": "00:05:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "There are two red roadblocks on the road you’re taking now.", + "E": "You’ll pass four red roadblocks along the road you’re on now." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_3", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "There are two red roadblocks on the road you’re taking now.", + "E": "You’ll pass four red roadblocks along the road you’re on now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001", + "video_length": "00:20:02", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many red roadblocks are there on the road I'm taking now?", + "answer": "Six.", + "question_chinese": "我现在走的这条路有多少个红色路障？", + "answer_chinese": "六个", + "evidence": [ + { + "description": "Six", + "timestamps": { + "start": "00:05:15", + "end": "00:05:16" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Six.", + "C": "Eight.", + "D": "There are four red roadblocks on the road you’re taking now.", + "E": "You’ll encounter eight red roadblocks along the road you’re on now." + }, + "answer_label": "B", + "answer_text": "Six.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_4", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Six.", + "C": "Eight.", + "D": "There are four red roadblocks on the road you’re taking now.", + "E": "You’ll encounter eight red roadblocks along the road you’re on now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001", + "video_length": "00:20:02", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_5", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is there a blue car behind me now?", + "answer": "Yes.", + "question_chinese": "我现在身后是否有一辆蓝色的车？", + "answer_chinese": "是的", + "evidence": [ + { + "description": "Yes.", + "timestamps": { + "start": "00:06:41", + "end": "00:06:42" + } + } + ], + "question_timestamps": "00:06:43", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_5", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 0, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001", + "video_length": "00:20:02", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_6", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is there a blue car behind me now?", + "answer": "No.", + "question_chinese": "我现在身后是否有一辆蓝色的车？", + "answer_chinese": "没有", + "evidence": [ + { + "description": "No.", + "timestamps": { + "start": "00:06:41", + "end": "00:06:42" + } + } + ], + "question_timestamps": "00:06:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-001_6", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/c1fca588-dd91-4f40-9b60-01c55e9e03c8-002-mcq.json b/QAs/c1fca588-dd91-4f40-9b60-01c55e9e03c8-002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2d2727bdec4fa88a83a811ae37d271c645cf2404 --- /dev/null +++ b/QAs/c1fca588-dd91-4f40-9b60-01c55e9e03c8-002-mcq.json @@ -0,0 +1,1017 @@ +[ + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_9", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many turns have I made since I met the red car?", + "answer": "Two.", + "question_chinese": "从我遇到了红色的车的时候到现在我一共拐了几个弯？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "The first", + "timestamps": { + "start": "00:19:14", + "end": "00:19:16" + } + }, + { + "description": "the second", + "timestamps": { + "start": "00:21:06", + "end": "00:21:08" + } + } + ], + "question_timestamps": "00:21:22", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One turn.", + "B": "Zero turns.", + "C": "Three turns in total since you first met the red car yesterday.", + "D": "Two.", + "E": "You made five turns after meeting the red car, counting every change in direction." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_9", + "correct_answer": "Two.", + "original_options": { + "A": "One turn.", + "B": "Zero turns.", + "C": "Three turns in total since you first met the red car yesterday.", + "D": "Two.", + "E": "You made five turns after meeting the red car, counting every change in direction." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 2, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_10", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many turns have I made since I met the red car?", + "answer": "Three.", + "question_chinese": "从我遇到了红色的车的时候到现在我一共拐了几个弯？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "The first", + "timestamps": { + "start": "00:19:14", + "end": "00:19:16" + } + }, + { + "description": "the second", + "timestamps": { + "start": "00:21:06", + "end": "00:21:08" + } + }, + { + "description": "the third", + "timestamps": { + "start": "00:21:54", + "end": "00:21:55" + } + } + ], + "question_timestamps": "00:22:24", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Two turns.", + "C": "Four turns.", + "D": "You made five turns since meeting the red car, including each time you changed direction.", + "E": "You made six turns in total after you met the red car and started navigating onward." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_10", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two turns.", + "C": "Four turns.", + "D": "You made five turns since meeting the red car, including each time you changed direction.", + "E": "You made six turns in total after you met the red car and started navigating onward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Since I met that passer-by in red, how many moving cars have I encountered in total?", + "answer": "One vehicles.", + "question_chinese": "从我遇到那位红色衣服的路人开始，我一共遇到了几辆行驶的轿车？", + "answer_chinese": "一辆", + "evidence": [ + { + "description": "The first", + "timestamps": { + "start": "00:17:11", + "end": "00:17:12" + } + } + ], + "question_timestamps": "00:17:22", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two vehicles.", + "B": "Zero vehicles.", + "C": "One vehicles.", + "D": "You encountered four moving cars in total after meeting the passer-by in red.", + "E": "You ran into five moving vehicles overall since you met that passer-by wearing red." + }, + "answer_label": "C", + "answer_text": "One vehicles.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_7", + "correct_answer": "One vehicles.", + "original_options": { + "A": "Two vehicles.", + "B": "Zero vehicles.", + "C": "One vehicles.", + "D": "You encountered four moving cars in total after meeting the passer-by in red.", + "E": "You ran into five moving vehicles overall since you met that passer-by wearing red." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 2, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_8", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Since I met that passer-by in red, how many moving cars have I encountered in total?", + "answer": "Three vehicles.", + "question_chinese": "从我遇到那位红色衣服的路人开始，我一共遇到了几辆行驶的轿车？", + "answer_chinese": "三辆", + "evidence": [ + { + "description": "The first", + "timestamps": { + "start": "00:17:11", + "end": "00:17:12" + } + }, + { + "description": "the second", + "timestamps": { + "start": "00:19:19", + "end": "00:19:21" + } + }, + { + "description": "the third", + "timestamps": { + "start": "00:19:17", + "end": "00:19:18" + } + } + ], + "question_timestamps": "00:21:03", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two vehicles.", + "B": "Three vehicles.", + "C": "Four vehicles.", + "D": "You encountered five moving cars in total after meeting the passer-by in red.", + "E": "You met only one moving vehicle overall since you first passed that person in red." + }, + "answer_label": "B", + "answer_text": "Three vehicles.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_8", + "correct_answer": "Three vehicles.", + "original_options": { + "A": "Two vehicles.", + "B": "Three vehicles.", + "C": "Four vehicles.", + "D": "You encountered five moving cars in total after meeting the passer-by in red.", + "E": "You met only one moving vehicle overall since you first passed that person in red." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my course of action at the fork just now?", + "answer": "You turned left.", + "question_chinese": "刚才的分岔路口我的行动轨迹是什么？", + "answer_chinese": "你向左拐了。", + "evidence": [ + { + "description": "You turned left.", + "timestamps": { + "start": "00:03:44", + "end": "00:03:46" + } + } + ], + "question_timestamps": "00:04:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You turned left.", + "B": "You turned right.", + "C": "You went straight.", + "D": "You paused briefly at the fork, then chose the right-hand path away from the camera.", + "E": "You turned around at the fork and headed back the way you came." + }, + "answer_label": "A", + "answer_text": "You turned left.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_1", + "correct_answer": "You turned left.", + "original_options": { + "A": "You turned left.", + "B": "You turned right.", + "C": "You went straight.", + "D": "You paused briefly at the fork, then chose the right-hand path away from the camera.", + "E": "You turned around at the fork and headed back the way you came." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 0, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 13 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my course of action at the fork just now?", + "answer": "You turned right.", + "question_chinese": "刚才的分岔路口我的行动轨迹是什么？", + "answer_chinese": "你向右拐了。", + "evidence": [ + { + "description": "You turned right.", + "timestamps": { + "start": "00:06:48", + "end": "00:06:49" + } + } + ], + "question_timestamps": "00:07:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You turned left.", + "B": "You went straight.", + "C": "You continued forward through the fork without turning, sticking to the main path.", + "D": "You turned right.", + "E": "You hesitated at the fork for a moment and then decided to take the left path." + }, + "answer_label": "D", + "answer_text": "You turned right.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_2", + "correct_answer": "You turned right.", + "original_options": { + "A": "You turned left.", + "B": "You went straight.", + "C": "You continued forward through the fork without turning, sticking to the main path.", + "D": "You turned right.", + "E": "You hesitated at the fork for a moment and then decided to take the left path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 0, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 16 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_6_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I parked just now, what did I do?", + "answer": "After you parked, you looked at the phone in your hand and also drank some water.", + "question_chinese": "我刚才停车后干了什么事情？", + "answer_chinese": "你停车后看了你手上的手机，还喝了水。", + "evidence": [ + { + "description": "You looked at your phone.", + "timestamps": { + "start": "00:24:19", + "end": "00:24:20" + } + }, + { + "description": "You drank water", + "timestamps": { + "start": "00:26:04", + "end": "00:26:07" + } + } + ], + "question_timestamps": "00:26:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You only checked your phone.", + "B": "You only drank some water.", + "C": "After you parked, you looked at the phone in your hand and also drank some water.", + "D": "After you parked, you checked your phone and then walked away without drinking any water.", + "E": "After parking, you drank water and then left without looking at the phone in your hand." + }, + "answer_label": "C", + "answer_text": "After you parked, you looked at the phone in your hand and also drank some water.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_6_T2", + "correct_answer": "After you parked, you looked at the phone in your hand and also drank some water.", + "original_options": { + "A": "You only checked your phone.", + "B": "You only drank some water.", + "C": "After you parked, you looked at the phone in your hand and also drank some water.", + "D": "After you parked, you checked your phone and then walked away without drinking any water.", + "E": "After parking, you drank water and then left without looking at the phone in your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_6_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I parked just now, what did I do?", + "answer": "You checked your phone, drank some water, and also used your hand to dig at the rocks on the ground in front of you.", + "question_chinese": "我刚才停车后干了什么事情？", + "answer_chinese": "你看了手机，喝了水，还用手挖了你面前地上的石块。", + "evidence": [ + { + "description": "You looked at your phone.", + "timestamps": { + "start": "00:24:19", + "end": "00:24:20" + } + }, + { + "description": "You drank water", + "timestamps": { + "start": "00:26:04", + "end": "00:26:07" + } + }, + { + "description": "You dug the stone with your hand.", + "timestamps": { + "start": "00:26:07", + "end": "00:26:09" + } + } + ], + "question_timestamps": "00:27:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You checked your phone and drank water.", + "B": "You checked your phone, drank some water, and also used your hand to dig at the rocks on the ground in front of you.", + "C": "You checked your phone and dug at rocks.", + "D": "You checked your phone, drank some water, and then kicked the rocks in front of you with your shoe.", + "E": "You checked your phone, drank some water, then dug at the rocks on the ground to your left." + }, + "answer_label": "B", + "answer_text": "You checked your phone, drank some water, and also used your hand to dig at the rocks on the ground in front of you.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_6_T1", + "correct_answer": "You checked your phone, drank some water, and also used your hand to dig at the rocks on the ground in front of you.", + "original_options": { + "A": "You checked your phone and drank water.", + "B": "You checked your phone, drank some water, and also used your hand to dig at the rocks on the ground in front of you.", + "C": "You checked your phone and dug at rocks.", + "D": "You checked your phone, drank some water, and then kicked the rocks in front of you with your shoe.", + "E": "You checked your phone, drank some water, then dug at the rocks on the ground to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 18 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_5", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many passers-by did I meet on this road just now?", + "answer": "One.", + "question_chinese": "我刚才在这条路上碰到了几个路人？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "One", + "timestamps": { + "start": "00:12:19", + "end": "00:12:20" + } + } + ], + "question_timestamps": "00:13:23", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "Two.", + "C": "I met three passers-by while walking down this road just now, heading straight ahead.", + "D": "I met four passers-by on this road just now, passing them as I continued forward.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_5", + "correct_answer": "One.", + "original_options": { + "A": "Zero.", + "B": "Two.", + "C": "I met three passers-by while walking down this road just now, heading straight ahead.", + "D": "I met four passers-by on this road just now, passing them as I continued forward.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_6", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many passers-by did I meet on this road just now?", + "answer": "Zero.", + "question_chinese": "我刚才在这条路上碰到了几个路人？", + "answer_chinese": "零个", + "evidence": [ + { + "description": "One", + "timestamps": { + "start": "00:12:19", + "end": "00:12:20" + } + } + ], + "question_timestamps": "00:14:42", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "I met three passers-by while walking down this road just now, heading straight ahead.", + "E": "I met four passers-by on this road just now, passing them as I continued forward." + }, + "answer_label": "A", + "answer_text": "Zero.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_6", + "correct_answer": "Zero.", + "original_options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "I met three passers-by while walking down this road just now, heading straight ahead.", + "E": "I met four passers-by on this road just now, passing them as I continued forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_3", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory at the crossroads?", + "answer": "You went straight ahead.", + "question_chinese": "我在十字路口的运动轨迹是什么？", + "answer_chinese": "你直走了。", + "evidence": [ + { + "description": "You went straight ahead.", + "timestamps": { + "start": "00:08:39", + "end": "00:08:41" + } + } + ], + "question_timestamps": "00:09:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You turned left.", + "B": "You turned right.", + "C": "You went straight ahead.", + "D": "You veered slightly left at the crossroads instead of continuing straight down the road.", + "E": "You took the right-hand branch at the crossroads and continued along that road." + }, + "answer_label": "C", + "answer_text": "You went straight ahead.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_3", + "correct_answer": "You went straight ahead.", + "original_options": { + "A": "You turned left.", + "B": "You turned right.", + "C": "You went straight ahead.", + "D": "You veered slightly left at the crossroads instead of continuing straight down the road.", + "E": "You took the right-hand branch at the crossroads and continued along that road." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 13 + ] + }, + { + "video_uid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002", + "video_length": "00:27:26", + "qid": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_4", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory at the crossroads?", + "answer": "You turned left.", + "question_chinese": "我在十字路口的运动轨迹是什么？", + "answer_chinese": "你向左拐了。", + "evidence": [ + { + "description": "You turned left.", + "timestamps": { + "start": "00:14:21", + "end": "00:14:22" + } + } + ], + "question_timestamps": "00:15:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You went straight.", + "B": "You turned left.", + "C": "You turned right.", + "D": "You continued straight through the crossroads rather than taking the left-hand turn.", + "E": "You turned right at the crossroads instead of turning left onto the side road." + }, + "answer_label": "B", + "answer_text": "You turned left.", + "q_id": "c1fca588-dd91-4f40-9b60-01c55e9e03c8-002_4", + "correct_answer": "You turned left.", + "original_options": { + "A": "You went straight.", + "B": "You turned left.", + "C": "You turned right.", + "D": "You continued straight through the crossroads rather than taking the left-hand turn.", + "E": "You turned right at the crossroads instead of turning left onto the side road." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 14 + ] + } +] diff --git a/QAs/d3a0899e-2093-454c-9f65-30087883193a-001-mcq.json b/QAs/d3a0899e-2093-454c-9f65-30087883193a-001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fea98c0617cbb3300474dffbe14892fa809bdd61 --- /dev/null +++ b/QAs/d3a0899e-2093-454c-9f65-30087883193a-001-mcq.json @@ -0,0 +1,906 @@ +[ + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-001", + "video_length": "00:20:00", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-001_1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I just get the spray bottle from?", + "answer": "You took the spray bottle from the floor on the right side of the swimming pool and put it back in its original position.", + "question_chinese": "我刚才把喷水壶从哪里拿到了哪里？", + "answer_chinese": "你把喷水壶从泳池右边的地板上拿走后又放回了原位。", + "evidence": [ + { + "description": "The spray bottle is on the right side of the swimming pool", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "You put the spray bottle back in its original place.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From the pool’s left-side floor.", + "B": "From the floor by the pool steps.", + "C": "You picked up the spray bottle off the floor on the left side of the swimming pool.", + "D": "You took the spray bottle from the floor on the right side of the swimming pool and put it back in its original position.", + "E": "You grabbed the spray bottle from the ground at the pool’s edge, then set it down somewhere else nearby." + }, + "answer_label": "D", + "answer_text": "You took the spray bottle from the floor on the right side of the swimming pool and put it back in its original position.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-001_1", + "correct_answer": "You took the spray bottle from the floor on the right side of the swimming pool and put it back in its original position.", + "original_options": { + "A": "From the pool’s left-side floor.", + "B": "From the floor by the pool steps.", + "C": "You picked up the spray bottle off the floor on the left side of the swimming pool.", + "D": "You took the spray bottle from the floor on the right side of the swimming pool and put it back in its original position.", + "E": "You grabbed the spray bottle from the ground at the pool’s edge, then set it down somewhere else nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 0, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 20 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-001", + "video_length": "00:20:00", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-001_2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I just get the spray bottle from?", + "answer": "You kept picking up the spray bottle from the ground and holding it in your hand.", + "question_chinese": "我刚才把喷水壶从哪里拿到了哪里？", + "answer_chinese": "你一直把喷水壶从地上捡起来后拿在了手里。", + "evidence": [ + { + "description": "The spray bottle is on the right side of the swimming pool", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "You put the spray bottle back in its original place.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:30" + } + }, + { + "description": "You kept picking up the spray bottle from the ground and holding it in your hand.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:01:42", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "From the counter.", + "B": "From the table.", + "C": "You kept picking up the spray bottle from the ground and holding it in your hand.", + "D": "You repeatedly picked up the spray bottle from the shelf and kept it in your hand.", + "E": "You kept lifting the spray bottle from the sink area and holding it instead of leaving it down." + }, + "answer_label": "C", + "answer_text": "You kept picking up the spray bottle from the ground and holding it in your hand.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-001_2", + "correct_answer": "You kept picking up the spray bottle from the ground and holding it in your hand.", + "original_options": { + "A": "From the counter.", + "B": "From the table.", + "C": "You kept picking up the spray bottle from the ground and holding it in your hand.", + "D": "You repeatedly picked up the spray bottle from the shelf and kept it in your hand.", + "E": "You kept lifting the spray bottle from the sink area and holding it instead of leaving it down." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 18 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-001", + "video_length": "00:20:00", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-001_3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many colors of cars did I just pass by?", + "answer": "One types.", + "question_chinese": "我刚才路过了几种颜色的轿车？", + "answer_chinese": "一种。", + "evidence": [ + { + "description": "The first.", + "timestamps": { + "start": "00:02:51", + "end": "00:02:52" + } + } + ], + "question_timestamps": "00:03:55", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two types.", + "B": "One types.", + "C": "Three types.", + "D": "You passed by cars of two different colors just now while continuing forward along the road.", + "E": "You passed by cars showing three different colors while moving forward past them in a row." + }, + "answer_label": "B", + "answer_text": "One types.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-001_3", + "correct_answer": "One types.", + "original_options": { + "A": "Two types.", + "B": "One types.", + "C": "Three types.", + "D": "You passed by cars of two different colors just now while continuing forward along the road.", + "E": "You passed by cars showing three different colors while moving forward past them in a row." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-001", + "video_length": "00:20:00", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-001_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many colors of cars did I just pass by?", + "answer": "Two types.", + "question_chinese": "我刚才路过了几种颜色的轿车？", + "answer_chinese": "两种。", + "evidence": [ + { + "description": "The first.", + "timestamps": { + "start": "00:02:51", + "end": "00:02:52" + } + }, + { + "description": "The second.", + "timestamps": { + "start": "00:02:59", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:04:38", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One type.", + "B": "Two types.", + "C": "Three types.", + "D": "You passed by cars of four different colors during that stretch as you kept moving ahead.", + "E": "You passed by cars in five distinct colors as you went along that section of road." + }, + "answer_label": "B", + "answer_text": "Two types.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-001_4", + "correct_answer": "Two types.", + "original_options": { + "A": "One type.", + "B": "Two types.", + "C": "Three types.", + "D": "You passed by cars of four different colors during that stretch as you kept moving ahead.", + "E": "You passed by cars in five distinct colors as you went along that section of road." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-001", + "video_length": "00:20:00", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-001_5", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What plant did I just squat down to clear?", + "answer": "It's the climbing ivy vine man.", + "question_chinese": "我刚才蹲下来清除的是什么植物？", + "answer_chinese": "是爬山虎藤曼。", + "evidence": [ + { + "description": "It's the climbing ivy vine man.", + "timestamps": { + "start": "00:11:51", + "end": "00:11:59" + } + } + ], + "question_timestamps": "00:12:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's a thorny rose bush, man.", + "B": "It's a bamboo stalk, man.", + "C": "It's the bright sunflower plant beside you, man, not the vine you were clearing.", + "D": "It's the flowering cactus plant behind you, man, the one you squatted to move aside.", + "E": "It's the climbing ivy vine man." + }, + "answer_label": "E", + "answer_text": "It's the climbing ivy vine man.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-001_5", + "correct_answer": "It's the climbing ivy vine man.", + "original_options": { + "A": "It's a thorny rose bush, man.", + "B": "It's a bamboo stalk, man.", + "C": "It's the bright sunflower plant beside you, man, not the vine you were clearing.", + "D": "It's the flowering cactus plant behind you, man, the one you squatted to move aside.", + "E": "It's the climbing ivy vine man." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 15 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-001", + "video_length": "00:20:00", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-001_6", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What plant did I just squat down to clear?", + "answer": "It's the morning glory.", + "question_chinese": "我刚才蹲下来清除的是什么植物？", + "answer_chinese": "是牵牛花。", + "evidence": [ + { + "description": "It's the morning glory.", + "timestamps": { + "start": "00:12:06", + "end": "00:12:11" + } + } + ], + "question_timestamps": "00:13:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's the bluebell.", + "B": "It's the tulip.", + "C": "It's the morning glory.", + "D": "It's the sunflower in front of you, the one you squatted down to clear.", + "E": "It's the daisy behind you, the small plant you just crouched to remove." + }, + "answer_label": "C", + "answer_text": "It's the morning glory.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-001_6", + "correct_answer": "It's the morning glory.", + "original_options": { + "A": "It's the bluebell.", + "B": "It's the tulip.", + "C": "It's the morning glory.", + "D": "It's the sunflower in front of you, the one you squatted down to clear.", + "E": "It's the daisy behind you, the small plant you just crouched to remove." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 2, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 13 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-001", + "video_length": "00:20:00", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-001_9", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many toy cars are there in my room?", + "answer": "Four vehicles.", + "question_chinese": "我的房间里放了几辆玩具车？", + "answer_chinese": "四辆。", + "evidence": [ + { + "description": "Three off-road toy cars.", + "timestamps": { + "start": "00:18:57", + "end": "00:18:58" + } + }, + { + "description": "A toy car of a small car.", + "timestamps": { + "start": "00:18:59", + "end": "00:19:00" + } + } + ], + "question_timestamps": "00:19:01", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three vehicles.", + "B": "Five vehicles.", + "C": "Two vehicles total, with one on the floor and another on a nearby surface.", + "D": "Six vehicles in the room, spread out across the floor and a shelf area.", + "E": "Four vehicles." + }, + "answer_label": "E", + "answer_text": "Four vehicles.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-001_9", + "correct_answer": "Four vehicles.", + "original_options": { + "A": "Three vehicles.", + "B": "Five vehicles.", + "C": "Two vehicles total, with one on the floor and another on a nearby surface.", + "D": "Six vehicles in the room, spread out across the floor and a shelf area.", + "E": "Four vehicles." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 2, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-001", + "video_length": "00:20:00", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-001_10", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many toy cars are there in my room?", + "answer": "Six vehicles.", + "question_chinese": "我的房间里放了几辆玩具车？", + "answer_chinese": "六辆。", + "evidence": [ + { + "description": "Three off-road toy cars.", + "timestamps": { + "start": "00:18:57", + "end": "00:18:58" + } + }, + { + "description": "A toy car of a small car.", + "timestamps": { + "start": "00:18:59", + "end": "00:19:00" + } + }, + { + "description": "Two toy bicycles", + "timestamps": { + "start": "00:19:00", + "end": "00:19:01" + } + } + ], + "question_timestamps": "00:19:53", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five vehicles.", + "B": "Eight vehicles.", + "C": "Six vehicles.", + "D": "Three vehicles total, with two together in one spot and one separate elsewhere.", + "E": "Seven vehicles in the room, with several clustered together and the rest spaced apart." + }, + "answer_label": "C", + "answer_text": "Six vehicles.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-001_10", + "correct_answer": "Six vehicles.", + "original_options": { + "A": "Five vehicles.", + "B": "Eight vehicles.", + "C": "Six vehicles.", + "D": "Three vehicles total, with two together in one spot and one separate elsewhere.", + "E": "Seven vehicles in the room, with several clustered together and the rest spaced apart." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-001", + "video_length": "00:20:00", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-001_7", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I need to use a trolley. Where is the trolley?", + "answer": "The trolley is in front of the green trash can.", + "question_chinese": "我需要用到推车，推车在哪里？", + "answer_chinese": "推车在绿色垃圾桶的前面。", + "evidence": [ + { + "description": "The trolley is in front of the green trash can.", + "timestamps": { + "start": "00:16:12", + "end": "00:16:13" + } + } + ], + "question_timestamps": "00:17:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The trolley is in front of the green trash can.", + "B": "Behind the green trash can.", + "C": "Next to the blue trash can.", + "D": "The trolley is to the left of the green trash can, positioned slightly off to the side.", + "E": "The trolley is near the yellow dumpster, not by the green trash can at all." + }, + "answer_label": "A", + "answer_text": "The trolley is in front of the green trash can.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-001_7", + "correct_answer": "The trolley is in front of the green trash can.", + "original_options": { + "A": "The trolley is in front of the green trash can.", + "B": "Behind the green trash can.", + "C": "Next to the blue trash can.", + "D": "The trolley is to the left of the green trash can, positioned slightly off to the side.", + "E": "The trolley is near the yellow dumpster, not by the green trash can at all." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 2, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 15 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-001", + "video_length": "00:20:00", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-001_8", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I need to use a trolley. Where is the trolley?", + "answer": "The trolley is on the right of the small red one.", + "question_chinese": "我需要用到推车，推车在哪里？", + "answer_chinese": "推车在红色小推车的右边。", + "evidence": [ + { + "description": "The trolley is on the right of the small red one.", + "timestamps": { + "start": "00:16:13", + "end": "00:16:14" + } + } + ], + "question_timestamps": "00:17:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left of the small red one.", + "B": "Behind the small red one.", + "C": "The trolley is in front of the large blue one, not aligned with the small red one.", + "D": "The trolley is below the small green one, positioned lower rather than to the side.", + "E": "The trolley is on the right of the small red one." + }, + "answer_label": "E", + "answer_text": "The trolley is on the right of the small red one.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-001_8", + "correct_answer": "The trolley is on the right of the small red one.", + "original_options": { + "A": "Left of the small red one.", + "B": "Behind the small red one.", + "C": "The trolley is in front of the large blue one, not aligned with the small red one.", + "D": "The trolley is below the small green one, positioned lower rather than to the side.", + "E": "The trolley is on the right of the small red one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 2, + "B": 1, + "C": 2, + "D": 1, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 15 + ] + } +] diff --git a/QAs/d3a0899e-2093-454c-9f65-30087883193a-002-mcq.json b/QAs/d3a0899e-2093-454c-9f65-30087883193a-002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2be55b3df877b667cf4e95c51e5ec5e3febb2382 --- /dev/null +++ b/QAs/d3a0899e-2093-454c-9f65-30087883193a-002-mcq.json @@ -0,0 +1,668 @@ +[ + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-002", + "video_length": "00:20:01", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-002_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the colorful sphere placed?", + "answer": "It was placed on top of the cart.", + "question_chinese": "彩色的球体放在了哪里？", + "answer_chinese": "放在了小车的上面。", + "evidence": [ + { + "description": "It was placed on top of the cart.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:27" + } + } + ], + "question_timestamps": "00:02:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Beside the cart.", + "B": "Under the cart.", + "C": "It was placed in front of the cart on the ground, not up on the cart.", + "D": "It was placed behind the cart against the shelf, instead of being set on top.", + "E": "It was placed on top of the cart." + }, + "answer_label": "E", + "answer_text": "It was placed on top of the cart.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-002_7", + "correct_answer": "It was placed on top of the cart.", + "original_options": { + "A": "Beside the cart.", + "B": "Under the cart.", + "C": "It was placed in front of the cart on the ground, not up on the cart.", + "D": "It was placed behind the cart against the shelf, instead of being set on top.", + "E": "It was placed on top of the cart." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 15 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-002", + "video_length": "00:20:01", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-002_8", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the colorful sphere placed?", + "answer": "Put it in front of the red rope.", + "question_chinese": "彩色的球体放在了哪里？", + "answer_chinese": "放在红色绳子的前面。", + "evidence": [ + { + "description": "Put it in front of the red rope.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:03:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put it behind the red rope.", + "B": "Put it left of the red rope.", + "C": "Put it in front of the red rope.", + "D": "Put it to the right of the red rope, not in front where you were instructed.", + "E": "Put it above the yellow rope instead, rather than placing it in front of the red rope." + }, + "answer_label": "C", + "answer_text": "Put it in front of the red rope.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-002_8", + "correct_answer": "Put it in front of the red rope.", + "original_options": { + "A": "Put it behind the red rope.", + "B": "Put it left of the red rope.", + "C": "Put it in front of the red rope.", + "D": "Put it to the right of the red rope, not in front where you were instructed.", + "E": "Put it above the yellow rope instead, rather than placing it in front of the red rope." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 2, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-002", + "video_length": "00:20:01", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-002_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boards did I rinse just now?", + "answer": "Two.", + "question_chinese": "我刚才冲洗了几块板子？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "The first.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:12" + } + }, + { + "description": "The second", + "timestamps": { + "start": "00:01:21", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "One.", + "B": "Three.", + "C": "You rinsed only one board before moving on to the next task afterward.", + "D": "Two.", + "E": "You rinsed three boards in a row before putting them aside to dry." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-002_5", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You rinsed only one board before moving on to the next task afterward.", + "D": "Two.", + "E": "You rinsed three boards in a row before putting them aside to dry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-002", + "video_length": "00:20:01", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-002_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boards did I rinse just now?", + "answer": "Four.", + "question_chinese": "我刚才冲洗了几块板子？", + "answer_chinese": "四个", + "evidence": [ + { + "description": "The first.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:12" + } + }, + { + "description": "The second", + "timestamps": { + "start": "00:01:21", + "end": "00:01:22" + } + }, + { + "description": "The third", + "timestamps": { + "start": "00:01:23", + "end": "00:01:24" + } + }, + { + "description": "The forth", + "timestamps": { + "start": "00:01:33", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "You rinsed three boards quickly and then stopped washing any more boards.", + "E": "You rinsed five boards total, finishing with the last one at the end." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-002_6", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "You rinsed three boards quickly and then stopped washing any more boards.", + "E": "You rinsed five boards total, finishing with the last one at the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-002", + "video_length": "00:20:01", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-002_2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the helmet directly when washing my hands?", + "answer": "No.", + "question_chinese": "我在洗手时能直接拿到头盔吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "The helmet is on the right side of the sink", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes, you can reach it directly." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-002_2", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes, you can reach it directly." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-002", + "video_length": "00:20:01", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-002_1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the helmet directly when washing my hands?", + "answer": "Yes.", + "question_chinese": "我在洗手时能直接拿到头盔吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The helmet is on the right side of the sink", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:34", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you can't reach it directly.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-002_1", + "correct_answer": "Yes.", + "original_options": { + "A": "No, you can't reach it directly.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 0, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-002", + "video_length": "00:20:01", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-002_3", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the orange stool placed?", + "answer": "It's at the lower left corner of the washbasin.", + "question_chinese": "橙色的凳子放在了哪里？", + "answer_chinese": "在洗手池的左下方。", + "evidence": [ + { + "description": "It's at the lower left corner of the washbasin.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:23" + } + }, + { + "description": "In front of the cabinet.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Upper left of the washbasin.", + "B": "Lower right of the washbasin.", + "C": "It's at the lower left corner of the washbasin.", + "D": "It's positioned at the upper right corner of the washbasin area near the edge.", + "E": "It's set near the middle of the washbasin, not down in the left corner." + }, + "answer_label": "C", + "answer_text": "It's at the lower left corner of the washbasin.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-002_3", + "correct_answer": "It's at the lower left corner of the washbasin.", + "original_options": { + "A": "Upper left of the washbasin.", + "B": "Lower right of the washbasin.", + "C": "It's at the lower left corner of the washbasin.", + "D": "It's positioned at the upper right corner of the washbasin area near the edge.", + "E": "It's set near the middle of the washbasin, not down in the left corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 0, + "C": 1, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-002", + "video_length": "00:20:01", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-002_4", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the orange stool placed?", + "answer": "In front of the cabinet.", + "question_chinese": "橙色的凳子放在了哪里？", + "answer_chinese": "橱柜的前面。", + "evidence": [ + { + "description": "It's at the lower left corner of the washbasin.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:23" + } + }, + { + "description": "In front of the cabinet.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:01:21", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To the left of the cabinet.", + "B": "In front of the cabinet.", + "C": "To the right of the cabinet.", + "D": "It's placed beside the blue sofa rather than directly in front of the cabinet.", + "E": "It's near the front of the desk, not lined up in front of the cabinet." + }, + "answer_label": "B", + "answer_text": "In front of the cabinet.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-002_4", + "correct_answer": "In front of the cabinet.", + "original_options": { + "A": "To the left of the cabinet.", + "B": "In front of the cabinet.", + "C": "To the right of the cabinet.", + "D": "It's placed beside the blue sofa rather than directly in front of the cabinet.", + "E": "It's near the front of the desk, not lined up in front of the cabinet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 0, + "B": 1, + "C": 1, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + } +] diff --git a/QAs/d3a0899e-2093-454c-9f65-30087883193a-003-mcq.json b/QAs/d3a0899e-2093-454c-9f65-30087883193a-003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8244589974c3f361c4eed03eb7520b3676519ca6 --- /dev/null +++ b/QAs/d3a0899e-2093-454c-9f65-30087883193a-003-mcq.json @@ -0,0 +1,1104 @@ +[ + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wooden steps did I paint just when I was on the grass?", + "answer": "One.", + "question_chinese": "刚在我在草地上时我给几层木板阶梯刷了油漆？", + "answer_chinese": "一层", + "evidence": [ + { + "description": "the first.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Four.", + "D": "You painted two wooden steps while you were standing on the grass.", + "E": "You painted four wooden steps on the grass before stopping and moving away." + }, + "answer_label": "A", + "answer_text": "One.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_1", + "correct_answer": "One.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Four.", + "D": "You painted two wooden steps while you were standing on the grass.", + "E": "You painted four wooden steps on the grass before stopping and moving away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 0, + "D": 0, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wooden steps did I paint just when I was on the grass?", + "answer": "Two.", + "question_chinese": "刚在我在草地上时我给几层木板阶梯刷了油漆？", + "answer_chinese": "两层", + "evidence": [ + { + "description": "the first.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:07:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "You painted one wooden step while you were on the grass near the edge.", + "D": "Two.", + "E": "You painted three wooden steps on the grass, completing them one by one." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_2", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "You painted one wooden step while you were on the grass near the edge.", + "D": "Two.", + "E": "You painted three wooden steps on the grass, completing them one by one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 0, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_13", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cars did I see in total just now?", + "answer": "One vehicle.", + "question_chinese": "我刚才一共看到了几辆车？", + "answer_chinese": "一辆", + "evidence": [ + { + "description": "The first.", + "timestamps": { + "start": "00:17:18", + "end": "00:17:20" + } + } + ], + "question_timestamps": "00:20:11", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two vehicles.", + "B": "Three vehicles.", + "C": "You saw two vehicles in total during that brief moment outside just now.", + "D": "You ended up seeing three vehicles altogether during that brief moment outside just now.", + "E": "One vehicle." + }, + "answer_label": "E", + "answer_text": "One vehicle.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_13", + "correct_answer": "One vehicle.", + "original_options": { + "A": "Two vehicles.", + "B": "Three vehicles.", + "C": "You saw two vehicles in total during that brief moment outside just now.", + "D": "You ended up seeing three vehicles altogether during that brief moment outside just now.", + "E": "One vehicle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 2, + "C": 2, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_14", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cars did I see in total just now?", + "answer": "Two vehicles.", + "question_chinese": "我刚才一共看到了几辆车？", + "answer_chinese": "两辆", + "evidence": [ + { + "description": "The first.", + "timestamps": { + "start": "00:17:18", + "end": "00:17:20" + } + }, + { + "description": "the second.", + "timestamps": { + "start": "00:20:24", + "end": "00:20:25" + } + } + ], + "question_timestamps": "00:21:12", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two vehicles.", + "B": "One vehicle.", + "C": "Three vehicles.", + "D": "You saw only one vehicle in total during that brief moment outside.", + "E": "You saw three vehicles altogether during that brief moment outside just now." + }, + "answer_label": "A", + "answer_text": "Two vehicles.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_14", + "correct_answer": "Two vehicles.", + "original_options": { + "A": "Two vehicles.", + "B": "One vehicle.", + "C": "Three vehicles.", + "D": "You saw only one vehicle in total during that brief moment outside.", + "E": "You saw three vehicles altogether during that brief moment outside just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 2, + "B": 2, + "C": 2, + "D": 2, + "E": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_7", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many objects did I use to deal with this can of paint just now?", + "answer": "One type.", + "question_chinese": "我刚才处理这罐油漆用了几种物体？", + "answer_chinese": "一种。", + "evidence": [ + { + "description": "You used a tissue.", + "timestamps": { + "start": "00:14:05", + "end": "00:14:06" + } + } + ], + "question_timestamps": "00:14:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Two types.", + "B": "Three types.", + "C": "You used two different types of objects to handle the paint can.", + "D": "One type.", + "E": "You used three types of objects while dealing with that can of paint." + }, + "answer_label": "D", + "answer_text": "One type.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_7", + "correct_answer": "One type.", + "original_options": { + "A": "Two types.", + "B": "Three types.", + "C": "You used two different types of objects to handle the paint can.", + "D": "One type.", + "E": "You used three types of objects while dealing with that can of paint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 2, + "D": 2, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_8", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many objects did I use to deal with this can of paint just now?", + "answer": "Two types.", + "question_chinese": "我刚才处理这罐油漆用了几种物体？", + "answer_chinese": "两种。", + "evidence": [ + { + "description": "You used a tissue.", + "timestamps": { + "start": "00:14:05", + "end": "00:14:06" + } + }, + { + "description": "You used a hammer.", + "timestamps": { + "start": "00:14:14", + "end": "00:14:15" + } + } + ], + "question_timestamps": "00:14:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "One type.", + "B": "Two types.", + "C": "Three types.", + "D": "You only used one type of object to deal with the paint can.", + "E": "You used three different types of objects while handling that can of paint." + }, + "answer_label": "B", + "answer_text": "Two types.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_8", + "correct_answer": "Two types.", + "original_options": { + "A": "One type.", + "B": "Two types.", + "C": "Three types.", + "D": "You only used one type of object to deal with the paint can.", + "E": "You used three different types of objects while handling that can of paint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking" + }, + "answer_slot_balanced": { + "A": 1, + "B": 2, + "C": 2, + "D": 2, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Did I see any other people when I passed by the black car just now?", + "answer": "Nobody.", + "question_chinese": "我刚才路过黑色轿车时看到有其他人吗？", + "answer_chinese": "没人。", + "evidence": [ + { + "description": "Nobody.", + "timestamps": { + "start": "00:13:50", + "end": "00:13:51" + } + } + ], + "question_timestamps": "00:14:28", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Nobody.", + "B": "Yes, you saw some people." + }, + "answer_label": "A", + "answer_text": "Nobody.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_11", + "correct_answer": "Nobody.", + "original_options": { + "A": "Nobody.", + "B": "Yes, you saw some people." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 2, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_12", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Did I see any other people when I passed by the black car just now?", + "answer": "Yes.", + "question_chinese": "我刚才路过黑色轿车时看到有其他人吗？", + "answer_chinese": "有人", + "evidence": [ + { + "description": "There is someone there.", + "timestamps": { + "start": "00:17:18", + "end": "00:17:20" + } + } + ], + "question_timestamps": "00:17:22", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you didn’t see anyone.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_12", + "correct_answer": "Yes.", + "original_options": { + "A": "No, you didn’t see anyone.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 2, + "B": 2 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_10", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the helmet directly when I'm in front of the washbasin?", + "answer": "No.", + "question_chinese": "我在洗手台前的时候能够直接拿到头盔吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The helmet is on the right side of the sink", + "timestamps": { + "start": "00:13:55", + "end": "00:13:56" + } + } + ], + "question_timestamps": "00:14:33", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can reach it directly.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_10", + "correct_answer": "No.", + "original_options": { + "A": "Yes, you can reach it directly.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_9", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the helmet directly when I'm in front of the washbasin?", + "answer": "Yes.", + "question_chinese": "我在洗手台前的时候能够直接拿到头盔吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The helmet is on the right side of the sink", + "timestamps": { + "start": "00:13:55", + "end": "00:13:56" + } + } + ], + "question_timestamps": "00:15:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No, you can’t reach it directly." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_9", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No, you can’t reach it directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "binary-llm" + ], + "answer_slot_balanced": { + "A": 1, + "B": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_5", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my hammer?", + "answer": "The hammer is in the upper left corner of the red chainsaw.", + "question_chinese": "我的锤子在哪里？", + "answer_chinese": "锤子在红色电锯的左上角。", + "evidence": [ + { + "description": "The hammer is in the upper left corner of the red chainsaw.", + "timestamps": { + "start": "00:14:07", + "end": "00:14:08" + } + } + ], + "question_timestamps": "00:14:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on the red chainsaw.", + "B": "It is under the red chainsaw.", + "C": "The hammer is in the upper right corner of the red chainsaw, not the upper left.", + "D": "The hammer is in the lower left corner of the red chainsaw, not in the upper corner.", + "E": "The hammer is in the upper left corner of the red chainsaw." + }, + "answer_label": "E", + "answer_text": "The hammer is in the upper left corner of the red chainsaw.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_5", + "correct_answer": "The hammer is in the upper left corner of the red chainsaw.", + "original_options": { + "A": "It is on the red chainsaw.", + "B": "It is under the red chainsaw.", + "C": "The hammer is in the upper right corner of the red chainsaw, not the upper left.", + "D": "The hammer is in the lower left corner of the red chainsaw, not in the upper corner.", + "E": "The hammer is in the upper left corner of the red chainsaw." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_6", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my hammer?", + "answer": "The hammer is in your hand.", + "question_chinese": "我的锤子在哪里？", + "answer_chinese": "锤子在你手上。", + "evidence": [ + { + "description": "The hammer is in the upper left corner of the red chainsaw.", + "timestamps": { + "start": "00:14:07", + "end": "00:14:08" + } + }, + { + "description": "The hammer is in your hand.", + "timestamps": { + "start": "00:14:12", + "end": "00:14:13" + } + } + ], + "question_timestamps": "00:14:16", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The hammer is on the table.", + "B": "The hammer is in the toolbox.", + "C": "The hammer is in your hand.", + "D": "The hammer is on the shelf beside you, not being held in your hand.", + "E": "The hammer is lying in front of your hand on the surface instead of in your grip." + }, + "answer_label": "C", + "answer_text": "The hammer is in your hand.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_6", + "correct_answer": "The hammer is in your hand.", + "original_options": { + "A": "The hammer is on the table.", + "B": "The hammer is in the toolbox.", + "C": "The hammer is in your hand.", + "D": "The hammer is on the shelf beside you, not being held in your hand.", + "E": "The hammer is lying in front of your hand on the surface instead of in your grip." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 2, + "D": 1, + "E": 1 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 17 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_3", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After entering the room, where did I see the colorful balls placed?", + "answer": "It was placed on top of the yellow toy car.", + "question_chinese": "进入房间后我看见彩色的球放在了哪里？", + "answer_chinese": "放在了黄色玩具车的上面。", + "evidence": [ + { + "description": "It was placed on top of the yellow toy car.", + "timestamps": { + "start": "00:13:42", + "end": "00:13:43" + } + } + ], + "question_timestamps": "00:13:45", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was under the yellow toy car.", + "B": "It was beside the yellow toy car.", + "C": "It was placed on top of the yellow toy car.", + "D": "It was placed in front of the yellow toy car rather than on top of it.", + "E": "It was placed behind the yellow toy car, not sitting on the top surface." + }, + "answer_label": "C", + "answer_text": "It was placed on top of the yellow toy car.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_3", + "correct_answer": "It was placed on top of the yellow toy car.", + "original_options": { + "A": "It was under the yellow toy car.", + "B": "It was beside the yellow toy car.", + "C": "It was placed on top of the yellow toy car.", + "D": "It was placed in front of the yellow toy car rather than on top of it.", + "E": "It was placed behind the yellow toy car, not sitting on the top surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 0, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 14 + ] + }, + { + "video_uid": "d3a0899e-2093-454c-9f65-30087883193a-003", + "video_length": "00:21:58", + "qid": "d3a0899e-2093-454c-9f65-30087883193a-003_4", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "After entering the room, where did I see the colorful balls placed?", + "answer": "It was placed at the back of the green bicycle.", + "question_chinese": "进入房间后我看见彩色的球放在了哪里？", + "answer_chinese": "放在了绿色自行车的后面。", + "evidence": [ + { + "description": "It was placed on top of the yellow toy car.", + "timestamps": { + "start": "00:13:42", + "end": "00:13:43" + } + }, + { + "description": "It was placed at the back of the green bicycle.", + "timestamps": { + "start": "00:13:50", + "end": "00:13:51" + } + } + ], + "question_timestamps": "00:14:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was in front of the green bicycle.", + "B": "It was placed at the back of the green bicycle.", + "C": "It was beside the green bicycle.", + "D": "It was placed to the left of the green bicycle rather than at the back of it.", + "E": "It was placed to the right of the green bicycle instead of positioned at the back." + }, + "answer_label": "B", + "answer_text": "It was placed at the back of the green bicycle.", + "q_id": "d3a0899e-2093-454c-9f65-30087883193a-003_4", + "correct_answer": "It was placed at the back of the green bicycle.", + "original_options": { + "A": "It was in front of the green bicycle.", + "B": "It was placed at the back of the green bicycle.", + "C": "It was beside the green bicycle.", + "D": "It was placed to the left of the green bicycle rather than at the back of it.", + "E": "It was placed to the right of the green bicycle instead of positioned at the back." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "meta": { + "language": "en", + "pipeline": [ + "neighbors((sub)category)", + "ans-sim-band", + "qa-crossencoder", + "nli-bidirectional", + "gpt-refine", + "mmr-diversity", + "post-check-regenerate", + "per-item-numeric-unify(digit)", + "align-entity" + ], + "params": { + "q_topk": 50, + "min_ans_sim": 0.25, + "max_ans_sim": 0.9, + "mmr_lambda": 0.7, + "group_level": "subcategory_qtype" + }, + "bucket": { + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory" + }, + "answer_slot_balanced": { + "A": 1, + "B": 1, + "C": 1, + "D": 1, + "E": 0 + }, + "numeric_style": "word" + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 17, + 16 + ] + } +] diff --git a/QAs/de8807ad-b86a-46bd-b185-3deba8f02f75-mcq.json b/QAs/de8807ad-b86a-46bd-b185-3deba8f02f75-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2834a32c726c922527561f490ba88a53e4cbd954 --- /dev/null +++ b/QAs/de8807ad-b86a-46bd-b185-3deba8f02f75-mcq.json @@ -0,0 +1,1006 @@ +[ + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_4", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, PAYDAY or AIRHEADS?", + "answer": "AIRHEADS is closer to you.", + "question_chinese": "PAYDAY和AIRHEADS哪个离我更近？", + "answer_chinese": "AIRHEADS离你更近。", + "evidence": [ + { + "description": "AIRHEADS is in front of you.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + }, + { + "description": "PAYDAY is on your left.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:01:05", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "PAYDAY is closer to you.", + "B": "AIRHEADS is closer to you." + }, + "answer_label": "B", + "answer_text": "AIRHEADS is closer to you.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_4", + "correct_answer": "AIRHEADS is closer to you.", + "original_options": { + "A": "PAYDAY is closer to you.", + "B": "AIRHEADS is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_5", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, PAYDAY or AIRHEADS?", + "answer": "PAYDAY is closer to you.", + "question_chinese": "PAYDAY和AIRHEADS哪个离我更近？", + "answer_chinese": "PAYDAY离你更近。", + "evidence": [ + { + "description": "PAYDAY is in front of you.", + "timestamps": { + "start": "00:01:54", + "end": "00:01:56" + } + }, + { + "description": "AIRHEADS is on your right.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:57", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "PAYDAY is closer to you.", + "B": "AIRHEADS is closer to you." + }, + "answer_label": "A", + "answer_text": "PAYDAY is closer to you.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_5", + "correct_answer": "PAYDAY is closer to you.", + "original_options": { + "A": "PAYDAY is closer to you.", + "B": "AIRHEADS is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_12", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the black chair?", + "answer": "No,you can't.", + "question_chinese": "我可以直接坐在黑色椅子上吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The chair is in front of you.", + "timestamps": { + "start": "00:18:54", + "end": "00:18:55" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:18:54", + "end": "00:18:55" + } + }, + { + "description": "You cannot sit directly on it.", + "timestamps": { + "start": "00:18:54", + "end": "00:18:55" + } + } + ], + "question_timestamps": "00:18:55", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_12", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_13", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the black chair?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接坐在黑色椅子上吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The chair is right next to you.", + "timestamps": { + "start": "00:18:59", + "end": "00:19:00" + } + }, + { + "description": "You can sit directly on it.", + "timestamps": { + "start": "00:18:59", + "end": "00:19:00" + } + } + ], + "question_timestamps": "00:19:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you cannot.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_13", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No, you cannot.", + "B": "Yes,you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_0", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I throw the trash directly into the bin?", + "answer": "No,you can't.", + "question_chinese": "我可以直接把垃圾扔到垃圾桶里吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The trash can is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You can't just throw the garbage inside.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:01", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_0", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I throw the trash directly into the bin?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接把垃圾扔到垃圾桶里吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The trash can is right next to you.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + }, + { + "description": "You can just throw the garbage inside.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:22", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_1", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_8", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the oranges directly?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接拿到橘子吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The orange is right in front of you.", + "timestamps": { + "start": "00:03:44", + "end": "00:03:45" + } + }, + { + "description": "You can get it directly.", + "timestamps": { + "start": "00:03:44", + "end": "00:03:45" + } + } + ], + "question_timestamps": "00:03:47", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_8", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No, you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_9", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the oranges directly?", + "answer": "No,you can't.", + "question_chinese": "我可以直接拿到橘子吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The orange is on your left.", + "timestamps": { + "start": "00:03:57", + "end": "00:03:58" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:03:57", + "end": "00:03:58" + } + }, + { + "description": "You cannot directly obtain it.", + "timestamps": { + "start": "00:03:57", + "end": "00:03:58" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_9", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_6", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many things have I held in my hand?", + "answer": "One.", + "question_chinese": "我手里拿过几种东西？", + "answer_chinese": "一种。", + "evidence": [ + { + "description": "You have something in your hand.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four things total, each held in your hand at different moments during the task.", + "E": "Five things total, each picked up and held briefly in your hand at separate times." + }, + "answer_label": "B", + "answer_text": "One.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_6", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four things total, each held in your hand at different moments during the task.", + "E": "Five things total, each picked up and held briefly in your hand at separate times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_7", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many things have I held in my hand?", + "answer": "Two.", + "question_chinese": "我手里拿过几种东西？", + "answer_chinese": "两种。", + "evidence": [ + { + "description": "You have something in your hand.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:07" + } + }, + { + "description": "You have something in your hand.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four different things total, each held in your hand separately at different moments in the video.", + "D": "Two.", + "E": "Five items altogether, each picked up and held in your hand at some point during the task." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_7", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four different things total, each held in your hand separately at different moments in the video.", + "D": "Two.", + "E": "Five items altogether, each picked up and held in your hand at some point during the task." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 17 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75_9_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk from the drinks fridge to the checkout counter?", + "answer": "First, walk to your left.", + "question_chinese": "我是怎么从饮料柜走到收银台的？", + "answer_chinese": "你先往左边走。", + "evidence": [ + { + "description": "Turn left first.", + "timestamps": { + "start": "00:19:00", + "end": "00:19:20" + } + } + ], + "question_timestamps": "00:19:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First, walk to your right.", + "B": "First, walk straight ahead.", + "C": "First, walk backward from the fridge, then continue straight until you reach the checkout counter.", + "D": "First, walk to your left.", + "E": "First, walk to your right, then continue forward until you reach the checkout counter." + }, + "answer_label": "D", + "answer_text": "First, walk to your left.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75_9_T2", + "correct_answer": "First, walk to your left.", + "original_options": { + "A": "First, walk to your right.", + "B": "First, walk straight ahead.", + "C": "First, walk backward from the fridge, then continue straight until you reach the checkout counter.", + "D": "First, walk to your left.", + "E": "First, walk to your right, then continue forward until you reach the checkout counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 14 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75_9_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk from the drinks fridge to the checkout counter?", + "answer": "First, walk to your left, then turn left again and you’ll arrive at the checkout counter.", + "question_chinese": "我是怎么从饮料柜走到收银台的？", + "answer_chinese": "你先往左边走，再往左转到了收银台。", + "evidence": [ + { + "description": "Turn left first.", + "timestamps": { + "start": "00:19:00", + "end": "00:19:20" + } + }, + { + "description": "Then you turn left.", + "timestamps": { + "start": "00:19:38", + "end": "00:19:42" + } + } + ], + "question_timestamps": "00:19:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First, walk to your right.", + "B": "First, walk straight ahead.", + "C": "First, walk to your right, then turn left again to reach the checkout counter.", + "D": "First, walk to your left, then turn left again and you’ll arrive at the checkout counter.", + "E": "First, walk straight ahead, then turn left again and you’ll arrive at the checkout counter." + }, + "answer_label": "D", + "answer_text": "First, walk to your left, then turn left again and you’ll arrive at the checkout counter.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75_9_T1", + "correct_answer": "First, walk to your left, then turn left again and you’ll arrive at the checkout counter.", + "original_options": { + "A": "First, walk to your right.", + "B": "First, walk straight ahead.", + "C": "First, walk to your right, then turn left again to reach the checkout counter.", + "D": "First, walk to your left, then turn left again and you’ll arrive at the checkout counter.", + "E": "First, walk straight ahead, then turn left again and you’ll arrive at the checkout counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 16 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "You turn back first, then go straight ahead.", + "question_chinese": "我的移动轨迹是怎样的？", + "answer_chinese": "你先往后转，然后直走。", + "evidence": [ + { + "description": "You turn back first", + "timestamps": { + "start": "00:00:19", + "end": "00:00:20" + } + }, + { + "description": "Then go straight ahead.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You go straight, then turn back.", + "B": "You turn right, then go straight.", + "C": "You go straight ahead first, then turn back to face where you started.", + "D": "You turn left first, then continue straight ahead without turning back at the beginning.", + "E": "You turn back first, then go straight ahead." + }, + "answer_label": "E", + "answer_text": "You turn back first, then go straight ahead.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_2", + "correct_answer": "You turn back first, then go straight ahead.", + "original_options": { + "A": "You go straight, then turn back.", + "B": "You turn right, then go straight.", + "C": "You go straight ahead first, then turn back to face where you started.", + "D": "You turn left first, then continue straight ahead without turning back at the beginning.", + "E": "You turn back first, then go straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 14 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_3", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "You turn back first, then straight ahead, and then turn right.", + "question_chinese": "我的移动轨迹是怎样的？", + "answer_chinese": "你先往后转，然后直走，然后右转。", + "evidence": [ + { + "description": "You turn back first.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:20" + } + }, + { + "description": "Then straight ahead.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:23" + } + }, + { + "description": "Then turn right.", + "timestamps": { + "start": "00:00:31", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turn back first, then straight ahead, and then turn right.", + "B": "You turn back, then turn left.", + "C": "You turn right, then go straight.", + "D": "You turn back first, then turn right immediately, and only afterward go straight ahead.", + "E": "You go straight ahead first, then turn back, and finish by turning right at the end." + }, + "answer_label": "A", + "answer_text": "You turn back first, then straight ahead, and then turn right.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_3", + "correct_answer": "You turn back first, then straight ahead, and then turn right.", + "original_options": { + "A": "You turn back first, then straight ahead, and then turn right.", + "B": "You turn back, then turn left.", + "C": "You turn right, then go straight.", + "D": "You turn back first, then turn right immediately, and only afterward go straight ahead.", + "E": "You go straight ahead first, then turn back, and finish by turning right at the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 16 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_14", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, eggplant or orange?", + "answer": "Eggplant is closer to you.", + "question_chinese": "茄子和橘子哪个离我更近？", + "answer_chinese": "茄子离你更近。", + "evidence": [ + { + "description": "Eggplant is right next to you.", + "timestamps": { + "start": "00:05:16", + "end": "00:05:17" + } + }, + { + "description": "Oranges are in front of you.", + "timestamps": { + "start": "00:05:16", + "end": "00:05:17" + } + }, + { + "description": "Eggplant is closer to you.", + "timestamps": { + "start": "00:05:16", + "end": "00:05:17" + } + } + ], + "question_timestamps": "00:05:18", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Eggplant is closer to you.", + "B": "Orange is closer to you." + }, + "answer_label": "A", + "answer_text": "Eggplant is closer to you.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_14", + "correct_answer": "Eggplant is closer to you.", + "original_options": { + "A": "Eggplant is closer to you.", + "B": "Orange is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_15", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, eggplant or orange?", + "answer": "Oranges are closer to you.", + "question_chinese": "茄子和橘子哪个离我更近？", + "answer_chinese": "橘子离你更近。", + "evidence": [ + { + "description": "Oranges are closer to you.", + "timestamps": { + "start": "00:05:49", + "end": "00:05:50" + } + }, + { + "description": "Oranges are in front of you.", + "timestamps": { + "start": "00:05:49", + "end": "00:05:50" + } + }, + { + "description": "Eggplant is behind you.", + "timestamps": { + "start": "00:05:49", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:05:51", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Eggplants are closer to you.", + "B": "Oranges are closer to you." + }, + "answer_label": "B", + "answer_text": "Oranges are closer to you.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_15", + "correct_answer": "Oranges are closer to you.", + "original_options": { + "A": "Eggplants are closer to you.", + "B": "Oranges are closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_10", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the beverage in for me?", + "answer": "The drink is on your right.", + "question_chinese": "饮料在我的哪个方向？", + "answer_chinese": "饮料在你右边。", + "evidence": [ + { + "description": "The drink is on your right.", + "timestamps": { + "start": "00:17:57", + "end": "00:18:00" + } + } + ], + "question_timestamps": "00:18:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The drink is on your left.", + "B": "The drink is behind you.", + "C": "The drink is on your right.", + "D": "The drink is directly in front of you, not positioned to your right side.", + "E": "The drink is to your left-hand side, so you would reach across your body to grab it." + }, + "answer_label": "C", + "answer_text": "The drink is on your right.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_10", + "correct_answer": "The drink is on your right.", + "original_options": { + "A": "The drink is on your left.", + "B": "The drink is behind you.", + "C": "The drink is on your right.", + "D": "The drink is directly in front of you, not positioned to your right side.", + "E": "The drink is to your left-hand side, so you would reach across your body to grab it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 17 + ] + }, + { + "video_uid": "de8807ad-b86a-46bd-b185-3deba8f02f75", + "video_length": "00:23:42", + "qid": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_11", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the beverage in for me?", + "answer": "The drink is in front of you.", + "question_chinese": "饮料在我的哪个方向？", + "answer_chinese": "饮料在你前面。", + "evidence": [ + { + "description": "The drink is in front of you.", + "timestamps": { + "start": "00:18:15", + "end": "00:18:16" + } + } + ], + "question_timestamps": "00:18:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The drink is behind you.", + "B": "The drink is on your right.", + "C": "The drink is off to your left side rather than directly in front of where you're facing.", + "D": "The drink is positioned behind you, so you would need to turn around to see it.", + "E": "The drink is in front of you." + }, + "answer_label": "E", + "answer_text": "The drink is in front of you.", + "q_id": "de8807ad-b86a-46bd-b185-3deba8f02f75-mcq_fixed_11", + "correct_answer": "The drink is in front of you.", + "original_options": { + "A": "The drink is behind you.", + "B": "The drink is on your right.", + "C": "The drink is off to your left side rather than directly in front of where you're facing.", + "D": "The drink is positioned behind you, so you would need to turn around to see it.", + "E": "The drink is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 16 + ] + } +] diff --git a/QAs/egolife_A1_001-mcq.json b/QAs/egolife_A1_001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..960dac35b4ea50712deb5726b69eae44cf5a9229 --- /dev/null +++ b/QAs/egolife_A1_001-mcq.json @@ -0,0 +1,1456 @@ +[ + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_2_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the reflector relative to me?", + "answer": "The reflector is to your front-left.", + "question_chinese": "反光板在我的什么位置？", + "answer_chinese": "反光板在你的左前方", + "evidence": [ + { + "description": "The reflector is on your front left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The reflector is front-right.", + "B": "The reflector is to your front-left.", + "C": "The reflector is back-left.", + "D": "The reflector is to your back-right, diagonally behind you rather than in front of you.", + "E": "The reflector is straight in front of you, not offset to the left side of your view." + }, + "answer_label": "B", + "answer_text": "The reflector is to your front-left.", + "q_id": "egolife_A1_001_2_T2", + "correct_answer": "The reflector is to your front-left.", + "original_options": { + "A": "The reflector is front-right.", + "B": "The reflector is to your front-left.", + "C": "The reflector is back-left.", + "D": "The reflector is to your back-right, diagonally behind you rather than in front of you.", + "E": "The reflector is straight in front of you, not offset to the left side of your view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_2_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the reflector relative to me?", + "answer": "The reflector is to your back-right.", + "question_chinese": "反光板在我的什么位置？", + "answer_chinese": "反光板在你的右后方", + "evidence": [ + { + "description": "The reflector is behind you on your right.", + "timestamps": { + "start": "00:01:50", + "end": "00:02:02" + } + } + ], + "question_timestamps": "00:02:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The reflector is back-left.", + "B": "The reflector is to your back-right.", + "C": "The reflector is front-right.", + "D": "The reflector is to your front-left, diagonally ahead of you rather than behind on your right side.", + "E": "The reflector is directly behind you, centered rather than offset to the right side of your body." + }, + "answer_label": "B", + "answer_text": "The reflector is to your back-right.", + "q_id": "egolife_A1_001_2_T1", + "correct_answer": "The reflector is to your back-right.", + "original_options": { + "A": "The reflector is back-left.", + "B": "The reflector is to your back-right.", + "C": "The reflector is front-right.", + "D": "The reflector is to your front-left, diagonally ahead of you rather than behind on your right side.", + "E": "The reflector is directly behind you, centered rather than offset to the right side of your body." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_5", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "If I want to retrieve the charging cable that was stored in the black data cable bag, what should I do?", + "answer": "You should turn to your left rear, walk straight for about four steps to reach the item cabinet. From the bottom layer, take out the black data cable package with an orange label on the surface, and open it. Then you will be able to obtain the data cable.", + "question_chinese": "如果我现在想去拿被存放到黑色数据线包中的数据线，我应该怎么做？", + "answer_chinese": "你应该向左后方转身，直走四步左右的距离到物品柜，从最底下一层，取出表面有橙色标签的黑色数据线包，并打开它，便可以获得数据线", + "evidence": [ + { + "description": "I stored the data cable in the black data cable bag with an orange label.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:00" + } + }, + { + "description": "I turned left, reached the wall, then turned around and walked to the table.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:11" + } + }, + { + "description": "I stacked the data cable packages with orange labels on top of those with green labels.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:13" + } + }, + { + "description": "I turned to the left rear, walked about four steps forward, and placed the black data cable package on the bottom shelf of the storage cabinet.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:20" + } + }, + { + "description": "I went to the bedroom, stayed for a while to explain, and then returned along the same route to the original position.", + "timestamps": { + "start": "00:02:13", + "end": "00:05:53" + } + } + ], + "question_timestamps": "00:05:57", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn to your back-right.", + "B": "You should turn to your left rear, walk straight for about four steps to reach the item cabinet. From the bottom layer, take out the black data cable package with an orange label on the surface, and open it. Then you will be able to obtain the data cable.", + "C": "Take the pouch from the top layer.", + "D": "Turn back-right, walk four steps to the cabinet, then take the black bag with a blue label.", + "E": "Turn back-left, walk five steps to the cabinet, then take the black bag with a red label." + }, + "answer_label": "B", + "answer_text": "You should turn to your left rear, walk straight for about four steps to reach the item cabinet. From the bottom layer, take out the black data cable package with an orange label on the surface, and open it. Then you will be able to obtain the data cable.", + "q_id": "egolife_A1_001_5", + "correct_answer": "You should turn to your left rear, walk straight for about four steps to reach the item cabinet. From the bottom layer, take out the black data cable package with an orange label on the surface, and open it. Then you will be able to obtain the data cable.", + "original_options": { + "A": "Turn to your back-right.", + "B": "You should turn to your left rear, walk straight for about four steps to reach the item cabinet. From the bottom layer, take out the black data cable package with an orange label on the surface, and open it. Then you will be able to obtain the data cable.", + "C": "Take the pouch from the top layer.", + "D": "Turn back-right, walk four steps to the cabinet, then take the black bag with a blue label.", + "E": "Turn back-left, walk five steps to the cabinet, then take the black bag with a red label." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_6", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "If I want to retrieve the charging cable stored in the black data cable bag right now, what should I do?", + "answer": "Turn right, walk straight for about six steps to the item cabinet. From the bottom layer, take out the black data cable package with an orange label on the surface, and open it. Then you can obtain the data cable.", + "question_chinese": "如果我现在想去拿被存放到黑色数据线包中的数据线，我应该怎么做？", + "answer_chinese": "向右转，直走六步左右的距离到物品柜，从最底下一层，取出表面有橙色标签的黑色数据线包，并打开它，便可以获得数据线", + "evidence": [ + { + "description": "I stored the data cable in the black data cable bag with an orange label.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:00" + } + }, + { + "description": "I turned left, reached the wall, then turned around and walked to the table.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:11" + } + }, + { + "description": "I stacked the data cable packages with orange labels on top of those with green labels.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:13" + } + }, + { + "description": "I turned to the left rear, walked about four steps forward, and placed the black data cable package on the bottom shelf of the storage cabinet.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:20" + } + }, + { + "description": "I went to the bedroom, stayed for a while to explain, and then returned along the same route to the original position.", + "timestamps": { + "start": "00:02:13", + "end": "00:05:53" + } + } + ], + "question_timestamps": "00:06:26", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left and walk six steps.", + "B": "Take it from the middle layer.", + "C": "Turn right, walk five steps to the cabinet, then take the black pouch with a red label and open it.", + "D": "Turn right, walk six steps to the cabinet, but pick up the blue data-cable package with the orange label instead.", + "E": "Turn right, walk straight for about six steps to the item cabinet. From the bottom layer, take out the black data cable package with an orange label on the surface, and open it. Then you can obtain the data cable." + }, + "answer_label": "E", + "answer_text": "Turn right, walk straight for about six steps to the item cabinet. From the bottom layer, take out the black data cable package with an orange label on the surface, and open it. Then you can obtain the data cable.", + "q_id": "egolife_A1_001_6", + "correct_answer": "Turn right, walk straight for about six steps to the item cabinet. From the bottom layer, take out the black data cable package with an orange label on the surface, and open it. Then you can obtain the data cable.", + "original_options": { + "A": "Turn left and walk six steps.", + "B": "Take it from the middle layer.", + "C": "Turn right, walk five steps to the cabinet, then take the black pouch with a red label and open it.", + "D": "Turn right, walk six steps to the cabinet, but pick up the blue data-cable package with the orange label instead.", + "E": "Turn right, walk straight for about six steps to the item cabinet. From the bottom layer, take out the black data cable package with an orange label on the surface, and open it. Then you can obtain the data cable." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_5_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the process like when I went to retrieve the cable that was stored in the black data-cable pouch?", + "answer": "You turned around toward your back-left, then walked straight about four steps to reach the storage cabinet.", + "question_chinese": "我去拿被存放到黑色数据线包中的数据线的过程是怎么样的？", + "answer_chinese": "你向左后方转身，直走四步左右的距离到达了物品柜", + "evidence": [ + { + "description": "I turned left, reached the wall, then turned around and walked to the table.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:11" + } + }, + { + "description": "I turned to the left rear, walked about four steps forward, and placed the black data cable package on the bottom shelf of the storage cabinet.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned toward your back-right.", + "B": "You turned around toward your back-left, then walked straight about four steps to reach the storage cabinet.", + "C": "You walked about five steps forward.", + "D": "You turned toward your front-left, then walked straight about four steps to reach the storage cabinet.", + "E": "You turned toward your back-left, then walked straight about three steps before arriving at the storage cabinet." + }, + "answer_label": "B", + "answer_text": "You turned around toward your back-left, then walked straight about four steps to reach the storage cabinet.", + "q_id": "egolife_A1_001_5_T2", + "correct_answer": "You turned around toward your back-left, then walked straight about four steps to reach the storage cabinet.", + "original_options": { + "A": "You turned toward your back-right.", + "B": "You turned around toward your back-left, then walked straight about four steps to reach the storage cabinet.", + "C": "You walked about five steps forward.", + "D": "You turned toward your front-left, then walked straight about four steps to reach the storage cabinet.", + "E": "You turned toward your back-left, then walked straight about three steps before arriving at the storage cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_5_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the process like when I went to retrieve the cable that was stored in the black data-cable pouch?", + "answer": "You should turn around toward your back-left and walk straight about four steps to the storage cabinet. From the very bottom shelf, take out the black data-cable pouch with an orange label on its surface and open it to get the data cable.", + "question_chinese": "我去拿被存放到黑色数据线包中的数据线的过程是怎么样的？", + "answer_chinese": "你应该向左后方转身，直走四步左右的距离到物品柜，从最底下一层，取出表面有橙色标签的黑色数据线包，并打开它，便可以获得数据线", + "evidence": [ + { + "description": "I stored the data cable in the black data cable bag with an orange label.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:00" + } + }, + { + "description": "I turned left, reached the wall, then turned around and walked to the table.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:11" + } + }, + { + "description": "I stacked the data cable packages with orange labels on top of those with green labels.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:13" + } + }, + { + "description": "I turned to the left rear, walked about four steps forward, and placed the black data cable package on the bottom shelf of the storage cabinet.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:20" + } + }, + { + "description": "I went to the bedroom, stayed for a while to explain, and then returned along the same route to the original position.", + "timestamps": { + "start": "00:02:20", + "end": "00:05:53" + } + } + ], + "question_timestamps": "00:05:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You should turn around toward your back-left and walk straight about four steps to the storage cabinet. From the very bottom shelf, take out the black data-cable pouch with an orange label on its surface and open it to get the data cable.", + "B": "Turn toward your back-right.", + "C": "Take it from the middle shelf.", + "D": "You should turn toward your front-left and walk about four steps to the storage cabinet to get it.", + "E": "You should turn toward your back-left, walk about four steps, then take the black pouch with a green label." + }, + "answer_label": "A", + "answer_text": "You should turn around toward your back-left and walk straight about four steps to the storage cabinet. From the very bottom shelf, take out the black data-cable pouch with an orange label on its surface and open it to get the data cable.", + "q_id": "egolife_A1_001_5_T1", + "correct_answer": "You should turn around toward your back-left and walk straight about four steps to the storage cabinet. From the very bottom shelf, take out the black data-cable pouch with an orange label on its surface and open it to get the data cable.", + "original_options": { + "A": "You should turn around toward your back-left and walk straight about four steps to the storage cabinet. From the very bottom shelf, take out the black data-cable pouch with an orange label on its surface and open it to get the data cable.", + "B": "Turn toward your back-right.", + "C": "Take it from the middle shelf.", + "D": "You should turn toward your front-left and walk about four steps to the storage cabinet to get it.", + "E": "You should turn toward your back-left, walk about four steps, then take the black pouch with a green label." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_6_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "I want to get the cable that’s stored in the black data-cable pouch—what should I do?", + "answer": "You should first turn right, then walk straight about six steps to the storage cabinet.", + "question_chinese": "我想去拿被存放到黑色数据线包中的数据线，我应该怎么做？", + "answer_chinese": "你应该先向右转，直走六步左右的距离到物品柜", + "evidence": [ + { + "description": "I turned left, reached the wall, then turned around and walked to the table.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:11" + } + }, + { + "description": "I stacked the data cable packages with orange labels on top of those with green labels.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:13" + } + }, + { + "description": "I turned to the left rear, walked about four steps forward, and placed the black data cable package on the bottom shelf of the storage cabinet.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then walk six steps.", + "B": "Turn right, then walk four steps.", + "C": "You should turn right, walk about six steps, and head to the item cabinet instead of the storage cabinet.", + "D": "You should turn right and walk about seven steps straight until you reach the storage cabinet.", + "E": "You should first turn right, then walk straight about six steps to the storage cabinet." + }, + "answer_label": "E", + "answer_text": "You should first turn right, then walk straight about six steps to the storage cabinet.", + "q_id": "egolife_A1_001_6_T2", + "correct_answer": "You should first turn right, then walk straight about six steps to the storage cabinet.", + "original_options": { + "A": "Turn left, then walk six steps.", + "B": "Turn right, then walk four steps.", + "C": "You should turn right, walk about six steps, and head to the item cabinet instead of the storage cabinet.", + "D": "You should turn right and walk about seven steps straight until you reach the storage cabinet.", + "E": "You should first turn right, then walk straight about six steps to the storage cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_6_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "I want to get the cable that’s stored in the black data-cable pouch—what should I do?", + "answer": "You should turn right and walk straight about six steps to the storage cabinet. From the very bottom shelf, take out the black data-cable pouch with an orange label on its surface and open it to get the data cable.", + "question_chinese": "我想去拿被存放到黑色数据线包中的数据线，我应该怎么做？", + "answer_chinese": "你应该向右转，直走六步左右的距离到物品柜，从最底下一层，取出表面有橙色标签的黑色数据线包，并打开它，便可以获得数据线", + "evidence": [ + { + "description": "I stored the data cable in the black data cable bag with an orange label.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:00" + } + }, + { + "description": "I turned left, reached the wall, then turned around and walked to the table.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:11" + } + }, + { + "description": "I stacked the data cable packages with orange labels on top of those with green labels.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:13" + } + }, + { + "description": "I turned to the left rear, walked about four steps forward, and placed the black data cable package on the bottom shelf of the storage cabinet.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:20" + } + }, + { + "description": "I went to the bedroom, stayed for a while to explain, and then returned along the same route to the original position.", + "timestamps": { + "start": "00:02:20", + "end": "00:05:53" + } + } + ], + "question_timestamps": "00:06:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left and walk six steps.", + "B": "Take it from the top shelf.", + "C": "Turn right, walk about six steps to the item cabinet, then take the black data-cable package with an orange label.", + "D": "You should turn right and walk straight about six steps to the storage cabinet. From the very bottom shelf, take out the black data-cable pouch with an orange label on its surface and open it to get the data cable.", + "E": "Turn right, walk about six steps to the storage cabinet, then grab the black pouch with a blue label and open it." + }, + "answer_label": "D", + "answer_text": "You should turn right and walk straight about six steps to the storage cabinet. From the very bottom shelf, take out the black data-cable pouch with an orange label on its surface and open it to get the data cable.", + "q_id": "egolife_A1_001_6_T1", + "correct_answer": "You should turn right and walk straight about six steps to the storage cabinet. From the very bottom shelf, take out the black data-cable pouch with an orange label on its surface and open it to get the data cable.", + "original_options": { + "A": "Turn left and walk six steps.", + "B": "Take it from the top shelf.", + "C": "Turn right, walk about six steps to the item cabinet, then take the black data-cable package with an orange label.", + "D": "You should turn right and walk straight about six steps to the storage cabinet. From the very bottom shelf, take out the black data-cable pouch with an orange label on its surface and open it to get the data cable.", + "E": "Turn right, walk about six steps to the storage cabinet, then grab the black pouch with a blue label and open it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 22 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_9", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction am I facing while sitting on the horse-shaped stool?", + "answer": "You are above the horse-shaped chair.", + "question_chinese": "我现在在马形凳子什么方位？", + "answer_chinese": "你在马形凳子的上方", + "evidence": [ + { + "description": "I picked up the stool which looked like a horse and had orange and black stripes on it, and moved it forward about four steps. Then I placed it down and sat on it.", + "timestamps": { + "start": "00:07:35", + "end": "00:07:43" + } + } + ], + "question_timestamps": "00:07:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are above the horse-shaped chair.", + "B": "You are in front of the stool.", + "C": "You are behind the stool.", + "D": "You are to the left of the horse-shaped chair while sitting on it, facing across the hallway.", + "E": "You are to the right of the horse-shaped chair, positioned beside it rather than above it." + }, + "answer_label": "A", + "answer_text": "You are above the horse-shaped chair.", + "q_id": "egolife_A1_001_9", + "correct_answer": "You are above the horse-shaped chair.", + "original_options": { + "A": "You are above the horse-shaped chair.", + "B": "You are in front of the stool.", + "C": "You are behind the stool.", + "D": "You are to the left of the horse-shaped chair while sitting on it, facing across the hallway.", + "E": "You are to the right of the horse-shaped chair, positioned beside it rather than above it." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_7_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the red mineral water bottle right now?", + "answer": "You’re standing in front of the table, so you can reach it. The red mineral-water bottle had always been placed at the left corner of the table by your left hand, so you could grab it just by reaching out.", + "question_chinese": "我现在能拿到红色矿泉水瓶吗？", + "answer_chinese": "你现在站在桌子面前，能拿到，红色矿泉水瓶之前一直放在你左手边的桌角，伸手就能拿到", + "evidence": [ + { + "description": "The red mineral water bottle has been placed in the corner of the table near the door.It is only 40 centimeters away from the current position.", + "timestamps": { + "start": "00:00:00", + "end": "00:08:10" + } + } + ], + "question_timestamps": "00:08:14", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You’re standing in front of the table, so you can reach it. The red mineral-water bottle had always been placed at the left corner of the table by your left hand, so you could grab it just by reaching out.", + "B": "No. You’re not close enough to the table, so the red mineral-water bottle is out of reach right now." + }, + "answer_label": "A", + "answer_text": "You’re standing in front of the table, so you can reach it. The red mineral-water bottle had always been placed at the left corner of the table by your left hand, so you could grab it just by reaching out.", + "q_id": "egolife_A1_001_7_T1", + "correct_answer": "You’re standing in front of the table, so you can reach it. The red mineral-water bottle had always been placed at the left corner of the table by your left hand, so you could grab it just by reaching out.", + "original_options": { + "A": "You’re standing in front of the table, so you can reach it. The red mineral-water bottle had always been placed at the left corner of the table by your left hand, so you could grab it just by reaching out.", + "B": "No. You’re not close enough to the table, so the red mineral-water bottle is out of reach right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_7_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the red mineral water bottle right now?", + "answer": "No. The red water bottle is on the table, and you’re about 3 meters away from it, so you can’t reach it.", + "question_chinese": "我现在能拿到红色矿泉水瓶吗？", + "answer_chinese": "不行，红色水瓶放在桌子上，你离瓶子大概有有3米的距离，不能拿到", + "evidence": [ + { + "description": "You walked three steps away from the table, out of reach of the bottle.", + "timestamps": { + "start": "00:08:08", + "end": "00:08:17" + } + } + ], + "question_timestamps": "00:08:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. You’re right in front of the table, and the red mineral-water bottle is at the left corner within easy reach.", + "B": "No. The red water bottle is on the table, and you’re about 3 meters away from it, so you can’t reach it." + }, + "answer_label": "B", + "answer_text": "No. The red water bottle is on the table, and you’re about 3 meters away from it, so you can’t reach it.", + "q_id": "egolife_A1_001_7_T2", + "correct_answer": "No. The red water bottle is on the table, and you’re about 3 meters away from it, so you can’t reach it.", + "original_options": { + "A": "Yes. You’re right in front of the table, and the red mineral-water bottle is at the left corner within easy reach.", + "B": "No. The red water bottle is on the table, and you’re about 3 meters away from it, so you can’t reach it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 22 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_10_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many unopened boxes did I see on the table?", + "answer": "When you were passing the phone back and forth, there were four unopened boxes on the table, and the table was in front of you.", + "question_chinese": "我看到桌上有几个未打开的盒子？", + "answer_chinese": "当你们互相传递手机时，桌子上有4个未打开的盒子，桌子在你前面", + "evidence": [ + { + "description": "When passing the mobile phone, it can be noticed that there are a total of four closed boxes on the table.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two unopened boxes.", + "B": "When you were passing the phone back and forth, there were four unopened boxes on the table, and the table was in front of you.", + "C": "Six unopened boxes.", + "D": "When you were passing the phone back and forth, there were three unopened boxes on the table in front of you.", + "E": "When you were passing the phone back and forth, there were five unopened boxes on the table in front of you." + }, + "answer_label": "B", + "answer_text": "When you were passing the phone back and forth, there were four unopened boxes on the table, and the table was in front of you.", + "q_id": "egolife_A1_001_10_T2", + "correct_answer": "When you were passing the phone back and forth, there were four unopened boxes on the table, and the table was in front of you.", + "original_options": { + "A": "Two unopened boxes.", + "B": "When you were passing the phone back and forth, there were four unopened boxes on the table, and the table was in front of you.", + "C": "Six unopened boxes.", + "D": "When you were passing the phone back and forth, there were three unopened boxes on the table in front of you.", + "E": "When you were passing the phone back and forth, there were five unopened boxes on the table in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_10_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many unopened boxes did I see on the table?", + "answer": "There are two unopened boxes on the table, and the table is in front of you.", + "question_chinese": "我看到桌上有几个未打开的盒子？", + "answer_chinese": "桌子上有2个未打开的盒子，桌子在你前面", + "evidence": [ + { + "description": "Opened two boxes one after another.", + "timestamps": { + "start": "00:00:49", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are two unopened boxes on the table, and the table is in front of you.", + "B": "One unopened box.", + "C": "Four unopened boxes.", + "D": "There are zero unopened boxes on the table, and the table is in front of you.", + "E": "There are three unopened boxes on the table, and the table is in front of you." + }, + "answer_label": "A", + "answer_text": "There are two unopened boxes on the table, and the table is in front of you.", + "q_id": "egolife_A1_001_10_T1", + "correct_answer": "There are two unopened boxes on the table, and the table is in front of you.", + "original_options": { + "A": "There are two unopened boxes on the table, and the table is in front of you.", + "B": "One unopened box.", + "C": "Four unopened boxes.", + "D": "There are zero unopened boxes on the table, and the table is in front of you.", + "E": "There are three unopened boxes on the table, and the table is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_4_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many opened white-and-blue boxes did I see on the table?", + "answer": "Right now, none of the boxes on the table are open, and the table is in front of you.", + "question_chinese": "我看到桌子上有几个打开的白蓝色的盒子？", + "answer_chinese": "桌子上现在没有盒子被打开，桌子在你前面", + "evidence": [ + { + "description": "The boxes on the table are unopened.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:01:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Right now, none of the boxes on the table are open, and the table is in front of you.", + "B": "One open box.", + "C": "Three open boxes.", + "D": "Right now, two of the white-and-blue boxes on the table are open, and the table is in front of you.", + "E": "Right now, four of the white-and-blue boxes on the table are open, and the table is in front of you." + }, + "answer_label": "A", + "answer_text": "Right now, none of the boxes on the table are open, and the table is in front of you.", + "q_id": "egolife_A1_001_4_T1", + "correct_answer": "Right now, none of the boxes on the table are open, and the table is in front of you.", + "original_options": { + "A": "Right now, none of the boxes on the table are open, and the table is in front of you.", + "B": "One open box.", + "C": "Three open boxes.", + "D": "Right now, two of the white-and-blue boxes on the table are open, and the table is in front of you.", + "E": "Right now, four of the white-and-blue boxes on the table are open, and the table is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_4_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many opened white-and-blue boxes did I see on the table?", + "answer": "There are four open boxes on the table, and the table is in front of you.", + "question_chinese": "我看到桌子上有几个打开的白蓝色的盒子？", + "answer_chinese": "桌子上有四个打开的盒子，桌子在你前面", + "evidence": [ + { + "description": "There are four open boxes on the table.", + "timestamps": { + "start": "00:01:25", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:02:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two open boxes.", + "B": "There are four open boxes on the table, and the table is in front of you.", + "C": "Six open boxes.", + "D": "There are three open boxes on the table, and the table is in front of you.", + "E": "There are five open boxes on the table, and the table is in front of you." + }, + "answer_label": "B", + "answer_text": "There are four open boxes on the table, and the table is in front of you.", + "q_id": "egolife_A1_001_4_T2", + "correct_answer": "There are four open boxes on the table, and the table is in front of you.", + "original_options": { + "A": "Two open boxes.", + "B": "There are four open boxes on the table, and the table is in front of you.", + "C": "Six open boxes.", + "D": "There are three open boxes on the table, and the table is in front of you.", + "E": "There are five open boxes on the table, and the table is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_11_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many black round boxes did I see on the table?", + "answer": "Right now you can see nine black boxes on the table, and the table is to your front-right.", + "question_chinese": "我看到桌子上有几个黑色的原形的盒子？", + "answer_chinese": "现在你看到桌子上有9个黑色的盒子，桌子在你右前方", + "evidence": [ + { + "description": "There are nine black boxes on the table.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Seven black boxes.", + "B": "Eleven black boxes.", + "C": "Right now you can see nine black boxes on the table, and the table is to your front-right.", + "D": "Right now you can see eight black boxes on the table, and the table is to your front-right.", + "E": "Right now you can see ten black boxes on the table, and the table is to your front-right." + }, + "answer_label": "C", + "answer_text": "Right now you can see nine black boxes on the table, and the table is to your front-right.", + "q_id": "egolife_A1_001_11_T2", + "correct_answer": "Right now you can see nine black boxes on the table, and the table is to your front-right.", + "original_options": { + "A": "Seven black boxes.", + "B": "Eleven black boxes.", + "C": "Right now you can see nine black boxes on the table, and the table is to your front-right.", + "D": "Right now you can see eight black boxes on the table, and the table is to your front-right.", + "E": "Right now you can see ten black boxes on the table, and the table is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_11_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many black round boxes did I see on the table?", + "answer": "Right now, you can see three black boxes on the table, and the table is to your front-right.", + "question_chinese": "我看到桌子上有几个黑色的原形的盒子？", + "answer_chinese": "现在你看到桌子上有3个黑色的盒子，桌子在你右前方", + "evidence": [ + { + "description": "Put the six black round boxes on the table onto the shelf.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:33" + } + } + ], + "question_timestamps": "00:02:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One black box.", + "B": "Right now, you can see three black boxes on the table, and the table is to your front-right.", + "C": "Five black boxes.", + "D": "Right now, you can see two black boxes on the table, and the table is to your front-right.", + "E": "Right now, you can see four black boxes on the table, and the table is to your front-right." + }, + "answer_label": "B", + "answer_text": "Right now, you can see three black boxes on the table, and the table is to your front-right.", + "q_id": "egolife_A1_001_11_T1", + "correct_answer": "Right now, you can see three black boxes on the table, and the table is to your front-right.", + "original_options": { + "A": "One black box.", + "B": "Right now, you can see three black boxes on the table, and the table is to your front-right.", + "C": "Five black boxes.", + "D": "Right now, you can see two black boxes on the table, and the table is to your front-right.", + "E": "Right now, you can see four black boxes on the table, and the table is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_12_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many laptop computers do I see in front of me?", + "answer": "In total, there are seven laptops in front of you.", + "question_chinese": "我面前看到多少台笔记本电脑？", + "answer_chinese": "你面前共有7台笔记本电脑", + "evidence": [ + { + "description": "There were a total of 7 laptops on the desk and the shelves.", + "timestamps": { + "start": "00:03:39", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:03:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Six laptops.", + "B": "Eight laptops.", + "C": "You see five laptops lined up across the table directly in front of you.", + "D": "In total, there are seven laptops in front of you.", + "E": "You see nine laptops spread out in a row directly ahead of you on the table." + }, + "answer_label": "D", + "answer_text": "In total, there are seven laptops in front of you.", + "q_id": "egolife_A1_001_12_T2", + "correct_answer": "In total, there are seven laptops in front of you.", + "original_options": { + "A": "Six laptops.", + "B": "Eight laptops.", + "C": "You see five laptops lined up across the table directly in front of you.", + "D": "In total, there are seven laptops in front of you.", + "E": "You see nine laptops spread out in a row directly ahead of you on the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_12_T1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many laptop computers do I see in front of me?", + "answer": "Right now, you have six laptops in front of you.", + "question_chinese": "我面前看到多少台笔记本电脑？", + "answer_chinese": "我现在面前有6台笔记本电脑", + "evidence": [ + { + "description": "There are four computers on the table, one on the shelf and one on the bed.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:47" + } + } + ], + "question_timestamps": "00:03:49", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five laptops.", + "B": "Seven laptops.", + "C": "Right now, you have six laptops in front of you.", + "D": "You have four laptops directly ahead of you, arranged in a neat row.", + "E": "You have eight laptops in front of you, sitting side by side across the surface." + }, + "answer_label": "C", + "answer_text": "Right now, you have six laptops in front of you.", + "q_id": "egolife_A1_001_12_T1", + "correct_answer": "Right now, you have six laptops in front of you.", + "original_options": { + "A": "Five laptops.", + "B": "Seven laptops.", + "C": "Right now, you have six laptops in front of you.", + "D": "You have four laptops directly ahead of you, arranged in a neat row.", + "E": "You have eight laptops in front of you, sitting side by side across the surface." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_1_T2", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the whiteboard relative to me?", + "answer": "When you were adjusting the phone countdown timer, the whiteboard was directly in front of you.", + "question_chinese": "白板在相对于我的什么位置？", + "answer_chinese": "当你在调节手机倒计时的时候，白板在你的正前方", + "evidence": [ + { + "description": "While holding the phone to adjust the countdown, he/she is facing the whiteboard directly.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was directly behind you.", + "B": "It was directly to your left.", + "C": "While you adjusted the phone countdown timer, the whiteboard was directly to your right.", + "D": "As you were setting the phone countdown timer, the whiteboard was above you rather than ahead.", + "E": "When you were adjusting the phone countdown timer, the whiteboard was directly in front of you." + }, + "answer_label": "E", + "answer_text": "When you were adjusting the phone countdown timer, the whiteboard was directly in front of you.", + "q_id": "egolife_A1_001_1_T2", + "correct_answer": "When you were adjusting the phone countdown timer, the whiteboard was directly in front of you.", + "original_options": { + "A": "It was directly behind you.", + "B": "It was directly to your left.", + "C": "While you adjusted the phone countdown timer, the whiteboard was directly to your right.", + "D": "As you were setting the phone countdown timer, the whiteboard was above you rather than ahead.", + "E": "When you were adjusting the phone countdown timer, the whiteboard was directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_1_T1", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the whiteboard relative to me?", + "answer": "When you were facing the window, the whiteboard was directly to your left.", + "question_chinese": "白板在相对于我的什么位置？", + "answer_chinese": "当你面向窗户时，白板在你的正左边", + "evidence": [ + { + "description": "When the meeting began, the whiteboard was right in front of me.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + }, + { + "description": "I walked around the table and moved towards the left front. I reached the position right in front of the window.", + "timestamps": { + "start": "00:01:45", + "end": "00:01:47" + } + }, + { + "description": "I walked to the position of the door, facing the window to my right front.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:54" + } + } + ], + "question_timestamps": "00:01:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was directly in front of you.", + "B": "It was directly behind you.", + "C": "When you were facing the window, the whiteboard was directly to your right.", + "D": "While you were looking toward the window, the whiteboard was positioned above you instead of beside you.", + "E": "When you were facing the window, the whiteboard was directly to your left." + }, + "answer_label": "E", + "answer_text": "When you were facing the window, the whiteboard was directly to your left.", + "q_id": "egolife_A1_001_1_T1", + "correct_answer": "When you were facing the window, the whiteboard was directly to your left.", + "original_options": { + "A": "It was directly in front of you.", + "B": "It was directly behind you.", + "C": "When you were facing the window, the whiteboard was directly to your right.", + "D": "While you were looking toward the window, the whiteboard was positioned above you instead of beside you.", + "E": "When you were facing the window, the whiteboard was directly to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_3_T2", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the yellow bottle on the windowsill relative to me?", + "answer": "The yellow water bottle on the windowsill is to your right.", + "question_chinese": "窗台上的黄色瓶子在相对于我的什么位置？", + "answer_chinese": "窗台上的黄色水瓶在你的右边", + "evidence": [ + { + "description": "The yellow bottle is on the windowsill, and you are standing to the left of the windowsill.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It is to your left.", + "B": "It is in front of you.", + "C": "The yellow water bottle on the windowsill is to your right.", + "D": "The yellow water bottle on the windowsill is behind you, not on your right side.", + "E": "The yellow water bottle on the windowsill is above you, higher up than your eye level." + }, + "answer_label": "C", + "answer_text": "The yellow water bottle on the windowsill is to your right.", + "q_id": "egolife_A1_001_3_T2", + "correct_answer": "The yellow water bottle on the windowsill is to your right.", + "original_options": { + "A": "It is to your left.", + "B": "It is in front of you.", + "C": "The yellow water bottle on the windowsill is to your right.", + "D": "The yellow water bottle on the windowsill is behind you, not on your right side.", + "E": "The yellow water bottle on the windowsill is above you, higher up than your eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_3_T1", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the yellow bottle on the windowsill relative to me?", + "answer": "The yellow bottle on the windowsill is in front of you.", + "question_chinese": "窗台上的黄色瓶子在相对于我的什么位置？", + "answer_chinese": "窗台上的黄色瓶子在你的前面", + "evidence": [ + { + "description": "You walked over to the opposite side of the windowsill.The yellow bottle is on the windowsill.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:49" + } + } + ], + "question_timestamps": "00:01:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It is to your right.", + "B": "It is behind you.", + "C": "The yellow bottle on the windowsill is to your left, not positioned straight ahead of you.", + "D": "The yellow bottle on the windowsill is above you, higher than where you’re looking forward.", + "E": "The yellow bottle on the windowsill is in front of you." + }, + "answer_label": "E", + "answer_text": "The yellow bottle on the windowsill is in front of you.", + "q_id": "egolife_A1_001_3_T1", + "correct_answer": "The yellow bottle on the windowsill is in front of you.", + "original_options": { + "A": "It is to your right.", + "B": "It is behind you.", + "C": "The yellow bottle on the windowsill is to your left, not positioned straight ahead of you.", + "D": "The yellow bottle on the windowsill is above you, higher than where you’re looking forward.", + "E": "The yellow bottle on the windowsill is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_8_T2", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the red mineral water bottle relative to me?", + "answer": "The red mineral-water bottle is to your front-left.", + "question_chinese": "红色矿泉水瓶在相对于我的什么位置？", + "answer_chinese": "红色矿泉水瓶在你的左前方", + "evidence": [ + { + "description": "The red mineral water bottle is on the left side of the table you are facing directly.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The red mineral-water bottle is to your front-left.", + "B": "The red mineral-water bottle is directly in front of you." + }, + "answer_label": "A", + "answer_text": "The red mineral-water bottle is to your front-left.", + "q_id": "egolife_A1_001_8_T2", + "correct_answer": "The red mineral-water bottle is to your front-left.", + "original_options": { + "A": "The red mineral-water bottle is to your front-left.", + "B": "The red mineral-water bottle is directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_001", + "video_length": "00:11:48", + "qid": "egolife_A1_001_8_T1", + "qid_num": 26, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the red mineral water bottle relative to me?", + "answer": "The red mineral-water bottle is directly in front of you.", + "question_chinese": "红色矿泉水瓶在相对于我的什么位置？", + "answer_chinese": "红色矿泉水瓶在你的正前方", + "evidence": [ + { + "description": "From the perspective where the red mineral water bottle is on your front left, you walked a few steps to the left, then turned around to face the red mineral water bottle directly.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:49" + } + } + ], + "question_timestamps": "00:01:51", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The red mineral-water bottle is to your front-left.", + "B": "The red mineral-water bottle is directly in front of you." + }, + "answer_label": "B", + "answer_text": "The red mineral-water bottle is directly in front of you.", + "q_id": "egolife_A1_001_8_T1", + "correct_answer": "The red mineral-water bottle is directly in front of you.", + "original_options": { + "A": "The red mineral-water bottle is to your front-left.", + "B": "The red mineral-water bottle is directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + } +] diff --git a/QAs/egolife_A1_002-mcq.json b/QAs/egolife_A1_002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..63135763633e440d118fbcd4c4ec18be46b05d3b --- /dev/null +++ b/QAs/egolife_A1_002-mcq.json @@ -0,0 +1,1427 @@ +[ + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_3_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the bedroom stool relative to me?", + "answer": "The bedroom stool is in front of you.", + "question_chinese": "卧室的凳子在我的哪里？", + "answer_chinese": "卧室的凳子在你的前面", + "evidence": [ + { + "description": "The stool in the bedroom is in front of you.", + "timestamps": { + "start": "00:02:42", + "end": "00:02:48" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The bedroom stool is in front of you.", + "D": "The bedroom stool sits slightly to your right as you face into the room from where you are standing.", + "E": "The bedroom stool is off to your back-left, just out of your direct line of sight." + }, + "answer_label": "C", + "answer_text": "The bedroom stool is in front of you.", + "q_id": "egolife_A1_002_3_T2", + "correct_answer": "The bedroom stool is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The bedroom stool is in front of you.", + "D": "The bedroom stool sits slightly to your right as you face into the room from where you are standing.", + "E": "The bedroom stool is off to your back-left, just out of your direct line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_3_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the bedroom stool relative to me?", + "answer": "The bedroom stool is to your left.", + "question_chinese": "卧室的凳子在我的哪里？", + "answer_chinese": "卧室的凳子在你的左边", + "evidence": [ + { + "description": "The stool in the bedroom is on your left.", + "timestamps": { + "start": "00:02:47", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:02:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your right.", + "C": "The bedroom stool is behind you, near where you just came from in the room.", + "D": "The bedroom stool is to your left.", + "E": "The bedroom stool is positioned ahead and slightly to your right rather than on your left." + }, + "answer_label": "D", + "answer_text": "The bedroom stool is to your left.", + "q_id": "egolife_A1_002_3_T1", + "correct_answer": "The bedroom stool is to your left.", + "original_options": { + "A": "In front of you.", + "B": "To your right.", + "C": "The bedroom stool is behind you, near where you just came from in the room.", + "D": "The bedroom stool is to your left.", + "E": "The bedroom stool is positioned ahead and slightly to your right rather than on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_11_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the bedroom doorway relative to me?", + "answer": "The bedroom doorway is behind you.", + "question_chinese": "卧室门口在我的哪里？", + "answer_chinese": "卧室门口在你的身后", + "evidence": [ + { + "description": "The doorway is behind you.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:01:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The bedroom doorway is behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "The bedroom doorway is to your right side, not behind you, as you face forward.", + "E": "The bedroom doorway is in front of you and slightly off-center, rather than at your back." + }, + "answer_label": "A", + "answer_text": "The bedroom doorway is behind you.", + "q_id": "egolife_A1_002_11_T1", + "correct_answer": "The bedroom doorway is behind you.", + "original_options": { + "A": "The bedroom doorway is behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "The bedroom doorway is to your right side, not behind you, as you face forward.", + "E": "The bedroom doorway is in front of you and slightly off-center, rather than at your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_11_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the bedroom doorway relative to me?", + "answer": "The bedroom doorway is to your front-left.", + "question_chinese": "卧室门口在我的哪里？", + "answer_chinese": "卧室门口在你的左前方", + "evidence": [ + { + "description": "The doorway is on your left front.", + "timestamps": { + "start": "00:05:39", + "end": "00:05:45" + } + } + ], + "question_timestamps": "00:05:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The bedroom doorway is to your front-left.", + "B": "Front-right of you.", + "C": "Back-left of you.", + "D": "The bedroom doorway is directly in front of you, not angled toward your front-left.", + "E": "The bedroom doorway sits behind you on the right side instead of being forward and left." + }, + "answer_label": "A", + "answer_text": "The bedroom doorway is to your front-left.", + "q_id": "egolife_A1_002_11_T2", + "correct_answer": "The bedroom doorway is to your front-left.", + "original_options": { + "A": "The bedroom doorway is to your front-left.", + "B": "Front-right of you.", + "C": "Back-left of you.", + "D": "The bedroom doorway is directly in front of you, not angled toward your front-left.", + "E": "The bedroom doorway sits behind you on the right side instead of being forward and left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_11", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go back to the bedroom, how should I get there?", + "answer": "You should walk left to exit the laundry room, then walk straight about eight steps forward-left to enter the bedroom.", + "question_chinese": "如果我想回卧室，我应该怎么走？", + "answer_chinese": "你应该向左走出洗衣房，然后向左前方直走8步左右的距离进入卧室。", + "evidence": [ + { + "description": "I stood up, walked about three steps forward-left, and entered the bedroom.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:43" + } + }, + { + "description": "I turned around, walked straight about ten steps forward-left, and entered the laundry room.", + "timestamps": { + "start": "00:02:51", + "end": "00:02:57" + } + } + ], + "question_timestamps": "00:03:01", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight.", + "B": "Go left, then forward-right.", + "C": "Walk right to exit the laundry room, then continue straight about eight steps to reach the bedroom.", + "D": "You should walk left to exit the laundry room, then walk straight about eight steps forward-left to enter the bedroom.", + "E": "Exit the laundry room by backing out to the right, then walk forward-left roughly six steps into the bedroom." + }, + "answer_label": "D", + "answer_text": "You should walk left to exit the laundry room, then walk straight about eight steps forward-left to enter the bedroom.", + "q_id": "egolife_A1_002_11", + "correct_answer": "You should walk left to exit the laundry room, then walk straight about eight steps forward-left to enter the bedroom.", + "original_options": { + "A": "Turn right, then go straight.", + "B": "Go left, then forward-right.", + "C": "Walk right to exit the laundry room, then continue straight about eight steps to reach the bedroom.", + "D": "You should walk left to exit the laundry room, then walk straight about eight steps forward-left to enter the bedroom.", + "E": "Exit the laundry room by backing out to the right, then walk forward-left roughly six steps into the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_12", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go back to the bedroom, how should I get there?", + "answer": "You should turn toward your rear-right. Walk straight for about five steps to enter the bedroom.", + "question_chinese": "如果我想回卧室，我应该怎么走？", + "answer_chinese": "你应当向右后方转弯。直走5步左右的距离进入卧室。", + "evidence": [ + { + "description": "I stood up, walked about three steps forward-left, and entered the bedroom.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:43" + } + }, + { + "description": "I turned around, walked straight about ten steps forward-left, and entered the laundry room.", + "timestamps": { + "start": "00:02:51", + "end": "00:02:57" + } + }, + { + "description": "After I turned left to exit the laundry room, I continued to turn left and walked toward the hallway.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:12" + } + } + ], + "question_timestamps": "00:03:16", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn toward your rear-left.", + "B": "Turn toward your front-right.", + "C": "Turn toward your rear-right and walk straight for about three steps to enter the bedroom.", + "D": "Turn toward your front-left, then walk straight about five steps until you reach the bedroom entrance.", + "E": "You should turn toward your rear-right. Walk straight for about five steps to enter the bedroom." + }, + "answer_label": "E", + "answer_text": "You should turn toward your rear-right. Walk straight for about five steps to enter the bedroom.", + "q_id": "egolife_A1_002_12", + "correct_answer": "You should turn toward your rear-right. Walk straight for about five steps to enter the bedroom.", + "original_options": { + "A": "Turn toward your rear-left.", + "B": "Turn toward your front-right.", + "C": "Turn toward your rear-right and walk straight for about three steps to enter the bedroom.", + "D": "Turn toward your front-left, then walk straight about five steps until you reach the bedroom entrance.", + "E": "You should turn toward your rear-right. Walk straight for about five steps to enter the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the window in my living room located?", + "answer": "The living room window is directly in front of you.", + "question_chinese": "客厅窗户在我的什么位置？", + "answer_chinese": "客厅窗户在你的正前方", + "evidence": [ + { + "description": "The living room window is directly in front of me.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Directly to your left.", + "B": "Directly behind you.", + "C": "The living room window is on your right side rather than centered directly in front of you.", + "D": "The living room window is directly in front of you.", + "E": "The window is slightly off to your left-front, not straight ahead in your direct line of sight." + }, + "answer_label": "D", + "answer_text": "The living room window is directly in front of you.", + "q_id": "egolife_A1_002_1", + "correct_answer": "The living room window is directly in front of you.", + "original_options": { + "A": "Directly to your left.", + "B": "Directly behind you.", + "C": "The living room window is on your right side rather than centered directly in front of you.", + "D": "The living room window is directly in front of you.", + "E": "The window is slightly off to your left-front, not straight ahead in your direct line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the window in my living room located?", + "answer": "The window is on your right side.", + "question_chinese": "客厅窗户在我的什么位置？", + "answer_chinese": "窗户在你的右侧", + "evidence": [ + { + "description": "The living room window is directly in front of me.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:11" + } + }, + { + "description": "I walked to the side of the table that faces the whiteboard directly.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:45" + } + }, + { + "description": "From my current perspective, the whiteboard is to the front-left of the window.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:46" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "On your left side.", + "B": "Directly behind you.", + "C": "The window is on your right side.", + "D": "The window is directly in front of you, not positioned along your right-hand side.", + "E": "The window sits on your left side as you face forward, opposite from where your right arm points." + }, + "answer_label": "C", + "answer_text": "The window is on your right side.", + "q_id": "egolife_A1_002_2", + "correct_answer": "The window is on your right side.", + "original_options": { + "A": "On your left side.", + "B": "Directly behind you.", + "C": "The window is on your right side.", + "D": "The window is directly in front of you, not positioned along your right-hand side.", + "E": "The window sits on your left side as you face forward, opposite from where your right arm points." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 18 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_7", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the scorecard with my hand?", + "answer": "Yes, the clapperboard has been placed on the edge of the table in front of you all the time.", + "question_chinese": "我可以用手够到场记牌吗？", + "answer_chinese": "可以，场记板一直放在你面前的桌子边缘。", + "evidence": [ + { + "description": "The clapperboard is on the edge of the table directly in front of me.", + "timestamps": { + "start": "00:09:53", + "end": "00:09:54" + } + } + ], + "question_timestamps": "00:09:58", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you can’t; it’s across the table, out of reach.", + "B": "Yes, the clapperboard has been placed on the edge of the table in front of you all the time." + }, + "answer_label": "B", + "answer_text": "Yes, the clapperboard has been placed on the edge of the table in front of you all the time.", + "q_id": "egolife_A1_002_7", + "correct_answer": "Yes, the clapperboard has been placed on the edge of the table in front of you all the time.", + "original_options": { + "A": "No, you can’t; it’s across the table, out of reach.", + "B": "Yes, the clapperboard has been placed on the edge of the table in front of you all the time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_6_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the clapperboard with my hand?", + "answer": "No—you can’t. The clapperboard is on the edge of the table across from you, about two to three steps away.", + "question_chinese": "我可以用手够到场记牌吗？", + "answer_chinese": "不可以，场记板在你对面桌子的边缘，离你两三步远。", + "evidence": [ + { + "description": "The clapperboard is on the edge of the table directly in front of me.", + "timestamps": { + "start": "00:09:53", + "end": "00:09:54" + } + }, + { + "description": "No one has moved the clapperboard during this period, so it remains on the edge of the table.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:25" + } + } + ], + "question_timestamps": "00:10:27", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. It’s on the table edge right in front of you, within arm’s reach.", + "B": "No—you can’t. The clapperboard is on the edge of the table across from you, about two to three steps away." + }, + "answer_label": "B", + "answer_text": "No—you can’t. The clapperboard is on the edge of the table across from you, about two to three steps away.", + "q_id": "egolife_A1_002_6_T2", + "correct_answer": "No—you can’t. The clapperboard is on the edge of the table across from you, about two to three steps away.", + "original_options": { + "A": "Yes—you can. It’s on the table edge right in front of you, within arm’s reach.", + "B": "No—you can’t. The clapperboard is on the edge of the table across from you, about two to three steps away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_6_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the clapperboard with my hand?", + "answer": "Yes—you can. The clapperboard is in front of you.", + "question_chinese": "我可以用手够到场记牌吗？", + "answer_chinese": "可以，现在场记板在你前面。", + "evidence": [ + { + "description": "The clapperboard is on the edge of the table directly in front of me.", + "timestamps": { + "start": "00:09:53", + "end": "00:09:54" + } + }, + { + "description": "I walked two or three steps forward-right along the table and reached the vicinity of the whiteboard.", + "timestamps": { + "start": "00:10:25", + "end": "00:10:29" + } + } + ], + "question_timestamps": "00:10:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. The clapperboard is in front of you.", + "B": "No—you can’t. The clapperboard is across from you, a few steps away on the table edge." + }, + "answer_label": "A", + "answer_text": "Yes—you can. The clapperboard is in front of you.", + "q_id": "egolife_A1_002_6_T1", + "correct_answer": "Yes—you can. The clapperboard is in front of you.", + "original_options": { + "A": "Yes—you can. The clapperboard is in front of you.", + "B": "No—you can’t. The clapperboard is across from you, a few steps away on the table edge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_7_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many opened blue-and-white boxes did I see?", + "answer": "You saw three open blue-and-white boxes, and they were in front of you.", + "question_chinese": "我看到几个打开的蓝白颜色的盒子？", + "answer_chinese": "你看到3个打开的蓝白颜色的盒子，它们在你前面", + "evidence": [ + { + "description": "You see three open blue and white boxes.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw three open blue-and-white boxes, and they were in front of you.", + "B": "Two open blue-and-white boxes.", + "C": "Four open blue-and-white boxes.", + "D": "You saw one open blue-and-white box, and it was placed directly in front of you.", + "E": "You saw five open blue-and-white boxes, and they were lined up in front of you." + }, + "answer_label": "A", + "answer_text": "You saw three open blue-and-white boxes, and they were in front of you.", + "q_id": "egolife_A1_002_7_T2", + "correct_answer": "You saw three open blue-and-white boxes, and they were in front of you.", + "original_options": { + "A": "You saw three open blue-and-white boxes, and they were in front of you.", + "B": "Two open blue-and-white boxes.", + "C": "Four open blue-and-white boxes.", + "D": "You saw one open blue-and-white box, and it was placed directly in front of you.", + "E": "You saw five open blue-and-white boxes, and they were lined up in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_7_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many opened blue-and-white boxes did I see?", + "answer": "You saw two open blue-and-white boxes, and they were in front of you.", + "question_chinese": "我看到几个打开的蓝白颜色的盒子？", + "answer_chinese": "你看到2个打开的蓝白颜色的盒子，它们在你前面", + "evidence": [ + { + "description": "You see two open blue and white boxes.", + "timestamps": { + "start": "00:00:41", + "end": "00:00:46" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One open blue-and-white box.", + "B": "Three open blue-and-white boxes.", + "C": "You saw zero open blue-and-white boxes, so there weren’t any open boxes in front of you.", + "D": "You saw four open blue-and-white boxes, and they were positioned in front of you.", + "E": "You saw two open blue-and-white boxes, and they were in front of you." + }, + "answer_label": "E", + "answer_text": "You saw two open blue-and-white boxes, and they were in front of you.", + "q_id": "egolife_A1_002_7_T1", + "correct_answer": "You saw two open blue-and-white boxes, and they were in front of you.", + "original_options": { + "A": "One open blue-and-white box.", + "B": "Three open blue-and-white boxes.", + "C": "You saw zero open blue-and-white boxes, so there weren’t any open boxes in front of you.", + "D": "You saw four open blue-and-white boxes, and they were positioned in front of you.", + "E": "You saw two open blue-and-white boxes, and they were in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_3", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many blue and white boxes are on the table?", + "answer": "Four.", + "question_chinese": "桌子上有几个蓝白颜色的盒子？", + "answer_chinese": "4", + "evidence": [ + { + "description": "At the very beginning of the video, there are a total of 4 blue-and-white boxes on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "Two blue-and-white boxes are on the table, sitting next to each other near the center.", + "E": "Six blue-and-white boxes are on the table, spread out across the surface in front of you." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "egolife_A1_002_3", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "Two blue-and-white boxes are on the table, sitting next to each other near the center.", + "E": "Six blue-and-white boxes are on the table, spread out across the surface in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_4", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many blue and white boxes are on the table?", + "answer": "Two.", + "question_chinese": "桌子上有几个蓝白颜色的盒子？", + "answer_chinese": "2", + "evidence": [ + { + "description": "At the very beginning of the video, there are a total of 4 blue-and-white boxes on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "I took two boxes from the living room to the bedroom.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four blue-and-white boxes are on the table, spaced apart across the tabletop in front of you.", + "E": "Five blue-and-white boxes are on the table, covering much of the surface near you." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "egolife_A1_002_4", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four blue-and-white boxes are on the table, spaced apart across the tabletop in front of you.", + "E": "Five blue-and-white boxes are on the table, covering much of the surface near you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_5_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the water bottle on the table relative to me?", + "answer": "The mineral water bottle on the table is to your front-right.", + "question_chinese": "桌子上的矿泉水瓶在我的哪里？", + "answer_chinese": "桌子上的矿泉水瓶在你的右前方", + "evidence": [ + { + "description": "The mineral water bottle on the table is on your right front.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The mineral water bottle on the table is to your front-left.", + "B": "The mineral water bottle on the table is to your front-right." + }, + "answer_label": "B", + "answer_text": "The mineral water bottle on the table is to your front-right.", + "q_id": "egolife_A1_002_5_T1", + "correct_answer": "The mineral water bottle on the table is to your front-right.", + "original_options": { + "A": "The mineral water bottle on the table is to your front-left.", + "B": "The mineral water bottle on the table is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_5_T2", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the water bottle on the table relative to me?", + "answer": "The mineral water bottle on the table is to your left.", + "question_chinese": "桌子上的矿泉水瓶在我的哪里？", + "answer_chinese": "桌子上的矿泉水瓶在你的左边", + "evidence": [ + { + "description": "The mineral water bottle on the table is on your left.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:10:22", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The mineral water bottle on the table is to your left.", + "B": "The mineral water bottle on the table is to your right." + }, + "answer_label": "A", + "answer_text": "The mineral water bottle on the table is to your left.", + "q_id": "egolife_A1_002_5_T2", + "correct_answer": "The mineral water bottle on the table is to your left.", + "original_options": { + "A": "The mineral water bottle on the table is to your left.", + "B": "The mineral water bottle on the table is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_8_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the vase on the table relative to me?", + "answer": "The vase on the table is in front of you.", + "question_chinese": "桌子上的花瓶在我的哪里？", + "answer_chinese": "桌子上的花瓶在你的前方", + "evidence": [ + { + "description": "The vase on the table is ahead of you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The vase is behind you.", + "B": "The vase is to your left.", + "C": "The vase on the table is positioned to your right as you face it.", + "D": "The vase on the table is in front of you.", + "E": "The vase on the table sits behind you, opposite your current facing direction." + }, + "answer_label": "D", + "answer_text": "The vase on the table is in front of you.", + "q_id": "egolife_A1_002_8_T2", + "correct_answer": "The vase on the table is in front of you.", + "original_options": { + "A": "The vase is behind you.", + "B": "The vase is to your left.", + "C": "The vase on the table is positioned to your right as you face it.", + "D": "The vase on the table is in front of you.", + "E": "The vase on the table sits behind you, opposite your current facing direction." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_8_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the vase on the table relative to me?", + "answer": "The vase on the table is to your front-left.", + "question_chinese": "桌子上的花瓶在我的哪里？", + "answer_chinese": "桌子上的花瓶在你的左前方", + "evidence": [ + { + "description": "The vase on the table is on your left front.", + "timestamps": { + "start": "00:00:27", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:00:34", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The vase is to your front-right.", + "B": "The vase on the table is to your front-left.", + "C": "The vase is to your back-left.", + "D": "The vase on the table is directly in front of you, not offset to the left.", + "E": "The vase on the table is behind and slightly left of you rather than ahead on the left." + }, + "answer_label": "B", + "answer_text": "The vase on the table is to your front-left.", + "q_id": "egolife_A1_002_8_T1", + "correct_answer": "The vase on the table is to your front-left.", + "original_options": { + "A": "The vase is to your front-right.", + "B": "The vase on the table is to your front-left.", + "C": "The vase is to your back-left.", + "D": "The vase on the table is directly in front of you, not offset to the left.", + "E": "The vase on the table is behind and slightly left of you rather than ahead on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_4_T2", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the table's position relative to me?", + "answer": "The table is now in front of you.", + "question_chinese": "桌子在我的哪个方位？", + "answer_chinese": "桌子现在在你的前面", + "evidence": [ + { + "description": "The table is in front of you now.", + "timestamps": { + "start": "00:01:06", + "end": "00:01:13" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The table is behind you.", + "B": "The table is to your left.", + "C": "The table is now in front of you.", + "D": "The table is now to your right, not positioned directly ahead of you.", + "E": "The table is now above you, rather than at your eye level in front." + }, + "answer_label": "C", + "answer_text": "The table is now in front of you.", + "q_id": "egolife_A1_002_4_T2", + "correct_answer": "The table is now in front of you.", + "original_options": { + "A": "The table is behind you.", + "B": "The table is to your left.", + "C": "The table is now in front of you.", + "D": "The table is now to your right, not positioned directly ahead of you.", + "E": "The table is now above you, rather than at your eye level in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_4_T1", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the table's position relative to me?", + "answer": "The table is now to your right.", + "question_chinese": "桌子在我的哪个方位？", + "answer_chinese": "桌子现在在你的右边。", + "evidence": [ + { + "description": "The table is on your right now.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:46" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The table is now to your right.", + "B": "The table is in front of you.", + "C": "The table is behind you.", + "D": "The table is now to your left, opposite the side you’re facing toward.", + "E": "The table is now above you, not at your right-hand side near the floor." + }, + "answer_label": "A", + "answer_text": "The table is now to your right.", + "q_id": "egolife_A1_002_4_T1", + "correct_answer": "The table is now to your right.", + "original_options": { + "A": "The table is now to your right.", + "B": "The table is in front of you.", + "C": "The table is behind you.", + "D": "The table is now to your left, opposite the side you’re facing toward.", + "E": "The table is now above you, not at your right-hand side near the floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_5", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "In which direction is the table located relative to me?", + "answer": "The table is in your front-left.", + "question_chinese": "桌子在我的哪个方位？", + "answer_chinese": "桌子在你的左前方。", + "evidence": [ + { + "description": "The table is in the exact center of the living room.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + }, + { + "description": "I walked straight two or three steps forward-left, then entered the bedroom. After that, I turned around and walked straight about ten steps forward-left again, and then entered the laundry room.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:48" + } + }, + { + "description": "I turned around in the laundry room and faced the original direction of the whiteboard.", + "timestamps": { + "start": "00:02:58", + "end": "00:02:59" + } + } + ], + "question_timestamps": "00:03:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The table is in your front-left.", + "B": "The table is in your front-right.", + "C": "The table is in your back-center.", + "D": "The table is directly in front of you, not angled toward your left side.", + "E": "The table is to your right side rather than positioned ahead and left of you." + }, + "answer_label": "A", + "answer_text": "The table is in your front-left.", + "q_id": "egolife_A1_002_5", + "correct_answer": "The table is in your front-left.", + "original_options": { + "A": "The table is in your front-left.", + "B": "The table is in your front-right.", + "C": "The table is in your back-center.", + "D": "The table is directly in front of you, not angled toward your left side.", + "E": "The table is to your right side rather than positioned ahead and left of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_6", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "In which direction is the table relative to me?", + "answer": "The table is now at your rear-right.", + "question_chinese": "桌子在我的哪个方位？", + "answer_chinese": "桌子现在在你的右后方。", + "evidence": [ + { + "description": "The table is in the exact center of the living room.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + }, + { + "description": "I walked straight two or three steps forward-left, then entered the bedroom. After that, I turned around and walked straight about ten steps forward-left again, and then entered the laundry room.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:48" + } + }, + { + "description": "I turned around in the laundry room and faced the original direction of the whiteboard.", + "timestamps": { + "start": "00:02:58", + "end": "00:02:59" + } + }, + { + "description": "After I turned left to exit the laundry room, I continued to turn left and walked toward the hallway.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:03:17", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The table is now at your rear-left.", + "B": "The table is now at your rear-right.", + "C": "The table is now at your front-left.", + "D": "The table is now at your front-center, not located behind you on the right.", + "E": "The table is now at your side-right, not diagonally behind you toward the right." + }, + "answer_label": "B", + "answer_text": "The table is now at your rear-right.", + "q_id": "egolife_A1_002_6", + "correct_answer": "The table is now at your rear-right.", + "original_options": { + "A": "The table is now at your rear-left.", + "B": "The table is now at your rear-right.", + "C": "The table is now at your front-left.", + "D": "The table is now at your front-center, not located behind you on the right.", + "E": "The table is now at your side-right, not diagonally behind you toward the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_12_T1", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the window relative to me?", + "answer": "The window is in front of you.", + "question_chinese": "窗户在我的哪里？", + "answer_chinese": "窗户在你的前方", + "evidence": [ + { + "description": "The window is ahead of you.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The window is in front of you.", + "B": "The window is behind you.", + "C": "The window is to your left.", + "D": "The window is to your right, not positioned straight ahead of you.", + "E": "The window is above you, rather than at your forward-facing eye-level view." + }, + "answer_label": "A", + "answer_text": "The window is in front of you.", + "q_id": "egolife_A1_002_12_T1", + "correct_answer": "The window is in front of you.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is behind you.", + "C": "The window is to your left.", + "D": "The window is to your right, not positioned straight ahead of you.", + "E": "The window is above you, rather than at your forward-facing eye-level view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_002", + "video_length": "00:12:00", + "qid": "egolife_A1_002_12_T2", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the window relative to me?", + "answer": "The window is to your right.", + "question_chinese": "窗户在我的哪里？", + "answer_chinese": "窗户在你的右边", + "evidence": [ + { + "description": "The window is on your right.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The window is directly in front of you, facing you as you look ahead.", + "D": "The window is to your right.", + "E": "The window is behind you, so you would need to turn around to see it." + }, + "answer_label": "D", + "answer_text": "The window is to your right.", + "q_id": "egolife_A1_002_12_T2", + "correct_answer": "The window is to your right.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The window is directly in front of you, facing you as you look ahead.", + "D": "The window is to your right.", + "E": "The window is behind you, so you would need to turn around to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 15 + ] + } +] diff --git a/QAs/egolife_A1_003-mcq.json b/QAs/egolife_A1_003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d5319908c7aaf0e62ed3fa1d9249d9ab8c58d639 --- /dev/null +++ b/QAs/egolife_A1_003-mcq.json @@ -0,0 +1,944 @@ +[ + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_6_T3", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which person is closest to me?", + "answer": "The boy in white is on your left.", + "question_chinese": "哪一个人离我最近?", + "answer_chinese": "白色衣服的男孩，在你的左方。", + "evidence": [ + { + "description": "It can be seen that there are 5 other people in the room besides me now.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I have come into the bedroom.", + "timestamps": { + "start": "00:02:07", + "end": "00:02:08" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The boy in orange is closest.", + "B": "The boy in white is on your left.", + "C": "The girl in blue is closest.", + "D": "The boy in white is closest, but he is on your right side instead of your left.", + "E": "The person wearing white is closest, standing on your right rather than the boy in white." + }, + "answer_label": "B", + "answer_text": "The boy in white is on your left.", + "q_id": "egolife_A1_003_6_T3", + "correct_answer": "The boy in white is on your left.", + "original_options": { + "A": "The boy in orange is closest.", + "B": "The boy in white is on your left.", + "C": "The girl in blue is closest.", + "D": "The boy in white is closest, but he is on your right side instead of your left.", + "E": "The person wearing white is closest, standing on your right rather than the boy in white." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_6_T4", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which person is closest to me?", + "answer": "The boy in orange is on your left.", + "question_chinese": "哪一个人离我最近?", + "answer_chinese": "橙色衣服的男孩，在你的左方。", + "evidence": [ + { + "description": "It can be seen that there are 5 other people in the room besides me now.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I have come into the bedroom.", + "timestamps": { + "start": "00:02:07", + "end": "00:02:08" + } + } + ], + "question_timestamps": "00:02:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The boy in white is closest.", + "B": "The girl in blue is closest.", + "C": "The boy in orange is on your left.", + "D": "The boy in orange is closest, but he is on your right side rather than your left.", + "E": "The closest person is the one wearing white on your right, not the boy in orange." + }, + "answer_label": "C", + "answer_text": "The boy in orange is on your left.", + "q_id": "egolife_A1_003_6_T4", + "correct_answer": "The boy in orange is on your left.", + "original_options": { + "A": "The boy in white is closest.", + "B": "The girl in blue is closest.", + "C": "The boy in orange is on your left.", + "D": "The boy in orange is closest, but he is on your right side rather than your left.", + "E": "The closest person is the one wearing white on your right, not the boy in orange." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_7", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the washing machine to do laundry now, how should I get there?", + "answer": "You should turn around and walk about five steps to get near the table, then walk about four steps toward the left front to enter the storage room. Finally, turn left and walk one or two steps, and you will see the washing machine.", + "question_chinese": "如果我现在想要去洗衣机洗衣服，应该怎么走?", + "answer_chinese": "你应该。转身走5步左右的距离到桌子附近，然后向左前方走4步左右的距离，进入储藏室。最后左转走一两步。就可以看到洗衣机了。", + "evidence": [ + { + "description": "The washing machine is about two steps away from the boxes piled up in the storage room, towards the window.", + "timestamps": { + "start": "00:01:25", + "end": "00:01:29" + } + }, + { + "description": "I turned around, walked about two or three steps, and entered the storage room.", + "timestamps": { + "start": "00:01:25", + "end": "00:01:28" + } + }, + { + "description": "I walked around the table toward the left front for about five or six steps and then entered the bedroom.", + "timestamps": { + "start": "00:02:03", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:09:08", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn around and enter the kitchen.", + "B": "You should turn around and walk about five steps to get near the table, then walk about four steps toward the left front to enter the storage room. Finally, turn left and walk one or two steps, and you will see the washing machine.", + "C": "Go forward into the hallway.", + "D": "Turn around, walk about five steps to the door, then head right into the hallway to look for it.", + "E": "Turn around, walk about five steps toward the table, then go right-front and turn right to find the machine." + }, + "answer_label": "B", + "answer_text": "You should turn around and walk about five steps to get near the table, then walk about four steps toward the left front to enter the storage room. Finally, turn left and walk one or two steps, and you will see the washing machine.", + "q_id": "egolife_A1_003_7", + "correct_answer": "You should turn around and walk about five steps to get near the table, then walk about four steps toward the left front to enter the storage room. Finally, turn left and walk one or two steps, and you will see the washing machine.", + "original_options": { + "A": "Turn around and enter the kitchen.", + "B": "You should turn around and walk about five steps to get near the table, then walk about four steps toward the left front to enter the storage room. Finally, turn left and walk one or two steps, and you will see the washing machine.", + "C": "Go forward into the hallway.", + "D": "Turn around, walk about five steps to the door, then head right into the hallway to look for it.", + "E": "Turn around, walk about five steps toward the table, then go right-front and turn right to find the machine." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_8", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the washing machine to do laundry now, how should I get there?", + "answer": "You should turn around and walk four steps to enter the storage room, then turn left and walk one or two steps, and you will be able to see the washing machine.", + "question_chinese": "如果我现在想要去洗衣机洗衣服，应该怎么走?", + "answer_chinese": "你应当转身走4步的距离进入储藏室，随后左转走一两步的距离便能看到洗衣机。", + "evidence": [ + { + "description": "The washing machine is about two steps away from the boxes piled up in the storage room, towards the window.", + "timestamps": { + "start": "00:01:25", + "end": "00:01:29" + } + }, + { + "description": "I turned around, walked about two or three steps, and entered the storage room.", + "timestamps": { + "start": "00:01:25", + "end": "00:01:28" + } + }, + { + "description": "I walked around the table toward the left front for about five or six steps and then entered the bedroom.", + "timestamps": { + "start": "00:02:03", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:11:43", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Walk straight into the living room.", + "B": "Turn left into the hallway.", + "C": "You should turn around and walk four steps to enter the storage room, then turn left and walk one or two steps, and you will be able to see the washing machine.", + "D": "Turn around and walk about four steps, but turn right afterward and continue a few steps to find it.", + "E": "Walk forward into another room first, then turn left and take several steps before you see the machine." + }, + "answer_label": "C", + "answer_text": "You should turn around and walk four steps to enter the storage room, then turn left and walk one or two steps, and you will be able to see the washing machine.", + "q_id": "egolife_A1_003_8", + "correct_answer": "You should turn around and walk four steps to enter the storage room, then turn left and walk one or two steps, and you will be able to see the washing machine.", + "original_options": { + "A": "Walk straight into the living room.", + "B": "Turn left into the hallway.", + "C": "You should turn around and walk four steps to enter the storage room, then turn left and walk one or two steps, and you will be able to see the washing machine.", + "D": "Turn around and walk about four steps, but turn right afterward and continue a few steps to find it.", + "E": "Walk forward into another room first, then turn left and take several steps before you see the machine." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_7_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bouquets did I see on the table in total?", + "answer": "Six, in front of you.", + "question_chinese": "我看到桌子上一共有几束花束？", + "answer_chinese": "6，在你的前方。", + "evidence": [ + { + "description": "There are no bouquets of flowers on the table.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:25" + } + }, + { + "description": "We took out a total of 6 bouquets of flowers from the box.", + "timestamps": { + "start": "00:01:47", + "end": "00:04:25" + } + } + ], + "question_timestamps": "00:07:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five bouquets.", + "B": "Six, in front of you.", + "C": "Seven bouquets.", + "D": "Four bouquets on the table in total, fewer than the six you later counted in front of you.", + "E": "Eight bouquets on the table in total, spread across the tabletop rather than just in front of you." + }, + "answer_label": "B", + "answer_text": "Six, in front of you.", + "q_id": "egolife_A1_003_7_T2", + "correct_answer": "Six, in front of you.", + "original_options": { + "A": "Five bouquets.", + "B": "Six, in front of you.", + "C": "Seven bouquets.", + "D": "Four bouquets on the table in total, fewer than the six you later counted in front of you.", + "E": "Eight bouquets on the table in total, spread across the tabletop rather than just in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_7_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bouquets did I see on the table in total?", + "answer": "Six, to your back-right.", + "question_chinese": "我看到桌子上一共有几束花束？", + "answer_chinese": "6，在你的右后方。", + "evidence": [ + { + "description": "There are no bouquets of flowers on the table.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:25" + } + }, + { + "description": "We took out a total of 6 bouquets of flowers from the box.", + "timestamps": { + "start": "00:01:47", + "end": "00:04:25" + } + } + ], + "question_timestamps": "00:09:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Six, to your back-right.", + "B": "Five, to your back-right.", + "C": "Seven, to your back-right.", + "D": "Four bouquets total, all sitting together on the table to your back-right.", + "E": "Eight bouquets in total, spread across the table area to your back-right." + }, + "answer_label": "A", + "answer_text": "Six, to your back-right.", + "q_id": "egolife_A1_003_7_T1", + "correct_answer": "Six, to your back-right.", + "original_options": { + "A": "Six, to your back-right.", + "B": "Five, to your back-right.", + "C": "Seven, to your back-right.", + "D": "Four bouquets total, all sitting together on the table to your back-right.", + "E": "Eight bouquets in total, spread across the table area to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_4_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the envelope with a heart drawn on it up there with my hand?", + "answer": "No—you can’t. It’s to your back-right.", + "question_chinese": "我能用手拿到上面画有爱心的信封吗？", + "answer_chinese": "不可以，它在你的右后方。", + "evidence": [ + { + "description": "Letter.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:10" + } + } + ], + "question_timestamps": "00:09:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t. It’s to your back-right.", + "B": "Yes—you can. It’s to your back-right." + }, + "answer_label": "A", + "answer_text": "No—you can’t. It’s to your back-right.", + "q_id": "egolife_A1_003_4_T2", + "correct_answer": "No—you can’t. It’s to your back-right.", + "original_options": { + "A": "No—you can’t. It’s to your back-right.", + "B": "Yes—you can. It’s to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_4_T1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the envelope with a heart drawn on it up there with my hand?", + "answer": "Yes—you can. He’s right by you, at the edge of the table next to you.", + "question_chinese": "我能用手拿到上面画有爱心的信封吗？", + "answer_chinese": "可以，他就在你身旁的桌子边缘", + "evidence": [ + { + "description": "I took out the envelope with a heart pattern printed on it from the box in the bedroom.", + "timestamps": { + "start": "00:09:08", + "end": "00:09:17" + } + }, + { + "description": "I placed the envelope with a heart drawn on it on the corner of the table.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:33" + } + } + ], + "question_timestamps": "00:11:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you can’t. It’s a bit too far from you, not within arm’s reach.", + "B": "Yes—you can. He’s right by you, at the edge of the table next to you." + }, + "answer_label": "B", + "answer_text": "Yes—you can. He’s right by you, at the edge of the table next to you.", + "q_id": "egolife_A1_003_4_T1", + "correct_answer": "Yes—you can. He’s right by you, at the edge of the table next to you.", + "original_options": { + "A": "No—you can’t. It’s a bit too far from you, not within arm’s reach.", + "B": "Yes—you can. He’s right by you, at the edge of the table next to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_5_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the envelope with a heart drawn on it using my hand?", + "answer": "Yes, it's right in front of you now.", + "question_chinese": "我能用手拿到上面画有爱心的信封吗？", + "answer_chinese": "可以，它现在就在你的正前方", + "evidence": [ + { + "description": "I took out the envelope with a heart pattern printed on it from the box in the bedroom.", + "timestamps": { + "start": "00:09:08", + "end": "00:09:17" + } + }, + { + "description": "I placed the envelope with a heart drawn on it on the corner of the table.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:33" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t—it’s too far away from you right now.", + "B": "Yes, it's right in front of you now." + }, + "answer_label": "B", + "answer_text": "Yes, it's right in front of you now.", + "q_id": "egolife_A1_003_5_T2", + "correct_answer": "Yes, it's right in front of you now.", + "original_options": { + "A": "No, you can’t—it’s too far away from you right now.", + "B": "Yes, it's right in front of you now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_5_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the envelope with a heart drawn on it up there with my hand?", + "answer": "Yes—you can. It’s прямо in front of you right now.", + "question_chinese": "我能用手拿到上面画有爱心的信封吗？", + "answer_chinese": "可以，它现在就在你的正前方", + "evidence": [ + { + "description": "I took out the envelope with a heart pattern printed on it from the box in the bedroom.", + "timestamps": { + "start": "00:09:08", + "end": "00:09:17" + } + }, + { + "description": "I placed the envelope with a heart drawn on it on the corner of the table.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:33" + } + } + ], + "question_timestamps": "00:12:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. It’s прямо in front of you right now.", + "B": "No—you can’t. It’s прямо in front of you right now." + }, + "answer_label": "A", + "answer_text": "Yes—you can. It’s прямо in front of you right now.", + "q_id": "egolife_A1_003_5_T1", + "correct_answer": "Yes—you can. It’s прямо in front of you right now.", + "original_options": { + "A": "Yes—you can. It’s прямо in front of you right now.", + "B": "No—you can’t. It’s прямо in front of you right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003-mcq_fixed_9", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the door to the bedroom closer to me, or the door to the stairwell?", + "answer": "The distance from you to the two doors is roughly the same now.", + "question_chinese": "是通往卧室的门离我更近，还是通往楼梯间的门离我更近？", + "answer_chinese": "现在你到两扇门的距离差不多。", + "evidence": [ + { + "description": "The door to the bedroom is in front of my left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "The door to the stairwell is directly in front of me.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:27" + } + } + ], + "question_timestamps": "00:02:53", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The distance from you to the two doors is roughly the same now.", + "B": "The door to the bedroom is closer to you than the door to the stairwell." + }, + "answer_label": "A", + "answer_text": "The distance from you to the two doors is roughly the same now.", + "q_id": "egolife_A1_003-mcq_fixed_9", + "correct_answer": "The distance from you to the two doors is roughly the same now.", + "original_options": { + "A": "The distance from you to the two doors is roughly the same now.", + "B": "The door to the bedroom is closer to you than the door to the stairwell." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003-mcq_fixed_10", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the door to the bedroom closer to me, or the door to the stairwell?", + "answer": "Now the door to the stairwell is closer to you.", + "question_chinese": "是通往卧室的门离我更近，还是通往楼梯间的门离我更近？", + "answer_chinese": "现在通往楼梯间的门离你更近。", + "evidence": [ + { + "description": "The door to the bedroom is in front of my left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "The door to the stairwell is directly in front of me.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:27" + } + } + ], + "question_timestamps": "00:02:54", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Now the door to the bedroom is closer to you.", + "B": "Now the door to the stairwell is closer to you." + }, + "answer_label": "B", + "answer_text": "Now the door to the stairwell is closer to you.", + "q_id": "egolife_A1_003-mcq_fixed_10", + "correct_answer": "Now the door to the stairwell is closer to you.", + "original_options": { + "A": "Now the door to the bedroom is closer to you.", + "B": "Now the door to the stairwell is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_5", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bouquets of flowers are there on the table?", + "answer": "Zero.", + "question_chinese": "桌子上一共有几束花束？", + "answer_chinese": "0", + "evidence": [ + { + "description": "There are no bouquets of flowers on the table.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:01:39", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Zero.", + "C": "Two.", + "D": "Three bouquets of flowers are visible on the table surface near the middle of the frame right now.", + "E": "Four bouquets of flowers are arranged on the table in front of you near the edge closest to you." + }, + "answer_label": "B", + "answer_text": "Zero.", + "q_id": "egolife_A1_003_5", + "correct_answer": "Zero.", + "original_options": { + "A": "One.", + "B": "Zero.", + "C": "Two.", + "D": "Three bouquets of flowers are visible on the table surface near the middle of the frame right now.", + "E": "Four bouquets of flowers are arranged on the table in front of you near the edge closest to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_6", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bouquets of flowers are there on the table?", + "answer": "Six.", + "question_chinese": "桌子上一共有几束花束？", + "answer_chinese": "6", + "evidence": [ + { + "description": "There are no bouquets of flowers on the table.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:25" + } + }, + { + "description": "We took out a total of 6 bouquets of flowers from the box.", + "timestamps": { + "start": "00:01:47", + "end": "00:04:25" + } + } + ], + "question_timestamps": "00:09:16", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "Five.", + "C": "Seven.", + "D": "There are four bouquets of flowers arranged neatly together on the table.", + "E": "There are eight bouquets of flowers spread across the table in a few small clusters." + }, + "answer_label": "A", + "answer_text": "Six.", + "q_id": "egolife_A1_003_6", + "correct_answer": "Six.", + "original_options": { + "A": "Six.", + "B": "Five.", + "C": "Seven.", + "D": "There are four bouquets of flowers arranged neatly together on the table.", + "E": "There are eight bouquets of flowers spread across the table in a few small clusters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the whiteboard from me?", + "answer": "The whiteboard is on your right.", + "question_chinese": "白板在我的什么方向？", + "answer_chinese": "白板在你的右侧", + "evidence": [ + { + "description": "The whiteboard is directly in front of me.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I stood up, turned left, and walked into the storage room.", + "timestamps": { + "start": "00:01:27", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The whiteboard is on your right.", + "B": "On your left.", + "C": "In front of you.", + "D": "The whiteboard is directly behind you as you face forward in the room.", + "E": "The whiteboard is above you, positioned overhead rather than to either side." + }, + "answer_label": "A", + "answer_text": "The whiteboard is on your right.", + "q_id": "egolife_A1_003_1", + "correct_answer": "The whiteboard is on your right.", + "original_options": { + "A": "The whiteboard is on your right.", + "B": "On your left.", + "C": "In front of you.", + "D": "The whiteboard is directly behind you as you face forward in the room.", + "E": "The whiteboard is above you, positioned overhead rather than to either side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_003", + "video_length": "00:12:00", + "qid": "egolife_A1_003_2", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the whiteboard from me?", + "answer": "The whiteboard is behind you.", + "question_chinese": "白板在我的什么方向？", + "answer_chinese": "白板在你的后方", + "evidence": [ + { + "description": "The whiteboard is directly in front of me.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I stood up, turned left, and walked into the storage room.", + "timestamps": { + "start": "00:01:27", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:03:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The whiteboard is in front of you.", + "B": "The whiteboard is to your left.", + "C": "The whiteboard is to your right, so you would look toward your right side to see it.", + "D": "The whiteboard is on the far wall to your left, visible if you turn your head slightly left.", + "E": "The whiteboard is behind you." + }, + "answer_label": "E", + "answer_text": "The whiteboard is behind you.", + "q_id": "egolife_A1_003_2", + "correct_answer": "The whiteboard is behind you.", + "original_options": { + "A": "The whiteboard is in front of you.", + "B": "The whiteboard is to your left.", + "C": "The whiteboard is to your right, so you would look toward your right side to see it.", + "D": "The whiteboard is on the far wall to your left, visible if you turn your head slightly left.", + "E": "The whiteboard is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 18 + ] + } +] diff --git a/QAs/egolife_A1_005-mcq.json b/QAs/egolife_A1_005-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b59ff2c8f02a788ee0e5a77bd11f080d84bae402 --- /dev/null +++ b/QAs/egolife_A1_005-mcq.json @@ -0,0 +1,734 @@ +[ + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_9", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many data cable bundles are there on the storage rack?", + "answer": "Two.", + "question_chinese": "储物架上有几个数据线包？", + "answer_chinese": "2", + "evidence": [ + { + "description": "There are two black data cable bags on the storage shelf.", + "timestamps": { + "start": "00:02:06", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:05:26", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "There are three data cable bundles sitting on the storage rack right now.", + "D": "Two.", + "E": "You can see only one data cable bundle on the rack, with no second bundle present." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "egolife_A1_005_9", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "There are three data cable bundles sitting on the storage rack right now.", + "D": "Two.", + "E": "You can see only one data cable bundle on the rack, with no second bundle present." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_10", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many data cable bundles are on the storage rack?", + "answer": "Zero.", + "question_chinese": "储物架上有几个数据线包？", + "answer_chinese": "0", + "evidence": [ + { + "description": "There are two black data cable bags on the storage shelf.", + "timestamps": { + "start": "00:02:06", + "end": "00:02:20" + } + }, + { + "description": "I took the data cable bag with an orange label attached from the storage shelf.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:03" + } + }, + { + "description": "I took the data cable bag with a green label attached from the storage shelf.", + "timestamps": { + "start": "00:07:25", + "end": "00:07:26" + } + } + ], + "question_timestamps": "00:11:31", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "There are two data cable bundles visible on the storage rack right now.", + "E": "You can spot three data cable bundles grouped together on the rack." + }, + "answer_label": "A", + "answer_text": "Zero.", + "q_id": "egolife_A1_005_10", + "correct_answer": "Zero.", + "original_options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "There are two data cable bundles visible on the storage rack right now.", + "E": "You can spot three data cable bundles grouped together on the rack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many laptops are on the bed in the bedroom?", + "answer": "One.", + "question_chinese": "卧室的床上有几台笔记本电脑？", + "answer_chinese": "1", + "evidence": [ + { + "description": "There is a laptop on the bed in the bedroom.", + "timestamps": { + "start": "00:06:54", + "end": "00:06:57" + } + } + ], + "question_timestamps": "00:09:02", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "There are two laptops resting on the bed in the bedroom right now.", + "D": "You can see three laptops on the bed, spaced apart rather than a single one.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_005_5", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "There are two laptops resting on the bed in the bedroom right now.", + "D": "You can see three laptops on the bed, spaced apart rather than a single one.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many laptops are on the bed in the bedroom?", + "answer": "Zero.", + "question_chinese": "卧室的床上有几台笔记本电脑？", + "answer_chinese": "0", + "evidence": [ + { + "description": "There is a laptop on the bed in the bedroom.", + "timestamps": { + "start": "00:06:54", + "end": "00:06:57" + } + } + ], + "question_timestamps": "00:11:51", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "There is one laptop on the bed in the bedroom at the moment.", + "D": "You can count two laptops placed on the bed, not an empty surface.", + "E": "Zero." + }, + "answer_label": "E", + "answer_text": "Zero.", + "q_id": "egolife_A1_005_6", + "correct_answer": "Zero.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "There is one laptop on the bed in the bedroom at the moment.", + "D": "You can count two laptops placed on the bed, not an empty surface.", + "E": "Zero." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_11", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the living room window from me?", + "answer": "The window of the living room is behind you.", + "question_chinese": "客厅的窗户在我的什么方向？", + "answer_chinese": "客厅的窗户在你的后方", + "evidence": [ + { + "description": "The window of the living room is on my right.", + "timestamps": { + "start": "00:00:29", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:06:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window of the living room is to your right when you look ahead.", + "D": "The window of the living room is behind you.", + "E": "The window of the living room is above you rather than behind your position." + }, + "answer_label": "D", + "answer_text": "The window of the living room is behind you.", + "q_id": "egolife_A1_005_11", + "correct_answer": "The window of the living room is behind you.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window of the living room is to your right when you look ahead.", + "D": "The window of the living room is behind you.", + "E": "The window of the living room is above you rather than behind your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_12", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the living room window from me?", + "answer": "The window of the living room is behind you.", + "question_chinese": "客厅的窗户在我的什么方向？", + "answer_chinese": "客厅的窗户在你的右侧", + "evidence": [ + { + "description": "The window of the living room is on my right.", + "timestamps": { + "start": "00:00:29", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:11:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "The window of the living room is behind you.", + "D": "The living room window is on your right side, beside you as you face forward.", + "E": "The living room window is ahead of you, slightly off to your left as you look into the room." + }, + "answer_label": "C", + "answer_text": "The window of the living room is behind you.", + "q_id": "egolife_A1_005_12", + "correct_answer": "The window of the living room is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "The window of the living room is behind you.", + "D": "The living room window is on your right side, beside you as you face forward.", + "E": "The living room window is ahead of you, slightly off to your left as you look into the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_7", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the bottle of mineral water with my hand now?", + "answer": "No, it is about five or six steps away from you right now.", + "question_chinese": "我现在能用手够到矿泉水瓶吗？", + "answer_chinese": "不行，它现在离你有五六步的距离。", + "evidence": [ + { + "description": "The mineral water bottle is at the corner of the table's edge.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:08:34", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, it is about five or six steps away from you right now.", + "B": "Yes, you can reach it; it's just an arm's length away from you right now." + }, + "answer_label": "A", + "answer_text": "No, it is about five or six steps away from you right now.", + "q_id": "egolife_A1_005_7", + "correct_answer": "No, it is about five or six steps away from you right now.", + "original_options": { + "A": "No, it is about five or six steps away from you right now.", + "B": "Yes, you can reach it; it's just an arm's length away from you right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_8", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the bottle of mineral water with my hand now?", + "answer": "Yes, it's right on the edge of the table.", + "question_chinese": "我现在能用手够到矿泉水瓶吗？", + "answer_chinese": "可以，它就在桌子的边缘。", + "evidence": [ + { + "description": "The mineral water bottle is at the corner of the table's edge.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:10:41", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you can’t reach it right now; it’s a few steps away from you, not within arm’s reach.", + "B": "Yes, it's right on the edge of the table." + }, + "answer_label": "B", + "answer_text": "Yes, it's right on the edge of the table.", + "q_id": "egolife_A1_005_8", + "correct_answer": "Yes, it's right on the edge of the table.", + "original_options": { + "A": "No, you can’t reach it right now; it’s a few steps away from you, not within arm’s reach.", + "B": "Yes, it's right on the edge of the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 21 + ] + }, + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_3", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How should I go to get a pot of pink flowers?", + "answer": "You should turn around and walk six steps, then turn left and walk three steps to reach the water fountain beside the window. The flowers are right in front of you.", + "question_chinese": "我现在要去拿一盆粉色的花，我应该怎么走？", + "answer_chinese": "你应当掉头走6步的距离，然后左拐走三步的距离，到达窗户旁的水池边，花就在你的正前方", + "evidence": [ + { + "description": "Close to the whiteboard, there is a pot of pink flowers.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + }, + { + "description": "I took this pot of pink flowers and placed it beside the water tank near the window.", + "timestamps": { + "start": "00:03:39", + "end": "00:03:49" + } + }, + { + "description": "The bedroom is about five steps to the left of me.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right, walk six steps.", + "B": "Turn around, walk five steps.", + "C": "You should turn around and walk six steps, then turn left and walk three steps to reach the water fountain beside the window. The flowers are right in front of you.", + "D": "Turn around and walk six steps, then turn right and walk three steps to the fountain by the window.", + "E": "Turn around and walk six steps, then keep going straight three steps to the water fountain beside the window." + }, + "answer_label": "C", + "answer_text": "You should turn around and walk six steps, then turn left and walk three steps to reach the water fountain beside the window. The flowers are right in front of you.", + "q_id": "egolife_A1_005_3", + "correct_answer": "You should turn around and walk six steps, then turn left and walk three steps to reach the water fountain beside the window. The flowers are right in front of you.", + "original_options": { + "A": "Turn right, walk six steps.", + "B": "Turn around, walk five steps.", + "C": "You should turn around and walk six steps, then turn left and walk three steps to reach the water fountain beside the window. The flowers are right in front of you.", + "D": "Turn around and walk six steps, then turn right and walk three steps to the fountain by the window.", + "E": "Turn around and walk six steps, then keep going straight three steps to the water fountain beside the window." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_4", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How should I get to where I can pick up a pot of pink flowers?", + "answer": "You should stand up and turn right. Walk two steps to reach the pool, and the flower will be right in front of you.", + "question_chinese": "我现在要去拿一盆粉色的花，我应该怎么走？", + "answer_chinese": "你应当起身向右转。走两步的距离到达水池旁，花就在你的正前方。", + "evidence": [ + { + "description": "Close to the whiteboard, there is a pot of pink flowers.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + }, + { + "description": "I took this pot of pink flowers and placed it beside the water tank near the window.", + "timestamps": { + "start": "00:03:39", + "end": "00:03:49" + } + }, + { + "description": "The bedroom is about five steps to the left of me.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:11:40", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Stand up and turn left.", + "B": "Walk three steps to the pool.", + "C": "Stand up and turn right, then walk two steps to the pool and the flower will be behind you.", + "D": "You should stand up and turn right. Walk two steps to reach the pool, and the flower will be right in front of you.", + "E": "Stand up and turn around, walk two steps to the pool, and the flower will be off to your left." + }, + "answer_label": "D", + "answer_text": "You should stand up and turn right. Walk two steps to reach the pool, and the flower will be right in front of you.", + "q_id": "egolife_A1_005_4", + "correct_answer": "You should stand up and turn right. Walk two steps to reach the pool, and the flower will be right in front of you.", + "original_options": { + "A": "Stand up and turn left.", + "B": "Walk three steps to the pool.", + "C": "Stand up and turn right, then walk two steps to the pool and the flower will be behind you.", + "D": "You should stand up and turn right. Walk two steps to reach the pool, and the flower will be right in front of you.", + "E": "Stand up and turn around, walk two steps to the pool, and the flower will be off to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the whiteboard from me?", + "answer": "The whiteboard is behind you on the left.", + "question_chinese": "白板在我的什么方向？", + "answer_chinese": "白板在你的左后方", + "evidence": [ + { + "description": "The whiteboard is right in front of me", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:03:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you on the left.", + "B": "The whiteboard is behind you on the left.", + "C": "Behind you on the right.", + "D": "The whiteboard is directly in front of you, slightly to your right as you face forward.", + "E": "The whiteboard is to your left side in front of you, not behind you at all." + }, + "answer_label": "B", + "answer_text": "The whiteboard is behind you on the left.", + "q_id": "egolife_A1_005_1", + "correct_answer": "The whiteboard is behind you on the left.", + "original_options": { + "A": "In front of you on the left.", + "B": "The whiteboard is behind you on the left.", + "C": "Behind you on the right.", + "D": "The whiteboard is directly in front of you, slightly to your right as you face forward.", + "E": "The whiteboard is to your left side in front of you, not behind you at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_005", + "video_length": "00:12:00", + "qid": "egolife_A1_005_2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In what direction is the whiteboard from me?", + "answer": "The whiteboard is on your right rear side.", + "question_chinese": "白板在我的什么方向？", + "answer_chinese": "白板在你的右后方", + "evidence": [ + { + "description": "The whiteboard is right in front of me", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:06:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Directly in front of you.", + "B": "On your left rear side.", + "C": "The whiteboard is on your right front side, ahead of you rather than behind your shoulder.", + "D": "The whiteboard is beside you on the left, close to your hip instead of on your right rear.", + "E": "The whiteboard is on your right rear side." + }, + "answer_label": "E", + "answer_text": "The whiteboard is on your right rear side.", + "q_id": "egolife_A1_005_2", + "correct_answer": "The whiteboard is on your right rear side.", + "original_options": { + "A": "Directly in front of you.", + "B": "On your left rear side.", + "C": "The whiteboard is on your right front side, ahead of you rather than behind your shoulder.", + "D": "The whiteboard is beside you on the left, close to your hip instead of on your right rear.", + "E": "The whiteboard is on your right rear side." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + } +] diff --git a/QAs/egolife_A1_009-mcq.json b/QAs/egolife_A1_009-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ec7ffd7d538ad1159b433afe7d92e59df2d7b071 --- /dev/null +++ b/QAs/egolife_A1_009-mcq.json @@ -0,0 +1,1742 @@ +[ + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009-mcq_fixed_12", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Taking the direction from which sunlight enters the window as the window's front, where am I located relative to the window?", + "answer": "You are directly in front of the window.", + "question_chinese": "从太阳光射进窗户的方向为窗户的正前方，我在窗户的哪个位置？", + "answer_chinese": "你在窗户的正前方", + "evidence": [ + { + "description": "The window is directly in front of me.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are directly in front of the window.", + "B": "You are directly behind the window.", + "C": "You are to the window's left.", + "D": "You are positioned diagonally in front of the window, so the sunlight reaches you from one side.", + "E": "You are on the right side of the window, where the sunlight comes past you rather than straight toward you." + }, + "answer_label": "A", + "answer_text": "You are directly in front of the window.", + "q_id": "egolife_A1_009-mcq_fixed_12", + "correct_answer": "You are directly in front of the window.", + "original_options": { + "A": "You are directly in front of the window.", + "B": "You are directly behind the window.", + "C": "You are to the window's left.", + "D": "You are positioned diagonally in front of the window, so the sunlight reaches you from one side.", + "E": "You are on the right side of the window, where the sunlight comes past you rather than straight toward you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_14", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From the direction of sunlight coming through the window, where am I positioned in relation to the window?", + "answer": "You are on the left side of the window.", + "question_chinese": "从太阳光射进窗户的方向为窗户的正前方，我在窗户的哪个位置？", + "answer_chinese": "你在窗户的左侧", + "evidence": [ + { + "description": "The window is directly in front of me.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:07:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are on the right side.", + "B": "You are directly in front of it.", + "C": "You are positioned behind the window area, with the sunlight coming from in front of you.", + "D": "You are standing to the right of the window, so the sunlight enters from your left side.", + "E": "You are on the left side of the window." + }, + "answer_label": "E", + "answer_text": "You are on the left side of the window.", + "q_id": "egolife_A1_009_14", + "correct_answer": "You are on the left side of the window.", + "original_options": { + "A": "You are on the right side.", + "B": "You are directly in front of it.", + "C": "You are positioned behind the window area, with the sunlight coming from in front of you.", + "D": "You are standing to the right of the window, so the sunlight enters from your left side.", + "E": "You are on the left side of the window." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the white power bank on the storage rack located in relation to me?", + "answer": "The white power bank is on your left.", + "question_chinese": "储物架上的白色充电宝在我的什么位置？", + "answer_chinese": "白色充电宝在你的左侧。", + "evidence": [ + { + "description": "There is a white power bank being charged on top of the storage cabinet.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:07:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your right.", + "B": "The white power bank is on your left.", + "C": "It is in front of you.", + "D": "The white power bank sits behind you on the rack, so you’d need to turn around to see it.", + "E": "The white power bank is to your front-right on the storage rack, slightly ahead of your position." + }, + "answer_label": "B", + "answer_text": "The white power bank is on your left.", + "q_id": "egolife_A1_009_1", + "correct_answer": "The white power bank is on your left.", + "original_options": { + "A": "It is on your right.", + "B": "The white power bank is on your left.", + "C": "It is in front of you.", + "D": "The white power bank sits behind you on the rack, so you’d need to turn around to see it.", + "E": "The white power bank is to your front-right on the storage rack, slightly ahead of your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 17 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the white power bank on the shelf located in relation to me?", + "answer": "The white power bank is behind you.", + "question_chinese": "储物架上的白色充电宝在我的什么位置？", + "answer_chinese": "白色充电宝在你的后方", + "evidence": [ + { + "description": "There is a white power bank being charged on top of the storage cabinet.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:11:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The white power bank is behind you.", + "B": "It is in front of you.", + "C": "It is to your right.", + "D": "The white power bank is on your left side on the shelf, roughly beside where you are standing.", + "E": "The white power bank is below you on the shelf area, closer to your feet than your back." + }, + "answer_label": "A", + "answer_text": "The white power bank is behind you.", + "q_id": "egolife_A1_009_2", + "correct_answer": "The white power bank is behind you.", + "original_options": { + "A": "The white power bank is behind you.", + "B": "It is in front of you.", + "C": "It is to your right.", + "D": "The white power bank is on your left side on the shelf, roughly beside where you are standing.", + "E": "The white power bank is below you on the shelf area, closer to your feet than your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_2_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the stool inside the bedroom relative to me?", + "answer": "The stool is in front of you.", + "question_chinese": "卧室里面的凳子在我的哪里？", + "answer_chinese": "凳子在你的前面", + "evidence": [ + { + "description": "It's in front of you.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stool is behind you.", + "B": "The stool is in front of you.", + "C": "The stool is to your left.", + "D": "The stool is off to your right side in the bedroom, not straight ahead in your path.", + "E": "The stool is positioned behind and slightly to the left of you, so it’s not in your direct view." + }, + "answer_label": "B", + "answer_text": "The stool is in front of you.", + "q_id": "egolife_A1_009_2_T1", + "correct_answer": "The stool is in front of you.", + "original_options": { + "A": "The stool is behind you.", + "B": "The stool is in front of you.", + "C": "The stool is to your left.", + "D": "The stool is off to your right side in the bedroom, not straight ahead in your path.", + "E": "The stool is positioned behind and slightly to the left of you, so it’s not in your direct view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_2_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the stool inside the bedroom relative to me?", + "answer": "It's to your left.", + "question_chinese": "卧室里面的凳子在我的哪里？", + "answer_chinese": "它在你的左边。", + "evidence": [ + { + "description": "It is on your left.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It's to your right.", + "B": "It's in front of you.", + "C": "It's behind you in the bedroom, so you would have to turn around to face it.", + "D": "It's to your left.", + "E": "It's slightly ahead and to your right, not on the left side of your body." + }, + "answer_label": "D", + "answer_text": "It's to your left.", + "q_id": "egolife_A1_009_2_T2", + "correct_answer": "It's to your left.", + "original_options": { + "A": "It's to your right.", + "B": "It's in front of you.", + "C": "It's behind you in the bedroom, so you would have to turn around to face it.", + "D": "It's to your left.", + "E": "It's slightly ahead and to your right, not on the left side of your body." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_3", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the locker to get something, how should I get there?", + "answer": "You should turn right. Walk about seven steps, and you will see the storage cabinet.", + "question_chinese": "如果我想去储物柜拿东西，我应该怎么走？", + "answer_chinese": "你应该向右转。走7步走的距离就能看到储物柜了.", + "evidence": [ + { + "description": "The green storage cabinet is directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "I walked from the living room to the bedroom.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:01:16", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left and walk five steps.", + "B": "Turn right and walk ten steps.", + "C": "Turn around, walk about six steps, and you will reach the storage cabinet straight ahead.", + "D": "Turn left and walk about eight steps; the storage cabinet will appear on the side as you approach.", + "E": "You should turn right. Walk about seven steps, and you will see the storage cabinet." + }, + "answer_label": "E", + "answer_text": "You should turn right. Walk about seven steps, and you will see the storage cabinet.", + "q_id": "egolife_A1_009_3", + "correct_answer": "You should turn right. Walk about seven steps, and you will see the storage cabinet.", + "original_options": { + "A": "Turn left and walk five steps.", + "B": "Turn right and walk ten steps.", + "C": "Turn around, walk about six steps, and you will reach the storage cabinet straight ahead.", + "D": "Turn left and walk about eight steps; the storage cabinet will appear on the side as you approach.", + "E": "You should turn right. Walk about seven steps, and you will see the storage cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_4", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the locker to get something, how should I get there?", + "answer": "You should turn right and walk one or two steps, then turn right again and walk about six steps, and you will see the storage cabinet.", + "question_chinese": "如果我想去储物柜拿东西，我应该怎么走？", + "answer_chinese": "你应该向右转走一两步，然后向右转走6步左右的距离，你就能看到储物柜了", + "evidence": [ + { + "description": "The green storage cabinet is directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "I walked from the living room to the bedroom.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:13" + } + }, + { + "description": "I walked from the bedroom to the kitchen.", + "timestamps": { + "start": "00:01:39", + "end": "00:02:07" + } + } + ], + "question_timestamps": "00:04:25", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, then left again.", + "B": "Turn right, then turn left.", + "C": "Turn left and walk one or two steps, then turn right and walk about five steps to the bookshelf.", + "D": "You should turn right and walk one or two steps, then turn right again and walk about six steps, and you will see the storage cabinet.", + "E": "Turn right and walk one step, then turn left and walk about ten steps until you reach the cabinet." + }, + "answer_label": "D", + "answer_text": "You should turn right and walk one or two steps, then turn right again and walk about six steps, and you will see the storage cabinet.", + "q_id": "egolife_A1_009_4", + "correct_answer": "You should turn right and walk one or two steps, then turn right again and walk about six steps, and you will see the storage cabinet.", + "original_options": { + "A": "Turn left, then left again.", + "B": "Turn right, then turn left.", + "C": "Turn left and walk one or two steps, then turn right and walk about five steps to the bookshelf.", + "D": "You should turn right and walk one or two steps, then turn right again and walk about six steps, and you will see the storage cabinet.", + "E": "Turn right and walk one step, then turn left and walk about ten steps until you reach the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_4_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I grab the laptop on the living-room table?", + "answer": "No. It’s about two to three steps away from you. You went into the bedroom, and the living-room table is outside the bedroom.", + "question_chinese": "我可以拿到客厅桌子上的笔记本吗？", + "answer_chinese": "不可以。它离你大概有两三步的距离，你进入卧室，客厅桌子在卧室外面。", + "evidence": [ + { + "description": "There is a laptop on the edge of the table in the living room.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The window in the living room is about two or three steps away from the table.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. It’s about two to three steps away from you. You went into the bedroom, and the living-room table is outside the bedroom.", + "B": "Yes. It’s within arm’s reach on the living-room table, so you can grab it without taking any steps." + }, + "answer_label": "A", + "answer_text": "No. It’s about two to three steps away from you. You went into the bedroom, and the living-room table is outside the bedroom.", + "q_id": "egolife_A1_009_4_T2", + "correct_answer": "No. It’s about two to three steps away from you. You went into the bedroom, and the living-room table is outside the bedroom.", + "original_options": { + "A": "No. It’s about two to three steps away from you. You went into the bedroom, and the living-room table is outside the bedroom.", + "B": "Yes. It’s within arm’s reach on the living-room table, so you can grab it without taking any steps." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_4_T1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I grab the laptop on the living-room table?", + "answer": "No. It’s about two to three steps away from you.", + "question_chinese": "我可以拿到客厅桌子上的笔记本吗？", + "answer_chinese": "不可以。它离你大概有两三步的距离。", + "evidence": [ + { + "description": "There is a laptop on the edge of the table in the living room.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The window in the living room is about two or three steps away from the table.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. It’s within arm’s reach from where you are.", + "B": "No. It’s about two to three steps away from you." + }, + "answer_label": "B", + "answer_text": "No. It’s about two to three steps away from you.", + "q_id": "egolife_A1_009_4_T1", + "correct_answer": "No. It’s about two to three steps away from you.", + "original_options": { + "A": "Yes. It’s within arm’s reach from where you are.", + "B": "No. It’s about two to three steps away from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_6", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "May I take the notebook from the table in the living room?", + "answer": "Yes, it's right on your right side.", + "question_chinese": "我可以拿到客厅桌子上的笔记本吗？", + "answer_chinese": "可以，它就在你的右手边。", + "evidence": [ + { + "description": "There is a laptop on the edge of the table in the living room.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The window in the living room is about two or three steps away from the table.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:07:08", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, it’s a few steps away from you.", + "B": "Yes, it's right on your right side." + }, + "answer_label": "B", + "answer_text": "Yes, it's right on your right side.", + "q_id": "egolife_A1_009_6", + "correct_answer": "Yes, it's right on your right side.", + "original_options": { + "A": "No, it’s a few steps away from you.", + "B": "Yes, it's right on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_9_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which white power bank in the kitchen am I closer to?", + "answer": "You are closer to the power bank on the table; it’s to your left.", + "question_chinese": "我离厨房的哪一个白色充电宝更近？", + "answer_chinese": "你离桌子上的那个充电宝更近，它在你左边。", + "evidence": [ + { + "description": "There is a white power bank on the table in the kitchen.", + "timestamps": { + "start": "00:02:07", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:02:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The power bank on the shelf.", + "B": "The power bank on the counter.", + "C": "You are closer to the power bank on the shelf; it’s directly in front of you.", + "D": "You are closer to the power bank on the table; it’s behind you on the far side.", + "E": "You are closer to the power bank on the table; it’s to your left." + }, + "answer_label": "E", + "answer_text": "You are closer to the power bank on the table; it’s to your left.", + "q_id": "egolife_A1_009_9_T2", + "correct_answer": "You are closer to the power bank on the table; it’s to your left.", + "original_options": { + "A": "The power bank on the shelf.", + "B": "The power bank on the counter.", + "C": "You are closer to the power bank on the shelf; it’s directly in front of you.", + "D": "You are closer to the power bank on the table; it’s behind you on the far side.", + "E": "You are closer to the power bank on the table; it’s to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_9_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which white power bank in the kitchen am I closer to?", + "answer": "You are closer to the power bank on the shelf; it’s in front of you.", + "question_chinese": "我离厨房的哪一个白色充电宝更近？", + "answer_chinese": "你离架子上的那个充电宝更近，他在你前面", + "evidence": [ + { + "description": "There is a white power bank on the table in the kitchen.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:09" + } + }, + { + "description": "There is another white power bank on the shelf in the kitchen.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:42" + } + } + ], + "question_timestamps": "00:02:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The power bank on the table.", + "B": "The power bank on the shelf behind you.", + "C": "You are closer to the power bank on the table; it’s to your left near the edge.", + "D": "You are closer to the power bank on the shelf; it’s in front of you.", + "E": "You are closer to the power bank on the shelf; it’s to your right rather than in front." + }, + "answer_label": "D", + "answer_text": "You are closer to the power bank on the shelf; it’s in front of you.", + "q_id": "egolife_A1_009_9_T1", + "correct_answer": "You are closer to the power bank on the shelf; it’s in front of you.", + "original_options": { + "A": "The power bank on the table.", + "B": "The power bank on the shelf behind you.", + "C": "You are closer to the power bank on the table; it’s to your left near the edge.", + "D": "You are closer to the power bank on the shelf; it’s in front of you.", + "E": "You are closer to the power bank on the shelf; it’s to your right rather than in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_11", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which white power bank in the kitchen is closer to me?", + "answer": "You are closer to the power bank on the shelf.", + "question_chinese": "我离厨房的哪一个白色充电宝更近？", + "answer_chinese": "你离架子上的那个充电宝更近", + "evidence": [ + { + "description": "There is a white power bank on the table in the kitchen.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:09" + } + }, + { + "description": "There is another white power bank on the shelf in the kitchen.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:42" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You are closer to the table one.", + "B": "You are closer to the floor one.", + "C": "You are closer to the power bank on the shelf.", + "D": "You are closer to the power bank on the kitchen table, not the one on the shelf.", + "E": "You are closer to the power bank on the countertop; the shelf one is farther away." + }, + "answer_label": "C", + "answer_text": "You are closer to the power bank on the shelf.", + "q_id": "egolife_A1_009_11", + "correct_answer": "You are closer to the power bank on the shelf.", + "original_options": { + "A": "You are closer to the table one.", + "B": "You are closer to the floor one.", + "C": "You are closer to the power bank on the shelf.", + "D": "You are closer to the power bank on the kitchen table, not the one on the shelf.", + "E": "You are closer to the power bank on the countertop; the shelf one is farther away." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_12", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which white power bank in the kitchen is closer to me?", + "answer": "You are closer to the power bank on the shelf.", + "question_chinese": "我离厨房的哪一个白色充电宝更近？", + "answer_chinese": "你离桌子上的那个充电宝更近", + "evidence": [ + { + "description": "There is a white power bank on the table in the kitchen.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:09" + } + }, + { + "description": "There is another white power bank on the shelf in the kitchen.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:42" + } + } + ], + "question_timestamps": "00:10:51", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You are closer to the power bank on the shelf.", + "B": "Closer to the table one.", + "C": "Closer to the drawer one.", + "D": "You are closer to the power bank on the countertop rather than the one on the shelf.", + "E": "You are closer to the power bank on the floor; the shelf one is farther from you." + }, + "answer_label": "A", + "answer_text": "You are closer to the power bank on the shelf.", + "q_id": "egolife_A1_009_12", + "correct_answer": "You are closer to the power bank on the shelf.", + "original_options": { + "A": "You are closer to the power bank on the shelf.", + "B": "Closer to the table one.", + "C": "Closer to the drawer one.", + "D": "You are closer to the power bank on the countertop rather than the one on the shelf.", + "E": "You are closer to the power bank on the floor; the shelf one is farther from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_5_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the green plant on the table relative to me?", + "answer": "He is in front of you.", + "question_chinese": "桌子上的绿植在我的哪里？", + "answer_chinese": "他在你的前面", + "evidence": [ + { + "description": "It is in front of you.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:46" + } + } + ], + "question_timestamps": "00:03:48", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "He is behind you. The green plant is on the table, and the table is behind you.", + "B": "He is in front of you." + }, + "answer_label": "B", + "answer_text": "He is in front of you.", + "q_id": "egolife_A1_009_5_T2", + "correct_answer": "He is in front of you.", + "original_options": { + "A": "He is behind you. The green plant is on the table, and the table is behind you.", + "B": "He is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_5_T1", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the green plant on the table relative to me?", + "answer": "He is behind you. The green plant is on the table, and the table is behind you.", + "question_chinese": "桌子上的绿植在我的哪里？", + "answer_chinese": "在你的后面，绿植在桌子上，桌子在我的后面。", + "evidence": [ + { + "description": "Behind you", + "timestamps": { + "start": "00:03:53", + "end": "00:03:59" + } + } + ], + "question_timestamps": "00:04:01", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "He is behind you. The green plant is on the table, and the table is behind you.", + "B": "He is in front of you." + }, + "answer_label": "A", + "answer_text": "He is behind you. The green plant is on the table, and the table is behind you.", + "q_id": "egolife_A1_009_5_T1", + "correct_answer": "He is behind you. The green plant is on the table, and the table is behind you.", + "original_options": { + "A": "He is behind you. The green plant is on the table, and the table is behind you.", + "B": "He is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_13_T2", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the table relative to me?", + "answer": "The table is to your left.", + "question_chinese": "桌子在我的哪里？", + "answer_chinese": "桌子在我的左边", + "evidence": [ + { + "description": "The table is on my left.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:51" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The table is to your left.", + "B": "The table is to your right.", + "C": "The table is in front of you.", + "D": "The table is directly behind you, not on your left side at all.", + "E": "The table sits ahead of you and slightly to the right as you face forward." + }, + "answer_label": "A", + "answer_text": "The table is to your left.", + "q_id": "egolife_A1_009_13_T2", + "correct_answer": "The table is to your left.", + "original_options": { + "A": "The table is to your left.", + "B": "The table is to your right.", + "C": "The table is in front of you.", + "D": "The table is directly behind you, not on your left side at all.", + "E": "The table sits ahead of you and slightly to the right as you face forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_13_T1", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the table relative to me?", + "answer": "The table is to your front-right.", + "question_chinese": "桌子在我的哪里？", + "answer_chinese": "桌子在你的右前方。", + "evidence": [ + { + "description": "The table is on your right front.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:04" + } + } + ], + "question_timestamps": "00:03:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The table is to your front-left.", + "B": "The table is to your back-right.", + "C": "The table is directly to your left side rather than in the front-right area.", + "D": "The table is to your front-right.", + "E": "The table is behind you on the left, not ahead of you to the right." + }, + "answer_label": "D", + "answer_text": "The table is to your front-right.", + "q_id": "egolife_A1_009_13_T1", + "correct_answer": "The table is to your front-right.", + "original_options": { + "A": "The table is to your front-left.", + "B": "The table is to your back-right.", + "C": "The table is directly to your left side rather than in the front-right area.", + "D": "The table is to your front-right.", + "E": "The table is behind you on the left, not ahead of you to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_10_T2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where are the stairs relative to me?", + "answer": "The stairs are to your right.", + "question_chinese": "楼梯在我的哪里？", + "answer_chinese": "楼梯在你的右边", + "evidence": [ + { + "description": "The stairs are on your right.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:12" + } + } + ], + "question_timestamps": "00:03:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stairs are to your left.", + "B": "The stairs are in front of you.", + "C": "The stairs are to your right.", + "D": "The stairs are positioned behind you rather than off to your right side.", + "E": "The stairs are above you, not located to the right at your level." + }, + "answer_label": "C", + "answer_text": "The stairs are to your right.", + "q_id": "egolife_A1_009_10_T2", + "correct_answer": "The stairs are to your right.", + "original_options": { + "A": "The stairs are to your left.", + "B": "The stairs are in front of you.", + "C": "The stairs are to your right.", + "D": "The stairs are positioned behind you rather than off to your right side.", + "E": "The stairs are above you, not located to the right at your level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_10_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where are the stairs relative to me?", + "answer": "The stairs are to your left.", + "question_chinese": "楼梯在我的哪里？", + "answer_chinese": "楼梯在你的左边", + "evidence": [ + { + "description": "The stairs are on your left.", + "timestamps": { + "start": "00:03:21", + "end": "00:03:24" + } + } + ], + "question_timestamps": "00:03:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The stairs are to your right.", + "B": "The stairs are to your left.", + "C": "The stairs are in front of you.", + "D": "The stairs are behind you rather than located off to your left side.", + "E": "The stairs are above you, not positioned to your left at floor level." + }, + "answer_label": "B", + "answer_text": "The stairs are to your left.", + "q_id": "egolife_A1_009_10_T1", + "correct_answer": "The stairs are to your left.", + "original_options": { + "A": "The stairs are to your right.", + "B": "The stairs are to your left.", + "C": "The stairs are in front of you.", + "D": "The stairs are behind you rather than located off to your left side.", + "E": "The stairs are above you, not positioned to your left at floor level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_9", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many white shelves are currently unfolded in the living room?", + "answer": "Zero shelves.", + "question_chinese": "现在客厅里有几个展开的白色的架子？", + "answer_chinese": "0个架子", + "evidence": [ + { + "description": "There is no unfolded white shelf in the living room now.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:06:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One shelf.", + "B": "Two shelves.", + "C": "Zero shelves.", + "D": "Three shelves are unfolded and visible in the living room right now for you.", + "E": "Four shelves are currently unfolded and visible in the living room at this moment." + }, + "answer_label": "C", + "answer_text": "Zero shelves.", + "q_id": "egolife_A1_009_9", + "correct_answer": "Zero shelves.", + "original_options": { + "A": "One shelf.", + "B": "Two shelves.", + "C": "Zero shelves.", + "D": "Three shelves are unfolded and visible in the living room right now for you.", + "E": "Four shelves are currently unfolded and visible in the living room at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_10", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many open white shelves are there in the living room now?", + "answer": "Two shelves.", + "question_chinese": "现在客厅里有几个展开的白色的架子？", + "answer_chinese": "两个架子", + "evidence": [ + { + "description": "There is no unfolded white shelf in the living room now.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "I unfolded two white shelves with the boy in orange clothes.", + "timestamps": { + "start": "00:07:55", + "end": "00:08:04" + } + } + ], + "question_timestamps": "00:10:54", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One shelf.", + "B": "Three shelves.", + "C": "Four shelves are open in the living room right now for you to see.", + "D": "Two shelves.", + "E": "Five shelves are currently open in the living room at this moment." + }, + "answer_label": "D", + "answer_text": "Two shelves.", + "q_id": "egolife_A1_009_10", + "correct_answer": "Two shelves.", + "original_options": { + "A": "One shelf.", + "B": "Three shelves.", + "C": "Four shelves are open in the living room right now for you to see.", + "D": "Two shelves.", + "E": "Five shelves are currently open in the living room at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 12 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_7", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the whiteboard located in relation to me?", + "answer": "The whiteboard is directly behind you.", + "question_chinese": "白板在我的什么位置？", + "answer_chinese": "白板在你的正后方。", + "evidence": [ + { + "description": "The whiteboard is directly in front of me.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:06:39", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The whiteboard is in front of you.", + "B": "The whiteboard is to your left.", + "C": "The whiteboard is directly behind you.", + "D": "The whiteboard is positioned above you rather than directly behind your back.", + "E": "The whiteboard is slightly in front of you, not directly behind you where you are standing." + }, + "answer_label": "C", + "answer_text": "The whiteboard is directly behind you.", + "q_id": "egolife_A1_009_7", + "correct_answer": "The whiteboard is directly behind you.", + "original_options": { + "A": "The whiteboard is in front of you.", + "B": "The whiteboard is to your left.", + "C": "The whiteboard is directly behind you.", + "D": "The whiteboard is positioned above you rather than directly behind your back.", + "E": "The whiteboard is slightly in front of you, not directly behind you where you are standing." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 16 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_8", + "qid_num": 25, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the whiteboard located in relation to me?", + "answer": "The whiteboard is behind your left side.", + "question_chinese": "白板在我的什么位置？", + "answer_chinese": "白板在你的左后方", + "evidence": [ + { + "description": "The whiteboard is directly in front of me.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:11:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The whiteboard is on your right side.", + "B": "The whiteboard is directly in front of you.", + "C": "The whiteboard is in front of your right side, not behind your left shoulder.", + "D": "The whiteboard is behind you on the right, rather than behind your left side.", + "E": "The whiteboard is behind your left side." + }, + "answer_label": "E", + "answer_text": "The whiteboard is behind your left side.", + "q_id": "egolife_A1_009_8", + "correct_answer": "The whiteboard is behind your left side.", + "original_options": { + "A": "The whiteboard is on your right side.", + "B": "The whiteboard is directly in front of you.", + "C": "The whiteboard is in front of your right side, not behind your left shoulder.", + "D": "The whiteboard is behind you on the right, rather than behind your left side.", + "E": "The whiteboard is behind your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_8_T2", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the window relative to me?", + "answer": "The window is in front of you.", + "question_chinese": "窗户在我的哪里？", + "answer_chinese": "窗户在你的前面", + "evidence": [ + { + "description": "The window is in front of you.", + "timestamps": { + "start": "00:02:07", + "end": "00:02:14" + } + } + ], + "question_timestamps": "00:02:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The window is positioned behind you, so you would need to turn around to face it directly.", + "D": "The window is off to your right side, meaning it isn’t in your direct line of sight.", + "E": "The window is in front of you." + }, + "answer_label": "E", + "answer_text": "The window is in front of you.", + "q_id": "egolife_A1_009_8_T2", + "correct_answer": "The window is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The window is positioned behind you, so you would need to turn around to face it directly.", + "D": "The window is off to your right side, meaning it isn’t in your direct line of sight.", + "E": "The window is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_8_T1", + "qid_num": 27, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the window relative to me?", + "answer": "The window is to your left.", + "question_chinese": "窗户在我的哪里？", + "answer_chinese": "窗户在你的左边", + "evidence": [ + { + "description": "The window is on your left.", + "timestamps": { + "start": "00:02:25", + "end": "00:02:29" + } + } + ], + "question_timestamps": "00:02:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The window is to your left.", + "B": "In front of you.", + "C": "To your right.", + "D": "The window is directly in front of you, so you can see it without turning your head.", + "E": "The window is behind you, requiring you to turn around to face it." + }, + "answer_label": "A", + "answer_text": "The window is to your left.", + "q_id": "egolife_A1_009_8_T1", + "correct_answer": "The window is to your left.", + "original_options": { + "A": "The window is to your left.", + "B": "In front of you.", + "C": "To your right.", + "D": "The window is directly in front of you, so you can see it without turning your head.", + "E": "The window is behind you, requiring you to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 13 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_7_T2", + "qid_num": 28, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What direction is the doorway from me?", + "answer": "The doorway is in front of you.", + "question_chinese": "门口在我的什么方位？", + "answer_chinese": "门口在你的前面", + "evidence": [ + { + "description": "The door is in front of you.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Behind you.", + "B": "To your right.", + "C": "The doorway is in front of you.", + "D": "The doorway is to your left side, so you’d need to look left to see it clearly.", + "E": "The doorway is behind you, meaning you have to turn around to walk through it." + }, + "answer_label": "C", + "answer_text": "The doorway is in front of you.", + "q_id": "egolife_A1_009_7_T2", + "correct_answer": "The doorway is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your right.", + "C": "The doorway is in front of you.", + "D": "The doorway is to your left side, so you’d need to look left to see it clearly.", + "E": "The doorway is behind you, meaning you have to turn around to walk through it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_7_T1", + "qid_num": 29, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What direction is the doorway from me?", + "answer": "The doorway is behind you.", + "question_chinese": "门口在我的什么方位？", + "answer_chinese": "门口在你的后面", + "evidence": [ + { + "description": "The door is behind you.", + "timestamps": { + "start": "00:01:44", + "end": "00:01:49" + } + } + ], + "question_timestamps": "00:01:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The doorway is behind you.", + "D": "The doorway is to your right, so it would be visible by glancing over your right shoulder.", + "E": "The doorway is above you, positioned higher than eye level rather than behind your current position." + }, + "answer_label": "C", + "answer_text": "The doorway is behind you.", + "q_id": "egolife_A1_009_7_T1", + "correct_answer": "The doorway is behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The doorway is behind you.", + "D": "The doorway is to your right, so it would be visible by glancing over your right shoulder.", + "E": "The doorway is above you, positioned higher than eye level rather than behind your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_1_T1", + "qid_num": 30, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the black-and-white checkered board relative to me?", + "answer": "The black-and-white checkered board is to your right.", + "question_chinese": "黑白格子的板子在我的哪里？", + "answer_chinese": "黑白格子的板子在你的右边", + "evidence": [ + { + "description": "The black and white checkered board is on your right.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:53" + } + } + ], + "question_timestamps": "00:01:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "Behind you.", + "C": "The black-and-white checkered board is in front of you, sitting within your forward view rather than at your side.", + "D": "The black-and-white checkered board is to your right.", + "E": "The black-and-white checkered board is above you, placed higher up instead of being beside you." + }, + "answer_label": "D", + "answer_text": "The black-and-white checkered board is to your right.", + "q_id": "egolife_A1_009_1_T1", + "correct_answer": "The black-and-white checkered board is to your right.", + "original_options": { + "A": "To your left.", + "B": "Behind you.", + "C": "The black-and-white checkered board is in front of you, sitting within your forward view rather than at your side.", + "D": "The black-and-white checkered board is to your right.", + "E": "The black-and-white checkered board is above you, placed higher up instead of being beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 20, + 16 + ] + }, + { + "video_uid": "egolife_A1_009", + "video_length": "00:12:00", + "qid": "egolife_A1_009_1_T2", + "qid_num": 31, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the black-and-white checkered board relative to me?", + "answer": "It is in front of you.", + "question_chinese": "黑白格子的板子在我的哪里？", + "answer_chinese": "它在你的前面", + "evidence": [ + { + "description": "It's in front of you.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:05" + } + } + ], + "question_timestamps": "00:03:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "It is off to your right, so you’d have to turn your head right to look at it.", + "D": "It is above you, positioned higher than your eye level rather than straight ahead.", + "E": "It is in front of you." + }, + "answer_label": "E", + "answer_text": "It is in front of you.", + "q_id": "egolife_A1_009_1_T2", + "correct_answer": "It is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "It is off to your right, so you’d have to turn your head right to look at it.", + "D": "It is above you, positioned higher than your eye level rather than straight ahead.", + "E": "It is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 19, + 14 + ] + } +] diff --git a/QAs/egolife_A1_010-mcq.json b/QAs/egolife_A1_010-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fc02168cedeba49659aec516540b79aab8bcb033 --- /dev/null +++ b/QAs/egolife_A1_010-mcq.json @@ -0,0 +1,921 @@ +[ + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010-mcq_fixed_8", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which door is closest to me?", + "answer": "The door leading to the storage room is closer to you.", + "question_chinese": "哪一扇门离我最近？", + "answer_chinese": "通往储物间的门离你更近", + "evidence": [ + { + "description": "The door leading to the storage room is on my right.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + }, + { + "description": "The door leading to the bedroom is on my left.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:05:56", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The kitchen door is closest.", + "B": "The conference room door is closest.", + "C": "The main entrance door behind you is closer, since it’s nearest your current position as you stand.", + "D": "The door leading to the storage room is closer to you.", + "E": "The narrow wooden door by the bookshelf and window is closest, appearing nearest because it sits right in your line of sight." + }, + "answer_label": "D", + "answer_text": "The door leading to the storage room is closer to you.", + "q_id": "egolife_A1_010-mcq_fixed_8", + "correct_answer": "The door leading to the storage room is closer to you.", + "original_options": { + "A": "The kitchen door is closest.", + "B": "The conference room door is closest.", + "C": "The main entrance door behind you is closer, since it’s nearest your current position as you stand.", + "D": "The door leading to the storage room is closer to you.", + "E": "The narrow wooden door by the bookshelf and window is closest, appearing nearest because it sits right in your line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 22 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010-mcq_fixed_9", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which door is closest to me?", + "answer": "The door leading to the bedroom is closer to you.", + "question_chinese": "哪一扇门离我最近？", + "answer_chinese": "通往卧室的门离你更近。", + "evidence": [ + { + "description": "The door leading to the storage room is on my right.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + }, + { + "description": "The door leading to the bedroom is on my left.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:11:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The kitchen door is closest.", + "B": "The storage room door is closest.", + "C": "The door near the living room couch is closest to you, positioned nearer than the other doorways around you.", + "D": "The door leading to the kitchen is closer to you, since it sits nearer in the shared hallway.", + "E": "The door leading to the bedroom is closer to you." + }, + "answer_label": "E", + "answer_text": "The door leading to the bedroom is closer to you.", + "q_id": "egolife_A1_010-mcq_fixed_9", + "correct_answer": "The door leading to the bedroom is closer to you.", + "original_options": { + "A": "The kitchen door is closest.", + "B": "The storage room door is closest.", + "C": "The door near the living room couch is closest to you, positioned nearer than the other doorways around you.", + "D": "The door leading to the kitchen is closer to you, since it sits nearer in the shared hallway.", + "E": "The door leading to the bedroom is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010-mcq_fixed_10", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the chessboard in?", + "answer": "It is behind your left side now.", + "question_chinese": "国际象棋的棋盘在我的什么方向？", + "answer_chinese": "它现在在你的左后方。", + "evidence": [ + { + "description": "There is a chessboard in front of my left side.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:05:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is behind your left side now.", + "B": "It is behind your right side now.", + "C": "It is in front of your left side.", + "D": "It is across the table toward your right, angled slightly away from your line of sight.", + "E": "You’ll find it just past your right shoulder, sitting closer to the near edge of the table." + }, + "answer_label": "A", + "answer_text": "It is behind your left side now.", + "q_id": "egolife_A1_010-mcq_fixed_10", + "correct_answer": "It is behind your left side now.", + "original_options": { + "A": "It is behind your left side now.", + "B": "It is behind your right side now.", + "C": "It is in front of your left side.", + "D": "It is across the table toward your right, angled slightly away from your line of sight.", + "E": "You’ll find it just past your right shoulder, sitting closer to the near edge of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010-mcq_fixed_11", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the chessboard in?", + "answer": "It is in front of your left side now.", + "question_chinese": "国际象棋的棋盘在我的什么方向？", + "answer_chinese": "它现在在你的左前方。", + "evidence": [ + { + "description": "There is a chessboard in front of my left side.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:09:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is in front of your left side now.", + "B": "It is in front of your right side.", + "C": "It is behind your left side now.", + "D": "You should see it diagonally behind you to the right, just past your shoulder and slightly off-center.", + "E": "It is positioned slightly behind your right side, closer to your rear than to your front." + }, + "answer_label": "A", + "answer_text": "It is in front of your left side now.", + "q_id": "egolife_A1_010-mcq_fixed_11", + "correct_answer": "It is in front of your left side now.", + "original_options": { + "A": "It is in front of your left side now.", + "B": "It is in front of your right side.", + "C": "It is behind your left side now.", + "D": "You should see it diagonally behind you to the right, just past your shoulder and slightly off-center.", + "E": "It is positioned slightly behind your right side, closer to your rear than to your front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010-mcq_fixed_6", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the living room table from me?", + "answer": "The table is right in front of you.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "桌子在你的正前方", + "evidence": [ + { + "description": "The table in the living room is right in front of me.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:06:39", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The table is behind you.", + "B": "The table is to your left.", + "C": "The table is slightly behind you, a bit off to one side rather than straight ahead.", + "D": "The table is right in front of you.", + "E": "The table is ahead but off to your right, not directly in front of where you’re facing." + }, + "answer_label": "D", + "answer_text": "The table is right in front of you.", + "q_id": "egolife_A1_010-mcq_fixed_6", + "correct_answer": "The table is right in front of you.", + "original_options": { + "A": "The table is behind you.", + "B": "The table is to your left.", + "C": "The table is slightly behind you, a bit off to one side rather than straight ahead.", + "D": "The table is right in front of you.", + "E": "The table is ahead but off to your right, not directly in front of where you’re facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010-mcq_fixed_7", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the living room table from me?", + "answer": "The table is behind you.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "桌子在你的后方。", + "evidence": [ + { + "description": "The table in the living room is right in front of me.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:11:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The table is in front of you.", + "B": "The table is to your left.", + "C": "The table is behind you.", + "D": "The table is diagonally to your right and a few steps ahead, not directly behind your position.", + "E": "The table is in front of you next to the sofa, within your forward view rather than behind you." + }, + "answer_label": "C", + "answer_text": "The table is behind you.", + "q_id": "egolife_A1_010-mcq_fixed_7", + "correct_answer": "The table is behind you.", + "original_options": { + "A": "The table is in front of you.", + "B": "The table is to your left.", + "C": "The table is behind you.", + "D": "The table is diagonally to your right and a few steps ahead, not directly behind your position.", + "E": "The table is in front of you next to the sofa, within your forward view rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010_5", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I go to get the parts for those white shelves?", + "answer": "You should turn around and walk about four steps, and then you’ll be able to get the parts.", + "question_chinese": "我现在去想拿那些白色架子的零件应该怎么走？", + "answer_chinese": "你应该向后转走4步左右的距离，然后就能拿到零件了。", + "evidence": [ + { + "description": "There is a pile of scattered shelf parts in front of me.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "I can get near the sink by walking straight forward for about five steps now.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "The table is only about one step away from the parts.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:07:07", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left and walk six steps.", + "B": "Turn around and walk two steps.", + "C": "You should turn around and walk about four steps, and then you’ll be able to get the parts.", + "D": "Turn around and walk about five steps, then you should be able to reach the parts.", + "E": "Turn right and walk about three steps, and you’ll be able to get the parts from there." + }, + "answer_label": "C", + "answer_text": "You should turn around and walk about four steps, and then you’ll be able to get the parts.", + "q_id": "egolife_A1_010_5", + "correct_answer": "You should turn around and walk about four steps, and then you’ll be able to get the parts.", + "original_options": { + "A": "Turn left and walk six steps.", + "B": "Turn around and walk two steps.", + "C": "You should turn around and walk about four steps, and then you’ll be able to get the parts.", + "D": "Turn around and walk about five steps, then you should be able to reach the parts.", + "E": "Turn right and walk about three steps, and you’ll be able to get the parts from there." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I go to get the parts for those white shelves?", + "answer": "You can turn left, then take one step, and you’ll be within reach of the parts.", + "question_chinese": "我现在去想拿那些白色架子的零件应该怎么走？", + "answer_chinese": "你可以向左转，然后走一步，走的距离就可以拿到零件了", + "evidence": [ + { + "description": "There is a pile of scattered shelf parts in front of me.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "I can get near the sink by walking straight forward for about five steps now.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "The table is only about one step away from the parts.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:09:34", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, then take two steps.", + "B": "Turn around, then take one step.", + "C": "Turn left, then take three steps, and you’ll be within reach of the parts.", + "D": "Turn right, then take one step, and you’ll be within reach of the parts without moving further.", + "E": "You can turn left, then take one step, and you’ll be within reach of the parts." + }, + "answer_label": "E", + "answer_text": "You can turn left, then take one step, and you’ll be within reach of the parts.", + "q_id": "egolife_A1_010_6", + "correct_answer": "You can turn left, then take one step, and you’ll be within reach of the parts.", + "original_options": { + "A": "Turn left, then take two steps.", + "B": "Turn around, then take one step.", + "C": "Turn left, then take three steps, and you’ll be within reach of the parts.", + "D": "Turn right, then take one step, and you’ll be within reach of the parts without moving further.", + "E": "You can turn left, then take one step, and you’ll be within reach of the parts." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010_3_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What am I doing right now?", + "answer": "You are assembling a shelf right now, and the shelf is in front of you.", + "question_chinese": "我现在在干嘛？", + "answer_chinese": "你现在在组装架子，架子在你的正前方。", + "evidence": [ + { + "description": "You are assembling shelves now.", + "timestamps": { + "start": "00:00:45", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are assembling a shelf right now, and the shelf is in front of you.", + "B": "You are talking with three people right now, and the three people are in front of you." + }, + "answer_label": "A", + "answer_text": "You are assembling a shelf right now, and the shelf is in front of you.", + "q_id": "egolife_A1_010_3_T1", + "correct_answer": "You are assembling a shelf right now, and the shelf is in front of you.", + "original_options": { + "A": "You are assembling a shelf right now, and the shelf is in front of you.", + "B": "You are talking with three people right now, and the three people are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010_3_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "What am I doing right now?", + "answer": "You are talking with three people right now, and the three people are in front of you.", + "question_chinese": "我现在在干嘛？", + "answer_chinese": "你现在在和三个人交谈，三个人在你的正前方。", + "evidence": [ + { + "description": "You are talking to someone else.", + "timestamps": { + "start": "00:06:38", + "end": "00:06:41" + } + } + ], + "question_timestamps": "00:06:44", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are assembling a shelf right now, and the shelf is in front of you.", + "B": "You are talking with three people right now, and the three people are in front of you." + }, + "answer_label": "B", + "answer_text": "You are talking with three people right now, and the three people are in front of you.", + "q_id": "egolife_A1_010_3_T2", + "correct_answer": "You are talking with three people right now, and the three people are in front of you.", + "original_options": { + "A": "You are assembling a shelf right now, and the shelf is in front of you.", + "B": "You are talking with three people right now, and the three people are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010_2_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the parts on the white shelf?", + "answer": "You can reach the parts of the white shelf; the parts are in front of you.", + "question_chinese": "我能碰到白色架子的零件吗？", + "answer_chinese": "你可以碰到白色架子的零件，白色架子的零件在你的正前方。", + "evidence": [ + { + "description": "There is a pile of scattered shelf parts in front of me.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the parts of the white shelf; the parts are in front of you.", + "B": "You cannot reach the parts of the white shelf; the parts are to your left." + }, + "answer_label": "A", + "answer_text": "You can reach the parts of the white shelf; the parts are in front of you.", + "q_id": "egolife_A1_010_2_T2", + "correct_answer": "You can reach the parts of the white shelf; the parts are in front of you.", + "original_options": { + "A": "You can reach the parts of the white shelf; the parts are in front of you.", + "B": "You cannot reach the parts of the white shelf; the parts are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010_3", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the parts of the white rack?", + "answer": "Yes.", + "question_chinese": "我能碰到白色架子的零件吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "There is a pile of scattered shelf parts in front of me.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:05:31", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_010_3", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010_2_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the parts on the white shelf?", + "answer": "You cannot reach the parts of the white shelf; the parts are to your left.", + "question_chinese": "我能碰到白色架子的零件吗？", + "answer_chinese": "你不可以碰到白色架子的零件，白色架子的零件在你的左边。", + "evidence": [ + { + "description": "You can't touch it.", + "timestamps": { + "start": "00:06:16", + "end": "00:06:22" + } + } + ], + "question_timestamps": "00:06:24", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the parts of the white shelf; the parts are in front of you.", + "B": "You cannot reach the parts of the white shelf; the parts are to your left." + }, + "answer_label": "B", + "answer_text": "You cannot reach the parts of the white shelf; the parts are to your left.", + "q_id": "egolife_A1_010_2_T1", + "correct_answer": "You cannot reach the parts of the white shelf; the parts are to your left.", + "original_options": { + "A": "You can reach the parts of the white shelf; the parts are in front of you.", + "B": "You cannot reach the parts of the white shelf; the parts are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010_4", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the parts of the white shelf?", + "answer": "No.", + "question_chinese": "我能碰到白色架子的零件吗？", + "answer_chinese": "不行", + "evidence": [ + { + "description": "There is a pile of scattered shelf parts in front of me.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:08:49", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_010_4", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010_1_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the window from me?", + "answer": "The window is in front of you.", + "question_chinese": "窗户在我的什么方向？", + "answer_chinese": "窗户在你的正前方。", + "evidence": [ + { + "description": "The window is in front of you.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is behind you.", + "B": "The window is to your left.", + "C": "The window is in front of you.", + "D": "The window is to your right, not directly in front of you.", + "E": "The window is above you, not in front of your line of sight." + }, + "answer_label": "C", + "answer_text": "The window is in front of you.", + "q_id": "egolife_A1_010_1_T2", + "correct_answer": "The window is in front of you.", + "original_options": { + "A": "The window is behind you.", + "B": "The window is to your left.", + "C": "The window is in front of you.", + "D": "The window is to your right, not directly in front of you.", + "E": "The window is above you, not in front of your line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010_1_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the window from me?", + "answer": "The window is to your left.", + "question_chinese": "窗户在我的什么方向？", + "answer_chinese": "窗户在你的左边。", + "evidence": [ + { + "description": "The window is right in front of me.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:04:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is to your right.", + "C": "The window is behind you, not on your left side at all.", + "D": "The window is above you, so it isn’t positioned to your left.", + "E": "The window is to your left." + }, + "answer_label": "E", + "answer_text": "The window is to your left.", + "q_id": "egolife_A1_010_1_T1", + "correct_answer": "The window is to your left.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is to your right.", + "C": "The window is behind you, not on your left side at all.", + "D": "The window is above you, so it isn’t positioned to your left.", + "E": "The window is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_010", + "video_length": "00:12:00", + "qid": "egolife_A1_010_2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the window from me?", + "answer": "The window is in front of your left.", + "question_chinese": "窗户在我的什么方向？", + "answer_chinese": "窗户在你的右后方", + "evidence": [ + { + "description": "The window is right in front of me.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:10:39", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is in front of your left.", + "B": "The window is behind your left.", + "C": "The window is in front of your right.", + "D": "The window is behind your right side, not toward the front-left of you.", + "E": "The window is directly in front of you rather than offset to the left." + }, + "answer_label": "A", + "answer_text": "The window is in front of your left.", + "q_id": "egolife_A1_010_2", + "correct_answer": "The window is in front of your left.", + "original_options": { + "A": "The window is in front of your left.", + "B": "The window is behind your left.", + "C": "The window is in front of your right.", + "D": "The window is behind your right side, not toward the front-left of you.", + "E": "The window is directly in front of you rather than offset to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 13, + 14 + ] + } +] diff --git a/QAs/egolife_A1_011-mcq.json b/QAs/egolife_A1_011-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ca752fba3bcc81a98354075dc5b90221d7076419 --- /dev/null +++ b/QAs/egolife_A1_011-mcq.json @@ -0,0 +1,1371 @@ +[ + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_5", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bedroom, how should I get there?", + "answer": "You should turn left and take one step forward, then you can enter the bedroom.", + "question_chinese": "如果我想去卧室，我应该怎么走？", + "answer_chinese": "你应该向左转走一步距离，然后就可以进入卧室了。", + "evidence": [ + { + "description": "The bedroom is about four or five steps away in front of my left side.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:41" + } + }, + { + "description": "The table in the living room is about two steps away to my right front.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:05:41", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You should turn left and take one step forward, then you can enter the bedroom.", + "B": "Turn right, then step forward once.", + "C": "Go straight ahead two steps.", + "D": "Turn left, walk two steps forward, and then enter the bedroom doorway.", + "E": "Turn right and take one step forward, then you can enter the bedroom." + }, + "answer_label": "A", + "answer_text": "You should turn left and take one step forward, then you can enter the bedroom.", + "q_id": "egolife_A1_011_5", + "correct_answer": "You should turn left and take one step forward, then you can enter the bedroom.", + "original_options": { + "A": "You should turn left and take one step forward, then you can enter the bedroom.", + "B": "Turn right, then step forward once.", + "C": "Go straight ahead two steps.", + "D": "Turn left, walk two steps forward, and then enter the bedroom doorway.", + "E": "Turn right and take one step forward, then you can enter the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_6", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bedroom, how should I get there?", + "answer": "You should move forward to your right front for about four steps until you reach the stairs, then go downstairs and exit the door. After that, walk about five or six steps to your left front and you will get to the bedroom.", + "question_chinese": "如果我想去卧室，我应该怎么走？", + "answer_chinese": "你应该向右前方前进4步左右的距离，然后到达楼梯，随后下楼出门，向左前方走五六步的距离就能到达卧室了。", + "evidence": [ + { + "description": "The bedroom is about four or five steps away in front of my left side.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:41" + } + }, + { + "description": "The table in the living room is about two steps away to my right front.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:41" + } + }, + { + "description": "I went up to the second floor during this period of time.", + "timestamps": { + "start": "00:08:48", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:11:41", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You should move forward to your right front for about four steps until you reach the stairs, then go downstairs and exit the door. After that, walk about five or six steps to your left front and you will get to the bedroom.", + "B": "Go upstairs, then turn left into the bedroom.", + "C": "Head left-front to the stairs, then up.", + "D": "Walk four steps to your left-front to the stairs, go upstairs, exit the door, then continue left-front.", + "E": "Move right-front until the stairs, go downstairs, but turn right-front after the door for five steps." + }, + "answer_label": "A", + "answer_text": "You should move forward to your right front for about four steps until you reach the stairs, then go downstairs and exit the door. After that, walk about five or six steps to your left front and you will get to the bedroom.", + "q_id": "egolife_A1_011_6", + "correct_answer": "You should move forward to your right front for about four steps until you reach the stairs, then go downstairs and exit the door. After that, walk about five or six steps to your left front and you will get to the bedroom.", + "original_options": { + "A": "You should move forward to your right front for about four steps until you reach the stairs, then go downstairs and exit the door. After that, walk about five or six steps to your left front and you will get to the bedroom.", + "B": "Go upstairs, then turn left into the bedroom.", + "C": "Head left-front to the stairs, then up.", + "D": "Walk four steps to your left-front to the stairs, go upstairs, exit the door, then continue left-front.", + "E": "Move right-front until the stairs, go downstairs, but turn right-front after the door for five steps." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_7_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the living room table from me?", + "answer": "The living-room table is below you.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "客厅的桌子在你的下方。", + "evidence": [ + { + "description": "The table in the living room is on my left front.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is behind you.", + "C": "The living-room table is to your left, roughly level with where you are standing.", + "D": "The living-room table sits to your right side, not below you from your current position.", + "E": "The living-room table is below you." + }, + "answer_label": "E", + "answer_text": "The living-room table is below you.", + "q_id": "egolife_A1_011_7_T1", + "correct_answer": "The living-room table is below you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is behind you.", + "C": "The living-room table is to your left, roughly level with where you are standing.", + "D": "The living-room table sits to your right side, not below you from your current position.", + "E": "The living-room table is below you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_7_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the living room table from me?", + "answer": "The living-room table is to your front-right.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "客厅的桌子在你的右前方。", + "evidence": [ + { + "description": "The living room table is on your right front.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:35" + } + } + ], + "question_timestamps": "00:02:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to your front-left.", + "B": "It is to your back-right.", + "C": "The living-room table is to your front-right.", + "D": "The living-room table is directly in front of you, not positioned to your front-right.", + "E": "The living-room table is off to your right but slightly behind you rather than ahead." + }, + "answer_label": "C", + "answer_text": "The living-room table is to your front-right.", + "q_id": "egolife_A1_011_7_T2", + "correct_answer": "The living-room table is to your front-right.", + "original_options": { + "A": "It is to your front-left.", + "B": "It is to your back-right.", + "C": "The living-room table is to your front-right.", + "D": "The living-room table is directly in front of you, not positioned to your front-right.", + "E": "The living-room table is off to your right but slightly behind you rather than ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_9", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the table in the living room relative to me?", + "answer": "The table in the living room is on your left front.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "客厅的桌子在你的左前方。", + "evidence": [ + { + "description": "The table in the living room is on my left front.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:06:29", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your right front.", + "B": "It is directly behind you.", + "C": "The table in the living room is on your left front.", + "D": "The table in the living room is on your left side, not out in front of you.", + "E": "The table in the living room is straight ahead of you rather than positioned toward your left front." + }, + "answer_label": "C", + "answer_text": "The table in the living room is on your left front.", + "q_id": "egolife_A1_011_9", + "correct_answer": "The table in the living room is on your left front.", + "original_options": { + "A": "It is on your right front.", + "B": "It is directly behind you.", + "C": "The table in the living room is on your left front.", + "D": "The table in the living room is on your left side, not out in front of you.", + "E": "The table in the living room is straight ahead of you rather than positioned toward your left front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011-mcq_fixed_7", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the living room table from me?", + "answer": "The table in the living room is below you.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "客厅的桌子在你的下方", + "evidence": [ + { + "description": "The table in the living room is on my left front.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:11:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is above you.", + "B": "It is to your left.", + "C": "The table in the living room is below you.", + "D": "The table in the living room is in front of you at about waist height, not below your position.", + "E": "The table in the living room is to your upper-right, appearing raised rather than situated below you." + }, + "answer_label": "C", + "answer_text": "The table in the living room is below you.", + "q_id": "egolife_A1_011-mcq_fixed_7", + "correct_answer": "The table in the living room is below you.", + "original_options": { + "A": "It is above you.", + "B": "It is to your left.", + "C": "The table in the living room is below you.", + "D": "The table in the living room is in front of you at about waist height, not below your position.", + "E": "The table in the living room is to your upper-right, appearing raised rather than situated below you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_8_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction am I from the whiteboard shelf?", + "answer": "The whiteboard shelf is straight in front of you.", + "question_chinese": "我在白板架子的什么方向？", + "answer_chinese": "白板架子在你的正前方。", + "evidence": [ + { + "description": "We are standing in the living room assembling a frame for the whiteboard.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is straight behind you.", + "B": "It is straight to your left.", + "C": "The whiteboard shelf is off to your right side rather than directly in front of you.", + "D": "The whiteboard shelf is above you, so it is not positioned straight ahead from where you stand.", + "E": "The whiteboard shelf is straight in front of you." + }, + "answer_label": "E", + "answer_text": "The whiteboard shelf is straight in front of you.", + "q_id": "egolife_A1_011_8_T2", + "correct_answer": "The whiteboard shelf is straight in front of you.", + "original_options": { + "A": "It is straight behind you.", + "B": "It is straight to your left.", + "C": "The whiteboard shelf is off to your right side rather than directly in front of you.", + "D": "The whiteboard shelf is above you, so it is not positioned straight ahead from where you stand.", + "E": "The whiteboard shelf is straight in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_8_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction am I from the whiteboard shelf?", + "answer": "The whiteboard shelf is to your left.", + "question_chinese": "我在白板架子的什么方向？", + "answer_chinese": "白板架子在你的左边。", + "evidence": [ + { + "description": "The whiteboard stand is on your left.", + "timestamps": { + "start": "00:02:21", + "end": "00:02:26" + } + } + ], + "question_timestamps": "00:02:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The whiteboard shelf is to your left.", + "B": "It is in front of you.", + "C": "It is behind you.", + "D": "The whiteboard shelf is to your right, not located on your left side.", + "E": "The whiteboard shelf is above you rather than positioned to your left from your viewpoint." + }, + "answer_label": "A", + "answer_text": "The whiteboard shelf is to your left.", + "q_id": "egolife_A1_011_8_T1", + "correct_answer": "The whiteboard shelf is to your left.", + "original_options": { + "A": "The whiteboard shelf is to your left.", + "B": "It is in front of you.", + "C": "It is behind you.", + "D": "The whiteboard shelf is to your right, not located on your left side.", + "E": "The whiteboard shelf is above you rather than positioned to your left from your viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011-mcq_fixed_8", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction am I relative to the whiteboard stand?", + "answer": "The whiteboard stand is directly in front of you.", + "question_chinese": "我在白板架子的什么方向？", + "answer_chinese": "白板架子在你的正前方。", + "evidence": [ + { + "description": "We are standing in the living room assembling a frame for the whiteboard.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:05:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is directly behind you.", + "B": "It is directly to your left.", + "C": "The whiteboard stand is directly in front of you.", + "D": "The whiteboard stand is to your right at an angle, not directly in front of you.", + "E": "The whiteboard stand is slightly off to your left side, so you are not positioned straight in front of it." + }, + "answer_label": "C", + "answer_text": "The whiteboard stand is directly in front of you.", + "q_id": "egolife_A1_011-mcq_fixed_8", + "correct_answer": "The whiteboard stand is directly in front of you.", + "original_options": { + "A": "It is directly behind you.", + "B": "It is directly to your left.", + "C": "The whiteboard stand is directly in front of you.", + "D": "The whiteboard stand is to your right at an angle, not directly in front of you.", + "E": "The whiteboard stand is slightly off to your left side, so you are not positioned straight in front of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 20 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_12", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In what direction is the whiteboard stand from me?", + "answer": "The whiteboard stand is directly in front of you.", + "question_chinese": "我在白板架子的什么方向？", + "answer_chinese": "白板的架子在你的正下方。", + "evidence": [ + { + "description": "We are standing in the living room assembling a frame for the whiteboard.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:11:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is directly behind you.", + "B": "The whiteboard stand is directly in front of you.", + "C": "It is to your right.", + "D": "The whiteboard stand is positioned to your front-left, slightly off to the side as you face forward.", + "E": "The whiteboard stand sits to your back-right, so you would need to turn around to see it." + }, + "answer_label": "B", + "answer_text": "The whiteboard stand is directly in front of you.", + "q_id": "egolife_A1_011_12", + "correct_answer": "The whiteboard stand is directly in front of you.", + "original_options": { + "A": "It is directly behind you.", + "B": "The whiteboard stand is directly in front of you.", + "C": "It is to your right.", + "D": "The whiteboard stand is positioned to your front-left, slightly off to the side as you face forward.", + "E": "The whiteboard stand sits to your back-right, so you would need to turn around to see it." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_5_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction am I from the black-and-white checkered board?", + "answer": "The black-and-white checkerboard is to your front-left.", + "question_chinese": "我在黑白方格板子的什么方向？", + "answer_chinese": "黑白方格板子在你的左前方。", + "evidence": [ + { + "description": "The black and white checkered board is on your left front.", + "timestamps": { + "start": "00:02:46", + "end": "00:02:49" + } + } + ], + "question_timestamps": "00:02:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are front-right of it.", + "B": "You are directly behind it.", + "C": "From the checkerboard’s position, you would be located to its back-left rather than near its front corner.", + "D": "The black-and-white checkerboard is to your front-left.", + "E": "You are positioned to the checkerboard’s front-right, meaning it would appear over your right shoulder." + }, + "answer_label": "D", + "answer_text": "The black-and-white checkerboard is to your front-left.", + "q_id": "egolife_A1_011_5_T2", + "correct_answer": "The black-and-white checkerboard is to your front-left.", + "original_options": { + "A": "You are front-right of it.", + "B": "You are directly behind it.", + "C": "From the checkerboard’s position, you would be located to its back-left rather than near its front corner.", + "D": "The black-and-white checkerboard is to your front-left.", + "E": "You are positioned to the checkerboard’s front-right, meaning it would appear over your right shoulder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_5_T1", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction am I from the black-and-white checkered board?", + "answer": "The black-and-white checkered board is in front of you.", + "question_chinese": "我在黑白方格板子的什么方向？", + "answer_chinese": "黑白格子的板子在你的前方", + "evidence": [ + { + "description": "The black and white checkered board is in front of you.", + "timestamps": { + "start": "00:03:23", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:03:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The black-and-white checkered board is in front of you.", + "B": "It is behind you.", + "C": "It is to your left.", + "D": "The black-and-white checkered board is off to your front-right, not centered in your forward view.", + "E": "The black-and-white checkered board is to your back-left, so you’d need to turn to face it." + }, + "answer_label": "A", + "answer_text": "The black-and-white checkered board is in front of you.", + "q_id": "egolife_A1_011_5_T1", + "correct_answer": "The black-and-white checkered board is in front of you.", + "original_options": { + "A": "The black-and-white checkered board is in front of you.", + "B": "It is behind you.", + "C": "It is to your left.", + "D": "The black-and-white checkered board is off to your front-right, not centered in your forward view.", + "E": "The black-and-white checkered board is to your back-left, so you’d need to turn to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_2_T1", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which door am I closer to right now?", + "answer": "You are closer to the door leading to the bedroom; it’s to your left.", + "question_chinese": "我现在离哪扇门更近？", + "answer_chinese": "你离通往卧室的门更近，它在你左边。", + "evidence": [ + { + "description": "Directly in front of me is the door leading to the second floor.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "What is in front of my left side is the door leading to the bedroom.", + "timestamps": { + "start": "00:01:39", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are closer to the second-floor door.", + "B": "You are closer to the right bedroom door.", + "C": "You are closer to the door leading to the second floor, and it is positioned in front of you.", + "D": "You are closer to the bedroom door, but it is behind you rather than located on your left.", + "E": "You are closer to the door leading to the bedroom; it’s to your left." + }, + "answer_label": "E", + "answer_text": "You are closer to the door leading to the bedroom; it’s to your left.", + "q_id": "egolife_A1_011_2_T1", + "correct_answer": "You are closer to the door leading to the bedroom; it’s to your left.", + "original_options": { + "A": "You are closer to the second-floor door.", + "B": "You are closer to the right bedroom door.", + "C": "You are closer to the door leading to the second floor, and it is positioned in front of you.", + "D": "You are closer to the bedroom door, but it is behind you rather than located on your left.", + "E": "You are closer to the door leading to the bedroom; it’s to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_3", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which door am I closer to now?", + "answer": "You are closer to the door leading to the bedroom.", + "question_chinese": "我现在离哪扇门更近？", + "answer_chinese": "你离通往卧室的门更近。", + "evidence": [ + { + "description": "Directly in front of me is the door leading to the second floor.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "What is in front of my left side is the door leading to the bedroom.", + "timestamps": { + "start": "00:01:39", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:05:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You are closer to the bathroom door.", + "B": "You are closer to the hallway door.", + "C": "You are closer to the dining room door, not the bedroom door, based on where you are standing now.", + "D": "You are closer to the garage door at the moment, making it the nearest option instead of the bedroom.", + "E": "You are closer to the door leading to the bedroom." + }, + "answer_label": "E", + "answer_text": "You are closer to the door leading to the bedroom.", + "q_id": "egolife_A1_011_3", + "correct_answer": "You are closer to the door leading to the bedroom.", + "original_options": { + "A": "You are closer to the bathroom door.", + "B": "You are closer to the hallway door.", + "C": "You are closer to the dining room door, not the bedroom door, based on where you are standing now.", + "D": "You are closer to the garage door at the moment, making it the nearest option instead of the bedroom.", + "E": "You are closer to the door leading to the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_2_T3", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which door am I closer to right now?", + "answer": "You are closer to the door on the second floor; it’s behind you.", + "question_chinese": "我现在离哪扇门更近？", + "answer_chinese": "你离2楼的门更近，它在你身后。", + "evidence": [ + { + "description": "Directly in front of me is the door leading to the second floor.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You are closer to the door on the second floor.", + "timestamps": { + "start": "00:09:02", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:09:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are closer to the bedroom door.", + "B": "You are closer to the door on the second floor; it’s behind you.", + "C": "You are closer to the front second-floor door.", + "D": "You are closer to the bedroom door, and it is positioned to your left rather than behind you.", + "E": "You are closer to the second-floor door, but it is in front of you instead of behind you." + }, + "answer_label": "B", + "answer_text": "You are closer to the door on the second floor; it’s behind you.", + "q_id": "egolife_A1_011_2_T3", + "correct_answer": "You are closer to the door on the second floor; it’s behind you.", + "original_options": { + "A": "You are closer to the bedroom door.", + "B": "You are closer to the door on the second floor; it’s behind you.", + "C": "You are closer to the front second-floor door.", + "D": "You are closer to the bedroom door, and it is positioned to your left rather than behind you.", + "E": "You are closer to the second-floor door, but it is in front of you instead of behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_4", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which door am I closer to now?", + "answer": "Including the vertical distance, you are closer to the door leading to the second floor.", + "question_chinese": "我现在离哪扇门更近？", + "answer_chinese": "算上垂直距离，你离通往2楼的门更近。", + "evidence": [ + { + "description": "Directly in front of me is the door leading to the second floor.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "What is in front of my left side is the door leading to the bedroom.", + "timestamps": { + "start": "00:01:39", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:09:16", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You are closer to the first-floor door.", + "B": "Including the vertical distance, you are closer to the door leading to the second floor.", + "C": "You are closer to the basement door.", + "D": "Including the vertical distance, you are closer to the garage door rather than the door leading upward.", + "E": "Including the vertical distance, you are closer to the kitchen door, making it the nearest overall option." + }, + "answer_label": "B", + "answer_text": "Including the vertical distance, you are closer to the door leading to the second floor.", + "q_id": "egolife_A1_011_4", + "correct_answer": "Including the vertical distance, you are closer to the door leading to the second floor.", + "original_options": { + "A": "You are closer to the first-floor door.", + "B": "Including the vertical distance, you are closer to the door leading to the second floor.", + "C": "You are closer to the basement door.", + "D": "Including the vertical distance, you are closer to the garage door rather than the door leading upward.", + "E": "Including the vertical distance, you are closer to the kitchen door, making it the nearest overall option." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_6_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many people did I see?", + "answer": "You saw three people, and they were all in front of you.", + "question_chinese": "我看到多少个人？", + "answer_chinese": "你看到三个人，他们都在你前面。", + "evidence": [ + { + "description": "You see three people, all in front of you.", + "timestamps": { + "start": "00:01:42", + "end": "00:01:48" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two people in front.", + "B": "You saw four people in front.", + "C": "You saw one person, and they were in front of you the entire time you were looking ahead.", + "D": "You saw five people, and they were all in front of you, grouped within your forward field of view.", + "E": "You saw three people, and they were all in front of you." + }, + "answer_label": "E", + "answer_text": "You saw three people, and they were all in front of you.", + "q_id": "egolife_A1_011_6_T2", + "correct_answer": "You saw three people, and they were all in front of you.", + "original_options": { + "A": "You saw two people in front.", + "B": "You saw four people in front.", + "C": "You saw one person, and they were in front of you the entire time you were looking ahead.", + "D": "You saw five people, and they were all in front of you, grouped within your forward field of view.", + "E": "You saw three people, and they were all in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_6_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many people did I see?", + "answer": "You saw two people, and they were in front of you.", + "question_chinese": "我看到多少个人？", + "answer_chinese": "你看到两个人，他们在你前方。", + "evidence": [ + { + "description": "You see two people ahead, a boy and a girl.", + "timestamps": { + "start": "00:03:26", + "end": "00:03:31" + } + } + ], + "question_timestamps": "00:03:34", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one person in front of you.", + "B": "You saw three people in front of you.", + "C": "You saw four people, all standing together in front of you as you looked ahead.", + "D": "You saw two people, and they were in front of you.", + "E": "You saw zero people, with no one visible in front of you at that moment." + }, + "answer_label": "D", + "answer_text": "You saw two people, and they were in front of you.", + "q_id": "egolife_A1_011_6_T1", + "correct_answer": "You saw two people, and they were in front of you.", + "original_options": { + "A": "You saw one person in front of you.", + "B": "You saw three people in front of you.", + "C": "You saw four people, all standing together in front of you as you looked ahead.", + "D": "You saw two people, and they were in front of you.", + "E": "You saw zero people, with no one visible in front of you at that moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_4_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the window relative to me?", + "answer": "The window is to your front-left.", + "question_chinese": "窗户在我的什么方向？", + "answer_chinese": "窗户在你的左前方。", + "evidence": [ + { + "description": "The window is on your left front.", + "timestamps": { + "start": "00:05:58", + "end": "00:06:05" + } + } + ], + "question_timestamps": "00:06:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is to your front-left.", + "B": "The window is to your front-right.", + "C": "The window is to your back-left.", + "D": "The window is directly to your left, not angled toward your front as you face forward.", + "E": "The window is behind you on the left side, so you would need to turn around to see it." + }, + "answer_label": "A", + "answer_text": "The window is to your front-left.", + "q_id": "egolife_A1_011_4_T2", + "correct_answer": "The window is to your front-left.", + "original_options": { + "A": "The window is to your front-left.", + "B": "The window is to your front-right.", + "C": "The window is to your back-left.", + "D": "The window is directly to your left, not angled toward your front as you face forward.", + "E": "The window is behind you on the left side, so you would need to turn around to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the window from me?", + "answer": "The window is in front of my left side.", + "question_chinese": "窗户在我的什么方向？", + "answer_chinese": "窗户在我的左前方。", + "evidence": [ + { + "description": "The window is directly in front of me.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:06:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is in front of my left side.", + "B": "The window is front-right.", + "C": "The window is behind my left side.", + "D": "The window is directly to my right, not positioned toward the front of my left side.", + "E": "The window is behind me on the right, requiring me to turn to face it." + }, + "answer_label": "A", + "answer_text": "The window is in front of my left side.", + "q_id": "egolife_A1_011_1", + "correct_answer": "The window is in front of my left side.", + "original_options": { + "A": "The window is in front of my left side.", + "B": "The window is front-right.", + "C": "The window is behind my left side.", + "D": "The window is directly to my right, not positioned toward the front of my left side.", + "E": "The window is behind me on the right, requiring me to turn to face it." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_4_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the window relative to me?", + "answer": "The window is to your left.", + "question_chinese": "窗户在我的什么方向？", + "answer_chinese": "窗户在你的左边", + "evidence": [ + { + "description": "The window is on your left.", + "timestamps": { + "start": "00:06:54", + "end": "00:06:57" + } + } + ], + "question_timestamps": "00:06:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is behind you.", + "D": "The window is to your right, on the opposite side from where your left shoulder is.", + "E": "The window is above you and slightly forward, not level with you on your left side." + }, + "answer_label": "B", + "answer_text": "The window is to your left.", + "q_id": "egolife_A1_011_4_T1", + "correct_answer": "The window is to your left.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is behind you.", + "D": "The window is to your right, on the opposite side from where your left shoulder is.", + "E": "The window is above you and slightly forward, not level with you on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011_2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the window from me?", + "answer": "The window is positioned slightly downward directly in front of you.", + "question_chinese": "窗户在我的什么方向？", + "answer_chinese": "窗户在你的正前方偏下的位置。", + "evidence": [ + { + "description": "The window is directly in front of me.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:11:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is slightly upward in front.", + "B": "The window is slightly to your left.", + "C": "The window is positioned slightly downward directly in front of you.", + "D": "The window is directly in front of you but positioned slightly upward rather than lower in your view.", + "E": "The window is behind you at about the same height, so it is not in front and downward." + }, + "answer_label": "C", + "answer_text": "The window is positioned slightly downward directly in front of you.", + "q_id": "egolife_A1_011_2", + "correct_answer": "The window is positioned slightly downward directly in front of you.", + "original_options": { + "A": "The window is slightly upward in front.", + "B": "The window is slightly to your left.", + "C": "The window is positioned slightly downward directly in front of you.", + "D": "The window is directly in front of you but positioned slightly upward rather than lower in your view.", + "E": "The window is behind you at about the same height, so it is not in front and downward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011-mcq_fixed_10", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the red refrigerator from me?", + "answer": "The red refrigerator is behind you on the right.", + "question_chinese": "红色的冰箱在我的什么方向？", + "answer_chinese": "红色的冰箱在你的右后方。", + "evidence": [ + { + "description": "The red refrigerator is in front of my left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:04:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The red refrigerator is behind you on the right.", + "B": "The red refrigerator is in front-right.", + "C": "The red refrigerator is behind-left.", + "D": "The red refrigerator is directly to your right at your side, not behind you as you face forward.", + "E": "The red refrigerator is behind you on the left, so it’s on the opposite side from your right." + }, + "answer_label": "A", + "answer_text": "The red refrigerator is behind you on the right.", + "q_id": "egolife_A1_011-mcq_fixed_10", + "correct_answer": "The red refrigerator is behind you on the right.", + "original_options": { + "A": "The red refrigerator is behind you on the right.", + "B": "The red refrigerator is in front-right.", + "C": "The red refrigerator is behind-left.", + "D": "The red refrigerator is directly to your right at your side, not behind you as you face forward.", + "E": "The red refrigerator is behind you on the left, so it’s on the opposite side from your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_011", + "video_length": "00:12:00", + "qid": "egolife_A1_011-mcq_fixed_11", + "qid_num": 24, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the red refrigerator from me?", + "answer": "The red refrigerator should be below you on the right.", + "question_chinese": "红色的冰箱在我的什么方向？", + "answer_chinese": "红色的冰箱应该在你的右下方。", + "evidence": [ + { + "description": "The red refrigerator is in front of my left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:09:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The red refrigerator is below-left.", + "B": "The red refrigerator is above-right.", + "C": "The red refrigerator is positioned diagonally above and a bit behind you rather than below you on the right.", + "D": "The red refrigerator should be below you on the right.", + "E": "The red refrigerator is directly above you on the left, not lower and to the right from your viewpoint." + }, + "answer_label": "D", + "answer_text": "The red refrigerator should be below you on the right.", + "q_id": "egolife_A1_011-mcq_fixed_11", + "correct_answer": "The red refrigerator should be below you on the right.", + "original_options": { + "A": "The red refrigerator is below-left.", + "B": "The red refrigerator is above-right.", + "C": "The red refrigerator is positioned diagonally above and a bit behind you rather than below you on the right.", + "D": "The red refrigerator should be below you on the right.", + "E": "The red refrigerator is directly above you on the left, not lower and to the right from your viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + } +] diff --git a/QAs/egolife_A1_012-mcq.json b/QAs/egolife_A1_012-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5f474efe2d4289f059bad0c5ca12b67d38ea6e8a --- /dev/null +++ b/QAs/egolife_A1_012-mcq.json @@ -0,0 +1,751 @@ +[ + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012_11", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the microwave on the second floor from me?", + "answer": "It's behind and above you.", + "question_chinese": "2楼的微波炉在我的什么方向？", + "answer_chinese": "在你的后上方。", + "evidence": [ + { + "description": "The microwave on the second floor is directly in front of me.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:05:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's in front and below you.", + "B": "It's beside and above you.", + "C": "It's behind you but down low, so it is not positioned above you on the second floor.", + "D": "It's behind and above you.", + "E": "It's in front of you and above you, meaning you wouldn't describe it as behind and above." + }, + "answer_label": "D", + "answer_text": "It's behind and above you.", + "q_id": "egolife_A1_012_11", + "correct_answer": "It's behind and above you.", + "original_options": { + "A": "It's in front and below you.", + "B": "It's beside and above you.", + "C": "It's behind you but down low, so it is not positioned above you on the second floor.", + "D": "It's behind and above you.", + "E": "It's in front of you and above you, meaning you wouldn't describe it as behind and above." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012_12", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the microwave on the second floor from me?", + "answer": "It's on your upper right.", + "question_chinese": "2楼的微波炉在我的什么方向？", + "answer_chinese": "在你的右上方", + "evidence": [ + { + "description": "The microwave on the second floor is directly in front of me.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:11:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's on your upper right.", + "B": "It's on your upper left.", + "C": "It's on your lower right.", + "D": "It's slightly ahead on your upper left, near where the corridor opens up on the second floor.", + "E": "It's behind you and a bit to the left, tucked near the wall as you face down the hallway." + }, + "answer_label": "A", + "answer_text": "It's on your upper right.", + "q_id": "egolife_A1_012_12", + "correct_answer": "It's on your upper right.", + "original_options": { + "A": "It's on your upper right.", + "B": "It's on your upper left.", + "C": "It's on your lower right.", + "D": "It's slightly ahead on your upper left, near where the corridor opens up on the second floor.", + "E": "It's behind you and a bit to the left, tucked near the wall as you face down the hallway." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012_1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the wooden bench on the second floor from me?", + "answer": "The wooden stool is above and in front of you.", + "question_chinese": "2楼的木质凳子在我的什么方向？", + "answer_chinese": "木质凳子在你的前上方", + "evidence": [ + { + "description": "There is a wooden stool in front of my left side.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "It is not difficult to see by comparison that the structures of the upper and lower floors are almost the same.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:05:49", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It's above and behind you.", + "B": "The wooden stool is above and in front of you.", + "C": "It's below and in front of you.", + "D": "The wooden stool is up and to your left, slightly ahead as you look toward the second-floor area.", + "E": "The wooden stool is in front of you but lower down, closer to the right side of your view." + }, + "answer_label": "B", + "answer_text": "The wooden stool is above and in front of you.", + "q_id": "egolife_A1_012_1", + "correct_answer": "The wooden stool is above and in front of you.", + "original_options": { + "A": "It's above and behind you.", + "B": "The wooden stool is above and in front of you.", + "C": "It's below and in front of you.", + "D": "The wooden stool is up and to your left, slightly ahead as you look toward the second-floor area.", + "E": "The wooden stool is in front of you but lower down, closer to the right side of your view." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012_2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the wooden bench on the second floor from me?", + "answer": "Now the wooden stool inside the row is almost directly above you.", + "question_chinese": "2楼的木质凳子在我的什么方向？", + "answer_chinese": "现在内排木质凳子，差不多在你的正上方", + "evidence": [ + { + "description": "There is a wooden stool in front of my left side.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "It is not difficult to see by comparison that the structures of the upper and lower floors are almost the same.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:10:41", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It's almost directly below you.", + "B": "It's up and to your right.", + "C": "Now the wooden stool inside the row is above you but shifted slightly forward, not directly overhead.", + "D": "Now the wooden stool inside the row is nearly overhead, but it sits a little to your left rather than centered.", + "E": "Now the wooden stool inside the row is almost directly above you." + }, + "answer_label": "E", + "answer_text": "Now the wooden stool inside the row is almost directly above you.", + "q_id": "egolife_A1_012_2", + "correct_answer": "Now the wooden stool inside the row is almost directly above you.", + "original_options": { + "A": "It's almost directly below you.", + "B": "It's up and to your right.", + "C": "Now the wooden stool inside the row is above you but shifted slightly forward, not directly overhead.", + "D": "Now the wooden stool inside the row is nearly overhead, but it sits a little to your left rather than centered.", + "E": "Now the wooden stool inside the row is almost directly above you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 21 + ] + }, + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012_5", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which of the two open doors in the living room on the first floor is closer to me?", + "answer": "The door leading to the stairs is closer to you.", + "question_chinese": "一楼客厅里打开的两扇门里面哪扇门离我更近？", + "answer_chinese": "通往楼梯的门离你更近。", + "evidence": [ + { + "description": "The door in front of me that leads to the living room is the one connecting the living room and the stairs.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + }, + { + "description": "The open door in front of me now is the one leading to the bedroom.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:05:59", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The hallway door is closer.", + "B": "The kitchen door is closer.", + "C": "The door leading to the stairs is closer to you.", + "D": "The door leading to the hallway is closer to you than the one that leads to the stairs.", + "E": "The door to the kitchen is the nearer one, while the door to the stairs is farther away from you." + }, + "answer_label": "C", + "answer_text": "The door leading to the stairs is closer to you.", + "q_id": "egolife_A1_012_5", + "correct_answer": "The door leading to the stairs is closer to you.", + "original_options": { + "A": "The hallway door is closer.", + "B": "The kitchen door is closer.", + "C": "The door leading to the stairs is closer to you.", + "D": "The door leading to the hallway is closer to you than the one that leads to the stairs.", + "E": "The door to the kitchen is the nearer one, while the door to the stairs is farther away from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012_6", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which of the two open doors in the living room on the first floor is closer to me?", + "answer": "Now the distance from the two doors to you is about the same.", + "question_chinese": "一楼客厅里打开的两扇门里面哪扇门离我更近？", + "answer_chinese": "现在两扇门到你的距离都差不多。", + "evidence": [ + { + "description": "The door in front of me that leads to the living room is the one connecting the living room and the stairs.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + }, + { + "description": "The open door in front of me now is the one leading to the bedroom.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:11:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The kitchen door is closer.", + "B": "The garage door is closer.", + "C": "Now the distance from the two doors to you is about the same.", + "D": "You are clearly closer to the kitchen door, with the other open door noticeably farther away from you.", + "E": "The two doors are not equally distant; one is much nearer to you than the other right now." + }, + "answer_label": "C", + "answer_text": "Now the distance from the two doors to you is about the same.", + "q_id": "egolife_A1_012_6", + "correct_answer": "Now the distance from the two doors to you is about the same.", + "original_options": { + "A": "The kitchen door is closer.", + "B": "The garage door is closer.", + "C": "Now the distance from the two doors to you is about the same.", + "D": "You are clearly closer to the kitchen door, with the other open door noticeably farther away from you.", + "E": "The two doors are not equally distant; one is much nearer to you than the other right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012-mcq_fixed_6", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many wooden chairs have I seen in total so far?", + "answer": "Three.", + "question_chinese": "到现在为止，我总共看到过几把木质椅子？", + "answer_chinese": "一共有3把椅子。", + "evidence": [ + { + "description": "rom this perspective, three wooden chairs are currently visible in the hall on the second floor.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You have seen a total of two wooden chairs in all so far during the video.", + "E": "You have seen four wooden chairs altogether up to this point in the video so far." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_012-mcq_fixed_6", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You have seen a total of two wooden chairs in all so far during the video.", + "E": "You have seen four wooden chairs altogether up to this point in the video so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012-mcq_fixed_7", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many wooden chairs have I seen in total so far?", + "answer": "Five.", + "question_chinese": "到现在为止，我总共看到过几把木质椅子？", + "answer_chinese": "一共见到过8把木制椅子。", + "evidence": [ + { + "description": "rom this perspective, three wooden chairs are currently visible in the hall on the second floor.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:18" + } + }, + { + "description": "There are another five wooden chairs in the hall on the first floor.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:10:46", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "You have seen a total of three wooden chairs in all so far during the video.", + "E": "You have seen seven wooden chairs altogether up to this point in the video so far." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_012-mcq_fixed_7", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "You have seen a total of three wooden chairs in all so far during the video.", + "E": "You have seen seven wooden chairs altogether up to this point in the video so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012-mcq_fixed_2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the second floor to get a wooden chair now, what should I do?", + "answer": "You should walk two or three steps forward to the right. When you get near the stairs, go up to the second floor. After exiting, turn left and walk three steps, and you will reach the side of the wooden chair.", + "question_chinese": "如果我现在想到二楼去拿一张木质椅子，我应该怎么走？", + "answer_chinese": "你应该向右前方走两三步。到达楼梯附近，然后上到二楼，出门向左转走三步的距离就能到达木质椅子的旁边。", + "evidence": [ + { + "description": "There is a wooden stool in front of my left side.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "The wooden chair on the second floor is about two or three steps away from the door leading to the stairs.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I went down from the second floor to the first floor.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:31" + } + }, + { + "description": "The table on the first floor is about three steps away from the door leading to the second floor.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:05:42", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Walk forward-left, upstairs, then turn right.", + "B": "Go straight to the stairs and ascend.", + "C": "You should walk two or three steps forward to the right. When you get near the stairs, go up to the second floor. After exiting, turn left and walk three steps, and you will reach the side of the wooden chair.", + "D": "Walk two or three steps forward to the left, go up the stairs, then turn right and walk three steps.", + "E": "Take two or three steps forward to the right, go upstairs, then turn left and walk about five steps to the chair." + }, + "answer_label": "C", + "answer_text": "You should walk two or three steps forward to the right. When you get near the stairs, go up to the second floor. After exiting, turn left and walk three steps, and you will reach the side of the wooden chair.", + "q_id": "egolife_A1_012-mcq_fixed_2", + "correct_answer": "You should walk two or three steps forward to the right. When you get near the stairs, go up to the second floor. After exiting, turn left and walk three steps, and you will reach the side of the wooden chair.", + "original_options": { + "A": "Walk forward-left, upstairs, then turn right.", + "B": "Go straight to the stairs and ascend.", + "C": "You should walk two or three steps forward to the right. When you get near the stairs, go up to the second floor. After exiting, turn left and walk three steps, and you will reach the side of the wooden chair.", + "D": "Walk two or three steps forward to the left, go up the stairs, then turn right and walk three steps.", + "E": "Take two or three steps forward to the right, go upstairs, then turn left and walk about five steps to the chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 22 + ] + }, + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012-mcq_fixed_3", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the second floor to get a wooden chair now, what should I do?", + "answer": "You should turn around, then walk three steps straight ahead to get near the stairs. After that, go up to the second floor. Exit and turn diagonally left forward, then walk three steps, and you will see the wooden chair.", + "question_chinese": "如果我现在想到二楼去拿一张木质椅子，我应该怎么走？", + "answer_chinese": "你应该调头，然后向正前方走三步的距离到达楼梯附近，然后上到二楼，出门向左前方转走三步的距离，就能看见木质的椅子了。", + "evidence": [ + { + "description": "There is a wooden stool in front of my left side.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "The wooden chair on the second floor is about two or three steps away from the door leading to the stairs.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "I went down from the second floor to the first floor.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:31" + } + }, + { + "description": "The table on the first floor is about three steps away from the door leading to the second floor.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:11:05", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn around and head for the stairs.", + "B": "You should turn around, then walk three steps straight ahead to get near the stairs. After that, go up to the second floor. Exit and turn diagonally left forward, then walk three steps, and you will see the wooden chair.", + "C": "Go upstairs, then turn right outside.", + "D": "Turn around, walk two steps to the stairs, go up, then turn diagonally right and walk three steps.", + "E": "Turn around, walk four steps to reach the stairs, go up, then turn diagonally left and walk two steps." + }, + "answer_label": "B", + "answer_text": "You should turn around, then walk three steps straight ahead to get near the stairs. After that, go up to the second floor. Exit and turn diagonally left forward, then walk three steps, and you will see the wooden chair.", + "q_id": "egolife_A1_012-mcq_fixed_3", + "correct_answer": "You should turn around, then walk three steps straight ahead to get near the stairs. After that, go up to the second floor. Exit and turn diagonally left forward, then walk three steps, and you will see the wooden chair.", + "original_options": { + "A": "Turn around and head for the stairs.", + "B": "You should turn around, then walk three steps straight ahead to get near the stairs. After that, go up to the second floor. Exit and turn diagonally left forward, then walk three steps, and you will see the wooden chair.", + "C": "Go upstairs, then turn right outside.", + "D": "Turn around, walk two steps to the stairs, go up, then turn diagonally right and walk three steps.", + "E": "Turn around, walk four steps to reach the stairs, go up, then turn diagonally left and walk two steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012_8_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the table relative to me?", + "answer": "The table is to your back-left.", + "question_chinese": "桌子在我的什么方向？", + "answer_chinese": "桌子在你的左后方。", + "evidence": [ + { + "description": "The table is at my right rear.", + "timestamps": { + "start": "00:02:23", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:02:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The table is to your back-right.", + "B": "The table is to your front-left.", + "C": "The table is directly in front of you, so you can face it without turning your body.", + "D": "The table is on your right side, a bit behind your shoulder rather than fully to the rear.", + "E": "The table is to your back-left." + }, + "answer_label": "E", + "answer_text": "The table is to your back-left.", + "q_id": "egolife_A1_012_8_T2", + "correct_answer": "The table is to your back-left.", + "original_options": { + "A": "The table is to your back-right.", + "B": "The table is to your front-left.", + "C": "The table is directly in front of you, so you can face it without turning your body.", + "D": "The table is on your right side, a bit behind your shoulder rather than fully to the rear.", + "E": "The table is to your back-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_012", + "video_length": "00:12:00", + "qid": "egolife_A1_012_8_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the table relative to me?", + "answer": "The table is directly in front of you.", + "question_chinese": "桌子在我的什么方向？", + "answer_chinese": "桌子在你的正前方。", + "evidence": [ + { + "description": "The table in the hall on the first floor is in front of me.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:11:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The table is directly behind you.", + "B": "The table is directly to your left.", + "C": "The table is on your right side, so you would need to look right to face it.", + "D": "The table is diagonally in front-left of you, not centered straight ahead in your view.", + "E": "The table is directly in front of you." + }, + "answer_label": "E", + "answer_text": "The table is directly in front of you.", + "q_id": "egolife_A1_012_8_T1", + "correct_answer": "The table is directly in front of you.", + "original_options": { + "A": "The table is directly behind you.", + "B": "The table is directly to your left.", + "C": "The table is on your right side, so you would need to look right to face it.", + "D": "The table is diagonally in front-left of you, not centered straight ahead in your view.", + "E": "The table is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 15 + ] + } +] diff --git a/QAs/egolife_A1_013-mcq.json b/QAs/egolife_A1_013-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e5ff42665444b65482444618f5e2b15416e42599 --- /dev/null +++ b/QAs/egolife_A1_013-mcq.json @@ -0,0 +1,823 @@ +[ + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013-mcq_fixed_10", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the table in the hall from me?", + "answer": "The table in the hall is on your left.", + "question_chinese": "大厅的桌子在我的什么方向？", + "answer_chinese": "大厅的桌子在你的左侧。", + "evidence": [ + { + "description": "The table in the hall is directly in front of me.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The table is on your right.", + "B": "The table is behind you.", + "C": "The table is directly in front of you, centered down the hallway as you face forward.", + "D": "The table in the hall is on your left.", + "E": "The table is diagonally ahead to your right, a few steps forward along the corridor." + }, + "answer_label": "D", + "answer_text": "The table in the hall is on your left.", + "q_id": "egolife_A1_013-mcq_fixed_10", + "correct_answer": "The table in the hall is on your left.", + "original_options": { + "A": "The table is on your right.", + "B": "The table is behind you.", + "C": "The table is directly in front of you, centered down the hallway as you face forward.", + "D": "The table in the hall is on your left.", + "E": "The table is diagonally ahead to your right, a few steps forward along the corridor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013-mcq_fixed_11", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the table in the hall from me?", + "answer": "The table in the hall is directly in front of you.", + "question_chinese": "大厅的桌子在我的什么方向？", + "answer_chinese": "大厅的桌子在你正前方", + "evidence": [ + { + "description": "The table in the hall is directly in front of me.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:11:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The table is behind you.", + "B": "The table is to your left.", + "C": "The table in the hall is directly in front of you.", + "D": "The table is on your right side, so you need to glance right to locate it.", + "E": "The table is diagonally behind you, near the corner wall at the back of the hallway." + }, + "answer_label": "C", + "answer_text": "The table in the hall is directly in front of you.", + "q_id": "egolife_A1_013-mcq_fixed_11", + "correct_answer": "The table in the hall is directly in front of you.", + "original_options": { + "A": "The table is behind you.", + "B": "The table is to your left.", + "C": "The table in the hall is directly in front of you.", + "D": "The table is on your right side, so you need to glance right to locate it.", + "E": "The table is diagonally behind you, near the corner wall at the back of the hallway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_5", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bedroom, how should I get there?", + "answer": "You should turn left, walk straight for about five steps until you get near the table, then turn right and walk for about four steps to enter the bedroom.", + "question_chinese": "如果我想去卧室，我应该怎么走？", + "answer_chinese": "你应该向左转，直走5步左右的距离到达桌子附近，然后向右转走4步左右的距离就能进入卧室。", + "evidence": [ + { + "description": "The bedroom is in my left front.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "The table in the living room is about two steps away from the bedroom.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "The sink beside the window is about three steps away from the central table.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:06:25", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You should turn left, walk straight for about five steps until you get near the table, then turn right and walk for about four steps to enter the bedroom.", + "B": "Turn right and walk to the bedroom.", + "C": "Turn left, then enter the first door.", + "D": "Turn left and walk about six steps to the table, then turn right and walk about three steps into the bedroom.", + "E": "Turn left, walk about five steps toward the table, then turn left and continue about four steps to reach the bedroom." + }, + "answer_label": "A", + "answer_text": "You should turn left, walk straight for about five steps until you get near the table, then turn right and walk for about four steps to enter the bedroom.", + "q_id": "egolife_A1_013_5", + "correct_answer": "You should turn left, walk straight for about five steps until you get near the table, then turn right and walk for about four steps to enter the bedroom.", + "original_options": { + "A": "You should turn left, walk straight for about five steps until you get near the table, then turn right and walk for about four steps to enter the bedroom.", + "B": "Turn right and walk to the bedroom.", + "C": "Turn left, then enter the first door.", + "D": "Turn left and walk about six steps to the table, then turn right and walk about three steps into the bedroom.", + "E": "Turn left, walk about five steps toward the table, then turn left and continue about four steps to reach the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_6", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bedroom, how should I get there?", + "answer": "You should turn right and walk two steps, then turn right again and walk about four more steps to enter the bedroom.", + "question_chinese": "如果我想去卧室，我应该怎么走？", + "answer_chinese": "你应该向右转走两步的距离，然后再向右转，再走4步左的距离就能进入卧室了。", + "evidence": [ + { + "description": "The bedroom is in my left front.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "The table in the living room is about two steps away from the bedroom.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "The sink beside the window is about three steps away from the central table.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:11:41", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You should turn right and walk two steps, then turn right again and walk about four more steps to enter the bedroom.", + "B": "Turn left, then walk into the bedroom.", + "C": "Go straight a few steps to the bedroom.", + "D": "Turn right and walk three steps, then turn left and continue about four steps to the bedroom.", + "E": "Turn right, take two steps, then turn left and walk about five steps to reach the bedroom doorway." + }, + "answer_label": "A", + "answer_text": "You should turn right and walk two steps, then turn right again and walk about four more steps to enter the bedroom.", + "q_id": "egolife_A1_013_6", + "correct_answer": "You should turn right and walk two steps, then turn right again and walk about four more steps to enter the bedroom.", + "original_options": { + "A": "You should turn right and walk two steps, then turn right again and walk about four more steps to enter the bedroom.", + "B": "Turn left, then walk into the bedroom.", + "C": "Go straight a few steps to the bedroom.", + "D": "Turn right and walk three steps, then turn left and continue about four steps to the bedroom.", + "E": "Turn right, take two steps, then turn left and walk about five steps to reach the bedroom doorway." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_7", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many laptops are on the table in the living room?", + "answer": "There is a laptop.", + "question_chinese": "客厅的桌子上有几台笔记本？", + "answer_chinese": "有一台笔记本", + "evidence": [ + { + "description": "A laptop can be seen on the table.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero laptops.", + "B": "Two laptops.", + "C": "There are three laptops on the table in the living room right now.", + "D": "There is a laptop.", + "E": "There are four laptops on the table in the living room at the moment." + }, + "answer_label": "D", + "answer_text": "There is a laptop.", + "q_id": "egolife_A1_013_7", + "correct_answer": "There is a laptop.", + "original_options": { + "A": "Zero laptops.", + "B": "Two laptops.", + "C": "There are three laptops on the table in the living room right now.", + "D": "There is a laptop.", + "E": "There are four laptops on the table in the living room at the moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_8", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many laptops are on the table in the living room?", + "answer": "There is no laptop on the living room table.", + "question_chinese": "客厅的桌子上有几台笔记本？", + "answer_chinese": "客厅的桌子上没有笔记本。", + "evidence": [ + { + "description": "A laptop can be seen on the table.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "I took the laptop from the table to the bed in the bedroom.", + "timestamps": { + "start": "00:01:14", + "end": "00:02:43" + } + } + ], + "question_timestamps": "00:10:48", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One laptop.", + "B": "Two laptops.", + "C": "There are three laptops on the living room table right now, placed side by side.", + "D": "There are four laptops on the living room table at the moment, arranged in a row.", + "E": "There is no laptop on the living room table." + }, + "answer_label": "E", + "answer_text": "There is no laptop on the living room table.", + "q_id": "egolife_A1_013_8", + "correct_answer": "There is no laptop on the living room table.", + "original_options": { + "A": "One laptop.", + "B": "Two laptops.", + "C": "There are three laptops on the living room table right now, placed side by side.", + "D": "There are four laptops on the living room table at the moment, arranged in a row.", + "E": "There is no laptop on the living room table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_4", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the door to the second floor closer to me, or is the door to the bedroom closer?", + "answer": "You are closer to the door near the bedroom.", + "question_chinese": "是通往二楼的门,还是通往卧室的门离我更近?", + "answer_chinese": "你离靠近卧室的门更近。", + "evidence": [ + { + "description": "The door leading to the second floor is in my left front.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The door in my left front is the one leading to the bedroom.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:03:42", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The door leading to the second floor is closer to you.", + "B": "You are closer to the door near the bedroom." + }, + "answer_label": "B", + "answer_text": "You are closer to the door near the bedroom.", + "q_id": "egolife_A1_013_4", + "correct_answer": "You are closer to the door near the bedroom.", + "original_options": { + "A": "The door leading to the second floor is closer to you.", + "B": "You are closer to the door near the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_3", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the door to the second floor closer to me, or is the door to the bedroom closer?", + "answer": "The door leading to the second floor is closer to you.", + "question_chinese": "是通往二楼的门,还是通往卧室的门离我更近?", + "answer_chinese": "通往2楼的门离你更近", + "evidence": [ + { + "description": "The door leading to the second floor is in my left front.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The door in my left front is the one leading to the bedroom.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:05:53", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You are closer to the door near the bedroom.", + "B": "The door leading to the second floor is closer to you." + }, + "answer_label": "B", + "answer_text": "The door leading to the second floor is closer to you.", + "q_id": "egolife_A1_013_3", + "correct_answer": "The door leading to the second floor is closer to you.", + "original_options": { + "A": "You are closer to the door near the bedroom.", + "B": "The door leading to the second floor is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the whiteboard from me?", + "answer": "The whiteboard is directly behind you.", + "question_chinese": "白板在我的什么方向？", + "answer_chinese": "白板在你的正后方。", + "evidence": [ + { + "description": "The whiteboard is directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:06:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Directly in front of you.", + "B": "The whiteboard is directly behind you.", + "C": "To your left side.", + "D": "The whiteboard is slightly ahead of you on the right, not behind your current position.", + "E": "The whiteboard is behind you but offset to your left, rather than directly behind you." + }, + "answer_label": "B", + "answer_text": "The whiteboard is directly behind you.", + "q_id": "egolife_A1_013_1", + "correct_answer": "The whiteboard is directly behind you.", + "original_options": { + "A": "Directly in front of you.", + "B": "The whiteboard is directly behind you.", + "C": "To your left side.", + "D": "The whiteboard is slightly ahead of you on the right, not behind your current position.", + "E": "The whiteboard is behind you but offset to your left, rather than directly behind you." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the whiteboard from me?", + "answer": "The whiteboard is in front of my left side.", + "question_chinese": "白板在我的什么方向？", + "answer_chinese": "白板在我的左前方", + "evidence": [ + { + "description": "The whiteboard is directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:11:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Directly in front of you.", + "B": "Behind you on the right.", + "C": "The whiteboard is directly to your left side, not positioned in front of your left side.", + "D": "The whiteboard is in front of my left side.", + "E": "The whiteboard is behind you to the left, rather than ahead and left of your position." + }, + "answer_label": "D", + "answer_text": "The whiteboard is in front of my left side.", + "q_id": "egolife_A1_013_2", + "correct_answer": "The whiteboard is in front of my left side.", + "original_options": { + "A": "Directly in front of you.", + "B": "Behind you on the right.", + "C": "The whiteboard is directly to your left side, not positioned in front of your left side.", + "D": "The whiteboard is in front of my left side.", + "E": "The whiteboard is behind you to the left, rather than ahead and left of your position." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_5_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the window relative to me?", + "answer": "The window is on the right side.", + "question_chinese": "窗户在我的什么方向？", + "answer_chinese": "窗户在你的右侧", + "evidence": [ + { + "description": "The window is directly in front of me.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:04:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The window is on your left side, not on the right side where you’re facing.", + "D": "The window is straight ahead of you, rather than positioned off to your right side.", + "E": "The window is on the right side." + }, + "answer_label": "E", + "answer_text": "The window is on the right side.", + "q_id": "egolife_A1_013_5_T1", + "correct_answer": "The window is on the right side.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The window is on your left side, not on the right side where you’re facing.", + "D": "The window is straight ahead of you, rather than positioned off to your right side.", + "E": "The window is on the right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_5_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the window relative to me?", + "answer": "The window is on the right side.", + "question_chinese": "窗户在我的什么方向？", + "answer_chinese": "窗户在你的右侧", + "evidence": [ + { + "description": "The window is directly in front of me.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:04:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is on the right side.", + "B": "It’s on your left.", + "C": "It’s behind you.", + "D": "The window is directly in front of you, not off to either side as you look forward.", + "E": "The window is on your left side, so you would turn slightly left to face it." + }, + "answer_label": "A", + "answer_text": "The window is on the right side.", + "q_id": "egolife_A1_013_5_T2", + "correct_answer": "The window is on the right side.", + "original_options": { + "A": "The window is on the right side.", + "B": "It’s on your left.", + "C": "It’s behind you.", + "D": "The window is directly in front of you, not off to either side as you look forward.", + "E": "The window is on your left side, so you would turn slightly left to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_6_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the window relative to me?", + "answer": "The window is to your back-left.", + "question_chinese": "窗户在我的什么方向？", + "answer_chinese": "窗户在你的左后方。", + "evidence": [ + { + "description": "The window is directly in front of me.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:11:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your front-right.", + "B": "It’s to your back-right.", + "C": "The window is to your back-left.", + "D": "The window is positioned in front of you on the left, closer to your forward-left side.", + "E": "The window is straight behind you, centered rather than offset toward your left shoulder." + }, + "answer_label": "C", + "answer_text": "The window is to your back-left.", + "q_id": "egolife_A1_013_6_T1", + "correct_answer": "The window is to your back-left.", + "original_options": { + "A": "It’s to your front-right.", + "B": "It’s to your back-right.", + "C": "The window is to your back-left.", + "D": "The window is positioned in front of you on the left, closer to your forward-left side.", + "E": "The window is straight behind you, centered rather than offset toward your left shoulder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_013", + "video_length": "00:12:00", + "qid": "egolife_A1_013_6_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the window relative to me?", + "answer": "The window is to your back-left.", + "question_chinese": "窗户在我的什么方向？", + "answer_chinese": "窗户在你的左后方。", + "evidence": [ + { + "description": "The window is directly in front of me.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:11:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your front-left.", + "B": "The window is to your back-left.", + "C": "It’s on your right side.", + "D": "The window is to your back-right, sitting behind you but offset toward your right shoulder.", + "E": "The window is directly in front of you, so you’d face it without turning around." + }, + "answer_label": "B", + "answer_text": "The window is to your back-left.", + "q_id": "egolife_A1_013_6_T2", + "correct_answer": "The window is to your back-left.", + "original_options": { + "A": "It’s to your front-left.", + "B": "The window is to your back-left.", + "C": "It’s on your right side.", + "D": "The window is to your back-right, sitting behind you but offset toward your right shoulder.", + "E": "The window is directly in front of you, so you’d face it without turning around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + } +] diff --git a/QAs/egolife_A1_014-mcq.json b/QAs/egolife_A1_014-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d77d49a3a219fc8ada9410582f31241ad399572b --- /dev/null +++ b/QAs/egolife_A1_014-mcq.json @@ -0,0 +1,1213 @@ +[ + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_6_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the chessboard from me?", + "answer": "The chessboard is to your front-left.", + "question_chinese": "国际象棋的棋盘在我的什么方向？", + "answer_chinese": "棋盘在你的左前方。", + "evidence": [ + { + "description": "It's on your left front.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your front-right.", + "B": "It’s behind you.", + "C": "The chessboard is to your back-left, meaning it’s behind your left shoulder rather than ahead.", + "D": "The chessboard is to your front-left.", + "E": "The chessboard is behind you on the right, so you’d turn around and look to your right." + }, + "answer_label": "D", + "answer_text": "The chessboard is to your front-left.", + "q_id": "egolife_A1_014_6_T1", + "correct_answer": "The chessboard is to your front-left.", + "original_options": { + "A": "It’s to your front-right.", + "B": "It’s behind you.", + "C": "The chessboard is to your back-left, meaning it’s behind your left shoulder rather than ahead.", + "D": "The chessboard is to your front-left.", + "E": "The chessboard is behind you on the right, so you’d turn around and look to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_6_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the chessboard from me?", + "answer": "The chessboard is behind you.", + "question_chinese": "国际象棋的棋盘在我的什么方向？", + "answer_chinese": "棋盘在你的后面。", + "evidence": [ + { + "description": "Behind you.", + "timestamps": { + "start": "00:00:55", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The chessboard is behind you.", + "B": "It’s in front of you.", + "C": "It’s to your left.", + "D": "The chessboard is to your right side, positioned beside you rather than behind you.", + "E": "The chessboard is above you, so you would need to look upward instead of turning around." + }, + "answer_label": "A", + "answer_text": "The chessboard is behind you.", + "q_id": "egolife_A1_014_6_T2", + "correct_answer": "The chessboard is behind you.", + "original_options": { + "A": "The chessboard is behind you.", + "B": "It’s in front of you.", + "C": "It’s to your left.", + "D": "The chessboard is to your right side, positioned beside you rather than behind you.", + "E": "The chessboard is above you, so you would need to look upward instead of turning around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_11", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the chessboard relative to me?", + "answer": "The chessboard is behind you on the left.", + "question_chinese": "国际象棋的棋盘在我的什么方向？", + "answer_chinese": "棋盘在你的左后方。", + "evidence": [ + { + "description": "There is a large chessboard in front of me on the left.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:05:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s behind you on the right.", + "C": "The chessboard is beside you on the left, not behind you, so it’s aligned with your left side.", + "D": "The chessboard is behind you in the center, meaning it’s straight back rather than back-left.", + "E": "The chessboard is behind you on the left." + }, + "answer_label": "E", + "answer_text": "The chessboard is behind you on the left.", + "q_id": "egolife_A1_014_11", + "correct_answer": "The chessboard is behind you on the left.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s behind you on the right.", + "C": "The chessboard is beside you on the left, not behind you, so it’s aligned with your left side.", + "D": "The chessboard is behind you in the center, meaning it’s straight back rather than back-left.", + "E": "The chessboard is behind you on the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_12", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the chessboard relative to me?", + "answer": "The chessboard is behind you on the left.", + "question_chinese": "国际象棋的棋盘在我的什么方向？", + "answer_chinese": "棋盘在你的右前方。", + "evidence": [ + { + "description": "There is a large chessboard in front of me on the left.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:11:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "The chessboard is behind you on the left.", + "C": "It’s on your right side.", + "D": "The chessboard is in front of you on the right, positioned ahead rather than behind your left shoulder.", + "E": "The chessboard is on your left side near the table, not behind you, so it’s left-adjacent." + }, + "answer_label": "B", + "answer_text": "The chessboard is behind you on the left.", + "q_id": "egolife_A1_014_12", + "correct_answer": "The chessboard is behind you on the left.", + "original_options": { + "A": "It’s in front of you.", + "B": "The chessboard is behind you on the left.", + "C": "It’s on your right side.", + "D": "The chessboard is in front of you on the right, positioned ahead rather than behind your left shoulder.", + "E": "The chessboard is on your left side near the table, not behind you, so it’s left-adjacent." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_1_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the table in the hall from me?", + "answer": "The table in the hall is in front of you.", + "question_chinese": "大厅的桌子在我的什么方向？", + "answer_chinese": "大厅的桌子在你的正前方。", + "evidence": [ + { + "description": "Right in front of you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s behind you.", + "B": "It’s to your left.", + "C": "The table in the hall is in front of you.", + "D": "The table in the hall is to your right, placed off your right side rather than straight ahead.", + "E": "The table in the hall is above you, so you would look upward instead of forward to see it." + }, + "answer_label": "C", + "answer_text": "The table in the hall is in front of you.", + "q_id": "egolife_A1_014_1_T2", + "correct_answer": "The table in the hall is in front of you.", + "original_options": { + "A": "It’s behind you.", + "B": "It’s to your left.", + "C": "The table in the hall is in front of you.", + "D": "The table in the hall is to your right, placed off your right side rather than straight ahead.", + "E": "The table in the hall is above you, so you would look upward instead of forward to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_1_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the table in the hall from me?", + "answer": "The table in the hall is behind you.", + "question_chinese": "大厅的桌子在我的什么方向？", + "answer_chinese": "大厅的桌子在你的后面。", + "evidence": [ + { + "description": "Behind you.", + "timestamps": { + "start": "00:00:53", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "It is off to your right, near the wall as you face down the hall.", + "D": "The table in the hall is behind you.", + "E": "It is ahead of you, slightly left as you stand in the hallway." + }, + "answer_label": "D", + "answer_text": "The table in the hall is behind you.", + "q_id": "egolife_A1_014_1_T1", + "correct_answer": "The table in the hall is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "It is off to your right, near the wall as you face down the hall.", + "D": "The table in the hall is behind you.", + "E": "It is ahead of you, slightly left as you stand in the hallway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In what direction is the table in the hall from me?", + "answer": "The table in the hall is on your right front.", + "question_chinese": "大厅的桌子在我的什么方向？", + "answer_chinese": "大厅的桌子在你的右前方", + "evidence": [ + { + "description": "The living room table is directly in front of me.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:02:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The table in the hall is on your right front.", + "B": "Behind you.", + "C": "To your left.", + "D": "The table is directly in front of you as you look down the hallway.", + "E": "The table is on your right side, more beside you than in front." + }, + "answer_label": "A", + "answer_text": "The table in the hall is on your right front.", + "q_id": "egolife_A1_014_1", + "correct_answer": "The table in the hall is on your right front.", + "original_options": { + "A": "The table in the hall is on your right front.", + "B": "Behind you.", + "C": "To your left.", + "D": "The table is directly in front of you as you look down the hallway.", + "E": "The table is on your right side, more beside you than in front." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014-mcq_fixed_1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the hall table from me?", + "answer": "The table in the hall is behind you.", + "question_chinese": "大厅的桌子在我的什么方向？", + "answer_chinese": "大厅的桌子在你的后方", + "evidence": [ + { + "description": "The living room table is directly in front of me.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:07:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "The table in the hall is behind you.", + "C": "To your right.", + "D": "The hall table is off to your left as you move forward into the hallway.", + "E": "You are facing the hall table, which sits ahead of you on the right." + }, + "answer_label": "B", + "answer_text": "The table in the hall is behind you.", + "q_id": "egolife_A1_014-mcq_fixed_1", + "correct_answer": "The table in the hall is behind you.", + "original_options": { + "A": "In front of you.", + "B": "The table in the hall is behind you.", + "C": "To your right.", + "D": "The hall table is off to your left as you move forward into the hallway.", + "E": "You are facing the hall table, which sits ahead of you on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_7", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mobile phones are there in total on the coffee table?", + "answer": "There is a mobile phone on the table.", + "question_chinese": "客厅的桌子上一共有几部手机？", + "answer_chinese": "桌子上有一部手机。", + "evidence": [ + { + "description": "You can now see a mobile phone on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:05:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two mobile phones on the table.", + "B": "There is a mobile phone on the table.", + "C": "There are three mobile phones on the table.", + "D": "There are four mobile phones on the table in total, grouped together on the coffee table.", + "E": "There are five mobile phones on the table, spread out across the surface of the coffee table." + }, + "answer_label": "B", + "answer_text": "There is a mobile phone on the table.", + "q_id": "egolife_A1_014_7", + "correct_answer": "There is a mobile phone on the table.", + "original_options": { + "A": "There are two mobile phones on the table.", + "B": "There is a mobile phone on the table.", + "C": "There are three mobile phones on the table.", + "D": "There are four mobile phones on the table in total, grouped together on the coffee table.", + "E": "There are five mobile phones on the table, spread out across the surface of the coffee table." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_8", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mobile phones are there in total on the table in the living room?", + "answer": "There are two mobile phones on the table.", + "question_chinese": "客厅的桌子上一共有几部手机？", + "answer_chinese": "桌子上有两部手机。", + "evidence": [ + { + "description": "You can now see a mobile phone on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + }, + { + "description": "I put the mobile phone in my hand on the table.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:26" + } + } + ], + "question_timestamps": "00:06:24", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There is one mobile phone on the table.", + "B": "There are three mobile phones on the table.", + "C": "There are four mobile phones on the table in total, with more than one placed close together.", + "D": "There are five mobile phones on the table in the living room, scattered across the tabletop.", + "E": "There are two mobile phones on the table." + }, + "answer_label": "E", + "answer_text": "There are two mobile phones on the table.", + "q_id": "egolife_A1_014_8", + "correct_answer": "There are two mobile phones on the table.", + "original_options": { + "A": "There is one mobile phone on the table.", + "B": "There are three mobile phones on the table.", + "C": "There are four mobile phones on the table in total, with more than one placed close together.", + "D": "There are five mobile phones on the table in the living room, scattered across the tabletop.", + "E": "There are two mobile phones on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_4_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What am I holding in my hand?", + "answer": "You are holding a marker in your hand.", + "question_chinese": "我手上拿是什么？", + "answer_chinese": "你手上拿着一只记号笔。", + "evidence": [ + { + "description": "You have a marker in your hand.", + "timestamps": { + "start": "00:03:04", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:03:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You are holding a pen in your hand.", + "B": "You are holding scissors in your hand.", + "C": "You are holding a mobile phone in your hand, gripping it in front of you.", + "D": "You are holding a stapler in your hand, with your fingers wrapped around it.", + "E": "You are holding a marker in your hand." + }, + "answer_label": "E", + "answer_text": "You are holding a marker in your hand.", + "q_id": "egolife_A1_014_4_T1", + "correct_answer": "You are holding a marker in your hand.", + "original_options": { + "A": "You are holding a pen in your hand.", + "B": "You are holding scissors in your hand.", + "C": "You are holding a mobile phone in your hand, gripping it in front of you.", + "D": "You are holding a stapler in your hand, with your fingers wrapped around it.", + "E": "You are holding a marker in your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_4_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What am I holding in my hand?", + "answer": "You are holding a pair of scissors in your hand.", + "question_chinese": "我手上拿是什么？", + "answer_chinese": "你手上拿着一把剪刀。", + "evidence": [ + { + "description": "You have a pair of scissors in your hand.", + "timestamps": { + "start": "00:03:32", + "end": "00:03:34" + } + } + ], + "question_timestamps": "00:03:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You are holding a pair of scissors in your hand.", + "B": "You are holding a marker in your hand.", + "C": "You are holding a pen in your hand.", + "D": "You are holding a mobile phone in your hand, with the screen facing up toward you.", + "E": "You are holding a ruler in your hand, pinched between your fingers as you look down." + }, + "answer_label": "A", + "answer_text": "You are holding a pair of scissors in your hand.", + "q_id": "egolife_A1_014_4_T2", + "correct_answer": "You are holding a pair of scissors in your hand.", + "original_options": { + "A": "You are holding a pair of scissors in your hand.", + "B": "You are holding a marker in your hand.", + "C": "You are holding a pen in your hand.", + "D": "You are holding a mobile phone in your hand, with the screen facing up toward you.", + "E": "You are holding a ruler in your hand, pinched between your fingers as you look down." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_3_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What is in front of me?", + "answer": "In front of you is a sink.", + "question_chinese": "我的前面是什么？", + "answer_chinese": "你的前面是水槽。", + "evidence": [ + { + "description": "In front of you is the sink.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In front of you is a door.", + "B": "In front of you is a table.", + "C": "In front of you is a chair, positioned directly across from where you are standing.", + "D": "In front of you is a sink.", + "E": "In front of you is a whiteboard, taking up most of the wall you are facing." + }, + "answer_label": "D", + "answer_text": "In front of you is a sink.", + "q_id": "egolife_A1_014_3_T1", + "correct_answer": "In front of you is a sink.", + "original_options": { + "A": "In front of you is a door.", + "B": "In front of you is a table.", + "C": "In front of you is a chair, positioned directly across from where you are standing.", + "D": "In front of you is a sink.", + "E": "In front of you is a whiteboard, taking up most of the wall you are facing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_3_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What is in front of me?", + "answer": "In front of you is a whiteboard.", + "question_chinese": "我的前面是什么？", + "answer_chinese": "你的前面是一块白板。", + "evidence": [ + { + "description": "In front of you is a whiteboard.", + "timestamps": { + "start": "00:00:55", + "end": "00:00:56" + } + } + ], + "question_timestamps": "00:05:58", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "A sink is in front of you.", + "B": "A table is in front of you.", + "C": "In front of you is a door, taking up most of the view straight ahead.", + "D": "In front of you is a chair, positioned directly ahead where you're facing.", + "E": "In front of you is a whiteboard." + }, + "answer_label": "E", + "answer_text": "In front of you is a whiteboard.", + "q_id": "egolife_A1_014_3_T2", + "correct_answer": "In front of you is a whiteboard.", + "original_options": { + "A": "A sink is in front of you.", + "B": "A table is in front of you.", + "C": "In front of you is a door, taking up most of the view straight ahead.", + "D": "In front of you is a chair, positioned directly ahead where you're facing.", + "E": "In front of you is a whiteboard." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_9_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mobile phones are on my table right now?", + "answer": "There is one phone on your table right now, and the phone is in front of you.", + "question_chinese": "我的桌子上现在有几部手机？", + "answer_chinese": "你的桌子上现在有一部手机,手机在你的正前方。", + "evidence": [ + { + "description": "You can now see a mobile phone on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero phones are on your table.", + "B": "Two phones are on your table.", + "C": "There are three phones on your table right now, and all three phones are in front of you.", + "D": "There are four phones on your table right now, and all four phones are in front of you.", + "E": "There is one phone on your table right now, and the phone is in front of you." + }, + "answer_label": "E", + "answer_text": "There is one phone on your table right now, and the phone is in front of you.", + "q_id": "egolife_A1_014_9_T1", + "correct_answer": "There is one phone on your table right now, and the phone is in front of you.", + "original_options": { + "A": "Zero phones are on your table.", + "B": "Two phones are on your table.", + "C": "There are three phones on your table right now, and all three phones are in front of you.", + "D": "There are four phones on your table right now, and all four phones are in front of you.", + "E": "There is one phone on your table right now, and the phone is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_9_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mobile phones are on my table right now?", + "answer": "There are two phones on your table right now, and the phones are behind you.", + "question_chinese": "我的桌子上现在有几部手机？", + "answer_chinese": "你的桌子上现在有两部手机，手机在你的后面。", + "evidence": [ + { + "description": "I put the mobile phone in my hand on the table.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:26" + } + } + ], + "question_timestamps": "00:05:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are two phones on your table right now, and the phones are behind you.", + "B": "Zero phones are on your table.", + "C": "One phone is on your table.", + "D": "There are three phones on your table right now, and the phones are behind you.", + "E": "There are four phones on your table right now, and the phones are behind you." + }, + "answer_label": "A", + "answer_text": "There are two phones on your table right now, and the phones are behind you.", + "q_id": "egolife_A1_014_9_T2", + "correct_answer": "There are two phones on your table right now, and the phones are behind you.", + "original_options": { + "A": "There are two phones on your table right now, and the phones are behind you.", + "B": "Zero phones are on your table.", + "C": "One phone is on your table.", + "D": "There are three phones on your table right now, and the phones are behind you.", + "E": "There are four phones on your table right now, and the phones are behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_8_T1", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which door is closer to me right now—the door to the bedroom, or the door to the second floor?", + "answer": "The door leading to the bedroom is closer to you.", + "question_chinese": "是通往卧室的门离我更近，还是通往二楼的门离我更近?", + "answer_chinese": "通往卧室的门离你更近。", + "evidence": [ + { + "description": "Directly in front of me is the door leading to the bedroom.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "The door leading to the second floor is on my right front.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:04:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The door leading to the second floor is closer to you.", + "B": "The door leading to the bedroom is closer to you." + }, + "answer_label": "B", + "answer_text": "The door leading to the bedroom is closer to you.", + "q_id": "egolife_A1_014_8_T1", + "correct_answer": "The door leading to the bedroom is closer to you.", + "original_options": { + "A": "The door leading to the second floor is closer to you.", + "B": "The door leading to the bedroom is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_8_T2", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which door is closer to me right now—the door to the bedroom, or the door to the second floor?", + "answer": "The door leading to the second floor is closer to you.", + "question_chinese": "是通往卧室的门离我更近，还是通往二楼的门离我更近?", + "answer_chinese": "通往2楼的门离你更近", + "evidence": [ + { + "description": "Directly in front of me is the door leading to the bedroom.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "The door leading to the second floor is on my right front.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:11:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The door leading to the bedroom is closer to you.", + "B": "The door leading to the second floor is closer to you." + }, + "answer_label": "B", + "answer_text": "The door leading to the second floor is closer to you.", + "q_id": "egolife_A1_014_8_T2", + "correct_answer": "The door leading to the second floor is closer to you.", + "original_options": { + "A": "The door leading to the bedroom is closer to you.", + "B": "The door leading to the second floor is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_7_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the oven relative to me?", + "answer": "The oven is behind you.", + "question_chinese": "烤箱相对于我在什么位置？", + "answer_chinese": "烤箱在你的后面。", + "evidence": [ + { + "description": "The oven is behind you.", + "timestamps": { + "start": "00:06:35", + "end": "00:06:38" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The oven is in front of you.", + "B": "The oven is behind you.", + "C": "The oven is to your left.", + "D": "The oven is to your right, positioned off to the side as you face forward.", + "E": "The oven is above you, higher than your eye level in your current view." + }, + "answer_label": "B", + "answer_text": "The oven is behind you.", + "q_id": "egolife_A1_014_7_T2", + "correct_answer": "The oven is behind you.", + "original_options": { + "A": "The oven is in front of you.", + "B": "The oven is behind you.", + "C": "The oven is to your left.", + "D": "The oven is to your right, positioned off to the side as you face forward.", + "E": "The oven is above you, higher than your eye level in your current view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_7_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the oven relative to me?", + "answer": "The oven is to your right.", + "question_chinese": "烤箱相对于我在什么位置？", + "answer_chinese": "烤箱在你的右边。", + "evidence": [ + { + "description": "You are washing your hands now..", + "timestamps": { + "start": "00:07:40", + "end": "00:07:44" + } + } + ], + "question_timestamps": "00:09:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The oven is in front of you.", + "B": "The oven is behind you.", + "C": "The oven is to your right.", + "D": "The oven is to your left, on the opposite side from where you're currently facing.", + "E": "The oven is above you, positioned higher up rather than beside you." + }, + "answer_label": "C", + "answer_text": "The oven is to your right.", + "q_id": "egolife_A1_014_7_T1", + "correct_answer": "The oven is to your right.", + "original_options": { + "A": "The oven is in front of you.", + "B": "The oven is behind you.", + "C": "The oven is to your right.", + "D": "The oven is to your left, on the opposite side from where you're currently facing.", + "E": "The oven is above you, positioned higher up rather than beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 12 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_9", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the whiteboard from me?", + "answer": "The whiteboard is behind you on the right.", + "question_chinese": "白板在我的什么方向？", + "answer_chinese": "白板在你的右后方", + "evidence": [ + { + "description": "The whiteboard is directly in front of me.", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:06:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you on the right.", + "B": "Behind you on the left.", + "C": "The whiteboard is behind you on the right.", + "D": "The whiteboard is in front of you on the left, visible when you look slightly ahead and leftward.", + "E": "The whiteboard is behind you in the center, directly back from your position rather than off to the right." + }, + "answer_label": "C", + "answer_text": "The whiteboard is behind you on the right.", + "q_id": "egolife_A1_014_9", + "correct_answer": "The whiteboard is behind you on the right.", + "original_options": { + "A": "In front of you on the right.", + "B": "Behind you on the left.", + "C": "The whiteboard is behind you on the right.", + "D": "The whiteboard is in front of you on the left, visible when you look slightly ahead and leftward.", + "E": "The whiteboard is behind you in the center, directly back from your position rather than off to the right." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_014", + "video_length": "00:12:00", + "qid": "egolife_A1_014_10", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the whiteboard from me?", + "answer": "The whiteboard is on your right.", + "question_chinese": "白板在我的什么方向？", + "answer_chinese": "白板在你的右侧。", + "evidence": [ + { + "description": "The whiteboard is directly in front of me.", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:10:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your left.", + "B": "The whiteboard is on your right.", + "C": "It is straight ahead of you.", + "D": "The whiteboard is behind you, slightly off to your left as you look around.", + "E": "The whiteboard is in front of you on the wall, not to either side." + }, + "answer_label": "B", + "answer_text": "The whiteboard is on your right.", + "q_id": "egolife_A1_014_10", + "correct_answer": "The whiteboard is on your right.", + "original_options": { + "A": "It is on your left.", + "B": "The whiteboard is on your right.", + "C": "It is straight ahead of you.", + "D": "The whiteboard is behind you, slightly off to your left as you look around.", + "E": "The whiteboard is in front of you on the wall, not to either side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 14 + ] + } +] diff --git a/QAs/egolife_A1_015-mcq.json b/QAs/egolife_A1_015-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..20b05a097b41620a30621e1beb94c0840b199ac6 --- /dev/null +++ b/QAs/egolife_A1_015-mcq.json @@ -0,0 +1,1194 @@ +[ + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_2_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go into the courtyard, how should I get there?", + "answer": "You should turn left and walk about two steps, then turn right into the hallway door; after that, turn left and go through the hallway to enter the courtyard.", + "question_chinese": "如果我想去庭院里面，应该怎么走？", + "answer_chinese": "你应该向左转走两步的距离，然后右转进入走廊的门，然后左转通过走廊就可以进入到庭院了。", + "evidence": [ + { + "description": "I can see the courtyard directly in front of me through the glass.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The distance from the pool to the door leading to the courtyard is about four steps.", + "timestamps": { + "start": "00:05:04", + "end": "00:05:05" + } + }, + { + "description": "You can see that the corridor leading to the courtyard through the door extends to the left.You can see that the corridor leading to the courtyard through the door extends to the left.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right, then right.", + "B": "Turn left, then left.", + "C": "You should turn left and walk about two steps, then turn right into the hallway door; after that, turn left and go through the hallway to enter the courtyard.", + "D": "Turn left about two steps, then turn left into the hallway door and continue straight toward the courtyard.", + "E": "Turn left and walk about two steps, enter the hallway door straight ahead, then turn right to reach the courtyard." + }, + "answer_label": "C", + "answer_text": "You should turn left and walk about two steps, then turn right into the hallway door; after that, turn left and go through the hallway to enter the courtyard.", + "q_id": "egolife_A1_015_2_T1", + "correct_answer": "You should turn left and walk about two steps, then turn right into the hallway door; after that, turn left and go through the hallway to enter the courtyard.", + "original_options": { + "A": "Turn right, then right.", + "B": "Turn left, then left.", + "C": "You should turn left and walk about two steps, then turn right into the hallway door; after that, turn left and go through the hallway to enter the courtyard.", + "D": "Turn left about two steps, then turn left into the hallway door and continue straight toward the courtyard.", + "E": "Turn left and walk about two steps, enter the hallway door straight ahead, then turn right to reach the courtyard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_2_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go into the courtyard, how should I get there?", + "answer": "You should turn right into the hallway door, then turn left and go through the hallway to enter the courtyard.", + "question_chinese": "如果我想去庭院里面，应该怎么走？", + "answer_chinese": "你应该右转进入走廊的门，然后左转通过走廊就可以进入到庭院了。", + "evidence": [ + { + "description": "I can see the courtyard directly in front of me through the glass.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The distance from the pool to the door leading to the courtyard is about four steps.", + "timestamps": { + "start": "00:05:04", + "end": "00:05:05" + } + }, + { + "description": "You can see that the corridor leading to the courtyard through the door extends to the left.You can see that the corridor leading to the courtyard through the door extends to the left.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:10:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left into the hallway door.", + "B": "You should turn right into the hallway door, then turn left and go through the hallway to enter the courtyard.", + "C": "Turn right, then turn right.", + "D": "Turn right into the hallway door, then go straight down the hallway without turning to reach the courtyard.", + "E": "Turn right into the hallway door, then make a right turn and follow the corridor until you reach the courtyard." + }, + "answer_label": "B", + "answer_text": "You should turn right into the hallway door, then turn left and go through the hallway to enter the courtyard.", + "q_id": "egolife_A1_015_2_T2", + "correct_answer": "You should turn right into the hallway door, then turn left and go through the hallway to enter the courtyard.", + "original_options": { + "A": "Turn left into the hallway door.", + "B": "You should turn right into the hallway door, then turn left and go through the hallway to enter the courtyard.", + "C": "Turn right, then turn right.", + "D": "Turn right into the hallway door, then go straight down the hallway without turning to reach the courtyard.", + "E": "Turn right into the hallway door, then make a right turn and follow the corridor until you reach the courtyard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_1_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go into the courtyard, how should I walk?", + "answer": "Walk about four steps to your front-left to reach the door leading to the courtyard, then turn left; go through the hallway and you'll reach the courtyard.", + "question_chinese": "如果我想去庭院里面，我应该怎么走？", + "answer_chinese": "向左前方走4步左右的距离到达通往庭院的门，然后向左转，穿过走廊就能到达庭院了。", + "evidence": [ + { + "description": "I can see the courtyard directly in front of me through the glass.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The distance from the pool to the door leading to the courtyard is about four steps.", + "timestamps": { + "start": "00:05:04", + "end": "00:05:05" + } + } + ], + "question_timestamps": "00:05:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk about four steps to your front-left to reach the door leading to the courtyard, then turn left; go through the hallway and you'll reach the courtyard.", + "B": "Walk four steps front-right.", + "C": "Walk four steps back-left.", + "D": "Walk about four steps to your front-left to reach the door, then turn right and continue through the hallway.", + "E": "Walk about three steps to your front-left to the courtyard door, then turn left and go through the hallway." + }, + "answer_label": "A", + "answer_text": "Walk about four steps to your front-left to reach the door leading to the courtyard, then turn left; go through the hallway and you'll reach the courtyard.", + "q_id": "egolife_A1_015_1_T2", + "correct_answer": "Walk about four steps to your front-left to reach the door leading to the courtyard, then turn left; go through the hallway and you'll reach the courtyard.", + "original_options": { + "A": "Walk about four steps to your front-left to reach the door leading to the courtyard, then turn left; go through the hallway and you'll reach the courtyard.", + "B": "Walk four steps front-right.", + "C": "Walk four steps back-left.", + "D": "Walk about four steps to your front-left to reach the door, then turn right and continue through the hallway.", + "E": "Walk about three steps to your front-left to the courtyard door, then turn left and go through the hallway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_1_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go into the courtyard, how should I walk?", + "answer": "You should turn around, walk about four steps to your front-left to reach the door leading to the courtyard, then turn left; go through the hallway and you'll reach the courtyard.", + "question_chinese": "如果我想去庭院里面，我应该怎么走？", + "answer_chinese": "你应该调头，向左前方走4步左右的距离到达通往庭院的门，然后向左转，穿过走廊就能到达庭院了。", + "evidence": [ + { + "description": "I can see the courtyard directly in front of me through the glass.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The distance from the pool to the door leading to the courtyard is about four steps.", + "timestamps": { + "start": "00:05:04", + "end": "00:05:05" + } + }, + { + "description": "You can see a door leading to the outside on the left.The distance from the whiteboard to the door leading to the courtyard is about two steps.You can see that the corridor leading to the courtyard through the door extends to the left.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:08:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around, then go front-right.", + "B": "Turn around, then go back-left.", + "C": "Without turning around, walk about four steps to your front-left to the door, then turn left through the hallway.", + "D": "You should turn around, walk about four steps to your front-left to reach the door leading to the courtyard, then turn left; go through the hallway and you'll reach the courtyard.", + "E": "Turn around and walk about four steps to your front-left, but turn right at the door and continue through the hallway." + }, + "answer_label": "D", + "answer_text": "You should turn around, walk about four steps to your front-left to reach the door leading to the courtyard, then turn left; go through the hallway and you'll reach the courtyard.", + "q_id": "egolife_A1_015_1_T1", + "correct_answer": "You should turn around, walk about four steps to your front-left to reach the door leading to the courtyard, then turn left; go through the hallway and you'll reach the courtyard.", + "original_options": { + "A": "Turn around, then go front-right.", + "B": "Turn around, then go back-left.", + "C": "Without turning around, walk about four steps to your front-left to the door, then turn left through the hallway.", + "D": "You should turn around, walk about four steps to your front-left to reach the door leading to the courtyard, then turn left; go through the hallway and you'll reach the courtyard.", + "E": "Turn around and walk about four steps to your front-left, but turn right at the door and continue through the hallway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_4", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the courtyard, how should I get there?", + "answer": "You should turn left and walk two steps, then turn right to enter the corridor door. After that, turn left again and go through the corridor to get into the courtyard.", + "question_chinese": "如果我想去庭院里面，我应该怎么走？", + "answer_chinese": "你应该向左转走两步的距离，然后右转进入走廊的门，然后左转通过走廊就可以进入到庭院了。", + "evidence": [ + { + "description": "I can see the courtyard directly in front of me through the glass.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You can see a door leading to the outside on the left.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:08" + } + }, + { + "description": "The distance from the pool to the door leading to the courtyard is about four steps.", + "timestamps": { + "start": "00:05:04", + "end": "00:05:08" + } + }, + { + "description": "The distance from the whiteboard to the door leading to the courtyard is about two steps.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:08" + } + }, + { + "description": "You can see that the corridor leading to the courtyard through the door extends to the left.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You should turn left and walk two steps, then turn right to enter the corridor door. After that, turn left again and go through the corridor to get into the courtyard.", + "B": "Turn right and walk two steps.", + "C": "Go straight two steps, then left.", + "D": "Turn left two steps, then turn left into the corridor door; then go through the corridor to the courtyard.", + "E": "Turn left and walk three steps, then turn right into the corridor door; then turn right and continue through the corridor." + }, + "answer_label": "A", + "answer_text": "You should turn left and walk two steps, then turn right to enter the corridor door. After that, turn left again and go through the corridor to get into the courtyard.", + "q_id": "egolife_A1_015_4", + "correct_answer": "You should turn left and walk two steps, then turn right to enter the corridor door. After that, turn left again and go through the corridor to get into the courtyard.", + "original_options": { + "A": "You should turn left and walk two steps, then turn right to enter the corridor door. After that, turn left again and go through the corridor to get into the courtyard.", + "B": "Turn right and walk two steps.", + "C": "Go straight two steps, then left.", + "D": "Turn left two steps, then turn left into the corridor door; then go through the corridor to the courtyard.", + "E": "Turn left and walk three steps, then turn right into the corridor door; then turn right and continue through the corridor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015-mcq_fixed_0", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What direction is the living room table from me?", + "answer": "The table is behind your left side.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "桌子在你的左后方。", + "evidence": [ + { + "description": "The table in the living room is in front of my right side.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:06:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is behind your right side.", + "C": "The table is behind your left side.", + "D": "The table is in front of your left side, not behind you as you stand there.", + "E": "The table is slightly to your right and ahead of you, closer than anything behind your left." + }, + "answer_label": "C", + "answer_text": "The table is behind your left side.", + "q_id": "egolife_A1_015-mcq_fixed_0", + "correct_answer": "The table is behind your left side.", + "original_options": { + "A": "It is in front of you.", + "B": "It is behind your right side.", + "C": "The table is behind your left side.", + "D": "The table is in front of your left side, not behind you as you stand there.", + "E": "The table is slightly to your right and ahead of you, closer than anything behind your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015-mcq_fixed_1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What direction is the living room table from me?", + "answer": "The table is directly behind you.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "桌子在你的正后方。", + "evidence": [ + { + "description": "The table in the living room is in front of my right side.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is directly in front of you.", + "B": "The table is directly behind you.", + "C": "It is directly to your right.", + "D": "The table is slightly behind you, but shifted toward your left rather than centered directly behind.", + "E": "The table is ahead of you near the middle of the room, not positioned behind you." + }, + "answer_label": "B", + "answer_text": "The table is directly behind you.", + "q_id": "egolife_A1_015-mcq_fixed_1", + "correct_answer": "The table is directly behind you.", + "original_options": { + "A": "It is directly in front of you.", + "B": "The table is directly behind you.", + "C": "It is directly to your right.", + "D": "The table is slightly behind you, but shifted toward your left rather than centered directly behind.", + "E": "The table is ahead of you near the middle of the room, not positioned behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_9_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many types of iron utensils did I see in the sink?", + "answer": "You saw two types of iron utensils.", + "question_chinese": "我在水池里看见几种铁质器皿？", + "answer_chinese": "你看见两种铁质器皿。", + "evidence": [ + { + "description": "From the current perspective, there are three types of ironware in total: bowls, iron trays, and iron steamers.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "I cleaned the iron steamer and took it out of the pool.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:38" + } + }, + { + "description": "I cleaned the iron tray and took it out of the pool.But there's still an iron tray left at the bottom of the pool.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:04:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two types of iron utensils.", + "B": "You saw one type of iron utensil.", + "C": "You saw three types of iron utensils.", + "D": "You saw four different types of iron utensils sitting together in the sink.", + "E": "You did not see any iron utensil types in the sink at all." + }, + "answer_label": "A", + "answer_text": "You saw two types of iron utensils.", + "q_id": "egolife_A1_015_9_T1", + "correct_answer": "You saw two types of iron utensils.", + "original_options": { + "A": "You saw two types of iron utensils.", + "B": "You saw one type of iron utensil.", + "C": "You saw three types of iron utensils.", + "D": "You saw four different types of iron utensils sitting together in the sink.", + "E": "You did not see any iron utensil types in the sink at all." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_9_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many types of iron utensils did I see in the sink?", + "answer": "You saw one iron utensil.", + "question_chinese": "我在水池里看见几种铁质器皿？", + "answer_chinese": "你看见一样铁制器皿。", + "evidence": [ + { + "description": "From the current perspective, there are three types of ironware in total: bowls, iron trays, and iron steamers.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "I cleaned the iron steamer and took it out of the pool.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:38" + } + }, + { + "description": "I cleaned the iron tray and took it out of the pool.But there's still an iron tray left at the bottom of the pool.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + }, + { + "description": "I have finished cleaning all the iron bowls and taken them all out of the pool.", + "timestamps": { + "start": "00:04:50", + "end": "00:05:01" + } + } + ], + "question_timestamps": "00:11:49", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw zero iron utensils.", + "B": "You saw one iron utensil.", + "C": "You saw two iron utensils.", + "D": "You saw three iron utensils in the sink when you looked down.", + "E": "You saw four iron utensils clustered together in the sink basin at once." + }, + "answer_label": "B", + "answer_text": "You saw one iron utensil.", + "q_id": "egolife_A1_015_9_T2", + "correct_answer": "You saw one iron utensil.", + "original_options": { + "A": "You saw zero iron utensils.", + "B": "You saw one iron utensil.", + "C": "You saw two iron utensils.", + "D": "You saw three iron utensils in the sink when you looked down.", + "E": "You saw four iron utensils clustered together in the sink basin at once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_7", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the table in the living room closer to me, or am I closer to the sink?", + "answer": "The pool is closer to you.", + "question_chinese": "是客厅里的桌子离我更近，还是水池里我更近？", + "answer_chinese": "水池离你更近。", + "evidence": [ + { + "description": "The pool is directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "The table is directly in front of me.", + "timestamps": { + "start": "00:00:56", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:06:37", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The table in the living room is closer to you.", + "B": "The pool is closer to you." + }, + "answer_label": "B", + "answer_text": "The pool is closer to you.", + "q_id": "egolife_A1_015_7", + "correct_answer": "The pool is closer to you.", + "original_options": { + "A": "The table in the living room is closer to you.", + "B": "The pool is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_8", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the table in the living room closer to me, or am I closer to the sink?", + "answer": "You are closer to the table now.", + "question_chinese": "是客厅里的桌子离我更近，还是水池里我更近？", + "answer_chinese": "你现在离桌子更近。", + "evidence": [ + { + "description": "The pool is directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "The table is directly in front of me.", + "timestamps": { + "start": "00:00:56", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:10:05", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You are closer to the table now.", + "B": "The sink is closer to you." + }, + "answer_label": "A", + "answer_text": "You are closer to the table now.", + "q_id": "egolife_A1_015_8", + "correct_answer": "You are closer to the table now.", + "original_options": { + "A": "You are closer to the table now.", + "B": "The sink is closer to you." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_5_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the chessboard from me?", + "answer": "The chessboard is in front of you.", + "question_chinese": "棋盘在我的什么方向？", + "answer_chinese": "棋盘在你的前方。", + "evidence": [ + { + "description": "You can see a large chessboard directly in front of me now.", + "timestamps": { + "start": "00:00:57", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:01:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The chessboard is behind you.", + "B": "The chessboard is in front of you.", + "C": "The chessboard is to your left.", + "D": "The chessboard is to your right side, not directly in front of you.", + "E": "The chessboard is above you, rather than at your eye level ahead." + }, + "answer_label": "B", + "answer_text": "The chessboard is in front of you.", + "q_id": "egolife_A1_015_5_T2", + "correct_answer": "The chessboard is in front of you.", + "original_options": { + "A": "The chessboard is behind you.", + "B": "The chessboard is in front of you.", + "C": "The chessboard is to your left.", + "D": "The chessboard is to your right side, not directly in front of you.", + "E": "The chessboard is above you, rather than at your eye level ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_6_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the chessboard from me?", + "answer": "The chessboard is to your left.", + "question_chinese": "棋盘在我的什么方向？", + "answer_chinese": "棋盘在你的左方。", + "evidence": [ + { + "description": "You can see a large chessboard directly in front of me now.", + "timestamps": { + "start": "00:00:57", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "Above you, not on either side as you look at the table.", + "D": "To your right side, opposite the left, when facing the chessboard's location.", + "E": "The chessboard is to your left." + }, + "answer_label": "E", + "answer_text": "The chessboard is to your left.", + "q_id": "egolife_A1_015_6_T2", + "correct_answer": "The chessboard is to your left.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "Above you, not on either side as you look at the table.", + "D": "To your right side, opposite the left, when facing the chessboard's location.", + "E": "The chessboard is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_5_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the chessboard from me?", + "answer": "The chessboard is to your back-left.", + "question_chinese": "棋盘在我的什么方向？", + "answer_chinese": "棋盘在你的左后方。", + "evidence": [ + { + "description": "You can see a large chessboard directly in front of me now.", + "timestamps": { + "start": "00:00:57", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:06:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The chessboard is to your back-right.", + "B": "The chessboard is to your back-left.", + "C": "The chessboard is to your front-left.", + "D": "The chessboard is directly in front of you, not angled behind your left shoulder.", + "E": "The chessboard is somewhere behind you but on the right side instead of the left." + }, + "answer_label": "B", + "answer_text": "The chessboard is to your back-left.", + "q_id": "egolife_A1_015_5_T1", + "correct_answer": "The chessboard is to your back-left.", + "original_options": { + "A": "The chessboard is to your back-right.", + "B": "The chessboard is to your back-left.", + "C": "The chessboard is to your front-left.", + "D": "The chessboard is directly in front of you, not angled behind your left shoulder.", + "E": "The chessboard is somewhere behind you but on the right side instead of the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_6_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the chessboard from me?", + "answer": "The chessboard is to your back-right.", + "question_chinese": "棋盘在我的什么方向？", + "answer_chinese": "棋盘在你的右后方。", + "evidence": [ + { + "description": "You can see a large chessboard directly in front of me now.", + "timestamps": { + "start": "00:00:57", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The chessboard is to your back-right.", + "B": "The chessboard is to your back-left.", + "C": "The chessboard is to your front-right.", + "D": "The chessboard is to your left side, not positioned behind your right shoulder.", + "E": "The chessboard is behind you but on the left, rather than behind you on the right." + }, + "answer_label": "A", + "answer_text": "The chessboard is to your back-right.", + "q_id": "egolife_A1_015_6_T1", + "correct_answer": "The chessboard is to your back-right.", + "original_options": { + "A": "The chessboard is to your back-right.", + "B": "The chessboard is to your back-left.", + "C": "The chessboard is to your front-right.", + "D": "The chessboard is to your left side, not positioned behind your right shoulder.", + "E": "The chessboard is behind you but on the left, rather than behind you on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_5", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many types of iron utensils are there in the pool?", + "answer": "There are two types of ironware in total.", + "question_chinese": "水池里一共有几种铁质器皿？", + "answer_chinese": "总共有两种铁质器皿。", + "evidence": [ + { + "description": "From the current perspective, there are three types of ironware in total: bowls, iron trays, and iron steamers.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "I cleaned the iron steamer and took it out of the pool.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:38" + } + }, + { + "description": "I cleaned the iron tray and took it out of the pool.But there's still an iron tray left at the bottom of the pool.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:04:33", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There is one type of ironware in total.", + "B": "There are three types of ironware in total.", + "C": "There are four types of ironware in total, so you can see four distinct iron utensil categories in the pool.", + "D": "There are two types of ironware in total.", + "E": "There are five types of ironware in total, meaning five different iron utensil kinds are present in the pool." + }, + "answer_label": "D", + "answer_text": "There are two types of ironware in total.", + "q_id": "egolife_A1_015_5", + "correct_answer": "There are two types of ironware in total.", + "original_options": { + "A": "There is one type of ironware in total.", + "B": "There are three types of ironware in total.", + "C": "There are four types of ironware in total, so you can see four distinct iron utensil categories in the pool.", + "D": "There are two types of ironware in total.", + "E": "There are five types of ironware in total, meaning five different iron utensil kinds are present in the pool." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 20, + 19 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_6", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many types of iron utensils are there in the pool?", + "answer": "There is one type of ironware in the pool.", + "question_chinese": "水池里一共有几种铁质器皿？", + "answer_chinese": "水池里有一样铁制器皿。", + "evidence": [ + { + "description": "From the current perspective, there are three types of ironware in total: bowls, iron trays, and iron steamers.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + }, + { + "description": "I cleaned the iron steamer and took it out of the pool.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:38" + } + }, + { + "description": "I cleaned the iron tray and took it out of the pool.But there's still an iron tray left at the bottom of the pool.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + }, + { + "description": "I have finished cleaning all the iron bowls and taken them all out of the pool.", + "timestamps": { + "start": "00:04:50", + "end": "00:05:01" + } + } + ], + "question_timestamps": "00:11:59", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two types in the pool.", + "B": "Three types in the pool.", + "C": "There is one type of ironware in the pool.", + "D": "There are four types of ironware in the pool, with four different iron utensil categories present there.", + "E": "There are five types of ironware in the pool, so you can identify five distinct iron utensil types." + }, + "answer_label": "C", + "answer_text": "There is one type of ironware in the pool.", + "q_id": "egolife_A1_015_6", + "correct_answer": "There is one type of ironware in the pool.", + "original_options": { + "A": "Two types in the pool.", + "B": "Three types in the pool.", + "C": "There is one type of ironware in the pool.", + "D": "There are four types of ironware in the pool, with four different iron utensil categories present there.", + "E": "There are five types of ironware in the pool, so you can identify five distinct iron utensil types." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_11", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the whiteboard from me?", + "answer": "The whiteboard is behind your right side.", + "question_chinese": "白板在我的什么方向？", + "answer_chinese": "白板在你的右后方", + "evidence": [ + { + "description": "The whiteboard is in front of my left side.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:06:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of your right side.", + "B": "Straight ahead of you.", + "C": "The whiteboard is behind your left side, slightly over your shoulder as you face forward.", + "D": "The whiteboard is behind your right side.", + "E": "The whiteboard is on your right side in front, not behind you as you look around." + }, + "answer_label": "D", + "answer_text": "The whiteboard is behind your right side.", + "q_id": "egolife_A1_015_11", + "correct_answer": "The whiteboard is behind your right side.", + "original_options": { + "A": "In front of your right side.", + "B": "Straight ahead of you.", + "C": "The whiteboard is behind your left side, slightly over your shoulder as you face forward.", + "D": "The whiteboard is behind your right side.", + "E": "The whiteboard is on your right side in front, not behind you as you look around." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_015", + "video_length": "00:12:00", + "qid": "egolife_A1_015_12", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the whiteboard from me?", + "answer": "The whiteboard is on your right side.", + "question_chinese": "白板在我的什么方向？", + "answer_chinese": "白板在你的右侧。", + "evidence": [ + { + "description": "The whiteboard is in front of my left side.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:12:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your left side.", + "B": "Behind your right side.", + "C": "The whiteboard is on your right side.", + "D": "The whiteboard is straight ahead of you, directly in front as you face forward.", + "E": "The whiteboard is above your head, higher than eye level when you look up." + }, + "answer_label": "C", + "answer_text": "The whiteboard is on your right side.", + "q_id": "egolife_A1_015_12", + "correct_answer": "The whiteboard is on your right side.", + "original_options": { + "A": "On your left side.", + "B": "Behind your right side.", + "C": "The whiteboard is on your right side.", + "D": "The whiteboard is straight ahead of you, directly in front as you face forward.", + "E": "The whiteboard is above your head, higher than eye level when you look up." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + } +] diff --git a/QAs/egolife_A1_016-mcq.json b/QAs/egolife_A1_016-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e50d4f56fed1a97ebc217d8206eb68ff097a945c --- /dev/null +++ b/QAs/egolife_A1_016-mcq.json @@ -0,0 +1,951 @@ +[ + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016-mcq_fixed_2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get to the main street, which way should I go?", + "answer": "You should turn around, go into the corridor, turn left to exit the door, then turn left again and walk straight ahead to get out of the courtyard gate. After that, turn left once more and keep walking—you’ll reach the main street.", + "question_chinese": "如果我想去大街上，我应该怎么走？", + "answer_chinese": "你应该向后转，然后进入走廊，左转出房门，之后再向左转直走，走出院门，然后再左转，一直走就能到大街上了。", + "evidence": [ + { + "description": "I walk out of the living room and go straight out the door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "I turn left, walk across the courtyard, and reach the courtyard gate.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:09" + } + }, + { + "description": "I go out, turn left, and walk straight until I reach the main street.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:35" + } + }, + { + "description": "I enter the house through the door, then walk straight and turn right into the living room.", + "timestamps": { + "start": "00:01:11", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:04:49", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right and head out to the street.", + "B": "You should turn around, go into the corridor, turn left to exit the door, then turn left again and walk straight ahead to get out of the courtyard gate. After that, turn left once more and keep walking—you’ll reach the main street.", + "C": "Go straight through the courtyard gate.", + "D": "Turn around, enter the corridor, turn right to exit the door, then keep straight to the courtyard gate.", + "E": "Turn around, go into the corridor, turn left to exit, then turn right outside and walk straight to the main street." + }, + "answer_label": "B", + "answer_text": "You should turn around, go into the corridor, turn left to exit the door, then turn left again and walk straight ahead to get out of the courtyard gate. After that, turn left once more and keep walking—you’ll reach the main street.", + "q_id": "egolife_A1_016-mcq_fixed_2", + "correct_answer": "You should turn around, go into the corridor, turn left to exit the door, then turn left again and walk straight ahead to get out of the courtyard gate. After that, turn left once more and keep walking—you’ll reach the main street.", + "original_options": { + "A": "Turn right and head out to the street.", + "B": "You should turn around, go into the corridor, turn left to exit the door, then turn left again and walk straight ahead to get out of the courtyard gate. After that, turn left once more and keep walking—you’ll reach the main street.", + "C": "Go straight through the courtyard gate.", + "D": "Turn around, enter the corridor, turn right to exit the door, then keep straight to the courtyard gate.", + "E": "Turn around, go into the corridor, turn left to exit, then turn right outside and walk straight to the main street." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 18, + 21 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016-mcq_fixed_3", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to get to the main street, which way should I go?", + "answer": "You should turn to your left rear, walk out of the courtyard gate, then turn left and keep walking straight until you reach the main street.", + "question_chinese": "如果我想去大街上，我应该怎么走？", + "answer_chinese": "你应当向左后方转，然后走出院门，随后左转，一直走就能走到大街上了。", + "evidence": [ + { + "description": "I walk out of the living room and go straight out the door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "I turn left, walk across the courtyard, and reach the courtyard gate.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:09" + } + }, + { + "description": "I go out, turn left, and walk straight until I reach the main street.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:35" + } + }, + { + "description": "I enter the house through the door, then walk straight and turn right into the living room.", + "timestamps": { + "start": "00:01:11", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:08:12", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right rear and exit the gate.", + "B": "Exit the gate, then turn right.", + "C": "Turn to your left rear, leave through the courtyard gate, then turn right and walk straight.", + "D": "Turn to your right rear, walk out of the courtyard gate, then turn left and continue straight.", + "E": "You should turn to your left rear, walk out of the courtyard gate, then turn left and keep walking straight until you reach the main street." + }, + "answer_label": "E", + "answer_text": "You should turn to your left rear, walk out of the courtyard gate, then turn left and keep walking straight until you reach the main street.", + "q_id": "egolife_A1_016-mcq_fixed_3", + "correct_answer": "You should turn to your left rear, walk out of the courtyard gate, then turn left and keep walking straight until you reach the main street.", + "original_options": { + "A": "Turn right rear and exit the gate.", + "B": "Exit the gate, then turn right.", + "C": "Turn to your left rear, leave through the courtyard gate, then turn right and walk straight.", + "D": "Turn to your right rear, walk out of the courtyard gate, then turn left and continue straight.", + "E": "You should turn to your left rear, walk out of the courtyard gate, then turn left and keep walking straight until you reach the main street." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_8_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the living room table from me?", + "answer": "The living room table is on your left side.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "客厅的桌子在你的左侧。", + "evidence": [ + { + "description": "The table in the living room is directly in front of me.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The living room table is on your left side.", + "B": "On your right side.", + "C": "Directly in front of you.", + "D": "The living room table is to your back-left, behind you rather than beside your left arm.", + "E": "The living room table is to your back-right, positioned behind you on the opposite side." + }, + "answer_label": "A", + "answer_text": "The living room table is on your left side.", + "q_id": "egolife_A1_016_8_T1", + "correct_answer": "The living room table is on your left side.", + "original_options": { + "A": "The living room table is on your left side.", + "B": "On your right side.", + "C": "Directly in front of you.", + "D": "The living room table is to your back-left, behind you rather than beside your left arm.", + "E": "The living room table is to your back-right, positioned behind you on the opposite side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_8_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the living room table from me?", + "answer": "The living room table is on your right side.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "客厅的桌子在你的右侧。", + "evidence": [ + { + "description": "The table in the living room is directly in front of me.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:05:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your left side.", + "B": "The living room table is on your right side.", + "C": "Directly in front of you.", + "D": "The living room table is to your back-left, slightly behind your left shoulder as you face forward.", + "E": "The living room table is to your back-right, behind you rather than beside your right arm." + }, + "answer_label": "B", + "answer_text": "The living room table is on your right side.", + "q_id": "egolife_A1_016_8_T2", + "correct_answer": "The living room table is on your right side.", + "original_options": { + "A": "On your left side.", + "B": "The living room table is on your right side.", + "C": "Directly in front of you.", + "D": "The living room table is to your back-left, slightly behind your left shoulder as you face forward.", + "E": "The living room table is to your back-right, behind you rather than beside your right arm." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_5_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the living room table from me?", + "answer": "The living room table is to your back-left.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "客厅的桌子在你的左后方。", + "evidence": [ + { + "description": "The table in the living room is directly in front of me.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:06:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your back-right.", + "B": "It’s to your front-left.", + "C": "The living room table is positioned behind you, slightly toward your right side as you face forward.", + "D": "The living room table is to your back-left.", + "E": "The living room table sits ahead of you on the left, so you would reach it by turning left." + }, + "answer_label": "D", + "answer_text": "The living room table is to your back-left.", + "q_id": "egolife_A1_016_5_T2", + "correct_answer": "The living room table is to your back-left.", + "original_options": { + "A": "It’s to your back-right.", + "B": "It’s to your front-left.", + "C": "The living room table is positioned behind you, slightly toward your right side as you face forward.", + "D": "The living room table is to your back-left.", + "E": "The living room table sits ahead of you on the left, so you would reach it by turning left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_5_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the living room table from me?", + "answer": "The living room table is to your back-right.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "客厅的桌子在你的右后方。", + "evidence": [ + { + "description": "The table in the living room is directly in front of me.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:08:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your back-left.", + "B": "It’s to your front-right.", + "C": "The living room table is to your back-right.", + "D": "The living room table is behind you and slightly to the left, not on your right side.", + "E": "The living room table is in front of you toward the right, so it’s not behind you." + }, + "answer_label": "C", + "answer_text": "The living room table is to your back-right.", + "q_id": "egolife_A1_016_5_T1", + "correct_answer": "The living room table is to your back-right.", + "original_options": { + "A": "It’s to your back-left.", + "B": "It’s to your front-right.", + "C": "The living room table is to your back-right.", + "D": "The living room table is behind you and slightly to the left, not on your right side.", + "E": "The living room table is in front of you toward the right, so it’s not behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_2_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which table am I closest to right now?", + "answer": "Right now, you’re closer to the table in the living room.", + "question_chinese": "我现在离哪张桌子最近？", + "answer_chinese": "你现在离客厅里的桌子更近。", + "evidence": [ + { + "description": "There is a table in the courtyard that lies ahead and to my right.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "There is a table in the living room directly in front of me.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:05:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The yard table.", + "B": "The dining table.", + "C": "Right now, you’re closer to the kitchen table than to the living room table.", + "D": "Right now, you’re about the same distance from both tables, so neither is closer.", + "E": "Right now, you’re closer to the table in the living room." + }, + "answer_label": "E", + "answer_text": "Right now, you’re closer to the table in the living room.", + "q_id": "egolife_A1_016_2_T1", + "correct_answer": "Right now, you’re closer to the table in the living room.", + "original_options": { + "A": "The yard table.", + "B": "The dining table.", + "C": "Right now, you’re closer to the kitchen table than to the living room table.", + "D": "Right now, you’re about the same distance from both tables, so neither is closer.", + "E": "Right now, you’re closer to the table in the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_2_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which table am I closest to right now?", + "answer": "Right now, you’re closer to the table in the living room.", + "question_chinese": "我现在离哪张桌子最近？", + "answer_chinese": "你现在离客厅里的桌子更近。", + "evidence": [ + { + "description": "There is a table in the courtyard that lies ahead and to my right.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "There is a table in the living room directly in front of me.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:05:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right now, you’re closer to the table in the living room.", + "B": "The yard table.", + "C": "The table behind you.", + "D": "Right now, you’re closest to the table on your left rather than the one in the living room.", + "E": "Right now, you’re closer to the table on your right, not the table located in the living room." + }, + "answer_label": "A", + "answer_text": "Right now, you’re closer to the table in the living room.", + "q_id": "egolife_A1_016_2_T2", + "correct_answer": "Right now, you’re closer to the table in the living room.", + "original_options": { + "A": "Right now, you’re closer to the table in the living room.", + "B": "The yard table.", + "C": "The table behind you.", + "D": "Right now, you’re closest to the table on your left rather than the one in the living room.", + "E": "Right now, you’re closer to the table on your right, not the table located in the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_2_T3", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which table am I closest to right now?", + "answer": "Right now, you’re closer to the table in the yard.", + "question_chinese": "我现在离哪张桌子最近？", + "answer_chinese": "你现在离院子里的桌子更近。", + "evidence": [ + { + "description": "There is a table in the courtyard that lies ahead and to my right.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "There is a table in the living room directly in front of me.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:08:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The living room table.", + "B": "Right now, you’re closer to the table in the yard.", + "C": "The table on your left.", + "D": "Right now, you’re closer to the table behind you than to the table out in the yard.", + "E": "Right now, you’re closest to the table on your right side, not the one in the yard." + }, + "answer_label": "B", + "answer_text": "Right now, you’re closer to the table in the yard.", + "q_id": "egolife_A1_016_2_T3", + "correct_answer": "Right now, you’re closer to the table in the yard.", + "original_options": { + "A": "The living room table.", + "B": "Right now, you’re closer to the table in the yard.", + "C": "The table on your left.", + "D": "Right now, you’re closer to the table behind you than to the table out in the yard.", + "E": "Right now, you’re closest to the table on your right side, not the one in the yard." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_2_T4", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which table am I closest to right now?", + "answer": "Right now, you’re closer to the table in the yard.", + "question_chinese": "我现在离哪张桌子最近？", + "answer_chinese": "你现在离院子里的桌子更近。", + "evidence": [ + { + "description": "There is a table in the courtyard that lies ahead and to my right.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "There is a table in the living room directly in front of me.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:08:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The living room table.", + "B": "The table behind you.", + "C": "Right now, you’re closer to the table in the yard.", + "D": "Right now, you’re closer to the table on your left than to the table in the yard.", + "E": "Right now, you’re closest to the table on your right side, rather than the table outside in the yard." + }, + "answer_label": "C", + "answer_text": "Right now, you’re closer to the table in the yard.", + "q_id": "egolife_A1_016_2_T4", + "correct_answer": "Right now, you’re closer to the table in the yard.", + "original_options": { + "A": "The living room table.", + "B": "The table behind you.", + "C": "Right now, you’re closer to the table in the yard.", + "D": "Right now, you’re closer to the table on your left than to the table in the yard.", + "E": "Right now, you’re closest to the table on your right side, rather than the table outside in the yard." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_7", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are currently in the house?", + "answer": "There is one person in the house.", + "question_chinese": "房子里现在有几个人？", + "answer_chinese": "房子里有一个人", + "evidence": [ + { + "description": "There are five people in the room now, including me.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:01:59", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one person in the house.", + "B": "Two people.", + "C": "Three people.", + "D": "There are four people currently in the house right now, including someone in another room.", + "E": "There are five people currently in the house right now, spread across different rooms." + }, + "answer_label": "A", + "answer_text": "There is one person in the house.", + "q_id": "egolife_A1_016_7", + "correct_answer": "There is one person in the house.", + "original_options": { + "A": "There is one person in the house.", + "B": "Two people.", + "C": "Three people.", + "D": "There are four people currently in the house right now, including someone in another room.", + "E": "There are five people currently in the house right now, spread across different rooms." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_8", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are currently in the house?", + "answer": "There is no one in the house now.", + "question_chinese": "房子里现在有几个人？", + "answer_chinese": "房子里现在没有人了", + "evidence": [ + { + "description": "There are five people in the room now, including me.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:20" + } + }, + { + "description": "Everyone has come out of the house.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:11" + } + } + ], + "question_timestamps": "00:08:42", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One person.", + "B": "There is no one in the house now.", + "C": "Two people.", + "D": "There are three people currently in the house right now, gathered together in one room.", + "E": "There are four people currently in the house right now, with at least one person in another room." + }, + "answer_label": "B", + "answer_text": "There is no one in the house now.", + "q_id": "egolife_A1_016_8", + "correct_answer": "There is no one in the house now.", + "original_options": { + "A": "One person.", + "B": "There is no one in the house now.", + "C": "Two people.", + "D": "There are three people currently in the house right now, gathered together in one room.", + "E": "There are four people currently in the house right now, with at least one person in another room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_1_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine relative to me?", + "answer": "The washing machine is on your right side.", + "question_chinese": "洗衣机在我的什么位置？", + "answer_chinese": "洗衣机在你的右侧。", + "evidence": [ + { + "description": "There is a washing machine on my right side.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:05:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The washing machine is on your left.", + "B": "The washing machine is on your right side.", + "C": "The washing machine is in front of you.", + "D": "The washing machine is behind you, not at your side, so you’d need to turn around.", + "E": "The washing machine is ahead on your left, so it’s not positioned on your right side." + }, + "answer_label": "B", + "answer_text": "The washing machine is on your right side.", + "q_id": "egolife_A1_016_1_T1", + "correct_answer": "The washing machine is on your right side.", + "original_options": { + "A": "The washing machine is on your left.", + "B": "The washing machine is on your right side.", + "C": "The washing machine is in front of you.", + "D": "The washing machine is behind you, not at your side, so you’d need to turn around.", + "E": "The washing machine is ahead on your left, so it’s not positioned on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_1_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine relative to me?", + "answer": "The washing machine is on your left side.", + "question_chinese": "洗衣机在我的什么位置？", + "answer_chinese": "洗衣机在你的左侧。", + "evidence": [ + { + "description": "There is a washing machine on my right side.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The washing machine is on your right.", + "B": "The washing machine is in front of you.", + "C": "The washing machine is behind you, so it isn’t positioned on your left side.", + "D": "The washing machine is on your left side.", + "E": "The washing machine is behind and to your right, not on your left, so you’d turn the other way." + }, + "answer_label": "D", + "answer_text": "The washing machine is on your left side.", + "q_id": "egolife_A1_016_1_T2", + "correct_answer": "The washing machine is on your left side.", + "original_options": { + "A": "The washing machine is on your right.", + "B": "The washing machine is in front of you.", + "C": "The washing machine is behind you, so it isn’t positioned on your left side.", + "D": "The washing machine is on your left side.", + "E": "The washing machine is behind and to your right, not on your left, so you’d turn the other way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 20 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_9", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the sunshade from me?", + "answer": "The sunshade is behind you.", + "question_chinese": "遮阳伞在我的什么方向？", + "answer_chinese": "遮阳伞在你的后方。", + "evidence": [ + { + "description": "A sunshade can be seen in the courtyard.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:05:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The sunshade is in front of you.", + "B": "The sunshade is to your left.", + "C": "The sunshade is off to your right side, so it isn’t located behind you.", + "D": "The sunshade is above and slightly in front of you, not positioned behind your back.", + "E": "The sunshade is behind you." + }, + "answer_label": "E", + "answer_text": "The sunshade is behind you.", + "q_id": "egolife_A1_016_9", + "correct_answer": "The sunshade is behind you.", + "original_options": { + "A": "The sunshade is in front of you.", + "B": "The sunshade is to your left.", + "C": "The sunshade is off to your right side, so it isn’t located behind you.", + "D": "The sunshade is above and slightly in front of you, not positioned behind your back.", + "E": "The sunshade is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_016", + "video_length": "00:11:45", + "qid": "egolife_A1_016_10", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the sunshade from me?", + "answer": "The sunshade is ahead and to your right.", + "question_chinese": "遮阳伞在我的什么方向？", + "answer_chinese": "遮阳伞在你的右前方。", + "evidence": [ + { + "description": "A sunshade can be seen in the courtyard.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:09:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The sunshade is ahead and left.", + "B": "The sunshade is behind you.", + "C": "The sunshade is ahead and to your right.", + "D": "The sunshade is directly to your right side, beside you rather than ahead of you.", + "E": "The sunshade is in front of you but to your left, so it isn’t ahead and to your right." + }, + "answer_label": "C", + "answer_text": "The sunshade is ahead and to your right.", + "q_id": "egolife_A1_016_10", + "correct_answer": "The sunshade is ahead and to your right.", + "original_options": { + "A": "The sunshade is ahead and left.", + "B": "The sunshade is behind you.", + "C": "The sunshade is ahead and to your right.", + "D": "The sunshade is directly to your right side, beside you rather than ahead of you.", + "E": "The sunshade is in front of you but to your left, so it isn’t ahead and to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 20 + ] + } +] diff --git a/QAs/egolife_A1_017-mcq.json b/QAs/egolife_A1_017-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..07a2196c0f5bf6a08f6542d4dd736d00be341ab7 --- /dev/null +++ b/QAs/egolife_A1_017-mcq.json @@ -0,0 +1,951 @@ +[ + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_4_T2", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which entrance to the mall is closer to me?", + "answer": "The entrance to your front-left of the underground parking garage exit is closer to you.", + "question_chinese": "商场的哪个入口离我更近？", + "answer_chinese": "地下停车场出口左前方的那个入口离你更近。", + "evidence": [ + { + "description": "The entrance on the front left of the underground parking exit is closer to me.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The front-right entrance is closer.", + "B": "The back-left entrance is closer.", + "C": "The entrance behind and to your right of the underground parking garage exit is the closer one.", + "D": "The entrance to your front-left of the underground parking garage exit is closer to you.", + "E": "The entrance in front-right of the underground parking garage exit is nearer to you than the front-left one." + }, + "answer_label": "D", + "answer_text": "The entrance to your front-left of the underground parking garage exit is closer to you.", + "q_id": "egolife_A1_017_4_T2", + "correct_answer": "The entrance to your front-left of the underground parking garage exit is closer to you.", + "original_options": { + "A": "The front-right entrance is closer.", + "B": "The back-left entrance is closer.", + "C": "The entrance behind and to your right of the underground parking garage exit is the closer one.", + "D": "The entrance to your front-left of the underground parking garage exit is closer to you.", + "E": "The entrance in front-right of the underground parking garage exit is nearer to you than the front-left one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_5_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which entrance to the mall is closer to me?", + "answer": "Right now, the mall entrance next to the escalator is closer to you.", + "question_chinese": "商场的哪个入口离我更近？", + "answer_chinese": "现在是扶梯旁边的那个商场入口离你更近。", + "evidence": [ + { + "description": "The entrance next to the escalator is closer to you.", + "timestamps": { + "start": "00:02:56", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:03:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The entrance by the garage exit is closer.", + "B": "The front-left entrance is closer.", + "C": "Right now, the entrance near the garage exit is closer than the one by the escalator.", + "D": "Right now, the entrance to your front-right near the underground garage exit is closer to you.", + "E": "Right now, the mall entrance next to the escalator is closer to you." + }, + "answer_label": "E", + "answer_text": "Right now, the mall entrance next to the escalator is closer to you.", + "q_id": "egolife_A1_017_5_T2", + "correct_answer": "Right now, the mall entrance next to the escalator is closer to you.", + "original_options": { + "A": "The entrance by the garage exit is closer.", + "B": "The front-left entrance is closer.", + "C": "Right now, the entrance near the garage exit is closer than the one by the escalator.", + "D": "Right now, the entrance to your front-right near the underground garage exit is closer to you.", + "E": "Right now, the mall entrance next to the escalator is closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_4_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which entrance to the mall is closer to me?", + "answer": "The entrance that’s closer to the escalator is closer to you.", + "question_chinese": "商场的哪个入口离我更近？", + "answer_chinese": "离扶手电梯近的那个入口离你更近。", + "evidence": [ + { + "description": "Translation: This entrance to the shopping mall directly in front of me is the one that is closer to the exit of the underground parking lot.", + "timestamps": { + "start": "00:02:27", + "end": "00:02:28" + } + }, + { + "description": "The shopping mall entrance ahead and to my right is the one closer to the escalator.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:53" + } + }, + { + "description": "We went up to the second floor by escalator.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:58" + } + } + ], + "question_timestamps": "00:04:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The garage-exit entrance is closer.", + "B": "The front-right entrance is closer.", + "C": "The entrance that’s closer to the escalator is closer to you.", + "D": "The entrance to your back-left of the underground parking garage exit is closer to you than the one by the escalator.", + "E": "The entrance that’s closer to the underground parking garage exit is the nearer choice, rather than the escalator-side entrance." + }, + "answer_label": "C", + "answer_text": "The entrance that’s closer to the escalator is closer to you.", + "q_id": "egolife_A1_017_4_T1", + "correct_answer": "The entrance that’s closer to the escalator is closer to you.", + "original_options": { + "A": "The garage-exit entrance is closer.", + "B": "The front-right entrance is closer.", + "C": "The entrance that’s closer to the escalator is closer to you.", + "D": "The entrance to your back-left of the underground parking garage exit is closer to you than the one by the escalator.", + "E": "The entrance that’s closer to the underground parking garage exit is the nearer choice, rather than the escalator-side entrance." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 20 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_5_T1", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which entrance to the mall is closer to me?", + "answer": "Right now, the mall entrance that’s closer to the underground parking garage exit is closer to you.", + "question_chinese": "商场的哪个入口离我更近？", + "answer_chinese": "现在是离地下停车场出口更近的那个商场入口离你更近。", + "evidence": [ + { + "description": "Translation: This entrance to the shopping mall directly in front of me is the one that is closer to the exit of the underground parking lot.", + "timestamps": { + "start": "00:02:27", + "end": "00:02:28" + } + }, + { + "description": "The shopping mall entrance ahead and to my right is the one closer to the escalator.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:53" + } + }, + { + "description": "We went up to the second floor by escalator.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:58" + } + } + ], + "question_timestamps": "00:06:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The escalator-side entrance is closer.", + "B": "The farther entrance is closer to you.", + "C": "Right now, the mall entrance that’s farther from the underground parking garage exit is closer to you.", + "D": "Right now, the mall entrance that’s closer to the underground parking garage exit is closer to you.", + "E": "Right now, the entrance located to your back-left of the underground parking garage exit is closer to you." + }, + "answer_label": "D", + "answer_text": "Right now, the mall entrance that’s closer to the underground parking garage exit is closer to you.", + "q_id": "egolife_A1_017_5_T1", + "correct_answer": "Right now, the mall entrance that’s closer to the underground parking garage exit is closer to you.", + "original_options": { + "A": "The escalator-side entrance is closer.", + "B": "The farther entrance is closer to you.", + "C": "Right now, the mall entrance that’s farther from the underground parking garage exit is closer to you.", + "D": "Right now, the mall entrance that’s closer to the underground parking garage exit is closer to you.", + "E": "Right now, the entrance located to your back-left of the underground parking garage exit is closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_2_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the exit of the underground parking garage relative to me?", + "answer": "The underground parking garage exit is to your front-left.", + "question_chinese": "地下停车场的出口在我的什么方向？", + "answer_chinese": "地下停车场的出口在你的左前方。", + "evidence": [ + { + "description": "The exit of the underground parking lot is on your front left.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The underground parking garage exit is to your front-left.", + "B": "It is to your front-right.", + "C": "It is directly behind you.", + "D": "The exit is on your right side, slightly ahead of you as you face forward.", + "E": "The exit is behind you to the right, so you would need to turn around to face it." + }, + "answer_label": "A", + "answer_text": "The underground parking garage exit is to your front-left.", + "q_id": "egolife_A1_017_2_T2", + "correct_answer": "The underground parking garage exit is to your front-left.", + "original_options": { + "A": "The underground parking garage exit is to your front-left.", + "B": "It is to your front-right.", + "C": "It is directly behind you.", + "D": "The exit is on your right side, slightly ahead of you as you face forward.", + "E": "The exit is behind you to the right, so you would need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_3_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the exit of the underground parking garage relative to me?", + "answer": "The underground parking garage exit is directly in front of you.", + "question_chinese": "地下停车场的出口在我的什么方向？", + "answer_chinese": "地下停车场的出口在你的正前方。", + "evidence": [ + { + "description": "The exit of the underground parking lot in front of me.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is directly behind you.", + "B": "The underground parking garage exit is directly in front of you.", + "C": "It is directly to your left.", + "D": "The exit is off to your right side, so you’d need to turn your body to face it.", + "E": "The exit is slightly behind you, meaning you should rotate around before heading toward it." + }, + "answer_label": "B", + "answer_text": "The underground parking garage exit is directly in front of you.", + "q_id": "egolife_A1_017_3_T2", + "correct_answer": "The underground parking garage exit is directly in front of you.", + "original_options": { + "A": "It is directly behind you.", + "B": "The underground parking garage exit is directly in front of you.", + "C": "It is directly to your left.", + "D": "The exit is off to your right side, so you’d need to turn your body to face it.", + "E": "The exit is slightly behind you, meaning you should rotate around before heading toward it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 15 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_2_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the exit of the underground parking garage relative to me?", + "answer": "The underground parking garage exit is to your back-left.", + "question_chinese": "地下停车场的出口在我的什么方向？", + "answer_chinese": "地下停车场的出口在你的左后方。", + "evidence": [ + { + "description": "The exit of the underground parking lot is ahead and to my left.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The underground parking garage exit is to your back-left.", + "B": "It is to your back-right.", + "C": "It is to your front-left.", + "D": "The exit is directly in front of you, so you can head forward without turning around.", + "E": "The exit is on your left but ahead of you, requiring only a slight turn left to align." + }, + "answer_label": "A", + "answer_text": "The underground parking garage exit is to your back-left.", + "q_id": "egolife_A1_017_2_T1", + "correct_answer": "The underground parking garage exit is to your back-left.", + "original_options": { + "A": "The underground parking garage exit is to your back-left.", + "B": "It is to your back-right.", + "C": "It is to your front-left.", + "D": "The exit is directly in front of you, so you can head forward without turning around.", + "E": "The exit is on your left but ahead of you, requiring only a slight turn left to align." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_3_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the exit of the underground parking garage relative to me?", + "answer": "The underground parking garage exit is roughly behind you, and slightly lower.", + "question_chinese": "地下停车场的出口在我的什么方向？", + "answer_chinese": "地下停车场的出口大约在你的后方，且偏下。", + "evidence": [ + { + "description": "The exit of the underground parking lot is ahead and to my left.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "We went up to the second floor by escalator.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:59" + } + } + ], + "question_timestamps": "00:08:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The underground parking garage exit is roughly behind you, and slightly lower.", + "B": "It is roughly in front of you.", + "C": "It is roughly above you.", + "D": "The exit is roughly in front of you and a bit higher, so you would look up slightly.", + "E": "The exit is roughly to your right and slightly higher, meaning it sits above your current level." + }, + "answer_label": "A", + "answer_text": "The underground parking garage exit is roughly behind you, and slightly lower.", + "q_id": "egolife_A1_017_3_T1", + "correct_answer": "The underground parking garage exit is roughly behind you, and slightly lower.", + "original_options": { + "A": "The underground parking garage exit is roughly behind you, and slightly lower.", + "B": "It is roughly in front of you.", + "C": "It is roughly above you.", + "D": "The exit is roughly in front of you and a bit higher, so you would look up slightly.", + "E": "The exit is roughly to your right and slightly higher, meaning it sits above your current level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_7_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go push another shopping cart, how should I walk?", + "answer": "You should turn around, then turn right to go out, and then keep walking straight.", + "question_chinese": "如果我想再去推一辆购物车，我应该怎么走？", + "answer_chinese": "你应该掉头，然后右转出门再直走。", + "evidence": [ + { + "description": "I walked left after I came in through the door.", + "timestamps": { + "start": "00:06:48", + "end": "00:08:35" + } + } + ], + "question_timestamps": "00:08:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around, then turn left.", + "B": "Turn around and walk straight.", + "C": "You should walk out of the supermarket toward your front-right, then angle toward your front-left for a few steps.", + "D": "You should turn around, then turn right to go out, and then keep walking straight.", + "E": "You should exit through the main entrance and then head toward your front-left for five or six steps." + }, + "answer_label": "D", + "answer_text": "You should turn around, then turn right to go out, and then keep walking straight.", + "q_id": "egolife_A1_017_7_T2", + "correct_answer": "You should turn around, then turn right to go out, and then keep walking straight.", + "original_options": { + "A": "Turn around, then turn left.", + "B": "Turn around and walk straight.", + "C": "You should walk out of the supermarket toward your front-right, then angle toward your front-left for a few steps.", + "D": "You should turn around, then turn right to go out, and then keep walking straight.", + "E": "You should exit through the main entrance and then head toward your front-left for five or six steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_7_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go push another shopping cart, how should I walk?", + "answer": "You should turn around, go out through the supermarket’s main entrance, then walk toward your front-left. After about five or six steps, you’ll see the shopping carts.", + "question_chinese": "如果我想再去推一辆购物车，我应该怎么走？", + "answer_chinese": "你应该掉头，然后出超市的大门，随后向左前方走，五六步的距离就能看到购物车了。", + "evidence": [ + { + "description": "There is a row of shopping carts directly in front of me.", + "timestamps": { + "start": "00:06:09", + "end": "00:06:10" + } + }, + { + "description": "This row of shopping carts is about five or six steps away from the supermarket door.", + "timestamps": { + "start": "00:06:09", + "end": "00:06:10" + } + } + ], + "question_timestamps": "00:11:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go out the main entrance, front-right.", + "B": "Go out the main entrance, back-left.", + "C": "You should turn around, then turn right to exit, and keep walking straight once you’re outside.", + "D": "You should turn around, go out through the supermarket’s main entrance, then walk toward your front-left. After about five or six steps, you’ll see the shopping carts.", + "E": "You should turn around and walk straight for a few steps, without heading out through the main entrance." + }, + "answer_label": "D", + "answer_text": "You should turn around, go out through the supermarket’s main entrance, then walk toward your front-left. After about five or six steps, you’ll see the shopping carts.", + "q_id": "egolife_A1_017_7_T1", + "correct_answer": "You should turn around, go out through the supermarket’s main entrance, then walk toward your front-left. After about five or six steps, you’ll see the shopping carts.", + "original_options": { + "A": "Go out the main entrance, front-right.", + "B": "Go out the main entrance, back-left.", + "C": "You should turn around, then turn right to exit, and keep walking straight once you’re outside.", + "D": "You should turn around, go out through the supermarket’s main entrance, then walk toward your front-left. After about five or six steps, you’ll see the shopping carts.", + "E": "You should turn around and walk straight for a few steps, without heading out through the main entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_1", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to return to the entrance of the underground parking lot now, how should I go?", + "answer": "You should turn around, then walk forward to the right to one of the shopping mall exits. After exiting, keep walking forward to the right along the mall until you reach the parking lot area, where you will see the entrance to the underground parking lot.", + "question_chinese": "如果现在我想回到地下停车场的入口，我应该怎么走？", + "answer_chinese": "你应该掉头，然后向右前方走到商场的其中一个出口，出门以后沿商场向右前方走到停车场的位置就能看到地下停车场的入口了。", + "evidence": [ + { + "description": "The entrance to the underground parking lot is ahead and to my left.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "We have come to the entrance of the shopping mall from the parking lot.", + "timestamps": { + "start": "00:00:54", + "end": "00:02:29" + } + } + ], + "question_timestamps": "00:03:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn around and head left to an exit.", + "B": "Go straight ahead to the mall exit.", + "C": "Turn around, reach a mall exit on your left, then keep left along the mall to the parking area.", + "D": "You should turn around, then walk forward to the right to one of the shopping mall exits. After exiting, keep walking forward to the right along the mall until you reach the parking lot area, where you will see the entrance to the underground parking lot.", + "E": "Turn around, take the nearest mall exit on your right, then continue straight until you pass the parking lot area." + }, + "answer_label": "D", + "answer_text": "You should turn around, then walk forward to the right to one of the shopping mall exits. After exiting, keep walking forward to the right along the mall until you reach the parking lot area, where you will see the entrance to the underground parking lot.", + "q_id": "egolife_A1_017_1", + "correct_answer": "You should turn around, then walk forward to the right to one of the shopping mall exits. After exiting, keep walking forward to the right along the mall until you reach the parking lot area, where you will see the entrance to the underground parking lot.", + "original_options": { + "A": "Turn around and head left to an exit.", + "B": "Go straight ahead to the mall exit.", + "C": "Turn around, reach a mall exit on your left, then keep left along the mall to the parking area.", + "D": "You should turn around, then walk forward to the right to one of the shopping mall exits. After exiting, keep walking forward to the right along the mall until you reach the parking lot area, where you will see the entrance to the underground parking lot.", + "E": "Turn around, take the nearest mall exit on your right, then continue straight until you pass the parking lot area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to return to the entrance of the underground parking lot now, how should I go?", + "answer": "You should turn around, then walk straight to the crossroads. After that, turn left and walk to the end, then turn right and head to the vicinity of the escalator. Take the escalator down to the first floor. Then walk forward to the right to one of the shopping mall exits. Once you exit, keep walking forward to the right along the mall until you reach the parking lot area, where you will see the entrance to the underground parking lot.", + "question_chinese": "如果现在我想回到地下停车场的入口，我应该怎么走？", + "answer_chinese": "你应该向后转，然后直走到十字路口，随后左转走到尽头再右转走到扶梯附近，乘坐扶手电梯下到1楼。然后向右前方走到商场的其中一个出口，出门以后沿商场向右前方走到停车场的位置就能看到地下停车场的入口了。", + "evidence": [ + { + "description": "The entrance to the underground parking lot is ahead and to my left.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "We have come to the entrance of the shopping mall from the parking lot.", + "timestamps": { + "start": "00:00:54", + "end": "00:02:29" + } + }, + { + "description": "We went up to the second floor by escalator.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:59" + } + } + ], + "question_timestamps": "00:06:17", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn around and go right at the crossroads.", + "B": "You should turn around, then walk straight to the crossroads. After that, turn left and walk to the end, then turn right and head to the vicinity of the escalator. Take the escalator down to the first floor. Then walk forward to the right to one of the shopping mall exits. Once you exit, keep walking forward to the right along the mall until you reach the parking lot area, where you will see the entrance to the underground parking lot.", + "C": "Take the escalator up instead of down.", + "D": "Turn around, go straight to the crossroads, turn right to the end, then take the escalator down and exit left.", + "E": "Turn around, reach the crossroads, turn left, then stay left past the escalator and leave through the nearest mall exit." + }, + "answer_label": "B", + "answer_text": "You should turn around, then walk straight to the crossroads. After that, turn left and walk to the end, then turn right and head to the vicinity of the escalator. Take the escalator down to the first floor. Then walk forward to the right to one of the shopping mall exits. Once you exit, keep walking forward to the right along the mall until you reach the parking lot area, where you will see the entrance to the underground parking lot.", + "q_id": "egolife_A1_017_2", + "correct_answer": "You should turn around, then walk straight to the crossroads. After that, turn left and walk to the end, then turn right and head to the vicinity of the escalator. Take the escalator down to the first floor. Then walk forward to the right to one of the shopping mall exits. Once you exit, keep walking forward to the right along the mall until you reach the parking lot area, where you will see the entrance to the underground parking lot.", + "original_options": { + "A": "Turn around and go right at the crossroads.", + "B": "You should turn around, then walk straight to the crossroads. After that, turn left and walk to the end, then turn right and head to the vicinity of the escalator. Take the escalator down to the first floor. Then walk forward to the right to one of the shopping mall exits. Once you exit, keep walking forward to the right along the mall until you reach the parking lot area, where you will see the entrance to the underground parking lot.", + "C": "Take the escalator up instead of down.", + "D": "Turn around, go straight to the crossroads, turn right to the end, then take the escalator down and exit left.", + "E": "Turn around, reach the crossroads, turn left, then stay left past the escalator and leave through the nearest mall exit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_11", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different items are in my shopping cart?", + "answer": "There is one kind of item.", + "question_chinese": "我的购物车里有几种物品？", + "answer_chinese": "有一种物品。", + "evidence": [ + { + "description": "The girl in yellow put a bottle of mineral water into my shopping cart.", + "timestamps": { + "start": "00:06:17", + "end": "00:06:18" + } + } + ], + "question_timestamps": "00:09:02", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two kinds of items.", + "B": "There are three kinds of items.", + "C": "There are four different kinds of items in your shopping cart right now.", + "D": "There are five different kinds of items in your shopping cart right now.", + "E": "There is one kind of item." + }, + "answer_label": "E", + "answer_text": "There is one kind of item.", + "q_id": "egolife_A1_017_11", + "correct_answer": "There is one kind of item.", + "original_options": { + "A": "There are two kinds of items.", + "B": "There are three kinds of items.", + "C": "There are four different kinds of items in your shopping cart right now.", + "D": "There are five different kinds of items in your shopping cart right now.", + "E": "There is one kind of item." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_12", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different items are in my shopping cart?", + "answer": "There are three kinds of items in total.", + "question_chinese": "我的购物车里有几种物品？", + "answer_chinese": "一共有3种物品。", + "evidence": [ + { + "description": "The girl in yellow put a bottle of mineral water into my shopping cart.", + "timestamps": { + "start": "00:06:17", + "end": "00:06:18" + } + }, + { + "description": "The girl in yellow put two boxes of strawberries into my shopping cart.", + "timestamps": { + "start": "00:09:03", + "end": "00:10:05" + } + }, + { + "description": "A box of grapes has been put into the shopping cart.", + "timestamps": { + "start": "00:10:30", + "end": "00:10:42" + } + } + ], + "question_timestamps": "00:11:09", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are two kinds in total.", + "B": "There are four kinds in total.", + "C": "There are five kinds of items in total in your shopping cart right now.", + "D": "There are six kinds of items in total in your shopping cart right now.", + "E": "There are three kinds of items in total." + }, + "answer_label": "E", + "answer_text": "There are three kinds of items in total.", + "q_id": "egolife_A1_017_12", + "correct_answer": "There are three kinds of items in total.", + "original_options": { + "A": "There are two kinds in total.", + "B": "There are four kinds in total.", + "C": "There are five kinds of items in total in your shopping cart right now.", + "D": "There are six kinds of items in total in your shopping cart right now.", + "E": "There are three kinds of items in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_8_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the shopping cart at the supermarket entrance relative to me?", + "answer": "The shopping carts at the supermarket entrance are to your back-left.", + "question_chinese": "超市门口的购物车在我的什么方向？", + "answer_chinese": "超市门口的购物车在你的左后方。", + "evidence": [ + { + "description": "There is a row of shopping carts is behind me.", + "timestamps": { + "start": "00:06:30", + "end": "00:06:52" + } + } + ], + "question_timestamps": "00:06:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They are to your back-right.", + "B": "They are to your front-left.", + "C": "The shopping carts at the supermarket entrance are to your back-left.", + "D": "The shopping carts at the entrance are directly in front of you, near the doorway.", + "E": "The shopping carts are to your front-right when you face the supermarket entrance." + }, + "answer_label": "C", + "answer_text": "The shopping carts at the supermarket entrance are to your back-left.", + "q_id": "egolife_A1_017_8_T2", + "correct_answer": "The shopping carts at the supermarket entrance are to your back-left.", + "original_options": { + "A": "They are to your back-right.", + "B": "They are to your front-left.", + "C": "The shopping carts at the supermarket entrance are to your back-left.", + "D": "The shopping carts at the entrance are directly in front of you, near the doorway.", + "E": "The shopping carts are to your front-right when you face the supermarket entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_017", + "video_length": "00:12:00", + "qid": "egolife_A1_017_8_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the shopping cart at the supermarket entrance relative to me?", + "answer": "The shopping carts at the supermarket entrance are to your back-left.", + "question_chinese": "超市门口的购物车在我的什么方向？", + "answer_chinese": "超市门口的购物车在你的左后方。", + "evidence": [ + { + "description": "There is a row of shopping carts ahead and to my right.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:11" + } + } + ], + "question_timestamps": "00:08:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They are to your back-right.", + "B": "They are to your front-left.", + "C": "The shopping carts at the supermarket entrance are directly in front of you.", + "D": "The shopping carts are off to your front-right as you stand near the entrance.", + "E": "The shopping carts at the supermarket entrance are to your back-left." + }, + "answer_label": "E", + "answer_text": "The shopping carts at the supermarket entrance are to your back-left.", + "q_id": "egolife_A1_017_8_T1", + "correct_answer": "The shopping carts at the supermarket entrance are to your back-left.", + "original_options": { + "A": "They are to your back-right.", + "B": "They are to your front-left.", + "C": "The shopping carts at the supermarket entrance are directly in front of you.", + "D": "The shopping carts are off to your front-right as you stand near the entrance.", + "E": "The shopping carts at the supermarket entrance are to your back-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + } +] diff --git a/QAs/egolife_A1_018-mcq.json b/QAs/egolife_A1_018-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..aa711e1085a1e537c04c91a8fa364608c722b9b1 --- /dev/null +++ b/QAs/egolife_A1_018-mcq.json @@ -0,0 +1,886 @@ +[ + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_3_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which sections of the supermarket have we visited so far?", + "answer": "Three sections in total.", + "question_chinese": "到目前为止，我们去了超市的哪几个分区？", + "answer_chinese": "总共三个区域。", + "evidence": [ + { + "description": "We are now in the fruit section.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "We have now entered the refrigerated section.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:41" + } + }, + { + "description": "We have now entered the raw food section.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:06:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four sections in total.", + "B": "Five sections in total.", + "C": "Six sections in total, meaning you’ve already visited six different supermarket sections so far.", + "D": "Three sections in total.", + "E": "Seven sections in total, so you’ve passed through seven distinct supermarket sections up to now." + }, + "answer_label": "D", + "answer_text": "Three sections in total.", + "q_id": "egolife_A1_018_3_T2", + "correct_answer": "Three sections in total.", + "original_options": { + "A": "Four sections in total.", + "B": "Five sections in total.", + "C": "Six sections in total, meaning you’ve already visited six different supermarket sections so far.", + "D": "Three sections in total.", + "E": "Seven sections in total, so you’ve passed through seven distinct supermarket sections up to now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_3_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, which sections of the supermarket have we been to?", + "answer": "There are 3 areas in total, located in front of you and to your back-left.", + "question_chinese": "到目前为止，我们去了超市的哪几个分区？", + "answer_chinese": "总共三个区域，在你的前方和左后方。", + "evidence": [ + { + "description": "We are now in the fruit section.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "We have now entered the refrigerated section.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:41" + } + }, + { + "description": "We have now entered the raw food section.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:06:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 2 areas in total.", + "B": "There are 4 areas in total.", + "C": "There are 3 areas in total, located in front of you and to your back-left.", + "D": "There are 2 areas in total, located in front of you and to your back-left.", + "E": "There are 4 areas in total, located in front of you and to your back-left." + }, + "answer_label": "C", + "answer_text": "There are 3 areas in total, located in front of you and to your back-left.", + "q_id": "egolife_A1_018_3_T1", + "correct_answer": "There are 3 areas in total, located in front of you and to your back-left.", + "original_options": { + "A": "There are 2 areas in total.", + "B": "There are 4 areas in total.", + "C": "There are 3 areas in total, located in front of you and to your back-left.", + "D": "There are 2 areas in total, located in front of you and to your back-left.", + "E": "There are 4 areas in total, located in front of you and to your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_4_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, which sections of the supermarket have we been to?", + "answer": "There are 5 areas in total.", + "question_chinese": "到目前为止，我们去了超市的哪几个分区？", + "answer_chinese": "总共5个区域", + "evidence": [ + { + "description": "We are now in the fruit section.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "We have now entered the refrigerated section.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:41" + } + }, + { + "description": "We have now entered the raw food section.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + }, + { + "description": "We have now entered the freshly cut meat section.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:11" + } + }, + { + "description": "We are now in the vegetable section.", + "timestamps": { + "start": "00:09:16", + "end": "00:09:17" + } + } + ], + "question_timestamps": "00:11:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 4 areas in total.", + "B": "There are 6 areas in total.", + "C": "There are 3 areas in total, not five so far at all.", + "D": "There are 5 areas in total.", + "E": "There are 7 areas in total that you have visited up to this point." + }, + "answer_label": "D", + "answer_text": "There are 5 areas in total.", + "q_id": "egolife_A1_018_4_T1", + "correct_answer": "There are 5 areas in total.", + "original_options": { + "A": "There are 4 areas in total.", + "B": "There are 6 areas in total.", + "C": "There are 3 areas in total, not five so far at all.", + "D": "There are 5 areas in total.", + "E": "There are 7 areas in total that you have visited up to this point." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_4_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, which sections of the supermarket have we visited?", + "answer": "Three sections in total.", + "question_chinese": "到目前为止，我们去了超市的哪几个分区？", + "answer_chinese": "总共5个区域", + "evidence": [ + { + "description": "We are now in the fruit section.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "We have now entered the refrigerated section.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:41" + } + }, + { + "description": "We have now entered the raw food section.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + }, + { + "description": "We have now entered the freshly cut meat section.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:11" + } + }, + { + "description": "We are now in the vegetable section.", + "timestamps": { + "start": "00:09:16", + "end": "00:09:17" + } + } + ], + "question_timestamps": "00:11:52", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four sections in total.", + "B": "Five sections in total.", + "C": "Six sections in total, meaning you’ve already been through six different supermarket sections so far.", + "D": "Seven sections in total, so you’ve visited seven distinct supermarket sections up to this point.", + "E": "Three sections in total." + }, + "answer_label": "E", + "answer_text": "Three sections in total.", + "q_id": "egolife_A1_018_4_T2", + "correct_answer": "Three sections in total.", + "original_options": { + "A": "Four sections in total.", + "B": "Five sections in total.", + "C": "Six sections in total, meaning you’ve already been through six different supermarket sections so far.", + "D": "Seven sections in total, so you’ve visited seven distinct supermarket sections up to this point.", + "E": "Three sections in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to leave the supermarket. How should I get there?", + "answer": "You should go straight to the end first, then turn left and walk to the fruit section, and then turn right and go straight again to get out of the supermarket.", + "question_chinese": "我想离开超市，我应该怎么走？", + "answer_chinese": "你应该先直走到尽头，随后左转，走到水果区，然后再右转直走就可以走出超市。", + "evidence": [ + { + "description": "The main entrance of the supermarket is on my right side.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:40" + } + }, + { + "description": "The refrigerated section is ahead and to the left of the fruit section.", + "timestamps": { + "start": "00:02:07", + "end": "00:02:08" + } + } + ], + "question_timestamps": "00:04:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn right.", + "B": "Turn left, then go straight.", + "C": "You should go straight to the end first, then turn left and walk to the fruit section, and then turn right and go straight again to get out of the supermarket.", + "D": "Go straight to the end, turn right to the fruit section, then turn left and walk straight out.", + "E": "Go straight to the end, turn left past the fruit section, then turn left again and continue straight to exit." + }, + "answer_label": "C", + "answer_text": "You should go straight to the end first, then turn left and walk to the fruit section, and then turn right and go straight again to get out of the supermarket.", + "q_id": "egolife_A1_018_1", + "correct_answer": "You should go straight to the end first, then turn left and walk to the fruit section, and then turn right and go straight again to get out of the supermarket.", + "original_options": { + "A": "Go straight, then turn right.", + "B": "Turn left, then go straight.", + "C": "You should go straight to the end first, then turn left and walk to the fruit section, and then turn right and go straight again to get out of the supermarket.", + "D": "Go straight to the end, turn right to the fruit section, then turn left and walk straight out.", + "E": "Go straight to the end, turn left past the fruit section, then turn left again and continue straight to exit." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to leave the supermarket. How should I get there?", + "answer": "You should turn right, walk onto the main aisle, then keep going straight right, and you’ll get out of the supermarket entrance.", + "question_chinese": "我想离开超市，我应该怎么走？", + "answer_chinese": "你应该向右转，走到大道上，然后再向右一直走，就能走出超市大门了。", + "evidence": [ + { + "description": "The main entrance of the supermarket is on my right side.", + "timestamps": { + "start": "00:00:39", + "end": "00:00:40" + } + }, + { + "description": "The refrigerated section is ahead and to the left of the fruit section.", + "timestamps": { + "start": "00:02:07", + "end": "00:02:08" + } + } + ], + "question_timestamps": "00:07:29", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left to the main aisle.", + "B": "Go straight onto the main aisle.", + "C": "Turn right onto the main aisle, then go straight left until you reach the entrance doors.", + "D": "Turn right, enter the main aisle, then turn left and continue straight to reach the supermarket entrance.", + "E": "You should turn right, walk onto the main aisle, then keep going straight right, and you’ll get out of the supermarket entrance." + }, + "answer_label": "E", + "answer_text": "You should turn right, walk onto the main aisle, then keep going straight right, and you’ll get out of the supermarket entrance.", + "q_id": "egolife_A1_018_2", + "correct_answer": "You should turn right, walk onto the main aisle, then keep going straight right, and you’ll get out of the supermarket entrance.", + "original_options": { + "A": "Turn left to the main aisle.", + "B": "Go straight onto the main aisle.", + "C": "Turn right onto the main aisle, then go straight left until you reach the entrance doors.", + "D": "Turn right, enter the main aisle, then turn left and continue straight to reach the supermarket entrance.", + "E": "You should turn right, walk onto the main aisle, then keep going straight right, and you’ll get out of the supermarket entrance." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_7", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the fruit-cutting bar or the raw preparation bar closer to me?", + "answer": "The freshly cut food preparation counter is closer to you now.", + "question_chinese": "是果切的吧台还是生切料理的吧台离我更近？", + "answer_chinese": "现在是生切料理的吧台离你更近。", + "evidence": [ + { + "description": "In front of us is the fruit cutting counter.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "In front of me now is the freshly cut food preparation counter.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:06:05", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The fruit-cutting bar is closer to you now.", + "B": "The freshly cut food preparation counter is closer to you now." + }, + "answer_label": "B", + "answer_text": "The freshly cut food preparation counter is closer to you now.", + "q_id": "egolife_A1_018_7", + "correct_answer": "The freshly cut food preparation counter is closer to you now.", + "original_options": { + "A": "The fruit-cutting bar is closer to you now.", + "B": "The freshly cut food preparation counter is closer to you now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_8", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the fruit-cutting bar or the raw preparation bar closer to me?", + "answer": "The fruit cutting counter is closer to you now.", + "question_chinese": "是果切的吧台还是生切料理的吧台离我更近？", + "answer_chinese": "现在是果切的吧台离你更近", + "evidence": [ + { + "description": "In front of us is the fruit cutting counter.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "In front of me now is the freshly cut food preparation counter.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:11:06", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The raw preparation bar is closer to you now.", + "B": "The fruit cutting counter is closer to you now." + }, + "answer_label": "B", + "answer_text": "The fruit cutting counter is closer to you now.", + "q_id": "egolife_A1_018_8", + "correct_answer": "The fruit cutting counter is closer to you now.", + "original_options": { + "A": "The raw preparation bar is closer to you now.", + "B": "The fruit cutting counter is closer to you now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018-mcq_fixed_8", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the fruit cutting counter from me?", + "answer": "The fruit cutting counter is behind your right side.", + "question_chinese": "果切的吧台在我的什么方向？", + "answer_chinese": "果切的吧台在你的右后方。", + "evidence": [ + { + "description": "The fruit cutting counter is right in front of me.", + "timestamps": { + "start": "00:00:49", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:05:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of your right side.", + "B": "Behind your left side.", + "C": "The fruit cutting counter is directly in front of you, slightly to the right as you face the aisle.", + "D": "The fruit cutting counter is behind your right side.", + "E": "The fruit cutting counter sits behind you to the left, so you’d turn left to reach it." + }, + "answer_label": "D", + "answer_text": "The fruit cutting counter is behind your right side.", + "q_id": "egolife_A1_018-mcq_fixed_8", + "correct_answer": "The fruit cutting counter is behind your right side.", + "original_options": { + "A": "In front of your right side.", + "B": "Behind your left side.", + "C": "The fruit cutting counter is directly in front of you, slightly to the right as you face the aisle.", + "D": "The fruit cutting counter is behind your right side.", + "E": "The fruit cutting counter sits behind you to the left, so you’d turn left to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018-mcq_fixed_9", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the fruit cutting counter from me?", + "answer": "The fruit cutting counter is right in front of your right side.", + "question_chinese": "果切的吧台在我的什么方向？", + "answer_chinese": "果切的吧台在你的右前方。", + "evidence": [ + { + "description": "The fruit cutting counter is right in front of me.", + "timestamps": { + "start": "00:00:49", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:09:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right behind your right side.", + "B": "In front of your left side.", + "C": "The fruit cutting counter is right in front of your right side.", + "D": "The fruit cutting counter is directly in front of you, centered ahead rather than off to your right.", + "E": "The fruit cutting counter is behind and to your left, so you’d need to pivot left to see it." + }, + "answer_label": "C", + "answer_text": "The fruit cutting counter is right in front of your right side.", + "q_id": "egolife_A1_018-mcq_fixed_9", + "correct_answer": "The fruit cutting counter is right in front of your right side.", + "original_options": { + "A": "Right behind your right side.", + "B": "In front of your left side.", + "C": "The fruit cutting counter is right in front of your right side.", + "D": "The fruit cutting counter is directly in front of you, centered ahead rather than off to your right.", + "E": "The fruit cutting counter is behind and to your left, so you’d need to pivot left to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_11", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the mango from me?", + "answer": "The mangoes are on your left side.", + "question_chinese": "芒果在我的什么方向？", + "answer_chinese": "芒果在你的左侧", + "evidence": [ + { + "description": "There are mangoes right in front of me.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:05:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your right side.", + "B": "Directly in front of you.", + "C": "The mangoes are on your left side.", + "D": "The mangoes are behind you on the left, so you’d have to turn around to spot them.", + "E": "The mangoes sit ahead to your right, slightly angled away from where you’re currently facing." + }, + "answer_label": "C", + "answer_text": "The mangoes are on your left side.", + "q_id": "egolife_A1_018_11", + "correct_answer": "The mangoes are on your left side.", + "original_options": { + "A": "On your right side.", + "B": "Directly in front of you.", + "C": "The mangoes are on your left side.", + "D": "The mangoes are behind you on the left, so you’d have to turn around to spot them.", + "E": "The mangoes sit ahead to your right, slightly angled away from where you’re currently facing." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_12", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the mango from me?", + "answer": "The mangoes are right in front of your right side.", + "question_chinese": "芒果在我的什么方向？", + "answer_chinese": "芒果在你的右前方。", + "evidence": [ + { + "description": "There are mangoes right in front of me.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:11:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right in front of you.", + "B": "Behind your right side.", + "C": "The mangoes are right in front of your left side, so they’re ahead but not on your right.", + "D": "The mangoes are right in front of your right side.", + "E": "The mangoes are to your right but slightly behind you, requiring a small turn back to see them." + }, + "answer_label": "D", + "answer_text": "The mangoes are right in front of your right side.", + "q_id": "egolife_A1_018_12", + "correct_answer": "The mangoes are right in front of your right side.", + "original_options": { + "A": "Right in front of you.", + "B": "Behind your right side.", + "C": "The mangoes are right in front of your left side, so they’re ahead but not on your right.", + "D": "The mangoes are right in front of your right side.", + "E": "The mangoes are to your right but slightly behind you, requiring a small turn back to see them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_3", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the supermarket's main entrance from me?", + "answer": "The main entrance of the supermarket is behind and to your left.", + "question_chinese": "超市的大门在我的什么方向？", + "answer_chinese": "超市的大门在你的左后方。", + "evidence": [ + { + "description": "The main entrance of the supermarket is ahead and to my right.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:07:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Ahead and to your left.", + "B": "Behind and to your right.", + "C": "The main entrance is directly in front of you, so you can walk straight without turning left.", + "D": "The main entrance of the supermarket is behind and to your left.", + "E": "The main entrance is in front and to your right, meaning you’d head forward then angle right." + }, + "answer_label": "D", + "answer_text": "The main entrance of the supermarket is behind and to your left.", + "q_id": "egolife_A1_018_3", + "correct_answer": "The main entrance of the supermarket is behind and to your left.", + "original_options": { + "A": "Ahead and to your left.", + "B": "Behind and to your right.", + "C": "The main entrance is directly in front of you, so you can walk straight without turning left.", + "D": "The main entrance of the supermarket is behind and to your left.", + "E": "The main entrance is in front and to your right, meaning you’d head forward then angle right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_018", + "video_length": "00:12:00", + "qid": "egolife_A1_018_4", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the supermarket's main entrance from me?", + "answer": "The main entrance of the supermarket is behind you.", + "question_chinese": "超市的大门在我的什么方向？", + "answer_chinese": "超市的大门在你的后方。", + "evidence": [ + { + "description": "The main entrance of the supermarket is ahead and to my right.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:11:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "The main entrance of the supermarket is behind you.", + "C": "On your left.", + "D": "The main entrance is ahead of you to the right, so you’d move forward and angle right.", + "E": "The main entrance is behind and to your left, meaning you’d turn left as you look back." + }, + "answer_label": "B", + "answer_text": "The main entrance of the supermarket is behind you.", + "q_id": "egolife_A1_018_4", + "correct_answer": "The main entrance of the supermarket is behind you.", + "original_options": { + "A": "In front of you.", + "B": "The main entrance of the supermarket is behind you.", + "C": "On your left.", + "D": "The main entrance is ahead of you to the right, so you’d move forward and angle right.", + "E": "The main entrance is behind and to your left, meaning you’d turn left as you look back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 18, + 18 + ] + } +] diff --git a/QAs/egolife_A1_019-mcq.json b/QAs/egolife_A1_019-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..078609fc4511ef182e85c28d569b8fbc98f18f2f --- /dev/null +++ b/QAs/egolife_A1_019-mcq.json @@ -0,0 +1,875 @@ +[ + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the freezer in the refrigerated and frozen section?", + "answer": "On your left side.", + "question_chinese": "冷藏冷冻区的冰柜在我的什么方向？", + "answer_chinese": "在你的左侧。", + "evidence": [ + { + "description": "What is in front of me is the freezer in the refrigerated and frozen section.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your right side.", + "B": "On your left side.", + "C": "Directly in front of you.", + "D": "The freezer is behind you on the left, so you’d need to turn around to reach it.", + "E": "The freezer is ahead and to your right, positioned diagonally forward from where you’re standing." + }, + "answer_label": "B", + "answer_text": "On your left side.", + "q_id": "egolife_A1_019_1", + "correct_answer": "On your left side.", + "original_options": { + "A": "On your right side.", + "B": "On your left side.", + "C": "Directly in front of you.", + "D": "The freezer is behind you on the left, so you’d need to turn around to reach it.", + "E": "The freezer is ahead and to your right, positioned diagonally forward from where you’re standing." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the freezer in the refrigerated and frozen section?", + "answer": "Right in front of your right side.", + "question_chinese": "冷藏冷冻区的冰柜在我的什么方向？", + "answer_chinese": "在你的右前方。", + "evidence": [ + { + "description": "What is in front of me is the freezer in the refrigerated and frozen section.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:11:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your left side.", + "B": "Behind your right side.", + "C": "The freezer is directly in front of you, centered ahead rather than positioned to your right.", + "D": "Right in front of your right side.", + "E": "The freezer is in front of your left side, so you’d step forward and angle left to reach it." + }, + "answer_label": "D", + "answer_text": "Right in front of your right side.", + "q_id": "egolife_A1_019_2", + "correct_answer": "Right in front of your right side.", + "original_options": { + "A": "On your left side.", + "B": "Behind your right side.", + "C": "The freezer is directly in front of you, centered ahead rather than positioned to your right.", + "D": "Right in front of your right side.", + "E": "The freezer is in front of your left side, so you’d step forward and angle left to reach it." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 20 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_13", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the ice cream freezer from me?", + "answer": "The ice cream freezer is right in front of you on your right.", + "question_chinese": "卖雪糕的冰柜在我的什么方向？", + "answer_chinese": "卖雪糕的冰柜在你的右前方", + "evidence": [ + { + "description": "There is an ice cream freezer right in front of me.", + "timestamps": { + "start": "00:00:41", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:05:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front on your left.", + "B": "Behind you on your right.", + "C": "The ice cream freezer is right in front of you on your right.", + "D": "The ice cream freezer is straight ahead, but it sits slightly to your left side.", + "E": "The ice cream freezer is behind you and a bit to your left, not in front on the right." + }, + "answer_label": "C", + "answer_text": "The ice cream freezer is right in front of you on your right.", + "q_id": "egolife_A1_019_13", + "correct_answer": "The ice cream freezer is right in front of you on your right.", + "original_options": { + "A": "In front on your left.", + "B": "Behind you on your right.", + "C": "The ice cream freezer is right in front of you on your right.", + "D": "The ice cream freezer is straight ahead, but it sits slightly to your left side.", + "E": "The ice cream freezer is behind you and a bit to your left, not in front on the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_14", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the ice cream freezer from me?", + "answer": "The ice cream freezer is right in front of you on your left.", + "question_chinese": "卖雪糕的冰柜在我的什么方向？", + "answer_chinese": "卖雪糕的冰柜在你的左前方。", + "evidence": [ + { + "description": "There is an ice cream freezer right in front of me.", + "timestamps": { + "start": "00:00:41", + "end": "00:00:42" + } + } + ], + "question_timestamps": "00:11:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front on your right.", + "B": "The ice cream freezer is right in front of you on your left.", + "C": "Behind you on your left.", + "D": "The ice cream freezer is in front of you, but it’s positioned more toward your right side.", + "E": "The ice cream freezer is off to your far left, rather than being right in front of you." + }, + "answer_label": "B", + "answer_text": "The ice cream freezer is right in front of you on your left.", + "q_id": "egolife_A1_019_14", + "correct_answer": "The ice cream freezer is right in front of you on your left.", + "original_options": { + "A": "In front on your right.", + "B": "The ice cream freezer is right in front of you on your left.", + "C": "Behind you on your left.", + "D": "The ice cream freezer is in front of you, but it’s positioned more toward your right side.", + "E": "The ice cream freezer is off to your far left, rather than being right in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_5", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the checkout area to pay, how should I get there?", + "answer": "You should turn left, go straight until you're near the gate, then turn left again and walk five steps to reach the checkout area.", + "question_chinese": "如果我想去收银区结账，我应该怎么走？", + "answer_chinese": "你应该向左转并直走到大门附近，然后向左转走5步的距离就能到达收银区了。", + "evidence": [ + { + "description": "The checkout area is right in front of your right side.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The distance from the checkout area to the gate is about half the width of the supermarket.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:06:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You should turn left, go straight until you're near the gate, then turn left again and walk five steps to reach the checkout area.", + "B": "Turn right and head to the gate.", + "C": "Go straight, then turn right twice.", + "D": "Turn left, walk straight past the gate, then turn left and take about ten steps to checkout.", + "E": "Turn left, go straight until you reach the gate, then turn right and walk six steps to the checkout." + }, + "answer_label": "A", + "answer_text": "You should turn left, go straight until you're near the gate, then turn left again and walk five steps to reach the checkout area.", + "q_id": "egolife_A1_019_5", + "correct_answer": "You should turn left, go straight until you're near the gate, then turn left again and walk five steps to reach the checkout area.", + "original_options": { + "A": "You should turn left, go straight until you're near the gate, then turn left again and walk five steps to reach the checkout area.", + "B": "Turn right and head to the gate.", + "C": "Go straight, then turn right twice.", + "D": "Turn left, walk straight past the gate, then turn left and take about ten steps to checkout.", + "E": "Turn left, go straight until you reach the gate, then turn right and walk six steps to the checkout." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_6", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the checkout area to pay, how should I get there?", + "answer": "You should go straight until you reach the main road ahead, then turn right and keep walking until you get to the gate. After that, turn left and walk five steps, and you will arrive at the checkout area.", + "question_chinese": "如果我想去收银区结账，我应该怎么走？", + "answer_chinese": "你应该直走到前面的主干道上，然后向右转，一直走到大门，随后左转走五步的距离就能到达收银区了。", + "evidence": [ + { + "description": "The checkout area is right in front of your right side.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The distance from the checkout area to the gate is about half the width of the supermarket.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:11:59", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn left at the gate.", + "B": "Turn right first, then walk to the gate.", + "C": "You should go straight until you reach the main road ahead, then turn right and keep walking until you get to the gate. After that, turn left and walk five steps, and you will arrive at the checkout area.", + "D": "Go straight until the main road, turn left, and keep walking until you reach the gate.", + "E": "Go straight to the main road, turn right, but then turn right again at the gate and continue forward." + }, + "answer_label": "C", + "answer_text": "You should go straight until you reach the main road ahead, then turn right and keep walking until you get to the gate. After that, turn left and walk five steps, and you will arrive at the checkout area.", + "q_id": "egolife_A1_019_6", + "correct_answer": "You should go straight until you reach the main road ahead, then turn right and keep walking until you get to the gate. After that, turn left and walk five steps, and you will arrive at the checkout area.", + "original_options": { + "A": "Go straight, then turn left at the gate.", + "B": "Turn right first, then walk to the gate.", + "C": "You should go straight until you reach the main road ahead, then turn right and keep walking until you get to the gate. After that, turn left and walk five steps, and you will arrive at the checkout area.", + "D": "Go straight until the main road, turn left, and keep walking until you reach the gate.", + "E": "Go straight to the main road, turn right, but then turn right again at the gate and continue forward." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many areas did we pass through in total?", + "answer": "We have passed through four area so far.", + "question_chinese": "我们一共经过了几个区域？", + "answer_chinese": "到目前为止经过了4个区域。", + "evidence": [ + { + "description": "We are now in the refrigerated and frozen section.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "We are now in the meat section.", + "timestamps": { + "start": "00:05:11", + "end": "00:05:12" + } + }, + { + "description": "We are passing through the deli section now.", + "timestamps": { + "start": "00:06:21", + "end": "00:06:26" + } + }, + { + "description": "We are now in the condiment section.", + "timestamps": { + "start": "00:06:51", + "end": "00:07:02" + } + } + ], + "question_timestamps": "00:08:49", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "We passed through three areas.", + "B": "We have passed through four area so far.", + "C": "We passed through five areas.", + "D": "We have passed through two areas so far in total, not four.", + "E": "We have passed through six areas so far, counting each distinct section we entered." + }, + "answer_label": "B", + "answer_text": "We have passed through four area so far.", + "q_id": "egolife_A1_019_3", + "correct_answer": "We have passed through four area so far.", + "original_options": { + "A": "We passed through three areas.", + "B": "We have passed through four area so far.", + "C": "We passed through five areas.", + "D": "We have passed through two areas so far in total, not four.", + "E": "We have passed through six areas so far, counting each distinct section we entered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many areas did we pass through in total?", + "answer": "We have passed through five areas so far.", + "question_chinese": "我们一共经过了几个区域？", + "answer_chinese": "到目前为止，经过了5个区域", + "evidence": [ + { + "description": "We are now in the refrigerated and frozen section.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "We are now in the meat section.", + "timestamps": { + "start": "00:05:11", + "end": "00:05:12" + } + }, + { + "description": "We are passing through the deli section now.", + "timestamps": { + "start": "00:06:21", + "end": "00:06:26" + } + }, + { + "description": "We are now in the condiment section.", + "timestamps": { + "start": "00:06:51", + "end": "00:07:02" + } + }, + { + "description": "We are passing through the chilled ready-to-eat section now.", + "timestamps": { + "start": "00:09:58", + "end": "00:09:59" + } + } + ], + "question_timestamps": "00:11:18", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "We have passed through five areas so far.", + "B": "We passed through four areas.", + "C": "We passed through six areas.", + "D": "We have passed through three areas so far in total, not five.", + "E": "We have passed through seven areas so far, including the last section you just entered." + }, + "answer_label": "A", + "answer_text": "We have passed through five areas so far.", + "q_id": "egolife_A1_019_4", + "correct_answer": "We have passed through five areas so far.", + "original_options": { + "A": "We have passed through five areas so far.", + "B": "We passed through four areas.", + "C": "We passed through six areas.", + "D": "We have passed through three areas so far in total, not five.", + "E": "We have passed through seven areas so far, including the last section you just entered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_9", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I take the frozen and refrigerated vegetables now?", + "answer": "Yes, it’s two steps away on your right.", + "question_chinese": "我现在想拿冷冻冷藏区的蔬菜，我可以拿到吗？", + "answer_chinese": "可以，他就在你右手边两步的距离。", + "evidence": [ + { + "description": "The refrigerated cabinets in the refrigerated and frozen section are right in front of me.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + }, + { + "description": "he refrigerated cabinet is about two steps away from me.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:04:14", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, it’s two steps away on your right.", + "B": "No, it’s too far away to reach from here." + }, + "answer_label": "A", + "answer_text": "Yes, it’s two steps away on your right.", + "q_id": "egolife_A1_019_9", + "correct_answer": "Yes, it’s two steps away on your right.", + "original_options": { + "A": "Yes, it’s two steps away on your right.", + "B": "No, it’s too far away to reach from here." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_10", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I take the frozen and refrigerated vegetables now?", + "answer": "No, the refrigerated cabinet is currently some distance away from you.", + "question_chinese": "我现在想拿冷冻冷藏区的蔬菜，我可以拿到吗？", + "answer_chinese": "不行，冷藏柜现在离你有一定的距离。", + "evidence": [ + { + "description": "The refrigerated cabinets in the refrigerated and frozen section are right in front of me.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + }, + { + "description": "he refrigerated cabinet is about two steps away from me.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:10:29", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, the refrigerated cabinet is within easy reach right now.", + "B": "No, the refrigerated cabinet is currently some distance away from you." + }, + "answer_label": "B", + "answer_text": "No, the refrigerated cabinet is currently some distance away from you.", + "q_id": "egolife_A1_019_10", + "correct_answer": "No, the refrigerated cabinet is currently some distance away from you.", + "original_options": { + "A": "Yes, the refrigerated cabinet is within easy reach right now.", + "B": "No, the refrigerated cabinet is currently some distance away from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_7", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the checkout area from where I am?", + "answer": "The checkout area is behind your right side now.", + "question_chinese": "收银区现在在我的什么方向？", + "answer_chinese": "收银区现在在你的右后方。", + "evidence": [ + { + "description": "The checkout area is right in front of your right side.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:05:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The checkout area is behind your right side now.", + "B": "The checkout area is beside your left side.", + "C": "The checkout area is behind your left side.", + "D": "The checkout area is in front of you, slightly toward your left as you face down the aisle.", + "E": "The checkout area is ahead and to your right, not behind you from where you’re standing." + }, + "answer_label": "A", + "answer_text": "The checkout area is behind your right side now.", + "q_id": "egolife_A1_019_7", + "correct_answer": "The checkout area is behind your right side now.", + "original_options": { + "A": "The checkout area is behind your right side now.", + "B": "The checkout area is beside your left side.", + "C": "The checkout area is behind your left side.", + "D": "The checkout area is in front of you, slightly toward your left as you face down the aisle.", + "E": "The checkout area is ahead and to your right, not behind you from where you’re standing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_8", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the checkout area from where I am?", + "answer": "The checkout area is right in front of your right side now.", + "question_chinese": "收银区现在在我的什么方向？", + "answer_chinese": "收银区现在在你的右前方", + "evidence": [ + { + "description": "The checkout area is right in front of your right side.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:11:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The checkout area is straight ahead.", + "B": "The checkout area is right in front of your right side now.", + "C": "The checkout area is behind you.", + "D": "The checkout area is on your left side now, roughly level with where you’re standing.", + "E": "The checkout area is in front of your left side, so you’d turn left slightly to face it." + }, + "answer_label": "B", + "answer_text": "The checkout area is right in front of your right side now.", + "q_id": "egolife_A1_019_8", + "correct_answer": "The checkout area is right in front of your right side now.", + "original_options": { + "A": "The checkout area is straight ahead.", + "B": "The checkout area is right in front of your right side now.", + "C": "The checkout area is behind you.", + "D": "The checkout area is on your left side now, roughly level with where you’re standing.", + "E": "The checkout area is in front of your left side, so you’d turn left slightly to face it." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_6_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the supermarket’s main entrance?", + "answer": "The supermarket’s main entrance was behind you.", + "question_chinese": "超市的大门在我的什么方向？", + "answer_chinese": "超市的大门在你的正后方。", + "evidence": [ + { + "description": "The supermarket gate is right in front of my right side.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:05:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The main entrance was in front of you.", + "B": "The main entrance was to your left.", + "C": "The supermarket’s main entrance was behind you.", + "D": "The supermarket’s main entrance was off to your right side as you looked down the aisle.", + "E": "The supermarket’s main entrance was ahead of you rather than behind, in the direction you were moving." + }, + "answer_label": "C", + "answer_text": "The supermarket’s main entrance was behind you.", + "q_id": "egolife_A1_019_6_T1", + "correct_answer": "The supermarket’s main entrance was behind you.", + "original_options": { + "A": "The main entrance was in front of you.", + "B": "The main entrance was to your left.", + "C": "The supermarket’s main entrance was behind you.", + "D": "The supermarket’s main entrance was off to your right side as you looked down the aisle.", + "E": "The supermarket’s main entrance was ahead of you rather than behind, in the direction you were moving." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_019", + "video_length": "00:12:00", + "qid": "egolife_A1_019_6_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the supermarket’s main entrance?", + "answer": "The supermarket’s main entrance was on the right side. The supermarket’s main entrance was on the right side.", + "question_chinese": "超市的大门在我的什么方向？", + "answer_chinese": "超市的大门在你的右侧。超市的大门在你的右侧。", + "evidence": [ + { + "description": "The supermarket gate is right in front of my right side.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:11:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The main entrance was on your left side.", + "B": "The main entrance was in front of you.", + "C": "The supermarket’s main entrance was behind you. The supermarket’s main entrance was behind you.", + "D": "The supermarket’s main entrance was on the right side. The supermarket’s main entrance was on the right side.", + "E": "The supermarket’s main entrance was on the left side. The supermarket’s main entrance was on the left side." + }, + "answer_label": "D", + "answer_text": "The supermarket’s main entrance was on the right side. The supermarket’s main entrance was on the right side.", + "q_id": "egolife_A1_019_6_T2", + "correct_answer": "The supermarket’s main entrance was on the right side. The supermarket’s main entrance was on the right side.", + "original_options": { + "A": "The main entrance was on your left side.", + "B": "The main entrance was in front of you.", + "C": "The supermarket’s main entrance was behind you. The supermarket’s main entrance was behind you.", + "D": "The supermarket’s main entrance was on the right side. The supermarket’s main entrance was on the right side.", + "E": "The supermarket’s main entrance was on the left side. The supermarket’s main entrance was on the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 20 + ] + } +] diff --git a/QAs/egolife_A1_020-mcq.json b/QAs/egolife_A1_020-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..37f4e07802c6f0c0e598359e7b51a626f5a3bcb6 --- /dev/null +++ b/QAs/egolife_A1_020-mcq.json @@ -0,0 +1,1150 @@ +[ + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_1_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many areas of the supermarket have we passed through in total?", + "answer": "You went through two areas in total: the snack section and the seafood section.", + "question_chinese": "到目前为止，我们总共经过了超市的几个区域？", + "answer_chinese": "你们总共经过了2个区域。经过了零食区和海鲜区。", + "evidence": [ + { + "description": "We're in the snack section right now.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:10" + } + }, + { + "description": "We're in the seafood section right now.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You went through one area in total.", + "B": "You went through three areas in total.", + "C": "You went through one area in total: the snack section, and you did not enter the seafood section.", + "D": "You went through three areas in total: the snack section, the seafood section, and the deli section.", + "E": "You went through two areas in total: the snack section and the seafood section." + }, + "answer_label": "E", + "answer_text": "You went through two areas in total: the snack section and the seafood section.", + "q_id": "egolife_A1_020_1_T2", + "correct_answer": "You went through two areas in total: the snack section and the seafood section.", + "original_options": { + "A": "You went through one area in total.", + "B": "You went through three areas in total.", + "C": "You went through one area in total: the snack section, and you did not enter the seafood section.", + "D": "You went through three areas in total: the snack section, the seafood section, and the deli section.", + "E": "You went through two areas in total: the snack section and the seafood section." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_1_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many areas of the supermarket have we passed through in total?", + "answer": "You went through three areas in total: the snack section, the seafood section, and the deli section.", + "question_chinese": "到目前为止，我们总共经过了超市的几个区域？", + "answer_chinese": "你们总共经过了三个区域。经过了零食区，海鲜区和熟食区。", + "evidence": [ + { + "description": "We're in the snack section right now.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:10" + } + }, + { + "description": "We're in the seafood section right now.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:08" + } + }, + { + "description": "We're passing through the deli section right now.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:16" + } + } + ], + "question_timestamps": "00:05:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You went through two areas in total.", + "B": "You went through four areas in total.", + "C": "You went through one area in total: the snack section, without passing through seafood or the deli.", + "D": "You went through three areas in total: the snack section, the seafood section, and the deli section.", + "E": "You went through five areas in total: the snack, seafood, deli, produce, and bakery sections." + }, + "answer_label": "D", + "answer_text": "You went through three areas in total: the snack section, the seafood section, and the deli section.", + "q_id": "egolife_A1_020_1_T1", + "correct_answer": "You went through three areas in total: the snack section, the seafood section, and the deli section.", + "original_options": { + "A": "You went through two areas in total.", + "B": "You went through four areas in total.", + "C": "You went through one area in total: the snack section, without passing through seafood or the deli.", + "D": "You went through three areas in total: the snack section, the seafood section, and the deli section.", + "E": "You went through five areas in total: the snack, seafood, deli, produce, and bakery sections." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_2_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many areas of the supermarket have we passed through in total?", + "answer": "So far, you’ve gone through 4 areas in total.", + "question_chinese": "到目前为止，我们总共经过了超市的几个区域？", + "answer_chinese": "到目前为止，你们总共经过了4个区域。", + "evidence": [ + { + "description": "We're in the snack section right now.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:10" + } + }, + { + "description": "We're in the seafood section right now.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:08" + } + }, + { + "description": "We're passing through the deli section right now.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:16" + } + }, + { + "description": "I'm at the fresh meat counter right now.", + "timestamps": { + "start": "00:07:28", + "end": "00:07:29" + } + }, + { + "description": "I'm passing through the frozen and refrigerated section right now.", + "timestamps": { + "start": "00:07:49", + "end": "00:07:51" + } + } + ], + "question_timestamps": "00:07:55", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, you’ve gone through 4 areas in total.", + "B": "So far, you’ve gone through 3 areas.", + "C": "So far, you’ve gone through 5 areas.", + "D": "So far, you’ve gone through 2 areas in total, not as many as four up to this point.", + "E": "So far, you’ve gone through 6 areas in total, meaning you passed two more areas than four." + }, + "answer_label": "A", + "answer_text": "So far, you’ve gone through 4 areas in total.", + "q_id": "egolife_A1_020_2_T2", + "correct_answer": "So far, you’ve gone through 4 areas in total.", + "original_options": { + "A": "So far, you’ve gone through 4 areas in total.", + "B": "So far, you’ve gone through 3 areas.", + "C": "So far, you’ve gone through 5 areas.", + "D": "So far, you’ve gone through 2 areas in total, not as many as four up to this point.", + "E": "So far, you’ve gone through 6 areas in total, meaning you passed two more areas than four." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_2_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many areas of the supermarket have we passed through in total?", + "answer": "So far, you’ve gone through 5 areas in total.", + "question_chinese": "到目前为止，我们总共经过了超市的几个区域？", + "answer_chinese": "到目前为止，总共经过了5个区域。", + "evidence": [ + { + "description": "We're in the snack section right now.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:10" + } + }, + { + "description": "We're in the seafood section right now.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:08" + } + }, + { + "description": "We're passing through the deli section right now.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:16" + } + }, + { + "description": "I'm at the fresh meat counter right now.", + "timestamps": { + "start": "00:07:28", + "end": "00:07:29" + } + }, + { + "description": "I'm passing through the frozen and refrigerated section right now.", + "timestamps": { + "start": "00:07:49", + "end": "00:07:51" + } + }, + { + "description": "I'm passing through the fruit section right now.", + "timestamps": { + "start": "00:08:15", + "end": "00:08:16" + } + } + ], + "question_timestamps": "00:09:52", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, you’ve gone through 4 areas.", + "B": "So far, you’ve gone through 6 areas.", + "C": "So far, you’ve gone through 3 areas in total, which is fewer than five up to now.", + "D": "So far, you’ve gone through 7 areas in total, meaning you passed two more areas than five.", + "E": "So far, you’ve gone through 5 areas in total." + }, + "answer_label": "E", + "answer_text": "So far, you’ve gone through 5 areas in total.", + "q_id": "egolife_A1_020_2_T1", + "correct_answer": "So far, you’ve gone through 5 areas in total.", + "original_options": { + "A": "So far, you’ve gone through 4 areas.", + "B": "So far, you’ve gone through 6 areas.", + "C": "So far, you’ve gone through 3 areas in total, which is fewer than five up to now.", + "D": "So far, you’ve gone through 7 areas in total, meaning you passed two more areas than five.", + "E": "So far, you’ve gone through 5 areas in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_5", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to find some snacks, where should I go?", + "answer": "You should walk forward about three steps, then turn right and you'll see the snack cabinet.", + "question_chinese": "我如果想去找一些零食，我应该去哪？", + "answer_chinese": "你应该向前走三步左右的距离，然后向右转就能看到零食柜子了。", + "evidence": [ + { + "description": "There is a row of cabinets for snacks in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "The snack cabinet is right beside the main aisle.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:04:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk forward two steps, then turn right.", + "B": "Walk forward three steps, then turn left.", + "C": "Walk forward about four steps, then turn right and you'll see the snack cabinet ahead of you.", + "D": "Take three steps forward, turn left, and the snack cabinet should be visible along that side.", + "E": "You should walk forward about three steps, then turn right and you'll see the snack cabinet." + }, + "answer_label": "E", + "answer_text": "You should walk forward about three steps, then turn right and you'll see the snack cabinet.", + "q_id": "egolife_A1_020_5", + "correct_answer": "You should walk forward about three steps, then turn right and you'll see the snack cabinet.", + "original_options": { + "A": "Walk forward two steps, then turn right.", + "B": "Walk forward three steps, then turn left.", + "C": "Walk forward about four steps, then turn right and you'll see the snack cabinet ahead of you.", + "D": "Take three steps forward, turn left, and the snack cabinet should be visible along that side.", + "E": "You should walk forward about three steps, then turn right and you'll see the snack cabinet." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_6", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to find some snacks, where should I go?", + "answer": "You should walk straight forward for six steps, then turn left and you'll see the snack cabinet.", + "question_chinese": "我如果想去找一些零食，我应该去哪？", + "answer_chinese": "你应该向前直走6步的距离，然后向左转就能看到放零食的柜子了。", + "evidence": [ + { + "description": "There is a row of cabinets for snacks in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "The snack cabinet is right beside the main aisle.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:08:37", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk six steps, then turn right.", + "B": "Walk five steps, then turn left.", + "C": "Walk straight forward for seven steps, then turn left and you'll see the snack cabinet near you.", + "D": "You should walk straight forward for six steps, then turn left and you'll see the snack cabinet.", + "E": "Go straight ahead for six steps, turn right, and you should spot the snack cabinet on that side." + }, + "answer_label": "D", + "answer_text": "You should walk straight forward for six steps, then turn left and you'll see the snack cabinet.", + "q_id": "egolife_A1_020_6", + "correct_answer": "You should walk straight forward for six steps, then turn left and you'll see the snack cabinet.", + "original_options": { + "A": "Walk six steps, then turn right.", + "B": "Walk five steps, then turn left.", + "C": "Walk straight forward for seven steps, then turn left and you'll see the snack cabinet near you.", + "D": "You should walk straight forward for six steps, then turn left and you'll see the snack cabinet.", + "E": "Go straight ahead for six steps, turn right, and you should spot the snack cabinet on that side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_7_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the cabinet for seafood relative to me?", + "answer": "The cabinet with seafood is in front of you.", + "question_chinese": "放海鲜的柜子在我的什么方向？", + "answer_chinese": "放海鲜的柜子在你的正前方。", + "evidence": [ + { + "description": "What's directly in front of me is the seafood cabinet.", + "timestamps": { + "start": "00:03:34", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:03:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The cabinet with seafood is in front of you.", + "B": "It is behind you.", + "C": "It is to your left.", + "D": "The seafood cabinet is back-left of you, so you would need to turn slightly to face it.", + "E": "The seafood cabinet is to your back-right, so you’d need to turn slightly to face it." + }, + "answer_label": "A", + "answer_text": "The cabinet with seafood is in front of you.", + "q_id": "egolife_A1_020_7_T2", + "correct_answer": "The cabinet with seafood is in front of you.", + "original_options": { + "A": "The cabinet with seafood is in front of you.", + "B": "It is behind you.", + "C": "It is to your left.", + "D": "The seafood cabinet is back-left of you, so you would need to turn slightly to face it.", + "E": "The seafood cabinet is to your back-right, so you’d need to turn slightly to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_11_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the cabinet for seafood relative to me?", + "answer": "The cabinet with seafood is to your back-right.", + "question_chinese": "放海鲜的柜子在我的什么方向？", + "answer_chinese": "放海鲜的柜子在你的右后方。", + "evidence": [ + { + "description": "The seafood cabinet is behind my right side.", + "timestamps": { + "start": "00:03:55", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:04:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your back-left.", + "B": "It’s to your front-right.", + "C": "The cabinet with seafood is in front of you, directly ahead in your current line of sight.", + "D": "The cabinet with seafood is to your back-right.", + "E": "The cabinet with seafood is to your front-left, so you should look forward and slightly left." + }, + "answer_label": "D", + "answer_text": "The cabinet with seafood is to your back-right.", + "q_id": "egolife_A1_020_11_T2", + "correct_answer": "The cabinet with seafood is to your back-right.", + "original_options": { + "A": "It’s to your back-left.", + "B": "It’s to your front-right.", + "C": "The cabinet with seafood is in front of you, directly ahead in your current line of sight.", + "D": "The cabinet with seafood is to your back-right.", + "E": "The cabinet with seafood is to your front-left, so you should look forward and slightly left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_7_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the cabinet for seafood relative to me?", + "answer": "The cabinet with seafood is to your back-left.", + "question_chinese": "放海鲜的柜子在我的什么方向？", + "answer_chinese": "放海鲜的柜子在你的左后方", + "evidence": [ + { + "description": "What's in front of me is the seafood cabinet.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:06:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your back-right.", + "B": "It’s to your front-left.", + "C": "The cabinet with seafood is in front of you, so you can see it without turning around.", + "D": "The cabinet with seafood is to your front-right, meaning it’s ahead of you and slightly right.", + "E": "The cabinet with seafood is to your back-left." + }, + "answer_label": "E", + "answer_text": "The cabinet with seafood is to your back-left.", + "q_id": "egolife_A1_020_7_T1", + "correct_answer": "The cabinet with seafood is to your back-left.", + "original_options": { + "A": "It’s to your back-right.", + "B": "It’s to your front-left.", + "C": "The cabinet with seafood is in front of you, so you can see it without turning around.", + "D": "The cabinet with seafood is to your front-right, meaning it’s ahead of you and slightly right.", + "E": "The cabinet with seafood is to your back-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_11_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the cabinet for seafood relative to me?", + "answer": "The cabinet with seafood is to your front-left.", + "question_chinese": "放海鲜的柜子在我的什么方向？", + "answer_chinese": "放海鲜的柜子在你的左前方。", + "evidence": [ + { + "description": "What's in front of me is the seafood cabinet.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:10:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your front-right.", + "B": "It’s to your back-left.", + "C": "The cabinet with seafood is to your back-right, so it’s behind you and slightly to the right.", + "D": "The cabinet with seafood is to your front-left.", + "E": "The cabinet with seafood is directly in front of you, straight ahead from where you’re standing." + }, + "answer_label": "D", + "answer_text": "The cabinet with seafood is to your front-left.", + "q_id": "egolife_A1_020_11_T1", + "correct_answer": "The cabinet with seafood is to your front-left.", + "original_options": { + "A": "It’s to your front-right.", + "B": "It’s to your back-left.", + "C": "The cabinet with seafood is to your back-right, so it’s behind you and slightly to the right.", + "D": "The cabinet with seafood is to your front-left.", + "E": "The cabinet with seafood is directly in front of you, straight ahead from where you’re standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_3_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the cabinet for snacks relative to me?", + "answer": "The snack cabinet is to your back-right.", + "question_chinese": "放零食的柜子在我的什么方向？", + "answer_chinese": "零食的柜子在你的右后方", + "evidence": [ + { + "description": "There is a row of cabinets for snacks in front of me.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:05:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The snack cabinet is to your back-right.", + "B": "It’s to your back-left.", + "C": "It’s to your front-right.", + "D": "The snack cabinet is to your front-left, so look ahead of you and slightly to the left.", + "E": "The snack cabinet is directly in front of you, so you can face forward to see it." + }, + "answer_label": "A", + "answer_text": "The snack cabinet is to your back-right.", + "q_id": "egolife_A1_020_3_T1", + "correct_answer": "The snack cabinet is to your back-right.", + "original_options": { + "A": "The snack cabinet is to your back-right.", + "B": "It’s to your back-left.", + "C": "It’s to your front-right.", + "D": "The snack cabinet is to your front-left, so look ahead of you and slightly to the left.", + "E": "The snack cabinet is directly in front of you, so you can face forward to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_3_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the cabinet for snacks relative to me?", + "answer": "The cabinet with snacks is to your front-left.", + "question_chinese": "放零食的柜子在我的什么方向？", + "answer_chinese": "放零食的柜子在你的左前方。", + "evidence": [ + { + "description": "There is a row of cabinets for snacks in front of me.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:11:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your front-right.", + "B": "It’s to your back-left.", + "C": "The cabinet with snacks is to your front-left.", + "D": "The cabinet with snacks is to your back-right, so it’s behind you and a bit to the right.", + "E": "The cabinet with snacks is straight in front of you, directly ahead of your current position." + }, + "answer_label": "C", + "answer_text": "The cabinet with snacks is to your front-left.", + "q_id": "egolife_A1_020_3_T2", + "correct_answer": "The cabinet with snacks is to your front-left.", + "original_options": { + "A": "It’s to your front-right.", + "B": "It’s to your back-left.", + "C": "The cabinet with snacks is to your front-left.", + "D": "The cabinet with snacks is to your back-right, so it’s behind you and a bit to the right.", + "E": "The cabinet with snacks is straight in front of you, directly ahead of your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_5_T2", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the cabinet for snacks or the cabinet for seafood?", + "answer": "Right now, the snack cabinet is closer to you. The snack cabinet is in front of you.", + "question_chinese": "是放零食的柜子还是放海鲜的柜子离我更近？", + "answer_chinese": "现在放零食的柜子离你更近。放零食的柜子在你前面。", + "evidence": [ + { + "description": "There is a row of snack cabinets in front of me.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:40" + } + } + ], + "question_timestamps": "00:01:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, the seafood cabinet is closer to you. The seafood cabinet is in front of you.", + "B": "Right now, the snack cabinet is closer to you. The snack cabinet is in front of you." + }, + "answer_label": "B", + "answer_text": "Right now, the snack cabinet is closer to you. The snack cabinet is in front of you.", + "q_id": "egolife_A1_020_5_T2", + "correct_answer": "Right now, the snack cabinet is closer to you. The snack cabinet is in front of you.", + "original_options": { + "A": "Right now, the seafood cabinet is closer to you. The seafood cabinet is in front of you.", + "B": "Right now, the snack cabinet is closer to you. The snack cabinet is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_6_T2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the cabinet for snacks or the cabinet for seafood?", + "answer": "The cabinet with seafood is closer to you right now.", + "question_chinese": "是放零食的柜子还是放海鲜的柜子离我更近？", + "answer_chinese": "现在放海鲜的柜子离你更近。", + "evidence": [ + { + "description": "The seafood cabinet is behind and to my right.", + "timestamps": { + "start": "00:03:48", + "end": "00:03:53" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The cabinet with seafood is closer to you right now.", + "B": "The cabinet with snacks is closer to you right now." + }, + "answer_label": "A", + "answer_text": "The cabinet with seafood is closer to you right now.", + "q_id": "egolife_A1_020_6_T2", + "correct_answer": "The cabinet with seafood is closer to you right now.", + "original_options": { + "A": "The cabinet with seafood is closer to you right now.", + "B": "The cabinet with snacks is closer to you right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_5_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the cabinet for snacks or the cabinet for seafood?", + "answer": "Right now, the seafood cabinet is closer to you. The seafood cabinet is in front of you.", + "question_chinese": "是放零食的柜子还是放海鲜的柜子离我更近？", + "answer_chinese": "现在放海鲜的柜子离你更近。放海鲜的柜子在你面前。", + "evidence": [ + { + "description": "There is a row of snack cabinets in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "What's in front of me is the seafood cabinet.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:07:31", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, the seafood cabinet is closer to you. The seafood cabinet is in front of you.", + "B": "Right now, the snack cabinet is closer to you. The snack cabinet is in front of you." + }, + "answer_label": "A", + "answer_text": "Right now, the seafood cabinet is closer to you. The seafood cabinet is in front of you.", + "q_id": "egolife_A1_020_5_T1", + "correct_answer": "Right now, the seafood cabinet is closer to you. The seafood cabinet is in front of you.", + "original_options": { + "A": "Right now, the seafood cabinet is closer to you. The seafood cabinet is in front of you.", + "B": "Right now, the snack cabinet is closer to you. The snack cabinet is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_6_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the cabinet for snacks or the cabinet for seafood?", + "answer": "The cabinet with snacks is closer to you right now.", + "question_chinese": "是放零食的柜子还是放海鲜的柜子离我更近？", + "answer_chinese": "现在放零食的柜子离你更近。", + "evidence": [ + { + "description": "There is a row of snack cabinets in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "What's in front of me is the seafood cabinet.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:08:48", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The cabinet with snacks is closer to you right now.", + "B": "The cabinet with seafood is closer to you right now." + }, + "answer_label": "A", + "answer_text": "The cabinet with snacks is closer to you right now.", + "q_id": "egolife_A1_020_6_T1", + "correct_answer": "The cabinet with snacks is closer to you right now.", + "original_options": { + "A": "The cabinet with snacks is closer to you right now.", + "B": "The cabinet with seafood is closer to you right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_8_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the sashimi counter relative to me?", + "answer": "The raw-food counter is to your left.", + "question_chinese": "生食料理的柜台在我的什么方向？", + "answer_chinese": "生食料理的柜台在你的左边。", + "evidence": [ + { + "description": "The raw food counter is on my left.", + "timestamps": { + "start": "00:01:43", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The raw-food counter is to your right, not on your left side.", + "D": "The raw-food counter is positioned in front of you, rather than off to your left.", + "E": "The raw-food counter is to your left." + }, + "answer_label": "E", + "answer_text": "The raw-food counter is to your left.", + "q_id": "egolife_A1_020_8_T2", + "correct_answer": "The raw-food counter is to your left.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The raw-food counter is to your right, not on your left side.", + "D": "The raw-food counter is positioned in front of you, rather than off to your left.", + "E": "The raw-food counter is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_12_T2", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the sashimi counter relative to me?", + "answer": "The raw-food counter is to your left.", + "question_chinese": "生食料理的柜台在我的什么方向？", + "answer_chinese": "生食料理的柜台在你的左边。", + "evidence": [ + { + "description": "The raw food counter is on my left.", + "timestamps": { + "start": "00:02:07", + "end": "00:02:11" + } + } + ], + "question_timestamps": "00:02:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The raw-food counter is to your left.", + "B": "In front of you.", + "C": "Behind you.", + "D": "The raw-food counter is to your right, not to your left side at all.", + "E": "The raw-food counter is directly ahead of you rather than positioned on your left side." + }, + "answer_label": "A", + "answer_text": "The raw-food counter is to your left.", + "q_id": "egolife_A1_020_12_T2", + "correct_answer": "The raw-food counter is to your left.", + "original_options": { + "A": "The raw-food counter is to your left.", + "B": "In front of you.", + "C": "Behind you.", + "D": "The raw-food counter is to your right, not to your left side at all.", + "E": "The raw-food counter is directly ahead of you rather than positioned on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_12_T1", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the sashimi counter relative to me?", + "answer": "The raw-food counter is to your back-right.", + "question_chinese": "生食料理的柜台在我的什么方向？", + "answer_chinese": "生食料理的柜台在你的右后方。", + "evidence": [ + { + "description": "The raw food counter is right in front of me on my left.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:06:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-left.", + "B": "The raw-food counter is to your back-right.", + "C": "To your front-right.", + "D": "The raw-food counter is to your front-left, not behind you on the right.", + "E": "The raw-food counter sits off to your left side, rather than in the back-right direction." + }, + "answer_label": "B", + "answer_text": "The raw-food counter is to your back-right.", + "q_id": "egolife_A1_020_12_T1", + "correct_answer": "The raw-food counter is to your back-right.", + "original_options": { + "A": "To your back-left.", + "B": "The raw-food counter is to your back-right.", + "C": "To your front-right.", + "D": "The raw-food counter is to your front-left, not behind you on the right.", + "E": "The raw-food counter sits off to your left side, rather than in the back-right direction." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_020", + "video_length": "00:12:00", + "qid": "egolife_A1_020_8_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the sashimi counter relative to me?", + "answer": "The raw-food counter is to your back-left.", + "question_chinese": "生食料理的柜台在我的什么方向？", + "answer_chinese": "生食料理的柜台在你的左后方。", + "evidence": [ + { + "description": "The raw food counter is right in front of me on my left.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:10:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-right.", + "B": "The raw-food counter is to your back-left.", + "C": "To your front-left.", + "D": "The raw-food counter is to your front-right, not behind you on the left.", + "E": "The raw-food counter is positioned to your right side, rather than back-left from where you are." + }, + "answer_label": "B", + "answer_text": "The raw-food counter is to your back-left.", + "q_id": "egolife_A1_020_8_T1", + "correct_answer": "The raw-food counter is to your back-left.", + "original_options": { + "A": "To your back-right.", + "B": "The raw-food counter is to your back-left.", + "C": "To your front-left.", + "D": "The raw-food counter is to your front-right, not behind you on the left.", + "E": "The raw-food counter is positioned to your right side, rather than back-left from where you are." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 16 + ] + } +] diff --git a/QAs/egolife_A1_021-mcq.json b/QAs/egolife_A1_021-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a0eb21836d2766e64162765fc0f10d689a1b4df5 --- /dev/null +++ b/QAs/egolife_A1_021-mcq.json @@ -0,0 +1,682 @@ +[ + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021_1_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was holding the rooster, where was the rooster relative to the shelf?", + "answer": "You are holding the rooster. You are in front of the shelf, and the rooster is directly in front of the shelf.", + "question_chinese": "我拿着公鸡时，公鸡现在在相对于货架的什么位置？", + "answer_chinese": "你拿着公鸡，你在货架的前面，公鸡在货架的正前方", + "evidence": [ + { + "description": "I am directly in front of the shelf.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "I'm holding a rooster.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:24" + } + }, + { + "description": "The rooster is directly in front of the shelf.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The rooster is behind the shelf.", + "B": "The rooster is left of the shelf.", + "C": "You are holding the rooster. You are in front of the shelf, and the rooster is directly in front of the shelf.", + "D": "You are holding the rooster behind the shelf, with the rooster directly behind the shelf.", + "E": "You are in front of the shelf, but the rooster is directly below the shelf while you hold it." + }, + "answer_label": "C", + "answer_text": "You are holding the rooster. You are in front of the shelf, and the rooster is directly in front of the shelf.", + "q_id": "egolife_A1_021_1_T1", + "correct_answer": "You are holding the rooster. You are in front of the shelf, and the rooster is directly in front of the shelf.", + "original_options": { + "A": "The rooster is behind the shelf.", + "B": "The rooster is left of the shelf.", + "C": "You are holding the rooster. You are in front of the shelf, and the rooster is directly in front of the shelf.", + "D": "You are holding the rooster behind the shelf, with the rooster directly behind the shelf.", + "E": "You are in front of the shelf, but the rooster is directly below the shelf while you hold it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021_1_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was holding the rooster, where was the rooster relative to the shelf?", + "answer": "The rooster is directly above the shelf, and you are in front of the shelf.", + "question_chinese": "我拿着公鸡时，公鸡现在在相对于货架的什么位置？", + "answer_chinese": "公鸡在货架的正上面，你在货架的前面", + "evidence": [ + { + "description": "I put the rooster on the eggs.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:35" + } + }, + { + "description": "The eggs are right above the shelf.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + }, + { + "description": "Then the rooster is directly above the shelf.", + "timestamps": { + "start": "00:02:33", + "end": "00:02:35" + } + } + ], + "question_timestamps": "00:02:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The rooster is directly below the shelf.", + "B": "The rooster is in front of the shelf.", + "C": "The rooster is directly above the shelf, but you are standing behind the shelf while holding it.", + "D": "The rooster is directly above the shelf, and you are in front of the shelf.", + "E": "The rooster is directly above the shelf, and you are positioned to the left of the shelf." + }, + "answer_label": "D", + "answer_text": "The rooster is directly above the shelf, and you are in front of the shelf.", + "q_id": "egolife_A1_021_1_T2", + "correct_answer": "The rooster is directly above the shelf, and you are in front of the shelf.", + "original_options": { + "A": "The rooster is directly below the shelf.", + "B": "The rooster is in front of the shelf.", + "C": "The rooster is directly above the shelf, but you are standing behind the shelf while holding it.", + "D": "The rooster is directly above the shelf, and you are in front of the shelf.", + "E": "The rooster is directly above the shelf, and you are positioned to the left of the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021_2_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Was the refrigerator cabinet door next to me ever opened?", + "answer": "The refrigerator door next to you was never opened, and the refrigerator is to your right.", + "question_chinese": "我旁边的冰箱柜门有打开过吗？", + "answer_chinese": "你旁边的冰箱柜门没有被打开过，冰箱在你的右边", + "evidence": [ + { + "description": "The door of the fridge next to you has never been opened.", + "timestamps": { + "start": "00:00:53", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The refrigerator door next to you was never opened, and the refrigerator is to your right.", + "B": "The refrigerator door next to you was opened, and the refrigerator is to your right." + }, + "answer_label": "A", + "answer_text": "The refrigerator door next to you was never opened, and the refrigerator is to your right.", + "q_id": "egolife_A1_021_2_T2", + "correct_answer": "The refrigerator door next to you was never opened, and the refrigerator is to your right.", + "original_options": { + "A": "The refrigerator door next to you was never opened, and the refrigerator is to your right.", + "B": "The refrigerator door next to you was opened, and the refrigerator is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021_2_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Was the refrigerator cabinet door next to me ever opened?", + "answer": "The refrigerator door next to you was opened, and the refrigerator is in front of you.", + "question_chinese": "我旁边的冰箱柜门有打开过吗？", + "answer_chinese": "你旁边的冰箱柜门有打开过，冰箱在你前面。", + "evidence": [ + { + "description": "The fridge door next to you has been opened.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:02:25", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The refrigerator door next to you was never opened, and the refrigerator is in front of you.", + "B": "The refrigerator door next to you was opened, and the refrigerator is in front of you." + }, + "answer_label": "B", + "answer_text": "The refrigerator door next to you was opened, and the refrigerator is in front of you.", + "q_id": "egolife_A1_021_2_T1", + "correct_answer": "The refrigerator door next to you was opened, and the refrigerator is in front of you.", + "original_options": { + "A": "The refrigerator door next to you was never opened, and the refrigerator is in front of you.", + "B": "The refrigerator door next to you was opened, and the refrigerator is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021_3_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Am I at the checkout counter right now?", + "answer": "You are not at the checkout counter right now; the checkout counter is to your front-left.", + "question_chinese": "我现在在收银台吗？", + "answer_chinese": "你现在不在收银台，收银台在我左前方。", + "evidence": [ + { + "description": "You are not at the checkout counter now. You are in the vegetable section.", + "timestamps": { + "start": "00:11:20", + "end": "00:11:23" + } + } + ], + "question_timestamps": "00:11:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Yes, you're at the checkout counter.", + "B": "No; it's to your front-right.", + "C": "You are not at the checkout counter right now; the checkout counter is to your back-left.", + "D": "You are not at the checkout counter right now; the checkout counter is to your front-left.", + "E": "You are not at the checkout counter right now; the checkout counter is directly behind you." + }, + "answer_label": "D", + "answer_text": "You are not at the checkout counter right now; the checkout counter is to your front-left.", + "q_id": "egolife_A1_021_3_T1", + "correct_answer": "You are not at the checkout counter right now; the checkout counter is to your front-left.", + "original_options": { + "A": "Yes, you're at the checkout counter.", + "B": "No; it's to your front-right.", + "C": "You are not at the checkout counter right now; the checkout counter is to your back-left.", + "D": "You are not at the checkout counter right now; the checkout counter is to your front-left.", + "E": "You are not at the checkout counter right now; the checkout counter is directly behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021_3_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Am I at the checkout counter right now?", + "answer": "You are at the checkout counter right now, and the checkout counter is in front of you.", + "question_chinese": "我现在在收银台吗？", + "answer_chinese": "你现在在收银台，收银台在你前面。", + "evidence": [ + { + "description": "You are at the checkout counter now.", + "timestamps": { + "start": "00:11:35", + "end": "00:11:45" + } + } + ], + "question_timestamps": "00:11:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You are at the checkout counter right now, and the checkout counter is in front of you.", + "B": "No, you're not at the counter.", + "C": "Yes; it’s to your left.", + "D": "You are at the checkout counter right now, but the checkout counter is behind you.", + "E": "You are at the checkout counter right now, and the checkout counter is positioned to your right." + }, + "answer_label": "A", + "answer_text": "You are at the checkout counter right now, and the checkout counter is in front of you.", + "q_id": "egolife_A1_021_3_T2", + "correct_answer": "You are at the checkout counter right now, and the checkout counter is in front of you.", + "original_options": { + "A": "You are at the checkout counter right now, and the checkout counter is in front of you.", + "B": "No, you're not at the counter.", + "C": "Yes; it’s to your left.", + "D": "You are at the checkout counter right now, but the checkout counter is behind you.", + "E": "You are at the checkout counter right now, and the checkout counter is positioned to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there a change in the quantity of drinks in the refrigerator now?", + "answer": "No.", + "question_chinese": "现在冰箱柜里饮料数量有变化吗", + "answer_chinese": "有", + "evidence": [ + { + "description": "A staff member took out a bottle of milk from the refrigerator.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:33" + } + }, + { + "description": "The staff put in a bottle of milk.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:33" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_021_5", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there a change in the quantity of drinks in the refrigerator now?", + "answer": "No.", + "question_chinese": "现在冰箱柜里饮料数量有变化吗", + "answer_chinese": "没有", + "evidence": [ + { + "description": "A staff member took out a bottle of milk from the refrigerator.", + "timestamps": { + "start": "00:02:30", + "end": "00:03:15" + } + }, + { + "description": "Another staff member put in another bottle of milk.", + "timestamps": { + "start": "00:02:30", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_021_6", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021-mcq_fixed_0", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the distance between my phone and me now?", + "answer": "The length of an arm.", + "question_chinese": "现在手机距离我的距离是多少", + "answer_chinese": "一手臂长", + "evidence": [ + { + "description": "I put my phone on the shelf.", + "timestamps": { + "start": "00:06:10", + "end": "00:07:00" + } + }, + { + "description": "I use a can to fix my phone.", + "timestamps": { + "start": "00:06:10", + "end": "00:07:00" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The length of an arm.", + "B": "A hand’s width away.", + "C": "About a chair’s height away.", + "D": "Roughly the width of a desk between you and the phone, like it’s resting just across the surface.", + "E": "About the span from your shoulder to your fingertips, as if the phone is set farther than a reach." + }, + "answer_label": "A", + "answer_text": "The length of an arm.", + "q_id": "egolife_A1_021-mcq_fixed_0", + "correct_answer": "The length of an arm.", + "original_options": { + "A": "The length of an arm.", + "B": "A hand’s width away.", + "C": "About a chair’s height away.", + "D": "Roughly the width of a desk between you and the phone, like it’s resting just across the surface.", + "E": "About the span from your shoulder to your fingertips, as if the phone is set farther than a reach." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021-mcq_fixed_1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the distance between my phone and me now?", + "answer": "About five centimeters from the chest.", + "question_chinese": "现在手机距离我的距离是多少", + "answer_chinese": "胸前五厘米左右", + "evidence": [ + { + "description": "I got my phone back", + "timestamps": { + "start": "00:07:07", + "end": "00:07:14" + } + }, + { + "description": "I held the phone in front of my chest to take a photo.", + "timestamps": { + "start": "00:07:07", + "end": "00:07:14" + } + } + ], + "question_timestamps": "00:07:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "About five centimeters from the chest.", + "B": "About ten centimeters from your chest.", + "C": "About three centimeters from your chest.", + "D": "Roughly twenty centimeters away from your face, as if you’re holding the phone out while looking down.", + "E": "About fifteen centimeters from your right hand, suggesting it’s not pressed close to your chest." + }, + "answer_label": "A", + "answer_text": "About five centimeters from the chest.", + "q_id": "egolife_A1_021-mcq_fixed_1", + "correct_answer": "About five centimeters from the chest.", + "original_options": { + "A": "About five centimeters from the chest.", + "B": "About ten centimeters from your chest.", + "C": "About three centimeters from your chest.", + "D": "Roughly twenty centimeters away from your face, as if you’re holding the phone out while looking down.", + "E": "About fifteen centimeters from your right hand, suggesting it’s not pressed close to your chest." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021_10", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the corn kernel in relation to me right now?", + "answer": "In my hand.", + "question_chinese": "现在玉米粒在相对于我的什么位置", + "answer_chinese": "在我的手上", + "evidence": [ + { + "description": "I picked up the corn kernels from the shopping cart at 00:01:00.", + "timestamps": { + "start": "00:02:00", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In my hand.", + "B": "In your pocket.", + "C": "In your bag.", + "D": "In your jacket pocket, kept there instead of being held in your hand right now.", + "E": "In your backpack, tucked away in a pocket you’d need to open to reach it." + }, + "answer_label": "A", + "answer_text": "In my hand.", + "q_id": "egolife_A1_021_10", + "correct_answer": "In my hand.", + "original_options": { + "A": "In my hand.", + "B": "In your pocket.", + "C": "In your bag.", + "D": "In your jacket pocket, kept there instead of being held in your hand right now.", + "E": "In your backpack, tucked away in a pocket you’d need to open to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_021", + "video_length": "00:12:00", + "qid": "egolife_A1_021_9", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the corn kernel in relation to me right now?", + "answer": "In the shopping cart directly in front of me.", + "question_chinese": "现在玉米粒在相对于我的什么位置", + "answer_chinese": "在我的正前方的购物车里面", + "evidence": [ + { + "description": "I will put the corn kernels back into the shopping cart at 00:07:00", + "timestamps": { + "start": "00:05:00", + "end": "00:07:00" + } + } + ], + "question_timestamps": "00:07:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the cart slightly to your right.", + "B": "In the cart slightly behind you.", + "C": "In the grocery cart ahead of you, but positioned off to the right rather than directly in front.", + "D": "In the shopping cart directly in front of me.", + "E": "In the shopping cart in front of you, but pushed a bit forward so it’s not directly at your position." + }, + "answer_label": "D", + "answer_text": "In the shopping cart directly in front of me.", + "q_id": "egolife_A1_021_9", + "correct_answer": "In the shopping cart directly in front of me.", + "original_options": { + "A": "In the cart slightly to your right.", + "B": "In the cart slightly behind you.", + "C": "In the grocery cart ahead of you, but positioned off to the right rather than directly in front.", + "D": "In the shopping cart directly in front of me.", + "E": "In the shopping cart in front of you, but pushed a bit forward so it’s not directly at your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 21 + ] + } +] diff --git a/QAs/egolife_A1_022-mcq.json b/QAs/egolife_A1_022-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a56e68e33b0e683c3e50be58e43d32edacd36768 --- /dev/null +++ b/QAs/egolife_A1_022-mcq.json @@ -0,0 +1,530 @@ +[ + { + "video_uid": "egolife_A1_022", + "video_length": "00:12:00", + "qid": "egolife_A1_022_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make from the checkout to the shopping cart area?", + "answer": "One.", + "question_chinese": "我想从收银台到购物车集中点要拐几个弯", + "answer_chinese": "一个", + "evidence": [ + { + "description": "After I turned left, I walked to the vicinity of the shopping cart.", + "timestamps": { + "start": "00:09:40", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:10:33", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two turns.", + "B": "Three turns.", + "C": "One.", + "D": "You need to make two turns from the checkout area to reach where the shopping carts are stored.", + "E": "You have to turn three times on the way, with each change of direction leading you closer to the cart area." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "egolife_A1_022_7", + "correct_answer": "One.", + "original_options": { + "A": "Two turns.", + "B": "Three turns.", + "C": "One.", + "D": "You need to make two turns from the checkout area to reach where the shopping carts are stored.", + "E": "You have to turn three times on the way, with each change of direction leading you closer to the cart area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 21 + ] + }, + { + "video_uid": "egolife_A1_022", + "video_length": "00:12:00", + "qid": "egolife_A1_022_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make from the checkout to the shopping cart area?", + "answer": "Twice.", + "question_chinese": "我想从收银台到购物车集中点要拐几个弯", + "answer_chinese": "一个", + "evidence": [ + { + "description": "I turned a corner from the checkout counter to the shopping cart collection point.", + "timestamps": { + "start": "00:09:40", + "end": "00:10:40" + } + }, + { + "description": "I turn a corner from the checkout counter to the service point.", + "timestamps": { + "start": "00:09:40", + "end": "00:10:40" + } + } + ], + "question_timestamps": "00:11:03", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "You only need to turn once from the checkout before arriving at the shopping cart area.", + "D": "You make three turns in total from checkout to the cart area, changing direction at each junction.", + "E": "Twice." + }, + "answer_label": "E", + "answer_text": "Twice.", + "q_id": "egolife_A1_022_8", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "You only need to turn once from the checkout before arriving at the shopping cart area.", + "D": "You make three turns in total from checkout to the cart area, changing direction at each junction.", + "E": "Twice." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_022", + "video_length": "00:12:00", + "qid": "egolife_A1_022_1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the power bank located in relation to the bag?", + "answer": "Right behind my bag.", + "question_chinese": "现在充电宝在相对于包的什么位置", + "answer_chinese": "在我的包的正后方", + "evidence": [ + { + "description": "I put the power bank in the shopping cart.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:40" + } + }, + { + "description": "The shopping cart is directly in front of the bag.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:40" + } + }, + { + "description": "I turned around 180 degrees.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:40" + } + }, + { + "description": "The bag is hanging on me.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Directly in front of my bag.", + "B": "To the left of my bag.", + "C": "It’s positioned slightly in front of my bag, not behind it, as if set down closer to me.", + "D": "The power bank sits to the left side of my bag, offset rather than aligned directly behind it.", + "E": "Right behind my bag." + }, + "answer_label": "E", + "answer_text": "Right behind my bag.", + "q_id": "egolife_A1_022_1", + "correct_answer": "Right behind my bag.", + "original_options": { + "A": "Directly in front of my bag.", + "B": "To the left of my bag.", + "C": "It’s positioned slightly in front of my bag, not behind it, as if set down closer to me.", + "D": "The power bank sits to the left side of my bag, offset rather than aligned directly behind it.", + "E": "Right behind my bag." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_022", + "video_length": "00:12:00", + "qid": "egolife_A1_022_2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the power bank located in relation to the bag?", + "answer": "Inside my bag.", + "question_chinese": "现在充电宝在相对于包的什么位置", + "answer_chinese": "在我的包里面", + "evidence": [ + { + "description": "I took the power bank out of the bag.", + "timestamps": { + "start": "00:01:40", + "end": "00:03:20" + } + }, + { + "description": "I put the power bank in the shopping cart.", + "timestamps": { + "start": "00:01:40", + "end": "00:03:20" + } + }, + { + "description": "I put the power bank back into my bag from the shopping cart.", + "timestamps": { + "start": "00:01:40", + "end": "00:03:20" + } + } + ], + "question_timestamps": "00:10:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "In front of my bag.", + "B": "Next to my bag.", + "C": "It’s placed in front of my bag on the surface, not tucked inside any compartment.", + "D": "Inside my bag.", + "E": "The power bank is beside my bag, close enough to touch without opening the bag at all." + }, + "answer_label": "D", + "answer_text": "Inside my bag.", + "q_id": "egolife_A1_022_2", + "correct_answer": "Inside my bag.", + "original_options": { + "A": "In front of my bag.", + "B": "Next to my bag.", + "C": "It’s placed in front of my bag on the surface, not tucked inside any compartment.", + "D": "Inside my bag.", + "E": "The power bank is beside my bag, close enough to touch without opening the bag at all." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_022", + "video_length": "00:12:00", + "qid": "egolife_A1_022_5", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the beef roll located in relation to the cash register?", + "answer": "Above the cash register.", + "question_chinese": "现在肥牛卷在相对于收银台哪里", + "answer_chinese": "在收银台的上方", + "evidence": [ + { + "description": "I placed the QR code of the fat beef rolls directly opposite the scanner.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:18" + } + }, + { + "description": "The scanner is directly above the cash register.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:02:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Above the cash register.", + "B": "Below the cash register.", + "C": "In front of the cash register.", + "D": "It is positioned behind the cash register, not hanging above it where you might expect.", + "E": "You see it to the right of the cash register, offset beside it rather than directly over it." + }, + "answer_label": "A", + "answer_text": "Above the cash register.", + "q_id": "egolife_A1_022_5", + "correct_answer": "Above the cash register.", + "original_options": { + "A": "Above the cash register.", + "B": "Below the cash register.", + "C": "In front of the cash register.", + "D": "It is positioned behind the cash register, not hanging above it where you might expect.", + "E": "You see it to the right of the cash register, offset beside it rather than directly over it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_022", + "video_length": "00:12:00", + "qid": "egolife_A1_022_6", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the beef roll located in relation to the cash register?", + "answer": "To the left-front diagonally.", + "question_chinese": "现在肥牛卷在相对于收银台哪里", + "answer_chinese": "在左斜前方", + "evidence": [ + { + "description": "I put the fat beef rolls into the plastic bag of the shopping cart.", + "timestamps": { + "start": "00:02:00", + "end": "00:06:00" + } + }, + { + "description": "I pushed the shopping cart to the left.", + "timestamps": { + "start": "00:02:00", + "end": "00:06:00" + } + } + ], + "question_timestamps": "00:08:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To the right-back diagonally.", + "B": "To the left-front diagonally.", + "C": "To your left-back diagonally.", + "D": "It is directly in front of the cash register, not offset diagonally toward the left side.", + "E": "You find it behind the cash register now, rather than sitting at the left-front diagonal position." + }, + "answer_label": "B", + "answer_text": "To the left-front diagonally.", + "q_id": "egolife_A1_022_6", + "correct_answer": "To the left-front diagonally.", + "original_options": { + "A": "To the right-back diagonally.", + "B": "To the left-front diagonally.", + "C": "To your left-back diagonally.", + "D": "It is directly in front of the cash register, not offset diagonally toward the left side.", + "E": "You find it behind the cash register now, rather than sitting at the left-front diagonal position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_022", + "video_length": "00:12:00", + "qid": "egolife_A1_022_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks are in the shopping cart now?", + "answer": "Seven bottles.", + "question_chinese": "现在这个购物车里有几瓶饮料", + "answer_chinese": "7瓶", + "evidence": [ + { + "description": "There are seven bottles of drinks in the shopping cart.", + "timestamps": { + "start": "00:04:35", + "end": "00:05:05" + } + } + ], + "question_timestamps": "00:05:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Six bottles.", + "B": "Eight bottles.", + "C": "There are five bottles of drinks in the cart right now, not as many as seven.", + "D": "You can count nine drink bottles in the cart at this moment, more than the correct total.", + "E": "Seven bottles." + }, + "answer_label": "E", + "answer_text": "Seven bottles.", + "q_id": "egolife_A1_022_3", + "correct_answer": "Seven bottles.", + "original_options": { + "A": "Six bottles.", + "B": "Eight bottles.", + "C": "There are five bottles of drinks in the cart right now, not as many as seven.", + "D": "You can count nine drink bottles in the cart at this moment, more than the correct total.", + "E": "Seven bottles." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_022", + "video_length": "00:12:00", + "qid": "egolife_A1_022_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks are in the shopping cart now?", + "answer": "Zero bottles.", + "question_chinese": "现在这个购物车里有几瓶饮料", + "answer_chinese": "0瓶", + "evidence": [ + { + "description": "Put the drinks in this shopping cart into another shopping cart.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:40" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero bottles.", + "B": "One bottle.", + "C": "Two bottles.", + "D": "There are three bottles of drinks in the shopping cart now instead of none.", + "E": "You have four bottles of drinks in the cart at the moment, not an empty cart." + }, + "answer_label": "A", + "answer_text": "Zero bottles.", + "q_id": "egolife_A1_022_4", + "correct_answer": "Zero bottles.", + "original_options": { + "A": "Zero bottles.", + "B": "One bottle.", + "C": "Two bottles.", + "D": "There are three bottles of drinks in the shopping cart now instead of none.", + "E": "You have four bottles of drinks in the cart at the moment, not an empty cart." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + } +] diff --git a/QAs/egolife_A1_023-mcq.json b/QAs/egolife_A1_023-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f330ae29dc570c34afba84e66ad8a84754042401 --- /dev/null +++ b/QAs/egolife_A1_023-mcq.json @@ -0,0 +1,1238 @@ +[ + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_10_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how far away is the sunflower from me right now?", + "answer": "10 centimeters.", + "question_chinese": "现在向日葵离我大概有多远？", + "answer_chinese": "10厘米", + "evidence": [ + { + "description": "I am standing 5 centimeters directly in front of the merchandise.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:10" + } + }, + { + "description": "Sunflowers are next to the products for sale.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:04:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "About 1 centimeter.", + "B": "About 30 centimeters.", + "C": "10 centimeters.", + "D": "It is roughly one meter away from you right now, farther than a hand’s width.", + "E": "You are about three meters away from the sunflower at the moment, not within a few centimeters." + }, + "answer_label": "C", + "answer_text": "10 centimeters.", + "q_id": "egolife_A1_023_10_T1", + "correct_answer": "10 centimeters.", + "original_options": { + "A": "About 1 centimeter.", + "B": "About 30 centimeters.", + "C": "10 centimeters.", + "D": "It is roughly one meter away from you right now, farther than a hand’s width.", + "E": "You are about three meters away from the sunflower at the moment, not within a few centimeters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_10_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how far away is the sunflower from me right now?", + "answer": "There was no distance.", + "question_chinese": "现在向日葵离我大概有多远？", + "answer_chinese": "没有距离", + "evidence": [ + { + "description": "Sunflowers are next to the products for sale.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:04:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "10 centimeters.", + "B": "About 1 meter.", + "C": "There was no distance.", + "D": "It is about three meters away to your left right now, so there is definitely distance.", + "E": "You are roughly 100 centimeters from the sunflower at the moment, not at zero separation." + }, + "answer_label": "C", + "answer_text": "There was no distance.", + "q_id": "egolife_A1_023_10_T2", + "correct_answer": "There was no distance.", + "original_options": { + "A": "10 centimeters.", + "B": "About 1 meter.", + "C": "There was no distance.", + "D": "It is about three meters away to your left right now, so there is definitely distance.", + "E": "You are roughly 100 centimeters from the sunflower at the moment, not at zero separation." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_1_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how far away is the sunflower from me right now?", + "answer": "10 centimeters.", + "question_chinese": "现在向日葵离我大概有多远？", + "answer_chinese": "10厘米", + "evidence": [ + { + "description": "I am standing 5 centimeters directly in front of the merchandise.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:10" + } + }, + { + "description": "Sunflowers are next to the products for sale.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:04:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "8 centimeters.", + "B": "10 centimeters.", + "C": "12 centimeters.", + "D": "It is about 9 centimeters away from you right now, just slightly closer than ten.", + "E": "You are approximately 11 centimeters from the sunflower at the moment, a bit farther than ten." + }, + "answer_label": "B", + "answer_text": "10 centimeters.", + "q_id": "egolife_A1_023_1_T1", + "correct_answer": "10 centimeters.", + "original_options": { + "A": "8 centimeters.", + "B": "10 centimeters.", + "C": "12 centimeters.", + "D": "It is about 9 centimeters away from you right now, just slightly closer than ten.", + "E": "You are approximately 11 centimeters from the sunflower at the moment, a bit farther than ten." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_1_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "About how far away is the sunflower from me right now?", + "answer": "100 meters.", + "question_chinese": "现在向日葵离我大概有多远？", + "answer_chinese": "100米", + "evidence": [ + { + "description": "I am standing 5 centimeters directly in front of the merchandise.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:07:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "99 meters.", + "B": "101 meters.", + "C": "100 meters.", + "D": "The sunflower is about 98 meters away from you right now, slightly less than one hundred.", + "E": "You are roughly 102 meters from the sunflower at the moment, a little farther than one hundred." + }, + "answer_label": "C", + "answer_text": "100 meters.", + "q_id": "egolife_A1_023_1_T2", + "correct_answer": "100 meters.", + "original_options": { + "A": "99 meters.", + "B": "101 meters.", + "C": "100 meters.", + "D": "The sunflower is about 98 meters away from you right now, slightly less than one hundred.", + "E": "You are roughly 102 meters from the sunflower at the moment, a little farther than one hundred." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_7", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the green onion located in relation to the kitchen wipes?", + "answer": "Directly above the kitchen wet wipes.", + "question_chinese": "现在大葱相对于厨房湿巾的位置", + "answer_chinese": "在厨房湿巾的正上方", + "evidence": [ + { + "description": "Put the kitchen wet wipes in the shopping cart first.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:40" + } + }, + { + "description": "Put the green onions into the shopping cart afterward.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:40" + } + }, + { + "description": "Put green onions and kitchen wet wipes in the same shopping cart.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:40" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Directly below the kitchen wet wipes.", + "B": "Slightly left of the kitchen wet wipes.", + "C": "Directly above the kitchen wet wipes.", + "D": "It is positioned a little to the right of the kitchen wet wipes, not stacked above them.", + "E": "It sits below and slightly behind the kitchen wet wipes, closer to the edge of the surface." + }, + "answer_label": "C", + "answer_text": "Directly above the kitchen wet wipes.", + "q_id": "egolife_A1_023_7", + "correct_answer": "Directly above the kitchen wet wipes.", + "original_options": { + "A": "Directly below the kitchen wet wipes.", + "B": "Slightly left of the kitchen wet wipes.", + "C": "Directly above the kitchen wet wipes.", + "D": "It is positioned a little to the right of the kitchen wet wipes, not stacked above them.", + "E": "It sits below and slightly behind the kitchen wet wipes, closer to the edge of the surface." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_8", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the green onion located in relation to the kitchen wipes?", + "answer": "Left side.", + "question_chinese": "现在大葱相对于厨房湿巾的位置", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The green Chinese onion is in the left hand.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:30" + } + }, + { + "description": "The kitchen wet wipes are in my right hand.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:30" + } + }, + { + "description": "Both hands are on the same horizontal line.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:30" + } + } + ], + "question_timestamps": "00:10:43", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "Left side.", + "C": "In front of you.", + "D": "It is positioned behind the kitchen wipes rather than on the left side.", + "E": "You see it on the right side of the wipes, slightly offset instead of aligned with them." + }, + "answer_label": "B", + "answer_text": "Left side.", + "q_id": "egolife_A1_023_8", + "correct_answer": "Left side.", + "original_options": { + "A": "Right side.", + "B": "Left side.", + "C": "In front of you.", + "D": "It is positioned behind the kitchen wipes rather than on the left side.", + "E": "You see it on the right side of the wipes, slightly offset instead of aligned with them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_6_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the scallion relative to my kitchen wet wipes?", + "answer": "Directly above the kitchen wet wipes.", + "question_chinese": "现在大葱相对于我厨房湿巾的位置？", + "answer_chinese": "在厨房湿巾的正上方", + "evidence": [ + { + "description": "Put the kitchen wet wipes in the shopping cart first.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:40" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly below the kitchen wet wipes.", + "B": "On the left side.", + "C": "It is to the right of the kitchen wet wipes, not placed above them.", + "D": "You have it behind the kitchen wet wipes, with the wipes blocking it from your direct view.", + "E": "Directly above the kitchen wet wipes." + }, + "answer_label": "E", + "answer_text": "Directly above the kitchen wet wipes.", + "q_id": "egolife_A1_023_6_T2", + "correct_answer": "Directly above the kitchen wet wipes.", + "original_options": { + "A": "Directly below the kitchen wet wipes.", + "B": "On the left side.", + "C": "It is to the right of the kitchen wet wipes, not placed above them.", + "D": "You have it behind the kitchen wet wipes, with the wipes blocking it from your direct view.", + "E": "Directly above the kitchen wet wipes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_6_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the scallion relative to my kitchen wet wipes?", + "answer": "Directly above the kitchen wet wipes.", + "question_chinese": "现在大葱相对于我厨房湿巾的位置？", + "answer_chinese": "在厨房湿巾的正上方", + "evidence": [ + { + "description": "Put the kitchen wet wipes in the shopping cart first.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:40" + } + }, + { + "description": "Put the green onions into the shopping cart afterward.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:40" + } + }, + { + "description": "Put green onions and kitchen wet wipes in the same shopping cart.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:40" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly below the kitchen wet wipes.", + "B": "Directly above the kitchen wet wipes.", + "C": "On the left side.", + "D": "It sits to the right of the kitchen wet wipes rather than directly above them.", + "E": "It is positioned behind the kitchen wet wipes, not stacked vertically over the top of them." + }, + "answer_label": "B", + "answer_text": "Directly above the kitchen wet wipes.", + "q_id": "egolife_A1_023_6_T1", + "correct_answer": "Directly above the kitchen wet wipes.", + "original_options": { + "A": "Directly below the kitchen wet wipes.", + "B": "Directly above the kitchen wet wipes.", + "C": "On the left side.", + "D": "It sits to the right of the kitchen wet wipes rather than directly above them.", + "E": "It is positioned behind the kitchen wet wipes, not stacked vertically over the top of them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_7_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the scallion relative to my kitchen wet wipes?", + "answer": "On the left side.", + "question_chinese": "现在大葱相对于我厨房湿巾的位置？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The green Chinese onion is in the left hand.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:30" + } + }, + { + "description": "The kitchen wet wipes are in my right hand.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:30" + } + }, + { + "description": "Both hands are on the same horizontal line.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:30" + } + } + ], + "question_timestamps": "00:10:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your right.", + "B": "On the left side.", + "C": "In front of you.", + "D": "It is behind you, not located on the left side of the kitchen wet wipes.", + "E": "You would find it on the right side of the kitchen wet wipes instead of on the left." + }, + "answer_label": "B", + "answer_text": "On the left side.", + "q_id": "egolife_A1_023_7_T1", + "correct_answer": "On the left side.", + "original_options": { + "A": "To your right.", + "B": "On the left side.", + "C": "In front of you.", + "D": "It is behind you, not located on the left side of the kitchen wet wipes.", + "E": "You would find it on the right side of the kitchen wet wipes instead of on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_7_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the scallion relative to my kitchen wet wipes?", + "answer": "On the left side.", + "question_chinese": "现在大葱相对于我厨房湿巾的位置？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The green Chinese onion is in the left hand.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:30" + } + } + ], + "question_timestamps": "00:10:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the left side.", + "B": "To your right.", + "C": "In front of you.", + "D": "It is behind you, rather than sitting on the left side of the kitchen wet wipes.", + "E": "It appears on the right side of the kitchen wet wipes instead of being positioned on the left." + }, + "answer_label": "A", + "answer_text": "On the left side.", + "q_id": "egolife_A1_023_7_T2", + "correct_answer": "On the left side.", + "original_options": { + "A": "On the left side.", + "B": "To your right.", + "C": "In front of you.", + "D": "It is behind you, rather than sitting on the left side of the kitchen wet wipes.", + "E": "It appears on the right side of the kitchen wet wipes instead of being positioned on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_8_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where is the child relative to me?", + "answer": "To your front-left.", + "question_chinese": "现在小孩相对于我的位置？", + "answer_chinese": "在我的左前方", + "evidence": [ + { + "description": "The child is directly in front of the first checkout counter.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:40" + } + }, + { + "description": "I am facing the third checkout counter.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:02:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "The child is directly in front of you rather than positioned toward your front-left.", + "D": "To your front-left.", + "E": "You see the child behind and to your right, not out toward the front-left corner." + }, + "answer_label": "D", + "answer_text": "To your front-left.", + "q_id": "egolife_A1_023_8_T1", + "correct_answer": "To your front-left.", + "original_options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "The child is directly in front of you rather than positioned toward your front-left.", + "D": "To your front-left.", + "E": "You see the child behind and to your right, not out toward the front-left corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_8_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where is the child relative to me?", + "answer": "To your front-left.", + "question_chinese": "现在小孩相对于我的位置？", + "answer_chinese": "在我的左前方", + "evidence": [ + { + "description": "The child is directly in front of the first checkout counter.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:02:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "To your front-left.", + "D": "The child is directly in front of you, not offset toward your front-left side.", + "E": "The child is behind you on the right, rather than standing out in front-left of you." + }, + "answer_label": "C", + "answer_text": "To your front-left.", + "q_id": "egolife_A1_023_8_T2", + "correct_answer": "To your front-left.", + "original_options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "To your front-left.", + "D": "The child is directly in front of you, not offset toward your front-left side.", + "E": "The child is behind you on the right, rather than standing out in front-left of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_9_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where is the child relative to me?", + "answer": "To your front-right.", + "question_chinese": "现在小孩相对我的位置？", + "answer_chinese": "右前方", + "evidence": [ + { + "description": "I stand at the checkout counter.", + "timestamps": { + "start": "00:02:50", + "end": "00:03:05" + } + }, + { + "description": "The child is on the right side of the checkout counter area.", + "timestamps": { + "start": "00:02:50", + "end": "00:03:05" + } + } + ], + "question_timestamps": "00:03:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-left.", + "B": "To your front-right.", + "C": "To your back-right.", + "D": "To your back-left, a little behind your shoulder as you look ahead.", + "E": "Directly in front of you, lined up with your gaze rather than off to the right." + }, + "answer_label": "B", + "answer_text": "To your front-right.", + "q_id": "egolife_A1_023_9_T1", + "correct_answer": "To your front-right.", + "original_options": { + "A": "To your front-left.", + "B": "To your front-right.", + "C": "To your back-right.", + "D": "To your back-left, a little behind your shoulder as you look ahead.", + "E": "Directly in front of you, lined up with your gaze rather than off to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 16 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_9_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where is the child relative to me?", + "answer": "To your front-right.", + "question_chinese": "现在小孩相对我的位置？", + "answer_chinese": "右前方", + "evidence": [ + { + "description": "The child is on the right side of the checkout counter area.", + "timestamps": { + "start": "00:02:50", + "end": "00:03:05" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-right.", + "B": "To your front-left.", + "C": "To your back-right.", + "D": "To your back-left, a little behind your shoulder as you look ahead.", + "E": "Directly in front of you, lined up with your gaze rather than off to the right." + }, + "answer_label": "A", + "answer_text": "To your front-right.", + "q_id": "egolife_A1_023_9_T2", + "correct_answer": "To your front-right.", + "original_options": { + "A": "To your front-right.", + "B": "To your front-left.", + "C": "To your back-right.", + "D": "To your back-left, a little behind your shoulder as you look ahead.", + "E": "Directly in front of you, lined up with your gaze rather than off to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 16 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_4_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, how many turns do I need to make to get from the checkout counter to the elevator?", + "answer": "You need three.", + "question_chinese": "现在我从收银台到电梯要拐几个弯？", + "answer_chinese": "你需要三个", + "evidence": [ + { + "description": "I went straight from the cash register, then turned left, then right, and finally right again to reach the elevator.", + "timestamps": { + "start": "00:07:00", + "end": "00:10:24" + } + } + ], + "question_timestamps": "00:10:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You need one.", + "B": "You need two.", + "C": "You need four, making an extra turn after leaving the checkout area before reaching the elevator.", + "D": "You need five, with two quick direction changes in the corridor before the final turn to the elevator.", + "E": "You need three." + }, + "answer_label": "E", + "answer_text": "You need three.", + "q_id": "egolife_A1_023_4_T1", + "correct_answer": "You need three.", + "original_options": { + "A": "You need one.", + "B": "You need two.", + "C": "You need four, making an extra turn after leaving the checkout area before reaching the elevator.", + "D": "You need five, with two quick direction changes in the corridor before the final turn to the elevator.", + "E": "You need three." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_4_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, how many turns do I need to make to get from the checkout counter to the elevator?", + "answer": "You need four.", + "question_chinese": "现在我从收银台到电梯要拐几个弯？", + "answer_chinese": "你需要四个", + "evidence": [ + { + "description": "I went straight from the cash register, then turned left, then right, and finally right again to reach the elevator.", + "timestamps": { + "start": "00:07:00", + "end": "00:10:24" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You need four.", + "B": "You need two.", + "C": "You need three.", + "D": "You need five, adding one more turn after passing the checkout counter to line up with the elevator.", + "E": "You need six, with several direction changes through the aisles before the final approach to the elevator." + }, + "answer_label": "A", + "answer_text": "You need four.", + "q_id": "egolife_A1_023_4_T2", + "correct_answer": "You need four.", + "original_options": { + "A": "You need four.", + "B": "You need two.", + "C": "You need three.", + "D": "You need five, adding one more turn after passing the checkout counter to line up with the elevator.", + "E": "You need six, with several direction changes through the aisles before the final approach to the elevator." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_5_T1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, how many turns do I need to make to get from the checkout counter to the elevator?", + "answer": "Three.", + "question_chinese": "现在我从收银台到电梯要拐几个弯？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "I went straight from the cash register, then turned left, then right, and finally right again to reach the elevator.", + "timestamps": { + "start": "00:11:00", + "end": "00:11:40" + } + } + ], + "question_timestamps": "00:12:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Three.", + "B": "One.", + "C": "Two.", + "D": "Four, with an additional turn after leaving the checkout area to face the elevator corridor.", + "E": "Five, requiring a couple extra direction changes through the path before reaching the elevator." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "egolife_A1_023_5_T1", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "One.", + "C": "Two.", + "D": "Four, with an additional turn after leaving the checkout area to face the elevator corridor.", + "E": "Five, requiring a couple extra direction changes through the path before reaching the elevator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_3", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many shopping carts do I have now?", + "answer": "Three.", + "question_chinese": "现在我拿了几个购物车", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There are three shopping carts in front of me with the items I bought.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "One cart, since you only have a single cart with you right now.", + "E": "Five carts, meaning you’re handling two more carts than a group of three." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_023_3", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "One cart, since you only have a single cart with you right now.", + "E": "Five carts, meaning you’re handling two more carts than a group of three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_4", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many shopping carts do I have now?", + "answer": "Zero.", + "question_chinese": "现在我拿了几个购物车", + "answer_chinese": "0个", + "evidence": [ + { + "description": "Now all commodities are held in hand.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:20" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Zero.", + "D": "Three carts, so you currently have multiple carts with you instead of none.", + "E": "Four carts, indicating you’ve gathered several carts rather than having zero." + }, + "answer_label": "C", + "answer_text": "Zero.", + "q_id": "egolife_A1_023_4", + "correct_answer": "Zero.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Zero.", + "D": "Three carts, so you currently have multiple carts with you instead of none.", + "E": "Four carts, indicating you’ve gathered several carts rather than having zero." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_2_T1", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far am I from the sunflower right now?", + "answer": "It was in front of you.", + "question_chinese": "现在我里向日葵有多远？", + "answer_chinese": "它在你面前", + "evidence": [ + { + "description": "I am now standing at the shopping cart collection point.", + "timestamps": { + "start": "00:07:10", + "end": "00:07:20" + } + }, + { + "description": "The sunflower hasn't moved; it's about a hundred meters from the shopping cart collection point.", + "timestamps": { + "start": "00:07:20", + "end": "00:07:30" + } + } + ], + "question_timestamps": "00:07:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "It was in front of you.", + "D": "Off to your right side, not directly ahead in your line of sight.", + "E": "Above you, so you had to look upward rather than straight ahead to see it." + }, + "answer_label": "C", + "answer_text": "It was in front of you.", + "q_id": "egolife_A1_023_2_T1", + "correct_answer": "It was in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "It was in front of you.", + "D": "Off to your right side, not directly ahead in your line of sight.", + "E": "Above you, so you had to look upward rather than straight ahead to see it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_023", + "video_length": "00:12:00", + "qid": "egolife_A1_023_2_T2", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far am I from the sunflower right now?", + "answer": "100 meters.", + "question_chinese": "现在我里向日葵有多远？", + "answer_chinese": "100米", + "evidence": [ + { + "description": "I am now standing at the shopping cart collection point.", + "timestamps": { + "start": "00:07:10", + "end": "00:07:20" + } + } + ], + "question_timestamps": "00:07:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "90 meters.", + "B": "110 meters.", + "C": "About 75 meters away from the sunflower, not as far as a full football field.", + "D": "100 meters.", + "E": "Roughly 130 meters from the sunflower, farther than you could reach in a short walk." + }, + "answer_label": "D", + "answer_text": "100 meters.", + "q_id": "egolife_A1_023_2_T2", + "correct_answer": "100 meters.", + "original_options": { + "A": "90 meters.", + "B": "110 meters.", + "C": "About 75 meters away from the sunflower, not as far as a full football field.", + "D": "100 meters.", + "E": "Roughly 130 meters from the sunflower, farther than you could reach in a short walk." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + } +] diff --git a/QAs/egolife_A1_024-mcq.json b/QAs/egolife_A1_024-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8971f55251787c8e357762f3aca9e9de57fe2562 --- /dev/null +++ b/QAs/egolife_A1_024-mcq.json @@ -0,0 +1,376 @@ +[ + { + "video_uid": "egolife_A1_024", + "video_length": "00:11:21", + "qid": "egolife_A1_024_9", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "The trajectory of a scallion.", + "answer": "From the elevator entrance to the supermarket entrance.", + "question_chinese": "大葱的运动轨迹", + "answer_chinese": "从电梯口处到超市门口", + "evidence": [ + { + "description": "I was holding a green Chinese onion and walked from the elevator entrance to the supermarket entrance.", + "timestamps": { + "start": "00:01:00", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:02:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the supermarket entrance to the elevator entrance.", + "B": "From the elevator entrance to the dairy section.", + "C": "From the elevator entrance to the supermarket entrance.", + "D": "From the elevator entrance to the bakery department, ending deeper inside the store.", + "E": "From the elevator entrance to the supermarket loading dock, moving past the entrance area." + }, + "answer_label": "C", + "answer_text": "From the elevator entrance to the supermarket entrance.", + "q_id": "egolife_A1_024_9", + "correct_answer": "From the elevator entrance to the supermarket entrance.", + "original_options": { + "A": "From the supermarket entrance to the elevator entrance.", + "B": "From the elevator entrance to the dairy section.", + "C": "From the elevator entrance to the supermarket entrance.", + "D": "From the elevator entrance to the bakery department, ending deeper inside the store.", + "E": "From the elevator entrance to the supermarket loading dock, moving past the entrance area." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_024", + "video_length": "00:11:21", + "qid": "egolife_A1_024_10", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "The trajectory of a scallion.", + "answer": "From the elevator entrance to the supermarket entrance and then to the trunk.", + "question_chinese": "大葱的运动轨迹", + "answer_chinese": "从电梯口到超市门口再到后备箱", + "evidence": [ + { + "description": "I held the green onions, walked from the elevator entrance to the supermarket entrance and then to the car, and put the green onions in the trunk.", + "timestamps": { + "start": "00:01:00", + "end": "00:04:30" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the elevator entrance to the supermarket entrance and then to the trunk.", + "B": "From the elevator entrance to the trunk.", + "C": "From the supermarket entrance to the trunk.", + "D": "From the elevator entrance to the supermarket entrance and then to the back seat of the car.", + "E": "From the elevator entrance to the parking lot entrance and then to the trunk, skipping the supermarket entrance." + }, + "answer_label": "A", + "answer_text": "From the elevator entrance to the supermarket entrance and then to the trunk.", + "q_id": "egolife_A1_024_10", + "correct_answer": "From the elevator entrance to the supermarket entrance and then to the trunk.", + "original_options": { + "A": "From the elevator entrance to the supermarket entrance and then to the trunk.", + "B": "From the elevator entrance to the trunk.", + "C": "From the supermarket entrance to the trunk.", + "D": "From the elevator entrance to the supermarket entrance and then to the back seat of the car.", + "E": "From the elevator entrance to the parking lot entrance and then to the trunk, skipping the supermarket entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_024", + "video_length": "00:11:21", + "qid": "egolife_A1_024-mcq_fixed_7", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is it from where I am to the ride-hailing pickup area?", + "answer": "Four.", + "question_chinese": "我现在到网约车停放处距离大概多远", + "answer_chinese": "1米", + "evidence": [ + { + "description": "I am standing in front of the car now.", + "timestamps": { + "start": "00:02:10", + "end": "00:04:20" + } + }, + { + "description": "One meter away from the car", + "timestamps": { + "start": "00:02:10", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:04:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two, meaning you’re only a couple steps away from the ride-hailing pickup area.", + "D": "Five, indicating you still have several steps to reach the ride-hailing pickup area.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "egolife_A1_024-mcq_fixed_7", + "correct_answer": "Four.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two, meaning you’re only a couple steps away from the ride-hailing pickup area.", + "D": "Five, indicating you still have several steps to reach the ride-hailing pickup area.", + "E": "Four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_024", + "video_length": "00:11:21", + "qid": "egolife_A1_024-mcq_fixed_6", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is it from where I am to the ride-hailing pickup area?", + "answer": "Five.", + "question_chinese": "我现在到网约车停放处距离大概多远", + "answer_chinese": "250米", + "evidence": [ + { + "description": "It's 200 meters from the entrance of the supermarket to the ride-hailing pick-up area.", + "timestamps": { + "start": "00:00:00", + "end": "00:04:20" + } + }, + { + "description": "It's 50 meters from the elevator to the entrance of the supermarket.", + "timestamps": { + "start": "00:00:00", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Four.", + "C": "Five.", + "D": "Three, so you’re not far but still need a few steps to the ride-hailing pickup area.", + "E": "Two, suggesting the ride-hailing pickup area is very close from where you are standing." + }, + "answer_label": "C", + "answer_text": "Five.", + "q_id": "egolife_A1_024-mcq_fixed_6", + "correct_answer": "Five.", + "original_options": { + "A": "One.", + "B": "Four.", + "C": "Five.", + "D": "Three, so you’re not far but still need a few steps to the ride-hailing pickup area.", + "E": "Two, suggesting the ride-hailing pickup area is very close from where you are standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 14 + ] + }, + { + "video_uid": "egolife_A1_024", + "video_length": "00:11:21", + "qid": "egolife_A1_024_4", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the shopping bag in relation to me right now?", + "answer": "Directly ahead.", + "question_chinese": "现在购物袋在相对于我的什么位置", + "answer_chinese": "正前方", + "evidence": [ + { + "description": "The shopping bag is in the trunk of the car.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:30" + } + }, + { + "description": "I am standing directly behind the car's trunk.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:30" + } + } + ], + "question_timestamps": "00:04:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Directly ahead.", + "B": "To your left.", + "C": "Behind you.", + "D": "To your right, slightly off to the side rather than centered in front of you.", + "E": "On the table nearby, not positioned in the space directly in front of you." + }, + "answer_label": "A", + "answer_text": "Directly ahead.", + "q_id": "egolife_A1_024_4", + "correct_answer": "Directly ahead.", + "original_options": { + "A": "Directly ahead.", + "B": "To your left.", + "C": "Behind you.", + "D": "To your right, slightly off to the side rather than centered in front of you.", + "E": "On the table nearby, not positioned in the space directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_024", + "video_length": "00:11:21", + "qid": "egolife_A1_024_3", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the shopping bag in relation to me right now?", + "answer": "Directly behind.", + "question_chinese": "现在购物袋在相对于我的什么位置", + "answer_chinese": "正后方", + "evidence": [ + { + "description": "The shopping bag is in the trunk.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:55" + } + }, + { + "description": "I am in the car.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:55" + } + } + ], + "question_timestamps": "00:06:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Directly ahead.", + "B": "To your left.", + "C": "To your right, positioned beside you instead of directly at your back.", + "D": "On the table beside you, not located in the space directly behind you.", + "E": "Directly behind." + }, + "answer_label": "E", + "answer_text": "Directly behind.", + "q_id": "egolife_A1_024_3", + "correct_answer": "Directly behind.", + "original_options": { + "A": "Directly ahead.", + "B": "To your left.", + "C": "To your right, positioned beside you instead of directly at your back.", + "D": "On the table beside you, not located in the space directly behind you.", + "E": "Directly behind." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 12, + 13 + ] + } +] diff --git a/QAs/egolife_A1_025-mcq.json b/QAs/egolife_A1_025-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..86989b464c2504c0e66de7037d86789e2a8decdb --- /dev/null +++ b/QAs/egolife_A1_025-mcq.json @@ -0,0 +1,788 @@ +[ + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_7", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to throw away some trash right now, where should I place the trash can?", + "answer": "It should be placed within easy reach, and the trash can should be placed next to the cabinet.", + "question_chinese": "如果我现在想丢垃圾，我应该把垃圾桶放在什么位置", + "answer_chinese": "应该放在触手可及的地方，把垃圾桶放在橱柜旁边", + "evidence": [ + { + "description": "I'm standing in the kitchen", + "timestamps": { + "start": "00:09:30", + "end": "00:09:35" + } + }, + { + "description": "I am 5 centimeters away from the cabinet.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:35" + } + }, + { + "description": "Put the trash bin next to the cabinet.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:35" + } + } + ], + "question_timestamps": "00:09:43", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Place it out of reach.", + "B": "Put it far from the cabinet.", + "C": "It should be placed within easy reach, but positioned in front of the cabinet instead of next to it.", + "D": "Keep it within easy reach, but place the trash can on the other side of the cabinet, not beside it.", + "E": "It should be placed within easy reach, and the trash can should be placed next to the cabinet." + }, + "answer_label": "E", + "answer_text": "It should be placed within easy reach, and the trash can should be placed next to the cabinet.", + "q_id": "egolife_A1_025_7", + "correct_answer": "It should be placed within easy reach, and the trash can should be placed next to the cabinet.", + "original_options": { + "A": "Place it out of reach.", + "B": "Put it far from the cabinet.", + "C": "It should be placed within easy reach, but positioned in front of the cabinet instead of next to it.", + "D": "Keep it within easy reach, but place the trash can on the other side of the cabinet, not beside it.", + "E": "It should be placed within easy reach, and the trash can should be placed next to the cabinet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_8", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to throw away some trash right now, where should I place the trash can?", + "answer": "Put it under the dining table.", + "question_chinese": "如果我现在想丢垃圾，我应该把垃圾桶放在什么位置", + "answer_chinese": "放在餐桌下面", + "evidence": [ + { + "description": "I stand beside the dining table.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:15" + } + }, + { + "description": "I am five centimeters away from the dining table.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:15" + } + }, + { + "description": "Put the trash can next to the dining table where I can reach it with my hand.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:15" + } + } + ], + "question_timestamps": "00:10:23", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Put it in front of the dining table.", + "B": "Place it beside the dining table.", + "C": "Put it under the dining table.", + "D": "Set the trash can next to the dining table so you can drop trash in without bending down.", + "E": "Put the trash can at the edge of the dining area, near the table, so it’s reachable but not underneath." + }, + "answer_label": "C", + "answer_text": "Put it under the dining table.", + "q_id": "egolife_A1_025_8", + "correct_answer": "Put it under the dining table.", + "original_options": { + "A": "Put it in front of the dining table.", + "B": "Place it beside the dining table.", + "C": "Put it under the dining table.", + "D": "Set the trash can next to the dining table so you can drop trash in without bending down.", + "E": "Put the trash can at the edge of the dining area, near the table, so it’s reachable but not underneath." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 18, + 21 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025-mcq_fixed_9", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to get to the kitchen from here?", + "answer": "One.", + "question_chinese": "我现在走到厨房要拐几个弯", + "answer_chinese": "一个", + "evidence": [ + { + "description": "I turn right in the living room to get to the kitchen.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:30" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Two turns: first turn left, then turn right to enter the kitchen from the main walkway.", + "D": "Three turns total, starting with a right turn, then another right, and a final left into the kitchen.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_025-mcq_fixed_9", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Two turns: first turn left, then turn right to enter the kitchen from the main walkway.", + "D": "Three turns total, starting with a right turn, then another right, and a final left into the kitchen.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025-mcq_fixed_8", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to get to the kitchen from here?", + "answer": "Three.", + "question_chinese": "我现在走到厨房要拐几个弯", + "answer_chinese": "三个", + "evidence": [ + { + "description": "I turned left from the kitchen to the interior door, then left again to the main door, and then left once more to the yard.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:55" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two turns.", + "B": "Four turns.", + "C": "Make two turns: one to leave this area, then a second to face into the kitchen.", + "D": "You need four turns overall, adjusting direction at each junction before you finally enter the kitchen.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "egolife_A1_025-mcq_fixed_8", + "correct_answer": "Three.", + "original_options": { + "A": "Two turns.", + "B": "Four turns.", + "C": "Make two turns: one to leave this area, then a second to face into the kitchen.", + "D": "You need four turns overall, adjusting direction at each junction before you finally enter the kitchen.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_4_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where did I see the can positioned relative to the scallions?", + "answer": "It is directly in front of the scallions, and to your left.", + "question_chinese": "我看到现在罐头在相对于大葱的什么位置？", + "answer_chinese": "在大葱的正前方，在你的左方", + "evidence": [ + { + "description": "The can is on the kitchen stove.The kitchen is on the left side of the green onions.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:30" + } + } + ], + "question_timestamps": "00:05:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of them, to your right.", + "B": "Behind them, to your left.", + "C": "It is directly in front of the scallions, and to your left.", + "D": "The can is behind the scallions, positioned slightly to your right when you looked at it.", + "E": "The can is behind the scallions and off to your left side, not in front of them." + }, + "answer_label": "C", + "answer_text": "It is directly in front of the scallions, and to your left.", + "q_id": "egolife_A1_025_4_T2", + "correct_answer": "It is directly in front of the scallions, and to your left.", + "original_options": { + "A": "In front of them, to your right.", + "B": "Behind them, to your left.", + "C": "It is directly in front of the scallions, and to your left.", + "D": "The can is behind the scallions, positioned slightly to your right when you looked at it.", + "E": "The can is behind the scallions and off to your left side, not in front of them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_4_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where did I see the can positioned relative to the scallions?", + "answer": "To your front-left; the canned food is to your back-right.", + "question_chinese": "我看到现在罐头在相对于大葱的什么位置？", + "answer_chinese": "左前方，罐头在你的右后方", + "evidence": [ + { + "description": "The can is on the kitchen stove.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:30" + } + } + ], + "question_timestamps": "00:06:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-right; back-right.", + "B": "To your front-left; the canned food is to your back-right.", + "C": "To your back-left; back-right.", + "D": "To your front-left; the canned food is positioned to your back-left, not behind you on the right.", + "E": "To your back-right; the canned food is to your front-left, meaning they swap positions from what you described." + }, + "answer_label": "B", + "answer_text": "To your front-left; the canned food is to your back-right.", + "q_id": "egolife_A1_025_4_T1", + "correct_answer": "To your front-left; the canned food is to your back-right.", + "original_options": { + "A": "To your front-right; back-right.", + "B": "To your front-left; the canned food is to your back-right.", + "C": "To your back-left; back-right.", + "D": "To your front-left; the canned food is positioned to your back-left, not behind you on the right.", + "E": "To your back-right; the canned food is to your front-left, meaning they swap positions from what you described." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_1_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I stand at the door now, which is closer to me, the backpack or the hat?", + "answer": "Hat.", + "question_chinese": "现在如果我站在门口，书包和帽子哪个离我更近？", + "answer_chinese": "帽子", + "evidence": [ + { + "description": "I took the hat off my head, and the hat was one centimeter away from me.", + "timestamps": { + "start": "00:00:40", + "end": "00:03:40" + } + }, + { + "description": "I took the schoolbag off my back, and it was two centimeters away from me.", + "timestamps": { + "start": "00:00:40", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:04:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Backpack.", + "B": "Hat." + }, + "answer_label": "B", + "answer_text": "Hat.", + "q_id": "egolife_A1_025_1_T2", + "correct_answer": "Hat.", + "original_options": { + "A": "Backpack.", + "B": "Hat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_1_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, if I’m standing at the doorway, which is closer to me: the backpack or the hat?", + "answer": "The hat is behind you.", + "question_chinese": "现在如果我站在门口，书包和帽子哪个离我更近？", + "answer_chinese": "帽子，在你的后方。", + "evidence": [ + { + "description": "I took the hat off my head, and the hat was one centimeter away from me.", + "timestamps": { + "start": "00:00:40", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:04:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The hat is behind you.", + "B": "The backpack is behind you." + }, + "answer_label": "A", + "answer_text": "The hat is behind you.", + "q_id": "egolife_A1_025_1_T1", + "correct_answer": "The hat is behind you.", + "original_options": { + "A": "The hat is behind you.", + "B": "The backpack is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_2_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, if I’m standing at the doorway, which is closer to me: the backpack or the hat?", + "answer": "The backpack is to your left.", + "question_chinese": "现在如果我站在门口，书包和帽子哪个离我更近？", + "answer_chinese": "书包，在你的左方。", + "evidence": [ + { + "description": "I put the hat on the chair in the corner, and the chair is 2 meters away from the door.", + "timestamps": { + "start": "00:00:40", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The hat is to your left.", + "B": "The backpack is to your left." + }, + "answer_label": "B", + "answer_text": "The backpack is to your left.", + "q_id": "egolife_A1_025_2_T1", + "correct_answer": "The backpack is to your left.", + "original_options": { + "A": "The hat is to your left.", + "B": "The backpack is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_2_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me right now if I stand at the door, the backpack or the hat?", + "answer": "Schoolbag.", + "question_chinese": "现在如果我站在门口，书包和帽子哪个离我更近？", + "answer_chinese": "书包", + "evidence": [ + { + "description": "I put the hat on the chair in the corner, and the chair is 2 meters away from the door.", + "timestamps": { + "start": "00:00:40", + "end": "00:03:40" + } + }, + { + "description": "I put my schoolbag on the chair near the door, and the chair is about 0.5 meters away from the door.", + "timestamps": { + "start": "00:00:40", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:07:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Hat.", + "B": "Schoolbag." + }, + "answer_label": "B", + "answer_text": "Schoolbag.", + "q_id": "egolife_A1_025_2_T2", + "correct_answer": "Schoolbag.", + "original_options": { + "A": "Hat.", + "B": "Schoolbag." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_5", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks are on the table now?", + "answer": "Seven bottles.", + "question_chinese": "现在桌子上有几瓶饮料", + "answer_chinese": "七瓶", + "evidence": [ + { + "description": "There are six bottles of Vitasoy and one bottle of fruit juice on the table.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:15" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Six bottles.", + "B": "Eight bottles.", + "C": "There are five bottles of drinks on the table right now, lined up in a row.", + "D": "Seven bottles.", + "E": "Nine bottles of drinks are currently sitting on the table in front of me." + }, + "answer_label": "D", + "answer_text": "Seven bottles.", + "q_id": "egolife_A1_025_5", + "correct_answer": "Seven bottles.", + "original_options": { + "A": "Six bottles.", + "B": "Eight bottles.", + "C": "There are five bottles of drinks on the table right now, lined up in a row.", + "D": "Seven bottles.", + "E": "Nine bottles of drinks are currently sitting on the table in front of me." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_6", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks are on the table now?", + "answer": "Zero.", + "question_chinese": "现在桌子上有几瓶饮料", + "answer_chinese": "0瓶", + "evidence": [ + { + "description": "I put the drink in the refrigerator.", + "timestamps": { + "start": "00:07:40", + "end": "00:07:50" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "Three bottles of drinks are still on the table right now, near the front edge.", + "E": "Four bottles of drinks remain on the table at the moment, placed together in a cluster." + }, + "answer_label": "A", + "answer_text": "Zero.", + "q_id": "egolife_A1_025_6", + "correct_answer": "Zero.", + "original_options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "Three bottles of drinks are still on the table right now, near the front edge.", + "E": "Four bottles of drinks remain on the table at the moment, placed together in a cluster." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_3_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the can relative to the green onion right now?", + "answer": "In front of the green Chinese onion.", + "question_chinese": "现在罐头在相对于大葱的什么位置？", + "answer_chinese": "在大葱的正前方", + "evidence": [ + { + "description": "The canned food is inside the refrigerator.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:05" + } + }, + { + "description": "The refrigerator is directly in front of the green onions.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:04:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind the green onion.", + "B": "Above the green onion.", + "C": "The can is positioned to the left of the green Chinese onion right now.", + "D": "In front of the green Chinese onion.", + "E": "The can sits next to the green Chinese onion, not in front of it." + }, + "answer_label": "D", + "answer_text": "In front of the green Chinese onion.", + "q_id": "egolife_A1_025_3_T1", + "correct_answer": "In front of the green Chinese onion.", + "original_options": { + "A": "Behind the green onion.", + "B": "Above the green onion.", + "C": "The can is positioned to the left of the green Chinese onion right now.", + "D": "In front of the green Chinese onion.", + "E": "The can sits next to the green Chinese onion, not in front of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_025", + "video_length": "00:12:00", + "qid": "egolife_A1_025_3_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the can relative to the green onion right now?", + "answer": "In front of the green Chinese onion.", + "question_chinese": "现在罐头在相对于大葱的什么位置？", + "answer_chinese": "在大葱的正前方", + "evidence": [ + { + "description": "The canned food is inside the refrigerator.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:05" + } + }, + { + "description": "The refrigerator is directly in front of the green onions.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:04:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind the green onion.", + "B": "Below the green onion.", + "C": "The can is positioned to the right of the green Chinese onion right now.", + "D": "In front of the green Chinese onion.", + "E": "The can is next to the green Chinese onion rather than in front of it." + }, + "answer_label": "D", + "answer_text": "In front of the green Chinese onion.", + "q_id": "egolife_A1_025_3_T2", + "correct_answer": "In front of the green Chinese onion.", + "original_options": { + "A": "Behind the green onion.", + "B": "Below the green onion.", + "C": "The can is positioned to the right of the green Chinese onion right now.", + "D": "In front of the green Chinese onion.", + "E": "The can is next to the green Chinese onion rather than in front of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + } +] diff --git a/QAs/egolife_A1_026-mcq.json b/QAs/egolife_A1_026-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..32626ef44786a078c0e398e67d2f2ed23e40c44f --- /dev/null +++ b/QAs/egolife_A1_026-mcq.json @@ -0,0 +1,800 @@ +[ + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the spoon on the bed?", + "answer": "Directly above the bed.", + "question_chinese": "勺子在床的什么位置", + "answer_chinese": "在床的正上方", + "evidence": [ + { + "description": "The spoon is in my pants pocket.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:55" + } + }, + { + "description": "I am in bed now.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:55" + } + } + ], + "question_timestamps": "00:03:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To the left of the bed.", + "B": "Directly below the bed.", + "C": "Directly above the bed.", + "D": "The spoon is positioned to the right side of the bed near the edge of the mattress.", + "E": "The spoon is in front of the bed rather than above it." + }, + "answer_label": "C", + "answer_text": "Directly above the bed.", + "q_id": "egolife_A1_026_7", + "correct_answer": "Directly above the bed.", + "original_options": { + "A": "To the left of the bed.", + "B": "Directly below the bed.", + "C": "Directly above the bed.", + "D": "The spoon is positioned to the right side of the bed near the edge of the mattress.", + "E": "The spoon is in front of the bed rather than above it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 17, + 12 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_8", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the spoon on the bed?", + "answer": "On the left side of the bed.", + "question_chinese": "勺子在床的什么位置", + "answer_chinese": "在床的左侧", + "evidence": [ + { + "description": "I put the spoon on the computer desk.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:55" + } + }, + { + "description": "The computer desk is on the left side of the bed.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:55" + } + } + ], + "question_timestamps": "00:08:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the right side.", + "B": "Near the pillow on the right.", + "C": "The spoon is on the far right corner of the bed close to the headboard.", + "D": "The spoon is on the right side of the bed, positioned just under the blanket.", + "E": "On the left side of the bed." + }, + "answer_label": "E", + "answer_text": "On the left side of the bed.", + "q_id": "egolife_A1_026_8", + "correct_answer": "On the left side of the bed.", + "original_options": { + "A": "On the right side.", + "B": "Near the pillow on the right.", + "C": "The spoon is on the far right corner of the bed close to the headboard.", + "D": "The spoon is on the right side of the bed, positioned just under the blanket.", + "E": "On the left side of the bed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_1_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, which location’s computer should I use?", + "answer": "On the sofa.", + "question_chinese": "如果我现在要使用电脑，我应该用哪个地方的电脑？", + "answer_chinese": "沙发上的", + "evidence": [ + { + "description": "I am standing in the living room now.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:21" + } + }, + { + "description": "Closest to the sofa", + "timestamps": { + "start": "00:00:15", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the sofa.", + "B": "On the chair.", + "C": "On the desk.", + "D": "Use the computer located under the bed right now, not the one on the sofa.", + "E": "Use the computer that is kept inside the drawer instead of the computer on the sofa." + }, + "answer_label": "A", + "answer_text": "On the sofa.", + "q_id": "egolife_A1_026_1_T1", + "correct_answer": "On the sofa.", + "original_options": { + "A": "On the sofa.", + "B": "On the chair.", + "C": "On the desk.", + "D": "Use the computer located under the bed right now, not the one on the sofa.", + "E": "Use the computer that is kept inside the drawer instead of the computer on the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_1_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, which computer should I use?", + "answer": "On the sofa.", + "question_chinese": "如果我现在要使用电脑，我应该用哪个地方的电脑？", + "answer_chinese": "沙发上的", + "evidence": [ + { + "description": "I am standing in the living room now.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:21" + } + }, + { + "description": "Closest to the sofa", + "timestamps": { + "start": "00:00:15", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the sofa.", + "B": "On the chair.", + "C": "On the desk.", + "D": "Use the computer on the chair since it’s closer than the others right now.", + "E": "Go with the computer on the desk because it’s the most accessible at the moment." + }, + "answer_label": "A", + "answer_text": "On the sofa.", + "q_id": "egolife_A1_026_1_T2", + "correct_answer": "On the sofa.", + "original_options": { + "A": "On the sofa.", + "B": "On the chair.", + "C": "On the desk.", + "D": "Use the computer on the chair since it’s closer than the others right now.", + "E": "Go with the computer on the desk because it’s the most accessible at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_2_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, which location’s computer should I use?", + "answer": "The computer under the desk lamp.", + "question_chinese": "如果我现在要使用电脑，我应该用哪个地方的电脑？", + "answer_chinese": "台灯下面的电脑", + "evidence": [ + { + "description": "I stand in front of the desk lamp.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:36" + } + }, + { + "description": "There is a computer placed under the desk lamp.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The laptop on the desk.", + "B": "The computer under the desk lamp.", + "C": "The computer to the left.", + "D": "Use the computer to the right of the desk lamp if you need one immediately.", + "E": "Choose the computer in front of the desk lamp because it’s the nearest option right now." + }, + "answer_label": "B", + "answer_text": "The computer under the desk lamp.", + "q_id": "egolife_A1_026_2_T1", + "correct_answer": "The computer under the desk lamp.", + "original_options": { + "A": "The laptop on the desk.", + "B": "The computer under the desk lamp.", + "C": "The computer to the left.", + "D": "Use the computer to the right of the desk lamp if you need one immediately.", + "E": "Choose the computer in front of the desk lamp because it’s the nearest option right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_2_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, which one should I use?", + "answer": "The computer under the desk lamp.", + "question_chinese": "如果我现在要使用电脑，我应该用哪个地方的电脑？", + "answer_chinese": "台灯下面的电脑", + "evidence": [ + { + "description": "I stand in front of the desk lamp.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:36" + } + }, + { + "description": "There is a computer placed under the desk lamp.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The laptop on the desk.", + "B": "The computer under the desk lamp.", + "C": "The computer to the right.", + "D": "Use the computer to the left of the desk lamp if you want the closest one.", + "E": "Pick the computer in front of the desk lamp since it’s the most reachable right now." + }, + "answer_label": "B", + "answer_text": "The computer under the desk lamp.", + "q_id": "egolife_A1_026_2_T2", + "correct_answer": "The computer under the desk lamp.", + "original_options": { + "A": "The laptop on the desk.", + "B": "The computer under the desk lamp.", + "C": "The computer to the right.", + "D": "Use the computer to the left of the desk lamp if you want the closest one.", + "E": "Pick the computer in front of the desk lamp since it’s the most reachable right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026-mcq_fixed_2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I stand in the restaurant, which one is closer to me, the power bank or the vase?", + "answer": "Charging pack.", + "question_chinese": "当我站在餐厅时，充电宝和花瓶哪个相对于我距离更近", + "answer_chinese": "充电包", + "evidence": [ + { + "description": "The power bank is at the bottom right corner of the dining table.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:18" + } + }, + { + "description": "The vase is on the far left of the dining table.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:18" + } + }, + { + "description": "I am standing at the lower right corner of the dining table.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Charging pack.", + "B": "Vase." + }, + "answer_label": "A", + "answer_text": "Charging pack.", + "q_id": "egolife_A1_026-mcq_fixed_2", + "correct_answer": "Charging pack.", + "original_options": { + "A": "Charging pack.", + "B": "Vase." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_3", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I am standing in the restaurant, which is closer to me, the power bank or the vase?", + "answer": "Power bank.", + "question_chinese": "当我站在餐厅时，充电宝和花瓶哪个相对于我距离更近", + "answer_chinese": "充电宝", + "evidence": [ + { + "description": "The power bank is at the bottom right corner of the dining table.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:18" + } + }, + { + "description": "The vase is on the far left of the dining table.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:18" + } + }, + { + "description": "I am standing at the lower right corner of the dining table.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Power bank.", + "B": "Vase." + }, + "answer_label": "A", + "answer_text": "Power bank.", + "q_id": "egolife_A1_026_3", + "correct_answer": "Power bank.", + "original_options": { + "A": "Power bank.", + "B": "Vase." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_4", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I am standing in the restaurant, which is closer to me, the power bank or the vase?", + "answer": "Vase.", + "question_chinese": "当我站在餐厅时，充电宝和花瓶哪个相对于我距离更近", + "answer_chinese": "花瓶", + "evidence": [ + { + "description": "The power bank is on the bed.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:50" + } + }, + { + "description": "The vase is on the left side of the dining table.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:50" + } + }, + { + "description": "I stand next to the dining table.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:50" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Vase.", + "B": "Power bank." + }, + "answer_label": "A", + "answer_text": "Vase.", + "q_id": "egolife_A1_026_4", + "correct_answer": "Vase.", + "original_options": { + "A": "Vase.", + "B": "Power bank." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_9", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I been to the second floor?", + "answer": "One.", + "question_chinese": "我上了几次二楼", + "answer_chinese": "一次", + "evidence": [ + { + "description": "I am now on the second floor.", + "timestamps": { + "start": "00:00:30", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:03:03", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "You went to the second floor twice during the trip, going up once and returning later.", + "D": "One.", + "E": "You visited the second floor three separate times, each time going upstairs from the ground level." + }, + "answer_label": "D", + "answer_text": "One.", + "q_id": "egolife_A1_026_9", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "You went to the second floor twice during the trip, going up once and returning later.", + "D": "One.", + "E": "You visited the second floor three separate times, each time going upstairs from the ground level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_10", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I been to the second floor?", + "answer": "Two.", + "question_chinese": "我上了几次二楼", + "answer_chinese": "两次", + "evidence": [ + { + "description": "I went to the second floor again.", + "timestamps": { + "start": "00:09:25", + "end": "00:10:00" + } + } + ], + "question_timestamps": "00:11:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Once.", + "B": "Two.", + "C": "Three times.", + "D": "You went to the second floor four times across the video, taking multiple separate trips upstairs.", + "E": "You visited the second floor five different times, returning upstairs again after coming back down." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "egolife_A1_026_10", + "correct_answer": "Two.", + "original_options": { + "A": "Once.", + "B": "Two.", + "C": "Three times.", + "D": "You went to the second floor four times across the video, taking multiple separate trips upstairs.", + "E": "You visited the second floor five different times, returning upstairs again after coming back down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_5", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are currently charging?", + "answer": "One.", + "question_chinese": "正在充电的充电宝有多少个", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There is a power bank charging on the bed.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:14" + } + } + ], + "question_timestamps": "00:02:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Two power banks are plugged in and charging right now on the table.", + "D": "Three power banks appear connected to chargers, with all indicator lights showing charging.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_026_5", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Two power banks are plugged in and charging right now on the table.", + "D": "Three power banks appear connected to chargers, with all indicator lights showing charging.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_026", + "video_length": "00:12:00", + "qid": "egolife_A1_026_6", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are currently charging?", + "answer": "Three.", + "question_chinese": "正在充电的充电宝有多少个", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There are three power banks charging on the bed now.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "Two power banks are plugged in and charging right now, not three.", + "E": "Four power banks are connected to chargers and actively charging at the moment." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_026_6", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "Two power banks are plugged in and charging right now, not three.", + "E": "Four power banks are connected to chargers and actively charging at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + } +] diff --git a/QAs/egolife_A1_027-mcq.json b/QAs/egolife_A1_027-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9bf5fa168df0801b17ff70cf27ba15ae185a1e6a --- /dev/null +++ b/QAs/egolife_A1_027-mcq.json @@ -0,0 +1,798 @@ +[ + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the power bank relative to the vase?", + "answer": "Directly to the right of the vase.", + "question_chinese": "充电宝相对于花瓶的位置是什么", + "answer_chinese": "在花瓶的正右方", + "evidence": [ + { + "description": "The vase is on the left side of the dining table.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:17" + } + }, + { + "description": "The power bank is on the right side of the dining table.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:17" + } + }, + { + "description": "The power bank is on the right side of the vase.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:17" + } + } + ], + "question_timestamps": "00:02:21", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Directly left of the vase.", + "B": "In front of the vase.", + "C": "It is slightly behind the vase while still resting on the same surface.", + "D": "It sits above the vase, aligned over it rather than placed beside it.", + "E": "Directly to the right of the vase." + }, + "answer_label": "E", + "answer_text": "Directly to the right of the vase.", + "q_id": "egolife_A1_027_1", + "correct_answer": "Directly to the right of the vase.", + "original_options": { + "A": "Directly left of the vase.", + "B": "In front of the vase.", + "C": "It is slightly behind the vase while still resting on the same surface.", + "D": "It sits above the vase, aligned over it rather than placed beside it.", + "E": "Directly to the right of the vase." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the power bank relative to the vase?", + "answer": "On the left side of the vase.", + "question_chinese": "充电宝相对于花瓶的位置是什么", + "answer_chinese": "在花瓶的左侧", + "evidence": [ + { + "description": "The power bank is on the bed.", + "timestamps": { + "start": "00:02:18", + "end": "00:02:45" + } + }, + { + "description": "The vase is on the dining table.", + "timestamps": { + "start": "00:02:18", + "end": "00:02:45" + } + }, + { + "description": "The bed is on the left side of the dining table.", + "timestamps": { + "start": "00:02:18", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:10:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "On the right side of the vase.", + "B": "On the left side of the vase.", + "C": "Directly in front of the vase.", + "D": "It is behind the vase, so the vase blocks it from your view.", + "E": "It is underneath the vase, placed lower rather than to either side." + }, + "answer_label": "B", + "answer_text": "On the left side of the vase.", + "q_id": "egolife_A1_027_2", + "correct_answer": "On the left side of the vase.", + "original_options": { + "A": "On the right side of the vase.", + "B": "On the left side of the vase.", + "C": "Directly in front of the vase.", + "D": "It is behind the vase, so the vase blocks it from your view.", + "E": "It is underneath the vase, placed lower rather than to either side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_3", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, where should I go to use one?", + "answer": "The computer on the sofa.", + "question_chinese": "如果我现在想使用电脑，我应该使用哪里的电脑", + "answer_chinese": "沙发上的电脑", + "evidence": [ + { + "description": "I am now standing in front of the sofa.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:19" + } + }, + { + "description": "Closest to the computer on the sofa", + "timestamps": { + "start": "00:00:15", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The computer in front of the sofa.", + "B": "The computer behind the sofa.", + "C": "Go to the computer to the left of the sofa to use it now.", + "D": "The computer on the sofa.", + "E": "Use the computer to the right of the sofa instead of the one on it." + }, + "answer_label": "D", + "answer_text": "The computer on the sofa.", + "q_id": "egolife_A1_027_3", + "correct_answer": "The computer on the sofa.", + "original_options": { + "A": "The computer in front of the sofa.", + "B": "The computer behind the sofa.", + "C": "Go to the computer to the left of the sofa to use it now.", + "D": "The computer on the sofa.", + "E": "Use the computer to the right of the sofa instead of the one on it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_4", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, where should I go to use one?", + "answer": "The computer on the computer desk.", + "question_chinese": "如果我现在想使用电脑，我应该使用哪里的电脑", + "answer_chinese": "电脑桌上的电脑", + "evidence": [ + { + "description": "I am standing in front of the computer desk now.", + "timestamps": { + "start": "00:11:45", + "end": "00:11:48" + } + }, + { + "description": "Closest to the computer on the desk", + "timestamps": { + "start": "00:11:45", + "end": "00:11:48" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The computer behind the desk.", + "B": "The computer in front of the desk.", + "C": "The computer on the computer desk.", + "D": "Use the laptop beside the computer desk if you need a computer right now.", + "E": "Go to the tablet to the left of the computer desk to use it." + }, + "answer_label": "C", + "answer_text": "The computer on the computer desk.", + "q_id": "egolife_A1_027_4", + "correct_answer": "The computer on the computer desk.", + "original_options": { + "A": "The computer behind the desk.", + "B": "The computer in front of the desk.", + "C": "The computer on the computer desk.", + "D": "Use the laptop beside the computer desk if you need a computer right now.", + "E": "Go to the tablet to the left of the computer desk to use it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_4_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now, was I holding a phone or gloves in my hand?", + "answer": "A phone.", + "question_chinese": "我刚手上拿的是手机还是手套？", + "answer_chinese": "手机", + "evidence": [ + { + "description": "What I was holding in my hand just now was a mobile phone.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:50" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Gloves.", + "B": "A phone." + }, + "answer_label": "B", + "answer_text": "A phone.", + "q_id": "egolife_A1_027_4_T1", + "correct_answer": "A phone.", + "original_options": { + "A": "Gloves.", + "B": "A phone." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_3_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now, was I holding a phone or gloves in my hand?", + "answer": "You were holding a phone in your hand just now.", + "question_chinese": "我刚手上拿的是手机还是手套？", + "answer_chinese": "你刚刚手上拿的是手机。", + "evidence": [ + { + "description": "The phone I just had in my hand", + "timestamps": { + "start": "00:05:10", + "end": "00:05:17" + } + } + ], + "question_timestamps": "00:05:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were holding gloves in your hand just now.", + "B": "You were holding a phone in your hand just now." + }, + "answer_label": "B", + "answer_text": "You were holding a phone in your hand just now.", + "q_id": "egolife_A1_027_3_T2", + "correct_answer": "You were holding a phone in your hand just now.", + "original_options": { + "A": "You were holding gloves in your hand just now.", + "B": "You were holding a phone in your hand just now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_4_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now, was I holding a phone or gloves in my hand?", + "answer": "The phone you were holding in your hand just now.", + "question_chinese": "我刚手上拿的是手机还是手套？", + "answer_chinese": "你刚刚手上拿的手机", + "evidence": [ + { + "description": "What I was holding in my hand just now was a mobile phone.", + "timestamps": { + "start": "00:07:25", + "end": "00:07:33" + } + } + ], + "question_timestamps": "00:07:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The phone you were holding in your hand just now.", + "B": "The gloves you were holding in your hand just now." + }, + "answer_label": "A", + "answer_text": "The phone you were holding in your hand just now.", + "q_id": "egolife_A1_027_4_T2", + "correct_answer": "The phone you were holding in your hand just now.", + "original_options": { + "A": "The phone you were holding in your hand just now.", + "B": "The gloves you were holding in your hand just now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_3_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now, was I holding a phone or gloves in my hand?", + "answer": "Gloves.", + "question_chinese": "我刚手上拿的是手机还是手套？", + "answer_chinese": "手套", + "evidence": [ + { + "description": "The gloves I just had in my hand", + "timestamps": { + "start": "00:10:25", + "end": "00:11:00" + } + } + ], + "question_timestamps": "00:11:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Gloves.", + "B": "A phone." + }, + "answer_label": "A", + "answer_text": "Gloves.", + "q_id": "egolife_A1_027_3_T1", + "correct_answer": "Gloves.", + "original_options": { + "A": "Gloves.", + "B": "A phone." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_8", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the can now?", + "answer": "Yes.", + "question_chinese": "我现在伸手可以拿到罐头吗", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The can is directly in front of me.", + "timestamps": { + "start": "00:11:28", + "end": "00:11:29" + } + }, + { + "description": "Less than five centimeters away from me", + "timestamps": { + "start": "00:11:28", + "end": "00:11:29" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_027_8", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_7", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the can now?", + "answer": "No, it's not allowed.", + "question_chinese": "我现在伸手可以拿到罐头吗", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "I am in the room now", + "timestamps": { + "start": "00:11:25", + "end": "00:11:50" + } + }, + { + "description": "The canned food in the kitchen is more than one meter away from the room.", + "timestamps": { + "start": "00:11:25", + "end": "00:11:50" + } + } + ], + "question_timestamps": "00:12:03", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, it's not allowed.", + "B": "Yes, you can reach it now." + }, + "answer_label": "A", + "answer_text": "No, it's not allowed.", + "q_id": "egolife_A1_027_7", + "correct_answer": "No, it's not allowed.", + "original_options": { + "A": "No, it's not allowed.", + "B": "Yes, you can reach it now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_9_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks do I see charging right now?", + "answer": "You saw 2 power banks charging, and the power banks were in front of you.", + "question_chinese": "我看见现在有几个充电宝在充电？", + "answer_chinese": "你看见有2个充电宝在充电，充电宝在你面前", + "evidence": [ + { + "description": "There is two power banks being charged.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:32" + } + } + ], + "question_timestamps": "00:02:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero power banks charging.", + "B": "Three power banks charging.", + "C": "You saw 1 power bank charging, and it was positioned in front of you.", + "D": "You saw 2 power banks charging, and the power banks were in front of you.", + "E": "You saw 4 power banks charging, and they were all located in front of you." + }, + "answer_label": "D", + "answer_text": "You saw 2 power banks charging, and the power banks were in front of you.", + "q_id": "egolife_A1_027_9_T1", + "correct_answer": "You saw 2 power banks charging, and the power banks were in front of you.", + "original_options": { + "A": "Zero power banks charging.", + "B": "Three power banks charging.", + "C": "You saw 1 power bank charging, and it was positioned in front of you.", + "D": "You saw 2 power banks charging, and the power banks were in front of you.", + "E": "You saw 4 power banks charging, and they were all located in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_9_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks do I see charging right now?", + "answer": "You saw 3 power banks charging.", + "question_chinese": "我看见现在有几个充电宝在充电？", + "answer_chinese": "你看见有3个充电宝在充电", + "evidence": [ + { + "description": "There are three power banks charging.", + "timestamps": { + "start": "00:02:34", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One power bank.", + "B": "Two power banks.", + "C": "You saw 4 power banks charging when you looked at them just now.", + "D": "You saw 5 power banks charging at that moment, rather than only three.", + "E": "You saw 3 power banks charging." + }, + "answer_label": "E", + "answer_text": "You saw 3 power banks charging.", + "q_id": "egolife_A1_027_9_T2", + "correct_answer": "You saw 3 power banks charging.", + "original_options": { + "A": "One power bank.", + "B": "Two power banks.", + "C": "You saw 4 power banks charging when you looked at them just now.", + "D": "You saw 5 power banks charging at that moment, rather than only three.", + "E": "You saw 3 power banks charging." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_10", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are currently charging?", + "answer": "One.", + "question_chinese": "现在有几个充电宝在充电", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There is a power bank being charged.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:32" + } + } + ], + "question_timestamps": "00:02:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four power banks are currently charging right now, not just one at the moment.", + "D": "One.", + "E": "Five power banks are currently charging, with several plugged in at once." + }, + "answer_label": "D", + "answer_text": "One.", + "q_id": "egolife_A1_027_10", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four power banks are currently charging right now, not just one at the moment.", + "D": "One.", + "E": "Five power banks are currently charging, with several plugged in at once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 12 + ] + }, + { + "video_uid": "egolife_A1_027", + "video_length": "00:12:00", + "qid": "egolife_A1_027_9", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are currently charging?", + "answer": "Three.", + "question_chinese": "现在有几个充电宝在充电", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There are three power banks charging.", + "timestamps": { + "start": "00:02:34", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Six power banks are currently charging, more than the three you think.", + "E": "Seven power banks are currently charging right now, not only three at the moment." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_027_9", + "correct_answer": "Three.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Six power banks are currently charging, more than the three you think.", + "E": "Seven power banks are currently charging right now, not only three at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + } +] diff --git a/QAs/egolife_A1_028-mcq.json b/QAs/egolife_A1_028-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4696167ad0292648da2c6d130de82d3cce0b09d8 --- /dev/null +++ b/QAs/egolife_A1_028-mcq.json @@ -0,0 +1,741 @@ +[ + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are charging on the second floor?", + "answer": "Two.", + "question_chinese": "二楼有几个电脑在充电", + "answer_chinese": "两个", + "evidence": [ + { + "description": "The computer on the sofa is not charging.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:10" + } + }, + { + "description": "The computers in the corner and on the table are charging.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four computers are charging on the second floor at the moment now.", + "D": "Two.", + "E": "Only one computer is charging on the second floor at the moment." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "egolife_A1_028_7", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four computers are charging on the second floor at the moment now.", + "D": "Two.", + "E": "Only one computer is charging on the second floor at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are charging on the second floor?", + "answer": "Three.", + "question_chinese": "二楼有几个电脑在充电", + "answer_chinese": "三个", + "evidence": [ + { + "description": "I charged the computer on the sofa.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Five computers are charging on the second floor at the moment now.", + "D": "Only two computers are charging on the second floor at the moment.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "egolife_A1_028_8", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Five computers are charging on the second floor at the moment now.", + "D": "Only two computers are charging on the second floor at the moment.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028-mcq_fixed_2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer now, which one should I use?", + "answer": "The computer on the sofa.", + "question_chinese": "如果我现在要使用电脑，我应该用哪台电脑", + "answer_chinese": "沙发上电脑", + "evidence": [ + { + "description": "I stand in front of the sofa.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:55" + } + }, + { + "description": "There is a computer on the sofa.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:55" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The computer on the sofa.", + "B": "The computer behind the sofa.", + "C": "The one on the desk.", + "D": "Use the computer on the desk beside the window since it’s closest to you right now.", + "E": "Choose the desktop in the corner because it’s the most accessible option at this moment." + }, + "answer_label": "A", + "answer_text": "The computer on the sofa.", + "q_id": "egolife_A1_028-mcq_fixed_2", + "correct_answer": "The computer on the sofa.", + "original_options": { + "A": "The computer on the sofa.", + "B": "The computer behind the sofa.", + "C": "The one on the desk.", + "D": "Use the computer on the desk beside the window since it’s closest to you right now.", + "E": "Choose the desktop in the corner because it’s the most accessible option at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028_4", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, which one should I use?", + "answer": "The computer on the chair.", + "question_chinese": "如果我现在要使用电脑，我应该用哪台电脑", + "answer_chinese": "椅子上电脑", + "evidence": [ + { + "description": "I stand in front of the chair.", + "timestamps": { + "start": "00:11:25", + "end": "00:11:30" + } + }, + { + "description": "There is a computer on the chair.", + "timestamps": { + "start": "00:11:25", + "end": "00:11:30" + } + } + ], + "question_timestamps": "00:11:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The computer behind the chair.", + "B": "The computer to the left.", + "C": "The computer on the chair.", + "D": "Use the computer in front of the chair since it’s the closest one to reach quickly.", + "E": "Pick the computer to the right of the chair because it’s easier to access from where you are." + }, + "answer_label": "C", + "answer_text": "The computer on the chair.", + "q_id": "egolife_A1_028_4", + "correct_answer": "The computer on the chair.", + "original_options": { + "A": "The computer behind the chair.", + "B": "The computer to the left.", + "C": "The computer on the chair.", + "D": "Use the computer in front of the chair since it’s the closest one to reach quickly.", + "E": "Pick the computer to the right of the chair because it’s easier to access from where you are." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028_2_T2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, which computer should I use?", + "answer": "You should use the computer on the chair; the computer is in front of you.", + "question_chinese": "如果我现在要使用电脑，我应该用哪台电脑？", + "answer_chinese": "你该用椅子上电脑，电脑在你前面。", + "evidence": [ + { + "description": "I stand in front of the chair.", + "timestamps": { + "start": "00:11:25", + "end": "00:11:30" + } + } + ], + "question_timestamps": "00:11:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You should use the computer on the chair; the computer is in front of you.", + "B": "Use the computer behind you.", + "C": "Use the computer to your left.", + "D": "You should use the computer on the chair; the computer is behind you.", + "E": "You should use the computer on the chair; the computer is to your right of you." + }, + "answer_label": "A", + "answer_text": "You should use the computer on the chair; the computer is in front of you.", + "q_id": "egolife_A1_028_2_T2", + "correct_answer": "You should use the computer on the chair; the computer is in front of you.", + "original_options": { + "A": "You should use the computer on the chair; the computer is in front of you.", + "B": "Use the computer behind you.", + "C": "Use the computer to your left.", + "D": "You should use the computer on the chair; the computer is behind you.", + "E": "You should use the computer on the chair; the computer is to your right of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028_2_T1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, which computer should I use?", + "answer": "You should use the computer on the desk; the computer is in front of you.", + "question_chinese": "如果我现在要使用电脑，我应该用哪台电脑？", + "answer_chinese": "你应该用桌子上电脑，电脑在你前面。", + "evidence": [ + { + "description": "There is a computer on the desk.", + "timestamps": { + "start": "00:11:43", + "end": "00:11:57" + } + } + ], + "question_timestamps": "00:11:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use the computer behind you.", + "B": "Use the computer to your right.", + "C": "You should use the computer on the desk; the computer is behind you.", + "D": "You should use the computer on the desk; the computer is in front of you.", + "E": "You should use the computer on the desk; the computer is to your left of you." + }, + "answer_label": "D", + "answer_text": "You should use the computer on the desk; the computer is in front of you.", + "q_id": "egolife_A1_028_2_T1", + "correct_answer": "You should use the computer on the desk; the computer is in front of you.", + "original_options": { + "A": "Use the computer behind you.", + "B": "Use the computer to your right.", + "C": "You should use the computer on the desk; the computer is behind you.", + "D": "You should use the computer on the desk; the computer is in front of you.", + "E": "You should use the computer on the desk; the computer is to your left of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028_5", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the blanket located in relation to the table lamp?", + "answer": "Directly below the desk lamp.", + "question_chinese": "毛毯相对于台灯在什么位置", + "answer_chinese": "在台灯的正下方", + "evidence": [ + { + "description": "The blanket is next to the computer in the corner.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:28" + } + }, + { + "description": "The computer is directly under the desk lamp.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:28" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To the left of the desk lamp.", + "B": "Directly behind the desk lamp.", + "C": "In front of the desk lamp on the floor, not underneath it.", + "D": "Above the desk lamp on the table, rather than hanging below it.", + "E": "Directly below the desk lamp." + }, + "answer_label": "E", + "answer_text": "Directly below the desk lamp.", + "q_id": "egolife_A1_028_5", + "correct_answer": "Directly below the desk lamp.", + "original_options": { + "A": "To the left of the desk lamp.", + "B": "Directly behind the desk lamp.", + "C": "In front of the desk lamp on the floor, not underneath it.", + "D": "Above the desk lamp on the table, rather than hanging below it.", + "E": "Directly below the desk lamp." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028_6", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the blanket located in relation to the table lamp?", + "answer": "Left front.", + "question_chinese": "毛毯相对于台灯在什么位置", + "answer_chinese": "左前方", + "evidence": [ + { + "description": "The blanket is directly in front of the sofa.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:50" + } + }, + { + "description": "The sofa is to the front-left of the desk lamp.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:50" + } + } + ], + "question_timestamps": "00:06:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Right front.", + "B": "Left front.", + "C": "Left back.", + "D": "Directly behind the desk lamp, not positioned toward the front-left area there.", + "E": "On the table beside the lamp instead of sitting left and in front." + }, + "answer_label": "B", + "answer_text": "Left front.", + "q_id": "egolife_A1_028_6", + "correct_answer": "Left front.", + "original_options": { + "A": "Right front.", + "B": "Left front.", + "C": "Left back.", + "D": "Directly behind the desk lamp, not positioned toward the front-left area there.", + "E": "On the table beside the lamp instead of sitting left and in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028_9", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get my phone now?", + "answer": "Yes.", + "question_chinese": "现在我想拿到手机可以拿的到吗", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The mobile phone is five centimeters away on the left side.", + "timestamps": { + "start": "00:07:50", + "end": "00:07:55" + } + } + ], + "question_timestamps": "00:08:03", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_028_9", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028_10", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get my phone now?", + "answer": "No.", + "question_chinese": "现在我想拿到手机可以拿的到吗", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "I'm sitting on a chair, and the chair is one meter away from the sofa, so I can't reach it.", + "timestamps": { + "start": "00:11:20", + "end": "00:11:34" + } + } + ], + "question_timestamps": "00:11:38", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_028_10", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028_1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the laptop charger located in relation to the dining table?", + "answer": "Left side.", + "question_chinese": "电脑充电器在相对于餐桌什么位置", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The computer charger is on the bed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "The bed is inside the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "The dining table is on the right side of the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Left side.", + "B": "Right side.", + "C": "In front of it.", + "D": "Behind the dining table, closer to the far edge than the near side.", + "E": "On the right side of the dining table, near the corner rather than centered." + }, + "answer_label": "A", + "answer_text": "Left side.", + "q_id": "egolife_A1_028_1", + "correct_answer": "Left side.", + "original_options": { + "A": "Left side.", + "B": "Right side.", + "C": "In front of it.", + "D": "Behind the dining table, closer to the far edge than the near side.", + "E": "On the right side of the dining table, near the corner rather than centered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_028", + "video_length": "00:12:00", + "qid": "egolife_A1_028_2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the laptop charger located in relation to the dining table?", + "answer": "Above.", + "question_chinese": "电脑充电器在相对于餐桌什么位置", + "answer_chinese": "上方", + "evidence": [ + { + "description": "The computer charger is on the sofa.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:55" + } + }, + { + "description": "The sofa is in the living room on the second floor.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:55" + } + }, + { + "description": "The second floor is directly above the dining table.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:55" + } + } + ], + "question_timestamps": "00:03:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Below.", + "B": "To the side.", + "C": "Behind the dining table, positioned lower than the tabletop and not above it.", + "D": "Above.", + "E": "Beside the dining table rather than above it, closer to the edge than the center." + }, + "answer_label": "D", + "answer_text": "Above.", + "q_id": "egolife_A1_028_2", + "correct_answer": "Above.", + "original_options": { + "A": "Below.", + "B": "To the side.", + "C": "Behind the dining table, positioned lower than the tabletop and not above it.", + "D": "Above.", + "E": "Beside the dining table rather than above it, closer to the edge than the center." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 3, + 13, + 15 + ] + } +] diff --git a/QAs/egolife_A1_029-mcq.json b/QAs/egolife_A1_029-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..81c9ed054f5e5f2c5385768d92808b5f8695078c --- /dev/null +++ b/QAs/egolife_A1_029-mcq.json @@ -0,0 +1,605 @@ +[ + { + "video_uid": "egolife_A1_029", + "video_length": "00:12:00", + "qid": "egolife_A1_029_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there a pot in the refrigerator right now?", + "answer": "Yes.", + "question_chinese": "冰箱里面现在有锅吗", + "answer_chinese": "有", + "evidence": [ + { + "description": "I just took the pot out of the refrigerator, which proves that there was a pot in the refrigerator before.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:09:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_029_7", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_029", + "video_length": "00:12:00", + "qid": "egolife_A1_029_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there a pot in the refrigerator right now?", + "answer": "No.", + "question_chinese": "冰箱里面现在有锅吗", + "answer_chinese": "没有", + "evidence": [ + { + "description": "I took the pot out of the refrigerator, and now there's no pot in the refrigerator.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:20" + } + } + ], + "question_timestamps": "00:11:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_029_8", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_029", + "video_length": "00:12:00", + "qid": "egolife_A1_029_5", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What is the position of the desk lamp relative to me?", + "answer": "Right behind me.", + "question_chinese": "台灯在相对于我的什么位置", + "answer_chinese": "在我的正后方", + "evidence": [ + { + "description": "I am right in front of the table.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:25" + } + }, + { + "description": "The desk lamp is directly in front of the table, and the distance from me to the table is less than the distance from the desk lamp to the table.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:01:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of me.", + "B": "Left behind me.", + "C": "Right behind me.", + "D": "Directly to your right side, not behind you where you can’t easily see it.", + "E": "In front of you and slightly to the right, rather than positioned behind your back." + }, + "answer_label": "C", + "answer_text": "Right behind me.", + "q_id": "egolife_A1_029_5", + "correct_answer": "Right behind me.", + "original_options": { + "A": "In front of me.", + "B": "Left behind me.", + "C": "Right behind me.", + "D": "Directly to your right side, not behind you where you can’t easily see it.", + "E": "In front of you and slightly to the right, rather than positioned behind your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_029", + "video_length": "00:12:00", + "qid": "egolife_A1_029_6", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What is the position of the desk lamp relative to me?", + "answer": "In front of me to the left.", + "question_chinese": "台灯在相对于我的什么位置", + "answer_chinese": "在我的左前方", + "evidence": [ + { + "description": "I am next to the sofa.", + "timestamps": { + "start": "00:07:30", + "end": "00:08:13" + } + }, + { + "description": "The desk lamp is to the left front relative to the sofa.", + "timestamps": { + "start": "00:07:30", + "end": "00:08:13" + } + } + ], + "question_timestamps": "00:08:19", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of me.", + "B": "Behind me right.", + "C": "In front of you to the right, not on the left side of your view.", + "D": "Behind you to the left, so it’s not positioned ahead of you at all.", + "E": "In front of me to the left." + }, + "answer_label": "E", + "answer_text": "In front of me to the left.", + "q_id": "egolife_A1_029_6", + "correct_answer": "In front of me to the left.", + "original_options": { + "A": "In front of me.", + "B": "Behind me right.", + "C": "In front of you to the right, not on the left side of your view.", + "D": "Behind you to the left, so it’s not positioned ahead of you at all.", + "E": "In front of me to the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_029", + "video_length": "00:12:00", + "qid": "egolife_A1_029_3", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I need to use a computer right now, which computer should I use?", + "answer": "The computer on the table.", + "question_chinese": "如果我现在需要使用电脑的话，应该用哪个地方电脑", + "answer_chinese": "桌子上的电脑", + "evidence": [ + { + "description": "I am sitting on a chair in front of the table.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:30" + } + }, + { + "description": "There is a computer directly above the table.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The computer near the window.", + "B": "The computer behind the table.", + "C": "Use the computer in front of the table since it’s closer than the one on the desk.", + "D": "The computer on the table.", + "E": "Use the computer on the desk beside the meeting room door because it’s the nearest option." + }, + "answer_label": "D", + "answer_text": "The computer on the table.", + "q_id": "egolife_A1_029_3", + "correct_answer": "The computer on the table.", + "original_options": { + "A": "The computer near the window.", + "B": "The computer behind the table.", + "C": "Use the computer in front of the table since it’s closer than the one on the desk.", + "D": "The computer on the table.", + "E": "Use the computer on the desk beside the meeting room door because it’s the nearest option." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_029", + "video_length": "00:12:00", + "qid": "egolife_A1_029_4", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I need to use a computer right now, which computer should I use?", + "answer": "The computer on the chair.", + "question_chinese": "如果我现在需要使用电脑的话，应该用哪个地方电脑", + "answer_chinese": "椅子上的电脑", + "evidence": [ + { + "description": "I am sitting five centimeters in front of the chair.", + "timestamps": { + "start": "00:05:10", + "end": "00:05:15" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The computer behind the chair.", + "B": "The computer on the chair.", + "C": "The computer to the left of the chair.", + "D": "Use the computer in front of the chair since it’s closest for you to reach right now.", + "E": "Use the computer to the right of the chair because it’s the easiest one to access immediately." + }, + "answer_label": "B", + "answer_text": "The computer on the chair.", + "q_id": "egolife_A1_029_4", + "correct_answer": "The computer on the chair.", + "original_options": { + "A": "The computer behind the chair.", + "B": "The computer on the chair.", + "C": "The computer to the left of the chair.", + "D": "Use the computer in front of the chair since it’s closest for you to reach right now.", + "E": "Use the computer to the right of the chair because it’s the easiest one to access immediately." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_029", + "video_length": "00:12:00", + "qid": "egolife_A1_029_1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach my phone right now?", + "answer": "No.", + "question_chinese": "我现在伸手可以拿到手机吗", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "The mobile phone is on the sofa.", + "timestamps": { + "start": "00:03:50", + "end": "00:04:00" + } + }, + { + "description": "I am standing 50 centimeters directly in front of the sofa.", + "timestamps": { + "start": "00:03:50", + "end": "00:04:00" + } + }, + { + "description": "The range I can reach by stretching out my hand is about fifteen centimeters.", + "timestamps": { + "start": "00:03:50", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:04:04", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_029_1", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_029", + "video_length": "00:12:00", + "qid": "egolife_A1_029_2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach my phone right now?", + "answer": "Yes.", + "question_chinese": "我现在伸手可以拿到手机吗", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The mobile phone is on the sofa.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:25" + } + }, + { + "description": "I sit five centimeters in front of the sofa.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:25" + } + }, + { + "description": "The range of my reach is fifteen centimeters.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:25" + } + } + ], + "question_timestamps": "00:04:33", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_029_2", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_029", + "video_length": "00:12:00", + "qid": "egolife_A1_029_9", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the pot's movement?", + "answer": "It hasn't moved yet.", + "question_chinese": "锅的运动轨迹是什么", + "answer_chinese": "此时还没有动", + "evidence": [ + { + "description": "The pot is inside the refrigerator.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:07" + } + } + ], + "question_timestamps": "00:09:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It rolled across the counter.", + "B": "It fell from the shelf.", + "C": "It moved from the sink area over to the table in a single motion.", + "D": "It hasn't moved yet.", + "E": "It slid along the countertop and then dropped down off the edge onto the floor." + }, + "answer_label": "D", + "answer_text": "It hasn't moved yet.", + "q_id": "egolife_A1_029_9", + "correct_answer": "It hasn't moved yet.", + "original_options": { + "A": "It rolled across the counter.", + "B": "It fell from the shelf.", + "C": "It moved from the sink area over to the table in a single motion.", + "D": "It hasn't moved yet.", + "E": "It slid along the countertop and then dropped down off the edge onto the floor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_029", + "video_length": "00:12:00", + "qid": "egolife_A1_029_10", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the pot's movement?", + "answer": "The pot goes from the refrigerator on the second floor to the kitchen on the first floor.", + "question_chinese": "锅的运动轨迹是什么", + "answer_chinese": "锅从二楼的冰箱到一楼的厨房", + "evidence": [ + { + "description": "I took the pot out of the refrigerator and used it in the kitchen on the first floor.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:30" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The pot goes from the refrigerator on the second floor to the kitchen on the first floor.", + "B": "From the bedroom upstairs to downstairs.", + "C": "From the first-floor kitchen to the pantry.", + "D": "The pot goes from the kitchen on the first floor up to the pantry on the second floor.", + "E": "The pot travels from a cupboard on the ground floor to the dining room on the second floor." + }, + "answer_label": "A", + "answer_text": "The pot goes from the refrigerator on the second floor to the kitchen on the first floor.", + "q_id": "egolife_A1_029_10", + "correct_answer": "The pot goes from the refrigerator on the second floor to the kitchen on the first floor.", + "original_options": { + "A": "The pot goes from the refrigerator on the second floor to the kitchen on the first floor.", + "B": "From the bedroom upstairs to downstairs.", + "C": "From the first-floor kitchen to the pantry.", + "D": "The pot goes from the kitchen on the first floor up to the pantry on the second floor.", + "E": "The pot travels from a cupboard on the ground floor to the dining room on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + } +] diff --git a/QAs/egolife_A1_030-mcq.json b/QAs/egolife_A1_030-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6d53dd090b1a9e7b1dd71d093d3b8bcb6da6ad0d --- /dev/null +++ b/QAs/egolife_A1_030-mcq.json @@ -0,0 +1,1071 @@ +[ + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the distance between the carpet on the floor and the table lamp?", + "answer": "50 centimeters.", + "question_chinese": "地上的地毯距离台灯的距离是多少", + "answer_chinese": "50厘米", + "evidence": [ + { + "description": "The blanket is directly in front of the left end of the sofa.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:56" + } + }, + { + "description": "The desk lamp is fifty centimeters to the left front of the sofa.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:56" + } + } + ], + "question_timestamps": "00:06:07", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "30 centimeters.", + "B": "50 centimeters.", + "C": "70 centimeters.", + "D": "The distance between the carpet on the floor and the table lamp is 80 centimeters.", + "E": "The carpet on the floor is only twenty centimeters away from the table lamp." + }, + "answer_label": "B", + "answer_text": "50 centimeters.", + "q_id": "egolife_A1_030_1", + "correct_answer": "50 centimeters.", + "original_options": { + "A": "30 centimeters.", + "B": "50 centimeters.", + "C": "70 centimeters.", + "D": "The distance between the carpet on the floor and the table lamp is 80 centimeters.", + "E": "The carpet on the floor is only twenty centimeters away from the table lamp." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the distance from the carpet on the floor to the table lamp?", + "answer": "60 centimeters.", + "question_chinese": "地上的地毯距离台灯的距离是多少", + "answer_chinese": "60厘米", + "evidence": [ + { + "description": "The blanket is directly in front of the right end of the sofa.", + "timestamps": { + "start": "00:05:55", + "end": "00:06:10" + } + }, + { + "description": "The right end of the sofa is 60 centimeters away from the desk lamp.", + "timestamps": { + "start": "00:05:55", + "end": "00:06:10" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "60 centimeters.", + "B": "75 centimeters.", + "C": "85 centimeters.", + "D": "The distance from the carpet on the floor to the table lamp is 90 centimeters.", + "E": "From the carpet on the floor to the table lamp, the distance measures 95 centimeters." + }, + "answer_label": "A", + "answer_text": "60 centimeters.", + "q_id": "egolife_A1_030_2", + "correct_answer": "60 centimeters.", + "original_options": { + "A": "60 centimeters.", + "B": "75 centimeters.", + "C": "85 centimeters.", + "D": "The distance from the carpet on the floor to the table lamp is 90 centimeters.", + "E": "From the carpet on the floor to the table lamp, the distance measures 95 centimeters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_5_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the power bank on the bed relative to me?", + "answer": "On your right-hand side.", + "question_chinese": "床上的充电宝在相对于我什么位置？", + "answer_chinese": "在你的右手边", + "evidence": [ + { + "description": "The power bank is on the bed, between the computer and the doll.", + "timestamps": { + "start": "00:01:09", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your right-hand side.", + "B": "On your left-hand side.", + "C": "Right in front of you.", + "D": "Slightly behind you, closer to your left side on the bed itself.", + "E": "In front of you on the bed, centered rather than off to your right side." + }, + "answer_label": "A", + "answer_text": "On your right-hand side.", + "q_id": "egolife_A1_030_5_T1", + "correct_answer": "On your right-hand side.", + "original_options": { + "A": "On your right-hand side.", + "B": "On your left-hand side.", + "C": "Right in front of you.", + "D": "Slightly behind you, closer to your left side on the bed itself.", + "E": "In front of you on the bed, centered rather than off to your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_5_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the power bank on the bed relative to me?", + "answer": "Right in front of you.", + "question_chinese": "床上的充电宝在相对于我什么位置？", + "answer_chinese": "在你的正前方", + "evidence": [ + { + "description": "The power bank is on the bed, between the computer and the doll.", + "timestamps": { + "start": "00:01:09", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:01:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right in front of you.", + "B": "On your left side.", + "C": "Right behind you.", + "D": "Off to your right side on the bed rather than directly in front of you.", + "E": "Slightly behind you, not in your direct line of sight on the bed." + }, + "answer_label": "A", + "answer_text": "Right in front of you.", + "q_id": "egolife_A1_030_5_T2", + "correct_answer": "Right in front of you.", + "original_options": { + "A": "Right in front of you.", + "B": "On your left side.", + "C": "Right behind you.", + "D": "Off to your right side on the bed rather than directly in front of you.", + "E": "Slightly behind you, not in your direct line of sight on the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_2_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far am I from the floor lamp in front of the sofa in the room?", + "answer": "One meter.", + "question_chinese": "我与房间里沙发面前的落地灯的距离是多少？", + "answer_chinese": "一米", + "evidence": [ + { + "description": "The floor lamp is in front of the sofa, about one meter away from me.", + "timestamps": { + "start": "00:05:55", + "end": "00:06:10" + } + } + ], + "question_timestamps": "00:06:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "50 centimeters.", + "B": "60 centimeters.", + "C": "About two meters away from the floor lamp in front of the sofa.", + "D": "One meter.", + "E": "Less than half a meter from the floor lamp near the sofa." + }, + "answer_label": "D", + "answer_text": "One meter.", + "q_id": "egolife_A1_030_2_T1", + "correct_answer": "One meter.", + "original_options": { + "A": "50 centimeters.", + "B": "60 centimeters.", + "C": "About two meters away from the floor lamp in front of the sofa.", + "D": "One meter.", + "E": "Less than half a meter from the floor lamp near the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_2_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far am I from the floor lamp in front of the sofa in the room?", + "answer": "More than five meters.", + "question_chinese": "我与房间里沙发面前的落地灯的距离是多少？", + "answer_chinese": "超过五米", + "evidence": [ + { + "description": "The floor lamp is in front of the sofa, about one meter away from me.", + "timestamps": { + "start": "00:05:55", + "end": "00:06:10" + } + }, + { + "description": "I am in the living room now, and the sofa is not here.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:31" + } + } + ], + "question_timestamps": "00:09:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "More than five meters.", + "B": "One meter.", + "C": "60 centimeters.", + "D": "About two meters away from the floor lamp in front of the sofa.", + "E": "Roughly half a meter from the floor lamp positioned by the sofa." + }, + "answer_label": "A", + "answer_text": "More than five meters.", + "q_id": "egolife_A1_030_2_T2", + "correct_answer": "More than five meters.", + "original_options": { + "A": "More than five meters.", + "B": "One meter.", + "C": "60 centimeters.", + "D": "About two meters away from the floor lamp in front of the sofa.", + "E": "Roughly half a meter from the floor lamp positioned by the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030-mcq_fixed_6", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to get into the kitchen now?", + "answer": "Three.", + "question_chinese": "我现在进入厨房要拐几个弯", + "answer_chinese": "三个", + "evidence": [ + { + "description": "I turn right three times to reach the kitchen.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:15" + } + } + ], + "question_timestamps": "00:07:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You only need to make one turn to enter the kitchen from here.", + "E": "You would make five turns total before reaching the kitchen entrance now." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_030-mcq_fixed_6", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You only need to make one turn to enter the kitchen from here.", + "E": "You would make five turns total before reaching the kitchen entrance now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030-mcq_fixed_7", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to get into the kitchen now?", + "answer": "One.", + "question_chinese": "我现在进入厨房要拐几个弯", + "answer_chinese": "一个", + "evidence": [ + { + "description": "I turned right one times to reach the kitchen.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:20" + } + } + ], + "question_timestamps": "00:11:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "You need to make four turns before you can enter the kitchen.", + "E": "It takes five turns in total to get into the kitchen from here." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "egolife_A1_030-mcq_fixed_7", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "You need to make four turns before you can enter the kitchen.", + "E": "It takes five turns in total to get into the kitchen from here." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_1_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the carpet on the floor from the desk lamp?", + "answer": "50 cm.", + "question_chinese": "我看见地上的地毯距离台灯的距离是多少？", + "answer_chinese": "50厘米", + "evidence": [ + { + "description": "The blanket is directly in front of the left end of the sofa.The desk lamp is fifty centimeters to the left front of the sofa.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:56" + } + } + ], + "question_timestamps": "00:06:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "60 cm.", + "B": "One meter.", + "C": "50 cm.", + "D": "About 30 cm, with the carpet almost touching the desk lamp area.", + "E": "Roughly 1.5 meters away, with a noticeable gap between carpet and desk lamp." + }, + "answer_label": "C", + "answer_text": "50 cm.", + "q_id": "egolife_A1_030_1_T1", + "correct_answer": "50 cm.", + "original_options": { + "A": "60 cm.", + "B": "One meter.", + "C": "50 cm.", + "D": "About 30 cm, with the carpet almost touching the desk lamp area.", + "E": "Roughly 1.5 meters away, with a noticeable gap between carpet and desk lamp." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_1_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the carpet on the floor from the desk lamp?", + "answer": "60 cm.", + "question_chinese": "我看见地上的地毯距离台灯的距离是多少？", + "answer_chinese": "60厘米", + "evidence": [ + { + "description": "The blanket is directly in front of the right end of the sofa.The right end of the sofa is 60 centimeters away from the desk lamp.", + "timestamps": { + "start": "00:05:55", + "end": "00:06:10" + } + } + ], + "question_timestamps": "00:06:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "50 cm.", + "B": "One meter.", + "C": "About 30 cm, so the carpet is very close to the desk lamp.", + "D": "Roughly 1.5 meters away, leaving a larger space between carpet and desk lamp.", + "E": "60 cm." + }, + "answer_label": "E", + "answer_text": "60 cm.", + "q_id": "egolife_A1_030_1_T2", + "correct_answer": "60 cm.", + "original_options": { + "A": "50 cm.", + "B": "One meter.", + "C": "About 30 cm, so the carpet is very close to the desk lamp.", + "D": "Roughly 1.5 meters away, leaving a larger space between carpet and desk lamp.", + "E": "60 cm." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_5", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What position is the power bank on the table in relation to the screen?", + "answer": "In front of the screen.", + "question_chinese": "桌子上的充电宝在相对于屏幕的什么位置", + "answer_chinese": "在屏幕的正前方", + "evidence": [ + { + "description": "The table is directly in front of the screen.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:06" + } + }, + { + "description": "The power bank is directly above the table.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:06" + } + }, + { + "description": "The power bank is directly in front of the screen.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "To the left of the screen.", + "B": "Behind the screen.", + "C": "In front of the screen.", + "D": "Sitting just to the right of the screen, not directly in front of it on the table.", + "E": "Placed behind the screen on the tabletop, so it is farther from you than the screen is." + }, + "answer_label": "C", + "answer_text": "In front of the screen.", + "q_id": "egolife_A1_030_5", + "correct_answer": "In front of the screen.", + "original_options": { + "A": "To the left of the screen.", + "B": "Behind the screen.", + "C": "In front of the screen.", + "D": "Sitting just to the right of the screen, not directly in front of it on the table.", + "E": "Placed behind the screen on the tabletop, so it is farther from you than the screen is." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 3, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_6", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the power bank on the table relative to the screen?", + "answer": "In the front left of the screen.", + "question_chinese": "桌子上的充电宝在相对于屏幕的什么位置", + "answer_chinese": "在屏幕的左前方", + "evidence": [ + { + "description": "The power bank is directly above the bed.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:20" + } + }, + { + "description": "The bed is in front of the left side of the screen.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:20" + } + }, + { + "description": "The power bank is in the front left of the screen.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:11:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "In the front right of the screen.", + "B": "In the back left of the screen.", + "C": "In the back right of the screen, positioned diagonally behind the screen rather than forward-left.", + "D": "In the front center of the screen, aligned straight ahead instead of being offset to the left.", + "E": "In the front left of the screen." + }, + "answer_label": "E", + "answer_text": "In the front left of the screen.", + "q_id": "egolife_A1_030_6", + "correct_answer": "In the front left of the screen.", + "original_options": { + "A": "In the front right of the screen.", + "B": "In the back left of the screen.", + "C": "In the back right of the screen, positioned diagonally behind the screen rather than forward-left.", + "D": "In the front center of the screen, aligned straight ahead instead of being offset to the left.", + "E": "In the front left of the screen." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_4_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the power bank on the table relative to me?", + "answer": "To your right.", + "question_chinese": "桌子上的充电宝在相对于我什么位置？", + "answer_chinese": "在你的右边", + "evidence": [ + { + "description": "The table is directly in front of the screen.The power bank is directly above the table.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:10:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "In front of you.", + "C": "Behind you on the table, so it is out of your direct forward view when you face the table.", + "D": "Slightly in front of you and centered, rather than offset to either side of your body.", + "E": "To your right." + }, + "answer_label": "E", + "answer_text": "To your right.", + "q_id": "egolife_A1_030_4_T1", + "correct_answer": "To your right.", + "original_options": { + "A": "To your left.", + "B": "In front of you.", + "C": "Behind you on the table, so it is out of your direct forward view when you face the table.", + "D": "Slightly in front of you and centered, rather than offset to either side of your body.", + "E": "To your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_4_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the power bank on the table relative to me?", + "answer": "In the living room behind you.", + "question_chinese": "桌子上的充电宝在相对于我什么位置？", + "answer_chinese": "在你的后边的客厅里", + "evidence": [ + { + "description": "The table is directly in front of the screen.The power bank is directly above the table.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:10:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the living room in front of you.", + "B": "In the living room to your left.", + "C": "In the living room to your right, positioned off to your side rather than behind your back.", + "D": "In the living room directly in front of you, so you would face it without needing to turn around.", + "E": "In the living room behind you." + }, + "answer_label": "E", + "answer_text": "In the living room behind you.", + "q_id": "egolife_A1_030_4_T2", + "correct_answer": "In the living room behind you.", + "original_options": { + "A": "In the living room in front of you.", + "B": "In the living room to your left.", + "C": "In the living room to your right, positioned off to your side rather than behind your back.", + "D": "In the living room directly in front of you, so you would face it without needing to turn around.", + "E": "In the living room behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_9", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the bowls inside the cabinet relative to the kitchen knife?", + "answer": "Directly below.", + "question_chinese": "橱柜里面的碗在相对于菜刀得什么位置", + "answer_chinese": "正下方", + "evidence": [ + { + "description": "The bowl is inside the cabinet.", + "timestamps": { + "start": "00:11:00", + "end": "00:11:08" + } + }, + { + "description": "The kitchen knife is directly above the cabinet.", + "timestamps": { + "start": "00:11:00", + "end": "00:11:08" + } + } + ], + "question_timestamps": "00:11:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Directly above.", + "B": "Directly below.", + "C": "To the left.", + "D": "To the right of the kitchen knife, sitting side-by-side instead of stacked below it in the cabinet.", + "E": "Behind the kitchen knife inside the cabinet, so the knife is closer to the cabinet door than the bowls." + }, + "answer_label": "B", + "answer_text": "Directly below.", + "q_id": "egolife_A1_030_9", + "correct_answer": "Directly below.", + "original_options": { + "A": "Directly above.", + "B": "Directly below.", + "C": "To the left.", + "D": "To the right of the kitchen knife, sitting side-by-side instead of stacked below it in the cabinet.", + "E": "Behind the kitchen knife inside the cabinet, so the knife is closer to the cabinet door than the bowls." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_10", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the bowls inside the cabinet relative to the kitchen knife?", + "answer": "Right side.", + "question_chinese": "橱柜里面的碗在相对于菜刀得什么位置", + "answer_chinese": "右侧", + "evidence": [ + { + "description": "The bowl is on the right side of the cutting board.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:20" + } + }, + { + "description": "The kitchen knife is directly above the cutting board.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:20" + } + } + ], + "question_timestamps": "00:11:26", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Left side.", + "B": "Front side.", + "C": "Directly above the kitchen knife, placed higher on the cabinet shelf rather than positioned to the side.", + "D": "Right side.", + "E": "Behind the kitchen knife in the cabinet, so the bowls sit farther back instead of on its right side." + }, + "answer_label": "D", + "answer_text": "Right side.", + "q_id": "egolife_A1_030_10", + "correct_answer": "Right side.", + "original_options": { + "A": "Left side.", + "B": "Front side.", + "C": "Directly above the kitchen knife, placed higher on the cabinet shelf rather than positioned to the side.", + "D": "Right side.", + "E": "Behind the kitchen knife in the cabinet, so the bowls sit farther back instead of on its right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_3", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "The number of computers that are currently turned on.", + "answer": "Five.", + "question_chinese": "现在开着机的电脑的个数", + "answer_chinese": "五个", + "evidence": [ + { + "description": "There are two on the sofa, one on the chair, one on the table, and one computer in the corner, all of which are turned on.", + "timestamps": { + "start": "00:00:00", + "end": "00:03:55" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "Only three computers are turned on right now; the remaining ones are powered off or asleep.", + "D": "Seven computers are currently on, with several screens lit up and running at the same time.", + "E": "Five." + }, + "answer_label": "E", + "answer_text": "Five.", + "q_id": "egolife_A1_030_3", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "Only three computers are turned on right now; the remaining ones are powered off or asleep.", + "D": "Seven computers are currently on, with several screens lit up and running at the same time.", + "E": "Five." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_030", + "video_length": "00:11:43", + "qid": "egolife_A1_030_4", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "The number of computers that are currently turned on.", + "answer": "Four.", + "question_chinese": "现在开着机的电脑的个数", + "answer_chinese": "四个", + "evidence": [ + { + "description": "I turned off one of the computers on the sofa.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:35" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Only two computers are turned on at the moment, while the others appear shut down or sleeping.", + "E": "Six computers are currently powered on, with multiple displays clearly lit and active." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "egolife_A1_030_4", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Only two computers are turned on at the moment, while the others appear shut down or sleeping.", + "E": "Six computers are currently powered on, with multiple displays clearly lit and active." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 13 + ] + } +] diff --git a/QAs/egolife_A1_031-mcq.json b/QAs/egolife_A1_031-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5682f8a63b98ee512bb24ef1e03358efef434801 --- /dev/null +++ b/QAs/egolife_A1_031-mcq.json @@ -0,0 +1,1239 @@ +[ + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the takeout bag relative to the rice cooker?", + "answer": "Right front.", + "question_chinese": "外卖袋相对于电饭煲的什么位置", + "answer_chinese": "右前方", + "evidence": [ + { + "description": "The takeout bag is on the dining table.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:45" + } + }, + { + "description": "The dining table is in front of and to the right of the rice cooker.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:02:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Left front.", + "B": "Right back.", + "C": "Right front.", + "D": "It sits behind the rice cooker, slightly offset to the right rather than toward the front.", + "E": "The bag is in front of the rice cooker, but shifted to the left side instead of the right." + }, + "answer_label": "C", + "answer_text": "Right front.", + "q_id": "egolife_A1_031_7", + "correct_answer": "Right front.", + "original_options": { + "A": "Left front.", + "B": "Right back.", + "C": "Right front.", + "D": "It sits behind the rice cooker, slightly offset to the right rather than toward the front.", + "E": "The bag is in front of the rice cooker, but shifted to the left side instead of the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_8", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the takeout bag in relation to the rice cooker?", + "answer": "Right side.", + "question_chinese": "外卖袋相对于电饭煲的什么位置", + "answer_chinese": "右侧", + "evidence": [ + { + "description": "Heating a takeout bag in the microwave", + "timestamps": { + "start": "00:06:00", + "end": "00:06:09" + } + }, + { + "description": "The microwave oven is directly to the right of the rice cooker.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:09" + } + } + ], + "question_timestamps": "00:10:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "Left side.", + "C": "In front.", + "D": "It is positioned behind the rice cooker rather than next to it on either side.", + "E": "The bag is above the rice cooker, not placed on the same level to its right." + }, + "answer_label": "A", + "answer_text": "Right side.", + "q_id": "egolife_A1_031_8", + "correct_answer": "Right side.", + "original_options": { + "A": "Right side.", + "B": "Left side.", + "C": "In front.", + "D": "It is positioned behind the rice cooker rather than next to it on either side.", + "E": "The bag is above the rice cooker, not placed on the same level to its right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031-mcq_fixed_8", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many flights of stairs do I need to climb to get to the second floor now?", + "answer": "Two layers.", + "question_chinese": "我现在上二楼要上几层楼梯", + "answer_chinese": "两层", + "evidence": [ + { + "description": "I climbed the stairs twice to reach the second floor.", + "timestamps": { + "start": "00:06:35", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "One layer.", + "B": "Three layers.", + "C": "Two layers.", + "D": "Zero layers, because you’re already on the second floor and don’t need any stairs.", + "E": "One continuous flight with a long landing, so it counts as a single layer to reach upstairs." + }, + "answer_label": "C", + "answer_text": "Two layers.", + "q_id": "egolife_A1_031-mcq_fixed_8", + "correct_answer": "Two layers.", + "original_options": { + "A": "One layer.", + "B": "Three layers.", + "C": "Two layers.", + "D": "Zero layers, because you’re already on the second floor and don’t need any stairs.", + "E": "One continuous flight with a long landing, so it counts as a single layer to reach upstairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031-mcq_fixed_9", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many flights of stairs do I need to climb to get to the second floor now?", + "answer": "Zero layer.", + "question_chinese": "我现在上二楼要上几层楼梯", + "answer_chinese": "零层", + "evidence": [ + { + "description": "I'm on the second floor right now.", + "timestamps": { + "start": "00:06:35", + "end": "00:06:55" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "One layer.", + "B": "Two layers.", + "C": "Zero layer.", + "D": "Three layers, requiring an extra set of stairs after the main landing to reach the second floor.", + "E": "Two flights total, with one short flight to a landing and another flight up to the second floor." + }, + "answer_label": "C", + "answer_text": "Zero layer.", + "q_id": "egolife_A1_031-mcq_fixed_9", + "correct_answer": "Zero layer.", + "original_options": { + "A": "One layer.", + "B": "Two layers.", + "C": "Zero layer.", + "D": "Three layers, requiring an extra set of stairs after the main landing to reach the second floor.", + "E": "Two flights total, with one short flight to a landing and another flight up to the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031-mcq_fixed_2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "I want to be able to reach the takeout bag just by stretching out my hand now. Where should I put the takeout bag?", + "answer": "Put the bag on the computer desk.", + "question_chinese": "我现在想伸手就能拿到外卖袋，我应该把外卖袋放在哪里", + "answer_chinese": "把袋子放在电脑桌上", + "evidence": [ + { + "description": "I am sitting in front of the computer desk now.", + "timestamps": { + "start": "00:03:30", + "end": "00:03:32" + } + }, + { + "description": "Five centimeters away from the computer desk", + "timestamps": { + "start": "00:03:30", + "end": "00:03:32" + } + } + ], + "question_timestamps": "00:03:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Put it under the desk.", + "B": "Put it on the nightstand.", + "C": "Place the takeout bag just behind the computer desk so you can reach back for it without standing.", + "D": "Put the bag on the computer desk.", + "E": "Put the takeout bag on the coffee table beside you instead of placing it on the computer desk." + }, + "answer_label": "D", + "answer_text": "Put the bag on the computer desk.", + "q_id": "egolife_A1_031-mcq_fixed_2", + "correct_answer": "Put the bag on the computer desk.", + "original_options": { + "A": "Put it under the desk.", + "B": "Put it on the nightstand.", + "C": "Place the takeout bag just behind the computer desk so you can reach back for it without standing.", + "D": "Put the bag on the computer desk.", + "E": "Put the takeout bag on the coffee table beside you instead of placing it on the computer desk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_7_T1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to be able to reach out and grab the takeout bag right now, where should I place the takeout bag?", + "answer": "You put the bag on the computer desk, and the computer desk was in front of you.", + "question_chinese": "我现在想伸手就能拿到外卖袋，我应该把外卖袋放在哪里？", + "answer_chinese": "把袋子放在电脑桌上，电脑桌在你面前。", + "evidence": [ + { + "description": "I am sitting in front of the computer desk now.", + "timestamps": { + "start": "00:03:30", + "end": "00:03:32" + } + } + ], + "question_timestamps": "00:03:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the desk behind you.", + "B": "On the desk to your left.", + "C": "You put the bag on the computer desk, but the desk was positioned to your right side instead of in front.", + "D": "You put the bag on the computer desk, and the computer desk was in front of you.", + "E": "You placed the bag on the computer desk, yet the desk was behind you so it wasn’t directly reachable forward." + }, + "answer_label": "D", + "answer_text": "You put the bag on the computer desk, and the computer desk was in front of you.", + "q_id": "egolife_A1_031_7_T1", + "correct_answer": "You put the bag on the computer desk, and the computer desk was in front of you.", + "original_options": { + "A": "On the desk behind you.", + "B": "On the desk to your left.", + "C": "You put the bag on the computer desk, but the desk was positioned to your right side instead of in front.", + "D": "You put the bag on the computer desk, and the computer desk was in front of you.", + "E": "You placed the bag on the computer desk, yet the desk was behind you so it wasn’t directly reachable forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_3_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to be able to reach out and grab the takeout bag right now, where should I place the takeout bag?", + "answer": "It was placed on the dining table, and the dining table was in front of you.", + "question_chinese": "我现在想伸手就能拿到外卖袋，我应该把外卖袋放在哪里？", + "answer_chinese": "放在餐桌上，餐桌在你面前。", + "evidence": [ + { + "description": "I stand in front of the dining table.", + "timestamps": { + "start": "00:05:50", + "end": "00:05:54" + } + } + ], + "question_timestamps": "00:05:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Dining table behind you.", + "B": "It was placed on the dining table, and the dining table was in front of you.", + "C": "Dining table to your left.", + "D": "It was placed on the dining table, but the dining table was to your right rather than directly in front.", + "E": "It was placed on the dining table, and the dining table was behind you so you’d have to turn around." + }, + "answer_label": "B", + "answer_text": "It was placed on the dining table, and the dining table was in front of you.", + "q_id": "egolife_A1_031_3_T1", + "correct_answer": "It was placed on the dining table, and the dining table was in front of you.", + "original_options": { + "A": "Dining table behind you.", + "B": "It was placed on the dining table, and the dining table was in front of you.", + "C": "Dining table to your left.", + "D": "It was placed on the dining table, but the dining table was to your right rather than directly in front.", + "E": "It was placed on the dining table, and the dining table was behind you so you’d have to turn around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_3_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to be able to reach out and grab the takeout bag right now, where should I place the takeout bag?", + "answer": "It was placed in the microwave. You were standing in front of the microwave.", + "question_chinese": "我现在想伸手就能拿到外卖袋，我应该把外卖袋放在哪里？", + "answer_chinese": "放在微波炉里。你站在微波炉面前。", + "evidence": [ + { + "description": "I stand in front of the microwave.", + "timestamps": { + "start": "00:06:50", + "end": "00:06:56" + } + } + ], + "question_timestamps": "00:06:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the microwave door.", + "B": "On the dining table in front.", + "C": "It was placed in the microwave. You were standing in front of the microwave.", + "D": "It was placed in the microwave, but you were standing behind the microwave instead of in front of it.", + "E": "You put the bag on the computer desk, and the desk was in front of you instead of using the microwave." + }, + "answer_label": "C", + "answer_text": "It was placed in the microwave. You were standing in front of the microwave.", + "q_id": "egolife_A1_031_3_T2", + "correct_answer": "It was placed in the microwave. You were standing in front of the microwave.", + "original_options": { + "A": "In the microwave door.", + "B": "On the dining table in front.", + "C": "It was placed in the microwave. You were standing in front of the microwave.", + "D": "It was placed in the microwave, but you were standing behind the microwave instead of in front of it.", + "E": "You put the bag on the computer desk, and the desk was in front of you instead of using the microwave." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_7_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to be able to reach out and grab the takeout bag right now, where should I place the takeout bag?", + "answer": "You put the bag on the sofa, and the sofa was in front of you.", + "question_chinese": "我现在想伸手就能拿到外卖袋，我应该把外卖袋放在哪里？", + "answer_chinese": "把袋子放在沙发上，沙发在你前面。", + "evidence": [ + { + "description": "I am sitting in front of the sofa now.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:06" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it on the sofa behind you.", + "B": "Put it on the sofa to your left.", + "C": "Put the bag on the sofa, but position the sofa to your right so it’s within reach.", + "D": "Place the takeout bag on the sofa above you, since that would still let you reach for it.", + "E": "You put the bag on the sofa, and the sofa was in front of you." + }, + "answer_label": "E", + "answer_text": "You put the bag on the sofa, and the sofa was in front of you.", + "q_id": "egolife_A1_031_7_T2", + "correct_answer": "You put the bag on the sofa, and the sofa was in front of you.", + "original_options": { + "A": "Put it on the sofa behind you.", + "B": "Put it on the sofa to your left.", + "C": "Put the bag on the sofa, but position the sofa to your right so it’s within reach.", + "D": "Place the takeout bag on the sofa above you, since that would still let you reach for it.", + "E": "You put the bag on the sofa, and the sofa was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_5_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the takeout bag relative to the rice cooker?", + "answer": "You saw the takeout bag to the rice cooker’s front-right.", + "question_chinese": "我看见外卖袋相对于电饭煲的什么位置？", + "answer_chinese": "你看见外卖袋在电饭煲的右前方。", + "evidence": [ + { + "description": "The dining table is in front of and to the right of the rice cooker.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:02:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw the takeout bag to the rice cooker’s front-right.", + "B": "Front-left of the rice cooker.", + "C": "Back-right of the rice cooker.", + "D": "You saw the takeout bag to the rice cooker’s back-left, slightly offset from its rear corner.", + "E": "You saw the takeout bag directly behind the rice cooker, rather than at its front-right." + }, + "answer_label": "A", + "answer_text": "You saw the takeout bag to the rice cooker’s front-right.", + "q_id": "egolife_A1_031_5_T1", + "correct_answer": "You saw the takeout bag to the rice cooker’s front-right.", + "original_options": { + "A": "You saw the takeout bag to the rice cooker’s front-right.", + "B": "Front-left of the rice cooker.", + "C": "Back-right of the rice cooker.", + "D": "You saw the takeout bag to the rice cooker’s back-left, slightly offset from its rear corner.", + "E": "You saw the takeout bag directly behind the rice cooker, rather than at its front-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_5_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the takeout bag relative to the rice cooker?", + "answer": "You saw the takeout bag inside the rice cooker.", + "question_chinese": "我看见外卖袋相对于电饭煲的什么位置？", + "answer_chinese": "你看见外卖袋在电饭煲里面", + "evidence": [ + { + "description": "The takeout bag is inside the rice cooker.", + "timestamps": { + "start": "00:06:28", + "end": "00:06:31" + } + } + ], + "question_timestamps": "00:06:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Outside the rice cooker.", + "B": "You saw the takeout bag inside the rice cooker.", + "C": "In front of the rice cooker.", + "D": "You saw the takeout bag to the rice cooker’s front-right, not placed inside it.", + "E": "You saw the takeout bag behind the rice cooker on the counter, rather than inside the cooker." + }, + "answer_label": "B", + "answer_text": "You saw the takeout bag inside the rice cooker.", + "q_id": "egolife_A1_031_5_T2", + "correct_answer": "You saw the takeout bag inside the rice cooker.", + "original_options": { + "A": "Outside the rice cooker.", + "B": "You saw the takeout bag inside the rice cooker.", + "C": "In front of the rice cooker.", + "D": "You saw the takeout bag to the rice cooker’s front-right, not placed inside it.", + "E": "You saw the takeout bag behind the rice cooker on the counter, rather than inside the cooker." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_6_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the takeout bag relative to the rice cooker?", + "answer": "To your front-right.", + "question_chinese": "我看见外卖袋相对于电饭煲的什么位置？", + "answer_chinese": "右前方", + "evidence": [ + { + "description": "The takeaway bag is in my hand,and I'm standing to the front right of the rice cooker.", + "timestamps": { + "start": "00:06:32", + "end": "00:06:35" + } + } + ], + "question_timestamps": "00:06:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "Off to your right side, but closer to your shoulder than your front-right.", + "D": "Directly behind you, so you would have had to turn around to see it.", + "E": "To your front-right." + }, + "answer_label": "E", + "answer_text": "To your front-right.", + "q_id": "egolife_A1_031_6_T2", + "correct_answer": "To your front-right.", + "original_options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "Off to your right side, but closer to your shoulder than your front-right.", + "D": "Directly behind you, so you would have had to turn around to see it.", + "E": "To your front-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_6_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the takeout bag relative to the rice cooker?", + "answer": "On the right side.", + "question_chinese": "我看见外卖袋相对于电饭煲的什么位置？", + "answer_chinese": "右侧", + "evidence": [ + { + "description": "Heating a takeout bag in the microwave", + "timestamps": { + "start": "00:06:00", + "end": "00:06:09" + } + }, + { + "description": "The microwave oven is directly to the right of the rice cooker.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:09" + } + } + ], + "question_timestamps": "00:10:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the left side.", + "B": "In front of you.", + "C": "Behind you relative to the rice cooker, not positioned on its right side.", + "D": "On the right side.", + "E": "On your left side of the scene, rather than on the right side near the rice cooker." + }, + "answer_label": "D", + "answer_text": "On the right side.", + "q_id": "egolife_A1_031_6_T1", + "correct_answer": "On the right side.", + "original_options": { + "A": "On the left side.", + "B": "In front of you.", + "C": "Behind you relative to the rice cooker, not positioned on its right side.", + "D": "On the right side.", + "E": "On your left side of the scene, rather than on the right side near the rice cooker." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_1_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the glasses relative to the dining table?", + "answer": "You saw the glasses above the dining table on the right side.", + "question_chinese": "我看见眼镜在相对于餐桌的什么位置？", + "answer_chinese": "你看见眼镜在餐桌的右上方", + "evidence": [ + { + "description": "The glasses are in my friend's head.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above the table on the left.", + "B": "Above the table in front of you.", + "C": "You saw the glasses above the dining table on the right side.", + "D": "You saw the glasses above the dining table behind you, not on the right side.", + "E": "You saw the glasses above the dining table on the left side instead of above it on the right." + }, + "answer_label": "C", + "answer_text": "You saw the glasses above the dining table on the right side.", + "q_id": "egolife_A1_031_1_T2", + "correct_answer": "You saw the glasses above the dining table on the right side.", + "original_options": { + "A": "Above the table on the left.", + "B": "Above the table in front of you.", + "C": "You saw the glasses above the dining table on the right side.", + "D": "You saw the glasses above the dining table behind you, not on the right side.", + "E": "You saw the glasses above the dining table on the left side instead of above it on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_1_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the glasses relative to the dining table?", + "answer": "Above the dining table, on the left side.", + "question_chinese": "我看见眼镜在相对于餐桌的什么位置？", + "answer_chinese": "在餐桌的左上方", + "evidence": [ + { + "description": "The glasses are in my hand.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + }, + { + "description": "I stand on the left side of the dining table.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + }, + { + "description": "My hand is at a certain distance from the horizontal line of the dining table.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above the dining table, on the right.", + "B": "Above the dining table, behind you.", + "C": "Above the dining table in front of you, rather than on the left side.", + "D": "Above the dining table on the right side, not on the left side where you’d expect.", + "E": "Above the dining table, on the left side." + }, + "answer_label": "E", + "answer_text": "Above the dining table, on the left side.", + "q_id": "egolife_A1_031_1_T1", + "correct_answer": "Above the dining table, on the left side.", + "original_options": { + "A": "Above the dining table, on the right.", + "B": "Above the dining table, behind you.", + "C": "Above the dining table in front of you, rather than on the left side.", + "D": "Above the dining table on the right side, not on the left side where you’d expect.", + "E": "Above the dining table, on the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_2_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the glasses relative to the dining table?", + "answer": "To the left of the dining table.", + "question_chinese": "我看见眼镜在相对于餐桌的什么位置？", + "answer_chinese": "在餐桌的左边", + "evidence": [ + { + "description": "The glasses are in the room on the left of the dining table.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To the right of the dining table.", + "B": "In front of the dining table.", + "C": "To the left of the dining table.", + "D": "Behind the dining table from your viewpoint, rather than positioned to its left.", + "E": "To your right of the dining table, not to the left side where you saw them." + }, + "answer_label": "C", + "answer_text": "To the left of the dining table.", + "q_id": "egolife_A1_031_2_T2", + "correct_answer": "To the left of the dining table.", + "original_options": { + "A": "To the right of the dining table.", + "B": "In front of the dining table.", + "C": "To the left of the dining table.", + "D": "Behind the dining table from your viewpoint, rather than positioned to its left.", + "E": "To your right of the dining table, not to the left side where you saw them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_2_T1", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I see the glasses relative to the dining table?", + "answer": "To the dining table’s front-right.", + "question_chinese": "我看见眼镜在相对于餐桌的什么位置？", + "answer_chinese": "在餐桌的右前方", + "evidence": [ + { + "description": "The glasses are directly above the suitcase.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:45" + } + }, + { + "description": "The suitcase is inside the room.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:45" + } + }, + { + "description": "The room is to the front right of the dining table.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:45" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To the dining table’s front-right.", + "B": "To the dining table’s front-left.", + "C": "To the dining table’s back-right.", + "D": "To the dining table’s back-left, slightly tucked closer to the far edge than the near side.", + "E": "Centered along the dining table’s right side, not in front of it, aligned with the table’s midpoint." + }, + "answer_label": "A", + "answer_text": "To the dining table’s front-right.", + "q_id": "egolife_A1_031_2_T1", + "correct_answer": "To the dining table’s front-right.", + "original_options": { + "A": "To the dining table’s front-right.", + "B": "To the dining table’s front-left.", + "C": "To the dining table’s back-right.", + "D": "To the dining table’s back-left, slightly tucked closer to the far edge than the near side.", + "E": "Centered along the dining table’s right side, not in front of it, aligned with the table’s midpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_5", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many takeout bags are on the table now?", + "answer": "Two.", + "question_chinese": "现在桌子上有几个外卖带袋子", + "answer_chinese": "两个", + "evidence": [ + { + "description": "There are now two takeaway bags on the table.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:53" + } + } + ], + "question_timestamps": "00:03:04", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four takeout bags are on the table now, spread out so none of them is stacked on another.", + "E": "Five takeout bags are on the table now, with one near each corner and one in the middle." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "egolife_A1_031_5", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four takeout bags are on the table now, spread out so none of them is stacked on another.", + "E": "Five takeout bags are on the table now, with one near each corner and one in the middle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_6", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many takeout bags are on the table now?", + "answer": "One.", + "question_chinese": "现在桌子上有几个外卖带袋子", + "answer_chinese": "一个", + "evidence": [ + { + "description": "I took one of the takeaways and put it in the microwave, and there was still one takeaway bag left on the table.", + "timestamps": { + "start": "00:05:50", + "end": "00:06:05" + } + } + ], + "question_timestamps": "00:06:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "Four takeout bags are on the table now, arranged in a loose cluster rather than a single bag.", + "E": "Five takeout bags are on the table now, filling most of the tabletop with multiple bags visible." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "egolife_A1_031_6", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "Four takeout bags are on the table now, arranged in a loose cluster rather than a single bag.", + "E": "Five takeout bags are on the table now, filling most of the tabletop with multiple bags visible." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the glasses in relation to the dining table?", + "answer": "At the upper left of the dining table.", + "question_chinese": "眼镜在相对于餐桌的什么位置", + "answer_chinese": "在餐桌的左上方", + "evidence": [ + { + "description": "The glasses are in my hand.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + }, + { + "description": "I stand on the left side of the dining table.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + }, + { + "description": "My hand is at a certain distance from the horizontal line of the dining table.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "At the upper right of the dining table.", + "B": "At the upper left of the dining table.", + "C": "At the lower left of the dining table.", + "D": "At the lower right of the dining table, nearer the closest edge instead of toward the top side.", + "E": "Along the left side of the dining table around the midpoint, not up in the upper corner area." + }, + "answer_label": "B", + "answer_text": "At the upper left of the dining table.", + "q_id": "egolife_A1_031_1", + "correct_answer": "At the upper left of the dining table.", + "original_options": { + "A": "At the upper right of the dining table.", + "B": "At the upper left of the dining table.", + "C": "At the lower left of the dining table.", + "D": "At the lower right of the dining table, nearer the closest edge instead of toward the top side.", + "E": "Along the left side of the dining table around the midpoint, not up in the upper corner area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_031", + "video_length": "00:12:00", + "qid": "egolife_A1_031_2", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the glasses relative to the dining table?", + "answer": "In front of the right side of the dining table.", + "question_chinese": "眼镜在相对于餐桌的什么位置", + "answer_chinese": "在餐桌的右前方", + "evidence": [ + { + "description": "The glasses are directly above the suitcase.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:45" + } + }, + { + "description": "The suitcase is inside the room.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:45" + } + }, + { + "description": "The room is to the front right of the dining table.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:45" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "In front of the right side of the dining table.", + "B": "Behind the left side.", + "C": "Behind the right side.", + "D": "In front of the left side of the dining table, not on the right-front area.", + "E": "Beside the middle section of the dining table, roughly centered on the long side without being in front." + }, + "answer_label": "A", + "answer_text": "In front of the right side of the dining table.", + "q_id": "egolife_A1_031_2", + "correct_answer": "In front of the right side of the dining table.", + "original_options": { + "A": "In front of the right side of the dining table.", + "B": "Behind the left side.", + "C": "Behind the right side.", + "D": "In front of the left side of the dining table, not on the right-front area.", + "E": "Beside the middle section of the dining table, roughly centered on the long side without being in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 18 + ] + } +] diff --git a/QAs/egolife_A1_032-mcq.json b/QAs/egolife_A1_032-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a68574f91b38bb336af49117b90769b6840b0d98 --- /dev/null +++ b/QAs/egolife_A1_032-mcq.json @@ -0,0 +1,634 @@ +[ + { + "video_uid": "egolife_A1_032", + "video_length": "00:11:45", + "qid": "egolife_A1_032_9", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the takeout bag positioned in relation to the desk lamp?", + "answer": "Left front.", + "question_chinese": "外卖袋相对于台灯在什么位置", + "answer_chinese": "左前方", + "evidence": [ + { + "description": "The takeaway bag is inside the microwave.", + "timestamps": { + "start": "00:01:50", + "end": "00:02:10" + } + }, + { + "description": "The microwave is to the left front of the desk lamp.", + "timestamps": { + "start": "00:01:50", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:02:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Right front.", + "B": "Left front.", + "C": "Left back.", + "D": "Directly behind the desk lamp, lined up so the bag sits straight in the lamp’s rear.", + "E": "To the right of the desk lamp and slightly forward, not on the left-front side of it." + }, + "answer_label": "B", + "answer_text": "Left front.", + "q_id": "egolife_A1_032_9", + "correct_answer": "Left front.", + "original_options": { + "A": "Right front.", + "B": "Left front.", + "C": "Left back.", + "D": "Directly behind the desk lamp, lined up so the bag sits straight in the lamp’s rear.", + "E": "To the right of the desk lamp and slightly forward, not on the left-front side of it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_032", + "video_length": "00:11:45", + "qid": "egolife_A1_032_10", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the takeout bag positioned in relation to the desk lamp?", + "answer": "Directly ahead.", + "question_chinese": "外卖袋相对于台灯在什么位置", + "answer_chinese": "正前方", + "evidence": [ + { + "description": "The takeout bag is directly above the table.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:15" + } + }, + { + "description": "The table is directly in front of the desk lamp.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:08:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind the desk lamp.", + "B": "To the left.", + "C": "To the right of the desk lamp, offset slightly rather than being straight in front of it.", + "D": "Just beside the desk lamp, positioned off to the side instead of directly ahead in the same line.", + "E": "Directly ahead." + }, + "answer_label": "E", + "answer_text": "Directly ahead.", + "q_id": "egolife_A1_032_10", + "correct_answer": "Directly ahead.", + "original_options": { + "A": "Behind the desk lamp.", + "B": "To the left.", + "C": "To the right of the desk lamp, offset slightly rather than being straight in front of it.", + "D": "Just beside the desk lamp, positioned off to the side instead of directly ahead in the same line.", + "E": "Directly ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_032", + "video_length": "00:11:45", + "qid": "egolife_A1_032-mcq_fixed_6", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer now, which one should I use?", + "answer": "A computer on the table.", + "question_chinese": "如果我现在想使用电脑，我应该用哪个电脑", + "answer_chinese": "桌子上电脑", + "evidence": [ + { + "description": "I am now standing next to the table.", + "timestamps": { + "start": "00:08:40", + "end": "00:08:47" + } + }, + { + "description": "Closest to the computer on the table", + "timestamps": { + "start": "00:08:40", + "end": "00:08:47" + } + } + ], + "question_timestamps": "00:08:48", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A computer beside the table.", + "B": "A computer under the table.", + "C": "Use the computer under the table, since it’s the one closest to you at the moment.", + "D": "A computer on the table.", + "E": "Use the computer beside the table, because it’s easier to reach than the one on top." + }, + "answer_label": "D", + "answer_text": "A computer on the table.", + "q_id": "egolife_A1_032-mcq_fixed_6", + "correct_answer": "A computer on the table.", + "original_options": { + "A": "A computer beside the table.", + "B": "A computer under the table.", + "C": "Use the computer under the table, since it’s the one closest to you at the moment.", + "D": "A computer on the table.", + "E": "Use the computer beside the table, because it’s easier to reach than the one on top." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_032", + "video_length": "00:11:45", + "qid": "egolife_A1_032_8", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, which one should I use?", + "answer": "The computer beside the corner.", + "question_chinese": "如果我现在想使用电脑，我应该用哪个电脑", + "answer_chinese": "角落旁的电脑", + "evidence": [ + { + "description": "I am standing in the corner now.", + "timestamps": { + "start": "00:10:20", + "end": "00:10:25" + } + }, + { + "description": "Closest to the computer in the corner", + "timestamps": { + "start": "00:10:20", + "end": "00:10:25" + } + } + ], + "question_timestamps": "00:10:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The laptop on the desk.", + "B": "The desktop on the shelf.", + "C": "The computer beside the corner.", + "D": "Use the computer in front of the corner since it’s closer than the one behind it.", + "E": "Pick the computer behind the corner because it’s the nearest one you can sit at immediately." + }, + "answer_label": "C", + "answer_text": "The computer beside the corner.", + "q_id": "egolife_A1_032_8", + "correct_answer": "The computer beside the corner.", + "original_options": { + "A": "The laptop on the desk.", + "B": "The desktop on the shelf.", + "C": "The computer beside the corner.", + "D": "Use the computer in front of the corner since it’s closer than the one behind it.", + "E": "Pick the computer behind the corner because it’s the nearest one you can sit at immediately." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_032", + "video_length": "00:11:45", + "qid": "egolife_A1_032_1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where should I go to get my backpack?", + "answer": "In front of the table.", + "question_chinese": "我想拿到书包我应该走到哪里", + "answer_chinese": "桌子前方", + "evidence": [ + { + "description": "The schoolbag is directly above the desk.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:07" + } + }, + { + "description": "I walked over to the table.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:07" + } + }, + { + "description": "Standing five centimeters away from the table", + "timestamps": { + "start": "00:04:05", + "end": "00:04:07" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Behind the table.", + "B": "Under the table.", + "C": "Go to the left side of the table, where the backpack is resting near the edge.", + "D": "Move to the right of the table to grab your backpack without needing to step around it.", + "E": "In front of the table." + }, + "answer_label": "E", + "answer_text": "In front of the table.", + "q_id": "egolife_A1_032_1", + "correct_answer": "In front of the table.", + "original_options": { + "A": "Behind the table.", + "B": "Under the table.", + "C": "Go to the left side of the table, where the backpack is resting near the edge.", + "D": "Move to the right of the table to grab your backpack without needing to step around it.", + "E": "In front of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_032", + "video_length": "00:11:45", + "qid": "egolife_A1_032_2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where should I go to get my backpack?", + "answer": "Walk directly in front of the sofa.", + "question_chinese": "我想拿到书包我应该走到哪里", + "answer_chinese": "走到沙发正前方", + "evidence": [ + { + "description": "The schoolbag is directly above the sofa.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:45" + } + }, + { + "description": "I walked to the front of the sofa.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:45" + } + }, + { + "description": "Five centimeters away from the sofa", + "timestamps": { + "start": "00:04:25", + "end": "00:04:45" + } + } + ], + "question_timestamps": "00:08:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Walk to the left of the sofa.", + "B": "Go to the right of the sofa.", + "C": "Head behind the sofa and look there, since the backpack isn’t positioned in front.", + "D": "Walk directly in front of the sofa.", + "E": "Walk up to the TV stand area and check beside it for your backpack instead." + }, + "answer_label": "D", + "answer_text": "Walk directly in front of the sofa.", + "q_id": "egolife_A1_032_2", + "correct_answer": "Walk directly in front of the sofa.", + "original_options": { + "A": "Walk to the left of the sofa.", + "B": "Go to the right of the sofa.", + "C": "Head behind the sofa and look there, since the backpack isn’t positioned in front.", + "D": "Walk directly in front of the sofa.", + "E": "Walk up to the TV stand area and check beside it for your backpack instead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_032", + "video_length": "00:11:45", + "qid": "egolife_A1_032-mcq_fixed_4", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many floors do I need to go down to get to the first floor now?", + "answer": "Two layers.", + "question_chinese": "我现在去一楼要下几层楼", + "answer_chinese": "两层", + "evidence": [ + { + "description": "I went down two floors from the second floor to the first floor.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Two layers.", + "B": "One layer.", + "C": "Three layers.", + "D": "You need to go down one floor, since you’re only one level above the first floor.", + "E": "Go down three floors because you’re currently above an extra intermediate level before reaching first." + }, + "answer_label": "A", + "answer_text": "Two layers.", + "q_id": "egolife_A1_032-mcq_fixed_4", + "correct_answer": "Two layers.", + "original_options": { + "A": "Two layers.", + "B": "One layer.", + "C": "Three layers.", + "D": "You need to go down one floor, since you’re only one level above the first floor.", + "E": "Go down three floors because you’re currently above an extra intermediate level before reaching first." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_032", + "video_length": "00:11:45", + "qid": "egolife_A1_032-mcq_fixed_5", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many floors do I need to go down to get to the first floor now?", + "answer": "Zero layers.", + "question_chinese": "我现在去一楼要下几层楼", + "answer_chinese": "零层", + "evidence": [ + { + "description": "I am on the first floor right now.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Zero layers.", + "B": "One layer.", + "C": "Two layers.", + "D": "Go down one floor, because the first floor is below where you are standing now.", + "E": "Descend three layers since there are multiple levels between you and the first-floor lobby." + }, + "answer_label": "A", + "answer_text": "Zero layers.", + "q_id": "egolife_A1_032-mcq_fixed_5", + "correct_answer": "Zero layers.", + "original_options": { + "A": "Zero layers.", + "B": "One layer.", + "C": "Two layers.", + "D": "Go down one floor, because the first floor is below where you are standing now.", + "E": "Descend three layers since there are multiple levels between you and the first-floor lobby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_032", + "video_length": "00:11:45", + "qid": "egolife_A1_032_3", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks are on the table now?", + "answer": "Zero.", + "question_chinese": "现在桌子上有几瓶饮料", + "answer_chinese": "零瓶", + "evidence": [ + { + "description": "There is no drink on the table now.", + "timestamps": { + "start": "00:00:00", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Zero.", + "C": "Two.", + "D": "Three bottles are on the table now, grouped together near the middle of the tabletop.", + "E": "Four bottles are on the table now, spread across the surface rather than being stacked together." + }, + "answer_label": "B", + "answer_text": "Zero.", + "q_id": "egolife_A1_032_3", + "correct_answer": "Zero.", + "original_options": { + "A": "One.", + "B": "Zero.", + "C": "Two.", + "D": "Three bottles are on the table now, grouped together near the middle of the tabletop.", + "E": "Four bottles are on the table now, spread across the surface rather than being stacked together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_032", + "video_length": "00:11:45", + "qid": "egolife_A1_032_4", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks are on the table now?", + "answer": "Two bottles.", + "question_chinese": "现在桌子上有几瓶饮料", + "answer_chinese": "两瓶", + "evidence": [ + { + "description": "There are two bottles of drinks on the table now.", + "timestamps": { + "start": "00:08:40", + "end": "00:08:50" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One bottle.", + "B": "Three bottles.", + "C": "Two bottles.", + "D": "Four bottles are on the table now, so there are more than the pair you had before.", + "E": "Five bottles are on the table now, filling most of the table surface near the front edge." + }, + "answer_label": "C", + "answer_text": "Two bottles.", + "q_id": "egolife_A1_032_4", + "correct_answer": "Two bottles.", + "original_options": { + "A": "One bottle.", + "B": "Three bottles.", + "C": "Two bottles.", + "D": "Four bottles are on the table now, so there are more than the pair you had before.", + "E": "Five bottles are on the table now, filling most of the table surface near the front edge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + } +] diff --git a/QAs/egolife_A1_033-mcq.json b/QAs/egolife_A1_033-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..80d75c0f5dc9922bda79cad4ec889a57fd1daa46 --- /dev/null +++ b/QAs/egolife_A1_033-mcq.json @@ -0,0 +1,724 @@ +[ + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_3_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to reach out and throw away trash right now, where should I place the trash can?", + "answer": "It was placed under the computer desk.", + "question_chinese": "如果我现在想伸手丢垃圾，我应该把垃圾桶放在哪里？", + "answer_chinese": "放在电脑桌下方", + "evidence": [ + { + "description": "I am standing in front of the computer desk.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:24" + } + }, + { + "description": "5 centimeters away from the computer desk", + "timestamps": { + "start": "00:04:20", + "end": "00:04:24" + } + } + ], + "question_timestamps": "00:04:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was placed beside the desk.", + "B": "It was placed behind the desk.", + "C": "It was placed in the corner by the desk, not underneath where your feet would be.", + "D": "It was placed under the computer desk.", + "E": "It was placed next to the computer desk so you could reach it without leaning under the tabletop." + }, + "answer_label": "D", + "answer_text": "It was placed under the computer desk.", + "q_id": "egolife_A1_033_3_T1", + "correct_answer": "It was placed under the computer desk.", + "original_options": { + "A": "It was placed beside the desk.", + "B": "It was placed behind the desk.", + "C": "It was placed in the corner by the desk, not underneath where your feet would be.", + "D": "It was placed under the computer desk.", + "E": "It was placed next to the computer desk so you could reach it without leaning under the tabletop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_3_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to reach out and throw away trash right now, where should I place the trash can?", + "answer": "It was placed under the computer desk.", + "question_chinese": "如果我现在想伸手丢垃圾，我应该把垃圾桶放在哪里？", + "answer_chinese": "放在电脑桌下方", + "evidence": [ + { + "description": "I am standing in front of the computer desk.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:24" + } + } + ], + "question_timestamps": "00:04:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On top of the computer desk.", + "B": "It was placed under the computer desk.", + "C": "Next to the computer desk.", + "D": "Place it beside the computer desk so you can reach it without leaning under anything.", + "E": "Put it near the dining table corner instead of under the desk, keeping it reachable from where you sit." + }, + "answer_label": "B", + "answer_text": "It was placed under the computer desk.", + "q_id": "egolife_A1_033_3_T2", + "correct_answer": "It was placed under the computer desk.", + "original_options": { + "A": "On top of the computer desk.", + "B": "It was placed under the computer desk.", + "C": "Next to the computer desk.", + "D": "Place it beside the computer desk so you can reach it without leaning under anything.", + "E": "Put it near the dining table corner instead of under the desk, keeping it reachable from where you sit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_4_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to reach out and throw away trash right now, where should I place the trash can?", + "answer": "It was placed on the corner of the dining table, and the dining table was in front of you.", + "question_chinese": "如果我现在想伸手丢垃圾，我应该把垃圾桶放在哪里？", + "answer_chinese": "放在餐桌角落，餐桌在你面前", + "evidence": [ + { + "description": "Five centimeters from the corner of the dining table", + "timestamps": { + "start": "00:07:20", + "end": "00:07:22" + } + } + ], + "question_timestamps": "00:07:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was placed on the corner of the dining table, and the dining table was in front of you.", + "B": "On the table behind you.", + "C": "On the table to your left.", + "D": "It was on the dining table corner, but the dining table was positioned behind you.", + "E": "It was on the dining table corner, with the dining table located to your right side." + }, + "answer_label": "A", + "answer_text": "It was placed on the corner of the dining table, and the dining table was in front of you.", + "q_id": "egolife_A1_033_4_T1", + "correct_answer": "It was placed on the corner of the dining table, and the dining table was in front of you.", + "original_options": { + "A": "It was placed on the corner of the dining table, and the dining table was in front of you.", + "B": "On the table behind you.", + "C": "On the table to your left.", + "D": "It was on the dining table corner, but the dining table was positioned behind you.", + "E": "It was on the dining table corner, with the dining table located to your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_4_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to reach out and throw away trash right now, where should I place the trash can?", + "answer": "It was placed under the computer desk, and the computer desk was behind you.", + "question_chinese": "如果我现在想伸手丢垃圾，我应该把垃圾桶放在哪里？", + "answer_chinese": "放在电脑桌下边，电脑桌在你后边", + "evidence": [ + { + "description": "Five centimeters from the corner of the dining table", + "timestamps": { + "start": "00:07:20", + "end": "00:07:22" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Under the desk in front.", + "B": "It was placed under the computer desk, and the computer desk was behind you.", + "C": "Under the desk to your left.", + "D": "It was placed under the computer desk, and the computer desk was in front of you.", + "E": "It was placed under the computer desk, and the computer desk was to your right." + }, + "answer_label": "B", + "answer_text": "It was placed under the computer desk, and the computer desk was behind you.", + "q_id": "egolife_A1_033_4_T2", + "correct_answer": "It was placed under the computer desk, and the computer desk was behind you.", + "original_options": { + "A": "Under the desk in front.", + "B": "It was placed under the computer desk, and the computer desk was behind you.", + "C": "Under the desk to your left.", + "D": "It was placed under the computer desk, and the computer desk was in front of you.", + "E": "It was placed under the computer desk, and the computer desk was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_7", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where should I go to pick up my glasses?", + "answer": "On the left side of the computer.", + "question_chinese": "我现在想拿眼镜我应该去哪里拿", + "answer_chinese": "在电脑左侧", + "evidence": [ + { + "description": "The glasses are placed on the box to the left of the computer.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:07" + } + } + ], + "question_timestamps": "00:04:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the right side of the computer.", + "B": "On the left side of the computer.", + "C": "In front of the computer.", + "D": "They are behind the computer, so go around to the back to pick them up.", + "E": "Look under the computer area, because the glasses were placed beneath it rather than beside it." + }, + "answer_label": "B", + "answer_text": "On the left side of the computer.", + "q_id": "egolife_A1_033_7", + "correct_answer": "On the left side of the computer.", + "original_options": { + "A": "On the right side of the computer.", + "B": "On the left side of the computer.", + "C": "In front of the computer.", + "D": "They are behind the computer, so go around to the back to pick them up.", + "E": "Look under the computer area, because the glasses were placed beneath it rather than beside it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_8", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where should I go to pick up my glasses?", + "answer": "Go get it from that girl's hand.", + "question_chinese": "我现在想拿眼镜我应该去哪里拿", + "answer_chinese": "去那个女孩手上拿", + "evidence": [ + { + "description": "I took the glasses and put them in the girl's hand.", + "timestamps": { + "start": "00:04:26", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:04:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go get it from that boy's backpack.", + "B": "Go get it from that woman's purse.", + "C": "Go get it from that man's briefcase, not from the girl standing nearby.", + "D": "Go get it from that girl's hand.", + "E": "Go get it from that boy's toy instead of taking it directly from the girl's hand." + }, + "answer_label": "D", + "answer_text": "Go get it from that girl's hand.", + "q_id": "egolife_A1_033_8", + "correct_answer": "Go get it from that girl's hand.", + "original_options": { + "A": "Go get it from that boy's backpack.", + "B": "Go get it from that woman's purse.", + "C": "Go get it from that man's briefcase, not from the girl standing nearby.", + "D": "Go get it from that girl's hand.", + "E": "Go get it from that boy's toy instead of taking it directly from the girl's hand." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_4", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cakes have strawberries on top?", + "answer": "Zero.", + "question_chinese": "现在有几个蛋糕上面有草莓", + "answer_chinese": "零个", + "evidence": [ + { + "description": "There are no strawberries on the cake now.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three cakes are topped with strawberries in the display you are looking at.", + "D": "Four cakes have strawberries on top of them, based on what is visible right now.", + "E": "Zero." + }, + "answer_label": "E", + "answer_text": "Zero.", + "q_id": "egolife_A1_033_4", + "correct_answer": "Zero.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three cakes are topped with strawberries in the display you are looking at.", + "D": "Four cakes have strawberries on top of them, based on what is visible right now.", + "E": "Zero." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_3", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cakes have strawberries on top?", + "answer": "Eight.", + "question_chinese": "现在有几个蛋糕上面有草莓", + "answer_chinese": "八个", + "evidence": [ + { + "description": "There are eight cakes decorated with strawberries now.", + "timestamps": { + "start": "00:08:46", + "end": "00:08:48" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Eight.", + "B": "Seven.", + "C": "Nine.", + "D": "Six cakes have strawberries on top, which is fewer than the total shown.", + "E": "Ten cakes have strawberries on top, counting all the ones you can see." + }, + "answer_label": "A", + "answer_text": "Eight.", + "q_id": "egolife_A1_033_3", + "correct_answer": "Eight.", + "original_options": { + "A": "Eight.", + "B": "Seven.", + "C": "Nine.", + "D": "Six cakes have strawberries on top, which is fewer than the total shown.", + "E": "Ten cakes have strawberries on top, counting all the ones you can see." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The position of the glasses relative to the computer.", + "answer": "On the left side of the computer.", + "question_chinese": "眼镜相对于电脑的位置", + "answer_chinese": "在电脑的左侧", + "evidence": [ + { + "description": "The glasses are directly above the box.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:29" + } + }, + { + "description": "The box is directly to the left of the computer.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:29" + } + }, + { + "description": "The height of the box is the same as that of the computer.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:29" + } + } + ], + "question_timestamps": "00:04:33", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "On the left side of the computer.", + "B": "On the right side of the computer.", + "C": "In front of the computer.", + "D": "Behind the computer on the desk, not positioned to its left side.", + "E": "Above the computer on the desk surface rather than sitting beside it." + }, + "answer_label": "A", + "answer_text": "On the left side of the computer.", + "q_id": "egolife_A1_033_2", + "correct_answer": "On the left side of the computer.", + "original_options": { + "A": "On the left side of the computer.", + "B": "On the right side of the computer.", + "C": "In front of the computer.", + "D": "Behind the computer on the desk, not positioned to its left side.", + "E": "Above the computer on the desk surface rather than sitting beside it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The position of the glasses relative to the computer.", + "answer": "Left front.", + "question_chinese": "眼镜相对于电脑的位置", + "answer_chinese": "左前方", + "evidence": [ + { + "description": "The computer is on the desk in the room.", + "timestamps": { + "start": "00:03:28", + "end": "00:04:35" + } + }, + { + "description": "The dining table is in front of and to the left of the desk in the room.", + "timestamps": { + "start": "00:03:28", + "end": "00:04:35" + } + }, + { + "description": "I stood by the dining table wearing glasses.", + "timestamps": { + "start": "00:03:38", + "end": "00:04:35" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Right front.", + "B": "Left back.", + "C": "The glasses are positioned directly in front of the computer, not off to the left side.", + "D": "Left front.", + "E": "They sit to the computer's right side, slightly forward, rather than on its left-front corner." + }, + "answer_label": "D", + "answer_text": "Left front.", + "q_id": "egolife_A1_033_1", + "correct_answer": "Left front.", + "original_options": { + "A": "Right front.", + "B": "Left back.", + "C": "The glasses are positioned directly in front of the computer, not off to the left side.", + "D": "Left front.", + "E": "They sit to the computer's right side, slightly forward, rather than on its left-front corner." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_6_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the cake from my hand?", + "answer": "There was no distance.", + "question_chinese": "蛋糕距离我手的距离是多少？", + "answer_chinese": "没有距离", + "evidence": [ + { + "description": "My hand is on the chair.", + "timestamps": { + "start": "00:09:40", + "end": "00:09:44" + } + } + ], + "question_timestamps": "00:09:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There was no distance.", + "B": "About 20 centimeters.", + "C": "Around three meters away.", + "D": "It was roughly 40 centimeters from your hand, close but not actually touching it.", + "E": "The cake was a few meters to your left, so you couldn't reach it without moving." + }, + "answer_label": "A", + "answer_text": "There was no distance.", + "q_id": "egolife_A1_033_6_T2", + "correct_answer": "There was no distance.", + "original_options": { + "A": "There was no distance.", + "B": "About 20 centimeters.", + "C": "Around three meters away.", + "D": "It was roughly 40 centimeters from your hand, close but not actually touching it.", + "E": "The cake was a few meters to your left, so you couldn't reach it without moving." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_033", + "video_length": "00:11:30", + "qid": "egolife_A1_033_6_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the cake from my hand?", + "answer": "20 centimeters.", + "question_chinese": "蛋糕距离我手的距离是多少？", + "answer_chinese": "二十厘米", + "evidence": [ + { + "description": "My hand is on the chair.", + "timestamps": { + "start": "00:09:40", + "end": "00:09:44" + } + }, + { + "description": "The chair is twenty centimeters away from the cake.", + "timestamps": { + "start": "00:09:40", + "end": "00:09:44" + } + } + ], + "question_timestamps": "00:09:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There was no distance.", + "B": "20 centimeters.", + "C": "About 40 centimeters.", + "D": "It was about three meters to your left, far beyond arm’s reach from your hand.", + "E": "The cake sat several meters away toward your front-left, not just a short 20-centimeter gap." + }, + "answer_label": "B", + "answer_text": "20 centimeters.", + "q_id": "egolife_A1_033_6_T1", + "correct_answer": "20 centimeters.", + "original_options": { + "A": "There was no distance.", + "B": "20 centimeters.", + "C": "About 40 centimeters.", + "D": "It was about three meters to your left, far beyond arm’s reach from your hand.", + "E": "The cake sat several meters away toward your front-left, not just a short 20-centimeter gap." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 15 + ] + } +] diff --git a/QAs/egolife_A1_034-mcq.json b/QAs/egolife_A1_034-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..23b6c799b0d8876adb1af0fc45b32de9e73be32a --- /dev/null +++ b/QAs/egolife_A1_034-mcq.json @@ -0,0 +1,1147 @@ +[ + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_10", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I would like to know the direction of the chair.", + "answer": "Directly ahead.", + "question_chinese": "我想对于椅子在什么方位", + "answer_chinese": "正前方", + "evidence": [ + { + "description": "The chair is directly in front of the bed.", + "timestamps": { + "start": "00:05:10", + "end": "00:05:14" + } + }, + { + "description": "I am sitting on the bed", + "timestamps": { + "start": "00:05:10", + "end": "00:05:14" + } + } + ], + "question_timestamps": "00:05:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The chair is off to your right side rather than straight in front of you.", + "D": "It sits slightly behind you, so you would need to turn around to face it.", + "E": "Directly ahead." + }, + "answer_label": "E", + "answer_text": "Directly ahead.", + "q_id": "egolife_A1_034_10", + "correct_answer": "Directly ahead.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The chair is off to your right side rather than straight in front of you.", + "D": "It sits slightly behind you, so you would need to turn around to face it.", + "E": "Directly ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_9", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "I would like to know the direction of the chair.", + "answer": "Right side.", + "question_chinese": "我想对于椅子在什么方位", + "answer_chinese": "右侧", + "evidence": [ + { + "description": "I stand at the door", + "timestamps": { + "start": "00:06:45", + "end": "00:06:47" + } + }, + { + "description": "The door is on the right side of the chair.", + "timestamps": { + "start": "00:06:45", + "end": "00:06:47" + } + } + ], + "question_timestamps": "00:06:51", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Right side.", + "B": "Left side.", + "C": "Directly ahead.", + "D": "The chair is behind you, not positioned on your right side at all.", + "E": "It’s in front of you near the table, rather than located to your right." + }, + "answer_label": "A", + "answer_text": "Right side.", + "q_id": "egolife_A1_034_9", + "correct_answer": "Right side.", + "original_options": { + "A": "Right side.", + "B": "Left side.", + "C": "Directly ahead.", + "D": "The chair is behind you, not positioned on your right side at all.", + "E": "It’s in front of you near the table, rather than located to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_1_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the strawberry in my hand from the vase?", + "answer": "60 centimeters.", + "question_chinese": "我手里拿的草莓与花瓶的距离是多少？", + "answer_chinese": "六十厘米", + "evidence": [ + { + "description": "The strawberry is directly below me.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:32" + } + }, + { + "description": "The position where I stand is 60 centimeters away from the vase.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:00:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "40 centimeters.", + "B": "60 centimeters.", + "C": "One meter.", + "D": "It was about three meters away to your left, much farther than a short reach.", + "E": "The strawberry in your hand was nearly touching the vase, leaving almost no gap between them." + }, + "answer_label": "B", + "answer_text": "60 centimeters.", + "q_id": "egolife_A1_034_1_T1", + "correct_answer": "60 centimeters.", + "original_options": { + "A": "40 centimeters.", + "B": "60 centimeters.", + "C": "One meter.", + "D": "It was about three meters away to your left, much farther than a short reach.", + "E": "The strawberry in your hand was nearly touching the vase, leaving almost no gap between them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_1_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the strawberry in my hand from the vase?", + "answer": "60 centimeters.", + "question_chinese": "我手里拿的草莓与花瓶的距离是多少？", + "answer_chinese": "六十厘米", + "evidence": [ + { + "description": "The strawberry is directly below me.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:32" + } + }, + { + "description": "The position where I stand is 60 centimeters away from the vase.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:00:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "40 centimeters.", + "B": "60 centimeters.", + "C": "One meter.", + "D": "It was about three meters away to your left, much farther than a short reach.", + "E": "The strawberry in your hand was nearly touching the vase, leaving almost no gap between them." + }, + "answer_label": "B", + "answer_text": "60 centimeters.", + "q_id": "egolife_A1_034_1_T2", + "correct_answer": "60 centimeters.", + "original_options": { + "A": "40 centimeters.", + "B": "60 centimeters.", + "C": "One meter.", + "D": "It was about three meters away to your left, much farther than a short reach.", + "E": "The strawberry in your hand was nearly touching the vase, leaving almost no gap between them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_1_T3", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the strawberry in my hand from the vase?", + "answer": "40 centimeters.", + "question_chinese": "我手里拿的草莓与花瓶的距离是多少？", + "answer_chinese": "四十厘米", + "evidence": [ + { + "description": "The strawberries are in the trash can under the dining table.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:45" + } + }, + { + "description": "The trash can is forty centimeters away from the vase.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "60 centimeters.", + "B": "40 centimeters.", + "C": "20 centimeters.", + "D": "It was about one meter from the vase, farther than a 40-centimeter separation.", + "E": "The distance was several meters to your left, so the vase was nowhere near your hand." + }, + "answer_label": "B", + "answer_text": "40 centimeters.", + "q_id": "egolife_A1_034_1_T3", + "correct_answer": "40 centimeters.", + "original_options": { + "A": "60 centimeters.", + "B": "40 centimeters.", + "C": "20 centimeters.", + "D": "It was about one meter from the vase, farther than a 40-centimeter separation.", + "E": "The distance was several meters to your left, so the vase was nowhere near your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_1_T4", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the strawberry in my hand from the vase?", + "answer": "40 centimeters.", + "question_chinese": "我手里拿的草莓与花瓶的距离是多少？", + "answer_chinese": "四十厘米", + "evidence": [ + { + "description": "The strawberries are in the trash can under the dining table.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:45" + } + }, + { + "description": "The trash can is forty centimeters away from the vase.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "60 centimeters.", + "B": "20 centimeters.", + "C": "About 25 centimeters from the vase, roughly at arm's reach in front of you.", + "D": "40 centimeters.", + "E": "Roughly 80 centimeters away from the vase, closer to your hand than to the table edge." + }, + "answer_label": "D", + "answer_text": "40 centimeters.", + "q_id": "egolife_A1_034_1_T4", + "correct_answer": "40 centimeters.", + "original_options": { + "A": "60 centimeters.", + "B": "20 centimeters.", + "C": "About 25 centimeters from the vase, roughly at arm's reach in front of you.", + "D": "40 centimeters.", + "E": "Roughly 80 centimeters away from the vase, closer to your hand than to the table edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034-mcq_fixed_7", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many flights of stairs do I need to climb to get to the second floor now?", + "answer": "Two layers.", + "question_chinese": "我现在上二楼要上几层楼梯", + "answer_chinese": "两层", + "evidence": [ + { + "description": "I went up two floors to the second floor.", + "timestamps": { + "start": "00:01:59", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "One layer.", + "B": "Two layers.", + "C": "Three layers.", + "D": "Just one flight of stairs takes you up to the second floor from where you are now.", + "E": "You need to climb three separate flights, with short landings between them before reaching the second floor." + }, + "answer_label": "B", + "answer_text": "Two layers.", + "q_id": "egolife_A1_034-mcq_fixed_7", + "correct_answer": "Two layers.", + "original_options": { + "A": "One layer.", + "B": "Two layers.", + "C": "Three layers.", + "D": "Just one flight of stairs takes you up to the second floor from where you are now.", + "E": "You need to climb three separate flights, with short landings between them before reaching the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034-mcq_fixed_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many flights of stairs do I need to climb to get to the second floor now?", + "answer": "Zero layers.", + "question_chinese": "我现在上二楼要上几层楼梯", + "answer_chinese": "零层", + "evidence": [ + { + "description": "I am now on the second floor..", + "timestamps": { + "start": "00:01:59", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "One layer.", + "B": "Two layers.", + "C": "You still need to climb one full flight of stairs to reach the second floor from your current level.", + "D": "Zero layers.", + "E": "It takes two flights to get to the second floor, with an intermediate landing separating the levels." + }, + "answer_label": "D", + "answer_text": "Zero layers.", + "q_id": "egolife_A1_034-mcq_fixed_6", + "correct_answer": "Zero layers.", + "original_options": { + "A": "One layer.", + "B": "Two layers.", + "C": "You still need to climb one full flight of stairs to reach the second floor from your current level.", + "D": "Zero layers.", + "E": "It takes two flights to get to the second floor, with an intermediate landing separating the levels." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_4_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my data cable relative to the cake?", + "answer": "The data cable in front of you was down-left of the cake.", + "question_chinese": "我的数据线相对于蛋糕的位置是什么？", + "answer_chinese": "你面前的数据线相对于蛋糕的左下方", + "evidence": [ + { + "description": "The data cable is directly above the sofa.", + "timestamps": { + "start": "00:02:11", + "end": "00:02:12" + } + }, + { + "description": "The sofa is on the second floor.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:11" + } + }, + { + "description": "The second floor is directly above the cake.", + "timestamps": { + "start": "00:02:09", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:02:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Up-right of the cake.", + "B": "Directly below the cake.", + "C": "The data cable in front of you was down-left of the cake.", + "D": "The data cable in front of you was directly to the left of the cake, not below it.", + "E": "The data cable in front of you was down-right of the cake, positioned lower but on the other side." + }, + "answer_label": "C", + "answer_text": "The data cable in front of you was down-left of the cake.", + "q_id": "egolife_A1_034_4_T2", + "correct_answer": "The data cable in front of you was down-left of the cake.", + "original_options": { + "A": "Up-right of the cake.", + "B": "Directly below the cake.", + "C": "The data cable in front of you was down-left of the cake.", + "D": "The data cable in front of you was directly to the left of the cake, not below it.", + "E": "The data cable in front of you was down-right of the cake, positioned lower but on the other side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_4_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my data cable relative to the cake?", + "answer": "The data cable in front of you was directly above the cake.", + "question_chinese": "我的数据线相对于蛋糕的位置是什么？", + "answer_chinese": "你面前的数据线相对于蛋糕的正上方", + "evidence": [ + { + "description": "The data cable is directly above the sofa.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:11" + } + }, + { + "description": "The sofa is on the second floor.", + "timestamps": { + "start": "00:02:11", + "end": "00:02:12" + } + }, + { + "description": "The second floor is directly above the cake.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:02:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly below the cake.", + "B": "Down-left of the cake.", + "C": "The data cable in front of you was directly to the right of the cake rather than above it.", + "D": "The data cable in front of you was slightly above-left of the cake, not centered directly over it.", + "E": "The data cable in front of you was directly above the cake." + }, + "answer_label": "E", + "answer_text": "The data cable in front of you was directly above the cake.", + "q_id": "egolife_A1_034_4_T1", + "correct_answer": "The data cable in front of you was directly above the cake.", + "original_options": { + "A": "Directly below the cake.", + "B": "Down-left of the cake.", + "C": "The data cable in front of you was directly to the right of the cake rather than above it.", + "D": "The data cable in front of you was slightly above-left of the cake, not centered directly over it.", + "E": "The data cable in front of you was directly above the cake." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_2_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cakes do I see in the plate right now?", + "answer": "There were three cakes in front of you.", + "question_chinese": "我看见现在盘子里有几个蛋糕？", + "answer_chinese": "你面前的有三个蛋糕", + "evidence": [ + { + "description": "There are three cakes on the plate now.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One cake.", + "B": "Two cakes.", + "C": "There were four cakes in front of you on the plate when you looked at it.", + "D": "There were five cakes in front of you on the plate, more than just a small group.", + "E": "There were three cakes in front of you." + }, + "answer_label": "E", + "answer_text": "There were three cakes in front of you.", + "q_id": "egolife_A1_034_2_T1", + "correct_answer": "There were three cakes in front of you.", + "original_options": { + "A": "One cake.", + "B": "Two cakes.", + "C": "There were four cakes in front of you on the plate when you looked at it.", + "D": "There were five cakes in front of you on the plate, more than just a small group.", + "E": "There were three cakes in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_2_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cakes do I see in the plate right now?", + "answer": "There was one cake in front of you.", + "question_chinese": "我看见现在盘子里有几个蛋糕？", + "answer_chinese": "你面前的有一个蛋糕。", + "evidence": [ + { + "description": "There are three cakes on the plate now.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There was one cake in front of you.", + "B": "Zero cakes.", + "C": "Two cakes.", + "D": "There were two cakes in front of you on the plate, side by side.", + "E": "There were three cakes in front of you on the plate, arranged together." + }, + "answer_label": "A", + "answer_text": "There was one cake in front of you.", + "q_id": "egolife_A1_034_2_T2", + "correct_answer": "There was one cake in front of you.", + "original_options": { + "A": "There was one cake in front of you.", + "B": "Zero cakes.", + "C": "Two cakes.", + "D": "There were two cakes in front of you on the plate, side by side.", + "E": "There were three cakes in front of you on the plate, arranged together." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_2_T3", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cakes do I see in the plate right now?", + "answer": "There was one cake in front of you.", + "question_chinese": "我看见现在盘子里有几个蛋糕？", + "answer_chinese": "你面前的有一个蛋糕", + "evidence": [ + { + "description": "There is a cake on the plate now.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:38" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero cakes.", + "B": "There was one cake in front of you.", + "C": "Two cakes.", + "D": "There were three cakes in front of you on the plate, clustered near the center.", + "E": "There were four cakes in front of you on the plate, filling most of it." + }, + "answer_label": "B", + "answer_text": "There was one cake in front of you.", + "q_id": "egolife_A1_034_2_T3", + "correct_answer": "There was one cake in front of you.", + "original_options": { + "A": "Zero cakes.", + "B": "There was one cake in front of you.", + "C": "Two cakes.", + "D": "There were three cakes in front of you on the plate, clustered near the center.", + "E": "There were four cakes in front of you on the plate, filling most of it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_3_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cakes are in the plate in front of me?", + "answer": "There were three cakes on the plate in front of you.", + "question_chinese": "我面前的盘子里有几个蛋糕？", + "answer_chinese": "你面前的盘子里有三个蛋糕。", + "evidence": [ + { + "description": "There is three cake on the plate now.", + "timestamps": { + "start": "00:01:13", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:01:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One cake.", + "B": "There were three cakes on the plate in front of you.", + "C": "Two cakes.", + "D": "There were four cakes on the plate in front of you, with one more than expected.", + "E": "There were five cakes on the plate in front of you, spread across the plate." + }, + "answer_label": "B", + "answer_text": "There were three cakes on the plate in front of you.", + "q_id": "egolife_A1_034_3_T2", + "correct_answer": "There were three cakes on the plate in front of you.", + "original_options": { + "A": "One cake.", + "B": "There were three cakes on the plate in front of you.", + "C": "Two cakes.", + "D": "There were four cakes on the plate in front of you, with one more than expected.", + "E": "There were five cakes on the plate in front of you, spread across the plate." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_3_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cakes are in the plate in front of me?", + "answer": "There was one cake on the plate in front of you.", + "question_chinese": "我面前的盘子里有几个蛋糕？", + "answer_chinese": "你面前的盘子里有一个蛋糕。", + "evidence": [ + { + "description": "There is a cake on the plate now.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:38" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There was one cake on the plate in front of you.", + "B": "Zero cakes.", + "C": "Two cakes.", + "D": "There were three cakes on the plate in front of you, grouped close together.", + "E": "There were four cakes on the plate in front of you, taking up the plate." + }, + "answer_label": "A", + "answer_text": "There was one cake on the plate in front of you.", + "q_id": "egolife_A1_034_3_T1", + "correct_answer": "There was one cake on the plate in front of you.", + "original_options": { + "A": "There was one cake on the plate in front of you.", + "B": "Zero cakes.", + "C": "Two cakes.", + "D": "There were three cakes on the plate in front of you, grouped close together.", + "E": "There were four cakes on the plate in front of you, taking up the plate." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_5", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the data cable relative to the cake?", + "answer": "Directly above.", + "question_chinese": "数据线相对于蛋糕的位置是什么", + "answer_chinese": "正上方", + "evidence": [ + { + "description": "The data cable is directly above the sofa.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:15" + } + }, + { + "description": "The sofa is on the second floor.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:15" + } + }, + { + "description": "The second floor is directly above the cake.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:02:19", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "To the left.", + "B": "To the right.", + "C": "Directly above.", + "D": "Right below the cake, aligned under it rather than above it in the scene.", + "E": "Right in front of the cake, closer to you than the cake is on the table." + }, + "answer_label": "C", + "answer_text": "Directly above.", + "q_id": "egolife_A1_034_5", + "correct_answer": "Directly above.", + "original_options": { + "A": "To the left.", + "B": "To the right.", + "C": "Directly above.", + "D": "Right below the cake, aligned under it rather than above it in the scene.", + "E": "Right in front of the cake, closer to you than the cake is on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_6", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the data cable in relation to the cake?", + "answer": "Left front.", + "question_chinese": "数据线相对于蛋糕的位置是什么", + "answer_chinese": "左前方", + "evidence": [ + { + "description": "The data cable is plugged into the power bank.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:20" + } + }, + { + "description": "The power bank is charging on the cabinet.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:20" + } + }, + { + "description": "The cabinet is to the front-left of the cake.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:20" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Right front.", + "B": "Behind the cake.", + "C": "Directly below the cake, not up toward the left side of the frame.", + "D": "Left front.", + "E": "To the right of the cake, offset sideways instead of diagonally to the left." + }, + "answer_label": "D", + "answer_text": "Left front.", + "q_id": "egolife_A1_034_6", + "correct_answer": "Left front.", + "original_options": { + "A": "Right front.", + "B": "Behind the cake.", + "C": "Directly below the cake, not up toward the left side of the frame.", + "D": "Left front.", + "E": "To the right of the cake, offset sideways instead of diagonally to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_3", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cakes are there on the plate now?", + "answer": "Three.", + "question_chinese": "现在盘子里有几个蛋糕", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There are three cakes on the plate now.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Six cakes are on the plate now, more than the three shown earlier.", + "E": "Seven cakes are on the plate now, covering most of the plate surface." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "egolife_A1_034_3", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Six cakes are on the plate now, more than the three shown earlier.", + "E": "Seven cakes are on the plate now, covering most of the plate surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_034", + "video_length": "00:12:00", + "qid": "egolife_A1_034_4", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cakes are on the plate now?", + "answer": "One.", + "question_chinese": "现在盘子里有几个蛋糕", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There is a cake on the plate now.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:38" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four cakes are on the plate now, not just a single cake.", + "D": "Five cakes are on the plate now, with several pieces visible together.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_034_4", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four cakes are on the plate now, not just a single cake.", + "D": "Five cakes are on the plate now, with several pieces visible together.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + } +] diff --git a/QAs/egolife_A1_035-mcq.json b/QAs/egolife_A1_035-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b55e334cefc0a72c1ecda2c42293d26a1ad19f26 --- /dev/null +++ b/QAs/egolife_A1_035-mcq.json @@ -0,0 +1,1003 @@ +[ + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where were the glasses just placed?", + "answer": "Put it on the dining table.", + "question_chinese": "刚刚眼镜的被放在什么地方", + "answer_chinese": "放在餐桌上", + "evidence": [ + { + "description": "The glasses are on the dining table.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:04:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put it on the kitchen counter.", + "B": "Put it on the dining table.", + "C": "Put it on the kitchen table.", + "D": "Put it on the living room shelf instead of the dining table where you were just standing.", + "E": "Put it on the kitchen shelf with the plant, not on the dining table." + }, + "answer_label": "B", + "answer_text": "Put it on the dining table.", + "q_id": "egolife_A1_035_7", + "correct_answer": "Put it on the dining table.", + "original_options": { + "A": "Put it on the kitchen counter.", + "B": "Put it on the dining table.", + "C": "Put it on the kitchen table.", + "D": "Put it on the living room shelf instead of the dining table where you were just standing.", + "E": "Put it on the kitchen shelf with the plant, not on the dining table." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 14 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where were the glasses just placed?", + "answer": "Put it on the computer desk.", + "question_chinese": "刚刚眼镜的被放在什么地方", + "answer_chinese": "放在电脑桌上", + "evidence": [ + { + "description": "I put my glasses on the computer desk.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:50" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Put it on the computer desk.", + "B": "Put it beside the computer desk.", + "C": "Put it behind the computer desk.", + "D": "Put it in front of the computer desk rather than placing it on the desktop surface.", + "E": "Put it in the drawer of the computer desk, not on top of it." + }, + "answer_label": "A", + "answer_text": "Put it on the computer desk.", + "q_id": "egolife_A1_035_8", + "correct_answer": "Put it on the computer desk.", + "original_options": { + "A": "Put it on the computer desk.", + "B": "Put it beside the computer desk.", + "C": "Put it behind the computer desk.", + "D": "Put it in front of the computer desk rather than placing it on the desktop surface.", + "E": "Put it in the drawer of the computer desk, not on top of it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_10", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the vase if I stretch out my hand now?", + "answer": "No.", + "question_chinese": "我现在伸手可以碰到花瓶吗", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "I'm sitting on a restaurant chair.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:45" + } + }, + { + "description": "Thirty centimeters away from the vase", + "timestamps": { + "start": "00:00:30", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:00:53", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_035_10", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_9", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the vase if I stretch out my hand now?", + "answer": "Yes.", + "question_chinese": "我现在伸手可以碰到花瓶吗", + "answer_chinese": "可以", + "evidence": [ + { + "description": "I am now standing in front of the vase.", + "timestamps": { + "start": "00:04:50", + "end": "00:04:55" + } + }, + { + "description": "Five centimeters away from the vase", + "timestamps": { + "start": "00:04:50", + "end": "00:04:55" + } + } + ], + "question_timestamps": "00:04:59", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_035_9", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_1_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to be able to reach the cake with my hand right now, where should I put the cake?", + "answer": "Put it on a chair in the dining room.", + "question_chinese": "我现在想伸手拿到蛋糕，我应该把蛋糕放在哪里？", + "answer_chinese": "放在餐厅的椅子上", + "evidence": [ + { + "description": "I am standing next to the chair in the restaurant.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:19" + } + }, + { + "description": "3 centimeters away from the chair", + "timestamps": { + "start": "00:03:10", + "end": "00:03:19" + } + } + ], + "question_timestamps": "00:03:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it on the dining table.", + "B": "Put it on the bedside table.", + "C": "Put it on a chair in the dining room.", + "D": "You can put the cake on the stovetop to your right so it stays within easy reach.", + "E": "Go straight for three meters, then turn right and go two meters to place it there." + }, + "answer_label": "C", + "answer_text": "Put it on a chair in the dining room.", + "q_id": "egolife_A1_035_1_T1", + "correct_answer": "Put it on a chair in the dining room.", + "original_options": { + "A": "Put it on the dining table.", + "B": "Put it on the bedside table.", + "C": "Put it on a chair in the dining room.", + "D": "You can put the cake on the stovetop to your right so it stays within easy reach.", + "E": "Go straight for three meters, then turn right and go two meters to place it there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_1_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to be able to reach the cake with my hand right now, where should I put the cake?", + "answer": "You can put the cake on the stovetop to your right.", + "question_chinese": "我现在想伸手拿到蛋糕，我应该把蛋糕放在哪里？", + "answer_chinese": "你可以把蛋糕放在右边灶台上", + "evidence": [ + { + "description": "I am standing next to the chair in the restaurant.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:14" + } + }, + { + "description": "3 centimeters away from the chair", + "timestamps": { + "start": "00:03:15", + "end": "00:03:19" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it on the left stovetop.", + "B": "You can put the cake on the stovetop to your right.", + "C": "Put it on the dining table.", + "D": "You can put the cake on the stovetop behind you instead of placing it to your right.", + "E": "You can put the cake on the counter in front of you, not on the stovetop to your right." + }, + "answer_label": "B", + "answer_text": "You can put the cake on the stovetop to your right.", + "q_id": "egolife_A1_035_1_T2", + "correct_answer": "You can put the cake on the stovetop to your right.", + "original_options": { + "A": "Put it on the left stovetop.", + "B": "You can put the cake on the stovetop to your right.", + "C": "Put it on the dining table.", + "D": "You can put the cake on the stovetop behind you instead of placing it to your right.", + "E": "You can put the cake on the counter in front of you, not on the stovetop to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_2_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to be able to reach the cake with my hand right now, where should I put the cake?", + "answer": "Put it on the dining table.", + "question_chinese": "我现在想伸手拿到蛋糕，我应该把蛋糕放在哪里？", + "answer_chinese": "放在餐桌上", + "evidence": [ + { + "description": "I am standing at the bottom-left corner of the dining table.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:10" + } + }, + { + "description": "Five centimeters from the dining table", + "timestamps": { + "start": "00:05:00", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:05:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it on the bedside table.", + "B": "Put it on a dining chair.", + "C": "You can put the cake on the stovetop to your right so it’s closer than the dining table.", + "D": "Go straight for three meters, then turn right and go two meters before putting it down.", + "E": "Put it on the dining table." + }, + "answer_label": "E", + "answer_text": "Put it on the dining table.", + "q_id": "egolife_A1_035_2_T1", + "correct_answer": "Put it on the dining table.", + "original_options": { + "A": "Put it on the bedside table.", + "B": "Put it on a dining chair.", + "C": "You can put the cake on the stovetop to your right so it’s closer than the dining table.", + "D": "Go straight for three meters, then turn right and go two meters before putting it down.", + "E": "Put it on the dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_2_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to be able to reach the cake with my hand right now, where should I put the cake?", + "answer": "You can put it on the bedside table.", + "question_chinese": "我现在想伸手拿到蛋糕，我应该把蛋糕放在哪里？", + "answer_chinese": "你可以放在床头柜上。", + "evidence": [ + { + "description": "I am standing at the bottom-left corner of the dining table.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:04" + } + }, + { + "description": "Five centimeters from the dining table", + "timestamps": { + "start": "00:05:05", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Put it on the dining table.", + "B": "You can put it on the bedside table.", + "C": "Put it on a dining chair.", + "D": "You can put the cake on the stovetop to your right instead of placing it on the bedside table.", + "E": "Go straight for three meters, then turn right and go two meters to set it down there." + }, + "answer_label": "B", + "answer_text": "You can put it on the bedside table.", + "q_id": "egolife_A1_035_2_T2", + "correct_answer": "You can put it on the bedside table.", + "original_options": { + "A": "Put it on the dining table.", + "B": "You can put it on the bedside table.", + "C": "Put it on a dining chair.", + "D": "You can put the cake on the stovetop to your right instead of placing it on the bedside table.", + "E": "Go straight for three meters, then turn right and go two meters to set it down there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_4_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people in front of me are wearing glasses right now?", + "answer": "All five people in front of you were wearing glasses.", + "question_chinese": "我面前现在有几个人带着眼镜？", + "answer_chinese": "你面前的五个人带着眼镜", + "evidence": [ + { + "description": "There are five people wearing glasses.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four people were wearing glasses.", + "B": "Six people were wearing glasses.", + "C": "All five people in front of you were wearing glasses.", + "D": "Only three people in front of you were wearing glasses, while the others were not.", + "E": "Seven people in front of you were wearing glasses, with none of them without glasses." + }, + "answer_label": "C", + "answer_text": "All five people in front of you were wearing glasses.", + "q_id": "egolife_A1_035_4_T1", + "correct_answer": "All five people in front of you were wearing glasses.", + "original_options": { + "A": "Four people were wearing glasses.", + "B": "Six people were wearing glasses.", + "C": "All five people in front of you were wearing glasses.", + "D": "Only three people in front of you were wearing glasses, while the others were not.", + "E": "Seven people in front of you were wearing glasses, with none of them without glasses." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_3_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people in front of me are wearing glasses right now?", + "answer": "There were five people wearing glasses in front of you.", + "question_chinese": "我面前现在有几个人带着眼镜？", + "answer_chinese": "你面前有五个人戴眼镜", + "evidence": [ + { + "description": "There are five people wearing glasses.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four people were wearing glasses.", + "B": "There were five people wearing glasses in front of you.", + "C": "Six people were wearing glasses.", + "D": "There were three people wearing glasses in front of you, and the rest were not wearing any.", + "E": "There were seven people wearing glasses in front of you, so no one in front lacked glasses." + }, + "answer_label": "B", + "answer_text": "There were five people wearing glasses in front of you.", + "q_id": "egolife_A1_035_3_T1", + "correct_answer": "There were five people wearing glasses in front of you.", + "original_options": { + "A": "Four people were wearing glasses.", + "B": "There were five people wearing glasses in front of you.", + "C": "Six people were wearing glasses.", + "D": "There were three people wearing glasses in front of you, and the rest were not wearing any.", + "E": "There were seven people wearing glasses in front of you, so no one in front lacked glasses." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_3_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people in front of me are wearing glasses right now?", + "answer": "There was only one person wearing glasses in front of you.", + "question_chinese": "我面前现在有几个人带着眼镜？", + "answer_chinese": "你面前只有一个人带眼镜。", + "evidence": [ + { + "description": "Only one person was wearing glasses.", + "timestamps": { + "start": "00:04:12", + "end": "00:04:13" + } + } + ], + "question_timestamps": "00:04:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero people were wearing glasses.", + "B": "Two people were wearing glasses.", + "C": "There was only one person wearing glasses in front of you.", + "D": "There were three people wearing glasses in front of you, not just a single person.", + "E": "There were four people wearing glasses in front of you, with several others not wearing glasses." + }, + "answer_label": "C", + "answer_text": "There was only one person wearing glasses in front of you.", + "q_id": "egolife_A1_035_3_T2", + "correct_answer": "There was only one person wearing glasses in front of you.", + "original_options": { + "A": "Zero people were wearing glasses.", + "B": "Two people were wearing glasses.", + "C": "There was only one person wearing glasses in front of you.", + "D": "There were three people wearing glasses in front of you, not just a single person.", + "E": "There were four people wearing glasses in front of you, with several others not wearing glasses." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_4_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people in front of me are wearing glasses right now?", + "answer": "The one person in front of you was wearing glasses.", + "question_chinese": "我面前现在有几个人带着眼镜？", + "answer_chinese": "你面前的一个人带着眼镜", + "evidence": [ + { + "description": "Now four people have taken off their glasses.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:45" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "No one wore glasses.", + "B": "Two people wore glasses.", + "C": "Three people in front of you were wearing glasses, rather than just a single person.", + "D": "The one person in front of you was wearing glasses.", + "E": "Four people in front of you were wearing glasses, not just one person standing there." + }, + "answer_label": "D", + "answer_text": "The one person in front of you was wearing glasses.", + "q_id": "egolife_A1_035_4_T2", + "correct_answer": "The one person in front of you was wearing glasses.", + "original_options": { + "A": "No one wore glasses.", + "B": "Two people wore glasses.", + "C": "Three people in front of you were wearing glasses, rather than just a single person.", + "D": "The one person in front of you was wearing glasses.", + "E": "Four people in front of you were wearing glasses, not just one person standing there." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_4_T3", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people in front of me are wearing glasses right now?", + "answer": "The one person in front of you was wearing glasses.", + "question_chinese": "我面前现在有几个人带着眼镜？", + "answer_chinese": "你面前的一个人带着眼镜", + "evidence": [ + { + "description": "Now four people have taken off their glasses.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:45" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "No one wore glasses.", + "B": "Two people wore glasses.", + "C": "Three people in front of you were wearing glasses, instead of only one person.", + "D": "Four people in front of you were wearing glasses, rather than a single person.", + "E": "The one person in front of you was wearing glasses." + }, + "answer_label": "E", + "answer_text": "The one person in front of you was wearing glasses.", + "q_id": "egolife_A1_035_4_T3", + "correct_answer": "The one person in front of you was wearing glasses.", + "original_options": { + "A": "No one wore glasses.", + "B": "Two people wore glasses.", + "C": "Three people in front of you were wearing glasses, instead of only one person.", + "D": "Four people in front of you were wearing glasses, rather than a single person.", + "E": "The one person in front of you was wearing glasses." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_3", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are currently wearing glasses?", + "answer": "Five.", + "question_chinese": "现在有几个人带着眼镜", + "answer_chinese": "五个", + "evidence": [ + { + "description": "There are five people wearing glasses.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "Three.", + "D": "Only one person is currently wearing glasses, not a group of several people.", + "E": "Two people are currently wearing glasses, with the others not wearing any." + }, + "answer_label": "B", + "answer_text": "Five.", + "q_id": "egolife_A1_035_3", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "Three.", + "D": "Only one person is currently wearing glasses, not a group of several people.", + "E": "Two people are currently wearing glasses, with the others not wearing any." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_4", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are wearing glasses now?", + "answer": "One.", + "question_chinese": "现在有几个人带着眼镜", + "answer_chinese": "一个", + "evidence": [ + { + "description": "Now four people have taken off their glasses.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:45" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four people are wearing glasses now, not just a single person present.", + "D": "Five people are wearing glasses now, with everyone else not wearing any.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_035_4", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four people are wearing glasses now, not just a single person present.", + "D": "Five people are wearing glasses now, with everyone else not wearing any.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_5", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the position of the glasses in relation to the vase?", + "answer": "Directly in front of the vase.", + "question_chinese": "眼镜相对于花瓶的位置在哪里", + "answer_chinese": "在花瓶的正前方", + "evidence": [ + { + "description": "The glasses are in my hand.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:00" + } + }, + { + "description": "I am standing right in front of the dining table.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:00" + } + }, + { + "description": "The vase is directly above the dining table.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:05:05", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "To the left of the vase.", + "B": "To the right of the vase.", + "C": "Behind the vase on the table, not positioned in front of it.", + "D": "Directly in front of the vase.", + "E": "Above the vase on a shelf, rather than sitting in front of it." + }, + "answer_label": "D", + "answer_text": "Directly in front of the vase.", + "q_id": "egolife_A1_035_5", + "correct_answer": "Directly in front of the vase.", + "original_options": { + "A": "To the left of the vase.", + "B": "To the right of the vase.", + "C": "Behind the vase on the table, not positioned in front of it.", + "D": "Directly in front of the vase.", + "E": "Above the vase on a shelf, rather than sitting in front of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_035", + "video_length": "00:06:19", + "qid": "egolife_A1_035_6", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the position of the glasses in relation to the vase?", + "answer": "Right front.", + "question_chinese": "眼镜相对于花瓶的位置在哪里", + "answer_chinese": "右前方", + "evidence": [ + { + "description": "The glasses are on the computer desk.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:07" + } + }, + { + "description": "The computer desk is inside the room.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:07" + } + }, + { + "description": "The room is in front of and to the right of the vase.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Left front.", + "B": "Right back.", + "C": "They are positioned to the left of the vase, slightly behind it from your viewpoint.", + "D": "The glasses sit behind the vase, closer to the center rather than in front of it.", + "E": "Right front." + }, + "answer_label": "E", + "answer_text": "Right front.", + "q_id": "egolife_A1_035_6", + "correct_answer": "Right front.", + "original_options": { + "A": "Left front.", + "B": "Right back.", + "C": "They are positioned to the left of the vase, slightly behind it from your viewpoint.", + "D": "The glasses sit behind the vase, closer to the center rather than in front of it.", + "E": "Right front." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D2_001-mcq.json b/QAs/egolife_A1_D2_001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..39c24863b07167b1725532bc5df6a2d2321569b6 --- /dev/null +++ b/QAs/egolife_A1_D2_001-mcq.json @@ -0,0 +1,862 @@ +[ + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_4", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to use the tablet now, where should I place it?", + "answer": "Put the tablet on the microwave oven.", + "question_chinese": "如果我现在想使用平板我应该把平板放在哪里", + "answer_chinese": "把平板放在微波炉上", + "evidence": [ + { + "description": "I am standing in the kitchen.", + "timestamps": { + "start": "00:05:45", + "end": "00:05:50" + } + }, + { + "description": "Five centimeters away from the microwave oven", + "timestamps": { + "start": "00:05:45", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:05:56", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Put the tablet on the coffee table.", + "B": "Put the tablet on the microwave oven.", + "C": "Put the tablet on the dining table.", + "D": "Place the tablet on the kitchen counter beside the sink so it’s within easy reach.", + "E": "Set the tablet on the shelf above the fridge to keep it out of the way." + }, + "answer_label": "B", + "answer_text": "Put the tablet on the microwave oven.", + "q_id": "egolife_A1_D2_001_4", + "correct_answer": "Put the tablet on the microwave oven.", + "original_options": { + "A": "Put the tablet on the coffee table.", + "B": "Put the tablet on the microwave oven.", + "C": "Put the tablet on the dining table.", + "D": "Place the tablet on the kitchen counter beside the sink so it’s within easy reach.", + "E": "Set the tablet on the shelf above the fridge to keep it out of the way." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_3", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to use the tablet now, where should I place it?", + "answer": "Put it on the dining table.", + "question_chinese": "如果我现在想使用平板我应该把平板放在哪里", + "answer_chinese": "放在餐桌上", + "evidence": [ + { + "description": "I am sitting at the dining table now.", + "timestamps": { + "start": "00:11:00", + "end": "00:11:05" + } + }, + { + "description": "Five centimeters away from the dining table", + "timestamps": { + "start": "00:11:00", + "end": "00:11:05" + } + } + ], + "question_timestamps": "00:11:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Put it on your lap.", + "B": "Put it beside the dining table.", + "C": "Place it in front of the dining table, not on top, so you can grab it quickly.", + "D": "Set it on your computer desktop instead of the dining table for easier viewing.", + "E": "Put it on the dining table." + }, + "answer_label": "E", + "answer_text": "Put it on the dining table.", + "q_id": "egolife_A1_D2_001_3", + "correct_answer": "Put it on the dining table.", + "original_options": { + "A": "Put it on your lap.", + "B": "Put it beside the dining table.", + "C": "Place it in front of the dining table, not on top, so you can grab it quickly.", + "D": "Set it on your computer desktop instead of the dining table for easier viewing.", + "E": "Put it on the dining table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_6_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the kettle relative to the black bag?", + "answer": "Left front.", + "question_chinese": "烧水壶相对于黑色包在什么位置？", + "answer_chinese": "左前方", + "evidence": [ + { + "description": "The black bag is placed above the dining table.", + "timestamps": { + "start": "00:01:50", + "end": "00:02:25" + } + }, + { + "description": "The kettle is above the cabinet.", + "timestamps": { + "start": "00:01:50", + "end": "00:02:25" + } + }, + { + "description": "The cabinet is in front of and to the left of the dining table.", + "timestamps": { + "start": "00:01:50", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right front.", + "B": "Right back.", + "C": "The kettle is behind the black bag, slightly to the left from where you are standing.", + "D": "Left front.", + "E": "It sits to the lower right of the black bag rather than in front of it." + }, + "answer_label": "D", + "answer_text": "Left front.", + "q_id": "egolife_A1_D2_001_6_T1", + "correct_answer": "Left front.", + "original_options": { + "A": "Right front.", + "B": "Right back.", + "C": "The kettle is behind the black bag, slightly to the left from where you are standing.", + "D": "Left front.", + "E": "It sits to the lower right of the black bag rather than in front of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_7_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the kettle relative to the black bag?", + "answer": "To the lower left of the black bag.", + "question_chinese": "烧水壶相对于黑色包在什么位置？", + "answer_chinese": "在黑色包的左下方", + "evidence": [ + { + "description": "The black bag is directly above the right side of the dining table.", + "timestamps": { + "start": "00:05:10", + "end": "00:05:29" + } + }, + { + "description": "The kettle is under the dining table.", + "timestamps": { + "start": "00:05:10", + "end": "00:05:29" + } + }, + { + "description": "The kettle is placed near the wall, and the left side of the dining table is against the wall.", + "timestamps": { + "start": "00:05:10", + "end": "00:05:29" + } + } + ], + "question_timestamps": "00:05:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To the lower left of the black bag.", + "B": "To the lower right of the black bag.", + "C": "Directly behind the black bag.", + "D": "The kettle is positioned directly in front of the black bag, not down to the left.", + "E": "The kettle is set a short distance away from the black bag, toward the center of the table." + }, + "answer_label": "A", + "answer_text": "To the lower left of the black bag.", + "q_id": "egolife_A1_D2_001_7_T1", + "correct_answer": "To the lower left of the black bag.", + "original_options": { + "A": "To the lower left of the black bag.", + "B": "To the lower right of the black bag.", + "C": "Directly behind the black bag.", + "D": "The kettle is positioned directly in front of the black bag, not down to the left.", + "E": "The kettle is set a short distance away from the black bag, toward the center of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 5, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_3_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many black bags are on the dining table right now?", + "answer": "Four, on the dining table on the right side.", + "question_chinese": "餐桌上现在有几个黑色的包？", + "answer_chinese": "四个，在你右侧的餐桌上", + "evidence": [ + { + "description": "There are four black bags on the table now.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:03:02", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, on the right side.", + "B": "Six, on the right side.", + "C": "Four, on the dining table on the right side.", + "D": "Three, all grouped together on the dining table on the right side.", + "E": "Five, spread out across the dining table but mostly on the right side." + }, + "answer_label": "C", + "answer_text": "Four, on the dining table on the right side.", + "q_id": "egolife_A1_D2_001_3_T1", + "correct_answer": "Four, on the dining table on the right side.", + "original_options": { + "A": "Two, on the right side.", + "B": "Six, on the right side.", + "C": "Four, on the dining table on the right side.", + "D": "Three, all grouped together on the dining table on the right side.", + "E": "Five, spread out across the dining table but mostly on the right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_5", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many black bags are there on the dining table now?", + "answer": "Four.", + "question_chinese": "餐桌上现在有几个黑色的包", + "answer_chinese": "四个", + "evidence": [ + { + "description": "There are four black bags on the table now.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three black bags are on the dining table right now, not four in total.", + "D": "Five black bags are on the dining table now, which is one more than four.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "egolife_A1_D2_001_5", + "correct_answer": "Four.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three black bags are on the dining table right now, not four in total.", + "D": "Five black bags are on the dining table now, which is one more than four.", + "E": "Four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_3_T3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many black bags are on the dining table right now?", + "answer": "Zero—there are no black bags left on the dining table to your front-left.", + "question_chinese": "餐桌上现在有几个黑色的包？", + "answer_chinese": "零个，你左前侧的餐桌上没有黑色的包了", + "evidence": [ + { + "description": "I put all the black bags on the bed, so there are no black bags on the dining table now.", + "timestamps": { + "start": "00:06:50", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One—there is one black bag left.", + "B": "Two—there are two black bags left.", + "C": "Three—there are three black bags left on the dining table to your front-left.", + "D": "Four—there are four black bags left on the dining table to your front-left.", + "E": "Zero—there are no black bags left on the dining table to your front-left." + }, + "answer_label": "E", + "answer_text": "Zero—there are no black bags left on the dining table to your front-left.", + "q_id": "egolife_A1_D2_001_3_T3", + "correct_answer": "Zero—there are no black bags left on the dining table to your front-left.", + "original_options": { + "A": "One—there is one black bag left.", + "B": "Two—there are two black bags left.", + "C": "Three—there are three black bags left on the dining table to your front-left.", + "D": "Four—there are four black bags left on the dining table to your front-left.", + "E": "Zero—there are no black bags left on the dining table to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many black bags are on the dining table now?", + "answer": "Zero.", + "question_chinese": "餐桌上现在有几个黑色的包", + "answer_chinese": "零个", + "evidence": [ + { + "description": "I put all the black bags on the bed, so there are no black bags on the dining table now.", + "timestamps": { + "start": "00:06:50", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "One black bag.", + "C": "Two black bags.", + "D": "Three black bags are on the dining table now, sitting near the center edge.", + "E": "Four black bags are on the dining table now, grouped together on one side." + }, + "answer_label": "A", + "answer_text": "Zero.", + "q_id": "egolife_A1_D2_001_6", + "correct_answer": "Zero.", + "original_options": { + "A": "Zero.", + "B": "One black bag.", + "C": "Two black bags.", + "D": "Three black bags are on the dining table now, sitting near the center edge.", + "E": "Four black bags are on the dining table now, grouped together on one side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_1_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the black bag from the refrigerator?", + "answer": "Forty centimeters.", + "question_chinese": "黑色包距离冰箱的距离是多少？", + "answer_chinese": "四十厘米", + "evidence": [ + { + "description": "The black bag is placed on the dining table.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + }, + { + "description": "The dining table is forty centimeters away from the refrigerator.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Twenty centimeters away.", + "B": "Eighty centimeters away.", + "C": "It is about one meter from the refrigerator, close enough to reach without moving much.", + "D": "Forty centimeters.", + "E": "It sits roughly three meters from the refrigerator, requiring you to take a few steps." + }, + "answer_label": "D", + "answer_text": "Forty centimeters.", + "q_id": "egolife_A1_D2_001_1_T1", + "correct_answer": "Forty centimeters.", + "original_options": { + "A": "Twenty centimeters away.", + "B": "Eighty centimeters away.", + "C": "It is about one meter from the refrigerator, close enough to reach without moving much.", + "D": "Forty centimeters.", + "E": "It sits roughly three meters from the refrigerator, requiring you to take a few steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the distance from the black bag to the refrigerator?", + "answer": "Forty centimeters.", + "question_chinese": "黑色包距离冰箱的距离是多少", + "answer_chinese": "四十厘米", + "evidence": [ + { + "description": "The black bag is placed on the dining table.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + }, + { + "description": "The dining table is forty centimeters away from the refrigerator.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Thirty centimeters.", + "B": "Sixty centimeters.", + "C": "Forty centimeters.", + "D": "It is about ninety centimeters from the refrigerator, less than a full meter away.", + "E": "The black bag is roughly one hundred twenty centimeters from the refrigerator, about an arm-and-a-half length." + }, + "answer_label": "C", + "answer_text": "Forty centimeters.", + "q_id": "egolife_A1_D2_001_1", + "correct_answer": "Forty centimeters.", + "original_options": { + "A": "Thirty centimeters.", + "B": "Sixty centimeters.", + "C": "Forty centimeters.", + "D": "It is about ninety centimeters from the refrigerator, less than a full meter away.", + "E": "The black bag is roughly one hundred twenty centimeters from the refrigerator, about an arm-and-a-half length." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_1_T3", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the black bag from the refrigerator?", + "answer": "Eighty centimeters.", + "question_chinese": "黑色包距离冰箱的距离是多少？", + "answer_chinese": "八十厘米", + "evidence": [ + { + "description": "The black bag is on the bed.", + "timestamps": { + "start": "00:06:50", + "end": "00:07:10" + } + }, + { + "description": "The bed is eighty centimeters away from the refrigerator.", + "timestamps": { + "start": "00:06:50", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Forty centimeters.", + "B": "One hundred twenty centimeters.", + "C": "It is about three meters from the refrigerator, so you would need to walk a few steps.", + "D": "The black bag is roughly ten meters from the refrigerator, far across the room from it.", + "E": "Eighty centimeters." + }, + "answer_label": "E", + "answer_text": "Eighty centimeters.", + "q_id": "egolife_A1_D2_001_1_T3", + "correct_answer": "Eighty centimeters.", + "original_options": { + "A": "Forty centimeters.", + "B": "One hundred twenty centimeters.", + "C": "It is about three meters from the refrigerator, so you would need to walk a few steps.", + "D": "The black bag is roughly ten meters from the refrigerator, far across the room from it.", + "E": "Eighty centimeters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the distance from the black bag to the refrigerator?", + "answer": "Eighty centimeters.", + "question_chinese": "黑色包距离冰箱的距离是多少", + "answer_chinese": "八十厘米", + "evidence": [ + { + "description": "The black bag is on the bed.", + "timestamps": { + "start": "00:06:50", + "end": "00:07:10" + } + }, + { + "description": "The bed is eighty centimeters away from the refrigerator.", + "timestamps": { + "start": "00:06:50", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Eighty centimeters.", + "B": "Sixty centimeters.", + "C": "Ninety centimeters.", + "D": "The distance from the black bag to the refrigerator is about fifty centimeters, roughly half a meter.", + "E": "The black bag is around one hundred twenty centimeters from the refrigerator, a little over one meter away." + }, + "answer_label": "A", + "answer_text": "Eighty centimeters.", + "q_id": "egolife_A1_D2_001_2", + "correct_answer": "Eighty centimeters.", + "original_options": { + "A": "Eighty centimeters.", + "B": "Sixty centimeters.", + "C": "Ninety centimeters.", + "D": "The distance from the black bag to the refrigerator is about fifty centimeters, roughly half a meter.", + "E": "The black bag is around one hundred twenty centimeters from the refrigerator, a little over one meter away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_7", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the black bag now?", + "answer": "At the dining table.", + "question_chinese": "黑色的包现在在什么位置", + "answer_chinese": "在餐桌上", + "evidence": [ + { + "description": "Now at the dining table", + "timestamps": { + "start": "00:01:00", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:03:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the kitchen counter.", + "B": "Under the dining table.", + "C": "The black bag is beside the dining table now, leaning against the table’s side.", + "D": "At the dining table.", + "E": "It is in front of the dining table now, positioned on the floor near the table edge." + }, + "answer_label": "D", + "answer_text": "At the dining table.", + "q_id": "egolife_A1_D2_001_7", + "correct_answer": "At the dining table.", + "original_options": { + "A": "On the kitchen counter.", + "B": "Under the dining table.", + "C": "The black bag is beside the dining table now, leaning against the table’s side.", + "D": "At the dining table.", + "E": "It is in front of the dining table now, positioned on the floor near the table edge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_001", + "video_length": "00:11:35", + "qid": "egolife_A1_D2_001_8", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the black bag now?", + "answer": "In bed.", + "question_chinese": "黑色的包现在在什么位置", + "answer_chinese": "在床上", + "evidence": [ + { + "description": "I put the black bag on the bed.", + "timestamps": { + "start": "00:06:50", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:09:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the chair.", + "B": "In the closet.", + "C": "The black bag is on the table now, placed near the front edge within reach.", + "D": "It is on the shelf now, resting on an open spot rather than on the bed.", + "E": "In bed." + }, + "answer_label": "E", + "answer_text": "In bed.", + "q_id": "egolife_A1_D2_001_8", + "correct_answer": "In bed.", + "original_options": { + "A": "On the chair.", + "B": "In the closet.", + "C": "The black bag is on the table now, placed near the front edge within reach.", + "D": "It is on the shelf now, resting on an open spot rather than on the bed.", + "E": "In bed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D2_002-mcq.json b/QAs/egolife_A1_D2_002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9c9347a343f2d2793eac95ae43c1b27a42f3b45e --- /dev/null +++ b/QAs/egolife_A1_D2_002-mcq.json @@ -0,0 +1,642 @@ +[ + { + "video_uid": "egolife_A1_D2_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_002-mcq_fixed_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to use a computer now, which one should I use?", + "answer": "The computer on the sofa.", + "question_chinese": "如果我现在想使用电脑，我应该用哪个电脑", + "answer_chinese": "沙发上的电脑", + "evidence": [ + { + "description": "I am sitting on the sofa now.", + "timestamps": { + "start": "00:03:50", + "end": "00:03:57" + } + }, + { + "description": "The closest to the computer on the sofa", + "timestamps": { + "start": "00:03:50", + "end": "00:03:57" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The computer on the sofa.", + "B": "The computer behind the sofa.", + "C": "The computer to the left of the sofa.", + "D": "Use the laptop in the study you were updating recently, since that is the one set up for work.", + "E": "Use the computer on the desk beside the printer, the one you used last week for tasks." + }, + "answer_label": "A", + "answer_text": "The computer on the sofa.", + "q_id": "egolife_A1_D2_002-mcq_fixed_5", + "correct_answer": "The computer on the sofa.", + "original_options": { + "A": "The computer on the sofa.", + "B": "The computer behind the sofa.", + "C": "The computer to the left of the sofa.", + "D": "Use the laptop in the study you were updating recently, since that is the one set up for work.", + "E": "Use the computer on the desk beside the printer, the one you used last week for tasks." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_002_5", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to use a computer right now, which one should I use?", + "answer": "The computer on the floor in the corner.", + "question_chinese": "如果我现在想使用电脑，我应该用哪个电脑", + "answer_chinese": "角落地上的电脑", + "evidence": [ + { + "description": "I am standing in the corner now.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:10" + } + }, + { + "description": "Closest to the computer in the corner", + "timestamps": { + "start": "00:07:00", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:07:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The computer on the shelf.", + "B": "The computer in front of the corner.", + "C": "Use the computer sitting on the floor just outside the corner, not tucked into it.", + "D": "Use the desktop computer placed on the shelf near the corner rather than the one on the floor.", + "E": "The computer on the floor in the corner." + }, + "answer_label": "E", + "answer_text": "The computer on the floor in the corner.", + "q_id": "egolife_A1_D2_002_5", + "correct_answer": "The computer on the floor in the corner.", + "original_options": { + "A": "The computer on the shelf.", + "B": "The computer in front of the corner.", + "C": "Use the computer sitting on the floor just outside the corner, not tucked into it.", + "D": "Use the desktop computer placed on the shelf near the corner rather than the one on the floor.", + "E": "The computer on the floor in the corner." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_002_8", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is my phone now?", + "answer": "At the dining table.", + "question_chinese": "手机现在在哪里", + "answer_chinese": "在餐桌上", + "evidence": [ + { + "description": "The mobile phone is on the dining table now.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:14" + } + } + ], + "question_timestamps": "00:03:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Beside the dining table.", + "B": "Behind the dining table.", + "C": "Your phone is on the floor next to the dining table rather than resting on the tabletop.", + "D": "At the dining table.", + "E": "The phone ended up to the left of the dining table, not placed directly at the table." + }, + "answer_label": "D", + "answer_text": "At the dining table.", + "q_id": "egolife_A1_D2_002_8", + "correct_answer": "At the dining table.", + "original_options": { + "A": "Beside the dining table.", + "B": "Behind the dining table.", + "C": "Your phone is on the floor next to the dining table rather than resting on the tabletop.", + "D": "At the dining table.", + "E": "The phone ended up to the left of the dining table, not placed directly at the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_002_7", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the phone now?", + "answer": "In front of the camera.", + "question_chinese": "手机现在在哪里", + "answer_chinese": "在摄像机前方", + "evidence": [ + { + "description": "The mobile phone is in front of the camera.", + "timestamps": { + "start": "00:11:00", + "end": "00:11:10" + } + } + ], + "question_timestamps": "00:11:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To the left of the camera.", + "B": "To the right of the camera.", + "C": "The phone is near the table’s edge instead of positioned directly in front of the camera.", + "D": "It’s behind the camera viewpoint, not in the camera’s forward line of sight.", + "E": "In front of the camera." + }, + "answer_label": "E", + "answer_text": "In front of the camera.", + "q_id": "egolife_A1_D2_002_7", + "correct_answer": "In front of the camera.", + "original_options": { + "A": "To the left of the camera.", + "B": "To the right of the camera.", + "C": "The phone is near the table’s edge instead of positioned directly in front of the camera.", + "D": "It’s behind the camera viewpoint, not in the camera’s forward line of sight.", + "E": "In front of the camera." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_002_10", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the distance from the milk to the vase?", + "answer": "Three centimeters.", + "question_chinese": "牛奶相对于花瓶的距离是多少", + "answer_chinese": "三厘米", + "evidence": [ + { + "description": "The milk is in front of the dining table.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + }, + { + "description": "The vase is at the back of the dining table.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + }, + { + "description": "The distance between the two is three centimeters.", + "timestamps": { + "start": "00:00:50", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:12", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Five centimeters.", + "B": "Twelve centimeters.", + "C": "Three centimeters.", + "D": "The milk is about twenty centimeters away from the vase, leaving a noticeable gap between them.", + "E": "The distance is closer to forty centimeters, with the milk sitting far from the vase on the surface." + }, + "answer_label": "C", + "answer_text": "Three centimeters.", + "q_id": "egolife_A1_D2_002_10", + "correct_answer": "Three centimeters.", + "original_options": { + "A": "Five centimeters.", + "B": "Twelve centimeters.", + "C": "Three centimeters.", + "D": "The milk is about twenty centimeters away from the vase, leaving a noticeable gap between them.", + "E": "The distance is closer to forty centimeters, with the milk sitting far from the vase on the surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_002_9", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What is the distance from the milk to the vase?", + "answer": "Fifteen centimeters.", + "question_chinese": "牛奶相对于花瓶的距离是多少", + "answer_chinese": "十五厘米", + "evidence": [ + { + "description": "I hold the milk in my hand.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:05" + } + }, + { + "description": "I stand in front of the dining table.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:05" + } + }, + { + "description": "Fifteen centimeters away from the dining table", + "timestamps": { + "start": "00:02:00", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:02:19", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Fifteen centimeters.", + "B": "Twenty-five centimeters.", + "C": "Forty centimeters.", + "D": "The milk is roughly seventy centimeters from the vase, almost a full arm’s length apart.", + "E": "The vase is about five meters away from the milk, across the room rather than nearby." + }, + "answer_label": "A", + "answer_text": "Fifteen centimeters.", + "q_id": "egolife_A1_D2_002_9", + "correct_answer": "Fifteen centimeters.", + "original_options": { + "A": "Fifteen centimeters.", + "B": "Twenty-five centimeters.", + "C": "Forty centimeters.", + "D": "The milk is roughly seventy centimeters from the vase, almost a full arm’s length apart.", + "E": "The vase is about five meters away from the milk, across the room rather than nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_002_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mobile phones are on the dining table now?", + "answer": "Three.", + "question_chinese": "现在餐桌上有几个手机", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There are three mobile phones.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:16", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "There are six mobile phones on the dining table now, spread out across the tabletop.", + "D": "Three.", + "E": "Seven phones are on the dining table at the moment, with multiple devices clustered together." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_002_1", + "correct_answer": "Three.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "There are six mobile phones on the dining table now, spread out across the tabletop.", + "D": "Three.", + "E": "Seven phones are on the dining table at the moment, with multiple devices clustered together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_002_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mobile phones are there on the dining table now?", + "answer": "Five.", + "question_chinese": "现在餐桌上有几个手机", + "answer_chinese": "五个", + "evidence": [ + { + "description": "There are five mobile phones on the dining table now.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "There are two mobile phones on the dining table now, with the rest removed from it.", + "E": "Only one phone is on the dining table at the moment, sitting by itself on the surface." + }, + "answer_label": "C", + "answer_text": "Five.", + "q_id": "egolife_A1_D2_002_2", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "There are two mobile phones on the dining table now, with the rest removed from it.", + "E": "Only one phone is on the dining table at the moment, sitting by itself on the surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_002_3", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What position is the laptop in relative to the power bank?", + "answer": "In front of the power bank.", + "question_chinese": "电脑在相对于充电宝的什么位置", + "answer_chinese": "在充电宝的正前方", + "evidence": [ + { + "description": "The power bank is 2 centimeters in front of the sofa.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:03" + } + }, + { + "description": "The computer is 5 centimeters in front of the sofa.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:03" + } + } + ], + "question_timestamps": "00:04:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind the power bank.", + "B": "To the left of the power bank.", + "C": "The laptop is positioned on the table beside the power bank rather than directly in front of it.", + "D": "In front of the power bank.", + "E": "The laptop is below the power bank on the floor, not placed in front of the power bank." + }, + "answer_label": "D", + "answer_text": "In front of the power bank.", + "q_id": "egolife_A1_D2_002_3", + "correct_answer": "In front of the power bank.", + "original_options": { + "A": "Behind the power bank.", + "B": "To the left of the power bank.", + "C": "The laptop is positioned on the table beside the power bank rather than directly in front of it.", + "D": "In front of the power bank.", + "E": "The laptop is below the power bank on the floor, not placed in front of the power bank." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_002_4", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What position is the computer in relative to the power bank?", + "answer": "Directly above the power bank.", + "question_chinese": "电脑在相对于充电宝的什么位置", + "answer_chinese": "在充电宝正上方", + "evidence": [ + { + "description": "The power bank is in front of the sofa.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:15" + } + }, + { + "description": "The computer is above the sofa.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:15" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Directly below the power bank.", + "B": "Directly above the power bank.", + "C": "To the left of the power bank.", + "D": "The computer is positioned to the right side of the power bank, not stacked vertically.", + "E": "The computer sits directly in front of the power bank, closer to you on the surface." + }, + "answer_label": "B", + "answer_text": "Directly above the power bank.", + "q_id": "egolife_A1_D2_002_4", + "correct_answer": "Directly above the power bank.", + "original_options": { + "A": "Directly below the power bank.", + "B": "Directly above the power bank.", + "C": "To the left of the power bank.", + "D": "The computer is positioned to the right side of the power bank, not stacked vertically.", + "E": "The computer sits directly in front of the power bank, closer to you on the surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 15, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D2_003-mcq.json b/QAs/egolife_A1_D2_003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..80d7e51c215d497fa526c2eec39e9c70336deb32 --- /dev/null +++ b/QAs/egolife_A1_D2_003-mcq.json @@ -0,0 +1,814 @@ +[ + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_10", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to get chopsticks now, where should I put them?", + "answer": "Put it on the takeaway box.", + "question_chinese": "如果我现在想拿到筷子，我应该把筷子放在哪里", + "answer_chinese": "放在外卖箱上", + "evidence": [ + { + "description": "I am now standing in front of the takeaway box.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + }, + { + "description": "Five centimeters away from the takeout box", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + }, + { + "description": "The range within my reach is ten centimeters.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:07:19", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Put it behind the takeaway box.", + "B": "Put it in front of the takeaway box.", + "C": "Put the chopsticks to the left of the takeaway box so they’re reachable without placing them on top.", + "D": "Put the chopsticks beside the takeaway box, keeping them next to it rather than on it.", + "E": "Put it on the takeaway box." + }, + "answer_label": "E", + "answer_text": "Put it on the takeaway box.", + "q_id": "egolife_A1_D2_003_10", + "correct_answer": "Put it on the takeaway box.", + "original_options": { + "A": "Put it behind the takeaway box.", + "B": "Put it in front of the takeaway box.", + "C": "Put the chopsticks to the left of the takeaway box so they’re reachable without placing them on top.", + "D": "Put the chopsticks beside the takeaway box, keeping them next to it rather than on it.", + "E": "Put it on the takeaway box." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_9", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I want to get chopsticks now, where should I put them?", + "answer": "Put it on the dining table.", + "question_chinese": "如果我现在想拿到筷子，我应该把筷子放在哪里", + "answer_chinese": "放在餐桌上", + "evidence": [ + { + "description": "I stand in front of the dining table.", + "timestamps": { + "start": "00:11:00", + "end": "00:11:24" + } + }, + { + "description": "Five centimeters away from the dining table", + "timestamps": { + "start": "00:11:00", + "end": "00:11:24" + } + }, + { + "description": "My hand is placed within ten centimeters.", + "timestamps": { + "start": "00:11:00", + "end": "00:11:24" + } + } + ], + "question_timestamps": "00:11:28", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Put it under the dining table.", + "B": "Put it behind the dining table.", + "C": "Put the chopsticks in front of the dining table, not on the tabletop surface itself.", + "D": "Put it on the dining table.", + "E": "Put the chopsticks in the utensil holder near the dining table instead of placing them on the table." + }, + "answer_label": "D", + "answer_text": "Put it on the dining table.", + "q_id": "egolife_A1_D2_003_9", + "correct_answer": "Put it on the dining table.", + "original_options": { + "A": "Put it under the dining table.", + "B": "Put it behind the dining table.", + "C": "Put the chopsticks in front of the dining table, not on the tabletop surface itself.", + "D": "Put it on the dining table.", + "E": "Put the chopsticks in the utensil holder near the dining table instead of placing them on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_4_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, which computer should I use?", + "answer": "The computer was on the sofa, and you were sitting on the sofa.", + "question_chinese": "如果我现在要使用电脑，我应该用哪里的电脑？", + "answer_chinese": "沙发上电脑，你现在坐在沙发上", + "evidence": [ + { + "description": "I am sitting on the sofa now.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer was on the sofa, and you were sitting on the sofa.", + "B": "Use the computer three meters left.", + "C": "Use the computer on the floor.", + "D": "Use the computer on the floor in the corner in front of you rather than the one on the sofa.", + "E": "Use the computer about 10 meters to your front-left, since it’s the one you can head toward." + }, + "answer_label": "A", + "answer_text": "The computer was on the sofa, and you were sitting on the sofa.", + "q_id": "egolife_A1_D2_003_4_T1", + "correct_answer": "The computer was on the sofa, and you were sitting on the sofa.", + "original_options": { + "A": "The computer was on the sofa, and you were sitting on the sofa.", + "B": "Use the computer three meters left.", + "C": "Use the computer on the floor.", + "D": "Use the computer on the floor in the corner in front of you rather than the one on the sofa.", + "E": "Use the computer about 10 meters to your front-left, since it’s the one you can head toward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_4_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, which computer should I use?", + "answer": "The computer was on the floor in the corner, and the corner was in front of you.", + "question_chinese": "如果我现在要使用电脑，我应该用哪里的电脑？", + "answer_chinese": "角落地上的电脑，角落在你面前", + "evidence": [ + { + "description": "I am sitting on the sofa now.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:04:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The corner was behind you.", + "B": "The computer was on the floor in the corner, and the corner was in front of you.", + "C": "Use the computer on the sofa.", + "D": "The computer was on the floor in the corner, but that corner was to your left, not ahead of you.", + "E": "The computer was on the floor in the corner, yet you’d need to turn around because the corner is behind you." + }, + "answer_label": "B", + "answer_text": "The computer was on the floor in the corner, and the corner was in front of you.", + "q_id": "egolife_A1_D2_003_4_T2", + "correct_answer": "The computer was on the floor in the corner, and the corner was in front of you.", + "original_options": { + "A": "The corner was behind you.", + "B": "The computer was on the floor in the corner, and the corner was in front of you.", + "C": "Use the computer on the sofa.", + "D": "The computer was on the floor in the corner, but that corner was to your left, not ahead of you.", + "E": "The computer was on the floor in the corner, yet you’d need to turn around because the corner is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 22 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_5_T2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I want to use a computer right now, which computer should I use?", + "answer": "The computer on the floor in the corner.", + "question_chinese": "如果我现在要使用电脑，我应该用哪里的电脑？", + "answer_chinese": "角落地上的电脑", + "evidence": [ + { + "description": "I am standing in the corner now.\nClosest to the computer in the corner", + "timestamps": { + "start": "00:04:30", + "end": "00:04:40" + } + } + ], + "question_timestamps": "00:04:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer on the floor in the corner.", + "B": "The computer on the sofa.", + "C": "The computer three meters left.", + "D": "Use the computer about 10 meters to your front-left instead of the one on the floor in the corner.", + "E": "Choose the computer on the sofa because it’s the nearer option rather than the one in the corner." + }, + "answer_label": "A", + "answer_text": "The computer on the floor in the corner.", + "q_id": "egolife_A1_D2_003_5_T2", + "correct_answer": "The computer on the floor in the corner.", + "original_options": { + "A": "The computer on the floor in the corner.", + "B": "The computer on the sofa.", + "C": "The computer three meters left.", + "D": "Use the computer about 10 meters to your front-left instead of the one on the floor in the corner.", + "E": "Choose the computer on the sofa because it’s the nearer option rather than the one in the corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_9_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From where I am right now, how many turns do I need to make to get to the yard entrance?", + "answer": "One.", + "question_chinese": "我现在到院子门口要转几个弯？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "I need to make one turns to get from the second floor to the entrance of the yard.", + "timestamps": { + "start": "00:05:50", + "end": "00:06:15" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "Three.", + "C": "You need to make two turns to line up with the path before reaching the yard entrance.", + "D": "You need to make three turns overall, changing direction at multiple points to arrive at the yard entrance.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_D2_003_9_T1", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "You need to make two turns to line up with the path before reaching the yard entrance.", + "D": "You need to make three turns overall, changing direction at multiple points to arrive at the yard entrance.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_9_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From where I am right now, how many turns do I need to make to get to the yard entrance?", + "answer": "You needed to turn once.", + "question_chinese": "我现在到院子门口要转几个弯？", + "answer_chinese": "你需要转一个", + "evidence": [ + { + "description": "I need to make one turns to get from the second floor to the entrance of the yard.", + "timestamps": { + "start": "00:05:50", + "end": "00:06:15" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You needed to turn twice.", + "B": "You needed to turn once.", + "C": "You don't need to make any turns.", + "D": "You needed to turn three times, adjusting direction at several points before reaching the yard entrance.", + "E": "You needed to turn four times in total, making multiple direction changes on the way to the yard entrance." + }, + "answer_label": "B", + "answer_text": "You needed to turn once.", + "q_id": "egolife_A1_D2_003_9_T2", + "correct_answer": "You needed to turn once.", + "original_options": { + "A": "You needed to turn twice.", + "B": "You needed to turn once.", + "C": "You don't need to make any turns.", + "D": "You needed to turn three times, adjusting direction at several points before reaching the yard entrance.", + "E": "You needed to turn four times in total, making multiple direction changes on the way to the yard entrance." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_8_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From where I am right now, how many turns do I need to make to get to the yard entrance?", + "answer": "Three.", + "question_chinese": "我现在到院子门口要转几个弯？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "I made three turns from the second floor to the gate of the yard.", + "timestamps": { + "start": "00:05:50", + "end": "00:06:15" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You only need to make two turns from here to reach the yard entrance.", + "E": "It would take four turns in total before you arrive at the yard entrance." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_003_8_T1", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You only need to make two turns from here to reach the yard entrance.", + "E": "It would take four turns in total before you arrive at the yard entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_8_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From where I am right now, how many turns do I need to make to get to the yard entrance?", + "answer": "You needed to turn three times.", + "question_chinese": "我现在到院子门口要转几个弯？", + "answer_chinese": "你需要转三个", + "evidence": [ + { + "description": "I made three turns from the second floor to the gate of the yard.", + "timestamps": { + "start": "00:05:50", + "end": "00:06:15" + } + } + ], + "question_timestamps": "00:11:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You needed to turn three times.", + "B": "You needed to turn once.", + "C": "You needed to turn twice.", + "D": "You needed to turn four times to get from here to the yard entrance.", + "E": "You needed to turn five times before reaching the yard entrance from your current spot." + }, + "answer_label": "A", + "answer_text": "You needed to turn three times.", + "q_id": "egolife_A1_D2_003_8_T2", + "correct_answer": "You needed to turn three times.", + "original_options": { + "A": "You needed to turn three times.", + "B": "You needed to turn once.", + "C": "You needed to turn twice.", + "D": "You needed to turn four times to get from here to the yard entrance.", + "E": "You needed to turn five times before reaching the yard entrance from your current spot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_3_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I positioned relative to the blanket?", + "answer": "It was directly above the blanket.", + "question_chinese": "我相对于毛毯在什么位置？", + "answer_chinese": "在毛毯的正上方", + "evidence": [ + { + "description": "I am sitting on the sofa.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:04" + } + }, + { + "description": "The blanket is on the sofa.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:04" + } + } + ], + "question_timestamps": "00:03:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was directly below the blanket.", + "B": "It was beside the blanket.", + "C": "It was positioned underneath the blanket, rather than directly above the blanket.", + "D": "It was directly above the blanket.", + "E": "It was located just behind the blanket, and not directly above it at all." + }, + "answer_label": "D", + "answer_text": "It was directly above the blanket.", + "q_id": "egolife_A1_D2_003_3_T1", + "correct_answer": "It was directly above the blanket.", + "original_options": { + "A": "It was directly below the blanket.", + "B": "It was beside the blanket.", + "C": "It was positioned underneath the blanket, rather than directly above the blanket.", + "D": "It was directly above the blanket.", + "E": "It was located just behind the blanket, and not directly above it at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_2_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I positioned relative to the blanket?", + "answer": "It was under the blanket.", + "question_chinese": "我相对于毛毯在什么位置？", + "answer_chinese": "在毛毯下方", + "evidence": [ + { + "description": "I'm standing on the first floor.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:20" + } + }, + { + "description": "The blanket is on the sofa, and the sofa is on the second floor.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:20" + } + } + ], + "question_timestamps": "00:11:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was directly above the blanket.", + "B": "It was beside the blanket.", + "C": "It was directly below the blanket, rather than actually being underneath the blanket.", + "D": "It was under the blanket.", + "E": "It was positioned behind the blanket, and not tucked under the blanket at all." + }, + "answer_label": "D", + "answer_text": "It was under the blanket.", + "q_id": "egolife_A1_D2_003_2_T1", + "correct_answer": "It was under the blanket.", + "original_options": { + "A": "It was directly above the blanket.", + "B": "It was beside the blanket.", + "C": "It was directly below the blanket, rather than actually being underneath the blanket.", + "D": "It was under the blanket.", + "E": "It was positioned behind the blanket, and not tucked under the blanket at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_2_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I positioned relative to the blanket?", + "answer": "It was under the blanket.", + "question_chinese": "我相对于毛毯在什么位置？", + "answer_chinese": "在毛毯下方", + "evidence": [ + { + "description": "The blanket is on the sofa, and the sofa is on the second floor.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:20" + } + } + ], + "question_timestamps": "00:11:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was directly above the blanket.", + "B": "It was next to the blanket.", + "C": "It was directly below the blanket instead of being covered by it.", + "D": "It was under the blanket.", + "E": "It was placed in front of the blanket, and not underneath the blanket at all." + }, + "answer_label": "D", + "answer_text": "It was under the blanket.", + "q_id": "egolife_A1_D2_003_2_T2", + "correct_answer": "It was under the blanket.", + "original_options": { + "A": "It was directly above the blanket.", + "B": "It was next to the blanket.", + "C": "It was directly below the blanket instead of being covered by it.", + "D": "It was under the blanket.", + "E": "It was placed in front of the blanket, and not underneath the blanket at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls are on the table?", + "answer": "Zero.", + "question_chinese": "桌子上有几个碗", + "answer_chinese": "零个", + "evidence": [ + { + "description": "There are no bowls on the dining table now.", + "timestamps": { + "start": "00:10:40", + "end": "00:10:45" + } + } + ], + "question_timestamps": "00:10:58", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One bowl.", + "B": "Two bowls.", + "C": "Zero.", + "D": "Three bowls are on the table, grouped together in one area near the center.", + "E": "Four bowls are on the table, spread across the tabletop in front of you." + }, + "answer_label": "C", + "answer_text": "Zero.", + "q_id": "egolife_A1_D2_003_1", + "correct_answer": "Zero.", + "original_options": { + "A": "One bowl.", + "B": "Two bowls.", + "C": "Zero.", + "D": "Three bowls are on the table, grouped together in one area near the center.", + "E": "Four bowls are on the table, spread across the tabletop in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_003_2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls are on the table?", + "answer": "Four bowls.", + "question_chinese": "桌子上有几个碗", + "answer_chinese": "四个碗", + "evidence": [ + { + "description": "There are four bowls on the dining table.", + "timestamps": { + "start": "00:11:27", + "end": "00:11:32" + } + } + ], + "question_timestamps": "00:11:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three bowls.", + "B": "Five bowls.", + "C": "Two bowls are on the table, sitting next to each other near the vase.", + "D": "One bowl is on the table, with no other bowls visible nearby.", + "E": "Four bowls." + }, + "answer_label": "E", + "answer_text": "Four bowls.", + "q_id": "egolife_A1_D2_003_2", + "correct_answer": "Four bowls.", + "original_options": { + "A": "Three bowls.", + "B": "Five bowls.", + "C": "Two bowls are on the table, sitting next to each other near the vase.", + "D": "One bowl is on the table, with no other bowls visible nearby.", + "E": "Four bowls." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 12 + ] + } +] diff --git a/QAs/egolife_A1_D2_004-mcq.json b/QAs/egolife_A1_D2_004-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..0b6e7bacb43f56c830014564c87c79622fbe2d57 --- /dev/null +++ b/QAs/egolife_A1_D2_004-mcq.json @@ -0,0 +1,618 @@ +[ + { + "video_uid": "egolife_A1_D2_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_004_4", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the power bank relative to the vase on the dining table?", + "answer": "On the left side of the dining table.", + "question_chinese": "充电宝相对于餐桌上的花瓶的什么位置", + "answer_chinese": "在餐桌的左侧", + "evidence": [ + { + "description": "The power bank is on the bed.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:54" + } + }, + { + "description": "The bed is on the left side of the vase.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:54" + } + } + ], + "question_timestamps": "00:01:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To the right of the vase.", + "B": "In front of the vase.", + "C": "On the left side of the dining table.", + "D": "It is behind the vase on the dining table, not on the left side.", + "E": "It sits near the center of the dining table rather than to the left of the vase." + }, + "answer_label": "C", + "answer_text": "On the left side of the dining table.", + "q_id": "egolife_A1_D2_004_4", + "correct_answer": "On the left side of the dining table.", + "original_options": { + "A": "To the right of the vase.", + "B": "In front of the vase.", + "C": "On the left side of the dining table.", + "D": "It is behind the vase on the dining table, not on the left side.", + "E": "It sits near the center of the dining table rather than to the left of the vase." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_004_3", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the power bank relative to the vase on the dining table?", + "answer": "To the right of the vase.", + "question_chinese": "充电宝相对于餐桌上的花瓶的什么位置", + "answer_chinese": "在花瓶的右侧", + "evidence": [ + { + "description": "The vase is on the left side of the dining table.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:40" + } + }, + { + "description": "The power bank is on the right side of the dining table.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:05:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To the left of the vase.", + "B": "To the right of the vase.", + "C": "Behind the vase.", + "D": "In front of the vase, close to the table edge, rather than on its right side.", + "E": "Slightly left of the vase, positioned nearer to you instead of being on the right." + }, + "answer_label": "B", + "answer_text": "To the right of the vase.", + "q_id": "egolife_A1_D2_004_3", + "correct_answer": "To the right of the vase.", + "original_options": { + "A": "To the left of the vase.", + "B": "To the right of the vase.", + "C": "Behind the vase.", + "D": "In front of the vase, close to the table edge, rather than on its right side.", + "E": "Slightly left of the vase, positioned nearer to you instead of being on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 3, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_004_7", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I walk in the door now, which is closer to me, the pot or my bowl?", + "answer": "My bowl is closer to me.", + "question_chinese": "如果我现在进门口，锅和我的碗哪个离我更近", + "answer_chinese": "我的碗离我更近", + "evidence": [ + { + "description": "The pot is sixty centimeters away from the door.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + }, + { + "description": "My bowl is fifty-five centimeters away from the door.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The pot is closer to you.", + "B": "My bowl is closer to me." + }, + "answer_label": "B", + "answer_text": "My bowl is closer to me.", + "q_id": "egolife_A1_D2_004_7", + "correct_answer": "My bowl is closer to me.", + "original_options": { + "A": "The pot is closer to you.", + "B": "My bowl is closer to me." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D2_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_004_8", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I walk in the door now, which is closer to me, the pot or my bowl?", + "answer": "Pot.", + "question_chinese": "如果我现在进门口，锅和我的碗哪个离我更近", + "answer_chinese": "锅", + "evidence": [ + { + "description": "The pot is fifty centimeters away from the door.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:10" + } + }, + { + "description": "My bowl is fifty-five centimeters away from the door.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:10" + } + } + ], + "question_timestamps": "00:10:24", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Your bowl is closer to you.", + "B": "Pot." + }, + "answer_label": "B", + "answer_text": "Pot.", + "q_id": "egolife_A1_D2_004_8", + "correct_answer": "Pot.", + "original_options": { + "A": "Your bowl is closer to you.", + "B": "Pot." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D2_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_004-mcq_fixed_8", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my current path to the gate?", + "answer": "Turn left when you go out, then go straight, and finally turn left.", + "question_chinese": "我现在到大门的路径是什么", + "answer_chinese": "出门左转，然后直走，最后左转", + "evidence": [ + { + "description": "I go out, turn left, then go straight, and finally turn left.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:30" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left when you go out, then go straight, and finally turn left.", + "B": "Turn right, then go straight.", + "C": "Go straight, then turn left.", + "D": "Turn left when you go out, then turn right, and finally go straight to reach the gate.", + "E": "Exit and turn right, continue straight for a bit, then make a final right turn at the end." + }, + "answer_label": "A", + "answer_text": "Turn left when you go out, then go straight, and finally turn left.", + "q_id": "egolife_A1_D2_004-mcq_fixed_8", + "correct_answer": "Turn left when you go out, then go straight, and finally turn left.", + "original_options": { + "A": "Turn left when you go out, then go straight, and finally turn left.", + "B": "Turn right, then go straight.", + "C": "Go straight, then turn left.", + "D": "Turn left when you go out, then turn right, and finally go straight to reach the gate.", + "E": "Exit and turn right, continue straight for a bit, then make a final right turn at the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_004-mcq_fixed_9", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my current path to the gate?", + "answer": "Go out of the room and turn left.Turn left when you go out, then go straight, and finally turn left.", + "question_chinese": "我现在到大门的路径是什么", + "answer_chinese": "出房间左转，出门左转，然后直走，最后左转", + "evidence": [ + { + "description": "I go out, turn left, then go straight, and finally turn left.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:30" + } + } + ], + "question_timestamps": "00:11:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go out and turn right.", + "B": "Go out and go straight.", + "C": "Go out of the room and turn left, then turn right, and finally turn left at the end.", + "D": "Step out of the room, go straight down the hallway, then turn right and continue until the gate.", + "E": "Go out of the room and turn left.Turn left when you go out, then go straight, and finally turn left." + }, + "answer_label": "E", + "answer_text": "Go out of the room and turn left.Turn left when you go out, then go straight, and finally turn left.", + "q_id": "egolife_A1_D2_004-mcq_fixed_9", + "correct_answer": "Go out of the room and turn left.Turn left when you go out, then go straight, and finally turn left.", + "original_options": { + "A": "Go out and turn right.", + "B": "Go out and go straight.", + "C": "Go out of the room and turn left, then turn right, and finally turn left at the end.", + "D": "Step out of the room, go straight down the hallway, then turn right and continue until the gate.", + "E": "Go out of the room and turn left.Turn left when you go out, then go straight, and finally turn left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_004_6", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are currently on the bed?", + "answer": "Three.", + "question_chinese": "现在床上充电宝的数量是多少", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There are three power banks on the bed now.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:54" + } + } + ], + "question_timestamps": "00:02:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "Five power banks are on the bed right now, spread out rather than grouped together.", + "E": "One power bank is currently on the bed, with the others no longer placed there." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_004_6", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "Five power banks are on the bed right now, spread out rather than grouped together.", + "E": "One power bank is currently on the bed, with the others no longer placed there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_004_5", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are currently on the bed?", + "answer": "Zero.", + "question_chinese": "现在床上充电宝的数量是多少", + "answer_chinese": "零个", + "evidence": [ + { + "description": "I took the three power banks from the bed and put them on the dining table.", + "timestamps": { + "start": "00:01:50", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "Three power banks are currently on the bed, sitting near each other rather than none at all.", + "E": "Four power banks are on the bed right now, not zero, occupying separate spots on the bedding." + }, + "answer_label": "A", + "answer_text": "Zero.", + "q_id": "egolife_A1_D2_004_5", + "correct_answer": "Zero.", + "original_options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "Three power banks are currently on the bed, sitting near each other rather than none at all.", + "E": "Four power banks are on the bed right now, not zero, occupying separate spots on the bedding." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_004_2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the sugar from the iron pot now?", + "answer": "Three centimeters.", + "question_chinese": "糖现在距离铁锅的距离是多少", + "answer_chinese": "三厘米", + "evidence": [ + { + "description": "The sugar is now directly above the bowl.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:45" + } + }, + { + "description": "The bowl is next to the iron pot.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:45" + } + }, + { + "description": "Three centimeters away from the iron pot", + "timestamps": { + "start": "00:00:20", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Two centimeters.", + "B": "Five centimeters.", + "C": "One centimeter, with the sugar almost touching the iron pot rather than being three centimeters away.", + "D": "Four centimeters, so the sugar sits noticeably farther from the iron pot than a three-centimeter gap.", + "E": "Three centimeters." + }, + "answer_label": "E", + "answer_text": "Three centimeters.", + "q_id": "egolife_A1_D2_004_2", + "correct_answer": "Three centimeters.", + "original_options": { + "A": "Two centimeters.", + "B": "Five centimeters.", + "C": "One centimeter, with the sugar almost touching the iron pot rather than being three centimeters away.", + "D": "Four centimeters, so the sugar sits noticeably farther from the iron pot than a three-centimeter gap.", + "E": "Three centimeters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_004_1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the sugar from the iron pot now?", + "answer": "Fifty centimeters.", + "question_chinese": "糖现在距离铁锅的距离是多少", + "answer_chinese": "五十厘米", + "evidence": [ + { + "description": "The sugar is on the cabinet now.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:20" + } + }, + { + "description": "The wok is directly above the dining table.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:20" + } + }, + { + "description": "The dining table is fifty centimeters away from the cabinet.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Forty centimeters.", + "B": "Sixty centimeters.", + "C": "About seventy centimeters away from the iron pot right now, not fifty centimeters.", + "D": "Fifty centimeters.", + "E": "Roughly thirty-five centimeters from the iron pot at the moment, not fifty centimeters." + }, + "answer_label": "D", + "answer_text": "Fifty centimeters.", + "q_id": "egolife_A1_D2_004_1", + "correct_answer": "Fifty centimeters.", + "original_options": { + "A": "Forty centimeters.", + "B": "Sixty centimeters.", + "C": "About seventy centimeters away from the iron pot right now, not fifty centimeters.", + "D": "Fifty centimeters.", + "E": "Roughly thirty-five centimeters from the iron pot at the moment, not fifty centimeters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D2_007-mcq.json b/QAs/egolife_A1_D2_007-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..49c530a351c424ff689038598259304c6a7cac20 --- /dev/null +++ b/QAs/egolife_A1_D2_007-mcq.json @@ -0,0 +1,792 @@ +[ + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are there in the living room?", + "answer": "One power bank.", + "question_chinese": "客厅里有几个充电宝？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There is a power bank on the table", + "timestamps": { + "start": "00:04:49", + "end": "00:04:59" + } + } + ], + "question_timestamps": "00:05:31", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two power banks.", + "B": "Three power banks.", + "C": "There are four power banks visible in the living room right now.", + "D": "One power bank.", + "E": "You can see five power banks placed around the living room area." + }, + "answer_label": "D", + "answer_text": "One power bank.", + "q_id": "egolife_A1_D2_007_1", + "correct_answer": "One power bank.", + "original_options": { + "A": "Two power banks.", + "B": "Three power banks.", + "C": "There are four power banks visible in the living room right now.", + "D": "One power bank.", + "E": "You can see five power banks placed around the living room area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are there in the living room?", + "answer": "Two power banks.", + "question_chinese": "客厅里有几个充电宝？", + "answer_chinese": "两个充电宝", + "evidence": [ + { + "description": "The second power bank on the table", + "timestamps": { + "start": "00:05:29", + "end": "00:05:41" + } + }, + { + "description": "The first power bank on the table", + "timestamps": { + "start": "00:04:49", + "end": "00:04:59" + } + } + ], + "question_timestamps": "00:06:47", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One power bank.", + "B": "Three power banks.", + "C": "There are four power banks in the living room at the moment.", + "D": "You can find five power banks in the living room right now.", + "E": "Two power banks." + }, + "answer_label": "E", + "answer_text": "Two power banks.", + "q_id": "egolife_A1_D2_007_2", + "correct_answer": "Two power banks.", + "original_options": { + "A": "One power bank.", + "B": "Three power banks.", + "C": "There are four power banks in the living room at the moment.", + "D": "You can find five power banks in the living room right now.", + "E": "Two power banks." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_8", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take while moving in the living room?", + "answer": "I walked from the side of the table far from the sink to the sink.", + "question_chinese": "我刚刚在客厅里是沿着什么路线移动的", + "answer_chinese": "我从桌子远离洗碗池的一侧走到洗碗池", + "evidence": [ + { + "description": "I walked from the side of the table far from the sink to the sink", + "timestamps": { + "start": "00:08:22", + "end": "00:08:28" + } + } + ], + "question_timestamps": "00:08:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I walked from the sink to the table.", + "B": "I walked from the side of the table far from the sink to the sink.", + "C": "I walked from the table to the couch.", + "D": "I walked from the side of the table close to the sink to the side far from it.", + "E": "I walked from the side of the table far from the sink to the side close to the couch." + }, + "answer_label": "B", + "answer_text": "I walked from the side of the table far from the sink to the sink.", + "q_id": "egolife_A1_D2_007_8", + "correct_answer": "I walked from the side of the table far from the sink to the sink.", + "original_options": { + "A": "I walked from the sink to the table.", + "B": "I walked from the side of the table far from the sink to the sink.", + "C": "I walked from the table to the couch.", + "D": "I walked from the side of the table close to the sink to the side far from it.", + "E": "I walked from the side of the table far from the sink to the side close to the couch." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_7", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take while moving in the living room?", + "answer": "I started from the bedroom door and walked counterclockwise around the table to the side close to the sink.", + "question_chinese": "我刚刚在客厅里是沿着什么路线移动的", + "answer_chinese": "我从卧室门口出发，绕着桌子逆时针来到靠近洗碗池的一边。", + "evidence": [ + { + "description": "I started from the bedroom door and walked counterclockwise around the table to the side close to the sink.", + "timestamps": { + "start": "00:10:34", + "end": "00:10:42" + } + } + ], + "question_timestamps": "00:10:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I walked clockwise around the table.", + "B": "I started from the bedroom door and walked counterclockwise around the table to the side close to the sink.", + "C": "I started from the living room door.", + "D": "I started from the bedroom door and walked clockwise around the table to the side close to the sink.", + "E": "I started from the bedroom door and walked counterclockwise around the table to the side far from the sink." + }, + "answer_label": "B", + "answer_text": "I started from the bedroom door and walked counterclockwise around the table to the side close to the sink.", + "q_id": "egolife_A1_D2_007_7", + "correct_answer": "I started from the bedroom door and walked counterclockwise around the table to the side close to the sink.", + "original_options": { + "A": "I walked clockwise around the table.", + "B": "I started from the bedroom door and walked counterclockwise around the table to the side close to the sink.", + "C": "I started from the living room door.", + "D": "I started from the bedroom door and walked clockwise around the table to the side close to the sink.", + "E": "I started from the bedroom door and walked counterclockwise around the table to the side far from the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_6", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I turn off the faucet now?", + "answer": "Yes.", + "question_chinese": "我现在能关闭水龙头吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "I'm right beside the faucet now and just turned it on at this spot", + "timestamps": { + "start": "00:08:53", + "end": "00:09:03" + } + } + ], + "question_timestamps": "00:09:18", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D2_007_6", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_5", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I turn off the faucet now?", + "answer": "No.", + "question_chinese": "我现在能关闭水龙头吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "I'm right beside the faucet now and just turned it on at this spot", + "timestamps": { + "start": "00:08:53", + "end": "00:09:03" + } + } + ], + "question_timestamps": "00:11:47", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D2_007_5", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007-mcq_fixed_11", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the room?", + "answer": "Five.", + "question_chinese": "房间里有几个垃圾桶？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There is a trash can under the table", + "timestamps": { + "start": "00:01:56", + "end": "00:02:09" + } + } + ], + "question_timestamps": "00:02:39", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "There are three trash cans in the room, not five in total.", + "E": "You can count seven trash cans in the room, not five overall." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D2_007-mcq_fixed_11", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "There are three trash cans in the room, not five in total.", + "E": "You can count seven trash cans in the room, not five overall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007-mcq_fixed_10", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the room?", + "answer": "Four.", + "question_chinese": "房间里有几个垃圾桶？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "There is a trash can under the table", + "timestamps": { + "start": "00:01:56", + "end": "00:02:09" + } + }, + { + "description": "The second trash can in the room", + "timestamps": { + "start": "00:11:05", + "end": "00:11:10" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "There are two trash cans in the room near the doorway and wall.", + "E": "You can see five trash cans placed around the room, including one by the desk." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "egolife_A1_D2_007-mcq_fixed_10", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "There are two trash cans in the room near the doorway and wall.", + "E": "You can see five trash cans placed around the room, including one by the desk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_2_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the iron pot from me?", + "answer": "The iron pot is on your right front.", + "question_chinese": "铁锅在我的什么方向？", + "answer_chinese": "铁锅在我的右前方", + "evidence": [ + { + "description": "The iron pot is on my right front", + "timestamps": { + "start": "00:03:45", + "end": "00:03:51" + } + } + ], + "question_timestamps": "00:04:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your left front.", + "B": "Directly behind you.", + "C": "The iron pot is on your right front.", + "D": "The iron pot is behind you on the counter, not in front.", + "E": "The iron pot sits on your left front, a bit ahead of where you stand." + }, + "answer_label": "C", + "answer_text": "The iron pot is on your right front.", + "q_id": "egolife_A1_D2_007_2_T1", + "correct_answer": "The iron pot is on your right front.", + "original_options": { + "A": "On your left front.", + "B": "Directly behind you.", + "C": "The iron pot is on your right front.", + "D": "The iron pot is behind you on the counter, not in front.", + "E": "The iron pot sits on your left front, a bit ahead of where you stand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_2_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the cast iron pot from me?", + "answer": "The iron pot is on your right front.", + "question_chinese": "铁锅在我的什么方向？", + "answer_chinese": "铁锅在我的右前方", + "evidence": [ + { + "description": "The iron pot is on my right front", + "timestamps": { + "start": "00:03:45", + "end": "00:03:51" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your left front.", + "B": "Directly behind you.", + "C": "The iron pot is on your right front.", + "D": "The iron pot is behind you on the counter, not in front.", + "E": "The iron pot sits on your left front, a bit ahead of where you stand." + }, + "answer_label": "C", + "answer_text": "The iron pot is on your right front.", + "q_id": "egolife_A1_D2_007_2_T2", + "correct_answer": "The iron pot is on your right front.", + "original_options": { + "A": "On your left front.", + "B": "Directly behind you.", + "C": "The iron pot is on your right front.", + "D": "The iron pot is behind you on the counter, not in front.", + "E": "The iron pot sits on your left front, a bit ahead of where you stand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_3_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the iron pot from me?", + "answer": "The iron pot is on your right.", + "question_chinese": "铁锅在我的什么方向？", + "answer_chinese": "铁锅在我的右边", + "evidence": [ + { + "description": "The iron pot is on my right", + "timestamps": { + "start": "00:09:34", + "end": "00:09:43" + } + } + ], + "question_timestamps": "00:09:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The iron pot is on your right.", + "B": "On your left.", + "C": "In front of you.", + "D": "The iron pot is behind you rather than on your right side.", + "E": "The iron pot is underneath you, not positioned to your right at all." + }, + "answer_label": "A", + "answer_text": "The iron pot is on your right.", + "q_id": "egolife_A1_D2_007_3_T1", + "correct_answer": "The iron pot is on your right.", + "original_options": { + "A": "The iron pot is on your right.", + "B": "On your left.", + "C": "In front of you.", + "D": "The iron pot is behind you rather than on your right side.", + "E": "The iron pot is underneath you, not positioned to your right at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_3_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In what direction is the cast iron pot from me?", + "answer": "The iron pot is on your right.", + "question_chinese": "铁锅在我的什么方向？", + "answer_chinese": "铁锅在我的右边", + "evidence": [ + { + "description": "The iron pot is on my right", + "timestamps": { + "start": "00:09:34", + "end": "00:09:43" + } + } + ], + "question_timestamps": "00:09:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The iron pot is on your right.", + "B": "On your left.", + "C": "In front of you.", + "D": "The iron pot is behind you rather than on your right side.", + "E": "The iron pot is underneath you, not positioned to your right at all." + }, + "answer_label": "A", + "answer_text": "The iron pot is on your right.", + "q_id": "egolife_A1_D2_007_3_T2", + "correct_answer": "The iron pot is on your right.", + "original_options": { + "A": "The iron pot is on your right.", + "B": "On your left.", + "C": "In front of you.", + "D": "The iron pot is behind you rather than on your right side.", + "E": "The iron pot is underneath you, not positioned to your right at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_9", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the mirror or the refrigerator?", + "answer": "Refrigerator.", + "question_chinese": "镜子和冰箱哪个理我更近一些", + "answer_chinese": "冰箱", + "evidence": [ + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + }, + { + "description": "The refrigerator is beside the curtain", + "timestamps": { + "start": "00:00:37", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:05:26", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Refrigerator.", + "B": "Mirror." + }, + "answer_label": "A", + "answer_text": "Refrigerator.", + "q_id": "egolife_A1_D2_007_9", + "correct_answer": "Refrigerator.", + "original_options": { + "A": "Refrigerator.", + "B": "Mirror." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_007_10", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the mirror or the refrigerator?", + "answer": "Mirror.", + "question_chinese": "镜子和冰箱哪个理我更近一些", + "answer_chinese": "镜子", + "evidence": [ + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + }, + { + "description": "The refrigerator is beside the curtain", + "timestamps": { + "start": "00:00:37", + "end": "00:00:47" + } + } + ], + "question_timestamps": "00:10:42", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Refrigerator.", + "B": "Mirror." + }, + "answer_label": "B", + "answer_text": "Mirror.", + "q_id": "egolife_A1_D2_007_10", + "correct_answer": "Mirror.", + "original_options": { + "A": "Refrigerator.", + "B": "Mirror." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D2_008-mcq.json b/QAs/egolife_A1_D2_008-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fec685fabe0146a9ab88922326f6873fc579a417 --- /dev/null +++ b/QAs/egolife_A1_D2_008-mcq.json @@ -0,0 +1,1101 @@ +[ + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are there in the bedroom?", + "answer": "One.", + "question_chinese": "卧室里一共有几个充电宝？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There is a power bank in the bedroom", + "timestamps": { + "start": "00:01:28", + "end": "00:01:37" + } + } + ], + "question_timestamps": "00:02:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "There are four power banks in the bedroom, lined up together on a nearby shelf.", + "D": "One.", + "E": "You have two power banks in the bedroom area, one by the bed and one on a table." + }, + "answer_label": "D", + "answer_text": "One.", + "q_id": "egolife_A1_D2_008_1", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "There are four power banks in the bedroom, lined up together on a nearby shelf.", + "D": "One.", + "E": "You have two power banks in the bedroom area, one by the bed and one on a table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are there in the bedroom?", + "answer": "Five.", + "question_chinese": "卧室里一共有几个充电宝？", + "answer_chinese": "五个", + "evidence": [ + { + "description": "There is a power bank in the bedroom", + "timestamps": { + "start": "00:01:28", + "end": "00:01:37" + } + }, + { + "description": "I brought in four more power banks from outside the bedroom", + "timestamps": { + "start": "00:05:21", + "end": "00:05:32" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "Three power banks are in the bedroom, grouped together on the same surface.", + "D": "Five.", + "E": "Seven power banks are in the bedroom, with several lined up and the rest stacked nearby." + }, + "answer_label": "D", + "answer_text": "Five.", + "q_id": "egolife_A1_D2_008_2", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "Three power banks are in the bedroom, grouped together on the same surface.", + "D": "Five.", + "E": "Seven power banks are in the bedroom, with several lined up and the rest stacked nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_9_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I picked up the sticky note just now, where did I put it?", + "answer": "You put it on the table in the bedroom.", + "question_chinese": "我刚刚拿起便签后将它放在了哪里？", + "answer_chinese": "我把它放在了卧室的桌子上", + "evidence": [ + { + "description": "I picked up the note and placed it on the table", + "timestamps": { + "start": "00:02:47", + "end": "00:02:54" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the bed.", + "B": "You put it on the table in the bedroom.", + "C": "On the dresser.", + "D": "You placed it in front of the table in the bedroom, rather than setting it down on top.", + "E": "You put it behind the table in the bedroom, not on the tabletop where you were working." + }, + "answer_label": "B", + "answer_text": "You put it on the table in the bedroom.", + "q_id": "egolife_A1_D2_008_9_T1", + "correct_answer": "You put it on the table in the bedroom.", + "original_options": { + "A": "On the bed.", + "B": "You put it on the table in the bedroom.", + "C": "On the dresser.", + "D": "You placed it in front of the table in the bedroom, rather than setting it down on top.", + "E": "You put it behind the table in the bedroom, not on the tabletop where you were working." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_9_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I put the sticky note just after I picked it up？", + "answer": "You put it on the table in the bedroom.", + "question_chinese": "我刚刚拿起便签后将它放在了哪里？", + "answer_chinese": "我把它放在了卧室的桌子上", + "evidence": [ + { + "description": "I picked up the note and placed it on the table", + "timestamps": { + "start": "00:02:47", + "end": "00:02:54" + } + } + ], + "question_timestamps": "00:03:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the bed.", + "B": "You put it on the table in the bedroom.", + "C": "On the dresser.", + "D": "You set it down on the coffee table beside the couch instead of placing it in the bedroom.", + "E": "You left it on the hallway shelf underneath the mirror rather than putting it on the bedroom table." + }, + "answer_label": "B", + "answer_text": "You put it on the table in the bedroom.", + "q_id": "egolife_A1_D2_008_9_T2", + "correct_answer": "You put it on the table in the bedroom.", + "original_options": { + "A": "On the bed.", + "B": "You put it on the table in the bedroom.", + "C": "On the dresser.", + "D": "You set it down on the coffee table beside the couch instead of placing it in the bedroom.", + "E": "You left it on the hallway shelf underneath the mirror rather than putting it on the bedroom table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_4_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I picked up the sticky note just now, where did I put it?", + "answer": "You put it on the computer table on the second floor.", + "question_chinese": "我刚刚拿起便签后将它放在了哪里？", + "answer_chinese": "我将它放在了二楼电脑的桌子上", + "evidence": [ + { + "description": "I put the sticky note on the computer on the second floor", + "timestamps": { + "start": "00:09:01", + "end": "00:09:07" + } + }, + { + "description": "I picked up the sticky note", + "timestamps": { + "start": "00:03:35", + "end": "00:03:49" + } + } + ], + "question_timestamps": "00:09:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the first floor.", + "B": "You put it on the computer table on the second floor.", + "C": "On the desk chair.", + "D": "You placed it under the computer table on the second floor, not on top of it.", + "E": "You put it on a downstairs table, instead of on the computer table on the second floor." + }, + "answer_label": "B", + "answer_text": "You put it on the computer table on the second floor.", + "q_id": "egolife_A1_D2_008_4_T1", + "correct_answer": "You put it on the computer table on the second floor.", + "original_options": { + "A": "On the first floor.", + "B": "You put it on the computer table on the second floor.", + "C": "On the desk chair.", + "D": "You placed it under the computer table on the second floor, not on top of it.", + "E": "You put it on a downstairs table, instead of on the computer table on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_4_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I put the sticky note after I just picked it up?", + "answer": "You put it on the computer table on the second floor.", + "question_chinese": "我刚刚拿起便签后将它放在了哪里？", + "answer_chinese": "我将它放在了二楼电脑的桌子上", + "evidence": [ + { + "description": "I put the sticky note on the computer on the second floor", + "timestamps": { + "start": "00:09:01", + "end": "00:09:07" + } + }, + { + "description": "I picked up the sticky note", + "timestamps": { + "start": "00:03:35", + "end": "00:03:49" + } + } + ], + "question_timestamps": "00:09:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the first floor.", + "B": "You put it on the computer table on the second floor.", + "C": "Behind the computer.", + "D": "You placed it on the computer table to the right of the window, not on the second-floor spot.", + "E": "You put it on the computer table in front of the chair instead of leaving it upstairs." + }, + "answer_label": "B", + "answer_text": "You put it on the computer table on the second floor.", + "q_id": "egolife_A1_D2_008_4_T2", + "correct_answer": "You put it on the computer table on the second floor.", + "original_options": { + "A": "On the first floor.", + "B": "You put it on the computer table on the second floor.", + "C": "Behind the computer.", + "D": "You placed it on the computer table to the right of the window, not on the second-floor spot.", + "E": "You put it on the computer table in front of the chair instead of leaving it upstairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_3", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the best place for me to use a computer right now?", + "answer": "Bedroom.", + "question_chinese": "我现在想要用电脑，最好去哪里？", + "answer_chinese": "卧室", + "evidence": [ + { + "description": "There are three computers in the bedroom", + "timestamps": { + "start": "00:02:15", + "end": "00:02:27" + } + } + ], + "question_timestamps": "00:06:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Living room.", + "B": "Bedroom." + }, + "answer_label": "B", + "answer_text": "Bedroom.", + "q_id": "egolife_A1_D2_008_3", + "correct_answer": "Bedroom.", + "original_options": { + "A": "Living room.", + "B": "Bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_4", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the best place for me to use a computer right now?", + "answer": "Living room on the second floor.", + "question_chinese": "我现在想要用电脑，最好去哪里？", + "answer_chinese": "二楼客厅", + "evidence": [ + { + "description": "There are three computers in the bedroom", + "timestamps": { + "start": "00:02:15", + "end": "00:02:27" + } + }, + { + "description": "There are some computers in the living room on the second floor", + "timestamps": { + "start": "00:09:00", + "end": "00:09:15" + } + }, + { + "description": "I'm in the living room on the second floor right now", + "timestamps": { + "start": "00:09:57", + "end": "00:10:10" + } + } + ], + "question_timestamps": "00:10:23", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Living room on the second floor.", + "B": "Kitchen on the first floor." + }, + "answer_label": "A", + "answer_text": "Living room on the second floor.", + "q_id": "egolife_A1_D2_008_4", + "correct_answer": "Living room on the second floor.", + "original_options": { + "A": "Living room on the second floor.", + "B": "Kitchen on the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_5_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the tape measure right now?", + "answer": "Yes.", + "question_chinese": "我现在能拿到卷尺吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The tape measure is on the table", + "timestamps": { + "start": "00:00:07", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D2_008_5_T1", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_5_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the tape measure now?", + "answer": "Yes.", + "question_chinese": "我现在能拿到卷尺吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The tape measure is on the table", + "timestamps": { + "start": "00:00:07", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D2_008_5_T2", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_6_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the tape measure right now?", + "answer": "No.", + "question_chinese": "我现在能拿到卷尺吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The tape measure is on the table", + "timestamps": { + "start": "00:00:07", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:01:36", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D2_008_6_T1", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_6_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the measuring tape now?", + "answer": "No.", + "question_chinese": "我现在能拿到卷尺吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The tape measure is on the table", + "timestamps": { + "start": "00:00:07", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:01:37", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D2_008_6_T2", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_7_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the house?", + "answer": "Six.", + "question_chinese": "房子里有几个垃圾桶？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There is a trash can under the table on the first floor", + "timestamps": { + "start": "00:03:24", + "end": "00:03:28" + } + } + ], + "question_timestamps": "00:03:41", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "Four.", + "C": "Five.", + "D": "Seven, so there are seven trash cans throughout the house in total.", + "E": "Eight, meaning you can count eight separate trash cans around the house." + }, + "answer_label": "A", + "answer_text": "Six.", + "q_id": "egolife_A1_D2_008_7_T1", + "correct_answer": "Six.", + "original_options": { + "A": "Six.", + "B": "Four.", + "C": "Five.", + "D": "Seven, so there are seven trash cans throughout the house in total.", + "E": "Eight, meaning you can count eight separate trash cans around the house." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_7_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the house?", + "answer": "One.", + "question_chinese": "房子里有几个垃圾桶？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There is a trash can under the table on the first floor", + "timestamps": { + "start": "00:03:24", + "end": "00:03:28" + } + } + ], + "question_timestamps": "00:03:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are two trash cans visible in the house overall right now.", + "E": "You can count three trash cans placed around the house during the walk-through." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "egolife_A1_D2_008_7_T2", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are two trash cans visible in the house overall right now.", + "E": "You can count three trash cans placed around the house during the walk-through." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_7_T3", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the house?", + "answer": "Six.", + "question_chinese": "房子里有几个垃圾桶？", + "answer_chinese": "四个", + "evidence": [ + { + "description": "There is a trash can under the table on the first floor", + "timestamps": { + "start": "00:03:24", + "end": "00:03:28" + } + }, + { + "description": "There is a trash can in the kitchen", + "timestamps": { + "start": "00:05:09", + "end": "00:05:12" + } + }, + { + "description": "There is a trash can under the table in the living room on the second floor", + "timestamps": { + "start": "00:08:57", + "end": "00:09:01" + } + }, + { + "description": "There is a trash can beside the windowsill on the second floor", + "timestamps": { + "start": "00:09:42", + "end": "00:09:48" + } + } + ], + "question_timestamps": "00:10:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "There are five trash cans total throughout the house in the video sequence.", + "D": "Six.", + "E": "You see seven trash cans in the house altogether across the different rooms shown." + }, + "answer_label": "D", + "answer_text": "Six.", + "q_id": "egolife_A1_D2_008_7_T3", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "There are five trash cans total throughout the house in the video sequence.", + "D": "Six.", + "E": "You see seven trash cans in the house altogether across the different rooms shown." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_7_T4", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the house?", + "answer": "Four.", + "question_chinese": "房子里有几个垃圾桶？", + "answer_chinese": "四个", + "evidence": [ + { + "description": "There is a trash can under the table on the first floor", + "timestamps": { + "start": "00:03:24", + "end": "00:03:28" + } + }, + { + "description": "There is a trash can in the kitchen", + "timestamps": { + "start": "00:05:09", + "end": "00:05:12" + } + }, + { + "description": "There is a trash can under the table in the living room on the second floor", + "timestamps": { + "start": "00:08:57", + "end": "00:09:01" + } + }, + { + "description": "There is a trash can beside the windowsill on the second floor", + "timestamps": { + "start": "00:09:42", + "end": "00:09:48" + } + } + ], + "question_timestamps": "00:10:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "There are three trash cans located in the house across the rooms shown.", + "D": "You can find five trash cans spread across the house during the walkthrough in the video.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "egolife_A1_D2_008_7_T4", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "There are three trash cans located in the house across the rooms shown.", + "D": "You can find five trash cans spread across the house during the walkthrough in the video.", + "E": "Four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_3_T1", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the mirror or the dining table?", + "answer": "Mirror.", + "question_chinese": "镜子和餐桌哪个离我更近？", + "answer_chinese": "镜子", + "evidence": [ + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:01:21", + "end": "00:01:25" + } + }, + { + "description": "The dining table is in the middle of the living room", + "timestamps": { + "start": "00:00:09", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:02:14", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Mirror.", + "B": "Dining table." + }, + "answer_label": "A", + "answer_text": "Mirror.", + "q_id": "egolife_A1_D2_008_3_T1", + "correct_answer": "Mirror.", + "original_options": { + "A": "Mirror.", + "B": "Dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_3_T2", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the mirror or the dining table?", + "answer": "Mirror.", + "question_chinese": "镜子和餐桌哪个离我更近？", + "answer_chinese": "镜子", + "evidence": [ + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:01:21", + "end": "00:01:25" + } + }, + { + "description": "The dining table is in the middle of the living room", + "timestamps": { + "start": "00:00:09", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:02:24", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Mirror.", + "B": "Dining table." + }, + "answer_label": "A", + "answer_text": "Mirror.", + "q_id": "egolife_A1_D2_008_3_T2", + "correct_answer": "Mirror.", + "original_options": { + "A": "Mirror.", + "B": "Dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_3_T3", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the mirror or the dining table?", + "answer": "Dining table.", + "question_chinese": "镜子和餐桌哪个离我更近？", + "answer_chinese": "餐桌", + "evidence": [ + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:01:21", + "end": "00:01:25" + } + }, + { + "description": "The dining table is in the middle of the living room", + "timestamps": { + "start": "00:00:09", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:05:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Mirror.", + "B": "Dining table." + }, + "answer_label": "B", + "answer_text": "Dining table.", + "q_id": "egolife_A1_D2_008_3_T3", + "correct_answer": "Dining table.", + "original_options": { + "A": "Mirror.", + "B": "Dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_008", + "video_length": "00:11:07", + "qid": "egolife_A1_D2_008_3_T4", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the mirror or the dining table?", + "answer": "Dining table.", + "question_chinese": "镜子和餐桌哪个离我更近？", + "answer_chinese": "餐桌", + "evidence": [ + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:01:21", + "end": "00:01:25" + } + }, + { + "description": "The dining table is in the middle of the living room", + "timestamps": { + "start": "00:00:09", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:05:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Mirror.", + "B": "Dining table." + }, + "answer_label": "B", + "answer_text": "Dining table.", + "q_id": "egolife_A1_D2_008_3_T4", + "correct_answer": "Dining table.", + "original_options": { + "A": "Mirror.", + "B": "Dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D2_009-mcq.json b/QAs/egolife_A1_D2_009-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5f80018780b613fcefaf7808d315323c2efcf72e --- /dev/null +++ b/QAs/egolife_A1_D2_009-mcq.json @@ -0,0 +1,805 @@ +[ + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_8", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many air conditioners are there in the house?", + "answer": "Two.", + "question_chinese": "家里一共有几个空调？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "There is an air conditioner in the bedroom", + "timestamps": { + "start": "00:01:05", + "end": "00:01:08" + } + }, + { + "description": "There is an air conditioner in the living room on the first floor", + "timestamps": { + "start": "00:01:36", + "end": "00:01:39" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Four air conditioners in total, with multiple units mounted in separate rooms across the house.", + "E": "Five air conditioners altogether, including several wall-mounted units spaced throughout different areas of the home." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "egolife_A1_D2_009_8", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Four air conditioners in total, with multiple units mounted in separate rooms across the house.", + "E": "Five air conditioners altogether, including several wall-mounted units spaced throughout different areas of the home." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_7", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many air conditioners are there in the house?", + "answer": "Three.", + "question_chinese": "家里一共有几个空调？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There is an air conditioner in the bedroom", + "timestamps": { + "start": "00:01:05", + "end": "00:01:08" + } + }, + { + "description": "There is an air conditioner in the living room on the second floor", + "timestamps": { + "start": "00:03:51", + "end": "00:04:05" + } + }, + { + "description": "There is an air conditioner in the living room on the first floor", + "timestamps": { + "start": "00:01:36", + "end": "00:01:39" + } + } + ], + "question_timestamps": "00:05:48", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Five air conditioners overall, with extra units installed in additional rooms besides the main living area.", + "D": "Six air conditioners in the house, including several wall-mounted units distributed across multiple rooms.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_009_7", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Five air conditioners overall, with extra units installed in additional rooms besides the main living area.", + "D": "Six air conditioners in the house, including several wall-mounted units distributed across multiple rooms.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_6_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "The last time I took out my phone, which direction was the windowsill from me?", + "answer": "Right rear.", + "question_chinese": "我最近一次拿出手机时，窗台在我的什么方向？", + "answer_chinese": "右后方", + "evidence": [ + { + "description": "The last time I took out my phone, the windowsill was behind me to the right", + "timestamps": { + "start": "00:02:05", + "end": "00:02:09" + } + } + ], + "question_timestamps": "00:03:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right front.", + "B": "Left rear.", + "C": "Behind you on the left side, the windowsill was positioned slightly out of your direct view.", + "D": "Right rear.", + "E": "In front of you toward the right, the windowsill sat closer to where you were facing at the time." + }, + "answer_label": "D", + "answer_text": "Right rear.", + "q_id": "egolife_A1_D2_009_6_T1", + "correct_answer": "Right rear.", + "original_options": { + "A": "Right front.", + "B": "Left rear.", + "C": "Behind you on the left side, the windowsill was positioned slightly out of your direct view.", + "D": "Right rear.", + "E": "In front of you toward the right, the windowsill sat closer to where you were facing at the time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_6_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I last took out my phone, which direction was the windowsill from me?", + "answer": "Right rear.", + "question_chinese": "我最近一次拿出手机时，窗台在我的什么方向？", + "answer_chinese": "右后方", + "evidence": [ + { + "description": "The last time I took out my phone, the windowsill was behind me to the right", + "timestamps": { + "start": "00:02:05", + "end": "00:02:09" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right front.", + "B": "Left rear.", + "C": "Behind you on the left side, the windowsill was positioned slightly out of your direct view.", + "D": "Right rear.", + "E": "In front of you toward the right, the windowsill sat closer to where you were facing at the time." + }, + "answer_label": "D", + "answer_text": "Right rear.", + "q_id": "egolife_A1_D2_009_6_T2", + "correct_answer": "Right rear.", + "original_options": { + "A": "Right front.", + "B": "Left rear.", + "C": "Behind you on the left side, the windowsill was positioned slightly out of your direct view.", + "D": "Right rear.", + "E": "In front of you toward the right, the windowsill sat closer to where you were facing at the time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_7_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "The last time I took out my phone, which direction was the windowsill from me?", + "answer": "Left.", + "question_chinese": "我最近一次拿出手机时，窗台在我的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "The last time I picked up my phone, the windowsill was on my left", + "timestamps": { + "start": "00:03:18", + "end": "00:03:26" + } + } + ], + "question_timestamps": "00:04:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right.", + "B": "Right rear.", + "C": "Left.", + "D": "Directly in front of you, the windowsill was aligned with your gaze when you took out the phone.", + "E": "Behind you to the right, the windowsill was positioned off to that side at that moment." + }, + "answer_label": "C", + "answer_text": "Left.", + "q_id": "egolife_A1_D2_009_7_T1", + "correct_answer": "Left.", + "original_options": { + "A": "Right.", + "B": "Right rear.", + "C": "Left.", + "D": "Directly in front of you, the windowsill was aligned with your gaze when you took out the phone.", + "E": "Behind you to the right, the windowsill was positioned off to that side at that moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_7_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction was the windowsill in when I last took out my phone?", + "answer": "Left.", + "question_chinese": "我最近一次拿出手机时，窗台在我的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "The last time I picked up my phone, the windowsill was on my left", + "timestamps": { + "start": "00:03:18", + "end": "00:03:26" + } + } + ], + "question_timestamps": "00:04:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right.", + "B": "In front of you.", + "C": "Left.", + "D": "The windowsill was on your right when you last pulled out your phone.", + "E": "When you last took out your phone, the windowsill was directly in front of you." + }, + "answer_label": "C", + "answer_text": "Left.", + "q_id": "egolife_A1_D2_009_7_T2", + "correct_answer": "Left.", + "original_options": { + "A": "Right.", + "B": "In front of you.", + "C": "Left.", + "D": "The windowsill was on your right when you last pulled out your phone.", + "E": "When you last took out your phone, the windowsill was directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 4, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_1_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "The last time I entered the first-floor living room, which direction was the writing board from me?", + "answer": "Right ahead.", + "question_chinese": "我最近一次进入一楼客厅时，写字板在我的什么方向？", + "answer_chinese": "正前方", + "evidence": [ + { + "description": "The last time I entered the living room, the writing board was right in front of me", + "timestamps": { + "start": "00:01:12", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Left.", + "B": "Behind you.", + "C": "Straight ahead and slightly to your left as you entered the room.", + "D": "Right ahead.", + "E": "To your right, but set back behind the sofa when you walked in." + }, + "answer_label": "D", + "answer_text": "Right ahead.", + "q_id": "egolife_A1_D2_009_1_T1", + "correct_answer": "Right ahead.", + "original_options": { + "A": "Left.", + "B": "Behind you.", + "C": "Straight ahead and slightly to your left as you entered the room.", + "D": "Right ahead.", + "E": "To your right, but set back behind the sofa when you walked in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_1_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I last entered the living room on the first floor, which direction was the writing board from me?", + "answer": "Right ahead.", + "question_chinese": "我最近一次进入一楼客厅时，写字板在我的什么方向？", + "answer_chinese": "正前方", + "evidence": [ + { + "description": "The last time I entered the living room, the writing board was right in front of me", + "timestamps": { + "start": "00:01:12", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Left.", + "B": "Behind you.", + "C": "Straight ahead and slightly to your left as you entered the room.", + "D": "Right ahead.", + "E": "To your right, but set back behind the sofa when you walked in." + }, + "answer_label": "D", + "answer_text": "Right ahead.", + "q_id": "egolife_A1_D2_009_1_T2", + "correct_answer": "Right ahead.", + "original_options": { + "A": "Left.", + "B": "Behind you.", + "C": "Straight ahead and slightly to your left as you entered the room.", + "D": "Right ahead.", + "E": "To your right, but set back behind the sofa when you walked in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_2_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "The last time I entered the first-floor living room, which direction was the writing board from me?", + "answer": "Left.", + "question_chinese": "我最近一次进入一楼客厅时，写字板在我的什么方向？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The last time I entered the living room on the first floor, the writing board was on my left", + "timestamps": { + "start": "00:08:54", + "end": "00:08:57" + } + } + ], + "question_timestamps": "00:09:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right.", + "B": "Straight ahead.", + "C": "Left.", + "D": "Behind you when you first stepped into the living room that time.", + "E": "Ahead of you and a bit to the right as you entered the living room again." + }, + "answer_label": "C", + "answer_text": "Left.", + "q_id": "egolife_A1_D2_009_2_T1", + "correct_answer": "Left.", + "original_options": { + "A": "Right.", + "B": "Straight ahead.", + "C": "Left.", + "D": "Behind you when you first stepped into the living room that time.", + "E": "Ahead of you and a bit to the right as you entered the living room again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_2_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "The last time I entered the living room on the first floor, which direction was the writing board in relation to me?", + "answer": "Left.", + "question_chinese": "我最近一次进入一楼客厅时，写字板在我的什么方向？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The last time I entered the living room on the first floor, the writing board was on my left", + "timestamps": { + "start": "00:08:54", + "end": "00:08:57" + } + } + ], + "question_timestamps": "00:09:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right.", + "B": "Straight ahead.", + "C": "Left.", + "D": "Behind you when you first stepped into the living room that time.", + "E": "Ahead of you and a bit to the right as you entered the living room again." + }, + "answer_label": "C", + "answer_text": "Left.", + "q_id": "egolife_A1_D2_009_2_T2", + "correct_answer": "Left.", + "original_options": { + "A": "Right.", + "B": "Straight ahead.", + "C": "Left.", + "D": "Behind you when you first stepped into the living room that time.", + "E": "Ahead of you and a bit to the right as you entered the living room again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_3_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to use the hair dryer now—how should I walk to get to it?", + "answer": "Turn backward to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, turn left and pass through the living room to reach the bedroom.", + "question_chinese": "我现在想要使用吹风机，应该怎么走？", + "answer_chinese": "向后转到达楼梯口，然后右转下楼，到达一楼后左转并穿过客厅到达卧室", + "evidence": [ + { + "description": "The hair dryer is in the bedroom on the first floor", + "timestamps": { + "start": "00:01:07", + "end": "00:01:14" + } + }, + { + "description": "To go from the living room on the first floor to the living room on the second floor, you need to first enter the staircase, then turn right and go upstairs. After reaching the second floor, turn left", + "timestamps": { + "start": "00:01:36", + "end": "00:01:57" + } + } + ], + "question_timestamps": "00:02:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn backward to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, turn left and pass through the living room to reach the bedroom.", + "B": "Go upstairs to the study.", + "C": "Turn left and head downstairs.", + "D": "Turn backward to the stairs, go downstairs, then turn right into the bathroom to find it.", + "E": "Turn around to the staircase, go downstairs, then pass the dining area to reach the guest bedroom." + }, + "answer_label": "A", + "answer_text": "Turn backward to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, turn left and pass through the living room to reach the bedroom.", + "q_id": "egolife_A1_D2_009_3_T1", + "correct_answer": "Turn backward to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, turn left and pass through the living room to reach the bedroom.", + "original_options": { + "A": "Turn backward to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, turn left and pass through the living room to reach the bedroom.", + "B": "Go upstairs to the study.", + "C": "Turn left and head downstairs.", + "D": "Turn backward to the stairs, go downstairs, then turn right into the bathroom to find it.", + "E": "Turn around to the staircase, go downstairs, then pass the dining area to reach the guest bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_3_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the hairdryer?", + "answer": "Turn backward to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, turn left and pass through the living room to reach the bedroom.", + "question_chinese": "我现在想要使用吹风机，应该怎么走？", + "answer_chinese": "向后转到达楼梯口，然后右转下楼，到达一楼后左转并穿过客厅到达卧室", + "evidence": [ + { + "description": "The hair dryer is in the bedroom on the first floor", + "timestamps": { + "start": "00:01:07", + "end": "00:01:14" + } + }, + { + "description": "To go from the living room on the first floor to the living room on the second floor, you need to first enter the staircase, then turn right and go upstairs. After reaching the second floor, turn left", + "timestamps": { + "start": "00:01:36", + "end": "00:01:57" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn backward to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, turn left and pass through the living room to reach the bedroom.", + "B": "Go upstairs to the study.", + "C": "Turn left and head downstairs.", + "D": "Turn backward to the stairs, go downstairs, then turn right into the bathroom to find it.", + "E": "Turn around to the staircase, go downstairs, then pass the dining area to reach the guest bedroom." + }, + "answer_label": "A", + "answer_text": "Turn backward to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, turn left and pass through the living room to reach the bedroom.", + "q_id": "egolife_A1_D2_009_3_T2", + "correct_answer": "Turn backward to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, turn left and pass through the living room to reach the bedroom.", + "original_options": { + "A": "Turn backward to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, turn left and pass through the living room to reach the bedroom.", + "B": "Go upstairs to the study.", + "C": "Turn left and head downstairs.", + "D": "Turn backward to the stairs, go downstairs, then turn right into the bathroom to find it.", + "E": "Turn around to the staircase, go downstairs, then pass the dining area to reach the guest bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_4_T1", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I want to use the hair dryer now—how should I walk to get to it?", + "answer": "Turn back into the bedroom.", + "question_chinese": "我现在想要使用吹风机，应该怎么走？", + "answer_chinese": "向后转进入卧室", + "evidence": [ + { + "description": "The hair dryer is in the bedroom on the first floor", + "timestamps": { + "start": "00:01:07", + "end": "00:01:14" + } + }, + { + "description": "To go from the living room on the first floor to the living room on the second floor, you need to first enter the staircase, then turn right and go upstairs. After reaching the second floor, turn left", + "timestamps": { + "start": "00:01:36", + "end": "00:01:57" + } + } + ], + "question_timestamps": "00:09:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left into the bathroom.", + "B": "Turn back into the bedroom.", + "C": "Turn right into the kitchen.", + "D": "Go straight ahead into the garage rather than turning back into the bedroom.", + "E": "Turn around and walk toward the dining room to look for it." + }, + "answer_label": "B", + "answer_text": "Turn back into the bedroom.", + "q_id": "egolife_A1_D2_009_4_T1", + "correct_answer": "Turn back into the bedroom.", + "original_options": { + "A": "Turn left into the bathroom.", + "B": "Turn back into the bedroom.", + "C": "Turn right into the kitchen.", + "D": "Go straight ahead into the garage rather than turning back into the bedroom.", + "E": "Turn around and walk toward the dining room to look for it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_009_4_T2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to where I can use a hairdryer?", + "answer": "Turn back into the bedroom.", + "question_chinese": "我现在想要使用吹风机，应该怎么走？", + "answer_chinese": "向后转进入卧室", + "evidence": [ + { + "description": "The hair dryer is in the bedroom on the first floor", + "timestamps": { + "start": "00:01:07", + "end": "00:01:14" + } + }, + { + "description": "To go from the living room on the first floor to the living room on the second floor, you need to first enter the staircase, then turn right and go upstairs. After reaching the second floor, turn left", + "timestamps": { + "start": "00:01:36", + "end": "00:01:57" + } + } + ], + "question_timestamps": "00:09:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left into the bathroom.", + "B": "Turn back into the bedroom.", + "C": "Turn right into the kitchen.", + "D": "Turn straight ahead into the garage to find the hairdryer for use there.", + "E": "Turn around and head toward the dining room area to use the hairdryer." + }, + "answer_label": "B", + "answer_text": "Turn back into the bedroom.", + "q_id": "egolife_A1_D2_009_4_T2", + "correct_answer": "Turn back into the bedroom.", + "original_options": { + "A": "Turn left into the bathroom.", + "B": "Turn back into the bedroom.", + "C": "Turn right into the kitchen.", + "D": "Turn straight ahead into the garage to find the hairdryer for use there.", + "E": "Turn around and head toward the dining room area to use the hairdryer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D2_010-mcq.json b/QAs/egolife_A1_D2_010-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d6493fb570880ed906f44887eacf2037134dd2c7 --- /dev/null +++ b/QAs/egolife_A1_D2_010-mcq.json @@ -0,0 +1,1076 @@ +[ + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_2_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are on my table?", + "answer": "Three.", + "question_chinese": "我桌上有几台电脑?", + "answer_chinese": "三台", + "evidence": [ + { + "description": "There are three computers on the table", + "timestamps": { + "start": "00:04:53", + "end": "00:05:02" + } + } + ], + "question_timestamps": "00:05:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Four computers are on your table, arranged across the surface in separate spots.", + "E": "Five computers are on your table, with several clustered together and one set apart." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_010_2_T1", + "correct_answer": "Three.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Four computers are on your table, arranged across the surface in separate spots.", + "E": "Five computers are on your table, with several clustered together and one set apart." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_2_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are on my table?", + "answer": "Three, located to your front-right and behind you.", + "question_chinese": "我桌上有几台电脑?", + "answer_chinese": "三台，分别你的右前方和后方。", + "evidence": [ + { + "description": "There are three computers on the table", + "timestamps": { + "start": "00:04:53", + "end": "00:05:02" + } + } + ], + "question_timestamps": "00:05:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One computer.", + "B": "Two computers.", + "C": "Three, located to your front-right and behind you.", + "D": "Four computers, with one to your front-right and the rest positioned behind you.", + "E": "Five computers, spread between your front-right side and behind you on the table." + }, + "answer_label": "C", + "answer_text": "Three, located to your front-right and behind you.", + "q_id": "egolife_A1_D2_010_2_T2", + "correct_answer": "Three, located to your front-right and behind you.", + "original_options": { + "A": "One computer.", + "B": "Two computers.", + "C": "Three, located to your front-right and behind you.", + "D": "Four computers, with one to your front-right and the rest positioned behind you.", + "E": "Five computers, spread between your front-right side and behind you on the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_2_T3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are on my table?", + "answer": "One, to your right.", + "question_chinese": "我桌上有几台电脑?", + "answer_chinese": "一台,在你的右边。", + "evidence": [ + { + "description": "There are one computers on the table", + "timestamps": { + "start": "00:10:13", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:11:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero, to your right.", + "B": "Two, to your right.", + "C": "Three computers are on your table, all positioned to your right side.", + "D": "One, to your right.", + "E": "Four computers sit on your table, grouped together on the right-hand side." + }, + "answer_label": "D", + "answer_text": "One, to your right.", + "q_id": "egolife_A1_D2_010_2_T3", + "correct_answer": "One, to your right.", + "original_options": { + "A": "Zero, to your right.", + "B": "Two, to your right.", + "C": "Three computers are on your table, all positioned to your right side.", + "D": "One, to your right.", + "E": "Four computers sit on your table, grouped together on the right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_8_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of water are on my table?", + "answer": "One bottle, in front of you.", + "question_chinese": "我桌子上有几瓶水？", + "answer_chinese": "一瓶，在你的前方", + "evidence": [ + { + "description": "There is a bottle of water on the table", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero bottles.", + "B": "Two bottles.", + "C": "One bottle, in front of you.", + "D": "Three bottles of water are on your table, lined up in front of you.", + "E": "Four bottles of water are on your table, clustered together directly in front of you." + }, + "answer_label": "C", + "answer_text": "One bottle, in front of you.", + "q_id": "egolife_A1_D2_010_8_T1", + "correct_answer": "One bottle, in front of you.", + "original_options": { + "A": "Zero bottles.", + "B": "Two bottles.", + "C": "One bottle, in front of you.", + "D": "Three bottles of water are on your table, lined up in front of you.", + "E": "Four bottles of water are on your table, clustered together directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_8_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of water are on my table?", + "answer": "One bottle, to your left.", + "question_chinese": "我桌子上有几瓶水？", + "answer_chinese": "一瓶，在你的左方。", + "evidence": [ + { + "description": "There is a bottle of water on the table", + "timestamps": { + "start": "00:01:03", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:52", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero bottles.", + "B": "Two bottles.", + "C": "Three bottles of water are on your table, grouped together to your left.", + "D": "One bottle, to your left.", + "E": "Four bottles of water are on your table, spread out along the left side." + }, + "answer_label": "D", + "answer_text": "One bottle, to your left.", + "q_id": "egolife_A1_D2_010_8_T2", + "correct_answer": "One bottle, to your left.", + "original_options": { + "A": "Zero bottles.", + "B": "Two bottles.", + "C": "Three bottles of water are on your table, grouped together to your left.", + "D": "One bottle, to your left.", + "E": "Four bottles of water are on your table, spread out along the left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_9", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the comb now?", + "answer": "No,I can't.", + "question_chinese": "我现在够得着梳子吗？", + "answer_chinese": "够不着", + "evidence": [ + { + "description": "The comb is at the right corner of the table, which is too far away", + "timestamps": { + "start": "00:10:12", + "end": "00:10:18" + } + } + ], + "question_timestamps": "00:10:25", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,I can't.", + "B": "Yes, I can." + }, + "answer_label": "A", + "answer_text": "No,I can't.", + "q_id": "egolife_A1_D2_010_9", + "correct_answer": "No,I can't.", + "original_options": { + "A": "No,I can't.", + "B": "Yes, I can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_10", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the comb now?", + "answer": "Yes,I can.", + "question_chinese": "我现在够得着梳子吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The comb is at the corner of the table, and I'm at the corner of the table right now", + "timestamps": { + "start": "00:11:12", + "end": "00:11:15" + } + } + ], + "question_timestamps": "00:11:29", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, I cannot.", + "B": "Yes,I can." + }, + "answer_label": "B", + "answer_text": "Yes,I can.", + "q_id": "egolife_A1_D2_010_10", + "correct_answer": "Yes,I can.", + "original_options": { + "A": "No, I cannot.", + "B": "Yes,I can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_5_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where did I see most of the computers mainly located?", + "answer": "On the sofa.", + "question_chinese": "我看到大部分电脑主要在什么地方？", + "answer_chinese": "沙发上", + "evidence": [ + { + "description": "There is a computer on the table and two computers on the sofa", + "timestamps": { + "start": "00:09:09", + "end": "00:09:12" + } + } + ], + "question_timestamps": "00:10:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the shelf.", + "B": "On the floor.", + "C": "On the sofa.", + "D": "In front of the sofa, not on it, with the computers clustered near the edge.", + "E": "Behind the sofa rather than on it, where the computers were mainly piled together." + }, + "answer_label": "C", + "answer_text": "On the sofa.", + "q_id": "egolife_A1_D2_010_5_T1", + "correct_answer": "On the sofa.", + "original_options": { + "A": "On the shelf.", + "B": "On the floor.", + "C": "On the sofa.", + "D": "In front of the sofa, not on it, with the computers clustered near the edge.", + "E": "Behind the sofa rather than on it, where the computers were mainly piled together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_5_T3", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where did I see most of the computers mainly located?", + "answer": "On the sofa to your back-left.", + "question_chinese": "我看到大部分电脑主要在什么地方？", + "answer_chinese": "你左后方的沙发上", + "evidence": [ + { + "description": "There is a computer on the table and two computers on the sofa", + "timestamps": { + "start": "00:09:09", + "end": "00:09:12" + } + } + ], + "question_timestamps": "00:10:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the sofa to your back-right.", + "B": "On the sofa to your back-left.", + "C": "On the sofa directly behind you.", + "D": "On the sofa to your front-left, where most of the computers were grouped together.", + "E": "On the sofa to your front-right, with the majority of the computers sitting there." + }, + "answer_label": "B", + "answer_text": "On the sofa to your back-left.", + "q_id": "egolife_A1_D2_010_5_T3", + "correct_answer": "On the sofa to your back-left.", + "original_options": { + "A": "On the sofa to your back-right.", + "B": "On the sofa to your back-left.", + "C": "On the sofa directly behind you.", + "D": "On the sofa to your front-left, where most of the computers were grouped together.", + "E": "On the sofa to your front-right, with the majority of the computers sitting there." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_5_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where did I see most of the computers mainly located?", + "answer": "On the sofa to your front-right.", + "question_chinese": "我看到大部分电脑主要在什么地方？", + "answer_chinese": "你右前方的沙发上", + "evidence": [ + { + "description": "There is a computer on the table and two computers on the sofa", + "timestamps": { + "start": "00:09:09", + "end": "00:09:12" + } + } + ], + "question_timestamps": "00:10:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the sofa to your front-right.", + "B": "On the sofa to your front-left.", + "C": "On the sofa to your back-right.", + "D": "On the sofa to your back-left, where most of the computers were mainly placed.", + "E": "On the sofa directly in front of you, with the computers concentrated in that spot." + }, + "answer_label": "A", + "answer_text": "On the sofa to your front-right.", + "q_id": "egolife_A1_D2_010_5_T2", + "correct_answer": "On the sofa to your front-right.", + "original_options": { + "A": "On the sofa to your front-right.", + "B": "On the sofa to your front-left.", + "C": "On the sofa to your back-right.", + "D": "On the sofa to your back-left, where most of the computers were mainly placed.", + "E": "On the sofa directly in front of you, with the computers concentrated in that spot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_3_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are on the table?", + "answer": "One.", + "question_chinese": "桌上有几台电脑?", + "answer_chinese": "一台", + "evidence": [ + { + "description": "There are one computers on the table", + "timestamps": { + "start": "00:10:13", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:10:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "There are four computers on the table, spread out rather than stacked together.", + "E": "There are five computers on the table, with multiple units visible at once." + }, + "answer_label": "B", + "answer_text": "One.", + "q_id": "egolife_A1_D2_010_3_T1", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "There are four computers on the table, spread out rather than stacked together.", + "E": "There are five computers on the table, with multiple units visible at once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_3_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are on the table?", + "answer": "Five.", + "question_chinese": "桌上有几台电脑?", + "answer_chinese": "一台", + "evidence": [ + { + "description": "There are one computers on the table", + "timestamps": { + "start": "00:10:13", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:10:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "One.", + "C": "Two.", + "D": "There are three computers on the table, not as many as a full set of five.", + "E": "There are four computers on the table, with one fewer than the maximum shown." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D2_010_3_T2", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "One.", + "C": "Two.", + "D": "There are three computers on the table, not as many as a full set of five.", + "E": "There are four computers on the table, with one fewer than the maximum shown." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_11", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls of rice are on the table?", + "answer": "Zero.", + "question_chinese": "桌上有几碗米饭？", + "answer_chinese": "零碗", + "evidence": [ + { + "description": "There is no rice on the table.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:04:09", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "There are three bowls of rice on the table, arranged close together in a small group.", + "E": "There are four bowls of rice on the table, making a noticeable cluster across the surface." + }, + "answer_label": "A", + "answer_text": "Zero.", + "q_id": "egolife_A1_D2_010_11", + "correct_answer": "Zero.", + "original_options": { + "A": "Zero.", + "B": "One.", + "C": "Two.", + "D": "There are three bowls of rice on the table, arranged close together in a small group.", + "E": "There are four bowls of rice on the table, making a noticeable cluster across the surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_12", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bowls of rice are on the table?", + "answer": "There are six bowls of rice on the table.", + "question_chinese": "桌上有几碗米饭？", + "answer_chinese": "六碗", + "evidence": [ + { + "description": "There are six bowls of rice on the table", + "timestamps": { + "start": "00:09:37", + "end": "00:09:41" + } + } + ], + "question_timestamps": "00:10:31", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are six bowls of rice on the table.", + "B": "Four bowls.", + "C": "Five bowls.", + "D": "There are seven bowls of rice on the table, with one more bowl than six.", + "E": "There are eight bowls of rice on the table, filling more of the tabletop space." + }, + "answer_label": "A", + "answer_text": "There are six bowls of rice on the table.", + "q_id": "egolife_A1_D2_010_12", + "correct_answer": "There are six bowls of rice on the table.", + "original_options": { + "A": "There are six bowls of rice on the table.", + "B": "Four bowls.", + "C": "Five bowls.", + "D": "There are seven bowls of rice on the table, with one more bowl than six.", + "E": "There are eight bowls of rice on the table, filling more of the tabletop space." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_1_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of water are on the table?", + "answer": "One.", + "question_chinese": "桌子上有几瓶水？", + "answer_chinese": "一瓶", + "evidence": [ + { + "description": "There is a bottle of water on the table", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are four bottles of water on the table, lined up rather than standing alone.", + "E": "There are five bottles of water on the table, taking up a large portion of the surface." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "egolife_A1_D2_010_1_T1", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are four bottles of water on the table, lined up rather than standing alone.", + "E": "There are five bottles of water on the table, taking up a large portion of the surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_1_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of water are on the table?", + "answer": "One.", + "question_chinese": "桌子上有几瓶水？", + "answer_chinese": "一瓶", + "evidence": [ + { + "description": "There is a bottle of water on the table", + "timestamps": { + "start": "00:01:03", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:29", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four bottles of water are on the table, grouped together near the center.", + "E": "Five bottles of water are on the table, spread out across the tabletop." + }, + "answer_label": "A", + "answer_text": "One.", + "q_id": "egolife_A1_D2_010_1_T2", + "correct_answer": "One.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four bottles of water are on the table, grouped together near the center.", + "E": "Five bottles of water are on the table, spread out across the tabletop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_5", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What is the position of the sofa relative to me?", + "answer": "Left rear.", + "question_chinese": "沙发在相对于我的什么位置？", + "answer_chinese": "左后方", + "evidence": [ + { + "description": "The sofa is behind me to the left", + "timestamps": { + "start": "00:02:39", + "end": "00:02:41" + } + } + ], + "question_timestamps": "00:07:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Front left.", + "B": "Right rear.", + "C": "The sofa is directly in front of you, slightly offset to your left side.", + "D": "Left rear.", + "E": "The sofa sits to your right side, a bit behind your shoulder line." + }, + "answer_label": "D", + "answer_text": "Left rear.", + "q_id": "egolife_A1_D2_010_5", + "correct_answer": "Left rear.", + "original_options": { + "A": "Front left.", + "B": "Right rear.", + "C": "The sofa is directly in front of you, slightly offset to your left side.", + "D": "Left rear.", + "E": "The sofa sits to your right side, a bit behind your shoulder line." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_6", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What is the position of the sofa relative to me?", + "answer": "Right astern.", + "question_chinese": "沙发在相对于我的什么位置？", + "answer_chinese": "正后方", + "evidence": [ + { + "description": "The sofa is right behind me", + "timestamps": { + "start": "00:10:19", + "end": "00:10:25" + } + } + ], + "question_timestamps": "00:10:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right astern.", + "B": "Left astern.", + "C": "Right ahead.", + "D": "The sofa is off your port quarter, behind you rather than on your right side.", + "E": "The sofa is forward of you on the starboard side, not behind your position." + }, + "answer_label": "A", + "answer_text": "Right astern.", + "q_id": "egolife_A1_D2_010_6", + "correct_answer": "Right astern.", + "original_options": { + "A": "Right astern.", + "B": "Left astern.", + "C": "Right ahead.", + "D": "The sofa is off your port quarter, behind you rather than on your right side.", + "E": "The sofa is forward of you on the starboard side, not behind your position." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_11_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the doll or the potted plant?", + "answer": "The doll is to your left.", + "question_chinese": "玩偶和盆栽哪个离我更近？", + "answer_chinese": "玩偶，在你的左边。", + "evidence": [ + { + "description": "The doll is on the sofa", + "timestamps": { + "start": "00:09:09", + "end": "00:09:11" + } + }, + { + "description": "The potted plant is in the corner.", + "timestamps": { + "start": "00:09:44", + "end": "00:09:48" + } + } + ], + "question_timestamps": "00:10:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The doll is to your left.", + "B": "The potted plant is to your right." + }, + "answer_label": "A", + "answer_text": "The doll is to your left.", + "q_id": "egolife_A1_D2_010_11_T1", + "correct_answer": "The doll is to your left.", + "original_options": { + "A": "The doll is to your left.", + "B": "The potted plant is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D2_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_010_11_T2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the doll or the potted plant?", + "answer": "The potted plant is behind you.", + "question_chinese": "玩偶和盆栽哪个离我更近？", + "answer_chinese": "盆栽，在你的身后", + "evidence": [ + { + "description": "The doll is on the sofa", + "timestamps": { + "start": "00:09:09", + "end": "00:09:11" + } + }, + { + "description": "The potted plant is in the corner.", + "timestamps": { + "start": "00:09:44", + "end": "00:09:48" + } + } + ], + "question_timestamps": "00:11:26", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The doll is to your left.", + "B": "The potted plant is behind you." + }, + "answer_label": "B", + "answer_text": "The potted plant is behind you.", + "q_id": "egolife_A1_D2_010_11_T2", + "correct_answer": "The potted plant is behind you.", + "original_options": { + "A": "The doll is to your left.", + "B": "The potted plant is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/egolife_A1_D2_011-mcq.json b/QAs/egolife_A1_D2_011-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a4ba9d2dfc278204ae8995cad969459de7c0d005 --- /dev/null +++ b/QAs/egolife_A1_D2_011-mcq.json @@ -0,0 +1,594 @@ +[ + { + "video_uid": "egolife_A1_D2_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_011_1_T3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the refrigerator relative to me?", + "answer": "The refrigerator is in front of you.", + "question_chinese": "冰箱在相对于我的什么位置？", + "answer_chinese": "冰箱在你的正前方。", + "evidence": [ + { + "description": "The refrigerator is in front of you.", + "timestamps": { + "start": "00:04:29", + "end": "00:04:31" + } + } + ], + "question_timestamps": "00:04:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The refrigerator is in front of you.", + "D": "The refrigerator is to your right side, positioned beside you rather than ahead.", + "E": "The refrigerator is behind you, so you would need to turn around to face it." + }, + "answer_label": "C", + "answer_text": "The refrigerator is in front of you.", + "q_id": "egolife_A1_D2_011_1_T3", + "correct_answer": "The refrigerator is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The refrigerator is in front of you.", + "D": "The refrigerator is to your right side, positioned beside you rather than ahead.", + "E": "The refrigerator is behind you, so you would need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_011_1_T4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the refrigerator relative to me?", + "answer": "The refrigerator is behind you.", + "question_chinese": "冰箱在相对于我的什么位置？", + "answer_chinese": "冰箱在你的后面。", + "evidence": [ + { + "description": "The refrigerator is behind you.", + "timestamps": { + "start": "00:04:32", + "end": "00:04:46" + } + } + ], + "question_timestamps": "00:04:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The refrigerator is in front of you.", + "B": "The refrigerator is to your left.", + "C": "The refrigerator is to your right, near your shoulder as you face forward.", + "D": "The refrigerator is above you, positioned overhead relative to where you are standing.", + "E": "The refrigerator is behind you." + }, + "answer_label": "E", + "answer_text": "The refrigerator is behind you.", + "q_id": "egolife_A1_D2_011_1_T4", + "correct_answer": "The refrigerator is behind you.", + "original_options": { + "A": "The refrigerator is in front of you.", + "B": "The refrigerator is to your left.", + "C": "The refrigerator is to your right, near your shoulder as you face forward.", + "D": "The refrigerator is above you, positioned overhead relative to where you are standing.", + "E": "The refrigerator is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_011_3", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to use the refrigerator, how should I get there?", + "answer": "Turn backward and move forward to the right front.", + "question_chinese": "如果我想要用冰箱，我应该怎么走？", + "answer_chinese": "向后转并向右前方前进", + "evidence": [ + { + "description": "The refrigerator is at the door of the staircase", + "timestamps": { + "start": "00:03:42", + "end": "00:03:45" + } + }, + { + "description": "The table and the refrigerator are respectively on both sides of the staircase door", + "timestamps": { + "start": "00:04:24", + "end": "00:04:26" + } + } + ], + "question_timestamps": "00:04:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn backward and go left front.", + "B": "Turn around and go straight ahead.", + "C": "Turn backward, then move forward toward the left front corner instead of angling to the right.", + "D": "Turn backward and move forward to the right front.", + "E": "Turn backward and walk forward to the left back area, not toward the right-front side." + }, + "answer_label": "D", + "answer_text": "Turn backward and move forward to the right front.", + "q_id": "egolife_A1_D2_011_3", + "correct_answer": "Turn backward and move forward to the right front.", + "original_options": { + "A": "Turn backward and go left front.", + "B": "Turn around and go straight ahead.", + "C": "Turn backward, then move forward toward the left front corner instead of angling to the right.", + "D": "Turn backward and move forward to the right front.", + "E": "Turn backward and walk forward to the left back area, not toward the right-front side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_011_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to use the refrigerator, how should I get there?", + "answer": "Go straight to the staircase entrance, then turn right after going upstairs.", + "question_chinese": "如果我想要用冰箱，我应该怎么走？", + "answer_chinese": "直行到达楼梯口，上楼后右转", + "evidence": [ + { + "description": "The refrigerator is at the door of the staircase", + "timestamps": { + "start": "00:03:42", + "end": "00:03:45" + } + } + ], + "question_timestamps": "00:09:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left after going upstairs.", + "B": "Turn right before going upstairs.", + "C": "Go straight to the staircase entrance, then turn right after going upstairs.", + "D": "Go straight to the staircase entrance, then turn left immediately before you start going upstairs.", + "E": "Go straight to the staircase entrance, go downstairs instead, then turn around at the bottom." + }, + "answer_label": "C", + "answer_text": "Go straight to the staircase entrance, then turn right after going upstairs.", + "q_id": "egolife_A1_D2_011_4", + "correct_answer": "Go straight to the staircase entrance, then turn right after going upstairs.", + "original_options": { + "A": "Turn left after going upstairs.", + "B": "Turn right before going upstairs.", + "C": "Go straight to the staircase entrance, then turn right after going upstairs.", + "D": "Go straight to the staircase entrance, then turn left immediately before you start going upstairs.", + "E": "Go straight to the staircase entrance, go downstairs instead, then turn around at the bottom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_011_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the living room?", + "answer": "One.", + "question_chinese": "客厅里有几个垃圾桶？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There is a trash can under the table", + "timestamps": { + "start": "00:01:12", + "end": "00:01:19" + } + } + ], + "question_timestamps": "00:02:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "There are two trash cans placed in the living room, so the total count is two.", + "D": "There are four trash cans in the living room, making the total number of bins four.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_D2_011_1", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "There are two trash cans placed in the living room, so the total count is two.", + "D": "There are four trash cans in the living room, making the total number of bins four.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_011_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the living room?", + "answer": "Two.", + "question_chinese": "客厅里有几个垃圾桶？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "There is a trash can under the table", + "timestamps": { + "start": "00:01:12", + "end": "00:01:19" + } + }, + { + "description": "There is a trash can by the windowsill", + "timestamps": { + "start": "00:03:36", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:05:48", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "There is only one trash can in the living room, so the count is one.", + "E": "There are four trash cans in the living room, so the total number present is four." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "egolife_A1_D2_011_2", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "There is only one trash can in the living room, so the count is one.", + "E": "There are four trash cans in the living room, so the total number present is four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_011_9", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my recent route?", + "answer": "I took a few steps straight ahead from the staircase, then slightly turned left and walked forward all the way to the table.", + "question_chinese": "我刚刚的移动路线是怎样的？", + "answer_chinese": "我从楼梯处向前走了几步，然后稍微向左转，一直走到桌子前。", + "evidence": [ + { + "description": "I took a few steps straight ahead from the staircase, then slightly turned left and walked forward all the way to the table", + "timestamps": { + "start": "00:04:42", + "end": "00:04:51" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "I turned right toward the table.", + "B": "I took a few steps straight ahead from the staircase, then slightly turned left and walked forward all the way to the table.", + "C": "I walked back from the staircase.", + "D": "I took a few steps straight from the staircase, then slightly turned right and continued forward to the table.", + "E": "I walked straight ahead from the staircase all the way to the table without making any left turn." + }, + "answer_label": "B", + "answer_text": "I took a few steps straight ahead from the staircase, then slightly turned left and walked forward all the way to the table.", + "q_id": "egolife_A1_D2_011_9", + "correct_answer": "I took a few steps straight ahead from the staircase, then slightly turned left and walked forward all the way to the table.", + "original_options": { + "A": "I turned right toward the table.", + "B": "I took a few steps straight ahead from the staircase, then slightly turned left and walked forward all the way to the table.", + "C": "I walked back from the staircase.", + "D": "I took a few steps straight from the staircase, then slightly turned right and continued forward to the table.", + "E": "I walked straight ahead from the staircase all the way to the table without making any left turn." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_011_10", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my recent route?", + "answer": "I went downstairs from the staircase, then turned right and walked straight to the table.", + "question_chinese": "我刚刚的移动路线是怎样的？", + "answer_chinese": "我从楼梯口下楼，然后右转并直行到桌子前", + "evidence": [ + { + "description": "I went downstairs from the staircase, then turned right and walked straight to the table", + "timestamps": { + "start": "00:05:40", + "end": "00:05:52" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "I went downstairs from the staircase, then turned right and walked straight to the table.", + "B": "I turned left to the table.", + "C": "I went upstairs, then turned right.", + "D": "I went downstairs from the staircase, then turned left and walked straight until I reached the table.", + "E": "I went downstairs from the staircase, then turned right and walked straight to the sofa instead of the table." + }, + "answer_label": "A", + "answer_text": "I went downstairs from the staircase, then turned right and walked straight to the table.", + "q_id": "egolife_A1_D2_011_10", + "correct_answer": "I went downstairs from the staircase, then turned right and walked straight to the table.", + "original_options": { + "A": "I went downstairs from the staircase, then turned right and walked straight to the table.", + "B": "I turned left to the table.", + "C": "I went upstairs, then turned right.", + "D": "I went downstairs from the staircase, then turned left and walked straight until I reached the table.", + "E": "I went downstairs from the staircase, then turned right and walked straight to the sofa instead of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_011_5", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the lion plush toy or the refrigerator?", + "answer": "Refrigerator.", + "question_chinese": "狮子玩偶和冰箱哪个离我更近", + "answer_chinese": "冰箱", + "evidence": [ + { + "description": "The lion doll is on the sofa", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "The refrigerator is at the door of the staircase", + "timestamps": { + "start": "00:04:25", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:04:43", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Lion plush toy.", + "B": "Refrigerator." + }, + "answer_label": "B", + "answer_text": "Refrigerator.", + "q_id": "egolife_A1_D2_011_5", + "correct_answer": "Refrigerator.", + "original_options": { + "A": "Lion plush toy.", + "B": "Refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D2_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_011_6", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the lion plush toy or the refrigerator?", + "answer": "Lion doll.", + "question_chinese": "狮子玩偶和冰箱哪个离我更近", + "answer_chinese": "狮子玩偶", + "evidence": [ + { + "description": "The lion doll is on the sofa", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "The refrigerator is at the door of the staircase", + "timestamps": { + "start": "00:04:25", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Lion doll.", + "B": "Refrigerator." + }, + "answer_label": "A", + "answer_text": "Lion doll.", + "q_id": "egolife_A1_D2_011_6", + "correct_answer": "Lion doll.", + "original_options": { + "A": "Lion doll.", + "B": "Refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D2_014-mcq.json b/QAs/egolife_A1_D2_014-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..998d3e7f0269bc8edf011b7ae7a3e2caae542ec8 --- /dev/null +++ b/QAs/egolife_A1_D2_014-mcq.json @@ -0,0 +1,439 @@ +[ + { + "video_uid": "egolife_A1_D2_014", + "video_length": "00:11:24", + "qid": "egolife_A1_D2_014_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were just on the sofa?", + "answer": "Two.", + "question_chinese": "刚刚沙发上有几个人？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "Just now there were two people lying on the sofa", + "timestamps": { + "start": "00:01:15", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four people were sitting on the sofa just a moment ago in the living room.", + "E": "Five people were sitting together on the sofa just before I looked over again." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "egolife_A1_D2_014_1", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four people were sitting on the sofa just a moment ago in the living room.", + "E": "Five people were sitting together on the sofa just before I looked over again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_014", + "video_length": "00:11:24", + "qid": "egolife_A1_D2_014_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were just on the sofa?", + "answer": "Zero.", + "question_chinese": "刚刚沙发上有几个人？", + "answer_chinese": "0个", + "evidence": [ + { + "description": "There was no one on the sofa just now", + "timestamps": { + "start": "00:05:44", + "end": "00:05:55" + } + } + ], + "question_timestamps": "00:06:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Zero.", + "D": "Three people were on the sofa a moment ago when I last checked the living room.", + "E": "Four people were sitting on the sofa right before, but they stood up quickly." + }, + "answer_label": "C", + "answer_text": "Zero.", + "q_id": "egolife_A1_D2_014_2", + "correct_answer": "Zero.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Zero.", + "D": "Three people were on the sofa a moment ago when I last checked the living room.", + "E": "Four people were sitting on the sofa right before, but they stood up quickly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_014", + "video_length": "00:11:24", + "qid": "egolife_A1_D2_014-mcq_fixed_2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where was the drink before I picked it up?", + "answer": "On the table in the living room.", + "question_chinese": "在我拿起饮料前，它在哪里？", + "answer_chinese": "客厅里的桌子上", + "evidence": [ + { + "description": "The drink is on the table in the living room", + "timestamps": { + "start": "00:01:17", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Under the table.", + "B": "On the kitchen counter.", + "C": "It was under the table in the living room before you picked it up.", + "D": "On the table in the living room.", + "E": "It was on a counter near the kitchen doorway before you grabbed it." + }, + "answer_label": "D", + "answer_text": "On the table in the living room.", + "q_id": "egolife_A1_D2_014-mcq_fixed_2", + "correct_answer": "On the table in the living room.", + "original_options": { + "A": "Under the table.", + "B": "On the kitchen counter.", + "C": "It was under the table in the living room before you picked it up.", + "D": "On the table in the living room.", + "E": "It was on a counter near the kitchen doorway before you grabbed it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_014", + "video_length": "00:11:24", + "qid": "egolife_A1_D2_014-mcq_fixed_3", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where was the drink before I picked it up?", + "answer": "On the table in the bedroom.", + "question_chinese": "在我拿起饮料前，它在哪里？", + "answer_chinese": "卧室里的桌子上", + "evidence": [ + { + "description": "The drink is on the table in the bedroom", + "timestamps": { + "start": "00:03:49", + "end": "00:03:53" + } + } + ], + "question_timestamps": "00:03:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Under the table.", + "B": "On the desk.", + "C": "On the table in the bedroom.", + "D": "It was next to the table in the bedroom before you picked it up.", + "E": "It was on the coffee table in the living room beside the sofa before you grabbed it." + }, + "answer_label": "C", + "answer_text": "On the table in the bedroom.", + "q_id": "egolife_A1_D2_014-mcq_fixed_3", + "correct_answer": "On the table in the bedroom.", + "original_options": { + "A": "Under the table.", + "B": "On the desk.", + "C": "On the table in the bedroom.", + "D": "It was next to the table in the bedroom before you picked it up.", + "E": "It was on the coffee table in the living room beside the sofa before you grabbed it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_014", + "video_length": "00:11:24", + "qid": "egolife_A1_D2_014_3_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was just at the doorway, which way did I turn to enter the living room?", + "answer": "To your right.", + "question_chinese": "我刚在门口是朝哪个方向转弯进入客厅的？", + "answer_chinese": "右边", + "evidence": [ + { + "description": "When you are in the bedroom, turn right to enter the living room.", + "timestamps": { + "start": "00:01:51", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:01:59", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "To your left.", + "B": "To your right." + }, + "answer_label": "B", + "answer_text": "To your right.", + "q_id": "egolife_A1_D2_014_3_T1", + "correct_answer": "To your right.", + "original_options": { + "A": "To your left.", + "B": "To your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D2_014", + "video_length": "00:11:24", + "qid": "egolife_A1_D2_014_3_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was just at the doorway, which way did I turn to enter the living room?", + "answer": "To your left.", + "question_chinese": "我刚在门口是朝哪个方向转弯进入客厅的？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "After going up the stairs, I turned right into the living room", + "timestamps": { + "start": "00:10:23", + "end": "00:10:26" + } + } + ], + "question_timestamps": "00:10:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "To your left.", + "B": "To your right." + }, + "answer_label": "A", + "answer_text": "To your left.", + "q_id": "egolife_A1_D2_014_3_T2", + "correct_answer": "To your left.", + "original_options": { + "A": "To your left.", + "B": "To your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D2_014", + "video_length": "00:11:24", + "qid": "egolife_A1_D2_014_5", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What steps do I need to take to obtain the black and white partition board?", + "answer": "Step out of the room door from the back and then walk to the right front.", + "question_chinese": "我现在想要拿到黑白间隔板子需要怎么走？", + "answer_chinese": "往后出房间门，再向右前方走", + "evidence": [ + { + "description": "The blackboard partition board is in the living room on the first floor", + "timestamps": { + "start": "00:01:23", + "end": "00:01:44" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go out the front door.", + "B": "Walk to the right back.", + "C": "Step out of the room door from the back and then walk to the right front.", + "D": "Step out of the room door from the front and then walk toward the left back.", + "E": "Step out of the room door from the back and then walk toward the left front." + }, + "answer_label": "C", + "answer_text": "Step out of the room door from the back and then walk to the right front.", + "q_id": "egolife_A1_D2_014_5", + "correct_answer": "Step out of the room door from the back and then walk to the right front.", + "original_options": { + "A": "Go out the front door.", + "B": "Walk to the right back.", + "C": "Step out of the room door from the back and then walk to the right front.", + "D": "Step out of the room door from the front and then walk toward the left back.", + "E": "Step out of the room door from the back and then walk toward the left front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_014", + "video_length": "00:11:24", + "qid": "egolife_A1_D2_014_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What steps do I need to take to obtain the black and white partition board?", + "answer": "Go down to the first floor from the back, turn left into the living room, and then move forward to the left.", + "question_chinese": "我现在想要拿到黑白间隔板子需要怎么走？", + "answer_chinese": "从后方下到一楼，左转进入客厅，再向左前方前进", + "evidence": [ + { + "description": "I'm at the staircase entrance on the second floor right now", + "timestamps": { + "start": "00:05:42", + "end": "00:05:48" + } + }, + { + "description": "The black and white partitioned version board is in the living room on the first floor", + "timestamps": { + "start": "00:01:23", + "end": "00:01:44" + } + } + ], + "question_timestamps": "00:05:56", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right into the living room.", + "B": "Go down to the first floor from the back, turn left into the living room, and then move forward to the left.", + "C": "Enter the hallway, then go left.", + "D": "Go down to the first floor from the front, turn left into the living room, then head forward left.", + "E": "Go down from the back, turn left into the living room, then continue straight past the dining area." + }, + "answer_label": "B", + "answer_text": "Go down to the first floor from the back, turn left into the living room, and then move forward to the left.", + "q_id": "egolife_A1_D2_014_6", + "correct_answer": "Go down to the first floor from the back, turn left into the living room, and then move forward to the left.", + "original_options": { + "A": "Turn right into the living room.", + "B": "Go down to the first floor from the back, turn left into the living room, and then move forward to the left.", + "C": "Enter the hallway, then go left.", + "D": "Go down to the first floor from the front, turn left into the living room, then head forward left.", + "E": "Go down from the back, turn left into the living room, then continue straight past the dining area." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D2_016-mcq.json b/QAs/egolife_A1_D2_016-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d0514b9c4e07d3348bfb8d75cf7ec85ec1a34b5c --- /dev/null +++ b/QAs/egolife_A1_D2_016-mcq.json @@ -0,0 +1,919 @@ +[ + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_3_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are in the bedroom?", + "answer": "Two, inside the bedroom behind you.", + "question_chinese": "卧室里有几把椅子？", + "answer_chinese": "两把，在你身后的卧室里面", + "evidence": [ + { + "description": "There are two chairs in the bedroom", + "timestamps": { + "start": "00:01:51", + "end": "00:01:54" + } + } + ], + "question_timestamps": "00:02:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, inside the bedroom behind you.", + "B": "Three, inside the bedroom behind you.", + "C": "Zero, inside the bedroom behind you; there aren’t any chairs visible in that room.", + "D": "Two, inside the bedroom behind you.", + "E": "Four, inside the bedroom behind you; you can spot multiple chairs grouped near the back." + }, + "answer_label": "D", + "answer_text": "Two, inside the bedroom behind you.", + "q_id": "egolife_A1_D2_016_3_T1", + "correct_answer": "Two, inside the bedroom behind you.", + "original_options": { + "A": "One, inside the bedroom behind you.", + "B": "Three, inside the bedroom behind you.", + "C": "Zero, inside the bedroom behind you; there aren’t any chairs visible in that room.", + "D": "Two, inside the bedroom behind you.", + "E": "Four, inside the bedroom behind you; you can spot multiple chairs grouped near the back." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_5", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the bedroom?", + "answer": "Two.", + "question_chinese": "卧室里有几把椅子？", + "answer_chinese": "两把", + "evidence": [ + { + "description": "There are two chairs in the bedroom", + "timestamps": { + "start": "00:01:51", + "end": "00:01:54" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four; you can count two pairs of chairs placed on opposite sides of the bedroom.", + "D": "Two.", + "E": "Five; several chairs are positioned around the bedroom, more than the two you’d expect." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "egolife_A1_D2_016_5", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four; you can count two pairs of chairs placed on opposite sides of the bedroom.", + "D": "Two.", + "E": "Five; several chairs are positioned around the bedroom, more than the two you’d expect." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_3_T4", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are in the bedroom?", + "answer": "Three, inside the bedroom at your back-right.", + "question_chinese": "卧室里有几把椅子？", + "answer_chinese": "三把，在你右后侧的卧室里面", + "evidence": [ + { + "description": "There are three chairs in the bedroom", + "timestamps": { + "start": "00:04:13", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:04:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, inside the bedroom at your back-right.", + "B": "Three, inside the bedroom at your back-right.", + "C": "Four, inside the bedroom at your back-right.", + "D": "One, inside the bedroom at your back-right, with only a single chair visible in that corner.", + "E": "Five, inside the bedroom at your back-right, with several chairs clustered together behind you." + }, + "answer_label": "B", + "answer_text": "Three, inside the bedroom at your back-right.", + "q_id": "egolife_A1_D2_016_3_T4", + "correct_answer": "Three, inside the bedroom at your back-right.", + "original_options": { + "A": "Two, inside the bedroom at your back-right.", + "B": "Three, inside the bedroom at your back-right.", + "C": "Four, inside the bedroom at your back-right.", + "D": "One, inside the bedroom at your back-right, with only a single chair visible in that corner.", + "E": "Five, inside the bedroom at your back-right, with several chairs clustered together behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the bedroom?", + "answer": "Three.", + "question_chinese": "卧室里有几把椅子？", + "answer_chinese": "三把", + "evidence": [ + { + "description": "There are three chairs in the bedroom", + "timestamps": { + "start": "00:04:13", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:04:36", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "Six; you can count six chairs in the bedroom, clearly more than three in total.", + "D": "Three.", + "E": "Seven; a large cluster of chairs fills the bedroom, well above three when you count them." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_016_6", + "correct_answer": "Three.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "Six; you can count six chairs in the bedroom, clearly more than three in total.", + "D": "Three.", + "E": "Seven; a large cluster of chairs fills the bedroom, well above three when you count them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_6_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the living-room microwave from me?", + "answer": "The right side.", + "question_chinese": "客厅的微波炉在我的什么方向？", + "answer_chinese": "右边", + "evidence": [ + { + "description": "The microwave oven is beside the windowsill", + "timestamps": { + "start": "00:01:37", + "end": "00:01:43" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The left side.", + "B": "The right side.", + "C": "Straight ahead.", + "D": "To your front-left, just past the living-room area from where you’re standing.", + "E": "Behind you on the left, so you’d need to turn around to face it." + }, + "answer_label": "B", + "answer_text": "The right side.", + "q_id": "egolife_A1_D2_016_6_T1", + "correct_answer": "The right side.", + "original_options": { + "A": "The left side.", + "B": "The right side.", + "C": "Straight ahead.", + "D": "To your front-left, just past the living-room area from where you’re standing.", + "E": "Behind you on the left, so you’d need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_7_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the living-room microwave from me?", + "answer": "To your front-right.", + "question_chinese": "客厅的微波炉在我的什么方向？", + "answer_chinese": "右前方", + "evidence": [ + { + "description": "The microwave appeared to the front-left of the door on the right side.", + "timestamps": { + "start": "00:01:28", + "end": "00:01:30" + } + }, + { + "description": "You walked up to the door, then turned your body to the left.", + "timestamps": { + "start": "00:02:25", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "To your front-right.", + "D": "On your left side, meaning you’d need to turn left to face the microwave.", + "E": "Directly behind you, so you have to turn around to see the living-room microwave." + }, + "answer_label": "C", + "answer_text": "To your front-right.", + "q_id": "egolife_A1_D2_016_7_T1", + "correct_answer": "To your front-right.", + "original_options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "To your front-right.", + "D": "On your left side, meaning you’d need to turn left to face the microwave.", + "E": "Directly behind you, so you have to turn around to see the living-room microwave." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many refrigerators are there in the house?", + "answer": "One.", + "question_chinese": "家里有几台冰箱？", + "answer_chinese": "一台", + "evidence": [ + { + "description": "There is a refrigerator at the entrance of the staircase on the second floor", + "timestamps": { + "start": "00:00:44", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:01:14", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four; there are multiple refrigerators placed in different rooms throughout the house, not just the kitchen.", + "E": "Five; you can find several refrigerators spread across the house, including extra units outside the kitchen." + }, + "answer_label": "A", + "answer_text": "One.", + "q_id": "egolife_A1_D2_016_1", + "correct_answer": "One.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four; there are multiple refrigerators placed in different rooms throughout the house, not just the kitchen.", + "E": "Five; you can find several refrigerators spread across the house, including extra units outside the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many refrigerators are there in the house?", + "answer": "Two.", + "question_chinese": "家里有几台冰箱？", + "answer_chinese": "两台", + "evidence": [ + { + "description": "There is a refrigerator at the entrance of the staircase on the second floor", + "timestamps": { + "start": "00:00:44", + "end": "00:00:54" + } + }, + { + "description": "There is a refrigerator beside the staircase on the first floor", + "timestamps": { + "start": "00:02:27", + "end": "00:02:31" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "There is only one refrigerator in the house, located in the kitchen near the counter.", + "D": "There are three refrigerators in the house, including one upstairs and two downstairs near the kitchen.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "egolife_A1_D2_016_2", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "There is only one refrigerator in the house, located in the kitchen near the counter.", + "D": "There are three refrigerators in the house, including one upstairs and two downstairs near the kitchen.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_7", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the washing machine?", + "answer": "I need to turn right out of the bedroom door and walk straight to the staircase, then turn right.", + "question_chinese": "我现在想要用洗衣机，应该怎么走？", + "answer_chinese": "我需要右转出卧室门，并直走到楼梯口，然后右转", + "evidence": [ + { + "description": "The washing machine is between the staircase on the first floor and the main entrance", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + }, + { + "description": "One needs to walk in a straight line from the bedroom to the staircase", + "timestamps": { + "start": "00:01:31", + "end": "00:01:35" + } + }, + { + "description": "I'm in the bedroom at the moment", + "timestamps": { + "start": "00:02:08", + "end": "00:02:13" + } + } + ], + "question_timestamps": "00:02:19", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, then turn right.", + "B": "Turn right, then turn left.", + "C": "I need to turn right out of the bedroom door and walk straight to the staircase, then turn right.", + "D": "Turn left out of the bedroom door, walk straight to the staircase, then turn right.", + "E": "Turn right out of the bedroom door, walk past the staircase, then turn left at the end." + }, + "answer_label": "C", + "answer_text": "I need to turn right out of the bedroom door and walk straight to the staircase, then turn right.", + "q_id": "egolife_A1_D2_016_7", + "correct_answer": "I need to turn right out of the bedroom door and walk straight to the staircase, then turn right.", + "original_options": { + "A": "Turn left, then turn right.", + "B": "Turn right, then turn left.", + "C": "I need to turn right out of the bedroom door and walk straight to the staircase, then turn right.", + "D": "Turn left out of the bedroom door, walk straight to the staircase, then turn right.", + "E": "Turn right out of the bedroom door, walk past the staircase, then turn left at the end." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_8", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the washing machine?", + "answer": "Turn right to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, walk straight for a short distance.", + "question_chinese": "我现在想要用洗衣机，应该怎么走？", + "answer_chinese": "右转到达楼梯口，然后右转并下楼，到达一楼后直行一小段距离", + "evidence": [ + { + "description": "The washing machine is between the staircase on the first floor and the main entrance", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + }, + { + "description": "To get to the living room on the second floor from the staircase on the first floor, you need to go upstairs and then turn left", + "timestamps": { + "start": "00:04:25", + "end": "00:04:40" + } + }, + { + "description": "I'm in the living room on the second floor right now", + "timestamps": { + "start": "00:04:41", + "end": "00:04:45" + } + } + ], + "question_timestamps": "00:04:52", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, walk straight for a short distance.", + "B": "Turn left, go downstairs.", + "C": "Go upstairs after turning right.", + "D": "Turn right to reach the staircase entrance, then turn left and go downstairs to the first floor.", + "E": "Turn right to the staircase, go downstairs, then turn left and continue straight for a short distance." + }, + "answer_label": "A", + "answer_text": "Turn right to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, walk straight for a short distance.", + "q_id": "egolife_A1_D2_016_8", + "correct_answer": "Turn right to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, walk straight for a short distance.", + "original_options": { + "A": "Turn right to reach the staircase entrance, then turn right and go downstairs. After reaching the first floor, walk straight for a short distance.", + "B": "Turn left, go downstairs.", + "C": "Go upstairs after turning right.", + "D": "Turn right to reach the staircase entrance, then turn left and go downstairs to the first floor.", + "E": "Turn right to the staircase, go downstairs, then turn left and continue straight for a short distance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_9", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get some bottled water now?", + "answer": "Yes.", + "question_chinese": "我现在能拿到矿泉水吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The mineral water is on the big table", + "timestamps": { + "start": "00:06:21", + "end": "00:06:31" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D2_016_9", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_10", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get some bottled water now?", + "answer": "No.", + "question_chinese": "我现在能拿到矿泉水吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The mineral water is on the big table", + "timestamps": { + "start": "00:06:21", + "end": "00:06:31" + } + } + ], + "question_timestamps": "00:07:36", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D2_016_10", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_3", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the table or the refrigerator?", + "answer": "Table.", + "question_chinese": "桌子和冰箱哪个离我更近一些？", + "answer_chinese": "桌子", + "evidence": [ + { + "description": "The table is beside the bedroom door and the refrigerator is beside the staircase", + "timestamps": { + "start": "00:01:30", + "end": "00:01:41" + } + }, + { + "description": "I'm in the bedroom at this moment", + "timestamps": { + "start": "00:04:13", + "end": "00:04:17" + } + } + ], + "question_timestamps": "00:04:29", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Table.", + "B": "Refrigerator." + }, + "answer_label": "A", + "answer_text": "Table.", + "q_id": "egolife_A1_D2_016_3", + "correct_answer": "Table.", + "original_options": { + "A": "Table.", + "B": "Refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_4", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the table or the refrigerator?", + "answer": "Refrigerator?", + "question_chinese": "桌子和冰箱哪个离我更近一些？", + "answer_chinese": "冰箱", + "evidence": [ + { + "description": "The table is beside the bedroom door and the refrigerator is beside the staircase", + "timestamps": { + "start": "00:01:30", + "end": "00:01:41" + } + }, + { + "description": "I'm on the stairs", + "timestamps": { + "start": "00:05:01", + "end": "00:05:06" + } + } + ], + "question_timestamps": "00:05:18", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Table.", + "B": "Refrigerator?" + }, + "answer_label": "B", + "answer_text": "Refrigerator?", + "q_id": "egolife_A1_D2_016_4", + "correct_answer": "Refrigerator?", + "original_options": { + "A": "Table.", + "B": "Refrigerator?" + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_2_T4", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the dining table or the refrigerator?", + "answer": "The refrigerator is on the left side.", + "question_chinese": "餐桌和冰箱哪个离我更近一些？", + "answer_chinese": "冰箱，在你的左侧", + "evidence": [ + { + "description": "The table is beside the bedroom door and the refrigerator is beside the staircase", + "timestamps": { + "start": "00:01:30", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The refrigerator is on the right side.", + "B": "The refrigerator is on the left side." + }, + "answer_label": "B", + "answer_text": "The refrigerator is on the left side.", + "q_id": "egolife_A1_D2_016_2_T4", + "correct_answer": "The refrigerator is on the left side.", + "original_options": { + "A": "The refrigerator is on the right side.", + "B": "The refrigerator is on the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D2_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_016_2_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the dining table or the refrigerator?", + "answer": "The dining table is at your back-right.", + "question_chinese": "餐桌和冰箱哪个离我更近一些？", + "answer_chinese": "餐桌，在你的右后侧", + "evidence": [ + { + "description": "The table is beside the bedroom door and the refrigerator is beside the staircase", + "timestamps": { + "start": "00:01:30", + "end": "00:01:41" + } + }, + { + "description": "I'm in the bedroom at this moment", + "timestamps": { + "start": "00:04:13", + "end": "00:04:17" + } + } + ], + "question_timestamps": "00:04:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The dining table is at your back-right.", + "B": "The refrigerator is closer to you." + }, + "answer_label": "A", + "answer_text": "The dining table is at your back-right.", + "q_id": "egolife_A1_D2_016_2_T1", + "correct_answer": "The dining table is at your back-right.", + "original_options": { + "A": "The dining table is at your back-right.", + "B": "The refrigerator is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/egolife_A1_D2_018-mcq.json b/QAs/egolife_A1_D2_018-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..78eecf759fdf1f0acdc8a069429fe1f7e9637884 --- /dev/null +++ b/QAs/egolife_A1_D2_018-mcq.json @@ -0,0 +1,480 @@ +[ + { + "video_uid": "egolife_A1_D2_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_018_5", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mirrors are there in the house?", + "answer": "One.", + "question_chinese": "家里一共有几面镜子？", + "answer_chinese": "一面", + "evidence": [ + { + "description": "There is a mirror in the living room on the first floor", + "timestamps": { + "start": "00:08:47", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:09:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are two mirrors in the house, one in the bathroom and another in the hallway.", + "E": "You can find three mirrors in the house, including one in the bathroom, bedroom, and hallway." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "egolife_A1_D2_018_5", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are two mirrors in the house, one in the bathroom and another in the hallway.", + "E": "You can find three mirrors in the house, including one in the bathroom, bedroom, and hallway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_018_6", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mirrors are there in the house?", + "answer": "Two.", + "question_chinese": "家里一共有几面镜子？", + "answer_chinese": "两面", + "evidence": [ + { + "description": "There is a mirror in the living room on the first floor", + "timestamps": { + "start": "00:08:47", + "end": "00:09:00" + } + }, + { + "description": "There is a mirror in the living room on the second floor", + "timestamps": { + "start": "00:11:41", + "end": "00:11:47" + } + } + ], + "question_timestamps": "00:12:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "There is only one mirror in the house, and it is in the bathroom by the sink.", + "E": "There are three mirrors throughout the house, placed in the bathroom, bedroom, and hallway." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "egolife_A1_D2_018_6", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "There is only one mirror in the house, and it is in the bathroom by the sink.", + "E": "There are three mirrors throughout the house, placed in the bathroom, bedroom, and hallway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_018_2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement pattern after I last entered the living room?", + "answer": "After entering the living room, I walked half a circle around the table and then returned to the bedroom by the same route.", + "question_chinese": "我最近一次进入客厅后的运动轨迹是怎样的？", + "answer_chinese": "我进入客厅后绕着桌子走了半圈，然后原路返回卧室", + "evidence": [ + { + "description": "After entering the living room, I walked half a circle around the table and then returned to the bedroom by the same route", + "timestamps": { + "start": "00:01:38", + "end": "00:02:09" + } + } + ], + "question_timestamps": "00:02:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "After entering the living room, I walked half a circle around the table and then returned to the bedroom by the same route.", + "B": "I walked straight through.", + "C": "I went back immediately.", + "D": "After entering the living room, I circled the table once and then went back to the bedroom.", + "E": "After entering the living room, I walked half a circle around the table and then went to the hallway." + }, + "answer_label": "A", + "answer_text": "After entering the living room, I walked half a circle around the table and then returned to the bedroom by the same route.", + "q_id": "egolife_A1_D2_018_2", + "correct_answer": "After entering the living room, I walked half a circle around the table and then returned to the bedroom by the same route.", + "original_options": { + "A": "After entering the living room, I walked half a circle around the table and then returned to the bedroom by the same route.", + "B": "I walked straight through.", + "C": "I went back immediately.", + "D": "After entering the living room, I circled the table once and then went back to the bedroom.", + "E": "After entering the living room, I walked half a circle around the table and then went to the hallway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_018_1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement pattern after I last entered the living room?", + "answer": "After entering the living room, I walked around on the spot and then returned to the bedroom.", + "question_chinese": "我最近一次进入客厅后的运动轨迹是怎样的？", + "answer_chinese": "进入客厅后，我在原地走了一圈，然后回到卧室", + "evidence": [ + { + "description": "After entering the living room, I walked around on the spot and then returned to the bedroom", + "timestamps": { + "start": "00:03:15", + "end": "00:03:19" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I walked around furniture.", + "B": "I headed to the hallway.", + "C": "After entering the living room, I stood still briefly and then returned to the bedroom.", + "D": "After entering the living room, I walked around on the spot and then returned to the bedroom.", + "E": "After entering the living room, I walked around on the spot and then moved toward the kitchen." + }, + "answer_label": "D", + "answer_text": "After entering the living room, I walked around on the spot and then returned to the bedroom.", + "q_id": "egolife_A1_D2_018_1", + "correct_answer": "After entering the living room, I walked around on the spot and then returned to the bedroom.", + "original_options": { + "A": "I walked around furniture.", + "B": "I headed to the hallway.", + "C": "After entering the living room, I stood still briefly and then returned to the bedroom.", + "D": "After entering the living room, I walked around on the spot and then returned to the bedroom.", + "E": "After entering the living room, I walked around on the spot and then moved toward the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_018_3", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I use the notepad now?", + "answer": "No.", + "question_chinese": "我现在能使用写字板吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The writing board is on the side of the living room near the main door", + "timestamps": { + "start": "00:02:12", + "end": "00:02:21" + } + }, + { + "description": "I'm in the bedroom at the moment and quite far from the writing board", + "timestamps": { + "start": "00:03:51", + "end": "00:04:02" + } + } + ], + "question_timestamps": "00:04:17", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D2_018_3", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_018_4", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I use the notepad now?", + "answer": "Yes.", + "question_chinese": "我现在能使用写字板吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The writing board is on the side of the living room near the main door", + "timestamps": { + "start": "00:02:12", + "end": "00:02:21" + } + }, + { + "description": "I have already walked to the side of the writing board", + "timestamps": { + "start": "00:09:15", + "end": "00:09:17" + } + } + ], + "question_timestamps": "00:09:23", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D2_018_4", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_018_7", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the table or the writing board?", + "answer": "Table.", + "question_chinese": "桌子和写字板哪个离我更近一些？", + "answer_chinese": "桌子", + "evidence": [ + { + "description": "The table is in the middle of the living room", + "timestamps": { + "start": "00:01:41", + "end": "00:01:45" + } + }, + { + "description": "The writing board is on the side of the living room close to the main door", + "timestamps": { + "start": "00:02:31", + "end": "00:02:39" + } + }, + { + "description": "The bedroom is on the side opposite the main door in the living room", + "timestamps": { + "start": "00:02:33", + "end": "00:02:51" + } + } + ], + "question_timestamps": "00:04:14", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Table.", + "B": "Writing board." + }, + "answer_label": "A", + "answer_text": "Table.", + "q_id": "egolife_A1_D2_018_7", + "correct_answer": "Table.", + "original_options": { + "A": "Table.", + "B": "Writing board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D2_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_018_8", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the table or the writing board?", + "answer": "Writing board.", + "question_chinese": "桌子和写字板哪个离我更近一些？", + "answer_chinese": "写字板", + "evidence": [ + { + "description": "The table is in the middle of the living room", + "timestamps": { + "start": "00:01:41", + "end": "00:01:45" + } + }, + { + "description": "The writing board is on the side of the living room close to the main door", + "timestamps": { + "start": "00:02:31", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:09:28", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Table.", + "B": "Writing board." + }, + "answer_label": "B", + "answer_text": "Writing board.", + "q_id": "egolife_A1_D2_018_8", + "correct_answer": "Writing board.", + "original_options": { + "A": "Table.", + "B": "Writing board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D2_019-mcq.json b/QAs/egolife_A1_D2_019-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..616bfa399b5bab160f30f68ad52a8a92d4d5191b --- /dev/null +++ b/QAs/egolife_A1_D2_019-mcq.json @@ -0,0 +1,935 @@ +[ + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_6_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the living room table from me?", + "answer": "Behind you.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "后面", + "evidence": [ + { + "description": "When I just picked up my phone, the table was on my left", + "timestamps": { + "start": "00:07:50", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:07:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "It’s slightly to your right, not directly behind you where you were standing.", + "D": "Behind you.", + "E": "The table is positioned ahead of you in the living room, so you face it rather than turn around." + }, + "answer_label": "D", + "answer_text": "Behind you.", + "q_id": "egolife_A1_D2_019_6_T1", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "It’s slightly to your right, not directly behind you where you were standing.", + "D": "Behind you.", + "E": "The table is positioned ahead of you in the living room, so you face it rather than turn around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_6_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the living room table from me?", + "answer": "In front of you.", + "question_chinese": "客厅的桌子在我的什么方向？", + "answer_chinese": "前面", + "evidence": [ + { + "description": "When I just picked up my phone, the table was on my left", + "timestamps": { + "start": "00:07:50", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:08:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "It’s off to your right side in the living room, so you’d turn right to face it.", + "E": "The table is behind your position, meaning you’d need to look back to see it." + }, + "answer_label": "B", + "answer_text": "In front of you.", + "q_id": "egolife_A1_D2_019_6_T2", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "It’s off to your right side in the living room, so you’d turn right to face it.", + "E": "The table is behind your position, meaning you’d need to look back to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_9", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the living room table from me when I just picked up my phone?", + "answer": "The right side.", + "question_chinese": "我刚刚拿起手机时，客厅的桌子在我的什么方向？", + "answer_chinese": "右边", + "evidence": [ + { + "description": "Just now when I picked up my phone, the table was on my right", + "timestamps": { + "start": "00:07:17", + "end": "00:07:19" + } + } + ], + "question_timestamps": "00:07:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The left side.", + "B": "The front side.", + "C": "It was in front of you rather than on your right, so you could see it without turning.", + "D": "The table was positioned to your left side at that moment, not on the right where you were facing.", + "E": "The right side." + }, + "answer_label": "E", + "answer_text": "The right side.", + "q_id": "egolife_A1_D2_019_9", + "correct_answer": "The right side.", + "original_options": { + "A": "The left side.", + "B": "The front side.", + "C": "It was in front of you rather than on your right, so you could see it without turning.", + "D": "The table was positioned to your left side at that moment, not on the right where you were facing.", + "E": "The right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_10", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I just picked up my phone, which direction was the table in the living room from me?", + "answer": "The left side.", + "question_chinese": "我刚刚拿起手机时，客厅的桌子在我的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "When I just picked up my phone, the table was on my left", + "timestamps": { + "start": "00:07:50", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:08:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The right side.", + "B": "The front side.", + "C": "It was on your right side when you picked up the phone, not to your left as you stood there.", + "D": "The left side.", + "E": "The table was in front of you at that moment, so you didn’t need to look to either side." + }, + "answer_label": "D", + "answer_text": "The left side.", + "q_id": "egolife_A1_D2_019_10", + "correct_answer": "The left side.", + "original_options": { + "A": "The right side.", + "B": "The front side.", + "C": "It was on your right side when you picked up the phone, not to your left as you stood there.", + "D": "The left side.", + "E": "The table was in front of you at that moment, so you didn’t need to look to either side." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to get to the living room on the second floor?", + "answer": "Three.", + "question_chinese": "我想要去二楼客厅，至少要拐几个弯？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There are three turns between the living room on the first floor and that on the second floor", + "timestamps": { + "start": "00:00:55", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:06:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "You only need to make two turns along the way before you enter the second-floor living room.", + "E": "You end up making four turns in total to reach the second-floor living room from where you start." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_019_3", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "You only need to make two turns along the way before you enter the second-floor living room.", + "E": "You end up making four turns in total to reach the second-floor living room from where you start." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_4", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to get to the living room on the second floor?", + "answer": "Four.", + "question_chinese": "我想要去二楼客厅，至少要拐几个弯？", + "answer_chinese": "四个", + "evidence": [ + { + "description": "There are three turns between the living room on the first floor and that on the second floor", + "timestamps": { + "start": "00:00:55", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:08:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Three.", + "B": "Five.", + "C": "You make three turns before arriving at the second-floor living room, not as many as four.", + "D": "It takes five turns overall to reach the second-floor living room, counting each change in direction.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "egolife_A1_D2_019_4", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "You make three turns before arriving at the second-floor living room, not as many as four.", + "D": "It takes five turns overall to reach the second-floor living room, counting each change in direction.", + "E": "Four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Can I get the vase now?", + "answer": "No.", + "question_chinese": "我现在可以拿到花瓶吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The vase is on the table in the living room", + "timestamps": { + "start": "00:01:19", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D2_019_1", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Can I get the vase now?", + "answer": "Yes.", + "question_chinese": "我现在可以拿到花瓶吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The vase is on the table in the living room", + "timestamps": { + "start": "00:01:19", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D2_019_2", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_7", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the microwave?", + "answer": "Turn right into the living room, go straight for a while and then turn left, and then go straight again.", + "question_chinese": "我现在想要用微波炉，应该怎么走？", + "answer_chinese": "向右转进入客厅，直走一会儿，然后向左拐，然后再直走", + "evidence": [ + { + "description": "The microwave oven is on the windowsill", + "timestamps": { + "start": "00:03:45", + "end": "00:03:50" + } + } + ], + "question_timestamps": "00:05:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right into the living room, go straight for a while and then turn left, and then go straight again.", + "B": "Turn left into the living room.", + "C": "Go straight, then turn right.", + "D": "Turn right into the living room, go straight for a while, then turn right, and continue straight.", + "E": "Go straight into the living room first, then turn left at the next opening and keep going forward." + }, + "answer_label": "A", + "answer_text": "Turn right into the living room, go straight for a while and then turn left, and then go straight again.", + "q_id": "egolife_A1_D2_019_7", + "correct_answer": "Turn right into the living room, go straight for a while and then turn left, and then go straight again.", + "original_options": { + "A": "Turn right into the living room, go straight for a while and then turn left, and then go straight again.", + "B": "Turn left into the living room.", + "C": "Go straight, then turn right.", + "D": "Turn right into the living room, go straight for a while, then turn right, and continue straight.", + "E": "Go straight into the living room first, then turn left at the next opening and keep going forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_8", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the microwave?", + "answer": "Go straight for a while, then turn left and go straight again.", + "question_chinese": "我现在想要用微波炉，应该怎么走？", + "answer_chinese": "直行一段距离再左转，再直行", + "evidence": [ + { + "description": "The microwave oven is on the windowsill", + "timestamps": { + "start": "00:03:45", + "end": "00:03:50" + } + } + ], + "question_timestamps": "00:07:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left immediately, then go straight.", + "B": "Go straight for a while, then turn left and go straight again.", + "C": "Go straight, then turn right.", + "D": "Go straight for a while, then turn right and go straight again until you reach it.", + "E": "Walk forward for a bit, turn left, then make another left before continuing straight." + }, + "answer_label": "B", + "answer_text": "Go straight for a while, then turn left and go straight again.", + "q_id": "egolife_A1_D2_019_8", + "correct_answer": "Go straight for a while, then turn left and go straight again.", + "original_options": { + "A": "Turn left immediately, then go straight.", + "B": "Go straight for a while, then turn left and go straight again.", + "C": "Go straight, then turn right.", + "D": "Go straight for a while, then turn right and go straight again until you reach it.", + "E": "Walk forward for a bit, turn left, then make another left before continuing straight." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_6", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the entire house?", + "answer": "Three.", + "question_chinese": "整个房子里有几个垃圾桶?", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There is a trash can beside the table on the second floor", + "timestamps": { + "start": "00:00:01", + "end": "00:00:06" + } + }, + { + "description": "There is a trash can under the table on the first floor", + "timestamps": { + "start": "00:03:57", + "end": "00:04:00" + } + }, + { + "description": "There is a trash can beside the curtain on the second floor", + "timestamps": { + "start": "00:00:18", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:05:12", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "There are five trash cans total across the house, including the kitchen and bathroom areas.", + "D": "Three.", + "E": "You have six trash cans in total, with one in each main room and an extra near the entry." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_019_6", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "There are five trash cans total across the house, including the kitchen and bathroom areas.", + "D": "Three.", + "E": "You have six trash cans in total, with one in each main room and an extra near the entry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_5", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the entire house?", + "answer": "Four.", + "question_chinese": "整个房子里有几个垃圾桶?", + "answer_chinese": "四个", + "evidence": [ + { + "description": "There is a trash can beside the table on the second floor", + "timestamps": { + "start": "00:00:01", + "end": "00:00:06" + } + }, + { + "description": "There is a trash can under the table on the first floor", + "timestamps": { + "start": "00:03:57", + "end": "00:04:00" + } + }, + { + "description": "There is a trash can beside the curtains on the first floor", + "timestamps": { + "start": "00:08:32", + "end": "00:08:40" + } + }, + { + "description": "There is a trash can beside the curtain on the second floor", + "timestamps": { + "start": "00:00:14", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:09:44", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "There are two trash cans in the house, one in the kitchen area and one elsewhere.", + "D": "Four.", + "E": "You can count five trash cans total, spread between the kitchen, bathroom, and another room." + }, + "answer_label": "D", + "answer_text": "Four.", + "q_id": "egolife_A1_D2_019_5", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "There are two trash cans in the house, one in the kitchen area and one elsewhere.", + "D": "Four.", + "E": "You can count five trash cans total, spread between the kitchen, bathroom, and another room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_5_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the living room table from me right now?", + "answer": "To your left.", + "question_chinese": "现在客厅的桌子在我的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "Just now when I picked up my phone, the table was on my right", + "timestamps": { + "start": "00:07:17", + "end": "00:07:19" + } + } + ], + "question_timestamps": "00:07:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "To your right.", + "D": "It’s behind you, so you’d need to turn around to face the living room table.", + "E": "It’s slightly in front of you and to the right, not on your left side." + }, + "answer_label": "B", + "answer_text": "To your left.", + "q_id": "egolife_A1_D2_019_5_T1", + "correct_answer": "To your left.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "To your right.", + "D": "It’s behind you, so you’d need to turn around to face the living room table.", + "E": "It’s slightly in front of you and to the right, not on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_5_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the living room table from me right now?", + "answer": "To your right.", + "question_chinese": "现在客厅的桌子在我的什么方向？", + "answer_chinese": "右边", + "evidence": [ + { + "description": "Just now when I picked up my phone, the table was on my right", + "timestamps": { + "start": "00:07:17", + "end": "00:07:19" + } + } + ], + "question_timestamps": "00:07:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "To your right.", + "C": "Behind you.", + "D": "It’s in front of you, so you don’t need to turn to see the living room table.", + "E": "It’s off to your left side rather than your right, near the living room seating area." + }, + "answer_label": "B", + "answer_text": "To your right.", + "q_id": "egolife_A1_D2_019_5_T2", + "correct_answer": "To your right.", + "original_options": { + "A": "To your left.", + "B": "To your right.", + "C": "Behind you.", + "D": "It’s in front of you, so you don’t need to turn to see the living room table.", + "E": "It’s off to your left side rather than your right, near the living room seating area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_12", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the mirror or the refrigerator?", + "answer": "Mirror.", + "question_chinese": "镜子和冰箱哪个离我更近？", + "answer_chinese": "镜子", + "evidence": [ + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:01:15", + "end": "00:01:26" + } + }, + { + "description": "The refrigerator is beside the staircase", + "timestamps": { + "start": "00:03:25", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:05:12", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Refrigerator.", + "B": "Mirror." + }, + "answer_label": "B", + "answer_text": "Mirror.", + "q_id": "egolife_A1_D2_019_12", + "correct_answer": "Mirror.", + "original_options": { + "A": "Refrigerator.", + "B": "Mirror." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_019_11", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the mirror or the refrigerator?", + "answer": "Refrigerator.", + "question_chinese": "镜子和冰箱哪个离我更近？", + "answer_chinese": "冰箱", + "evidence": [ + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:01:15", + "end": "00:01:26" + } + }, + { + "description": "The refrigerator is beside the staircase", + "timestamps": { + "start": "00:03:25", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:08:29", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Refrigerator.", + "B": "Mirror." + }, + "answer_label": "A", + "answer_text": "Refrigerator.", + "q_id": "egolife_A1_D2_019_11", + "correct_answer": "Refrigerator.", + "original_options": { + "A": "Refrigerator.", + "B": "Mirror." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D2_020-mcq.json b/QAs/egolife_A1_D2_020-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a32e3bf0504b7839288a3f9e5e4a15bc76fe00ab --- /dev/null +++ b/QAs/egolife_A1_D2_020-mcq.json @@ -0,0 +1,862 @@ +[ + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_1_T4", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the refrigerator or the table?", + "answer": "The table is close to you.", + "question_chinese": "冰箱和桌子哪个离我近？", + "answer_chinese": "桌子离你近", + "evidence": [ + { + "description": "The table is in the middle of the living room", + "timestamps": { + "start": "00:03:32", + "end": "00:03:36" + } + }, + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:02:29", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:05:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The table is close to you.", + "B": "The refrigerator is close to you." + }, + "answer_label": "A", + "answer_text": "The table is close to you.", + "q_id": "egolife_A1_D2_020_1_T4", + "correct_answer": "The table is close to you.", + "original_options": { + "A": "The table is close to you.", + "B": "The refrigerator is close to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_1_T3", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the refrigerator or the table?", + "answer": "The fridge is close to you.", + "question_chinese": "冰箱和桌子哪个离我近？", + "answer_chinese": "冰箱离你近", + "evidence": [ + { + "description": "The table is in the middle of the living room", + "timestamps": { + "start": "00:03:32", + "end": "00:03:36" + } + }, + { + "description": "The refrigerator is on the side of the living room close to the main door", + "timestamps": { + "start": "00:04:05", + "end": "00:04:16" + } + }, + { + "description": "As soon as you step out of the bedroom, there is a table", + "timestamps": { + "start": "00:07:13", + "end": "00:07:21" + } + }, + { + "description": "You took two steps from the table to the refrigerator.", + "timestamps": { + "start": "00:07:21", + "end": "00:07:53" + } + } + ], + "question_timestamps": "00:07:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The table is close to you.", + "B": "The fridge is close to you." + }, + "answer_label": "B", + "answer_text": "The fridge is close to you.", + "q_id": "egolife_A1_D2_020_1_T3", + "correct_answer": "The fridge is close to you.", + "original_options": { + "A": "The table is close to you.", + "B": "The fridge is close to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_1_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the refrigerator or the table?", + "answer": "The table is closer to you.", + "question_chinese": "冰箱和桌子哪个离我近？", + "answer_chinese": "桌子离你更近", + "evidence": [ + { + "description": "The table is in the middle of the living room", + "timestamps": { + "start": "00:03:32", + "end": "00:03:36" + } + }, + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:02:29", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:11:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The table is closer to you.", + "B": "The refrigerator is closer to you." + }, + "answer_label": "A", + "answer_text": "The table is closer to you.", + "q_id": "egolife_A1_D2_020_1_T2", + "correct_answer": "The table is closer to you.", + "original_options": { + "A": "The table is closer to you.", + "B": "The refrigerator is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_4", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the refrigerator or the table?", + "answer": "Table.", + "question_chinese": "冰箱和桌子哪个离我近？", + "answer_chinese": "桌子", + "evidence": [ + { + "description": "The table is in the middle of the living room", + "timestamps": { + "start": "00:03:32", + "end": "00:03:36" + } + }, + { + "description": "The refrigerator is on the side of the living room close to the main door", + "timestamps": { + "start": "00:04:05", + "end": "00:04:16" + } + }, + { + "description": "As soon as you step out of the bedroom, there is a table", + "timestamps": { + "start": "00:07:13", + "end": "00:07:21" + } + } + ], + "question_timestamps": "00:11:27", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Table.", + "B": "Refrigerator." + }, + "answer_label": "A", + "answer_text": "Table.", + "q_id": "egolife_A1_D2_020_4", + "correct_answer": "Table.", + "original_options": { + "A": "Table.", + "B": "Refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_5_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the yellow-background wall art in the bedroom relative to me?", + "answer": "The painting is on the wall in front of you.", + "question_chinese": "卧室里的黄色背景挂画在相对于我的什么位置？", + "answer_chinese": "画在你的前方的墙壁上", + "evidence": [ + { + "description": "The painting is hung on the wall directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The painting is on the wall behind you, not the one directly in your line of sight.", + "D": "The painting is mounted on the wall to your right as you face the bed, not straight ahead.", + "E": "The painting is on the wall in front of you." + }, + "answer_label": "E", + "answer_text": "The painting is on the wall in front of you.", + "q_id": "egolife_A1_D2_020_5_T2", + "correct_answer": "The painting is on the wall in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The painting is on the wall behind you, not the one directly in your line of sight.", + "D": "The painting is mounted on the wall to your right as you face the bed, not straight ahead.", + "E": "The painting is on the wall in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_5_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the yellow-background wall art in the bedroom relative to me?", + "answer": "It is on the wall to your left.", + "question_chinese": "卧室里的黄色背景挂画在相对于我的什么位置？", + "answer_chinese": "他在你的左边的墙壁上", + "evidence": [ + { + "description": "The painting is hung on the wall directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "It is on the wall to your left.", + "C": "Behind you.", + "D": "It is on the wall to your right, across from where you are standing in the bedroom.", + "E": "It is hanging on the wall in front of you, directly ahead rather than off to your left." + }, + "answer_label": "B", + "answer_text": "It is on the wall to your left.", + "q_id": "egolife_A1_D2_020_5_T1", + "correct_answer": "It is on the wall to your left.", + "original_options": { + "A": "In front of you.", + "B": "It is on the wall to your left.", + "C": "Behind you.", + "D": "It is on the wall to your right, across from where you are standing in the bedroom.", + "E": "It is hanging on the wall in front of you, directly ahead rather than off to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020-mcq_fixed_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there at home? Where are they respectively??", + "answer": "Three.", + "question_chinese": "家里有几个垃圾桶？分别在哪里？", + "answer_chinese": "一个，在客厅的桌子下", + "evidence": [ + { + "description": "There is a trash can under the table in the living room", + "timestamps": { + "start": "00:01:48", + "end": "00:01:53" + } + } + ], + "question_timestamps": "00:02:46", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "Two trash cans: one in the kitchen and one near the bedroom door.", + "E": "Four trash cans: one in the kitchen, one in the bathroom, and two near the living room table." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_020-mcq_fixed_2", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "Two trash cans: one in the kitchen and one near the bedroom door.", + "E": "Four trash cans: one in the kitchen, one in the bathroom, and two near the living room table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020-mcq_fixed_3", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there at home? Where are they respectively??", + "answer": "Six.", + "question_chinese": "家里有几个垃圾桶？分别在哪里？", + "answer_chinese": "两个，一个在客厅的桌子下，另一个在窗台旁", + "evidence": [ + { + "description": "There is a trash can under the table in the living room", + "timestamps": { + "start": "00:01:48", + "end": "00:01:53" + } + }, + { + "description": "There is a trash can beside the windowsill", + "timestamps": { + "start": "00:05:31", + "end": "00:05:49" + } + } + ], + "question_timestamps": "00:06:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "Five trash cans: two in the kitchen, one in the bathroom, and two by the hallway.", + "D": "Six.", + "E": "Seven trash cans: three in the kitchen, two in the living room, and two near the bedroom." + }, + "answer_label": "D", + "answer_text": "Six.", + "q_id": "egolife_A1_D2_020-mcq_fixed_3", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "Five trash cans: two in the kitchen, one in the bathroom, and two by the hallway.", + "D": "Six.", + "E": "Seven trash cans: three in the kitchen, two in the living room, and two near the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_2_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the house, and where are they relative to me?", + "answer": "There is only one trash can in the home, and it is now under the table to your left.", + "question_chinese": "家里有几个垃圾桶？它在先对于我的什么位置？", + "answer_chinese": "家里只有客厅一个垃圾桶，它现在在你左边的桌子下面", + "evidence": [ + { + "description": "There is a trash can under the table in the living room", + "timestamps": { + "start": "00:01:48", + "end": "00:01:53" + } + } + ], + "question_timestamps": "00:02:01", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero trash cans.", + "B": "Two trash cans.", + "C": "There are two trash cans in the home, and both are under the table to your left.", + "D": "There is only one trash can in the home, and it is now under the table to your left.", + "E": "There are no trash cans in the home, so nothing is under the table to your left." + }, + "answer_label": "D", + "answer_text": "There is only one trash can in the home, and it is now under the table to your left.", + "q_id": "egolife_A1_D2_020_2_T2", + "correct_answer": "There is only one trash can in the home, and it is now under the table to your left.", + "original_options": { + "A": "Zero trash cans.", + "B": "Two trash cans.", + "C": "There are two trash cans in the home, and both are under the table to your left.", + "D": "There is only one trash can in the home, and it is now under the table to your left.", + "E": "There are no trash cans in the home, so nothing is under the table to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_2_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the house, and where are they relative to me?", + "answer": "There is only one trash can in the home, and it is under the table to your right.", + "question_chinese": "家里有几个垃圾桶？它在先对于我的什么位置？", + "answer_chinese": "家里只有客厅一个垃圾桶，它在你右边的桌子下面", + "evidence": [ + { + "description": "There is a trash can under the table in the living room", + "timestamps": { + "start": "00:01:48", + "end": "00:01:53" + } + }, + { + "description": "The trash can has stayed under the table the whole time, and you are on the left side of the table.", + "timestamps": { + "start": "00:02:50", + "end": "00:03:05" + } + } + ], + "question_timestamps": "00:04:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero trash cans.", + "B": "Two trash cans.", + "C": "There is only one trash can in the home, and it is under the table to your right.", + "D": "There are two trash cans in the home, and both are under the table to your right.", + "E": "There are no trash cans in the home, so nothing is under the table to your right." + }, + "answer_label": "C", + "answer_text": "There is only one trash can in the home, and it is under the table to your right.", + "q_id": "egolife_A1_D2_020_2_T1", + "correct_answer": "There is only one trash can in the home, and it is under the table to your right.", + "original_options": { + "A": "Zero trash cans.", + "B": "Two trash cans.", + "C": "There is only one trash can in the home, and it is under the table to your right.", + "D": "There are two trash cans in the home, and both are under the table to your right.", + "E": "There are no trash cans in the home, so nothing is under the table to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_7", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was I the last time I picked up the water bottle?", + "answer": "Bedroom door.", + "question_chinese": "我最近一次拿起水瓶时，在什么位置？", + "answer_chinese": "卧室门口", + "evidence": [ + { + "description": "The last time I picked up a water bottle was at the bedroom door", + "timestamps": { + "start": "00:03:15", + "end": "00:03:25" + } + } + ], + "question_timestamps": "00:03:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the hallway.", + "B": "Bedroom door.", + "C": "By the kitchen counter.", + "D": "You were standing next to the living room table when you picked up the water bottle.", + "E": "You were near the bathroom entrance the last time you picked up the water bottle." + }, + "answer_label": "B", + "answer_text": "Bedroom door.", + "q_id": "egolife_A1_D2_020_7", + "correct_answer": "Bedroom door.", + "original_options": { + "A": "In the hallway.", + "B": "Bedroom door.", + "C": "By the kitchen counter.", + "D": "You were standing next to the living room table when you picked up the water bottle.", + "E": "You were near the bathroom entrance the last time you picked up the water bottle." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_8", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was I the last time I picked up the water bottle?", + "answer": "The last time I picked up the water bottle was in the bedroom.", + "question_chinese": "我最近一次拿起水瓶时，在什么位置？", + "answer_chinese": "卧室里", + "evidence": [ + { + "description": "The last time I picked up the water bottle was in the bedroom", + "timestamps": { + "start": "00:07:10", + "end": "00:07:14" + } + } + ], + "question_timestamps": "00:07:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The last time I picked up the water bottle was in the bedroom.", + "B": "The last time was in the bathroom.", + "C": "The last time was in the hallway.", + "D": "The last time I picked up the water bottle was in the kitchen, near the counter.", + "E": "The last time I picked up the water bottle was in the living room, next to the table." + }, + "answer_label": "A", + "answer_text": "The last time I picked up the water bottle was in the bedroom.", + "q_id": "egolife_A1_D2_020_8", + "correct_answer": "The last time I picked up the water bottle was in the bedroom.", + "original_options": { + "A": "The last time I picked up the water bottle was in the bedroom.", + "B": "The last time was in the bathroom.", + "C": "The last time was in the hallway.", + "D": "The last time I picked up the water bottle was in the kitchen, near the counter.", + "E": "The last time I picked up the water bottle was in the living room, next to the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_4_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to use the microwave right now, how should I walk there?", + "answer": "You should first turn around and leave the bedroom.", + "question_chinese": "我现在想要用微波炉，应该怎么走？", + "answer_chinese": "你应该先向后转出卧室", + "evidence": [ + { + "description": "The microwave oven is beside the windowsill", + "timestamps": { + "start": "00:05:37", + "end": "00:05:41" + } + } + ], + "question_timestamps": "00:05:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk straight into the hallway.", + "B": "Turn left and stay in the bedroom.", + "C": "Turn around and go straight past the table, then turn right and continue forward to the microwave.", + "D": "Leave the bedroom, go around the table to the right, then turn left and walk straight to the microwave.", + "E": "You should first turn around and leave the bedroom." + }, + "answer_label": "E", + "answer_text": "You should first turn around and leave the bedroom.", + "q_id": "egolife_A1_D2_020_4_T2", + "correct_answer": "You should first turn around and leave the bedroom.", + "original_options": { + "A": "Walk straight into the hallway.", + "B": "Turn left and stay in the bedroom.", + "C": "Turn around and go straight past the table, then turn right and continue forward to the microwave.", + "D": "Leave the bedroom, go around the table to the right, then turn left and walk straight to the microwave.", + "E": "You should first turn around and leave the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_4_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to use the microwave right now, how should I walk there?", + "answer": "Turn around to leave the bedroom. You should walk straight for a bit to go around the table, then turn left and keep going straight.", + "question_chinese": "我现在想要用微波炉，应该怎么走？", + "answer_chinese": "向后转出卧室，你应该直走一段距离绕过桌子，然后左转并直行", + "evidence": [ + { + "description": "The microwave oven is beside the windowsill", + "timestamps": { + "start": "00:05:37", + "end": "00:05:41" + } + }, + { + "description": "The bedroom door to the refrigerator needs to pass diagonally through the living room and around the table", + "timestamps": { + "start": "00:06:03", + "end": "00:06:09" + } + } + ], + "question_timestamps": "00:06:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right after the table.", + "B": "Turn around to leave the bedroom. You should walk straight for a bit to go around the table, then turn left and keep going straight.", + "C": "Go left without turning around.", + "D": "Turn around to leave the bedroom, go around the table, then turn right and keep walking straight.", + "E": "Turn around and exit the bedroom, walk straight past the table, then turn left twice before continuing forward." + }, + "answer_label": "B", + "answer_text": "Turn around to leave the bedroom. You should walk straight for a bit to go around the table, then turn left and keep going straight.", + "q_id": "egolife_A1_D2_020_4_T1", + "correct_answer": "Turn around to leave the bedroom. You should walk straight for a bit to go around the table, then turn left and keep going straight.", + "original_options": { + "A": "Turn right after the table.", + "B": "Turn around to leave the bedroom. You should walk straight for a bit to go around the table, then turn left and keep going straight.", + "C": "Go left without turning around.", + "D": "Turn around to leave the bedroom, go around the table, then turn right and keep walking straight.", + "E": "Turn around and exit the bedroom, walk straight past the table, then turn left twice before continuing forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_020_10", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How do I get to the microwave?", + "answer": "Turn backward and take a step.", + "question_chinese": "我现在想要用微波炉，应该怎么走？", + "answer_chinese": "向后转并走一步", + "evidence": [ + { + "description": "The microwave oven is beside the windowsill", + "timestamps": { + "start": "00:05:37", + "end": "00:05:41" + } + }, + { + "description": "The microwave oven is behind the table compared to the bedroom door", + "timestamps": { + "start": "00:07:28", + "end": "00:07:35" + } + } + ], + "question_timestamps": "00:10:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn backward and take a step.", + "B": "Turn left and take a step.", + "C": "Turn right and take a step.", + "D": "Face forward, then take one step toward the microwave without turning around first.", + "E": "Turn around halfway, then step sideways and approach the microwave from the counter side." + }, + "answer_label": "A", + "answer_text": "Turn backward and take a step.", + "q_id": "egolife_A1_D2_020_10", + "correct_answer": "Turn backward and take a step.", + "original_options": { + "A": "Turn backward and take a step.", + "B": "Turn left and take a step.", + "C": "Turn right and take a step.", + "D": "Face forward, then take one step toward the microwave without turning around first.", + "E": "Turn around halfway, then step sideways and approach the microwave from the counter side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D2_021-mcq.json b/QAs/egolife_A1_D2_021-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a87c363a8e21f1bc90165a48645d58655b56bc69 --- /dev/null +++ b/QAs/egolife_A1_D2_021-mcq.json @@ -0,0 +1,1031 @@ +[ + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_10", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was I when I just picked up my phone?", + "answer": "In the bedroom.", + "question_chinese": "我刚刚拿起手机时在什么地方", + "answer_chinese": "卧室里", + "evidence": [ + { + "description": "I picked up my mobile phone in the bedroom", + "timestamps": { + "start": "00:00:43", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the hallway.", + "B": "In the living room.", + "C": "In the study, standing by the desk, with your laptop nearby on the tabletop.", + "D": "In the bedroom.", + "E": "In the kitchen, right by the counter where you were preparing a drink at that moment." + }, + "answer_label": "D", + "answer_text": "In the bedroom.", + "q_id": "egolife_A1_D2_021_10", + "correct_answer": "In the bedroom.", + "original_options": { + "A": "In the hallway.", + "B": "In the living room.", + "C": "In the study, standing by the desk, with your laptop nearby on the tabletop.", + "D": "In the bedroom.", + "E": "In the kitchen, right by the counter where you were preparing a drink at that moment." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_9", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was I when I just picked up my phone?", + "answer": "I was at the bedroom door at that time.", + "question_chinese": "我刚刚拿起手机时在什么地方", + "answer_chinese": "我当时在卧室门口", + "evidence": [ + { + "description": "I picked up my mobile phone when I was at the bedroom door", + "timestamps": { + "start": "00:07:49", + "end": "00:07:52" + } + } + ], + "question_timestamps": "00:08:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I was by the sink.", + "B": "I was in the living room.", + "C": "I was near the kitchen table at that time, facing toward the counter.", + "D": "I was at the bedroom door at that time.", + "E": "I was beside the window at that time, a few steps away from the bed." + }, + "answer_label": "D", + "answer_text": "I was at the bedroom door at that time.", + "q_id": "egolife_A1_D2_021_9", + "correct_answer": "I was at the bedroom door at that time.", + "original_options": { + "A": "I was by the sink.", + "B": "I was in the living room.", + "C": "I was near the kitchen table at that time, facing toward the counter.", + "D": "I was at the bedroom door at that time.", + "E": "I was beside the window at that time, a few steps away from the bed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_11", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I get a bowl?", + "answer": "On the table.", + "question_chinese": "我在哪里可以拿到碗？", + "answer_chinese": "桌子上", + "evidence": [ + { + "description": "The bowl is on the table", + "timestamps": { + "start": "00:02:19", + "end": "00:02:29" + } + } + ], + "question_timestamps": "00:02:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the table.", + "B": "On the counter.", + "C": "In the cabinet.", + "D": "On the shelf by the window, near the other dishes laid out in a small stack.", + "E": "On the cabinet near the door, next to the stacked plates and a few cups." + }, + "answer_label": "A", + "answer_text": "On the table.", + "q_id": "egolife_A1_D2_021_11", + "correct_answer": "On the table.", + "original_options": { + "A": "On the table.", + "B": "On the counter.", + "C": "In the cabinet.", + "D": "On the shelf by the window, near the other dishes laid out in a small stack.", + "E": "On the cabinet near the door, next to the stacked plates and a few cups." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_12", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where can I get a bowl?", + "answer": "Washing-up sink.", + "question_chinese": "我在哪里可以拿到碗？", + "answer_chinese": "洗碗池", + "evidence": [ + { + "description": "The bowl is in the sink", + "timestamps": { + "start": "00:08:35", + "end": "00:08:53" + } + } + ], + "question_timestamps": "00:10:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Kitchen cabinet.", + "B": "On the kitchen island.", + "C": "Washing-up sink.", + "D": "Under the dining table, tucked in close to the chair legs where you might miss it.", + "E": "On the refrigerator shelf, placed behind other containers inside the fridge compartment." + }, + "answer_label": "C", + "answer_text": "Washing-up sink.", + "q_id": "egolife_A1_D2_021_12", + "correct_answer": "Washing-up sink.", + "original_options": { + "A": "Kitchen cabinet.", + "B": "On the kitchen island.", + "C": "Washing-up sink.", + "D": "Under the dining table, tucked in close to the chair legs where you might miss it.", + "E": "On the refrigerator shelf, placed behind other containers inside the fridge compartment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 16, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_5", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to use the washing machine now?", + "answer": "One.", + "question_chinese": "我现在想要用洗衣机，至少需要拐几个弯？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "The washing machine is at the main entrance of the house", + "timestamps": { + "start": "00:02:26", + "end": "00:02:46" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four turns are needed before you can face the washing machine directly and use it now.", + "D": "Five turns are required to line yourself up with the washing machine controls in front of you.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_D2_021_5", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four turns are needed before you can face the washing machine directly and use it now.", + "D": "Five turns are required to line yourself up with the washing machine controls in front of you.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_6", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many turns do I need to make to use the washing machine now?", + "answer": "Two.", + "question_chinese": "我现在想要用洗衣机，至少需要拐几个弯？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "The washing machine is at the main entrance of the house", + "timestamps": { + "start": "00:02:26", + "end": "00:02:46" + } + } + ], + "question_timestamps": "00:07:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Seven turns are needed before you can reach the washing machine and start using it.", + "E": "Eight turns are required to end up facing the washing machine directly, ready to operate it." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "egolife_A1_D2_021_6", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Seven turns are needed before you can reach the washing machine and start using it.", + "E": "Eight turns are required to end up facing the washing machine directly, ready to operate it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get some dish soap now?", + "answer": "Yes.", + "question_chinese": "我现在能否拿到洗洁精？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The detergent is by the sink", + "timestamps": { + "start": "00:08:37", + "end": "00:08:41" + } + } + ], + "question_timestamps": "00:09:06", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D2_021_1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get some dish soap now?", + "answer": "No.", + "question_chinese": "我现在能否拿到洗洁精？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The detergent is by the sink", + "timestamps": { + "start": "00:08:37", + "end": "00:08:41" + } + } + ], + "question_timestamps": "00:11:56", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D2_021_2", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_4_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers do I see in the whole room?", + "answer": "You saw five of them in the room in front of you.", + "question_chinese": "我看见整个屋子有几台电脑？", + "answer_chinese": "你看见有五台，在你面前的屋子里", + "evidence": [ + { + "description": "There are seven computers in the bedroom", + "timestamps": { + "start": "00:00:01", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "You saw five of them in the room in front of you.", + "C": "Four.", + "D": "You saw six of them in the room in front of you.", + "E": "You saw seven of them in the room in front of you." + }, + "answer_label": "B", + "answer_text": "You saw five of them in the room in front of you.", + "q_id": "egolife_A1_D2_021_4_T2", + "correct_answer": "You saw five of them in the room in front of you.", + "original_options": { + "A": "Three.", + "B": "You saw five of them in the room in front of you.", + "C": "Four.", + "D": "You saw six of them in the room in front of you.", + "E": "You saw seven of them in the room in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_5_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers do I see in the whole room?", + "answer": "You saw seven of them in the room in front of you.", + "question_chinese": "我看见整个屋子有几台电脑？", + "answer_chinese": "你看见有七台，在你面前的屋子里", + "evidence": [ + { + "description": "There are seven computers in the bedroom", + "timestamps": { + "start": "00:06:15", + "end": "00:06:37" + } + } + ], + "question_timestamps": "00:06:39", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Six.", + "C": "You saw seven of them in the room in front of you.", + "D": "You saw eight of them in the room in front of you.", + "E": "You saw nine of them in the room in front of you." + }, + "answer_label": "C", + "answer_text": "You saw seven of them in the room in front of you.", + "q_id": "egolife_A1_D2_021_5_T2", + "correct_answer": "You saw seven of them in the room in front of you.", + "original_options": { + "A": "Five.", + "B": "Six.", + "C": "You saw seven of them in the room in front of you.", + "D": "You saw eight of them in the room in front of you.", + "E": "You saw nine of them in the room in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_4_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers do I see in the whole room?", + "answer": "You saw seven of them in the room in front of you.", + "question_chinese": "我看见整个屋子有几台电脑？", + "answer_chinese": "你看见有七台，在你面前的屋子里", + "evidence": [ + { + "description": "There are seven computers in the bedroom", + "timestamps": { + "start": "00:06:15", + "end": "00:06:37" + } + } + ], + "question_timestamps": "00:07:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw seven of them in the room in front of you.", + "B": "Five.", + "C": "Six.", + "D": "You saw eight of them in the room in front of you.", + "E": "You saw nine of them in the room in front of you." + }, + "answer_label": "A", + "answer_text": "You saw seven of them in the room in front of you.", + "q_id": "egolife_A1_D2_021_4_T1", + "correct_answer": "You saw seven of them in the room in front of you.", + "original_options": { + "A": "You saw seven of them in the room in front of you.", + "B": "Five.", + "C": "Six.", + "D": "You saw eight of them in the room in front of you.", + "E": "You saw nine of them in the room in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_5_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers do I see in the whole room?", + "answer": "You saw eight of them in the room in front of you.", + "question_chinese": "我看见整个屋子有几台电脑？", + "answer_chinese": "你看见有八台，在你面前的屋子里", + "evidence": [ + { + "description": "There are seven computers in the bedroom", + "timestamps": { + "start": "00:06:15", + "end": "00:06:37" + } + }, + { + "description": "There is a computer on the table in the living room", + "timestamps": { + "start": "00:07:52", + "end": "00:08:07" + } + } + ], + "question_timestamps": "00:08:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "You saw eight of them in the room in front of you.", + "C": "Seven.", + "D": "You saw nine of them in the room in front of you.", + "E": "You saw ten of them in the room in front of you." + }, + "answer_label": "B", + "answer_text": "You saw eight of them in the room in front of you.", + "q_id": "egolife_A1_D2_021_5_T1", + "correct_answer": "You saw eight of them in the room in front of you.", + "original_options": { + "A": "Six.", + "B": "You saw eight of them in the room in front of you.", + "C": "Seven.", + "D": "You saw nine of them in the room in front of you.", + "E": "You saw ten of them in the room in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021-mcq_fixed_12", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the room? Where are they respectively?", + "answer": "Four.", + "question_chinese": "房间里有几个垃圾桶？分别在哪里？", + "answer_chinese": "一个，在桌子下面", + "evidence": [ + { + "description": "There is a trash can under the table", + "timestamps": { + "start": "00:02:19", + "end": "00:02:29" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One trash can.", + "B": "Four.", + "C": "Two trash cans.", + "D": "Three trash cans: one by the door, one near the table, and one beside the refrigerator.", + "E": "Five trash cans: one by the door, two near the table, one beside the refrigerator, and one in the corner." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "egolife_A1_D2_021-mcq_fixed_12", + "correct_answer": "Four.", + "original_options": { + "A": "One trash can.", + "B": "Four.", + "C": "Two trash cans.", + "D": "Three trash cans: one by the door, one near the table, and one beside the refrigerator.", + "E": "Five trash cans: one by the door, two near the table, one beside the refrigerator, and one in the corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021-mcq_fixed_13", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the room? Where are they respectively?", + "answer": "Five.", + "question_chinese": "房间里有几个垃圾桶？分别在哪里？", + "answer_chinese": "两个，一个在桌子下面，另一个在洗碗池旁", + "evidence": [ + { + "description": "There is a trash can under the table", + "timestamps": { + "start": "00:02:19", + "end": "00:02:29" + } + }, + { + "description": "There is a trash can beside the sink", + "timestamps": { + "start": "00:09:15", + "end": "00:09:21" + } + } + ], + "question_timestamps": "00:09:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Two.", + "C": "Four.", + "D": "There are three trash cans in the room, one by the sink, one near the door, and one by a desk.", + "E": "There are six trash cans in the room, placed near the sink, door, desk, and other corners." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D2_021-mcq_fixed_13", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "Two.", + "C": "Four.", + "D": "There are three trash cans in the room, one by the sink, one near the door, and one by a desk.", + "E": "There are six trash cans in the room, placed near the sink, door, desk, and other corners." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 21, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_7", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are there in the whole house?", + "answer": "Seven.", + "question_chinese": "整个屋子有几台电脑？", + "answer_chinese": "七台", + "evidence": [ + { + "description": "There are seven computers in the bedroom", + "timestamps": { + "start": "00:06:15", + "end": "00:06:37" + } + } + ], + "question_timestamps": "00:07:14", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Six.", + "C": "There are eight computers in the whole house in total, including the ones on the desks and in bedrooms.", + "D": "Seven.", + "E": "There are nine computers in the whole house in total, including the ones on the desks and in bedrooms." + }, + "answer_label": "D", + "answer_text": "Seven.", + "q_id": "egolife_A1_D2_021_7", + "correct_answer": "Seven.", + "original_options": { + "A": "Five.", + "B": "Six.", + "C": "There are eight computers in the whole house in total, including the ones on the desks and in bedrooms.", + "D": "Seven.", + "E": "There are nine computers in the whole house in total, including the ones on the desks and in bedrooms." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_8", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are there in the entire house?", + "answer": "Eight.", + "question_chinese": "整个屋子有几台电脑？", + "answer_chinese": "八台", + "evidence": [ + { + "description": "There are seven computers in the bedroom", + "timestamps": { + "start": "00:06:15", + "end": "00:06:37" + } + }, + { + "description": "There is a computer on the table in the living room", + "timestamps": { + "start": "00:07:52", + "end": "00:08:07" + } + } + ], + "question_timestamps": "00:08:26", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Eight.", + "B": "Seven.", + "C": "Nine.", + "D": "Six computers total across the entire house, counted in every room and workspace area.", + "E": "Ten computers total throughout the house after checking all rooms, desks, and common areas." + }, + "answer_label": "A", + "answer_text": "Eight.", + "q_id": "egolife_A1_D2_021_8", + "correct_answer": "Eight.", + "original_options": { + "A": "Eight.", + "B": "Seven.", + "C": "Nine.", + "D": "Six computers total across the entire house, counted in every room and workspace area.", + "E": "Ten computers total throughout the house after checking all rooms, desks, and common areas." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_3", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the mirror or the microwave?", + "answer": "Mirror.", + "question_chinese": "镜子和微波炉哪个理我更近一些？", + "answer_chinese": "镜子", + "evidence": [ + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:07:30", + "end": "00:07:35" + } + }, + { + "description": "The microwave oven is by the window", + "timestamps": { + "start": "00:02:28", + "end": "00:02:32" + } + } + ], + "question_timestamps": "00:08:01", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Microwave oven.", + "B": "Mirror." + }, + "answer_label": "B", + "answer_text": "Mirror.", + "q_id": "egolife_A1_D2_021_3", + "correct_answer": "Mirror.", + "original_options": { + "A": "Microwave oven.", + "B": "Mirror." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D2_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_021_4", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the mirror or the microwave?", + "answer": "Microwave oven.", + "question_chinese": "镜子和微波炉哪个理我更近一些？", + "answer_chinese": "微波炉", + "evidence": [ + { + "description": "The mirror is beside the bedroom door", + "timestamps": { + "start": "00:07:30", + "end": "00:07:35" + } + }, + { + "description": "The microwave oven is by the window", + "timestamps": { + "start": "00:02:28", + "end": "00:02:32" + } + } + ], + "question_timestamps": "00:09:31", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Microwave oven.", + "B": "Mirror." + }, + "answer_label": "A", + "answer_text": "Microwave oven.", + "q_id": "egolife_A1_D2_021_4", + "correct_answer": "Microwave oven.", + "original_options": { + "A": "Microwave oven.", + "B": "Mirror." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D2_023-mcq.json b/QAs/egolife_A1_D2_023-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1564413ee7009f4ef0a6664a40ff298619713bdc --- /dev/null +++ b/QAs/egolife_A1_D2_023-mcq.json @@ -0,0 +1,1102 @@ +[ + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_8", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the refrigerator or the table?", + "answer": "The table.", + "question_chinese": "冰箱和桌子哪个离我更近？", + "answer_chinese": "桌子", + "evidence": [ + { + "description": "The table is on the side of the living room far from the windowsill", + "timestamps": { + "start": "00:06:05", + "end": "00:06:12" + } + }, + { + "description": "The refrigerator is on the side of the room near the windowsill", + "timestamps": { + "start": "00:07:09", + "end": "00:07:12" + } + }, + { + "description": "I'm right beside the table now", + "timestamps": { + "start": "00:10:00", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:10:17", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The table.", + "B": "The refrigerator." + }, + "answer_label": "A", + "answer_text": "The table.", + "q_id": "egolife_A1_D2_023_8", + "correct_answer": "The table.", + "original_options": { + "A": "The table.", + "B": "The refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_5_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the refrigerator or the table?", + "answer": "The fridge is closer to you.", + "question_chinese": "冰箱和桌子哪个离我更近？", + "answer_chinese": "冰箱离你更近", + "evidence": [ + { + "description": "The table is on the side of the living room far from the windowsill", + "timestamps": { + "start": "00:06:05", + "end": "00:06:12" + } + }, + { + "description": "The refrigerator is on the side of the room near the windowsill", + "timestamps": { + "start": "00:07:09", + "end": "00:07:12" + } + }, + { + "description": "I'm right beside the table now", + "timestamps": { + "start": "00:10:00", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The fridge is closer to you.", + "B": "The table is closer to you." + }, + "answer_label": "A", + "answer_text": "The fridge is closer to you.", + "q_id": "egolife_A1_D2_023_5_T2", + "correct_answer": "The fridge is closer to you.", + "original_options": { + "A": "The fridge is closer to you.", + "B": "The table is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_5_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the refrigerator or the table?", + "answer": "The table is closer to you.", + "question_chinese": "冰箱和桌子哪个离我更近？", + "answer_chinese": "桌子离你更近", + "evidence": [ + { + "description": "The table is on the side of the living room far from the windowsill", + "timestamps": { + "start": "00:06:05", + "end": "00:06:12" + } + }, + { + "description": "The refrigerator is on the side of the room near the windowsill", + "timestamps": { + "start": "00:07:09", + "end": "00:07:12" + } + }, + { + "description": "I'm right beside the table now", + "timestamps": { + "start": "00:10:00", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:10:27", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The table is closer to you.", + "B": "The fridge is closer to you." + }, + "answer_label": "A", + "answer_text": "The table is closer to you.", + "q_id": "egolife_A1_D2_023_5_T1", + "correct_answer": "The table is closer to you.", + "original_options": { + "A": "The table is closer to you.", + "B": "The fridge is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_3", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the house?", + "answer": "One.", + "question_chinese": "家里一共有几个垃圾桶？", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There is a trash can under the table on the first floor", + "timestamps": { + "start": "00:05:46", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:06:09", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four trash cans in total around the house, including the kitchen, bathroom, and other rooms checked.", + "D": "One.", + "E": "Five trash cans altogether when you check every room, including the kitchen and both bathrooms." + }, + "answer_label": "D", + "answer_text": "One.", + "q_id": "egolife_A1_D2_023_3", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four trash cans in total around the house, including the kitchen, bathroom, and other rooms checked.", + "D": "One.", + "E": "Five trash cans altogether when you check every room, including the kitchen and both bathrooms." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_4", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash bins are there in the house?", + "answer": "Three.", + "question_chinese": "家里一共有几个垃圾桶？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There is a trash can under the table on the first floor", + "timestamps": { + "start": "00:05:46", + "end": "00:05:50" + } + }, + { + "description": "There is a trash can under the table on the second floor", + "timestamps": { + "start": "00:06:04", + "end": "00:06:12" + } + }, + { + "description": "There is a trash can in front of the windowsill on the second floor", + "timestamps": { + "start": "00:07:07", + "end": "00:07:14" + } + } + ], + "question_timestamps": "00:10:36", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "Three.", + "D": "Six trash bins total throughout the house, counted across the kitchen, bathrooms, and other rooms.", + "E": "Seven trash bins altogether when you count every room, including kitchen areas and all bathrooms." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_023_4", + "correct_answer": "Three.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "Three.", + "D": "Six trash bins total throughout the house, counted across the kitchen, bathrooms, and other rooms.", + "E": "Seven trash bins altogether when you count every room, including kitchen areas and all bathrooms." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_5", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I just picked up the foam board, which direction is the windowsill from me?", + "answer": "Right astern.", + "question_chinese": "我刚刚拿起泡沫板时，窗台在我的什么方向？", + "answer_chinese": "正后方", + "evidence": [ + { + "description": "When I just picked up the foam board, the windowsill was right behind me", + "timestamps": { + "start": "00:06:54", + "end": "00:07:01" + } + } + ], + "question_timestamps": "00:07:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left astern.", + "B": "Right astern.", + "C": "Right ahead.", + "D": "Slightly off your left shoulder, behind you, not directly to the right of your back.", + "E": "Ahead and to the left, as if you had turned away from the window before lifting the board." + }, + "answer_label": "B", + "answer_text": "Right astern.", + "q_id": "egolife_A1_D2_023_5", + "correct_answer": "Right astern.", + "original_options": { + "A": "Left astern.", + "B": "Right astern.", + "C": "Right ahead.", + "D": "Slightly off your left shoulder, behind you, not directly to the right of your back.", + "E": "Ahead and to the left, as if you had turned away from the window before lifting the board." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_6", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I just picked up the foam board, which direction is the windowsill from me?", + "answer": "The left side.", + "question_chinese": "我刚刚拿起泡沫板时，窗台在我的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "When I just picked up the foam board, the windowsill was on my left", + "timestamps": { + "start": "00:10:32", + "end": "00:10:38" + } + } + ], + "question_timestamps": "00:11:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The right side.", + "B": "Straight behind you.", + "C": "The left side.", + "D": "Slightly in front of you on the right, closer to your front-right than your left side.", + "E": "Behind and a bit to the left, as if the window was over your left shoulder rather than beside you." + }, + "answer_label": "C", + "answer_text": "The left side.", + "q_id": "egolife_A1_D2_023_6", + "correct_answer": "The left side.", + "original_options": { + "A": "The right side.", + "B": "Straight behind you.", + "C": "The left side.", + "D": "Slightly in front of you on the right, closer to your front-right than your left side.", + "E": "Behind and a bit to the left, as if the window was over your left shoulder rather than beside you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_6_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the motion trajectory when I picked up the foam board?", + "answer": "You picked it up from beside the bathroom door.", + "question_chinese": "我拿泡沫板的运动轨迹是怎样的？", + "answer_chinese": "它被你从卫生间门旁拿起来了", + "evidence": [ + { + "description": "You took the foam board away from the bathroom door.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:03" + } + } + ], + "question_timestamps": "00:07:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked it up from beside the bathroom door.", + "B": "You picked it up from the seat cushion.", + "C": "You picked it up from the floor.", + "D": "You first picked it up from the seat cushion, then moved it in a straight line to the bathroom doorway area.", + "E": "After lifting it beside the bathroom door, you carried it in an arc and placed it down near the living room center." + }, + "answer_label": "A", + "answer_text": "You picked it up from beside the bathroom door.", + "q_id": "egolife_A1_D2_023_6_T2", + "correct_answer": "You picked it up from beside the bathroom door.", + "original_options": { + "A": "You picked it up from beside the bathroom door.", + "B": "You picked it up from the seat cushion.", + "C": "You picked it up from the floor.", + "D": "You first picked it up from the seat cushion, then moved it in a straight line to the bathroom doorway area.", + "E": "After lifting it beside the bathroom door, you carried it in an arc and placed it down near the living room center." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 21, + 22 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_6_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the motion trajectory when I picked up the foam board?", + "answer": "You first picked it up from beside the bathroom door, then it moved in a straight line to the seat cushion.", + "question_chinese": "我拿泡沫板的运动轨迹是怎样的？", + "answer_chinese": "它被你先从卫生间门旁拿起来，然后直线运动到了坐垫上", + "evidence": [ + { + "description": "You first picked it up from beside the bathroom door, and then moved it in a straight line onto the seat cushion.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:18" + } + } + ], + "question_timestamps": "00:07:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked it up from the seat cushion.", + "B": "You first picked it up from beside the bathroom door, then it moved in a straight line to the seat cushion.", + "C": "You picked it up from the floor center.", + "D": "You picked it up beside the bathroom door and moved it in a straight line to the living room floor center.", + "E": "You first picked it up from the seat cushion, then moved it in a straight line to beside the bathroom door." + }, + "answer_label": "B", + "answer_text": "You first picked it up from beside the bathroom door, then it moved in a straight line to the seat cushion.", + "q_id": "egolife_A1_D2_023_6_T1", + "correct_answer": "You first picked it up from beside the bathroom door, then it moved in a straight line to the seat cushion.", + "original_options": { + "A": "You picked it up from the seat cushion.", + "B": "You first picked it up from beside the bathroom door, then it moved in a straight line to the seat cushion.", + "C": "You picked it up from the floor center.", + "D": "You picked it up beside the bathroom door and moved it in a straight line to the living room floor center.", + "E": "You first picked it up from the seat cushion, then moved it in a straight line to beside the bathroom door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the water cup now?", + "answer": "Yes.", + "question_chinese": "我现在可以拿到水杯吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The water cup is on the table", + "timestamps": { + "start": "00:00:35", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:01:44", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D2_023_1", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the water cup now?", + "answer": "No.", + "question_chinese": "我现在可以拿到水杯吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The water cup is on the table", + "timestamps": { + "start": "00:00:35", + "end": "00:00:43" + } + }, + { + "description": "I'm at the bedroom door at this moment, at a certain distance from the table", + "timestamps": { + "start": "00:05:26", + "end": "00:05:35" + } + } + ], + "question_timestamps": "00:05:40", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D2_023_2", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_7_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the motion trajectory as I’m picking up the foam board right now?", + "answer": "After it was picked up from the seat cushion, it moved in a straight line to the floor at the very center of the living room.", + "question_chinese": "我现在拿泡沫板的运动轨迹是怎样的？", + "answer_chinese": "它被从坐垫拿起来后，直线运动到了客厅正中心的地板上", + "evidence": [ + { + "description": "After being lifted from the cushion, it moved in a straight line to the floor right in the center of the living room", + "timestamps": { + "start": "00:10:18", + "end": "00:10:23" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Picked up beside bathroom door.", + "B": "After it was picked up from the seat cushion, it moved in a straight line to the floor at the very center of the living room.", + "C": "Moved onto the seat cushion.", + "D": "You first picked it up beside the bathroom door, then moved it in a straight line to the seat cushion.", + "E": "After picking it up from the seat cushion, it moved in a straight line to the floor near the living room edge." + }, + "answer_label": "B", + "answer_text": "After it was picked up from the seat cushion, it moved in a straight line to the floor at the very center of the living room.", + "q_id": "egolife_A1_D2_023_7_T1", + "correct_answer": "After it was picked up from the seat cushion, it moved in a straight line to the floor at the very center of the living room.", + "original_options": { + "A": "Picked up beside bathroom door.", + "B": "After it was picked up from the seat cushion, it moved in a straight line to the floor at the very center of the living room.", + "C": "Moved onto the seat cushion.", + "D": "You first picked it up beside the bathroom door, then moved it in a straight line to the seat cushion.", + "E": "After picking it up from the seat cushion, it moved in a straight line to the floor near the living room edge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 22 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_7_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the motion trajectory as I’m picking up the foam board right now?", + "answer": "You picked it up from the seat cushion.", + "question_chinese": "我现在拿泡沫板的运动轨迹是怎样的？", + "answer_chinese": "它被从你从坐垫拿起来了", + "evidence": [ + { + "description": "After being lifted from the cushion", + "timestamps": { + "start": "00:10:18", + "end": "00:10:19" + } + } + ], + "question_timestamps": "00:10:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From bathroom door area.", + "B": "From the floor center.", + "C": "You picked it up from the seat cushion.", + "D": "You first picked it up beside the bathroom door, then moved it in a straight line to the seat cushion.", + "E": "After lifting it from the seat cushion, it moved in a straight line to the floor at the very center." + }, + "answer_label": "C", + "answer_text": "You picked it up from the seat cushion.", + "q_id": "egolife_A1_D2_023_7_T2", + "correct_answer": "You picked it up from the seat cushion.", + "original_options": { + "A": "From bathroom door area.", + "B": "From the floor center.", + "C": "You picked it up from the seat cushion.", + "D": "You first picked it up beside the bathroom door, then moved it in a straight line to the seat cushion.", + "E": "After lifting it from the seat cushion, it moved in a straight line to the floor at the very center." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_9", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the foam board from the moment it is picked up until it is set down?", + "answer": "It was picked up from beside the bathroom door and then moved in a straight line onto the cushion.", + "question_chinese": "泡沫板从被拿起到被放下期间的运动轨迹是怎样的？", + "answer_chinese": "它被从卫生间门旁拿起来，然后直线运动到了坐垫上", + "evidence": [ + { + "description": "What is the movement trajectory of the foam board from being picked up to being put down?", + "timestamps": { + "start": "00:07:00", + "end": "00:07:18" + } + } + ], + "question_timestamps": "00:07:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was picked up from beside the bathroom door and then moved in a straight line onto the cushion.", + "B": "It was carried to the floor.", + "C": "It was placed on the cushion edge.", + "D": "It was picked up beside the bathroom door, then carried diagonally and set down on the cushion.", + "E": "It was lifted from near the bathroom door and moved straight before being set on the floor by the cushion." + }, + "answer_label": "A", + "answer_text": "It was picked up from beside the bathroom door and then moved in a straight line onto the cushion.", + "q_id": "egolife_A1_D2_023_9", + "correct_answer": "It was picked up from beside the bathroom door and then moved in a straight line onto the cushion.", + "original_options": { + "A": "It was picked up from beside the bathroom door and then moved in a straight line onto the cushion.", + "B": "It was carried to the floor.", + "C": "It was placed on the cushion edge.", + "D": "It was picked up beside the bathroom door, then carried diagonally and set down on the cushion.", + "E": "It was lifted from near the bathroom door and moved straight before being set on the floor by the cushion." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_10", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the foam board from the moment it is picked up to when it is set down?", + "answer": "After being lifted from the cushion, it moved in a straight line to the floor right in the center of the living room.", + "question_chinese": "泡沫板从被拿起到被放下期间的运动轨迹是怎样的？", + "answer_chinese": "它被从坐垫拿起来后，直线运动到了客厅正中心的地板上", + "evidence": [ + { + "description": "After being lifted from the cushion, it moved in a straight line to the floor right in the center of the living room", + "timestamps": { + "start": "00:10:18", + "end": "00:10:23" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was moved onto the cushion.", + "B": "It was set down near the wall.", + "C": "After being lifted from the cushion, it traveled in a straight line to the floor near the sofa.", + "D": "After being lifted from the cushion, it was carried straight and placed on the floor along the living room edge.", + "E": "After being lifted from the cushion, it moved in a straight line to the floor right in the center of the living room." + }, + "answer_label": "E", + "answer_text": "After being lifted from the cushion, it moved in a straight line to the floor right in the center of the living room.", + "q_id": "egolife_A1_D2_023_10", + "correct_answer": "After being lifted from the cushion, it moved in a straight line to the floor right in the center of the living room.", + "original_options": { + "A": "It was moved onto the cushion.", + "B": "It was set down near the wall.", + "C": "After being lifted from the cushion, it traveled in a straight line to the floor near the sofa.", + "D": "After being lifted from the cushion, it was carried straight and placed on the floor along the living room edge.", + "E": "After being lifted from the cushion, it moved in a straight line to the floor right in the center of the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_2_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many trash cans have I seen in the house in total?", + "answer": "So far, you haven’t seen any trash cans in the home.", + "question_chinese": "目前为止，我看到家里一共有几个垃圾桶？", + "answer_chinese": "你目前没看到家里有垃圾桶", + "evidence": [ + { + "description": "You still haven't seen the trash can at home.", + "timestamps": { + "start": "00:00:00", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One trash can.", + "B": "So far, you haven’t seen any trash cans in the home.", + "C": "Two trash cans.", + "D": "So far, you’ve seen one trash can in the home in total.", + "E": "Up to now, you’ve seen a total of two trash cans somewhere in the home." + }, + "answer_label": "B", + "answer_text": "So far, you haven’t seen any trash cans in the home.", + "q_id": "egolife_A1_D2_023_2_T2", + "correct_answer": "So far, you haven’t seen any trash cans in the home.", + "original_options": { + "A": "One trash can.", + "B": "So far, you haven’t seen any trash cans in the home.", + "C": "Two trash cans.", + "D": "So far, you’ve seen one trash can in the home in total.", + "E": "Up to now, you’ve seen a total of two trash cans somewhere in the home." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_2_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many trash cans have I seen in the house in total?", + "answer": "So far, I have seen one trash can in the home.", + "question_chinese": "目前为止，我看到家里一共有几个垃圾桶？", + "answer_chinese": "目前为止，我看到家里有一个垃圾桶", + "evidence": [ + { + "description": "There is a trash can under the table on the first floor", + "timestamps": { + "start": "00:05:46", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:06:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, I have seen one trash can in the home.", + "B": "Zero trash cans.", + "C": "Two trash cans.", + "D": "So far, I have seen three trash cans in the home total.", + "E": "Up to now, I have seen a total of four trash cans in the home." + }, + "answer_label": "A", + "answer_text": "So far, I have seen one trash can in the home.", + "q_id": "egolife_A1_D2_023_2_T1", + "correct_answer": "So far, I have seen one trash can in the home.", + "original_options": { + "A": "So far, I have seen one trash can in the home.", + "B": "Zero trash cans.", + "C": "Two trash cans.", + "D": "So far, I have seen three trash cans in the home total.", + "E": "Up to now, I have seen a total of four trash cans in the home." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_3_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many trash cans have I seen in the house in total?", + "answer": "Up to now, you’ve seen a total of two trash cans in the home.", + "question_chinese": "目前为止，我看到家里一共有几个垃圾桶？", + "answer_chinese": "到目前，你看到家里一共有两个垃圾桶", + "evidence": [ + { + "description": "There is a trash can under the table on the first floor", + "timestamps": { + "start": "00:05:46", + "end": "00:05:50" + } + }, + { + "description": "There is a trash can under the table on the second floor", + "timestamps": { + "start": "00:06:04", + "end": "00:06:12" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero trash cans.", + "B": "One trash can.", + "C": "Up to now, you’ve seen a total of two trash cans in the home.", + "D": "Up to now, you’ve seen a total of three trash cans in the home.", + "E": "Up to now, you’ve seen a total of four trash cans throughout the home." + }, + "answer_label": "C", + "answer_text": "Up to now, you’ve seen a total of two trash cans in the home.", + "q_id": "egolife_A1_D2_023_3_T2", + "correct_answer": "Up to now, you’ve seen a total of two trash cans in the home.", + "original_options": { + "A": "Zero trash cans.", + "B": "One trash can.", + "C": "Up to now, you’ve seen a total of two trash cans in the home.", + "D": "Up to now, you’ve seen a total of three trash cans in the home.", + "E": "Up to now, you’ve seen a total of four trash cans throughout the home." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_023_3_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many trash cans have I seen in the house in total?", + "answer": "Up to now, you’ve seen three trash cans in the home.", + "question_chinese": "目前为止，我看到家里一共有几个垃圾桶？", + "answer_chinese": "目前你看到家里有三个垃圾桶", + "evidence": [ + { + "description": "There is a trash can under the table on the first floor", + "timestamps": { + "start": "00:05:46", + "end": "00:05:50" + } + }, + { + "description": "There is a trash can under the table on the second floor", + "timestamps": { + "start": "00:06:04", + "end": "00:06:12" + } + }, + { + "description": "There is a trash can in front of the windowsill on the second floor", + "timestamps": { + "start": "00:07:07", + "end": "00:07:14" + } + } + ], + "question_timestamps": "00:10:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One trash can.", + "B": "Up to now, you’ve seen three trash cans in the home.", + "C": "Two trash cans.", + "D": "Up to now, you’ve seen four trash cans in the home.", + "E": "Up to now, you’ve seen five trash cans total in the home." + }, + "answer_label": "B", + "answer_text": "Up to now, you’ve seen three trash cans in the home.", + "q_id": "egolife_A1_D2_023_3_T1", + "correct_answer": "Up to now, you’ve seen three trash cans in the home.", + "original_options": { + "A": "One trash can.", + "B": "Up to now, you’ve seen three trash cans in the home.", + "C": "Two trash cans.", + "D": "Up to now, you’ve seen four trash cans in the home.", + "E": "Up to now, you’ve seen five trash cans total in the home." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D2_024-mcq.json b/QAs/egolife_A1_D2_024-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6f77ac98de2f1457d5a11d09c33d0972ae91d50d --- /dev/null +++ b/QAs/egolife_A1_D2_024-mcq.json @@ -0,0 +1,873 @@ +[ + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of mineral water are on the table in the living room on the second floor?", + "answer": "Two.", + "question_chinese": "二楼客厅桌子上有几瓶矿泉水？", + "answer_chinese": "两瓶", + "evidence": [ + { + "description": "There are two bottles of mineral water on the table", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four bottles of mineral water are on the table in the upstairs living room.", + "D": "Two.", + "E": "Five bottles of mineral water are sitting on the living room table on the second floor." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "egolife_A1_D2_024_1", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four bottles of mineral water are on the table in the upstairs living room.", + "D": "Two.", + "E": "Five bottles of mineral water are sitting on the living room table on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of mineral water are on the table in the living room on the second floor?", + "answer": "Zero.", + "question_chinese": "二楼客厅桌子上有几瓶矿泉水？", + "answer_chinese": "零", + "evidence": [ + { + "description": "There is no mineral water on the table", + "timestamps": { + "start": "00:07:04", + "end": "00:07:09" + } + } + ], + "question_timestamps": "00:07:29", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three bottles of mineral water are on the table in the upstairs living room.", + "D": "Four bottles of mineral water are on the table in the second-floor living room.", + "E": "Zero." + }, + "answer_label": "E", + "answer_text": "Zero.", + "q_id": "egolife_A1_D2_024_2", + "correct_answer": "Zero.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three bottles of mineral water are on the table in the upstairs living room.", + "D": "Four bottles of mineral water are on the table in the second-floor living room.", + "E": "Zero." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_5_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wooden chairs are on the second floor?", + "answer": "Five.", + "question_chinese": "二楼有几个木椅？", + "answer_chinese": "五个", + "evidence": [ + { + "description": "There are five wooden chairs on the second floor", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "Only three wooden chairs are upstairs in that area on the second floor.", + "E": "There are seven wooden chairs on the second floor in total, not counting any stools." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D2_024_5_T1", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "Only three wooden chairs are upstairs in that area on the second floor.", + "E": "There are seven wooden chairs on the second floor in total, not counting any stools." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_5_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wooden chairs are on the second floor?", + "answer": "Five.", + "question_chinese": "二楼有几个木椅？", + "answer_chinese": "五个", + "evidence": [ + { + "description": "There are five wooden chairs on the second floor", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:49", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "Only three wooden chairs are upstairs in that area on the second floor.", + "E": "There are seven wooden chairs on the second floor in total, not counting any stools." + }, + "answer_label": "C", + "answer_text": "Five.", + "q_id": "egolife_A1_D2_024_5_T2", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "Only three wooden chairs are upstairs in that area on the second floor.", + "E": "There are seven wooden chairs on the second floor in total, not counting any stools." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_5_T3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wooden chairs are on the second floor?", + "answer": "Six.", + "question_chinese": "二楼有几个木椅？", + "answer_chinese": "六个", + "evidence": [ + { + "description": "There are six wooden chairs on the second floor", + "timestamps": { + "start": "00:06:48", + "end": "00:07:05" + } + }, + { + "description": "There are five wooden chairs on the second floor", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + }, + { + "description": "Someone brought up a chair", + "timestamps": { + "start": "00:02:28", + "end": "00:02:32" + } + } + ], + "question_timestamps": "00:07:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "There are only four wooden chairs upstairs on the second floor, not in any other rooms.", + "D": "There are eight wooden chairs on the second floor altogether, spread across different parts of the floor.", + "E": "Six." + }, + "answer_label": "E", + "answer_text": "Six.", + "q_id": "egolife_A1_D2_024_5_T3", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "There are only four wooden chairs upstairs on the second floor, not in any other rooms.", + "D": "There are eight wooden chairs on the second floor altogether, spread across different parts of the floor.", + "E": "Six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_5_T4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many wooden chairs are on the second floor?", + "answer": "Six.", + "question_chinese": "二楼有几个木椅？", + "answer_chinese": "六个", + "evidence": [ + { + "description": "There are six wooden chairs on the second floor", + "timestamps": { + "start": "00:06:48", + "end": "00:07:05" + } + }, + { + "description": "There are five wooden chairs on the second floor", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + }, + { + "description": "Someone brought up a chair", + "timestamps": { + "start": "00:02:28", + "end": "00:02:32" + } + } + ], + "question_timestamps": "00:07:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "There are only four wooden chairs upstairs on the second floor, not in any other rooms.", + "D": "Six.", + "E": "There are eight wooden chairs on the second floor altogether, spread across different parts of the floor." + }, + "answer_label": "D", + "answer_text": "Six.", + "q_id": "egolife_A1_D2_024_5_T4", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "There are only four wooden chairs upstairs on the second floor, not in any other rooms.", + "D": "Six.", + "E": "There are eight wooden chairs on the second floor altogether, spread across different parts of the floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_3_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction from me is the refrigerator right now?", + "answer": "The fridge was directly in front of you.", + "question_chinese": "冰箱目前在我的什么方向？", + "answer_chinese": "冰箱在你的正前方。", + "evidence": [ + { + "description": "The refrigerator is right in front of you.", + "timestamps": { + "start": "00:02:34", + "end": "00:02:36" + } + } + ], + "question_timestamps": "00:02:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge was behind you.", + "B": "The fridge was to your left.", + "C": "The fridge was directly in front of you.", + "D": "The fridge was directly to your right as you looked around the room.", + "E": "The fridge was far behind you, not straight ahead in front from your viewpoint." + }, + "answer_label": "C", + "answer_text": "The fridge was directly in front of you.", + "q_id": "egolife_A1_D2_024_3_T1", + "correct_answer": "The fridge was directly in front of you.", + "original_options": { + "A": "The fridge was behind you.", + "B": "The fridge was to your left.", + "C": "The fridge was directly in front of you.", + "D": "The fridge was directly to your right as you looked around the room.", + "E": "The fridge was far behind you, not straight ahead in front from your viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_3_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction from me is the refrigerator right now?", + "answer": "The refrigerator was to your front-left.", + "question_chinese": "冰箱目前在我的什么方向？", + "answer_chinese": "冰箱在你的左前方。", + "evidence": [ + { + "description": "The refrigerator is right in front of you.", + "timestamps": { + "start": "00:02:34", + "end": "00:02:36" + } + } + ], + "question_timestamps": "00:03:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator was front-right.", + "B": "The refrigerator was back-left.", + "C": "The refrigerator was directly in front of you, not angled to your left.", + "D": "The refrigerator sat behind you on the left side rather than in your forward-left view.", + "E": "The refrigerator was to your front-left." + }, + "answer_label": "E", + "answer_text": "The refrigerator was to your front-left.", + "q_id": "egolife_A1_D2_024_3_T2", + "correct_answer": "The refrigerator was to your front-left.", + "original_options": { + "A": "The refrigerator was front-right.", + "B": "The refrigerator was back-left.", + "C": "The refrigerator was directly in front of you, not angled to your left.", + "D": "The refrigerator sat behind you on the left side rather than in your forward-left view.", + "E": "The refrigerator was to your front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_2_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the second floor, which is closer to me, the refrigerator or the sofa?", + "answer": "The sofa—you were next to the table.", + "question_chinese": "在二楼，冰箱和沙发哪个离我更近一些？", + "answer_chinese": "沙发，你现在在桌子旁边。", + "evidence": [ + { + "description": "The refrigerator is on the side close to the windowsill, while the sofa is on the side far from the windowsill", + "timestamps": { + "start": "00:02:30", + "end": "00:02:44" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The refrigerator—you were next to the table.", + "B": "The sofa—you were next to the table." + }, + "answer_label": "B", + "answer_text": "The sofa—you were next to the table.", + "q_id": "egolife_A1_D2_024_2_T1", + "correct_answer": "The sofa—you were next to the table.", + "original_options": { + "A": "The refrigerator—you were next to the table.", + "B": "The sofa—you were next to the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_2_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the second floor, which is closer to me, the refrigerator or the sofa?", + "answer": "The sofa.", + "question_chinese": "在二楼，冰箱和沙发哪个离我更近一些？", + "answer_chinese": "沙发", + "evidence": [ + { + "description": "The refrigerator is on the side close to the windowsill, while the sofa is on the side far from the windowsill", + "timestamps": { + "start": "00:02:30", + "end": "00:02:44" + } + } + ], + "question_timestamps": "00:02:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The refrigerator.", + "B": "The sofa." + }, + "answer_label": "B", + "answer_text": "The sofa.", + "q_id": "egolife_A1_D2_024_2_T2", + "correct_answer": "The sofa.", + "original_options": { + "A": "The refrigerator.", + "B": "The sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_2_T3", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the second floor, which is closer to me, the refrigerator or the sofa?", + "answer": "The refrigerator—you were by the window.", + "question_chinese": "在二楼，冰箱和沙发哪个离我更近一些？", + "answer_chinese": "冰箱，你现在在窗户边。", + "evidence": [ + { + "description": "The refrigerator is on the side close to the windowsill, while the sofa is on the side far from the windowsill", + "timestamps": { + "start": "00:02:30", + "end": "00:02:44" + } + } + ], + "question_timestamps": "00:06:37", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sofa—you were by the window.", + "B": "The refrigerator—you were by the window." + }, + "answer_label": "B", + "answer_text": "The refrigerator—you were by the window.", + "q_id": "egolife_A1_D2_024_2_T3", + "correct_answer": "The refrigerator—you were by the window.", + "original_options": { + "A": "The sofa—you were by the window.", + "B": "The refrigerator—you were by the window." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_2_T4", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the second floor, which is closer to me, the refrigerator or the sofa?", + "answer": "The refrigerator.", + "question_chinese": "在二楼，冰箱和沙发哪个离我更近一些？", + "answer_chinese": "冰箱", + "evidence": [ + { + "description": "The refrigerator is on the side close to the windowsill, while the sofa is on the side far from the windowsill", + "timestamps": { + "start": "00:02:30", + "end": "00:02:44" + } + } + ], + "question_timestamps": "00:06:47", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sofa.", + "B": "The refrigerator." + }, + "answer_label": "B", + "answer_text": "The refrigerator.", + "q_id": "egolife_A1_D2_024_2_T4", + "correct_answer": "The refrigerator.", + "original_options": { + "A": "The sofa.", + "B": "The refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_7", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I just picked up my phone, which direction is the refrigerator from me?", + "answer": "Front-left.", + "question_chinese": "我刚刚拿起手机时，冰箱在我的什么方向？", + "answer_chinese": "左前方", + "evidence": [ + { + "description": "When I just picked up my phone, the refrigerator was on my left front", + "timestamps": { + "start": "00:02:15", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Front-right.", + "B": "Back-left.", + "C": "Directly to your right, slightly ahead of where you were standing right then.", + "D": "Front-left.", + "E": "Behind you and slightly to the right when you picked up your phone." + }, + "answer_label": "D", + "answer_text": "Front-left.", + "q_id": "egolife_A1_D2_024_7", + "correct_answer": "Front-left.", + "original_options": { + "A": "Front-right.", + "B": "Back-left.", + "C": "Directly to your right, slightly ahead of where you were standing right then.", + "D": "Front-left.", + "E": "Behind you and slightly to the right when you picked up your phone." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_8", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I just picked up my phone, which direction was the refrigerator from me?", + "answer": "Right fron.", + "question_chinese": "我刚刚拿起手机时，冰箱在我的什么方向？", + "answer_chinese": "右前方", + "evidence": [ + { + "description": "When I just picked up my phone, the refrigerator was right in front of me", + "timestamps": { + "start": "00:10:09", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:10:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Front-left.", + "B": "Back-right.", + "C": "Directly in front of you, not off to either side at the moment you grabbed your phone.", + "D": "Right fron.", + "E": "Behind you on the left side when you picked up your phone." + }, + "answer_label": "D", + "answer_text": "Right fron.", + "q_id": "egolife_A1_D2_024_8", + "correct_answer": "Right fron.", + "original_options": { + "A": "Front-left.", + "B": "Back-right.", + "C": "Directly in front of you, not off to either side at the moment you grabbed your phone.", + "D": "Right fron.", + "E": "Behind you on the left side when you picked up your phone." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_4_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people do I see on the second floor right now?", + "answer": "You can see one person upstairs right now: the guy in a white short-sleeve shirt to your left.", + "question_chinese": "我看到二楼现在有多少人？", + "answer_chinese": "你看到现在二楼有一个人。在你左边穿白色短袖的男生。", + "evidence": [ + { + "description": "one", + "timestamps": { + "start": "00:02:18", + "end": "00:02:21" + } + } + ], + "question_timestamps": "00:02:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero people.", + "B": "Two people.", + "C": "You can see two people upstairs right now: the guy in a white short-sleeve shirt and another person nearby.", + "D": "You can see three people upstairs right now: the guy in a white short-sleeve shirt plus two other people.", + "E": "You can see one person upstairs right now: the guy in a white short-sleeve shirt to your left." + }, + "answer_label": "E", + "answer_text": "You can see one person upstairs right now: the guy in a white short-sleeve shirt to your left.", + "q_id": "egolife_A1_D2_024_4_T2", + "correct_answer": "You can see one person upstairs right now: the guy in a white short-sleeve shirt to your left.", + "original_options": { + "A": "Zero people.", + "B": "Two people.", + "C": "You can see two people upstairs right now: the guy in a white short-sleeve shirt and another person nearby.", + "D": "You can see three people upstairs right now: the guy in a white short-sleeve shirt plus two other people.", + "E": "You can see one person upstairs right now: the guy in a white short-sleeve shirt to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_024_4_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people do I see on the second floor right now?", + "answer": "You can see two people upstairs right now: the guy in a white short-sleeve shirt on your left-hand side, and a girl with long hair.", + "question_chinese": "我看到二楼现在有多少人？", + "answer_chinese": "你看到现在二楼有两个人。在你左手边穿白色短袖的男生和长发的女生。", + "evidence": [ + { + "description": "one", + "timestamps": { + "start": "00:02:18", + "end": "00:02:21" + } + }, + { + "description": "two", + "timestamps": { + "start": "00:02:30", + "end": "00:02:35" + } + } + ], + "question_timestamps": "00:03:31", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero people.", + "B": "One person.", + "C": "You can see three people upstairs right now: the guy in a white short-sleeve shirt, the long-haired girl, and one more.", + "D": "You can see two people upstairs right now: the guy in a white short-sleeve shirt on your left-hand side, and a girl with long hair.", + "E": "You can see four people upstairs right now: the guy in a white short-sleeve shirt, the long-haired girl, and two more people." + }, + "answer_label": "D", + "answer_text": "You can see two people upstairs right now: the guy in a white short-sleeve shirt on your left-hand side, and a girl with long hair.", + "q_id": "egolife_A1_D2_024_4_T1", + "correct_answer": "You can see two people upstairs right now: the guy in a white short-sleeve shirt on your left-hand side, and a girl with long hair.", + "original_options": { + "A": "Zero people.", + "B": "One person.", + "C": "You can see three people upstairs right now: the guy in a white short-sleeve shirt, the long-haired girl, and one more.", + "D": "You can see two people upstairs right now: the guy in a white short-sleeve shirt on your left-hand side, and a girl with long hair.", + "E": "You can see four people upstairs right now: the guy in a white short-sleeve shirt, the long-haired girl, and two more people." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 21, + 22 + ] + } +] diff --git a/QAs/egolife_A1_D2_025-mcq.json b/QAs/egolife_A1_D2_025-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..cd656c3994b46451455828c998bac4b4fc7e5577 --- /dev/null +++ b/QAs/egolife_A1_D2_025-mcq.json @@ -0,0 +1,924 @@ +[ + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_9", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the second floor, which is closer to me, the sofa or the microwave?", + "answer": "Sofa.", + "question_chinese": "在二楼，沙发和微波炉哪个离我更近？", + "answer_chinese": "沙发", + "evidence": [ + { + "description": "The microwave oven is beside the windowsill", + "timestamps": { + "start": "00:03:34", + "end": "00:03:38" + } + }, + { + "description": "The sofa is on the side where the main door of the living room is", + "timestamps": { + "start": "00:04:10", + "end": "00:04:15" + } + } + ], + "question_timestamps": "00:06:02", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Microwave oven.", + "B": "Sofa." + }, + "answer_label": "B", + "answer_text": "Sofa.", + "q_id": "egolife_A1_D2_025_9", + "correct_answer": "Sofa.", + "original_options": { + "A": "Microwave oven.", + "B": "Sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_10", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the second floor, which is closer to me, the sofa or the microwave?", + "answer": "Microwave oven.", + "question_chinese": "在二楼，沙发和微波炉哪个离我更近？", + "answer_chinese": "微波炉", + "evidence": [ + { + "description": "The microwave oven is beside the windowsill", + "timestamps": { + "start": "00:03:34", + "end": "00:03:38" + } + }, + { + "description": "The sofa is on the side where the main door of the living room is", + "timestamps": { + "start": "00:04:10", + "end": "00:04:15" + } + } + ], + "question_timestamps": "00:06:30", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Microwave oven.", + "B": "Sofa." + }, + "answer_label": "A", + "answer_text": "Microwave oven.", + "q_id": "egolife_A1_D2_025_10", + "correct_answer": "Microwave oven.", + "original_options": { + "A": "Microwave oven.", + "B": "Sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_2_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to use the refrigerator right now, what is the fastest way to get there?", + "answer": "Stand up, then turn left.", + "question_chinese": "如果我现在想要使用冰箱，最快应该怎么走？", + "answer_chinese": "站起来，然后左转", + "evidence": [ + { + "description": "There is a refrigerator beside the staircase on the first floor", + "timestamps": { + "start": "00:03:05", + "end": "00:03:10" + } + }, + { + "description": "There is a refrigerator beside the staircase on the second floor", + "timestamps": { + "start": "00:03:40", + "end": "00:03:45" + } + } + ], + "question_timestamps": "00:07:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Stand up, then turn right.", + "B": "Stay seated and turn left.", + "C": "Stand up, then turn left.", + "D": "Stand up, walk forward to the doorway, then turn left toward the refrigerator.", + "E": "Stand up, turn right first, then loop around the table and turn left to reach the refrigerator." + }, + "answer_label": "C", + "answer_text": "Stand up, then turn left.", + "q_id": "egolife_A1_D2_025_2_T1", + "correct_answer": "Stand up, then turn left.", + "original_options": { + "A": "Stand up, then turn right.", + "B": "Stay seated and turn left.", + "C": "Stand up, then turn left.", + "D": "Stand up, walk forward to the doorway, then turn left toward the refrigerator.", + "E": "Stand up, turn right first, then loop around the table and turn left to reach the refrigerator." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_2_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to use the refrigerator right now, what is the quickest way to get there?", + "answer": "Stand up and then turn left.", + "question_chinese": "如果我现在想要使用冰箱，最快应该怎么走？", + "answer_chinese": "站起来，然后左转", + "evidence": [ + { + "description": "There is a refrigerator beside the staircase on the first floor", + "timestamps": { + "start": "00:03:05", + "end": "00:03:10" + } + }, + { + "description": "There is a refrigerator beside the staircase on the second floor", + "timestamps": { + "start": "00:03:40", + "end": "00:03:45" + } + } + ], + "question_timestamps": "00:07:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Stand up, then turn right.", + "B": "Stay seated and turn left.", + "C": "Stand up, walk straight a few steps, then turn left to reach the refrigerator.", + "D": "Stand up and then turn left.", + "E": "Stand up, turn right out of the bedroom door, then turn left after passing the table." + }, + "answer_label": "D", + "answer_text": "Stand up and then turn left.", + "q_id": "egolife_A1_D2_025_2_T2", + "correct_answer": "Stand up and then turn left.", + "original_options": { + "A": "Stand up, then turn right.", + "B": "Stay seated and turn left.", + "C": "Stand up, walk straight a few steps, then turn left to reach the refrigerator.", + "D": "Stand up and then turn left.", + "E": "Stand up, turn right out of the bedroom door, then turn left after passing the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_3_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to use the refrigerator right now, what is the fastest way to get there?", + "answer": "Stand up, turn right and go out the bedroom door, then walk straight for a bit to go around the table, then turn left and walk about five steps.", + "question_chinese": "如果我现在想要使用冰箱，最快应该怎么走？", + "answer_chinese": "站起来，右转出卧室门，然后直走一段距离绕过桌子，然后左转走五步左右", + "evidence": [ + { + "description": "There is a refrigerator beside the staircase on the first floor", + "timestamps": { + "start": "00:03:05", + "end": "00:03:10" + } + }, + { + "description": "There is a refrigerator beside the staircase on the second floor", + "timestamps": { + "start": "00:03:40", + "end": "00:03:45" + } + } + ], + "question_timestamps": "00:09:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left and walk five steps.", + "B": "Stand up, turn right, then stop.", + "C": "Stand up, turn right out the bedroom door, then turn right after the table and walk five steps.", + "D": "Stand up, turn right and go out the bedroom door, then walk straight for a bit to go around the table, then turn left and walk about five steps.", + "E": "Stand up, turn right and exit the bedroom, walk around the table, then turn left and walk about three steps." + }, + "answer_label": "D", + "answer_text": "Stand up, turn right and go out the bedroom door, then walk straight for a bit to go around the table, then turn left and walk about five steps.", + "q_id": "egolife_A1_D2_025_3_T1", + "correct_answer": "Stand up, turn right and go out the bedroom door, then walk straight for a bit to go around the table, then turn left and walk about five steps.", + "original_options": { + "A": "Turn left and walk five steps.", + "B": "Stand up, turn right, then stop.", + "C": "Stand up, turn right out the bedroom door, then turn right after the table and walk five steps.", + "D": "Stand up, turn right and go out the bedroom door, then walk straight for a bit to go around the table, then turn left and walk about five steps.", + "E": "Stand up, turn right and exit the bedroom, walk around the table, then turn left and walk about three steps." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_3_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to use the refrigerator now, what is the quickest way to get there?", + "answer": "Stand up, turn right out of the bedroom door, then walk straight for a while around the table, and then turn left and walk about five steps.", + "question_chinese": "如果我现在想要使用冰箱，最快应该怎么走？", + "answer_chinese": "站起来，右转出卧室门，然后直走一段距离绕过桌子，然后左转走五步左右", + "evidence": [ + { + "description": "There is a refrigerator beside the staircase on the first floor", + "timestamps": { + "start": "00:03:05", + "end": "00:03:10" + } + }, + { + "description": "There is a refrigerator beside the staircase on the second floor", + "timestamps": { + "start": "00:03:40", + "end": "00:03:45" + } + } + ], + "question_timestamps": "00:09:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Stand up, then turn left.", + "B": "Stand up, then turn right.", + "C": "Stand up, turn right out of the bedroom door, then walk straight for a while around the table, and then turn left and walk about five steps.", + "D": "Stand up, turn left out of the bedroom door, walk around the table, then turn left and walk five steps.", + "E": "Stand up, turn right out of the bedroom door, walk around the table, then turn right and walk about five steps." + }, + "answer_label": "C", + "answer_text": "Stand up, turn right out of the bedroom door, then walk straight for a while around the table, and then turn left and walk about five steps.", + "q_id": "egolife_A1_D2_025_3_T2", + "correct_answer": "Stand up, turn right out of the bedroom door, then walk straight for a while around the table, and then turn left and walk about five steps.", + "original_options": { + "A": "Stand up, then turn left.", + "B": "Stand up, then turn right.", + "C": "Stand up, turn right out of the bedroom door, then walk straight for a while around the table, and then turn left and walk about five steps.", + "D": "Stand up, turn left out of the bedroom door, walk around the table, then turn left and walk five steps.", + "E": "Stand up, turn right out of the bedroom door, walk around the table, then turn right and walk about five steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025-mcq_fixed_0", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there at home? Where are they respectively located?", + "answer": "Four.", + "question_chinese": "家里有几个垃圾桶？分别在什么位置？", + "answer_chinese": "两个，一个在一楼客厅桌子下，一个在一楼窗台旁", + "evidence": [ + { + "description": "There is a trash can under the table in the living room on the first floor", + "timestamps": { + "start": "00:03:04", + "end": "00:03:08" + } + }, + { + "description": "There is a trash can beside the windowsill on the first floor", + "timestamps": { + "start": "00:03:07", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One trash can total.", + "B": "Three trash cans total.", + "C": "Four.", + "D": "Two trash cans: one in the kitchen and one in the bathroom.", + "E": "Five trash cans: one in the kitchen, one in the bathroom, one by the bedroom, and two in the living room." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "egolife_A1_D2_025-mcq_fixed_0", + "correct_answer": "Four.", + "original_options": { + "A": "One trash can total.", + "B": "Three trash cans total.", + "C": "Four.", + "D": "Two trash cans: one in the kitchen and one in the bathroom.", + "E": "Five trash cans: one in the kitchen, one in the bathroom, one by the bedroom, and two in the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 21 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025-mcq_fixed_1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there at home? Where are they respectively located?", + "answer": "Five.", + "question_chinese": "家里有几个垃圾桶？分别在什么位置？", + "answer_chinese": "三个，一个在一楼客厅桌子下，一个在一楼窗台旁，还有一个在二楼窗台旁", + "evidence": [ + { + "description": "There is a trash can under the table in the living room on the first floor", + "timestamps": { + "start": "00:03:04", + "end": "00:03:08" + } + }, + { + "description": "There is a trash can beside the windowsill on the first floor", + "timestamps": { + "start": "00:03:07", + "end": "00:03:10" + } + }, + { + "description": "There is a trash can beside the windowsill on the second floor", + "timestamps": { + "start": "00:03:34", + "end": "00:03:39" + } + } + ], + "question_timestamps": "00:05:59", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "One trash can total.", + "C": "Four trash cans total.", + "D": "Two trash cans: one in the kitchen and one in the bathroom.", + "E": "Three trash cans: one in the kitchen, one in the bathroom, and one near the bedroom door." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D2_025-mcq_fixed_1", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "One trash can total.", + "C": "Four trash cans total.", + "D": "Two trash cans: one in the kitchen and one in the bathroom.", + "E": "Three trash cans: one in the kitchen, one in the bathroom, and one near the bedroom door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_1_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans do I have in my home?", + "answer": "Three, all behind you.", + "question_chinese": "我家里有几个垃圾桶？", + "answer_chinese": "三个，都在你的后方。", + "evidence": [ + { + "description": "There is a trash can under the table in the living room on the first floor", + "timestamps": { + "start": "00:03:04", + "end": "00:03:08" + } + }, + { + "description": "There is a trash can beside the windowsill on the first floor", + "timestamps": { + "start": "00:03:09", + "end": "00:03:10" + } + }, + { + "description": "There is a trash can beside the windowsill on the second floor", + "timestamps": { + "start": "00:03:34", + "end": "00:03:39" + } + } + ], + "question_timestamps": "00:03:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, all behind you.", + "B": "Three, all behind you.", + "C": "Four, all behind you.", + "D": "One trash can directly behind you near the wall, with the others not visible from here.", + "E": "Five trash cans lined up behind you along the back wall in a single row." + }, + "answer_label": "B", + "answer_text": "Three, all behind you.", + "q_id": "egolife_A1_D2_025_1_T1", + "correct_answer": "Three, all behind you.", + "original_options": { + "A": "Two, all behind you.", + "B": "Three, all behind you.", + "C": "Four, all behind you.", + "D": "One trash can directly behind you near the wall, with the others not visible from here.", + "E": "Five trash cans lined up behind you along the back wall in a single row." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_1_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans do I have in my home?", + "answer": "Three, all to your back-left.", + "question_chinese": "我家里有几个垃圾桶？", + "answer_chinese": "三个，都在你的左后方。", + "evidence": [ + { + "description": "There is a trash can under the table in the living room on the first floor", + "timestamps": { + "start": "00:03:04", + "end": "00:03:08" + } + }, + { + "description": "There is a trash can beside the windowsill on the first floor", + "timestamps": { + "start": "00:03:09", + "end": "00:03:10" + } + }, + { + "description": "There is a trash can beside the windowsill on the second floor", + "timestamps": { + "start": "00:03:34", + "end": "00:03:39" + } + } + ], + "question_timestamps": "00:03:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, all to your back-left.", + "B": "Four, all to your back-left.", + "C": "Three, all to your back-left.", + "D": "One trash can sits to your back-left in the corner, with no others nearby.", + "E": "Five trash cans are clustered together to your back-left near the wall in one group." + }, + "answer_label": "C", + "answer_text": "Three, all to your back-left.", + "q_id": "egolife_A1_D2_025_1_T2", + "correct_answer": "Three, all to your back-left.", + "original_options": { + "A": "Two, all to your back-left.", + "B": "Four, all to your back-left.", + "C": "Three, all to your back-left.", + "D": "One trash can sits to your back-left in the corner, with no others nearby.", + "E": "Five trash cans are clustered together to your back-left near the wall in one group." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_5", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I last entered the living room on the first floor, was the windowsill on my left or my right?", + "answer": "Left.", + "question_chinese": "我最近一次进入一楼客厅时，窗台在我的左边还是右边？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "The last time I entered the living room on the first floor, the windowsill was on my left", + "timestamps": { + "start": "00:03:05", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:04:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Right.", + "B": "Left." + }, + "answer_label": "B", + "answer_text": "Left.", + "q_id": "egolife_A1_D2_025_5", + "correct_answer": "Left.", + "original_options": { + "A": "Right.", + "B": "Left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_6", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I last entered the living room on the first floor, was the windowsill on my left or my right?", + "answer": "Right.", + "question_chinese": "我最近一次进入一楼客厅时，窗台在我的左边还是右边？", + "answer_chinese": "右边", + "evidence": [ + { + "description": "The last time I entered the living room, the windowsill was on my right", + "timestamps": { + "start": "00:09:30", + "end": "00:09:36" + } + } + ], + "question_timestamps": "00:10:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Right.", + "B": "Left." + }, + "answer_label": "A", + "answer_text": "Right.", + "q_id": "egolife_A1_D2_025_6", + "correct_answer": "Right.", + "original_options": { + "A": "Right.", + "B": "Left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_7", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the trash can now?", + "answer": "Yes.", + "question_chinese": "我现在能拿到垃圾桶吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The trash can is beside the cushion", + "timestamps": { + "start": "00:03:33", + "end": "00:03:49" + } + }, + { + "description": "I sat on the cushion", + "timestamps": { + "start": "00:06:16", + "end": "00:06:23" + } + } + ], + "question_timestamps": "00:07:26", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D2_025_7", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_8", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the trash can now?", + "answer": "No.", + "question_chinese": "我现在能拿到垃圾桶吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The trash can is beside the cushion", + "timestamps": { + "start": "00:03:33", + "end": "00:03:49" + } + } + ], + "question_timestamps": "00:09:31", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D2_025_8", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_11", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the doll now?", + "answer": "No.", + "question_chinese": "我现在能拿到玩偶吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The dolls are all on the sofa", + "timestamps": { + "start": "00:04:07", + "end": "00:04:17" + } + } + ], + "question_timestamps": "00:06:21", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D2_025_11", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_025_12", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the doll now?", + "answer": "Yes.", + "question_chinese": "我现在能拿到玩偶吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The dolls are all on the sofa", + "timestamps": { + "start": "00:04:07", + "end": "00:04:17" + } + } + ], + "question_timestamps": "00:11:25", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D2_025_12", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D2_026-mcq.json b/QAs/egolife_A1_D2_026-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e8480b82f98062d6e1318e1a6263f6d7429ba04f --- /dev/null +++ b/QAs/egolife_A1_D2_026-mcq.json @@ -0,0 +1,644 @@ +[ + { + "video_uid": "egolife_A1_D2_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_026_1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the window next to the second floor now, how should I get there?", + "answer": "You should walk four steps backward to your left, and you’ll reach the side of the window.", + "question_chinese": "如果现在我想去2楼的窗户旁边，那我应该怎么走？", + "answer_chinese": "你应该向左后方走4步的距离就能到达窗户旁边了。", + "evidence": [ + { + "description": "The window on the second floor is on my left.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + }, + { + "description": "The door leading to the stairs is about four steps away from the window on the second floor.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:05:46", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk four steps forward.", + "B": "Walk three steps backward left.", + "C": "Walk four steps backward to your right, and you’ll reach the side of the window.", + "D": "Walk five steps backward to your left, and you’ll reach the side of the window.", + "E": "You should walk four steps backward to your left, and you’ll reach the side of the window." + }, + "answer_label": "E", + "answer_text": "You should walk four steps backward to your left, and you’ll reach the side of the window.", + "q_id": "egolife_A1_D2_026_1", + "correct_answer": "You should walk four steps backward to your left, and you’ll reach the side of the window.", + "original_options": { + "A": "Walk four steps forward.", + "B": "Walk three steps backward left.", + "C": "Walk four steps backward to your right, and you’ll reach the side of the window.", + "D": "Walk five steps backward to your left, and you’ll reach the side of the window.", + "E": "You should walk four steps backward to your left, and you’ll reach the side of the window." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_026_2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the window next to the second floor now, how should I get there?", + "answer": "You should walk about seven steps backward to your right, enter the stairwell, go up to the second floor, then walk forward to your right for about four steps, and you’ll reach the side of the window on the second floor.", + "question_chinese": "如果现在我想去2楼的窗户旁边，那我应该怎么走？", + "answer_chinese": "你应当向右后方走7步左右的距离。进入楼梯间，然后上到二楼，随后向右前方走，4步左右的距离就能到达二楼的窗户旁边了。", + "evidence": [ + { + "description": "The window on the second floor is on my left.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + }, + { + "description": "The door leading to the stairs is about four steps away from the window on the second floor.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "I went down from the second floor to the first floor.", + "timestamps": { + "start": "00:06:24", + "end": "00:06:38" + } + }, + { + "description": "The door leading to the second floor is about seven steps away from the bedroom.", + "timestamps": { + "start": "00:06:38", + "end": "00:06:43" + } + } + ], + "question_timestamps": "00:09:55", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Go up, then turn left.", + "B": "Enter stairwell after five steps.", + "C": "You should walk about seven steps backward to your right, enter the stairwell, go up to the second floor, then walk forward to your right for about four steps, and you’ll reach the side of the window on the second floor.", + "D": "Walk about seven steps backward to your left, go upstairs, then walk forward right four steps to the window.", + "E": "Walk about seven steps backward to your right, enter the stairwell, go up, then walk forward left for four steps." + }, + "answer_label": "C", + "answer_text": "You should walk about seven steps backward to your right, enter the stairwell, go up to the second floor, then walk forward to your right for about four steps, and you’ll reach the side of the window on the second floor.", + "q_id": "egolife_A1_D2_026_2", + "correct_answer": "You should walk about seven steps backward to your right, enter the stairwell, go up to the second floor, then walk forward to your right for about four steps, and you’ll reach the side of the window on the second floor.", + "original_options": { + "A": "Go up, then turn left.", + "B": "Enter stairwell after five steps.", + "C": "You should walk about seven steps backward to your right, enter the stairwell, go up to the second floor, then walk forward to your right for about four steps, and you’ll reach the side of the window on the second floor.", + "D": "Walk about seven steps backward to your left, go upstairs, then walk forward right four steps to the window.", + "E": "Walk about seven steps backward to your right, enter the stairwell, go up, then walk forward left for four steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_026_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cameras have I seen so far?", + "answer": "I have seen four cameras in total.", + "question_chinese": "我到目前为止总共见过几个摄像头？", + "answer_chinese": "总共见过4个摄像头", + "evidence": [ + { + "description": "The first camera is to my right front.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "There is a camera each to the left front and right front.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:31" + } + }, + { + "description": "The fourth camera is directly in front of me.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:49" + } + } + ], + "question_timestamps": "00:02:55", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I have seen three cameras.", + "B": "I have seen five cameras.", + "C": "I have seen two cameras in total so far, not counting any others you passed earlier.", + "D": "I have seen four cameras in total.", + "E": "I have seen six cameras in total, including the ones you noticed near the end." + }, + "answer_label": "D", + "answer_text": "I have seen four cameras in total.", + "q_id": "egolife_A1_D2_026_5", + "correct_answer": "I have seen four cameras in total.", + "original_options": { + "A": "I have seen three cameras.", + "B": "I have seen five cameras.", + "C": "I have seen two cameras in total so far, not counting any others you passed earlier.", + "D": "I have seen four cameras in total.", + "E": "I have seen six cameras in total, including the ones you noticed near the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_026_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cameras have I seen so far?", + "answer": "The fifth camera is to my right front.", + "question_chinese": "我到目前为止总共见过几个摄像头？", + "answer_chinese": "我的右前方是第五个摄像头。", + "evidence": [ + { + "description": "The first camera is to my right front.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "There is a camera each to the left front and right front.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:31" + } + }, + { + "description": "The fourth camera is directly in front of me.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:49" + } + } + ], + "question_timestamps": "00:04:46", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four cameras so far.", + "B": "The fifth camera is to my right front.", + "C": "Six cameras so far.", + "D": "I have seen three cameras so far, and the next one should be somewhere ahead of you.", + "E": "I have seen seven cameras so far, so the fifth would no longer be the latest one you noticed." + }, + "answer_label": "B", + "answer_text": "The fifth camera is to my right front.", + "q_id": "egolife_A1_D2_026_6", + "correct_answer": "The fifth camera is to my right front.", + "original_options": { + "A": "Four cameras so far.", + "B": "The fifth camera is to my right front.", + "C": "Six cameras so far.", + "D": "I have seen three cameras so far, and the next one should be somewhere ahead of you.", + "E": "I have seen seven cameras so far, so the fifth would no longer be the latest one you noticed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_026_3", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the projection screen from me?", + "answer": "The projection screen is on your left.", + "question_chinese": "投影屏幕在我的什么方向？", + "answer_chinese": "投影屏幕在你的左侧。", + "evidence": [ + { + "description": "The projection screen is to my right front.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:05:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "The projection screen is on your left.", + "C": "It is on your right.", + "D": "The projection screen is behind you, so you would need to turn around to face it.", + "E": "The projection screen is above you and slightly forward, not positioned to your left." + }, + "answer_label": "B", + "answer_text": "The projection screen is on your left.", + "q_id": "egolife_A1_D2_026_3", + "correct_answer": "The projection screen is on your left.", + "original_options": { + "A": "It is in front of you.", + "B": "The projection screen is on your left.", + "C": "It is on your right.", + "D": "The projection screen is behind you, so you would need to turn around to face it.", + "E": "The projection screen is above you and slightly forward, not positioned to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_026_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the projection screen from me?", + "answer": "The projection screen is on your upper left.", + "question_chinese": "投影屏幕在我的什么方向？", + "answer_chinese": "投影屏幕在你的左上方。", + "evidence": [ + { + "description": "The projection screen is to my right front.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:10:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your upper right.", + "B": "It is directly in front of you.", + "C": "The projection screen is on your lower left, closer to the ground than your eye level.", + "D": "The projection screen is on your upper left.", + "E": "The projection screen is behind you to the left, not up in your upper-left field of view." + }, + "answer_label": "D", + "answer_text": "The projection screen is on your upper left.", + "q_id": "egolife_A1_D2_026_4", + "correct_answer": "The projection screen is on your upper left.", + "original_options": { + "A": "It is on your upper right.", + "B": "It is directly in front of you.", + "C": "The projection screen is on your lower left, closer to the ground than your eye level.", + "D": "The projection screen is on your upper left.", + "E": "The projection screen is behind you to the left, not up in your upper-left field of view." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_026_7", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the window on the second floor closer to me, or is the door to the stairwell on the second floor closer to me?", + "answer": "He door to the second-floor stairwell is closer to you.", + "question_chinese": "是二楼的窗户离我更近，还是二楼楼梯间的门离我更近？", + "answer_chinese": "2楼楼梯间的门离你更近。", + "evidence": [ + { + "description": "The window on the second floor is on my left side.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The door to the stairwell is forward and to my right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:06:02", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "He door to the second-floor stairwell is closer to you.", + "B": "The window on the second floor is closer to you." + }, + "answer_label": "A", + "answer_text": "He door to the second-floor stairwell is closer to you.", + "q_id": "egolife_A1_D2_026_7", + "correct_answer": "He door to the second-floor stairwell is closer to you.", + "original_options": { + "A": "He door to the second-floor stairwell is closer to you.", + "B": "The window on the second floor is closer to you." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D2_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_026_8", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the window on the second floor closer to me, or is the door to the stairwell on the second floor closer to me?", + "answer": "If vertical distance is taken into account, the window on the second floor is closer to you.", + "question_chinese": "是二楼的窗户离我更近，还是二楼楼梯间的门离我更近？", + "answer_chinese": "算上垂直距离的话，二楼的窗户离你更近。", + "evidence": [ + { + "description": "The window on the second floor is on my left side.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The door to the stairwell is forward and to my right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:11:07", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "If vertical distance is taken into account, the door to the second-floor stairwell is closer to you.", + "B": "If vertical distance is taken into account, the window on the second floor is closer to you." + }, + "answer_label": "B", + "answer_text": "If vertical distance is taken into account, the window on the second floor is closer to you.", + "q_id": "egolife_A1_D2_026_8", + "correct_answer": "If vertical distance is taken into account, the window on the second floor is closer to you.", + "original_options": { + "A": "If vertical distance is taken into account, the door to the second-floor stairwell is closer to you.", + "B": "If vertical distance is taken into account, the window on the second floor is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_026_9", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the red refrigerator from me?", + "answer": "The red refrigerator is on your left.", + "question_chinese": "红色的冰箱在我的什么方向？", + "answer_chinese": "红色的冰箱在你的左侧", + "evidence": [ + { + "description": "There is a red refrigerator in front of me to the right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:06:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is on your right.", + "C": "The red refrigerator is behind you, so you would need to turn around to see it.", + "D": "The red refrigerator is above you and slightly forward, not located to your left side.", + "E": "The red refrigerator is on your left." + }, + "answer_label": "E", + "answer_text": "The red refrigerator is on your left.", + "q_id": "egolife_A1_D2_026_9", + "correct_answer": "The red refrigerator is on your left.", + "original_options": { + "A": "It is in front of you.", + "B": "It is on your right.", + "C": "The red refrigerator is behind you, so you would need to turn around to see it.", + "D": "The red refrigerator is above you and slightly forward, not located to your left side.", + "E": "The red refrigerator is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_026_10", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the red refrigerator from me?", + "answer": "The red refrigerator is on your upper right.", + "question_chinese": "红色的冰箱在我的什么方向？", + "answer_chinese": "红色的冰箱在你的右上方。", + "evidence": [ + { + "description": "There is a red refrigerator in front of me to the right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:11:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on your upper left.", + "B": "It is behind you now.", + "C": "The red refrigerator is on your upper right.", + "D": "The red refrigerator is directly in front of you at about eye level, not up to your right.", + "E": "The red refrigerator is on your lower right, below your line of sight rather than on your upper right." + }, + "answer_label": "C", + "answer_text": "The red refrigerator is on your upper right.", + "q_id": "egolife_A1_D2_026_10", + "correct_answer": "The red refrigerator is on your upper right.", + "original_options": { + "A": "It is on your upper left.", + "B": "It is behind you now.", + "C": "The red refrigerator is on your upper right.", + "D": "The red refrigerator is directly in front of you at about eye level, not up to your right.", + "E": "The red refrigerator is on your lower right, below your line of sight rather than on your upper right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 19 + ] + } +] diff --git a/QAs/egolife_A1_D2_027-mcq.json b/QAs/egolife_A1_D2_027-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..097d57ce7cf40dade0d889baa9097d1557509e6a --- /dev/null +++ b/QAs/egolife_A1_D2_027-mcq.json @@ -0,0 +1,230 @@ +[ + { + "video_uid": "egolife_A1_D2_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_027_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many plates were in the kitchen sink just now?", + "answer": "One.", + "question_chinese": "刚才厨房的水池中有几个盘子", + "answer_chinese": "一个", + "evidence": [ + { + "description": "When I stand in front of the pool, there is a plate.", + "timestamps": { + "start": "00:10:16", + "end": "00:10:18" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "There were two plates sitting in the kitchen sink just now, not one.", + "D": "You had three plates in the kitchen sink a moment ago, not just one.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_D2_027_7", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "There were two plates sitting in the kitchen sink just now, not one.", + "D": "You had three plates in the kitchen sink a moment ago, not just one.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_027_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many plates were in the kitchen sink just now?", + "answer": "One.", + "question_chinese": "刚才厨房的水池中有几个盘子", + "answer_chinese": "一个", + "evidence": [ + { + "description": "When I stand in front of the pool, there is a plate.", + "timestamps": { + "start": "00:10:16", + "end": "00:10:18" + } + } + ], + "question_timestamps": "00:11:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "Two.", + "C": "There were four plates in the kitchen sink just a moment ago.", + "D": "One.", + "E": "You saw two plates sitting in the sink just now, not one." + }, + "answer_label": "D", + "answer_text": "One.", + "q_id": "egolife_A1_D2_027_8", + "correct_answer": "One.", + "original_options": { + "A": "Zero.", + "B": "Two.", + "C": "There were four plates in the kitchen sink just a moment ago.", + "D": "One.", + "E": "You saw two plates sitting in the sink just now, not one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_027_1_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now, how many plates did I see in the kitchen sink?", + "answer": "You just saw 3 plates in the sink. The sink was to your front-right.", + "question_chinese": "我刚才看见厨房的水池中有几个盘子？", + "answer_chinese": "你刚才看见水池中有3个盘子。水池在你右前方", + "evidence": [ + { + "description": "When I stand in front of the pool, there is three plate.", + "timestamps": { + "start": "00:10:25", + "end": "00:10:30" + } + } + ], + "question_timestamps": "00:10:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just saw 2 plates in the sink.", + "B": "You just saw 4 plates in the sink.", + "C": "You just saw 2 plates in the sink, and the sink was to your front-right.", + "D": "You just saw 3 plates in the sink. The sink was to your front-right.", + "E": "You just saw 5 plates in the sink, with the sink located to your front-right." + }, + "answer_label": "D", + "answer_text": "You just saw 3 plates in the sink. The sink was to your front-right.", + "q_id": "egolife_A1_D2_027_1_T1", + "correct_answer": "You just saw 3 plates in the sink. The sink was to your front-right.", + "original_options": { + "A": "You just saw 2 plates in the sink.", + "B": "You just saw 4 plates in the sink.", + "C": "You just saw 2 plates in the sink, and the sink was to your front-right.", + "D": "You just saw 3 plates in the sink. The sink was to your front-right.", + "E": "You just saw 5 plates in the sink, with the sink located to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_027_1_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now, how many plates did I see in the kitchen sink?", + "answer": "You just saw one plate in the sink.", + "question_chinese": "我刚才看见厨房的水池中有几个盘子？", + "answer_chinese": "你刚才看见水池中有一个盘子。", + "evidence": [ + { + "description": "When I stand in front of the pool, there is a plate.", + "timestamps": { + "start": "00:10:16", + "end": "00:10:18" + } + } + ], + "question_timestamps": "00:11:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just saw zero plates in the sink.", + "B": "You just saw two plates in the sink.", + "C": "You just saw three plates in the sink when you looked into it just now.", + "D": "You just saw four plates in the sink a moment ago, not a single plate.", + "E": "You just saw one plate in the sink." + }, + "answer_label": "E", + "answer_text": "You just saw one plate in the sink.", + "q_id": "egolife_A1_D2_027_1_T2", + "correct_answer": "You just saw one plate in the sink.", + "original_options": { + "A": "You just saw zero plates in the sink.", + "B": "You just saw two plates in the sink.", + "C": "You just saw three plates in the sink when you looked into it just now.", + "D": "You just saw four plates in the sink a moment ago, not a single plate.", + "E": "You just saw one plate in the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D2_028-mcq.json b/QAs/egolife_A1_D2_028-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..13ba3138fb4ce0f6b0735c491be8d2e7b9ffe8f8 --- /dev/null +++ b/QAs/egolife_A1_D2_028-mcq.json @@ -0,0 +1,641 @@ +[ + { + "video_uid": "egolife_A1_D2_028", + "video_length": "00:11:34", + "qid": "egolife_A1_D2_028_5", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are on the second floor?", + "answer": "There are eight people on the second floor.", + "question_chinese": "二楼有几个人？", + "answer_chinese": "二楼有8个人", + "evidence": [ + { + "description": "There are seven people in my field of vision", + "timestamps": { + "start": "00:04:02", + "end": "00:04:03" + } + }, + { + "description": "I'm here too.", + "timestamps": { + "start": "00:04:03", + "end": "00:04:04" + } + } + ], + "question_timestamps": "00:05:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Seven people.", + "B": "Nine people.", + "C": "There are eight people on the second floor.", + "D": "There are seven people on the second floor right now, not eight in total.", + "E": "There are ten people on the second floor at the moment you checked." + }, + "answer_label": "C", + "answer_text": "There are eight people on the second floor.", + "q_id": "egolife_A1_D2_028_5", + "correct_answer": "There are eight people on the second floor.", + "original_options": { + "A": "Seven people.", + "B": "Nine people.", + "C": "There are eight people on the second floor.", + "D": "There are seven people on the second floor right now, not eight in total.", + "E": "There are ten people on the second floor at the moment you checked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_028", + "video_length": "00:11:34", + "qid": "egolife_A1_D2_028_6", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are on the second floor?", + "answer": "There are four people on the second floor.", + "question_chinese": "二楼有几个人？", + "answer_chinese": "二楼有4个人", + "evidence": [ + { + "description": "There are three people in my field of vision", + "timestamps": { + "start": "00:07:22", + "end": "00:07:23" + } + }, + { + "description": "I'm here too.", + "timestamps": { + "start": "00:07:23", + "end": "00:07:24" + } + } + ], + "question_timestamps": "00:09:40", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three people.", + "B": "Five people.", + "C": "There are two people on the second floor right now, not four in total.", + "D": "There are four people on the second floor.", + "E": "There are six people on the second floor based on what you just counted." + }, + "answer_label": "D", + "answer_text": "There are four people on the second floor.", + "q_id": "egolife_A1_D2_028_6", + "correct_answer": "There are four people on the second floor.", + "original_options": { + "A": "Three people.", + "B": "Five people.", + "C": "There are two people on the second floor right now, not four in total.", + "D": "There are four people on the second floor.", + "E": "There are six people on the second floor based on what you just counted." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_028", + "video_length": "00:11:34", + "qid": "egolife_A1_D2_028_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding in my hand?", + "answer": "The plate of raw meat I was holding just now.", + "question_chinese": "我刚才手里拿的什么？", + "answer_chinese": "我刚才手里拿的一盘生肉", + "evidence": [ + { + "description": "The plate of raw meat I'm holding in my hand now", + "timestamps": { + "start": "00:00:11", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:01:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A bowl of cooked vegetables.", + "B": "A bowl of fresh fruit.", + "C": "You were holding a basket of fresh herbs in your hand just now.", + "D": "The item in your hand was a bowl of mixed nuts you were carrying a moment ago.", + "E": "The plate of raw meat I was holding just now." + }, + "answer_label": "E", + "answer_text": "The plate of raw meat I was holding just now.", + "q_id": "egolife_A1_D2_028_1", + "correct_answer": "The plate of raw meat I was holding just now.", + "original_options": { + "A": "A bowl of cooked vegetables.", + "B": "A bowl of fresh fruit.", + "C": "You were holding a basket of fresh herbs in your hand just now.", + "D": "The item in your hand was a bowl of mixed nuts you were carrying a moment ago.", + "E": "The plate of raw meat I was holding just now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_028", + "video_length": "00:11:34", + "qid": "egolife_A1_D2_028_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding in my hand?", + "answer": "The wine bottle I was holding just now.", + "question_chinese": "我刚才手里拿的什么？", + "answer_chinese": "我刚才手里拿的酒瓶", + "evidence": [ + { + "description": "The wine bottle in my hand now", + "timestamps": { + "start": "00:03:25", + "end": "00:03:30" + } + } + ], + "question_timestamps": "00:04:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The wine bottle I was holding just now.", + "B": "The red book.", + "C": "The green notebook.", + "D": "You were holding the blue tablet in your hand just a moment ago.", + "E": "The item you had in your hand was the yellow folder you were carrying just now." + }, + "answer_label": "A", + "answer_text": "The wine bottle I was holding just now.", + "q_id": "egolife_A1_D2_028_2", + "correct_answer": "The wine bottle I was holding just now.", + "original_options": { + "A": "The wine bottle I was holding just now.", + "B": "The red book.", + "C": "The green notebook.", + "D": "You were holding the blue tablet in your hand just a moment ago.", + "E": "The item you had in your hand was the yellow folder you were carrying just now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_028", + "video_length": "00:11:34", + "qid": "egolife_A1_D2_028_3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I just walked around the table half a circle clockwise.", + "question_chinese": "我刚才走的什么路线？", + "answer_chinese": "我刚才顺时针绕桌子半圈", + "evidence": [ + { + "description": "I'm walking around the table half a circle clockwise", + "timestamps": { + "start": "00:01:09", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I walked half a circle counterclockwise.", + "B": "I walked a full circle clockwise.", + "C": "I just walked around the table three-quarters of a circle clockwise instead of stopping halfway.", + "D": "I just walked around the table half a circle clockwise.", + "E": "I just walked around the table half a circle counterclockwise, circling the opposite direction from before." + }, + "answer_label": "D", + "answer_text": "I just walked around the table half a circle clockwise.", + "q_id": "egolife_A1_D2_028_3", + "correct_answer": "I just walked around the table half a circle clockwise.", + "original_options": { + "A": "I walked half a circle counterclockwise.", + "B": "I walked a full circle clockwise.", + "C": "I just walked around the table three-quarters of a circle clockwise instead of stopping halfway.", + "D": "I just walked around the table half a circle clockwise.", + "E": "I just walked around the table half a circle counterclockwise, circling the opposite direction from before." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_028", + "video_length": "00:11:34", + "qid": "egolife_A1_D2_028_4", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I just went from the second floor to the first floor, from the living room to the kitchen.", + "question_chinese": "我刚才走的什么路线？", + "answer_chinese": "我刚才从二楼到一楼客厅到厨房", + "evidence": [ + { + "description": "I'm going from the second floor to the first floor, from the living room to the kitchen", + "timestamps": { + "start": "00:02:54", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:03:37", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From first floor to second.", + "B": "I just went from the second floor to the first floor, from the living room to the kitchen.", + "C": "From kitchen to living room.", + "D": "I went from the second floor to the first floor, ending in the living room instead of the kitchen.", + "E": "I went from the first floor to the second floor, heading from the living room toward the kitchen upstairs." + }, + "answer_label": "B", + "answer_text": "I just went from the second floor to the first floor, from the living room to the kitchen.", + "q_id": "egolife_A1_D2_028_4", + "correct_answer": "I just went from the second floor to the first floor, from the living room to the kitchen.", + "original_options": { + "A": "From first floor to second.", + "B": "I just went from the second floor to the first floor, from the living room to the kitchen.", + "C": "From kitchen to living room.", + "D": "I went from the second floor to the first floor, ending in the living room instead of the kitchen.", + "E": "I went from the first floor to the second floor, heading from the living room toward the kitchen upstairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_028", + "video_length": "00:11:34", + "qid": "egolife_A1_D2_028_9", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I encounter the red refrigerator without going through the door?", + "answer": "I can touch the red refrigerator without going through the door.", + "question_chinese": "我能不经过门就碰到红色冰箱吗？", + "answer_chinese": "我能不经过门就碰到红色冰箱", + "evidence": [ + { + "description": "The red refrigerator is in the living room on the first floor", + "timestamps": { + "start": "00:03:10", + "end": "00:03:11" + } + }, + { + "description": "I'm here too.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:08" + } + } + ], + "question_timestamps": "00:06:23", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I can't touch the red refrigerator without going through the door.", + "B": "I can touch the red refrigerator without going through the door." + }, + "answer_label": "B", + "answer_text": "I can touch the red refrigerator without going through the door.", + "q_id": "egolife_A1_D2_028_9", + "correct_answer": "I can touch the red refrigerator without going through the door.", + "original_options": { + "A": "I can't touch the red refrigerator without going through the door.", + "B": "I can touch the red refrigerator without going through the door." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D2_028", + "video_length": "00:11:34", + "qid": "egolife_A1_D2_028_10", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I encounter the red refrigerator without going through the door?", + "answer": "I couldn't help bumping into the red refrigerator without passing through the door.", + "question_chinese": "我能不经过门就碰到红色冰箱吗？", + "answer_chinese": "我不能不经过门就碰到红色冰箱", + "evidence": [ + { + "description": "The red refrigerator is in the living room on the first floor", + "timestamps": { + "start": "00:03:10", + "end": "00:03:11" + } + }, + { + "description": "I'm also on the second floor.", + "timestamps": { + "start": "00:06:38", + "end": "00:06:44" + } + } + ], + "question_timestamps": "00:06:49", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I couldn't encounter the red refrigerator without going through the door.", + "B": "I couldn't help bumping into the red refrigerator without passing through the door." + }, + "answer_label": "B", + "answer_text": "I couldn't help bumping into the red refrigerator without passing through the door.", + "q_id": "egolife_A1_D2_028_10", + "correct_answer": "I couldn't help bumping into the red refrigerator without passing through the door.", + "original_options": { + "A": "I couldn't encounter the red refrigerator without going through the door.", + "B": "I couldn't help bumping into the red refrigerator without passing through the door." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D2_028", + "video_length": "00:11:34", + "qid": "egolife_A1_D2_028_7", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I passed by the chair opposite the window?", + "answer": "So far, I have passed by the chair opposite the window three times.", + "question_chinese": "目前我经过窗户对的椅子几次？", + "answer_chinese": "目前我经过窗户对的椅子3次", + "evidence": [ + { + "description": "I'm passing by the chair opposite the window", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + }, + { + "description": "I'm passing by the chair opposite the window", + "timestamps": { + "start": "00:00:44", + "end": "00:00:46" + } + }, + { + "description": "I'm passing by the chair opposite the window", + "timestamps": { + "start": "00:00:56", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I passed it twice.", + "B": "So far, I have passed by the chair opposite the window three times.", + "C": "I passed it four times.", + "D": "So far, I have passed by the chair opposite the window five times during my movement.", + "E": "So far, I have passed by the chair opposite the window only once, and never again afterward." + }, + "answer_label": "B", + "answer_text": "So far, I have passed by the chair opposite the window three times.", + "q_id": "egolife_A1_D2_028_7", + "correct_answer": "So far, I have passed by the chair opposite the window three times.", + "original_options": { + "A": "I passed it twice.", + "B": "So far, I have passed by the chair opposite the window three times.", + "C": "I passed it four times.", + "D": "So far, I have passed by the chair opposite the window five times during my movement.", + "E": "So far, I have passed by the chair opposite the window only once, and never again afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_028", + "video_length": "00:11:34", + "qid": "egolife_A1_D2_028_8", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I passed by the chair opposite the window?", + "answer": "So far, I have passed by the chair opposite the window four times.", + "question_chinese": "目前我经过窗户对的椅子几次？", + "answer_chinese": "目前我经过窗户对的椅子4次", + "evidence": [ + { + "description": "I'm passing by the chair opposite the window", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + }, + { + "description": "I'm passing by the chair opposite the window", + "timestamps": { + "start": "00:00:44", + "end": "00:00:46" + } + }, + { + "description": "I'm passing by the chair opposite the window", + "timestamps": { + "start": "00:00:56", + "end": "00:00:59" + } + }, + { + "description": "I'm passing by the chair opposite the window", + "timestamps": { + "start": "00:01:13", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I passed it three times.", + "B": "I passed it five times.", + "C": "So far, I have passed by the chair opposite the window two times during my route.", + "D": "So far, I have passed by the chair opposite the window six times as I went back and forth.", + "E": "So far, I have passed by the chair opposite the window four times." + }, + "answer_label": "E", + "answer_text": "So far, I have passed by the chair opposite the window four times.", + "q_id": "egolife_A1_D2_028_8", + "correct_answer": "So far, I have passed by the chair opposite the window four times.", + "original_options": { + "A": "I passed it three times.", + "B": "I passed it five times.", + "C": "So far, I have passed by the chair opposite the window two times during my route.", + "D": "So far, I have passed by the chair opposite the window six times as I went back and forth.", + "E": "So far, I have passed by the chair opposite the window four times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 19 + ] + } +] diff --git a/QAs/egolife_A1_D2_029-mcq.json b/QAs/egolife_A1_D2_029-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..dc143b6b4ddfe3913a2bfd57f9addf2fbb8f0abe --- /dev/null +++ b/QAs/egolife_A1_D2_029-mcq.json @@ -0,0 +1,876 @@ +[ + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029_1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I just walked from the second floor to the first floor.", + "question_chinese": "我刚走了什么路线？", + "answer_chinese": "我刚从二楼走到一楼", + "evidence": [ + { + "description": "I'm walking from the second floor to the first floor", + "timestamps": { + "start": "00:02:17", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:03:54", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I just walked from the second floor to the first floor.", + "B": "First floor to second floor.", + "C": "Second floor to third floor.", + "D": "I just walked from the first floor down to the basement, going lower instead of descending one level.", + "E": "I just walked from the second floor up to the rooftop, moving upward rather than down to the first floor." + }, + "answer_label": "A", + "answer_text": "I just walked from the second floor to the first floor.", + "q_id": "egolife_A1_D2_029_1", + "correct_answer": "I just walked from the second floor to the first floor.", + "original_options": { + "A": "I just walked from the second floor to the first floor.", + "B": "First floor to second floor.", + "C": "Second floor to third floor.", + "D": "I just walked from the first floor down to the basement, going lower instead of descending one level.", + "E": "I just walked from the second floor up to the rooftop, moving upward rather than down to the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029_2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I just walked from the first floor to the second floor.", + "question_chinese": "我刚走了什么路线？", + "answer_chinese": "我刚才从一楼走到二楼", + "evidence": [ + { + "description": "I'm walking from the first floor to the second floor", + "timestamps": { + "start": "00:03:16", + "end": "00:03:30" + } + } + ], + "question_timestamps": "00:04:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Second floor to first floor.", + "B": "First floor to third floor.", + "C": "I just walked from the first floor to the second floor.", + "D": "I just walked from the second floor to the third floor, moving up one level from where I started.", + "E": "I just walked from the first floor to the kitchen on the first floor, not upstairs to the second floor." + }, + "answer_label": "C", + "answer_text": "I just walked from the first floor to the second floor.", + "q_id": "egolife_A1_D2_029_2", + "correct_answer": "I just walked from the first floor to the second floor.", + "original_options": { + "A": "Second floor to first floor.", + "B": "First floor to third floor.", + "C": "I just walked from the first floor to the second floor.", + "D": "I just walked from the second floor to the third floor, moving up one level from where I started.", + "E": "I just walked from the first floor to the kitchen on the first floor, not upstairs to the second floor." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many doors have I passed through so far?", + "answer": "I have passed through the door three times so far.", + "question_chinese": "我目前经过了几次门", + "answer_chinese": "我目前经过了3次门", + "evidence": [ + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:18", + "end": "00:02:19" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:33", + "end": "00:02:35" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:59", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I’ve passed through once.", + "B": "I’ve passed through twice.", + "C": "I have passed through the door four times so far, counting each entry and exit as separate.", + "D": "I have passed through the door three times so far.", + "E": "I have passed through the door five times so far, including the last time you crossed into this area." + }, + "answer_label": "D", + "answer_text": "I have passed through the door three times so far.", + "q_id": "egolife_A1_D2_029_7", + "correct_answer": "I have passed through the door three times so far.", + "original_options": { + "A": "I’ve passed through once.", + "B": "I’ve passed through twice.", + "C": "I have passed through the door four times so far, counting each entry and exit as separate.", + "D": "I have passed through the door three times so far.", + "E": "I have passed through the door five times so far, including the last time you crossed into this area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029_6_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Up to now, how many doors have I gone through?", + "answer": "So far you’ve passed through doors five times, and all those doors are behind you.", + "question_chinese": "我目前经过了几次门？", + "answer_chinese": "我目前经过了5次门，门都在你的身后。", + "evidence": [ + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:18", + "end": "00:02:19" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:33", + "end": "00:02:35" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:59", + "end": "00:03:00" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:03:15", + "end": "00:03:16" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:03:26", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far you’ve passed through doors five times, and all those doors are behind you.", + "B": "You’ve gone through three doors.", + "C": "You’ve gone through four doors.", + "D": "So far you’ve passed through doors six times, and every one of those doors is now behind you.", + "E": "So far you’ve passed through doors seven times, and all those doors are behind you as you continue forward." + }, + "answer_label": "A", + "answer_text": "So far you’ve passed through doors five times, and all those doors are behind you.", + "q_id": "egolife_A1_D2_029_6_T1", + "correct_answer": "So far you’ve passed through doors five times, and all those doors are behind you.", + "original_options": { + "A": "So far you’ve passed through doors five times, and all those doors are behind you.", + "B": "You’ve gone through three doors.", + "C": "You’ve gone through four doors.", + "D": "So far you’ve passed through doors six times, and every one of those doors is now behind you.", + "E": "So far you’ve passed through doors seven times, and all those doors are behind you as you continue forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029_8", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many doors have I passed through so far?", + "answer": "I have passed through the door five times so far.", + "question_chinese": "我目前经过了几次门", + "answer_chinese": "我目前经过了5次门", + "evidence": [ + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:18", + "end": "00:02:19" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:33", + "end": "00:02:35" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:59", + "end": "00:03:00" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:03:15", + "end": "00:03:16" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:03:27", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:08:05", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I have passed through the door five times so far.", + "B": "I’ve passed through once.", + "C": "I’ve passed through twice.", + "D": "I have passed through the door three times so far, not counting any time you only opened it.", + "E": "I have passed through the door four times so far, including the most recent crossing you made." + }, + "answer_label": "A", + "answer_text": "I have passed through the door five times so far.", + "q_id": "egolife_A1_D2_029_8", + "correct_answer": "I have passed through the door five times so far.", + "original_options": { + "A": "I have passed through the door five times so far.", + "B": "I’ve passed through once.", + "C": "I’ve passed through twice.", + "D": "I have passed through the door three times so far, not counting any time you only opened it.", + "E": "I have passed through the door four times so far, including the most recent crossing you made." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029_6_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Up to now, how many doors have I gone through?", + "answer": "So far you’ve passed through doors five times, and all those doors are behind you.", + "question_chinese": "我目前经过了几次门？", + "answer_chinese": "我目前经过了5次门，门都在你的身后。", + "evidence": [ + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:18", + "end": "00:02:19" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:33", + "end": "00:02:35" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:59", + "end": "00:03:00" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:03:15", + "end": "00:03:16" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:03:26", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:11:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You’ve gone through three doors.", + "B": "So far you’ve passed through doors five times, and all those doors are behind you.", + "C": "You’ve gone through four doors.", + "D": "So far you’ve passed through doors six times, and all those doors are behind you at this point.", + "E": "So far you’ve passed through doors seven times, and all those doors are behind you as you move on." + }, + "answer_label": "B", + "answer_text": "So far you’ve passed through doors five times, and all those doors are behind you.", + "q_id": "egolife_A1_D2_029_6_T2", + "correct_answer": "So far you’ve passed through doors five times, and all those doors are behind you.", + "original_options": { + "A": "You’ve gone through three doors.", + "B": "So far you’ve passed through doors five times, and all those doors are behind you.", + "C": "You’ve gone through four doors.", + "D": "So far you’ve passed through doors six times, and all those doors are behind you at this point.", + "E": "So far you’ve passed through doors seven times, and all those doors are behind you as you move on." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029-mcq_fixed_2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I touch the whiteboard without going through the door?", + "answer": "YOU can touch the whiteboard without going through the door.", + "question_chinese": "我能不经过门就碰到白板吗？", + "answer_chinese": "我能不经过门就碰到白板", + "evidence": [ + { + "description": "The whiteboard is in the living room on the first floor.", + "timestamps": { + "start": "00:02:46", + "end": "00:02:47" + } + }, + { + "description": "I'm here too.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:41" + } + } + ], + "question_timestamps": "00:03:09", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "YOU can touch the whiteboard without going through the door.", + "B": "YOU cannot touch the whiteboard without going through the door." + }, + "answer_label": "A", + "answer_text": "YOU can touch the whiteboard without going through the door.", + "q_id": "egolife_A1_D2_029-mcq_fixed_2", + "correct_answer": "YOU can touch the whiteboard without going through the door.", + "original_options": { + "A": "YOU can touch the whiteboard without going through the door.", + "B": "YOU cannot touch the whiteboard without going through the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029-mcq_fixed_3", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I touch the whiteboard without going through the door?", + "answer": "YOU couldn't help bumping into the whiteboard without passing through the door.", + "question_chinese": "我能不经过门就碰到白板吗？", + "answer_chinese": "我不能不经过门就碰到白板", + "evidence": [ + { + "description": "The whiteboard is in the living room on the first floor.", + "timestamps": { + "start": "00:02:46", + "end": "00:02:47" + } + }, + { + "description": "I'm on the second floor.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "YOU have to pass through the door to touch the whiteboard.", + "B": "YOU couldn't help bumping into the whiteboard without passing through the door." + }, + "answer_label": "B", + "answer_text": "YOU couldn't help bumping into the whiteboard without passing through the door.", + "q_id": "egolife_A1_D2_029-mcq_fixed_3", + "correct_answer": "YOU couldn't help bumping into the whiteboard without passing through the door.", + "original_options": { + "A": "YOU have to pass through the door to touch the whiteboard.", + "B": "YOU couldn't help bumping into the whiteboard without passing through the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029_7_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the window from me?", + "answer": "The window is behind you.", + "question_chinese": "窗户在我哪个方向？", + "answer_chinese": "窗户在你后面", + "evidence": [ + { + "description": "The window was in front of me and I turned it 180 degrees", + "timestamps": { + "start": "00:01:41", + "end": "00:01:43" + } + } + ], + "question_timestamps": "00:01:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is to your right, facing you as you look across the room from your current spot.", + "D": "The window is ahead of you, positioned so you’d see it straight on without turning around.", + "E": "The window is behind you." + }, + "answer_label": "E", + "answer_text": "The window is behind you.", + "q_id": "egolife_A1_D2_029_7_T1", + "correct_answer": "The window is behind you.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is to your right, facing you as you look across the room from your current spot.", + "D": "The window is ahead of you, positioned so you’d see it straight on without turning around.", + "E": "The window is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029_7_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the window from me?", + "answer": "The window is to your right.", + "question_chinese": "窗户在我哪个方向？", + "answer_chinese": "窗户在你右方。", + "evidence": [ + { + "description": "The window was in front of me and I turned it 180 degrees", + "timestamps": { + "start": "00:01:41", + "end": "00:01:43" + } + } + ], + "question_timestamps": "00:01:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is to your left.", + "B": "The window is behind you.", + "C": "The window is to your right.", + "D": "The window is directly in front of you, visible without turning your head to either side.", + "E": "The window is in front-left of you, requiring a slight turn to the left to face it." + }, + "answer_label": "C", + "answer_text": "The window is to your right.", + "q_id": "egolife_A1_D2_029_7_T2", + "correct_answer": "The window is to your right.", + "original_options": { + "A": "The window is to your left.", + "B": "The window is behind you.", + "C": "The window is to your right.", + "D": "The window is directly in front of you, visible without turning your head to either side.", + "E": "The window is in front-left of you, requiring a slight turn to the left to face it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029_3_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the window facing?", + "answer": "The window is right in front of you.", + "question_chinese": "窗户在我哪个方向？", + "answer_chinese": "窗户在我正面", + "evidence": [ + { + "description": "The window is right in front of me", + "timestamps": { + "start": "00:10:52", + "end": "00:10:53" + } + } + ], + "question_timestamps": "00:11:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is to your left.", + "B": "The window is to your right.", + "C": "The window is right in front of you.", + "D": "The window is behind you, so you’d need to turn around to see it clearly.", + "E": "The window is above you, positioned overhead rather than at eye level in front." + }, + "answer_label": "C", + "answer_text": "The window is right in front of you.", + "q_id": "egolife_A1_D2_029_3_T1", + "correct_answer": "The window is right in front of you.", + "original_options": { + "A": "The window is to your left.", + "B": "The window is to your right.", + "C": "The window is right in front of you.", + "D": "The window is behind you, so you’d need to turn around to see it clearly.", + "E": "The window is above you, positioned overhead rather than at eye level in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029_3_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the window facing?", + "answer": "The window is right in front of you.", + "question_chinese": "窗户在我哪个方向？", + "answer_chinese": "窗户在我正面", + "evidence": [ + { + "description": "The window is right in front of me", + "timestamps": { + "start": "00:10:52", + "end": "00:10:53" + } + } + ], + "question_timestamps": "00:11:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is to your left.", + "B": "The window is to your right.", + "C": "The window is right in front of you.", + "D": "The window is behind you, meaning it’s not in your forward line of sight.", + "E": "The window is overhead, so you would look up instead of straight ahead to face it." + }, + "answer_label": "C", + "answer_text": "The window is right in front of you.", + "q_id": "egolife_A1_D2_029_3_T2", + "correct_answer": "The window is right in front of you.", + "original_options": { + "A": "The window is to your left.", + "B": "The window is to your right.", + "C": "The window is right in front of you.", + "D": "The window is behind you, meaning it’s not in your forward line of sight.", + "E": "The window is overhead, so you would look up instead of straight ahead to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029-mcq_fixed_4", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are there here?", + "answer": "Six.", + "question_chinese": "这里有几个人？", + "answer_chinese": "这里有5个人", + "evidence": [ + { + "description": "There are four people in my field of vision", + "timestamps": { + "start": "00:02:06", + "end": "00:02:07" + } + }, + { + "description": "I'm here too", + "timestamps": { + "start": "00:02:00", + "end": "00:02:02" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Six.", + "C": "Seven.", + "D": "There are four people standing here in the scene, fewer than six in total.", + "E": "There are eight people present here, which is more than six in the scene." + }, + "answer_label": "B", + "answer_text": "Six.", + "q_id": "egolife_A1_D2_029-mcq_fixed_4", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Six.", + "C": "Seven.", + "D": "There are four people standing here in the scene, fewer than six in total.", + "E": "There are eight people present here, which is more than six in the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_029-mcq_fixed_5", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are there here?", + "answer": "One.", + "question_chinese": "这里有几个人？", + "answer_chinese": "这里有3个人", + "evidence": [ + { + "description": "There are two people in my field of vision", + "timestamps": { + "start": "00:02:47", + "end": "00:02:50" + } + }, + { + "description": "I'm here too", + "timestamps": { + "start": "00:02:50", + "end": "00:02:51" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "There are four people here in total, which is more than one person.", + "D": "One.", + "E": "There are five people here altogether, not just a single person in the scene." + }, + "answer_label": "D", + "answer_text": "One.", + "q_id": "egolife_A1_D2_029-mcq_fixed_5", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "There are four people here in total, which is more than one person.", + "D": "One.", + "E": "There are five people here altogether, not just a single person in the scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D2_030-mcq.json b/QAs/egolife_A1_D2_030-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..83e5f768cb3011f1744012c17b826aa7d9d93440 --- /dev/null +++ b/QAs/egolife_A1_D2_030-mcq.json @@ -0,0 +1,696 @@ +[ + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding just now?", + "answer": "The broom I was holding just now.", + "question_chinese": "刚才我手上拿的什么?", + "answer_chinese": "刚才我手上拿的扫把", + "evidence": [ + { + "description": "The broom in my hand now", + "timestamps": { + "start": "00:07:56", + "end": "00:08:00" + } + } + ], + "question_timestamps": "00:08:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The shovel I was holding.", + "B": "The rake I was holding.", + "C": "The broom I was holding just now.", + "D": "The bucket I was holding just now, not a long-handled cleaning tool like a broom.", + "E": "The spade I was holding just now, rather than a broom used for sweeping the floor." + }, + "answer_label": "C", + "answer_text": "The broom I was holding just now.", + "q_id": "egolife_A1_D2_030_7", + "correct_answer": "The broom I was holding just now.", + "original_options": { + "A": "The shovel I was holding.", + "B": "The rake I was holding.", + "C": "The broom I was holding just now.", + "D": "The bucket I was holding just now, not a long-handled cleaning tool like a broom.", + "E": "The spade I was holding just now, rather than a broom used for sweeping the floor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030-mcq_fixed_7", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "The mobile phone you were holding just now.", + "question_chinese": "刚才我手上拿的什么?", + "answer_chinese": "刚才我手上拿的手机", + "evidence": [ + { + "description": "The mobile phone I'm holding in my hand now", + "timestamps": { + "start": "00:08:20", + "end": "00:08:23" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The mobile phone you were holding just now.", + "B": "The key you were holding.", + "C": "The notebook you were holding.", + "D": "The tablet device you were holding just now, larger than a phone and held with both hands.", + "E": "The paperback book you were holding just now, something you’d read rather than tap like a phone." + }, + "answer_label": "A", + "answer_text": "The mobile phone you were holding just now.", + "q_id": "egolife_A1_D2_030-mcq_fixed_7", + "correct_answer": "The mobile phone you were holding just now.", + "original_options": { + "A": "The mobile phone you were holding just now.", + "B": "The key you were holding.", + "C": "The notebook you were holding.", + "D": "The tablet device you were holding just now, larger than a phone and held with both hands.", + "E": "The paperback book you were holding just now, something you’d read rather than tap like a phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I just walked from the bathroom to the living room.", + "question_chinese": "我刚才走了什么路线？", + "answer_chinese": "我刚才从卫生间走到客厅", + "evidence": [ + { + "description": "I'm walking from the bathroom to the living room", + "timestamps": { + "start": "00:00:43", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Bathroom to dining room.", + "B": "Bathroom to hallway.", + "C": "I just walked from the living room to the bathroom, reversing the route you took a moment ago.", + "D": "I just walked from the living room to the kitchen, ending somewhere different than the living room.", + "E": "I just walked from the bathroom to the living room." + }, + "answer_label": "E", + "answer_text": "I just walked from the bathroom to the living room.", + "q_id": "egolife_A1_D2_030_1", + "correct_answer": "I just walked from the bathroom to the living room.", + "original_options": { + "A": "Bathroom to dining room.", + "B": "Bathroom to hallway.", + "C": "I just walked from the living room to the bathroom, reversing the route you took a moment ago.", + "D": "I just walked from the living room to the kitchen, ending somewhere different than the living room.", + "E": "I just walked from the bathroom to the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I just walked from the living room to the bedroom.", + "question_chinese": "我刚才走了什么路线？", + "answer_chinese": "我刚才从客厅走到卧室", + "evidence": [ + { + "description": "I'm walking from the living room to the bedroom", + "timestamps": { + "start": "00:01:57", + "end": "00:02:09" + } + } + ], + "question_timestamps": "00:03:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Living room to dining room.", + "B": "Living room to hallway.", + "C": "I just walked from the hallway to the bedroom, starting in a different place than the living room.", + "D": "I just walked from the living room to the bedroom.", + "E": "I just walked from the bedroom to the living room, which is the opposite direction of your movement." + }, + "answer_label": "D", + "answer_text": "I just walked from the living room to the bedroom.", + "q_id": "egolife_A1_D2_030_2", + "correct_answer": "I just walked from the living room to the bedroom.", + "original_options": { + "A": "Living room to dining room.", + "B": "Living room to hallway.", + "C": "I just walked from the hallway to the bedroom, starting in a different place than the living room.", + "D": "I just walked from the living room to the bedroom.", + "E": "I just walked from the bedroom to the living room, which is the opposite direction of your movement." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030_4_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see here?", + "answer": "You saw two people here. These two people were in front of you.", + "question_chinese": "我看到这里有几个人?", + "answer_chinese": "你看到这里有两个人。这两个人在你的前面。", + "evidence": [ + { + "description": "You see there are two people here.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one person here.", + "B": "You saw two people here. These two people were in front of you.", + "C": "You saw three people here.", + "D": "You saw four people here, and all of them were standing in front of you.", + "E": "You saw zero people here, with no one visible anywhere in front of you." + }, + "answer_label": "B", + "answer_text": "You saw two people here. These two people were in front of you.", + "q_id": "egolife_A1_D2_030_4_T2", + "correct_answer": "You saw two people here. These two people were in front of you.", + "original_options": { + "A": "You saw one person here.", + "B": "You saw two people here. These two people were in front of you.", + "C": "You saw three people here.", + "D": "You saw four people here, and all of them were standing in front of you.", + "E": "You saw zero people here, with no one visible anywhere in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030_4_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people did I see here?", + "answer": "You saw one person here. This person was to your left.", + "question_chinese": "我看到这里有几个人?", + "answer_chinese": "你看到这里有一个人。这一个人在你的左边。", + "evidence": [ + { + "description": "I'm here", + "timestamps": { + "start": "00:06:26", + "end": "00:06:29" + } + } + ], + "question_timestamps": "00:06:31", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one person here. This person was to your left.", + "B": "You saw zero people here.", + "C": "You saw two people here.", + "D": "You saw three people here, and all three were positioned to your left.", + "E": "You saw four people here, with everyone clustered together on your left side." + }, + "answer_label": "A", + "answer_text": "You saw one person here. This person was to your left.", + "q_id": "egolife_A1_D2_030_4_T1", + "correct_answer": "You saw one person here. This person was to your left.", + "original_options": { + "A": "You saw one person here. This person was to your left.", + "B": "You saw zero people here.", + "C": "You saw two people here.", + "D": "You saw three people here, and all three were positioned to your left.", + "E": "You saw four people here, with everyone clustered together on your left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030-mcq_fixed_9", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the whiteboard without going through the door?", + "answer": "YOU couldn't help bumping into the whiteboard without passing through the door.", + "question_chinese": "我能不经过门就碰到白板吗?", + "answer_chinese": "我不能不经过门就碰到白板", + "evidence": [ + { + "description": "The whiteboard is in the living room on the first floor", + "timestamps": { + "start": "00:01:25", + "end": "00:01:27" + } + }, + { + "description": "There is a door between the living room and the bedroom", + "timestamps": { + "start": "00:01:37", + "end": "00:01:38" + } + }, + { + "description": "I'm in the bedroom.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:52" + } + } + ], + "question_timestamps": "00:05:09", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "YOU couldn't help bumping into the whiteboard without passing through the door.", + "B": "YOU managed to reach the whiteboard only after going through the door." + }, + "answer_label": "A", + "answer_text": "YOU couldn't help bumping into the whiteboard without passing through the door.", + "q_id": "egolife_A1_D2_030-mcq_fixed_9", + "correct_answer": "YOU couldn't help bumping into the whiteboard without passing through the door.", + "original_options": { + "A": "YOU couldn't help bumping into the whiteboard without passing through the door.", + "B": "YOU managed to reach the whiteboard only after going through the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030-mcq_fixed_8", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the whiteboard without going through the door?", + "answer": "YOU can touch the whiteboard without going through the door.", + "question_chinese": "我能不经过门就碰到白板吗?", + "answer_chinese": "我能不经过门就碰到白板", + "evidence": [ + { + "description": "The whiteboard is in the living room on the first floor", + "timestamps": { + "start": "00:01:25", + "end": "00:01:27" + } + }, + { + "description": "I'm also in the living room on the first floor", + "timestamps": { + "start": "00:07:54", + "end": "00:07:55" + } + } + ], + "question_timestamps": "00:09:24", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "YOU cannot touch the whiteboard without going through the door.", + "B": "YOU can touch the whiteboard without going through the door." + }, + "answer_label": "B", + "answer_text": "YOU can touch the whiteboard without going through the door.", + "q_id": "egolife_A1_D2_030-mcq_fixed_8", + "correct_answer": "YOU can touch the whiteboard without going through the door.", + "original_options": { + "A": "YOU cannot touch the whiteboard without going through the door.", + "B": "YOU can touch the whiteboard without going through the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030_3", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the window from me?", + "answer": "The window is behind me.", + "question_chinese": "窗户在我什么方向?", + "answer_chinese": "窗户在我后面", + "evidence": [ + { + "description": "The window is right in front of me", + "timestamps": { + "start": "00:01:27", + "end": "00:01:28" + } + }, + { + "description": "I'm turning 180 degrees", + "timestamps": { + "start": "00:01:29", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:01:39", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The window is in front of me.", + "B": "The window is behind me.", + "C": "The window is to my left.", + "D": "The window is to my right, not behind me from where I'm standing.", + "E": "The window is above my line of sight, so it isn't located behind me." + }, + "answer_label": "B", + "answer_text": "The window is behind me.", + "q_id": "egolife_A1_D2_030_3", + "correct_answer": "The window is behind me.", + "original_options": { + "A": "The window is in front of me.", + "B": "The window is behind me.", + "C": "The window is to my left.", + "D": "The window is to my right, not behind me from where I'm standing.", + "E": "The window is above my line of sight, so it isn't located behind me." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030_4", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the window facing from my position?", + "answer": "The window is on my left.", + "question_chinese": "窗户在我什么方向?", + "answer_chinese": "窗户在我的左边", + "evidence": [ + { + "description": "I'm turning 90 degrees", + "timestamps": { + "start": "00:02:53", + "end": "00:03:20" + } + }, + { + "description": "The window is in front of me", + "timestamps": { + "start": "00:02:21", + "end": "00:02:42" + } + } + ], + "question_timestamps": "00:03:38", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The window is on my right.", + "B": "The window is on my left." + }, + "answer_label": "B", + "answer_text": "The window is on my left.", + "q_id": "egolife_A1_D2_030_4", + "correct_answer": "The window is on my left.", + "original_options": { + "A": "The window is on my right.", + "B": "The window is on my left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030_5", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are here?", + "answer": "There is one person here.", + "question_chinese": "这里有几个人?", + "answer_chinese": "这里有1个人", + "evidence": [ + { + "description": "There is no one in sight", + "timestamps": { + "start": "00:06:00", + "end": "00:06:05" + } + }, + { + "description": "I'm here", + "timestamps": { + "start": "00:06:05", + "end": "00:06:06" + } + } + ], + "question_timestamps": "00:06:53", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are two persons here.", + "B": "There are three persons here.", + "C": "There are four persons here, standing together in the same area you are looking at.", + "D": "There are five persons here, with multiple people visible at the same time.", + "E": "There is one person here." + }, + "answer_label": "E", + "answer_text": "There is one person here.", + "q_id": "egolife_A1_D2_030_5", + "correct_answer": "There is one person here.", + "original_options": { + "A": "There are two persons here.", + "B": "There are three persons here.", + "C": "There are four persons here, standing together in the same area you are looking at.", + "D": "There are five persons here, with multiple people visible at the same time.", + "E": "There is one person here." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_030-mcq_fixed_5", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are there here?", + "answer": "Six.", + "question_chinese": "这里有几个人?", + "answer_chinese": "这里有三个人", + "evidence": [ + { + "description": "There are two people in sight", + "timestamps": { + "start": "00:07:07", + "end": "00:07:08" + } + }, + { + "description": "I'm here", + "timestamps": { + "start": "00:07:08", + "end": "00:07:09" + } + } + ], + "question_timestamps": "00:07:32", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "Seven people are present, which is one more than the correct count of six.", + "E": "Eight people are present, which is two more than the correct count of six." + }, + "answer_label": "B", + "answer_text": "Six.", + "q_id": "egolife_A1_D2_030-mcq_fixed_5", + "correct_answer": "Six.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "Seven people are present, which is one more than the correct count of six.", + "E": "Eight people are present, which is two more than the correct count of six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D2_034-mcq.json b/QAs/egolife_A1_D2_034-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e51f2751014ae459ce8e8f2ea820d876651a4e35 --- /dev/null +++ b/QAs/egolife_A1_D2_034-mcq.json @@ -0,0 +1,619 @@ +[ + { + "video_uid": "egolife_A1_D2_034", + "video_length": "00:11:30", + "qid": "egolife_A1_D2_034-mcq_fixed_2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the first floor to the second floor?", + "answer": "At the beginning, You were standing on the first floor, then YOU went up two flights of stairs and arrived at the second floor.", + "question_chinese": "从一楼到二楼的路径是什么？", + "answer_chinese": "开始时，我站在一楼，然后我上了俩层楼梯，到达二楼。", + "evidence": [ + { + "description": "At the beginning, I was standing on the first floor.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:36" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:05:36", + "end": "00:06:23" + } + } + ], + "question_timestamps": "00:06:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took one flight of stairs.", + "B": "You took three flights of stairs.", + "C": "You started on the first floor, climbed one flight of stairs, and ended up on the third floor.", + "D": "From the first floor, you climbed three flights of stairs before reaching the third floor landing.", + "E": "At the beginning, You were standing on the first floor, then YOU went up two flights of stairs and arrived at the second floor." + }, + "answer_label": "E", + "answer_text": "At the beginning, You were standing on the first floor, then YOU went up two flights of stairs and arrived at the second floor.", + "q_id": "egolife_A1_D2_034-mcq_fixed_2", + "correct_answer": "At the beginning, You were standing on the first floor, then YOU went up two flights of stairs and arrived at the second floor.", + "original_options": { + "A": "You took one flight of stairs.", + "B": "You took three flights of stairs.", + "C": "You started on the first floor, climbed one flight of stairs, and ended up on the third floor.", + "D": "From the first floor, you climbed three flights of stairs before reaching the third floor landing.", + "E": "At the beginning, You were standing on the first floor, then YOU went up two flights of stairs and arrived at the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_034", + "video_length": "00:11:30", + "qid": "egolife_A1_D2_034-mcq_fixed_3", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the first floor to the second floor?", + "answer": "You go from the first floor to the second floor.", + "question_chinese": "从一楼到二楼的路径是什么？", + "answer_chinese": "我从一楼，到达二楼。", + "evidence": [ + { + "description": "At the beginning, I was standing on the first floor.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:36" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:05:36", + "end": "00:06:23" + } + } + ], + "question_timestamps": "00:11:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From first floor to basement.", + "B": "From second floor to first.", + "C": "You go from the first floor to the second floor.", + "D": "You leave the first floor, head down to the basement, then come back up to the second floor.", + "E": "You start on the first floor, walk along a corridor, and return back to the first floor." + }, + "answer_label": "C", + "answer_text": "You go from the first floor to the second floor.", + "q_id": "egolife_A1_D2_034-mcq_fixed_3", + "correct_answer": "You go from the first floor to the second floor.", + "original_options": { + "A": "From first floor to basement.", + "B": "From second floor to first.", + "C": "You go from the first floor to the second floor.", + "D": "You leave the first floor, head down to the basement, then come back up to the second floor.", + "E": "You start on the first floor, walk along a corridor, and return back to the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_034", + "video_length": "00:11:30", + "qid": "egolife_A1_D2_034-mcq_fixed_0", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I just stepped out of the bathroom door, where was the toilet in relation to me?", + "answer": "Behind you.", + "question_chinese": "当我刚出厕所门时，马桶在我的什么位置？", + "answer_chinese": "在我的身后。", + "evidence": [ + { + "description": "When I stepped through the bathroom door, the toilet was directly in front of me.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:32" + } + }, + { + "description": "I turned 180 degrees.", + "timestamps": { + "start": "00:02:38", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:02:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "It was in front of you as you stepped out, just past the bathroom doorway.", + "D": "It sat to your right side when you exited the bathroom, near where you had been standing.", + "E": "Behind you." + }, + "answer_label": "E", + "answer_text": "Behind you.", + "q_id": "egolife_A1_D2_034-mcq_fixed_0", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "It was in front of you as you stepped out, just past the bathroom doorway.", + "D": "It sat to your right side when you exited the bathroom, near where you had been standing.", + "E": "Behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_034", + "video_length": "00:11:30", + "qid": "egolife_A1_D2_034-mcq_fixed_1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I just stepped out of the bathroom door, where was the toilet in relation to me?", + "answer": "To your left rear.", + "question_chinese": "当我刚出厕所门时，马桶在我的什么位置？", + "answer_chinese": "在我的左后方。", + "evidence": [ + { + "description": "When I stepped through the bathroom door, the toilet was directly in front of me.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:31" + } + }, + { + "description": "I turned 180 degrees.", + "timestamps": { + "start": "00:02:38", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:08:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To your left front.", + "B": "Directly behind you.", + "C": "It was slightly behind and to your right as you stepped out of the bathroom doorway.", + "D": "The toilet was off to your left side, more in line with your shoulder than behind you.", + "E": "To your left rear." + }, + "answer_label": "E", + "answer_text": "To your left rear.", + "q_id": "egolife_A1_D2_034-mcq_fixed_1", + "correct_answer": "To your left rear.", + "original_options": { + "A": "To your left front.", + "B": "Directly behind you.", + "C": "It was slightly behind and to your right as you stepped out of the bathroom doorway.", + "D": "The toilet was off to your left side, more in line with your shoulder than behind you.", + "E": "To your left rear." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_034", + "video_length": "00:11:30", + "qid": "egolife_A1_D2_034-mcq_fixed_8", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many shampoos/body washes are there?", + "answer": "Four.", + "question_chinese": "有几个洗发水/沐浴露？", + "answer_chinese": "四个。", + "evidence": [ + { + "description": "I entered the second-floor restroom.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:31" + } + }, + { + "description": "I saw four bottles of shampoo/body wash on the windowsill of the toilet.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:36" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Five, counting an extra bottle as another shampoo/body wash in the set.", + "D": "Three, if you treat one bottle as conditioner only and don’t count it as shampoo/body wash.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "egolife_A1_D2_034-mcq_fixed_8", + "correct_answer": "Four.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Five, counting an extra bottle as another shampoo/body wash in the set.", + "D": "Three, if you treat one bottle as conditioner only and don’t count it as shampoo/body wash.", + "E": "Four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_034", + "video_length": "00:11:30", + "qid": "egolife_A1_D2_034-mcq_fixed_9", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many shampoos/body washes are there?", + "answer": "Three.", + "question_chinese": "有几个洗发水/沐浴露？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "I entered the second-floor restroom.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:32" + } + }, + { + "description": "I saw three bottles of shampoo/body wash on the windowsill of the toilet.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:36" + } + } + ], + "question_timestamps": "00:11:23", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "Five, counting two similar bottles separately as shampoos/body washes in the total.", + "E": "Six, if you include every bottle on the shelf as either shampoo or body wash." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_034-mcq_fixed_9", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "Five, counting two similar bottles separately as shampoos/body washes in the total.", + "E": "Six, if you include every bottle on the shelf as either shampoo or body wash." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_034", + "video_length": "00:11:30", + "qid": "egolife_A1_D2_034_7", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many people have I seen in total?", + "answer": "A person.", + "question_chinese": "此时，我一共看见了几个人？", + "answer_chinese": "一个人。", + "evidence": [ + { + "description": "As I was going out through the living room door, I brushed past a person.", + "timestamps": { + "start": "00:03:06", + "end": "00:03:07" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two people.", + "B": "Three people.", + "C": "A person.", + "D": "Four people, counting the individuals you passed earlier as well as the one you just saw.", + "E": "Five people, if you include everyone briefly visible in the background as separate people." + }, + "answer_label": "C", + "answer_text": "A person.", + "q_id": "egolife_A1_D2_034_7", + "correct_answer": "A person.", + "original_options": { + "A": "Two people.", + "B": "Three people.", + "C": "A person.", + "D": "Four people, counting the individuals you passed earlier as well as the one you just saw.", + "E": "Five people, if you include everyone briefly visible in the background as separate people." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_034", + "video_length": "00:11:30", + "qid": "egolife_A1_D2_034_8", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many people have I seen in total?", + "answer": "Four people.", + "question_chinese": "此时，我一共看见了几个人？", + "answer_chinese": "四个人。", + "evidence": [ + { + "description": "I saw two people in the living room.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:20" + } + }, + { + "description": "I opened the door and there were two boys.", + "timestamps": { + "start": "00:10:09", + "end": "00:10:10" + } + } + ], + "question_timestamps": "00:10:34", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two people.", + "B": "Three people.", + "C": "Four people.", + "D": "Five people, assuming you count an additional person you glimpsed only briefly in passing.", + "E": "One person, if you only count the most recent individual you saw and ignore earlier encounters." + }, + "answer_label": "C", + "answer_text": "Four people.", + "q_id": "egolife_A1_D2_034_8", + "correct_answer": "Four people.", + "original_options": { + "A": "Two people.", + "B": "Three people.", + "C": "Four people.", + "D": "Five people, assuming you count an additional person you glimpsed only briefly in passing.", + "E": "One person, if you only count the most recent individual you saw and ignore earlier encounters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_034", + "video_length": "00:11:30", + "qid": "egolife_A1_D2_034_5", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine located in relation to me right now?", + "answer": "The washing machine is to my right rear.", + "question_chinese": "此时洗衣机相对于我在什么位置？", + "answer_chinese": "洗衣机在我的右后方。", + "evidence": [ + { + "description": "As I walked toward the door leading outside, I saw the washing machine in front of me to the right.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:11" + } + }, + { + "description": "I walked to the door.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:03:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The washing machine is to my left rear.", + "B": "The washing machine is in front of me.", + "C": "The washing machine is off to my right side, roughly even with where I'm standing right now.", + "D": "The washing machine is to my right rear.", + "E": "The washing machine sits to my left front, slightly ahead of me as I face forward." + }, + "answer_label": "D", + "answer_text": "The washing machine is to my right rear.", + "q_id": "egolife_A1_D2_034_5", + "correct_answer": "The washing machine is to my right rear.", + "original_options": { + "A": "The washing machine is to my left rear.", + "B": "The washing machine is in front of me.", + "C": "The washing machine is off to my right side, roughly even with where I'm standing right now.", + "D": "The washing machine is to my right rear.", + "E": "The washing machine sits to my left front, slightly ahead of me as I face forward." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_034", + "video_length": "00:11:30", + "qid": "egolife_A1_D2_034_6", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine located in relation to me right now?", + "answer": "The washing machine is to my left rear.", + "question_chinese": "此时洗衣机相对于我在什么位置？", + "answer_chinese": "洗衣机在我的左后方。", + "evidence": [ + { + "description": "When I walked backward toward the stairs, the washing machine was in front of me to the right.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:42" + } + }, + { + "description": "When I turn 180 degrees.", + "timestamps": { + "start": "00:04:41", + "end": "00:04:42" + } + } + ], + "question_timestamps": "00:07:35", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The washing machine is to my left rear.", + "B": "The washing machine is to my right rear.", + "C": "The washing machine is to my left front.", + "D": "The washing machine is directly in front of me, centered ahead as I look forward.", + "E": "The washing machine is on my right side, beside me rather than behind my shoulder." + }, + "answer_label": "A", + "answer_text": "The washing machine is to my left rear.", + "q_id": "egolife_A1_D2_034_6", + "correct_answer": "The washing machine is to my left rear.", + "original_options": { + "A": "The washing machine is to my left rear.", + "B": "The washing machine is to my right rear.", + "C": "The washing machine is to my left front.", + "D": "The washing machine is directly in front of me, centered ahead as I look forward.", + "E": "The washing machine is on my right side, beside me rather than behind my shoulder." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D2_035-mcq.json b/QAs/egolife_A1_D2_035-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3fd04332503434406b555e5fe11b3521160c3e7c --- /dev/null +++ b/QAs/egolife_A1_D2_035-mcq.json @@ -0,0 +1,746 @@ +[ + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035-mcq_fixed_8", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the first floor to the second floor?", + "answer": "At the beginning, You were standing on the first floor, then YOU went up two flights of stairs and arrived at the second floor.", + "question_chinese": "从一楼到二楼的路径是什么？", + "answer_chinese": "开始时，我站在一楼，然后我上了俩层楼梯，到达二楼。", + "evidence": [ + { + "description": "At the beginning, I was standing on the first floor.", + "timestamps": { + "start": "00:10:54", + "end": "00:10:55" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:05" + } + } + ], + "question_timestamps": "00:11:07", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went up one flight of stairs.", + "B": "You went up three flights of stairs.", + "C": "At the beginning, You were standing on the first floor, then YOU went up two flights of stairs and arrived at the second floor.", + "D": "You started on the first floor, took a single flight of stairs, and ended up on the second floor.", + "E": "You began on the first floor, climbed three stair flights in a row, and then arrived at the second floor." + }, + "answer_label": "C", + "answer_text": "At the beginning, You were standing on the first floor, then YOU went up two flights of stairs and arrived at the second floor.", + "q_id": "egolife_A1_D2_035-mcq_fixed_8", + "correct_answer": "At the beginning, You were standing on the first floor, then YOU went up two flights of stairs and arrived at the second floor.", + "original_options": { + "A": "You went up one flight of stairs.", + "B": "You went up three flights of stairs.", + "C": "At the beginning, You were standing on the first floor, then YOU went up two flights of stairs and arrived at the second floor.", + "D": "You started on the first floor, took a single flight of stairs, and ended up on the second floor.", + "E": "You began on the first floor, climbed three stair flights in a row, and then arrived at the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035-mcq_fixed_9", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the first floor to the second floor?", + "answer": "You go from the first floor to the second floor.", + "question_chinese": "从一楼到二楼的路径是什么？", + "answer_chinese": "我从一楼到达二楼。", + "evidence": [ + { + "description": "At the beginning, I was standing on the first floor.", + "timestamps": { + "start": "00:10:54", + "end": "00:10:56" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:10:56", + "end": "00:11:06" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You go second floor to first.", + "B": "You go basement to second floor.", + "C": "You go from the first floor to the second floor.", + "D": "You start on the first floor, go down to the basement, then take stairs up to the second floor.", + "E": "You begin on the first floor, walk through a long corridor, then reach a different stairwell to the second floor." + }, + "answer_label": "C", + "answer_text": "You go from the first floor to the second floor.", + "q_id": "egolife_A1_D2_035-mcq_fixed_9", + "correct_answer": "You go from the first floor to the second floor.", + "original_options": { + "A": "You go second floor to first.", + "B": "You go basement to second floor.", + "C": "You go from the first floor to the second floor.", + "D": "You start on the first floor, go down to the basement, then take stairs up to the second floor.", + "E": "You begin on the first floor, walk through a long corridor, then reach a different stairwell to the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035-mcq_fixed_6", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I am at the door, where is the green suitcase relative to me?", + "answer": "To your left rear.", + "question_chinese": "当我在门口时，绿色行李箱相对于我在哪个位置？", + "answer_chinese": "在我的左后方。", + "evidence": [ + { + "description": "When I first walked in, the green suitcase was in front of me to the right.", + "timestamps": { + "start": "00:09:08", + "end": "00:09:09" + } + }, + { + "description": "I turned around and went out.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:29" + } + } + ], + "question_timestamps": "00:09:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To your right rear.", + "B": "To your left front.", + "C": "To your left rear.", + "D": "Directly in front of you, sitting near the doorway as you face out.", + "E": "On your right side, slightly behind you, close to the entrance area by the wall." + }, + "answer_label": "C", + "answer_text": "To your left rear.", + "q_id": "egolife_A1_D2_035-mcq_fixed_6", + "correct_answer": "To your left rear.", + "original_options": { + "A": "To your right rear.", + "B": "To your left front.", + "C": "To your left rear.", + "D": "Directly in front of you, sitting near the doorway as you face out.", + "E": "On your right side, slightly behind you, close to the entrance area by the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035-mcq_fixed_7", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I am at the door, where is the green suitcase relative to me?", + "answer": "In front of you.", + "question_chinese": "当我在门口时，绿色行李箱相对于我在哪个位置？", + "answer_chinese": "在我的前方。", + "evidence": [ + { + "description": "When I first walked in, the green suitcase was in front of me to the right.", + "timestamps": { + "start": "00:09:08", + "end": "00:09:09" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left side.", + "C": "To your right front, a little ahead of you as you stand at the door.", + "D": "In front of you.", + "E": "Just behind your left shoulder, positioned near the entry area rather than in front of you." + }, + "answer_label": "D", + "answer_text": "In front of you.", + "q_id": "egolife_A1_D2_035-mcq_fixed_7", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left side.", + "C": "To your right front, a little ahead of you as you stand at the door.", + "D": "In front of you.", + "E": "Just behind your left shoulder, positioned near the entry area rather than in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035-mcq_fixed_4", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "The laptop is behind you.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "笔记本电脑在我的后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:05:39", + "end": "00:05:40" + } + }, + { + "description": "I turned 90 degrees and left the dormitory.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:16" + } + } + ], + "question_timestamps": "00:06:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The laptop is in front of you.", + "B": "The laptop is to your left.", + "C": "The laptop is to your right side, roughly beside you as you step out of the bedroom door.", + "D": "The laptop is behind you.", + "E": "The laptop is ahead and slightly to your right, positioned in front of you rather than behind." + }, + "answer_label": "D", + "answer_text": "The laptop is behind you.", + "q_id": "egolife_A1_D2_035-mcq_fixed_4", + "correct_answer": "The laptop is behind you.", + "original_options": { + "A": "The laptop is in front of you.", + "B": "The laptop is to your left.", + "C": "The laptop is to your right side, roughly beside you as you step out of the bedroom door.", + "D": "The laptop is behind you.", + "E": "The laptop is ahead and slightly to your right, positioned in front of you rather than behind." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035-mcq_fixed_5", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "When You stepped out of the bedroom door, the laptop was behind you to the left.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "我刚出卧室门时，笔记本电脑在我的左后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:05:52", + "end": "00:05:53" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:16" + } + } + ], + "question_timestamps": "00:08:54", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you to the right.", + "B": "Behind you to the right.", + "C": "When you stepped out of the bedroom door, the laptop was in front of you to the left.", + "D": "When you stepped out of the bedroom door, the laptop was behind you, but positioned on your right side.", + "E": "When You stepped out of the bedroom door, the laptop was behind you to the left." + }, + "answer_label": "E", + "answer_text": "When You stepped out of the bedroom door, the laptop was behind you to the left.", + "q_id": "egolife_A1_D2_035-mcq_fixed_5", + "correct_answer": "When You stepped out of the bedroom door, the laptop was behind you to the left.", + "original_options": { + "A": "In front of you to the right.", + "B": "Behind you to the right.", + "C": "When you stepped out of the bedroom door, the laptop was in front of you to the left.", + "D": "When you stepped out of the bedroom door, the laptop was behind you, but positioned on your right side.", + "E": "When You stepped out of the bedroom door, the laptop was behind you to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035_2_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many cell phones are on my desk?", + "answer": "There was one on your table, and the table was front-right of you.", + "question_chinese": "此时，我的桌子上有几个手机？", + "answer_chinese": "你的桌子上有一个，桌子在你右前边", + "evidence": [ + { + "description": "At the beginning, there was a mobile phone in the upper left corner of the table.", + "timestamps": { + "start": "00:00:56", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There were two on your table.", + "B": "There were zero on your table.", + "C": "There were three on your table, and the table was front-right of you when you looked over.", + "D": "There was one on your table, and the table was front-right of you.", + "E": "There were four on your table, with the table positioned front-right of you at that moment." + }, + "answer_label": "D", + "answer_text": "There was one on your table, and the table was front-right of you.", + "q_id": "egolife_A1_D2_035_2_T1", + "correct_answer": "There was one on your table, and the table was front-right of you.", + "original_options": { + "A": "There were two on your table.", + "B": "There were zero on your table.", + "C": "There were three on your table, and the table was front-right of you when you looked over.", + "D": "There was one on your table, and the table was front-right of you.", + "E": "There were four on your table, with the table positioned front-right of you at that moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035_2_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many cell phones are on my desk?", + "answer": "There are two on your table, and the table is in front of you.", + "question_chinese": "此时，我的桌子上有几个手机？", + "answer_chinese": "你的桌子上有两个，桌子在你前面。", + "evidence": [ + { + "description": "A boy placed his mobile phone on the table.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + }, + { + "description": "A girl placed her phone on the table.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:08:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There is one on your table.", + "B": "There are three on your table.", + "C": "There are zero on your table, and the table is in front of you as you face it.", + "D": "There are four on your table, and the table is directly in front of you right now.", + "E": "There are two on your table, and the table is in front of you." + }, + "answer_label": "E", + "answer_text": "There are two on your table, and the table is in front of you.", + "q_id": "egolife_A1_D2_035_2_T2", + "correct_answer": "There are two on your table, and the table is in front of you.", + "original_options": { + "A": "There is one on your table.", + "B": "There are three on your table.", + "C": "There are zero on your table, and the table is in front of you as you face it.", + "D": "There are four on your table, and the table is directly in front of you right now.", + "E": "There are two on your table, and the table is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035_1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At this moment, what items do I see on the table?", + "answer": "Two.", + "question_chinese": "此时，我看见桌子上有几种物品？", + "answer_chinese": "俩个。", + "evidence": [ + { + "description": "I see many flowers on the table, and these are plants.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "A KFC bag is placed in the corner of the table.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Four items, clearly visible on the table as you look down at it.", + "E": "Five items, all currently arranged and visible on the table in front of you." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "egolife_A1_D2_035_1", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Four items, clearly visible on the table as you look down at it.", + "E": "Five items, all currently arranged and visible on the table in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035_2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At this moment, what items do I see on the table?", + "answer": "Four.", + "question_chinese": "此时，我看见桌子上有几种物品？", + "answer_chinese": "四个。", + "evidence": [ + { + "description": "There are plants and flowers.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + }, + { + "description": "There is food, grapes.", + "timestamps": { + "start": "00:02:46", + "end": "00:02:47" + } + }, + { + "description": "There are electronic products such as mobile phones, power banks, etc.", + "timestamps": { + "start": "00:02:46", + "end": "00:02:47" + } + }, + { + "description": "There are furnishings, such as plates.", + "timestamps": { + "start": "00:02:45", + "end": "00:02:46" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Four.", + "C": "Three.", + "D": "Two items, clearly visible on the table as you look down at it.", + "E": "Five items, all currently present and visible on the table in front of you." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "egolife_A1_D2_035_2", + "correct_answer": "Four.", + "original_options": { + "A": "One.", + "B": "Four.", + "C": "Three.", + "D": "Two items, clearly visible on the table as you look down at it.", + "E": "Five items, all currently present and visible on the table in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035_3", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many mobile phones are on the table?", + "answer": "One.", + "question_chinese": "此时，桌子上有几个手机？", + "answer_chinese": "一个。", + "evidence": [ + { + "description": "At the beginning, there was a mobile phone in the upper left corner of the table.", + "timestamps": { + "start": "00:00:56", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:01:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Zero.", + "C": "Two.", + "D": "Three phones are lying on the table at this moment as you view it.", + "E": "Four phones are placed on the table right now, spread out in front of you." + }, + "answer_label": "A", + "answer_text": "One.", + "q_id": "egolife_A1_D2_035_3", + "correct_answer": "One.", + "original_options": { + "A": "One.", + "B": "Zero.", + "C": "Two.", + "D": "Three phones are lying on the table at this moment as you view it.", + "E": "Four phones are placed on the table right now, spread out in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D2_035_4", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many mobile phones are on the table?", + "answer": "One.", + "question_chinese": "此时，桌子上有几个手机？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "A boy placed his mobile phone on the table.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + }, + { + "description": "A girl placed her phone on the table.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + }, + { + "description": "There was another mobile phone.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:08:47", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "Two.", + "C": "Three phones are visible on the table at this moment when you look over.", + "D": "Five phones are sitting on the table right now, spread out in front of you.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_D2_035_4", + "correct_answer": "One.", + "original_options": { + "A": "Zero.", + "B": "Two.", + "C": "Three phones are visible on the table at this moment when you look over.", + "D": "Five phones are sitting on the table right now, spread out in front of you.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D2_036-mcq.json b/QAs/egolife_A1_D2_036-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..620a5178c58553932b0bf1aa26a42cde1e199c14 --- /dev/null +++ b/QAs/egolife_A1_D2_036-mcq.json @@ -0,0 +1,857 @@ +[ + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036-mcq_fixed_2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the second floor to the first floor?", + "answer": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "question_chinese": "我从二楼到一楼的路径是什么？", + "answer_chinese": "开始时，我在二楼，下了俩层楼梯之后我到达了一楼。", + "evidence": [ + { + "description": "I went down two flights of stairs", + "timestamps": { + "start": "00:03:36", + "end": "00:03:50" + } + }, + { + "description": "At the beginning, we were all on the second floor.", + "timestamps": { + "start": "00:03:30", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went down one flight of stairs.", + "B": "You went down three flights of stairs.", + "C": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "D": "You started on the second floor and descended one flight of stairs to reach the first floor.", + "E": "You started on the second floor, descended three flights of stairs, and then arrived at the first floor." + }, + "answer_label": "C", + "answer_text": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "q_id": "egolife_A1_D2_036-mcq_fixed_2", + "correct_answer": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "original_options": { + "A": "You went down one flight of stairs.", + "B": "You went down three flights of stairs.", + "C": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "D": "You started on the second floor and descended one flight of stairs to reach the first floor.", + "E": "You started on the second floor, descended three flights of stairs, and then arrived at the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036-mcq_fixed_3", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the second floor to the first floor?", + "answer": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "question_chinese": "我从二楼到一楼的路径是什么？", + "answer_chinese": "开始时，我在二楼，下了俩层楼梯之后我到达了一楼。", + "evidence": [ + { + "description": "I went down two flights of stairs", + "timestamps": { + "start": "00:03:36", + "end": "00:03:50" + } + }, + { + "description": "At the beginning, we were all on the second floor.", + "timestamps": { + "start": "00:03:30", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:06:02", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "B": "You went down one flight of stairs.", + "C": "You went down three flights of stairs.", + "D": "You took an elevator from the second floor straight down to the first floor.", + "E": "You went down one flight to a mezzanine level, crossed over, then descended the remaining steps to the first floor." + }, + "answer_label": "A", + "answer_text": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "q_id": "egolife_A1_D2_036-mcq_fixed_3", + "correct_answer": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "original_options": { + "A": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "B": "You went down one flight of stairs.", + "C": "You went down three flights of stairs.", + "D": "You took an elevator from the second floor straight down to the first floor.", + "E": "You went down one flight to a mezzanine level, crossed over, then descended the remaining steps to the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 20 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036_2_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take from the second floor to where I am now?", + "answer": "At the beginning, you were on the second floor, walking toward the stairwell.", + "question_chinese": "我从二楼到目前位置的路径是什么？", + "answer_chinese": "开始时，我在二楼，在往楼梯口走去", + "evidence": [ + { + "description": "You were walking toward the stairwell.", + "timestamps": { + "start": "00:03:28", + "end": "00:03:32" + } + } + ], + "question_timestamps": "00:03:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went straight past the stairwell.", + "B": "You stayed on the second floor.", + "C": "You started on the second floor, turned away from the stairwell, and continued along the hallway instead.", + "D": "At the beginning, you were on the second floor, walking toward the stairwell.", + "E": "You began upstairs and walked toward a corridor, never descending the stairs before reaching your current spot." + }, + "answer_label": "D", + "answer_text": "At the beginning, you were on the second floor, walking toward the stairwell.", + "q_id": "egolife_A1_D2_036_2_T1", + "correct_answer": "At the beginning, you were on the second floor, walking toward the stairwell.", + "original_options": { + "A": "You went straight past the stairwell.", + "B": "You stayed on the second floor.", + "C": "You started on the second floor, turned away from the stairwell, and continued along the hallway instead.", + "D": "At the beginning, you were on the second floor, walking toward the stairwell.", + "E": "You began upstairs and walked toward a corridor, never descending the stairs before reaching your current spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036_2_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take from the second floor to where I am now?", + "answer": "At the beginning, you were on the second floor. After going down two flights of stairs, you arrived on the first floor.", + "question_chinese": "我从二楼到目前位置的路径是什么？", + "answer_chinese": "开始时，我在二楼，下了俩层楼梯之后我到达了一楼。", + "evidence": [ + { + "description": "At the beginning, we were all on the second floor.", + "timestamps": { + "start": "00:03:28", + "end": "00:03:32" + } + }, + { + "description": "I went down two flights of stairs", + "timestamps": { + "start": "00:03:36", + "end": "00:03:50" + } + } + ], + "question_timestamps": "00:06:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went down one flight of stairs.", + "B": "At the beginning, you were on the second floor. After going down two flights of stairs, you arrived on the first floor.", + "C": "You took the elevator to the first floor.", + "D": "You started on the second floor and descended a single flight, stopping on the next level down.", + "E": "From the second floor, you walked away from the stairwell and ended up on the same floor you started." + }, + "answer_label": "B", + "answer_text": "At the beginning, you were on the second floor. After going down two flights of stairs, you arrived on the first floor.", + "q_id": "egolife_A1_D2_036_2_T2", + "correct_answer": "At the beginning, you were on the second floor. After going down two flights of stairs, you arrived on the first floor.", + "original_options": { + "A": "You went down one flight of stairs.", + "B": "At the beginning, you were on the second floor. After going down two flights of stairs, you arrived on the first floor.", + "C": "You took the elevator to the first floor.", + "D": "You started on the second floor and descended a single flight, stopping on the next level down.", + "E": "From the second floor, you walked away from the stairwell and ended up on the same floor you started." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036-mcq_fixed_6", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many guitars can I see?", + "answer": "Six.", + "question_chinese": "我看见有几把吉他？", + "answer_chinese": "俩把。", + "evidence": [ + { + "description": "When I first stepped into the boys' bedroom, I saw a guitar lying on the floor.", + "timestamps": { + "start": "00:04:15", + "end": "00:04:18" + } + }, + { + "description": "When I fully entered the bedroom, I saw that there was still one on the bed.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:19" + } + } + ], + "question_timestamps": "00:04:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "Six.", + "D": "Only four guitars are visible from your current viewpoint, with the rest out of frame.", + "E": "You can count eight guitars if you include the ones hanging farther back on the wall." + }, + "answer_label": "C", + "answer_text": "Six.", + "q_id": "egolife_A1_D2_036-mcq_fixed_6", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "Six.", + "D": "Only four guitars are visible from your current viewpoint, with the rest out of frame.", + "E": "You can count eight guitars if you include the ones hanging farther back on the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036-mcq_fixed_7", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many guitars can I see?", + "answer": "Three.", + "question_chinese": "我看见有几把吉他？", + "answer_chinese": "一把。", + "evidence": [ + { + "description": "When I first stepped into the boys' bedroom, I saw a guitar lying on the floor.", + "timestamps": { + "start": "00:04:15", + "end": "00:04:18" + } + }, + { + "description": "When I fully entered the bedroom, I saw that there was still one on the bed.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:19" + } + } + ], + "question_timestamps": "00:05:32", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Five guitars are visible in the scene when you look across the wall display carefully.", + "D": "Six guitars can be seen when you scan the area, including those partly behind furniture.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "egolife_A1_D2_036-mcq_fixed_7", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Five guitars are visible in the scene when you look across the wall display carefully.", + "D": "Six guitars can be seen when you scan the area, including those partly behind furniture.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036-mcq_fixed_4", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many types of items do I see on the dining table?", + "answer": "Six.", + "question_chinese": "我看见餐桌上有几种物品？", + "answer_chinese": "三种。", + "evidence": [ + { + "description": "Electronic products, such as tablet computers.", + "timestamps": { + "start": "00:03:53", + "end": "00:03:54" + } + }, + { + "description": "Plants, flowers.", + "timestamps": { + "start": "00:03:53", + "end": "00:03:55" + } + }, + { + "description": "Tableware: such as plates, cups.", + "timestamps": { + "start": "00:03:53", + "end": "00:03:55" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five types.", + "B": "Seven types.", + "C": "Only four different item types are on the dining table, with the rest elsewhere in the room.", + "D": "Six.", + "E": "Eight distinct item types are visible on the dining table when you count every category separately." + }, + "answer_label": "D", + "answer_text": "Six.", + "q_id": "egolife_A1_D2_036-mcq_fixed_4", + "correct_answer": "Six.", + "original_options": { + "A": "Five types.", + "B": "Seven types.", + "C": "Only four different item types are on the dining table, with the rest elsewhere in the room.", + "D": "Six.", + "E": "Eight distinct item types are visible on the dining table when you count every category separately." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036-mcq_fixed_5", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many types of items do I see on the dining table?", + "answer": "Five.", + "question_chinese": "我看见餐桌上有几种物品？", + "answer_chinese": "四种。", + "evidence": [ + { + "description": "Electronic products, such as tablet computers.", + "timestamps": { + "start": "00:03:53", + "end": "00:03:55" + } + }, + { + "description": "Plants, flowers.", + "timestamps": { + "start": "00:03:53", + "end": "00:03:55" + } + }, + { + "description": "Tableware: such as plates, cups.", + "timestamps": { + "start": "00:03:53", + "end": "00:03:55" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four types.", + "B": "Six types.", + "C": "Three different item types are on the dining table, with other objects not on the table.", + "D": "Five.", + "E": "Two distinct item types are visible on the dining table when you group similar items together." + }, + "answer_label": "D", + "answer_text": "Five.", + "q_id": "egolife_A1_D2_036-mcq_fixed_5", + "correct_answer": "Five.", + "original_options": { + "A": "Four types.", + "B": "Six types.", + "C": "Three different item types are on the dining table, with other objects not on the table.", + "D": "Five.", + "E": "Two distinct item types are visible on the dining table when you group similar items together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036-mcq_fixed_8", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment, which is closer to me, the power bank or the plate?", + "answer": "Power bank.", + "question_chinese": "此时，充电宝和餐盘那个离我更近一些？", + "answer_chinese": "充电宝。", + "evidence": [ + { + "description": "The power bank and the plate are on the dining table.", + "timestamps": { + "start": "00:05:48", + "end": "00:05:49" + } + }, + { + "description": "The power bank is to the right of the table, and the plate is to the left of the table.", + "timestamps": { + "start": "00:05:48", + "end": "00:05:49" + } + }, + { + "description": "I am standing to the right of the table.", + "timestamps": { + "start": "00:05:48", + "end": "00:05:49" + } + } + ], + "question_timestamps": "00:05:50", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Power bank.", + "B": "Plate." + }, + "answer_label": "A", + "answer_text": "Power bank.", + "q_id": "egolife_A1_D2_036-mcq_fixed_8", + "correct_answer": "Power bank.", + "original_options": { + "A": "Power bank.", + "B": "Plate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036-mcq_fixed_9", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment, which is closer to me, the power bank or the plate?", + "answer": "Dish.", + "question_chinese": "此时，充电宝和餐盘那个离我更近一些？", + "answer_chinese": "餐盘。", + "evidence": [ + { + "description": "I am sitting by the table, and the power bank is right next to my hand.", + "timestamps": { + "start": "00:06:47", + "end": "00:06:48" + } + }, + { + "description": "The Dish is in front of me, one arm's length away.", + "timestamps": { + "start": "00:06:47", + "end": "00:06:48" + } + } + ], + "question_timestamps": "00:06:55", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Power bank.", + "B": "Dish." + }, + "answer_label": "B", + "answer_text": "Dish.", + "q_id": "egolife_A1_D2_036-mcq_fixed_9", + "correct_answer": "Dish.", + "original_options": { + "A": "Power bank.", + "B": "Dish." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036_1_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many bedrooms have I visited in total?", + "answer": "One, at your back-left.", + "question_chinese": "此时，我一共参观了几个卧室？", + "answer_chinese": "一个，在你的左后侧", + "evidence": [ + { + "description": "At the beginning, we stood at the door of a girl's bedroom and listened to her explain her bedroom.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero, at your back-left.", + "B": "Two, at your back-left.", + "C": "Three total, with it still at your back-left as you stand now.", + "D": "One, at your back-left.", + "E": "Four total, and the most recent one is positioned at your back-left." + }, + "answer_label": "D", + "answer_text": "One, at your back-left.", + "q_id": "egolife_A1_D2_036_1_T1", + "correct_answer": "One, at your back-left.", + "original_options": { + "A": "Zero, at your back-left.", + "B": "Two, at your back-left.", + "C": "Three total, with it still at your back-left as you stand now.", + "D": "One, at your back-left.", + "E": "Four total, and the most recent one is positioned at your back-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036_1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this point, how many bedrooms have I visited in total?", + "answer": "One.", + "question_chinese": "此时，我一共参观了几个卧室？", + "answer_chinese": "一个。", + "evidence": [ + { + "description": "At the beginning, we stood at the door of a girl's bedroom and listened to her explain her bedroom.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four bedrooms in total so far, counting each distinct bedroom you entered.", + "E": "Five bedrooms altogether up to this point, including the most recently visited one." + }, + "answer_label": "B", + "answer_text": "One.", + "q_id": "egolife_A1_D2_036_1", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four bedrooms in total so far, counting each distinct bedroom you entered.", + "E": "Five bedrooms altogether up to this point, including the most recently visited one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036_2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this point, how many bedrooms have I visited in total?", + "answer": "Four.", + "question_chinese": "此时，我一共参观了几个卧室？", + "answer_chinese": "四个。", + "evidence": [ + { + "description": "We arrived at the second girl's bedroom.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:47" + } + }, + { + "description": "We visited the third girl's bedroom.", + "timestamps": { + "start": "00:02:02", + "end": "00:02:03" + } + }, + { + "description": "We visited a boy's bedroom on the first floor.", + "timestamps": { + "start": "00:03:57", + "end": "00:03:58" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Two bedrooms total so far, meaning you have only entered two distinct bedrooms.", + "E": "Six bedrooms in total by now, counting every separate bedroom you visited." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "egolife_A1_D2_036_2", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Two bedrooms total so far, meaning you have only entered two distinct bedrooms.", + "E": "Six bedrooms in total by now, counting every separate bedroom you visited." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D2_036", + "video_length": "00:06:55", + "qid": "egolife_A1_D2_036_1_T4", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many bedrooms have I visited in total?", + "answer": "Four total: three are on the second floor, and the one you just referenced is on the first floor behind you.", + "question_chinese": "此时，我一共参观了几个卧室？", + "answer_chinese": "四个，三个在二楼，刚刚参考的这个在一楼你的身后", + "evidence": [ + { + "description": "At the beginning, we stood at the door of a girl's bedroom and listened to her explain her bedroom.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:25" + } + }, + { + "description": "We arrived at the second girl's bedroom.", + "timestamps": { + "start": "00:00:45", + "end": "00:00:47" + } + }, + { + "description": "We visited the third girl's bedroom.", + "timestamps": { + "start": "00:02:02", + "end": "00:02:03" + } + }, + { + "description": "We visited a boy's bedroom on the first floor.", + "timestamps": { + "start": "00:03:57", + "end": "00:03:58" + } + } + ], + "question_timestamps": "00:04:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three total.", + "B": "Four total: three are on the second floor, and the one you just referenced is on the first floor behind you.", + "C": "Five total.", + "D": "Two total: one is on the second floor, and the one you referenced is on the first floor behind you.", + "E": "Six total: five are on the second floor, and the one you just referenced is on the first floor behind you." + }, + "answer_label": "B", + "answer_text": "Four total: three are on the second floor, and the one you just referenced is on the first floor behind you.", + "q_id": "egolife_A1_D2_036_1_T4", + "correct_answer": "Four total: three are on the second floor, and the one you just referenced is on the first floor behind you.", + "original_options": { + "A": "Three total.", + "B": "Four total: three are on the second floor, and the one you just referenced is on the first floor behind you.", + "C": "Five total.", + "D": "Two total: one is on the second floor, and the one you referenced is on the first floor behind you.", + "E": "Six total: five are on the second floor, and the one you just referenced is on the first floor behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 20, + 21 + ] + } +] diff --git a/QAs/egolife_A1_D3_001-mcq.json b/QAs/egolife_A1_D3_001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3562fc1febaf4bde22293d1b44f436225614e890 --- /dev/null +++ b/QAs/egolife_A1_D3_001-mcq.json @@ -0,0 +1,1062 @@ +[ + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_5_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the bedroom relative to me?", + "answer": "The bedroom is behind you.", + "question_chinese": "卧室在我的什么位置？", + "answer_chinese": "卧室在你的后面。", + "evidence": [ + { + "description": "The bedroom is in behind of you.", + "timestamps": { + "start": "00:06:01", + "end": "00:06:02" + } + } + ], + "question_timestamps": "00:06:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The bedroom is to your right as you face forward, not behind where you are standing now.", + "D": "The bedroom is behind you.", + "E": "The bedroom is above you rather than behind you, so it is not located to the rear." + }, + "answer_label": "D", + "answer_text": "The bedroom is behind you.", + "q_id": "egolife_A1_D3_001_5_T2", + "correct_answer": "The bedroom is behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The bedroom is to your right as you face forward, not behind where you are standing now.", + "D": "The bedroom is behind you.", + "E": "The bedroom is above you rather than behind you, so it is not located to the rear." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_5_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the bedroom relative to me?", + "answer": "The bedroom is in front of you.", + "question_chinese": "卧室在我的什么位置？", + "answer_chinese": "卧室在你的前方。", + "evidence": [ + { + "description": "The bedroom is in front of you.", + "timestamps": { + "start": "00:06:12", + "end": "00:06:20" + } + } + ], + "question_timestamps": "00:06:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your right.", + "C": "The bedroom is to your left relative to where you are facing.", + "D": "The bedroom is in front of you.", + "E": "The bedroom is above you rather than straight ahead in your view." + }, + "answer_label": "D", + "answer_text": "The bedroom is in front of you.", + "q_id": "egolife_A1_D3_001_5_T1", + "correct_answer": "The bedroom is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your right.", + "C": "The bedroom is to your left relative to where you are facing.", + "D": "The bedroom is in front of you.", + "E": "The bedroom is above you rather than straight ahead in your view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001-mcq_fixed_4", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the second floor to the first floor?", + "answer": "You walked from beside the table towards the door of the second floor, went down two flights of stairs in a row, and arrived at the first floor.", + "question_chinese": "我从二楼到一楼的路径是什么？", + "answer_chinese": "我从桌子旁边走向二楼门口，接连下了俩层楼梯，到达一楼。", + "evidence": [ + { + "description": "At the beginning, I was standing next to the table on the second floor.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:03" + } + }, + { + "description": "I went down two flights of stairs.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:13" + } + } + ], + "question_timestamps": "00:02:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went down one flight only.", + "B": "You walked away from the door.", + "C": "You left beside the table, headed into the corridor, and descended a single staircase straight to the first floor.", + "D": "You started near the table, turned the wrong way from the door, then went down two flights and reached the first floor.", + "E": "You walked from beside the table towards the door of the second floor, went down two flights of stairs in a row, and arrived at the first floor." + }, + "answer_label": "E", + "answer_text": "You walked from beside the table towards the door of the second floor, went down two flights of stairs in a row, and arrived at the first floor.", + "q_id": "egolife_A1_D3_001-mcq_fixed_4", + "correct_answer": "You walked from beside the table towards the door of the second floor, went down two flights of stairs in a row, and arrived at the first floor.", + "original_options": { + "A": "You went down one flight only.", + "B": "You walked away from the door.", + "C": "You left beside the table, headed into the corridor, and descended a single staircase straight to the first floor.", + "D": "You started near the table, turned the wrong way from the door, then went down two flights and reached the first floor.", + "E": "You walked from beside the table towards the door of the second floor, went down two flights of stairs in a row, and arrived at the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 22 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001-mcq_fixed_5", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the second floor to the first floor?", + "answer": "You went downstairs to the first floor..", + "question_chinese": "我从二楼到一楼的路径是什么？", + "answer_chinese": "我下楼到达一楼。", + "evidence": [ + { + "description": "At the beginning, I was on the second floor.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:04" + } + }, + { + "description": "I went down two flights of stairs.", + "timestamps": { + "start": "00:02:11", + "end": "00:02:13" + } + } + ], + "question_timestamps": "00:06:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went upstairs to the first floor.", + "B": "You went downstairs to the second floor.", + "C": "You walked along the corridor and took stairs that led down into the basement instead of the first floor.", + "D": "You went downstairs to the first floor..", + "E": "You crossed the hallway and used a different staircase that took you up rather than down." + }, + "answer_label": "D", + "answer_text": "You went downstairs to the first floor..", + "q_id": "egolife_A1_D3_001-mcq_fixed_5", + "correct_answer": "You went downstairs to the first floor..", + "original_options": { + "A": "You went upstairs to the first floor.", + "B": "You went downstairs to the second floor.", + "C": "You walked along the corridor and took stairs that led down into the basement instead of the first floor.", + "D": "You went downstairs to the first floor..", + "E": "You crossed the hallway and used a different staircase that took you up rather than down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001-mcq_fixed_6", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement path of the pen near the whiteboard on the first floor that I see?", + "answer": "YOU picked up a pen from next to the whiteboard on the first floor and climbed two flights of stairs.", + "question_chinese": "我看见在一楼白板旁边的签字笔的移动路径是什么？", + "answer_chinese": "我从一楼白板旁边拿起签字笔，上了俩层楼梯。", + "evidence": [ + { + "description": "At the beginning, I was by the whiteboard on the first floor, with a pen on the whiteboard.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:16" + } + }, + { + "description": "I am on the stairs, holding a pen in my hand.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:23" + } + } + ], + "question_timestamps": "00:02:32", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "YOU picked up a pen from next to the whiteboard on the first floor and climbed two flights of stairs.", + "B": "You carried the pen across the room.", + "C": "You climbed one flight with the pen.", + "D": "You picked up the pen by the whiteboard and only climbed one flight of stairs before stopping.", + "E": "You grabbed the pen next to the whiteboard on the first floor, then went up three flights of stairs." + }, + "answer_label": "A", + "answer_text": "YOU picked up a pen from next to the whiteboard on the first floor and climbed two flights of stairs.", + "q_id": "egolife_A1_D3_001-mcq_fixed_6", + "correct_answer": "YOU picked up a pen from next to the whiteboard on the first floor and climbed two flights of stairs.", + "original_options": { + "A": "YOU picked up a pen from next to the whiteboard on the first floor and climbed two flights of stairs.", + "B": "You carried the pen across the room.", + "C": "You climbed one flight with the pen.", + "D": "You picked up the pen by the whiteboard and only climbed one flight of stairs before stopping.", + "E": "You grabbed the pen next to the whiteboard on the first floor, then went up three flights of stairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001-mcq_fixed_7", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement path of the pen near the whiteboard on the first floor that I see?", + "answer": "You picked up a pen from next to the whiteboard on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "question_chinese": "我看见在一楼白板旁边的签字笔的移动路径是什么？", + "answer_chinese": "我从一楼白板旁边拿起签字笔，上了俩层楼梯，到达二楼。", + "evidence": [ + { + "description": "At the beginning, I was by the whiteboard on the first floor, with a pen on the whiteboard.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:15" + } + }, + { + "description": "I am on the stairs, holding a pen in my hand.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:23" + } + }, + { + "description": "I took a pen and went up two flights of stairs.", + "timestamps": { + "start": "00:02:24", + "end": "00:02:43" + } + } + ], + "question_timestamps": "00:08:12", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You climbed one flight of stairs.", + "B": "You climbed three flights of stairs.", + "C": "You picked up a pen from next to the whiteboard on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "D": "You picked up the pen by the whiteboard and climbed one flight of stairs before reaching the second floor.", + "E": "You grabbed the pen next to the whiteboard, climbed three flights of stairs, and still ended up on the second floor." + }, + "answer_label": "C", + "answer_text": "You picked up a pen from next to the whiteboard on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "q_id": "egolife_A1_D3_001-mcq_fixed_7", + "correct_answer": "You picked up a pen from next to the whiteboard on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "original_options": { + "A": "You climbed one flight of stairs.", + "B": "You climbed three flights of stairs.", + "C": "You picked up a pen from next to the whiteboard on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "D": "You picked up the pen by the whiteboard and climbed one flight of stairs before reaching the second floor.", + "E": "You grabbed the pen next to the whiteboard, climbed three flights of stairs, and still ended up on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_4_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the reflector relative to me?", + "answer": "The reflector panel is behind you.", + "question_chinese": "打光板在我的什么位置？", + "answer_chinese": "打光板在你的后方。", + "evidence": [ + { + "description": "The reflector is in behind of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The reflector panel is in front of you.", + "B": "The reflector panel is to your left.", + "C": "The reflector panel is positioned to your right as you face forward.", + "D": "The reflector panel is above you, hanging higher than your eye level.", + "E": "The reflector panel is behind you." + }, + "answer_label": "E", + "answer_text": "The reflector panel is behind you.", + "q_id": "egolife_A1_D3_001_4_T1", + "correct_answer": "The reflector panel is behind you.", + "original_options": { + "A": "The reflector panel is in front of you.", + "B": "The reflector panel is to your left.", + "C": "The reflector panel is positioned to your right as you face forward.", + "D": "The reflector panel is above you, hanging higher than your eye level.", + "E": "The reflector panel is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_4_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the reflector relative to me?", + "answer": "The reflector panel is in front of you.", + "question_chinese": "打光板在我的什么位置？", + "answer_chinese": "打光板在你的前方。", + "evidence": [ + { + "description": "The reflector is in front of you.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The reflector panel is behind you.", + "B": "The reflector panel is to your right.", + "C": "The reflector panel is positioned to your left as you face forward.", + "D": "The reflector panel is above you, hanging higher than your eye level.", + "E": "The reflector panel is in front of you." + }, + "answer_label": "E", + "answer_text": "The reflector panel is in front of you.", + "q_id": "egolife_A1_D3_001_4_T2", + "correct_answer": "The reflector panel is in front of you.", + "original_options": { + "A": "The reflector panel is behind you.", + "B": "The reflector panel is to your right.", + "C": "The reflector panel is positioned to your left as you face forward.", + "D": "The reflector panel is above you, hanging higher than your eye level.", + "E": "The reflector panel is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_9", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the light board positioned in relation to me?", + "answer": "The light board is behind me.", + "question_chinese": "打光板在我的什么位置？", + "answer_chinese": "打光板在我的后方。", + "evidence": [ + { + "description": "When I walked around the table, the light panel was in front of me to the right.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + }, + { + "description": "As I walked towards the door.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The light board is in front of you.", + "B": "The light board is behind me.", + "C": "The light board is to your left.", + "D": "The light board is positioned to your right as you face forward.", + "E": "The light board is above you, hanging higher than your eye level." + }, + "answer_label": "B", + "answer_text": "The light board is behind me.", + "q_id": "egolife_A1_D3_001_9", + "correct_answer": "The light board is behind me.", + "original_options": { + "A": "The light board is in front of you.", + "B": "The light board is behind me.", + "C": "The light board is to your left.", + "D": "The light board is positioned to your right as you face forward.", + "E": "The light board is above you, hanging higher than your eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_10", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the light board located in my setup?", + "answer": "The lighting board is behind me on the left.", + "question_chinese": "打光板在我的什么位置？", + "answer_chinese": "打光板在我的左后方。", + "evidence": [ + { + "description": "When I first walked in, the lighting board was directly in front of me.", + "timestamps": { + "start": "00:02:44", + "end": "00:02:45" + } + }, + { + "description": "I am standing near the whiteboard.", + "timestamps": { + "start": "00:06:01", + "end": "00:06:02" + } + } + ], + "question_timestamps": "00:10:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind you on the right.", + "B": "In front on the left.", + "C": "The lighting board sits in front of you on the right side of your setup.", + "D": "The lighting board is above you on the left, not behind you in the arrangement.", + "E": "The lighting board is behind me on the left." + }, + "answer_label": "E", + "answer_text": "The lighting board is behind me on the left.", + "q_id": "egolife_A1_D3_001_10", + "correct_answer": "The lighting board is behind me on the left.", + "original_options": { + "A": "Behind you on the right.", + "B": "In front on the left.", + "C": "The lighting board sits in front of you on the right side of your setup.", + "D": "The lighting board is above you on the left, not behind you in the arrangement.", + "E": "The lighting board is behind me on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many disposable paper cups are on the table?", + "answer": "Two.", + "question_chinese": "此时，一共有几个一次性纸杯放在了桌子上？", + "answer_chinese": "俩个。", + "evidence": [ + { + "description": "I picked up a disposable paper cup from the table.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + }, + { + "description": "I placed two paper cups in a row on the table.", + "timestamps": { + "start": "00:00:31", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Four disposable paper cups are on the table right now, clearly visible together near the center.", + "E": "Five disposable paper cups are on the table at this moment, arranged close to each other on the surface." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "egolife_A1_D3_001_1", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "Four disposable paper cups are on the table right now, clearly visible together near the center.", + "E": "Five disposable paper cups are on the table at this moment, arranged close to each other on the surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many disposable paper cups are on the table?", + "answer": "Sixteen.", + "question_chinese": "此时，一共有几个一次性纸杯放在了桌子上？", + "answer_chinese": "十六个。", + "evidence": [ + { + "description": "I saw a boy who took out fourteen cups.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:45" + } + }, + { + "description": "The cups are divided into three piles, with five in each group.", + "timestamps": { + "start": "00:01:45", + "end": "00:01:46" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Sixteen.", + "C": "Eight.", + "D": "Seven disposable paper cups are on the table at this moment, grouped near the front edge together.", + "E": "Five disposable paper cups are on the table right now, spaced apart across the tabletop in view." + }, + "answer_label": "B", + "answer_text": "Sixteen.", + "q_id": "egolife_A1_D3_001_2", + "correct_answer": "Sixteen.", + "original_options": { + "A": "Four.", + "B": "Sixteen.", + "C": "Eight.", + "D": "Seven disposable paper cups are on the table at this moment, grouped near the front edge together.", + "E": "Five disposable paper cups are on the table right now, spaced apart across the tabletop in view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_2_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many bottles of water did I put on the table?", + "answer": "At this moment, you put 3 bottles of water on the table.", + "question_chinese": "此时，我放几瓶水在桌子上？", + "answer_chinese": "此时，你放3瓶水在桌子上。", + "evidence": [ + { + "description": "I took three bottles of water out of the bag one after another and placed them on the table.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You put 2 bottles on the table.", + "B": "You put 4 bottles on the table.", + "C": "At this moment, you put 3 bottles of water on the table.", + "D": "At this moment, you put 1 bottle of water on the table, not several.", + "E": "At this moment, you put 5 bottles of water on the table, lined up together." + }, + "answer_label": "C", + "answer_text": "At this moment, you put 3 bottles of water on the table.", + "q_id": "egolife_A1_D3_001_2_T1", + "correct_answer": "At this moment, you put 3 bottles of water on the table.", + "original_options": { + "A": "You put 2 bottles on the table.", + "B": "You put 4 bottles on the table.", + "C": "At this moment, you put 3 bottles of water on the table.", + "D": "At this moment, you put 1 bottle of water on the table, not several.", + "E": "At this moment, you put 5 bottles of water on the table, lined up together." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_2_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many bottles of water did I put on the table?", + "answer": "At this moment, you put 10 bottles of water on the table.", + "question_chinese": "此时，我放几瓶水在桌子上？", + "answer_chinese": "此时，你放10瓶水在桌子上？", + "evidence": [ + { + "description": "I took the bottles of water out of the bag one after another and placed them on the table.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You put 9 bottles on the table.", + "B": "You put 11 bottles on the table.", + "C": "At this moment, you put 8 bottles of water on the table, fewer than ten.", + "D": "At this moment, you put 10 bottles of water on the table.", + "E": "At this moment, you put 12 bottles of water on the table, more than ten total." + }, + "answer_label": "D", + "answer_text": "At this moment, you put 10 bottles of water on the table.", + "q_id": "egolife_A1_D3_001_2_T2", + "correct_answer": "At this moment, you put 10 bottles of water on the table.", + "original_options": { + "A": "You put 9 bottles on the table.", + "B": "You put 11 bottles on the table.", + "C": "At this moment, you put 8 bottles of water on the table, fewer than ten.", + "D": "At this moment, you put 10 bottles of water on the table.", + "E": "At this moment, you put 12 bottles of water on the table, more than ten total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_3_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, can I touch the cup?", + "answer": "At this moment, you can't reach the cup; the cup is in front of you.", + "question_chinese": "此时，我能接触到杯子吗？", + "answer_chinese": "此时，你不能接触到杯子，杯子在你前面。", + "evidence": [ + { + "description": "You can not touch the cup.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:01:29", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Yes, you can touch the cup.", + "B": "You can reach it; it’s behind you.", + "C": "At this moment, you can reach the cup easily because it is in front of you.", + "D": "At this moment, you can reach the cup since it is positioned to your left side.", + "E": "At this moment, you can't reach the cup; the cup is in front of you." + }, + "answer_label": "E", + "answer_text": "At this moment, you can't reach the cup; the cup is in front of you.", + "q_id": "egolife_A1_D3_001_3_T2", + "correct_answer": "At this moment, you can't reach the cup; the cup is in front of you.", + "original_options": { + "A": "Yes, you can touch the cup.", + "B": "You can reach it; it’s behind you.", + "C": "At this moment, you can reach the cup easily because it is in front of you.", + "D": "At this moment, you can reach the cup since it is positioned to your left side.", + "E": "At this moment, you can't reach the cup; the cup is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_3_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, can I touch the cup?", + "answer": "At this moment, you can reach the cup; the cup is in front of you.", + "question_chinese": "此时，我能接触到杯子吗？", + "answer_chinese": "此时，你能接触到杯子，杯子在你前面。", + "evidence": [ + { + "description": "You can touch the cup.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:44" + } + } + ], + "question_timestamps": "00:01:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "No, you can’t touch the cup.", + "B": "You can reach it; it’s behind you.", + "C": "At this moment, you can reach the cup, but it is positioned to your left rather than ahead.", + "D": "At this moment, you can reach the cup; the cup is in front of you.", + "E": "At this moment, you can reach the cup, but it is to your right instead of in front." + }, + "answer_label": "D", + "answer_text": "At this moment, you can reach the cup; the cup is in front of you.", + "q_id": "egolife_A1_D3_001_3_T1", + "correct_answer": "At this moment, you can reach the cup; the cup is in front of you.", + "original_options": { + "A": "No, you can’t touch the cup.", + "B": "You can reach it; it’s behind you.", + "C": "At this moment, you can reach the cup, but it is positioned to your left rather than ahead.", + "D": "At this moment, you can reach the cup; the cup is in front of you.", + "E": "At this moment, you can reach the cup, but it is to your right instead of in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_3", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many bottles of water are placed on the table?", + "answer": "Three.", + "question_chinese": "此时，有几瓶水被放在了桌子上？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "I saw a bag of drinks on the chair next to the table.", + "timestamps": { + "start": "00:00:49", + "end": "00:00:50" + } + }, + { + "description": "I took three bottles of water out of the bag one after another and placed them on the table.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Six bottles of water are placed on the table right now, not just three.", + "E": "Seven bottles of water are currently placed on the table at this moment." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_001_3", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Six bottles of water are placed on the table right now, not just three.", + "E": "Seven bottles of water are currently placed on the table at this moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_001", + "video_length": "00:11:32", + "qid": "egolife_A1_D3_001_4", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many bottles of water have been placed on the table?", + "answer": "Ten bottles.", + "question_chinese": "此时，有几瓶水被放在了桌子上？", + "answer_chinese": "十瓶。", + "evidence": [ + { + "description": "I took out seven bottles of water in a row.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:01:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Nine bottles.", + "B": "Eleven bottles.", + "C": "Ten bottles.", + "D": "Eight bottles of water have been placed on the table so far.", + "E": "Three bottles of water have been placed on the table at this moment." + }, + "answer_label": "C", + "answer_text": "Ten bottles.", + "q_id": "egolife_A1_D3_001_4", + "correct_answer": "Ten bottles.", + "original_options": { + "A": "Nine bottles.", + "B": "Eleven bottles.", + "C": "Ten bottles.", + "D": "Eight bottles of water have been placed on the table so far.", + "E": "Three bottles of water have been placed on the table at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D3_002-mcq.json b/QAs/egolife_A1_D3_002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..debc36df8c1e37f9cf23120c870b199c2990246f --- /dev/null +++ b/QAs/egolife_A1_D3_002-mcq.json @@ -0,0 +1,538 @@ +[ + { + "video_uid": "egolife_A1_D3_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_002-mcq_fixed_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the white trash can in the living room on the first floor?", + "answer": "The white trash can on the first floor is directly in front of the living room door.", + "question_chinese": "一楼客厅的白色垃圾桶现在在哪里？", + "answer_chinese": "一楼的白色垃圾桶在客厅门的正前方。", + "evidence": [ + { + "description": "The white trash can on the first floor is directly in front of the living room door.", + "timestamps": { + "start": "00:03:59", + "end": "00:04:00" + } + }, + { + "description": "The white trash can on the first floor is directly in front of the living room door.", + "timestamps": { + "start": "00:05:45", + "end": "00:05:46" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It’s behind the living room door.", + "B": "It’s near the living room window.", + "C": "The white trash can on the first floor is just behind the television stand near the corner.", + "D": "The white trash can on the first floor is directly in front of the living room door.", + "E": "The white trash can on the first floor sits beside the coffee table near the couch." + }, + "answer_label": "D", + "answer_text": "The white trash can on the first floor is directly in front of the living room door.", + "q_id": "egolife_A1_D3_002-mcq_fixed_5", + "correct_answer": "The white trash can on the first floor is directly in front of the living room door.", + "original_options": { + "A": "It’s behind the living room door.", + "B": "It’s near the living room window.", + "C": "The white trash can on the first floor is just behind the television stand near the corner.", + "D": "The white trash can on the first floor is directly in front of the living room door.", + "E": "The white trash can on the first floor sits beside the coffee table near the couch." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_002-mcq_fixed_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the white trash can in the living room on the first floor?", + "answer": "The white trash can on the first floor is next to the dining table.", + "question_chinese": "一楼客厅的白色垃圾桶现在在哪里？", + "answer_chinese": "一楼的白色垃圾桶在餐桌旁。", + "evidence": [ + { + "description": "At this moment, the white trash can on the first floor can be seen next to the dining table.", + "timestamps": { + "start": "00:03:59", + "end": "00:04:00" + } + }, + { + "description": "At this moment, the white trash can on the first floor can be seen next to the dining table.", + "timestamps": { + "start": "00:05:45", + "end": "00:05:46" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The white trash can on the first floor is next to the dining table.", + "B": "It’s by the living room window.", + "C": "It’s near the living room couch.", + "D": "The white trash can on the first floor is tucked in the corner next to the TV stand.", + "E": "It’s right beside the small coffee table near the couch where magazines are kept." + }, + "answer_label": "A", + "answer_text": "The white trash can on the first floor is next to the dining table.", + "q_id": "egolife_A1_D3_002-mcq_fixed_4", + "correct_answer": "The white trash can on the first floor is next to the dining table.", + "original_options": { + "A": "The white trash can on the first floor is next to the dining table.", + "B": "It’s by the living room window.", + "C": "It’s near the living room couch.", + "D": "The white trash can on the first floor is tucked in the corner next to the TV stand.", + "E": "It’s right beside the small coffee table near the couch where magazines are kept." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_002-mcq_fixed_6", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many computers have you seen in the second-floor living room?", + "answer": "Three.", + "question_chinese": "二楼客厅里到目前为止你看到了几个电脑？", + "answer_chinese": "到目前为止你看到了一个电脑。", + "evidence": [ + { + "description": "At this moment, a computer can be seen on the floor at the kitchen entrance.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "Five computers have been seen in the second-floor living room so far today.", + "E": "Up to now, you have seen only one computer in the second-floor living room." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_002-mcq_fixed_6", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "Five computers have been seen in the second-floor living room so far today.", + "E": "Up to now, you have seen only one computer in the second-floor living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_002-mcq_fixed_7", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many computers have you seen in the second-floor living room?", + "answer": "One.", + "question_chinese": "二楼客厅里到目前为止你看到了几个电脑？", + "answer_chinese": "到目前为止你看到了两个电脑。", + "evidence": [ + { + "description": "At this moment, a computer can be seen on the floor at the kitchen entrance.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + }, + { + "description": "At this moment, another computer can be seen on the floor beside the sofa.", + "timestamps": { + "start": "00:08:15", + "end": "00:08:16" + } + } + ], + "question_timestamps": "00:10:47", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Zero.", + "D": "Three computers have appeared in the second-floor living room so far today.", + "E": "You have already seen four computers in the second-floor living room up to this point." + }, + "answer_label": "B", + "answer_text": "One.", + "q_id": "egolife_A1_D3_002-mcq_fixed_7", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Zero.", + "D": "Three computers have appeared in the second-floor living room so far today.", + "E": "You have already seen four computers in the second-floor living room up to this point." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_002-mcq_fixed_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I scan the camera in the living room on the first floor just now?", + "answer": "Three.", + "question_chinese": "我刚刚在一楼客厅一共扫了几次摄像头？", + "answer_chinese": "你刚刚在一楼客厅一共扫了四次摄像头.", + "evidence": [ + { + "description": "At this moment, you scanned two cameras.", + "timestamps": { + "start": "00:05:11", + "end": "00:05:22" + } + }, + { + "description": "At this moment, you scanned four cameras.", + "timestamps": { + "start": "00:05:27", + "end": "00:05:32" + } + } + ], + "question_timestamps": "00:05:33", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You scanned the camera five times in the first-floor living room just now.", + "E": "In the first-floor living room, you did a single long scan that counts as one time." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_002-mcq_fixed_1", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You scanned the camera five times in the first-floor living room just now.", + "E": "In the first-floor living room, you did a single long scan that counts as one time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_002-mcq_fixed_0", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I scan the camera in the living room on the first floor just now?", + "answer": "One.", + "question_chinese": "我刚刚在一楼客厅一共扫了几次摄像头？", + "answer_chinese": "你刚刚一共扫了八个摄像头。", + "evidence": [ + { + "description": "At this moment, you scanned two cameras.", + "timestamps": { + "start": "00:05:11", + "end": "00:05:22" + } + }, + { + "description": "At this moment, you scanned four cameras.", + "timestamps": { + "start": "00:05:27", + "end": "00:05:32" + } + }, + { + "description": "So far, you have scanned a total of eight cameras.", + "timestamps": { + "start": "00:05:34", + "end": "00:05:53" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Zero.", + "C": "You scanned the camera three times in the first-floor living room just now.", + "D": "In the first-floor living room, you scanned the camera four separate times a moment ago.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_D3_002-mcq_fixed_0", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Zero.", + "C": "You scanned the camera three times in the first-floor living room just now.", + "D": "In the first-floor living room, you scanned the camera four separate times a moment ago.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_002-mcq_fixed_8", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there water in the mineral water bottle on my left?", + "answer": "Yes, there is water in the mineral water bottle on your left.", + "question_chinese": "此时我左手拿的矿泉水里有水吗？", + "answer_chinese": "有，你左手拿的矿泉水有水。", + "evidence": [ + { + "description": "At this moment, you can see water in the mineral water bottle in your left hand.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:41", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, there is water in the mineral water bottle on your left.", + "B": "No, there is no water in the mineral water bottle on your left." + }, + "answer_label": "A", + "answer_text": "Yes, there is water in the mineral water bottle on your left.", + "q_id": "egolife_A1_D3_002-mcq_fixed_8", + "correct_answer": "Yes, there is water in the mineral water bottle on your left.", + "original_options": { + "A": "Yes, there is water in the mineral water bottle on your left.", + "B": "No, there is no water in the mineral water bottle on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_002-mcq_fixed_9", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there water in the mineral water bottle on my left?", + "answer": "No, there is no water in the mineral water bottle on your left.", + "question_chinese": "此时我左手拿的矿泉水里有水吗？", + "answer_chinese": "没有，你左手拿的矿泉水没有水。", + "evidence": [ + { + "description": "At this moment, you are pouring the water from the mineral water bottle into the sink.", + "timestamps": { + "start": "00:01:49", + "end": "00:01:51" + } + } + ], + "question_timestamps": "00:02:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, there is water in the mineral water bottle on your left.", + "B": "No, there is no water in the mineral water bottle on your left." + }, + "answer_label": "B", + "answer_text": "No, there is no water in the mineral water bottle on your left.", + "q_id": "egolife_A1_D3_002-mcq_fixed_9", + "correct_answer": "No, there is no water in the mineral water bottle on your left.", + "original_options": { + "A": "Yes, there is water in the mineral water bottle on your left.", + "B": "No, there is no water in the mineral water bottle on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_002-mcq_fixed_2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there a doll on the sofa on the second floor?", + "answer": "Yes, there is a doll on the sofa on the second floor, and it is on the right side of the sofa.", + "question_chinese": "现在二楼的沙发上有玩偶吗？", + "answer_chinese": "是的，二楼的沙发上有玩偶，它在沙发的右边。", + "evidence": [ + { + "description": "At this moment, a lion doll can be seen on the sofa on the second floor.", + "timestamps": { + "start": "00:07:12", + "end": "00:07:13" + } + } + ], + "question_timestamps": "00:08:54", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, there is a doll on the sofa on the second floor, and it is on the right side of the sofa.", + "B": "No, there is no doll on the sofa on the second floor." + }, + "answer_label": "A", + "answer_text": "Yes, there is a doll on the sofa on the second floor, and it is on the right side of the sofa.", + "q_id": "egolife_A1_D3_002-mcq_fixed_2", + "correct_answer": "Yes, there is a doll on the sofa on the second floor, and it is on the right side of the sofa.", + "original_options": { + "A": "Yes, there is a doll on the sofa on the second floor, and it is on the right side of the sofa.", + "B": "No, there is no doll on the sofa on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_002-mcq_fixed_3", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there a doll on the sofa on the second floor?", + "answer": "Yes, there is a doll on the sofa on the second floor, and it is above the white little flower doll.", + "question_chinese": "现在二楼的沙发上有玩偶吗？", + "answer_chinese": "是的，二楼的沙发上有玩偶，它在白色小花玩偶上面。", + "evidence": [ + { + "description": "At this moment, a lion doll can be seen on the sofa on the second floor.", + "timestamps": { + "start": "00:07:12", + "end": "00:07:13" + } + } + ], + "question_timestamps": "00:10:09", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, there is no doll on the sofa on the second floor, and nothing is above the white little flower doll.", + "B": "Yes, there is a doll on the sofa on the second floor, and it is above the white little flower doll." + }, + "answer_label": "B", + "answer_text": "Yes, there is a doll on the sofa on the second floor, and it is above the white little flower doll.", + "q_id": "egolife_A1_D3_002-mcq_fixed_3", + "correct_answer": "Yes, there is a doll on the sofa on the second floor, and it is above the white little flower doll.", + "original_options": { + "A": "No, there is no doll on the sofa on the second floor, and nothing is above the white little flower doll.", + "B": "Yes, there is a doll on the sofa on the second floor, and it is above the white little flower doll." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 21 + ] + } +] diff --git a/QAs/egolife_A1_D3_003-mcq.json b/QAs/egolife_A1_D3_003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d5379618e616fd444f28b1890904c88379579c98 --- /dev/null +++ b/QAs/egolife_A1_D3_003-mcq.json @@ -0,0 +1,559 @@ +[ + { + "video_uid": "egolife_A1_D3_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_003-mcq_fixed_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the refrigerator on the second floor located?", + "answer": "On the second floor, the refrigerator is against the wall on the right side of the kitchen.", + "question_chinese": "二楼的冰箱现在在哪个位置？", + "answer_chinese": "二楼的冰箱在厨房右边的墙边。", + "evidence": [ + { + "description": "On the second floor, the refrigerator is against the wall on the right side of the kitchen.", + "timestamps": { + "start": "00:08:13", + "end": "00:08:14" + } + }, + { + "description": "On the second floor, the refrigerator is against the wall on the right side of the kitchen.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:14" + } + } + ], + "question_timestamps": "00:10:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the left wall of the kitchen.", + "B": "Centered along the kitchen wall.", + "C": "It’s against the kitchen wall on the left side, opposite where you would normally stand to cook.", + "D": "On the second floor, the refrigerator is against the wall on the right side of the kitchen.", + "E": "The refrigerator sits against the center wall of the kitchen, not specifically on the right side." + }, + "answer_label": "D", + "answer_text": "On the second floor, the refrigerator is against the wall on the right side of the kitchen.", + "q_id": "egolife_A1_D3_003-mcq_fixed_7", + "correct_answer": "On the second floor, the refrigerator is against the wall on the right side of the kitchen.", + "original_options": { + "A": "On the left wall of the kitchen.", + "B": "Centered along the kitchen wall.", + "C": "It’s against the kitchen wall on the left side, opposite where you would normally stand to cook.", + "D": "On the second floor, the refrigerator is against the wall on the right side of the kitchen.", + "E": "The refrigerator sits against the center wall of the kitchen, not specifically on the right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_003-mcq_fixed_6", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the refrigerator on the second floor located?", + "answer": "The refrigerator on the second floor is behind the door at the stairway exit.", + "question_chinese": "二楼的冰箱现在在哪个位置？", + "answer_chinese": "二楼的冰箱在楼梯出口的门后。", + "evidence": [ + { + "description": "At this moment, you can see the refrigerator on the second floor behind the door at the stairway exit.", + "timestamps": { + "start": "00:08:13", + "end": "00:08:14" + } + }, + { + "description": "At this moment, you can see the refrigerator on the second floor behind the door at the stairway exit.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:14" + } + } + ], + "question_timestamps": "00:11:49", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Beside the stairway exit door.", + "B": "The refrigerator on the second floor is behind the door at the stairway exit.", + "C": "Across from the stairway exit.", + "D": "It’s positioned behind a curtain near the stairway exit rather than being hidden directly behind the door.", + "E": "The refrigerator is tucked behind a bookshelf at the stairway exit, so you see it only after walking around it." + }, + "answer_label": "B", + "answer_text": "The refrigerator on the second floor is behind the door at the stairway exit.", + "q_id": "egolife_A1_D3_003-mcq_fixed_6", + "correct_answer": "The refrigerator on the second floor is behind the door at the stairway exit.", + "original_options": { + "A": "Beside the stairway exit door.", + "B": "The refrigerator on the second floor is behind the door at the stairway exit.", + "C": "Across from the stairway exit.", + "D": "It’s positioned behind a curtain near the stairway exit rather than being hidden directly behind the door.", + "E": "The refrigerator is tucked behind a bookshelf at the stairway exit, so you see it only after walking around it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_003-mcq_fixed_11", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there mineral water on the sofa on the second floor?", + "answer": "Yes, there is mineral water on the sofa; it is to the right of the lion plush toy.", + "question_chinese": "二楼的沙发上现在有矿泉水吗？", + "answer_chinese": "有，沙发上有矿泉水，它在小狮子玩偶的右边。", + "evidence": [ + { + "description": "At this moment, you can see that there is mineral water on the sofa.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + }, + { + "description": "At this moment, you can see that there is mineral water on the sofa.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:06:02", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, there is no mineral water on the sofa on the second floor.", + "B": "Yes, there is mineral water on the sofa; it is to the right of the lion plush toy." + }, + "answer_label": "B", + "answer_text": "Yes, there is mineral water on the sofa; it is to the right of the lion plush toy.", + "q_id": "egolife_A1_D3_003-mcq_fixed_11", + "correct_answer": "Yes, there is mineral water on the sofa; it is to the right of the lion plush toy.", + "original_options": { + "A": "No, there is no mineral water on the sofa on the second floor.", + "B": "Yes, there is mineral water on the sofa; it is to the right of the lion plush toy." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_003-mcq_fixed_10", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there mineral water on the sofa on the second floor?", + "answer": "Yes, there is mineral water on the sofa; it is on the far right of the sofa.", + "question_chinese": "二楼的沙发上现在有矿泉水吗？", + "answer_chinese": "有，沙发上有矿泉水，它在沙发最右侧。", + "evidence": [ + { + "description": "At this moment, you can see that there is mineral water on the sofa.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + }, + { + "description": "At this moment, you can see that there is mineral water on the sofa.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:08:42", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, there is mineral water on the sofa; it is on the far right of the sofa.", + "B": "No, there is no mineral water on the sofa on the second floor." + }, + "answer_label": "A", + "answer_text": "Yes, there is mineral water on the sofa; it is on the far right of the sofa.", + "q_id": "egolife_A1_D3_003-mcq_fixed_10", + "correct_answer": "Yes, there is mineral water on the sofa; it is on the far right of the sofa.", + "original_options": { + "A": "Yes, there is mineral water on the sofa; it is on the far right of the sofa.", + "B": "No, there is no mineral water on the sofa on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_003-mcq_fixed_0", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the current relative position of the pen (which was originally on the sofa) to the dining table?", + "answer": "The pen is still on the sofa, located two meters to the front-left of the dining table.", + "question_chinese": "原本放在沙发上的笔现在和餐桌的相对位置是什么？", + "answer_chinese": "笔现在还在沙发上，位于餐桌的左前方两米处。", + "evidence": [ + { + "description": "At this moment, you can see the pen on the sofa, and the sofa is in the front-left of the dining table.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:49" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Two meters behind the dining table.", + "B": "The pen is still on the sofa, located two meters to the front-left of the dining table.", + "C": "Two meters to the front-right.", + "D": "The pen is still on the sofa, but it’s about two meters to the back-right of the dining table.", + "E": "The pen has been moved off the sofa and is now roughly two meters directly behind the dining table." + }, + "answer_label": "B", + "answer_text": "The pen is still on the sofa, located two meters to the front-left of the dining table.", + "q_id": "egolife_A1_D3_003-mcq_fixed_0", + "correct_answer": "The pen is still on the sofa, located two meters to the front-left of the dining table.", + "original_options": { + "A": "Two meters behind the dining table.", + "B": "The pen is still on the sofa, located two meters to the front-left of the dining table.", + "C": "Two meters to the front-right.", + "D": "The pen is still on the sofa, but it’s about two meters to the back-right of the dining table.", + "E": "The pen has been moved off the sofa and is now roughly two meters directly behind the dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_003-mcq_fixed_1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the current relative position of the pen (which was originally on the sofa) to the dining table?", + "answer": "The pen is now on the dining table, positioned at the far left.", + "question_chinese": "原本放在沙发上的笔现在和餐桌的相对位置是什么？", + "answer_chinese": "笔现在在餐桌上面，且位于最左侧。", + "evidence": [ + { + "description": "At this moment, you can see the pen on the sofa, and the sofa is in the front-left of the dining table.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:49" + } + }, + { + "description": "At this moment, you placed the pen to the left of the dining table.", + "timestamps": { + "start": "00:02:03", + "end": "00:02:04" + } + } + ], + "question_timestamps": "00:03:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the dining table’s far right.", + "B": "Near the center of the table.", + "C": "The pen is on the dining table, placed slightly right of center rather than at the far left edge.", + "D": "You’ll find the pen on the dining table near the middle, closer to where the coasters and bowl usually sit.", + "E": "The pen is now on the dining table, positioned at the far left." + }, + "answer_label": "E", + "answer_text": "The pen is now on the dining table, positioned at the far left.", + "q_id": "egolife_A1_D3_003-mcq_fixed_1", + "correct_answer": "The pen is now on the dining table, positioned at the far left.", + "original_options": { + "A": "On the dining table’s far right.", + "B": "Near the center of the table.", + "C": "The pen is on the dining table, placed slightly right of center rather than at the far left edge.", + "D": "You’ll find the pen on the dining table near the middle, closer to where the coasters and bowl usually sit.", + "E": "The pen is now on the dining table, positioned at the far left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D3_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_003_3_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I gone downstairs at any point so far?", + "answer": "No—you haven’t gone downstairs so far.", + "question_chinese": "我到目前为止下楼了吗？", + "answer_chinese": "没有，你到目前为止没有下楼。", + "evidence": [ + { + "description": "At this moment, you turned back at the staircase and went to the second floor without going downstairs.", + "timestamps": { + "start": "00:02:07", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you haven’t gone downstairs so far.", + "B": "Yes—you’ve gone downstairs at least once so far." + }, + "answer_label": "A", + "answer_text": "No—you haven’t gone downstairs so far.", + "q_id": "egolife_A1_D3_003_3_T1", + "correct_answer": "No—you haven’t gone downstairs so far.", + "original_options": { + "A": "No—you haven’t gone downstairs so far.", + "B": "Yes—you’ve gone downstairs at least once so far." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D3_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_003_3_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I gone downstairs at any point so far?", + "answer": "No—you haven’t gone downstairs so far. The most recent time you went down the stairs, you turned back.", + "question_chinese": "我到目前为止下楼了吗？", + "answer_chinese": "没有，你到目前为止没有下楼，你最近的一次下楼梯但又折回来了。", + "evidence": [ + { + "description": "At this moment, you turned back at the staircase and went to the second floor without going downstairs.", + "timestamps": { + "start": "00:02:07", + "end": "00:02:10" + } + }, + { + "description": "At this moment, you went down one floor but turned back, without reaching the first floor.", + "timestamps": { + "start": "00:03:15", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you went downstairs so far, and you kept going instead of turning back.", + "B": "No—you haven’t gone downstairs so far. The most recent time you went down the stairs, you turned back." + }, + "answer_label": "B", + "answer_text": "No—you haven’t gone downstairs so far. The most recent time you went down the stairs, you turned back.", + "q_id": "egolife_A1_D3_003_3_T2", + "correct_answer": "No—you haven’t gone downstairs so far. The most recent time you went down the stairs, you turned back.", + "original_options": { + "A": "Yes—you went downstairs so far, and you kept going instead of turning back.", + "B": "No—you haven’t gone downstairs so far. The most recent time you went down the stairs, you turned back." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_003-mcq_fixed_3", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of water did I mark with the pen so far?", + "answer": "Six.", + "question_chinese": "我到目前为止用笔一共标记了几瓶水？", + "answer_chinese": "你一共标记了三瓶水。", + "evidence": [ + { + "description": "Right now, you have marked a total of one bottle of water.", + "timestamps": { + "start": "00:04:21", + "end": "00:04:29" + } + }, + { + "description": "Right now, you have marked a total of three bottles of water.", + "timestamps": { + "start": "00:04:40", + "end": "00:04:47" + } + } + ], + "question_timestamps": "00:04:51", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "You marked five bottles of water with the pen by this point in the video.", + "D": "You have marked seven water bottles so far, counting each one you already labeled.", + "E": "Six." + }, + "answer_label": "E", + "answer_text": "Six.", + "q_id": "egolife_A1_D3_003-mcq_fixed_3", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "You marked five bottles of water with the pen by this point in the video.", + "D": "You have marked seven water bottles so far, counting each one you already labeled.", + "E": "Six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_003-mcq_fixed_2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of water did I mark with the pen so far?", + "answer": "Four.", + "question_chinese": "我到目前为止用笔一共标记了几瓶水？", + "answer_chinese": "你一共标记了六瓶水。", + "evidence": [ + { + "description": "Right now, you have marked a total of one bottle of water.", + "timestamps": { + "start": "00:04:21", + "end": "00:04:29" + } + }, + { + "description": "Right now, you have marked a total of three bottles of water.", + "timestamps": { + "start": "00:04:40", + "end": "00:04:47" + } + }, + { + "description": "Right now, you have marked a total of six bottles of water.", + "timestamps": { + "start": "00:04:51", + "end": "00:05:18" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "You marked three bottles of water with the pen up to this point in the video.", + "E": "So far, you have marked five water bottles, including the ones you labeled earlier." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_003-mcq_fixed_2", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "You marked three bottles of water with the pen up to this point in the video.", + "E": "So far, you have marked five water bottles, including the ones you labeled earlier." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D3_004-mcq.json b/QAs/egolife_A1_D3_004-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..30dd3a99fb655f0f4105c39fec67943dcaa41eb8 --- /dev/null +++ b/QAs/egolife_A1_D3_004-mcq.json @@ -0,0 +1,545 @@ +[ + { + "video_uid": "egolife_A1_D3_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_004-mcq_fixed_8", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many rooms are there on the second floor besides the kitchen and the living room?", + "answer": "Six.", + "question_chinese": "二楼除了厨房和客厅外有几个房间？", + "answer_chinese": "二楼除了厨房和客厅外有三个房间。", + "evidence": [ + { + "description": "Right now, you can see two rooms to the left of the kitchen.", + "timestamps": { + "start": "00:01:12", + "end": "00:01:13" + } + }, + { + "description": "Right now, you can see two doors to the right of the kitchen.", + "timestamps": { + "start": "00:02:45", + "end": "00:02:46" + } + }, + { + "description": "Right now, you can see that one door on the right side of the kitchen leads to the downstairs entrance, so there are three rooms in total.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:23" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "Six.", + "D": "There are five rooms on the second floor besides the kitchen and living room.", + "E": "Besides the kitchen and the living room, there are seven other rooms on the second floor." + }, + "answer_label": "C", + "answer_text": "Six.", + "q_id": "egolife_A1_D3_004-mcq_fixed_8", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "Six.", + "D": "There are five rooms on the second floor besides the kitchen and living room.", + "E": "Besides the kitchen and the living room, there are seven other rooms on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_004-mcq_fixed_9", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many rooms are there on the second floor besides the kitchen and the living room?", + "answer": "Three.", + "question_chinese": "二楼除了厨房和客厅外有几个房间？", + "answer_chinese": "有三个房间。", + "evidence": [ + { + "description": "Right now, you can see two rooms to the left of the kitchen.", + "timestamps": { + "start": "00:01:12", + "end": "00:01:13" + } + }, + { + "description": "Right now, you can see two doors to the right of the kitchen.", + "timestamps": { + "start": "00:02:45", + "end": "00:02:46" + } + }, + { + "description": "Right now, you can see that one door on the right side of the kitchen leads to the downstairs entrance, so there are three rooms in total.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:23" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "There are four rooms on the second floor besides the kitchen and living room.", + "D": "Three.", + "E": "Besides the kitchen and the living room, there are six other rooms on the second floor." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_004-mcq_fixed_9", + "correct_answer": "Three.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "There are four rooms on the second floor besides the kitchen and living room.", + "D": "Three.", + "E": "Besides the kitchen and the living room, there are six other rooms on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_004-mcq_fixed_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of water did I take from the kitchen earlier?", + "answer": "Four.", + "question_chinese": "我之前从厨房拿了几瓶水？", + "answer_chinese": "你之前从厨房没有拿一瓶水。", + "evidence": [ + { + "description": "Around this time, you did not take a bottle of water from the kitchen.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:04:42", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "You took three bottles of water from the kitchen earlier in the video.", + "E": "Earlier, you took five bottles of water from the kitchen before leaving the room." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_004-mcq_fixed_7", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "You took three bottles of water from the kitchen earlier in the video.", + "E": "Earlier, you took five bottles of water from the kitchen before leaving the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_004-mcq_fixed_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of water did I take from the kitchen earlier?", + "answer": "One.", + "question_chinese": "我之前从厨房拿了几瓶水？", + "answer_chinese": "你之前从厨房拿了四瓶水。", + "evidence": [ + { + "description": "Right now, you can be seen taking four bottles of water from the kitchen.", + "timestamps": { + "start": "00:05:26", + "end": "00:05:27" + } + }, + { + "description": "Right now, you are putting the four bottles of water you got on the dining table.", + "timestamps": { + "start": "00:05:41", + "end": "00:05:42" + } + } + ], + "question_timestamps": "00:10:11", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You took two bottles of water from the kitchen earlier in the video.", + "E": "Earlier, you took three bottles of water from the kitchen, not just a single one." + }, + "answer_label": "A", + "answer_text": "One.", + "q_id": "egolife_A1_D3_004-mcq_fixed_6", + "correct_answer": "One.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You took two bottles of water from the kitchen earlier in the video.", + "E": "Earlier, you took three bottles of water from the kitchen, not just a single one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_004-mcq_fixed_4", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the pizza I brought up from the second floor?", + "answer": "You put it on the end of the table on the second floor that's far from the kitchen.", + "question_chinese": "我从二楼拿的披萨放到了哪里？", + "answer_chinese": "你放在了二楼桌子远离厨房的一端。", + "evidence": [ + { + "description": "Right now, you are putting the pizza on the end of the dining table far from the kitchen.", + "timestamps": { + "start": "00:05:09", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:06:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it on the end of the table on the second floor that's far from the kitchen.", + "B": "On the table end near the kitchen.", + "C": "On the middle of the second-floor table.", + "D": "You put it on the end of the table on the second floor, but on the side closest to the kitchen.", + "E": "You set it down on the middle of the table on the second floor, far from the kitchen instead of the end." + }, + "answer_label": "A", + "answer_text": "You put it on the end of the table on the second floor that's far from the kitchen.", + "q_id": "egolife_A1_D3_004-mcq_fixed_4", + "correct_answer": "You put it on the end of the table on the second floor that's far from the kitchen.", + "original_options": { + "A": "You put it on the end of the table on the second floor that's far from the kitchen.", + "B": "On the table end near the kitchen.", + "C": "On the middle of the second-floor table.", + "D": "You put it on the end of the table on the second floor, but on the side closest to the kitchen.", + "E": "You set it down on the middle of the table on the second floor, far from the kitchen instead of the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 22 + ] + }, + { + "video_uid": "egolife_A1_D3_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_004-mcq_fixed_5", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I put the pizza I brought up from the second floor?", + "answer": "The pizza was placed in the middle of the table, arranged in a row.", + "question_chinese": "我从二楼拿的披萨放到了哪里？", + "answer_chinese": "披萨放到了桌子中间摆成一列。", + "evidence": [ + { + "description": "Right now, you are putting the pizza on the end of the dining table far from the kitchen.", + "timestamps": { + "start": "00:05:09", + "end": "00:05:10" + } + }, + { + "description": "At this moment, you can see the pizza arranged in a row.", + "timestamps": { + "start": "00:07:26", + "end": "00:07:27" + } + } + ], + "question_timestamps": "00:08:42", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the left side of the table.", + "B": "The pizza was placed in the middle of the table, arranged in a row.", + "C": "Stacked in a pile on the table.", + "D": "The pizza was placed on the left side of the table, arranged in a neat row.", + "E": "You left the pizza on the table but stacked the boxes in a pile rather than lining them up." + }, + "answer_label": "B", + "answer_text": "The pizza was placed in the middle of the table, arranged in a row.", + "q_id": "egolife_A1_D3_004-mcq_fixed_5", + "correct_answer": "The pizza was placed in the middle of the table, arranged in a row.", + "original_options": { + "A": "On the left side of the table.", + "B": "The pizza was placed in the middle of the table, arranged in a row.", + "C": "Stacked in a pile on the table.", + "D": "The pizza was placed on the left side of the table, arranged in a neat row.", + "E": "You left the pizza on the table but stacked the boxes in a pile rather than lining them up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_004-mcq_fixed_2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just sit on the sofa on the second floor?", + "answer": "No, you didn't just sit on the sofa on the second floor.", + "question_chinese": "我刚刚坐过二楼的沙发吗？", + "answer_chinese": "没有，你刚刚没有坐过二楼的沙发。", + "evidence": [ + { + "description": "From this point until now, I have not sat on the sofa on the second floor.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:03:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you just sat on the sofa on the second floor.", + "B": "No, you didn't just sit on the sofa on the second floor." + }, + "answer_label": "B", + "answer_text": "No, you didn't just sit on the sofa on the second floor.", + "q_id": "egolife_A1_D3_004-mcq_fixed_2", + "correct_answer": "No, you didn't just sit on the sofa on the second floor.", + "original_options": { + "A": "Yes, you just sat on the sofa on the second floor.", + "B": "No, you didn't just sit on the sofa on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D3_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_004-mcq_fixed_3", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just sit on the sofa on the second floor?", + "answer": "Yes, you just sat on the sofa on the second floor.", + "question_chinese": "我刚刚坐过二楼的沙发吗？", + "answer_chinese": "是的，你刚刚坐过二楼的沙发。", + "evidence": [ + { + "description": "Right now, you are sitting on the sofa on the second floor.", + "timestamps": { + "start": "00:04:01", + "end": "00:04:02" + } + } + ], + "question_timestamps": "00:11:21", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you just sat on the sofa on the second floor.", + "B": "No, you did not just sit on the sofa on the second floor." + }, + "answer_label": "A", + "answer_text": "Yes, you just sat on the sofa on the second floor.", + "q_id": "egolife_A1_D3_004-mcq_fixed_3", + "correct_answer": "Yes, you just sat on the sofa on the second floor.", + "original_options": { + "A": "Yes, you just sat on the sofa on the second floor.", + "B": "No, you did not just sit on the sofa on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_004-mcq_fixed_0", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I eaten pizza on the second floor so far?", + "answer": "No, you haven't.", + "question_chinese": "我到目前为止在二楼吃过披萨吗？", + "answer_chinese": "没有，你没有吃过。", + "evidence": [ + { + "description": "Right now, the pizza is on the table, and you are not eating it.", + "timestamps": { + "start": "00:05:47", + "end": "00:05:48" + } + } + ], + "question_timestamps": "00:05:49", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you have.", + "B": "No, you haven't." + }, + "answer_label": "B", + "answer_text": "No, you haven't.", + "q_id": "egolife_A1_D3_004-mcq_fixed_0", + "correct_answer": "No, you haven't.", + "original_options": { + "A": "Yes, you have.", + "B": "No, you haven't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_004-mcq_fixed_1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I eaten pizza on the second floor so far?", + "answer": "Yes, you ate a piece of pizza.", + "question_chinese": "我到目前为止在二楼吃过披萨吗？", + "answer_chinese": "是的，你吃了一块披萨。", + "evidence": [ + { + "description": "Right now, the pizza is on the table, and you are not eating it.", + "timestamps": { + "start": "00:05:47", + "end": "00:05:48" + } + }, + { + "description": "Right now, you are putting on gloves and taking a piece of pizza to eat.", + "timestamps": { + "start": "00:09:42", + "end": "00:09:45" + } + } + ], + "question_timestamps": "00:11:45", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you ate a piece of pizza.", + "B": "No, you haven't eaten pizza there yet." + }, + "answer_label": "A", + "answer_text": "Yes, you ate a piece of pizza.", + "q_id": "egolife_A1_D3_004-mcq_fixed_1", + "correct_answer": "Yes, you ate a piece of pizza.", + "original_options": { + "A": "Yes, you ate a piece of pizza.", + "B": "No, you haven't eaten pizza there yet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + } +] diff --git a/QAs/egolife_A1_D3_005-mcq.json b/QAs/egolife_A1_D3_005-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..891d1c33e8f7a0bc438682bcf6b2eca829f36930 --- /dev/null +++ b/QAs/egolife_A1_D3_005-mcq.json @@ -0,0 +1,694 @@ +[ + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005-mcq_fixed_9", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from the living room on the second floor to my current location?", + "answer": "You went out through the door next to the kitchen on the second floor and went downstairs to your current position.", + "question_chinese": "我从二楼客厅到现在的位置经过怎样的路径？", + "answer_chinese": "你从二楼厨房旁的门出去下楼到了现在的位置。", + "evidence": [ + { + "description": "At this moment, you leave the living room on the second floor.", + "timestamps": { + "start": "00:11:14", + "end": "00:11:15" + } + }, + { + "description": "At this moment, you are going downstairs.", + "timestamps": { + "start": "00:11:16", + "end": "00:11:24" + } + } + ], + "question_timestamps": "00:11:26", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went out through the door next to the kitchen on the second floor and went downstairs to your current position.", + "B": "You stayed on the second floor.", + "C": "You went downstairs through the living-room door.", + "D": "You went out through the door beside the bathroom and went downstairs to your current position.", + "E": "You exited through the door by the bedroom and went downstairs to where you are now." + }, + "answer_label": "A", + "answer_text": "You went out through the door next to the kitchen on the second floor and went downstairs to your current position.", + "q_id": "egolife_A1_D3_005-mcq_fixed_9", + "correct_answer": "You went out through the door next to the kitchen on the second floor and went downstairs to your current position.", + "original_options": { + "A": "You went out through the door next to the kitchen on the second floor and went downstairs to your current position.", + "B": "You stayed on the second floor.", + "C": "You went downstairs through the living-room door.", + "D": "You went out through the door beside the bathroom and went downstairs to your current position.", + "E": "You exited through the door by the bedroom and went downstairs to where you are now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005-mcq_fixed_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take from the living room on the second floor to my current location?", + "answer": "After exiting the door beside the second-floor kitchen, going downstairs, and turning left, you enter the first-floor living room.", + "question_chinese": "我从二楼客厅到现在的位置经过怎样的路径？", + "answer_chinese": "你从二楼厨房旁的门出去下楼后左转便进入了一楼客厅。", + "evidence": [ + { + "description": "At this moment, you leave the living room on the second floor.", + "timestamps": { + "start": "00:11:14", + "end": "00:11:15" + } + }, + { + "description": "At this moment, you are going downstairs.", + "timestamps": { + "start": "00:11:16", + "end": "00:11:24" + } + }, + { + "description": "At this moment, you enter the first-floor living room.", + "timestamps": { + "start": "00:11:30", + "end": "00:11:31" + } + } + ], + "question_timestamps": "00:11:35", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turned right after the stairs.", + "B": "You went straight after going downstairs.", + "C": "After exiting the second-floor kitchen door and going downstairs, you turned right and entered the first-floor living room.", + "D": "You left by the second-floor kitchen door, went downstairs, then continued straight into the first-floor living room.", + "E": "After exiting the door beside the second-floor kitchen, going downstairs, and turning left, you enter the first-floor living room." + }, + "answer_label": "E", + "answer_text": "After exiting the door beside the second-floor kitchen, going downstairs, and turning left, you enter the first-floor living room.", + "q_id": "egolife_A1_D3_005-mcq_fixed_8", + "correct_answer": "After exiting the door beside the second-floor kitchen, going downstairs, and turning left, you enter the first-floor living room.", + "original_options": { + "A": "You turned right after the stairs.", + "B": "You went straight after going downstairs.", + "C": "After exiting the second-floor kitchen door and going downstairs, you turned right and entered the first-floor living room.", + "D": "You left by the second-floor kitchen door, went downstairs, then continued straight into the first-floor living room.", + "E": "After exiting the door beside the second-floor kitchen, going downstairs, and turning left, you enter the first-floor living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005_1_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was eating, how many boxes of pizza were on the table?", + "answer": "There are six pizza boxes on the table, to your right.", + "question_chinese": "我吃饭时桌子上有几盒披萨？", + "answer_chinese": "桌子上有六盒披萨，在你的右方。", + "evidence": [ + { + "description": "At this moment, you see six boxes of pizza on the table.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:30" + } + }, + { + "description": "At this moment, you see six boxes of pizza on the table.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:07:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are six pizza boxes on the table, to your right.", + "B": "Four pizza boxes.", + "C": "Eight pizza boxes.", + "D": "There are five pizza boxes on the table, to your right side.", + "E": "There are seven pizza boxes on the table, positioned to your right while you were eating." + }, + "answer_label": "A", + "answer_text": "There are six pizza boxes on the table, to your right.", + "q_id": "egolife_A1_D3_005_1_T1", + "correct_answer": "There are six pizza boxes on the table, to your right.", + "original_options": { + "A": "There are six pizza boxes on the table, to your right.", + "B": "Four pizza boxes.", + "C": "Eight pizza boxes.", + "D": "There are five pizza boxes on the table, to your right side.", + "E": "There are seven pizza boxes on the table, positioned to your right while you were eating." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005_1_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "When I was eating, how many boxes of pizza were on the table?", + "answer": "There are six pizza boxes on the table, to your front-left.", + "question_chinese": "我吃饭时桌子上有几盒披萨？", + "answer_chinese": "桌子上有六盒披萨，在你的左前方。", + "evidence": [ + { + "description": "At this moment, you see six boxes of pizza on the table.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "At this moment, you see six boxes of pizza on the table.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:09:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four pizza boxes.", + "B": "There are six pizza boxes on the table, to your front-left.", + "C": "Eight pizza boxes.", + "D": "There are five pizza boxes on the table, to your front-left side.", + "E": "There are seven pizza boxes on the table, located to your front-left while you were eating." + }, + "answer_label": "B", + "answer_text": "There are six pizza boxes on the table, to your front-left.", + "q_id": "egolife_A1_D3_005_1_T2", + "correct_answer": "There are six pizza boxes on the table, to your front-left.", + "original_options": { + "A": "Four pizza boxes.", + "B": "There are six pizza boxes on the table, to your front-left.", + "C": "Eight pizza boxes.", + "D": "There are five pizza boxes on the table, to your front-left side.", + "E": "There are seven pizza boxes on the table, located to your front-left while you were eating." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005_5_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are to my left?", + "answer": "There are two people on your left-hand side.", + "question_chinese": "我左手边有几个人？", + "answer_chinese": "你左手边有二个人。", + "evidence": [ + { + "description": "There are three people on your left.", + "timestamps": { + "start": "00:03:28", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are two people on your left-hand side.", + "B": "Zero people on your left.", + "C": "One person on your left.", + "D": "Three people are standing on your left-hand side right next to you.", + "E": "Four people are on your left-hand side, grouped together close to your shoulder." + }, + "answer_label": "A", + "answer_text": "There are two people on your left-hand side.", + "q_id": "egolife_A1_D3_005_5_T2", + "correct_answer": "There are two people on your left-hand side.", + "original_options": { + "A": "There are two people on your left-hand side.", + "B": "Zero people on your left.", + "C": "One person on your left.", + "D": "Three people are standing on your left-hand side right next to you.", + "E": "Four people are on your left-hand side, grouped together close to your shoulder." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005_5_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are to my left?", + "answer": "There are three people to your left.", + "question_chinese": "我左手边有几个人？", + "answer_chinese": "你左手边有三个人。", + "evidence": [ + { + "description": "There are three people on your left.", + "timestamps": { + "start": "00:08:28", + "end": "00:08:29" + } + } + ], + "question_timestamps": "00:08:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One person.", + "B": "Two people.", + "C": "There are four people positioned to your left at this moment right now.", + "D": "There are five people standing together to your left-hand side right now.", + "E": "There are three people to your left." + }, + "answer_label": "E", + "answer_text": "There are three people to your left.", + "q_id": "egolife_A1_D3_005_5_T1", + "correct_answer": "There are three people to your left.", + "original_options": { + "A": "One person.", + "B": "Two people.", + "C": "There are four people positioned to your left at this moment right now.", + "D": "There are five people standing together to your left-hand side right now.", + "E": "There are three people to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005-mcq_fixed_4", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the garbage bag for the pizza box from?", + "answer": "Your garbage bag was from beside the trash can on the left side of the kitchen.", + "question_chinese": "我手里拿的放披萨盒的垃圾袋从哪里拿的？", + "answer_chinese": "你的垃圾袋是从厨房左侧的垃圾桶旁边拿到的。", + "evidence": [ + { + "description": "At this moment, you pick up the garbage bag from the floor.", + "timestamps": { + "start": "00:09:03", + "end": "00:09:04" + } + } + ], + "question_timestamps": "00:09:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From under the sink.", + "B": "Your garbage bag was from beside the trash can on the left side of the kitchen.", + "C": "From the right-side trash can.", + "D": "You grabbed the garbage bag from a drawer near the refrigerator on the kitchen side.", + "E": "You picked up the garbage bag from beside a blue rug on the kitchen’s left side." + }, + "answer_label": "B", + "answer_text": "Your garbage bag was from beside the trash can on the left side of the kitchen.", + "q_id": "egolife_A1_D3_005-mcq_fixed_4", + "correct_answer": "Your garbage bag was from beside the trash can on the left side of the kitchen.", + "original_options": { + "A": "From under the sink.", + "B": "Your garbage bag was from beside the trash can on the left side of the kitchen.", + "C": "From the right-side trash can.", + "D": "You grabbed the garbage bag from a drawer near the refrigerator on the kitchen side.", + "E": "You picked up the garbage bag from beside a blue rug on the kitchen’s left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005-mcq_fixed_5", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the garbage bag for the pizza box from?", + "answer": "Beside the trash can on the left side of the kitchen.", + "question_chinese": "我手里拿的放披萨盒的垃圾袋从哪里拿的？", + "answer_chinese": "厨房左侧的垃圾桶旁边。", + "evidence": [ + { + "description": "At this moment, you pick up the garbage bag from the floor.", + "timestamps": { + "start": "00:09:03", + "end": "00:09:04" + } + } + ], + "question_timestamps": "00:11:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Beside the right-side trash can.", + "B": "Beside the microwave on the left.", + "C": "You pulled the garbage bag from the drawer under the sink with the cleaning supplies.", + "D": "You got the garbage bag from inside the cabinet above the microwave in the kitchen.", + "E": "Beside the trash can on the left side of the kitchen." + }, + "answer_label": "E", + "answer_text": "Beside the trash can on the left side of the kitchen.", + "q_id": "egolife_A1_D3_005-mcq_fixed_5", + "correct_answer": "Beside the trash can on the left side of the kitchen.", + "original_options": { + "A": "Beside the right-side trash can.", + "B": "Beside the microwave on the left.", + "C": "You pulled the garbage bag from the drawer under the sink with the cleaning supplies.", + "D": "You got the garbage bag from inside the cabinet above the microwave in the kitchen.", + "E": "Beside the trash can on the left side of the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005-mcq_fixed_2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is my relative position to the dining table at this moment?", + "answer": "You are currently at the end of the dining table near the kitchen, with the table on your right side.", + "question_chinese": "我此时与餐桌的相对位置是什么？", + "answer_chinese": "你此时在餐桌靠近厨房的一端，餐桌在你的右手边。", + "evidence": [ + { + "description": "You are currently at the end of the dining table near the kitchen, with the table on your right side.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are currently at the end of the dining table near the kitchen, with the table on your right side.", + "B": "At table’s center, on your left.", + "C": "At the kitchen end, table left.", + "D": "You are at the far end near the window, with the kitchen entrance diagonally behind you.", + "E": "You are beside the center of the dining table, facing the hallway and away from the kitchen." + }, + "answer_label": "A", + "answer_text": "You are currently at the end of the dining table near the kitchen, with the table on your right side.", + "q_id": "egolife_A1_D3_005-mcq_fixed_2", + "correct_answer": "You are currently at the end of the dining table near the kitchen, with the table on your right side.", + "original_options": { + "A": "You are currently at the end of the dining table near the kitchen, with the table on your right side.", + "B": "At table’s center, on your left.", + "C": "At the kitchen end, table left.", + "D": "You are at the far end near the window, with the kitchen entrance diagonally behind you.", + "E": "You are beside the center of the dining table, facing the hallway and away from the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005-mcq_fixed_3", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is my relative position to the dining table at this moment?", + "answer": "You are currently sitting at the end of the dining table farthest from the kitchen, with the table directly in front of you.", + "question_chinese": "我此时与餐桌的相对位置是什么？", + "answer_chinese": "你此时坐在餐桌远离厨房的一端，餐桌在你的正前方。", + "evidence": [ + { + "description": "You are currently at the end of the dining table near the kitchen, with the table on your right side.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + }, + { + "description": "You are currently sitting at the end of the dining table farthest from the kitchen, with the table directly in front of you.", + "timestamps": { + "start": "00:01:38", + "end": "00:01:39" + } + } + ], + "question_timestamps": "00:02:07", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Nearest the kitchen, table in front.", + "B": "Farthest from the kitchen, table to your left.", + "C": "You are seated along the side closest to the kitchen, with the table running parallel to your left arm.", + "D": "You are currently sitting at the end of the dining table farthest from the kitchen, with the table directly in front of you.", + "E": "You are halfway down the table on the kitchen side, with your back partly toward the living room." + }, + "answer_label": "D", + "answer_text": "You are currently sitting at the end of the dining table farthest from the kitchen, with the table directly in front of you.", + "q_id": "egolife_A1_D3_005-mcq_fixed_3", + "correct_answer": "You are currently sitting at the end of the dining table farthest from the kitchen, with the table directly in front of you.", + "original_options": { + "A": "Nearest the kitchen, table in front.", + "B": "Farthest from the kitchen, table to your left.", + "C": "You are seated along the side closest to the kitchen, with the table running parallel to your left arm.", + "D": "You are currently sitting at the end of the dining table farthest from the kitchen, with the table directly in front of you.", + "E": "You are halfway down the table on the kitchen side, with your back partly toward the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005_4_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are on my left now?", + "answer": "Three.", + "question_chinese": "此时我左手边有几个人？", + "answer_chinese": "你左手边有两个人。", + "evidence": [ + { + "description": "At this moment, you can see one person sitting across from you, two people on your left, and three people on your right.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:49" + } + }, + { + "description": "At this moment, you can see one person sitting across from you, two people on your left, and three people on your right.", + "timestamps": { + "start": "00:02:44", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "There are five people on your left side right now today there.", + "E": "There are six people standing on your left side currently right now." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_005_4_T1", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "There are five people on your left side right now today there.", + "E": "There are six people standing on your left side currently right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_005_4_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are on my left now?", + "answer": "Three.", + "question_chinese": "此时我左手边有几个人？", + "answer_chinese": "你左手边有两个人。", + "evidence": [ + { + "description": "At this moment, you can see one person sitting across from you, two people on your left, and three people on your right.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:49" + } + }, + { + "description": "At this moment, you can see one person sitting across from you, two people on your left, and three people on your right.", + "timestamps": { + "start": "00:02:44", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:04:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "There are five people on your left side right now today there.", + "E": "There are six people standing on your left side currently right now." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_005_4_T2", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "There are five people on your left side right now today there.", + "E": "There are six people standing on your left side currently right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + } +] diff --git a/QAs/egolife_A1_D3_006-mcq.json b/QAs/egolife_A1_D3_006-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..87bc2dbe3deebd97e3b0d3ec00ae009fc4b8d68d --- /dev/null +++ b/QAs/egolife_A1_D3_006-mcq.json @@ -0,0 +1,963 @@ +[ + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_2_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the fridge relative to me?", + "answer": "On your left-hand side.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在你的左手边。", + "evidence": [ + { + "description": "When I face the door, the refrigerator is directly on my left.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + }, + { + "description": "As I walked toward the door, I turned to the whiteboard.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your right-hand side.", + "B": "Directly in front of you.", + "C": "Behind you, slightly to the left as you face forward in the room.", + "D": "In front of you and slightly to the right when you look out from your position.", + "E": "On your left-hand side." + }, + "answer_label": "E", + "answer_text": "On your left-hand side.", + "q_id": "egolife_A1_D3_006_2_T1", + "correct_answer": "On your left-hand side.", + "original_options": { + "A": "On your right-hand side.", + "B": "Directly in front of you.", + "C": "Behind you, slightly to the left as you face forward in the room.", + "D": "In front of you and slightly to the right when you look out from your position.", + "E": "On your left-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_2_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the fridge relative to me?", + "answer": "To your back-left.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在你的左后方。", + "evidence": [ + { + "description": "When I face the door, the refrigerator is directly on my left.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + }, + { + "description": "As I walked toward the door, I turned to the whiteboard.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-left.", + "B": "To your back-right.", + "C": "To your front-left.", + "D": "On your right-hand side, a little behind you as you turn from where you stand.", + "E": "Directly behind you rather than off to the left, closer to your centerline." + }, + "answer_label": "A", + "answer_text": "To your back-left.", + "q_id": "egolife_A1_D3_006_2_T2", + "correct_answer": "To your back-left.", + "original_options": { + "A": "To your back-left.", + "B": "To your back-right.", + "C": "To your front-left.", + "D": "On your right-hand side, a little behind you as you turn from where you stand.", + "E": "Directly behind you rather than off to the left, closer to your centerline." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_3_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the fridge relative to me?", + "answer": "To your back-right.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在你的右后方。", + "evidence": [ + { + "description": "The refrigerator is between the door and the countertop, and the door is on my right hand side now.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:02:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-left.", + "B": "To your back-right.", + "C": "To your front-right.", + "D": "On your right-hand side but slightly in front of you, not behind your shoulder.", + "E": "Directly behind you, centered, rather than being offset to your back-right side." + }, + "answer_label": "B", + "answer_text": "To your back-right.", + "q_id": "egolife_A1_D3_006_3_T1", + "correct_answer": "To your back-right.", + "original_options": { + "A": "To your back-left.", + "B": "To your back-right.", + "C": "To your front-right.", + "D": "On your right-hand side but slightly in front of you, not behind your shoulder.", + "E": "Directly behind you, centered, rather than being offset to your back-right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_3_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the fridge relative to me?", + "answer": "To your back-left.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在你的左后方。", + "evidence": [ + { + "description": "The refrigerator is between the door and the countertop, and the door is on my right hand side now.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:06:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-right.", + "B": "To your front-left.", + "C": "On your left-hand side, but more level with you rather than behind you.", + "D": "In front of you to the right, instead of being located behind you on the left.", + "E": "To your back-left." + }, + "answer_label": "E", + "answer_text": "To your back-left.", + "q_id": "egolife_A1_D3_006_3_T2", + "correct_answer": "To your back-left.", + "original_options": { + "A": "To your back-right.", + "B": "To your front-left.", + "C": "On your left-hand side, but more level with you rather than behind you.", + "D": "In front of you to the right, instead of being located behind you on the left.", + "E": "To your back-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_8", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator located?", + "answer": "In front of me to the right.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在我的右前方。", + "evidence": [ + { + "description": "I am standing in front of the table.", + "timestamps": { + "start": "00:07:20", + "end": "00:07:21" + } + } + ], + "question_timestamps": "00:07:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To the left of me.", + "B": "Directly in front of me.", + "C": "Behind me to the right, not ahead of me as I’m facing the space.", + "D": "In front of me to the left, rather than in front of me on the right side.", + "E": "In front of me to the right." + }, + "answer_label": "E", + "answer_text": "In front of me to the right.", + "q_id": "egolife_A1_D3_006_8", + "correct_answer": "In front of me to the right.", + "original_options": { + "A": "To the left of me.", + "B": "Directly in front of me.", + "C": "Behind me to the right, not ahead of me as I’m facing the space.", + "D": "In front of me to the left, rather than in front of me on the right side.", + "E": "In front of me to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006-mcq_fixed_4", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the first floor to the second floor?", + "answer": "At the beginning, You were on the first floor, then YOU climbed two flights of stairs and arrived at the second floor.", + "question_chinese": "我从一楼到二楼的路径是什么？", + "answer_chinese": "开始时，我在一楼，然后上了俩层楼梯，到达二楼。", + "evidence": [ + { + "description": "At the beginning, I was on the first floor.", + "timestamps": { + "start": "00:03:13", + "end": "00:03:14" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:03:21", + "end": "00:03:34" + } + } + ], + "question_timestamps": "00:03:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "At the beginning, You were on the first floor, then YOU climbed two flights of stairs and arrived at the second floor.", + "B": "You climbed one flight of stairs.", + "C": "You climbed three flights of stairs.", + "D": "You started on the first floor, took one long staircase, and reached the second floor in a single climb.", + "E": "You began on the first floor, went to a landing, then climbed one more short staircase to reach the second floor." + }, + "answer_label": "A", + "answer_text": "At the beginning, You were on the first floor, then YOU climbed two flights of stairs and arrived at the second floor.", + "q_id": "egolife_A1_D3_006-mcq_fixed_4", + "correct_answer": "At the beginning, You were on the first floor, then YOU climbed two flights of stairs and arrived at the second floor.", + "original_options": { + "A": "At the beginning, You were on the first floor, then YOU climbed two flights of stairs and arrived at the second floor.", + "B": "You climbed one flight of stairs.", + "C": "You climbed three flights of stairs.", + "D": "You started on the first floor, took one long staircase, and reached the second floor in a single climb.", + "E": "You began on the first floor, went to a landing, then climbed one more short staircase to reach the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006-mcq_fixed_5", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the first floor to the second floor?", + "answer": "You went from the first floor to the second floor.", + "question_chinese": "我从一楼到二楼的路径是什么？", + "answer_chinese": "我从一楼，到达二楼。", + "evidence": [ + { + "description": "At the beginning, I was on the first floor.", + "timestamps": { + "start": "00:03:13", + "end": "00:03:14" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:03:21", + "end": "00:03:34" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went from second to first.", + "B": "You went from the first floor to the second floor.", + "C": "You went from first to garden.", + "D": "You took an elevator and stopped at a mezzanine, never actually reaching the second floor.", + "E": "You stayed on the first floor and only walked through the lobby toward the stairs without going up." + }, + "answer_label": "B", + "answer_text": "You went from the first floor to the second floor.", + "q_id": "egolife_A1_D3_006-mcq_fixed_5", + "correct_answer": "You went from the first floor to the second floor.", + "original_options": { + "A": "You went from second to first.", + "B": "You went from the first floor to the second floor.", + "C": "You went from first to garden.", + "D": "You took an elevator and stopped at a mezzanine, never actually reaching the second floor.", + "E": "You stayed on the first floor and only walked through the lobby toward the stairs without going up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006-mcq_fixed_2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "我刚出卧室门时，笔记本电脑在我的左后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:02:29", + "end": "00:02:30" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind you to the right.", + "B": "In front of you.", + "C": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "D": "When you stepped out of the bedroom door, the laptop was diagonally in front of you, closer to the right wall.", + "E": "When you stepped out of the bedroom door, the laptop was in front of you to the right, partly visible." + }, + "answer_label": "C", + "answer_text": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "q_id": "egolife_A1_D3_006-mcq_fixed_2", + "correct_answer": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "original_options": { + "A": "Behind you to the right.", + "B": "In front of you.", + "C": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "D": "When you stepped out of the bedroom door, the laptop was diagonally in front of you, closer to the right wall.", + "E": "When you stepped out of the bedroom door, the laptop was in front of you to the right, partly visible." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 21, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006-mcq_fixed_3", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "The laptop is behind you.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "笔记本电脑在我的后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:31" + } + }, + { + "description": "I turned 90 degrees and left the dormitory.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:09:38", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The laptop is in front of you.", + "B": "The laptop is behind you.", + "C": "The laptop is to your left.", + "D": "The laptop is ahead of you, slightly to the right as you step out of the bedroom doorway.", + "E": "The laptop sits off to your front-left, so you face it as soon as you exit the bedroom." + }, + "answer_label": "B", + "answer_text": "The laptop is behind you.", + "q_id": "egolife_A1_D3_006-mcq_fixed_3", + "correct_answer": "The laptop is behind you.", + "original_options": { + "A": "The laptop is in front of you.", + "B": "The laptop is behind you.", + "C": "The laptop is to your left.", + "D": "The laptop is ahead of you, slightly to the right as you step out of the bedroom doorway.", + "E": "The laptop sits off to your front-left, so you face it as soon as you exit the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_9", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the paper cup or the power bank?", + "answer": "Cup.", + "question_chinese": "纸杯和充电宝那个距我更近？", + "answer_chinese": "杯子。", + "evidence": [ + { + "description": "The paper cup is directly in front of me where I am standing.", + "timestamps": { + "start": "00:06:18", + "end": "00:06:19" + } + }, + { + "description": "The power bank is in front of me to the right.", + "timestamps": { + "start": "00:06:18", + "end": "00:06:19" + } + } + ], + "question_timestamps": "00:06:30", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Power bank.", + "B": "Cup." + }, + "answer_label": "B", + "answer_text": "Cup.", + "q_id": "egolife_A1_D3_006_9", + "correct_answer": "Cup.", + "original_options": { + "A": "Power bank.", + "B": "Cup." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_10", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the paper cup or the power bank?", + "answer": "Paper cup.", + "question_chinese": "纸杯和充电宝那个距我更近？", + "answer_chinese": "纸杯。", + "evidence": [ + { + "description": "A paper cup is by my hand.", + "timestamps": { + "start": "00:10:40", + "end": "00:10:50" + } + }, + { + "description": "The power bank is in the upper left corner of the table.", + "timestamps": { + "start": "00:10:40", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:11:01", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Paper cup.", + "B": "Power bank." + }, + "answer_label": "A", + "answer_text": "Paper cup.", + "q_id": "egolife_A1_D3_006_10", + "correct_answer": "Paper cup.", + "original_options": { + "A": "Paper cup.", + "B": "Power bank." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path of the cardboard box?", + "answer": "A boy packed the cardboard box and handed it to me, then I put it in the microwave.", + "question_chinese": "纸盒的路径是什么？", + "answer_chinese": "一个男生装好纸盒，递给了我，然后被我放到了微波炉了。", + "evidence": [ + { + "description": "At the beginning, two cardboard boxes were on the table.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:11" + } + }, + { + "description": "A boy handed me two cardboard boxes.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:19" + } + }, + { + "description": "Finally, I put the cardboard box into the microwave oven.", + "timestamps": { + "start": "00:01:31", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:01:38", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "I put it in the sink.", + "B": "I put it in the closet.", + "C": "A boy packed the cardboard box and handed it to me, then I set it down on the table.", + "D": "A boy packed the cardboard box and handed it to me, then I put it in the microwave.", + "E": "A boy packed the cardboard box, handed it to me, and afterward I placed it outside in the garden." + }, + "answer_label": "D", + "answer_text": "A boy packed the cardboard box and handed it to me, then I put it in the microwave.", + "q_id": "egolife_A1_D3_006_1", + "correct_answer": "A boy packed the cardboard box and handed it to me, then I put it in the microwave.", + "original_options": { + "A": "I put it in the sink.", + "B": "I put it in the closet.", + "C": "A boy packed the cardboard box and handed it to me, then I set it down on the table.", + "D": "A boy packed the cardboard box and handed it to me, then I put it in the microwave.", + "E": "A boy packed the cardboard box, handed it to me, and afterward I placed it outside in the garden." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path of the cardboard box?", + "answer": "A boy handed me two cardboard boxes, and finally I put the boxes into the microwave oven.", + "question_chinese": "纸盒的路径是什么？", + "answer_chinese": "一个男生把俩个纸盒递给了我，最后我把纸盒放进了微波炉。", + "evidence": [ + { + "description": "At the beginning, two cardboard boxes were on the table.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:11" + } + }, + { + "description": "A boy handed me two cardboard boxes.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:19" + } + }, + { + "description": "Finally, I put the cardboard box into the microwave oven.", + "timestamps": { + "start": "00:01:31", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:08:37", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "A boy handed me one box.", + "B": "A girl handed me two boxes.", + "C": "A boy handed me two cardboard boxes, and finally I put the boxes into the microwave oven.", + "D": "A boy handed me two cardboard boxes, and finally I placed the boxes on the kitchen counter instead.", + "E": "A boy handed me two cardboard boxes, and in the end I left them stacked on the floor." + }, + "answer_label": "C", + "answer_text": "A boy handed me two cardboard boxes, and finally I put the boxes into the microwave oven.", + "q_id": "egolife_A1_D3_006_2", + "correct_answer": "A boy handed me two cardboard boxes, and finally I put the boxes into the microwave oven.", + "original_options": { + "A": "A boy handed me one box.", + "B": "A girl handed me two boxes.", + "C": "A boy handed me two cardboard boxes, and finally I put the boxes into the microwave oven.", + "D": "A boy handed me two cardboard boxes, and finally I placed the boxes on the kitchen counter instead.", + "E": "A boy handed me two cardboard boxes, and in the end I left them stacked on the floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_4_T1", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the tabletop in front of me, which is closer to me: the paper cup or the power bank?", + "answer": "The paper cup on the tabletop in front of you is closer to you.", + "question_chinese": "面前桌面上的纸杯和充电宝那个距我更近？", + "answer_chinese": "面前桌面上纸杯离你更近。", + "evidence": [ + { + "description": "The paper cup is directly in front of me where I am standing.The power bank is in front of me to the right.", + "timestamps": { + "start": "00:06:18", + "end": "00:06:19" + } + } + ], + "question_timestamps": "00:06:21", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The paper cup on the tabletop in front of you is closer to you.", + "B": "The power bank on the tabletop in front of you is closer to you." + }, + "answer_label": "A", + "answer_text": "The paper cup on the tabletop in front of you is closer to you.", + "q_id": "egolife_A1_D3_006_4_T1", + "correct_answer": "The paper cup on the tabletop in front of you is closer to you.", + "original_options": { + "A": "The paper cup on the tabletop in front of you is closer to you.", + "B": "The power bank on the tabletop in front of you is closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_4_T2", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the tabletop in front of me, which is closer to me: the paper cup or the power bank?", + "answer": "The power bank on the tabletop in front of you is closer to you.", + "question_chinese": "面前桌面上的纸杯和充电宝那个距我更近？", + "answer_chinese": "面前桌面上充电宝离你更近。", + "evidence": [ + { + "description": "I walked from the right rear of the table to the left rear, getting closer to the power bank.", + "timestamps": { + "start": "00:06:53", + "end": "00:06:54" + } + } + ], + "question_timestamps": "00:06:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The paper cup on the tabletop in front of you is closer to you.", + "B": "The power bank on the tabletop in front of you is closer to you." + }, + "answer_label": "B", + "answer_text": "The power bank on the tabletop in front of you is closer to you.", + "q_id": "egolife_A1_D3_006_4_T2", + "correct_answer": "The power bank on the tabletop in front of you is closer to you.", + "original_options": { + "A": "The paper cup on the tabletop in front of you is closer to you.", + "B": "The power bank on the tabletop in front of you is closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_4_T4", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the tabletop in front of me, which is closer to me: the paper cup or the power bank?", + "answer": "The power bank on the tabletop in front of you is closer to you.", + "question_chinese": "面前桌面上的纸杯和充电宝那个距我更近？", + "answer_chinese": "面前桌面上充电宝离你更近。", + "evidence": [ + { + "description": "I got up from the sofa and walked towards the dining table. The power bank is closer than the paper cup.", + "timestamps": { + "start": "00:09:05", + "end": "00:09:06" + } + } + ], + "question_timestamps": "00:09:08", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The paper cup on the tabletop in front of you is closer to you.", + "B": "The power bank on the tabletop in front of you is closer to you." + }, + "answer_label": "B", + "answer_text": "The power bank on the tabletop in front of you is closer to you.", + "q_id": "egolife_A1_D3_006_4_T4", + "correct_answer": "The power bank on the tabletop in front of you is closer to you.", + "original_options": { + "A": "The paper cup on the tabletop in front of you is closer to you.", + "B": "The power bank on the tabletop in front of you is closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_006_4_T3", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the tabletop in front of me, which is closer to me: the paper cup or the power bank?", + "answer": "The paper cup on the tabletop in front of you is closer to you.", + "question_chinese": "面前桌面上的纸杯和充电宝那个距我更近？", + "answer_chinese": "面前桌面上纸杯离你更近。", + "evidence": [ + { + "description": "A paper cup is by my hand.The power bank is in the upper left corner of the table.", + "timestamps": { + "start": "00:10:40", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:10:52", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The paper cup on the tabletop in front of you is closer to you.", + "B": "The power bank on the tabletop in front of you is closer to you." + }, + "answer_label": "A", + "answer_text": "The paper cup on the tabletop in front of you is closer to you.", + "q_id": "egolife_A1_D3_006_4_T3", + "correct_answer": "The paper cup on the tabletop in front of you is closer to you.", + "original_options": { + "A": "The paper cup on the tabletop in front of you is closer to you.", + "B": "The power bank on the tabletop in front of you is closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + } +] diff --git a/QAs/egolife_A1_D3_007-mcq.json b/QAs/egolife_A1_D3_007-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2a8d20890f7e05bca23f218643c6b50a79b1fe25 --- /dev/null +++ b/QAs/egolife_A1_D3_007-mcq.json @@ -0,0 +1,592 @@ +[ + { + "video_uid": "egolife_A1_D3_007", + "video_length": "00:11:13", + "qid": "egolife_A1_D3_007-mcq_fixed_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the second floor to the first floor?", + "answer": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "question_chinese": "我从二楼到一楼的路径是什么？", + "answer_chinese": "开始时，我在二楼，下了俩层楼梯之后我到达了一楼。", + "evidence": [ + { + "description": "At the beginning, we were all on the second floor.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:12" + } + }, + { + "description": "I went down two flights of stairs", + "timestamps": { + "start": "00:08:13", + "end": "00:08:30" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took one flight of stairs.", + "B": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "C": "You took three flights of stairs.", + "D": "Starting on the second floor, you took the elevator down and walked straight out onto the first floor.", + "E": "From the second floor, you descended one long staircase and stepped directly onto the first floor without another flight." + }, + "answer_label": "B", + "answer_text": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "q_id": "egolife_A1_D3_007-mcq_fixed_6", + "correct_answer": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "original_options": { + "A": "You took one flight of stairs.", + "B": "At the beginning, You were on the second floor. After going down two flights of stairs, YOU arrived at the first floor.", + "C": "You took three flights of stairs.", + "D": "Starting on the second floor, you took the elevator down and walked straight out onto the first floor.", + "E": "From the second floor, you descended one long staircase and stepped directly onto the first floor without another flight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_007", + "video_length": "00:11:13", + "qid": "egolife_A1_D3_007-mcq_fixed_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the second floor to the first floor?", + "answer": "You went downstairs and arrived at the first floor.", + "question_chinese": "我从二楼到一楼的路径是什么？", + "answer_chinese": "下楼到达了一楼。", + "evidence": [ + { + "description": "At the beginning, we were all on the second floor.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:13" + } + }, + { + "description": "I went down two flights of stairs", + "timestamps": { + "start": "00:08:20", + "end": "00:08:30" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went downstairs and arrived at the first floor.", + "B": "You went upstairs to the first floor.", + "C": "You stayed on the second floor.", + "D": "You took an escalator down and ended up on the ground level instead of the first floor.", + "E": "You walked along the second-floor hallway, then used an exterior ramp to reach the first-floor entrance." + }, + "answer_label": "A", + "answer_text": "You went downstairs and arrived at the first floor.", + "q_id": "egolife_A1_D3_007-mcq_fixed_7", + "correct_answer": "You went downstairs and arrived at the first floor.", + "original_options": { + "A": "You went downstairs and arrived at the first floor.", + "B": "You went upstairs to the first floor.", + "C": "You stayed on the second floor.", + "D": "You took an escalator down and ended up on the ground level instead of the first floor.", + "E": "You walked along the second-floor hallway, then used an exterior ramp to reach the first-floor entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_007", + "video_length": "00:11:13", + "qid": "egolife_A1_D3_007-mcq_fixed_8", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "我刚出卧室门时，笔记本电脑在我的左后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:08:54", + "end": "00:08:55" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:08:56", + "end": "00:08:57" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you, left.", + "B": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "C": "Behind you, right.", + "D": "As you stepped out of the bedroom, the laptop was directly in front of you and slightly to the right.", + "E": "When you exited the bedroom doorway, the laptop sat ahead of you on your left, near the hallway side." + }, + "answer_label": "B", + "answer_text": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "q_id": "egolife_A1_D3_007-mcq_fixed_8", + "correct_answer": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "original_options": { + "A": "In front of you, left.", + "B": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "C": "Behind you, right.", + "D": "As you stepped out of the bedroom, the laptop was directly in front of you and slightly to the right.", + "E": "When you exited the bedroom doorway, the laptop sat ahead of you on your left, near the hallway side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 20, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_007", + "video_length": "00:11:13", + "qid": "egolife_A1_D3_007-mcq_fixed_9", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "The laptop is behind you.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "笔记本电脑在我的后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:08:54", + "end": "00:08:55" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:08:56", + "end": "00:08:57" + } + } + ], + "question_timestamps": "00:11:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The laptop is behind you.", + "B": "The laptop is in front of you.", + "C": "The laptop is to your left.", + "D": "The laptop is directly ahead of you in the corner, next to the window visible from the doorway.", + "E": "The laptop is off to your right side as you step out, closer to the hallway than the bedroom." + }, + "answer_label": "A", + "answer_text": "The laptop is behind you.", + "q_id": "egolife_A1_D3_007-mcq_fixed_9", + "correct_answer": "The laptop is behind you.", + "original_options": { + "A": "The laptop is behind you.", + "B": "The laptop is in front of you.", + "C": "The laptop is to your left.", + "D": "The laptop is directly ahead of you in the corner, next to the window visible from the doorway.", + "E": "The laptop is off to your right side as you step out, closer to the hallway than the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_007", + "video_length": "00:11:13", + "qid": "egolife_A1_D3_007_6", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the bottled water?", + "answer": "No.", + "question_chinese": "我能否够到矿泉水？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "Mineral water is in front of me.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:08" + } + }, + { + "description": "Mineral water is to the left of the table, and I am at the far right of the table.", + "timestamps": { + "start": "00:05:08", + "end": "00:05:09" + } + } + ], + "question_timestamps": "00:05:13", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D3_007_6", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_007", + "video_length": "00:11:13", + "qid": "egolife_A1_D3_007_5", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the bottled water?", + "answer": "Yes.", + "question_chinese": "我能否够到矿泉水？", + "answer_chinese": "能。", + "evidence": [ + { + "description": "Mineral water is in front of me.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:36" + } + }, + { + "description": "I am one arm's length away from the mineral water.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:36" + } + } + ], + "question_timestamps": "00:05:40", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D3_007_5", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_007", + "video_length": "00:11:13", + "qid": "egolife_A1_D3_007-mcq_fixed_2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many items are on the table?", + "answer": "Six.", + "question_chinese": "桌子上有几种物品？", + "answer_chinese": "四种。", + "evidence": [ + { + "description": "There are electrical devices such as mobile phones and power banks.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:48" + } + }, + { + "description": "There are paper cups.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:48" + } + }, + { + "description": "There are drinks.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:48" + } + }, + { + "description": "There are playing cards.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:48" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four items.", + "B": "Five items.", + "C": "Seven items are on the table in total, including everything visible on its surface.", + "D": "Six.", + "E": "Eight items are on the table altogether, counting all the objects placed on top." + }, + "answer_label": "D", + "answer_text": "Six.", + "q_id": "egolife_A1_D3_007-mcq_fixed_2", + "correct_answer": "Six.", + "original_options": { + "A": "Four items.", + "B": "Five items.", + "C": "Seven items are on the table in total, including everything visible on its surface.", + "D": "Six.", + "E": "Eight items are on the table altogether, counting all the objects placed on top." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_007", + "video_length": "00:11:13", + "qid": "egolife_A1_D3_007-mcq_fixed_3", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many items are on the table?", + "answer": "Four.", + "question_chinese": "桌子上有几种物品？", + "answer_chinese": "五种。", + "evidence": [ + { + "description": "There are electrical devices such as mobile phones and power banks.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:48" + } + }, + { + "description": "There are paper cups.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:48" + } + }, + { + "description": "There are playing cards.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:48" + } + }, + { + "description": "There are drinks.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:48" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "Two items are on the table, with the rest of the surface clear.", + "D": "Six items are on the table, spread out rather than clustered together.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_007-mcq_fixed_3", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "Two items are on the table, with the rest of the surface clear.", + "D": "Six items are on the table, spread out rather than clustered together.", + "E": "Four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_007", + "video_length": "00:11:13", + "qid": "egolife_A1_D3_007_1_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me, the paper cup or the phone?", + "answer": "The paper cup is in front of you.", + "question_chinese": "此时，纸杯和手机那个距我更近一些？", + "answer_chinese": "纸杯，在你的前方。", + "evidence": [ + { + "description": "Paper cups and receipts are on the table, both directly in front of me.The mobile phone is in front of the paper cup.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The paper cup is in front of you.", + "B": "The phone is directly to your right." + }, + "answer_label": "A", + "answer_text": "The paper cup is in front of you.", + "q_id": "egolife_A1_D3_007_1_T1", + "correct_answer": "The paper cup is in front of you.", + "original_options": { + "A": "The paper cup is in front of you.", + "B": "The phone is directly to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D3_007", + "video_length": "00:11:13", + "qid": "egolife_A1_D3_007_1_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me, the paper cup or the phone?", + "answer": "The phone is directly to your right.", + "question_chinese": "此时，纸杯和手机那个距我更近一些？", + "answer_chinese": "手机，在你的正右方。", + "evidence": [ + { + "description": "I stood in front of the table and saw the mobile phone and paper cups directly in front of me.The mobile phone is lying on the corner of the table.", + "timestamps": { + "start": "00:06:48", + "end": "00:06:49" + } + } + ], + "question_timestamps": "00:07:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The paper cup is in front of you.", + "B": "The phone is directly to your right." + }, + "answer_label": "B", + "answer_text": "The phone is directly to your right.", + "q_id": "egolife_A1_D3_007_1_T2", + "correct_answer": "The phone is directly to your right.", + "original_options": { + "A": "The paper cup is in front of you.", + "B": "The phone is directly to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + } +] diff --git a/QAs/egolife_A1_D3_008-mcq.json b/QAs/egolife_A1_D3_008-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..978acbe1dcc6351d033635b7f6878dcaac78b141 --- /dev/null +++ b/QAs/egolife_A1_D3_008-mcq.json @@ -0,0 +1,563 @@ +[ + { + "video_uid": "egolife_A1_D3_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_008-mcq_fixed_4", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is placed at the entrance of the pavilion?", + "answer": "There is an orange-red children's scooter placed at the entrance of the pavilion.", + "question_chinese": "亭子门口放了一个什么东西？", + "answer_chinese": "亭子门口放了一个橙红色的儿童滑板车。", + "evidence": [ + { + "description": "Based on the current scene, you can see a children's scooter parked at the entrance of the pavilion.", + "timestamps": { + "start": "00:09:58", + "end": "00:10:00" + } + }, + { + "description": "Based on the current scene, you can see a children's scooter parked at the entrance of the pavilion.", + "timestamps": { + "start": "00:10:19", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:10:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A yellow skateboard.", + "B": "A purple beach ball.", + "C": "There is an orange-red children's scooter placed at the entrance of the pavilion.", + "D": "There is a small red-and-yellow pushcart parked right at the entrance of the pavilion.", + "E": "There is a low wooden bench positioned at the entrance of the pavilion, right by the doorway." + }, + "answer_label": "C", + "answer_text": "There is an orange-red children's scooter placed at the entrance of the pavilion.", + "q_id": "egolife_A1_D3_008-mcq_fixed_4", + "correct_answer": "There is an orange-red children's scooter placed at the entrance of the pavilion.", + "original_options": { + "A": "A yellow skateboard.", + "B": "A purple beach ball.", + "C": "There is an orange-red children's scooter placed at the entrance of the pavilion.", + "D": "There is a small red-and-yellow pushcart parked right at the entrance of the pavilion.", + "E": "There is a low wooden bench positioned at the entrance of the pavilion, right by the doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_008-mcq_fixed_5", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is placed at the entrance of the pavilion?", + "answer": "There is an orange-red children's scooter placed.", + "question_chinese": "亭子门口放了一个什么东西？", + "answer_chinese": "一个橙红色的儿童滑板车。", + "evidence": [ + { + "description": "Based on the current scene, you can see a children's scooter parked at the entrance of the pavilion.", + "timestamps": { + "start": "00:09:58", + "end": "00:10:00" + } + }, + { + "description": "Based on the current scene, you can see a children's scooter parked at the entrance of the pavilion.", + "timestamps": { + "start": "00:10:19", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:11:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A green tricycle.", + "B": "A blue bicycle.", + "C": "There is an orange-red children's scooter placed.", + "D": "A worn pedal-powered go-kart with chipped paint is parked right where you enter the pavilion.", + "E": "A bright yellow skateboard is leaning against the side of the entrance near the doorway." + }, + "answer_label": "C", + "answer_text": "There is an orange-red children's scooter placed.", + "q_id": "egolife_A1_D3_008-mcq_fixed_5", + "correct_answer": "There is an orange-red children's scooter placed.", + "original_options": { + "A": "A green tricycle.", + "B": "A blue bicycle.", + "C": "There is an orange-red children's scooter placed.", + "D": "A worn pedal-powered go-kart with chipped paint is parked right where you enter the pavilion.", + "E": "A bright yellow skateboard is leaning against the side of the entrance near the doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_008-mcq_fixed_8", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many life rings were there on the lakeside railing I just passed by?", + "answer": "Six.", + "question_chinese": "刚刚我经过的湖旁护栏有几个救生圈？", + "answer_chinese": "湖旁护栏有一个救生圈。", + "evidence": [ + { + "description": "At this moment, you can see a life ring on the railing.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "Five.", + "C": "Seven.", + "D": "Four life rings were on the lakeside railing as you walked past that section.", + "E": "Eight life rings were lined along the lakeside railing you just passed, evenly spaced out." + }, + "answer_label": "A", + "answer_text": "Six.", + "q_id": "egolife_A1_D3_008-mcq_fixed_8", + "correct_answer": "Six.", + "original_options": { + "A": "Six.", + "B": "Five.", + "C": "Seven.", + "D": "Four life rings were on the lakeside railing as you walked past that section.", + "E": "Eight life rings were lined along the lakeside railing you just passed, evenly spaced out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_008-mcq_fixed_9", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many life rings were there on the lakeside railing I just passed by?", + "answer": "One.", + "question_chinese": "刚刚我经过的湖旁护栏有几个救生圈？", + "answer_chinese": "一个救生圈。", + "evidence": [ + { + "description": "At this moment, you can see a life ring on the railing.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four life rings were on the lakeside railing you just passed, placed along one stretch.", + "D": "Five life rings were attached to the lakeside railing you walked by, one after another.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_D3_008-mcq_fixed_9", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four life rings were on the lakeside railing you just passed, placed along one stretch.", + "D": "Five life rings were attached to the lakeside railing you walked by, one after another.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_008-mcq_fixed_7", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take from the fish-watching spot to my current location?", + "answer": "You went straight along the path to the left of the fish-watching spot, turned right at the bend, and continued straight to your current location.", + "question_chinese": "我从观鱼处到现在的位置经历了怎么的路线？", + "answer_chinese": "你从观鱼处左边的路直走到拐弯处右转直走到了现在的位置。", + "evidence": [ + { + "description": "During this period of time, you walked straight from the fish-watching spot to the first bend.", + "timestamps": { + "start": "00:07:18", + "end": "00:07:33" + } + }, + { + "description": "At this moment, turn right at the junction and go straight.", + "timestamps": { + "start": "00:07:38", + "end": "00:07:39" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took the path on the right.", + "B": "You turned left at the bend.", + "C": "You went straight along the path to the right of the fish-watching spot, turned right at the bend, then continued straight.", + "D": "You went straight along the path to the left of the fish-watching spot, turned right at the bend, and continued straight to your current location.", + "E": "You went straight on the left-side path, turned left at the bend, and then kept going straight to here." + }, + "answer_label": "D", + "answer_text": "You went straight along the path to the left of the fish-watching spot, turned right at the bend, and continued straight to your current location.", + "q_id": "egolife_A1_D3_008-mcq_fixed_7", + "correct_answer": "You went straight along the path to the left of the fish-watching spot, turned right at the bend, and continued straight to your current location.", + "original_options": { + "A": "You took the path on the right.", + "B": "You turned left at the bend.", + "C": "You went straight along the path to the right of the fish-watching spot, turned right at the bend, then continued straight.", + "D": "You went straight along the path to the left of the fish-watching spot, turned right at the bend, and continued straight to your current location.", + "E": "You went straight on the left-side path, turned left at the bend, and then kept going straight to here." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 21, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_008-mcq_fixed_6", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take from the fish-watching spot to my current location?", + "answer": "You go straight along the left path from the fish-watching spot, turn right at the bend, and keep walking straight until you can see the pavilion on your right hand side.", + "question_chinese": "我从观鱼处到现在的位置经历了怎么的路线？", + "answer_chinese": "你从观鱼处左边的路直走到拐弯处右转直走一直到可以看到右手边的亭子。", + "evidence": [ + { + "description": "During this period of time, you walked straight from the fish-watching spot to the first bend.", + "timestamps": { + "start": "00:07:18", + "end": "00:07:33" + } + }, + { + "description": "At this moment, turn right at the junction and go straight.", + "timestamps": { + "start": "00:07:38", + "end": "00:07:39" + } + }, + { + "description": "At this moment, you can see the pavilion on your right.", + "timestamps": { + "start": "00:09:51", + "end": "00:09:52" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Take the right path and continue straight.", + "B": "Turn left at the bend, then continue.", + "C": "Go straight along the left path, turn left at the bend, then walk on until the pavilion appears on your right.", + "D": "You go straight along the left path from the fish-watching spot, turn right at the bend, and keep walking straight until you can see the pavilion on your right hand side.", + "E": "Go straight on the left path, turn right at the bend, and continue until the pavilion is on your left." + }, + "answer_label": "D", + "answer_text": "You go straight along the left path from the fish-watching spot, turn right at the bend, and keep walking straight until you can see the pavilion on your right hand side.", + "q_id": "egolife_A1_D3_008-mcq_fixed_6", + "correct_answer": "You go straight along the left path from the fish-watching spot, turn right at the bend, and keep walking straight until you can see the pavilion on your right hand side.", + "original_options": { + "A": "Take the right path and continue straight.", + "B": "Turn left at the bend, then continue.", + "C": "Go straight along the left path, turn left at the bend, then walk on until the pavilion appears on your right.", + "D": "You go straight along the left path from the fish-watching spot, turn right at the bend, and keep walking straight until you can see the pavilion on your right hand side.", + "E": "Go straight on the left path, turn right at the bend, and continue until the pavilion is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_008-mcq_fixed_1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just eat pizza by the lake?", + "answer": "You did not just eat pizza.", + "question_chinese": "我刚刚在湖边吃披萨了吗？", + "answer_chinese": "你刚刚没有吃披萨。", + "evidence": [ + { + "description": "There is no evidence that you ate pizza.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You did not just eat pizza.", + "B": "You did just eat pizza by the lake." + }, + "answer_label": "A", + "answer_text": "You did not just eat pizza.", + "q_id": "egolife_A1_D3_008-mcq_fixed_1", + "correct_answer": "You did not just eat pizza.", + "original_options": { + "A": "You did not just eat pizza.", + "B": "You did just eat pizza by the lake." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D3_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_008-mcq_fixed_0", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just eat pizza by the lake?", + "answer": "Yes，you just ate pizza.", + "question_chinese": "我刚刚在湖边吃披萨了吗？", + "answer_chinese": "你刚刚吃了披萨。", + "evidence": [ + { + "description": "At this moment, you took a piece of pizza from the pizza box on the stone bench.", + "timestamps": { + "start": "00:03:24", + "end": "00:03:26" + } + }, + { + "description": "At this moment, you put the pizza into your mouth.", + "timestamps": { + "start": "00:03:30", + "end": "00:03:31" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you did not just eat pizza.", + "B": "Yes，you just ate pizza." + }, + "answer_label": "B", + "answer_text": "Yes，you just ate pizza.", + "q_id": "egolife_A1_D3_008-mcq_fixed_0", + "correct_answer": "Yes，you just ate pizza.", + "original_options": { + "A": "No, you did not just eat pizza.", + "B": "Yes，you just ate pizza." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D3_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_008-mcq_fixed_3", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the pizza box now?", + "answer": "The pizza box is on the stone bench beside the fence.", + "question_chinese": "披萨盒现在在哪个位置？", + "answer_chinese": "披萨盒在围栏旁边的石凳上。", + "evidence": [ + { + "description": "At this moment, the pizza can be seen on the stone bench beside the fence.", + "timestamps": { + "start": "00:04:06", + "end": "00:04:07" + } + } + ], + "question_timestamps": "00:06:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the wooden bench by the fence.", + "B": "On the metal bench by the fence.", + "C": "The pizza box is on the stone bench beside the fence.", + "D": "The pizza box is on the wooden picnic table near the gate, not on the bench beside the fence.", + "E": "You left the pizza box under the pergola beside the patio chairs, rather than on the stone bench." + }, + "answer_label": "C", + "answer_text": "The pizza box is on the stone bench beside the fence.", + "q_id": "egolife_A1_D3_008-mcq_fixed_3", + "correct_answer": "The pizza box is on the stone bench beside the fence.", + "original_options": { + "A": "On the wooden bench by the fence.", + "B": "On the metal bench by the fence.", + "C": "The pizza box is on the stone bench beside the fence.", + "D": "The pizza box is on the wooden picnic table near the gate, not on the bench beside the fence.", + "E": "You left the pizza box under the pergola beside the patio chairs, rather than on the stone bench." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_008-mcq_fixed_2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the pizza box now?", + "answer": "The pizza box is in the trash can by the roadside.", + "question_chinese": "披萨盒现在在哪个位置？", + "answer_chinese": "披萨盒在路边的垃圾桶里。", + "evidence": [ + { + "description": "At this moment, you threw the box into the trash can.", + "timestamps": { + "start": "00:08:36", + "end": "00:08:38" + } + }, + { + "description": "At this moment, the pizza can be seen on the stone bench beside the fence.", + "timestamps": { + "start": "00:04:06", + "end": "00:04:07" + } + } + ], + "question_timestamps": "00:11:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In your car’s back seat.", + "B": "On the kitchen counter.", + "C": "The pizza box is in the garden by the roadside, not inside the trash can there.", + "D": "The pizza box is in the trash can by the roadside.", + "E": "You put the pizza box in the oven by the roadside area instead of throwing it into the trash can." + }, + "answer_label": "D", + "answer_text": "The pizza box is in the trash can by the roadside.", + "q_id": "egolife_A1_D3_008-mcq_fixed_2", + "correct_answer": "The pizza box is in the trash can by the roadside.", + "original_options": { + "A": "In your car’s back seat.", + "B": "On the kitchen counter.", + "C": "The pizza box is in the garden by the roadside, not inside the trash can there.", + "D": "The pizza box is in the trash can by the roadside.", + "E": "You put the pizza box in the oven by the roadside area instead of throwing it into the trash can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 16, + 20 + ] + } +] diff --git a/QAs/egolife_A1_D3_009-mcq.json b/QAs/egolife_A1_D3_009-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3014ce1d3905f8ec9deb680232cbfb3ca1d14203 --- /dev/null +++ b/QAs/egolife_A1_D3_009-mcq.json @@ -0,0 +1,715 @@ +[ + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009-mcq_fixed_0", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the elderly man in the pavilion standing in relation to me?", + "answer": "He is now standing in front of you and to your left.", + "question_chinese": "亭子里大爷现在站在我的哪个位置？", + "answer_chinese": "他现在站在我的左前方。", + "evidence": [ + { + "description": "During this time, Grandpa was standing to my left front.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "He is now standing in front of you and to your left.", + "B": "In front of you and to your right.", + "C": "Behind you and to your left.", + "D": "He is slightly behind you on your right side, a few steps back from where you are standing.", + "E": "He is standing in front of you but off to your right, not in front-left relative to you." + }, + "answer_label": "A", + "answer_text": "He is now standing in front of you and to your left.", + "q_id": "egolife_A1_D3_009-mcq_fixed_0", + "correct_answer": "He is now standing in front of you and to your left.", + "original_options": { + "A": "He is now standing in front of you and to your left.", + "B": "In front of you and to your right.", + "C": "Behind you and to your left.", + "D": "He is slightly behind you on your right side, a few steps back from where you are standing.", + "E": "He is standing in front of you but off to your right, not in front-left relative to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009-mcq_fixed_1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the elderly man in the pavilion standing in relation to me?", + "answer": "Now he is behind you.", + "question_chinese": "亭子里大爷现在站在我的哪个位置？", + "answer_chinese": "现在在你的后边。", + "evidence": [ + { + "description": "During this time, Grandpa was standing to my left front.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:33" + } + }, + { + "description": "Now he is behind you.", + "timestamps": { + "start": "00:01:42", + "end": "00:01:43" + } + } + ], + "question_timestamps": "00:01:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Now he is in front of you.", + "B": "Now he is to your left.", + "C": "Now he is behind you.", + "D": "He is in front and a little to your left, close to the pavilion's edge where the garden path begins.", + "E": "He is a few steps to your right next to the wooden railing, not positioned behind you." + }, + "answer_label": "C", + "answer_text": "Now he is behind you.", + "q_id": "egolife_A1_D3_009-mcq_fixed_1", + "correct_answer": "Now he is behind you.", + "original_options": { + "A": "Now he is in front of you.", + "B": "Now he is to your left.", + "C": "Now he is behind you.", + "D": "He is in front and a little to your left, close to the pavilion's edge where the garden path begins.", + "E": "He is a few steps to your right next to the wooden railing, not positioned behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009-mcq_fixed_9", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many turns did I make from the pavilion entrance to the current bridge?", + "answer": "One.", + "question_chinese": "从亭子门口到现在的位置我一共拐了几个弯？", + "answer_chinese": "你一共拐了一个弯。", + "evidence": [ + { + "description": "Now, you have made the first turn.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:02:21", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "You made two turns from the pavilion entrance before reaching the current bridge.", + "D": "You made three turns along the way from the pavilion entrance to arrive at the bridge.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_D3_009-mcq_fixed_9", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "You made two turns from the pavilion entrance before reaching the current bridge.", + "D": "You made three turns along the way from the pavilion entrance to arrive at the bridge.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009_5_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From the pavilion entrance to my current position, how many turns have I made in total?", + "answer": "You made one turn in total.", + "question_chinese": "从亭子门口到现在的位置我一共拐了几个弯？", + "answer_chinese": "你一共拐了一个弯。", + "evidence": [ + { + "description": "Now, you have made the first turn.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:18" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You made zero turns in total.", + "B": "You made two turns in total.", + "C": "You made three turns in total along the way from the pavilion entrance to where you are now.", + "D": "You made four turns in total while walking from the pavilion entrance to your current position.", + "E": "You made one turn in total." + }, + "answer_label": "E", + "answer_text": "You made one turn in total.", + "q_id": "egolife_A1_D3_009_5_T2", + "correct_answer": "You made one turn in total.", + "original_options": { + "A": "You made zero turns in total.", + "B": "You made two turns in total.", + "C": "You made three turns in total along the way from the pavilion entrance to where you are now.", + "D": "You made four turns in total while walking from the pavilion entrance to your current position.", + "E": "You made one turn in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009_5_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From the pavilion entrance to my current position, how many turns have I made in total?", + "answer": "You made two turns in total.", + "question_chinese": "从亭子门口到现在的位置我一共拐了几个弯？", + "answer_chinese": "你一共拐了两个弯。", + "evidence": [ + { + "description": "Now, you have made the first turn.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:18" + } + }, + { + "description": "Now, you have made the second turn.", + "timestamps": { + "start": "00:02:23", + "end": "00:02:26" + } + } + ], + "question_timestamps": "00:03:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You made zero turns in total.", + "B": "You made one turn in total.", + "C": "You made three turns in total from the pavilion entrance to your current position.", + "D": "You made two turns in total.", + "E": "You made four turns in total during the route from the pavilion entrance to where you are now." + }, + "answer_label": "D", + "answer_text": "You made two turns in total.", + "q_id": "egolife_A1_D3_009_5_T1", + "correct_answer": "You made two turns in total.", + "original_options": { + "A": "You made zero turns in total.", + "B": "You made one turn in total.", + "C": "You made three turns in total from the pavilion entrance to your current position.", + "D": "You made two turns in total.", + "E": "You made four turns in total during the route from the pavilion entrance to where you are now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009_6_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the rockery relative to me?", + "answer": "The rockery is to your left now.", + "question_chinese": "假山在相对于我的什么位置？", + "answer_chinese": "假山现在在你的左边", + "evidence": [ + { + "description": "The rockery is directly opposite the lake.", + "timestamps": { + "start": "00:00:05", + "end": "00:01:27" + } + } + ], + "question_timestamps": "00:01:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The rockery is in front of you now.", + "B": "The rockery is behind you now.", + "C": "The rockery is to your right now, not on your left side at the moment.", + "D": "The rockery is above you now, rather than being positioned to either side of you.", + "E": "The rockery is to your left now." + }, + "answer_label": "E", + "answer_text": "The rockery is to your left now.", + "q_id": "egolife_A1_D3_009_6_T2", + "correct_answer": "The rockery is to your left now.", + "original_options": { + "A": "The rockery is in front of you now.", + "B": "The rockery is behind you now.", + "C": "The rockery is to your right now, not on your left side at the moment.", + "D": "The rockery is above you now, rather than being positioned to either side of you.", + "E": "The rockery is to your left now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009_6_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the rockery relative to me?", + "answer": "The rockery is behind you now.", + "question_chinese": "假山在相对于我的什么位置？", + "answer_chinese": "假山现在在你的后面。", + "evidence": [ + { + "description": "The rockery is behind you.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:24" + } + } + ], + "question_timestamps": "00:02:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The rockery is behind you now.", + "B": "The rockery is in front of you now.", + "C": "The rockery is to your left now.", + "D": "The rockery is to your right now, instead of being located behind your current position.", + "E": "The rockery is above you now, rather than sitting behind you along the path." + }, + "answer_label": "A", + "answer_text": "The rockery is behind you now.", + "q_id": "egolife_A1_D3_009_6_T1", + "correct_answer": "The rockery is behind you now.", + "original_options": { + "A": "The rockery is behind you now.", + "B": "The rockery is in front of you now.", + "C": "The rockery is to your left now.", + "D": "The rockery is to your right now, instead of being located behind your current position.", + "E": "The rockery is above you now, rather than sitting behind you along the path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009-mcq_fixed_7", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many streetlights did I pass from the pavilion entrance to this spot?", + "answer": "Three.", + "question_chinese": "刚刚从亭子门口到现在这个位置我一共经过了几个路灯？", + "answer_chinese": "你一共经过了两个路灯。", + "evidence": [ + { + "description": "Now, you have passed the first streetlight.", + "timestamps": { + "start": "00:02:02", + "end": "00:02:03" + } + }, + { + "description": "Now, you have passed the second streetlight.", + "timestamps": { + "start": "00:02:25", + "end": "00:02:26" + } + } + ], + "question_timestamps": "00:02:38", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Five streetlights in total from the pavilion entrance to this spot you reached at the end.", + "D": "Six streetlights in total that you passed between the pavilion entrance and where you are now.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_009-mcq_fixed_7", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Five streetlights in total from the pavilion entrance to this spot you reached at the end.", + "D": "Six streetlights in total that you passed between the pavilion entrance and where you are now.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009-mcq_fixed_6", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many streetlights did I pass from the pavilion entrance to this spot?", + "answer": "Four.", + "question_chinese": "刚刚从亭子门口到现在这个位置我一共经过了几个路灯？", + "answer_chinese": "你一共经过了三个路灯。", + "evidence": [ + { + "description": "Now, you have passed the first streetlight.", + "timestamps": { + "start": "00:02:02", + "end": "00:02:03" + } + }, + { + "description": "Now, you have passed the second streetlight.", + "timestamps": { + "start": "00:02:25", + "end": "00:02:26" + } + }, + { + "description": "Now, you have passed the first streetlight.", + "timestamps": { + "start": "00:02:42", + "end": "00:02:43" + } + } + ], + "question_timestamps": "00:04:15", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "Two streetlights in total that you passed from the pavilion entrance to this spot.", + "E": "Six streetlights in total along the route from the pavilion entrance to where you are now." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_009-mcq_fixed_6", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "Two streetlights in total that you passed from the pavilion entrance to this spot.", + "E": "Six streetlights in total along the route from the pavilion entrance to where you are now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009-mcq_fixed_4", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I stand on the stone pile by the bridge I just passed?", + "answer": "No, you didn't stand on it.", + "question_chinese": "刚刚经过的桥旁石堆我站上去了吗？", + "answer_chinese": "没有，你没有站上去。", + "evidence": [ + { + "description": "At this moment, the girl climbed up onto it, while you remained standing on the path.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:12" + } + } + ], + "question_timestamps": "00:07:26", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you stood on it.", + "B": "No, you didn't stand on it." + }, + "answer_label": "B", + "answer_text": "No, you didn't stand on it.", + "q_id": "egolife_A1_D3_009-mcq_fixed_4", + "correct_answer": "No, you didn't stand on it.", + "original_options": { + "A": "Yes, you stood on it.", + "B": "No, you didn't stand on it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009-mcq_fixed_5", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I stand on the stone pile by the bridge I just passed?", + "answer": "No.", + "question_chinese": "刚刚经过的桥旁石堆我站上去了吗？", + "answer_chinese": "没有。", + "evidence": [ + { + "description": "At this moment, the girl climbed up onto it, while you remained standing on the path.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:12" + } + } + ], + "question_timestamps": "00:09:56", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D3_009-mcq_fixed_5", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009-mcq_fixed_2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the pavilion from me now?", + "answer": "Now the pavilion is in front of you and to your right.", + "question_chinese": "现在亭子在我的哪个方向？", + "answer_chinese": "现在亭子在你的右前方。", + "evidence": [ + { + "description": "Now the pavilion is in front of you and to your right.", + "timestamps": { + "start": "00:09:45", + "end": "00:09:46" + } + } + ], + "question_timestamps": "00:09:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind you on the left.", + "B": "In front of you on the left.", + "C": "The pavilion is directly behind you, slightly offset toward your left side as you face forward.", + "D": "Now the pavilion is in front of you and to your right.", + "E": "The pavilion sits ahead of you but more toward your left, not on the right-hand side." + }, + "answer_label": "D", + "answer_text": "Now the pavilion is in front of you and to your right.", + "q_id": "egolife_A1_D3_009-mcq_fixed_2", + "correct_answer": "Now the pavilion is in front of you and to your right.", + "original_options": { + "A": "Behind you on the left.", + "B": "In front of you on the left.", + "C": "The pavilion is directly behind you, slightly offset toward your left side as you face forward.", + "D": "Now the pavilion is in front of you and to your right.", + "E": "The pavilion sits ahead of you but more toward your left, not on the right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_009-mcq_fixed_3", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the pavilion from me now?", + "answer": "The pavilion is now behind and to your right.", + "question_chinese": "现在亭子在我的哪个方向？", + "answer_chinese": "现在亭子在你的右后方。", + "evidence": [ + { + "description": "During this time, you walked past the pavilion, which was on your right side.", + "timestamps": { + "start": "00:09:49", + "end": "00:10:02" + } + }, + { + "description": "During this time, you have already passed the pavilion and are facing the front path, so the pavilion is behind and to your right.", + "timestamps": { + "start": "00:10:02", + "end": "00:10:08" + } + } + ], + "question_timestamps": "00:10:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The pavilion is now behind and to your right.", + "B": "In front of you to the right.", + "C": "Behind you to the left.", + "D": "The pavilion is straight ahead of you, positioned slightly to your right as you look forward.", + "E": "The pavilion is behind you but shifted toward your left, rather than being on your right side." + }, + "answer_label": "A", + "answer_text": "The pavilion is now behind and to your right.", + "q_id": "egolife_A1_D3_009-mcq_fixed_3", + "correct_answer": "The pavilion is now behind and to your right.", + "original_options": { + "A": "The pavilion is now behind and to your right.", + "B": "In front of you to the right.", + "C": "Behind you to the left.", + "D": "The pavilion is straight ahead of you, positioned slightly to your right as you look forward.", + "E": "The pavilion is behind you but shifted toward your left, rather than being on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D3_010-mcq.json b/QAs/egolife_A1_D3_010-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9d02f1ef0597fd014e9f84eddc8f8b8d6f3bb16b --- /dev/null +++ b/QAs/egolife_A1_D3_010-mcq.json @@ -0,0 +1,1602 @@ +[ + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_13_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was along the roadside that I just passed?", + "answer": "An outdoor parking lot and a small red kiosk.", + "question_chinese": "我刚刚经过的路边有什么？", + "answer_chinese": "一个室外停车场和一个红色小亭子", + "evidence": [ + { + "description": "There is a small outdoor parking lot there", + "timestamps": { + "start": "00:10:44", + "end": "00:11:06" + } + } + ], + "question_timestamps": "00:11:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A parking lot and a red gazebo.", + "B": "A parking lot and a blue kiosk.", + "C": "An outdoor parking lot and a small red kiosk.", + "D": "An outdoor parking lot with a small red shed along the roadside you just passed.", + "E": "An outdoor parking lot and a small red tent were along the roadside as you went by." + }, + "answer_label": "C", + "answer_text": "An outdoor parking lot and a small red kiosk.", + "q_id": "egolife_A1_D3_010_13_T1", + "correct_answer": "An outdoor parking lot and a small red kiosk.", + "original_options": { + "A": "A parking lot and a red gazebo.", + "B": "A parking lot and a blue kiosk.", + "C": "An outdoor parking lot and a small red kiosk.", + "D": "An outdoor parking lot with a small red shed along the roadside you just passed.", + "E": "An outdoor parking lot and a small red tent were along the roadside as you went by." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_13_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is there by the roadside I just passed?", + "answer": "An outdoor parking lot and a small red pavilion.", + "question_chinese": "我刚刚经过的路边有什么？", + "answer_chinese": "一个室外停车场和一个红色小亭子", + "evidence": [ + { + "description": "There is a small outdoor parking lot there", + "timestamps": { + "start": "00:10:44", + "end": "00:11:06" + } + } + ], + "question_timestamps": "00:11:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "An outdoor parking lot and a red kiosk.", + "B": "An outdoor parking lot and a red shed.", + "C": "An outdoor parking lot and a small red pavilion.", + "D": "An outdoor parking lot with a small red booth was set close to the roadside you just passed.", + "E": "An outdoor parking lot and a small red tent were right by the road as you went by." + }, + "answer_label": "C", + "answer_text": "An outdoor parking lot and a small red pavilion.", + "q_id": "egolife_A1_D3_010_13_T2", + "correct_answer": "An outdoor parking lot and a small red pavilion.", + "original_options": { + "A": "An outdoor parking lot and a red kiosk.", + "B": "An outdoor parking lot and a red shed.", + "C": "An outdoor parking lot and a small red pavilion.", + "D": "An outdoor parking lot with a small red booth was set close to the roadside you just passed.", + "E": "An outdoor parking lot and a small red tent were right by the road as you went by." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_5_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was along the roadside that I just passed?", + "answer": "An outdoor parking lot and a small red kiosk.", + "question_chinese": "我刚刚经过的路边有什么？", + "answer_chinese": "一个室外停车场和一个红色小亭子", + "evidence": [ + { + "description": "There is a small outdoor parking lot there", + "timestamps": { + "start": "00:10:44", + "end": "00:11:06" + } + } + ], + "question_timestamps": "00:11:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "An outdoor parking lot and a small red kiosk.", + "B": "A parking lot and a red booth.", + "C": "A parking lot and a red pavilion.", + "D": "An outdoor parking lot with a small red shed was along the roadside you passed a moment ago.", + "E": "An outdoor parking lot and a small red tent were positioned right off the road as you went by." + }, + "answer_label": "A", + "answer_text": "An outdoor parking lot and a small red kiosk.", + "q_id": "egolife_A1_D3_010_5_T1", + "correct_answer": "An outdoor parking lot and a small red kiosk.", + "original_options": { + "A": "An outdoor parking lot and a small red kiosk.", + "B": "A parking lot and a red booth.", + "C": "A parking lot and a red pavilion.", + "D": "An outdoor parking lot with a small red shed was along the roadside you passed a moment ago.", + "E": "An outdoor parking lot and a small red tent were positioned right off the road as you went by." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_5_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was by the roadside that I just passed?", + "answer": "An outdoor parking lot and a small red pavilion.", + "question_chinese": "我刚刚经过的路边有什么？", + "answer_chinese": "一个室外停车场和一个红色小亭子", + "evidence": [ + { + "description": "There is a small outdoor parking lot there", + "timestamps": { + "start": "00:10:44", + "end": "00:11:06" + } + } + ], + "question_timestamps": "00:11:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "An outdoor parking lot and a red kiosk.", + "B": "An outdoor parking lot and a red booth.", + "C": "An outdoor parking lot and a small red shed were by the roadside as you passed through the area.", + "D": "An outdoor parking lot and a small red pavilion.", + "E": "An outdoor parking lot with a small red tent stood near the roadside just before you moved on." + }, + "answer_label": "D", + "answer_text": "An outdoor parking lot and a small red pavilion.", + "q_id": "egolife_A1_D3_010_5_T2", + "correct_answer": "An outdoor parking lot and a small red pavilion.", + "original_options": { + "A": "An outdoor parking lot and a red kiosk.", + "B": "An outdoor parking lot and a red booth.", + "C": "An outdoor parking lot and a small red shed were by the roadside as you passed through the area.", + "D": "An outdoor parking lot and a small red pavilion.", + "E": "An outdoor parking lot with a small red tent stood near the roadside just before you moved on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_7_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I saw a car earlier—what direction is it from me now?", + "answer": "He was to your back-left.", + "question_chinese": "我刚才看到一辆车，它现在在我的什么方向？", + "answer_chinese": "他在我的左后方", + "evidence": [ + { + "description": "When I saw the car,it was behind me to the left", + "timestamps": { + "start": "00:03:01", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:04:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "He was to your back-right.", + "B": "He was to your front-left.", + "C": "He was to your back-left.", + "D": "He was positioned ahead of you and slightly to your right, not behind on the left.", + "E": "He was directly behind you rather than off to your back-left side as you looked forward." + }, + "answer_label": "C", + "answer_text": "He was to your back-left.", + "q_id": "egolife_A1_D3_010_7_T1", + "correct_answer": "He was to your back-left.", + "original_options": { + "A": "He was to your back-right.", + "B": "He was to your front-left.", + "C": "He was to your back-left.", + "D": "He was positioned ahead of you and slightly to your right, not behind on the left.", + "E": "He was directly behind you rather than off to your back-left side as you looked forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_7_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "I just saw a car; in which direction is it from me now?", + "answer": "It is behind you to the left.", + "question_chinese": "我刚才看到一辆车，它现在在我的什么方向？", + "answer_chinese": "他在我的左后方", + "evidence": [ + { + "description": "When I saw the car,it was behind me to the left", + "timestamps": { + "start": "00:03:01", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:04:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you to the left.", + "B": "It is behind you to the right.", + "C": "It is ahead of you to the left.", + "D": "It is in front of you and slightly to the right, not behind you on the left side.", + "E": "It is beside you on your left rather than behind you, so it is not back-left anymore." + }, + "answer_label": "A", + "answer_text": "It is behind you to the left.", + "q_id": "egolife_A1_D3_010_7_T2", + "correct_answer": "It is behind you to the left.", + "original_options": { + "A": "It is behind you to the left.", + "B": "It is behind you to the right.", + "C": "It is ahead of you to the left.", + "D": "It is in front of you and slightly to the right, not behind you on the left side.", + "E": "It is beside you on your left rather than behind you, so it is not back-left anymore." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_10_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, how many stone-paved paths did I see?", + "answer": "There was only one.", + "question_chinese": "我刚才看到了几条石头铺成的路？", + "answer_chinese": "只有一条", + "evidence": [ + { + "description": "There is a path paved with pebbles in front of me", + "timestamps": { + "start": "00:00:06", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:03:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There were none.", + "B": "There were two.", + "C": "You saw no stone-paved paths at all, just an unpaved stretch instead.", + "D": "You passed two separate stone-paved paths, one after the other, rather than a single one.", + "E": "There was only one." + }, + "answer_label": "E", + "answer_text": "There was only one.", + "q_id": "egolife_A1_D3_010_10_T1", + "correct_answer": "There was only one.", + "original_options": { + "A": "There were none.", + "B": "There were two.", + "C": "You saw no stone-paved paths at all, just an unpaved stretch instead.", + "D": "You passed two separate stone-paved paths, one after the other, rather than a single one.", + "E": "There was only one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_10_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stone-laid paths did I see?", + "answer": "Four.", + "question_chinese": "我刚才看到了几条石头铺成的路】？", + "answer_chinese": "只有一条", + "evidence": [ + { + "description": "There is a path paved with pebbles in front of me", + "timestamps": { + "start": "00:00:06", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:03:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Four.", + "C": "Three.", + "D": "You only saw two stone-laid paths, not a cluster of four different ones.", + "E": "You counted five stone-laid paths in total, including one more branching off to the side." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_010_10_T2", + "correct_answer": "Four.", + "original_options": { + "A": "One.", + "B": "Four.", + "C": "Three.", + "D": "You only saw two stone-laid paths, not a cluster of four different ones.", + "E": "You counted five stone-laid paths in total, including one more branching off to the side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_3", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I just saw a path made of stones.", + "answer": "You noticed one.", + "question_chinese": "我刚才看到了几条石头铺成的路】", + "answer_chinese": "只有一条", + "evidence": [ + { + "description": "There is a path paved with pebbles in front of me", + "timestamps": { + "start": "00:00:06", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:03:26", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You noticed none.", + "B": "You noticed two.", + "C": "You noticed three stone-made paths, not just a single stone path in view.", + "D": "You noticed one.", + "E": "You noticed four stone paths overall, with multiple segments appearing as you moved forward." + }, + "answer_label": "D", + "answer_text": "You noticed one.", + "q_id": "egolife_A1_D3_010_3", + "correct_answer": "You noticed one.", + "original_options": { + "A": "You noticed none.", + "B": "You noticed two.", + "C": "You noticed three stone-made paths, not just a single stone path in view.", + "D": "You noticed one.", + "E": "You noticed four stone paths overall, with multiple segments appearing as you moved forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_11_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Just now, how many stone-paved paths did I see?", + "answer": "Two.", + "question_chinese": "我刚才看到了几条石头铺成的路？", + "answer_chinese": "两条", + "evidence": [ + { + "description": "There is a path paved with stones", + "timestamps": { + "start": "00:00:08", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:09:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Zero.", + "C": "One.", + "D": "You saw three stone-paved paths, not the pair you were expecting to recall.", + "E": "You saw four stone-paved paths in total, spread across the area rather than just two." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "egolife_A1_D3_010_11_T1", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "Zero.", + "C": "One.", + "D": "You saw three stone-paved paths, not the pair you were expecting to recall.", + "E": "You saw four stone-paved paths in total, spread across the area rather than just two." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_6_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The car I saw earlier—what direction is it from me now?", + "answer": "It was to your back-right.", + "question_chinese": "我刚才看到的车，现在在我的什么方向？", + "answer_chinese": "它在我的右后方", + "evidence": [ + { + "description": "At first,the car was in front of me,when I turned left and went stright all the way.Now,the car is behind me on the right", + "timestamps": { + "start": "00:04:54", + "end": "00:05:04" + } + } + ], + "question_timestamps": "00:07:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was to your back-left.", + "B": "It was to your back-right.", + "C": "It was to your front-right.", + "D": "It was off your front-left, so you would have had to turn left to face it.", + "E": "It was directly behind you, centered, not angled toward your right side." + }, + "answer_label": "B", + "answer_text": "It was to your back-right.", + "q_id": "egolife_A1_D3_010_6_T1", + "correct_answer": "It was to your back-right.", + "original_options": { + "A": "It was to your back-left.", + "B": "It was to your back-right.", + "C": "It was to your front-right.", + "D": "It was off your front-left, so you would have had to turn left to face it.", + "E": "It was directly behind you, centered, not angled toward your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_6_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the car I just saw in now?", + "answer": "It is behind you to the right.", + "question_chinese": "我刚才看到的车，现在在我的什么方向？", + "answer_chinese": "它在我的右后方", + "evidence": [ + { + "description": "At first,the car was in front of me,when I turned left and went stright all the way.Now,the car is behind me on the right", + "timestamps": { + "start": "00:04:54", + "end": "00:05:04" + } + } + ], + "question_timestamps": "00:07:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you to the left.", + "B": "In front of you to the right.", + "C": "It is behind you to the right.", + "D": "It is directly behind you, lined up with your center rather than off to the right.", + "E": "It is in front of you to the left, so you would have to turn left to face it." + }, + "answer_label": "C", + "answer_text": "It is behind you to the right.", + "q_id": "egolife_A1_D3_010_6_T2", + "correct_answer": "It is behind you to the right.", + "original_options": { + "A": "Behind you to the left.", + "B": "In front of you to the right.", + "C": "It is behind you to the right.", + "D": "It is directly behind you, lined up with your center rather than off to the right.", + "E": "It is in front of you to the left, so you would have to turn left to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_11_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many roads paved with stones I have saw just now?", + "answer": "Three.", + "question_chinese": "我刚看到了几条石头铺成的路？", + "answer_chinese": "两条", + "evidence": [ + { + "description": "There is a path paved with stones", + "timestamps": { + "start": "00:00:08", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:09:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You saw five stone-paved roads just now, not merely three across the scene.", + "E": "You counted six roads paved with stones, with several branching off from the main way." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_010_11_T2", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You saw five stone-paved roads just now, not merely three across the scene.", + "E": "You counted six roads paved with stones, with several branching off from the main way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_4", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "I just saw a path made of stones.", + "answer": "Two.", + "question_chinese": "我刚看到了几条石头铺成的路", + "answer_chinese": "两条", + "evidence": [ + { + "description": "There is a path paved with stones", + "timestamps": { + "start": "00:00:08", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:09:29", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You just saw four stone-made paths, which is more than the number you’re recalling.", + "E": "You just saw five paths made of stones, spread out rather than only a couple." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "egolife_A1_D3_010_4", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You just saw four stone-made paths, which is more than the number you’re recalling.", + "E": "You just saw five paths made of stones, spread out rather than only a couple." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_12_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was looking at the map just now, what was around me?", + "answer": "There were your companions, trees, and cars.", + "question_chinese": "我刚看地图的时候，周围有什么？", + "answer_chinese": "有我的伙伴，有树，有车", + "evidence": [ + { + "description": "My friends and I are looking at the map", + "timestamps": { + "start": "00:02:16", + "end": "00:02:28" + } + }, + { + "description": "There are many trees and a car", + "timestamps": { + "start": "00:02:39", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:05:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Only trees and cars were around you.", + "B": "There were your companions, trees, and cars.", + "C": "Your companions and cars, no trees.", + "D": "You were surrounded by your companions and trees, but there weren’t any cars nearby.", + "E": "Around you were trees and parked cars, while your companions weren’t standing nearby." + }, + "answer_label": "B", + "answer_text": "There were your companions, trees, and cars.", + "q_id": "egolife_A1_D3_010_12_T1", + "correct_answer": "There were your companions, trees, and cars.", + "original_options": { + "A": "Only trees and cars were around you.", + "B": "There were your companions, trees, and cars.", + "C": "Your companions and cars, no trees.", + "D": "You were surrounded by your companions and trees, but there weren’t any cars nearby.", + "E": "Around you were trees and parked cars, while your companions weren’t standing nearby." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_12_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was looking at the map,what was around me?", + "answer": "Your friends,a carand the trees.", + "question_chinese": "我刚看地图的时候，周围有什么？", + "answer_chinese": "有我的伙伴，有树，有车", + "evidence": [ + { + "description": "My friends and I are looking at the map", + "timestamps": { + "start": "00:02:16", + "end": "00:02:28" + } + }, + { + "description": "There are many trees and a car", + "timestamps": { + "start": "00:02:39", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:05:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your friends and the trees only.", + "B": "Your friends,a carand the trees.", + "C": "Your friends and a car only.", + "D": "You had your friends with you, but there wasn’t a car nearby—just trees around the area.", + "E": "You were by a car and some trees, but your friends weren’t around you at that moment." + }, + "answer_label": "B", + "answer_text": "Your friends,a carand the trees.", + "q_id": "egolife_A1_D3_010_12_T2", + "correct_answer": "Your friends,a carand the trees.", + "original_options": { + "A": "Your friends and the trees only.", + "B": "Your friends,a carand the trees.", + "C": "Your friends and a car only.", + "D": "You had your friends with you, but there wasn’t a car nearby—just trees around the area.", + "E": "You were by a car and some trees, but your friends weren’t around you at that moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_4_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was looking at the map just now, what was around me?", + "answer": "There were your companions, trees, and cars.", + "question_chinese": "我刚看地图的时候，周围有什么？", + "answer_chinese": "有我的伙伴，有树，有车", + "evidence": [ + { + "description": "My friends and I are looking at the map", + "timestamps": { + "start": "00:02:16", + "end": "00:02:28" + } + }, + { + "description": "There are many trees and a car", + "timestamps": { + "start": "00:02:39", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:05:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Only trees and cars were around you.", + "B": "Your companions and trees, no cars.", + "C": "You were with your companions and cars, but there weren’t any trees around you.", + "D": "Around you were trees and cars, while your companions weren’t close by at that time.", + "E": "There were your companions, trees, and cars." + }, + "answer_label": "E", + "answer_text": "There were your companions, trees, and cars.", + "q_id": "egolife_A1_D3_010_4_T1", + "correct_answer": "There were your companions, trees, and cars.", + "original_options": { + "A": "Only trees and cars were around you.", + "B": "Your companions and trees, no cars.", + "C": "You were with your companions and cars, but there weren’t any trees around you.", + "D": "Around you were trees and cars, while your companions weren’t close by at that time.", + "E": "There were your companions, trees, and cars." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_4_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was around me when I just looked at the map?", + "answer": "Your friends,a carand the trees.", + "question_chinese": "我刚看地图的时候，周围有什么？", + "answer_chinese": "有我的伙伴，有树，有车", + "evidence": [ + { + "description": "My friends and I are looking at the map", + "timestamps": { + "start": "00:02:16", + "end": "00:02:28" + } + }, + { + "description": "There are many trees and a car", + "timestamps": { + "start": "00:02:39", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:05:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your friends,a carand the trees.", + "B": "Your friends and trees only.", + "C": "Your friends and a car only.", + "D": "You were with your friends and a bus nearby, with trees around instead of a car.", + "E": "You had your friends with you, but you were near fences rather than trees by the car." + }, + "answer_label": "A", + "answer_text": "Your friends,a carand the trees.", + "q_id": "egolife_A1_D3_010_4_T2", + "correct_answer": "Your friends,a carand the trees.", + "original_options": { + "A": "Your friends,a carand the trees.", + "B": "Your friends and trees only.", + "C": "Your friends and a car only.", + "D": "You were with your friends and a bus nearby, with trees around instead of a car.", + "E": "You had your friends with you, but you were near fences rather than trees by the car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_14_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was standing under the streetlight just now, was I closer to the car or to the map?", + "answer": "You were closer to the car.", + "question_chinese": "我刚站在路灯下的时候，是距离车更近，还是距离地图更近？", + "answer_chinese": "我距离车更近", + "evidence": [ + { + "description": "The map is about 8 to 9 meters in front of me,and the car is about 5 to 6 meters to my right,so the car is closer to me", + "timestamps": { + "start": "00:04:33", + "end": "00:04:52" + } + } + ], + "question_timestamps": "00:05:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were closer to the car.", + "B": "You were closer to the map." + }, + "answer_label": "A", + "answer_text": "You were closer to the car.", + "q_id": "egolife_A1_D3_010_14_T1", + "correct_answer": "You were closer to the car.", + "original_options": { + "A": "You were closer to the car.", + "B": "You were closer to the map." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_14_T2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I first stood under the street lamp,was I closer to the car or to the map?", + "answer": "You were closer to the car.", + "question_chinese": "我刚站在路灯下的时候，是距离车更近，还是距离地图更近？", + "answer_chinese": "我距离车更近", + "evidence": [ + { + "description": "The map is about 8 to 9 meters in front of me,and the car is about 5 to 6 meters to my right,so the car is closer to me", + "timestamps": { + "start": "00:04:33", + "end": "00:04:52" + } + } + ], + "question_timestamps": "00:05:51", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were closer to the car.", + "B": "You were closer to the map." + }, + "answer_label": "A", + "answer_text": "You were closer to the car.", + "q_id": "egolife_A1_D3_010_14_T2", + "correct_answer": "You were closer to the car.", + "original_options": { + "A": "You were closer to the car.", + "B": "You were closer to the map." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_8_T1", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was standing under the streetlight just now, was I closer to the car or to the map?", + "answer": "You were closer to the car.", + "question_chinese": "我刚站在路灯下的时候，是距离车更近，还是距离地图更近？", + "answer_chinese": "我距离车更近", + "evidence": [ + { + "description": "The map is about 8 to 9 meters in front of me,and the car is about 5 to 6 meters to my right,so the car is closer to me", + "timestamps": { + "start": "00:04:33", + "end": "00:04:52" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were closer to the car.", + "B": "You were closer to the map." + }, + "answer_label": "A", + "answer_text": "You were closer to the car.", + "q_id": "egolife_A1_D3_010_8_T1", + "correct_answer": "You were closer to the car.", + "original_options": { + "A": "You were closer to the car.", + "B": "You were closer to the map." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_8_T2", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was just standing under the streetlight, was I closer to the car or closer to the map?", + "answer": "You were closer to the car.", + "question_chinese": "我刚站在路灯下的时候，是距离车更近，还是距离地图更近？", + "answer_chinese": "我距离车更近", + "evidence": [ + { + "description": "The map is about 8 to 9 meters in front of me,and the car is about 5 to 6 meters to my right,so the car is closer to me", + "timestamps": { + "start": "00:04:33", + "end": "00:04:52" + } + } + ], + "question_timestamps": "00:06:01", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were closer to the car.", + "B": "You were closer to the map." + }, + "answer_label": "A", + "answer_text": "You were closer to the car.", + "q_id": "egolife_A1_D3_010_8_T2", + "correct_answer": "You were closer to the car.", + "original_options": { + "A": "You were closer to the car.", + "B": "You were closer to the map." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_15_T1", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, am I closer to the map or to the car?", + "answer": "You were closer to the map.", + "question_chinese": "现在我是距离地图更近还是车更近？", + "answer_chinese": "我距离地图更近", + "evidence": [ + { + "description": "At this moment,they are all behind me,and I am closer to the map", + "timestamps": { + "start": "00:05:13", + "end": "00:05:18" + } + }, + { + "description": "During this period,I kept going straight", + "timestamps": { + "start": "00:06:06", + "end": "00:06:52" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were closer to the car.", + "B": "You were closer to the map." + }, + "answer_label": "B", + "answer_text": "You were closer to the map.", + "q_id": "egolife_A1_D3_010_15_T1", + "correct_answer": "You were closer to the map.", + "original_options": { + "A": "You were closer to the car.", + "B": "You were closer to the map." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_15_T2", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Now am I closer to the map or the car?", + "answer": "You are closer to the map.", + "question_chinese": "现在我是距离地图更近还是车更近？", + "answer_chinese": "我距离地图更近", + "evidence": [ + { + "description": "At this moment,they are all behind me,and I am closer to the map", + "timestamps": { + "start": "00:05:13", + "end": "00:05:18" + } + }, + { + "description": "During this period,I kept going straight", + "timestamps": { + "start": "00:06:06", + "end": "00:06:52" + } + } + ], + "question_timestamps": "00:07:31", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are closer to the car.", + "B": "You are closer to the map." + }, + "answer_label": "B", + "answer_text": "You are closer to the map.", + "q_id": "egolife_A1_D3_010_15_T2", + "correct_answer": "You are closer to the map.", + "original_options": { + "A": "You are closer to the car.", + "B": "You are closer to the map." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_9_T1", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, am I closer to the map or to the car?", + "answer": "You were closer to the map.", + "question_chinese": "现在我是距离地图更近还是车更近？", + "answer_chinese": "我距离地图更近", + "evidence": [ + { + "description": "At this moment,they are all behind me,and I am closer to the map", + "timestamps": { + "start": "00:05:13", + "end": "00:05:18" + } + }, + { + "description": "During this period,I kept going straight", + "timestamps": { + "start": "00:06:06", + "end": "00:06:52" + } + } + ], + "question_timestamps": "00:07:40", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were closer to the map.", + "B": "You were closer to the car." + }, + "answer_label": "A", + "answer_text": "You were closer to the map.", + "q_id": "egolife_A1_D3_010_9_T1", + "correct_answer": "You were closer to the map.", + "original_options": { + "A": "You were closer to the map.", + "B": "You were closer to the car." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_9_T2", + "qid_num": 26, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the map now or to the car?", + "answer": "You are closer to the map.", + "question_chinese": "现在我是距离地图更近还是车更近？", + "answer_chinese": "我距离地图更近", + "evidence": [ + { + "description": "At this moment,they are all behind me,and I am closer to the map", + "timestamps": { + "start": "00:05:13", + "end": "00:05:18" + } + }, + { + "description": "During this period,I kept going straight", + "timestamps": { + "start": "00:06:06", + "end": "00:06:52" + } + } + ], + "question_timestamps": "00:07:41", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are closer to the car.", + "B": "You are closer to the map." + }, + "answer_label": "B", + "answer_text": "You are closer to the map.", + "q_id": "egolife_A1_D3_010_9_T2", + "correct_answer": "You are closer to the map.", + "original_options": { + "A": "You are closer to the car.", + "B": "You are closer to the map." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_2_T1", + "qid_num": 27, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to now, how many intersections have I passed?", + "answer": "You have passed four intersections.", + "question_chinese": "目前为止，我经历了几个岔路口？", + "answer_chinese": "我经历了四个岔路口", + "evidence": [ + { + "description": "I am coming out of a crossroads", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "When faced with two paths I had never taken beore,I chose the smoother one", + "timestamps": { + "start": "00:00:29", + "end": "00:00:41" + } + }, + { + "description": "Facing this crossroads,I chose the left one", + "timestamps": { + "start": "00:00:58", + "end": "00:01:12" + } + }, + { + "description": "I walked out of this fork in the road", + "timestamps": { + "start": "00:01:24", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:04:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have passed three intersections.", + "B": "You have passed five intersections.", + "C": "You have passed four intersections.", + "D": "Up to now, you have passed two intersections along your route so far.", + "E": "So far, you have passed six intersections as you kept moving forward." + }, + "answer_label": "C", + "answer_text": "You have passed four intersections.", + "q_id": "egolife_A1_D3_010_2_T1", + "correct_answer": "You have passed four intersections.", + "original_options": { + "A": "You have passed three intersections.", + "B": "You have passed five intersections.", + "C": "You have passed four intersections.", + "D": "Up to now, you have passed two intersections along your route so far.", + "E": "So far, you have passed six intersections as you kept moving forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_1_T1", + "qid_num": 28, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to now, how many intersections have I passed?", + "answer": "Up to now, you have passed eight intersections.", + "question_chinese": "目前为止，我经历了几个岔路口？", + "answer_chinese": "到现在，我经历了八个岔路口", + "evidence": [ + { + "description": "I chose the intersection on the right", + "timestamps": { + "start": "00:06:41", + "end": "00:07:05" + } + }, + { + "description": "This time,I chose the path on the left", + "timestamps": { + "start": "00:07:55", + "end": "00:08:06" + } + }, + { + "description": "This time,I still chose the one on the left", + "timestamps": { + "start": "00:08:22", + "end": "00:08:39" + } + }, + { + "description": "After hesitation,I took the road on the right", + "timestamps": { + "start": "00:09:07", + "end": "00:09:19" + } + } + ], + "question_timestamps": "00:10:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have passed seven intersections.", + "B": "Up to now, you have passed eight intersections.", + "C": "You have passed nine intersections.", + "D": "Up to now, you have passed six intersections on your way so far.", + "E": "So far, you have passed ten intersections as you continued along the route." + }, + "answer_label": "B", + "answer_text": "Up to now, you have passed eight intersections.", + "q_id": "egolife_A1_D3_010_1_T1", + "correct_answer": "Up to now, you have passed eight intersections.", + "original_options": { + "A": "You have passed seven intersections.", + "B": "Up to now, you have passed eight intersections.", + "C": "You have passed nine intersections.", + "D": "Up to now, you have passed six intersections on your way so far.", + "E": "So far, you have passed ten intersections as you continued along the route." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_010_1_T2", + "qid_num": 29, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "So far, I have encountered several crossroads?", + "answer": "Up to now,l have came through eight forks in the road.", + "question_chinese": "目前为止，我经历了几个岔路口？", + "answer_chinese": "到现在，我经历了八个岔路口", + "evidence": [ + { + "description": "I chose the intersection on the right", + "timestamps": { + "start": "00:06:41", + "end": "00:07:05" + } + }, + { + "description": "This time,I chose the path on the left", + "timestamps": { + "start": "00:07:55", + "end": "00:08:06" + } + }, + { + "description": "This time,I still chose the one on the left", + "timestamps": { + "start": "00:08:22", + "end": "00:08:39" + } + }, + { + "description": "After hesitation,I took the road on the right", + "timestamps": { + "start": "00:09:07", + "end": "00:09:19" + } + } + ], + "question_timestamps": "00:10:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Came through seven forks.", + "B": "Came through nine forks.", + "C": "Up to now, I have came through six forks in the road so far.", + "D": "Up to now,l have came through eight forks in the road.", + "E": "Up to now, I have came through ten forks in the road during your route." + }, + "answer_label": "D", + "answer_text": "Up to now,l have came through eight forks in the road.", + "q_id": "egolife_A1_D3_010_1_T2", + "correct_answer": "Up to now,l have came through eight forks in the road.", + "original_options": { + "A": "Came through seven forks.", + "B": "Came through nine forks.", + "C": "Up to now, I have came through six forks in the road so far.", + "D": "Up to now,l have came through eight forks in the road.", + "E": "Up to now, I have came through ten forks in the road during your route." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D3_011-mcq.json b/QAs/egolife_A1_D3_011-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6e1a2d145cbebb9513e1799a795a15328502787a --- /dev/null +++ b/QAs/egolife_A1_D3_011-mcq.json @@ -0,0 +1,749 @@ +[ + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_8", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is around the place where I just stopped?", + "answer": "There were my companions, stones, and shrub hedges.", + "question_chinese": "刚刚我停留的地方我周围有什么？", + "answer_chinese": "有我的伙伴，石头和灌木绿篱", + "evidence": [ + { + "description": "There are my companions, stones, and shrub hedges.", + "timestamps": { + "start": "00:01:09", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:04:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There were companions and fences nearby.", + "B": "There were stones and grass clumps.", + "C": "There were my companions, stones, and shrub hedges.", + "D": "There were your companions around you, with boulders nearby and patches of grass clumps.", + "E": "There were your buddies near where you stopped, with fences nearby and a few daisies around." + }, + "answer_label": "C", + "answer_text": "There were my companions, stones, and shrub hedges.", + "q_id": "egolife_A1_D3_011_8", + "correct_answer": "There were my companions, stones, and shrub hedges.", + "original_options": { + "A": "There were companions and fences nearby.", + "B": "There were stones and grass clumps.", + "C": "There were my companions, stones, and shrub hedges.", + "D": "There were your companions around you, with boulders nearby and patches of grass clumps.", + "E": "There were your buddies near where you stopped, with fences nearby and a few daisies around." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_7", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is around the place where I just stopped?", + "answer": "There were my companions, bicycles, and trees.", + "question_chinese": "刚刚我停留的地方我周围有什么？", + "answer_chinese": "有我的伙伴，自行车和树", + "evidence": [ + { + "description": "There are my companions, bicycles, and trees.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:18" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There were stones and shrubs.", + "B": "There were my companions, bicycles, and trees.", + "C": "There were only my companions.", + "D": "There were bicycles and trees around, but none of my companions were visible nearby.", + "E": "There were my companions and bicycles around where you stopped, but there were no trees in sight." + }, + "answer_label": "B", + "answer_text": "There were my companions, bicycles, and trees.", + "q_id": "egolife_A1_D3_011_7", + "correct_answer": "There were my companions, bicycles, and trees.", + "original_options": { + "A": "There were stones and shrubs.", + "B": "There were my companions, bicycles, and trees.", + "C": "There were only my companions.", + "D": "There were bicycles and trees around, but none of my companions were visible nearby.", + "E": "There were my companions and bicycles around where you stopped, but there were no trees in sight." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding in my hand?", + "answer": "Bags, data cables, and power banks.", + "question_chinese": "刚刚我手上拿的是什么？", + "answer_chinese": "包，数据线和充电宝", + "evidence": [ + { + "description": "I was holding a bag in my hand, and then I took out the data cable from it.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:10" + } + }, + { + "description": "I took out my power bank from my bag again.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A wallet and keys.", + "B": "Books and notebooks.", + "C": "You were holding a laptop with its charger, plus headphones bundled together in your hand.", + "D": "Bags, data cables, and power banks.", + "E": "In your hand you had a stack of papers and a tablet, along with a coffee mug." + }, + "answer_label": "D", + "answer_text": "Bags, data cables, and power banks.", + "q_id": "egolife_A1_D3_011_2", + "correct_answer": "Bags, data cables, and power banks.", + "original_options": { + "A": "A wallet and keys.", + "B": "Books and notebooks.", + "C": "You were holding a laptop with its charger, plus headphones bundled together in your hand.", + "D": "Bags, data cables, and power banks.", + "E": "In your hand you had a stack of papers and a tablet, along with a coffee mug." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding in my hand?", + "answer": "My phone.", + "question_chinese": "刚刚我手上拿的是什么？", + "answer_chinese": "手机", + "evidence": [ + { + "description": "I was looking at the navigation on my phone.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:11" + } + } + ], + "question_timestamps": "00:10:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "My wallet.", + "B": "My keys.", + "C": "I was holding a set of keys in my hand, not looking at any screen.", + "D": "My phone.", + "E": "I had a small wallet in my hand and kept it close while walking." + }, + "answer_label": "D", + "answer_text": "My phone.", + "q_id": "egolife_A1_D3_011_1", + "correct_answer": "My phone.", + "original_options": { + "A": "My wallet.", + "B": "My keys.", + "C": "I was holding a set of keys in my hand, not looking at any screen.", + "D": "My phone.", + "E": "I had a small wallet in my hand and kept it close while walking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_3", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the river I just saw?", + "answer": "On my right.", + "question_chinese": "刚刚看到的江在我的什么方向？", + "answer_chinese": "在我的右边", + "evidence": [ + { + "description": "At this moment, the river is on my right, and I continue walking straight ahead.", + "timestamps": { + "start": "00:06:30", + "end": "00:06:36" + } + } + ], + "question_timestamps": "00:08:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On my left.", + "B": "Straight ahead.", + "C": "It was slightly behind you, closer to your left side as you looked around.", + "D": "You had already passed it, so the river was behind you rather than on your right.", + "E": "On my right." + }, + "answer_label": "E", + "answer_text": "On my right.", + "q_id": "egolife_A1_D3_011_3", + "correct_answer": "On my right.", + "original_options": { + "A": "On my left.", + "B": "Straight ahead.", + "C": "It was slightly behind you, closer to your left side as you looked around.", + "D": "You had already passed it, so the river was behind you rather than on your right.", + "E": "On my right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the river I just saw?", + "answer": "It is behind me.", + "question_chinese": "刚刚看到的江在我的什么方向？", + "answer_chinese": "在我的后面", + "evidence": [ + { + "description": "At this moment, the river was on my right, and then I kept going straight.", + "timestamps": { + "start": "00:09:20", + "end": "00:09:40" + } + }, + { + "description": "I turned left and then kept going straight. Now the river is behind me.", + "timestamps": { + "start": "00:10:20", + "end": "00:10:26" + } + } + ], + "question_timestamps": "00:10:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It is in front of me.", + "B": "It is behind me.", + "C": "It is to my left.", + "D": "The river was off to your right side as you moved, not behind you.", + "E": "It was straight ahead of you, and you were walking toward it rather than away from it." + }, + "answer_label": "B", + "answer_text": "It is behind me.", + "q_id": "egolife_A1_D3_011_4", + "correct_answer": "It is behind me.", + "original_options": { + "A": "It is in front of me.", + "B": "It is behind me.", + "C": "It is to my left.", + "D": "The river was off to your right side as you moved, not behind you.", + "E": "It was straight ahead of you, and you were walking toward it rather than away from it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many of the trash cans I just passed by can I touch?", + "answer": "Two.", + "question_chinese": "刚刚经过的垃圾桶有几个是我可以碰到的？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "The first trash bin within my reach", + "timestamps": { + "start": "00:02:32", + "end": "00:02:35" + } + }, + { + "description": "The second trash bin within my reach", + "timestamps": { + "start": "00:03:32", + "end": "00:03:35" + } + } + ], + "question_timestamps": "00:05:41", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four of the trash cans were close enough that you could reach out and touch them.", + "E": "Five were within arm’s reach as you passed, so you could touch all five." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "egolife_A1_D3_011_5", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four of the trash cans were close enough that you could reach out and touch them.", + "E": "Five were within arm’s reach as you passed, so you could touch all five." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many of the trash cans I just passed by can I touch?", + "answer": "Three.", + "question_chinese": "刚刚经过的垃圾桶有几个是我可以碰到的？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "The first trash bin within my reach", + "timestamps": { + "start": "00:02:32", + "end": "00:02:35" + } + }, + { + "description": "The second trash bin within my reach", + "timestamps": { + "start": "00:03:32", + "end": "00:03:35" + } + }, + { + "description": "The third trash bin within my reach", + "timestamps": { + "start": "00:07:49", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:09:34", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Five of the trash cans were close enough for you to reach and touch as you walked by.", + "D": "Three.", + "E": "Seven were lined up within arm’s reach, so you could touch seven of them." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_011_6", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Five of the trash cans were close enough for you to reach and touch as you walked by.", + "D": "Three.", + "E": "Seven were lined up within arm’s reach, so you could touch seven of them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_6_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did my backpack just take?", + "answer": "Your friend was holding it for you.", + "question_chinese": "我的书包刚刚经历了什么路线？", + "answer_chinese": "你的朋友帮你拿着", + "evidence": [ + { + "description": "My schoolbag was taken by my partner first.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:44" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You carried it yourself.", + "B": "You set it down briefly.", + "C": "You picked up the backpack yourself and kept it on you instead of letting your friend hold it.", + "D": "Your friend took it, then you put it through the security scanner before taking it back.", + "E": "Your friend was holding it for you." + }, + "answer_label": "E", + "answer_text": "Your friend was holding it for you.", + "q_id": "egolife_A1_D3_011_6_T2", + "correct_answer": "Your friend was holding it for you.", + "original_options": { + "A": "You carried it yourself.", + "B": "You set it down briefly.", + "C": "You picked up the backpack yourself and kept it on you instead of letting your friend hold it.", + "D": "Your friend took it, then you put it through the security scanner before taking it back.", + "E": "Your friend was holding it for you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_6_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did my backpack just take?", + "answer": "Your friend held it for you, and then gave it back to you.", + "question_chinese": "我的书包刚刚经历了什么路线？", + "answer_chinese": "你的朋友帮你拿着，然后又还给了你", + "evidence": [ + { + "description": "My schoolbag was taken by my partner first.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:44" + } + }, + { + "description": "Then she handed my schoolbag back to me.", + "timestamps": { + "start": "00:04:08", + "end": "00:04:11" + } + } + ], + "question_timestamps": "00:06:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your friend held it for you, and then gave it back to you.", + "B": "You kept it the whole time.", + "C": "You handed it to your friend.", + "D": "You took the backpack from your friend, then you carried it without handing it back again.", + "E": "You set it down briefly and picked it up again, without your friend ever taking it." + }, + "answer_label": "A", + "answer_text": "Your friend held it for you, and then gave it back to you.", + "q_id": "egolife_A1_D3_011_6_T1", + "correct_answer": "Your friend held it for you, and then gave it back to you.", + "original_options": { + "A": "Your friend held it for you, and then gave it back to you.", + "B": "You kept it the whole time.", + "C": "You handed it to your friend.", + "D": "You took the backpack from your friend, then you carried it without handing it back again.", + "E": "You set it down briefly and picked it up again, without your friend ever taking it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_5_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to now, how many turns have I made?", + "answer": "You made three turns.", + "question_chinese": "目前为止，我拐了几个弯？", + "answer_chinese": "你拐了三个弯。", + "evidence": [ + { + "description": "I made a left turn first.", + "timestamps": { + "start": "00:07:18", + "end": "00:07:20" + } + }, + { + "description": "After turning right, I made a left turn.", + "timestamps": { + "start": "00:07:21", + "end": "00:07:22" + } + } + ], + "question_timestamps": "00:08:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You made one turn.", + "B": "You made two turns.", + "C": "You made three turns.", + "D": "You made four turns in total while moving forward through the area and changing direction several times.", + "E": "You made five turns overall, including an extra change of direction near the end of your path." + }, + "answer_label": "C", + "answer_text": "You made three turns.", + "q_id": "egolife_A1_D3_011_5_T2", + "correct_answer": "You made three turns.", + "original_options": { + "A": "You made one turn.", + "B": "You made two turns.", + "C": "You made three turns.", + "D": "You made four turns in total while moving forward through the area and changing direction several times.", + "E": "You made five turns overall, including an extra change of direction near the end of your path." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_011_5_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to now, how many turns have I made?", + "answer": "You made four turns.", + "question_chinese": "目前为止，我拐了几个弯？", + "answer_chinese": "你拐了四个弯。", + "evidence": [ + { + "description": "I made a left turn first.", + "timestamps": { + "start": "00:07:18", + "end": "00:07:20" + } + }, + { + "description": "After turning right, I made a left turn.", + "timestamps": { + "start": "00:07:21", + "end": "00:07:22" + } + }, + { + "description": "I made a left turn.", + "timestamps": { + "start": "00:10:23", + "end": "00:10:25" + } + } + ], + "question_timestamps": "00:11:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You made two turns.", + "B": "You made four turns.", + "C": "You made three turns.", + "D": "You made five turns total while navigating, with one additional change of direction before stopping.", + "E": "You made six turns overall, counting a couple of quick direction changes along the way." + }, + "answer_label": "B", + "answer_text": "You made four turns.", + "q_id": "egolife_A1_D3_011_5_T1", + "correct_answer": "You made four turns.", + "original_options": { + "A": "You made two turns.", + "B": "You made four turns.", + "C": "You made three turns.", + "D": "You made five turns total while navigating, with one additional change of direction before stopping.", + "E": "You made six turns overall, counting a couple of quick direction changes along the way." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D3_012-mcq.json b/QAs/egolife_A1_D3_012-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..aa484708d2954d60b4d6bcb3372b48a9ee3dbf4e --- /dev/null +++ b/QAs/egolife_A1_D3_012-mcq.json @@ -0,0 +1,930 @@ +[ + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding in my hand?", + "answer": "The bag，the date cable and the power bank.", + "question_chinese": "刚刚我的手上拿的是什么？", + "answer_chinese": "包，数据线和充电宝", + "evidence": [ + { + "description": "I'm carrying my schoolbag in my hand.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:10" + } + }, + { + "description": "I took the power bank out of my bag.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The bag and the power bank.", + "B": "The bag，the date cable and the power bank.", + "C": "Only the date cable.", + "D": "The bag, the USB cable, and a wall charger were all in your hand at the same time.", + "E": "You were holding the bag along with the power bank and a set of keys, not the date cable." + }, + "answer_label": "B", + "answer_text": "The bag，the date cable and the power bank.", + "q_id": "egolife_A1_D3_012_2", + "correct_answer": "The bag，the date cable and the power bank.", + "original_options": { + "A": "The bag and the power bank.", + "B": "The bag，the date cable and the power bank.", + "C": "Only the date cable.", + "D": "The bag, the USB cable, and a wall charger were all in your hand at the same time.", + "E": "You were holding the bag along with the power bank and a set of keys, not the date cable." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding in my hand?", + "answer": "I was just holding my phone.", + "question_chinese": "刚刚我的手上拿的是什么？", + "answer_chinese": "拿着手机", + "evidence": [ + { + "description": "I'm looking at the navigation on my phone.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:11" + } + } + ], + "question_timestamps": "00:10:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I was just holding my tablet.", + "B": "I was just holding my glasses.", + "C": "I was just holding my book in my hand as I looked down at it.", + "D": "I was just holding my phone.", + "E": "I was just holding a phone case in my hand, not the phone itself." + }, + "answer_label": "D", + "answer_text": "I was just holding my phone.", + "q_id": "egolife_A1_D3_012_1", + "correct_answer": "I was just holding my phone.", + "original_options": { + "A": "I was just holding my tablet.", + "B": "I was just holding my glasses.", + "C": "I was just holding my book in my hand as I looked down at it.", + "D": "I was just holding my phone.", + "E": "I was just holding a phone case in my hand, not the phone itself." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_4", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the river I just saw from my current location?", + "answer": "It was on my right.", + "question_chinese": "刚刚看到的江现在在我的什么方向？", + "answer_chinese": "在我的右边", + "evidence": [ + { + "description": "At this moment, the river is on my right, and I am continuing to move straight ahead.", + "timestamps": { + "start": "00:06:30", + "end": "00:06:36" + } + } + ], + "question_timestamps": "00:08:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was on my left.", + "B": "It was behind me.", + "C": "It was on my right.", + "D": "It was straight ahead of me, visible past the edge of the path as I looked forward.", + "E": "It was on my left, running alongside the area I had just walked through." + }, + "answer_label": "C", + "answer_text": "It was on my right.", + "q_id": "egolife_A1_D3_012_4", + "correct_answer": "It was on my right.", + "original_options": { + "A": "It was on my left.", + "B": "It was behind me.", + "C": "It was on my right.", + "D": "It was straight ahead of me, visible past the edge of the path as I looked forward.", + "E": "It was on my left, running alongside the area I had just walked through." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_3", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the river I just saw from my current location?", + "answer": "It was behind me.", + "question_chinese": "刚刚看到的江现在在我的什么方向？", + "answer_chinese": "在我的后面", + "evidence": [ + { + "description": "At this moment, the river is on my right, and I have been moving straight ahead.", + "timestamps": { + "start": "00:09:20", + "end": "00:09:40" + } + }, + { + "description": "I turned left, then kept going straight, and now the river is behind me.", + "timestamps": { + "start": "00:10:20", + "end": "00:10:26" + } + } + ], + "question_timestamps": "00:10:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was behind me.", + "B": "It was in front of you.", + "C": "It was to your left.", + "D": "It was to your right, just past the narrow dirt road from where you were standing.", + "E": "It was off to your left, partly hidden by trees as you looked away from the road." + }, + "answer_label": "A", + "answer_text": "It was behind me.", + "q_id": "egolife_A1_D3_012_3", + "correct_answer": "It was behind me.", + "original_options": { + "A": "It was behind me.", + "B": "It was in front of you.", + "C": "It was to your left.", + "D": "It was to your right, just past the narrow dirt road from where you were standing.", + "E": "It was off to your left, partly hidden by trees as you looked away from the road." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_2_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the “Stormy Surge” I just saw from me now?", + "answer": "On your right-hand side.", + "question_chinese": "刚刚看到的“风起云涌”现在在我的什么方向？", + "answer_chinese": "在你的右手边", + "evidence": [ + { + "description": "The signboard **Feng Qi Yun Yong** is on the right-hand side when entering the door.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:12" + } + } + ], + "question_timestamps": "00:08:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your left-hand side.", + "B": "Directly in front of you.", + "C": "On your right-hand side.", + "D": "Behind you on your left-hand side, so you would need to turn around to see it again.", + "E": "Straight ahead of you, positioned slightly left of center from where you are now." + }, + "answer_label": "C", + "answer_text": "On your right-hand side.", + "q_id": "egolife_A1_D3_012_2_T1", + "correct_answer": "On your right-hand side.", + "original_options": { + "A": "On your left-hand side.", + "B": "Directly in front of you.", + "C": "On your right-hand side.", + "D": "Behind you on your left-hand side, so you would need to turn around to see it again.", + "E": "Straight ahead of you, positioned slightly left of center from where you are now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_2_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the “Stormy Surge” I just saw from me now?", + "answer": "To your back-right.", + "question_chinese": "刚刚看到的“风起云涌”现在在我的什么方向？", + "answer_chinese": "在你的右后方", + "evidence": [ + { + "description": "The signboard **Feng Qi Yun Yong** is on the right-hand side when entering the door.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:12" + } + } + ], + "question_timestamps": "00:08:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-left.", + "B": "To your back-right.", + "C": "To your front-right.", + "D": "It’s directly behind you from where you’re standing now, not off to the right.", + "E": "It’s on your right side but slightly ahead of you, rather than behind you." + }, + "answer_label": "B", + "answer_text": "To your back-right.", + "q_id": "egolife_A1_D3_012_2_T2", + "correct_answer": "To your back-right.", + "original_options": { + "A": "To your back-left.", + "B": "To your back-right.", + "C": "To your front-right.", + "D": "It’s directly behind you from where you’re standing now, not off to the right.", + "E": "It’s on your right side but slightly ahead of you, rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_3_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the Stormy Surge I just saw from me now?", + "answer": "To your back-right.", + "question_chinese": "刚刚看到的风起云涌现在在我的什么方向？", + "answer_chinese": "在你的右后边", + "evidence": [ + { + "description": "The signboard **Feng Qi Yun Yong** is on the right-hand side when entering the door.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:12" + } + } + ], + "question_timestamps": "00:08:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-right.", + "B": "To your back-left.", + "C": "To your front-right.", + "D": "It’s straight behind you from your current position, not angled to the right.", + "E": "It’s off your right side in front of you, not behind your right shoulder." + }, + "answer_label": "A", + "answer_text": "To your back-right.", + "q_id": "egolife_A1_D3_012_3_T1", + "correct_answer": "To your back-right.", + "original_options": { + "A": "To your back-right.", + "B": "To your back-left.", + "C": "To your front-right.", + "D": "It’s straight behind you from your current position, not angled to the right.", + "E": "It’s off your right side in front of you, not behind your right shoulder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_3_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the Stormy Surge I just saw from me now?", + "answer": "Behind you.", + "question_chinese": "刚刚看到的风起云涌现在在我的什么方向？", + "answer_chinese": "在你的后边", + "evidence": [ + { + "description": "The signboard **Feng Qi Yun Yong** is on the right-hand side when entering the door.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:12" + } + } + ], + "question_timestamps": "00:08:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "Behind you.", + "D": "It was over to your right side, not directly behind you where you can’t see it.", + "E": "It was in front of you and slightly to the left, not behind your current viewpoint." + }, + "answer_label": "C", + "answer_text": "Behind you.", + "q_id": "egolife_A1_D3_012_3_T2", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "Behind you.", + "D": "It was over to your right side, not directly behind you where you can’t see it.", + "E": "It was in front of you and slightly to the left, not behind your current viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_6_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the place I just passed through, what was around me?", + "answer": "In front of you, there were your companions and trees.", + "question_chinese": "刚刚经过的地方我的周围有什么？", + "answer_chinese": "你的面前有你的伙伴们和树", + "evidence": [ + { + "description": "There are my friends and trees.", + "timestamps": { + "start": "00:03:34", + "end": "00:03:35" + } + } + ], + "question_timestamps": "00:03:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you were companions and trees.", + "B": "To your left were companions and trees.", + "C": "To your right, you had your companions and trees around you as you passed through.", + "D": "Above you, you could see trees and your companions nearby instead of them being in front.", + "E": "In front of you, there were your companions and trees." + }, + "answer_label": "E", + "answer_text": "In front of you, there were your companions and trees.", + "q_id": "egolife_A1_D3_012_6_T2", + "correct_answer": "In front of you, there were your companions and trees.", + "original_options": { + "A": "Behind you were companions and trees.", + "B": "To your left were companions and trees.", + "C": "To your right, you had your companions and trees around you as you passed through.", + "D": "Above you, you could see trees and your companions nearby instead of them being in front.", + "E": "In front of you, there were your companions and trees." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_6_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the place I just passed through, what was around me?", + "answer": "In front of you, there were your companions, bicycles, and trees.", + "question_chinese": "刚刚经过的地方我的周围有什么？", + "answer_chinese": "你的面前有你的伙伴们，自行车和树", + "evidence": [ + { + "description": "There are my friends and trees.", + "timestamps": { + "start": "00:03:34", + "end": "00:03:35" + } + }, + { + "description": "bicycles", + "timestamps": { + "start": "00:04:08", + "end": "00:04:09" + } + } + ], + "question_timestamps": "00:10:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you were companions, bicycles, and trees.", + "B": "To your left were companions, bicycles, and trees.", + "C": "To your right, there were your companions along with bicycles and trees as you moved past.", + "D": "In front of you, there were your companions, bicycles, and trees.", + "E": "Above you, there were your companions, bicycles, and trees rather than them being in front of you." + }, + "answer_label": "D", + "answer_text": "In front of you, there were your companions, bicycles, and trees.", + "q_id": "egolife_A1_D3_012_6_T1", + "correct_answer": "In front of you, there were your companions, bicycles, and trees.", + "original_options": { + "A": "Behind you were companions, bicycles, and trees.", + "B": "To your left were companions, bicycles, and trees.", + "C": "To your right, there were your companions along with bicycles and trees as you moved past.", + "D": "In front of you, there were your companions, bicycles, and trees.", + "E": "Above you, there were your companions, bicycles, and trees rather than them being in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_9_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the place I just walked past, what was around me?", + "answer": "There were your companions and trees.", + "question_chinese": "刚刚路过的地方我的周围有什么？", + "answer_chinese": "有你的伙伴们和树", + "evidence": [ + { + "description": "There are my friends and trees.", + "timestamps": { + "start": "00:03:34", + "end": "00:03:35" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Only your companions.", + "B": "There were your companions and trees.", + "C": "Only trees around you.", + "D": "There were your companions and a row of trees directly in front of you.", + "E": "You were surrounded by your companions and trees, with everyone clustered mostly behind you." + }, + "answer_label": "B", + "answer_text": "There were your companions and trees.", + "q_id": "egolife_A1_D3_012_9_T2", + "correct_answer": "There were your companions and trees.", + "original_options": { + "A": "Only your companions.", + "B": "There were your companions and trees.", + "C": "Only trees around you.", + "D": "There were your companions and a row of trees directly in front of you.", + "E": "You were surrounded by your companions and trees, with everyone clustered mostly behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_9_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At the place I just walked past, what was around me?", + "answer": "There were your companions, bicycles, and trees.", + "question_chinese": "刚刚路过的地方我的周围有什么？", + "answer_chinese": "有你的伙伴们，自行车和树", + "evidence": [ + { + "description": "There are my friends and trees.", + "timestamps": { + "start": "00:03:34", + "end": "00:03:35" + } + }, + { + "description": "bicycles", + "timestamps": { + "start": "00:04:08", + "end": "00:04:09" + } + } + ], + "question_timestamps": "00:10:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There were your companions, bicycles, and trees.", + "B": "There were companions and trees.", + "C": "There were bicycles and trees.", + "D": "There were your companions and bicycles, but no trees around you at the place you passed.", + "E": "In front of you, there were your companions and trees, rather than bicycles being there too." + }, + "answer_label": "A", + "answer_text": "There were your companions, bicycles, and trees.", + "q_id": "egolife_A1_D3_012_9_T1", + "correct_answer": "There were your companions, bicycles, and trees.", + "original_options": { + "A": "There were your companions, bicycles, and trees.", + "B": "There were companions and trees.", + "C": "There were bicycles and trees.", + "D": "There were your companions and bicycles, but no trees around you at the place you passed.", + "E": "In front of you, there were your companions and trees, rather than bicycles being there too." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_4_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Of the trash cans I just passed, how many were within my reach?", + "answer": "You could touch one trash can on the right side.", + "question_chinese": "我刚刚经过的垃圾桶有几个是我能碰到的？", + "answer_chinese": "你可以碰到一个在你右侧的垃圾桶", + "evidence": [ + { + "description": "The first trash bin within reach", + "timestamps": { + "start": "00:02:32", + "end": "00:02:36" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You could touch zero trash cans.", + "B": "You could touch two trash cans.", + "C": "You could touch one trash can on the right side.", + "D": "You could touch three trash cans on the right side as you went past them.", + "E": "You could touch four trash cans on the right side within arm’s reach at that moment." + }, + "answer_label": "C", + "answer_text": "You could touch one trash can on the right side.", + "q_id": "egolife_A1_D3_012_4_T2", + "correct_answer": "You could touch one trash can on the right side.", + "original_options": { + "A": "You could touch zero trash cans.", + "B": "You could touch two trash cans.", + "C": "You could touch one trash can on the right side.", + "D": "You could touch three trash cans on the right side as you went past them.", + "E": "You could touch four trash cans on the right side within arm’s reach at that moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_5_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Of the trash cans I just passed, how many were within my reach?", + "answer": "You could touch two trash cans on the right side.", + "question_chinese": "我刚刚经过的垃圾桶有几个是我能碰到的？", + "answer_chinese": "你可以碰到两个在你右侧的垃圾桶", + "evidence": [ + { + "description": "The first trash bin within reach", + "timestamps": { + "start": "00:02:32", + "end": "00:02:36" + } + }, + { + "description": "The second trash bin within reach", + "timestamps": { + "start": "00:03:32", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You could touch one trash can.", + "B": "You could touch three trash cans.", + "C": "You could touch one trash can on the right side as you passed the others out of reach.", + "D": "You could touch two trash cans on the right side.", + "E": "You could touch three trash cans on the right side, with each one close enough to reach." + }, + "answer_label": "D", + "answer_text": "You could touch two trash cans on the right side.", + "q_id": "egolife_A1_D3_012_5_T2", + "correct_answer": "You could touch two trash cans on the right side.", + "original_options": { + "A": "You could touch one trash can.", + "B": "You could touch three trash cans.", + "C": "You could touch one trash can on the right side as you passed the others out of reach.", + "D": "You could touch two trash cans on the right side.", + "E": "You could touch three trash cans on the right side, with each one close enough to reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_4_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Of the trash cans I just passed, how many were within my reach?", + "answer": "You could touch two trash cans on the right side.", + "question_chinese": "我刚刚经过的垃圾桶有几个是我能碰到的？", + "answer_chinese": "你可以碰到两个在你右侧的垃圾桶", + "evidence": [ + { + "description": "The first trash bin within reach", + "timestamps": { + "start": "00:02:32", + "end": "00:02:36" + } + }, + { + "description": "The second trash bin within reach", + "timestamps": { + "start": "00:03:32", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:06:42", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You could touch zero trash cans.", + "B": "You could touch two trash cans on the right side.", + "C": "You could touch one trash can.", + "D": "You could touch three trash cans on the right side, all close enough to reach as you walked by.", + "E": "You could touch zero trash cans on the right side because they stayed beyond arm’s reach." + }, + "answer_label": "B", + "answer_text": "You could touch two trash cans on the right side.", + "q_id": "egolife_A1_D3_012_4_T1", + "correct_answer": "You could touch two trash cans on the right side.", + "original_options": { + "A": "You could touch zero trash cans.", + "B": "You could touch two trash cans on the right side.", + "C": "You could touch one trash can.", + "D": "You could touch three trash cans on the right side, all close enough to reach as you walked by.", + "E": "You could touch zero trash cans on the right side because they stayed beyond arm’s reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_012_5_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Of the trash cans I just passed, how many were within my reach?", + "answer": "You could touch three trash cans on the right side.", + "question_chinese": "我刚刚经过的垃圾桶有几个是我能碰到的？", + "answer_chinese": "你可以碰到三个在你右侧的垃圾桶", + "evidence": [ + { + "description": "The first trash bin within reach", + "timestamps": { + "start": "00:02:32", + "end": "00:02:36" + } + }, + { + "description": "The second trash bin within reach", + "timestamps": { + "start": "00:03:32", + "end": "00:03:36" + } + }, + { + "description": "The third trash bin within reach", + "timestamps": { + "start": "00:07:49", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You could touch one trash can.", + "B": "You could touch two trash cans.", + "C": "You could touch three trash cans on the right side.", + "D": "You could touch four trash cans on the right side, with each one within arm’s reach.", + "E": "You could touch five trash cans on the right side as you passed along the row beside you." + }, + "answer_label": "C", + "answer_text": "You could touch three trash cans on the right side.", + "q_id": "egolife_A1_D3_012_5_T1", + "correct_answer": "You could touch three trash cans on the right side.", + "original_options": { + "A": "You could touch one trash can.", + "B": "You could touch two trash cans.", + "C": "You could touch three trash cans on the right side.", + "D": "You could touch four trash cans on the right side, with each one within arm’s reach.", + "E": "You could touch five trash cans on the right side as you passed along the row beside you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D3_013-mcq.json b/QAs/egolife_A1_D3_013-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8e1a777dac4e9a3d1838a1d841617e52897c97e3 --- /dev/null +++ b/QAs/egolife_A1_D3_013-mcq.json @@ -0,0 +1,1331 @@ +[ + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_5_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was just watching my friend play the drums, could I touch the drum he was playing?", + "answer": "Yes.", + "question_chinese": "刚刚看朋友打鼓时，我可以碰到他打的鼓吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "I can touch it, the drum is very close to me.", + "timestamps": { + "start": "00:03:22", + "end": "00:03:28" + } + } + ], + "question_timestamps": "00:03:40", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D3_013_5_T1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_5_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was just watching my friend play the drums, could I touch the drum he was playing?", + "answer": "No.", + "question_chinese": "刚刚看朋友打鼓时，我可以碰到他打的鼓吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "I can touch it, the drum is very close to me.", + "timestamps": { + "start": "00:03:22", + "end": "00:03:28" + } + }, + { + "description": "I can‘t touch it.", + "timestamps": { + "start": "00:03:51", + "end": "00:03:54" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D3_013_5_T2", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_7_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was just watching my friend play the drums, which direction from him was I?", + "answer": "To your left.", + "question_chinese": "刚刚看朋友打鼓的时候我在他的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "I was on his left", + "timestamps": { + "start": "00:03:16", + "end": "00:03:22" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your right, standing slightly off to the side while he played the drums.", + "D": "Directly in front of you, facing him as you watched him play the drums.", + "E": "To your left." + }, + "answer_label": "E", + "answer_text": "To your left.", + "q_id": "egolife_A1_D3_013_7_T2", + "correct_answer": "To your left.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your right, standing slightly off to the side while he played the drums.", + "D": "Directly in front of you, facing him as you watched him play the drums.", + "E": "To your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_7_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was just watching my friend play the drums, which direction from him was I?", + "answer": "Left and right.", + "question_chinese": "刚刚看朋友打鼓的时候我在他的什么方向？", + "answer_chinese": "左边和右边", + "evidence": [ + { + "description": "I was on his left", + "timestamps": { + "start": "00:03:16", + "end": "00:03:22" + } + }, + { + "description": "I was on his right", + "timestamps": { + "start": "00:05:03", + "end": "00:05:13" + } + } + ], + "question_timestamps": "00:07:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Left and left.", + "B": "Right and right.", + "C": "To your left, not switching sides while you watched him play the drums.", + "D": "Left and right.", + "E": "You were on his right side the whole time while watching him play the drums." + }, + "answer_label": "D", + "answer_text": "Left and right.", + "q_id": "egolife_A1_D3_013_7_T1", + "correct_answer": "Left and right.", + "original_options": { + "A": "Left and left.", + "B": "Right and right.", + "C": "To your left, not switching sides while you watched him play the drums.", + "D": "Left and right.", + "E": "You were on his right side the whole time while watching him play the drums." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_13_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was just watching my friend play the drums, what was around me?", + "answer": "There are trash cans and a dance arcade machine around you.", + "question_chinese": "刚刚看朋友打鼓的时候，我的周围有什么？", + "answer_chinese": "你周围有垃圾桶和舞蹈机", + "evidence": [ + { + "description": "There is a trash can by my left foot.", + "timestamps": { + "start": "00:03:29", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:03:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Only trash cans were around you.", + "B": "Only a dance arcade machine was nearby.", + "C": "There were trash cans and a claw machine around you, but no dance arcade machine beside you.", + "D": "There were trash cans and a dance arcade machine around you, along with a claw machine nearby.", + "E": "There are trash cans and a dance arcade machine around you." + }, + "answer_label": "E", + "answer_text": "There are trash cans and a dance arcade machine around you.", + "q_id": "egolife_A1_D3_013_13_T2", + "correct_answer": "There are trash cans and a dance arcade machine around you.", + "original_options": { + "A": "Only trash cans were around you.", + "B": "Only a dance arcade machine was nearby.", + "C": "There were trash cans and a claw machine around you, but no dance arcade machine beside you.", + "D": "There were trash cans and a dance arcade machine around you, along with a claw machine nearby.", + "E": "There are trash cans and a dance arcade machine around you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_9_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was just watching my friend play the drums, what was around me?", + "answer": "Around you, there was a trash can.", + "question_chinese": "刚刚看朋友打鼓的时候，我的周围有什么？", + "answer_chinese": "你周围有垃圾桶。", + "evidence": [ + { + "description": "There is a trash can by my left foot.", + "timestamps": { + "start": "00:03:29", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Around you, there was no trash can.", + "B": "Around you, there were two trash cans.", + "C": "Around you, there was a trash can.", + "D": "Around you, there was a trash can and a dance machine near where you were watching.", + "E": "Around you, there was a trash can plus a dance machine and a claw machine beside the drums." + }, + "answer_label": "C", + "answer_text": "Around you, there was a trash can.", + "q_id": "egolife_A1_D3_013_9_T2", + "correct_answer": "Around you, there was a trash can.", + "original_options": { + "A": "Around you, there was no trash can.", + "B": "Around you, there were two trash cans.", + "C": "Around you, there was a trash can.", + "D": "Around you, there was a trash can and a dance machine near where you were watching.", + "E": "Around you, there was a trash can plus a dance machine and a claw machine beside the drums." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_13_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was just watching my friend play the drums, what was around me?", + "answer": "There are trash cans, a dance arcade machine, and a claw machine plush around you.", + "question_chinese": "刚刚看朋友打鼓的时候，我的周围有什么？", + "answer_chinese": "你周围有垃圾桶和舞蹈机和抓娃机娃", + "evidence": [ + { + "description": "There is a trash can by my left foot.", + "timestamps": { + "start": "00:03:29", + "end": "00:03:41" + } + }, + { + "description": "The claw machine is on the left, and the dancing machine is on the right.", + "timestamps": { + "start": "00:04:11", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:06:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There are trash cans, a dance arcade machine, and a claw machine plush around you.", + "B": "Around you, there was a trash can.", + "C": "Around you, there were yellow chairs.", + "D": "Around you, there were trash cans and a dance arcade machine around you while you watched the drumming.", + "E": "Around you, there were a trash can, a dance machine, and a claw machine close by." + }, + "answer_label": "A", + "answer_text": "There are trash cans, a dance arcade machine, and a claw machine plush around you.", + "q_id": "egolife_A1_D3_013_13_T1", + "correct_answer": "There are trash cans, a dance arcade machine, and a claw machine plush around you.", + "original_options": { + "A": "There are trash cans, a dance arcade machine, and a claw machine plush around you.", + "B": "Around you, there was a trash can.", + "C": "Around you, there were yellow chairs.", + "D": "Around you, there were trash cans and a dance arcade machine around you while you watched the drumming.", + "E": "Around you, there were a trash can, a dance machine, and a claw machine close by." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_9_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was just watching my friend play the drums, what was around me?", + "answer": "Around you, there were a trash can, a dance machine, and a claw machine.", + "question_chinese": "刚刚看朋友打鼓的时候，我的周围有什么？", + "answer_chinese": "你周围有垃圾桶和舞蹈机和抓娃娃机。", + "evidence": [ + { + "description": "There is a trash can by my left foot.", + "timestamps": { + "start": "00:03:29", + "end": "00:03:41" + } + }, + { + "description": "The claw machine is on the left, and the dancing machine is on the right.", + "timestamps": { + "start": "00:04:11", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:06:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Around you, there was a trash can.", + "B": "Around you, there were yellow chairs.", + "C": "Around you, there were trash cans and a dance arcade machine around you as you watched your friend.", + "D": "Around you, there were a trash can, a dance machine, and a claw machine.", + "E": "Around you, there were trash cans, a dance arcade machine, and a claw machine plush near you." + }, + "answer_label": "D", + "answer_text": "Around you, there were a trash can, a dance machine, and a claw machine.", + "q_id": "egolife_A1_D3_013_9_T1", + "correct_answer": "Around you, there were a trash can, a dance machine, and a claw machine.", + "original_options": { + "A": "Around you, there was a trash can.", + "B": "Around you, there were yellow chairs.", + "C": "Around you, there were trash cans and a dance arcade machine around you as you watched your friend.", + "D": "Around you, there were a trash can, a dance machine, and a claw machine.", + "E": "Around you, there were trash cans, a dance arcade machine, and a claw machine plush near you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_8_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I just walked in through the door, what was around me?", + "answer": "Around you, there were yellow chairs.", + "question_chinese": "刚刚进门的时候，我周围有什么？", + "answer_chinese": "你周围有黄色椅子", + "evidence": [ + { + "description": "There is a yellow chair in the front left.", + "timestamps": { + "start": "00:11:33", + "end": "00:11:35" + } + } + ], + "question_timestamps": "00:11:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Around you, there was a trash can.", + "B": "Around you, there were trash cans.", + "C": "Around you, there were blue chairs as you entered through the door.", + "D": "Around you, there were trash cans and a dance arcade machine positioned near the entrance.", + "E": "Around you, there were yellow chairs." + }, + "answer_label": "E", + "answer_text": "Around you, there were yellow chairs.", + "q_id": "egolife_A1_D3_013_8_T2", + "correct_answer": "Around you, there were yellow chairs.", + "original_options": { + "A": "Around you, there was a trash can.", + "B": "Around you, there were trash cans.", + "C": "Around you, there were blue chairs as you entered through the door.", + "D": "Around you, there were trash cans and a dance arcade machine positioned near the entrance.", + "E": "Around you, there were yellow chairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_12_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I just walked in through the door, what was around me?", + "answer": "There are yellow chairs around you.", + "question_chinese": "刚刚进门的时候，我周围有什么？", + "answer_chinese": "你周边有黄色椅子", + "evidence": [ + { + "description": "There is a yellow chair in the front left.", + "timestamps": { + "start": "00:11:35", + "end": "00:11:36" + } + } + ], + "question_timestamps": "00:11:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Around you, there were trash cans.", + "B": "Around you, there was a trash can.", + "C": "Around you, there were blue chairs and entertainment facilities around you just after you walked in.", + "D": "There are yellow chairs around you.", + "E": "Around you, there were trash cans and a dance arcade machine set up around you." + }, + "answer_label": "D", + "answer_text": "There are yellow chairs around you.", + "q_id": "egolife_A1_D3_013_12_T2", + "correct_answer": "There are yellow chairs around you.", + "original_options": { + "A": "Around you, there were trash cans.", + "B": "Around you, there was a trash can.", + "C": "Around you, there were blue chairs and entertainment facilities around you just after you walked in.", + "D": "There are yellow chairs around you.", + "E": "Around you, there were trash cans and a dance arcade machine set up around you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_8_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I just walked in through the door, what was around me?", + "answer": "Around you, there were yellow chairs and entertainment facilities.", + "question_chinese": "刚刚进门的时候，我周围有什么？", + "answer_chinese": "你周围有黄色椅子和娱乐设施", + "evidence": [ + { + "description": "There is a yellow chair in the front left.", + "timestamps": { + "start": "00:11:33", + "end": "00:11:35" + } + }, + { + "description": "There are many entertainment facilities around.", + "timestamps": { + "start": "00:11:41", + "end": "00:11:52" + } + } + ], + "question_timestamps": "00:11:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Around you, there were blue chairs.", + "B": "Around you, there were trash cans.", + "C": "Around you, there were yellow chairs and entertainment facilities.", + "D": "Around you, there were yellow chairs and vending machines as you came in the door.", + "E": "Around you, there were trash cans and a dance arcade machine nearby when you entered." + }, + "answer_label": "C", + "answer_text": "Around you, there were yellow chairs and entertainment facilities.", + "q_id": "egolife_A1_D3_013_8_T1", + "correct_answer": "Around you, there were yellow chairs and entertainment facilities.", + "original_options": { + "A": "Around you, there were blue chairs.", + "B": "Around you, there were trash cans.", + "C": "Around you, there were yellow chairs and entertainment facilities.", + "D": "Around you, there were yellow chairs and vending machines as you came in the door.", + "E": "Around you, there were trash cans and a dance arcade machine nearby when you entered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_12_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I just walked in through the door, what was around me?", + "answer": "There are yellow chairs and amusement facilities around you.", + "question_chinese": "刚刚进门的时候，我周围有什么？", + "answer_chinese": "你周边有黄色椅子和娱乐设施", + "evidence": [ + { + "description": "There is a yellow chair in the front left.", + "timestamps": { + "start": "00:11:35", + "end": "00:11:37" + } + }, + { + "description": "There are many entertainment facilities around.", + "timestamps": { + "start": "00:11:41", + "end": "00:11:52" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Around you, there were blue chairs.", + "B": "Around you, there was a trash can.", + "C": "Around you, there were yellow chairs and vending machines around you just inside the doorway.", + "D": "There are yellow chairs and amusement facilities around you.", + "E": "Around you, there were trash cans and a dance arcade machine around you when you walked in." + }, + "answer_label": "D", + "answer_text": "There are yellow chairs and amusement facilities around you.", + "q_id": "egolife_A1_D3_013_12_T1", + "correct_answer": "There are yellow chairs and amusement facilities around you.", + "original_options": { + "A": "Around you, there were blue chairs.", + "B": "Around you, there was a trash can.", + "C": "Around you, there were yellow chairs and vending machines around you just inside the doorway.", + "D": "There are yellow chairs and amusement facilities around you.", + "E": "Around you, there were trash cans and a dance arcade machine around you when you walked in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_9", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was around me when I just walked in?", + "answer": "There is a yellow chair and entertainment facilities.", + "question_chinese": "刚刚进门的时候，我周围有什么？", + "answer_chinese": "黄色椅子和娱乐设施", + "evidence": [ + { + "description": "There is a yellow chair in the front left.", + "timestamps": { + "start": "00:11:35", + "end": "00:11:37" + } + }, + { + "description": "There are many entertainment facilities around.", + "timestamps": { + "start": "00:11:41", + "end": "00:11:52" + } + } + ], + "question_timestamps": "00:12:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "There were blue chairs around you.", + "B": "There were trash cans around you.", + "C": "There was a yellow chair and a dance arcade machine around you when you first walked in.", + "D": "There were trash cans and a dance arcade machine around you right after you entered.", + "E": "There is a yellow chair and entertainment facilities." + }, + "answer_label": "E", + "answer_text": "There is a yellow chair and entertainment facilities.", + "q_id": "egolife_A1_D3_013_9", + "correct_answer": "There is a yellow chair and entertainment facilities.", + "original_options": { + "A": "There were blue chairs around you.", + "B": "There were trash cans around you.", + "C": "There was a yellow chair and a dance arcade machine around you when you first walked in.", + "D": "There were trash cans and a dance arcade machine around you right after you entered.", + "E": "There is a yellow chair and entertainment facilities." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_1_T2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From when I saw that friend in green clothes to when I accompanied her to the claw machines, what route have I taken?", + "answer": "I turned left first, then right, then left, then left.", + "question_chinese": "我从看到那个绿色衣服的朋友到陪她去抓娃娃，当前经历了什么路线？", + "answer_chinese": "我先左拐，再右拐，左拐，左拐", + "evidence": [ + { + "description": "I turned left first", + "timestamps": { + "start": "00:06:38", + "end": "00:06:45" + } + }, + { + "description": "Then I turned right", + "timestamps": { + "start": "00:06:46", + "end": "00:06:48" + } + }, + { + "description": "Then I turned left", + "timestamps": { + "start": "00:06:49", + "end": "00:06:57" + } + }, + { + "description": "Then I turned left", + "timestamps": { + "start": "00:06:57", + "end": "00:06:59" + } + } + ], + "question_timestamps": "00:07:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "I turned right, then left, then left.", + "B": "I turned left, then left, then right.", + "C": "I turned left first, then left, then right, then left, making a different sequence to reach them.", + "D": "I turned left first, then right, then left, then left.", + "E": "I turned right first, then left, then right, then left, before arriving at the claw machines." + }, + "answer_label": "D", + "answer_text": "I turned left first, then right, then left, then left.", + "q_id": "egolife_A1_D3_013_1_T2", + "correct_answer": "I turned left first, then right, then left, then left.", + "original_options": { + "A": "I turned right, then left, then left.", + "B": "I turned left, then left, then right.", + "C": "I turned left first, then left, then right, then left, making a different sequence to reach them.", + "D": "I turned left first, then right, then left, then left.", + "E": "I turned right first, then left, then right, then left, before arriving at the claw machines." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_1_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From when I saw that friend in green clothes to when I accompanied her to the claw machines, what route have I taken?", + "answer": "I turned left first, then right, then left, then left, and finally right.", + "question_chinese": "我从看到那个绿色衣服的朋友到陪她去抓娃娃，当前经历了什么路线？", + "answer_chinese": "我先左拐，再右拐，左拐，左拐，最后右拐", + "evidence": [ + { + "description": "I turned left first", + "timestamps": { + "start": "00:06:38", + "end": "00:06:45" + } + }, + { + "description": "Then I turned right", + "timestamps": { + "start": "00:06:46", + "end": "00:06:48" + } + }, + { + "description": "Then I turned left", + "timestamps": { + "start": "00:06:49", + "end": "00:06:57" + } + }, + { + "description": "Then I turned left", + "timestamps": { + "start": "00:06:57", + "end": "00:06:59" + } + }, + { + "description": "At last,I turned rigth", + "timestamps": { + "start": "00:07:02", + "end": "00:07:04" + } + } + ], + "question_timestamps": "00:08:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "I turned left, then right, then left.", + "B": "I turned right, then right, then left.", + "C": "I turned left first, then left, then left, then right, and only then went over to the claw machines.", + "D": "I turned right first, then left, then left, then right, and ended up at the machines afterward.", + "E": "I turned left first, then right, then left, then left, and finally right." + }, + "answer_label": "E", + "answer_text": "I turned left first, then right, then left, then left, and finally right.", + "q_id": "egolife_A1_D3_013_1_T1", + "correct_answer": "I turned left first, then right, then left, then left, and finally right.", + "original_options": { + "A": "I turned left, then right, then left.", + "B": "I turned right, then right, then left.", + "C": "I turned left first, then left, then left, then right, and only then went over to the claw machines.", + "D": "I turned right first, then left, then left, then right, and ended up at the machines afterward.", + "E": "I turned left first, then right, then left, then left, and finally right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_3_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many claw machines did I stop in front of just now?", + "answer": "You stopped in front of one claw machine.", + "question_chinese": "我刚刚在几个抓娃娃机面前做了停留？", + "answer_chinese": "你在一个抓娃娃机前做了停留", + "evidence": [ + { + "description": "I'm playing with this girl in green clothes to catch dolls.", + "timestamps": { + "start": "00:07:15", + "end": "00:07:28" + } + } + ], + "question_timestamps": "00:07:31", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You stopped in front of zero claw machines.", + "B": "You stopped in front of two claw machines.", + "C": "You stopped in front of one claw machine.", + "D": "You stopped in front of three claw machines before moving on to the next spot.", + "E": "You stopped in front of four claw machines in a row, pausing briefly at each one." + }, + "answer_label": "C", + "answer_text": "You stopped in front of one claw machine.", + "q_id": "egolife_A1_D3_013_3_T2", + "correct_answer": "You stopped in front of one claw machine.", + "original_options": { + "A": "You stopped in front of zero claw machines.", + "B": "You stopped in front of two claw machines.", + "C": "You stopped in front of one claw machine.", + "D": "You stopped in front of three claw machines before moving on to the next spot.", + "E": "You stopped in front of four claw machines in a row, pausing briefly at each one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_11_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many claw machines did I stop in front of just now?", + "answer": "You stopped in front of two claw machines.", + "question_chinese": "我刚刚在几个抓娃娃机面前做了停留？", + "answer_chinese": "你在两个抓娃娃机前做了停留", + "evidence": [ + { + "description": "I'm playing with this girl in green clothes to catch dolls.", + "timestamps": { + "start": "00:07:15", + "end": "00:07:28" + } + }, + { + "description": "I accompanied this friend to another claw machin.e", + "timestamps": { + "start": "00:09:55", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:10:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You stopped in front of one claw machine.", + "B": "You stopped in front of three claw machines.", + "C": "You stopped in front of zero claw machines and kept walking without pausing.", + "D": "You stopped in front of four claw machines, stopping briefly at each before continuing.", + "E": "You stopped in front of two claw machines." + }, + "answer_label": "E", + "answer_text": "You stopped in front of two claw machines.", + "q_id": "egolife_A1_D3_013_11_T2", + "correct_answer": "You stopped in front of two claw machines.", + "original_options": { + "A": "You stopped in front of one claw machine.", + "B": "You stopped in front of three claw machines.", + "C": "You stopped in front of zero claw machines and kept walking without pausing.", + "D": "You stopped in front of four claw machines, stopping briefly at each before continuing.", + "E": "You stopped in front of two claw machines." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_11_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many claw machines did I stop in front of just now?", + "answer": "You stopped in front of three claw machines.", + "question_chinese": "我刚刚在几个抓娃娃机面前做了停留？", + "answer_chinese": "你在三个抓娃娃机前做了停留", + "evidence": [ + { + "description": "I'm playing with this girl in green clothes to catch dolls.", + "timestamps": { + "start": "00:07:15", + "end": "00:07:28" + } + }, + { + "description": "I accompanied this friend to another claw machin.e", + "timestamps": { + "start": "00:09:55", + "end": "00:10:04" + } + }, + { + "description": "I am watching another friend catch a pachinko ball.", + "timestamps": { + "start": "00:10:21", + "end": "00:10:28" + } + } + ], + "question_timestamps": "00:11:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You stopped in front of two claw machines.", + "B": "You stopped in front of four claw machines.", + "C": "You stopped in front of one claw machine and then moved along without stopping again.", + "D": "You stopped in front of three claw machines.", + "E": "You stopped in front of five claw machines, pausing at each one before walking away." + }, + "answer_label": "D", + "answer_text": "You stopped in front of three claw machines.", + "q_id": "egolife_A1_D3_013_11_T1", + "correct_answer": "You stopped in front of three claw machines.", + "original_options": { + "A": "You stopped in front of two claw machines.", + "B": "You stopped in front of four claw machines.", + "C": "You stopped in front of one claw machine and then moved along without stopping again.", + "D": "You stopped in front of three claw machines.", + "E": "You stopped in front of five claw machines, pausing at each one before walking away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_3_T1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many claw machines did I stop in front of just now?", + "answer": "You stopped in front of three claw machines.", + "question_chinese": "我刚刚在几个抓娃娃机面前做了停留？", + "answer_chinese": "你在三个抓娃娃机前做了停留", + "evidence": [ + { + "description": "I'm playing with this girl in green clothes to catch dolls.", + "timestamps": { + "start": "00:07:15", + "end": "00:07:28" + } + }, + { + "description": "I accompanied this friend to another claw machin.e", + "timestamps": { + "start": "00:09:55", + "end": "00:10:04" + } + }, + { + "description": "I am watching another friend catch a pachinko ball.", + "timestamps": { + "start": "00:10:21", + "end": "00:10:28" + } + } + ], + "question_timestamps": "00:11:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You stopped in front of two claw machines.", + "B": "You stopped in front of three claw machines.", + "C": "You stopped in front of four claw machines.", + "D": "You stopped in front of one claw machine and then continued walking past the rest.", + "E": "You stopped in front of five claw machines, pausing briefly in front of each one." + }, + "answer_label": "B", + "answer_text": "You stopped in front of three claw machines.", + "q_id": "egolife_A1_D3_013_3_T1", + "correct_answer": "You stopped in front of three claw machines.", + "original_options": { + "A": "You stopped in front of two claw machines.", + "B": "You stopped in front of three claw machines.", + "C": "You stopped in front of four claw machines.", + "D": "You stopped in front of one claw machine and then continued walking past the rest.", + "E": "You stopped in front of five claw machines, pausing briefly in front of each one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_4_T1", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Right now, can I touch the drum he’s playing?", + "answer": "Yes.", + "question_chinese": "我目前可以碰到他打的鼓吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "yes", + "timestamps": { + "start": "00:05:00", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:05:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D3_013_4_T1", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_4_T2", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Right now, can I touch the drum he’s playing?", + "answer": "No.", + "question_chinese": "我目前可以碰到他打的鼓吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "yes", + "timestamps": { + "start": "00:05:00", + "end": "00:05:10" + } + }, + { + "description": "no", + "timestamps": { + "start": "00:05:27", + "end": "00:05:45" + } + } + ], + "question_timestamps": "00:06:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D3_013_4_T2", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_6_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was watching my friend play the drums, which direction from him was I?", + "answer": "You were on his right.", + "question_chinese": "看朋友打鼓的时候，我在他的什么方向？", + "answer_chinese": "你在他的右边", + "evidence": [ + { + "description": "I was on his left.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:24" + } + } + ], + "question_timestamps": "00:03:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were on his right.", + "B": "You were on his left.", + "C": "You were in front of him.", + "D": "You were standing behind him, watching as he played the drums from that position.", + "E": "You were positioned slightly to his left and behind him while he played the drums." + }, + "answer_label": "A", + "answer_text": "You were on his right.", + "q_id": "egolife_A1_D3_013_6_T2", + "correct_answer": "You were on his right.", + "original_options": { + "A": "You were on his right.", + "B": "You were on his left.", + "C": "You were in front of him.", + "D": "You were standing behind him, watching as he played the drums from that position.", + "E": "You were positioned slightly to his left and behind him while he played the drums." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_013_6_T1", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was watching my friend play the drums, which direction from him was I?", + "answer": "You were on his right.", + "question_chinese": "看朋友打鼓的时候，我在他的什么方向？", + "answer_chinese": "你在他的右边", + "evidence": [ + { + "description": "I was on his right", + "timestamps": { + "start": "00:11:13", + "end": "00:11:19" + } + } + ], + "question_timestamps": "00:11:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were on his right.", + "B": "You were on his left.", + "C": "You were behind him.", + "D": "You were standing in front of him, watching as he played the drums from the opposite side.", + "E": "You were positioned slightly to his left, so you viewed his drumming from an angled side position." + }, + "answer_label": "A", + "answer_text": "You were on his right.", + "q_id": "egolife_A1_D3_013_6_T1", + "correct_answer": "You were on his right.", + "original_options": { + "A": "You were on his right.", + "B": "You were on his left.", + "C": "You were behind him.", + "D": "You were standing in front of him, watching as he played the drums from the opposite side.", + "E": "You were positioned slightly to his left, so you viewed his drumming from an angled side position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D3_014-mcq.json b/QAs/egolife_A1_D3_014-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..85203e8a8fd72f975601cbaf646b11d6f017a259 --- /dev/null +++ b/QAs/egolife_A1_D3_014-mcq.json @@ -0,0 +1,323 @@ +[ + { + "video_uid": "egolife_A1_D3_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_014_1_T2", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was just playing the curved-surface ball-and-stick arcade game with the friend in green, could I reach out and touch her?", + "answer": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could reach her. Your friend was to your right.", + "question_chinese": "刚刚跟绿色衣服朋友一起玩曲面曲球棍游戏机的时候，我可以碰到她吗？", + "answer_chinese": "刚刚跟绿色衣服朋友一起玩曲面曲球棍游戏机的时候，你可以碰到她。朋友在你的右边。", + "evidence": [ + { + "description": "She is a little away from me", + "timestamps": { + "start": "00:02:21", + "end": "00:02:26" + } + } + ], + "question_timestamps": "00:02:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could not reach her. Your friend was to your right.", + "B": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could reach her. Your friend was to your right." + }, + "answer_label": "B", + "answer_text": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could reach her. Your friend was to your right.", + "q_id": "egolife_A1_D3_014_1_T2", + "correct_answer": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could reach her. Your friend was to your right.", + "original_options": { + "A": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could not reach her. Your friend was to your right.", + "B": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could reach her. Your friend was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 27 + ] + }, + { + "video_uid": "egolife_A1_D3_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_014_1_T1", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was just playing the curved-surface ball-and-stick arcade game with the friend in green, could I reach out and touch her?", + "answer": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could not reach her. Your friend was in front of you.", + "question_chinese": "刚刚跟绿色衣服朋友一起玩曲面曲球棍游戏机的时候，我可以碰到她吗？", + "answer_chinese": "刚刚跟绿色衣服朋友一起玩曲面曲球棍游戏机的时候，你不可以碰到她。朋友在你的前面。", + "evidence": [ + { + "description": "She is a little away from me", + "timestamps": { + "start": "00:02:21", + "end": "00:02:38" + } + } + ], + "question_timestamps": "00:02:40", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could not reach her. Your friend was in front of you.", + "B": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could reach her. Your friend was in front of you." + }, + "answer_label": "A", + "answer_text": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could not reach her. Your friend was in front of you.", + "q_id": "egolife_A1_D3_014_1_T1", + "correct_answer": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could not reach her. Your friend was in front of you.", + "original_options": { + "A": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could not reach her. Your friend was in front of you.", + "B": "When you were just playing the curved-surface ball-and-club arcade game with your friend in green, you could reach her. Your friend was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 27 + ] + }, + { + "video_uid": "egolife_A1_D3_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_014-mcq_fixed_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where have I watched friends playing video games, and how many friends were there in total?", + "answer": "Two,both were at the racing game machines.", + "question_chinese": "我刚刚在哪些地点一共看过几个朋友玩游戏机？", + "answer_chinese": "两个，都在赛车游戏机旁", + "evidence": [ + { + "description": "I was watching my friend on my left playing the racing game machine.", + "timestamps": { + "start": "00:06:32", + "end": "00:06:34" + } + }, + { + "description": "I was watching my friend on my right playing the racing game machine.", + "timestamps": { + "start": "00:06:45", + "end": "00:06:49" + } + } + ], + "question_timestamps": "00:08:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "One, at the racing game machines.", + "B": "Three, all at the racing game machines.", + "C": "Two friends total, and you watched them at the shooting game machines beside the racing setup.", + "D": "Four friends total, with two at the racing game machines and two at another nearby arcade cabinet.", + "E": "Two,both were at the racing game machines." + }, + "answer_label": "E", + "answer_text": "Two,both were at the racing game machines.", + "q_id": "egolife_A1_D3_014-mcq_fixed_7", + "correct_answer": "Two,both were at the racing game machines.", + "original_options": { + "A": "One, at the racing game machines.", + "B": "Three, all at the racing game machines.", + "C": "Two friends total, and you watched them at the shooting game machines beside the racing setup.", + "D": "Four friends total, with two at the racing game machines and two at another nearby arcade cabinet.", + "E": "Two,both were at the racing game machines." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_014-mcq_fixed_8", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where have I watched friends playing video games, and how many friends were there in total?", + "answer": "Three friends, two at the racing game machines and one at the blue machine.", + "question_chinese": "我刚刚在哪些地点一共看过几个朋友玩游戏机？", + "answer_chinese": "三个，分别在赛车游戏机旁和一个蓝色机器旁", + "evidence": [ + { + "description": "I am watching my friend on my left play the racing game machine.", + "timestamps": { + "start": "00:06:31", + "end": "00:06:34" + } + }, + { + "description": "I am watching my friend on my right play the racing game machine.", + "timestamps": { + "start": "00:06:40", + "end": "00:06:45" + } + }, + { + "description": "I am watching the friend playing the machine", + "timestamps": { + "start": "00:10:04", + "end": "00:10:11" + } + } + ], + "question_timestamps": "00:11:35", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Two friends, both at the racing machines.", + "B": "Four friends, all at the blue machine.", + "C": "Three friends total, with one at the racing machines and two gathered at the blue machine.", + "D": "Three friends, two at the racing game machines and one at the blue machine.", + "E": "Five friends total, with two at the racing game machines and three clustered around the blue machine." + }, + "answer_label": "D", + "answer_text": "Three friends, two at the racing game machines and one at the blue machine.", + "q_id": "egolife_A1_D3_014-mcq_fixed_8", + "correct_answer": "Three friends, two at the racing game machines and one at the blue machine.", + "original_options": { + "A": "Two friends, both at the racing machines.", + "B": "Four friends, all at the blue machine.", + "C": "Three friends total, with one at the racing machines and two gathered at the blue machine.", + "D": "Three friends, two at the racing game machines and one at the blue machine.", + "E": "Five friends total, with two at the racing game machines and three clustered around the blue machine." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_014_9_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I just walked in through that yellow door, how many racing arcade machines did I see?", + "answer": "When you just went through that yellow door, you saw three racing arcade machines. The racing arcade machines are behind you.", + "question_chinese": "我刚刚进那个黄色的门的时候，我看见了几个赛车游戏机？", + "answer_chinese": "你刚刚进那个黄色的门的时候，你看见了三个赛车游戏机。赛车游戏机在你的后面。", + "evidence": [ + { + "description": "At this moment, there was a row of chairs and game machines in front of me.", + "timestamps": { + "start": "00:05:56", + "end": "00:05:59" + } + } + ], + "question_timestamps": "00:06:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw one racing arcade machine.", + "B": "You saw five racing arcade machines.", + "C": "When you just went through that yellow door, you saw two racing arcade machines. They were behind you.", + "D": "When you just went through that yellow door, you saw three racing arcade machines. The racing arcade machines are behind you.", + "E": "When you just went through that yellow door, you saw four racing arcade machines. The machines were behind you." + }, + "answer_label": "D", + "answer_text": "When you just went through that yellow door, you saw three racing arcade machines. The racing arcade machines are behind you.", + "q_id": "egolife_A1_D3_014_9_T1", + "correct_answer": "When you just went through that yellow door, you saw three racing arcade machines. The racing arcade machines are behind you.", + "original_options": { + "A": "You saw one racing arcade machine.", + "B": "You saw five racing arcade machines.", + "C": "When you just went through that yellow door, you saw two racing arcade machines. They were behind you.", + "D": "When you just went through that yellow door, you saw three racing arcade machines. The racing arcade machines are behind you.", + "E": "When you just went through that yellow door, you saw four racing arcade machines. The machines were behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_014_9_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I just walked in through that yellow door, how many racing arcade machines did I see?", + "answer": "When you just went in through that yellow door, you saw four racing arcade machines. The racing arcade machines were in front of you.", + "question_chinese": "我刚刚进那个黄色的门的时候，我看见了几个赛车游戏机？", + "answer_chinese": "你刚刚进那个黄色的门的时候，你看见了四个赛车游戏机。赛车游戏机在你的前面。", + "evidence": [ + { + "description": "After entering the door, there were four rows of racing game machines in front of me", + "timestamps": { + "start": "00:06:04", + "end": "00:06:10" + } + } + ], + "question_timestamps": "00:06:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw two racing arcade machines.", + "B": "You saw six racing arcade machines.", + "C": "When you just went in through that yellow door, you saw three racing arcade machines. They were in front of you.", + "D": "When you just went in through that yellow door, you saw five racing arcade machines. The machines were in front of you.", + "E": "When you just went in through that yellow door, you saw four racing arcade machines. The racing arcade machines were in front of you." + }, + "answer_label": "E", + "answer_text": "When you just went in through that yellow door, you saw four racing arcade machines. The racing arcade machines were in front of you.", + "q_id": "egolife_A1_D3_014_9_T2", + "correct_answer": "When you just went in through that yellow door, you saw four racing arcade machines. The racing arcade machines were in front of you.", + "original_options": { + "A": "You saw two racing arcade machines.", + "B": "You saw six racing arcade machines.", + "C": "When you just went in through that yellow door, you saw three racing arcade machines. They were in front of you.", + "D": "When you just went in through that yellow door, you saw five racing arcade machines. The machines were in front of you.", + "E": "When you just went in through that yellow door, you saw four racing arcade machines. The racing arcade machines were in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 22 + ] + } +] diff --git a/QAs/egolife_A1_D3_015-mcq.json b/QAs/egolife_A1_D3_015-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..90e17cd9e28fc62e0cbf2c06b967e521676f5d65 --- /dev/null +++ b/QAs/egolife_A1_D3_015-mcq.json @@ -0,0 +1,1290 @@ +[ + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_1_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Just now, what route did the red box in my hand take?", + "answer": "It was placed on the bowling machine by me.", + "question_chinese": "刚刚我手中的红框子经历了什么路线？", + "answer_chinese": "它被我放到了保龄球机上.", + "evidence": [ + { + "description": "I placed it on the bowling machine,", + "timestamps": { + "start": "00:02:30", + "end": "00:02:41" + } + } + ], + "question_timestamps": "00:02:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was placed on the bowling machine by me.", + "B": "You picked it up.", + "C": "You set it down elsewhere.", + "D": "You placed the red box down briefly and then brought it back into your hand afterward.", + "E": "You moved the red box from your hand toward the machine area, but never actually set it on the bowling machine." + }, + "answer_label": "A", + "answer_text": "It was placed on the bowling machine by me.", + "q_id": "egolife_A1_D3_015_1_T2", + "correct_answer": "It was placed on the bowling machine by me.", + "original_options": { + "A": "It was placed on the bowling machine by me.", + "B": "You picked it up.", + "C": "You set it down elsewhere.", + "D": "You placed the red box down briefly and then brought it back into your hand afterward.", + "E": "You moved the red box from your hand toward the machine area, but never actually set it on the bowling machine." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 21 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_1_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Just now, what route did the red box in my hand take?", + "answer": "It was picked up by you.", + "question_chinese": "刚刚我手中的红框子经历了什么路线？", + "answer_chinese": "它被你拿了起来.", + "evidence": [ + { + "description": "Then it was picked by me", + "timestamps": { + "start": "00:06:15", + "end": "00:06:36" + } + } + ], + "question_timestamps": "00:06:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You placed it down.", + "B": "You dropped it.", + "C": "You set the red box on the bowling machine and left it there instead of picking it up.", + "D": "It was picked up by you.", + "E": "You placed it on the bowling machine first, then picked it up again, so it moved twice." + }, + "answer_label": "D", + "answer_text": "It was picked up by you.", + "q_id": "egolife_A1_D3_015_1_T1", + "correct_answer": "It was picked up by you.", + "original_options": { + "A": "You placed it down.", + "B": "You dropped it.", + "C": "You set the red box on the bowling machine and left it there instead of picking it up.", + "D": "It was picked up by you.", + "E": "You placed it on the bowling machine first, then picked it up again, so it moved twice." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did the red box in my hand just take?", + "answer": "I first placed it on the bowling machine, and then picked it up again.", + "question_chinese": "刚刚我手中的红框子经历了什么路线？", + "answer_chinese": "它先被我放到了保龄球机上，然后又被我拿了起来", + "evidence": [ + { + "description": "Then it was picked by me", + "timestamps": { + "start": "00:06:15", + "end": "00:06:36" + } + }, + { + "description": "I placed it on the bowling machine,", + "timestamps": { + "start": "00:02:30", + "end": "00:02:41" + } + } + ], + "question_timestamps": "00:08:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was placed on the machine.", + "B": "It was picked up by you.", + "C": "I first placed it on the bowling machine, and then picked it up again.", + "D": "I put the red box on the bowling machine and did not pick it back up afterward.", + "E": "You picked up the red box and kept holding it, never setting it on the bowling machine." + }, + "answer_label": "C", + "answer_text": "I first placed it on the bowling machine, and then picked it up again.", + "q_id": "egolife_A1_D3_015_1", + "correct_answer": "I first placed it on the bowling machine, and then picked it up again.", + "original_options": { + "A": "It was placed on the machine.", + "B": "It was picked up by you.", + "C": "I first placed it on the bowling machine, and then picked it up again.", + "D": "I put the red box on the bowling machine and did not pick it back up afterward.", + "E": "You picked up the red box and kept holding it, never setting it on the bowling machine." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_8_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was watching my friend in green bowl just now, could I touch the pillar on this machine?", + "answer": "You can touch the pillar of this machine; the pillar is in front of you.", + "question_chinese": "刚刚看绿色衣服朋友掷保龄球的时候，我可以碰到这个机器的柱子吗？", + "answer_chinese": "你可以碰到这个机器的柱子，柱子在你前面。", + "evidence": [ + { + "description": "Yes, it was very close to me.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:11" + } + } + ], + "question_timestamps": "00:05:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You cannot touch the pillar of this machine; the pillar is in front of you.", + "B": "You can touch the pillar of this machine; the pillar is in front of you." + }, + "answer_label": "B", + "answer_text": "You can touch the pillar of this machine; the pillar is in front of you.", + "q_id": "egolife_A1_D3_015_8_T2", + "correct_answer": "You can touch the pillar of this machine; the pillar is in front of you.", + "original_options": { + "A": "You cannot touch the pillar of this machine; the pillar is in front of you.", + "B": "You can touch the pillar of this machine; the pillar is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_8_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was watching my friend in green bowl just now, could I touch the pillar on this machine?", + "answer": "You can't touch the pillar of this machine; the pillar is in front of you.", + "question_chinese": "刚刚看绿色衣服朋友掷保龄球的时候，我可以碰到这个机器的柱子吗？", + "answer_chinese": "你不可以碰到这个机器的柱子，柱子在你前面。", + "evidence": [ + { + "description": "Yes, it was very close to me.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:23" + } + } + ], + "question_timestamps": "00:05:26", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can't touch the pillar of this machine; the pillar is in front of you.", + "B": "You can touch the pillar of this machine; the pillar is in front of you." + }, + "answer_label": "A", + "answer_text": "You can't touch the pillar of this machine; the pillar is in front of you.", + "q_id": "egolife_A1_D3_015_8_T1", + "correct_answer": "You can't touch the pillar of this machine; the pillar is in front of you.", + "original_options": { + "A": "You can't touch the pillar of this machine; the pillar is in front of you.", + "B": "You can touch the pillar of this machine; the pillar is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_9", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the pole of this machine while watching my friend in the green shirt bowl?", + "answer": "Yes，I can.", + "question_chinese": "刚刚看绿色衣服朋友掷保龄球的时候，我可以碰到这个机器的柱子吗", + "answer_chinese": "以可", + "evidence": [ + { + "description": "Yes, it was very close to me.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:11" + } + } + ], + "question_timestamps": "00:07:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes，I can.", + "B": "No, I cannot." + }, + "answer_label": "A", + "answer_text": "Yes，I can.", + "q_id": "egolife_A1_D3_015_9", + "correct_answer": "Yes，I can.", + "original_options": { + "A": "Yes，I can.", + "B": "No, I cannot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_5_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I just walked past the basketball arcade machine, was it to my left or to my right?", + "answer": "It's in front of you.", + "question_chinese": "我刚刚经过篮球机的时候，他在我的左边还是右边？", + "answer_chinese": "它在你的前面。", + "evidence": [ + { + "description": "It was in front of you.", + "timestamps": { + "start": "00:06:36", + "end": "00:06:40" + } + } + ], + "question_timestamps": "00:06:43", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It's in front of you.", + "B": "It's to your left." + }, + "answer_label": "A", + "answer_text": "It's in front of you.", + "q_id": "egolife_A1_D3_015_5_T2", + "correct_answer": "It's in front of you.", + "original_options": { + "A": "It's in front of you.", + "B": "It's to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_5_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I just walked past the basketball arcade machine, was it to my left or to my right?", + "answer": "It's to your left.", + "question_chinese": "我刚刚经过篮球机的时候，他在我的左边还是右边？", + "answer_chinese": "它在我的左边", + "evidence": [ + { + "description": "It was on my right", + "timestamps": { + "start": "00:06:36", + "end": "00:06:40" + } + } + ], + "question_timestamps": "00:06:44", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It's to your right.", + "B": "It's to your left." + }, + "answer_label": "B", + "answer_text": "It's to your left.", + "q_id": "egolife_A1_D3_015_5_T1", + "correct_answer": "It's to your left.", + "original_options": { + "A": "It's to your right.", + "B": "It's to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_6", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I just passed by the basketball machine, was he on my left or my right?", + "answer": "It was on my right.", + "question_chinese": "我刚刚经过篮球机的时候，他在我的左边还是右边", + "answer_chinese": "它在我的左边", + "evidence": [ + { + "description": "It was on my right", + "timestamps": { + "start": "00:06:36", + "end": "00:06:40" + } + } + ], + "question_timestamps": "00:07:33", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "It was on my left.", + "B": "It was on my right." + }, + "answer_label": "B", + "answer_text": "It was on my right.", + "q_id": "egolife_A1_D3_015_6", + "correct_answer": "It was on my right.", + "original_options": { + "A": "It was on my left.", + "B": "It was on my right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_9_T1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the friend wearing patterned clothes?", + "answer": "You can't touch the friend wearing patterned clothes; he is to your back-right.", + "question_chinese": "我可以碰到花色衣服的朋友吗？", + "answer_chinese": "你不可以碰到花色衣服的朋友，他在你右后方。", + "evidence": [ + { + "description": "No, you can not.", + "timestamps": { + "start": "00:09:38", + "end": "00:09:40" + } + } + ], + "question_timestamps": "00:09:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can touch the friend wearing patterned clothes; he is to your back-right.", + "B": "You can't touch the friend wearing patterned clothes; he is to your back-right." + }, + "answer_label": "B", + "answer_text": "You can't touch the friend wearing patterned clothes; he is to your back-right.", + "q_id": "egolife_A1_D3_015_9_T1", + "correct_answer": "You can't touch the friend wearing patterned clothes; he is to your back-right.", + "original_options": { + "A": "You can touch the friend wearing patterned clothes; he is to your back-right.", + "B": "You can't touch the friend wearing patterned clothes; he is to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_9_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the friend wearing patterned clothes?", + "answer": "You can touch the friend wearing patterned clothes; he is to your right.", + "question_chinese": "我可以碰到花色衣服的朋友吗？", + "answer_chinese": "你可以碰到花色衣服的朋友，他在你右边。", + "evidence": [ + { + "description": "No, it was a bit far away from me", + "timestamps": { + "start": "00:10:06", + "end": "00:10:24" + } + } + ], + "question_timestamps": "00:10:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can touch the friend wearing patterned clothes; he is to your right.", + "B": "You can't touch the friend wearing patterned clothes; he is to your right." + }, + "answer_label": "A", + "answer_text": "You can touch the friend wearing patterned clothes; he is to your right.", + "q_id": "egolife_A1_D3_015_9_T2", + "correct_answer": "You can touch the friend wearing patterned clothes; he is to your right.", + "original_options": { + "A": "You can touch the friend wearing patterned clothes; he is to your right.", + "B": "You can't touch the friend wearing patterned clothes; he is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_6_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I walked past the basketball arcade machine, was it to my left or to my right?", + "answer": "It's to your right.", + "question_chinese": "我经过篮球机的时候，它在我的左边还是右边？", + "answer_chinese": "它在你的右边.", + "evidence": [ + { + "description": "It was on my left", + "timestamps": { + "start": "00:01:31", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:47", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It's to your left.", + "B": "It's to your right." + }, + "answer_label": "B", + "answer_text": "It's to your right.", + "q_id": "egolife_A1_D3_015_6_T2", + "correct_answer": "It's to your right.", + "original_options": { + "A": "It's to your left.", + "B": "It's to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_6_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I walked past the basketball arcade machine, was it to my left or to my right?", + "answer": "It's to your left.", + "question_chinese": "我经过篮球机的时候，它在我的左边还是右边？", + "answer_chinese": "它在你的左边。", + "evidence": [ + { + "description": "It was on my left", + "timestamps": { + "start": "00:06:38", + "end": "00:06:42" + } + } + ], + "question_timestamps": "00:06:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It's to your right.", + "B": "It's to your left." + }, + "answer_label": "B", + "answer_text": "It's to your left.", + "q_id": "egolife_A1_D3_015_6_T1", + "correct_answer": "It's to your left.", + "original_options": { + "A": "It's to your right.", + "B": "It's to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_4_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Can I touch the game console directly?", + "answer": "You can directly reach the arcade machine; the arcade machine is in front of you.", + "question_chinese": "我能直接接触到游戏机吗？", + "answer_chinese": "你能直接接触到游戏机，游戏机在你前面。", + "evidence": [ + { + "description": "You can directly reach the game console.", + "timestamps": { + "start": "00:06:17", + "end": "00:06:28" + } + } + ], + "question_timestamps": "00:06:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can directly reach the arcade machine; the arcade machine is in front of you.", + "B": "You can't reach it directly.", + "C": "It's to your left.", + "D": "You can directly reach the arcade machine, but it is positioned to your left, not in front of you.", + "E": "You can directly reach the arcade machine, though it is actually behind you rather than in front of you." + }, + "answer_label": "A", + "answer_text": "You can directly reach the arcade machine; the arcade machine is in front of you.", + "q_id": "egolife_A1_D3_015_4_T2", + "correct_answer": "You can directly reach the arcade machine; the arcade machine is in front of you.", + "original_options": { + "A": "You can directly reach the arcade machine; the arcade machine is in front of you.", + "B": "You can't reach it directly.", + "C": "It's to your left.", + "D": "You can directly reach the arcade machine, but it is positioned to your left, not in front of you.", + "E": "You can directly reach the arcade machine, though it is actually behind you rather than in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_4_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Can I touch the game console directly?", + "answer": "You can't directly reach the arcade machine; the arcade machine is in front of you.", + "question_chinese": "我能直接接触到游戏机吗？", + "answer_chinese": "你不能直接接触到游戏机,游戏机在你前面。", + "evidence": [ + { + "description": "You can not directly reach the game console.", + "timestamps": { + "start": "00:06:30", + "end": "00:06:34" + } + } + ], + "question_timestamps": "00:06:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You can reach it directly.", + "B": "It's to your left.", + "C": "You can't directly reach the arcade machine; it is to your left rather than in front of you.", + "D": "You can't directly reach the arcade machine; it is behind you, not in front of you.", + "E": "You can't directly reach the arcade machine; the arcade machine is in front of you." + }, + "answer_label": "E", + "answer_text": "You can't directly reach the arcade machine; the arcade machine is in front of you.", + "q_id": "egolife_A1_D3_015_4_T1", + "correct_answer": "You can't directly reach the arcade machine; the arcade machine is in front of you.", + "original_options": { + "A": "You can reach it directly.", + "B": "It's to your left.", + "C": "You can't directly reach the arcade machine; it is to your left rather than in front of you.", + "D": "You can't directly reach the arcade machine; it is behind you, not in front of you.", + "E": "You can't directly reach the arcade machine; the arcade machine is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_3_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, in which direction from me is the emergency exit passage?", + "answer": "At this moment, the emergency exit passage is to your right.", + "question_chinese": "此时，安全通道在我的什么方向？", + "answer_chinese": "此时，安全通道在你右边。", + "evidence": [ + { + "description": "The emergency exit is on your right.", + "timestamps": { + "start": "00:08:23", + "end": "00:08:26" + } + } + ], + "question_timestamps": "00:08:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to your left.", + "B": "It is behind you.", + "C": "At this moment, the emergency exit passage is directly in front of you, not to your right.", + "D": "At this moment, the emergency exit passage is behind you, so it isn't on your right side.", + "E": "At this moment, the emergency exit passage is to your right." + }, + "answer_label": "E", + "answer_text": "At this moment, the emergency exit passage is to your right.", + "q_id": "egolife_A1_D3_015_3_T2", + "correct_answer": "At this moment, the emergency exit passage is to your right.", + "original_options": { + "A": "It is to your left.", + "B": "It is behind you.", + "C": "At this moment, the emergency exit passage is directly in front of you, not to your right.", + "D": "At this moment, the emergency exit passage is behind you, so it isn't on your right side.", + "E": "At this moment, the emergency exit passage is to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_3_T1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, in which direction from me is the emergency exit passage?", + "answer": "At this moment, the emergency exit passage is behind you.", + "question_chinese": "此时，安全通道在我的什么方向？", + "answer_chinese": "此时，安全通道在你后面。", + "evidence": [ + { + "description": "The emergency exit is behind of you.", + "timestamps": { + "start": "00:08:33", + "end": "00:08:40" + } + } + ], + "question_timestamps": "00:08:43", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "At this moment, the emergency exit passage is behind you.", + "B": "It is in front of you.", + "C": "It is to your left.", + "D": "At this moment, the emergency exit passage is to your right, not behind you.", + "E": "Right now, the emergency exit passage is above you, not behind you." + }, + "answer_label": "A", + "answer_text": "At this moment, the emergency exit passage is behind you.", + "q_id": "egolife_A1_D3_015_3_T1", + "correct_answer": "At this moment, the emergency exit passage is behind you.", + "original_options": { + "A": "At this moment, the emergency exit passage is behind you.", + "B": "It is in front of you.", + "C": "It is to your left.", + "D": "At this moment, the emergency exit passage is to your right, not behind you.", + "E": "Right now, the emergency exit passage is above you, not behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_7_T2", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Right now, which of my friends is shooting?", + "answer": "At this moment, your friend wearing a hat is bowling, and she is in front of you.", + "question_chinese": "此时，我的哪个朋友在投球？", + "answer_chinese": "此时，你戴帽子的朋友在投球，她在你前面。", + "evidence": [ + { + "description": "Your friend in white is throwing the ball.", + "timestamps": { + "start": "00:09:43", + "end": "00:10:09" + } + } + ], + "question_timestamps": "00:10:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Friend in hat behind you.", + "B": "Friend in hat to your left.", + "C": "At this moment, your friend wearing a hat is bowling, and she is to your right.", + "D": "At this moment, your friend wearing a hat is bowling, and she is behind you rather than in front.", + "E": "At this moment, your friend wearing a hat is bowling, and she is in front of you." + }, + "answer_label": "E", + "answer_text": "At this moment, your friend wearing a hat is bowling, and she is in front of you.", + "q_id": "egolife_A1_D3_015_7_T2", + "correct_answer": "At this moment, your friend wearing a hat is bowling, and she is in front of you.", + "original_options": { + "A": "Friend in hat behind you.", + "B": "Friend in hat to your left.", + "C": "At this moment, your friend wearing a hat is bowling, and she is to your right.", + "D": "At this moment, your friend wearing a hat is bowling, and she is behind you rather than in front.", + "E": "At this moment, your friend wearing a hat is bowling, and she is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_7_T1", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Right now, which of my friends is shooting?", + "answer": "At this moment, your friend in green is bowling, and she is in front of you.", + "question_chinese": "此时，我的哪个朋友在投球？", + "answer_chinese": "此时，你穿绿色衣服的朋友在投球，她在你前面。", + "evidence": [ + { + "description": "Your friend in grenn is throwing the ball.", + "timestamps": { + "start": "00:10:13", + "end": "00:10:43" + } + } + ], + "question_timestamps": "00:10:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your friend in green is behind you.", + "B": "At this moment, your friend in green is bowling, and she is in front of you.", + "C": "Your friend in green is to your left.", + "D": "At this moment, your friend in green is bowling, and she is to your right.", + "E": "At this moment, your friend in green is bowling, but she is behind you instead of in front." + }, + "answer_label": "B", + "answer_text": "At this moment, your friend in green is bowling, and she is in front of you.", + "q_id": "egolife_A1_D3_015_7_T1", + "correct_answer": "At this moment, your friend in green is bowling, and she is in front of you.", + "original_options": { + "A": "Your friend in green is behind you.", + "B": "At this moment, your friend in green is bowling, and she is in front of you.", + "C": "Your friend in green is to your left.", + "D": "At this moment, your friend in green is bowling, and she is to your right.", + "E": "At this moment, your friend in green is bowling, but she is behind you instead of in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_10_T1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many balls in total has my friend just thrown at the bowling machine?", + "answer": "At this moment, your friend just threw 0 balls in total in front of the bowling machine, and the bowling machine is in front of you.", + "question_chinese": "此时，我的朋友刚刚在保龄球机前一共掷了几个球？", + "answer_chinese": "此时，你的朋友刚刚在保龄球机前一共掷了0个球，保龄球机在你前面。", + "evidence": [ + { + "description": "The first one", + "timestamps": { + "start": "00:02:28", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Your friend threw 1 ball total.", + "B": "Your friend threw 2 balls total.", + "C": "At this moment, your friend just threw 3 balls total at the bowling machine in front of you.", + "D": "At this moment, your friend just threw 4 balls total at the bowling machine in front of you.", + "E": "At this moment, your friend just threw 0 balls in total in front of the bowling machine, and the bowling machine is in front of you." + }, + "answer_label": "E", + "answer_text": "At this moment, your friend just threw 0 balls in total in front of the bowling machine, and the bowling machine is in front of you.", + "q_id": "egolife_A1_D3_015_10_T1", + "correct_answer": "At this moment, your friend just threw 0 balls in total in front of the bowling machine, and the bowling machine is in front of you.", + "original_options": { + "A": "Your friend threw 1 ball total.", + "B": "Your friend threw 2 balls total.", + "C": "At this moment, your friend just threw 3 balls total at the bowling machine in front of you.", + "D": "At this moment, your friend just threw 4 balls total at the bowling machine in front of you.", + "E": "At this moment, your friend just threw 0 balls in total in front of the bowling machine, and the bowling machine is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_10_T2", + "qid_num": 21, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many balls in total has my friend just thrown at the bowling machine?", + "answer": "At this moment, your friend just threw 5 balls in total in front of the bowling machine, and the bowling machine is in front of you.", + "question_chinese": "此时，我的朋友刚刚在保龄球机前一共掷了几个球？", + "answer_chinese": "此时，你的朋友刚刚在保龄球机前一共掷了5个球，保龄球机在你前面。", + "evidence": [ + { + "description": "The first one", + "timestamps": { + "start": "00:02:49", + "end": "00:02:56" + } + }, + { + "description": "The second one", + "timestamps": { + "start": "00:04:16", + "end": "00:04:20" + } + }, + { + "description": "The third one", + "timestamps": { + "start": "00:04:30", + "end": "00:04:40" + } + }, + { + "description": "The fourth one", + "timestamps": { + "start": "00:04:56", + "end": "00:04:59" + } + }, + { + "description": "The fifth one", + "timestamps": { + "start": "00:06:06", + "end": "00:06:11" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "At this moment, your friend just threw 5 balls in total in front of the bowling machine, and the bowling machine is in front of you.", + "B": "Your friend threw 4 balls total.", + "C": "Your friend threw 6 balls total.", + "D": "At this moment, your friend just threw 3 balls total at the bowling machine in front of you.", + "E": "At this moment, your friend just threw 7 balls total at the bowling machine in front of you." + }, + "answer_label": "A", + "answer_text": "At this moment, your friend just threw 5 balls in total in front of the bowling machine, and the bowling machine is in front of you.", + "q_id": "egolife_A1_D3_015_10_T2", + "correct_answer": "At this moment, your friend just threw 5 balls in total in front of the bowling machine, and the bowling machine is in front of you.", + "original_options": { + "A": "At this moment, your friend just threw 5 balls in total in front of the bowling machine, and the bowling machine is in front of you.", + "B": "Your friend threw 4 balls total.", + "C": "Your friend threw 6 balls total.", + "D": "At this moment, your friend just threw 3 balls total at the bowling machine in front of you.", + "E": "At this moment, your friend just threw 7 balls total at the bowling machine in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_2_T2", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, in which direction from me is the red stool?", + "answer": "At this moment, the red stool is to your right.", + "question_chinese": "此时，红色凳子在我的什么方向？", + "answer_chinese": "此时，红色凳子在你右边.", + "evidence": [ + { + "description": "The red chair in your right.", + "timestamps": { + "start": "00:01:13", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "At this moment, the red stool is to your right.", + "C": "It is behind you.", + "D": "At this moment, the red stool is to your left rather than to your right.", + "E": "Right now, the red stool is positioned in front of you, not off to your right side." + }, + "answer_label": "B", + "answer_text": "At this moment, the red stool is to your right.", + "q_id": "egolife_A1_D3_015_2_T2", + "correct_answer": "At this moment, the red stool is to your right.", + "original_options": { + "A": "It is in front of you.", + "B": "At this moment, the red stool is to your right.", + "C": "It is behind you.", + "D": "At this moment, the red stool is to your left rather than to your right.", + "E": "Right now, the red stool is positioned in front of you, not off to your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_2_T1", + "qid_num": 23, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, in which direction from me is the red stool?", + "answer": "At this moment, the red stool is in front of you.", + "question_chinese": "此时，红色凳子在我的什么方向？", + "answer_chinese": "此时，红色凳子在你前面。", + "evidence": [ + { + "description": "The red chair is in front of you.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is to your right.", + "C": "At this moment, the red stool is to your left instead of being in front of you.", + "D": "Right now, the red stool is above you, not located straight ahead in your view.", + "E": "At this moment, the red stool is in front of you." + }, + "answer_label": "E", + "answer_text": "At this moment, the red stool is in front of you.", + "q_id": "egolife_A1_D3_015_2_T1", + "correct_answer": "At this moment, the red stool is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is to your right.", + "C": "At this moment, the red stool is to your left instead of being in front of you.", + "D": "Right now, the red stool is above you, not located straight ahead in your view.", + "E": "At this moment, the red stool is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_11_T1", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Which direction is the guy in patterned clothes relative to me?", + "answer": "The boy wearing patterned clothes is behind you.", + "question_chinese": "穿花衣服的男生在我的哪个方向？", + "answer_chinese": "穿花衣服的男生在你后面。", + "evidence": [ + { + "description": "The first one", + "timestamps": { + "start": "00:08:23", + "end": "00:08:26" + } + } + ], + "question_timestamps": "00:08:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The boy wearing patterned clothes is behind you.", + "B": "He is in front of you.", + "C": "He is to your left.", + "D": "The boy wearing patterned clothes is to your right rather than behind you.", + "E": "The boy wearing patterned clothes is in front of you, not positioned behind your back." + }, + "answer_label": "A", + "answer_text": "The boy wearing patterned clothes is behind you.", + "q_id": "egolife_A1_D3_015_11_T1", + "correct_answer": "The boy wearing patterned clothes is behind you.", + "original_options": { + "A": "The boy wearing patterned clothes is behind you.", + "B": "He is in front of you.", + "C": "He is to your left.", + "D": "The boy wearing patterned clothes is to your right rather than behind you.", + "E": "The boy wearing patterned clothes is in front of you, not positioned behind your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_015_11_T2", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Which direction is the guy in patterned clothes relative to me?", + "answer": "The boy wearing patterned clothes is in front of you.", + "question_chinese": "穿花衣服的男生在我的哪个方向？", + "answer_chinese": "穿花衣服的男生在你前面。", + "evidence": [ + { + "description": "The first one", + "timestamps": { + "start": "00:08:34", + "end": "00:08:38" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The boy is behind you.", + "B": "The boy is to your left.", + "C": "The boy wearing patterned clothes is in front of you.", + "D": "The boy is to your right, staying beside you as you move forward.", + "E": "The boy is above you, positioned higher up than your eye level." + }, + "answer_label": "C", + "answer_text": "The boy wearing patterned clothes is in front of you.", + "q_id": "egolife_A1_D3_015_11_T2", + "correct_answer": "The boy wearing patterned clothes is in front of you.", + "original_options": { + "A": "The boy is behind you.", + "B": "The boy is to your left.", + "C": "The boy wearing patterned clothes is in front of you.", + "D": "The boy is to your right, staying beside you as you move forward.", + "E": "The boy is above you, positioned higher up than your eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 12 + ] + } +] diff --git a/QAs/egolife_A1_D3_016-mcq.json b/QAs/egolife_A1_D3_016-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..592eed22f415c17181f39a72da66b6d542f6af71 --- /dev/null +++ b/QAs/egolife_A1_D3_016-mcq.json @@ -0,0 +1,1140 @@ +[ + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_5_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the freezer relative to me?", + "answer": "The freezer is in front of you.", + "question_chinese": "冰柜在我的哪里？", + "answer_chinese": "冰柜在你的前面", + "evidence": [ + { + "description": "The freezer is in front of you.", + "timestamps": { + "start": "00:07:15", + "end": "00:07:19" + } + } + ], + "question_timestamps": "00:07:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The freezer is behind you.", + "B": "The freezer is to your left.", + "C": "The freezer is in front of you.", + "D": "The freezer is to your right, sitting along the aisle next to you as you face forward.", + "E": "The freezer is above you, higher than your head and not directly at your level." + }, + "answer_label": "C", + "answer_text": "The freezer is in front of you.", + "q_id": "egolife_A1_D3_016_5_T2", + "correct_answer": "The freezer is in front of you.", + "original_options": { + "A": "The freezer is behind you.", + "B": "The freezer is to your left.", + "C": "The freezer is in front of you.", + "D": "The freezer is to your right, sitting along the aisle next to you as you face forward.", + "E": "The freezer is above you, higher than your head and not directly at your level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_5_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the freezer relative to me?", + "answer": "The freezer is to your left.", + "question_chinese": "冰柜在我的哪里？", + "answer_chinese": "冰柜在你的左边", + "evidence": [ + { + "description": "The freezer is on your left.", + "timestamps": { + "start": "00:07:24", + "end": "00:07:30" + } + } + ], + "question_timestamps": "00:07:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The freezer is to your left.", + "B": "The freezer is in front of you.", + "C": "The freezer is behind you.", + "D": "The freezer is to your right, positioned alongside you as you look straight ahead.", + "E": "The freezer is above you, located higher than your eye level rather than beside you." + }, + "answer_label": "A", + "answer_text": "The freezer is to your left.", + "q_id": "egolife_A1_D3_016_5_T1", + "correct_answer": "The freezer is to your left.", + "original_options": { + "A": "The freezer is to your left.", + "B": "The freezer is in front of you.", + "C": "The freezer is behind you.", + "D": "The freezer is to your right, positioned alongside you as you look straight ahead.", + "E": "The freezer is above you, located higher than your eye level rather than beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What area of the supermarket am I in right now?", + "answer": "Beverage section.", + "question_chinese": "我刚刚在超市的什么区域？", + "answer_chinese": "饮品区", + "evidence": [ + { + "description": "The fridge is full of drinks.", + "timestamps": { + "start": "00:05:57", + "end": "00:06:03" + } + }, + { + "description": "The cabinet is filled with drinks.", + "timestamps": { + "start": "00:06:29", + "end": "00:06:42" + } + } + ], + "question_timestamps": "00:07:59", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Frozen foods section.", + "B": "Produce section.", + "C": "Beverage section.", + "D": "You are in the dairy section, where refrigerated items line the shelves near you.", + "E": "You are in the snack aisle, with chips and packaged treats displayed along the racks." + }, + "answer_label": "C", + "answer_text": "Beverage section.", + "q_id": "egolife_A1_D3_016_7", + "correct_answer": "Beverage section.", + "original_options": { + "A": "Frozen foods section.", + "B": "Produce section.", + "C": "Beverage section.", + "D": "You are in the dairy section, where refrigerated items line the shelves near you.", + "E": "You are in the snack aisle, with chips and packaged treats displayed along the racks." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_8", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What section of the supermarket am I in right now?", + "answer": "Snack Section.", + "question_chinese": "我刚刚在超市的什么区域？", + "answer_chinese": "零食区", + "evidence": [ + { + "description": "The shelves are stocked with snacks.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:33" + } + }, + { + "description": "The shelves are stocked with snacks.", + "timestamps": { + "start": "00:08:41", + "end": "00:08:57" + } + } + ], + "question_timestamps": "00:10:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Snack Section.", + "B": "Beverage section.", + "C": "Frozen foods section.", + "D": "You are in the bakery section, with breads and pastries displayed on the nearby shelves.", + "E": "You are in the produce section, surrounded by fresh fruits and vegetables in open bins." + }, + "answer_label": "A", + "answer_text": "Snack Section.", + "q_id": "egolife_A1_D3_016_8", + "correct_answer": "Snack Section.", + "original_options": { + "A": "Snack Section.", + "B": "Beverage section.", + "C": "Frozen foods section.", + "D": "You are in the bakery section, with breads and pastries displayed on the nearby shelves.", + "E": "You are in the produce section, surrounded by fresh fruits and vegetables in open bins." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_12_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I doing?", + "answer": "You were looking at an arcade machine, and it was in front of you.", + "question_chinese": "我在干什么？", + "answer_chinese": "你在观察游戏机，游戏机在你前面", + "evidence": [ + { + "description": "You are looking at the game console, which is in front of you.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:07" + } + } + ], + "question_timestamps": "00:03:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were looking away from the arcade machine.", + "B": "You were walking past the arcade machine.", + "C": "You were looking at an arcade machine, and it was in front of you.", + "D": "You were looking at an arcade machine, but it was positioned behind you rather than ahead.", + "E": "You were looking at an arcade machine, with it off to your right side instead of straight ahead." + }, + "answer_label": "C", + "answer_text": "You were looking at an arcade machine, and it was in front of you.", + "q_id": "egolife_A1_D3_016_12_T1", + "correct_answer": "You were looking at an arcade machine, and it was in front of you.", + "original_options": { + "A": "You were looking away from the arcade machine.", + "B": "You were walking past the arcade machine.", + "C": "You were looking at an arcade machine, and it was in front of you.", + "D": "You were looking at an arcade machine, but it was positioned behind you rather than ahead.", + "E": "You were looking at an arcade machine, with it off to your right side instead of straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_12_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What am I doing?", + "answer": "You are taking an elevator downstairs, and the elevator is in front of you.", + "question_chinese": "我在干什么？", + "answer_chinese": "你在坐电梯下楼，电梯在你前面", + "evidence": [ + { + "description": "I'm taking the elevator downstairs. The elevator is in front of you.", + "timestamps": { + "start": "00:11:35", + "end": "00:11:42" + } + } + ], + "question_timestamps": "00:11:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are taking an elevator upstairs.", + "B": "You are waiting outside the elevator.", + "C": "You are taking the stairs downstairs, with the stairwell in front of you instead of an elevator.", + "D": "You are taking an elevator downstairs, and the elevator is in front of you.", + "E": "You are taking an elevator downstairs, but the elevator doors are behind you instead of in front." + }, + "answer_label": "D", + "answer_text": "You are taking an elevator downstairs, and the elevator is in front of you.", + "q_id": "egolife_A1_D3_016_12_T2", + "correct_answer": "You are taking an elevator downstairs, and the elevator is in front of you.", + "original_options": { + "A": "You are taking an elevator upstairs.", + "B": "You are waiting outside the elevator.", + "C": "You are taking the stairs downstairs, with the stairwell in front of you instead of an elevator.", + "D": "You are taking an elevator downstairs, and the elevator is in front of you.", + "E": "You are taking an elevator downstairs, but the elevator doors are behind you instead of in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_7_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which area of the supermarket am I in?", + "answer": "You were in the drinks section, with the drinks in front of you.", + "question_chinese": "我在超市的什么区域？", + "answer_chinese": "你在饮品区，饮品在你前面", + "evidence": [ + { + "description": "You are in the drink area.", + "timestamps": { + "start": "00:05:48", + "end": "00:05:57" + } + } + ], + "question_timestamps": "00:05:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Drinks are behind you.", + "B": "Drinks are to your left.", + "C": "You were in the drinks section, with the drinks to your right as you stood facing down the aisle.", + "D": "You were in the drinks section, with the drinks above you rather than directly in front of you.", + "E": "You were in the drinks section, with the drinks in front of you." + }, + "answer_label": "E", + "answer_text": "You were in the drinks section, with the drinks in front of you.", + "q_id": "egolife_A1_D3_016_7_T1", + "correct_answer": "You were in the drinks section, with the drinks in front of you.", + "original_options": { + "A": "Drinks are behind you.", + "B": "Drinks are to your left.", + "C": "You were in the drinks section, with the drinks to your right as you stood facing down the aisle.", + "D": "You were in the drinks section, with the drinks above you rather than directly in front of you.", + "E": "You were in the drinks section, with the drinks in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_7_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which area of the supermarket am I in?", + "answer": "You were in the snacks section, with the snacks to your right.", + "question_chinese": "我在超市的什么区域？", + "answer_chinese": "你在零食区，零食在你的右边", + "evidence": [ + { + "description": "The shelves are stocked with snacks.", + "timestamps": { + "start": "00:08:25", + "end": "00:08:32" + } + } + ], + "question_timestamps": "00:08:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were in the drinks section.", + "B": "You were in the bakery area.", + "C": "You were in the produce aisle, with the fruits and vegetables lined up beside you on both sides.", + "D": "You were in the checkout area, with the registers ahead as you moved down the front lane.", + "E": "You were in the snacks section, with the snacks to your right." + }, + "answer_label": "E", + "answer_text": "You were in the snacks section, with the snacks to your right.", + "q_id": "egolife_A1_D3_016_7_T2", + "correct_answer": "You were in the snacks section, with the snacks to your right.", + "original_options": { + "A": "You were in the drinks section.", + "B": "You were in the bakery area.", + "C": "You were in the produce aisle, with the fruits and vegetables lined up beside you on both sides.", + "D": "You were in the checkout area, with the registers ahead as you moved down the front lane.", + "E": "You were in the snacks section, with the snacks to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_6_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many people did I see?", + "answer": "You saw a boy in front of you.", + "question_chinese": "我看到多少人？", + "answer_chinese": "你看到一个男生，他在你前面", + "evidence": [ + { + "description": "You see a boy in front of you.", + "timestamps": { + "start": "00:06:26", + "end": "00:06:30" + } + } + ], + "question_timestamps": "00:06:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw no one.", + "B": "You saw two people.", + "C": "You saw two people in front of you, standing close together as you moved toward them.", + "D": "You saw a boy in front of you.", + "E": "You saw three people ahead of you, with the boy among them near the center." + }, + "answer_label": "D", + "answer_text": "You saw a boy in front of you.", + "q_id": "egolife_A1_D3_016_6_T2", + "correct_answer": "You saw a boy in front of you.", + "original_options": { + "A": "You saw no one.", + "B": "You saw two people.", + "C": "You saw two people in front of you, standing close together as you moved toward them.", + "D": "You saw a boy in front of you.", + "E": "You saw three people ahead of you, with the boy among them near the center." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_6_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many people did I see?", + "answer": "You saw three people in front of you.", + "question_chinese": "我看到多少人？", + "answer_chinese": "你看到三个人，他们在你前面", + "evidence": [ + { + "description": "You see two man and a woman ahead of you.", + "timestamps": { + "start": "00:06:37", + "end": "00:06:47" + } + } + ], + "question_timestamps": "00:06:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw two people.", + "B": "You saw four people.", + "C": "You saw three people in front of you.", + "D": "You saw one person in front of you, standing alone as you approached the aisle.", + "E": "You saw five people in front of you, grouped together while you looked straight ahead." + }, + "answer_label": "C", + "answer_text": "You saw three people in front of you.", + "q_id": "egolife_A1_D3_016_6_T1", + "correct_answer": "You saw three people in front of you.", + "original_options": { + "A": "You saw two people.", + "B": "You saw four people.", + "C": "You saw three people in front of you.", + "D": "You saw one person in front of you, standing alone as you approached the aisle.", + "E": "You saw five people in front of you, grouped together while you looked straight ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_4_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the coin pusher machine relative to me?", + "answer": "The coin pusher machine is in front of you.", + "question_chinese": "推币机在我的哪里？", + "answer_chinese": "推币机在你的前面", + "evidence": [ + { + "description": "The coin pusher is in front of you.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:03:08", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The coin pusher machine is in front of you.", + "B": "The coin pusher machine is behind you; you already walked past it." + }, + "answer_label": "A", + "answer_text": "The coin pusher machine is in front of you.", + "q_id": "egolife_A1_D3_016_4_T2", + "correct_answer": "The coin pusher machine is in front of you.", + "original_options": { + "A": "The coin pusher machine is in front of you.", + "B": "The coin pusher machine is behind you; you already walked past it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_4_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the coin pusher machine relative to me?", + "answer": "The coin pusher machine is behind you; you already walked past it.", + "question_chinese": "推币机在我的哪里？", + "answer_chinese": "推币机在你的身后，你经过了它", + "evidence": [ + { + "description": "The coin pusher is behind you.", + "timestamps": { + "start": "00:03:22", + "end": "00:03:26" + } + } + ], + "question_timestamps": "00:03:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The coin pusher machine is in front of you.", + "B": "The coin pusher machine is behind you; you already walked past it." + }, + "answer_label": "B", + "answer_text": "The coin pusher machine is behind you; you already walked past it.", + "q_id": "egolife_A1_D3_016_4_T1", + "correct_answer": "The coin pusher machine is behind you; you already walked past it.", + "original_options": { + "A": "The coin pusher machine is in front of you.", + "B": "The coin pusher machine is behind you; you already walked past it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_13_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the checkout counter relative to me?", + "answer": "The checkout counter is in front of you.", + "question_chinese": "收银台在我哪里？", + "answer_chinese": "收银台在你前面", + "evidence": [ + { + "description": "The cashier desk is in front of you.", + "timestamps": { + "start": "00:10:30", + "end": "00:10:38" + } + } + ], + "question_timestamps": "00:10:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The checkout counter is behind you.", + "B": "The checkout counter is in front of you.", + "C": "The checkout counter is to your left.", + "D": "The checkout counter is to your right as you stand in the aisle facing forward.", + "E": "The checkout counter is off to your left, slightly behind you as you turn away from it." + }, + "answer_label": "B", + "answer_text": "The checkout counter is in front of you.", + "q_id": "egolife_A1_D3_016_13_T2", + "correct_answer": "The checkout counter is in front of you.", + "original_options": { + "A": "The checkout counter is behind you.", + "B": "The checkout counter is in front of you.", + "C": "The checkout counter is to your left.", + "D": "The checkout counter is to your right as you stand in the aisle facing forward.", + "E": "The checkout counter is off to your left, slightly behind you as you turn away from it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_13_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the checkout counter relative to me?", + "answer": "The checkout counter is behind you.", + "question_chinese": "收银台在我哪里？", + "answer_chinese": "收银台在你身后", + "evidence": [ + { + "description": "The checkout counter is behind you.", + "timestamps": { + "start": "00:10:56", + "end": "00:11:03" + } + } + ], + "question_timestamps": "00:11:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The checkout counter is in front of you.", + "B": "The checkout counter is to your left.", + "C": "The checkout counter is to your right as you face down the aisle toward the shelves.", + "D": "The checkout counter is behind you.", + "E": "The checkout counter is ahead of you, positioned in front as you approach the end of the lane." + }, + "answer_label": "D", + "answer_text": "The checkout counter is behind you.", + "q_id": "egolife_A1_D3_016_13_T1", + "correct_answer": "The checkout counter is behind you.", + "original_options": { + "A": "The checkout counter is in front of you.", + "B": "The checkout counter is to your left.", + "C": "The checkout counter is to your right as you face down the aisle toward the shelves.", + "D": "The checkout counter is behind you.", + "E": "The checkout counter is ahead of you, positioned in front as you approach the end of the lane." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_8_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the drink area from me?", + "answer": "To your right.", + "question_chinese": "放饮料的地方在我的什么方向？", + "answer_chinese": "在你的右边", + "evidence": [ + { + "description": "It is on my right", + "timestamps": { + "start": "00:05:45", + "end": "00:05:51" + } + } + ], + "question_timestamps": "00:05:53", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "To your left.", + "B": "To your right." + }, + "answer_label": "B", + "answer_text": "To your right.", + "q_id": "egolife_A1_D3_016_8_T1", + "correct_answer": "To your right.", + "original_options": { + "A": "To your left.", + "B": "To your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_8_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the drink area from me?", + "answer": "It was in front of you.", + "question_chinese": "放饮料的地方在我的什么方向？", + "answer_chinese": "在你的前面", + "evidence": [ + { + "description": "In front of you", + "timestamps": { + "start": "00:05:54", + "end": "00:06:00" + } + } + ], + "question_timestamps": "00:06:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It was in front of you.", + "B": "It was behind you." + }, + "answer_label": "A", + "answer_text": "It was in front of you.", + "q_id": "egolife_A1_D3_016_8_T2", + "correct_answer": "It was in front of you.", + "original_options": { + "A": "It was in front of you.", + "B": "It was behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_9_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Is the pillar to my left or to my right?", + "answer": "To your right.", + "question_chinese": "柱子在我的左边还是右边？", + "answer_chinese": "在你的右边", + "evidence": [ + { + "description": "It is on my right", + "timestamps": { + "start": "00:11:10", + "end": "00:11:24" + } + } + ], + "question_timestamps": "00:11:26", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It’s to your left.", + "B": "To your right." + }, + "answer_label": "B", + "answer_text": "To your right.", + "q_id": "egolife_A1_D3_016_9_T2", + "correct_answer": "To your right.", + "original_options": { + "A": "It’s to your left.", + "B": "To your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_9_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Is the pillar to my left or to my right?", + "answer": "It’s to your left.", + "question_chinese": "柱子在我的左边还是右边？", + "answer_chinese": "在你的左边", + "evidence": [ + { + "description": "It is on my left", + "timestamps": { + "start": "00:11:25", + "end": "00:11:31" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It’s to your left.", + "B": "To your right." + }, + "answer_label": "A", + "answer_text": "It’s to your left.", + "q_id": "egolife_A1_D3_016_9_T1", + "correct_answer": "It’s to your left.", + "original_options": { + "A": "It’s to your left.", + "B": "To your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_2_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the elevator relative to me?", + "answer": "The elevator is to your front-left.", + "question_chinese": "电梯在我哪里？", + "answer_chinese": "电梯在你左前方", + "evidence": [ + { + "description": "The elevator is on your left front.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:48" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Front-right.", + "B": "Back-left.", + "C": "The elevator is directly to your left, not in front of you.", + "D": "The elevator is behind you on the right side, not at your front-left.", + "E": "The elevator is to your front-left." + }, + "answer_label": "E", + "answer_text": "The elevator is to your front-left.", + "q_id": "egolife_A1_D3_016_2_T2", + "correct_answer": "The elevator is to your front-left.", + "original_options": { + "A": "Front-right.", + "B": "Back-left.", + "C": "The elevator is directly to your left, not in front of you.", + "D": "The elevator is behind you on the right side, not at your front-left.", + "E": "The elevator is to your front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_2_T1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the elevator relative to me?", + "answer": "The elevator is in front of you.", + "question_chinese": "电梯在我哪里？", + "answer_chinese": "电梯在你前面", + "evidence": [ + { + "description": "The elevator is in front of you.", + "timestamps": { + "start": "00:11:30", + "end": "00:11:35" + } + } + ], + "question_timestamps": "00:11:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The elevator is slightly behind you, not directly in front of you.", + "D": "The elevator is off to your right side rather than directly in front of you.", + "E": "The elevator is in front of you." + }, + "answer_label": "E", + "answer_text": "The elevator is in front of you.", + "q_id": "egolife_A1_D3_016_2_T1", + "correct_answer": "The elevator is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The elevator is slightly behind you, not directly in front of you.", + "D": "The elevator is off to your right side rather than directly in front of you.", + "E": "The elevator is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_3_T2", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the doorway relative to me?", + "answer": "The doorway is to your right.", + "question_chinese": "门口在我的哪里？", + "answer_chinese": "门口在你的右边", + "evidence": [ + { + "description": "The doorway is on your right.", + "timestamps": { + "start": "00:02:27", + "end": "00:02:33" + } + } + ], + "question_timestamps": "00:02:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The doorway is behind you, not positioned on your right side at the moment.", + "D": "The doorway is to your right.", + "E": "The doorway is ahead and slightly left of you rather than to your right side." + }, + "answer_label": "D", + "answer_text": "The doorway is to your right.", + "q_id": "egolife_A1_D3_016_3_T2", + "correct_answer": "The doorway is to your right.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The doorway is behind you, not positioned on your right side at the moment.", + "D": "The doorway is to your right.", + "E": "The doorway is ahead and slightly left of you rather than to your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_016_3_T1", + "qid_num": 22, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the doorway relative to me?", + "answer": "The doorway is in front of you.", + "question_chinese": "门口在我的哪里？", + "answer_chinese": "门口在你的前面", + "evidence": [ + { + "description": "The door is in front of you.", + "timestamps": { + "start": "00:03:23", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:03:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The doorway is in front of you.", + "B": "Behind you.", + "C": "To your right.", + "D": "The doorway is to your left side instead of directly in front of you.", + "E": "The doorway is behind you and slightly to the left, not straight ahead." + }, + "answer_label": "A", + "answer_text": "The doorway is in front of you.", + "q_id": "egolife_A1_D3_016_3_T1", + "correct_answer": "The doorway is in front of you.", + "original_options": { + "A": "The doorway is in front of you.", + "B": "Behind you.", + "C": "To your right.", + "D": "The doorway is to your left side instead of directly in front of you.", + "E": "The doorway is behind you and slightly to the left, not straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 14, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D3_017-mcq.json b/QAs/egolife_A1_D3_017-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..67caba239588258dfe4a5d6f74e5c5af4d3bd9cf --- /dev/null +++ b/QAs/egolife_A1_D3_017-mcq.json @@ -0,0 +1,938 @@ +[ + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017-mcq_fixed_8", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people pushed this shopping cart through which section of the supermarket just now?", + "answer": "One.", + "question_chinese": "刚刚有几个人在超市的什么区推过这个手推车？", + "answer_chinese": "两个，分别在冷藏区和水果区", + "evidence": [ + { + "description": "I am the first one to push the cart.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:29" + } + }, + { + "description": "Now it's the friend in dark green who is pushing the cart.", + "timestamps": { + "start": "00:09:26", + "end": "00:09:33" + } + } + ], + "question_timestamps": "00:10:21", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four people pushed the shopping cart through that supermarket section just now.", + "E": "Five people were pushing the shopping cart through the section just now." + }, + "answer_label": "B", + "answer_text": "One.", + "q_id": "egolife_A1_D3_017-mcq_fixed_8", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four people pushed the shopping cart through that supermarket section just now.", + "E": "Five people were pushing the shopping cart through the section just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017-mcq_fixed_9", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people pushed this shopping cart through which section of the supermarket just now?", + "answer": "Three.", + "question_chinese": "刚刚有几个人在超市的什么区推过这个手推车？", + "answer_chinese": "三个，分别在冷藏区，水果区和饮品区", + "evidence": [ + { + "description": "I am the first one to push the cart.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:29" + } + }, + { + "description": "Now it's the friend in dark green who is pushing the cart.", + "timestamps": { + "start": "00:09:26", + "end": "00:09:33" + } + }, + { + "description": "Now it's the girl in white who is pushing the cart.", + "timestamps": { + "start": "00:10:29", + "end": "00:10:44" + } + } + ], + "question_timestamps": "00:11:19", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "Two people pushed the shopping cart through that section just now, not three.", + "E": "Four people pushed the cart through the section you just passed, rather than only three." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_017-mcq_fixed_9", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "Two people pushed the shopping cart through that section just now, not three.", + "E": "Four people pushed the cart through the section you just passed, rather than only three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_6_T2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was next to the freezer just now, could I touch the products inside it?", + "answer": "No—you were still a few meters away from the freezer.", + "question_chinese": "刚才在冰柜旁的时候，我可以碰到里面的商品吗？", + "answer_chinese": "不可以，冰柜离你还有几米远", + "evidence": [ + { + "description": "The freezer was close to me and the door was open.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:26" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—you were still a few meters away from the freezer.", + "B": "Yes—you were close enough to reach into the freezer and touch the products inside." + }, + "answer_label": "A", + "answer_text": "No—you were still a few meters away from the freezer.", + "q_id": "egolife_A1_D3_017_6_T2", + "correct_answer": "No—you were still a few meters away from the freezer.", + "original_options": { + "A": "No—you were still a few meters away from the freezer.", + "B": "Yes—you were close enough to reach into the freezer and touch the products inside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_6_T1", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was next to the freezer just now, could I touch the products inside it?", + "answer": "Yes.", + "question_chinese": "刚才在冰柜旁的时候，我可以碰到里面的商品吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The freezer was close to me and the door was open.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:26" + } + }, + { + "description": "The freezer was close to me with no obstacles in between.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:44" + } + } + ], + "question_timestamps": "00:04:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D3_017_6_T1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_7_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was next to the freezer just now, could I touch the products inside it?", + "answer": "No.", + "question_chinese": "刚才在冰柜旁的时候，我可以碰到里面的商品吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "The freezer that passed by here was closed, so I couldn't touch the items inside.", + "timestamps": { + "start": "00:04:24", + "end": "00:04:37" + } + }, + { + "description": "The freezer that passed by here was closed as well, so I couldn't touch the items inside.", + "timestamps": { + "start": "00:05:05", + "end": "00:05:29" + } + } + ], + "question_timestamps": "00:06:51", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D3_017_7_T1", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_7_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was next to the freezer just now, could I touch the products inside it?", + "answer": "Yes, the freezer was right in front of you.", + "question_chinese": "刚才在冰柜旁的时候，我可以碰到里面的商品吗？", + "answer_chinese": "可以，冰柜就在你面前。", + "evidence": [ + { + "description": "The freezer that passed by here was closed, so I couldn't touch the items inside.", + "timestamps": { + "start": "00:04:24", + "end": "00:04:37" + } + }, + { + "description": "The freezer that passed by here was closed as well, so I couldn't touch the items inside.", + "timestamps": { + "start": "00:05:05", + "end": "00:05:29" + } + }, + { + "description": "Refrigerator for storing vegetables", + "timestamps": { + "start": "00:07:41", + "end": "00:07:43" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, the freezer was right in front of you.", + "B": "No, you were still a few meters away from the freezer." + }, + "answer_label": "A", + "answer_text": "Yes, the freezer was right in front of you.", + "q_id": "egolife_A1_D3_017_7_T2", + "correct_answer": "Yes, the freezer was right in front of you.", + "original_options": { + "A": "Yes, the freezer was right in front of you.", + "B": "No, you were still a few meters away from the freezer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What section of the supermarket am I in right now?", + "answer": "Meat and Poultry Sectino.", + "question_chinese": "我刚刚在超市的什么区域？", + "answer_chinese": "肉食品区域", + "evidence": [ + { + "description": "This is in the Meat and Poultry Section.", + "timestamps": { + "start": "00:01:06", + "end": "00:01:12" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Meat and Poultry Sectino.", + "B": "Seafood section.", + "C": "Deli counter.", + "D": "You are in the dairy and eggs section right now, looking at refrigerated cases.", + "E": "You are in the produce section right now, with fruits and vegetables displayed nearby." + }, + "answer_label": "A", + "answer_text": "Meat and Poultry Sectino.", + "q_id": "egolife_A1_D3_017_5", + "correct_answer": "Meat and Poultry Sectino.", + "original_options": { + "A": "Meat and Poultry Sectino.", + "B": "Seafood section.", + "C": "Deli counter.", + "D": "You are in the dairy and eggs section right now, looking at refrigerated cases.", + "E": "You are in the produce section right now, with fruits and vegetables displayed nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What section of the supermarket am I in right now?", + "answer": "The Produce Section.", + "question_chinese": "我刚刚在超市的什么区域？", + "answer_chinese": "水果蔬区菜", + "evidence": [ + { + "description": "This is in the Produce Section.", + "timestamps": { + "start": "00:07:34", + "end": "00:07:58" + } + } + ], + "question_timestamps": "00:08:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The Produce Section.", + "B": "The Dairy Section.", + "C": "The Frozen Foods.", + "D": "You are in the canned goods aisle right now, facing shelves of packaged items.", + "E": "You are in the juice section right now, standing near bottles and cartons on shelves." + }, + "answer_label": "A", + "answer_text": "The Produce Section.", + "q_id": "egolife_A1_D3_017_6", + "correct_answer": "The Produce Section.", + "original_options": { + "A": "The Produce Section.", + "B": "The Dairy Section.", + "C": "The Frozen Foods.", + "D": "You are in the canned goods aisle right now, facing shelves of packaged items.", + "E": "You are in the juice section right now, standing near bottles and cartons on shelves." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_4_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take just now?", + "answer": "You kept going straight at first.", + "question_chinese": "我刚才经历了什么路线？", + "answer_chinese": "你先一直在直行", + "evidence": [ + { + "description": "Igo straight firts", + "timestamps": { + "start": "00:00:21", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left immediately.", + "B": "You kept going straight at first.", + "C": "You turned right first.", + "D": "You walked straight, then turned left and stopped shortly afterward near an aisle end.", + "E": "You made an early right turn, continued forward, and then veered left near the end." + }, + "answer_label": "B", + "answer_text": "You kept going straight at first.", + "q_id": "egolife_A1_D3_017_4_T2", + "correct_answer": "You kept going straight at first.", + "original_options": { + "A": "You turned left immediately.", + "B": "You kept going straight at first.", + "C": "You turned right first.", + "D": "You walked straight, then turned left and stopped shortly afterward near an aisle end.", + "E": "You made an early right turn, continued forward, and then veered left near the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_4_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take just now?", + "answer": "You went straight first, made a small turn, and then continued straight.", + "question_chinese": "我刚才经历了什么路线？", + "answer_chinese": "我先直行，拐了一个小弯后继续直行", + "evidence": [ + { + "description": "Igo straight firts", + "timestamps": { + "start": "00:00:21", + "end": "00:00:31" + } + }, + { + "description": "Make a small turn and then keep going straight.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:49" + } + } + ], + "question_timestamps": "00:02:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went straight the whole time.", + "B": "You turned left, then went straight.", + "C": "You went straight at first, then made a larger turn before continuing straight without any additional turns.", + "D": "You went straight first, made a small turn, and then continued straight.", + "E": "You started by turning slightly, walked straight for a while, and finished with a left turn." + }, + "answer_label": "D", + "answer_text": "You went straight first, made a small turn, and then continued straight.", + "q_id": "egolife_A1_D3_017_4_T1", + "correct_answer": "You went straight first, made a small turn, and then continued straight.", + "original_options": { + "A": "You went straight the whole time.", + "B": "You turned left, then went straight.", + "C": "You went straight at first, then made a larger turn before continuing straight without any additional turns.", + "D": "You went straight first, made a small turn, and then continued straight.", + "E": "You started by turning slightly, walked straight for a while, and finished with a left turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_5_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take just now?", + "answer": "You went straight first, made a small turn and continued straight, then turned left and went straight.", + "question_chinese": "我刚才经历了什么路线？", + "answer_chinese": "我先直行，拐了一个小弯后继续直行，左转后然后直行", + "evidence": [ + { + "description": "Igo straight firts", + "timestamps": { + "start": "00:00:21", + "end": "00:00:31" + } + }, + { + "description": "Make a small turn and then keep going straight.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:49" + } + }, + { + "description": "I continue going straight.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:24" + } + }, + { + "description": "Turn left and go straight.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:13" + } + } + ], + "question_timestamps": "00:04:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You kept going straight.", + "B": "You turned right near the end.", + "C": "You went straight first, made a small turn and continued straight, then turned left and went straight.", + "D": "You went straight first, made a small turn, then turned right and continued straight afterward.", + "E": "You went straight, made a small turn, continued straight, and finished without taking the left turn." + }, + "answer_label": "C", + "answer_text": "You went straight first, made a small turn and continued straight, then turned left and went straight.", + "q_id": "egolife_A1_D3_017_5_T1", + "correct_answer": "You went straight first, made a small turn and continued straight, then turned left and went straight.", + "original_options": { + "A": "You kept going straight.", + "B": "You turned right near the end.", + "C": "You went straight first, made a small turn and continued straight, then turned left and went straight.", + "D": "You went straight first, made a small turn, then turned right and continued straight afterward.", + "E": "You went straight, made a small turn, continued straight, and finished without taking the left turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_5_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take just now?", + "answer": "You went straight first, made a small turn and continued straight, then turned left and went straight, and then turned right.", + "question_chinese": "我刚才经历了什么路线？", + "answer_chinese": "我先直行，拐了一个小弯后继续直行，左转后然后直行，然后在向右转", + "evidence": [ + { + "description": "Igo straight firts", + "timestamps": { + "start": "00:00:21", + "end": "00:00:31" + } + }, + { + "description": "Make a small turn and then keep going straight.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:49" + } + }, + { + "description": "I continue going straight.", + "timestamps": { + "start": "00:02:08", + "end": "00:02:24" + } + }, + { + "description": "Turn left and go straight.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:13" + } + }, + { + "description": "You turned right.", + "timestamps": { + "start": "00:04:40", + "end": "00:04:42" + } + } + ], + "question_timestamps": "00:04:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You ended with a left turn.", + "B": "You turned right twice.", + "C": "You went straight, made a small turn, continued straight, then turned left and finished straight without turning again.", + "D": "You went straight first, made a small turn, continued straight, then turned right and ended with another right turn.", + "E": "You went straight first, made a small turn and continued straight, then turned left and went straight, and then turned right." + }, + "answer_label": "E", + "answer_text": "You went straight first, made a small turn and continued straight, then turned left and went straight, and then turned right.", + "q_id": "egolife_A1_D3_017_5_T2", + "correct_answer": "You went straight first, made a small turn and continued straight, then turned left and went straight, and then turned right.", + "original_options": { + "A": "You ended with a left turn.", + "B": "You turned right twice.", + "C": "You went straight, made a small turn, continued straight, then turned left and finished straight without turning again.", + "D": "You went straight first, made a small turn, continued straight, then turned right and ended with another right turn.", + "E": "You went straight first, made a small turn and continued straight, then turned left and went straight, and then turned right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_1_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which products have I touched?", + "answer": "You were only pushing a supermarket shopping cart.", + "question_chinese": "我都碰过些什么商品？", + "answer_chinese": "你手中只推了个超市推车。", + "evidence": [ + { + "description": "You pushed a supermarket cart.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You touched the meat products.", + "B": "You were only pushing a supermarket shopping cart.", + "C": "You touched the fruit display.", + "D": "You handled the meat products in front of you and also touched some fruit nearby.", + "E": "You touched meat products, fruit, and a green pepper while standing in front of the display." + }, + "answer_label": "B", + "answer_text": "You were only pushing a supermarket shopping cart.", + "q_id": "egolife_A1_D3_017_1_T2", + "correct_answer": "You were only pushing a supermarket shopping cart.", + "original_options": { + "A": "You touched the meat products.", + "B": "You were only pushing a supermarket shopping cart.", + "C": "You touched the fruit display.", + "D": "You handled the meat products in front of you and also touched some fruit nearby.", + "E": "You touched meat products, fruit, and a green pepper while standing in front of the display." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_1_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which products have I touched?", + "answer": "The meat products in front of you.", + "question_chinese": "我都碰过些什么商品？", + "answer_chinese": "你面前的肉类商品", + "evidence": [ + { + "description": "I handled these meat products for a while.", + "timestamps": { + "start": "00:01:52", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:04:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The meat products behind you.", + "B": "The meat products to your left.", + "C": "The meat products in front of you.", + "D": "You touched the meat products to your right rather than the ones directly in front of you.", + "E": "You only handled the meat products behind you, not the ones positioned in front of you." + }, + "answer_label": "C", + "answer_text": "The meat products in front of you.", + "q_id": "egolife_A1_D3_017_1_T1", + "correct_answer": "The meat products in front of you.", + "original_options": { + "A": "The meat products behind you.", + "B": "The meat products to your left.", + "C": "The meat products in front of you.", + "D": "You touched the meat products to your right rather than the ones directly in front of you.", + "E": "You only handled the meat products behind you, not the ones positioned in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_2_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which products have I touched?", + "answer": "You touched the meat products, fruit, and green pepper in front of you.", + "question_chinese": "我都碰过些什么商品？", + "answer_chinese": "你碰了你面前的肉类和水果和青椒。", + "evidence": [ + { + "description": "I handled these meat products for a while.", + "timestamps": { + "start": "00:01:52", + "end": "00:02:00" + } + }, + { + "description": "You touched the green pepper.", + "timestamps": { + "start": "00:08:06", + "end": "00:08:08" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You touched the meat products, fruit, and green pepper in front of you.", + "B": "You touched only the meat products.", + "C": "You touched meat and fruit only.", + "D": "You touched the meat products, fruit, and green pepper to your left rather than in front of you.", + "E": "You handled the meat products and fruit in front of you, but you never touched the green pepper." + }, + "answer_label": "A", + "answer_text": "You touched the meat products, fruit, and green pepper in front of you.", + "q_id": "egolife_A1_D3_017_2_T2", + "correct_answer": "You touched the meat products, fruit, and green pepper in front of you.", + "original_options": { + "A": "You touched the meat products, fruit, and green pepper in front of you.", + "B": "You touched only the meat products.", + "C": "You touched meat and fruit only.", + "D": "You touched the meat products, fruit, and green pepper to your left rather than in front of you.", + "E": "You handled the meat products and fruit in front of you, but you never touched the green pepper." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_017_2_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Which products have I touched?", + "answer": "The meat products and fruit in front of you.", + "question_chinese": "我都碰过些什么商品？", + "answer_chinese": "你面前的肉类和水果", + "evidence": [ + { + "description": "I handled these meat products for a while.", + "timestamps": { + "start": "00:01:52", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The meat products behind you.", + "B": "The meat products and fruit in front of you.", + "C": "The fruit to your left.", + "D": "You touched the meat products and fruit to your right, not the items directly in front of you.", + "E": "You handled fruit in front of you but did not touch any meat products at that spot." + }, + "answer_label": "B", + "answer_text": "The meat products and fruit in front of you.", + "q_id": "egolife_A1_D3_017_2_T1", + "correct_answer": "The meat products and fruit in front of you.", + "original_options": { + "A": "The meat products behind you.", + "B": "The meat products and fruit in front of you.", + "C": "The fruit to your left.", + "D": "You touched the meat products and fruit to your right, not the items directly in front of you.", + "E": "You handled fruit in front of you but did not touch any meat products at that spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D3_018-mcq.json b/QAs/egolife_A1_D3_018-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..bfa5f0209f7b0d539655cb6b44b2199a6b1177e9 --- /dev/null +++ b/QAs/egolife_A1_D3_018-mcq.json @@ -0,0 +1,999 @@ +[ + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the freezer at the place we just stopped at closed or open?", + "answer": "It was close.", + "question_chinese": "刚刚停留的那处的冰柜是关着的还是开着的？", + "answer_chinese": "是关着的", + "evidence": [ + { + "description": "The freezer was closed.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "It was close.", + "B": "It was open." + }, + "answer_label": "A", + "answer_text": "It was close.", + "q_id": "egolife_A1_D3_018_1", + "correct_answer": "It was close.", + "original_options": { + "A": "It was close.", + "B": "It was open." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the freezer at the place we just stopped at closed or open?", + "answer": "The freezer was open.", + "question_chinese": "刚刚停留的那处的冰柜是关着的还是开着的？", + "answer_chinese": "冰柜是开着的", + "evidence": [ + { + "description": "The freezer was closed.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:04:27", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The freezer was closed.", + "B": "The freezer was open." + }, + "answer_label": "B", + "answer_text": "The freezer was open.", + "q_id": "egolife_A1_D3_018_2", + "correct_answer": "The freezer was open.", + "original_options": { + "A": "The freezer was closed.", + "B": "The freezer was open." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_2_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route is that box of fruit I was holding just now currently taking?", + "answer": "You held it in your hand.", + "question_chinese": "刚刚我拿的那盒水果正经历了什么路线？", + "answer_chinese": "你把它拿在手里。", + "evidence": [ + { + "description": "It was first held in my hand", + "timestamps": { + "start": "00:00:49", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You put it onto the cart.", + "B": "You placed it in the cart.", + "C": "You held it in your hand.", + "D": "You held it briefly, then set it down on the cart before moving on down the aisle.", + "E": "You stopped holding it and placed it into the shopping cart, leaving your hands free afterward." + }, + "answer_label": "C", + "answer_text": "You held it in your hand.", + "q_id": "egolife_A1_D3_018_2_T1", + "correct_answer": "You held it in your hand.", + "original_options": { + "A": "You put it onto the cart.", + "B": "You placed it in the cart.", + "C": "You held it in your hand.", + "D": "You held it briefly, then set it down on the cart before moving on down the aisle.", + "E": "You stopped holding it and placed it into the shopping cart, leaving your hands free afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_2_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route is that box of fruit I was holding just now currently taking?", + "answer": "You held it in your hand, then put it onto the cart.", + "question_chinese": "刚刚我拿的那盒水果正经历了什么路线？", + "answer_chinese": "你把它拿在手里,然后放到了手推车上", + "evidence": [ + { + "description": "It was first held in my hand", + "timestamps": { + "start": "00:00:49", + "end": "00:01:03" + } + }, + { + "description": "Then I put it on the handcart.", + "timestamps": { + "start": "00:03:32", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:04:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You held it in your hand, then put it onto the cart.", + "B": "You held it in your hand.", + "C": "You placed it in the cart.", + "D": "You held it in your hand first and kept carrying it as you continued walking down the aisle.", + "E": "You moved it off your hand and set it into the shopping cart without placing it on top." + }, + "answer_label": "A", + "answer_text": "You held it in your hand, then put it onto the cart.", + "q_id": "egolife_A1_D3_018_2_T2", + "correct_answer": "You held it in your hand, then put it onto the cart.", + "original_options": { + "A": "You held it in your hand, then put it onto the cart.", + "B": "You held it in your hand.", + "C": "You placed it in the cart.", + "D": "You held it in your hand first and kept carrying it as you continued walking down the aisle.", + "E": "You moved it off your hand and set it into the shopping cart without placing it on top." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_6_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did that box of fruit I was holding just now take?", + "answer": "You held it in your hand first.", + "question_chinese": "刚刚我拿的那盒水果经历了什么路线？", + "answer_chinese": "它先是被你拿在手上", + "evidence": [ + { + "description": "It was first held in my hand", + "timestamps": { + "start": "00:00:49", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:03:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You carried it in the cart.", + "B": "You held it in your hand first.", + "C": "You put it onto the cart.", + "D": "You placed it on the cart first, then carried it in your hand afterward.", + "E": "You picked it up from the cart, then carried it in your hand for the rest of the aisle." + }, + "answer_label": "B", + "answer_text": "You held it in your hand first.", + "q_id": "egolife_A1_D3_018_6_T1", + "correct_answer": "You held it in your hand first.", + "original_options": { + "A": "You carried it in the cart.", + "B": "You held it in your hand first.", + "C": "You put it onto the cart.", + "D": "You placed it on the cart first, then carried it in your hand afterward.", + "E": "You picked it up from the cart, then carried it in your hand for the rest of the aisle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_6_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did that box of fruit I was holding just now take?", + "answer": "You held it in your hand first, then put it into the shopping cart.", + "question_chinese": "刚刚我拿的那盒水果经历了什么路线？", + "answer_chinese": "它先是被你拿在手上，放进来购物车。", + "evidence": [ + { + "description": "It was first held in my hand", + "timestamps": { + "start": "00:00:49", + "end": "00:01:03" + } + }, + { + "description": "Then I put it on the handcart.", + "timestamps": { + "start": "00:03:32", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:08:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You held it in your hand.", + "B": "You put it onto the cart.", + "C": "You held it in your hand first, then put it into the shopping cart.", + "D": "You held it in your hand first, then set it down on top of the cart instead.", + "E": "You carried it in your hand the whole time and never placed it into the shopping cart." + }, + "answer_label": "C", + "answer_text": "You held it in your hand first, then put it into the shopping cart.", + "q_id": "egolife_A1_D3_018_6_T2", + "correct_answer": "You held it in your hand first, then put it into the shopping cart.", + "original_options": { + "A": "You held it in your hand.", + "B": "You put it onto the cart.", + "C": "You held it in your hand first, then put it into the shopping cart.", + "D": "You held it in your hand first, then set it down on top of the cart instead.", + "E": "You carried it in your hand the whole time and never placed it into the shopping cart." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding in my hand?", + "answer": "I took a box of fruits and my phone.", + "question_chinese": "刚刚我的手上拿了什么？", + "answer_chinese": "拿了一盒水果和手机", + "evidence": [ + { + "description": "I'm holding a box of fruits in my hand.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:57" + } + }, + { + "description": "I took out my phone again.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:13" + } + } + ], + "question_timestamps": "00:06:44", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A box of fruits only.", + "B": "I took a box of fruits and my phone.", + "C": "Your phone only.", + "D": "You were holding a box of fruits and your keys in the same hand.", + "E": "You picked up a box of fruits along with a power bank, not the phone." + }, + "answer_label": "B", + "answer_text": "I took a box of fruits and my phone.", + "q_id": "egolife_A1_D3_018_5", + "correct_answer": "I took a box of fruits and my phone.", + "original_options": { + "A": "A box of fruits only.", + "B": "I took a box of fruits and my phone.", + "C": "Your phone only.", + "D": "You were holding a box of fruits and your keys in the same hand.", + "E": "You picked up a box of fruits along with a power bank, not the phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018-mcq_fixed_5", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "Fruits, the phone and the power bank.", + "question_chinese": "刚刚我的手上拿了什么？", + "answer_chinese": "水果手机和充宝电", + "evidence": [ + { + "description": "I'm holding a box of fruits in my hand.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:57" + } + }, + { + "description": "I took out my phone again.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:13" + } + }, + { + "description": "I'm holding a power bank in my hand.", + "timestamps": { + "start": "00:07:39", + "end": "00:07:53" + } + } + ], + "question_timestamps": "00:08:23", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Fruits and the phone.", + "B": "Phone and power bank.", + "C": "Fruits, the phone and the power bank.", + "D": "You were holding fruits along with your keys and wallet instead of the phone and power bank.", + "E": "You carried fruits and a phone, but the third item was your charger, not a power bank." + }, + "answer_label": "C", + "answer_text": "Fruits, the phone and the power bank.", + "q_id": "egolife_A1_D3_018-mcq_fixed_5", + "correct_answer": "Fruits, the phone and the power bank.", + "original_options": { + "A": "Fruits and the phone.", + "B": "Phone and power bank.", + "C": "Fruits, the phone and the power bank.", + "D": "You were holding fruits along with your keys and wallet instead of the phone and power bank.", + "E": "You carried fruits and a phone, but the third item was your charger, not a power bank." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018-mcq_fixed_2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did that box of fruit take just now?", + "answer": "It was first held in your hand.", + "question_chinese": "刚刚那盒水果经历了什么路线？", + "answer_chinese": "它先是被我拿在手上", + "evidence": [ + { + "description": "It was first held in my hand", + "timestamps": { + "start": "00:00:49", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:03:33", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it onto the cart.", + "B": "You placed it in the cart.", + "C": "You held it in your hand, then set it onto the cart right afterward.", + "D": "You first held it in your hand, then put it into the shopping cart to carry it along.", + "E": "It was first held in your hand." + }, + "answer_label": "E", + "answer_text": "It was first held in your hand.", + "q_id": "egolife_A1_D3_018-mcq_fixed_2", + "correct_answer": "It was first held in your hand.", + "original_options": { + "A": "You put it onto the cart.", + "B": "You placed it in the cart.", + "C": "You held it in your hand, then set it onto the cart right afterward.", + "D": "You first held it in your hand, then put it into the shopping cart to carry it along.", + "E": "It was first held in your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_4", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did that box of fruit just take?", + "answer": "I held it in my hand first and then put it on the trolley.", + "question_chinese": "刚刚那盒水果经历了什么路线？", + "answer_chinese": "我先把它拿在手里,然后放到了手推车上", + "evidence": [ + { + "description": "It was first held in my hand", + "timestamps": { + "start": "00:00:49", + "end": "00:01:03" + } + }, + { + "description": "Then I put it on the handcart.", + "timestamps": { + "start": "00:03:32", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:04:49", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I put it straight on the counter.", + "B": "I set it down on the floor.", + "C": "I carried it from the shelf and placed it on the counter after holding it briefly.", + "D": "I lifted it off the ground and moved it into the cabinet before letting go.", + "E": "I held it in my hand first and then put it on the trolley." + }, + "answer_label": "E", + "answer_text": "I held it in my hand first and then put it on the trolley.", + "q_id": "egolife_A1_D3_018_4", + "correct_answer": "I held it in my hand first and then put it on the trolley.", + "original_options": { + "A": "I put it straight on the counter.", + "B": "I set it down on the floor.", + "C": "I carried it from the shelf and placed it on the counter after holding it briefly.", + "D": "I lifted it off the ground and moved it into the cabinet before letting go.", + "E": "I held it in my hand first and then put it on the trolley." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_4_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After that brief stop just now, what route am I currently taking?", + "answer": "You turned left first.", + "question_chinese": "刚才短暂停留后我正在经历什么路线？", + "answer_chinese": "你先左转", + "evidence": [ + { + "description": "I turn left first.", + "timestamps": { + "start": "00:05:56", + "end": "00:06:08" + } + } + ], + "question_timestamps": "00:06:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right first.", + "B": "You turned left first.", + "C": "You walked straight.", + "D": "You went straight ahead instead of turning left at the start of the route.", + "E": "You turned right first and continued forward rather than beginning with a left turn." + }, + "answer_label": "B", + "answer_text": "You turned left first.", + "q_id": "egolife_A1_D3_018_4_T2", + "correct_answer": "You turned left first.", + "original_options": { + "A": "You turned right first.", + "B": "You turned left first.", + "C": "You walked straight.", + "D": "You went straight ahead instead of turning left at the start of the route.", + "E": "You turned right first and continued forward rather than beginning with a left turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_4_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After that brief stop just now, what route am I currently taking?", + "answer": "You turned left first, then turned right, and then turned right again.", + "question_chinese": "刚才短暂停留后我正在经历什么路线？", + "answer_chinese": "你先左转，然后右转然后再右转", + "evidence": [ + { + "description": "I turn left first.", + "timestamps": { + "start": "00:05:56", + "end": "00:06:08" + } + }, + { + "description": "Then I turn right.", + "timestamps": { + "start": "00:06:48", + "end": "00:06:54" + } + }, + { + "description": "Then I turned right again.", + "timestamps": { + "start": "00:07:05", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right first.", + "B": "You turned left twice first.", + "C": "You turned left first, then turned right, and then turned right again.", + "D": "You turned left first, then turned left again, and finally turned right once.", + "E": "You turned right first, then turned right again, and then turned right one more time." + }, + "answer_label": "C", + "answer_text": "You turned left first, then turned right, and then turned right again.", + "q_id": "egolife_A1_D3_018_4_T1", + "correct_answer": "You turned left first, then turned right, and then turned right again.", + "original_options": { + "A": "You turned right first.", + "B": "You turned left twice first.", + "C": "You turned left first, then turned right, and then turned right again.", + "D": "You turned left first, then turned left again, and finally turned right once.", + "E": "You turned right first, then turned right again, and then turned right one more time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_8_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After that brief stop just now, what route did I take?", + "answer": "You walked straight.", + "question_chinese": "刚才短暂停留后我经历了什么路线？", + "answer_chinese": "你直走了。", + "evidence": [ + { + "description": "I go straight first.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:48" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left first.", + "B": "You walked straight.", + "C": "You turned right first.", + "D": "You walked straight first, then turned around and headed back the other way.", + "E": "You turned left first, then turned right, and then turned right again." + }, + "answer_label": "B", + "answer_text": "You walked straight.", + "q_id": "egolife_A1_D3_018_8_T2", + "correct_answer": "You walked straight.", + "original_options": { + "A": "You turned left first.", + "B": "You walked straight.", + "C": "You turned right first.", + "D": "You walked straight first, then turned around and headed back the other way.", + "E": "You turned left first, then turned right, and then turned right again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_8_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After that brief stop just now, what route did I take?", + "answer": "You walked straight first, then turned around.", + "question_chinese": "刚才短暂停留后我经历了什么路线？", + "answer_chinese": "你先直走，然后转身", + "evidence": [ + { + "description": "I go straight first.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:48" + } + }, + { + "description": "Then I turn around.", + "timestamps": { + "start": "00:03:56", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:05:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked straight.", + "B": "You turned around immediately.", + "C": "You turned left first, then turned right, and then turned right again.", + "D": "You walked straight first, then turned around.", + "E": "You turned left first and kept going without walking straight and turning back." + }, + "answer_label": "D", + "answer_text": "You walked straight first, then turned around.", + "q_id": "egolife_A1_D3_018_8_T1", + "correct_answer": "You walked straight first, then turned around.", + "original_options": { + "A": "You walked straight.", + "B": "You turned around immediately.", + "C": "You turned left first, then turned right, and then turned right again.", + "D": "You walked straight first, then turned around.", + "E": "You turned left first and kept going without walking straight and turning back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_8", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I experience after a brief stop just now?", + "answer": "I turned left first, then right, and right again.", + "question_chinese": "刚才短暂停留后我经历了什么路线？", + "answer_chinese": "我先左转，然后右转然后再右转", + "evidence": [ + { + "description": "I turn left first.", + "timestamps": { + "start": "00:05:56", + "end": "00:06:08" + } + }, + { + "description": "Then I turn right.", + "timestamps": { + "start": "00:06:48", + "end": "00:06:54" + } + }, + { + "description": "Then I turned right again.", + "timestamps": { + "start": "00:07:05", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "I turned right first.", + "B": "I turned left first, then right, and right again.", + "C": "I walked straight ahead.", + "D": "I turned left first, then turned right, and then turned left again instead of right.", + "E": "I turned right first, then turned left, and then turned left again after the stop." + }, + "answer_label": "B", + "answer_text": "I turned left first, then right, and right again.", + "q_id": "egolife_A1_D3_018_8", + "correct_answer": "I turned left first, then right, and right again.", + "original_options": { + "A": "I turned right first.", + "B": "I turned left first, then right, and right again.", + "C": "I walked straight ahead.", + "D": "I turned left first, then turned right, and then turned left again instead of right.", + "E": "I turned right first, then turned left, and then turned left again after the stop." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_3_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What have I held in my hand?", + "answer": "You picked up a box of fruit.", + "question_chinese": "我的手上拿过了什么？", + "answer_chinese": "你拿了一盒水果。", + "evidence": [ + { + "description": "I'm holding a box of fruits in my hand.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:02:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked up a box of fruit.", + "B": "You held a phone.", + "C": "You held a shopping bag.", + "D": "You picked up a phone first and then set it down again soon after.", + "E": "You held a phone and then carried it while walking, without picking up any fruit box." + }, + "answer_label": "A", + "answer_text": "You picked up a box of fruit.", + "q_id": "egolife_A1_D3_018_3_T2", + "correct_answer": "You picked up a box of fruit.", + "original_options": { + "A": "You picked up a box of fruit.", + "B": "You held a phone.", + "C": "You held a shopping bag.", + "D": "You picked up a phone first and then set it down again soon after.", + "E": "You held a phone and then carried it while walking, without picking up any fruit box." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_018_3_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What have I held in my hand?", + "answer": "You picked up a box of fruit and a phone.", + "question_chinese": "我的手上拿过了什么？", + "answer_chinese": "你拿了一盒水果和手机。", + "evidence": [ + { + "description": "I'm holding a box of fruits in my hand.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:57" + } + }, + { + "description": "I took out my phone again.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:13" + } + } + ], + "question_timestamps": "00:06:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You picked up a box of fruit.", + "B": "You picked up a phone.", + "C": "You only picked up the box of fruit and never held a phone in your hand.", + "D": "You picked up a box of fruit and a phone.", + "E": "You held a phone briefly but didn’t pick up the box of fruit at any point." + }, + "answer_label": "D", + "answer_text": "You picked up a box of fruit and a phone.", + "q_id": "egolife_A1_D3_018_3_T1", + "correct_answer": "You picked up a box of fruit and a phone.", + "original_options": { + "A": "You picked up a box of fruit.", + "B": "You picked up a phone.", + "C": "You only picked up the box of fruit and never held a phone in your hand.", + "D": "You picked up a box of fruit and a phone.", + "E": "You held a phone briefly but didn’t pick up the box of fruit at any point." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D3_019-mcq.json b/QAs/egolife_A1_D3_019-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c18c9969a6a454d13da161aa725f8104b447bc3e --- /dev/null +++ b/QAs/egolife_A1_D3_019-mcq.json @@ -0,0 +1,618 @@ +[ + { + "video_uid": "egolife_A1_D3_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_019_5", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route have I taken from the moment I finished collecting the silver until now?", + "answer": "I went straight first, then turned right.", + "question_chinese": "从收完银到现在，我经历了什么路线？", + "answer_chinese": "我先直行，再右拐", + "evidence": [ + { + "description": "I go straight first.", + "timestamps": { + "start": "00:10:51", + "end": "00:10:57" + } + }, + { + "description": "Then turn right.", + "timestamps": { + "start": "00:11:01", + "end": "00:11:12" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I went straight, then turned left.", + "B": "I turned right, then went straight.", + "C": "I went straight at first, then turned left after passing the corner.", + "D": "I went straight first, then turned right.", + "E": "I turned right immediately, then continued straight for a short distance afterward." + }, + "answer_label": "D", + "answer_text": "I went straight first, then turned right.", + "q_id": "egolife_A1_D3_019_5", + "correct_answer": "I went straight first, then turned right.", + "original_options": { + "A": "I went straight, then turned left.", + "B": "I turned right, then went straight.", + "C": "I went straight at first, then turned left after passing the corner.", + "D": "I went straight first, then turned right.", + "E": "I turned right immediately, then continued straight for a short distance afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_019_6", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route have I taken from the time I finished collecting the money until now?", + "answer": "I went go straight, thened turn right, and took the escalator.", + "question_chinese": "从收完银到现在，我经历了什么路线？", + "answer_chinese": "我先直行再右转再上扶梯", + "evidence": [ + { + "description": "I go straight first.", + "timestamps": { + "start": "00:10:51", + "end": "00:10:57" + } + }, + { + "description": "Then turn right.", + "timestamps": { + "start": "00:11:01", + "end": "00:11:12" + } + }, + { + "description": "Then I take the escalator.", + "timestamps": { + "start": "00:11:53", + "end": "00:12:00" + } + } + ], + "question_timestamps": "00:12:03", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "I went go straight, thened turn right, and took the escalator.", + "B": "I went straight, then turned left.", + "C": "I went straight, then took the stairs.", + "D": "I went straight first, turned right afterward, but then used the stairs instead of the escalator.", + "E": "I went straight ahead, then turned left, and finally took the elevator rather than the escalator." + }, + "answer_label": "A", + "answer_text": "I went go straight, thened turn right, and took the escalator.", + "q_id": "egolife_A1_D3_019_6", + "correct_answer": "I went go straight, thened turn right, and took the escalator.", + "original_options": { + "A": "I went go straight, thened turn right, and took the escalator.", + "B": "I went straight, then turned left.", + "C": "I went straight, then took the stairs.", + "D": "I went straight first, turned right afterward, but then used the stairs instead of the escalator.", + "E": "I went straight ahead, then turned left, and finally took the elevator rather than the escalator." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_019_7", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the items on the shelf just now?", + "answer": "Yes.", + "question_chinese": "刚才我能碰到货架上的物品吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "When passing by the shelf, I was very close to it and could reach the items on top.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:15" + } + }, + { + "description": "When passing by the shelf, I was very close to it and could reach the items on top.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:49" + } + } + ], + "question_timestamps": "00:08:24", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D3_019_7", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_019_8", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the items on the shelf just now?", + "answer": "No.", + "question_chinese": "刚才我能碰到货架上的物品吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "I was just checking out and was a bit far from the shelf.", + "timestamps": { + "start": "00:09:44", + "end": "00:09:55" + } + }, + { + "description": "I was just checking out and was a bit far from the shelf.", + "timestamps": { + "start": "00:10:08", + "end": "00:10:16" + } + } + ], + "question_timestamps": "00:11:02", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D3_019_8", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_019_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What area of the supermarket am I in right now?", + "answer": "ERfrigeration block.", + "question_chinese": "我刚刚在超市的什么区域？", + "answer_chinese": "冷藏区", + "evidence": [ + { + "description": "This is the refrigerated section of the supermarket.", + "timestamps": { + "start": "00:01:45", + "end": "00:01:54" + } + } + ], + "question_timestamps": "00:03:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "ERfrigeration block.", + "B": "Dairy section.", + "C": "Beverage aisle.", + "D": "Meat section near the bakery area, not by the refrigeration block at all.", + "E": "Produce section near the bakery area rather than staying by the refrigeration block now." + }, + "answer_label": "A", + "answer_text": "ERfrigeration block.", + "q_id": "egolife_A1_D3_019_1", + "correct_answer": "ERfrigeration block.", + "original_options": { + "A": "ERfrigeration block.", + "B": "Dairy section.", + "C": "Beverage aisle.", + "D": "Meat section near the bakery area, not by the refrigeration block at all.", + "E": "Produce section near the bakery area rather than staying by the refrigeration block now." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_019_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What area of the supermarket am I in right now?", + "answer": "Fruit section.", + "question_chinese": "我刚刚在超市的什么区域？", + "answer_chinese": "水果区", + "evidence": [ + { + "description": "This is the fruit section of the supermarket.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:05:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Vegetable section.", + "B": "Fruit section.", + "C": "Bakery section.", + "D": "Fresh juice section, not the fruit section you’re thinking of right now.", + "E": "Deli counter area rather than standing in the fruit section right now." + }, + "answer_label": "B", + "answer_text": "Fruit section.", + "q_id": "egolife_A1_D3_019_2", + "correct_answer": "Fruit section.", + "original_options": { + "A": "Vegetable section.", + "B": "Fruit section.", + "C": "Bakery section.", + "D": "Fresh juice section, not the fruit section you’re thinking of right now.", + "E": "Deli counter area rather than standing in the fruit section right now." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_019_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What do I just have in my hand?", + "answer": "I held a helmet and a phone.", + "question_chinese": "我的手里刚刚拿了什么？", + "answer_chinese": "拿着头盔和手机", + "evidence": [ + { + "description": "Holding a helmet and a phone", + "timestamps": { + "start": "00:02:32", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:04:39", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I held a wallet and keys.", + "B": "I held a cup and flashlight.", + "C": "I held a book in one hand and a remote control in the other.", + "D": "I held a tablet and a pair of glasses just a moment ago.", + "E": "I held a helmet and a phone." + }, + "answer_label": "E", + "answer_text": "I held a helmet and a phone.", + "q_id": "egolife_A1_D3_019_3", + "correct_answer": "I held a helmet and a phone.", + "original_options": { + "A": "I held a wallet and keys.", + "B": "I held a cup and flashlight.", + "C": "I held a book in one hand and a remote control in the other.", + "D": "I held a tablet and a pair of glasses just a moment ago.", + "E": "I held a helmet and a phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_019_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What do I just have in my hand?", + "answer": "I carried a helmet and phone, along with a plastic bag and shopping cart.", + "question_chinese": "我的手里刚刚拿了什么？", + "answer_chinese": "我拿着头盔手机，塑料袋和手推车", + "evidence": [ + { + "description": "Holding a helmet and a phone", + "timestamps": { + "start": "00:02:32", + "end": "00:02:45" + } + }, + { + "description": "At this moment, I am holding a plastic bag.", + "timestamps": { + "start": "00:08:31", + "end": "00:08:43" + } + }, + { + "description": "At this moment, I am pushing a shopping cart.", + "timestamps": { + "start": "00:10:06", + "end": "00:10:17" + } + } + ], + "question_timestamps": "00:11:05", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "I carried a backpack and tablet.", + "B": "I held a basket and keys.", + "C": "I carried a helmet and phone, along with a plastic bag and shopping cart.", + "D": "I just carried a purse and wallet, along with a magazine and sunglasses.", + "E": "I held a camera and laptop, along with a thermos and a jacket." + }, + "answer_label": "C", + "answer_text": "I carried a helmet and phone, along with a plastic bag and shopping cart.", + "q_id": "egolife_A1_D3_019_4", + "correct_answer": "I carried a helmet and phone, along with a plastic bag and shopping cart.", + "original_options": { + "A": "I carried a backpack and tablet.", + "B": "I held a basket and keys.", + "C": "I carried a helmet and phone, along with a plastic bag and shopping cart.", + "D": "I just carried a purse and wallet, along with a magazine and sunglasses.", + "E": "I held a camera and laptop, along with a thermos and a jacket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_019-mcq_fixed_8", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have pushed our shopping cart through which section of the supermarket so far?", + "answer": "Five.", + "question_chinese": "目前为止，一共有几个人在超市的什么区推过我们的那辆手推车？", + "answer_chinese": "一个，在冷藏区", + "evidence": [ + { + "description": "The friend in white is pushing the shopping cart.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:52" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "Three people have pushed our shopping cart through that supermarket section so far.", + "D": "Five.", + "E": "Seven different people have taken turns pushing our cart through that supermarket section so far." + }, + "answer_label": "D", + "answer_text": "Five.", + "q_id": "egolife_A1_D3_019-mcq_fixed_8", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "Three people have pushed our shopping cart through that supermarket section so far.", + "D": "Five.", + "E": "Seven different people have taken turns pushing our cart through that supermarket section so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_019-mcq_fixed_9", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have pushed our shopping cart through which section of the supermarket so far?", + "answer": "Four.", + "question_chinese": "目前为止，一共有几个人在超市的什么区推过我们的那辆手推车？", + "answer_chinese": "两个，分别在冷藏区和饮品区", + "evidence": [ + { + "description": "The friend in white is pushing the shopping cart.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:52" + } + }, + { + "description": "The friend in the patterned clothes is pushing the shopping cart\n.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:01" + } + } + ], + "question_timestamps": "00:04:29", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Two people have pushed our shopping cart through that supermarket section so far.", + "E": "Six people have pushed our shopping cart through that supermarket section up to now." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_019-mcq_fixed_9", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Two people have pushed our shopping cart through that supermarket section so far.", + "E": "Six people have pushed our shopping cart through that supermarket section up to now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D3_020-mcq.json b/QAs/egolife_A1_D3_020-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5ef5d921ba3d1d0804ec0e347bdbc5b015918811 --- /dev/null +++ b/QAs/egolife_A1_D3_020-mcq.json @@ -0,0 +1,844 @@ +[ + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020-mcq_fixed_8", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people touched my phone and where did it happen just now?", + "answer": "Four.", + "question_chinese": "刚刚有几个人在什么地方碰过我的手机？", + "answer_chinese": "只有我一个，在墙旁边", + "evidence": [ + { + "description": "My phone has only been touched by me.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:24" + } + } + ], + "question_timestamps": "00:04:44", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Two people touched your phone during that moment just now, not four people in total.", + "E": "Six people ended up touching your phone, all during that moment just now." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_020-mcq_fixed_8", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Two people touched your phone during that moment just now, not four people in total.", + "E": "Six people ended up touching your phone, all during that moment just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020-mcq_fixed_9", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people touched my phone and where did it happen just now?", + "answer": "Three.", + "question_chinese": "刚刚有几个人在什么地方碰过我的手机？", + "answer_chinese": "两个人，在墙边", + "evidence": [ + { + "description": "My phone has only been touched by me.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:24" + } + }, + { + "description": "The friend in colorful clothes is using my phone at the moment.", + "timestamps": { + "start": "00:06:39", + "end": "00:07:06" + } + } + ], + "question_timestamps": "00:10:51", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "Five people touched your phone during that moment just now, not three people total.", + "E": "Six people have touched your phone during the moment that just happened." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_020-mcq_fixed_9", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "Five people touched your phone during that moment just now, not three people total.", + "E": "Six people have touched your phone during the moment that just happened." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_9", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have you just taken the escalator?", + "answer": "Once.", + "question_chinese": "刚刚走了几次扶梯？", + "answer_chinese": "一次", + "evidence": [ + { + "description": "This is my first time taking the escalato.r", + "timestamps": { + "start": "00:00:01", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Three times.", + "C": "You took the escalator two times just now, not once during that stretch.", + "D": "You ended up taking the escalator four separate times during that short stretch.", + "E": "Once." + }, + "answer_label": "E", + "answer_text": "Once.", + "q_id": "egolife_A1_D3_020_9", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Three times.", + "C": "You took the escalator two times just now, not once during that stretch.", + "D": "You ended up taking the escalator four separate times during that short stretch.", + "E": "Once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_10", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many escalators have you just gone up and down?", + "answer": "Twice.", + "question_chinese": "刚刚走了几次扶梯？", + "answer_chinese": "两次", + "evidence": [ + { + "description": "This is my first time taking the escalato.r", + "timestamps": { + "start": "00:00:01", + "end": "00:00:22" + } + }, + { + "description": "This is my second time taking the escalator.", + "timestamps": { + "start": "00:00:51", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:03:46", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "You went up and down on one escalator only, not two in total.", + "E": "You went up and down using four escalators in total just now." + }, + "answer_label": "A", + "answer_text": "Twice.", + "q_id": "egolife_A1_D3_020_10", + "correct_answer": "Twice.", + "original_options": { + "A": "Twice.", + "B": "Once.", + "C": "Three times.", + "D": "You went up and down on one escalator only, not two in total.", + "E": "You went up and down using four escalators in total just now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the door I just entered from now?", + "answer": "It is on my left.", + "question_chinese": "刚才进的门现在我的什么方向？", + "answer_chinese": "在我的左边", + "evidence": [ + { + "description": "After entering this door, we have been staying here, with the door now behind me.", + "timestamps": { + "start": "00:01:40", + "end": "00:02:05" + } + }, + { + "description": "Then I turned sideways, and now the door is on my left\n.", + "timestamps": { + "start": "00:02:39", + "end": "00:02:49" + } + } + ], + "question_timestamps": "00:04:39", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It is on my right.", + "B": "It is on my left.", + "C": "It is straight ahead.", + "D": "It is behind me now, back the way I came in, not on my left side.", + "E": "It is slightly to my right, near where I first entered the room." + }, + "answer_label": "B", + "answer_text": "It is on my left.", + "q_id": "egolife_A1_D3_020_2", + "correct_answer": "It is on my left.", + "original_options": { + "A": "It is on my right.", + "B": "It is on my left.", + "C": "It is straight ahead.", + "D": "It is behind me now, back the way I came in, not on my left side.", + "E": "It is slightly to my right, near where I first entered the room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the door I just entered from now?", + "answer": "It is onm y ight.", + "question_chinese": "刚才进的门现在我的什么方向？", + "answer_chinese": "在我的右边", + "evidence": [ + { + "description": "After entering this door, we have been staying here, with the door now behind me.", + "timestamps": { + "start": "00:01:40", + "end": "00:02:05" + } + }, + { + "description": "Then I turned sideways, and now the door is on my left\n.", + "timestamps": { + "start": "00:02:39", + "end": "00:02:49" + } + }, + { + "description": "After walking straight ahead, I turned around, and now the door is in front of me.", + "timestamps": { + "start": "00:04:18", + "end": "00:05:05" + } + }, + { + "description": "I turned sideways again, and now the door is on my right.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:39" + } + } + ], + "question_timestamps": "00:07:42", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "It is on my left.", + "B": "It is straight ahead.", + "C": "It is behind me now, in the direction I walked from earlier.", + "D": "It is in front of me now, roughly aligned with where I entered.", + "E": "It is onm y ight." + }, + "answer_label": "E", + "answer_text": "It is onm y ight.", + "q_id": "egolife_A1_D3_020_4", + "correct_answer": "It is onm y ight.", + "original_options": { + "A": "It is on my left.", + "B": "It is straight ahead.", + "C": "It is behind me now, in the direction I walked from earlier.", + "D": "It is in front of me now, roughly aligned with where I entered.", + "E": "It is onm y ight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_7", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just meet my friend in the green clothes?", + "answer": "Yes.", + "question_chinese": "我刚刚可以碰到绿色衣服的朋友吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The friend in green clothes is right beside my feet.", + "timestamps": { + "start": "00:03:38", + "end": "00:03:51" + } + } + ], + "question_timestamps": "00:05:24", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D3_020_7", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_8", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just meet my friend in the green clothes?", + "answer": "No.", + "question_chinese": "我刚刚可以碰到绿色衣服的朋友吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "She is in front of me, but a little far away.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:48" + } + } + ], + "question_timestamps": "00:08:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D3_020_8", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_4_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I just ride the escalator?", + "answer": "You just took the escalator once, and the escalator is behind you.", + "question_chinese": "我刚刚走了几次扶梯？", + "answer_chinese": "你刚刚走了1次扶梯，扶梯在你后面。", + "evidence": [ + { + "description": "This is my first time taking the escalato.r", + "timestamps": { + "start": "00:00:01", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:00:26", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "You just took the escalator once, and the escalator is behind you.", + "C": "Three times.", + "D": "You just took the escalator twice, and the escalator is behind you.", + "E": "You didn't ride the escalator at all, and the escalator is still ahead of you." + }, + "answer_label": "B", + "answer_text": "You just took the escalator once, and the escalator is behind you.", + "q_id": "egolife_A1_D3_020_4_T2", + "correct_answer": "You just took the escalator once, and the escalator is behind you.", + "original_options": { + "A": "Twice.", + "B": "You just took the escalator once, and the escalator is behind you.", + "C": "Three times.", + "D": "You just took the escalator twice, and the escalator is behind you.", + "E": "You didn't ride the escalator at all, and the escalator is still ahead of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_4_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I just ride the escalator?", + "answer": "You just took the escalator twice, and the escalator is behind you.", + "question_chinese": "我刚刚走了几次扶梯？", + "answer_chinese": "你刚刚走了两次扶梯，扶梯在你后面。", + "evidence": [ + { + "description": "This is my first time taking the escalato.r", + "timestamps": { + "start": "00:00:01", + "end": "00:00:22" + } + }, + { + "description": "This is my second time taking the escalato.r", + "timestamps": { + "start": "00:01:00", + "end": "00:01:29" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You just took the escalator twice, and the escalator is behind you.", + "B": "Once.", + "C": "Three times.", + "D": "You just took the escalator once, and the escalator is behind you.", + "E": "You just took the escalator four times, and the escalator is behind you." + }, + "answer_label": "A", + "answer_text": "You just took the escalator twice, and the escalator is behind you.", + "q_id": "egolife_A1_D3_020_4_T1", + "correct_answer": "You just took the escalator twice, and the escalator is behind you.", + "original_options": { + "A": "You just took the escalator twice, and the escalator is behind you.", + "B": "Once.", + "C": "Three times.", + "D": "You just took the escalator once, and the escalator is behind you.", + "E": "You just took the escalator four times, and the escalator is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_3", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did my phone just go through?", + "answer": "It was first held in my hand and then placed on the groun.d.", + "question_chinese": "我的手机刚刚经历了什么路线？", + "answer_chinese": "它先一直被我拿在手上，然后被我放在地上", + "evidence": [ + { + "description": "I held it in my hand continuously at first.", + "timestamps": { + "start": "00:02:29", + "end": "00:02:53" + } + }, + { + "description": "Then I placed it on the ground.", + "timestamps": { + "start": "00:05:05", + "end": "00:05:18" + } + } + ], + "question_timestamps": "00:07:19", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Hand to pocket.", + "B": "Pocket to ground.", + "C": "It was first held in my hand and then placed on the groun.d.", + "D": "It was first kept in your pocket, then taken out and held in your hand.", + "E": "It was first held in your hand, then placed into your pocket, not on the ground." + }, + "answer_label": "C", + "answer_text": "It was first held in my hand and then placed on the groun.d.", + "q_id": "egolife_A1_D3_020_3", + "correct_answer": "It was first held in my hand and then placed on the groun.d.", + "original_options": { + "A": "Hand to pocket.", + "B": "Pocket to ground.", + "C": "It was first held in my hand and then placed on the groun.d.", + "D": "It was first kept in your pocket, then taken out and held in your hand.", + "E": "It was first held in your hand, then placed into your pocket, not on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did my phone just go through?", + "answer": "I held it in my hand continuously, then placed it on the ground, and finally picked it up.", + "question_chinese": "我的手机刚刚经历了什么路线？", + "answer_chinese": "我先一直拿在手上，然后放在地上，最后拿起来", + "evidence": [ + { + "description": "I held it in my hand continuously at first.", + "timestamps": { + "start": "00:02:29", + "end": "00:02:53" + } + }, + { + "description": "Then I placed it on the ground.", + "timestamps": { + "start": "00:05:05", + "end": "00:05:18" + } + }, + { + "description": "Then I picked it up again.", + "timestamps": { + "start": "00:07:45", + "end": "00:08:11" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Hand, then pocket.", + "B": "I held it in my hand continuously, then placed it on the ground, and finally picked it up.", + "C": "Pocket, then ground.", + "D": "You held it in your hand, put it in your pocket, and then took it back out.", + "E": "You set it on the ground first, then held it in your hand without picking it up again." + }, + "answer_label": "B", + "answer_text": "I held it in my hand continuously, then placed it on the ground, and finally picked it up.", + "q_id": "egolife_A1_D3_020_1", + "correct_answer": "I held it in my hand continuously, then placed it on the ground, and finally picked it up.", + "original_options": { + "A": "Hand, then pocket.", + "B": "I held it in my hand continuously, then placed it on the ground, and finally picked it up.", + "C": "Pocket, then ground.", + "D": "You held it in your hand, put it in your pocket, and then took it back out.", + "E": "You set it on the ground first, then held it in your hand without picking it up again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_5_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, is the phone in my hand?", + "answer": "Right now, the phone is not in your hand, and the phone is to your right.", + "question_chinese": "此时，手机在我手上吗？", + "answer_chinese": "此时，手机不在你手上，手机在你右边。", + "evidence": [ + { + "description": "The phone is not your hand.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:14" + } + } + ], + "question_timestamps": "00:02:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In your hand.", + "B": "Not in hand.", + "C": "Right now, the phone is not in your hand, and the phone is to your right.", + "D": "Right now, the phone is in your hand, and the phone is to your right.", + "E": "Right now, the phone is not in your hand, and the phone is to your left." + }, + "answer_label": "C", + "answer_text": "Right now, the phone is not in your hand, and the phone is to your right.", + "q_id": "egolife_A1_D3_020_5_T2", + "correct_answer": "Right now, the phone is not in your hand, and the phone is to your right.", + "original_options": { + "A": "In your hand.", + "B": "Not in hand.", + "C": "Right now, the phone is not in your hand, and the phone is to your right.", + "D": "Right now, the phone is in your hand, and the phone is to your right.", + "E": "Right now, the phone is not in your hand, and the phone is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_020_5_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, is the phone in my hand?", + "answer": "Right now, the phone is in your hand, and it is in front of you.", + "question_chinese": "此时，手机在我手上吗？", + "answer_chinese": "此时，手机在你手上，它在你前面。", + "evidence": [ + { + "description": "The phone is on your hand.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:02:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Not in hand.", + "B": "Right now, the phone is in your hand, and it is in front of you.", + "C": "In your hand, left.", + "D": "Right now, the phone is not in your hand, and it is in front of you.", + "E": "Right now, the phone is in your hand, but it is behind you instead of in front of you." + }, + "answer_label": "B", + "answer_text": "Right now, the phone is in your hand, and it is in front of you.", + "q_id": "egolife_A1_D3_020_5_T1", + "correct_answer": "Right now, the phone is in your hand, and it is in front of you.", + "original_options": { + "A": "Not in hand.", + "B": "Right now, the phone is in your hand, and it is in front of you.", + "C": "In your hand, left.", + "D": "Right now, the phone is not in your hand, and it is in front of you.", + "E": "Right now, the phone is in your hand, but it is behind you instead of in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 19 + ] + } +] diff --git a/QAs/egolife_A1_D3_021-mcq.json b/QAs/egolife_A1_D3_021-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5e398cb02494f096e84d228eeca01604250fce52 --- /dev/null +++ b/QAs/egolife_A1_D3_021-mcq.json @@ -0,0 +1,744 @@ +[ + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021_7_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was in my hand just now?", + "answer": "You took a phone and a plastic bag. The phone was in your pants pocket, and you handed the plastic bag to a friend.", + "question_chinese": "刚刚我的手里拿了什么？", + "answer_chinese": "你拿了手机以及塑料袋。手机在你的裤子口袋里，塑料袋你递给了朋友。", + "evidence": [ + { + "description": "At this moment, I am holding my phone.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:11" + } + }, + { + "description": "I picked up the plastic bag.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took a phone and a plastic bag. The phone was in your pants pocket, and you handed the plastic bag to a friend.", + "B": "Only your phone.", + "C": "Only a plastic bag.", + "D": "You took a phone from your pocket and kept it in your hand, without holding any bag.", + "E": "You pulled out a plastic bag and held it briefly, but you never took your phone out." + }, + "answer_label": "A", + "answer_text": "You took a phone and a plastic bag. The phone was in your pants pocket, and you handed the plastic bag to a friend.", + "q_id": "egolife_A1_D3_021_7_T2", + "correct_answer": "You took a phone and a plastic bag. The phone was in your pants pocket, and you handed the plastic bag to a friend.", + "original_options": { + "A": "You took a phone and a plastic bag. The phone was in your pants pocket, and you handed the plastic bag to a friend.", + "B": "Only your phone.", + "C": "Only a plastic bag.", + "D": "You took a phone from your pocket and kept it in your hand, without holding any bag.", + "E": "You pulled out a plastic bag and held it briefly, but you never took your phone out." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021_5", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding in my hand?", + "answer": "I took my phone and the plastic bag.", + "question_chinese": "刚刚我的手里拿了什么？", + "answer_chinese": "我拿了手机和塑料袋", + "evidence": [ + { + "description": "At this moment, I am holding my phone.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:11" + } + }, + { + "description": "I picked up the plastic bag.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:04:01", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I took my phone and the plastic bag.", + "B": "I was holding only my phone.", + "C": "I was holding only a plastic bag.", + "D": "I had my phone in my hand, while the plastic bag stayed elsewhere and wasn’t being held.", + "E": "I was holding the plastic bag in my hand, but my phone remained put away the whole time." + }, + "answer_label": "A", + "answer_text": "I took my phone and the plastic bag.", + "q_id": "egolife_A1_D3_021_5", + "correct_answer": "I took my phone and the plastic bag.", + "original_options": { + "A": "I took my phone and the plastic bag.", + "B": "I was holding only my phone.", + "C": "I was holding only a plastic bag.", + "D": "I had my phone in my hand, while the plastic bag stayed elsewhere and wasn’t being held.", + "E": "I was holding the plastic bag in my hand, but my phone remained put away the whole time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021_7_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was in my hand just now?", + "answer": "You picked up a backpack and a helmet. The backpack and helmet were in your hand.", + "question_chinese": "刚刚我的手里拿了什么？", + "answer_chinese": "你拿了书包和头盔。书包和头盔在你的手里。", + "evidence": [ + { + "description": "I picked up the schoolbag and helmet.", + "timestamps": { + "start": "00:09:44", + "end": "00:10:01" + } + } + ], + "question_timestamps": "00:10:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Only the backpack.", + "B": "Only the helmet.", + "C": "You held a backpack together with a plastic bag in your hand, not a helmet.", + "D": "You carried a helmet in your hand, while the backpack stayed on your back the entire time.", + "E": "You picked up a backpack and a helmet. The backpack and helmet were in your hand." + }, + "answer_label": "E", + "answer_text": "You picked up a backpack and a helmet. The backpack and helmet were in your hand.", + "q_id": "egolife_A1_D3_021_7_T1", + "correct_answer": "You picked up a backpack and a helmet. The backpack and helmet were in your hand.", + "original_options": { + "A": "Only the backpack.", + "B": "Only the helmet.", + "C": "You held a backpack together with a plastic bag in your hand, not a helmet.", + "D": "You carried a helmet in your hand, while the backpack stayed on your back the entire time.", + "E": "You picked up a backpack and a helmet. The backpack and helmet were in your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding in my hand?", + "answer": "Mobile phone, helmet, and my schoolbag.", + "question_chinese": "刚刚我的手里拿了什么？", + "answer_chinese": "手机，头盔和我的书包", + "evidence": [ + { + "description": "At this moment, I am holding my phone.", + "timestamps": { + "start": "00:06:19", + "end": "00:06:28" + } + }, + { + "description": "I am holding the helmet and my bag", + "timestamps": { + "start": "00:09:13", + "end": "00:09:35" + } + } + ], + "question_timestamps": "00:10:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Mobile phone and helmet only.", + "B": "Helmet and schoolbag only.", + "C": "You were holding a mobile phone and your schoolbag, but you weren’t holding the helmet.", + "D": "You had your helmet and schoolbag in hand, while the mobile phone stayed in your pocket.", + "E": "Mobile phone, helmet, and my schoolbag." + }, + "answer_label": "E", + "answer_text": "Mobile phone, helmet, and my schoolbag.", + "q_id": "egolife_A1_D3_021_6", + "correct_answer": "Mobile phone, helmet, and my schoolbag.", + "original_options": { + "A": "Mobile phone and helmet only.", + "B": "Helmet and schoolbag only.", + "C": "You were holding a mobile phone and your schoolbag, but you weren’t holding the helmet.", + "D": "You had your helmet and schoolbag in hand, while the mobile phone stayed in your pocket.", + "E": "Mobile phone, helmet, and my schoolbag." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is around the place where I just stopped?", + "answer": "There were my friends and the things I bought, with a wall on the left.", + "question_chinese": "我刚刚停留的地方周围有什么？", + "answer_chinese": "有我的朋友和买的东西，左边有一面墙", + "evidence": [ + { + "description": "My friends and the things we bought are here.", + "timestamps": { + "start": "00:00:29", + "end": "00:00:32" + } + }, + { + "description": "There is a wall on the left.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A wall was on your right.", + "B": "There were my friends and the things I bought, with a wall on the left.", + "C": "A wall was behind you.", + "D": "Your friends and purchases were nearby, but the wall was straight ahead instead of on the left.", + "E": "You stopped near your friends and the items you bought, with the wall to your right side." + }, + "answer_label": "B", + "answer_text": "There were my friends and the things I bought, with a wall on the left.", + "q_id": "egolife_A1_D3_021_1", + "correct_answer": "There were my friends and the things I bought, with a wall on the left.", + "original_options": { + "A": "A wall was on your right.", + "B": "There were my friends and the things I bought, with a wall on the left.", + "C": "A wall was behind you.", + "D": "Your friends and purchases were nearby, but the wall was straight ahead instead of on the left.", + "E": "You stopped near your friends and the items you bought, with the wall to your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is around the place where I just stopped?", + "answer": "There were railings, cars, friends, and pedestrians.", + "question_chinese": "我刚刚停留的地方周围有什么？", + "answer_chinese": "有栏杆，车，朋友和行人", + "evidence": [ + { + "description": "There were railings and friends", + "timestamps": { + "start": "00:03:11", + "end": "00:03:13" + } + }, + { + "description": "There were also cars and pedestrians.", + "timestamps": { + "start": "00:03:16", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:06:28", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Only railings and cars.", + "B": "Only friends and pedestrians.", + "C": "You stopped near railings, cars, and friends, but there were no pedestrians around you at that moment.", + "D": "There were railings, cars, friends, and pedestrians.", + "E": "You were by cars and pedestrians, while railings weren’t present and your friends weren’t nearby." + }, + "answer_label": "D", + "answer_text": "There were railings, cars, friends, and pedestrians.", + "q_id": "egolife_A1_D3_021_2", + "correct_answer": "There were railings, cars, friends, and pedestrians.", + "original_options": { + "A": "Only railings and cars.", + "B": "Only friends and pedestrians.", + "C": "You stopped near railings, cars, and friends, but there were no pedestrians around you at that moment.", + "D": "There were railings, cars, friends, and pedestrians.", + "E": "You were by cars and pedestrians, while railings weren’t present and your friends weren’t nearby." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021_3", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "Went straight ahead and exited through the gate, then turned right and continue straight, finally turned left and then turned right again.", + "question_chinese": "我刚刚走了什么路线？", + "answer_chinese": "直行出门，然后右转直行，左转后右转", + "evidence": [ + { + "description": "First, I went straight ahead and then exited through a gate.", + "timestamps": { + "start": "00:00:26", + "end": "00:01:26" + } + }, + { + "description": "Then we turned right.", + "timestamps": { + "start": "00:01:51", + "end": "00:01:59" + } + }, + { + "description": "I went straight ahead and then turned left.", + "timestamps": { + "start": "00:02:34", + "end": "00:02:39" + } + }, + { + "description": "Then I turned right.", + "timestamps": { + "start": "00:03:51", + "end": "00:04:02" + } + } + ], + "question_timestamps": "00:05:53", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Exit the gate, then turn left.", + "B": "Exit the gate, then turn right.", + "C": "Went straight ahead and exited through the gate, then turned right and continue straight, finally turned left and then turned right again.", + "D": "You went straight through the gate, turned right, continued straight, then turned right and finally turned left.", + "E": "You went straight through the gate, turned left, continued straight, then turned right and then turned left again." + }, + "answer_label": "C", + "answer_text": "Went straight ahead and exited through the gate, then turned right and continue straight, finally turned left and then turned right again.", + "q_id": "egolife_A1_D3_021_3", + "correct_answer": "Went straight ahead and exited through the gate, then turned right and continue straight, finally turned left and then turned right again.", + "original_options": { + "A": "Exit the gate, then turn left.", + "B": "Exit the gate, then turn right.", + "C": "Went straight ahead and exited through the gate, then turned right and continue straight, finally turned left and then turned right again.", + "D": "You went straight through the gate, turned right, continued straight, then turned right and finally turned left.", + "E": "You went straight through the gate, turned left, continued straight, then turned right and then turned left again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021_4", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I turned right first, then went straight ahead and turned right again.", + "question_chinese": "我刚刚走了什么路线？", + "answer_chinese": "我先右转，直走后再右转", + "evidence": [ + { + "description": "I turned right first.", + "timestamps": { + "start": "00:10:03", + "end": "00:10:15" + } + }, + { + "description": "Then I kept going straight ahead.", + "timestamps": { + "start": "00:10:39", + "end": "00:10:43" + } + }, + { + "description": "Finally, I turned right.", + "timestamps": { + "start": "00:10:47", + "end": "00:11:00" + } + } + ], + "question_timestamps": "00:11:07", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "I turned left first.", + "B": "I went straight, then turned right.", + "C": "I turned right first, went straight ahead, and then turned left instead of turning right again.", + "D": "I turned left first, then went straight ahead and turned right once, without making a second right turn.", + "E": "I turned right first, then went straight ahead and turned right again." + }, + "answer_label": "E", + "answer_text": "I turned right first, then went straight ahead and turned right again.", + "q_id": "egolife_A1_D3_021_4", + "correct_answer": "I turned right first, then went straight ahead and turned right again.", + "original_options": { + "A": "I turned left first.", + "B": "I went straight, then turned right.", + "C": "I turned right first, went straight ahead, and then turned left instead of turning right again.", + "D": "I turned left first, then went straight ahead and turned right once, without making a second right turn.", + "E": "I turned right first, then went straight ahead and turned right again." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021_7", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "May I touch the car door?", + "answer": "Yes,I can.", + "question_chinese": "我可以碰到车门吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The car is very close to me, I can touch it.", + "timestamps": { + "start": "00:07:39", + "end": "00:07:46" + } + } + ], + "question_timestamps": "00:09:39", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, I can't.", + "B": "Yes,I can." + }, + "answer_label": "B", + "answer_text": "Yes,I can.", + "q_id": "egolife_A1_D3_021_7", + "correct_answer": "Yes,I can.", + "original_options": { + "A": "No, I can't.", + "B": "Yes,I can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021_8", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "May I touch the car door?", + "answer": "No,I can not.", + "question_chinese": "我可以碰到车门吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "I have moved away, and I can't reach the door anymore.", + "timestamps": { + "start": "00:10:04", + "end": "00:10:24" + } + } + ], + "question_timestamps": "00:11:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,I can not.", + "B": "Yes, I can." + }, + "answer_label": "A", + "answer_text": "No,I can not.", + "q_id": "egolife_A1_D3_021_8", + "correct_answer": "No,I can not.", + "original_options": { + "A": "No,I can not.", + "B": "Yes, I can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021-mcq_fixed_8", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the friend in colorful clothes just now?", + "answer": "The friend in colorful clothes was in front of you and slightly to your left.", + "question_chinese": "花色衣服的朋友刚在我的什么方向？", + "answer_chinese": "在我的左前边", + "evidence": [ + { + "description": "The friend in colorful clothes was in front of me and slightly to my left.", + "timestamps": { + "start": "00:04:15", + "end": "00:04:26" + } + } + ], + "question_timestamps": "00:06:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you, slightly right.", + "B": "Behind you, slightly left.", + "C": "The friend in colorful clothes was directly in front of you, a bit to your right.", + "D": "The friend in colorful clothes was behind you and slightly to your right as you looked ahead.", + "E": "The friend in colorful clothes was in front of you and slightly to your left." + }, + "answer_label": "E", + "answer_text": "The friend in colorful clothes was in front of you and slightly to your left.", + "q_id": "egolife_A1_D3_021-mcq_fixed_8", + "correct_answer": "The friend in colorful clothes was in front of you and slightly to your left.", + "original_options": { + "A": "In front of you, slightly right.", + "B": "Behind you, slightly left.", + "C": "The friend in colorful clothes was directly in front of you, a bit to your right.", + "D": "The friend in colorful clothes was behind you and slightly to your right as you looked ahead.", + "E": "The friend in colorful clothes was in front of you and slightly to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_021", + "video_length": "00:11:05", + "qid": "egolife_A1_D3_021_10", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is my friend in the colorful clothes?", + "answer": "The friend in colorful clothes was on my left.", + "question_chinese": "花色衣服的朋友刚在我的什么方向？", + "answer_chinese": "在我的左边", + "evidence": [ + { + "description": "The friend in colorful clothes is on my left.", + "timestamps": { + "start": "00:06:02", + "end": "00:06:20" + } + } + ], + "question_timestamps": "00:11:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The friend was on my right.", + "B": "The friend in colorful clothes was on my left.", + "C": "The friend was behind me.", + "D": "The friend in colorful clothes was in front of me, not off to my left side.", + "E": "The friend in colorful clothes was to my right, a little behind my shoulder as I faced forward." + }, + "answer_label": "B", + "answer_text": "The friend in colorful clothes was on my left.", + "q_id": "egolife_A1_D3_021_10", + "correct_answer": "The friend in colorful clothes was on my left.", + "original_options": { + "A": "The friend was on my right.", + "B": "The friend in colorful clothes was on my left.", + "C": "The friend was behind me.", + "D": "The friend in colorful clothes was in front of me, not off to my left side.", + "E": "The friend in colorful clothes was to my right, a little behind my shoulder as I faced forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D3_022-mcq.json b/QAs/egolife_A1_D3_022-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d5ed8458d28aa7ff90bae5705000f987e3487617 --- /dev/null +++ b/QAs/egolife_A1_D3_022-mcq.json @@ -0,0 +1,1146 @@ +[ + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What route did you just take in the stairwell?", + "answer": "I turned left first, then turned left again.", + "question_chinese": "刚刚在楼梯间经历了什么路线？", + "answer_chinese": "我先左转，再左转", + "evidence": [ + { + "description": "I turn left first.", + "timestamps": { + "start": "00:10:12", + "end": "00:10:15" + } + }, + { + "description": "I turn left again", + "timestamps": { + "start": "00:10:19", + "end": "00:10:21" + } + } + ], + "question_timestamps": "00:10:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I turned right, then right.", + "B": "I turned left first, then turned left again.", + "C": "I turned right, then left.", + "D": "I turned left first, then made a right turn the second time in the stairwell.", + "E": "I went straight at first and then turned left when I reached the next landing." + }, + "answer_label": "B", + "answer_text": "I turned left first, then turned left again.", + "q_id": "egolife_A1_D3_022_7", + "correct_answer": "I turned left first, then turned left again.", + "original_options": { + "A": "I turned right, then right.", + "B": "I turned left first, then turned left again.", + "C": "I turned right, then left.", + "D": "I turned left first, then made a right turn the second time in the stairwell.", + "E": "I went straight at first and then turned left when I reached the next landing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_8", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What route did you just take in the stairwell?", + "answer": "I turned right first, then turned right again.", + "question_chinese": "刚刚在楼梯间经历了什么路线？", + "answer_chinese": "我先右转，再右转", + "evidence": [ + { + "description": "I turned right first", + "timestamps": { + "start": "00:10:32", + "end": "00:10:35" + } + }, + { + "description": "I turned right again", + "timestamps": { + "start": "00:10:37", + "end": "00:10:40" + } + } + ], + "question_timestamps": "00:10:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I turned left, then left.", + "B": "I went straight, then right.", + "C": "I turned right first, but the second turn was to the left instead of right.", + "D": "I turned left initially and then turned right on the next landing in the stairwell.", + "E": "I turned right first, then turned right again." + }, + "answer_label": "E", + "answer_text": "I turned right first, then turned right again.", + "q_id": "egolife_A1_D3_022_8", + "correct_answer": "I turned right first, then turned right again.", + "original_options": { + "A": "I turned left, then left.", + "B": "I went straight, then right.", + "C": "I turned right first, but the second turn was to the left instead of right.", + "D": "I turned left initially and then turned right on the next landing in the stairwell.", + "E": "I turned right first, then turned right again." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_3", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did the item in my hand just take?", + "answer": "I placed the thing in my hand and then my schoolbag on the ground.", + "question_chinese": "刚刚我手上的东西经历了什么路线？", + "answer_chinese": "我把手上的东西和书包先后放在了地上", + "evidence": [ + { + "description": "At this moment, I placed the thing in my hand on the ground.", + "timestamps": { + "start": "00:01:49", + "end": "00:01:56" + } + }, + { + "description": "At this moment, I placed my schoolbag on the ground.", + "timestamps": { + "start": "00:02:05", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:04:01", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I put only the bag down.", + "B": "I put only the item down.", + "C": "I placed my schoolbag on the ground first and then set the thing in my hand down.", + "D": "I placed the thing in my hand and then my schoolbag on the ground.", + "E": "I kept the thing in my hand and put my schoolbag down somewhere else instead of on the ground." + }, + "answer_label": "D", + "answer_text": "I placed the thing in my hand and then my schoolbag on the ground.", + "q_id": "egolife_A1_D3_022_3", + "correct_answer": "I placed the thing in my hand and then my schoolbag on the ground.", + "original_options": { + "A": "I put only the bag down.", + "B": "I put only the item down.", + "C": "I placed my schoolbag on the ground first and then set the thing in my hand down.", + "D": "I placed the thing in my hand and then my schoolbag on the ground.", + "E": "I kept the thing in my hand and put my schoolbag down somewhere else instead of on the ground." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_4", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did the item in my hand just take?", + "answer": "I plugged the data cable into the computer and placed my glasses and other items on the chair.", + "question_chinese": "刚刚我手上的东西经历了什么路线？", + "answer_chinese": "我将数据线插到了电脑上，眼镜等东西放在椅子上", + "evidence": [ + { + "description": "I took a bunch of data cables and then plugged them into the compute.r", + "timestamps": { + "start": "00:05:46", + "end": "00:05:55" + } + }, + { + "description": "I put my glasses and other things from my hand onto the box on the chai.r", + "timestamps": { + "start": "00:06:25", + "end": "00:07:56" + } + } + ], + "question_timestamps": "00:10:51", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I plugged it into a laptop.", + "B": "I placed the glasses on the desk.", + "C": "I plugged the data cable into a monitor, then set my glasses and other items onto a shelf.", + "D": "I connected the data cable to a tablet, then placed my glasses and other items on the floor.", + "E": "I plugged the data cable into the computer and placed my glasses and other items on the chair." + }, + "answer_label": "E", + "answer_text": "I plugged the data cable into the computer and placed my glasses and other items on the chair.", + "q_id": "egolife_A1_D3_022_4", + "correct_answer": "I plugged the data cable into the computer and placed my glasses and other items on the chair.", + "original_options": { + "A": "I plugged it into a laptop.", + "B": "I placed the glasses on the desk.", + "C": "I plugged the data cable into a monitor, then set my glasses and other items onto a shelf.", + "D": "I connected the data cable to a tablet, then placed my glasses and other items on the floor.", + "E": "I plugged the data cable into the computer and placed my glasses and other items on the chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_5", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just touch the table in the living room?", + "answer": "No,I can not.", + "question_chinese": "刚刚我能碰到客厅上的桌子吗？", + "answer_chinese": "不行", + "evidence": [ + { + "description": "Standing next to the table, it was some distance away from me.", + "timestamps": { + "start": "00:01:44", + "end": "00:01:51" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,I can not.", + "B": "Yes, I can." + }, + "answer_label": "A", + "answer_text": "No,I can not.", + "q_id": "egolife_A1_D3_022_5", + "correct_answer": "No,I can not.", + "original_options": { + "A": "No,I can not.", + "B": "Yes, I can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_6", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just touch the table in the living room?", + "answer": "Yes,I can.", + "question_chinese": "刚刚我能碰到客厅上的桌子吗？", + "answer_chinese": "以可", + "evidence": [ + { + "description": "At this moment, I was close enough to the table to touch it.", + "timestamps": { + "start": "00:03:57", + "end": "00:04:02" + } + } + ], + "question_timestamps": "00:05:17", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, I can't.", + "B": "Yes,I can." + }, + "answer_label": "B", + "answer_text": "Yes,I can.", + "q_id": "egolife_A1_D3_022_6", + "correct_answer": "Yes,I can.", + "original_options": { + "A": "No, I can't.", + "B": "Yes,I can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_7_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the chess table?", + "answer": "It’s to your left.", + "question_chinese": "国际象棋桌在我的什么方向？", + "answer_chinese": "在你的左边。", + "evidence": [ + { + "description": "On your left.", + "timestamps": { + "start": "00:01:54", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:01:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s to your right.", + "C": "It’s directly behind you, not on your left side where your arm would reach.", + "D": "It’s above your eye level, so it isn’t positioned to your left at all.", + "E": "It’s to your left." + }, + "answer_label": "E", + "answer_text": "It’s to your left.", + "q_id": "egolife_A1_D3_022_7_T2", + "correct_answer": "It’s to your left.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s to your right.", + "C": "It’s directly behind you, not on your left side where your arm would reach.", + "D": "It’s above your eye level, so it isn’t positioned to your left at all.", + "E": "It’s to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_7_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the chess table?", + "answer": "It’s behind you.", + "question_chinese": "国际象棋桌在我的什么方向？", + "answer_chinese": "在你的后面。", + "evidence": [ + { + "description": "Behind you.", + "timestamps": { + "start": "00:02:01", + "end": "00:02:04" + } + } + ], + "question_timestamps": "00:02:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s to your left.", + "C": "It’s behind you.", + "D": "It’s off to your right side, not located behind you where you’d need to turn around.", + "E": "It’s ahead of you in your line of sight, rather than behind your back." + }, + "answer_label": "C", + "answer_text": "It’s behind you.", + "q_id": "egolife_A1_D3_022_7_T1", + "correct_answer": "It’s behind you.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s to your left.", + "C": "It’s behind you.", + "D": "It’s off to your right side, not located behind you where you’d need to turn around.", + "E": "It’s ahead of you in your line of sight, rather than behind your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_6_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, in which direction from me is the living room table?", + "answer": "It’s to your left.", + "question_chinese": "现在客厅的桌子在我的什么方向？", + "answer_chinese": "在你的左边。", + "evidence": [ + { + "description": "On your left.", + "timestamps": { + "start": "00:03:33", + "end": "00:03:37" + } + } + ], + "question_timestamps": "00:03:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your left.", + "B": "It’s to your right.", + "C": "It’s behind you.", + "D": "It’s positioned in front of you right now, not on your left side at this moment.", + "E": "It’s slightly behind and to your right, so it isn’t to your left currently." + }, + "answer_label": "A", + "answer_text": "It’s to your left.", + "q_id": "egolife_A1_D3_022_6_T2", + "correct_answer": "It’s to your left.", + "original_options": { + "A": "It’s to your left.", + "B": "It’s to your right.", + "C": "It’s behind you.", + "D": "It’s positioned in front of you right now, not on your left side at this moment.", + "E": "It’s slightly behind and to your right, so it isn’t to your left currently." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_6_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, in which direction from me is the living room table?", + "answer": "It’s to your right.", + "question_chinese": "现在客厅的桌子在我的什么方向？", + "answer_chinese": "在你的右边。", + "evidence": [ + { + "description": "The living room table is on the right side of the computer, and now I'm in the computer room, with the table still on my right.", + "timestamps": { + "start": "00:03:49", + "end": "00:03:51" + } + } + ], + "question_timestamps": "00:03:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your left.", + "B": "It’s in front of you.", + "C": "It’s to your right.", + "D": "It’s behind you at the moment, not on your right side where you’d reach over.", + "E": "It’s above you rather than to your right, so the right-side direction is incorrect." + }, + "answer_label": "C", + "answer_text": "It’s to your right.", + "q_id": "egolife_A1_D3_022_6_T1", + "correct_answer": "It’s to your right.", + "original_options": { + "A": "It’s to your left.", + "B": "It’s in front of you.", + "C": "It’s to your right.", + "D": "It’s behind you at the moment, not on your right side where you’d reach over.", + "E": "It’s above you rather than to your right, so the right-side direction is incorrect." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_10", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the table in the living room from where I am?", + "answer": "It is behind me.", + "question_chinese": "现在客厅的桌子在我的什么方向？", + "answer_chinese": "在我后面", + "evidence": [ + { + "description": "The table is on the left side of the kitchen. At this moment, I'm facing the kitchen, and the table is behind me.", + "timestamps": { + "start": "00:03:54", + "end": "00:04:02" + } + } + ], + "question_timestamps": "00:05:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is in front of me.", + "B": "It is to my left.", + "C": "It is off to my right side, not behind me where I would have to turn around.", + "D": "It is behind me.", + "E": "It is positioned across in front of me, rather than behind me from my current spot." + }, + "answer_label": "D", + "answer_text": "It is behind me.", + "q_id": "egolife_A1_D3_022_10", + "correct_answer": "It is behind me.", + "original_options": { + "A": "It is in front of me.", + "B": "It is to my left.", + "C": "It is off to my right side, not behind me where I would have to turn around.", + "D": "It is behind me.", + "E": "It is positioned across in front of me, rather than behind me from my current spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_9", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the table in the living room from where I am?", + "answer": "Iti s on my right.", + "question_chinese": "现在客厅的桌子在我的什么方向？", + "answer_chinese": "在我的右边", + "evidence": [ + { + "description": "The living room table is on the right side of the computer, and now I'm in the computer room, with the table still on my right.", + "timestamps": { + "start": "00:03:33", + "end": "00:03:37" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is in front of me.", + "B": "Iti s on my right.", + "C": "It is to my left now.", + "D": "It is behind me instead of on my right, so I’d need to turn around to face it.", + "E": "It is slightly in front and left of me, not on my right side from where I’m standing." + }, + "answer_label": "B", + "answer_text": "Iti s on my right.", + "q_id": "egolife_A1_D3_022_9", + "correct_answer": "Iti s on my right.", + "original_options": { + "A": "It is in front of me.", + "B": "Iti s on my right.", + "C": "It is to my left now.", + "D": "It is behind me instead of on my right, so I’d need to turn around to face it.", + "E": "It is slightly in front and left of me, not on my right side from where I’m standing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_2_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the computer room relative to me?", + "answer": "The computer room is in front of you.", + "question_chinese": "电脑房相对于我在什么位置？", + "answer_chinese": "电脑房在你的正前方。", + "evidence": [ + { + "description": "Right in front of you.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:49" + } + } + ], + "question_timestamps": "00:01:52", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The computer room is in front of you.", + "B": "The computer room is behind you.", + "C": "The computer room is to your left.", + "D": "The computer room is to your right, not directly ahead of you.", + "E": "The computer room is above you, rather than located in front of you." + }, + "answer_label": "A", + "answer_text": "The computer room is in front of you.", + "q_id": "egolife_A1_D3_022_2_T2", + "correct_answer": "The computer room is in front of you.", + "original_options": { + "A": "The computer room is in front of you.", + "B": "The computer room is behind you.", + "C": "The computer room is to your left.", + "D": "The computer room is to your right, not directly ahead of you.", + "E": "The computer room is above you, rather than located in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_2_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the computer room relative to me?", + "answer": "The computer room is behind you.", + "question_chinese": "电脑房相对于我在什么位置？", + "answer_chinese": "电脑房在你的后面。", + "evidence": [ + { + "description": "Behind you.", + "timestamps": { + "start": "00:01:49", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:01:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The computer room is behind you.", + "B": "The computer room is in front of you.", + "C": "The computer room is to your right.", + "D": "The computer room is to your left, not positioned behind you at all.", + "E": "The computer room is above you, instead of being behind your current position." + }, + "answer_label": "A", + "answer_text": "The computer room is behind you.", + "q_id": "egolife_A1_D3_022_2_T1", + "correct_answer": "The computer room is behind you.", + "original_options": { + "A": "The computer room is behind you.", + "B": "The computer room is in front of you.", + "C": "The computer room is to your right.", + "D": "The computer room is to your left, not positioned behind you at all.", + "E": "The computer room is above you, instead of being behind your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_1_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I entered the computer room in total?", + "answer": "So far, you have entered three times in total, and the computer is in front of you.", + "question_chinese": "目前为止，我一共进了几次电脑房？", + "answer_chinese": "目前为止，你一共进了三次，电脑在你的正前方。", + "evidence": [ + { + "description": "The first time I entered the room", + "timestamps": { + "start": "00:01:58", + "end": "00:02:09" + } + }, + { + "description": "The second time I entered the room", + "timestamps": { + "start": "00:05:49", + "end": "00:05:57" + } + }, + { + "description": "The third time I entered the room", + "timestamps": { + "start": "00:07:20", + "end": "00:07:30" + } + } + ], + "question_timestamps": "00:07:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two times total.", + "B": "Four times total.", + "C": "So far, you have entered two times in total, and the computer is in front of you.", + "D": "So far, you have entered four times in total, and the computer is in front of you.", + "E": "So far, you have entered three times in total, and the computer is in front of you." + }, + "answer_label": "E", + "answer_text": "So far, you have entered three times in total, and the computer is in front of you.", + "q_id": "egolife_A1_D3_022_1_T1", + "correct_answer": "So far, you have entered three times in total, and the computer is in front of you.", + "original_options": { + "A": "Two times total.", + "B": "Four times total.", + "C": "So far, you have entered two times in total, and the computer is in front of you.", + "D": "So far, you have entered four times in total, and the computer is in front of you.", + "E": "So far, you have entered three times in total, and the computer is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_1_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I entered the computer room in total?", + "answer": "So far, you’ve gone in five times in total, and the computer is to your left.", + "question_chinese": "目前为止，我一共进了几次电脑房？", + "answer_chinese": "目前为止，你一共进了五次，电脑在你的左边。", + "evidence": [ + { + "description": "The first time I entered the room", + "timestamps": { + "start": "00:01:58", + "end": "00:02:09" + } + }, + { + "description": "The second time I entered the room", + "timestamps": { + "start": "00:05:49", + "end": "00:05:57" + } + }, + { + "description": "The third time I entered the room", + "timestamps": { + "start": "00:07:23", + "end": "00:07:30" + } + }, + { + "description": "The forth time I entered the room", + "timestamps": { + "start": "00:08:30", + "end": "00:08:40" + } + }, + { + "description": "The fifth time I entered the room", + "timestamps": { + "start": "00:09:48", + "end": "00:09:57" + } + } + ], + "question_timestamps": "00:10:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, you’ve gone in five times in total, and the computer is to your left.", + "B": "Four times total.", + "C": "Six times total.", + "D": "So far, you’ve gone in four times in total, and the computer is to your left.", + "E": "So far, you’ve gone in six times in total, and the computer is to your left." + }, + "answer_label": "A", + "answer_text": "So far, you’ve gone in five times in total, and the computer is to your left.", + "q_id": "egolife_A1_D3_022_1_T2", + "correct_answer": "So far, you’ve gone in five times in total, and the computer is to your left.", + "original_options": { + "A": "So far, you’ve gone in five times in total, and the computer is to your left.", + "B": "Four times total.", + "C": "Six times total.", + "D": "So far, you’ve gone in four times in total, and the computer is to your left.", + "E": "So far, you’ve gone in six times in total, and the computer is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_1_T3", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I entered the computer room in total?", + "answer": "So far, you’ve gone in seven times in total, and the computer is in front of you.", + "question_chinese": "目前为止，我一共进了几次电脑房？", + "answer_chinese": "目前为止，你一共进了七次，电脑在你的正前方。", + "evidence": [ + { + "description": "The first time I entered the room", + "timestamps": { + "start": "00:01:58", + "end": "00:02:09" + } + }, + { + "description": "The second time I entered the room", + "timestamps": { + "start": "00:05:49", + "end": "00:05:57" + } + }, + { + "description": "The third time I entered the room", + "timestamps": { + "start": "00:07:23", + "end": "00:07:30" + } + }, + { + "description": "The forth time I entered the room", + "timestamps": { + "start": "00:08:30", + "end": "00:08:40" + } + }, + { + "description": "The fifth time I entered the room", + "timestamps": { + "start": "00:09:48", + "end": "00:09:57" + } + }, + { + "description": "The sixth time I entered the room", + "timestamps": { + "start": "00:10:52", + "end": "00:11:01" + } + }, + { + "description": "The seventh time I entered the room", + "timestamps": { + "start": "00:11:49", + "end": "00:11:54" + } + } + ], + "question_timestamps": "00:11:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Six times total.", + "B": "So far, you’ve gone in seven times in total, and the computer is in front of you.", + "C": "Eight times total.", + "D": "So far, you’ve gone in six times in total, and the computer is in front of you.", + "E": "So far, you’ve gone in eight times in total, and the computer is in front of you." + }, + "answer_label": "B", + "answer_text": "So far, you’ve gone in seven times in total, and the computer is in front of you.", + "q_id": "egolife_A1_D3_022_1_T3", + "correct_answer": "So far, you’ve gone in seven times in total, and the computer is in front of you.", + "original_options": { + "A": "Six times total.", + "B": "So far, you’ve gone in seven times in total, and the computer is in front of you.", + "C": "Eight times total.", + "D": "So far, you’ve gone in six times in total, and the computer is in front of you.", + "E": "So far, you’ve gone in eight times in total, and the computer is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_022_2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I entered the computer room so far?", + "answer": "Seventh.", + "question_chinese": "目前为止，我一共进了几次电脑房？", + "answer_chinese": "七次", + "evidence": [ + { + "description": "The first time I entered the room", + "timestamps": { + "start": "00:01:58", + "end": "00:02:09" + } + }, + { + "description": "The second time I entered the room", + "timestamps": { + "start": "00:05:49", + "end": "00:05:57" + } + }, + { + "description": "The third time I entered the room", + "timestamps": { + "start": "00:07:23", + "end": "00:07:30" + } + }, + { + "description": "The forth time I entered the room", + "timestamps": { + "start": "00:08:30", + "end": "00:08:40" + } + }, + { + "description": "The fifth time I entered the room", + "timestamps": { + "start": "00:09:48", + "end": "00:09:57" + } + }, + { + "description": "The sixth time I entered the room", + "timestamps": { + "start": "00:10:52", + "end": "00:11:01" + } + }, + { + "description": "The seventh time I entered the room", + "timestamps": { + "start": "00:11:49", + "end": "00:11:54" + } + } + ], + "question_timestamps": "00:12:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Fifth.", + "B": "Seventh.", + "C": "Sixth.", + "D": "You have entered the computer room for the eighth time so far.", + "E": "You have entered the computer room for the ninth time up to now." + }, + "answer_label": "B", + "answer_text": "Seventh.", + "q_id": "egolife_A1_D3_022_2", + "correct_answer": "Seventh.", + "original_options": { + "A": "Fifth.", + "B": "Seventh.", + "C": "Sixth.", + "D": "You have entered the computer room for the eighth time so far.", + "E": "You have entered the computer room for the ninth time up to now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D3_023-mcq.json b/QAs/egolife_A1_D3_023-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..db180b58d0c4cb1842807827f3ebfb2343c7c699 --- /dev/null +++ b/QAs/egolife_A1_D3_023-mcq.json @@ -0,0 +1,844 @@ +[ + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_6", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was around you just now when you were sitting?", + "answer": "Computers, beverage, bed.", + "question_chinese": "刚刚坐着的时候，周围有什么？", + "answer_chinese": "电脑饮料和床", + "evidence": [ + { + "description": "There are computers and a bottle on the tabl.e", + "timestamps": { + "start": "00:00:17", + "end": "00:00:30" + } + }, + { + "description": "There is a bed on my right", + "timestamps": { + "start": "00:02:59", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:05:37", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Computers, bed, pillows.", + "B": "Computers, beverage, chair.", + "C": "Computers, a beverage, and a chair were around you when you were sitting.", + "D": "Computers, beverage, bed.", + "E": "There were computers nearby, with a drink and some pillows around you as you sat." + }, + "answer_label": "D", + "answer_text": "Computers, beverage, bed.", + "q_id": "egolife_A1_D3_023_6", + "correct_answer": "Computers, beverage, bed.", + "original_options": { + "A": "Computers, bed, pillows.", + "B": "Computers, beverage, chair.", + "C": "Computers, a beverage, and a chair were around you when you were sitting.", + "D": "Computers, beverage, bed.", + "E": "There were computers nearby, with a drink and some pillows around you as you sat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_5", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was around you just now while you were sitting?", + "answer": "There were pillows, mineral water bottles, cardboard boxes, and chairs next to me.", + "question_chinese": "刚刚坐着的时候，周围有什么？", + "answer_chinese": "有抱枕，矿泉水瓶，纸箱子和椅子在我旁边", + "evidence": [ + { + "description": "I'm sitting on the sofa, and there are pillows on it.", + "timestamps": { + "start": "00:05:56", + "end": "00:05:59" + } + }, + { + "description": "There are mineral water bottles and cardboard boxes on the sofa.", + "timestamps": { + "start": "00:06:03", + "end": "00:06:12" + } + }, + { + "description": "There is a chair on my right", + "timestamps": { + "start": "00:06:17", + "end": "00:06:19" + } + } + ], + "question_timestamps": "00:07:39", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "There were pillows, mineral water bottles, cardboard boxes, and chairs next to me.", + "B": "Pillows, water bottles, chairs.", + "C": "Boxes and chairs next to you.", + "D": "There were pillows, mineral water bottles, and chairs next to me, but no cardboard boxes.", + "E": "There were pillows and cardboard boxes next to me, but you were surrounded by tables instead of chairs." + }, + "answer_label": "A", + "answer_text": "There were pillows, mineral water bottles, cardboard boxes, and chairs next to me.", + "q_id": "egolife_A1_D3_023_5", + "correct_answer": "There were pillows, mineral water bottles, cardboard boxes, and chairs next to me.", + "original_options": { + "A": "There were pillows, mineral water bottles, cardboard boxes, and chairs next to me.", + "B": "Pillows, water bottles, chairs.", + "C": "Boxes and chairs next to you.", + "D": "There were pillows, mineral water bottles, and chairs next to me, but no cardboard boxes.", + "E": "There were pillows and cardboard boxes next to me, but you were surrounded by tables instead of chairs." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_4", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the computer just now?", + "answer": "The computers were placed on both the table and the cabinet.", + "question_chinese": "刚刚的电脑是放在哪里的？", + "answer_chinese": "放在桌子和柜子上", + "evidence": [ + { + "description": "These computers were placed on the table", + "timestamps": { + "start": "00:00:54", + "end": "00:00:56" + } + }, + { + "description": "Some of the computers are placed on the cabinet", + "timestamps": { + "start": "00:01:00", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:03:39", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the table only.", + "B": "On the cabinet only.", + "C": "The computers were placed on the cabinet and then moved down onto the floor.", + "D": "The computers were placed on both the table and the cabinet.", + "E": "The computers were set on the table and later shifted to the floor beside it." + }, + "answer_label": "D", + "answer_text": "The computers were placed on both the table and the cabinet.", + "q_id": "egolife_A1_D3_023_4", + "correct_answer": "The computers were placed on both the table and the cabinet.", + "original_options": { + "A": "On the table only.", + "B": "On the cabinet only.", + "C": "The computers were placed on the cabinet and then moved down onto the floor.", + "D": "The computers were placed on both the table and the cabinet.", + "E": "The computers were set on the table and later shifted to the floor beside it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_3", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the computer just now?", + "answer": "These computers were placed on the table, cabinet, and floor.", + "question_chinese": "刚刚的电脑是放在哪里的？", + "answer_chinese": "电脑被放在桌子柜子和地上", + "evidence": [ + { + "description": "The computers are placed on the table", + "timestamps": { + "start": "00:00:46", + "end": "00:00:50" + } + }, + { + "description": "The computers are placed on the cabinet.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:09" + } + }, + { + "description": "These computers were placed on the floor.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:13" + } + } + ], + "question_timestamps": "00:07:51", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "On the table and cabinet only.", + "B": "On the cabinet and floor only.", + "C": "These computers were placed on the table and floor, with none left sitting on the cabinet.", + "D": "These computers were arranged on the cabinet and table, and later moved off the floor.", + "E": "These computers were placed on the table, cabinet, and floor." + }, + "answer_label": "E", + "answer_text": "These computers were placed on the table, cabinet, and floor.", + "q_id": "egolife_A1_D3_023_3", + "correct_answer": "These computers were placed on the table, cabinet, and floor.", + "original_options": { + "A": "On the table and cabinet only.", + "B": "On the cabinet and floor only.", + "C": "These computers were placed on the table and floor, with none left sitting on the cabinet.", + "D": "These computers were arranged on the cabinet and table, and later moved off the floor.", + "E": "These computers were placed on the table, cabinet, and floor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_5_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was on the chair by the doorway just now?", + "answer": "At the beginning there was a black strip-shaped item, and later there was nothing behind you.", + "question_chinese": "刚刚门口的椅子上什么？", + "answer_chinese": "开始有一条黑色的条装物，后面没东西了", + "evidence": [ + { + "description": "At this moment, there was a black strip-like object on the chair.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:03:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A black strip-shaped item stayed there.", + "B": "Nothing was on it the whole time.", + "C": "At the beginning there was nothing on the chair, and later a black strip-shaped item appeared.", + "D": "At the beginning there was a black strip-shaped item, and later there was nothing behind you.", + "E": "At first you saw nothing on the chair, and later a black strip-shaped item was there." + }, + "answer_label": "D", + "answer_text": "At the beginning there was a black strip-shaped item, and later there was nothing behind you.", + "q_id": "egolife_A1_D3_023_5_T1", + "correct_answer": "At the beginning there was a black strip-shaped item, and later there was nothing behind you.", + "original_options": { + "A": "A black strip-shaped item stayed there.", + "B": "Nothing was on it the whole time.", + "C": "At the beginning there was nothing on the chair, and later a black strip-shaped item appeared.", + "D": "At the beginning there was a black strip-shaped item, and later there was nothing behind you.", + "E": "At first you saw nothing on the chair, and later a black strip-shaped item was there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_5_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was on the chair by the doorway just now?", + "answer": "At the beginning there was a black strip-shaped item, and later there was nothing behind you.", + "question_chinese": "刚刚门口的椅子上什么？", + "answer_chinese": "开始有一条黑色的条装物，后面没东西了", + "evidence": [ + { + "description": "At this moment, there was a black strip-like object on the chair.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:03:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A black strip-shaped item remained on it.", + "B": "It was empty the entire time.", + "C": "At first the chair was empty, and later a black strip-shaped item was placed on it.", + "D": "At the beginning there was a black strip-shaped item, and later there was nothing behind you.", + "E": "At the beginning the chair was empty, and later a black strip-shaped item was on it." + }, + "answer_label": "D", + "answer_text": "At the beginning there was a black strip-shaped item, and later there was nothing behind you.", + "q_id": "egolife_A1_D3_023_5_T2", + "correct_answer": "At the beginning there was a black strip-shaped item, and later there was nothing behind you.", + "original_options": { + "A": "A black strip-shaped item remained on it.", + "B": "It was empty the entire time.", + "C": "At first the chair was empty, and later a black strip-shaped item was placed on it.", + "D": "At the beginning there was a black strip-shaped item, and later there was nothing behind you.", + "E": "At the beginning the chair was empty, and later a black strip-shaped item was on it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_5_T3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was on the chair by the doorway just now?", + "answer": "At first there was a black strip-shaped item on the chair, and later a girl sat on it.", + "question_chinese": "刚刚门口的椅子上什么？", + "answer_chinese": "椅子上先有一条黑色条状物，后面有个女孩子坐在上面", + "evidence": [ + { + "description": "At this moment, there was a black strip-like object on the chair.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:09" + } + }, + { + "description": "111At this moment, there was a girl sitting on the chair.", + "timestamps": { + "start": "00:04:19", + "end": "00:04:22" + } + } + ], + "question_timestamps": "00:07:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "At first there was a black strip-shaped item on the chair, and later a girl sat on it.", + "B": "Only the black strip-shaped item.", + "C": "A girl sat there the whole time.", + "D": "At first there was nothing on the chair, and later a black strip-shaped item was left on it.", + "E": "At first a black strip-shaped item was on the chair, and later the chair was left empty." + }, + "answer_label": "A", + "answer_text": "At first there was a black strip-shaped item on the chair, and later a girl sat on it.", + "q_id": "egolife_A1_D3_023_5_T3", + "correct_answer": "At first there was a black strip-shaped item on the chair, and later a girl sat on it.", + "original_options": { + "A": "At first there was a black strip-shaped item on the chair, and later a girl sat on it.", + "B": "Only the black strip-shaped item.", + "C": "A girl sat there the whole time.", + "D": "At first there was nothing on the chair, and later a black strip-shaped item was left on it.", + "E": "At first a black strip-shaped item was on the chair, and later the chair was left empty." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_5_T4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was on the chair by the doorway just now?", + "answer": "At first there was a black strip-shaped item on the chair, and later a girl sat on it.", + "question_chinese": "刚刚门口的椅子上什么？", + "answer_chinese": "椅子上先有一条黑色条状物，后面有个女孩子坐在上面", + "evidence": [ + { + "description": "At this moment, there was a black strip-like object on the chair.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:09" + } + }, + { + "description": "111At this moment, there was a girl sitting on the chair.", + "timestamps": { + "start": "00:04:19", + "end": "00:04:22" + } + } + ], + "question_timestamps": "00:07:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Just a black strip-shaped item.", + "B": "At first there was a black strip-shaped item on the chair, and later a girl sat on it.", + "C": "Nothing was on the chair.", + "D": "At first the chair was empty, and later a black strip-shaped item appeared on it.", + "E": "At first a girl sat on the chair, and later only a black strip-shaped item remained there." + }, + "answer_label": "B", + "answer_text": "At first there was a black strip-shaped item on the chair, and later a girl sat on it.", + "q_id": "egolife_A1_D3_023_5_T4", + "correct_answer": "At first there was a black strip-shaped item on the chair, and later a girl sat on it.", + "original_options": { + "A": "Just a black strip-shaped item.", + "B": "At first there was a black strip-shaped item on the chair, and later a girl sat on it.", + "C": "Nothing was on the chair.", + "D": "At first the chair was empty, and later a black strip-shaped item appeared on it.", + "E": "At first a girl sat on the chair, and later only a black strip-shaped item remained there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did the white power bank just take?", + "answer": "I packed them into the box and then placed it on the sofa.", + "question_chinese": "刚才白色充电宝经历了什么路线？", + "answer_chinese": "我把它们收拾进了盒子，然后放到了沙发上", + "evidence": [ + { + "description": "I collected them and put them in the box", + "timestamps": { + "start": "00:05:20", + "end": "00:05:27" + } + }, + { + "description": "I put them on the sofa.", + "timestamps": { + "start": "00:05:55", + "end": "00:06:00" + } + } + ], + "question_timestamps": "00:06:28", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Into a box, then into a drawer.", + "B": "I packed them into the box and then placed it on the sofa.", + "C": "Into a box, then onto the table.", + "D": "I packed it into the box and then placed the box down on the floor.", + "E": "I first put it on the sofa and then packed it back into the box." + }, + "answer_label": "B", + "answer_text": "I packed them into the box and then placed it on the sofa.", + "q_id": "egolife_A1_D3_023_2", + "correct_answer": "I packed them into the box and then placed it on the sofa.", + "original_options": { + "A": "Into a box, then into a drawer.", + "B": "I packed them into the box and then placed it on the sofa.", + "C": "Into a box, then onto the table.", + "D": "I packed it into the box and then placed the box down on the floor.", + "E": "I first put it on the sofa and then packed it back into the box." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did the white power bank just take?", + "answer": "I placed them in various locations around the room and plugged them in.", + "question_chinese": "刚才白色充电宝经历了什么路线？", + "answer_chinese": "然后我把它们放到了房间的各个地方插上电", + "evidence": [ + { + "description": "Then I placed them in various locations around the room and plugged them in.", + "timestamps": { + "start": "00:06:42", + "end": "00:06:59" + } + } + ], + "question_timestamps": "00:09:35", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You left them unplugged.", + "B": "You kept them in one spot.", + "C": "I placed them in various locations around the room and plugged them in.", + "D": "I placed them in various locations around the room but did not plug them in anywhere.", + "E": "I placed them in various locations around the room and then turned them off instead of plugging them in." + }, + "answer_label": "C", + "answer_text": "I placed them in various locations around the room and plugged them in.", + "q_id": "egolife_A1_D3_023_1", + "correct_answer": "I placed them in various locations around the room and plugged them in.", + "original_options": { + "A": "You left them unplugged.", + "B": "You kept them in one spot.", + "C": "I placed them in various locations around the room and plugged them in.", + "D": "I placed them in various locations around the room but did not plug them in anywhere.", + "E": "I placed them in various locations around the room and then turned them off instead of plugging them in." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_9", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just touch the table in the restaurant?", + "answer": "No,I can not.", + "question_chinese": "我刚刚可以碰到餐厅的桌子吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "The table is in the living room, while I'm in my room, so I can't reach the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:02:26", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,I can not.", + "B": "Yes, I can." + }, + "answer_label": "A", + "answer_text": "No,I can not.", + "q_id": "egolife_A1_D3_023_9", + "correct_answer": "No,I can not.", + "original_options": { + "A": "No,I can not.", + "B": "Yes, I can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_10", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just touch the table in the restaurant?", + "answer": "Yes,I can.", + "question_chinese": "我刚刚可以碰到餐厅的桌子吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "When passing by the table, it was close to me, so I could reach it.", + "timestamps": { + "start": "00:03:27", + "end": "00:03:32" + } + } + ], + "question_timestamps": "00:08:18", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, I can not.", + "B": "Yes,I can." + }, + "answer_label": "B", + "answer_text": "Yes,I can.", + "q_id": "egolife_A1_D3_023_10", + "correct_answer": "Yes,I can.", + "original_options": { + "A": "No, I can not.", + "B": "Yes,I can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_4_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was sitting just now, what was around me?", + "answer": "Around you there were a computer and a drink, and they were directly in front of you.", + "question_chinese": "我刚刚坐着的时候，周围有什么？", + "answer_chinese": "你周围有电脑饮料，电脑饮料在你的正前方", + "evidence": [ + { + "description": "There are computers and a bottle on the tabl.e", + "timestamps": { + "start": "00:00:17", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A computer and a drink behind you.", + "B": "A computer and a drink to your left.", + "C": "Around you there were a computer and a drink, and they were positioned directly behind you.", + "D": "Around you there was only a computer, and it was placed directly in front of you.", + "E": "Around you there were a computer and a drink, and they were directly in front of you." + }, + "answer_label": "E", + "answer_text": "Around you there were a computer and a drink, and they were directly in front of you.", + "q_id": "egolife_A1_D3_023_4_T2", + "correct_answer": "Around you there were a computer and a drink, and they were directly in front of you.", + "original_options": { + "A": "A computer and a drink behind you.", + "B": "A computer and a drink to your left.", + "C": "Around you there were a computer and a drink, and they were positioned directly behind you.", + "D": "Around you there was only a computer, and it was placed directly in front of you.", + "E": "Around you there were a computer and a drink, and they were directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_023", + "video_length": "00:11:37", + "qid": "egolife_A1_D3_023_4_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was sitting just now, what was around me?", + "answer": "Next to you there were a computer, a drink, and a bed. The computer and drink were directly in front of you, and the bed was on the right side.", + "question_chinese": "我刚刚坐着的时候，周围有什么？", + "answer_chinese": "你旁边有电脑饮料和床，电脑饮料在你的正前方，床在你的右侧。", + "evidence": [ + { + "description": "There are computers and a bottle on the tabl.e", + "timestamps": { + "start": "00:00:17", + "end": "00:00:30" + } + }, + { + "description": "There is a bed on my right", + "timestamps": { + "start": "00:02:59", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:05:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bed was on your left.", + "B": "The computer and drink were behind you.", + "C": "Next to you there were a computer, a drink, and a bed; the bed was on your left side.", + "D": "Next to you there were a computer, a drink, and a bed. The computer and drink were directly in front of you, and the bed was on the right side.", + "E": "Next to you there were a computer, a drink, and a bed; the computer and drink were directly behind you." + }, + "answer_label": "D", + "answer_text": "Next to you there were a computer, a drink, and a bed. The computer and drink were directly in front of you, and the bed was on the right side.", + "q_id": "egolife_A1_D3_023_4_T1", + "correct_answer": "Next to you there were a computer, a drink, and a bed. The computer and drink were directly in front of you, and the bed was on the right side.", + "original_options": { + "A": "The bed was on your left.", + "B": "The computer and drink were behind you.", + "C": "Next to you there were a computer, a drink, and a bed; the bed was on your left side.", + "D": "Next to you there were a computer, a drink, and a bed. The computer and drink were directly in front of you, and the bed was on the right side.", + "E": "Next to you there were a computer, a drink, and a bed; the computer and drink were directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 20 + ] + } +] diff --git a/QAs/egolife_A1_D3_024-mcq.json b/QAs/egolife_A1_D3_024-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f413ad6b1912a5fd732ff10eb3571e8ef17d3c9e --- /dev/null +++ b/QAs/egolife_A1_D3_024-mcq.json @@ -0,0 +1,1366 @@ +[ + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_5", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did you take after just arriving at the stairwell?", + "answer": "I turned right and then turn right again.", + "question_chinese": "刚刚到楼梯间后经历了什么路线？", + "answer_chinese": "右转再右转", + "evidence": [ + { + "description": "I turned right first", + "timestamps": { + "start": "00:10:01", + "end": "00:10:03" + } + }, + { + "description": "Then I turnedr ight again", + "timestamps": { + "start": "00:10:09", + "end": "00:10:12" + } + } + ], + "question_timestamps": "00:10:57", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I turned right and then turn right again.", + "B": "I turned left twice.", + "C": "I turned right, then left.", + "D": "I went straight ahead at first, and then turned right after reaching the next corner.", + "E": "I turned right initially, then continued straight instead of making a second right turn." + }, + "answer_label": "A", + "answer_text": "I turned right and then turn right again.", + "q_id": "egolife_A1_D3_024_5", + "correct_answer": "I turned right and then turn right again.", + "original_options": { + "A": "I turned right and then turn right again.", + "B": "I turned left twice.", + "C": "I turned right, then left.", + "D": "I went straight ahead at first, and then turned right after reaching the next corner.", + "E": "I turned right initially, then continued straight instead of making a second right turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_6", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did you take after just arriving at the stairwell?", + "answer": "I turned left and then turn left again.", + "question_chinese": "刚刚到楼梯间后经历了什么路线？", + "answer_chinese": "我左转再左转", + "evidence": [ + { + "description": "I turned left first", + "timestamps": { + "start": "00:11:18", + "end": "00:11:26" + } + }, + { + "description": "Then，I turned left again", + "timestamps": { + "start": "00:11:27", + "end": "00:11:31" + } + } + ], + "question_timestamps": "00:11:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I turned right twice.", + "B": "I turned left, then right.", + "C": "I went straight ahead first, and then turned right after walking a short distance.", + "D": "I turned left and then turn left again.", + "E": "I turned left initially, then continued straight instead of making a second left turn." + }, + "answer_label": "D", + "answer_text": "I turned left and then turn left again.", + "q_id": "egolife_A1_D3_024_6", + "correct_answer": "I turned left and then turn left again.", + "original_options": { + "A": "I turned right twice.", + "B": "I turned left, then right.", + "C": "I went straight ahead first, and then turned right after walking a short distance.", + "D": "I turned left and then turn left again.", + "E": "I turned left initially, then continued straight instead of making a second left turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024-mcq_fixed_6", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was on the sofa just now, in which direction was the compute？", + "answer": "It is in front of you.", + "question_chinese": "刚刚在沙发上时，电脑在我的什么方向？", + "answer_chinese": "在我的前面", + "evidence": [ + { + "description": "The computer is in front of me.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is in front of you.", + "C": "It is to your left.", + "D": "It was behind you, closer to the wall than the sofa when you were sitting down.", + "E": "It was on your right side, roughly aligned with the sofa arm as you looked forward." + }, + "answer_label": "B", + "answer_text": "It is in front of you.", + "q_id": "egolife_A1_D3_024-mcq_fixed_6", + "correct_answer": "It is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is in front of you.", + "C": "It is to your left.", + "D": "It was behind you, closer to the wall than the sofa when you were sitting down.", + "E": "It was on your right side, roughly aligned with the sofa arm as you looked forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_8", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was just on the sofa, which direction was my computer in?", + "answer": "It is behind me.", + "question_chinese": "刚刚在沙发上时，电脑在我的什么方向？", + "answer_chinese": "在我的后面", + "evidence": [ + { + "description": "There is a computer here.", + "timestamps": { + "start": "00:04:52", + "end": "00:04:53" + } + }, + { + "description": "At this moment, when facing the sofa, the computer is behind me.", + "timestamps": { + "start": "00:04:57", + "end": "00:05:05" + } + } + ], + "question_timestamps": "00:05:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It is in front of me.", + "B": "It is to my right.", + "C": "It was to your left side while you sat on the sofa, not behind you.", + "D": "It was in front of you, positioned where you could see it without turning around.", + "E": "It is behind me." + }, + "answer_label": "E", + "answer_text": "It is behind me.", + "q_id": "egolife_A1_D3_024_8", + "correct_answer": "It is behind me.", + "original_options": { + "A": "It is in front of me.", + "B": "It is to my right.", + "C": "It was to your left side while you sat on the sofa, not behind you.", + "D": "It was in front of you, positioned where you could see it without turning around.", + "E": "It is behind me." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was on the sofa I just saw?", + "answer": "There is a sofa and a power bank.", + "question_chinese": "刚才看到的沙发上有什么？", + "answer_chinese": "有沙发和充电宝", + "evidence": [ + { + "description": "There are pillows on the sofa.", + "timestamps": { + "start": "00:03:14", + "end": "00:03:17" + } + }, + { + "description": "There is a power bank that is fully charged.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:04:26", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "There is a sofa and a book.", + "B": "There is a sofa and a tablet.", + "C": "There is a sofa and a camera placed on it, with nothing like a power bank present.", + "D": "There is a sofa and a power bank.", + "E": "There is a sofa with a charger on it, but you did not see any power bank there." + }, + "answer_label": "D", + "answer_text": "There is a sofa and a power bank.", + "q_id": "egolife_A1_D3_024_3", + "correct_answer": "There is a sofa and a power bank.", + "original_options": { + "A": "There is a sofa and a book.", + "B": "There is a sofa and a tablet.", + "C": "There is a sofa and a camera placed on it, with nothing like a power bank present.", + "D": "There is a sofa and a power bank.", + "E": "There is a sofa with a charger on it, but you did not see any power bank there." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was on the sofa that I just saw?", + "answer": "There was a charger and a black hair dryer.", + "question_chinese": "刚才看到的沙发上有什么？", + "answer_chinese": "有充电器和黑色吹风机", + "evidence": [ + { + "description": "There was a charger and a black hair drye.r", + "timestamps": { + "start": "00:04:57", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:06:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Shoes and a yellow hair dryer.", + "B": "There was a charger and a black hair dryer.", + "C": "A tablet and a red hair dryer.", + "D": "There was a charger and a white hair dryer sitting together on the sofa.", + "E": "There were books on the sofa along with a blue hair dryer placed beside them." + }, + "answer_label": "B", + "answer_text": "There was a charger and a black hair dryer.", + "q_id": "egolife_A1_D3_024_4", + "correct_answer": "There was a charger and a black hair dryer.", + "original_options": { + "A": "Shoes and a yellow hair dryer.", + "B": "There was a charger and a black hair dryer.", + "C": "A tablet and a red hair dryer.", + "D": "There was a charger and a white hair dryer sitting together on the sofa.", + "E": "There were books on the sofa along with a blue hair dryer placed beside them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_7_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I just walked past the table, could I reach out and touch it?", + "answer": "Yes—you can. The table is to your right.", + "question_chinese": "刚经过桌子的时候，我可以碰到它吗？", + "answer_chinese": "可以，桌子在你右边", + "evidence": [ + { + "description": "This table is close to me, and I can reach it", + "timestamps": { + "start": "00:08:20", + "end": "00:08:28" + } + } + ], + "question_timestamps": "00:08:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. The table is to your right.", + "B": "No—you can't. The table is to your right." + }, + "answer_label": "A", + "answer_text": "Yes—you can. The table is to your right.", + "q_id": "egolife_A1_D3_024_7_T1", + "correct_answer": "Yes—you can. The table is to your right.", + "original_options": { + "A": "Yes—you can. The table is to your right.", + "B": "No—you can't. The table is to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_9", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch it as I just walk past the table?", + "answer": "No,I can not.", + "question_chinese": "刚经过桌子的时候，我可以碰到它吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "This table is some distance away from me", + "timestamps": { + "start": "00:10:14", + "end": "00:10:18" + } + } + ], + "question_timestamps": "00:11:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, I can touch it.", + "B": "No,I can not." + }, + "answer_label": "B", + "answer_text": "No,I can not.", + "q_id": "egolife_A1_D3_024_9", + "correct_answer": "No,I can not.", + "original_options": { + "A": "Yes, I can touch it.", + "B": "No,I can not." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_7_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I just walked past the table, could I reach out and touch it?", + "answer": "No—you can't. The table is next to the area outside the bedroom, and you are inside the bedroom.", + "question_chinese": "刚经过桌子的时候，我可以碰到它吗？", + "answer_chinese": "不可以，桌子在卧室外面的旁边，我在卧室里面。", + "evidence": [ + { + "description": "I can't reach it.", + "timestamps": { + "start": "00:11:06", + "end": "00:11:09" + } + } + ], + "question_timestamps": "00:11:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you can. The table is next to the area outside the bedroom, and you are inside the bedroom.", + "B": "No—you can't. The table is next to the area outside the bedroom, and you are inside the bedroom." + }, + "answer_label": "B", + "answer_text": "No—you can't. The table is next to the area outside the bedroom, and you are inside the bedroom.", + "q_id": "egolife_A1_D3_024_7_T2", + "correct_answer": "No—you can't. The table is next to the area outside the bedroom, and you are inside the bedroom.", + "original_options": { + "A": "Yes—you can. The table is next to the area outside the bedroom, and you are inside the bedroom.", + "B": "No—you can't. The table is next to the area outside the bedroom, and you are inside the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_10", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch it when I just walk past the table?", + "answer": "Yes,I can.", + "question_chinese": "刚经过桌子的时候，我可以碰到它吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "This table is close to me, and I can reach it", + "timestamps": { + "start": "00:11:01", + "end": "00:11:03" + } + } + ], + "question_timestamps": "00:11:32", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,I can.", + "B": "No, I cannot." + }, + "answer_label": "A", + "answer_text": "Yes,I can.", + "q_id": "egolife_A1_D3_024_10", + "correct_answer": "Yes,I can.", + "original_options": { + "A": "Yes,I can.", + "B": "No, I cannot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_3_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just see?", + "answer": "You saw a white power bank to your left.", + "question_chinese": "我刚才看到什么？", + "answer_chinese": "你看到白色的充电宝，在你的左边", + "evidence": [ + { + "description": "There is a power bank that is fully charged.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A white power bank in front.", + "B": "A white power bank to your right.", + "C": "You saw a white power bank behind you, just off to the side as you looked around.", + "D": "You saw a white power bank above you, positioned higher than your eye level.", + "E": "You saw a white power bank to your left." + }, + "answer_label": "E", + "answer_text": "You saw a white power bank to your left.", + "q_id": "egolife_A1_D3_024_3_T1", + "correct_answer": "You saw a white power bank to your left.", + "original_options": { + "A": "A white power bank in front.", + "B": "A white power bank to your right.", + "C": "You saw a white power bank behind you, just off to the side as you looked around.", + "D": "You saw a white power bank above you, positioned higher than your eye level.", + "E": "You saw a white power bank to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_3_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just see?", + "answer": "You saw a charger and a black hair dryer, and they were both in front of you.", + "question_chinese": "我刚才看到什么？", + "answer_chinese": "你看到充电器和黑色吹风机，它们都在你前面。", + "evidence": [ + { + "description": "There was a charger and a black hair drye.r", + "timestamps": { + "start": "00:04:57", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:05:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They were both behind you.", + "B": "You saw a charger and a black hair dryer, and they were both in front of you.", + "C": "They were both to your left.", + "D": "You saw a charger and a black hair dryer, but they were both behind you rather than in front.", + "E": "You saw a charger and a black hair dryer, and both items were off to your right side." + }, + "answer_label": "B", + "answer_text": "You saw a charger and a black hair dryer, and they were both in front of you.", + "q_id": "egolife_A1_D3_024_3_T2", + "correct_answer": "You saw a charger and a black hair dryer, and they were both in front of you.", + "original_options": { + "A": "They were both behind you.", + "B": "You saw a charger and a black hair dryer, and they were both in front of you.", + "C": "They were both to your left.", + "D": "You saw a charger and a black hair dryer, but they were both behind you rather than in front.", + "E": "You saw a charger and a black hair dryer, and both items were off to your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_2_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see on the sofa?", + "answer": "You saw a doll and a power bank on the sofa; the power bank was to your left, and the doll was behind you.", + "question_chinese": "我看到沙发上有什么？", + "answer_chinese": "你看到沙发上有玩偶和充电宝,充电宝在你左边，玩偶在你身后。", + "evidence": [ + { + "description": "There is a power bank and dolls on the sofa.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:16" + } + } + ], + "question_timestamps": "00:03:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The doll was in front.", + "B": "The doll was to your right.", + "C": "You saw a doll and a power bank; the power bank was left, and the doll was in front.", + "D": "You saw a doll and a power bank; the power bank was left, and the doll was to your right.", + "E": "You saw a doll and a power bank on the sofa; the power bank was to your left, and the doll was behind you." + }, + "answer_label": "E", + "answer_text": "You saw a doll and a power bank on the sofa; the power bank was to your left, and the doll was behind you.", + "q_id": "egolife_A1_D3_024_2_T2", + "correct_answer": "You saw a doll and a power bank on the sofa; the power bank was to your left, and the doll was behind you.", + "original_options": { + "A": "The doll was in front.", + "B": "The doll was to your right.", + "C": "You saw a doll and a power bank; the power bank was left, and the doll was in front.", + "D": "You saw a doll and a power bank; the power bank was left, and the doll was to your right.", + "E": "You saw a doll and a power bank on the sofa; the power bank was to your left, and the doll was behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_2_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I see on the sofa?", + "answer": "You saw a hair dryer on the sofa, and it was in front of you.", + "question_chinese": "我看到沙发上有什么？", + "answer_chinese": "你看到沙发上有吹风机，它在你前面", + "evidence": [ + { + "description": "You see a hair dryer on the sofa.", + "timestamps": { + "start": "00:05:04", + "end": "00:05:09" + } + } + ], + "question_timestamps": "00:05:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was behind you.", + "B": "It was to your left.", + "C": "You saw a hair dryer on the sofa, and it was off to your right side.", + "D": "You saw a hair dryer on the sofa, and it was in front of you.", + "E": "You saw a hair dryer on the sofa, and it was above you rather than in front." + }, + "answer_label": "D", + "answer_text": "You saw a hair dryer on the sofa, and it was in front of you.", + "q_id": "egolife_A1_D3_024_2_T1", + "correct_answer": "You saw a hair dryer on the sofa, and it was in front of you.", + "original_options": { + "A": "It was behind you.", + "B": "It was to your left.", + "C": "You saw a hair dryer on the sofa, and it was off to your right side.", + "D": "You saw a hair dryer on the sofa, and it was in front of you.", + "E": "You saw a hair dryer on the sofa, and it was above you rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_9_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the computer from me?", + "answer": "The computer is in front of you.", + "question_chinese": "电脑在我的什么方向？", + "answer_chinese": "电脑在你的前面", + "evidence": [ + { + "description": "The computer is in front of me.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:0:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer is behind you.", + "B": "The computer is in front of you.", + "C": "The computer is to your left.", + "D": "The computer is to your right side, not directly ahead of you.", + "E": "The computer is above you, positioned higher than where you're looking now." + }, + "answer_label": "B", + "answer_text": "The computer is in front of you.", + "q_id": "egolife_A1_D3_024_9_T2", + "correct_answer": "The computer is in front of you.", + "original_options": { + "A": "The computer is behind you.", + "B": "The computer is in front of you.", + "C": "The computer is to your left.", + "D": "The computer is to your right side, not directly ahead of you.", + "E": "The computer is above you, positioned higher than where you're looking now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_9_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the computer from me?", + "answer": "The computer is behind you.", + "question_chinese": "电脑在我的什么方向？", + "answer_chinese": "电脑在你的身后", + "evidence": [ + { + "description": "The computer is in front of me.", + "timestamps": { + "start": "00:02:46", + "end": "00:02:49" + } + } + ], + "question_timestamps": "00:02:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer is in front of you.", + "B": "The computer is to your right.", + "C": "The computer is to your left side, not behind where you're standing.", + "D": "The computer is above you, positioned higher than your line of sight.", + "E": "The computer is behind you." + }, + "answer_label": "E", + "answer_text": "The computer is behind you.", + "q_id": "egolife_A1_D3_024_9_T1", + "correct_answer": "The computer is behind you.", + "original_options": { + "A": "The computer is in front of you.", + "B": "The computer is to your right.", + "C": "The computer is to your left side, not behind where you're standing.", + "D": "The computer is above you, positioned higher than your line of sight.", + "E": "The computer is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_1_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I touched the computer in total?", + "answer": "So far, you've touched the computer twice in total, and the computer is in front of you.", + "question_chinese": "目前为止，我一共碰过几次电脑？", + "answer_chinese": "目前为止，你一共碰过两次电脑，电脑在你前面", + "evidence": [ + { + "description": "The first time I touched a computer.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:27" + } + }, + { + "description": "The second time I touched a computer.", + "timestamps": { + "start": "00:01:51", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:01:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You've touched it once total.", + "B": "So far, you've touched the computer twice in total, and the computer is in front of you.", + "C": "You've touched it three times total.", + "D": "So far, you've touched the computer four times in total, and the computer is in front of you.", + "E": "So far, you've touched the computer twice in total, but the computer is behind you." + }, + "answer_label": "B", + "answer_text": "So far, you've touched the computer twice in total, and the computer is in front of you.", + "q_id": "egolife_A1_D3_024_1_T1", + "correct_answer": "So far, you've touched the computer twice in total, and the computer is in front of you.", + "original_options": { + "A": "You've touched it once total.", + "B": "So far, you've touched the computer twice in total, and the computer is in front of you.", + "C": "You've touched it three times total.", + "D": "So far, you've touched the computer four times in total, and the computer is in front of you.", + "E": "So far, you've touched the computer twice in total, but the computer is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_1_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times have I touched the computer in total?", + "answer": "So far, you've touched the computer three times in total, and the computer is to your front-left.", + "question_chinese": "目前为止，我一共碰过几次电脑？", + "answer_chinese": "目前为止，你一共碰过三次电脑，电脑在你左前方", + "evidence": [ + { + "description": "The first time I touched a computer.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:27" + } + }, + { + "description": "The third time I touched a computer.", + "timestamps": { + "start": "00:03:27", + "end": "00:03:35" + } + }, + { + "description": "The second time I touched a computer.", + "timestamps": { + "start": "00:01:51", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:03:41", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You've touched it twice total.", + "B": "You've touched it four times total.", + "C": "So far, you've touched the computer three times in total, and the computer is to your front-left.", + "D": "So far, you've touched the computer five times in total, and the computer is to your front-left.", + "E": "So far, you've touched the computer one time in total, and the computer is to your front-left." + }, + "answer_label": "C", + "answer_text": "So far, you've touched the computer three times in total, and the computer is to your front-left.", + "q_id": "egolife_A1_D3_024_1_T2", + "correct_answer": "So far, you've touched the computer three times in total, and the computer is to your front-left.", + "original_options": { + "A": "You've touched it twice total.", + "B": "You've touched it four times total.", + "C": "So far, you've touched the computer three times in total, and the computer is to your front-left.", + "D": "So far, you've touched the computer five times in total, and the computer is to your front-left.", + "E": "So far, you've touched the computer one time in total, and the computer is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_1", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used a computer so far?", + "answer": "Twice.", + "question_chinese": "目前为止，我一共碰过几次电脑？", + "answer_chinese": "三次", + "evidence": [ + { + "description": "The first time I touched a computer.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:27" + } + }, + { + "description": "The third time I touched a computer.", + "timestamps": { + "start": "00:03:27", + "end": "00:03:35" + } + }, + { + "description": "The second time I touched a computer.", + "timestamps": { + "start": "00:01:51", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:05:46", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Twice.", + "D": "Four times so far, counting each separate time you interacted with a computer.", + "E": "Five times in total, including brief moments when you checked something on the computer." + }, + "answer_label": "C", + "answer_text": "Twice.", + "q_id": "egolife_A1_D3_024_1", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Twice.", + "D": "Four times so far, counting each separate time you interacted with a computer.", + "E": "Five times in total, including brief moments when you checked something on the computer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024-mcq_fixed_1", + "qid_num": 20, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many computers have I encountered?", + "answer": "Four.", + "question_chinese": "目前为止，我一共碰过几次电脑？", + "answer_chinese": "六次", + "evidence": [ + { + "description": "The first time I touched a computer.", + "timestamps": { + "start": "00:00:19", + "end": "00:00:27" + } + }, + { + "description": "The third time I touched a computer.", + "timestamps": { + "start": "00:03:27", + "end": "00:03:35" + } + }, + { + "description": "The second time I touched a computer.", + "timestamps": { + "start": "00:01:51", + "end": "00:01:55" + } + }, + { + "description": "The forth time I touched a computer.", + "timestamps": { + "start": "00:00:00", + "end": "00:05:50" + } + }, + { + "description": "The fifth time I touched a computer.", + "timestamps": { + "start": "00:07:39", + "end": "00:07:52" + } + }, + { + "description": "The sixth time I touched a computer.", + "timestamps": { + "start": "00:08:58", + "end": "00:09:07" + } + } + ], + "question_timestamps": "00:11:39", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Two computers so far, including the ones you passed by without using them.", + "E": "Six computers total, counting each separate computer you noticed during the video." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_024-mcq_fixed_1", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Two computers so far, including the ones you passed by without using them.", + "E": "Six computers total, counting each separate computer you noticed during the video." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_5_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the vase from me?", + "answer": "The vase is in front of you.", + "question_chinese": "花瓶在我的什么方向？", + "answer_chinese": "花瓶在你的前面", + "evidence": [ + { + "description": "The vase is in front of you.", + "timestamps": { + "start": "00:10:26", + "end": "00:10:29" + } + } + ], + "question_timestamps": "00:10:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The vase is to your right, off to the side as you look ahead.", + "D": "The vase is above you, placed on a higher surface than your current eye level.", + "E": "The vase is in front of you." + }, + "answer_label": "E", + "answer_text": "The vase is in front of you.", + "q_id": "egolife_A1_D3_024_5_T1", + "correct_answer": "The vase is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The vase is to your right, off to the side as you look ahead.", + "D": "The vase is above you, placed on a higher surface than your current eye level.", + "E": "The vase is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_5_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the vase from me?", + "answer": "The vase is to your left.", + "question_chinese": "花瓶在我的什么方向？", + "answer_chinese": "花瓶在你的左边", + "evidence": [ + { + "description": "The vase is on your left.", + "timestamps": { + "start": "00:11:12", + "end": "00:11:17" + } + } + ], + "question_timestamps": "00:11:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The vase is to your right, on the opposite side from where you are facing.", + "D": "The vase is to your left.", + "E": "The vase is above you, sitting higher up than your current viewpoint." + }, + "answer_label": "D", + "answer_text": "The vase is to your left.", + "q_id": "egolife_A1_D3_024_5_T2", + "correct_answer": "The vase is to your left.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The vase is to your right, on the opposite side from where you are facing.", + "D": "The vase is to your left.", + "E": "The vase is above you, sitting higher up than your current viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 15, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_6_T2", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the doorway relative to me?", + "answer": "The doorway is in front of you.", + "question_chinese": "门口在我的哪里？", + "answer_chinese": "门口在你的前面", + "evidence": [ + { + "description": "The door is in front of you.", + "timestamps": { + "start": "00:02:51", + "end": "00:02:56" + } + } + ], + "question_timestamps": "00:02:58", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The doorway is behind you.", + "B": "The doorway is in front of you." + }, + "answer_label": "B", + "answer_text": "The doorway is in front of you.", + "q_id": "egolife_A1_D3_024_6_T2", + "correct_answer": "The doorway is in front of you.", + "original_options": { + "A": "The doorway is behind you.", + "B": "The doorway is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D3_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_024_6_T1", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the doorway relative to me?", + "answer": "The doorway is to your right.", + "question_chinese": "门口在我的哪里？", + "answer_chinese": "门口在你右边", + "evidence": [ + { + "description": "This table is some distance away from me", + "timestamps": { + "start": "00:04:38", + "end": "00:04:42" + } + } + ], + "question_timestamps": "00:04:44", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The doorway is to your right.", + "B": "The doorway is in front of you." + }, + "answer_label": "A", + "answer_text": "The doorway is to your right.", + "q_id": "egolife_A1_D3_024_6_T1", + "correct_answer": "The doorway is to your right.", + "original_options": { + "A": "The doorway is to your right.", + "B": "The doorway is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + } +] diff --git a/QAs/egolife_A1_D3_025-mcq.json b/QAs/egolife_A1_D3_025-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fbae686c173a23856237c942495a354498397af0 --- /dev/null +++ b/QAs/egolife_A1_D3_025-mcq.json @@ -0,0 +1,884 @@ +[ + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_10", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just touch the pizza on the table?", + "answer": "No,I can not.", + "question_chinese": "刚刚我可以碰到桌子上的披萨吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The pizza on the table is a bit far from me.", + "timestamps": { + "start": "00:02:49", + "end": "00:02:55" + } + } + ], + "question_timestamps": "00:05:28", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,I can not.", + "B": "Yes, I can." + }, + "answer_label": "A", + "answer_text": "No,I can not.", + "q_id": "egolife_A1_D3_025_10", + "correct_answer": "No,I can not.", + "original_options": { + "A": "No,I can not.", + "B": "Yes, I can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_9", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just touch the pizza on the table?", + "answer": "Yes,I can.", + "question_chinese": "刚刚我可以碰到桌子上的披萨吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The pizza is very close to me.", + "timestamps": { + "start": "00:07:22", + "end": "00:07:28" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, I cannot.", + "B": "Yes,I can." + }, + "answer_label": "B", + "answer_text": "Yes,I can.", + "q_id": "egolife_A1_D3_025_9", + "correct_answer": "Yes,I can.", + "original_options": { + "A": "No, I cannot.", + "B": "Yes,I can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_10_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now, what did I have in my hand?", + "answer": "You had held the glasses in your hand just now.", + "question_chinese": "刚刚我的手里拿过什么？", + "answer_chinese": "你刚刚手里拿过眼镜。", + "evidence": [ + { + "description": "At this moment, I am holding a pair of glasses in my hand.", + "timestamps": { + "start": "00:03:56", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You had held the glasses in your hand just now.", + "B": "You were holding the remote control.", + "C": "You had a pot in your hand.", + "D": "You were holding a remote control in your hand just now, not the glasses.", + "E": "You had a trash bag in your hand just now while you were near the sink." + }, + "answer_label": "A", + "answer_text": "You had held the glasses in your hand just now.", + "q_id": "egolife_A1_D3_025_10_T2", + "correct_answer": "You had held the glasses in your hand just now.", + "original_options": { + "A": "You had held the glasses in your hand just now.", + "B": "You were holding the remote control.", + "C": "You had a pot in your hand.", + "D": "You were holding a remote control in your hand just now, not the glasses.", + "E": "You had a trash bag in your hand just now while you were near the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_10_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now, what did I have in my hand?", + "answer": "A remote control and glasses.", + "question_chinese": "刚刚我的手里拿过什么？", + "answer_chinese": "遥控器和眼镜", + "evidence": [ + { + "description": "At this moment, I am holding a remote control in my hand.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:42" + } + }, + { + "description": "At this moment, I am holding a pair of glasses in my hand.", + "timestamps": { + "start": "00:03:56", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:06:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were holding the glasses.", + "B": "You held only the remote control.", + "C": "You were holding a pot in your hand just now while standing in front of the sink.", + "D": "A remote control and glasses.", + "E": "You had a trash bag and a pot in your hand just now, but no remote control." + }, + "answer_label": "D", + "answer_text": "A remote control and glasses.", + "q_id": "egolife_A1_D3_025_10_T1", + "correct_answer": "A remote control and glasses.", + "original_options": { + "A": "You were holding the glasses.", + "B": "You held only the remote control.", + "C": "You were holding a pot in your hand just now while standing in front of the sink.", + "D": "A remote control and glasses.", + "E": "You had a trash bag and a pot in your hand just now, but no remote control." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just have in my hand?", + "answer": "Remote control and glasses.", + "question_chinese": "刚刚我的手里拿过什么？", + "answer_chinese": "遥控器和眼镜", + "evidence": [ + { + "description": "At this moment, I am holding a remote control in my hand.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:42" + } + }, + { + "description": "At this moment, I am holding a pair of glasses in my hand.", + "timestamps": { + "start": "00:03:56", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Remote control and a flashlight.", + "B": "Remote control and glasses.", + "C": "Remote control and a watch.", + "D": "You were holding the remote control along with a notebook, not the glasses.", + "E": "You had a remote control and a pair of gloves in your hand just now." + }, + "answer_label": "B", + "answer_text": "Remote control and glasses.", + "q_id": "egolife_A1_D3_025_2", + "correct_answer": "Remote control and glasses.", + "original_options": { + "A": "Remote control and a flashlight.", + "B": "Remote control and glasses.", + "C": "Remote control and a watch.", + "D": "You were holding the remote control along with a notebook, not the glasses.", + "E": "You had a remote control and a pair of gloves in your hand just now." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_11_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You were holding a pot in your hand just now, and you were standing in front of the sink.", + "question_chinese": "刚才我手里拿着什么？", + "answer_chinese": "刚刚你手里拿着锅，你站在水槽前面。", + "evidence": [ + { + "description": "At this moment, I am holding a pot in my hand.", + "timestamps": { + "start": "00:09:34", + "end": "00:09:44" + } + } + ], + "question_timestamps": "00:09:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were holding the glasses.", + "B": "You were holding a trash bag.", + "C": "You were holding a pot in your hand just now, but you were standing behind the sink.", + "D": "You were holding a pot just now, and you were positioned to the side of the sink instead of in front.", + "E": "You were holding a pot in your hand just now, and you were standing in front of the sink." + }, + "answer_label": "E", + "answer_text": "You were holding a pot in your hand just now, and you were standing in front of the sink.", + "q_id": "egolife_A1_D3_025_11_T2", + "correct_answer": "You were holding a pot in your hand just now, and you were standing in front of the sink.", + "original_options": { + "A": "You were holding the glasses.", + "B": "You were holding a trash bag.", + "C": "You were holding a pot in your hand just now, but you were standing behind the sink.", + "D": "You were holding a pot just now, and you were positioned to the side of the sink instead of in front.", + "E": "You were holding a pot in your hand just now, and you were standing in front of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 21 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_11_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "A trash bag, a pot, and a scouring pad.", + "question_chinese": "刚才我手里拿着什么？", + "answer_chinese": "垃圾袋，锅和清洁球", + "evidence": [ + { + "description": "At this moment, I am holding a garbage bag in my hand.", + "timestamps": { + "start": "00:08:23", + "end": "00:08:40" + } + }, + { + "description": "At this moment, I am holding a pot in my hand.", + "timestamps": { + "start": "00:09:34", + "end": "00:09:44" + } + }, + { + "description": "At this moment, I am holding a cleaning ball in my hand.", + "timestamps": { + "start": "00:10:15", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:11:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A trash bag, a pot, and a scouring pad.", + "B": "A pot and a scouring pad.", + "C": "A trash bag and a pot.", + "D": "You were holding a remote control and glasses in your hand just now, not the cleaning items.", + "E": "You were holding only a pot in your hand just now while standing in front of the sink." + }, + "answer_label": "A", + "answer_text": "A trash bag, a pot, and a scouring pad.", + "q_id": "egolife_A1_D3_025_11_T1", + "correct_answer": "A trash bag, a pot, and a scouring pad.", + "original_options": { + "A": "A trash bag, a pot, and a scouring pad.", + "B": "A pot and a scouring pad.", + "C": "A trash bag and a pot.", + "D": "You were holding a remote control and glasses in your hand just now, not the cleaning items.", + "E": "You were holding only a pot in your hand just now while standing in front of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_3", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding just now?", + "answer": "Garbage bag, pot, and cleaning ball.", + "question_chinese": "刚才我手里拿着什么？", + "answer_chinese": "垃圾袋，锅和清洁球", + "evidence": [ + { + "description": "At this moment, I am holding a garbage bag in my hand.", + "timestamps": { + "start": "00:08:23", + "end": "00:08:40" + } + }, + { + "description": "At this moment, I am holding a pot in my hand.", + "timestamps": { + "start": "00:09:34", + "end": "00:09:44" + } + }, + { + "description": "At this moment, I am holding a cleaning ball in my hand.", + "timestamps": { + "start": "00:10:15", + "end": "00:10:20" + } + } + ], + "question_timestamps": "00:11:29", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Garbage bag and pot only.", + "B": "Pot and cleaning ball only.", + "C": "You were holding a trash bag and a scouring pad just now, but not the pot.", + "D": "Garbage bag, pot, and cleaning ball.", + "E": "You were holding the pot just now, but you did not have the garbage bag or the cleaning ball." + }, + "answer_label": "D", + "answer_text": "Garbage bag, pot, and cleaning ball.", + "q_id": "egolife_A1_D3_025_3", + "correct_answer": "Garbage bag, pot, and cleaning ball.", + "original_options": { + "A": "Garbage bag and pot only.", + "B": "Pot and cleaning ball only.", + "C": "You were holding a trash bag and a scouring pad just now, but not the pot.", + "D": "Garbage bag, pot, and cleaning ball.", + "E": "You were holding the pot just now, but you did not have the garbage bag or the cleaning ball." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_5_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just went out to throw away the trash?", + "answer": "You went straight first and then turned left, and the trash can was to your front-left.", + "question_chinese": "我刚出门丢垃圾经历了什么路线？", + "answer_chinese": "你先直行然后左转，垃圾桶在你左前方。", + "evidence": [ + { + "description": "I'll go straight first and then turn left.", + "timestamps": { + "start": "00:08:59", + "end": "00:09:01" + } + } + ], + "question_timestamps": "00:09:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went straight, then turned right.", + "B": "The trash can was back-left.", + "C": "You turned left first and then walked straight, and the trash can ended up to your front-left.", + "D": "You went straight first and then turned left, but the trash can was to your front-right instead.", + "E": "You went straight first and then turned left, and the trash can was to your front-left." + }, + "answer_label": "E", + "answer_text": "You went straight first and then turned left, and the trash can was to your front-left.", + "q_id": "egolife_A1_D3_025_5_T2", + "correct_answer": "You went straight first and then turned left, and the trash can was to your front-left.", + "original_options": { + "A": "You went straight, then turned right.", + "B": "The trash can was back-left.", + "C": "You turned left first and then walked straight, and the trash can ended up to your front-left.", + "D": "You went straight first and then turned left, but the trash can was to your front-right instead.", + "E": "You went straight first and then turned left, and the trash can was to your front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_5_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just went out to throw away the trash?", + "answer": "I left the room first, then went straight, and finally turned left.", + "question_chinese": "我刚出门丢垃圾经历了什么路线？", + "answer_chinese": "我先出房间，然后直行最后左转", + "evidence": [ + { + "description": "I first gathered up the garbage and went outside", + "timestamps": { + "start": "00:08:38", + "end": "00:08:56" + } + }, + { + "description": "Then kept straight ahead.", + "timestamps": { + "start": "00:08:59", + "end": "00:09:01" + } + }, + { + "description": "Then I turned left.", + "timestamps": { + "start": "00:09:01", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:10:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "I turned left before going straight.", + "B": "I left the room first, then went straight, and finally turned left.", + "C": "I stayed in the room.", + "D": "I left the room first, then turned right, and finally went straight toward the trash can.", + "E": "I left the room, went straight for a moment, and then turned around instead of turning left." + }, + "answer_label": "B", + "answer_text": "I left the room first, then went straight, and finally turned left.", + "q_id": "egolife_A1_D3_025_5_T1", + "correct_answer": "I left the room first, then went straight, and finally turned left.", + "original_options": { + "A": "I turned left before going straight.", + "B": "I left the room first, then went straight, and finally turned left.", + "C": "I stayed in the room.", + "D": "I left the room first, then turned right, and finally went straight toward the trash can.", + "E": "I left the room, went straight for a moment, and then turned around instead of turning left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_7", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I just entered the server room?", + "answer": "Once.", + "question_chinese": "我刚进了几次机房？", + "answer_chinese": "一次", + "evidence": [ + { + "description": "I entered the computer room.", + "timestamps": { + "start": "00:02:58", + "end": "00:03:04" + } + } + ], + "question_timestamps": "00:03:31", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Three times.", + "C": "You entered the server room four times in total, not just once.", + "D": "You have entered the server room two times so far, not just a single time.", + "E": "Once." + }, + "answer_label": "E", + "answer_text": "Once.", + "q_id": "egolife_A1_D3_025_7", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Three times.", + "C": "You entered the server room four times in total, not just once.", + "D": "You have entered the server room two times so far, not just a single time.", + "E": "Once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_6", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I just entered the server room?", + "answer": "Twice.", + "question_chinese": "我刚进了几次机房？", + "answer_chinese": "两次", + "evidence": [ + { + "description": "I entered the computer room.", + "timestamps": { + "start": "00:02:58", + "end": "00:03:04" + } + }, + { + "description": "I entered the computer room again", + "timestamps": { + "start": "00:03:29", + "end": "00:03:32" + } + } + ], + "question_timestamps": "00:06:12", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times, including the first entry earlier and three more quick re-entries afterward.", + "D": "Twice.", + "E": "Seven times, counting every time you crossed the doorway into the server room during the sequence." + }, + "answer_label": "D", + "answer_text": "Twice.", + "q_id": "egolife_A1_D3_025_6", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times, including the first entry earlier and three more quick re-entries afterward.", + "D": "Twice.", + "E": "Seven times, counting every time you crossed the doorway into the server room during the sequence." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_13_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I first walked into the computer room, what was around me?", + "answer": "There was a computer and a bed around you, and also a lamp. The bed and the computer were to your right, and the lamp was above your head.", + "question_chinese": "我刚进电脑房的时候，周围有什么？", + "answer_chinese": "你周围有电脑和床，还有灯，床和电脑在你右边，等在你头顶。", + "evidence": [ + { + "description": "There are a bed and a computer around you, with a lamp overhead.", + "timestamps": { + "start": "00:02:57", + "end": "00:02:59" + } + } + ], + "question_timestamps": "00:03:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Bed and computer on your left.", + "B": "Bed and computer in front of you.", + "C": "There was a computer and a bed around you, and also a lamp. The bed and the computer were to your right, and the lamp was above your head.", + "D": "There was a computer and a bed, with the lamp above your head, but both main items were behind you.", + "E": "A computer and bed were nearby, and the lamp was overhead, though the computer and bed were to your left." + }, + "answer_label": "C", + "answer_text": "There was a computer and a bed around you, and also a lamp. The bed and the computer were to your right, and the lamp was above your head.", + "q_id": "egolife_A1_D3_025_13_T2", + "correct_answer": "There was a computer and a bed around you, and also a lamp. The bed and the computer were to your right, and the lamp was above your head.", + "original_options": { + "A": "Bed and computer on your left.", + "B": "Bed and computer in front of you.", + "C": "There was a computer and a bed around you, and also a lamp. The bed and the computer were to your right, and the lamp was above your head.", + "D": "There was a computer and a bed, with the lamp above your head, but both main items were behind you.", + "E": "A computer and bed were nearby, and the lamp was overhead, though the computer and bed were to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_13_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I first walked into the computer room, what was around me?", + "answer": "There was a computer and a bed.", + "question_chinese": "我刚进电脑房的时候，周围有什么？", + "answer_chinese": "有电脑和床", + "evidence": [ + { + "description": "There were many computers in front of me and a bed", + "timestamps": { + "start": "00:02:57", + "end": "00:02:59" + } + } + ], + "question_timestamps": "00:05:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "There was a computer and a bed.", + "B": "There was only a computer.", + "C": "There was only a bed.", + "D": "There was a computer and a bed around you, and a lamp was also present above your head.", + "E": "You were surrounded by a computer setup and a lamp, but there was no bed in the room." + }, + "answer_label": "A", + "answer_text": "There was a computer and a bed.", + "q_id": "egolife_A1_D3_025_13_T1", + "correct_answer": "There was a computer and a bed.", + "original_options": { + "A": "There was a computer and a bed.", + "B": "There was only a computer.", + "C": "There was only a bed.", + "D": "There was a computer and a bed around you, and a lamp was also present above your head.", + "E": "You were surrounded by a computer setup and a lamp, but there was no bed in the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_025_8", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was around me when I first entered the computer room?", + "answer": "There were computers and a bed.", + "question_chinese": "我刚进电脑房的时候，周围有什么？", + "answer_chinese": "有电脑和床", + "evidence": [ + { + "description": "There were many computers in front of me and a bed", + "timestamps": { + "start": "00:02:57", + "end": "00:02:59" + } + } + ], + "question_timestamps": "00:05:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There were computers and a bed.", + "B": "There were laptops and a couch.", + "C": "There were desks and a television.", + "D": "There was a bed and a lamp, but no computers were around you when you entered.", + "E": "You first saw computers and a couch in the room, with no bed near you at that moment." + }, + "answer_label": "A", + "answer_text": "There were computers and a bed.", + "q_id": "egolife_A1_D3_025_8", + "correct_answer": "There were computers and a bed.", + "original_options": { + "A": "There were computers and a bed.", + "B": "There were laptops and a couch.", + "C": "There were desks and a television.", + "D": "There was a bed and a lamp, but no computers were around you when you entered.", + "E": "You first saw computers and a couch in the room, with no bed near you at that moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D3_026-mcq.json b/QAs/egolife_A1_D3_026-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..48ee1fa60a25a6052961d63a66ec113a651f4b04 --- /dev/null +++ b/QAs/egolife_A1_D3_026-mcq.json @@ -0,0 +1,1556 @@ +[ + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_13_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Just now, what route did the pot in my hand take?", + "answer": "You first held it in your hand, and then you handed it to your friend.", + "question_chinese": "刚刚我手里的锅经历了什么路线？", + "answer_chinese": "它先被我拿在手里，然后被我递给了我的朋友", + "evidence": [ + { + "description": "The pot was initially held in my hand.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:36" + } + }, + { + "description": "The pot was passed to my friend by me", + "timestamps": { + "start": "00:01:24", + "end": "00:01:52" + } + } + ], + "question_timestamps": "00:02:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You set it down on the table.", + "B": "You placed it on the stovetop.", + "C": "You held the pot briefly, then set it down on a nearby surface instead of handing it over.", + "D": "You started with it in your hand, but then you put it down rather than passing it to your friend.", + "E": "You first held it in your hand, and then you handed it to your friend." + }, + "answer_label": "E", + "answer_text": "You first held it in your hand, and then you handed it to your friend.", + "q_id": "egolife_A1_D3_026_13_T1", + "correct_answer": "You first held it in your hand, and then you handed it to your friend.", + "original_options": { + "A": "You set it down on the table.", + "B": "You placed it on the stovetop.", + "C": "You held the pot briefly, then set it down on a nearby surface instead of handing it over.", + "D": "You started with it in your hand, but then you put it down rather than passing it to your friend.", + "E": "You first held it in your hand, and then you handed it to your friend." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_13_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did the pot in my hand take just now?", + "answer": "It was first held in your hand, and then passed to your friend.", + "question_chinese": "刚刚我手里的锅经历了什么路线？", + "answer_chinese": "它先被我拿在手里，然后被我递给了我的朋友", + "evidence": [ + { + "description": "The pot was initially held in my hand.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:36" + } + }, + { + "description": "The pot was passed to my friend by me", + "timestamps": { + "start": "00:01:24", + "end": "00:01:52" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was first held in your hand, and then passed to your friend.", + "B": "It was set down on the table.", + "C": "It stayed in your hand the whole time.", + "D": "It was held in your hand, then placed on a nearby surface instead of being passed to your friend.", + "E": "You carried it briefly, set it down, and did not pass it directly to your friend afterward." + }, + "answer_label": "A", + "answer_text": "It was first held in your hand, and then passed to your friend.", + "q_id": "egolife_A1_D3_026_13_T2", + "correct_answer": "It was first held in your hand, and then passed to your friend.", + "original_options": { + "A": "It was first held in your hand, and then passed to your friend.", + "B": "It was set down on the table.", + "C": "It stayed in your hand the whole time.", + "D": "It was held in your hand, then placed on a nearby surface instead of being passed to your friend.", + "E": "You carried it briefly, set it down, and did not pass it directly to your friend afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_5_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Just now, what route did the pot in my hand take?", + "answer": "It was placed on the stovetop, and then it was placed onto the table.", + "question_chinese": "刚刚我手里的锅经历了什么路线？", + "answer_chinese": "它被放在灶台上，又被放到了桌子上", + "evidence": [ + { + "description": "At this moment，the pot is on the stove", + "timestamps": { + "start": "00:02:53", + "end": "00:02:56" + } + }, + { + "description": "At this moment，the pot is on the table", + "timestamps": { + "start": "00:05:07", + "end": "00:05:11" + } + } + ], + "question_timestamps": "00:07:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It went from table to stovetop.", + "B": "It was handed to your friend afterward.", + "C": "It was placed on the stovetop, and then it was placed onto the table.", + "D": "It was placed on the stovetop and then picked back up into your hand rather than being set onto the table.", + "E": "After being put on the table, you moved it to the stovetop, reversing the order of placements." + }, + "answer_label": "C", + "answer_text": "It was placed on the stovetop, and then it was placed onto the table.", + "q_id": "egolife_A1_D3_026_5_T1", + "correct_answer": "It was placed on the stovetop, and then it was placed onto the table.", + "original_options": { + "A": "It went from table to stovetop.", + "B": "It was handed to your friend afterward.", + "C": "It was placed on the stovetop, and then it was placed onto the table.", + "D": "It was placed on the stovetop and then picked back up into your hand rather than being set onto the table.", + "E": "After being put on the table, you moved it to the stovetop, reversing the order of placements." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_5_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did the pot in my hand just go through?", + "answer": "It was placed on the stove, and then put on the table.", + "question_chinese": "刚刚我手里的锅经历了什么路线?", + "answer_chinese": "它被放在灶台上，又被放到了桌子上", + "evidence": [ + { + "description": "At this moment，the pot is on the stove", + "timestamps": { + "start": "00:02:53", + "end": "00:02:56" + } + }, + { + "description": "At this moment，the pot is on the table", + "timestamps": { + "start": "00:05:07", + "end": "00:05:11" + } + } + ], + "question_timestamps": "00:07:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It went from the stove to the counter.", + "B": "It went from the stove to the sink.", + "C": "It was moved off the stove onto the counter, and later placed back onto the stove again.", + "D": "It was set down on the table first, then lifted and placed onto the stove afterward.", + "E": "It was placed on the stove, and then put on the table." + }, + "answer_label": "E", + "answer_text": "It was placed on the stove, and then put on the table.", + "q_id": "egolife_A1_D3_026_5_T2", + "correct_answer": "It was placed on the stove, and then put on the table.", + "original_options": { + "A": "It went from the stove to the counter.", + "B": "It went from the stove to the sink.", + "C": "It was moved off the stove onto the counter, and later placed back onto the stove again.", + "D": "It was set down on the table first, then lifted and placed onto the stove afterward.", + "E": "It was placed on the stove, and then put on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_9_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was in my hand just now?", + "answer": "I picked up a pot, an egg, a phone, and some tissues.", + "question_chinese": "刚刚我的手里拿了什么？", + "answer_chinese": "我拿了锅，蛋，手机，纸巾", + "evidence": [ + { + "description": "At this moment, I am holding a pot in my hand", + "timestamps": { + "start": "00:00:08", + "end": "00:00:22" + } + }, + { + "description": "I'm holding paper and wiping the pot.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:41" + } + }, + { + "description": "I'm holding a phone", + "timestamps": { + "start": "00:02:07", + "end": "00:02:13" + } + }, + { + "description": "I am beating eggs", + "timestamps": { + "start": "00:02:47", + "end": "00:03:03" + } + } + ], + "question_timestamps": "00:05:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "I picked up a pot and a phone.", + "B": "I picked up a pot, an egg, a phone, and some tissues.", + "C": "I picked up a pot and some tissues.", + "D": "I picked up a pot, two eggs, a phone, and some tissues.", + "E": "I picked up a pot, an egg, some tissues, and left the phone behind." + }, + "answer_label": "B", + "answer_text": "I picked up a pot, an egg, a phone, and some tissues.", + "q_id": "egolife_A1_D3_026_9_T1", + "correct_answer": "I picked up a pot, an egg, a phone, and some tissues.", + "original_options": { + "A": "I picked up a pot and a phone.", + "B": "I picked up a pot, an egg, a phone, and some tissues.", + "C": "I picked up a pot and some tissues.", + "D": "I picked up a pot, two eggs, a phone, and some tissues.", + "E": "I picked up a pot, an egg, some tissues, and left the phone behind." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_9_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was in my hand just now?", + "answer": "You took the pot, eggs, phone, and tissues.", + "question_chinese": "刚刚我的手里拿了什么？", + "answer_chinese": "我拿了锅，蛋，手机，纸巾", + "evidence": [ + { + "description": "At this moment, I am holding a pot in my hand", + "timestamps": { + "start": "00:00:08", + "end": "00:00:22" + } + }, + { + "description": "I'm holding paper and wiping the pot.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:41" + } + }, + { + "description": "I'm holding a phone", + "timestamps": { + "start": "00:02:07", + "end": "00:02:13" + } + }, + { + "description": "I am beating eggs", + "timestamps": { + "start": "00:02:47", + "end": "00:03:03" + } + } + ], + "question_timestamps": "00:05:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took the pot and phone.", + "B": "You took eggs and tissues.", + "C": "You took the pot, eggs, phone, and tissues.", + "D": "You took the pot, a single egg, and tissues, but you weren’t holding a phone.", + "E": "You took the phone and tissues, while the pot and eggs stayed where they were." + }, + "answer_label": "C", + "answer_text": "You took the pot, eggs, phone, and tissues.", + "q_id": "egolife_A1_D3_026_9_T2", + "correct_answer": "You took the pot, eggs, phone, and tissues.", + "original_options": { + "A": "You took the pot and phone.", + "B": "You took eggs and tissues.", + "C": "You took the pot, eggs, phone, and tissues.", + "D": "You took the pot, a single egg, and tissues, but you weren’t holding a phone.", + "E": "You took the phone and tissues, while the pot and eggs stayed where they were." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_1_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was in my hand just now?", + "answer": "I picked up a pot, an egg, a phone, and some tissues.", + "question_chinese": "刚刚我的手里拿了什么？", + "answer_chinese": "我拿了锅，蛋，手机，纸巾", + "evidence": [ + { + "description": "At this moment, I am holding a pot in my hand", + "timestamps": { + "start": "00:00:08", + "end": "00:00:22" + } + }, + { + "description": "I'm holding paper and wiping the pot.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:41" + } + }, + { + "description": "I'm holding a phone", + "timestamps": { + "start": "00:02:07", + "end": "00:02:13" + } + }, + { + "description": "I am beating eggs", + "timestamps": { + "start": "00:02:47", + "end": "00:03:03" + } + } + ], + "question_timestamps": "00:05:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "I picked up only a phone.", + "B": "I picked up a pot and an egg.", + "C": "I picked up a pot, an egg, and some tissues, but I didn’t pick up a phone.", + "D": "I picked up a pot, an egg, a phone, and some tissues.", + "E": "I picked up a pot, two eggs, a phone, and some tissues instead of just one egg." + }, + "answer_label": "D", + "answer_text": "I picked up a pot, an egg, a phone, and some tissues.", + "q_id": "egolife_A1_D3_026_1_T1", + "correct_answer": "I picked up a pot, an egg, a phone, and some tissues.", + "original_options": { + "A": "I picked up only a phone.", + "B": "I picked up a pot and an egg.", + "C": "I picked up a pot, an egg, and some tissues, but I didn’t pick up a phone.", + "D": "I picked up a pot, an egg, a phone, and some tissues.", + "E": "I picked up a pot, two eggs, a phone, and some tissues instead of just one egg." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_1_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding in my hand?", + "answer": "You took the pot, eggs, phone, and tissues.", + "question_chinese": "刚刚我的手里拿了什么？", + "answer_chinese": "我拿了锅，蛋，手机，纸巾", + "evidence": [ + { + "description": "At this moment, I am holding a pot in my hand", + "timestamps": { + "start": "00:00:08", + "end": "00:00:22" + } + }, + { + "description": "I'm holding paper and wiping the pot.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:41" + } + }, + { + "description": "I'm holding a phone", + "timestamps": { + "start": "00:02:07", + "end": "00:02:13" + } + }, + { + "description": "I am beating eggs", + "timestamps": { + "start": "00:02:47", + "end": "00:03:03" + } + } + ], + "question_timestamps": "00:05:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took the pot, eggs, phone, and tissues.", + "B": "You took the pot and phone.", + "C": "You took eggs and tissues.", + "D": "You took the pot, a single egg, and tissues, leaving the phone behind.", + "E": "You took the phone and tissues in hand, but you never picked up the pot or eggs." + }, + "answer_label": "A", + "answer_text": "You took the pot, eggs, phone, and tissues.", + "q_id": "egolife_A1_D3_026_1_T2", + "correct_answer": "You took the pot, eggs, phone, and tissues.", + "original_options": { + "A": "You took the pot, eggs, phone, and tissues.", + "B": "You took the pot and phone.", + "C": "You took eggs and tissues.", + "D": "You took the pot, a single egg, and tissues, leaving the phone behind.", + "E": "You took the phone and tissues in hand, but you never picked up the pot or eggs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_8_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I just saw the pot, could I reach out and touch it?", + "answer": "No.", + "question_chinese": "刚刚看到锅的时候，我可以碰到它吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "At this moment, the pot was on the stove, some distance away from me.", + "timestamps": { + "start": "00:02:54", + "end": "00:02:56" + } + }, + { + "description": "At this moment, the pot was on the stove, some distance away from me.", + "timestamps": { + "start": "00:03:17", + "end": "00:03:34" + } + } + ], + "question_timestamps": "00:04:51", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D3_026_8_T1", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_7_T1", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I just saw the pot, could I reach out and touch it?", + "answer": "Yes.", + "question_chinese": "刚刚看到锅的时候，我可以碰到它吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The pot was right on the sink, very close to me, within easy reach.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:16" + } + } + ], + "question_timestamps": "00:07:25", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D3_026_7_T1", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_7_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the pot when I just see it?", + "answer": "Yes.", + "question_chinese": "刚刚看到锅的时候我可以碰到它吗?", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The pot was right on the sink, very close to me, within easy reach.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:16" + } + } + ], + "question_timestamps": "00:07:26", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D3_026_7_T2", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_11_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I been to the computer room in total just now?", + "answer": "Once.", + "question_chinese": "我刚一共去了几次电脑房？", + "answer_chinese": "一次", + "evidence": [ + { + "description": "I arrived at the computer room", + "timestamps": { + "start": "00:08:28", + "end": "00:08:41" + } + } + ], + "question_timestamps": "00:09:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Zero times.", + "C": "Three times, because you entered the computer room on three separate visits.", + "D": "Once.", + "E": "Four times, counting each entry into the computer room as a separate trip overall." + }, + "answer_label": "D", + "answer_text": "Once.", + "q_id": "egolife_A1_D3_026_11_T2", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Zero times.", + "C": "Three times, because you entered the computer room on three separate visits.", + "D": "Once.", + "E": "Four times, counting each entry into the computer room as a separate trip overall." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_3_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I been to the computer lab in total?", + "answer": "Once.", + "question_chinese": "我刚一共去了几次电脑房？", + "answer_chinese": "一次", + "evidence": [ + { + "description": "I arrived at the computer room", + "timestamps": { + "start": "00:08:28", + "end": "00:08:41" + } + } + ], + "question_timestamps": "00:09:56", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Four times.", + "C": "Three times, since you went into the computer lab on three different occasions.", + "D": "Once.", + "E": "Six times, if you count every visit into the computer lab throughout the whole sequence." + }, + "answer_label": "D", + "answer_text": "Once.", + "q_id": "egolife_A1_D3_026_3_T2", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Four times.", + "C": "Three times, since you went into the computer lab on three different occasions.", + "D": "Once.", + "E": "Six times, if you count every visit into the computer lab throughout the whole sequence." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_11_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many times did I go to the computer room just now?", + "answer": "Once.", + "question_chinese": "我刚刚一共去了几次电脑房？", + "answer_chinese": "一次", + "evidence": [ + { + "description": "I arrived at the computer room", + "timestamps": { + "start": "00:08:28", + "end": "00:08:41" + } + } + ], + "question_timestamps": "00:09:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Zero times.", + "C": "Once.", + "D": "Three times, because you returned to the computer room for three separate visits.", + "E": "Four times, counting each time you stepped into the computer room as another trip." + }, + "answer_label": "C", + "answer_text": "Once.", + "q_id": "egolife_A1_D3_026_11_T1", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Zero times.", + "C": "Once.", + "D": "Three times, because you returned to the computer room for three separate visits.", + "E": "Four times, counting each time you stepped into the computer room as another trip." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_3_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many times did I go to the computer room just now?", + "answer": "Once.", + "question_chinese": "我刚刚一共去了几次电脑房？", + "answer_chinese": "一次", + "evidence": [ + { + "description": "I arrived at the computer room", + "timestamps": { + "start": "00:08:28", + "end": "00:08:41" + } + } + ], + "question_timestamps": "00:09:55", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Twice.", + "B": "Three times.", + "C": "Four times, since you went back into the computer room multiple separate times overall.", + "D": "Once.", + "E": "Six times, if you include every time you entered the computer room during the entire moment." + }, + "answer_label": "D", + "answer_text": "Once.", + "q_id": "egolife_A1_D3_026_3_T1", + "correct_answer": "Once.", + "original_options": { + "A": "Twice.", + "B": "Three times.", + "C": "Four times, since you went back into the computer room multiple separate times overall.", + "D": "Once.", + "E": "Six times, if you include every time you entered the computer room during the entire moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_12_T1", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many times did I go to the computer room just now?", + "answer": "Twice.", + "question_chinese": "我刚刚一共去了几次电脑房？", + "answer_chinese": "两次", + "evidence": [ + { + "description": "The first time I went to the computer room", + "timestamps": { + "start": "00:08:39", + "end": "00:08:49" + } + }, + { + "description": "The second time I went to the computer room", + "timestamps": { + "start": "00:09:52", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:11:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times, because you entered the computer room on four separate occasions in total.", + "D": "Twice.", + "E": "Zero times, meaning you never actually went into the computer room at all." + }, + "answer_label": "D", + "answer_text": "Twice.", + "q_id": "egolife_A1_D3_026_12_T1", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times, because you entered the computer room on four separate occasions in total.", + "D": "Twice.", + "E": "Zero times, meaning you never actually went into the computer room at all." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_12_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I been to the computer room in total just now?\"", + "answer": "Twice.", + "question_chinese": "我刚刚一共去了几次电脑房？", + "answer_chinese": "两次", + "evidence": [ + { + "description": "The first time I went to the computer room", + "timestamps": { + "start": "00:08:39", + "end": "00:08:49" + } + }, + { + "description": "The second time I went to the computer room", + "timestamps": { + "start": "00:09:52", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:11:31", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times, since you went into the computer room four distinct times altogether.", + "D": "Zero times, meaning you never entered the computer room during that whole period.", + "E": "Twice." + }, + "answer_label": "E", + "answer_text": "Twice.", + "q_id": "egolife_A1_D3_026_12_T2", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times, since you went into the computer room four distinct times altogether.", + "D": "Zero times, meaning you never entered the computer room during that whole period.", + "E": "Twice." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_4_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, how many times did I go to the computer room just now?", + "answer": "Twice.", + "question_chinese": "我刚刚一共去了几次电脑房？", + "answer_chinese": "两次", + "evidence": [ + { + "description": "The first time I went to the computer room", + "timestamps": { + "start": "00:08:39", + "end": "00:08:49" + } + }, + { + "description": "The second time I went to the computer room", + "timestamps": { + "start": "00:09:52", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:11:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Twice.", + "C": "Three times.", + "D": "Five times, because you visited the computer room repeatedly and entered it five separate times.", + "E": "Zero times, indicating you did not go to the computer room at any point just now." + }, + "answer_label": "B", + "answer_text": "Twice.", + "q_id": "egolife_A1_D3_026_4_T1", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Twice.", + "C": "Three times.", + "D": "Five times, because you visited the computer room repeatedly and entered it five separate times.", + "E": "Zero times, indicating you did not go to the computer room at any point just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_4_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I gone to the computer room in total?", + "answer": "Twice.", + "question_chinese": "我刚刚一共去了几次电脑房？", + "answer_chinese": "两次", + "evidence": [ + { + "description": "The first time I went to the computer room", + "timestamps": { + "start": "00:08:39", + "end": "00:08:49" + } + }, + { + "description": "The second time I went to the computer room", + "timestamps": { + "start": "00:09:52", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:11:41", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Twice.", + "C": "Three times.", + "D": "Five times, since you made five separate visits into the computer room in total.", + "E": "Zero times, meaning you never went into the computer room during the entire sequence." + }, + "answer_label": "B", + "answer_text": "Twice.", + "q_id": "egolife_A1_D3_026_4_T2", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Twice.", + "C": "Three times.", + "D": "Five times, since you made five separate visits into the computer room in total.", + "E": "Zero times, meaning you never went into the computer room during the entire sequence." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_14_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was at the sink just now, which direction was the pot from me?", + "answer": "In front of you.", + "question_chinese": "我刚刚在洗水池的时候，锅在我的什么方向？", + "answer_chinese": "在我的前面", + "evidence": [ + { + "description": "At this moment,the sink is in front of me,with the pot on it", + "timestamps": { + "start": "00:04:57", + "end": "00:05:04" + } + } + ], + "question_timestamps": "00:06:52", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "To your right as you stood at the sink, not directly in front of you.", + "E": "Right beside you on the counter, not in front as you faced the sink." + }, + "answer_label": "A", + "answer_text": "In front of you.", + "q_id": "egolife_A1_D3_026_14_T1", + "correct_answer": "In front of you.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "To your right as you stood at the sink, not directly in front of you.", + "E": "Right beside you on the counter, not in front as you faced the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_6_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was at the sink just now, which direction was the pot from me?", + "answer": "To your left.", + "question_chinese": "我刚刚在洗水池的时候，锅在我的什么方向？", + "answer_chinese": "在我的左边", + "evidence": [ + { + "description": "At this moment,the pot is on the stove,which is located on my left", + "timestamps": { + "start": "00:09:18", + "end": "00:09:23" + } + } + ], + "question_timestamps": "00:10:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your right.", + "B": "Behind you.", + "C": "In front of you as you faced the sink, not off to your left.", + "D": "To your left.", + "E": "Next to you on the counter, not specifically left or right of you." + }, + "answer_label": "D", + "answer_text": "To your left.", + "q_id": "egolife_A1_D3_026_6_T1", + "correct_answer": "To your left.", + "original_options": { + "A": "To your right.", + "B": "Behind you.", + "C": "In front of you as you faced the sink, not off to your left.", + "D": "To your left.", + "E": "Next to you on the counter, not specifically left or right of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_6_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the pot in relation to me when I was just at the sink?", + "answer": "On your left.", + "question_chinese": "我刚刚在洗水池的时候，锅在我的什么方向？", + "answer_chinese": "在我的左边", + "evidence": [ + { + "description": "At this moment,the pot is on the stove,which is located on my left", + "timestamps": { + "start": "00:09:18", + "end": "00:09:23" + } + } + ], + "question_timestamps": "00:10:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your right.", + "B": "Behind you.", + "C": "In front of you while you stood at the sink, not on your left side.", + "D": "On your left.", + "E": "Right next to you by the sink, not clearly on the left or right side." + }, + "answer_label": "D", + "answer_text": "On your left.", + "q_id": "egolife_A1_D3_026_6_T2", + "correct_answer": "On your left.", + "original_options": { + "A": "On your right.", + "B": "Behind you.", + "C": "In front of you while you stood at the sink, not on your left side.", + "D": "On your left.", + "E": "Right next to you by the sink, not clearly on the left or right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_10_T1", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just have in my hand?", + "answer": "I picked up a box, a phone, and a bowl.", + "question_chinese": "我刚刚手里拿过什么？", + "answer_chinese": "我拿过箱子手机和碗", + "evidence": [ + { + "description": "I am holding a box in my hand", + "timestamps": { + "start": "00:05:52", + "end": "00:06:03" + } + }, + { + "description": "I am holding a phone in my hand", + "timestamps": { + "start": "00:07:50", + "end": "00:08:00" + } + }, + { + "description": "I am holding a bowl in my hand", + "timestamps": { + "start": "00:08:11", + "end": "00:08:21" + } + } + ], + "question_timestamps": "00:09:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "I picked up a box and a phone.", + "B": "I picked up a box, a phone, and a bowl.", + "C": "I picked up a phone and a bowl.", + "D": "I picked up a box and a bowl, but you didn’t have a phone.", + "E": "I picked up a box, a phone, and a pot instead of the bowl." + }, + "answer_label": "B", + "answer_text": "I picked up a box, a phone, and a bowl.", + "q_id": "egolife_A1_D3_026_10_T1", + "correct_answer": "I picked up a box, a phone, and a bowl.", + "original_options": { + "A": "I picked up a box and a phone.", + "B": "I picked up a box, a phone, and a bowl.", + "C": "I picked up a phone and a bowl.", + "D": "I picked up a box and a bowl, but you didn’t have a phone.", + "E": "I picked up a box, a phone, and a pot instead of the bowl." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_10_T2", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I have in my hand just now?", + "answer": "You took the box, phone, and bowl.", + "question_chinese": "我刚刚手里拿过什么？", + "answer_chinese": "我拿过箱子手机和碗", + "evidence": [ + { + "description": "I am holding a box in my hand", + "timestamps": { + "start": "00:05:52", + "end": "00:06:03" + } + }, + { + "description": "I am holding a phone in my hand", + "timestamps": { + "start": "00:07:50", + "end": "00:08:00" + } + }, + { + "description": "I am holding a bowl in my hand", + "timestamps": { + "start": "00:08:11", + "end": "00:08:21" + } + } + ], + "question_timestamps": "00:09:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You took the box and phone.", + "B": "You took the phone and bowl.", + "C": "You took the box, phone, and bowl.", + "D": "You took the box and bowl, but you didn’t pick up the phone.", + "E": "You took the box, phone, and a pot, not a bowl, from the same spot." + }, + "answer_label": "C", + "answer_text": "You took the box, phone, and bowl.", + "q_id": "egolife_A1_D3_026_10_T2", + "correct_answer": "You took the box, phone, and bowl.", + "original_options": { + "A": "You took the box and phone.", + "B": "You took the phone and bowl.", + "C": "You took the box, phone, and bowl.", + "D": "You took the box and bowl, but you didn’t pick up the phone.", + "E": "You took the box, phone, and a pot, not a bowl, from the same spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_2_T1", + "qid_num": 25, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just have in my hand?", + "answer": "I picked up a box, a phone, and a bowl.", + "question_chinese": "我刚刚手里拿过什么？", + "answer_chinese": "我拿过箱子手机和碗", + "evidence": [ + { + "description": "I am holding a box in my hand", + "timestamps": { + "start": "00:05:52", + "end": "00:06:03" + } + }, + { + "description": "I am holding a phone in my hand", + "timestamps": { + "start": "00:07:50", + "end": "00:08:00" + } + }, + { + "description": "I am holding a bowl in my hand", + "timestamps": { + "start": "00:08:11", + "end": "00:08:21" + } + } + ], + "question_timestamps": "00:09:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "I picked up a box and a phone.", + "B": "I picked up a box and a bowl.", + "C": "I picked up a phone and a bowl, but there wasn’t a box in your hand.", + "D": "I picked up a box, a phone, and a pot rather than the bowl.", + "E": "I picked up a box, a phone, and a bowl." + }, + "answer_label": "E", + "answer_text": "I picked up a box, a phone, and a bowl.", + "q_id": "egolife_A1_D3_026_2_T1", + "correct_answer": "I picked up a box, a phone, and a bowl.", + "original_options": { + "A": "I picked up a box and a phone.", + "B": "I picked up a box and a bowl.", + "C": "I picked up a phone and a bowl, but there wasn’t a box in your hand.", + "D": "I picked up a box, a phone, and a pot rather than the bowl.", + "E": "I picked up a box, a phone, and a bowl." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_026_2_T2", + "qid_num": 26, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just have in my hand?", + "answer": "You took the box, phone, and bowl.", + "question_chinese": "我刚刚手里拿过什么？", + "answer_chinese": "我拿过箱子手机和碗", + "evidence": [ + { + "description": "I am holding a box in my hand", + "timestamps": { + "start": "00:05:52", + "end": "00:06:03" + } + }, + { + "description": "I am holding a phone in my hand", + "timestamps": { + "start": "00:07:50", + "end": "00:08:00" + } + }, + { + "description": "I am holding a bowl in my hand", + "timestamps": { + "start": "00:08:11", + "end": "00:08:21" + } + } + ], + "question_timestamps": "00:09:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You held the box, phone, and cup.", + "B": "You took the box, tablet, and bowl.", + "C": "You took the box and phone, but no bowl was involved at all.", + "D": "You grabbed the box and bowl, leaving the phone behind on the counter.", + "E": "You took the box, phone, and bowl." + }, + "answer_label": "E", + "answer_text": "You took the box, phone, and bowl.", + "q_id": "egolife_A1_D3_026_2_T2", + "correct_answer": "You took the box, phone, and bowl.", + "original_options": { + "A": "You held the box, phone, and cup.", + "B": "You took the box, tablet, and bowl.", + "C": "You took the box and phone, but no bowl was involved at all.", + "D": "You grabbed the box and bowl, leaving the phone behind on the counter.", + "E": "You took the box, phone, and bowl." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D3_027-mcq.json b/QAs/egolife_A1_D3_027-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a29e810dd9137f28a2f18f834504cb53302819f2 --- /dev/null +++ b/QAs/egolife_A1_D3_027-mcq.json @@ -0,0 +1,1139 @@ +[ + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_5_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just now when I was in the kitchen, where was the sink relative to me?", + "answer": "It was in front of you.", + "question_chinese": "刚刚在厨房的时候，洗水池在我的什么方地？", + "answer_chinese": "在我的前方", + "evidence": [ + { + "description": "It is in front of me", + "timestamps": { + "start": "00:09:33", + "end": "00:09:48" + } + } + ], + "question_timestamps": "00:11:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "It was behind you.", + "C": "It was to your left.", + "D": "It was to your right as you stood in the kitchen, not straight ahead of you.", + "E": "It was off to the side next to you, not directly in front of where you were." + }, + "answer_label": "A", + "answer_text": "It was in front of you.", + "q_id": "egolife_A1_D3_027_5_T1", + "correct_answer": "It was in front of you.", + "original_options": { + "A": "It was in front of you.", + "B": "It was behind you.", + "C": "It was to your left.", + "D": "It was to your right as you stood in the kitchen, not straight ahead of you.", + "E": "It was off to the side next to you, not directly in front of where you were." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_5_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just now when I was in the kitchen, where was the sink relative to me?", + "answer": "It was in front of you.", + "question_chinese": "刚刚在厨房的时候，洗水池在我的什么方地？", + "answer_chinese": "在我的前方", + "evidence": [ + { + "description": "It is in front of me", + "timestamps": { + "start": "00:09:33", + "end": "00:09:48" + } + } + ], + "question_timestamps": "00:11:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was behind you.", + "B": "It was to your left.", + "C": "It was in front of you.", + "D": "It was off to your right side, not directly in front of you as you stood there.", + "E": "It was behind your position in the kitchen, so you would have had to turn around to face it." + }, + "answer_label": "C", + "answer_text": "It was in front of you.", + "q_id": "egolife_A1_D3_027_5_T2", + "correct_answer": "It was in front of you.", + "original_options": { + "A": "It was behind you.", + "B": "It was to your left.", + "C": "It was in front of you.", + "D": "It was off to your right side, not directly in front of you as you stood there.", + "E": "It was behind your position in the kitchen, so you would have had to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_4_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just now in the kitchen, where was the sink relative to me?", + "answer": "It was to your right.", + "question_chinese": "刚刚在厨房里，洗水池在我的什么地方？", + "answer_chinese": "在我的右边", + "evidence": [ + { + "description": "The sink was located in front and slightly to the right of me in the kitchen.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:03:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was to your left.", + "B": "It was in front of you.", + "C": "It was behind you in the kitchen, so you would need to turn around to see it.", + "D": "It was to your right.", + "E": "It was positioned straight ahead of you rather than on either side as you faced the counter." + }, + "answer_label": "D", + "answer_text": "It was to your right.", + "q_id": "egolife_A1_D3_027_4_T1", + "correct_answer": "It was to your right.", + "original_options": { + "A": "It was to your left.", + "B": "It was in front of you.", + "C": "It was behind you in the kitchen, so you would need to turn around to see it.", + "D": "It was to your right.", + "E": "It was positioned straight ahead of you rather than on either side as you faced the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_4_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Just now in the kitchen, where was the sink relative to me?", + "answer": "It was in front of you.", + "question_chinese": "刚刚在厨房里，洗水池在我的什么地方？", + "answer_chinese": "在我的前边", + "evidence": [ + { + "description": "The sink was located in front and slightly to the right of me in the kitchen.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:11:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "It was behind you.", + "C": "It was to your right.", + "D": "It was off to your left side rather than directly ahead when you were standing in the kitchen.", + "E": "It was behind your position, meaning you would have had to rotate around to face the sink." + }, + "answer_label": "A", + "answer_text": "It was in front of you.", + "q_id": "egolife_A1_D3_027_4_T2", + "correct_answer": "It was in front of you.", + "original_options": { + "A": "It was in front of you.", + "B": "It was behind you.", + "C": "It was to your right.", + "D": "It was off to your left side rather than directly ahead when you were standing in the kitchen.", + "E": "It was behind your position, meaning you would have had to rotate around to face the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_12_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was next to the table just now, which friend was closer to me—the friend in white or the friend in green?", + "answer": "The friend in green clothes was closer to you.", + "question_chinese": "刚刚在桌子旁边的时候，是白色衣服的朋友离我更近还是绿色衣服的朋友离我更近？", + "answer_chinese": "绿色衣服的朋友离我更近", + "evidence": [ + { + "description": "The friend in green clothes was closer to me.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:06" + } + } + ], + "question_timestamps": "00:10:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The friend in white.", + "B": "They were equally close.", + "C": "Neither friend was clearly closer; both stood about the same distance from you by the table.", + "D": "Both were near you, but the friend in white clothes ended up slightly closer than the friend in green.", + "E": "The friend in green clothes was closer to you." + }, + "answer_label": "E", + "answer_text": "The friend in green clothes was closer to you.", + "q_id": "egolife_A1_D3_027_12_T1", + "correct_answer": "The friend in green clothes was closer to you.", + "original_options": { + "A": "The friend in white.", + "B": "They were equally close.", + "C": "Neither friend was clearly closer; both stood about the same distance from you by the table.", + "D": "Both were near you, but the friend in white clothes ended up slightly closer than the friend in green.", + "E": "The friend in green clothes was closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_12_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was next to the table just now, which friend was closer to me—the friend in white or the friend in green?", + "answer": "The friend in green clothes was closer to you.", + "question_chinese": "刚刚在桌子旁边的时候，是白色衣服的朋友离我更近还是绿色衣服的朋友离我更近？", + "answer_chinese": "绿色衣服的朋友离我更近", + "evidence": [ + { + "description": "The friend in green clothes was closer to me.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:06" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The friend in green clothes was closer to you.", + "B": "The friend in white was closer.", + "C": "They were far from you.", + "D": "Neither one was closer; the friend in white and the friend in green were standing about the same distance away.", + "E": "Both friends were near the table, but the friend in white was the one positioned nearer to you." + }, + "answer_label": "A", + "answer_text": "The friend in green clothes was closer to you.", + "q_id": "egolife_A1_D3_027_12_T2", + "correct_answer": "The friend in green clothes was closer to you.", + "original_options": { + "A": "The friend in green clothes was closer to you.", + "B": "The friend in white was closer.", + "C": "They were far from you.", + "D": "Neither one was closer; the friend in white and the friend in green were standing about the same distance away.", + "E": "Both friends were near the table, but the friend in white was the one positioned nearer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_7_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was next to the table just now, which friend was closer to me—the friend in white or the friend in green?", + "answer": "The friend in green clothes was closer to you.", + "question_chinese": "刚刚在桌子旁边的时候，是白色衣服的朋友离我更近还是绿色衣服的朋友离我更近？", + "answer_chinese": "绿色衣服的朋友离我更近", + "evidence": [ + { + "description": "The friend in green clothes was closer to me.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:06" + } + } + ], + "question_timestamps": "00:10:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The friend in green clothes was closer to you.", + "B": "The friend in white was closer.", + "C": "They were far away.", + "D": "Neither friend was closer; the friend in white and the friend in green were equally close to you.", + "E": "Both friends were close by the table, but the friend dressed in white was slightly nearer than the one in green." + }, + "answer_label": "A", + "answer_text": "The friend in green clothes was closer to you.", + "q_id": "egolife_A1_D3_027_7_T1", + "correct_answer": "The friend in green clothes was closer to you.", + "original_options": { + "A": "The friend in green clothes was closer to you.", + "B": "The friend in white was closer.", + "C": "They were far away.", + "D": "Neither friend was closer; the friend in white and the friend in green were equally close to you.", + "E": "Both friends were close by the table, but the friend dressed in white was slightly nearer than the one in green." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 18, + 21 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_7_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was next to the table just now, which friend was closer to me—the friend in white or the friend in green?", + "answer": "The friend in green clothes was closer to you.", + "question_chinese": "刚刚在桌子旁边的时候，是白色衣服的朋友离我更近还是绿色衣服的朋友离我更近？", + "answer_chinese": "绿色衣服的朋友离我更近", + "evidence": [ + { + "description": "The friend in green clothes was closer to me.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:06" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The friend in white was closer.", + "B": "The friend in green clothes was closer to you.", + "C": "They were far away.", + "D": "Neither friend was closer; you were about the same distance from the friend in white and the friend in green.", + "E": "Both were near you by the table, but the friend in white was a bit closer than the friend in green." + }, + "answer_label": "B", + "answer_text": "The friend in green clothes was closer to you.", + "q_id": "egolife_A1_D3_027_7_T2", + "correct_answer": "The friend in green clothes was closer to you.", + "original_options": { + "A": "The friend in white was closer.", + "B": "The friend in green clothes was closer to you.", + "C": "They were far away.", + "D": "Neither friend was closer; you were about the same distance from the friend in white and the friend in green.", + "E": "Both were near you by the table, but the friend in white was a bit closer than the friend in green." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_11_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was next to the table just now, which friend was closer to me—the one wearing green or the one wearing white?", + "answer": "The friend in white clothes was closer to you.", + "question_chinese": "刚在桌子旁边的时候，是绿色衣服朋友离我近还是白色衣服朋友离我近？", + "answer_chinese": "白色衣服的朋友离我更近", + "evidence": [ + { + "description": "The friend in white clothes was in front of me.", + "timestamps": { + "start": "00:02:49", + "end": "00:02:51" + } + }, + { + "description": "The person in green clothes was at a farther distance from me.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:05:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The friend in green was closer.", + "B": "The friend stood farther away.", + "C": "Neither friend was closer; the one wearing green and the one wearing white were at about the same distance from you.", + "D": "The friend in white clothes was closer to you.", + "E": "Both friends were near you by the table, but the friend wearing green stood slightly closer than the friend in white." + }, + "answer_label": "D", + "answer_text": "The friend in white clothes was closer to you.", + "q_id": "egolife_A1_D3_027_11_T1", + "correct_answer": "The friend in white clothes was closer to you.", + "original_options": { + "A": "The friend in green was closer.", + "B": "The friend stood farther away.", + "C": "Neither friend was closer; the one wearing green and the one wearing white were at about the same distance from you.", + "D": "The friend in white clothes was closer to you.", + "E": "Both friends were near you by the table, but the friend wearing green stood slightly closer than the friend in white." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_11_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was next to the table just now, which friend was closer to me—the one wearing green or the one wearing white?", + "answer": "The friend in white clothes was closer to you.", + "question_chinese": "刚在桌子旁边的时候，是绿色衣服朋友离我近还是白色衣服朋友离我近？", + "answer_chinese": "白色衣服的朋友离我更近", + "evidence": [ + { + "description": "The friend in white clothes was in front of me.", + "timestamps": { + "start": "00:02:49", + "end": "00:02:51" + } + }, + { + "description": "The person in green clothes was at a farther distance from me.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:05:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The friend in green was closer.", + "B": "They were about equally close.", + "C": "Both friends seemed the same distance from you while you stood next to the table.", + "D": "You were nearer to the friend in green, with the friend in white a bit farther away.", + "E": "The friend in white clothes was closer to you." + }, + "answer_label": "E", + "answer_text": "The friend in white clothes was closer to you.", + "q_id": "egolife_A1_D3_027_11_T2", + "correct_answer": "The friend in white clothes was closer to you.", + "original_options": { + "A": "The friend in green was closer.", + "B": "They were about equally close.", + "C": "Both friends seemed the same distance from you while you stood next to the table.", + "D": "You were nearer to the friend in green, with the friend in white a bit farther away.", + "E": "The friend in white clothes was closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_6_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was next to the table just now, which friend was closer to me—the one wearing green or the one wearing white?", + "answer": "The friend in white clothes was closer to you.", + "question_chinese": "刚在桌子旁边的时候，是绿色衣服朋友离我近还是白色衣服朋友离我近？", + "answer_chinese": "白色衣服的朋友离我更近", + "evidence": [ + { + "description": "The friend in white clothes was in front of me.", + "timestamps": { + "start": "00:02:49", + "end": "00:02:51" + } + }, + { + "description": "The person in green clothes was at a farther distance from me.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:05:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The friend in green.", + "B": "They were equally close.", + "C": "Neither friend appeared closer, since both were standing at roughly the same distance from you beside the table.", + "D": "Both friends were close to you, but the one wearing green was positioned nearer than the one wearing white.", + "E": "The friend in white clothes was closer to you." + }, + "answer_label": "E", + "answer_text": "The friend in white clothes was closer to you.", + "q_id": "egolife_A1_D3_027_6_T1", + "correct_answer": "The friend in white clothes was closer to you.", + "original_options": { + "A": "The friend in green.", + "B": "They were equally close.", + "C": "Neither friend appeared closer, since both were standing at roughly the same distance from you beside the table.", + "D": "Both friends were close to you, but the one wearing green was positioned nearer than the one wearing white.", + "E": "The friend in white clothes was closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_6_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I was next to the table just now, which friend was closer to me—the one wearing green or the one wearing white?", + "answer": "The friend in green clothes was closer to you.", + "question_chinese": "刚在桌子旁边的时候，是绿色衣服朋友离我近还是白色衣服朋友离我近？", + "answer_chinese": "绿色衣服的朋友离我更近", + "evidence": [ + { + "description": "The person in green clothes was at a farther distance from me.", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:10:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The friend in green clothes was closer to you.", + "B": "The friend in white.", + "C": "They were equally close.", + "D": "Neither friend was closer because they were standing at similar distances from you when you were by the table.", + "E": "Both friends were close to you, but the friend in white ended up slightly nearer than the friend in green." + }, + "answer_label": "A", + "answer_text": "The friend in green clothes was closer to you.", + "q_id": "egolife_A1_D3_027_6_T2", + "correct_answer": "The friend in green clothes was closer to you.", + "original_options": { + "A": "The friend in green clothes was closer to you.", + "B": "The friend in white.", + "C": "They were equally close.", + "D": "Neither friend was closer because they were standing at similar distances from you when you were by the table.", + "E": "Both friends were close to you, but the friend in white ended up slightly nearer than the friend in green." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_2_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What route did I take when I just came back indoors from outside?", + "answer": "I went straight first, then turned right, then turned right again.", + "question_chinese": "我刚从室外回到室内经历了什么路线？", + "answer_chinese": "我先直行右转再右转", + "evidence": [ + { + "description": "I went straight first and then turned right.", + "timestamps": { + "start": "00:05:56", + "end": "00:06:03" + } + }, + { + "description": "Finally, turn right.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:11" + } + } + ], + "question_timestamps": "00:07:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Straight, left, then right.", + "B": "Straight, right, then left.", + "C": "You went straight first, then turned left, and then turned left again to get back inside.", + "D": "I went straight first, then turned right, then turned right again.", + "E": "You walked straight, turned right, and then made a left turn before continuing indoors." + }, + "answer_label": "D", + "answer_text": "I went straight first, then turned right, then turned right again.", + "q_id": "egolife_A1_D3_027_2_T1", + "correct_answer": "I went straight first, then turned right, then turned right again.", + "original_options": { + "A": "Straight, left, then right.", + "B": "Straight, right, then left.", + "C": "You went straight first, then turned left, and then turned left again to get back inside.", + "D": "I went straight first, then turned right, then turned right again.", + "E": "You walked straight, turned right, and then made a left turn before continuing indoors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_2_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What route did I take when I just came back indoors from outside?", + "answer": "I went straight first, then turned right, then turned right again.", + "question_chinese": "我刚从室外回到室内经历了什么路线？", + "answer_chinese": "我先直行右转再右转", + "evidence": [ + { + "description": "I went straight first and then turned right.", + "timestamps": { + "start": "00:05:56", + "end": "00:06:03" + } + }, + { + "description": "Finally, turn right.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:11" + } + } + ], + "question_timestamps": "00:07:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Straight, left, then right.", + "B": "I went straight first, then turned right, then turned right again.", + "C": "Straight, right, then left.", + "D": "You went straight first, then turned left, and then turned left again to come back indoors.", + "E": "You walked straight, turned right, and then made a left turn before you ended up indoors." + }, + "answer_label": "B", + "answer_text": "I went straight first, then turned right, then turned right again.", + "q_id": "egolife_A1_D3_027_2_T2", + "correct_answer": "I went straight first, then turned right, then turned right again.", + "original_options": { + "A": "Straight, left, then right.", + "B": "I went straight first, then turned right, then turned right again.", + "C": "Straight, right, then left.", + "D": "You went straight first, then turned left, and then turned left again to come back indoors.", + "E": "You walked straight, turned right, and then made a left turn before you ended up indoors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_9_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I go to the sink in total just now?", + "answer": "Twice.", + "question_chinese": "我刚刚一共去了几次洗水池？", + "answer_chinese": "两次", + "evidence": [ + { + "description": "I am washing the blender and the pot.", + "timestamps": { + "start": "00:09:34", + "end": "00:09:55" + } + }, + { + "description": "I put a bowl down.", + "timestamps": { + "start": "00:03:03", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:11:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Twice.", + "D": "You went to the sink four times in total during that sequence.", + "E": "You never went to the sink at all; you stayed away from it the whole time." + }, + "answer_label": "C", + "answer_text": "Twice.", + "q_id": "egolife_A1_D3_027_9_T1", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Twice.", + "D": "You went to the sink four times in total during that sequence.", + "E": "You never went to the sink at all; you stayed away from it the whole time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_9_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I go to the sink in total just now?", + "answer": "You went twice.", + "question_chinese": "我刚刚一共去了几次洗水池？", + "answer_chinese": "你去了两次", + "evidence": [ + { + "description": "I am washing the blender and the pot.", + "timestamps": { + "start": "00:09:34", + "end": "00:09:55" + } + }, + { + "description": "I put a bowl down.", + "timestamps": { + "start": "00:03:03", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:11:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went once.", + "B": "You went three times.", + "C": "You went four times in total, making two extra trips beyond the first two.", + "D": "You went twice.", + "E": "You didn’t go to the sink at all; there were no sink visits in that moment." + }, + "answer_label": "D", + "answer_text": "You went twice.", + "q_id": "egolife_A1_D3_027_9_T2", + "correct_answer": "You went twice.", + "original_options": { + "A": "You went once.", + "B": "You went three times.", + "C": "You went four times in total, making two extra trips beyond the first two.", + "D": "You went twice.", + "E": "You didn’t go to the sink at all; there were no sink visits in that moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_8_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I go to the washbasin in total just now?", + "answer": "You went once.", + "question_chinese": "我刚刚一共去了几次洗漱台？", + "answer_chinese": "你去了一次", + "evidence": [ + { + "description": "I just put a bowl in the washbasin.", + "timestamps": { + "start": "00:03:04", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:05:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went twice.", + "B": "You went once.", + "C": "You didn’t go at all.", + "D": "You went to the washbasin three times in total during that segment.", + "E": "You went to the washbasin four times altogether, returning repeatedly instead of just once." + }, + "answer_label": "B", + "answer_text": "You went once.", + "q_id": "egolife_A1_D3_027_8_T1", + "correct_answer": "You went once.", + "original_options": { + "A": "You went twice.", + "B": "You went once.", + "C": "You didn’t go at all.", + "D": "You went to the washbasin three times in total during that segment.", + "E": "You went to the washbasin four times altogether, returning repeatedly instead of just once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 6, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_8_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I go to the washbasin in total just now?", + "answer": "You went twice.", + "question_chinese": "我刚刚一共去了几次洗漱台？", + "answer_chinese": "你去了两次", + "evidence": [ + { + "description": "I just put a bowl in the washbasin.", + "timestamps": { + "start": "00:03:04", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:11:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went once.", + "B": "You went three times.", + "C": "You went twice.", + "D": "You went four times in total, returning to the washbasin after each brief stop away.", + "E": "You never went to the washbasin; you stayed away from it the entire time just now." + }, + "answer_label": "C", + "answer_text": "You went twice.", + "q_id": "egolife_A1_D3_027_8_T2", + "correct_answer": "You went twice.", + "original_options": { + "A": "You went once.", + "B": "You went three times.", + "C": "You went twice.", + "D": "You went four times in total, returning to the washbasin after each brief stop away.", + "E": "You never went to the washbasin; you stayed away from it the entire time just now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_3_T2", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take just now, from when I put down the bowl to when I went outside?", + "answer": "I went straight first, then turned left, and finally turned left.", + "question_chinese": "我刚刚从放下碗到去室外经历了什么路线？", + "answer_chinese": "我先直行，再左转，最后左转", + "evidence": [ + { + "description": "I go stright first", + "timestamps": { + "start": "00:03:09", + "end": "00:03:10" + } + }, + { + "description": "Then I turned left", + "timestamps": { + "start": "00:03:10", + "end": "00:03:12" + } + }, + { + "description": "At last，I turned left", + "timestamps": { + "start": "00:03:12", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:05:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Straight, then right, then left.", + "B": "Straight, then left, then right.", + "C": "You went straight at first, turned right at the corner, and then turned right again before going outside.", + "D": "I went straight first, then turned left, and finally turned left.", + "E": "You walked straight ahead, made a left turn, and then continued straight without making the final left turn." + }, + "answer_label": "D", + "answer_text": "I went straight first, then turned left, and finally turned left.", + "q_id": "egolife_A1_D3_027_3_T2", + "correct_answer": "I went straight first, then turned left, and finally turned left.", + "original_options": { + "A": "Straight, then right, then left.", + "B": "Straight, then left, then right.", + "C": "You went straight at first, turned right at the corner, and then turned right again before going outside.", + "D": "I went straight first, then turned left, and finally turned left.", + "E": "You walked straight ahead, made a left turn, and then continued straight without making the final left turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_027_3_T1", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take just now, from when I put down the bowl to when I went outside?", + "answer": "I went straight first, then turned left, and finally turned left.", + "question_chinese": "我刚刚从放下碗到去室外经历了什么路线？", + "answer_chinese": "我先直行，再左转，最后左转", + "evidence": [ + { + "description": "I go stright first", + "timestamps": { + "start": "00:03:09", + "end": "00:03:10" + } + }, + { + "description": "Then I turned left", + "timestamps": { + "start": "00:03:10", + "end": "00:03:12" + } + }, + { + "description": "At last，I turned left", + "timestamps": { + "start": "00:03:12", + "end": "00:03:13" + } + } + ], + "question_timestamps": "00:05:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Straight, then right, then left.", + "B": "Straight, then left, then right.", + "C": "I went straight first, then turned left, and finally turned left.", + "D": "You went straight first, turned right near the doorway, and then turned left as you headed outside.", + "E": "You walked straight ahead, turned left once, and then went straight out without turning left a second time." + }, + "answer_label": "C", + "answer_text": "I went straight first, then turned left, and finally turned left.", + "q_id": "egolife_A1_D3_027_3_T1", + "correct_answer": "I went straight first, then turned left, and finally turned left.", + "original_options": { + "A": "Straight, then right, then left.", + "B": "Straight, then left, then right.", + "C": "I went straight first, then turned left, and finally turned left.", + "D": "You went straight first, turned right near the doorway, and then turned left as you headed outside.", + "E": "You walked straight ahead, turned left once, and then went straight out without turning left a second time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D3_028-mcq.json b/QAs/egolife_A1_D3_028-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3e0b866013f24b2809870483fba27cfe1052a03e --- /dev/null +++ b/QAs/egolife_A1_D3_028-mcq.json @@ -0,0 +1,1285 @@ +[ + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_11_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take from the kitchen to the bedroom just now?", + "answer": "YOU went straight first and then turned right.", + "question_chinese": "刚刚从厨房到卧室，我经历了什么路线？", + "answer_chinese": "我先直行再右转", + "evidence": [ + { + "description": "I went straight first", + "timestamps": { + "start": "00:07:30", + "end": "00:07:34" + } + }, + { + "description": "And then turned right", + "timestamps": { + "start": "00:07:34", + "end": "00:07:37" + } + } + ], + "question_timestamps": "00:09:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "YOU went straight first and then turned right.", + "B": "You went straight, then left.", + "C": "You turned right first, then straight.", + "D": "You turned left as you left the kitchen, then went straight down the hallway into the bedroom.", + "E": "You went straight out of the kitchen, continued forward, and didn’t make the right turn before reaching the bedroom." + }, + "answer_label": "A", + "answer_text": "YOU went straight first and then turned right.", + "q_id": "egolife_A1_D3_028_11_T1", + "correct_answer": "YOU went straight first and then turned right.", + "original_options": { + "A": "YOU went straight first and then turned right.", + "B": "You went straight, then left.", + "C": "You turned right first, then straight.", + "D": "You turned left as you left the kitchen, then went straight down the hallway into the bedroom.", + "E": "You went straight out of the kitchen, continued forward, and didn’t make the right turn before reaching the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_11_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take from the kitchen to the bedroom just now?", + "answer": "YOU went straight first and then turned right.", + "question_chinese": "刚刚从厨房到卧室，我经历了什么路线？", + "answer_chinese": "我先直行再右转", + "evidence": [ + { + "description": "I went straight first", + "timestamps": { + "start": "00:07:30", + "end": "00:07:34" + } + }, + { + "description": "And then turned right", + "timestamps": { + "start": "00:07:34", + "end": "00:07:37" + } + } + ], + "question_timestamps": "00:09:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "YOU went straight first and then turned right.", + "B": "You went straight, then left.", + "C": "You turned right first, then straight.", + "D": "You exited the kitchen by turning left, then walked straight until you reached the bedroom doorway.", + "E": "You went straight from the kitchen and then turned left instead of turning right to get to the bedroom." + }, + "answer_label": "A", + "answer_text": "YOU went straight first and then turned right.", + "q_id": "egolife_A1_D3_028_11_T2", + "correct_answer": "YOU went straight first and then turned right.", + "original_options": { + "A": "YOU went straight first and then turned right.", + "B": "You went straight, then left.", + "C": "You turned right first, then straight.", + "D": "You exited the kitchen by turning left, then walked straight until you reached the bedroom doorway.", + "E": "You went straight from the kitchen and then turned left instead of turning right to get to the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_5_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did the book I just picked up take?", + "answer": "It was first picked up by you from the table, and then you put it back on the table.", + "question_chinese": "刚刚我拿的那本书经历了什么路线？", + "answer_chinese": "它先被我从桌子上拿起来，然后又被我放到了桌子上", + "evidence": [ + { + "description": "It was first picked up by me from the table,", + "timestamps": { + "start": "00:06:56", + "end": "00:07:01" + } + }, + { + "description": "And then I put it back on the table.", + "timestamps": { + "start": "00:07:27", + "end": "00:07:31" + } + } + ], + "question_timestamps": "00:08:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Table to bookshelf.", + "B": "It was first picked up by you from the table, and then you put it back on the table.", + "C": "Desk to desk.", + "D": "You picked the book up from the table and then set it down on the counter instead of returning it.", + "E": "You picked it up from the table and placed it on a shelf rather than the table." + }, + "answer_label": "B", + "answer_text": "It was first picked up by you from the table, and then you put it back on the table.", + "q_id": "egolife_A1_D3_028_5_T1", + "correct_answer": "It was first picked up by you from the table, and then you put it back on the table.", + "original_options": { + "A": "Table to bookshelf.", + "B": "It was first picked up by you from the table, and then you put it back on the table.", + "C": "Desk to desk.", + "D": "You picked the book up from the table and then set it down on the counter instead of returning it.", + "E": "You picked it up from the table and placed it on a shelf rather than the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 20, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_5_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did the book I just picked up take?", + "answer": "It was first picked up by you from the table, and then you put it back on the table.", + "question_chinese": "刚刚我拿的那本书经历了什么路线？", + "answer_chinese": "它先被我从桌子上拿起来，然后又被我放到了桌子上", + "evidence": [ + { + "description": "It was first picked up by me from the table,", + "timestamps": { + "start": "00:06:56", + "end": "00:07:01" + } + }, + { + "description": "And then I put it back on the table.", + "timestamps": { + "start": "00:07:27", + "end": "00:07:31" + } + } + ], + "question_timestamps": "00:08:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Table to bookshelf.", + "B": "It was first picked up by you from the table, and then you put it back on the table.", + "C": "Table to counter.", + "D": "You grabbed the book from the desk and then returned it to the desk rather than using the table.", + "E": "You picked it up from the table, moved it away briefly, and then left it on the counter instead of the table." + }, + "answer_label": "B", + "answer_text": "It was first picked up by you from the table, and then you put it back on the table.", + "q_id": "egolife_A1_D3_028_5_T2", + "correct_answer": "It was first picked up by you from the table, and then you put it back on the table.", + "original_options": { + "A": "Table to bookshelf.", + "B": "It was first picked up by you from the table, and then you put it back on the table.", + "C": "Table to counter.", + "D": "You grabbed the book from the desk and then returned it to the desk rather than using the table.", + "E": "You picked it up from the table, moved it away briefly, and then left it on the counter instead of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 19, + 22 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_8_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I have in my hand just now?", + "answer": "A white object, plastic wrap, and a phone.", + "question_chinese": "刚刚我的手里拿了什么？", + "answer_chinese": "一个白色物体，保鲜膜和手机", + "evidence": [ + { + "description": "I am holding a white object.", + "timestamps": { + "start": "00:03:23", + "end": "00:03:32" + } + }, + { + "description": "I am holding plastic wrap.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:31" + } + }, + { + "description": "I am holding my phone", + "timestamps": { + "start": "00:05:33", + "end": "00:05:37" + } + } + ], + "question_timestamps": "00:06:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A white object and a phone.", + "B": "Plastic wrap and a phone.", + "C": "A white object, plastic wrap, and a phone.", + "D": "You were holding a black object along with plastic wrap and a phone in your hand just now.", + "E": "You had a white object and plastic wrap, but instead of a phone you were holding a small calculator." + }, + "answer_label": "C", + "answer_text": "A white object, plastic wrap, and a phone.", + "q_id": "egolife_A1_D3_028_8_T1", + "correct_answer": "A white object, plastic wrap, and a phone.", + "original_options": { + "A": "A white object and a phone.", + "B": "Plastic wrap and a phone.", + "C": "A white object, plastic wrap, and a phone.", + "D": "You were holding a black object along with plastic wrap and a phone in your hand just now.", + "E": "You had a white object and plastic wrap, but instead of a phone you were holding a small calculator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_8_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I have in my hand just now?", + "answer": "A white object, plastic wrap, and a phone.", + "question_chinese": "刚刚我的手里拿了什么？", + "answer_chinese": "一个白色物体，保鲜膜和手机", + "evidence": [ + { + "description": "I am holding a white object.", + "timestamps": { + "start": "00:03:23", + "end": "00:03:32" + } + }, + { + "description": "I am holding plastic wrap.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:31" + } + }, + { + "description": "I am holding my phone", + "timestamps": { + "start": "00:05:33", + "end": "00:05:37" + } + } + ], + "question_timestamps": "00:06:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A black object and a phone.", + "B": "Plastic wrap and a notebook.", + "C": "A white object, plastic wrap, and a phone.", + "D": "A white object, bubble wrap, and a phone you held together briefly.", + "E": "A white object, plastic wrap, and a small notebook gathered in your hand briefly." + }, + "answer_label": "C", + "answer_text": "A white object, plastic wrap, and a phone.", + "q_id": "egolife_A1_D3_028_8_T2", + "correct_answer": "A white object, plastic wrap, and a phone.", + "original_options": { + "A": "A black object and a phone.", + "B": "Plastic wrap and a notebook.", + "C": "A white object, plastic wrap, and a phone.", + "D": "A white object, bubble wrap, and a phone you held together briefly.", + "E": "A white object, plastic wrap, and a small notebook gathered in your hand briefly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_9", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I first walked in, was the refrigerator on my left or my right?", + "answer": "On my left.", + "question_chinese": "刚进门的时候，冰箱在我的左边还是右边？", + "answer_chinese": "在我的左边", + "evidence": [ + { + "description": "The refrigerator is on the left side of the door. When I enter, it is on my left.", + "timestamps": { + "start": "00:07:46", + "end": "00:07:55" + } + } + ], + "question_timestamps": "00:09:13", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "On my right.", + "B": "On my left." + }, + "answer_label": "B", + "answer_text": "On my left.", + "q_id": "egolife_A1_D3_028_9", + "correct_answer": "On my left.", + "original_options": { + "A": "On my right.", + "B": "On my left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_10", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I first walked in, was the refrigerator on my left or my right?", + "answer": "On my right.", + "question_chinese": "刚进门的时候，冰箱在我的左边还是右边？", + "answer_chinese": "在我的右边", + "evidence": [ + { + "description": "At this moment, the refrigerator is on the right side of the door, which means it is on my right.", + "timestamps": { + "start": "00:10:13", + "end": "00:10:21" + } + } + ], + "question_timestamps": "00:10:33", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "On my right.", + "B": "On my left." + }, + "answer_label": "A", + "answer_text": "On my right.", + "q_id": "egolife_A1_D3_028_10", + "correct_answer": "On my right.", + "original_options": { + "A": "On my right.", + "B": "On my left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_10_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I been outside in total?", + "answer": "Four.", + "question_chinese": "我一共去过几次室外？", + "answer_chinese": "三次", + "evidence": [ + { + "description": "The first time I went outside", + "timestamps": { + "start": "00:05:19", + "end": "00:05:29" + } + }, + { + "description": "The second time I went outside", + "timestamps": { + "start": "00:07:51", + "end": "00:08:01" + } + }, + { + "description": "The third time I went outside", + "timestamps": { + "start": "00:11:46", + "end": "00:11:54" + } + } + ], + "question_timestamps": "00:11:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "Two times in total, counting each separate trip outside you made during the video.", + "E": "One time overall, meaning you only stepped outdoors once across all the scenes shown." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_028_10_T1", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "Four.", + "D": "Two times in total, counting each separate trip outside you made during the video.", + "E": "One time overall, meaning you only stepped outdoors once across all the scenes shown." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_10_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I been outside in total?", + "answer": "Four.", + "question_chinese": "我一共去过几次室外？", + "answer_chinese": "三次", + "evidence": [ + { + "description": "The first time I went outside", + "timestamps": { + "start": "00:05:19", + "end": "00:05:29" + } + }, + { + "description": "The second time I went outside", + "timestamps": { + "start": "00:07:51", + "end": "00:08:01" + } + }, + { + "description": "The third time I went outside", + "timestamps": { + "start": "00:11:46", + "end": "00:11:54" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Five.", + "C": "Four.", + "D": "Three times altogether, treating each exit outdoors as a separate instance across the whole clip.", + "E": "Two times total, counting only the moments you fully went outside rather than pausing at the doorway." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_028_10_T2", + "correct_answer": "Four.", + "original_options": { + "A": "One.", + "B": "Five.", + "C": "Four.", + "D": "Three times altogether, treating each exit outdoors as a separate instance across the whole clip.", + "E": "Two times total, counting only the moments you fully went outside rather than pausing at the doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_6", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I been outdoors in total?", + "answer": "Third.", + "question_chinese": "我一共去过几次室外？", + "answer_chinese": "三次", + "evidence": [ + { + "description": "The first time I went outside", + "timestamps": { + "start": "00:05:19", + "end": "00:05:29" + } + }, + { + "description": "The second time I went outside", + "timestamps": { + "start": "00:07:51", + "end": "00:08:01" + } + }, + { + "description": "The third time I went outside", + "timestamps": { + "start": "00:11:46", + "end": "00:11:54" + } + } + ], + "question_timestamps": "00:12:09", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Third.", + "B": "One time.", + "C": "Two times.", + "D": "Four times in total, counting each separate moment you went outdoors during the sequence.", + "E": "Five times altogether, including every distinct trip outside you made from start to finish." + }, + "answer_label": "A", + "answer_text": "Third.", + "q_id": "egolife_A1_D3_028_6", + "correct_answer": "Third.", + "original_options": { + "A": "Third.", + "B": "One time.", + "C": "Two times.", + "D": "Four times in total, counting each separate moment you went outdoors during the sequence.", + "E": "Five times altogether, including every distinct trip outside you made from start to finish." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_9_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I been outside in total just now?", + "answer": "Five.", + "question_chinese": "我刚刚一共去过几次室外？", + "answer_chinese": "两次", + "evidence": [ + { + "description": "The first time I went outside", + "timestamps": { + "start": "00:05:15", + "end": "00:05:29" + } + }, + { + "description": "The second time I went outside", + "timestamps": { + "start": "00:07:48", + "end": "00:08:01" + } + } + ], + "question_timestamps": "00:10:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "Three.", + "D": "Two times in total, counting the separate occasions you went outdoors up to that moment.", + "E": "One time overall, meaning you had only stepped outside once just before the question was asked." + }, + "answer_label": "B", + "answer_text": "Five.", + "q_id": "egolife_A1_D3_028_9_T1", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "Three.", + "D": "Two times in total, counting the separate occasions you went outdoors up to that moment.", + "E": "One time overall, meaning you had only stepped outside once just before the question was asked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_9_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I been outside in total just now?", + "answer": "Twice.", + "question_chinese": "我刚刚一共去过几次室外？", + "answer_chinese": "两次", + "evidence": [ + { + "description": "The first time I went outside", + "timestamps": { + "start": "00:05:15", + "end": "00:05:29" + } + }, + { + "description": "The second time I went outside", + "timestamps": { + "start": "00:07:48", + "end": "00:08:01" + } + } + ], + "question_timestamps": "00:10:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times total, if you count each separate exit outdoors as its own distinct trip.", + "D": "Five times altogether, including every brief step outside along with the longer outdoor segments.", + "E": "Twice." + }, + "answer_label": "E", + "answer_text": "Twice.", + "q_id": "egolife_A1_D3_028_9_T2", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Four times total, if you count each separate exit outdoors as its own distinct trip.", + "D": "Five times altogether, including every brief step outside along with the longer outdoor segments.", + "E": "Twice." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_5", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I been outdoors in total?", + "answer": "Twice.", + "question_chinese": "我刚刚一共去过几次室外？", + "answer_chinese": "两次", + "evidence": [ + { + "description": "The first time I went outside", + "timestamps": { + "start": "00:05:15", + "end": "00:05:29" + } + }, + { + "description": "The second time I went outside", + "timestamps": { + "start": "00:07:48", + "end": "00:08:01" + } + } + ], + "question_timestamps": "00:11:03", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "Twice.", + "D": "Four times in total, counting each separate time you stepped outside during the video.", + "E": "Five times overall, including every moment you went outside and came back indoors." + }, + "answer_label": "C", + "answer_text": "Twice.", + "q_id": "egolife_A1_D3_028_5", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "Twice.", + "D": "Four times in total, counting each separate time you stepped outside during the video.", + "E": "Five times overall, including every moment you went outside and came back indoors." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_7_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "Pizza and phone.", + "question_chinese": "我刚刚手里拿了什么东西？", + "answer_chinese": "披萨和手机", + "evidence": [ + { + "description": "I'm holding a pizza in my hand.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:06" + } + }, + { + "description": "I'm holding a phone in my hand.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:03:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Notebook and pen.", + "B": "Book and watch.", + "C": "You were holding a sandwich and your phone in your hand just a moment ago.", + "D": "You had a warm sub in one hand while keeping a phone in the other.", + "E": "Pizza and phone." + }, + "answer_label": "E", + "answer_text": "Pizza and phone.", + "q_id": "egolife_A1_D3_028_7_T1", + "correct_answer": "Pizza and phone.", + "original_options": { + "A": "Notebook and pen.", + "B": "Book and watch.", + "C": "You were holding a sandwich and your phone in your hand just a moment ago.", + "D": "You had a warm sub in one hand while keeping a phone in the other.", + "E": "Pizza and phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_7_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "Pizza and phone.", + "question_chinese": "我刚刚手里拿了什么东西？", + "answer_chinese": "披萨和手机", + "evidence": [ + { + "description": "I'm holding a pizza in my hand.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:06" + } + }, + { + "description": "I'm holding a phone in my hand.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:03:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Notebook and pen.", + "B": "Book and watch.", + "C": "You were holding a sandwich and your phone in your hand just a moment ago.", + "D": "You had a warm sub in one hand while keeping a phone in the other.", + "E": "Pizza and phone." + }, + "answer_label": "E", + "answer_text": "Pizza and phone.", + "q_id": "egolife_A1_D3_028_7_T2", + "correct_answer": "Pizza and phone.", + "original_options": { + "A": "Notebook and pen.", + "B": "Book and watch.", + "C": "You were holding a sandwich and your phone in your hand just a moment ago.", + "D": "You had a warm sub in one hand while keeping a phone in the other.", + "E": "Pizza and phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_1_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just have in my hand?", + "answer": "Pizza and phone.", + "question_chinese": "我刚刚手里拿了什么东西？", + "answer_chinese": "披萨和手机", + "evidence": [ + { + "description": "I'm holding a pizza in my hand.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:06" + } + }, + { + "description": "I'm holding a phone in my hand.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:03:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Pizza and keys.", + "B": "Pizza and wallet.", + "C": "You were holding a pizza along with a watch in your hand just before this moment.", + "D": "You had a pizza in one hand and a tablet in the other a second ago.", + "E": "Pizza and phone." + }, + "answer_label": "E", + "answer_text": "Pizza and phone.", + "q_id": "egolife_A1_D3_028_1_T1", + "correct_answer": "Pizza and phone.", + "original_options": { + "A": "Pizza and keys.", + "B": "Pizza and wallet.", + "C": "You were holding a pizza along with a watch in your hand just before this moment.", + "D": "You had a pizza in one hand and a tablet in the other a second ago.", + "E": "Pizza and phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_1_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just have in my hand?", + "answer": "Pizza and phone.", + "question_chinese": "我刚刚手里拿了什么东西？", + "answer_chinese": "披萨和手机", + "evidence": [ + { + "description": "I'm holding a pizza in my hand.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:06" + } + }, + { + "description": "I'm holding a phone in my hand.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:03:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Pizza and keys.", + "B": "Pizza and wallet.", + "C": "You were holding a pizza along with a watch in your hand just before this moment.", + "D": "You had a pizza in one hand and a tablet in the other a second ago.", + "E": "Pizza and phone." + }, + "answer_label": "E", + "answer_text": "Pizza and phone.", + "q_id": "egolife_A1_D3_028_1_T2", + "correct_answer": "Pizza and phone.", + "original_options": { + "A": "Pizza and keys.", + "B": "Pizza and wallet.", + "C": "You were holding a pizza along with a watch in your hand just before this moment.", + "D": "You had a pizza in one hand and a tablet in the other a second ago.", + "E": "Pizza and phone." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_2_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the grill when I just walk up to it?", + "answer": "Yes，you can.", + "question_chinese": "我刚刚走向烧烤架的时候，可以碰到它吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The barbecue grill is right in front of me now, and I can touch it.", + "timestamps": { + "start": "00:05:29", + "end": "00:05:39" + } + } + ], + "question_timestamps": "00:07:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you cannot.", + "B": "Yes，you can." + }, + "answer_label": "B", + "answer_text": "Yes，you can.", + "q_id": "egolife_A1_D3_028_2_T1", + "correct_answer": "Yes，you can.", + "original_options": { + "A": "No, you cannot.", + "B": "Yes，you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_2_T2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the grill when I just walk up to it?", + "answer": "Yes.", + "question_chinese": "我刚刚走向烧烤架的时候，可以碰到它吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The barbecue grill is right in front of me now, and I can touch it.", + "timestamps": { + "start": "00:05:29", + "end": "00:05:39" + } + } + ], + "question_timestamps": "00:07:51", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D3_028_2_T2", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_3_T1", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "As I was walking towards the barbecue grill, can I touch him?", + "answer": "No，you can not.", + "question_chinese": "我刚刚走向烧烤架的时候，我可以碰到他吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "It is in front of me, but still at a distance.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:38" + } + } + ], + "question_timestamps": "00:10:37", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No，you can not.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No，you can not.", + "q_id": "egolife_A1_D3_028_3_T1", + "correct_answer": "No，you can not.", + "original_options": { + "A": "No，you can not.", + "B": "Yes, you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D3_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_028_3_T2", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "As I was walking towards the barbecue grill, can I touch him?", + "answer": "No.", + "question_chinese": "我刚刚走向烧烤架的时候，我可以碰到他吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "It is in front of me, but still at a distance.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:38" + } + } + ], + "question_timestamps": "00:10:38", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D3_028_3_T2", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D3_029-mcq.json b/QAs/egolife_A1_D3_029-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..34cd188ed5c8135c7b8b44ba51ff622599fb8f1b --- /dev/null +++ b/QAs/egolife_A1_D3_029-mcq.json @@ -0,0 +1,1087 @@ +[ + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_8_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was by the barbecue grill just now, what did the fire look like?", + "answer": "You saw that the fire in front of you was out.", + "question_chinese": "刚刚在烧烤架旁时，我看见火是什么状态？", + "answer_chinese": "你看见面前的火是灭的", + "evidence": [ + { + "description": "You see the fire on the grill is out.", + "timestamps": { + "start": "00:03:22", + "end": "00:03:23" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was burning.", + "B": "It was smoky.", + "C": "You saw that the fire in front of you was out.", + "D": "You saw that the fire behind you was still burning as you stood near the barbecue grill.", + "E": "You noticed the fire to your left was lit, with small flames flickering as you approached the grill." + }, + "answer_label": "C", + "answer_text": "You saw that the fire in front of you was out.", + "q_id": "egolife_A1_D3_029_8_T2", + "correct_answer": "You saw that the fire in front of you was out.", + "original_options": { + "A": "It was burning.", + "B": "It was smoky.", + "C": "You saw that the fire in front of you was out.", + "D": "You saw that the fire behind you was still burning as you stood near the barbecue grill.", + "E": "You noticed the fire to your left was lit, with small flames flickering as you approached the grill." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_3_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was by the barbecue grill just now, what did the fire look like?", + "answer": "You saw that the fire in front of you was out.", + "question_chinese": "刚刚在烧烤架旁时，我看见火是什么状态？", + "answer_chinese": "你看见面前的火是灭的", + "evidence": [ + { + "description": "The fire by the barbecue grill was extinguished.", + "timestamps": { + "start": "00:03:26", + "end": "00:03:34" + } + } + ], + "question_timestamps": "00:06:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was burning.", + "B": "It was smoky.", + "C": "You saw that the fire behind you was still burning as you stood near the barbecue grill.", + "D": "You noticed the fire to your left was lit, with small flames flickering as you approached the grill.", + "E": "You saw that the fire in front of you was out." + }, + "answer_label": "E", + "answer_text": "You saw that the fire in front of you was out.", + "q_id": "egolife_A1_D3_029_3_T1", + "correct_answer": "You saw that the fire in front of you was out.", + "original_options": { + "A": "It was burning.", + "B": "It was smoky.", + "C": "You saw that the fire behind you was still burning as you stood near the barbecue grill.", + "D": "You noticed the fire to your left was lit, with small flames flickering as you approached the grill.", + "E": "You saw that the fire in front of you was out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_3_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was by the barbecue grill just now, what did the fire look like?", + "answer": "You saw that the fire in front of you was out.", + "question_chinese": "刚刚在烧烤架旁时，我看见火是什么状态？", + "answer_chinese": "你看见面前的火是灭的", + "evidence": [ + { + "description": "I see the fire is burning.", + "timestamps": { + "start": "00:06:33", + "end": "00:06:34" + } + } + ], + "question_timestamps": "00:06:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was burning.", + "B": "It was smoky.", + "C": "You saw that the fire behind you was still burning as you stood near the barbecue grill.", + "D": "You noticed the fire to your left was lit, with small flames flickering as you approached the grill.", + "E": "You saw that the fire in front of you was out." + }, + "answer_label": "E", + "answer_text": "You saw that the fire in front of you was out.", + "q_id": "egolife_A1_D3_029_3_T2", + "correct_answer": "You saw that the fire in front of you was out.", + "original_options": { + "A": "It was burning.", + "B": "It was smoky.", + "C": "You saw that the fire behind you was still burning as you stood near the barbecue grill.", + "D": "You noticed the fire to your left was lit, with small flames flickering as you approached the grill.", + "E": "You saw that the fire in front of you was out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_4_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was standing next to the barbecue grill just now, what did the fire look like?", + "answer": "You saw that the fire in front of you was out.", + "question_chinese": "刚刚站在烧烤架旁的时候，我看见火是什么状态？", + "answer_chinese": "你看见面前的火是灭着的", + "evidence": [ + { + "description": "You see the fire on the grill is out.", + "timestamps": { + "start": "00:02:24", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:02:47", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You saw that the fire in front of you was out.", + "B": "You saw that the fire in front of you was burning." + }, + "answer_label": "A", + "answer_text": "You saw that the fire in front of you was out.", + "q_id": "egolife_A1_D3_029_4_T2", + "correct_answer": "You saw that the fire in front of you was out.", + "original_options": { + "A": "You saw that the fire in front of you was out.", + "B": "You saw that the fire in front of you was burning." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_4_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was standing next to the barbecue grill just now, what did the fire look like?", + "answer": "You saw that the fire in front of you was burning.", + "question_chinese": "刚刚站在烧烤架旁的时候，我看见火是什么状态？", + "answer_chinese": "你看见面前的火是烧着的", + "evidence": [ + { + "description": "The fire was burning when I was standing by the barbecue grill just now.", + "timestamps": { + "start": "00:06:33", + "end": "00:06:36" + } + } + ], + "question_timestamps": "00:09:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You saw that the fire in front of you was out.", + "B": "You saw that the fire in front of you was burning." + }, + "answer_label": "B", + "answer_text": "You saw that the fire in front of you was burning.", + "q_id": "egolife_A1_D3_029_4_T1", + "correct_answer": "You saw that the fire in front of you was burning.", + "original_options": { + "A": "You saw that the fire in front of you was out.", + "B": "You saw that the fire in front of you was burning." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_10_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just wash at the sink again?", + "answer": "You washed the knife in the sink in front of you.", + "question_chinese": "我刚又在洗水池洗了什么？", + "answer_chinese": "你在面前的洗碗池洗了刀", + "evidence": [ + { + "description": "I washed the knife", + "timestamps": { + "start": "00:10:06", + "end": "00:10:18" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed the knife behind you.", + "B": "You washed the knife in the sink in front of you.", + "C": "You washed the knife to your left.", + "D": "You washed the knife in the sink to your right, rinsing it again under running water.", + "E": "You washed the knife in the sink behind you, scrubbing it briefly before setting it down." + }, + "answer_label": "B", + "answer_text": "You washed the knife in the sink in front of you.", + "q_id": "egolife_A1_D3_029_10_T1", + "correct_answer": "You washed the knife in the sink in front of you.", + "original_options": { + "A": "You washed the knife behind you.", + "B": "You washed the knife in the sink in front of you.", + "C": "You washed the knife to your left.", + "D": "You washed the knife in the sink to your right, rinsing it again under running water.", + "E": "You washed the knife in the sink behind you, scrubbing it briefly before setting it down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_10", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just wash in the sink?", + "answer": "You washed the knife and mangoes.", + "question_chinese": "我刚又在洗水池洗了什么？", + "answer_chinese": "洗了刀和芒果", + "evidence": [ + { + "description": "I washed the knife", + "timestamps": { + "start": "00:10:06", + "end": "00:10:18" + } + }, + { + "description": "I washed the mangoes.", + "timestamps": { + "start": "00:10:25", + "end": "00:10:28" + } + } + ], + "question_timestamps": "00:11:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You washed the knife and oranges.", + "B": "You washed the fork and mangoes.", + "C": "You washed the knife and peaches together, rinsing both items under the sink water.", + "D": "You washed the knife and mangoes.", + "E": "You washed the knife and apples in the sink, cleaning them before putting them aside." + }, + "answer_label": "D", + "answer_text": "You washed the knife and mangoes.", + "q_id": "egolife_A1_D3_029_10", + "correct_answer": "You washed the knife and mangoes.", + "original_options": { + "A": "You washed the knife and oranges.", + "B": "You washed the fork and mangoes.", + "C": "You washed the knife and peaches together, rinsing both items under the sink water.", + "D": "You washed the knife and mangoes.", + "E": "You washed the knife and apples in the sink, cleaning them before putting them aside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_10_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just wash at the sink again?", + "answer": "You washed a knife and a mango in the sink in front of you.", + "question_chinese": "我刚又在洗水池洗了什么？", + "answer_chinese": "你在面前的洗碗池洗了刀和芒果", + "evidence": [ + { + "description": "I washed the knife", + "timestamps": { + "start": "00:10:06", + "end": "00:10:18" + } + }, + { + "description": "I washed mangoes", + "timestamps": { + "start": "00:10:25", + "end": "00:10:26" + } + } + ], + "question_timestamps": "00:11:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed a knife and a mango in the sink in front of you.", + "B": "You washed a spoon and a mango.", + "C": "You washed a knife and an apple.", + "D": "You washed a knife and a mango in the sink behind you, not the one directly in front.", + "E": "You washed a knife along with a mango, but you did it at the sink to your left." + }, + "answer_label": "A", + "answer_text": "You washed a knife and a mango in the sink in front of you.", + "q_id": "egolife_A1_D3_029_10_T2", + "correct_answer": "You washed a knife and a mango in the sink in front of you.", + "original_options": { + "A": "You washed a knife and a mango in the sink in front of you.", + "B": "You washed a spoon and a mango.", + "C": "You washed a knife and an apple.", + "D": "You washed a knife and a mango in the sink behind you, not the one directly in front.", + "E": "You washed a knife along with a mango, but you did it at the sink to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_5_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just wash at the sink?", + "answer": "You washed your hands and bowls in the sink in front of you.", + "question_chinese": "我刚都在洗水池洗了什么？", + "answer_chinese": "你在面前的洗碗池洗了手和碗", + "evidence": [ + { + "description": "I washed the bowls", + "timestamps": { + "start": "00:06:16", + "end": "00:06:19" + } + }, + { + "description": "I washed my hands", + "timestamps": { + "start": "00:05:46", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed only your hands.", + "B": "You washed only the bowls.", + "C": "You washed your hands and bowls in the sink in front of you.", + "D": "You washed your hands and bowls in the sink behind you, not the one in front.", + "E": "You washed your hands along with the bowls, but at the sink to your right." + }, + "answer_label": "C", + "answer_text": "You washed your hands and bowls in the sink in front of you.", + "q_id": "egolife_A1_D3_029_5_T1", + "correct_answer": "You washed your hands and bowls in the sink in front of you.", + "original_options": { + "A": "You washed only your hands.", + "B": "You washed only the bowls.", + "C": "You washed your hands and bowls in the sink in front of you.", + "D": "You washed your hands and bowls in the sink behind you, not the one in front.", + "E": "You washed your hands along with the bowls, but at the sink to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_5_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just wash at the sink?", + "answer": "You washed your hands and bowls in the sink in front of you.", + "question_chinese": "我刚都在洗水池洗了什么？", + "answer_chinese": "你在面前的洗碗池洗了手和碗", + "evidence": [ + { + "description": "I washed the bowls", + "timestamps": { + "start": "00:06:16", + "end": "00:06:19" + } + }, + { + "description": "I washed my hands", + "timestamps": { + "start": "00:05:46", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:07:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed only your hands.", + "B": "You washed only the bowls.", + "C": "You washed your hands and bowls in the sink behind you, not the one in front.", + "D": "You washed your hands along with the bowls, but at the sink to your left.", + "E": "You washed your hands and bowls in the sink in front of you." + }, + "answer_label": "E", + "answer_text": "You washed your hands and bowls in the sink in front of you.", + "q_id": "egolife_A1_D3_029_5_T2", + "correct_answer": "You washed your hands and bowls in the sink in front of you.", + "original_options": { + "A": "You washed only your hands.", + "B": "You washed only the bowls.", + "C": "You washed your hands and bowls in the sink behind you, not the one in front.", + "D": "You washed your hands along with the bowls, but at the sink to your left.", + "E": "You washed your hands and bowls in the sink in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_5_T3", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just wash at the sink?", + "answer": "You washed a knife and a mango in the sink in front of you.", + "question_chinese": "我刚都在洗水池洗了什么？", + "answer_chinese": "你在面前的洗碗池洗了刀和芒果", + "evidence": [ + { + "description": "I washed the knife", + "timestamps": { + "start": "00:10:06", + "end": "00:10:18" + } + }, + { + "description": "I washed the mangoes.", + "timestamps": { + "start": "00:10:25", + "end": "00:10:28" + } + } + ], + "question_timestamps": "00:11:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed a spoon and a mango.", + "B": "You washed a knife and an apple.", + "C": "You washed a knife and a mango in the sink in front of you.", + "D": "You washed a knife and a mango in the sink to your right, not the one ahead.", + "E": "You washed a knife along with a mango, but you did it at the sink behind you." + }, + "answer_label": "C", + "answer_text": "You washed a knife and a mango in the sink in front of you.", + "q_id": "egolife_A1_D3_029_5_T3", + "correct_answer": "You washed a knife and a mango in the sink in front of you.", + "original_options": { + "A": "You washed a spoon and a mango.", + "B": "You washed a knife and an apple.", + "C": "You washed a knife and a mango in the sink in front of you.", + "D": "You washed a knife and a mango in the sink to your right, not the one ahead.", + "E": "You washed a knife along with a mango, but you did it at the sink behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_9_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What does the fire in front of me look like?", + "answer": "You saw that the fire in front of you was burning.", + "question_chinese": "我看见面前的火是什么状态？", + "answer_chinese": "你看见面前的火是烧着的", + "evidence": [ + { + "description": "The fire was burning when I was standing by the barbecue grill just now.", + "timestamps": { + "start": "00:06:33", + "end": "00:06:34" + } + } + ], + "question_timestamps": "00:06:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It looked like it was out.", + "B": "It looked like it was smoking.", + "C": "You saw the fire in front of you was completely out, with no flames visible anymore.", + "D": "You saw that the fire in front of you was burning.", + "E": "You saw the fire ahead of you flickering weakly, giving off smoke rather than burning strongly." + }, + "answer_label": "D", + "answer_text": "You saw that the fire in front of you was burning.", + "q_id": "egolife_A1_D3_029_9_T1", + "correct_answer": "You saw that the fire in front of you was burning.", + "original_options": { + "A": "It looked like it was out.", + "B": "It looked like it was smoking.", + "C": "You saw the fire in front of you was completely out, with no flames visible anymore.", + "D": "You saw that the fire in front of you was burning.", + "E": "You saw the fire ahead of you flickering weakly, giving off smoke rather than burning strongly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_9_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What does the fire in front of me look like?", + "answer": "You couldn't see the fire now.", + "question_chinese": "我看见面前的火是什么状态？", + "answer_chinese": "你现在看不到火", + "evidence": [ + { + "description": "The fire was burning when I was standing by the barbecue grill just now.", + "timestamps": { + "start": "00:06:33", + "end": "00:06:34" + } + } + ], + "question_timestamps": "00:07:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was burning in front of you.", + "B": "It was out in front of you.", + "C": "You couldn't see the fire now.", + "D": "You could still see the fire in front of you burning clearly, with flames visible.", + "E": "You could see the fire in front of you, but it had gone out and wasn’t burning anymore." + }, + "answer_label": "C", + "answer_text": "You couldn't see the fire now.", + "q_id": "egolife_A1_D3_029_9_T2", + "correct_answer": "You couldn't see the fire now.", + "original_options": { + "A": "It was burning in front of you.", + "B": "It was out in front of you.", + "C": "You couldn't see the fire now.", + "D": "You could still see the fire in front of you burning clearly, with flames visible.", + "E": "You could see the fire in front of you, but it had gone out and wasn’t burning anymore." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_6_T2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the white trash can closer to me, or is the table closer to me?", + "answer": "The white trash can.", + "question_chinese": "是白色垃圾桶离我更近还是桌子离我更近？", + "answer_chinese": "白色垃圾桶", + "evidence": [ + { + "description": "The white trash can and the table were both in front of me, with the trash can closer to me\n.", + "timestamps": { + "start": "00:07:21", + "end": "00:07:22" + } + } + ], + "question_timestamps": "00:07:42", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The table.", + "B": "The white trash can." + }, + "answer_label": "B", + "answer_text": "The white trash can.", + "q_id": "egolife_A1_D3_029_6_T2", + "correct_answer": "The white trash can.", + "original_options": { + "A": "The table.", + "B": "The white trash can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_6_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the white trash can closer to me, or is the table closer to me?", + "answer": "The table.", + "question_chinese": "是白色垃圾桶离我更近还是桌子离我更近？", + "answer_chinese": "桌子", + "evidence": [ + { + "description": "The white trash can and the table were both in front of me but I walked closer to the table.\n.", + "timestamps": { + "start": "00:07:40", + "end": "00:07:41" + } + } + ], + "question_timestamps": "00:07:43", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The table.", + "B": "The white trash can." + }, + "answer_label": "A", + "answer_text": "The table.", + "q_id": "egolife_A1_D3_029_6_T1", + "correct_answer": "The table.", + "original_options": { + "A": "The table.", + "B": "The white trash can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_7_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the blue trash can closer to me, or is the dining table closer to me?", + "answer": "The blue trash can was closer.", + "question_chinese": "是蓝色垃圾桶离我更近还是餐桌离我更近？", + "answer_chinese": "蓝色垃圾桶更近", + "evidence": [ + { + "description": "The blue trash can is between the dining table and the countertop.", + "timestamps": { + "start": "00:05:40", + "end": "00:05:43" + } + }, + { + "description": "I walked up to the blue trash can.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:11" + } + } + ], + "question_timestamps": "00:06:52", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The dining table was closer.", + "B": "The blue trash can was closer." + }, + "answer_label": "B", + "answer_text": "The blue trash can was closer.", + "q_id": "egolife_A1_D3_029_7_T1", + "correct_answer": "The blue trash can was closer.", + "original_options": { + "A": "The dining table was closer.", + "B": "The blue trash can was closer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_7_T2", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the blue trash can closer to me, or is the dining table closer to me?", + "answer": "The dining table was closer.", + "question_chinese": "是蓝色垃圾桶离我更近还是餐桌离我更近？", + "answer_chinese": "餐桌更近", + "evidence": [ + { + "description": "The blue trash can is between the dining table and the countertop.", + "timestamps": { + "start": "00:05:40", + "end": "00:05:43" + } + }, + { + "description": "I walked closer to the table.", + "timestamps": { + "start": "00:07:22", + "end": "00:07:23" + } + } + ], + "question_timestamps": "00:07:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The dining table was closer.", + "B": "The blue trash can was closer." + }, + "answer_label": "A", + "answer_text": "The dining table was closer.", + "q_id": "egolife_A1_D3_029_7_T2", + "correct_answer": "The dining table was closer.", + "original_options": { + "A": "The dining table was closer.", + "B": "The blue trash can was closer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_2_T2", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the blue trash can in front of me or the table?", + "answer": "The blue trash can in front of you is much closer to you.", + "question_chinese": "是面前的蓝色垃圾桶离我更近还是桌子离我更近？", + "answer_chinese": "面前的蓝色垃圾桶离你更近更近", + "evidence": [ + { + "description": "The blue trash can is between the dining table and the countertop.", + "timestamps": { + "start": "00:05:56", + "end": "00:05:57" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The blue trash can in front of you is much closer to you.", + "B": "The table in front of you is much closer to you." + }, + "answer_label": "A", + "answer_text": "The blue trash can in front of you is much closer to you.", + "q_id": "egolife_A1_D3_029_2_T2", + "correct_answer": "The blue trash can in front of you is much closer to you.", + "original_options": { + "A": "The blue trash can in front of you is much closer to you.", + "B": "The table in front of you is much closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_2_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the blue trash can in front of me or the table?", + "answer": "The table in front of you is much closer to you.", + "question_chinese": "是面前的蓝色垃圾桶离我更近还是桌子离我更近？", + "answer_chinese": "面前的桌子离你更近更近", + "evidence": [ + { + "description": "The table was right in front of me, within arm's reach.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:01" + } + }, + { + "description": "The trash can was behind me, at a slight distance.", + "timestamps": { + "start": "00:05:03", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:07:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The blue trash can in front of you is much closer to you.", + "B": "The table in front of you is much closer to you." + }, + "answer_label": "B", + "answer_text": "The table in front of you is much closer to you.", + "q_id": "egolife_A1_D3_029_2_T1", + "correct_answer": "The table in front of you is much closer to you.", + "original_options": { + "A": "The blue trash can in front of you is much closer to you.", + "B": "The table in front of you is much closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_1_T2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me: the trash can in front of me or the table?", + "answer": "The table in front of you is closer to you.", + "question_chinese": "现在是面前的垃圾桶离我更近还是桌子离我更近？", + "answer_chinese": "面前桌子离你更近", + "evidence": [ + { + "description": "The white trash can and the table were both in front of me，but I walked closer to the table.", + "timestamps": { + "start": "00:07:21", + "end": "00:07:22" + } + } + ], + "question_timestamps": "00:07:24", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The trash can in front of you is closer to you.", + "B": "The table in front of you is closer to you." + }, + "answer_label": "B", + "answer_text": "The table in front of you is closer to you.", + "q_id": "egolife_A1_D3_029_1_T2", + "correct_answer": "The table in front of you is closer to you.", + "original_options": { + "A": "The trash can in front of you is closer to you.", + "B": "The table in front of you is closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D3_029", + "video_length": "00:11:53", + "qid": "egolife_A1_D3_029_1_T1", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, which is closer to me: the trash can in front of me or the table?", + "answer": "The trash can in front of you is closer to you.", + "question_chinese": "现在是面前的垃圾桶离我更近还是桌子离我更近？", + "answer_chinese": "面前垃圾桶离你更近", + "evidence": [ + { + "description": "The white trash can and the table were both in front of me, with the trash can closer to me\n.", + "timestamps": { + "start": "00:07:21", + "end": "00:07:22" + } + } + ], + "question_timestamps": "00:08:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The trash can in front of you is closer to you.", + "B": "The table in front of you is closer to you." + }, + "answer_label": "A", + "answer_text": "The trash can in front of you is closer to you.", + "q_id": "egolife_A1_D3_029_1_T1", + "correct_answer": "The trash can in front of you is closer to you.", + "original_options": { + "A": "The trash can in front of you is closer to you.", + "B": "The table in front of you is closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + } +] diff --git a/QAs/egolife_A1_D3_030-mcq.json b/QAs/egolife_A1_D3_030-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2f5f5b81b7d856b7fa483fb86b936451bf64ef80 --- /dev/null +++ b/QAs/egolife_A1_D3_030-mcq.json @@ -0,0 +1,1286 @@ +[ + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_1_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Did I end up placing the cup to my right, or to my left?", + "answer": "You put the cup on the table on the right.", + "question_chinese": "我到底把杯子放在我右边了，还是放在左边了？", + "answer_chinese": "你把杯子放在了右边的桌子上。", + "evidence": [ + { + "description": "At this moment, the cup has remained on your right side.", + "timestamps": { + "start": "00:06:30", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:07:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You put the cup on the table on the left.", + "B": "You put the cup on the table on the right." + }, + "answer_label": "B", + "answer_text": "You put the cup on the table on the right.", + "q_id": "egolife_A1_D3_030_1_T1", + "correct_answer": "You put the cup on the table on the right.", + "original_options": { + "A": "You put the cup on the table on the left.", + "B": "You put the cup on the table on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_1_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Did I put the cup on my right side or on my left side?", + "answer": "You put the cup on the table on the right.", + "question_chinese": "我到底把杯子放在我右边了，还是放在左边了？", + "answer_chinese": "你把杯子放在了右边的桌子上。", + "evidence": [ + { + "description": "At this moment, the cup has remained on your right side.", + "timestamps": { + "start": "00:06:30", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:07:34", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You put the cup on the table on the left.", + "B": "You put the cup on the table on the right." + }, + "answer_label": "B", + "answer_text": "You put the cup on the table on the right.", + "q_id": "egolife_A1_D3_030_1_T2", + "correct_answer": "You put the cup on the table on the right.", + "original_options": { + "A": "You put the cup on the table on the left.", + "B": "You put the cup on the table on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_2_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Did I end up placing the cup to my right, or to my left?", + "answer": "You put the cup on the right side of the table.", + "question_chinese": "我到底把杯子放在我右边了，还是放在左边了？", + "answer_chinese": "你把杯子放在了桌子的右方", + "evidence": [ + { + "description": "At this moment, the cup has been on the table on your right hand side.", + "timestamps": { + "start": "00:06:54", + "end": "00:06:58" + } + } + ], + "question_timestamps": "00:09:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You put the cup on the right side of the table.", + "B": "You put the cup on the left side of the table." + }, + "answer_label": "A", + "answer_text": "You put the cup on the right side of the table.", + "q_id": "egolife_A1_D3_030_2_T1", + "correct_answer": "You put the cup on the right side of the table.", + "original_options": { + "A": "You put the cup on the right side of the table.", + "B": "You put the cup on the left side of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_2_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Did I put the cup on my right side or my left side?", + "answer": "You put the cup on the right side of the table.", + "question_chinese": "我到底把杯子放在我右边了，还是放在左边了？", + "answer_chinese": "你把杯子放在了桌子的右方", + "evidence": [ + { + "description": "At this moment, the cup has been on the table on your right hand side.", + "timestamps": { + "start": "00:06:54", + "end": "00:06:58" + } + } + ], + "question_timestamps": "00:09:04", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You put the cup on the right side of the table.", + "B": "You put the cup on the left side of the table." + }, + "answer_label": "A", + "answer_text": "You put the cup on the right side of the table.", + "q_id": "egolife_A1_D3_030_2_T2", + "correct_answer": "You put the cup on the right side of the table.", + "original_options": { + "A": "You put the cup on the right side of the table.", + "B": "You put the cup on the left side of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030-mcq_fixed_10", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I eat mangoes while I was by the sink so far?", + "answer": "Five.", + "question_chinese": "我在水槽旁边时，到目前为止吃了几次芒果？", + "answer_chinese": "一次", + "evidence": [ + { + "description": "I put the mango into my mouth at this moment", + "timestamps": { + "start": "00:00:14", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:54", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four times.", + "B": "Six times.", + "C": "You ate mangoes by the sink four times in total so far.", + "D": "You ate mangoes by the sink six times altogether up to this point.", + "E": "Five." + }, + "answer_label": "E", + "answer_text": "Five.", + "q_id": "egolife_A1_D3_030-mcq_fixed_10", + "correct_answer": "Five.", + "original_options": { + "A": "Four times.", + "B": "Six times.", + "C": "You ate mangoes by the sink four times in total so far.", + "D": "You ate mangoes by the sink six times altogether up to this point.", + "E": "Five." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030-mcq_fixed_11", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I eat mangoes while I was by the sink so far?", + "answer": "Four.", + "question_chinese": "我在水槽旁边时，到目前为止吃了几次芒果？", + "answer_chinese": "两次", + "evidence": [ + { + "description": "I put the mango into my mouth at this moment.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:19" + } + }, + { + "description": "I'm putting the mango into my mouth again to eat it.", + "timestamps": { + "start": "00:09:57", + "end": "00:10:25" + } + } + ], + "question_timestamps": "00:11:08", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three times.", + "C": "Five times.", + "D": "You ate mangoes by the sink three times in total so far.", + "E": "You ate mangoes by the sink five times altogether up to this point." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_030-mcq_fixed_11", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Three times.", + "C": "Five times.", + "D": "You ate mangoes by the sink three times in total so far.", + "E": "You ate mangoes by the sink five times altogether up to this point." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_7", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Am I sitting on a chair right now?", + "answer": "No, you've been standing here all this time, and the chair is on your right.", + "question_chinese": "我此时坐在椅子上了吗？", + "answer_chinese": "没有，你此时一直在站着，椅子在你右边。", + "evidence": [ + { + "description": "At this moment, you have been standing all along.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:07" + } + }, + { + "description": "The girl in green clothes is standing next to the chair, which is unoccupied.", + "timestamps": { + "start": "00:06:08", + "end": "00:06:10" + } + } + ], + "question_timestamps": "00:06:18", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you've been standing here all this time, and the chair is on your right.", + "B": "Yes, you're sitting on a chair right now, with the chair directly under you." + }, + "answer_label": "A", + "answer_text": "No, you've been standing here all this time, and the chair is on your right.", + "q_id": "egolife_A1_D3_030_7", + "correct_answer": "No, you've been standing here all this time, and the chair is on your right.", + "original_options": { + "A": "No, you've been standing here all this time, and the chair is on your right.", + "B": "Yes, you're sitting on a chair right now, with the chair directly under you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_8", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Am I sitting on a chair right now?", + "answer": "Yes, you are sitting on the chair.", + "question_chinese": "我此时坐在椅子上了吗？", + "answer_chinese": "是的，你坐在椅子上了", + "evidence": [ + { + "description": "At this moment, you have been standing there arranging items.", + "timestamps": { + "start": "00:06:09", + "end": "00:06:10" + } + }, + { + "description": "After the girl left, you picked up the chair on the right and placed it beside you before sitting down.", + "timestamps": { + "start": "00:06:18", + "end": "00:06:21" + } + } + ], + "question_timestamps": "00:06:37", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you are not sitting on the chair.", + "B": "Yes, you are sitting on the chair." + }, + "answer_label": "B", + "answer_text": "Yes, you are sitting on the chair.", + "q_id": "egolife_A1_D3_030_8", + "correct_answer": "Yes, you are sitting on the chair.", + "original_options": { + "A": "No, you are not sitting on the chair.", + "B": "Yes, you are sitting on the chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030-mcq_fixed_8", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the bowl of batter take to get to this position?", + "answer": "The bowl was taken by you from the center of the table's edge into your hands, hovering over a corner of the table.", + "question_chinese": "放有面糊的碗到现在这个位置经历了什么途径？", + "answer_chinese": "碗从桌子边缘的正中间被你拿到了手中，悬在桌子一角。", + "evidence": [ + { + "description": "At this moment, the bowl containing batter on the table is picked up by you from the middle of the table's edge.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:49" + } + }, + { + "description": "During this time, you have been holding the bowl and pouring batter into a corner of the table.", + "timestamps": { + "start": "00:00:52", + "end": "00:00:53" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You slid it along the tabletop.", + "B": "You moved it to the table center.", + "C": "You picked the bowl up from a table corner and carried it in your hands toward the table’s center edge.", + "D": "You kept the bowl on the table, dragging it from the corner toward the center edge without lifting it.", + "E": "The bowl was taken by you from the center of the table's edge into your hands, hovering over a corner of the table." + }, + "answer_label": "E", + "answer_text": "The bowl was taken by you from the center of the table's edge into your hands, hovering over a corner of the table.", + "q_id": "egolife_A1_D3_030-mcq_fixed_8", + "correct_answer": "The bowl was taken by you from the center of the table's edge into your hands, hovering over a corner of the table.", + "original_options": { + "A": "You slid it along the tabletop.", + "B": "You moved it to the table center.", + "C": "You picked the bowl up from a table corner and carried it in your hands toward the table’s center edge.", + "D": "You kept the bowl on the table, dragging it from the corner toward the center edge without lifting it.", + "E": "The bowl was taken by you from the center of the table's edge into your hands, hovering over a corner of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030-mcq_fixed_9", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the bowl of batter take to get to this position?", + "answer": "The bowl was taken by you from the center of the table's edge into your hands, hovering over a corner of the table. Finally, you placed the bowl in front of the plate with the bread.", + "question_chinese": "放有面糊的碗到现在这个位置经历了什么途径？", + "answer_chinese": "碗从桌子边缘的正中间被你拿到了手中，悬在桌子一角。最后你将碗放到了放面包的盘子的前面。", + "evidence": [ + { + "description": "At this moment, the bowl containing batter on the table is picked up by you from the middle of the table's edge.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:49" + } + }, + { + "description": "During this time, you have been holding the bowl and pouring batter into a corner of the table.", + "timestamps": { + "start": "00:00:52", + "end": "00:00:53" + } + }, + { + "description": "At this moment, the bowl was handed over to the girl wearing green, who placed it in front of the plate containing the bread dough.", + "timestamps": { + "start": "00:05:31", + "end": "00:05:34" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You slid it to the water glass.", + "B": "You set it beside the napkins.", + "C": "You lifted the bowl from the far corner, moved it toward your side, then placed it next to the water glass.", + "D": "You picked up the bowl at the center edge and set it down behind the bread plate after hovering over the corner.", + "E": "The bowl was taken by you from the center of the table's edge into your hands, hovering over a corner of the table. Finally, you placed the bowl in front of the plate with the bread." + }, + "answer_label": "E", + "answer_text": "The bowl was taken by you from the center of the table's edge into your hands, hovering over a corner of the table. Finally, you placed the bowl in front of the plate with the bread.", + "q_id": "egolife_A1_D3_030-mcq_fixed_9", + "correct_answer": "The bowl was taken by you from the center of the table's edge into your hands, hovering over a corner of the table. Finally, you placed the bowl in front of the plate with the bread.", + "original_options": { + "A": "You slid it to the water glass.", + "B": "You set it beside the napkins.", + "C": "You lifted the bowl from the far corner, moved it toward your side, then placed it next to the water glass.", + "D": "You picked up the bowl at the center edge and set it down behind the bread plate after hovering over the corner.", + "E": "The bowl was taken by you from the center of the table's edge into your hands, hovering over a corner of the table. Finally, you placed the bowl in front of the plate with the bread." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 21, + 22 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_8_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many cake layers have I poured?", + "answer": "Four.", + "question_chinese": "此时，我已经倒了几个蛋糕胚？", + "answer_chinese": "你此时刚好倒完了六个蛋糕胚。", + "evidence": [ + { + "description": "At this moment, not a single cake batter had been poured correctly on the table.", + "timestamps": { + "start": "00:00:51", + "end": "00:00:53" + } + }, + { + "description": "At this moment, you have just finished pouring the cake batter.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:24" + } + }, + { + "description": "At this moment, you have already successfully prepared four cake batter mixtures in a row.", + "timestamps": { + "start": "00:01:56", + "end": "00:01:59" + } + }, + { + "description": "At this moment, you have already successfully prepared six cake batter mixtures consecutively.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:17" + } + } + ], + "question_timestamps": "00:02:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "You have poured only three cake layers so far, and have not poured the fourth yet.", + "E": "You have poured five cake layers already, meaning you added one more layer after the fourth." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_030_8_T1", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "You have poured only three cake layers so far, and have not poured the fourth yet.", + "E": "You have poured five cake layers already, meaning you added one more layer after the fourth." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_8_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this point, how many cake bases have I already poured？", + "answer": "You have just finished pouring six cake batter molds at this moment.", + "question_chinese": "此时，我已经倒了几个蛋糕胚？", + "answer_chinese": "你此时刚好倒完了六个蛋糕胚。", + "evidence": [ + { + "description": "At this moment, not a single cake batter had been poured correctly on the table.", + "timestamps": { + "start": "00:00:51", + "end": "00:00:53" + } + }, + { + "description": "At this moment, you have just finished pouring the cake batter.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:24" + } + }, + { + "description": "At this moment, you have already successfully prepared four cake batter mixtures in a row.", + "timestamps": { + "start": "00:01:56", + "end": "00:01:59" + } + }, + { + "description": "At this moment, you have already successfully prepared six cake batter mixtures consecutively.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:17" + } + } + ], + "question_timestamps": "00:02:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have poured five molds.", + "B": "You have just finished pouring six cake batter molds at this moment.", + "C": "You have poured seven molds.", + "D": "You have just finished pouring five cake batter molds at this moment.", + "E": "You have just finished pouring seven cake batter molds at this moment." + }, + "answer_label": "B", + "answer_text": "You have just finished pouring six cake batter molds at this moment.", + "q_id": "egolife_A1_D3_030_8_T2", + "correct_answer": "You have just finished pouring six cake batter molds at this moment.", + "original_options": { + "A": "You have poured five molds.", + "B": "You have just finished pouring six cake batter molds at this moment.", + "C": "You have poured seven molds.", + "D": "You have just finished pouring five cake batter molds at this moment.", + "E": "You have just finished pouring seven cake batter molds at this moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_4_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many cake layers have I poured?", + "answer": "You have just finished pouring six cake batter molds at this moment.", + "question_chinese": "此时，我已经倒了几个蛋糕胚？", + "answer_chinese": "你此时刚好倒完了六个蛋糕胚。", + "evidence": [ + { + "description": "At this moment, not a single cake batter had been poured correctly on the table.", + "timestamps": { + "start": "00:00:51", + "end": "00:00:53" + } + }, + { + "description": "At this moment, you have just finished pouring the cake batter.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:24" + } + }, + { + "description": "At this moment, you have already successfully prepared four cake batter mixtures in a row.", + "timestamps": { + "start": "00:01:56", + "end": "00:01:59" + } + }, + { + "description": "At this moment, you have already successfully prepared six cake batter mixtures consecutively.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:17" + } + } + ], + "question_timestamps": "00:02:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have poured five molds.", + "B": "You have poured seven molds.", + "C": "You have just finished pouring six cake batter molds at this moment.", + "D": "You have just finished pouring five cake batter molds at this moment.", + "E": "You have just finished pouring seven cake batter molds at this moment." + }, + "answer_label": "C", + "answer_text": "You have just finished pouring six cake batter molds at this moment.", + "q_id": "egolife_A1_D3_030_4_T1", + "correct_answer": "You have just finished pouring six cake batter molds at this moment.", + "original_options": { + "A": "You have poured five molds.", + "B": "You have poured seven molds.", + "C": "You have just finished pouring six cake batter molds at this moment.", + "D": "You have just finished pouring five cake batter molds at this moment.", + "E": "You have just finished pouring seven cake batter molds at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_4_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this point, how many cake layers have I already made?", + "answer": "You have just finished pouring six cake batter molds at this moment.", + "question_chinese": "此时，我已经倒了几个蛋糕胚？", + "answer_chinese": "你此时刚好倒完了六个蛋糕胚。", + "evidence": [ + { + "description": "At this moment, not a single cake batter had been poured correctly on the table.", + "timestamps": { + "start": "00:00:51", + "end": "00:00:53" + } + }, + { + "description": "At this moment, you have just finished pouring the cake batter.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:24" + } + }, + { + "description": "At this moment, you have already successfully prepared four cake batter mixtures in a row.", + "timestamps": { + "start": "00:01:56", + "end": "00:01:59" + } + }, + { + "description": "At this moment, you have already successfully prepared six cake batter mixtures consecutively.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:17" + } + } + ], + "question_timestamps": "00:02:29", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have poured five molds.", + "B": "You have poured seven molds.", + "C": "You have just finished pouring six cake batter molds at this moment.", + "D": "You have just finished pouring five cake batter molds at this moment.", + "E": "You have just finished pouring seven cake batter molds at this moment." + }, + "answer_label": "C", + "answer_text": "You have just finished pouring six cake batter molds at this moment.", + "q_id": "egolife_A1_D3_030_4_T2", + "correct_answer": "You have just finished pouring six cake batter molds at this moment.", + "original_options": { + "A": "You have poured five molds.", + "B": "You have poured seven molds.", + "C": "You have just finished pouring six cake batter molds at this moment.", + "D": "You have just finished pouring five cake batter molds at this moment.", + "E": "You have just finished pouring seven cake batter molds at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_9_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many cake layers have I poured?", + "answer": "Six.", + "question_chinese": "此时，我已经倒了几个蛋糕胚？", + "answer_chinese": "你此时刚好倒完了十二个蛋糕胚。", + "evidence": [ + { + "description": "At this moment, not a single cake base had been prepared on the table.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:06" + } + }, + { + "description": "You have just finished pouring six cake batter molds at this moment.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:17" + } + }, + { + "description": "You have just finished pouring twelve cake batter molds at this moment.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:03" + } + } + ], + "question_timestamps": "00:03:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "You have poured only five cake layers so far, stopping one layer short of six.", + "D": "You have already poured seven cake layers by this point, which is one more than six.", + "E": "Six." + }, + "answer_label": "E", + "answer_text": "Six.", + "q_id": "egolife_A1_D3_030_9_T1", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "You have poured only five cake layers so far, stopping one layer short of six.", + "D": "You have already poured seven cake layers by this point, which is one more than six.", + "E": "Six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_9_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this point, how many cake bases have I already made?", + "answer": "You have just finished pouring twelve cake batter molds at this moment.", + "question_chinese": "此时，我已经倒了几个蛋糕胚？", + "answer_chinese": "你此时刚好倒完了十二个蛋糕胚。", + "evidence": [ + { + "description": "At this moment, not a single cake base had been prepared on the table.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:06" + } + }, + { + "description": "You have just finished pouring six cake batter molds at this moment.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:17" + } + }, + { + "description": "You have just finished pouring twelve cake batter molds at this moment.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:03" + } + } + ], + "question_timestamps": "00:03:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have poured eleven molds.", + "B": "You have poured thirteen molds.", + "C": "You have just finished pouring twelve cake batter molds at this moment.", + "D": "You have just finished pouring eleven cake batter molds at this moment.", + "E": "You have just finished pouring thirteen cake batter molds at this moment." + }, + "answer_label": "C", + "answer_text": "You have just finished pouring twelve cake batter molds at this moment.", + "q_id": "egolife_A1_D3_030_9_T2", + "correct_answer": "You have just finished pouring twelve cake batter molds at this moment.", + "original_options": { + "A": "You have poured eleven molds.", + "B": "You have poured thirteen molds.", + "C": "You have just finished pouring twelve cake batter molds at this moment.", + "D": "You have just finished pouring eleven cake batter molds at this moment.", + "E": "You have just finished pouring thirteen cake batter molds at this moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_5_T1", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many cake layers have I poured?", + "answer": "You have just finished pouring twelve cake batter molds at this moment.", + "question_chinese": "此时，我已经倒了几个蛋糕胚？", + "answer_chinese": "你此时刚好倒完了十二个蛋糕胚。", + "evidence": [ + { + "description": "At this moment, not a single cake base had been prepared on the table.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:06" + } + }, + { + "description": "You have just finished pouring six cake batter molds at this moment.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:17" + } + }, + { + "description": "You have just finished pouring twelve cake batter molds at this moment.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:03" + } + } + ], + "question_timestamps": "00:03:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have poured eleven molds.", + "B": "You have poured thirteen molds.", + "C": "You have just finished pouring twelve cake batter molds at this moment.", + "D": "You have just finished pouring eleven cake batter molds at this moment.", + "E": "You have just finished pouring thirteen cake batter molds at this moment." + }, + "answer_label": "C", + "answer_text": "You have just finished pouring twelve cake batter molds at this moment.", + "q_id": "egolife_A1_D3_030_5_T1", + "correct_answer": "You have just finished pouring twelve cake batter molds at this moment.", + "original_options": { + "A": "You have poured eleven molds.", + "B": "You have poured thirteen molds.", + "C": "You have just finished pouring twelve cake batter molds at this moment.", + "D": "You have just finished pouring eleven cake batter molds at this moment.", + "E": "You have just finished pouring thirteen cake batter molds at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_5_T2", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this point, how many cake layers have I already baked?", + "answer": "You have just finished pouring twelve cake batter molds at this moment.", + "question_chinese": "此时，我已经倒了几个蛋糕胚？", + "answer_chinese": "你此时刚好倒完了十二个蛋糕胚。", + "evidence": [ + { + "description": "At this moment, not a single cake base had been prepared on the table.", + "timestamps": { + "start": "00:01:05", + "end": "00:01:06" + } + }, + { + "description": "You have just finished pouring six cake batter molds at this moment.", + "timestamps": { + "start": "00:02:16", + "end": "00:02:17" + } + }, + { + "description": "You have just finished pouring twelve cake batter molds at this moment.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:03" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have poured eleven molds.", + "B": "You have poured thirteen molds.", + "C": "You have just finished pouring twelve cake batter molds at this moment.", + "D": "You have just finished pouring eleven cake batter molds at this moment.", + "E": "You have just finished pouring thirteen cake batter molds at this moment." + }, + "answer_label": "C", + "answer_text": "You have just finished pouring twelve cake batter molds at this moment.", + "q_id": "egolife_A1_D3_030_5_T2", + "correct_answer": "You have just finished pouring twelve cake batter molds at this moment.", + "original_options": { + "A": "You have poured eleven molds.", + "B": "You have poured thirteen molds.", + "C": "You have just finished pouring twelve cake batter molds at this moment.", + "D": "You have just finished pouring eleven cake batter molds at this moment.", + "E": "You have just finished pouring thirteen cake batter molds at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_6_T1", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, is the girl in green to my right?", + "answer": "No, she is on your left.", + "question_chinese": "此时，绿衣服的女生在我的右边吗？", + "answer_chinese": "不是，她在你的左边", + "evidence": [ + { + "description": "At this moment, the girl stood on my left.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:01:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, she is on your left.", + "B": "Yes, she is on your right." + }, + "answer_label": "A", + "answer_text": "No, she is on your left.", + "q_id": "egolife_A1_D3_030_6_T1", + "correct_answer": "No, she is on your left.", + "original_options": { + "A": "No, she is on your left.", + "B": "Yes, she is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_6_T2", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Is the girl in the green dress on my right?", + "answer": "No, she is on your left.", + "question_chinese": "此时，绿衣服的女生在我的右边吗？", + "answer_chinese": "不是，她在你的左边", + "evidence": [ + { + "description": "At this moment, the girl stood on my left.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, she is on your left.", + "B": "Yes, she is on your right." + }, + "answer_label": "A", + "answer_text": "No, she is on your left.", + "q_id": "egolife_A1_D3_030_6_T2", + "correct_answer": "No, she is on your left.", + "original_options": { + "A": "No, she is on your left.", + "B": "Yes, she is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_7_T1", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, is the girl in green to my right?", + "answer": "No, she is on your left.", + "question_chinese": "此时，绿衣服的女生在我的右边吗？", + "answer_chinese": "是的，她在你的右边。", + "evidence": [ + { + "description": "At this moment, the girl stood on my left.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:26" + } + }, + { + "description": "During this period, the girl in green started from my left side, walked half a circle around me, and came to my right side.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, she is on your left.", + "B": "Yes, she is on your right." + }, + "answer_label": "A", + "answer_text": "No, she is on your left.", + "q_id": "egolife_A1_D3_030_7_T1", + "correct_answer": "No, she is on your left.", + "original_options": { + "A": "No, she is on your left.", + "B": "Yes, she is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_030_7_T2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "At this moment, is the girl in the green clothes to my right?", + "answer": "No, she is on your left.", + "question_chinese": "此时，绿衣服的女生在我的右边吗？", + "answer_chinese": "是的，她在你的右边。", + "evidence": [ + { + "description": "At this moment, the girl stood on my left.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:26" + } + }, + { + "description": "During this period, the girl in green started from my left side, walked half a circle around me, and came to my right side.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:02:01", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, she is on your left.", + "B": "Yes, she is on your right." + }, + "answer_label": "A", + "answer_text": "No, she is on your left.", + "q_id": "egolife_A1_D3_030_7_T2", + "correct_answer": "No, she is on your left.", + "original_options": { + "A": "No, she is on your left.", + "B": "Yes, she is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/egolife_A1_D3_031-mcq.json b/QAs/egolife_A1_D3_031-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7d3e3ee50ca864b7b1416562ee6db626e4a5d0eb --- /dev/null +++ b/QAs/egolife_A1_D3_031-mcq.json @@ -0,0 +1,1006 @@ +[ + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031-mcq_fixed_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "So far, have I been to the barbecue area?", + "answer": "No, you have been cutting mangoes inside the house the whole time.", + "question_chinese": "到目前为止，我去过烧烤的地方吗？", + "answer_chinese": "没有，你一直在屋子里切芒果。", + "evidence": [ + { + "description": "During this time, you have been constantly cutting mangoes.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:04:26", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you have been sitting in the barbecue area enjoying the afternoon.", + "B": "No, you have been cutting mangoes inside the house the whole time." + }, + "answer_label": "B", + "answer_text": "No, you have been cutting mangoes inside the house the whole time.", + "q_id": "egolife_A1_D3_031-mcq_fixed_6", + "correct_answer": "No, you have been cutting mangoes inside the house the whole time.", + "original_options": { + "A": "Yes, you have been sitting in the barbecue area enjoying the afternoon.", + "B": "No, you have been cutting mangoes inside the house the whole time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031-mcq_fixed_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "So far, have I been to the barbecue area?", + "answer": "Yes, you have been there.", + "question_chinese": "到目前为止，我去过烧烤的地方吗？", + "answer_chinese": "是的，你去过。", + "evidence": [ + { + "description": "During this time, you have been constantly cutting mangoes.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + }, + { + "description": "At this moment, you can be seen arriving at the barbecue area.", + "timestamps": { + "start": "00:05:42", + "end": "00:05:43" + } + } + ], + "question_timestamps": "00:06:37", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you have been there.", + "B": "No, you have not been there." + }, + "answer_label": "A", + "answer_text": "Yes, you have been there.", + "q_id": "egolife_A1_D3_031-mcq_fixed_7", + "correct_answer": "Yes, you have been there.", + "original_options": { + "A": "Yes, you have been there.", + "B": "No, you have not been there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_8", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "I just went to the second floor a few times?", + "answer": "You just went to the second floor once.", + "question_chinese": "我刚刚去了几次二楼？", + "answer_chinese": "你刚刚去了一次二楼。", + "evidence": [ + { + "description": "At this moment, you went to the second floor for the first time.", + "timestamps": { + "start": "00:06:29", + "end": "00:06:30" + } + } + ], + "question_timestamps": "00:08:11", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You just went to the second floor once.", + "B": "You went twice.", + "C": "You went three times.", + "D": "You just went to the second floor twice, going up and coming back down one time.", + "E": "You just went to the second floor three times, making multiple trips upstairs during this period." + }, + "answer_label": "A", + "answer_text": "You just went to the second floor once.", + "q_id": "egolife_A1_D3_031_8", + "correct_answer": "You just went to the second floor once.", + "original_options": { + "A": "You just went to the second floor once.", + "B": "You went twice.", + "C": "You went three times.", + "D": "You just went to the second floor twice, going up and coming back down one time.", + "E": "You just went to the second floor three times, making multiple trips upstairs during this period." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_9", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I been to the second floor just now?", + "answer": "You just went to the second floor once.", + "question_chinese": "我刚刚去了几次二楼？", + "answer_chinese": "你刚刚去了两次二楼。", + "evidence": [ + { + "description": "At this moment, you went to the second floor for the first time.", + "timestamps": { + "start": "00:06:29", + "end": "00:06:30" + } + }, + { + "description": "At this moment, you went to the second floor for the second time.", + "timestamps": { + "start": "00:11:25", + "end": "00:11:27" + } + } + ], + "question_timestamps": "00:11:52", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You just went to the second floor once.", + "B": "Twice.", + "C": "Three times.", + "D": "You just went to the second floor twice, with one additional trip upstairs besides the first.", + "E": "You just went to the second floor three times, repeating the upstairs trip several times just now." + }, + "answer_label": "A", + "answer_text": "You just went to the second floor once.", + "q_id": "egolife_A1_D3_031_9", + "correct_answer": "You just went to the second floor once.", + "original_options": { + "A": "You just went to the second floor once.", + "B": "Twice.", + "C": "Three times.", + "D": "You just went to the second floor twice, with one additional trip upstairs besides the first.", + "E": "You just went to the second floor three times, repeating the upstairs trip several times just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031-mcq_fixed_4", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after I just finished taking the light reflector?", + "answer": "You put it in the barbecue house, turned it on, and recorded a video of the barbecue with your phone.", + "question_chinese": "我刚刚拿完打光板后做了什么？", + "answer_chinese": "你把它放到烧烤屋里打开并用手机拍了烧烤的视频。", + "evidence": [ + { + "description": "At this moment, you took the light reflector into the barbecue house and turned it on.", + "timestamps": { + "start": "00:07:43", + "end": "00:07:53" + } + }, + { + "description": "At this moment, you filmed the barbecue video with your phone.", + "timestamps": { + "start": "00:08:33", + "end": "00:08:42" + } + } + ], + "question_timestamps": "00:08:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turned it off and walked away.", + "B": "You left it outside by the grill.", + "C": "You carried it into the barbecue house, left it off, and then filmed the barbecue on your phone.", + "D": "You put it in the barbecue house, turned it on, and recorded a video of the barbecue with your phone.", + "E": "You set it inside the barbecue house, turned it on briefly, then took a few photos instead of recording video." + }, + "answer_label": "D", + "answer_text": "You put it in the barbecue house, turned it on, and recorded a video of the barbecue with your phone.", + "q_id": "egolife_A1_D3_031-mcq_fixed_4", + "correct_answer": "You put it in the barbecue house, turned it on, and recorded a video of the barbecue with your phone.", + "original_options": { + "A": "You turned it off and walked away.", + "B": "You left it outside by the grill.", + "C": "You carried it into the barbecue house, left it off, and then filmed the barbecue on your phone.", + "D": "You put it in the barbecue house, turned it on, and recorded a video of the barbecue with your phone.", + "E": "You set it inside the barbecue house, turned it on briefly, then took a few photos instead of recording video." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031-mcq_fixed_5", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I do after I just finished taking the light reflector?", + "answer": "You placed the ring light next to the grill, turned on the light, then took out your phone to film a video, and finally went to the bathroom to wash your hands.", + "question_chinese": "我刚刚拿完打光板后做了什么？", + "answer_chinese": "你把打光灯放到了烧烤架旁边并打开了灯，然后拿出了手机拍视频，最后去洗手间洗了手。", + "evidence": [ + { + "description": "At this moment, you filmed the barbecue video with your phone.", + "timestamps": { + "start": "00:08:33", + "end": "00:08:42" + } + }, + { + "description": "At this moment, you took the light reflector into the barbecue house and turned it on.", + "timestamps": { + "start": "00:07:43", + "end": "00:07:53" + } + }, + { + "description": "At this moment, you come to the bathroom to wash your hands.", + "timestamps": { + "start": "00:09:13", + "end": "00:09:14" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You put it beside the grill and left.", + "B": "You placed the ring light next to the grill, turned on the light, then took out your phone to film a video, and finally went to the bathroom to wash your hands.", + "C": "You turned the light off and filmed.", + "D": "You placed the ring light next to the grill, turned it on, checked your phone, and then stayed to watch the grill.", + "E": "You set the ring light near the grill, left the light off, filmed a video anyway, and then washed your hands elsewhere." + }, + "answer_label": "B", + "answer_text": "You placed the ring light next to the grill, turned on the light, then took out your phone to film a video, and finally went to the bathroom to wash your hands.", + "q_id": "egolife_A1_D3_031-mcq_fixed_5", + "correct_answer": "You placed the ring light next to the grill, turned on the light, then took out your phone to film a video, and finally went to the bathroom to wash your hands.", + "original_options": { + "A": "You put it beside the grill and left.", + "B": "You placed the ring light next to the grill, turned on the light, then took out your phone to film a video, and finally went to the bathroom to wash your hands.", + "C": "You turned the light off and filmed.", + "D": "You placed the ring light next to the grill, turned it on, checked your phone, and then stayed to watch the grill.", + "E": "You set the ring light near the grill, left the light off, filmed a video anyway, and then washed your hands elsewhere." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 22, + 22 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031-mcq_fixed_8", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just enter the room with several computers?", + "answer": "No, you just did not enter the room with several computers.", + "question_chinese": "我刚刚进入了放有多台电脑的房间里了吗？", + "answer_chinese": "没有，你刚刚没有进入了放有多台电脑的房间里。", + "evidence": [ + { + "description": "You have been constantly cutting mangoes, so you did not enter the room.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:03:38", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you just did not enter the room with several computers.", + "B": "Yes, you just entered the room with several computers." + }, + "answer_label": "A", + "answer_text": "No, you just did not enter the room with several computers.", + "q_id": "egolife_A1_D3_031-mcq_fixed_8", + "correct_answer": "No, you just did not enter the room with several computers.", + "original_options": { + "A": "No, you just did not enter the room with several computers.", + "B": "Yes, you just entered the room with several computers." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_2_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just enter a room that has multiple computers?", + "answer": "Yes. You just entered the room with multiple computers, and the room is behind you.", + "question_chinese": "我刚刚进入了放有多台电脑的房间里了吗？", + "answer_chinese": "是的，你刚刚进入了放有多台电脑的房间里，房间在你身后。", + "evidence": [ + { + "description": "At this moment, you entered the room with several computers.", + "timestamps": { + "start": "00:04:34", + "end": "00:04:37" + } + } + ], + "question_timestamps": "00:04:39", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. You just entered the room with multiple computers, and the room is behind you.", + "B": "No. You didn't enter the room with multiple computers; you were in front of the whiteboard, facing the opposite room." + }, + "answer_label": "A", + "answer_text": "Yes. You just entered the room with multiple computers, and the room is behind you.", + "q_id": "egolife_A1_D3_031_2_T1", + "correct_answer": "Yes. You just entered the room with multiple computers, and the room is behind you.", + "original_options": { + "A": "Yes. You just entered the room with multiple computers, and the room is behind you.", + "B": "No. You didn't enter the room with multiple computers; you were in front of the whiteboard, facing the opposite room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_2_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just enter a room that has multiple computers?", + "answer": "No. You didn't enter the room with multiple computers. You were in front of the whiteboard, and the room with multiple computers was opposite the whiteboard.", + "question_chinese": "我刚刚进入了放有多台电脑的房间里了吗？", + "answer_chinese": "没有，你没有进入放有多台电脑的房间里，你在白板前面，放有多台电脑的房间在白板的对面。", + "evidence": [ + { + "description": "You not entered the room with several computers.", + "timestamps": { + "start": "00:04:43", + "end": "00:04:48" + } + } + ], + "question_timestamps": "00:04:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. You just entered the room with multiple computers, and the room is behind you.", + "B": "No. You didn't enter the room with multiple computers. You were in front of the whiteboard, and the room with multiple computers was opposite the whiteboard." + }, + "answer_label": "B", + "answer_text": "No. You didn't enter the room with multiple computers. You were in front of the whiteboard, and the room with multiple computers was opposite the whiteboard.", + "q_id": "egolife_A1_D3_031_2_T2", + "correct_answer": "No. You didn't enter the room with multiple computers. You were in front of the whiteboard, and the room with multiple computers was opposite the whiteboard.", + "original_options": { + "A": "Yes. You just entered the room with multiple computers, and the room is behind you.", + "B": "No. You didn't enter the room with multiple computers. You were in front of the whiteboard, and the room with multiple computers was opposite the whiteboard." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_3", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just enter a room with multiple computers?", + "answer": "Yes, you just entered the room with several computers.", + "question_chinese": "我刚刚进入了放有多台电脑的房间里了吗？", + "answer_chinese": "是的，你刚刚进入了放有多台电脑的房间里。", + "evidence": [ + { + "description": "At this moment, you entered the room with several computers.", + "timestamps": { + "start": "00:04:34", + "end": "00:04:37" + } + } + ], + "question_timestamps": "00:08:41", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you just entered the room with several computers.", + "B": "No, you just exited the room with several computers." + }, + "answer_label": "A", + "answer_text": "Yes, you just entered the room with several computers.", + "q_id": "egolife_A1_D3_031_3", + "correct_answer": "Yes, you just entered the room with several computers.", + "original_options": { + "A": "Yes, you just entered the room with several computers.", + "B": "No, you just exited the room with several computers." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_4", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just enter a room with multiple computers?", + "answer": "Yes, you just went into the room with multiple computers.", + "question_chinese": "我刚刚进入了放有多台电脑的房间里了吗？", + "answer_chinese": "是的，你刚刚进入了放有多台电脑的房间里。", + "evidence": [ + { + "description": "At this moment, you entered the room with several computers.", + "timestamps": { + "start": "00:04:34", + "end": "00:04:37" + } + } + ], + "question_timestamps": "00:10:41", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you just left the room without entering the one with multiple computers.", + "B": "Yes, you just went into the room with multiple computers." + }, + "answer_label": "B", + "answer_text": "Yes, you just went into the room with multiple computers.", + "q_id": "egolife_A1_D3_031_4", + "correct_answer": "Yes, you just went into the room with multiple computers.", + "original_options": { + "A": "No, you just left the room without entering the one with multiple computers.", + "B": "Yes, you just went into the room with multiple computers." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_3_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What am I doing?", + "answer": "You are chopping vegetables, and the cutting board and knife are in front of you.", + "question_chinese": "我在干什么？", + "answer_chinese": "你在切菜，菜板和菜刀在你前面。", + "evidence": [ + { + "description": "You are cutting vegetables.", + "timestamps": { + "start": "00:00:23", + "end": "00:00:30" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You are washing vegetables at the sink.", + "B": "You are chopping vegetables, and the cutting board and knife are in front of you.", + "C": "You are peeling vegetables with a knife.", + "D": "You are arranging vegetables on the cutting board, and the board is in front of you without chopping.", + "E": "You are carrying vegetables to the counter, and the cutting board and knife are not being used yet." + }, + "answer_label": "B", + "answer_text": "You are chopping vegetables, and the cutting board and knife are in front of you.", + "q_id": "egolife_A1_D3_031_3_T1", + "correct_answer": "You are chopping vegetables, and the cutting board and knife are in front of you.", + "original_options": { + "A": "You are washing vegetables at the sink.", + "B": "You are chopping vegetables, and the cutting board and knife are in front of you.", + "C": "You are peeling vegetables with a knife.", + "D": "You are arranging vegetables on the cutting board, and the board is in front of you without chopping.", + "E": "You are carrying vegetables to the counter, and the cutting board and knife are not being used yet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_3_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What am I doing?", + "answer": "You are going up the stairs, and the stairs are under your feet.", + "question_chinese": "我在干什么？", + "answer_chinese": "你在上楼梯，楼梯在你脚下。", + "evidence": [ + { + "description": "You are going upstairs.", + "timestamps": { + "start": "00:06:29", + "end": "00:06:34" + } + } + ], + "question_timestamps": "00:06:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You are going down the stairs.", + "B": "You are standing on the landing.", + "C": "You are going up the stairs, and the stairs are under your feet.", + "D": "You are walking along the hallway, with the flat floor under your feet instead of stairs.", + "E": "You are stepping off the stairs onto the landing, with the landing under your feet." + }, + "answer_label": "C", + "answer_text": "You are going up the stairs, and the stairs are under your feet.", + "q_id": "egolife_A1_D3_031_3_T2", + "correct_answer": "You are going up the stairs, and the stairs are under your feet.", + "original_options": { + "A": "You are going down the stairs.", + "B": "You are standing on the landing.", + "C": "You are going up the stairs, and the stairs are under your feet.", + "D": "You are walking along the hallway, with the flat floor under your feet instead of stairs.", + "E": "You are stepping off the stairs onto the landing, with the landing under your feet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_7_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Is anyone next to me?", + "answer": "No one is next to you. You're in the kitchen; the dining table is behind you, and there is someone next to the dining table.", + "question_chinese": "我旁边有人吗？", + "answer_chinese": "你旁边没有人，你在厨房里面，餐桌在你身后，餐桌旁边有人。", + "evidence": [ + { + "description": "There's nobody beside you.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:00:59", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. Someone is next to you in the kitchen, standing beside you rather than by the dining table behind you.", + "B": "No one is next to you. You're in the kitchen; the dining table is behind you, and there is someone next to the dining table." + }, + "answer_label": "B", + "answer_text": "No one is next to you. You're in the kitchen; the dining table is behind you, and there is someone next to the dining table.", + "q_id": "egolife_A1_D3_031_7_T1", + "correct_answer": "No one is next to you. You're in the kitchen; the dining table is behind you, and there is someone next to the dining table.", + "original_options": { + "A": "Yes. Someone is next to you in the kitchen, standing beside you rather than by the dining table behind you.", + "B": "No one is next to you. You're in the kitchen; the dining table is behind you, and there is someone next to the dining table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_7_T2", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Is anyone next to me?", + "answer": "Yes. There is a girl next to you, and she is to your left.", + "question_chinese": "我旁边有人吗？", + "answer_chinese": "有，你旁边有一位女生，她在你左边。", + "evidence": [ + { + "description": "Yes, there is a girl beside you.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. There is a girl next to you, and she is to your left.", + "B": "No one is next to you. You're in the kitchen; the dining table is behind you, and there is someone next to it." + }, + "answer_label": "A", + "answer_text": "Yes. There is a girl next to you, and she is to your left.", + "q_id": "egolife_A1_D3_031_7_T2", + "correct_answer": "Yes. There is a girl next to you, and she is to your left.", + "original_options": { + "A": "Yes. There is a girl next to you, and she is to your left.", + "B": "No one is next to you. You're in the kitchen; the dining table is behind you, and there is someone next to it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 23 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_1", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I last put the knife?", + "answer": "You put it on the cutting board by the sink.", + "question_chinese": "我最后一次把刀放在了哪里？", + "answer_chinese": "你把它放到了水槽边的切菜板上。", + "evidence": [ + { + "description": "At this moment, you put it on the cutting board by the sink.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:19" + } + } + ], + "question_timestamps": "00:08:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the dining table.", + "B": "You put it on the cutting board by the sink.", + "C": "On the stove-side cutting board.", + "D": "You last put the knife on the dining table next to a chair, not by the sink.", + "E": "You last left it on a silver tray in the pantry rather than on the cutting board by the sink." + }, + "answer_label": "B", + "answer_text": "You put it on the cutting board by the sink.", + "q_id": "egolife_A1_D3_031_1", + "correct_answer": "You put it on the cutting board by the sink.", + "original_options": { + "A": "On the dining table.", + "B": "You put it on the cutting board by the sink.", + "C": "On the stove-side cutting board.", + "D": "You last put the knife on the dining table next to a chair, not by the sink.", + "E": "You last left it on a silver tray in the pantry rather than on the cutting board by the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_2", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I last put the knife?", + "answer": "You put the chef's knife on the table in the barbecue house.", + "question_chinese": "我最后一次把刀放在了哪里？", + "answer_chinese": "你把菜刀放到了烧烤屋里的桌子上。", + "evidence": [ + { + "description": "At this moment, you put it on the cutting board by the sink.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:19" + } + }, + { + "description": "At this moment, you put the chef's knife on the table in the barbecue house.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:11" + } + } + ], + "question_timestamps": "00:11:52", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the counter.", + "B": "On the shelf.", + "C": "You put the chef's knife on the table in the barbecue house.", + "D": "You put the chef's knife on the counter in the barbecue house instead of on the table.", + "E": "You put the butter knife on the table in the barbecue house, not the chef's knife." + }, + "answer_label": "C", + "answer_text": "You put the chef's knife on the table in the barbecue house.", + "q_id": "egolife_A1_D3_031_2", + "correct_answer": "You put the chef's knife on the table in the barbecue house.", + "original_options": { + "A": "On the counter.", + "B": "On the shelf.", + "C": "You put the chef's knife on the table in the barbecue house.", + "D": "You put the chef's knife on the counter in the barbecue house instead of on the table.", + "E": "You put the butter knife on the table in the barbecue house, not the chef's knife." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_4_T1", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the black-and-white checkered board relative to me?", + "answer": "The black-and-white checkered board is in front of you.", + "question_chinese": "黑白格子的板子在我的哪里？", + "answer_chinese": "黑白格子的板子在你的前面", + "evidence": [ + { + "description": "The black and white checkered board is in front of you.", + "timestamps": { + "start": "00:04:26", + "end": "00:04:30" + } + } + ], + "question_timestamps": "00:04:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "To your right.", + "B": "Behind you.", + "C": "The black-and-white checkered board is in front of you.", + "D": "The black-and-white checkered board is to your left rather than positioned directly in front of you.", + "E": "The black-and-white checkered board is under your feet, not ahead of you in your line of sight." + }, + "answer_label": "C", + "answer_text": "The black-and-white checkered board is in front of you.", + "q_id": "egolife_A1_D3_031_4_T1", + "correct_answer": "The black-and-white checkered board is in front of you.", + "original_options": { + "A": "To your right.", + "B": "Behind you.", + "C": "The black-and-white checkered board is in front of you.", + "D": "The black-and-white checkered board is to your left rather than positioned directly in front of you.", + "E": "The black-and-white checkered board is under your feet, not ahead of you in your line of sight." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_031_4_T2", + "qid_num": 19, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the black-and-white checkered board relative to me?", + "answer": "The black-and-white checkered board is to your right.", + "question_chinese": "黑白格子的板子在我的哪里？", + "answer_chinese": "黑白格子的板子在你的右边", + "evidence": [ + { + "description": "The black and white checkered board is on your right.", + "timestamps": { + "start": "00:04:38", + "end": "00:04:42" + } + } + ], + "question_timestamps": "00:04:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The black-and-white checkered board is behind you instead of being positioned to your right.", + "D": "The black-and-white checkered board is to your right.", + "E": "The black-and-white checkered board is under your feet rather than beside you on your right." + }, + "answer_label": "D", + "answer_text": "The black-and-white checkered board is to your right.", + "q_id": "egolife_A1_D3_031_4_T2", + "correct_answer": "The black-and-white checkered board is to your right.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The black-and-white checkered board is behind you instead of being positioned to your right.", + "D": "The black-and-white checkered board is to your right.", + "E": "The black-and-white checkered board is under your feet rather than beside you on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D3_032-mcq.json b/QAs/egolife_A1_D3_032-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f473949485426d153b11aae722838e6a70e47eb0 --- /dev/null +++ b/QAs/egolife_A1_D3_032-mcq.json @@ -0,0 +1,788 @@ +[ + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032_3_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I am, which is closer to me: the plate or the vase?", + "answer": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you.", + "question_chinese": "从我的位置看，盘子和花瓶之间哪个离我更近？", + "answer_chinese": "花瓶离你更近，此时，花瓶在你右手边，而盘子在你前面。", + "evidence": [ + { + "description": "At this moment, the vase is on your right side, while the plate is in front of you.", + "timestamps": { + "start": "00:00:49", + "end": "00:00:50" + } + }, + { + "description": "The plate is relatively farther away compared to the vase.", + "timestamps": { + "start": "00:00:49", + "end": "00:00:51" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you.", + "B": "The plate is closer to you. Right now, the plate is by your right hand, and the vase is farther away in front of you." + }, + "answer_label": "A", + "answer_text": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you.", + "q_id": "egolife_A1_D3_032_3_T1", + "correct_answer": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you.", + "original_options": { + "A": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you.", + "B": "The plate is closer to you. Right now, the plate is by your right hand, and the vase is farther away in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 25 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032_3_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I am, which is closer to me: the plate or the vase?", + "answer": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you.", + "question_chinese": "从我的位置看，盘子和花瓶之间哪个离我更近？", + "answer_chinese": "花瓶离你更近，此时，花瓶在你右手边，而盘子在你前面。", + "evidence": [ + { + "description": "At this moment, the vase is on your right side, while the plate is in front of you.", + "timestamps": { + "start": "00:00:49", + "end": "00:00:50" + } + }, + { + "description": "The plate is relatively farther away compared to the vase.", + "timestamps": { + "start": "00:00:49", + "end": "00:00:51" + } + } + ], + "question_timestamps": "00:01:02", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The plate is closer to you. Right now, the plate is by your right hand, and the vase is in front of you.", + "B": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you." + }, + "answer_label": "B", + "answer_text": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you.", + "q_id": "egolife_A1_D3_032_3_T2", + "correct_answer": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you.", + "original_options": { + "A": "The plate is closer to you. Right now, the plate is by your right hand, and the vase is in front of you.", + "B": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 23 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032_3_T3", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I am, which is closer to me: the plate or the vase?", + "answer": "The plate is closer to you. Right now, the plate is by your left hand, and the vase is farther away in front of you.", + "question_chinese": "从我的位置看，盘子和花瓶之间哪个离我更近？", + "answer_chinese": "盘子离你更近，此时盘子在你左手边，而花瓶在你前方较远处。", + "evidence": [ + { + "description": "At this moment, the plate is on your left side, while the vase is further away in front of you.", + "timestamps": { + "start": "00:05:34", + "end": "00:05:37" + } + } + ], + "question_timestamps": "00:05:39", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The plate is closer to you. Right now, the plate is by your left hand, and the vase is farther away in front of you.", + "B": "The vase is closer to you. Right now, the vase is by your left hand, and the plate is farther away in front of you." + }, + "answer_label": "A", + "answer_text": "The plate is closer to you. Right now, the plate is by your left hand, and the vase is farther away in front of you.", + "q_id": "egolife_A1_D3_032_3_T3", + "correct_answer": "The plate is closer to you. Right now, the plate is by your left hand, and the vase is farther away in front of you.", + "original_options": { + "A": "The plate is closer to you. Right now, the plate is by your left hand, and the vase is farther away in front of you.", + "B": "The vase is closer to you. Right now, the vase is by your left hand, and the plate is farther away in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 25 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032_3_T4", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I am, which is closer to me: the plate or the vase?", + "answer": "The plate is closer to you. Right now, the plate is by your left hand, and the vase is farther away in front of you.", + "question_chinese": "从我的位置看，盘子和花瓶之间哪个离我更近？", + "answer_chinese": "盘子离你更近，此时盘子在你左手边，而花瓶在你前方较远处。", + "evidence": [ + { + "description": "At this moment, the plate is on your left side, while the vase is further away in front of you.", + "timestamps": { + "start": "00:05:34", + "end": "00:05:37" + } + } + ], + "question_timestamps": "00:05:49", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you.", + "B": "The plate is closer to you. Right now, the plate is by your left hand, and the vase is farther away in front of you." + }, + "answer_label": "B", + "answer_text": "The plate is closer to you. Right now, the plate is by your left hand, and the vase is farther away in front of you.", + "q_id": "egolife_A1_D3_032_3_T4", + "correct_answer": "The plate is closer to you. Right now, the plate is by your left hand, and the vase is farther away in front of you.", + "original_options": { + "A": "The vase is closer to you. Right now, the vase is by your right hand, and the plate is in front of you.", + "B": "The plate is closer to you. Right now, the plate is by your left hand, and the vase is farther away in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 23 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032-mcq_fixed_10", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just take out from the cabinet in the kitchen on the first floor?", + "answer": "You did not open the cabinet in the first-floor kitchen just now，so you didn't pick up anything.", + "question_chinese": "我刚刚从一楼厨房的柜子里拿出了什么？", + "answer_chinese": "你刚刚没有打开过一楼厨房的柜子，所以什么也没有拿。", + "evidence": [ + { + "description": "At this moment, you have just arrived at the kitchen and have not opened the cabinet.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took out a green plate.", + "B": "You took out a red bowl.", + "C": "You did not open the cabinet in the first-floor kitchen just now，so you didn't pick up anything.", + "D": "You opened the first-floor kitchen cabinet and took out a box of herbal tea bags from inside.", + "E": "You opened the cabinet in the first-floor kitchen and removed a set of white ceramic mugs from the middle shelf." + }, + "answer_label": "C", + "answer_text": "You did not open the cabinet in the first-floor kitchen just now，so you didn't pick up anything.", + "q_id": "egolife_A1_D3_032-mcq_fixed_10", + "correct_answer": "You did not open the cabinet in the first-floor kitchen just now，so you didn't pick up anything.", + "original_options": { + "A": "You took out a green plate.", + "B": "You took out a red bowl.", + "C": "You did not open the cabinet in the first-floor kitchen just now，so you didn't pick up anything.", + "D": "You opened the first-floor kitchen cabinet and took out a box of herbal tea bags from inside.", + "E": "You opened the cabinet in the first-floor kitchen and removed a set of white ceramic mugs from the middle shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032-mcq_fixed_11", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just take out from the cabinet in the kitchen on the first floor?", + "answer": "You just took out a pair of scissors and a chef's knife from the kitchen cabinet.", + "question_chinese": "我刚刚从一楼厨房的柜子里拿出了什么？", + "answer_chinese": "At this moment, you took the chef's knife out of the cabinet.", + "evidence": [ + { + "description": "At this moment, you took the scissors out of the cabinet.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:24" + } + }, + { + "description": "At this moment, you took the chef's knife out of the cabinet.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:07:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took out a rolling pin.", + "B": "You took out a whisk.", + "C": "You just took out a pair of scissors and a chef's knife from the kitchen cabinet.", + "D": "You just took out a serrated bread knife and a rolling pin from the kitchen cabinet on the first floor.", + "E": "You removed a whisk, a serrated bread knife, and the matching cutting board from the lower part of the cabinet." + }, + "answer_label": "C", + "answer_text": "You just took out a pair of scissors and a chef's knife from the kitchen cabinet.", + "q_id": "egolife_A1_D3_032-mcq_fixed_11", + "correct_answer": "You just took out a pair of scissors and a chef's knife from the kitchen cabinet.", + "original_options": { + "A": "You took out a rolling pin.", + "B": "You took out a whisk.", + "C": "You just took out a pair of scissors and a chef's knife from the kitchen cabinet.", + "D": "You just took out a serrated bread knife and a rolling pin from the kitchen cabinet on the first floor.", + "E": "You removed a whisk, a serrated bread knife, and the matching cutting board from the lower part of the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many layers of cake have I already cut?", + "answer": "You have already cut three layers of cake.", + "question_chinese": "我已经切了几层蛋糕？", + "answer_chinese": "你已经切了三层蛋糕。", + "evidence": [ + { + "description": "There are exactly three layers of cake that you have cut.", + "timestamps": { + "start": "00:05:37", + "end": "00:05:45" + } + } + ], + "question_timestamps": "00:06:02", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have already cut three layers of cake.", + "B": "You cut two layers.", + "C": "You cut four layers.", + "D": "You have already cut five layers of cake, so there are five sliced sections visible.", + "E": "You have already cut six layers of cake, with each layer separated into distinct cut portions." + }, + "answer_label": "A", + "answer_text": "You have already cut three layers of cake.", + "q_id": "egolife_A1_D3_032_1", + "correct_answer": "You have already cut three layers of cake.", + "original_options": { + "A": "You have already cut three layers of cake.", + "B": "You cut two layers.", + "C": "You cut four layers.", + "D": "You have already cut five layers of cake, so there are five sliced sections visible.", + "E": "You have already cut six layers of cake, with each layer separated into distinct cut portions." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many layers of cake have I already cut?", + "answer": "At this moment, exactly four layers of cake have been cut by you.", + "question_chinese": "我已经切了几层蛋糕？", + "answer_chinese": "你已经切了四层蛋糕。", + "evidence": [ + { + "description": "At this moment, exactly four layers of cake have been cut by you.", + "timestamps": { + "start": "00:07:10", + "end": "00:07:16" + } + } + ], + "question_timestamps": "00:08:26", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "At this moment, exactly four layers of cake have been cut by you.", + "B": "Exactly three layers.", + "C": "Exactly five layers.", + "D": "At this moment, exactly two layers of cake have been cut by you, not four layers.", + "E": "At this moment, exactly one layer of cake has been cut by you, with the rest still uncut." + }, + "answer_label": "A", + "answer_text": "At this moment, exactly four layers of cake have been cut by you.", + "q_id": "egolife_A1_D3_032_2", + "correct_answer": "At this moment, exactly four layers of cake have been cut by you.", + "original_options": { + "A": "At this moment, exactly four layers of cake have been cut by you.", + "B": "Exactly three layers.", + "C": "Exactly five layers.", + "D": "At this moment, exactly two layers of cake have been cut by you, not four layers.", + "E": "At this moment, exactly one layer of cake has been cut by you, with the rest still uncut." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032-mcq_fixed_6", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position after picking up the light?", + "answer": "You took the light from the second floor, turned around, went downstairs, and walked to this staircase entrance.", + "question_chinese": "我拿起打光灯后是怎么走到现在的位置的？", + "answer_chinese": "你从二楼拿走打光灯后向后转下楼后走到现在这个楼梯口。", + "evidence": [ + { + "description": "At this moment, after you picked up the light, you turned around.", + "timestamps": { + "start": "00:11:04", + "end": "00:11:08" + } + }, + { + "description": "At this moment, you walked to the door, started going downstairs, and reached the staircase entrance.", + "timestamps": { + "start": "00:11:09", + "end": "00:11:20" + } + } + ], + "question_timestamps": "00:11:21", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You stayed upstairs after taking the light.", + "B": "You took the light from the second floor, turned around, went downstairs, and walked to this staircase entrance.", + "C": "You went upstairs after picking up the light.", + "D": "You took the light from the second floor, walked down the stairs, and headed toward the garden entrance instead.", + "E": "You picked up the light on the first floor, turned around, went upstairs, and walked to the kitchen entrance." + }, + "answer_label": "B", + "answer_text": "You took the light from the second floor, turned around, went downstairs, and walked to this staircase entrance.", + "q_id": "egolife_A1_D3_032-mcq_fixed_6", + "correct_answer": "You took the light from the second floor, turned around, went downstairs, and walked to this staircase entrance.", + "original_options": { + "A": "You stayed upstairs after taking the light.", + "B": "You took the light from the second floor, turned around, went downstairs, and walked to this staircase entrance.", + "C": "You went upstairs after picking up the light.", + "D": "You took the light from the second floor, walked down the stairs, and headed toward the garden entrance instead.", + "E": "You picked up the light on the first floor, turned around, went upstairs, and walked to the kitchen entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032-mcq_fixed_7", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position after picking up the light?", + "answer": "You took the light from the second floor, turned around, went downstairs to the stairwell, walked straight to the end, turned left, and arrived at your current position.", + "question_chinese": "我拿起打光灯后是怎么走到现在的位置的？", + "answer_chinese": "你从二楼拿走打光灯后向后转下楼后走到楼梯口再直走到尽头左转来到了现在的位置。", + "evidence": [ + { + "description": "At this moment, after you picked up the light, you turned around.", + "timestamps": { + "start": "00:11:04", + "end": "00:11:08" + } + }, + { + "description": "At this moment, you walked to the door, started going downstairs, and reached the staircase entrance.", + "timestamps": { + "start": "00:11:09", + "end": "00:11:21" + } + }, + { + "description": "At this moment, you went straight from the stairwell, turned left at the end, and arrived at your current position.", + "timestamps": { + "start": "00:11:23", + "end": "00:11:26" + } + } + ], + "question_timestamps": "00:11:27", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You took the light from the second floor, turned around, went downstairs to the stairwell, walked straight to the end, turned left, and arrived at your current position.", + "B": "You went upstairs and turned right.", + "C": "You stayed on the first floor.", + "D": "You took the light from the second floor, went downstairs, walked to the end, then turned right to arrive.", + "E": "You grabbed the light upstairs, headed down to the stairwell, turned left too early, and stopped midway." + }, + "answer_label": "A", + "answer_text": "You took the light from the second floor, turned around, went downstairs to the stairwell, walked straight to the end, turned left, and arrived at your current position.", + "q_id": "egolife_A1_D3_032-mcq_fixed_7", + "correct_answer": "You took the light from the second floor, turned around, went downstairs to the stairwell, walked straight to the end, turned left, and arrived at your current position.", + "original_options": { + "A": "You took the light from the second floor, turned around, went downstairs to the stairwell, walked straight to the end, turned left, and arrived at your current position.", + "B": "You went upstairs and turned right.", + "C": "You stayed on the first floor.", + "D": "You took the light from the second floor, went downstairs, walked to the end, then turned right to arrive.", + "E": "You grabbed the light upstairs, headed down to the stairwell, turned left too early, and stopped midway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032-mcq_fixed_8", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes were there in the box next to the sink now?", + "answer": "Four.", + "question_chinese": "现在水槽旁边的盒子里有几个芒果？", + "answer_chinese": "现在盒子里一共有四个芒果。", + "evidence": [ + { + "description": "At this moment, you haven't taken any mangoes yet, and there are four mangoes in the box.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:12" + } + } + ], + "question_timestamps": "00:08:14", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "One.", + "C": "Two.", + "D": "Three mangoes were in the box next to the sink now, not four as you might expect.", + "E": "Five mangoes were sitting in the box by the sink now, filling it more than before." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "egolife_A1_D3_032-mcq_fixed_8", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "One.", + "C": "Two.", + "D": "Three mangoes were in the box next to the sink now, not four as you might expect.", + "E": "Five mangoes were sitting in the box by the sink now, filling it more than before." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032-mcq_fixed_9", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes were there in the box next to the sink now?", + "answer": "Six.", + "question_chinese": "现在水槽旁边的盒子里有几个芒果？", + "answer_chinese": "现在还有三个芒果。", + "evidence": [ + { + "description": "At this moment, you haven't taken any mangoes yet, and there are four mangoes in the box.", + "timestamps": { + "start": "00:08:11", + "end": "00:08:12" + } + }, + { + "description": "At this moment, you took out one mango, so there are still three mangoes left in the box.", + "timestamps": { + "start": "00:08:15", + "end": "00:08:16" + } + } + ], + "question_timestamps": "00:09:55", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "Seven mangoes were in the box next to the sink now, more than the correct count.", + "D": "Eight mangoes were in the box by the sink now, making it look more crowded than before.", + "E": "Six." + }, + "answer_label": "E", + "answer_text": "Six.", + "q_id": "egolife_A1_D3_032-mcq_fixed_9", + "correct_answer": "Six.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "Seven mangoes were in the box next to the sink now, more than the correct count.", + "D": "Eight mangoes were in the box by the sink now, making it look more crowded than before.", + "E": "Six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032_5", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where are the red wet wipes now?", + "answer": "The red wet wipe is now in the hand of the girl in green clothes who is sitting on the stool.", + "question_chinese": "红色湿纸巾现在在哪里？", + "answer_chinese": "红色湿纸巾现在在坐在凳子上的绿衣服女孩手里。", + "evidence": [ + { + "description": "At this moment, the girl in green is holding a wet wipe.", + "timestamps": { + "start": "00:00:46", + "end": "00:00:48" + } + } + ], + "question_timestamps": "00:00:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the girl's hand on the chair.", + "B": "In the girl's hand beside the stool.", + "C": "The red wet wipe is in the hand of the girl in green clothes while she is standing by the stool.", + "D": "The red wet wipe is in the hand of the girl in green clothes as she kneels near the stool.", + "E": "The red wet wipe is now in the hand of the girl in green clothes who is sitting on the stool." + }, + "answer_label": "E", + "answer_text": "The red wet wipe is now in the hand of the girl in green clothes who is sitting on the stool.", + "q_id": "egolife_A1_D3_032_5", + "correct_answer": "The red wet wipe is now in the hand of the girl in green clothes who is sitting on the stool.", + "original_options": { + "A": "In the girl's hand on the chair.", + "B": "In the girl's hand beside the stool.", + "C": "The red wet wipe is in the hand of the girl in green clothes while she is standing by the stool.", + "D": "The red wet wipe is in the hand of the girl in green clothes as she kneels near the stool.", + "E": "The red wet wipe is now in the hand of the girl in green clothes who is sitting on the stool." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_032_6", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where are the red wet wipes now?", + "answer": "The red wet wipe is now under the table in the living room on the first floor.", + "question_chinese": "红色湿纸巾现在在哪里？", + "answer_chinese": "红色湿纸巾现在在一楼客厅桌子下面。", + "evidence": [ + { + "description": "At this moment, the girl in green is holding a wet wipe.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:49" + } + }, + { + "description": "At this moment, the girl in green put the wet wipe under the table, and it has never been touched again since then.", + "timestamps": { + "start": "00:00:51", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:06:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Under the sofa in the living room.", + "B": "Under a chair on the second floor.", + "C": "The red wet wipe is now under the table in the living room on the first floor.", + "D": "The red wet wipe is under the table in the dining room on the first floor, not in the living room.", + "E": "The red wet wipe is under the table in the hallway on the first floor, close to where you are." + }, + "answer_label": "C", + "answer_text": "The red wet wipe is now under the table in the living room on the first floor.", + "q_id": "egolife_A1_D3_032_6", + "correct_answer": "The red wet wipe is now under the table in the living room on the first floor.", + "original_options": { + "A": "Under the sofa in the living room.", + "B": "Under a chair on the second floor.", + "C": "The red wet wipe is now under the table in the living room on the first floor.", + "D": "The red wet wipe is under the table in the dining room on the first floor, not in the living room.", + "E": "The red wet wipe is under the table in the hallway on the first floor, close to where you are." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 20 + ] + } +] diff --git a/QAs/egolife_A1_D3_033-mcq.json b/QAs/egolife_A1_D3_033-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d4679c58efafdf9f838417c2c9f1f6a7571b32ba --- /dev/null +++ b/QAs/egolife_A1_D3_033-mcq.json @@ -0,0 +1,791 @@ +[ + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033_3_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is that box of strawberries from earlier, relative to me now?", + "answer": "The box of strawberries is now in the hands of the girl in green behind you.", + "question_chinese": "刚刚的那盒草莓现在在相对我的哪个位置？", + "answer_chinese": "那盒草莓现在在你身后的绿衣服女孩手里。", + "evidence": [ + { + "description": "At this moment, the girl in green came to the sink, holding strawberries in her hand.", + "timestamps": { + "start": "00:02:55", + "end": "00:02:56" + } + } + ], + "question_timestamps": "00:03:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The box of strawberries is now in the hands of the girl in green to your right, not behind you.", + "D": "The box of strawberries is now in the hands of the girl in green above you, out of your direct reach.", + "E": "The box of strawberries is now in the hands of the girl in green behind you." + }, + "answer_label": "E", + "answer_text": "The box of strawberries is now in the hands of the girl in green behind you.", + "q_id": "egolife_A1_D3_033_3_T1", + "correct_answer": "The box of strawberries is now in the hands of the girl in green behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The box of strawberries is now in the hands of the girl in green to your right, not behind you.", + "D": "The box of strawberries is now in the hands of the girl in green above you, out of your direct reach.", + "E": "The box of strawberries is now in the hands of the girl in green behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033_2_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is that box of strawberries from earlier, relative to me now?", + "answer": "Where is the box of strawberries that was just here, relative to your position now?", + "question_chinese": "刚刚的那盒草莓现在在相对我的哪个位置？", + "answer_chinese": "那盒草莓在我右后方的冰箱里。", + "evidence": [ + { + "description": "At this moment, the girl in green came to the sink, holding strawberries in her hand.", + "timestamps": { + "start": "00:02:55", + "end": "00:02:56" + } + }, + { + "description": "At this moment, the girl in green put the remaining strawberries back into the fridge.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:26" + } + } + ], + "question_timestamps": "00:07:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s directly in front of you.", + "B": "It’s just to your left.", + "C": "It’s a few steps behind you, on the ground near where you were standing.", + "D": "It’s on your right side now, slightly behind your shoulder as you face forward.", + "E": "Where is the box of strawberries that was just here, relative to your position now?" + }, + "answer_label": "E", + "answer_text": "Where is the box of strawberries that was just here, relative to your position now?", + "q_id": "egolife_A1_D3_033_2_T1", + "correct_answer": "Where is the box of strawberries that was just here, relative to your position now?", + "original_options": { + "A": "It’s directly in front of you.", + "B": "It’s just to your left.", + "C": "It’s a few steps behind you, on the ground near where you were standing.", + "D": "It’s on your right side now, slightly behind your shoulder as you face forward.", + "E": "Where is the box of strawberries that was just here, relative to your position now?" + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033_9", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the relative position of Coca-Cola and oyster sauce?", + "answer": "Coca-Cola and oyster sauce are close to each other, both on the side of the sink.", + "question_chinese": "可口可乐和蚝油的相对位置是怎样的？", + "answer_chinese": "可口可乐和蚝油彼此靠近，都在水槽一侧。", + "evidence": [ + { + "description": "At this point, the Coca-Cola has been staying close to the oyster sauce.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:25" + } + } + ], + "question_timestamps": "00:08:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They are far apart.", + "B": "Coca-Cola and oyster sauce are close to each other, both on the side of the sink.", + "C": "Both are under the sink.", + "D": "Coca-Cola is by the sink, but the oyster sauce is across the counter where you prepare food.", + "E": "Coca-Cola and oyster sauce are together to the right of the sink, not simply on the sink’s side." + }, + "answer_label": "B", + "answer_text": "Coca-Cola and oyster sauce are close to each other, both on the side of the sink.", + "q_id": "egolife_A1_D3_033_9", + "correct_answer": "Coca-Cola and oyster sauce are close to each other, both on the side of the sink.", + "original_options": { + "A": "They are far apart.", + "B": "Coca-Cola and oyster sauce are close to each other, both on the side of the sink.", + "C": "Both are under the sink.", + "D": "Coca-Cola is by the sink, but the oyster sauce is across the counter where you prepare food.", + "E": "Coca-Cola and oyster sauce are together to the right of the sink, not simply on the sink’s side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033_11", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the relative position of Coca-Cola and oyster sauce?", + "answer": "The oyster sauce is about one sink's distance to the front-left of the Coca-Cola.", + "question_chinese": "可口可乐和蚝油的相对位置是怎样的？", + "answer_chinese": "蚝油在可口可乐的左前方大约一个水槽的距离。", + "evidence": [ + { + "description": "At this point, the Coca-Cola has been staying close to the oyster sauce.", + "timestamps": { + "start": "00:08:19", + "end": "00:08:20" + } + }, + { + "description": "At this point, the oyster sauce fell on the ground.", + "timestamps": { + "start": "00:08:21", + "end": "00:08:22" + } + }, + { + "description": "At this point, the oyster sauce was picked up and placed to the right of the sink, in front-left of the Coca-Cola.", + "timestamps": { + "start": "00:08:23", + "end": "00:08:28" + } + } + ], + "question_timestamps": "00:10:51", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "In front-right of the Coca-Cola.", + "B": "Behind-left of the Coca-Cola.", + "C": "The oyster sauce is about one sink's distance to the front-left of the Coca-Cola.", + "D": "The oyster sauce is about one sink's distance to the front-right of the Coca-Cola.", + "E": "The oyster sauce is about half a sink's distance to the front-left of the Coca-Cola." + }, + "answer_label": "C", + "answer_text": "The oyster sauce is about one sink's distance to the front-left of the Coca-Cola.", + "q_id": "egolife_A1_D3_033_11", + "correct_answer": "The oyster sauce is about one sink's distance to the front-left of the Coca-Cola.", + "original_options": { + "A": "In front-right of the Coca-Cola.", + "B": "Behind-left of the Coca-Cola.", + "C": "The oyster sauce is about one sink's distance to the front-left of the Coca-Cola.", + "D": "The oyster sauce is about one sink's distance to the front-right of the Coca-Cola.", + "E": "The oyster sauce is about half a sink's distance to the front-left of the Coca-Cola." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033-mcq_fixed_11", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get from here to the barbecue place?", + "answer": "You should first turn around, then turn left, walk to the end, turn left again, and you will arrive at the barbecue area.", + "question_chinese": "我从这里到烧烤的位置应该怎么走？", + "answer_chinese": "你应该先向后转，再左转，走到尽头左转就到了烧烤的地方。", + "evidence": [ + { + "description": "At this point, you turn right again and go out the door.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:16" + } + }, + { + "description": "At this point, you turn right from the barbecue place, go straight until you reach the corner, and then turn right again.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:01:17", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn around, then turn right.", + "B": "Walk straight, then turn left twice.", + "C": "You should first turn around, then turn left, walk to the end, turn left again, and you will arrive at the barbecue area.", + "D": "Turn around, turn left, walk to the end, then turn right to reach the barbecue area.", + "E": "Turn around, turn left, walk halfway, turn left again, and continue past the end before stopping at the barbecue area." + }, + "answer_label": "C", + "answer_text": "You should first turn around, then turn left, walk to the end, turn left again, and you will arrive at the barbecue area.", + "q_id": "egolife_A1_D3_033-mcq_fixed_11", + "correct_answer": "You should first turn around, then turn left, walk to the end, turn left again, and you will arrive at the barbecue area.", + "original_options": { + "A": "Turn around, then turn right.", + "B": "Walk straight, then turn left twice.", + "C": "You should first turn around, then turn left, walk to the end, turn left again, and you will arrive at the barbecue area.", + "D": "Turn around, turn left, walk to the end, then turn right to reach the barbecue area.", + "E": "Turn around, turn left, walk halfway, turn left again, and continue past the end before stopping at the barbecue area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033-mcq_fixed_10", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get from here to the barbecue place?", + "answer": "You should turn around first, walk to the first door, turn left, then turn left again, and turn left at the end to reach the barbecue place.", + "question_chinese": "我从这里到烧烤的位置应该怎么走？", + "answer_chinese": "你应该先向后转，走到第一个门向左转，再左转，走到尽头左转就到了烧烤的地方。", + "evidence": [ + { + "description": "At this point, you turn right again and go out the door.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:16" + } + }, + { + "description": "At this point, you turn right from the barbecue place, go straight until you reach the corner, and then turn right again.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:11" + } + }, + { + "description": "At this point, you turn right again and reach the kitchen.", + "timestamps": { + "start": "00:01:17", + "end": "00:01:18" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Go to the second door first.", + "B": "Turn right at the first door.", + "C": "You should turn around first, walk to the first door, turn left, then turn left again, and turn left at the end to reach the barbecue place.", + "D": "Turn around, walk to the first door, turn right, then turn right again, and turn right at the end.", + "E": "Turn around, walk to the first door, turn left, then turn right, and turn left at the end to arrive." + }, + "answer_label": "C", + "answer_text": "You should turn around first, walk to the first door, turn left, then turn left again, and turn left at the end to reach the barbecue place.", + "q_id": "egolife_A1_D3_033-mcq_fixed_10", + "correct_answer": "You should turn around first, walk to the first door, turn left, then turn left again, and turn left at the end to reach the barbecue place.", + "original_options": { + "A": "Go to the second door first.", + "B": "Turn right at the first door.", + "C": "You should turn around first, walk to the first door, turn left, then turn left again, and turn left at the end to reach the barbecue place.", + "D": "Turn around, walk to the first door, turn right, then turn right again, and turn right at the end.", + "E": "Turn around, walk to the first door, turn left, then turn right, and turn left at the end to arrive." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033_9_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the fruit plate relative to me now?", + "answer": "At this moment, you placed the fruit plate on the scale on the table.", + "question_chinese": "水果盘现在在相对我的哪个位置？", + "answer_chinese": "在你后面桌子的称子上。", + "evidence": [ + { + "description": "At this moment, you placed the fruit plate on the scale on the table.", + "timestamps": { + "start": "00:07:55", + "end": "00:07:56" + } + } + ], + "question_timestamps": "00:09:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "At this moment, you placed the fruit plate on the scale on the table.", + "B": "It’s on your right-hand side.", + "C": "It’s on the edge of the table.", + "D": "You left the fruit plate on the table without placing it on the scale.", + "E": "At this moment, you put the fruit plate on the counter beside the table, not on the scale." + }, + "answer_label": "A", + "answer_text": "At this moment, you placed the fruit plate on the scale on the table.", + "q_id": "egolife_A1_D3_033_9_T1", + "correct_answer": "At this moment, you placed the fruit plate on the scale on the table.", + "original_options": { + "A": "At this moment, you placed the fruit plate on the scale on the table.", + "B": "It’s on your right-hand side.", + "C": "It’s on the edge of the table.", + "D": "You left the fruit plate on the table without placing it on the scale.", + "E": "At this moment, you put the fruit plate on the counter beside the table, not on the scale." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 14, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033_4_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the fruit plate relative to me now?", + "answer": "It's on your right-hand side.", + "question_chinese": "水果盘现在在相对我的哪个位置？", + "answer_chinese": "在你右手边上。", + "evidence": [ + { + "description": "At this moment, the girl in green took the fruit plate to the sink.", + "timestamps": { + "start": "00:07:55", + "end": "00:07:56" + } + }, + { + "description": "At this moment, the girl in green took the fruit plate to the sink.", + "timestamps": { + "start": "00:09:47", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:10:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It's on your left-hand side.", + "B": "It's behind you.", + "C": "It's on your right-hand side.", + "D": "It's directly in front of you, centered rather than off to either side.", + "E": "It's slightly behind and to your left, not on your right-hand side." + }, + "answer_label": "C", + "answer_text": "It's on your right-hand side.", + "q_id": "egolife_A1_D3_033_4_T1", + "correct_answer": "It's on your right-hand side.", + "original_options": { + "A": "It's on your left-hand side.", + "B": "It's behind you.", + "C": "It's on your right-hand side.", + "D": "It's directly in front of you, centered rather than off to either side.", + "E": "It's slightly behind and to your left, not on your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033_5_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where are the tissues that were next to the sink now?", + "answer": "The tissue was on the ground, next to your right foot.", + "question_chinese": "水槽旁边的纸巾现在在哪里？", + "answer_chinese": "纸巾在地上，你的右脚边。", + "evidence": [ + { + "description": "At this moment, the paper towel around the sink is knocked by you onto the floor.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Next to your left foot.", + "B": "In front of your feet.", + "C": "The tissue was on the ground, next to your left foot rather than your right foot.", + "D": "The tissue was on the ground, next to your right foot.", + "E": "The tissue was on the ground behind you, a little away from your feet instead of beside your right foot." + }, + "answer_label": "D", + "answer_text": "The tissue was on the ground, next to your right foot.", + "q_id": "egolife_A1_D3_033_5_T1", + "correct_answer": "The tissue was on the ground, next to your right foot.", + "original_options": { + "A": "Next to your left foot.", + "B": "In front of your feet.", + "C": "The tissue was on the ground, next to your left foot rather than your right foot.", + "D": "The tissue was on the ground, next to your right foot.", + "E": "The tissue was on the ground behind you, a little away from your feet instead of beside your right foot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033_8", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the paper towel next to the sink now?", + "answer": "The paper towel is around the sink.", + "question_chinese": "水槽旁边的纸巾现在在哪里？", + "answer_chinese": "纸巾在水槽周围。", + "evidence": [ + { + "description": "At this moment, the paper towel around the sink is knocked by you onto the floor.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:22" + } + }, + { + "description": "At this moment, you picked up the paper towel that had fallen on the ground and placed it back around the sink.", + "timestamps": { + "start": "00:01:27", + "end": "00:01:28" + } + } + ], + "question_timestamps": "00:04:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The paper towel is around the sink.", + "B": "On the counter by the sink.", + "C": "In the drawer next to the sink.", + "D": "The paper towel is on the shelf above the sink, not around the sink area.", + "E": "The paper towel ended up beside the stove across from the sink instead of staying near the sink." + }, + "answer_label": "A", + "answer_text": "The paper towel is around the sink.", + "q_id": "egolife_A1_D3_033_8", + "correct_answer": "The paper towel is around the sink.", + "original_options": { + "A": "The paper towel is around the sink.", + "B": "On the counter by the sink.", + "C": "In the drawer next to the sink.", + "D": "The paper towel is on the shelf above the sink, not around the sink area.", + "E": "The paper towel ended up beside the stove across from the sink instead of staying near the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033_6_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where are the tissues that were next to the sink now?", + "answer": "The tissue was on the right side of the sink, to your front-right.", + "question_chinese": "水槽旁边的纸巾现在在哪里？", + "answer_chinese": "纸巾在你右前方的水槽右侧。", + "evidence": [ + { + "description": "At this moment, the paper towel around the sink is knocked by you onto the floor.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:22" + } + }, + { + "description": "At this moment, you picked up the paper towel that had fallen on the ground and placed it back around the sink.", + "timestamps": { + "start": "00:01:27", + "end": "00:01:28" + } + } + ], + "question_timestamps": "00:05:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the left side of the sink.", + "B": "On the front side of the sink.", + "C": "The tissue was on the right side of the sink, to your front-right.", + "D": "The tissue was on the left side of the sink, positioned to your front-left rather than front-right.", + "E": "The tissue was on the right side of the sink but to your back-right, not to your front-right." + }, + "answer_label": "C", + "answer_text": "The tissue was on the right side of the sink, to your front-right.", + "q_id": "egolife_A1_D3_033_6_T1", + "correct_answer": "The tissue was on the right side of the sink, to your front-right.", + "original_options": { + "A": "On the left side of the sink.", + "B": "On the front side of the sink.", + "C": "The tissue was on the right side of the sink, to your front-right.", + "D": "The tissue was on the left side of the sink, positioned to your front-left rather than front-right.", + "E": "The tissue was on the right side of the sink but to your back-right, not to your front-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033_2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes are left in the box next to the sink?", + "answer": "There are three mangoes left in the box next to the sink now.", + "question_chinese": "现在水槽旁边的盒子里还有多少个芒果？", + "answer_chinese": "现在水槽旁边的盒子里还有三个芒果。", + "evidence": [ + { + "description": "At this moment, there are still three mangoes in the box.", + "timestamps": { + "start": "00:07:18", + "end": "00:07:19" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two mangoes left now.", + "B": "There are four mangoes left now.", + "C": "There are three mangoes left in the box next to the sink now.", + "D": "There are five mangoes left in the box next to the sink now, not three.", + "E": "There are six mangoes left in the box beside the sink now, so the box isn’t nearly empty." + }, + "answer_label": "C", + "answer_text": "There are three mangoes left in the box next to the sink now.", + "q_id": "egolife_A1_D3_033_2", + "correct_answer": "There are three mangoes left in the box next to the sink now.", + "original_options": { + "A": "There are two mangoes left now.", + "B": "There are four mangoes left now.", + "C": "There are three mangoes left in the box next to the sink now.", + "D": "There are five mangoes left in the box next to the sink now, not three.", + "E": "There are six mangoes left in the box beside the sink now, so the box isn’t nearly empty." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_033", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_033_1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes are left in the box next to the sink?", + "answer": "There are twe mangoes left in the box next to the sink now.", + "question_chinese": "现在水槽旁边的盒子里还有多少个芒果？", + "answer_chinese": "现在水槽旁边的盒子里还有两个芒果。", + "evidence": [ + { + "description": "At this moment, you took one of the three mangoes from the box, leaving two.", + "timestamps": { + "start": "00:07:19", + "end": "00:07:23" + } + }, + { + "description": "At this moment, it is clear that there are two mangoes left in the box.", + "timestamps": { + "start": "00:08:16", + "end": "00:08:17" + } + } + ], + "question_timestamps": "00:10:21", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There is one mango left now.", + "B": "There are three mangoes left now.", + "C": "There are five mangoes left in the box next to the sink now, not two.", + "D": "There are twe mangoes left in the box next to the sink now.", + "E": "There are eight mangoes left in the box beside the sink now, so you still have plenty." + }, + "answer_label": "D", + "answer_text": "There are twe mangoes left in the box next to the sink now.", + "q_id": "egolife_A1_D3_033_1", + "correct_answer": "There are twe mangoes left in the box next to the sink now.", + "original_options": { + "A": "There is one mango left now.", + "B": "There are three mangoes left now.", + "C": "There are five mangoes left in the box next to the sink now, not two.", + "D": "There are twe mangoes left in the box next to the sink now.", + "E": "There are eight mangoes left in the box beside the sink now, so you still have plenty." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D3_034-mcq.json b/QAs/egolife_A1_D3_034-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7a4eb13ff0591adcbbe85a82c2d851c9bcaaf2d1 --- /dev/null +++ b/QAs/egolife_A1_D3_034-mcq.json @@ -0,0 +1,897 @@ +[ + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034_9", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the knife come to be in its current position?", + "answer": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.", + "question_chinese": "刀经过怎样的途径来到现在的位置？", + "answer_chinese": "从厨房的切菜板上被我拿到了身后约两米处的餐桌上。", + "evidence": [ + { + "description": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:24" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.", + "B": "You moved it to the counter.", + "C": "You took it from the drawer.", + "D": "You picked up the knife from the cutting board in the kitchen and set it on the counter in front of you.", + "E": "You carried the knife from the kitchen countertop to the dining table about two meters in front of you." + }, + "answer_label": "A", + "answer_text": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.", + "q_id": "egolife_A1_D3_034_9", + "correct_answer": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.", + "original_options": { + "A": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.", + "B": "You moved it to the counter.", + "C": "You took it from the drawer.", + "D": "You picked up the knife from the cutting board in the kitchen and set it on the counter in front of you.", + "E": "You carried the knife from the kitchen countertop to the dining table about two meters in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034_10", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the knife arrive at its current position?", + "answer": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.and then was taken by me from the dining table back to the kitchen sink.", + "question_chinese": "刀经过怎样的途径来到现在的位置？", + "answer_chinese": "刀从厨房的切菜板上被我拿到了身后约两米处的餐桌上，从餐桌上又被我拿回了厨房水槽里。", + "evidence": [ + { + "description": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:24" + } + }, + { + "description": "The knife was taken by me from the dining table back to the kitchen sink.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You moved it only once.", + "B": "Someone else carried it.", + "C": "You took it from the cutting board to the dining table behind you, then brought it back to the counter.", + "D": "You carried it from the cutting board to a countertop behind you, then brought it back and left it at the sink.", + "E": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.and then was taken by me from the dining table back to the kitchen sink." + }, + "answer_label": "E", + "answer_text": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.and then was taken by me from the dining table back to the kitchen sink.", + "q_id": "egolife_A1_D3_034_10", + "correct_answer": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.and then was taken by me from the dining table back to the kitchen sink.", + "original_options": { + "A": "You moved it only once.", + "B": "Someone else carried it.", + "C": "You took it from the cutting board to the dining table behind you, then brought it back to the counter.", + "D": "You carried it from the cutting board to a countertop behind you, then brought it back and left it at the sink.", + "E": "The knife was taken by me from the cutting board in the kitchen to the dining table about two meters behind me.and then was taken by me from the dining table back to the kitchen sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 20, + 22 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many layers of cream have I applied so far?", + "answer": "You have applied two layers of cream in total so far.", + "question_chinese": "我到现在为止一共涂了几层奶油？", + "answer_chinese": "你到现在为止一共涂了两层奶油。", + "evidence": [ + { + "description": "Now you have finished applying a layer of cream.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:49" + } + }, + { + "description": "Now you have finished applying two layers of cream.", + "timestamps": { + "start": "00:03:31", + "end": "00:03:33" + } + } + ], + "question_timestamps": "00:04:29", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have applied one layer so far.", + "B": "You have applied three layers so far.", + "C": "You have applied four layers of cream in total so far, not two, based on your repeated spreading.", + "D": "You have applied five layers of cream in total so far, with multiple passes covering the same area again.", + "E": "You have applied two layers of cream in total so far." + }, + "answer_label": "E", + "answer_text": "You have applied two layers of cream in total so far.", + "q_id": "egolife_A1_D3_034_5", + "correct_answer": "You have applied two layers of cream in total so far.", + "original_options": { + "A": "You have applied one layer so far.", + "B": "You have applied three layers so far.", + "C": "You have applied four layers of cream in total so far, not two, based on your repeated spreading.", + "D": "You have applied five layers of cream in total so far, with multiple passes covering the same area again.", + "E": "You have applied two layers of cream in total so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034_1_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many layers of cream have I applied in total?", + "answer": "You have spread two layers of cream in total so far. The cream is in front of you.", + "question_chinese": "我到现在为止一共涂了几层奶油？", + "answer_chinese": "你到现在为止一共涂了两层奶油。奶油在你的前面。", + "evidence": [ + { + "description": "Now you have finished applying a layer of cream.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:49" + } + }, + { + "description": "Now you have finished applying two layers of cream.", + "timestamps": { + "start": "00:03:31", + "end": "00:03:33" + } + } + ], + "question_timestamps": "00:04:31", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have spread one layer so far.", + "B": "You have spread three layers so far.", + "C": "You have spread two layers of cream in total so far. The cream is in front of you.", + "D": "You have spread zero layers of cream in total so far. The cream is in front of you.", + "E": "You have spread four layers of cream in total so far. The cream is in front of you." + }, + "answer_label": "C", + "answer_text": "You have spread two layers of cream in total so far. The cream is in front of you.", + "q_id": "egolife_A1_D3_034_1_T1", + "correct_answer": "You have spread two layers of cream in total so far. The cream is in front of you.", + "original_options": { + "A": "You have spread one layer so far.", + "B": "You have spread three layers so far.", + "C": "You have spread two layers of cream in total so far. The cream is in front of you.", + "D": "You have spread zero layers of cream in total so far. The cream is in front of you.", + "E": "You have spread four layers of cream in total so far. The cream is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034_1_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many layers of cream have I applied in total?", + "answer": "You have spread four layers of cream in total so far. The cream is in front of you.", + "question_chinese": "我到现在为止一共涂了几层奶油？", + "answer_chinese": "你到现在为止一共涂了四层奶油。奶油在你的前面。", + "evidence": [ + { + "description": "Now you have finished applying a layer of cream.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:50" + } + }, + { + "description": "Now, you have finished applying two layers of cream.", + "timestamps": { + "start": "00:03:31", + "end": "00:03:33" + } + }, + { + "description": "Now, you have finished applying three layers of cream.", + "timestamps": { + "start": "00:04:31", + "end": "00:04:33" + } + }, + { + "description": "Now, you have finished applying four layers of cream.", + "timestamps": { + "start": "00:07:22", + "end": "00:07:24" + } + } + ], + "question_timestamps": "00:10:22", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have spread three layers so far.", + "B": "You have spread five layers so far.", + "C": "You have spread two layers of cream in total so far. The cream is in front of you.", + "D": "You have spread four layers of cream in total so far. The cream is in front of you.", + "E": "You have spread six layers of cream in total so far. The cream is in front of you." + }, + "answer_label": "D", + "answer_text": "You have spread four layers of cream in total so far. The cream is in front of you.", + "q_id": "egolife_A1_D3_034_1_T2", + "correct_answer": "You have spread four layers of cream in total so far. The cream is in front of you.", + "original_options": { + "A": "You have spread three layers so far.", + "B": "You have spread five layers so far.", + "C": "You have spread two layers of cream in total so far. The cream is in front of you.", + "D": "You have spread four layers of cream in total so far. The cream is in front of you.", + "E": "You have spread six layers of cream in total so far. The cream is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034-mcq_fixed_3", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the bench get here?", + "answer": "You started from the dining table, walked halfway around it into the room, picked up a stool near the computer, and placed it near the dining table .", + "question_chinese": "板凳是经过怎样的途径拿过来的？", + "answer_chinese": "你从餐桌出发，绕桌子半圈走到房间里，在电脑附近拿起板凳到了餐桌附近.", + "evidence": [ + { + "description": "At this moment, you set off from the dining table and walked halfway around it into the room.", + "timestamps": { + "start": "00:04:04", + "end": "00:04:11" + } + }, + { + "description": "At this moment, you picked up the stool from beside the computer and walked back to sit beside the dining table.", + "timestamps": { + "start": "00:04:14", + "end": "00:04:23" + } + } + ], + "question_timestamps": "00:04:24", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You carried it from the living room.", + "B": "You dragged it over from the hallway.", + "C": "You started at the dining table, walked around it, picked up a stool near the computer, and set it by the window.", + "D": "You began near the computer, lifted the stool, walked around the dining table, and placed it back near the computer.", + "E": "You started from the dining table, walked halfway around it into the room, picked up a stool near the computer, and placed it near the dining table ." + }, + "answer_label": "E", + "answer_text": "You started from the dining table, walked halfway around it into the room, picked up a stool near the computer, and placed it near the dining table .", + "q_id": "egolife_A1_D3_034-mcq_fixed_3", + "correct_answer": "You started from the dining table, walked halfway around it into the room, picked up a stool near the computer, and placed it near the dining table .", + "original_options": { + "A": "You carried it from the living room.", + "B": "You dragged it over from the hallway.", + "C": "You started at the dining table, walked around it, picked up a stool near the computer, and set it by the window.", + "D": "You began near the computer, lifted the stool, walked around the dining table, and placed it back near the computer.", + "E": "You started from the dining table, walked halfway around it into the room, picked up a stool near the computer, and placed it near the dining table ." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 22, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034-mcq_fixed_2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the bench get here?", + "answer": "You started from the dining table, walked halfway around it into the room, picked up a stool near the computer, and placed it near the dining table before sitting down.", + "question_chinese": "板凳是经过怎样的途径拿过来的？", + "answer_chinese": "你从餐桌出发，绕桌子半圈走到房间里，在电脑附近拿过来板凳并放到了餐桌附近坐下。", + "evidence": [ + { + "description": "At this moment, you set off from the dining table and walked halfway around it into the room.", + "timestamps": { + "start": "00:04:04", + "end": "00:04:11" + } + }, + { + "description": "At this moment, you picked up the stool from beside the computer and walked back to sit beside the dining table.", + "timestamps": { + "start": "00:04:14", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:07:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You carried it from the kitchen.", + "B": "You dragged it in from the hallway.", + "C": "You walked around the dining table, picked up a chair near the computer, and set it beside the table.", + "D": "You started from the dining table, walked halfway around it into the room, picked up a stool near the computer, and placed it near the dining table before sitting down.", + "E": "You started at the dining table, grabbed a stool by the computer, and placed it near the couch before sitting down." + }, + "answer_label": "D", + "answer_text": "You started from the dining table, walked halfway around it into the room, picked up a stool near the computer, and placed it near the dining table before sitting down.", + "q_id": "egolife_A1_D3_034-mcq_fixed_2", + "correct_answer": "You started from the dining table, walked halfway around it into the room, picked up a stool near the computer, and placed it near the dining table before sitting down.", + "original_options": { + "A": "You carried it from the kitchen.", + "B": "You dragged it in from the hallway.", + "C": "You walked around the dining table, picked up a chair near the computer, and set it beside the table.", + "D": "You started from the dining table, walked halfway around it into the room, picked up a stool near the computer, and placed it near the dining table before sitting down.", + "E": "You started at the dining table, grabbed a stool by the computer, and placed it near the couch before sitting down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034_7", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates are there on the dining table next to the kitchen?", + "answer": "At this moment, there are two plates on the dining table in total.", + "question_chinese": "此时厨房旁边的餐桌上一共有几个盘子？", + "answer_chinese": "此时餐桌上一共有两个盘子。", + "evidence": [ + { + "description": "At this moment, there are two plates on the dining table in total. One is for fruit, and the other is for cake.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:04:29", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "At this moment, there are two plates on the dining table in total.", + "B": "There is one plate.", + "C": "There are three plates.", + "D": "At this moment, there are four plates on the dining table in total.", + "E": "At this moment, there are five plates on the dining table in total." + }, + "answer_label": "A", + "answer_text": "At this moment, there are two plates on the dining table in total.", + "q_id": "egolife_A1_D3_034_7", + "correct_answer": "At this moment, there are two plates on the dining table in total.", + "original_options": { + "A": "At this moment, there are two plates on the dining table in total.", + "B": "There is one plate.", + "C": "There are three plates.", + "D": "At this moment, there are four plates on the dining table in total.", + "E": "At this moment, there are five plates on the dining table in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034_8", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many plates are there on the dining table next to the kitchen at this moment?", + "answer": "At this moment, there are two plates on the dining table in total.", + "question_chinese": "此时厨房旁边的餐桌上一共有几个盘子？", + "answer_chinese": "此时餐桌上一共有三个盘子。", + "evidence": [ + { + "description": "At this moment, there are two plates on the dining table in total. One is for fruit, and the other is for cake.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:10" + } + }, + { + "description": "At this moment, there are three plates on the dining table in total: one for fruit, one for cake, and one for barbecue.", + "timestamps": { + "start": "00:08:38", + "end": "00:08:40" + } + } + ], + "question_timestamps": "00:10:36", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There is one plate.", + "B": "At this moment, there are two plates on the dining table in total.", + "C": "There are three plates.", + "D": "At this moment, there are four plates on the dining table in total.", + "E": "At this moment, there are five plates on the dining table in total." + }, + "answer_label": "B", + "answer_text": "At this moment, there are two plates on the dining table in total.", + "q_id": "egolife_A1_D3_034_8", + "correct_answer": "At this moment, there are two plates on the dining table in total.", + "original_options": { + "A": "There is one plate.", + "B": "At this moment, there are two plates on the dining table in total.", + "C": "There are three plates.", + "D": "At this moment, there are four plates on the dining table in total.", + "E": "At this moment, there are five plates on the dining table in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034_3_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many plates in total do I see on the dining table next to the kitchen?", + "answer": "Right now, you see one plate in total on the dining table. The plate is to your front-right.", + "question_chinese": "此时我看见厨房旁边的餐桌上一共有几个盘子？", + "answer_chinese": "此时你看见餐桌上一共有一个盘子。盘子在你的右前方。", + "evidence": [ + { + "description": "At this moment, there is a plate on the dining table in total.", + "timestamps": { + "start": "00:00:43", + "end": "00:01:28" + } + } + ], + "question_timestamps": "00:01:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Right now, you see one plate in total on the dining table. The plate is to your front-right.", + "B": "You see zero plates.", + "C": "You see two plates.", + "D": "Right now, you see three plates in total on the dining table, and they are to your front-right.", + "E": "Right now, you see four plates in total on the dining table, with the plates positioned to your front-right." + }, + "answer_label": "A", + "answer_text": "Right now, you see one plate in total on the dining table. The plate is to your front-right.", + "q_id": "egolife_A1_D3_034_3_T2", + "correct_answer": "Right now, you see one plate in total on the dining table. The plate is to your front-right.", + "original_options": { + "A": "Right now, you see one plate in total on the dining table. The plate is to your front-right.", + "B": "You see zero plates.", + "C": "You see two plates.", + "D": "Right now, you see three plates in total on the dining table, and they are to your front-right.", + "E": "Right now, you see four plates in total on the dining table, with the plates positioned to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034_3_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many plates in total do I see on the dining table next to the kitchen?", + "answer": "Right now, you see three plates in total on the dining table. The plates are in front of you.", + "question_chinese": "此时我看见厨房旁边的餐桌上一共有几个盘子？", + "answer_chinese": "此时你看见餐桌上一共有三个盘子。盘子在你的前面。", + "evidence": [ + { + "description": "At this moment, there are two plates on the dining table in total. One is for fruit, and the other is for cake.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:10" + } + }, + { + "description": "At this moment, there are three plates on the dining table in total: one for fruit, one for cake, and one for barbecue.", + "timestamps": { + "start": "00:08:38", + "end": "00:08:40" + } + } + ], + "question_timestamps": "00:10:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You see one plate.", + "B": "You see two plates.", + "C": "Right now, you see four plates in total on the dining table, and the plates are in front of you.", + "D": "Right now, you see five plates in total on the dining table, with the plates directly in front of you.", + "E": "Right now, you see three plates in total on the dining table. The plates are in front of you." + }, + "answer_label": "E", + "answer_text": "Right now, you see three plates in total on the dining table. The plates are in front of you.", + "q_id": "egolife_A1_D3_034_3_T1", + "correct_answer": "Right now, you see three plates in total on the dining table. The plates are in front of you.", + "original_options": { + "A": "You see one plate.", + "B": "You see two plates.", + "C": "Right now, you see four plates in total on the dining table, and the plates are in front of you.", + "D": "Right now, you see five plates in total on the dining table, with the plates directly in front of you.", + "E": "Right now, you see three plates in total on the dining table. The plates are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034_2_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many bowls do I see on the dining table next to the kitchen?", + "answer": "Right now, you don’t see a single bowl on the dining table. The dining table is to your front-right.", + "question_chinese": "此时我看见厨房旁边的餐桌上有几个碗？", + "answer_chinese": "此时你看见餐桌上没有一个碗。餐桌在你的右前方。", + "evidence": [ + { + "description": "At this moment, there is no plate on the dining table in total.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:52" + } + } + ], + "question_timestamps": "00:03:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You see one bowl.", + "B": "You see two bowls.", + "C": "Right now, you see three bowls on the dining table, and the bowls are to your front-right.", + "D": "Right now, you don’t see a single bowl on the dining table. The dining table is to your front-right.", + "E": "Right now, you see four bowls on the dining table, with the bowls positioned to your front-right." + }, + "answer_label": "D", + "answer_text": "Right now, you don’t see a single bowl on the dining table. The dining table is to your front-right.", + "q_id": "egolife_A1_D3_034_2_T1", + "correct_answer": "Right now, you don’t see a single bowl on the dining table. The dining table is to your front-right.", + "original_options": { + "A": "You see one bowl.", + "B": "You see two bowls.", + "C": "Right now, you see three bowls on the dining table, and the bowls are to your front-right.", + "D": "Right now, you don’t see a single bowl on the dining table. The dining table is to your front-right.", + "E": "Right now, you see four bowls on the dining table, with the bowls positioned to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034_2_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many bowls do I see on the dining table next to the kitchen?", + "answer": "Right now, you see one bowl on the dining table. The bowl is to your front-right.", + "question_chinese": "此时我看见厨房旁边的餐桌上有几个碗？", + "answer_chinese": "此时你看见餐桌上有一个碗。碗在你的右前方。", + "evidence": [ + { + "description": "At this moment, there is one bowl on the dining table.", + "timestamps": { + "start": "00:08:41", + "end": "00:08:50" + } + } + ], + "question_timestamps": "00:09:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You see zero bowls.", + "B": "You see two bowls.", + "C": "Right now, you see three bowls on the dining table, and the bowls are to your front-right.", + "D": "Right now, you see four bowls on the dining table, with the bowls positioned to your front-right.", + "E": "Right now, you see one bowl on the dining table. The bowl is to your front-right." + }, + "answer_label": "E", + "answer_text": "Right now, you see one bowl on the dining table. The bowl is to your front-right.", + "q_id": "egolife_A1_D3_034_2_T2", + "correct_answer": "Right now, you see one bowl on the dining table. The bowl is to your front-right.", + "original_options": { + "A": "You see zero bowls.", + "B": "You see two bowls.", + "C": "Right now, you see three bowls on the dining table, and the bowls are to your front-right.", + "D": "Right now, you see four bowls on the dining table, with the bowls positioned to your front-right.", + "E": "Right now, you see one bowl on the dining table. The bowl is to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034-mcq_fixed_7", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What fruit is on the first layer of the cake?", + "answer": "The first layer of the cake has not had any fruit placed on it so far.", + "question_chinese": "第一层蛋糕放的是什么水果？", + "answer_chinese": "第一层蛋糕到目前为止还没有放任何水果。", + "evidence": [ + { + "description": "During this period, you are piping the cream, and have not yet reached the step of adding fruit.", + "timestamps": { + "start": "00:01:31", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Strawberries are on the first layer.", + "B": "Blueberries are on the first layer.", + "C": "The first layer has thin kiwi slices laid near the center, partly covered by frosting.", + "D": "The first layer of the cake has not had any fruit placed on it so far.", + "E": "The first layer has a few strawberry slices arranged along the edge, with some blueberries tucked beside them." + }, + "answer_label": "D", + "answer_text": "The first layer of the cake has not had any fruit placed on it so far.", + "q_id": "egolife_A1_D3_034-mcq_fixed_7", + "correct_answer": "The first layer of the cake has not had any fruit placed on it so far.", + "original_options": { + "A": "Strawberries are on the first layer.", + "B": "Blueberries are on the first layer.", + "C": "The first layer has thin kiwi slices laid near the center, partly covered by frosting.", + "D": "The first layer of the cake has not had any fruit placed on it so far.", + "E": "The first layer has a few strawberry slices arranged along the edge, with some blueberries tucked beside them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_034", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_034-mcq_fixed_6", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What fruit is on the first layer of the cake?", + "answer": "The first layer of the cake has mango on it.", + "question_chinese": "第一层蛋糕放的是什么水果？", + "answer_chinese": "第一层蛋糕放的是芒果。", + "evidence": [ + { + "description": "At this moment, the girl placed the mango from her plate onto the cream of the first layer of the cake.", + "timestamps": { + "start": "00:02:27", + "end": "00:02:29" + } + } + ], + "question_timestamps": "00:08:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The first layer of the cake has mango on it.", + "B": "The first layer has strawberries on it.", + "C": "The first layer has blueberries on it.", + "D": "The first layer of the cake is topped with peach slices arranged neatly across the surface.", + "E": "The first layer of the cake has papaya pieces scattered on it for a bright orange look." + }, + "answer_label": "A", + "answer_text": "The first layer of the cake has mango on it.", + "q_id": "egolife_A1_D3_034-mcq_fixed_6", + "correct_answer": "The first layer of the cake has mango on it.", + "original_options": { + "A": "The first layer of the cake has mango on it.", + "B": "The first layer has strawberries on it.", + "C": "The first layer has blueberries on it.", + "D": "The first layer of the cake is topped with peach slices arranged neatly across the surface.", + "E": "The first layer of the cake has papaya pieces scattered on it for a bright orange look." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D3_035-mcq.json b/QAs/egolife_A1_D3_035-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e36dc64122cb8e7ec50b145ab5a4957e5cd1445d --- /dev/null +++ b/QAs/egolife_A1_D3_035-mcq.json @@ -0,0 +1,606 @@ +[ + { + "video_uid": "egolife_A1_D3_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_035-mcq_fixed_6", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many strawberries were placed on top of the cake just now?", + "answer": "One.", + "question_chinese": "刚刚的蛋糕顶层放了几个草莓？", + "answer_chinese": "蛋糕顶层放了三个草莓。", + "evidence": [ + { + "description": "At this moment, the girl placed three strawberries from the plate onto the cake.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:39" + } + } + ], + "question_timestamps": "00:01:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four strawberries were placed on top of the cake during that moment, not just a single one.", + "D": "Five strawberries were placed on top of the cake just now, spread out across the surface.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_D3_035-mcq_fixed_6", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four strawberries were placed on top of the cake during that moment, not just a single one.", + "D": "Five strawberries were placed on top of the cake just now, spread out across the surface.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_035-mcq_fixed_7", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many strawberries were placed on top of the cake just now?", + "answer": "Three.", + "question_chinese": "刚刚的蛋糕顶层放了几个草莓？", + "answer_chinese": "蛋糕顶层放了六个草莓。", + "evidence": [ + { + "description": "At this moment, the girl placed all six strawberries from the plate onto the cake.", + "timestamps": { + "start": "00:01:26", + "end": "00:01:46" + } + } + ], + "question_timestamps": "00:09:54", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "Six strawberries were placed on top of the cake just now, not only three in total.", + "D": "Seven strawberries were placed on top of the cake just now, covering most of the top.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "egolife_A1_D3_035-mcq_fixed_7", + "correct_answer": "Three.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "Six strawberries were placed on top of the cake just now, not only three in total.", + "D": "Seven strawberries were placed on top of the cake just now, covering most of the top.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_035_3", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I go from sitting at the dining table to being in this position?", + "answer": "You turned around from beside the dining table, walked straight out the door, and arrived at your current position.", + "question_chinese": "我是怎么从餐桌旁到现在这个位置？", + "answer_chinese": "你从餐桌旁向后转直走出门到了现在的位置。", + "evidence": [ + { + "description": "Right now, you turn around from beside the dining table.", + "timestamps": { + "start": "00:11:31", + "end": "00:11:36" + } + }, + { + "description": "Right now, you walk straight forward three or four steps and go out the door.", + "timestamps": { + "start": "00:11:37", + "end": "00:11:38" + } + } + ], + "question_timestamps": "00:11:42", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You walked through the hallway to get here.", + "B": "You headed toward the sofa and arrived here.", + "C": "You stepped away from the dining table, looped around the coffee table, then went through the hallway to your spot.", + "D": "You turned around from beside the dining table, walked straight out the door, and arrived at your current position.", + "E": "You turned from beside the armchair, walked into the kitchen area, and then continued on to your current position." + }, + "answer_label": "D", + "answer_text": "You turned around from beside the dining table, walked straight out the door, and arrived at your current position.", + "q_id": "egolife_A1_D3_035_3", + "correct_answer": "You turned around from beside the dining table, walked straight out the door, and arrived at your current position.", + "original_options": { + "A": "You walked through the hallway to get here.", + "B": "You headed toward the sofa and arrived here.", + "C": "You stepped away from the dining table, looped around the coffee table, then went through the hallway to your spot.", + "D": "You turned around from beside the dining table, walked straight out the door, and arrived at your current position.", + "E": "You turned from beside the armchair, walked into the kitchen area, and then continued on to your current position." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 20, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_035_4", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I go from sitting at the dining table to being in this position?", + "answer": "You turned around from beside the dining table, walked straight out the door, immediately turned left, and kept walking straight until you reached the table at this moment.", + "question_chinese": "我是怎么从餐桌旁到现在这个位置？", + "answer_chinese": "你从餐桌旁向后转直走出门立即左转直走一直到此时的桌子旁。", + "evidence": [ + { + "description": "Right now, you turn around from beside the dining table.", + "timestamps": { + "start": "00:11:31", + "end": "00:11:36" + } + }, + { + "description": "you walk straight forward three or four steps and go out the door.", + "timestamps": { + "start": "00:11:37", + "end": "00:11:38" + } + }, + { + "description": "you turn left immediately and walk straight to the table.", + "timestamps": { + "start": "00:11:40", + "end": "00:11:49" + } + } + ], + "question_timestamps": "00:11:53", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turned right after exiting the door.", + "B": "You turned left after a few steps outside.", + "C": "You left the dining table, went out the door, immediately turned right, and kept walking straight to this table.", + "D": "You walked out from the dining table, passed the doorway, then turned left after a short distance and continued until this table.", + "E": "You turned around from beside the dining table, walked straight out the door, immediately turned left, and kept walking straight until you reached the table at this moment." + }, + "answer_label": "E", + "answer_text": "You turned around from beside the dining table, walked straight out the door, immediately turned left, and kept walking straight until you reached the table at this moment.", + "q_id": "egolife_A1_D3_035_4", + "correct_answer": "You turned around from beside the dining table, walked straight out the door, immediately turned left, and kept walking straight until you reached the table at this moment.", + "original_options": { + "A": "You turned right after exiting the door.", + "B": "You turned left after a few steps outside.", + "C": "You left the dining table, went out the door, immediately turned right, and kept walking straight to this table.", + "D": "You walked out from the dining table, passed the doorway, then turned left after a short distance and continued until this table.", + "E": "You turned around from beside the dining table, walked straight out the door, immediately turned left, and kept walking straight until you reached the table at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 22 + ] + }, + { + "video_uid": "egolife_A1_D3_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_035-mcq_fixed_8", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the skewers on the plate on the dining table now?", + "answer": "You can reach the meat skewers on the plate; it's about one plate's distance to your left.", + "question_chinese": "我现在可以够到餐桌上的盘子里的肉串吗？", + "answer_chinese": "你可以够到盘子里的肉串，它就在你左手边大约一个盘子的距离。", + "evidence": [ + { + "description": "it's about one plate's distance to your left.", + "timestamps": { + "start": "00:07:30", + "end": "00:07:32" + } + } + ], + "question_timestamps": "00:07:33", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot reach the meat skewers on the plate; it's about one plate's distance to your left.", + "B": "You can reach the meat skewers on the plate; it's about one plate's distance to your left." + }, + "answer_label": "B", + "answer_text": "You can reach the meat skewers on the plate; it's about one plate's distance to your left.", + "q_id": "egolife_A1_D3_035-mcq_fixed_8", + "correct_answer": "You can reach the meat skewers on the plate; it's about one plate's distance to your left.", + "original_options": { + "A": "You cannot reach the meat skewers on the plate; it's about one plate's distance to your left.", + "B": "You can reach the meat skewers on the plate; it's about one plate's distance to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_035-mcq_fixed_9", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the skewers on the plate on the dining table now?", + "answer": "No, you cannot reach the skewers on the plate.", + "question_chinese": "我现在可以够到餐桌上的盘子里的肉串吗？", + "answer_chinese": "不，你不能够到盘子里的肉串。", + "evidence": [ + { + "description": "At this moment, you have arrived at the kitchen, approximately three meters away from the skewers on the table.", + "timestamps": { + "start": "00:09:22", + "end": "00:09:23" + } + } + ], + "question_timestamps": "00:10:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you cannot reach the skewers on the plate.", + "B": "Yes, you can reach the skewers on the plate." + }, + "answer_label": "A", + "answer_text": "No, you cannot reach the skewers on the plate.", + "q_id": "egolife_A1_D3_035-mcq_fixed_9", + "correct_answer": "No, you cannot reach the skewers on the plate.", + "original_options": { + "A": "No, you cannot reach the skewers on the plate.", + "B": "Yes, you can reach the skewers on the plate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_D3_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_035_7", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path should I take to get something from the refrigerator?", + "answer": "You should turn right from your current position, walk straight for two steps, and the fridge is behind the door.", + "question_chinese": "我现在要从冰箱里拿东西应该走什么路径？", + "answer_chinese": "你应该从当前位置右转直走两步，冰箱就在门后。", + "evidence": [ + { + "description": "At this moment, the refrigerator can be seen opposite the dining table, behind the door.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:29" + } + }, + { + "description": "At this moment, the refrigerator can be seen opposite the dining table, behind the door.", + "timestamps": { + "start": "00:09:58", + "end": "00:09:59" + } + } + ], + "question_timestamps": "00:10:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You should turn right from your current position, walk straight for two steps, and the fridge is behind the door.", + "B": "Turn left and take two steps.", + "C": "Turn right and take three steps.", + "D": "You should turn left from where you are, walk straight two steps, and the fridge will be behind the counter.", + "E": "You should turn around from your current position, walk straight for two steps, and the fridge is behind the table." + }, + "answer_label": "A", + "answer_text": "You should turn right from your current position, walk straight for two steps, and the fridge is behind the door.", + "q_id": "egolife_A1_D3_035_7", + "correct_answer": "You should turn right from your current position, walk straight for two steps, and the fridge is behind the door.", + "original_options": { + "A": "You should turn right from your current position, walk straight for two steps, and the fridge is behind the door.", + "B": "Turn left and take two steps.", + "C": "Turn right and take three steps.", + "D": "You should turn left from where you are, walk straight two steps, and the fridge will be behind the counter.", + "E": "You should turn around from your current position, walk straight for two steps, and the fridge is behind the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_035_8", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path should I take to get something from the refrigerator?", + "answer": "You should turn around and walk straight for two steps; the fridge will be on your left.", + "question_chinese": "我现在要从冰箱里拿东西应该走什么路径？", + "answer_chinese": "你应该向后转直走两步，冰箱就在你的左手边。", + "evidence": [ + { + "description": "At this moment, the refrigerator can be seen opposite the dining table, behind the door.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:29" + } + }, + { + "description": "At this moment, the refrigerator can be seen opposite the dining table, behind the door.", + "timestamps": { + "start": "00:09:58", + "end": "00:09:59" + } + } + ], + "question_timestamps": "00:11:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn around and take three steps.", + "B": "You should turn around and walk straight for two steps; the fridge will be on your left.", + "C": "Turn around; the fridge is on your right.", + "D": "Turn around, walk straight for one step, and the refrigerator should be directly in front of you.", + "E": "Turn around, walk forward for four steps, and you will find the refrigerator on your left side." + }, + "answer_label": "B", + "answer_text": "You should turn around and walk straight for two steps; the fridge will be on your left.", + "q_id": "egolife_A1_D3_035_8", + "correct_answer": "You should turn around and walk straight for two steps; the fridge will be on your left.", + "original_options": { + "A": "Turn around and take three steps.", + "B": "You should turn around and walk straight for two steps; the fridge will be on your left.", + "C": "Turn around; the fridge is on your right.", + "D": "Turn around, walk straight for one step, and the refrigerator should be directly in front of you.", + "E": "Turn around, walk forward for four steps, and you will find the refrigerator on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_035_9", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the plate with the cake end up in this position?", + "answer": "You took the plate from the girl, turned around, and walked two steps to the stove.", + "question_chinese": "装有蛋糕的盘子是怎么到现在这个位置的？", + "answer_chinese": "你从女孩手里接过盘子，转身走了两步到达了灶台。", + "evidence": [ + { + "description": "The girl passed the plate to you, who were sitting beside the dining table.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:14" + } + }, + { + "description": "You picked up the plate, turned around, walked straight to one end of the stove, and put the plate down.", + "timestamps": { + "start": "00:09:15", + "end": "00:09:23" + } + } + ], + "question_timestamps": "00:09:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You carried it to the table.", + "B": "You walked three steps to the stove.", + "C": "You took the plate from the girl, turned around, and walked two steps to the stove.", + "D": "You took the plate from the girl, turned around, and walked one step before setting it on the stove.", + "E": "You accepted the plate, turned around, and walked two steps toward the refrigerator instead of the stove." + }, + "answer_label": "C", + "answer_text": "You took the plate from the girl, turned around, and walked two steps to the stove.", + "q_id": "egolife_A1_D3_035_9", + "correct_answer": "You took the plate from the girl, turned around, and walked two steps to the stove.", + "original_options": { + "A": "You carried it to the table.", + "B": "You walked three steps to the stove.", + "C": "You took the plate from the girl, turned around, and walked two steps to the stove.", + "D": "You took the plate from the girl, turned around, and walked one step before setting it on the stove.", + "E": "You accepted the plate, turned around, and walked two steps toward the refrigerator instead of the stove." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_035", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_035_10", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the plate with the cake end up in this position?", + "answer": "You took the plate from the girl, turned around, and walked two steps to the stove. Then you turned again and placed the plate at one end of the dining table.", + "question_chinese": "装有蛋糕的盘子是怎么到现在这个位置的？", + "answer_chinese": "你从女孩手里接过盘子，转身走了两步到达了灶台。然后又转身将盘子放到了餐桌一端。", + "evidence": [ + { + "description": "The girl passed the plate to you, who were sitting beside the dining table.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:14" + } + }, + { + "description": "You picked up the plate, turned around, walked straight to one end of the stove, and put the plate down.", + "timestamps": { + "start": "00:09:15", + "end": "00:09:24" + } + }, + { + "description": "You picked up the plate from the stove, turned around, and placed it at one end of the dining table behind you.", + "timestamps": { + "start": "00:09:25", + "end": "00:09:33" + } + } + ], + "question_timestamps": "00:09:37", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You left it on the stove.", + "B": "You placed it at the table center.", + "C": "You took the plate from the girl, turned around, walked three steps to the stove, then set it at the table center.", + "D": "You took the plate, turned around, walked two steps to the counter, then placed it at the corner of the dining table.", + "E": "You took the plate from the girl, turned around, and walked two steps to the stove. Then you turned again and placed the plate at one end of the dining table." + }, + "answer_label": "E", + "answer_text": "You took the plate from the girl, turned around, and walked two steps to the stove. Then you turned again and placed the plate at one end of the dining table.", + "q_id": "egolife_A1_D3_035_10", + "correct_answer": "You took the plate from the girl, turned around, and walked two steps to the stove. Then you turned again and placed the plate at one end of the dining table.", + "original_options": { + "A": "You left it on the stove.", + "B": "You placed it at the table center.", + "C": "You took the plate from the girl, turned around, walked three steps to the stove, then set it at the table center.", + "D": "You took the plate, turned around, walked two steps to the counter, then placed it at the corner of the dining table.", + "E": "You took the plate from the girl, turned around, and walked two steps to the stove. Then you turned again and placed the plate at one end of the dining table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 22, + 22 + ] + } +] diff --git a/QAs/egolife_A1_D3_036-mcq.json b/QAs/egolife_A1_D3_036-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b9984ab4b8603aa9457755524a4c4ff5ae226b9f --- /dev/null +++ b/QAs/egolife_A1_D3_036-mcq.json @@ -0,0 +1,669 @@ +[ + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did i put the chopsticks i just washed?", + "answer": "You just placed the washed chopsticks on the plate to the left of the kitchen stove.", + "question_chinese": "刚刚洗好的筷子被我放到了哪里？", + "answer_chinese": "刚刚洗好的筷子被你放到了厨房灶台左侧的盘子上。", + "evidence": [ + { + "description": "Now, you place the washed chopsticks on the plate to the left of the stove table.", + "timestamps": { + "start": "00:08:27", + "end": "00:08:37" + } + } + ], + "question_timestamps": "00:09:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just placed the washed chopsticks on the plate to the left of the kitchen stove.", + "B": "You put them in the dish rack.", + "C": "You left them on the countertop.", + "D": "You placed the washed chopsticks on a plate behind the kitchen stove, not on the left-side plate.", + "E": "You set the washed chopsticks on the plate with the spoon next to the stove, rather than the plate on the left." + }, + "answer_label": "A", + "answer_text": "You just placed the washed chopsticks on the plate to the left of the kitchen stove.", + "q_id": "egolife_A1_D3_036-mcq_fixed_6", + "correct_answer": "You just placed the washed chopsticks on the plate to the left of the kitchen stove.", + "original_options": { + "A": "You just placed the washed chopsticks on the plate to the left of the kitchen stove.", + "B": "You put them in the dish rack.", + "C": "You left them on the countertop.", + "D": "You placed the washed chopsticks on a plate behind the kitchen stove, not on the left-side plate.", + "E": "You set the washed chopsticks on the plate with the spoon next to the stove, rather than the plate on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 22 + ] + }, + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did i put the chopsticks i just washed?", + "answer": "The chopsticks you just washed have been placed by you on the table in the barbecue yard.", + "question_chinese": "刚刚洗好的筷子被我放到了哪里？", + "answer_chinese": "刚刚洗好的筷子被你放到了烧烤的院子里的桌子上。", + "evidence": [ + { + "description": "Now, you place the washed chopsticks on the plate to the left of the stove table.", + "timestamps": { + "start": "00:08:27", + "end": "00:08:31" + } + }, + { + "description": "At this moment, you have placed the chopsticks on the table in the barbecue yard.", + "timestamps": { + "start": "00:11:04", + "end": "00:11:05" + } + } + ], + "question_timestamps": "00:11:17", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You left them on the grill.", + "B": "The chopsticks you just washed have been placed by you on the table in the barbecue yard.", + "C": "You put them under a chair.", + "D": "After washing, you placed the chopsticks in the utensil holder next to the stove instead of taking them outside.", + "E": "You carried the chopsticks to the balcony and set them beside the potted plants to let them air-dry." + }, + "answer_label": "B", + "answer_text": "The chopsticks you just washed have been placed by you on the table in the barbecue yard.", + "q_id": "egolife_A1_D3_036-mcq_fixed_7", + "correct_answer": "The chopsticks you just washed have been placed by you on the table in the barbecue yard.", + "original_options": { + "A": "You left them on the grill.", + "B": "The chopsticks you just washed have been placed by you on the table in the barbecue yard.", + "C": "You put them under a chair.", + "D": "After washing, you placed the chopsticks in the utensil holder next to the stove instead of taking them outside.", + "E": "You carried the chopsticks to the balcony and set them beside the potted plants to let them air-dry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_0", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I just get the chopsticks from?", + "answer": "You have not picked up the chopsticks so far.", + "question_chinese": "我刚刚从哪里拿的筷子？", + "answer_chinese": "你到目前为止并没有拿过筷子。", + "evidence": [ + { + "description": "You have not picked up the chopsticks so far.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:01" + } + } + ], + "question_timestamps": "00:06:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You have not picked up the chopsticks so far.", + "B": "From the utensil organizer.", + "C": "From a bamboo case.", + "D": "You picked them up from the utensil organizer behind your glass of water right before sitting down.", + "E": "You lifted them from a woven bamboo case placed between the soy sauce bottle and the noodle bowl." + }, + "answer_label": "A", + "answer_text": "You have not picked up the chopsticks so far.", + "q_id": "egolife_A1_D3_036-mcq_fixed_0", + "correct_answer": "You have not picked up the chopsticks so far.", + "original_options": { + "A": "You have not picked up the chopsticks so far.", + "B": "From the utensil organizer.", + "C": "From a bamboo case.", + "D": "You picked them up from the utensil organizer behind your glass of water right before sitting down.", + "E": "You lifted them from a woven bamboo case placed between the soy sauce bottle and the noodle bowl." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I just get the chopsticks from?", + "answer": "You got the chopsticks from the cabinet in the kitchen on the second floor.", + "question_chinese": "我刚刚从哪里拿的筷子？", + "answer_chinese": "你从二楼厨房的柜子里拿的筷子。", + "evidence": [ + { + "description": "You are now going upstairs to the second floor.", + "timestamps": { + "start": "00:07:41", + "end": "00:07:53" + } + }, + { + "description": "Now, you have arrived at the kitchen on the second floor, opened the cabinet door, and taken out the chopsticks.", + "timestamps": { + "start": "00:07:55", + "end": "00:07:57" + } + } + ], + "question_timestamps": "00:11:09", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From the drawer in the second-floor kitchen.", + "B": "From a box in the second-floor kitchen.", + "C": "You took the chopsticks from a wooden box on the counter in the main kitchen near the living room.", + "D": "You got the chopsticks from the cabinet in the kitchen on the second floor.", + "E": "You grabbed the chopsticks from a storage shelf beside the dining table in the first-floor kitchen." + }, + "answer_label": "D", + "answer_text": "You got the chopsticks from the cabinet in the kitchen on the second floor.", + "q_id": "egolife_A1_D3_036-mcq_fixed_1", + "correct_answer": "You got the chopsticks from the cabinet in the kitchen on the second floor.", + "original_options": { + "A": "From the drawer in the second-floor kitchen.", + "B": "From a box in the second-floor kitchen.", + "C": "You took the chopsticks from a wooden box on the counter in the main kitchen near the living room.", + "D": "You got the chopsticks from the cabinet in the kitchen on the second floor.", + "E": "You grabbed the chopsticks from a storage shelf beside the dining table in the first-floor kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_9", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just go to the second floor to get the chopsticks?", + "answer": "No, you have not gone to the second floor to get the chopsticks so far.", + "question_chinese": "我刚刚去过二楼拿筷子吗？", + "answer_chinese": "没有，你到目前为止没有去过二楼拿筷子。", + "evidence": [ + { + "description": "During this time, you have been active on the first floor.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:01" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you have not gone to the second floor to get the chopsticks so far.", + "B": "Yes, you have already gone to the second floor to get the chopsticks." + }, + "answer_label": "A", + "answer_text": "No, you have not gone to the second floor to get the chopsticks so far.", + "q_id": "egolife_A1_D3_036-mcq_fixed_9", + "correct_answer": "No, you have not gone to the second floor to get the chopsticks so far.", + "original_options": { + "A": "No, you have not gone to the second floor to get the chopsticks so far.", + "B": "Yes, you have already gone to the second floor to get the chopsticks." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_8", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just go to the second floor to get the chopsticks?", + "answer": "Yes, you went to the second floor to get the chopsticks.", + "question_chinese": "我刚刚去过二楼拿筷子吗？", + "answer_chinese": "是的，你去过二楼拿筷子。", + "evidence": [ + { + "description": "At this moment, you have come to the second floor to get the chopsticks.", + "timestamps": { + "start": "00:07:56", + "end": "00:07:57" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you did not go to the second floor to get the chopsticks.", + "B": "Yes, you went to the second floor to get the chopsticks." + }, + "answer_label": "B", + "answer_text": "Yes, you went to the second floor to get the chopsticks.", + "q_id": "egolife_A1_D3_036-mcq_fixed_8", + "correct_answer": "Yes, you went to the second floor to get the chopsticks.", + "original_options": { + "A": "No, you did not go to the second floor to get the chopsticks.", + "B": "Yes, you went to the second floor to get the chopsticks." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_4", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I just wash the dishes?", + "answer": "You just washed the dishes in the sink on your right.", + "question_chinese": "我刚刚在哪里洗的盘子？", + "answer_chinese": "你刚刚在右手边的水槽里洗的盘子。", + "evidence": [ + { + "description": "You are washing dishes in the sink.", + "timestamps": { + "start": "00:06:11", + "end": "00:06:30" + } + } + ], + "question_timestamps": "00:08:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the sink on your left.", + "B": "In the sink behind you.", + "C": "You washed the dishes at the small washing station near the pantry, not at the right-side sink.", + "D": "You washed the dishes in a deep tub near the refrigerator, instead of the sink to your right.", + "E": "You just washed the dishes in the sink on your right." + }, + "answer_label": "E", + "answer_text": "You just washed the dishes in the sink on your right.", + "q_id": "egolife_A1_D3_036-mcq_fixed_4", + "correct_answer": "You just washed the dishes in the sink on your right.", + "original_options": { + "A": "In the sink on your left.", + "B": "In the sink behind you.", + "C": "You washed the dishes at the small washing station near the pantry, not at the right-side sink.", + "D": "You washed the dishes in a deep tub near the refrigerator, instead of the sink to your right.", + "E": "You just washed the dishes in the sink on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_5", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where did I just wash the dishes?", + "answer": "You just washed the dishes in the sink in front of you.", + "question_chinese": "我刚刚在哪里洗的盘子？", + "answer_chinese": "你刚刚在面前的水槽里洗的盘子。", + "evidence": [ + { + "description": "You are washing dishes in the sink.", + "timestamps": { + "start": "00:06:11", + "end": "00:06:30" + } + } + ], + "question_timestamps": "00:10:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the sink behind you.", + "B": "In the sink to your left.", + "C": "You just washed the dishes in the sink in front of you.", + "D": "You washed the dishes in a basin near the back door, not in the sink directly ahead.", + "E": "You washed the dishes in a deep trough next to the garden shed, rather than the sink in front." + }, + "answer_label": "C", + "answer_text": "You just washed the dishes in the sink in front of you.", + "q_id": "egolife_A1_D3_036-mcq_fixed_5", + "correct_answer": "You just washed the dishes in the sink in front of you.", + "original_options": { + "A": "In the sink behind you.", + "B": "In the sink to your left.", + "C": "You just washed the dishes in the sink in front of you.", + "D": "You washed the dishes in a basin near the back door, not in the sink directly ahead.", + "E": "You washed the dishes in a deep trough next to the garden shed, rather than the sink in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_10", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just eat while sitting at the dining table, and have I finished eating?", + "answer": "You just ate cake at the table, and you did not finish it.", + "question_chinese": "我刚刚坐在餐桌旁吃了什么，吃完了吗？", + "answer_chinese": "你刚刚在餐桌旁吃了蛋糕，没吃完。", + "evidence": [ + { + "description": "You are now starting to eat cake at the table.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:27" + } + }, + { + "description": "At this moment, you are still eating the cake.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:11" + } + } + ], + "question_timestamps": "00:04:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You ate cake and finished it.", + "B": "You just ate cake at the table, and you did not finish it.", + "C": "You ate sushi and didn’t finish it.", + "D": "You ate a pastry at the dining table, and you finished it completely with no leftovers.", + "E": "You ate steak while sitting at the dining table, and you finished it rather than leaving some." + }, + "answer_label": "B", + "answer_text": "You just ate cake at the table, and you did not finish it.", + "q_id": "egolife_A1_D3_036-mcq_fixed_10", + "correct_answer": "You just ate cake at the table, and you did not finish it.", + "original_options": { + "A": "You ate cake and finished it.", + "B": "You just ate cake at the table, and you did not finish it.", + "C": "You ate sushi and didn’t finish it.", + "D": "You ate a pastry at the dining table, and you finished it completely with no leftovers.", + "E": "You ate steak while sitting at the dining table, and you finished it rather than leaving some." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_11", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just eat while sitting at the dining table, and have I finished eating?", + "answer": "You just ate cake at the table, and you finished it.", + "question_chinese": "我刚刚坐在餐桌旁吃了什么，吃完了吗？", + "answer_chinese": "你刚刚在餐桌旁吃了蛋糕，吃完了。", + "evidence": [ + { + "description": "You are now starting to eat cake at the table.", + "timestamps": { + "start": "00:01:22", + "end": "00:01:37" + } + }, + { + "description": "You have just finished the cake right now.", + "timestamps": { + "start": "00:05:09", + "end": "00:05:11" + } + } + ], + "question_timestamps": "00:10:22", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You ate cake and didn’t finish it.", + "B": "You ate bread and finished it.", + "C": "You ate a slice of pie at the dining table, and you still had a few bites left.", + "D": "You finished a bowl of fruit salad while sitting at the dining table, not cake.", + "E": "You just ate cake at the table, and you finished it." + }, + "answer_label": "E", + "answer_text": "You just ate cake at the table, and you finished it.", + "q_id": "egolife_A1_D3_036-mcq_fixed_11", + "correct_answer": "You just ate cake at the table, and you finished it.", + "original_options": { + "A": "You ate cake and didn’t finish it.", + "B": "You ate bread and finished it.", + "C": "You ate a slice of pie at the dining table, and you still had a few bites left.", + "D": "You finished a bowl of fruit salad while sitting at the dining table, not cake.", + "E": "You just ate cake at the table, and you finished it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_3", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What's the best way to get to the kitchen on the first floor to pick something up?", + "answer": "You should turn around, go straight, turn right, enter the door, turn right again, and you will arrive at the kitchen.", + "question_chinese": "我现在要去一楼厨房拿东西应该走什么途径？", + "answer_chinese": "你应该向后转直走右转，进门再右转就到了厨房。", + "evidence": [ + { + "description": "Now, you head straight from the kitchen, turn left, and go out the door.", + "timestamps": { + "start": "00:05:19", + "end": "00:05:20" + } + }, + { + "description": "Now, you immediately turn left at the door and walk straight ahead.", + "timestamps": { + "start": "00:05:21", + "end": "00:05:24" + } + } + ], + "question_timestamps": "00:05:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn around and take a left.", + "B": "Go straight and enter the left door.", + "C": "You should turn around, go straight, turn right, enter the door, turn right again, and you will arrive at the kitchen.", + "D": "Turn around, go straight, turn right, enter the door, then turn left instead of right to reach it.", + "E": "Turn around, go straight, turn left, enter the door, then turn right again to get to the kitchen." + }, + "answer_label": "C", + "answer_text": "You should turn around, go straight, turn right, enter the door, turn right again, and you will arrive at the kitchen.", + "q_id": "egolife_A1_D3_036-mcq_fixed_3", + "correct_answer": "You should turn around, go straight, turn right, enter the door, turn right again, and you will arrive at the kitchen.", + "original_options": { + "A": "Turn around and take a left.", + "B": "Go straight and enter the left door.", + "C": "You should turn around, go straight, turn right, enter the door, turn right again, and you will arrive at the kitchen.", + "D": "Turn around, go straight, turn right, enter the door, then turn left instead of right to reach it.", + "E": "Turn around, go straight, turn left, enter the door, then turn right again to get to the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_036-mcq_fixed_2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What's the best way to get to the kitchen on the first floor to pick something up?", + "answer": "You should turn around, head straight towards the front-left, then make a right turn, followed by another right turn. Exit the door and turn right again, and you'll reach the kitchen.", + "question_chinese": "我现在要去一楼厨房拿东西应该走什么途径？", + "answer_chinese": "你应该向后转向左前方直走右转再右转，出门再右转就到了厨房。", + "evidence": [ + { + "description": "Now, you head straight from the kitchen, turn left, and go out the door.", + "timestamps": { + "start": "00:05:19", + "end": "00:05:20" + } + }, + { + "description": "Now, you immediately turn left at the door and walk straight ahead.", + "timestamps": { + "start": "00:05:21", + "end": "00:05:24" + } + }, + { + "description": "Now, after reaching the corner, you walk straight towards the front-left direction. You have arrived at your current position.", + "timestamps": { + "start": "00:05:25", + "end": "00:05:29" + } + } + ], + "question_timestamps": "00:09:36", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, then exit the door.", + "B": "Head front-right, then turn left.", + "C": "You should turn around, head straight towards the front-left, then make a right turn, followed by another right turn. Exit the door and turn right again, and you'll reach the kitchen.", + "D": "Turn around, head straight toward the front-left, then make a left turn, then a right, exit and turn right.", + "E": "Turn around, head straight toward the front-left, make a right turn, then a left turn, exit the door and turn left." + }, + "answer_label": "C", + "answer_text": "You should turn around, head straight towards the front-left, then make a right turn, followed by another right turn. Exit the door and turn right again, and you'll reach the kitchen.", + "q_id": "egolife_A1_D3_036-mcq_fixed_2", + "correct_answer": "You should turn around, head straight towards the front-left, then make a right turn, followed by another right turn. Exit the door and turn right again, and you'll reach the kitchen.", + "original_options": { + "A": "Turn left, then exit the door.", + "B": "Head front-right, then turn left.", + "C": "You should turn around, head straight towards the front-left, then make a right turn, followed by another right turn. Exit the door and turn right again, and you'll reach the kitchen.", + "D": "Turn around, head straight toward the front-left, then make a left turn, then a right, exit and turn right.", + "E": "Turn around, head straight toward the front-left, make a right turn, then a left turn, exit the door and turn left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 21 + ] + } +] diff --git a/QAs/egolife_A1_D3_037-mcq.json b/QAs/egolife_A1_D3_037-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a328147fe855cd169486710846bca387ecd462d9 --- /dev/null +++ b/QAs/egolife_A1_D3_037-mcq.json @@ -0,0 +1,600 @@ +[ + { + "video_uid": "egolife_A1_D3_037", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_037-mcq_fixed_2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What did the girl in the green clothes just do?", + "answer": "The girl in the green clothes just helped you hold the bag.", + "question_chinese": "刚刚绿色衣服的女孩做了什么？", + "answer_chinese": "绿衣服女生刚刚帮你撑袋子。", + "evidence": [ + { + "description": "At this moment, the girl in the green shirt is continuously helping you hold the bag.", + "timestamps": { + "start": "00:03:21", + "end": "00:03:24" + } + } + ], + "question_timestamps": "00:04:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "She showed you where to place it.", + "B": "The girl in the green clothes just helped you hold the bag.", + "C": "She told you where to store the bag.", + "D": "The girl in the green clothes explained where to store the bag and suggested a safer way to handle it.", + "E": "The girl in the green clothes pointed out where to put your belongings instead of helping you hold the bag." + }, + "answer_label": "B", + "answer_text": "The girl in the green clothes just helped you hold the bag.", + "q_id": "egolife_A1_D3_037-mcq_fixed_2", + "correct_answer": "The girl in the green clothes just helped you hold the bag.", + "original_options": { + "A": "She showed you where to place it.", + "B": "The girl in the green clothes just helped you hold the bag.", + "C": "She told you where to store the bag.", + "D": "The girl in the green clothes explained where to store the bag and suggested a safer way to handle it.", + "E": "The girl in the green clothes pointed out where to put your belongings instead of helping you hold the bag." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_037", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_037-mcq_fixed_3", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What did the girl in the green clothes just do?", + "answer": "The girl in the green clothes just helped you hold the bag and then went to the kitchen to get a plate of strawberries.", + "question_chinese": "刚刚绿色衣服的女孩做了什么？", + "answer_chinese": "刚刚绿色衣服的女孩帮你撑袋子，然后去厨房拿了一盘草莓。", + "evidence": [ + { + "description": "At this moment, the girl in the green shirt is continuously helping you hold the bag.", + "timestamps": { + "start": "00:03:21", + "end": "00:03:24" + } + }, + { + "description": "At this moment, the girl in the green clothes has brought over a plate of strawberries.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:31" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The girl in the green clothes just helped you hold the bag and then went to the kitchen to get a plate of strawberries.", + "B": "She brought you a bowl of apples.", + "C": "She got you a glass of water.", + "D": "She helped you hold the bag, then went to the kitchen to bring back a bowl of apples.", + "E": "She took the bag from you briefly, then headed into the kitchen to fetch a glass of water." + }, + "answer_label": "A", + "answer_text": "The girl in the green clothes just helped you hold the bag and then went to the kitchen to get a plate of strawberries.", + "q_id": "egolife_A1_D3_037-mcq_fixed_3", + "correct_answer": "The girl in the green clothes just helped you hold the bag and then went to the kitchen to get a plate of strawberries.", + "original_options": { + "A": "The girl in the green clothes just helped you hold the bag and then went to the kitchen to get a plate of strawberries.", + "B": "She brought you a bowl of apples.", + "C": "She got you a glass of water.", + "D": "She helped you hold the bag, then went to the kitchen to bring back a bowl of apples.", + "E": "She took the bag from you briefly, then headed into the kitchen to fetch a glass of water." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_037", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_037-mcq_fixed_4", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I just do with the washed knife I was holding?", + "answer": "You took the washed knife and cut the plastic bag's opening.", + "question_chinese": "我刚刚拿着洗好的菜刀干了什么？", + "answer_chinese": "你拿着洗好的菜刀切塑料袋口。", + "evidence": [ + { + "description": "You took the washed knife and cut the plastic bag's opening.", + "timestamps": { + "start": "00:01:39", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You put it back in the drawer.", + "B": "You dried it with a towel.", + "C": "You took the washed knife and cut the plastic bag's opening.", + "D": "You used the washed knife to slice an apple, then set the knife down by the sink.", + "E": "You peeled an orange with the washed knife and then placed it away without cutting any plastic." + }, + "answer_label": "C", + "answer_text": "You took the washed knife and cut the plastic bag's opening.", + "q_id": "egolife_A1_D3_037-mcq_fixed_4", + "correct_answer": "You took the washed knife and cut the plastic bag's opening.", + "original_options": { + "A": "You put it back in the drawer.", + "B": "You dried it with a towel.", + "C": "You took the washed knife and cut the plastic bag's opening.", + "D": "You used the washed knife to slice an apple, then set the knife down by the sink.", + "E": "You peeled an orange with the washed knife and then placed it away without cutting any plastic." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_037", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_037-mcq_fixed_5", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What did I just do with the washed knife I was holding?", + "answer": "You used the washed knife to cut the opening of the plastic bag, and then placed the knife on the plate inside the sink.", + "question_chinese": "我刚刚拿着洗好的菜刀干了什么？", + "answer_chinese": "你拿着洗好的菜刀切塑料袋口，然后把菜刀放到了洗手槽里的盘子上。", + "evidence": [ + { + "description": "You took the washed knife and cut the plastic bag's opening.", + "timestamps": { + "start": "00:01:39", + "end": "00:01:42" + } + }, + { + "description": "At this moment, you placed the kitchen knife onto the plate inside the sink.", + "timestamps": { + "start": "00:01:46", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:07:41", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You used the washed knife to cut the opening of the plastic bag, and then placed the knife on the plate inside the sink.", + "B": "You put it back on the counter.", + "C": "You set it on the cutting board.", + "D": "You used the washed knife to open the plastic bag, then left the knife on the countertop by the sink.", + "E": "You cut the plastic bag open and then returned the washed knife to the drawer instead of leaving it in the sink." + }, + "answer_label": "A", + "answer_text": "You used the washed knife to cut the opening of the plastic bag, and then placed the knife on the plate inside the sink.", + "q_id": "egolife_A1_D3_037-mcq_fixed_5", + "correct_answer": "You used the washed knife to cut the opening of the plastic bag, and then placed the knife on the plate inside the sink.", + "original_options": { + "A": "You used the washed knife to cut the opening of the plastic bag, and then placed the knife on the plate inside the sink.", + "B": "You put it back on the counter.", + "C": "You set it on the cutting board.", + "D": "You used the washed knife to open the plastic bag, then left the knife on the countertop by the sink.", + "E": "You cut the plastic bag open and then returned the washed knife to the drawer instead of leaving it in the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 22 + ] + }, + { + "video_uid": "egolife_A1_D3_037", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_037-mcq_fixed_6", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is my current relative position to the kitchen?", + "answer": "At this moment, the kitchen is on your right side, and you are sitting facing the wall.", + "question_chinese": "此时我和厨房的相对位置是什么？", + "answer_chinese": "此时厨房在你的右手边，你面朝墙壁坐着。", + "evidence": [ + { + "description": "At this moment, the kitchen is on your right side, and you are sitting facing the wall.", + "timestamps": { + "start": "00:02:59", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:03:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The kitchen is on your left.", + "B": "The kitchen is behind you.", + "C": "Right now the kitchen is in front of you, and you are sitting facing the refrigerator.", + "D": "You are turned toward a window, with the kitchen positioned behind your left shoulder as you sit.", + "E": "At this moment, the kitchen is on your right side, and you are sitting facing the wall." + }, + "answer_label": "E", + "answer_text": "At this moment, the kitchen is on your right side, and you are sitting facing the wall.", + "q_id": "egolife_A1_D3_037-mcq_fixed_6", + "correct_answer": "At this moment, the kitchen is on your right side, and you are sitting facing the wall.", + "original_options": { + "A": "The kitchen is on your left.", + "B": "The kitchen is behind you.", + "C": "Right now the kitchen is in front of you, and you are sitting facing the refrigerator.", + "D": "You are turned toward a window, with the kitchen positioned behind your left shoulder as you sit.", + "E": "At this moment, the kitchen is on your right side, and you are sitting facing the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_037", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_037-mcq_fixed_7", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is my current relative position to the kitchen?", + "answer": "At this moment, the kitchen is behind you, and you are sitting facing the dining table.", + "question_chinese": "此时我和厨房的相对位置是什么？", + "answer_chinese": "此时，厨房在你的后面，你面朝餐桌坐着。", + "evidence": [ + { + "description": "At this moment, the kitchen is on your right side, and you are sitting facing the wall.", + "timestamps": { + "start": "00:02:59", + "end": "00:03:00" + } + }, + { + "description": "At this moment, the kitchen is behind you, and you are sitting facing the dining table.", + "timestamps": { + "start": "00:09:01", + "end": "00:09:02" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The kitchen is in front of you.", + "B": "The kitchen is on your left.", + "C": "At this moment, the kitchen is behind you, and you are sitting facing the dining table.", + "D": "The kitchen is directly to your right, and you are sitting angled slightly toward the living room area.", + "E": "You are facing the hallway, with the kitchen located just past your left shoulder while you sit." + }, + "answer_label": "C", + "answer_text": "At this moment, the kitchen is behind you, and you are sitting facing the dining table.", + "q_id": "egolife_A1_D3_037-mcq_fixed_7", + "correct_answer": "At this moment, the kitchen is behind you, and you are sitting facing the dining table.", + "original_options": { + "A": "The kitchen is in front of you.", + "B": "The kitchen is on your left.", + "C": "At this moment, the kitchen is behind you, and you are sitting facing the dining table.", + "D": "The kitchen is directly to your right, and you are sitting angled slightly toward the living room area.", + "E": "You are facing the hallway, with the kitchen located just past your left shoulder while you sit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_037", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_037-mcq_fixed_1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the plate get here?", + "answer": "After being washed by you, the plate left the kitchen sink, turned around, went straight for two steps, turned left, and arrived at its current position.", + "question_chinese": "盘子是经过怎样的途径到达现在的位置？", + "answer_chinese": "盘子被你清洗后从厨房水槽出发，向后转直走两步后向左转到了现在的位置。", + "evidence": [ + { + "description": "Now, you have finished washing the plate.", + "timestamps": { + "start": "00:07:32", + "end": "00:07:44" + } + }, + { + "description": "Now, you turn around with the plate in hand, walk two steps straight, and then turn left.", + "timestamps": { + "start": "00:07:45", + "end": "00:07:47" + } + } + ], + "question_timestamps": "00:07:48", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It went three steps, then left.", + "B": "It went two steps, then right.", + "C": "After you washed it, the plate left the sink, turned around, walked three steps straight, then turned left to arrive.", + "D": "After being washed by you, the plate left the sink, turned around, went straight two steps, turned right, and reached its spot.", + "E": "After being washed by you, the plate left the kitchen sink, turned around, went straight for two steps, turned left, and arrived at its current position." + }, + "answer_label": "E", + "answer_text": "After being washed by you, the plate left the kitchen sink, turned around, went straight for two steps, turned left, and arrived at its current position.", + "q_id": "egolife_A1_D3_037-mcq_fixed_1", + "correct_answer": "After being washed by you, the plate left the kitchen sink, turned around, went straight for two steps, turned left, and arrived at its current position.", + "original_options": { + "A": "It went three steps, then left.", + "B": "It went two steps, then right.", + "C": "After you washed it, the plate left the sink, turned around, walked three steps straight, then turned left to arrive.", + "D": "After being washed by you, the plate left the sink, turned around, went straight two steps, turned right, and reached its spot.", + "E": "After being washed by you, the plate left the kitchen sink, turned around, went straight for two steps, turned left, and arrived at its current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 22 + ] + }, + { + "video_uid": "egolife_A1_D3_037", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_037-mcq_fixed_0", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the plate get here?", + "answer": "You washed the plate, took it from the sink, turned back, walked two steps straight, turned left, exited the door, made an immediate left, and walked straight to the balcony table.", + "question_chinese": "盘子是经过怎样的途径到达现在的位置？", + "answer_chinese": "盘子被你清洗后从厨房水槽出发，向后转直走两步后向左转出门立即左转直走到尽头到达院子的桌子上。", + "evidence": [ + { + "description": "Now, you have finished washing the plate.", + "timestamps": { + "start": "00:07:32", + "end": "00:07:44" + } + }, + { + "description": "Now, you turn around with the plate in hand, walk two steps straight, and then turn left.", + "timestamps": { + "start": "00:07:45", + "end": "00:07:47" + } + }, + { + "description": "Now, you immediately turn left, walk straight to the end, and arrive at the table on the balcony.", + "timestamps": { + "start": "00:07:49", + "end": "00:07:56" + } + } + ], + "question_timestamps": "00:10:05", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You washed the plate, took it from the sink, turned back, walked two steps straight, turned left, exited the door, made an immediate left, and walked straight to the balcony table.", + "B": "You walked three steps before turning.", + "C": "You turned right after two steps.", + "D": "You washed the plate, took it from the sink, walked three steps, turned left, exited, turned left, and reached the balcony table.", + "E": "You washed it, carried it out the door, but turned right immediately and then walked straight to a different table." + }, + "answer_label": "A", + "answer_text": "You washed the plate, took it from the sink, turned back, walked two steps straight, turned left, exited the door, made an immediate left, and walked straight to the balcony table.", + "q_id": "egolife_A1_D3_037-mcq_fixed_0", + "correct_answer": "You washed the plate, took it from the sink, turned back, walked two steps straight, turned left, exited the door, made an immediate left, and walked straight to the balcony table.", + "original_options": { + "A": "You washed the plate, took it from the sink, turned back, walked two steps straight, turned left, exited the door, made an immediate left, and walked straight to the balcony table.", + "B": "You walked three steps before turning.", + "C": "You turned right after two steps.", + "D": "You washed the plate, took it from the sink, walked three steps, turned left, exited, turned left, and reached the balcony table.", + "E": "You washed it, carried it out the door, but turned right immediately and then walked straight to a different table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 20 + ] + }, + { + "video_uid": "egolife_A1_D3_037", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_037_5", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many of the small cakes with piped cream did I decorate?", + "answer": "One of them is the one you piped.", + "question_chinese": "目前为止挤好奶油的小蛋糕里有几个是我挤的？", + "answer_chinese": "有一个是你挤的。", + "evidence": [ + { + "description": "At this moment, the cream on six small cakes was all piped by the girl.", + "timestamps": { + "start": "00:08:24", + "end": "00:08:33" + } + }, + { + "description": "Now, there are seven small cakes with whipped cream in total, so I piped one of them.", + "timestamps": { + "start": "00:09:41", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:10:09", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two of them are the ones you piped.", + "B": "One of them is the one you piped.", + "C": "Three of them are the ones you piped.", + "D": "Four of them are the ones you piped so far, including the ones you decorated earlier.", + "E": "Five of them are the ones you piped up to now, counting each small cake with cream you decorated." + }, + "answer_label": "B", + "answer_text": "One of them is the one you piped.", + "q_id": "egolife_A1_D3_037_5", + "correct_answer": "One of them is the one you piped.", + "original_options": { + "A": "Two of them are the ones you piped.", + "B": "One of them is the one you piped.", + "C": "Three of them are the ones you piped.", + "D": "Four of them are the ones you piped so far, including the ones you decorated earlier.", + "E": "Five of them are the ones you piped up to now, counting each small cake with cream you decorated." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_037", + "video_length": "00:12:00", + "qid": "egolife_A1_D3_037_6", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many of the small cakes with whipped cream that have been decorated so far did I decorate?", + "answer": "Two of them are the ones you piped.", + "question_chinese": "目前为止挤好奶油的小蛋糕里有几个是我挤的？", + "answer_chinese": "有两个是你挤的。", + "evidence": [ + { + "description": "At this moment, the cream on six small cakes was all piped by the girl.", + "timestamps": { + "start": "00:08:24", + "end": "00:08:33" + } + }, + { + "description": "Now, there are eight small cakes with whipped cream in total, so I piped two of them.", + "timestamps": { + "start": "00:10:55", + "end": "00:10:56" + } + } + ], + "question_timestamps": "00:11:11", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One is the one you piped.", + "B": "Three are the ones you piped.", + "C": "Four of them are the ones you piped, and the remaining cakes were decorated by someone else.", + "D": "Five of them are the ones you piped, with only a couple decorated by another person.", + "E": "Two of them are the ones you piped." + }, + "answer_label": "E", + "answer_text": "Two of them are the ones you piped.", + "q_id": "egolife_A1_D3_037_6", + "correct_answer": "Two of them are the ones you piped.", + "original_options": { + "A": "One is the one you piped.", + "B": "Three are the ones you piped.", + "C": "Four of them are the ones you piped, and the remaining cakes were decorated by someone else.", + "D": "Five of them are the ones you piped, with only a couple decorated by another person.", + "E": "Two of them are the ones you piped." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D3_038-mcq.json b/QAs/egolife_A1_D3_038-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7a7a5652eaef709e3e5865763a2fe56e70c0629c --- /dev/null +++ b/QAs/egolife_A1_D3_038-mcq.json @@ -0,0 +1,920 @@ +[ + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just see someone playing a musical instrument on the balcony?", + "answer": "No, you didn't see anyone playing a musical instrument.", + "question_chinese": "刚刚我见过阳台上有人弹乐器吗？", + "answer_chinese": "没有，你没见过有人弹乐器。", + "evidence": [ + { + "description": "So far, no one has played a musical instrument.", + "timestamps": { + "start": "00:03:29", + "end": "00:03:30" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you saw someone playing a musical instrument.", + "B": "No, you didn't see anyone playing a musical instrument." + }, + "answer_label": "B", + "answer_text": "No, you didn't see anyone playing a musical instrument.", + "q_id": "egolife_A1_D3_038_7", + "correct_answer": "No, you didn't see anyone playing a musical instrument.", + "original_options": { + "A": "Yes, you saw someone playing a musical instrument.", + "B": "No, you didn't see anyone playing a musical instrument." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just see someone playing an instrument on the balcony?", + "answer": "Yes, you saw someone playing a musical instrument just now.", + "question_chinese": "刚刚我见过阳台上有人弹乐器吗？", + "answer_chinese": "是的，你看见有人刚刚弹乐器了。", + "evidence": [ + { + "description": "So far, no one has played a musical instrument.", + "timestamps": { + "start": "00:03:29", + "end": "00:03:30" + } + }, + { + "description": "At this moment, a boy is playing a musical instrument.", + "timestamps": { + "start": "00:07:12", + "end": "00:08:21" + } + } + ], + "question_timestamps": "00:09:38", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you saw someone playing a musical instrument just now.", + "B": "No, you did not see anyone playing a musical instrument just now." + }, + "answer_label": "A", + "answer_text": "Yes, you saw someone playing a musical instrument just now.", + "q_id": "egolife_A1_D3_038_8", + "correct_answer": "Yes, you saw someone playing a musical instrument just now.", + "original_options": { + "A": "Yes, you saw someone playing a musical instrument just now.", + "B": "No, you did not see anyone playing a musical instrument just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038-mcq_fixed_7", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the chopsticks just now?", + "answer": "YOU took the chopsticks from the tabletop on the left side of the kitchen and then washed them in the sink.", + "question_chinese": "刚刚筷子经过了怎么样的运动轨迹？", + "answer_chinese": "筷子从厨房左手边的桌面上被我拿出来，然后被我放到了水槽里洗。", + "evidence": [ + { + "description": "At this moment, I have taken the chopsticks from the kitchen table.", + "timestamps": { + "start": "00:03:50", + "end": "00:03:54" + } + }, + { + "description": "Right now, I am quickly washing the chopsticks.", + "timestamps": { + "start": "00:03:55", + "end": "00:03:58" + } + } + ], + "question_timestamps": "00:03:59", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You moved them to the middle counter.", + "B": "You washed them on the countertop.", + "C": "You took the chopsticks from the tabletop on the right side of the kitchen and then washed them in the sink.", + "D": "You picked up the chopsticks from the dish rack and set them down on the left-side tabletop without washing.", + "E": "YOU took the chopsticks from the tabletop on the left side of the kitchen and then washed them in the sink." + }, + "answer_label": "E", + "answer_text": "YOU took the chopsticks from the tabletop on the left side of the kitchen and then washed them in the sink.", + "q_id": "egolife_A1_D3_038-mcq_fixed_7", + "correct_answer": "YOU took the chopsticks from the tabletop on the left side of the kitchen and then washed them in the sink.", + "original_options": { + "A": "You moved them to the middle counter.", + "B": "You washed them on the countertop.", + "C": "You took the chopsticks from the tabletop on the right side of the kitchen and then washed them in the sink.", + "D": "You picked up the chopsticks from the dish rack and set them down on the left-side tabletop without washing.", + "E": "YOU took the chopsticks from the tabletop on the left side of the kitchen and then washed them in the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 21, + 19 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038-mcq_fixed_6", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the chopsticks just now?", + "answer": "YOU took the chopsticks from the tabletop on the left side of the kitchen, washed them in the sink, and then gave them to the girl in the white shirt.", + "question_chinese": "刚刚筷子经过了怎么样的运动轨迹？", + "answer_chinese": "筷子从厨房左手边的桌面上被我拿出来，然后被我放到了水槽里洗，之后被我给了白衣服女孩。", + "evidence": [ + { + "description": "At this moment, I have taken the chopsticks from the kitchen table.", + "timestamps": { + "start": "00:03:50", + "end": "00:03:54" + } + }, + { + "description": "Right now, I am quickly washing the chopsticks.", + "timestamps": { + "start": "00:03:55", + "end": "00:03:58" + } + }, + { + "description": "At this moment, I am handing the chopsticks to the girl in the white shirt.", + "timestamps": { + "start": "00:04:01", + "end": "00:04:03" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You gave them away without washing.", + "B": "You washed them but kept them.", + "C": "You took the chopsticks from the right-side tabletop, washed them in the sink, then gave them to the girl.", + "D": "You took the chopsticks from the left-side tabletop, washed them in the sink, then placed them back down.", + "E": "YOU took the chopsticks from the tabletop on the left side of the kitchen, washed them in the sink, and then gave them to the girl in the white shirt." + }, + "answer_label": "E", + "answer_text": "YOU took the chopsticks from the tabletop on the left side of the kitchen, washed them in the sink, and then gave them to the girl in the white shirt.", + "q_id": "egolife_A1_D3_038-mcq_fixed_6", + "correct_answer": "YOU took the chopsticks from the tabletop on the left side of the kitchen, washed them in the sink, and then gave them to the girl in the white shirt.", + "original_options": { + "A": "You gave them away without washing.", + "B": "You washed them but kept them.", + "C": "You took the chopsticks from the right-side tabletop, washed them in the sink, then gave them to the girl.", + "D": "You took the chopsticks from the left-side tabletop, washed them in the sink, then placed them back down.", + "E": "YOU took the chopsticks from the tabletop on the left side of the kitchen, washed them in the sink, and then gave them to the girl in the white shirt." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038-mcq_fixed_8", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I eaten the cake on the table in the yard so far?", + "answer": "No.", + "question_chinese": "我吃了院子桌子上的蛋糕吗到目前为止？", + "answer_chinese": "没有，你没有吃院子桌子上的蛋糕。", + "evidence": [ + { + "description": "So far, you have not eaten the cake on the table in the yard.", + "timestamps": { + "start": "00:06:16", + "end": "00:06:17" + } + } + ], + "question_timestamps": "00:08:18", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D3_038-mcq_fixed_8", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038-mcq_fixed_9", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I eaten the cake on the table in the yard so far?", + "answer": "Yes, you have eaten the cake on the table in the yard.", + "question_chinese": "我吃了院子桌子上的蛋糕吗到目前为止？", + "answer_chinese": "是的，你吃过院子桌子上的蛋糕。", + "evidence": [ + { + "description": "So far, you have not eaten the cake on the table in the yard.", + "timestamps": { + "start": "00:06:16", + "end": "00:06:17" + } + }, + { + "description": "Right now, you are eating the cake.", + "timestamps": { + "start": "00:09:16", + "end": "00:09:17" + } + } + ], + "question_timestamps": "00:10:18", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you have eaten the cake on the table in the yard.", + "B": "No, you have not eaten the cake on the table in the yard." + }, + "answer_label": "A", + "answer_text": "Yes, you have eaten the cake on the table in the yard.", + "q_id": "egolife_A1_D3_038-mcq_fixed_9", + "correct_answer": "Yes, you have eaten the cake on the table in the yard.", + "original_options": { + "A": "Yes, you have eaten the cake on the table in the yard.", + "B": "No, you have not eaten the cake on the table in the yard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_6_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where am I relative to the guy in the white shirt?", + "answer": "You are behind the boy in white.", + "question_chinese": "我在白色衣服男生的什么方向？", + "answer_chinese": "你在白色衣服男生的后面。", + "evidence": [ + { + "description": "You are behind the boy in white.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:30" + } + } + ], + "question_timestamps": "00:04:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are behind the boy in white.", + "B": "You are in front of the boy in white." + }, + "answer_label": "A", + "answer_text": "You are behind the boy in white.", + "q_id": "egolife_A1_D3_038_6_T2", + "correct_answer": "You are behind the boy in white.", + "original_options": { + "A": "You are behind the boy in white.", + "B": "You are in front of the boy in white." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_6_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where am I relative to the guy in the white shirt?", + "answer": "You are in front of the boy in white.", + "question_chinese": "我在白色衣服男生的什么方向？", + "answer_chinese": "你在白色衣服男生的前面。", + "evidence": [ + { + "description": "You are in front of the boy in white.", + "timestamps": { + "start": "00:05:45", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:05:53", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are behind the boy in white.", + "B": "You are in front of the boy in white." + }, + "answer_label": "B", + "answer_text": "You are in front of the boy in white.", + "q_id": "egolife_A1_D3_038_6_T1", + "correct_answer": "You are in front of the boy in white.", + "original_options": { + "A": "You are behind the boy in white.", + "B": "You are in front of the boy in white." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_4_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many pairs of chopsticks did I wash?", + "answer": "You washed 0 pairs of chopsticks, and the chopsticks are in front of you.", + "question_chinese": "我洗了几双筷子？", + "answer_chinese": "你洗了0双筷子，筷子在你前面。", + "evidence": [ + { + "description": "You washed zero pairs of chopsticks.", + "timestamps": { + "start": "00:03:29", + "end": "00:03:32" + } + } + ], + "question_timestamps": "00:03:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed 0 pairs of chopsticks, and the chopsticks are in front of you.", + "B": "You washed 1 pair of chopsticks.", + "C": "You washed 2 pairs of chopsticks.", + "D": "You washed 3 pairs of chopsticks, and the chopsticks are still in front of you afterward.", + "E": "You washed 4 pairs of chopsticks, and the chopsticks remain in front of you when you finish." + }, + "answer_label": "A", + "answer_text": "You washed 0 pairs of chopsticks, and the chopsticks are in front of you.", + "q_id": "egolife_A1_D3_038_4_T2", + "correct_answer": "You washed 0 pairs of chopsticks, and the chopsticks are in front of you.", + "original_options": { + "A": "You washed 0 pairs of chopsticks, and the chopsticks are in front of you.", + "B": "You washed 1 pair of chopsticks.", + "C": "You washed 2 pairs of chopsticks.", + "D": "You washed 3 pairs of chopsticks, and the chopsticks are still in front of you afterward.", + "E": "You washed 4 pairs of chopsticks, and the chopsticks remain in front of you when you finish." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_4_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many pairs of chopsticks did I wash?", + "answer": "You washed 1 pair of chopsticks, and the chopsticks are in front of you.", + "question_chinese": "我洗了几双筷子？", + "answer_chinese": "你洗了1双筷子，筷子在你前面。", + "evidence": [ + { + "description": "You washed a pairs of chopsticks.", + "timestamps": { + "start": "00:03:55", + "end": "00:03:58" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You washed 0 pairs of chopsticks.", + "B": "You washed 2 pairs of chopsticks.", + "C": "You washed 3 pairs of chopsticks, and the chopsticks are in front of you when you’re done.", + "D": "You washed 4 pairs of chopsticks, and the chopsticks are placed in front of you afterward.", + "E": "You washed 1 pair of chopsticks, and the chopsticks are in front of you." + }, + "answer_label": "E", + "answer_text": "You washed 1 pair of chopsticks, and the chopsticks are in front of you.", + "q_id": "egolife_A1_D3_038_4_T1", + "correct_answer": "You washed 1 pair of chopsticks, and the chopsticks are in front of you.", + "original_options": { + "A": "You washed 0 pairs of chopsticks.", + "B": "You washed 2 pairs of chopsticks.", + "C": "You washed 3 pairs of chopsticks, and the chopsticks are in front of you when you’re done.", + "D": "You washed 4 pairs of chopsticks, and the chopsticks are placed in front of you afterward.", + "E": "You washed 1 pair of chopsticks, and the chopsticks are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_1_T2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I touch the faucet?", + "answer": "You can't touch the faucet; the faucet is in front of you.", + "question_chinese": "我能碰到水龙头吗？", + "answer_chinese": "你不能碰到水龙头，水龙头在你前面。", + "evidence": [ + { + "description": "You must not touch the tap.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:03:46", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can touch the faucet; the faucet is in front of you.", + "B": "You can't touch the faucet; the faucet is in front of you." + }, + "answer_label": "B", + "answer_text": "You can't touch the faucet; the faucet is in front of you.", + "q_id": "egolife_A1_D3_038_1_T2", + "correct_answer": "You can't touch the faucet; the faucet is in front of you.", + "original_options": { + "A": "You can touch the faucet; the faucet is in front of you.", + "B": "You can't touch the faucet; the faucet is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_1_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I touch the faucet?", + "answer": "You can touch the faucet; the faucet is to your right.", + "question_chinese": "我能碰到水龙头吗？", + "answer_chinese": "你能碰到水龙头，水龙头在你右边。", + "evidence": [ + { + "description": "You can touch the tap.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:47" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can touch the faucet; the faucet is to your right.", + "B": "You can't touch the faucet; the faucet is to your right." + }, + "answer_label": "A", + "answer_text": "You can touch the faucet; the faucet is to your right.", + "q_id": "egolife_A1_D3_038_1_T1", + "correct_answer": "You can touch the faucet; the faucet is to your right.", + "original_options": { + "A": "You can touch the faucet; the faucet is to your right.", + "B": "You can't touch the faucet; the faucet is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038-mcq_fixed_2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is my current relative position to the barbecue grill?", + "answer": "At this moment, the barbecue grill is right in front of you.", + "question_chinese": "此时我和烧烤架的相对位置是什么？", + "answer_chinese": "此时烧烤架在你的正前面。", + "evidence": [ + { + "description": "At this moment, the barbecue grill is right in front of you.", + "timestamps": { + "start": "00:06:27", + "end": "00:06:28" + } + } + ], + "question_timestamps": "00:06:33", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The barbecue grill is to your left.", + "B": "The barbecue grill is behind you.", + "C": "At this moment, the barbecue grill is off to your right side rather than directly in front.", + "D": "At this moment, the barbecue grill is right in front of you.", + "E": "Right now, the barbecue grill sits several feet behind you, not in front of your current position." + }, + "answer_label": "D", + "answer_text": "At this moment, the barbecue grill is right in front of you.", + "q_id": "egolife_A1_D3_038-mcq_fixed_2", + "correct_answer": "At this moment, the barbecue grill is right in front of you.", + "original_options": { + "A": "The barbecue grill is to your left.", + "B": "The barbecue grill is behind you.", + "C": "At this moment, the barbecue grill is off to your right side rather than directly in front.", + "D": "At this moment, the barbecue grill is right in front of you.", + "E": "Right now, the barbecue grill sits several feet behind you, not in front of your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038-mcq_fixed_3", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is my current relative position to the barbecue grill?", + "answer": "At this moment, the barbecue grill is behind you.", + "question_chinese": "此时我和烧烤架的相对位置是什么？", + "answer_chinese": "此时烧烤架在你的后面。", + "evidence": [ + { + "description": "At this moment, the barbecue grill is right in front of you.", + "timestamps": { + "start": "00:06:27", + "end": "00:06:28" + } + }, + { + "description": "You turned left and walked straight, so the barbecue grill is now behind you.", + "timestamps": { + "start": "00:06:42", + "end": "00:06:43" + } + } + ], + "question_timestamps": "00:06:52", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The barbecue grill is in front of you.", + "B": "The barbecue grill is to your right.", + "C": "At this moment, the barbecue grill is slightly to your left, not positioned behind you.", + "D": "At this moment, the barbecue grill is behind you.", + "E": "The barbecue grill is to your left, so it isn’t behind your current position." + }, + "answer_label": "D", + "answer_text": "At this moment, the barbecue grill is behind you.", + "q_id": "egolife_A1_D3_038-mcq_fixed_3", + "correct_answer": "At this moment, the barbecue grill is behind you.", + "original_options": { + "A": "The barbecue grill is in front of you.", + "B": "The barbecue grill is to your right.", + "C": "At this moment, the barbecue grill is slightly to your left, not positioned behind you.", + "D": "At this moment, the barbecue grill is behind you.", + "E": "The barbecue grill is to your left, so it isn’t behind your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_2_T2", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, how many other people are in the room besides me?", + "answer": "Right now, you are the only person in the room.", + "question_chinese": "此时房间里除了我还有多少人？", + "answer_chinese": "此时房间里只有你。", + "evidence": [ + { + "description": "Nothing.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:53" + } + } + ], + "question_timestamps": "00:03:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, you are the only person in the room.", + "B": "Right now, there are two other people in the room besides you." + }, + "answer_label": "A", + "answer_text": "Right now, you are the only person in the room.", + "q_id": "egolife_A1_D3_038_2_T2", + "correct_answer": "Right now, you are the only person in the room.", + "original_options": { + "A": "Right now, you are the only person in the room.", + "B": "Right now, there are two other people in the room besides you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_2_T1", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, how many other people are in the room besides me?", + "answer": "Right now, there are two other people in the room besides you.", + "question_chinese": "此时房间里除了我还有多少人？", + "answer_chinese": "此时房间里除了你还有2个人。", + "evidence": [ + { + "description": "two pelpols", + "timestamps": { + "start": "00:04:00", + "end": "00:04:08" + } + } + ], + "question_timestamps": "00:04:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, there is one other person in the room besides you.", + "B": "Right now, there are two other people in the room besides you." + }, + "answer_label": "B", + "answer_text": "Right now, there are two other people in the room besides you.", + "q_id": "egolife_A1_D3_038_2_T1", + "correct_answer": "Right now, there are two other people in the room besides you.", + "original_options": { + "A": "Right now, there is one other person in the room besides you.", + "B": "Right now, there are two other people in the room besides you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_3_T2", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Which direction is the sink relative to me?", + "answer": "The sink is in front of you.", + "question_chinese": "洗手台在我的什么方向？", + "answer_chinese": "洗手台在你前面。", + "evidence": [ + { + "description": "The sink in front of you.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:48" + } + } + ], + "question_timestamps": "00:03:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sink is behind you.", + "B": "The sink is to your left.", + "C": "The sink is to your right, rather than directly in front of you.", + "D": "The sink is above you, rather than positioned directly in front of you.", + "E": "The sink is in front of you." + }, + "answer_label": "E", + "answer_text": "The sink is in front of you.", + "q_id": "egolife_A1_D3_038_3_T2", + "correct_answer": "The sink is in front of you.", + "original_options": { + "A": "The sink is behind you.", + "B": "The sink is to your left.", + "C": "The sink is to your right, rather than directly in front of you.", + "D": "The sink is above you, rather than positioned directly in front of you.", + "E": "The sink is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D3_038", + "video_length": "00:10:33", + "qid": "egolife_A1_D3_038_3_T1", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Which direction is the sink relative to me?", + "answer": "The sink is behind you.", + "question_chinese": "洗手台在我的什么方向？", + "answer_chinese": "洗手台在你后面。", + "evidence": [ + { + "description": "The sink is on behind of you.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:09" + } + } + ], + "question_timestamps": "00:04:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The sink is in front of you.", + "B": "The sink is behind you.", + "C": "The sink is to your left.", + "D": "The sink is to your right, rather than directly behind you now.", + "E": "The sink is above you rather than located behind your current position." + }, + "answer_label": "B", + "answer_text": "The sink is behind you.", + "q_id": "egolife_A1_D3_038_3_T1", + "correct_answer": "The sink is behind you.", + "original_options": { + "A": "The sink is in front of you.", + "B": "The sink is behind you.", + "C": "The sink is to your left.", + "D": "The sink is to your right, rather than directly behind you now.", + "E": "The sink is above you rather than located behind your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 12 + ] + } +] diff --git a/QAs/egolife_A1_D4_001-mcq.json b/QAs/egolife_A1_D4_001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..14fdf7db493d5d501ec8374bdb1fbad7878c13bd --- /dev/null +++ b/QAs/egolife_A1_D4_001-mcq.json @@ -0,0 +1,602 @@ +[ + { + "video_uid": "egolife_A1_D4_001", + "video_length": "00:11:40", + "qid": "egolife_A1_D4_001-mcq_fixed_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What are the locations of the two computers on the second floor?", + "answer": "One is on the floor to the left of the sofa, and the other is on the ground between the kitchen and the desk.", + "question_chinese": "二楼两个电脑的位置是什么？", + "answer_chinese": "一个在沙发左边的地下，一个在厨房和餐厅之间的地上。", + "evidence": [ + { + "description": "One is on the floor to the left of the sofa.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:35" + } + }, + { + "description": "The other is on the ground between the kitchen and the sofa.", + "timestamps": { + "start": "00:08:32", + "end": "00:08:33" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "One is on the floor to the left of the sofa, and the other is on the ground between the kitchen and the desk.", + "B": "Both computers are by the sofa.", + "C": "One is on the desk.", + "D": "One computer is on the chair left of the sofa, and the other is on the table by the kitchen.", + "E": "One computer is on the coffee table by the sofa, and the other is between the dining table and kitchen counter." + }, + "answer_label": "A", + "answer_text": "One is on the floor to the left of the sofa, and the other is on the ground between the kitchen and the desk.", + "q_id": "egolife_A1_D4_001-mcq_fixed_5", + "correct_answer": "One is on the floor to the left of the sofa, and the other is on the ground between the kitchen and the desk.", + "original_options": { + "A": "One is on the floor to the left of the sofa, and the other is on the ground between the kitchen and the desk.", + "B": "Both computers are by the sofa.", + "C": "One is on the desk.", + "D": "One computer is on the chair left of the sofa, and the other is on the table by the kitchen.", + "E": "One computer is on the coffee table by the sofa, and the other is between the dining table and kitchen counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D4_001", + "video_length": "00:11:40", + "qid": "egolife_A1_D4_001-mcq_fixed_6", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What are the locations of the two computers on the second floor?", + "answer": "One is on the floor to the left of the sofa, and the other is on the floor against the wall between the kitchen and the dining table.", + "question_chinese": "二楼两个电脑的位置是什么？", + "answer_chinese": "一个在沙发左边的地下，一个在厨房和餐桌之间的地上靠墙脚处。", + "evidence": [ + { + "description": "One is on the floor to the left of the sofa.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:35" + } + }, + { + "description": "The other is on the ground between the kitchen and the sofa.", + "timestamps": { + "start": "00:08:32", + "end": "00:08:33" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Both computers are on the desk.", + "B": "One is by the hallway entrance.", + "C": "One computer is on the floor to the right of the sofa, and the other is between the kitchen and dining table.", + "D": "One computer is on the floor left of the sofa, and the other is against the wall near the hallway entrance.", + "E": "One is on the floor to the left of the sofa, and the other is on the floor against the wall between the kitchen and the dining table." + }, + "answer_label": "E", + "answer_text": "One is on the floor to the left of the sofa, and the other is on the floor against the wall between the kitchen and the dining table.", + "q_id": "egolife_A1_D4_001-mcq_fixed_6", + "correct_answer": "One is on the floor to the left of the sofa, and the other is on the floor against the wall between the kitchen and the dining table.", + "original_options": { + "A": "Both computers are on the desk.", + "B": "One is by the hallway entrance.", + "C": "One computer is on the floor to the right of the sofa, and the other is between the kitchen and dining table.", + "D": "One computer is on the floor left of the sofa, and the other is against the wall near the hallway entrance.", + "E": "One is on the floor to the left of the sofa, and the other is on the floor against the wall between the kitchen and the dining table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 21 + ] + }, + { + "video_uid": "egolife_A1_D4_001", + "video_length": "00:11:40", + "qid": "egolife_A1_D4_001-mcq_fixed_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position from the entrance just now?", + "answer": "You turned left when you went out and walked past two houses to get to your current position.", + "question_chinese": "我刚刚从门口出发怎么到达现在的位置？", + "answer_chinese": "你出门左转走过两栋房屋后到达现在的位置。", + "evidence": [ + { + "description": "Right now, you turn left when you go out.", + "timestamps": { + "start": "00:02:56", + "end": "00:03:02" + } + }, + { + "description": "At this moment, you are walking straight and passing two houses.", + "timestamps": { + "start": "00:03:03", + "end": "00:03:30" + } + } + ], + "question_timestamps": "00:03:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turned right and passed two houses.", + "B": "You went straight and passed two houses.", + "C": "You went out, turned left, and walked past three houses to reach your current position.", + "D": "You went out, walked past one house, then turned left and continued until you arrived here.", + "E": "You turned left when you went out and walked past two houses to get to your current position." + }, + "answer_label": "E", + "answer_text": "You turned left when you went out and walked past two houses to get to your current position.", + "q_id": "egolife_A1_D4_001-mcq_fixed_1", + "correct_answer": "You turned left when you went out and walked past two houses to get to your current position.", + "original_options": { + "A": "You turned right and passed two houses.", + "B": "You went straight and passed two houses.", + "C": "You went out, turned left, and walked past three houses to reach your current position.", + "D": "You went out, walked past one house, then turned left and continued until you arrived here.", + "E": "You turned left when you went out and walked past two houses to get to your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_001", + "video_length": "00:11:40", + "qid": "egolife_A1_D4_001-mcq_fixed_0", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get to my current position from the entrance just now?", + "answer": "You go out, turn left, walk past two houses, then turn right and go straight to the place where you pick up the takeout.", + "question_chinese": "我刚刚从门口出发怎么到达现在的位置？", + "answer_chinese": "你出门左转走过两栋房屋后右转直走到达拿外卖的地方。", + "evidence": [ + { + "description": "Right now, you turn left when you go out.", + "timestamps": { + "start": "00:02:56", + "end": "00:03:02" + } + }, + { + "description": "At this moment, you are walking straight and passing two houses.", + "timestamps": { + "start": "00:03:03", + "end": "00:03:30" + } + }, + { + "description": "At this moment, you are turning right, walking straight, and arriving at the place to pick up the takeout.", + "timestamps": { + "start": "00:03:21", + "end": "00:03:37" + } + } + ], + "question_timestamps": "00:09:55", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turned right after going out.", + "B": "You passed three houses before turning.", + "C": "You go out, turn left, walk past one house, then turn right and go straight to pick up the takeout.", + "D": "You go out, turn left, walk past two houses, then turn right and go straight to the place where you pick up the takeout.", + "E": "You go out, turn right, walk past two houses, then turn left and go straight to the takeout pickup spot." + }, + "answer_label": "D", + "answer_text": "You go out, turn left, walk past two houses, then turn right and go straight to the place where you pick up the takeout.", + "q_id": "egolife_A1_D4_001-mcq_fixed_0", + "correct_answer": "You go out, turn left, walk past two houses, then turn right and go straight to the place where you pick up the takeout.", + "original_options": { + "A": "You turned right after going out.", + "B": "You passed three houses before turning.", + "C": "You go out, turn left, walk past one house, then turn right and go straight to pick up the takeout.", + "D": "You go out, turn left, walk past two houses, then turn right and go straight to the place where you pick up the takeout.", + "E": "You go out, turn right, walk past two houses, then turn left and go straight to the takeout pickup spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_001", + "video_length": "00:11:40", + "qid": "egolife_A1_D4_001-mcq_fixed_8", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers did I just debug in the bedroom on the first floor?", + "answer": "One.", + "question_chinese": "我刚刚调试了一楼卧室的几台电脑？", + "answer_chinese": "你刚刚调试了一楼卧室的两台电脑。", + "evidence": [ + { + "description": "Right now, you have finished debugging the computer at the bottom.", + "timestamps": { + "start": "00:04:58", + "end": "00:05:00" + } + }, + { + "description": "Right now, you have debugged the computer on the second-to-last row.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You debugged two computers in the bedroom on the first floor just now.", + "E": "You debugged three computers in the bedroom on the first floor just now." + }, + "answer_label": "A", + "answer_text": "One.", + "q_id": "egolife_A1_D4_001-mcq_fixed_8", + "correct_answer": "One.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You debugged two computers in the bedroom on the first floor just now.", + "E": "You debugged three computers in the bedroom on the first floor just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_001", + "video_length": "00:11:40", + "qid": "egolife_A1_D4_001-mcq_fixed_7", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers did I just debug in the bedroom on the first floor?", + "answer": "Three.", + "question_chinese": "我刚刚调试了一楼卧室的几台电脑？", + "answer_chinese": "你刚刚调试了一楼卧室的三台电脑。", + "evidence": [ + { + "description": "Right now, you have finished debugging the computer at the bottom.", + "timestamps": { + "start": "00:04:58", + "end": "00:05:00" + } + }, + { + "description": "Right now, you have debugged the computer on the second-to-last row.", + "timestamps": { + "start": "00:05:07", + "end": "00:05:08" + } + }, + { + "description": "Right now, you have debugged the computer on the right-hand side of the first row.", + "timestamps": { + "start": "00:05:15", + "end": "00:05:16" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "You debugged two computers in that bedroom on the first floor earlier today.", + "E": "You debugged four computers in the first-floor bedroom and then stopped debugging for a while." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "egolife_A1_D4_001-mcq_fixed_7", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "You debugged two computers in that bedroom on the first floor earlier today.", + "E": "You debugged four computers in the first-floor bedroom and then stopped debugging for a while." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_001", + "video_length": "00:11:40", + "qid": "egolife_A1_D4_001_3_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many times have I debugged the second-floor computer?", + "answer": "You already debugged it once.", + "question_chinese": "我目前为止已经调试了二楼的电脑几次？", + "answer_chinese": "你已经调试了一次。", + "evidence": [ + { + "description": "At this moment, you are debugging a computer.", + "timestamps": { + "start": "00:06:58", + "end": "00:08:27" + } + } + ], + "question_timestamps": "00:08:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You already debugged it zero times.", + "B": "You already debugged it twice.", + "C": "So far, you have debugged the second-floor computer three times in total.", + "D": "Up to now, you have debugged the second-floor computer four separate times.", + "E": "You already debugged it once." + }, + "answer_label": "E", + "answer_text": "You already debugged it once.", + "q_id": "egolife_A1_D4_001_3_T1", + "correct_answer": "You already debugged it once.", + "original_options": { + "A": "You already debugged it zero times.", + "B": "You already debugged it twice.", + "C": "So far, you have debugged the second-floor computer three times in total.", + "D": "Up to now, you have debugged the second-floor computer four separate times.", + "E": "You already debugged it once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_001", + "video_length": "00:11:40", + "qid": "egolife_A1_D4_001_3_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many times have I debugged the second-floor computer?", + "answer": "You already debugged it twice.", + "question_chinese": "我目前为止已经调试了二楼的电脑几次？", + "answer_chinese": "你已经调试了两次。", + "evidence": [ + { + "description": "At this moment, you are debugging a computer.", + "timestamps": { + "start": "00:06:58", + "end": "00:08:27" + } + }, + { + "description": "You debugged the computer for the second time.", + "timestamps": { + "start": "00:08:37", + "end": "00:08:38" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You already debugged it twice.", + "B": "You already debugged it once.", + "C": "You already debugged it three times.", + "D": "So far, you have debugged the second-floor computer four times altogether today.", + "E": "Up to this point, you haven’t debugged the second-floor computer at all yet." + }, + "answer_label": "A", + "answer_text": "You already debugged it twice.", + "q_id": "egolife_A1_D4_001_3_T2", + "correct_answer": "You already debugged it twice.", + "original_options": { + "A": "You already debugged it twice.", + "B": "You already debugged it once.", + "C": "You already debugged it three times.", + "D": "So far, you have debugged the second-floor computer four times altogether today.", + "E": "Up to this point, you haven’t debugged the second-floor computer at all yet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_001", + "video_length": "00:11:40", + "qid": "egolife_A1_D4_001-mcq_fixed_2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did the glasses take?", + "answer": "YOU took the three pairs of glasses from the table in the bedroom on the first floor.", + "question_chinese": "眼镜经过了什么样的途径？", + "answer_chinese": "三个眼镜被我从一楼卧室的桌子上拿了出来。", + "evidence": [ + { + "description": "I took the three pairs of glasses from the table in the bedroom on the first floor.", + "timestamps": { + "start": "00:05:42", + "end": "00:05:45" + } + } + ], + "question_timestamps": "00:05:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "YOU took the three pairs of glasses from the table in the bedroom on the first floor.", + "B": "You took one pair of glasses.", + "C": "You took three glasses from the table.", + "D": "You picked up three pairs of glasses from a dresser upstairs on the second floor.", + "E": "You carried the three pairs of glasses from the first-floor bedroom table into the hallway, then left them there." + }, + "answer_label": "A", + "answer_text": "YOU took the three pairs of glasses from the table in the bedroom on the first floor.", + "q_id": "egolife_A1_D4_001-mcq_fixed_2", + "correct_answer": "YOU took the three pairs of glasses from the table in the bedroom on the first floor.", + "original_options": { + "A": "YOU took the three pairs of glasses from the table in the bedroom on the first floor.", + "B": "You took one pair of glasses.", + "C": "You took three glasses from the table.", + "D": "You picked up three pairs of glasses from a dresser upstairs on the second floor.", + "E": "You carried the three pairs of glasses from the first-floor bedroom table into the hallway, then left them there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_001", + "video_length": "00:11:40", + "qid": "egolife_A1_D4_001-mcq_fixed_3", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did the glasses take?", + "answer": "YOU took the three pairs of glasses from the table in the bedroom on the first floor and then placed them on the table on the second floor, arranging them.", + "question_chinese": "眼镜经过了什么样的途径？", + "answer_chinese": "三个眼镜被我从一楼卧室的桌子上拿了出来，然后被我放到了二楼桌子上摆开。", + "evidence": [ + { + "description": "I took the three pairs of glasses from the table in the bedroom on the first floor.", + "timestamps": { + "start": "00:05:42", + "end": "00:05:45" + } + }, + { + "description": "Right now, I have placed the glasses on the second-floor table and arranged them.", + "timestamps": { + "start": "00:06:29", + "end": "00:06:31" + } + } + ], + "question_timestamps": "00:07:01", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You carried them downstairs to the living room.", + "B": "YOU took the three pairs of glasses from the table in the bedroom on the first floor and then placed them on the table on the second floor, arranging them.", + "C": "You left the glasses on the first-floor table.", + "D": "You took three pairs of glasses from the first-floor bedroom table and put them onto a shelf upstairs.", + "E": "You collected the glasses from the second-floor table and brought them back down to the first-floor bedroom." + }, + "answer_label": "B", + "answer_text": "YOU took the three pairs of glasses from the table in the bedroom on the first floor and then placed them on the table on the second floor, arranging them.", + "q_id": "egolife_A1_D4_001-mcq_fixed_3", + "correct_answer": "YOU took the three pairs of glasses from the table in the bedroom on the first floor and then placed them on the table on the second floor, arranging them.", + "original_options": { + "A": "You carried them downstairs to the living room.", + "B": "YOU took the three pairs of glasses from the table in the bedroom on the first floor and then placed them on the table on the second floor, arranging them.", + "C": "You left the glasses on the first-floor table.", + "D": "You took three pairs of glasses from the first-floor bedroom table and put them onto a shelf upstairs.", + "E": "You collected the glasses from the second-floor table and brought them back down to the first-floor bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D4_002-mcq.json b/QAs/egolife_A1_D4_002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..513002352790eaae420e5f8291f4cfecea629dd2 --- /dev/null +++ b/QAs/egolife_A1_D4_002-mcq.json @@ -0,0 +1,613 @@ +[ + { + "video_uid": "egolife_A1_D4_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_002-mcq_fixed_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the refrigerator on the second floor located?", + "answer": "The refrigerator is placed behind the door.", + "question_chinese": "二楼的冰箱放在哪里？", + "answer_chinese": "冰箱放在门的后面。", + "evidence": [ + { + "description": "At this moment, you can see the refrigerator placed behind the door", + "timestamps": { + "start": "00:09:22", + "end": "00:09:23" + } + } + ], + "question_timestamps": "00:10:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is placed behind the door.", + "B": "It is under the window.", + "C": "It is behind the couch.", + "D": "The refrigerator is positioned next to the stairwell on the second floor, not behind any door.", + "E": "On the second floor, the refrigerator sits between the lounge seating area and the vending machine area." + }, + "answer_label": "A", + "answer_text": "The refrigerator is placed behind the door.", + "q_id": "egolife_A1_D4_002-mcq_fixed_7", + "correct_answer": "The refrigerator is placed behind the door.", + "original_options": { + "A": "The refrigerator is placed behind the door.", + "B": "It is under the window.", + "C": "It is behind the couch.", + "D": "The refrigerator is positioned next to the stairwell on the second floor, not behind any door.", + "E": "On the second floor, the refrigerator sits between the lounge seating area and the vending machine area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_002-mcq_fixed_6", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the refrigerator on the second floor located?", + "answer": "The refrigerator is placed in front of the kitchen.", + "question_chinese": "二楼的冰箱放在哪里？", + "answer_chinese": "冰箱放在厨房前面。", + "evidence": [ + { + "description": "At this moment, you can see the refrigerator placed in front of the kitchen.", + "timestamps": { + "start": "00:09:22", + "end": "00:09:23" + } + }, + { + "description": "At this moment, you can see the refrigerator placed in front of the kitchen.", + "timestamps": { + "start": "00:00:51", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:11:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is placed in front of the kitchen.", + "B": "It is behind the kitchen.", + "C": "It is left of the kitchen.", + "D": "The refrigerator is placed against the wall beside the pantry door, rather than in front of the kitchen.", + "E": "You will find the refrigerator by the stairwell next to shelves, not directly in front of the kitchen." + }, + "answer_label": "A", + "answer_text": "The refrigerator is placed in front of the kitchen.", + "q_id": "egolife_A1_D4_002-mcq_fixed_6", + "correct_answer": "The refrigerator is placed in front of the kitchen.", + "original_options": { + "A": "The refrigerator is placed in front of the kitchen.", + "B": "It is behind the kitchen.", + "C": "It is left of the kitchen.", + "D": "The refrigerator is placed against the wall beside the pantry door, rather than in front of the kitchen.", + "E": "You will find the refrigerator by the stairwell next to shelves, not directly in front of the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_002_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just tidy up the sofa?", + "answer": "No, you didn't just tidy up the sofa.", + "question_chinese": "我刚刚收拾沙发了吗？", + "answer_chinese": "没有，你刚刚没有收拾沙发。", + "evidence": [ + { + "description": "So far, you haven't tidied the sofa.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:01:58", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you just tidied up the sofa.", + "B": "No, you didn't just tidy up the sofa." + }, + "answer_label": "B", + "answer_text": "No, you didn't just tidy up the sofa.", + "q_id": "egolife_A1_D4_002_1", + "correct_answer": "No, you didn't just tidy up the sofa.", + "original_options": { + "A": "Yes, you just tidied up the sofa.", + "B": "No, you didn't just tidy up the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D4_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_002_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just tidy up the sofa?", + "answer": "Yes, you just tidied up the sofa.", + "question_chinese": "我刚刚收拾沙发了吗？", + "answer_chinese": "是的，你刚刚收拾了沙发。", + "evidence": [ + { + "description": "At this moment, you are tidying up the sofa.", + "timestamps": { + "start": "00:04:59", + "end": "00:05:20" + } + } + ], + "question_timestamps": "00:08:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you just tidied up the sofa.", + "B": "No, you did not tidy up the sofa." + }, + "answer_label": "A", + "answer_text": "Yes, you just tidied up the sofa.", + "q_id": "egolife_A1_D4_002_2", + "correct_answer": "Yes, you just tidied up the sofa.", + "original_options": { + "A": "Yes, you just tidied up the sofa.", + "B": "No, you did not tidy up the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D4_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_002_7", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route should I take to get from the sofa to the kitchen?", + "answer": "You should get up, turn right, and go straight for two steps to reach the kitchen.", + "question_chinese": "我如果想要从沙发到厨房要经过什么途径？", + "answer_chinese": "你应该起身右转直走两步到达厨房。", + "evidence": [ + { + "description": "At this moment, you get up from the sofa.", + "timestamps": { + "start": "00:05:41", + "end": "00:05:42" + } + }, + { + "description": "At this moment, you begin to turn right.", + "timestamps": { + "start": "00:05:43", + "end": "00:05:44" + } + }, + { + "description": "At this moment, you walk straight to the front of the kitchen, which is on your right side.", + "timestamps": { + "start": "00:05:45", + "end": "00:05:46" + } + } + ], + "question_timestamps": "00:07:58", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Get up and turn left.", + "B": "You should get up, turn right, and go straight for two steps to reach the kitchen.", + "C": "Turn right and take three steps.", + "D": "You should stand up, turn left, and walk straight for two steps to reach the kitchen.", + "E": "You should get up, turn right, and go straight for one step to reach the kitchen." + }, + "answer_label": "B", + "answer_text": "You should get up, turn right, and go straight for two steps to reach the kitchen.", + "q_id": "egolife_A1_D4_002_7", + "correct_answer": "You should get up, turn right, and go straight for two steps to reach the kitchen.", + "original_options": { + "A": "Get up and turn left.", + "B": "You should get up, turn right, and go straight for two steps to reach the kitchen.", + "C": "Turn right and take three steps.", + "D": "You should stand up, turn left, and walk straight for two steps to reach the kitchen.", + "E": "You should get up, turn right, and go straight for one step to reach the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_002_8", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route do I need to take to get from the sofa to the kitchen?", + "answer": "You should get up, turn right, and go straight for two steps to reach the kitchen.", + "question_chinese": "我如果想要从沙发到厨房要经过什么途径？", + "answer_chinese": "你应该起身右转直走两步到达厨房。", + "evidence": [ + { + "description": "At this moment, you get up from the sofa.", + "timestamps": { + "start": "00:05:41", + "end": "00:05:42" + } + }, + { + "description": "At this moment, you begin to turn right.", + "timestamps": { + "start": "00:05:43", + "end": "00:05:44" + } + }, + { + "description": "At this moment, you walk straight to the front of the kitchen, which is on your right side.", + "timestamps": { + "start": "00:05:45", + "end": "00:05:46" + } + } + ], + "question_timestamps": "00:10:22", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left and go straight.", + "B": "Turn right and take one step.", + "C": "You should get up, turn right, and go straight for three steps to reach the kitchen.", + "D": "You should get up, turn right, and go straight for two steps to reach the kitchen.", + "E": "You should stand up, turn left, and go straight for four steps to reach the kitchen." + }, + "answer_label": "D", + "answer_text": "You should get up, turn right, and go straight for two steps to reach the kitchen.", + "q_id": "egolife_A1_D4_002_8", + "correct_answer": "You should get up, turn right, and go straight for two steps to reach the kitchen.", + "original_options": { + "A": "Turn left and go straight.", + "B": "Turn right and take one step.", + "C": "You should get up, turn right, and go straight for three steps to reach the kitchen.", + "D": "You should get up, turn right, and go straight for two steps to reach the kitchen.", + "E": "You should stand up, turn left, and go straight for four steps to reach the kitchen." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_002_4_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I walk straight ahead right now, can I reach the kitchen?", + "answer": "You could reach the kitchen. The kitchen was in front of you.", + "question_chinese": "我现在直走可以到达厨房吗？", + "answer_chinese": "你可以到达厨房。厨房在你的前边。", + "evidence": [ + { + "description": "At this moment, the kitchen is on your right. You need to turn right and go straight to reach the kitchen.", + "timestamps": { + "start": "00:02:04", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:02:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You could not reach the kitchen. The kitchen was to your right.", + "B": "You could reach the kitchen. The kitchen was in front of you." + }, + "answer_label": "B", + "answer_text": "You could reach the kitchen. The kitchen was in front of you.", + "q_id": "egolife_A1_D4_002_4_T1", + "correct_answer": "You could reach the kitchen. The kitchen was in front of you.", + "original_options": { + "A": "You could not reach the kitchen. The kitchen was to your right.", + "B": "You could reach the kitchen. The kitchen was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_002-mcq_fixed_2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get to the kitchen by going straight now?", + "answer": "You can not reach the kitchen.You need to turn right and go straight to reach the kitchen.", + "question_chinese": "我现在直走可以到达厨房吗？", + "answer_chinese": "你不可以到达厨房，你需要右转直走才能到达。", + "evidence": [ + { + "description": "At this moment, you are sitting on the sofa, with the bedroom opposite you.", + "timestamps": { + "start": "00:04:51", + "end": "00:04:52" + } + }, + { + "description": "At this moment, the kitchen is on your right. You need to turn right and go straight to reach the kitchen.", + "timestamps": { + "start": "00:04:51", + "end": "00:04:52" + } + } + ], + "question_timestamps": "00:04:53", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can reach the kitchen by going straight without turning right first.", + "B": "You can not reach the kitchen.You need to turn right and go straight to reach the kitchen." + }, + "answer_label": "B", + "answer_text": "You can not reach the kitchen.You need to turn right and go straight to reach the kitchen.", + "q_id": "egolife_A1_D4_002-mcq_fixed_2", + "correct_answer": "You can not reach the kitchen.You need to turn right and go straight to reach the kitchen.", + "original_options": { + "A": "You can reach the kitchen by going straight without turning right first.", + "B": "You can not reach the kitchen.You need to turn right and go straight to reach the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_002_4_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I walk straight ahead right now, can I reach the kitchen?", + "answer": "You could not reach the kitchen. The kitchen was to your right.", + "question_chinese": "我现在直走可以到达厨房吗？", + "answer_chinese": "你不可以到达厨房。厨房在你的右边。", + "evidence": [ + { + "description": "At this moment, you are sitting on the sofa, with the bedroom opposite you. the kitchen is on your right. You need to turn right and go straight to reach the kitchen.", + "timestamps": { + "start": "00:04:51", + "end": "00:04:52" + } + } + ], + "question_timestamps": "00:04:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You could not reach the kitchen. The kitchen was to your right.", + "B": "You could reach the kitchen. The kitchen was in front of you." + }, + "answer_label": "A", + "answer_text": "You could not reach the kitchen. The kitchen was to your right.", + "q_id": "egolife_A1_D4_002_4_T2", + "correct_answer": "You could not reach the kitchen. The kitchen was to your right.", + "original_options": { + "A": "You could not reach the kitchen. The kitchen was to your right.", + "B": "You could reach the kitchen. The kitchen was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_002-mcq_fixed_8", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my phone now?", + "answer": "Your phone is now at the left end of the sofa.", + "question_chinese": "我的手机现在在哪里？", + "answer_chinese": "你的手机现在在沙发的左端。", + "evidence": [ + { + "description": "At this moment, you placed the phone at the far left end of the sofa.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:04:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Your phone is at the right end.", + "B": "Your phone is on the coffee table.", + "C": "Your phone is now at the left end of the sofa.", + "D": "Your phone is resting on the coffee table directly in front of the sofa, not on the cushions.", + "E": "Your phone is inside your backpack next to the right end of the sofa, not placed on the sofa." + }, + "answer_label": "C", + "answer_text": "Your phone is now at the left end of the sofa.", + "q_id": "egolife_A1_D4_002-mcq_fixed_8", + "correct_answer": "Your phone is now at the left end of the sofa.", + "original_options": { + "A": "Your phone is at the right end.", + "B": "Your phone is on the coffee table.", + "C": "Your phone is now at the left end of the sofa.", + "D": "Your phone is resting on the coffee table directly in front of the sofa, not on the cushions.", + "E": "Your phone is inside your backpack next to the right end of the sofa, not placed on the sofa." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_002-mcq_fixed_9", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my phone now?", + "answer": "Your phone is on you right now.", + "question_chinese": "我的手机现在在哪里？", + "answer_chinese": "你的手机现在你身上。", + "evidence": [ + { + "description": "At this moment, you placed the phone at the far left end of the sofa.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + }, + { + "description": "Right now, you picked up the phone from the sofa and put it in your pocket.", + "timestamps": { + "start": "00:05:42", + "end": "00:05:43" + } + } + ], + "question_timestamps": "00:06:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Your phone is beside you right now.", + "B": "Your phone is under you right now.", + "C": "Your phone is probably still on your desk where you last used it, not on your body.", + "D": "Your phone is likely in your bag under your notebook, rather than being on you right now.", + "E": "Your phone is on you right now." + }, + "answer_label": "E", + "answer_text": "Your phone is on you right now.", + "q_id": "egolife_A1_D4_002-mcq_fixed_9", + "correct_answer": "Your phone is on you right now.", + "original_options": { + "A": "Your phone is beside you right now.", + "B": "Your phone is under you right now.", + "C": "Your phone is probably still on your desk where you last used it, not on your body.", + "D": "Your phone is likely in your bag under your notebook, rather than being on you right now.", + "E": "Your phone is on you right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D4_003-mcq.json b/QAs/egolife_A1_D4_003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fe4fcbb24df67df13b1dab2f77da73cbcd057ada --- /dev/null +++ b/QAs/egolife_A1_D4_003-mcq.json @@ -0,0 +1,461 @@ +[ + { + "video_uid": "egolife_A1_D4_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_003-mcq_fixed_2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the relative position of the computer closest to me among the two computers on the second floor?", + "answer": "The computer closest to you, which is near the kitchen, is on the floor at your feet.", + "question_chinese": "二楼两个电脑离我最近的一个和我的相对位置什么？", + "answer_chinese": "在靠近厨房的一台电脑离我最近，在我脚下的地上放着。", + "evidence": [ + { + "description": "From this point in time up until the moment of the question, you have been sitting on a stool next to the computer near the kitchen, so the computer is at your feet.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's on the floor to your right.", + "B": "It's on the floor behind you.", + "C": "The computer closest to you, which is near the kitchen, is on the floor at your feet.", + "D": "The closest computer, near the kitchen, is sitting on a low table just in front of you.", + "E": "The computer nearest you by the kitchen is tucked beside the wall, a step ahead on the floor." + }, + "answer_label": "C", + "answer_text": "The computer closest to you, which is near the kitchen, is on the floor at your feet.", + "q_id": "egolife_A1_D4_003-mcq_fixed_2", + "correct_answer": "The computer closest to you, which is near the kitchen, is on the floor at your feet.", + "original_options": { + "A": "It's on the floor to your right.", + "B": "It's on the floor behind you.", + "C": "The computer closest to you, which is near the kitchen, is on the floor at your feet.", + "D": "The closest computer, near the kitchen, is sitting on a low table just in front of you.", + "E": "The computer nearest you by the kitchen is tucked beside the wall, a step ahead on the floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_003-mcq_fixed_3", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the relative position of the computer closest to me among the two computers on the second floor?", + "answer": "The computer closest to you, which is near the sofa, is on the floor less than a meter to your left.", + "question_chinese": "二楼两个电脑离我最近的一个和我的相对位置什么？", + "answer_chinese": "靠近沙发的电脑离我最近，它就在我左边不到一米的地下放着。", + "evidence": [ + { + "description": "From this point in time up until the moment of the question, you have been sitting on a stool next to the computer near the kitchen, so the computer is at your feet.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "Right now, I'm lying on the sofa with the computer in front of me.", + "timestamps": { + "start": "00:02:09", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:04:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It's on the floor to your right.", + "B": "It's about a meter behind you.", + "C": "The closest computer near the sofa is on the floor a little over a meter in front of you.", + "D": "The computer closest to you, which is near the sofa, is on the floor less than a meter to your left.", + "E": "The computer nearest you by the sofa is positioned on the floor just ahead and slightly to your right." + }, + "answer_label": "D", + "answer_text": "The computer closest to you, which is near the sofa, is on the floor less than a meter to your left.", + "q_id": "egolife_A1_D4_003-mcq_fixed_3", + "correct_answer": "The computer closest to you, which is near the sofa, is on the floor less than a meter to your left.", + "original_options": { + "A": "It's on the floor to your right.", + "B": "It's about a meter behind you.", + "C": "The closest computer near the sofa is on the floor a little over a meter in front of you.", + "D": "The computer closest to you, which is near the sofa, is on the floor less than a meter to your left.", + "E": "The computer nearest you by the sofa is positioned on the floor just ahead and slightly to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_003-mcq_fixed_8", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I won playing games with the girl in the second-floor living room so far?", + "answer": "Three.", + "question_chinese": "我到目前为止在二楼客厅和女生玩游戏赢了几次？", + "answer_chinese": "你玩游戏赢了两次。", + "evidence": [ + { + "description": "At this moment, you won for the first time.", + "timestamps": { + "start": "00:05:48", + "end": "00:05:54" + } + }, + { + "description": "At this moment, you won for the second time.", + "timestamps": { + "start": "00:05:55", + "end": "00:05:58" + } + } + ], + "question_timestamps": "00:05:59", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You've won two times so far against the girl in the second-floor living room.", + "E": "You have four wins so far from the games you've played with the girl upstairs in the living room." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "egolife_A1_D4_003-mcq_fixed_8", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "You've won two times so far against the girl in the second-floor living room.", + "E": "You have four wins so far from the games you've played with the girl upstairs in the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_003-mcq_fixed_9", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I won playing games with the girl in the second-floor living room so far?", + "answer": "Four.", + "question_chinese": "我到目前为止在二楼客厅和女生玩游戏赢了几次？", + "answer_chinese": "你刚刚玩游戏赢了三次。", + "evidence": [ + { + "description": "At this moment, you won for the first time.", + "timestamps": { + "start": "00:05:48", + "end": "00:05:54" + } + }, + { + "description": "At this moment, you won for the second time.", + "timestamps": { + "start": "00:05:55", + "end": "00:05:58" + } + }, + { + "description": "At this moment, you won for the third time.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:02" + } + } + ], + "question_timestamps": "00:10:04", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "So far, you've won three times while playing games with the girl in the second-floor living room.", + "E": "Up to now, you have five total wins from your games with the girl in the upstairs living room." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "egolife_A1_D4_003-mcq_fixed_9", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "So far, you've won three times while playing games with the girl in the second-floor living room.", + "E": "Up to now, you have five total wins from your games with the girl in the upstairs living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_003-mcq_fixed_5", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I find on the table after I finished playing the game?", + "answer": "You have just finished the game, haven't approached the table yet, and haven't discovered anything.", + "question_chinese": "我玩完游戏后在桌子上发现了什么？", + "answer_chinese": "你刚刚结束游戏，还没有靠近桌子，什么也没有发现。", + "evidence": [ + { + "description": "You have just finished the game, haven't approached the table yet, and haven't discovered anything.", + "timestamps": { + "start": "00:06:03", + "end": "00:06:04" + } + } + ], + "question_timestamps": "00:06:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You found a notebook.", + "B": "You found some coins.", + "C": "You have just finished the game, haven't approached the table yet, and haven't discovered anything.", + "D": "After finishing the game, you walked over to the table and noticed a blue notebook lying there.", + "E": "Once the game ended, you approached the table and spotted a small pile of coins left on top." + }, + "answer_label": "C", + "answer_text": "You have just finished the game, haven't approached the table yet, and haven't discovered anything.", + "q_id": "egolife_A1_D4_003-mcq_fixed_5", + "correct_answer": "You have just finished the game, haven't approached the table yet, and haven't discovered anything.", + "original_options": { + "A": "You found a notebook.", + "B": "You found some coins.", + "C": "You have just finished the game, haven't approached the table yet, and haven't discovered anything.", + "D": "After finishing the game, you walked over to the table and noticed a blue notebook lying there.", + "E": "Once the game ended, you approached the table and spotted a small pile of coins left on top." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_003-mcq_fixed_4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I find on the table after I finished playing the game?", + "answer": "You found a hairpin.", + "question_chinese": "我玩完游戏后在桌子上发现了什么？", + "answer_chinese": "你发现了一个簪子。", + "evidence": [ + { + "description": "At this moment, you find a hairpin.", + "timestamps": { + "start": "00:06:17", + "end": "00:06:18" + } + } + ], + "question_timestamps": "00:09:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You found a notebook.", + "B": "You found a pen.", + "C": "You picked up a pencil case from the table after the game ended.", + "D": "You found a hairpin.", + "E": "You came across a small metal-and-plastic accessory on the table, but it was a different clip." + }, + "answer_label": "D", + "answer_text": "You found a hairpin.", + "q_id": "egolife_A1_D4_003-mcq_fixed_4", + "correct_answer": "You found a hairpin.", + "original_options": { + "A": "You found a notebook.", + "B": "You found a pen.", + "C": "You picked up a pencil case from the table after the game ended.", + "D": "You found a hairpin.", + "E": "You came across a small metal-and-plastic accessory on the table, but it was a different clip." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_003-mcq_fixed_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many computers on the second floor have I debugged?", + "answer": "Five.", + "question_chinese": "我目前为止已经调试了二楼的几个电脑？", + "answer_chinese": "你已经调试了二楼的一个电脑。", + "evidence": [ + { + "description": "At this moment, you are debugging a computer.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:01:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "So far, you have debugged four computers on the second floor in total.", + "D": "Five.", + "E": "Up to this point, you've only debugged six different computers on the second floor." + }, + "answer_label": "D", + "answer_text": "Five.", + "q_id": "egolife_A1_D4_003-mcq_fixed_1", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "So far, you have debugged four computers on the second floor in total.", + "D": "Five.", + "E": "Up to this point, you've only debugged six different computers on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_003-mcq_fixed_0", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many computers on the second floor have I debugged?", + "answer": "Six.", + "question_chinese": "我目前为止已经调试了二楼的几个电脑？", + "answer_chinese": "你已经调试了二楼的两个电脑。", + "evidence": [ + { + "description": "At this moment, you are debugging a computer.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:44" + } + }, + { + "description": "At this moment, you are debugging another computer.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:33" + } + } + ], + "question_timestamps": "00:10:04", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "So far, you have debugged five computers on the second floor altogether.", + "D": "Six.", + "E": "Up to now, you've debugged seven computers on the second floor, not six." + }, + "answer_label": "D", + "answer_text": "Six.", + "q_id": "egolife_A1_D4_003-mcq_fixed_0", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "So far, you have debugged five computers on the second floor altogether.", + "D": "Six.", + "E": "Up to now, you've debugged seven computers on the second floor, not six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D4_004-mcq.json b/QAs/egolife_A1_D4_004-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..04e31a44cf40ea2c2f37dc6e14c5bed1e096ab00 --- /dev/null +++ b/QAs/egolife_A1_D4_004-mcq.json @@ -0,0 +1,708 @@ +[ + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004-mcq_fixed_9", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the bedroom on the first floor now in relation to my position?", + "answer": "The bedroom on the first floor is one or two meters behind you.", + "question_chinese": "一楼的卧室现在在我的哪个位置？", + "answer_chinese": "一楼卧室在你的后面一两米处。", + "evidence": [ + { + "description": "At this moment, you can see the bedroom is to the left of the dining table.", + "timestamps": { + "start": "00:06:26", + "end": "00:06:27" + } + }, + { + "description": "Since the dining table is on your left hand side at this moment, the bedroom is behind you.", + "timestamps": { + "start": "00:07:02", + "end": "00:07:03" + } + } + ], + "question_timestamps": "00:07:04", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "The bedroom on the first floor is one or two meters behind you.", + "C": "About two meters to your right.", + "D": "The first-floor bedroom is a few meters ahead, past the hallway opening from where you stand.", + "E": "The first-floor bedroom is slightly left of you, roughly one to two meters away along the same wall." + }, + "answer_label": "B", + "answer_text": "The bedroom on the first floor is one or two meters behind you.", + "q_id": "egolife_A1_D4_004-mcq_fixed_9", + "correct_answer": "The bedroom on the first floor is one or two meters behind you.", + "original_options": { + "A": "In front of you.", + "B": "The bedroom on the first floor is one or two meters behind you.", + "C": "About two meters to your right.", + "D": "The first-floor bedroom is a few meters ahead, past the hallway opening from where you stand.", + "E": "The first-floor bedroom is slightly left of you, roughly one to two meters away along the same wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004-mcq_fixed_8", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the bedroom on the first floor now in relation to my position?", + "answer": "The bedroom on the first floor is two or three meters away to your left.", + "question_chinese": "一楼的卧室现在在我的哪个位置？", + "answer_chinese": "一楼的卧室在你的左边两三米的距离。", + "evidence": [ + { + "description": "At this moment, the bedroom can be seen to the left of the dining table.", + "timestamps": { + "start": "00:07:19", + "end": "00:07:20" + } + }, + { + "description": "Now you are standing near the dining table, so the bedroom is on your left.", + "timestamps": { + "start": "00:07:49", + "end": "00:07:50" + } + } + ], + "question_timestamps": "00:08:51", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It’s about two meters behind you.", + "B": "The bedroom on the first floor is two or three meters away to your left.", + "C": "Two or three meters to your right.", + "D": "The first-floor bedroom is directly in front of you, a couple of meters away at the end of the hall.", + "E": "The first-floor bedroom is a few meters away behind you, just across the hallway from your current spot." + }, + "answer_label": "B", + "answer_text": "The bedroom on the first floor is two or three meters away to your left.", + "q_id": "egolife_A1_D4_004-mcq_fixed_8", + "correct_answer": "The bedroom on the first floor is two or three meters away to your left.", + "original_options": { + "A": "It’s about two meters behind you.", + "B": "The bedroom on the first floor is two or three meters away to your left.", + "C": "Two or three meters to your right.", + "D": "The first-floor bedroom is directly in front of you, a couple of meters away at the end of the hall.", + "E": "The first-floor bedroom is a few meters away behind you, just across the hallway from your current spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004_2_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "So far, what trajectory has the bread on the plate on the first-floor dining table gone through?", + "answer": "You picked up the bread on the plate and ate it.", + "question_chinese": "一楼餐桌的盘子里的面包到目前为止经历了怎样的运动轨迹？", + "answer_chinese": "盘子上的面包被你拿起来吃。", + "evidence": [ + { + "description": "You picked up the bread on the plate and ate it.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:04:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You left it on the plate.", + "B": "You moved it to the table edge.", + "C": "You picked up the bread, looked at it briefly, and placed it back onto the plate without eating.", + "D": "You picked up the bread on the plate and ate it.", + "E": "You lifted the bread off the plate and set it down on the table surface beside the plate." + }, + "answer_label": "D", + "answer_text": "You picked up the bread on the plate and ate it.", + "q_id": "egolife_A1_D4_004_2_T1", + "correct_answer": "You picked up the bread on the plate and ate it.", + "original_options": { + "A": "You left it on the plate.", + "B": "You moved it to the table edge.", + "C": "You picked up the bread, looked at it briefly, and placed it back onto the plate without eating.", + "D": "You picked up the bread on the plate and ate it.", + "E": "You lifted the bread off the plate and set it down on the table surface beside the plate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004-mcq_fixed_1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What has been the trajectory of the bread on the plate on the dining table on the first floor so far?", + "answer": "You picked up and ate the bread on the upper layer of the plate; the rest is still in the plate.", + "question_chinese": "一楼餐桌的盘子里的面包到目前为止经历了怎样的运动轨迹？", + "answer_chinese": "盘子上层的面包被你拿起来吃了，剩下的还在盘子里。", + "evidence": [ + { + "description": "At this moment, you take the bread from the plate next to the computer on the dining table.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:20" + } + }, + { + "description": "At this moment, you can see that the bread on the plate is still in its original place.", + "timestamps": { + "start": "00:04:09", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You ate all the bread.", + "B": "You didn’t eat any bread.", + "C": "You picked up the bread from the upper layer, moved it to the table, and left the rest on the plate.", + "D": "You picked up and ate the bread on the upper layer of the plate; the rest is still in the plate.", + "E": "You took a piece from the upper layer, but you put it back and left everything still sitting on the plate." + }, + "answer_label": "D", + "answer_text": "You picked up and ate the bread on the upper layer of the plate; the rest is still in the plate.", + "q_id": "egolife_A1_D4_004-mcq_fixed_1", + "correct_answer": "You picked up and ate the bread on the upper layer of the plate; the rest is still in the plate.", + "original_options": { + "A": "You ate all the bread.", + "B": "You didn’t eat any bread.", + "C": "You picked up the bread from the upper layer, moved it to the table, and left the rest on the plate.", + "D": "You picked up and ate the bread on the upper layer of the plate; the rest is still in the plate.", + "E": "You took a piece from the upper layer, but you put it back and left everything still sitting on the plate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004-mcq_fixed_4", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I walk if I want to get something from the refrigerator right now?", + "answer": "You should turn right, take one step, then turn left and take two steps; the refrigerator will be right in front of you.", + "question_chinese": "如果我此时要从冰箱里拿东西应该怎么走？", + "answer_chinese": "你应该右转走一步再左转走两步转，冰箱就在你面前。", + "evidence": [ + { + "description": "At this moment, you can see the refrigerator behind the door.", + "timestamps": { + "start": "00:05:45", + "end": "00:05:47" + } + }, + { + "description": "At this moment, you can see the refrigerator behind the door.", + "timestamps": { + "start": "00:10:13", + "end": "00:10:14" + } + } + ], + "question_timestamps": "00:10:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You should turn right, take one step, then turn left and take two steps; the refrigerator will be right in front of you.", + "B": "Turn left and take two steps.", + "C": "Go straight three steps to it.", + "D": "Turn right, take two steps, then turn left and take one step; the refrigerator will be on your left.", + "E": "Turn right first, walk forward two steps, then turn right again; the refrigerator will be behind you." + }, + "answer_label": "A", + "answer_text": "You should turn right, take one step, then turn left and take two steps; the refrigerator will be right in front of you.", + "q_id": "egolife_A1_D4_004-mcq_fixed_4", + "correct_answer": "You should turn right, take one step, then turn left and take two steps; the refrigerator will be right in front of you.", + "original_options": { + "A": "You should turn right, take one step, then turn left and take two steps; the refrigerator will be right in front of you.", + "B": "Turn left and take two steps.", + "C": "Go straight three steps to it.", + "D": "Turn right, take two steps, then turn left and take one step; the refrigerator will be on your left.", + "E": "Turn right first, walk forward two steps, then turn right again; the refrigerator will be behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004-mcq_fixed_5", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I walk if I want to get something from the refrigerator right now?", + "answer": "You should turn around; the refrigerator is right behind the door.", + "question_chinese": "如果我此时要从冰箱里拿东西应该怎么走？", + "answer_chinese": "你应该后转，冰箱就在门后面。", + "evidence": [ + { + "description": "At this moment, you can see the refrigerator behind the door.", + "timestamps": { + "start": "00:10:13", + "end": "00:10:14" + } + } + ], + "question_timestamps": "00:10:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right; it’s beside the door.", + "B": "Walk straight; it’s in front of you.", + "C": "You should turn around; the refrigerator is right behind the door.", + "D": "Take a few steps forward past the counter, then turn left; the refrigerator will be along that wall.", + "E": "Turn around and move a step toward the counter; the refrigerator is behind the kitchen counter, not the door." + }, + "answer_label": "C", + "answer_text": "You should turn around; the refrigerator is right behind the door.", + "q_id": "egolife_A1_D4_004-mcq_fixed_5", + "correct_answer": "You should turn around; the refrigerator is right behind the door.", + "original_options": { + "A": "Turn right; it’s beside the door.", + "B": "Walk straight; it’s in front of you.", + "C": "You should turn around; the refrigerator is right behind the door.", + "D": "Take a few steps forward past the counter, then turn left; the refrigerator will be along that wall.", + "E": "Turn around and move a step toward the counter; the refrigerator is behind the kitchen counter, not the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004_1_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get from the second floor to where I am now?", + "answer": "You started from the second floor, turned right, went straight, turned right again, went out the door, went downstairs, and arrived at your current position.", + "question_chinese": "我从二楼出发怎么到达现在的位置？", + "answer_chinese": "你从二楼出发右转直走到了现在的位置。", + "evidence": [ + { + "description": "Now turn right and go straight ahead.", + "timestamps": { + "start": "00:02:38", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:02:41", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left, then went downstairs.", + "B": "You started from the second floor, turned right, went straight, turned right again, went out the door, went downstairs, and arrived at your current position.", + "C": "You went straight out before turning.", + "D": "You started on the second floor, turned left, went straight, turned right, went outside, went downstairs, and stopped there.", + "E": "You began upstairs, turned right, went straight, turned left, exited through the door, went downstairs, then continued straight to here." + }, + "answer_label": "B", + "answer_text": "You started from the second floor, turned right, went straight, turned right again, went out the door, went downstairs, and arrived at your current position.", + "q_id": "egolife_A1_D4_004_1_T1", + "correct_answer": "You started from the second floor, turned right, went straight, turned right again, went out the door, went downstairs, and arrived at your current position.", + "original_options": { + "A": "You turned left, then went downstairs.", + "B": "You started from the second floor, turned right, went straight, turned right again, went out the door, went downstairs, and arrived at your current position.", + "C": "You went straight out before turning.", + "D": "You started on the second floor, turned left, went straight, turned right, went outside, went downstairs, and stopped there.", + "E": "You began upstairs, turned right, went straight, turned left, exited through the door, went downstairs, then continued straight to here." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004_7_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get from the second floor to where I am now?", + "answer": "You started from the second floor, turned right and went straight, then turned right again, went out, went downstairs, then turned left, went in through the door, and reached the first-floor living room.", + "question_chinese": "我从二楼出发怎么到达现在的位置？", + "answer_chinese": "你从二楼出发右转直走再右转出门下楼然后左转进门到一楼客厅。", + "evidence": [ + { + "description": "Now turn right and go straight ahead.", + "timestamps": { + "start": "00:02:38", + "end": "00:02:40" + } + }, + { + "description": "Now turn right and go downstairs.", + "timestamps": { + "start": "00:02:41", + "end": "00:02:52" + } + }, + { + "description": "Now turn left, go in, and enter the living room on the first floor.", + "timestamps": { + "start": "00:02:53", + "end": "00:02:58" + } + } + ], + "question_timestamps": "00:03:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right after going downstairs.", + "B": "You turned left first upstairs.", + "C": "You started from the second floor, turned right and went straight, then turned right again, went out, went downstairs, then turned left, went in through the door, and reached the first-floor living room.", + "D": "You started on the second floor, turned right, went straight, turned left, went out, went downstairs, then entered the living room.", + "E": "You began upstairs, turned left, went straight, turned right, went out, went downstairs, then turned right and entered." + }, + "answer_label": "C", + "answer_text": "You started from the second floor, turned right and went straight, then turned right again, went out, went downstairs, then turned left, went in through the door, and reached the first-floor living room.", + "q_id": "egolife_A1_D4_004_7_T1", + "correct_answer": "You started from the second floor, turned right and went straight, then turned right again, went out, went downstairs, then turned left, went in through the door, and reached the first-floor living room.", + "original_options": { + "A": "You turned right after going downstairs.", + "B": "You turned left first upstairs.", + "C": "You started from the second floor, turned right and went straight, then turned right again, went out, went downstairs, then turned left, went in through the door, and reached the first-floor living room.", + "D": "You started on the second floor, turned right, went straight, turned left, went out, went downstairs, then entered the living room.", + "E": "You began upstairs, turned left, went straight, turned right, went out, went downstairs, then turned right and entered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 21, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004-mcq_fixed_6", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cameras did I just scan on the first floor?", + "answer": "Four.", + "question_chinese": "我刚刚在一楼拿手机扫了多少个摄像头？", + "answer_chinese": "你刚刚拿手机扫了一个摄像头。", + "evidence": [ + { + "description": "You are scanning the first camera right now.", + "timestamps": { + "start": "00:03:31", + "end": "00:03:32" + } + } + ], + "question_timestamps": "00:03:34", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "You scanned three cameras on the first floor, finishing after the third one.", + "E": "You scanned five cameras on the first floor, moving on only after the fifth scan." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "egolife_A1_D4_004-mcq_fixed_6", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "You scanned three cameras on the first floor, finishing after the third one.", + "E": "You scanned five cameras on the first floor, moving on only after the fifth scan." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004-mcq_fixed_7", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many cameras did I just scan on the first floor?", + "answer": "Six.", + "question_chinese": "我刚刚在一楼拿手机扫了多少个摄像头？", + "answer_chinese": "你刚刚拿手机扫了两个摄像头。", + "evidence": [ + { + "description": "You are scanning the first camera right now.", + "timestamps": { + "start": "00:03:31", + "end": "00:03:32" + } + }, + { + "description": "You are scanning the second camera right now.", + "timestamps": { + "start": "00:03:36", + "end": "00:03:37" + } + } + ], + "question_timestamps": "00:06:21", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "Six.", + "D": "You scanned five cameras on the first floor before stopping and moving on.", + "E": "You scanned seven cameras on the first floor, counting each scan as you went." + }, + "answer_label": "C", + "answer_text": "Six.", + "q_id": "egolife_A1_D4_004-mcq_fixed_7", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "Six.", + "D": "You scanned five cameras on the first floor before stopping and moving on.", + "E": "You scanned seven cameras on the first floor, counting each scan as you went." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004-mcq_fixed_2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path has the broom just taken?", + "answer": "The broom was originally placed on the floor under the whiteboard in the living room on the first floor. Afterwards, you picked it up and handed it to a girl. She then walked straight to the kitchen door and used it.", + "question_chinese": "扫帚刚刚经历了怎样的运动路线？", + "answer_chinese": "扫帚原本放在一楼客厅大白板下面的地上，之后你把它拿起来递给了一个女生，她拿着它直走到厨房门口使用。", + "evidence": [ + { + "description": "Right now, you pick up the broom.", + "timestamps": { + "start": "00:05:22", + "end": "00:05:23" + } + }, + { + "description": "Right now, you pick it up and hand it to a girl, who then takes it, walks straight to the kitchen door, and uses it.", + "timestamps": { + "start": "00:05:25", + "end": "00:05:30" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You carried it upstairs.", + "B": "She used it on the porch.", + "C": "The broom was originally placed on the floor under the whiteboard in the living room on the first floor. Afterwards, you picked it up and handed it to a girl. She then walked straight to the kitchen door and used it.", + "D": "The broom started by the hallway coat rack, then you gave it to the girl to use at the garden entrance.", + "E": "The broom began under a living-room table, then you handed it to the girl, who swept near the porch door." + }, + "answer_label": "C", + "answer_text": "The broom was originally placed on the floor under the whiteboard in the living room on the first floor. Afterwards, you picked it up and handed it to a girl. She then walked straight to the kitchen door and used it.", + "q_id": "egolife_A1_D4_004-mcq_fixed_2", + "correct_answer": "The broom was originally placed on the floor under the whiteboard in the living room on the first floor. Afterwards, you picked it up and handed it to a girl. She then walked straight to the kitchen door and used it.", + "original_options": { + "A": "You carried it upstairs.", + "B": "She used it on the porch.", + "C": "The broom was originally placed on the floor under the whiteboard in the living room on the first floor. Afterwards, you picked it up and handed it to a girl. She then walked straight to the kitchen door and used it.", + "D": "The broom started by the hallway coat rack, then you gave it to the girl to use at the garden entrance.", + "E": "The broom began under a living-room table, then you handed it to the girl, who swept near the porch door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 21, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_004-mcq_fixed_3", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path has the broom just taken?", + "answer": "The broom was originally placed on the floor under the whiteboard in the living room on the first floor. Afterwards, you picked it up and handed it to a girl. She then walked straight to the kitchen door and used it. After use, it was placed by the left side of the kitchen door.", + "question_chinese": "扫帚刚刚经历了怎样的运动路线？", + "answer_chinese": "扫帚原本放在一楼客厅大白板下面的地上，之后你把它拿起来递给了一个女生，她拿着它直走到厨房门口使用，在使用之后被放到了厨房左边门口。", + "evidence": [ + { + "description": "Right now, you pick up the broom.", + "timestamps": { + "start": "00:05:22", + "end": "00:05:23" + } + }, + { + "description": "Right now, you pick it up and hand it to a girl, who then takes it, walks straight to the kitchen door, and uses it.", + "timestamps": { + "start": "00:05:25", + "end": "00:05:30" + } + }, + { + "description": "At this moment, you can see that the broom, having been used, has been placed by the left side of the kitchen door.", + "timestamps": { + "start": "00:09:33", + "end": "00:09:34" + } + } + ], + "question_timestamps": "00:10:08", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The broom was originally placed on the floor under the whiteboard in the living room on the first floor. Afterwards, you picked it up and handed it to a girl. She then walked straight to the kitchen door and used it. After use, it was placed by the left side of the kitchen door.", + "B": "It ended by the right side.", + "C": "She returned it to the living room.", + "D": "It began under a dining-room table, then she used it at the kitchen door and left it on the right side.", + "E": "It started by the hallway shoe rack; you carried it upstairs briefly, then it was left near the balcony door." + }, + "answer_label": "A", + "answer_text": "The broom was originally placed on the floor under the whiteboard in the living room on the first floor. Afterwards, you picked it up and handed it to a girl. She then walked straight to the kitchen door and used it. After use, it was placed by the left side of the kitchen door.", + "q_id": "egolife_A1_D4_004-mcq_fixed_3", + "correct_answer": "The broom was originally placed on the floor under the whiteboard in the living room on the first floor. Afterwards, you picked it up and handed it to a girl. She then walked straight to the kitchen door and used it. After use, it was placed by the left side of the kitchen door.", + "original_options": { + "A": "The broom was originally placed on the floor under the whiteboard in the living room on the first floor. Afterwards, you picked it up and handed it to a girl. She then walked straight to the kitchen door and used it. After use, it was placed by the left side of the kitchen door.", + "B": "It ended by the right side.", + "C": "She returned it to the living room.", + "D": "It began under a dining-room table, then she used it at the kitchen door and left it on the right side.", + "E": "It started by the hallway shoe rack; you carried it upstairs briefly, then it was left near the balcony door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 20 + ] + } +] diff --git a/QAs/egolife_A1_D4_005-mcq.json b/QAs/egolife_A1_D4_005-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1d8895ed211d7ac0a994a387feea55f52311df3b --- /dev/null +++ b/QAs/egolife_A1_D4_005-mcq.json @@ -0,0 +1,798 @@ +[ + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005_1_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the microwave relative to me?", + "answer": "The microwave is in front of you.", + "question_chinese": "微波炉相对于我在什么位置？", + "answer_chinese": "微波炉在你的正前方。", + "evidence": [ + { + "description": "Right in front of you.", + "timestamps": { + "start": "00:0:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The microwave is behind you.", + "B": "The microwave is to your left.", + "C": "The microwave is in front of you.", + "D": "The microwave is to your right as you face forward in the kitchen area nearby.", + "E": "The microwave is above you, mounted higher than your head at the cabinet level." + }, + "answer_label": "C", + "answer_text": "The microwave is in front of you.", + "q_id": "egolife_A1_D4_005_1_T2", + "correct_answer": "The microwave is in front of you.", + "original_options": { + "A": "The microwave is behind you.", + "B": "The microwave is to your left.", + "C": "The microwave is in front of you.", + "D": "The microwave is to your right as you face forward in the kitchen area nearby.", + "E": "The microwave is above you, mounted higher than your head at the cabinet level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005_1_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the microwave relative to me?", + "answer": "The microwave is behind you.", + "question_chinese": "微波炉相对于我在什么位置？", + "answer_chinese": "微波炉在你的后面。", + "evidence": [ + { + "description": "Behind you.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The microwave is in front of you.", + "B": "The microwave is to your right.", + "C": "The microwave is to your left as you are standing there, just off to the side near you.", + "D": "The microwave is behind you.", + "E": "The microwave is above you, mounted higher than where you are looking while you stand there." + }, + "answer_label": "D", + "answer_text": "The microwave is behind you.", + "q_id": "egolife_A1_D4_005_1_T1", + "correct_answer": "The microwave is behind you.", + "original_options": { + "A": "The microwave is in front of you.", + "B": "The microwave is to your right.", + "C": "The microwave is to your left as you are standing there, just off to the side near you.", + "D": "The microwave is behind you.", + "E": "The microwave is above you, mounted higher than where you are looking while you stand there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005-mcq_fixed_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the bread I just ate?", + "answer": "You didn't just eat bread, so you don't know where the bread is.", + "question_chinese": "我刚刚吃的面包从哪里拿的？", + "answer_chinese": "你刚刚没有吃面包，所以不知道面包在哪里。", + "evidence": [ + { + "description": "So far, there is no evidence that you ate bread.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the grocery store.", + "B": "From the bakery.", + "C": "You grabbed the bread from a plate on the table just before eating it.", + "D": "You took the bread from the outer end of the table opposite the kitchen and ate it right away.", + "E": "You didn't just eat bread, so you don't know where the bread is." + }, + "answer_label": "E", + "answer_text": "You didn't just eat bread, so you don't know where the bread is.", + "q_id": "egolife_A1_D4_005-mcq_fixed_1", + "correct_answer": "You didn't just eat bread, so you don't know where the bread is.", + "original_options": { + "A": "From the grocery store.", + "B": "From the bakery.", + "C": "You grabbed the bread from a plate on the table just before eating it.", + "D": "You took the bread from the outer end of the table opposite the kitchen and ate it right away.", + "E": "You didn't just eat bread, so you don't know where the bread is." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005-mcq_fixed_0", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the bread I just ate?", + "answer": "The bread you just ate was taken from the plate on the outer end of the table opposite the kitchen on the first floor.", + "question_chinese": "我刚刚吃的面包从哪里拿的？", + "answer_chinese": "你刚刚吃的面包是从一楼厨房对面餐桌靠外端的盘子里拿的。", + "evidence": [ + { + "description": "You pick up the bread from the plate right now.", + "timestamps": { + "start": "00:02:07", + "end": "00:02:08" + } + } + ], + "question_timestamps": "00:09:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the inner end plate.", + "B": "From a basket on that table.", + "C": "You took the bread from the plate on the inner end of the table opposite the kitchen on the first floor.", + "D": "The bread you just ate was taken from the plate on the outer end of the table opposite the kitchen on the first floor.", + "E": "You reached for the bread in the middle of the table, between the salt shaker and the water pitcher." + }, + "answer_label": "D", + "answer_text": "The bread you just ate was taken from the plate on the outer end of the table opposite the kitchen on the first floor.", + "q_id": "egolife_A1_D4_005-mcq_fixed_0", + "correct_answer": "The bread you just ate was taken from the plate on the outer end of the table opposite the kitchen on the first floor.", + "original_options": { + "A": "From the inner end plate.", + "B": "From a basket on that table.", + "C": "You took the bread from the plate on the inner end of the table opposite the kitchen on the first floor.", + "D": "The bread you just ate was taken from the plate on the outer end of the table opposite the kitchen on the first floor.", + "E": "You reached for the bread in the middle of the table, between the salt shaker and the water pitcher." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 21, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005-mcq_fixed_3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the refrigerator on the first floor just now?", + "answer": "Three.", + "question_chinese": "我刚刚打开了几次一楼的冰箱？", + "answer_chinese": "你刚刚打开了一次一楼的冰箱。", + "evidence": [ + { + "description": "At this moment, you are opening the refrigerator on the first floor for the first time.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:01:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Two times.", + "C": "Four times.", + "D": "You opened the first-floor refrigerator two times, checking inside and then closing it once.", + "E": "You opened the first-floor refrigerator four times, returning to it repeatedly before finally leaving it shut." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "egolife_A1_D4_005-mcq_fixed_3", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two times.", + "C": "Four times.", + "D": "You opened the first-floor refrigerator two times, checking inside and then closing it once.", + "E": "You opened the first-floor refrigerator four times, returning to it repeatedly before finally leaving it shut." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005-mcq_fixed_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I open the refrigerator on the first floor just now?", + "answer": "Five.", + "question_chinese": "我刚刚打开了几次一楼的冰箱？", + "answer_chinese": "你刚刚打开了两次一楼的冰箱。", + "evidence": [ + { + "description": "At this moment, you are opening the refrigerator on the first floor for the first time.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:19" + } + }, + { + "description": "At this moment, you are opening the refrigerator on the first floor for the second time.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:10:42", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four times.", + "B": "Six times.", + "C": "Five.", + "D": "You opened the first-floor refrigerator four times, going back once fewer than you thought.", + "E": "You opened the first-floor refrigerator six times, making an extra stop to look inside again." + }, + "answer_label": "C", + "answer_text": "Five.", + "q_id": "egolife_A1_D4_005-mcq_fixed_2", + "correct_answer": "Five.", + "original_options": { + "A": "Four times.", + "B": "Six times.", + "C": "Five.", + "D": "You opened the first-floor refrigerator four times, going back once fewer than you thought.", + "E": "You opened the first-floor refrigerator six times, making an extra stop to look inside again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005-mcq_fixed_5", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I get the last power bank I just took?", + "answer": "It is the last power bank you took from the bottom shelf of the rack in the living room on the first floor.", + "question_chinese": "我刚刚最后一个充电宝是从哪里拿的？", + "answer_chinese": "你从一楼客厅架子最底层拿的最后一个充电宝。", + "evidence": [ + { + "description": "It is the last power bank you took from the bottom shelf of the rack in the living room on the first floor.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:19" + } + } + ], + "question_timestamps": "00:04:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the middle shelf of the rack.", + "B": "It is the last power bank you took from the bottom shelf of the rack in the living room on the first floor.", + "C": "From the top shelf of the rack.", + "D": "You took it from the middle shelf of the rack in the first-floor living room.", + "E": "You took it from the bottom shelf of the rack in the second-floor living room." + }, + "answer_label": "B", + "answer_text": "It is the last power bank you took from the bottom shelf of the rack in the living room on the first floor.", + "q_id": "egolife_A1_D4_005-mcq_fixed_5", + "correct_answer": "It is the last power bank you took from the bottom shelf of the rack in the living room on the first floor.", + "original_options": { + "A": "From the middle shelf of the rack.", + "B": "It is the last power bank you took from the bottom shelf of the rack in the living room on the first floor.", + "C": "From the top shelf of the rack.", + "D": "You took it from the middle shelf of the rack in the first-floor living room.", + "E": "You took it from the bottom shelf of the rack in the second-floor living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005-mcq_fixed_4", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where did I get the last power bank I just took?", + "answer": "You got the last power bank from the bed in the bedroom on the first floor where there are multiple computers.", + "question_chinese": "我刚刚最后一个充电宝是从哪里拿的？", + "answer_chinese": "你从一楼放有多台电脑的卧室的床上拿的最后一个充电宝。", + "evidence": [ + { + "description": "At this moment, you are picking up the last power bank from the bed.", + "timestamps": { + "start": "00:06:22", + "end": "00:06:24" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the chair in the living room.", + "B": "From the coffee table by the kitchen.", + "C": "You got the last power bank from the chair in the living room on the first floor where there are multiple lamps.", + "D": "You got the last power bank from the bed in the bedroom on the first floor where there are multiple computers.", + "E": "You took the last power bank from the shelf near the window in the upstairs guest bedroom." + }, + "answer_label": "D", + "answer_text": "You got the last power bank from the bed in the bedroom on the first floor where there are multiple computers.", + "q_id": "egolife_A1_D4_005-mcq_fixed_4", + "correct_answer": "You got the last power bank from the bed in the bedroom on the first floor where there are multiple computers.", + "original_options": { + "A": "From the chair in the living room.", + "B": "From the coffee table by the kitchen.", + "C": "You got the last power bank from the chair in the living room on the first floor where there are multiple lamps.", + "D": "You got the last power bank from the bed in the bedroom on the first floor where there are multiple computers.", + "E": "You took the last power bank from the shelf near the window in the upstairs guest bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 22, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005-mcq_fixed_7", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I steady the light stand in the yard?", + "answer": "Five.", + "question_chinese": "我在院子里扶了几次打光灯？", + "answer_chinese": "你在院子里扶了一次打光灯。", + "evidence": [ + { + "description": "At this moment, you are steadying the light stand for the first time.", + "timestamps": { + "start": "00:07:58", + "end": "00:08:01" + } + } + ], + "question_timestamps": "00:09:59", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Four times.", + "C": "Six times.", + "D": "You steadied the light stand in the yard four times, stopping one adjustment earlier than five.", + "E": "You steadied the light stand in the yard six times, making one more stabilizing touch than expected." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D4_005-mcq_fixed_7", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "Four times.", + "C": "Six times.", + "D": "You steadied the light stand in the yard four times, stopping one adjustment earlier than five.", + "E": "You steadied the light stand in the yard six times, making one more stabilizing touch than expected." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005-mcq_fixed_6", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I steady the light stand in the yard?", + "answer": "Four.", + "question_chinese": "我在院子里扶了几次打光灯？", + "answer_chinese": "你在院子里扶了两次打光灯。", + "evidence": [ + { + "description": "At this moment, you are steadying the light stand for the first time.", + "timestamps": { + "start": "00:07:58", + "end": "00:08:01" + } + }, + { + "description": "At this moment, you are steadying the light stand for the second time.", + "timestamps": { + "start": "00:10:20", + "end": "00:10:23" + } + } + ], + "question_timestamps": "00:11:31", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three times.", + "B": "Five times.", + "C": "You steadied the light stand in the yard three times, correcting it only after the biggest wobbles.", + "D": "Four.", + "E": "You steadied the light stand in the yard five times, repeatedly fixing it before it stayed in place." + }, + "answer_label": "D", + "answer_text": "Four.", + "q_id": "egolife_A1_D4_005-mcq_fixed_6", + "correct_answer": "Four.", + "original_options": { + "A": "Three times.", + "B": "Five times.", + "C": "You steadied the light stand in the yard three times, correcting it only after the biggest wobbles.", + "D": "Four.", + "E": "You steadied the light stand in the yard five times, repeatedly fixing it before it stayed in place." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005_10", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get from the dining table on the first floor to the table in the yard?", + "answer": "You should first turn right, go out the door, then immediately turn left, go out another door, and walk straight toward the left front.", + "question_chinese": "我现在如果要从一楼餐桌到院子里的桌子旁应该怎么走？", + "answer_chinese": "你应该先右转出门后立即左转出门然后向左前方直走。", + "evidence": [ + { + "description": "At this moment, you are turning right and going out the door from the dining table.", + "timestamps": { + "start": "00:07:39", + "end": "00:07:41" + } + }, + { + "description": "Turn left and go out the door immediately.", + "timestamps": { + "start": "00:07:42", + "end": "00:07:45" + } + }, + { + "description": "At this moment, you are walking straight toward the left front, and the table is right in front of you.", + "timestamps": { + "start": "00:07:46", + "end": "00:07:50" + } + } + ], + "question_timestamps": "00:09:24", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, then right, then go straight.", + "B": "Turn right, then right, then go straight.", + "C": "First turn left, go out the door, then turn left and walk straight to the back yard.", + "D": "First turn right, go out the door, then turn right and walk straight toward the left front.", + "E": "You should first turn right, go out the door, then immediately turn left, go out another door, and walk straight toward the left front." + }, + "answer_label": "E", + "answer_text": "You should first turn right, go out the door, then immediately turn left, go out another door, and walk straight toward the left front.", + "q_id": "egolife_A1_D4_005_10", + "correct_answer": "You should first turn right, go out the door, then immediately turn left, go out another door, and walk straight toward the left front.", + "original_options": { + "A": "Turn left, then right, then go straight.", + "B": "Turn right, then right, then go straight.", + "C": "First turn left, go out the door, then turn left and walk straight to the back yard.", + "D": "First turn right, go out the door, then turn right and walk straight toward the left front.", + "E": "You should first turn right, go out the door, then immediately turn left, go out another door, and walk straight toward the left front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005_11", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get from the dining table on the first floor to the table in the yard?", + "answer": "Ou should first turn right, go out the door, then immediately turn left, go out another door, and walk straight toward the left.", + "question_chinese": "我现在如果要从一楼餐桌到院子里的桌子旁应该怎么走？", + "answer_chinese": "你应该先右转出门后立即左转出门然后向左前方直走。", + "evidence": [ + { + "description": "At this moment, you are turning right and going out the door from the dining table.", + "timestamps": { + "start": "00:07:39", + "end": "00:07:41" + } + }, + { + "description": "Turn left and go out the door immediately.", + "timestamps": { + "start": "00:07:42", + "end": "00:07:45" + } + }, + { + "description": "At this moment, you are walking straight toward the left front, and the table is right in front of you.", + "timestamps": { + "start": "00:07:46", + "end": "00:07:50" + } + } + ], + "question_timestamps": "00:11:25", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, go outside, then right.", + "B": "Go straight out both doors, then left.", + "C": "You should turn right, go out the first door, turn right again, exit the second door, and walk straight ahead.", + "D": "You should turn left from the dining table, exit the door, immediately turn left, go through the next door, and head rightward.", + "E": "Ou should first turn right, go out the door, then immediately turn left, go out another door, and walk straight toward the left." + }, + "answer_label": "E", + "answer_text": "Ou should first turn right, go out the door, then immediately turn left, go out another door, and walk straight toward the left.", + "q_id": "egolife_A1_D4_005_11", + "correct_answer": "Ou should first turn right, go out the door, then immediately turn left, go out another door, and walk straight toward the left.", + "original_options": { + "A": "Turn left, go outside, then right.", + "B": "Go straight out both doors, then left.", + "C": "You should turn right, go out the first door, turn right again, exit the second door, and walk straight ahead.", + "D": "You should turn left from the dining table, exit the door, immediately turn left, go through the next door, and head rightward.", + "E": "Ou should first turn right, go out the door, then immediately turn left, go out another door, and walk straight toward the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 22 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005_2_T1", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is on my dining table?", + "answer": "On your dining table, there’s a plate of boiled eggs and a plate of pizza, and the boiled eggs and pizza are in front of you.", + "question_chinese": "我的餐桌上有什么东西？", + "answer_chinese": "你的餐桌上有一盘煮鸡蛋和一盘披萨,煮鸡蛋和披萨在你的正前方。", + "evidence": [ + { + "description": "There is a plate of boiled eggs and a plate of pizza on your dining table.", + "timestamps": { + "start": "00:02:05", + "end": "00:02:07" + } + } + ], + "question_timestamps": "00:02:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your dining table, there’s a plate of boiled eggs and a plate of pizza, and the boiled eggs and pizza are in front of you.", + "B": "A plate of boiled eggs only.", + "C": "A plate of pizza only.", + "D": "There’s a plate of boiled eggs and a plate of pizza, and both plates are behind you.", + "E": "You have a plate of boiled eggs and a plate of pizza on the table, but they’re set to your left." + }, + "answer_label": "A", + "answer_text": "On your dining table, there’s a plate of boiled eggs and a plate of pizza, and the boiled eggs and pizza are in front of you.", + "q_id": "egolife_A1_D4_005_2_T1", + "correct_answer": "On your dining table, there’s a plate of boiled eggs and a plate of pizza, and the boiled eggs and pizza are in front of you.", + "original_options": { + "A": "On your dining table, there’s a plate of boiled eggs and a plate of pizza, and the boiled eggs and pizza are in front of you.", + "B": "A plate of boiled eggs only.", + "C": "A plate of pizza only.", + "D": "There’s a plate of boiled eggs and a plate of pizza, and both plates are behind you.", + "E": "You have a plate of boiled eggs and a plate of pizza on the table, but they’re set to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 22 + ] + }, + { + "video_uid": "egolife_A1_D4_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_005_2_T2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is on my dining table?", + "answer": "On your dining table, there’s a plate of boiled eggs, a plate of pizza, and some flowers; the boiled eggs, the plate of pizza, and the flowers are all in front of you.", + "question_chinese": "我的餐桌上有什么东西？", + "answer_chinese": "你的餐桌子上有一盘煮鸡蛋和一盘披萨，还有一些花，煮鸡蛋和一盘披萨、一些话在你的正前方。", + "evidence": [ + { + "description": "There are a plate of boiled eggs and a plate of pizza on your dining table, and some flowers..", + "timestamps": { + "start": "00:02:12", + "end": "00:02:13" + } + } + ], + "question_timestamps": "00:02:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Boiled eggs and pizza only.", + "B": "Just flowers on the table.", + "C": "There’s a plate of boiled eggs, a plate of pizza, and some flowers, and everything is behind you.", + "D": "You see boiled eggs, pizza, and flowers on the table, but they’re all arranged to your right.", + "E": "On your dining table, there’s a plate of boiled eggs, a plate of pizza, and some flowers; the boiled eggs, the plate of pizza, and the flowers are all in front of you." + }, + "answer_label": "E", + "answer_text": "On your dining table, there’s a plate of boiled eggs, a plate of pizza, and some flowers; the boiled eggs, the plate of pizza, and the flowers are all in front of you.", + "q_id": "egolife_A1_D4_005_2_T2", + "correct_answer": "On your dining table, there’s a plate of boiled eggs, a plate of pizza, and some flowers; the boiled eggs, the plate of pizza, and the flowers are all in front of you.", + "original_options": { + "A": "Boiled eggs and pizza only.", + "B": "Just flowers on the table.", + "C": "There’s a plate of boiled eggs, a plate of pizza, and some flowers, and everything is behind you.", + "D": "You see boiled eggs, pizza, and flowers on the table, but they’re all arranged to your right.", + "E": "On your dining table, there’s a plate of boiled eggs, a plate of pizza, and some flowers; the boiled eggs, the plate of pizza, and the flowers are all in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D4_006-mcq.json b/QAs/egolife_A1_D4_006-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..672df79393e1ad8310e2d8fb3d02ca8f2e0d3b83 --- /dev/null +++ b/QAs/egolife_A1_D4_006-mcq.json @@ -0,0 +1,722 @@ +[ + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006-mcq_fixed_2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the mirror in the living room on the first floor?", + "answer": "The mirror is on the wall between the two bedrooms.", + "question_chinese": "一楼客厅里的镜子在哪里?", + "answer_chinese": "镜子在两个卧室之间的墙上。", + "evidence": [ + { + "description": "At this moment, you can see the mirror on the wall between the two bedrooms.", + "timestamps": { + "start": "00:03:57", + "end": "00:03:59" + } + }, + { + "description": "At this moment, you can also see the mirror on the wall between the two bedrooms.The girl is looking in the mirror.", + "timestamps": { + "start": "00:05:15", + "end": "00:05:19" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The mirror is above the fireplace.", + "B": "The mirror is on the wall between the two bedrooms.", + "C": "The mirror is next to the bookshelf.", + "D": "The mirror is mounted on the wall with a vintage clock, facing toward the seating area.", + "E": "The mirror hangs on the wall with a family portrait, reflecting the living room as you pass." + }, + "answer_label": "B", + "answer_text": "The mirror is on the wall between the two bedrooms.", + "q_id": "egolife_A1_D4_006-mcq_fixed_2", + "correct_answer": "The mirror is on the wall between the two bedrooms.", + "original_options": { + "A": "The mirror is above the fireplace.", + "B": "The mirror is on the wall between the two bedrooms.", + "C": "The mirror is next to the bookshelf.", + "D": "The mirror is mounted on the wall with a vintage clock, facing toward the seating area.", + "E": "The mirror hangs on the wall with a family portrait, reflecting the living room as you pass." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006-mcq_fixed_3", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the mirror in the living room on the first floor?", + "answer": "The mirror is on the wall between the two bedrooms on your left.", + "question_chinese": "一楼客厅里的镜子在哪里?", + "answer_chinese": "镜子在你左手边两个卧室之间的墙上。", + "evidence": [ + { + "description": "At this moment, you can see the mirror on the wall between the two bedrooms.", + "timestamps": { + "start": "00:03:57", + "end": "00:03:59" + } + }, + { + "description": "At this moment, you can also see the mirror on the wall between the two bedrooms.The girl is looking in the mirror.", + "timestamps": { + "start": "00:05:15", + "end": "00:05:19" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The mirror is on the wall between the two bedrooms on your left.", + "B": "Between two bedrooms on your right.", + "C": "Between two paintings on your left.", + "D": "The mirror stands on a narrow console table near the staircase, so you notice it while walking by.", + "E": "The mirror is mounted above the fireplace near the middle of the living room, just past the bookshelf on your right." + }, + "answer_label": "A", + "answer_text": "The mirror is on the wall between the two bedrooms on your left.", + "q_id": "egolife_A1_D4_006-mcq_fixed_3", + "correct_answer": "The mirror is on the wall between the two bedrooms on your left.", + "original_options": { + "A": "The mirror is on the wall between the two bedrooms on your left.", + "B": "Between two bedrooms on your right.", + "C": "Between two paintings on your left.", + "D": "The mirror stands on a narrow console table near the staircase, so you notice it while walking by.", + "E": "The mirror is mounted above the fireplace near the middle of the living room, just past the bookshelf on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 21 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006_4_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the bottle I picked up in the courtyard just now?", + "answer": "The bottle you just picked up is in your right hand.", + "question_chinese": "我刚刚在院子里捡的瓶子现在在哪里？", + "answer_chinese": "你刚刚捡到的瓶子在你的右手上。", + "evidence": [ + { + "description": "The bottle you just picked up is in your right hand.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In your left hand.", + "B": "On the ground nearby.", + "C": "The bottle you just picked up is in your right hand.", + "D": "You already tossed the bottle into a nearby trash can that’s positioned behind you in the courtyard.", + "E": "The bottle you picked up is now in the trash can near the yard entrance, slightly to your left." + }, + "answer_label": "C", + "answer_text": "The bottle you just picked up is in your right hand.", + "q_id": "egolife_A1_D4_006_4_T2", + "correct_answer": "The bottle you just picked up is in your right hand.", + "original_options": { + "A": "In your left hand.", + "B": "On the ground nearby.", + "C": "The bottle you just picked up is in your right hand.", + "D": "You already tossed the bottle into a nearby trash can that’s positioned behind you in the courtyard.", + "E": "The bottle you picked up is now in the trash can near the yard entrance, slightly to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006_4_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the bottle I picked up in the courtyard just now?", + "answer": "The bottle you just picked up is in the trash can, and the trash can is behind you.", + "question_chinese": "我刚刚在院子里捡的瓶子现在在哪里？", + "answer_chinese": "你刚刚捡到的瓶子在垃圾桶里，垃圾桶在你后面。", + "evidence": [ + { + "description": "You throw the bottle into the trash can.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the trash can in front of you.", + "B": "In the trash can to your left.", + "C": "The bottle you picked up is in the trash can, and the trash can is to your right as you turn around.", + "D": "The bottle you just picked up is in the trash can, and the trash can is behind you.", + "E": "You threw the bottle into the trash can, but the trash can is positioned ahead of you near the path." + }, + "answer_label": "D", + "answer_text": "The bottle you just picked up is in the trash can, and the trash can is behind you.", + "q_id": "egolife_A1_D4_006_4_T1", + "correct_answer": "The bottle you just picked up is in the trash can, and the trash can is behind you.", + "original_options": { + "A": "In the trash can in front of you.", + "B": "In the trash can to your left.", + "C": "The bottle you picked up is in the trash can, and the trash can is to your right as you turn around.", + "D": "The bottle you just picked up is in the trash can, and the trash can is behind you.", + "E": "You threw the bottle into the trash can, but the trash can is positioned ahead of you near the path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 22, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006-mcq_fixed_4", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where is the bottle I just picked up in the yard now?", + "answer": "You just threw the bottle you picked up into the trash can on the left side of the yard entrance.", + "question_chinese": "我刚刚在院子里捡的瓶子现在在哪里？", + "answer_chinese": "你刚刚捡到的瓶子扔到了院子门口左侧的垃圾桶里。", + "evidence": [ + { + "description": "At this moment, you are throwing the bottle into the trash can on the left side of the yard entrance.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:05:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just threw the bottle you picked up into the trash can on the left side of the yard entrance.", + "B": "You threw it into the right trash can.", + "C": "You left it near the yard entrance.", + "D": "You just threw the bottle you picked up into the trash can behind the yard entrance, not the one at the side.", + "E": "You carried the bottle into the shed and left it on the workbench near the right-hand gate, planning to use it later." + }, + "answer_label": "A", + "answer_text": "You just threw the bottle you picked up into the trash can on the left side of the yard entrance.", + "q_id": "egolife_A1_D4_006-mcq_fixed_4", + "correct_answer": "You just threw the bottle you picked up into the trash can on the left side of the yard entrance.", + "original_options": { + "A": "You just threw the bottle you picked up into the trash can on the left side of the yard entrance.", + "B": "You threw it into the right trash can.", + "C": "You left it near the yard entrance.", + "D": "You just threw the bottle you picked up into the trash can behind the yard entrance, not the one at the side.", + "E": "You carried the bottle into the shed and left it on the workbench near the right-hand gate, planning to use it later." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 22, + 22 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006-mcq_fixed_1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just sit on the animal-shaped stool beside the table?", + "answer": "No，you just did not sit on the animal-shaped stool beside the table.", + "question_chinese": "我刚刚坐过餐桌旁边的动物板凳吗？", + "answer_chinese": "没有，你刚刚没有坐过餐桌旁边的动物板凳。", + "evidence": [ + { + "description": "So far, there is no evidence that you have sat on the animal stool.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you just did sit on the animal-shaped stool beside the table.", + "B": "No，you just did not sit on the animal-shaped stool beside the table." + }, + "answer_label": "B", + "answer_text": "No，you just did not sit on the animal-shaped stool beside the table.", + "q_id": "egolife_A1_D4_006-mcq_fixed_1", + "correct_answer": "No，you just did not sit on the animal-shaped stool beside the table.", + "original_options": { + "A": "Yes, you just did sit on the animal-shaped stool beside the table.", + "B": "No，you just did not sit on the animal-shaped stool beside the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006-mcq_fixed_0", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Did I just sit on the animal-shaped stool beside the table?", + "answer": "Yes, you just sat on the animal-shaped stool beside the table.", + "question_chinese": "我刚刚坐过餐桌旁边的动物板凳吗？", + "answer_chinese": "是的，你刚刚坐过餐桌旁边的动物板凳。", + "evidence": [ + { + "description": "At this moment, you see the animal-shaped stool.", + "timestamps": { + "start": "00:07:41", + "end": "00:07:48" + } + }, + { + "description": "At this moment, you are sitting on the stool.", + "timestamps": { + "start": "00:07:49", + "end": "00:08:18" + } + }, + { + "description": "At this moment, you are getting up from the stool.", + "timestamps": { + "start": "00:08:19", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:09:19", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Yes, you just sat on the animal-shaped stool beside the table.", + "B": "No, you did not just sit on the animal-shaped stool beside the table." + }, + "answer_label": "A", + "answer_text": "Yes, you just sat on the animal-shaped stool beside the table.", + "q_id": "egolife_A1_D4_006-mcq_fixed_0", + "correct_answer": "Yes, you just sat on the animal-shaped stool beside the table.", + "original_options": { + "A": "Yes, you just sat on the animal-shaped stool beside the table.", + "B": "No, you did not just sit on the animal-shaped stool beside the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006-mcq_fixed_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just wash?", + "answer": "You just washed nothing.", + "question_chinese": "我刚刚洗了什么？", + "answer_chinese": "你刚刚什么也没洗。", + "evidence": [ + { + "description": "You just washed nothing.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:05:24", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just washed dishes.", + "B": "You just washed a cup.", + "C": "You just washed nothing.", + "D": "You just washed a couple of utensils at the sink before setting them aside to dry.", + "E": "You just washed the frying pan briefly, scrubbing it and rinsing it clean in the sink." + }, + "answer_label": "C", + "answer_text": "You just washed nothing.", + "q_id": "egolife_A1_D4_006-mcq_fixed_6", + "correct_answer": "You just washed nothing.", + "original_options": { + "A": "You just washed dishes.", + "B": "You just washed a cup.", + "C": "You just washed nothing.", + "D": "You just washed a couple of utensils at the sink before setting them aside to dry.", + "E": "You just washed the frying pan briefly, scrubbing it and rinsing it clean in the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006-mcq_fixed_7", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What did I just wash?", + "answer": "You just washed the egg beater.", + "question_chinese": "我刚刚洗了什么？", + "answer_chinese": "你刚刚洗了搅蛋器。", + "evidence": [ + { + "description": "At this moment, you are washing the whisk.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:36" + } + } + ], + "question_timestamps": "00:11:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You just washed the dough hook.", + "B": "You just washed the frying pan.", + "C": "You just washed the egg beater.", + "D": "You just washed the salad spinner, rinsing it thoroughly and shaking off the remaining water.", + "E": "You just washed the metal spatula, scrubbing off residue before rinsing it clean at the sink." + }, + "answer_label": "C", + "answer_text": "You just washed the egg beater.", + "q_id": "egolife_A1_D4_006-mcq_fixed_7", + "correct_answer": "You just washed the egg beater.", + "original_options": { + "A": "You just washed the dough hook.", + "B": "You just washed the frying pan.", + "C": "You just washed the egg beater.", + "D": "You just washed the salad spinner, rinsing it thoroughly and shaking off the remaining water.", + "E": "You just washed the metal spatula, scrubbing off residue before rinsing it clean at the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006-mcq_fixed_8", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the whisk take?", + "answer": "The whisk was taken from the boy sitting beside the dining table and brought to the sink to be washed.", + "question_chinese": "搅蛋器经历了怎样的运动轨迹？", + "answer_chinese": "搅蛋器被你从餐桌旁坐的男孩手中拿过来去水槽清洗了一下.", + "evidence": [ + { + "description": "At this moment, you are taking the whisk from the hands of the boy sitting beside the table.", + "timestamps": { + "start": "00:06:22", + "end": "00:06:26" + } + }, + { + "description": "At this moment, you start to go to the sink to wash the whisk.", + "timestamps": { + "start": "00:06:26", + "end": "00:06:34" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It went from the counter to the sink.", + "B": "It went from the table to a drawer.", + "C": "The whisk was picked up from the counter near the stove and placed on the shelf above the sink.", + "D": "The whisk was taken from the boy sitting beside the dining table and brought to the sink to be washed.", + "E": "The whisk was taken from the boy beside the dining table and carried to the dish rack instead of the sink." + }, + "answer_label": "D", + "answer_text": "The whisk was taken from the boy sitting beside the dining table and brought to the sink to be washed.", + "q_id": "egolife_A1_D4_006-mcq_fixed_8", + "correct_answer": "The whisk was taken from the boy sitting beside the dining table and brought to the sink to be washed.", + "original_options": { + "A": "It went from the counter to the sink.", + "B": "It went from the table to a drawer.", + "C": "The whisk was picked up from the counter near the stove and placed on the shelf above the sink.", + "D": "The whisk was taken from the boy sitting beside the dining table and brought to the sink to be washed.", + "E": "The whisk was taken from the boy beside the dining table and carried to the dish rack instead of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006-mcq_fixed_9", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the whisk take?", + "answer": "The whisk was taken from the boy sitting beside the dining table and brought to the sink to be washed,and then the whisk you washed has been placed in the basin inside the sink.", + "question_chinese": "搅蛋器经历了怎样的运动轨迹？", + "answer_chinese": "搅蛋器被你从餐桌旁坐的男孩手中拿过来去水槽清洗了一下，然后洗好的搅蛋器被你放到了水槽里的盆里。", + "evidence": [ + { + "description": "At this moment, you are taking the whisk from the hands of the boy sitting beside the table.", + "timestamps": { + "start": "00:06:22", + "end": "00:06:26" + } + }, + { + "description": "At this moment, you start to go to the sink to wash the whisk.", + "timestamps": { + "start": "00:06:26", + "end": "00:06:34" + } + }, + { + "description": "Right now, the whisk you washed has been placed in the basin inside the sink.", + "timestamps": { + "start": "00:06:41", + "end": "00:06:48" + } + } + ], + "question_timestamps": "00:09:27", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "It ended up under the drying rack.", + "B": "It ended up in the refrigerator.", + "C": "The whisk was taken from the boy sitting beside the dining table and brought to the sink to be washed,and then the whisk you washed has been placed in the basin inside the sink.", + "D": "The whisk was taken from the boy beside the dining table to the sink, then placed on the counter to dry.", + "E": "The whisk was moved from the countertop to the dish rack after a quick rinse, rather than left in the sink basin." + }, + "answer_label": "C", + "answer_text": "The whisk was taken from the boy sitting beside the dining table and brought to the sink to be washed,and then the whisk you washed has been placed in the basin inside the sink.", + "q_id": "egolife_A1_D4_006-mcq_fixed_9", + "correct_answer": "The whisk was taken from the boy sitting beside the dining table and brought to the sink to be washed,and then the whisk you washed has been placed in the basin inside the sink.", + "original_options": { + "A": "It ended up under the drying rack.", + "B": "It ended up in the refrigerator.", + "C": "The whisk was taken from the boy sitting beside the dining table and brought to the sink to be washed,and then the whisk you washed has been placed in the basin inside the sink.", + "D": "The whisk was taken from the boy beside the dining table to the sink, then placed on the counter to dry.", + "E": "The whisk was moved from the countertop to the dish rack after a quick rinse, rather than left in the sink basin." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 21, + 22 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006_3_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Relative to me, where is the trash can in the courtyard?", + "answer": "The trash can is to your front-left.", + "question_chinese": "相对于我的位置，院子里的垃圾桶在哪里？", + "answer_chinese": "垃圾桶在你的左前方。", + "evidence": [ + { + "description": "The trash can is in front left of you.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can is to your front-right.", + "B": "The trash can is to your front-left.", + "C": "The trash can is to your back-left.", + "D": "The trash can is behind you, slightly to your left as you stand in the courtyard.", + "E": "The trash can is in front of you on your right side, a short distance away in the courtyard." + }, + "answer_label": "B", + "answer_text": "The trash can is to your front-left.", + "q_id": "egolife_A1_D4_006_3_T2", + "correct_answer": "The trash can is to your front-left.", + "original_options": { + "A": "The trash can is to your front-right.", + "B": "The trash can is to your front-left.", + "C": "The trash can is to your back-left.", + "D": "The trash can is behind you, slightly to your left as you stand in the courtyard.", + "E": "The trash can is in front of you on your right side, a short distance away in the courtyard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_006_3_T1", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Relative to me, where is the trash can in the courtyard?", + "answer": "The trash can is behind you.", + "question_chinese": "相对于我的位置，院子里的垃圾桶在哪里？", + "answer_chinese": "垃圾桶在你的身后。", + "evidence": [ + { + "description": "The trash can is behind you.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can is in front of you.", + "B": "The trash can is to your left.", + "C": "The trash can is to your front-right as you face the courtyard, not behind you.", + "D": "The trash can is behind you.", + "E": "The trash can is off to your right side, roughly parallel with your position in the courtyard." + }, + "answer_label": "D", + "answer_text": "The trash can is behind you.", + "q_id": "egolife_A1_D4_006_3_T1", + "correct_answer": "The trash can is behind you.", + "original_options": { + "A": "The trash can is in front of you.", + "B": "The trash can is to your left.", + "C": "The trash can is to your front-right as you face the courtyard, not behind you.", + "D": "The trash can is behind you.", + "E": "The trash can is off to your right side, roughly parallel with your position in the courtyard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D4_007-mcq.json b/QAs/egolife_A1_D4_007-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b79b349b8c537d918389c3135e8d6faf49fce960 --- /dev/null +++ b/QAs/egolife_A1_D4_007-mcq.json @@ -0,0 +1,575 @@ +[ + { + "video_uid": "egolife_A1_D4_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_007-mcq_fixed_8", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What's on the table on the second floor?", + "answer": "There are cups, mineral water, and paper towels on the table on the second floor.", + "question_chinese": "二楼的桌子上放的有什么东西？", + "answer_chinese": "二楼的桌子上放了杯子，矿泉水和纸巾。", + "evidence": [ + { + "description": "At this moment, you can see that there are cups, mineral water, and paper towels on the table on the second floor.", + "timestamps": { + "start": "00:05:21", + "end": "00:05:22" + } + }, + { + "description": "At this moment, you can see that there are cups, mineral water, and paper towels on the table on the second floor.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:42" + } + }, + { + "description": "At this moment, you can see that there are cups, mineral water, and paper towels on the table on the second floor.", + "timestamps": { + "start": "00:04:35", + "end": "00:04:38" + } + } + ], + "question_timestamps": "00:08:52", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "There are plates and napkins on the table.", + "B": "There are bottles of soda and mugs.", + "C": "There are cups, mineral water, and paper towels on the table on the second floor.", + "D": "There are candles, lemonade, and paper towels arranged together on the table on the second floor.", + "E": "On the second floor table, you see flowers and sparkling water placed beside a small stack of books." + }, + "answer_label": "C", + "answer_text": "There are cups, mineral water, and paper towels on the table on the second floor.", + "q_id": "egolife_A1_D4_007-mcq_fixed_8", + "correct_answer": "There are cups, mineral water, and paper towels on the table on the second floor.", + "original_options": { + "A": "There are plates and napkins on the table.", + "B": "There are bottles of soda and mugs.", + "C": "There are cups, mineral water, and paper towels on the table on the second floor.", + "D": "There are candles, lemonade, and paper towels arranged together on the table on the second floor.", + "E": "On the second floor table, you see flowers and sparkling water placed beside a small stack of books." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_007-mcq_fixed_9", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What's on the table on the second floor?", + "answer": "There are cups, mineral water, and paper towels on the table.", + "question_chinese": "二楼的桌子上放的有什么东西？", + "answer_chinese": "桌子上放了杯子，矿泉水和纸巾。", + "evidence": [ + { + "description": "At this moment, you can see that there are cups, mineral water, and paper towels on the table on the second floor.", + "timestamps": { + "start": "00:04:35", + "end": "00:04:38" + } + }, + { + "description": "At this moment, you can see that there are cups, mineral water, and paper towels on the table on the second floor.", + "timestamps": { + "start": "00:05:21", + "end": "00:05:23" + } + }, + { + "description": "At this moment, you can see that there are cups, mineral water, and paper towels on the table on the second floor.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:42" + } + } + ], + "question_timestamps": "00:10:40", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Plates, sparkling water, and napkins.", + "B": "Mugs, tap water, and a towel.", + "C": "There are plates, a bottle of sparkling water, and napkins laid out on the table upstairs.", + "D": "On the second-floor table, you can see mugs, a jug of water, and a folded cloth towel.", + "E": "There are cups, mineral water, and paper towels on the table." + }, + "answer_label": "E", + "answer_text": "There are cups, mineral water, and paper towels on the table.", + "q_id": "egolife_A1_D4_007-mcq_fixed_9", + "correct_answer": "There are cups, mineral water, and paper towels on the table.", + "original_options": { + "A": "Plates, sparkling water, and napkins.", + "B": "Mugs, tap water, and a towel.", + "C": "There are plates, a bottle of sparkling water, and napkins laid out on the table upstairs.", + "D": "On the second-floor table, you can see mugs, a jug of water, and a folded cloth towel.", + "E": "There are cups, mineral water, and paper towels on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_007-mcq_fixed_2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the thing the guy just gave me?", + "answer": "The item has been placed on the desk in the bedroom where the computer is.", + "question_chinese": "刚刚男生送的东西被我放到了哪里？", + "answer_chinese": "东西被你送到了放有电脑的卧室的桌子上。", + "evidence": [ + { + "description": "At this moment, the boy gave me a piece of paper with a pattern on it.", + "timestamps": { + "start": "00:02:43", + "end": "00:02:44" + } + }, + { + "description": "At this moment, I placed this piece of paper on the desk in the bedroom where the computer is.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:42" + } + } + ], + "question_timestamps": "00:08:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the hallway shelf.", + "B": "On the kitchen counter.", + "C": "The item has been placed on the desk in the bedroom where the computer is.", + "D": "You left it on the hallway shelf beside the stack of unopened mail near the bedrooms.", + "E": "The item is on the desk in the bedroom, but on the side where the mirror is." + }, + "answer_label": "C", + "answer_text": "The item has been placed on the desk in the bedroom where the computer is.", + "q_id": "egolife_A1_D4_007-mcq_fixed_2", + "correct_answer": "The item has been placed on the desk in the bedroom where the computer is.", + "original_options": { + "A": "On the hallway shelf.", + "B": "On the kitchen counter.", + "C": "The item has been placed on the desk in the bedroom where the computer is.", + "D": "You left it on the hallway shelf beside the stack of unopened mail near the bedrooms.", + "E": "The item is on the desk in the bedroom, but on the side where the mirror is." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_007-mcq_fixed_3", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I put the thing the guy just gave me?", + "answer": "The item the guy just gave you has been placed on the desk in the bedroom where the computer is.", + "question_chinese": "刚刚男生送的东西被我放到了哪里？", + "answer_chinese": "刚刚男生送的东西被你送到了放有电脑的卧室的桌子上。", + "evidence": [ + { + "description": "At this moment, the boy gave me a piece of paper with a pattern on it.", + "timestamps": { + "start": "00:02:43", + "end": "00:02:44" + } + }, + { + "description": "At this moment, I placed this piece of paper on the desk in the bedroom where the computer is.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:42" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The item the guy just gave you has been placed on the desk in the bedroom where the computer is.", + "B": "On the bedroom shelf.", + "C": "On the kitchen counter.", + "D": "You set it down on the hallway shelf near the stack of unopened mail after walking away.", + "E": "You placed it on the chair in the bedroom, away from the desk with the computer." + }, + "answer_label": "A", + "answer_text": "The item the guy just gave you has been placed on the desk in the bedroom where the computer is.", + "q_id": "egolife_A1_D4_007-mcq_fixed_3", + "correct_answer": "The item the guy just gave you has been placed on the desk in the bedroom where the computer is.", + "original_options": { + "A": "The item the guy just gave you has been placed on the desk in the bedroom where the computer is.", + "B": "On the bedroom shelf.", + "C": "On the kitchen counter.", + "D": "You set it down on the hallway shelf near the stack of unopened mail after walking away.", + "E": "You placed it on the chair in the bedroom, away from the desk with the computer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_007-mcq_fixed_4", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How was the animal-shaped stool just placed into the bedroom?", + "answer": "It was kicked by you from beside the living room table straight into the bedroom.", + "question_chinese": "动物板凳刚刚是怎么放到卧室里的？", + "answer_chinese": "它你从客厅桌子旁直走踢到了卧室里。", + "evidence": [ + { + "description": "Right now, you are kicking the animal-shaped stool into the bedroom.", + "timestamps": { + "start": "00:03:58", + "end": "00:04:07" + } + } + ], + "question_timestamps": "00:08:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You pushed it into the bedroom.", + "B": "You carried it into the bedroom.", + "C": "You slid the stool from beside the living room table and guided it across the floor into the bedroom.", + "D": "It was kicked by you from beside the living room table straight into the bedroom.", + "E": "It was kicked from beside the dining room table straight through the doorway into the bedroom." + }, + "answer_label": "D", + "answer_text": "It was kicked by you from beside the living room table straight into the bedroom.", + "q_id": "egolife_A1_D4_007-mcq_fixed_4", + "correct_answer": "It was kicked by you from beside the living room table straight into the bedroom.", + "original_options": { + "A": "You pushed it into the bedroom.", + "B": "You carried it into the bedroom.", + "C": "You slid the stool from beside the living room table and guided it across the floor into the bedroom.", + "D": "It was kicked by you from beside the living room table straight into the bedroom.", + "E": "It was kicked from beside the dining room table straight through the doorway into the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_007-mcq_fixed_5", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How was the animal-shaped stool just placed into the bedroom?", + "answer": "The animal-shaped stool was kicked by you from beside the living room table straight into the bedroom.", + "question_chinese": "动物板凳刚刚是怎么放到卧室里的？", + "answer_chinese": "动物板凳被你从客厅桌子旁直走踢到了卧室里。", + "evidence": [ + { + "description": "Right now, you are kicking the animal-shaped stool into the bedroom.", + "timestamps": { + "start": "00:03:58", + "end": "00:04:07" + } + } + ], + "question_timestamps": "00:10:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It was pushed into the bedroom.", + "B": "It was kicked out of the bedroom.", + "C": "You picked up the animal-shaped stool near the sofa and set it down inside the bedroom carefully.", + "D": "You slid the animal-shaped stool away from the kitchen counter and across the hallway floor into the bedroom.", + "E": "The animal-shaped stool was kicked by you from beside the living room table straight into the bedroom." + }, + "answer_label": "E", + "answer_text": "The animal-shaped stool was kicked by you from beside the living room table straight into the bedroom.", + "q_id": "egolife_A1_D4_007-mcq_fixed_5", + "correct_answer": "The animal-shaped stool was kicked by you from beside the living room table straight into the bedroom.", + "original_options": { + "A": "It was pushed into the bedroom.", + "B": "It was kicked out of the bedroom.", + "C": "You picked up the animal-shaped stool near the sofa and set it down inside the bedroom carefully.", + "D": "You slid the animal-shaped stool away from the kitchen counter and across the hallway floor into the bedroom.", + "E": "The animal-shaped stool was kicked by you from beside the living room table straight into the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_007-mcq_fixed_7", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I sat on the second-floor sofa just now?", + "answer": "No, you just did not sit on the sofa on the second floor.", + "question_chinese": "我刚刚坐没坐过过二楼的沙发？", + "answer_chinese": "没有，你刚刚没有坐过二楼的沙发。", + "evidence": [ + { + "description": "There is no evidence that you have sat on the sofa.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:04:09", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you just did not sit on the sofa on the second floor.", + "B": "Yes, you just sat on the sofa on the second floor." + }, + "answer_label": "A", + "answer_text": "No, you just did not sit on the sofa on the second floor.", + "q_id": "egolife_A1_D4_007-mcq_fixed_7", + "correct_answer": "No, you just did not sit on the sofa on the second floor.", + "original_options": { + "A": "No, you just did not sit on the sofa on the second floor.", + "B": "Yes, you just sat on the sofa on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D4_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_007-mcq_fixed_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Have I sat on the second-floor sofa just now?", + "answer": "You just sat on the sofa on the second floor.", + "question_chinese": "我刚刚坐没坐过过二楼的沙发？", + "answer_chinese": "你刚刚坐过二楼的沙发。", + "evidence": [ + { + "description": "Right now, you are getting up from the sofa.", + "timestamps": { + "start": "00:07:01", + "end": "00:07:05" + } + }, + { + "description": "Right now, you are sitting on the sofa and starting to play with your phone.", + "timestamps": { + "start": "00:05:26", + "end": "00:05:38" + } + } + ], + "question_timestamps": "00:09:50", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You did not just sit on the sofa on the second floor.", + "B": "You just sat on the sofa on the second floor." + }, + "answer_label": "B", + "answer_text": "You just sat on the sofa on the second floor.", + "q_id": "egolife_A1_D4_007-mcq_fixed_6", + "correct_answer": "You just sat on the sofa on the second floor.", + "original_options": { + "A": "You did not just sit on the sofa on the second floor.", + "B": "You just sat on the sofa on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_007_1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I positioned in relation to the dining table?", + "answer": "You are now standing about one meter to the left or right of the dining table, facing the kitchen.", + "question_chinese": "我现在相对于餐桌的位置在哪里?", + "answer_chinese": "你现在站在餐桌左右侧约1米处,面向厨房方向。", + "evidence": [ + { + "description": "You are now standing about 1 meter to the left or right of the dining table, facing the kitchen.", + "timestamps": { + "start": "00:08:09", + "end": "00:08:12" + } + } + ], + "question_timestamps": "00:08:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Two meters left of the dining table.", + "B": "You are now standing about one meter to the left or right of the dining table, facing the kitchen.", + "C": "One meter in front of the dining table.", + "D": "You are standing about one meter to the right of the dining table, facing the living room area.", + "E": "You are standing about one meter behind the dining table, facing the kitchen as if preparing to walk past it." + }, + "answer_label": "B", + "answer_text": "You are now standing about one meter to the left or right of the dining table, facing the kitchen.", + "q_id": "egolife_A1_D4_007_1", + "correct_answer": "You are now standing about one meter to the left or right of the dining table, facing the kitchen.", + "original_options": { + "A": "Two meters left of the dining table.", + "B": "You are now standing about one meter to the left or right of the dining table, facing the kitchen.", + "C": "One meter in front of the dining table.", + "D": "You are standing about one meter to the right of the dining table, facing the living room area.", + "E": "You are standing about one meter behind the dining table, facing the kitchen as if preparing to walk past it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_007_2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I positioned in relation to the dining table?", + "answer": "You are now sitting on the bench on the right side of the dining table, facing the direction of the living room door.", + "question_chinese": "我现在相对于餐桌的位置在哪里?", + "answer_chinese": "你现在站坐在餐桌右侧板凳上,面向客厅大门方向。", + "evidence": [ + { + "description": "You are now sitting on the bench on the right side of the dining table, facing the direction of the living room door.", + "timestamps": { + "start": "00:09:35", + "end": "00:09:37" + } + } + ], + "question_timestamps": "00:09:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Sitting on the left bench.", + "B": "Standing at the table’s end.", + "C": "You are seated on the left-side bench, facing toward the living room door instead of the right side.", + "D": "You are now sitting on the bench on the right side of the dining table, facing the direction of the living room door.", + "E": "You are sitting on the right-side bench, but you’re facing toward the kitchen rather than the living room door." + }, + "answer_label": "D", + "answer_text": "You are now sitting on the bench on the right side of the dining table, facing the direction of the living room door.", + "q_id": "egolife_A1_D4_007_2", + "correct_answer": "You are now sitting on the bench on the right side of the dining table, facing the direction of the living room door.", + "original_options": { + "A": "Sitting on the left bench.", + "B": "Standing at the table’s end.", + "C": "You are seated on the left-side bench, facing toward the living room door instead of the right side.", + "D": "You are now sitting on the bench on the right side of the dining table, facing the direction of the living room door.", + "E": "You are sitting on the right-side bench, but you’re facing toward the kitchen rather than the living room door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 20 + ] + } +] diff --git a/QAs/egolife_A1_D4_008-mcq.json b/QAs/egolife_A1_D4_008-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..96d41b1e6c2d521b8219156cc41e71753515279f --- /dev/null +++ b/QAs/egolife_A1_D4_008-mcq.json @@ -0,0 +1,794 @@ +[ + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008-mcq_fixed_2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I throw all the trash I just picked up?", + "answer": "You threw all the trash you just picked up into the trash can behind you under the table in the yard.", + "question_chinese": "我刚刚捡起的垃圾都扔到了哪里？", + "answer_chinese": "你刚刚捡起的垃圾都扔到了你身后院子里的桌子下面的垃圾桶。", + "evidence": [ + { + "description": "At this moment, you threw all the trash you picked up into the trash can under the table in the yard.", + "timestamps": { + "start": "00:08:45", + "end": "00:08:48" + } + }, + { + "description": "At this moment, you threw all the trash you picked up into the trash can under the table in the yard.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Into a red bag under the table.", + "B": "Into a bin in front of you.", + "C": "You tossed all the trash into a red bag behind you, tucked under the table in the yard.", + "D": "You threw all the trash you just picked up into the trash can behind you under the table in the yard.", + "E": "You put all the trash into a covered container beside the garage, near where the garden tools are kept." + }, + "answer_label": "D", + "answer_text": "You threw all the trash you just picked up into the trash can behind you under the table in the yard.", + "q_id": "egolife_A1_D4_008-mcq_fixed_2", + "correct_answer": "You threw all the trash you just picked up into the trash can behind you under the table in the yard.", + "original_options": { + "A": "Into a red bag under the table.", + "B": "Into a bin in front of you.", + "C": "You tossed all the trash into a red bag behind you, tucked under the table in the yard.", + "D": "You threw all the trash you just picked up into the trash can behind you under the table in the yard.", + "E": "You put all the trash into a covered container beside the garage, near where the garden tools are kept." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008-mcq_fixed_3", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where did I throw all the trash I just picked up?", + "answer": "You threw all the trash you just picked up into the trash can in front of you under the table in the yard.", + "question_chinese": "我刚刚捡起的垃圾都扔到了哪里？", + "answer_chinese": "你刚刚捡起的垃圾都扔到了你对面院子里的桌子下面的垃圾桶。", + "evidence": [ + { + "description": "At this moment, you threw all the trash you picked up into the trash can under the table in the yard.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:19" + } + }, + { + "description": "At this moment, you threw all the trash you picked up into the trash can under the table in the yard.", + "timestamps": { + "start": "00:08:45", + "end": "00:08:46" + } + } + ], + "question_timestamps": "00:11:09", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Into the bin to your left.", + "B": "Into a wooden crate under the table.", + "C": "You threw all the trash you picked up into the trash can to your left under the table in the yard.", + "D": "You threw all the trash you just picked up into the trash can in front of you under the table in the yard.", + "E": "You tossed all the trash into a wooden crate in front of you under the table in the yard." + }, + "answer_label": "D", + "answer_text": "You threw all the trash you just picked up into the trash can in front of you under the table in the yard.", + "q_id": "egolife_A1_D4_008-mcq_fixed_3", + "correct_answer": "You threw all the trash you just picked up into the trash can in front of you under the table in the yard.", + "original_options": { + "A": "Into the bin to your left.", + "B": "Into a wooden crate under the table.", + "C": "You threw all the trash you picked up into the trash can to your left under the table in the yard.", + "D": "You threw all the trash you just picked up into the trash can in front of you under the table in the yard.", + "E": "You tossed all the trash into a wooden crate in front of you under the table in the yard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008_6_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Was the door to the room with the washing machine open or closed when I saw it?", + "answer": "The door is closed right now, and you’re about to open it.", + "question_chinese": "我看到放置洗衣机的房间的门是开着的还是关着的？", + "answer_chinese": "现在门是关着的，你正准备打开它。", + "evidence": [ + { + "description": "The door to the room where the washing machine is placed is closed.", + "timestamps": { + "start": "00:03:46", + "end": "00:03:48" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The door is closed right now, and you’re about to open it.", + "B": "The door is open right now; after you opened it, you didn’t close it." + }, + "answer_label": "A", + "answer_text": "The door is closed right now, and you’re about to open it.", + "q_id": "egolife_A1_D4_008_6_T2", + "correct_answer": "The door is closed right now, and you’re about to open it.", + "original_options": { + "A": "The door is closed right now, and you’re about to open it.", + "B": "The door is open right now; after you opened it, you didn’t close it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008_6_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Was the door to the room with the washing machine open or closed when I saw it?", + "answer": "The door is open right now; after you opened it, you didn’t close it.", + "question_chinese": "我看到放置洗衣机的房间的门是开着的还是关着的？", + "answer_chinese": "现在门是开着的，你打开之后没有关闭。", + "evidence": [ + { + "description": "The door of the room where the washing machine is placed is open; you didn't close it after opening it.", + "timestamps": { + "start": "00:03:48", + "end": "00:03:52" + } + } + ], + "question_timestamps": "00:04:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The door is closed right now, and you’re about to open it.", + "B": "The door is open right now; after you opened it, you didn’t close it." + }, + "answer_label": "B", + "answer_text": "The door is open right now; after you opened it, you didn’t close it.", + "q_id": "egolife_A1_D4_008_6_T1", + "correct_answer": "The door is open right now; after you opened it, you didn’t close it.", + "original_options": { + "A": "The door is closed right now, and you’re about to open it.", + "B": "The door is open right now; after you opened it, you didn’t close it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008_1_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes did I see on the table in the yard?", + "answer": "There are no boxes on the table in the yard.", + "question_chinese": "我看到院子里的桌子上有几个盒子？", + "answer_chinese": "院子里的桌子上没有盒子。", + "evidence": [ + { + "description": "At this moment, you can see that there are no boxes on the table.", + "timestamps": { + "start": "00:05:48", + "end": "00:05:49" + } + } + ], + "question_timestamps": "00:06:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One box on the table.", + "B": "Three boxes on the table.", + "C": "At this moment, you can see four boxes sitting on the table in the yard.", + "D": "At this moment, you can see seven boxes arranged across the table in the yard.", + "E": "There are no boxes on the table in the yard." + }, + "answer_label": "E", + "answer_text": "There are no boxes on the table in the yard.", + "q_id": "egolife_A1_D4_008_1_T1", + "correct_answer": "There are no boxes on the table in the yard.", + "original_options": { + "A": "One box on the table.", + "B": "Three boxes on the table.", + "C": "At this moment, you can see four boxes sitting on the table in the yard.", + "D": "At this moment, you can see seven boxes arranged across the table in the yard.", + "E": "There are no boxes on the table in the yard." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008_1_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes did I see on the table in the yard?", + "answer": "There are four boxes on the table in the yard.", + "question_chinese": "我看到院子里的桌子上有几个盒子？", + "answer_chinese": "院子里的桌子上有四个盒子。", + "evidence": [ + { + "description": "Brought over four boxes and put them all on the table.", + "timestamps": { + "start": "00:09:30", + "end": "00:09:55" + } + } + ], + "question_timestamps": "00:09:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are four boxes on the table in the yard.", + "B": "Two boxes on the table.", + "C": "Six boxes on the table.", + "D": "There are three boxes sitting on the table in the yard, not four.", + "E": "There are five boxes on the table in the yard, so the count is higher than four." + }, + "answer_label": "A", + "answer_text": "There are four boxes on the table in the yard.", + "q_id": "egolife_A1_D4_008_1_T2", + "correct_answer": "There are four boxes on the table in the yard.", + "original_options": { + "A": "There are four boxes on the table in the yard.", + "B": "Two boxes on the table.", + "C": "Six boxes on the table.", + "D": "There are three boxes sitting on the table in the yard, not four.", + "E": "There are five boxes on the table in the yard, so the count is higher than four." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008_2_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many boxes do I see on the table in the yard?", + "answer": "At this moment, you can see three boxes on the table.", + "question_chinese": "此时我看到院子里的桌子上有几个盒子？", + "answer_chinese": "此时你可以看到桌子上有三个盒子。", + "evidence": [ + { + "description": "Brought over four boxes, put three on the table, the girl in white is holding one box.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:43" + } + } + ], + "question_timestamps": "00:09:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "At this moment, you can see three boxes on the table.", + "B": "One box on the table.", + "C": "Five boxes on the table.", + "D": "At this moment, you can see two boxes on the table, not three.", + "E": "At this moment, you can see four boxes on the table, so there is one more than three." + }, + "answer_label": "A", + "answer_text": "At this moment, you can see three boxes on the table.", + "q_id": "egolife_A1_D4_008_2_T1", + "correct_answer": "At this moment, you can see three boxes on the table.", + "original_options": { + "A": "At this moment, you can see three boxes on the table.", + "B": "One box on the table.", + "C": "Five boxes on the table.", + "D": "At this moment, you can see two boxes on the table, not three.", + "E": "At this moment, you can see four boxes on the table, so there is one more than three." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008_2_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many boxes do I see on the table in the yard?", + "answer": "At this moment, you can see seven boxes on the table.", + "question_chinese": "此时我看到院子里的桌子上有几个盒子？", + "answer_chinese": "此时你可以看到桌子上有七个盒子。", + "evidence": [ + { + "description": "There were originally three boxes on the table. At this moment, the girl placed the box in her hand onto the table, so now there are four boxes on the table.", + "timestamps": { + "start": "00:09:57", + "end": "00:09:58" + } + } + ], + "question_timestamps": "00:10:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five boxes.", + "B": "At this moment, you can see seven boxes on the table.", + "C": "Nine boxes.", + "D": "At this moment, you can see six boxes on the table in the yard.", + "E": "At this moment, you can see eight boxes lined up on the table in the yard." + }, + "answer_label": "B", + "answer_text": "At this moment, you can see seven boxes on the table.", + "q_id": "egolife_A1_D4_008_2_T2", + "correct_answer": "At this moment, you can see seven boxes on the table.", + "original_options": { + "A": "Five boxes.", + "B": "At this moment, you can see seven boxes on the table.", + "C": "Nine boxes.", + "D": "At this moment, you can see six boxes on the table in the yard.", + "E": "At this moment, you can see eight boxes lined up on the table in the yard." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008_5", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes are on the table in the yard?", + "answer": "There are no boxes on the table in the yard.", + "question_chinese": "此时院子里的桌子上有几个盒子？", + "answer_chinese": "院子里的桌子上没有盒子。", + "evidence": [ + { + "description": "At this moment, you can see that there are no boxes on the table.", + "timestamps": { + "start": "00:05:48", + "end": "00:05:49" + } + } + ], + "question_timestamps": "00:06:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two boxes.", + "B": "There are no boxes on the table in the yard.", + "C": "Five boxes.", + "D": "There is one box on the table in the yard right now.", + "E": "There are three boxes on the table in the yard at the moment." + }, + "answer_label": "B", + "answer_text": "There are no boxes on the table in the yard.", + "q_id": "egolife_A1_D4_008_5", + "correct_answer": "There are no boxes on the table in the yard.", + "original_options": { + "A": "Two boxes.", + "B": "There are no boxes on the table in the yard.", + "C": "Five boxes.", + "D": "There is one box on the table in the yard right now.", + "E": "There are three boxes on the table in the yard at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008_6", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes are on the table in the yard right now?", + "answer": "At this moment, you can see four boxes on the table.", + "question_chinese": "此时院子里的桌子上有几个盒子？", + "answer_chinese": "此时你可以看到桌子上有四个盒子。", + "evidence": [ + { + "description": "There were originally three boxes on the table. At this moment, the girl placed the box in her hand onto the table, so now there are four boxes on the table.", + "timestamps": { + "start": "00:09:57", + "end": "00:09:58" + } + } + ], + "question_timestamps": "00:10:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One box.", + "B": "At this moment, you can see four boxes on the table.", + "C": "Five boxes.", + "D": "At this moment, you can see two boxes on the table in the yard.", + "E": "At this moment, you can see three boxes sitting on the table in the yard." + }, + "answer_label": "B", + "answer_text": "At this moment, you can see four boxes on the table.", + "q_id": "egolife_A1_D4_008_6", + "correct_answer": "At this moment, you can see four boxes on the table.", + "original_options": { + "A": "One box.", + "B": "At this moment, you can see four boxes on the table.", + "C": "Five boxes.", + "D": "At this moment, you can see two boxes on the table in the yard.", + "E": "At this moment, you can see three boxes sitting on the table in the yard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008-mcq_fixed_8", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many light stands are next to the barbecue grill?", + "answer": "Five.", + "question_chinese": "烧烤架旁有几个打光灯？", + "answer_chinese": "烧烤架旁有两个打光灯。", + "evidence": [ + { + "description": "You can see two light stands next to the barbecue grill, one of which has fallen over.", + "timestamps": { + "start": "00:04:36", + "end": "00:04:37" + } + }, + { + "description": "You can see two light stands next to the barbecue grill; the fallen one has been picked up by me.", + "timestamps": { + "start": "00:07:17", + "end": "00:07:18" + } + }, + { + "description": "You can see two light stands next to the barbecue grill, one of which has fallen over.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:35" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Two.", + "C": "Four.", + "D": "There are three light stands next to the barbecue grill right now.", + "E": "You can see only one light stand positioned next to the barbecue grill." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D4_008-mcq_fixed_8", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "Two.", + "C": "Four.", + "D": "There are three light stands next to the barbecue grill right now.", + "E": "You can see only one light stand positioned next to the barbecue grill." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008-mcq_fixed_7", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many light stands are next to the barbecue grill?", + "answer": "One.", + "question_chinese": "烧烤架旁有几个打光灯？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "You can see two light stands next to the barbecue grill, one of which has fallen over.", + "timestamps": { + "start": "00:04:36", + "end": "00:04:37" + } + }, + { + "description": "You can see two light stands next to the barbecue grill; the fallen one has been picked up by me.", + "timestamps": { + "start": "00:07:17", + "end": "00:07:18" + } + }, + { + "description": "You can see two light stands next to the barbecue grill, one of which has fallen over.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:35" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Five.", + "D": "There are three light stands next to the barbecue grill right now.", + "E": "You can see four light stands positioned next to the barbecue grill." + }, + "answer_label": "A", + "answer_text": "One.", + "q_id": "egolife_A1_D4_008-mcq_fixed_7", + "correct_answer": "One.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Five.", + "D": "There are three light stands next to the barbecue grill right now.", + "E": "You can see four light stands positioned next to the barbecue grill." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008-mcq_fixed_4", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is placed in the corner of the wall to the right of the yard gate?", + "answer": "There is a round table and a chair placed in the corner of the wall to the right of the yard gate.", + "question_chinese": "院子大门右侧墙角放的有什么？", + "answer_chinese": "院子大门右侧墙角放的有一个圆桌和一个椅子。", + "evidence": [ + { + "description": "At this moment, you can see that there is a round table and a chair placed in the corner of the wall to the right of the yard gate.", + "timestamps": { + "start": "00:05:18", + "end": "00:05:20" + } + }, + { + "description": "At this moment, you can see that there is a round table and a chair placed in the corner of the wall to the right of the yard gate.", + "timestamps": { + "start": "00:06:45", + "end": "00:06:46" + } + } + ], + "question_timestamps": "00:09:22", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A square table and two chairs.", + "B": "A bench along the wall.", + "C": "There is a small storage cabinet and a chair placed in the corner to the right of the yard gate.", + "D": "There is a potted plant and a narrow shelf set in the corner of the wall.", + "E": "There is a round table and a chair placed in the corner of the wall to the right of the yard gate." + }, + "answer_label": "E", + "answer_text": "There is a round table and a chair placed in the corner of the wall to the right of the yard gate.", + "q_id": "egolife_A1_D4_008-mcq_fixed_4", + "correct_answer": "There is a round table and a chair placed in the corner of the wall to the right of the yard gate.", + "original_options": { + "A": "A square table and two chairs.", + "B": "A bench along the wall.", + "C": "There is a small storage cabinet and a chair placed in the corner to the right of the yard gate.", + "D": "There is a potted plant and a narrow shelf set in the corner of the wall.", + "E": "There is a round table and a chair placed in the corner of the wall to the right of the yard gate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_008-mcq_fixed_5", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is placed in the corner of the wall to the right of the yard gate?", + "answer": "There is a round table and a chair.", + "question_chinese": "院子大门右侧墙角放的有什么？", + "answer_chinese": "有一个圆桌和一个椅子。", + "evidence": [ + { + "description": "At this moment, you can see that there is a round table and a chair placed in the corner of the wall to the right of the yard gate.", + "timestamps": { + "start": "00:05:18", + "end": "00:05:20" + } + }, + { + "description": "At this moment, you can see that there is a round table and a chair placed in the corner of the wall to the right of the yard gate.", + "timestamps": { + "start": "00:08:45", + "end": "00:08:46" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A square table and a lamp.", + "B": "A triangular table and a chair.", + "C": "There is a round table and a chair.", + "D": "A stack of neatly arranged firewood is placed in the corner of the wall to the right of the yard gate.", + "E": "An outdoor storage cabinet with gardening tools is placed there, beside a small ornamental shrub." + }, + "answer_label": "C", + "answer_text": "There is a round table and a chair.", + "q_id": "egolife_A1_D4_008-mcq_fixed_5", + "correct_answer": "There is a round table and a chair.", + "original_options": { + "A": "A square table and a lamp.", + "B": "A triangular table and a chair.", + "C": "There is a round table and a chair.", + "D": "A stack of neatly arranged firewood is placed in the corner of the wall to the right of the yard gate.", + "E": "An outdoor storage cabinet with gardening tools is placed there, beside a small ornamental shrub." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D4_009-mcq.json b/QAs/egolife_A1_D4_009-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..926c6d7667068a5f20c0ea6486adf4caecfd7f68 --- /dev/null +++ b/QAs/egolife_A1_D4_009-mcq.json @@ -0,0 +1,1224 @@ +[ + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_6_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the large whiteboard relative to me?", + "answer": "It’s to your left.", + "question_chinese": "大白板相对于我在哪里？", + "answer_chinese": "在你的左边。", + "evidence": [ + { + "description": "On your left.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s behind you.", + "C": "It’s to your right, just off to the side as you face forward.", + "D": "It’s directly ahead of you, positioned in front of where you are standing.", + "E": "It’s to your left." + }, + "answer_label": "E", + "answer_text": "It’s to your left.", + "q_id": "egolife_A1_D4_009_6_T2", + "correct_answer": "It’s to your left.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s behind you.", + "C": "It’s to your right, just off to the side as you face forward.", + "D": "It’s directly ahead of you, positioned in front of where you are standing.", + "E": "It’s to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_6_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the large whiteboard relative to me?", + "answer": "It’s behind you.", + "question_chinese": "大白板相对于我在哪里？", + "answer_chinese": "在你的后面。", + "evidence": [ + { + "description": "Behind you.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:00:42", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s to your left.", + "C": "It’s to your right, near where you’re facing as you stand.", + "D": "It’s behind you.", + "E": "It’s above you, positioned higher up than your current eye level." + }, + "answer_label": "D", + "answer_text": "It’s behind you.", + "q_id": "egolife_A1_D4_009_6_T1", + "correct_answer": "It’s behind you.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s to your left.", + "C": "It’s to your right, near where you’re facing as you stand.", + "D": "It’s behind you.", + "E": "It’s above you, positioned higher up than your current eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_6_T3", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the large whiteboard relative to me?", + "answer": "It’s to your right.", + "question_chinese": "大白板相对于我在哪里？", + "answer_chinese": "在你的右边。", + "evidence": [ + { + "description": "On your right.", + "timestamps": { + "start": "00:01:17", + "end": "00:01:18" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to your right.", + "B": "It’s to your left.", + "C": "It’s behind you.", + "D": "It’s in front of you, directly along the direction you’re currently facing.", + "E": "It’s above you, hanging higher than your head from where you’re standing." + }, + "answer_label": "A", + "answer_text": "It’s to your right.", + "q_id": "egolife_A1_D4_009_6_T3", + "correct_answer": "It’s to your right.", + "original_options": { + "A": "It’s to your right.", + "B": "It’s to your left.", + "C": "It’s behind you.", + "D": "It’s in front of you, directly along the direction you’re currently facing.", + "E": "It’s above you, hanging higher than your head from where you’re standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009-mcq_fixed_0", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I just get the scissors I gave to the girl?", + "answer": "The scissors were taken from the bed in the bedroom where there are multiple computers.", + "question_chinese": "我刚刚给女生的剪刀是从哪里拿的？", + "answer_chinese": "剪刀是从放有多台电脑的卧室的床上拿的。", + "evidence": [ + { + "description": "At this moment, you pick up the scissors from the bed.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From a drawer in that bedroom.", + "B": "From a shelf in that bedroom.", + "C": "You picked up the scissors from the chair in the bedroom with multiple computers.", + "D": "The scissors were taken from the bed in the bedroom where there are multiple computers.", + "E": "You retrieved the scissors from a closet in the bedroom where there are multiple computers." + }, + "answer_label": "D", + "answer_text": "The scissors were taken from the bed in the bedroom where there are multiple computers.", + "q_id": "egolife_A1_D4_009-mcq_fixed_0", + "correct_answer": "The scissors were taken from the bed in the bedroom where there are multiple computers.", + "original_options": { + "A": "From a drawer in that bedroom.", + "B": "From a shelf in that bedroom.", + "C": "You picked up the scissors from the chair in the bedroom with multiple computers.", + "D": "The scissors were taken from the bed in the bedroom where there are multiple computers.", + "E": "You retrieved the scissors from a closet in the bedroom where there are multiple computers." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009-mcq_fixed_1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I just get the scissors I gave to the girl?", + "answer": "The scissors you just gave the girl were taken from the bed in the bedroom where there are multiple computers.", + "question_chinese": "我刚刚给女生的剪刀是从哪里拿的？", + "answer_chinese": "你刚刚给女生的剪刀是从放有多台电脑的卧室的床上拿的。", + "evidence": [ + { + "description": "At this moment, you pick up the scissors from the bed.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The scissors you just gave the girl were taken from the bed in the bedroom where there are multiple computers.", + "B": "They were taken from the chair.", + "C": "They were taken from the closet.", + "D": "You grabbed the scissors from a drawer in the bedroom that has multiple computers.", + "E": "You took the scissors from a shelf in the bedroom where there are multiple computers." + }, + "answer_label": "A", + "answer_text": "The scissors you just gave the girl were taken from the bed in the bedroom where there are multiple computers.", + "q_id": "egolife_A1_D4_009-mcq_fixed_1", + "correct_answer": "The scissors you just gave the girl were taken from the bed in the bedroom where there are multiple computers.", + "original_options": { + "A": "The scissors you just gave the girl were taken from the bed in the bedroom where there are multiple computers.", + "B": "They were taken from the chair.", + "C": "They were taken from the closet.", + "D": "You grabbed the scissors from a drawer in the bedroom that has multiple computers.", + "E": "You took the scissors from a shelf in the bedroom where there are multiple computers." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_7", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get from the bedroom with multiple computers to where I am now?", + "answer": "You exited the bedroom and walked most of the way around the dining table to reach your current position.", + "question_chinese": "我是怎么从放有多台电脑的卧室里到达现在这个位置的？", + "answer_chinese": "你从卧室出门，沿着餐桌走大半圈到达了现在的位置。", + "evidence": [ + { + "description": "At this moment, you come out of the bedroom, and the rack is opposite the dining table.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + }, + { + "description": "At this moment, you walk straight two steps along the table.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:34" + } + }, + { + "description": "At this moment, you take one step left, then turn left again and walk straight to your current position.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:49", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You walked straight past the dining table.", + "B": "You went through the hallway to get here.", + "C": "You exited the bedroom and walked only a short way around the dining table to arrive.", + "D": "You left the bedroom and looped completely around the dining table before stopping where you are now.", + "E": "You exited the bedroom and walked most of the way around the dining table to reach your current position." + }, + "answer_label": "E", + "answer_text": "You exited the bedroom and walked most of the way around the dining table to reach your current position.", + "q_id": "egolife_A1_D4_009_7", + "correct_answer": "You exited the bedroom and walked most of the way around the dining table to reach your current position.", + "original_options": { + "A": "You walked straight past the dining table.", + "B": "You went through the hallway to get here.", + "C": "You exited the bedroom and walked only a short way around the dining table to arrive.", + "D": "You left the bedroom and looped completely around the dining table before stopping where you are now.", + "E": "You exited the bedroom and walked most of the way around the dining table to reach your current position." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009-mcq_fixed_7", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get from the bedroom with multiple computers to my current location?", + "answer": "You exited the bedroom, walked most of the way around the dining table, then turned back and went straight toward the front-right to reach your current position.", + "question_chinese": "我是怎么从放有多台电脑的卧室里到达现在这个位置的？", + "answer_chinese": "你从卧室出门，沿着餐桌走大半圈，然后又后转向右前方直走到达了现在的位置。", + "evidence": [ + { + "description": "At this moment, you come out of the bedroom, and the rack is opposite the dining table.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + }, + { + "description": "At this moment, you walk straight two steps along the table.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:34" + } + }, + { + "description": "At this moment, you take one step left, then turn left again and walk straight to your current position.", + "timestamps": { + "start": "00:00:40", + "end": "00:00:44" + } + }, + { + "description": "You then turned back, walked straight toward the front-right, and arrived at your current position.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You went straight toward the front-left.", + "B": "You went straight toward the back-right.", + "C": "You exited the bedroom, circled most of the dining table, then headed straight toward the back-left to stop.", + "D": "You left the bedroom, went almost all the way around the dining table, then continued forward without turning back.", + "E": "You exited the bedroom, walked most of the way around the dining table, then turned back and went straight toward the front-right to reach your current position." + }, + "answer_label": "E", + "answer_text": "You exited the bedroom, walked most of the way around the dining table, then turned back and went straight toward the front-right to reach your current position.", + "q_id": "egolife_A1_D4_009-mcq_fixed_7", + "correct_answer": "You exited the bedroom, walked most of the way around the dining table, then turned back and went straight toward the front-right to reach your current position.", + "original_options": { + "A": "You went straight toward the front-left.", + "B": "You went straight toward the back-right.", + "C": "You exited the bedroom, circled most of the dining table, then headed straight toward the back-left to stop.", + "D": "You left the bedroom, went almost all the way around the dining table, then continued forward without turning back.", + "E": "You exited the bedroom, walked most of the way around the dining table, then turned back and went straight toward the front-right to reach your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_8", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the bench now?", + "answer": "You can sit on the stool.", + "question_chinese": "此时我可以直接坐到板凳上吗？", + "answer_chinese": "你可以坐到板凳上。", + "evidence": [ + { + "description": "There is a chair placed diagonally to the left of the table.", + "timestamps": { + "start": "00:01:28", + "end": "00:01:30" + } + }, + { + "description": "You are now standing in front of the chair placed diagonally to the left of the table, so you can sit on the stool.", + "timestamps": { + "start": "00:01:32", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:39", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You cannot sit on the stool.", + "B": "You can sit on the stool." + }, + "answer_label": "B", + "answer_text": "You can sit on the stool.", + "q_id": "egolife_A1_D4_009_8", + "correct_answer": "You can sit on the stool.", + "original_options": { + "A": "You cannot sit on the stool.", + "B": "You can sit on the stool." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009-mcq_fixed_11", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit on the stool directly now?", + "answer": "Yes，the stool is on your right.", + "question_chinese": "此时我可以直接坐到板凳上吗？", + "answer_chinese": "可以，板凳在你的右手边。", + "evidence": [ + { + "description": "The stool is on your right side now.", + "timestamps": { + "start": "00:08:41", + "end": "00:08:42" + } + } + ], + "question_timestamps": "00:08:58", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, the stool is on your left.", + "B": "Yes，the stool is on your right." + }, + "answer_label": "B", + "answer_text": "Yes，the stool is on your right.", + "q_id": "egolife_A1_D4_009-mcq_fixed_11", + "correct_answer": "Yes，the stool is on your right.", + "original_options": { + "A": "No, the stool is on your left.", + "B": "Yes，the stool is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_2_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit down on the chair directly right now?", + "answer": "You can sit down on the chair directly; the chair is in front of you.", + "question_chinese": "此时我可以直接坐到椅子上吗？", + "answer_chinese": "你可以直接坐到椅子上，椅子在你的正前方。", + "evidence": [ + { + "description": "You can sit in the chair.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t sit down on the chair directly; the chair is to your left.", + "B": "You can sit down on the chair directly; the chair is in front of you." + }, + "answer_label": "B", + "answer_text": "You can sit down on the chair directly; the chair is in front of you.", + "q_id": "egolife_A1_D4_009_2_T2", + "correct_answer": "You can sit down on the chair directly; the chair is in front of you.", + "original_options": { + "A": "You can’t sit down on the chair directly; the chair is to your left.", + "B": "You can sit down on the chair directly; the chair is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_2_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit down on the chair directly right now?", + "answer": "You can’t sit down on the chair directly; the chair is to your left.", + "question_chinese": "此时我可以直接坐到椅子上吗？", + "answer_chinese": "你不可以直接坐到椅子上，椅子在你的左边。", + "evidence": [ + { + "description": "You can't just sit on the bench.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t sit down on the chair directly; the chair is to your left.", + "B": "You can sit down on the chair directly; the chair is in front of you." + }, + "answer_label": "A", + "answer_text": "You can’t sit down on the chair directly; the chair is to your left.", + "q_id": "egolife_A1_D4_009_2_T1", + "correct_answer": "You can’t sit down on the chair directly; the chair is to your left.", + "original_options": { + "A": "You can’t sit down on the chair directly; the chair is to your left.", + "B": "You can sit down on the chair directly; the chair is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009-mcq_fixed_5", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there to my left at this moment?", + "answer": "Six.", + "question_chinese": "此时我左边有几个椅子？", + "answer_chinese": "你左边没有椅子。", + "evidence": [ + { + "description": "There is a wall to your left now, not a chair.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "There are four chairs to your left at this moment right now.", + "D": "You have eight chairs to your left at this moment right now.", + "E": "Six." + }, + "answer_label": "E", + "answer_text": "Six.", + "q_id": "egolife_A1_D4_009-mcq_fixed_5", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "There are four chairs to your left at this moment right now.", + "D": "You have eight chairs to your left at this moment right now.", + "E": "Six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009-mcq_fixed_4", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there to my left at this moment?", + "answer": "Six.", + "question_chinese": "此时我左边有几个椅子？", + "answer_chinese": "你左边有两个椅子。", + "evidence": [ + { + "description": "There are three chairs on the side of the table that is against the wall.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:30" + } + }, + { + "description": "You sat in the rightmost one, so there are two chairs to your left.", + "timestamps": { + "start": "00:01:39", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:01:48", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "Four.", + "C": "Eight.", + "D": "There are five chairs to your left at this moment right now.", + "E": "You can count seven chairs lined up to your left right now today." + }, + "answer_label": "A", + "answer_text": "Six.", + "q_id": "egolife_A1_D4_009-mcq_fixed_4", + "correct_answer": "Six.", + "original_options": { + "A": "Six.", + "B": "Four.", + "C": "Eight.", + "D": "There are five chairs to your left at this moment right now.", + "E": "You can count seven chairs lined up to your left right now today." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009-mcq_fixed_9", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment,which one is closer to me: the girl in jeans or the girl in the white long-sleeved top?", + "answer": "The girl wearing a white long-sleeved top is closer to you.", + "question_chinese": "此时穿牛仔衣服的女生和穿白色长袖的女生哪个离我更近?", + "answer_chinese": "穿白色长袖衣服的女生离你更近,距离大约一米。穿牛仔衣服的女生在你前方,距离约三\n两米。", + "evidence": [ + { + "description": "The girl wearing a white long-sleeved top is closer to you, at a distance of about one meter. The girl wearing jeans is in front of you, at a distance of about three to four meters.", + "timestamps": { + "start": "00:09:08", + "end": "00:09:10" + } + } + ], + "question_timestamps": "00:09:28", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The girl wearing a white long-sleeved top is closer to you.", + "B": "The girl wearing jeans is closer to you, at a distance of about one meter." + }, + "answer_label": "A", + "answer_text": "The girl wearing a white long-sleeved top is closer to you.", + "q_id": "egolife_A1_D4_009-mcq_fixed_9", + "correct_answer": "The girl wearing a white long-sleeved top is closer to you.", + "original_options": { + "A": "The girl wearing a white long-sleeved top is closer to you.", + "B": "The girl wearing jeans is closer to you, at a distance of about one meter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009-mcq_fixed_8", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment,which one is closer to me: the girl in jeans or the girl in the white long-sleeved top?", + "answer": "The girl in jeans is closer to you, at a distance of about two meters.", + "question_chinese": "此时穿牛仔衣服的女生和穿白色长袖的女生哪个离我更近?", + "answer_chinese": "穿牛仔衣服的女生离你更近,距离大约两米。穿白色长袖的女生在你左前方,距离约三米。", + "evidence": [ + { + "description": "The girl in jeans is closer to you, at a distance of about two meters.", + "timestamps": { + "start": "00:11:36", + "end": "00:11:37" + } + }, + { + "description": "The girl in the white long-sleeved top is in front and to your left, about three meters away.", + "timestamps": { + "start": "00:11:38", + "end": "00:11:39" + } + } + ], + "question_timestamps": "00:11:40", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The girl wearing a white long-sleeved top is closer to you.", + "B": "The girl in jeans is closer to you, at a distance of about two meters." + }, + "answer_label": "B", + "answer_text": "The girl in jeans is closer to you, at a distance of about two meters.", + "q_id": "egolife_A1_D4_009-mcq_fixed_8", + "correct_answer": "The girl in jeans is closer to you, at a distance of about two meters.", + "original_options": { + "A": "The girl wearing a white long-sleeved top is closer to you.", + "B": "The girl in jeans is closer to you, at a distance of about two meters." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_8_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I close to the sink right now, or far from it?", + "answer": "Right now, you’re farther from the sink, and the sink is in front of you.", + "question_chinese": "现在我离水槽远/近？", + "answer_chinese": "现在你离水槽远，水槽在你的正前方。", + "evidence": [ + { + "description": "Now you stay away from the sink.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, you’re farther from the sink, and the sink is in front of you.", + "B": "Right now, you’re closer to the sink, and the sink is to your left." + }, + "answer_label": "A", + "answer_text": "Right now, you’re farther from the sink, and the sink is in front of you.", + "q_id": "egolife_A1_D4_009_8_T2", + "correct_answer": "Right now, you’re farther from the sink, and the sink is in front of you.", + "original_options": { + "A": "Right now, you’re farther from the sink, and the sink is in front of you.", + "B": "Right now, you’re closer to the sink, and the sink is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_8_T1", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I close to the sink right now, or far from it?", + "answer": "Right now, you’re closer to the sink, and the sink is to your left.", + "question_chinese": "现在我离水槽远/近？", + "answer_chinese": "现在你离水槽近，水槽在你的左边。", + "evidence": [ + { + "description": "Now you are near the sink.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:00:54", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, you’re farther from the sink, and the sink is in front of you.", + "B": "Right now, you’re closer to the sink, and the sink is to your left." + }, + "answer_label": "B", + "answer_text": "Right now, you’re closer to the sink, and the sink is to your left.", + "q_id": "egolife_A1_D4_009_8_T1", + "correct_answer": "Right now, you’re closer to the sink, and the sink is to your left.", + "original_options": { + "A": "Right now, you’re farther from the sink, and the sink is in front of you.", + "B": "Right now, you’re closer to the sink, and the sink is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_5_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, in which direction from me is the chair?", + "answer": "Right now, the chair is in front of you.", + "question_chinese": "现在椅子在我的哪个方向？", + "answer_chinese": "现在椅子在你的正前方。", + "evidence": [ + { + "description": "Right in front of you.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Right now, the chair is behind you.", + "B": "Right now, the chair is in front of you.", + "C": "Right now, the chair is to your left.", + "D": "Right now, the chair is to your right, not directly in front of you.", + "E": "Right now, the chair is above you, rather than being positioned in front." + }, + "answer_label": "B", + "answer_text": "Right now, the chair is in front of you.", + "q_id": "egolife_A1_D4_009_5_T1", + "correct_answer": "Right now, the chair is in front of you.", + "original_options": { + "A": "Right now, the chair is behind you.", + "B": "Right now, the chair is in front of you.", + "C": "Right now, the chair is to your left.", + "D": "Right now, the chair is to your right, not directly in front of you.", + "E": "Right now, the chair is above you, rather than being positioned in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_5_T2", + "qid_num": 19, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, in which direction from me is the chair?", + "answer": "Right now, the chair is behind you.", + "question_chinese": "现在椅子在我的哪个方向？", + "answer_chinese": "现在椅子在你的后面。", + "evidence": [ + { + "description": "Behind you.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:38" + } + } + ], + "question_timestamps": "00:00:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The chair is to your left.", + "B": "Right now, the chair is behind you.", + "C": "The chair is in front.", + "D": "The chair is to your right, not directly behind you at this moment.", + "E": "The chair is above you, rather than behind you, at this moment right now." + }, + "answer_label": "B", + "answer_text": "Right now, the chair is behind you.", + "q_id": "egolife_A1_D4_009_5_T2", + "correct_answer": "Right now, the chair is behind you.", + "original_options": { + "A": "The chair is to your left.", + "B": "Right now, the chair is behind you.", + "C": "The chair is in front.", + "D": "The chair is to your right, not directly behind you at this moment.", + "E": "The chair is above you, rather than behind you, at this moment right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_1_T2", + "qid_num": 20, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the flower on the dining table relative to me?", + "answer": "It’s in front of you.", + "question_chinese": "餐桌上的花相对于我在什么位置？", + "answer_chinese": "在你的正前方。", + "evidence": [ + { + "description": "Right in front of you.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s behind you.", + "B": "It’s to your left.", + "C": "It’s to your right on the dining table, not positioned in front of you.", + "D": "It’s in front of you.", + "E": "It’s above you rather than on the dining table in front of your viewpoint." + }, + "answer_label": "D", + "answer_text": "It’s in front of you.", + "q_id": "egolife_A1_D4_009_1_T2", + "correct_answer": "It’s in front of you.", + "original_options": { + "A": "It’s behind you.", + "B": "It’s to your left.", + "C": "It’s to your right on the dining table, not positioned in front of you.", + "D": "It’s in front of you.", + "E": "It’s above you rather than on the dining table in front of your viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009_1_T1", + "qid_num": 21, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the flower on the dining table relative to me?", + "answer": "It’s behind you.", + "question_chinese": "餐桌上的花相对于我在什么位置？", + "answer_chinese": "在你的后面。", + "evidence": [ + { + "description": "Behind you.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s behind you.", + "B": "It’s in front of you.", + "C": "It’s to your left.", + "D": "It’s to your right on the dining table, not located behind you.", + "E": "It’s above you instead of being behind you relative to where you are standing." + }, + "answer_label": "A", + "answer_text": "It’s behind you.", + "q_id": "egolife_A1_D4_009_1_T1", + "correct_answer": "It’s behind you.", + "original_options": { + "A": "It’s behind you.", + "B": "It’s in front of you.", + "C": "It’s to your left.", + "D": "It’s to your right on the dining table, not located behind you.", + "E": "It’s above you instead of being behind you relative to where you are standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009-mcq_fixed_2", + "qid_num": 22, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the yellow screwdriver on the table taken from?", + "answer": "The yellow screwdriver is on the second shelf of the rack to the right of the whiteboard in the living room.", + "question_chinese": "黄色螺丝刀现在在哪里？", + "answer_chinese": "黄色螺丝刀在客厅小白板右侧架子第二层。", + "evidence": [ + { + "description": "The yellow screwdriver is on the second shelf of the rack to the right of the whiteboard in the living room.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:02:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the top shelf of the rack.", + "B": "The yellow screwdriver is on the second shelf of the rack to the right of the whiteboard in the living room.", + "C": "From the rack left of the whiteboard.", + "D": "The yellow screwdriver is on the third shelf of the rack to the right of the whiteboard in the living room.", + "E": "The yellow screwdriver is on the second shelf of the rack to the left of the whiteboard in the living room." + }, + "answer_label": "B", + "answer_text": "The yellow screwdriver is on the second shelf of the rack to the right of the whiteboard in the living room.", + "q_id": "egolife_A1_D4_009-mcq_fixed_2", + "correct_answer": "The yellow screwdriver is on the second shelf of the rack to the right of the whiteboard in the living room.", + "original_options": { + "A": "From the top shelf of the rack.", + "B": "The yellow screwdriver is on the second shelf of the rack to the right of the whiteboard in the living room.", + "C": "From the rack left of the whiteboard.", + "D": "The yellow screwdriver is on the third shelf of the rack to the right of the whiteboard in the living room.", + "E": "The yellow screwdriver is on the second shelf of the rack to the left of the whiteboard in the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D4_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_009-mcq_fixed_3", + "qid_num": 23, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the yellow screwdriver on the table taken from?", + "answer": "The yellow screwdriver is now on the table in the yard.", + "question_chinese": "黄色螺丝刀现在在哪里？", + "answer_chinese": "黄色螺丝刀现在在院子的桌上。", + "evidence": [ + { + "description": "The yellow screwdriver is on the second shelf of the rack to the right of the whiteboard in the living room.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:02" + } + }, + { + "description": "The yellow screwdriver is now on the table in the yard.", + "timestamps": { + "start": "00:06:03", + "end": "00:06:04" + } + } + ], + "question_timestamps": "00:07:38", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From behind the table.", + "B": "The yellow screwdriver is now on the table in the yard.", + "C": "From under the table.", + "D": "It was taken from the garage workbench where it was used to fix the lawnmower.", + "E": "You picked it up near the kitchen sink after repairing the leaky faucet earlier." + }, + "answer_label": "B", + "answer_text": "The yellow screwdriver is now on the table in the yard.", + "q_id": "egolife_A1_D4_009-mcq_fixed_3", + "correct_answer": "The yellow screwdriver is now on the table in the yard.", + "original_options": { + "A": "From behind the table.", + "B": "The yellow screwdriver is now on the table in the yard.", + "C": "From under the table.", + "D": "It was taken from the garage workbench where it was used to fix the lawnmower.", + "E": "You picked it up near the kitchen sink after repairing the leaky faucet earlier." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D4_010-mcq.json b/QAs/egolife_A1_D4_010-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2057e624afa59ca1bba02a1afe0904a84781961f --- /dev/null +++ b/QAs/egolife_A1_D4_010-mcq.json @@ -0,0 +1,752 @@ +[ + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010-mcq_fixed_6", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the cutting board on the second floor placed?", + "answer": "The cutting board on the second floor is placed on the tabletop to the left of the sink, leaning against the glass.", + "question_chinese": "二楼的切菜板放在了哪里？", + "answer_chinese": "二楼的切菜板放到了水槽左边的桌面上，靠着玻璃立着。", + "evidence": [ + { + "description": "Right now, you can see the cutting board is placed on the tabletop to the left of the sink, leaning against the glass.", + "timestamps": { + "start": "00:01:52", + "end": "00:01:53" + } + }, + { + "description": "Right now, you can see the cutting board is placed on the tabletop to the left of the sink, leaning against the glass.", + "timestamps": { + "start": "00:02:06", + "end": "00:02:07" + } + }, + { + "description": "Right now, you can see the cutting board is placed on the tabletop to the left of the sink, leaning against the glass.", + "timestamps": { + "start": "00:03:16", + "end": "00:03:17" + } + } + ], + "question_timestamps": "00:05:41", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "On the tabletop right of the sink.", + "B": "Leaning against the fruit bowl.", + "C": "The cutting board on the second floor is placed on the tabletop to the right of the sink, leaning against the vase.", + "D": "The cutting board on the second floor is placed on the tabletop to the left of the sink, leaning against the glass.", + "E": "Positioned upright against the backsplash between the stove and the bread box on the second floor." + }, + "answer_label": "D", + "answer_text": "The cutting board on the second floor is placed on the tabletop to the left of the sink, leaning against the glass.", + "q_id": "egolife_A1_D4_010-mcq_fixed_6", + "correct_answer": "The cutting board on the second floor is placed on the tabletop to the left of the sink, leaning against the glass.", + "original_options": { + "A": "On the tabletop right of the sink.", + "B": "Leaning against the fruit bowl.", + "C": "The cutting board on the second floor is placed on the tabletop to the right of the sink, leaning against the vase.", + "D": "The cutting board on the second floor is placed on the tabletop to the left of the sink, leaning against the glass.", + "E": "Positioned upright against the backsplash between the stove and the bread box on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 22, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010-mcq_fixed_7", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the cutting board on the second floor placed?", + "answer": "The cutting board on the second floor is placed on the tabletop to the left of the sink.", + "question_chinese": "二楼的切菜板放在了哪里？", + "answer_chinese": "二楼的切菜板放到了水槽左边的桌面上。", + "evidence": [ + { + "description": "Right now, you can see the cutting board is placed on the tabletop to the left of the sink, leaning against the glass.", + "timestamps": { + "start": "00:01:52", + "end": "00:01:53" + } + }, + { + "description": "Right now, you can see the cutting board is placed on the tabletop to the left of the sink, leaning against the glass.", + "timestamps": { + "start": "00:02:06", + "end": "00:02:07" + } + }, + { + "description": "Right now, you can see the cutting board is placed on the tabletop to the left of the sink, leaning against the glass.", + "timestamps": { + "start": "00:03:16", + "end": "00:03:17" + } + } + ], + "question_timestamps": "00:09:08", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "On the tabletop to the right.", + "B": "On the kitchen island center.", + "C": "The cutting board on the second floor is placed on the tabletop with a blue placemat, not specifically left of the sink.", + "D": "The cutting board on the second floor is placed on the tabletop to the left of the sink.", + "E": "The cutting board on the second floor is placed on the tabletop with a red bowl, rather than left of the sink." + }, + "answer_label": "D", + "answer_text": "The cutting board on the second floor is placed on the tabletop to the left of the sink.", + "q_id": "egolife_A1_D4_010-mcq_fixed_7", + "correct_answer": "The cutting board on the second floor is placed on the tabletop to the left of the sink.", + "original_options": { + "A": "On the tabletop to the right.", + "B": "On the kitchen island center.", + "C": "The cutting board on the second floor is placed on the tabletop with a blue placemat, not specifically left of the sink.", + "D": "The cutting board on the second floor is placed on the tabletop to the left of the sink.", + "E": "The cutting board on the second floor is placed on the tabletop with a red bowl, rather than left of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 22, + 22 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010-mcq_fixed_8", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I sit on the small sofa next to the kitchen, can I reach the range hood?", + "answer": "You cannot reach the range hood directly.", + "question_chinese": "当我坐在厨房旁边的小沙发上时，我能够够到油烟机吗？", + "answer_chinese": "你不能直接够到油烟机。", + "evidence": [ + { + "description": "The range hood is at the left end of the sink.", + "timestamps": { + "start": "00:02:57", + "end": "00:02:58" + } + }, + { + "description": "The sofa is to the right of the sink, and it's about five meters away from the range hood, so you can't reach it.", + "timestamps": { + "start": "00:08:02", + "end": "00:08:03" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Yes, you can reach it.", + "B": "You can touch it by stretching.", + "C": "You can reach the range hood if you lean forward on the sofa and fully extend your arm.", + "D": "From the small sofa by the kitchen, you can just barely reach the range hood with a long stretch.", + "E": "You cannot reach the range hood directly." + }, + "answer_label": "E", + "answer_text": "You cannot reach the range hood directly.", + "q_id": "egolife_A1_D4_010-mcq_fixed_8", + "correct_answer": "You cannot reach the range hood directly.", + "original_options": { + "A": "Yes, you can reach it.", + "B": "You can touch it by stretching.", + "C": "You can reach the range hood if you lean forward on the sofa and fully extend your arm.", + "D": "From the small sofa by the kitchen, you can just barely reach the range hood with a long stretch.", + "E": "You cannot reach the range hood directly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010-mcq_fixed_9", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I sit on the small sofa next to the kitchen, can I reach the range hood?", + "answer": "You can not.", + "question_chinese": "当我坐在厨房旁边的小沙发上时，我能够够到油烟机吗？", + "answer_chinese": "你不能直接够到油烟机。", + "evidence": [ + { + "description": "The range hood is at the left end of the sink.", + "timestamps": { + "start": "00:02:57", + "end": "00:02:58" + } + }, + { + "description": "The sofa is to the right of the sink, and it's about five meters away from the range hood, so you can't reach it.", + "timestamps": { + "start": "00:08:02", + "end": "00:08:03" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can not.", + "B": "You can." + }, + "answer_label": "A", + "answer_text": "You can not.", + "q_id": "egolife_A1_D4_010-mcq_fixed_9", + "correct_answer": "You can not.", + "original_options": { + "A": "You can not.", + "B": "You can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010-mcq_fixed_2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the coffee I'm drinking right now?", + "answer": "The coffee you are drinking now was taken from the left side of the table in the kitchen on the second floor.", + "question_chinese": "我此时喝的咖啡是从哪里拿到的？", + "answer_chinese": "你此时喝的咖啡是从二楼厨房左侧桌面上拿到的。", + "evidence": [ + { + "description": "Right now, you pick up the prepared coffee from the kitchen tabletop.", + "timestamps": { + "start": "00:11:07", + "end": "00:11:08" + } + } + ], + "question_timestamps": "00:11:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From the right side of the table.", + "B": "From the first-floor kitchen table.", + "C": "You picked up the coffee from the right side of the table in the second-floor kitchen.", + "D": "The coffee you are drinking now was taken from the left side of the table in the kitchen on the second floor.", + "E": "You grabbed the coffee from the left side of the table in the kitchen on the first floor." + }, + "answer_label": "D", + "answer_text": "The coffee you are drinking now was taken from the left side of the table in the kitchen on the second floor.", + "q_id": "egolife_A1_D4_010-mcq_fixed_2", + "correct_answer": "The coffee you are drinking now was taken from the left side of the table in the kitchen on the second floor.", + "original_options": { + "A": "From the right side of the table.", + "B": "From the first-floor kitchen table.", + "C": "You picked up the coffee from the right side of the table in the second-floor kitchen.", + "D": "The coffee you are drinking now was taken from the left side of the table in the kitchen on the second floor.", + "E": "You grabbed the coffee from the left side of the table in the kitchen on the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010-mcq_fixed_3", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I get the coffee I'm drinking right now?", + "answer": "The coffee was taken from the left side of the table in the kitchen on the second floor.", + "question_chinese": "我此时喝的咖啡是从哪里拿到的？", + "answer_chinese": "咖啡是从二楼厨房左侧桌面上拿到的。", + "evidence": [ + { + "description": "Right now, you pick up the prepared coffee from the kitchen tabletop.", + "timestamps": { + "start": "00:11:07", + "end": "00:11:08" + } + } + ], + "question_timestamps": "00:11:59", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The coffee was taken from the left side of the table in the kitchen on the second floor.", + "B": "From the right side of the table.", + "C": "From the first-floor kitchen table.", + "D": "You took the coffee from the right side of the table in the kitchen on the second floor.", + "E": "You got the coffee from the left side of the table in the kitchen on the first floor." + }, + "answer_label": "A", + "answer_text": "The coffee was taken from the left side of the table in the kitchen on the second floor.", + "q_id": "egolife_A1_D4_010-mcq_fixed_3", + "correct_answer": "The coffee was taken from the left side of the table in the kitchen on the second floor.", + "original_options": { + "A": "The coffee was taken from the left side of the table in the kitchen on the second floor.", + "B": "From the right side of the table.", + "C": "From the first-floor kitchen table.", + "D": "You took the coffee from the right side of the table in the kitchen on the second floor.", + "E": "You got the coffee from the left side of the table in the kitchen on the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010-mcq_fixed_0", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "At this moment, where is the sink in relation to me?", + "answer": "The sink is on your right side.", + "question_chinese": "此时水槽在我的哪个位置？", + "answer_chinese": "水槽在你的右手边。", + "evidence": [ + { + "description": "The sink is on the boy's right side. You are standing facing the boy, so the sink is on your right side.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:48" + } + } + ], + "question_timestamps": "00:01:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The sink is on your left.", + "B": "The sink is behind you.", + "C": "The sink is on your right side.", + "D": "The sink is slightly behind you on the left side, near the corner of the counter.", + "E": "The sink is behind your right shoulder, positioned along the counter near the cabinets." + }, + "answer_label": "C", + "answer_text": "The sink is on your right side.", + "q_id": "egolife_A1_D4_010-mcq_fixed_0", + "correct_answer": "The sink is on your right side.", + "original_options": { + "A": "The sink is on your left.", + "B": "The sink is behind you.", + "C": "The sink is on your right side.", + "D": "The sink is slightly behind you on the left side, near the corner of the counter.", + "E": "The sink is behind your right shoulder, positioned along the counter near the cabinets." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010-mcq_fixed_1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "At this moment, where is the sink in relation to me?", + "answer": "The sink is right in front of you.", + "question_chinese": "此时水槽在我的哪个位置？", + "answer_chinese": "水槽在你正前面。", + "evidence": [ + { + "description": "The sink is on the boy's right side. You are standing facing the boy, so the sink is on your right side.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:48" + } + }, + { + "description": "The sink is right in front of you.", + "timestamps": { + "start": "00:04:42", + "end": "00:04:43" + } + } + ], + "question_timestamps": "00:04:44", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The sink is to your left.", + "B": "The sink is behind you.", + "C": "The sink is directly to your right side, lined up with the cabinets and counter edge.", + "D": "The sink is right in front of you.", + "E": "The sink is slightly behind you to the left, tucked into the counter near the corner." + }, + "answer_label": "D", + "answer_text": "The sink is right in front of you.", + "q_id": "egolife_A1_D4_010-mcq_fixed_1", + "correct_answer": "The sink is right in front of you.", + "original_options": { + "A": "The sink is to your left.", + "B": "The sink is behind you.", + "C": "The sink is directly to your right side, lined up with the cabinets and counter edge.", + "D": "The sink is right in front of you.", + "E": "The sink is slightly behind you to the left, tucked into the counter near the corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010-mcq_fixed_4", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kettles are there to the right of the sink?", + "answer": "Five.", + "question_chinese": "水槽右边有几个水壶？", + "answer_chinese": "有两个水壶。", + "evidence": [ + { + "description": "Right now, the boy takes a kettle from the right side of the sink, and there is still one kettle remaining on the right.", + "timestamps": { + "start": "00:04:42", + "end": "00:04:44" + } + }, + { + "description": "At this moment, the boy has returned the used kettle to its original place, so there are now two kettles on the right side of the sink.", + "timestamps": { + "start": "00:04:54", + "end": "00:04:56" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "There are three kettles positioned to the right of the sink on the counter.", + "E": "You can see seven kettles to the right of the sink, grouped together along the counter." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D4_010-mcq_fixed_4", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "There are three kettles positioned to the right of the sink on the counter.", + "E": "You can see seven kettles to the right of the sink, grouped together along the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010-mcq_fixed_5", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kettles are there to the right of the sink?", + "answer": "Six.", + "question_chinese": "水槽右边有几个水壶？", + "answer_chinese": "水槽右边有两个水壶。", + "evidence": [ + { + "description": "Right now, the boy takes a kettle from the right side of the sink, and there is still one kettle remaining on the right.", + "timestamps": { + "start": "00:04:42", + "end": "00:04:43" + } + }, + { + "description": "At this moment, the boy has returned the used kettle to its original place, so there are now two kettles on the right side of the sink.", + "timestamps": { + "start": "00:04:54", + "end": "00:04:55" + } + } + ], + "question_timestamps": "00:11:47", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "Five.", + "C": "Seven.", + "D": "There are four kettles placed to the right of the sink on the countertop.", + "E": "You can count eight kettles to the right of the sink, lined up next to each other." + }, + "answer_label": "A", + "answer_text": "Six.", + "q_id": "egolife_A1_D4_010-mcq_fixed_5", + "correct_answer": "Six.", + "original_options": { + "A": "Six.", + "B": "Five.", + "C": "Seven.", + "D": "There are four kettles placed to the right of the sink on the countertop.", + "E": "You can count eight kettles to the right of the sink, lined up next to each other." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010_5_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Up to now, how many people have I run into?", + "answer": "You ran into two people, and they were to your front-right.", + "question_chinese": "至今为止我遇到了几个人？", + "answer_chinese": "你遇到了两个人，他们在你的右前方", + "evidence": [ + { + "description": "You came across two people.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You ran into one person.", + "B": "You ran into three people.", + "C": "You ran into two people, and they were to your front-right.", + "D": "You ran into four people overall, and they were to your front-right when you encountered them.", + "E": "You ran into zero people up to now, so there wasn’t anyone positioned to your front-right." + }, + "answer_label": "C", + "answer_text": "You ran into two people, and they were to your front-right.", + "q_id": "egolife_A1_D4_010_5_T2", + "correct_answer": "You ran into two people, and they were to your front-right.", + "original_options": { + "A": "You ran into one person.", + "B": "You ran into three people.", + "C": "You ran into two people, and they were to your front-right.", + "D": "You ran into four people overall, and they were to your front-right when you encountered them.", + "E": "You ran into zero people up to now, so there wasn’t anyone positioned to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010_5_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Up to now, how many people have I run into?", + "answer": "You ran into three people, and they were directly in front of you.", + "question_chinese": "至今为止我遇到了几个人？", + "answer_chinese": "你遇到了三个人，他们在你的正前方", + "evidence": [ + { + "description": "You encountered three people.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You ran into two people.", + "B": "You ran into three people, and they were directly in front of you.", + "C": "You ran into four people.", + "D": "You ran into one person, and they were directly in front of you when you crossed paths.", + "E": "You ran into five people in total, and each time they were positioned directly in front of you." + }, + "answer_label": "B", + "answer_text": "You ran into three people, and they were directly in front of you.", + "q_id": "egolife_A1_D4_010_5_T1", + "correct_answer": "You ran into three people, and they were directly in front of you.", + "original_options": { + "A": "You ran into two people.", + "B": "You ran into three people, and they were directly in front of you.", + "C": "You ran into four people.", + "D": "You ran into one person, and they were directly in front of you when you crossed paths.", + "E": "You ran into five people in total, and each time they were positioned directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_010_5_T3", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Up to now, how many people have I run into?", + "answer": "You ran into four people, and they were directly in front of you.", + "question_chinese": "至今为止我遇到了几个人？", + "answer_chinese": "你遇到了四个人，他们在你的正前方", + "evidence": [ + { + "description": "You came across two people.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Three people downstairs and four people upstairs.", + "timestamps": { + "start": "00:01:34", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:09:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You ran into three people.", + "B": "You ran into five people.", + "C": "You ran into two people, and they were directly in front of you when you ran into them.", + "D": "You ran into four people, and they were directly in front of you.", + "E": "You ran into six people up to now, and they were directly in front of you at each encounter." + }, + "answer_label": "D", + "answer_text": "You ran into four people, and they were directly in front of you.", + "q_id": "egolife_A1_D4_010_5_T3", + "correct_answer": "You ran into four people, and they were directly in front of you.", + "original_options": { + "A": "You ran into three people.", + "B": "You ran into five people.", + "C": "You ran into two people, and they were directly in front of you when you ran into them.", + "D": "You ran into four people, and they were directly in front of you.", + "E": "You ran into six people up to now, and they were directly in front of you at each encounter." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 19 + ] + } +] diff --git a/QAs/egolife_A1_D4_011-mcq.json b/QAs/egolife_A1_D4_011-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..72d961fd9d29cc060b7b71410644b647333de051 --- /dev/null +++ b/QAs/egolife_A1_D4_011-mcq.json @@ -0,0 +1,588 @@ +[ + { + "video_uid": "egolife_A1_D4_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_011-mcq_fixed_4", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What's on the small sofa next to the kitchen on the second floor?", + "answer": "There is a hair dryer on the small sofa in the kitchen on the second floor.", + "question_chinese": "二楼厨房旁边的小沙发上放了什么?", + "answer_chinese": "二楼厨房的小沙发上放了一个吹风机。", + "evidence": [ + { + "description": "At this moment, you can see a hair dryer on the small sofa in the second-floor kitchen.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:04" + } + }, + { + "description": "At this moment, you can see a hair dryer on the small sofa in the second-floor kitchen.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A folded blanket.", + "B": "A stack of magazines.", + "C": "There is a hair dryer on the small sofa in the kitchen on the second floor.", + "D": "There is a book on the small sofa next to the kitchen on the second floor.", + "E": "There is a magazine on the small sofa beside the kitchen on the second floor." + }, + "answer_label": "C", + "answer_text": "There is a hair dryer on the small sofa in the kitchen on the second floor.", + "q_id": "egolife_A1_D4_011-mcq_fixed_4", + "correct_answer": "There is a hair dryer on the small sofa in the kitchen on the second floor.", + "original_options": { + "A": "A folded blanket.", + "B": "A stack of magazines.", + "C": "There is a hair dryer on the small sofa in the kitchen on the second floor.", + "D": "There is a book on the small sofa next to the kitchen on the second floor.", + "E": "There is a magazine on the small sofa beside the kitchen on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_011-mcq_fixed_5", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What's on the small sofa next to the kitchen on the second floor?", + "answer": "A hair dryer.", + "question_chinese": "二楼厨房旁边的小沙发上放了什么?", + "answer_chinese": "一个吹风机。", + "evidence": [ + { + "description": "At this moment, you can see a hair dryer on the small sofa in the second-floor kitchen.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:04" + } + }, + { + "description": "At this moment, you can see a hair dryer on the small sofa in the second-floor kitchen.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A book.", + "B": "A hair dryer.", + "C": "A guitar.", + "D": "A decorative porcelain vase with artificial flowers is sitting on the small sofa next to the kitchen.", + "E": "A small storage box filled with grooming accessories is on the small sofa by the upstairs kitchen." + }, + "answer_label": "B", + "answer_text": "A hair dryer.", + "q_id": "egolife_A1_D4_011-mcq_fixed_5", + "correct_answer": "A hair dryer.", + "original_options": { + "A": "A book.", + "B": "A hair dryer.", + "C": "A guitar.", + "D": "A decorative porcelain vase with artificial flowers is sitting on the small sofa next to the kitchen.", + "E": "A small storage box filled with grooming accessories is on the small sofa by the upstairs kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_011-mcq_fixed_0", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the takeout I got from outside finally placed?", + "answer": "The takeout was finally placed on the dining table in the living room on the second floor.", + "question_chinese": "我从外面拿到的外卖最后被放到了哪里？", + "answer_chinese": "外卖最后被放到了二楼客厅的餐桌上。", + "evidence": [ + { + "description": "At this moment, you placed the two takeout orders on the table.", + "timestamps": { + "start": "00:00:56", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:06:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the coffee table downstairs.", + "B": "In the refrigerator upstairs.", + "C": "The takeout was finally placed on the storage shelf next to the refrigerator in the kitchen area.", + "D": "The takeout was finally placed on the dining table in the living room on the second floor.", + "E": "The takeout was finally placed on the floor in front of the dining table in the second-floor living room." + }, + "answer_label": "D", + "answer_text": "The takeout was finally placed on the dining table in the living room on the second floor.", + "q_id": "egolife_A1_D4_011-mcq_fixed_0", + "correct_answer": "The takeout was finally placed on the dining table in the living room on the second floor.", + "original_options": { + "A": "On the coffee table downstairs.", + "B": "In the refrigerator upstairs.", + "C": "The takeout was finally placed on the storage shelf next to the refrigerator in the kitchen area.", + "D": "The takeout was finally placed on the dining table in the living room on the second floor.", + "E": "The takeout was finally placed on the floor in front of the dining table in the second-floor living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_011-mcq_fixed_1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the takeout I got from outside finally placed?", + "answer": "On the dining table in the living room on the second floor.", + "question_chinese": "我从外面拿到的外卖最后被放到了哪里？", + "answer_chinese": "二楼客厅的餐桌上。", + "evidence": [ + { + "description": "At this moment, you placed the two takeout orders on the table.", + "timestamps": { + "start": "00:00:56", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:09:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the dining table in the living room on the second floor.", + "B": "On the first-floor dining table.", + "C": "On the third-floor dining table.", + "D": "On the breakfast bar adjoining the open kitchen, where you set it down before moving it elsewhere.", + "E": "On the sideboard in the hallway by the stairs, where you placed it after coming in." + }, + "answer_label": "A", + "answer_text": "On the dining table in the living room on the second floor.", + "q_id": "egolife_A1_D4_011-mcq_fixed_1", + "correct_answer": "On the dining table in the living room on the second floor.", + "original_options": { + "A": "On the dining table in the living room on the second floor.", + "B": "On the first-floor dining table.", + "C": "On the third-floor dining table.", + "D": "On the breakfast bar adjoining the open kitchen, where you set it down before moving it elsewhere.", + "E": "On the sideboard in the hallway by the stairs, where you placed it after coming in." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_011-mcq_fixed_8", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many takeout orders did I just pick up?", + "answer": "Five.", + "question_chinese": "我刚刚拿了几个外卖？", + "answer_chinese": "你刚刚拿了两个外卖。", + "evidence": [ + { + "description": "At this moment, you can be seen holding a takeout bag in each hand.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + }, + { + "description": "At this moment, you placed the two takeout orders on the table.", + "timestamps": { + "start": "00:00:56", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:05:50", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "One.", + "C": "Two.", + "D": "Three orders in total, picked up together from the same place just before heading back.", + "E": "Four orders altogether, gathered at once and carried away immediately after collecting the takeout." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D4_011-mcq_fixed_8", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "One.", + "C": "Two.", + "D": "Three orders in total, picked up together from the same place just before heading back.", + "E": "Four orders altogether, gathered at once and carried away immediately after collecting the takeout." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_011-mcq_fixed_9", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many takeout orders did I just pick up?", + "answer": "One.", + "question_chinese": "我刚刚拿了几个外卖？", + "answer_chinese": "两个外卖。", + "evidence": [ + { + "description": "At this moment, you can be seen holding a takeout bag in each hand.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + }, + { + "description": "At this moment, you placed the two takeout orders on the table.", + "timestamps": { + "start": "00:00:56", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:08:32", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "You picked up four takeout orders in total, not just a single order.", + "E": "You just picked up five separate takeout orders, rather than only one." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "egolife_A1_D4_011-mcq_fixed_9", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "You picked up four takeout orders in total, not just a single order.", + "E": "You just picked up five separate takeout orders, rather than only one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_011-mcq_fixed_6", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go from the courtyard gate to the living room on the second floor, what path should I take?", + "answer": "You should go straight into the courtyard, turn right at the end, go straight and go upstairs. After going upstairs, turn left to reach the living room on the second floor.", + "question_chinese": "我如果要从院门口到二楼客厅应该走什么路径？", + "answer_chinese": "你应该进院直走到尽头右拐直走上楼，上楼后左转到了二楼客厅。", + "evidence": [ + { + "description": "At this moment, you enter the courtyard, walk straight to the end, and turn right.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "At this moment, you walk straight upstairs.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:32" + } + }, + { + "description": "At this moment, you turn left and enter the living room.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:49" + } + } + ], + "question_timestamps": "00:05:42", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left at the end.", + "B": "You should go straight into the courtyard, turn right at the end, go straight and go upstairs. After going upstairs, turn left to reach the living room on the second floor.", + "C": "Go upstairs, then turn right.", + "D": "Go straight into the courtyard, turn left at the end, then go upstairs and turn left.", + "E": "Enter the courtyard, turn right immediately, take the stairs up, then turn left to reach the living room." + }, + "answer_label": "B", + "answer_text": "You should go straight into the courtyard, turn right at the end, go straight and go upstairs. After going upstairs, turn left to reach the living room on the second floor.", + "q_id": "egolife_A1_D4_011-mcq_fixed_6", + "correct_answer": "You should go straight into the courtyard, turn right at the end, go straight and go upstairs. After going upstairs, turn left to reach the living room on the second floor.", + "original_options": { + "A": "Turn left at the end.", + "B": "You should go straight into the courtyard, turn right at the end, go straight and go upstairs. After going upstairs, turn left to reach the living room on the second floor.", + "C": "Go upstairs, then turn right.", + "D": "Go straight into the courtyard, turn left at the end, then go upstairs and turn left.", + "E": "Enter the courtyard, turn right immediately, take the stairs up, then turn left to reach the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_011-mcq_fixed_7", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go from the courtyard gate to the living room on the second floor, what path should I take?", + "answer": "Go straight into the courtyard, turn right at the end, go straight and go upstairs. After going upstairs, turn left to reach the living room on the second floor.", + "question_chinese": "我如果要从院门口到二楼客厅应该走什么路径？", + "answer_chinese": "进院直走到尽头右拐直走上楼，上楼后左转到了二楼客厅。", + "evidence": [ + { + "description": "At this moment, you enter the courtyard, walk straight to the end, and turn right.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "At this moment, you walk straight upstairs.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:32" + } + }, + { + "description": "At this moment, you turn left and enter the living room.", + "timestamps": { + "start": "00:00:47", + "end": "00:00:49" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left at the end.", + "B": "Go straight into the courtyard, turn right at the end, go straight and go upstairs. After going upstairs, turn left to reach the living room on the second floor.", + "C": "Turn right after upstairs.", + "D": "Go straight into the courtyard, turn right immediately, then go upstairs and turn left on the second floor.", + "E": "Enter the courtyard, turn right at the end, go upstairs, then turn right to reach the second-floor living room." + }, + "answer_label": "B", + "answer_text": "Go straight into the courtyard, turn right at the end, go straight and go upstairs. After going upstairs, turn left to reach the living room on the second floor.", + "q_id": "egolife_A1_D4_011-mcq_fixed_7", + "correct_answer": "Go straight into the courtyard, turn right at the end, go straight and go upstairs. After going upstairs, turn left to reach the living room on the second floor.", + "original_options": { + "A": "Turn left at the end.", + "B": "Go straight into the courtyard, turn right at the end, go straight and go upstairs. After going upstairs, turn left to reach the living room on the second floor.", + "C": "Turn right after upstairs.", + "D": "Go straight into the courtyard, turn right immediately, then go upstairs and turn left on the second floor.", + "E": "Enter the courtyard, turn right at the end, go upstairs, then turn right to reach the second-floor living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_011-mcq_fixed_2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I standing now?", + "answer": "You are currently on the right side of the living room table on the first floor, facing the kitchen.", + "question_chinese": "我现在站在哪里？", + "answer_chinese": "你现在在一楼客厅桌子右侧，面向厨房。", + "evidence": [ + { + "description": "At this moment, you can be seen standing on the right side of the living room table on the first floor, facing the kitchen.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:09" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left side of the living room table.", + "B": "You are currently on the right side of the living room table on the first floor, facing the kitchen.", + "C": "Right side of the table, facing the bookshelf.", + "D": "You are in front of the living room table on the first floor, facing toward the staircase.", + "E": "You are near the television area in the living room, positioned away from the table and facing the dining side." + }, + "answer_label": "B", + "answer_text": "You are currently on the right side of the living room table on the first floor, facing the kitchen.", + "q_id": "egolife_A1_D4_011-mcq_fixed_2", + "correct_answer": "You are currently on the right side of the living room table on the first floor, facing the kitchen.", + "original_options": { + "A": "Left side of the living room table.", + "B": "You are currently on the right side of the living room table on the first floor, facing the kitchen.", + "C": "Right side of the table, facing the bookshelf.", + "D": "You are in front of the living room table on the first floor, facing toward the staircase.", + "E": "You are near the television area in the living room, positioned away from the table and facing the dining side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_011", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_011-mcq_fixed_3", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I standing now?", + "answer": "You are currently about one meter to the right of the living room table on the first floor, facing the kitchen.", + "question_chinese": "我现在站在哪里？", + "answer_chinese": "你现在在一楼客厅桌子右侧一米左右，面向厨房。", + "evidence": [ + { + "description": "At this moment, you can be seen standing on the right side of the living room table on the first floor, facing the kitchen.", + "timestamps": { + "start": "00:07:14", + "end": "00:07:15" + } + } + ], + "question_timestamps": "00:07:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "One meter left of the table.", + "B": "Right of the table, facing the living room.", + "C": "You are beside the couch along the left wall, standing diagonally across from the living room table.", + "D": "You are near the bookshelf a few steps from the table, facing toward the hallway rather than the kitchen.", + "E": "You are currently about one meter to the right of the living room table on the first floor, facing the kitchen." + }, + "answer_label": "E", + "answer_text": "You are currently about one meter to the right of the living room table on the first floor, facing the kitchen.", + "q_id": "egolife_A1_D4_011-mcq_fixed_3", + "correct_answer": "You are currently about one meter to the right of the living room table on the first floor, facing the kitchen.", + "original_options": { + "A": "One meter left of the table.", + "B": "Right of the table, facing the living room.", + "C": "You are beside the couch along the left wall, standing diagonally across from the living room table.", + "D": "You are near the bookshelf a few steps from the table, facing toward the hallway rather than the kitchen.", + "E": "You are currently about one meter to the right of the living room table on the first floor, facing the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 17, + 19 + ] + } +] diff --git a/QAs/egolife_A1_D4_012-mcq.json b/QAs/egolife_A1_D4_012-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9ca6608128e7e3c4206cd7e83d671ed4cc4e9e07 --- /dev/null +++ b/QAs/egolife_A1_D4_012-mcq.json @@ -0,0 +1,1165 @@ +[ + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012-mcq_fixed_2", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Who was closer to me the last time I stood still?", + "answer": "The last time You stood still, the girl in pink was closer to you.", + "question_chinese": "我上一次站着不动的时候谁离我更近？", + "answer_chinese": "我上一次站着不动的时候粉衣服的女生离我更近", + "evidence": [ + { + "description": "The girl in pink is closer to me", + "timestamps": { + "start": "00:02:00", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The tall man was closer.", + "B": "The boy in the blue jacket was closer.", + "C": "The last time You stood still, the girl in pink was closer to you.", + "D": "The last time you stood still, the girl in pink was farther away than the other person nearby.", + "E": "The last time you stood still, the nearest person to you was the man with sunglasses and a backpack." + }, + "answer_label": "C", + "answer_text": "The last time You stood still, the girl in pink was closer to you.", + "q_id": "egolife_A1_D4_012-mcq_fixed_2", + "correct_answer": "The last time You stood still, the girl in pink was closer to you.", + "original_options": { + "A": "The tall man was closer.", + "B": "The boy in the blue jacket was closer.", + "C": "The last time You stood still, the girl in pink was closer to you.", + "D": "The last time you stood still, the girl in pink was farther away than the other person nearby.", + "E": "The last time you stood still, the nearest person to you was the man with sunglasses and a backpack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_7_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "The last time I was standing still, who was closer to me?", + "answer": "The last time you were standing still, the girl in pink was closer to you. She was in front of you.", + "question_chinese": "我上一次站着不动的时候谁离我更近？", + "answer_chinese": "你上一次站着不动的时候粉衣服的女生离我更近，她在你前面。", + "evidence": [ + { + "description": "The girl in pink is closer to me", + "timestamps": { + "start": "00:02:00", + "end": "00:02:06" + } + } + ], + "question_timestamps": "00:02:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "She was behind you.", + "B": "The last time you were standing still, the girl in pink was closer to you. She was in front of you.", + "C": "She was to your left.", + "D": "The last time you were standing still, the girl in pink was closer to you, but she was to your right.", + "E": "The last time you were standing still, the girl in pink was closer to you, positioned behind you rather than ahead." + }, + "answer_label": "B", + "answer_text": "The last time you were standing still, the girl in pink was closer to you. She was in front of you.", + "q_id": "egolife_A1_D4_012_7_T2", + "correct_answer": "The last time you were standing still, the girl in pink was closer to you. She was in front of you.", + "original_options": { + "A": "She was behind you.", + "B": "The last time you were standing still, the girl in pink was closer to you. She was in front of you.", + "C": "She was to your left.", + "D": "The last time you were standing still, the girl in pink was closer to you, but she was to your right.", + "E": "The last time you were standing still, the girl in pink was closer to you, positioned behind you rather than ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_7_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "The last time I was standing still, who was closer to me?", + "answer": "The last time you were standing still, the girl in blue was closer to you. She was in front of you.", + "question_chinese": "我上一次站着不动的时候谁离我更近？", + "answer_chinese": "你上一次站着不动的时候蓝衣服的女生离我更近，她在你前面。", + "evidence": [ + { + "description": "The girl in blue is closer to me", + "timestamps": { + "start": "00:02:37", + "end": "00:02:42" + } + } + ], + "question_timestamps": "00:03:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "She was behind you.", + "B": "She was to your left.", + "C": "The last time you were standing still, the girl in blue was closer to you, but she was to your right.", + "D": "The last time you were standing still, the girl in blue was closer to you. She was in front of you.", + "E": "The last time you were standing still, the girl in blue was closer to you, standing behind you instead of in front." + }, + "answer_label": "D", + "answer_text": "The last time you were standing still, the girl in blue was closer to you. She was in front of you.", + "q_id": "egolife_A1_D4_012_7_T1", + "correct_answer": "The last time you were standing still, the girl in blue was closer to you. She was in front of you.", + "original_options": { + "A": "She was behind you.", + "B": "She was to your left.", + "C": "The last time you were standing still, the girl in blue was closer to you, but she was to your right.", + "D": "The last time you were standing still, the girl in blue was closer to you. She was in front of you.", + "E": "The last time you were standing still, the girl in blue was closer to you, standing behind you instead of in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 21, + 22 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Who was closest to me the last time I stood still?", + "answer": "The last time I stood still, the girl in blue was closer to me.", + "question_chinese": "我上一次站着不动的时候谁离我更近？", + "answer_chinese": "我上一次站着不动的时候蓝衣服女生离我更近", + "evidence": [ + { + "description": "The girl in blue is closer to me", + "timestamps": { + "start": "00:02:25", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The last time I stood still, the girl in blue was closer to me.", + "B": "The last time I stood still, the girl in blue was not closer to me." + }, + "answer_label": "A", + "answer_text": "The last time I stood still, the girl in blue was closer to me.", + "q_id": "egolife_A1_D4_012_2", + "correct_answer": "The last time I stood still, the girl in blue was closer to me.", + "original_options": { + "A": "The last time I stood still, the girl in blue was closer to me.", + "B": "The last time I stood still, the girl in blue was not closer to me." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_8", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many laptops are there where I am?", + "answer": "There is a laptop where I am.", + "question_chinese": "我在的地方有几个笔记本电脑？", + "answer_chinese": "我在的地方有1个笔记本电脑", + "evidence": [ + { + "description": "Here is a laptop", + "timestamps": { + "start": "00:01:58", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:04:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two laptops here.", + "B": "There are three laptops here.", + "C": "There are four laptops where I am right now, spread out on different surfaces around me.", + "D": "There is a laptop where I am.", + "E": "There are five laptops in the area where I am, so multiple computers are nearby." + }, + "answer_label": "D", + "answer_text": "There is a laptop where I am.", + "q_id": "egolife_A1_D4_012_8", + "correct_answer": "There is a laptop where I am.", + "original_options": { + "A": "There are two laptops here.", + "B": "There are three laptops here.", + "C": "There are four laptops where I am right now, spread out on different surfaces around me.", + "D": "There is a laptop where I am.", + "E": "There are five laptops in the area where I am, so multiple computers are nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_9", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many laptops are there where I am?", + "answer": "There are seven laptops here.", + "question_chinese": "我在的地方有几个笔记本电脑？", + "answer_chinese": "这有7个笔记本电脑", + "evidence": [ + { + "description": "There are seven laptops here", + "timestamps": { + "start": "00:01:49", + "end": "00:01:50" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are five laptops here.", + "B": "There are seven laptops here.", + "C": "There are six laptops here.", + "D": "There are eight laptops in the area where I am, with some sitting close together nearby.", + "E": "There are nine laptops here where I am right now, making the space feel crowded with devices." + }, + "answer_label": "B", + "answer_text": "There are seven laptops here.", + "q_id": "egolife_A1_D4_012_9", + "correct_answer": "There are seven laptops here.", + "original_options": { + "A": "There are five laptops here.", + "B": "There are seven laptops here.", + "C": "There are six laptops here.", + "D": "There are eight laptops in the area where I am, with some sitting close together nearby.", + "E": "There are nine laptops here where I am right now, making the space feel crowded with devices." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_6_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the living room from where I am right now?", + "answer": "Turn left and walk straight to the living room. you're inside the living room.", + "question_chinese": "我现在去客厅怎么走？", + "answer_chinese": "你左转直走到客厅，我在客厅里面。", + "evidence": [ + { + "description": "Turn left and face the door.", + "timestamps": { + "start": "00:01:16", + "end": "00:01:23" + } + }, + { + "description": "Then go straight to the living room.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:01:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight.", + "B": "Turn around and walk straight.", + "C": "Turn left and walk straight to the living room. you're inside the living room.", + "D": "Turn left and walk straight to the living room, but you end up outside it.", + "E": "Turn right and walk straight to the living room until you're inside the living room." + }, + "answer_label": "C", + "answer_text": "Turn left and walk straight to the living room. you're inside the living room.", + "q_id": "egolife_A1_D4_012_6_T2", + "correct_answer": "Turn left and walk straight to the living room. you're inside the living room.", + "original_options": { + "A": "Turn right, then go straight.", + "B": "Turn around and walk straight.", + "C": "Turn left and walk straight to the living room. you're inside the living room.", + "D": "Turn left and walk straight to the living room, but you end up outside it.", + "E": "Turn right and walk straight to the living room until you're inside the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_6_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the living room from where I am right now?", + "answer": "Turn around and walk straight to the living room. you're inside the living room.", + "question_chinese": "我现在去客厅怎么走？", + "answer_chinese": "你向后转弯直走到客厅，我在客厅里面。", + "evidence": [ + { + "description": "Turn around to face the door.", + "timestamps": { + "start": "00:04:50", + "end": "00:04:53" + } + }, + { + "description": "Then go straight to the living room", + "timestamps": { + "start": "00:04:54", + "end": "00:05:07" + } + } + ], + "question_timestamps": "00:05:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left and walk straight.", + "B": "Turn around and walk straight to the living room. you're inside the living room.", + "C": "Turn right and walk straight.", + "D": "Turn around and walk straight to the living room, but you end up outside it.", + "E": "Turn left and keep walking straight until you reach the living room and step inside." + }, + "answer_label": "B", + "answer_text": "Turn around and walk straight to the living room. you're inside the living room.", + "q_id": "egolife_A1_D4_012_6_T1", + "correct_answer": "Turn around and walk straight to the living room. you're inside the living room.", + "original_options": { + "A": "Turn left and walk straight.", + "B": "Turn around and walk straight to the living room. you're inside the living room.", + "C": "Turn right and walk straight.", + "D": "Turn around and walk straight to the living room, but you end up outside it.", + "E": "Turn left and keep walking straight until you reach the living room and step inside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_2_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in the room I’m in right now?", + "answer": "There are five people in the room you're in right now, and four of them are in front of you.", + "question_chinese": "我现在在的这个房间里有几个人？", + "answer_chinese": "你现在在的这个房间里有5个人，四个人在你的前面。", + "evidence": [ + { + "description": "There are four people in my field of vision", + "timestamps": { + "start": "00:02:55", + "end": "00:03:00" + } + }, + { + "description": "I'm in the room", + "timestamps": { + "start": "00:02:37", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are five people in the room you're in right now, and four of them are in front of you.", + "B": "There are three people here.", + "C": "There are four people here.", + "D": "There are six people in the room you're in right now, and four of them are in front of you.", + "E": "There are seven people in the room you're in right now, and five of them are in front of you." + }, + "answer_label": "A", + "answer_text": "There are five people in the room you're in right now, and four of them are in front of you.", + "q_id": "egolife_A1_D4_012_2_T1", + "correct_answer": "There are five people in the room you're in right now, and four of them are in front of you.", + "original_options": { + "A": "There are five people in the room you're in right now, and four of them are in front of you.", + "B": "There are three people here.", + "C": "There are four people here.", + "D": "There are six people in the room you're in right now, and four of them are in front of you.", + "E": "There are seven people in the room you're in right now, and five of them are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_2_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in the room I’m in right now?", + "answer": "There are two people in the room you're in right now, and one of them is to your right.", + "question_chinese": "我现在在的这个房间里有几个人？", + "answer_chinese": "你现在在的这个房间里有两个人，其中一个人在我的右边。", + "evidence": [ + { + "description": "The boy in the black coat enters the room.", + "timestamps": { + "start": "00:04:13", + "end": "00:04:18" + } + }, + { + "description": "I'm in the room", + "timestamps": { + "start": "00:04:00", + "end": "00:04:10" + } + } + ], + "question_timestamps": "00:04:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There is one person here.", + "B": "There are two people in the room you're in right now, and one of them is to your right.", + "C": "There are three people here.", + "D": "There are four people in the room you're in right now, and one of them is to your right.", + "E": "There are zero people in the room you're in right now, so nobody is to your right." + }, + "answer_label": "B", + "answer_text": "There are two people in the room you're in right now, and one of them is to your right.", + "q_id": "egolife_A1_D4_012_2_T2", + "correct_answer": "There are two people in the room you're in right now, and one of them is to your right.", + "original_options": { + "A": "There is one person here.", + "B": "There are two people in the room you're in right now, and one of them is to your right.", + "C": "There are three people here.", + "D": "There are four people in the room you're in right now, and one of them is to your right.", + "E": "There are zero people in the room you're in right now, so nobody is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_3_T4", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where are the stairs relative to me?", + "answer": "The mirror is in front of you.", + "question_chinese": "楼梯在我的哪里？", + "answer_chinese": "镜子在你的前面", + "evidence": [ + { + "description": "The mirror is in front of you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The stairs are behind you.", + "B": "The stairs are to your left.", + "C": "The mirror is in front of you.", + "D": "The stairs are to your right as you stand where you are right now in the room.", + "E": "The stairs are above you rather than in front of you, so you must look upward." + }, + "answer_label": "C", + "answer_text": "The mirror is in front of you.", + "q_id": "egolife_A1_D4_012_3_T4", + "correct_answer": "The mirror is in front of you.", + "original_options": { + "A": "The stairs are behind you.", + "B": "The stairs are to your left.", + "C": "The mirror is in front of you.", + "D": "The stairs are to your right as you stand where you are right now in the room.", + "E": "The stairs are above you rather than in front of you, so you must look upward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_3_T3", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where are the stairs relative to me?", + "answer": "The mirror is behind you.", + "question_chinese": "楼梯在我的哪里？", + "answer_chinese": "镜子在你的后面", + "evidence": [ + { + "description": "The mirror is behind you.", + "timestamps": { + "start": "00:00:56", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:01:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The mirror is in front of you.", + "B": "The mirror is to your left.", + "C": "The mirror is behind you.", + "D": "The mirror is to your right, not behind you at all here.", + "E": "The mirror is above you, so you have to look upward to see it clearly." + }, + "answer_label": "C", + "answer_text": "The mirror is behind you.", + "q_id": "egolife_A1_D4_012_3_T3", + "correct_answer": "The mirror is behind you.", + "original_options": { + "A": "The mirror is in front of you.", + "B": "The mirror is to your left.", + "C": "The mirror is behind you.", + "D": "The mirror is to your right, not behind you at all here.", + "E": "The mirror is above you, so you have to look upward to see it clearly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_3_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where are the stairs relative to me?", + "answer": "The stairs are in front of you.", + "question_chinese": "楼梯在我的哪里？", + "answer_chinese": "楼梯在你的前面", + "evidence": [ + { + "description": "You have a black box in your hand.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The stairs are behind you.", + "B": "The stairs are to your left.", + "C": "The stairs are to your right, rather than directly in front of you.", + "D": "The stairs are above you, so you would need to look up to spot them.", + "E": "The stairs are in front of you." + }, + "answer_label": "E", + "answer_text": "The stairs are in front of you.", + "q_id": "egolife_A1_D4_012_3_T2", + "correct_answer": "The stairs are in front of you.", + "original_options": { + "A": "The stairs are behind you.", + "B": "The stairs are to your left.", + "C": "The stairs are to your right, rather than directly in front of you.", + "D": "The stairs are above you, so you would need to look up to spot them.", + "E": "The stairs are in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_3_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where are the stairs relative to me?", + "answer": "The stairs are to your left.", + "question_chinese": "楼梯在我的哪里？", + "answer_chinese": "楼梯在你的左边", + "evidence": [ + { + "description": "The stairs are on your left.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The stairs are in front of you.", + "B": "The stairs are behind you.", + "C": "The stairs are to your right, not positioned on your left side.", + "D": "The stairs are above you, meaning they are higher up rather than beside you.", + "E": "The stairs are to your left." + }, + "answer_label": "E", + "answer_text": "The stairs are to your left.", + "q_id": "egolife_A1_D4_012_3_T1", + "correct_answer": "The stairs are to your left.", + "original_options": { + "A": "The stairs are in front of you.", + "B": "The stairs are behind you.", + "C": "The stairs are to your right, not positioned on your left side.", + "D": "The stairs are above you, meaning they are higher up rather than beside you.", + "E": "The stairs are to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_3", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in this room right now?", + "answer": "There are five people in this room now.", + "question_chinese": "现在这个房间里有几个人？", + "answer_chinese": "5个人", + "evidence": [ + { + "description": "There are four people in my field of vision", + "timestamps": { + "start": "00:02:55", + "end": "00:03:00" + } + }, + { + "description": "I'm in the room", + "timestamps": { + "start": "00:02:33", + "end": "00:02:42" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are five people in this room now.", + "B": "There are four people now.", + "C": "There are six people now.", + "D": "There are three people in this room now, not five as stated.", + "E": "There are seven people in this room now, which is more than five." + }, + "answer_label": "A", + "answer_text": "There are five people in this room now.", + "q_id": "egolife_A1_D4_012_3", + "correct_answer": "There are five people in this room now.", + "original_options": { + "A": "There are five people in this room now.", + "B": "There are four people now.", + "C": "There are six people now.", + "D": "There are three people in this room now, not five as stated.", + "E": "There are seven people in this room now, which is more than five." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_4", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are there in this room now?", + "answer": "There are six people in this room now.", + "question_chinese": "现在这个房间里有几个人？", + "answer_chinese": "现在这个房间里有六个人", + "evidence": [ + { + "description": "There are four people in my field of vision", + "timestamps": { + "start": "00:02:55", + "end": "00:03:00" + } + }, + { + "description": "I'm in the room", + "timestamps": { + "start": "00:02:33", + "end": "00:02:42" + } + }, + { + "description": "A person came into the room", + "timestamps": { + "start": "00:03:18", + "end": "00:03:21" + } + } + ], + "question_timestamps": "00:03:52", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are five people now.", + "B": "There are seven people now.", + "C": "There are four people in this room now, not a total of six.", + "D": "There are eight people in this room now, which is more than six.", + "E": "There are six people in this room now." + }, + "answer_label": "E", + "answer_text": "There are six people in this room now.", + "q_id": "egolife_A1_D4_012_4", + "correct_answer": "There are six people in this room now.", + "original_options": { + "A": "There are five people now.", + "B": "There are seven people now.", + "C": "There are four people in this room now, not a total of six.", + "D": "There are eight people in this room now, which is more than six.", + "E": "There are six people in this room now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_6", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I passed through the living room door so far?", + "answer": "So far, I have passed by the living room door twice.", + "question_chinese": "目前为止我经过过几次客厅门？", + "answer_chinese": "目前为止我经过过2次客厅门", + "evidence": [ + { + "description": "I'm passing by the living room door", + "timestamps": { + "start": "00:00:37", + "end": "00:00:44" + } + }, + { + "description": "I'm passing by the living room door", + "timestamps": { + "start": "00:00:47", + "end": "00:00:51" + } + } + ], + "question_timestamps": "00:01:01", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "I passed by it once.", + "B": "I passed by it three times.", + "C": "So far, I have passed by the living room door twice.", + "D": "So far, I have passed by the living room door four times in total.", + "E": "So far, I have passed by the living room door five times altogether." + }, + "answer_label": "C", + "answer_text": "So far, I have passed by the living room door twice.", + "q_id": "egolife_A1_D4_012_6", + "correct_answer": "So far, I have passed by the living room door twice.", + "original_options": { + "A": "I passed by it once.", + "B": "I passed by it three times.", + "C": "So far, I have passed by the living room door twice.", + "D": "So far, I have passed by the living room door four times in total.", + "E": "So far, I have passed by the living room door five times altogether." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_7", + "qid_num": 18, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I passed through the living room door so far?", + "answer": "So far, I have passed by the living room door three times.", + "question_chinese": "目前为止我经过过几次客厅门？", + "answer_chinese": "目前为止我经过过3次客厅门", + "evidence": [ + { + "description": "I'm passing by the living room door", + "timestamps": { + "start": "00:00:37", + "end": "00:00:44" + } + }, + { + "description": "I'm passing by the living room door", + "timestamps": { + "start": "00:00:47", + "end": "00:00:51" + } + }, + { + "description": "I'm passing by the living room door", + "timestamps": { + "start": "00:01:00", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:01:23", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I passed by it twice.", + "B": "I passed by it four times.", + "C": "So far, I have passed by the living room door five times, not three.", + "D": "So far, I have passed by the living room door three times.", + "E": "So far, I have passed by the living room door six times in total." + }, + "answer_label": "D", + "answer_text": "So far, I have passed by the living room door three times.", + "q_id": "egolife_A1_D4_012_7", + "correct_answer": "So far, I have passed by the living room door three times.", + "original_options": { + "A": "I passed by it twice.", + "B": "I passed by it four times.", + "C": "So far, I have passed by the living room door five times, not three.", + "D": "So far, I have passed by the living room door three times.", + "E": "So far, I have passed by the living room door six times in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_1_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Relative to me, where is the doorway?", + "answer": "The doorway is in front of you.", + "question_chinese": "相对于我的位置，门口在哪里？", + "answer_chinese": "门口在你的前面", + "evidence": [ + { + "description": "The door is in front of you.", + "timestamps": { + "start": "00:01:08", + "end": "00:01:12" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The doorway is behind you.", + "B": "The doorway is to your left.", + "C": "The doorway is to your right, rather than straight ahead of you.", + "D": "The doorway is above you, meaning it is higher up instead of in front.", + "E": "The doorway is in front of you." + }, + "answer_label": "E", + "answer_text": "The doorway is in front of you.", + "q_id": "egolife_A1_D4_012_1_T1", + "correct_answer": "The doorway is in front of you.", + "original_options": { + "A": "The doorway is behind you.", + "B": "The doorway is to your left.", + "C": "The doorway is to your right, rather than straight ahead of you.", + "D": "The doorway is above you, meaning it is higher up instead of in front.", + "E": "The doorway is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_012_1_T2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Relative to me, where is the doorway?", + "answer": "The doorway is to your right.", + "question_chinese": "相对于我的位置，门口在哪里？", + "answer_chinese": "门口在你的右边。", + "evidence": [ + { + "description": "The door is on your right.", + "timestamps": { + "start": "00:03:06", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The doorway is in front of you.", + "B": "The doorway is to your left.", + "C": "The doorway is to your right.", + "D": "The doorway is behind you, not off to either side from where you are standing.", + "E": "The doorway is above you, higher than your current eye level, rather than to the right." + }, + "answer_label": "C", + "answer_text": "The doorway is to your right.", + "q_id": "egolife_A1_D4_012_1_T2", + "correct_answer": "The doorway is to your right.", + "original_options": { + "A": "The doorway is in front of you.", + "B": "The doorway is to your left.", + "C": "The doorway is to your right.", + "D": "The doorway is behind you, not off to either side from where you are standing.", + "E": "The doorway is above you, higher than your current eye level, rather than to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D4_013-mcq.json b/QAs/egolife_A1_D4_013-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..aa5a1d779cb0970126c40f019b7a9513cf8b06b5 --- /dev/null +++ b/QAs/egolife_A1_D4_013-mcq.json @@ -0,0 +1,734 @@ +[ + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_6", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many people are on the second floor?", + "answer": "There are seven people on the second floor.", + "question_chinese": "二楼有几个人？", + "answer_chinese": "二楼有七个人", + "evidence": [ + { + "description": "I'm also on the second floor", + "timestamps": { + "start": "00:01:05", + "end": "00:01:09" + } + }, + { + "description": "There are six people in my field of vision", + "timestamps": { + "start": "00:00:59", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are six people.", + "B": "There are eight people.", + "C": "There are five people on the second floor, spread out along the hallway.", + "D": "There are nine people on the second floor, with a few standing near the railing.", + "E": "There are seven people on the second floor." + }, + "answer_label": "E", + "answer_text": "There are seven people on the second floor.", + "q_id": "egolife_A1_D4_013_6", + "correct_answer": "There are seven people on the second floor.", + "original_options": { + "A": "There are six people.", + "B": "There are eight people.", + "C": "There are five people on the second floor, spread out along the hallway.", + "D": "There are nine people on the second floor, with a few standing near the railing.", + "E": "There are seven people on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_5", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many people are on the second floor?", + "answer": "There are six people on the second floor.", + "question_chinese": "二楼有几个人？", + "answer_chinese": "二楼有六个人", + "evidence": [ + { + "description": "There are five people in my field of vision", + "timestamps": { + "start": "00:07:09", + "end": "00:07:12" + } + }, + { + "description": "I'm also on the second floor", + "timestamps": { + "start": "00:07:13", + "end": "00:07:18" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are five people.", + "B": "There are six people on the second floor.", + "C": "There are seven people.", + "D": "There are four people on the second floor, with no one else up there.", + "E": "There are eight people on the second floor, including a small group clustered together." + }, + "answer_label": "B", + "answer_text": "There are six people on the second floor.", + "q_id": "egolife_A1_D4_013_5", + "correct_answer": "There are six people on the second floor.", + "original_options": { + "A": "There are five people.", + "B": "There are six people on the second floor.", + "C": "There are seven people.", + "D": "There are four people on the second floor, with no one else up there.", + "E": "There are eight people on the second floor, including a small group clustered together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_3_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "The last time I stood still, who was I closer to?", + "answer": "The girl in the black dress was the closest to me, and she was to your right.", + "question_chinese": "我上一次站着不动的时候我离谁更近？", + "answer_chinese": "黑裙子的女生离我最近，黑裙子女生在我右边。", + "evidence": [ + { + "description": "The girl in the black dress is the closest to me", + "timestamps": { + "start": "00:00:58", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:03:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The girl in the white dress.", + "B": "The guy in black clothes.", + "C": "The girl in the black dress was the closest to me, and she was to your right.", + "D": "The girl in the black dress was the closest to me, and she was to your left.", + "E": "The girl in the black dress was the closest to me, and she was directly in front of you." + }, + "answer_label": "C", + "answer_text": "The girl in the black dress was the closest to me, and she was to your right.", + "q_id": "egolife_A1_D4_013_3_T1", + "correct_answer": "The girl in the black dress was the closest to me, and she was to your right.", + "original_options": { + "A": "The girl in the white dress.", + "B": "The guy in black clothes.", + "C": "The girl in the black dress was the closest to me, and she was to your right.", + "D": "The girl in the black dress was the closest to me, and she was to your left.", + "E": "The girl in the black dress was the closest to me, and she was directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_3_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "The last time I stood still, who was I closer to?", + "answer": "The guy in black clothes was the closest to me, and he was to your left.", + "question_chinese": "我上一次站着不动的时候我离谁更近？", + "answer_chinese": "黑衣服的男生离我最近，他在我左边。", + "evidence": [ + { + "description": "The boy in the black coat is the closest to me", + "timestamps": { + "start": "00:04:10", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:04:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The girl in the black dress.", + "B": "The guy in black clothes was the closest to me, and he was to your left.", + "C": "The guy in white clothes.", + "D": "The guy in black clothes was the closest to me, and he was to your right.", + "E": "The guy in black clothes was the closest to me, and he was standing behind you." + }, + "answer_label": "B", + "answer_text": "The guy in black clothes was the closest to me, and he was to your left.", + "q_id": "egolife_A1_D4_013_3_T2", + "correct_answer": "The guy in black clothes was the closest to me, and he was to your left.", + "original_options": { + "A": "The girl in the black dress.", + "B": "The guy in black clothes was the closest to me, and he was to your left.", + "C": "The guy in white clothes.", + "D": "The guy in black clothes was the closest to me, and he was to your right.", + "E": "The guy in black clothes was the closest to me, and he was standing behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_5_T2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What was I holding in my hand just now?", + "answer": "You were holding two big lights in your hand.", + "question_chinese": "我刚才手上拿着什么？", + "answer_chinese": "我手上拿着两个大灯", + "evidence": [ + { + "description": "I'm holding two big lights in my hand", + "timestamps": { + "start": "00:01:49", + "end": "00:01:59" + } + } + ], + "question_timestamps": "00:03:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A bottle of vinegar.", + "B": "One big light.", + "C": "You were holding two big lights in your hand.", + "D": "You were holding a small bottle in your hand instead of any lights.", + "E": "You were holding two small lights in your hand, not the big ones." + }, + "answer_label": "C", + "answer_text": "You were holding two big lights in your hand.", + "q_id": "egolife_A1_D4_013_5_T2", + "correct_answer": "You were holding two big lights in your hand.", + "original_options": { + "A": "A bottle of vinegar.", + "B": "One big light.", + "C": "You were holding two big lights in your hand.", + "D": "You were holding a small bottle in your hand instead of any lights.", + "E": "You were holding two small lights in your hand, not the big ones." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_5_T4", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "What was I holding in my hand just now?", + "answer": "You were holding a bottle of vinegar in your hand.", + "question_chinese": "我刚才手上拿着什么？", + "answer_chinese": "我手上拿着一瓶醋", + "evidence": [ + { + "description": "I'm holding a bottle of vinegar in my hand", + "timestamps": { + "start": "00:07:29", + "end": "00:07:32" + } + } + ], + "question_timestamps": "00:07:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were holding a bottle of vinegar in your hand.", + "B": "Two big lights.", + "C": "A water bottle.", + "D": "You were holding two big lights in your hand instead of any kind of bottle.", + "E": "You were holding a different bottle, not the vinegar one, in your hand." + }, + "answer_label": "A", + "answer_text": "You were holding a bottle of vinegar in your hand.", + "q_id": "egolife_A1_D4_013_5_T4", + "correct_answer": "You were holding a bottle of vinegar in your hand.", + "original_options": { + "A": "You were holding a bottle of vinegar in your hand.", + "B": "Two big lights.", + "C": "A water bottle.", + "D": "You were holding two big lights in your hand instead of any kind of bottle.", + "E": "You were holding a different bottle, not the vinegar one, in your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I went up the stairs from the first floor to the second floor.", + "question_chinese": "我刚走了什么路线", + "answer_chinese": "我从一楼通过楼梯到达二楼", + "evidence": [ + { + "description": "I went up the stairs from the first floor to the second floor", + "timestamps": { + "start": "00:00:33", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I went downstairs to the first floor.", + "B": "I went up to the third floor.", + "C": "I went up the stairs from the first floor to the second floor.", + "D": "I went down the stairs from the second floor to the first floor.", + "E": "I went up the stairs from the ground floor to the first floor instead." + }, + "answer_label": "C", + "answer_text": "I went up the stairs from the first floor to the second floor.", + "q_id": "egolife_A1_D4_013_1", + "correct_answer": "I went up the stairs from the first floor to the second floor.", + "original_options": { + "A": "I went downstairs to the first floor.", + "B": "I went up to the third floor.", + "C": "I went up the stairs from the first floor to the second floor.", + "D": "I went down the stairs from the second floor to the first floor.", + "E": "I went up the stairs from the ground floor to the first floor instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I went from the second floor to the first floor via the stairs and then went out into the yard.", + "question_chinese": "我刚走了什么路线", + "answer_chinese": "我从二楼通过楼梯到一楼并且出门走进院子", + "evidence": [ + { + "description": "I went from the second floor to the first floor via the stairs and then went out into the yard", + "timestamps": { + "start": "00:01:21", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:01:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I went out into the kitchen.", + "B": "I went out into the garage.", + "C": "I went downstairs from the second floor and ended up going out into the hallway instead of outside.", + "D": "I went from the second floor to the first floor via the stairs and then went out into the yard.", + "E": "I went from the second floor to the first floor via the stairs, then headed into the living room afterward." + }, + "answer_label": "D", + "answer_text": "I went from the second floor to the first floor via the stairs and then went out into the yard.", + "q_id": "egolife_A1_D4_013_2", + "correct_answer": "I went from the second floor to the first floor via the stairs and then went out into the yard.", + "original_options": { + "A": "I went out into the kitchen.", + "B": "I went out into the garage.", + "C": "I went downstairs from the second floor and ended up going out into the hallway instead of outside.", + "D": "I went from the second floor to the first floor via the stairs and then went out into the yard.", + "E": "I went from the second floor to the first floor via the stairs, then headed into the living room afterward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_2_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to now, how many times have I gone upstairs?", + "answer": "So far, you’ve gone upstairs once.", + "question_chinese": "目前为止我上了几次楼？", + "answer_chinese": "目前为止你上了1次楼.", + "evidence": [ + { + "description": "The first time going upstairs", + "timestamps": { + "start": "00:00:34", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero times.", + "B": "Twice.", + "C": "So far, you’ve gone upstairs once.", + "D": "I have gone upstairs two times so far during this video today.", + "E": "Up to this point, I've gone upstairs four times in total so far." + }, + "answer_label": "C", + "answer_text": "So far, you’ve gone upstairs once.", + "q_id": "egolife_A1_D4_013_2_T2", + "correct_answer": "So far, you’ve gone upstairs once.", + "original_options": { + "A": "Zero times.", + "B": "Twice.", + "C": "So far, you’ve gone upstairs once.", + "D": "I have gone upstairs two times so far during this video today.", + "E": "Up to this point, I've gone upstairs four times in total so far." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_2_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to now, how many times have I gone upstairs?", + "answer": "So far, I’ve gone upstairs twice.", + "question_chinese": "目前为止我上了几次楼？", + "answer_chinese": "目前为止我上了2次楼", + "evidence": [ + { + "description": "The first time going upstairs", + "timestamps": { + "start": "00:00:34", + "end": "00:00:45" + } + }, + { + "description": "The second time going upstairs", + "timestamps": { + "start": "00:02:14", + "end": "00:02:52" + } + } + ], + "question_timestamps": "00:11:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero times.", + "B": "So far, I’ve gone upstairs twice.", + "C": "Once.", + "D": "Up to this point, I've gone upstairs three times in total so far.", + "E": "So far in the video, I've gone upstairs four times altogether up to now." + }, + "answer_label": "B", + "answer_text": "So far, I’ve gone upstairs twice.", + "q_id": "egolife_A1_D4_013_2_T1", + "correct_answer": "So far, I’ve gone upstairs twice.", + "original_options": { + "A": "Zero times.", + "B": "So far, I’ve gone upstairs twice.", + "C": "Once.", + "D": "Up to this point, I've gone upstairs three times in total so far.", + "E": "So far in the video, I've gone upstairs four times altogether up to now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_9", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many gates have you passed through so far?", + "answer": "So far, I have passed through three doors.", + "question_chinese": "目前为止经过了几个门？", + "answer_chinese": "目前为止经过了3个门", + "evidence": [ + { + "description": "I'm passing through one door", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + }, + { + "description": "I'm passing through one door", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + }, + { + "description": "I'm passing through one door", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:50", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four doors.", + "B": "Five doors.", + "C": "So far, I have passed through six doors in total up to this point.", + "D": "So far, I have passed through three doors.", + "E": "Up to now, I have passed through seven doors in total so far." + }, + "answer_label": "D", + "answer_text": "So far, I have passed through three doors.", + "q_id": "egolife_A1_D4_013_9", + "correct_answer": "So far, I have passed through three doors.", + "original_options": { + "A": "Four doors.", + "B": "Five doors.", + "C": "So far, I have passed through six doors in total up to this point.", + "D": "So far, I have passed through three doors.", + "E": "Up to now, I have passed through seven doors in total so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_013_10", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many gates have you passed through so far?", + "answer": "So far, I have passed through four doors.", + "question_chinese": "目前为止经过了几个门？", + "answer_chinese": "目前为止经过了4个门", + "evidence": [ + { + "description": "I'm passing through one door", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + }, + { + "description": "I'm passing through one door", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + }, + { + "description": "I'm passing through one door", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + }, + { + "description": "I'm passing through one door", + "timestamps": { + "start": "00:00:45", + "end": "00:00:46" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "So far, I have passed through four doors.", + "B": "One door.", + "C": "Two doors.", + "D": "So far, I have passed through three doors in total up to this point.", + "E": "Up to now, I have passed through five doors in total so far." + }, + "answer_label": "A", + "answer_text": "So far, I have passed through four doors.", + "q_id": "egolife_A1_D4_013_10", + "correct_answer": "So far, I have passed through four doors.", + "original_options": { + "A": "So far, I have passed through four doors.", + "B": "One door.", + "C": "Two doors.", + "D": "So far, I have passed through three doors in total up to this point.", + "E": "Up to now, I have passed through five doors in total so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D4_014-mcq.json b/QAs/egolife_A1_D4_014-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9895d3f435e6ebea4a3215f6926aef9d69ed3e3c --- /dev/null +++ b/QAs/egolife_A1_D4_014-mcq.json @@ -0,0 +1,732 @@ +[ + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Did I just walk from my left side or my right side?", + "answer": "I just walked from my right.", + "question_chinese": "我刚从自己的左边还是右边走动的？", + "answer_chinese": "我刚从自己的右边走动的", + "evidence": [ + { + "description": "I just walked from my right", + "timestamps": { + "start": "00:01:55", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:02:49", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "I just walked from my left.", + "B": "I just walked from my right." + }, + "answer_label": "B", + "answer_text": "I just walked from my right.", + "q_id": "egolife_A1_D4_014_5", + "correct_answer": "I just walked from my right.", + "original_options": { + "A": "I just walked from my left.", + "B": "I just walked from my right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014_6", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Did I just walk from my left side or my right side?", + "answer": "I just walked from my left.", + "question_chinese": "我刚从自己的左边还是右边走动的？", + "answer_chinese": "我刚从自己的左边走动的", + "evidence": [ + { + "description": "I just walked from my left", + "timestamps": { + "start": "00:03:36", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:04:18", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "I just walked from my left.", + "B": "I just walked from my right." + }, + "answer_label": "A", + "answer_text": "I just walked from my left.", + "q_id": "egolife_A1_D4_014_6", + "correct_answer": "I just walked from my left.", + "original_options": { + "A": "I just walked from my left.", + "B": "I just walked from my right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was I just holding while moving around?", + "answer": "Holding chopsticks.", + "question_chinese": "我刚拿着什么东西走动？", + "answer_chinese": "拿着筷子", + "evidence": [ + { + "description": "Holding chopsticks", + "timestamps": { + "start": "00:00:09", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Holding a spoon.", + "B": "Holding a plate.", + "C": "Holding chopsticks.", + "D": "I was holding a tennis racket while walking around the area nearby.", + "E": "I was holding a baseball bat while moving around the room earlier." + }, + "answer_label": "C", + "answer_text": "Holding chopsticks.", + "q_id": "egolife_A1_D4_014_1", + "correct_answer": "Holding chopsticks.", + "original_options": { + "A": "Holding a spoon.", + "B": "Holding a plate.", + "C": "Holding chopsticks.", + "D": "I was holding a tennis racket while walking around the area nearby.", + "E": "I was holding a baseball bat while moving around the room earlier." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014_2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was I just holding while moving around?", + "answer": "I walked around with chopsticks and a bowl in my hand.", + "question_chinese": "我刚拿着什么东西走动？", + "answer_chinese": "我拿着筷子和碗走动", + "evidence": [ + { + "description": "I walked around with chopsticks and a bowl in my hand", + "timestamps": { + "start": "00:01:56", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:03:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I carried a bowl only.", + "B": "I held chopsticks and a cup.", + "C": "I walked around holding a spoon and a bowl in my hand instead of chopsticks.", + "D": "I moved around with chopsticks and a plate in my hand, not carrying a bowl.", + "E": "I walked around with chopsticks and a bowl in my hand." + }, + "answer_label": "E", + "answer_text": "I walked around with chopsticks and a bowl in my hand.", + "q_id": "egolife_A1_D4_014_2", + "correct_answer": "I walked around with chopsticks and a bowl in my hand.", + "original_options": { + "A": "I carried a bowl only.", + "B": "I held chopsticks and a cup.", + "C": "I walked around holding a spoon and a bowl in my hand instead of chopsticks.", + "D": "I moved around with chopsticks and a plate in my hand, not carrying a bowl.", + "E": "I walked around with chopsticks and a bowl in my hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014-mcq_fixed_9", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer for me to go to the pink clothes now, to the left or to the right?", + "answer": "It's closer from the left side.", + "question_chinese": "我现在要去粉色衣服旁从左边走更近还是右边更近?", + "answer_chinese": "从左手边更近", + "evidence": [ + { + "description": "The boy in pink is on my left", + "timestamps": { + "start": "00:00:39", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "It's closer from the left side.", + "B": "It's closer from the right side." + }, + "answer_label": "A", + "answer_text": "It's closer from the left side.", + "q_id": "egolife_A1_D4_014-mcq_fixed_9", + "correct_answer": "It's closer from the left side.", + "original_options": { + "A": "It's closer from the left side.", + "B": "It's closer from the right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014-mcq_fixed_8", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer for me to go to the pink clothes now, to the left or to the right?", + "answer": "It's closer from the right.", + "question_chinese": "我现在要去粉色衣服旁从左边走更近还是右边更近?", + "answer_chinese": "从右边更近", + "evidence": [ + { + "description": "The boy in pink is on my right hand side", + "timestamps": { + "start": "00:04:47", + "end": "00:04:49" + } + } + ], + "question_timestamps": "00:05:09", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "It's closer from the left.", + "B": "It's closer from the right." + }, + "answer_label": "B", + "answer_text": "It's closer from the right.", + "q_id": "egolife_A1_D4_014-mcq_fixed_8", + "correct_answer": "It's closer from the right.", + "original_options": { + "A": "It's closer from the left.", + "B": "It's closer from the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014_3_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the refrigerator or closer to the reflector?", + "answer": "You were closer to the pink-haired girl.", + "question_chinese": "我距离冰箱近还是离反光板近？", + "answer_chinese": "我是离粉头发女生近", + "evidence": [ + { + "description": "The girl with pink hair is beside me", + "timestamps": { + "start": "00:00:28", + "end": "00:00:31" + } + }, + { + "description": "There were two people between me and the boy with yellow hair", + "timestamps": { + "start": "00:00:33", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:01:55", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were closer to the pink-haired girl.", + "B": "You were closer to the yellow-haired boy." + }, + "answer_label": "A", + "answer_text": "You were closer to the pink-haired girl.", + "q_id": "egolife_A1_D4_014_3_T1", + "correct_answer": "You were closer to the pink-haired girl.", + "original_options": { + "A": "You were closer to the pink-haired girl.", + "B": "You were closer to the yellow-haired boy." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014_3_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the refrigerator or closer to the reflector?", + "answer": "You were closer to the yellow-haired boy.", + "question_chinese": "我距离冰箱近还是离反光板近？", + "answer_chinese": "我距离黄色头发的男孩近", + "evidence": [ + { + "description": "I am standing next to the boy with yellow hair", + "timestamps": { + "start": "00:02:06", + "end": "00:02:08" + } + }, + { + "description": "There are two people between the pink-haired girl and me.", + "timestamps": { + "start": "00:02:21", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:02:51", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were closer to the yellow-haired boy.", + "B": "You were closer to the reflector." + }, + "answer_label": "A", + "answer_text": "You were closer to the yellow-haired boy.", + "q_id": "egolife_A1_D4_014_3_T2", + "correct_answer": "You were closer to the yellow-haired boy.", + "original_options": { + "A": "You were closer to the yellow-haired boy.", + "B": "You were closer to the reflector." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014_3_T3", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the refrigerator or closer to the reflector?", + "answer": "You are closer to the reflector.", + "question_chinese": "我距离冰箱近还是离反光板近？", + "answer_chinese": "你距离反光板近", + "evidence": [ + { + "description": "You walked near the reflector, and the refrigerator is two tables away from you.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:53" + } + } + ], + "question_timestamps": "00:08:45", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are closer to the refrigerator.", + "B": "You are closer to the reflector." + }, + "answer_label": "B", + "answer_text": "You are closer to the reflector.", + "q_id": "egolife_A1_D4_014_3_T3", + "correct_answer": "You are closer to the reflector.", + "original_options": { + "A": "You are closer to the refrigerator.", + "B": "You are closer to the reflector." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014_3_T4", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the refrigerator or closer to the reflector?", + "answer": "You are closer to the fridge.", + "question_chinese": "我距离冰箱近还是离反光板近？", + "answer_chinese": "你距离冰箱近", + "evidence": [ + { + "description": "You walked near the refrigerator, with the reflector a table's distance away from you.", + "timestamps": { + "start": "00:08:45", + "end": "00:08:59" + } + } + ], + "question_timestamps": "00:09:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You are closer to the reflector.", + "B": "You are closer to the fridge." + }, + "answer_label": "B", + "answer_text": "You are closer to the fridge.", + "q_id": "egolife_A1_D4_014_3_T4", + "correct_answer": "You are closer to the fridge.", + "original_options": { + "A": "You are closer to the reflector.", + "B": "You are closer to the fridge." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014_7", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the girl with pink hair or the boy with yellow hair?", + "answer": "I'm close to girls with pink hair.", + "question_chinese": "我距离粉色头发的女孩近还是和黄色头发的男孩近？", + "answer_chinese": "我是离粉头发女生近", + "evidence": [ + { + "description": "The girl with pink hair is beside me", + "timestamps": { + "start": "00:00:28", + "end": "00:00:31" + } + }, + { + "description": "There were two people between me and the boy with yellow hair", + "timestamps": { + "start": "00:00:33", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:02:05", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I'm close to boys with yellow hair.", + "B": "I'm close to girls with pink hair." + }, + "answer_label": "B", + "answer_text": "I'm close to girls with pink hair.", + "q_id": "egolife_A1_D4_014_7", + "correct_answer": "I'm close to girls with pink hair.", + "original_options": { + "A": "I'm close to boys with yellow hair.", + "B": "I'm close to girls with pink hair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014_8", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the girl with pink hair or the boy with yellow hair?", + "answer": "I was close to the boy with yellow hair.", + "question_chinese": "我距离粉色头发的女孩近还是和黄色头发的男孩近？", + "answer_chinese": "我距离黄色头发的男孩近", + "evidence": [ + { + "description": "I'm standing beside the boy with yellow hair", + "timestamps": { + "start": "00:02:06", + "end": "00:02:08" + } + }, + { + "description": "There are two people between me and the girl with pink hair", + "timestamps": { + "start": "00:02:21", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I was close to the boy with yellow hair.", + "B": "I was close to the girl with pink hair." + }, + "answer_label": "A", + "answer_text": "I was close to the boy with yellow hair.", + "q_id": "egolife_A1_D4_014_8", + "correct_answer": "I was close to the boy with yellow hair.", + "original_options": { + "A": "I was close to the boy with yellow hair.", + "B": "I was close to the girl with pink hair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014-mcq_fixed_4", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the window in my direction?", + "answer": "The window is behind you.", + "question_chinese": "窗户在我的哪个方向", + "answer_chinese": "窗户在我身后", + "evidence": [ + { + "description": "The window is in front of me", + "timestamps": { + "start": "00:08:41", + "end": "00:08:42" + } + }, + { + "description": "I turn 180 degrees", + "timestamps": { + "start": "00:08:46", + "end": "00:08:56" + } + } + ], + "question_timestamps": "00:09:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is behind you.", + "D": "The window is diagonally in front of you, slightly to the right, so you can see it when facing forward.", + "E": "The window is above you on the wall, so you have to look up to see it clearly." + }, + "answer_label": "C", + "answer_text": "The window is behind you.", + "q_id": "egolife_A1_D4_014-mcq_fixed_4", + "correct_answer": "The window is behind you.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is behind you.", + "D": "The window is diagonally in front of you, slightly to the right, so you can see it when facing forward.", + "E": "The window is above you on the wall, so you have to look up to see it clearly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_014-mcq_fixed_5", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the window in my direction?", + "answer": "The window is on your right.", + "question_chinese": "窗户在我的哪个方向", + "answer_chinese": "窗户在我右边", + "evidence": [ + { + "description": "The window is in front of me", + "timestamps": { + "start": "00:09:52", + "end": "00:09:53" + } + }, + { + "description": "I turned 90 degrees to the right", + "timestamps": { + "start": "00:09:54", + "end": "00:10:02" + } + } + ], + "question_timestamps": "00:10:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The window is on your left.", + "B": "The window is behind you.", + "C": "The window is directly in front of you, so you face it when you look straight ahead.", + "D": "The window is above you on the wall, requiring you to tilt your head up to see it.", + "E": "The window is on your right." + }, + "answer_label": "E", + "answer_text": "The window is on your right.", + "q_id": "egolife_A1_D4_014-mcq_fixed_5", + "correct_answer": "The window is on your right.", + "original_options": { + "A": "The window is on your left.", + "B": "The window is behind you.", + "C": "The window is directly in front of you, so you face it when you look straight ahead.", + "D": "The window is above you on the wall, requiring you to tilt your head up to see it.", + "E": "The window is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D4_015-mcq.json b/QAs/egolife_A1_D4_015-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..17c43d74f616bc8d5e795f3301884ca542109f9b --- /dev/null +++ b/QAs/egolife_A1_D4_015-mcq.json @@ -0,0 +1,633 @@ +[ + { + "video_uid": "egolife_A1_D4_015", + "video_length": "00:10:55", + "qid": "egolife_A1_D4_015_2_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "The last time I sat down, how many beverage bottles in front of me were within reach to touch?", + "answer": "The last time I sat down, there were 6 drink bottles in front of you that I could reach.", + "question_chinese": "我上一次坐下时面前可以碰到的饮料瓶子有几个？", + "answer_chinese": "我上一次坐下时面前可以碰到的饮料瓶子有6个", + "evidence": [ + { + "description": "There are four beverage bottles and a thermos in front of me. One transparent cup can be touched", + "timestamps": { + "start": "00:00:02", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:01:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Five bottles were within reach.", + "B": "Seven bottles were within reach.", + "C": "The last time I sat down, there were 4 drink bottles in front of you that I could reach.", + "D": "The last time I sat down, there were 6 drink bottles in front of you that I could reach.", + "E": "The last time I sat down, there were 8 drink bottles in front of you that I could reach." + }, + "answer_label": "D", + "answer_text": "The last time I sat down, there were 6 drink bottles in front of you that I could reach.", + "q_id": "egolife_A1_D4_015_2_T1", + "correct_answer": "The last time I sat down, there were 6 drink bottles in front of you that I could reach.", + "original_options": { + "A": "Five bottles were within reach.", + "B": "Seven bottles were within reach.", + "C": "The last time I sat down, there were 4 drink bottles in front of you that I could reach.", + "D": "The last time I sat down, there were 6 drink bottles in front of you that I could reach.", + "E": "The last time I sat down, there were 8 drink bottles in front of you that I could reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_015", + "video_length": "00:10:55", + "qid": "egolife_A1_D4_015_2_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "The last time I sat down, how many beverage bottles in front of me were within reach to touch?", + "answer": "The last time you sat down, there were 0 drink bottles in front of you that you could reach.", + "question_chinese": "我上一次坐下时面前可以碰到的饮料瓶子有几个？", + "answer_chinese": "你上一次坐下时面前可以碰到的饮料瓶子有0个", + "evidence": [ + { + "description": "There are no beverage bottle can be touched.", + "timestamps": { + "start": "00:06:40", + "end": "00:06:46" + } + } + ], + "question_timestamps": "00:06:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One bottle was within reach.", + "B": "Two bottles were within reach.", + "C": "The last time you sat down, there were 3 drink bottles in front of you that you could reach.", + "D": "The last time you sat down, there were 4 drink bottles in front of you that you could reach.", + "E": "The last time you sat down, there were 0 drink bottles in front of you that you could reach." + }, + "answer_label": "E", + "answer_text": "The last time you sat down, there were 0 drink bottles in front of you that you could reach.", + "q_id": "egolife_A1_D4_015_2_T2", + "correct_answer": "The last time you sat down, there were 0 drink bottles in front of you that you could reach.", + "original_options": { + "A": "One bottle was within reach.", + "B": "Two bottles were within reach.", + "C": "The last time you sat down, there were 3 drink bottles in front of you that you could reach.", + "D": "The last time you sat down, there were 4 drink bottles in front of you that you could reach.", + "E": "The last time you sat down, there were 0 drink bottles in front of you that you could reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_015", + "video_length": "00:10:55", + "qid": "egolife_A1_D4_015_7", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get to the yard?", + "answer": "Go straight and then turn right out of the living room. Then turn left out and you'll reach the courtyard.", + "question_chinese": "我想要去院子里应该怎么走", + "answer_chinese": "直走然后右拐出客厅，接着左拐出门就是庭院", + "evidence": [ + { + "description": "Now my position is the same as in the video", + "timestamps": { + "start": "00:05:23", + "end": "00:05:24" + } + }, + { + "description": "I walked straight and then turned right out of the living room. After that, I turned left out of the door and it was the courtyard", + "timestamps": { + "start": "00:05:24", + "end": "00:05:36" + } + } + ], + "question_timestamps": "00:06:01", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn left.", + "B": "Turn right out of the living room.", + "C": "Go straight and turn left out of the living room, then turn right outside to reach the courtyard.", + "D": "Go straight past the living room, turn right at the next exit, then turn right again to the yard.", + "E": "Go straight and then turn right out of the living room. Then turn left out and you'll reach the courtyard." + }, + "answer_label": "E", + "answer_text": "Go straight and then turn right out of the living room. Then turn left out and you'll reach the courtyard.", + "q_id": "egolife_A1_D4_015_7", + "correct_answer": "Go straight and then turn right out of the living room. Then turn left out and you'll reach the courtyard.", + "original_options": { + "A": "Go straight, then turn left.", + "B": "Turn right out of the living room.", + "C": "Go straight and turn left out of the living room, then turn right outside to reach the courtyard.", + "D": "Go straight past the living room, turn right at the next exit, then turn right again to the yard.", + "E": "Go straight and then turn right out of the living room. Then turn left out and you'll reach the courtyard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_015", + "video_length": "00:10:55", + "qid": "egolife_A1_D4_015_8", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get to the yard?", + "answer": "Turn back and walk straight, then go out and you'll reach the courtyard.", + "question_chinese": "我想要去院子里应该怎么走", + "answer_chinese": "向后转直走然后出门就是庭院", + "evidence": [ + { + "description": "The stairs are in front of me", + "timestamps": { + "start": "00:05:39", + "end": "00:05:45" + } + }, + { + "description": "In front of me is a courtyard", + "timestamps": { + "start": "00:05:33", + "end": "00:05:35" + } + }, + { + "description": "I turned around", + "timestamps": { + "start": "00:05:37", + "end": "00:05:39" + } + } + ], + "question_timestamps": "00:09:01", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn back and walk straight, then go out and you'll reach the courtyard.", + "B": "Turn around and go left.", + "C": "Keep going forward and exit.", + "D": "Turn back and walk straight, then turn right before going out to reach the courtyard.", + "E": "Turn back and walk straight, then go out and you'll reach the front porch instead." + }, + "answer_label": "A", + "answer_text": "Turn back and walk straight, then go out and you'll reach the courtyard.", + "q_id": "egolife_A1_D4_015_8", + "correct_answer": "Turn back and walk straight, then go out and you'll reach the courtyard.", + "original_options": { + "A": "Turn back and walk straight, then go out and you'll reach the courtyard.", + "B": "Turn around and go left.", + "C": "Keep going forward and exit.", + "D": "Turn back and walk straight, then turn right before going out to reach the courtyard.", + "E": "Turn back and walk straight, then go out and you'll reach the front porch instead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_015", + "video_length": "00:10:55", + "qid": "egolife_A1_D4_015_5", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the window facing now?", + "answer": "Now the window is behind me.", + "question_chinese": "现在窗户在我的哪个方位", + "answer_chinese": "现在窗户在我的后面", + "evidence": [ + { + "description": "I'm sitting in the same position as I am now", + "timestamps": { + "start": "00:00:00", + "end": "00:00:16" + } + }, + { + "description": "I got up and turned 180 degrees before sitting down", + "timestamps": { + "start": "00:00:20", + "end": "00:00:32" + } + }, + { + "description": "The window is right opposite me now", + "timestamps": { + "start": "00:00:33", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Now the window is in front of me.", + "B": "Now the window is behind me.", + "C": "Now the window is to my left.", + "D": "Now the window is above me, so I have to look up to see it.", + "E": "Now the window is on my right side, right next to where I'm standing." + }, + "answer_label": "B", + "answer_text": "Now the window is behind me.", + "q_id": "egolife_A1_D4_015_5", + "correct_answer": "Now the window is behind me.", + "original_options": { + "A": "Now the window is in front of me.", + "B": "Now the window is behind me.", + "C": "Now the window is to my left.", + "D": "Now the window is above me, so I have to look up to see it.", + "E": "Now the window is on my right side, right next to where I'm standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_015", + "video_length": "00:10:55", + "qid": "egolife_A1_D4_015_6", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the window facing now?", + "answer": "The window is on my right now.", + "question_chinese": "现在窗户在我的哪个方位", + "answer_chinese": "现在窗户在我右边", + "evidence": [ + { + "description": "The window is right opposite me now", + "timestamps": { + "start": "00:05:57", + "end": "00:05:59" + } + }, + { + "description": "I turned left by 90 degrees", + "timestamps": { + "start": "00:08:03", + "end": "00:08:04" + } + } + ], + "question_timestamps": "00:08:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The window is on my left now.", + "B": "The window is in front of me now.", + "C": "The window is on my right now.", + "D": "The window is behind me right now, not on my right side.", + "E": "The window is slightly to my left now as I face the room." + }, + "answer_label": "C", + "answer_text": "The window is on my right now.", + "q_id": "egolife_A1_D4_015_6", + "correct_answer": "The window is on my right now.", + "original_options": { + "A": "The window is on my left now.", + "B": "The window is in front of me now.", + "C": "The window is on my right now.", + "D": "The window is behind me right now, not on my right side.", + "E": "The window is slightly to my left now as I face the room." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_015", + "video_length": "00:10:55", + "qid": "egolife_A1_D4_015_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are currently in the living room?", + "answer": "So far, there are five people in the living room.", + "question_chinese": "目前为止客厅里面有几个人？", + "answer_chinese": "目前为止客厅里有5个人", + "evidence": [ + { + "description": "There are three people sitting on the chairs here. I'm standing alone and I'm also in the living room", + "timestamps": { + "start": "00:05:56", + "end": "00:05:58" + } + } + ], + "question_timestamps": "00:06:17", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are four people.", + "B": "There are six people.", + "C": "So far, there are three people in the living room at the moment.", + "D": "So far, there are seven people in the living room right now.", + "E": "So far, there are five people in the living room." + }, + "answer_label": "E", + "answer_text": "So far, there are five people in the living room.", + "q_id": "egolife_A1_D4_015_3", + "correct_answer": "So far, there are five people in the living room.", + "original_options": { + "A": "There are four people.", + "B": "There are six people.", + "C": "So far, there are three people in the living room at the moment.", + "D": "So far, there are seven people in the living room right now.", + "E": "So far, there are five people in the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_015", + "video_length": "00:10:55", + "qid": "egolife_A1_D4_015_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are currently in the living room?", + "answer": "So far, there are four people in the living room.", + "question_chinese": "目前为止客厅里面有几个人？", + "answer_chinese": "目前为止客厅里面有四个人", + "evidence": [ + { + "description": "There are three people sitting on the chairs here. I'm standing alone and I'm also in the living room", + "timestamps": { + "start": "00:05:56", + "end": "00:05:58" + } + }, + { + "description": "A person came out of the room", + "timestamps": { + "start": "00:06:23", + "end": "00:06:26" + } + } + ], + "question_timestamps": "00:06:53", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are three people.", + "B": "There are five people.", + "C": "So far, there are four people in the living room.", + "D": "So far, there are two people in the living room right now.", + "E": "So far, there are six people in the living room at the moment." + }, + "answer_label": "C", + "answer_text": "So far, there are four people in the living room.", + "q_id": "egolife_A1_D4_015_4", + "correct_answer": "So far, there are four people in the living room.", + "original_options": { + "A": "There are three people.", + "B": "There are five people.", + "C": "So far, there are four people in the living room.", + "D": "So far, there are two people in the living room right now.", + "E": "So far, there are six people in the living room at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_015", + "video_length": "00:10:55", + "qid": "egolife_A1_D4_015_9", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many chairs with backs are there in this room?", + "answer": "There are four chairs with backrests in this room.", + "question_chinese": "这个房间里面有几个带靠背的椅子？", + "answer_chinese": "这个房间里面有4个带靠背的椅子", + "evidence": [ + { + "description": "There are four chairs with backrests in this room", + "timestamps": { + "start": "00:06:14", + "end": "00:06:25" + } + } + ], + "question_timestamps": "00:08:35", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are four chairs with backrests in this room.", + "B": "There are three chairs.", + "C": "There are five chairs.", + "D": "There are two chairs with backrests in this room right now, not four.", + "E": "There are six chairs with backrests in this room at the moment, not four." + }, + "answer_label": "A", + "answer_text": "There are four chairs with backrests in this room.", + "q_id": "egolife_A1_D4_015_9", + "correct_answer": "There are four chairs with backrests in this room.", + "original_options": { + "A": "There are four chairs with backrests in this room.", + "B": "There are three chairs.", + "C": "There are five chairs.", + "D": "There are two chairs with backrests in this room right now, not four.", + "E": "There are six chairs with backrests in this room at the moment, not four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_015", + "video_length": "00:10:55", + "qid": "egolife_A1_D4_015_10", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many chairs with backs are there in this room?", + "answer": "There are eight chairs with backrests in this room.", + "question_chinese": "这个房间里面有几个带靠背的椅子？", + "answer_chinese": "这个房间里面有8个带靠背的椅子", + "evidence": [ + { + "description": "There are eight chairs with backrests in this room", + "timestamps": { + "start": "00:09:24", + "end": "00:09:26" + } + } + ], + "question_timestamps": "00:09:54", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are eight chairs with backrests in this room.", + "B": "There are seven chairs.", + "C": "There are nine chairs.", + "D": "There are six chairs with backrests in this room right now, not eight.", + "E": "There are ten chairs with backrests in this room at the moment, not eight." + }, + "answer_label": "A", + "answer_text": "There are eight chairs with backrests in this room.", + "q_id": "egolife_A1_D4_015_10", + "correct_answer": "There are eight chairs with backrests in this room.", + "original_options": { + "A": "There are eight chairs with backrests in this room.", + "B": "There are seven chairs.", + "C": "There are nine chairs.", + "D": "There are six chairs with backrests in this room right now, not eight.", + "E": "There are ten chairs with backrests in this room at the moment, not eight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D4_016-mcq.json b/QAs/egolife_A1_D4_016-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..eb983462e84d0372ab4bda7c282deb2154d919af --- /dev/null +++ b/QAs/egolife_A1_D4_016-mcq.json @@ -0,0 +1,664 @@ +[ + { + "video_uid": "egolife_A1_D4_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_016_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the kitchen from here?", + "answer": "The kitchen is behind me.", + "question_chinese": "厨房在我哪个方向？", + "answer_chinese": "厨房在我背面", + "evidence": [ + { + "description": "The kitchen is right in front of me", + "timestamps": { + "start": "00:06:20", + "end": "00:06:21" + } + }, + { + "description": "I turned 180 degrees", + "timestamps": { + "start": "00:06:25", + "end": "00:06:29" + } + } + ], + "question_timestamps": "00:06:42", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The kitchen is behind me.", + "B": "The kitchen is to my left.", + "C": "The kitchen is in front of me.", + "D": "The kitchen is off to my right from where I'm standing now.", + "E": "The kitchen is ahead of me a bit, not behind my current position." + }, + "answer_label": "A", + "answer_text": "The kitchen is behind me.", + "q_id": "egolife_A1_D4_016_7", + "correct_answer": "The kitchen is behind me.", + "original_options": { + "A": "The kitchen is behind me.", + "B": "The kitchen is to my left.", + "C": "The kitchen is in front of me.", + "D": "The kitchen is off to my right from where I'm standing now.", + "E": "The kitchen is ahead of me a bit, not behind my current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_016_8", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the kitchen from here?", + "answer": "The kitchen is right in front of me.", + "question_chinese": "厨房在我哪个方向？", + "answer_chinese": "厨房在我正面", + "evidence": [ + { + "description": "The kitchen is right in front of me", + "timestamps": { + "start": "00:09:17", + "end": "00:09:18" + } + } + ], + "question_timestamps": "00:09:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The kitchen is behind me.", + "B": "The kitchen is to my left.", + "C": "The kitchen is right in front of me.", + "D": "The kitchen is off to my right side from where I'm standing right now.", + "E": "The kitchen is behind and slightly to my left from this position." + }, + "answer_label": "C", + "answer_text": "The kitchen is right in front of me.", + "q_id": "egolife_A1_D4_016_8", + "correct_answer": "The kitchen is right in front of me.", + "original_options": { + "A": "The kitchen is behind me.", + "B": "The kitchen is to my left.", + "C": "The kitchen is right in front of me.", + "D": "The kitchen is off to my right side from where I'm standing right now.", + "E": "The kitchen is behind and slightly to my left from this position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_016_3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the place where I am?", + "answer": "There are eight chairs in the place where I am.", + "question_chinese": "我在的这个地方有几个椅子？", + "answer_chinese": "我在的这个地方有8个椅子", + "evidence": [ + { + "description": "There are seven chairs", + "timestamps": { + "start": "00:00:19", + "end": "00:00:20" + } + }, + { + "description": "There is one chair", + "timestamps": { + "start": "00:02:15", + "end": "00:02:16" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are seven chairs.", + "B": "There are eight chairs in the place where I am.", + "C": "There are nine chairs.", + "D": "There are six chairs in the place where I am right now.", + "E": "There are ten chairs in the place where I am at the moment." + }, + "answer_label": "B", + "answer_text": "There are eight chairs in the place where I am.", + "q_id": "egolife_A1_D4_016_3", + "correct_answer": "There are eight chairs in the place where I am.", + "original_options": { + "A": "There are seven chairs.", + "B": "There are eight chairs in the place where I am.", + "C": "There are nine chairs.", + "D": "There are six chairs in the place where I am right now.", + "E": "There are ten chairs in the place where I am at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_016_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the place where I am?", + "answer": "There are four chairs in the place where I am.", + "question_chinese": "我在的这个地方有几个椅子？", + "answer_chinese": "我在的这个地方有4个椅子", + "evidence": [ + { + "description": "There are four chairs in the place where I am", + "timestamps": { + "start": "00:00:47", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:11:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three chairs here.", + "B": "There are five chairs here.", + "C": "There are two chairs in the place where I am right now.", + "D": "There are four chairs in the place where I am.", + "E": "There are six chairs in the place where I am, spread around the room." + }, + "answer_label": "D", + "answer_text": "There are four chairs in the place where I am.", + "q_id": "egolife_A1_D4_016_4", + "correct_answer": "There are four chairs in the place where I am.", + "original_options": { + "A": "There are three chairs here.", + "B": "There are five chairs here.", + "C": "There are two chairs in the place where I am right now.", + "D": "There are four chairs in the place where I am.", + "E": "There are six chairs in the place where I am, spread around the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_016_5", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I taken out the trash so far?", + "answer": "I've been in the trash once so far.", + "question_chinese": "我目前到了几次垃圾？", + "answer_chinese": "我目前到了1次垃圾", + "evidence": [ + { + "description": "I'm taking out the trash", + "timestamps": { + "start": "00:04:29", + "end": "00:04:32" + } + } + ], + "question_timestamps": "00:07:53", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "I've been in the trash twice.", + "B": "I've been in the trash three times.", + "C": "I've been in the trash once so far.", + "D": "I've been in the trash four times so far during this video.", + "E": "I've been in the trash five times so far, going back and forth repeatedly." + }, + "answer_label": "C", + "answer_text": "I've been in the trash once so far.", + "q_id": "egolife_A1_D4_016_5", + "correct_answer": "I've been in the trash once so far.", + "original_options": { + "A": "I've been in the trash twice.", + "B": "I've been in the trash three times.", + "C": "I've been in the trash once so far.", + "D": "I've been in the trash four times so far during this video.", + "E": "I've been in the trash five times so far, going back and forth repeatedly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_016_6", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I taken out the trash so far?", + "answer": "I've been in the trash twice so far.", + "question_chinese": "我目前到了几次垃圾？", + "answer_chinese": "我目前到了2次垃圾", + "evidence": [ + { + "description": "I'm taking out the trash", + "timestamps": { + "start": "00:04:29", + "end": "00:04:32" + } + }, + { + "description": "I'm taking out the trash", + "timestamps": { + "start": "00:08:22", + "end": "00:08:25" + } + } + ], + "question_timestamps": "00:09:03", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "I've been in the trash once.", + "B": "I've been in the trash three times.", + "C": "I've been in the trash four times so far during this video.", + "D": "I've been in the trash twice so far.", + "E": "I've been in the trash five times so far, taking it out repeatedly." + }, + "answer_label": "D", + "answer_text": "I've been in the trash twice so far.", + "q_id": "egolife_A1_D4_016_6", + "correct_answer": "I've been in the trash twice so far.", + "original_options": { + "A": "I've been in the trash once.", + "B": "I've been in the trash three times.", + "C": "I've been in the trash four times so far during this video.", + "D": "I've been in the trash twice so far.", + "E": "I've been in the trash five times so far, taking it out repeatedly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_016_9", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors did I pass through?", + "answer": "I passed through the door four times.", + "question_chinese": "我经过了几次门？", + "answer_chinese": "我经过了4次门", + "evidence": [ + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:00", + "end": "00:02:01" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:13", + "end": "00:02:14" + } + } + ], + "question_timestamps": "00:04:23", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I passed through the door once.", + "B": "I passed through the door three times.", + "C": "I passed through the door two times over the course of the video.", + "D": "I passed through the door five times while moving between the areas.", + "E": "I passed through the door four times." + }, + "answer_label": "E", + "answer_text": "I passed through the door four times.", + "q_id": "egolife_A1_D4_016_9", + "correct_answer": "I passed through the door four times.", + "original_options": { + "A": "I passed through the door once.", + "B": "I passed through the door three times.", + "C": "I passed through the door two times over the course of the video.", + "D": "I passed through the door five times while moving between the areas.", + "E": "I passed through the door four times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_016_10", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors did I pass through?", + "answer": "I passed through the door five times.", + "question_chinese": "我经过了几次门？", + "answer_chinese": "我经过了5次门", + "evidence": [ + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:00", + "end": "00:02:01" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:02:13", + "end": "00:02:14" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:04:23", + "end": "00:04:24" + } + } + ], + "question_timestamps": "00:04:43", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I passed through the door four times.", + "B": "I passed through the door six times.", + "C": "I passed through the door three times while moving around the house.", + "D": "I passed through the door two times before stopping in the next room.", + "E": "I passed through the door five times." + }, + "answer_label": "E", + "answer_text": "I passed through the door five times.", + "q_id": "egolife_A1_D4_016_10", + "correct_answer": "I passed through the door five times.", + "original_options": { + "A": "I passed through the door four times.", + "B": "I passed through the door six times.", + "C": "I passed through the door three times while moving around the house.", + "D": "I passed through the door two times before stopping in the next room.", + "E": "I passed through the door five times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_016_6_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many people have I seen?", + "answer": "So far, I have seen 1 person.", + "question_chinese": "目前为止我看到了几个人？", + "answer_chinese": "目前为止我看到了1个人", + "evidence": [ + { + "description": "There is a man in black standing in the field of vision", + "timestamps": { + "start": "00:00:24", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, I have seen 0 people.", + "B": "So far, I have seen 2 people.", + "C": "So far, I have seen 3 people while moving through the space.", + "D": "So far, I have seen 4 people during the video up to this point.", + "E": "So far, I have seen 1 person." + }, + "answer_label": "E", + "answer_text": "So far, I have seen 1 person.", + "q_id": "egolife_A1_D4_016_6_T2", + "correct_answer": "So far, I have seen 1 person.", + "original_options": { + "A": "So far, I have seen 0 people.", + "B": "So far, I have seen 2 people.", + "C": "So far, I have seen 3 people while moving through the space.", + "D": "So far, I have seen 4 people during the video up to this point.", + "E": "So far, I have seen 1 person." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_016_6_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many people have I seen?", + "answer": "So far, I have seen 6 people.", + "question_chinese": "目前为止我看到了几个人？", + "answer_chinese": "目前为止我看到了6个人", + "evidence": [ + { + "description": "There is a man in black standing in the field of vision", + "timestamps": { + "start": "00:00:24", + "end": "00:00:33" + } + }, + { + "description": "There are four people in the field of vision", + "timestamps": { + "start": "00:00:49", + "end": "00:00:53" + } + }, + { + "description": "A boy is leaning against the door frame.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, I have seen 5 people.", + "B": "So far, I have seen 7 people.", + "C": "So far, I have seen 4 people during the video up to now.", + "D": "So far, I have seen 6 people.", + "E": "So far, I have seen 8 people as I moved through different areas." + }, + "answer_label": "D", + "answer_text": "So far, I have seen 6 people.", + "q_id": "egolife_A1_D4_016_6_T1", + "correct_answer": "So far, I have seen 6 people.", + "original_options": { + "A": "So far, I have seen 5 people.", + "B": "So far, I have seen 7 people.", + "C": "So far, I have seen 4 people during the video up to now.", + "D": "So far, I have seen 6 people.", + "E": "So far, I have seen 8 people as I moved through different areas." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D4_017-mcq.json b/QAs/egolife_A1_D4_017-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f69057502bc825a6cfb7e003ecf15196cefe244f --- /dev/null +++ b/QAs/egolife_A1_D4_017-mcq.json @@ -0,0 +1,685 @@ +[ + { + "video_uid": "egolife_A1_D4_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_017_9", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the table on the second floor?", + "answer": "The table on the second floor is right in the middle of the house.", + "question_chinese": "二楼的桌子在哪？", + "answer_chinese": "二楼的桌子在房子正中间", + "evidence": [ + { + "description": "The table on the second floor is right in the middle of the house", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:02:29", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "It is against the wall.", + "B": "The table on the second floor is right in the middle of the house.", + "C": "It is near the window.", + "D": "The table on the second floor is tucked into a corner along the wall.", + "E": "The table on the second floor is positioned near the window rather than centrally." + }, + "answer_label": "B", + "answer_text": "The table on the second floor is right in the middle of the house.", + "q_id": "egolife_A1_D4_017_9", + "correct_answer": "The table on the second floor is right in the middle of the house.", + "original_options": { + "A": "It is against the wall.", + "B": "The table on the second floor is right in the middle of the house.", + "C": "It is near the window.", + "D": "The table on the second floor is tucked into a corner along the wall.", + "E": "The table on the second floor is positioned near the window rather than centrally." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_017_8", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the table on the second floor?", + "answer": "The table on the second floor is by the wall.", + "question_chinese": "二楼的桌子在哪？", + "answer_chinese": "二楼的桌子在墙边", + "evidence": [ + { + "description": "The table on the second floor is right in the middle of the house", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + }, + { + "description": "The three of us moved the table to the wall", + "timestamps": { + "start": "00:02:20", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:03:35", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Near the staircase.", + "B": "Across from the door.", + "C": "The table on the second floor is under the window near the corner.", + "D": "The table on the second floor is by the wall.", + "E": "The table on the second floor sits in the middle of the room, away from any walls." + }, + "answer_label": "D", + "answer_text": "The table on the second floor is by the wall.", + "q_id": "egolife_A1_D4_017_8", + "correct_answer": "The table on the second floor is by the wall.", + "original_options": { + "A": "Near the staircase.", + "B": "Across from the door.", + "C": "The table on the second floor is under the window near the corner.", + "D": "The table on the second floor is by the wall.", + "E": "The table on the second floor sits in the middle of the room, away from any walls." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_017_1_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just went to get a drink?", + "answer": "You were walking around the living room holding a drink.", + "question_chinese": "我刚拿饮料走了什么路线？", + "answer_chinese": "我拿着饮料在客厅里转", + "evidence": [ + { + "description": "I was walking around the living room with a drink in my hand", + "timestamps": { + "start": "00:00:31", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You carried it upstairs.", + "B": "You walked straight to the kitchen.", + "C": "You were walking around the living room holding a drink.", + "D": "You turned right and walked a couple of steps to grab the drink, then came back.", + "E": "You turned around and walked backward a few steps to get the drink, then returned forward." + }, + "answer_label": "C", + "answer_text": "You were walking around the living room holding a drink.", + "q_id": "egolife_A1_D4_017_1_T1", + "correct_answer": "You were walking around the living room holding a drink.", + "original_options": { + "A": "You carried it upstairs.", + "B": "You walked straight to the kitchen.", + "C": "You were walking around the living room holding a drink.", + "D": "You turned right and walked a couple of steps to grab the drink, then came back.", + "E": "You turned around and walked backward a few steps to get the drink, then returned forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_017_1_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just went to get a drink?", + "answer": "You carried a drink from the first floor to the second floor.", + "question_chinese": "我刚拿饮料走了什么路线？", + "answer_chinese": "你拿着饮料从一楼到二楼", + "evidence": [ + { + "description": "I was walking around the living room with a drink in my hand", + "timestamps": { + "start": "00:00:31", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:01:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You stayed in the living room.", + "B": "You carried a drink from the first floor to the second floor.", + "C": "You went downstairs for it.", + "D": "You turned right, walked two steps, and grabbed the drink without changing floors.", + "E": "You moved around the living room with the drink and never went up to the second floor." + }, + "answer_label": "B", + "answer_text": "You carried a drink from the first floor to the second floor.", + "q_id": "egolife_A1_D4_017_1_T2", + "correct_answer": "You carried a drink from the first floor to the second floor.", + "original_options": { + "A": "You stayed in the living room.", + "B": "You carried a drink from the first floor to the second floor.", + "C": "You went downstairs for it.", + "D": "You turned right, walked two steps, and grabbed the drink without changing floors.", + "E": "You moved around the living room with the drink and never went up to the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_017_6_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just went to get a drink?", + "answer": "You carried a drink from the first floor to the second floor.", + "question_chinese": "我刚拿饮料走了什么路线？", + "answer_chinese": "你拿着饮料从一楼到二楼", + "evidence": [ + { + "description": "I carried a drink from the first floor to the second floor", + "timestamps": { + "start": "00:01:01", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:01:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You stayed downstairs with it.", + "B": "You just walked two steps right.", + "C": "You carried a drink from the first floor to the second floor.", + "D": "You were walking around the living room holding a drink instead of taking it upstairs.", + "E": "You turned around and walked backward a couple of steps to get the drink, not upstairs." + }, + "answer_label": "C", + "answer_text": "You carried a drink from the first floor to the second floor.", + "q_id": "egolife_A1_D4_017_6_T2", + "correct_answer": "You carried a drink from the first floor to the second floor.", + "original_options": { + "A": "You stayed downstairs with it.", + "B": "You just walked two steps right.", + "C": "You carried a drink from the first floor to the second floor.", + "D": "You were walking around the living room holding a drink instead of taking it upstairs.", + "E": "You turned around and walked backward a couple of steps to get the drink, not upstairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_017_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How many cameras did I point my phone at?", + "answer": "I pointed my phone at the two cameras.", + "question_chinese": "我把手机对准了几个摄像头？", + "answer_chinese": "我把手机对准了2个摄像头", + "evidence": [ + { + "description": "I pointed my mobile phone at one camera", + "timestamps": { + "start": "00:04:41", + "end": "00:04:45" + } + }, + { + "description": "I pointed my mobile phone at one camera", + "timestamps": { + "start": "00:04:45", + "end": "00:04:49" + } + } + ], + "question_timestamps": "00:05:04", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One camera.", + "B": "Three cameras.", + "C": "I pointed my phone at the two cameras.", + "D": "I pointed my phone at four cameras during the scan in one continuous sweep.", + "E": "I pointed my phone at five cameras, moving the phone from one to the next." + }, + "answer_label": "C", + "answer_text": "I pointed my phone at the two cameras.", + "q_id": "egolife_A1_D4_017_4", + "correct_answer": "I pointed my phone at the two cameras.", + "original_options": { + "A": "One camera.", + "B": "Three cameras.", + "C": "I pointed my phone at the two cameras.", + "D": "I pointed my phone at four cameras during the scan in one continuous sweep.", + "E": "I pointed my phone at five cameras, moving the phone from one to the next." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_017_5", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "How many cameras did I point my phone at?", + "answer": "I pointed my phone at the four cameras.", + "question_chinese": "我把手机对准了几个摄像头？", + "answer_chinese": "我把手机对准了4个摄像头", + "evidence": [ + { + "description": "I pointed my phone at the two cameras", + "timestamps": { + "start": "00:04:41", + "end": "00:04:49" + } + } + ], + "question_timestamps": "00:06:53", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "I pointed my phone at two cameras.", + "B": "I pointed my phone at five cameras.", + "C": "I pointed my phone at the one camera for a close-up view.", + "D": "I pointed my phone at the four cameras.", + "E": "I pointed my phone at the three cameras, pausing briefly on each one." + }, + "answer_label": "D", + "answer_text": "I pointed my phone at the four cameras.", + "q_id": "egolife_A1_D4_017_5", + "correct_answer": "I pointed my phone at the four cameras.", + "original_options": { + "A": "I pointed my phone at two cameras.", + "B": "I pointed my phone at five cameras.", + "C": "I pointed my phone at the one camera for a close-up view.", + "D": "I pointed my phone at the four cameras.", + "E": "I pointed my phone at the three cameras, pausing briefly on each one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_017_7", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Am I on the first floor or the second floor now?", + "answer": "I'm on the second floor.", + "question_chinese": "我现在在1楼还是2楼？", + "answer_chinese": "我在2楼", + "evidence": [ + { + "description": "I reached the second floor by stairs", + "timestamps": { + "start": "00:01:01", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "I'm on the second floor.", + "B": "I'm on the first floor." + }, + "answer_label": "A", + "answer_text": "I'm on the second floor.", + "q_id": "egolife_A1_D4_017_7", + "correct_answer": "I'm on the second floor.", + "original_options": { + "A": "I'm on the second floor.", + "B": "I'm on the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D4_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_017_6", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Am I on the first floor or the second floor now?", + "answer": "I'm on the first floor now.", + "question_chinese": "我现在在1楼还是2楼？", + "answer_chinese": "我现在在1楼", + "evidence": [ + { + "description": "I came down from the second floor", + "timestamps": { + "start": "00:04:53", + "end": "00:05:02" + } + }, + { + "description": "I haven't gone upstairs", + "timestamps": { + "start": "00:05:00", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:09:35", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I'm on the second floor now.", + "B": "I'm on the first floor now." + }, + "answer_label": "B", + "answer_text": "I'm on the first floor now.", + "q_id": "egolife_A1_D4_017_6", + "correct_answer": "I'm on the first floor now.", + "original_options": { + "A": "I'm on the second floor now.", + "B": "I'm on the first floor now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D4_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_017_2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs have I moved on the second floor?", + "answer": "I have moved three chairs on the second floor so far.", + "question_chinese": "我目前在二楼搬过几个椅子？", + "answer_chinese": "我目前在二楼搬过3个椅子", + "evidence": [ + { + "description": "I moved the first chair", + "timestamps": { + "start": "00:02:09", + "end": "00:02:12" + } + }, + { + "description": "I moved the second chair", + "timestamps": { + "start": "00:02:16", + "end": "00:02:19" + } + }, + { + "description": "I moved the third chair", + "timestamps": { + "start": "00:02:19", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I moved two chairs upstairs.", + "B": "I moved four chairs upstairs.", + "C": "I have moved three chairs on the second floor so far.", + "D": "I have moved five chairs on the second floor so far, not three.", + "E": "I have moved six chairs on the second floor so far, counting every chair you relocated." + }, + "answer_label": "C", + "answer_text": "I have moved three chairs on the second floor so far.", + "q_id": "egolife_A1_D4_017_2", + "correct_answer": "I have moved three chairs on the second floor so far.", + "original_options": { + "A": "I moved two chairs upstairs.", + "B": "I moved four chairs upstairs.", + "C": "I have moved three chairs on the second floor so far.", + "D": "I have moved five chairs on the second floor so far, not three.", + "E": "I have moved six chairs on the second floor so far, counting every chair you relocated." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_017_3", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs have I moved on the second floor?", + "answer": "I have moved six chairs on the second floor at present.", + "question_chinese": "我目前在二楼搬过几个椅子？", + "answer_chinese": "我目前在2楼搬了6个椅子", + "evidence": [ + { + "description": "I moved the first chair", + "timestamps": { + "start": "00:02:09", + "end": "00:02:12" + } + }, + { + "description": "I moved the second chair", + "timestamps": { + "start": "00:02:16", + "end": "00:02:19" + } + }, + { + "description": "I moved the third chair", + "timestamps": { + "start": "00:02:19", + "end": "00:02:22" + } + }, + { + "description": "I moved the fourth chair", + "timestamps": { + "start": "00:02:46", + "end": "00:02:54" + } + }, + { + "description": "I moved the fifth chair", + "timestamps": { + "start": "00:02:57", + "end": "00:03:03" + } + }, + { + "description": "I moved the sixth chair", + "timestamps": { + "start": "00:03:10", + "end": "00:03:16" + } + } + ], + "question_timestamps": "00:03:33", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I moved five chairs upstairs.", + "B": "I moved seven chairs upstairs.", + "C": "I have moved four chairs on the second floor at present, not six.", + "D": "I have moved six chairs on the second floor at present.", + "E": "I have moved eight chairs on the second floor at present, after moving two more recently." + }, + "answer_label": "D", + "answer_text": "I have moved six chairs on the second floor at present.", + "q_id": "egolife_A1_D4_017_3", + "correct_answer": "I have moved six chairs on the second floor at present.", + "original_options": { + "A": "I moved five chairs upstairs.", + "B": "I moved seven chairs upstairs.", + "C": "I have moved four chairs on the second floor at present, not six.", + "D": "I have moved six chairs on the second floor at present.", + "E": "I have moved eight chairs on the second floor at present, after moving two more recently." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D4_018-mcq.json b/QAs/egolife_A1_D4_018-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3552202ec10bfd875223593f0253c33f3aede3ab --- /dev/null +++ b/QAs/egolife_A1_D4_018-mcq.json @@ -0,0 +1,963 @@ +[ + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018-mcq_fixed_0", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are there in the living room on the first floor?", + "answer": "One.", + "question_chinese": "一楼客厅里有几个人？", + "answer_chinese": "一楼客厅里有6个人", + "evidence": [ + { + "description": "There are five people in my field of vision", + "timestamps": { + "start": "00:01:00", + "end": "00:01:02" + } + }, + { + "description": "I'm here too.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:01:37", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are four people in the living room on the first floor right now.", + "E": "There are five people in the living room on the first floor at the moment." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "egolife_A1_D4_018-mcq_fixed_0", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are four people in the living room on the first floor right now.", + "E": "There are five people in the living room on the first floor at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018_1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in the living room on the first floor?", + "answer": "There are seven people in the living room on the first floor.", + "question_chinese": "一楼客厅里有几个人？", + "answer_chinese": "一楼客厅里有7个人", + "evidence": [ + { + "description": "There are six people in my field of vision", + "timestamps": { + "start": "00:01:00", + "end": "00:01:02" + } + }, + { + "description": "I'm here too.", + "timestamps": { + "start": "00:01:03", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:01:47", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Six people.", + "B": "There are seven people in the living room on the first floor.", + "C": "Eight people.", + "D": "There are five people in the living room on the first floor, not seven.", + "E": "There are nine people in the living room on the first floor when you count everyone present." + }, + "answer_label": "B", + "answer_text": "There are seven people in the living room on the first floor.", + "q_id": "egolife_A1_D4_018_1", + "correct_answer": "There are seven people in the living room on the first floor.", + "original_options": { + "A": "Six people.", + "B": "There are seven people in the living room on the first floor.", + "C": "Eight people.", + "D": "There are five people in the living room on the first floor, not seven.", + "E": "There are nine people in the living room on the first floor when you count everyone present." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018-mcq_fixed_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are there in the living room on the first floor?", + "answer": "Six.", + "question_chinese": "一楼客厅里有几个人？", + "answer_chinese": "一楼客厅里有5个人", + "evidence": [ + { + "description": "There are five people in my field of vision and I'm here too.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:02" + } + }, + { + "description": "I'm leaving.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:05:50", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "There are four people in the living room on the first floor right now.", + "D": "There are eight people in the living room on the first floor at the moment.", + "E": "Six." + }, + "answer_label": "E", + "answer_text": "Six.", + "q_id": "egolife_A1_D4_018-mcq_fixed_1", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "There are four people in the living room on the first floor right now.", + "D": "There are eight people in the living room on the first floor at the moment.", + "E": "Six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in the living room on the first floor?", + "answer": "There are six people in the living room on the first floor.", + "question_chinese": "一楼客厅里有几个人？", + "answer_chinese": "一楼客厅里有6个人", + "evidence": [ + { + "description": "There are six people in my field of vision and I'm here too.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:02" + } + }, + { + "description": "I'm leaving.", + "timestamps": { + "start": "00:01:40", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four people.", + "B": "Seven people.", + "C": "There are six people in the living room on the first floor.", + "D": "There are five people in the living room on the first floor, not six.", + "E": "There are eight people in the living room on the first floor when you count everyone there." + }, + "answer_label": "C", + "answer_text": "There are six people in the living room on the first floor.", + "q_id": "egolife_A1_D4_018_2", + "correct_answer": "There are six people in the living room on the first floor.", + "original_options": { + "A": "Four people.", + "B": "Seven people.", + "C": "There are six people in the living room on the first floor.", + "D": "There are five people in the living room on the first floor, not six.", + "E": "There are eight people in the living room on the first floor when you count everyone there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What do I have in my hand?", + "answer": "The mobile phone I just held in my hand.", + "question_chinese": "我刚手上拿的什么？", + "answer_chinese": "我刚手上拿的手机", + "evidence": [ + { + "description": "The mobile phone I'm holding now", + "timestamps": { + "start": "00:03:36", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:05:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The mobile phone I just held in my hand.", + "B": "The book in my hand.", + "C": "The tablet in my hand.", + "D": "The flashlight I just held in my hand, not a mobile phone.", + "E": "The wallet I just held in my hand instead of the mobile phone." + }, + "answer_label": "A", + "answer_text": "The mobile phone I just held in my hand.", + "q_id": "egolife_A1_D4_018_9", + "correct_answer": "The mobile phone I just held in my hand.", + "original_options": { + "A": "The mobile phone I just held in my hand.", + "B": "The book in my hand.", + "C": "The tablet in my hand.", + "D": "The flashlight I just held in my hand, not a mobile phone.", + "E": "The wallet I just held in my hand instead of the mobile phone." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018_10", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What do I have in my hand?", + "answer": "The mobile phone and power bank I was holding just now.", + "question_chinese": "我刚手上拿的什么？", + "answer_chinese": "我刚才手上拿的手机和充电宝", + "evidence": [ + { + "description": "The mobile phone and power bank I'm holding now", + "timestamps": { + "start": "00:10:20", + "end": "00:10:24" + } + } + ], + "question_timestamps": "00:11:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A tablet and charger.", + "B": "A laptop and mouse.", + "C": "The tablet and headphones I was holding just now, not a mobile phone and power bank.", + "D": "The mobile phone and power bank I was holding just now.", + "E": "The tablet and keyboard I was holding just now instead of the mobile phone and power bank." + }, + "answer_label": "D", + "answer_text": "The mobile phone and power bank I was holding just now.", + "q_id": "egolife_A1_D4_018_10", + "correct_answer": "The mobile phone and power bank I was holding just now.", + "original_options": { + "A": "A tablet and charger.", + "B": "A laptop and mouse.", + "C": "The tablet and headphones I was holding just now, not a mobile phone and power bank.", + "D": "The mobile phone and power bank I was holding just now.", + "E": "The tablet and keyboard I was holding just now instead of the mobile phone and power bank." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018_5", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I went from the first floor to the second floor.", + "question_chinese": "我刚才走了什么路线？", + "answer_chinese": "我从一楼上到二楼", + "evidence": [ + { + "description": "I went from the first floor to the second floor", + "timestamps": { + "start": "00:01:41", + "end": "00:01:54" + } + } + ], + "question_timestamps": "00:02:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Second to first floor.", + "B": "First floor to basement.", + "C": "I stayed on the first floor the whole time, without going up to the second floor.", + "D": "I went from the first floor to the second floor.", + "E": "I went from the first floor to the third floor, skipping the second floor entirely." + }, + "answer_label": "D", + "answer_text": "I went from the first floor to the second floor.", + "q_id": "egolife_A1_D4_018_5", + "correct_answer": "I went from the first floor to the second floor.", + "original_options": { + "A": "Second to first floor.", + "B": "First floor to basement.", + "C": "I stayed on the first floor the whole time, without going up to the second floor.", + "D": "I went from the first floor to the second floor.", + "E": "I went from the first floor to the third floor, skipping the second floor entirely." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I went from the second floor to the living room on the first floor.", + "question_chinese": "我刚才走了什么路线？", + "answer_chinese": "我从二楼到一楼客厅", + "evidence": [ + { + "description": "I went from the second floor to the living room on the first floor", + "timestamps": { + "start": "00:05:56", + "end": "00:06:08" + } + } + ], + "question_timestamps": "00:07:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From first floor to living room.", + "B": "Second floor to kitchen downstairs.", + "C": "I went from the second floor to the living room on the first floor.", + "D": "I stayed on the second floor and went to the living room without going down to the first floor.", + "E": "I went from the third floor to the living room on the first floor, coming down more than one level." + }, + "answer_label": "C", + "answer_text": "I went from the second floor to the living room on the first floor.", + "q_id": "egolife_A1_D4_018_6", + "correct_answer": "I went from the second floor to the living room on the first floor.", + "original_options": { + "A": "From first floor to living room.", + "B": "Second floor to kitchen downstairs.", + "C": "I went from the second floor to the living room on the first floor.", + "D": "I stayed on the second floor and went to the living room without going down to the first floor.", + "E": "I went from the third floor to the living room on the first floor, coming down more than one level." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018-mcq_fixed_9", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get back to my bedroom now?", + "answer": "Turn left, go downstairs, turn left into the living room and then go straight.", + "question_chinese": "我现在要回卧室怎么走？", + "answer_chinese": "左转下楼左转进入客厅然后直走", + "evidence": [ + { + "description": "To the left of where I'm sitting is the door", + "timestamps": { + "start": "00:05:10", + "end": "00:05:11" + } + }, + { + "description": "The bedroom faces the door leading out of the living room", + "timestamps": { + "start": "00:01:00", + "end": "00:01:02" + } + }, + { + "description": "I went out through the door, turned right and went upstairs", + "timestamps": { + "start": "00:01:40", + "end": "00:01:55" + } + } + ], + "question_timestamps": "00:05:24", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, go downstairs, turn left into the living room and then go straight.", + "B": "Turn right, then go downstairs.", + "C": "Go downstairs, then turn right.", + "D": "Turn left, go downstairs, turn right into the living room, and then continue straight ahead.", + "E": "Turn left, go upstairs, turn left into the living room, and then go straight until you reach it." + }, + "answer_label": "A", + "answer_text": "Turn left, go downstairs, turn left into the living room and then go straight.", + "q_id": "egolife_A1_D4_018-mcq_fixed_9", + "correct_answer": "Turn left, go downstairs, turn left into the living room and then go straight.", + "original_options": { + "A": "Turn left, go downstairs, turn left into the living room and then go straight.", + "B": "Turn right, then go downstairs.", + "C": "Go downstairs, then turn right.", + "D": "Turn left, go downstairs, turn right into the living room, and then continue straight ahead.", + "E": "Turn left, go upstairs, turn left into the living room, and then go straight until you reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018-mcq_fixed_8", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get back to my bedroom now?", + "answer": "Walk half a circle around the table and then go straight into the bedroom.", + "question_chinese": "我现在要回卧室怎么走？", + "answer_chinese": "绕桌子走半圈然后直走进入卧室", + "evidence": [ + { + "description": "This door is for the bedroom", + "timestamps": { + "start": "00:07:48", + "end": "00:07:53" + } + }, + { + "description": "The same door is right in front of me", + "timestamps": { + "start": "00:08:59", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:09:06", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk half a circle around the table and then go straight into the bedroom.", + "B": "Walk a full circle around the table.", + "C": "Walk a quarter circle around the table.", + "D": "Walk half a circle around the table and then turn left to enter the bedroom instead of going straight.", + "E": "Walk three-quarters around the table, then angle right toward the bedroom doorway before entering." + }, + "answer_label": "A", + "answer_text": "Walk half a circle around the table and then go straight into the bedroom.", + "q_id": "egolife_A1_D4_018-mcq_fixed_8", + "correct_answer": "Walk half a circle around the table and then go straight into the bedroom.", + "original_options": { + "A": "Walk half a circle around the table and then go straight into the bedroom.", + "B": "Walk a full circle around the table.", + "C": "Walk a quarter circle around the table.", + "D": "Walk half a circle around the table and then turn left to enter the bedroom instead of going straight.", + "E": "Walk three-quarters around the table, then angle right toward the bedroom doorway before entering." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018_3_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see on the first floor?", + "answer": "2, in front of you.", + "question_chinese": "我看到一楼有几个椅子？", + "answer_chinese": "2个，在你的前方。", + "evidence": [ + { + "description": "There are two chairs here", + "timestamps": { + "start": "00:06:11", + "end": "00:06:12" + } + } + ], + "question_timestamps": "00:06:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "1, in front of you.", + "B": "3, in front of you.", + "C": "0, in front of you; you didn’t see any chairs there on the first floor.", + "D": "2, in front of you.", + "E": "4, in front of you; there were four chairs grouped together ahead of you." + }, + "answer_label": "D", + "answer_text": "2, in front of you.", + "q_id": "egolife_A1_D4_018_3_T2", + "correct_answer": "2, in front of you.", + "original_options": { + "A": "1, in front of you.", + "B": "3, in front of you.", + "C": "0, in front of you; you didn’t see any chairs there on the first floor.", + "D": "2, in front of you.", + "E": "4, in front of you; there were four chairs grouped together ahead of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018_3_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see on the first floor?", + "answer": "4, in front of you.", + "question_chinese": "我看到一楼有几个椅子？", + "answer_chinese": "4个，在你的前方。", + "evidence": [ + { + "description": "There are two chairs here", + "timestamps": { + "start": "00:06:11", + "end": "00:06:12" + } + }, + { + "description": "There are two chairs here", + "timestamps": { + "start": "00:09:02", + "end": "00:09:03" + } + } + ], + "question_timestamps": "00:11:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "3, in front of you.", + "B": "5, in front of you.", + "C": "4, in front of you.", + "D": "2, in front of you; you only saw two chairs positioned ahead of you on the first floor.", + "E": "6, in front of you; there were six chairs lined up in front of you there." + }, + "answer_label": "C", + "answer_text": "4, in front of you.", + "q_id": "egolife_A1_D4_018_3_T1", + "correct_answer": "4, in front of you.", + "original_options": { + "A": "3, in front of you.", + "B": "5, in front of you.", + "C": "4, in front of you.", + "D": "2, in front of you; you only saw two chairs positioned ahead of you on the first floor.", + "E": "6, in front of you; there were six chairs lined up in front of you there." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018_2_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see on the second floor?", + "answer": "3, behind you.", + "question_chinese": "我看到二楼有几个椅子？", + "answer_chinese": "3个，在你后方。", + "evidence": [ + { + "description": "There are three chairs here", + "timestamps": { + "start": "00:01:57", + "end": "00:01:58" + } + } + ], + "question_timestamps": "00:02:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "2, behind you.", + "B": "4, behind you.", + "C": "3, behind you.", + "D": "1, behind you; there was only a single chair visible behind you on the second floor.", + "E": "5, behind you; you saw five chairs grouped behind you on the second floor." + }, + "answer_label": "C", + "answer_text": "3, behind you.", + "q_id": "egolife_A1_D4_018_2_T2", + "correct_answer": "3, behind you.", + "original_options": { + "A": "2, behind you.", + "B": "4, behind you.", + "C": "3, behind you.", + "D": "1, behind you; there was only a single chair visible behind you on the second floor.", + "E": "5, behind you; you saw five chairs grouped behind you on the second floor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018_2_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see on the second floor?", + "answer": "8, to your right.", + "question_chinese": "我看到二楼有几个椅子？", + "answer_chinese": "8个，在你右边。", + "evidence": [ + { + "description": "There are three chairs here", + "timestamps": { + "start": "00:01:57", + "end": "00:01:58" + } + }, + { + "description": "There are five chairs here", + "timestamps": { + "start": "00:02:09", + "end": "00:02:11" + } + } + ], + "question_timestamps": "00:05:55", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "7, to your right.", + "B": "8, to your right.", + "C": "9, to your right.", + "D": "6, to your right; only six chairs were visible on the right side on the second floor.", + "E": "10, to your right; you counted ten chairs positioned along the right side there." + }, + "answer_label": "B", + "answer_text": "8, to your right.", + "q_id": "egolife_A1_D4_018_2_T1", + "correct_answer": "8, to your right.", + "original_options": { + "A": "7, to your right.", + "B": "8, to your right.", + "C": "9, to your right.", + "D": "6, to your right; only six chairs were visible on the right side on the second floor.", + "E": "10, to your right; you counted ten chairs positioned along the right side there." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018-mcq_fixed_2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there here?", + "answer": "Four.", + "question_chinese": "这里有几个椅子？", + "answer_chinese": "这里有7个椅子", + "evidence": [ + { + "description": "There are two chairs here", + "timestamps": { + "start": "00:01:57", + "end": "00:01:58" + } + }, + { + "description": "There are five chairs here", + "timestamps": { + "start": "00:02:09", + "end": "00:02:11" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "There are only two chairs visible in this area right now, not four.", + "D": "You can count five chairs in this area, with one tucked behind the table.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "egolife_A1_D4_018-mcq_fixed_2", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "There are only two chairs visible in this area right now, not four.", + "D": "You can count five chairs in this area, with one tucked behind the table.", + "E": "Four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_018-mcq_fixed_3", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there here?", + "answer": "Three.", + "question_chinese": "这里有几个椅子？", + "answer_chinese": "这里有4个椅子", + "evidence": [ + { + "description": "There are two chairs here", + "timestamps": { + "start": "00:06:11", + "end": "00:06:12" + } + }, + { + "description": "There are two chairs here", + "timestamps": { + "start": "00:09:02", + "end": "00:09:03" + } + } + ], + "question_timestamps": "00:11:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "There are five chairs in this spot, including one partially hidden behind the table.", + "D": "Three.", + "E": "You can see six chairs placed here, lined up along the wall beside the seating area." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "egolife_A1_D4_018-mcq_fixed_3", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "There are five chairs in this spot, including one partially hidden behind the table.", + "D": "Three.", + "E": "You can see six chairs placed here, lined up along the wall beside the seating area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D4_019-mcq.json b/QAs/egolife_A1_D4_019-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2685a057f3828450b88c2046f078cb76e5c957ba --- /dev/null +++ b/QAs/egolife_A1_D4_019-mcq.json @@ -0,0 +1,602 @@ +[ + { + "video_uid": "egolife_A1_D4_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_019_1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where was I sitting last time and on which floor?", + "answer": "The last time I sat on the sofa on the second floor.", + "question_chinese": "我上一次坐在几楼的哪里？", + "answer_chinese": "我上一次坐在二楼的沙发", + "evidence": [ + { + "description": "I'm sitting on the sofa", + "timestamps": { + "start": "00:00:35", + "end": "00:00:40" + } + }, + { + "description": "I'm going downstairs.", + "timestamps": { + "start": "00:02:39", + "end": "00:02:49" + } + } + ], + "question_timestamps": "00:02:54", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On a chair downstairs.", + "B": "On the sofa upstairs.", + "C": "The last time I sat on a chair on the second floor, not the sofa.", + "D": "The last time I sat on the sofa on the first floor instead of the second.", + "E": "The last time I sat on the sofa on the second floor." + }, + "answer_label": "E", + "answer_text": "The last time I sat on the sofa on the second floor.", + "q_id": "egolife_A1_D4_019_1", + "correct_answer": "The last time I sat on the sofa on the second floor.", + "original_options": { + "A": "On a chair downstairs.", + "B": "On the sofa upstairs.", + "C": "The last time I sat on a chair on the second floor, not the sofa.", + "D": "The last time I sat on the sofa on the first floor instead of the second.", + "E": "The last time I sat on the sofa on the second floor." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_019_2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where was I sitting last time and on which floor?", + "answer": "The last time I sat on the chair on the first floor.", + "question_chinese": "我上一次坐在几楼的哪里？", + "answer_chinese": "我上一次坐在一楼的椅子上", + "evidence": [ + { + "description": "I'm going downstairs.", + "timestamps": { + "start": "00:02:39", + "end": "00:02:49" + } + }, + { + "description": "I'm sitting on a chair", + "timestamps": { + "start": "00:02:55", + "end": "00:02:57" + } + } + ], + "question_timestamps": "00:07:57", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The last time I sat on the chair on the first floor.", + "B": "On the sofa upstairs.", + "C": "On the bench downstairs.", + "D": "The last time I sat on the chair on the second floor, not the first.", + "E": "The last time I sat on a stool on the first floor instead of a chair." + }, + "answer_label": "A", + "answer_text": "The last time I sat on the chair on the first floor.", + "q_id": "egolife_A1_D4_019_2", + "correct_answer": "The last time I sat on the chair on the first floor.", + "original_options": { + "A": "The last time I sat on the chair on the first floor.", + "B": "On the sofa upstairs.", + "C": "On the bench downstairs.", + "D": "The last time I sat on the chair on the second floor, not the first.", + "E": "The last time I sat on a stool on the first floor instead of a chair." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_019_6", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Am I on the first floor or the second floor now?", + "answer": "I'm on the first floor now.", + "question_chinese": "我现在在一楼还是二楼？", + "answer_chinese": "我现在在一楼", + "evidence": [ + { + "description": "I'm going downstairs.", + "timestamps": { + "start": "00:02:39", + "end": "00:02:49" + } + } + ], + "question_timestamps": "00:08:03", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "I'm on the second floor now.", + "B": "I'm on the first floor now." + }, + "answer_label": "B", + "answer_text": "I'm on the first floor now.", + "q_id": "egolife_A1_D4_019_6", + "correct_answer": "I'm on the first floor now.", + "original_options": { + "A": "I'm on the second floor now.", + "B": "I'm on the first floor now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D4_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_019_5", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Am I on the first floor or the second floor now?", + "answer": "I'm on the second floor now.", + "question_chinese": "我现在在一楼还是二楼？", + "answer_chinese": "我现在在二楼", + "evidence": [ + { + "description": "I'm going upstairs.", + "timestamps": { + "start": "00:09:09", + "end": "00:09:16" + } + } + ], + "question_timestamps": "00:09:50", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "I'm on the second floor now.", + "B": "I'm on the first floor now." + }, + "answer_label": "A", + "answer_text": "I'm on the second floor now.", + "q_id": "egolife_A1_D4_019_5", + "correct_answer": "I'm on the second floor now.", + "original_options": { + "A": "I'm on the second floor now.", + "B": "I'm on the first floor now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D4_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_019_9", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I touch the cutting board without going through the door?", + "answer": "I can touch the cutting board without going through the door.", + "question_chinese": "我能不经过门就碰到菜板嘛", + "answer_chinese": "我能不经过门就碰到菜板", + "evidence": [ + { + "description": "There is a cutting board on the second floor", + "timestamps": { + "start": "00:09:21", + "end": "00:09:24" + } + }, + { + "description": "I'm going upstairs.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:14" + } + }, + { + "description": "I have already passed through the only door", + "timestamps": { + "start": "00:09:14", + "end": "00:09:17" + } + } + ], + "question_timestamps": "00:09:50", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "I can touch the cutting board without going through the door.", + "B": "I can't touch the cutting board without going through the door." + }, + "answer_label": "A", + "answer_text": "I can touch the cutting board without going through the door.", + "q_id": "egolife_A1_D4_019_9", + "correct_answer": "I can touch the cutting board without going through the door.", + "original_options": { + "A": "I can touch the cutting board without going through the door.", + "B": "I can't touch the cutting board without going through the door." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D4_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_019_10", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I touch the cutting board without going through the door?", + "answer": "I couldn't help bumping into the cutting board without passing through the door.", + "question_chinese": "我能不经过门就碰到菜板嘛", + "answer_chinese": "我不能不经过门就碰到菜板", + "evidence": [ + { + "description": "There is a cutting board on the first floor", + "timestamps": { + "start": "00:08:31", + "end": "00:08:33" + } + }, + { + "description": "I'm in the yard", + "timestamps": { + "start": "00:10:54", + "end": "00:10:58" + } + }, + { + "description": "There is a door between the interior and the yard", + "timestamps": { + "start": "00:10:52", + "end": "00:10:55" + } + } + ], + "question_timestamps": "00:11:53", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "I can touch the cutting board without going through the door.", + "B": "I couldn't help bumping into the cutting board without passing through the door." + }, + "answer_label": "B", + "answer_text": "I couldn't help bumping into the cutting board without passing through the door.", + "q_id": "egolife_A1_D4_019_10", + "correct_answer": "I couldn't help bumping into the cutting board without passing through the door.", + "original_options": { + "A": "I can touch the cutting board without going through the door.", + "B": "I couldn't help bumping into the cutting board without passing through the door." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D4_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_019_8", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route should I take to get here?", + "answer": "I got here from the kitchen on the first floor through the living room and the stairs.", + "question_chinese": "我走什么路线到达这里？", + "answer_chinese": "我从一楼的厨房经过客厅和楼梯到达这里", + "evidence": [ + { + "description": "I got here from the kitchen on the first floor through the living room and the stairs", + "timestamps": { + "start": "00:08:50", + "end": "00:09:23" + } + } + ], + "question_timestamps": "00:09:47", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I came from the living room downstairs.", + "B": "I got here from the first-floor kitchen.", + "C": "I got here from the kitchen on the first floor through the living room and the stairs.", + "D": "I got here from the kitchen on the first floor through the hallway and then up the stairs.", + "E": "I got here from the first floor by taking the stairs first and then crossing the living room." + }, + "answer_label": "C", + "answer_text": "I got here from the kitchen on the first floor through the living room and the stairs.", + "q_id": "egolife_A1_D4_019_8", + "correct_answer": "I got here from the kitchen on the first floor through the living room and the stairs.", + "original_options": { + "A": "I came from the living room downstairs.", + "B": "I got here from the first-floor kitchen.", + "C": "I got here from the kitchen on the first floor through the living room and the stairs.", + "D": "I got here from the kitchen on the first floor through the hallway and then up the stairs.", + "E": "I got here from the first floor by taking the stairs first and then crossing the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_019_7", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route should I take to get here?", + "answer": "I got here from the second floor via the stairs and the living room.", + "question_chinese": "我走什么路线到达这里？", + "answer_chinese": "我从二楼经过楼梯和客厅到达这里", + "evidence": [ + { + "description": "I got here from the second floor via the stairs and the living room", + "timestamps": { + "start": "00:09:45", + "end": "00:10:01" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I got here from the first floor.", + "B": "I got here from the second floor via the stairs and the living room.", + "C": "I got here via the hallway and stairs.", + "D": "I got here from the second floor via the stairs and the hallway, not through the living room.", + "E": "I got here from the third floor via the stairs and then passed through the living room." + }, + "answer_label": "B", + "answer_text": "I got here from the second floor via the stairs and the living room.", + "q_id": "egolife_A1_D4_019_7", + "correct_answer": "I got here from the second floor via the stairs and the living room.", + "original_options": { + "A": "I got here from the first floor.", + "B": "I got here from the second floor via the stairs and the living room.", + "C": "I got here via the hallway and stairs.", + "D": "I got here from the second floor via the stairs and the hallway, not through the living room.", + "E": "I got here from the third floor via the stairs and then passed through the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_019_3", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "\"Who am I closest to right now?\"", + "answer": "Now I'm closest to the boy in black.", + "question_chinese": "现在我离谁最近？", + "answer_chinese": "现在我离黑衣服男生最近", + "evidence": [ + { + "description": "Now I'm closest to the boy in black", + "timestamps": { + "start": "00:03:30", + "end": "00:03:35" + } + } + ], + "question_timestamps": "00:05:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Now I'm closest to the girl in red.", + "B": "Now I'm closest to the girl in blue.", + "C": "Now I'm closest to the girl in green, standing nearest to you right now.", + "D": "Now I'm closest to the boy in black.", + "E": "Now I'm closest to the girl in yellow, not the boy in black." + }, + "answer_label": "D", + "answer_text": "Now I'm closest to the boy in black.", + "q_id": "egolife_A1_D4_019_3", + "correct_answer": "Now I'm closest to the boy in black.", + "original_options": { + "A": "Now I'm closest to the girl in red.", + "B": "Now I'm closest to the girl in blue.", + "C": "Now I'm closest to the girl in green, standing nearest to you right now.", + "D": "Now I'm closest to the boy in black.", + "E": "Now I'm closest to the girl in yellow, not the boy in black." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_019_4", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "\"Who am I closest to right now?\"", + "answer": "Now I'm closest to the girl in pink.", + "question_chinese": "现在我离谁最近？", + "answer_chinese": "现在我离粉衣服女生最近", + "evidence": [ + { + "description": "Now I'm closest to the girl in pink", + "timestamps": { + "start": "00:10:10", + "end": "00:10:11" + } + } + ], + "question_timestamps": "00:10:25", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Now I'm closest to the girl in blue.", + "B": "Now I'm closest to the girl in yellow.", + "C": "Now I'm closest to the girl in pink.", + "D": "Now I'm closest to the girl in white, who is nearest to you right now.", + "E": "Now I'm closest to the girl in green, not the girl in pink." + }, + "answer_label": "C", + "answer_text": "Now I'm closest to the girl in pink.", + "q_id": "egolife_A1_D4_019_4", + "correct_answer": "Now I'm closest to the girl in pink.", + "original_options": { + "A": "Now I'm closest to the girl in blue.", + "B": "Now I'm closest to the girl in yellow.", + "C": "Now I'm closest to the girl in pink.", + "D": "Now I'm closest to the girl in white, who is nearest to you right now.", + "E": "Now I'm closest to the girl in green, not the girl in pink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D4_020-mcq.json b/QAs/egolife_A1_D4_020-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8b0ee9b035966fdb54aa1fe418475f0d1dc5e325 --- /dev/null +++ b/QAs/egolife_A1_D4_020-mcq.json @@ -0,0 +1,837 @@ +[ + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_6_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I walk forward, will I get closer to the main gate or farther away from it?", + "answer": "If you walk forward, you’ll be farther from the main entrance.", + "question_chinese": "如果我往前走，是离大门更近还是更远？", + "answer_chinese": "如果我往前走，是离大门更远", + "evidence": [ + { + "description": "I'm sitting on a bench", + "timestamps": { + "start": "00:07:00", + "end": "00:07:16" + } + }, + { + "description": "The stool I was sitting on was facing away from the gate", + "timestamps": { + "start": "00:02:10", + "end": "00:02:11" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You'll be closer to the main entrance.", + "B": "You'll stay the same distance.", + "C": "If you walk forward, you’ll be closer to the main entrance because it’s straight ahead of you.", + "D": "If you walk forward, you’ll be farther from the main entrance.", + "E": "If you walk forward, you’ll get farther from the front gate because it’s behind you." + }, + "answer_label": "D", + "answer_text": "If you walk forward, you’ll be farther from the main entrance.", + "q_id": "egolife_A1_D4_020_6_T1", + "correct_answer": "If you walk forward, you’ll be farther from the main entrance.", + "original_options": { + "A": "You'll be closer to the main entrance.", + "B": "You'll stay the same distance.", + "C": "If you walk forward, you’ll be closer to the main entrance because it’s straight ahead of you.", + "D": "If you walk forward, you’ll be farther from the main entrance.", + "E": "If you walk forward, you’ll get farther from the front gate because it’s behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_7_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I walk forward, will I get closer to the main gate or farther away from it?", + "answer": "If you walk forward, you’ll be closer to the front gate; the gate is to your right.", + "question_chinese": "如果我往前走，是离大门更近还是更远？", + "answer_chinese": "如果我往前走，是离大门更近，大门在我右边", + "evidence": [ + { + "description": "I face the gate.", + "timestamps": { + "start": "00:09:22", + "end": "00:09:29" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "If you walk forward, you’ll be closer to the front gate; the gate is to your right.", + "B": "You'll be closer; the gate is left.", + "C": "You'll be farther from the main entrance.", + "D": "If you walk forward, you’ll be closer to the front gate, but it’s in front of you.", + "E": "If you walk forward, you’ll be closer to the back gate even though it’s to your right." + }, + "answer_label": "A", + "answer_text": "If you walk forward, you’ll be closer to the front gate; the gate is to your right.", + "q_id": "egolife_A1_D4_020_7_T2", + "correct_answer": "If you walk forward, you’ll be closer to the front gate; the gate is to your right.", + "original_options": { + "A": "If you walk forward, you’ll be closer to the front gate; the gate is to your right.", + "B": "You'll be closer; the gate is left.", + "C": "You'll be farther from the main entrance.", + "D": "If you walk forward, you’ll be closer to the front gate, but it’s in front of you.", + "E": "If you walk forward, you’ll be closer to the back gate even though it’s to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_7_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "If I walk forward, will I get closer to the main gate or farther away from it?", + "answer": "If you walk forward, you’ll be closer to the front gate; the gate is in front of you.", + "question_chinese": "如果我往前走，是离大门更近还是更远？", + "answer_chinese": "如果我往前走，是离大门更近，大门在我前面", + "evidence": [ + { + "description": "I face the gate.", + "timestamps": { + "start": "00:09:22", + "end": "00:09:29" + } + } + ], + "question_timestamps": "00:10:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You'll be farther from the main entrance.", + "B": "You'll be closer; the gate is right.", + "C": "If you walk forward, you’ll be closer to the front gate, but the gate is behind you.", + "D": "If you walk forward, you’ll be closer to the back gate even though it’s in front of you.", + "E": "If you walk forward, you’ll be closer to the front gate; the gate is in front of you." + }, + "answer_label": "E", + "answer_text": "If you walk forward, you’ll be closer to the front gate; the gate is in front of you.", + "q_id": "egolife_A1_D4_020_7_T1", + "correct_answer": "If you walk forward, you’ll be closer to the front gate; the gate is in front of you.", + "original_options": { + "A": "You'll be farther from the main entrance.", + "B": "You'll be closer; the gate is right.", + "C": "If you walk forward, you’ll be closer to the front gate, but the gate is behind you.", + "D": "If you walk forward, you’ll be closer to the back gate even though it’s in front of you.", + "E": "If you walk forward, you’ll be closer to the front gate; the gate is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I was just walking straight in the alley.", + "question_chinese": "我刚走了什么了路线", + "answer_chinese": "我刚在小巷子里直走", + "evidence": [ + { + "description": "I was just walking straight in the alley", + "timestamps": { + "start": "00:00:00", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You turned left in the alley.", + "B": "I was just walking straight in the alley.", + "C": "I was just walking in a hallway.", + "D": "You walked from the alley out onto the street instead of continuing straight.", + "E": "You turned right first and then continued walking straight for the rest of the way." + }, + "answer_label": "B", + "answer_text": "I was just walking straight in the alley.", + "q_id": "egolife_A1_D4_020_1", + "correct_answer": "I was just walking straight in the alley.", + "original_options": { + "A": "You turned left in the alley.", + "B": "I was just walking straight in the alley.", + "C": "I was just walking in a hallway.", + "D": "You walked from the alley out onto the street instead of continuing straight.", + "E": "You turned right first and then continued walking straight for the rest of the way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I had just stepped back two steps to the chair and sat down.", + "question_chinese": "我刚走了什么了路线", + "answer_chinese": "我刚后退两步走到椅子前坐下", + "evidence": [ + { + "description": "I had just stepped back two steps to the chair and sat down", + "timestamps": { + "start": "00:03:55", + "end": "00:04:04" + } + } + ], + "question_timestamps": "00:06:33", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I stepped forward to the chair.", + "B": "I stepped back one step and sat.", + "C": "I walked back three steps to the chair, then sat down after reaching it.", + "D": "I stepped back two steps toward the chair but stayed standing instead of sitting down.", + "E": "I had just stepped back two steps to the chair and sat down." + }, + "answer_label": "E", + "answer_text": "I had just stepped back two steps to the chair and sat down.", + "q_id": "egolife_A1_D4_020_2", + "correct_answer": "I had just stepped back two steps to the chair and sat down.", + "original_options": { + "A": "I stepped forward to the chair.", + "B": "I stepped back one step and sat.", + "C": "I walked back three steps to the chair, then sat down after reaching it.", + "D": "I stepped back two steps toward the chair but stayed standing instead of sitting down.", + "E": "I had just stepped back two steps to the chair and sat down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_5", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the girl in the red dress or the girl in the brown coat?", + "answer": "I'm now close to the girl in the coffee-colored coat.", + "question_chinese": "我现在是离红衣服女生近还是离咖色外套的女生近？", + "answer_chinese": "我现在离咖色外套的女生近", + "evidence": [ + { + "description": "I walked towards the direction of the girl in the coffee-colored coat", + "timestamps": { + "start": "00:02:49", + "end": "00:02:55" + } + } + ], + "question_timestamps": "00:03:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "I'm now close to the girl in the coffee-colored coat.", + "B": "I'm now close to the girl in the red dress." + }, + "answer_label": "A", + "answer_text": "I'm now close to the girl in the coffee-colored coat.", + "q_id": "egolife_A1_D4_020_5", + "correct_answer": "I'm now close to the girl in the coffee-colored coat.", + "original_options": { + "A": "I'm now close to the girl in the coffee-colored coat.", + "B": "I'm now close to the girl in the red dress." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_6", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the girl in the red dress or the girl in the brown coat?", + "answer": "I'm now closest to the girl in red.", + "question_chinese": "我现在是离红衣服女生近还是离咖色外套的女生近？", + "answer_chinese": "我现在是离红衣服女生近", + "evidence": [ + { + "description": "I walked behind the girl in red", + "timestamps": { + "start": "00:09:20", + "end": "00:09:25" + } + } + ], + "question_timestamps": "00:10:02", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "I'm now closest to the girl in the brown coat.", + "B": "I'm now closest to the girl in red." + }, + "answer_label": "B", + "answer_text": "I'm now closest to the girl in red.", + "q_id": "egolife_A1_D4_020_6", + "correct_answer": "I'm now closest to the girl in red.", + "original_options": { + "A": "I'm now closest to the girl in the brown coat.", + "B": "I'm now closest to the girl in red." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_5_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in the yard right now?", + "answer": "There are 5 people in the yard now.", + "question_chinese": "我现在院子有几个人？", + "answer_chinese": "现在院子有5个人", + "evidence": [ + { + "description": "There are four people in my field of vision", + "timestamps": { + "start": "00:06:57", + "end": "00:06:58" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 4 people.", + "B": "There are 6 people.", + "C": "There are 5 people in the yard now.", + "D": "There are 3 people in the yard right now, not counting anyone outside the yard.", + "E": "There are 7 people in the yard right now, with two standing together near the entrance." + }, + "answer_label": "C", + "answer_text": "There are 5 people in the yard now.", + "q_id": "egolife_A1_D4_020_5_T1", + "correct_answer": "There are 5 people in the yard now.", + "original_options": { + "A": "There are 4 people.", + "B": "There are 6 people.", + "C": "There are 5 people in the yard now.", + "D": "There are 3 people in the yard right now, not counting anyone outside the yard.", + "E": "There are 7 people in the yard right now, with two standing together near the entrance." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_4_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in the yard right now?", + "answer": "There are 7 people in your yard now.", + "question_chinese": "我现在院子有几个人？", + "answer_chinese": "你现在院子有7个人", + "evidence": [ + { + "description": "There are two people in my field of vision", + "timestamps": { + "start": "00:11:00", + "end": "00:11:51" + } + }, + { + "description": "There are four people in my field of vision and I'm also in the yard", + "timestamps": { + "start": "00:10:46", + "end": "00:11:00" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 6 people.", + "B": "There are 8 people.", + "C": "There are 7 people in your yard now.", + "D": "There are 5 people in your yard right now, with a smaller group gathered near the middle.", + "E": "There are 9 people in your yard right now, with several clustered together by the edge." + }, + "answer_label": "C", + "answer_text": "There are 7 people in your yard now.", + "q_id": "egolife_A1_D4_020_4_T2", + "correct_answer": "There are 7 people in your yard now.", + "original_options": { + "A": "There are 6 people.", + "B": "There are 8 people.", + "C": "There are 7 people in your yard now.", + "D": "There are 5 people in your yard right now, with a smaller group gathered near the middle.", + "E": "There are 9 people in your yard right now, with several clustered together by the edge." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_4_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in the yard right now?", + "answer": "There are 7 people in the yard now.", + "question_chinese": "我现在院子有几个人？", + "answer_chinese": "现在院子有7个人", + "evidence": [ + { + "description": "There are two people in my field of vision", + "timestamps": { + "start": "00:11:00", + "end": "00:11:51" + } + }, + { + "description": "There are four people in my field of vision and I'm also in the yard", + "timestamps": { + "start": "00:10:46", + "end": "00:11:00" + } + } + ], + "question_timestamps": "00:12:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 6 people.", + "B": "There are 7 people in the yard now.", + "C": "There are 8 people.", + "D": "There are 5 people in the yard now, with just a small group visible in front of you.", + "E": "There are 9 people in the yard now, including a couple standing slightly behind the others." + }, + "answer_label": "B", + "answer_text": "There are 7 people in the yard now.", + "q_id": "egolife_A1_D4_020_4_T1", + "correct_answer": "There are 7 people in the yard now.", + "original_options": { + "A": "There are 6 people.", + "B": "There are 7 people in the yard now.", + "C": "There are 8 people.", + "D": "There are 5 people in the yard now, with just a small group visible in front of you.", + "E": "There are 9 people in the yard now, including a couple standing slightly behind the others." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_3", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I petted a dog so far?", + "answer": "I have touched the dog three times so far.", + "question_chinese": "我目前为止摸了几次狗？", + "answer_chinese": "我目前为止摸了3次狗", + "evidence": [ + { + "description": "I touched the dog once", + "timestamps": { + "start": "00:03:25", + "end": "00:03:27" + } + }, + { + "description": "I touched the dog once", + "timestamps": { + "start": "00:05:04", + "end": "00:05:18" + } + }, + { + "description": "I touched the dog once", + "timestamps": { + "start": "00:05:37", + "end": "00:05:41" + } + } + ], + "question_timestamps": "00:06:33", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I touched it twice.", + "B": "I touched it four times.", + "C": "I have touched the dog five times so far, with each pet counted as a separate touch.", + "D": "I have touched the dog three times so far.", + "E": "I have touched the dog six times so far, including the quick pats you did earlier." + }, + "answer_label": "D", + "answer_text": "I have touched the dog three times so far.", + "q_id": "egolife_A1_D4_020_3", + "correct_answer": "I have touched the dog three times so far.", + "original_options": { + "A": "I touched it twice.", + "B": "I touched it four times.", + "C": "I have touched the dog five times so far, with each pet counted as a separate touch.", + "D": "I have touched the dog three times so far.", + "E": "I have touched the dog six times so far, including the quick pats you did earlier." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_4", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I petted a dog so far?", + "answer": "I have touched the dog four times so far.", + "question_chinese": "我目前为止摸了几次狗？", + "answer_chinese": "我目前为止摸了4次狗", + "evidence": [ + { + "description": "I touched the dog once", + "timestamps": { + "start": "00:03:25", + "end": "00:03:27" + } + }, + { + "description": "I touched the dog once", + "timestamps": { + "start": "00:05:04", + "end": "00:05:18" + } + }, + { + "description": "I touched the dog once", + "timestamps": { + "start": "00:05:37", + "end": "00:05:41" + } + }, + { + "description": "I touched the dog once", + "timestamps": { + "start": "00:07:08", + "end": "00:07:15" + } + } + ], + "question_timestamps": "00:07:33", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I touched it three times.", + "B": "I touched it five times.", + "C": "I have touched the dog one time so far, counting only the single clear pet you gave it.", + "D": "I have touched the dog two times so far, including the brief second touch right after the first.", + "E": "I have touched the dog four times so far." + }, + "answer_label": "E", + "answer_text": "I have touched the dog four times so far.", + "q_id": "egolife_A1_D4_020_4", + "correct_answer": "I have touched the dog four times so far.", + "original_options": { + "A": "I touched it three times.", + "B": "I touched it five times.", + "C": "I have touched the dog one time so far, counting only the single clear pet you gave it.", + "D": "I have touched the dog two times so far, including the brief second touch right after the first.", + "E": "I have touched the dog four times so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_8", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in the yard right now?", + "answer": "There are five people in the yard now.", + "question_chinese": "现在院子有几个人？", + "answer_chinese": "现在院子有5个人", + "evidence": [ + { + "description": "There are four people in my field of vision", + "timestamps": { + "start": "00:06:57", + "end": "00:06:58" + } + } + ], + "question_timestamps": "00:07:30", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are five people in the yard now.", + "B": "There are four people in the yard.", + "C": "There are six people in the yard.", + "D": "There are three people in the yard right now, standing near each other.", + "E": "There are seven people in the yard now, with a couple more gathered by the side." + }, + "answer_label": "A", + "answer_text": "There are five people in the yard now.", + "q_id": "egolife_A1_D4_020_8", + "correct_answer": "There are five people in the yard now.", + "original_options": { + "A": "There are five people in the yard now.", + "B": "There are four people in the yard.", + "C": "There are six people in the yard.", + "D": "There are three people in the yard right now, standing near each other.", + "E": "There are seven people in the yard now, with a couple more gathered by the side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_020_7", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in the yard now?", + "answer": "There are seven people in the yard now.", + "question_chinese": "现在院子有几个人？", + "answer_chinese": "现在院子有7个人", + "evidence": [ + { + "description": "There are two people in my field of vision", + "timestamps": { + "start": "00:11:00", + "end": "00:11:51" + } + }, + { + "description": "There are four people in my field of vision and I'm also in the yard", + "timestamps": { + "start": "00:10:46", + "end": "00:11:00" + } + } + ], + "question_timestamps": "00:12:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are seven people in the yard now.", + "B": "There are six people in the yard.", + "C": "There are eight people in the yard.", + "D": "There are five people in the yard now, with only a small group present.", + "E": "There are nine people in the yard right now, with two more than you expected." + }, + "answer_label": "A", + "answer_text": "There are seven people in the yard now.", + "q_id": "egolife_A1_D4_020_7", + "correct_answer": "There are seven people in the yard now.", + "original_options": { + "A": "There are seven people in the yard now.", + "B": "There are six people in the yard.", + "C": "There are eight people in the yard.", + "D": "There are five people in the yard now, with only a small group present.", + "E": "There are nine people in the yard right now, with two more than you expected." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D4_021-mcq.json b/QAs/egolife_A1_D4_021-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9a6c84e2c43f866e39725da0df526ae99f5bff15 --- /dev/null +++ b/QAs/egolife_A1_D4_021-mcq.json @@ -0,0 +1,647 @@ +[ + { + "video_uid": "egolife_A1_D4_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_021_9", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Who was the last person to pet the dog?", + "answer": "The last person who petted a dog was a boy in black with grey hair.", + "question_chinese": "上一个撸狗的人是谁？", + "answer_chinese": "上一个撸狗的人是黑衣服灰头发的男生", + "evidence": [ + { + "description": "A boy in black with grey hair is petting a dog", + "timestamps": { + "start": "00:04:51", + "end": "00:04:53" + } + } + ], + "question_timestamps": "00:05:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A man in black with grey hair.", + "B": "A boy in red with grey hair.", + "C": "The last person who petted the dog was a girl in black with grey hair.", + "D": "The last person who petted a dog was a boy in black with grey hair.", + "E": "The last person who petted the dog was a girl in red with dark hair." + }, + "answer_label": "D", + "answer_text": "The last person who petted a dog was a boy in black with grey hair.", + "q_id": "egolife_A1_D4_021_9", + "correct_answer": "The last person who petted a dog was a boy in black with grey hair.", + "original_options": { + "A": "A man in black with grey hair.", + "B": "A boy in red with grey hair.", + "C": "The last person who petted the dog was a girl in black with grey hair.", + "D": "The last person who petted a dog was a boy in black with grey hair.", + "E": "The last person who petted the dog was a girl in red with dark hair." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_021_10", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Who was the last person to pet the dog?", + "answer": "The last person who petted the dog was a girl in a blue coat.", + "question_chinese": "上一个撸狗的人是谁？", + "answer_chinese": "上一个撸狗的人是蓝外套的女生", + "evidence": [ + { + "description": "The girl in the blue coat is petting a dog", + "timestamps": { + "start": "00:05:23", + "end": "00:05:24" + } + } + ], + "question_timestamps": "00:07:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A boy in a blue coat.", + "B": "A girl in a yellow coat.", + "C": "The last person who petted the dog was a boy wearing a red coat.", + "D": "The last person who petted the dog was a boy in a green coat.", + "E": "The last person who petted the dog was a girl in a blue coat." + }, + "answer_label": "E", + "answer_text": "The last person who petted the dog was a girl in a blue coat.", + "q_id": "egolife_A1_D4_021_10", + "correct_answer": "The last person who petted the dog was a girl in a blue coat.", + "original_options": { + "A": "A boy in a blue coat.", + "B": "A girl in a yellow coat.", + "C": "The last person who petted the dog was a boy wearing a red coat.", + "D": "The last person who petted the dog was a boy in a green coat.", + "E": "The last person who petted the dog was a girl in a blue coat." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_021_7", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the main entrance?", + "answer": "The gate is on my left.", + "question_chinese": "大门在我哪个方向？", + "answer_chinese": "大门在我左边", + "evidence": [ + { + "description": "The door is in my view", + "timestamps": { + "start": "00:02:57", + "end": "00:03:00" + } + }, + { + "description": "I'm turning my head left", + "timestamps": { + "start": "00:02:55", + "end": "00:02:56" + } + } + ], + "question_timestamps": "00:04:18", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The gate is on my right.", + "B": "The gate is on my left.", + "C": "The gate is in front of me.", + "D": "The gate is directly behind me, opposite where I am facing right now.", + "E": "The gate is on my far right, slightly past the side of the yard." + }, + "answer_label": "B", + "answer_text": "The gate is on my left.", + "q_id": "egolife_A1_D4_021_7", + "correct_answer": "The gate is on my left.", + "original_options": { + "A": "The gate is on my right.", + "B": "The gate is on my left.", + "C": "The gate is in front of me.", + "D": "The gate is directly behind me, opposite where I am facing right now.", + "E": "The gate is on my far right, slightly past the side of the yard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_021_8", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the main entrance?", + "answer": "The gate is behind me.", + "question_chinese": "大门在我哪个方向？", + "answer_chinese": "大门在我后面", + "evidence": [ + { + "description": "I'm turning my head back", + "timestamps": { + "start": "00:07:14", + "end": "00:07:18" + } + }, + { + "description": "The gate is in my view", + "timestamps": { + "start": "00:07:18", + "end": "00:07:19" + } + } + ], + "question_timestamps": "00:07:28", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The gate is behind me.", + "B": "The gate is in front of me.", + "C": "The gate is to my left.", + "D": "The gate is to my right, off to the side as I face forward.", + "E": "The gate is beside me, not behind me, but near my shoulder right now." + }, + "answer_label": "A", + "answer_text": "The gate is behind me.", + "q_id": "egolife_A1_D4_021_8", + "correct_answer": "The gate is behind me.", + "original_options": { + "A": "The gate is behind me.", + "B": "The gate is in front of me.", + "C": "The gate is to my left.", + "D": "The gate is to my right, off to the side as I face forward.", + "E": "The gate is beside me, not behind me, but near my shoulder right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_021-mcq_fixed_8", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are there in the place where I am?", + "answer": "Five.", + "question_chinese": "我所在的地方有几个人？", + "answer_chinese": "5", + "evidence": [ + { + "description": "There are four people in the field of vision", + "timestamps": { + "start": "00:01:00", + "end": "00:01:02" + } + }, + { + "description": "I'm here too", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:59", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "There are three people in the place where I am right now.", + "D": "Five.", + "E": "There are seven people in the place where I am at the moment." + }, + "answer_label": "D", + "answer_text": "Five.", + "q_id": "egolife_A1_D4_021-mcq_fixed_8", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "There are three people in the place where I am right now.", + "D": "Five.", + "E": "There are seven people in the place where I am at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_021-mcq_fixed_9", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are there in the place where I am?", + "answer": "Three.", + "question_chinese": "我所在的地方有几个人？", + "answer_chinese": "7", + "evidence": [ + { + "description": "There are six people in the field of vision", + "timestamps": { + "start": "00:02:53", + "end": "00:03:05" + } + }, + { + "description": "I'm here too", + "timestamps": { + "start": "00:03:05", + "end": "00:03:07" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "There are five people in the place where I am right now.", + "D": "Three.", + "E": "There are six people where I am, with a larger group than three." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "egolife_A1_D4_021-mcq_fixed_9", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "There are five people in the place where I am right now.", + "D": "Three.", + "E": "There are six people where I am, with a larger group than three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_021_6", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "\"Who am I closest to right now?\"", + "answer": "I'm closest to the one in black clothes and with grey hair now.", + "question_chinese": "我现在离谁最近", + "answer_chinese": "我现在离黑衣服灰头发的最近", + "evidence": [ + { + "description": "I'm closest to the one in black clothes and with grey hair now", + "timestamps": { + "start": "00:00:27", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I'm closest to the one with brown hair.", + "B": "I'm closest to the one with blonde hair.", + "C": "I'm closest to the one in black clothes and with grey hair now.", + "D": "I'm closest to the one in black clothes with black hair, standing nearest to me right now.", + "E": "I'm closest to the one in black clothes with red hair, not the grey-haired person beside them." + }, + "answer_label": "C", + "answer_text": "I'm closest to the one in black clothes and with grey hair now.", + "q_id": "egolife_A1_D4_021_6", + "correct_answer": "I'm closest to the one in black clothes and with grey hair now.", + "original_options": { + "A": "I'm closest to the one with brown hair.", + "B": "I'm closest to the one with blonde hair.", + "C": "I'm closest to the one in black clothes and with grey hair now.", + "D": "I'm closest to the one in black clothes with black hair, standing nearest to me right now.", + "E": "I'm closest to the one in black clothes with red hair, not the grey-haired person beside them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_021_5", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "\"Who am I closest to right now?\"", + "answer": "I'm closest to the girl in red now.", + "question_chinese": "我现在离谁最近", + "answer_chinese": "我现在离红衣服的女生最近", + "evidence": [ + { + "description": "I'm closest to the girl in red now", + "timestamps": { + "start": "00:03:51", + "end": "00:03:52" + } + } + ], + "question_timestamps": "00:04:59", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I'm closest to the girl in blue.", + "B": "I'm closest to the girl in white.", + "C": "I'm closest to the girl in pink right now, with the red-dressed girl a bit farther away.", + "D": "I'm closest to the girl in red now.", + "E": "I'm closest to the girl in green now, while the girl in red is standing farther from me." + }, + "answer_label": "D", + "answer_text": "I'm closest to the girl in red now.", + "q_id": "egolife_A1_D4_021_5", + "correct_answer": "I'm closest to the girl in red now.", + "original_options": { + "A": "I'm closest to the girl in blue.", + "B": "I'm closest to the girl in white.", + "C": "I'm closest to the girl in pink right now, with the red-dressed girl a bit farther away.", + "D": "I'm closest to the girl in red now.", + "E": "I'm closest to the girl in green now, while the girl in red is standing farther from me." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_021_1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I pet the dog with my hand?", + "answer": "I petted the dog by hand three times.", + "question_chinese": "我用手撸了几次狗？", + "answer_chinese": "我用手撸了3次狗", + "evidence": [ + { + "description": "I'm petting a dog", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + }, + { + "description": "I'm petting a dog", + "timestamps": { + "start": "00:00:34", + "end": "00:00:41" + } + }, + { + "description": "I'm petting a dog", + "timestamps": { + "start": "00:01:25", + "end": "00:01:28" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I petted the dog once.", + "B": "I petted the dog twice.", + "C": "I petted the dog by hand four times, with four distinct strokes during the interaction.", + "D": "I petted the dog by hand three times.", + "E": "I petted the dog by hand five times, finishing with a final pat before stopping." + }, + "answer_label": "D", + "answer_text": "I petted the dog by hand three times.", + "q_id": "egolife_A1_D4_021_1", + "correct_answer": "I petted the dog by hand three times.", + "original_options": { + "A": "I petted the dog once.", + "B": "I petted the dog twice.", + "C": "I petted the dog by hand four times, with four distinct strokes during the interaction.", + "D": "I petted the dog by hand three times.", + "E": "I petted the dog by hand five times, finishing with a final pat before stopping." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_021_2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times did I pet the dog with my hand?", + "answer": "I petted the dog by hand four times.", + "question_chinese": "我用手撸了几次狗？", + "answer_chinese": "我用手撸了4次狗", + "evidence": [ + { + "description": "I'm petting a dog", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + }, + { + "description": "I'm petting a dog", + "timestamps": { + "start": "00:00:34", + "end": "00:00:41" + } + }, + { + "description": "I'm petting a dog", + "timestamps": { + "start": "00:01:25", + "end": "00:01:28" + } + }, + { + "description": "I'm petting a dog", + "timestamps": { + "start": "00:01:48", + "end": "00:01:49" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I petted the dog once.", + "B": "I petted the dog by hand four times.", + "C": "I petted the dog twice.", + "D": "I petted the dog by hand three times, pausing briefly between each separate pet.", + "E": "I petted the dog by hand five times, giving it one extra pat at the end." + }, + "answer_label": "B", + "answer_text": "I petted the dog by hand four times.", + "q_id": "egolife_A1_D4_021_2", + "correct_answer": "I petted the dog by hand four times.", + "original_options": { + "A": "I petted the dog once.", + "B": "I petted the dog by hand four times.", + "C": "I petted the dog twice.", + "D": "I petted the dog by hand three times, pausing briefly between each separate pet.", + "E": "I petted the dog by hand five times, giving it one extra pat at the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D4_022-mcq.json b/QAs/egolife_A1_D4_022-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6e82f42b88a97ab5939a96729479f87bf2aa150f --- /dev/null +++ b/QAs/egolife_A1_D4_022-mcq.json @@ -0,0 +1,619 @@ +[ + { + "video_uid": "egolife_A1_D4_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_022_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I went from the courtyard through the corridor to the living room.", + "question_chinese": "我刚走了什么路线？", + "answer_chinese": "我从庭院通过走廊到客厅", + "evidence": [ + { + "description": "I went from the courtyard through the corridor to the living room", + "timestamps": { + "start": "00:07:27", + "end": "00:07:37" + } + } + ], + "question_timestamps": "00:07:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I went to the dining room.", + "B": "I went to the hallway.", + "C": "I went from the courtyard through the corridor to the living room.", + "D": "I went from the courtyard through the corridor to the kitchen, ending there instead of the living room.", + "E": "I went from the courtyard through the hallway to the living room, taking a different passage than the corridor." + }, + "answer_label": "C", + "answer_text": "I went from the courtyard through the corridor to the living room.", + "q_id": "egolife_A1_D4_022_7", + "correct_answer": "I went from the courtyard through the corridor to the living room.", + "original_options": { + "A": "I went to the dining room.", + "B": "I went to the hallway.", + "C": "I went from the courtyard through the corridor to the living room.", + "D": "I went from the courtyard through the corridor to the kitchen, ending there instead of the living room.", + "E": "I went from the courtyard through the hallway to the living room, taking a different passage than the corridor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_022_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I just walked from the corridor to the bedroom.", + "question_chinese": "我刚走了什么路线？", + "answer_chinese": "我刚从走廊到卧室", + "evidence": [ + { + "description": "I just walked from the corridor to the bedroom", + "timestamps": { + "start": "00:07:55", + "end": "00:08:07" + } + } + ], + "question_timestamps": "00:08:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I walked from the corridor to the hallway.", + "B": "I just walked from the corridor to the bedroom.", + "C": "I walked from the bedroom to the corridor.", + "D": "I just walked from the corridor to the dining room, not into the bedroom at the end.", + "E": "I just walked from the hallway to the bedroom, starting in a different place than the corridor." + }, + "answer_label": "B", + "answer_text": "I just walked from the corridor to the bedroom.", + "q_id": "egolife_A1_D4_022_8", + "correct_answer": "I just walked from the corridor to the bedroom.", + "original_options": { + "A": "I walked from the corridor to the hallway.", + "B": "I just walked from the corridor to the bedroom.", + "C": "I walked from the bedroom to the corridor.", + "D": "I just walked from the corridor to the dining room, not into the bedroom at the end.", + "E": "I just walked from the hallway to the bedroom, starting in a different place than the corridor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_022_6", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I access the sink now?", + "answer": "I can reach the sink now.", + "question_chinese": "我现在能碰到洗碗池吗？", + "answer_chinese": "我现在能碰到洗碗池", + "evidence": [ + { + "description": "The sink is in the kitchen", + "timestamps": { + "start": "00:05:58", + "end": "00:06:02" + } + }, + { + "description": "I'm in the kitchen now, too", + "timestamps": { + "start": "00:07:37", + "end": "00:07:39" + } + } + ], + "question_timestamps": "00:07:44", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I can't reach the sink right now.", + "B": "I can reach the sink now." + }, + "answer_label": "B", + "answer_text": "I can reach the sink now.", + "q_id": "egolife_A1_D4_022_6", + "correct_answer": "I can reach the sink now.", + "original_options": { + "A": "I can't reach the sink right now.", + "B": "I can reach the sink now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D4_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_022_5", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I access the sink now?", + "answer": "I can't reach the sink right now.", + "question_chinese": "我现在能碰到洗碗池吗？", + "answer_chinese": "我现在不能碰到洗碗池", + "evidence": [ + { + "description": "The sink is in the kitchen", + "timestamps": { + "start": "00:05:58", + "end": "00:06:02" + } + }, + { + "description": "I'm in the yard.", + "timestamps": { + "start": "00:08:45", + "end": "00:08:50" + } + } + ], + "question_timestamps": "00:09:43", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I can't reach the sink right now.", + "B": "I can reach the sink now." + }, + "answer_label": "A", + "answer_text": "I can't reach the sink right now.", + "q_id": "egolife_A1_D4_022_5", + "correct_answer": "I can't reach the sink right now.", + "original_options": { + "A": "I can't reach the sink right now.", + "B": "I can reach the sink now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D4_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_022_9", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Am I facing the bedroom right now?", + "answer": "What I'm facing now is not a bedroom.", + "question_chinese": "我现在面对的是不是卧室？", + "answer_chinese": "我现在面对的不是卧室", + "evidence": [ + { + "description": "I turned and walked towards the courtyard", + "timestamps": { + "start": "00:07:50", + "end": "00:07:53" + } + } + ], + "question_timestamps": "00:07:59", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Now I'm facing the bedroom.", + "B": "What I'm facing now is not a bedroom." + }, + "answer_label": "B", + "answer_text": "What I'm facing now is not a bedroom.", + "q_id": "egolife_A1_D4_022_9", + "correct_answer": "What I'm facing now is not a bedroom.", + "original_options": { + "A": "Now I'm facing the bedroom.", + "B": "What I'm facing now is not a bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D4_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_022_10", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Am I facing the bedroom right now?", + "answer": "Now I'm facing the bedroom.", + "question_chinese": "我现在面对的是不是卧室？", + "answer_chinese": "我现在面对的是卧室", + "evidence": [ + { + "description": "Now I'm facing the bedroom", + "timestamps": { + "start": "00:08:03", + "end": "00:08:05" + } + } + ], + "question_timestamps": "00:08:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "What I'm facing now is not a bedroom.", + "B": "Now I'm facing the bedroom." + }, + "answer_label": "B", + "answer_text": "Now I'm facing the bedroom.", + "q_id": "egolife_A1_D4_022_10", + "correct_answer": "Now I'm facing the bedroom.", + "original_options": { + "A": "What I'm facing now is not a bedroom.", + "B": "Now I'm facing the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D4_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_022_1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many underground drainage pipes have I passed through?", + "answer": "I walked through three underground drainage pipes.", + "question_chinese": "我走过了几个地下排水管道？", + "answer_chinese": "我走过了3个地下排水管道", + "evidence": [ + { + "description": "I walked through the first underground drainage pipe", + "timestamps": { + "start": "00:00:42", + "end": "00:00:45" + } + }, + { + "description": "I walked through the second underground drainage pipe", + "timestamps": { + "start": "00:01:09", + "end": "00:01:11" + } + }, + { + "description": "I walked through the third underground drainage pipe", + "timestamps": { + "start": "00:01:33", + "end": "00:01:39" + } + } + ], + "question_timestamps": "00:02:53", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I walked through two underground drainage pipes.", + "B": "I walked through four underground drainage pipes.", + "C": "I walked through three underground drainage pipes.", + "D": "I walked through five underground drainage pipes in total as I moved along the underground passage.", + "E": "I walked through six underground drainage pipes overall before reaching the next open area." + }, + "answer_label": "C", + "answer_text": "I walked through three underground drainage pipes.", + "q_id": "egolife_A1_D4_022_1", + "correct_answer": "I walked through three underground drainage pipes.", + "original_options": { + "A": "I walked through two underground drainage pipes.", + "B": "I walked through four underground drainage pipes.", + "C": "I walked through three underground drainage pipes.", + "D": "I walked through five underground drainage pipes in total as I moved along the underground passage.", + "E": "I walked through six underground drainage pipes overall before reaching the next open area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_022_2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How many underground drainage pipes have I passed through?", + "answer": "I walked through four underground drainage pipes.", + "question_chinese": "我走过了几个地下排水管道？", + "answer_chinese": "我走过了4个地下排水管道", + "evidence": [ + { + "description": "I walked through the first underground drainage pipe", + "timestamps": { + "start": "00:00:42", + "end": "00:00:45" + } + }, + { + "description": "I walked through the second underground drainage pipe", + "timestamps": { + "start": "00:01:09", + "end": "00:01:11" + } + }, + { + "description": "I walked through the third underground drainage pipe", + "timestamps": { + "start": "00:01:33", + "end": "00:01:39" + } + }, + { + "description": "I walked through the fourth underground drainage pipe", + "timestamps": { + "start": "00:03:34", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I walked through three underground drainage pipes.", + "B": "I walked through five underground drainage pipes.", + "C": "I walked through two underground drainage pipes before coming out to the next section of the route.", + "D": "I walked through four underground drainage pipes.", + "E": "I walked through six underground drainage pipes altogether during that part of the walk." + }, + "answer_label": "D", + "answer_text": "I walked through four underground drainage pipes.", + "q_id": "egolife_A1_D4_022_2", + "correct_answer": "I walked through four underground drainage pipes.", + "original_options": { + "A": "I walked through three underground drainage pipes.", + "B": "I walked through five underground drainage pipes.", + "C": "I walked through two underground drainage pipes before coming out to the next section of the route.", + "D": "I walked through four underground drainage pipes.", + "E": "I walked through six underground drainage pipes altogether during that part of the walk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_022_3", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are around me?", + "answer": "There are five people around me.", + "question_chinese": "我身边有几个人？", + "answer_chinese": "我身边有5个人", + "evidence": [ + { + "description": "There are four people in my field of vision", + "timestamps": { + "start": "00:02:35", + "end": "00:02:37" + } + }, + { + "description": "I'm with them", + "timestamps": { + "start": "00:00:00", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are four people around me.", + "B": "There are six people around me.", + "C": "There are three people around me, standing close enough to be considered nearby.", + "D": "There are five people around me.", + "E": "There are seven people around me, gathered around in the immediate area." + }, + "answer_label": "D", + "answer_text": "There are five people around me.", + "q_id": "egolife_A1_D4_022_3", + "correct_answer": "There are five people around me.", + "original_options": { + "A": "There are four people around me.", + "B": "There are six people around me.", + "C": "There are three people around me, standing close enough to be considered nearby.", + "D": "There are five people around me.", + "E": "There are seven people around me, gathered around in the immediate area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_022_4", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are around me?", + "answer": "There are three people around me.", + "question_chinese": "我身边有几个人？", + "answer_chinese": "我身边有3个人", + "evidence": [ + { + "description": "There are two people in my field of vision", + "timestamps": { + "start": "00:06:05", + "end": "00:06:12" + } + }, + { + "description": "There are two people in my field of vision", + "timestamps": { + "start": "00:06:05", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:06:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are two people around me.", + "B": "There are four people around me.", + "C": "There are five people around me, positioned near you in the same space.", + "D": "There are six people around me, close enough to count as around you.", + "E": "There are three people around me." + }, + "answer_label": "E", + "answer_text": "There are three people around me.", + "q_id": "egolife_A1_D4_022_4", + "correct_answer": "There are three people around me.", + "original_options": { + "A": "There are two people around me.", + "B": "There are four people around me.", + "C": "There are five people around me, positioned near you in the same space.", + "D": "There are six people around me, close enough to count as around you.", + "E": "There are three people around me." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D4_023-mcq.json b/QAs/egolife_A1_D4_023-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ea28031ea72d1738658a31e7ecc8f610cbc8e787 --- /dev/null +++ b/QAs/egolife_A1_D4_023-mcq.json @@ -0,0 +1,639 @@ +[ + { + "video_uid": "egolife_A1_D4_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_023-mcq_fixed_9", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand the last time I passed by the living room door?", + "answer": "The last time YOU passed by the living room door, You were holding a black data cable in your hand.", + "question_chinese": "我上一次经过客厅的门的时候手里拿着什么？", + "answer_chinese": "我上一次经过客厅的门的时候手里拿着黑色数据线", + "evidence": [ + { + "description": "I'm picking up the black data cable", + "timestamps": { + "start": "00:00:48", + "end": "00:00:52" + } + }, + { + "description": "I'm passing by the living room door", + "timestamps": { + "start": "00:00:56", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:01:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You were holding a red apple.", + "B": "The last time YOU passed by the living room door, You were holding a black data cable in your hand.", + "C": "You were holding a green water bottle.", + "D": "The last time YOU passed by the living room door, you were holding a white charging cable in your hand.", + "E": "The last time YOU passed by the living room door, you were holding a grey HDMI cord in your hand." + }, + "answer_label": "B", + "answer_text": "The last time YOU passed by the living room door, You were holding a black data cable in your hand.", + "q_id": "egolife_A1_D4_023-mcq_fixed_9", + "correct_answer": "The last time YOU passed by the living room door, You were holding a black data cable in your hand.", + "original_options": { + "A": "You were holding a red apple.", + "B": "The last time YOU passed by the living room door, You were holding a black data cable in your hand.", + "C": "You were holding a green water bottle.", + "D": "The last time YOU passed by the living room door, you were holding a white charging cable in your hand.", + "E": "The last time YOU passed by the living room door, you were holding a grey HDMI cord in your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_023-mcq_fixed_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand the last time I passed by the living room door?", + "answer": "The last time YOU passed by the living room door, your hands were empty.", + "question_chinese": "我上一次经过客厅的门的时候手里拿着什么？", + "answer_chinese": "我上一次经过客厅的门的时候手里是空的", + "evidence": [ + { + "description": "I put down the black data cable", + "timestamps": { + "start": "00:01:08", + "end": "00:01:09" + } + }, + { + "description": "I'm passing by the living room door", + "timestamps": { + "start": "00:01:18", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:02:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You were holding keys.", + "B": "You were holding some paper.", + "C": "The last time YOU passed by the living room door, you were holding a small paperback book in your hand.", + "D": "The last time YOU passed by the living room door, you were holding a notepad with notes in your hand.", + "E": "The last time YOU passed by the living room door, your hands were empty." + }, + "answer_label": "E", + "answer_text": "The last time YOU passed by the living room door, your hands were empty.", + "q_id": "egolife_A1_D4_023-mcq_fixed_8", + "correct_answer": "The last time YOU passed by the living room door, your hands were empty.", + "original_options": { + "A": "You were holding keys.", + "B": "You were holding some paper.", + "C": "The last time YOU passed by the living room door, you were holding a small paperback book in your hand.", + "D": "The last time YOU passed by the living room door, you were holding a notepad with notes in your hand.", + "E": "The last time YOU passed by the living room door, your hands were empty." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_023_9", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many doors did I just pass by?", + "answer": "I just passed through three doors.", + "question_chinese": "我刚经过了几个门？", + "answer_chinese": "我刚经过了3个门", + "evidence": [ + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:00:38", + "end": "00:00:41" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:00:56", + "end": "00:00:58" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:00:59", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "I just passed through two doors.", + "B": "I just passed through four doors.", + "C": "I just passed through one door before stopping near the next doorway along the way.", + "D": "I just passed through five doors in a row, moving from one room into the next.", + "E": "I just passed through three doors." + }, + "answer_label": "E", + "answer_text": "I just passed through three doors.", + "q_id": "egolife_A1_D4_023_9", + "correct_answer": "I just passed through three doors.", + "original_options": { + "A": "I just passed through two doors.", + "B": "I just passed through four doors.", + "C": "I just passed through one door before stopping near the next doorway along the way.", + "D": "I just passed through five doors in a row, moving from one room into the next.", + "E": "I just passed through three doors." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_023_10", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many doors did I just pass by?", + "answer": "I just passed through six doors.", + "question_chinese": "我刚经过了几个门？", + "answer_chinese": "我刚经过了6个门", + "evidence": [ + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:00:38", + "end": "00:00:41" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:00:56", + "end": "00:00:58" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:00:59", + "end": "00:01:02" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:01:14", + "end": "00:01:15" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:01:18", + "end": "00:01:19" + } + }, + { + "description": "I'm passing by the door", + "timestamps": { + "start": "00:01:24", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "I just passed through six doors.", + "B": "I just passed through five doors.", + "C": "I just passed through seven doors.", + "D": "I just passed through four doors before reaching the area I entered last.", + "E": "I just passed through eight doors, continuing through multiple doorways without turning back." + }, + "answer_label": "A", + "answer_text": "I just passed through six doors.", + "q_id": "egolife_A1_D4_023_10", + "correct_answer": "I just passed through six doors.", + "original_options": { + "A": "I just passed through six doors.", + "B": "I just passed through five doors.", + "C": "I just passed through seven doors.", + "D": "I just passed through four doors before reaching the area I entered last.", + "E": "I just passed through eight doors, continuing through multiple doorways without turning back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_023_7", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I just walked from the kitchen to the yard.", + "question_chinese": "我刚走了什么路线？", + "answer_chinese": "我刚从厨房走到院子", + "evidence": [ + { + "description": "I just walked from the kitchen to the yard", + "timestamps": { + "start": "00:02:01", + "end": "00:02:13" + } + } + ], + "question_timestamps": "00:02:28", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I walked from the yard to the kitchen.", + "B": "I just walked from the kitchen to the yard.", + "C": "I walked from the kitchen to the porch.", + "D": "I just walked from the kitchen to the garage, leaving the cooking area behind you.", + "E": "I just walked from the yard to the hallway, moving indoors rather than heading outside." + }, + "answer_label": "B", + "answer_text": "I just walked from the kitchen to the yard.", + "q_id": "egolife_A1_D4_023_7", + "correct_answer": "I just walked from the kitchen to the yard.", + "original_options": { + "A": "I walked from the yard to the kitchen.", + "B": "I just walked from the kitchen to the yard.", + "C": "I walked from the kitchen to the porch.", + "D": "I just walked from the kitchen to the garage, leaving the cooking area behind you.", + "E": "I just walked from the yard to the hallway, moving indoors rather than heading outside." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_023_8", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I just take?", + "answer": "I just walked from the bedroom to the living room and then to the yard.", + "question_chinese": "我刚走了什么路线？", + "answer_chinese": "我刚从卧室走到客厅再到院子", + "evidence": [ + { + "description": "I just walked from the bedroom to the living room and then to the yard", + "timestamps": { + "start": "00:03:00", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:05:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Bedroom to yard, then living room.", + "B": "Living room to bedroom, then yard.", + "C": "I just walked from the bedroom to the yard first, then went into the living room afterward.", + "D": "I just walked from the bedroom to the living room and then to the yard.", + "E": "I just walked from the yard to the living room and then continued on to the bedroom." + }, + "answer_label": "D", + "answer_text": "I just walked from the bedroom to the living room and then to the yard.", + "q_id": "egolife_A1_D4_023_8", + "correct_answer": "I just walked from the bedroom to the living room and then to the yard.", + "original_options": { + "A": "Bedroom to yard, then living room.", + "B": "Living room to bedroom, then yard.", + "C": "I just walked from the bedroom to the yard first, then went into the living room afterward.", + "D": "I just walked from the bedroom to the living room and then to the yard.", + "E": "I just walked from the yard to the living room and then continued on to the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_023_4", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "\"Who am I closest to?\"", + "answer": "I'm closest to the girl in red.", + "question_chinese": "我离谁最近？", + "answer_chinese": "我离红衣服的女生最近", + "evidence": [ + { + "description": "I'm closest to the girl in red", + "timestamps": { + "start": "00:02:44", + "end": "00:02:46" + } + } + ], + "question_timestamps": "00:03:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I'm closest to the girl to my left.", + "B": "I'm closest to the girl in red.", + "C": "I'm closest to the girl behind me.", + "D": "I'm closest to the girl in front of you, not the one wearing red nearby.", + "E": "I'm closest to the girl to your right, with the girl in red standing farther away." + }, + "answer_label": "B", + "answer_text": "I'm closest to the girl in red.", + "q_id": "egolife_A1_D4_023_4", + "correct_answer": "I'm closest to the girl in red.", + "original_options": { + "A": "I'm closest to the girl to my left.", + "B": "I'm closest to the girl in red.", + "C": "I'm closest to the girl behind me.", + "D": "I'm closest to the girl in front of you, not the one wearing red nearby.", + "E": "I'm closest to the girl to your right, with the girl in red standing farther away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_023_3", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "\"Who am I closest to?\"", + "answer": "I'm closest to boys in black clothes and with grey hair.", + "question_chinese": "我离谁最近？", + "answer_chinese": "我离黑衣服灰头发的男生最近", + "evidence": [ + { + "description": "I'm closest to boys in black clothes and with grey hair", + "timestamps": { + "start": "00:04:43", + "end": "00:04:45" + } + } + ], + "question_timestamps": "00:06:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I'm closest to boys in blue outfits.", + "B": "I'm closest to girls in white dresses.", + "C": "I'm closest to boys in black clothes and with grey hair.", + "D": "I'm closest to boys in grey jackets with red hair, not the boys dressed in black.", + "E": "I'm closest to girls in striped shirts with brown hair, standing nearer than any of the boys." + }, + "answer_label": "C", + "answer_text": "I'm closest to boys in black clothes and with grey hair.", + "q_id": "egolife_A1_D4_023_3", + "correct_answer": "I'm closest to boys in black clothes and with grey hair.", + "original_options": { + "A": "I'm closest to boys in blue outfits.", + "B": "I'm closest to girls in white dresses.", + "C": "I'm closest to boys in black clothes and with grey hair.", + "D": "I'm closest to boys in grey jackets with red hair, not the boys dressed in black.", + "E": "I'm closest to girls in striped shirts with brown hair, standing nearer than any of the boys." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_023_5", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I reach the clothes drying rack without going through the door?", + "answer": "I can touch the clothes drying rack without going through the door.", + "question_chinese": "我能不经过门就碰到晾衣架吗？", + "answer_chinese": "我能不经过门就碰到晾衣架", + "evidence": [ + { + "description": "The clothes drying rack is in the yard", + "timestamps": { + "start": "00:04:10", + "end": "00:04:13" + } + }, + { + "description": "I'm also in the yard", + "timestamps": { + "start": "00:04:18", + "end": "00:04:19" + } + } + ], + "question_timestamps": "00:07:07", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I can touch the clothes drying rack without going through the door.", + "B": "I can't touch the clothes drying rack without going through the door." + }, + "answer_label": "A", + "answer_text": "I can touch the clothes drying rack without going through the door.", + "q_id": "egolife_A1_D4_023_5", + "correct_answer": "I can touch the clothes drying rack without going through the door.", + "original_options": { + "A": "I can touch the clothes drying rack without going through the door.", + "B": "I can't touch the clothes drying rack without going through the door." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_023_6", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I reach the clothes drying rack without going through the door?", + "answer": "I couldn't avoid bumping into the clothes drying rack just by passing through the door.", + "question_chinese": "我能不经过门就碰到晾衣架吗？", + "answer_chinese": "我不能不经过门就碰到晾衣架", + "evidence": [ + { + "description": "The clothes drying rack is in the yard", + "timestamps": { + "start": "00:04:10", + "end": "00:04:13" + } + }, + { + "description": "I'm indoors.", + "timestamps": { + "start": "00:10:31", + "end": "00:10:32" + } + } + ], + "question_timestamps": "00:10:48", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I couldn't avoid bumping into the clothes drying rack just by passing through the door.", + "B": "I could avoid bumping into the clothes drying rack just by passing through the door." + }, + "answer_label": "A", + "answer_text": "I couldn't avoid bumping into the clothes drying rack just by passing through the door.", + "q_id": "egolife_A1_D4_023_6", + "correct_answer": "I couldn't avoid bumping into the clothes drying rack just by passing through the door.", + "original_options": { + "A": "I couldn't avoid bumping into the clothes drying rack just by passing through the door.", + "B": "I could avoid bumping into the clothes drying rack just by passing through the door." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + } +] diff --git a/QAs/egolife_A1_D4_025-mcq.json b/QAs/egolife_A1_D4_025-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..0c8176bb80c8987f4ead4ecb8c666c070ca09f7a --- /dev/null +++ b/QAs/egolife_A1_D4_025-mcq.json @@ -0,0 +1,872 @@ +[ + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025_3", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I just arrive here?", + "answer": "I went from the courtyard through the corridor to the living room.", + "question_chinese": "我刚怎么到达这里的？", + "answer_chinese": "我从庭院经过走廊到达客厅", + "evidence": [ + { + "description": "I go from the courtyard through the corridor to the living room", + "timestamps": { + "start": "00:00:31", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I went to the dining room.", + "B": "I came from the attic.", + "C": "I went from the courtyard through the corridor to the dining room, not the living room.", + "D": "I came from the attic into the courtyard, then took the corridor before entering the dining room.", + "E": "I went from the courtyard through the corridor to the living room." + }, + "answer_label": "E", + "answer_text": "I went from the courtyard through the corridor to the living room.", + "q_id": "egolife_A1_D4_025_3", + "correct_answer": "I went from the courtyard through the corridor to the living room.", + "original_options": { + "A": "I went to the dining room.", + "B": "I came from the attic.", + "C": "I went from the courtyard through the corridor to the dining room, not the living room.", + "D": "I came from the attic into the courtyard, then took the corridor before entering the dining room.", + "E": "I went from the courtyard through the corridor to the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025_4", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I just arrive here?", + "answer": "I reached the courtyard from the bedroom through the living room and the corridor.", + "question_chinese": "我刚怎么到达这里的？", + "answer_chinese": "我从卧室经过客厅和走廊到达庭院", + "evidence": [ + { + "description": "I reach the courtyard from the bedroom through the living room and the corridor", + "timestamps": { + "start": "00:01:28", + "end": "00:01:50" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I came from the bathroom.", + "B": "I got here from the garage.", + "C": "I reached the courtyard from the bathroom through the living room and the corridor, not from the bedroom.", + "D": "I reached the courtyard from the bedroom through the living room and the corridor.", + "E": "I arrived at the courtyard from the bedroom via the living room and the front porch, skipping the corridor." + }, + "answer_label": "D", + "answer_text": "I reached the courtyard from the bedroom through the living room and the corridor.", + "q_id": "egolife_A1_D4_025_4", + "correct_answer": "I reached the courtyard from the bedroom through the living room and the corridor.", + "original_options": { + "A": "I came from the bathroom.", + "B": "I got here from the garage.", + "C": "I reached the courtyard from the bathroom through the living room and the corridor, not from the bedroom.", + "D": "I reached the courtyard from the bedroom through the living room and the corridor.", + "E": "I arrived at the courtyard from the bedroom via the living room and the front porch, skipping the corridor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025_5", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get here just now?", + "answer": "I went from the courtyard through the corridor to the living room.", + "question_chinese": "我刚才怎么到达这里的？", + "answer_chinese": "我从庭院经过走廊到客厅", + "evidence": [ + { + "description": "I went from the courtyard through the corridor to the living room", + "timestamps": { + "start": "00:00:31", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I went from the courtyard through the corridor to the living room.", + "B": "I came from the garden.", + "C": "I walked in from the closet.", + "D": "I got here from the basement by going up the stairs and then through the hallway.", + "E": "I arrived from the second floor by coming down the stairs and entering the kitchen first." + }, + "answer_label": "A", + "answer_text": "I went from the courtyard through the corridor to the living room.", + "q_id": "egolife_A1_D4_025_5", + "correct_answer": "I went from the courtyard through the corridor to the living room.", + "original_options": { + "A": "I went from the courtyard through the corridor to the living room.", + "B": "I came from the garden.", + "C": "I walked in from the closet.", + "D": "I got here from the basement by going up the stairs and then through the hallway.", + "E": "I arrived from the second floor by coming down the stairs and entering the kitchen first." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025_6", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I get here just now?", + "answer": "I reached the courtyard from the bedroom through the living room and the corridor.", + "question_chinese": "我刚才怎么到达这里的？", + "answer_chinese": "我从卧室经过客厅和走廊到达庭院", + "evidence": [ + { + "description": "I reached the courtyard from the bedroom through the living room and the corridor", + "timestamps": { + "start": "00:01:28", + "end": "00:01:50" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I came from the bathroom.", + "B": "I reached the courtyard from the bedroom through the living room and the corridor.", + "C": "I arrived from the kitchen.", + "D": "I reached the courtyard from the bathroom through the hallway and then the living room, not from the bedroom.", + "E": "I arrived in the courtyard from the garage by passing through the living room and the corridor." + }, + "answer_label": "B", + "answer_text": "I reached the courtyard from the bedroom through the living room and the corridor.", + "q_id": "egolife_A1_D4_025_6", + "correct_answer": "I reached the courtyard from the bedroom through the living room and the corridor.", + "original_options": { + "A": "I came from the bathroom.", + "B": "I reached the courtyard from the bedroom through the living room and the corridor.", + "C": "I arrived from the kitchen.", + "D": "I reached the courtyard from the bathroom through the hallway and then the living room, not from the bedroom.", + "E": "I arrived in the courtyard from the garage by passing through the living room and the corridor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025_4_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors have I passed through so far?", + "answer": "You've passed through four doors now.", + "question_chinese": "我现在经过了几扇门？", + "answer_chinese": "我现在经过了4扇门", + "evidence": [ + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:30", + "end": "00:00:33" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:37", + "end": "00:00:40" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:56", + "end": "00:00:58" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:11", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:01:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You've passed through one door.", + "B": "You've passed through two doors.", + "C": "You've passed through three doors now, so you only went through three separate doorways so far.", + "D": "You've passed through four doors now.", + "E": "You've passed through five doors now, meaning you crossed five different door thresholds up to this point." + }, + "answer_label": "D", + "answer_text": "You've passed through four doors now.", + "q_id": "egolife_A1_D4_025_4_T1", + "correct_answer": "You've passed through four doors now.", + "original_options": { + "A": "You've passed through one door.", + "B": "You've passed through two doors.", + "C": "You've passed through three doors now, so you only went through three separate doorways so far.", + "D": "You've passed through four doors now.", + "E": "You've passed through five doors now, meaning you crossed five different door thresholds up to this point." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025_4_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors have I passed through now?", + "answer": "You've passed through four doors now.", + "question_chinese": "我现在经过了几扇门？", + "answer_chinese": "我现在经过了4扇门", + "evidence": [ + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:30", + "end": "00:00:33" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:37", + "end": "00:00:40" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:56", + "end": "00:00:58" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:11", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You've passed through one door.", + "B": "You've passed through three doors.", + "C": "You've passed through two doors now, so you have only crossed two doorways up to this moment.", + "D": "You've passed through four doors now.", + "E": "You've passed through five doors now, meaning you entered through five distinct doors along your route." + }, + "answer_label": "D", + "answer_text": "You've passed through four doors now.", + "q_id": "egolife_A1_D4_025_4_T2", + "correct_answer": "You've passed through four doors now.", + "original_options": { + "A": "You've passed through one door.", + "B": "You've passed through three doors.", + "C": "You've passed through two doors now, so you have only crossed two doorways up to this moment.", + "D": "You've passed through four doors now.", + "E": "You've passed through five doors now, meaning you entered through five distinct doors along your route." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025_5_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors have I passed through so far?", + "answer": "You've passed through eight doors now.", + "question_chinese": "我现在经过了几扇门？", + "answer_chinese": "我现在经过了8扇门", + "evidence": [ + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:30", + "end": "00:00:33" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:37", + "end": "00:00:40" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:56", + "end": "00:00:58" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:11", + "end": "00:01:14" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:25", + "end": "00:01:27" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:32", + "end": "00:01:34" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:39", + "end": "00:01:41" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:43", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:02:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You've passed through eight doors now.", + "B": "You've passed through six doors.", + "C": "You've passed through seven doors.", + "D": "You've passed through nine doors now, so you have crossed nine separate doorways so far.", + "E": "You've passed through ten doors now, meaning you went through ten distinct door thresholds up to now." + }, + "answer_label": "A", + "answer_text": "You've passed through eight doors now.", + "q_id": "egolife_A1_D4_025_5_T1", + "correct_answer": "You've passed through eight doors now.", + "original_options": { + "A": "You've passed through eight doors now.", + "B": "You've passed through six doors.", + "C": "You've passed through seven doors.", + "D": "You've passed through nine doors now, so you have crossed nine separate doorways so far.", + "E": "You've passed through ten doors now, meaning you went through ten distinct door thresholds up to now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025_5_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors have I passed through?", + "answer": "You've passed through eight doors now.", + "question_chinese": "我现在经过了几扇门？", + "answer_chinese": "我现在经过了8扇门", + "evidence": [ + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:30", + "end": "00:00:33" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:37", + "end": "00:00:40" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:00:56", + "end": "00:00:58" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:11", + "end": "00:01:14" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:25", + "end": "00:01:27" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:32", + "end": "00:01:34" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:39", + "end": "00:01:41" + } + }, + { + "description": "I passed through a door", + "timestamps": { + "start": "00:01:43", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:02:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You've passed through eight doors now.", + "B": "You've passed through six doors.", + "C": "You've passed through nine doors.", + "D": "You've passed through seven doors now, so you only crossed seven different doorways along the way.", + "E": "You've passed through ten doors now, meaning you have gone through ten distinct door thresholds so far." + }, + "answer_label": "A", + "answer_text": "You've passed through eight doors now.", + "q_id": "egolife_A1_D4_025_5_T2", + "correct_answer": "You've passed through eight doors now.", + "original_options": { + "A": "You've passed through eight doors now.", + "B": "You've passed through six doors.", + "C": "You've passed through nine doors.", + "D": "You've passed through seven doors now, so you only crossed seven different doorways along the way.", + "E": "You've passed through ten doors now, meaning you have gone through ten distinct door thresholds so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025-mcq_fixed_7", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I get the guitar without going through the door now?", + "answer": "Yes, YOU can.", + "question_chinese": "我现在能不经过门就拿到吉他吗？", + "answer_chinese": "是的，我可以", + "evidence": [ + { + "description": "The guitar is in the yard.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "I'm also in the yard", + "timestamps": { + "start": "00:02:30", + "end": "00:02:31" + } + } + ], + "question_timestamps": "00:02:44", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, YOU can.", + "B": "No, YOU can't." + }, + "answer_label": "A", + "answer_text": "Yes, YOU can.", + "q_id": "egolife_A1_D4_025-mcq_fixed_7", + "correct_answer": "Yes, YOU can.", + "original_options": { + "A": "Yes, YOU can.", + "B": "No, YOU can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025-mcq_fixed_6", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I get the guitar without going through the door now?", + "answer": "No, YOU can't.", + "question_chinese": "我现在能不经过门就拿到吉他吗？", + "answer_chinese": "不，我不行", + "evidence": [ + { + "description": "The guitar is in the yard.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "I'm indoors.", + "timestamps": { + "start": "00:04:45", + "end": "00:04:50" + } + } + ], + "question_timestamps": "00:07:04", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, YOU can.", + "B": "No, YOU can't." + }, + "answer_label": "B", + "answer_text": "No, YOU can't.", + "q_id": "egolife_A1_D4_025-mcq_fixed_6", + "correct_answer": "No, YOU can't.", + "original_options": { + "A": "Yes, YOU can.", + "B": "No, YOU can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025_7_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are here?", + "answer": "Five.", + "question_chinese": "这里有几个人？", + "answer_chinese": "这里有5个人", + "evidence": [ + { + "description": "I'm here", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "There are four people in my field of vision", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "There are three people standing here together right now in this area.", + "E": "There are seven people here in total, gathered around the scene at the moment." + }, + "answer_label": "C", + "answer_text": "Five.", + "q_id": "egolife_A1_D4_025_7_T1", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "There are three people standing here together right now in this area.", + "E": "There are seven people here in total, gathered around the scene at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025_1_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are here?", + "answer": "There are two people here.", + "question_chinese": "这里有几个人？", + "answer_chinese": "这里有2个人", + "evidence": [ + { + "description": "There is one person in my field of vision", + "timestamps": { + "start": "00:00:38", + "end": "00:00:39" + } + }, + { + "description": "I'm here too", + "timestamps": { + "start": "00:00:40", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:01:34", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One person.", + "B": "There are two people here.", + "C": "Three people.", + "D": "There are four people here in total, standing together in this area right now.", + "E": "There are five people here altogether, gathered in the scene at this moment." + }, + "answer_label": "B", + "answer_text": "There are two people here.", + "q_id": "egolife_A1_D4_025_1_T1", + "correct_answer": "There are two people here.", + "original_options": { + "A": "One person.", + "B": "There are two people here.", + "C": "Three people.", + "D": "There are four people here in total, standing together in this area right now.", + "E": "There are five people here altogether, gathered in the scene at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_025_1_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are here?", + "answer": "There are two people here.", + "question_chinese": "这里有几个人？", + "answer_chinese": "这里有2个人", + "evidence": [ + { + "description": "There is one person in my field of vision", + "timestamps": { + "start": "00:00:38", + "end": "00:00:39" + } + }, + { + "description": "I'm here too", + "timestamps": { + "start": "00:00:40", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One person.", + "B": "There are two people here.", + "C": "Three people.", + "D": "There are four people here in total, standing together in this area right now.", + "E": "There are five people here altogether, gathered in the scene at this moment." + }, + "answer_label": "B", + "answer_text": "There are two people here.", + "q_id": "egolife_A1_D4_025_1_T2", + "correct_answer": "There are two people here.", + "original_options": { + "A": "One person.", + "B": "There are two people here.", + "C": "Three people.", + "D": "There are four people here in total, standing together in this area right now.", + "E": "There are five people here altogether, gathered in the scene at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D4_027-mcq.json b/QAs/egolife_A1_D4_027-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9bc15c9630eee8cee9ae4e1d1f5e0164de87ceeb --- /dev/null +++ b/QAs/egolife_A1_D4_027-mcq.json @@ -0,0 +1,952 @@ +[ + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_4_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I go to the yard right now, how should I walk there?", + "answer": "You should turn right, go out through the door by your right hand, then turn left and go straight.", + "question_chinese": "如果我现在去院子应该怎么走？", + "answer_chinese": "你应该想右转从右手的门出去左转直走。", + "evidence": [ + { + "description": "Go out through this door", + "timestamps": { + "start": "00:02:22", + "end": "00:02:26" + } + }, + { + "description": "Turn left and go straight into the yard", + "timestamps": { + "start": "00:02:27", + "end": "00:02:30" + } + }, + { + "description": "Right beside the table, I see a door on the left front", + "timestamps": { + "start": "00:07:52", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:08:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go straight.", + "B": "You should turn right, go out through the door by your right hand, then turn left and go straight.", + "C": "Turn right, then turn right.", + "D": "Turn right, exit through the door by your right hand, then turn right and continue straight.", + "E": "Turn left first, go out through the door by your right hand, then turn left and walk straight." + }, + "answer_label": "B", + "answer_text": "You should turn right, go out through the door by your right hand, then turn left and go straight.", + "q_id": "egolife_A1_D4_027_4_T2", + "correct_answer": "You should turn right, go out through the door by your right hand, then turn left and go straight.", + "original_options": { + "A": "Turn left, then go straight.", + "B": "You should turn right, go out through the door by your right hand, then turn left and go straight.", + "C": "Turn right, then turn right.", + "D": "Turn right, exit through the door by your right hand, then turn right and continue straight.", + "E": "Turn left first, go out through the door by your right hand, then turn left and walk straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_3_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I go to the yard right now, how should I walk there?", + "answer": "If you go to the yard now, you should turn around, go out through the door to your left, turn left, and then go straight.", + "question_chinese": "如果我现在去院子应该怎么走？", + "answer_chinese": "如果我现在去院子应该向后转从左边的门出去左转再直走", + "evidence": [ + { + "description": "Turn backward at the same position. There is a door on the left", + "timestamps": { + "start": "00:07:16", + "end": "00:07:18" + } + }, + { + "description": "Go out through this door", + "timestamps": { + "start": "00:02:22", + "end": "00:02:26" + } + }, + { + "description": "Turn left and go straight into the yard", + "timestamps": { + "start": "00:02:27", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:08:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use the door to your right.", + "B": "Go out the door in front.", + "C": "Turn around, go out through the door to your right, then turn left and go straight.", + "D": "If you go to the yard now, you should turn around, go out through the door to your left, turn left, and then go straight.", + "E": "Turn around, exit through the door behind you, then turn left and continue straight ahead." + }, + "answer_label": "D", + "answer_text": "If you go to the yard now, you should turn around, go out through the door to your left, turn left, and then go straight.", + "q_id": "egolife_A1_D4_027_3_T1", + "correct_answer": "If you go to the yard now, you should turn around, go out through the door to your left, turn left, and then go straight.", + "original_options": { + "A": "Use the door to your right.", + "B": "Go out the door in front.", + "C": "Turn around, go out through the door to your right, then turn left and go straight.", + "D": "If you go to the yard now, you should turn around, go out through the door to your left, turn left, and then go straight.", + "E": "Turn around, exit through the door behind you, then turn left and continue straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_3_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I go to the yard right now, how should I walk there?", + "answer": "If you go to the yard now, you should turn left, go out through the door to your left, turn left, and then go straight.", + "question_chinese": "如果我现在去院子应该怎么走？", + "answer_chinese": "如果我现在去院子应该向左转从左边的门出去左转再直走", + "evidence": [ + { + "description": "Turn backward at the same position. There is a door on the left", + "timestamps": { + "start": "00:07:16", + "end": "00:07:18" + } + }, + { + "description": "Go out through this door", + "timestamps": { + "start": "00:02:22", + "end": "00:02:26" + } + }, + { + "description": "Turn left and go straight into the yard", + "timestamps": { + "start": "00:02:27", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:09:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "If you go to the yard now, you should turn left, go out through the door to your left, turn left, and then go straight.", + "B": "Exit through the front door.", + "C": "Use the door to your right.", + "D": "Turn left, go out through the door in front of you, then turn left and go straight.", + "E": "Turn left, exit through the door behind you, then turn left and continue straight ahead." + }, + "answer_label": "A", + "answer_text": "If you go to the yard now, you should turn left, go out through the door to your left, turn left, and then go straight.", + "q_id": "egolife_A1_D4_027_3_T2", + "correct_answer": "If you go to the yard now, you should turn left, go out through the door to your left, turn left, and then go straight.", + "original_options": { + "A": "If you go to the yard now, you should turn left, go out through the door to your left, turn left, and then go straight.", + "B": "Exit through the front door.", + "C": "Use the door to your right.", + "D": "Turn left, go out through the door in front of you, then turn left and go straight.", + "E": "Turn left, exit through the door behind you, then turn left and continue straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_4_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I go to the yard right now, how should I walk there?", + "answer": "Turn around and go straight, exit through the door to your front-left, then turn left and go straight.", + "question_chinese": "如果我现在去院子应该怎么走？", + "answer_chinese": "向后转直走，从左前方门出去左转直走", + "evidence": [ + { + "description": "Walk from the table to the current position", + "timestamps": { + "start": "00:09:10", + "end": "00:09:13" + } + }, + { + "description": "Go out through this door", + "timestamps": { + "start": "00:02:22", + "end": "00:02:26" + } + }, + { + "description": "Turn left and go straight into the yard", + "timestamps": { + "start": "00:02:27", + "end": "00:02:30" + } + }, + { + "description": "Right beside the table, I see a door on the left front", + "timestamps": { + "start": "00:07:52", + "end": "00:07:54" + } + } + ], + "question_timestamps": "00:09:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around and go straight, exit through the door to your front-left, then turn left and go straight.", + "B": "Turn around, exit front-right, then left.", + "C": "Turn around, exit the door behind you.", + "D": "Turn around and go straight, exit through the door to your front-right, then turn right and go straight.", + "E": "Turn around and go straight, go out the door to your left, then turn left and continue straight." + }, + "answer_label": "A", + "answer_text": "Turn around and go straight, exit through the door to your front-left, then turn left and go straight.", + "q_id": "egolife_A1_D4_027_4_T1", + "correct_answer": "Turn around and go straight, exit through the door to your front-left, then turn left and go straight.", + "original_options": { + "A": "Turn around and go straight, exit through the door to your front-left, then turn left and go straight.", + "B": "Turn around, exit front-right, then left.", + "C": "Turn around, exit the door behind you.", + "D": "Turn around and go straight, exit through the door to your front-right, then turn right and go straight.", + "E": "Turn around and go straight, go out the door to your left, then turn left and continue straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding while moving around?", + "answer": "I was just walking around with chopsticks in hand.", + "question_chinese": "我刚拿着什么东西走动？", + "answer_chinese": "我刚拿着筷子走动", + "evidence": [ + { + "description": "I'm walking around with chopsticks in hand", + "timestamps": { + "start": "00:00:16", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A fork in your hand.", + "B": "I was just walking around with chopsticks in hand.", + "C": "A book in your hand.", + "D": "You were walking around holding a sandwich behind your back the whole time.", + "E": "You were moving around with two pencils kept in your pocket, not carrying anything in hand." + }, + "answer_label": "B", + "answer_text": "I was just walking around with chopsticks in hand.", + "q_id": "egolife_A1_D4_027_1", + "correct_answer": "I was just walking around with chopsticks in hand.", + "original_options": { + "A": "A fork in your hand.", + "B": "I was just walking around with chopsticks in hand.", + "C": "A book in your hand.", + "D": "You were walking around holding a sandwich behind your back the whole time.", + "E": "You were moving around with two pencils kept in your pocket, not carrying anything in hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I just holding while moving around?", + "answer": "I was just walking around with my phone in hand.", + "question_chinese": "我刚拿着什么东西走动？", + "answer_chinese": "我刚拿着手机走动", + "evidence": [ + { + "description": "I'm walking around with my mobile phone in hand", + "timestamps": { + "start": "00:02:25", + "end": "00:02:27" + } + } + ], + "question_timestamps": "00:03:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Your keys in your hand.", + "B": "An umbrella behind your back.", + "C": "You were walking around with your backpack on your shoulder, not holding anything in your hand.", + "D": "I was just walking around with my phone in hand.", + "E": "You were moving around with your wallet in your pocket rather than carrying a device in hand." + }, + "answer_label": "D", + "answer_text": "I was just walking around with my phone in hand.", + "q_id": "egolife_A1_D4_027_2", + "correct_answer": "I was just walking around with my phone in hand.", + "original_options": { + "A": "Your keys in your hand.", + "B": "An umbrella behind your back.", + "C": "You were walking around with your backpack on your shoulder, not holding anything in your hand.", + "D": "I was just walking around with my phone in hand.", + "E": "You were moving around with your wallet in your pocket rather than carrying a device in hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027-mcq_fixed_9", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I run into a vase in this space now?", + "answer": "You're not allowed to touch vases in this space now.", + "question_chinese": "我现在可以在这个空间内碰到花瓶吗？", + "answer_chinese": "我现在不可以在这个空间内碰到花瓶", + "evidence": [ + { + "description": "The vase is on the living room table", + "timestamps": { + "start": "00:00:10", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:05:59", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You're not allowed to touch vases in this space now.", + "B": "You're allowed to touch vases in this space now." + }, + "answer_label": "A", + "answer_text": "You're not allowed to touch vases in this space now.", + "q_id": "egolife_A1_D4_027-mcq_fixed_9", + "correct_answer": "You're not allowed to touch vases in this space now.", + "original_options": { + "A": "You're not allowed to touch vases in this space now.", + "B": "You're allowed to touch vases in this space now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027-mcq_fixed_8", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I run into a vase in this space now?", + "answer": "YOU can now run into vases in this space.", + "question_chinese": "我现在可以在这个空间内碰到花瓶吗？", + "answer_chinese": "我现在可以在这个空间内碰到花瓶", + "evidence": [ + { + "description": "The vase is on the living room table", + "timestamps": { + "start": "00:00:10", + "end": "00:00:13" + } + }, + { + "description": "I'm in the living room.", + "timestamps": { + "start": "00:06:39", + "end": "00:06:40" + } + } + ], + "question_timestamps": "00:06:59", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "YOU cannot run into vases in this space now.", + "B": "YOU can now run into vases in this space." + }, + "answer_label": "B", + "answer_text": "YOU can now run into vases in this space.", + "q_id": "egolife_A1_D4_027-mcq_fixed_8", + "correct_answer": "YOU can now run into vases in this space.", + "original_options": { + "A": "YOU cannot run into vases in this space now.", + "B": "YOU can now run into vases in this space." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_5_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs do I see here?", + "answer": "You see 9 chairs here.", + "question_chinese": "我看到这里有几个椅子？", + "answer_chinese": "你看到这里有9个椅子", + "evidence": [ + { + "description": "There are nine chairs here", + "timestamps": { + "start": "00:02:35", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:06:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You see 10 chairs here.", + "B": "You see 8 chairs here.", + "C": "You see 11 chairs here, with most grouped together and a couple set apart nearby.", + "D": "You see 7 chairs here, with several around the area and the rest off to the side.", + "E": "You see 9 chairs here." + }, + "answer_label": "E", + "answer_text": "You see 9 chairs here.", + "q_id": "egolife_A1_D4_027_5_T1", + "correct_answer": "You see 9 chairs here.", + "original_options": { + "A": "You see 10 chairs here.", + "B": "You see 8 chairs here.", + "C": "You see 11 chairs here, with most grouped together and a couple set apart nearby.", + "D": "You see 7 chairs here, with several around the area and the rest off to the side.", + "E": "You see 9 chairs here." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_5_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs do I see here?", + "answer": "You see 9 chairs here.", + "question_chinese": "我看到这里有几个椅子？", + "answer_chinese": "你看到这里有9个椅子", + "evidence": [ + { + "description": "There are nine chairs here", + "timestamps": { + "start": "00:02:35", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You see 9 chairs here.", + "B": "You see 10 chairs here.", + "C": "You see 8 chairs here.", + "D": "You see 11 chairs here, with most grouped together and a couple set apart nearby.", + "E": "You see 7 chairs here, with several around the area and the rest off to the side." + }, + "answer_label": "A", + "answer_text": "You see 9 chairs here.", + "q_id": "egolife_A1_D4_027_5_T2", + "correct_answer": "You see 9 chairs here.", + "original_options": { + "A": "You see 9 chairs here.", + "B": "You see 10 chairs here.", + "C": "You see 8 chairs here.", + "D": "You see 11 chairs here, with most grouped together and a couple set apart nearby.", + "E": "You see 7 chairs here, with several around the area and the rest off to the side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_5_T3", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs do I see here?", + "answer": "You see 4 chairs here.", + "question_chinese": "我看到这里有几个椅子？", + "answer_chinese": "你看到这里有4个椅子", + "evidence": [ + { + "description": "There are four chairs here", + "timestamps": { + "start": "00:06:39", + "end": "00:06:45" + } + } + ], + "question_timestamps": "00:07:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You see 5 chairs here.", + "B": "You see 4 chairs here.", + "C": "You see 3 chairs here.", + "D": "You see 6 chairs here, with a few close together and the rest spaced out nearby.", + "E": "You see 2 chairs here, both positioned in front of you with no others visible." + }, + "answer_label": "B", + "answer_text": "You see 4 chairs here.", + "q_id": "egolife_A1_D4_027_5_T3", + "correct_answer": "You see 4 chairs here.", + "original_options": { + "A": "You see 5 chairs here.", + "B": "You see 4 chairs here.", + "C": "You see 3 chairs here.", + "D": "You see 6 chairs here, with a few close together and the rest spaced out nearby.", + "E": "You see 2 chairs here, both positioned in front of you with no others visible." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_5_T4", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs do I see here?", + "answer": "You see 4 chairs here.", + "question_chinese": "我看到这里有几个椅子？", + "answer_chinese": "你看到这里有4个椅子", + "evidence": [ + { + "description": "There are four chairs here", + "timestamps": { + "start": "00:06:39", + "end": "00:06:45" + } + } + ], + "question_timestamps": "00:07:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You see 3 chairs here.", + "B": "You see 5 chairs here.", + "C": "You see 4 chairs here.", + "D": "You see 2 chairs here, with two seats visible beside the table.", + "E": "You see 6 chairs here, spread around the room and partly out of frame." + }, + "answer_label": "C", + "answer_text": "You see 4 chairs here.", + "q_id": "egolife_A1_D4_027_5_T4", + "correct_answer": "You see 4 chairs here.", + "original_options": { + "A": "You see 3 chairs here.", + "B": "You see 5 chairs here.", + "C": "You see 4 chairs here.", + "D": "You see 2 chairs here, with two seats visible beside the table.", + "E": "You see 6 chairs here, spread around the room and partly out of frame." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_6", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the whiteboard from me?", + "answer": "The whiteboard is on my right.", + "question_chinese": "白板在我什么方向？", + "answer_chinese": "白板在我右边", + "evidence": [ + { + "description": "The whiteboard is on my left front", + "timestamps": { + "start": "00:07:20", + "end": "00:07:22" + } + }, + { + "description": "I'm walking forward, turning right, walking forward and turning right again", + "timestamps": { + "start": "00:07:22", + "end": "00:07:24" + } + } + ], + "question_timestamps": "00:07:44", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The whiteboard is on my left.", + "B": "The whiteboard is in front of me.", + "C": "The whiteboard is behind me on the left side as I look around the room.", + "D": "The whiteboard is on my right.", + "E": "The whiteboard is directly behind me, not to either side when I'm facing forward." + }, + "answer_label": "D", + "answer_text": "The whiteboard is on my right.", + "q_id": "egolife_A1_D4_027_6", + "correct_answer": "The whiteboard is on my right.", + "original_options": { + "A": "The whiteboard is on my left.", + "B": "The whiteboard is in front of me.", + "C": "The whiteboard is behind me on the left side as I look around the room.", + "D": "The whiteboard is on my right.", + "E": "The whiteboard is directly behind me, not to either side when I'm facing forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_5", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the whiteboard from me?", + "answer": "The whiteboard is behind me to the right.", + "question_chinese": "白板在我什么方向？", + "answer_chinese": "白板在我右后方", + "evidence": [ + { + "description": "The whiteboard is in front of me", + "timestamps": { + "start": "00:09:21", + "end": "00:09:22" + } + }, + { + "description": "I'm turning left.", + "timestamps": { + "start": "00:09:25", + "end": "00:09:27" + } + } + ], + "question_timestamps": "00:09:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The whiteboard is behind me to the right.", + "B": "The whiteboard is on my left.", + "C": "The whiteboard is in front of me.", + "D": "The whiteboard is behind me to the left, slightly off to that side.", + "E": "The whiteboard is in front-right of me, not behind me when I'm facing forward." + }, + "answer_label": "A", + "answer_text": "The whiteboard is behind me to the right.", + "q_id": "egolife_A1_D4_027_5", + "correct_answer": "The whiteboard is behind me to the right.", + "original_options": { + "A": "The whiteboard is behind me to the right.", + "B": "The whiteboard is on my left.", + "C": "The whiteboard is in front of me.", + "D": "The whiteboard is behind me to the left, slightly off to that side.", + "E": "The whiteboard is in front-right of me, not behind me when I'm facing forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_10", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there here?", + "answer": "There are nine chairs here.", + "question_chinese": "这里有几个椅子？", + "answer_chinese": "这里有9个椅子", + "evidence": [ + { + "description": "There are nine chairs here", + "timestamps": { + "start": "00:02:35", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are eight chairs here.", + "B": "There are nine chairs here.", + "C": "There are ten chairs here.", + "D": "There are seven chairs here, with four on one side and three on the other.", + "E": "There are 11 chairs here, including a couple pushed back away from the group." + }, + "answer_label": "B", + "answer_text": "There are nine chairs here.", + "q_id": "egolife_A1_D4_027_10", + "correct_answer": "There are nine chairs here.", + "original_options": { + "A": "There are eight chairs here.", + "B": "There are nine chairs here.", + "C": "There are ten chairs here.", + "D": "There are seven chairs here, with four on one side and three on the other.", + "E": "There are 11 chairs here, including a couple pushed back away from the group." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_027", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_027_9", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there here?", + "answer": "There are four chairs here.", + "question_chinese": "这里有几个椅子？", + "answer_chinese": "这里有4个椅子", + "evidence": [ + { + "description": "There are four chairs here", + "timestamps": { + "start": "00:06:39", + "end": "00:06:45" + } + } + ], + "question_timestamps": "00:07:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three chairs here.", + "B": "There are five chairs here.", + "C": "There are four chairs here.", + "D": "There are two chairs here, one near the table and one against the wall.", + "E": "There is one chair here, sitting by itself without any others nearby." + }, + "answer_label": "C", + "answer_text": "There are four chairs here.", + "q_id": "egolife_A1_D4_027_9", + "correct_answer": "There are four chairs here.", + "original_options": { + "A": "There are three chairs here.", + "B": "There are five chairs here.", + "C": "There are four chairs here.", + "D": "There are two chairs here, one near the table and one against the wall.", + "E": "There is one chair here, sitting by itself without any others nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 12 + ] + } +] diff --git a/QAs/egolife_A1_D4_028-mcq.json b/QAs/egolife_A1_D4_028-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1a51f5cbb345c91d0155c38928cf36744b14558a --- /dev/null +++ b/QAs/egolife_A1_D4_028-mcq.json @@ -0,0 +1,833 @@ +[ + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_9", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sweet potatoes were inside the baking tray the last time I saw it?", + "answer": "The last time I saw the baking tray, there were seven sweet potatoes in it.", + "question_chinese": "我上次看见烤盘时里面有几个红薯？", + "answer_chinese": "我上次看见烤盘时里面有7个红薯", + "evidence": [ + { + "description": "There are seven sweet potatoes in the baking pan", + "timestamps": { + "start": "00:02:45", + "end": "00:02:52" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There were six sweet potatoes in it.", + "B": "There were eight sweet potatoes in it.", + "C": "The last time I saw the baking tray, there were seven sweet potatoes in it.", + "D": "The last time I saw the baking tray, there were five sweet potatoes arranged in a single layer.", + "E": "The last time I saw the baking tray, there were nine sweet potatoes filling most of the tray." + }, + "answer_label": "C", + "answer_text": "The last time I saw the baking tray, there were seven sweet potatoes in it.", + "q_id": "egolife_A1_D4_028_9", + "correct_answer": "The last time I saw the baking tray, there were seven sweet potatoes in it.", + "original_options": { + "A": "There were six sweet potatoes in it.", + "B": "There were eight sweet potatoes in it.", + "C": "The last time I saw the baking tray, there were seven sweet potatoes in it.", + "D": "The last time I saw the baking tray, there were five sweet potatoes arranged in a single layer.", + "E": "The last time I saw the baking tray, there were nine sweet potatoes filling most of the tray." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_10", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sweet potatoes were inside the baking tray the last time I saw it?", + "answer": "The last time I saw the baking tray, there were six sweet potatoes in it.", + "question_chinese": "我上次看见烤盘时里面有几个红薯？", + "answer_chinese": "我上次看见烤盘时里面有6个红薯", + "evidence": [ + { + "description": "There are six\nThere are six sweet potatoes in the baking tray", + "timestamps": { + "start": "00:04:16", + "end": "00:04:24" + } + } + ], + "question_timestamps": "00:10:07", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There were five sweet potatoes in it.", + "B": "There were seven sweet potatoes in it.", + "C": "The last time I saw the baking tray, there were six sweet potatoes in it.", + "D": "The last time I saw the baking tray, there were four sweet potatoes spaced apart in the tray.", + "E": "The last time I saw the baking tray, there were eight sweet potatoes packed closely together." + }, + "answer_label": "C", + "answer_text": "The last time I saw the baking tray, there were six sweet potatoes in it.", + "q_id": "egolife_A1_D4_028_10", + "correct_answer": "The last time I saw the baking tray, there were six sweet potatoes in it.", + "original_options": { + "A": "There were five sweet potatoes in it.", + "B": "There were seven sweet potatoes in it.", + "C": "The last time I saw the baking tray, there were six sweet potatoes in it.", + "D": "The last time I saw the baking tray, there were four sweet potatoes spaced apart in the tray.", + "E": "The last time I saw the baking tray, there were eight sweet potatoes packed closely together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_4_T2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, am I closer to the bedroom or closer to the yard?", + "answer": "You were closer to the bedroom, and the bedroom was to your front-right.", + "question_chinese": "我此时离卧室更近还是离院子更近？", + "answer_chinese": "我离卧室更近，卧室在我右前方", + "evidence": [ + { + "description": "I'm standing beside the table. The bedroom is right in front and to the right of it.", + "timestamps": { + "start": "00:10:08", + "end": "00:10:16" + } + } + ], + "question_timestamps": "00:10:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were closer to the yard, and the yard was to your front-left.", + "B": "You were closer to the bedroom, and the bedroom was to your front-right." + }, + "answer_label": "B", + "answer_text": "You were closer to the bedroom, and the bedroom was to your front-right.", + "q_id": "egolife_A1_D4_028_4_T2", + "correct_answer": "You were closer to the bedroom, and the bedroom was to your front-right.", + "original_options": { + "A": "You were closer to the yard, and the yard was to your front-left.", + "B": "You were closer to the bedroom, and the bedroom was to your front-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_4_T1", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, am I closer to the bedroom or closer to the yard?", + "answer": "You were closer to the yard, and the yard was in front of you.", + "question_chinese": "我此时离卧室更近还是离院子更近？", + "answer_chinese": "我离院子更近，院子在你前面。", + "evidence": [ + { + "description": "The yard is outside the glass right opposite me", + "timestamps": { + "start": "00:02:55", + "end": "00:02:59" + } + } + ], + "question_timestamps": "00:11:44", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You were closer to the bedroom, and the bedroom was in front of you.", + "B": "You were closer to the yard, and the yard was in front of you." + }, + "answer_label": "B", + "answer_text": "You were closer to the yard, and the yard was in front of you.", + "q_id": "egolife_A1_D4_028_4_T1", + "correct_answer": "You were closer to the yard, and the yard was in front of you.", + "original_options": { + "A": "You were closer to the bedroom, and the bedroom was in front of you.", + "B": "You were closer to the yard, and the yard was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get some dishwashing liquid now?", + "answer": "I can touch dishwashing liquid.", + "question_chinese": "我现在能碰到洗洁精吗？", + "answer_chinese": "我可以碰到洗洁精", + "evidence": [ + { + "description": "The detergent is on the counter above the sink", + "timestamps": { + "start": "00:02:40", + "end": "00:03:09" + } + }, + { + "description": "I have been standing by the sink all the time", + "timestamps": { + "start": "00:03:10", + "end": "00:03:12" + } + } + ], + "question_timestamps": "00:03:28", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I can't touch dishwashing liquid.", + "B": "I can touch dishwashing liquid." + }, + "answer_label": "B", + "answer_text": "I can touch dishwashing liquid.", + "q_id": "egolife_A1_D4_028_1", + "correct_answer": "I can touch dishwashing liquid.", + "original_options": { + "A": "I can't touch dishwashing liquid.", + "B": "I can touch dishwashing liquid." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_1_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the dishwashing liquid right now?", + "answer": "I could reach the dish soap, and it was in front of you.", + "question_chinese": "我现在能碰到洗洁精吗？", + "answer_chinese": "我可以碰到洗洁精，洗洁精在我前方", + "evidence": [ + { + "description": "The detergent is on the counter above the sink", + "timestamps": { + "start": "00:02:40", + "end": "00:03:09" + } + }, + { + "description": "I have been standing by the sink all the time", + "timestamps": { + "start": "00:03:10", + "end": "00:03:12" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "I couldn't reach the dish soap, and it was in front of you.", + "B": "I could reach the dish soap, and it was in front of you." + }, + "answer_label": "B", + "answer_text": "I could reach the dish soap, and it was in front of you.", + "q_id": "egolife_A1_D4_028_1_T2", + "correct_answer": "I could reach the dish soap, and it was in front of you.", + "original_options": { + "A": "I couldn't reach the dish soap, and it was in front of you.", + "B": "I could reach the dish soap, and it was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_1_T3", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the dishwashing liquid right now?", + "answer": "You couldn’t reach the dish soap—it was in the kitchen, and you were in the bathroom at that moment.", + "question_chinese": "我现在能碰到洗洁精吗？", + "answer_chinese": "你不可以碰到洗洁精，洗洁精在厨房，而我此时在洗手间。", + "evidence": [ + { + "description": "You can't touch the detergent. It's in the kitchen while you're in the bathroom now.", + "timestamps": { + "start": "00:10:45", + "end": "00:10:50" + } + } + ], + "question_timestamps": "00:10:52", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn’t reach the dish soap—it was in the kitchen, and you were in the bathroom at that moment.", + "B": "You could reach the dish soap—it was in the kitchen with you at that moment." + }, + "answer_label": "A", + "answer_text": "You couldn’t reach the dish soap—it was in the kitchen, and you were in the bathroom at that moment.", + "q_id": "egolife_A1_D4_028_1_T3", + "correct_answer": "You couldn’t reach the dish soap—it was in the kitchen, and you were in the bathroom at that moment.", + "original_options": { + "A": "You couldn’t reach the dish soap—it was in the kitchen, and you were in the bathroom at that moment.", + "B": "You could reach the dish soap—it was in the kitchen with you at that moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get some dish soap right now?", + "answer": "I can't touch the detergent.", + "question_chinese": "我现在能碰到洗洁精吗？", + "answer_chinese": "我不能碰到洗洁精", + "evidence": [ + { + "description": "The detergent is on the counter above the sink", + "timestamps": { + "start": "00:02:40", + "end": "00:03:09" + } + }, + { + "description": "I'm in my bedroom now", + "timestamps": { + "start": "00:10:58", + "end": "00:11:00" + } + } + ], + "question_timestamps": "00:11:59", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I can touch the detergent.", + "B": "I can't touch the detergent." + }, + "answer_label": "B", + "answer_text": "I can't touch the detergent.", + "q_id": "egolife_A1_D4_028_2", + "correct_answer": "I can't touch the detergent.", + "original_options": { + "A": "I can touch the detergent.", + "B": "I can't touch the detergent." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_5_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many chairs did I see in the yard?", + "answer": "There are 9 chairs in the yard. Eight chairs are around the table, and one chair is in the corner above you to your left.", + "question_chinese": "我看见院子里面有几把椅子？", + "answer_chinese": "院子里面有9把椅子，8把椅子在桌子四周，还有一个椅子在我左上方的墙角", + "evidence": [ + { + "description": "There are seven chairs in the yard", + "timestamps": { + "start": "00:00:58", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:04:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 9 chairs in the yard. Eight chairs are around the table, and one chair is in the corner above you to your left.", + "B": "There are 8 chairs.", + "C": "There are 10 chairs.", + "D": "There are 8 chairs in the yard: seven around the table and one chair in the corner above you to your left.", + "E": "There are 10 chairs in the yard: eight around the table and two chairs in the corner above you to your left." + }, + "answer_label": "A", + "answer_text": "There are 9 chairs in the yard. Eight chairs are around the table, and one chair is in the corner above you to your left.", + "q_id": "egolife_A1_D4_028_5_T1", + "correct_answer": "There are 9 chairs in the yard. Eight chairs are around the table, and one chair is in the corner above you to your left.", + "original_options": { + "A": "There are 9 chairs in the yard. Eight chairs are around the table, and one chair is in the corner above you to your left.", + "B": "There are 8 chairs.", + "C": "There are 10 chairs.", + "D": "There are 8 chairs in the yard: seven around the table and one chair in the corner above you to your left.", + "E": "There are 10 chairs in the yard: eight around the table and two chairs in the corner above you to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 22, + 22 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_5_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many chairs did I see in the yard?", + "answer": "There are 7 chairs in the yard.", + "question_chinese": "我看见院子里面有几把椅子？", + "answer_chinese": "院子里面有7把椅子", + "evidence": [ + { + "description": "There are seven chairs in the yard", + "timestamps": { + "start": "00:00:58", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:04:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 6 chairs.", + "B": "There are 7 chairs in the yard.", + "C": "There are 8 chairs.", + "D": "There are 5 chairs in the yard, which is fewer than seven chairs total.", + "E": "There are 9 chairs in the yard, which is more than seven chairs total." + }, + "answer_label": "B", + "answer_text": "There are 7 chairs in the yard.", + "q_id": "egolife_A1_D4_028_5_T2", + "correct_answer": "There are 7 chairs in the yard.", + "original_options": { + "A": "There are 6 chairs.", + "B": "There are 7 chairs in the yard.", + "C": "There are 8 chairs.", + "D": "There are 5 chairs in the yard, which is fewer than seven chairs total.", + "E": "There are 9 chairs in the yard, which is more than seven chairs total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_8", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is my bedroom in now?", + "answer": "The bedroom is on my left.", + "question_chinese": "现在卧室在我的哪个方向？", + "answer_chinese": "卧室在我的左手边", + "evidence": [ + { + "description": "I turned 90 degrees to the right", + "timestamps": { + "start": "00:02:11", + "end": "00:02:14" + } + }, + { + "description": "The bedroom is right in front of me", + "timestamps": { + "start": "00:02:09", + "end": "00:02:11" + } + } + ], + "question_timestamps": "00:02:17", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The bedroom is on my right.", + "B": "The bedroom is in front of me.", + "C": "The bedroom is behind me rather than on my left side at this moment.", + "D": "The bedroom is above me instead of being positioned to my left right now.", + "E": "The bedroom is on my left." + }, + "answer_label": "E", + "answer_text": "The bedroom is on my left.", + "q_id": "egolife_A1_D4_028_8", + "correct_answer": "The bedroom is on my left.", + "original_options": { + "A": "The bedroom is on my right.", + "B": "The bedroom is in front of me.", + "C": "The bedroom is behind me rather than on my left side at this moment.", + "D": "The bedroom is above me instead of being positioned to my left right now.", + "E": "The bedroom is on my left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_7", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is my bedroom in now?", + "answer": "The bedroom is now behind me to the left.", + "question_chinese": "现在卧室在我的哪个方向？", + "answer_chinese": "现在卧室在我的左后方", + "evidence": [ + { + "description": "The bedroom is on my right front", + "timestamps": { + "start": "00:03:27", + "end": "00:03:30" + } + }, + { + "description": "I turned 180 degrees backward", + "timestamps": { + "start": "00:03:25", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:05:46", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of me to the right.", + "B": "Behind me to the right.", + "C": "The bedroom is now behind me to the left.", + "D": "The bedroom is now directly in front of me, slightly to the left as I face forward.", + "E": "The bedroom is now off to my left side, about even with me rather than behind." + }, + "answer_label": "C", + "answer_text": "The bedroom is now behind me to the left.", + "q_id": "egolife_A1_D4_028_7", + "correct_answer": "The bedroom is now behind me to the left.", + "original_options": { + "A": "In front of me to the right.", + "B": "Behind me to the right.", + "C": "The bedroom is now behind me to the left.", + "D": "The bedroom is now directly in front of me, slightly to the left as I face forward.", + "E": "The bedroom is now off to my left side, about even with me rather than behind." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_3", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors have I passed through so far?", + "answer": "So far, I have passed through four doors.", + "question_chinese": "目前为止我经过了几扇门", + "answer_chinese": "目前为止我经过四扇门", + "evidence": [ + { + "description": "I went through the door and into the bedroom", + "timestamps": { + "start": "00:00:38", + "end": "00:00:40" + } + }, + { + "description": "I went through the door and walked out of the bedroom", + "timestamps": { + "start": "00:00:45", + "end": "00:00:47" + } + }, + { + "description": "I walked out of the living room through the door", + "timestamps": { + "start": "00:00:50", + "end": "00:00:53" + } + }, + { + "description": "I walked past the door to the yard", + "timestamps": { + "start": "00:00:54", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:01:33", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I passed through three doors.", + "B": "I passed through five doors.", + "C": "So far, I have passed through two doors during the route up to this point.", + "D": "So far, I have passed through four doors.", + "E": "So far, I have passed through six doors, counting each doorway I went through while moving between rooms." + }, + "answer_label": "D", + "answer_text": "So far, I have passed through four doors.", + "q_id": "egolife_A1_D4_028_3", + "correct_answer": "So far, I have passed through four doors.", + "original_options": { + "A": "I passed through three doors.", + "B": "I passed through five doors.", + "C": "So far, I have passed through two doors during the route up to this point.", + "D": "So far, I have passed through four doors.", + "E": "So far, I have passed through six doors, counting each doorway I went through while moving between rooms." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_028_4", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors have I passed through so far?", + "answer": "So far, I have passed through six doors.", + "question_chinese": "目前为止我经过了几扇门", + "answer_chinese": "目前为止我经过了6扇门", + "evidence": [ + { + "description": "I went through the door and into the bedroom", + "timestamps": { + "start": "00:00:38", + "end": "00:00:40" + } + }, + { + "description": "I went through the door and walked out of the bedroom", + "timestamps": { + "start": "00:00:45", + "end": "00:00:47" + } + }, + { + "description": "I walked out of the living room through the door", + "timestamps": { + "start": "00:00:50", + "end": "00:00:53" + } + }, + { + "description": "I walked past the door to the yard", + "timestamps": { + "start": "00:00:54", + "end": "00:00:58" + } + }, + { + "description": "I passed through the door and returned to the house", + "timestamps": { + "start": "00:01:56", + "end": "00:02:00" + } + }, + { + "description": "I passed through the door and returned to the living room", + "timestamps": { + "start": "00:02:07", + "end": "00:02:09" + } + } + ], + "question_timestamps": "00:04:33", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I passed through five doors.", + "B": "I passed through seven doors.", + "C": "So far, I have passed through four doors on my way through the rooms so far.", + "D": "So far, I have passed through eight doors, counting every doorway I crossed during the path.", + "E": "So far, I have passed through six doors." + }, + "answer_label": "E", + "answer_text": "So far, I have passed through six doors.", + "q_id": "egolife_A1_D4_028_4", + "correct_answer": "So far, I have passed through six doors.", + "original_options": { + "A": "I passed through five doors.", + "B": "I passed through seven doors.", + "C": "So far, I have passed through four doors on my way through the rooms so far.", + "D": "So far, I have passed through eight doors, counting every doorway I crossed during the path.", + "E": "So far, I have passed through six doors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D4_029-mcq.json b/QAs/egolife_A1_D4_029-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8da7d307686e4a75a1ee93f6627c171a267b3e84 --- /dev/null +++ b/QAs/egolife_A1_D4_029-mcq.json @@ -0,0 +1,970 @@ +[ + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029_6", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "\"Who among you is closer to me?\"", + "answer": "A woman in pink.", + "question_chinese": "在座的各位谁离我更近", + "answer_chinese": "粉衣服的女人", + "evidence": [ + { + "description": "The woman in pink is sitting right beside me", + "timestamps": { + "start": "00:03:42", + "end": "00:03:45" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A woman in green.", + "B": "A man in blue.", + "C": "A woman in pink.", + "D": "The person closest to me is the child wearing yellow, not the woman in pink.", + "E": "It is the person dressed in white who is closer to me than the woman in pink." + }, + "answer_label": "C", + "answer_text": "A woman in pink.", + "q_id": "egolife_A1_D4_029_6", + "correct_answer": "A woman in pink.", + "original_options": { + "A": "A woman in green.", + "B": "A man in blue.", + "C": "A woman in pink.", + "D": "The person closest to me is the child wearing yellow, not the woman in pink.", + "E": "It is the person dressed in white who is closer to me than the woman in pink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029_5", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "\"Who among you is closest to me?\"", + "answer": "A person in black.", + "question_chinese": "在座的各位谁离我更近", + "answer_chinese": "穿黑衣服的人", + "evidence": [ + { + "description": "All three of them were sitting opposite me.", + "timestamps": { + "start": "00:05:28", + "end": "00:05:46" + } + }, + { + "description": "But the person in black was closer to the one directly opposite me", + "timestamps": { + "start": "00:05:28", + "end": "00:05:46" + } + } + ], + "question_timestamps": "00:08:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A person in front of me.", + "B": "A person to the left of me.", + "C": "A person in black.", + "D": "The closest one is the person behind me rather than the person in black.", + "E": "The person to my right is closest to me, not the person dressed in black." + }, + "answer_label": "C", + "answer_text": "A person in black.", + "q_id": "egolife_A1_D4_029_5", + "correct_answer": "A person in black.", + "original_options": { + "A": "A person in front of me.", + "B": "A person to the left of me.", + "C": "A person in black.", + "D": "The closest one is the person behind me rather than the person in black.", + "E": "The person to my right is closest to me, not the person dressed in black." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029_3", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The last time I passed by the bedroom door, which side was the bed on?", + "answer": "The bed is on my left.", + "question_chinese": "我上一次经过卧室门的时候，床在我的哪边", + "answer_chinese": "床在我的左边", + "evidence": [ + { + "description": "The bed is on my left", + "timestamps": { + "start": "00:00:37", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:01:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The bed is on my right.", + "B": "The bed is on my left.", + "C": "The bed is straight ahead.", + "D": "The bed was on my right side when I last passed the bedroom door.", + "E": "When I went by the bedroom door last time, the bed was positioned on the far side, not to my left." + }, + "answer_label": "B", + "answer_text": "The bed is on my left.", + "q_id": "egolife_A1_D4_029_3", + "correct_answer": "The bed is on my left.", + "original_options": { + "A": "The bed is on my right.", + "B": "The bed is on my left.", + "C": "The bed is straight ahead.", + "D": "The bed was on my right side when I last passed the bedroom door.", + "E": "When I went by the bedroom door last time, the bed was positioned on the far side, not to my left." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 21 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029_4", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "The last time I passed by the bedroom door, which side was the bed on?", + "answer": "The bed is on my right.", + "question_chinese": "我上一次经过卧室门的时候，床在我的哪边", + "answer_chinese": "床在我的右边", + "evidence": [ + { + "description": "The bed is on my right", + "timestamps": { + "start": "00:01:32", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:01:39", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The bed is on my left.", + "B": "The bed is on my right.", + "C": "The bed is straight ahead.", + "D": "The bed was on my left side the last time I passed by the bedroom door.", + "E": "When I last went past the bedroom door, the bed was on the opposite side of the room, not to my right." + }, + "answer_label": "B", + "answer_text": "The bed is on my right.", + "q_id": "egolife_A1_D4_029_4", + "correct_answer": "The bed is on my right.", + "original_options": { + "A": "The bed is on my left.", + "B": "The bed is on my right.", + "C": "The bed is straight ahead.", + "D": "The bed was on my left side the last time I passed by the bedroom door.", + "E": "When I last went past the bedroom door, the bed was on the opposite side of the room, not to my right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 22 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029-mcq_fixed_6", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take just now?", + "answer": "YOU just walked from the bedroom to the living room and then to the bathroom.", + "question_chinese": "我刚才走了什么路线？", + "answer_chinese": "我刚才从卧室走到客厅再到卫生间", + "evidence": [ + { + "description": "I'm walking from the bedroom to the living room and then to the bathroom", + "timestamps": { + "start": "00:01:36", + "end": "00:01:45" + } + } + ], + "question_timestamps": "00:02:29", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Living room to bedroom.", + "B": "YOU just walked from the bedroom to the living room and then to the bathroom.", + "C": "Bedroom to bathroom.", + "D": "YOU just walked from the bedroom into the bathroom first, then continued on into the living room afterward.", + "E": "YOU just walked from the living room to the bedroom and then went on to the bathroom at the end." + }, + "answer_label": "B", + "answer_text": "YOU just walked from the bedroom to the living room and then to the bathroom.", + "q_id": "egolife_A1_D4_029-mcq_fixed_6", + "correct_answer": "YOU just walked from the bedroom to the living room and then to the bathroom.", + "original_options": { + "A": "Living room to bedroom.", + "B": "YOU just walked from the bedroom to the living room and then to the bathroom.", + "C": "Bedroom to bathroom.", + "D": "YOU just walked from the bedroom into the bathroom first, then continued on into the living room afterward.", + "E": "YOU just walked from the living room to the bedroom and then went on to the bathroom at the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029-mcq_fixed_7", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take just now?", + "answer": "YOU just walked from the bedroom to the living room.", + "question_chinese": "我刚才走了什么路线？", + "answer_chinese": "我刚才从卧室走到客厅", + "evidence": [ + { + "description": "I'm walking from the bedroom to the living room", + "timestamps": { + "start": "00:03:29", + "end": "00:03:42" + } + } + ], + "question_timestamps": "00:04:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From living room to bedroom.", + "B": "YOU just walked from the bedroom to the living room.", + "C": "From bedroom to kitchen.", + "D": "You walked out of the bedroom, went down the hallway, and ended up in the kitchen.", + "E": "You walked out of the living room, went down the hallway, and ended up in the bedroom." + }, + "answer_label": "B", + "answer_text": "YOU just walked from the bedroom to the living room.", + "q_id": "egolife_A1_D4_029-mcq_fixed_7", + "correct_answer": "YOU just walked from the bedroom to the living room.", + "original_options": { + "A": "From living room to bedroom.", + "B": "YOU just walked from the bedroom to the living room.", + "C": "From bedroom to kitchen.", + "D": "You walked out of the bedroom, went down the hallway, and ended up in the kitchen.", + "E": "You walked out of the living room, went down the hallway, and ended up in the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029-mcq_fixed_9", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I been in the bedroom so far?", + "answer": "Six.", + "question_chinese": "我目前在卧室待了几次？", + "answer_chinese": "我目前在卧室待了1次", + "evidence": [ + { + "description": "I'm in the bedroom now", + "timestamps": { + "start": "00:00:00", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "Five.", + "C": "Seven.", + "D": "You have been in the bedroom four times so far in total.", + "E": "You have been in the bedroom eight times so far in total." + }, + "answer_label": "A", + "answer_text": "Six.", + "q_id": "egolife_A1_D4_029-mcq_fixed_9", + "correct_answer": "Six.", + "original_options": { + "A": "Six.", + "B": "Five.", + "C": "Seven.", + "D": "You have been in the bedroom four times so far in total.", + "E": "You have been in the bedroom eight times so far in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029-mcq_fixed_8", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many times have I been in the bedroom so far?", + "answer": "One.", + "question_chinese": "我目前在卧室待了几次？", + "answer_chinese": "我目前在卧室待了2次", + "evidence": [ + { + "description": "I'm in the bedroom now", + "timestamps": { + "start": "00:00:00", + "end": "00:00:35" + } + }, + { + "description": "I'm in the bedroom now", + "timestamps": { + "start": "00:01:34", + "end": "00:01:36" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "You have been in the bedroom four times so far in total.", + "E": "You have been in the bedroom five times so far in total." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "egolife_A1_D4_029-mcq_fixed_8", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "You have been in the bedroom four times so far in total.", + "E": "You have been in the bedroom five times so far in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029_9", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the restroom now?", + "answer": "Walked straight and turned right, then entered the bathroom through the left door.", + "question_chinese": "现在我要去卫生间该怎么走", + "answer_chinese": "直走然后右转，进入左手边的房子就进入了卫生间", + "evidence": [ + { + "description": "At this moment, I passed through the same door", + "timestamps": { + "start": "00:01:36", + "end": "00:01:37" + } + }, + { + "description": "I walked straight and turned right, then entered the bathroom through the left door", + "timestamps": { + "start": "00:01:37", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:03:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn left.", + "B": "Walked straight and turned right, then entered the bathroom through the left door.", + "C": "Turn right, then use the right door.", + "D": "Walk straight and turn right, then enter the bathroom through the right-side door instead of the left.", + "E": "Walk straight, turn left at the end, and enter the bathroom using the left door you see." + }, + "answer_label": "B", + "answer_text": "Walked straight and turned right, then entered the bathroom through the left door.", + "q_id": "egolife_A1_D4_029_9", + "correct_answer": "Walked straight and turned right, then entered the bathroom through the left door.", + "original_options": { + "A": "Go straight, then turn left.", + "B": "Walked straight and turned right, then entered the bathroom through the left door.", + "C": "Turn right, then use the right door.", + "D": "Walk straight and turn right, then enter the bathroom through the right-side door instead of the left.", + "E": "Walk straight, turn left at the end, and enter the bathroom using the left door you see." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029_10", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the restroom now?", + "answer": "Walk towards the area with air conditioning in the view, and then enter the room on the same side as in front of the air conditioning.", + "question_chinese": "现在我要去卫生间该怎么走", + "answer_chinese": "朝着视野中有空调的地方走，然后进入空调前面同一侧的房间", + "evidence": [ + { + "description": "I walked towards the bathroom", + "timestamps": { + "start": "00:01:38", + "end": "00:01:41" + } + }, + { + "description": "There is an air conditioner on the wall diagonally above the entrance door", + "timestamps": { + "start": "00:01:38", + "end": "00:01:42" + } + } + ], + "question_timestamps": "00:04:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Head toward the air conditioning, then turn away.", + "B": "Go toward the air conditioning and enter opposite.", + "C": "Walk toward the air conditioning you see, then enter the room on the opposite side from where it stands.", + "D": "Move toward the air conditioning in view, then enter the room on the other side instead of in front of it.", + "E": "Walk towards the area with air conditioning in the view, and then enter the room on the same side as in front of the air conditioning." + }, + "answer_label": "E", + "answer_text": "Walk towards the area with air conditioning in the view, and then enter the room on the same side as in front of the air conditioning.", + "q_id": "egolife_A1_D4_029_10", + "correct_answer": "Walk towards the area with air conditioning in the view, and then enter the room on the same side as in front of the air conditioning.", + "original_options": { + "A": "Head toward the air conditioning, then turn away.", + "B": "Go toward the air conditioning and enter opposite.", + "C": "Walk toward the air conditioning you see, then enter the room on the opposite side from where it stands.", + "D": "Move toward the air conditioning in view, then enter the room on the other side instead of in front of it.", + "E": "Walk towards the area with air conditioning in the view, and then enter the room on the same side as in front of the air conditioning." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029-mcq_fixed_5", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the bathroom now", + "answer": "Get up and enter the door in view, then walk straight, turn left into the living room, and then turn left into the first door on the left.", + "question_chinese": "现在我要去卫生间该怎么走", + "answer_chinese": "起身进入视野中的门然后直走左转进入客厅再左转，进入左手边第一个门", + "evidence": [ + { + "description": "The bathroom is in the first door on the side close to the living room door", + "timestamps": { + "start": "00:01:37", + "end": "00:01:43" + } + }, + { + "description": "I went out through the living room door, turned left and walked straight into the yard", + "timestamps": { + "start": "00:04:55", + "end": "00:05:11" + } + } + ], + "question_timestamps": "00:07:57", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right into the living room first.", + "B": "Use the second door on the left.", + "C": "Enter the door in view, walk straight, turn left into the living room, then turn right into the first door.", + "D": "Go through the visible door, continue straight, turn right into the living room, then take the first door on the left.", + "E": "Get up and enter the door in view, then walk straight, turn left into the living room, and then turn left into the first door on the left." + }, + "answer_label": "E", + "answer_text": "Get up and enter the door in view, then walk straight, turn left into the living room, and then turn left into the first door on the left.", + "q_id": "egolife_A1_D4_029-mcq_fixed_5", + "correct_answer": "Get up and enter the door in view, then walk straight, turn left into the living room, and then turn left into the first door on the left.", + "original_options": { + "A": "Turn right into the living room first.", + "B": "Use the second door on the left.", + "C": "Enter the door in view, walk straight, turn left into the living room, then turn right into the first door.", + "D": "Go through the visible door, continue straight, turn right into the living room, then take the first door on the left.", + "E": "Get up and enter the door in view, then walk straight, turn left into the living room, and then turn left into the first door on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029_7", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors have I passed through so far?", + "answer": "So far, I have passed through four doors.", + "question_chinese": "目前为止我经过了几扇门", + "answer_chinese": "目前为止我经过了4扇门", + "evidence": [ + { + "description": "I passed through the bedroom door and left the bedroom", + "timestamps": { + "start": "00:00:38", + "end": "00:00:40" + } + }, + { + "description": "I entered the bedroom through the bedroom door", + "timestamps": { + "start": "00:01:32", + "end": "00:01:34" + } + }, + { + "description": "I passed through the bedroom door and left the bedroom", + "timestamps": { + "start": "00:01:36", + "end": "00:01:38" + } + }, + { + "description": "I entered the bathroom through the bathroom door", + "timestamps": { + "start": "00:01:41", + "end": "00:01:43" + } + } + ], + "question_timestamps": "00:03:08", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "So far, I have passed through four doors.", + "B": "One door.", + "C": "Three doors.", + "D": "So far, I have passed through two doors in total so far.", + "E": "Up to now, I have passed through five doors in total so far." + }, + "answer_label": "A", + "answer_text": "So far, I have passed through four doors.", + "q_id": "egolife_A1_D4_029_7", + "correct_answer": "So far, I have passed through four doors.", + "original_options": { + "A": "So far, I have passed through four doors.", + "B": "One door.", + "C": "Three doors.", + "D": "So far, I have passed through two doors in total so far.", + "E": "Up to now, I have passed through five doors in total so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029_8", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many doors have I passed through so far?", + "answer": "So far, I have entered seven doors.", + "question_chinese": "目前为止我经过了几扇门", + "answer_chinese": "目前为止我经过七扇门", + "evidence": [ + { + "description": "I passed through the bedroom door and left the bedroom", + "timestamps": { + "start": "00:00:38", + "end": "00:00:40" + } + }, + { + "description": "I entered the bedroom through the bedroom door", + "timestamps": { + "start": "00:01:32", + "end": "00:01:34" + } + }, + { + "description": "I passed through the bedroom door and left the bedroom", + "timestamps": { + "start": "00:01:36", + "end": "00:01:38" + } + }, + { + "description": "I entered the bathroom through the bathroom door", + "timestamps": { + "start": "00:01:41", + "end": "00:01:43" + } + }, + { + "description": "I passed by the bathroom door and then walked out of it", + "timestamps": { + "start": "00:03:14", + "end": "00:03:18" + } + }, + { + "description": "I passed through the bedroom door and then walked into it", + "timestamps": { + "start": "00:03:21", + "end": "00:03:23" + } + }, + { + "description": "I passed through the bedroom door and then walked out of it", + "timestamps": { + "start": "00:03:30", + "end": "00:03:32" + } + } + ], + "question_timestamps": "00:04:53", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Six doors.", + "B": "So far, I have entered seven doors.", + "C": "Eight doors.", + "D": "So far, I have entered five doors in total so far today.", + "E": "So far, I have entered nine doors in total so far today." + }, + "answer_label": "B", + "answer_text": "So far, I have entered seven doors.", + "q_id": "egolife_A1_D4_029_8", + "correct_answer": "So far, I have entered seven doors.", + "original_options": { + "A": "Six doors.", + "B": "So far, I have entered seven doors.", + "C": "Eight doors.", + "D": "So far, I have entered five doors in total so far today.", + "E": "So far, I have entered nine doors in total so far today." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029_1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have I seen so far?", + "answer": "So far, I have met three people.", + "question_chinese": "目前为止我见过几个人？", + "answer_chinese": "目前为止我见过3个人", + "evidence": [ + { + "description": "There are three people in the living room", + "timestamps": { + "start": "00:01:23", + "end": "00:01:30" + } + } + ], + "question_timestamps": "00:05:09", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "So far, I have met two people.", + "B": "So far, I have met four people.", + "C": "So far, I have met five people while moving around and interacting briefly along the way.", + "D": "So far, I have met six people in total, including everyone you have spoken to up to now.", + "E": "So far, I have met three people." + }, + "answer_label": "E", + "answer_text": "So far, I have met three people.", + "q_id": "egolife_A1_D4_029_1", + "correct_answer": "So far, I have met three people.", + "original_options": { + "A": "So far, I have met two people.", + "B": "So far, I have met four people.", + "C": "So far, I have met five people while moving around and interacting briefly along the way.", + "D": "So far, I have met six people in total, including everyone you have spoken to up to now.", + "E": "So far, I have met three people." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_029_2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have I seen so far?", + "answer": "So far, I have met six people.", + "question_chinese": "目前为止我见过几个人？", + "answer_chinese": "目前为止我见过6个人", + "evidence": [ + { + "description": "There are three people in the living room", + "timestamps": { + "start": "00:01:23", + "end": "00:01:30" + } + }, + { + "description": "There are three people in the yard", + "timestamps": { + "start": "00:05:08", + "end": "00:05:27" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "So far, I have met five people.", + "B": "So far, I have met six people.", + "C": "So far, I have met seven people.", + "D": "So far, I have met four people total, counting only the people you actually interacted with.", + "E": "So far, I have met eight people in all, including those you passed by without stopping." + }, + "answer_label": "B", + "answer_text": "So far, I have met six people.", + "q_id": "egolife_A1_D4_029_2", + "correct_answer": "So far, I have met six people.", + "original_options": { + "A": "So far, I have met five people.", + "B": "So far, I have met six people.", + "C": "So far, I have met seven people.", + "D": "So far, I have met four people total, counting only the people you actually interacted with.", + "E": "So far, I have met eight people in all, including those you passed by without stopping." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D4_030-mcq.json b/QAs/egolife_A1_D4_030-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..181dfe2ea407b50ebf0f20d9fdbd14c6cc8de848 --- /dev/null +++ b/QAs/egolife_A1_D4_030-mcq.json @@ -0,0 +1,706 @@ +[ + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to go to the bedroom, how should I get there?", + "answer": "Go straigh.", + "question_chinese": "如果我想走向卧室我该怎么走", + "answer_chinese": "直走", + "evidence": [ + { + "description": "The position I am facing at the moment is where I am standing now", + "timestamps": { + "start": "00:01:13", + "end": "00:01:15" + } + }, + { + "description": "I was walking out of the bedroom at this moment", + "timestamps": { + "start": "00:01:05", + "end": "00:01:15" + } + } + ], + "question_timestamps": "00:04:38", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left.", + "B": "Turn right.", + "C": "Turn around first, then walk forward until you reach the bedroom entrance.", + "D": "Go straight for a bit, then take a left to enter the bedroom.", + "E": "Go straigh." + }, + "answer_label": "E", + "answer_text": "Go straigh.", + "q_id": "egolife_A1_D4_030_7", + "correct_answer": "Go straigh.", + "original_options": { + "A": "Turn left.", + "B": "Turn right.", + "C": "Turn around first, then walk forward until you reach the bedroom entrance.", + "D": "Go straight for a bit, then take a left to enter the bedroom.", + "E": "Go straigh." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030_8", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to go to the bedroom, how should I get there?", + "answer": "Table, turn right again, walk to the table, turn right again, and go straight.", + "question_chinese": "如果我想走向卧室我该怎么走", + "answer_chinese": "先左转，走到桌子边缘再右转，走到桌子边缘在右转，然后直走。", + "evidence": [ + { + "description": "I passed the table on the side where the chairs were placed and entered the bedroom", + "timestamps": { + "start": "00:01:00", + "end": "00:01:03" + } + }, + { + "description": "I am now opposite the side of the table with the chairs.", + "timestamps": { + "start": "00:00:00", + "end": "00:06:33" + } + } + ], + "question_timestamps": "00:06:47", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go straight.", + "B": "Walk to the table, then left.", + "C": "Table, turn right again, walk to the table, turn right again, and go straight.", + "D": "Table, turn left again, walk to the table, turn right again, and go straight.", + "E": "Table, turn right again, walk to the chair, turn right again, then go straight into the bedroom." + }, + "answer_label": "C", + "answer_text": "Table, turn right again, walk to the table, turn right again, and go straight.", + "q_id": "egolife_A1_D4_030_8", + "correct_answer": "Table, turn right again, walk to the table, turn right again, and go straight.", + "original_options": { + "A": "Turn left, then go straight.", + "B": "Walk to the table, then left.", + "C": "Table, turn right again, walk to the table, turn right again, and go straight.", + "D": "Table, turn left again, walk to the table, turn right again, and go straight.", + "E": "Table, turn right again, walk to the chair, turn right again, then go straight into the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030_6", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand the last time I was sitting down?", + "answer": "I haad my mobile phone.", + "question_chinese": "我上一次坐着的时候手上拿着什么", + "answer_chinese": "我拿着手机", + "evidence": [ + { + "description": "I sat in a chair outdoors with my mobile phone in my hand.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I had my computer.", + "B": "I had a card.", + "C": "I haad my mobile phone.", + "D": "The last time I sat down, I was holding a bowl in my hand.", + "E": "When I was sitting down last, I had a spatula in my hand instead." + }, + "answer_label": "C", + "answer_text": "I haad my mobile phone.", + "q_id": "egolife_A1_D4_030_6", + "correct_answer": "I haad my mobile phone.", + "original_options": { + "A": "I had my computer.", + "B": "I had a card.", + "C": "I haad my mobile phone.", + "D": "The last time I sat down, I was holding a bowl in my hand.", + "E": "When I was sitting down last, I had a spatula in my hand instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030_5", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand the last time I was sitting down?", + "answer": "I had my computer.", + "question_chinese": "我上一次坐着的时候手上拿着什么", + "answer_chinese": "我的电脑", + "evidence": [ + { + "description": "I sat in a chair in the living room with a computer.", + "timestamps": { + "start": "00:01:20", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I had my tablet.", + "B": "I had my book.", + "C": "I had my glasses in my hand the last time I was sitting down.", + "D": "I had my game controller in my hand when I last sat down.", + "E": "I had my computer." + }, + "answer_label": "E", + "answer_text": "I had my computer.", + "q_id": "egolife_A1_D4_030_5", + "correct_answer": "I had my computer.", + "original_options": { + "A": "I had my tablet.", + "B": "I had my book.", + "C": "I had my glasses in my hand the last time I was sitting down.", + "D": "I had my game controller in my hand when I last sat down.", + "E": "I had my computer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030_3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did I just take with my laptop?", + "answer": "I took from bedroom to living room.", + "question_chinese": "我刚拿着我的笔记本电脑走了什么路线", + "answer_chinese": "我把它从卧室拿到客厅", + "evidence": [ + { + "description": "I took my notebook out of the quilt in my bedroom and went to the living room to sit down.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:19" + } + } + ], + "question_timestamps": "00:02:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I took from bedroom to living room.", + "B": "I took from living room to bedroom.", + "C": "I took from bedroom to kitchen.", + "D": "I took from hallway to living room with my laptop before stopping there.", + "E": "I took from bedroom to dining room, carrying my laptop the whole way." + }, + "answer_label": "A", + "answer_text": "I took from bedroom to living room.", + "q_id": "egolife_A1_D4_030_3", + "correct_answer": "I took from bedroom to living room.", + "original_options": { + "A": "I took from bedroom to living room.", + "B": "I took from living room to bedroom.", + "C": "I took from bedroom to kitchen.", + "D": "I took from hallway to living room with my laptop before stopping there.", + "E": "I took from bedroom to dining room, carrying my laptop the whole way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030_4", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What route did I just take with my laptop?", + "answer": "I took it to get a roasted sweet potato and put it back on my seat.", + "question_chinese": "我刚拿着我的笔记本电脑走了什么路线", + "answer_chinese": "我拿起我的笔记本起来去拿了一个红薯，并且又走回去把电脑放回座位上。", + "evidence": [ + { + "description": "I picked up my notebook and got up to get a sweet potato, and walked back to put the computer back on my seat.", + "timestamps": { + "start": "00:02:52", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I got a sandwich and returned it.", + "B": "I got a hot drink and returned it.", + "C": "I took it to get a roasted sweet potato and put it back on my seat.", + "D": "I took it to get a salad and then put it back in my bag afterward.", + "E": "I took it to get a chocolate croissant and put it back on the counter when done." + }, + "answer_label": "C", + "answer_text": "I took it to get a roasted sweet potato and put it back on my seat.", + "q_id": "egolife_A1_D4_030_4", + "correct_answer": "I took it to get a roasted sweet potato and put it back on my seat.", + "original_options": { + "A": "I got a sandwich and returned it.", + "B": "I got a hot drink and returned it.", + "C": "I took it to get a roasted sweet potato and put it back on my seat.", + "D": "I took it to get a salad and then put it back in my bag afterward.", + "E": "I took it to get a chocolate croissant and put it back on the counter when done." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030-mcq_fixed_7", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the computer right where I am now?", + "answer": "YOU can't touch the computer right where You are now.", + "question_chinese": "我现在在原地能碰到电脑吗？", + "answer_chinese": "我现在在原地不能碰到电脑", + "evidence": [ + { + "description": "There is a computer on the stool near the whiteboard", + "timestamps": { + "start": "00:03:00", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "YOU can't touch the computer right where You are now.", + "B": "YOU can touch the computer right where You are now." + }, + "answer_label": "A", + "answer_text": "YOU can't touch the computer right where You are now.", + "q_id": "egolife_A1_D4_030-mcq_fixed_7", + "correct_answer": "YOU can't touch the computer right where You are now.", + "original_options": { + "A": "YOU can't touch the computer right where You are now.", + "B": "YOU can touch the computer right where You are now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030-mcq_fixed_6", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the computer right where I am now?", + "answer": "YOU can touch the computer right where You are now.", + "question_chinese": "我现在在原地能碰到电脑吗？", + "answer_chinese": "我现在在原地能碰到电脑", + "evidence": [ + { + "description": "There are many computers in the bedroom", + "timestamps": { + "start": "00:10:12", + "end": "00:10:14" + } + }, + { + "description": "I'm in the bedroom.", + "timestamps": { + "start": "00:10:15", + "end": "00:10:17" + } + } + ], + "question_timestamps": "00:11:20", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "YOU cannot touch the computer right where You are now.", + "B": "YOU can touch the computer right where You are now." + }, + "answer_label": "B", + "answer_text": "YOU can touch the computer right where You are now.", + "q_id": "egolife_A1_D4_030-mcq_fixed_6", + "correct_answer": "YOU can touch the computer right where You are now.", + "original_options": { + "A": "YOU cannot touch the computer right where You are now.", + "B": "YOU can touch the computer right where You are now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030_10", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the girl in the red dress or the boy in the light blue shirt?", + "answer": "I am closer to the boy in light blue.", + "question_chinese": "我现在是离红衣服女生更近还是离浅蓝色男生更近", + "answer_chinese": "我离浅蓝色衣服的男生更近", + "evidence": [ + { + "description": "The girl in red is sitting on the stool in front of me", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + }, + { + "description": "The boy in blue passed where the girl in red was and walked on", + "timestamps": { + "start": "00:01:28", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I am closer to the girl in the red dress.", + "B": "I am closer to the boy in light blue." + }, + "answer_label": "B", + "answer_text": "I am closer to the boy in light blue.", + "q_id": "egolife_A1_D4_030_10", + "correct_answer": "I am closer to the boy in light blue.", + "original_options": { + "A": "I am closer to the girl in the red dress.", + "B": "I am closer to the boy in light blue." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030_9", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Am I closer to the girl in the red dress or the boy in the light blue shirt?", + "answer": "Closer to the girl in red.", + "question_chinese": "我现在是离红衣服女生更近还是离浅蓝色男生更近", + "answer_chinese": "离红衣服女生更近", + "evidence": [ + { + "description": "Now this girl's arm is very shocked from me, no more than 5cm", + "timestamps": { + "start": "00:08:29", + "end": "00:08:31" + } + }, + { + "description": "IIt's about half a meter away from the boys", + "timestamps": { + "start": "00:08:32", + "end": "00:08:39" + } + } + ], + "question_timestamps": "00:09:27", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Closer to the girl in red.", + "B": "Closer to the boy in light blue." + }, + "answer_label": "A", + "answer_text": "Closer to the girl in red.", + "q_id": "egolife_A1_D4_030_9", + "correct_answer": "Closer to the girl in red.", + "original_options": { + "A": "Closer to the girl in red.", + "B": "Closer to the boy in light blue." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030-mcq_fixed_8", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the yard now?", + "answer": "Turn right and go straight out of the living room. Then turn right and go straight out again.", + "question_chinese": "我现在要去院子该怎么走？", + "answer_chinese": "右转出门直走出客厅最转再直走出门", + "evidence": [ + { + "description": "I walked straight through the entrance of the yard", + "timestamps": { + "start": "00:00:53", + "end": "00:00:57" + } + }, + { + "description": "Turn left into the living room", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + }, + { + "description": "Walk straight into the bedroom", + "timestamps": { + "start": "00:00:59", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go straight out.", + "B": "Go straight out, then turn right.", + "C": "Turn right and go straight out of the living room. Then turn right and go straight out again.", + "D": "Turn right and go out of the living room, then turn left and go straight out again.", + "E": "Turn left out of the living room, walk straight, then turn right and go straight out again." + }, + "answer_label": "C", + "answer_text": "Turn right and go straight out of the living room. Then turn right and go straight out again.", + "q_id": "egolife_A1_D4_030-mcq_fixed_8", + "correct_answer": "Turn right and go straight out of the living room. Then turn right and go straight out again.", + "original_options": { + "A": "Turn left, then go straight out.", + "B": "Go straight out, then turn right.", + "C": "Turn right and go straight out of the living room. Then turn right and go straight out again.", + "D": "Turn right and go out of the living room, then turn left and go straight out again.", + "E": "Turn left out of the living room, walk straight, then turn right and go straight out again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_030", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_030-mcq_fixed_9", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the yard now?", + "answer": "Go out through the door on the right, turn left and walk straight out.", + "question_chinese": "我现在要去院子该怎么走？", + "answer_chinese": "从右手边的门出去左转直走出门", + "evidence": [ + { + "description": "I walked straight through the entrance of the yard", + "timestamps": { + "start": "00:00:53", + "end": "00:00:57" + } + }, + { + "description": "Turn left into the living room", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Use the door on the left.", + "B": "Go right, then turn right.", + "C": "Go out through the door on the right, turn left and walk straight out.", + "D": "Go out through the door on the left, turn right, and walk straight out from there.", + "E": "Go out through the door on the right, turn right, and keep walking straight to exit." + }, + "answer_label": "C", + "answer_text": "Go out through the door on the right, turn left and walk straight out.", + "q_id": "egolife_A1_D4_030-mcq_fixed_9", + "correct_answer": "Go out through the door on the right, turn left and walk straight out.", + "original_options": { + "A": "Use the door on the left.", + "B": "Go right, then turn right.", + "C": "Go out through the door on the right, turn left and walk straight out.", + "D": "Go out through the door on the left, turn right, and walk straight out from there.", + "E": "Go out through the door on the right, turn right, and keep walking straight to exit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D4_031-mcq.json b/QAs/egolife_A1_D4_031-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..11965b193c0bfa7a7f3e711fda1d4b78da07f5d8 --- /dev/null +++ b/QAs/egolife_A1_D4_031-mcq.json @@ -0,0 +1,1113 @@ +[ + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_6_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk to the courtyard?", + "answer": "You turned left first.", + "question_chinese": "我去庭院是怎么走的？", + "answer_chinese": "你先左转", + "evidence": [ + { + "description": "I walked out the same door.", + "timestamps": { + "start": "00:03:04", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:03:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right first.", + "B": "You went straight first.", + "C": "You turned right first, then walked straight, opened the door, and entered the courtyard.", + "D": "You turned left first.", + "E": "You turned around first, went downstairs, then continued straight until you reached the courtyard." + }, + "answer_label": "D", + "answer_text": "You turned left first.", + "q_id": "egolife_A1_D4_031_6_T2", + "correct_answer": "You turned left first.", + "original_options": { + "A": "You turned right first.", + "B": "You went straight first.", + "C": "You turned right first, then walked straight, opened the door, and entered the courtyard.", + "D": "You turned left first.", + "E": "You turned around first, went downstairs, then continued straight until you reached the courtyard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_6_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk to the courtyard?", + "answer": "You turned left first, then walked straight, opened the door, and arrived at the courtyard.", + "question_chinese": "我去庭院是怎么走的？", + "answer_chinese": "你先左转了，然后直走开门到达庭院", + "evidence": [ + { + "description": "I walked out the same door. Then I turned left to go out and then entered the courtyard", + "timestamps": { + "start": "00:03:06", + "end": "00:03:12" + } + } + ], + "question_timestamps": "00:07:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right first.", + "B": "You went downstairs first.", + "C": "You turned left first, walked straight, then turned around before opening the door to the courtyard.", + "D": "You turned left first, walked straight past the door, then turned back and entered the courtyard.", + "E": "You turned left first, then walked straight, opened the door, and arrived at the courtyard." + }, + "answer_label": "E", + "answer_text": "You turned left first, then walked straight, opened the door, and arrived at the courtyard.", + "q_id": "egolife_A1_D4_031_6_T1", + "correct_answer": "You turned left first, then walked straight, opened the door, and arrived at the courtyard.", + "original_options": { + "A": "You turned right first.", + "B": "You went downstairs first.", + "C": "You turned left first, walked straight, then turned around before opening the door to the courtyard.", + "D": "You turned left first, walked straight past the door, then turned back and entered the courtyard.", + "E": "You turned left first, then walked straight, opened the door, and arrived at the courtyard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the table now?", + "answer": "Yes,the table is about eight cm from you.", + "question_chinese": "我现在能碰到桌子吗", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The table is clase to me ,not more than 20 cm.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:53" + } + } + ], + "question_timestamps": "00:00:58", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,the table is about eight cm from you.", + "B": "No, the table is too far to reach." + }, + "answer_label": "A", + "answer_text": "Yes,the table is about eight cm from you.", + "q_id": "egolife_A1_D4_031_2", + "correct_answer": "Yes,the table is about eight cm from you.", + "original_options": { + "A": "Yes,the table is about eight cm from you.", + "B": "No, the table is too far to reach." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_1", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the table now?", + "answer": "No,the table is more than one meter from you.", + "question_chinese": "我现在能碰到桌子吗", + "answer_chinese": "不行", + "evidence": [ + { + "description": "The chair is far from the table and I sat down on the chair", + "timestamps": { + "start": "00:00:57", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:01:42", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can reach it from here.", + "B": "No,the table is more than one meter from you." + }, + "answer_label": "B", + "answer_text": "No,the table is more than one meter from you.", + "q_id": "egolife_A1_D4_031_1", + "correct_answer": "No,the table is more than one meter from you.", + "original_options": { + "A": "Yes, you can reach it from here.", + "B": "No,the table is more than one meter from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_7", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in this room right now?", + "answer": "There are three people.", + "question_chinese": "此时这个房间有几个人", + "answer_chinese": "3个", + "evidence": [ + { + "description": "There are actually two people on the sofa.", + "timestamps": { + "start": "00:10:20", + "end": "00:10:33" + } + }, + { + "description": "I am in this room.", + "timestamps": { + "start": "00:10:20", + "end": "00:10:33" + } + } + ], + "question_timestamps": "00:10:44", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are two people.", + "B": "There are four people.", + "C": "There are five people standing in the room right now with you.", + "D": "There are three people.", + "E": "There are two people in this room at the moment, not counting you." + }, + "answer_label": "D", + "answer_text": "There are three people.", + "q_id": "egolife_A1_D4_031_7", + "correct_answer": "There are three people.", + "original_options": { + "A": "There are two people.", + "B": "There are four people.", + "C": "There are five people standing in the room right now with you.", + "D": "There are three people.", + "E": "There are two people in this room at the moment, not counting you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_8", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in this room right now?", + "answer": "There are four people.", + "question_chinese": "此时这个房间有几个人", + "answer_chinese": "4个人", + "evidence": [ + { + "description": "At this time, a woman enters the door.", + "timestamps": { + "start": "00:11:42", + "end": "00:11:43" + } + }, + { + "description": "There are actually two people on the sofa and I am in this room.", + "timestamps": { + "start": "00:10:20", + "end": "00:10:32" + } + } + ], + "question_timestamps": "00:11:53", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are three people.", + "B": "There are five people.", + "C": "There are two people in this room right now, including the camera wearer.", + "D": "There are six people in the room at the moment, including you as well.", + "E": "There are four people." + }, + "answer_label": "E", + "answer_text": "There are four people.", + "q_id": "egolife_A1_D4_031_8", + "correct_answer": "There are four people.", + "original_options": { + "A": "There are three people.", + "B": "There are five people.", + "C": "There are two people in this room right now, including the camera wearer.", + "D": "There are six people in the room at the moment, including you as well.", + "E": "There are four people." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_7_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the courtyard from where I am right now?", + "answer": "You first turned around and went downstairs.", + "question_chinese": "现在我要怎么到达庭院？", + "answer_chinese": "你首先转身下楼了", + "evidence": [ + { + "description": "I went downstairs to face the door which leading to the courtyard", + "timestamps": { + "start": "00:05:05", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left first.", + "B": "You went upstairs first.", + "C": "You first walk straight ahead, then turn left to head toward the courtyard.", + "D": "You first turned around and went downstairs.", + "E": "You first turn right, then go downstairs, and continue until you reach the courtyard." + }, + "answer_label": "D", + "answer_text": "You first turned around and went downstairs.", + "q_id": "egolife_A1_D4_031_7_T2", + "correct_answer": "You first turned around and went downstairs.", + "original_options": { + "A": "You turned left first.", + "B": "You went upstairs first.", + "C": "You first walk straight ahead, then turn left to head toward the courtyard.", + "D": "You first turned around and went downstairs.", + "E": "You first turn right, then go downstairs, and continue until you reach the courtyard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_7_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the courtyard from where I am right now?", + "answer": "You first turned around, went downstairs, and then walked straight to reach the courtyard.", + "question_chinese": "现在我要怎么到达庭院？", + "answer_chinese": "你首先转身下楼然后直走到达庭院", + "evidence": [ + { + "description": "I went downstairs to face the door which leading to the courtyard", + "timestamps": { + "start": "00:05:07", + "end": "00:05:23" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left first.", + "B": "You went upstairs first.", + "C": "You first turned around, went downstairs, and then walked straight to reach the courtyard.", + "D": "You turn around and walk straight first, then go downstairs to get to the courtyard.", + "E": "You turn around, go downstairs, then turn left before continuing toward the courtyard." + }, + "answer_label": "C", + "answer_text": "You first turned around, went downstairs, and then walked straight to reach the courtyard.", + "q_id": "egolife_A1_D4_031_7_T1", + "correct_answer": "You first turned around, went downstairs, and then walked straight to reach the courtyard.", + "original_options": { + "A": "You turned left first.", + "B": "You went upstairs first.", + "C": "You first turned around, went downstairs, and then walked straight to reach the courtyard.", + "D": "You turn around and walk straight first, then go downstairs to get to the courtyard.", + "E": "You turn around, go downstairs, then turn left before continuing toward the courtyard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_9_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, how many doors have I passed through?", + "answer": "So far, you have passed through one door.", + "question_chinese": "目前为止我经过了几扇门？", + "answer_chinese": "目前你经过了一扇门", + "evidence": [ + { + "description": "I passed the bedroom door.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero doors.", + "B": "Two doors.", + "C": "So far, you have already passed through three different doors in total.", + "D": "Up to this point, you have passed through four doors while moving around the building.", + "E": "So far, you have passed through one door." + }, + "answer_label": "E", + "answer_text": "So far, you have passed through one door.", + "q_id": "egolife_A1_D4_031_9_T2", + "correct_answer": "So far, you have passed through one door.", + "original_options": { + "A": "Zero doors.", + "B": "Two doors.", + "C": "So far, you have already passed through three different doors in total.", + "D": "Up to this point, you have passed through four doors while moving around the building.", + "E": "So far, you have passed through one door." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_9_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, how many doors have I passed through?", + "answer": "So far, you have passed through three doors.", + "question_chinese": "目前为止我经过了几扇门？", + "answer_chinese": "目前你经过了三扇门", + "evidence": [ + { + "description": "I passed the living room door.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:07" + } + }, + { + "description": "I passed the bedroom door.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + }, + { + "description": "I passed the door leading to the yard.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:12" + } + } + ], + "question_timestamps": "00:03:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One door.", + "B": "Two doors.", + "C": "So far, you have already passed through four different doors in total.", + "D": "So far, you have passed through three doors.", + "E": "Up to this point, you have passed through five doors while walking through the building." + }, + "answer_label": "D", + "answer_text": "So far, you have passed through three doors.", + "q_id": "egolife_A1_D4_031_9_T1", + "correct_answer": "So far, you have passed through three doors.", + "original_options": { + "A": "One door.", + "B": "Two doors.", + "C": "So far, you have already passed through four different doors in total.", + "D": "So far, you have passed through three doors.", + "E": "Up to this point, you have passed through five doors while walking through the building." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_4", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many doors have I passed through so far?", + "answer": "I passed four doors.", + "question_chinese": "目前为止我经过了几扇门", + "answer_chinese": "4", + "evidence": [ + { + "description": "I passed the living room door.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:07" + } + }, + { + "description": "I passed the bedroom door.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + }, + { + "description": "I passed the door leading to the yard.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:12" + } + }, + { + "description": "I passed the door of my home.", + "timestamps": { + "start": "00:03:15", + "end": "00:03:16" + } + } + ], + "question_timestamps": "00:03:35", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I passed three doors.", + "B": "I passed four doors.", + "C": "I passed five doors.", + "D": "I passed two doors before reaching this point, not four in total.", + "E": "I passed six doors altogether so far, counting each doorway you went through." + }, + "answer_label": "B", + "answer_text": "I passed four doors.", + "q_id": "egolife_A1_D4_031_4", + "correct_answer": "I passed four doors.", + "original_options": { + "A": "I passed three doors.", + "B": "I passed four doors.", + "C": "I passed five doors.", + "D": "I passed two doors before reaching this point, not four in total.", + "E": "I passed six doors altogether so far, counting each doorway you went through." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_8_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, how many doors have I passed through?", + "answer": "So far, you have passed through four doors.", + "question_chinese": "目前为止我经过了几扇门？", + "answer_chinese": "目前你经过了四扇门", + "evidence": [ + { + "description": "I passed the living room door.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:07" + } + }, + { + "description": "I passed the bedroom door.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + }, + { + "description": "I passed the door leading to the yard.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:12" + } + }, + { + "description": "I passed the door of my room.", + "timestamps": { + "start": "00:03:15", + "end": "00:03:16" + } + } + ], + "question_timestamps": "00:04:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, you have passed through three doors.", + "B": "So far, you have passed through five doors.", + "C": "So far, you have passed through two doors, not four, up to this moment.", + "D": "So far, you have passed through four doors.", + "E": "So far, you have passed through six doors in total along the route you've taken." + }, + "answer_label": "D", + "answer_text": "So far, you have passed through four doors.", + "q_id": "egolife_A1_D4_031_8_T2", + "correct_answer": "So far, you have passed through four doors.", + "original_options": { + "A": "So far, you have passed through three doors.", + "B": "So far, you have passed through five doors.", + "C": "So far, you have passed through two doors, not four, up to this moment.", + "D": "So far, you have passed through four doors.", + "E": "So far, you have passed through six doors in total along the route you've taken." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_8_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "So far, how many doors have I passed through?", + "answer": "So far, you have passed through five doors.", + "question_chinese": "目前为止我经过了几扇门？", + "answer_chinese": "目前你经过了五扇门", + "evidence": [ + { + "description": "I passed the living room door.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:07" + } + }, + { + "description": "I passed the bedroom door.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + }, + { + "description": "I passed the door leading to the yard.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:12" + } + }, + { + "description": "I passed the door of my room.", + "timestamps": { + "start": "00:03:15", + "end": "00:03:16" + } + }, + { + "description": "I went home through the gate.", + "timestamps": { + "start": "00:04:31", + "end": "00:04:33" + } + } + ], + "question_timestamps": "00:04:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "So far, you have passed through four doors.", + "B": "So far, you have passed through six doors.", + "C": "So far, you have passed through five doors.", + "D": "So far, you have passed through three doors, not five, up to this point.", + "E": "So far, you have passed through seven doors altogether while moving between the areas." + }, + "answer_label": "C", + "answer_text": "So far, you have passed through five doors.", + "q_id": "egolife_A1_D4_031_8_T1", + "correct_answer": "So far, you have passed through five doors.", + "original_options": { + "A": "So far, you have passed through four doors.", + "B": "So far, you have passed through six doors.", + "C": "So far, you have passed through five doors.", + "D": "So far, you have passed through three doors, not five, up to this point.", + "E": "So far, you have passed through seven doors altogether while moving between the areas." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_3", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many doors have I passed through so far?", + "answer": "I passed five doors.", + "question_chinese": "目前为止我经过了几扇门", + "answer_chinese": "我经过了5扇门", + "evidence": [ + { + "description": "I passed the living room door.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:07" + } + }, + { + "description": "I passed the bedroom door.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + }, + { + "description": "I passed the door leading to the yard.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:12" + } + }, + { + "description": "I passed the door of my room.", + "timestamps": { + "start": "00:03:15", + "end": "00:03:16" + } + }, + { + "description": "I went home through the gate.", + "timestamps": { + "start": "00:04:31", + "end": "00:04:33" + } + } + ], + "question_timestamps": "00:04:47", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I passed four doors.", + "B": "I passed six doors.", + "C": "I passed five doors.", + "D": "I passed three doors so far, not five, before reaching where you are now.", + "E": "I passed seven doors in total, counting every doorway you went through along the way." + }, + "answer_label": "C", + "answer_text": "I passed five doors.", + "q_id": "egolife_A1_D4_031_3", + "correct_answer": "I passed five doors.", + "original_options": { + "A": "I passed four doors.", + "B": "I passed six doors.", + "C": "I passed five doors.", + "D": "I passed three doors so far, not five, before reaching where you are now.", + "E": "I passed seven doors in total, counting every doorway you went through along the way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_3_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Relative to where I am right now, where is the window?", + "answer": "The window is to your left.", + "question_chinese": "相对于我现在的位置，窗户在哪？", + "answer_chinese": "窗户在你左边", + "evidence": [ + { + "description": "Now the window is in front of me.", + "timestamps": { + "start": "00:10:42", + "end": "00:10:43" + } + } + ], + "question_timestamps": "00:10:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is to your left.", + "B": "The window is to your right.", + "C": "The window is behind you.", + "D": "The window is straight in front of you relative to your current position.", + "E": "The window is above you rather than off to the side where you're standing." + }, + "answer_label": "A", + "answer_text": "The window is to your left.", + "q_id": "egolife_A1_D4_031_3_T2", + "correct_answer": "The window is to your left.", + "original_options": { + "A": "The window is to your left.", + "B": "The window is to your right.", + "C": "The window is behind you.", + "D": "The window is straight in front of you relative to your current position.", + "E": "The window is above you rather than off to the side where you're standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_3_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Relative to where I am right now, where is the window?", + "answer": "The window is behind you.", + "question_chinese": "相对于我现在的位置，窗户在哪？", + "answer_chinese": "窗户在你后面", + "evidence": [ + { + "description": "Now the window is in front of me.", + "timestamps": { + "start": "00:10:42", + "end": "00:10:43" + } + }, + { + "description": "Turned to face the same table.", + "timestamps": { + "start": "00:10:43", + "end": "00:10:44" + } + } + ], + "question_timestamps": "00:11:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is behind you.", + "D": "The window is to your right relative to where you are standing now.", + "E": "The window is above you, not behind you, from your current viewpoint." + }, + "answer_label": "C", + "answer_text": "The window is behind you.", + "q_id": "egolife_A1_D4_031_3_T1", + "correct_answer": "The window is behind you.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is behind you.", + "D": "The window is to your right relative to where you are standing now.", + "E": "The window is above you, not behind you, from your current viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_4_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Relative to me, where is the boy playing the guitar?", + "answer": "To your left.", + "question_chinese": "相对于我的位置，弹吉他的男孩在哪？", + "answer_chinese": "在你左边", + "evidence": [ + { + "description": "I entered the house through the same door.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:08" + } + } + ], + "question_timestamps": "00:11:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your right.", + "B": "Behind you.", + "C": "In front of you, facing the same direction you are as you look around the room.", + "D": "To your left.", + "E": "Above you, positioned higher up rather than beside you on the left at eye level." + }, + "answer_label": "D", + "answer_text": "To your left.", + "q_id": "egolife_A1_D4_031_4_T2", + "correct_answer": "To your left.", + "original_options": { + "A": "To your right.", + "B": "Behind you.", + "C": "In front of you, facing the same direction you are as you look around the room.", + "D": "To your left.", + "E": "Above you, positioned higher up rather than beside you on the left at eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D4_031", + "video_length": "00:11:43", + "qid": "egolife_A1_D4_031_4_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Relative to me, where is the boy playing the guitar?", + "answer": "In front of you.", + "question_chinese": "相对于我的位置，弹吉他的男孩在哪？", + "answer_chinese": "在你前面", + "evidence": [ + { + "description": "The boy playing the guitar is in front of me.", + "timestamps": { + "start": "00:11:14", + "end": "00:11:21" + } + } + ], + "question_timestamps": "00:11:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "To your right, off to the side rather than directly ahead of you.", + "E": "Above you, positioned higher up instead of being in front at eye level." + }, + "answer_label": "A", + "answer_text": "In front of you.", + "q_id": "egolife_A1_D4_031_4_T1", + "correct_answer": "In front of you.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "To your right, off to the side rather than directly ahead of you.", + "E": "Above you, positioned higher up instead of being in front at eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D4_032-mcq.json b/QAs/egolife_A1_D4_032-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..151cbd20f6f3875d8028c6479d85cd30b23b21b2 --- /dev/null +++ b/QAs/egolife_A1_D4_032-mcq.json @@ -0,0 +1,238 @@ +[ + { + "video_uid": "egolife_A1_D4_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_032_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the guitar?", + "answer": "In the hands of the person on the sofa.", + "question_chinese": "吉他的运动轨迹是什么", + "answer_chinese": "在沙发上的人手里", + "evidence": [ + { + "description": "The person on the sofa was playing the guitar.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:03:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In the doorway person's hands.", + "B": "In the hands of the person behind you.", + "C": "In the hands of the person on the sofa.", + "D": "In the hands of the person standing behind the sofa, not the one sitting on it.", + "E": "In the hands of the person sitting on the floor beside the sofa, rather than on the sofa." + }, + "answer_label": "C", + "answer_text": "In the hands of the person on the sofa.", + "q_id": "egolife_A1_D4_032_7", + "correct_answer": "In the hands of the person on the sofa.", + "original_options": { + "A": "In the doorway person's hands.", + "B": "In the hands of the person behind you.", + "C": "In the hands of the person on the sofa.", + "D": "In the hands of the person standing behind the sofa, not the one sitting on it.", + "E": "In the hands of the person sitting on the floor beside the sofa, rather than on the sofa." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_032_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the guitar?", + "answer": "The guitar was picked up and carried around.", + "question_chinese": "吉他的运动轨迹是什么", + "answer_chinese": "吉他被拿起来走了一圈", + "evidence": [ + { + "description": "The boy picked up the guitar and walked around with it.", + "timestamps": { + "start": "00:05:28", + "end": "00:05:53" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The guitar was set down on the sofa.", + "B": "The guitar was left leaning on the wall.", + "C": "The guitar was picked up briefly, then put back down in the same spot without being carried around.", + "D": "The guitar was picked up and carried around.", + "E": "The guitar was lifted and moved a little, but it was not carried around from place to place." + }, + "answer_label": "D", + "answer_text": "The guitar was picked up and carried around.", + "q_id": "egolife_A1_D4_032_8", + "correct_answer": "The guitar was picked up and carried around.", + "original_options": { + "A": "The guitar was set down on the sofa.", + "B": "The guitar was left leaning on the wall.", + "C": "The guitar was picked up briefly, then put back down in the same spot without being carried around.", + "D": "The guitar was picked up and carried around.", + "E": "The guitar was lifted and moved a little, but it was not carried around from place to place." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D4_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_032_3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many people are on the sofa?", + "answer": "Two.", + "question_chinese": "沙发上有多少人", + "answer_chinese": "两个人", + "evidence": [ + { + "description": "There were two girls sitting on the sofa.", + "timestamps": { + "start": "00:01:50", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:02:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four people are sitting on the sofa at once during the scene.", + "E": "Five people are sitting on the sofa together at the same time." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "egolife_A1_D4_032_3", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four people are sitting on the sofa at once during the scene.", + "E": "Five people are sitting on the sofa together at the same time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D4_032", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_032_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many people are on the sofa?", + "answer": "One.", + "question_chinese": "沙发上有多少人", + "answer_chinese": "一个", + "evidence": [ + { + "description": "There was a girl sitting on the sofa.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:10" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four people are sitting on the sofa at once during the scene.", + "E": "Five people are sitting on the sofa together at the same time." + }, + "answer_label": "B", + "answer_text": "One.", + "q_id": "egolife_A1_D4_032_4", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four people are sitting on the sofa at once during the scene.", + "E": "Five people are sitting on the sofa together at the same time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + } +] diff --git a/QAs/egolife_A1_D4_036-mcq.json b/QAs/egolife_A1_D4_036-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..cf831c2b85f3d07763f190a3d34f62bc918aaf2f --- /dev/null +++ b/QAs/egolife_A1_D4_036-mcq.json @@ -0,0 +1,614 @@ +[ + { + "video_uid": "egolife_A1_D4_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_036_4_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was washing my hands, where was the girl relative to me?", + "answer": "To your right.", + "question_chinese": "我洗手时女生在我什么方位？", + "answer_chinese": "在我右边", + "evidence": [ + { + "description": "The girl made a phone call next to me.", + "timestamps": { + "start": "00:01:38", + "end": "00:01:50" + } + }, + { + "description": "She told me to make a phone call while she was on my right.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:03:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "To your right.", + "C": "Behind you.", + "D": "Directly across the sink from you while you were washing your hands at the sink.", + "E": "To your left side near the counter while you were washing your hands." + }, + "answer_label": "B", + "answer_text": "To your right.", + "q_id": "egolife_A1_D4_036_4_T1", + "correct_answer": "To your right.", + "original_options": { + "A": "To your left.", + "B": "To your right.", + "C": "Behind you.", + "D": "Directly across the sink from you while you were washing your hands at the sink.", + "E": "To your left side near the counter while you were washing your hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_036_4_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I was washing my hands, where was the girl relative to me?", + "answer": "Right.", + "question_chinese": "我洗手时女生在我什么方位？", + "answer_chinese": "在我右边", + "evidence": [ + { + "description": "The girl made a phone call next to me.", + "timestamps": { + "start": "00:01:38", + "end": "00:01:50" + } + }, + { + "description": "She told me to make a phone call while she was on my right.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:10" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right.", + "B": "Left.", + "C": "Behind you.", + "D": "In front of you, near the sink area while you were washing your hands.", + "E": "To your left side as you stood washing your hands at the sink." + }, + "answer_label": "A", + "answer_text": "Right.", + "q_id": "egolife_A1_D4_036_4_T2", + "correct_answer": "Right.", + "original_options": { + "A": "Right.", + "B": "Left.", + "C": "Behind you.", + "D": "In front of you, near the sink area while you were washing your hands.", + "E": "To your left side as you stood washing your hands at the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_036_2_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take when I went to wash my hands?", + "answer": "You went to the bathroom first, then backed to the kitchen.", + "question_chinese": "我洗手经过的路径是什么样的？", + "answer_chinese": "我先在厨房，再去卫生间，再去厨房", + "evidence": [ + { + "description": "I washed my hands in the bathroom first.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + }, + { + "description": "I walked out of the bathroom after dispensing the hand soap.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:13" + } + }, + { + "description": "I went to the kitchen to wash my hands.", + "timestamps": { + "start": "00:00:30", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:03:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went to the bathroom first, then backed to the kitchen.", + "B": "You went to the kitchen first.", + "C": "You went to the hallway first.", + "D": "You walked straight to the kitchen area first, then went into the bathroom to wash your hands.", + "E": "You headed into the bathroom after passing through the living room, then returned forward into the kitchen." + }, + "answer_label": "A", + "answer_text": "You went to the bathroom first, then backed to the kitchen.", + "q_id": "egolife_A1_D4_036_2_T1", + "correct_answer": "You went to the bathroom first, then backed to the kitchen.", + "original_options": { + "A": "You went to the bathroom first, then backed to the kitchen.", + "B": "You went to the kitchen first.", + "C": "You went to the hallway first.", + "D": "You walked straight to the kitchen area first, then went into the bathroom to wash your hands.", + "E": "You headed into the bathroom after passing through the living room, then returned forward into the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_036_2_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path I take when I wash my hands?", + "answer": "You went to the bathroom first, then backed to the kitchen.", + "question_chinese": "我洗手经过的路径是什么样的？", + "answer_chinese": "我先在厨房，再去卫生间，再去厨房", + "evidence": [ + { + "description": "I washed my hands in the bathroom first.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + }, + { + "description": "I walked out of the bathroom after dispensing the hand soap.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:13" + } + }, + { + "description": "I went to the kitchen to wash my hands.", + "timestamps": { + "start": "00:00:30", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:03:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went to the bathroom first, then backed to the kitchen.", + "B": "You went to the living room first.", + "C": "You went to the hallway first.", + "D": "You left the kitchen and went straight into the bathroom, then walked forward into the dining area afterward.", + "E": "You went into the bathroom after turning toward the hallway, then came back to the kitchen without backing up." + }, + "answer_label": "A", + "answer_text": "You went to the bathroom first, then backed to the kitchen.", + "q_id": "egolife_A1_D4_036_2_T2", + "correct_answer": "You went to the bathroom first, then backed to the kitchen.", + "original_options": { + "A": "You went to the bathroom first, then backed to the kitchen.", + "B": "You went to the living room first.", + "C": "You went to the hallway first.", + "D": "You left the kitchen and went straight into the bathroom, then walked forward into the dining area afterward.", + "E": "You went into the bathroom after turning toward the hallway, then came back to the kitchen without backing up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D4_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_036_3_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take when I went to wash my hands?", + "answer": "First in the bathroom, then went to the kitchen.", + "question_chinese": "我洗手经过的路径是什么样的？", + "answer_chinese": "先在卫生间，然后去了厨房", + "evidence": [ + { + "description": "I washed my hands in the bathroom first.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + }, + { + "description": "I walked out of the bathroom after dispensing the hand soap.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:13" + } + }, + { + "description": "I went to the kitchen to wash my hands.", + "timestamps": { + "start": "00:00:30", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "First in the kitchen, then the bathroom.", + "B": "First in the bathroom, then the hallway.", + "C": "You started in the bathroom, washed up there, and then walked into the living room afterward.", + "D": "You went to the bathroom first, then stopped by the entryway before heading into the kitchen area.", + "E": "First in the bathroom, then went to the kitchen." + }, + "answer_label": "E", + "answer_text": "First in the bathroom, then went to the kitchen.", + "q_id": "egolife_A1_D4_036_3_T1", + "correct_answer": "First in the bathroom, then went to the kitchen.", + "original_options": { + "A": "First in the kitchen, then the bathroom.", + "B": "First in the bathroom, then the hallway.", + "C": "You started in the bathroom, washed up there, and then walked into the living room afterward.", + "D": "You went to the bathroom first, then stopped by the entryway before heading into the kitchen area.", + "E": "First in the bathroom, then went to the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_036_3_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What does the path I take when washing my hands look like?", + "answer": "First in the bathroom, then went to the kitchen.", + "question_chinese": "我洗手经过的路径是什么样的？", + "answer_chinese": "先在卫生间，然后去了厨房", + "evidence": [ + { + "description": "I washed my hands in the bathroom first.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + }, + { + "description": "I walked out of the bathroom after dispensing the hand soap.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:13" + } + }, + { + "description": "I went to the kitchen to wash my hands.", + "timestamps": { + "start": "00:00:30", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:05:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Kitchen first, then bathroom.", + "B": "Bathroom first, then hallway.", + "C": "You begin in the bathroom and then head straight into the living room once you finish washing.", + "D": "You go to the bathroom first, then detour through the entryway before going anywhere near the kitchen.", + "E": "First in the bathroom, then went to the kitchen." + }, + "answer_label": "E", + "answer_text": "First in the bathroom, then went to the kitchen.", + "q_id": "egolife_A1_D4_036_3_T2", + "correct_answer": "First in the bathroom, then went to the kitchen.", + "original_options": { + "A": "Kitchen first, then bathroom.", + "B": "Bathroom first, then hallway.", + "C": "You begin in the bathroom and then head straight into the living room once you finish washing.", + "D": "You go to the bathroom first, then detour through the entryway before going anywhere near the kitchen.", + "E": "First in the bathroom, then went to the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_036_5_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes are in the box?", + "answer": "Two mangoes.", + "question_chinese": "盒子里有几个芒果？", + "answer_chinese": "两个芒果", + "evidence": [ + { + "description": "When I took it out of the fridge, there were two mangoes in the box.", + "timestamps": { + "start": "00:07:20", + "end": "00:07:30" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two mangoes.", + "B": "One mango.", + "C": "Three mangoes.", + "D": "There are four ripe mangoes sitting in the box right now for a quick snack.", + "E": "There are five mangoes packed together inside the box, filling most of the space." + }, + "answer_label": "A", + "answer_text": "Two mangoes.", + "q_id": "egolife_A1_D4_036_5_T1", + "correct_answer": "Two mangoes.", + "original_options": { + "A": "Two mangoes.", + "B": "One mango.", + "C": "Three mangoes.", + "D": "There are four ripe mangoes sitting in the box right now for a quick snack.", + "E": "There are five mangoes packed together inside the box, filling most of the space." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_036_5_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes are in the box?", + "answer": "Two mangoes.", + "question_chinese": "盒子里有几个芒果？", + "answer_chinese": "一个芒果", + "evidence": [ + { + "description": "I took one of the mangoes out.", + "timestamps": { + "start": "00:07:40", + "end": "00:07:48" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One mango.", + "B": "Two mangoes.", + "C": "Three mangoes.", + "D": "There are four ripe mangoes sitting in the box right now for a quick snack.", + "E": "There are five mangoes packed together inside the box, filling most of the space." + }, + "answer_label": "B", + "answer_text": "Two mangoes.", + "q_id": "egolife_A1_D4_036_5_T2", + "correct_answer": "Two mangoes.", + "original_options": { + "A": "One mango.", + "B": "Two mangoes.", + "C": "Three mangoes.", + "D": "There are four ripe mangoes sitting in the box right now for a quick snack.", + "E": "There are five mangoes packed together inside the box, filling most of the space." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D4_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_036_1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the toilet in relation to the sink?", + "answer": "To the left of the sink.", + "question_chinese": "马桶在洗手台的什么位置", + "answer_chinese": "在洗手台左边", + "evidence": [ + { + "description": "I saw the toilet to the left of the sink while I was washing my hands in the bathroom.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To the right of the sink.", + "B": "In front of the sink.", + "C": "To the left of the sink.", + "D": "The toilet is behind the sink, positioned further back along the same wall in the bathroom.", + "E": "The toilet is in front of the sink, closer to the bathroom doorway than the sink." + }, + "answer_label": "C", + "answer_text": "To the left of the sink.", + "q_id": "egolife_A1_D4_036_1", + "correct_answer": "To the left of the sink.", + "original_options": { + "A": "To the right of the sink.", + "B": "In front of the sink.", + "C": "To the left of the sink.", + "D": "The toilet is behind the sink, positioned further back along the same wall in the bathroom.", + "E": "The toilet is in front of the sink, closer to the bathroom doorway than the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D4_036", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_036_2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the toilet in relation to the sink?", + "answer": "Left.", + "question_chinese": "马桶在洗手台的什么位置", + "answer_chinese": "左边", + "evidence": [ + { + "description": "I saw the toilet to the left of the sink while I was washing my hands in the bathroom.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right.", + "B": "In front.", + "C": "The toilet is behind the sink rather than beside it on either side.", + "D": "Left.", + "E": "The toilet is across from the sink, not positioned to the left." + }, + "answer_label": "D", + "answer_text": "Left.", + "q_id": "egolife_A1_D4_036_2", + "correct_answer": "Left.", + "original_options": { + "A": "Right.", + "B": "In front.", + "C": "The toilet is behind the sink rather than beside it on either side.", + "D": "Left.", + "E": "The toilet is across from the sink, not positioned to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 12 + ] + } +] diff --git a/QAs/egolife_A1_D4_038-mcq.json b/QAs/egolife_A1_D4_038-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d20097842ac3974c691c9474f5ef4836dabba6a8 --- /dev/null +++ b/QAs/egolife_A1_D4_038-mcq.json @@ -0,0 +1,136 @@ +[ + { + "video_uid": "egolife_A1_D4_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_038_1_T1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the chopsticks' movement?", + "answer": "Take it out of the cabinet, wash it, and then take it upstairs.", + "question_chinese": "筷子的运动轨迹是什么？", + "answer_chinese": "从柜子里拿出来，然后洗完后拿到二楼", + "evidence": [ + { + "description": "I took the chopsticks out of the cabinet.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + }, + { + "description": "I washed the chopsticks under the tap.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:30" + } + }, + { + "description": "I took the chopsticks upstairs.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Take it out, dry it, downstairs.", + "B": "Take it out of the cabinet, wash it, and then take it upstairs.", + "C": "Take it out, rinse it, outside.", + "D": "You take the chopsticks from the drawer, dry them off, and then carry them downstairs.", + "E": "You pull the chopsticks from the cabinet, rinse them briefly, and then leave them on the counter." + }, + "answer_label": "B", + "answer_text": "Take it out of the cabinet, wash it, and then take it upstairs.", + "q_id": "egolife_A1_D4_038_1_T1", + "correct_answer": "Take it out of the cabinet, wash it, and then take it upstairs.", + "original_options": { + "A": "Take it out, dry it, downstairs.", + "B": "Take it out of the cabinet, wash it, and then take it upstairs.", + "C": "Take it out, rinse it, outside.", + "D": "You take the chopsticks from the drawer, dry them off, and then carry them downstairs.", + "E": "You pull the chopsticks from the cabinet, rinse them briefly, and then leave them on the counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D4_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_038_1_T2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the chopsticks' movement?", + "answer": "Take it out of the cabinet, wash it, and then take it upstairs.", + "question_chinese": "筷子的运动轨迹是什么？", + "answer_chinese": "从柜子里拿出来，然后洗完后拿到二楼", + "evidence": [ + { + "description": "I took the chopsticks out of the cabinet.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + }, + { + "description": "I washed the chopsticks under the tap.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:30" + } + }, + { + "description": "I took the chopsticks upstairs.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Remove it, clean it, living room.", + "B": "Take it out of the cabinet, wash it, and then take it upstairs.", + "C": "Take it out, polish it, basement.", + "D": "You remove the chopsticks from the counter, clean them, and then bring them into the living room.", + "E": "You bring the chopsticks out of the cupboard, polish them, and then take them down to the basement." + }, + "answer_label": "B", + "answer_text": "Take it out of the cabinet, wash it, and then take it upstairs.", + "q_id": "egolife_A1_D4_038_1_T2", + "correct_answer": "Take it out of the cabinet, wash it, and then take it upstairs.", + "original_options": { + "A": "Remove it, clean it, living room.", + "B": "Take it out of the cabinet, wash it, and then take it upstairs.", + "C": "Take it out, polish it, basement.", + "D": "You remove the chopsticks from the counter, clean them, and then bring them into the living room.", + "E": "You bring the chopsticks out of the cupboard, polish them, and then take them down to the basement." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D4_039-mcq.json b/QAs/egolife_A1_D4_039-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..780f1d471ec95410d905da3fb2959b8b2c05d9f0 --- /dev/null +++ b/QAs/egolife_A1_D4_039-mcq.json @@ -0,0 +1,134 @@ +[ + { + "video_uid": "egolife_A1_D4_039", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_039_3", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the computer screen facing?", + "answer": "Facing the door on the second floor.", + "question_chinese": "电脑屏幕朝向哪", + "answer_chinese": "朝向二楼的门", + "evidence": [ + { + "description": "The computer was in front of me.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:10" + } + }, + { + "description": "The computer screen was facing the door.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Facing the door on the first floor.", + "B": "Facing the door on the third floor.", + "C": "Facing the door to your left side rather than toward the second-floor doorway.", + "D": "Facing the door to your right, not aligned with the second-floor door ahead.", + "E": "Facing the door on the second floor." + }, + "answer_label": "E", + "answer_text": "Facing the door on the second floor.", + "q_id": "egolife_A1_D4_039_3", + "correct_answer": "Facing the door on the second floor.", + "original_options": { + "A": "Facing the door on the first floor.", + "B": "Facing the door on the third floor.", + "C": "Facing the door to your left side rather than toward the second-floor doorway.", + "D": "Facing the door to your right, not aligned with the second-floor door ahead.", + "E": "Facing the door on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D4_039", + "video_length": "00:12:00", + "qid": "egolife_A1_D4_039_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the computer screen facing?", + "answer": "The entrance to the second floor.", + "question_chinese": "电脑屏幕朝向哪", + "answer_chinese": "二楼门口", + "evidence": [ + { + "description": "The computer was in front of me.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:10" + } + }, + { + "description": "The computer screen was facing the door.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:10" + } + } + ], + "question_timestamps": "00:04:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The entrance to the first floor.", + "B": "The entrance to the second floor.", + "C": "The entrance is behind you.", + "D": "The screen faces the entrance on your right side instead of toward the second-floor entrance.", + "E": "It is oriented toward the left-side entrance, not aimed at the second-floor entrance." + }, + "answer_label": "B", + "answer_text": "The entrance to the second floor.", + "q_id": "egolife_A1_D4_039_4", + "correct_answer": "The entrance to the second floor.", + "original_options": { + "A": "The entrance to the first floor.", + "B": "The entrance to the second floor.", + "C": "The entrance is behind you.", + "D": "The screen faces the entrance on your right side instead of toward the second-floor entrance.", + "E": "It is oriented toward the left-side entrance, not aimed at the second-floor entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D5_001-mcq.json b/QAs/egolife_A1_D5_001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d433477d0b60554f0305b06e70c05901d4bf3d1e --- /dev/null +++ b/QAs/egolife_A1_D5_001-mcq.json @@ -0,0 +1,374 @@ +[ + { + "video_uid": "egolife_A1_D5_001", + "video_length": "00:11:59", + "qid": "egolife_A1_D5_001_5", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the condition of the curtains in the bedroom?", + "answer": "The curtains are drawn.", + "question_chinese": "卧室里窗帘的状态是什么样的", + "answer_chinese": "窗帘是拉上的", + "evidence": [ + { + "description": "I walked into the bedroom and saw that the curtains were drawn.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:30" + } + }, + { + "description": "Then I got onto the bed and pulled the curtains open.", + "timestamps": { + "start": "00:02:38", + "end": "00:02:52" + } + } + ], + "question_timestamps": "00:02:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The curtains are drawn.", + "B": "The curtains are wide open.", + "C": "The curtains are slightly open.", + "D": "The curtains are only partially drawn, leaving a noticeable gap for light to come through.", + "E": "They are drawn back to the sides, leaving most of the window uncovered." + }, + "answer_label": "A", + "answer_text": "The curtains are drawn.", + "q_id": "egolife_A1_D5_001_5", + "correct_answer": "The curtains are drawn.", + "original_options": { + "A": "The curtains are drawn.", + "B": "The curtains are wide open.", + "C": "The curtains are slightly open.", + "D": "The curtains are only partially drawn, leaving a noticeable gap for light to come through.", + "E": "They are drawn back to the sides, leaving most of the window uncovered." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_001", + "video_length": "00:11:59", + "qid": "egolife_A1_D5_001_6", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the condition of the curtains in the bedroom?", + "answer": "The curtains are open.", + "question_chinese": "卧室里窗帘的状态是什么样的", + "answer_chinese": "窗帘是拉开的", + "evidence": [ + { + "description": "I saw the curtains drawn to the side of the window.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:40" + } + }, + { + "description": "I saw the sunlight streaming onto the bed from outside the window.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:40" + } + } + ], + "question_timestamps": "00:08:52", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The curtains are completely drawn.", + "B": "The curtains are completely closed.", + "C": "The curtains are partially drawn, leaving part of the window covered and part exposed.", + "D": "The curtains are open.", + "E": "They are loosely gathered to one side, not spread open across the window." + }, + "answer_label": "D", + "answer_text": "The curtains are open.", + "q_id": "egolife_A1_D5_001_6", + "correct_answer": "The curtains are open.", + "original_options": { + "A": "The curtains are completely drawn.", + "B": "The curtains are completely closed.", + "C": "The curtains are partially drawn, leaving part of the window covered and part exposed.", + "D": "The curtains are open.", + "E": "They are loosely gathered to one side, not spread open across the window." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_001", + "video_length": "00:11:59", + "qid": "egolife_A1_D5_001_4_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where am I, and where am I going?", + "answer": "You'm standing in the middle of the living room on the second floor, and you'm going downstairs.", + "question_chinese": "我在什么位置，将要去哪里？", + "answer_chinese": "我站在二楼客厅中间，将要下楼", + "evidence": [ + { + "description": "I came up from the first floor, am standing in the middle of the living room, and am facing the window.", + "timestamps": { + "start": "00:00:41", + "end": "00:00:51" + } + }, + { + "description": "I turn right, go out the door, and walk down the stairs.", + "timestamps": { + "start": "00:00:52", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You'm standing in the middle of the living room on the second floor, and you'm going downstairs.", + "B": "In the hallway upstairs, heading bathroom.", + "C": "In the living room, going outside.", + "D": "You are standing in the doorway of the living room on the first floor, and you are going to the garage.", + "E": "You are sitting on the couch in the second-floor living room, and you are heading upstairs instead." + }, + "answer_label": "A", + "answer_text": "You'm standing in the middle of the living room on the second floor, and you'm going downstairs.", + "q_id": "egolife_A1_D5_001_4_T1", + "correct_answer": "You'm standing in the middle of the living room on the second floor, and you'm going downstairs.", + "original_options": { + "A": "You'm standing in the middle of the living room on the second floor, and you'm going downstairs.", + "B": "In the hallway upstairs, heading bathroom.", + "C": "In the living room, going outside.", + "D": "You are standing in the doorway of the living room on the first floor, and you are going to the garage.", + "E": "You are sitting on the couch in the second-floor living room, and you are heading upstairs instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_001", + "video_length": "00:11:59", + "qid": "egolife_A1_D5_001_4_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where am I, and where am I going?", + "answer": "You'm standing in the middle of the living room on the second floor, and you'm going downstairs.", + "question_chinese": "我在什么位置，将要去哪里？", + "answer_chinese": "我站在二楼客厅中间，将要下楼", + "evidence": [ + { + "description": "I came up from the first floor, am standing in the middle of the living room, and am facing the window.", + "timestamps": { + "start": "00:00:41", + "end": "00:00:51" + } + }, + { + "description": "I turn right, go out the door, and walk down the stairs.", + "timestamps": { + "start": "00:00:52", + "end": "00:01:04" + } + } + ], + "question_timestamps": "00:01:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You'm standing in the middle of the living room on the second floor, and you'm going downstairs.", + "B": "In the corner downstairs, going outside.", + "C": "In the hallway upstairs, heading bathroom.", + "D": "You are standing in the middle of the hallway on the second floor, and you are going to the bathroom next.", + "E": "You are standing in the doorway of the living room on the first floor, and you are going to the garage." + }, + "answer_label": "A", + "answer_text": "You'm standing in the middle of the living room on the second floor, and you'm going downstairs.", + "q_id": "egolife_A1_D5_001_4_T2", + "correct_answer": "You'm standing in the middle of the living room on the second floor, and you'm going downstairs.", + "original_options": { + "A": "You'm standing in the middle of the living room on the second floor, and you'm going downstairs.", + "B": "In the corner downstairs, going outside.", + "C": "In the hallway upstairs, heading bathroom.", + "D": "You are standing in the middle of the hallway on the second floor, and you are going to the bathroom next.", + "E": "You are standing in the doorway of the living room on the first floor, and you are going to the garage." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D5_001", + "video_length": "00:11:59", + "qid": "egolife_A1_D5_001_5_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where am I and where am I going?", + "answer": "You'm standing in the middle of the living room on the first floor, about to head to the kitchen.", + "question_chinese": "我在什么位置，将要去哪里？", + "answer_chinese": "我站在一楼客厅中间，正要走向厨房", + "evidence": [ + { + "description": "After going downstairs, I stayed on the first floor the whole time.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:20" + } + }, + { + "description": "After handing over the items, I walk straight towards the window until I reach the kitchen.", + "timestamps": { + "start": "00:11:22", + "end": "00:11:35" + } + } + ], + "question_timestamps": "00:11:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the living room, heading to dining room.", + "B": "In the living room, heading to hallway.", + "C": "You'm standing in the middle of the living room on the first floor, about to head to the kitchen.", + "D": "You're in the middle of the living room on the first floor, about to go to the pantry.", + "E": "You're standing mid-living room on the first floor, and you're about to head down to the basement." + }, + "answer_label": "C", + "answer_text": "You'm standing in the middle of the living room on the first floor, about to head to the kitchen.", + "q_id": "egolife_A1_D5_001_5_T1", + "correct_answer": "You'm standing in the middle of the living room on the first floor, about to head to the kitchen.", + "original_options": { + "A": "In the living room, heading to dining room.", + "B": "In the living room, heading to hallway.", + "C": "You'm standing in the middle of the living room on the first floor, about to head to the kitchen.", + "D": "You're in the middle of the living room on the first floor, about to go to the pantry.", + "E": "You're standing mid-living room on the first floor, and you're about to head down to the basement." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_001", + "video_length": "00:11:59", + "qid": "egolife_A1_D5_001_5_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where am I and where am I going?", + "answer": "You'm standing in the middle of the living room on the first floor, about to head to the kitchen.", + "question_chinese": "我在什么位置，将要去哪里？", + "answer_chinese": "我站在一楼客厅中间，正要走向厨房", + "evidence": [ + { + "description": "After going downstairs, I stayed on the first floor the whole time.", + "timestamps": { + "start": "00:10:00", + "end": "00:10:20" + } + }, + { + "description": "After handing over the items, I walk straight towards the window until I reach the kitchen.", + "timestamps": { + "start": "00:11:22", + "end": "00:11:35" + } + } + ], + "question_timestamps": "00:11:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the living room, heading to dining room.", + "B": "In the living room, heading to hallway.", + "C": "You'm standing in the middle of the living room on the first floor, about to head to the kitchen.", + "D": "You're in the middle of the living room on the first floor, about to go to the pantry.", + "E": "You're standing mid-living room on the first floor, and you're about to head down to the basement." + }, + "answer_label": "C", + "answer_text": "You'm standing in the middle of the living room on the first floor, about to head to the kitchen.", + "q_id": "egolife_A1_D5_001_5_T2", + "correct_answer": "You'm standing in the middle of the living room on the first floor, about to head to the kitchen.", + "original_options": { + "A": "In the living room, heading to dining room.", + "B": "In the living room, heading to hallway.", + "C": "You'm standing in the middle of the living room on the first floor, about to head to the kitchen.", + "D": "You're in the middle of the living room on the first floor, about to go to the pantry.", + "E": "You're standing mid-living room on the first floor, and you're about to head down to the basement." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D5_002-mcq.json b/QAs/egolife_A1_D5_002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..281dddb1acdb4902978679ec94d58780e761852e --- /dev/null +++ b/QAs/egolife_A1_D5_002-mcq.json @@ -0,0 +1,120 @@ +[ + { + "video_uid": "egolife_A1_D5_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_002_8", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I go up to the second floor, what direction is the mirror in relative to me?", + "answer": "Front-right.", + "question_chinese": "当我上二楼时，镜子在我什么方位", + "answer_chinese": "右前方", + "evidence": [ + { + "description": "When I went up to the second floor, I saw the mirror in the front-right.", + "timestamps": { + "start": "00:08:52", + "end": "00:08:54" + } + } + ], + "question_timestamps": "00:09:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Front-left.", + "B": "Back-right.", + "C": "When you reach the second floor, the mirror is directly in front of you, not off to the side.", + "D": "Front-right.", + "E": "As you go up to the second floor, the mirror ends up behind you on the right-hand side." + }, + "answer_label": "D", + "answer_text": "Front-right.", + "q_id": "egolife_A1_D5_002_8", + "correct_answer": "Front-right.", + "original_options": { + "A": "Front-left.", + "B": "Back-right.", + "C": "When you reach the second floor, the mirror is directly in front of you, not off to the side.", + "D": "Front-right.", + "E": "As you go up to the second floor, the mirror ends up behind you on the right-hand side." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_002_9", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I go to the second floor, what direction is the mirror in relative to me?", + "answer": "Front-left.", + "question_chinese": "当我上二楼时，镜子在我什么方位", + "answer_chinese": "左前方", + "evidence": [ + { + "description": "When I went up to the second floor, I saw the mirror in front-left of me.", + "timestamps": { + "start": "00:11:35", + "end": "00:11:36" + } + } + ], + "question_timestamps": "00:11:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Front-right.", + "B": "Front-left.", + "C": "Back-left.", + "D": "When you get to the second floor, the mirror is straight ahead of you rather than to the left.", + "E": "After arriving on the second floor, the mirror is behind you on the right, not in front-left." + }, + "answer_label": "B", + "answer_text": "Front-left.", + "q_id": "egolife_A1_D5_002_9", + "correct_answer": "Front-left.", + "original_options": { + "A": "Front-right.", + "B": "Front-left.", + "C": "Back-left.", + "D": "When you get to the second floor, the mirror is straight ahead of you rather than to the left.", + "E": "After arriving on the second floor, the mirror is behind you on the right, not in front-left." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 19, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D5_003-mcq.json b/QAs/egolife_A1_D5_003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a86763d3de9f5b19f925c9e3aa4be42b47a419a1 --- /dev/null +++ b/QAs/egolife_A1_D5_003-mcq.json @@ -0,0 +1,379 @@ +[ + { + "video_uid": "egolife_A1_D5_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_003_1_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Which direction is the trash can relative to me?", + "answer": "The trash can is in front of you.", + "question_chinese": "垃圾桶在我什么方向？", + "answer_chinese": "垃圾桶在你前面。", + "evidence": [ + { + "description": "The trash can in front of you.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:05" + } + } + ], + "question_timestamps": "00:07:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can is behind you.", + "B": "The trash can is to your left.", + "C": "The trash can is off to your right side, not directly ahead of you.", + "D": "The trash can is positioned above you in the scene rather than in front of you.", + "E": "The trash can is in front of you." + }, + "answer_label": "E", + "answer_text": "The trash can is in front of you.", + "q_id": "egolife_A1_D5_003_1_T2", + "correct_answer": "The trash can is in front of you.", + "original_options": { + "A": "The trash can is behind you.", + "B": "The trash can is to your left.", + "C": "The trash can is off to your right side, not directly ahead of you.", + "D": "The trash can is positioned above you in the scene rather than in front of you.", + "E": "The trash can is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_003_1_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Which direction is the trash can relative to me?", + "answer": "The trash can is behind you.", + "question_chinese": "垃圾桶在我什么方向？", + "answer_chinese": "垃圾桶在你后面。", + "evidence": [ + { + "description": "The trash is behind of you.", + "timestamps": { + "start": "00:07:07", + "end": "00:07:12" + } + } + ], + "question_timestamps": "00:07:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can is behind you.", + "B": "The trash can is in front of you.", + "C": "The trash can is to your right.", + "D": "The trash can is located to your left side instead of being behind you.", + "E": "The trash can is above you in the view, not behind you where you'd expect it." + }, + "answer_label": "A", + "answer_text": "The trash can is behind you.", + "q_id": "egolife_A1_D5_003_1_T1", + "correct_answer": "The trash can is behind you.", + "original_options": { + "A": "The trash can is behind you.", + "B": "The trash can is in front of you.", + "C": "The trash can is to your right.", + "D": "The trash can is located to your left side instead of being behind you.", + "E": "The trash can is above you in the view, not behind you where you'd expect it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_003_3_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was going downstairs, where was the girl sitting on the sofa relative to me?", + "answer": "She is in front of you.", + "question_chinese": "我下楼时，坐在沙发上的女生在我什么位置？", + "answer_chinese": "她在你前面。", + "evidence": [ + { + "description": "She is in front of you.", + "timestamps": { + "start": "00:05:40", + "end": "00:05:47" + } + } + ], + "question_timestamps": "00:05:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "She is behind you.", + "B": "She is to your left.", + "C": "While you were going downstairs, she was to your right rather than directly in front of you.", + "D": "She is in front of you.", + "E": "As you went downstairs, she was positioned above you in the space, not in front of you." + }, + "answer_label": "D", + "answer_text": "She is in front of you.", + "q_id": "egolife_A1_D5_003_3_T2", + "correct_answer": "She is in front of you.", + "original_options": { + "A": "She is behind you.", + "B": "She is to your left.", + "C": "While you were going downstairs, she was to your right rather than directly in front of you.", + "D": "She is in front of you.", + "E": "As you went downstairs, she was positioned above you in the space, not in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_003_3_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I was going downstairs, where was the girl sitting on the sofa relative to me?", + "answer": "She is to your back-right.", + "question_chinese": "我下楼时，坐在沙发上的女生在我什么位置？", + "answer_chinese": "她在你右后方。", + "evidence": [ + { + "description": "She is on your right rear.", + "timestamps": { + "start": "00:05:48", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:05:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "She is to your back-left.", + "B": "She is to your front-right.", + "C": "She is to your back-right.", + "D": "While you were going downstairs, she was directly in front of you, not back-right behind you.", + "E": "As you moved downstairs, she was behind you on the left side rather than on your back-right." + }, + "answer_label": "C", + "answer_text": "She is to your back-right.", + "q_id": "egolife_A1_D5_003_3_T1", + "correct_answer": "She is to your back-right.", + "original_options": { + "A": "She is to your back-left.", + "B": "She is to your front-right.", + "C": "She is to your back-right.", + "D": "While you were going downstairs, she was directly in front of you, not back-right behind you.", + "E": "As you moved downstairs, she was behind you on the left side rather than on your back-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_003_10", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the girl sitting on the sofa in relation to me when I went downstairs?", + "answer": "Behind and to my right.", + "question_chinese": "我下楼时，坐在沙发上的女生在我什么位置", + "answer_chinese": "右后方", + "evidence": [ + { + "description": "Right before I went downstairs, she had just stood up from the sofa.", + "timestamps": { + "start": "00:05:40", + "end": "00:05:46" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind and to your left.", + "B": "In front and to your right.", + "C": "She was behind you but positioned on your left side as you went downstairs.", + "D": "She was in front of you on the sofa, slightly offset to your right when you headed down.", + "E": "Behind and to my right." + }, + "answer_label": "E", + "answer_text": "Behind and to my right.", + "q_id": "egolife_A1_D5_003_10", + "correct_answer": "Behind and to my right.", + "original_options": { + "A": "Behind and to your left.", + "B": "In front and to your right.", + "C": "She was behind you but positioned on your left side as you went downstairs.", + "D": "She was in front of you on the sofa, slightly offset to your right when you headed down.", + "E": "Behind and to my right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_003_2_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where are the grapes relative to me?", + "answer": "Right now, the grapes are in front of you.", + "question_chinese": "此时，葡萄在我什么方位？", + "answer_chinese": "此时，葡萄在你前面", + "evidence": [ + { + "description": "The graps are in front of you.", + "timestamps": { + "start": "00:09:38", + "end": "00:09:42" + } + } + ], + "question_timestamps": "00:09:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right now, the grapes are behind you.", + "B": "Right now, the grapes are to your left.", + "C": "Right now, the grapes are to your right, not centered in front of you.", + "D": "Right now, the grapes are above you, rather than sitting directly in your forward view.", + "E": "Right now, the grapes are in front of you." + }, + "answer_label": "E", + "answer_text": "Right now, the grapes are in front of you.", + "q_id": "egolife_A1_D5_003_2_T2", + "correct_answer": "Right now, the grapes are in front of you.", + "original_options": { + "A": "Right now, the grapes are behind you.", + "B": "Right now, the grapes are to your left.", + "C": "Right now, the grapes are to your right, not centered in front of you.", + "D": "Right now, the grapes are above you, rather than sitting directly in your forward view.", + "E": "Right now, the grapes are in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_003_2_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Right now, where are the grapes relative to me?", + "answer": "Right now, the grapes are to your right.", + "question_chinese": "此时，葡萄在我什么方位？", + "answer_chinese": "此时，葡萄在你右边。", + "evidence": [ + { + "description": "The graps are on your right.", + "timestamps": { + "start": "00:09:45", + "end": "00:09:48" + } + } + ], + "question_timestamps": "00:09:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Right now, the grapes are to your right.", + "B": "Right now, the grapes are ahead.", + "C": "Right now, the grapes are behind you.", + "D": "Right now, the grapes are to your left, not on your right side.", + "E": "Right now, the grapes are above you instead of being positioned to your right." + }, + "answer_label": "A", + "answer_text": "Right now, the grapes are to your right.", + "q_id": "egolife_A1_D5_003_2_T1", + "correct_answer": "Right now, the grapes are to your right.", + "original_options": { + "A": "Right now, the grapes are to your right.", + "B": "Right now, the grapes are ahead.", + "C": "Right now, the grapes are behind you.", + "D": "Right now, the grapes are to your left, not on your right side.", + "E": "Right now, the grapes are above you instead of being positioned to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D5_004-mcq.json b/QAs/egolife_A1_D5_004-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9bc39d65095e2adaabdd547ae89dde0ffdd7f1e9 --- /dev/null +++ b/QAs/egolife_A1_D5_004-mcq.json @@ -0,0 +1,565 @@ +[ + { + "video_uid": "egolife_A1_D5_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_004_1_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the seat cushion relative to me?", + "answer": "On the ground to your front-right.", + "question_chinese": "坐垫在我什么方位？", + "answer_chinese": "在我右前方的地上", + "evidence": [ + { + "description": "I saw the cushion in front of the sofa.", + "timestamps": { + "start": "00:00:52", + "end": "00:00:53" + } + }, + { + "description": "I sat down and saw the cushion right in front of me.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:57" + } + } + ], + "question_timestamps": "00:01:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the ground to your front-left.", + "B": "On the ground to your front-right.", + "C": "On the ground directly in front of you.", + "D": "On the ground to your back-right, rather than in your front-right area.", + "E": "On the ground to your back-left, not placed diagonally in front of you on the right." + }, + "answer_label": "B", + "answer_text": "On the ground to your front-right.", + "q_id": "egolife_A1_D5_004_1_T1", + "correct_answer": "On the ground to your front-right.", + "original_options": { + "A": "On the ground to your front-left.", + "B": "On the ground to your front-right.", + "C": "On the ground directly in front of you.", + "D": "On the ground to your back-right, rather than in your front-right area.", + "E": "On the ground to your back-left, not placed diagonally in front of you on the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 12, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_004_1_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the seat cushion relative to me?", + "answer": "On the ground to your back-left.", + "question_chinese": "坐垫在我什么方位？", + "answer_chinese": "在我左后方的地上。", + "evidence": [ + { + "description": "I see a cushion.", + "timestamps": { + "start": "00:07:38", + "end": "00:07:44" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the ground to your back-right.", + "B": "On the ground to your front-left.", + "C": "On the ground to your back-left.", + "D": "On the ground to your front-right, rather than behind you on the left side.", + "E": "On the ground directly behind you, not diagonally behind you toward the left." + }, + "answer_label": "C", + "answer_text": "On the ground to your back-left.", + "q_id": "egolife_A1_D5_004_1_T2", + "correct_answer": "On the ground to your back-left.", + "original_options": { + "A": "On the ground to your back-right.", + "B": "On the ground to your front-left.", + "C": "On the ground to your back-left.", + "D": "On the ground to your front-right, rather than behind you on the left side.", + "E": "On the ground directly behind you, not diagonally behind you toward the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_004_5_T1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I take the photography light to?", + "answer": "You moved the fill light from the corner of the first-floor living room: you turned left and went straight, passed the stairs, and finally brought it into the downstairs room.", + "question_chinese": "我将打光灯拿去了哪？", + "answer_chinese": "你将打光灯从一楼客厅角落，左转直走，经过楼梯，最后拿到楼下的房间内。", + "evidence": [ + { + "description": "I took the light to the living room on the first floor.", + "timestamps": { + "start": "00:10:30", + "end": "00:11:23" + } + } + ], + "question_timestamps": "00:11:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved the fill light from the corner of the first-floor living room: you turned left and went straight, passed the stairs, and finally brought it into the downstairs room.", + "B": "You brought it into the upstairs room.", + "C": "You returned it to the living room.", + "D": "You turned right and went straight, passed the stairs, and finally brought it into the downstairs room.", + "E": "You turned left and went straight, but brought it into the downstairs room before passing the stairs." + }, + "answer_label": "A", + "answer_text": "You moved the fill light from the corner of the first-floor living room: you turned left and went straight, passed the stairs, and finally brought it into the downstairs room.", + "q_id": "egolife_A1_D5_004_5_T1", + "correct_answer": "You moved the fill light from the corner of the first-floor living room: you turned left and went straight, passed the stairs, and finally brought it into the downstairs room.", + "original_options": { + "A": "You moved the fill light from the corner of the first-floor living room: you turned left and went straight, passed the stairs, and finally brought it into the downstairs room.", + "B": "You brought it into the upstairs room.", + "C": "You returned it to the living room.", + "D": "You turned right and went straight, passed the stairs, and finally brought it into the downstairs room.", + "E": "You turned left and went straight, but brought it into the downstairs room before passing the stairs." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_004_5_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Where did I take the photography light to?", + "answer": "You moved the fill light from the corner of the first-floor living room: you turned left and went straight, passed the stairs, and finally brought it into the downstairs room.", + "question_chinese": "我将打光灯拿去了哪？", + "answer_chinese": "你将打光灯从一楼客厅角落，左转直走，经过楼梯，最后拿到楼下的房间内。", + "evidence": [ + { + "description": "I took the light to the living room on the first floor.", + "timestamps": { + "start": "00:10:30", + "end": "00:11:23" + } + } + ], + "question_timestamps": "00:11:58", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You left it at the stairs.", + "B": "You moved the fill light from the corner of the first-floor living room: you turned left and went straight, passed the stairs, and finally brought it into the downstairs room.", + "C": "You set it down in the hallway.", + "D": "You moved the fill light from the living room corner, turned right, went straight, and took it into the downstairs room.", + "E": "You moved the fill light from the living room corner, turned left, went straight, and brought it into the upstairs room." + }, + "answer_label": "B", + "answer_text": "You moved the fill light from the corner of the first-floor living room: you turned left and went straight, passed the stairs, and finally brought it into the downstairs room.", + "q_id": "egolife_A1_D5_004_5_T2", + "correct_answer": "You moved the fill light from the corner of the first-floor living room: you turned left and went straight, passed the stairs, and finally brought it into the downstairs room.", + "original_options": { + "A": "You left it at the stairs.", + "B": "You moved the fill light from the corner of the first-floor living room: you turned left and went straight, passed the stairs, and finally brought it into the downstairs room.", + "C": "You set it down in the hallway.", + "D": "You moved the fill light from the living room corner, turned right, went straight, and took it into the downstairs room.", + "E": "You moved the fill light from the living room corner, turned left, went straight, and brought it into the upstairs room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D5_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_004_4_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the plugboard in my hand?", + "answer": "The insert board was moved from the ground in front of you onto the chair.", + "question_chinese": "我手上的插板移动轨迹是？", + "answer_chinese": "插板从你前方的地上放到了椅子上。", + "evidence": [ + { + "description": "I picked up the power strip from the floor.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:12" + } + } + ], + "question_timestamps": "00:10:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was moved onto the tabletop.", + "B": "It was moved from chair to ground.", + "C": "The insert board was picked up from the ground in front of you and placed at the downstairs end of the tabletop.", + "D": "The insert board was moved from the ground behind you and then set down onto the chair.", + "E": "The insert board was moved from the ground in front of you onto the chair." + }, + "answer_label": "E", + "answer_text": "The insert board was moved from the ground in front of you onto the chair.", + "q_id": "egolife_A1_D5_004_4_T1", + "correct_answer": "The insert board was moved from the ground in front of you onto the chair.", + "original_options": { + "A": "It was moved onto the tabletop.", + "B": "It was moved from chair to ground.", + "C": "The insert board was picked up from the ground in front of you and placed at the downstairs end of the tabletop.", + "D": "The insert board was moved from the ground behind you and then set down onto the chair.", + "E": "The insert board was moved from the ground in front of you onto the chair." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 22, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_004_4_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement trajectory of the plugboard in my hand?", + "answer": "The insert board was picked up from the ground in front of you, then taken downstairs and placed at the downstairs end of the tabletop.", + "question_chinese": "我手上的插板移动轨迹是？", + "answer_chinese": "插板从你前方的地上拿起，然后移动到楼下，放到了桌面端的楼下。", + "evidence": [ + { + "description": "I picked up the power strip from the floor.", + "timestamps": { + "start": "00:10:10", + "end": "00:10:12" + } + } + ], + "question_timestamps": "00:11:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The insert board was picked up from the ground in front of you, then taken downstairs and placed at the downstairs end of the tabletop.", + "B": "You carried it upstairs.", + "C": "You picked it up behind you.", + "D": "You picked it up in front, carried it upstairs, and placed it on the tabletop.", + "E": "You lifted it from behind you, went downstairs, and set it at the tabletop end." + }, + "answer_label": "A", + "answer_text": "The insert board was picked up from the ground in front of you, then taken downstairs and placed at the downstairs end of the tabletop.", + "q_id": "egolife_A1_D5_004_4_T2", + "correct_answer": "The insert board was picked up from the ground in front of you, then taken downstairs and placed at the downstairs end of the tabletop.", + "original_options": { + "A": "The insert board was picked up from the ground in front of you, then taken downstairs and placed at the downstairs end of the tabletop.", + "B": "You carried it upstairs.", + "C": "You picked it up behind you.", + "D": "You picked it up in front, carried it upstairs, and placed it on the tabletop.", + "E": "You lifted it from behind you, went downstairs, and set it at the tabletop end." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_004_7", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which girl is the boy in the orange shirt closer to?", + "answer": "Closer to the girl on the right.", + "question_chinese": "橙衣服男孩离哪位女生更近", + "answer_chinese": "离右边女生更近", + "evidence": [ + { + "description": "The girl on the left was on the sofa.", + "timestamps": { + "start": "00:06:36", + "end": "00:06:39" + } + }, + { + "description": "The girl on the right sat on the chair and was closer to the boy.", + "timestamps": { + "start": "00:06:37", + "end": "00:06:39" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Closer to the girl on the left.", + "B": "Closer to the girl on the right.", + "C": "Closer to the girl in the red dress.", + "D": "Closer to the girl in the green dress than to the girl on the right.", + "E": "Closer to the girl holding the book, rather than the girl standing on the right." + }, + "answer_label": "B", + "answer_text": "Closer to the girl on the right.", + "q_id": "egolife_A1_D5_004_7", + "correct_answer": "Closer to the girl on the right.", + "original_options": { + "A": "Closer to the girl on the left.", + "B": "Closer to the girl on the right.", + "C": "Closer to the girl in the red dress.", + "D": "Closer to the girl in the green dress than to the girl on the right.", + "E": "Closer to the girl holding the book, rather than the girl standing on the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_004_8", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which girl is the boy in the orange shirt closer to?", + "answer": "He is closer to the girl on the right.", + "question_chinese": "橙衣服男孩离哪位女生更近", + "answer_chinese": "右边女生", + "evidence": [ + { + "description": "The girl on the left was on the sofa.", + "timestamps": { + "start": "00:06:35", + "end": "00:06:37" + } + }, + { + "description": "The girl on the right sat on the chair and was closer to the boy.", + "timestamps": { + "start": "00:06:37", + "end": "00:06:39" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "He is closer to the girl on the right.", + "B": "Closer to the girl on the left.", + "C": "Closer to the girl in blue.", + "D": "He is closer to the girl in the green shirt, not the girl standing on the right.", + "E": "He is closer to the girl on the floor than to the girl positioned on the right." + }, + "answer_label": "A", + "answer_text": "He is closer to the girl on the right.", + "q_id": "egolife_A1_D5_004_8", + "correct_answer": "He is closer to the girl on the right.", + "original_options": { + "A": "He is closer to the girl on the right.", + "B": "Closer to the girl on the left.", + "C": "Closer to the girl in blue.", + "D": "He is closer to the girl in the green shirt, not the girl standing on the right.", + "E": "He is closer to the girl on the floor than to the girl positioned on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_004_2_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the laptop relative to me?", + "answer": "on your left-hand side.", + "question_chinese": "笔记本电脑在我什么方位？", + "answer_chinese": "在我的左手边", + "evidence": [ + { + "description": "I see a laptop computer\n.", + "timestamps": { + "start": "00:01:00", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your right-hand side.", + "B": "Directly in front of you.", + "C": "on your left-hand side.", + "D": "Behind you on the ground, out of your immediate view behind your position.", + "E": "On your front-right side, angled slightly ahead of where you are currently facing." + }, + "answer_label": "C", + "answer_text": "on your left-hand side.", + "q_id": "egolife_A1_D5_004_2_T1", + "correct_answer": "on your left-hand side.", + "original_options": { + "A": "On your right-hand side.", + "B": "Directly in front of you.", + "C": "on your left-hand side.", + "D": "Behind you on the ground, out of your immediate view behind your position.", + "E": "On your front-right side, angled slightly ahead of where you are currently facing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_004", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_004_2_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the laptop relative to me?", + "answer": "One is to your left, and the other is to your front-right.", + "question_chinese": "笔记本电脑在我什么方位？", + "answer_chinese": "一个在你的左边，另一个在你的右前方。", + "evidence": [ + { + "description": "I saw the laptop computer again.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:07" + } + } + ], + "question_timestamps": "00:04:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Both are in front of you.", + "B": "Both are on your right.", + "C": "One is to your left, and the other is to your front-right.", + "D": "One is behind you, and the other is to your front-right, not on your left side.", + "E": "One is directly in front of you, and the other sits slightly to your right." + }, + "answer_label": "C", + "answer_text": "One is to your left, and the other is to your front-right.", + "q_id": "egolife_A1_D5_004_2_T2", + "correct_answer": "One is to your left, and the other is to your front-right.", + "original_options": { + "A": "Both are in front of you.", + "B": "Both are on your right.", + "C": "One is to your left, and the other is to your front-right.", + "D": "One is behind you, and the other is to your front-right, not on your left side.", + "E": "One is directly in front of you, and the other sits slightly to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D5_005-mcq.json b/QAs/egolife_A1_D5_005-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a3714bd91dfc43fe44ef416c056a1b44fdee0124 --- /dev/null +++ b/QAs/egolife_A1_D5_005-mcq.json @@ -0,0 +1,621 @@ +[ + { + "video_uid": "egolife_A1_D5_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_005_4_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just went out to pick up the takeout?", + "answer": "You walked out of the yard, pushed the door open, and went to your left.", + "question_chinese": "我刚刚出门拿外卖的路线是什么？", + "answer_chinese": "你走出院子，推开了门，走向左边", + "evidence": [ + { + "description": "I walked out of the yard and opened the door.", + "timestamps": { + "start": "00:08:58", + "end": "00:09:04" + } + }, + { + "description": "I went straight to the left.", + "timestamps": { + "start": "00:09:05", + "end": "00:09:20" + } + } + ], + "question_timestamps": "00:09:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went straight ahead after opening.", + "B": "You walked out of the yard, pushed the door open, and went to your left.", + "C": "You turned right after the door.", + "D": "You walked out of the yard, pushed the door open, and then went to your right.", + "E": "You walked out of the yard, pushed the door open, and continued straight in front of you." + }, + "answer_label": "B", + "answer_text": "You walked out of the yard, pushed the door open, and went to your left.", + "q_id": "egolife_A1_D5_005_4_T2", + "correct_answer": "You walked out of the yard, pushed the door open, and went to your left.", + "original_options": { + "A": "You went straight ahead after opening.", + "B": "You walked out of the yard, pushed the door open, and went to your left.", + "C": "You turned right after the door.", + "D": "You walked out of the yard, pushed the door open, and then went to your right.", + "E": "You walked out of the yard, pushed the door open, and continued straight in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_005_4_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I just went out to pick up the takeout?", + "answer": "You walked out of the yard, pushed the door open, and went straight to your left. You ran into a delivery person on the way.", + "question_chinese": "我刚刚出门拿外卖的路线是什么？", + "answer_chinese": "你走出院子，推开了门，径直走向左边，在路上遇到了配送员。", + "evidence": [ + { + "description": "I walked out of the yard and opened the door.", + "timestamps": { + "start": "00:08:58", + "end": "00:09:04" + } + }, + { + "description": "I went straight to the left.", + "timestamps": { + "start": "00:09:05", + "end": "00:09:23" + } + }, + { + "description": "I met the delivery person on the way.", + "timestamps": { + "start": "00:09:24", + "end": "00:09:27" + } + } + ], + "question_timestamps": "00:11:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You went straight ahead.", + "B": "You turned right afterward.", + "C": "You walked out of the yard, pushed the door open, and went straight to your left. You ran into a delivery person on the way.", + "D": "You walked out of the yard, pushed the door open, went straight to your right, and met a delivery person.", + "E": "You walked out of the yard, pushed the door open, went straight ahead, and met a delivery person." + }, + "answer_label": "C", + "answer_text": "You walked out of the yard, pushed the door open, and went straight to your left. You ran into a delivery person on the way.", + "q_id": "egolife_A1_D5_005_4_T1", + "correct_answer": "You walked out of the yard, pushed the door open, and went straight to your left. You ran into a delivery person on the way.", + "original_options": { + "A": "You went straight ahead.", + "B": "You turned right afterward.", + "C": "You walked out of the yard, pushed the door open, and went straight to your left. You ran into a delivery person on the way.", + "D": "You walked out of the yard, pushed the door open, went straight to your right, and met a delivery person.", + "E": "You walked out of the yard, pushed the door open, went straight ahead, and met a delivery person." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_005_2_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes do I see in the box right now?", + "answer": "You saw one inside the box in front of you.", + "question_chinese": "我看见现在盒子里有几个芒果？", + "answer_chinese": "你在面前的盒子里看见了一个", + "evidence": [ + { + "description": "There are two mangoes in the box.You took one out of the box.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:16" + } + } + ], + "question_timestamps": "00:02:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero mangoes.", + "B": "Two mangoes.", + "C": "You saw three mangoes inside the box in front of you right now.", + "D": "You saw four mangoes inside the box in front of you right now.", + "E": "You saw one inside the box in front of you." + }, + "answer_label": "E", + "answer_text": "You saw one inside the box in front of you.", + "q_id": "egolife_A1_D5_005_2_T2", + "correct_answer": "You saw one inside the box in front of you.", + "original_options": { + "A": "Zero mangoes.", + "B": "Two mangoes.", + "C": "You saw three mangoes inside the box in front of you right now.", + "D": "You saw four mangoes inside the box in front of you right now.", + "E": "You saw one inside the box in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_005_2_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes do I see in the box right now?", + "answer": "You saw there were no mangoes in the box in front of you.", + "question_chinese": "我看见现在盒子里有几个芒果？", + "answer_chinese": "你看见面前的盒子里没有芒果", + "evidence": [ + { + "description": "I took out the mangoes to wash them.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:37" + } + }, + { + "description": "I cut the two mangoes in half.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:50" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one mango in the box.", + "B": "You saw two mangoes in the box.", + "C": "You saw three mangoes inside the box in front of you when you looked inside it.", + "D": "You saw there were no mangoes in the box in front of you.", + "E": "You saw four mangoes inside the box in front of you, spread out rather than stacked together." + }, + "answer_label": "D", + "answer_text": "You saw there were no mangoes in the box in front of you.", + "q_id": "egolife_A1_D5_005_2_T1", + "correct_answer": "You saw there were no mangoes in the box in front of you.", + "original_options": { + "A": "You saw one mango in the box.", + "B": "You saw two mangoes in the box.", + "C": "You saw three mangoes inside the box in front of you when you looked inside it.", + "D": "You saw there were no mangoes in the box in front of you.", + "E": "You saw four mangoes inside the box in front of you, spread out rather than stacked together." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_005_1_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes did I see in the box?", + "answer": "You saw one inside the box in front of you.", + "question_chinese": "我看见盒子里有几个芒果？", + "answer_chinese": "你在面前的盒子里看见了一个", + "evidence": [ + { + "description": "There are two mangoes in the box by the kitchen sink.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:14" + } + }, + { + "description": "You took one out of the box.", + "timestamps": { + "start": "00:02:14", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:02:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw zero mangoes in the box.", + "B": "You saw two mangoes in the box.", + "C": "You saw three mangoes inside the box in front of you when you checked the contents.", + "D": "You saw one inside the box in front of you.", + "E": "You saw four mangoes inside the box in front of you as you looked down into it." + }, + "answer_label": "D", + "answer_text": "You saw one inside the box in front of you.", + "q_id": "egolife_A1_D5_005_1_T2", + "correct_answer": "You saw one inside the box in front of you.", + "original_options": { + "A": "You saw zero mangoes in the box.", + "B": "You saw two mangoes in the box.", + "C": "You saw three mangoes inside the box in front of you when you checked the contents.", + "D": "You saw one inside the box in front of you.", + "E": "You saw four mangoes inside the box in front of you as you looked down into it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_005_1_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes did I see in the box?", + "answer": "You saw two inside the box in front of you.", + "question_chinese": "我看见盒子里有几个芒果？", + "answer_chinese": "你在面前的盒子里看见了两个", + "evidence": [ + { + "description": "There are two mangoes in the box by the kitchen sink.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:15" + } + }, + { + "description": "I took out the mangoes to wash them.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw zero mangoes in the box.", + "B": "You saw one mango in the box.", + "C": "You saw two inside the box in front of you.", + "D": "You saw three mangoes inside the box in front of you when you looked into it.", + "E": "You saw four mangoes inside the box in front of you after checking the box’s contents." + }, + "answer_label": "C", + "answer_text": "You saw two inside the box in front of you.", + "q_id": "egolife_A1_D5_005_1_T1", + "correct_answer": "You saw two inside the box in front of you.", + "original_options": { + "A": "You saw zero mangoes in the box.", + "B": "You saw one mango in the box.", + "C": "You saw two inside the box in front of you.", + "D": "You saw three mangoes inside the box in front of you when you looked into it.", + "E": "You saw four mangoes inside the box in front of you after checking the box’s contents." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_005_3_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After picking up the takeout, what route did I take?", + "answer": "You turned left and went straight, then turned right into the yard, then turned around and shut the door properly.", + "question_chinese": "拿完外卖后，我的路径是什么", + "answer_chinese": "左转直走，右转走进院子，掉头把门关好", + "evidence": [ + { + "description": "Walk to the left.", + "timestamps": { + "start": "00:05:46", + "end": "00:05:47" + } + }, + { + "description": "Turn right and walk into the yard.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:11" + } + } + ], + "question_timestamps": "00:06:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right, then went straight.", + "B": "You turned left and went straight, then turned right into the yard, then turned around and shut the door properly.", + "C": "You turned left into the yard.", + "D": "You turned right and went straight, then turned right into the yard before turning around to close the door.", + "E": "You turned left and went straight, then turned left into the yard, then turned around and shut the door properly." + }, + "answer_label": "B", + "answer_text": "You turned left and went straight, then turned right into the yard, then turned around and shut the door properly.", + "q_id": "egolife_A1_D5_005_3_T2", + "correct_answer": "You turned left and went straight, then turned right into the yard, then turned around and shut the door properly.", + "original_options": { + "A": "You turned right, then went straight.", + "B": "You turned left and went straight, then turned right into the yard, then turned around and shut the door properly.", + "C": "You turned left into the yard.", + "D": "You turned right and went straight, then turned right into the yard before turning around to close the door.", + "E": "You turned left and went straight, then turned left into the yard, then turned around and shut the door properly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D5_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_005_3_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After picking up the takeout, what route did I take?", + "answer": "You turned left and went straight, then turned right into the yard, then turned around and shut the door properly.", + "question_chinese": "拿完外卖后，我的路径是什么", + "answer_chinese": "左转直走，右转走进院子，掉头把门关好", + "evidence": [ + { + "description": "Walk to the left.", + "timestamps": { + "start": "00:05:46", + "end": "00:05:47" + } + }, + { + "description": "Turn right and walk into the yard.", + "timestamps": { + "start": "00:06:10", + "end": "00:06:11" + } + }, + { + "description": "Turn around and close the door tightly.", + "timestamps": { + "start": "00:06:15", + "end": "00:06:17" + } + } + ], + "question_timestamps": "00:06:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right, then went straight.", + "B": "You went straight into the yard.", + "C": "You turned left and went straight, then turned right into the yard, then turned around and shut the door properly.", + "D": "You walked out of the yard, pushed the door open, and then headed to your left afterward.", + "E": "You turned left and went straight, then turned left into the yard, then turned around and shut the door properly." + }, + "answer_label": "C", + "answer_text": "You turned left and went straight, then turned right into the yard, then turned around and shut the door properly.", + "q_id": "egolife_A1_D5_005_3_T1", + "correct_answer": "You turned left and went straight, then turned right into the yard, then turned around and shut the door properly.", + "original_options": { + "A": "You turned right, then went straight.", + "B": "You went straight into the yard.", + "C": "You turned left and went straight, then turned right into the yard, then turned around and shut the door properly.", + "D": "You walked out of the yard, pushed the door open, and then headed to your left afterward.", + "E": "You turned left and went straight, then turned left into the yard, then turned around and shut the door properly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D5_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_005_3", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes are there in the box?", + "answer": "Two.", + "question_chinese": "盒子里有几个芒果", + "answer_chinese": "两个", + "evidence": [ + { + "description": "There are two mangoes in the box by the kitchen sink.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:15" + } + }, + { + "description": "I took out the mangoes to wash them.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:37" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four mangoes were in the box in front of you when you looked inside it.", + "D": "Two.", + "E": "Five mangoes were in the box in front of you, which is more than two." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "egolife_A1_D5_005_3", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four mangoes were in the box in front of you when you looked inside it.", + "D": "Two.", + "E": "Five mangoes were in the box in front of you, which is more than two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_005_4", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mangoes are in the box?", + "answer": "Two mangoes.", + "question_chinese": "盒子里有几个芒果", + "answer_chinese": "零个", + "evidence": [ + { + "description": "I took out the mangoes to wash them.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:37" + } + }, + { + "description": "I cut the two mangoes in half.", + "timestamps": { + "start": "00:03:45", + "end": "00:03:50" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One mango.", + "B": "Two mangoes.", + "C": "Three mangoes.", + "D": "Four mangoes were in the box in front of you when you checked it.", + "E": "Five mangoes were in the box in front of you when you looked into it." + }, + "answer_label": "B", + "answer_text": "Two mangoes.", + "q_id": "egolife_A1_D5_005_4", + "correct_answer": "Two mangoes.", + "original_options": { + "A": "One mango.", + "B": "Two mangoes.", + "C": "Three mangoes.", + "D": "Four mangoes were in the box in front of you when you checked it.", + "E": "Five mangoes were in the box in front of you when you looked into it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D5_006-mcq.json b/QAs/egolife_A1_D5_006-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ef64a28928aa015eb92a8a7605d068647c522459 --- /dev/null +++ b/QAs/egolife_A1_D5_006-mcq.json @@ -0,0 +1,339 @@ +[ + { + "video_uid": "egolife_A1_D5_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_006_3_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks did we take out of the fridge?", + "answer": "You all took one bottle from the refrigerator in front of you.", + "question_chinese": "我们从冰箱里拿了多少瓶饮料？", + "answer_chinese": "你们从面前的冰箱里拿了一瓶", + "evidence": [ + { + "description": "I took one bottles of drinks from the fridge.", + "timestamps": { + "start": "00:11:22", + "end": "00:11:25" + } + } + ], + "question_timestamps": "00:11:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You all took zero bottles from the fridge.", + "B": "You all took one bottle from the refrigerator in front of you.", + "C": "You all took two bottles from the fridge.", + "D": "You all took three bottles from the refrigerator in front of you when you grabbed the drinks.", + "E": "You all took four bottles from the refrigerator in front of you, taking more than just a single bottle." + }, + "answer_label": "B", + "answer_text": "You all took one bottle from the refrigerator in front of you.", + "q_id": "egolife_A1_D5_006_3_T2", + "correct_answer": "You all took one bottle from the refrigerator in front of you.", + "original_options": { + "A": "You all took zero bottles from the fridge.", + "B": "You all took one bottle from the refrigerator in front of you.", + "C": "You all took two bottles from the fridge.", + "D": "You all took three bottles from the refrigerator in front of you when you grabbed the drinks.", + "E": "You all took four bottles from the refrigerator in front of you, taking more than just a single bottle." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_006_3_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks did we take out of the fridge?", + "answer": "You all took six bottles from the refrigerator in front of you.", + "question_chinese": "我们从冰箱里拿了多少瓶饮料？", + "answer_chinese": "你们从面前的冰箱里拿了六瓶", + "evidence": [ + { + "description": "I took one bottles of drinks from the fridge.", + "timestamps": { + "start": "00:11:22", + "end": "00:11:25" + } + }, + { + "description": "The girl took five bottles of drinks from the fridge.", + "timestamps": { + "start": "00:11:28", + "end": "00:11:33" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You all took six bottles from the refrigerator in front of you.", + "B": "Four bottles.", + "C": "Eight bottles.", + "D": "You all took five bottles from the refrigerator in front of you, not six.", + "E": "You all took seven bottles out of the fridge in front of you before closing it." + }, + "answer_label": "A", + "answer_text": "You all took six bottles from the refrigerator in front of you.", + "q_id": "egolife_A1_D5_006_3_T1", + "correct_answer": "You all took six bottles from the refrigerator in front of you.", + "original_options": { + "A": "You all took six bottles from the refrigerator in front of you.", + "B": "Four bottles.", + "C": "Eight bottles.", + "D": "You all took five bottles from the refrigerator in front of you, not six.", + "E": "You all took seven bottles out of the fridge in front of you before closing it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_006_2_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in my view?", + "answer": "There were four people in front of you.", + "question_chinese": "我的视野里有多少人？", + "answer_chinese": "你面前有四个人", + "evidence": [ + { + "description": "There are three girls and two boys by the table.There is a girl on the sofa.", + "timestamps": { + "start": "00:09:40", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two people.", + "B": "Six people.", + "C": "There were three people standing in front of you in your view.", + "D": "There were five people in front of you, filling most of your line of sight.", + "E": "There were four people in front of you." + }, + "answer_label": "E", + "answer_text": "There were four people in front of you.", + "q_id": "egolife_A1_D5_006_2_T1", + "correct_answer": "There were four people in front of you.", + "original_options": { + "A": "Two people.", + "B": "Six people.", + "C": "There were three people standing in front of you in your view.", + "D": "There were five people in front of you, filling most of your line of sight.", + "E": "There were four people in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_006_2_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in my view?", + "answer": "There were five people in front of you.", + "question_chinese": "我的视野里有多少人？", + "answer_chinese": "你面前有五个人", + "evidence": [ + { + "description": "There are three girls and two boys by the table.There is a girl on the sofa.", + "timestamps": { + "start": "00:09:40", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:10:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There were five people in front of you.", + "B": "Three people.", + "C": "Seven people.", + "D": "There were four people in front of you in your view, not five.", + "E": "There were six people gathered in front of you within your view." + }, + "answer_label": "A", + "answer_text": "There were five people in front of you.", + "q_id": "egolife_A1_D5_006_2_T2", + "correct_answer": "There were five people in front of you.", + "original_options": { + "A": "There were five people in front of you.", + "B": "Three people.", + "C": "Seven people.", + "D": "There were four people in front of you in your view, not five.", + "E": "There were six people gathered in front of you within your view." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D5_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_006_7", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there by the table?", + "answer": "Three.", + "question_chinese": "桌子边有多少把椅子", + "answer_chinese": "三把", + "evidence": [ + { + "description": "I went upstairs and saw three chairs by the table.", + "timestamps": { + "start": "00:05:30", + "end": "00:05:35" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "There were five chairs placed by the table, not three chairs total.", + "D": "There were seven chairs by the table, lined up around its edges.", + "E": "Three." + }, + "answer_label": "E", + "answer_text": "Three.", + "q_id": "egolife_A1_D5_006_7", + "correct_answer": "Three.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "There were five chairs placed by the table, not three chairs total.", + "D": "There were seven chairs by the table, lined up around its edges.", + "E": "Three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D5_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_006_8", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there by the table?", + "answer": "Five.", + "question_chinese": "桌子边有多少把椅子", + "answer_chinese": "五把", + "evidence": [ + { + "description": "A girl and I each carried a chair to the table.", + "timestamps": { + "start": "00:06:05", + "end": "00:06:13" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Two.", + "C": "Four.", + "D": "There were three chairs by the table, rather than five in total.", + "E": "There was only one chair by the table instead of a group of five." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D5_006_8", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "Two.", + "C": "Four.", + "D": "There were three chairs by the table, rather than five in total.", + "E": "There was only one chair by the table instead of a group of five." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D5_007-mcq.json b/QAs/egolife_A1_D5_007-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ae3446511ab0223ee4e8beda581655a0058f2ecb --- /dev/null +++ b/QAs/egolife_A1_D5_007-mcq.json @@ -0,0 +1,220 @@ +[ + { + "video_uid": "egolife_A1_D5_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_007_9", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can the lady to my left touch the plants?", + "answer": "Yes.", + "question_chinese": "我左边的女士能不能碰到植物", + "answer_chinese": "可以", + "evidence": [ + { + "description": "I saw her touch a leaf.", + "timestamps": { + "start": "00:09:05", + "end": "00:09:07" + } + } + ], + "question_timestamps": "00:10:03", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D5_007_9", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D5_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_007_10", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can the lady to my left touch the plants?", + "answer": "Yes.", + "question_chinese": "我左边的女士能不能碰到植物", + "answer_chinese": "能", + "evidence": [ + { + "description": "I saw her touch a leaf.", + "timestamps": { + "start": "00:09:05", + "end": "00:09:07" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D5_007_10", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D5_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_007_7", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many women are in my line of sight?", + "answer": "Three.", + "question_chinese": "我视野里有多少位女士", + "answer_chinese": "三位", + "evidence": [ + { + "description": "I saw three ladies sitting at the table.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "Six.", + "D": "Five women were in your line of sight, rather than only three women.", + "E": "Seven women were visible in front of you within your line of sight." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "egolife_A1_D5_007_7", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "Six.", + "D": "Five women were in your line of sight, rather than only three women.", + "E": "Seven women were visible in front of you within your line of sight." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_007_8", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many women are in my line of sight?", + "answer": "Four.", + "question_chinese": "我视野里有多少位女士", + "answer_chinese": "四位", + "evidence": [ + { + "description": "Three ladies at the table, one lady on the sofa.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:30" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "One.", + "C": "Two.", + "D": "Three women are in your line of sight right now, standing close together in front of you.", + "E": "Five women are in your line of sight, with a couple off to the side and others ahead." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "egolife_A1_D5_007_8", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "One.", + "C": "Two.", + "D": "Three women are in your line of sight right now, standing close together in front of you.", + "E": "Five women are in your line of sight, with a couple off to the side and others ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D5_008-mcq.json b/QAs/egolife_A1_D5_008-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..faf73e87ba7405f448e3c4ae11769900184c8ebe --- /dev/null +++ b/QAs/egolife_A1_D5_008-mcq.json @@ -0,0 +1,134 @@ +[ + { + "video_uid": "egolife_A1_D5_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_008_3", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my left knee pointing right now?", + "answer": "The door on the second floor.", + "question_chinese": "我刚刚左膝盖朝向哪里", + "answer_chinese": "二楼的门", + "evidence": [ + { + "description": "I saw that my left knee was pointing toward the door on the second floor.", + "timestamps": { + "start": "00:01:58", + "end": "00:01:59" + } + }, + { + "description": "I stood with my legs apart.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:51" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The window on the second floor.", + "B": "The door on the first floor.", + "C": "Your left knee is pointing toward the wall on your left side as you face forward.", + "D": "The door on the second floor.", + "E": "Your left knee is pointing toward the stairs on the third floor, angled slightly upward." + }, + "answer_label": "D", + "answer_text": "The door on the second floor.", + "q_id": "egolife_A1_D5_008_3", + "correct_answer": "The door on the second floor.", + "original_options": { + "A": "The window on the second floor.", + "B": "The door on the first floor.", + "C": "Your left knee is pointing toward the wall on your left side as you face forward.", + "D": "The door on the second floor.", + "E": "Your left knee is pointing toward the stairs on the third floor, angled slightly upward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_008_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my left knee pointing right now?", + "answer": "Straight ahead.", + "question_chinese": "我刚刚左膝盖朝向哪里", + "answer_chinese": "正前方", + "evidence": [ + { + "description": "I closed my legs slightly.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:51" + } + }, + { + "description": "My left knee is pointing at the girl in front of me.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:51" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Straight ahead.", + "B": "Slightly to the left.", + "C": "Straight down.", + "D": "Your left knee is pointing toward your green backpack, angled a bit off-center.", + "E": "Your left knee is pointing slightly backward and to your left, not aligned with your forward direction." + }, + "answer_label": "A", + "answer_text": "Straight ahead.", + "q_id": "egolife_A1_D5_008_4", + "correct_answer": "Straight ahead.", + "original_options": { + "A": "Straight ahead.", + "B": "Slightly to the left.", + "C": "Straight down.", + "D": "Your left knee is pointing toward your green backpack, angled a bit off-center.", + "E": "Your left knee is pointing slightly backward and to your left, not aligned with your forward direction." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 13, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D5_009-mcq.json b/QAs/egolife_A1_D5_009-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fb20ab6f6b61ec243f9cf22548f6611e4819b8c3 --- /dev/null +++ b/QAs/egolife_A1_D5_009-mcq.json @@ -0,0 +1,494 @@ +[ + { + "video_uid": "egolife_A1_D5_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_009_1", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "My recent activity trail.", + "answer": "Walk two circles clockwise.", + "question_chinese": "我刚刚的行动轨迹", + "answer_chinese": "顺时针走两圈", + "evidence": [ + { + "description": "I walked two circles clockwise along the second floor.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:37" + } + } + ], + "question_timestamps": "00:02:33", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk one circle clockwise.", + "B": "Walk two circles counterclockwise.", + "C": "Walk two circles clockwise.", + "D": "Walk two circles clockwise, then stop halfway through a third loop before returning to where you started.", + "E": "Walk in a clockwise circle once, reverse direction, and complete a second circle counterclockwise." + }, + "answer_label": "C", + "answer_text": "Walk two circles clockwise.", + "q_id": "egolife_A1_D5_009_1", + "correct_answer": "Walk two circles clockwise.", + "original_options": { + "A": "Walk one circle clockwise.", + "B": "Walk two circles counterclockwise.", + "C": "Walk two circles clockwise.", + "D": "Walk two circles clockwise, then stop halfway through a third loop before returning to where you started.", + "E": "Walk in a clockwise circle once, reverse direction, and complete a second circle counterclockwise." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 14 + ] + }, + { + "video_uid": "egolife_A1_D5_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_009_2", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "My recent activity trail.", + "answer": "Go around clockwise twice.", + "question_chinese": "我刚刚的行动轨迹", + "answer_chinese": "顺时针走两圈", + "evidence": [ + { + "description": "I walked two circles clockwise along the second floor.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:37" + } + } + ], + "question_timestamps": "00:06:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go around counterclockwise twice.", + "B": "Go around clockwise twice.", + "C": "Go around clockwise once.", + "D": "Go around clockwise three times, completing one extra loop beyond what you just did.", + "E": "Go around counterclockwise once, reversing direction and stopping after a single circuit." + }, + "answer_label": "B", + "answer_text": "Go around clockwise twice.", + "q_id": "egolife_A1_D5_009_2", + "correct_answer": "Go around clockwise twice.", + "original_options": { + "A": "Go around counterclockwise twice.", + "B": "Go around clockwise twice.", + "C": "Go around clockwise once.", + "D": "Go around clockwise three times, completing one extra loop beyond what you just did.", + "E": "Go around counterclockwise once, reversing direction and stopping after a single circuit." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 12 + ] + }, + { + "video_uid": "egolife_A1_D5_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_009_3_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many phones were in my field of view?", + "answer": "There were 2 in your view. They were in the hands of two people in front of you.", + "question_chinese": "我视野里有多少个手机？", + "answer_chinese": "你的视野里有2个。在你前面两个人的手里。", + "evidence": [ + { + "description": "There were two ladies in my field of vision.Each of them was holding a phone.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "One.", + "C": "Three phones were in your view, with one person holding two phones in front of you.", + "D": "There were 2 in your view. They were in the hands of two people in front of you.", + "E": "Four phones were in your view, with each of the two people holding two phones." + }, + "answer_label": "D", + "answer_text": "There were 2 in your view. They were in the hands of two people in front of you.", + "q_id": "egolife_A1_D5_009_3_T1", + "correct_answer": "There were 2 in your view. They were in the hands of two people in front of you.", + "original_options": { + "A": "Zero.", + "B": "One.", + "C": "Three phones were in your view, with one person holding two phones in front of you.", + "D": "There were 2 in your view. They were in the hands of two people in front of you.", + "E": "Four phones were in your view, with each of the two people holding two phones." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_009_7", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many phones are in my line of sight?", + "answer": "Two.", + "question_chinese": "我视野里有多少个手机", + "answer_chinese": "两个", + "evidence": [ + { + "description": "There were two ladies in my field of vision.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:37" + } + }, + { + "description": "Each of them was holding a phone.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four phones are in your line of sight, with a couple clustered together and the rest nearby.", + "D": "Two.", + "E": "Five phones are in your line of sight, spread across the area in front of you." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "egolife_A1_D5_009_7", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four phones are in your line of sight, with a couple clustered together and the rest nearby.", + "D": "Two.", + "E": "Five phones are in your line of sight, spread across the area in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_009_3_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many phones were in my field of view?", + "answer": "There was 1 in your view. You were holding it in your hand.", + "question_chinese": "我视野里有多少个手机？", + "answer_chinese": "你的视野里有1个。在你的手里拿着。", + "evidence": [ + { + "description": "I was looking at my phone.the only thing in my field of vision was my phone.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:08:23", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "Two.", + "C": "Three phones were in your view, with one in your hand and two visible nearby.", + "D": "Four phones were in your view, with one in your hand and three visible nearby.", + "E": "There was 1 in your view. You were holding it in your hand." + }, + "answer_label": "E", + "answer_text": "There was 1 in your view. You were holding it in your hand.", + "q_id": "egolife_A1_D5_009_3_T2", + "correct_answer": "There was 1 in your view. You were holding it in your hand.", + "original_options": { + "A": "Zero.", + "B": "Two.", + "C": "Three phones were in your view, with one in your hand and two visible nearby.", + "D": "Four phones were in your view, with one in your hand and three visible nearby.", + "E": "There was 1 in your view. You were holding it in your hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_009_8", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mobile phones are in my view?", + "answer": "One.", + "question_chinese": "我视野里有多少个手机", + "answer_chinese": "一个", + "evidence": [ + { + "description": "I was looking at my phone.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:20" + } + }, + { + "description": "The only thing in my field of vision was my phone.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four mobile phones are visible in your view at the same time.", + "D": "Five mobile phones appear in your view, grouped closely together on the table.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_D5_009_8", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four mobile phones are visible in your view at the same time.", + "D": "Five mobile phones appear in your view, grouped closely together on the table.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_009_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are in my line of sight?", + "answer": "Two.", + "question_chinese": "我视野里有多少把椅子", + "answer_chinese": "两把", + "evidence": [ + { + "description": "There weree two chairs in my field of vision, in the kitchen.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:40" + } + }, + { + "description": "No other chairs.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:40" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "There are four chairs visible in your line of sight right now.", + "E": "You have five chairs clearly visible in your line of sight right now." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "egolife_A1_D5_009_5", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "There are four chairs visible in your line of sight right now.", + "E": "You have five chairs clearly visible in your line of sight right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_009_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are in my line of sight?", + "answer": "Six.", + "question_chinese": "我视野里有多少把椅子", + "answer_chinese": "六把", + "evidence": [ + { + "description": "There are four chairs in front of the table.", + "timestamps": { + "start": "00:04:17", + "end": "00:04:18" + } + }, + { + "description": "There are two chairs on either side of the table.", + "timestamps": { + "start": "00:04:17", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "There are four chairs visible in your line of sight right now.", + "D": "You can see eight chairs in your line of sight right now.", + "E": "Six." + }, + "answer_label": "E", + "answer_text": "Six.", + "q_id": "egolife_A1_D5_009_6", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "There are four chairs visible in your line of sight right now.", + "D": "You can see eight chairs in your line of sight right now.", + "E": "Six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + } +] diff --git a/QAs/egolife_A1_D5_010-mcq.json b/QAs/egolife_A1_D5_010-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..bedc2f812ec845e968da8763c2a222633cbd24d5 --- /dev/null +++ b/QAs/egolife_A1_D5_010-mcq.json @@ -0,0 +1,120 @@ +[ + { + "video_uid": "egolife_A1_D5_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_010_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I sit down, where is the refrigerator in relation to my position?", + "answer": "The fridge is right in front of me.", + "question_chinese": "我坐下时，冰箱相对我的位置", + "answer_chinese": "冰箱在我前面", + "evidence": [ + { + "description": "When I sit down, the fridge is in front of me.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "The fridge is right in front of me.", + "C": "Behind you.", + "D": "The refrigerator is off to your right as you sit down now.", + "E": "When you sit down, the refrigerator is behind you rather than in front." + }, + "answer_label": "B", + "answer_text": "The fridge is right in front of me.", + "q_id": "egolife_A1_D5_010_1", + "correct_answer": "The fridge is right in front of me.", + "original_options": { + "A": "To your left.", + "B": "The fridge is right in front of me.", + "C": "Behind you.", + "D": "The refrigerator is off to your right as you sit down now.", + "E": "When you sit down, the refrigerator is behind you rather than in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_010_2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I sit down, where is the refrigerator in relation to my position?", + "answer": "The fridge is in front of me.", + "question_chinese": "我坐下时，冰箱相对我的位置", + "answer_chinese": "在我前面", + "evidence": [ + { + "description": "When I sit down, the fridge is in front of me.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:05:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The refrigerator is positioned to your right when you sit down now.", + "D": "As you sit down, the refrigerator ends up behind you, not in front.", + "E": "The fridge is in front of me." + }, + "answer_label": "E", + "answer_text": "The fridge is in front of me.", + "q_id": "egolife_A1_D5_010_2", + "correct_answer": "The fridge is in front of me.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The refrigerator is positioned to your right when you sit down now.", + "D": "As you sit down, the refrigerator ends up behind you, not in front.", + "E": "The fridge is in front of me." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 12, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D5_011-mcq.json b/QAs/egolife_A1_D5_011-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4a988c195a7f266801759049623c2ad3ec2451d8 --- /dev/null +++ b/QAs/egolife_A1_D5_011-mcq.json @@ -0,0 +1,148 @@ +[ + { + "video_uid": "egolife_A1_D5_011", + "video_length": "00:11:43", + "qid": "egolife_A1_D5_011_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in the shadows?", + "answer": "Five.", + "question_chinese": "阴影里有多少人", + "answer_chinese": "五个人", + "evidence": [ + { + "description": "The two girls opposite were sitting in the shadows.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + }, + { + "description": "A boy in front of me was sitting in the shadows.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + }, + { + "description": "I was standing in the shadows with a boy.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:02:40", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "Three people are visible as shadows, with two shadow figures missing from the group.", + "E": "Seven people appear in the shadows, making the group larger than the correct count." + }, + "answer_label": "C", + "answer_text": "Five.", + "q_id": "egolife_A1_D5_011_3", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "Three people are visible as shadows, with two shadow figures missing from the group.", + "E": "Seven people appear in the shadows, making the group larger than the correct count." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D5_011", + "video_length": "00:11:43", + "qid": "egolife_A1_D5_011_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are in the shadows?", + "answer": "Four.", + "question_chinese": "阴影里有多少人", + "answer_chinese": "四个", + "evidence": [ + { + "description": "The two girls opposite were sitting in the shadows.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + }, + { + "description": "A boy in front of me was sitting in the shadows.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + }, + { + "description": "I was standing in the shadows.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Two people are visible in the shadows, with only a pair of silhouettes present.", + "E": "Six people appear as shadows, exceeding the correct number by two people overall." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "egolife_A1_D5_011_4", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Four.", + "C": "Five.", + "D": "Two people are visible in the shadows, with only a pair of silhouettes present.", + "E": "Six people appear as shadows, exceeding the correct number by two people overall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D5_012-mcq.json b/QAs/egolife_A1_D5_012-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c5b7d6d5fbd2feeddc1ab7fb1f61e0e55281d110 --- /dev/null +++ b/QAs/egolife_A1_D5_012-mcq.json @@ -0,0 +1,705 @@ +[ + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are on the table in the living room?", + "answer": "There are two power banks on the table in the living room.", + "question_chinese": "客厅的桌子上有几个充电宝？", + "answer_chinese": "客厅的桌子有两个充电宝。", + "evidence": [ + { + "description": "There are two power banks on the table in the living room.", + "timestamps": { + "start": "00:08:05", + "end": "00:08:06" + } + } + ], + "question_timestamps": "00:08:18", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one power bank.", + "B": "There are two power banks on the table in the living room.", + "C": "There are three power banks.", + "D": "There are four power banks on the table in the living room.", + "E": "There are five power banks sitting on the table in the living room." + }, + "answer_label": "B", + "answer_text": "There are two power banks on the table in the living room.", + "q_id": "egolife_A1_D5_012_3", + "correct_answer": "There are two power banks on the table in the living room.", + "original_options": { + "A": "There is one power bank.", + "B": "There are two power banks on the table in the living room.", + "C": "There are three power banks.", + "D": "There are four power banks on the table in the living room.", + "E": "There are five power banks sitting on the table in the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are on the table in the living room?", + "answer": "There is no power bank on the table in the living room.", + "question_chinese": "客厅的桌子上有几个充电宝？", + "answer_chinese": "客厅的桌子上没有充电宝。", + "evidence": [ + { + "description": "There are two power banks on the table in the living room.", + "timestamps": { + "start": "00:08:05", + "end": "00:08:06" + } + }, + { + "description": "There is no power bank on the table in the living room.", + "timestamps": { + "start": "00:10:20", + "end": "00:10:22" + } + } + ], + "question_timestamps": "00:10:35", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One power bank.", + "B": "Two power banks.", + "C": "There are three power banks on the table in the living room, lined up near the edge.", + "D": "There is no power bank on the table in the living room.", + "E": "There are four power banks on the table in the living room, placed together beside each other." + }, + "answer_label": "D", + "answer_text": "There is no power bank on the table in the living room.", + "q_id": "egolife_A1_D5_012_4", + "correct_answer": "There is no power bank on the table in the living room.", + "original_options": { + "A": "One power bank.", + "B": "Two power banks.", + "C": "There are three power banks on the table in the living room, lined up near the edge.", + "D": "There is no power bank on the table in the living room.", + "E": "There are four power banks on the table in the living room, placed together beside each other." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I gone from outside to inside?", + "answer": "You went from outside to inside once.", + "question_chinese": "我从屋外到室内了几次？", + "answer_chinese": "你屋外到室内1次。", + "evidence": [ + { + "description": "You went from outside to inside once.", + "timestamps": { + "start": "00:06:28", + "end": "00:06:30" + } + } + ], + "question_timestamps": "00:06:41", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You went from outside to inside twice.", + "B": "You went from outside to inside three times.", + "C": "You went from outside to inside once.", + "D": "You went from outside to inside four times over the course of the video, returning indoors each time.", + "E": "You went from outside to inside five times, repeatedly coming back into the house after being outdoors." + }, + "answer_label": "C", + "answer_text": "You went from outside to inside once.", + "q_id": "egolife_A1_D5_012_5", + "correct_answer": "You went from outside to inside once.", + "original_options": { + "A": "You went from outside to inside twice.", + "B": "You went from outside to inside three times.", + "C": "You went from outside to inside once.", + "D": "You went from outside to inside four times over the course of the video, returning indoors each time.", + "E": "You went from outside to inside five times, repeatedly coming back into the house after being outdoors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I gone from outside to inside?", + "answer": "You went in and out of my house three times.", + "question_chinese": "我从屋外到室内了几次？", + "answer_chinese": "你屋外到室内3次。", + "evidence": [ + { + "description": "You went from outside to inside once.", + "timestamps": { + "start": "00:06:28", + "end": "00:06:30" + } + }, + { + "description": "You went from outside to inside once.", + "timestamps": { + "start": "00:07:51", + "end": "00:07:53" + } + }, + { + "description": "You went from outside to inside once.", + "timestamps": { + "start": "00:10:37", + "end": "00:10:39" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You went in and out twice.", + "B": "You went in and out four times.", + "C": "You went in and out of my house five times, crossing from outdoors to indoors repeatedly.", + "D": "You went in and out of my house six times, entering the house again after each trip outside.", + "E": "You went in and out of my house three times." + }, + "answer_label": "E", + "answer_text": "You went in and out of my house three times.", + "q_id": "egolife_A1_D5_012_6", + "correct_answer": "You went in and out of my house three times.", + "original_options": { + "A": "You went in and out twice.", + "B": "You went in and out four times.", + "C": "You went in and out of my house five times, crossing from outdoors to indoors repeatedly.", + "D": "You went in and out of my house six times, entering the house again after each trip outside.", + "E": "You went in and out of my house three times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012-mcq_fixed_5", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many friends and I are in the yard?", + "answer": "Four.", + "question_chinese": "我和多少个朋友在院子里面？", + "answer_chinese": "你和7个朋友在院子里面", + "evidence": [ + { + "description": "There are seven people in front of you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Four.", + "D": "Three people total were in the yard, including you and two friends.", + "E": "Five people total were in the yard, with you standing there alongside four friends." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "egolife_A1_D5_012-mcq_fixed_5", + "correct_answer": "Four.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Four.", + "D": "Three people total were in the yard, including you and two friends.", + "E": "Five people total were in the yard, with you standing there alongside four friends." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012-mcq_fixed_6", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many friends and I are in the yard?", + "answer": "Five.", + "question_chinese": "我和多少个朋友在院子里面？", + "answer_chinese": "你和8个朋友在院子里面", + "evidence": [ + { + "description": "There are seven people in front of you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "There are five people in front of you.", + "timestamps": { + "start": "00:03:08", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:03:12", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three people total were in the yard, including you and two friends standing nearby.", + "D": "Six people total were in the yard, with you and five friends gathered together outside.", + "E": "Five." + }, + "answer_label": "E", + "answer_text": "Five.", + "q_id": "egolife_A1_D5_012-mcq_fixed_6", + "correct_answer": "Five.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three people total were in the yard, including you and two friends standing nearby.", + "D": "Six people total were in the yard, with you and five friends gathered together outside.", + "E": "Five." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012_4_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the laptop charger right now?", + "answer": "You could reach the computer charger in front of you; it was directly in front of you.", + "question_chinese": "我现在能够拿的到电脑充电器吗？", + "answer_chinese": "你能够拿到你面前的电脑充电器，电脑充电器就在你的正前方。", + "evidence": [ + { + "description": "The computer charger is on the table in the yard.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:03:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn’t reach the computer charger in front of you; it was on the table downstairs.", + "B": "You could reach the computer charger in front of you; it was directly in front of you." + }, + "answer_label": "B", + "answer_text": "You could reach the computer charger in front of you; it was directly in front of you.", + "q_id": "egolife_A1_D5_012_4_T1", + "correct_answer": "You could reach the computer charger in front of you; it was directly in front of you.", + "original_options": { + "A": "You couldn’t reach the computer charger in front of you; it was on the table downstairs.", + "B": "You could reach the computer charger in front of you; it was directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012_4_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the laptop charger right now?", + "answer": "You couldn’t reach the computer charger behind you; it was on the table downstairs.", + "question_chinese": "我现在能够拿的到电脑充电器吗？", + "answer_chinese": "你不能够拿到你后面的电脑充电器，电脑充电器在你楼下的桌子上。", + "evidence": [ + { + "description": "The computer charger is on the table in the yard.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:08:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn’t reach the computer charger behind you; it was on the table downstairs.", + "B": "You could reach the computer charger behind you; it was right within arm’s reach." + }, + "answer_label": "A", + "answer_text": "You couldn’t reach the computer charger behind you; it was on the table downstairs.", + "q_id": "egolife_A1_D5_012_4_T2", + "correct_answer": "You couldn’t reach the computer charger behind you; it was on the table downstairs.", + "original_options": { + "A": "You couldn’t reach the computer charger behind you; it was on the table downstairs.", + "B": "You could reach the computer charger behind you; it was right within arm’s reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012_10", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks do I have with me right now?", + "answer": "You currently have two power banks on me.", + "question_chinese": "我现在身上有几个充电宝？", + "answer_chinese": "你现在身上有2个充电宝", + "evidence": [ + { + "description": "You took the two power banks on the table in the living room.", + "timestamps": { + "start": "00:08:06", + "end": "00:08:08" + } + }, + { + "description": "You gave one charging pad to my friend.", + "timestamps": { + "start": "00:10:04", + "end": "00:10:06" + } + }, + { + "description": "You took one charging battery from the table in the yard.", + "timestamps": { + "start": "00:10:20", + "end": "00:10:22" + } + } + ], + "question_timestamps": "00:10:35", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have one power bank.", + "B": "You have three power banks.", + "C": "You currently have four power banks on you, so you’re carrying more than just a pair.", + "D": "You currently have two power banks on me.", + "E": "You currently have five power banks on you, meaning there are several backups with you right now." + }, + "answer_label": "D", + "answer_text": "You currently have two power banks on me.", + "q_id": "egolife_A1_D5_012_10", + "correct_answer": "You currently have two power banks on me.", + "original_options": { + "A": "You have one power bank.", + "B": "You have three power banks.", + "C": "You currently have four power banks on you, so you’re carrying more than just a pair.", + "D": "You currently have two power banks on me.", + "E": "You currently have five power banks on you, meaning there are several backups with you right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012_9", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks do I have with me right now?", + "answer": "You have one power bank on me now.", + "question_chinese": "我现在身上有几个充电宝？", + "answer_chinese": "你现在身上有1个充电宝。", + "evidence": [ + { + "description": "You took the two power banks on the table in the living room.", + "timestamps": { + "start": "00:08:06", + "end": "00:08:08" + } + }, + { + "description": "You gave one charging pad to my friend.", + "timestamps": { + "start": "00:10:04", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:11:05", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have two power banks.", + "B": "You have one power bank on me now.", + "C": "You have three power banks.", + "D": "You have four power banks on you now, so you’re carrying multiple spares at the moment.", + "E": "You have five power banks on you now, which is far more than a single one." + }, + "answer_label": "B", + "answer_text": "You have one power bank on me now.", + "q_id": "egolife_A1_D5_012_9", + "correct_answer": "You have one power bank on me now.", + "original_options": { + "A": "You have two power banks.", + "B": "You have one power bank on me now.", + "C": "You have three power banks.", + "D": "You have four power banks on you now, so you’re carrying multiple spares at the moment.", + "E": "You have five power banks on you now, which is far more than a single one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012-mcq_fixed_10", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the washing machine at this moment in relation to me?", + "answer": "The washing machine is on your left.", + "question_chinese": "此时此刻，洗衣机在我的哪边？", + "answer_chinese": "洗衣机在你的左手边。", + "evidence": [ + { + "description": "The washing machine is on your left.", + "timestamps": { + "start": "00:06:30", + "end": "00:06:32" + } + } + ], + "question_timestamps": "00:06:33", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The washing machine is on your right.", + "B": "The washing machine is on your left.", + "C": "The washing machine is behind you.", + "D": "The washing machine is directly in front of you, so you’d face it without turning left.", + "E": "The washing machine is in front of you to the right, positioned off your right shoulder." + }, + "answer_label": "B", + "answer_text": "The washing machine is on your left.", + "q_id": "egolife_A1_D5_012-mcq_fixed_10", + "correct_answer": "The washing machine is on your left.", + "original_options": { + "A": "The washing machine is on your right.", + "B": "The washing machine is on your left.", + "C": "The washing machine is behind you.", + "D": "The washing machine is directly in front of you, so you’d face it without turning left.", + "E": "The washing machine is in front of you to the right, positioned off your right shoulder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_012-mcq_fixed_11", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the washing machine at this moment in relation to me?", + "answer": "The washing machine is on your right hand side.", + "question_chinese": "此时此刻，洗衣机在我的哪边？", + "answer_chinese": "洗衣机在你的右手边。", + "evidence": [ + { + "description": "The washing machine is on your right hand side.", + "timestamps": { + "start": "00:07:19", + "end": "00:07:20" + } + } + ], + "question_timestamps": "00:07:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The washing machine is on your left.", + "B": "The washing machine is behind you.", + "C": "The washing machine is on your right hand side.", + "D": "The washing machine is directly in front of you, not off to your right hand side.", + "E": "The washing machine is in front of you toward the left, so it isn’t on your right." + }, + "answer_label": "C", + "answer_text": "The washing machine is on your right hand side.", + "q_id": "egolife_A1_D5_012-mcq_fixed_11", + "correct_answer": "The washing machine is on your right hand side.", + "original_options": { + "A": "The washing machine is on your left.", + "B": "The washing machine is behind you.", + "C": "The washing machine is on your right hand side.", + "D": "The washing machine is directly in front of you, not off to your right hand side.", + "E": "The washing machine is in front of you toward the left, so it isn’t on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D5_013-mcq.json b/QAs/egolife_A1_D5_013-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7d5ef91c75f60fde85549a7af6c80c4b4dcd25c1 --- /dev/null +++ b/QAs/egolife_A1_D5_013-mcq.json @@ -0,0 +1,740 @@ +[ + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013_3_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there a paper lion at the entrance of my house?", + "answer": "There are no paper lions at the entrance of the house; it is at the doorway in front of you.", + "question_chinese": "我房子门口有纸狮子吗？", + "answer_chinese": "房子门口没有纸狮子，它在你前面的门口里。", + "evidence": [ + { + "description": "There is no paper lion at the entrance of the house.", + "timestamps": { + "start": "00:04:26", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:04:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There is one paper lion at the entrance.", + "B": "There are two paper lions at the entrance.", + "C": "There are three paper lions at the entrance of the house; they are not at the doorway in front of you.", + "D": "There are no paper lions at the entrance of the house; it is at the doorway in front of you.", + "E": "There are five paper lions at the entrance of the house; they are not at the doorway in front of you." + }, + "answer_label": "D", + "answer_text": "There are no paper lions at the entrance of the house; it is at the doorway in front of you.", + "q_id": "egolife_A1_D5_013_3_T1", + "correct_answer": "There are no paper lions at the entrance of the house; it is at the doorway in front of you.", + "original_options": { + "A": "There is one paper lion at the entrance.", + "B": "There are two paper lions at the entrance.", + "C": "There are three paper lions at the entrance of the house; they are not at the doorway in front of you.", + "D": "There are no paper lions at the entrance of the house; it is at the doorway in front of you.", + "E": "There are five paper lions at the entrance of the house; they are not at the doorway in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013_3_T4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there a paper lion at the entrance of my house?", + "answer": "There are two paper lions at the entrance of the house, and they are to your left.", + "question_chinese": "我房子门口有纸狮子吗？", + "answer_chinese": "房子门口有两只纸狮子，它在你左边。", + "evidence": [ + { + "description": "Your friend placed two paper lions at the door.", + "timestamps": { + "start": "00:04:29", + "end": "00:04:30" + } + } + ], + "question_timestamps": "00:04:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There is one paper lion to your left.", + "B": "There are two paper lions at the entrance of the house, and they are to your left.", + "C": "There are three paper lions to your left.", + "D": "There are zero paper lions at the entrance of the house, so nothing is sitting to your left.", + "E": "There are four paper lions at the entrance of the house, and they are grouped to your left." + }, + "answer_label": "B", + "answer_text": "There are two paper lions at the entrance of the house, and they are to your left.", + "q_id": "egolife_A1_D5_013_3_T4", + "correct_answer": "There are two paper lions at the entrance of the house, and they are to your left.", + "original_options": { + "A": "There is one paper lion to your left.", + "B": "There are two paper lions at the entrance of the house, and they are to your left.", + "C": "There are three paper lions to your left.", + "D": "There are zero paper lions at the entrance of the house, so nothing is sitting to your left.", + "E": "There are four paper lions at the entrance of the house, and they are grouped to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013_3_T3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there a paper lion at the entrance of my house?", + "answer": "There are two paper lions at the entrance of the house.", + "question_chinese": "我房子门口有纸狮子吗？", + "answer_chinese": "房子门口有两只纸狮子。", + "evidence": [ + { + "description": "There is no paper lion at the entrance of the house.", + "timestamps": { + "start": "00:04:26", + "end": "00:04:28" + } + }, + { + "description": "Your friend placed two paper lions at the door.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:32" + } + } + ], + "question_timestamps": "00:08:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "There are no paper lions at the entrance of the house.", + "B": "There are two paper lions at the entrance of the house." + }, + "answer_label": "B", + "answer_text": "There are two paper lions at the entrance of the house.", + "q_id": "egolife_A1_D5_013_3_T3", + "correct_answer": "There are two paper lions at the entrance of the house.", + "original_options": { + "A": "There are no paper lions at the entrance of the house.", + "B": "There are two paper lions at the entrance of the house." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013-mcq_fixed_8", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many friends am I with now?", + "answer": "Three.", + "question_chinese": "我现在和多少个朋友在一起？", + "answer_chinese": "你现在和7个朋友在一起。", + "evidence": [ + { + "description": "You are currently with 7 friends.", + "timestamps": { + "start": "00:08:45", + "end": "00:08:50" + } + } + ], + "question_timestamps": "00:08:51", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "Right now, you are with two friends, not three, in the current scene.", + "E": "At the moment, you are with four friends rather than three, based on what is visible." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_D5_013-mcq_fixed_8", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "Right now, you are with two friends, not three, in the current scene.", + "E": "At the moment, you are with four friends rather than three, based on what is visible." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013-mcq_fixed_7", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many friends am I with now?", + "answer": "Five.", + "question_chinese": "我现在和多少个朋友在一起？", + "answer_chinese": "你现在和3个朋友在一起。", + "evidence": [ + { + "description": "You are currently with 3 friends.", + "timestamps": { + "start": "00:09:24", + "end": "00:09:26" + } + }, + { + "description": "You are currently with 3 friends and are shopping in the supermarket.", + "timestamps": { + "start": "00:11:06", + "end": "00:11:08" + } + } + ], + "question_timestamps": "00:11:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "You are with four friends right now, not five, based on the current view.", + "D": "Five.", + "E": "Right now, you are with six friends rather than five, according to the scene you see." + }, + "answer_label": "D", + "answer_text": "Five.", + "q_id": "egolife_A1_D5_013-mcq_fixed_7", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "You are with four friends right now, not five, based on the current view.", + "D": "Five.", + "E": "Right now, you are with six friends rather than five, according to the scene you see." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013_3", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the living room now?", + "answer": "Now turn back, enter the door, go straight through the corridor, then turn right and go straight again, and you will reach the living room.", + "question_chinese": "我现在怎么可以到达客厅？", + "answer_chinese": "你现在向后转身进门直走穿过走廊，然后右转再直走，就可以到达客厅。", + "evidence": [ + { + "description": "You are now crossing the corridor.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:06" + } + }, + { + "description": "Now you turn right and walk straight to the living room.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:02:17", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Now turn back, enter the door, go straight through the corridor, then turn right and go straight again, and you will reach the living room.", + "B": "Turn back and turn left.", + "C": "Turn back, then go right.", + "D": "Turn back, enter the door, go through the corridor, then turn left and continue straight to the living room.", + "E": "Turn back, enter the door, walk straight down the corridor, then turn right and immediately turn left to reach the living room." + }, + "answer_label": "A", + "answer_text": "Now turn back, enter the door, go straight through the corridor, then turn right and go straight again, and you will reach the living room.", + "q_id": "egolife_A1_D5_013_3", + "correct_answer": "Now turn back, enter the door, go straight through the corridor, then turn right and go straight again, and you will reach the living room.", + "original_options": { + "A": "Now turn back, enter the door, go straight through the corridor, then turn right and go straight again, and you will reach the living room.", + "B": "Turn back and turn left.", + "C": "Turn back, then go right.", + "D": "Turn back, enter the door, go through the corridor, then turn left and continue straight to the living room.", + "E": "Turn back, enter the door, walk straight down the corridor, then turn right and immediately turn left to reach the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 22 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013_4", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the living room now?", + "answer": "Now walk to the front of the door on your right, enter the room, walk straight along the corridor, then turn right and walk straight again, and you will reach the living room.", + "question_chinese": "我现在怎么可以到达客厅？", + "answer_chinese": "你现在向右前面走到门口，再进门进门直走穿过走廊，然后右转再直走，就可以到达客厅。", + "evidence": [ + { + "description": "You are now crossing the corridor.", + "timestamps": { + "start": "00:01:04", + "end": "00:01:06" + } + }, + { + "description": "Now you turn right and walk straight to the living room.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:08" + } + }, + { + "description": "The door is on your right front.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:03:47", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Use the door on your left.", + "B": "Enter and turn left afterward.", + "C": "Walk to the front of the door on your right, enter, go straight, then turn left and continue to the living room.", + "D": "Go to the door on your right, enter the room, walk straight along the corridor, then turn right twice before going straight.", + "E": "Now walk to the front of the door on your right, enter the room, walk straight along the corridor, then turn right and walk straight again, and you will reach the living room." + }, + "answer_label": "E", + "answer_text": "Now walk to the front of the door on your right, enter the room, walk straight along the corridor, then turn right and walk straight again, and you will reach the living room.", + "q_id": "egolife_A1_D5_013_4", + "correct_answer": "Now walk to the front of the door on your right, enter the room, walk straight along the corridor, then turn right and walk straight again, and you will reach the living room.", + "original_options": { + "A": "Use the door on your left.", + "B": "Enter and turn left afterward.", + "C": "Walk to the front of the door on your right, enter, go straight, then turn left and continue to the living room.", + "D": "Go to the door on your right, enter the room, walk straight along the corridor, then turn right twice before going straight.", + "E": "Now walk to the front of the door on your right, enter the room, walk straight along the corridor, then turn right and walk straight again, and you will reach the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 22, + 22 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013_1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I access a computer now?", + "answer": "Yes, there are computers in front of you and on your right. They are about half a meter away from you.", + "question_chinese": "我现在能够的到电脑吗？", + "answer_chinese": "能够到，你的前方与右手边都有电脑，离你半米左右。", + "evidence": [ + { + "description": "You are in your bedroom. There are computers on the desk in the bedroom and beside the bed.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:51" + } + } + ], + "question_timestamps": "00:01:55", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, there are computers in front of you and on your right. They are about half a meter away from you.", + "B": "No, you cannot access a computer right now; there isn't one within reach." + }, + "answer_label": "A", + "answer_text": "Yes, there are computers in front of you and on your right. They are about half a meter away from you.", + "q_id": "egolife_A1_D5_013_1", + "correct_answer": "Yes, there are computers in front of you and on your right. They are about half a meter away from you.", + "original_options": { + "A": "Yes, there are computers in front of you and on your right. They are about half a meter away from you.", + "B": "No, you cannot access a computer right now; there isn't one within reach." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013_2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I access a computer now?", + "answer": "You can't get the computer.", + "question_chinese": "我现在能够的到电脑吗？", + "answer_chinese": "你不能够得到电脑。", + "evidence": [ + { + "description": "You are in your bedroom. There are computers on the desk in the bedroom and beside the bed.", + "timestamps": { + "start": "00:01:50", + "end": "00:01:51" + } + }, + { + "description": "You are in the yard now, but the computer is in the bedroom.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:04:16", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't get the computer.", + "B": "Yes, you can access the computer right now." + }, + "answer_label": "A", + "answer_text": "You can't get the computer.", + "q_id": "egolife_A1_D5_013_2", + "correct_answer": "You can't get the computer.", + "original_options": { + "A": "You can't get the computer.", + "B": "Yes, you can access the computer right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013_7", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there a paper lion at the entrance of the house?", + "answer": "There is no paper lion at the entrance of the house.", + "question_chinese": "房子门口有纸狮子吗？", + "answer_chinese": "房子门口没有纸狮子。", + "evidence": [ + { + "description": "There is no paper lion at the entrance of the house.", + "timestamps": { + "start": "00:04:26", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:04:40", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "There is no paper lion at the entrance of the house.", + "B": "There is a paper lion at the entrance of the house." + }, + "answer_label": "A", + "answer_text": "There is no paper lion at the entrance of the house.", + "q_id": "egolife_A1_D5_013_7", + "correct_answer": "There is no paper lion at the entrance of the house.", + "original_options": { + "A": "There is no paper lion at the entrance of the house.", + "B": "There is a paper lion at the entrance of the house." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013_8", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is there a paper lion at the entrance of the house?", + "answer": "There are two paper lions at the entrance of the house.", + "question_chinese": "房子门口有纸狮子吗？", + "answer_chinese": "房子门口有两只纸狮子。", + "evidence": [ + { + "description": "There is no paper lion at the entrance of the house.", + "timestamps": { + "start": "00:04:26", + "end": "00:04:28" + } + }, + { + "description": "Your friend placed two paper lions at the door.", + "timestamps": { + "start": "00:08:30", + "end": "00:08:32" + } + } + ], + "question_timestamps": "00:08:45", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One paper lion.", + "B": "Three paper lions.", + "C": "There are two paper lions at the entrance of the house.", + "D": "There is one paper lion at the entrance of the house, visible near the doorway right now.", + "E": "There are three paper lions placed at the entrance of the house, standing together by the door." + }, + "answer_label": "C", + "answer_text": "There are two paper lions at the entrance of the house.", + "q_id": "egolife_A1_D5_013_8", + "correct_answer": "There are two paper lions at the entrance of the house.", + "original_options": { + "A": "One paper lion.", + "B": "Three paper lions.", + "C": "There are two paper lions at the entrance of the house.", + "D": "There is one paper lion at the entrance of the house, visible near the doorway right now.", + "E": "There are three paper lions placed at the entrance of the house, standing together by the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013-mcq_fixed_4", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Based on my current position, is the refrigerator in that direction?", + "answer": "The refrigerator is in your left front.", + "question_chinese": "根据我现在的位置，冰箱在我的那个方向？", + "answer_chinese": "冰箱在你的左前方。", + "evidence": [ + { + "description": "The refrigerator is in the front left of you.", + "timestamps": { + "start": "00:05:41", + "end": "00:05:42" + } + } + ], + "question_timestamps": "00:05:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is straight ahead.", + "B": "The refrigerator is in your right front.", + "C": "The refrigerator is in your left front.", + "D": "The refrigerator is behind you, slightly toward your right side from where you're standing now.", + "E": "The refrigerator is on your left rear side, not in front of you from your current position." + }, + "answer_label": "C", + "answer_text": "The refrigerator is in your left front.", + "q_id": "egolife_A1_D5_013-mcq_fixed_4", + "correct_answer": "The refrigerator is in your left front.", + "original_options": { + "A": "The refrigerator is straight ahead.", + "B": "The refrigerator is in your right front.", + "C": "The refrigerator is in your left front.", + "D": "The refrigerator is behind you, slightly toward your right side from where you're standing now.", + "E": "The refrigerator is on your left rear side, not in front of you from your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_013", + "video_length": "00:11:39", + "qid": "egolife_A1_D5_013-mcq_fixed_9", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Based on my current position, is the refrigerator in that direction?", + "answer": "The refrigerator is on your left rear side.", + "question_chinese": "根据我现在的位置，冰箱在我的那个方向？", + "answer_chinese": "冰箱在你的左后方。", + "evidence": [ + { + "description": "You turned 135 degrees clockwise facing the refrigerator.", + "timestamps": { + "start": "00:06:52", + "end": "00:06:54" + } + } + ], + "question_timestamps": "00:07:02", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is in your left front.", + "B": "The refrigerator is directly ahead of you.", + "C": "The refrigerator is on your left rear side.", + "D": "The refrigerator is behind you on the right side, angled away from your current facing direction.", + "E": "The refrigerator is in front of you and slightly to the right from where you are standing now." + }, + "answer_label": "C", + "answer_text": "The refrigerator is on your left rear side.", + "q_id": "egolife_A1_D5_013-mcq_fixed_9", + "correct_answer": "The refrigerator is on your left rear side.", + "original_options": { + "A": "The refrigerator is in your left front.", + "B": "The refrigerator is directly ahead of you.", + "C": "The refrigerator is on your left rear side.", + "D": "The refrigerator is behind you on the right side, angled away from your current facing direction.", + "E": "The refrigerator is in front of you and slightly to the right from where you are standing now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D5_014-mcq.json b/QAs/egolife_A1_D5_014-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a89c72772efe5871a55ac02404ad59780635f17e --- /dev/null +++ b/QAs/egolife_A1_D5_014-mcq.json @@ -0,0 +1,887 @@ +[ + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_1_T2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I took a shopping cart at the entrance, what route did I walk inside the supermarket?", + "answer": "After you entered the supermarket, you went straight.", + "question_chinese": "从门口拿了购物车后，我在超市里面走了什么样的路线？", + "answer_chinese": "你进超市后直行。", + "evidence": [ + { + "description": "The moment you enter the store, walk straight ahead.", + "timestamps": { + "start": "00:00:55", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "After you entered, you turned right.", + "B": "After you entered, you turned left.", + "C": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight.", + "D": "After you entered the supermarket, you went straight, then turned right, and then continued straight down the next aisle.", + "E": "After you entered the supermarket, you went straight." + }, + "answer_label": "E", + "answer_text": "After you entered the supermarket, you went straight.", + "q_id": "egolife_A1_D5_014_1_T2", + "correct_answer": "After you entered the supermarket, you went straight.", + "original_options": { + "A": "After you entered, you turned right.", + "B": "After you entered, you turned left.", + "C": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight.", + "D": "After you entered the supermarket, you went straight, then turned right, and then continued straight down the next aisle.", + "E": "After you entered the supermarket, you went straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_1_T1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I took a shopping cart at the entrance, what route did I walk inside the supermarket?", + "answer": "After you entered the supermarket, you went straight, then turned right and immediately turned left, and then went straight.", + "question_chinese": "从门口拿了购物车后，我在超市里面走了什么样的路线？", + "answer_chinese": "你进超市后直行，然后右转在立即左转，再直行。", + "evidence": [ + { + "description": "The moment you enter the store, walk straight ahead.", + "timestamps": { + "start": "00:00:55", + "end": "00:01:00" + } + }, + { + "description": "Now, turn right and immediately turn left. You are walking through the snack section.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:01:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "After you entered, you went straight.", + "B": "After you entered, you turned left.", + "C": "After you entered the supermarket, you went straight, then turned right and immediately turned right, and then went straight.", + "D": "After you entered the supermarket, you went straight, then turned right and immediately turned left, and then went straight.", + "E": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right." + }, + "answer_label": "D", + "answer_text": "After you entered the supermarket, you went straight, then turned right and immediately turned left, and then went straight.", + "q_id": "egolife_A1_D5_014_1_T1", + "correct_answer": "After you entered the supermarket, you went straight, then turned right and immediately turned left, and then went straight.", + "original_options": { + "A": "After you entered, you went straight.", + "B": "After you entered, you turned left.", + "C": "After you entered the supermarket, you went straight, then turned right and immediately turned right, and then went straight.", + "D": "After you entered the supermarket, you went straight, then turned right and immediately turned left, and then went straight.", + "E": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_2_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I took a shopping cart at the entrance, what route did I walk inside the supermarket?", + "answer": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right.", + "question_chinese": "从门口拿了购物车后，我在超市里面走了什么样的路线？", + "answer_chinese": "你进入超市后直行，然后右转立即左转，再直行，再右转。", + "evidence": [ + { + "description": "The moment you enter the store, walk straight ahead.", + "timestamps": { + "start": "00:00:55", + "end": "00:01:00" + } + }, + { + "description": "Now, turn right and immediately turn left. You are walking through the snack section.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:17" + } + }, + { + "description": "Now, you have walked straight from the snack section to the beauty and hygiene products section, then turn right .", + "timestamps": { + "start": "00:02:58", + "end": "00:03:00" + } + } + ], + "question_timestamps": "00:03:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right.", + "B": "After you entered, you went straight.", + "C": "After you entered, you turned left.", + "D": "After you entered the supermarket, you went straight, then turned right and immediately turned left, and then went straight.", + "E": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then turned right." + }, + "answer_label": "A", + "answer_text": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right.", + "q_id": "egolife_A1_D5_014_2_T2", + "correct_answer": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right.", + "original_options": { + "A": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right.", + "B": "After you entered, you went straight.", + "C": "After you entered, you turned left.", + "D": "After you entered the supermarket, you went straight, then turned right and immediately turned left, and then went straight.", + "E": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then turned right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_2_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I took a shopping cart at the entrance, what route did I walk inside the supermarket?", + "answer": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right, then went straight.", + "question_chinese": "从门口拿了购物车后，我在超市里面走了什么样的路线？", + "answer_chinese": "你进入超市后直行，然后右转立即左转，再直行，再右转，再直行。", + "evidence": [ + { + "description": "The moment you enter the store, walk straight ahead.", + "timestamps": { + "start": "00:00:55", + "end": "00:01:00" + } + }, + { + "description": "Now, turn right and immediately turn left. You are walking through the snack section.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:17" + } + }, + { + "description": "Now, you have walked straight from the snack section to the beauty and hygiene products section, then turn right and walk straight.", + "timestamps": { + "start": "00:02:58", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:03:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "After you entered, you went straight.", + "B": "After you entered, you turned right.", + "C": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then turned right, then went straight.", + "D": "After you entered the supermarket, you went straight, then turned left and immediately turned left, then went straight.", + "E": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right, then went straight." + }, + "answer_label": "E", + "answer_text": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right, then went straight.", + "q_id": "egolife_A1_D5_014_2_T1", + "correct_answer": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right, then went straight.", + "original_options": { + "A": "After you entered, you went straight.", + "B": "After you entered, you turned right.", + "C": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then turned right, then went straight.", + "D": "After you entered the supermarket, you went straight, then turned left and immediately turned left, then went straight.", + "E": "After you entered the supermarket, you went straight, then turned right and immediately turned left, then went straight, then turned right, then went straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_5", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get a drink now?", + "answer": "You can't get the drink.", + "question_chinese": "我现在能够拿的到饮料吗？", + "answer_chinese": "你拿不到饮料。", + "evidence": [ + { + "description": "You are beside the elevator now. The elevator is not in the snack area.\n2min12 - 2min17 The drink is in the snack area.", + "timestamps": { + "start": "00:03:24", + "end": "00:03:25" + } + } + ], + "question_timestamps": "00:03:31", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't get the drink.", + "B": "You can get the drink now." + }, + "answer_label": "A", + "answer_text": "You can't get the drink.", + "q_id": "egolife_A1_D5_014_5", + "correct_answer": "You can't get the drink.", + "original_options": { + "A": "You can't get the drink.", + "B": "You can get the drink now." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_6", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get a drink now?", + "answer": "You can get the drink now.", + "question_chinese": "我现在能够拿的到饮料吗？", + "answer_chinese": "你能够拿到饮料。", + "evidence": [ + { + "description": "You are beside the elevator now. The elevator is not in the snack area.\n2min12 - 2min17 The drink is in the snack area.", + "timestamps": { + "start": "00:03:24", + "end": "00:03:25" + } + }, + { + "description": "You are in the snack area now. The drink vending machine is on your left.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:23" + } + } + ], + "question_timestamps": "00:06:27", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't get the drink.", + "B": "You can get the drink now." + }, + "answer_label": "B", + "answer_text": "You can get the drink now.", + "q_id": "egolife_A1_D5_014_6", + "correct_answer": "You can get the drink now.", + "original_options": { + "A": "You can't get the drink.", + "B": "You can get the drink now." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014-mcq_fixed_9", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "At this moment, where is the clothing store in relation to me in the supermarket?", + "answer": "The clothing store is on your left.", + "question_chinese": "此时此刻，超市的服装店在我的哪个方位？", + "answer_chinese": "服装店在你的左手边。", + "evidence": [ + { + "description": "Now you enter the supermarket through the main door. The clothing store is on your left.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The clothing store is on your right.", + "B": "The clothing store is behind you.", + "C": "The clothing store is straight ahead of you, across the main aisle near the front of the supermarket.", + "D": "The clothing store is to your right-hand side, a short walk past the next aisle opening.", + "E": "The clothing store is on your left." + }, + "answer_label": "E", + "answer_text": "The clothing store is on your left.", + "q_id": "egolife_A1_D5_014-mcq_fixed_9", + "correct_answer": "The clothing store is on your left.", + "original_options": { + "A": "The clothing store is on your right.", + "B": "The clothing store is behind you.", + "C": "The clothing store is straight ahead of you, across the main aisle near the front of the supermarket.", + "D": "The clothing store is to your right-hand side, a short walk past the next aisle opening.", + "E": "The clothing store is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_8_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, which direction is the supermarket’s clothing store from me?", + "answer": "The clothing store was on your left-hand side.", + "question_chinese": "此时此刻，超市的服装店在我的哪个方位？", + "answer_chinese": "服装店在你的左手边。", + "evidence": [ + { + "description": "Now you enter the supermarket through the main door. The clothing store is on your left.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:00:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The clothing store was on your right.", + "B": "The clothing store was behind you.", + "C": "The clothing store was on your left-hand side.", + "D": "The clothing store was directly in front of you, across the aisle from where you were standing.", + "E": "The clothing store was on your right-hand side, just a few steps away along the same aisle." + }, + "answer_label": "C", + "answer_text": "The clothing store was on your left-hand side.", + "q_id": "egolife_A1_D5_014_8_T1", + "correct_answer": "The clothing store was on your left-hand side.", + "original_options": { + "A": "The clothing store was on your right.", + "B": "The clothing store was behind you.", + "C": "The clothing store was on your left-hand side.", + "D": "The clothing store was directly in front of you, across the aisle from where you were standing.", + "E": "The clothing store was on your right-hand side, just a few steps away along the same aisle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014-mcq_fixed_6", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the clothing store in relation to the entrance of the supermarket where I entered?", + "answer": "The clothing store is on your right hand side.", + "question_chinese": "此时此刻，超市的服装店在我的哪个方位？", + "answer_chinese": "服装店在你的右手边。", + "evidence": [ + { + "description": "You are now walking towards the supermarket entrance from inside the supermarket. The clothing store is on your right hand side.", + "timestamps": { + "start": "00:08:04", + "end": "00:08:07" + } + } + ], + "question_timestamps": "00:08:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The clothing store is on your left.", + "B": "The clothing store is straight ahead.", + "C": "The clothing store is behind the entrance area, positioned just to the left as you walk in.", + "D": "The clothing store is on your right hand side.", + "E": "The clothing store is straight ahead from the entrance, located near the back after you pass several aisles." + }, + "answer_label": "D", + "answer_text": "The clothing store is on your right hand side.", + "q_id": "egolife_A1_D5_014-mcq_fixed_6", + "correct_answer": "The clothing store is on your right hand side.", + "original_options": { + "A": "The clothing store is on your left.", + "B": "The clothing store is straight ahead.", + "C": "The clothing store is behind the entrance area, positioned just to the left as you walk in.", + "D": "The clothing store is on your right hand side.", + "E": "The clothing store is straight ahead from the entrance, located near the back after you pass several aisles." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_6_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, which direction is the supermarket’s clothing store from me?", + "answer": "The clothing store was on your right-hand side.", + "question_chinese": "此时此刻，超市的服装店在我的哪个方位？", + "answer_chinese": "服装店在你的右手边。", + "evidence": [ + { + "description": "You are now walking towards the supermarket entrance from inside the supermarket. The clothing store is on your right hand side.", + "timestamps": { + "start": "00:08:04", + "end": "00:08:07" + } + } + ], + "question_timestamps": "00:08:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The clothing store was on your left.", + "B": "The clothing store was behind you.", + "C": "The clothing store was directly in front of you, across the aisle from where you were standing.", + "D": "The clothing store was on your left-hand side, a short distance away along the same aisle.", + "E": "The clothing store was on your right-hand side." + }, + "answer_label": "E", + "answer_text": "The clothing store was on your right-hand side.", + "q_id": "egolife_A1_D5_014_6_T1", + "correct_answer": "The clothing store was on your right-hand side.", + "original_options": { + "A": "The clothing store was on your left.", + "B": "The clothing store was behind you.", + "C": "The clothing store was directly in front of you, across the aisle from where you were standing.", + "D": "The clothing store was on your left-hand side, a short distance away along the same aisle.", + "E": "The clothing store was on your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_6_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, which direction is the supermarket’s clothing store from me?", + "answer": "The clothing store was directly in front of you.", + "question_chinese": "此时此刻，超市的服装店在我的哪个方位？", + "answer_chinese": "服装店在你的正前方。", + "evidence": [ + { + "description": "You are now walking towards the supermarket entrance from inside the supermarket. The clothing store is on your right hand side.", + "timestamps": { + "start": "00:08:04", + "end": "00:08:07" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The clothing store was behind you.", + "B": "The clothing store was to your left.", + "C": "The clothing store was on your right-hand side, just off the aisle you were facing at the time.", + "D": "The clothing store was to your left, down the aisle you had just walked through moments earlier.", + "E": "The clothing store was directly in front of you." + }, + "answer_label": "E", + "answer_text": "The clothing store was directly in front of you.", + "q_id": "egolife_A1_D5_014_6_T2", + "correct_answer": "The clothing store was directly in front of you.", + "original_options": { + "A": "The clothing store was behind you.", + "B": "The clothing store was to your left.", + "C": "The clothing store was on your right-hand side, just off the aisle you were facing at the time.", + "D": "The clothing store was to your left, down the aisle you had just walked through moments earlier.", + "E": "The clothing store was directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_8_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, which direction is the supermarket’s clothing store from me?", + "answer": "The clothing store was behind you.", + "question_chinese": "此时此刻，超市的服装店在我的哪个方位？", + "answer_chinese": "服装店在你的后面边。", + "evidence": [ + { + "description": "Now you enter the supermarket through the main door. The clothing store is on your left.", + "timestamps": { + "start": "00:00:50", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:08:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The clothing store was in front of you.", + "B": "The clothing store was to your left.", + "C": "The clothing store was behind you.", + "D": "The clothing store was directly in front of you, positioned at the end of the aisle you were facing.", + "E": "The clothing store was to your right-hand side, across the aisle from where you were standing." + }, + "answer_label": "C", + "answer_text": "The clothing store was behind you.", + "q_id": "egolife_A1_D5_014_8_T2", + "correct_answer": "The clothing store was behind you.", + "original_options": { + "A": "The clothing store was in front of you.", + "B": "The clothing store was to your left.", + "C": "The clothing store was behind you.", + "D": "The clothing store was directly in front of you, positioned at the end of the aisle you were facing.", + "E": "The clothing store was to your right-hand side, across the aisle from where you were standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014-mcq_fixed_7", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Now, is there an elevator on my right side?", + "answer": "There is an elevator on your right side.", + "question_chinese": "现在，我的右手边有电梯吗？", + "answer_chinese": "你的右手边没有电梯。", + "evidence": [ + { + "description": "The elevator going downstairs is on your right side.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:44" + } + } + ], + "question_timestamps": "00:03:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "There is no elevator on your right side.", + "B": "There is an elevator on your right side." + }, + "answer_label": "B", + "answer_text": "There is an elevator on your right side.", + "q_id": "egolife_A1_D5_014-mcq_fixed_7", + "correct_answer": "There is an elevator on your right side.", + "original_options": { + "A": "There is no elevator on your right side.", + "B": "There is an elevator on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014-mcq_fixed_8", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Now, is there an elevator on my right side?", + "answer": "There is no elevator on your right side.", + "question_chinese": "现在，我的右手边有电梯吗？", + "answer_chinese": "你的右手边没有电梯", + "evidence": [ + { + "description": "There is no elevator around you.", + "timestamps": { + "start": "00:06:58", + "end": "00:07:00" + } + } + ], + "question_timestamps": "00:07:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "There is no elevator on your right side.", + "B": "There is an elevator on your right side." + }, + "answer_label": "A", + "answer_text": "There is no elevator on your right side.", + "q_id": "egolife_A1_D5_014-mcq_fixed_8", + "correct_answer": "There is no elevator on your right side.", + "original_options": { + "A": "There is no elevator on your right side.", + "B": "There is an elevator on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_7", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many bottles of drinks are in the shopping cart now?", + "answer": "There are no drinks in the shopping cart now.", + "question_chinese": "现在购物车里面有几瓶饮料？", + "answer_chinese": "现在购物车里面没有饮料", + "evidence": [ + { + "description": "There are no drinks in the shopping cart yet.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:12" + } + } + ], + "question_timestamps": "00:09:23", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are no drinks in the shopping cart now.", + "B": "There is one drink in the cart.", + "C": "There are two bottles of drinks now.", + "D": "There are three bottles of drinks in the shopping cart now, sitting together near the front.", + "E": "There are four bottles of drinks in the shopping cart now, lined up along the side of the basket." + }, + "answer_label": "A", + "answer_text": "There are no drinks in the shopping cart now.", + "q_id": "egolife_A1_D5_014_7", + "correct_answer": "There are no drinks in the shopping cart now.", + "original_options": { + "A": "There are no drinks in the shopping cart now.", + "B": "There is one drink in the cart.", + "C": "There are two bottles of drinks now.", + "D": "There are three bottles of drinks in the shopping cart now, sitting together near the front.", + "E": "There are four bottles of drinks in the shopping cart now, lined up along the side of the basket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_014_8", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many bottles of drinks are in the shopping cart now?", + "answer": "There are now four bottles of drinks in the shopping cart.", + "question_chinese": "现在购物车里面有几瓶饮料？", + "answer_chinese": "现在购物车里有4瓶饮料。", + "evidence": [ + { + "description": "There are no drinks in the shopping cart yet.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:12" + } + }, + { + "description": "Your friend has put 4 bottles of drinks in the shopping cart.", + "timestamps": { + "start": "00:10:48", + "end": "00:10:51" + } + } + ], + "question_timestamps": "00:11:05", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are three bottles of drinks now.", + "B": "There are now four bottles of drinks in the shopping cart.", + "C": "There are five bottles of drinks now.", + "D": "There are two bottles of drinks in the shopping cart now, placed next to each other.", + "E": "There is one bottle of drink in the shopping cart now, with the rest of the cart empty." + }, + "answer_label": "B", + "answer_text": "There are now four bottles of drinks in the shopping cart.", + "q_id": "egolife_A1_D5_014_8", + "correct_answer": "There are now four bottles of drinks in the shopping cart.", + "original_options": { + "A": "There are three bottles of drinks now.", + "B": "There are now four bottles of drinks in the shopping cart.", + "C": "There are five bottles of drinks now.", + "D": "There are two bottles of drinks in the shopping cart now, placed next to each other.", + "E": "There is one bottle of drink in the shopping cart now, with the rest of the cart empty." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D5_015-mcq.json b/QAs/egolife_A1_D5_015-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c7c4bd3916677afe2479a6f343320c6a73f6b11c --- /dev/null +++ b/QAs/egolife_A1_D5_015-mcq.json @@ -0,0 +1,807 @@ +[ + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015-mcq_fixed_8", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "So far, what has been the course of my movement?", + "answer": "You first turned right and then walked backward, and then you reached the elevator on the upper floor. You stayed here for a while.", + "question_chinese": "到目前为止，我的运动轨迹是怎样的？", + "answer_chinese": "你先右转再倒着走路，然后走到了上楼的电梯旁，你在这里逗留了会。", + "evidence": [ + { + "description": "You turned 90 degrees counterclockwise at the shelf with the cola on it, and then walked backward.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "You reached the elevator on the second floor.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + }, + { + "description": "You stayed by the elevator for a while.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:40" + } + } + ], + "question_timestamps": "00:01:41", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You first turned right and then walked backward, and then you reached the elevator on the upper floor. You stayed here for a while.", + "B": "You turned left, then walked backward.", + "C": "You turned right, then walked forward.", + "D": "You first turned left and walked backward, then reached the elevator on the upper floor and waited there.", + "E": "You turned right, walked backward, and reached the elevator on the lower floor, staying there for a while." + }, + "answer_label": "A", + "answer_text": "You first turned right and then walked backward, and then you reached the elevator on the upper floor. You stayed here for a while.", + "q_id": "egolife_A1_D5_015-mcq_fixed_8", + "correct_answer": "You first turned right and then walked backward, and then you reached the elevator on the upper floor. You stayed here for a while.", + "original_options": { + "A": "You first turned right and then walked backward, and then you reached the elevator on the upper floor. You stayed here for a while.", + "B": "You turned left, then walked backward.", + "C": "You turned right, then walked forward.", + "D": "You first turned left and walked backward, then reached the elevator on the upper floor and waited there.", + "E": "You turned right, walked backward, and reached the elevator on the lower floor, staying there for a while." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015-mcq_fixed_9", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "So far, what has been the course of my movement?", + "answer": "First, you turn right and walk backwards. Then you reach the elevator leading to the upper floor. You and your friend stay here for a while, and then take the elevator to the second floor.", + "question_chinese": "到目前为止，我的运动轨迹是怎样的？", + "answer_chinese": "你先右转再倒着走路，然后走到了上楼的电梯旁，你和你的朋友在这里逗留了会，然后乘坐去二楼的电梯。", + "evidence": [ + { + "description": "You turned 90 degrees counterclockwise at the shelf with the cola on it, and then walked backward.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "You reached the elevator on the second floor.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:20" + } + }, + { + "description": "You stayed by the elevator for a while.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:40" + } + }, + { + "description": "You are currently on your way to the second floor by elevator.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:03" + } + } + ], + "question_timestamps": "00:02:05", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turned left and walked backwards.", + "B": "First, you turn right and walk backwards. Then you reach the elevator leading to the upper floor. You and your friend stay here for a while, and then take the elevator to the second floor.", + "C": "You turned right and walked forwards.", + "D": "You turned right and walked backward to the elevator, stayed briefly, then walked away without taking it up.", + "E": "You turned right and walked backward, reached the elevator, stayed with your friend, then took it to the first floor." + }, + "answer_label": "B", + "answer_text": "First, you turn right and walk backwards. Then you reach the elevator leading to the upper floor. You and your friend stay here for a while, and then take the elevator to the second floor.", + "q_id": "egolife_A1_D5_015-mcq_fixed_9", + "correct_answer": "First, you turn right and walk backwards. Then you reach the elevator leading to the upper floor. You and your friend stay here for a while, and then take the elevator to the second floor.", + "original_options": { + "A": "You turned left and walked backwards.", + "B": "First, you turn right and walk backwards. Then you reach the elevator leading to the upper floor. You and your friend stay here for a while, and then take the elevator to the second floor.", + "C": "You turned right and walked forwards.", + "D": "You turned right and walked backward to the elevator, stayed briefly, then walked away without taking it up.", + "E": "You turned right and walked backward, reached the elevator, stayed with your friend, then took it to the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_3", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What fruits are in front of me to the right when I just took the elevator to the second floor of the supermarket?", + "answer": "There were oranges and bananas in your right front.", + "question_chinese": "我刚坐电梯到达超市二楼的时候我的右前方有什么水果？", + "answer_chinese": "你的右前方有橙子和香蕉。", + "evidence": [ + { + "description": "When you turn right when going upstairs, there are oranges and bananas in front of you.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:06" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There were apples and grapes there.", + "B": "There were oranges and bananas in your right front.", + "C": "There were lemons and limes there.", + "D": "There were strawberries and peaches in your right front, displayed together on the stand.", + "E": "There were cherries and pears in your right front when you stepped out of the elevator." + }, + "answer_label": "B", + "answer_text": "There were oranges and bananas in your right front.", + "q_id": "egolife_A1_D5_015_3", + "correct_answer": "There were oranges and bananas in your right front.", + "original_options": { + "A": "There were apples and grapes there.", + "B": "There were oranges and bananas in your right front.", + "C": "There were lemons and limes there.", + "D": "There were strawberries and peaches in your right front, displayed together on the stand.", + "E": "There were cherries and pears in your right front when you stepped out of the elevator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_4", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What fruits are in front of me to the right when I just arrived at the second floor of the supermarket by elevator?", + "answer": "On your right front there are oranges, bananas and blueberries.", + "question_chinese": "我刚坐电梯到达超市二楼的时候我的右前方有什么水果？", + "answer_chinese": "你的右前方有橙子，香蕉和蓝莓。", + "evidence": [ + { + "description": "When you turn right when going upstairs, there are oranges and bananas in front of you.", + "timestamps": { + "start": "00:03:02", + "end": "00:03:06" + } + }, + { + "description": "There are blueberries in front of you.", + "timestamps": { + "start": "00:03:21", + "end": "00:03:24" + } + } + ], + "question_timestamps": "00:03:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On your right front are apples and bananas.", + "B": "On your right front are bananas and strawberries.", + "C": "On your right front there are oranges, bananas and blueberries.", + "D": "On your right front there are oranges, bananas, and blackberries arranged together in the fruit section.", + "E": "On your right front there are oranges, grapes, and strawberries as you arrive on the second floor." + }, + "answer_label": "C", + "answer_text": "On your right front there are oranges, bananas and blueberries.", + "q_id": "egolife_A1_D5_015_4", + "correct_answer": "On your right front there are oranges, bananas and blueberries.", + "original_options": { + "A": "On your right front are apples and bananas.", + "B": "On your right front are bananas and strawberries.", + "C": "On your right front there are oranges, bananas and blueberries.", + "D": "On your right front there are oranges, bananas, and blackberries arranged together in the fruit section.", + "E": "On your right front there are oranges, grapes, and strawberries as you arrive on the second floor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_7", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What foods have I eaten at the supermarket?", + "answer": "You ate dates at the supermarket.", + "question_chinese": "我在超市吃过哪些食物？", + "answer_chinese": "你在超市吃过红枣。", + "evidence": [ + { + "description": "You are eating dates now.", + "timestamps": { + "start": "00:08:26", + "end": "00:08:28" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You ate apples at the supermarket.", + "B": "You ate bananas at the supermarket.", + "C": "You ate dates behind the supermarket, not inside the store during shopping.", + "D": "You ate dates at the supermarket.", + "E": "You ate dates in front of the supermarket, outside the entrance rather than inside." + }, + "answer_label": "D", + "answer_text": "You ate dates at the supermarket.", + "q_id": "egolife_A1_D5_015_7", + "correct_answer": "You ate dates at the supermarket.", + "original_options": { + "A": "You ate apples at the supermarket.", + "B": "You ate bananas at the supermarket.", + "C": "You ate dates behind the supermarket, not inside the store during shopping.", + "D": "You ate dates at the supermarket.", + "E": "You ate dates in front of the supermarket, outside the entrance rather than inside." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_8", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What foods have I eaten at the supermarket?", + "answer": "You ate dates and instant noodles at the supermarket.", + "question_chinese": "我在超市吃过哪些食物？", + "answer_chinese": "你在超市吃过红枣和方便面。", + "evidence": [ + { + "description": "You are eating dates now.", + "timestamps": { + "start": "00:08:26", + "end": "00:08:28" + } + }, + { + "description": "You are eating instant noodles.", + "timestamps": { + "start": "00:09:55", + "end": "00:09:58" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You ate dates at the supermarket.", + "B": "You ate dates and instant noodles at the supermarket.", + "C": "You ate instant noodles at the supermarket.", + "D": "You ate dates and chips at the supermarket, snacking while walking through the aisles.", + "E": "You ate instant noodles and rice at the supermarket, instead of having any dates there." + }, + "answer_label": "B", + "answer_text": "You ate dates and instant noodles at the supermarket.", + "q_id": "egolife_A1_D5_015_8", + "correct_answer": "You ate dates and instant noodles at the supermarket.", + "original_options": { + "A": "You ate dates at the supermarket.", + "B": "You ate dates and instant noodles at the supermarket.", + "C": "You ate instant noodles at the supermarket.", + "D": "You ate dates and chips at the supermarket, snacking while walking through the aisles.", + "E": "You ate instant noodles and rice at the supermarket, instead of having any dates there." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_9", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get durian now?", + "answer": "You can get durian.", + "question_chinese": "我现在能够的到榴莲吗？", + "answer_chinese": "你能够得到榴莲。", + "evidence": [ + { + "description": "Your friend is holding a durian in front of you.", + "timestamps": { + "start": "00:03:42", + "end": "00:03:44" + } + } + ], + "question_timestamps": "00:03:48", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can get durian.", + "B": "You can't reach the durian." + }, + "answer_label": "A", + "answer_text": "You can get durian.", + "q_id": "egolife_A1_D5_015_9", + "correct_answer": "You can get durian.", + "original_options": { + "A": "You can get durian.", + "B": "You can't reach the durian." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_10", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get durian now?", + "answer": "You can't reach the durian.", + "question_chinese": "我现在能够的到榴莲吗？", + "answer_chinese": "你够不到榴莲。", + "evidence": [ + { + "description": "The durian is in the fruit section of the supermarket.", + "timestamps": { + "start": "00:04:26", + "end": "00:04:30" + } + }, + { + "description": "You are now in the snack food section, and there is no durian here.", + "timestamps": { + "start": "00:11:29", + "end": "00:11:30" + } + } + ], + "question_timestamps": "00:11:35", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't reach the durian.", + "B": "You can get durian." + }, + "answer_label": "A", + "answer_text": "You can't reach the durian.", + "q_id": "egolife_A1_D5_015_10", + "correct_answer": "You can't reach the durian.", + "original_options": { + "A": "You can't reach the durian.", + "B": "You can get durian." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_3_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Based on where I am right now, which is closer to me—the watermelon or the Hami melon?", + "answer": "The watermelon is closer to you.", + "question_chinese": "根据我现在的位置，西瓜和哈密瓜哪个离我更近？", + "answer_chinese": "西瓜离你更近。", + "evidence": [ + { + "description": "The watermelon is right in front of you, while the honeydew melon is in your right front.", + "timestamps": { + "start": "00:05:33", + "end": "00:05:35" + } + } + ], + "question_timestamps": "00:05:46", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The Hami melon is closer to you.", + "B": "The watermelon is closer to you." + }, + "answer_label": "B", + "answer_text": "The watermelon is closer to you.", + "q_id": "egolife_A1_D5_015_3_T1", + "correct_answer": "The watermelon is closer to you.", + "original_options": { + "A": "The Hami melon is closer to you.", + "B": "The watermelon is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_3_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Based on where I am right now, which is closer to me—the watermelon or the Hami melon?", + "answer": "The Hami melon is closer to you.", + "question_chinese": "根据我现在的位置，西瓜和哈密瓜哪个离我更近？", + "answer_chinese": "哈密瓜离你更近。", + "evidence": [ + { + "description": "The watermelon is right in front of you, while the honeydew melon is in your right front.", + "timestamps": { + "start": "00:05:33", + "end": "00:05:35" + } + }, + { + "description": "The honeydew melon is on your left side, while the watermelon is on your left rear.", + "timestamps": { + "start": "00:05:49", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:06:01", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The watermelon is closer to you.", + "B": "The Hami melon is closer to you." + }, + "answer_label": "B", + "answer_text": "The Hami melon is closer to you.", + "q_id": "egolife_A1_D5_015_3_T2", + "correct_answer": "The Hami melon is closer to you.", + "original_options": { + "A": "The watermelon is closer to you.", + "B": "The Hami melon is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_1_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is to my right right now?", + "answer": "To your right, there’s a shelf full of plastic storage bags.", + "question_chinese": "现在我的右边有什么？", + "answer_chinese": "你的右边有一货架的保鲜袋。", + "evidence": [ + { + "description": "There is a shelf full of fresh-keeping bags on your right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The shelf is ahead.", + "B": "To your right, there’s a shelf full of plastic storage bags.", + "C": "The shelf is behind you.", + "D": "To your left, there’s a shelf full of plastic storage bags, not on your right.", + "E": "Above you, there’s a shelf full of plastic storage bags rather than to your right." + }, + "answer_label": "B", + "answer_text": "To your right, there’s a shelf full of plastic storage bags.", + "q_id": "egolife_A1_D5_015_1_T2", + "correct_answer": "To your right, there’s a shelf full of plastic storage bags.", + "original_options": { + "A": "The shelf is ahead.", + "B": "To your right, there’s a shelf full of plastic storage bags.", + "C": "The shelf is behind you.", + "D": "To your left, there’s a shelf full of plastic storage bags, not on your right.", + "E": "Above you, there’s a shelf full of plastic storage bags rather than to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_1_T1", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is to my right right now?", + "answer": "To your right, there’s a shelf full of carbonated drinks.", + "question_chinese": "现在我的右边有什么？", + "answer_chinese": "你的右边有一货架的碳酸饮料。", + "evidence": [ + { + "description": "There is a shelf full of carbonated drinks on your right.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A shelf of carbonated drinks is ahead.", + "B": "A shelf of carbonated drinks is behind you.", + "C": "To your left, there’s a shelf full of carbonated drinks instead of to your right.", + "D": "Above you, there’s a shelf full of carbonated drinks rather than to your right.", + "E": "To your right, there’s a shelf full of carbonated drinks." + }, + "answer_label": "E", + "answer_text": "To your right, there’s a shelf full of carbonated drinks.", + "q_id": "egolife_A1_D5_015_1_T1", + "correct_answer": "To your right, there’s a shelf full of carbonated drinks.", + "original_options": { + "A": "A shelf of carbonated drinks is ahead.", + "B": "A shelf of carbonated drinks is behind you.", + "C": "To your left, there’s a shelf full of carbonated drinks instead of to your right.", + "D": "Above you, there’s a shelf full of carbonated drinks rather than to your right.", + "E": "To your right, there’s a shelf full of carbonated drinks." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_2_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are my hands placed right now?", + "answer": "Right now, your hand is on the pile of mung beans, and the pile of mung beans is in front of you.", + "question_chinese": "现在我的手放在哪里？", + "answer_chinese": "现在你的手放在绿豆堆上，绿豆堆在你的正前方。", + "evidence": [ + { + "description": "Now put your hand on the mung bean.", + "timestamps": { + "start": "00:07:12", + "end": "00:07:14" + } + } + ], + "question_timestamps": "00:07:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your hand is behind you.", + "B": "Your hand is to your left.", + "C": "Right now, your hand is on the pile of mung beans, and the pile is to your right.", + "D": "Right now, your hand is on the pile of mung beans, and the pile of mung beans is in front of you.", + "E": "Right now, your hand is on the pile of mung beans, and the pile of mung beans is above you." + }, + "answer_label": "D", + "answer_text": "Right now, your hand is on the pile of mung beans, and the pile of mung beans is in front of you.", + "q_id": "egolife_A1_D5_015_2_T2", + "correct_answer": "Right now, your hand is on the pile of mung beans, and the pile of mung beans is in front of you.", + "original_options": { + "A": "Your hand is behind you.", + "B": "Your hand is to your left.", + "C": "Right now, your hand is on the pile of mung beans, and the pile is to your right.", + "D": "Right now, your hand is on the pile of mung beans, and the pile of mung beans is in front of you.", + "E": "Right now, your hand is on the pile of mung beans, and the pile of mung beans is above you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D5_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_015_2_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are my hands placed right now?", + "answer": "Right now, your hand is on the pile of red dates, and the pile of red dates is in front of you.", + "question_chinese": "现在我的手放在哪里？", + "answer_chinese": "现在你的手放在红枣堆上，红枣堆在你的正前方。", + "evidence": [ + { + "description": "Now put your hand on the pile of red dates.", + "timestamps": { + "start": "00:08:05", + "end": "00:08:07" + } + } + ], + "question_timestamps": "00:08:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your hand is behind you.", + "B": "Right now, your hand is on the pile of red dates, and the pile of red dates is in front of you.", + "C": "Your hand is to your left.", + "D": "Right now, your hand is on the pile of red dates, and the pile is to your right.", + "E": "Right now, your hand is on the pile of red dates, and the pile of red dates is above you." + }, + "answer_label": "B", + "answer_text": "Right now, your hand is on the pile of red dates, and the pile of red dates is in front of you.", + "q_id": "egolife_A1_D5_015_2_T1", + "correct_answer": "Right now, your hand is on the pile of red dates, and the pile of red dates is in front of you.", + "original_options": { + "A": "Your hand is behind you.", + "B": "Right now, your hand is on the pile of red dates, and the pile of red dates is in front of you.", + "C": "Your hand is to your left.", + "D": "Right now, your hand is on the pile of red dates, and the pile is to your right.", + "E": "Right now, your hand is on the pile of red dates, and the pile of red dates is above you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 20 + ] + } +] diff --git a/QAs/egolife_A1_D5_016-mcq.json b/QAs/egolife_A1_D5_016-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d214942d61b1f7eaa2fc0c459c0e07e6349f47b8 --- /dev/null +++ b/QAs/egolife_A1_D5_016-mcq.json @@ -0,0 +1,959 @@ +[ + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016-mcq_fixed_6", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When my friend found the shopping cart, what was around me?", + "answer": "There was soy sauce and instant noodles around you.", + "question_chinese": "当我的朋友找到购物车的时候，我的四周有什么？", + "answer_chinese": "你的四周有酱油和方便面。", + "evidence": [ + { + "description": "In front of you was soy sauce and to your left front was instant noodles.", + "timestamps": { + "start": "00:02:04", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There was vinegar around you.", + "B": "There was soy sauce and instant noodles around you.", + "C": "There were dried noodles around you.", + "D": "There were soy sauce bottles and bags of rice around you when your friend located the cart.", + "E": "There was instant noodles and canned beans around you while your friend found the shopping cart." + }, + "answer_label": "B", + "answer_text": "There was soy sauce and instant noodles around you.", + "q_id": "egolife_A1_D5_016-mcq_fixed_6", + "correct_answer": "There was soy sauce and instant noodles around you.", + "original_options": { + "A": "There was vinegar around you.", + "B": "There was soy sauce and instant noodles around you.", + "C": "There were dried noodles around you.", + "D": "There were soy sauce bottles and bags of rice around you when your friend located the cart.", + "E": "There was instant noodles and canned beans around you while your friend found the shopping cart." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016-mcq_fixed_7", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When my friend found the shopping cart, what was around me?", + "answer": "Around you were soy sauce, instant noodles and dried noodles.", + "question_chinese": "当我的朋友找到购物车的时候，我的四周有什么？", + "answer_chinese": "你的四周有酱油，方便面和挂面。", + "evidence": [ + { + "description": "In front of you was soy sauce and to your left front was instant noodles.", + "timestamps": { + "start": "00:02:04", + "end": "00:02:05" + } + }, + { + "description": "On your left side, there were a lot of dried noodles.", + "timestamps": { + "start": "00:02:21", + "end": "00:02:22" + } + } + ], + "question_timestamps": "00:02:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Around you were soy sauce and instant noodles.", + "B": "Around you were instant noodles and dried noodles.", + "C": "Around you were soy sauce, dried noodles, and bags of rice as your friend discovered the shopping cart.", + "D": "Around you were soy sauce, instant noodles, and ramen seasoning packets when your friend finally found the cart.", + "E": "Around you were soy sauce, instant noodles and dried noodles." + }, + "answer_label": "E", + "answer_text": "Around you were soy sauce, instant noodles and dried noodles.", + "q_id": "egolife_A1_D5_016-mcq_fixed_7", + "correct_answer": "Around you were soy sauce, instant noodles and dried noodles.", + "original_options": { + "A": "Around you were soy sauce and instant noodles.", + "B": "Around you were instant noodles and dried noodles.", + "C": "Around you were soy sauce, dried noodles, and bags of rice as your friend discovered the shopping cart.", + "D": "Around you were soy sauce, instant noodles, and ramen seasoning packets when your friend finally found the cart.", + "E": "Around you were soy sauce, instant noodles and dried noodles." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_7", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was I when I first saw Wangzai Milk?", + "answer": "You were in front of the shelf with the socket and to the left of the shelf with the large bottle of mineral water.", + "question_chinese": "当我第一次看到旺仔牛奶的时候，我在什么位置？", + "answer_chinese": "你在放有插座货架的前面，放有大瓶矿泉水货架的左边。", + "evidence": [ + { + "description": "There was a large bottle of mineral water on the shelf on your right.", + "timestamps": { + "start": "00:08:34", + "end": "00:08:35" + } + }, + { + "description": "There was a socket on the shelf in front of you.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:36" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind the socket shelf.", + "B": "Right of the mineral water shelf.", + "C": "You were in front of the shelf with the socket and to the left of the shelf with the large bottle of mineral water.", + "D": "In front of the socket shelf and to the right of the large mineral water bottle shelf.", + "E": "Behind the socket shelf and to the right of the shelf with the large bottle of mineral water." + }, + "answer_label": "C", + "answer_text": "You were in front of the shelf with the socket and to the left of the shelf with the large bottle of mineral water.", + "q_id": "egolife_A1_D5_016_7", + "correct_answer": "You were in front of the shelf with the socket and to the left of the shelf with the large bottle of mineral water.", + "original_options": { + "A": "Behind the socket shelf.", + "B": "Right of the mineral water shelf.", + "C": "You were in front of the shelf with the socket and to the left of the shelf with the large bottle of mineral water.", + "D": "In front of the socket shelf and to the right of the large mineral water bottle shelf.", + "E": "Behind the socket shelf and to the right of the shelf with the large bottle of mineral water." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_8", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was I when I first saw Wangzai Milk?", + "answer": "You were in front of the shelf with the socket, to the left of the shelf with the large bottle of mineral water, and to the right of the shelf with the small bottle of mineral water.", + "question_chinese": "当我第一次看到旺仔牛奶的时候，我在什么位置？", + "answer_chinese": "你在放有插座货架的前面，放有大瓶矿泉水货架的左边，以及放有小瓶矿泉水货架的右边。", + "evidence": [ + { + "description": "There was a large bottle of mineral water on the shelf on your right.", + "timestamps": { + "start": "00:08:34", + "end": "00:08:35" + } + }, + { + "description": "There was a socket on the shelf in front of you.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:36" + } + }, + { + "description": "From your perspective, relative to seeing the Wangzai Milk, you have rotated 90 degrees counterclockwise. There is a small bottle of mineral water directly in front of you.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:02" + } + } + ], + "question_timestamps": "00:09:08", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You were in front of the shelf with the socket, to the left of the shelf with the large bottle of mineral water, and to the right of the shelf with the small bottle of mineral water.", + "B": "Behind the socket shelf.", + "C": "Left of the small mineral water.", + "D": "In front of the socket shelf, right of the large mineral water, and left of the small bottle.", + "E": "Behind the socket shelf, left of the large mineral water bottle, and right of the small bottle." + }, + "answer_label": "A", + "answer_text": "You were in front of the shelf with the socket, to the left of the shelf with the large bottle of mineral water, and to the right of the shelf with the small bottle of mineral water.", + "q_id": "egolife_A1_D5_016_8", + "correct_answer": "You were in front of the shelf with the socket, to the left of the shelf with the large bottle of mineral water, and to the right of the shelf with the small bottle of mineral water.", + "original_options": { + "A": "You were in front of the shelf with the socket, to the left of the shelf with the large bottle of mineral water, and to the right of the shelf with the small bottle of mineral water.", + "B": "Behind the socket shelf.", + "C": "Left of the small mineral water.", + "D": "In front of the socket shelf, right of the large mineral water, and left of the small bottle.", + "E": "Behind the socket shelf, left of the large mineral water bottle, and right of the small bottle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_1_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "After my friend in blue and I found the shopping cart, how many more items did we add?", + "answer": "There are already two items in the shopping cart, and the shopping cart is by you.", + "question_chinese": "我和我穿蓝色衣服的朋友找到购物车后，又新放了多少个商品？", + "answer_chinese": "购物车里面已经放了两件商品，购物车在你。", + "evidence": [ + { + "description": "You are putting a bottle of milk into the shopping cart.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:29" + } + }, + { + "description": "Your friend in red puts a box of plums into the shopping cart.", + "timestamps": { + "start": "00:02:41", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:02:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero items added.", + "B": "Four items added.", + "C": "There are already zero items in the shopping cart, and the shopping cart is by you.", + "D": "There are already four items in the shopping cart, and the shopping cart is by you.", + "E": "There are already two items in the shopping cart, and the shopping cart is by you." + }, + "answer_label": "E", + "answer_text": "There are already two items in the shopping cart, and the shopping cart is by you.", + "q_id": "egolife_A1_D5_016_1_T1", + "correct_answer": "There are already two items in the shopping cart, and the shopping cart is by you.", + "original_options": { + "A": "Zero items added.", + "B": "Four items added.", + "C": "There are already zero items in the shopping cart, and the shopping cart is by you.", + "D": "There are already four items in the shopping cart, and the shopping cart is by you.", + "E": "There are already two items in the shopping cart, and the shopping cart is by you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_3", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many new items did my friend in the blue shirt and I add after we found the shopping cart?", + "answer": "There are already two items in the shopping cart.", + "question_chinese": "我和我穿蓝色衣服的朋友找到购物车后，又新放了多少个商品？", + "answer_chinese": "购物车里面已经放了两件商品。", + "evidence": [ + { + "description": "You are putting a bottle of milk into the shopping cart.", + "timestamps": { + "start": "00:02:27", + "end": "00:02:28" + } + }, + { + "description": "Your friend in red puts a box of plums into the shopping cart.", + "timestamps": { + "start": "00:04:42", + "end": "00:04:44" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Zero items.", + "B": "One item.", + "C": "There were already one item in the cart, so we added one more item afterward.", + "D": "There are already two items in the shopping cart.", + "E": "There were already five items in the cart, so we added five more items after finding it." + }, + "answer_label": "D", + "answer_text": "There are already two items in the shopping cart.", + "q_id": "egolife_A1_D5_016_3", + "correct_answer": "There are already two items in the shopping cart.", + "original_options": { + "A": "Zero items.", + "B": "One item.", + "C": "There were already one item in the cart, so we added one more item afterward.", + "D": "There are already two items in the shopping cart.", + "E": "There were already five items in the cart, so we added five more items after finding it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_1_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "After my friend in blue and I found the shopping cart, how many more items did we add?", + "answer": "There are already three items in the shopping cart, and the shopping cart is in front of you.", + "question_chinese": "我和我穿蓝色衣服的朋友找到购物车后，又新放了多少个商品？", + "answer_chinese": "购物车里面已经放了三件商品，购物车在你前面。", + "evidence": [ + { + "description": "You are putting a bottle of milk into the shopping cart.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:28" + } + }, + { + "description": "Your friend in red puts a box of plums into the shopping cart.", + "timestamps": { + "start": "00:02:41", + "end": "00:02:44" + } + }, + { + "description": "Your friend handed you a box of pineapples, and after taking it, you placed it in the shopping cart.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:06" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are already three items in the shopping cart, and the shopping cart is in front of you.", + "B": "One item added.", + "C": "Five items added.", + "D": "There is already one item in the shopping cart, and the shopping cart is in front of you.", + "E": "There are already five items in the shopping cart, and the shopping cart is in front of you." + }, + "answer_label": "A", + "answer_text": "There are already three items in the shopping cart, and the shopping cart is in front of you.", + "q_id": "egolife_A1_D5_016_1_T2", + "correct_answer": "There are already three items in the shopping cart, and the shopping cart is in front of you.", + "original_options": { + "A": "There are already three items in the shopping cart, and the shopping cart is in front of you.", + "B": "One item added.", + "C": "Five items added.", + "D": "There is already one item in the shopping cart, and the shopping cart is in front of you.", + "E": "There are already five items in the shopping cart, and the shopping cart is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_4", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many new items did my friend in the blue shirt and I add after we found the shopping cart?", + "answer": "There are already two items in the shopping cart.", + "question_chinese": "我和我穿蓝色衣服的朋友找到购物车后，又新放了多少个商品？", + "answer_chinese": "购物车里面已经放了三件商品。", + "evidence": [ + { + "description": "You are putting a bottle of milk into the shopping cart.", + "timestamps": { + "start": "00:02:27", + "end": "00:02:28" + } + }, + { + "description": "Your friend in red puts a box of plums into the shopping cart.", + "timestamps": { + "start": "00:04:42", + "end": "00:04:44" + } + }, + { + "description": "Your friend handed you a box of pineapples, and after taking it, you placed it in the shopping cart.", + "timestamps": { + "start": "00:05:04", + "end": "00:05:05" + } + } + ], + "question_timestamps": "00:05:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are already two items in the shopping cart.", + "B": "Zero items.", + "C": "One item.", + "D": "There were already one item in the cart, so we added one more item afterward.", + "E": "There were already five items in the cart, so we added five more items after finding it." + }, + "answer_label": "A", + "answer_text": "There are already two items in the shopping cart.", + "q_id": "egolife_A1_D5_016_4", + "correct_answer": "There are already two items in the shopping cart.", + "original_options": { + "A": "There are already two items in the shopping cart.", + "B": "Zero items.", + "C": "One item.", + "D": "There were already one item in the cart, so we added one more item afterward.", + "E": "There were already five items in the cart, so we added five more items after finding it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_4_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I find my friend?", + "answer": "Turn around, then turn left. Your friend is to your front-right.", + "question_chinese": "我怎么找到的我的朋友？", + "answer_chinese": "你向后转，然向左转，你的朋友在你的右前方。", + "evidence": [ + { + "description": "You turn around and turn left.", + "timestamps": { + "start": "00:01:58", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:02:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around, then turn right.", + "B": "Turn left, then turn around.", + "C": "Turn around, then turn left, but your friend ends up to your front-left instead of front-right.", + "D": "Turn around, then turn left, and you find your friend behind you on the right, not ahead.", + "E": "Turn around, then turn left. Your friend is to your front-right." + }, + "answer_label": "E", + "answer_text": "Turn around, then turn left. Your friend is to your front-right.", + "q_id": "egolife_A1_D5_016_4_T1", + "correct_answer": "Turn around, then turn left. Your friend is to your front-right.", + "original_options": { + "A": "Turn around, then turn right.", + "B": "Turn left, then turn around.", + "C": "Turn around, then turn left, but your friend ends up to your front-left instead of front-right.", + "D": "Turn around, then turn left, and you find your friend behind you on the right, not ahead.", + "E": "Turn around, then turn left. Your friend is to your front-right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_4_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I find my friend?", + "answer": "Turn around, walk straight, then turn left, and continue straight. Your friend is to your front-right.", + "question_chinese": "我怎么找到的我的朋友？", + "answer_chinese": "你向后转，然后直走，再向左转，接着直走，你的朋友在你的右前方。", + "evidence": [ + { + "description": "You turn around and walk straight.", + "timestamps": { + "start": "00:10:57", + "end": "00:11:05" + } + }, + { + "description": "You turn left beside the shelf with Bai Shui Shan mineral water and walk straight. Your friend appears in front of you on the right.", + "timestamps": { + "start": "00:11:09", + "end": "00:11:15" + } + } + ], + "question_timestamps": "00:11:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn around, then turn left.", + "B": "Turn around, then turn right.", + "C": "Turn around, walk straight, then turn right and continue straight; your friend is to your front-right.", + "D": "Turn around, walk straight, then turn left, and continue straight. Your friend is to your front-right.", + "E": "Turn around, walk straight, then turn left and continue straight, but your friend is to your front-left." + }, + "answer_label": "D", + "answer_text": "Turn around, walk straight, then turn left, and continue straight. Your friend is to your front-right.", + "q_id": "egolife_A1_D5_016_4_T2", + "correct_answer": "Turn around, walk straight, then turn left, and continue straight. Your friend is to your front-right.", + "original_options": { + "A": "Turn around, then turn left.", + "B": "Turn around, then turn right.", + "C": "Turn around, walk straight, then turn right and continue straight; your friend is to your front-right.", + "D": "Turn around, walk straight, then turn left, and continue straight. Your friend is to your front-right.", + "E": "Turn around, walk straight, then turn left and continue straight, but your friend is to your front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_9", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I find my friend after I get the Wangzai milk?", + "answer": "You turn around, then walk straight, then turn left, and walk straight again. Your friend is in front of you on the right.", + "question_chinese": "我拿到旺仔牛奶后，我怎么找到的我的朋友？", + "answer_chinese": "你向后转，然后直走，再向左转，接着直走，你的朋友在你的右前方。", + "evidence": [ + { + "description": "You turn around and walk straight.", + "timestamps": { + "start": "00:10:59", + "end": "00:11:05" + } + }, + { + "description": "You turn left beside the shelf with Bai Shui Shan mineral water and walk straight. Your friend appears in front of you on the right.", + "timestamps": { + "start": "00:11:09", + "end": "00:11:13" + } + } + ], + "question_timestamps": "00:11:17", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn around, then turn right.", + "B": "Turn around, then stop.", + "C": "You turn around, walk straight, then turn left and walk straight again, but your friend is in front-left.", + "D": "You turn around, walk straight, then turn right and walk straight again; your friend is in front of you on the right.", + "E": "You turn around, then walk straight, then turn left, and walk straight again. Your friend is in front of you on the right." + }, + "answer_label": "E", + "answer_text": "You turn around, then walk straight, then turn left, and walk straight again. Your friend is in front of you on the right.", + "q_id": "egolife_A1_D5_016_9", + "correct_answer": "You turn around, then walk straight, then turn left, and walk straight again. Your friend is in front of you on the right.", + "original_options": { + "A": "Turn around, then turn right.", + "B": "Turn around, then stop.", + "C": "You turn around, walk straight, then turn left and walk straight again, but your friend is in front-left.", + "D": "You turn around, walk straight, then turn right and walk straight again; your friend is in front of you on the right.", + "E": "You turn around, then walk straight, then turn left, and walk straight again. Your friend is in front of you on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 22 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_10", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I find my friend after I got the Wangzai milk?", + "answer": "You turn around, walk straight, then turn left, and walk straight again. Your friend is in front of you on the right.", + "question_chinese": "我拿到旺仔牛奶后，我怎么找到的我的朋友？", + "answer_chinese": "你向后转，然后直走，再向左转，接着直走，你的朋友在你的右前方。", + "evidence": [ + { + "description": "You turn around and walk straight.", + "timestamps": { + "start": "00:10:59", + "end": "00:11:05" + } + }, + { + "description": "You turn left beside the shelf with Bai Shui Shan mineral water and walk straight. Your friend appears in front of you on the right.", + "timestamps": { + "start": "00:11:09", + "end": "00:11:13" + } + } + ], + "question_timestamps": "00:11:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left, then walk straight.", + "B": "Turn around, then turn right.", + "C": "You turn around, walk straight, then turn right and walk straight again; your friend is in front on the left.", + "D": "You turn around, walk straight, then turn left, and walk straight again. Your friend is in front of you on the right.", + "E": "You turn around, walk straight, then turn left and walk back again; your friend ends up behind you on the right." + }, + "answer_label": "D", + "answer_text": "You turn around, walk straight, then turn left, and walk straight again. Your friend is in front of you on the right.", + "q_id": "egolife_A1_D5_016_10", + "correct_answer": "You turn around, walk straight, then turn left, and walk straight again. Your friend is in front of you on the right.", + "original_options": { + "A": "Turn left, then walk straight.", + "B": "Turn around, then turn right.", + "C": "You turn around, walk straight, then turn right and walk straight again; your friend is in front on the left.", + "D": "You turn around, walk straight, then turn left, and walk straight again. Your friend is in front of you on the right.", + "E": "You turn around, walk straight, then turn left and walk back again; your friend ends up behind you on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_3_T1", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many friends are around me?", + "answer": "Zero. You don’t have any friends next to you; your friend is in front of you, a few meters away.", + "question_chinese": "我身边有多少个朋友？", + "answer_chinese": "零个，你身边没有朋友，你的朋友在你前面，有几米远。", + "evidence": [ + { + "description": "There are zero. You have no friends around you.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One friend is next to you.", + "B": "Zero. You don’t have any friends next to you; your friend is in front of you, a few meters away.", + "C": "Two friends are around you.", + "D": "One. You have a friend standing right beside you, close enough to be considered around you.", + "E": "Three. You are surrounded by three friends nearby, with at least one positioned close at your side." + }, + "answer_label": "B", + "answer_text": "Zero. You don’t have any friends next to you; your friend is in front of you, a few meters away.", + "q_id": "egolife_A1_D5_016_3_T1", + "correct_answer": "Zero. You don’t have any friends next to you; your friend is in front of you, a few meters away.", + "original_options": { + "A": "One friend is next to you.", + "B": "Zero. You don’t have any friends next to you; your friend is in front of you, a few meters away.", + "C": "Two friends are around you.", + "D": "One. You have a friend standing right beside you, close enough to be considered around you.", + "E": "Three. You are surrounded by three friends nearby, with at least one positioned close at your side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016_3_T2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many friends are around me?", + "answer": "You have a friend next to you, and your friend is to your left.", + "question_chinese": "我身边有多少个朋友？", + "answer_chinese": "你身边有一个朋友，你的朋友在你的左边。", + "evidence": [ + { + "description": "There is a friend beside you, and he is wearing blue clothes.", + "timestamps": { + "start": "00:02:23", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:02:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You have no friends around you.", + "B": "Your friend is to your right.", + "C": "You have a friend next to you, and your friend is to your left.", + "D": "You have a friend next to you, but your friend is directly in front of you rather than to the left.", + "E": "You have a friend next to you, and your friend is behind you, not positioned on your left side." + }, + "answer_label": "C", + "answer_text": "You have a friend next to you, and your friend is to your left.", + "q_id": "egolife_A1_D5_016_3_T2", + "correct_answer": "You have a friend next to you, and your friend is to your left.", + "original_options": { + "A": "You have no friends around you.", + "B": "Your friend is to your right.", + "C": "You have a friend next to you, and your friend is to your left.", + "D": "You have a friend next to you, but your friend is directly in front of you rather than to the left.", + "E": "You have a friend next to you, and your friend is behind you, not positioned on your left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016-mcq_fixed_8", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "According to my current position, can I reach the laundry detergent?", + "answer": "You can reach the laundry detergent.", + "question_chinese": "根据我现在的位置，我可以够得到洗衣液吗？", + "answer_chinese": "你够得到洗衣液。", + "evidence": [ + { + "description": "After you get off the elevator, there is laundry detergent in front of you.", + "timestamps": { + "start": "00:06:58", + "end": "00:06:59" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can reach the laundry detergent.", + "B": "You cannot reach the laundry detergent." + }, + "answer_label": "A", + "answer_text": "You can reach the laundry detergent.", + "q_id": "egolife_A1_D5_016-mcq_fixed_8", + "correct_answer": "You can reach the laundry detergent.", + "original_options": { + "A": "You can reach the laundry detergent.", + "B": "You cannot reach the laundry detergent." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D5_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_016-mcq_fixed_9", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "According to my current position, can I reach the laundry detergent?", + "answer": "You can't reach the laundry detergent.", + "question_chinese": "根据我现在的位置，我可以够得到洗衣液吗？", + "answer_chinese": "你够不到洗衣液。", + "evidence": [ + { + "description": "There is no laundry detergent around you.", + "timestamps": { + "start": "00:07:58", + "end": "00:08:00" + } + } + ], + "question_timestamps": "00:08:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can reach the laundry detergent.", + "B": "You can't reach the laundry detergent." + }, + "answer_label": "B", + "answer_text": "You can't reach the laundry detergent.", + "q_id": "egolife_A1_D5_016-mcq_fixed_9", + "correct_answer": "You can't reach the laundry detergent.", + "original_options": { + "A": "You can reach the laundry detergent.", + "B": "You can't reach the laundry detergent." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/egolife_A1_D5_017-mcq.json b/QAs/egolife_A1_D5_017-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4c5de2b7ca529959010761c392e3222a9e7636ea --- /dev/null +++ b/QAs/egolife_A1_D5_017-mcq.json @@ -0,0 +1,570 @@ +[ + { + "video_uid": "egolife_A1_D5_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_017-mcq_fixed_2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "So far, what has been my movement pattern?", + "answer": "You turn right on the spot, then proceed straight, and finally walk around the cashier counter.", + "question_chinese": "到目前为止，我的运动轨迹是怎样的？", + "answer_chinese": "你原地右转，然后直行，最后在收银台旁边踱步。", + "evidence": [ + { + "description": "You turn right on the spot, then proceed straight along with your friend.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:30" + } + }, + { + "description": "You walk around the cashier counter.", + "timestamps": { + "start": "00:00:54", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:01", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turn left, then go straight.", + "B": "You go straight, then turn right.", + "C": "You turn right in place, walk straight ahead, then loop around a nearby display instead of the cashier.", + "D": "You pivot left on the spot, proceed straight, and then walk around the cashier counter before stopping.", + "E": "You turn right on the spot, then proceed straight, and finally walk around the cashier counter." + }, + "answer_label": "E", + "answer_text": "You turn right on the spot, then proceed straight, and finally walk around the cashier counter.", + "q_id": "egolife_A1_D5_017-mcq_fixed_2", + "correct_answer": "You turn right on the spot, then proceed straight, and finally walk around the cashier counter.", + "original_options": { + "A": "You turn left, then go straight.", + "B": "You go straight, then turn right.", + "C": "You turn right in place, walk straight ahead, then loop around a nearby display instead of the cashier.", + "D": "You pivot left on the spot, proceed straight, and then walk around the cashier counter before stopping.", + "E": "You turn right on the spot, then proceed straight, and finally walk around the cashier counter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_017-mcq_fixed_3", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "So far, what has been my movement pattern?", + "answer": "You turn right on the spot, then proceed straight, then walk around the cashier counter again, and finally wait for checkout at the cashier counter.", + "question_chinese": "到目前为止，我的运动轨迹是怎样的？", + "answer_chinese": "你原地右转，然后直行，接着在收银台旁边踱步，最后在收银台旁等待结账。", + "evidence": [ + { + "description": "You turn right on the spot, then proceed straight along with your friend.", + "timestamps": { + "start": "00:00:26", + "end": "00:00:30" + } + }, + { + "description": "You walk around the cashier counter.", + "timestamps": { + "start": "00:00:54", + "end": "00:01:00" + } + }, + { + "description": "You and your friend are waiting for checkout.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:02:21", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turn left, then go straight.", + "B": "You circle the cashier counter once.", + "C": "You turn right on the spot, go straight, circle the cashier counter twice, then wait at checkout.", + "D": "You turn right on the spot, then proceed straight, then walk around the cashier counter again, and finally wait for checkout at the cashier counter.", + "E": "You turn right on the spot, proceed straight, then head away from the cashier counter and stop without waiting." + }, + "answer_label": "D", + "answer_text": "You turn right on the spot, then proceed straight, then walk around the cashier counter again, and finally wait for checkout at the cashier counter.", + "q_id": "egolife_A1_D5_017-mcq_fixed_3", + "correct_answer": "You turn right on the spot, then proceed straight, then walk around the cashier counter again, and finally wait for checkout at the cashier counter.", + "original_options": { + "A": "You turn left, then go straight.", + "B": "You circle the cashier counter once.", + "C": "You turn right on the spot, go straight, circle the cashier counter twice, then wait at checkout.", + "D": "You turn right on the spot, then proceed straight, then walk around the cashier counter again, and finally wait for checkout at the cashier counter.", + "E": "You turn right on the spot, proceed straight, then head away from the cashier counter and stop without waiting." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_017_1_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks did I put into the shopping bag in total?", + "answer": "You put four bottles of beverages into the shopping bag.", + "question_chinese": "我一共将多少瓶饮料放到了购物袋里面？", + "answer_chinese": "你将4瓶饮料放入了购物袋中。", + "evidence": [ + { + "description": "You put 4 bottles of Wanjia milk into the shopping bag.", + "timestamps": { + "start": "00:03:58", + "end": "00:04:00" + } + } + ], + "question_timestamps": "00:04:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You put three bottles in the bag.", + "B": "You put five bottles in the bag.", + "C": "You put six bottles of beverages into the shopping bag in total.", + "D": "You put four bottles of beverages into the shopping bag.", + "E": "You put seven bottles of beverages into the shopping bag in total." + }, + "answer_label": "D", + "answer_text": "You put four bottles of beverages into the shopping bag.", + "q_id": "egolife_A1_D5_017_1_T1", + "correct_answer": "You put four bottles of beverages into the shopping bag.", + "original_options": { + "A": "You put three bottles in the bag.", + "B": "You put five bottles in the bag.", + "C": "You put six bottles of beverages into the shopping bag in total.", + "D": "You put four bottles of beverages into the shopping bag.", + "E": "You put seven bottles of beverages into the shopping bag in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D5_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_017_2_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of drinks did I put into the shopping bag in total?", + "answer": "You put eleven bottles of beverages into the shopping bag.", + "question_chinese": "我一共将多少瓶饮料放到了购物袋里面？", + "answer_chinese": "你将11瓶饮料放入了购物袋中。", + "evidence": [ + { + "description": "You put 4 bottles of Wanjia milk into the shopping bag.", + "timestamps": { + "start": "00:03:58", + "end": "00:04:00" + } + }, + { + "description": "You put 5 bottles of beverages into the shopping bag.", + "timestamps": { + "start": "00:04:15", + "end": "00:04:29" + } + }, + { + "description": "You put 2 bottles of beverages into the shopping bag.", + "timestamps": { + "start": "00:04:35", + "end": "00:04:38" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You put ten bottles in the bag.", + "B": "You put eleven bottles of beverages into the shopping bag.", + "C": "You put twelve bottles in the bag.", + "D": "You put nine bottles of beverages into the shopping bag in total.", + "E": "You put thirteen bottles of beverages into the shopping bag in total." + }, + "answer_label": "B", + "answer_text": "You put eleven bottles of beverages into the shopping bag.", + "q_id": "egolife_A1_D5_017_2_T1", + "correct_answer": "You put eleven bottles of beverages into the shopping bag.", + "original_options": { + "A": "You put ten bottles in the bag.", + "B": "You put eleven bottles of beverages into the shopping bag.", + "C": "You put twelve bottles in the bag.", + "D": "You put nine bottles of beverages into the shopping bag in total.", + "E": "You put thirteen bottles of beverages into the shopping bag in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D5_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_017-mcq_fixed_4", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I parted from my friend, how has my movement trajectory been up to now?", + "answer": "As YOU faced the elevator and turned counterclockwise by 135 degrees, then walked straight, and then saw the \"Multi-Point Self-Purchase\" sign, YOU turned left and then walked straight to return the shopping cart.", + "question_chinese": "我和我的朋友分别后，到现在我的运动轨迹是怎样的？", + "answer_chinese": "在面对着电梯逆时针转过135度，然后直走，再看到“多点自助购”时，向左转，然后直走将购物车归还。", + "evidence": [ + { + "description": "You faced the elevator and turned counterclockwise by 135 degrees, then pushed the shopping cart and walked straight.", + "timestamps": { + "start": "00:07:34", + "end": "00:07:40" + } + }, + { + "description": "You and your friend were having a conversation in front of the \"Multi-Point Self-Purchase\" sign, then you looked to the left front, and you saw the place where the shopping cart was returned.", + "timestamps": { + "start": "00:08:32", + "end": "00:08:38" + } + } + ], + "question_timestamps": "00:08:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turn clockwise, then walk straight.", + "B": "You turn 90 degrees, then walk straight.", + "C": "As YOU faced the elevator and turned counterclockwise by 135 degrees, then walked straight, and then saw the \"Multi-Point Self-Purchase\" sign, YOU turned left and then walked straight to return the shopping cart.", + "D": "Facing the elevator, YOU turn clockwise 135 degrees, walk straight, then turn right at the sign to return the cart.", + "E": "YOU face the elevator, rotate counterclockwise 135 degrees, walk straight past the sign, then turn right and continue forward." + }, + "answer_label": "C", + "answer_text": "As YOU faced the elevator and turned counterclockwise by 135 degrees, then walked straight, and then saw the \"Multi-Point Self-Purchase\" sign, YOU turned left and then walked straight to return the shopping cart.", + "q_id": "egolife_A1_D5_017-mcq_fixed_4", + "correct_answer": "As YOU faced the elevator and turned counterclockwise by 135 degrees, then walked straight, and then saw the \"Multi-Point Self-Purchase\" sign, YOU turned left and then walked straight to return the shopping cart.", + "original_options": { + "A": "You turn clockwise, then walk straight.", + "B": "You turn 90 degrees, then walk straight.", + "C": "As YOU faced the elevator and turned counterclockwise by 135 degrees, then walked straight, and then saw the \"Multi-Point Self-Purchase\" sign, YOU turned left and then walked straight to return the shopping cart.", + "D": "Facing the elevator, YOU turn clockwise 135 degrees, walk straight, then turn right at the sign to return the cart.", + "E": "YOU face the elevator, rotate counterclockwise 135 degrees, walk straight past the sign, then turn right and continue forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_017-mcq_fixed_5", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I parted from my friend, how has my movement trajectory been up to now?", + "answer": "As YOU faced the elevator and turned counterclockwise by 135 degrees, then walked straight, and then saw the \"Multi-Point Self-Purchase\" sign, YOU turned left and then walked straight to return the shopping cart, and finally turned your back.", + "question_chinese": "我和我的朋友分别后，到现在我的运动轨迹是怎样的？", + "answer_chinese": "在面对着电梯逆时针转过135度，然后直走，再看到“多点自助购”时，向左转，然后直走将购物车归还，最后向后转身。", + "evidence": [ + { + "description": "You faced the elevator and turned counterclockwise by 135 degrees, then pushed the shopping cart and walked straight.", + "timestamps": { + "start": "00:07:34", + "end": "00:07:40" + } + }, + { + "description": "You and your friend were having a conversation in front of the \"Multi-Point Self-Purchase\" sign, then you looked to the left front, and you saw the place where the shopping cart was returned.", + "timestamps": { + "start": "00:08:32", + "end": "00:08:38" + } + }, + { + "description": "You turned your back.", + "timestamps": { + "start": "00:08:46", + "end": "00:08:50" + } + } + ], + "question_timestamps": "00:08:52", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turn clockwise, then return the cart.", + "B": "You turn 90 degrees, then return the cart.", + "C": "As YOU face the elevator, YOU turn clockwise 135 degrees, walk straight, turn right at the sign, then turn your back.", + "D": "YOU face the elevator, turn counterclockwise 135 degrees, walk straight, then turn right at the sign and finally turn your back.", + "E": "As YOU faced the elevator and turned counterclockwise by 135 degrees, then walked straight, and then saw the \"Multi-Point Self-Purchase\" sign, YOU turned left and then walked straight to return the shopping cart, and finally turned your back." + }, + "answer_label": "E", + "answer_text": "As YOU faced the elevator and turned counterclockwise by 135 degrees, then walked straight, and then saw the \"Multi-Point Self-Purchase\" sign, YOU turned left and then walked straight to return the shopping cart, and finally turned your back.", + "q_id": "egolife_A1_D5_017-mcq_fixed_5", + "correct_answer": "As YOU faced the elevator and turned counterclockwise by 135 degrees, then walked straight, and then saw the \"Multi-Point Self-Purchase\" sign, YOU turned left and then walked straight to return the shopping cart, and finally turned your back.", + "original_options": { + "A": "You turn clockwise, then return the cart.", + "B": "You turn 90 degrees, then return the cart.", + "C": "As YOU face the elevator, YOU turn clockwise 135 degrees, walk straight, turn right at the sign, then turn your back.", + "D": "YOU face the elevator, turn counterclockwise 135 degrees, walk straight, then turn right at the sign and finally turn your back.", + "E": "As YOU faced the elevator and turned counterclockwise by 135 degrees, then walked straight, and then saw the \"Multi-Point Self-Purchase\" sign, YOU turned left and then walked straight to return the shopping cart, and finally turned your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D5_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_017-mcq_fixed_8", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Based on my current location, which is closer to the supermarket entrance or the clothing store?", + "answer": "The clothing store is closer to you.", + "question_chinese": "根据我现在的位置，超市入口和服装店哪个离我更近？", + "answer_chinese": "服装店离你更近。", + "evidence": [ + { + "description": "The clothing store is closer to you.", + "timestamps": { + "start": "00:10:48", + "end": "00:10:51" + } + } + ], + "question_timestamps": "00:10:52", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The clothing store is closer to you.", + "B": "The supermarket entrance is closer to you." + }, + "answer_label": "A", + "answer_text": "The clothing store is closer to you.", + "q_id": "egolife_A1_D5_017-mcq_fixed_8", + "correct_answer": "The clothing store is closer to you.", + "original_options": { + "A": "The clothing store is closer to you.", + "B": "The supermarket entrance is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D5_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_017-mcq_fixed_9", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Based on my current location, which is closer to the supermarket entrance or the clothing store?", + "answer": "The supermarket entrance is closer to you.", + "question_chinese": "根据我现在的位置，超市入口和服装店哪个离我更近？", + "answer_chinese": "超市入口离你更近。", + "evidence": [ + { + "description": "The supermarket entrance is closer to you.", + "timestamps": { + "start": "00:11:10", + "end": "00:11:12" + } + } + ], + "question_timestamps": "00:11:13", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The clothing store is closer to you.", + "B": "The supermarket entrance is closer to you." + }, + "answer_label": "B", + "answer_text": "The supermarket entrance is closer to you.", + "q_id": "egolife_A1_D5_017-mcq_fixed_9", + "correct_answer": "The supermarket entrance is closer to you.", + "original_options": { + "A": "The clothing store is closer to you.", + "B": "The supermarket entrance is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D5_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_017-mcq_fixed_6", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Now, what goods are there directly in front of me?", + "answer": "There are clothing items in front of you.", + "question_chinese": "现在，我的正前方有什么商品？", + "answer_chinese": "你的前方有服装商品。", + "evidence": [ + { + "description": "There are many clothes in front of you.", + "timestamps": { + "start": "00:09:44", + "end": "00:09:45" + } + } + ], + "question_timestamps": "00:09:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There are shoes in front of you.", + "B": "There are hats in front of you.", + "C": "There are clothing items in front of you.", + "D": "In front of you, there is a neat row of folded towels laid out for display.", + "E": "Directly ahead, you see a display of small household items arranged together on the shelf." + }, + "answer_label": "C", + "answer_text": "There are clothing items in front of you.", + "q_id": "egolife_A1_D5_017-mcq_fixed_6", + "correct_answer": "There are clothing items in front of you.", + "original_options": { + "A": "There are shoes in front of you.", + "B": "There are hats in front of you.", + "C": "There are clothing items in front of you.", + "D": "In front of you, there is a neat row of folded towels laid out for display.", + "E": "Directly ahead, you see a display of small household items arranged together on the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_017-mcq_fixed_7", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Now, what goods are there directly in front of me?", + "answer": "There are jewelry items in front of you.", + "question_chinese": "现在，我的正前方有什么商品？", + "answer_chinese": "你的前方有珠宝首饰商品。", + "evidence": [ + { + "description": "There is a jewelry store in front of you.", + "timestamps": { + "start": "00:10:13", + "end": "00:10:14" + } + } + ], + "question_timestamps": "00:10:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There are watches in front of you.", + "B": "There are jewelry items in front of you.", + "C": "There are hair accessories in front of you.", + "D": "Right in front of you, there is a rack of scarves and small fashion accessories hanging together.", + "E": "Directly ahead, you can see small souvenir trinkets lined up in a display case." + }, + "answer_label": "B", + "answer_text": "There are jewelry items in front of you.", + "q_id": "egolife_A1_D5_017-mcq_fixed_7", + "correct_answer": "There are jewelry items in front of you.", + "original_options": { + "A": "There are watches in front of you.", + "B": "There are jewelry items in front of you.", + "C": "There are hair accessories in front of you.", + "D": "Right in front of you, there is a rack of scarves and small fashion accessories hanging together.", + "E": "Directly ahead, you can see small souvenir trinkets lined up in a display case." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D5_018-mcq.json b/QAs/egolife_A1_D5_018-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5e92fe797832762af2188aa2113666e4bcb63944 --- /dev/null +++ b/QAs/egolife_A1_D5_018-mcq.json @@ -0,0 +1,564 @@ +[ + { + "video_uid": "egolife_A1_D5_018", + "video_length": "00:10:36", + "qid": "egolife_A1_D5_018-mcq_fixed_2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was your movement trajectory from the moment you got off the vehicle until now?", + "answer": "You walked along the current road, then turned left into an alley and walked straight.", + "question_chinese": "下车后到当前时刻，我的运动轨迹是怎样的？", + "answer_chinese": "你沿当前到道路前行，然后左拐进了一条小巷子并直走", + "evidence": [ + { + "description": "You were walking along the current road.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + }, + { + "description": "You turned left into an alley and walked straight.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:02" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turned right into an alley.", + "B": "You went straight without turning.", + "C": "You walked along the current road, then turned left into an alley and walked straight.", + "D": "You walked along the current road, turned right into an alley, and then continued straight ahead.", + "E": "You walked along the current road, crossed to the other side, then entered an alley and kept walking." + }, + "answer_label": "C", + "answer_text": "You walked along the current road, then turned left into an alley and walked straight.", + "q_id": "egolife_A1_D5_018-mcq_fixed_2", + "correct_answer": "You walked along the current road, then turned left into an alley and walked straight.", + "original_options": { + "A": "You turned right into an alley.", + "B": "You went straight without turning.", + "C": "You walked along the current road, then turned left into an alley and walked straight.", + "D": "You walked along the current road, turned right into an alley, and then continued straight ahead.", + "E": "You walked along the current road, crossed to the other side, then entered an alley and kept walking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_018", + "video_length": "00:10:36", + "qid": "egolife_A1_D5_018-mcq_fixed_3", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was your movement trajectory from the moment you got off the vehicle until now?", + "answer": "You walked along the current road, then turned left into an alley and walked straight, and then turned left and returned to your place.", + "question_chinese": "下车后到当前时刻，我的运动轨迹是怎样的？", + "answer_chinese": "你沿当前到道路前行，然后左拐进了一条小巷子并直走，接着左拐回到了你住的地方。", + "evidence": [ + { + "description": "You were walking along the current road.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:35" + } + }, + { + "description": "You turned left into an alley and walked straight.", + "timestamps": { + "start": "00:01:57", + "end": "00:02:02" + } + }, + { + "description": "You continued walking straight in the alley, then turned left and returned to your house.", + "timestamps": { + "start": "00:03:48", + "end": "00:03:54" + } + } + ], + "question_timestamps": "00:03:56", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turned right and returned.", + "B": "You never returned to your place.", + "C": "You walked along the current road, turned left into an alley, went straight, then turned right to return.", + "D": "You walked along the current road, then turned left into an alley and walked straight, and then turned left and returned to your place.", + "E": "You walked along the current road, turned right into an alley, went straight, then turned left back to your place." + }, + "answer_label": "D", + "answer_text": "You walked along the current road, then turned left into an alley and walked straight, and then turned left and returned to your place.", + "q_id": "egolife_A1_D5_018-mcq_fixed_3", + "correct_answer": "You walked along the current road, then turned left into an alley and walked straight, and then turned left and returned to your place.", + "original_options": { + "A": "You turned right and returned.", + "B": "You never returned to your place.", + "C": "You walked along the current road, turned left into an alley, went straight, then turned right to return.", + "D": "You walked along the current road, then turned left into an alley and walked straight, and then turned left and returned to your place.", + "E": "You walked along the current road, turned right into an alley, went straight, then turned left back to your place." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D5_018", + "video_length": "00:10:36", + "qid": "egolife_A1_D5_018-mcq_fixed_0", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "So far, what's my movement trajectory like?", + "answer": "You go straight along the current road, then turn left at the T-junction and keep going straight.", + "question_chinese": "到目前为止，我的运动轨迹是怎样的？", + "answer_chinese": "你沿当前道路前行，然后遇到T字形路口时向左拐，继续直行。", + "evidence": [ + { + "description": "You go straight along the current road, then turn left at the T-junction.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:15" + } + }, + { + "description": "You keep going straight and look left and right.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:00:35", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turn right at the T-junction.", + "B": "You turn left, then turn right.", + "C": "You go straight along the current road, turn right at the T-junction, and continue straight afterward.", + "D": "You go straight on the current road, turn left at the T-junction, then make a U-turn shortly after.", + "E": "You go straight along the current road, then turn left at the T-junction and keep going straight." + }, + "answer_label": "E", + "answer_text": "You go straight along the current road, then turn left at the T-junction and keep going straight.", + "q_id": "egolife_A1_D5_018-mcq_fixed_0", + "correct_answer": "You go straight along the current road, then turn left at the T-junction and keep going straight.", + "original_options": { + "A": "You turn right at the T-junction.", + "B": "You turn left, then turn right.", + "C": "You go straight along the current road, turn right at the T-junction, and continue straight afterward.", + "D": "You go straight on the current road, turn left at the T-junction, then make a U-turn shortly after.", + "E": "You go straight along the current road, then turn left at the T-junction and keep going straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_018", + "video_length": "00:10:36", + "qid": "egolife_A1_D5_018-mcq_fixed_1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "So far, what's my movement trajectory like?", + "answer": "You go straight along the current road, then turn left at the T-junction and keep going straight, and finally get into a ride-hailing car.", + "question_chinese": "到目前为止，我的运动轨迹是怎样的？", + "answer_chinese": "你沿当前道路前行，然后遇到T字形路口时向左拐，继续直行，最后坐到了网约车上。", + "evidence": [ + { + "description": "You go straight along the current road, then turn left at the T-junction.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:15" + } + }, + { + "description": "You keep going straight and look left and right.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:34" + } + }, + { + "description": "You get into a ride-hailing car.", + "timestamps": { + "start": "00:00:55", + "end": "00:01:00" + } + } + ], + "question_timestamps": "00:01:01", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You get into a bus at the end.", + "B": "You stop walking before any ride.", + "C": "You go straight along the current road, turn left at the T-junction, keep going, then get onto a public bus.", + "D": "You go straight along the current road, turn right at the T-junction, continue straight, and finally get into a ride-hailing car.", + "E": "You go straight along the current road, then turn left at the T-junction and keep going straight, and finally get into a ride-hailing car." + }, + "answer_label": "E", + "answer_text": "You go straight along the current road, then turn left at the T-junction and keep going straight, and finally get into a ride-hailing car.", + "q_id": "egolife_A1_D5_018-mcq_fixed_1", + "correct_answer": "You go straight along the current road, then turn left at the T-junction and keep going straight, and finally get into a ride-hailing car.", + "original_options": { + "A": "You get into a bus at the end.", + "B": "You stop walking before any ride.", + "C": "You go straight along the current road, turn left at the T-junction, keep going, then get onto a public bus.", + "D": "You go straight along the current road, turn right at the T-junction, continue straight, and finally get into a ride-hailing car.", + "E": "You go straight along the current road, then turn left at the T-junction and keep going straight, and finally get into a ride-hailing car." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D5_018", + "video_length": "00:10:36", + "qid": "egolife_A1_D5_018-mcq_fixed_8", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Am I now on the side of the living room table that is far from the kitchen, or the side that is close to the kitchen?", + "answer": "You are on the side of the living room table that is close to the kitchen.", + "question_chinese": "我现在是在客厅桌子离厨房远的一侧，还是近的一侧", + "answer_chinese": "你在客厅桌子离厨房近的一侧。", + "evidence": [ + { + "description": "You are on the side of the living room table that is close to the kitchen.", + "timestamps": { + "start": "00:09:01", + "end": "00:09:03" + } + } + ], + "question_timestamps": "00:09:04", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You are on the side of the living room table that is close to the kitchen.", + "B": "You are on the side of the living room table that is far from the kitchen." + }, + "answer_label": "A", + "answer_text": "You are on the side of the living room table that is close to the kitchen.", + "q_id": "egolife_A1_D5_018-mcq_fixed_8", + "correct_answer": "You are on the side of the living room table that is close to the kitchen.", + "original_options": { + "A": "You are on the side of the living room table that is close to the kitchen.", + "B": "You are on the side of the living room table that is far from the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_018", + "video_length": "00:10:36", + "qid": "egolife_A1_D5_018-mcq_fixed_9", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Am I now on the side of the living room table that is far from the kitchen, or the side that is close to the kitchen?", + "answer": "You are on the side of the dining table in the living room that is far from the kitchen.", + "question_chinese": "我现在是在客厅桌子离厨房远的一侧，还是近的一侧", + "answer_chinese": "你在客厅桌子离厨房远的一侧", + "evidence": [ + { + "description": "You are on the side of the dining table in the living room that is far from the kitchen.", + "timestamps": { + "start": "00:09:16", + "end": "00:09:18" + } + } + ], + "question_timestamps": "00:09:19", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You are on the side of the dining table in the living room that is close to the kitchen.", + "B": "You are on the side of the dining table in the living room that is far from the kitchen." + }, + "answer_label": "B", + "answer_text": "You are on the side of the dining table in the living room that is far from the kitchen.", + "q_id": "egolife_A1_D5_018-mcq_fixed_9", + "correct_answer": "You are on the side of the dining table in the living room that is far from the kitchen.", + "original_options": { + "A": "You are on the side of the dining table in the living room that is close to the kitchen.", + "B": "You are on the side of the dining table in the living room that is far from the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_018", + "video_length": "00:10:36", + "qid": "egolife_A1_D5_018_5", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many friends are currently in the living room with me?", + "answer": "There are six friends with you in the living room now.", + "question_chinese": "现在一共有多少个朋友和我在客厅？", + "answer_chinese": "现在有6个朋友和你在客厅。", + "evidence": [ + { + "description": "There are 6 of your friends in the living room.", + "timestamps": { + "start": "00:04:40", + "end": "00:04:45" + } + } + ], + "question_timestamps": "00:04:56", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are five friends with you now.", + "B": "There are seven friends with you now.", + "C": "There are six friends with you in the living room now.", + "D": "There are four friends with you in the living room at the moment.", + "E": "There are eight friends with you in the living room right now, not six." + }, + "answer_label": "C", + "answer_text": "There are six friends with you in the living room now.", + "q_id": "egolife_A1_D5_018_5", + "correct_answer": "There are six friends with you in the living room now.", + "original_options": { + "A": "There are five friends with you now.", + "B": "There are seven friends with you now.", + "C": "There are six friends with you in the living room now.", + "D": "There are four friends with you in the living room at the moment.", + "E": "There are eight friends with you in the living room right now, not six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D5_018", + "video_length": "00:10:36", + "qid": "egolife_A1_D5_018_6", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many friends are currently in the living room with me?", + "answer": "There are eight friends with you in the living room now.", + "question_chinese": "现在一共有多少个朋友和我在客厅？", + "answer_chinese": "现在有8个朋友和你在客厅。", + "evidence": [ + { + "description": "There are 6 of your friends in the living room.", + "timestamps": { + "start": "00:04:40", + "end": "00:04:45" + } + }, + { + "description": "2 of the friends who were with you in the yard earlier also came into the living room.", + "timestamps": { + "start": "00:04:52", + "end": "00:04:54" + } + } + ], + "question_timestamps": "00:05:05", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are seven friends with you now.", + "B": "There are nine friends with you now.", + "C": "There are eight friends with you in the living room now.", + "D": "There are six friends with you in the living room at the moment.", + "E": "There are ten friends with you in the living room right now, not eight." + }, + "answer_label": "C", + "answer_text": "There are eight friends with you in the living room now.", + "q_id": "egolife_A1_D5_018_6", + "correct_answer": "There are eight friends with you in the living room now.", + "original_options": { + "A": "There are seven friends with you now.", + "B": "There are nine friends with you now.", + "C": "There are eight friends with you in the living room now.", + "D": "There are six friends with you in the living room at the moment.", + "E": "There are ten friends with you in the living room right now, not eight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D5_018", + "video_length": "00:10:36", + "qid": "egolife_A1_D5_018-mcq_fixed_6", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the glass now?", + "answer": "You can reach the glasses.", + "question_chinese": "现在我能够得到玻璃杯吗？", + "answer_chinese": "你够的到玻璃杯。", + "evidence": [ + { + "description": "You put two glasses beside the kitchen sink.", + "timestamps": { + "start": "00:04:46", + "end": "00:04:48" + } + } + ], + "question_timestamps": "00:04:49", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You cannot reach the glasses.", + "B": "You can reach the glasses." + }, + "answer_label": "B", + "answer_text": "You can reach the glasses.", + "q_id": "egolife_A1_D5_018-mcq_fixed_6", + "correct_answer": "You can reach the glasses.", + "original_options": { + "A": "You cannot reach the glasses.", + "B": "You can reach the glasses." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D5_018", + "video_length": "00:10:36", + "qid": "egolife_A1_D5_018-mcq_fixed_7", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the glass now?", + "answer": "You can't reach the glasses.", + "question_chinese": "现在我能够得到玻璃杯吗？", + "answer_chinese": "你够不到玻璃杯。", + "evidence": [ + { + "description": "You put two glasses beside the kitchen sink.", + "timestamps": { + "start": "00:04:46", + "end": "00:04:48" + } + }, + { + "description": "You are beside the table in the living room.", + "timestamps": { + "start": "00:05:08", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:05:11", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't reach the glasses.", + "B": "You can reach the glasses." + }, + "answer_label": "A", + "answer_text": "You can't reach the glasses.", + "q_id": "egolife_A1_D5_018-mcq_fixed_7", + "correct_answer": "You can't reach the glasses.", + "original_options": { + "A": "You can't reach the glasses.", + "B": "You can reach the glasses." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + } +] diff --git a/QAs/egolife_A1_D5_019-mcq.json b/QAs/egolife_A1_D5_019-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d0080af55ae36f97966937c0a4a837e21a76ee0f --- /dev/null +++ b/QAs/egolife_A1_D5_019-mcq.json @@ -0,0 +1,601 @@ +[ + { + "video_uid": "egolife_A1_D5_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_019_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I been to the second floor so far?", + "answer": "You have gone to the second floor a total of two times.", + "question_chinese": "到现在为止，我去了几次二楼？", + "answer_chinese": "你一共去了2次二楼。", + "evidence": [ + { + "description": "You are currently going from the first floor to the second floor.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:56" + } + }, + { + "description": "You are currently going from the first floor to the second floor.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:17" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have gone once.", + "B": "You have gone three times.", + "C": "You have gone to the second floor a total of four times so far.", + "D": "You have gone to the second floor a total of five times up to now.", + "E": "You have gone to the second floor a total of two times." + }, + "answer_label": "E", + "answer_text": "You have gone to the second floor a total of two times.", + "q_id": "egolife_A1_D5_019_1", + "correct_answer": "You have gone to the second floor a total of two times.", + "original_options": { + "A": "You have gone once.", + "B": "You have gone three times.", + "C": "You have gone to the second floor a total of four times so far.", + "D": "You have gone to the second floor a total of five times up to now.", + "E": "You have gone to the second floor a total of two times." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_019_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I been to the second floor so far?", + "answer": "You have gone to the second floor a total of three times.", + "question_chinese": "到现在为止，我去了几次二楼？", + "answer_chinese": "你一共去了3次二楼", + "evidence": [ + { + "description": "You are currently going from the first floor to the second floor.", + "timestamps": { + "start": "00:01:47", + "end": "00:01:56" + } + }, + { + "description": "You are currently going from the first floor to the second floor.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:17" + } + }, + { + "description": "You are currently carrying a glass cup and heading to the second floor.", + "timestamps": { + "start": "00:04:48", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have gone twice.", + "B": "You have gone to the second floor a total of three times.", + "C": "You have gone four times.", + "D": "You have gone to the second floor a total of five times so far.", + "E": "You have gone to the second floor a total of six times up to now." + }, + "answer_label": "B", + "answer_text": "You have gone to the second floor a total of three times.", + "q_id": "egolife_A1_D5_019_2", + "correct_answer": "You have gone to the second floor a total of three times.", + "original_options": { + "A": "You have gone twice.", + "B": "You have gone to the second floor a total of three times.", + "C": "You have gone four times.", + "D": "You have gone to the second floor a total of five times so far.", + "E": "You have gone to the second floor a total of six times up to now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_019_7", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the computer now?", + "answer": "You can get the computer.", + "question_chinese": "我现在可以够得到电脑吗？", + "answer_chinese": "你能够得到电脑。", + "evidence": [ + { + "description": "There is a computer at the corner on your right front.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:02" + } + }, + { + "description": "You are using the computer.", + "timestamps": { + "start": "00:02:18", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:02:24", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can get the computer.", + "B": "You cannot get the computer." + }, + "answer_label": "A", + "answer_text": "You can get the computer.", + "q_id": "egolife_A1_D5_019_7", + "correct_answer": "You can get the computer.", + "original_options": { + "A": "You can get the computer.", + "B": "You cannot get the computer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D5_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_019_8", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the computer now?", + "answer": "You cannot get the computer.", + "question_chinese": "我现在可以够得到电脑吗？", + "answer_chinese": "你不能够得到电脑。", + "evidence": [ + { + "description": "There is a computer at the corner on your right front.", + "timestamps": { + "start": "00:02:00", + "end": "00:02:02" + } + }, + { + "description": "You and your friend are on the first floor.", + "timestamps": { + "start": "00:04:16", + "end": "00:04:19" + } + } + ], + "question_timestamps": "00:04:23", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can get the computer.", + "B": "You cannot get the computer." + }, + "answer_label": "B", + "answer_text": "You cannot get the computer.", + "q_id": "egolife_A1_D5_019_8", + "correct_answer": "You cannot get the computer.", + "original_options": { + "A": "You can get the computer.", + "B": "You cannot get the computer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D5_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_019_3_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many eggs has my friend and I each cracked?", + "answer": "You and your friend have cracked a total of one egg so far.", + "question_chinese": "现在我和我的朋友一个打了多少个鸡蛋？", + "answer_chinese": "你和你的朋友一共已经打了1个鸡蛋。", + "evidence": [ + { + "description": "You are watching your friend crack the first egg.", + "timestamps": { + "start": "00:09:41", + "end": "00:09:45" + } + } + ], + "question_timestamps": "00:09:58", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two eggs total.", + "B": "Zero eggs total.", + "C": "You and your friend have cracked a total of two eggs so far.", + "D": "You and your friend have cracked a total of three eggs up to this point.", + "E": "You and your friend have cracked a total of one egg so far." + }, + "answer_label": "E", + "answer_text": "You and your friend have cracked a total of one egg so far.", + "q_id": "egolife_A1_D5_019_3_T1", + "correct_answer": "You and your friend have cracked a total of one egg so far.", + "original_options": { + "A": "Two eggs total.", + "B": "Zero eggs total.", + "C": "You and your friend have cracked a total of two eggs so far.", + "D": "You and your friend have cracked a total of three eggs up to this point.", + "E": "You and your friend have cracked a total of one egg so far." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_019_4_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many eggs has my friend and I each cracked?", + "answer": "You and your friend have cracked a total of 3 eggs so far.", + "question_chinese": "现在我和我的朋友一个打了多少个鸡蛋？", + "answer_chinese": "你和你的朋友一共已经打了3个鸡蛋。", + "evidence": [ + { + "description": "You are watching your friend crack the first egg.", + "timestamps": { + "start": "00:09:41", + "end": "00:09:45" + } + }, + { + "description": "Your friend has one egg in his hand, and there are 3 eggs on the table.", + "timestamps": { + "start": "00:09:58", + "end": "00:10:01" + } + } + ], + "question_timestamps": "00:10:03", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two eggs total.", + "B": "You and your friend have cracked a total of 3 eggs so far.", + "C": "Five eggs total.", + "D": "You and your friend have cracked a total of two eggs so far.", + "E": "You and your friend have cracked a total of five eggs up to this point." + }, + "answer_label": "B", + "answer_text": "You and your friend have cracked a total of 3 eggs so far.", + "q_id": "egolife_A1_D5_019_4_T2", + "correct_answer": "You and your friend have cracked a total of 3 eggs so far.", + "original_options": { + "A": "Two eggs total.", + "B": "You and your friend have cracked a total of 3 eggs so far.", + "C": "Five eggs total.", + "D": "You and your friend have cracked a total of two eggs so far.", + "E": "You and your friend have cracked a total of five eggs up to this point." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_019_3_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many eggs has my friend and I each cracked?", + "answer": "You and your friend have cracked a total of 2 eggs so far.", + "question_chinese": "现在我和我的朋友一个打了多少个鸡蛋？", + "answer_chinese": "你和你的朋友一共已经打了2个鸡蛋。", + "evidence": [ + { + "description": "You are watching your friend crack the first egg.", + "timestamps": { + "start": "00:09:41", + "end": "00:09:45" + } + }, + { + "description": "Your friend beat the second egg.", + "timestamps": { + "start": "00:09:59", + "end": "00:10:30" + } + } + ], + "question_timestamps": "00:10:40", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You and your friend have cracked a total of 2 eggs so far.", + "B": "One egg total.", + "C": "Three eggs total.", + "D": "You and your friend have cracked a total of one egg so far.", + "E": "You and your friend have cracked a total of four eggs up to this point." + }, + "answer_label": "A", + "answer_text": "You and your friend have cracked a total of 2 eggs so far.", + "q_id": "egolife_A1_D5_019_3_T2", + "correct_answer": "You and your friend have cracked a total of 2 eggs so far.", + "original_options": { + "A": "You and your friend have cracked a total of 2 eggs so far.", + "B": "One egg total.", + "C": "Three eggs total.", + "D": "You and your friend have cracked a total of one egg so far.", + "E": "You and your friend have cracked a total of four eggs up to this point." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_019_4_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "By now, how many eggs has my friend and I each cracked?", + "answer": "You and your friend have cracked a total of 5 eggs so far.", + "question_chinese": "现在我和我的朋友一个打了多少个鸡蛋？", + "answer_chinese": "你和你的朋友一共已经打了5个鸡蛋。", + "evidence": [ + { + "description": "You are watching your friend crack the first egg.", + "timestamps": { + "start": "00:09:41", + "end": "00:09:45" + } + }, + { + "description": "Your friend has one egg in his hand, and there are 3 eggs on the table.", + "timestamps": { + "start": "00:09:58", + "end": "00:10:01" + } + }, + { + "description": "There are no eggs on the table or in your friend's hand anymore.", + "timestamps": { + "start": "00:11:35", + "end": "00:11:40" + } + } + ], + "question_timestamps": "00:11:52", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You and your friend have cracked a total of 5 eggs so far.", + "B": "Four eggs total.", + "C": "Six eggs total.", + "D": "You and your friend have cracked a total of four eggs so far.", + "E": "You and your friend have cracked a total of seven eggs up to this point." + }, + "answer_label": "A", + "answer_text": "You and your friend have cracked a total of 5 eggs so far.", + "q_id": "egolife_A1_D5_019_4_T1", + "correct_answer": "You and your friend have cracked a total of 5 eggs so far.", + "original_options": { + "A": "You and your friend have cracked a total of 5 eggs so far.", + "B": "Four eggs total.", + "C": "Six eggs total.", + "D": "You and your friend have cracked a total of four eggs so far.", + "E": "You and your friend have cracked a total of seven eggs up to this point." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_019-mcq_fixed_6", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is around me now?", + "answer": "There is a chair on your left and many water bottles and plates in front of you.", + "question_chinese": "现在我的周围有什么？", + "answer_chinese": "你的的左边有把椅子并且前方有许多水瓶和盘子。", + "evidence": [ + { + "description": "When you reach the operation table in the kitchen on the second floor, there is a chair beside the table, and there are many water bottles on the table and many plates in the sink.", + "timestamps": { + "start": "00:03:19", + "end": "00:03:22" + } + } + ], + "question_timestamps": "00:03:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A sofa left, glasses in front.", + "B": "A table right with cups ahead.", + "C": "There is a sofa on your left, and several glasses and napkins are set in front of you.", + "D": "There is a chair on your left and many water bottles and plates in front of you.", + "E": "There is a table on your right with many cups and forks placed in front of you." + }, + "answer_label": "D", + "answer_text": "There is a chair on your left and many water bottles and plates in front of you.", + "q_id": "egolife_A1_D5_019-mcq_fixed_6", + "correct_answer": "There is a chair on your left and many water bottles and plates in front of you.", + "original_options": { + "A": "A sofa left, glasses in front.", + "B": "A table right with cups ahead.", + "C": "There is a sofa on your left, and several glasses and napkins are set in front of you.", + "D": "There is a chair on your left and many water bottles and plates in front of you.", + "E": "There is a table on your right with many cups and forks placed in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_019", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_019-mcq_fixed_7", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is around me now?", + "answer": "There is a computer and some other electronic devices around you.", + "question_chinese": "现在我的周围有什么？", + "answer_chinese": "你的周围有一台电脑和其他一些电子设备。", + "evidence": [ + { + "description": "There is a computer and some other electronic devices around you.", + "timestamps": { + "start": "00:03:34", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:03:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Mostly kitchen appliances and utensils.", + "B": "A television with household items.", + "C": "There are mainly stationery supplies, notebooks, and printed documents nearby, with no prominent electronics around you.", + "D": "There is a printer nearby along with several office supplies, suggesting an office setup rather than devices.", + "E": "There is a computer and some other electronic devices around you." + }, + "answer_label": "E", + "answer_text": "There is a computer and some other electronic devices around you.", + "q_id": "egolife_A1_D5_019-mcq_fixed_7", + "correct_answer": "There is a computer and some other electronic devices around you.", + "original_options": { + "A": "Mostly kitchen appliances and utensils.", + "B": "A television with household items.", + "C": "There are mainly stationery supplies, notebooks, and printed documents nearby, with no prominent electronics around you.", + "D": "There is a printer nearby along with several office supplies, suggesting an office setup rather than devices.", + "E": "There is a computer and some other electronic devices around you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D5_020-mcq.json b/QAs/egolife_A1_D5_020-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c829fe8b6643f0c09791584129d575f909f17057 --- /dev/null +++ b/QAs/egolife_A1_D5_020-mcq.json @@ -0,0 +1,585 @@ +[ + { + "video_uid": "egolife_A1_D5_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_020_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mobile phones have I used so far?", + "answer": "You have used one phone.", + "question_chinese": "到目前为止，我一共使用过几部手机？", + "answer_chinese": "你使用过一部手机。", + "evidence": [ + { + "description": "You are currently using your own phone.", + "timestamps": { + "start": "00:06:54", + "end": "00:06:56" + } + } + ], + "question_timestamps": "00:07:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have used one phone.", + "B": "Two phones.", + "C": "Three phones.", + "D": "You have used two phones so far during the activity, switching once between devices.", + "E": "You have used four phones so far, switching between different devices during the activity." + }, + "answer_label": "A", + "answer_text": "You have used one phone.", + "q_id": "egolife_A1_D5_020_7", + "correct_answer": "You have used one phone.", + "original_options": { + "A": "You have used one phone.", + "B": "Two phones.", + "C": "Three phones.", + "D": "You have used two phones so far during the activity, switching once between devices.", + "E": "You have used four phones so far, switching between different devices during the activity." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D5_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_020_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many mobile phones have I used so far?", + "answer": "You have used two phones.", + "question_chinese": "到目前为止，我一共使用过几部手机？", + "answer_chinese": "你使用过两部手机。", + "evidence": [ + { + "description": "You are currently using your own phone.", + "timestamps": { + "start": "00:06:54", + "end": "00:06:56" + } + }, + { + "description": "You received your friend's phone and are using this one.", + "timestamps": { + "start": "00:08:50", + "end": "00:08:54" + } + } + ], + "question_timestamps": "00:09:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have used two phones.", + "B": "One phone.", + "C": "Three phones.", + "D": "You have used one phone so far, without switching to another device.", + "E": "You have used four phones so far, changing devices multiple times during the activity." + }, + "answer_label": "A", + "answer_text": "You have used two phones.", + "q_id": "egolife_A1_D5_020_8", + "correct_answer": "You have used two phones.", + "original_options": { + "A": "You have used two phones.", + "B": "One phone.", + "C": "Three phones.", + "D": "You have used one phone so far, without switching to another device.", + "E": "You have used four phones so far, changing devices multiple times during the activity." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D5_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_020_5", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What is in front of me when I stand in front of the kitchen countertop?", + "answer": "There is a chopping board and many seasonings in front of you.", + "question_chinese": "当我站在厨房操作台前方时，我的前面有什么？", + "answer_chinese": "你的前面有砧板和许多调料。", + "evidence": [ + { + "description": "There is a chopping board and many seasonings in front of you.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:22" + } + } + ], + "question_timestamps": "00:05:26", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "There is a chopping board and many seasonings in front of you.", + "B": "There is a chopping board to your left.", + "C": "There are many seasonings behind you.", + "D": "There is a chopping board and many seasonings to your right, not directly in front of you.", + "E": "In front of you are several bowls and spices, while the chopping board sits off to the side." + }, + "answer_label": "A", + "answer_text": "There is a chopping board and many seasonings in front of you.", + "q_id": "egolife_A1_D5_020_5", + "correct_answer": "There is a chopping board and many seasonings in front of you.", + "original_options": { + "A": "There is a chopping board and many seasonings in front of you.", + "B": "There is a chopping board to your left.", + "C": "There are many seasonings behind you.", + "D": "There is a chopping board and many seasonings to your right, not directly in front of you.", + "E": "In front of you are several bowls and spices, while the chopping board sits off to the side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_020_6", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What is in front of me when I stand in front of the kitchen countertop?", + "answer": "There is a chopping board and many seasonings in front of you, and one of your friends is outside the window in front of you.", + "question_chinese": "当我站在厨房操作台前方时，我的前面有什么？", + "answer_chinese": "你的前面有有砧板和许多调料，并且你的一个朋友在你前方窗户的外面。", + "evidence": [ + { + "description": "There is a chopping board and many seasonings in front of you.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:22" + } + }, + { + "description": "One of your friends is walking in front of the kitchen window.", + "timestamps": { + "start": "00:05:24", + "end": "00:05:25" + } + } + ], + "question_timestamps": "00:05:29", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "A knife and fruit are in front.", + "B": "Your friend is inside ahead.", + "C": "There is a chopping board and many seasonings to your left, and your friend is outside the window ahead.", + "D": "There is a chopping board and many seasonings behind you, while your friend remains outside the window in front.", + "E": "There is a chopping board and many seasonings in front of you, and one of your friends is outside the window in front of you." + }, + "answer_label": "E", + "answer_text": "There is a chopping board and many seasonings in front of you, and one of your friends is outside the window in front of you.", + "q_id": "egolife_A1_D5_020_6", + "correct_answer": "There is a chopping board and many seasonings in front of you, and one of your friends is outside the window in front of you.", + "original_options": { + "A": "A knife and fruit are in front.", + "B": "Your friend is inside ahead.", + "C": "There is a chopping board and many seasonings to your left, and your friend is outside the window ahead.", + "D": "There is a chopping board and many seasonings behind you, while your friend remains outside the window in front.", + "E": "There is a chopping board and many seasonings in front of you, and one of your friends is outside the window in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_020_9", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How can I move the cup from the table to the kitchen countertop?", + "answer": "You pick up the glass cup on the table, then go counterclockwise around the table, and then walk straight ahead.", + "question_chinese": "我现在怎么可以把桌子上的杯子放到厨房的操作台上？", + "answer_chinese": "你拿起桌上的玻璃杯，然后逆时针绕过桌子，再直走", + "evidence": [ + { + "description": "When you are sitting near the glass cup, your front is towards the kitchen.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:08" + } + } + ], + "question_timestamps": "00:03:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go clockwise around the table.", + "B": "Walk left after circling the table.", + "C": "You pick up the glass cup, go clockwise around the table, then walk straight ahead to the countertop.", + "D": "You pick up the glass cup on the table, then go counterclockwise around the table, and then walk straight ahead.", + "E": "You pick up the glass cup, go counterclockwise around the kitchen instead of the table, then walk straight ahead." + }, + "answer_label": "D", + "answer_text": "You pick up the glass cup on the table, then go counterclockwise around the table, and then walk straight ahead.", + "q_id": "egolife_A1_D5_020_9", + "correct_answer": "You pick up the glass cup on the table, then go counterclockwise around the table, and then walk straight ahead.", + "original_options": { + "A": "Go clockwise around the table.", + "B": "Walk left after circling the table.", + "C": "You pick up the glass cup, go clockwise around the table, then walk straight ahead to the countertop.", + "D": "You pick up the glass cup on the table, then go counterclockwise around the table, and then walk straight ahead.", + "E": "You pick up the glass cup, go counterclockwise around the kitchen instead of the table, then walk straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_020_10", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How can I move the cup from the table to the kitchen countertop?", + "answer": "Now you stand up, go clockwise to the opposite side of the table to pick up the cup, then turn around, go around the table and walk straight ahead.", + "question_chinese": "我现在怎么可以把桌子上的杯子放到厨房的操作台上？", + "answer_chinese": "你现在起身，顺时针绕到桌子的对面拿起杯子，然后转身，绕过桌子再直走。", + "evidence": [ + { + "description": "When you are sitting near the glass cup, your front is towards the kitchen.", + "timestamps": { + "start": "00:03:05", + "end": "00:03:08" + } + }, + { + "description": "The glass cup is on the table on your left, and the kitchen worktop is on your right.", + "timestamps": { + "start": "00:09:16", + "end": "00:09:18" + } + }, + { + "description": "You are sitting opposite the position where you sat before.", + "timestamps": { + "start": "00:11:20", + "end": "00:11:22" + } + } + ], + "question_timestamps": "00:11:26", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Now you stand up, go clockwise to the opposite side of the table to pick up the cup, then turn around, go around the table and walk straight ahead.", + "B": "Sit down and go counterclockwise.", + "C": "Pick it up from the left side.", + "D": "Stand up, go clockwise to the right side to grab the cup, then walk left after circling the table.", + "E": "Bend down, go counterclockwise to the opposite side to reach the cup, then go behind the table straight ahead." + }, + "answer_label": "A", + "answer_text": "Now you stand up, go clockwise to the opposite side of the table to pick up the cup, then turn around, go around the table and walk straight ahead.", + "q_id": "egolife_A1_D5_020_10", + "correct_answer": "Now you stand up, go clockwise to the opposite side of the table to pick up the cup, then turn around, go around the table and walk straight ahead.", + "original_options": { + "A": "Now you stand up, go clockwise to the opposite side of the table to pick up the cup, then turn around, go around the table and walk straight ahead.", + "B": "Sit down and go counterclockwise.", + "C": "Pick it up from the left side.", + "D": "Stand up, go clockwise to the right side to grab the cup, then walk left after circling the table.", + "E": "Bend down, go counterclockwise to the opposite side to reach the cup, then go behind the table straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_020-mcq_fixed_8", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many of my friends directly participated in the making of the protein batter?", + "answer": "Four.", + "question_chinese": "有多少个我的朋友直接参与了蛋白糊的制作？", + "answer_chinese": "有2个你的朋友直接参与了蛋糕的制作。", + "evidence": [ + { + "description": "Two of your friends are configuring the ingredients for the egg yolk batter.", + "timestamps": { + "start": "00:01:48", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:02:01", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Four.", + "C": "Two.", + "D": "Three friends directly participated in making the protein batter, helping with mixing and adding ingredients together.", + "E": "Five friends directly participated in making the protein batter, working together on the mixing steps." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "egolife_A1_D5_020-mcq_fixed_8", + "correct_answer": "Four.", + "original_options": { + "A": "One.", + "B": "Four.", + "C": "Two.", + "D": "Three friends directly participated in making the protein batter, helping with mixing and adding ingredients together.", + "E": "Five friends directly participated in making the protein batter, working together on the mixing steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_020-mcq_fixed_9", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many of my friends directly participated in the making of the protein batter?", + "answer": "Six.", + "question_chinese": "有多少个我的朋友直接参与了蛋白糊的制作？", + "answer_chinese": "有3个你的朋友直接参与了蛋糕的制作。", + "evidence": [ + { + "description": "Two of your friends are configuring the ingredients for the egg yolk batter.", + "timestamps": { + "start": "00:01:48", + "end": "00:02:00" + } + }, + { + "description": "Your friend in the white clothes is stirring the egg yolk batter.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:30" + } + } + ], + "question_timestamps": "00:04:31", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "Seven friends directly participated in making the protein batter, assisting with mixing and adding the ingredients.", + "E": "Eight friends directly participated in making the protein batter, taking turns mixing and preparing the batter." + }, + "answer_label": "B", + "answer_text": "Six.", + "q_id": "egolife_A1_D5_020-mcq_fixed_9", + "correct_answer": "Six.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "Five.", + "D": "Seven friends directly participated in making the protein batter, assisting with mixing and adding the ingredients.", + "E": "Eight friends directly participated in making the protein batter, taking turns mixing and preparing the batter." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_020-mcq_fixed_6", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Based on my position, which is closer to me, the strawberries or the refrigerator?", + "answer": "The strawberry is closer to you.", + "question_chinese": "根据我的位置，草莓和冰箱哪个离我更近？", + "answer_chinese": "草莓离你更近。", + "evidence": [ + { + "description": "The strawberries are closer to you.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The strawberry is closer to you.", + "B": "The refrigerator is closer to you." + }, + "answer_label": "A", + "answer_text": "The strawberry is closer to you.", + "q_id": "egolife_A1_D5_020-mcq_fixed_6", + "correct_answer": "The strawberry is closer to you.", + "original_options": { + "A": "The strawberry is closer to you.", + "B": "The refrigerator is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D5_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_020-mcq_fixed_7", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Based on my position, which is closer to me, the strawberries or the refrigerator?", + "answer": "The refrigerator is closer to you.", + "question_chinese": "根据我的位置，草莓和冰箱哪个离我更近？", + "answer_chinese": "冰箱离你更近", + "evidence": [ + { + "description": "The refrigerator is closer to you.", + "timestamps": { + "start": "00:04:24", + "end": "00:04:26" + } + } + ], + "question_timestamps": "00:04:27", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The strawberries are closer to you.", + "B": "The refrigerator is closer to you." + }, + "answer_label": "B", + "answer_text": "The refrigerator is closer to you.", + "q_id": "egolife_A1_D5_020-mcq_fixed_7", + "correct_answer": "The refrigerator is closer to you.", + "original_options": { + "A": "The strawberries are closer to you.", + "B": "The refrigerator is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/egolife_A1_D5_021-mcq.json b/QAs/egolife_A1_D5_021-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2f5912a8e35f79e2b9f7052985624e6ced7c9439 --- /dev/null +++ b/QAs/egolife_A1_D5_021-mcq.json @@ -0,0 +1,580 @@ +[ + { + "video_uid": "egolife_A1_D5_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_021_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stones have I thrown in the yard so far?", + "answer": "You threw a total of four pebbles in the yard.", + "question_chinese": "到目前为止，我在院子里面一共扔了多少个石子？", + "answer_chinese": "你在院子里一共扔了4个石子。", + "evidence": [ + { + "description": "You threw 3 pebbles in the yard.", + "timestamps": { + "start": "00:04:42", + "end": "00:05:00" + } + }, + { + "description": "You threw one pebble in the yard.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:42" + } + } + ], + "question_timestamps": "00:05:53", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You threw two pebbles in the yard.", + "B": "You threw five pebbles in the yard.", + "C": "So far, you have thrown three pebbles into the yard, one after another across separate tosses.", + "D": "You threw a total of four pebbles in the yard.", + "E": "Up to this point, you only tossed six small pebbles into the yard in total." + }, + "answer_label": "D", + "answer_text": "You threw a total of four pebbles in the yard.", + "q_id": "egolife_A1_D5_021_3", + "correct_answer": "You threw a total of four pebbles in the yard.", + "original_options": { + "A": "You threw two pebbles in the yard.", + "B": "You threw five pebbles in the yard.", + "C": "So far, you have thrown three pebbles into the yard, one after another across separate tosses.", + "D": "You threw a total of four pebbles in the yard.", + "E": "Up to this point, you only tossed six small pebbles into the yard in total." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_021_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many stones have I thrown in the yard so far?", + "answer": "You threw a total of five pebbles in the yard.", + "question_chinese": "到目前为止，我在院子里面一共扔了多少个石子？", + "answer_chinese": "你在院子里一共扔了5个石子。", + "evidence": [ + { + "description": "You threw 3 pebbles in the yard.", + "timestamps": { + "start": "00:04:42", + "end": "00:05:00" + } + }, + { + "description": "You threw one pebble in the yard.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:42" + } + }, + { + "description": "You threw one pebble in the yard.", + "timestamps": { + "start": "00:06:17", + "end": "00:06:20" + } + } + ], + "question_timestamps": "00:06:31", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You threw four pebbles in the yard.", + "B": "You threw three pebbles in the yard.", + "C": "So far, you have thrown two pebbles into the yard, ending your throws after the second toss.", + "D": "You threw a total of five pebbles in the yard.", + "E": "Up to now, you have tossed six pebbles into the yard in total, not five." + }, + "answer_label": "D", + "answer_text": "You threw a total of five pebbles in the yard.", + "q_id": "egolife_A1_D5_021_4", + "correct_answer": "You threw a total of five pebbles in the yard.", + "original_options": { + "A": "You threw four pebbles in the yard.", + "B": "You threw three pebbles in the yard.", + "C": "So far, you have thrown two pebbles into the yard, ending your throws after the second toss.", + "D": "You threw a total of five pebbles in the yard.", + "E": "Up to now, you have tossed six pebbles into the yard in total, not five." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_021-mcq_fixed_4", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the wine in my hand so far?", + "answer": "The wine was handed to your friend.", + "question_chinese": "我拿出包装里面的酒后，它是怎么被放进冰箱的？", + "answer_chinese": "酒被你递给了你的朋友。", + "evidence": [ + { + "description": "You took out the wine from the package at the entrance of the living room and handed it to your friend.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:02" + } + } + ], + "question_timestamps": "00:09:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The wine stayed in your hand.", + "B": "You set the wine down.", + "C": "The wine moved from your hand onto a nearby surface, and you did not pass it to anyone.", + "D": "The wine was handed to your friend.", + "E": "You kept holding the wine, briefly raised it, and then brought it back down without handing it off." + }, + "answer_label": "D", + "answer_text": "The wine was handed to your friend.", + "q_id": "egolife_A1_D5_021-mcq_fixed_4", + "correct_answer": "The wine was handed to your friend.", + "original_options": { + "A": "The wine stayed in your hand.", + "B": "You set the wine down.", + "C": "The wine moved from your hand onto a nearby surface, and you did not pass it to anyone.", + "D": "The wine was handed to your friend.", + "E": "You kept holding the wine, briefly raised it, and then brought it back down without handing it off." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_021-mcq_fixed_5", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the wine in my hand so far?", + "answer": "The wine was handed to your friend, and then your friend placed it in the refrigerator.", + "question_chinese": "我拿出包装里面的酒后，它是怎么被放进冰箱的？", + "answer_chinese": "酒被你递给了你的朋友，然后你的朋友将它放到了冰箱里。", + "evidence": [ + { + "description": "You took out the wine from the package at the entrance of the living room and handed it to your friend.", + "timestamps": { + "start": "00:08:55", + "end": "00:09:02" + } + }, + { + "description": "Your friend put the wine in the refrigerator.", + "timestamps": { + "start": "00:09:04", + "end": "00:09:08" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Your friend placed it in a cupboard.", + "B": "Your friend put it in a basket.", + "C": "You handed the wine to your friend, who set it down on the counter instead of refrigerating it.", + "D": "After you passed it to your friend, they kept it out and opened it rather than putting it in the refrigerator.", + "E": "The wine was handed to your friend, and then your friend placed it in the refrigerator." + }, + "answer_label": "E", + "answer_text": "The wine was handed to your friend, and then your friend placed it in the refrigerator.", + "q_id": "egolife_A1_D5_021-mcq_fixed_5", + "correct_answer": "The wine was handed to your friend, and then your friend placed it in the refrigerator.", + "original_options": { + "A": "Your friend placed it in a cupboard.", + "B": "Your friend put it in a basket.", + "C": "You handed the wine to your friend, who set it down on the counter instead of refrigerating it.", + "D": "After you passed it to your friend, they kept it out and opened it rather than putting it in the refrigerator.", + "E": "The wine was handed to your friend, and then your friend placed it in the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 21 + ] + }, + { + "video_uid": "egolife_A1_D5_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_021-mcq_fixed_6", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Based on my current position, which is closer to me, the computer or the sofa on the second floor?", + "answer": "The sofa is closer to you.", + "question_chinese": "根据我现在的位置，二楼的电脑和沙发哪个离我更近？", + "answer_chinese": "沙发离你更近。", + "evidence": [ + { + "description": "There are sofas on both sides of you.", + "timestamps": { + "start": "00:11:28", + "end": "00:11:30" + } + } + ], + "question_timestamps": "00:11:31", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The sofa is closer to you.", + "B": "The computer is closer to you." + }, + "answer_label": "A", + "answer_text": "The sofa is closer to you.", + "q_id": "egolife_A1_D5_021-mcq_fixed_6", + "correct_answer": "The sofa is closer to you.", + "original_options": { + "A": "The sofa is closer to you.", + "B": "The computer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D5_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_021-mcq_fixed_7", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Based on my current position, which is closer to me, the computer or the sofa on the second floor?", + "answer": "The computer is closer to you.", + "question_chinese": "根据我现在的位置，二楼的电脑和沙发哪个离我更近？", + "answer_chinese": "电脑离你更近。", + "evidence": [ + { + "description": "The computer in the corner of the living room on the second floor is closer to you.", + "timestamps": { + "start": "00:11:40", + "end": "00:11:41" + } + } + ], + "question_timestamps": "00:11:42", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The sofa on the second floor is closer to you.", + "B": "The computer is closer to you." + }, + "answer_label": "B", + "answer_text": "The computer is closer to you.", + "q_id": "egolife_A1_D5_021-mcq_fixed_7", + "correct_answer": "The computer is closer to you.", + "original_options": { + "A": "The sofa on the second floor is closer to you.", + "B": "The computer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D5_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_021_2_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kitchen knives are in my sink in total right now?", + "answer": "Right now, there are two kitchen knives in your sink, and the knives are in front of you.", + "question_chinese": "现在我的水槽里面一共有多少把菜刀？", + "answer_chinese": "现在你的水槽里面有两把菜刀，菜刀在你的正前方。", + "evidence": [ + { + "description": "There are two kitchen knives in the sink.", + "timestamps": { + "start": "00:10:12", + "end": "00:10:13" + } + } + ], + "question_timestamps": "00:10:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There is one kitchen knife in your sink.", + "B": "Right now, there are two kitchen knives in your sink, and the knives are in front of you.", + "C": "There are three kitchen knives in your sink.", + "D": "Right now, there are four kitchen knives in your sink, and they are positioned in front of you.", + "E": "At the moment, there are zero kitchen knives in your sink, even though you are facing the sink." + }, + "answer_label": "B", + "answer_text": "Right now, there are two kitchen knives in your sink, and the knives are in front of you.", + "q_id": "egolife_A1_D5_021_2_T1", + "correct_answer": "Right now, there are two kitchen knives in your sink, and the knives are in front of you.", + "original_options": { + "A": "There is one kitchen knife in your sink.", + "B": "Right now, there are two kitchen knives in your sink, and the knives are in front of you.", + "C": "There are three kitchen knives in your sink.", + "D": "Right now, there are four kitchen knives in your sink, and they are positioned in front of you.", + "E": "At the moment, there are zero kitchen knives in your sink, even though you are facing the sink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_021_2_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kitchen knives are in my sink in total right now?", + "answer": "Right now, there is one kitchen knife in your sink, and the knife is to your right.", + "question_chinese": "现在我的水槽里面一共有多少把菜刀？", + "answer_chinese": "现在你的水槽里面有一把菜刀，菜刀在你的右边。", + "evidence": [ + { + "description": "There are two kitchen knives in the sink.", + "timestamps": { + "start": "00:10:12", + "end": "00:10:13" + } + }, + { + "description": "You moved one kitchen knife from the sink to the countertop.", + "timestamps": { + "start": "00:10:19", + "end": "00:10:23" + } + } + ], + "question_timestamps": "00:10:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are two kitchen knives in your sink.", + "B": "There are zero kitchen knives in your sink.", + "C": "Right now, there is one kitchen knife in your sink, and the knife is to your right.", + "D": "Right now, there are three kitchen knives in your sink, and the knives are to your right side.", + "E": "At the moment, there are four kitchen knives in your sink, and they are located to your right." + }, + "answer_label": "C", + "answer_text": "Right now, there is one kitchen knife in your sink, and the knife is to your right.", + "q_id": "egolife_A1_D5_021_2_T2", + "correct_answer": "Right now, there is one kitchen knife in your sink, and the knife is to your right.", + "original_options": { + "A": "There are two kitchen knives in your sink.", + "B": "There are zero kitchen knives in your sink.", + "C": "Right now, there is one kitchen knife in your sink, and the knife is to your right.", + "D": "Right now, there are three kitchen knives in your sink, and the knives are to your right side.", + "E": "At the moment, there are four kitchen knives in your sink, and they are located to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_021_7", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kitchen knives are there in the sink right now?", + "answer": "There are two kitchen knives in the sink now.", + "question_chinese": "现在水槽里面一共有多少把菜刀？", + "answer_chinese": "水槽里面现在有两把菜刀。", + "evidence": [ + { + "description": "There are two kitchen knives in the sink.", + "timestamps": { + "start": "00:10:12", + "end": "00:10:13" + } + } + ], + "question_timestamps": "00:10:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one kitchen knife in the sink.", + "B": "There are two kitchen knives in the sink now.", + "C": "There are three kitchen knives in the sink.", + "D": "There are four kitchen knives in the sink right now, lying together in the basin.", + "E": "There are five kitchen knives in the sink now, with several blades visible near the drain." + }, + "answer_label": "B", + "answer_text": "There are two kitchen knives in the sink now.", + "q_id": "egolife_A1_D5_021_7", + "correct_answer": "There are two kitchen knives in the sink now.", + "original_options": { + "A": "There is one kitchen knife in the sink.", + "B": "There are two kitchen knives in the sink now.", + "C": "There are three kitchen knives in the sink.", + "D": "There are four kitchen knives in the sink right now, lying together in the basin.", + "E": "There are five kitchen knives in the sink now, with several blades visible near the drain." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_021_8", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many kitchen knives are there in the sink right now?", + "answer": "There are two kitchen knives in the sink now.", + "question_chinese": "现在水槽里面一共有多少把菜刀？", + "answer_chinese": "水槽里面现在有一把菜刀", + "evidence": [ + { + "description": "There are two kitchen knives in the sink.", + "timestamps": { + "start": "00:10:12", + "end": "00:10:13" + } + }, + { + "description": "You moved one kitchen knife from the sink to the countertop.", + "timestamps": { + "start": "00:10:19", + "end": "00:10:21" + } + } + ], + "question_timestamps": "00:10:35", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There is one kitchen knife in the sink.", + "B": "There are three kitchen knives in the sink.", + "C": "There are four kitchen knives in the sink right now, sitting side by side in the basin.", + "D": "There are five kitchen knives in the sink now, clustered together near the bottom of the sink.", + "E": "There are two kitchen knives in the sink now." + }, + "answer_label": "E", + "answer_text": "There are two kitchen knives in the sink now.", + "q_id": "egolife_A1_D5_021_8", + "correct_answer": "There are two kitchen knives in the sink now.", + "original_options": { + "A": "There is one kitchen knife in the sink.", + "B": "There are three kitchen knives in the sink.", + "C": "There are four kitchen knives in the sink right now, sitting side by side in the basin.", + "D": "There are five kitchen knives in the sink now, clustered together near the bottom of the sink.", + "E": "There are two kitchen knives in the sink now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D5_022-mcq.json b/QAs/egolife_A1_D5_022-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..00263c1ab0ae93c82ea2ac1b991e8df43106e661 --- /dev/null +++ b/QAs/egolife_A1_D5_022-mcq.json @@ -0,0 +1,587 @@ +[ + { + "video_uid": "egolife_A1_D5_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_022-mcq_fixed_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement trajectory of the mango on the kitchen countertop until now?", + "answer": "You placed the mango from the kitchen countertop onto the table in the living room.", + "question_chinese": "厨房操作台上的芒果，到现在的运动轨迹是怎样的？", + "answer_chinese": "芒果被你从厨房操作台放到了客厅的桌子上。", + "evidence": [ + { + "description": "You picked up a mango from the kitchen countertop.", + "timestamps": { + "start": "00:06:50", + "end": "00:06:52" + } + }, + { + "description": "You placed the mango in your hand onto the table in the living room.", + "timestamps": { + "start": "00:07:10", + "end": "00:07:12" + } + } + ], + "question_timestamps": "00:07:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You left the mango on the countertop.", + "B": "You placed the mango from the kitchen countertop onto the table in the living room.", + "C": "You moved the mango to the kitchen table.", + "D": "You picked up the mango from the countertop and set it down again in the same spot.", + "E": "You moved the mango off the countertop and placed it on a chair in the living room instead." + }, + "answer_label": "B", + "answer_text": "You placed the mango from the kitchen countertop onto the table in the living room.", + "q_id": "egolife_A1_D5_022-mcq_fixed_6", + "correct_answer": "You placed the mango from the kitchen countertop onto the table in the living room.", + "original_options": { + "A": "You left the mango on the countertop.", + "B": "You placed the mango from the kitchen countertop onto the table in the living room.", + "C": "You moved the mango to the kitchen table.", + "D": "You picked up the mango from the countertop and set it down again in the same spot.", + "E": "You moved the mango off the countertop and placed it on a chair in the living room instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_022-mcq_fixed_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement trajectory of the mango on the kitchen countertop until now?", + "answer": "The mango was placed from the kitchen countertop onto the table in the living room, and then you picked it up and put it down again.", + "question_chinese": "厨房操作台上的芒果，到现在的运动轨迹是怎样的？", + "answer_chinese": "芒果被你从厨房操作台放到了客厅的桌子上，然后你将其拿起后又放下。", + "evidence": [ + { + "description": "You picked up a mango from the kitchen countertop.", + "timestamps": { + "start": "00:06:50", + "end": "00:06:52" + } + }, + { + "description": "You placed the mango in your hand onto the table in the living room.", + "timestamps": { + "start": "00:07:10", + "end": "00:07:12" + } + }, + { + "description": "You picked up the mango from the table in the living room and put it down again.", + "timestamps": { + "start": "00:07:20", + "end": "00:07:25" + } + } + ], + "question_timestamps": "00:07:27", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "It stayed on the countertop the whole time.", + "B": "The mango was placed from the kitchen countertop onto the table in the living room, and then you picked it up and put it down again.", + "C": "You moved it to the refrigerator and back.", + "D": "The mango was placed from the kitchen countertop onto a chair in the living room, then you picked it up again.", + "E": "The mango was moved from the kitchen countertop onto a side table in the living room, and later set down again." + }, + "answer_label": "B", + "answer_text": "The mango was placed from the kitchen countertop onto the table in the living room, and then you picked it up and put it down again.", + "q_id": "egolife_A1_D5_022-mcq_fixed_7", + "correct_answer": "The mango was placed from the kitchen countertop onto the table in the living room, and then you picked it up and put it down again.", + "original_options": { + "A": "It stayed on the countertop the whole time.", + "B": "The mango was placed from the kitchen countertop onto the table in the living room, and then you picked it up and put it down again.", + "C": "You moved it to the refrigerator and back.", + "D": "The mango was placed from the kitchen countertop onto a chair in the living room, then you picked it up again.", + "E": "The mango was moved from the kitchen countertop onto a side table in the living room, and later set down again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D5_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_022_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many egg tart shells do I have in total?", + "answer": "There are a total of four egg tart shells on your hand.", + "question_chinese": "我手上一共放了多少个蛋挞皮？", + "answer_chinese": "你的手上一共放了四个蛋挞皮。", + "evidence": [ + { + "description": "Your friend has placed 4 egg tart shells on your hand.", + "timestamps": { + "start": "00:03:32", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:03:54", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one egg tart shell total.", + "B": "There are three egg tart shells total.", + "C": "There are two egg tart shells on your hand in total, held together in your palm.", + "D": "There are a total of four egg tart shells on your hand.", + "E": "There are five egg tart shells on your hand in total, stacked and carried at the same time." + }, + "answer_label": "D", + "answer_text": "There are a total of four egg tart shells on your hand.", + "q_id": "egolife_A1_D5_022_5", + "correct_answer": "There are a total of four egg tart shells on your hand.", + "original_options": { + "A": "There is one egg tart shell total.", + "B": "There are three egg tart shells total.", + "C": "There are two egg tart shells on your hand in total, held together in your palm.", + "D": "There are a total of four egg tart shells on your hand.", + "E": "There are five egg tart shells on your hand in total, stacked and carried at the same time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_022_6", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many egg tart shells do I have in total?", + "answer": "There are a total of three egg tart shells on your hand.", + "question_chinese": "我手上一共放了多少个蛋挞皮？", + "answer_chinese": "你的手上一共放了3个蛋挞皮。", + "evidence": [ + { + "description": "Your friend has placed 4 egg tart shells on your hand.", + "timestamps": { + "start": "00:03:32", + "end": "00:03:43" + } + }, + { + "description": "Your friend took away one egg tart shell from your hand.", + "timestamps": { + "start": "00:03:43", + "end": "00:03:44" + } + } + ], + "question_timestamps": "00:04:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are two egg tart shells total.", + "B": "There are a total of three egg tart shells on your hand.", + "C": "There are four egg tart shells total.", + "D": "There are five egg tart shells on your hand in total, held together while you carry them.", + "E": "There are six egg tart shells on your hand in total, stacked and balanced as you hold them." + }, + "answer_label": "B", + "answer_text": "There are a total of three egg tart shells on your hand.", + "q_id": "egolife_A1_D5_022_6", + "correct_answer": "There are a total of three egg tart shells on your hand.", + "original_options": { + "A": "There are two egg tart shells total.", + "B": "There are a total of three egg tart shells on your hand.", + "C": "There are four egg tart shells total.", + "D": "There are five egg tart shells on your hand in total, held together while you carry them.", + "E": "There are six egg tart shells on your hand in total, stacked and balanced as you hold them." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_022-mcq_fixed_4", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Based on my current position, where is the glove?", + "answer": "The glove is on the table in front of you.", + "question_chinese": "根据我现在的位置，手套在我的什么方位？", + "answer_chinese": "手套在你面前的桌子上。", + "evidence": [ + { + "description": "The glove is on the table in front of you.", + "timestamps": { + "start": "00:02:03", + "end": "00:02:04" + } + } + ], + "question_timestamps": "00:02:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The glove is behind you.", + "B": "The glove is on your left.", + "C": "The glove is on the table behind you, near the far edge away from your hands.", + "D": "The glove is on the chair to your left, tucked against the backrest beside you.", + "E": "The glove is on the table in front of you." + }, + "answer_label": "E", + "answer_text": "The glove is on the table in front of you.", + "q_id": "egolife_A1_D5_022-mcq_fixed_4", + "correct_answer": "The glove is on the table in front of you.", + "original_options": { + "A": "The glove is behind you.", + "B": "The glove is on your left.", + "C": "The glove is on the table behind you, near the far edge away from your hands.", + "D": "The glove is on the chair to your left, tucked against the backrest beside you.", + "E": "The glove is on the table in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_022-mcq_fixed_5", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Based on my current position, where is the glove?", + "answer": "The glove was put on your right hand.", + "question_chinese": "根据我现在的位置，手套在我的什么方位？", + "answer_chinese": "手套被你戴到了右手上。", + "evidence": [ + { + "description": "The glove was put on your right hand.", + "timestamps": { + "start": "00:02:26", + "end": "00:02:28" + } + } + ], + "question_timestamps": "00:02:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The glove was put on your left hand.", + "B": "The glove was put on your right hand.", + "C": "The glove is tucked into your left pocket.", + "D": "You left the glove on the chair behind you after moving away from where you first picked it up.", + "E": "The glove was placed on the floor by your feet, rather than being worn on your hand." + }, + "answer_label": "B", + "answer_text": "The glove was put on your right hand.", + "q_id": "egolife_A1_D5_022-mcq_fixed_5", + "correct_answer": "The glove was put on your right hand.", + "original_options": { + "A": "The glove was put on your left hand.", + "B": "The glove was put on your right hand.", + "C": "The glove is tucked into your left pocket.", + "D": "You left the glove on the chair behind you after moving away from where you first picked it up.", + "E": "The glove was placed on the floor by your feet, rather than being worn on your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_022_1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "At this moment, where did I put the backpack on the table?", + "answer": "The schoolbag was placed beside the entrance of the living room by you.", + "question_chinese": "此时此刻，桌子上的书包被我放在了哪里？", + "answer_chinese": "书包被你放在了客厅的入口旁。", + "evidence": [ + { + "description": "You threw the schoolbag on the table and placed it beside the entrance of the living room.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Near the kitchen counter.", + "B": "The schoolbag was placed beside the entrance of the living room by you.", + "C": "In front of the television.", + "D": "The schoolbag was tossed onto a dining-room chair, not placed on the table.", + "E": "The schoolbag was dropped on the hallway floor instead of beside the living room entrance." + }, + "answer_label": "B", + "answer_text": "The schoolbag was placed beside the entrance of the living room by you.", + "q_id": "egolife_A1_D5_022_1", + "correct_answer": "The schoolbag was placed beside the entrance of the living room by you.", + "original_options": { + "A": "Near the kitchen counter.", + "B": "The schoolbag was placed beside the entrance of the living room by you.", + "C": "In front of the television.", + "D": "The schoolbag was tossed onto a dining-room chair, not placed on the table.", + "E": "The schoolbag was dropped on the hallway floor instead of beside the living room entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_022_2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "At this moment, where did I put the backpack on the table?", + "answer": "The schoolbag was hung on the blackboard in the yard.", + "question_chinese": "此时此刻，桌子上的书包被我放在了哪里？", + "answer_chinese": "书包背挂在了院子里的黑板上。", + "evidence": [ + { + "description": "You threw the schoolbag on the table and placed it beside the entrance of the living room.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:03" + } + }, + { + "description": "You picked up the schoolbag and took it to the yard.", + "timestamps": { + "start": "00:09:12", + "end": "00:09:20" + } + }, + { + "description": "You hung the schoolbag on the blackboard in the yard.", + "timestamps": { + "start": "00:09:48", + "end": "00:09:51" + } + } + ], + "question_timestamps": "00:09:56", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The schoolbag was placed under the desk.", + "B": "The schoolbag was left beside the sofa.", + "C": "The schoolbag was found in the hallway near the stairs rather than being hung up in the yard.", + "D": "The schoolbag was dropped in the corner of the cafeteria instead of being hung on the blackboard outside.", + "E": "The schoolbag was hung on the blackboard in the yard." + }, + "answer_label": "E", + "answer_text": "The schoolbag was hung on the blackboard in the yard.", + "q_id": "egolife_A1_D5_022_2", + "correct_answer": "The schoolbag was hung on the blackboard in the yard.", + "original_options": { + "A": "The schoolbag was placed under the desk.", + "B": "The schoolbag was left beside the sofa.", + "C": "The schoolbag was found in the hallway near the stairs rather than being hung up in the yard.", + "D": "The schoolbag was dropped in the corner of the cafeteria instead of being hung on the blackboard outside.", + "E": "The schoolbag was hung on the blackboard in the yard." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_022-mcq_fixed_8", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment, which is closer to me, the landline clothes dryer or the courier service?", + "answer": "The package is closer to you.", + "question_chinese": "此时此刻，置地晾衣架和快递哪个离我更近？", + "answer_chinese": "快递离你更近。", + "evidence": [ + { + "description": "The courier on the table in the yard is closer to you.", + "timestamps": { + "start": "00:10:08", + "end": "00:10:10" + } + }, + { + "description": "You are sitting at the table in the yard.", + "timestamps": { + "start": "00:10:19", + "end": "00:10:21" + } + } + ], + "question_timestamps": "00:10:22", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The package is closer to you.", + "B": "The landline clothes dryer is closer to you." + }, + "answer_label": "A", + "answer_text": "The package is closer to you.", + "q_id": "egolife_A1_D5_022-mcq_fixed_8", + "correct_answer": "The package is closer to you.", + "original_options": { + "A": "The package is closer to you.", + "B": "The landline clothes dryer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D5_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_022-mcq_fixed_9", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment, which is closer to me, the landline clothes dryer or the courier service?", + "answer": "The landline clothes dryer is closer to you.", + "question_chinese": "此时此刻，置地晾衣架和快递哪个离我更近？", + "answer_chinese": "置地晾衣架离你更近。", + "evidence": [ + { + "description": "You are beside the landline clothes dryer at the door.", + "timestamps": { + "start": "00:11:42", + "end": "00:11:44" + } + } + ], + "question_timestamps": "00:11:45", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The courier service is closer to you.", + "B": "The landline clothes dryer is closer to you." + }, + "answer_label": "B", + "answer_text": "The landline clothes dryer is closer to you.", + "q_id": "egolife_A1_D5_022-mcq_fixed_9", + "correct_answer": "The landline clothes dryer is closer to you.", + "original_options": { + "A": "The courier service is closer to you.", + "B": "The landline clothes dryer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + } +] diff --git a/QAs/egolife_A1_D5_024-mcq.json b/QAs/egolife_A1_D5_024-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fa4882dd8eb19ba7e15ae7601807f0f022094914 --- /dev/null +++ b/QAs/egolife_A1_D5_024-mcq.json @@ -0,0 +1,601 @@ +[ + { + "video_uid": "egolife_A1_D5_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_024_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many three-wheeled electric vehicles have I seen so far?", + "answer": "You saw one tricycle.", + "question_chinese": "到现在为止我一共看到了几辆三轮电动车？", + "answer_chinese": "你看到了一辆三轮电动车。", + "evidence": [ + { + "description": "There was a tricycle parked in front of you.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:28" + } + } + ], + "question_timestamps": "00:03:39", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You saw two tricycles.", + "B": "You saw three tricycles.", + "C": "You saw four tricycles in total as you continued moving through the area.", + "D": "You saw five tricycles altogether, counting each three-wheeled electric vehicle you passed.", + "E": "You saw one tricycle." + }, + "answer_label": "E", + "answer_text": "You saw one tricycle.", + "q_id": "egolife_A1_D5_024_3", + "correct_answer": "You saw one tricycle.", + "original_options": { + "A": "You saw two tricycles.", + "B": "You saw three tricycles.", + "C": "You saw four tricycles in total as you continued moving through the area.", + "D": "You saw five tricycles altogether, counting each three-wheeled electric vehicle you passed.", + "E": "You saw one tricycle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D5_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_024_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many three-wheeled electric vehicles have I seen so far?", + "answer": "You saw one tricycle.", + "question_chinese": "到现在为止我一共看到了几辆三轮电动车？", + "answer_chinese": "你看到了两辆三轮电动车", + "evidence": [ + { + "description": "There was a tricycle parked in front of you.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:28" + } + }, + { + "description": "There was a tricycle parked beside the pavilion in front of you.", + "timestamps": { + "start": "00:04:26", + "end": "00:04:28" + } + } + ], + "question_timestamps": "00:04:40", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You saw two tricycles.", + "B": "You saw three tricycles.", + "C": "You saw four tricycles in total during the time you were watching the street.", + "D": "You saw five tricycles altogether, adding up every three-wheeled electric vehicle you noticed.", + "E": "You saw one tricycle." + }, + "answer_label": "E", + "answer_text": "You saw one tricycle.", + "q_id": "egolife_A1_D5_024_4", + "correct_answer": "You saw one tricycle.", + "original_options": { + "A": "You saw two tricycles.", + "B": "You saw three tricycles.", + "C": "You saw four tricycles in total during the time you were watching the street.", + "D": "You saw five tricycles altogether, adding up every three-wheeled electric vehicle you noticed.", + "E": "You saw one tricycle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D5_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_024-mcq_fixed_8", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At the present moment, which one is closer to me among the front desk and the tables in the Yangtou Restaurant?", + "answer": "You are closer to the table inside.", + "question_chinese": "当前时刻，羊汤馆里的前台和桌子那个离我近？", + "answer_chinese": "桌子里你更近。", + "evidence": [ + { + "description": "The table in the Yangtou Restaurant is closer to you.", + "timestamps": { + "start": "00:11:30", + "end": "00:11:32" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are closer to the front desk.", + "B": "You are closer to the table inside.", + "C": "You are closer to the table outside.", + "D": "You are closer to the front desk than the tables in the Yangtou Restaurant at this moment.", + "E": "You are closer to the front desk area, with the tables positioned farther away from you right now." + }, + "answer_label": "B", + "answer_text": "You are closer to the table inside.", + "q_id": "egolife_A1_D5_024-mcq_fixed_8", + "correct_answer": "You are closer to the table inside.", + "original_options": { + "A": "You are closer to the front desk.", + "B": "You are closer to the table inside.", + "C": "You are closer to the table outside.", + "D": "You are closer to the front desk than the tables in the Yangtou Restaurant at this moment.", + "E": "You are closer to the front desk area, with the tables positioned farther away from you right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_024-mcq_fixed_9", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At the present moment, which one is closer to me among the front desk and the tables in the Yangtou Restaurant?", + "answer": "The front desk is closer to you.", + "question_chinese": "当前时刻，羊汤馆里的前台和桌子那个离我近？", + "answer_chinese": "前台离你更近", + "evidence": [ + { + "description": "The front desk in the Yangtou Restaurant is closer to you.", + "timestamps": { + "start": "00:11:40", + "end": "00:11:42" + } + } + ], + "question_timestamps": "00:11:43", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The front desk is closer to you.", + "B": "The tables are closer to you." + }, + "answer_label": "A", + "answer_text": "The front desk is closer to you.", + "q_id": "egolife_A1_D5_024-mcq_fixed_9", + "correct_answer": "The front desk is closer to you.", + "original_options": { + "A": "The front desk is closer to you.", + "B": "The tables are closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D5_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_024_1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is there a trash can to my right?", + "answer": "There is a trash can on your right.", + "question_chinese": "我现在右手边有垃圾桶吗？", + "answer_chinese": "你的右手边有垃圾桶。", + "evidence": [ + { + "description": "The trash can is on your right.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:50" + } + } + ], + "question_timestamps": "00:00:54", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "There is no trash can on your right.", + "B": "There is a trash can on your right." + }, + "answer_label": "B", + "answer_text": "There is a trash can on your right.", + "q_id": "egolife_A1_D5_024_1", + "correct_answer": "There is a trash can on your right.", + "original_options": { + "A": "There is no trash can on your right.", + "B": "There is a trash can on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D5_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_024_2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Is there a trash can to my right?", + "answer": "The trash can is on your right.", + "question_chinese": "我现在右手边有垃圾桶吗？", + "answer_chinese": "垃圾桶你的右手边。", + "evidence": [ + { + "description": "There is a trash can on your right.", + "timestamps": { + "start": "00:01:18", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:01:24", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The trash can is on your right.", + "B": "The trash can is on your left." + }, + "answer_label": "A", + "answer_text": "The trash can is on your right.", + "q_id": "egolife_A1_D5_024_2", + "correct_answer": "The trash can is on your right.", + "original_options": { + "A": "The trash can is on your right.", + "B": "The trash can is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D5_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_024-mcq_fixed_3", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Do I have a trash can on my right now?", + "answer": "There is no trash can on your right.", + "question_chinese": "我现在右手边有垃圾桶吗？", + "answer_chinese": "你的右手边没有垃圾桶。", + "evidence": [ + { + "description": "There is no trash can on your right.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:01:26", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "There is a trash can on your right.", + "B": "There is no trash can on your right." + }, + "answer_label": "B", + "answer_text": "There is no trash can on your right.", + "q_id": "egolife_A1_D5_024-mcq_fixed_3", + "correct_answer": "There is no trash can on your right.", + "original_options": { + "A": "There is a trash can on your right.", + "B": "There is no trash can on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D5_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_024-mcq_fixed_6", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "At this moment, what is around me?", + "answer": "There were many trees, cars and houses around you.", + "question_chinese": "此时此刻，我的周围有什么？", + "answer_chinese": "你的周围有许多树，车和居民楼。", + "evidence": [ + { + "description": "There were many trees, cars and houses around you.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:38" + } + } + ], + "question_timestamps": "00:05:39", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Few clouds are above you.", + "B": "There were many trees, cars and houses around you.", + "C": "Some bikes are in front of you.", + "D": "There were clouds, birds, and planes above you rather than trees, cars, and houses nearby.", + "E": "The surroundings looked like rocks, bikes, and boats in front of you, not trees and houses." + }, + "answer_label": "B", + "answer_text": "There were many trees, cars and houses around you.", + "q_id": "egolife_A1_D5_024-mcq_fixed_6", + "correct_answer": "There were many trees, cars and houses around you.", + "original_options": { + "A": "Few clouds are above you.", + "B": "There were many trees, cars and houses around you.", + "C": "Some bikes are in front of you.", + "D": "There were clouds, birds, and planes above you rather than trees, cars, and houses nearby.", + "E": "The surroundings looked like rocks, bikes, and boats in front of you, not trees and houses." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_024-mcq_fixed_7", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "At this moment, what is around me?", + "answer": "There are many cars and residential buildings around you, as well as a tower-shaped building.", + "question_chinese": "此时此刻，我的周围有什么？", + "answer_chinese": "你的周围有许多车和居民楼，还有一座塔状建筑", + "evidence": [ + { + "description": "There are many cars and residential buildings around you, as well as a tower-shaped building.", + "timestamps": { + "start": "00:09:45", + "end": "00:09:47" + } + } + ], + "question_timestamps": "00:09:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Only parked motorcycles are nearby.", + "B": "Buses and office buildings surround you.", + "C": "There are many bicycles and commercial buildings around you, along with a dome-shaped building.", + "D": "There are many motorcycles and industrial buildings behind you, along with a pyramid-shaped building.", + "E": "There are many cars and residential buildings around you, as well as a tower-shaped building." + }, + "answer_label": "E", + "answer_text": "There are many cars and residential buildings around you, as well as a tower-shaped building.", + "q_id": "egolife_A1_D5_024-mcq_fixed_7", + "correct_answer": "There are many cars and residential buildings around you, as well as a tower-shaped building.", + "original_options": { + "A": "Only parked motorcycles are nearby.", + "B": "Buses and office buildings surround you.", + "C": "There are many bicycles and commercial buildings around you, along with a dome-shaped building.", + "D": "There are many motorcycles and industrial buildings behind you, along with a pyramid-shaped building.", + "E": "There are many cars and residential buildings around you, as well as a tower-shaped building." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D5_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_024-mcq_fixed_4", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory like since the video started?", + "answer": "Walk straight. First, turn left at the intersection where you see the pavilion. Then, turn right at the first crossroads.", + "question_chinese": "视频开始到现在，我的运动轨迹是怎样的？", + "answer_chinese": "你直走，首先在看到亭子的路口左转，接着在第一个十字路口右转。", + "evidence": [ + { + "description": "You turn left at an intersection with a pavilion.", + "timestamps": { + "start": "00:04:26", + "end": "00:04:32" + } + }, + { + "description": "You turn right at the crossroads.", + "timestamps": { + "start": "00:06:45", + "end": "00:06:52" + } + } + ], + "question_timestamps": "00:06:53", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right, then left.", + "B": "Turn left, then left.", + "C": "Walk straight, turn right at the pavilion intersection, then turn left at the first crossroads you reach.", + "D": "Walk straight. First, turn left at the intersection where you see the pavilion. Then, turn right at the first crossroads.", + "E": "Walk straight until you see the pavilion, turn left there, then turn right at the next crossroads afterward." + }, + "answer_label": "D", + "answer_text": "Walk straight. First, turn left at the intersection where you see the pavilion. Then, turn right at the first crossroads.", + "q_id": "egolife_A1_D5_024-mcq_fixed_4", + "correct_answer": "Walk straight. First, turn left at the intersection where you see the pavilion. Then, turn right at the first crossroads.", + "original_options": { + "A": "Turn right, then left.", + "B": "Turn left, then left.", + "C": "Walk straight, turn right at the pavilion intersection, then turn left at the first crossroads you reach.", + "D": "Walk straight. First, turn left at the intersection where you see the pavilion. Then, turn right at the first crossroads.", + "E": "Walk straight until you see the pavilion, turn left there, then turn right at the next crossroads afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_024-mcq_fixed_5", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory like since the video started?", + "answer": "Walk straight. First, turn left at the intersection where you see the pavilion, then turn right at the first crossroads, and continue walking straight.", + "question_chinese": "视频开始到现在，我的运动轨迹是怎样的？", + "answer_chinese": "你直走，首先在看到亭子的路口左转，接着在第一个十字路口右转，并继续直走。", + "evidence": [ + { + "description": "You turn left at an intersection with a pavilion.", + "timestamps": { + "start": "00:04:26", + "end": "00:04:32" + } + }, + { + "description": "You turn right at the crossroads.", + "timestamps": { + "start": "00:06:45", + "end": "00:06:52" + } + }, + { + "description": "You keep walking straight and saw a tower-shaped building.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:30" + } + } + ], + "question_timestamps": "00:09:31", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Walk straight. First, turn left at the intersection where you see the pavilion, then turn right at the first crossroads, and continue walking straight.", + "B": "Turn left, then left.", + "C": "Turn right, then left.", + "D": "Head toward the pavilion, turn right just before it, then go straight until the second intersection.", + "E": "Walk straight to the pavilion, turn left, pass the first crossroads, then turn right at the next junction." + }, + "answer_label": "A", + "answer_text": "Walk straight. First, turn left at the intersection where you see the pavilion, then turn right at the first crossroads, and continue walking straight.", + "q_id": "egolife_A1_D5_024-mcq_fixed_5", + "correct_answer": "Walk straight. First, turn left at the intersection where you see the pavilion, then turn right at the first crossroads, and continue walking straight.", + "original_options": { + "A": "Walk straight. First, turn left at the intersection where you see the pavilion, then turn right at the first crossroads, and continue walking straight.", + "B": "Turn left, then left.", + "C": "Turn right, then left.", + "D": "Head toward the pavilion, turn right just before it, then go straight until the second intersection.", + "E": "Walk straight to the pavilion, turn left, pass the first crossroads, then turn right at the next junction." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D5_025-mcq.json b/QAs/egolife_A1_D5_025-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..27ab40a67995e0a57b76a581625df9106662db9e --- /dev/null +++ b/QAs/egolife_A1_D5_025-mcq.json @@ -0,0 +1,648 @@ +[ + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025_1_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many scrapers do my friend and I have in total for scratching lottery tickets?", + "answer": "There is 1 shovel in total. The shovel is in the hands of the friend facing you.", + "question_chinese": "到目前为止，我和我的朋友刮彩票一共有几个铲子？", + "answer_chinese": "一共一个铲子。铲子在你对面的朋友手上。", + "evidence": [ + { + "description": "Your friend is using a metal shovel for the lottery drawing.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:17" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 0 shovels in total.", + "B": "There are 2 shovels in total.", + "C": "There are 3 shovels in total, and the shovels are in the hands of the friend facing you.", + "D": "There are 4 shovels in total, and the shovels are in the hands of the friend facing you.", + "E": "There is 1 shovel in total. The shovel is in the hands of the friend facing you." + }, + "answer_label": "E", + "answer_text": "There is 1 shovel in total. The shovel is in the hands of the friend facing you.", + "q_id": "egolife_A1_D5_025_1_T1", + "correct_answer": "There is 1 shovel in total. The shovel is in the hands of the friend facing you.", + "original_options": { + "A": "There are 0 shovels in total.", + "B": "There are 2 shovels in total.", + "C": "There are 3 shovels in total, and the shovels are in the hands of the friend facing you.", + "D": "There are 4 shovels in total, and the shovels are in the hands of the friend facing you.", + "E": "There is 1 shovel in total. The shovel is in the hands of the friend facing you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025_1_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many scrapers do my friend and I have in total for scratching lottery tickets?", + "answer": "You have 2 shovels in total. The boy to your left and the blonde girl each has one.", + "question_chinese": "到目前为止，我和我的朋友刮彩票一共有几个铲子？", + "answer_chinese": "你们一共2个铲子。你左边的男生和黄头发女生每人拿了一个。", + "evidence": [ + { + "description": "Your friend is using two shovel for the lottery drawing.", + "timestamps": { + "start": "00:02:48", + "end": "00:02:55" + } + } + ], + "question_timestamps": "00:02:58", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have 2 shovels in total. The boy to your left and the blonde girl each has one.", + "B": "You have 1 shovel in total.", + "C": "You have 3 shovels in total.", + "D": "You have 1 shovel in total. Only the boy to your left has one, and the blonde girl has none.", + "E": "You have 4 shovels in total. The boy and blonde girl each has one, plus two more with you." + }, + "answer_label": "A", + "answer_text": "You have 2 shovels in total. The boy to your left and the blonde girl each has one.", + "q_id": "egolife_A1_D5_025_1_T2", + "correct_answer": "You have 2 shovels in total. The boy to your left and the blonde girl each has one.", + "original_options": { + "A": "You have 2 shovels in total. The boy to your left and the blonde girl each has one.", + "B": "You have 1 shovel in total.", + "C": "You have 3 shovels in total.", + "D": "You have 1 shovel in total. Only the boy to your left has one, and the blonde girl has none.", + "E": "You have 4 shovels in total. The boy and blonde girl each has one, plus two more with you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025_6", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many scratch-off tickets have my friends and I scratched?", + "answer": "There were two shovels in total.", + "question_chinese": "到目前为止，我和我的朋友刮彩票一共有几个铲子？", + "answer_chinese": "一共两个铲子。", + "evidence": [ + { + "description": "Your friend is using a metal shovel for the lottery drawing.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:17" + } + }, + { + "description": "One of your friends brought a plastic shovel.", + "timestamps": { + "start": "00:02:48", + "end": "00:02:51" + } + } + ], + "question_timestamps": "00:03:02", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There was one shovel in total.", + "B": "There were three shovels in total.", + "C": "There were two shovels in total.", + "D": "There were four shovels in total, counted after everyone finished scratching their tickets.", + "E": "There were five shovels in total, with several laid out together as the group worked." + }, + "answer_label": "C", + "answer_text": "There were two shovels in total.", + "q_id": "egolife_A1_D5_025_6", + "correct_answer": "There were two shovels in total.", + "original_options": { + "A": "There was one shovel in total.", + "B": "There were three shovels in total.", + "C": "There were two shovels in total.", + "D": "There were four shovels in total, counted after everyone finished scratching their tickets.", + "E": "There were five shovels in total, with several laid out together as the group worked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025-mcq_fixed_0", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people in total bought the lottery tickets with me?", + "answer": "Three.", + "question_chinese": "我和我的朋友一共有多少人买了彩票？", + "answer_chinese": "一共五个人买了彩票。", + "evidence": [ + { + "description": "Your three friends bought lottery tickets.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:14" + } + }, + { + "description": "The other two of your friends bought lottery tickets.", + "timestamps": { + "start": "00:06:16", + "end": "00:06:18" + } + } + ], + "question_timestamps": "00:06:19", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Five people bought the lottery tickets with you in total, counting everyone who was there at the counter.", + "D": "Three.", + "E": "Six people bought the lottery tickets with you in total, including you and the friends standing beside you." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "egolife_A1_D5_025-mcq_fixed_0", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Five people bought the lottery tickets with you in total, counting everyone who was there at the counter.", + "D": "Three.", + "E": "Six people bought the lottery tickets with you in total, including you and the friends standing beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025-mcq_fixed_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people in total bought the lottery tickets with me?", + "answer": "Six.", + "question_chinese": "我和我的朋友一共有多少人买了彩票？", + "answer_chinese": "一共六个人了买彩票。", + "evidence": [ + { + "description": "Your three friends bought lottery tickets.", + "timestamps": { + "start": "00:02:12", + "end": "00:02:14" + } + }, + { + "description": "The other two of your friends bought lottery tickets.", + "timestamps": { + "start": "00:06:16", + "end": "00:06:18" + } + }, + { + "description": "You bought one lottery ticket.", + "timestamps": { + "start": "00:07:05", + "end": "00:07:08" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "Four people bought the lottery tickets with you in total, so there were fewer people than you first thought.", + "D": "Six.", + "E": "Eight people bought the lottery tickets with you in total, making the group larger than the one you remember." + }, + "answer_label": "D", + "answer_text": "Six.", + "q_id": "egolife_A1_D5_025-mcq_fixed_1", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "Four people bought the lottery tickets with you in total, so there were fewer people than you first thought.", + "D": "Six.", + "E": "Eight people bought the lottery tickets with you in total, making the group larger than the one you remember." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025-mcq_fixed_4", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "At this moment, can I get a pencil?", + "answer": "You can't reach the pencil.", + "question_chinese": "此时此刻，我能够够得到铅笔吗？", + "answer_chinese": "你够不到铅笔。", + "evidence": [ + { + "description": "You got up and left the table. Now you can't reach the pencil on the table.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:36" + } + } + ], + "question_timestamps": "00:06:37", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "You can't reach the pencil.", + "B": "You can reach the pencil." + }, + "answer_label": "A", + "answer_text": "You can't reach the pencil.", + "q_id": "egolife_A1_D5_025-mcq_fixed_4", + "correct_answer": "You can't reach the pencil.", + "original_options": { + "A": "You can't reach the pencil.", + "B": "You can reach the pencil." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025_5_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "At this moment, can I reach the pencil?", + "answer": "You can’t reach the pencil. The pencil is on the table, and you are to the left of the table.", + "question_chinese": "此时此刻，我能够够得到铅笔吗？", + "answer_chinese": "你够不到铅笔。铅笔在桌子上，而你在桌子的左边", + "evidence": [ + { + "description": "You got up and left the table. Now you can't reach the pencil on the table.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:36" + } + } + ], + "question_timestamps": "00:06:38", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the pencil. The pencil is on the table, and you are to the left of the table.", + "B": "You can’t reach the pencil. The pencil is on the table, and you are to the left of the table." + }, + "answer_label": "B", + "answer_text": "You can’t reach the pencil. The pencil is on the table, and you are to the left of the table.", + "q_id": "egolife_A1_D5_025_5_T1", + "correct_answer": "You can’t reach the pencil. The pencil is on the table, and you are to the left of the table.", + "original_options": { + "A": "You can reach the pencil. The pencil is on the table, and you are to the left of the table.", + "B": "You can’t reach the pencil. The pencil is on the table, and you are to the left of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025_5_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "At this moment, can I reach the pencil?", + "answer": "You can reach the pencil. The pencil is to the right of your right hand.", + "question_chinese": "此时此刻，我能够够得到铅笔吗？", + "answer_chinese": "你能够到铅笔。铅笔在你右手边。", + "evidence": [ + { + "description": "You are sitting by the table. The pencil is on your right hand side.", + "timestamps": { + "start": "00:07:20", + "end": "00:07:30" + } + } + ], + "question_timestamps": "00:07:35", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the pencil. The pencil is to the right of your right hand.", + "B": "You cannot reach the pencil. The pencil is to the right of your right hand." + }, + "answer_label": "A", + "answer_text": "You can reach the pencil. The pencil is to the right of your right hand.", + "q_id": "egolife_A1_D5_025_5_T2", + "correct_answer": "You can reach the pencil. The pencil is to the right of your right hand.", + "original_options": { + "A": "You can reach the pencil. The pencil is to the right of your right hand.", + "B": "You cannot reach the pencil. The pencil is to the right of your right hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025-mcq_fixed_5", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment, which is closer to me, the calculator or the trash can?", + "answer": "The calculator is closer to you.", + "question_chinese": "此时此刻，计算器和垃圾桶哪个离我更近？", + "answer_chinese": "计算器离你更近。", + "evidence": [ + { + "description": "The calculator on the counter is closer to you.", + "timestamps": { + "start": "00:08:21", + "end": "00:08:22" + } + } + ], + "question_timestamps": "00:08:23", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The trash can is closer to you.", + "B": "The calculator is closer to you." + }, + "answer_label": "B", + "answer_text": "The calculator is closer to you.", + "q_id": "egolife_A1_D5_025-mcq_fixed_5", + "correct_answer": "The calculator is closer to you.", + "original_options": { + "A": "The trash can is closer to you.", + "B": "The calculator is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025-mcq_fixed_6", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment, which is closer to me, the calculator or the trash can?", + "answer": "The trash can is closer to you.", + "question_chinese": "此时此刻，计算器和垃圾桶哪个离我更近？", + "answer_chinese": "垃圾桶离你更近", + "evidence": [ + { + "description": "The trash can beside the table is closer to you.", + "timestamps": { + "start": "00:09:19", + "end": "00:09:20" + } + } + ], + "question_timestamps": "00:09:21", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The trash can is closer to you.", + "B": "The calculator is closer to you." + }, + "answer_label": "A", + "answer_text": "The trash can is closer to you.", + "q_id": "egolife_A1_D5_025-mcq_fixed_6", + "correct_answer": "The trash can is closer to you.", + "original_options": { + "A": "The trash can is closer to you.", + "B": "The calculator is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025-mcq_fixed_7", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After leaving the store, what has been my movement trajectory so far?", + "answer": "You turn right, then go straight to the side of the supermarket, then turn left and go straight again, and then turn left again and go straight.", + "question_chinese": "走出商店后，到现在我的运动轨迹是怎样的？", + "answer_chinese": "你右转，然后直行到超市旁，然后再左转，并直行。", + "evidence": [ + { + "description": "After you leave the store, turn right and then go straight.", + "timestamps": { + "start": "00:09:24", + "end": "00:09:28" + } + }, + { + "description": "You turn left beside the supermarket and then go straight.", + "timestamps": { + "start": "00:10:20", + "end": "00:10:25" + } + } + ], + "question_timestamps": "00:10:26", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You turn left, then go straight.", + "B": "You turn right, then turn right again.", + "C": "You turn right, then go straight to the side of the supermarket, then turn left and go straight again, and then turn left again and go straight.", + "D": "You turn right, go straight along the supermarket, then turn right and continue straight to the next corner.", + "E": "You turn left after exiting, walk straight beside the supermarket, then turn right twice while continuing straight." + }, + "answer_label": "C", + "answer_text": "You turn right, then go straight to the side of the supermarket, then turn left and go straight again, and then turn left again and go straight.", + "q_id": "egolife_A1_D5_025-mcq_fixed_7", + "correct_answer": "You turn right, then go straight to the side of the supermarket, then turn left and go straight again, and then turn left again and go straight.", + "original_options": { + "A": "You turn left, then go straight.", + "B": "You turn right, then turn right again.", + "C": "You turn right, then go straight to the side of the supermarket, then turn left and go straight again, and then turn left again and go straight.", + "D": "You turn right, go straight along the supermarket, then turn right and continue straight to the next corner.", + "E": "You turn left after exiting, walk straight beside the supermarket, then turn right twice while continuing straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_025", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_025-mcq_fixed_8", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After leaving the store, what has been my movement trajectory so far?", + "answer": "You turn right, then go straight to the side of the supermarket, then turn left and go straight to the middle of the road, then turn left again and go straight.", + "question_chinese": "走出商店后，到现在我的运动轨迹是怎样的？", + "answer_chinese": "你右转，然后直行到超市旁，然后再左转，并直行到马路中间，接着左转身并直走。", + "evidence": [ + { + "description": "After you leave the store, turn right and then go straight.", + "timestamps": { + "start": "00:09:24", + "end": "00:09:28" + } + }, + { + "description": "You turn left beside the supermarket and then go straight.", + "timestamps": { + "start": "00:10:20", + "end": "00:10:25" + } + }, + { + "description": "You turn left twice in the middle of the road and then go straight.", + "timestamps": { + "start": "00:10:55", + "end": "00:11:00" + } + } + ], + "question_timestamps": "00:11:01", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You go straight, then turn left.", + "B": "You turn right, then turn right again.", + "C": "You turn right, go straight by the supermarket, then turn left and keep straight without turning left again.", + "D": "You turn right, then go straight to the side of the supermarket, then turn left and go straight to the middle of the road, then turn left again and go straight.", + "E": "You turn left after leaving, walk along the supermarket side, then turn right toward the middle of the road." + }, + "answer_label": "D", + "answer_text": "You turn right, then go straight to the side of the supermarket, then turn left and go straight to the middle of the road, then turn left again and go straight.", + "q_id": "egolife_A1_D5_025-mcq_fixed_8", + "correct_answer": "You turn right, then go straight to the side of the supermarket, then turn left and go straight to the middle of the road, then turn left again and go straight.", + "original_options": { + "A": "You go straight, then turn left.", + "B": "You turn right, then turn right again.", + "C": "You turn right, go straight by the supermarket, then turn left and keep straight without turning left again.", + "D": "You turn right, then go straight to the side of the supermarket, then turn left and go straight to the middle of the road, then turn left again and go straight.", + "E": "You turn left after leaving, walk along the supermarket side, then turn right toward the middle of the road." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + } +] diff --git a/QAs/egolife_A1_D5_026-mcq.json b/QAs/egolife_A1_D5_026-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e538ef55f6e09647a6650146b4b7e19f5caf1692 --- /dev/null +++ b/QAs/egolife_A1_D5_026-mcq.json @@ -0,0 +1,134 @@ +[ + { + "video_uid": "egolife_A1_D5_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_026_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the package reach your friend?", + "answer": "The express delivery moved from the shelf to the express delivery outbound device and then to your friend's hands.", + "question_chinese": "快递是怎么到你朋友手上的？", + "answer_chinese": "快递从货架上移动到了快递出库仪上，再到了你朋友的手上。", + "evidence": [ + { + "description": "The owner of the express delivery station found the package from the shelf.", + "timestamps": { + "start": "00:08:50", + "end": "00:08:54" + } + }, + { + "description": "The owner of the express delivery station handed the package over to your friend after taking it out of the warehouse.", + "timestamps": { + "start": "00:08:56", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:09:04", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Shelf straight to friend.", + "B": "Outbound device back to shelf.", + "C": "The express delivery moved from the shelf to the express delivery outbound device and then to your friend's hands.", + "D": "The express delivery moved from the shelf to your friend's hands first, then was placed onto the outbound device.", + "E": "The express delivery moved from the outbound device to the shelf, and only afterward ended up in your friend's hands." + }, + "answer_label": "C", + "answer_text": "The express delivery moved from the shelf to the express delivery outbound device and then to your friend's hands.", + "q_id": "egolife_A1_D5_026_7", + "correct_answer": "The express delivery moved from the shelf to the express delivery outbound device and then to your friend's hands.", + "original_options": { + "A": "Shelf straight to friend.", + "B": "Outbound device back to shelf.", + "C": "The express delivery moved from the shelf to the express delivery outbound device and then to your friend's hands.", + "D": "The express delivery moved from the shelf to your friend's hands first, then was placed onto the outbound device.", + "E": "The express delivery moved from the outbound device to the shelf, and only afterward ended up in your friend's hands." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D5_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_026_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How did the delivery reach your friend?", + "answer": "The express delivery was moved from the shelf to the express delivery outbound instrument and then to your friend's hands.", + "question_chinese": "快递是怎么到你朋友手上的？", + "answer_chinese": "快递从货架上移动到了快递出库仪上，再到了你朋友的手上。", + "evidence": [ + { + "description": "The owner of the express delivery station found the package from the shelf.", + "timestamps": { + "start": "00:08:50", + "end": "00:08:54" + } + }, + { + "description": "The owner of the express delivery station handed the package over to your friend after taking it out of the warehouse.", + "timestamps": { + "start": "00:08:56", + "end": "00:09:00" + } + } + ], + "question_timestamps": "00:09:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The express delivery was moved from the shelf to the express delivery outbound instrument and then to your friend's hands.", + "B": "It went to your friend's mailbox.", + "C": "It ended up in your friend's backpack.", + "D": "The express delivery was moved from the shelf to the outbound instrument, then placed into your friend's car.", + "E": "The express delivery was moved from the shelf to the outbound instrument and then given to your friend's pet." + }, + "answer_label": "A", + "answer_text": "The express delivery was moved from the shelf to the express delivery outbound instrument and then to your friend's hands.", + "q_id": "egolife_A1_D5_026_8", + "correct_answer": "The express delivery was moved from the shelf to the express delivery outbound instrument and then to your friend's hands.", + "original_options": { + "A": "The express delivery was moved from the shelf to the express delivery outbound instrument and then to your friend's hands.", + "B": "It went to your friend's mailbox.", + "C": "It ended up in your friend's backpack.", + "D": "The express delivery was moved from the shelf to the outbound instrument, then placed into your friend's car.", + "E": "The express delivery was moved from the shelf to the outbound instrument and then given to your friend's pet." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + } +] diff --git a/QAs/egolife_A1_D5_028-mcq.json b/QAs/egolife_A1_D5_028-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ad2ba8dcdd83eb9d7add771e11773adcca3ffc09 --- /dev/null +++ b/QAs/egolife_A1_D5_028-mcq.json @@ -0,0 +1,601 @@ +[ + { + "video_uid": "egolife_A1_D5_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_028-mcq_fixed_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path have I taken with my package since I entered the living room?", + "answer": "You walk from the living room entrance to the kitchen.", + "question_chinese": "从我进入客厅到现在，我拿着我手上的快递经过了什么路径？", + "answer_chinese": "你从客厅入口走到了厨房。", + "evidence": [ + { + "description": "You take the delivery from the entrance of the living room to the kitchen.", + "timestamps": { + "start": "00:07:42", + "end": "00:07:45" + } + } + ], + "question_timestamps": "00:07:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You walk from the entrance to hallway.", + "B": "You walk from the entrance to dining room.", + "C": "You walk from the living room entrance into the hallway, then continue past it toward the stairs.", + "D": "You walk from the living room entrance to the kitchen.", + "E": "You carry the package from the living room entrance to the dining table, stopping there instead of going onward." + }, + "answer_label": "D", + "answer_text": "You walk from the living room entrance to the kitchen.", + "q_id": "egolife_A1_D5_028-mcq_fixed_6", + "correct_answer": "You walk from the living room entrance to the kitchen.", + "original_options": { + "A": "You walk from the entrance to hallway.", + "B": "You walk from the entrance to dining room.", + "C": "You walk from the living room entrance into the hallway, then continue past it toward the stairs.", + "D": "You walk from the living room entrance to the kitchen.", + "E": "You carry the package from the living room entrance to the dining table, stopping there instead of going onward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_028-mcq_fixed_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path have I taken with my package since I entered the living room?", + "answer": "You walk from the entrance of the living room to the kitchen, and then from the kitchen to the projection screen in the living room.", + "question_chinese": "从我进入客厅到现在，我拿着我手上的快递经过了什么路径？", + "answer_chinese": "你从客厅入口走到了厨房，再从厨房走到了客厅的投影幕前。", + "evidence": [ + { + "description": "You take the delivery from the entrance of the living room to the kitchen.", + "timestamps": { + "start": "00:07:42", + "end": "00:07:45" + } + }, + { + "description": "You walk from the kitchen to the projection screen in the living room.", + "timestamps": { + "start": "00:07:55", + "end": "00:08:02" + } + } + ], + "question_timestamps": "00:08:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You go straight to the sofa.", + "B": "You never go into the kitchen.", + "C": "You walk from the living room entrance to the kitchen, then from the kitchen to the sofa in the living room.", + "D": "You walk from the entrance of the living room to the kitchen, and then from the kitchen to the projection screen in the living room.", + "E": "You walk from the living room entrance to the kitchen, then return to the dining area instead of the projection screen." + }, + "answer_label": "D", + "answer_text": "You walk from the entrance of the living room to the kitchen, and then from the kitchen to the projection screen in the living room.", + "q_id": "egolife_A1_D5_028-mcq_fixed_7", + "correct_answer": "You walk from the entrance of the living room to the kitchen, and then from the kitchen to the projection screen in the living room.", + "original_options": { + "A": "You go straight to the sofa.", + "B": "You never go into the kitchen.", + "C": "You walk from the living room entrance to the kitchen, then from the kitchen to the sofa in the living room.", + "D": "You walk from the entrance of the living room to the kitchen, and then from the kitchen to the projection screen in the living room.", + "E": "You walk from the living room entrance to the kitchen, then return to the dining area instead of the projection screen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D5_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_028_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many red traffic lights have I encountered so far?", + "answer": "You have encountered a total of three red traffic lights.", + "question_chinese": "到目前为止，我遇到了几个红色的路灯？", + "answer_chinese": "你一共遇到了3个红色路灯。", + "evidence": [ + { + "description": "You encountered two red traffic lights.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:44" + } + }, + { + "description": "You encountered another red traffic light.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:58" + } + } + ], + "question_timestamps": "00:02:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have encountered one red traffic light.", + "B": "You have encountered two red traffic lights.", + "C": "You have encountered a total of four red traffic lights so far during your route.", + "D": "You have encountered a total of three red traffic lights.", + "E": "You have encountered a total of five red traffic lights up to this point in the trip." + }, + "answer_label": "D", + "answer_text": "You have encountered a total of three red traffic lights.", + "q_id": "egolife_A1_D5_028_1", + "correct_answer": "You have encountered a total of three red traffic lights.", + "original_options": { + "A": "You have encountered one red traffic light.", + "B": "You have encountered two red traffic lights.", + "C": "You have encountered a total of four red traffic lights so far during your route.", + "D": "You have encountered a total of three red traffic lights.", + "E": "You have encountered a total of five red traffic lights up to this point in the trip." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_028_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many red traffic lights have I encountered so far?", + "answer": "You have encountered a total of five red traffic lights.", + "question_chinese": "到目前为止，我遇到了几个红色的路灯？", + "answer_chinese": "你一共遇到了5个红色路灯。", + "evidence": [ + { + "description": "You encountered two red traffic lights.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:44" + } + }, + { + "description": "You encountered another red traffic light.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:58" + } + }, + { + "description": "You encountered two more red traffic lights.", + "timestamps": { + "start": "00:02:50", + "end": "00:02:52" + } + } + ], + "question_timestamps": "00:03:03", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "You have encountered three red traffic lights.", + "B": "You have encountered four red traffic lights.", + "C": "You have encountered a total of two red traffic lights so far during your trip.", + "D": "You have encountered a total of five red traffic lights.", + "E": "You have encountered a total of six red traffic lights up to this point in your journey." + }, + "answer_label": "D", + "answer_text": "You have encountered a total of five red traffic lights.", + "q_id": "egolife_A1_D5_028_2", + "correct_answer": "You have encountered a total of five red traffic lights.", + "original_options": { + "A": "You have encountered three red traffic lights.", + "B": "You have encountered four red traffic lights.", + "C": "You have encountered a total of two red traffic lights so far during your trip.", + "D": "You have encountered a total of five red traffic lights.", + "E": "You have encountered a total of six red traffic lights up to this point in your journey." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_028-mcq_fixed_4", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my trajectory since I realized I was going the wrong way?", + "answer": "You turn back and follow the current road and then enter a small, bright alley.", + "question_chinese": "我发现走错路后，到现在我的运动轨迹是怎样的？", + "answer_chinese": "你回头后沿当前道路前行接着进入了一条明亮的小巷子", + "evidence": [ + { + "description": "You turn back and follow your current path.", + "timestamps": { + "start": "00:04:36", + "end": "00:04:41" + } + }, + { + "description": "You enter a bright alley.", + "timestamps": { + "start": "00:07:03", + "end": "00:07:06" + } + } + ], + "question_timestamps": "00:07:07", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You keep walking forward.", + "B": "You turn left into the main street.", + "C": "You turn back and follow the current road and then enter a small, bright alley.", + "D": "You turn back, follow the road, and enter a dim alley instead of a bright one.", + "E": "You reverse direction, follow the road a short way, then enter a narrow alley that isn’t brightly lit." + }, + "answer_label": "C", + "answer_text": "You turn back and follow the current road and then enter a small, bright alley.", + "q_id": "egolife_A1_D5_028-mcq_fixed_4", + "correct_answer": "You turn back and follow the current road and then enter a small, bright alley.", + "original_options": { + "A": "You keep walking forward.", + "B": "You turn left into the main street.", + "C": "You turn back and follow the current road and then enter a small, bright alley.", + "D": "You turn back, follow the road, and enter a dim alley instead of a bright one.", + "E": "You reverse direction, follow the road a short way, then enter a narrow alley that isn’t brightly lit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D5_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_028-mcq_fixed_5", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my trajectory since I realized I was going the wrong way?", + "answer": "You turn back and follow the current road into a small, bright alley. After half a minute or so, turn right.", + "question_chinese": "我发现走错路后，到现在我的运动轨迹是怎样的？", + "answer_chinese": "你回头后沿当前道路前行进入了一条明亮的小巷子，半分钟左右后右拐。", + "evidence": [ + { + "description": "You turn back and follow your current path.", + "timestamps": { + "start": "00:04:36", + "end": "00:04:41" + } + }, + { + "description": "You enter a bright alley.", + "timestamps": { + "start": "00:07:03", + "end": "00:07:06" + } + }, + { + "description": "You turn right from the alley into the yard.", + "timestamps": { + "start": "00:07:28", + "end": "00:07:30" + } + } + ], + "question_timestamps": "00:07:31", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You turn left after half a minute.", + "B": "You keep going straight after half a minute.", + "C": "You turn back into the bright alley, walk about half a minute, and then turn left at the opening.", + "D": "You turn back and follow the current road into a small, bright alley. After half a minute or so, turn right.", + "E": "You retrace onto the same road, enter the small bright alley, and continue without turning right afterward." + }, + "answer_label": "D", + "answer_text": "You turn back and follow the current road into a small, bright alley. After half a minute or so, turn right.", + "q_id": "egolife_A1_D5_028-mcq_fixed_5", + "correct_answer": "You turn back and follow the current road into a small, bright alley. After half a minute or so, turn right.", + "original_options": { + "A": "You turn left after half a minute.", + "B": "You keep going straight after half a minute.", + "C": "You turn back into the bright alley, walk about half a minute, and then turn left at the opening.", + "D": "You turn back and follow the current road into a small, bright alley. After half a minute or so, turn right.", + "E": "You retrace onto the same road, enter the small bright alley, and continue without turning right afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D5_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_028-mcq_fixed_8", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I in the house now?", + "answer": "You are in the bathroom now.", + "question_chinese": "我现在在房子里的哪里？", + "answer_chinese": "你现在在卫生间。", + "evidence": [ + { + "description": "You wash your hands in the bathroom", + "timestamps": { + "start": "00:09:00", + "end": "00:09:02" + } + } + ], + "question_timestamps": "00:09:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are in the kitchen now.", + "B": "You are in the living room now.", + "C": "You are in the bathroom now.", + "D": "You are in the hallway right now, standing near the closet area rather than inside the bathroom.", + "E": "You are in a bedroom now, away from the sink and shower area of the bathroom." + }, + "answer_label": "C", + "answer_text": "You are in the bathroom now.", + "q_id": "egolife_A1_D5_028-mcq_fixed_8", + "correct_answer": "You are in the bathroom now.", + "original_options": { + "A": "You are in the kitchen now.", + "B": "You are in the living room now.", + "C": "You are in the bathroom now.", + "D": "You are in the hallway right now, standing near the closet area rather than inside the bathroom.", + "E": "You are in a bedroom now, away from the sink and shower area of the bathroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D5_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_028-mcq_fixed_9", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I in the house now?", + "answer": "You are in the living room now.", + "question_chinese": "我现在在房子里的哪里？", + "answer_chinese": "你现在在客厅里。", + "evidence": [ + { + "description": "You are at the living room table", + "timestamps": { + "start": "00:10:02", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:10:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are in the kitchen now.", + "B": "You are in the bedroom now.", + "C": "You are in the dining room now, seated near the table instead of being in the living room.", + "D": "You are in the living room now.", + "E": "You are in the hallway area now, between rooms, not settled in the living room." + }, + "answer_label": "D", + "answer_text": "You are in the living room now.", + "q_id": "egolife_A1_D5_028-mcq_fixed_9", + "correct_answer": "You are in the living room now.", + "original_options": { + "A": "You are in the kitchen now.", + "B": "You are in the bedroom now.", + "C": "You are in the dining room now, seated near the table instead of being in the living room.", + "D": "You are in the living room now.", + "E": "You are in the hallway area now, between rooms, not settled in the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_D5_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_028_9", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I take the pot on the table now?", + "answer": "You can reach the pot on the table.", + "question_chinese": "现在我能够拿到桌子上的锅吗？", + "answer_chinese": "你能够到桌子上的锅。", + "evidence": [ + { + "description": "You are beside the table in the living room.", + "timestamps": { + "start": "00:09:42", + "end": "00:09:44" + } + }, + { + "description": "The pot is on the table in the living room.", + "timestamps": { + "start": "00:08:48", + "end": "00:08:50" + } + } + ], + "question_timestamps": "00:09:48", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can't reach the pot on the table.", + "B": "You can reach the pot on the table." + }, + "answer_label": "B", + "answer_text": "You can reach the pot on the table.", + "q_id": "egolife_A1_D5_028_9", + "correct_answer": "You can reach the pot on the table.", + "original_options": { + "A": "You can't reach the pot on the table.", + "B": "You can reach the pot on the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D5_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_028_10", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I take the pot on the table now?", + "answer": "You can't reach the pot on the table.", + "question_chinese": "现在我能够拿到桌子上的锅吗？", + "answer_chinese": "你不能够到桌子上的锅。", + "evidence": [ + { + "description": "The pot is on the table in the living room.", + "timestamps": { + "start": "00:08:48", + "end": "00:08:50" + } + }, + { + "description": "You are beside the projector screen far from the table in the living room.", + "timestamps": { + "start": "00:11:40", + "end": "00:11:41" + } + } + ], + "question_timestamps": "00:11:45", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can reach the pot on the table.", + "B": "You can't reach the pot on the table." + }, + "answer_label": "B", + "answer_text": "You can't reach the pot on the table.", + "q_id": "egolife_A1_D5_028_10", + "correct_answer": "You can't reach the pot on the table.", + "original_options": { + "A": "You can reach the pot on the table.", + "B": "You can't reach the pot on the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + } +] diff --git a/QAs/egolife_A1_D5_031-mcq.json b/QAs/egolife_A1_D5_031-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3241d34e71f5bd5a86b546ccc2a6a59b138088d1 --- /dev/null +++ b/QAs/egolife_A1_D5_031-mcq.json @@ -0,0 +1,280 @@ +[ + { + "video_uid": "egolife_A1_D5_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_031_1_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, onto how many egg tarts did I sift the green matcha powder?", + "answer": "You sifted matcha powder onto 3 egg tarts.", + "question_chinese": "我将绿色的抹茶粉一共筛到了几个蛋挞上？", + "answer_chinese": "你将抹茶粉筛到了3个蛋挞上。", + "evidence": [ + { + "description": "You sprinkled the matcha powder onto 3 egg tarts.", + "timestamps": { + "start": "00:05:28", + "end": "00:05:42" + } + } + ], + "question_timestamps": "00:05:54", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You sifted matcha powder onto 2 egg tarts.", + "B": "You sifted matcha powder onto 3 egg tarts.", + "C": "You sifted matcha powder onto 4 egg tarts.", + "D": "You sifted matcha powder onto 1 egg tart in total during the dusting step.", + "E": "You sifted matcha powder onto 5 egg tarts overall before you finished the sifting." + }, + "answer_label": "B", + "answer_text": "You sifted matcha powder onto 3 egg tarts.", + "q_id": "egolife_A1_D5_031_1_T1", + "correct_answer": "You sifted matcha powder onto 3 egg tarts.", + "original_options": { + "A": "You sifted matcha powder onto 2 egg tarts.", + "B": "You sifted matcha powder onto 3 egg tarts.", + "C": "You sifted matcha powder onto 4 egg tarts.", + "D": "You sifted matcha powder onto 1 egg tart in total during the dusting step.", + "E": "You sifted matcha powder onto 5 egg tarts overall before you finished the sifting." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D5_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_031_1_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In total, onto how many egg tarts did I sift the green matcha powder?", + "answer": "You sifted matcha powder onto 4 egg tarts.", + "question_chinese": "我将绿色的抹茶粉一共筛到了几个蛋挞上？", + "answer_chinese": "你将抹茶粉筛到了4个蛋挞上。", + "evidence": [ + { + "description": "You sprinkled the matcha powder onto 4 egg tarts.", + "timestamps": { + "start": "00:05:28", + "end": "00:06:42" + } + } + ], + "question_timestamps": "00:06:45", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You sifted matcha powder onto 2 egg tarts.", + "B": "You sifted matcha powder onto 6 egg tarts.", + "C": "You sifted matcha powder onto 4 egg tarts.", + "D": "You sifted matcha powder onto 3 egg tarts in total, covering three pastries with a light green dusting.", + "E": "You sifted matcha powder onto 5 egg tarts altogether, moving the sifter across five tarts one by one." + }, + "answer_label": "C", + "answer_text": "You sifted matcha powder onto 4 egg tarts.", + "q_id": "egolife_A1_D5_031_1_T2", + "correct_answer": "You sifted matcha powder onto 4 egg tarts.", + "original_options": { + "A": "You sifted matcha powder onto 2 egg tarts.", + "B": "You sifted matcha powder onto 6 egg tarts.", + "C": "You sifted matcha powder onto 4 egg tarts.", + "D": "You sifted matcha powder onto 3 egg tarts in total, covering three pastries with a light green dusting.", + "E": "You sifted matcha powder onto 5 egg tarts altogether, moving the sifter across five tarts one by one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_031_4", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many egg tarts did I sift the green matcha powder onto in total?", + "answer": "You sprinkled the matcha powder onto four egg tarts.", + "question_chinese": "我将绿色的抹茶粉一共筛到了几个蛋挞上？", + "answer_chinese": "你将抹茶粉筛到了4个蛋挞上。", + "evidence": [ + { + "description": "You sprinkled the matcha powder onto 3 egg tarts.", + "timestamps": { + "start": "00:05:28", + "end": "00:05:42" + } + }, + { + "description": "You sprinkled the matcha powder onto 1 egg tart.", + "timestamps": { + "start": "00:06:31", + "end": "00:06:35" + } + } + ], + "question_timestamps": "00:06:54", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You sprinkled it onto two egg tarts.", + "B": "You sprinkled the matcha powder onto four egg tarts.", + "C": "You sprinkled it onto five egg tarts.", + "D": "You sprinkled the matcha powder onto three egg tarts in total, dusting only three of the pastries.", + "E": "You sprinkled the matcha powder onto one egg tart overall, focusing the green powder on a single tart." + }, + "answer_label": "B", + "answer_text": "You sprinkled the matcha powder onto four egg tarts.", + "q_id": "egolife_A1_D5_031_4", + "correct_answer": "You sprinkled the matcha powder onto four egg tarts.", + "original_options": { + "A": "You sprinkled it onto two egg tarts.", + "B": "You sprinkled the matcha powder onto four egg tarts.", + "C": "You sprinkled it onto five egg tarts.", + "D": "You sprinkled the matcha powder onto three egg tarts in total, dusting only three of the pastries.", + "E": "You sprinkled the matcha powder onto one egg tart overall, focusing the green powder on a single tart." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D5_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_031_2_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "While I was sprinkling the matcha powder, how many egg tarts got taken away?", + "answer": "While you were sprinkling matcha powder, 0 egg tarts were taken away, and the egg tarts were in front of you.", + "question_chinese": "我撒抹茶粉时有多少个蛋挞被拿走了？", + "answer_chinese": "你撒抹茶粉时有0个蛋挞被拿走了，蛋挞在你前面。", + "evidence": [ + { + "description": "Zero egg tarts were taken away.", + "timestamps": { + "start": "00:05:28", + "end": "00:05:42" + } + } + ], + "question_timestamps": "00:05:46", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "1 egg tart was taken away.", + "B": "2 egg tarts were taken away.", + "C": "While you were sprinkling matcha powder, 3 egg tarts were taken away, and the egg tarts were in front of you.", + "D": "While you were sprinkling matcha powder, 0 egg tarts were taken away, and the egg tarts were in front of you.", + "E": "While you were sprinkling matcha powder, 4 egg tarts were taken away, and the egg tarts were in front of you." + }, + "answer_label": "D", + "answer_text": "While you were sprinkling matcha powder, 0 egg tarts were taken away, and the egg tarts were in front of you.", + "q_id": "egolife_A1_D5_031_2_T2", + "correct_answer": "While you were sprinkling matcha powder, 0 egg tarts were taken away, and the egg tarts were in front of you.", + "original_options": { + "A": "1 egg tart was taken away.", + "B": "2 egg tarts were taken away.", + "C": "While you were sprinkling matcha powder, 3 egg tarts were taken away, and the egg tarts were in front of you.", + "D": "While you were sprinkling matcha powder, 0 egg tarts were taken away, and the egg tarts were in front of you.", + "E": "While you were sprinkling matcha powder, 4 egg tarts were taken away, and the egg tarts were in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D5_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D5_031_2_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "While I was sprinkling the matcha powder, how many egg tarts got taken away?", + "answer": "While you were sprinkling matcha powder, 4 egg tarts were taken away, and the egg tarts were in front of you.", + "question_chinese": "我撒抹茶粉时有多少个蛋挞被拿走了？", + "answer_chinese": "你撒抹茶粉时有4个蛋挞被拿走了，蛋挞在你前面。", + "evidence": [ + { + "description": "Four egg tarts were taken away.", + "timestamps": { + "start": "00:06:56", + "end": "00:06:59" + } + } + ], + "question_timestamps": "00:07:02", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "2 egg tarts were taken away.", + "B": "6 egg tarts were taken away.", + "C": "While you were sprinkling matcha powder, 4 egg tarts were taken away, and the egg tarts were in front of you.", + "D": "While you were sprinkling matcha powder, 3 egg tarts were taken away, and the egg tarts were in front of you.", + "E": "While you were sprinkling matcha powder, 5 egg tarts were taken away, and the egg tarts were in front of you." + }, + "answer_label": "C", + "answer_text": "While you were sprinkling matcha powder, 4 egg tarts were taken away, and the egg tarts were in front of you.", + "q_id": "egolife_A1_D5_031_2_T1", + "correct_answer": "While you were sprinkling matcha powder, 4 egg tarts were taken away, and the egg tarts were in front of you.", + "original_options": { + "A": "2 egg tarts were taken away.", + "B": "6 egg tarts were taken away.", + "C": "While you were sprinkling matcha powder, 4 egg tarts were taken away, and the egg tarts were in front of you.", + "D": "While you were sprinkling matcha powder, 3 egg tarts were taken away, and the egg tarts were in front of you.", + "E": "While you were sprinkling matcha powder, 5 egg tarts were taken away, and the egg tarts were in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 21 + ] + } +] diff --git a/QAs/egolife_A1_D6_012-mcq.json b/QAs/egolife_A1_D6_012-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3b9b69221aa75893688fd55a70c43d4594b23c49 --- /dev/null +++ b/QAs/egolife_A1_D6_012-mcq.json @@ -0,0 +1,1392 @@ +[ + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_3", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I get up, what direction are the items on the chair in relation to me?", + "answer": "After I stood up, the items on the chair were to my rear.", + "question_chinese": "在我起身之后，椅子上的物品相对我在什么方位？", + "answer_chinese": "在我起身之后，椅子上的物品在我的后方。", + "evidence": [ + { + "description": "When I was sitting, the items on the chair were in front of me to the right.", + "timestamps": { + "start": "00:03:16", + "end": "00:03:20" + } + }, + { + "description": "Turn 90 degrees to face the table.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:23" + } + } + ], + "question_timestamps": "00:03:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To my left.", + "B": "In front of me.", + "C": "After I stood up, the items on the chair were to my rear.", + "D": "After I stood up, the items on the chair were to my right, sitting off to the side of my body.", + "E": "After I stood up, the items on the chair were directly ahead of me, positioned in my forward view." + }, + "answer_label": "C", + "answer_text": "After I stood up, the items on the chair were to my rear.", + "q_id": "egolife_A1_D6_012_3", + "correct_answer": "After I stood up, the items on the chair were to my rear.", + "original_options": { + "A": "To my left.", + "B": "In front of me.", + "C": "After I stood up, the items on the chair were to my rear.", + "D": "After I stood up, the items on the chair were to my right, sitting off to the side of my body.", + "E": "After I stood up, the items on the chair were directly ahead of me, positioned in my forward view." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 21, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "After I get up, what direction are the items on the chair in relation to me?", + "answer": "After I stood up, the item was behind me.", + "question_chinese": "在我起身之后，椅子上的物品相对我在什么方位？", + "answer_chinese": "在我起身之后，物品在我的后方。", + "evidence": [ + { + "description": "When I am sitting, the object is in front of me to the right.", + "timestamps": { + "start": "00:03:19", + "end": "00:03:20" + } + }, + { + "description": "Turn 90 degrees to face the table.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:23" + } + } + ], + "question_timestamps": "00:05:42", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The item was in front of me.", + "B": "The item was to my left.", + "C": "After I stood up, the item was to my right, positioned off to the side of my body.", + "D": "After I stood up, the item was behind me.", + "E": "After I stood up, the item was directly in front of me, within my forward line of sight." + }, + "answer_label": "D", + "answer_text": "After I stood up, the item was behind me.", + "q_id": "egolife_A1_D6_012_4", + "correct_answer": "After I stood up, the item was behind me.", + "original_options": { + "A": "The item was in front of me.", + "B": "The item was to my left.", + "C": "After I stood up, the item was to my right, positioned off to the side of my body.", + "D": "After I stood up, the item was behind me.", + "E": "After I stood up, the item was directly in front of me, within my forward line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_7", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I am sitting in the chair, which is closer to me, the drink or the phone?", + "answer": "The phone is closer to me, about 50 centimeters away. The drink is to my left, approximately 150 centimeters away.", + "question_chinese": "当我坐在椅子上时，饮料和手机那个离我更近？", + "answer_chinese": "手机离我更近，大约在50厘米左右，饮料在我的左边，距离大约是150厘米左右。", + "evidence": [ + { + "description": "The phone is on the table, in the position directly in front of me.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:21" + } + }, + { + "description": "The drink is on the table, in front of me to the left.", + "timestamps": { + "start": "00:01:21", + "end": "00:01:22" + } + }, + { + "description": "The distance between the beverage and the mobile phone is about 100 centimeters.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:01:33", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The phone is closer to me, about 50 centimeters away. The drink is to my left, approximately 150 centimeters away.", + "B": "The drink is closer to me, about 50 centimeters away. The phone is to my left, approximately 150 centimeters away." + }, + "answer_label": "A", + "answer_text": "The phone is closer to me, about 50 centimeters away. The drink is to my left, approximately 150 centimeters away.", + "q_id": "egolife_A1_D6_012_7", + "correct_answer": "The phone is closer to me, about 50 centimeters away. The drink is to my left, approximately 150 centimeters away.", + "original_options": { + "A": "The phone is closer to me, about 50 centimeters away. The drink is to my left, approximately 150 centimeters away.", + "B": "The drink is closer to me, about 50 centimeters away. The phone is to my left, approximately 150 centimeters away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_8", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "When I am sitting in the chair, which is closer to me, the drink or the phone?", + "answer": "The phone is closer to me, about 50 centimeters away. The drink is to my left, approximately 150 centimeters away.", + "question_chinese": "当我坐在椅子上时，饮料和手机那个离我更近？", + "answer_chinese": "手机离我更近，大约在50厘米左右，饮料在我的左边，距离大约是150厘米左右。", + "evidence": [ + { + "description": "The mobile phone is on the table, directly in front of me.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:21" + } + }, + { + "description": "The drink is on the table in front of me to the left.", + "timestamps": { + "start": "00:01:21", + "end": "00:01:22" + } + }, + { + "description": "The distance between the beverage and the mobile phone is about 100 centimeters.", + "timestamps": { + "start": "00:01:20", + "end": "00:01:22" + } + } + ], + "question_timestamps": "00:09:50", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The drink is closer to me, about 50 centimeters away. The phone is to my left, approximately 150 centimeters away.", + "B": "The phone is closer to me, about 50 centimeters away. The drink is to my left, approximately 150 centimeters away." + }, + "answer_label": "B", + "answer_text": "The phone is closer to me, about 50 centimeters away. The drink is to my left, approximately 150 centimeters away.", + "q_id": "egolife_A1_D6_012_8", + "correct_answer": "The phone is closer to me, about 50 centimeters away. The drink is to my left, approximately 150 centimeters away.", + "original_options": { + "A": "The drink is closer to me, about 50 centimeters away. The phone is to my left, approximately 150 centimeters away.", + "B": "The phone is closer to me, about 50 centimeters away. The drink is to my left, approximately 150 centimeters away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_5", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take before I sat down in front of the first laptop?", + "answer": "I first walked straight from outside into the room, went up the stairs, entered the hall, placed the items in my hand on the table, and then turned around to arrive.", + "question_chinese": "当我坐在第一台笔记本电脑之前，我经过了什么路线？", + "answer_chinese": "我先径直从室外走入室内，上楼梯后进入大厅，将手中物品放在桌子上后，转身到达。", + "evidence": [ + { + "description": "When I placed the item on the table, the laptop was behind me.", + "timestamps": { + "start": "00:04:48", + "end": "00:04:50" + } + } + ], + "question_timestamps": "00:05:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I first walked straight from outside into the room, went up the stairs, entered the hall, placed the items in my hand on the table, and then turned around to arrive.", + "B": "You went down the stairs first.", + "C": "You set the items on a chair.", + "D": "You walked in from outside, went up the stairs, entered the hall, set your items on the floor, then turned around.", + "E": "You came in from outside, climbed the stairs, entered the hall, put what you carried on the sofa, then turned to arrive." + }, + "answer_label": "A", + "answer_text": "I first walked straight from outside into the room, went up the stairs, entered the hall, placed the items in my hand on the table, and then turned around to arrive.", + "q_id": "egolife_A1_D6_012_5", + "correct_answer": "I first walked straight from outside into the room, went up the stairs, entered the hall, placed the items in my hand on the table, and then turned around to arrive.", + "original_options": { + "A": "I first walked straight from outside into the room, went up the stairs, entered the hall, placed the items in my hand on the table, and then turned around to arrive.", + "B": "You went down the stairs first.", + "C": "You set the items on a chair.", + "D": "You walked in from outside, went up the stairs, entered the hall, set your items on the floor, then turned around.", + "E": "You came in from outside, climbed the stairs, entered the hall, put what you carried on the sofa, then turned to arrive." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 22 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_6", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take before I sat down in front of the first laptop?", + "answer": "I first walked straight from outside into the room, went up the stairs, entered the hall, placed the items in my hand on the table, and then turned around to arrive.", + "question_chinese": "当我坐在第一台笔记本电脑之前，我经过了什么路线？", + "answer_chinese": "我先径直从室外走入室内，上楼梯后进入大厅，将手中物品放在桌子上后，转身到达。", + "evidence": [ + { + "description": "When placing the item on the table, the laptop is behind me.", + "timestamps": { + "start": "00:04:48", + "end": "00:04:50" + } + }, + { + "description": "Turn the laptop 180 degrees so it's in front of me.", + "timestamps": { + "start": "00:04:51", + "end": "00:04:51" + } + } + ], + "question_timestamps": "00:08:41", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went down the stairs.", + "B": "You turned right at the doorway.", + "C": "I first walked straight from outside into the room, went up the stairs, entered the hall, placed the items in my hand on the table, and then turned around to arrive.", + "D": "You walked in from outside, went up the stairs, entered the hall, put your items on a chair, then turned around.", + "E": "You came in from outside, went up the stairs, entered the hall, set the items on a shelf, then turned around." + }, + "answer_label": "C", + "answer_text": "I first walked straight from outside into the room, went up the stairs, entered the hall, placed the items in my hand on the table, and then turned around to arrive.", + "q_id": "egolife_A1_D6_012_6", + "correct_answer": "I first walked straight from outside into the room, went up the stairs, entered the hall, placed the items in my hand on the table, and then turned around to arrive.", + "original_options": { + "A": "You went down the stairs.", + "B": "You turned right at the doorway.", + "C": "I first walked straight from outside into the room, went up the stairs, entered the hall, placed the items in my hand on the table, and then turned around to arrive.", + "D": "You walked in from outside, went up the stairs, entered the hall, put your items on a chair, then turned around.", + "E": "You came in from outside, went up the stairs, entered the hall, set the items on a shelf, then turned around." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_3_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I put the notebook on the table, what route did I take?", + "answer": "You walked backward for a while, then doubled back a short distance, went toward the doorway and turned left, and headed toward the computer.", + "question_chinese": "当我把本子放在桌子上后，我经过了什么路线？", + "answer_chinese": "你向后走了一段距离，又往回折返了一小段路，走向门口后向左转，走向电脑", + "evidence": [ + { + "description": "Walked backward for a certain distance.", + "timestamps": { + "start": "00:04:50", + "end": "00:04:52" + } + }, + { + "description": "Turned back a short distance again.", + "timestamps": { + "start": "00:04:53", + "end": "00:04:54" + } + }, + { + "description": "Walk towards the door, then turn left and head to the computer.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:05:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right at the doorway.", + "B": "You walked forward the whole time.", + "C": "You walked backward for a while, doubled back a short distance, went to the doorway, turned right, then headed toward the computer.", + "D": "You walked backward for a while, then doubled back a short distance, went toward the doorway and turned left, and headed toward the computer.", + "E": "You walked backward briefly, doubled back, went toward the doorway and turned left, and stopped before reaching the computer." + }, + "answer_label": "D", + "answer_text": "You walked backward for a while, then doubled back a short distance, went toward the doorway and turned left, and headed toward the computer.", + "q_id": "egolife_A1_D6_012_3_T2", + "correct_answer": "You walked backward for a while, then doubled back a short distance, went toward the doorway and turned left, and headed toward the computer.", + "original_options": { + "A": "You turned right at the doorway.", + "B": "You walked forward the whole time.", + "C": "You walked backward for a while, doubled back a short distance, went to the doorway, turned right, then headed toward the computer.", + "D": "You walked backward for a while, then doubled back a short distance, went toward the doorway and turned left, and headed toward the computer.", + "E": "You walked backward briefly, doubled back, went toward the doorway and turned left, and stopped before reaching the computer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 22, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_3_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I put the notebook on the table, what route did I take?", + "answer": "You walked backward for a while, then doubled back a short distance, went toward the doorway and turned left, and headed toward the computer. You sat on the floor and turned on the computer.", + "question_chinese": "当我把本子放在桌子上后，我经过了什么路线？", + "answer_chinese": "你向后走了一段距离，又往回折返了一小段路，走向门口后向左转，走向电脑，坐在地板上打开电脑", + "evidence": [ + { + "description": "Walked backward for a certain distance.", + "timestamps": { + "start": "00:04:50", + "end": "00:04:52" + } + }, + { + "description": "Turned back a short distance again.", + "timestamps": { + "start": "00:04:53", + "end": "00:04:54" + } + }, + { + "description": "Walk towards the door, then turn left and head to the computer.", + "timestamps": { + "start": "00:04:55", + "end": "00:05:00" + } + }, + { + "description": "Sit on the floor and turn on the computer.", + "timestamps": { + "start": "00:05:01", + "end": "00:05:07" + } + } + ], + "question_timestamps": "00:05:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned right at the doorway.", + "B": "You never sat on the floor.", + "C": "You walked backward for a while, then doubled back a short distance, went toward the doorway and turned left, and headed toward the computer. You sat on the floor and turned on the computer.", + "D": "You walked backward for a while, doubled back a short distance, turned right at the doorway, and headed toward the computer.", + "E": "You walked backward for a while, then doubled back, turned left at the doorway, and reached the computer without turning it on." + }, + "answer_label": "C", + "answer_text": "You walked backward for a while, then doubled back a short distance, went toward the doorway and turned left, and headed toward the computer. You sat on the floor and turned on the computer.", + "q_id": "egolife_A1_D6_012_3_T1", + "correct_answer": "You walked backward for a while, then doubled back a short distance, went toward the doorway and turned left, and headed toward the computer. You sat on the floor and turned on the computer.", + "original_options": { + "A": "You turned right at the doorway.", + "B": "You never sat on the floor.", + "C": "You walked backward for a while, then doubled back a short distance, went toward the doorway and turned left, and headed toward the computer. You sat on the floor and turned on the computer.", + "D": "You walked backward for a while, doubled back a short distance, turned right at the doorway, and headed toward the computer.", + "E": "You walked backward for a while, then doubled back, turned left at the doorway, and reached the computer without turning it on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 22 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_6_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times did I use the laptop in the second-floor lobby?", + "answer": "You used it once.", + "question_chinese": "我在二楼大厅使用了几次笔记本？", + "answer_chinese": "你使用了一次。", + "evidence": [ + { + "description": "After placing the items and turning around, I saw the first one in the corner to the left of the hall.", + "timestamps": { + "start": "00:04:48", + "end": "00:05:02" + } + }, + { + "description": "After standing up, I saw the second laptop in the right corner of the hall.", + "timestamps": { + "start": "00:05:52", + "end": "00:05:53" + } + } + ], + "question_timestamps": "00:05:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You used it twice.", + "B": "You didn’t use it at all.", + "C": "You used it once.", + "D": "You used the laptop three separate times while you were in the second-floor lobby.", + "E": "You used the laptop four times in the second-floor lobby before moving on." + }, + "answer_label": "C", + "answer_text": "You used it once.", + "q_id": "egolife_A1_D6_012_6_T2", + "correct_answer": "You used it once.", + "original_options": { + "A": "You used it twice.", + "B": "You didn’t use it at all.", + "C": "You used it once.", + "D": "You used the laptop three separate times while you were in the second-floor lobby.", + "E": "You used the laptop four times in the second-floor lobby before moving on." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_6_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times did I use the laptop in the second-floor lobby?", + "answer": "You used it twice.", + "question_chinese": "我在二楼大厅使用了几次笔记本？", + "answer_chinese": "你使用了两次。", + "evidence": [ + { + "description": "After placing the items and turning around, I saw the first one in the corner to the left of the hall.", + "timestamps": { + "start": "00:04:48", + "end": "00:05:02" + } + }, + { + "description": "After standing up, I saw the second laptop in the right corner of the hall.", + "timestamps": { + "start": "00:05:52", + "end": "00:06:01" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You used it once.", + "B": "You used it twice.", + "C": "You didn’t use it at all.", + "D": "You used the laptop three times while you were in the second-floor lobby.", + "E": "You used the laptop four separate times while you were in the second-floor lobby." + }, + "answer_label": "B", + "answer_text": "You used it twice.", + "q_id": "egolife_A1_D6_012_6_T1", + "correct_answer": "You used it twice.", + "original_options": { + "A": "You used it once.", + "B": "You used it twice.", + "C": "You didn’t use it at all.", + "D": "You used the laptop three times while you were in the second-floor lobby.", + "E": "You used the laptop four separate times while you were in the second-floor lobby." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 7, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_5_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was in the second-floor lobby, how many times did I use the laptop?", + "answer": "You used it once.", + "question_chinese": "我在二楼大厅的时候用了几次笔记本电脑？", + "answer_chinese": "你使用了一次。", + "evidence": [ + { + "description": "After I put things down , I used the first computer.", + "timestamps": { + "start": "00:04:56", + "end": "00:05:02" + } + } + ], + "question_timestamps": "00:05:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used it zero times.", + "B": "You used it once.", + "C": "You used it twice.", + "D": "You used the laptop three times while you were in the second-floor lobby.", + "E": "You used the laptop four times in the second-floor lobby before leaving." + }, + "answer_label": "B", + "answer_text": "You used it once.", + "q_id": "egolife_A1_D6_012_5_T2", + "correct_answer": "You used it once.", + "original_options": { + "A": "You used it zero times.", + "B": "You used it once.", + "C": "You used it twice.", + "D": "You used the laptop three times while you were in the second-floor lobby.", + "E": "You used the laptop four times in the second-floor lobby before leaving." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_10_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was in the second-floor lobby, how many times did I use the laptop?", + "answer": "You used the computer once.", + "question_chinese": "我在二楼大厅的时候用了几次笔记本电脑？", + "answer_chinese": "你用了一次电脑。", + "evidence": [ + { + "description": "After I put things down , I used the first computer.", + "timestamps": { + "start": "00:04:56", + "end": "00:05:02" + } + } + ], + "question_timestamps": "00:05:33", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used the computer zero times.", + "B": "You used the computer twice.", + "C": "You used the computer three times while you were in the second-floor lobby.", + "D": "You used the computer four times in the second-floor lobby before moving on.", + "E": "You used the computer once." + }, + "answer_label": "E", + "answer_text": "You used the computer once.", + "q_id": "egolife_A1_D6_012_10_T1", + "correct_answer": "You used the computer once.", + "original_options": { + "A": "You used the computer zero times.", + "B": "You used the computer twice.", + "C": "You used the computer three times while you were in the second-floor lobby.", + "D": "You used the computer four times in the second-floor lobby before moving on.", + "E": "You used the computer once." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_5_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was in the second-floor lobby, how many times did I use the laptop?", + "answer": "You used it twice.", + "question_chinese": "我在二楼大厅的时候用了几次笔记本电脑？", + "answer_chinese": "你使用了两次。", + "evidence": [ + { + "description": "After I put things down , I used the first computer.", + "timestamps": { + "start": "00:04:56", + "end": "00:05:02" + } + }, + { + "description": "I walked through the hall to the opposite corner and used the second laptop.", + "timestamps": { + "start": "00:05:52", + "end": "00:06:01" + } + } + ], + "question_timestamps": "00:08:47", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used it twice.", + "B": "You used it once.", + "C": "You used it three times.", + "D": "You did not use the laptop at all while you were in the second-floor lobby.", + "E": "You used the laptop four times during your time in the second-floor lobby." + }, + "answer_label": "A", + "answer_text": "You used it twice.", + "q_id": "egolife_A1_D6_012_5_T1", + "correct_answer": "You used it twice.", + "original_options": { + "A": "You used it twice.", + "B": "You used it once.", + "C": "You used it three times.", + "D": "You did not use the laptop at all while you were in the second-floor lobby.", + "E": "You used the laptop four times during your time in the second-floor lobby." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_10_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I was in the second-floor lobby, how many times did I use the laptop?", + "answer": "You used the computer twice.", + "question_chinese": "我在二楼大厅的时候用了几次笔记本电脑？", + "answer_chinese": "你用了两次电脑。", + "evidence": [ + { + "description": "After placing the items and turning around, I saw the first one in the corner to the left of the hall.", + "timestamps": { + "start": "00:04:48", + "end": "00:05:02" + } + }, + { + "description": "After standing up, I saw the second laptop in the right corner of the hall.", + "timestamps": { + "start": "00:05:52", + "end": "00:06:01" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used the computer twice.", + "B": "You used it once.", + "C": "You used it three times.", + "D": "You didn’t use the computer at all in the second-floor lobby.", + "E": "You used the computer four times while you were in the second-floor lobby." + }, + "answer_label": "A", + "answer_text": "You used the computer twice.", + "q_id": "egolife_A1_D6_012_10_T2", + "correct_answer": "You used the computer twice.", + "original_options": { + "A": "You used the computer twice.", + "B": "You used it once.", + "C": "You used it three times.", + "D": "You didn’t use the computer at all in the second-floor lobby.", + "E": "You used the computer four times while you were in the second-floor lobby." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_7_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the laptop relative to me right now?", + "answer": "The laptop was in the room behind you.", + "question_chinese": "现在笔记本电脑相对我在什么方位？", + "answer_chinese": "笔记本电脑在你后面的房间里。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:09:01", + "end": "00:09:02" + } + }, + { + "description": "I turned 90 degrees and left the dormitory.", + "timestamps": { + "start": "00:09:15", + "end": "00:09:16" + } + } + ], + "question_timestamps": "00:09:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the room in front of you.", + "B": "In the room to your left.", + "C": "The laptop was in the room to your right when you checked your position.", + "D": "The laptop was in the room behind you.", + "E": "The laptop was in the room above you rather than behind you." + }, + "answer_label": "D", + "answer_text": "The laptop was in the room behind you.", + "q_id": "egolife_A1_D6_012_7_T2", + "correct_answer": "The laptop was in the room behind you.", + "original_options": { + "A": "In the room in front of you.", + "B": "In the room to your left.", + "C": "The laptop was in the room to your right when you checked your position.", + "D": "The laptop was in the room behind you.", + "E": "The laptop was in the room above you rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_7_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the laptop relative to me right now?", + "answer": "The laptop was in the room downstairs from you.", + "question_chinese": "现在笔记本电脑相对我在什么方位？", + "answer_chinese": "笔记本电脑在你楼下的房间里。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:09:01", + "end": "00:09:02" + } + }, + { + "description": "I turned 90 degrees and left the dormitory.", + "timestamps": { + "start": "00:09:15", + "end": "00:09:16" + } + }, + { + "description": "You went upstairs.", + "timestamps": { + "start": "00:10:13", + "end": "00:10:14" + } + } + ], + "question_timestamps": "00:10:38", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the room upstairs from you.", + "B": "The laptop was in the room downstairs from you.", + "C": "In the room behind you.", + "D": "The laptop was positioned in front of you instead of downstairs from you.", + "E": "The laptop was in the bedroom to your back-left, not in a downstairs room." + }, + "answer_label": "B", + "answer_text": "The laptop was in the room downstairs from you.", + "q_id": "egolife_A1_D6_012_7_T1", + "correct_answer": "The laptop was in the room downstairs from you.", + "original_options": { + "A": "In the room upstairs from you.", + "B": "The laptop was in the room downstairs from you.", + "C": "In the room behind you.", + "D": "The laptop was positioned in front of you instead of downstairs from you.", + "E": "The laptop was in the bedroom to your back-left, not in a downstairs room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_1_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the laptop relative to me right now?", + "answer": "The laptop was in front of you.", + "question_chinese": "现在笔记本电脑相对我来说在什么方位？", + "answer_chinese": "笔记本电脑在你的面前。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:09:01", + "end": "00:09:02" + } + } + ], + "question_timestamps": "00:09:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The laptop was in front of you.", + "B": "The laptop was behind you.", + "C": "The laptop was to your left.", + "D": "The laptop was to your right when you were looking around the room.", + "E": "The laptop was above you rather than directly in front of you." + }, + "answer_label": "A", + "answer_text": "The laptop was in front of you.", + "q_id": "egolife_A1_D6_012_1_T2", + "correct_answer": "The laptop was in front of you.", + "original_options": { + "A": "The laptop was in front of you.", + "B": "The laptop was behind you.", + "C": "The laptop was to your left.", + "D": "The laptop was to your right when you were looking around the room.", + "E": "The laptop was above you rather than directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_1_T1", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the laptop relative to me right now?", + "answer": "The laptop was in the bedroom to your back-left.", + "question_chinese": "现在笔记本电脑相对我来说在什么方位？", + "answer_chinese": "笔记本电脑在你的左后方的卧室里。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:09:01", + "end": "00:09:02" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:09:15", + "end": "00:09:16" + } + } + ], + "question_timestamps": "00:09:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In the bedroom to your back-right.", + "B": "In the bedroom to your front-left.", + "C": "The laptop was in front of you, not in the bedroom to your back-left.", + "D": "The laptop was in the room behind you rather than back-left in the bedroom.", + "E": "The laptop was in the bedroom to your back-left." + }, + "answer_label": "E", + "answer_text": "The laptop was in the bedroom to your back-left.", + "q_id": "egolife_A1_D6_012_1_T1", + "correct_answer": "The laptop was in the bedroom to your back-left.", + "original_options": { + "A": "In the bedroom to your back-right.", + "B": "In the bedroom to your front-left.", + "C": "The laptop was in front of you, not in the bedroom to your back-left.", + "D": "The laptop was in the room behind you rather than back-left in the bedroom.", + "E": "The laptop was in the bedroom to your back-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_8_T1", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the table in front of me, which is closer to me: the handicraft item or the phone?", + "answer": "The handicraft on the table in front of you was closer to you.", + "question_chinese": "面前桌上的手工作品和手机哪个离我更近？", + "answer_chinese": "面前桌上的手工作品离我更近", + "evidence": [ + { + "description": "I'm doing handcrafts with my hands, and the phone is on the desktop.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:01:23", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "answer_label": "A", + "answer_text": "The handicraft on the table in front of you was closer to you.", + "q_id": "egolife_A1_D6_012_8_T1", + "correct_answer": "The handicraft on the table in front of you was closer to you.", + "original_options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_8_T2", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the table in front of me, which is closer to me: the handicraft item or the phone?", + "answer": "The phone on the table in front of you was closer to you.", + "question_chinese": "面前桌上的手工作品和手机哪个离我更近？", + "answer_chinese": "面前桌上的手机离我更近", + "evidence": [ + { + "description": "I'm doing handcrafts with my hands, and the phone is on the desktop.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "You put down the handmade work and picked up the phone from the table.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:23" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "answer_label": "B", + "answer_text": "The phone on the table in front of you was closer to you.", + "q_id": "egolife_A1_D6_012_8_T2", + "correct_answer": "The phone on the table in front of you was closer to you.", + "original_options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_4_T1", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the table in front of me, which is closer to me: the handicraft item or the phone?", + "answer": "The handicraft on the table in front of you was closer to you.", + "question_chinese": "面前桌上的手工作品和手机那个离我更近？", + "answer_chinese": "面前桌上的手工作品离你更近", + "evidence": [ + { + "description": "I'm doing handcrafts with my hands, and the phone is on the desktop.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:01:33", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "answer_label": "A", + "answer_text": "The handicraft on the table in front of you was closer to you.", + "q_id": "egolife_A1_D6_012_4_T1", + "correct_answer": "The handicraft on the table in front of you was closer to you.", + "original_options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_9_T1", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the table in front of me, which is closer to me: the handicraft item or the phone?", + "answer": "The phone on the table in front of you was closer to you.", + "question_chinese": "面前桌上的手工作品和手机那个离我更近？", + "answer_chinese": "面前桌上的手机离你更近", + "evidence": [ + { + "description": "You put down the handmade work and picked up the phone from the table.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:23" + } + } + ], + "question_timestamps": "00:03:25", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "answer_label": "B", + "answer_text": "The phone on the table in front of you was closer to you.", + "q_id": "egolife_A1_D6_012_9_T1", + "correct_answer": "The phone on the table in front of you was closer to you.", + "original_options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_4_T2", + "qid_num": 23, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the table in front of me, which is closer to me: the handicraft item or the phone?", + "answer": "The phone on the table in front of you was closer to you.", + "question_chinese": "面前桌上的手工作品和手机那个离我更近？", + "answer_chinese": "面前桌上的手机离你更近", + "evidence": [ + { + "description": "You put down the handmade work and picked up the phone from the table.", + "timestamps": { + "start": "00:03:22", + "end": "00:03:24" + } + } + ], + "question_timestamps": "00:03:26", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "answer_label": "B", + "answer_text": "The phone on the table in front of you was closer to you.", + "q_id": "egolife_A1_D6_012_4_T2", + "correct_answer": "The phone on the table in front of you was closer to you.", + "original_options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_012", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_012_9_T2", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the table in front of me, which is closer to me: the handicraft item or the phone?", + "answer": "The handicraft on the table in front of you was closer to you.", + "question_chinese": "面前桌上的手工作品和手机那个离我更近？", + "answer_chinese": "面前桌上的手工作品离你更近", + "evidence": [ + { + "description": "You put down the handmade work and picked up the phone from the table.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:23" + } + }, + { + "description": "You took over the handmade work.", + "timestamps": { + "start": "00:04:04", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:04:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "answer_label": "A", + "answer_text": "The handicraft on the table in front of you was closer to you.", + "q_id": "egolife_A1_D6_012_9_T2", + "correct_answer": "The handicraft on the table in front of you was closer to you.", + "original_options": { + "A": "The handicraft on the table in front of you was closer to you.", + "B": "The phone on the table in front of you was closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + } +] diff --git a/QAs/egolife_A1_D6_013-mcq.json b/QAs/egolife_A1_D6_013-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e5d273e4ecda6aecfe919b8030606935392a6559 --- /dev/null +++ b/QAs/egolife_A1_D6_013-mcq.json @@ -0,0 +1,626 @@ +[ + { + "video_uid": "egolife_A1_D6_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_013_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times was it filmed in total?", + "answer": "Five times.", + "question_chinese": "一共拍了几次？", + "answer_chinese": "五次。", + "evidence": [ + { + "description": "Tapped three times, one at a time.", + "timestamps": { + "start": "00:04:46", + "end": "00:04:48" + } + } + ], + "question_timestamps": "00:04:59", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four times.", + "B": "Six times.", + "C": "It was filmed a total of three times during the entire recording session.", + "D": "Five times.", + "E": "In total, it was captured on video seven separate times across the whole scene." + }, + "answer_label": "D", + "answer_text": "Five times.", + "q_id": "egolife_A1_D6_013_3", + "correct_answer": "Five times.", + "original_options": { + "A": "Four times.", + "B": "Six times.", + "C": "It was filmed a total of three times during the entire recording session.", + "D": "Five times.", + "E": "In total, it was captured on video seven separate times across the whole scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_013_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times was it filmed in total?", + "answer": "Fourteen times.", + "question_chinese": "一共拍了几次？", + "answer_chinese": "十四次。", + "evidence": [ + { + "description": "I saw the girl take six photos first.", + "timestamps": { + "start": "00:04:52", + "end": "00:04:57" + } + }, + { + "description": "I then clapped three more times.", + "timestamps": { + "start": "00:04:57", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:11:11", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Thirteen times.", + "B": "Fifteen times.", + "C": "Altogether, it was filmed a total of twelve times during the entire session.", + "D": "Fourteen times.", + "E": "In total, it was recorded on video sixteen different times across the whole scene." + }, + "answer_label": "D", + "answer_text": "Fourteen times.", + "q_id": "egolife_A1_D6_013_4", + "correct_answer": "Fourteen times.", + "original_options": { + "A": "Thirteen times.", + "B": "Fifteen times.", + "C": "Altogether, it was filmed a total of twelve times during the entire session.", + "D": "Fourteen times.", + "E": "In total, it was recorded on video sixteen different times across the whole scene." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_013-mcq_fixed_8", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my path when approaching others for help at this moment?", + "answer": "YOU went from the second floor to the first floor.", + "question_chinese": "当我为了寻求帮助而走向他人时，我的路径是什么？", + "answer_chinese": "我从二楼到一楼。", + "evidence": [ + { + "description": "I went down the stairs twice ", + "timestamps": { + "start": "00:08:16", + "end": "00:09:05" + } + }, + { + "description": "I went up the stairs once.", + "timestamps": { + "start": "00:08:16", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:09:06", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "YOU went upstairs.", + "B": "YOU stayed on the second floor.", + "C": "YOU went from the second floor to the first floor.", + "D": "YOU went from the first floor to the second floor to ask someone for help.", + "E": "YOU started down the stairs but turned back before reaching the first floor." + }, + "answer_label": "C", + "answer_text": "YOU went from the second floor to the first floor.", + "q_id": "egolife_A1_D6_013-mcq_fixed_8", + "correct_answer": "YOU went from the second floor to the first floor.", + "original_options": { + "A": "YOU went upstairs.", + "B": "YOU stayed on the second floor.", + "C": "YOU went from the second floor to the first floor.", + "D": "YOU went from the first floor to the second floor to ask someone for help.", + "E": "YOU started down the stairs but turned back before reaching the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 6, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_013-mcq_fixed_9", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my path when approaching others for help at this moment?", + "answer": "YOU went from the first floor to the second floor.", + "question_chinese": "当我为了寻求帮助而走向他人时，我的路径是什么？", + "answer_chinese": "我从一楼到二楼。", + "evidence": [ + { + "description": "First went down twice.", + "timestamps": { + "start": "00:08:16", + "end": "00:11:55" + } + }, + { + "description": "I then up once, then up again, then down once.", + "timestamps": { + "start": "00:08:16", + "end": "00:09:05" + } + }, + { + "description": "I then up again, then down once.", + "timestamps": { + "start": "00:08:16", + "end": "00:09:05" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "YOU went downstairs.", + "B": "YOU stayed on the first floor.", + "C": "YOU went from the first floor to the second floor.", + "D": "YOU went from the second floor to the first floor to ask someone for help.", + "E": "YOU headed toward the stairs but turned back before reaching the second floor." + }, + "answer_label": "C", + "answer_text": "YOU went from the first floor to the second floor.", + "q_id": "egolife_A1_D6_013-mcq_fixed_9", + "correct_answer": "YOU went from the first floor to the second floor.", + "original_options": { + "A": "YOU went downstairs.", + "B": "YOU stayed on the first floor.", + "C": "YOU went from the first floor to the second floor.", + "D": "YOU went from the second floor to the first floor to ask someone for help.", + "E": "YOU headed toward the stairs but turned back before reaching the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 6, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_013-mcq_fixed_6", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take when I went to get water?", + "answer": "First, come out of the bedroom, walk to the laptop and sit down, then go to the side of the balcony to get water.", + "question_chinese": "我去接水时经过了什么路径？", + "answer_chinese": "先从卧室出来，走到笔记本旁坐下，然后走到阳台旁边，接水。", + "evidence": [ + { + "description": "I am sitting next to my laptop and using it.", + "timestamps": { + "start": "00:06:46", + "end": "00:06:59" + } + }, + { + "description": "I am standing by the balcony.", + "timestamps": { + "start": "00:07:24", + "end": "00:07:30" + } + } + ], + "question_timestamps": "00:10:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went straight to the balcony.", + "B": "First, come out of the bedroom, walk to the laptop and sit down, then go to the side of the balcony to get water.", + "C": "You got water from the kitchen.", + "D": "You left the bedroom, sat at the laptop briefly, then turned toward the living room before getting water.", + "E": "You came out of the bedroom, walked to the laptop to sit, then returned toward the hallway instead of the balcony side." + }, + "answer_label": "B", + "answer_text": "First, come out of the bedroom, walk to the laptop and sit down, then go to the side of the balcony to get water.", + "q_id": "egolife_A1_D6_013-mcq_fixed_6", + "correct_answer": "First, come out of the bedroom, walk to the laptop and sit down, then go to the side of the balcony to get water.", + "original_options": { + "A": "You went straight to the balcony.", + "B": "First, come out of the bedroom, walk to the laptop and sit down, then go to the side of the balcony to get water.", + "C": "You got water from the kitchen.", + "D": "You left the bedroom, sat at the laptop briefly, then turned toward the living room before getting water.", + "E": "You came out of the bedroom, walked to the laptop to sit, then returned toward the hallway instead of the balcony side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 22 + ] + }, + { + "video_uid": "egolife_A1_D6_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_013-mcq_fixed_7", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take when I went to get water?", + "answer": "First, come out of the bedroom, walk to the laptop and sit down, then go to the side of the balcony, then go downstairs, walk to the kitchen, go upstairs, go downstairs again, enter the bedroom, go upstairs, and get water.", + "question_chinese": "我去接水时经过了什么路径？", + "answer_chinese": "先从卧室出来，然后走到阳台旁边，然后下楼，走到厨房，上楼，下楼，进入卧室，上楼，接水。", + "evidence": [ + { + "description": "I am sitting next to my laptop and using it.", + "timestamps": { + "start": "00:06:46", + "end": "00:06:59" + } + }, + { + "description": "I am standing by the balcony.", + "timestamps": { + "start": "00:07:24", + "end": "00:07:30" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went downstairs once and got water.", + "B": "You skipped the balcony and went downstairs.", + "C": "You came out of the bedroom, sat at the laptop, then went downstairs to the kitchen and got water there.", + "D": "First, come out of the bedroom, walk to the laptop and sit down, then go to the side of the balcony, then go downstairs, walk to the kitchen, go upstairs, go downstairs again, enter the bedroom, go upstairs, and get water.", + "E": "You left the bedroom, went to the laptop, visited the balcony side, then went upstairs once and got water without looping again." + }, + "answer_label": "D", + "answer_text": "First, come out of the bedroom, walk to the laptop and sit down, then go to the side of the balcony, then go downstairs, walk to the kitchen, go upstairs, go downstairs again, enter the bedroom, go upstairs, and get water.", + "q_id": "egolife_A1_D6_013-mcq_fixed_7", + "correct_answer": "First, come out of the bedroom, walk to the laptop and sit down, then go to the side of the balcony, then go downstairs, walk to the kitchen, go upstairs, go downstairs again, enter the bedroom, go upstairs, and get water.", + "original_options": { + "A": "You went downstairs once and got water.", + "B": "You skipped the balcony and went downstairs.", + "C": "You came out of the bedroom, sat at the laptop, then went downstairs to the kitchen and got water there.", + "D": "First, come out of the bedroom, walk to the laptop and sit down, then go to the side of the balcony, then go downstairs, walk to the kitchen, go upstairs, go downstairs again, enter the bedroom, go upstairs, and get water.", + "E": "You left the bedroom, went to the laptop, visited the balcony side, then went upstairs once and got water without looping again." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 22 + ] + }, + { + "video_uid": "egolife_A1_D6_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_013-mcq_fixed_4", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the notebook on the table in relation to me?", + "answer": "The notebook is behind you to the right.", + "question_chinese": "桌子上的笔记本在在我的什么位置？", + "answer_chinese": "笔记本在我的右后方。", + "evidence": [ + { + "description": "When I first arrived at the door, the laptop was in front of me to the right.", + "timestamps": { + "start": "00:05:09", + "end": "00:05:14" + } + }, + { + "description": "When I face the balcony, the laptop is behind me.", + "timestamps": { + "start": "00:05:19", + "end": "00:05:25" + } + }, + { + "description": "I turned 90 degrees.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:25" + } + } + ], + "question_timestamps": "00:05:46", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "The notebook is in front of you.", + "B": "The notebook is behind you left.", + "C": "The notebook is directly to your right side on the table, not behind your body.", + "D": "The notebook is behind you to the right.", + "E": "The notebook sits in front-left of you on the tabletop, closer to the edge nearest your hands." + }, + "answer_label": "D", + "answer_text": "The notebook is behind you to the right.", + "q_id": "egolife_A1_D6_013-mcq_fixed_4", + "correct_answer": "The notebook is behind you to the right.", + "original_options": { + "A": "The notebook is in front of you.", + "B": "The notebook is behind you left.", + "C": "The notebook is directly to your right side on the table, not behind your body.", + "D": "The notebook is behind you to the right.", + "E": "The notebook sits in front-left of you on the tabletop, closer to the edge nearest your hands." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_013-mcq_fixed_5", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the notebook on the table in relation to me?", + "answer": "The laptop is directly in front of you.", + "question_chinese": "桌子上的笔记本在在我的什么位置？", + "answer_chinese": "笔记本在我的正前方。", + "evidence": [ + { + "description": "When I first arrived at the door, the laptop was in front of me to the right.", + "timestamps": { + "start": "00:05:09", + "end": "00:05:10" + } + }, + { + "description": "I turned 90 degrees.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:21" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The laptop is directly behind you.", + "B": "The laptop is directly to your left.", + "C": "The laptop is slightly off to your right, closer to your dominant hand than the center of the table.", + "D": "The laptop sits diagonally in front of you, angled toward the far left edge of the table.", + "E": "The laptop is directly in front of you." + }, + "answer_label": "E", + "answer_text": "The laptop is directly in front of you.", + "q_id": "egolife_A1_D6_013-mcq_fixed_5", + "correct_answer": "The laptop is directly in front of you.", + "original_options": { + "A": "The laptop is directly behind you.", + "B": "The laptop is directly to your left.", + "C": "The laptop is slightly off to your right, closer to your dominant hand than the center of the table.", + "D": "The laptop sits diagonally in front of you, angled toward the far left edge of the table.", + "E": "The laptop is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_013_1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many dolls are on the ground now?", + "answer": "Two.", + "question_chinese": "现在地上有几个玩偶？", + "answer_chinese": "俩个。", + "evidence": [ + { + "description": "I saw a doll being thrown off the bed.", + "timestamps": { + "start": "00:01:07", + "end": "00:01:08" + } + }, + { + "description": "I saw another doll being thrown on the ground.", + "timestamps": { + "start": "00:01:08", + "end": "00:01:09" + } + } + ], + "question_timestamps": "00:01:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four dolls are on the ground now, spread out near each other rather than in a single pile.", + "E": "Five dolls are on the ground now, with several more visible than before after they were moved." + }, + "answer_label": "A", + "answer_text": "Two.", + "q_id": "egolife_A1_D6_013_1", + "correct_answer": "Two.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "Four dolls are on the ground now, spread out near each other rather than in a single pile.", + "E": "Five dolls are on the ground now, with several more visible than before after they were moved." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_013", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_013_2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many dolls are on the ground now?", + "answer": "Four.", + "question_chinese": "现在地上有几个玩偶？", + "answer_chinese": "四个。", + "evidence": [ + { + "description": "A pink and an orange doll were thrown to the ground。", + "timestamps": { + "start": "00:01:14", + "end": "00:01:16" + } + } + ], + "question_timestamps": "00:10:16", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Two.", + "C": "Three.", + "D": "Only one doll is on the ground right now, not four at all.", + "E": "Five dolls are on the ground now, which is more than four." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "egolife_A1_D6_013_2", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Two.", + "C": "Three.", + "D": "Only one doll is on the ground right now, not four at all.", + "E": "Five dolls are on the ground now, which is more than four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + } +] diff --git a/QAs/egolife_A1_D6_014-mcq.json b/QAs/egolife_A1_D6_014-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c7bb52fdb1cc06593d577659662c41fa0b59b1fb --- /dev/null +++ b/QAs/egolife_A1_D6_014-mcq.json @@ -0,0 +1,1104 @@ +[ + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014-mcq_fixed_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path do you need to take to go from the first floor to the second floor?", + "answer": "Passed through two steps.", + "question_chinese": "从一楼到二楼要经过什么路径？", + "answer_chinese": "经过俩层台阶。", + "evidence": [ + { + "description": "First went up one floor，It's another level up.", + "timestamps": { + "start": "00:04:04", + "end": "00:04:12" + } + } + ], + "question_timestamps": "00:04:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Passed through two steps.", + "B": "Passed through one step.", + "C": "Passed through three steps.", + "D": "You go up by passing through four steps from the first floor to the second floor.", + "E": "You reach the second floor by passing through a single long staircase rather than just two steps." + }, + "answer_label": "A", + "answer_text": "Passed through two steps.", + "q_id": "egolife_A1_D6_014-mcq_fixed_6", + "correct_answer": "Passed through two steps.", + "original_options": { + "A": "Passed through two steps.", + "B": "Passed through one step.", + "C": "Passed through three steps.", + "D": "You go up by passing through four steps from the first floor to the second floor.", + "E": "You reach the second floor by passing through a single long staircase rather than just two steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014-mcq_fixed_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path do you need to take to go from the first floor to the second floor?", + "answer": "YOU first went up one flight of stairs, then another flight of stairs.", + "question_chinese": "从一楼到二楼要经过什么路径？", + "answer_chinese": "我先上了一层楼梯，又上了第二层楼梯。", + "evidence": [ + { + "description": "First went up one floor，It's another level up.", + "timestamps": { + "start": "00:04:04", + "end": "00:04:14" + } + } + ], + "question_timestamps": "00:10:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You took the elevator to the second floor.", + "B": "You went up one flight, then down.", + "C": "YOU first went up one flight of stairs, then another flight of stairs.", + "D": "You first went down one flight of stairs, then went up another flight of stairs to reach the second floor.", + "E": "You went up a single flight of stairs and reached the second floor without taking a second flight." + }, + "answer_label": "C", + "answer_text": "YOU first went up one flight of stairs, then another flight of stairs.", + "q_id": "egolife_A1_D6_014-mcq_fixed_7", + "correct_answer": "YOU first went up one flight of stairs, then another flight of stairs.", + "original_options": { + "A": "You took the elevator to the second floor.", + "B": "You went up one flight, then down.", + "C": "YOU first went up one flight of stairs, then another flight of stairs.", + "D": "You first went down one flight of stairs, then went up another flight of stairs to reach the second floor.", + "E": "You went up a single flight of stairs and reached the second floor without taking a second flight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 20, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_5", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the bread on the refrigerator from me now?", + "answer": "To my right.", + "question_chinese": "冰箱上的面包现在我的哪个方位？", + "answer_chinese": "在我的右方。", + "evidence": [ + { + "description": "When I walked in front of the refrigerator, the bread was right in front of me.", + "timestamps": { + "start": "00:03:32", + "end": "00:03:33" + } + }, + { + "description": "Turn around 180 degrees.", + "timestamps": { + "start": "00:03:34", + "end": "00:03:35" + } + } + ], + "question_timestamps": "00:03:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To my left.", + "B": "In front of me.", + "C": "It was slightly behind me on the left side as I faced the refrigerator.", + "D": "To my right.", + "E": "It was above and a bit to the left of me, near the top edge of the refrigerator." + }, + "answer_label": "D", + "answer_text": "To my right.", + "q_id": "egolife_A1_D6_014_5", + "correct_answer": "To my right.", + "original_options": { + "A": "To my left.", + "B": "In front of me.", + "C": "It was slightly behind me on the left side as I faced the refrigerator.", + "D": "To my right.", + "E": "It was above and a bit to the left of me, near the top edge of the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_6", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the bread on the refrigerator from me now?", + "answer": "To my left rear.", + "question_chinese": "冰箱上的面包现在我的哪个方位？", + "answer_chinese": "在我的左后方。", + "evidence": [ + { + "description": "When I walked towards the refrigerator, the bread was directly in front of me.", + "timestamps": { + "start": "00:03:39", + "end": "00:03:40" + } + }, + { + "description": "Keep a distance from the refrigerator.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:08:37", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To my right rear.", + "B": "Directly behind me.", + "C": "It was in front of you and slightly to the left as you stood near the refrigerator.", + "D": "To my left rear.", + "E": "It was to your right side, a little behind you, relative to where you were standing." + }, + "answer_label": "D", + "answer_text": "To my left rear.", + "q_id": "egolife_A1_D6_014_6", + "correct_answer": "To my left rear.", + "original_options": { + "A": "To my right rear.", + "B": "Directly behind me.", + "C": "It was in front of you and slightly to the left as you stood near the refrigerator.", + "D": "To my left rear.", + "E": "It was to your right side, a little behind you, relative to where you were standing." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014-mcq_fixed_8", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "There are several large dolls on the bed.", + "answer": "Five.", + "question_chinese": "床上有几个大的玩偶。", + "answer_chinese": "四个。", + "evidence": [ + { + "description": "There is a lion by the pillow, and three by the wall.", + "timestamps": { + "start": "00:05:09", + "end": "00:05:15" + } + } + ], + "question_timestamps": "00:05:16", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "There are four large dolls arranged together on the bed in a neat row.", + "D": "Five.", + "E": "You can count six large dolls sitting together on the bed near the pillows." + }, + "answer_label": "D", + "answer_text": "Five.", + "q_id": "egolife_A1_D6_014-mcq_fixed_8", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "There are four large dolls arranged together on the bed in a neat row.", + "D": "Five.", + "E": "You can count six large dolls sitting together on the bed near the pillows." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014-mcq_fixed_9", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "There are several large dolls on the bed.", + "answer": "Six.", + "question_chinese": "床上有几个大的玩偶。", + "answer_chinese": "五个。", + "evidence": [ + { + "description": "There is a lion by the pillow, and three by the wall，there are also some small dolls", + "timestamps": { + "start": "00:05:09", + "end": "00:05:15" + } + } + ], + "question_timestamps": "00:10:07", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "Six.", + "D": "There are five large dolls placed on the bed, lined up close together.", + "E": "You can see seven large dolls spread out across the bed near the foot area." + }, + "answer_label": "C", + "answer_text": "Six.", + "q_id": "egolife_A1_D6_014-mcq_fixed_9", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "Six.", + "D": "There are five large dolls placed on the bed, lined up close together.", + "E": "You can see seven large dolls spread out across the bed near the foot area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014-mcq_fixed_2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "When YOU stepped out of the bedroom door, the laptop was behind you.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "我刚出卧室门时，笔记本电脑在我的后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:03:04", + "end": "00:03:05" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:03:14", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:03:16", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "It was to your left.", + "C": "When YOU stepped out of the bedroom door, the laptop was ahead of you, slightly to the right.", + "D": "When YOU stepped out of the bedroom door, the laptop was behind you.", + "E": "When YOU stepped out of the bedroom door, the laptop was off to your left side in the hallway." + }, + "answer_label": "D", + "answer_text": "When YOU stepped out of the bedroom door, the laptop was behind you.", + "q_id": "egolife_A1_D6_014-mcq_fixed_2", + "correct_answer": "When YOU stepped out of the bedroom door, the laptop was behind you.", + "original_options": { + "A": "It was in front of you.", + "B": "It was to your left.", + "C": "When YOU stepped out of the bedroom door, the laptop was ahead of you, slightly to the right.", + "D": "When YOU stepped out of the bedroom door, the laptop was behind you.", + "E": "When YOU stepped out of the bedroom door, the laptop was off to your left side in the hallway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_2_T3", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I had just walked out of the bedroom door, where was the laptop relative to me?", + "answer": "Right now you were turned sideways to the door, and the laptop was directly to your left.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "你现在侧对着门，笔记本电脑在你的正左边。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:03:04", + "end": "00:03:05" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:03:14", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:03:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The laptop was in front of you.", + "B": "Right now you were turned sideways to the door, and the laptop was directly to your left.", + "C": "The laptop was to your right.", + "D": "Right now you were turned sideways to the door, and the laptop was directly behind you.", + "E": "Right now you were turned sideways to the door, and the laptop was slightly in front of your right side." + }, + "answer_label": "B", + "answer_text": "Right now you were turned sideways to the door, and the laptop was directly to your left.", + "q_id": "egolife_A1_D6_014_2_T3", + "correct_answer": "Right now you were turned sideways to the door, and the laptop was directly to your left.", + "original_options": { + "A": "The laptop was in front of you.", + "B": "Right now you were turned sideways to the door, and the laptop was directly to your left.", + "C": "The laptop was to your right.", + "D": "Right now you were turned sideways to the door, and the laptop was directly behind you.", + "E": "Right now you were turned sideways to the door, and the laptop was slightly in front of your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_5_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I had just walked out of the bedroom door, where was the laptop relative to me?", + "answer": "Right now you were on the stairs below the bedroom, and the computer was above you.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "你现在在楼梯上卧室的下面，电脑在你的上方", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:03:04", + "end": "00:03:05" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:03:14", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:04:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer was in front of you.", + "B": "The computer was behind you.", + "C": "Right now you were on the stairs below the bedroom, and the computer was to your left.", + "D": "Right now you were on the stairs below the bedroom, and the computer was above you.", + "E": "Right now you were on the stairs below the bedroom, and the computer was down the stairs behind you." + }, + "answer_label": "D", + "answer_text": "Right now you were on the stairs below the bedroom, and the computer was above you.", + "q_id": "egolife_A1_D6_014_5_T1", + "correct_answer": "Right now you were on the stairs below the bedroom, and the computer was above you.", + "original_options": { + "A": "The computer was in front of you.", + "B": "The computer was behind you.", + "C": "Right now you were on the stairs below the bedroom, and the computer was to your left.", + "D": "Right now you were on the stairs below the bedroom, and the computer was above you.", + "E": "Right now you were on the stairs below the bedroom, and the computer was down the stairs behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_4", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I just stepped out of the bedroom, what direction was my laptop in relative to me?", + "answer": "When I stepped out of the bedroom door, the laptop was behind me to the left.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "我刚出卧室门时，笔记本电脑在我的左后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:03:04", + "end": "00:03:05" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:03:14", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:10:04", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was behind me to the right.", + "B": "When I stepped out of the bedroom door, the laptop was behind me to the left.", + "C": "It was in front of me.", + "D": "When I stepped out of the bedroom door, the laptop was ahead of me to the left.", + "E": "When I stepped out of the bedroom door, the laptop was directly to my right, slightly in front of me." + }, + "answer_label": "B", + "answer_text": "When I stepped out of the bedroom door, the laptop was behind me to the left.", + "q_id": "egolife_A1_D6_014_4", + "correct_answer": "When I stepped out of the bedroom door, the laptop was behind me to the left.", + "original_options": { + "A": "It was behind me to the right.", + "B": "When I stepped out of the bedroom door, the laptop was behind me to the left.", + "C": "It was in front of me.", + "D": "When I stepped out of the bedroom door, the laptop was ahead of me to the left.", + "E": "When I stepped out of the bedroom door, the laptop was directly to my right, slightly in front of me." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_5_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I had just walked out of the bedroom door, where was the laptop relative to me?", + "answer": "When you had just walked out of the bedroom door, the laptop was behind you.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "我刚出卧室门时，笔记本电脑在我的后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:03:04", + "end": "00:03:05" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:03:14", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:11:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "It was to your left.", + "C": "Right after leaving the bedroom, the laptop was positioned ahead of you, near your line of sight.", + "D": "When you had just walked out of the bedroom door, the laptop was behind you.", + "E": "As you stepped out of the bedroom door, the laptop sat off to your right side rather than behind you." + }, + "answer_label": "D", + "answer_text": "When you had just walked out of the bedroom door, the laptop was behind you.", + "q_id": "egolife_A1_D6_014_5_T2", + "correct_answer": "When you had just walked out of the bedroom door, the laptop was behind you.", + "original_options": { + "A": "It was in front of you.", + "B": "It was to your left.", + "C": "Right after leaving the bedroom, the laptop was positioned ahead of you, near your line of sight.", + "D": "When you had just walked out of the bedroom door, the laptop was behind you.", + "E": "As you stepped out of the bedroom door, the laptop sat off to your right side rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_2_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I had just walked out of the bedroom door, where was the laptop relative to me?", + "answer": "When you had just walked out of the bedroom door, the laptop was to your back-left.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "你刚出卧室门时，笔记本电脑在你的左后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:03:04", + "end": "00:03:05" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:03:14", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:11:54", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "When you had just walked out of the bedroom door, the laptop was to your back-left.", + "B": "It was to your back-right.", + "C": "It was to your front-left.", + "D": "Just after you exited the bedroom, the laptop was positioned in front of you on the left side.", + "E": "As you walked out the bedroom door, the laptop was behind you but offset to your right." + }, + "answer_label": "A", + "answer_text": "When you had just walked out of the bedroom door, the laptop was to your back-left.", + "q_id": "egolife_A1_D6_014_2_T2", + "correct_answer": "When you had just walked out of the bedroom door, the laptop was to your back-left.", + "original_options": { + "A": "When you had just walked out of the bedroom door, the laptop was to your back-left.", + "B": "It was to your back-right.", + "C": "It was to your front-left.", + "D": "Just after you exited the bedroom, the laptop was positioned in front of you on the left side.", + "E": "As you walked out the bedroom door, the laptop was behind you but offset to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_1_T2", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I’m standing right now, which is closer: the drinks or the chopsticks?", + "answer": "The drinks were closer. The drinks were to your front-right, and the chopsticks were directly in front of you.", + "question_chinese": "我现在站的位置距离饮料和筷子那个比较近？", + "answer_chinese": "距离饮料更近，饮料在你右前方，筷子在你正前方。", + "evidence": [ + { + "description": "The drink is directly in front of me and to the left of the table.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + }, + { + "description": "Chopsticks are in front of me on the right, to the right of the table.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:21" + } + } + ], + "question_timestamps": "00:03:01", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The chopsticks were closer.", + "B": "The drinks were farther.", + "C": "The chopsticks were closer, since they were directly in front of you while the drinks sat off to your front-right.", + "D": "The chopsticks ended up nearer to you; the drinks were slightly farther away despite being at your front-right.", + "E": "The drinks were closer. The drinks were to your front-right, and the chopsticks were directly in front of you." + }, + "answer_label": "E", + "answer_text": "The drinks were closer. The drinks were to your front-right, and the chopsticks were directly in front of you.", + "q_id": "egolife_A1_D6_014_1_T2", + "correct_answer": "The drinks were closer. The drinks were to your front-right, and the chopsticks were directly in front of you.", + "original_options": { + "A": "The chopsticks were closer.", + "B": "The drinks were farther.", + "C": "The chopsticks were closer, since they were directly in front of you while the drinks sat off to your front-right.", + "D": "The chopsticks ended up nearer to you; the drinks were slightly farther away despite being at your front-right.", + "E": "The drinks were closer. The drinks were to your front-right, and the chopsticks were directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_4_T1", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I’m standing right now, which is closer: the drinks or the chopsticks?", + "answer": "The chopsticks were closer.", + "question_chinese": "我现在站的位置距离饮料和筷子那个比较近？", + "answer_chinese": "距离筷子更近。", + "evidence": [ + { + "description": "The drink is to my left front, about three meters away from me.", + "timestamps": { + "start": "00:03:16", + "end": "00:03:17" + } + }, + { + "description": "Chopsticks are in front of me, about four meters away.", + "timestamps": { + "start": "00:03:17", + "end": "00:03:18" + } + }, + { + "description": "The distance between the drink and the chopsticks is about 1 meter.", + "timestamps": { + "start": "00:03:18", + "end": "00:03:19" + } + } + ], + "question_timestamps": "00:03:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The drinks were closer.", + "B": "The chopsticks were closer.", + "C": "The drinks were nearer.", + "D": "The drinks were closer than the chopsticks, even though the chopsticks were still within reach in front of you.", + "E": "The drinks were the nearer item, with the chopsticks positioned a bit farther away from where you stood." + }, + "answer_label": "B", + "answer_text": "The chopsticks were closer.", + "q_id": "egolife_A1_D6_014_4_T1", + "correct_answer": "The chopsticks were closer.", + "original_options": { + "A": "The drinks were closer.", + "B": "The chopsticks were closer.", + "C": "The drinks were nearer.", + "D": "The drinks were closer than the chopsticks, even though the chopsticks were still within reach in front of you.", + "E": "The drinks were the nearer item, with the chopsticks positioned a bit farther away from where you stood." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to where I am standing now, the drinks or the chopsticks?", + "answer": "It is about two steps away from the chopsticks and about three steps away from the drink.", + "question_chinese": "我现在站的位置距离饮料和筷子那个比较近？", + "answer_chinese": "距离筷子更近，大概俩歩远，距离饮料大概三步远。", + "evidence": [ + { + "description": "The drink is directly in front of me and to the left of the table.", + "timestamps": { + "start": "00:03:19", + "end": "00:03:20" + } + }, + { + "description": "Chopsticks are in front of me on the right, to the right of the table.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:21" + } + }, + { + "description": "The distance between the drink and the chopsticks is about 1 meter.", + "timestamps": { + "start": "00:03:19", + "end": "00:03:20" + } + } + ], + "question_timestamps": "00:03:31", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "It is about three steps away from the chopsticks and about two steps away from the drink.", + "B": "It is about two steps away from the chopsticks and about three steps away from the drink." + }, + "answer_label": "B", + "answer_text": "It is about two steps away from the chopsticks and about three steps away from the drink.", + "q_id": "egolife_A1_D6_014_1", + "correct_answer": "It is about two steps away from the chopsticks and about three steps away from the drink.", + "original_options": { + "A": "It is about three steps away from the chopsticks and about two steps away from the drink.", + "B": "It is about two steps away from the chopsticks and about three steps away from the drink." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_1_T3", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I’m standing right now, which is closer: the drinks or the chopsticks?", + "answer": "The drinks were closer.", + "question_chinese": "我现在站的位置距离饮料和筷子那个比较近？", + "answer_chinese": "距离饮料更近。", + "evidence": [ + { + "description": "The drink is to my left front, about three meters away from me.", + "timestamps": { + "start": "00:03:36", + "end": "00:03:37" + } + }, + { + "description": "Chopsticks are in front of me, about four meters away.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:38" + } + }, + { + "description": "The distance between the drink and the chopsticks is about 1 meter.", + "timestamps": { + "start": "00:03:36", + "end": "00:03:38" + } + } + ], + "question_timestamps": "00:03:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The drinks were closer.", + "B": "The chopsticks were closer.", + "C": "The drinks were farther.", + "D": "The chopsticks were closer than the drinks, with the drinks sitting a bit farther away off to the right.", + "E": "The chopsticks were the nearer item from where you stood, and the drinks were slightly farther in comparison." + }, + "answer_label": "A", + "answer_text": "The drinks were closer.", + "q_id": "egolife_A1_D6_014_1_T3", + "correct_answer": "The drinks were closer.", + "original_options": { + "A": "The drinks were closer.", + "B": "The chopsticks were closer.", + "C": "The drinks were farther.", + "D": "The chopsticks were closer than the drinks, with the drinks sitting a bit farther away off to the right.", + "E": "The chopsticks were the nearer item from where you stood, and the drinks were slightly farther in comparison." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_1_T1", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I’m standing right now, which is closer: the drinks or the chopsticks?", + "answer": "The chopsticks were closer—about two steps away—while the drinks were about three steps away.", + "question_chinese": "我现在站的位置距离饮料和筷子那个比较近？", + "answer_chinese": "距离筷子更近，大概俩歩远，距离饮料大概三步远。", + "evidence": [ + { + "description": "The drink is directly in front of me and to the left of the table.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:21" + } + }, + { + "description": "Chopsticks are in front of me on the right, to the right of the table.", + "timestamps": { + "start": "00:03:21", + "end": "00:03:22" + } + }, + { + "description": "The distance between the drink and the chopsticks is about 1 meter.", + "timestamps": { + "start": "00:03:19", + "end": "00:03:20" + } + } + ], + "question_timestamps": "00:03:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The drinks were closer.", + "B": "The drinks were two steps away.", + "C": "The chopsticks were closer—about one step away—while the drinks were about two steps away.", + "D": "The chopsticks were closer—about two steps away—while the drinks were about three steps away.", + "E": "The drinks were closer—around two steps away—while the chopsticks were farther at roughly three steps." + }, + "answer_label": "D", + "answer_text": "The chopsticks were closer—about two steps away—while the drinks were about three steps away.", + "q_id": "egolife_A1_D6_014_1_T1", + "correct_answer": "The chopsticks were closer—about two steps away—while the drinks were about three steps away.", + "original_options": { + "A": "The drinks were closer.", + "B": "The drinks were two steps away.", + "C": "The chopsticks were closer—about one step away—while the drinks were about two steps away.", + "D": "The chopsticks were closer—about two steps away—while the drinks were about three steps away.", + "E": "The drinks were closer—around two steps away—while the chopsticks were farther at roughly three steps." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_014", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_014_4_T2", + "qid_num": 18, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I’m standing right now, which is closer: the drinks or the chopsticks?", + "answer": "The drinks were closer.", + "question_chinese": "我现在站的位置距离饮料和筷子那个比较近？", + "answer_chinese": "距离饮料更近。", + "evidence": [ + { + "description": "The drink is to my left front, about three meters away from me.", + "timestamps": { + "start": "00:03:39", + "end": "00:03:40" + } + }, + { + "description": "Chopsticks are in front of me, about four meters away.", + "timestamps": { + "start": "00:03:38", + "end": "00:03:39" + } + }, + { + "description": "The distance between the drink and the chopsticks is about 1 meter.", + "timestamps": { + "start": "00:03:36", + "end": "00:03:38" + } + } + ], + "question_timestamps": "00:08:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The chopsticks were closer.", + "B": "The chopsticks were nearer.", + "C": "The drinks were closer.", + "D": "The chopsticks were closer—about two steps away—while the drinks were about three steps away.", + "E": "The chopsticks were the nearer option from your spot, and the drinks were positioned slightly farther away." + }, + "answer_label": "C", + "answer_text": "The drinks were closer.", + "q_id": "egolife_A1_D6_014_4_T2", + "correct_answer": "The drinks were closer.", + "original_options": { + "A": "The chopsticks were closer.", + "B": "The chopsticks were nearer.", + "C": "The drinks were closer.", + "D": "The chopsticks were closer—about two steps away—while the drinks were about three steps away.", + "E": "The chopsticks were the nearer option from your spot, and the drinks were positioned slightly farther away." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D6_015-mcq.json b/QAs/egolife_A1_D6_015-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..38e5cbe84562336ab3dba76acc4fe3447b843216 --- /dev/null +++ b/QAs/egolife_A1_D6_015-mcq.json @@ -0,0 +1,955 @@ +[ + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different types of dolls are there?", + "answer": "One.", + "question_chinese": "一共有几种类型的玩偶？", + "answer_chinese": "一个。", + "evidence": [ + { + "description": "There are two strings hanging on the door frame, and both are green.", + "timestamps": { + "start": "00:01:24", + "end": "00:01:25" + } + } + ], + "question_timestamps": "00:01:36", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "There are two different types of dolls visible overall in the scene right now.", + "D": "You can see three distinct types of dolls, each differing in style or appearance overall.", + "E": "One." + }, + "answer_label": "E", + "answer_text": "One.", + "q_id": "egolife_A1_D6_015_3", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "There are two different types of dolls visible overall in the scene right now.", + "D": "You can see three distinct types of dolls, each differing in style or appearance overall.", + "E": "One." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_10", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different types of dolls are there?", + "answer": "Four types.", + "question_chinese": "一共有几种类型的玩偶？", + "answer_chinese": "四种。", + "evidence": [ + { + "description": "There are two yellow ones.", + "timestamps": { + "start": "00:03:14", + "end": "00:03:15" + } + }, + { + "description": "There is a white one.", + "timestamps": { + "start": "00:03:15", + "end": "00:03:16" + } + }, + { + "description": "I assemble pink dolls.", + "timestamps": { + "start": "00:03:28", + "end": "00:03:43" + } + } + ], + "question_timestamps": "00:04:15", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Three types.", + "B": "Five types.", + "C": "Four types.", + "D": "There are two different types of dolls visible overall in the scene right now.", + "E": "You can identify only one type of doll overall, with no other distinct kinds present." + }, + "answer_label": "C", + "answer_text": "Four types.", + "q_id": "egolife_A1_D6_015_10", + "correct_answer": "Four types.", + "original_options": { + "A": "Three types.", + "B": "Five types.", + "C": "Four types.", + "D": "There are two different types of dolls visible overall in the scene right now.", + "E": "You can identify only one type of doll overall, with no other distinct kinds present." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_1_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From my point of view right now, how many adhesive hooks are on the door in total?", + "answer": "From your point of view, there are two adhesive hooks on the door right now. The adhesive hooks are in front of you.", + "question_chinese": "在我的视角，现在门上共有几个粘钩？", + "answer_chinese": "在你的视角，现在门上有两个粘钩。粘钩在你的前面。", + "evidence": [ + { + "description": "I stuck another one on the left side of the door frame.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + }, + { + "description": "I see one on the right side of the door frame.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:27", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One hook.", + "B": "Three hooks.", + "C": "From your point of view, there is one adhesive hook on the door right now in front of you.", + "D": "From your point of view, there are four adhesive hooks on the door right now, and they are in front of you.", + "E": "From your point of view, there are two adhesive hooks on the door right now. The adhesive hooks are in front of you." + }, + "answer_label": "E", + "answer_text": "From your point of view, there are two adhesive hooks on the door right now. The adhesive hooks are in front of you.", + "q_id": "egolife_A1_D6_015_1_T1", + "correct_answer": "From your point of view, there are two adhesive hooks on the door right now. The adhesive hooks are in front of you.", + "original_options": { + "A": "One hook.", + "B": "Three hooks.", + "C": "From your point of view, there is one adhesive hook on the door right now in front of you.", + "D": "From your point of view, there are four adhesive hooks on the door right now, and they are in front of you.", + "E": "From your point of view, there are two adhesive hooks on the door right now. The adhesive hooks are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 19, + 22 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_1_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From my point of view right now, how many adhesive hooks are on the door in total?", + "answer": "From your point of view, there are five adhesive hooks on the door right now. The adhesive hooks are to your front-right.", + "question_chinese": "在我的视角，现在门上共有几个粘钩？", + "answer_chinese": "在你的视角，现在门上有五个粘钩。粘钩在你的右前方。", + "evidence": [ + { + "description": "I stuck three more evenly on the door frame.", + "timestamps": { + "start": "00:00:52", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "From your point of view, there are five adhesive hooks on the door right now. The adhesive hooks are to your front-right.", + "B": "Four hooks.", + "C": "Six hooks.", + "D": "From your point of view, there are three adhesive hooks on the door right now, and they are to your front-right.", + "E": "From your point of view, there are seven adhesive hooks on the door right now, positioned to your front-right." + }, + "answer_label": "A", + "answer_text": "From your point of view, there are five adhesive hooks on the door right now. The adhesive hooks are to your front-right.", + "q_id": "egolife_A1_D6_015_1_T2", + "correct_answer": "From your point of view, there are five adhesive hooks on the door right now. The adhesive hooks are to your front-right.", + "original_options": { + "A": "From your point of view, there are five adhesive hooks on the door right now. The adhesive hooks are to your front-right.", + "B": "Four hooks.", + "C": "Six hooks.", + "D": "From your point of view, there are three adhesive hooks on the door right now, and they are to your front-right.", + "E": "From your point of view, there are seven adhesive hooks on the door right now, positioned to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 21, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015-mcq_fixed_3", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I sit in front of the laptop, where is the kettle relative to me?", + "answer": "In front of you.", + "question_chinese": "当我坐在笔记本前时，水壶在我的什么方位？", + "answer_chinese": "在我的前方。", + "evidence": [ + { + "description": "The kettle is to the right of the refrigerator, and when I face the refrigerator, it is directly in front of me.", + "timestamps": { + "start": "00:06:12", + "end": "00:06:13" + } + }, + { + "description": "When I faced the balcony, I saw a watering can directly in front of me.", + "timestamps": { + "start": "00:05:46", + "end": "00:05:47" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "Behind you.", + "C": "It is slightly to your right, near the edge of the table.", + "D": "In front of you.", + "E": "It is behind you on the left, placed between the desk and the wall." + }, + "answer_label": "D", + "answer_text": "In front of you.", + "q_id": "egolife_A1_D6_015-mcq_fixed_3", + "correct_answer": "In front of you.", + "original_options": { + "A": "To your left.", + "B": "Behind you.", + "C": "It is slightly to your right, near the edge of the table.", + "D": "In front of you.", + "E": "It is behind you on the left, placed between the desk and the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015-mcq_fixed_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I sit in front of the laptop, where is the kettle relative to me?", + "answer": "In front of you to the left.", + "question_chinese": "当我坐在笔记本前时，水壶在我的什么方位？", + "answer_chinese": "在我的左前方。", + "evidence": [ + { + "description": "When I faced the balcony, I saw a watering can directly in front of me.", + "timestamps": { + "start": "00:05:46", + "end": "00:05:47" + } + }, + { + "description": "The kettle is to the right of the refrigerator, and when I face the refrigerator, it is directly in front of me.", + "timestamps": { + "start": "00:06:12", + "end": "00:06:13" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "In front of you to the left.", + "C": "Behind you to the left.", + "D": "It is in front of you to the right, not on your left side.", + "E": "It is slightly behind you to the right, near the edge of the desk." + }, + "answer_label": "B", + "answer_text": "In front of you to the left.", + "q_id": "egolife_A1_D6_015-mcq_fixed_4", + "correct_answer": "In front of you to the left.", + "original_options": { + "A": "In front of you.", + "B": "In front of you to the left.", + "C": "Behind you to the left.", + "D": "It is in front of you to the right, not on your left side.", + "E": "It is slightly behind you to the right, near the edge of the desk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_3_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the laptop in total?", + "answer": "You have used the laptop one time in total. The computer is in front of you.", + "question_chinese": "我一共使用了几次笔记本电脑？", + "answer_chinese": "你一共使用了一次笔记本电脑。电脑在你的前面。", + "evidence": [ + { + "description": "After tidying up the dolls, I went to a corner of the living room and used the laptop for the first time.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:40" + } + } + ], + "question_timestamps": "00:05:52", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero times.", + "B": "Two times.", + "C": "You have used the laptop one time in total. The computer is in front of you.", + "D": "You have used the laptop three times in total. The computer is in front of you.", + "E": "You have used the laptop four times in total, and the computer is in front of you." + }, + "answer_label": "C", + "answer_text": "You have used the laptop one time in total. The computer is in front of you.", + "q_id": "egolife_A1_D6_015_3_T2", + "correct_answer": "You have used the laptop one time in total. The computer is in front of you.", + "original_options": { + "A": "Zero times.", + "B": "Two times.", + "C": "You have used the laptop one time in total. The computer is in front of you.", + "D": "You have used the laptop three times in total. The computer is in front of you.", + "E": "You have used the laptop four times in total, and the computer is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_6_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the laptop in total?", + "answer": "You used the laptop two times in total. The computer is behind you.", + "question_chinese": "我一共使用了几次笔记本电脑？", + "answer_chinese": "你一共使用了两次笔记本电脑。电脑在你的后面。", + "evidence": [ + { + "description": "You have used the laptop twice.", + "timestamps": { + "start": "00:05:35", + "end": "00:06:12" + } + } + ], + "question_timestamps": "00:06:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You used the laptop two times in total. The computer is behind you.", + "B": "One time.", + "C": "Three times.", + "D": "You used the laptop zero times in total. The computer is behind you.", + "E": "You used the laptop four times in total, and the computer is behind you." + }, + "answer_label": "A", + "answer_text": "You used the laptop two times in total. The computer is behind you.", + "q_id": "egolife_A1_D6_015_6_T2", + "correct_answer": "You used the laptop two times in total. The computer is behind you.", + "original_options": { + "A": "You used the laptop two times in total. The computer is behind you.", + "B": "One time.", + "C": "Three times.", + "D": "You used the laptop zero times in total. The computer is behind you.", + "E": "You used the laptop four times in total, and the computer is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_3_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the laptop in total?", + "answer": "You used the laptop three times in total. The computer is behind you.", + "question_chinese": "我一共使用了几次笔记本电脑？", + "answer_chinese": "你一共使用了三次笔记本电脑。电脑在你的后面。", + "evidence": [ + { + "description": "You have used the laptop three times in total.", + "timestamps": { + "start": "00:05:38", + "end": "00:06:47" + } + } + ], + "question_timestamps": "00:06:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One time.", + "B": "You used the laptop three times in total. The computer is behind you.", + "C": "Two times.", + "D": "You used the laptop four times in total, and the computer is behind you.", + "E": "You used the laptop five times in total, with the computer still positioned behind you." + }, + "answer_label": "B", + "answer_text": "You used the laptop three times in total. The computer is behind you.", + "q_id": "egolife_A1_D6_015_3_T1", + "correct_answer": "You used the laptop three times in total. The computer is behind you.", + "original_options": { + "A": "One time.", + "B": "You used the laptop three times in total. The computer is behind you.", + "C": "Two times.", + "D": "You used the laptop four times in total, and the computer is behind you.", + "E": "You used the laptop five times in total, with the computer still positioned behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_6_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the laptop in total?", + "answer": "You used the laptop four times in total. The computer is in front of you.", + "question_chinese": "我一共使用了几次笔记本电脑？", + "answer_chinese": "你一共使用了四次笔记本电脑。电脑在你的前面。", + "evidence": [ + { + "description": "You have used the laptop four times in total.", + "timestamps": { + "start": "00:05:53", + "end": "00:08:07" + } + } + ], + "question_timestamps": "00:08:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Three times.", + "C": "You used the laptop five times in total, and the computer is in front of you.", + "D": "You used the laptop six times in total, with the computer still in front of you.", + "E": "You used the laptop four times in total. The computer is in front of you." + }, + "answer_label": "E", + "answer_text": "You used the laptop four times in total. The computer is in front of you.", + "q_id": "egolife_A1_D6_015_6_T1", + "correct_answer": "You used the laptop four times in total. The computer is in front of you.", + "original_options": { + "A": "Two times.", + "B": "Three times.", + "C": "You used the laptop five times in total, and the computer is in front of you.", + "D": "You used the laptop six times in total, with the computer still in front of you.", + "E": "You used the laptop four times in total. The computer is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_3_T3", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the laptop in total?", + "answer": "You have used the laptop five times in total. The computer is in front of you.", + "question_chinese": "我一共使用了几次笔记本电脑？", + "answer_chinese": "你一共使用了五次笔记本电脑。电脑在你的前面。", + "evidence": [ + { + "description": "After using a laptop for the first time, I switched to another laptop and used it intermittently three times.", + "timestamps": { + "start": "00:05:53", + "end": "00:09:43" + } + }, + { + "description": "In the end, I used the laptop I started with.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:19" + } + } + ], + "question_timestamps": "00:10:21", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three times.", + "B": "Four times.", + "C": "You have used the laptop six times in total, and the computer is in front of you.", + "D": "You have used the laptop five times in total. The computer is in front of you.", + "E": "You have used the laptop seven times in total, with the computer still in front of you." + }, + "answer_label": "D", + "answer_text": "You have used the laptop five times in total. The computer is in front of you.", + "q_id": "egolife_A1_D6_015_3_T3", + "correct_answer": "You have used the laptop five times in total. The computer is in front of you.", + "original_options": { + "A": "Three times.", + "B": "Four times.", + "C": "You have used the laptop six times in total, and the computer is in front of you.", + "D": "You have used the laptop five times in total. The computer is in front of you.", + "E": "You have used the laptop seven times in total, with the computer still in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_9", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times have I used the laptop in total?", + "answer": "Five times.", + "question_chinese": "我一共使用了几次笔记本电脑？", + "answer_chinese": "五次。", + "evidence": [ + { + "description": "After using a laptop for the first time, I switched to another laptop and used it intermittently three times.", + "timestamps": { + "start": "00:05:53", + "end": "00:09:43" + } + }, + { + "description": "In the end, I used the laptop I started with.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:19" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One time.", + "B": "Two times.", + "C": "I used the laptop three times in total, which is fewer than five.", + "D": "Five times.", + "E": "I used the laptop four times altogether, which is fewer than five." + }, + "answer_label": "D", + "answer_text": "Five times.", + "q_id": "egolife_A1_D6_015_9", + "correct_answer": "Five times.", + "original_options": { + "A": "One time.", + "B": "Two times.", + "C": "I used the laptop three times in total, which is fewer than five.", + "D": "Five times.", + "E": "I used the laptop four times altogether, which is fewer than five." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_7", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the pink toy?", + "answer": "From the girl's hand to mine, I held it and walked out of the bedroom, appearing in the living room.", + "question_chinese": "我看见粉色玩偶的运动路径是什么？", + "answer_chinese": "从女生手上到了我的手上，我拿着它走出了卧室，出现在客厅。", + "evidence": [ + { + "description": "The girl gave me a pink doll.", + "timestamps": { + "start": "00:03:21", + "end": "00:03:24" + } + } + ], + "question_timestamps": "00:03:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "From her hand to mine, hallway.", + "B": "From her hand to mine, bedroom.", + "C": "From the girl's hand to mine, I held it and walked out of the bedroom, appearing in the living room.", + "D": "From her hand to my friend’s, then we walked out of the bedroom into the living room.", + "E": "From the girl's hand to mine, I set it down in the bedroom, then walked into the living room." + }, + "answer_label": "C", + "answer_text": "From the girl's hand to mine, I held it and walked out of the bedroom, appearing in the living room.", + "q_id": "egolife_A1_D6_015_7", + "correct_answer": "From the girl's hand to mine, I held it and walked out of the bedroom, appearing in the living room.", + "original_options": { + "A": "From her hand to mine, hallway.", + "B": "From her hand to mine, bedroom.", + "C": "From the girl's hand to mine, I held it and walked out of the bedroom, appearing in the living room.", + "D": "From her hand to my friend’s, then we walked out of the bedroom into the living room.", + "E": "From the girl's hand to mine, I set it down in the bedroom, then walked into the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_8", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the pink doll?", + "answer": "I put it on the table and then picked it up from the table.", + "question_chinese": "我看见粉色玩偶的运动路径是什么？", + "answer_chinese": "我把它放在了桌子上，之后又从桌子上拿起.", + "evidence": [ + { + "description": "I walked out of the bedroom, put it on the table, and continued stringing.", + "timestamps": { + "start": "00:04:19", + "end": "00:05:24" + } + }, + { + "description": "After I finished using my laptop, I picked it up from the desk again.", + "timestamps": { + "start": "00:11:09", + "end": "00:11:10" + } + } + ], + "question_timestamps": "00:11:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I put it on the table and then picked it up from the table.", + "B": "Put it on the shelf.", + "C": "Placed it under the table.", + "D": "I left it on the table and walked away without picking it up again.", + "E": "I picked it up first and then placed it on the table afterward." + }, + "answer_label": "A", + "answer_text": "I put it on the table and then picked it up from the table.", + "q_id": "egolife_A1_D6_015_8", + "correct_answer": "I put it on the table and then picked it up from the table.", + "original_options": { + "A": "I put it on the table and then picked it up from the table.", + "B": "Put it on the shelf.", + "C": "Placed it under the table.", + "D": "I left it on the table and walked away without picking it up again.", + "E": "I picked it up first and then placed it on the table afterward." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hooks are there on the door now?", + "answer": "Two.", + "question_chinese": "现在门上共有几个粘钩？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "I stuck another one on the left side of the door frame.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:22" + } + }, + { + "description": "I see one on the right side of the door frame.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "There are four hooks on the door now, which is more than two.", + "D": "Two.", + "E": "There are five hooks on the door now, which is more than two." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "egolife_A1_D6_015_1", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "There are four hooks on the door now, which is more than two.", + "D": "Two.", + "E": "There are five hooks on the door now, which is more than two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_015", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_015_2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hooks are there on the door now?", + "answer": "Two.", + "question_chinese": "现在门上共有几个粘钩？", + "answer_chinese": "五个。", + "evidence": [ + { + "description": "I stuck three more evenly on the door frame.", + "timestamps": { + "start": "00:00:52", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:08:57", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "There are four hooks on the door now, which is more than two.", + "D": "There are five hooks on the door now, which is more than two.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "egolife_A1_D6_015_2", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "There are four hooks on the door now, which is more than two.", + "D": "There are five hooks on the door now, which is more than two.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D6_016-mcq.json b/QAs/egolife_A1_D6_016-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..dedca8dda9653a4de65cf150342eaba3bf18e458 --- /dev/null +++ b/QAs/egolife_A1_D6_016-mcq.json @@ -0,0 +1,939 @@ +[ + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016-mcq_fixed_8", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the living room door, where was the violin in relation to me?", + "answer": "Behind you.", + "question_chinese": "当我刚出客厅门时，小提琴在我的什么方位？", + "answer_chinese": "在我的后方", + "evidence": [ + { + "description": "When I walked towards the violin, it was directly in front of me", + "timestamps": { + "start": "00:10:19", + "end": "00:10:20" + } + }, + { + "description": "As I walked toward the door, the violin was to my right", + "timestamps": { + "start": "00:10:20", + "end": "00:10:22" + } + } + ], + "question_timestamps": "00:10:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "It was off to your right side as you stepped through the living room doorway.", + "E": "It was ahead of you in the hallway area as you exited, not trailing behind you." + }, + "answer_label": "A", + "answer_text": "Behind you.", + "q_id": "egolife_A1_D6_016-mcq_fixed_8", + "correct_answer": "Behind you.", + "original_options": { + "A": "Behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "It was off to your right side as you stepped through the living room doorway.", + "E": "It was ahead of you in the hallway area as you exited, not trailing behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016-mcq_fixed_9", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the living room door, where was the violin in relation to me?", + "answer": "To your right rear.", + "question_chinese": "当我刚出客厅门时，小提琴在我的什么方位？", + "answer_chinese": "在我的右后方", + "evidence": [ + { + "description": "When I walked towards the violin, it was directly in front of me", + "timestamps": { + "start": "00:10:19", + "end": "00:10:20" + } + }, + { + "description": "As I walked toward the door, the violin was to my right", + "timestamps": { + "start": "00:10:20", + "end": "00:10:22" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To your left rear.", + "B": "To your right front.", + "C": "It was directly behind you as you stepped out, not angled over your right shoulder.", + "D": "It sat toward your front-left side when you exited the door, rather than behind on your right.", + "E": "To your right rear." + }, + "answer_label": "E", + "answer_text": "To your right rear.", + "q_id": "egolife_A1_D6_016-mcq_fixed_9", + "correct_answer": "To your right rear.", + "original_options": { + "A": "To your left rear.", + "B": "To your right front.", + "C": "It was directly behind you as you stepped out, not angled over your right shoulder.", + "D": "It sat toward your front-left side when you exited the door, rather than behind on your right.", + "E": "To your right rear." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_5", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me when I pull the curtain, the doll or the alarm clock?", + "answer": "Alarm clock.", + "question_chinese": "当我按窗帘时，玩偶和闹钟哪一个离我比较近？", + "answer_chinese": "闹钟", + "evidence": [ + { + "description": "When I was to the left of the window, the alarm clock was directly in front of me.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:01" + } + }, + { + "description": "The doll is to my right, about 1 meter away from me.", + "timestamps": { + "start": "00:03:00", + "end": "00:03:02" + } + }, + { + "description": "The alarm clock and the doll are about 1 meter apart.", + "timestamps": { + "start": "00:03:01", + "end": "00:03:02" + } + } + ], + "question_timestamps": "00:03:20", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Doll.", + "B": "Alarm clock." + }, + "answer_label": "B", + "answer_text": "Alarm clock.", + "q_id": "egolife_A1_D6_016_5", + "correct_answer": "Alarm clock.", + "original_options": { + "A": "Doll.", + "B": "Alarm clock." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_6", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me when I pull the curtain, the doll or the alarm clock?", + "answer": "Doll.", + "question_chinese": "当我按窗帘时，玩偶和闹钟哪一个离我比较近？", + "answer_chinese": "玩偶", + "evidence": [ + { + "description": "When I was installing the curtains on the right, the doll was directly below me.", + "timestamps": { + "start": "00:05:22", + "end": "00:05:23" + } + }, + { + "description": "The alarm clock is to my left, about 1 meter away.", + "timestamps": { + "start": "00:05:22", + "end": "00:05:23" + } + }, + { + "description": "The alarm clock is about 1 meter away from the doll.", + "timestamps": { + "start": "00:05:22", + "end": "00:05:23" + } + } + ], + "question_timestamps": "00:05:34", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Alarm clock.", + "B": "Doll." + }, + "answer_label": "B", + "answer_text": "Doll.", + "q_id": "egolife_A1_D6_016_6", + "correct_answer": "Doll.", + "original_options": { + "A": "Alarm clock.", + "B": "Doll." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016-mcq_fixed_6", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I went to pick up the takeout?", + "answer": "YOU went out of the bedroom, went downstairs, went outside, exited through the gate, and walked straight down the path.", + "question_chinese": "我去取外卖时经过了什么路径？", + "answer_chinese": "我从卧室出去，下楼，出去室外，出了大门，进入小道直走", + "evidence": [ + { + "description": "I opened the living room door and walked outside, then went out through the front door.", + "timestamps": { + "start": "00:09:07", + "end": "00:09:14" + } + }, + { + "description": "When I was going downstairs, I received a call from the food delivery service.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:02" + } + } + ], + "question_timestamps": "00:09:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "YOU went out of the bedroom, went downstairs, went outside, exited through the gate, and walked straight down the path.", + "B": "You left the bedroom and went outside.", + "C": "You went downstairs, then turned back inside.", + "D": "YOU went out of the bedroom, went downstairs, exited through a front door, then turned left down the path.", + "E": "YOU went out of the bedroom, passed through the living room, went outside, and walked down the driveway instead of the path." + }, + "answer_label": "A", + "answer_text": "YOU went out of the bedroom, went downstairs, went outside, exited through the gate, and walked straight down the path.", + "q_id": "egolife_A1_D6_016-mcq_fixed_6", + "correct_answer": "YOU went out of the bedroom, went downstairs, went outside, exited through the gate, and walked straight down the path.", + "original_options": { + "A": "YOU went out of the bedroom, went downstairs, went outside, exited through the gate, and walked straight down the path.", + "B": "You left the bedroom and went outside.", + "C": "You went downstairs, then turned back inside.", + "D": "YOU went out of the bedroom, went downstairs, exited through a front door, then turned left down the path.", + "E": "YOU went out of the bedroom, passed through the living room, went outside, and walked down the driveway instead of the path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 22 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016-mcq_fixed_7", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route did I take when I went to pick up the takeout?", + "answer": "YOU went outside and walked straight down the path..", + "question_chinese": "我去取外卖时经过了什么路径？", + "answer_chinese": "我到室外，在小道直走", + "evidence": [ + { + "description": "When I was going downstairs, I received a call from the food delivery service.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:02" + } + }, + { + "description": "I opened the living room door and walked outside, then went out through the front door.", + "timestamps": { + "start": "00:09:07", + "end": "00:09:14" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "YOU went outside and turned left.", + "B": "YOU went outside and turned right.", + "C": "YOU stepped outside and walked along the driveway toward the street instead of going straight down the path.", + "D": "YOU went outside and walked straight down the path..", + "E": "YOU went outside, crossed the yard area, then headed to the street corner rather than continuing straight down the path." + }, + "answer_label": "D", + "answer_text": "YOU went outside and walked straight down the path..", + "q_id": "egolife_A1_D6_016-mcq_fixed_7", + "correct_answer": "YOU went outside and walked straight down the path..", + "original_options": { + "A": "YOU went outside and turned left.", + "B": "YOU went outside and turned right.", + "C": "YOU stepped outside and walked along the driveway toward the street instead of going straight down the path.", + "D": "YOU went outside and walked straight down the path..", + "E": "YOU went outside, crossed the yard area, then headed to the street corner rather than continuing straight down the path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_2_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many adhesive hooks are there on the window in total?", + "answer": "You saw one.", + "question_chinese": "我现在窗户上共有几个粘钩？", + "answer_chinese": "你看到一个。", + "evidence": [ + { + "description": "A girl gave me a sticky hook, and then I stuck it to the upper left corner of the window", + "timestamps": { + "start": "00:03:36", + "end": "00:03:37" + } + } + ], + "question_timestamps": "00:03:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw zero.", + "B": "You saw one.", + "C": "You saw two.", + "D": "You saw three adhesive hooks attached to the window at the same time.", + "E": "You saw four adhesive hooks on the window in total right now." + }, + "answer_label": "B", + "answer_text": "You saw one.", + "q_id": "egolife_A1_D6_016_2_T1", + "correct_answer": "You saw one.", + "original_options": { + "A": "You saw zero.", + "B": "You saw one.", + "C": "You saw two.", + "D": "You saw three adhesive hooks attached to the window at the same time.", + "E": "You saw four adhesive hooks on the window in total right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_2_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many adhesive hooks are there on the window in total?", + "answer": "One.", + "question_chinese": "我现在窗户上共有几个粘钩？", + "answer_chinese": "一个。", + "evidence": [ + { + "description": "A girl gave me a sticky hook, and then I stuck it to the upper left corner of the window", + "timestamps": { + "start": "00:03:36", + "end": "00:03:37" + } + } + ], + "question_timestamps": "00:03:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are four adhesive hooks on the window in total right now.", + "E": "There are five adhesive hooks stuck on the window at the moment." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "egolife_A1_D6_016_2_T2", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are four adhesive hooks on the window in total right now.", + "E": "There are five adhesive hooks stuck on the window at the moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_2_T3", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many adhesive hooks are there on the window in total?", + "answer": "Two.", + "question_chinese": "我现在窗户上共有几个粘钩？", + "answer_chinese": "俩个", + "evidence": [ + { + "description": "I stuck another adhesive hook in the upper right corner of the window.", + "timestamps": { + "start": "00:05:10", + "end": "00:05:12" + } + } + ], + "question_timestamps": "00:06:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "There are four adhesive hooks stuck on the window in total right now.", + "E": "You can count five adhesive hooks on the window at the moment, not two." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "egolife_A1_D6_016_2_T3", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "There are four adhesive hooks stuck on the window in total right now.", + "E": "You can count five adhesive hooks on the window at the moment, not two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_3_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many adhesive hooks are there on the window in total?", + "answer": "You saw two, and the window is to your left.", + "question_chinese": "我现在窗户上共有几个粘钩？", + "answer_chinese": "你看到俩个，窗户在你左边", + "evidence": [ + { + "description": "I stuck another adhesive hook in the upper right corner of the window.", + "timestamps": { + "start": "00:05:10", + "end": "00:05:12" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One hook.", + "B": "Three hooks.", + "C": "You saw two, and the window is to your left.", + "D": "You saw zero adhesive hooks, and the window is still positioned to your left.", + "E": "You saw four adhesive hooks, and the window remains on your left side." + }, + "answer_label": "C", + "answer_text": "You saw two, and the window is to your left.", + "q_id": "egolife_A1_D6_016_3_T1", + "correct_answer": "You saw two, and the window is to your left.", + "original_options": { + "A": "One hook.", + "B": "Three hooks.", + "C": "You saw two, and the window is to your left.", + "D": "You saw zero adhesive hooks, and the window is still positioned to your left.", + "E": "You saw four adhesive hooks, and the window remains on your left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_3_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many adhesive hooks are there on the window in total?", + "answer": "You saw two, and the window is in front of you.", + "question_chinese": "我现在窗户上共有几个粘钩？", + "answer_chinese": "你看到俩个，窗户在你前面", + "evidence": [ + { + "description": "I stuck another adhesive hook in the upper right corner of the window.", + "timestamps": { + "start": "00:05:10", + "end": "00:05:12" + } + }, + { + "description": "I have a laptop computer.", + "timestamps": { + "start": "00:03:24", + "end": "00:03:28" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One hook.", + "B": "Three hooks.", + "C": "You saw zero adhesive hooks, and the window is directly in front of you.", + "D": "You saw two, and the window is in front of you.", + "E": "You saw four adhesive hooks, and the window remains straight ahead in front of you." + }, + "answer_label": "D", + "answer_text": "You saw two, and the window is in front of you.", + "q_id": "egolife_A1_D6_016_3_T2", + "correct_answer": "You saw two, and the window is in front of you.", + "original_options": { + "A": "One hook.", + "B": "Three hooks.", + "C": "You saw zero adhesive hooks, and the window is directly in front of you.", + "D": "You saw two, and the window is in front of you.", + "E": "You saw four adhesive hooks, and the window remains straight ahead in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_1_T1", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What movement path did I see the pink doll take?", + "answer": "Starting from your hand, it was placed on the table.", + "question_chinese": "我看见粉色玩偶的运动路径是什么？", + "answer_chinese": "从我手中，到被放在了桌子上。", + "evidence": [ + { + "description": "A doll couldn't be strung up, so I put it on the table.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:52" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was hung on the door.", + "B": "It was set down on the floor.", + "C": "Starting from your hand, it was placed on the chair instead of being put on the table.", + "D": "Starting from your hand, it was moved to the window area and stuck there rather than placed on the table.", + "E": "Starting from your hand, it was placed on the table." + }, + "answer_label": "E", + "answer_text": "Starting from your hand, it was placed on the table.", + "q_id": "egolife_A1_D6_016_1_T1", + "correct_answer": "Starting from your hand, it was placed on the table.", + "original_options": { + "A": "It was hung on the door.", + "B": "It was set down on the floor.", + "C": "Starting from your hand, it was placed on the chair instead of being put on the table.", + "D": "Starting from your hand, it was moved to the window area and stuck there rather than placed on the table.", + "E": "Starting from your hand, it was placed on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_1", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the trajectory of the pink doll?", + "answer": "From my hands, one was placed on the table and the other was hung on the door.", + "question_chinese": "我看见粉色玩偶的运动路径是什么？", + "answer_chinese": "从我手中，到一个被放在了桌子上，另一个被挂在了门上。", + "evidence": [ + { + "description": "A doll couldn't be strung up, so I put it on the table.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:52" + } + }, + { + "description": "A doll was brought into my bedroom by me and then hung on the door.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:50" + } + } + ], + "question_timestamps": "00:01:55", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Floor and door.", + "B": "Table and wall.", + "C": "From my hands, one was placed on the chair and the other was hung on the door.", + "D": "From my hands, one was placed on the table and the other was hung on the door.", + "E": "From my hands, one was placed on the table and the other was hung on the window, not the door." + }, + "answer_label": "D", + "answer_text": "From my hands, one was placed on the table and the other was hung on the door.", + "q_id": "egolife_A1_D6_016_1", + "correct_answer": "From my hands, one was placed on the table and the other was hung on the door.", + "original_options": { + "A": "Floor and door.", + "B": "Table and wall.", + "C": "From my hands, one was placed on the chair and the other was hung on the door.", + "D": "From my hands, one was placed on the table and the other was hung on the door.", + "E": "From my hands, one was placed on the table and the other was hung on the window, not the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_1_T2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What movement path did I see the pink doll take?", + "answer": "Starting from your hand: one was placed on the table, and the other was hung on the door.", + "question_chinese": "我看见粉色玩偶的运动路径是什么？", + "answer_chinese": "从我手中，到一个被放在了桌子上，另一个被挂在了门上。", + "evidence": [ + { + "description": "A doll couldn't be strung up, so I put it on the table.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:52" + } + }, + { + "description": "A doll was brought into my bedroom by me and then hung on the door.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:50" + } + } + ], + "question_timestamps": "00:04:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Only hung on door.", + "B": "Starting from your hand: one was placed on the table, and the other was hung on the door.", + "C": "Only placed on floor.", + "D": "Starting from your hand: one was placed on the table, and the other was hung on the wall.", + "E": "Starting from your hand: one was set on the chair, and the other was hung on the door." + }, + "answer_label": "B", + "answer_text": "Starting from your hand: one was placed on the table, and the other was hung on the door.", + "q_id": "egolife_A1_D6_016_1_T2", + "correct_answer": "Starting from your hand: one was placed on the table, and the other was hung on the door.", + "original_options": { + "A": "Only hung on door.", + "B": "Starting from your hand: one was placed on the table, and the other was hung on the door.", + "C": "Only placed on floor.", + "D": "Starting from your hand: one was placed on the table, and the other was hung on the wall.", + "E": "Starting from your hand: one was set on the chair, and the other was hung on the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_3", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many adhesive hooks are there on the window now?", + "answer": "One.", + "question_chinese": "现在窗户上共有几个粘钩？", + "answer_chinese": "一个。", + "evidence": [ + { + "description": "A girl gave me a sticky hook, and then I stuck it to the upper left corner of the window", + "timestamps": { + "start": "00:03:36", + "end": "00:03:37" + } + } + ], + "question_timestamps": "00:03:48", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are four adhesive hooks on the window now when you count them all.", + "E": "You can see five adhesive hooks on the window at this moment, not just one." + }, + "answer_label": "C", + "answer_text": "One.", + "q_id": "egolife_A1_D6_016_3", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "One.", + "D": "There are four adhesive hooks on the window now when you count them all.", + "E": "You can see five adhesive hooks on the window at this moment, not just one." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_016", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_016_4", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hooks are there in total on the window now?", + "answer": "Two.", + "question_chinese": "现在窗户上共有几个粘钩？", + "answer_chinese": "俩个", + "evidence": [ + { + "description": "I stuck another adhesive hook in the upper right corner of the window.", + "timestamps": { + "start": "00:05:10", + "end": "00:05:12" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "There are four hooks in total on the window now when you count them.", + "E": "You can see five hooks on the window right now, which is more than two." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "egolife_A1_D6_016_4", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "There are four hooks in total on the window now when you count them.", + "E": "You can see five hooks on the window right now, which is more than two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D6_017-mcq.json b/QAs/egolife_A1_D6_017-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e7893990a0efc011cb54037af5b1a399d2ca6a22 --- /dev/null +++ b/QAs/egolife_A1_D6_017-mcq.json @@ -0,0 +1,807 @@ +[ + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017-mcq_fixed_2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the mobile path of a power bank?", + "answer": "It went from the table to your hand and then back to the table.", + "question_chinese": "充电宝的移动路径是什么？", + "answer_chinese": "从桌子上到了我的手上然后又放回了桌子上", + "evidence": [ + { + "description": "I see the power bank charging on the table", + "timestamps": { + "start": "00:00:33", + "end": "00:00:34" + } + }, + { + "description": "I picked up the power bank to check its charging status.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:35" + } + }, + { + "description": "I put down the power bank after I finished watching it.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "From the table to your pocket.", + "B": "From your hand to the shelf.", + "C": "It went from the table to your hand and then back to the table.", + "D": "It moved from the table into your pocket, then you set it down on the table again.", + "E": "You picked it up from your hand and placed it on a nearby shelf before returning it to your hand." + }, + "answer_label": "C", + "answer_text": "It went from the table to your hand and then back to the table.", + "q_id": "egolife_A1_D6_017-mcq_fixed_2", + "correct_answer": "It went from the table to your hand and then back to the table.", + "original_options": { + "A": "From the table to your pocket.", + "B": "From your hand to the shelf.", + "C": "It went from the table to your hand and then back to the table.", + "D": "It moved from the table into your pocket, then you set it down on the table again.", + "E": "You picked it up from your hand and placed it on a nearby shelf before returning it to your hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017-mcq_fixed_3", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the mobile path of a power bank?", + "answer": "First on your hand and then back on the table.", + "question_chinese": "充电宝的移动路径是什么？", + "answer_chinese": "先在我的手上然后又放回了桌子上", + "evidence": [ + { + "description": "I picked up the power bank to check its charging status.", + "timestamps": { + "start": "00:00:34", + "end": "00:00:35" + } + }, + { + "description": "I put down the power bank after I finished watching it.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:36" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "From your pocket to the table.", + "B": "From the table to your hand.", + "C": "It started on the table, you picked it up into your hand, and then you put it in your pocket.", + "D": "First on your hand and then back on the table.", + "E": "You took it from your hand, set it on a shelf briefly, and then left it there instead of the table." + }, + "answer_label": "D", + "answer_text": "First on your hand and then back on the table.", + "q_id": "egolife_A1_D6_017-mcq_fixed_3", + "correct_answer": "First on your hand and then back on the table.", + "original_options": { + "A": "From your pocket to the table.", + "B": "From the table to your hand.", + "C": "It started on the table, you picked it up into your hand, and then you put it in your pocket.", + "D": "First on your hand and then back on the table.", + "E": "You took it from your hand, set it on a shelf briefly, and then left it there instead of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017_3_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many software programs did I open on the computer in total?", + "answer": "You opened two.", + "question_chinese": "我一共打开了几个软件在电脑上？", + "answer_chinese": "你打开俩个", + "evidence": [ + { + "description": "I saw the notebook's meeting editing software", + "timestamps": { + "start": "00:00:57", + "end": "00:02:38" + } + }, + { + "description": "After that, I opened the WeChat app", + "timestamps": { + "start": "00:02:44", + "end": "00:03:36" + } + } + ], + "question_timestamps": "00:04:44", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You opened zero.", + "B": "You opened one.", + "C": "You opened three programs in total while using the computer during the session.", + "D": "You opened two.", + "E": "You opened four different programs overall, switching between them as you worked." + }, + "answer_label": "D", + "answer_text": "You opened two.", + "q_id": "egolife_A1_D6_017_3_T1", + "correct_answer": "You opened two.", + "original_options": { + "A": "You opened zero.", + "B": "You opened one.", + "C": "You opened three programs in total while using the computer during the session.", + "D": "You opened two.", + "E": "You opened four different programs overall, switching between them as you worked." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017_3_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many software programs did I open on the computer in total?", + "answer": "You opened four.", + "question_chinese": "我一共打开了几个软件在电脑上？", + "answer_chinese": "你打开四个", + "evidence": [ + { + "description": "I saw the notebook's meeting editing software", + "timestamps": { + "start": "00:00:57", + "end": "00:02:38" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You opened two.", + "B": "You opened three.", + "C": "You opened five programs in total on the computer during the session.", + "D": "You opened six different programs overall, moving through several windows as you worked.", + "E": "You opened four." + }, + "answer_label": "E", + "answer_text": "You opened four.", + "q_id": "egolife_A1_D6_017_3_T2", + "correct_answer": "You opened four.", + "original_options": { + "A": "You opened two.", + "B": "You opened three.", + "C": "You opened five programs in total on the computer during the session.", + "D": "You opened six different programs overall, moving through several windows as you worked.", + "E": "You opened four." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017_4_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many software programs did I open on the computer in total?", + "answer": "Four.", + "question_chinese": "我一共打开了几个软件在电脑上？", + "answer_chinese": "四个", + "evidence": [ + { + "description": "Then I opened the search software", + "timestamps": { + "start": "00:05:09", + "end": "00:08:38" + } + }, + { + "description": "Then I opened the QQ app again", + "timestamps": { + "start": "00:08:49", + "end": "00:09:50" + } + } + ], + "question_timestamps": "00:12:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Two.", + "C": "Three.", + "D": "You opened five programs in total on the computer during the session.", + "E": "You opened six different programs overall, moving between multiple applications as you worked." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "egolife_A1_D6_017_4_T1", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Two.", + "C": "Three.", + "D": "You opened five programs in total on the computer during the session.", + "E": "You opened six different programs overall, moving between multiple applications as you worked." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017-mcq_fixed_0", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "The laptop is behind you.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "笔记本电脑在我的后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The laptop is behind you.", + "B": "The laptop is in front of you.", + "C": "The laptop is to your left.", + "D": "The laptop is slightly to your right as you step out, positioned near the doorway and visible immediately.", + "E": "As you exited the bedroom, the laptop sat ahead of you and a bit to the left, within sight." + }, + "answer_label": "A", + "answer_text": "The laptop is behind you.", + "q_id": "egolife_A1_D6_017-mcq_fixed_0", + "correct_answer": "The laptop is behind you.", + "original_options": { + "A": "The laptop is behind you.", + "B": "The laptop is in front of you.", + "C": "The laptop is to your left.", + "D": "The laptop is slightly to your right as you step out, positioned near the doorway and visible immediately.", + "E": "As you exited the bedroom, the laptop sat ahead of you and a bit to the left, within sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017-mcq_fixed_1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "我刚出卧室门时，笔记本电脑在我的左后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:04:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "B": "Behind you to the right.", + "C": "In front of you to the right.", + "D": "As you stepped out, the laptop was straight ahead near the wall, slightly to your right of center.", + "E": "The laptop sat in front of you and angled a little left beyond the doorway, not behind you at all." + }, + "answer_label": "A", + "answer_text": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "q_id": "egolife_A1_D6_017-mcq_fixed_1", + "correct_answer": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "original_options": { + "A": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "B": "Behind you to the right.", + "C": "In front of you to the right.", + "D": "As you stepped out, the laptop was straight ahead near the wall, slightly to your right of center.", + "E": "The laptop sat in front of you and angled a little left beyond the doorway, not behind you at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017_1_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "About how many different kinds of items are in my bedroom?", + "answer": "You have five categories.", + "question_chinese": "我的卧室大概有几种物品？", + "answer_chinese": "你有五类", + "evidence": [ + { + "description": "I saw plastic bags, a notebook, a bag, a plastic board, and a charger on the bed.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You have three categories.", + "B": "You have five categories.", + "C": "You have four categories.", + "D": "You have six categories of items in the bedroom when you group similar objects together.", + "E": "You have seven categories overall, counting furniture, electronics, clothing, and other distinct item types." + }, + "answer_label": "B", + "answer_text": "You have five categories.", + "q_id": "egolife_A1_D6_017_1_T1", + "correct_answer": "You have five categories.", + "original_options": { + "A": "You have three categories.", + "B": "You have five categories.", + "C": "You have four categories.", + "D": "You have six categories of items in the bedroom when you group similar objects together.", + "E": "You have seven categories overall, counting furniture, electronics, clothing, and other distinct item types." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017_1_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "About how many different kinds of items are in my bedroom?", + "answer": "You had seven categories.", + "question_chinese": "我的卧室大概有几种物品？", + "answer_chinese": "你有七类", + "evidence": [ + { + "description": "I saw plastic bags, a notebook, a bag, plastic sheets, a charger, a mobile phone, and a storage box on the bed.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:11:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You had six categories.", + "B": "You had eight categories.", + "C": "You had seven categories.", + "D": "You had about five different categories of items overall in the bedroom.", + "E": "You had roughly nine categories, counting each distinct type of item you saw in the room." + }, + "answer_label": "C", + "answer_text": "You had seven categories.", + "q_id": "egolife_A1_D6_017_1_T2", + "correct_answer": "You had seven categories.", + "original_options": { + "A": "You had six categories.", + "B": "You had eight categories.", + "C": "You had seven categories.", + "D": "You had about five different categories of items overall in the bedroom.", + "E": "You had roughly nine categories, counting each distinct type of item you saw in the room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017_2_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "About how many different kinds of items are in my bedroom?", + "answer": "Your bedroom had about seven categories.", + "question_chinese": "我的卧室大概有几种物品？", + "answer_chinese": "你卧室大概有七类", + "evidence": [ + { + "description": "I saw plastic bags, a notebook, a bag, plastic sheets, a charger, a mobile phone, and a storage box on the bed.", + "timestamps": { + "start": "00:01:19", + "end": "00:01:20" + } + } + ], + "question_timestamps": "00:11:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your bedroom had about six categories.", + "B": "Your bedroom had about eight categories.", + "C": "Your bedroom had roughly five categories of items based on the different types you noticed.", + "D": "Your bedroom had about seven categories.", + "E": "Your bedroom had about nine categories when you grouped the items by kind." + }, + "answer_label": "D", + "answer_text": "Your bedroom had about seven categories.", + "q_id": "egolife_A1_D6_017_2_T2", + "correct_answer": "Your bedroom had about seven categories.", + "original_options": { + "A": "Your bedroom had about six categories.", + "B": "Your bedroom had about eight categories.", + "C": "Your bedroom had roughly five categories of items based on the different types you noticed.", + "D": "Your bedroom had about seven categories.", + "E": "Your bedroom had about nine categories when you grouped the items by kind." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017_5_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many laptop computers did I see?", + "answer": "Eight.", + "question_chinese": "我看见了几台笔记本电脑？", + "answer_chinese": "八个。", + "evidence": [ + { + "description": "I saw six on the table shelf and two on the small table.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Seven.", + "B": "Nine.", + "C": "Eight.", + "D": "You saw seven laptop computers in total across the room during the walkthrough.", + "E": "You saw nine laptop computers altogether during the video while scanning the desk and bed." + }, + "answer_label": "C", + "answer_text": "Eight.", + "q_id": "egolife_A1_D6_017_5_T1", + "correct_answer": "Eight.", + "original_options": { + "A": "Seven.", + "B": "Nine.", + "C": "Eight.", + "D": "You saw seven laptop computers in total across the room during the walkthrough.", + "E": "You saw nine laptop computers altogether during the video while scanning the desk and bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017_5_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many laptop computers did I see?", + "answer": "Eight.", + "question_chinese": "我看见了几台笔记本电脑？", + "answer_chinese": "八个。", + "evidence": [ + { + "description": "I saw six on the table shelf and two on the small table.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Six.", + "B": "Ten.", + "C": "Eight.", + "D": "You saw six laptop computers over the whole sequence while moving through the bedroom.", + "E": "You saw ten laptop computers in total during your walkthrough, including ones on the bed." + }, + "answer_label": "C", + "answer_text": "Eight.", + "q_id": "egolife_A1_D6_017_5_T2", + "correct_answer": "Eight.", + "original_options": { + "A": "Six.", + "B": "Ten.", + "C": "Eight.", + "D": "You saw six laptop computers over the whole sequence while moving through the bedroom.", + "E": "You saw ten laptop computers in total during your walkthrough, including ones on the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017_5_T3", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many laptop computers did I see?", + "answer": "Nine.", + "question_chinese": "我看见了几台笔记本电脑？", + "answer_chinese": "九台", + "evidence": [ + { + "description": "I saw six on the table shelf and two on the small table.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "I saw there was still a machine on the bed", + "timestamps": { + "start": "00:00:54", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:10:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Eight.", + "B": "Ten.", + "C": "You saw eight laptop computers in total across the scene while looking around the room.", + "D": "Nine.", + "E": "You saw ten laptop computers altogether during the video, including those set on the desk." + }, + "answer_label": "D", + "answer_text": "Nine.", + "q_id": "egolife_A1_D6_017_5_T3", + "correct_answer": "Nine.", + "original_options": { + "A": "Eight.", + "B": "Ten.", + "C": "You saw eight laptop computers in total across the scene while looking around the room.", + "D": "Nine.", + "E": "You saw ten laptop computers altogether during the video, including those set on the desk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_017", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_017_5_T4", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many laptop computers did I see?", + "answer": "Nine.", + "question_chinese": "我看见了几台笔记本电脑？", + "answer_chinese": "九台", + "evidence": [ + { + "description": "I saw six on the table shelf and two on the small table.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "I saw there was still a machine on the bed", + "timestamps": { + "start": "00:00:54", + "end": "00:00:55" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Seven.", + "B": "Nine.", + "C": "Eleven.", + "D": "You saw seven laptop computers across the scene while turning and checking each surface.", + "E": "You saw eleven laptop computers in total during the recording, counting every one you passed." + }, + "answer_label": "B", + "answer_text": "Nine.", + "q_id": "egolife_A1_D6_017_5_T4", + "correct_answer": "Nine.", + "original_options": { + "A": "Seven.", + "B": "Nine.", + "C": "Eleven.", + "D": "You saw seven laptop computers across the scene while turning and checking each surface.", + "E": "You saw eleven laptop computers in total during the recording, counting every one you passed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D6_018-mcq.json b/QAs/egolife_A1_D6_018-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8fee9ed6e295a3053d0dde308785e0278a1d04c6 --- /dev/null +++ b/QAs/egolife_A1_D6_018-mcq.json @@ -0,0 +1,831 @@ +[ + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018_6_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the power bank relative to me?", + "answer": "To your right.", + "question_chinese": "充电宝在我什么方位？", + "answer_chinese": "在我右面。", + "evidence": [ + { + "description": "After I got up from bed, the power bank was in front of me.", + "timestamps": { + "start": "00:07:19", + "end": "00:07:20" + } + }, + { + "description": "When I turn 90 degrees.", + "timestamps": { + "start": "00:07:31", + "end": "00:07:32" + } + } + ], + "question_timestamps": "00:07:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "In front of you.", + "C": "Slightly behind you on your left side, near the pillow at the head of the bed.", + "D": "To your right.", + "E": "Just in front of you beside the foot of the bed, within your reach." + }, + "answer_label": "D", + "answer_text": "To your right.", + "q_id": "egolife_A1_D6_018_6_T1", + "correct_answer": "To your right.", + "original_options": { + "A": "To your left.", + "B": "In front of you.", + "C": "Slightly behind you on your left side, near the pillow at the head of the bed.", + "D": "To your right.", + "E": "Just in front of you beside the foot of the bed, within your reach." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018_6_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the power bank relative to me?", + "answer": "To your right.", + "question_chinese": "充电宝在我什么方位？", + "answer_chinese": "在我前面。", + "evidence": [ + { + "description": "After I got up from bed, the power bank was in front of me.", + "timestamps": { + "start": "00:07:19", + "end": "00:07:20" + } + }, + { + "description": "When I turn 90 degrees.", + "timestamps": { + "start": "00:07:32", + "end": "00:07:33" + } + } + ], + "question_timestamps": "00:11:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "Behind you.", + "C": "Near your left foot, tucked between the blanket and the bed edge.", + "D": "To your right.", + "E": "In front of you on the desk, with its cable trailing toward the bed." + }, + "answer_label": "D", + "answer_text": "To your right.", + "q_id": "egolife_A1_D6_018_6_T2", + "correct_answer": "To your right.", + "original_options": { + "A": "To your left.", + "B": "Behind you.", + "C": "Near your left foot, tucked between the blanket and the bed edge.", + "D": "To your right.", + "E": "In front of you on the desk, with its cable trailing toward the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018_7_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction did I see the backpack from me?", + "answer": "In front of you to the right.", + "question_chinese": "我看见书包在我的什么方位？", + "answer_chinese": "在我的右前方。", + "evidence": [ + { + "description": "I saw a gray bag in front of me to the right", + "timestamps": { + "start": "00:03:33", + "end": "00:03:34" + } + } + ], + "question_timestamps": "00:03:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you to the left.", + "B": "In front of you to the right.", + "C": "Directly behind you.", + "D": "Off to your left in front, not on the right side where you were looking.", + "E": "Behind you and slightly left, as if the backpack was near the wall behind your shoulder." + }, + "answer_label": "B", + "answer_text": "In front of you to the right.", + "q_id": "egolife_A1_D6_018_7_T1", + "correct_answer": "In front of you to the right.", + "original_options": { + "A": "In front of you to the left.", + "B": "In front of you to the right.", + "C": "Directly behind you.", + "D": "Off to your left in front, not on the right side where you were looking.", + "E": "Behind you and slightly left, as if the backpack was near the wall behind your shoulder." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 3, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018_7_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction did I see the backpack from me?", + "answer": "In front of you to the right.", + "question_chinese": "我看见书包在我的什么方位？", + "answer_chinese": "在我的后方。", + "evidence": [ + { + "description": "I saw a gray bag in front of me to the right", + "timestamps": { + "start": "00:07:48", + "end": "00:07:49" + } + }, + { + "description": "I turn 180 degrees.", + "timestamps": { + "start": "00:07:48", + "end": "00:07:49" + } + } + ], + "question_timestamps": "00:07:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you to the left.", + "B": "Directly to your left.", + "C": "In front of you to the right.", + "D": "Slightly behind you on the left side, rather than ahead of you on the right.", + "E": "In front but centered, not angled to your right the way you remembered seeing it." + }, + "answer_label": "C", + "answer_text": "In front of you to the right.", + "q_id": "egolife_A1_D6_018_7_T2", + "correct_answer": "In front of you to the right.", + "original_options": { + "A": "In front of you to the left.", + "B": "Directly to your left.", + "C": "In front of you to the right.", + "D": "Slightly behind you on the left side, rather than ahead of you on the right.", + "E": "In front but centered, not angled to your right the way you remembered seeing it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 4, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018_4_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How can I find my glasses?", + "answer": "YOU saw a person pick up their eyes from the table and then put them on the bed.", + "question_chinese": "我该如何找到我的眼镜?", + "answer_chinese": "我看见一个人讲眼睛从桌子上拿起，然后又有放在了床上", + "evidence": [ + { + "description": "I saw a person pick up glasses from the table.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:01" + } + }, + { + "description": "Then he put the glasses on the bed.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:06:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "YOU saw a person pick up their eyes from the table and then put them on the bed.", + "B": "They moved them to the floor.", + "C": "They put them on a chair.", + "D": "YOU saw a person pick up their eyes from the bed and then set them down on the table.", + "E": "YOU watched someone take their eyes off the table and place them onto a nearby chair instead." + }, + "answer_label": "A", + "answer_text": "YOU saw a person pick up their eyes from the table and then put them on the bed.", + "q_id": "egolife_A1_D6_018_4_T1", + "correct_answer": "YOU saw a person pick up their eyes from the table and then put them on the bed.", + "original_options": { + "A": "YOU saw a person pick up their eyes from the table and then put them on the bed.", + "B": "They moved them to the floor.", + "C": "They put them on a chair.", + "D": "YOU saw a person pick up their eyes from the bed and then set them down on the table.", + "E": "YOU watched someone take their eyes off the table and place them onto a nearby chair instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018_1_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How can I find my glasses?", + "answer": "You saw a person pick up their glasses from the table and then put them on the bed.", + "question_chinese": "我该如何找到我的眼镜?", + "answer_chinese": "我看见一个人将眼镜从桌子上拿起，然后又放在了床上", + "evidence": [ + { + "description": "I saw a person pick up glasses from the table.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:01" + } + }, + { + "description": "Then he put the glasses on the bed.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:06:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They left them on the table.", + "B": "They put them on the couch.", + "C": "You saw a person pick up their glasses from the bed and then put them back on the table.", + "D": "You saw a person pick up their glasses from the table and then put them on the bed.", + "E": "You watched someone take the glasses from the table and set them down on a nearby chair instead." + }, + "answer_label": "D", + "answer_text": "You saw a person pick up their glasses from the table and then put them on the bed.", + "q_id": "egolife_A1_D6_018_1_T1", + "correct_answer": "You saw a person pick up their glasses from the table and then put them on the bed.", + "original_options": { + "A": "They left them on the table.", + "B": "They put them on the couch.", + "C": "You saw a person pick up their glasses from the bed and then put them back on the table.", + "D": "You saw a person pick up their glasses from the table and then put them on the bed.", + "E": "You watched someone take the glasses from the table and set them down on a nearby chair instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018_1_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How should I find my glasses?", + "answer": "You saw a person pick up their glasses from the table and then put them on the bed.", + "question_chinese": "我该如何找到我的眼镜?", + "answer_chinese": "我看见一个人将眼镜从桌子上拿起，然后又放在了床上", + "evidence": [ + { + "description": "I saw a person pick up glasses from the table.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:01" + } + }, + { + "description": "Then he put the glasses on the bed.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:06:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They moved them to the desk.", + "B": "They set them on the counter.", + "C": "You saw a person pick up their glasses from the bed and then put them down on the table.", + "D": "You saw a person pick up their glasses from the table and then put them on the bed.", + "E": "You noticed someone lift the glasses from the table and place them on the couch rather than the bed." + }, + "answer_label": "D", + "answer_text": "You saw a person pick up their glasses from the table and then put them on the bed.", + "q_id": "egolife_A1_D6_018_1_T2", + "correct_answer": "You saw a person pick up their glasses from the table and then put them on the bed.", + "original_options": { + "A": "They moved them to the desk.", + "B": "They set them on the counter.", + "C": "You saw a person pick up their glasses from the bed and then put them down on the table.", + "D": "You saw a person pick up their glasses from the table and then put them on the bed.", + "E": "You noticed someone lift the glasses from the table and place them on the couch rather than the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018_4_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How can I find my glasses?", + "answer": "YOU saw a person pick up their eyes from the table and then put them on the bed.", + "question_chinese": "我该如何找到我的眼镜?", + "answer_chinese": "我看见一个人讲眼睛从桌子上拿起，然后又有放在了床上", + "evidence": [ + { + "description": "I saw a person pick up glasses from the table", + "timestamps": { + "start": "00:06:04", + "end": "00:06:06" + } + }, + { + "description": "Then he put the glasses on the bed.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:08:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They put them on the shelf.", + "B": "They left them on the table.", + "C": "YOU saw a person pick up their eyes from the table and then put them down on the floor.", + "D": "YOU saw a person pick up their eyes from the table and then put them on the bed.", + "E": "YOU watched someone take their eyes from the bed and place them back on the table afterward." + }, + "answer_label": "D", + "answer_text": "YOU saw a person pick up their eyes from the table and then put them on the bed.", + "q_id": "egolife_A1_D6_018_4_T2", + "correct_answer": "YOU saw a person pick up their eyes from the table and then put them on the bed.", + "original_options": { + "A": "They put them on the shelf.", + "B": "They left them on the table.", + "C": "YOU saw a person pick up their eyes from the table and then put them down on the floor.", + "D": "YOU saw a person pick up their eyes from the table and then put them on the bed.", + "E": "YOU watched someone take their eyes from the bed and place them back on the table afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018_2_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How can I find my glasses?", + "answer": "Glasses from the table and then put them on the bed.", + "question_chinese": "我该如何找到我的眼镜?", + "answer_chinese": "我看见一个人将眼镜从桌子上拿起，然后又放在了床上", + "evidence": [ + { + "description": "I saw a person pick up glasses from the table", + "timestamps": { + "start": "00:06:04", + "end": "00:06:06" + } + }, + { + "description": "Then he put the glasses on the bed.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:08:23", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Glasses from the bed to the table.", + "B": "Glasses from the floor to the couch.", + "C": "Glasses from the table and then put them on the bed.", + "D": "Glasses from the chair and then put them on the shelf, not taken from the table to the bed.", + "E": "Glasses from the counter and then put them on the dresser, instead of ending up on the bed." + }, + "answer_label": "C", + "answer_text": "Glasses from the table and then put them on the bed.", + "q_id": "egolife_A1_D6_018_2_T1", + "correct_answer": "Glasses from the table and then put them on the bed.", + "original_options": { + "A": "Glasses from the bed to the table.", + "B": "Glasses from the floor to the couch.", + "C": "Glasses from the table and then put them on the bed.", + "D": "Glasses from the chair and then put them on the shelf, not taken from the table to the bed.", + "E": "Glasses from the counter and then put them on the dresser, instead of ending up on the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018_2_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "How can I find my glasses?", + "answer": "You saw a person pick up their glasses from the table and then put them on the bed.", + "question_chinese": "我该如何找到我的眼镜?", + "answer_chinese": "我看见一个人将眼镜从桌子上拿起，然后又放在了床上", + "evidence": [ + { + "description": "I saw a person pick up glasses from the table", + "timestamps": { + "start": "00:06:04", + "end": "00:06:06" + } + }, + { + "description": "Then he put the glasses on the bed.", + "timestamps": { + "start": "00:06:06", + "end": "00:06:07" + } + } + ], + "question_timestamps": "00:08:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "They moved them from the bed.", + "B": "They placed them on your legs.", + "C": "You saw a person pick up their glasses from the bed and then put them back on the table afterward.", + "D": "It went from your legs onto the bed and then stayed on your legs, not moved from the table.", + "E": "You saw a person pick up their glasses from the table and then put them on the bed." + }, + "answer_label": "E", + "answer_text": "You saw a person pick up their glasses from the table and then put them on the bed.", + "q_id": "egolife_A1_D6_018_2_T2", + "correct_answer": "You saw a person pick up their glasses from the table and then put them on the bed.", + "original_options": { + "A": "They moved them from the bed.", + "B": "They placed them on your legs.", + "C": "You saw a person pick up their glasses from the bed and then put them back on the table afterward.", + "D": "It went from your legs onto the bed and then stayed on your legs, not moved from the table.", + "E": "You saw a person pick up their glasses from the table and then put them on the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018-mcq_fixed_0", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Now, am I closer to the power bank or the drink?", + "answer": "You are close to the power bank.", + "question_chinese": "现在我距充电宝近还是饮料近？", + "answer_chinese": "我离充电宝近。", + "evidence": [ + { + "description": "The power bank is in front of me, about 20 centimeters away.", + "timestamps": { + "start": "00:08:15", + "end": "00:08:16" + } + }, + { + "description": "The power bank is about 1 meter away from the cola.", + "timestamps": { + "start": "00:08:15", + "end": "00:08:16" + } + }, + { + "description": "The cola is in front of me to the left, about 150 centimeters away from me.", + "timestamps": { + "start": "00:08:15", + "end": "00:08:16" + } + } + ], + "question_timestamps": "00:08:17", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "You are close to the power bank.", + "B": "You are close to the drink." + }, + "answer_label": "A", + "answer_text": "You are close to the power bank.", + "q_id": "egolife_A1_D6_018-mcq_fixed_0", + "correct_answer": "You are close to the power bank.", + "original_options": { + "A": "You are close to the power bank.", + "B": "You are close to the drink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018-mcq_fixed_1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Now, am I closer to the power bank or the drink?", + "answer": "You are close to the drink.", + "question_chinese": "现在我距充电宝近还是饮料近？", + "answer_chinese": "我离饮料近。", + "evidence": [ + { + "description": "The the drink is in front of me, about 120 centimeters away.", + "timestamps": { + "start": "00:08:15", + "end": "00:08:16" + } + }, + { + "description": "The the drink is about 1 meter away from the cola.", + "timestamps": { + "start": "00:08:15", + "end": "00:08:16" + } + }, + { + "description": "The cola is in front of me to the left, about 50 centimeters away from me.", + "timestamps": { + "start": "00:08:15", + "end": "00:08:16" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "You are close to the power bank.", + "B": "You are close to the drink." + }, + "answer_label": "B", + "answer_text": "You are close to the drink.", + "q_id": "egolife_A1_D6_018-mcq_fixed_1", + "correct_answer": "You are close to the drink.", + "original_options": { + "A": "You are close to the power bank.", + "B": "You are close to the drink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018-mcq_fixed_4", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path for the laptop?", + "answer": "It was placed from your legs onto the bed, then put back onto your legs.", + "question_chinese": "笔记本电脑的路径？", + "answer_chinese": "从我的腿上被我放到了床上，然后又放回了腿上。", + "evidence": [ + { + "description": "I put my laptop on the bed first.", + "timestamps": { + "start": "00:08:02", + "end": "00:08:04" + } + }, + { + "description": "After chatting, I put the notebook back on my lap.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:36" + } + } + ], + "question_timestamps": "00:08:38", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was placed from your legs onto the bed, then put back onto your legs.", + "B": "From legs to the floor.", + "C": "From bed to your legs.", + "D": "You moved it off your legs onto the bed, then left it there instead of returning it.", + "E": "You lifted it from your legs, set it on the bed briefly, then shifted it to the side of the bed." + }, + "answer_label": "A", + "answer_text": "It was placed from your legs onto the bed, then put back onto your legs.", + "q_id": "egolife_A1_D6_018-mcq_fixed_4", + "correct_answer": "It was placed from your legs onto the bed, then put back onto your legs.", + "original_options": { + "A": "It was placed from your legs onto the bed, then put back onto your legs.", + "B": "From legs to the floor.", + "C": "From bed to your legs.", + "D": "You moved it off your legs onto the bed, then left it there instead of returning it.", + "E": "You lifted it from your legs, set it on the bed briefly, then shifted it to the side of the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_018", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_018-mcq_fixed_5", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path for the laptop?", + "answer": "First on the bed and then put back on the legs.", + "question_chinese": "笔记本电脑的路径？", + "answer_chinese": "先在床上然后又放回了腿上。", + "evidence": [ + { + "description": "After chatting, I put the notebook back on my lap.", + "timestamps": { + "start": "00:08:35", + "end": "00:08:36" + } + }, + { + "description": "I put my laptop on the bed first.", + "timestamps": { + "start": "00:08:04", + "end": "00:08:05" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "First on the legs, then on the bed.", + "B": "First on the chair, then on the legs.", + "C": "It started on your legs and was placed onto the bed, but it was not returned to your legs.", + "D": "First on the bed and then put back on the legs.", + "E": "You put it on the bed briefly, then moved it onto a nearby table rather than back onto your legs." + }, + "answer_label": "D", + "answer_text": "First on the bed and then put back on the legs.", + "q_id": "egolife_A1_D6_018-mcq_fixed_5", + "correct_answer": "First on the bed and then put back on the legs.", + "original_options": { + "A": "First on the legs, then on the bed.", + "B": "First on the chair, then on the legs.", + "C": "It started on your legs and was placed onto the bed, but it was not returned to your legs.", + "D": "First on the bed and then put back on the legs.", + "E": "You put it on the bed briefly, then moved it onto a nearby table rather than back onto your legs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 19, + 20 + ] + } +] diff --git a/QAs/egolife_A1_D6_019-mcq.json b/QAs/egolife_A1_D6_019-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..0677414e491225c8cc2b663c10966653b446e3e5 --- /dev/null +++ b/QAs/egolife_A1_D6_019-mcq.json @@ -0,0 +1,589 @@ +[ + { + "video_uid": "egolife_A1_D6_019", + "video_length": "00:11:40", + "qid": "egolife_A1_D6_019-mcq_fixed_4", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I walk towards the whiteboard, where is the bucket in relation to me?", + "answer": "To your left rear.", + "question_chinese": "当我走向白板时，水桶在我的什么方位？", + "answer_chinese": "在我的左后方", + "evidence": [ + { + "description": "After that, I stood up", + "timestamps": { + "start": "00:05:22", + "end": "00:05:23" + } + }, + { + "description": "When I was using the computer, I saw a bucket directly in front of me", + "timestamps": { + "start": "00:04:54", + "end": "00:04:55" + } + } + ], + "question_timestamps": "00:05:44", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To your right rear.", + "B": "To your left front.", + "C": "It is directly in front of you as you approach the whiteboard, slightly offset to your right.", + "D": "It sits behind you on the right side when you head toward the whiteboard, near where you started.", + "E": "To your left rear." + }, + "answer_label": "E", + "answer_text": "To your left rear.", + "q_id": "egolife_A1_D6_019-mcq_fixed_4", + "correct_answer": "To your left rear.", + "original_options": { + "A": "To your right rear.", + "B": "To your left front.", + "C": "It is directly in front of you as you approach the whiteboard, slightly offset to your right.", + "D": "It sits behind you on the right side when you head toward the whiteboard, near where you started.", + "E": "To your left rear." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_019", + "video_length": "00:11:40", + "qid": "egolife_A1_D6_019-mcq_fixed_5", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I walk towards the whiteboard, where is the bucket in relation to me?", + "answer": "Behind you.", + "question_chinese": "当我走向白板时，水桶在我的什么方位？", + "answer_chinese": "在我的后方", + "evidence": [ + { + "description": "When I was using the computer, I saw a bucket directly in front of me", + "timestamps": { + "start": "00:04:54", + "end": "00:04:55" + } + }, + { + "description": "After that, I stood up", + "timestamps": { + "start": "00:05:22", + "end": "00:05:23" + } + } + ], + "question_timestamps": "00:09:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left side.", + "C": "It is ahead of you near the base of the whiteboard, so you would pass it as you walk up.", + "D": "It remains to your right side as you move toward the whiteboard, roughly level with your shoulder.", + "E": "Behind you." + }, + "answer_label": "E", + "answer_text": "Behind you.", + "q_id": "egolife_A1_D6_019-mcq_fixed_5", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left side.", + "C": "It is ahead of you near the base of the whiteboard, so you would pass it as you walk up.", + "D": "It remains to your right side as you move toward the whiteboard, roughly level with your shoulder.", + "E": "Behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_019", + "video_length": "00:11:40", + "qid": "egolife_A1_D6_019-mcq_fixed_8", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the second floor to the kitchen?", + "answer": "Walk down two flights of stairs from the second floor to the kitchen.", + "question_chinese": "我从二楼到厨房的路径？", + "answer_chinese": "从二楼下俩层楼梯，走到厨房。", + "evidence": [ + { + "description": "I went down two flights of stairs", + "timestamps": { + "start": "00:06:40", + "end": "00:06:49" + } + }, + { + "description": "Then go into the living room to reach the kitchen", + "timestamps": { + "start": "00:06:49", + "end": "00:06:52" + } + }, + { + "description": "I went out from the living room on the second floor", + "timestamps": { + "start": "00:06:39", + "end": "00:06:40" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Walk down one flight of stairs.", + "B": "Walk down three flights of stairs.", + "C": "Go down a single flight from the second floor and you will arrive directly at the kitchen entrance.", + "D": "Walk down two flights of stairs from the second floor to the kitchen.", + "E": "Descend three separate stair flights from the second floor before reaching the kitchen area." + }, + "answer_label": "D", + "answer_text": "Walk down two flights of stairs from the second floor to the kitchen.", + "q_id": "egolife_A1_D6_019-mcq_fixed_8", + "correct_answer": "Walk down two flights of stairs from the second floor to the kitchen.", + "original_options": { + "A": "Walk down one flight of stairs.", + "B": "Walk down three flights of stairs.", + "C": "Go down a single flight from the second floor and you will arrive directly at the kitchen entrance.", + "D": "Walk down two flights of stairs from the second floor to the kitchen.", + "E": "Descend three separate stair flights from the second floor before reaching the kitchen area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_019", + "video_length": "00:11:40", + "qid": "egolife_A1_D6_019-mcq_fixed_9", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the second floor to the kitchen?", + "answer": "Enter the hall and walk to the kitchen.", + "question_chinese": "我从二楼到厨房的路径？", + "answer_chinese": "进入大厅走到厨房。", + "evidence": [ + { + "description": "I went out from the living room on the second floor", + "timestamps": { + "start": "00:06:39", + "end": "00:06:40" + } + }, + { + "description": "Then go into the living room to reach the kitchen", + "timestamps": { + "start": "00:06:49", + "end": "00:06:52" + } + } + ], + "question_timestamps": "00:11:40", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Enter the hall and turn left.", + "B": "Enter the hall and go upstairs.", + "C": "Enter the hall, walk past the mirror, and continue into the pantry instead of the kitchen.", + "D": "You enter the hall, head through the lounge area, and only then continue on to the kitchen.", + "E": "Enter the hall and walk to the kitchen." + }, + "answer_label": "E", + "answer_text": "Enter the hall and walk to the kitchen.", + "q_id": "egolife_A1_D6_019-mcq_fixed_9", + "correct_answer": "Enter the hall and walk to the kitchen.", + "original_options": { + "A": "Enter the hall and turn left.", + "B": "Enter the hall and go upstairs.", + "C": "Enter the hall, walk past the mirror, and continue into the pantry instead of the kitchen.", + "D": "You enter the hall, head through the lounge area, and only then continue on to the kitchen.", + "E": "Enter the hall and walk to the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_019", + "video_length": "00:11:40", + "qid": "egolife_A1_D6_019-mcq_fixed_0", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "The laptop is behind you.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "笔记本电脑在我的后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + }, + { + "description": "I turned 90 degrees and left the dormitory.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The laptop is in front of you.", + "B": "The laptop is behind you.", + "C": "The laptop is to your left.", + "D": "The laptop is ahead of you on the desk as you step through the doorway into the room.", + "E": "The laptop is off to your right as you exit the bedroom door, near the side of the room." + }, + "answer_label": "B", + "answer_text": "The laptop is behind you.", + "q_id": "egolife_A1_D6_019-mcq_fixed_0", + "correct_answer": "The laptop is behind you.", + "original_options": { + "A": "The laptop is in front of you.", + "B": "The laptop is behind you.", + "C": "The laptop is to your left.", + "D": "The laptop is ahead of you on the desk as you step through the doorway into the room.", + "E": "The laptop is off to your right as you exit the bedroom door, near the side of the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_019", + "video_length": "00:11:40", + "qid": "egolife_A1_D6_019-mcq_fixed_1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "我刚出卧室门时，笔记本电脑在我的左后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:03:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "It was behind you to the right.", + "C": "When you stepped out of the bedroom door, the laptop was directly behind you without being on the left side.", + "D": "When you stepped out of the bedroom door, the laptop was slightly ahead and to your left instead of behind you.", + "E": "When YOU stepped out of the bedroom door, the laptop was behind you to the left." + }, + "answer_label": "E", + "answer_text": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "q_id": "egolife_A1_D6_019-mcq_fixed_1", + "correct_answer": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "original_options": { + "A": "It was in front of you.", + "B": "It was behind you to the right.", + "C": "When you stepped out of the bedroom door, the laptop was directly behind you without being on the left side.", + "D": "When you stepped out of the bedroom door, the laptop was slightly ahead and to your left instead of behind you.", + "E": "When YOU stepped out of the bedroom door, the laptop was behind you to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_019", + "video_length": "00:11:40", + "qid": "egolife_A1_D6_019_7", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is farther from where I am standing right now, the toilet paper or the egg tart?", + "answer": "Stay away from the toilet paper.", + "question_chinese": "我现在站着的位置，距离卫生纸和蛋挞哪个远？", + "answer_chinese": "离卫生纸远一些", + "evidence": [ + { + "description": "Egg tarts and toilet paper are directly in front of me", + "timestamps": { + "start": "00:07:45", + "end": "00:07:46" + } + }, + { + "description": "At my position, the toilet paper is in front of the egg tart.", + "timestamps": { + "start": "00:07:45", + "end": "00:07:46" + } + } + ], + "question_timestamps": "00:07:57", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Stay away from the toilet paper.", + "B": "Stay away from the egg tarts a bit." + }, + "answer_label": "A", + "answer_text": "Stay away from the toilet paper.", + "q_id": "egolife_A1_D6_019_7", + "correct_answer": "Stay away from the toilet paper.", + "original_options": { + "A": "Stay away from the toilet paper.", + "B": "Stay away from the egg tarts a bit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D6_019", + "video_length": "00:11:40", + "qid": "egolife_A1_D6_019_8", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is farther from where I am standing right now, the toilet paper or the egg tart?", + "answer": "Stay away from the egg tarts a bit.", + "question_chinese": "我现在站着的位置，距离卫生纸和蛋挞哪个远？", + "answer_chinese": "离蛋挞远一些", + "evidence": [ + { + "description": "They are all right in front of me", + "timestamps": { + "start": "00:08:57", + "end": "00:08:58" + } + }, + { + "description": "In my opinion, the egg tart is in front of the toilet paper", + "timestamps": { + "start": "00:08:57", + "end": "00:08:58" + } + } + ], + "question_timestamps": "00:09:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Stay away from the toilet paper.", + "B": "Stay away from the egg tarts a bit." + }, + "answer_label": "B", + "answer_text": "Stay away from the egg tarts a bit.", + "q_id": "egolife_A1_D6_019_8", + "correct_answer": "Stay away from the egg tarts a bit.", + "original_options": { + "A": "Stay away from the toilet paper.", + "B": "Stay away from the egg tarts a bit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D6_019", + "video_length": "00:11:40", + "qid": "egolife_A1_D6_019-mcq_fixed_2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path of the white small power bank?", + "answer": "YOU first picked it up from the table, then took it into two bedrooms one after another, and finally plugged it into an outlet in the living room.", + "question_chinese": "白色小充电宝的路径是什么？", + "answer_chinese": "我先从桌子上拿起了它，我拿着它接连走进了俩个卧室，最后插到了客厅的插座上", + "evidence": [ + { + "description": "It was first on the dining table", + "timestamps": { + "start": "00:00:27", + "end": "00:00:28" + } + } + ], + "question_timestamps": "00:00:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You left it on the table.", + "B": "You took it to one bedroom.", + "C": "You picked it up from the table, went to one bedroom, then plugged it into a bedroom outlet.", + "D": "You picked it up, visited the living room first, then took it into two bedrooms and never plugged it in.", + "E": "YOU first picked it up from the table, then took it into two bedrooms one after another, and finally plugged it into an outlet in the living room." + }, + "answer_label": "E", + "answer_text": "YOU first picked it up from the table, then took it into two bedrooms one after another, and finally plugged it into an outlet in the living room.", + "q_id": "egolife_A1_D6_019-mcq_fixed_2", + "correct_answer": "YOU first picked it up from the table, then took it into two bedrooms one after another, and finally plugged it into an outlet in the living room.", + "original_options": { + "A": "You left it on the table.", + "B": "You took it to one bedroom.", + "C": "You picked it up from the table, went to one bedroom, then plugged it into a bedroom outlet.", + "D": "You picked it up, visited the living room first, then took it into two bedrooms and never plugged it in.", + "E": "YOU first picked it up from the table, then took it into two bedrooms one after another, and finally plugged it into an outlet in the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_019", + "video_length": "00:11:40", + "qid": "egolife_A1_D6_019-mcq_fixed_3", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path of the white small power bank?", + "answer": "YOU plugged him into the socket in the living room.", + "question_chinese": "白色小充电宝的路径是什么？", + "answer_chinese": "我把他插到了客厅的插座上", + "evidence": [ + { + "description": "It was first on the dining table", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:03:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "YOU plugged him into the socket in the living room.", + "B": "You plugged it in the kitchen.", + "C": "You plugged it in the bedroom.", + "D": "You carried the power bank to the kitchen and plugged it into a socket there instead of the living room.", + "E": "You didn’t plug it in at all; you put the power bank away in a drawer after using it." + }, + "answer_label": "A", + "answer_text": "YOU plugged him into the socket in the living room.", + "q_id": "egolife_A1_D6_019-mcq_fixed_3", + "correct_answer": "YOU plugged him into the socket in the living room.", + "original_options": { + "A": "YOU plugged him into the socket in the living room.", + "B": "You plugged it in the kitchen.", + "C": "You plugged it in the bedroom.", + "D": "You carried the power bank to the kitchen and plugged it into a socket there instead of the living room.", + "E": "You didn’t plug it in at all; you put the power bank away in a drawer after using it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 20 + ] + } +] diff --git a/QAs/egolife_A1_D6_020-mcq.json b/QAs/egolife_A1_D6_020-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f112e08fd670fcaefa40d13a6535bd439c2beb68 --- /dev/null +++ b/QAs/egolife_A1_D6_020-mcq.json @@ -0,0 +1,879 @@ +[ + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_4_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From where I am right now, can I reach the toilet paper?", + "answer": "No.", + "question_chinese": "在这个地方，我是否能够到卫生纸？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "Toilet paper is on the dining table.", + "timestamps": { + "start": "00:09:14", + "end": "00:09:15" + } + }, + { + "description": "I am about 1 meter away from the table.", + "timestamps": { + "start": "00:09:14", + "end": "00:09:15" + } + } + ], + "question_timestamps": "00:09:26", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No.", + "B": "Yes.", + "C": "Yes, easily.", + "D": "Yes, you can reach it from where you are without standing up or taking a step.", + "E": "Yes, it’s within arm’s reach right now, so you could grab it by leaning forward slightly." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D6_020_4_T1", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes.", + "C": "Yes, easily.", + "D": "Yes, you can reach it from where you are without standing up or taking a step.", + "E": "Yes, it’s within arm’s reach right now, so you could grab it by leaning forward slightly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_4_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "\"Is there a place where I can get toilet paper?\"", + "answer": "No.", + "question_chinese": "在这个地方，我是否能够到卫生纸？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "Toilet paper is on the dining table.", + "timestamps": { + "start": "00:09:14", + "end": "00:09:15" + } + }, + { + "description": "I am about 1 meter away from the table.", + "timestamps": { + "start": "00:09:14", + "end": "00:09:15" + } + } + ], + "question_timestamps": "00:09:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No.", + "B": "Yes, in the bathroom.", + "C": "Yes, in the kitchen.", + "D": "Yes, you can get toilet paper by going to the bathroom area where it is kept for use.", + "E": "Yes, there’s a place to get toilet paper if you head to the kitchen side and check nearby storage." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D6_020_4_T2", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes, in the bathroom.", + "C": "Yes, in the kitchen.", + "D": "Yes, you can get toilet paper by going to the bathroom area where it is kept for use.", + "E": "Yes, there’s a place to get toilet paper if you head to the kitchen side and check nearby storage." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_5_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "From where I am right now, can I reach the toilet paper?", + "answer": "Yes.", + "question_chinese": "在这个地方，我是否能够到卫生纸？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "I am about 50 centimeters away from the toilet paper.", + "timestamps": { + "start": "00:09:20", + "end": "00:09:21" + } + } + ], + "question_timestamps": "00:09:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No.", + "B": "No, it’s out of reach.", + "C": "Yes.", + "D": "No, you can see the toilet paper, but you can’t reach it from where you are standing now.", + "E": "No, you would need to move closer before you could grab the toilet paper from your current position." + }, + "answer_label": "C", + "answer_text": "Yes.", + "q_id": "egolife_A1_D6_020_5_T1", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "No, it’s out of reach.", + "C": "Yes.", + "D": "No, you can see the toilet paper, but you can’t reach it from where you are standing now.", + "E": "No, you would need to move closer before you could grab the toilet paper from your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 6, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_5_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "\"Can I get toilet paper in this place?\"", + "answer": "Yes.", + "question_chinese": "在这个地方，我是否能够到卫生纸？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "I am about 50 centimeters away from the toilet paper.", + "timestamps": { + "start": "00:09:20", + "end": "00:09:21" + } + } + ], + "question_timestamps": "00:09:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No.", + "B": "No, not here.", + "C": "Yes.", + "D": "No, there isn’t any toilet paper available for you to get in this place right now.", + "E": "No, even if you look around, you won’t be able to get toilet paper from this location." + }, + "answer_label": "C", + "answer_text": "Yes.", + "q_id": "egolife_A1_D6_020_5_T2", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "No, not here.", + "C": "Yes.", + "D": "No, there isn’t any toilet paper available for you to get in this place right now.", + "E": "No, even if you look around, you won’t be able to get toilet paper from this location." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 3, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_6_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I turn around in the second-floor main kitchen?", + "answer": "Four.", + "question_chinese": "我从二楼大厨房进行了几次转身？", + "answer_chinese": "俩次。", + "evidence": [ + { + "description": "I go straight first, then turn around to face the others.", + "timestamps": { + "start": "00:08:52", + "end": "00:08:56" + } + }, + { + "description": "Then I turned around and walked into the staircase.", + "timestamps": { + "start": "00:08:56", + "end": "00:08:57" + } + } + ], + "question_timestamps": "00:08:58", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Once.", + "C": "Twice.", + "D": "You turned around three times while you were in the second-floor main kitchen.", + "E": "You turned around five times in the second-floor main kitchen before moving on." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "egolife_A1_D6_020_6_T1", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Once.", + "C": "Twice.", + "D": "You turned around three times while you were in the second-floor main kitchen.", + "E": "You turned around five times in the second-floor main kitchen before moving on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_6_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I turn around in the second-floor main kitchen?", + "answer": "Twice.", + "question_chinese": "我从二楼大厨房进行了几次转身？", + "answer_chinese": "俩次。", + "evidence": [ + { + "description": "I go straight first, then turn around to face the others.", + "timestamps": { + "start": "00:08:52", + "end": "00:08:56" + } + }, + { + "description": "Then I turned around and walked into the staircase.", + "timestamps": { + "start": "00:08:56", + "end": "00:08:57" + } + } + ], + "question_timestamps": "00:09:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Once.", + "B": "Three times.", + "C": "You turned around four times while you were in the second-floor main kitchen.", + "D": "You turned around five times in the second-floor main kitchen during that segment.", + "E": "Twice." + }, + "answer_label": "E", + "answer_text": "Twice.", + "q_id": "egolife_A1_D6_020_6_T2", + "correct_answer": "Twice.", + "original_options": { + "A": "Once.", + "B": "Three times.", + "C": "You turned around four times while you were in the second-floor main kitchen.", + "D": "You turned around five times in the second-floor main kitchen during that segment.", + "E": "Twice." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_6_T3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I turn around in the second-floor main kitchen?", + "answer": "Six times.", + "question_chinese": "我从二楼大厨房进行了几次转身？", + "answer_chinese": "六次", + "evidence": [ + { + "description": "I passed two corners while going down the stairs.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:05" + } + }, + { + "description": "Turn around when entering the door.", + "timestamps": { + "start": "00:09:05", + "end": "00:09:06" + } + }, + { + "description": "Turned two small corners in the hall.", + "timestamps": { + "start": "00:09:07", + "end": "00:09:08" + } + } + ], + "question_timestamps": "00:12:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Four times.", + "B": "Seven times.", + "C": "You turned around twice in the second-floor main kitchen during that moment.", + "D": "Six times.", + "E": "You turned around eight times in the second-floor main kitchen before leaving the area." + }, + "answer_label": "D", + "answer_text": "Six times.", + "q_id": "egolife_A1_D6_020_6_T3", + "correct_answer": "Six times.", + "original_options": { + "A": "Four times.", + "B": "Seven times.", + "C": "You turned around twice in the second-floor main kitchen during that moment.", + "D": "Six times.", + "E": "You turned around eight times in the second-floor main kitchen before leaving the area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_3", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In what direction is the doll on the shelf located from where I am standing?", + "answer": "At this time, Play is to my left.", + "question_chinese": "我站在这个位置在架子上放着的玩偶位于我的什么方位？", + "answer_chinese": "此时玩我在我的左边。", + "evidence": [ + { + "description": "After I put down the wine bottle and walked straight ahead, I saw a doll in front of me.", + "timestamps": { + "start": "00:00:43", + "end": "00:00:44" + } + }, + { + "description": "I turned my body another 90 degrees.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To my right.", + "B": "Behind me.", + "C": "At this time, Play is directly in front of me on the shelf.", + "D": "At this time, Play is behind me and slightly to the right from where I am standing.", + "E": "At this time, Play is to my left." + }, + "answer_label": "E", + "answer_text": "At this time, Play is to my left.", + "q_id": "egolife_A1_D6_020_3", + "correct_answer": "At this time, Play is to my left.", + "original_options": { + "A": "To my right.", + "B": "Behind me.", + "C": "At this time, Play is directly in front of me on the shelf.", + "D": "At this time, Play is behind me and slightly to the right from where I am standing.", + "E": "At this time, Play is to my left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_4", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In what direction is the doll on the shelf located from where I am standing?", + "answer": "When I was playing with my laptop at the table, the doll was behind me to the right.", + "question_chinese": "我站在这个位置在架子上放着的玩偶位于我的什么方位？", + "answer_chinese": "当我在桌子前玩笔记本时，玩偶在我的右后方。", + "evidence": [ + { + "description": "After I put down the wine bottle and walked straight ahead, I saw a doll in front of me.", + "timestamps": { + "start": "00:00:44", + "end": "00:00:45" + } + }, + { + "description": "As I sat in front of the laptop, with my back to the doll.", + "timestamps": { + "start": "00:04:48", + "end": "00:04:50" + } + } + ], + "question_timestamps": "00:04:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front-left.", + "B": "When I was playing with my laptop at the table, the doll was behind me to the right.", + "C": "Behind-left.", + "D": "When I was playing with my laptop at the table, the doll was directly behind me, not off to the right.", + "E": "When I was playing with my laptop at the table, the doll was behind me to the left." + }, + "answer_label": "B", + "answer_text": "When I was playing with my laptop at the table, the doll was behind me to the right.", + "q_id": "egolife_A1_D6_020_4", + "correct_answer": "When I was playing with my laptop at the table, the doll was behind me to the right.", + "original_options": { + "A": "In front-left.", + "B": "When I was playing with my laptop at the table, the doll was behind me to the right.", + "C": "Behind-left.", + "D": "When I was playing with my laptop at the table, the doll was directly behind me, not off to the right.", + "E": "When I was playing with my laptop at the table, the doll was behind me to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 1, + 21, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_2_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many collection bags have I punched holes in?", + "answer": "A collection bag.", + "question_chinese": "直到现在，我已经在几个收集袋上打孔了？", + "answer_chinese": "一个收集袋。", + "evidence": [ + { + "description": "I took a storage bag from a girl's hand and made holes in it.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:48" + } + } + ], + "question_timestamps": "00:05:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two collection bags.", + "B": "Four collection bags.", + "C": "A collection bag.", + "D": "Up to now, you have punched holes in five collection bags total.", + "E": "So far, you have punched holes in ten collection bags in total." + }, + "answer_label": "C", + "answer_text": "A collection bag.", + "q_id": "egolife_A1_D6_020_2_T1", + "correct_answer": "A collection bag.", + "original_options": { + "A": "Two collection bags.", + "B": "Four collection bags.", + "C": "A collection bag.", + "D": "Up to now, you have punched holes in five collection bags total.", + "E": "So far, you have punched holes in ten collection bags in total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_2_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Until now, how many holes have I made in the collection bags?", + "answer": "A collection bag.", + "question_chinese": "直到现在，我已经在几个收集袋上打孔了？", + "answer_chinese": "一个收集袋。", + "evidence": [ + { + "description": "I took a storage bag from a girl's hand and made holes in it.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:48" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two collection bags.", + "B": "Four collection bags.", + "C": "A collection bag.", + "D": "You have made holes in two collection bags so far, puncturing each one once.", + "E": "You have made holes in five collection bags up to now, spreading the punctures across them." + }, + "answer_label": "C", + "answer_text": "A collection bag.", + "q_id": "egolife_A1_D6_020_2_T2", + "correct_answer": "A collection bag.", + "original_options": { + "A": "Two collection bags.", + "B": "Four collection bags.", + "C": "A collection bag.", + "D": "You have made holes in two collection bags so far, puncturing each one once.", + "E": "You have made holes in five collection bags up to now, spreading the punctures across them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_3_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many collection bags have I punched holes in?", + "answer": "Four storage bags.", + "question_chinese": "直到现在，我已经在几个收集袋上打孔了？", + "answer_chinese": "四个收纳袋。", + "evidence": [ + { + "description": "I first took two separately and hit these two first.", + "timestamps": { + "start": "00:05:38", + "end": "00:06:14" + } + }, + { + "description": "Then I took a dozen and hit two of them.", + "timestamps": { + "start": "00:06:36", + "end": "00:07:38" + } + } + ], + "question_timestamps": "00:07:49", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two storage bags.", + "B": "Five storage bags.", + "C": "You have punched holes in three storage bags so far, working through them one by one.", + "D": "Four storage bags.", + "E": "You have punched holes in one storage bag up to now, leaving the rest untouched." + }, + "answer_label": "D", + "answer_text": "Four storage bags.", + "q_id": "egolife_A1_D6_020_3_T1", + "correct_answer": "Four storage bags.", + "original_options": { + "A": "Two storage bags.", + "B": "Five storage bags.", + "C": "You have punched holes in three storage bags so far, working through them one by one.", + "D": "Four storage bags.", + "E": "You have punched holes in one storage bag up to now, leaving the rest untouched." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020_3_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "\"How many collection bags have I punctured so far?\"", + "answer": "Four storage bags.", + "question_chinese": "直到现在，我已经在几个收集袋上打孔了？", + "answer_chinese": "四个收纳袋。", + "evidence": [ + { + "description": "I first took two separately and hit these two first.", + "timestamps": { + "start": "00:05:38", + "end": "00:06:14" + } + }, + { + "description": "Then I took a dozen and hit two of them.", + "timestamps": { + "start": "00:06:36", + "end": "00:07:38" + } + } + ], + "question_timestamps": "00:07:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One storage bag.", + "B": "Three storage bags.", + "C": "You have punctured two storage bags so far, putting a hole in each bag.", + "D": "Four storage bags.", + "E": "You have punctured five storage bags up to now, making holes across all of them." + }, + "answer_label": "D", + "answer_text": "Four storage bags.", + "q_id": "egolife_A1_D6_020_3_T2", + "correct_answer": "Four storage bags.", + "original_options": { + "A": "One storage bag.", + "B": "Three storage bags.", + "C": "You have punctured two storage bags so far, putting a hole in each bag.", + "D": "Four storage bags.", + "E": "You have punctured five storage bags up to now, making holes across all of them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020-mcq_fixed_0", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the route of the wine bottle?", + "answer": "YOU moved the wine bottle from the left corner of the table to the right corner of the table.", + "question_chinese": "酒瓶的路线是什么？", + "answer_chinese": "我把酒瓶从桌子的左角边拿到了桌子上的右角边。", + "evidence": [ + { + "description": "I walked to the left side of the table and saw the wine bottle.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:00:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "YOU moved the wine bottle from the left corner of the table to the right corner of the table.", + "B": "YOU moved it right corner to left.", + "C": "YOU moved it from center to left corner.", + "D": "YOU moved the wine bottle from the middle of the table to the near edge in front of you.", + "E": "YOU moved the wine bottle from the left corner of the table to the center of the table." + }, + "answer_label": "A", + "answer_text": "YOU moved the wine bottle from the left corner of the table to the right corner of the table.", + "q_id": "egolife_A1_D6_020-mcq_fixed_0", + "correct_answer": "YOU moved the wine bottle from the left corner of the table to the right corner of the table.", + "original_options": { + "A": "YOU moved the wine bottle from the left corner of the table to the right corner of the table.", + "B": "YOU moved it right corner to left.", + "C": "YOU moved it from center to left corner.", + "D": "YOU moved the wine bottle from the middle of the table to the near edge in front of you.", + "E": "YOU moved the wine bottle from the left corner of the table to the center of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_020", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_020-mcq_fixed_1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the route of the wine bottle?", + "answer": "YOU moved it to the bottom right corner.", + "question_chinese": "酒瓶的路线是什么？", + "answer_chinese": "我把它移动到右角边。", + "evidence": [ + { + "description": "I walked to the left side of the table and saw the wine bottle.", + "timestamps": { + "start": "00:00:36", + "end": "00:00:37" + } + } + ], + "question_timestamps": "00:05:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "YOU moved it to the top right corner.", + "B": "YOU moved it to the bottom left corner.", + "C": "YOU moved it to the top left corner, away from where it started.", + "D": "YOU moved it to the bottom right corner.", + "E": "YOU moved it to the center area instead of placing it in any corner." + }, + "answer_label": "D", + "answer_text": "YOU moved it to the bottom right corner.", + "q_id": "egolife_A1_D6_020-mcq_fixed_1", + "correct_answer": "YOU moved it to the bottom right corner.", + "original_options": { + "A": "YOU moved it to the top right corner.", + "B": "YOU moved it to the bottom left corner.", + "C": "YOU moved it to the top left corner, away from where it started.", + "D": "YOU moved it to the bottom right corner.", + "E": "YOU moved it to the center area instead of placing it in any corner." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 13, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D6_021-mcq.json b/QAs/egolife_A1_D6_021-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d84f67a5add39145293fd88e303b5c6e4a39648d --- /dev/null +++ b/QAs/egolife_A1_D6_021-mcq.json @@ -0,0 +1,678 @@ +[ + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021_4_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the food on the plate?", + "answer": "Yes.", + "question_chinese": "在我站着的地方我是否能够到餐盘里的东西？", + "answer_chinese": "能。", + "evidence": [ + { + "description": "The plate is directly in front of me, about one meter away.", + "timestamps": { + "start": "00:07:18", + "end": "00:07:19" + } + } + ], + "question_timestamps": "00:07:23", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D6_021_4_T1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021_4_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the items in the tray from where I am standing?", + "answer": "Yes.", + "question_chinese": "在我站着的地方我是否能够到餐盘里的东西？", + "answer_chinese": "能。", + "evidence": [ + { + "description": "The plate is directly in front of me, about one meter away.", + "timestamps": { + "start": "00:07:18", + "end": "00:07:19" + } + } + ], + "question_timestamps": "00:07:24", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D6_021_4_T2", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021_5_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the food on the plate?", + "answer": "No.", + "question_chinese": "在我站着的地方我是否能够到餐盘里的东西？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The plate is in front of me to the right.", + "timestamps": { + "start": "00:07:45", + "end": "00:07:46" + } + }, + { + "description": "I am about 1 meter away from the plate.", + "timestamps": { + "start": "00:07:45", + "end": "00:07:46" + } + } + ], + "question_timestamps": "00:07:50", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D6_021_5_T1", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021_5_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the items in the tray from where I'm standing?", + "answer": "No.", + "question_chinese": "在我站着的地方我是否能够到餐盘里的东西？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The plate is in front of me to the right.", + "timestamps": { + "start": "00:07:45", + "end": "00:07:46" + } + }, + { + "description": "I am about 1 meter away from the plate.", + "timestamps": { + "start": "00:07:45", + "end": "00:07:46" + } + } + ], + "question_timestamps": "00:07:51", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D6_021_5_T2", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021-mcq_fixed_5", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the living room door, where was the violin in relation to me?", + "answer": "To your right rear.", + "question_chinese": "当我刚出客厅门时，小提琴在我的什么方位？", + "answer_chinese": "在我的右后方", + "evidence": [ + { + "description": "When I walked towards the violin, it was directly in front of me", + "timestamps": { + "start": "00:04:36", + "end": "00:04:37" + } + }, + { + "description": "As I walked toward the door, the violin was to my right", + "timestamps": { + "start": "00:04:38", + "end": "00:04:39" + } + } + ], + "question_timestamps": "00:04:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To your left rear.", + "B": "To your right front.", + "C": "In front of you, slightly right, near the doorway as you stepped out of the living room.", + "D": "To your right rear.", + "E": "Directly behind you, a bit left, positioned close to the hallway area outside the living room." + }, + "answer_label": "D", + "answer_text": "To your right rear.", + "q_id": "egolife_A1_D6_021-mcq_fixed_5", + "correct_answer": "To your right rear.", + "original_options": { + "A": "To your left rear.", + "B": "To your right front.", + "C": "In front of you, slightly right, near the doorway as you stepped out of the living room.", + "D": "To your right rear.", + "E": "Directly behind you, a bit left, positioned close to the hallway area outside the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021-mcq_fixed_9", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the living room door, where was the violin in relation to me?", + "answer": "Behind you.", + "question_chinese": "当我刚出客厅门时，小提琴在我的什么方位？", + "answer_chinese": "在我的后方", + "evidence": [ + { + "description": "When I walked towards the violin, it was directly in front of me", + "timestamps": { + "start": "00:04:35", + "end": "00:04:36" + } + }, + { + "description": "As I walked toward the door, the violin was to my right", + "timestamps": { + "start": "00:04:36", + "end": "00:04:37" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "In front of you and slightly to the right, near where the living room opens into the hallway.", + "E": "To your right, a little forward, positioned near the edge of the doorway as you exited." + }, + "answer_label": "A", + "answer_text": "Behind you.", + "q_id": "egolife_A1_D6_021-mcq_fixed_9", + "correct_answer": "Behind you.", + "original_options": { + "A": "Behind you.", + "B": "In front of you.", + "C": "To your left.", + "D": "In front of you and slightly to the right, near where the living room opens into the hallway.", + "E": "To your right, a little forward, positioned near the edge of the doorway as you exited." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021_1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement path of the small laptop on the table?", + "answer": "I picked it up from the table and carried it through the living room, then sat down in the chair.", + "question_chinese": "桌子上小笔记本电脑的移动路径是什么？", + "answer_chinese": "我从桌子上把它拿起，并拿着它走过客厅，坐到了椅子上。", + "evidence": [ + { + "description": "The laptop was originally placed on the table.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:13" + } + }, + { + "description": "I am sitting on a chair and using a laptop.", + "timestamps": { + "start": "00:00:24", + "end": "00:00:25" + } + } + ], + "question_timestamps": "00:00:29", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I carried it into the hallway.", + "B": "I left it on the table.", + "C": "I picked it up from the table, walked through the hallway, and then set it down on the chair.", + "D": "I picked it up from the table and carried it through the living room, then sat down in the chair.", + "E": "I grabbed it off the table, carried it around the kitchen area, and then sat down without it." + }, + "answer_label": "D", + "answer_text": "I picked it up from the table and carried it through the living room, then sat down in the chair.", + "q_id": "egolife_A1_D6_021_1", + "correct_answer": "I picked it up from the table and carried it through the living room, then sat down in the chair.", + "original_options": { + "A": "I carried it into the hallway.", + "B": "I left it on the table.", + "C": "I picked it up from the table, walked through the hallway, and then set it down on the chair.", + "D": "I picked it up from the table and carried it through the living room, then sat down in the chair.", + "E": "I grabbed it off the table, carried it around the kitchen area, and then sat down without it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021_2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement path of the small laptop on the table?", + "answer": "When I sat on the chair, the notebook fell off my legs, and then I put it back on the table.", + "question_chinese": "桌子上小笔记本电脑的移动路径是什么？", + "answer_chinese": "当我坐在椅子上时，笔记本从我的腿上，然后被我放回到桌子上。", + "evidence": [ + { + "description": "I am sitting on the chair using the laptop", + "timestamps": { + "start": "00:01:33", + "end": "00:01:34" + } + }, + { + "description": "Finally, it was placed on the table.", + "timestamps": { + "start": "00:02:24", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:03:08", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It stayed on your legs.", + "B": "You put it on the chair.", + "C": "When you sat on the chair, it slipped off your lap and you left it on the floor.", + "D": "As you sat down, it fell from your legs onto the chair, and then you moved it to the couch.", + "E": "When I sat on the chair, the notebook fell off my legs, and then I put it back on the table." + }, + "answer_label": "E", + "answer_text": "When I sat on the chair, the notebook fell off my legs, and then I put it back on the table.", + "q_id": "egolife_A1_D6_021_2", + "correct_answer": "When I sat on the chair, the notebook fell off my legs, and then I put it back on the table.", + "original_options": { + "A": "It stayed on your legs.", + "B": "You put it on the chair.", + "C": "When you sat on the chair, it slipped off your lap and you left it on the floor.", + "D": "As you sat down, it fell from your legs onto the chair, and then you moved it to the couch.", + "E": "When I sat on the chair, the notebook fell off my legs, and then I put it back on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021_3_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to this point, how many turns have I made in total?", + "answer": "Three.", + "question_chinese": "此时我一共经过了几个弯？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "I made my first turn just as I stepped out of the gate.", + "timestamps": { + "start": "00:08:43", + "end": "00:08:44" + } + }, + { + "description": "Then I went through two more corners.", + "timestamps": { + "start": "00:09:10", + "end": "00:09:20" + } + } + ], + "question_timestamps": "00:09:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You made two turns total before reaching this point in your route.", + "E": "You have made four turns overall up to now, counting each change in direction." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_D6_021_3_T1", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You made two turns total before reaching this point in your route.", + "E": "You have made four turns overall up to now, counting each change in direction." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021_3_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to this point, how many turns have I made in total?", + "answer": "Three.", + "question_chinese": "此时我一共经过了几个弯？", + "answer_chinese": "五个。", + "evidence": [ + { + "description": "I first took three turns.", + "timestamps": { + "start": "00:09:21", + "end": "00:09:22" + } + }, + { + "description": "I made the fourth turn in the white car.", + "timestamps": { + "start": "00:09:31", + "end": "00:09:32" + } + } + ], + "question_timestamps": "00:10:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Three.", + "B": "Two.", + "C": "Five.", + "D": "You have only made two turns so far during the movement up to this point.", + "E": "You made five turns in total up to now, including every time you changed direction." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "egolife_A1_D6_021_3_T2", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two.", + "C": "Five.", + "D": "You have only made two turns so far during the movement up to this point.", + "E": "You made five turns in total up to now, including every time you changed direction." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021_3", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment, which is closer to me, the drink or the scissors?", + "answer": "I am a bit closer to the drink.", + "question_chinese": "此时我离饮料和剪刀哪个近？", + "answer_chinese": "我离饮料更近一些。", + "evidence": [ + { + "description": "The drink is in front of me, about 1 meter away from where I'm standing.", + "timestamps": { + "start": "00:05:09", + "end": "00:05:10" + } + }, + { + "description": "At this moment, the place where I am standing is about 1.5 meters away from the scissors.", + "timestamps": { + "start": "00:05:09", + "end": "00:05:10" + } + }, + { + "description": "The distance between the drink and the scissors is about 2 meters.", + "timestamps": { + "start": "00:05:09", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:05:21", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "I am a bit closer to the scissors.", + "B": "I am a bit closer to the drink." + }, + "answer_label": "B", + "answer_text": "I am a bit closer to the drink.", + "q_id": "egolife_A1_D6_021_3", + "correct_answer": "I am a bit closer to the drink.", + "original_options": { + "A": "I am a bit closer to the scissors.", + "B": "I am a bit closer to the drink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D6_021", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_021_9", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment, which is closer to me, the drink or the scissors?", + "answer": "Get closer to the scissors.", + "question_chinese": "此时我离饮料和剪刀哪个近？", + "answer_chinese": "离剪刀更近一些。", + "evidence": [ + { + "description": "At this moment, I am standing further away from the table, with the scissors in front of the drink.", + "timestamps": { + "start": "00:06:28", + "end": "00:06:29" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Get closer to the scissors.", + "B": "Get closer to the drink." + }, + "answer_label": "A", + "answer_text": "Get closer to the scissors.", + "q_id": "egolife_A1_D6_021_9", + "correct_answer": "Get closer to the scissors.", + "original_options": { + "A": "Get closer to the scissors.", + "B": "Get closer to the drink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + } +] diff --git a/QAs/egolife_A1_D6_022-mcq.json b/QAs/egolife_A1_D6_022-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..47544100d5acb62f5bbb471e09ca1c3afd44993c --- /dev/null +++ b/QAs/egolife_A1_D6_022-mcq.json @@ -0,0 +1,836 @@ +[ + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022-mcq_fixed_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from outside to the second floor?", + "answer": "YOU opened the main door, entered the room, then walked up two flights of stairs to reach the second floor.", + "question_chinese": "我从室外到二楼的路径是什么？", + "answer_chinese": "我打开大门，进入室内，然后走俩层楼梯，到达二楼。", + "evidence": [ + { + "description": "I closed the door and walked through the corridor.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You took one flight of stairs.", + "B": "YOU opened the main door, entered the room, then walked up two flights of stairs to reach the second floor.", + "C": "You took three flights of stairs.", + "D": "YOU opened the main door, entered the room, then walked up one flight of stairs to reach the second floor.", + "E": "YOU opened the main door, entered the room, then walked up three flights of stairs before arriving on the second floor." + }, + "answer_label": "B", + "answer_text": "YOU opened the main door, entered the room, then walked up two flights of stairs to reach the second floor.", + "q_id": "egolife_A1_D6_022-mcq_fixed_6", + "correct_answer": "YOU opened the main door, entered the room, then walked up two flights of stairs to reach the second floor.", + "original_options": { + "A": "You took one flight of stairs.", + "B": "YOU opened the main door, entered the room, then walked up two flights of stairs to reach the second floor.", + "C": "You took three flights of stairs.", + "D": "YOU opened the main door, entered the room, then walked up one flight of stairs to reach the second floor.", + "E": "YOU opened the main door, entered the room, then walked up three flights of stairs before arriving on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022-mcq_fixed_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from outside to the second floor?", + "answer": "YOU walked into the room and reached the second floor..", + "question_chinese": "我从室外到二楼的路径是什么？", + "answer_chinese": "我走入室内，到达二楼。", + "evidence": [ + { + "description": "I closed the door and walked through the corridor.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:08:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You walked into the blue car.", + "B": "YOU walked into the room and reached the second floor..", + "C": "You walked into the garden.", + "D": "You walked in from the rear entrance, crossed the kitchen, and climbed the narrow metal stairs to the second floor.", + "E": "You came in through the main hall, turned left, and went up the spiral staircase to reach the second floor." + }, + "answer_label": "B", + "answer_text": "YOU walked into the room and reached the second floor..", + "q_id": "egolife_A1_D6_022-mcq_fixed_7", + "correct_answer": "YOU walked into the room and reached the second floor..", + "original_options": { + "A": "You walked into the blue car.", + "B": "YOU walked into the room and reached the second floor..", + "C": "You walked into the garden.", + "D": "You walked in from the rear entrance, crossed the kitchen, and climbed the narrow metal stairs to the second floor.", + "E": "You came in through the main hall, turned left, and went up the spiral staircase to reach the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022-mcq_fixed_0", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "我刚出卧室门时，笔记本电脑在我的左后方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:04:06", + "end": "00:04:07" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:23" + } + } + ], + "question_timestamps": "00:04:24", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you to the right.", + "C": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "D": "When YOU stepped out of the bedroom door, the laptop was ahead of you to the right.", + "E": "When you stepped out of the bedroom door, the laptop was directly in front of you on a small desk." + }, + "answer_label": "C", + "answer_text": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "q_id": "egolife_A1_D6_022-mcq_fixed_0", + "correct_answer": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "original_options": { + "A": "In front of you.", + "B": "Behind you to the right.", + "C": "When YOU stepped out of the bedroom door, the laptop was behind you to the left.", + "D": "When YOU stepped out of the bedroom door, the laptop was ahead of you to the right.", + "E": "When you stepped out of the bedroom door, the laptop was directly in front of you on a small desk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022-mcq_fixed_1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "The laptop is to your left..", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "笔记本电脑在我的左方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:04:06", + "end": "00:04:07" + } + }, + { + "description": "I turned 0 degrees and left the dormitory.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:23" + } + } + ], + "question_timestamps": "00:08:43", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The laptop is behind you.", + "B": "The laptop is to your right.", + "C": "The laptop is to your left..", + "D": "The laptop is located diagonally in front of you to the right, near the bookshelf.", + "E": "As you leave the bedroom, the laptop is behind you, placed near the window and partially obscured by the chair." + }, + "answer_label": "C", + "answer_text": "The laptop is to your left..", + "q_id": "egolife_A1_D6_022-mcq_fixed_1", + "correct_answer": "The laptop is to your left..", + "original_options": { + "A": "The laptop is behind you.", + "B": "The laptop is to your right.", + "C": "The laptop is to your left..", + "D": "The laptop is located diagonally in front of you to the right, near the bookshelf.", + "E": "As you leave the bedroom, the laptop is behind you, placed near the window and partially obscured by the chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022_5", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many egg tart shells have I removed in total?", + "answer": "Three.", + "question_chinese": "我此时一共摘掉了几个蛋挞壳？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "I picked three egg tart shells in a row.", + "timestamps": { + "start": "00:07:55", + "end": "00:08:14" + } + } + ], + "question_timestamps": "00:08:25", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You have removed a total of two egg tart shells so far.", + "E": "Altogether, you have taken out four egg tart shells in total so far." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_D6_022_5", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "You have removed a total of two egg tart shells so far.", + "E": "Altogether, you have taken out four egg tart shells in total so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022_6", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many egg tart shells have I removed in total?", + "answer": "Three.", + "question_chinese": "我此时一共摘掉了几个蛋挞壳？", + "answer_chinese": "五个。", + "evidence": [ + { + "description": "I plucked one from the bottom right corner.", + "timestamps": { + "start": "00:08:50", + "end": "00:08:51" + } + }, + { + "description": "Three were picked off in front, then two more were picked off.", + "timestamps": { + "start": "00:08:14", + "end": "00:08:51" + } + } + ], + "question_timestamps": "00:09:02", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "In total, you have removed two egg tart shells altogether so far.", + "E": "You have taken out four egg tart shells in total up to now." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "egolife_A1_D6_022_6", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "In total, you have removed two egg tart shells altogether so far.", + "E": "You have taken out four egg tart shells in total up to now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022_3_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the refrigerator relative to me?", + "answer": "It’s in front of you.", + "question_chinese": "此时冰箱在我的哪个位置？", + "answer_chinese": "在你的正前方。", + "evidence": [ + { + "description": "Right in front of you.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s behind you.", + "C": "It’s to your left.", + "D": "It’s to your right, along the wall next to you now.", + "E": "It’s above you, mounted high on the wall nearby right now." + }, + "answer_label": "A", + "answer_text": "It’s in front of you.", + "q_id": "egolife_A1_D6_022_3_T1", + "correct_answer": "It’s in front of you.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s behind you.", + "C": "It’s to your left.", + "D": "It’s to your right, along the wall next to you now.", + "E": "It’s above you, mounted high on the wall nearby right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022_9", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator located right now?", + "answer": "To my left.", + "question_chinese": "此时冰箱在我的哪个位置？", + "answer_chinese": "在我的左边。", + "evidence": [ + { + "description": "As I walked toward the door, the refrigerator was directly in front of me.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + }, + { + "description": "When I turn 90 degrees.", + "timestamps": { + "start": "00:00:31", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:00:36", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of me.", + "B": "To my right.", + "C": "To my left.", + "D": "Behind me, slightly to the right side near the doorway, not on my left at all.", + "E": "Straight ahead of me, closer than the counter, rather than being located to my left." + }, + "answer_label": "C", + "answer_text": "To my left.", + "q_id": "egolife_A1_D6_022_9", + "correct_answer": "To my left.", + "original_options": { + "A": "In front of me.", + "B": "To my right.", + "C": "To my left.", + "D": "Behind me, slightly to the right side near the doorway, not on my left at all.", + "E": "Straight ahead of me, closer than the counter, rather than being located to my left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022_3_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the refrigerator relative to me?", + "answer": "It’s to your left.", + "question_chinese": "此时冰箱在我的哪个位置？", + "answer_chinese": "在你的左边。", + "evidence": [ + { + "description": "on your left", + "timestamps": { + "start": "00:00:59", + "end": "00:01:01" + } + } + ], + "question_timestamps": "00:01:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s to your right.", + "C": "It’s directly behind you at the moment, not on your left side where you would normally reach.", + "D": "It’s to your left.", + "E": "It’s ahead of you, slightly off to the right side, rather than being positioned to your left." + }, + "answer_label": "D", + "answer_text": "It’s to your left.", + "q_id": "egolife_A1_D6_022_3_T2", + "correct_answer": "It’s to your left.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s to your right.", + "C": "It’s directly behind you at the moment, not on your left side where you would normally reach.", + "D": "It’s to your left.", + "E": "It’s ahead of you, slightly off to the right side, rather than being positioned to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022_10", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator located right now?", + "answer": "To my left.", + "question_chinese": "此时冰箱在我的哪个位置？", + "answer_chinese": "在我的左面。", + "evidence": [ + { + "description": "As I walked toward the door, the refrigerator was directly in front of me.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + }, + { + "description": "When I turn 90 degrees.", + "timestamps": { + "start": "00:00:31", + "end": "00:00:32" + } + } + ], + "question_timestamps": "00:05:42", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of me.", + "B": "To my right.", + "C": "Behind me, a little off to the right, rather than being placed on my left side right now.", + "D": "Directly in front of me, not off to either side, instead of being located to my left.", + "E": "To my left." + }, + "answer_label": "E", + "answer_text": "To my left.", + "q_id": "egolife_A1_D6_022_10", + "correct_answer": "To my left.", + "original_options": { + "A": "In front of me.", + "B": "To my right.", + "C": "Behind me, a little off to the right, rather than being placed on my left side right now.", + "D": "Directly in front of me, not off to either side, instead of being located to my left.", + "E": "To my left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022_3", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the toilet in relation to me right now?", + "answer": "In front of me to the left.", + "question_chinese": "此时马桶在我的什么方位？", + "answer_chinese": "在我的左前方。", + "evidence": [ + { + "description": "I am standing in front of the sink.", + "timestamps": { + "start": "00:06:44", + "end": "00:06:45" + } + }, + { + "description": "The toilet is to the left of the sink.", + "timestamps": { + "start": "00:06:44", + "end": "00:06:45" + } + } + ], + "question_timestamps": "00:06:49", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind me to the right.", + "B": "In front of me to the left.", + "C": "In front of me to the right.", + "D": "Directly behind me, slightly angled toward my left side, not in front of me at all.", + "E": "In front of me but closer to the right side than the left, not front-left." + }, + "answer_label": "B", + "answer_text": "In front of me to the left.", + "q_id": "egolife_A1_D6_022_3", + "correct_answer": "In front of me to the left.", + "original_options": { + "A": "Behind me to the right.", + "B": "In front of me to the left.", + "C": "In front of me to the right.", + "D": "Directly behind me, slightly angled toward my left side, not in front of me at all.", + "E": "In front of me but closer to the right side than the left, not front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022_4", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction is the toilet in relation to me right now?", + "answer": "The toilet is behind me.", + "question_chinese": "此时马桶在我的什么方位？", + "answer_chinese": "马桶在我的后方。", + "evidence": [ + { + "description": "When I turn 180 degrees.", + "timestamps": { + "start": "00:07:04", + "end": "00:07:05" + } + } + ], + "question_timestamps": "00:07:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The toilet is in front of me.", + "B": "The toilet is to my left.", + "C": "The toilet is slightly in front of me, not behind me, so I would see it without turning around.", + "D": "The toilet is off to my right side rather than behind me, so it isn’t directly behind.", + "E": "The toilet is behind me." + }, + "answer_label": "E", + "answer_text": "The toilet is behind me.", + "q_id": "egolife_A1_D6_022_4", + "correct_answer": "The toilet is behind me.", + "original_options": { + "A": "The toilet is in front of me.", + "B": "The toilet is to my left.", + "C": "The toilet is slightly in front of me, not behind me, so I would see it without turning around.", + "D": "The toilet is off to my right side rather than behind me, so it isn’t directly behind.", + "E": "The toilet is behind me." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022_2_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many egg tart shells have I removed in total?", + "answer": "Right now, you’ve removed three egg tart shells in total, and the egg tart shells are in front of you.", + "question_chinese": "现在，我一共摘掉了几个蛋挞壳？", + "answer_chinese": "现在，你一共摘掉了三个蛋挞壳，蛋挞壳在你的正前方。", + "evidence": [ + { + "description": "Now, you have taken off three altogether.", + "timestamps": { + "start": "00:07:55", + "end": "00:08:14" + } + } + ], + "question_timestamps": "00:08:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two egg tart shells.", + "B": "Four egg tart shells.", + "C": "Right now, you’ve removed two egg tart shells in total, and the egg tart shells are in front of you.", + "D": "Right now, you’ve removed three egg tart shells in total, and the egg tart shells are in front of you.", + "E": "Right now, you’ve removed four egg tart shells in total, and the egg tart shells are in front of you." + }, + "answer_label": "D", + "answer_text": "Right now, you’ve removed three egg tart shells in total, and the egg tart shells are in front of you.", + "q_id": "egolife_A1_D6_022_2_T1", + "correct_answer": "Right now, you’ve removed three egg tart shells in total, and the egg tart shells are in front of you.", + "original_options": { + "A": "Two egg tart shells.", + "B": "Four egg tart shells.", + "C": "Right now, you’ve removed two egg tart shells in total, and the egg tart shells are in front of you.", + "D": "Right now, you’ve removed three egg tart shells in total, and the egg tart shells are in front of you.", + "E": "Right now, you’ve removed four egg tart shells in total, and the egg tart shells are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_022", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_022_2_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many egg tart shells have I removed in total?", + "answer": "Right now, you’ve removed five egg tart shells in total, and the egg tart shells are in front of you.", + "question_chinese": "现在，我一共摘掉了几个蛋挞壳？", + "answer_chinese": "现在，你一共摘掉了五个蛋挞壳，蛋挞壳在你的正前方。", + "evidence": [ + { + "description": "Now you've removed five altogether.", + "timestamps": { + "start": "00:07:50", + "end": "00:08:42" + } + } + ], + "question_timestamps": "00:08:44", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Four egg tart shells.", + "B": "Six egg tart shells.", + "C": "Right now, you’ve removed four egg tart shells in total, and the egg tart shells are in front of you.", + "D": "Right now, you’ve removed six egg tart shells in total, and the egg tart shells are in front of you.", + "E": "Right now, you’ve removed five egg tart shells in total, and the egg tart shells are in front of you." + }, + "answer_label": "E", + "answer_text": "Right now, you’ve removed five egg tart shells in total, and the egg tart shells are in front of you.", + "q_id": "egolife_A1_D6_022_2_T2", + "correct_answer": "Right now, you’ve removed five egg tart shells in total, and the egg tart shells are in front of you.", + "original_options": { + "A": "Four egg tart shells.", + "B": "Six egg tart shells.", + "C": "Right now, you’ve removed four egg tart shells in total, and the egg tart shells are in front of you.", + "D": "Right now, you’ve removed six egg tart shells in total, and the egg tart shells are in front of you.", + "E": "Right now, you’ve removed five egg tart shells in total, and the egg tart shells are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 21, + 21 + ] + } +] diff --git a/QAs/egolife_A1_D6_023-mcq.json b/QAs/egolife_A1_D6_023-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f43dd2e1056670e7bba3c1e3e5b363c246480b26 --- /dev/null +++ b/QAs/egolife_A1_D6_023-mcq.json @@ -0,0 +1,675 @@ +[ + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023_9", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many strawberry tarts did I put on the device in total?", + "answer": "Two.", + "question_chinese": "我一共放了几个草莓蛋挞在装置上？", + "answer_chinese": "俩个。", + "evidence": [ + { + "description": "I picked up the strawberry cream puff from the top left corner of the table.", + "timestamps": { + "start": "00:03:30", + "end": "00:03:41" + } + }, + { + "description": "Now there are two egg tarts on the plate.", + "timestamps": { + "start": "00:03:40", + "end": "00:03:41" + } + } + ], + "question_timestamps": "00:03:52", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You put four strawberry tarts on the device in total, not just two overall.", + "E": "In total, you placed five strawberry tarts onto the device during the whole process." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "egolife_A1_D6_023_9", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "You put four strawberry tarts on the device in total, not just two overall.", + "E": "In total, you placed five strawberry tarts onto the device during the whole process." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023_10", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many strawberry tarts did I put on the device in total?", + "answer": "Four.", + "question_chinese": "我一共放了几个草莓蛋挞在装置上？", + "answer_chinese": "四个。", + "evidence": [ + { + "description": "After coming down from the second floor, I put another one of his items on the second shelf.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:05:23", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "Two, because you only set down one tart twice on the device.", + "D": "Six, since you loaded three tarts first and then added three more later.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "egolife_A1_D6_023_10", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "Two, because you only set down one tart twice on the device.", + "D": "Six, since you loaded three tarts first and then added three more later.", + "E": "Four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023-mcq_fixed_6", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the first floor to the second floor?", + "answer": "Go out through the door on the first floor, then take two flights of stairs to reach the second floor.", + "question_chinese": "我从一楼到二楼的路径是什么？", + "answer_chinese": "从一楼门出去，然后上了俩层楼，到达二楼。", + "evidence": [ + { + "description": "I climbed two flights of stairs in a row.", + "timestamps": { + "start": "00:05:40", + "end": "00:05:51" + } + } + ], + "question_timestamps": "00:05:52", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Take the elevator upstairs.", + "B": "Use three flights of stairs.", + "C": "Go out through the first-floor door, then climb one long staircase to reach the second floor.", + "D": "Exit the first floor through the door and follow the corridor to a stairwell with only one flight up.", + "E": "Go out through the door on the first floor, then take two flights of stairs to reach the second floor." + }, + "answer_label": "E", + "answer_text": "Go out through the door on the first floor, then take two flights of stairs to reach the second floor.", + "q_id": "egolife_A1_D6_023-mcq_fixed_6", + "correct_answer": "Go out through the door on the first floor, then take two flights of stairs to reach the second floor.", + "original_options": { + "A": "Take the elevator upstairs.", + "B": "Use three flights of stairs.", + "C": "Go out through the first-floor door, then climb one long staircase to reach the second floor.", + "D": "Exit the first floor through the door and follow the corridor to a stairwell with only one flight up.", + "E": "Go out through the door on the first floor, then take two flights of stairs to reach the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023-mcq_fixed_7", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the first floor to the second floor?", + "answer": "YOU went out of the lobby and arrived at the second floor.", + "question_chinese": "我从一楼到二楼的路径是什么？", + "answer_chinese": "我出了大厅，到达二楼。", + "evidence": [ + { + "description": "I climbed two flights of stairs in a row.", + "timestamps": { + "start": "00:05:42", + "end": "00:05:51" + } + } + ], + "question_timestamps": "00:10:21", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "YOU arrived at the first floor.", + "B": "YOU went out of the lobby and arrived at the second floor.", + "C": "YOU arrived at the basement.", + "D": "YOU stayed inside the lobby and never went up to the second floor.", + "E": "YOU left the lobby, walked down the corridor, and ended up on the third floor instead." + }, + "answer_label": "B", + "answer_text": "YOU went out of the lobby and arrived at the second floor.", + "q_id": "egolife_A1_D6_023-mcq_fixed_7", + "correct_answer": "YOU went out of the lobby and arrived at the second floor.", + "original_options": { + "A": "YOU arrived at the first floor.", + "B": "YOU went out of the lobby and arrived at the second floor.", + "C": "YOU arrived at the basement.", + "D": "YOU stayed inside the lobby and never went up to the second floor.", + "E": "YOU left the lobby, walked down the corridor, and ended up on the third floor instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023-mcq_fixed_0", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the moving path of the book?", + "answer": "A person put Ben's book on the table. YOU picked it up and placed it on the small table next to it.", + "question_chinese": "本子的移动路径？", + "answer_chinese": "一个人把本在放到了桌子上，我把他拿起后，放在了旁边的小桌子上。", + "evidence": [ + { + "description": "I saw a girl put her notebook on the table.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "After I picked him up, I placed him on the small table by the door.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "YOU left it on the table.", + "B": "A person put Ben's book on the table. YOU picked it up and placed it on the small table next to it.", + "C": "YOU put it on a chair.", + "D": "A person put Ben's book on the table, and YOU moved it onto the small table behind it.", + "E": "Someone set Ben’s book on the table, then YOU carried it away and placed it on a nearby red chair." + }, + "answer_label": "B", + "answer_text": "A person put Ben's book on the table. YOU picked it up and placed it on the small table next to it.", + "q_id": "egolife_A1_D6_023-mcq_fixed_0", + "correct_answer": "A person put Ben's book on the table. YOU picked it up and placed it on the small table next to it.", + "original_options": { + "A": "YOU left it on the table.", + "B": "A person put Ben's book on the table. YOU picked it up and placed it on the small table next to it.", + "C": "YOU put it on a chair.", + "D": "A person put Ben's book on the table, and YOU moved it onto the small table behind it.", + "E": "Someone set Ben’s book on the table, then YOU carried it away and placed it on a nearby red chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023-mcq_fixed_1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the moving path of the book?", + "answer": "After picking him up, YOU placed him on the small table next to you.", + "question_chinese": "本子的移动路径？", + "answer_chinese": "我把他拿起后，放在了旁边的小桌子上。", + "evidence": [ + { + "description": "I saw a girl put her notebook on the table.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "After I picked him up, I placed him on the small table by the door.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:06:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You set it behind you.", + "B": "You set it on a large table.", + "C": "After picking him up, YOU placed him on the small table next to you.", + "D": "After picking him up, YOU carried him across the room and placed him on the coffee table.", + "E": "After picking him up, YOU walked to the kitchen counter and set him down there." + }, + "answer_label": "C", + "answer_text": "After picking him up, YOU placed him on the small table next to you.", + "q_id": "egolife_A1_D6_023-mcq_fixed_1", + "correct_answer": "After picking him up, YOU placed him on the small table next to you.", + "original_options": { + "A": "You set it behind you.", + "B": "You set it on a large table.", + "C": "After picking him up, YOU placed him on the small table next to you.", + "D": "After picking him up, YOU carried him across the room and placed him on the coffee table.", + "E": "After picking him up, YOU walked to the kitchen counter and set him down there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023_5", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator located right now?", + "answer": "The refrigerator is behind me.", + "question_chinese": "此时冰箱在我的哪个位置？", + "answer_chinese": "冰箱在我的后面。", + "evidence": [ + { + "description": "When I am on the right side of the table, the table is in front of and to the left of me.", + "timestamps": { + "start": "00:04:55", + "end": "00:04:56" + } + }, + { + "description": "I turned around at the door.", + "timestamps": { + "start": "00:05:20", + "end": "00:05:21" + } + } + ], + "question_timestamps": "00:05:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is in front of me.", + "B": "The refrigerator is to my right.", + "C": "The refrigerator is to my left, positioned beside you rather than behind your current stance.", + "D": "The refrigerator is behind me.", + "E": "The refrigerator is ahead of you across the room, not behind you where you’re facing." + }, + "answer_label": "D", + "answer_text": "The refrigerator is behind me.", + "q_id": "egolife_A1_D6_023_5", + "correct_answer": "The refrigerator is behind me.", + "original_options": { + "A": "The refrigerator is in front of me.", + "B": "The refrigerator is to my right.", + "C": "The refrigerator is to my left, positioned beside you rather than behind your current stance.", + "D": "The refrigerator is behind me.", + "E": "The refrigerator is ahead of you across the room, not behind you where you’re facing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023_6", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator located right now?", + "answer": "To my left.", + "question_chinese": "此时冰箱在我的哪个位置？", + "answer_chinese": "在我的右面。", + "evidence": [ + { + "description": "When I stood at the door, facing the table.", + "timestamps": { + "start": "00:05:32", + "end": "00:05:33" + } + } + ], + "question_timestamps": "00:06:43", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To my right.", + "B": "To my left.", + "C": "In front of me.", + "D": "Behind me, positioned at my back rather than on my left side.", + "E": "Above me, mounted higher up instead of being located to my left." + }, + "answer_label": "B", + "answer_text": "To my left.", + "q_id": "egolife_A1_D6_023_6", + "correct_answer": "To my left.", + "original_options": { + "A": "To my right.", + "B": "To my left.", + "C": "In front of me.", + "D": "Behind me, positioned at my back rather than on my left side.", + "E": "Above me, mounted higher up instead of being located to my left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023-mcq_fixed_2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment, which is closer to my location, the blueberries or the strawberries?", + "answer": "Blueberry.", + "question_chinese": "此时我的位置离蓝莓和草莓哪个近一些？", + "answer_chinese": "蓝莓。", + "evidence": [ + { + "description": "Blueberries are to my right, about 1 meter away.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + }, + { + "description": "Strawberries are in the upper left corner of the table, about 2 meters away from me.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + }, + { + "description": "Blueberries are to the right of strawberries.", + "timestamps": { + "start": "00:00:22", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:24", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Strawberry.", + "B": "Blueberry." + }, + "answer_label": "B", + "answer_text": "Blueberry.", + "q_id": "egolife_A1_D6_023-mcq_fixed_2", + "correct_answer": "Blueberry.", + "original_options": { + "A": "Strawberry.", + "B": "Blueberry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023-mcq_fixed_3", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "At this moment, which is closer to my location, the blueberries or the strawberries?", + "answer": "Blueberry.", + "question_chinese": "此时我的位置离蓝莓和草莓哪个近一些？", + "answer_chinese": "草莓。", + "evidence": [ + { + "description": "Blueberries and strawberries are both in front of me.", + "timestamps": { + "start": "00:05:13", + "end": "00:05:14" + } + }, + { + "description": "The blueberry is a bit farther from me than the strawberry.", + "timestamps": { + "start": "00:05:13", + "end": "00:05:14" + } + } + ], + "question_timestamps": "00:05:15", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Blueberry.", + "B": "Strawberry." + }, + "answer_label": "A", + "answer_text": "Blueberry.", + "q_id": "egolife_A1_D6_023-mcq_fixed_3", + "correct_answer": "Blueberry.", + "original_options": { + "A": "Blueberry.", + "B": "Strawberry." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023_4_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which side of me are the blueberries on?", + "answer": "The blueberries are in front of you.", + "question_chinese": "蓝莓在我的哪边？", + "answer_chinese": "蓝莓在你的前面。", + "evidence": [ + { + "description": "The blueberry is in front of you.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:33" + } + } + ], + "question_timestamps": "00:00:37", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The blueberries are behind you.", + "B": "The blueberries are in front of you." + }, + "answer_label": "B", + "answer_text": "The blueberries are in front of you.", + "q_id": "egolife_A1_D6_023_4_T2", + "correct_answer": "The blueberries are in front of you.", + "original_options": { + "A": "The blueberries are behind you.", + "B": "The blueberries are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D6_023", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_023_4_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which side of me are the blueberries on?", + "answer": "The blueberries are to your right.", + "question_chinese": "蓝莓在我的哪边？", + "answer_chinese": "蓝莓在你的右边", + "evidence": [ + { + "description": "The blueberry is on your right.", + "timestamps": { + "start": "00:00:57", + "end": "00:01:07" + } + } + ], + "question_timestamps": "00:01:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The blueberries are to your right.", + "B": "The blueberries are to your left." + }, + "answer_label": "A", + "answer_text": "The blueberries are to your right.", + "q_id": "egolife_A1_D6_023_4_T1", + "correct_answer": "The blueberries are to your right.", + "original_options": { + "A": "The blueberries are to your right.", + "B": "The blueberries are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/egolife_A1_D6_024-mcq.json b/QAs/egolife_A1_D6_024-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c864e98edcad3c4e30cb3be08d7cb13da10a0b18 --- /dev/null +++ b/QAs/egolife_A1_D6_024-mcq.json @@ -0,0 +1,921 @@ +[ + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024-mcq_fixed_6", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator located in my place?", + "answer": "In front of you to the left.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在我的左前方。", + "evidence": [ + { + "description": "I am facing the door.", + "timestamps": { + "start": "00:03:28", + "end": "00:03:29" + } + }, + { + "description": "The refrigerator is to the left of the door.", + "timestamps": { + "start": "00:03:28", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind you to the left.", + "B": "In front of you to the right.", + "C": "In front of you to the left.", + "D": "Directly behind you, slightly left of center, near the dining table area.", + "E": "In front of you on the right side, positioned near the counter and cabinets." + }, + "answer_label": "C", + "answer_text": "In front of you to the left.", + "q_id": "egolife_A1_D6_024-mcq_fixed_6", + "correct_answer": "In front of you to the left.", + "original_options": { + "A": "Behind you to the left.", + "B": "In front of you to the right.", + "C": "In front of you to the left.", + "D": "Directly behind you, slightly left of center, near the dining table area.", + "E": "In front of you on the right side, positioned near the counter and cabinets." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024-mcq_fixed_7", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator located in my place?", + "answer": "Behind you.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在我的后方。", + "evidence": [ + { + "description": "I turned around to face the large screen.", + "timestamps": { + "start": "00:04:12", + "end": "00:04:13" + } + } + ], + "question_timestamps": "00:04:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "Behind you.", + "D": "Ahead of you, slightly to the left, next to the counter with small items on it.", + "E": "To your right side, near the kitchen counter area where you prepare food." + }, + "answer_label": "C", + "answer_text": "Behind you.", + "q_id": "egolife_A1_D6_024-mcq_fixed_7", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "Behind you.", + "D": "Ahead of you, slightly to the left, next to the counter with small items on it.", + "E": "To your right side, near the kitchen counter area where you prepare food." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_3_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the fridge relative to me?", + "answer": "To your back-left.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在我的左后方。", + "evidence": [ + { + "description": "The refrigerator is behind the door of the room.", + "timestamps": { + "start": "00:04:12", + "end": "00:04:18" + } + } + ], + "question_timestamps": "00:04:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-left.", + "B": "To your back-right.", + "C": "To your front-left.", + "D": "On your left-hand side, roughly beside you rather than diagonally behind you.", + "E": "In front of you on the right, so you would face it without turning around." + }, + "answer_label": "A", + "answer_text": "To your back-left.", + "q_id": "egolife_A1_D6_024_3_T2", + "correct_answer": "To your back-left.", + "original_options": { + "A": "To your back-left.", + "B": "To your back-right.", + "C": "To your front-left.", + "D": "On your left-hand side, roughly beside you rather than diagonally behind you.", + "E": "In front of you on the right, so you would face it without turning around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_3_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the fridge relative to me?", + "answer": "To your front-right.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在你的右前方。", + "evidence": [ + { + "description": "The refrigerator is behind the door of the room.", + "timestamps": { + "start": "00:03:13", + "end": "00:03:34" + } + } + ], + "question_timestamps": "00:05:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-left.", + "B": "To your front-right.", + "C": "To your back-right.", + "D": "To your back-left, meaning it is behind you and off to the left side.", + "E": "Straight ahead of you, slightly left of center, not on the right side." + }, + "answer_label": "B", + "answer_text": "To your front-right.", + "q_id": "egolife_A1_D6_024_3_T1", + "correct_answer": "To your front-right.", + "original_options": { + "A": "To your front-left.", + "B": "To your front-right.", + "C": "To your back-right.", + "D": "To your back-left, meaning it is behind you and off to the left side.", + "E": "Straight ahead of you, slightly left of center, not on the right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_2_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can relative to me?", + "answer": "on your left side.", + "question_chinese": "垃圾桶在我的什么方位？", + "answer_chinese": "在我的左侧。", + "evidence": [ + { + "description": "When I face the balcony, the trash can is directly in front of me.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:26" + } + }, + { + "description": "I turned 90 degrees.", + "timestamps": { + "start": "00:03:26", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:03:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "on your left side.", + "B": "On your right side.", + "C": "In front of you.", + "D": "Behind you, a bit to the right, so you would need to turn to reach it.", + "E": "In front of you to the left, slightly ahead rather than directly beside you." + }, + "answer_label": "A", + "answer_text": "on your left side.", + "q_id": "egolife_A1_D6_024_2_T1", + "correct_answer": "on your left side.", + "original_options": { + "A": "on your left side.", + "B": "On your right side.", + "C": "In front of you.", + "D": "Behind you, a bit to the right, so you would need to turn to reach it.", + "E": "In front of you to the left, slightly ahead rather than directly beside you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_2_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can relative to me?", + "answer": "on your right side.", + "question_chinese": "垃圾桶在我的什么方位？", + "answer_chinese": "在我的右侧。", + "evidence": [ + { + "description": "When I face the balcony, the trash can is directly in front of me.", + "timestamps": { + "start": "00:03:25", + "end": "00:03:26" + } + }, + { + "description": "I turned 90 degrees.", + "timestamps": { + "start": "00:03:26", + "end": "00:03:27" + } + } + ], + "question_timestamps": "00:04:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your left side.", + "B": "on your right side.", + "C": "Behind you.", + "D": "In front of you, slightly to the left as you face forward.", + "E": "Behind you and a bit to your left, not directly beside you." + }, + "answer_label": "B", + "answer_text": "on your right side.", + "q_id": "egolife_A1_D6_024_2_T2", + "correct_answer": "on your right side.", + "original_options": { + "A": "On your left side.", + "B": "on your right side.", + "C": "Behind you.", + "D": "In front of you, slightly to the left as you face forward.", + "E": "Behind you and a bit to your left, not directly beside you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_4_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take to get to the doorway?", + "answer": "I turned left, walked straight, then turned right and walked straight.", + "question_chinese": "我走到门口的路径？", + "answer_chinese": "我向左转身，直走，然后右转直走。", + "evidence": [ + { + "description": "At first, I was standing in front of the large screen.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:26" + } + } + ], + "question_timestamps": "00:04:57", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "I turned right first.", + "B": "I turned left twice.", + "C": "I turned left, walked straight, then turned right and walked straight.", + "D": "I turned left, walked straight, then turned left and walked straight again.", + "E": "I turned right, walked straight, then turned right and walked straight to the doorway." + }, + "answer_label": "C", + "answer_text": "I turned left, walked straight, then turned right and walked straight.", + "q_id": "egolife_A1_D6_024_4_T1", + "correct_answer": "I turned left, walked straight, then turned right and walked straight.", + "original_options": { + "A": "I turned right first.", + "B": "I turned left twice.", + "C": "I turned left, walked straight, then turned right and walked straight.", + "D": "I turned left, walked straight, then turned left and walked straight again.", + "E": "I turned right, walked straight, then turned right and walked straight to the doorway." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_4_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I take to get to the doorway?", + "answer": "I turned left, walked straight, paused there, then turned right and walked straight.", + "question_chinese": "我走到门口的路径？", + "answer_chinese": "我向左转身，直走，停留，然后右转直走。", + "evidence": [ + { + "description": "My movement trajectory.", + "timestamps": { + "start": "00:04:05", + "end": "00:04:24" + } + } + ], + "question_timestamps": "00:05:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "I didn’t pause there.", + "B": "I turned right first.", + "C": "I turned left, walked straight, then turned right and walked straight without stopping.", + "D": "I turned left, walked straight, paused there, then turned right and walked straight.", + "E": "I turned left, walked straight, paused there, then turned left and walked straight to the doorway." + }, + "answer_label": "D", + "answer_text": "I turned left, walked straight, paused there, then turned right and walked straight.", + "q_id": "egolife_A1_D6_024_4_T2", + "correct_answer": "I turned left, walked straight, paused there, then turned right and walked straight.", + "original_options": { + "A": "I didn’t pause there.", + "B": "I turned right first.", + "C": "I turned left, walked straight, then turned right and walked straight without stopping.", + "D": "I turned left, walked straight, paused there, then turned right and walked straight.", + "E": "I turned left, walked straight, paused there, then turned left and walked straight to the doorway." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_5_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I walk along?", + "answer": "You walked from the doorway to where you placed the stand, and then went back to the doorway.", + "question_chinese": "我走动的路径？", + "answer_chinese": "你从门口位置，走到放支架的位置，又回到了门口。", + "evidence": [ + { + "description": "I started by walking from the door to the corner where the laptop is placed.", + "timestamps": { + "start": "00:04:48", + "end": "00:05:00" + } + }, + { + "description": "I walked from the corner to the door.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:05" + } + } + ], + "question_timestamps": "00:05:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You kept spinning around.", + "B": "You walked straight to the doorway.", + "C": "You turned left, walked straight, then turned right and continued forward past the stand location.", + "D": "You walked from the doorway to where you placed the stand, and then went back to the doorway.", + "E": "You went from the doorway to the stand location, then continued past it instead of returning." + }, + "answer_label": "D", + "answer_text": "You walked from the doorway to where you placed the stand, and then went back to the doorway.", + "q_id": "egolife_A1_D6_024_5_T1", + "correct_answer": "You walked from the doorway to where you placed the stand, and then went back to the doorway.", + "original_options": { + "A": "You kept spinning around.", + "B": "You walked straight to the doorway.", + "C": "You turned left, walked straight, then turned right and continued forward past the stand location.", + "D": "You walked from the doorway to where you placed the stand, and then went back to the doorway.", + "E": "You went from the doorway to the stand location, then continued past it instead of returning." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_5_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What path did I walk along?", + "answer": "You kept spinning around in place.", + "question_chinese": "我走动的路径？", + "answer_chinese": "你一直在原地转圈。", + "evidence": [ + { + "description": "The viewpoint kept spinning in circles in place.", + "timestamps": { + "start": "00:06:07", + "end": "00:06:41" + } + } + ], + "question_timestamps": "00:06:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked to the doorway.", + "B": "You walked to the stand.", + "C": "You turned left, walked straight, then turned right and walked straight to the doorway.", + "D": "You walked from the doorway to where you placed the stand, then returned to the doorway.", + "E": "You kept spinning around in place." + }, + "answer_label": "E", + "answer_text": "You kept spinning around in place.", + "q_id": "egolife_A1_D6_024_5_T2", + "correct_answer": "You kept spinning around in place.", + "original_options": { + "A": "You walked to the doorway.", + "B": "You walked to the stand.", + "C": "You turned left, walked straight, then turned right and walked straight to the doorway.", + "D": "You walked from the doorway to where you placed the stand, then returned to the doorway.", + "E": "You kept spinning around in place." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At this moment and in this place, how many categories of items can I see?", + "answer": "Four categories.", + "question_chinese": "此时此地，我目之所及共有几类物品？", + "answer_chinese": "四类。", + "evidence": [ + { + "description": "There are electrical appliances, such as air conditioners, lighting panels, and so on.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "Musical instruments, such as electronic keyboards.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "There are pieces of furniture, such as tables, chairs, and sofas.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "There are beverages, with alcohol and water placed on the table.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:26", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Three categories.", + "B": "Two categories.", + "C": "Four categories.", + "D": "Seven categories of items are visible in this spot right now here.", + "E": "Only one category of items can be seen at this moment and location." + }, + "answer_label": "C", + "answer_text": "Four categories.", + "q_id": "egolife_A1_D6_024_1", + "correct_answer": "Four categories.", + "original_options": { + "A": "Three categories.", + "B": "Two categories.", + "C": "Four categories.", + "D": "Seven categories of items are visible in this spot right now here.", + "E": "Only one category of items can be seen at this moment and location." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "At this moment and in this place, how many categories of items can I see?", + "answer": "Three categories.", + "question_chinese": "此时此地，我目之所及共有几类物品？", + "answer_chinese": "三类。", + "evidence": [ + { + "description": "There is food on the storage counter.", + "timestamps": { + "start": "00:03:13", + "end": "00:03:14" + } + }, + { + "description": "There are furnishings, such as trash cans.", + "timestamps": { + "start": "00:03:14", + "end": "00:03:15" + } + }, + { + "description": "There are electrical appliances, such as microwave ovens.", + "timestamps": { + "start": "00:03:13", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:03:26", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Two categories.", + "B": "Three categories.", + "C": "Four categories.", + "D": "Six categories of items are visible here at this moment right now.", + "E": "Nine categories of items can be seen in this place right now." + }, + "answer_label": "B", + "answer_text": "Three categories.", + "q_id": "egolife_A1_D6_024_2", + "correct_answer": "Three categories.", + "original_options": { + "A": "Two categories.", + "B": "Three categories.", + "C": "Four categories.", + "D": "Six categories of items are visible here at this moment right now.", + "E": "Nine categories of items can be seen in this place right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_7_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the air conditioner relative to me?", + "answer": "Above you to the left.", + "question_chinese": "空调在我的什么方位？", + "answer_chinese": "在你左上方。", + "evidence": [ + { + "description": "The air conditioner is mounted on the wall on the left side of the room.", + "timestamps": { + "start": "00:04:52", + "end": "00:05:11" + } + } + ], + "question_timestamps": "00:05:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above you to the right.", + "B": "Above you to the left.", + "C": "Behind you.", + "D": "Below you to the left, closer to your lower field of view than overhead.", + "E": "In front of you to the left, rather than up high above your head." + }, + "answer_label": "B", + "answer_text": "Above you to the left.", + "q_id": "egolife_A1_D6_024_7_T2", + "correct_answer": "Above you to the left.", + "original_options": { + "A": "Above you to the right.", + "B": "Above you to the left.", + "C": "Behind you.", + "D": "Below you to the left, closer to your lower field of view than overhead.", + "E": "In front of you to the left, rather than up high above your head." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024_7_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the air conditioner relative to me?", + "answer": "Above you to the right.", + "question_chinese": "空调在我的什么方位？", + "answer_chinese": "在你的右上方。", + "evidence": [ + { + "description": "The air conditioner is mounted on the wall on the left side of the room.", + "timestamps": { + "start": "00:04:52", + "end": "00:05:11" + } + } + ], + "question_timestamps": "00:08:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above you to the right.", + "B": "Above you to the left.", + "C": "Behind you.", + "D": "Below you to the right, closer to your shoulder than your head.", + "E": "In front of you to the left, roughly at eye level when you look up." + }, + "answer_label": "A", + "answer_text": "Above you to the right.", + "q_id": "egolife_A1_D6_024_7_T1", + "correct_answer": "Above you to the right.", + "original_options": { + "A": "Above you to the right.", + "B": "Above you to the left.", + "C": "Behind you.", + "D": "Below you to the right, closer to your shoulder than your head.", + "E": "In front of you to the left, roughly at eye level when you look up." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 2, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024-mcq_fixed_4", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the drink or the notebook?", + "answer": "Beverage.", + "question_chinese": "饮料和笔记本哪个距我更近？", + "answer_chinese": "饮料。", + "evidence": [ + { + "description": "The drink is in front of the notebook.", + "timestamps": { + "start": "00:07:27", + "end": "00:07:28" + } + }, + { + "description": "The drinks and notebook are right in front of me.", + "timestamps": { + "start": "00:07:26", + "end": "00:07:27" + } + } + ], + "question_timestamps": "00:07:28", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Beverage.", + "B": "Notebook." + }, + "answer_label": "A", + "answer_text": "Beverage.", + "q_id": "egolife_A1_D6_024-mcq_fixed_4", + "correct_answer": "Beverage.", + "original_options": { + "A": "Beverage.", + "B": "Notebook." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_024", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_024-mcq_fixed_5", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the drink or the notebook?", + "answer": "Notebook.", + "question_chinese": "饮料和笔记本哪个距我更近？", + "answer_chinese": "笔记本。", + "evidence": [ + { + "description": "The drinks and notebook are right in front of me.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:29" + } + }, + { + "description": "The notebook is in front of the drink.", + "timestamps": { + "start": "00:09:28", + "end": "00:09:29" + } + } + ], + "question_timestamps": "00:09:30", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Drink.", + "B": "Notebook." + }, + "answer_label": "B", + "answer_text": "Notebook.", + "q_id": "egolife_A1_D6_024-mcq_fixed_5", + "correct_answer": "Notebook.", + "original_options": { + "A": "Drink.", + "B": "Notebook." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D6_025-mcq.json b/QAs/egolife_A1_D6_025-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..bbccf9347e2f9e1765e8d51d5731edc45af82ed1 --- /dev/null +++ b/QAs/egolife_A1_D6_025-mcq.json @@ -0,0 +1,1113 @@ +[ + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_3_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the fridge relative to me?", + "answer": "It is to your left.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在你的左面。", + "evidence": [ + { + "description": "The fridge is on your left.", + "timestamps": { + "start": "00:00:53", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:00:56", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to your left.", + "B": "It is to your right.", + "C": "It is behind you.", + "D": "It is in front of you, straight ahead at about chest height.", + "E": "It is above you, positioned higher than your head when you look up." + }, + "answer_label": "A", + "answer_text": "It is to your left.", + "q_id": "egolife_A1_D6_025_3_T2", + "correct_answer": "It is to your left.", + "original_options": { + "A": "It is to your left.", + "B": "It is to your right.", + "C": "It is behind you.", + "D": "It is in front of you, straight ahead at about chest height.", + "E": "It is above you, positioned higher than your head when you look up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_5", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator located?", + "answer": "To my left.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在我的左面。", + "evidence": [ + { + "description": "When I walked towards the balcony, the refrigerator was directly in front of me.", + "timestamps": { + "start": "00:00:53", + "end": "00:00:54" + } + }, + { + "description": "I turned 90 degrees.", + "timestamps": { + "start": "00:00:53", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:00:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To my right.", + "B": "To my left.", + "C": "Behind me.", + "D": "In front of me, slightly off-center as I face the room from where I'm standing.", + "E": "Above me, mounted higher than my head near the ceiling line in this room." + }, + "answer_label": "B", + "answer_text": "To my left.", + "q_id": "egolife_A1_D6_025_5", + "correct_answer": "To my left.", + "original_options": { + "A": "To my right.", + "B": "To my left.", + "C": "Behind me.", + "D": "In front of me, slightly off-center as I face the room from where I'm standing.", + "E": "Above me, mounted higher than my head near the ceiling line in this room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_3_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the fridge relative to me?", + "answer": "It is behind you.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在你的后面。", + "evidence": [ + { + "description": "The fridge is behind pf you.", + "timestamps": { + "start": "00:00:59", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "It is behind you.", + "D": "It is to your right, off to the side rather than directly behind you.", + "E": "It is above you, positioned higher than your head when you look upward." + }, + "answer_label": "C", + "answer_text": "It is behind you.", + "q_id": "egolife_A1_D6_025_3_T1", + "correct_answer": "It is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "It is behind you.", + "D": "It is to your right, off to the side rather than directly behind you.", + "E": "It is above you, positioned higher than your head when you look upward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025-mcq_fixed_3", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the refrigerator located in my place?", + "answer": "In front of you to the left.", + "question_chinese": "冰箱在我的什么位置？", + "answer_chinese": "在我的前方。", + "evidence": [ + { + "description": "I stood near the table, facing the door.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:56" + } + }, + { + "description": "The refrigerator is to the left of the door.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:56" + } + } + ], + "question_timestamps": "00:01:57", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Behind you to the left.", + "B": "In front of you to the left.", + "C": "In front of you to the right.", + "D": "Directly in front of you, not offset left or right as you face it.", + "E": "Slightly behind you on the right, close enough that you’d need to turn around." + }, + "answer_label": "B", + "answer_text": "In front of you to the left.", + "q_id": "egolife_A1_D6_025-mcq_fixed_3", + "correct_answer": "In front of you to the left.", + "original_options": { + "A": "Behind you to the left.", + "B": "In front of you to the left.", + "C": "In front of you to the right.", + "D": "Directly in front of you, not offset left or right as you face it.", + "E": "Slightly behind you on the right, close enough that you’d need to turn around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_5_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Am I holding the phone in my hand?", + "answer": "You are holding the phone in your hand; it is in front of you.", + "question_chinese": "我手上拿着手机吗？", + "answer_chinese": "你手上拿着手机，它在你前面。", + "evidence": [ + { + "description": "You are holding a mobile phone in your hand.", + "timestamps": { + "start": "00:06:45", + "end": "00:06:55" + } + } + ], + "question_timestamps": "00:07:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are holding the phone in your hand; it is in front of you.", + "B": "You are not holding the phone.", + "C": "You are holding it behind you.", + "D": "You are not holding the phone in your hand; it sits down-right of you, out of reach.", + "E": "You are holding the phone in your hand, but it is down-left of you rather than in front." + }, + "answer_label": "A", + "answer_text": "You are holding the phone in your hand; it is in front of you.", + "q_id": "egolife_A1_D6_025_5_T2", + "correct_answer": "You are holding the phone in your hand; it is in front of you.", + "original_options": { + "A": "You are holding the phone in your hand; it is in front of you.", + "B": "You are not holding the phone.", + "C": "You are holding it behind you.", + "D": "You are not holding the phone in your hand; it sits down-right of you, out of reach.", + "E": "You are holding the phone in your hand, but it is down-left of you rather than in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_5_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Am I holding the phone in my hand?", + "answer": "You are not holding the phone in your hand; it is down-right of you.", + "question_chinese": "我手上拿着手机吗？", + "answer_chinese": "你手上没有拿着手机,它在你右下方。", + "evidence": [ + { + "description": "You are not holding a mobile phone in your hand.", + "timestamps": { + "start": "00:07:04", + "end": "00:07:05" + } + } + ], + "question_timestamps": "00:07:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are holding the phone.", + "B": "The phone is up-left of you.", + "C": "You are holding the phone in your hand, and it is directly in front of you.", + "D": "You are not holding the phone in your hand; it is down-left of you instead.", + "E": "You are not holding the phone in your hand; it is down-right of you." + }, + "answer_label": "E", + "answer_text": "You are not holding the phone in your hand; it is down-right of you.", + "q_id": "egolife_A1_D6_025_5_T1", + "correct_answer": "You are not holding the phone in your hand; it is down-right of you.", + "original_options": { + "A": "You are holding the phone.", + "B": "The phone is up-left of you.", + "C": "You are holding the phone in your hand, and it is directly in front of you.", + "D": "You are not holding the phone in your hand; it is down-left of you instead.", + "E": "You are not holding the phone in your hand; it is down-right of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_4_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I reach the computer?", + "answer": "You can reach the computer; the computer is down to your right.", + "question_chinese": "我能接触到电脑吗？", + "answer_chinese": "你能接触到电脑，电脑在你右下方。", + "evidence": [ + { + "description": "You can touch the computer.", + "timestamps": { + "start": "00:01:10", + "end": "00:01:47" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You cannot reach the computer.", + "B": "You can reach it on your left.", + "C": "You cannot reach the computer; it is positioned behind you and out of your reach.", + "D": "You can reach the computer; the computer is down to your right.", + "E": "You can reach the computer, but it is directly in front of you rather than down-right." + }, + "answer_label": "D", + "answer_text": "You can reach the computer; the computer is down to your right.", + "q_id": "egolife_A1_D6_025_4_T1", + "correct_answer": "You can reach the computer; the computer is down to your right.", + "original_options": { + "A": "You cannot reach the computer.", + "B": "You can reach it on your left.", + "C": "You cannot reach the computer; it is positioned behind you and out of your reach.", + "D": "You can reach the computer; the computer is down to your right.", + "E": "You can reach the computer, but it is directly in front of you rather than down-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_4_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I reach the computer?", + "answer": "You cannot reach the computer; the computer is behind you.", + "question_chinese": "我能接触到电脑吗？", + "answer_chinese": "你不能接触到电脑，电脑在你后面。", + "evidence": [ + { + "description": "You can not touch the computer.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:03" + } + } + ], + "question_timestamps": "00:07:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can reach the computer.", + "B": "It is in front of you.", + "C": "You cannot reach the computer; it is down to your right, not behind you.", + "D": "You can reach the computer because it is in front of you, not behind you.", + "E": "You cannot reach the computer; the computer is behind you." + }, + "answer_label": "E", + "answer_text": "You cannot reach the computer; the computer is behind you.", + "q_id": "egolife_A1_D6_025_4_T2", + "correct_answer": "You cannot reach the computer; the computer is behind you.", + "original_options": { + "A": "You can reach the computer.", + "B": "It is in front of you.", + "C": "You cannot reach the computer; it is down to your right, not behind you.", + "D": "You can reach the computer because it is in front of you, not behind you.", + "E": "You cannot reach the computer; the computer is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_7", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path I take?", + "answer": "I walked from beside the table to near the balcony and then returned to my original spot.", + "question_chinese": "我走动的路径？", + "answer_chinese": "我从桌子旁边走到了阳台附近，然后又回到原处。", + "evidence": [ + { + "description": "After that, I stood by the balcony for a while.", + "timestamps": { + "start": "00:00:54", + "end": "00:00:56" + } + }, + { + "description": "I started by standing in front of the large screen, holding a notebook in my hand.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:43" + } + }, + { + "description": "In the end, I returned to my original place.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:03" + } + } + ], + "question_timestamps": "00:01:07", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "I walked to the balcony only.", + "B": "I walked from the balcony to the table.", + "C": "I walked from beside the table to near the balcony and then returned to my original spot.", + "D": "I walked from beside the table to near the balcony, but I did not return to my starting spot.", + "E": "I walked from beside the table toward the balcony, then continued past it instead of returning." + }, + "answer_label": "C", + "answer_text": "I walked from beside the table to near the balcony and then returned to my original spot.", + "q_id": "egolife_A1_D6_025_7", + "correct_answer": "I walked from beside the table to near the balcony and then returned to my original spot.", + "original_options": { + "A": "I walked to the balcony only.", + "B": "I walked from the balcony to the table.", + "C": "I walked from beside the table to near the balcony and then returned to my original spot.", + "D": "I walked from beside the table to near the balcony, but I did not return to my starting spot.", + "E": "I walked from beside the table toward the balcony, then continued past it instead of returning." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_8", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path I take when I move?", + "answer": "I walked from near the large screen to the door.", + "question_chinese": "我走动的路径？", + "answer_chinese": "我从大屏幕附近走到了门口。", + "evidence": [ + { + "description": "Originally, I stood near the large screen.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:55" + } + }, + { + "description": "Finally, I walked to the door.", + "timestamps": { + "start": "00:07:04", + "end": "00:07:05" + } + } + ], + "question_timestamps": "00:07:09", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I walked from the door to the screen.", + "B": "I walked from the screen to the hallway.", + "C": "I walked from near the large screen to the door.", + "D": "I walked from near the large screen to the center of the room, not to the door.", + "E": "I walked from near the large screen toward the door but stopped short before reaching it." + }, + "answer_label": "C", + "answer_text": "I walked from near the large screen to the door.", + "q_id": "egolife_A1_D6_025_8", + "correct_answer": "I walked from near the large screen to the door.", + "original_options": { + "A": "I walked from the door to the screen.", + "B": "I walked from the screen to the hallway.", + "C": "I walked from near the large screen to the door.", + "D": "I walked from near the large screen to the center of the room, not to the door.", + "E": "I walked from near the large screen toward the door but stopped short before reaching it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_2_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which direction is the projector relative to me?", + "answer": "The projector is to your right.", + "question_chinese": "投影仪在我的哪个方向？", + "answer_chinese": "投影仪在你右边。", + "evidence": [ + { + "description": "The projector is on my right.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:40" + } + } + ], + "question_timestamps": "00:00:43", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The projector is to your left.", + "B": "The projector is to your right." + }, + "answer_label": "B", + "answer_text": "The projector is to your right.", + "q_id": "egolife_A1_D6_025_2_T2", + "correct_answer": "The projector is to your right.", + "original_options": { + "A": "The projector is to your left.", + "B": "The projector is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_2_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which direction is the projector relative to me?", + "answer": "The projector is in front of you.", + "question_chinese": "投影仪在我的哪个方向？", + "answer_chinese": "投影仪在你前面。", + "evidence": [ + { + "description": "The projector is in front of you.", + "timestamps": { + "start": "00:00:55", + "end": "00:00:59" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The projector is in front of you.", + "B": "The projector is behind you." + }, + "answer_label": "A", + "answer_text": "The projector is in front of you.", + "q_id": "egolife_A1_D6_025_2_T1", + "correct_answer": "The projector is in front of you.", + "original_options": { + "A": "The projector is in front of you.", + "B": "The projector is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_3", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the stand or the mineral water?", + "answer": "Move the bracket closer.", + "question_chinese": "支架与矿泉水那个离我更近一些？", + "answer_chinese": "支架更近一些。", + "evidence": [ + { + "description": "The bracket and the mineral water are directly in front of me.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:06" + } + }, + { + "description": "The support is about a step away from me.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:06" + } + }, + { + "description": "The mineral water is about 1 meter away from me.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:10:17", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Move the bracket closer.", + "B": "Move the mineral water closer." + }, + "answer_label": "A", + "answer_text": "Move the bracket closer.", + "q_id": "egolife_A1_D6_025_3", + "correct_answer": "Move the bracket closer.", + "original_options": { + "A": "Move the bracket closer.", + "B": "Move the mineral water closer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_4", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the stand or the bottled water?", + "answer": "Bracket.", + "question_chinese": "支架与矿泉水那个离我更近一些？", + "answer_chinese": "支架。", + "evidence": [ + { + "description": "The bracket and the mineral water are directly in front of me.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:06" + } + }, + { + "description": "The support is about a step away from me.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:06" + } + }, + { + "description": "The mineral water is about 1 meter away from me.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:06" + } + } + ], + "question_timestamps": "00:11:33", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Mineral water.", + "B": "Bracket." + }, + "answer_label": "B", + "answer_text": "Bracket.", + "q_id": "egolife_A1_D6_025_4", + "correct_answer": "Bracket.", + "original_options": { + "A": "Mineral water.", + "B": "Bracket." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_1_T2", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, is my friend who is singing sitting or standing?", + "answer": "Right now, the friend who is singing is sitting, and he is in front of you.", + "question_chinese": "此时，我正在唱歌的朋友是坐着还是站着？", + "answer_chinese": "此时，你正在唱歌的朋友是坐着，他在你前面。", + "evidence": [ + { + "description": "Your friend who is singing is sitting.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:10:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, the friend who is singing is sitting, and he is in front of you.", + "B": "Right now, the friend who is singing is standing, and he is in front of you." + }, + "answer_label": "A", + "answer_text": "Right now, the friend who is singing is sitting, and he is in front of you.", + "q_id": "egolife_A1_D6_025_1_T2", + "correct_answer": "Right now, the friend who is singing is sitting, and he is in front of you.", + "original_options": { + "A": "Right now, the friend who is singing is sitting, and he is in front of you.", + "B": "Right now, the friend who is singing is standing, and he is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_1_T1", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Right now, is my friend who is singing sitting or standing?", + "answer": "Right now, the friend who is singing is standing, and he is in front of you.", + "question_chinese": "此时，我正在唱歌的朋友是坐着还是站着？", + "answer_chinese": "此时，你正在唱歌的朋友是站着，他在你前面。", + "evidence": [ + { + "description": "Your friend who is singing is standing.", + "timestamps": { + "start": "00:10:18", + "end": "00:10:35" + } + } + ], + "question_timestamps": "00:10:38", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, the friend who is singing is sitting, and he is in front of you.", + "B": "Right now, the friend who is singing is standing, and he is in front of you." + }, + "answer_label": "B", + "answer_text": "Right now, the friend who is singing is standing, and he is in front of you.", + "q_id": "egolife_A1_D6_025_1_T1", + "correct_answer": "Right now, the friend who is singing is standing, and he is in front of you.", + "original_options": { + "A": "Right now, the friend who is singing is sitting, and he is in front of you.", + "B": "Right now, the friend who is singing is standing, and he is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_9", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many egg tarts do I see left on the plate?", + "answer": "Seven.", + "question_chinese": "此时，我看见还有几个蛋挞在餐盘上？", + "answer_chinese": "七个。", + "evidence": [ + { + "description": "I saw a boy carrying a tray, walking towards the crowd, with seven egg tarts on the tray.", + "timestamps": { + "start": "00:07:38", + "end": "00:07:39" + } + } + ], + "question_timestamps": "00:07:50", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Seven.", + "B": "Six.", + "C": "Eight.", + "D": "You can see six egg tarts left on the plate at this moment.", + "E": "You can see eight egg tarts left on the plate at this moment." + }, + "answer_label": "A", + "answer_text": "Seven.", + "q_id": "egolife_A1_D6_025_9", + "correct_answer": "Seven.", + "original_options": { + "A": "Seven.", + "B": "Six.", + "C": "Eight.", + "D": "You can see six egg tarts left on the plate at this moment.", + "E": "You can see eight egg tarts left on the plate at this moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_10", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "At this moment, how many egg tarts do I see left on the plate?", + "answer": "Five.", + "question_chinese": "此时，我看见还有几个蛋挞在餐盘上？", + "answer_chinese": "五个。", + "evidence": [ + { + "description": "I saw two people take the egg tarts one after another.", + "timestamps": { + "start": "00:07:40", + "end": "00:08:10" + } + } + ], + "question_timestamps": "00:08:21", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "At this moment, you can see four egg tarts left on the plate.", + "D": "At this moment, you can see six egg tarts left on the plate.", + "E": "Five." + }, + "answer_label": "E", + "answer_text": "Five.", + "q_id": "egolife_A1_D6_025_10", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "At this moment, you can see four egg tarts left on the plate.", + "D": "At this moment, you can see six egg tarts left on the plate.", + "E": "Five." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_8_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the air conditioner relative to me?", + "answer": "The air conditioner is in front of you.", + "question_chinese": "空调在我的什么位置？", + "answer_chinese": "空调在你前面。", + "evidence": [ + { + "description": "The air conditioner is in front of you.", + "timestamps": { + "start": "00:01:55", + "end": "00:01:59" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The air conditioner is in front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "The air conditioner is to your right relative to you right now.", + "E": "The air conditioner is above you relative to your position right now." + }, + "answer_label": "A", + "answer_text": "The air conditioner is in front of you.", + "q_id": "egolife_A1_D6_025_8_T2", + "correct_answer": "The air conditioner is in front of you.", + "original_options": { + "A": "The air conditioner is in front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "The air conditioner is to your right relative to you right now.", + "E": "The air conditioner is above you relative to your position right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_025", + "video_length": "00:11:23", + "qid": "egolife_A1_D6_025_8_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the air conditioner relative to me?", + "answer": "The air conditioner is to your right.", + "question_chinese": "空调在我的什么位置？", + "answer_chinese": "空调在你右边。", + "evidence": [ + { + "description": "The air conditional is on your right.", + "timestamps": { + "start": "00:07:05", + "end": "00:07:10" + } + } + ], + "question_timestamps": "00:07:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "The air conditioner is to your right.", + "C": "To your left.", + "D": "The air conditioner is behind you relative to where you are standing.", + "E": "The air conditioner is above you relative to your current position right now." + }, + "answer_label": "B", + "answer_text": "The air conditioner is to your right.", + "q_id": "egolife_A1_D6_025_8_T1", + "correct_answer": "The air conditioner is to your right.", + "original_options": { + "A": "In front of you.", + "B": "The air conditioner is to your right.", + "C": "To your left.", + "D": "The air conditioner is behind you relative to where you are standing.", + "E": "The air conditioner is above you relative to your current position right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D6_026-mcq.json b/QAs/egolife_A1_D6_026-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4e94181468747fa4dd22786aeb677672f1406ab4 --- /dev/null +++ b/QAs/egolife_A1_D6_026-mcq.json @@ -0,0 +1,697 @@ +[ + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026_1_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the bottled water on the floor, or the power bank on the sofa?", + "answer": "The mineral water is closer to you.", + "question_chinese": "地上的矿泉水与沙发上的充电宝哪个距我比较近？", + "answer_chinese": "矿泉水离你近。", + "evidence": [ + { + "description": "Mineral water is directly in front of me, about 1 meter away.The power bank is to my left, about 1.5 meters away from me.", + "timestamps": { + "start": "00:00:28", + "end": "00:00:29" + } + } + ], + "question_timestamps": "00:00:40", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The power bank is closer to you.", + "B": "The mineral water is closer to you." + }, + "answer_label": "B", + "answer_text": "The mineral water is closer to you.", + "q_id": "egolife_A1_D6_026_1_T1", + "correct_answer": "The mineral water is closer to you.", + "original_options": { + "A": "The power bank is closer to you.", + "B": "The mineral water is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026_1_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the bottled water on the floor, or the power bank on the sofa?", + "answer": "The power bank is closer to you.", + "question_chinese": "地上的矿泉水与沙发上的充电宝哪个距我比较近？", + "answer_chinese": "充电宝离你近。", + "evidence": [ + { + "description": "The power bank is to my left hand. I can reach it as long as I lower my hand.", + "timestamps": { + "start": "00:06:00", + "end": "00:06:01" + } + }, + { + "description": "Mineral water is in front of me to the right, about a step away from me.", + "timestamps": { + "start": "00:06:01", + "end": "00:06:05" + } + } + ], + "question_timestamps": "00:06:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The bottled water on the floor is closer to you.", + "B": "The power bank is closer to you." + }, + "answer_label": "B", + "answer_text": "The power bank is closer to you.", + "q_id": "egolife_A1_D6_026_1_T2", + "correct_answer": "The power bank is closer to you.", + "original_options": { + "A": "The bottled water on the floor is closer to you.", + "B": "The power bank is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026-mcq_fixed_4", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can in relation to me?", + "answer": "Behind you.", + "question_chinese": "垃圾桶在我的什么方位？", + "answer_chinese": "在我的后面。", + "evidence": [ + { + "description": "When I walked into the room, the trash can was directly in front of me.", + "timestamps": { + "start": "00:10:26", + "end": "00:10:27" + } + }, + { + "description": "I turned around 180 degrees.", + "timestamps": { + "start": "00:10:28", + "end": "00:10:29" + } + } + ], + "question_timestamps": "00:10:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "Directly to your right, close to where you are standing right now.", + "E": "In front of you, a little past the coffee table area from where you are." + }, + "answer_label": "B", + "answer_text": "Behind you.", + "q_id": "egolife_A1_D6_026-mcq_fixed_4", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "Directly to your right, close to where you are standing right now.", + "E": "In front of you, a little past the coffee table area from where you are." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026_5_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can relative to me?", + "answer": "It is behind you.", + "question_chinese": "垃圾桶在我的什么方位？", + "answer_chinese": "在你的后面。", + "evidence": [ + { + "description": "When I walked into the room, the trash can was directly in front of me.", + "timestamps": { + "start": "00:10:26", + "end": "00:10:27" + } + }, + { + "description": "I turned around 180 degrees.", + "timestamps": { + "start": "00:10:28", + "end": "00:10:29" + } + } + ], + "question_timestamps": "00:10:31", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "It is to your right, close beside you as you are standing there.", + "D": "It is above you, positioned higher than your eye level right now.", + "E": "It is behind you." + }, + "answer_label": "E", + "answer_text": "It is behind you.", + "q_id": "egolife_A1_D6_026_5_T1", + "correct_answer": "It is behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is to your left.", + "C": "It is to your right, close beside you as you are standing there.", + "D": "It is above you, positioned higher than your eye level right now.", + "E": "It is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026_5_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can relative to me?", + "answer": "It was in front of you.", + "question_chinese": "垃圾桶在我的什么方位？", + "answer_chinese": "在你的前边。", + "evidence": [ + { + "description": "When I walked into the room, the trash can was directly in front of me.", + "timestamps": { + "start": "00:10:27", + "end": "00:10:28" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was behind you.", + "B": "It was to your left.", + "C": "It was to your right, off to the side as you faced forward.", + "D": "It was in front of you.", + "E": "It was above you, positioned higher up than where you were looking." + }, + "answer_label": "D", + "answer_text": "It was in front of you.", + "q_id": "egolife_A1_D6_026_5_T2", + "correct_answer": "It was in front of you.", + "original_options": { + "A": "It was behind you.", + "B": "It was to your left.", + "C": "It was to your right, off to the side as you faced forward.", + "D": "It was in front of you.", + "E": "It was above you, positioned higher up than where you were looking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 12 + ] + }, + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026_2_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many egg tarts do I see left on the plate?", + "answer": "You saw four.", + "question_chinese": "此时我看见餐盘里还有几个蛋挞？", + "answer_chinese": "你看到四个。", + "evidence": [ + { + "description": "When I saw a person walking towards me holding a egg tart, I saw five egg tarts in the plate.", + "timestamps": { + "start": "00:10:02", + "end": "00:10:03" + } + }, + { + "description": "I took an egg tart from the plate.", + "timestamps": { + "start": "00:10:03", + "end": "00:10:04" + } + } + ], + "question_timestamps": "00:10:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw three.", + "B": "You saw five.", + "C": "You saw four.", + "D": "You saw two egg tarts left on the plate right now at this moment.", + "E": "You saw six egg tarts remaining on the plate right now at this moment." + }, + "answer_label": "C", + "answer_text": "You saw four.", + "q_id": "egolife_A1_D6_026_2_T1", + "correct_answer": "You saw four.", + "original_options": { + "A": "You saw three.", + "B": "You saw five.", + "C": "You saw four.", + "D": "You saw two egg tarts left on the plate right now at this moment.", + "E": "You saw six egg tarts remaining on the plate right now at this moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026_2_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many egg tarts do I see left on the plate?", + "answer": "You saw two.", + "question_chinese": "此时我看见餐盘里还有几个蛋挞？", + "answer_chinese": "你看到两个。", + "evidence": [ + { + "description": "After that, I saw that person give the remaining three people in succession.", + "timestamps": { + "start": "00:10:05", + "end": "00:10:30" + } + } + ], + "question_timestamps": "00:11:48", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one.", + "B": "You saw three.", + "C": "You saw zero egg tarts left on the plate right now at this moment.", + "D": "You saw two.", + "E": "You saw four egg tarts remaining on the plate right now at this moment." + }, + "answer_label": "D", + "answer_text": "You saw two.", + "q_id": "egolife_A1_D6_026_2_T2", + "correct_answer": "You saw two.", + "original_options": { + "A": "You saw one.", + "B": "You saw three.", + "C": "You saw zero egg tarts left on the plate right now at this moment.", + "D": "You saw two.", + "E": "You saw four egg tarts remaining on the plate right now at this moment." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026-mcq_fixed_0", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "At this moment, where is the laptop that is placed on the stairs relative to me?", + "answer": "To the left of your left hand.", + "question_chinese": "此时，放在楼梯上的笔记本在我的什么位置？", + "answer_chinese": "在我的左手边。", + "evidence": [ + { + "description": "When I faced the stairs, I saw the notebook directly in front of me.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:07" + } + }, + { + "description": "I turned 90 degrees.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To the right of your left hand.", + "B": "To the left of your left hand.", + "C": "In front of your left hand.", + "D": "To your right side, close to where your foot touches the lower step.", + "E": "Slightly behind you on the higher steps, near the railing where you last leaned." + }, + "answer_label": "B", + "answer_text": "To the left of your left hand.", + "q_id": "egolife_A1_D6_026-mcq_fixed_0", + "correct_answer": "To the left of your left hand.", + "original_options": { + "A": "To the right of your left hand.", + "B": "To the left of your left hand.", + "C": "In front of your left hand.", + "D": "To your right side, close to where your foot touches the lower step.", + "E": "Slightly behind you on the higher steps, near the railing where you last leaned." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026_3_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the notebook on the stairs relative to me?", + "answer": "On your left-hand side.", + "question_chinese": "此时，放在楼梯上的笔记本在我的什么位置？", + "answer_chinese": "在你的左手边。", + "evidence": [ + { + "description": "When I faced the stairs, I saw the notebook directly in front of me.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:07" + } + }, + { + "description": "I turned 90 degrees.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On your left-hand side.", + "B": "On your right-hand side.", + "C": "In front of you.", + "D": "Behind you, on the stairs directly behind your back right now, not beside you.", + "E": "In front of you on the stairs, slightly closer than your feet right now." + }, + "answer_label": "A", + "answer_text": "On your left-hand side.", + "q_id": "egolife_A1_D6_026_3_T1", + "correct_answer": "On your left-hand side.", + "original_options": { + "A": "On your left-hand side.", + "B": "On your right-hand side.", + "C": "In front of you.", + "D": "Behind you, on the stairs directly behind your back right now, not beside you.", + "E": "In front of you on the stairs, slightly closer than your feet right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026_3_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the notebook on the stairs relative to me?", + "answer": "It was on your right.", + "question_chinese": "此时，放在楼梯上的笔记本在我的什么位置？", + "answer_chinese": "在你的右手边。", + "evidence": [ + { + "description": "When I faced the stairs, I saw the notebook directly in front of me.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:07" + } + }, + { + "description": "I turned 270 degrees.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:05:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "It was to your left.", + "C": "It was behind you on the stairs as you faced upward toward the next step.", + "D": "It was slightly behind you on the stairs, closer to your left side.", + "E": "It was on your right." + }, + "answer_label": "E", + "answer_text": "It was on your right.", + "q_id": "egolife_A1_D6_026_3_T2", + "correct_answer": "It was on your right.", + "original_options": { + "A": "It was in front of you.", + "B": "It was to your left.", + "C": "It was behind you on the stairs as you faced upward toward the next step.", + "D": "It was slightly behind you on the stairs, closer to your left side.", + "E": "It was on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026-mcq_fixed_8", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path of the cardboard box?", + "answer": "A girl was handed from one boy's hand to another and finally placed on the ground.", + "question_chinese": "纸箱的路径是什么？", + "answer_chinese": "一个女生手上移交到了一个男生手上，最后被放在了地上。", + "evidence": [ + { + "description": "The cardboard box was originally lying on the ground next to the girl's feet.", + "timestamps": { + "start": "00:11:42", + "end": "00:11:43" + } + }, + { + "description": "I saw the boy put it on the ground.", + "timestamps": { + "start": "00:11:45", + "end": "00:11:46" + } + }, + { + "description": "The girl handed the cardboard box to the boy.", + "timestamps": { + "start": "00:11:43", + "end": "00:11:44" + } + } + ], + "question_timestamps": "00:11:46", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "It went straight to the ground.", + "B": "It stayed with one boy.", + "C": "A box was passed from one boy to another, then carried away instead of being placed on the ground.", + "D": "A box moved from the ground into a boy's hands, then was handed off once and kept held.", + "E": "A girl was handed from one boy's hand to another and finally placed on the ground." + }, + "answer_label": "E", + "answer_text": "A girl was handed from one boy's hand to another and finally placed on the ground.", + "q_id": "egolife_A1_D6_026-mcq_fixed_8", + "correct_answer": "A girl was handed from one boy's hand to another and finally placed on the ground.", + "original_options": { + "A": "It went straight to the ground.", + "B": "It stayed with one boy.", + "C": "A box was passed from one boy to another, then carried away instead of being placed on the ground.", + "D": "A box moved from the ground into a boy's hands, then was handed off once and kept held.", + "E": "A girl was handed from one boy's hand to another and finally placed on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_026", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_026-mcq_fixed_9", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path of the cardboard box?", + "answer": "Finally, it was placed on the ground.", + "question_chinese": "纸箱的路径是什么？", + "answer_chinese": "最后被放在了地上。", + "evidence": [ + { + "description": "The cardboard box was originally lying on the ground next to the girl's feet.", + "timestamps": { + "start": "00:11:43", + "end": "00:11:44" + } + }, + { + "description": "The girl handed the cardboard box to the boy.", + "timestamps": { + "start": "00:11:44", + "end": "00:11:45" + } + }, + { + "description": "I saw the boy put it on the ground.", + "timestamps": { + "start": "00:11:45", + "end": "00:11:46" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Finally, it was placed on the ground.", + "B": "Finally, it was set on a table.", + "C": "Finally, it was put on a shelf.", + "D": "In the end, it was carried away and kept in someone's hands rather than being placed down.", + "E": "It was moved upstairs and left leaning against a cabinet, not placed on the ground." + }, + "answer_label": "A", + "answer_text": "Finally, it was placed on the ground.", + "q_id": "egolife_A1_D6_026-mcq_fixed_9", + "correct_answer": "Finally, it was placed on the ground.", + "original_options": { + "A": "Finally, it was placed on the ground.", + "B": "Finally, it was set on a table.", + "C": "Finally, it was put on a shelf.", + "D": "In the end, it was carried away and kept in someone's hands rather than being placed down.", + "E": "It was moved upstairs and left leaning against a cabinet, not placed on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D6_027-mcq.json b/QAs/egolife_A1_D6_027-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b745f8ec07a206711411d86ba8ce07c958c7f85a --- /dev/null +++ b/QAs/egolife_A1_D6_027-mcq.json @@ -0,0 +1,679 @@ +[ + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027-mcq_fixed_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the first floor to the second floor?", + "answer": "After finishing charging the power bank, YOU walked out of the first-floor living room, went up two floors, and arrived at the second floor.", + "question_chinese": "我从一楼到二楼的路径是什么？", + "answer_chinese": "我放完充电宝之后，走出一楼客厅，上了俩层楼，到达二楼。", + "evidence": [ + { + "description": "I put my power bank on the first floor to charge it.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:13" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You went up one floor.", + "B": "After finishing charging the power bank, YOU walked out of the first-floor living room, went up two floors, and arrived at the second floor.", + "C": "You went up three floors.", + "D": "After charging the power bank, you left the first-floor living room and reached the second floor without going up two floors.", + "E": "After finishing charging, you exited the first-floor living room, took only one flight of stairs, and arrived on the second floor." + }, + "answer_label": "B", + "answer_text": "After finishing charging the power bank, YOU walked out of the first-floor living room, went up two floors, and arrived at the second floor.", + "q_id": "egolife_A1_D6_027-mcq_fixed_6", + "correct_answer": "After finishing charging the power bank, YOU walked out of the first-floor living room, went up two floors, and arrived at the second floor.", + "original_options": { + "A": "You went up one floor.", + "B": "After finishing charging the power bank, YOU walked out of the first-floor living room, went up two floors, and arrived at the second floor.", + "C": "You went up three floors.", + "D": "After charging the power bank, you left the first-floor living room and reached the second floor without going up two floors.", + "E": "After finishing charging, you exited the first-floor living room, took only one flight of stairs, and arrived on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027-mcq_fixed_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the first floor to the second floor?", + "answer": "YOU walked out of the first-floor living room, went upstairs, and arrived at the second floor.", + "question_chinese": "我从一楼到二楼的路径是什么？", + "answer_chinese": "我走出一楼客厅，上楼，到达二楼。", + "evidence": [ + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:29" + } + }, + { + "description": "I am charging my power bank on the first floor.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:15" + } + } + ], + "question_timestamps": "00:09:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You left the first-floor bathroom.", + "B": "You exited the first-floor green room.", + "C": "You walked out of the first-floor hallway, went upstairs, and arrived at the second floor instead of leaving the living room.", + "D": "You stepped out of the first-floor living room but went downstairs first, then returned to reach the second floor.", + "E": "YOU walked out of the first-floor living room, went upstairs, and arrived at the second floor." + }, + "answer_label": "E", + "answer_text": "YOU walked out of the first-floor living room, went upstairs, and arrived at the second floor.", + "q_id": "egolife_A1_D6_027-mcq_fixed_7", + "correct_answer": "YOU walked out of the first-floor living room, went upstairs, and arrived at the second floor.", + "original_options": { + "A": "You left the first-floor bathroom.", + "B": "You exited the first-floor green room.", + "C": "You walked out of the first-floor hallway, went upstairs, and arrived at the second floor instead of leaving the living room.", + "D": "You stepped out of the first-floor living room but went downstairs first, then returned to reach the second floor.", + "E": "YOU walked out of the first-floor living room, went upstairs, and arrived at the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 21, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027-mcq_fixed_4", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement path of the power bank in my hand?", + "answer": "The power bank passed through two flights of stairs from your hand and reached the first floor, where it was placed on the ground.", + "question_chinese": "我手中充电宝的移动路径是什么？", + "answer_chinese": "充电宝从我手中经过了俩层楼梯，到达一楼，放在了地上。", + "evidence": [ + { + "description": "I took it to the first floor to charge it.", + "timestamps": { + "start": "00:03:12", + "end": "00:03:13" + } + }, + { + "description": "At first, the power bank was in my hand.", + "timestamps": { + "start": "00:02:56", + "end": "00:02:57" + } + } + ], + "question_timestamps": "00:03:15", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It went down one flight.", + "B": "It went down three flights.", + "C": "The power bank went down two flights of stairs but ended up placed on a chair, not on the ground.", + "D": "The power bank stayed in your hand through the stairs and was set on a bench on the first floor.", + "E": "The power bank passed through two flights of stairs from your hand and reached the first floor, where it was placed on the ground." + }, + "answer_label": "E", + "answer_text": "The power bank passed through two flights of stairs from your hand and reached the first floor, where it was placed on the ground.", + "q_id": "egolife_A1_D6_027-mcq_fixed_4", + "correct_answer": "The power bank passed through two flights of stairs from your hand and reached the first floor, where it was placed on the ground.", + "original_options": { + "A": "It went down one flight.", + "B": "It went down three flights.", + "C": "The power bank went down two flights of stairs but ended up placed on a chair, not on the ground.", + "D": "The power bank stayed in your hand through the stairs and was set on a bench on the first floor.", + "E": "The power bank passed through two flights of stairs from your hand and reached the first floor, where it was placed on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027-mcq_fixed_5", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement path of the power bank in my hand?", + "answer": "Finally, it was placed on the ground.", + "question_chinese": "我手中充电宝的移动路径是什么？", + "answer_chinese": "最后被放到了地上。", + "evidence": [ + { + "description": "At first, the power bank was in my hand.", + "timestamps": { + "start": "00:02:56", + "end": "00:02:57" + } + }, + { + "description": "I took it to the first floor to charge it.", + "timestamps": { + "start": "00:03:10", + "end": "00:03:12" + } + } + ], + "question_timestamps": "00:08:56", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Finally, it was placed on a chair.", + "B": "Finally, it was set on the couch.", + "C": "Finally, it was placed on the ground.", + "D": "You kept holding it at the end and did not place it on the ground.", + "E": "You put it down briefly on a bench, then picked it back up instead of leaving it on the ground." + }, + "answer_label": "C", + "answer_text": "Finally, it was placed on the ground.", + "q_id": "egolife_A1_D6_027-mcq_fixed_5", + "correct_answer": "Finally, it was placed on the ground.", + "original_options": { + "A": "Finally, it was placed on a chair.", + "B": "Finally, it was set on the couch.", + "C": "Finally, it was placed on the ground.", + "D": "You kept holding it at the end and did not place it on the ground.", + "E": "You put it down briefly on a bench, then picked it back up instead of leaving it on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027_2_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the laptop relative to me?", + "answer": "To the left of your left hand.", + "question_chinese": "此时，笔记本在我的什么位置？", + "answer_chinese": "在我的左手边。", + "evidence": [ + { + "description": "When I faced the stairs, I saw the notebook directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "I turned 90 degrees.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To the right of your right hand.", + "B": "In front of your hands.", + "C": "To the left of your left hand.", + "D": "To the left of your right hand, closer to the center than to your left hand.", + "E": "Near your right elbow at a diagonal, positioned closer to the middle of the table than your left side." + }, + "answer_label": "C", + "answer_text": "To the left of your left hand.", + "q_id": "egolife_A1_D6_027_2_T1", + "correct_answer": "To the left of your left hand.", + "original_options": { + "A": "To the right of your right hand.", + "B": "In front of your hands.", + "C": "To the left of your left hand.", + "D": "To the left of your right hand, closer to the center than to your left hand.", + "E": "Near your right elbow at a diagonal, positioned closer to the middle of the table than your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027_2_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the laptop relative to me?", + "answer": "To the left of your left hand.", + "question_chinese": "此时，笔记本在我的什么位置？", + "answer_chinese": "在我的前边。", + "evidence": [ + { + "description": "When I faced the stairs, I saw the notebook directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:05:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To the right of your left hand.", + "B": "In front of your hands.", + "C": "To the left of your left hand.", + "D": "It sits to the left of your right hand, positioned between your hands rather than outside your left hand.", + "E": "It is on your right side, slightly ahead of your hands, within easy reach on the table." + }, + "answer_label": "C", + "answer_text": "To the left of your left hand.", + "q_id": "egolife_A1_D6_027_2_T2", + "correct_answer": "To the left of your left hand.", + "original_options": { + "A": "To the right of your left hand.", + "B": "In front of your hands.", + "C": "To the left of your left hand.", + "D": "It sits to the left of your right hand, positioned between your hands rather than outside your left hand.", + "E": "It is on your right side, slightly ahead of your hands, within easy reach on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 19, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027_1_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the bottled water or the cardboard box?", + "answer": "The mineral water bottle is nearby.", + "question_chinese": "矿泉水瓶与纸箱那个距我更近？", + "answer_chinese": "矿泉水瓶近。", + "evidence": [ + { + "description": "Both of them are directly in front of me.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + }, + { + "description": "Mineral water is in front of the cardboard box.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The mineral water bottle is nearby.", + "B": "The cardboard box is nearby." + }, + "answer_label": "A", + "answer_text": "The mineral water bottle is nearby.", + "q_id": "egolife_A1_D6_027_1_T1", + "correct_answer": "The mineral water bottle is nearby.", + "original_options": { + "A": "The mineral water bottle is nearby.", + "B": "The cardboard box is nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027_1_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the bottled water or the cardboard box?", + "answer": "The mineral water bottle is nearby.", + "question_chinese": "矿泉水瓶与纸箱那个距我更近？", + "answer_chinese": "矿泉水瓶近。", + "evidence": [ + { + "description": "Both of them are directly in front of me.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + }, + { + "description": "Mineral water is in front of the cardboard box.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:23", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The mineral water bottle is nearby.", + "B": "The cardboard box is nearby." + }, + "answer_label": "A", + "answer_text": "The mineral water bottle is nearby.", + "q_id": "egolife_A1_D6_027_1_T2", + "correct_answer": "The mineral water bottle is nearby.", + "original_options": { + "A": "The mineral water bottle is nearby.", + "B": "The cardboard box is nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027_1_T3", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the bottled water or the cardboard box?", + "answer": "The mineral water bottle is nearby.", + "question_chinese": "矿泉水瓶与纸箱那个距我更近？", + "answer_chinese": "矿泉水瓶。", + "evidence": [ + { + "description": "Both of them are directly in front of me.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + }, + { + "description": "Mineral water is in front of the cardboard box.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:07:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The cardboard box is nearby.", + "B": "The mineral water bottle is nearby." + }, + "answer_label": "B", + "answer_text": "The mineral water bottle is nearby.", + "q_id": "egolife_A1_D6_027_1_T3", + "correct_answer": "The mineral water bottle is nearby.", + "original_options": { + "A": "The cardboard box is nearby.", + "B": "The mineral water bottle is nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027_1_T4", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the bottled water or the cardboard box?", + "answer": "Which is closer to you, the mineral water bottle or the cardboard box?", + "question_chinese": "矿泉水瓶与纸箱那个距我更近？", + "answer_chinese": "矿泉水瓶。", + "evidence": [ + { + "description": "Both of them are directly in front of me.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + }, + { + "description": "Mineral water is in front of the cardboard box.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:07:39", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Which is closer to you, the mineral water bottle or the cardboard box?", + "B": "The mineral water bottle is closer to you." + }, + "answer_label": "A", + "answer_text": "Which is closer to you, the mineral water bottle or the cardboard box?", + "q_id": "egolife_A1_D6_027_1_T4", + "correct_answer": "Which is closer to you, the mineral water bottle or the cardboard box?", + "original_options": { + "A": "Which is closer to you, the mineral water bottle or the cardboard box?", + "B": "The mineral water bottle is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027_5_T1", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the black power bank take?", + "answer": "YOU picked up the black power bank from near the door on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "question_chinese": "黑色充电宝的路径是什么？", + "answer_chinese": "我从一楼靠近门口的地方拿起了黑色充电宝，上了俩层楼梯，到达了二楼。", + "evidence": [ + { + "description": "A black power bank is placed on a small table by the door on the first floor.", + "timestamps": { + "start": "00:03:17", + "end": "00:03:18" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:03:18", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You climbed one flight upstairs.", + "B": "You arrived on the third floor.", + "C": "You picked it up near the staircase on the first floor, then went up to the second floor.", + "D": "You picked it up near the door on the first floor, climbed two flights, and arrived on the third floor.", + "E": "YOU picked up the black power bank from near the door on the first floor, climbed two flights of stairs, and arrived on the second floor." + }, + "answer_label": "E", + "answer_text": "YOU picked up the black power bank from near the door on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "q_id": "egolife_A1_D6_027_5_T1", + "correct_answer": "YOU picked up the black power bank from near the door on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "original_options": { + "A": "You climbed one flight upstairs.", + "B": "You arrived on the third floor.", + "C": "You picked it up near the staircase on the first floor, then went up to the second floor.", + "D": "You picked it up near the door on the first floor, climbed two flights, and arrived on the third floor.", + "E": "YOU picked up the black power bank from near the door on the first floor, climbed two flights of stairs, and arrived on the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_027", + "video_length": "00:11:18", + "qid": "egolife_A1_D6_027_5_T2", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What path did the black power bank take?", + "answer": "YOU picked up the black power bank from near the door on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "question_chinese": "黑色充电宝的路径是什么？", + "answer_chinese": "我从一楼靠近门口的地方拿起了黑色充电宝，上了俩层楼梯，到达了二楼。", + "evidence": [ + { + "description": "A black power bank is placed on a small table by the door on the first floor.", + "timestamps": { + "start": "00:03:17", + "end": "00:03:18" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:03:20", + "end": "00:03:29" + } + } + ], + "question_timestamps": "00:11:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You climbed one flight upstairs.", + "B": "You ended up on the third floor.", + "C": "YOU picked up the black power bank from near the door on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "D": "You picked it up near the door on the first floor, climbed three flights, and arrived on the second floor.", + "E": "You grabbed it near the door on the first floor and reached the second floor without using stairs." + }, + "answer_label": "C", + "answer_text": "YOU picked up the black power bank from near the door on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "q_id": "egolife_A1_D6_027_5_T2", + "correct_answer": "YOU picked up the black power bank from near the door on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "original_options": { + "A": "You climbed one flight upstairs.", + "B": "You ended up on the third floor.", + "C": "YOU picked up the black power bank from near the door on the first floor, climbed two flights of stairs, and arrived on the second floor.", + "D": "You picked it up near the door on the first floor, climbed three flights, and arrived on the second floor.", + "E": "You grabbed it near the door on the first floor and reached the second floor without using stairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 20, + 18 + ] + } +] diff --git a/QAs/egolife_A1_D6_028-mcq.json b/QAs/egolife_A1_D6_028-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..242ca02da36dcc0681e1c5b287a29063efb781d3 --- /dev/null +++ b/QAs/egolife_A1_D6_028-mcq.json @@ -0,0 +1,613 @@ +[ + { + "video_uid": "egolife_A1_D6_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_028-mcq_fixed_4", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement path of the seat cushion?", + "answer": "YOU moved the thing You were doing from the floor on your left side to somewhere near the laptop.", + "question_chinese": "坐垫的移动路径是什么？", + "answer_chinese": "我把做点从左手边的地上拿到了靠近笔记本电脑的附近。", + "evidence": [ + { + "description": "The cushion was originally placed near the large screen.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:01" + } + }, + { + "description": "Finally, I placed the cushion in front of the laptop.", + "timestamps": { + "start": "00:04:03", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:04:06", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From the right-side floor.", + "B": "From near the laptop to left.", + "C": "You picked it up from the floor on your left and placed it on the chair beside you.", + "D": "YOU moved the thing You were doing from the floor on your left side to somewhere near the laptop.", + "E": "You moved it from the floor on your left side to the far edge of the table, away from the laptop." + }, + "answer_label": "D", + "answer_text": "YOU moved the thing You were doing from the floor on your left side to somewhere near the laptop.", + "q_id": "egolife_A1_D6_028-mcq_fixed_4", + "correct_answer": "YOU moved the thing You were doing from the floor on your left side to somewhere near the laptop.", + "original_options": { + "A": "From the right-side floor.", + "B": "From near the laptop to left.", + "C": "You picked it up from the floor on your left and placed it on the chair beside you.", + "D": "YOU moved the thing You were doing from the floor on your left side to somewhere near the laptop.", + "E": "You moved it from the floor on your left side to the far edge of the table, away from the laptop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 18, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_028-mcq_fixed_5", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement path of the seat cushion?", + "answer": "YOU took it close to the laptop.", + "question_chinese": "坐垫的移动路径是什么？", + "answer_chinese": "我把它拿到了靠近笔记本电脑。", + "evidence": [ + { + "description": "The cushion was originally placed near the large screen.", + "timestamps": { + "start": "00:04:00", + "end": "00:04:02" + } + }, + { + "description": "Finally, I placed the cushion in front of the laptop.", + "timestamps": { + "start": "00:04:03", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "YOU took it under the table.", + "B": "YOU took it over the rug.", + "C": "YOU took it close to the laptop.", + "D": "YOU moved it toward the desk edge, stopping nearer the power outlet behind the desk.", + "E": "YOU carried the seat cushion to the table’s side, ending up closer to your coffee cup." + }, + "answer_label": "C", + "answer_text": "YOU took it close to the laptop.", + "q_id": "egolife_A1_D6_028-mcq_fixed_5", + "correct_answer": "YOU took it close to the laptop.", + "original_options": { + "A": "YOU took it under the table.", + "B": "YOU took it over the rug.", + "C": "YOU took it close to the laptop.", + "D": "YOU moved it toward the desk edge, stopping nearer the power outlet behind the desk.", + "E": "YOU carried the seat cushion to the table’s side, ending up closer to your coffee cup." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_028_3", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the microwave located?", + "answer": "To my left rear.", + "question_chinese": "微波炉在我的什么位置？", + "answer_chinese": "在我的左后方。", + "evidence": [ + { + "description": "When I face the balcony towards the morning sun, the microwave oven is directly in front of me.", + "timestamps": { + "start": "00:05:02", + "end": "00:05:03" + } + }, + { + "description": "I turned to the large screen.", + "timestamps": { + "start": "00:05:05", + "end": "00:05:06" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To my right rear.", + "B": "Directly to my left.", + "C": "To my left rear.", + "D": "In front of me on the counter, slightly left of center when I look ahead.", + "E": "At my right rear corner, set back behind me rather than off to my left." + }, + "answer_label": "C", + "answer_text": "To my left rear.", + "q_id": "egolife_A1_D6_028_3", + "correct_answer": "To my left rear.", + "original_options": { + "A": "To my right rear.", + "B": "Directly to my left.", + "C": "To my left rear.", + "D": "In front of me on the counter, slightly left of center when I look ahead.", + "E": "At my right rear corner, set back behind me rather than off to my left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_028_4", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the microwave located?", + "answer": "Behind me.", + "question_chinese": "微波炉在我的什么位置？", + "answer_chinese": "在我的身后。", + "evidence": [ + { + "description": "As I walked toward the door, the microwave was in front of me to the left.", + "timestamps": { + "start": "00:11:45", + "end": "00:11:46" + } + }, + { + "description": "I turned 90 degrees.", + "timestamps": { + "start": "00:11:46", + "end": "00:11:47" + } + } + ], + "question_timestamps": "00:12:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your right.", + "C": "Off to your left side, not behind you, so you’d see it when you turn left.", + "D": "Behind me.", + "E": "Ahead of you above the stove area, positioned in front rather than directly behind your body." + }, + "answer_label": "D", + "answer_text": "Behind me.", + "q_id": "egolife_A1_D6_028_4", + "correct_answer": "Behind me.", + "original_options": { + "A": "In front of you.", + "B": "To your right.", + "C": "Off to your left side, not behind you, so you’d see it when you turn left.", + "D": "Behind me.", + "E": "Ahead of you above the stove area, positioned in front rather than directly behind your body." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_028_7", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "\"Can I reach the trash can at this moment?\"", + "answer": "No.", + "question_chinese": "此时我是否能够到垃圾桶？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The trash can is placed next to the balcony.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:01" + } + }, + { + "description": "The trash can is about 2 meters away from me.", + "timestamps": { + "start": "00:05:01", + "end": "00:05:02" + } + } + ], + "question_timestamps": "00:05:06", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D6_028_7", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_028_8", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "\"Can I get to the trash can right now?\"", + "answer": "No.", + "question_chinese": "此时我是否能够到垃圾桶？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The trash can is directly in front of me.", + "timestamps": { + "start": "00:07:03", + "end": "00:07:04" + } + }, + { + "description": "The trash can is about an arm's length away from me.", + "timestamps": { + "start": "00:07:03", + "end": "00:07:04" + } + } + ], + "question_timestamps": "00:07:08", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D6_028_8", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_028-mcq_fixed_0", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "At this moment, where is the laptop that is placed on the stairs relative to me?", + "answer": "To the left of your left hand.", + "question_chinese": "此时，放在楼梯上的笔记本在我的什么位置？", + "answer_chinese": "在我的左手边。", + "evidence": [ + { + "description": "When I faced the stairs, I saw the notebook directly in front of me.", + "timestamps": { + "start": "00:03:01", + "end": "00:03:02" + } + }, + { + "description": "I turned 90 degrees.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:03:11", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To your right hand.", + "B": "In front of you.", + "C": "To the left of your right hand, sitting across your body rather than beside your left hand.", + "D": "A few steps below you on the right side of the stairs, not tucked near your left hand.", + "E": "To the left of your left hand." + }, + "answer_label": "E", + "answer_text": "To the left of your left hand.", + "q_id": "egolife_A1_D6_028-mcq_fixed_0", + "correct_answer": "To the left of your left hand.", + "original_options": { + "A": "To your right hand.", + "B": "In front of you.", + "C": "To the left of your right hand, sitting across your body rather than beside your left hand.", + "D": "A few steps below you on the right side of the stairs, not tucked near your left hand.", + "E": "To the left of your left hand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_028-mcq_fixed_1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "At this moment, where is the laptop that is placed on the stairs relative to me?", + "answer": "In front of you.", + "question_chinese": "此时，放在楼梯上的笔记本在我的什么位置？", + "answer_chinese": "在我的前边。", + "evidence": [ + { + "description": "When I faced the stairs, I saw the notebook directly in front of me.", + "timestamps": { + "start": "00:03:07", + "end": "00:03:08" + } + }, + { + "description": "I turned 90 degrees.", + "timestamps": { + "start": "00:03:09", + "end": "00:03:10" + } + } + ], + "question_timestamps": "00:08:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your right.", + "D": "Off to your left side on the stair tread, not directly in front of where you’re facing.", + "E": "Just behind you on the lower step, so it sits behind rather than straight ahead." + }, + "answer_label": "A", + "answer_text": "In front of you.", + "q_id": "egolife_A1_D6_028-mcq_fixed_1", + "correct_answer": "In front of you.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "To your right.", + "D": "Off to your left side on the stair tread, not directly in front of where you’re facing.", + "E": "Just behind you on the lower step, so it sits behind rather than straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_028_9", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "\"Where is the speaker located in relation to me?\"", + "answer": "To my right.", + "question_chinese": "音响在我的那个位置？", + "answer_chinese": "在我的右面。", + "evidence": [ + { + "description": "As I walked towards the table, I saw the speaker placed in the upper left corner of the table.", + "timestamps": { + "start": "00:06:34", + "end": "00:06:35" + } + }, + { + "description": "I am sitting in front of my laptop.", + "timestamps": { + "start": "00:06:35", + "end": "00:06:36" + } + } + ], + "question_timestamps": "00:06:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To my left.", + "B": "Behind me.", + "C": "In front of me near the desk edge, not positioned over on my right side.", + "D": "To my right.", + "E": "Slightly behind and to my left, so I’d have to turn left to face it." + }, + "answer_label": "D", + "answer_text": "To my right.", + "q_id": "egolife_A1_D6_028_9", + "correct_answer": "To my right.", + "original_options": { + "A": "To my left.", + "B": "Behind me.", + "C": "In front of me near the desk edge, not positioned over on my right side.", + "D": "To my right.", + "E": "Slightly behind and to my left, so I’d have to turn left to face it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_028", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_028_10", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "\"Where is the speaker located in my position?\"", + "answer": "Behind me.", + "question_chinese": "音响在我的那个位置？", + "answer_chinese": "在我的后方。", + "evidence": [ + { + "description": "I turned around and walked towards the balcony.", + "timestamps": { + "start": "00:06:46", + "end": "00:06:48" + } + } + ], + "question_timestamps": "00:08:49", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of me.", + "B": "Behind me.", + "C": "To my left.", + "D": "The speaker is positioned in front of you, facing you as you stand in your current spot.", + "E": "The speaker is slightly to your right, close enough that you could turn and address them easily." + }, + "answer_label": "B", + "answer_text": "Behind me.", + "q_id": "egolife_A1_D6_028_10", + "correct_answer": "Behind me.", + "original_options": { + "A": "In front of me.", + "B": "Behind me.", + "C": "To my left.", + "D": "The speaker is positioned in front of you, facing you as you stand in your current spot.", + "E": "The speaker is slightly to your right, close enough that you could turn and address them easily." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D6_029-mcq.json b/QAs/egolife_A1_D6_029-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..10b8925ffa2d79acfb28d35dadeb21a82191792d --- /dev/null +++ b/QAs/egolife_A1_D6_029-mcq.json @@ -0,0 +1,820 @@ +[ + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029-mcq_fixed_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the second floor to the first floor?", + "answer": "YOU walked straight from near the table on the second floor, turned right, then went down two flights of stairs to reach the second floor.", + "question_chinese": "我从二楼到一楼的路径是什么？", + "answer_chinese": "我从靠近二楼桌子的地方，直走，右转，然后下了两层楼梯，到达二楼。", + "evidence": [ + { + "description": "I got the bracket in the bedroom on the first floor.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + }, + { + "description": "At the beginning, I stood near the table on the second floor.", + "timestamps": { + "start": "00:02:02", + "end": "00:02:03" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turned left, then downstairs.", + "B": "YOU walked straight from near the table on the second floor, turned right, then went down two flights of stairs to reach the second floor.", + "C": "Went down one flight of stairs.", + "D": "YOU walked straight from the table area, turned right, and went down one flight to reach the first floor.", + "E": "YOU started near the table upstairs, turned left at the hallway, then descended two flights of stairs to the first floor." + }, + "answer_label": "B", + "answer_text": "YOU walked straight from near the table on the second floor, turned right, then went down two flights of stairs to reach the second floor.", + "q_id": "egolife_A1_D6_029-mcq_fixed_6", + "correct_answer": "YOU walked straight from near the table on the second floor, turned right, then went down two flights of stairs to reach the second floor.", + "original_options": { + "A": "Turned left, then downstairs.", + "B": "YOU walked straight from near the table on the second floor, turned right, then went down two flights of stairs to reach the second floor.", + "C": "Went down one flight of stairs.", + "D": "YOU walked straight from the table area, turned right, and went down one flight to reach the first floor.", + "E": "YOU started near the table upstairs, turned left at the hallway, then descended two flights of stairs to the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029-mcq_fixed_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from the second floor to the first floor?", + "answer": "YOU arrived at the first floor from near the table on the second floor.", + "question_chinese": "我从二楼到一楼的路径是什么？", + "answer_chinese": "我从靠近二楼桌子的地方，到达一楼。", + "evidence": [ + { + "description": "At the beginning, I stood near the table on the second floor.", + "timestamps": { + "start": "00:02:02", + "end": "00:02:03" + } + }, + { + "description": "I got the bracket in the bedroom on the first floor.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + } + ], + "question_timestamps": "00:08:47", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From near the sofa upstairs.", + "B": "From near the clock upstairs.", + "C": "YOU arrived at the first floor after starting near the sofa on the second floor rather than the table.", + "D": "YOU arrived at the first floor from near the table on the second floor.", + "E": "YOU arrived at the first floor after starting near the clock on the second floor, not from near the table." + }, + "answer_label": "D", + "answer_text": "YOU arrived at the first floor from near the table on the second floor.", + "q_id": "egolife_A1_D6_029-mcq_fixed_7", + "correct_answer": "YOU arrived at the first floor from near the table on the second floor.", + "original_options": { + "A": "From near the sofa upstairs.", + "B": "From near the clock upstairs.", + "C": "YOU arrived at the first floor after starting near the sofa on the second floor rather than the table.", + "D": "YOU arrived at the first floor from near the table on the second floor.", + "E": "YOU arrived at the first floor after starting near the clock on the second floor, not from near the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029-mcq_fixed_2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "When YOU stepped out of the bedroom door, the laptop was behind you.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "我刚出卧室门时，笔记本电脑在我的后方。", + "evidence": [ + { + "description": "I see laptops in front of me to the right.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:21" + } + }, + { + "description": "I turned 180 degrees and left the dormitory.", + "timestamps": { + "start": "00:02:25", + "end": "00:02:26" + } + } + ], + "question_timestamps": "00:02:27", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "When YOU stepped out of the bedroom door, the laptop was behind you.", + "B": "It was in front of you.", + "C": "It was to your left.", + "D": "When YOU stepped out of the bedroom door, the laptop was diagonally ahead of you in the hallway.", + "E": "When YOU stepped out of the bedroom door, the laptop was off to your right, just outside the doorway." + }, + "answer_label": "A", + "answer_text": "When YOU stepped out of the bedroom door, the laptop was behind you.", + "q_id": "egolife_A1_D6_029-mcq_fixed_2", + "correct_answer": "When YOU stepped out of the bedroom door, the laptop was behind you.", + "original_options": { + "A": "When YOU stepped out of the bedroom door, the laptop was behind you.", + "B": "It was in front of you.", + "C": "It was to your left.", + "D": "When YOU stepped out of the bedroom door, the laptop was diagonally ahead of you in the hallway.", + "E": "When YOU stepped out of the bedroom door, the laptop was off to your right, just outside the doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029-mcq_fixed_3", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I stepped out of the bedroom door, in what direction was the laptop computer relative to me?", + "answer": "The laptop was behind you.", + "question_chinese": "我刚出卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "笔记本电脑在我的后方。", + "evidence": [ + { + "description": "I see laptops in front of me to the right.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + }, + { + "description": "I turned 90 degrees and left the dormitory.", + "timestamps": { + "start": "00:02:25", + "end": "00:02:26" + } + } + ], + "question_timestamps": "00:07:38", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The laptop was in front of you.", + "B": "The laptop was to your right.", + "C": "The laptop was slightly to your left, positioned beside you as you stepped out of the bedroom door.", + "D": "It was diagonally ahead of you, so you would have walked toward it when leaving the bedroom.", + "E": "The laptop was behind you." + }, + "answer_label": "E", + "answer_text": "The laptop was behind you.", + "q_id": "egolife_A1_D6_029-mcq_fixed_3", + "correct_answer": "The laptop was behind you.", + "original_options": { + "A": "The laptop was in front of you.", + "B": "The laptop was to your right.", + "C": "The laptop was slightly to your left, positioned beside you as you stepped out of the bedroom door.", + "D": "It was diagonally ahead of you, so you would have walked toward it when leaving the bedroom.", + "E": "The laptop was behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029_1_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the photographer relative to me?", + "answer": "The photographer is to your right.", + "question_chinese": "摄影师在我的什么位置？", + "answer_chinese": "摄影师在你的右边。", + "evidence": [ + { + "description": "The photographer is on your right.", + "timestamps": { + "start": "00:03:26", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:03:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The photographer is to your right.", + "B": "The photographer is to your left.", + "C": "The photographer is behind you.", + "D": "The photographer is directly in front of you, within your forward view as you face ahead.", + "E": "The photographer is in front of you, but only visible after you turn around." + }, + "answer_label": "A", + "answer_text": "The photographer is to your right.", + "q_id": "egolife_A1_D6_029_1_T2", + "correct_answer": "The photographer is to your right.", + "original_options": { + "A": "The photographer is to your right.", + "B": "The photographer is to your left.", + "C": "The photographer is behind you.", + "D": "The photographer is directly in front of you, within your forward view as you face ahead.", + "E": "The photographer is in front of you, but only visible after you turn around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029_1_T1", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the photographer relative to me?", + "answer": "The photographer is in front of you.", + "question_chinese": "摄影师在我的什么位置？", + "answer_chinese": "摄影师在你的前方。", + "evidence": [ + { + "description": "When I walked down the stairs, the refrigerator was directly in front of me.", + "timestamps": { + "start": "00:03:50", + "end": "00:03:55" + } + } + ], + "question_timestamps": "00:03:59", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The photographer is in front of you.", + "B": "The photographer is behind you.", + "C": "The photographer is to your right.", + "D": "The photographer is positioned to your left, at your side rather than straight ahead of you.", + "E": "The photographer is behind you, meaning you would have to turn around to face them." + }, + "answer_label": "A", + "answer_text": "The photographer is in front of you.", + "q_id": "egolife_A1_D6_029_1_T1", + "correct_answer": "The photographer is in front of you.", + "original_options": { + "A": "The photographer is in front of you.", + "B": "The photographer is behind you.", + "C": "The photographer is to your right.", + "D": "The photographer is positioned to your left, at your side rather than straight ahead of you.", + "E": "The photographer is behind you, meaning you would have to turn around to face them." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029-mcq_fixed_4", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path of the bracket?", + "answer": "YOU picked it up from the bed in the first-floor bedroom, climbed two flights of stairs, and placed it on the balcony of the second floor.", + "question_chinese": "支架的路径是什么？", + "answer_chinese": "我从一楼卧室的床上拿起，上了俩层楼梯，放在了二楼阳台上。", + "evidence": [ + { + "description": "The bracket is on the bed in the first-floor bedroom.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + }, + { + "description": "After arriving on the second floor, I placed it near the balcony.", + "timestamps": { + "start": "00:04:12", + "end": "00:04:13" + } + } + ], + "question_timestamps": "00:04:14", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Climbed one flight of stairs.", + "B": "Placed it in the bedroom upstairs.", + "C": "YOU picked it up from the bed in the first-floor bedroom, climbed two flights of stairs, and placed it on the balcony of the second floor.", + "D": "YOU picked it up from the bed, climbed one flight of stairs, and placed it on the balcony.", + "E": "YOU picked it up from the bed, climbed three flights of stairs, and placed it on the second-floor balcony." + }, + "answer_label": "C", + "answer_text": "YOU picked it up from the bed in the first-floor bedroom, climbed two flights of stairs, and placed it on the balcony of the second floor.", + "q_id": "egolife_A1_D6_029-mcq_fixed_4", + "correct_answer": "YOU picked it up from the bed in the first-floor bedroom, climbed two flights of stairs, and placed it on the balcony of the second floor.", + "original_options": { + "A": "Climbed one flight of stairs.", + "B": "Placed it in the bedroom upstairs.", + "C": "YOU picked it up from the bed in the first-floor bedroom, climbed two flights of stairs, and placed it on the balcony of the second floor.", + "D": "YOU picked it up from the bed, climbed one flight of stairs, and placed it on the balcony.", + "E": "YOU picked it up from the bed, climbed three flights of stairs, and placed it on the second-floor balcony." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029-mcq_fixed_5", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path of the bracket?", + "answer": "YOU put it on the second-floor balcony..", + "question_chinese": "支架的路径是什么？", + "answer_chinese": "我把它放在了二楼阳台上。", + "evidence": [ + { + "description": "The bracket is on the bed in the first-floor bedroom.", + "timestamps": { + "start": "00:02:19", + "end": "00:02:20" + } + }, + { + "description": "After arriving on the second floor, I placed it near the balcony.", + "timestamps": { + "start": "00:04:12", + "end": "00:04:13" + } + } + ], + "question_timestamps": "00:07:23", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "YOU put it beside the balcony.", + "B": "YOU put it in the second-floor garden.", + "C": "YOU carried the bracket upstairs and set it just inside the balcony doorway, not out on the balcony.", + "D": "YOU moved the bracket to the balcony area but left it on the floor by the railing.", + "E": "YOU put it on the second-floor balcony.." + }, + "answer_label": "E", + "answer_text": "YOU put it on the second-floor balcony..", + "q_id": "egolife_A1_D6_029-mcq_fixed_5", + "correct_answer": "YOU put it on the second-floor balcony..", + "original_options": { + "A": "YOU put it beside the balcony.", + "B": "YOU put it in the second-floor garden.", + "C": "YOU carried the bracket upstairs and set it just inside the balcony doorway, not out on the balcony.", + "D": "YOU moved the bracket to the balcony area but left it on the floor by the railing.", + "E": "YOU put it on the second-floor balcony.." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029-mcq_fixed_0", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the moving path of the chair?", + "answer": "YOU have moved it from the hall to the corner of the balcony.", + "question_chinese": "椅子的移动路径是什么？", + "answer_chinese": "我把它从大厅已到了阳台角落。", + "evidence": [ + { + "description": "The chair was originally facing the hall.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:30" + } + }, + { + "description": "I moved it to the balcony, near the trash can.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:01:32", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "From the hall to the balcony center.", + "B": "From the hall to the garden center.", + "C": "YOU moved the chair from the hall onto the balcony, but positioned it along the railing instead of the corner.", + "D": "YOU carried the chair out of the hall and placed it at the balcony entrance, not all the way into the corner.", + "E": "YOU have moved it from the hall to the corner of the balcony." + }, + "answer_label": "E", + "answer_text": "YOU have moved it from the hall to the corner of the balcony.", + "q_id": "egolife_A1_D6_029-mcq_fixed_0", + "correct_answer": "YOU have moved it from the hall to the corner of the balcony.", + "original_options": { + "A": "From the hall to the balcony center.", + "B": "From the hall to the garden center.", + "C": "YOU moved the chair from the hall onto the balcony, but positioned it along the railing instead of the corner.", + "D": "YOU carried the chair out of the hall and placed it at the balcony entrance, not all the way into the corner.", + "E": "YOU have moved it from the hall to the corner of the balcony." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 20, + 22 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029-mcq_fixed_1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the moving path of the chair?", + "answer": "Moved to the side of the table.", + "question_chinese": "椅子的移动路径是什么？", + "answer_chinese": "移到了桌子旁边。", + "evidence": [ + { + "description": "The chair was originally facing the hall.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:30" + } + }, + { + "description": "I moved it next to the table.", + "timestamps": { + "start": "00:01:30", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:05:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Moved to the front of the table.", + "B": "Moved to the back of the room.", + "C": "You slid the chair closer to where you were sitting before, rather than relocating it beside the table.", + "D": "You rotated the chair to face the other side and left it in place, not moving it beside the table.", + "E": "Moved to the side of the table." + }, + "answer_label": "E", + "answer_text": "Moved to the side of the table.", + "q_id": "egolife_A1_D6_029-mcq_fixed_1", + "correct_answer": "Moved to the side of the table.", + "original_options": { + "A": "Moved to the front of the table.", + "B": "Moved to the back of the room.", + "C": "You slid the chair closer to where you were sitting before, rather than relocating it beside the table.", + "D": "You rotated the chair to face the other side and left it in place, not moving it beside the table.", + "E": "Moved to the side of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029_9", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my washing machine?", + "answer": "To my right rear.", + "question_chinese": "洗衣机在我的什么位置？", + "answer_chinese": "在我的右后方。", + "evidence": [ + { + "description": "When I walked down the stairs, the refrigerator was directly in front of me.", + "timestamps": { + "start": "00:02:15", + "end": "00:02:16" + } + }, + { + "description": "I turned towards the door.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:16" + } + } + ], + "question_timestamps": "00:02:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To my left rear.", + "B": "To my right front.", + "C": "It’s directly behind me, slightly off to my left side rather than on my right rear.", + "D": "To my right rear.", + "E": "It’s in front of me across the room, not positioned behind my right shoulder." + }, + "answer_label": "D", + "answer_text": "To my right rear.", + "q_id": "egolife_A1_D6_029_9", + "correct_answer": "To my right rear.", + "original_options": { + "A": "To my left rear.", + "B": "To my right front.", + "C": "It’s directly behind me, slightly off to my left side rather than on my right rear.", + "D": "To my right rear.", + "E": "It’s in front of me across the room, not positioned behind my right shoulder." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029-mcq_fixed_9", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine in my place?", + "answer": "Behind you.", + "question_chinese": "洗衣机在我的什么位置？", + "answer_chinese": "在我的后方。", + "evidence": [ + { + "description": "When I first went out, the washing machine was in front of me to the right.", + "timestamps": { + "start": "00:02:30", + "end": "00:02:31" + } + }, + { + "description": "I turned towards the stairs.", + "timestamps": { + "start": "00:02:31", + "end": "00:03:32" + } + } + ], + "question_timestamps": "00:04:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "It’s off to your right side, not behind you where you’d expect it to be.", + "D": "Behind you.", + "E": "It’s ahead of you near the doorway area, rather than located behind your position." + }, + "answer_label": "D", + "answer_text": "Behind you.", + "q_id": "egolife_A1_D6_029-mcq_fixed_9", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "It’s off to your right side, not behind you where you’d expect it to be.", + "D": "Behind you.", + "E": "It’s ahead of you near the doorway area, rather than located behind your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029_6_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the doorway relative to me?", + "answer": "The doorway is in front of you.", + "question_chinese": "门口在我的什么位置？", + "answer_chinese": "门口在你的前方。", + "evidence": [ + { + "description": "The door is in front left of you.", + "timestamps": { + "start": "00:02:25", + "end": "00:02:30" + } + } + ], + "question_timestamps": "00:02:34", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The doorway is behind you.", + "B": "The doorway is to your left.", + "C": "The doorway is off to your right side, not straight ahead of you.", + "D": "The doorway is in front of you.", + "E": "The doorway is overhead above you, rather than being located in front at eye level." + }, + "answer_label": "D", + "answer_text": "The doorway is in front of you.", + "q_id": "egolife_A1_D6_029_6_T2", + "correct_answer": "The doorway is in front of you.", + "original_options": { + "A": "The doorway is behind you.", + "B": "The doorway is to your left.", + "C": "The doorway is off to your right side, not straight ahead of you.", + "D": "The doorway is in front of you.", + "E": "The doorway is overhead above you, rather than being located in front at eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_029", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_029_6_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the doorway relative to me?", + "answer": "The doorway is to your left.", + "question_chinese": "门口在我的什么位置？", + "answer_chinese": "门口在你的左边。", + "evidence": [ + { + "description": "When I go upstairs, the door is on my left.", + "timestamps": { + "start": "00:02:42", + "end": "00:02:45" + } + } + ], + "question_timestamps": "00:02:47", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The doorway is in front of you.", + "B": "The doorway is behind you.", + "C": "The doorway is to your right, not on your left side where you’re facing.", + "D": "The doorway is above you overhead, rather than positioned to your left at the same level.", + "E": "The doorway is to your left." + }, + "answer_label": "E", + "answer_text": "The doorway is to your left.", + "q_id": "egolife_A1_D6_029_6_T1", + "correct_answer": "The doorway is to your left.", + "original_options": { + "A": "The doorway is in front of you.", + "B": "The doorway is behind you.", + "C": "The doorway is to your right, not on your left side where you’re facing.", + "D": "The doorway is above you overhead, rather than positioned to your left at the same level.", + "E": "The doorway is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D6_030-mcq.json b/QAs/egolife_A1_D6_030-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..ca5541f201621f91773981e0f7df01eaaca60b6e --- /dev/null +++ b/QAs/egolife_A1_D6_030-mcq.json @@ -0,0 +1,616 @@ +[ + { + "video_uid": "egolife_A1_D6_030", + "video_length": "00:11:54", + "qid": "egolife_A1_D6_030-mcq_fixed_8", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from outside to the first-floor living room?", + "answer": "YOU approached from near the table, weaved through the crowd, went in, walked straight, turned right into the living room.", + "question_chinese": "我从室外进入一楼客厅的路径是什么？", + "answer_chinese": "我从桌子附近，绕过人群，进门，直走，右转进入客厅。", + "evidence": [ + { + "description": "I passed by the crowd.", + "timestamps": { + "start": "00:10:27", + "end": "00:10:28" + } + } + ], + "question_timestamps": "00:10:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turned left into the living room.", + "B": "YOU approached from near the table, weaved through the crowd, went in, walked straight, turned right into the living room.", + "C": "Entered, then turned left into the living room.", + "D": "YOU approached from near the table, weaved through the crowd, went in, walked straight, then turned left into the living room.", + "E": "YOU came in from near the front porch, walked straight past the entry, then turned left before reaching the living room." + }, + "answer_label": "B", + "answer_text": "YOU approached from near the table, weaved through the crowd, went in, walked straight, turned right into the living room.", + "q_id": "egolife_A1_D6_030-mcq_fixed_8", + "correct_answer": "YOU approached from near the table, weaved through the crowd, went in, walked straight, turned right into the living room.", + "original_options": { + "A": "Turned left into the living room.", + "B": "YOU approached from near the table, weaved through the crowd, went in, walked straight, turned right into the living room.", + "C": "Entered, then turned left into the living room.", + "D": "YOU approached from near the table, weaved through the crowd, went in, walked straight, then turned left into the living room.", + "E": "YOU came in from near the front porch, walked straight past the entry, then turned left before reaching the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_030", + "video_length": "00:11:54", + "qid": "egolife_A1_D6_030-mcq_fixed_9", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from outside to the first-floor living room?", + "answer": "YOU turned right upon entering the room and went into the living room.", + "question_chinese": "我从室外进入一楼客厅的路径是什么？", + "answer_chinese": "我进入室内右转进入客厅。", + "evidence": [ + { + "description": "I wove through the crowd.", + "timestamps": { + "start": "00:10:27", + "end": "00:10:28" + } + } + ], + "question_timestamps": "00:11:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You went straight into the living room.", + "B": "You turned left into the living room.", + "C": "You entered, turned left into the hallway, and then walked through an opening to the living room.", + "D": "You entered, walked straight past the entry area, and reached the living room without turning right.", + "E": "YOU turned right upon entering the room and went into the living room." + }, + "answer_label": "E", + "answer_text": "YOU turned right upon entering the room and went into the living room.", + "q_id": "egolife_A1_D6_030-mcq_fixed_9", + "correct_answer": "YOU turned right upon entering the room and went into the living room.", + "original_options": { + "A": "You went straight into the living room.", + "B": "You turned left into the living room.", + "C": "You entered, turned left into the hallway, and then walked through an opening to the living room.", + "D": "You entered, walked straight past the entry area, and reached the living room without turning right.", + "E": "YOU turned right upon entering the room and went into the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_030", + "video_length": "00:11:54", + "qid": "egolife_A1_D6_030-mcq_fixed_6", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I entered the bedroom, in what direction was the laptop computer relative to me?", + "answer": "The laptop is directly in front of you on the right side.", + "question_chinese": "我刚进卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "笔记本电脑在我的右正前方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:10:36", + "end": "00:10:37" + } + }, + { + "description": "I turned my laptop around so it was behind me.", + "timestamps": { + "start": "00:10:40", + "end": "00:10:41" + } + } + ], + "question_timestamps": "00:10:42", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Directly in front of you on the left.", + "B": "Directly behind you on the right.", + "C": "It was slightly behind you on the left, so you would need to turn around to face it.", + "D": "The laptop is directly in front of you on the right side.", + "E": "It sat diagonally ahead to your left, not directly in front of you when you entered." + }, + "answer_label": "D", + "answer_text": "The laptop is directly in front of you on the right side.", + "q_id": "egolife_A1_D6_030-mcq_fixed_6", + "correct_answer": "The laptop is directly in front of you on the right side.", + "original_options": { + "A": "Directly in front of you on the left.", + "B": "Directly behind you on the right.", + "C": "It was slightly behind you on the left, so you would need to turn around to face it.", + "D": "The laptop is directly in front of you on the right side.", + "E": "It sat diagonally ahead to your left, not directly in front of you when you entered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_030", + "video_length": "00:11:54", + "qid": "egolife_A1_D6_030-mcq_fixed_7", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I entered the bedroom, in what direction was the laptop computer relative to me?", + "answer": "When YOU stepped into the bedroom, the laptop was right in front of you.", + "question_chinese": "我刚进卧室门时，笔记本电脑相对我在什么方位？", + "answer_chinese": "我刚进卧室门时，笔记本电脑在我的正前方。", + "evidence": [ + { + "description": "I see two laptops in front of me to the right.", + "timestamps": { + "start": "00:10:36", + "end": "00:10:37" + } + }, + { + "description": "I turned my laptop around so it was behind me.", + "timestamps": { + "start": "00:10:40", + "end": "00:10:41" + } + } + ], + "question_timestamps": "00:11:54", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It was to your left.", + "B": "When YOU stepped into the bedroom, the laptop was right in front of you.", + "C": "It was behind you.", + "D": "As you entered, the laptop was diagonally in front of you to the right, not straight ahead.", + "E": "When you stepped in, the laptop was off to your left side, so it wasn’t in your direct line of sight." + }, + "answer_label": "B", + "answer_text": "When YOU stepped into the bedroom, the laptop was right in front of you.", + "q_id": "egolife_A1_D6_030-mcq_fixed_7", + "correct_answer": "When YOU stepped into the bedroom, the laptop was right in front of you.", + "original_options": { + "A": "It was to your left.", + "B": "When YOU stepped into the bedroom, the laptop was right in front of you.", + "C": "It was behind you.", + "D": "As you entered, the laptop was diagonally in front of you to the right, not straight ahead.", + "E": "When you stepped in, the laptop was off to your left side, so it wasn’t in your direct line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 22 + ] + }, + { + "video_uid": "egolife_A1_D6_030", + "video_length": "00:11:54", + "qid": "egolife_A1_D6_030_1_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Standing here, am I closer to the power bank on the chair or the wine bottle on the table?", + "answer": "You’re closer to the power bank. The power bank is in front of you.", + "question_chinese": "我站在这里，离椅子上的充电宝近还是桌子上的酒瓶近？", + "answer_chinese": "你离充电宝近。充电宝在你正前方。", + "evidence": [ + { + "description": "The power bank is directly in front of me.", + "timestamps": { + "start": "00:05:36", + "end": "00:05:37" + } + } + ], + "question_timestamps": "00:05:48", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You’re closer to the power bank. The power bank is in front of you.", + "B": "You’re closer to the wine bottle on the table. The wine bottle is in front of you." + }, + "answer_label": "A", + "answer_text": "You’re closer to the power bank. The power bank is in front of you.", + "q_id": "egolife_A1_D6_030_1_T1", + "correct_answer": "You’re closer to the power bank. The power bank is in front of you.", + "original_options": { + "A": "You’re closer to the power bank. The power bank is in front of you.", + "B": "You’re closer to the wine bottle on the table. The wine bottle is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_030", + "video_length": "00:11:54", + "qid": "egolife_A1_D6_030_1_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Standing here, am I closer to the power bank on the chair or the wine bottle on the table?", + "answer": "You’re closer to the wine bottle on the table. The wine bottle is in front of you.", + "question_chinese": "我站在这里，离椅子上的充电宝近还是桌子上的酒瓶近？", + "answer_chinese": "你离桌子上的酒瓶更近，酒瓶在你正前方。", + "evidence": [ + { + "description": "The wine bottle is directly in front of me.", + "timestamps": { + "start": "00:07:44", + "end": "00:07:45" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You’re closer to the wine bottle on the table. The wine bottle is in front of you.", + "B": "You’re closer to the power bank. The power bank is in front of you." + }, + "answer_label": "A", + "answer_text": "You’re closer to the wine bottle on the table. The wine bottle is in front of you.", + "q_id": "egolife_A1_D6_030_1_T2", + "correct_answer": "You’re closer to the wine bottle on the table. The wine bottle is in front of you.", + "original_options": { + "A": "You’re closer to the wine bottle on the table. The wine bottle is in front of you.", + "B": "You’re closer to the power bank. The power bank is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_030", + "video_length": "00:11:54", + "qid": "egolife_A1_D6_030-mcq_fixed_0", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "At this moment, is the refrigerator in my direction?", + "answer": "To your left.", + "question_chinese": "此时冰箱的位置在我的方位？", + "answer_chinese": "在我左面。", + "evidence": [ + { + "description": "The refrigerator is to the left of the door.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:23" + } + }, + { + "description": "When I sit in the corner of the second-floor living room facing the door, the refrigerator is directly in front of me.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:23" + } + }, + { + "description": "I am facing the large screen.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:26" + } + } + ], + "question_timestamps": "00:04:27", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "To your right.", + "B": "To your left.", + "C": "In front of you.", + "D": "Directly behind you, close to the wall, so you’d have to turn around to see it.", + "E": "Diagonally in front of you to the right, near the kitchen entry rather than on your left." + }, + "answer_label": "B", + "answer_text": "To your left.", + "q_id": "egolife_A1_D6_030-mcq_fixed_0", + "correct_answer": "To your left.", + "original_options": { + "A": "To your right.", + "B": "To your left.", + "C": "In front of you.", + "D": "Directly behind you, close to the wall, so you’d have to turn around to see it.", + "E": "Diagonally in front of you to the right, near the kitchen entry rather than on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_030", + "video_length": "00:11:54", + "qid": "egolife_A1_D6_030_2_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the refrigerator relative to me?", + "answer": "The refrigerator is on your left.", + "question_chinese": "此时冰箱的位置在我的方位？", + "answer_chinese": "冰箱在你左面。", + "evidence": [ + { + "description": "The refrigerator is to the left of me.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:23" + } + }, + { + "description": "I am facing the large screen.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:26" + } + } + ], + "question_timestamps": "00:04:29", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is on your left.", + "B": "The refrigerator is in front of you.", + "C": "The refrigerator is on your right.", + "D": "The refrigerator is directly behind you, so you’d need to turn around to face it.", + "E": "The refrigerator is diagonally in front of you to the right, not positioned on your left side." + }, + "answer_label": "A", + "answer_text": "The refrigerator is on your left.", + "q_id": "egolife_A1_D6_030_2_T1", + "correct_answer": "The refrigerator is on your left.", + "original_options": { + "A": "The refrigerator is on your left.", + "B": "The refrigerator is in front of you.", + "C": "The refrigerator is on your right.", + "D": "The refrigerator is directly behind you, so you’d need to turn around to face it.", + "E": "The refrigerator is diagonally in front of you to the right, not positioned on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_030", + "video_length": "00:11:54", + "qid": "egolife_A1_D6_030_2_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Right now, where is the refrigerator relative to me?", + "answer": "The refrigerator is in front of you.", + "question_chinese": "此时冰箱的位置在我的方位？", + "answer_chinese": "冰箱在我前面。", + "evidence": [ + { + "description": "The refrigerator is to the left of the door.", + "timestamps": { + "start": "00:04:22", + "end": "00:04:23" + } + } + ], + "question_timestamps": "00:05:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The refrigerator is behind you.", + "B": "The refrigerator is on your left.", + "C": "The refrigerator is to your right side, not in your direct line of sight ahead.", + "D": "The refrigerator is diagonally behind you to the left, so it isn’t directly in front of you.", + "E": "The refrigerator is in front of you." + }, + "answer_label": "E", + "answer_text": "The refrigerator is in front of you.", + "q_id": "egolife_A1_D6_030_2_T2", + "correct_answer": "The refrigerator is in front of you.", + "original_options": { + "A": "The refrigerator is behind you.", + "B": "The refrigerator is on your left.", + "C": "The refrigerator is to your right side, not in your direct line of sight ahead.", + "D": "The refrigerator is diagonally behind you to the left, so it isn’t directly in front of you.", + "E": "The refrigerator is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_030", + "video_length": "00:11:54", + "qid": "egolife_A1_D6_030-mcq_fixed_4", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the washing machine in my place?", + "answer": "To your left rear.", + "question_chinese": "洗衣机在我的什么位置？", + "answer_chinese": "在我的左后方。", + "evidence": [ + { + "description": "When I first walked in, I saw the washing machine in front of me to the left.", + "timestamps": { + "start": "00:10:30", + "end": "00:10:31" + } + }, + { + "description": "I walked past the washing machine.", + "timestamps": { + "start": "00:10:31", + "end": "00:10:32" + } + } + ], + "question_timestamps": "00:10:33", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To your left front.", + "B": "To your right rear.", + "C": "It sits behind you on the right side, slightly back from where you’re standing.", + "D": "You’ll find it ahead of you and a bit to the left, positioned near the wall.", + "E": "To your left rear." + }, + "answer_label": "E", + "answer_text": "To your left rear.", + "q_id": "egolife_A1_D6_030-mcq_fixed_4", + "correct_answer": "To your left rear.", + "original_options": { + "A": "To your left front.", + "B": "To your right rear.", + "C": "It sits behind you on the right side, slightly back from where you’re standing.", + "D": "You’ll find it ahead of you and a bit to the left, positioned near the wall.", + "E": "To your left rear." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_030", + "video_length": "00:11:54", + "qid": "egolife_A1_D6_030-mcq_fixed_5", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the washing machine in my place?", + "answer": "When YOU went out, the washing machine was to your right rear.", + "question_chinese": "洗衣机在我的什么位置？", + "answer_chinese": "当我出门时，洗衣机在我的右后方。", + "evidence": [ + { + "description": "As I walked down the stairs, I saw the washing machine in front of me to the right.", + "timestamps": { + "start": "00:10:45", + "end": "00:10:46" + } + }, + { + "description": "I walked down the corridor to the door.", + "timestamps": { + "start": "00:10:46", + "end": "00:10:47" + } + } + ], + "question_timestamps": "00:10:48", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "When YOU went out, it was left rear.", + "B": "When YOU went out, it was right front.", + "C": "When YOU went out, the washing machine was to your right rear.", + "D": "When YOU went out, the washing machine was directly in front of you as you exited the doorway.", + "E": "When YOU went out, the washing machine was to your left front, slightly ahead along the wall." + }, + "answer_label": "C", + "answer_text": "When YOU went out, the washing machine was to your right rear.", + "q_id": "egolife_A1_D6_030-mcq_fixed_5", + "correct_answer": "When YOU went out, the washing machine was to your right rear.", + "original_options": { + "A": "When YOU went out, it was left rear.", + "B": "When YOU went out, it was right front.", + "C": "When YOU went out, the washing machine was to your right rear.", + "D": "When YOU went out, the washing machine was directly in front of you as you exited the doorway.", + "E": "When YOU went out, the washing machine was to your left front, slightly ahead along the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D6_031-mcq.json b/QAs/egolife_A1_D6_031-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3ae28061223d1b8ac3dec64f35420bc7b339fa29 --- /dev/null +++ b/QAs/egolife_A1_D6_031-mcq.json @@ -0,0 +1,829 @@ +[ + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031-mcq_fixed_8", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path of the charging cable?", + "answer": "YOU moved it from next to the table on the second floor to the first floor.", + "question_chinese": "充电线的路径是什么？", + "answer_chinese": "我把它从二楼桌子旁边拿到了一楼。", + "evidence": [ + { + "description": "The charger is under the table on the second floor.", + "timestamps": { + "start": "00:07:00", + "end": "00:07:01" + } + }, + { + "description": "Finally, I plugged it into the socket by the front door.", + "timestamps": { + "start": "00:07:56", + "end": "00:07:57" + } + } + ], + "question_timestamps": "00:07:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "YOU moved it from first to second floor.", + "B": "YOU left it on the second floor.", + "C": "YOU moved it from next to the table on the second floor to the first floor.", + "D": "YOU moved it from next to the lamp on the second floor down to the first floor.", + "E": "YOU carried it from beside the table downstairs and brought it up the stairs to the second floor." + }, + "answer_label": "C", + "answer_text": "YOU moved it from next to the table on the second floor to the first floor.", + "q_id": "egolife_A1_D6_031-mcq_fixed_8", + "correct_answer": "YOU moved it from next to the table on the second floor to the first floor.", + "original_options": { + "A": "YOU moved it from first to second floor.", + "B": "YOU left it on the second floor.", + "C": "YOU moved it from next to the table on the second floor to the first floor.", + "D": "YOU moved it from next to the lamp on the second floor down to the first floor.", + "E": "YOU carried it from beside the table downstairs and brought it up the stairs to the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031-mcq_fixed_9", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the path of the charging cable?", + "answer": "YOU took it from the second floor to the first floor.", + "question_chinese": "充电线的路径是什么？", + "answer_chinese": "我把它从二楼拿到了一楼。", + "evidence": [ + { + "description": "The charger is under the table on the second floor.", + "timestamps": { + "start": "00:07:01", + "end": "00:07:02" + } + }, + { + "description": "Finally, I plugged it into the socket by the main door.", + "timestamps": { + "start": "00:07:56", + "end": "00:07:57" + } + } + ], + "question_timestamps": "00:12:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "YOU took it from first to second floor.", + "B": "YOU took it from the second floor outside.", + "C": "YOU took it from the second floor down to the ground floor and left it near the entry.", + "D": "YOU took it from the first floor up to the second floor, bringing it back upstairs.", + "E": "YOU took it from the second floor to the first floor." + }, + "answer_label": "E", + "answer_text": "YOU took it from the second floor to the first floor.", + "q_id": "egolife_A1_D6_031-mcq_fixed_9", + "correct_answer": "YOU took it from the second floor to the first floor.", + "original_options": { + "A": "YOU took it from first to second floor.", + "B": "YOU took it from the second floor outside.", + "C": "YOU took it from the second floor down to the ground floor and left it near the entry.", + "D": "YOU took it from the first floor up to the second floor, bringing it back upstairs.", + "E": "YOU took it from the second floor to the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031-mcq_fixed_6", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from outside to the second floor?", + "answer": "YOU entered the door, then went up two flights of stairs to reach the second floor.", + "question_chinese": "我从室外到二楼的路径是什么？", + "answer_chinese": "我进门，然后上了俩层楼梯，到达二楼。", + "evidence": [ + { + "description": "I entered the room from outside.", + "timestamps": { + "start": "00:04:03", + "end": "00:04:04" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:04:06", + "end": "00:04:11" + } + } + ], + "question_timestamps": "00:04:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "YOU entered the door, then went up two flights of stairs to reach the second floor.", + "B": "YOU went up one flight of stairs.", + "C": "YOU went up three flights of stairs.", + "D": "YOU entered the door, then went up one flight of stairs to reach the second floor.", + "E": "YOU entered the door, then climbed three flights of stairs before arriving at the second floor." + }, + "answer_label": "A", + "answer_text": "YOU entered the door, then went up two flights of stairs to reach the second floor.", + "q_id": "egolife_A1_D6_031-mcq_fixed_6", + "correct_answer": "YOU entered the door, then went up two flights of stairs to reach the second floor.", + "original_options": { + "A": "YOU entered the door, then went up two flights of stairs to reach the second floor.", + "B": "YOU went up one flight of stairs.", + "C": "YOU went up three flights of stairs.", + "D": "YOU entered the door, then went up one flight of stairs to reach the second floor.", + "E": "YOU entered the door, then climbed three flights of stairs before arriving at the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031-mcq_fixed_7", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the path from outside to the second floor?", + "answer": "YOU entered the room and arrived at the second floor.", + "question_chinese": "我从室外到二楼的路径是什么？", + "answer_chinese": "我进门，到达二楼。", + "evidence": [ + { + "description": "I entered the room from outside.", + "timestamps": { + "start": "00:05:54", + "end": "00:05:55" + } + }, + { + "description": "I climbed two flights of stairs.", + "timestamps": { + "start": "00:06:01", + "end": "00:06:16" + } + } + ], + "question_timestamps": "00:06:17", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "YOU arrived at first floor.", + "B": "YOU arrived at ground floor.", + "C": "YOU entered the room and arrived at the second floor.", + "D": "YOU entered the room and arrived at the basement level instead of reaching the second floor.", + "E": "YOU entered the room, but you stopped on a lower level and never made it to the second floor." + }, + "answer_label": "C", + "answer_text": "YOU entered the room and arrived at the second floor.", + "q_id": "egolife_A1_D6_031-mcq_fixed_7", + "correct_answer": "YOU entered the room and arrived at the second floor.", + "original_options": { + "A": "YOU arrived at first floor.", + "B": "YOU arrived at ground floor.", + "C": "YOU entered the room and arrived at the second floor.", + "D": "YOU entered the room and arrived at the basement level instead of reaching the second floor.", + "E": "YOU entered the room, but you stopped on a lower level and never made it to the second floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031_2_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement path of my light panel?", + "answer": "You moved the lighting reflector that was at the doorway to in front of the main entrance.", + "question_chinese": "我打光板的移动路径是什么？", + "answer_chinese": "我把放在门口的打光板移动到了大门前面。", + "evidence": [ + { + "description": "I moved the light board from the door to the front.", + "timestamps": { + "start": "00:09:34", + "end": "00:09:36" + } + } + ], + "question_timestamps": "00:09:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved it from doorway to inside.", + "B": "You moved it from entrance to doorway.", + "C": "You moved the lighting reflector that was at the doorway to behind the main entrance.", + "D": "You moved the lighting reflector that was at the doorway to in front of the main entrance.", + "E": "You moved the lighting reflector from the doorway to the side of the main entrance." + }, + "answer_label": "D", + "answer_text": "You moved the lighting reflector that was at the doorway to in front of the main entrance.", + "q_id": "egolife_A1_D6_031_2_T1", + "correct_answer": "You moved the lighting reflector that was at the doorway to in front of the main entrance.", + "original_options": { + "A": "You moved it from doorway to inside.", + "B": "You moved it from entrance to doorway.", + "C": "You moved the lighting reflector that was at the doorway to behind the main entrance.", + "D": "You moved the lighting reflector that was at the doorway to in front of the main entrance.", + "E": "You moved the lighting reflector from the doorway to the side of the main entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031_2_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement path of my light panel?", + "answer": "You moved the lighting reflector next to you to in front of the main entrance.", + "question_chinese": "我打光板的移动路径是什么？", + "answer_chinese": "你把你旁边的打光板移动到了大门前面。", + "evidence": [ + { + "description": "I moved the light board from the door to the front.", + "timestamps": { + "start": "00:09:34", + "end": "00:09:36" + } + } + ], + "question_timestamps": "00:09:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You moved it next to you behind.", + "B": "You moved it from doorway to front.", + "C": "You moved the lighting reflector next to you to in front of the main entrance.", + "D": "You moved the lighting reflector next to you to behind the main entrance, not out in front.", + "E": "You moved the lighting reflector that was at the doorway to in front of the main entrance instead." + }, + "answer_label": "C", + "answer_text": "You moved the lighting reflector next to you to in front of the main entrance.", + "q_id": "egolife_A1_D6_031_2_T2", + "correct_answer": "You moved the lighting reflector next to you to in front of the main entrance.", + "original_options": { + "A": "You moved it next to you behind.", + "B": "You moved it from doorway to front.", + "C": "You moved the lighting reflector next to you to in front of the main entrance.", + "D": "You moved the lighting reflector next to you to behind the main entrance, not out in front.", + "E": "You moved the lighting reflector that was at the doorway to in front of the main entrance instead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031_1_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement path of my light panel?", + "answer": "The lighting reflector was moved by you from to your left of the second-floor table, down two flights of stairs.", + "question_chinese": "我的打光板的移动路径是什么？", + "answer_chinese": "打光板被我从二楼桌子的左边，下俩层楼梯。", + "evidence": [ + { + "description": "The light board is on the left side of the table.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:19" + } + }, + { + "description": "I turned on the light and walked down the stairs, then went outside.", + "timestamps": { + "start": "00:04:45", + "end": "00:04:59" + } + } + ], + "question_timestamps": "00:05:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the right side, down two flights.", + "B": "From the front, down two flights.", + "C": "The lighting reflector was moved by you from to your left of the second-floor table, down two flights of stairs.", + "D": "You moved the reflector from behind the second-floor table, carrying it down two flights of stairs.", + "E": "You carried the reflector from the left of the second-floor table down only one flight of stairs." + }, + "answer_label": "C", + "answer_text": "The lighting reflector was moved by you from to your left of the second-floor table, down two flights of stairs.", + "q_id": "egolife_A1_D6_031_1_T2", + "correct_answer": "The lighting reflector was moved by you from to your left of the second-floor table, down two flights of stairs.", + "original_options": { + "A": "From the right side, down two flights.", + "B": "From the front, down two flights.", + "C": "The lighting reflector was moved by you from to your left of the second-floor table, down two flights of stairs.", + "D": "You moved the reflector from behind the second-floor table, carrying it down two flights of stairs.", + "E": "You carried the reflector from the left of the second-floor table down only one flight of stairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031_1_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What was the movement path of my light panel?", + "answer": "The lighting reflector was moved by you from to your left of the second-floor table, down two flights of stairs, to a spot outdoors near the entrance.", + "question_chinese": "我的打光板的移动路径是什么？", + "answer_chinese": "打光板被我从二楼桌子的左边，下俩层楼梯之后，移动到了室外靠进们的位置。", + "evidence": [ + { + "description": "The light board is on the left side of the table.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:19" + } + }, + { + "description": "I turned on the light and walked down the stairs, then went outside.", + "timestamps": { + "start": "00:04:45", + "end": "00:04:59" + } + }, + { + "description": "I put the lamp outside the door.", + "timestamps": { + "start": "00:05:14", + "end": "00:05:15" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "From the table’s right side to outdoors.", + "B": "The lighting reflector was moved by you from to your left of the second-floor table, down two flights of stairs, to a spot outdoors near the entrance.", + "C": "From behind the table to outdoors.", + "D": "You moved the lighting reflector from the left of the second-floor table down one flight, ending outdoors near the entrance.", + "E": "You carried the lighting reflector from the left of the second-floor table down two flights, ending indoors near the entrance." + }, + "answer_label": "B", + "answer_text": "The lighting reflector was moved by you from to your left of the second-floor table, down two flights of stairs, to a spot outdoors near the entrance.", + "q_id": "egolife_A1_D6_031_1_T1", + "correct_answer": "The lighting reflector was moved by you from to your left of the second-floor table, down two flights of stairs, to a spot outdoors near the entrance.", + "original_options": { + "A": "From the table’s right side to outdoors.", + "B": "The lighting reflector was moved by you from to your left of the second-floor table, down two flights of stairs, to a spot outdoors near the entrance.", + "C": "From behind the table to outdoors.", + "D": "You moved the lighting reflector from the left of the second-floor table down one flight, ending outdoors near the entrance.", + "E": "You carried the lighting reflector from the left of the second-floor table down two flights, ending indoors near the entrance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031_1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement path of the light board?", + "answer": "I moved the lighting board from the left side of the table on the second floor, down two flights of stairs, to a position outdoors near the door.", + "question_chinese": "打光板的移动路径是什么？", + "answer_chinese": "打光板被我从二楼桌子的左边，下俩层楼梯之后，移动到了室外靠进们的位置。", + "evidence": [ + { + "description": "The light board is on the left side of the table.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:19" + } + }, + { + "description": "I turned on the light and walked down the stairs, then went outside.", + "timestamps": { + "start": "00:04:45", + "end": "00:04:59" + } + }, + { + "description": "I put the lamp outside the door.", + "timestamps": { + "start": "00:05:14", + "end": "00:05:15" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "I moved the lighting board from the left side of the table on the second floor, down two flights of stairs, to a position outdoors near the door.", + "B": "From the right side, down two flights.", + "C": "From the left side, down one flight.", + "D": "I moved the lighting board from the left table, down three flights, to outdoors near the door.", + "E": "I moved the lighting board from the left table, down two flights, to indoors near the door." + }, + "answer_label": "A", + "answer_text": "I moved the lighting board from the left side of the table on the second floor, down two flights of stairs, to a position outdoors near the door.", + "q_id": "egolife_A1_D6_031_1", + "correct_answer": "I moved the lighting board from the left side of the table on the second floor, down two flights of stairs, to a position outdoors near the door.", + "original_options": { + "A": "I moved the lighting board from the left side of the table on the second floor, down two flights of stairs, to a position outdoors near the door.", + "B": "From the right side, down two flights.", + "C": "From the left side, down one flight.", + "D": "I moved the lighting board from the left table, down three flights, to outdoors near the door.", + "E": "I moved the lighting board from the left table, down two flights, to indoors near the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031_2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "What is the movement path of the lighting board?", + "answer": "I moved the lighting board that was placed by the door to in front of the main gate.", + "question_chinese": "打光板的移动路径是什么？", + "answer_chinese": "我把放在门口的打光板移动到了大门前面。", + "evidence": [ + { + "description": "I moved the light board from the door to the front.", + "timestamps": { + "start": "00:09:34", + "end": "00:09:36" + } + } + ], + "question_timestamps": "00:09:40", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To behind the main gate.", + "B": "To the left of the main gate.", + "C": "I moved the lighting board from by the door to the right side of the main gate.", + "D": "I moved the lighting board that was placed by the door to in front of the main gate.", + "E": "I moved the lighting board from by the door to in front of the side gate instead of the main gate." + }, + "answer_label": "D", + "answer_text": "I moved the lighting board that was placed by the door to in front of the main gate.", + "q_id": "egolife_A1_D6_031_2", + "correct_answer": "I moved the lighting board that was placed by the door to in front of the main gate.", + "original_options": { + "A": "To behind the main gate.", + "B": "To the left of the main gate.", + "C": "I moved the lighting board from by the door to the right side of the main gate.", + "D": "I moved the lighting board that was placed by the door to in front of the main gate.", + "E": "I moved the lighting board from by the door to in front of the side gate instead of the main gate." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 17, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031-mcq_fixed_2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine in my place?", + "answer": "To your left rear.", + "question_chinese": "洗衣机在我的什么位置？", + "answer_chinese": "在我的左后方。", + "evidence": [ + { + "description": "When I first walked in, I saw the washing machine in front of me to the left.", + "timestamps": { + "start": "00:04:03", + "end": "00:04:04" + } + }, + { + "description": "I walked past the washing machine.", + "timestamps": { + "start": "00:04:04", + "end": "00:04:05" + } + } + ], + "question_timestamps": "00:04:06", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "To your right rear.", + "B": "To your left rear.", + "C": "To your left front.", + "D": "Behind you, roughly centered between the dining table and the balcony entrance.", + "E": "In the left corner near the sink and shelves, just past where you’re standing." + }, + "answer_label": "B", + "answer_text": "To your left rear.", + "q_id": "egolife_A1_D6_031-mcq_fixed_2", + "correct_answer": "To your left rear.", + "original_options": { + "A": "To your right rear.", + "B": "To your left rear.", + "C": "To your left front.", + "D": "Behind you, roughly centered between the dining table and the balcony entrance.", + "E": "In the left corner near the sink and shelves, just past where you’re standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031-mcq_fixed_3", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the washing machine in my place?", + "answer": "The washing machine was to your right rear.", + "question_chinese": "洗衣机在我的什么位置？", + "answer_chinese": "洗衣机在我的右后方。", + "evidence": [ + { + "description": "As I walked down the stairs, I saw the washing machine in front of me to the right.", + "timestamps": { + "start": "00:04:58", + "end": "00:04:59" + } + }, + { + "description": "I walked down the corridor to the door.", + "timestamps": { + "start": "00:04:49", + "end": "00:05:01" + } + } + ], + "question_timestamps": "00:05:02", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The washing machine was to your left rear.", + "B": "The washing machine was to your right front.", + "C": "You would find the washing machine behind the door, tucked into the corner on your left.", + "D": "If you turned from the entrance, the washing machine would be along the left wall, under the detergent shelf.", + "E": "The washing machine was to your right rear." + }, + "answer_label": "E", + "answer_text": "The washing machine was to your right rear.", + "q_id": "egolife_A1_D6_031-mcq_fixed_3", + "correct_answer": "The washing machine was to your right rear.", + "original_options": { + "A": "The washing machine was to your left rear.", + "B": "The washing machine was to your right front.", + "C": "You would find the washing machine behind the door, tucked into the corner on your left.", + "D": "If you turned from the entrance, the washing machine would be along the left wall, under the detergent shelf.", + "E": "The washing machine was to your right rear." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031-mcq_fixed_4", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the water cup while standing here?", + "answer": "Yes.", + "question_chinese": "站在这里我是否能拿到水杯？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The water cup is directly in front of me.", + "timestamps": { + "start": "00:02:38", + "end": "00:02:39" + } + }, + { + "description": "The water cup is about 50 centimeters away from me.", + "timestamps": { + "start": "00:02:38", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:02:40", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D6_031-mcq_fixed_4", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_031", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_031-mcq_fixed_5", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the water cup while standing here?", + "answer": "No.", + "question_chinese": "站在这里我是否能拿到水杯？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The water cup is directly in front of me.", + "timestamps": { + "start": "00:02:38", + "end": "00:02:39" + } + }, + { + "description": "The water cup is about 100 centimeters away from me.", + "timestamps": { + "start": "00:02:38", + "end": "00:02:39" + } + } + ], + "question_timestamps": "00:06:32", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D6_031-mcq_fixed_5", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D6_038-mcq.json b/QAs/egolife_A1_D6_038-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..10799f43e967e8f1abcc9b7aa765b5a4d7934ea3 --- /dev/null +++ b/QAs/egolife_A1_D6_038-mcq.json @@ -0,0 +1,530 @@ +[ + { + "video_uid": "egolife_A1_D6_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_038_6", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the microwave or the refrigerator?", + "answer": "Refrigerator.", + "question_chinese": "微波炉和冰箱，哪个离我更近一些？", + "answer_chinese": "冰箱", + "evidence": [ + { + "description": "The microwave oven is beside the windowsill", + "timestamps": { + "start": "00:01:55", + "end": "00:02:00" + } + }, + { + "description": "The refrigerator is beside the main door", + "timestamps": { + "start": "00:01:49", + "end": "00:01:53" + } + } + ], + "question_timestamps": "00:06:53", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Microwave oven.", + "B": "Refrigerator." + }, + "answer_label": "B", + "answer_text": "Refrigerator.", + "q_id": "egolife_A1_D6_038_6", + "correct_answer": "Refrigerator.", + "original_options": { + "A": "Microwave oven.", + "B": "Refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D6_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_038_5", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the microwave or the refrigerator?", + "answer": "Microwave oven.", + "question_chinese": "微波炉和冰箱，哪个离我更近一些？", + "answer_chinese": "微波炉", + "evidence": [ + { + "description": "The microwave oven is beside the windowsill", + "timestamps": { + "start": "00:01:55", + "end": "00:02:00" + } + }, + { + "description": "The refrigerator is beside the main door", + "timestamps": { + "start": "00:01:49", + "end": "00:01:53" + } + } + ], + "question_timestamps": "00:11:27", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Microwave oven.", + "B": "Refrigerator." + }, + "answer_label": "A", + "answer_text": "Microwave oven.", + "q_id": "egolife_A1_D6_038_5", + "correct_answer": "Microwave oven.", + "original_options": { + "A": "Microwave oven.", + "B": "Refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_038_1_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I last entered the living room on the first floor, which side was the windowsill on?", + "answer": "Left.", + "question_chinese": "我最近一次进入一楼客厅时，窗台在我的哪一侧？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The last time I entered the living room, the windowsill was on my left", + "timestamps": { + "start": "00:00:10", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:29", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right.", + "B": "Left." + }, + "answer_label": "B", + "answer_text": "Left.", + "q_id": "egolife_A1_D6_038_1_T1", + "correct_answer": "Left.", + "original_options": { + "A": "Right.", + "B": "Left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_038_1_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I last entered the living room on the first floor, which side was the windowsill on?", + "answer": "Left.", + "question_chinese": "我最近一次进入一楼客厅时，窗台在我的哪一侧？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The last time I entered the living room, the windowsill was on my left", + "timestamps": { + "start": "00:00:10", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right.", + "B": "Left." + }, + "answer_label": "B", + "answer_text": "Left.", + "q_id": "egolife_A1_D6_038_1_T2", + "correct_answer": "Left.", + "original_options": { + "A": "Right.", + "B": "Left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_038_2_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the place where I can use headphones?", + "answer": "Turn left and go straight into the bedroom.", + "question_chinese": "我现在想要用耳机，应该怎么走？", + "answer_chinese": "左转，直走进入卧室", + "evidence": [ + { + "description": "The headphones are on the table in the bedroom on the first floor", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:02:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right into the bedroom.", + "B": "Turn left into the bathroom.", + "C": "Turn left and go straight into the bedroom.", + "D": "Turn left and go straight into the hallway instead of entering the bedroom.", + "E": "Turn right and keep going straight until you reach the corridor, not the bedroom." + }, + "answer_label": "C", + "answer_text": "Turn left and go straight into the bedroom.", + "q_id": "egolife_A1_D6_038_2_T1", + "correct_answer": "Turn left and go straight into the bedroom.", + "original_options": { + "A": "Turn right into the bedroom.", + "B": "Turn left into the bathroom.", + "C": "Turn left and go straight into the bedroom.", + "D": "Turn left and go straight into the hallway instead of entering the bedroom.", + "E": "Turn right and keep going straight until you reach the corridor, not the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_038_2_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the place where I can use headphones?", + "answer": "Turn left and go straight into the bedroom.", + "question_chinese": "我现在想要用耳机，应该怎么走？", + "answer_chinese": "左转，直走进入卧室", + "evidence": [ + { + "description": "The headphones are on the table in the bedroom on the first floor", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:02:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn right into the bedroom.", + "B": "Turn left into the hallway.", + "C": "Turn left and go straight into the bedroom.", + "D": "Turn left and go straight into the bathroom rather than continuing into the bedroom.", + "E": "Turn right and walk straight down the corridor, stopping before you reach the bedroom." + }, + "answer_label": "C", + "answer_text": "Turn left and go straight into the bedroom.", + "q_id": "egolife_A1_D6_038_2_T2", + "correct_answer": "Turn left and go straight into the bedroom.", + "original_options": { + "A": "Turn right into the bedroom.", + "B": "Turn left into the hallway.", + "C": "Turn left and go straight into the bedroom.", + "D": "Turn left and go straight into the bathroom rather than continuing into the bedroom.", + "E": "Turn right and walk straight down the corridor, stopping before you reach the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_038_3_T1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to where I can use headphones?", + "answer": "Turn right, then go straight into the house. At the staircase, turn right into the living room and then walk forward into the bedroom.", + "question_chinese": "我现在想要用耳机，应该怎么走？", + "answer_chinese": "右转，然后直行进入房子，在楼梯口右转进入客厅，然后向前方走进卧室", + "evidence": [ + { + "description": "The headphones are on the table in the bedroom on the first floor", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + }, + { + "description": "To get from the living room to the yard, you need to first go to the staircase, then turn left and go straight", + "timestamps": { + "start": "00:10:42", + "end": "00:11:01" + } + } + ], + "question_timestamps": "00:11:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go into the house.", + "B": "Turn right, then go straight into the house. At the staircase, turn right into the living room and then walk forward into the bedroom.", + "C": "Go straight into the living room.", + "D": "Turn right and go straight into the house, then turn left at the staircase into the living room.", + "E": "Turn right into the house, but at the staircase go left and continue forward into the bedroom." + }, + "answer_label": "B", + "answer_text": "Turn right, then go straight into the house. At the staircase, turn right into the living room and then walk forward into the bedroom.", + "q_id": "egolife_A1_D6_038_3_T1", + "correct_answer": "Turn right, then go straight into the house. At the staircase, turn right into the living room and then walk forward into the bedroom.", + "original_options": { + "A": "Turn left, then go into the house.", + "B": "Turn right, then go straight into the house. At the staircase, turn right into the living room and then walk forward into the bedroom.", + "C": "Go straight into the living room.", + "D": "Turn right and go straight into the house, then turn left at the staircase into the living room.", + "E": "Turn right into the house, but at the staircase go left and continue forward into the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_038_3_T2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to where I can use headphones?", + "answer": "Turn right, then go straight into the house. At the staircase, turn right into the living room and then walk forward into the bedroom.", + "question_chinese": "我现在想要用耳机，应该怎么走？", + "answer_chinese": "右转，然后直行进入房子，在楼梯口右转进入客厅，然后向前方走进卧室", + "evidence": [ + { + "description": "The headphones are on the table in the bedroom on the first floor", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + }, + { + "description": "To get from the living room to the yard, you need to first go to the staircase, then turn left and go straight", + "timestamps": { + "start": "00:10:42", + "end": "00:11:01" + } + } + ], + "question_timestamps": "00:11:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then enter the house.", + "B": "Turn right, then go straight into the house. At the staircase, turn right into the living room and then walk forward into the bedroom.", + "C": "Turn right into the living room.", + "D": "Turn right, go straight into the house, then at the staircase turn right into the bedroom directly.", + "E": "Turn right and go straight into the house, then turn left at the staircase and walk forward into the bedroom." + }, + "answer_label": "B", + "answer_text": "Turn right, then go straight into the house. At the staircase, turn right into the living room and then walk forward into the bedroom.", + "q_id": "egolife_A1_D6_038_3_T2", + "correct_answer": "Turn right, then go straight into the house. At the staircase, turn right into the living room and then walk forward into the bedroom.", + "original_options": { + "A": "Turn left, then enter the house.", + "B": "Turn right, then go straight into the house. At the staircase, turn right into the living room and then walk forward into the bedroom.", + "C": "Turn right into the living room.", + "D": "Turn right, go straight into the house, then at the staircase turn right into the bedroom directly.", + "E": "Turn right and go straight into the house, then turn left at the staircase and walk forward into the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_038_7", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the cake now?", + "answer": "Yes.", + "question_chinese": "我现在能否拿到蛋糕？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The cake is on the table in the living room on the first floor", + "timestamps": { + "start": "00:06:56", + "end": "00:07:01" + } + } + ], + "question_timestamps": "00:07:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D6_038_7", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D6_038", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_038_8", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the cake now?", + "answer": "No.", + "question_chinese": "我现在能否拿到蛋糕？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The cake is on the table in the living room on the first floor", + "timestamps": { + "start": "00:06:56", + "end": "00:07:01" + } + } + ], + "question_timestamps": "00:10:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D6_038_8", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D6_042-mcq.json b/QAs/egolife_A1_D6_042-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1d4469cabd56a5f41ae232ed8d314f088cd5fcd5 --- /dev/null +++ b/QAs/egolife_A1_D6_042-mcq.json @@ -0,0 +1,762 @@ +[ + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_11", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the chessboard from me?", + "answer": "The chessboard is directly in front of you.", + "question_chinese": "国际象棋的棋盘在我的什么方向？", + "answer_chinese": "棋盘在你的正前方。", + "evidence": [ + { + "description": "There is a huge chessboard directly in front of me.", + "timestamps": { + "start": "00:00:53", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:05:57", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The chessboard is directly in front of you.", + "D": "The chessboard is directly behind you, so you would need to turn around to face it.", + "E": "The chessboard is off to your right side rather than straight ahead of you." + }, + "answer_label": "C", + "answer_text": "The chessboard is directly in front of you.", + "q_id": "egolife_A1_D6_042_11", + "correct_answer": "The chessboard is directly in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The chessboard is directly in front of you.", + "D": "The chessboard is directly behind you, so you would need to turn around to face it.", + "E": "The chessboard is off to your right side rather than straight ahead of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_12", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the chessboard in relation to me?", + "answer": "The chessboard is behind you on the right.", + "question_chinese": "国际象棋的棋盘在我的什么方向？", + "answer_chinese": "棋盘在你的右后方。", + "evidence": [ + { + "description": "There is a huge chessboard directly in front of me.", + "timestamps": { + "start": "00:00:53", + "end": "00:00:54" + } + } + ], + "question_timestamps": "00:11:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The chessboard is behind you on the right.", + "B": "In front of you.", + "C": "Behind you on the left.", + "D": "The chessboard is directly in front of you, so you can see it without turning around.", + "E": "The chessboard is behind you on the left, meaning you’d need to turn left to face it." + }, + "answer_label": "A", + "answer_text": "The chessboard is behind you on the right.", + "q_id": "egolife_A1_D6_042_12", + "correct_answer": "The chessboard is behind you on the right.", + "original_options": { + "A": "The chessboard is behind you on the right.", + "B": "In front of you.", + "C": "Behind you on the left.", + "D": "The chessboard is directly in front of you, so you can see it without turning around.", + "E": "The chessboard is behind you on the left, meaning you’d need to turn left to face it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_7", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Which door is closer to me, the one near the courtyard or the one near the bedroom, between the two open doors in the living room?", + "answer": "The door leading to the courtyard is closer to you.", + "question_chinese": "在房子客厅的两扇打开的门之中，是靠近庭院的门，还是靠近卧室的门离我更近？", + "answer_chinese": "靠近庭院的门离你更近。", + "evidence": [ + { + "description": "The door leading to the bedroom is on my left.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + }, + { + "description": "The door leading to the courtyard is directly in front of me on the left.", + "timestamps": { + "start": "00:00:57", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:06:34", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The door leading to the courtyard is closer to you.", + "B": "The door leading to the bedroom is closer to you." + }, + "answer_label": "A", + "answer_text": "The door leading to the courtyard is closer to you.", + "q_id": "egolife_A1_D6_042_7", + "correct_answer": "The door leading to the courtyard is closer to you.", + "original_options": { + "A": "The door leading to the courtyard is closer to you.", + "B": "The door leading to the bedroom is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_8", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Which door is closer to me between the two open doors in the living room: the one leading to the courtyard or the one leading to the bedroom?", + "answer": "The door leading to the courtyard is closer to you.", + "question_chinese": "在房子客厅的两扇打开的门之中，是靠近庭院的门，还是靠近卧室的门离我更近？", + "answer_chinese": "靠近卧室的门离你更近", + "evidence": [ + { + "description": "The door leading to the bedroom is on my left.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + }, + { + "description": "The door leading to the courtyard is directly in front of me on the left.", + "timestamps": { + "start": "00:00:57", + "end": "00:00:58" + } + } + ], + "question_timestamps": "00:11:19", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The door leading to the courtyard is closer to you.", + "B": "The door leading to the bedroom is closer to you." + }, + "answer_label": "A", + "answer_text": "The door leading to the courtyard is closer to you.", + "q_id": "egolife_A1_D6_042_8", + "correct_answer": "The door leading to the courtyard is closer to you.", + "original_options": { + "A": "The door leading to the courtyard is closer to you.", + "B": "The door leading to the bedroom is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_3", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bedroom, how should I get there?", + "answer": "You should turn left to enter the door, walk straight for about four steps, then turn right into the living room, and afterward walk forward to your left for roughly six steps to reach the doorway of the bedroom.", + "question_chinese": "如果我想去卧室，我应该怎么走？", + "answer_chinese": "你应该左转进门，直走4步的距离，然后右转进入客厅，随后向左前方走6步左右的距离就能到达卧室门口了。", + "evidence": [ + { + "description": "The bedroom is on my left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The table in the living room is about two steps away from the bedroom.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The table in the living room is about two steps away from the bedroom.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "I walked from inside the house to the courtyard.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:06:56", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right into the door.", + "B": "Go straight into the living room.", + "C": "Turn left into the door, walk four steps, then turn left into the living room and continue six steps.", + "D": "You should turn left to enter the door, walk straight for about four steps, then turn right into the living room, and afterward walk forward to your left for roughly six steps to reach the doorway of the bedroom.", + "E": "Turn left to enter, walk about two steps, turn right into the living room, then go forward left about six steps." + }, + "answer_label": "D", + "answer_text": "You should turn left to enter the door, walk straight for about four steps, then turn right into the living room, and afterward walk forward to your left for roughly six steps to reach the doorway of the bedroom.", + "q_id": "egolife_A1_D6_042_3", + "correct_answer": "You should turn left to enter the door, walk straight for about four steps, then turn right into the living room, and afterward walk forward to your left for roughly six steps to reach the doorway of the bedroom.", + "original_options": { + "A": "Turn right into the door.", + "B": "Go straight into the living room.", + "C": "Turn left into the door, walk four steps, then turn left into the living room and continue six steps.", + "D": "You should turn left to enter the door, walk straight for about four steps, then turn right into the living room, and afterward walk forward to your left for roughly six steps to reach the doorway of the bedroom.", + "E": "Turn left to enter, walk about two steps, turn right into the living room, then go forward left about six steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_4", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bedroom, how should I get there?", + "answer": "You should turn left and walk—after about two steps, you’ll reach the bedroom.", + "question_chinese": "如果我想去卧室，我应该怎么走？", + "answer_chinese": "你应该向左转走两步的距离就能进入卧室了。", + "evidence": [ + { + "description": "The bedroom is on my left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The table in the living room is about two steps away from the bedroom.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The table in the living room is about two steps away from the bedroom.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "I walked from inside the house to the courtyard.", + "timestamps": { + "start": "00:00:58", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:11:11", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right and walk.", + "B": "You should turn left and walk—after about two steps, you’ll reach the bedroom.", + "C": "Go straight two steps.", + "D": "You should turn left and walk—after about four steps, you’ll reach the bedroom.", + "E": "You should turn left and walk—after about one step, you’ll reach the bedroom." + }, + "answer_label": "B", + "answer_text": "You should turn left and walk—after about two steps, you’ll reach the bedroom.", + "q_id": "egolife_A1_D6_042_4", + "correct_answer": "You should turn left and walk—after about two steps, you’ll reach the bedroom.", + "original_options": { + "A": "Turn right and walk.", + "B": "You should turn left and walk—after about two steps, you’ll reach the bedroom.", + "C": "Go straight two steps.", + "D": "You should turn left and walk—after about four steps, you’ll reach the bedroom.", + "E": "You should turn left and walk—after about one step, you’ll reach the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the notebook on the living room table from me?", + "answer": "It's right in front of you on your left now.", + "question_chinese": "客厅桌子上的笔记本在我的什么方向？", + "answer_chinese": "它现在在你的左前方。", + "evidence": [ + { + "description": "There is a laptop directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:06:34", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's right in front of you on your left now.", + "B": "It's in front of you on your right.", + "C": "It's behind you on your left.", + "D": "It's directly in front of you, but shifted to your right rather than on your left.", + "E": "It's behind you and slightly to the right, not in front of you on the left." + }, + "answer_label": "A", + "answer_text": "It's right in front of you on your left now.", + "q_id": "egolife_A1_D6_042_1", + "correct_answer": "It's right in front of you on your left now.", + "original_options": { + "A": "It's right in front of you on your left now.", + "B": "It's in front of you on your right.", + "C": "It's behind you on your left.", + "D": "It's directly in front of you, but shifted to your right rather than on your left.", + "E": "It's behind you and slightly to the right, not in front of you on the left." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the notebook on the living room table relative to me?", + "answer": "It's on your left now.", + "question_chinese": "客厅桌子上的笔记本在我的什么方向？", + "answer_chinese": "它现在在你的左侧。", + "evidence": [ + { + "description": "There is a laptop directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:11:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's on your right now.", + "B": "It's directly in front of you.", + "C": "It's on your left now.", + "D": "It's behind you right now, not positioned on your left side as you look forward.", + "E": "It's off to your right side at the moment, rather than being located on your left." + }, + "answer_label": "C", + "answer_text": "It's on your left now.", + "q_id": "egolife_A1_D6_042_2", + "correct_answer": "It's on your left now.", + "original_options": { + "A": "It's on your right now.", + "B": "It's directly in front of you.", + "C": "It's on your left now.", + "D": "It's behind you right now, not positioned on your left side as you look forward.", + "E": "It's off to your right side at the moment, rather than being located on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_9", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the microwave in the living room from me?", + "answer": "The microwave is in front of you on the left.", + "question_chinese": "客厅里的微波炉在我的什么方向？", + "answer_chinese": "微波炉在你的左前方。", + "evidence": [ + { + "description": "There are two microwaves stacked on top of each other in front of me on the right.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:06:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's in front of you on the right.", + "B": "It's behind you on the left.", + "C": "The microwave is directly behind you on the right, not in front of you on the left.", + "D": "The microwave is in front of you on the left.", + "E": "The microwave sits in front of you but on your right side rather than the left." + }, + "answer_label": "D", + "answer_text": "The microwave is in front of you on the left.", + "q_id": "egolife_A1_D6_042_9", + "correct_answer": "The microwave is in front of you on the left.", + "original_options": { + "A": "It's in front of you on the right.", + "B": "It's behind you on the left.", + "C": "The microwave is directly behind you on the right, not in front of you on the left.", + "D": "The microwave is in front of you on the left.", + "E": "The microwave sits in front of you but on your right side rather than the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_10", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the microwave in the living room from me?", + "answer": "The microwave is on your right side.", + "question_chinese": "客厅里的微波炉在我的什么方向？", + "answer_chinese": "微波炉在你的右侧。", + "evidence": [ + { + "description": "There are two microwaves stacked on top of each other in front of me on the right.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:10:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The microwave is on your left side.", + "B": "The microwave is directly behind you.", + "C": "The microwave is on your right side.", + "D": "The microwave is in front of you, not positioned on your right side as you face forward.", + "E": "The microwave is behind and slightly to your left, rather than being on your right side." + }, + "answer_label": "C", + "answer_text": "The microwave is on your right side.", + "q_id": "egolife_A1_D6_042_10", + "correct_answer": "The microwave is on your right side.", + "original_options": { + "A": "The microwave is on your left side.", + "B": "The microwave is directly behind you.", + "C": "The microwave is on your right side.", + "D": "The microwave is in front of you, not positioned on your right side as you face forward.", + "E": "The microwave is behind and slightly to your left, rather than being on your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_5", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are currently in the house?", + "answer": "Right now, there are three people in the house.", + "question_chinese": "房子里现在有几个人？", + "answer_chinese": "房子里一共有3个人，现在。", + "evidence": [ + { + "description": "At the moment, there are two people in the house including me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Two people entered the house, and I walked out of it.", + "timestamps": { + "start": "00:00:53", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are two people in the house.", + "B": "Right now, there are three people in the house.", + "C": "There are four people in the house.", + "D": "Right now, there are five people in the house, not three as you might think.", + "E": "At the moment, seven people are in the house, which is more than the correct total." + }, + "answer_label": "B", + "answer_text": "Right now, there are three people in the house.", + "q_id": "egolife_A1_D6_042_5", + "correct_answer": "Right now, there are three people in the house.", + "original_options": { + "A": "There are two people in the house.", + "B": "Right now, there are three people in the house.", + "C": "There are four people in the house.", + "D": "Right now, there are five people in the house, not three as you might think.", + "E": "At the moment, seven people are in the house, which is more than the correct total." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_042", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_042_6", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people are currently in the house?", + "answer": "There are six people in the house now in total.", + "question_chinese": "房子里现在有几个人？", + "answer_chinese": "现在房子里一共有6个人。", + "evidence": [ + { + "description": "At the moment, there are two people in the house including me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Two people entered the house, and I walked out of it.", + "timestamps": { + "start": "00:00:53", + "end": "00:01:05" + } + } + ], + "question_timestamps": "00:11:17", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are five people total.", + "B": "There are six people in the house now in total.", + "C": "There are seven people total.", + "D": "There are four people in the house now in total, not six at this moment.", + "E": "There are eight people in the house now in total, which is higher than the actual count." + }, + "answer_label": "B", + "answer_text": "There are six people in the house now in total.", + "q_id": "egolife_A1_D6_042_6", + "correct_answer": "There are six people in the house now in total.", + "original_options": { + "A": "There are five people total.", + "B": "There are six people in the house now in total.", + "C": "There are seven people total.", + "D": "There are four people in the house now in total, not six at this moment.", + "E": "There are eight people in the house now in total, which is higher than the actual count." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D6_043-mcq.json b/QAs/egolife_A1_D6_043-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..98b4df5da5b2d207ad13251353c06f4db681bf77 --- /dev/null +++ b/QAs/egolife_A1_D6_043-mcq.json @@ -0,0 +1,1056 @@ +[ + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_1_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many people have appeared in my field of view in total?", + "answer": "A total of four people appeared in your field of view. All four are to your left.", + "question_chinese": "到目前为止，我总共视野里出现过几个人？", + "answer_chinese": "你总共视野出现了四个人。这四个人在你的左边。", + "evidence": [ + { + "description": "Four people have been seen so far.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "A total of four people appeared in your field of view. All four are to your left.", + "B": "Three people appeared, all on your left.", + "C": "Five people appeared, all on your left.", + "D": "A total of two people appeared in your field of view, and both of them are to your left.", + "E": "A total of six people appeared in your field of view, and all six are positioned to your left." + }, + "answer_label": "A", + "answer_text": "A total of four people appeared in your field of view. All four are to your left.", + "q_id": "egolife_A1_D6_043_1_T1", + "correct_answer": "A total of four people appeared in your field of view. All four are to your left.", + "original_options": { + "A": "A total of four people appeared in your field of view. All four are to your left.", + "B": "Three people appeared, all on your left.", + "C": "Five people appeared, all on your left.", + "D": "A total of two people appeared in your field of view, and both of them are to your left.", + "E": "A total of six people appeared in your field of view, and all six are positioned to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_1_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many people have appeared in my field of view in total?", + "answer": "A total of one person appeared in your field of view. That person is to your left.", + "question_chinese": "到目前为止，我总共视野里出现过几个人？", + "answer_chinese": "你总共视野出现了一个人。这一个人在你的左边。", + "evidence": [ + { + "description": "One people have been seen so far.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero people appeared in your view.", + "B": "A total of one person appeared in your field of view. That person is to your left.", + "C": "Two people appeared, both on your left.", + "D": "A total of three people appeared in your field of view, and all three are to your left.", + "E": "A total of four people appeared in your field of view, with all four located to your left side." + }, + "answer_label": "B", + "answer_text": "A total of one person appeared in your field of view. That person is to your left.", + "q_id": "egolife_A1_D6_043_1_T2", + "correct_answer": "A total of one person appeared in your field of view. That person is to your left.", + "original_options": { + "A": "Zero people appeared in your view.", + "B": "A total of one person appeared in your field of view. That person is to your left.", + "C": "Two people appeared, both on your left.", + "D": "A total of three people appeared in your field of view, and all three are to your left.", + "E": "A total of four people appeared in your field of view, with all four located to your left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many people have appeared in my field of vision?", + "answer": "A total of four people have appeared.", + "question_chinese": "到目前为止，我总共视野里出现过几个人。", + "answer_chinese": "总共出现了4个人。", + "evidence": [ + { + "description": "Four people have been seen so far.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:05:18", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "A total of five people have appeared.", + "B": "A total of three people have appeared.", + "C": "A total of four people have appeared.", + "D": "A total of two people have appeared in your view so far.", + "E": "A total of six people have appeared in your field of vision so far." + }, + "answer_label": "C", + "answer_text": "A total of four people have appeared.", + "q_id": "egolife_A1_D6_043_1", + "correct_answer": "A total of four people have appeared.", + "original_options": { + "A": "A total of five people have appeared.", + "B": "A total of three people have appeared.", + "C": "A total of four people have appeared.", + "D": "A total of two people have appeared in your view so far.", + "E": "A total of six people have appeared in your field of vision so far." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_2_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many people have appeared in my field of view in total?", + "answer": "A total of eight people appeared in your field of view. All eight are in front of you.", + "question_chinese": "到目前为止，我总共视野里出现过几个人？", + "answer_chinese": "你总共视野出现了八个人。这八个人在你的前面。", + "evidence": [ + { + "description": "Four people have been seen so far.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "This is the fifth person.", + "timestamps": { + "start": "00:06:57", + "end": "00:06:58" + } + }, + { + "description": "Two more people have appeared.", + "timestamps": { + "start": "00:08:28", + "end": "00:08:30" + } + }, + { + "description": "This is the eighth person.", + "timestamps": { + "start": "00:08:55", + "end": "00:08:57" + } + } + ], + "question_timestamps": "00:08:59", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Seven people appeared, all in front.", + "B": "Nine people appeared, all in front.", + "C": "A total of eight people appeared in your field of view. All eight are in front of you.", + "D": "A total of six people appeared in your field of view, and all six are in front of you.", + "E": "A total of ten people appeared in your field of view, and all ten are in front of you." + }, + "answer_label": "C", + "answer_text": "A total of eight people appeared in your field of view. All eight are in front of you.", + "q_id": "egolife_A1_D6_043_2_T2", + "correct_answer": "A total of eight people appeared in your field of view. All eight are in front of you.", + "original_options": { + "A": "Seven people appeared, all in front.", + "B": "Nine people appeared, all in front.", + "C": "A total of eight people appeared in your field of view. All eight are in front of you.", + "D": "A total of six people appeared in your field of view, and all six are in front of you.", + "E": "A total of ten people appeared in your field of view, and all ten are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_2_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many people have appeared in my field of view in total?", + "answer": "A total of nine people appeared in your field of view. All nine are in front of you.", + "question_chinese": "到目前为止，我总共视野里出现过几个人？", + "answer_chinese": "你总共视野出现了九个人。这九个人在你的前面。", + "evidence": [ + { + "description": "Four people have been seen so far.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "This is the fifth person.", + "timestamps": { + "start": "00:06:57", + "end": "00:06:58" + } + }, + { + "description": "Two more people have appeared.", + "timestamps": { + "start": "00:08:28", + "end": "00:08:30" + } + }, + { + "description": "This is the eighth person.", + "timestamps": { + "start": "00:08:55", + "end": "00:08:57" + } + }, + { + "description": "This is the ninth person.", + "timestamps": { + "start": "00:10:12", + "end": "00:10:14" + } + } + ], + "question_timestamps": "00:10:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Eight people appeared, all in front.", + "B": "A total of nine people appeared in your field of view. All nine are in front of you.", + "C": "Ten people appeared, all in front.", + "D": "A total of seven people appeared in your field of view, and all seven are in front of you.", + "E": "A total of eleven people appeared in your field of view, and all eleven are in front of you." + }, + "answer_label": "B", + "answer_text": "A total of nine people appeared in your field of view. All nine are in front of you.", + "q_id": "egolife_A1_D6_043_2_T1", + "correct_answer": "A total of nine people appeared in your field of view. All nine are in front of you.", + "original_options": { + "A": "Eight people appeared, all in front.", + "B": "A total of nine people appeared in your field of view. All nine are in front of you.", + "C": "Ten people appeared, all in front.", + "D": "A total of seven people appeared in your field of view, and all seven are in front of you.", + "E": "A total of eleven people appeared in your field of view, and all eleven are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many people have appeared in my view?", + "answer": "A total of nine people have appeared.", + "question_chinese": "到目前为止，我总共视野里出现过几个人。", + "answer_chinese": "一共出现了9个人。", + "evidence": [ + { + "description": "Four people have been seen so far.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "This is the fifth person.", + "timestamps": { + "start": "00:06:57", + "end": "00:06:58" + } + }, + { + "description": "Two more people have appeared.", + "timestamps": { + "start": "00:08:28", + "end": "00:08:30" + } + }, + { + "description": "This is the eighth person.", + "timestamps": { + "start": "00:08:55", + "end": "00:08:57" + } + }, + { + "description": "This is the ninth person.", + "timestamps": { + "start": "00:10:12", + "end": "00:10:14" + } + } + ], + "question_timestamps": "00:11:46", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "A total of eight people have appeared.", + "B": "A total of ten people have appeared.", + "C": "A total of seven people have appeared in your view so far.", + "D": "A total of eleven people have appeared in your field of vision so far.", + "E": "A total of nine people have appeared." + }, + "answer_label": "E", + "answer_text": "A total of nine people have appeared.", + "q_id": "egolife_A1_D6_043_2", + "correct_answer": "A total of nine people have appeared.", + "original_options": { + "A": "A total of eight people have appeared.", + "B": "A total of ten people have appeared.", + "C": "A total of seven people have appeared in your view so far.", + "D": "A total of eleven people have appeared in your field of vision so far.", + "E": "A total of nine people have appeared." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_5", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bedroom to get something, how should I get there?", + "answer": "You should stand up, then turn left and walk two steps to enter the bedroom.", + "question_chinese": "如果我想去卧室拿东西，我应该怎么走？", + "answer_chinese": "你应当起身，然后向左转走两步就能进入卧室了。", + "evidence": [ + { + "description": "The bedroom is on my left.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:02" + } + }, + { + "description": "The bedroom is about two steps away from the stool I'm sitting on.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:05:34", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Stand up, turn right, walk two steps.", + "B": "Stand up, turn left, walk one step.", + "C": "Stand up, then turn right and walk one step to enter the bedroom.", + "D": "You should stand up, then turn left and walk two steps to enter the bedroom.", + "E": "Stand up, then turn left and walk three steps before entering the bedroom." + }, + "answer_label": "D", + "answer_text": "You should stand up, then turn left and walk two steps to enter the bedroom.", + "q_id": "egolife_A1_D6_043_5", + "correct_answer": "You should stand up, then turn left and walk two steps to enter the bedroom.", + "original_options": { + "A": "Stand up, turn right, walk two steps.", + "B": "Stand up, turn left, walk one step.", + "C": "Stand up, then turn right and walk one step to enter the bedroom.", + "D": "You should stand up, then turn left and walk two steps to enter the bedroom.", + "E": "Stand up, then turn left and walk three steps before entering the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the bedroom to get something, how should I get there?", + "answer": "You should walk four steps forward to your left, and you will be able to enter the bedroom.", + "question_chinese": "如果我想去卧室拿东西，我应该怎么走？", + "answer_chinese": "你应当向左前方走4步的距离就能进入卧室了。", + "evidence": [ + { + "description": "The bedroom is on my left.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:02" + } + }, + { + "description": "The bedroom is about two steps away from the stool I'm sitting on.", + "timestamps": { + "start": "00:01:01", + "end": "00:01:02" + } + } + ], + "question_timestamps": "00:11:29", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk four steps backward to your left.", + "B": "Walk four steps forward to your right.", + "C": "Walk three steps forward to your right, and you will be able to enter the bedroom.", + "D": "You should walk four steps forward to your left, and you will be able to enter the bedroom.", + "E": "Walk five steps forward to your left, and you will be able to enter the bedroom." + }, + "answer_label": "D", + "answer_text": "You should walk four steps forward to your left, and you will be able to enter the bedroom.", + "q_id": "egolife_A1_D6_043_6", + "correct_answer": "You should walk four steps forward to your left, and you will be able to enter the bedroom.", + "original_options": { + "A": "Walk four steps backward to your left.", + "B": "Walk four steps forward to your right.", + "C": "Walk three steps forward to your right, and you will be able to enter the bedroom.", + "D": "You should walk four steps forward to your left, and you will be able to enter the bedroom.", + "E": "Walk five steps forward to your left, and you will be able to enter the bedroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_11", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the flower on the living room table from me?", + "answer": "That pot of flowers is on your right.", + "question_chinese": "客厅桌子上的花在我什么方向？", + "answer_chinese": "那盆花在你的右侧。", + "evidence": [ + { + "description": "There is a pot of pink flowers on the table in front of my left.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:08:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your left.", + "B": "In front of you.", + "C": "The pot of flowers is behind you, on the far side of the living room table.", + "D": "That pot of flowers is on your right.", + "E": "The pot of flowers is straight ahead, centered on the living room table in front of you." + }, + "answer_label": "D", + "answer_text": "That pot of flowers is on your right.", + "q_id": "egolife_A1_D6_043_11", + "correct_answer": "That pot of flowers is on your right.", + "original_options": { + "A": "On your left.", + "B": "In front of you.", + "C": "The pot of flowers is behind you, on the far side of the living room table.", + "D": "That pot of flowers is on your right.", + "E": "The pot of flowers is straight ahead, centered on the living room table in front of you." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_12", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the flower on the living room table from me?", + "answer": "That pot of flowers is on your left.", + "question_chinese": "客厅桌子上的花在我什么方向？", + "answer_chinese": "那盆花在你的左侧。", + "evidence": [ + { + "description": "There is a pot of pink flowers on the table in front of my left.", + "timestamps": { + "start": "00:00:33", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:11:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your right.", + "B": "That pot of flowers is on your left.", + "C": "In front of you.", + "D": "That pot of flowers is behind you, not on your left side.", + "E": "That pot of flowers is straight ahead of you rather than to the left." + }, + "answer_label": "B", + "answer_text": "That pot of flowers is on your left.", + "q_id": "egolife_A1_D6_043_12", + "correct_answer": "That pot of flowers is on your left.", + "original_options": { + "A": "On your right.", + "B": "That pot of flowers is on your left.", + "C": "In front of you.", + "D": "That pot of flowers is behind you, not on your left side.", + "E": "That pot of flowers is straight ahead of you rather than to the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_9", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the living room window from me?", + "answer": "The window is directly in front of you.", + "question_chinese": "客厅的窗户在我的什么方向？", + "answer_chinese": "窗户在你的正前方。", + "evidence": [ + { + "description": "There is a window directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:06:23", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is to your left.", + "B": "The window is behind you.", + "C": "The window is slightly to your right, not directly in front of you where you are facing.", + "D": "The window sits off to your left side, so you would need to turn to see it clearly.", + "E": "The window is directly in front of you." + }, + "answer_label": "E", + "answer_text": "The window is directly in front of you.", + "q_id": "egolife_A1_D6_043_9", + "correct_answer": "The window is directly in front of you.", + "original_options": { + "A": "The window is to your left.", + "B": "The window is behind you.", + "C": "The window is slightly to your right, not directly in front of you where you are facing.", + "D": "The window sits off to your left side, so you would need to turn to see it clearly.", + "E": "The window is directly in front of you." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_10", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the living room window from me?", + "answer": "The living room window is in front of your right.", + "question_chinese": "客厅的窗户在我的什么方向？", + "answer_chinese": "客厅的窗户在你的右前方。", + "evidence": [ + { + "description": "There is a window directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:09:04", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The living room window is in front of your right.", + "B": "The window is directly in front.", + "C": "The window is behind your left.", + "D": "The window is in front of your left side, so it isn’t angled toward your right.", + "E": "The window is straight behind you, meaning you’d have to turn around to face it." + }, + "answer_label": "A", + "answer_text": "The living room window is in front of your right.", + "q_id": "egolife_A1_D6_043_10", + "correct_answer": "The living room window is in front of your right.", + "original_options": { + "A": "The living room window is in front of your right.", + "B": "The window is directly in front.", + "C": "The window is behind your left.", + "D": "The window is in front of your left side, so it isn’t angled toward your right.", + "E": "The window is straight behind you, meaning you’d have to turn around to face it." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_5_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is my laptop relative to me?", + "answer": "Your laptop is in front of you.", + "question_chinese": "我的笔记本电脑在我的什么方向？", + "answer_chinese": "你的笔记本在你的前面。", + "evidence": [ + { + "description": "The laptop in front of me is mine.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:00:45", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your laptop is to your left.", + "B": "Your laptop is behind you.", + "C": "Your laptop is off to your right side, not positioned straight ahead of your view.", + "D": "Your laptop is in front of you.", + "E": "Your laptop is behind your left side, so it isn’t in front of you where you’re facing." + }, + "answer_label": "D", + "answer_text": "Your laptop is in front of you.", + "q_id": "egolife_A1_D6_043_5_T2", + "correct_answer": "Your laptop is in front of you.", + "original_options": { + "A": "Your laptop is to your left.", + "B": "Your laptop is behind you.", + "C": "Your laptop is off to your right side, not positioned straight ahead of your view.", + "D": "Your laptop is in front of you.", + "E": "Your laptop is behind your left side, so it isn’t in front of you where you’re facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_5_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is my laptop relative to me?", + "answer": "Your laptop is on the right side.", + "question_chinese": "我的笔记本电脑在我的什么方向？", + "answer_chinese": "你的笔记本在你的右侧。", + "evidence": [ + { + "description": "The laptop in front of me is mine.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:07:51", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Your laptop is on the left.", + "B": "Your laptop is in front of you.", + "C": "Your laptop is behind you, so it isn’t sitting on your right-hand side.", + "D": "Your laptop is slightly to your left and a bit behind, not located on the right side.", + "E": "Your laptop is on the right side." + }, + "answer_label": "E", + "answer_text": "Your laptop is on the right side.", + "q_id": "egolife_A1_D6_043_5_T1", + "correct_answer": "Your laptop is on the right side.", + "original_options": { + "A": "Your laptop is on the left.", + "B": "Your laptop is in front of you.", + "C": "Your laptop is behind you, so it isn’t sitting on your right-hand side.", + "D": "Your laptop is slightly to your left and a bit behind, not located on the right side.", + "E": "Your laptop is on the right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_8", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In what direction is my laptop?", + "answer": "Your laptop is behind your left side.", + "question_chinese": "我的笔记本电脑在我的什么方向？", + "answer_chinese": "你的笔记本电脑在你的左后方。", + "evidence": [ + { + "description": "The laptop in front of me is mine.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:43" + } + } + ], + "question_timestamps": "00:11:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Your laptop is in front of you.", + "B": "Your laptop is behind your right.", + "C": "Your laptop is behind your left side.", + "D": "Your laptop is on your left side but in front of you, rather than positioned behind your left.", + "E": "Your laptop is behind you and slightly to the right, so it’s not behind your left side." + }, + "answer_label": "C", + "answer_text": "Your laptop is behind your left side.", + "q_id": "egolife_A1_D6_043_8", + "correct_answer": "Your laptop is behind your left side.", + "original_options": { + "A": "Your laptop is in front of you.", + "B": "Your laptop is behind your right.", + "C": "Your laptop is behind your left side.", + "D": "Your laptop is on your left side but in front of you, rather than positioned behind your left.", + "E": "Your laptop is behind you and slightly to the right, so it’s not behind your left side." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_3", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the door to the bedroom closer to me, or is the door to the courtyard closer to me?", + "answer": "The door leading to the bedroom is closer to you.", + "question_chinese": "是到卧室的门离我更近，还是到庭院的门离我更近？", + "answer_chinese": "到卧室的门离你更近。", + "evidence": [ + { + "description": "The door in front of me on the right leads to the courtyard.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "The door on my left leads to the bedroom.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:06:42", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The door leading to the bedroom is closer to you.", + "B": "The door leading to the courtyard is closer to you." + }, + "answer_label": "A", + "answer_text": "The door leading to the bedroom is closer to you.", + "q_id": "egolife_A1_D6_043_3", + "correct_answer": "The door leading to the bedroom is closer to you.", + "original_options": { + "A": "The door leading to the bedroom is closer to you.", + "B": "The door leading to the courtyard is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D6_043", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_043_4", + "qid_num": 17, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the door to the bedroom closer to me, or is the door to the courtyard closer to me?", + "answer": "The door leading to the courtyard is closer to you.", + "question_chinese": "是到卧室的门离我更近，还是到庭院的门离我更近？", + "answer_chinese": "到庭院的门离你更近。", + "evidence": [ + { + "description": "The door in front of me on the right leads to the courtyard.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "The door on my left leads to the bedroom.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:08:07", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "The door leading to the bedroom is closer to you.", + "B": "The door leading to the courtyard is closer to you." + }, + "answer_label": "B", + "answer_text": "The door leading to the courtyard is closer to you.", + "q_id": "egolife_A1_D6_043_4", + "correct_answer": "The door leading to the courtyard is closer to you.", + "original_options": { + "A": "The door leading to the bedroom is closer to you.", + "B": "The door leading to the courtyard is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + } +] diff --git a/QAs/egolife_A1_D6_044-mcq.json b/QAs/egolife_A1_D6_044-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3d0928f2da276688a8ca95e623556265b1f97f91 --- /dev/null +++ b/QAs/egolife_A1_D6_044-mcq.json @@ -0,0 +1,1451 @@ +[ + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_8_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the first-floor bedroom?", + "answer": "The bedroom is on the left side.", + "question_chinese": "一楼的卧室在我的什么方向？", + "answer_chinese": "卧室在你的左侧", + "evidence": [ + { + "description": "The bedroom on the first floor is in front of my left.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bedroom is on the left side.", + "B": "The bedroom is on the right.", + "C": "The bedroom is in front of you.", + "D": "The bedroom is behind you, so it isn’t positioned on your left side from where you stand.", + "E": "The bedroom is on your right side, meaning you would turn right instead of left to reach it." + }, + "answer_label": "A", + "answer_text": "The bedroom is on the left side.", + "q_id": "egolife_A1_D6_044_8_T2", + "correct_answer": "The bedroom is on the left side.", + "original_options": { + "A": "The bedroom is on the left side.", + "B": "The bedroom is on the right.", + "C": "The bedroom is in front of you.", + "D": "The bedroom is behind you, so it isn’t positioned on your left side from where you stand.", + "E": "The bedroom is on your right side, meaning you would turn right instead of left to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_8_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the first-floor bedroom?", + "answer": "The bedroom is in front of you.", + "question_chinese": "一楼的卧室在我的什么方向？", + "answer_chinese": "卧室在你的正前方。", + "evidence": [ + { + "description": "The bedroom is in front of you.", + "timestamps": { + "start": "00:00:30", + "end": "00:00:31" + } + } + ], + "question_timestamps": "00:00:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bedroom is in front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "The bedroom is to your right as you stand in the hallway now.", + "E": "The bedroom is above you, on a higher level from your current spot." + }, + "answer_label": "A", + "answer_text": "The bedroom is in front of you.", + "q_id": "egolife_A1_D6_044_8_T1", + "correct_answer": "The bedroom is in front of you.", + "original_options": { + "A": "The bedroom is in front of you.", + "B": "Behind you.", + "C": "To your left.", + "D": "The bedroom is to your right as you stand in the hallway now.", + "E": "The bedroom is above you, on a higher level from your current spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_11", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the bedroom on the first floor from me?", + "answer": "The bedroom is on your left.", + "question_chinese": "一楼的卧室在我的什么方向？", + "answer_chinese": "卧室在你的左侧", + "evidence": [ + { + "description": "The bedroom on the first floor is in front of my left.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:06:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your right.", + "C": "The bedroom is behind you from where you are standing right now.", + "D": "The bedroom is straight ahead of you as you face forward right now.", + "E": "The bedroom is on your left." + }, + "answer_label": "E", + "answer_text": "The bedroom is on your left.", + "q_id": "egolife_A1_D6_044_11", + "correct_answer": "The bedroom is on your left.", + "original_options": { + "A": "In front of you.", + "B": "To your right.", + "C": "The bedroom is behind you from where you are standing right now.", + "D": "The bedroom is straight ahead of you as you face forward right now.", + "E": "The bedroom is on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_12", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the bedroom on the first floor from my location?", + "answer": "The bedroom is behind and below you.", + "question_chinese": "一楼的卧室在我的什么方向？", + "answer_chinese": "卧室在你的后下方。", + "evidence": [ + { + "description": "The bedroom on the first floor is in front of my left.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:11:31", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front and above you.", + "B": "The bedroom is behind and below you.", + "C": "Behind and above you.", + "D": "The bedroom is in front of you but lower down than your current position.", + "E": "The bedroom is behind you, but it is at the same level as you." + }, + "answer_label": "B", + "answer_text": "The bedroom is behind and below you.", + "q_id": "egolife_A1_D6_044_12", + "correct_answer": "The bedroom is behind and below you.", + "original_options": { + "A": "In front and above you.", + "B": "The bedroom is behind and below you.", + "C": "Behind and above you.", + "D": "The bedroom is in front of you but lower down than your current position.", + "E": "The bedroom is behind you, but it is at the same level as you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_5", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many people have appeared in my line of sight?", + "answer": "A total of five people have appeared so far.", + "question_chinese": "到目前为止，我的视野里一共出现过几个人？", + "answer_chinese": "一共出现过5个人，到目前为止。", + "evidence": [ + { + "description": "There are four people in my field of vision now in total.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "This is the fifth person.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Four people.", + "B": "Six people.", + "C": "A total of five people have appeared so far.", + "D": "A total of three people have appeared so far in your line of sight.", + "E": "A total of eight people have appeared so far within your field of view." + }, + "answer_label": "C", + "answer_text": "A total of five people have appeared so far.", + "q_id": "egolife_A1_D6_044_5", + "correct_answer": "A total of five people have appeared so far.", + "original_options": { + "A": "Four people.", + "B": "Six people.", + "C": "A total of five people have appeared so far.", + "D": "A total of three people have appeared so far in your line of sight.", + "E": "A total of eight people have appeared so far within your field of view." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_6", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "So far, how many people have appeared in my sight?", + "answer": "A total of seven people have appeared so far.", + "question_chinese": "到目前为止，我的视野里一共出现过几个人？", + "answer_chinese": "到目前为止，总共出现过7个人。", + "evidence": [ + { + "description": "There are four people in my field of vision now in total.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "This is the fifth person.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "This is the sixth person now.", + "timestamps": { + "start": "00:03:53", + "end": "00:03:55" + } + }, + { + "description": "This is the seventh person.", + "timestamps": { + "start": "00:04:56", + "end": "00:04:58" + } + } + ], + "question_timestamps": "00:07:05", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "A total of seven people have appeared so far.", + "B": "Six people.", + "C": "Eight people.", + "D": "A total of five people have appeared so far in your sight.", + "E": "A total of nine people have appeared so far in your field of view." + }, + "answer_label": "A", + "answer_text": "A total of seven people have appeared so far.", + "q_id": "egolife_A1_D6_044_6", + "correct_answer": "A total of seven people have appeared so far.", + "original_options": { + "A": "A total of seven people have appeared so far.", + "B": "Six people.", + "C": "Eight people.", + "D": "A total of five people have appeared so far in your sight.", + "E": "A total of nine people have appeared so far in your field of view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_3_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many people have appeared in my field of view?", + "answer": "Up to now, four people have appeared in your view, and all four are in front of you.", + "question_chinese": "到目前为止，我的视野里出现过几个人？", + "answer_chinese": "到目前为止,你的视野里出现过四个人，四个人在你的正前方。", + "evidence": [ + { + "description": "So far, there have been four people in your field of vision.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Three people, all in front.", + "B": "Five people, all in front.", + "C": "Up to now, two people have appeared in your view, and both are in front of you.", + "D": "Up to now, six people have appeared in your view, and all six are in front of you.", + "E": "Up to now, four people have appeared in your view, and all four are in front of you." + }, + "answer_label": "E", + "answer_text": "Up to now, four people have appeared in your view, and all four are in front of you.", + "q_id": "egolife_A1_D6_044_3_T2", + "correct_answer": "Up to now, four people have appeared in your view, and all four are in front of you.", + "original_options": { + "A": "Three people, all in front.", + "B": "Five people, all in front.", + "C": "Up to now, two people have appeared in your view, and both are in front of you.", + "D": "Up to now, six people have appeared in your view, and all six are in front of you.", + "E": "Up to now, four people have appeared in your view, and all four are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_3_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many people have appeared in my field of view?", + "answer": "Up to now, five people have appeared in your view, and all five are in front of you.", + "question_chinese": "到目前为止，我的视野里出现过几个人？", + "answer_chinese": "到目前为止，你的视野里出现过五个人，五个人在你的正前方。", + "evidence": [ + { + "description": "So far, there have been five people in your field of vision.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Four people, all in front.", + "B": "Six people, all in front.", + "C": "Up to now, three people have appeared in your view, and all three are in front of you.", + "D": "Up to now, five people have appeared in your view, and all five are in front of you.", + "E": "Up to now, seven people have appeared in your view, and all seven are in front of you." + }, + "answer_label": "D", + "answer_text": "Up to now, five people have appeared in your view, and all five are in front of you.", + "q_id": "egolife_A1_D6_044_3_T1", + "correct_answer": "Up to now, five people have appeared in your view, and all five are in front of you.", + "original_options": { + "A": "Four people, all in front.", + "B": "Six people, all in front.", + "C": "Up to now, three people have appeared in your view, and all three are in front of you.", + "D": "Up to now, five people have appeared in your view, and all five are in front of you.", + "E": "Up to now, seven people have appeared in your view, and all seven are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_3_T3", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many people have appeared in my field of view?", + "answer": "Up to now, three people have appeared in your view, and all three are in front of you.", + "question_chinese": "到目前为止，我的视野里出现过几个人？", + "answer_chinese": "到目前为止，你的视野里出现过三个人，三个人在你的正前方。", + "evidence": [ + { + "description": "So far, there have been three people in your field of vision.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:21", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Up to now, three people have appeared in your view, and all three are in front of you.", + "B": "Two people, both in front.", + "C": "Four people, all in front.", + "D": "Up to now, one person has appeared in your view, and that person is in front of you.", + "E": "Up to now, five people have appeared in your view, and all five are in front of you." + }, + "answer_label": "A", + "answer_text": "Up to now, three people have appeared in your view, and all three are in front of you.", + "q_id": "egolife_A1_D6_044_3_T3", + "correct_answer": "Up to now, three people have appeared in your view, and all three are in front of you.", + "original_options": { + "A": "Up to now, three people have appeared in your view, and all three are in front of you.", + "B": "Two people, both in front.", + "C": "Four people, all in front.", + "D": "Up to now, one person has appeared in your view, and that person is in front of you.", + "E": "Up to now, five people have appeared in your view, and all five are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_3", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to go to the bedroom now, how should I get there?", + "answer": "You should turn left and then walk about four steps to get into the bedroom.", + "question_chinese": "如果现在我想去卧室，我应该怎么走？", + "answer_chinese": "你应当向左转，然后走4步左右的距离就能进入卧室了。", + "evidence": [ + { + "description": "The door leading to the bedroom is in front of my left.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "The distance from the bedroom to the living room table is about four steps.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:05:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You should turn left and then walk about four steps to get into the bedroom.", + "B": "Turn right and walk four steps.", + "C": "Turn left and walk five steps.", + "D": "Turn left, walk about four steps, and you will end up in the hallway instead of the bedroom.", + "E": "Turn left and keep going for around seven steps before you reach the bedroom doorway." + }, + "answer_label": "A", + "answer_text": "You should turn left and then walk about four steps to get into the bedroom.", + "q_id": "egolife_A1_D6_044_3", + "correct_answer": "You should turn left and then walk about four steps to get into the bedroom.", + "original_options": { + "A": "You should turn left and then walk about four steps to get into the bedroom.", + "B": "Turn right and walk four steps.", + "C": "Turn left and walk five steps.", + "D": "Turn left, walk about four steps, and you will end up in the hallway instead of the bedroom.", + "E": "Turn left and keep going for around seven steps before you reach the bedroom doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_4", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "If I want to go to the bedroom now, how should I get there?", + "answer": "You should stand up, turn right, enter the stairwell, go down to the first floor, then exit the door and walk forward to your left for about six steps, and you’ll reach the bedroom.", + "question_chinese": "如果现在我想去卧室，我应该怎么走？", + "answer_chinese": "你应当起身向右转，进入楼梯间，然后下到一楼，随后出门向左前方走，6步左右的距离就能进入卧室了。", + "evidence": [ + { + "description": "The door leading to the bedroom is in front of my left.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "The distance from the bedroom to the living room table is about four steps.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "I went up from the first floor to the second floor.", + "timestamps": { + "start": "00:07:34", + "end": "00:08:04" + } + } + ], + "question_timestamps": "00:10:14", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Stand up and turn left first.", + "B": "Go down, then walk straight six steps.", + "C": "Stand up, turn right, enter the stairwell, go down to the second floor, then exit and walk left six steps.", + "D": "You should stand up, turn right, enter the stairwell, go down to the first floor, then exit the door and walk forward to your left for about six steps, and you’ll reach the bedroom.", + "E": "Stand up, turn right, enter the stairwell, go down to the first floor, exit the door, then walk forward about ten steps." + }, + "answer_label": "D", + "answer_text": "You should stand up, turn right, enter the stairwell, go down to the first floor, then exit the door and walk forward to your left for about six steps, and you’ll reach the bedroom.", + "q_id": "egolife_A1_D6_044_4", + "correct_answer": "You should stand up, turn right, enter the stairwell, go down to the first floor, then exit the door and walk forward to your left for about six steps, and you’ll reach the bedroom.", + "original_options": { + "A": "Stand up and turn left first.", + "B": "Go down, then walk straight six steps.", + "C": "Stand up, turn right, enter the stairwell, go down to the second floor, then exit and walk left six steps.", + "D": "You should stand up, turn right, enter the stairwell, go down to the first floor, then exit the door and walk forward to your left for about six steps, and you’ll reach the bedroom.", + "E": "Stand up, turn right, enter the stairwell, go down to the first floor, exit the door, then walk forward about ten steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 20, + 22 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_7", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the notebook on the coffee table relative to me?", + "answer": "The laptop is on your left.", + "question_chinese": "客厅桌上的笔记本在我的什么方向？", + "answer_chinese": "笔记本在你的左侧。", + "evidence": [ + { + "description": "There is a laptop directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:05:42", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The laptop is on your right.", + "B": "The laptop is in front of you.", + "C": "The laptop is behind you, on the far side of the coffee table relative to where you are standing.", + "D": "The laptop is underneath the coffee table, directly below the tabletop rather than on the surface.", + "E": "The laptop is on your left." + }, + "answer_label": "E", + "answer_text": "The laptop is on your left.", + "q_id": "egolife_A1_D6_044_7", + "correct_answer": "The laptop is on your left.", + "original_options": { + "A": "The laptop is on your right.", + "B": "The laptop is in front of you.", + "C": "The laptop is behind you, on the far side of the coffee table relative to where you are standing.", + "D": "The laptop is underneath the coffee table, directly below the tabletop rather than on the surface.", + "E": "The laptop is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_8", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the notebook on the coffee table from me?", + "answer": "The laptop is roughly right below you now.", + "question_chinese": "客厅桌上的笔记本在我的什么方向？", + "answer_chinese": "现在笔记本差不多在你的下方。", + "evidence": [ + { + "description": "There is a laptop directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:10:27", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The laptop is roughly above you now.", + "B": "The laptop is roughly to your left now.", + "C": "The laptop is roughly right below you now.", + "D": "The laptop is roughly in front of you now, sitting ahead rather than directly below your position.", + "E": "The laptop is roughly behind you now, not beneath you, relative to where you are facing." + }, + "answer_label": "C", + "answer_text": "The laptop is roughly right below you now.", + "q_id": "egolife_A1_D6_044_8", + "correct_answer": "The laptop is roughly right below you now.", + "original_options": { + "A": "The laptop is roughly above you now.", + "B": "The laptop is roughly to your left now.", + "C": "The laptop is roughly right below you now.", + "D": "The laptop is roughly in front of you now, sitting ahead rather than directly below your position.", + "E": "The laptop is roughly behind you now, not beneath you, relative to where you are facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_5_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the laptop on the living room table?", + "answer": "The laptop on the living-room table is on the left side.", + "question_chinese": "客厅桌上的笔记本电脑在我的什么方向？", + "answer_chinese": "客厅桌上的笔记本电脑在你的左侧。", + "evidence": [ + { + "description": "There is a laptop directly in front of me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to your right.", + "B": "It is behind you.", + "C": "The laptop on the living-room table is on the left side.", + "D": "The laptop on the living-room table is in front of you, not off to the left side.", + "E": "The laptop on the living-room table is slightly behind and to your right from where you are standing." + }, + "answer_label": "C", + "answer_text": "The laptop on the living-room table is on the left side.", + "q_id": "egolife_A1_D6_044_5_T2", + "correct_answer": "The laptop on the living-room table is on the left side.", + "original_options": { + "A": "It is to your right.", + "B": "It is behind you.", + "C": "The laptop on the living-room table is on the left side.", + "D": "The laptop on the living-room table is in front of you, not off to the left side.", + "E": "The laptop on the living-room table is slightly behind and to your right from where you are standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_5_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the laptop on the living room table?", + "answer": "The laptop on the living-room table is in front of you.", + "question_chinese": "客厅桌上的笔记本电脑在我的什么方向？", + "answer_chinese": "客厅桌上的笔记本电脑在你的正前方。", + "evidence": [ + { + "description": "The laptop is in front of you.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "The laptop on the living-room table is in front of you.", + "C": "It is to your left.", + "D": "The laptop on the living-room table is to your right, not directly in front of you.", + "E": "The laptop on the living-room table is slightly behind you, off to the side rather than ahead." + }, + "answer_label": "B", + "answer_text": "The laptop on the living-room table is in front of you.", + "q_id": "egolife_A1_D6_044_5_T1", + "correct_answer": "The laptop on the living-room table is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "The laptop on the living-room table is in front of you.", + "C": "It is to your left.", + "D": "The laptop on the living-room table is to your right, not directly in front of you.", + "E": "The laptop on the living-room table is slightly behind you, off to the side rather than ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_6_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the red refrigerator in the living room?", + "answer": "The red fridge in the living room is to your right.", + "question_chinese": "客厅的红冰箱在我的什么方向？", + "answer_chinese": "客厅的红冰箱在你的右边。", + "evidence": [ + { + "description": "On your right.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to your left.", + "B": "It is behind you.", + "C": "The red fridge in the living room is to your right.", + "D": "The red fridge in the living room is in front of you, not positioned off to your right side.", + "E": "The red fridge in the living room is slightly behind and to your left relative to where you are facing." + }, + "answer_label": "C", + "answer_text": "The red fridge in the living room is to your right.", + "q_id": "egolife_A1_D6_044_6_T1", + "correct_answer": "The red fridge in the living room is to your right.", + "original_options": { + "A": "It is to your left.", + "B": "It is behind you.", + "C": "The red fridge in the living room is to your right.", + "D": "The red fridge in the living room is in front of you, not positioned off to your right side.", + "E": "The red fridge in the living room is slightly behind and to your left relative to where you are facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 19, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_6_T2", + "qid_num": 17, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from me is the red refrigerator in the living room?", + "answer": "The red fridge in the living room is in front of you.", + "question_chinese": "客厅的红冰箱在我的什么方向？", + "answer_chinese": "客厅的红冰箱在你的正前方。", + "evidence": [ + { + "description": "The red refrigerator in the living room is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is to your right.", + "C": "The red fridge in the living room is in front of you.", + "D": "The red fridge in the living room is to your left, not straight ahead in front of you.", + "E": "The red fridge in the living room is behind you and off to one side rather than directly in front." + }, + "answer_label": "C", + "answer_text": "The red fridge in the living room is in front of you.", + "q_id": "egolife_A1_D6_044_6_T2", + "correct_answer": "The red fridge in the living room is in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is to your right.", + "C": "The red fridge in the living room is in front of you.", + "D": "The red fridge in the living room is to your left, not straight ahead in front of you.", + "E": "The red fridge in the living room is behind you and off to one side rather than directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_9", + "qid_num": 18, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the microwave from me?", + "answer": "The microwave is in front of your right.", + "question_chinese": "微波炉在我的什么方向？", + "answer_chinese": "微波炉在你的右前方。", + "evidence": [ + { + "description": "There are two microwaves stacked on top of each other directly in front of me.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:04:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The microwave is to your left.", + "B": "The microwave is behind you.", + "C": "The microwave is slightly behind you on the right, not in front of your right side.", + "D": "The microwave is in front of your right.", + "E": "The microwave is straight ahead of you, centered in your view rather than positioned front-right." + }, + "answer_label": "D", + "answer_text": "The microwave is in front of your right.", + "q_id": "egolife_A1_D6_044_9", + "correct_answer": "The microwave is in front of your right.", + "original_options": { + "A": "The microwave is to your left.", + "B": "The microwave is behind you.", + "C": "The microwave is slightly behind you on the right, not in front of your right side.", + "D": "The microwave is in front of your right.", + "E": "The microwave is straight ahead of you, centered in your view rather than positioned front-right." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_10", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the microwave from me?", + "answer": "The microwave is at the bottom right of you.", + "question_chinese": "微波炉在我的什么方向？", + "answer_chinese": "微波炉在你的右下方。", + "evidence": [ + { + "description": "There are two microwaves stacked on top of each other directly in front of me.", + "timestamps": { + "start": "00:00:32", + "end": "00:00:34" + } + } + ], + "question_timestamps": "00:09:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The microwave is at your top left.", + "B": "The microwave is at the bottom right of you.", + "C": "The microwave is at your bottom left.", + "D": "The microwave is at the top right of you, higher in view instead of down low.", + "E": "The microwave is centered directly in front of you, not positioned at your bottom-right side." + }, + "answer_label": "B", + "answer_text": "The microwave is at the bottom right of you.", + "q_id": "egolife_A1_D6_044_10", + "correct_answer": "The microwave is at the bottom right of you.", + "original_options": { + "A": "The microwave is at your top left.", + "B": "The microwave is at the bottom right of you.", + "C": "The microwave is at your bottom left.", + "D": "The microwave is at the top right of you, higher in view instead of down low.", + "E": "The microwave is centered directly in front of you, not positioned at your bottom-right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_9_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where am I sitting right now?", + "answer": "You’re currently sitting on a chair in the first-floor living room—the chair is right where you are, because you’re sitting on it.", + "question_chinese": "我现在坐在哪里？", + "answer_chinese": "你现在坐在一楼客厅的椅子上，椅子就在你这里，你坐着。", + "evidence": [ + { + "description": "You are sitting in a chair in the living room on the first floor.", + "timestamps": { + "start": "00:04:07", + "end": "00:04:11" + } + } + ], + "question_timestamps": "00:04:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’re sitting on the sofa.", + "B": "You’re sitting on the floor.", + "C": "You’re sitting on a chair in the second-floor living room, not the first-floor living room.", + "D": "You’re sitting on a sofa cushion in the first-floor living room, rather than on a chair.", + "E": "You’re currently sitting on a chair in the first-floor living room—the chair is right where you are, because you’re sitting on it." + }, + "answer_label": "E", + "answer_text": "You’re currently sitting on a chair in the first-floor living room—the chair is right where you are, because you’re sitting on it.", + "q_id": "egolife_A1_D6_044_9_T1", + "correct_answer": "You’re currently sitting on a chair in the first-floor living room—the chair is right where you are, because you’re sitting on it.", + "original_options": { + "A": "You’re sitting on the sofa.", + "B": "You’re sitting on the floor.", + "C": "You’re sitting on a chair in the second-floor living room, not the first-floor living room.", + "D": "You’re sitting on a sofa cushion in the first-floor living room, rather than on a chair.", + "E": "You’re currently sitting on a chair in the first-floor living room—the chair is right where you are, because you’re sitting on it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_9_T2", + "qid_num": 21, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where am I sitting right now?", + "answer": "You’re currently sitting on a sofa cushion in the second-floor living room—the sofa cushion is right where you are, because you’re sitting on it.", + "question_chinese": "我现在坐在哪里？", + "answer_chinese": "你现在坐在二楼客厅的沙发垫上，沙发垫就在你这里，你坐着。", + "evidence": [ + { + "description": "Go to the second floor.", + "timestamps": { + "start": "00:07:40", + "end": "00:07:44" + } + }, + { + "description": "You are sitting on the sofa cushion in the living room on the second floor.", + "timestamps": { + "start": "00:08:18", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:08:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You’re sitting on a chair.", + "B": "You’re currently sitting on a sofa cushion in the second-floor living room—the sofa cushion is right where you are, because you’re sitting on it.", + "C": "You’re sitting on the floor.", + "D": "You’re sitting on a sofa cushion in the first-floor living room, not the second-floor living room.", + "E": "You’re sitting on a chair in the second-floor living room, rather than on a sofa cushion." + }, + "answer_label": "B", + "answer_text": "You’re currently sitting on a sofa cushion in the second-floor living room—the sofa cushion is right where you are, because you’re sitting on it.", + "q_id": "egolife_A1_D6_044_9_T2", + "correct_answer": "You’re currently sitting on a sofa cushion in the second-floor living room—the sofa cushion is right where you are, because you’re sitting on it.", + "original_options": { + "A": "You’re sitting on a chair.", + "B": "You’re currently sitting on a sofa cushion in the second-floor living room—the sofa cushion is right where you are, because you’re sitting on it.", + "C": "You’re sitting on the floor.", + "D": "You’re sitting on a sofa cushion in the first-floor living room, not the second-floor living room.", + "E": "You’re sitting on a chair in the second-floor living room, rather than on a sofa cushion." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_4_T1", + "qid_num": 22, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What is directly in front of me?", + "answer": "In front of you, there’s a laptop.", + "question_chinese": "我的正前方有什么东西？", + "answer_chinese": "你的正前方有一台笔记本电脑。", + "evidence": [ + { + "description": "There is a laptop in front of you.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left, there’s a laptop.", + "B": "Behind you, there’s a laptop.", + "C": "In front of you, there’s a laptop.", + "D": "To your right, there’s a laptop rather than directly in front of you.", + "E": "Above you, there’s a laptop, not positioned straight ahead in your line of sight." + }, + "answer_label": "C", + "answer_text": "In front of you, there’s a laptop.", + "q_id": "egolife_A1_D6_044_4_T1", + "correct_answer": "In front of you, there’s a laptop.", + "original_options": { + "A": "To your left, there’s a laptop.", + "B": "Behind you, there’s a laptop.", + "C": "In front of you, there’s a laptop.", + "D": "To your right, there’s a laptop rather than directly in front of you.", + "E": "Above you, there’s a laptop, not positioned straight ahead in your line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_4_T2", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "What is directly in front of me?", + "answer": "In front of you, there’s a picture frame and a table.", + "question_chinese": "我的正前方有什么东西？", + "answer_chinese": "你的正前方有一幅画框和一张桌子。", + "evidence": [ + { + "description": "There is a picture frame and a table in front of you.", + "timestamps": { + "start": "00:00:21", + "end": "00:00:22" + } + } + ], + "question_timestamps": "00:00:24", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you, there’s a chair.", + "B": "In front of you, there’s a couch.", + "C": "In front of you, there’s only a picture frame, with no table directly ahead of you.", + "D": "In front of you, there’s only a table, and the picture frame is not directly in front.", + "E": "In front of you, there’s a picture frame and a table." + }, + "answer_label": "E", + "answer_text": "In front of you, there’s a picture frame and a table.", + "q_id": "egolife_A1_D6_044_4_T2", + "correct_answer": "In front of you, there’s a picture frame and a table.", + "original_options": { + "A": "In front of you, there’s a chair.", + "B": "In front of you, there’s a couch.", + "C": "In front of you, there’s only a picture frame, with no table directly ahead of you.", + "D": "In front of you, there’s only a table, and the picture frame is not directly in front.", + "E": "In front of you, there’s a picture frame and a table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_1", + "qid_num": 24, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the door to the bedroom closer to me, or is the door to the second floor closer to me?", + "answer": "The door to the bedroom is closer to you.", + "question_chinese": "是到卧室的门离我更近，还是到二楼的门离我更近？", + "answer_chinese": "到卧室的门离你更近。", + "evidence": [ + { + "description": "The door leading to the bedroom is in front of my left.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "The door leading to the second floor is in front of my right.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:06:09", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "If the vertical distance is taken into account, the door leading to the second floor should be closer to you.", + "B": "The door to the bedroom is closer to you." + }, + "answer_label": "B", + "answer_text": "The door to the bedroom is closer to you.", + "q_id": "egolife_A1_D6_044_1", + "correct_answer": "The door to the bedroom is closer to you.", + "original_options": { + "A": "If the vertical distance is taken into account, the door leading to the second floor should be closer to you.", + "B": "The door to the bedroom is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_044", + "video_length": "00:12:00", + "qid": "egolife_A1_D6_044_2", + "qid_num": 25, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the door to the bedroom closer to me, or is the door to the second floor closer to me?", + "answer": "If the vertical distance is taken into account, the door leading to the second floor should be closer to you.", + "question_chinese": "是到卧室的门离我更近，还是到二楼的门离我更近？", + "answer_chinese": "算上垂直距离的话，应该是到二楼的门离你更近。", + "evidence": [ + { + "description": "The door leading to the bedroom is in front of my left.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "The door leading to the second floor is in front of my right.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:11:09", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "If the vertical distance is taken into account, the door leading to the second floor should be closer to you.", + "B": "The door to the bedroom is closer to you." + }, + "answer_label": "A", + "answer_text": "If the vertical distance is taken into account, the door leading to the second floor should be closer to you.", + "q_id": "egolife_A1_D6_044_2", + "correct_answer": "If the vertical distance is taken into account, the door leading to the second floor should be closer to you.", + "original_options": { + "A": "If the vertical distance is taken into account, the door leading to the second floor should be closer to you.", + "B": "The door to the bedroom is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + } +] diff --git a/QAs/egolife_A1_D6_045-mcq.json b/QAs/egolife_A1_D6_045-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..dc72beb90811a8cb399afd28856df18a29a57945 --- /dev/null +++ b/QAs/egolife_A1_D6_045-mcq.json @@ -0,0 +1,843 @@ +[ + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_1_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the guitar relative to me?", + "answer": "The guitar should be above you to your right.", + "question_chinese": "吉他在我的什么方向？", + "answer_chinese": "吉他应该在你的右上方。", + "evidence": [ + { + "description": "There is a guitar directly in front of me.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:03:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above you to your left.", + "B": "Above you in front of you.", + "C": "The guitar should be above you and slightly behind you, not on your right side.", + "D": "The guitar should be above you to your right.", + "E": "It’s positioned above you but closer to your left, rather than above-right relative to you." + }, + "answer_label": "D", + "answer_text": "The guitar should be above you to your right.", + "q_id": "egolife_A1_D6_045_1_T1", + "correct_answer": "The guitar should be above you to your right.", + "original_options": { + "A": "Above you to your left.", + "B": "Above you in front of you.", + "C": "The guitar should be above you and slightly behind you, not on your right side.", + "D": "The guitar should be above you to your right.", + "E": "It’s positioned above you but closer to your left, rather than above-right relative to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_1_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the guitar relative to me?", + "answer": "The guitar should be above you to your left.", + "question_chinese": "吉他在我的什么方向？", + "answer_chinese": "吉他应该在你的左上方", + "evidence": [ + { + "description": "There is a guitar directly in front of me.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:05:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Above you to your right.", + "B": "The guitar should be above you to your left.", + "C": "Above you behind you.", + "D": "The guitar should be above you and in front of you, not over your left shoulder.", + "E": "It’s above you but offset to the right side, instead of being above-left relative to you." + }, + "answer_label": "B", + "answer_text": "The guitar should be above you to your left.", + "q_id": "egolife_A1_D6_045_1_T2", + "correct_answer": "The guitar should be above you to your left.", + "original_options": { + "A": "Above you to your right.", + "B": "The guitar should be above you to your left.", + "C": "Above you behind you.", + "D": "The guitar should be above you and in front of you, not over your left shoulder.", + "E": "It’s above you but offset to the right side, instead of being above-left relative to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_2_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go up to the second floor to get a chair, how should I go?", + "answer": "You should turn right, pass the dining table, go up the stairs, then step outside and after one or two steps you’ll see the chair.", + "question_chinese": "如果想到二楼去拿一把椅子，我应该怎么走？", + "answer_chinese": "你应该向右转，经过餐桌，上楼梯，然后出门走一两步就能看到椅子了。", + "evidence": [ + { + "description": "You can see there are some chairs on the second floor.", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + }, + { + "description": "I went down from the second floor to the first floor.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:51" + } + }, + { + "description": "After going downstairs and opening the door, the dining table is right in front.", + "timestamps": { + "start": "00:02:52", + "end": "00:02:53" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left and go upstairs.", + "B": "Go upstairs without passing the table.", + "C": "Turn right, pass the dining table, go up the stairs, then step inside to find the chair.", + "D": "Turn right, go past the dining table, climb the stairs, then take several steps outside before spotting the chair.", + "E": "You should turn right, pass the dining table, go up the stairs, then step outside and after one or two steps you’ll see the chair." + }, + "answer_label": "E", + "answer_text": "You should turn right, pass the dining table, go up the stairs, then step outside and after one or two steps you’ll see the chair.", + "q_id": "egolife_A1_D6_045_2_T2", + "correct_answer": "You should turn right, pass the dining table, go up the stairs, then step outside and after one or two steps you’ll see the chair.", + "original_options": { + "A": "Turn left and go upstairs.", + "B": "Go upstairs without passing the table.", + "C": "Turn right, pass the dining table, go up the stairs, then step inside to find the chair.", + "D": "Turn right, go past the dining table, climb the stairs, then take several steps outside before spotting the chair.", + "E": "You should turn right, pass the dining table, go up the stairs, then step outside and after one or two steps you’ll see the chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_3_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go up to the second floor to get a chair, how should I go?", + "answer": "You should walk about three steps to the right, go in through the door, then go up the stairs in front of you to the second floor, and you’ll see the chair.", + "question_chinese": "如果想到二楼去拿一把椅子，我应该怎么走？", + "answer_chinese": "你应该向右走三步的距离，然后进门，随后从面前的楼梯上到二楼，就能看到椅子了。", + "evidence": [ + { + "description": "You can see there are some chairs on the second floor.", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + }, + { + "description": "I went down from the second floor to the first floor.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:51" + } + } + ], + "question_timestamps": "00:03:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Go up the stairs behind you.", + "B": "You should walk about three steps to the right, go in through the door, then go up the stairs in front of you to the second floor, and you’ll see the chair.", + "C": "Use the stairs to your left.", + "D": "Walk about three steps to the right, enter through the door, then take the stairs to your right.", + "E": "Walk about three steps to the right, go in through the door, then go upstairs using the stairs behind you." + }, + "answer_label": "B", + "answer_text": "You should walk about three steps to the right, go in through the door, then go up the stairs in front of you to the second floor, and you’ll see the chair.", + "q_id": "egolife_A1_D6_045_3_T2", + "correct_answer": "You should walk about three steps to the right, go in through the door, then go up the stairs in front of you to the second floor, and you’ll see the chair.", + "original_options": { + "A": "Go up the stairs behind you.", + "B": "You should walk about three steps to the right, go in through the door, then go up the stairs in front of you to the second floor, and you’ll see the chair.", + "C": "Use the stairs to your left.", + "D": "Walk about three steps to the right, enter through the door, then take the stairs to your right.", + "E": "Walk about three steps to the right, go in through the door, then go upstairs using the stairs behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_2_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go up to the second floor to get a chair, how should I go?", + "answer": "You should go up the stairs in front of you to the second floor, then step outside and after one or two steps you’ll see the chair.", + "question_chinese": "如果想到二楼去拿一把椅子，我应该怎么走？", + "answer_chinese": "你应该从面前的楼梯上到二楼，然后出门走一两步就能看到椅子了。", + "evidence": [ + { + "description": "You can see there are some chairs on the second floor.", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + }, + { + "description": "I went down from the second floor to the first floor.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:51" + } + } + ], + "question_timestamps": "00:03:46", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Take the stairs behind you.", + "B": "You should go up the stairs in front of you to the second floor, then step outside and after one or two steps you’ll see the chair.", + "C": "Go up the stairs to your left.", + "D": "Go up the stairs to your right to the second floor, then step outside and look for the chair.", + "E": "Head upstairs using the stairs behind you, then step outside and after one or two steps you’ll see the chair." + }, + "answer_label": "B", + "answer_text": "You should go up the stairs in front of you to the second floor, then step outside and after one or two steps you’ll see the chair.", + "q_id": "egolife_A1_D6_045_2_T1", + "correct_answer": "You should go up the stairs in front of you to the second floor, then step outside and after one or two steps you’ll see the chair.", + "original_options": { + "A": "Take the stairs behind you.", + "B": "You should go up the stairs in front of you to the second floor, then step outside and after one or two steps you’ll see the chair.", + "C": "Go up the stairs to your left.", + "D": "Go up the stairs to your right to the second floor, then step outside and look for the chair.", + "E": "Head upstairs using the stairs behind you, then step outside and after one or two steps you’ll see the chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_3_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go up to the second floor to get a chair, how should I go?", + "answer": "You should walk about five steps toward your front-left, go in through the door, then go up the stairs in front of you to the second floor, and you’ll see the chair.", + "question_chinese": "如果想到二楼去拿一把椅子，我应该怎么走？", + "answer_chinese": "你应该像左前方走5步的距离，然后进门，随后从面前的楼梯上到二楼，就能看到椅子了。", + "evidence": [ + { + "description": "You can see there are some chairs on the second floor.", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + }, + { + "description": "I went down from the second floor to the first floor.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:51" + } + } + ], + "question_timestamps": "00:05:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk five steps front-right.", + "B": "Go through the door immediately.", + "C": "You should walk about five steps toward your front-left, go in through the door, then go up the stairs in front of you to the second floor, and you’ll see the chair.", + "D": "Walk about five steps toward your front-left, enter the door, then take the stairs to your left.", + "E": "Walk about five steps toward your front-left, go in through the door, then go upstairs using stairs behind you." + }, + "answer_label": "C", + "answer_text": "You should walk about five steps toward your front-left, go in through the door, then go up the stairs in front of you to the second floor, and you’ll see the chair.", + "q_id": "egolife_A1_D6_045_3_T1", + "correct_answer": "You should walk about five steps toward your front-left, go in through the door, then go up the stairs in front of you to the second floor, and you’ll see the chair.", + "original_options": { + "A": "Walk five steps front-right.", + "B": "Go through the door immediately.", + "C": "You should walk about five steps toward your front-left, go in through the door, then go up the stairs in front of you to the second floor, and you’ll see the chair.", + "D": "Walk about five steps toward your front-left, enter the door, then take the stairs to your left.", + "E": "Walk about five steps toward your front-left, go in through the door, then go upstairs using stairs behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_5", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to go to the second floor to get a chair, how should I get there?", + "answer": "You should go up to the second floor via the stairs in front of you, then go outside and walk a step or two, and you’ll see the chairs.", + "question_chinese": "我如果想到二楼去拿一把椅子，我应该怎么走？", + "answer_chinese": "你应该从面前的楼梯上到二楼，然后出门走一两步就能看到椅子了。", + "evidence": [ + { + "description": "You can see there are some chairs on the second floor.", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + }, + { + "description": "I went down from the second floor to the first floor.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:51" + } + } + ], + "question_timestamps": "00:03:46", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Use the stairs behind you.", + "B": "Go downstairs first.", + "C": "Go up the front stairs, then stay inside and walk a step or two to find chairs.", + "D": "Go up the left stairs to the second floor, then go outside and walk a step or two.", + "E": "You should go up to the second floor via the stairs in front of you, then go outside and walk a step or two, and you’ll see the chairs." + }, + "answer_label": "E", + "answer_text": "You should go up to the second floor via the stairs in front of you, then go outside and walk a step or two, and you’ll see the chairs.", + "q_id": "egolife_A1_D6_045_5", + "correct_answer": "You should go up to the second floor via the stairs in front of you, then go outside and walk a step or two, and you’ll see the chairs.", + "original_options": { + "A": "Use the stairs behind you.", + "B": "Go downstairs first.", + "C": "Go up the front stairs, then stay inside and walk a step or two to find chairs.", + "D": "Go up the left stairs to the second floor, then go outside and walk a step or two.", + "E": "You should go up to the second floor via the stairs in front of you, then go outside and walk a step or two, and you’ll see the chairs." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I go to the second floor to get a chair?", + "answer": "You should walk five steps forward to your left, then enter the door, and afterwards go up to the second floor via the stairs in front of you—you will see the chairs there.", + "question_chinese": "我如果想到二楼去拿一把椅子，我应该怎么走？", + "answer_chinese": "你应该像左前方走5步的距离，然后进门，随后从面前的楼梯上到二楼，就能看到椅子了。", + "evidence": [ + { + "description": "You can see there are some chairs on the second floor.", + "timestamps": { + "start": "00:00:58", + "end": "00:00:59" + } + }, + { + "description": "I went down from the second floor to the first floor.", + "timestamps": { + "start": "00:02:40", + "end": "00:02:51" + } + } + ], + "question_timestamps": "00:05:28", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk five steps forward-right.", + "B": "Enter the door, then take stairs behind you.", + "C": "You should walk five steps forward to your left, then enter the door, and afterwards go up to the second floor via the stairs in front of you—you will see the chairs there.", + "D": "Walk five steps forward-left, enter the door, then go up using the stairs on your left.", + "E": "Walk six steps forward-left, go through the door, then take the stairs in front of you." + }, + "answer_label": "C", + "answer_text": "You should walk five steps forward to your left, then enter the door, and afterwards go up to the second floor via the stairs in front of you—you will see the chairs there.", + "q_id": "egolife_A1_D6_045_6", + "correct_answer": "You should walk five steps forward to your left, then enter the door, and afterwards go up to the second floor via the stairs in front of you—you will see the chairs there.", + "original_options": { + "A": "Walk five steps forward-right.", + "B": "Enter the door, then take stairs behind you.", + "C": "You should walk five steps forward to your left, then enter the door, and afterwards go up to the second floor via the stairs in front of you—you will see the chairs there.", + "D": "Walk five steps forward-left, enter the door, then go up using the stairs on your left.", + "E": "Walk six steps forward-left, go through the door, then take the stairs in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 8, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_9", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the projection screen from me?", + "answer": "The projection screen is behind and above you.", + "question_chinese": "投影屏幕在我的什么方向？", + "answer_chinese": "投影屏幕在你的后上方。", + "evidence": [ + { + "description": "There is a projection screen directly in front of me.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:03:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is directly in front of you.", + "B": "It is below and behind you.", + "C": "The projection screen is in front of you at about eye level, not positioned above your head.", + "D": "The projection screen is behind and above you.", + "E": "The projection screen sits to your right side and slightly below you, rather than behind you." + }, + "answer_label": "D", + "answer_text": "The projection screen is behind and above you.", + "q_id": "egolife_A1_D6_045_9", + "correct_answer": "The projection screen is behind and above you.", + "original_options": { + "A": "It is directly in front of you.", + "B": "It is below and behind you.", + "C": "The projection screen is in front of you at about eye level, not positioned above your head.", + "D": "The projection screen is behind and above you.", + "E": "The projection screen sits to your right side and slightly below you, rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_10", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the projection screen from me?", + "answer": "The projection screen is above and to your left.", + "question_chinese": "投影屏幕在我的什么方向？", + "answer_chinese": "投影屏幕在你的左上方，", + "evidence": [ + { + "description": "There is a projection screen directly in front of me.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:05:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is below and to your right.", + "B": "The projection screen is above and to your left.", + "C": "It is behind and to your left.", + "D": "The projection screen is above you but on your right side, not to your left.", + "E": "The projection screen is in front of you at a lower height, rather than above you on the left." + }, + "answer_label": "B", + "answer_text": "The projection screen is above and to your left.", + "q_id": "egolife_A1_D6_045_10", + "correct_answer": "The projection screen is above and to your left.", + "original_options": { + "A": "It is below and to your right.", + "B": "The projection screen is above and to your left.", + "C": "It is behind and to your left.", + "D": "The projection screen is above you but on your right side, not to your left.", + "E": "The projection screen is in front of you at a lower height, rather than above you on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_7", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red refrigerator located?", + "answer": "The red refrigerator is on your left now.", + "question_chinese": "红色冰箱在我的什么位置？", + "answer_chinese": "红色的冰箱现在在你的左侧。", + "evidence": [ + { + "description": "There is a red refrigerator in front of my right.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:02:44", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The red refrigerator is on your left now.", + "B": "It is on your right now.", + "C": "It is directly in front of you.", + "D": "The red refrigerator is behind you now, not positioned on your left side.", + "E": "The red refrigerator is to your upper right now, rather than on your left at your current position." + }, + "answer_label": "A", + "answer_text": "The red refrigerator is on your left now.", + "q_id": "egolife_A1_D6_045_7", + "correct_answer": "The red refrigerator is on your left now.", + "original_options": { + "A": "The red refrigerator is on your left now.", + "B": "It is on your right now.", + "C": "It is directly in front of you.", + "D": "The red refrigerator is behind you now, not positioned on your left side.", + "E": "The red refrigerator is to your upper right now, rather than on your left at your current position." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 18 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045_8", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the red refrigerator located?", + "answer": "The red refrigerator should be on your upper left now.", + "question_chinese": "红色冰箱在我的什么位置？", + "answer_chinese": "红色的冰箱现在应该在你的左上方。", + "evidence": [ + { + "description": "There is a red refrigerator in front of my right.", + "timestamps": { + "start": "00:01:23", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:05:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It should be on your lower left.", + "B": "It should be on your upper right.", + "C": "The red refrigerator should be directly behind you now, not up and to your left.", + "D": "The red refrigerator should be in front of you at about chest height, rather than on the upper left.", + "E": "The red refrigerator should be on your upper left now." + }, + "answer_label": "E", + "answer_text": "The red refrigerator should be on your upper left now.", + "q_id": "egolife_A1_D6_045_8", + "correct_answer": "The red refrigerator should be on your upper left now.", + "original_options": { + "A": "It should be on your lower left.", + "B": "It should be on your upper right.", + "C": "The red refrigerator should be directly behind you now, not up and to your left.", + "D": "The red refrigerator should be in front of you at about chest height, rather than on the upper left.", + "E": "The red refrigerator should be on your upper left now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 19 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045-mcq_fixed_8", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are stored in the black box in total?", + "answer": "Five.", + "question_chinese": "黑色的盒子里一共收纳了多少个充电宝？", + "answer_chinese": "一共有6个充电宝", + "evidence": [ + { + "description": "I put six power banks into a box.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:06" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "Six.", + "D": "There are three power banks stored in the black box in total, not five.", + "E": "Seven power banks are stored in the black box altogether, which is more than five." + }, + "answer_label": "B", + "answer_text": "Five.", + "q_id": "egolife_A1_D6_045-mcq_fixed_8", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "Six.", + "D": "There are three power banks stored in the black box in total, not five.", + "E": "Seven power banks are stored in the black box altogether, which is more than five." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D6_045", + "video_length": "00:05:59", + "qid": "egolife_A1_D6_045-mcq_fixed_9", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are stored in the black box in total?", + "answer": "Six.", + "question_chinese": "黑色的盒子里一共收纳了多少个充电宝？", + "answer_chinese": "一共有8个充电宝", + "evidence": [ + { + "description": "I put six power banks into a box.", + "timestamps": { + "start": "00:00:00", + "end": "00:01:13" + } + }, + { + "description": "I put another power bank into the black box.", + "timestamps": { + "start": "00:01:17", + "end": "00:01:18" + } + }, + { + "description": "I put another power bank into the black box.", + "timestamps": { + "start": "00:01:31", + "end": "00:01:33" + } + } + ], + "question_timestamps": "00:05:59", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "Six.", + "D": "There are four power banks stored in the black box in total, not six.", + "E": "Eight power banks are stored in the black box altogether, which is more than six." + }, + "answer_label": "C", + "answer_text": "Six.", + "q_id": "egolife_A1_D6_045-mcq_fixed_9", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "Six.", + "D": "There are four power banks stored in the black box in total, not six.", + "E": "Eight power banks are stored in the black box altogether, which is more than six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + } +] diff --git a/QAs/egolife_A1_D7_001-mcq.json b/QAs/egolife_A1_D7_001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1d2059ff73c137ee3ada861b91ee62c305a5777a --- /dev/null +++ b/QAs/egolife_A1_D7_001-mcq.json @@ -0,0 +1,1010 @@ +[ + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "The cup I was just holding still has soda in it.", + "answer": "Cup.", + "question_chinese": "我刚刚手里拿的杯子还是汽水", + "answer_chinese": "杯子", + "evidence": [ + { + "description": "I am holding a cup in my hand.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:15" + } + } + ], + "question_timestamps": "00:09:30", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Cup.", + "B": "Bottle." + }, + "answer_label": "A", + "answer_text": "Cup.", + "q_id": "egolife_A1_D7_001_7", + "correct_answer": "Cup.", + "original_options": { + "A": "Cup.", + "B": "Bottle." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "The cup I was just holding still has soda in it.", + "answer": "Water.", + "question_chinese": "我刚刚手里拿的杯子还是汽水", + "answer_chinese": "汽水", + "evidence": [ + { + "description": "The soda in my hand", + "timestamps": { + "start": "00:10:30", + "end": "00:10:45" + } + } + ], + "question_timestamps": "00:11:00", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Soda.", + "B": "Water." + }, + "answer_label": "B", + "answer_text": "Water.", + "q_id": "egolife_A1_D7_001_8", + "correct_answer": "Water.", + "original_options": { + "A": "Soda.", + "B": "Water." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 90, + "reasons": [ + "not_interrogative" + ] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_1_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far apart do I see the chair and the tree are?", + "answer": "One meter.", + "question_chinese": "我看见椅子距离树的距离是多远？", + "answer_chinese": "一米", + "evidence": [ + { + "description": "The chair is in the right corner of the yard.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:21" + } + }, + { + "description": "The tree is in the left corner of the yard.", + "timestamps": { + "start": "00:02:22", + "end": "00:02:23" + } + }, + { + "description": "The yard is about one meter long.", + "timestamps": { + "start": "00:02:24", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:02:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Ten centimeters.", + "B": "Three meters.", + "C": "About three meters apart, with the chair slightly closer than the tree from your viewpoint.", + "D": "One meter.", + "E": "Roughly ten meters apart, with a noticeable gap separating the chair and the tree." + }, + "answer_label": "D", + "answer_text": "One meter.", + "q_id": "egolife_A1_D7_001_1_T1", + "correct_answer": "One meter.", + "original_options": { + "A": "Ten centimeters.", + "B": "Three meters.", + "C": "About three meters apart, with the chair slightly closer than the tree from your viewpoint.", + "D": "One meter.", + "E": "Roughly ten meters apart, with a noticeable gap separating the chair and the tree." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_1_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far apart do I see the chair and the tree are?", + "answer": "One meter.", + "question_chinese": "我看见椅子距离树的距离是多远？", + "answer_chinese": "一米", + "evidence": [ + { + "description": "The chair is in the right corner of the yard.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:25" + } + }, + { + "description": "The tree is in the left corner of the yard.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:25" + } + }, + { + "description": "The yard is about one meter long.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One meter.", + "B": "Ten centimeters.", + "C": "Three meters.", + "D": "They look about three meters apart, not close enough to seem within a single step.", + "E": "They appear roughly ten meters apart, with plenty of space between the chair and the tree." + }, + "answer_label": "A", + "answer_text": "One meter.", + "q_id": "egolife_A1_D7_001_1_T2", + "correct_answer": "One meter.", + "original_options": { + "A": "One meter.", + "B": "Ten centimeters.", + "C": "Three meters.", + "D": "They look about three meters apart, not close enough to seem within a single step.", + "E": "They appear roughly ten meters apart, with plenty of space between the chair and the tree." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_1_T3", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far apart do I see the chair and the tree are?", + "answer": "Ten centimeters.", + "question_chinese": "我看见椅子距离树的距离是多远？", + "answer_chinese": "十厘米", + "evidence": [ + { + "description": "The chair is at the lower left of the shed.", + "timestamps": { + "start": "00:02:28", + "end": "00:02:50" + } + }, + { + "description": "The shed is to the right of the tree.", + "timestamps": { + "start": "00:02:28", + "end": "00:02:50" + } + }, + { + "description": "The lower left of the shed is ten centimeters away from the tree.", + "timestamps": { + "start": "00:02:28", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One meter.", + "B": "Three meters.", + "C": "They appear about one meter apart, separated by a small but clearly visible gap.", + "D": "Ten centimeters.", + "E": "They look roughly three meters apart, far more than just a hand’s width." + }, + "answer_label": "D", + "answer_text": "Ten centimeters.", + "q_id": "egolife_A1_D7_001_1_T3", + "correct_answer": "Ten centimeters.", + "original_options": { + "A": "One meter.", + "B": "Three meters.", + "C": "They appear about one meter apart, separated by a small but clearly visible gap.", + "D": "Ten centimeters.", + "E": "They look roughly three meters apart, far more than just a hand’s width." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_1_T4", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far apart do I see the chair and the tree are?", + "answer": "Ten centimeters.", + "question_chinese": "我看见椅子距离树的距离是多远？", + "answer_chinese": "十厘米", + "evidence": [ + { + "description": "The chair is at the lower left of the shed.", + "timestamps": { + "start": "00:02:28", + "end": "00:02:50" + } + }, + { + "description": "The shed is to the right of the tree.", + "timestamps": { + "start": "00:02:28", + "end": "00:02:50" + } + }, + { + "description": "The lower left of the shed is ten centimeters away from the tree.", + "timestamps": { + "start": "00:02:28", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One meter.", + "B": "Ten centimeters.", + "C": "Three meters.", + "D": "They seem about one meter apart, not nearly as close as a few centimeters.", + "E": "They appear roughly ten meters apart, with the chair and tree clearly separated by distance." + }, + "answer_label": "B", + "answer_text": "Ten centimeters.", + "q_id": "egolife_A1_D7_001_1_T4", + "correct_answer": "Ten centimeters.", + "original_options": { + "A": "One meter.", + "B": "Ten centimeters.", + "C": "Three meters.", + "D": "They seem about one meter apart, not nearly as close as a few centimeters.", + "E": "They appear roughly ten meters apart, with the chair and tree clearly separated by distance." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_10", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are placed in front of the table?", + "answer": "Seven.", + "question_chinese": "放在桌子跟前的椅子的数量是多少", + "answer_chinese": "七个", + "evidence": [ + { + "description": "There are seven chairs in front of the table now.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:28" + } + } + ], + "question_timestamps": "00:02:40", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Seven.", + "B": "Six.", + "C": "Eight.", + "D": "Five chairs are placed in front of the table, forming a shorter row than expected.", + "E": "Nine chairs are placed in front of the table, making the row look longer than seven." + }, + "answer_label": "A", + "answer_text": "Seven.", + "q_id": "egolife_A1_D7_001_10", + "correct_answer": "Seven.", + "original_options": { + "A": "Seven.", + "B": "Six.", + "C": "Eight.", + "D": "Five chairs are placed in front of the table, forming a shorter row than expected.", + "E": "Nine chairs are placed in front of the table, making the row look longer than seven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_9", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are placed in front of the table?", + "answer": "Eight.", + "question_chinese": "放在桌子跟前的椅子的数量是多少", + "answer_chinese": "八个", + "evidence": [ + { + "description": "Now there are eight people sitting at the table.", + "timestamps": { + "start": "00:08:00", + "end": "00:08:30" + } + } + ], + "question_timestamps": "00:10:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Eight.", + "B": "Seven.", + "C": "Nine.", + "D": "Six chairs are placed in front of the table, leaving two fewer seats than eight.", + "E": "Ten chairs are placed in front of the table, extending the row beyond what you saw." + }, + "answer_label": "A", + "answer_text": "Eight.", + "q_id": "egolife_A1_D7_001_9", + "correct_answer": "Eight.", + "original_options": { + "A": "Eight.", + "B": "Seven.", + "C": "Nine.", + "D": "Six chairs are placed in front of the table, leaving two fewer seats than eight.", + "E": "Ten chairs are placed in front of the table, extending the row beyond what you saw." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_5", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the cup located in relation to the toilet paper?", + "answer": "Left front.", + "question_chinese": "杯子相对于卫生纸在什么位置", + "answer_chinese": "左前方", + "evidence": [ + { + "description": "The cup is on the dining table in the living room.", + "timestamps": { + "start": "00:08:40", + "end": "00:08:45" + } + }, + { + "description": "The dining table is to the front-left of the toilet paper.", + "timestamps": { + "start": "00:08:40", + "end": "00:08:45" + } + } + ], + "question_timestamps": "00:08:49", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Right front.", + "B": "Left front.", + "C": "Left behind.", + "D": "The cup is positioned to the right of the toilet paper, slightly forward from it.", + "E": "The cup sits behind the toilet paper, offset a bit to the left rather than in front." + }, + "answer_label": "B", + "answer_text": "Left front.", + "q_id": "egolife_A1_D7_001_5", + "correct_answer": "Left front.", + "original_options": { + "A": "Right front.", + "B": "Left front.", + "C": "Left behind.", + "D": "The cup is positioned to the right of the toilet paper, slightly forward from it.", + "E": "The cup sits behind the toilet paper, offset a bit to the left rather than in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_6", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the cup located in relation to the toilet paper?", + "answer": "Right above the toilet paper.", + "question_chinese": "杯子相对于卫生纸在什么位置", + "answer_chinese": "在卫生纸的正上方", + "evidence": [ + { + "description": "The cup is in my hand.", + "timestamps": { + "start": "00:09:00", + "end": "00:09:02" + } + }, + { + "description": "My hand is directly above the toilet paper", + "timestamps": { + "start": "00:09:00", + "end": "00:09:02" + } + } + ], + "question_timestamps": "00:09:06", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Directly below.", + "B": "Left above.", + "C": "The cup is located to the left of the toilet paper, at about the same height as it.", + "D": "Right above the toilet paper.", + "E": "The cup is positioned in front of the toilet paper instead of above it, closer toward you." + }, + "answer_label": "D", + "answer_text": "Right above the toilet paper.", + "q_id": "egolife_A1_D7_001_6", + "correct_answer": "Right above the toilet paper.", + "original_options": { + "A": "Directly below.", + "B": "Left above.", + "C": "The cup is located to the left of the toilet paper, at about the same height as it.", + "D": "Right above the toilet paper.", + "E": "The cup is positioned in front of the toilet paper instead of above it, closer toward you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 18, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the chair from the tree?", + "answer": "One meter.", + "question_chinese": "椅子距离树的距离是多远", + "answer_chinese": "一米", + "evidence": [ + { + "description": "The chair is in the right corner of the yard.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:25" + } + }, + { + "description": "The tree is in the left corner of the yard.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:25" + } + }, + { + "description": "The yard is about one meter long.", + "timestamps": { + "start": "00:02:20", + "end": "00:02:25" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Two meters away.", + "B": "Half a meter away.", + "C": "About three meters from the tree, leaving a noticeable gap between the chair and trunk.", + "D": "One meter.", + "E": "Roughly five meters from the tree, far enough that you would need several steps to reach it." + }, + "answer_label": "D", + "answer_text": "One meter.", + "q_id": "egolife_A1_D7_001_1", + "correct_answer": "One meter.", + "original_options": { + "A": "Two meters away.", + "B": "Half a meter away.", + "C": "About three meters from the tree, leaving a noticeable gap between the chair and trunk.", + "D": "One meter.", + "E": "Roughly five meters from the tree, far enough that you would need several steps to reach it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How far is the chair from the tree?", + "answer": "Ten centimeters.", + "question_chinese": "椅子距离树的距离是多远", + "answer_chinese": "十厘米", + "evidence": [ + { + "description": "The chair is at the lower left of the shed.", + "timestamps": { + "start": "00:02:28", + "end": "00:02:50" + } + }, + { + "description": "The shed is to the right of the tree.", + "timestamps": { + "start": "00:02:28", + "end": "00:02:50" + } + }, + { + "description": "The lower left of the shed is ten centimeters away from the tree.", + "timestamps": { + "start": "00:02:28", + "end": "00:02:50" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Five centimeters away.", + "B": "Twenty centimeters away.", + "C": "Ten centimeters.", + "D": "About fifteen centimeters from the tree, with a small but visible space separating them.", + "E": "Around thirteen centimeters from the tree, close but not quite touching the trunk." + }, + "answer_label": "C", + "answer_text": "Ten centimeters.", + "q_id": "egolife_A1_D7_001_2", + "correct_answer": "Ten centimeters.", + "original_options": { + "A": "Five centimeters away.", + "B": "Twenty centimeters away.", + "C": "Ten centimeters.", + "D": "About fifteen centimeters from the tree, with a small but visible space separating them.", + "E": "Around thirteen centimeters from the tree, close but not quite touching the trunk." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_2_T2", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the soda relative to me?", + "answer": "It was directly in front of you.", + "question_chinese": "汽水相对于我在我的什么位置？", + "answer_chinese": "在我的正前方", + "evidence": [ + { + "description": "I am now standing facing the right corner of the yard.", + "timestamps": { + "start": "00:02:11", + "end": "00:02:12" + } + }, + { + "description": "The soda is on the table, and the table is directly behind me.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:11" + } + } + ], + "question_timestamps": "00:02:14", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly behind you.", + "B": "Directly to your left.", + "C": "It was directly to your right, positioned beside you rather than straight ahead.", + "D": "It was directly in front of you.", + "E": "It was directly above you, not at eye level in front of your body." + }, + "answer_label": "D", + "answer_text": "It was directly in front of you.", + "q_id": "egolife_A1_D7_001_2_T2", + "correct_answer": "It was directly in front of you.", + "original_options": { + "A": "Directly behind you.", + "B": "Directly to your left.", + "C": "It was directly to your right, positioned beside you rather than straight ahead.", + "D": "It was directly in front of you.", + "E": "It was directly above you, not at eye level in front of your body." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_2_T1", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the soda relative to me?", + "answer": "It was directly behind you.", + "question_chinese": "汽水相对于我在我的什么位置？", + "answer_chinese": "在我的正后方", + "evidence": [ + { + "description": "I am now standing facing the right corner of the yard.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:15" + } + }, + { + "description": "The soda is on the table, and the table is directly behind me.", + "timestamps": { + "start": "00:02:13", + "end": "00:02:15" + } + } + ], + "question_timestamps": "00:02:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly in front of you.", + "B": "Directly to your left.", + "C": "It was directly to your right, off to the side instead of behind you.", + "D": "It was directly above you, positioned overhead rather than behind your back.", + "E": "It was directly behind you." + }, + "answer_label": "E", + "answer_text": "It was directly behind you.", + "q_id": "egolife_A1_D7_001_2_T1", + "correct_answer": "It was directly behind you.", + "original_options": { + "A": "Directly in front of you.", + "B": "Directly to your left.", + "C": "It was directly to your right, off to the side instead of behind you.", + "D": "It was directly above you, positioned overhead rather than behind your back.", + "E": "It was directly behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 12 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_3_T1", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the soda relative to me?", + "answer": "It was directly in front of you.", + "question_chinese": "汽水相对于我在我的什么位置？", + "answer_chinese": "在我的正前方", + "evidence": [ + { + "description": "I sat right across the table.", + "timestamps": { + "start": "00:06:25", + "end": "00:06:30" + } + }, + { + "description": "The soda is on the table.", + "timestamps": { + "start": "00:06:25", + "end": "00:06:30" + } + } + ], + "question_timestamps": "00:06:35", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly behind you.", + "B": "It was directly in front of you.", + "C": "Directly to your right.", + "D": "It was directly to your left, positioned beside you rather than in front.", + "E": "It was directly above you, not straight ahead in your field of view." + }, + "answer_label": "B", + "answer_text": "It was directly in front of you.", + "q_id": "egolife_A1_D7_001_3_T1", + "correct_answer": "It was directly in front of you.", + "original_options": { + "A": "Directly behind you.", + "B": "It was directly in front of you.", + "C": "Directly to your right.", + "D": "It was directly to your left, positioned beside you rather than in front.", + "E": "It was directly above you, not straight ahead in your field of view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_001", + "video_length": "00:11:51", + "qid": "egolife_A1_D7_001_3_T2", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the soda relative to me?", + "answer": "It was in your hand.", + "question_chinese": "汽水相对于我在我的什么位置？", + "answer_chinese": "在我的手上", + "evidence": [ + { + "description": "I sat right across the table.", + "timestamps": { + "start": "00:06:25", + "end": "00:06:30" + } + }, + { + "description": "The soda is on the table.", + "timestamps": { + "start": "00:06:20", + "end": "00:06:24" + } + } + ], + "question_timestamps": "00:10:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly in front of you.", + "B": "Directly behind you.", + "C": "It was in your hand.", + "D": "It was directly to your left, not being held in your hand.", + "E": "It was directly above you, positioned overhead rather than in your grasp." + }, + "answer_label": "C", + "answer_text": "It was in your hand.", + "q_id": "egolife_A1_D7_001_3_T2", + "correct_answer": "It was in your hand.", + "original_options": { + "A": "Directly in front of you.", + "B": "Directly behind you.", + "C": "It was in your hand.", + "D": "It was directly to your left, not being held in your hand.", + "E": "It was directly above you, positioned overhead rather than in your grasp." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 3, + 12, + 12 + ] + } +] diff --git a/QAs/egolife_A1_D7_002-mcq.json b/QAs/egolife_A1_D7_002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9b465980ca381b9e3a66ca9dfae4d8e3798e3eea --- /dev/null +++ b/QAs/egolife_A1_D7_002-mcq.json @@ -0,0 +1,910 @@ +[ + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_3", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Who was the last person to stand up?", + "answer": "The last person to stand up was the first girl on my right.", + "question_chinese": "上一个站起来的人是谁？", + "answer_chinese": "上一个站起来的人是我右边的第一个女生", + "evidence": [ + { + "description": "The first girl on my right stood up", + "timestamps": { + "start": "00:03:30", + "end": "00:03:34" + } + } + ], + "question_timestamps": "00:05:58", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The second girl on my right.", + "B": "The first girl on my left.", + "C": "The last person to stand up was the second girl on my left, not the first girl on my right.", + "D": "The last person to stand up was the first girl on my right.", + "E": "The last person to stand up was the last girl on my left, rather than the first girl on my right." + }, + "answer_label": "D", + "answer_text": "The last person to stand up was the first girl on my right.", + "q_id": "egolife_A1_D7_002_3", + "correct_answer": "The last person to stand up was the first girl on my right.", + "original_options": { + "A": "The second girl on my right.", + "B": "The first girl on my left.", + "C": "The last person to stand up was the second girl on my left, not the first girl on my right.", + "D": "The last person to stand up was the first girl on my right.", + "E": "The last person to stand up was the last girl on my left, rather than the first girl on my right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_4", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Who was the last person to stand up?", + "answer": "The last girl to stand up was the second one on my right.", + "question_chinese": "上一个站起来的人是谁？", + "answer_chinese": "上一个站起来的是我右边的第二个女生", + "evidence": [ + { + "description": "The second girl on my right stood up", + "timestamps": { + "start": "00:05:57", + "end": "00:06:02" + } + } + ], + "question_timestamps": "00:07:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The third one on my right.", + "B": "The first one in blue.", + "C": "The last girl to stand up was the second one on my right.", + "D": "The last girl to stand up was the first one wearing red, not the second one on my right.", + "E": "The last girl to stand up was the first one sitting down, not the second one on my right." + }, + "answer_label": "C", + "answer_text": "The last girl to stand up was the second one on my right.", + "q_id": "egolife_A1_D7_002_4", + "correct_answer": "The last girl to stand up was the second one on my right.", + "original_options": { + "A": "The third one on my right.", + "B": "The first one in blue.", + "C": "The last girl to stand up was the second one on my right.", + "D": "The last girl to stand up was the first one wearing red, not the second one on my right.", + "E": "The last girl to stand up was the first one sitting down, not the second one on my right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 19 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_3_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "The chopsticks you were holding in your hand just now.", + "question_chinese": "刚才我手里拿的什么？", + "answer_chinese": "刚才你手里拿的筷子", + "evidence": [ + { + "description": "The chopsticks in my hand now", + "timestamps": { + "start": "00:04:57", + "end": "00:04:58" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "A bowl.", + "B": "A spoon.", + "C": "A bowl and a pan in your hand, not chopsticks at all.", + "D": "The chopsticks you were holding in your hand just now.", + "E": "A paper handicraft you were carrying in your hand, not any eating utensils." + }, + "answer_label": "D", + "answer_text": "The chopsticks you were holding in your hand just now.", + "q_id": "egolife_A1_D7_002_3_T2", + "correct_answer": "The chopsticks you were holding in your hand just now.", + "original_options": { + "A": "A bowl.", + "B": "A spoon.", + "C": "A bowl and a pan in your hand, not chopsticks at all.", + "D": "The chopsticks you were holding in your hand just now.", + "E": "A paper handicraft you were carrying in your hand, not any eating utensils." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_3_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "The bowl and chopsticks you were holding in your hand just now.", + "question_chinese": "刚才我手里拿的什么？", + "answer_chinese": "刚才你手里拿的碗和筷子", + "evidence": [ + { + "description": "The bowl and chopsticks in my hand now", + "timestamps": { + "start": "00:07:09", + "end": "00:07:14" + } + } + ], + "question_timestamps": "00:08:39", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Just the chopsticks.", + "B": "The bowl and chopsticks you were holding in your hand just now.", + "C": "A paper handicraft.", + "D": "You were holding only a bowl in your hand, without any chopsticks with it.", + "E": "You were holding a bowl and a pan together, keeping both items in your hand at once." + }, + "answer_label": "B", + "answer_text": "The bowl and chopsticks you were holding in your hand just now.", + "q_id": "egolife_A1_D7_002_3_T1", + "correct_answer": "The bowl and chopsticks you were holding in your hand just now.", + "original_options": { + "A": "Just the chopsticks.", + "B": "The bowl and chopsticks you were holding in your hand just now.", + "C": "A paper handicraft.", + "D": "You were holding only a bowl in your hand, without any chopsticks with it.", + "E": "You were holding a bowl and a pan together, keeping both items in your hand at once." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002-mcq_fixed_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You were holding a paper handicraft in your hand just now.", + "question_chinese": "刚才我手里拿的什么？", + "answer_chinese": "我刚才手里拿着纸质手工品", + "evidence": [ + { + "description": "I'm holding a paper handicraft in my hand now", + "timestamps": { + "start": "00:09:15", + "end": "00:09:19" + } + } + ], + "question_timestamps": "00:09:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "A plastic toy.", + "B": "A metal key.", + "C": "A brochure or pamphlet with colorful printed designs, like something you picked up from an exhibit.", + "D": "A decorative postcard with intricate printed patterns, not a handmade paper craft.", + "E": "You were holding a paper handicraft in your hand just now." + }, + "answer_label": "E", + "answer_text": "You were holding a paper handicraft in your hand just now.", + "q_id": "egolife_A1_D7_002-mcq_fixed_9", + "correct_answer": "You were holding a paper handicraft in your hand just now.", + "original_options": { + "A": "A plastic toy.", + "B": "A metal key.", + "C": "A brochure or pamphlet with colorful printed designs, like something you picked up from an exhibit.", + "D": "A decorative postcard with intricate printed patterns, not a handmade paper craft.", + "E": "You were holding a paper handicraft in your hand just now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 12 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_7", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where am I in relation to the girl wearing a denim jacket?", + "answer": "I'm on the right of the girl in the denim jacket.", + "question_chinese": "我在牛仔外套女生的什么方位？", + "answer_chinese": "我在牛仔外套女生的右边", + "evidence": [ + { + "description": "He is on my left", + "timestamps": { + "start": "00:00:37", + "end": "00:00:38" + } + }, + { + "description": "We face the same direction", + "timestamps": { + "start": "00:00:38", + "end": "00:00:39" + } + } + ], + "question_timestamps": "00:02:03", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I'm on her left.", + "B": "I'm behind her.", + "C": "I'm on the right of the girl in the denim jacket.", + "D": "I'm standing directly in front of the girl in the denim jacket.", + "E": "I'm positioned opposite the girl in the denim jacket, facing her across the space." + }, + "answer_label": "C", + "answer_text": "I'm on the right of the girl in the denim jacket.", + "q_id": "egolife_A1_D7_002_7", + "correct_answer": "I'm on the right of the girl in the denim jacket.", + "original_options": { + "A": "I'm on her left.", + "B": "I'm behind her.", + "C": "I'm on the right of the girl in the denim jacket.", + "D": "I'm standing directly in front of the girl in the denim jacket.", + "E": "I'm positioned opposite the girl in the denim jacket, facing her across the space." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_8", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where am I positioned in relation to the girl in the denim jacket?", + "answer": "I'm right opposite her.", + "question_chinese": "我在牛仔外套女生的什么方位？", + "answer_chinese": "我在她正对面", + "evidence": [ + { + "description": "She is opposite me", + "timestamps": { + "start": "00:08:21", + "end": "00:08:22" + } + }, + { + "description": "She is facing me", + "timestamps": { + "start": "00:08:23", + "end": "00:08:26" + } + } + ], + "question_timestamps": "00:09:02", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "I'm to her left.", + "B": "I'm right opposite her.", + "C": "I'm behind her.", + "D": "I'm standing in front of her rather than directly across from her.", + "E": "I'm positioned to her right side, not facing her from the opposite position." + }, + "answer_label": "B", + "answer_text": "I'm right opposite her.", + "q_id": "egolife_A1_D7_002_8", + "correct_answer": "I'm right opposite her.", + "original_options": { + "A": "I'm to her left.", + "B": "I'm right opposite her.", + "C": "I'm behind her.", + "D": "I'm standing in front of her rather than directly across from her.", + "E": "I'm positioned to her right side, not facing her from the opposite position." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_5_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the girl in the denim jacket relative to me?", + "answer": "She is to your left.", + "question_chinese": "牛仔外套女生在我的什么方位？", + "answer_chinese": "她在你的左边", + "evidence": [ + { + "description": "She is on my left", + "timestamps": { + "start": "00:04:50", + "end": "00:04:51" + } + } + ], + "question_timestamps": "00:04:53", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "She is to your right.", + "B": "She is to your left.", + "C": "She is behind you.", + "D": "She is directly in front of you rather than standing to your left side.", + "E": "She is across from you, facing you, instead of being positioned on your left." + }, + "answer_label": "B", + "answer_text": "She is to your left.", + "q_id": "egolife_A1_D7_002_5_T2", + "correct_answer": "She is to your left.", + "original_options": { + "A": "She is to your right.", + "B": "She is to your left.", + "C": "She is behind you.", + "D": "She is directly in front of you rather than standing to your left side.", + "E": "She is across from you, facing you, instead of being positioned on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_5_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where was the girl in the denim jacket relative to me?", + "answer": "She is directly in front of you.", + "question_chinese": "牛仔外套女生在我的什么方位？", + "answer_chinese": "她在你的正对面", + "evidence": [ + { + "description": "She is opposite me", + "timestamps": { + "start": "00:08:21", + "end": "00:08:22" + } + }, + { + "description": "She is facing me", + "timestamps": { + "start": "00:08:23", + "end": "00:08:26" + } + } + ], + "question_timestamps": "00:09:02", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "She is directly behind you.", + "B": "She is directly to your left.", + "C": "She is directly in front of you.", + "D": "She is directly to your right rather than standing straight in front of you.", + "E": "She is positioned behind you, so you would need to turn around to face her." + }, + "answer_label": "C", + "answer_text": "She is directly in front of you.", + "q_id": "egolife_A1_D7_002_5_T1", + "correct_answer": "She is directly in front of you.", + "original_options": { + "A": "She is directly behind you.", + "B": "She is directly to your left.", + "C": "She is directly in front of you.", + "D": "She is directly to your right rather than standing straight in front of you.", + "E": "She is positioned behind you, so you would need to turn around to face her." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bites of food have you had so far?", + "answer": "So far, I've eaten three bites.", + "question_chinese": "目前为止吃了几口东西？", + "answer_chinese": "目前为止吃了3口东西", + "evidence": [ + { + "description": "I'm eating", + "timestamps": { + "start": "00:00:17", + "end": "00:00:20" + } + }, + { + "description": "I'm eating", + "timestamps": { + "start": "00:00:25", + "end": "00:00:27" + } + }, + { + "description": "I'm eating", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + } + ], + "question_timestamps": "00:01:15", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "So far, I've eaten two bites.", + "B": "So far, I've eaten three bites.", + "C": "So far, I've eaten four bites.", + "D": "Up to this point, I've only taken one bite of food in total.", + "E": "By now, I've already had five bites, counting each time I chewed and swallowed." + }, + "answer_label": "B", + "answer_text": "So far, I've eaten three bites.", + "q_id": "egolife_A1_D7_002_1", + "correct_answer": "So far, I've eaten three bites.", + "original_options": { + "A": "So far, I've eaten two bites.", + "B": "So far, I've eaten three bites.", + "C": "So far, I've eaten four bites.", + "D": "Up to this point, I've only taken one bite of food in total.", + "E": "By now, I've already had five bites, counting each time I chewed and swallowed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bites of food have you had so far?", + "answer": "So far, I've eaten five bites.", + "question_chinese": "目前为止吃了几口东西？", + "answer_chinese": "目前为止吃了5口东西", + "evidence": [ + { + "description": "I'm eating", + "timestamps": { + "start": "00:00:17", + "end": "00:00:20" + } + }, + { + "description": "I'm eating", + "timestamps": { + "start": "00:00:25", + "end": "00:00:27" + } + }, + { + "description": "I'm eating", + "timestamps": { + "start": "00:00:33", + "end": "00:00:35" + } + }, + { + "description": "I'm eating", + "timestamps": { + "start": "00:01:08", + "end": "00:01:11" + } + }, + { + "description": "I'm eating", + "timestamps": { + "start": "00:01:13", + "end": "00:01:14" + } + } + ], + "question_timestamps": "00:01:25", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "So far, I've eaten five bites.", + "B": "So far, I've eaten four bites.", + "C": "So far, I've eaten six bites.", + "D": "So far, I've eaten three bites, not counting any sips or drinks.", + "E": "At this point, I've had seven bites altogether, including the small ones I just took." + }, + "answer_label": "A", + "answer_text": "So far, I've eaten five bites.", + "q_id": "egolife_A1_D7_002_2", + "correct_answer": "So far, I've eaten five bites.", + "original_options": { + "A": "So far, I've eaten five bites.", + "B": "So far, I've eaten four bites.", + "C": "So far, I've eaten six bites.", + "D": "So far, I've eaten three bites, not counting any sips or drinks.", + "E": "At this point, I've had seven bites altogether, including the small ones I just took." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_6_T2", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Besides me, what route did the person who was walking around just now take?", + "answer": "You saw her leave the table.", + "question_chinese": "除我之外，刚才走动的人走了什么路线？", + "answer_chinese": "你看见她离开了桌子", + "evidence": [ + { + "description": "She is leaving.", + "timestamps": { + "start": "00:03:33", + "end": "00:03:35" + } + } + ], + "question_timestamps": "00:04:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "She circled half the table.", + "B": "She walked around the whole table.", + "C": "Besides you, she went around half the table starting from the right-hand side.", + "D": "You saw her leave the table.", + "E": "She walked around half the table from the left-hand side, then turned and returned the same way." + }, + "answer_label": "D", + "answer_text": "You saw her leave the table.", + "q_id": "egolife_A1_D7_002_6_T2", + "correct_answer": "You saw her leave the table.", + "original_options": { + "A": "She circled half the table.", + "B": "She walked around the whole table.", + "C": "Besides you, she went around half the table starting from the right-hand side.", + "D": "You saw her leave the table.", + "E": "She walked around half the table from the left-hand side, then turned and returned the same way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_6_T1", + "qid_num": 15, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Besides me, what route did the person who was walking around just now take?", + "answer": "After leaving, she came back from inside.", + "question_chinese": "除我之外，刚才走动的人走了什么路线？", + "answer_chinese": "她离开后又从室内回来", + "evidence": [ + { + "description": "She is leaving.", + "timestamps": { + "start": "00:03:33", + "end": "00:03:35" + } + }, + { + "description": "She is coming back from indoors", + "timestamps": { + "start": "00:05:13", + "end": "00:05:17" + } + } + ], + "question_timestamps": "00:06:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw her leave the table.", + "B": "After leaving, she came back from outside.", + "C": "Besides you, she went around half the table starting from the left-hand side.", + "D": "After leaving, she walked around half the table from the right-hand side, then stopped near the table.", + "E": "After leaving, she came back from inside." + }, + "answer_label": "E", + "answer_text": "After leaving, she came back from inside.", + "q_id": "egolife_A1_D7_002_6_T1", + "correct_answer": "After leaving, she came back from inside.", + "original_options": { + "A": "You saw her leave the table.", + "B": "After leaving, she came back from outside.", + "C": "Besides you, she went around half the table starting from the left-hand side.", + "D": "After leaving, she walked around half the table from the right-hand side, then stopped near the table.", + "E": "After leaving, she came back from inside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_7_T2", + "qid_num": 16, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Besides me, what route did the person who was walking around just now take?", + "answer": "Besides you, the person who was moving around just now went around half the table starting from the left-hand side.", + "question_chinese": "除我之外，刚才走动的人走了什么路线？", + "answer_chinese": "除你之外，刚才走动的人从左手边绕桌子半圈", + "evidence": [ + { + "description": "She walks half a circle around the table from his left hand", + "timestamps": { + "start": "00:08:10", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:08:22", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw her leave the table.", + "B": "Besides you, the person who was moving around just now went around half the table starting from the left-hand side.", + "C": "She started from the right-hand side.", + "D": "After leaving, she came back from inside rather than continuing around the table.", + "E": "Besides you, she went around half the table from the left-hand side, then returned the same way." + }, + "answer_label": "B", + "answer_text": "Besides you, the person who was moving around just now went around half the table starting from the left-hand side.", + "q_id": "egolife_A1_D7_002_7_T2", + "correct_answer": "Besides you, the person who was moving around just now went around half the table starting from the left-hand side.", + "original_options": { + "A": "You saw her leave the table.", + "B": "Besides you, the person who was moving around just now went around half the table starting from the left-hand side.", + "C": "She started from the right-hand side.", + "D": "After leaving, she came back from inside rather than continuing around the table.", + "E": "Besides you, she went around half the table from the left-hand side, then returned the same way." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_002", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_002_7_T1", + "qid_num": 17, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Besides me, what route did the person who was walking around just now take?", + "answer": "Besides you, the person who was moving around just now went around half the table starting from the left-hand side, then returned the same way.", + "question_chinese": "除我之外，刚才走动的人走了什么路线？", + "answer_chinese": "除你之外，刚才走动的人从左手边绕桌子半圈然后原路返回", + "evidence": [ + { + "description": "She walks half a circle around the table from his left hand", + "timestamps": { + "start": "00:08:10", + "end": "00:08:20" + } + }, + { + "description": "She retraces her steps", + "timestamps": { + "start": "00:09:05", + "end": "00:09:10" + } + } + ], + "question_timestamps": "00:09:33", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "She only went around half the table.", + "B": "Besides you, the person who was moving around just now went around half the table starting from the left-hand side, then returned the same way.", + "C": "She returned by a different route.", + "D": "Besides you, she went around half the table starting from the right-hand side, then returned the same way.", + "E": "Besides you, she walked around the whole table once and then came back to her original spot." + }, + "answer_label": "B", + "answer_text": "Besides you, the person who was moving around just now went around half the table starting from the left-hand side, then returned the same way.", + "q_id": "egolife_A1_D7_002_7_T1", + "correct_answer": "Besides you, the person who was moving around just now went around half the table starting from the left-hand side, then returned the same way.", + "original_options": { + "A": "She only went around half the table.", + "B": "Besides you, the person who was moving around just now went around half the table starting from the left-hand side, then returned the same way.", + "C": "She returned by a different route.", + "D": "Besides you, she went around half the table starting from the right-hand side, then returned the same way.", + "E": "Besides you, she walked around the whole table once and then came back to her original spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 17 + ] + } +] diff --git a/QAs/egolife_A1_D7_003-mcq.json b/QAs/egolife_A1_D7_003-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..74af0e9a6fe19be2c2940b9b7802dd7f4bd1f6f0 --- /dev/null +++ b/QAs/egolife_A1_D7_003-mcq.json @@ -0,0 +1,559 @@ +[ + { + "video_uid": "egolife_A1_D7_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_003_10", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Who was the last person to stand up?", + "answer": "The last one to stand up was the first girl on my right.", + "question_chinese": "上一个站起来的是谁？", + "answer_chinese": "上一个站起来的是我右手边第一个女生", + "evidence": [ + { + "description": "The first girl on my left is standing up", + "timestamps": { + "start": "00:04:18", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:05:53", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Second girl on my right.", + "B": "The last one to stand up was the first girl on my right.", + "C": "Last girl on my right.", + "D": "The last one to stand up was the first girl on my left, not the one on my right.", + "E": "The last one to stand up was the second girl on my left, after everyone else stayed seated." + }, + "answer_label": "B", + "answer_text": "The last one to stand up was the first girl on my right.", + "q_id": "egolife_A1_D7_003_10", + "correct_answer": "The last one to stand up was the first girl on my right.", + "original_options": { + "A": "Second girl on my right.", + "B": "The last one to stand up was the first girl on my right.", + "C": "Last girl on my right.", + "D": "The last one to stand up was the first girl on my left, not the one on my right.", + "E": "The last one to stand up was the second girl on my left, after everyone else stayed seated." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_003_9", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "Who was the last person to stand up?", + "answer": "The last girl to stand up was the second one on my right.", + "question_chinese": "上一个站起来的是谁？", + "answer_chinese": "上一个站起来的是我右手边第二个女生", + "evidence": [ + { + "description": "The second girl on my right is standing up", + "timestamps": { + "start": "00:09:42", + "end": "00:09:44" + } + } + ], + "question_timestamps": "00:10:48", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The first girl on my right.", + "B": "The first girl on my left.", + "C": "The last girl to stand up was the second one on my right.", + "D": "The last girl to stand up was the third one on my right, closest to the edge of the group.", + "E": "The last girl to stand up was the second one on my left, after the others had already risen." + }, + "answer_label": "C", + "answer_text": "The last girl to stand up was the second one on my right.", + "q_id": "egolife_A1_D7_003_9", + "correct_answer": "The last girl to stand up was the second one on my right.", + "original_options": { + "A": "The first girl on my right.", + "B": "The first girl on my left.", + "C": "The last girl to stand up was the second one on my right.", + "D": "The last girl to stand up was the third one on my right, closest to the edge of the group.", + "E": "The last girl to stand up was the second one on my left, after the others had already risen." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 19 + ] + }, + { + "video_uid": "egolife_A1_D7_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_003_2_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You were holding your mobile phone in your hand just now.", + "question_chinese": "我刚才手里拿着什么？", + "answer_chinese": "我刚才手里拿着手机", + "evidence": [ + { + "description": "I'm holding my mobile phone in my hand now", + "timestamps": { + "start": "00:03:59", + "end": "00:04:02" + } + } + ], + "question_timestamps": "00:04:50", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were holding your wallet.", + "B": "You were holding a set of keys.", + "C": "You were holding your mobile phone in your hand just now.", + "D": "You were holding a tablet in your hand just now, not your phone.", + "E": "You were holding a book in your hand just now while looking down at it." + }, + "answer_label": "C", + "answer_text": "You were holding your mobile phone in your hand just now.", + "q_id": "egolife_A1_D7_003_2_T1", + "correct_answer": "You were holding your mobile phone in your hand just now.", + "original_options": { + "A": "You were holding your wallet.", + "B": "You were holding a set of keys.", + "C": "You were holding your mobile phone in your hand just now.", + "D": "You were holding a tablet in your hand just now, not your phone.", + "E": "You were holding a book in your hand just now while looking down at it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_003_1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding just now?", + "answer": "I was holding my mobile phone in my hand just now.", + "question_chinese": "我刚才手里拿着什么？", + "answer_chinese": "我刚才手里拿着手机", + "evidence": [ + { + "description": "I'm holding my mobile phone in my hand now", + "timestamps": { + "start": "00:03:59", + "end": "00:04:02" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I was holding my tablet.", + "B": "I was holding my wallet.", + "C": "I was holding my mobile phone in my hand just now.", + "D": "I was holding a set of keys in my hand just now, not a phone.", + "E": "I was holding a book in my hand just now while I moved around." + }, + "answer_label": "C", + "answer_text": "I was holding my mobile phone in my hand just now.", + "q_id": "egolife_A1_D7_003_1", + "correct_answer": "I was holding my mobile phone in my hand just now.", + "original_options": { + "A": "I was holding my tablet.", + "B": "I was holding my wallet.", + "C": "I was holding my mobile phone in my hand just now.", + "D": "I was holding a set of keys in my hand just now, not a phone.", + "E": "I was holding a book in my hand just now while I moved around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_003_2_T3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What was I holding in my hand just now?", + "answer": "You were holding a bowl and chopsticks in your hand just now.", + "question_chinese": "我刚才手里拿着什么？", + "answer_chinese": "我刚才手里拿着碗和筷子", + "evidence": [ + { + "description": "I'm holding a bowl and chopsticks in my hand now", + "timestamps": { + "start": "00:07:10", + "end": "00:07:17" + } + } + ], + "question_timestamps": "00:07:55", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were holding your mobile phone.", + "B": "You were holding a bowl and chopsticks in your hand just now.", + "C": "You were holding a bowl only.", + "D": "You were holding a bowl and a spoon in your hand just now instead of chopsticks.", + "E": "You were holding chopsticks in your hand just now, but not a bowl at the same time." + }, + "answer_label": "B", + "answer_text": "You were holding a bowl and chopsticks in your hand just now.", + "q_id": "egolife_A1_D7_003_2_T3", + "correct_answer": "You were holding a bowl and chopsticks in your hand just now.", + "original_options": { + "A": "You were holding your mobile phone.", + "B": "You were holding a bowl and chopsticks in your hand just now.", + "C": "You were holding a bowl only.", + "D": "You were holding a bowl and a spoon in your hand just now instead of chopsticks.", + "E": "You were holding chopsticks in your hand just now, but not a bowl at the same time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_003_7", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many bites of food did I eat?", + "answer": "I took three bites.", + "question_chinese": "我吃了几口东西？", + "answer_chinese": "我吃了3口东西", + "evidence": [ + { + "description": "I'm taking a bite", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "I'm taking a bite", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + }, + { + "description": "I'm taking a bite", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:01:03", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I took two bites.", + "B": "I took four bites.", + "C": "I took five bites in total, with brief pauses between each bite.", + "D": "I took six bites overall, finishing the last bite after a short pause.", + "E": "I took three bites." + }, + "answer_label": "E", + "answer_text": "I took three bites.", + "q_id": "egolife_A1_D7_003_7", + "correct_answer": "I took three bites.", + "original_options": { + "A": "I took two bites.", + "B": "I took four bites.", + "C": "I took five bites in total, with brief pauses between each bite.", + "D": "I took six bites overall, finishing the last bite after a short pause.", + "E": "I took three bites." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_003_8", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How many bites of food did I eat?", + "answer": "I ate five bites.", + "question_chinese": "我吃了几口东西？", + "answer_chinese": "我吃了5口东西", + "evidence": [ + { + "description": "I'm taking a bite", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "I'm taking a bite", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + }, + { + "description": "I'm taking a bite", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + }, + { + "description": "I'm taking a bite", + "timestamps": { + "start": "00:01:24", + "end": "00:01:25" + } + }, + { + "description": "I'm taking a bite", + "timestamps": { + "start": "00:01:30", + "end": "00:01:31" + } + } + ], + "question_timestamps": "00:01:53", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "I ate three bites.", + "B": "I ate four bites.", + "C": "I ate five bites.", + "D": "I ate two bites total, taking only a couple mouthfuls before stopping.", + "E": "I ate six bites altogether, continuing for one more bite than you remembered." + }, + "answer_label": "C", + "answer_text": "I ate five bites.", + "q_id": "egolife_A1_D7_003_8", + "correct_answer": "I ate five bites.", + "original_options": { + "A": "I ate three bites.", + "B": "I ate four bites.", + "C": "I ate five bites.", + "D": "I ate two bites total, taking only a couple mouthfuls before stopping.", + "E": "I ate six bites altogether, continuing for one more bite than you remembered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_003_3", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the umbrella without going through the door?", + "answer": "I can run into an umbrella without going through the door.", + "question_chinese": "我能不经过门就碰到遮阳伞吗？", + "answer_chinese": "我能不经过门就碰到遮阳伞", + "evidence": [ + { + "description": "The sunshade is in the yard", + "timestamps": { + "start": "00:06:40", + "end": "00:06:45" + } + }, + { + "description": "I'm also in the yard", + "timestamps": { + "start": "00:06:58", + "end": "00:07:00" + } + } + ], + "question_timestamps": "00:09:03", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I can run into an umbrella without going through the door.", + "B": "No, you can’t reach it without going through the door." + }, + "answer_label": "A", + "answer_text": "I can run into an umbrella without going through the door.", + "q_id": "egolife_A1_D7_003_3", + "correct_answer": "I can run into an umbrella without going through the door.", + "original_options": { + "A": "I can run into an umbrella without going through the door.", + "B": "No, you can’t reach it without going through the door." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "egolife_A1_D7_003", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_003_4", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the umbrella without going through the door?", + "answer": "I couldn't avoid bumping into an umbrella just by passing through the door.", + "question_chinese": "我能不经过门就碰到遮阳伞吗？", + "answer_chinese": "我不能不经过门就碰到遮阳伞", + "evidence": [ + { + "description": "The sunshade is in the yard", + "timestamps": { + "start": "00:06:40", + "end": "00:06:45" + } + }, + { + "description": "I'm entering the room", + "timestamps": { + "start": "00:11:45", + "end": "00:11:46" + } + } + ], + "question_timestamps": "00:11:58", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "I couldn't avoid bumping into an umbrella just by passing through the door.", + "B": "Yes, you can reach the umbrella without going through the door." + }, + "answer_label": "A", + "answer_text": "I couldn't avoid bumping into an umbrella just by passing through the door.", + "q_id": "egolife_A1_D7_003_4", + "correct_answer": "I couldn't avoid bumping into an umbrella just by passing through the door.", + "original_options": { + "A": "I couldn't avoid bumping into an umbrella just by passing through the door.", + "B": "Yes, you can reach the umbrella without going through the door." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + } +] diff --git a/QAs/egolife_A1_D7_005-mcq.json b/QAs/egolife_A1_D7_005-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b44af4a774fec865722a79e45a08bda9881a4117 --- /dev/null +++ b/QAs/egolife_A1_D7_005-mcq.json @@ -0,0 +1,809 @@ +[ + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of mineral water are there in the bedroom on the first floor?", + "answer": "One bottle.", + "question_chinese": "一楼卧室里有几瓶矿泉水？", + "answer_chinese": "一瓶", + "evidence": [ + { + "description": "There is a bottle of mineral water on the table in the bedroom on the first floor", + "timestamps": { + "start": "00:10:16", + "end": "00:10:24" + } + } + ], + "question_timestamps": "00:10:47", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two bottles.", + "B": "Three bottles.", + "C": "Four bottles of mineral water are in the bedroom on the first floor.", + "D": "One bottle.", + "E": "There are five bottles of mineral water in the bedroom on the first floor." + }, + "answer_label": "D", + "answer_text": "One bottle.", + "q_id": "egolife_A1_D7_005_7", + "correct_answer": "One bottle.", + "original_options": { + "A": "Two bottles.", + "B": "Three bottles.", + "C": "Four bottles of mineral water are in the bedroom on the first floor.", + "D": "One bottle.", + "E": "There are five bottles of mineral water in the bedroom on the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of mineral water are there in the bedroom on the first floor?", + "answer": "Zero bottle.", + "question_chinese": "一楼卧室里有几瓶矿泉水？", + "answer_chinese": "零", + "evidence": [ + { + "description": "There is a bottle of mineral water on the table in the bedroom on the first floor", + "timestamps": { + "start": "00:10:16", + "end": "00:10:24" + } + }, + { + "description": "I took a bottle of mineral water from the bedroom table", + "timestamps": { + "start": "00:11:16", + "end": "00:11:29" + } + } + ], + "question_timestamps": "00:11:55", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One bottle.", + "B": "Two bottles.", + "C": "Three bottles of mineral water are in the bedroom on the first floor.", + "D": "Zero bottle.", + "E": "There are four bottles of mineral water in the bedroom on the first floor." + }, + "answer_label": "D", + "answer_text": "Zero bottle.", + "q_id": "egolife_A1_D7_005_8", + "correct_answer": "Zero bottle.", + "original_options": { + "A": "One bottle.", + "B": "Two bottles.", + "C": "Three bottles of mineral water are in the bedroom on the first floor.", + "D": "Zero bottle.", + "E": "There are four bottles of mineral water in the bedroom on the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_3", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to where I can get some juice?", + "answer": "Walk straight into the staircase, then turn right and go downstairs. After reaching the first floor, walk straight to the table in the yard Walk straight into the staircase, then turn right and go downstairs. After reaching the first floor, walk straight to the table in the yard.", + "question_chinese": "我怎么走才能拿到果汁？", + "answer_chinese": "直走进入楼梯口，然后右转并下楼，到达一楼后直走到院子的桌子旁", + "evidence": [ + { + "description": "The juice is on the table in the yard", + "timestamps": { + "start": "00:03:51", + "end": "00:04:06" + } + }, + { + "description": "To get from the yard to the living room on the second floor, you need to go straight up the stairs first, then turn left after reaching the second floor", + "timestamps": { + "start": "00:07:40", + "end": "00:07:58" + } + } + ], + "question_timestamps": "00:08:48", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Go to the stairwell, then turn left.", + "B": "Head to the staircase and go upstairs.", + "C": "Walk straight into the staircase, then turn right and go downstairs. After reaching the first floor, walk straight to the table in the yard Walk straight into the staircase, then turn right and go downstairs. After reaching the first floor, walk straight to the table in the yard.", + "D": "Walk straight into the staircase, turn left, go downstairs, then continue straight to the table in the yard.", + "E": "Walk into the stairwell, turn right, stay on the first floor, and go straight to the table in the yard." + }, + "answer_label": "C", + "answer_text": "Walk straight into the staircase, then turn right and go downstairs. After reaching the first floor, walk straight to the table in the yard Walk straight into the staircase, then turn right and go downstairs. After reaching the first floor, walk straight to the table in the yard.", + "q_id": "egolife_A1_D7_005_3", + "correct_answer": "Walk straight into the staircase, then turn right and go downstairs. After reaching the first floor, walk straight to the table in the yard Walk straight into the staircase, then turn right and go downstairs. After reaching the first floor, walk straight to the table in the yard.", + "original_options": { + "A": "Go to the stairwell, then turn left.", + "B": "Head to the staircase and go upstairs.", + "C": "Walk straight into the staircase, then turn right and go downstairs. After reaching the first floor, walk straight to the table in the yard Walk straight into the staircase, then turn right and go downstairs. After reaching the first floor, walk straight to the table in the yard.", + "D": "Walk straight into the staircase, turn left, go downstairs, then continue straight to the table in the yard.", + "E": "Walk into the stairwell, turn right, stay on the first floor, and go straight to the table in the yard." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 18, + 20 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_4", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the juice?", + "answer": "Walk straight into the staircase, then turn left and go straight to the table in the yard.", + "question_chinese": "我怎么走才能拿到果汁？", + "answer_chinese": "直走进入楼梯口，然后左转并直行到达院子里的桌子旁", + "evidence": [ + { + "description": "The juice is on the table in the yard", + "timestamps": { + "start": "00:03:51", + "end": "00:04:06" + } + }, + { + "description": "To walk from the living room on the first floor to the courtyard, you need to turn left once and then go straight", + "timestamps": { + "start": "00:10:06", + "end": "00:10:13" + } + } + ], + "question_timestamps": "00:11:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right at the staircase.", + "B": "Go down the stairs first.", + "C": "Walk straight into the staircase, then turn right and go straight to the table in the yard.", + "D": "Walk straight into the staircase, then turn left and go straight to the table in the yard.", + "E": "Walk into the stairwell, turn left, then keep going straight to the table in the living room." + }, + "answer_label": "D", + "answer_text": "Walk straight into the staircase, then turn left and go straight to the table in the yard.", + "q_id": "egolife_A1_D7_005_4", + "correct_answer": "Walk straight into the staircase, then turn left and go straight to the table in the yard.", + "original_options": { + "A": "Turn right at the staircase.", + "B": "Go down the stairs first.", + "C": "Walk straight into the staircase, then turn right and go straight to the table in the yard.", + "D": "Walk straight into the staircase, then turn left and go straight to the table in the yard.", + "E": "Walk into the stairwell, turn left, then keep going straight to the table in the living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_2_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk to get the juice?", + "answer": "First, go straight into the stairwell, then turn right and go downstairs.", + "question_chinese": "我是怎么走拿到果汁的？", + "answer_chinese": "首先你需要直走进入楼梯口，然后右转并下楼", + "evidence": [ + { + "description": "To get from the yard to the living room on the second floor, you need to go straight up the stairs first, then turn left after reaching the second floor", + "timestamps": { + "start": "00:07:40", + "end": "00:07:58" + } + } + ], + "question_timestamps": "00:08:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go downstairs.", + "B": "Go straight, then go upstairs.", + "C": "First, go straight into the stairwell, then turn left and go downstairs to reach the juice.", + "D": "First, go straight into the stairwell, then turn right and go downstairs.", + "E": "First, go straight into the stairwell, then turn right and go upstairs instead of going downstairs." + }, + "answer_label": "D", + "answer_text": "First, go straight into the stairwell, then turn right and go downstairs.", + "q_id": "egolife_A1_D7_005_2_T2", + "correct_answer": "First, go straight into the stairwell, then turn right and go downstairs.", + "original_options": { + "A": "Turn left, then go downstairs.", + "B": "Go straight, then go upstairs.", + "C": "First, go straight into the stairwell, then turn left and go downstairs to reach the juice.", + "D": "First, go straight into the stairwell, then turn right and go downstairs.", + "E": "First, go straight into the stairwell, then turn right and go upstairs instead of going downstairs." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_2_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I walk to get the juice?", + "answer": "First, go straight into the stairwell, then turn right and go downstairs. After you reach the first floor, go straight to the table in the yard.", + "question_chinese": "我是怎么走拿到果汁的？", + "answer_chinese": "首先你需要直走进入楼梯口，然后右转并下楼，到达一楼后直走到院子的桌子旁", + "evidence": [ + { + "description": "The juice is on the table in the yard", + "timestamps": { + "start": "00:03:51", + "end": "00:04:06" + } + }, + { + "description": "To get from the yard to the living room on the second floor, you need to go straight up the stairs first, then turn left after reaching the second floor", + "timestamps": { + "start": "00:07:40", + "end": "00:07:58" + } + } + ], + "question_timestamps": "00:08:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go downstairs.", + "B": "Go downstairs without turning right.", + "C": "First, go straight into the stairwell, then turn right and go downstairs, but stop before reaching the first floor.", + "D": "First, go straight into the stairwell, then turn right and go downstairs. After you reach the first floor, go straight to the table in the yard.", + "E": "First, go straight into the stairwell, turn right and go downstairs, then go straight to the table in the living room." + }, + "answer_label": "D", + "answer_text": "First, go straight into the stairwell, then turn right and go downstairs. After you reach the first floor, go straight to the table in the yard.", + "q_id": "egolife_A1_D7_005_2_T1", + "correct_answer": "First, go straight into the stairwell, then turn right and go downstairs. After you reach the first floor, go straight to the table in the yard.", + "original_options": { + "A": "Turn left, then go downstairs.", + "B": "Go downstairs without turning right.", + "C": "First, go straight into the stairwell, then turn right and go downstairs, but stop before reaching the first floor.", + "D": "First, go straight into the stairwell, then turn right and go downstairs. After you reach the first floor, go straight to the table in the yard.", + "E": "First, go straight into the stairwell, turn right and go downstairs, then go straight to the table in the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_5", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I last entered the living room on the first floor, which direction was the table from me?", + "answer": "Right ahead.", + "question_chinese": "我最近一次进入一楼客厅时，桌子在我的什么方向？", + "answer_chinese": "正前方", + "evidence": [ + { + "description": "The last time I entered the living room on the first floor, the table was right in front of me", + "timestamps": { + "start": "00:10:07", + "end": "00:10:11" + } + } + ], + "question_timestamps": "00:10:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right ahead.", + "B": "Left ahead.", + "C": "Directly behind you.", + "D": "The table was slightly to your left as you stepped into the living room.", + "E": "It was behind you on the right when you last entered the first-floor living room." + }, + "answer_label": "A", + "answer_text": "Right ahead.", + "q_id": "egolife_A1_D7_005_5", + "correct_answer": "Right ahead.", + "original_options": { + "A": "Right ahead.", + "B": "Left ahead.", + "C": "Directly behind you.", + "D": "The table was slightly to your left as you stepped into the living room.", + "E": "It was behind you on the right when you last entered the first-floor living room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_6", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I last entered the living room on the first floor, which direction was the table from me?", + "answer": "Front-left.", + "question_chinese": "我最近一次进入一楼客厅时，桌子在我的什么方向？", + "answer_chinese": "左前方", + "evidence": [ + { + "description": "The last time I entered the living room on the first floor, the table was in front of me to the left", + "timestamps": { + "start": "00:11:23", + "end": "00:11:27" + } + } + ], + "question_timestamps": "00:11:41", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Front-right.", + "B": "Front-left.", + "C": "Back-left.", + "D": "The table was to your back-right when you last entered the living room on the first floor.", + "E": "It was straight ahead and slightly right as you stepped into the first-floor living room." + }, + "answer_label": "B", + "answer_text": "Front-left.", + "q_id": "egolife_A1_D7_005_6", + "correct_answer": "Front-left.", + "original_options": { + "A": "Front-right.", + "B": "Front-left.", + "C": "Back-left.", + "D": "The table was to your back-right when you last entered the living room on the first floor.", + "E": "It was straight ahead and slightly right as you stepped into the first-floor living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get some juice now?", + "answer": "Yes.", + "question_chinese": "我现在可以拿到果汁吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The juice is on the table in the yard", + "timestamps": { + "start": "00:07:22", + "end": "00:07:28" + } + } + ], + "question_timestamps": "00:07:35", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D7_005_1", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get some juice now?", + "answer": "No.", + "question_chinese": "我现在可以拿到果汁吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The juice is on the table in the yard", + "timestamps": { + "start": "00:07:22", + "end": "00:07:28" + } + } + ], + "question_timestamps": "00:09:11", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D7_005_2", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_5_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of water did I see in the first-floor bedroom?", + "answer": "You saw one bottle of mineral water. It was on the table to your left.", + "question_chinese": "我看到一楼卧室里有几瓶矿泉水？", + "answer_chinese": "你看到有一瓶矿泉水，矿泉水在你左边的桌子上", + "evidence": [ + { + "description": "There is a bottle of mineral water on the table in the bedroom on the first floor", + "timestamps": { + "start": "00:10:16", + "end": "00:10:24" + } + } + ], + "question_timestamps": "00:10:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two bottles.", + "B": "No bottles.", + "C": "You saw three bottles of mineral water on the table to your left.", + "D": "You saw one bottle of mineral water. It was on the table to your left.", + "E": "You saw four bottles of mineral water, all grouped together on the table to your left." + }, + "answer_label": "D", + "answer_text": "You saw one bottle of mineral water. It was on the table to your left.", + "q_id": "egolife_A1_D7_005_5_T1", + "correct_answer": "You saw one bottle of mineral water. It was on the table to your left.", + "original_options": { + "A": "Two bottles.", + "B": "No bottles.", + "C": "You saw three bottles of mineral water on the table to your left.", + "D": "You saw one bottle of mineral water. It was on the table to your left.", + "E": "You saw four bottles of mineral water, all grouped together on the table to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_5_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of water did I see in the first-floor bedroom?", + "answer": "There wasn't any mineral water. You took a bottle from the bedroom, and the bedroom was behind you.", + "question_chinese": "我看到一楼卧室里有几瓶矿泉水？", + "answer_chinese": "没有矿泉水，你从卧室拿走了一瓶，卧室在你的后面", + "evidence": [ + { + "description": "There is a bottle of mineral water on the table in the bedroom on the first floor", + "timestamps": { + "start": "00:10:16", + "end": "00:10:24" + } + }, + { + "description": "I took a bottle of mineral water from the bedroom table", + "timestamps": { + "start": "00:11:16", + "end": "00:11:29" + } + } + ], + "question_timestamps": "00:11:51", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One bottle.", + "B": "Two bottles.", + "C": "You saw one bottle of mineral water in the first-floor bedroom before you turned around to leave.", + "D": "You saw three bottles of water in the bedroom, grouped together in the room as you passed through.", + "E": "There wasn't any mineral water. You took a bottle from the bedroom, and the bedroom was behind you." + }, + "answer_label": "E", + "answer_text": "There wasn't any mineral water. You took a bottle from the bedroom, and the bedroom was behind you.", + "q_id": "egolife_A1_D7_005_5_T2", + "correct_answer": "There wasn't any mineral water. You took a bottle from the bedroom, and the bedroom was behind you.", + "original_options": { + "A": "One bottle.", + "B": "Two bottles.", + "C": "You saw one bottle of mineral water in the first-floor bedroom before you turned around to leave.", + "D": "You saw three bottles of water in the bedroom, grouped together in the room as you passed through.", + "E": "There wasn't any mineral water. You took a bottle from the bedroom, and the bedroom was behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_3_T2", + "qid_num": 13, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the girl in blue relative to me?", + "answer": "In front of you.", + "question_chinese": "穿蓝色衣服的女孩在相对于我的什么位置？", + "answer_chinese": "在你的前面", + "evidence": [ + { + "description": "The girl in blue is in front of you", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:32", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "In front of you.", + "D": "She is standing to your right, not directly in front of you.", + "E": "She is positioned behind you, so you would need to turn around to face her." + }, + "answer_label": "C", + "answer_text": "In front of you.", + "q_id": "egolife_A1_D7_005_3_T2", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "In front of you.", + "D": "She is standing to your right, not directly in front of you.", + "E": "She is positioned behind you, so you would need to turn around to face her." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_005", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_005_3_T1", + "qid_num": 14, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Where is the girl in blue relative to me?", + "answer": "Behind you.", + "question_chinese": "穿蓝色衣服的女孩在相对于我的什么位置？", + "answer_chinese": "在你的后面", + "evidence": [ + { + "description": "The girl wearing blue clothes is behind you.", + "timestamps": { + "start": "00:07:30", + "end": "00:07:38" + } + } + ], + "question_timestamps": "00:07:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your right.", + "C": "Behind you.", + "D": "She is off to your left side rather than directly behind you.", + "E": "She is in front of you, within your forward view, not trailing behind you." + }, + "answer_label": "C", + "answer_text": "Behind you.", + "q_id": "egolife_A1_D7_005_3_T1", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your right.", + "C": "Behind you.", + "D": "She is off to your left side rather than directly behind you.", + "E": "She is in front of you, within your forward view, not trailing behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 14 + ] + } +] diff --git a/QAs/egolife_A1_D7_006-mcq.json b/QAs/egolife_A1_D7_006-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d5e07855a0a6cfbc1da3c8833ccb47037519f986 --- /dev/null +++ b/QAs/egolife_A1_D7_006-mcq.json @@ -0,0 +1,623 @@ +[ + { + "video_uid": "egolife_A1_D7_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_006_5", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What movements did I make from the moment I picked up my phone to when I put it down?", + "answer": "I walked straight from the bedroom door to the shelf.", + "question_chinese": "从我拿起手机到放下，这期间我是怎样运动的？", + "answer_chinese": "我从卧室门口径直走到架子前", + "evidence": [ + { + "description": "I walked straight from the bedroom door to the shelf", + "timestamps": { + "start": "00:06:00", + "end": "00:06:11" + } + } + ], + "question_timestamps": "00:06:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I walked to the couch.", + "B": "I walked to the armchair.", + "C": "I walked straight from the bedroom door to the kitchen counter, then stopped there.", + "D": "I walked straight from the bedroom door to the lamp, ending the movement there.", + "E": "I walked straight from the bedroom door to the shelf." + }, + "answer_label": "E", + "answer_text": "I walked straight from the bedroom door to the shelf.", + "q_id": "egolife_A1_D7_006_5", + "correct_answer": "I walked straight from the bedroom door to the shelf.", + "original_options": { + "A": "I walked to the couch.", + "B": "I walked to the armchair.", + "C": "I walked straight from the bedroom door to the kitchen counter, then stopped there.", + "D": "I walked straight from the bedroom door to the lamp, ending the movement there.", + "E": "I walked straight from the bedroom door to the shelf." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_006_6", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What movements did I make from the moment I picked up my phone to when I put it down?", + "answer": "I have been standing still.", + "question_chinese": "从我拿起手机到放下，这期间我是怎样运动的？", + "answer_chinese": "我一直站着不动", + "evidence": [ + { + "description": "I have been standing still", + "timestamps": { + "start": "00:06:55", + "end": "00:07:40" + } + } + ], + "question_timestamps": "00:07:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I walked forward.", + "B": "I turned around.", + "C": "I walked a few steps to the right before putting the phone down.", + "D": "I have been standing still.", + "E": "I moved toward the left side of the room instead of remaining in the same spot." + }, + "answer_label": "D", + "answer_text": "I have been standing still.", + "q_id": "egolife_A1_D7_006_6", + "correct_answer": "I have been standing still.", + "original_options": { + "A": "I walked forward.", + "B": "I turned around.", + "C": "I walked a few steps to the right before putting the phone down.", + "D": "I have been standing still.", + "E": "I moved toward the left side of the room instead of remaining in the same spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 13, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_006_9", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are currently charging at home?", + "answer": "Two.", + "question_chinese": "家里一共有几个充电宝正在充电？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "There is a power bank charging on the table", + "timestamps": { + "start": "00:04:01", + "end": "00:04:08" + } + }, + { + "description": "There is a power bank charging on the shelf", + "timestamps": { + "start": "00:04:32", + "end": "00:04:42" + } + } + ], + "question_timestamps": "00:08:19", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four power banks are plugged in and charging at home right now.", + "D": "Five power banks are currently charging at home on the power strip by the wall.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "egolife_A1_D7_006_9", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four power banks are plugged in and charging at home right now.", + "D": "Five power banks are currently charging at home on the power strip by the wall.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_006_10", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are currently charging at home?", + "answer": "Six.", + "question_chinese": "家里一共有几个充电宝正在充电？", + "answer_chinese": "六个", + "evidence": [ + { + "description": "There is a power bank charging on the table", + "timestamps": { + "start": "00:04:01", + "end": "00:04:08" + } + }, + { + "description": "There is a power bank charging on the shelf", + "timestamps": { + "start": "00:04:32", + "end": "00:04:42" + } + }, + { + "description": "There is a power bank on the ground charging", + "timestamps": { + "start": "00:11:22", + "end": "00:11:28" + } + }, + { + "description": "There is a power bank on the ground charging", + "timestamps": { + "start": "00:11:29", + "end": "00:11:35" + } + }, + { + "description": "There is a power bank charging on each side of the windowsill", + "timestamps": { + "start": "00:11:36", + "end": "00:11:51" + } + } + ], + "question_timestamps": "00:12:03", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Six.", + "C": "Seven.", + "D": "Four power banks are currently charging at home on the desk by the wall.", + "E": "Eight power banks are plugged in and charging at home right now." + }, + "answer_label": "B", + "answer_text": "Six.", + "q_id": "egolife_A1_D7_006_10", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Six.", + "C": "Seven.", + "D": "Four power banks are currently charging at home on the desk by the wall.", + "E": "Eight power banks are plugged in and charging at home right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 12 + ] + }, + { + "video_uid": "egolife_A1_D7_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_006_3", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was I the last time I picked up my phone?", + "answer": "Bedroom.", + "question_chinese": "我最近一次拿起手机时，是在什么地方？", + "answer_chinese": "卧室", + "evidence": [ + { + "description": "The last time I picked up my mobile phone was in the bedroom", + "timestamps": { + "start": "00:05:00", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:05:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Bedroom.", + "B": "Kitchen.", + "C": "Living room.", + "D": "You were in the hallway when you last picked up your phone.", + "E": "You last picked up your phone while standing in the bathroom near the sink." + }, + "answer_label": "A", + "answer_text": "Bedroom.", + "q_id": "egolife_A1_D7_006_3", + "correct_answer": "Bedroom.", + "original_options": { + "A": "Bedroom.", + "B": "Kitchen.", + "C": "Living room.", + "D": "You were in the hallway when you last picked up your phone.", + "E": "You last picked up your phone while standing in the bathroom near the sink." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_006_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where was I the last time I picked up my phone?", + "answer": "Living room.", + "question_chinese": "我最近一次拿起手机时，是在什么地方？", + "answer_chinese": "客厅", + "evidence": [ + { + "description": "The last time I picked up my mobile phone was in the living room", + "timestamps": { + "start": "00:09:40", + "end": "00:10:10" + } + } + ], + "question_timestamps": "00:10:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Bedroom.", + "B": "Living room.", + "C": "Kitchen.", + "D": "You were on the porch when you last picked up your phone.", + "E": "You last picked up your phone while standing in the hallway near the doorway." + }, + "answer_label": "B", + "answer_text": "Living room.", + "q_id": "egolife_A1_D7_006_4", + "correct_answer": "Living room.", + "original_options": { + "A": "Bedroom.", + "B": "Living room.", + "C": "Kitchen.", + "D": "You were on the porch when you last picked up your phone.", + "E": "You last picked up your phone while standing in the hallway near the doorway." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_006_11", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I last picked up my phone, which direction was the windowsill in relation to me?", + "answer": "Left.", + "question_chinese": "我最近一次拿起手机时，窗台在我的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "When I picked up my phone, the windowsill was on my left", + "timestamps": { + "start": "00:06:49", + "end": "00:06:57" + } + }, + { + "description": "The windowsill is on the side of the living room facing the yard", + "timestamps": { + "start": "00:03:14", + "end": "00:03:20" + } + } + ], + "question_timestamps": "00:08:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Left.", + "B": "Right.", + "C": "Behind you.", + "D": "The windowsill was directly in front of you when you picked up your phone.", + "E": "When you picked up your phone, the windowsill was on your right side." + }, + "answer_label": "A", + "answer_text": "Left.", + "q_id": "egolife_A1_D7_006_11", + "correct_answer": "Left.", + "original_options": { + "A": "Left.", + "B": "Right.", + "C": "Behind you.", + "D": "The windowsill was directly in front of you when you picked up your phone.", + "E": "When you picked up your phone, the windowsill was on your right side." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_006_12", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "What direction was the windowsill from me the last time I picked up my phone?", + "answer": "Right ahead.", + "question_chinese": "我最近一次拿起手机时，窗台在我的什么方向？", + "answer_chinese": "正前方", + "evidence": [ + { + "description": "Just as I picked up my phone, the windowsill was right in front of me", + "timestamps": { + "start": "00:09:41", + "end": "00:10:07" + } + } + ], + "question_timestamps": "00:10:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left ahead.", + "B": "Right beside you.", + "C": "The windowsill was directly behind you when you last picked up your phone.", + "D": "When you picked up your phone, the windowsill was a bit to your left.", + "E": "Right ahead." + }, + "answer_label": "E", + "answer_text": "Right ahead.", + "q_id": "egolife_A1_D7_006_12", + "correct_answer": "Right ahead.", + "original_options": { + "A": "Left ahead.", + "B": "Right beside you.", + "C": "The windowsill was directly behind you when you last picked up your phone.", + "D": "When you picked up your phone, the windowsill was a bit to your left.", + "E": "Right ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_006_1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the water bottle now?", + "answer": "Yes.", + "question_chinese": "我现在能拿到水瓶吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The water bottle is on the table", + "timestamps": { + "start": "00:03:16", + "end": "00:03:21" + } + } + ], + "question_timestamps": "00:03:48", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D7_006_1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D7_006", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_006_2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the water bottle now?", + "answer": "No.", + "question_chinese": "我现在能拿到水瓶吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The water bottle is on the table", + "timestamps": { + "start": "00:03:16", + "end": "00:03:21" + } + }, + { + "description": "I'm in the bedroom at the moment and quite far from the water bottle", + "timestamps": { + "start": "00:06:42", + "end": "00:06:50" + } + } + ], + "question_timestamps": "00:06:57", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D7_006_2", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D7_007-mcq.json b/QAs/egolife_A1_D7_007-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e2ab7d13bcd6430ee512ba88f48954c34bf84901 --- /dev/null +++ b/QAs/egolife_A1_D7_007-mcq.json @@ -0,0 +1,1284 @@ +[ + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_4", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are charging at home?", + "answer": "Two.", + "question_chinese": "家里有多少充电宝在充电？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "There is a power bank charging on the shelf", + "timestamps": { + "start": "00:00:15", + "end": "00:00:23" + } + }, + { + "description": "There is a power bank charging on the left side of the windowsill", + "timestamps": { + "start": "00:00:54", + "end": "00:01:03" + } + }, + { + "description": "There is a power bank charging on the shelf on the right side of the windowsill", + "timestamps": { + "start": "00:01:17", + "end": "00:01:20" + } + }, + { + "description": "I took away the power bank on the left side of the windowsill", + "timestamps": { + "start": "00:01:09", + "end": "00:01:23" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four power banks were charging at home simultaneously during that moment in the video.", + "E": "Five power banks were plugged in at home and charging together at the same time." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "egolife_A1_D7_007_4", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four power banks were charging at home simultaneously during that moment in the video.", + "E": "Five power banks were plugged in at home and charging together at the same time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_3", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many power banks are charging at home?", + "answer": "Five.", + "question_chinese": "家里有多少充电宝在充电？", + "answer_chinese": "五个", + "evidence": [ + { + "description": "There is a power bank charging on the shelf", + "timestamps": { + "start": "00:00:15", + "end": "00:00:23" + } + }, + { + "description": "There is a power bank charging on the left side of the windowsill", + "timestamps": { + "start": "00:00:54", + "end": "00:01:03" + } + }, + { + "description": "There is a power bank charging on the shelf on the right side of the windowsill", + "timestamps": { + "start": "00:01:17", + "end": "00:01:20" + } + }, + { + "description": "I took away the power bank on the left side of the windowsill", + "timestamps": { + "start": "00:01:09", + "end": "00:01:23" + } + }, + { + "description": "There is a power bank charging on the ground", + "timestamps": { + "start": "00:01:20", + "end": "00:01:22" + } + }, + { + "description": "There is a power bank charging on the table", + "timestamps": { + "start": "00:01:25", + "end": "00:01:28" + } + }, + { + "description": "I took a power bank and charged it", + "timestamps": { + "start": "00:01:54", + "end": "00:02:02" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Four.", + "C": "Three.", + "D": "Two power banks were charging at home, not a complete set of five total.", + "E": "Only one power bank was plugged in and charging at home at that moment." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D7_007_3", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "Four.", + "C": "Three.", + "D": "Two power banks were charging at home, not a complete set of five total.", + "E": "Only one power bank was plugged in and charging at home at that moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_1_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I put down the power bank just now, which direction was the windowsill from me?", + "answer": "The living-room window was to your right.", + "question_chinese": "我刚刚放下充电宝时，窗台在我的什么方向？", + "answer_chinese": "客厅的窗户在你的右方", + "evidence": [ + { + "description": "Just when I put down the power bank, the windowsill was right behind me", + "timestamps": { + "start": "00:01:20", + "end": "00:01:23" + } + }, + { + "description": "The windowsill is on the side of the living room close to the yard", + "timestamps": { + "start": "00:00:06", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:01:25", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The living-room window was to your right.", + "B": "It was to your left.", + "C": "It was in front of you.", + "D": "The living-room window was behind you when you set the power bank down.", + "E": "The windowsill was above you rather than to your right at that moment." + }, + "answer_label": "A", + "answer_text": "The living-room window was to your right.", + "q_id": "egolife_A1_D7_007_1_T2", + "correct_answer": "The living-room window was to your right.", + "original_options": { + "A": "The living-room window was to your right.", + "B": "It was to your left.", + "C": "It was in front of you.", + "D": "The living-room window was behind you when you set the power bank down.", + "E": "The windowsill was above you rather than to your right at that moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_1_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I put down the power bank just now, which direction was the windowsill from me?", + "answer": "Directly behind you.", + "question_chinese": "我刚刚放下充电宝时，窗台在我的什么方向？", + "answer_chinese": "正后方", + "evidence": [ + { + "description": "Just when I put down the power bank, the windowsill was right behind me", + "timestamps": { + "start": "00:01:20", + "end": "00:01:24" + } + }, + { + "description": "The windowsill is on the side of the living room close to the yard", + "timestamps": { + "start": "00:00:06", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:01:28", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly in front of you.", + "B": "Directly behind you.", + "C": "Directly to your left.", + "D": "Directly to your right, not behind you, when you put the power bank down.", + "E": "Directly above you, rather than directly behind you, at that moment in time." + }, + "answer_label": "B", + "answer_text": "Directly behind you.", + "q_id": "egolife_A1_D7_007_1_T1", + "correct_answer": "Directly behind you.", + "original_options": { + "A": "Directly in front of you.", + "B": "Directly behind you.", + "C": "Directly to your left.", + "D": "Directly to your right, not behind you, when you put the power bank down.", + "E": "Directly above you, rather than directly behind you, at that moment in time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_2_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I put down the power bank just now, which direction was the windowsill from me?", + "answer": "To your front-right.", + "question_chinese": "我刚刚放下充电宝时，窗台在我的什么方向？", + "answer_chinese": "右前方", + "evidence": [ + { + "description": "When I just put down the power bank, the windowsill was right in front of me", + "timestamps": { + "start": "00:01:57", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:02:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "It was directly behind you instead of being to your front-right at that moment.", + "D": "To your front-right.", + "E": "The living-room window was to your right, not positioned at your front-right when you set it down." + }, + "answer_label": "D", + "answer_text": "To your front-right.", + "q_id": "egolife_A1_D7_007_2_T1", + "correct_answer": "To your front-right.", + "original_options": { + "A": "To your front-left.", + "B": "To your back-right.", + "C": "It was directly behind you instead of being to your front-right at that moment.", + "D": "To your front-right.", + "E": "The living-room window was to your right, not positioned at your front-right when you set it down." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_2_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I put down the power bank just now, which direction was the windowsill from me?", + "answer": "The living-room window was behind you.", + "question_chinese": "我刚刚放下充电宝时，窗台在我的什么方向？", + "answer_chinese": "客厅的窗户在你的后方", + "evidence": [ + { + "description": "When I just put down the power bank, the windowsill was right in front of me", + "timestamps": { + "start": "00:01:57", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:02:48", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The living-room window was in front of you.", + "B": "The living-room window was to your left.", + "C": "The living-room window was behind you.", + "D": "The living-room window was to your right when you put the power bank down.", + "E": "The living-room window was above you, not behind you at that moment." + }, + "answer_label": "C", + "answer_text": "The living-room window was behind you.", + "q_id": "egolife_A1_D7_007_2_T2", + "correct_answer": "The living-room window was behind you.", + "original_options": { + "A": "The living-room window was in front of you.", + "B": "The living-room window was to your left.", + "C": "The living-room window was behind you.", + "D": "The living-room window was to your right when you put the power bank down.", + "E": "The living-room window was above you, not behind you at that moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 12 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_7_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "The last time I went up the stairs, how many people besides me were on the staircase?", + "answer": "There was no one on the stairs.", + "question_chinese": "我最近一次上楼梯时，楼梯里除我之外有几个人？", + "answer_chinese": "楼梯上没有人。", + "evidence": [ + { + "description": "When I just went up the stairs, I was the only one there", + "timestamps": { + "start": "00:06:18", + "end": "00:06:32" + } + } + ], + "question_timestamps": "00:07:11", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "There was no one on the stairs.", + "D": "There were three people on the stairs besides you as you went up.", + "E": "There were four other people on the staircase when you last climbed up." + }, + "answer_label": "C", + "answer_text": "There was no one on the stairs.", + "q_id": "egolife_A1_D7_007_7_T1", + "correct_answer": "There was no one on the stairs.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "There was no one on the stairs.", + "D": "There were three people on the stairs besides you as you went up.", + "E": "There were four other people on the staircase when you last climbed up." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_7_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "The last time I went up the stairs, how many people besides me were on the staircase?", + "answer": "At the stair entrance, you ran into a girl wearing yellow.", + "question_chinese": "我最近一次上楼梯时，楼梯里除我之外有几个人？", + "answer_chinese": "你在楼梯口碰见有穿黄衣服的女孩", + "evidence": [ + { + "description": "When I just went up the stairs, I was the only one there", + "timestamps": { + "start": "00:06:18", + "end": "00:06:32" + } + } + ], + "question_timestamps": "00:07:37", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You went up alone.", + "B": "You met a boy in blue.", + "C": "You ran into one person on the stairs, but not at the stair entrance.", + "D": "At the stair entrance, you ran into a girl wearing yellow.", + "E": "You ran into two people on the staircase while going up, not just a single girl in yellow." + }, + "answer_label": "D", + "answer_text": "At the stair entrance, you ran into a girl wearing yellow.", + "q_id": "egolife_A1_D7_007_7_T2", + "correct_answer": "At the stair entrance, you ran into a girl wearing yellow.", + "original_options": { + "A": "You went up alone.", + "B": "You met a boy in blue.", + "C": "You ran into one person on the stairs, but not at the stair entrance.", + "D": "At the stair entrance, you ran into a girl wearing yellow.", + "E": "You ran into two people on the staircase while going up, not just a single girl in yellow." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 14, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_8_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "The last time I went up the stairs, how many people besides me were on the staircase?", + "answer": "You ran into one person on the stairs.", + "question_chinese": "我最近一次上楼梯时，楼梯里除我之外有几个人？", + "answer_chinese": "你在楼梯上碰到了一个人。", + "evidence": [ + { + "description": "When I just went up the stairs, there was only one person in the stairs besides me", + "timestamps": { + "start": "00:09:17", + "end": "00:09:35" + } + } + ], + "question_timestamps": "00:09:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw no one on the stairs.", + "B": "You ran into two people on the stairs.", + "C": "You ran into three different people while going up the staircase, passing them as you climbed.", + "D": "You ran into one person on the stairs.", + "E": "You encountered two people on the staircase as you went up, one ahead of you and one coming down." + }, + "answer_label": "D", + "answer_text": "You ran into one person on the stairs.", + "q_id": "egolife_A1_D7_007_8_T1", + "correct_answer": "You ran into one person on the stairs.", + "original_options": { + "A": "You saw no one on the stairs.", + "B": "You ran into two people on the stairs.", + "C": "You ran into three different people while going up the staircase, passing them as you climbed.", + "D": "You ran into one person on the stairs.", + "E": "You encountered two people on the staircase as you went up, one ahead of you and one coming down." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_8_T2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "The last time I went up the stairs, how many people besides me were on the staircase?", + "answer": "There was nobody—you went down the stairs by yourself.", + "question_chinese": "我最近一次上楼梯时，楼梯里除我之外有几个人？", + "answer_chinese": "没有人，你一个人下的楼梯", + "evidence": [ + { + "description": "When I just went up the stairs, there was only one person in the stairs besides me", + "timestamps": { + "start": "00:09:17", + "end": "00:09:35" + } + } + ], + "question_timestamps": "00:09:53", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There was one person on the stairs.", + "B": "There were two people on the staircase.", + "C": "There were three people besides you on the staircase, so you were not alone while going down.", + "D": "You went down the stairs with one other person present, passing them as you moved along the steps.", + "E": "There was nobody—you went down the stairs by yourself." + }, + "answer_label": "E", + "answer_text": "There was nobody—you went down the stairs by yourself.", + "q_id": "egolife_A1_D7_007_8_T2", + "correct_answer": "There was nobody—you went down the stairs by yourself.", + "original_options": { + "A": "There was one person on the stairs.", + "B": "There were two people on the staircase.", + "C": "There were three people besides you on the staircase, so you were not alone while going down.", + "D": "You went down the stairs with one other person present, passing them as you moved along the steps.", + "E": "There was nobody—you went down the stairs by yourself." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_5", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the washing machine?", + "answer": "Turn left to the main gate, then turn right into the house and walk about three steps.", + "question_chinese": "我现在想要用洗衣机，应该怎么走？", + "answer_chinese": "向左转到正门，然后向右转进入房子，再走大约三步", + "evidence": [ + { + "description": "The washing machine is between the staircase on the first floor and the main entrance", + "timestamps": { + "start": "00:07:41", + "end": "00:08:02" + } + } + ], + "question_timestamps": "00:08:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right at the main gate.", + "B": "Go straight into the house.", + "C": "Turn left to the main gate, then go straight inside and walk about two steps forward.", + "D": "Turn left toward the main gate, then turn right into the house and walk about five steps.", + "E": "Turn left to the main gate, then turn right into the house and walk about three steps." + }, + "answer_label": "E", + "answer_text": "Turn left to the main gate, then turn right into the house and walk about three steps.", + "q_id": "egolife_A1_D7_007_5", + "correct_answer": "Turn left to the main gate, then turn right into the house and walk about three steps.", + "original_options": { + "A": "Turn right at the main gate.", + "B": "Go straight into the house.", + "C": "Turn left to the main gate, then go straight inside and walk about two steps forward.", + "D": "Turn left toward the main gate, then turn right into the house and walk about five steps.", + "E": "Turn left to the main gate, then turn right into the house and walk about three steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_6", + "qid_num": 12, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the washing machine?", + "answer": "Walk straight to the staircase on the second floor, turn right and go downstairs. After reaching the first floor, walk a few more steps forward.", + "question_chinese": "我现在想要用洗衣机，应该怎么走？", + "answer_chinese": "直走到达二楼楼梯口，右转下楼，到达一楼后再往前走几步", + "evidence": [ + { + "description": "The washing machine is between the staircase on the first floor and the main entrance", + "timestamps": { + "start": "00:07:41", + "end": "00:08:02" + } + }, + { + "description": "To get from the staircase on the first floor to the living room on the second floor, you need to go up the stairs and then turn left", + "timestamps": { + "start": "00:09:18", + "end": "00:09:37" + } + } + ], + "question_timestamps": "00:09:58", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Walk straight to the staircase on the second floor, turn right and go downstairs. After reaching the first floor, walk a few more steps forward.", + "B": "Turn left and go downstairs.", + "C": "Go upstairs to the third floor.", + "D": "Walk to the staircase on the second floor, turn left, go downstairs, then take a few steps to the right.", + "E": "Head to the staircase on the first floor, turn left and go downstairs to the ground floor, then step forward." + }, + "answer_label": "A", + "answer_text": "Walk straight to the staircase on the second floor, turn right and go downstairs. After reaching the first floor, walk a few more steps forward.", + "q_id": "egolife_A1_D7_007_6", + "correct_answer": "Walk straight to the staircase on the second floor, turn right and go downstairs. After reaching the first floor, walk a few more steps forward.", + "original_options": { + "A": "Walk straight to the staircase on the second floor, turn right and go downstairs. After reaching the first floor, walk a few more steps forward.", + "B": "Turn left and go downstairs.", + "C": "Go upstairs to the third floor.", + "D": "Walk to the staircase on the second floor, turn left, go downstairs, then take a few steps to the right.", + "E": "Head to the staircase on the first floor, turn left and go downstairs to the ground floor, then step forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_9", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I use the microwave now?", + "answer": "No.", + "question_chinese": "我现在能够到微波炉吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The microwave oven is on the right side of the windowsill", + "timestamps": { + "start": "00:00:35", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:01:05", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "egolife_A1_D7_007_9", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_10", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I use the microwave now?", + "answer": "Yes.", + "question_chinese": "我现在能够到微波炉吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The microwave oven is on the right side of the windowsill", + "timestamps": { + "start": "00:00:35", + "end": "00:00:41" + } + } + ], + "question_timestamps": "00:02:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "egolife_A1_D7_007_10", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_3_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In my home, how many power banks are charging?", + "answer": "Right now, there were two power banks charging at your home, on the tables to your left and directly in front of you.", + "question_chinese": "我的家里有多少充电宝在充电？", + "answer_chinese": "你现在的家里有两个充电宝在充电，分别在你左方和正前方的桌子上。", + "evidence": [ + { + "description": "There is a power bank charging on the shelf", + "timestamps": { + "start": "00:00:15", + "end": "00:00:23" + } + }, + { + "description": "There is a power bank charging on the left side of the windowsill", + "timestamps": { + "start": "00:00:54", + "end": "00:01:03" + } + }, + { + "description": "There is a power bank charging on the shelf on the right side of the windowsill", + "timestamps": { + "start": "00:01:17", + "end": "00:01:20" + } + }, + { + "description": "I took away the power bank on the left side of the windowsill", + "timestamps": { + "start": "00:01:20", + "end": "00:01:23" + } + } + ], + "question_timestamps": "00:01:25", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero power banks are charging.", + "B": "One power bank is charging.", + "C": "Right now, there are three power banks charging at home: one on your left table and two in front of you.", + "D": "Right now, there were two power banks charging at your home, on the tables to your left and directly in front of you.", + "E": "Right now, there are four power banks charging at your home, spread across the tables to your left and in front." + }, + "answer_label": "D", + "answer_text": "Right now, there were two power banks charging at your home, on the tables to your left and directly in front of you.", + "q_id": "egolife_A1_D7_007_3_T1", + "correct_answer": "Right now, there were two power banks charging at your home, on the tables to your left and directly in front of you.", + "original_options": { + "A": "Zero power banks are charging.", + "B": "One power bank is charging.", + "C": "Right now, there are three power banks charging at home: one on your left table and two in front of you.", + "D": "Right now, there were two power banks charging at your home, on the tables to your left and directly in front of you.", + "E": "Right now, there are four power banks charging at your home, spread across the tables to your left and in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 21 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_3_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In my home, how many power banks are charging?", + "answer": "Right now at home, there are two power banks charging: one by the window on your left side, and one on the cabinet on the right side.", + "question_chinese": "我的家里有多少充电宝在充电？", + "answer_chinese": "你现在的家里有两个充电宝在充电，在你左侧的窗子上，和右侧的柜子上。", + "evidence": [ + { + "description": "There is a power bank charging on the shelf", + "timestamps": { + "start": "00:00:15", + "end": "00:00:23" + } + }, + { + "description": "There is a power bank charging on the left side of the windowsill", + "timestamps": { + "start": "00:00:54", + "end": "00:01:03" + } + }, + { + "description": "There is a power bank charging on the shelf on the right side of the windowsill", + "timestamps": { + "start": "00:01:17", + "end": "00:01:20" + } + }, + { + "description": "I took away the power bank on the left side of the windowsill", + "timestamps": { + "start": "00:01:20", + "end": "00:01:23" + } + } + ], + "question_timestamps": "00:01:35", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "No power banks are charging.", + "B": "Only one power bank is charging.", + "C": "Right now at home, three power banks are charging: two by your left window and one on the right cabinet.", + "D": "Right now at home, four power banks are charging: two by the left window and two on your right cabinet.", + "E": "Right now at home, there are two power banks charging: one by the window on your left side, and one on the cabinet on the right side." + }, + "answer_label": "E", + "answer_text": "Right now at home, there are two power banks charging: one by the window on your left side, and one on the cabinet on the right side.", + "q_id": "egolife_A1_D7_007_3_T2", + "correct_answer": "Right now at home, there are two power banks charging: one by the window on your left side, and one on the cabinet on the right side.", + "original_options": { + "A": "No power banks are charging.", + "B": "Only one power bank is charging.", + "C": "Right now at home, three power banks are charging: two by your left window and one on the right cabinet.", + "D": "Right now at home, four power banks are charging: two by the left window and two on your right cabinet.", + "E": "Right now at home, there are two power banks charging: one by the window on your left side, and one on the cabinet on the right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 20, + 20 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_3_T3", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In my home, how many power banks are charging?", + "answer": "Right now at home, there are five power banks charging: by the window on your left side, on the cabinet on the right side, and on the table in front of you.", + "question_chinese": "我的家里有多少充电宝在充电？", + "answer_chinese": "你现在的家里有五个充电宝在充电，在你左侧的窗子上，和右侧的柜子上，在你面前的桌子上。", + "evidence": [ + { + "description": "There is a power bank charging on the shelf", + "timestamps": { + "start": "00:00:15", + "end": "00:00:23" + } + }, + { + "description": "There is a power bank charging on the left side of the windowsill", + "timestamps": { + "start": "00:00:54", + "end": "00:01:03" + } + }, + { + "description": "There is a power bank charging on the shelf on the right side of the windowsill", + "timestamps": { + "start": "00:01:17", + "end": "00:01:20" + } + }, + { + "description": "I took away the power bank on the left side of the windowsill", + "timestamps": { + "start": "00:01:21", + "end": "00:01:23" + } + }, + { + "description": "There is a power bank charging on the table", + "timestamps": { + "start": "00:01:25", + "end": "00:01:28" + } + }, + { + "description": "I took a power bank and charged it", + "timestamps": { + "start": "00:01:54", + "end": "00:02:02" + } + } + ], + "question_timestamps": "00:02:50", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Right now at home, there are five power banks charging: by the window on your left side, on the cabinet on the right side, and on the table in front of you.", + "B": "There are four power banks charging.", + "C": "There are six power banks charging.", + "D": "Right now there are three power banks charging: two by the window and one on the table.", + "E": "Right now there are seven power banks charging: three by the window, two on cabinet, two on table." + }, + "answer_label": "A", + "answer_text": "Right now at home, there are five power banks charging: by the window on your left side, on the cabinet on the right side, and on the table in front of you.", + "q_id": "egolife_A1_D7_007_3_T3", + "correct_answer": "Right now at home, there are five power banks charging: by the window on your left side, on the cabinet on the right side, and on the table in front of you.", + "original_options": { + "A": "Right now at home, there are five power banks charging: by the window on your left side, on the cabinet on the right side, and on the table in front of you.", + "B": "There are four power banks charging.", + "C": "There are six power banks charging.", + "D": "Right now there are three power banks charging: two by the window and one on the table.", + "E": "Right now there are seven power banks charging: three by the window, two on cabinet, two on table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_3_T4", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "In my home, how many power banks are charging?", + "answer": "Right now at home, there are five power banks charging: by the window on your left side, on the cabinet on the right side, and on the table in front of you.", + "question_chinese": "我的家里有多少充电宝在充电？", + "answer_chinese": "你现在的家里有五个充电宝在充电，在你左侧的窗子上，和右侧的柜子上，在你面前的桌子上。", + "evidence": [ + { + "description": "There is a power bank charging on the shelf", + "timestamps": { + "start": "00:00:15", + "end": "00:00:23" + } + }, + { + "description": "There is a power bank charging on the left side of the windowsill", + "timestamps": { + "start": "00:00:54", + "end": "00:01:03" + } + }, + { + "description": "There is a power bank charging on the shelf on the right side of the windowsill", + "timestamps": { + "start": "00:01:17", + "end": "00:01:20" + } + }, + { + "description": "I took away the power bank on the left side of the windowsill", + "timestamps": { + "start": "00:01:21", + "end": "00:01:23" + } + }, + { + "description": "There is a power bank charging on the table", + "timestamps": { + "start": "00:01:25", + "end": "00:01:28" + } + }, + { + "description": "I took a power bank and charged it", + "timestamps": { + "start": "00:01:54", + "end": "00:02:02" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are four power banks charging.", + "B": "Right now at home, there are five power banks charging: by the window on your left side, on the cabinet on the right side, and on the table in front of you.", + "C": "There are six power banks charging.", + "D": "Right now there are three power banks charging: two by the window and one on the table.", + "E": "Right now there are seven power banks charging: three by the window, two on cabinet, two on table." + }, + "answer_label": "B", + "answer_text": "Right now at home, there are five power banks charging: by the window on your left side, on the cabinet on the right side, and on the table in front of you.", + "q_id": "egolife_A1_D7_007_3_T4", + "correct_answer": "Right now at home, there are five power banks charging: by the window on your left side, on the cabinet on the right side, and on the table in front of you.", + "original_options": { + "A": "There are four power banks charging.", + "B": "Right now at home, there are five power banks charging: by the window on your left side, on the cabinet on the right side, and on the table in front of you.", + "C": "There are six power banks charging.", + "D": "Right now there are three power banks charging: two by the window and one on the table.", + "E": "Right now there are seven power banks charging: three by the window, two on cabinet, two on table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_7", + "qid_num": 19, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the table or the microwave?", + "answer": "Table.", + "question_chinese": "桌子和微波炉哪个离我近？", + "answer_chinese": "桌子", + "evidence": [ + { + "description": "The table is in the middle of the living room", + "timestamps": { + "start": "00:00:01", + "end": "00:00:09" + } + }, + { + "description": "The microwave oven is on the side of the living room near the windowsill", + "timestamps": { + "start": "00:00:43", + "end": "00:00:56" + } + } + ], + "question_timestamps": "00:02:59", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Microwave oven.", + "B": "Table." + }, + "answer_label": "B", + "answer_text": "Table.", + "q_id": "egolife_A1_D7_007_7", + "correct_answer": "Table.", + "original_options": { + "A": "Microwave oven.", + "B": "Table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "egolife_A1_D7_007", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_007_8", + "qid_num": 20, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the table or the microwave?", + "answer": "Microwave oven.", + "question_chinese": "桌子和微波炉哪个离我近？", + "answer_chinese": "微波炉", + "evidence": [ + { + "description": "The table is in the middle of the living room", + "timestamps": { + "start": "00:00:01", + "end": "00:00:09" + } + }, + { + "description": "The microwave oven is on the side of the living room near the windowsill", + "timestamps": { + "start": "00:00:43", + "end": "00:00:56" + } + } + ], + "question_timestamps": "00:08:43", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Microwave oven.", + "B": "Table." + }, + "answer_label": "A", + "answer_text": "Microwave oven.", + "q_id": "egolife_A1_D7_007_8", + "correct_answer": "Microwave oven.", + "original_options": { + "A": "Microwave oven.", + "B": "Table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/egolife_A1_D7_008-mcq.json b/QAs/egolife_A1_D7_008-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b4309e5fa99f56ec76c44d1c59a72d85dc0b9ed6 --- /dev/null +++ b/QAs/egolife_A1_D7_008-mcq.json @@ -0,0 +1,862 @@ +[ + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008-mcq_fixed_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my trajectory from the time I had the trash bag?", + "answer": "After you finish collecting garbage in the living room on the second floor, go downstairs with the garbage bag.", + "question_chinese": "从我带着垃圾袋，到现在我的运动轨迹是怎样的？", + "answer_chinese": "你在二楼客厅收集完垃圾后，带着垃圾袋下楼。", + "evidence": [ + { + "description": "You are collecting garbage into garbage bags in the living room on the second floor.", + "timestamps": { + "start": "00:01:55", + "end": "00:02:10" + } + }, + { + "description": "You're coming downstairs with a garbage bag.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:15" + } + } + ], + "question_timestamps": "00:04:16", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You stayed upstairs with the bag.", + "B": "After you finish collecting garbage in the living room on the second floor, go downstairs with the garbage bag.", + "C": "You went upstairs with the bag.", + "D": "After collecting garbage in the living room upstairs, you leave the bag by the front door.", + "E": "After collecting garbage in the living room upstairs, you carry the bag to the balcony and keep it." + }, + "answer_label": "B", + "answer_text": "After you finish collecting garbage in the living room on the second floor, go downstairs with the garbage bag.", + "q_id": "egolife_A1_D7_008-mcq_fixed_6", + "correct_answer": "After you finish collecting garbage in the living room on the second floor, go downstairs with the garbage bag.", + "original_options": { + "A": "You stayed upstairs with the bag.", + "B": "After you finish collecting garbage in the living room on the second floor, go downstairs with the garbage bag.", + "C": "You went upstairs with the bag.", + "D": "After collecting garbage in the living room upstairs, you leave the bag by the front door.", + "E": "After collecting garbage in the living room upstairs, you carry the bag to the balcony and keep it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008-mcq_fixed_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my trajectory from the time I had the trash bag?", + "answer": "After you collect garbage in the living room on the second floor, go downstairs with the garbage bag and put it next to the yard exit.", + "question_chinese": "从我带着垃圾袋，到现在我的运动轨迹是怎样的？", + "answer_chinese": "你在二楼客厅收集完垃圾后，带着垃圾袋下楼后将其放到院子出口旁。", + "evidence": [ + { + "description": "You are collecting garbage into garbage bags in the living room on the second floor.", + "timestamps": { + "start": "00:01:55", + "end": "00:02:10" + } + }, + { + "description": "You're coming downstairs with a garbage bag.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:15" + } + }, + { + "description": "You put the garbage bag by the yard exit.", + "timestamps": { + "start": "00:04:30", + "end": "00:04:36" + } + } + ], + "question_timestamps": "00:04:40", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You left it by the yard exit.", + "B": "After you collect garbage in the living room on the second floor, go downstairs with the garbage bag and put it next to the yard exit.", + "C": "You put it behind the yard exit.", + "D": "After collecting garbage upstairs, you go downstairs and set the bag by the front door.", + "E": "After collecting garbage upstairs, you carry the bag downstairs and place it inside the hallway near the exit." + }, + "answer_label": "B", + "answer_text": "After you collect garbage in the living room on the second floor, go downstairs with the garbage bag and put it next to the yard exit.", + "q_id": "egolife_A1_D7_008-mcq_fixed_7", + "correct_answer": "After you collect garbage in the living room on the second floor, go downstairs with the garbage bag and put it next to the yard exit.", + "original_options": { + "A": "You left it by the yard exit.", + "B": "After you collect garbage in the living room on the second floor, go downstairs with the garbage bag and put it next to the yard exit.", + "C": "You put it behind the yard exit.", + "D": "After collecting garbage upstairs, you go downstairs and set the bag by the front door.", + "E": "After collecting garbage upstairs, you carry the bag downstairs and place it inside the hallway near the exit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 15, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008-mcq_fixed_8", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my trajectory since I picked up my glasses?", + "answer": "You walk from the yard to the living room on the first floor.", + "question_chinese": "从我拿起眼镜，到现在我的运动轨迹是怎样的？", + "answer_chinese": "你从院子里走到了一楼客厅。", + "evidence": [ + { + "description": "You take your glasses from the yard to the living room on the first floor.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:50" + } + } + ], + "question_timestamps": "00:05:51", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Kitchen to living room.", + "B": "Living room to yard.", + "C": "You walk from the yard into the kitchen first, then continue toward the dining room instead.", + "D": "You walk from the yard to the living room on the first floor.", + "E": "You cross the yard and enter the hallway near the staircase, stopping before the living room." + }, + "answer_label": "D", + "answer_text": "You walk from the yard to the living room on the first floor.", + "q_id": "egolife_A1_D7_008-mcq_fixed_8", + "correct_answer": "You walk from the yard to the living room on the first floor.", + "original_options": { + "A": "Kitchen to living room.", + "B": "Living room to yard.", + "C": "You walk from the yard into the kitchen first, then continue toward the dining room instead.", + "D": "You walk from the yard to the living room on the first floor.", + "E": "You cross the yard and enter the hallway near the staircase, stopping before the living room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008-mcq_fixed_9", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my trajectory since I picked up my glasses?", + "answer": "You walk from the yard to the living room on the first floor and then into the bedroom nearest the living room table.", + "question_chinese": "从我拿起眼镜，到现在我的运动轨迹是怎样的？", + "answer_chinese": "你从院子里走到了一楼客厅，然后走进了离客厅桌子最近的卧室。", + "evidence": [ + { + "description": "You take your glasses from the yard to the living room on the first floor.", + "timestamps": { + "start": "00:05:38", + "end": "00:05:50" + } + }, + { + "description": "You walk from the living room into the bedroom nearest the desk with your glasses on.", + "timestamps": { + "start": "00:05:54", + "end": "00:05:59" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You go to the guest room.", + "B": "You go to the office.", + "C": "You walk from the yard into the kitchen, then enter the bedroom with the yellow lamp.", + "D": "You start in the yard, enter the living room, then head into the bedroom with green wallpaper instead.", + "E": "You walk from the yard to the living room on the first floor and then into the bedroom nearest the living room table." + }, + "answer_label": "E", + "answer_text": "You walk from the yard to the living room on the first floor and then into the bedroom nearest the living room table.", + "q_id": "egolife_A1_D7_008-mcq_fixed_9", + "correct_answer": "You walk from the yard to the living room on the first floor and then into the bedroom nearest the living room table.", + "original_options": { + "A": "You go to the guest room.", + "B": "You go to the office.", + "C": "You walk from the yard into the kitchen, then enter the bedroom with the yellow lamp.", + "D": "You start in the yard, enter the living room, then head into the bedroom with green wallpaper instead.", + "E": "You walk from the yard to the living room on the first floor and then into the bedroom nearest the living room table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008_1_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times in total have I gone downstairs?", + "answer": "In total, you went downstairs once, and the stairs are in front of you.", + "question_chinese": "到目前为止，我一共下楼过几次？", + "answer_chinese": "你一共下楼了一次，楼梯在你前面。", + "evidence": [ + { + "description": "You are going downstairs.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:15" + } + } + ], + "question_timestamps": "00:04:30", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You went downstairs zero times.", + "B": "You went downstairs twice.", + "C": "In total, you went downstairs once, and the stairs are in front of you.", + "D": "In total, you went downstairs three times so far, and the stairs are still in front of you.", + "E": "Altogether, you went downstairs four times up to now, with the staircase remaining directly ahead." + }, + "answer_label": "C", + "answer_text": "In total, you went downstairs once, and the stairs are in front of you.", + "q_id": "egolife_A1_D7_008_1_T1", + "correct_answer": "In total, you went downstairs once, and the stairs are in front of you.", + "original_options": { + "A": "You went downstairs zero times.", + "B": "You went downstairs twice.", + "C": "In total, you went downstairs once, and the stairs are in front of you.", + "D": "In total, you went downstairs three times so far, and the stairs are still in front of you.", + "E": "Altogether, you went downstairs four times up to now, with the staircase remaining directly ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008_1_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Up to now, how many times in total have I gone downstairs?", + "answer": "In total, you went downstairs twice, and the stairs are in front of you.", + "question_chinese": "到目前为止，我一共下楼过几次？", + "answer_chinese": "你一共下楼了两次，楼梯在你前面。", + "evidence": [ + { + "description": "You are going downstairs.", + "timestamps": { + "start": "00:04:10", + "end": "00:04:15" + } + }, + { + "description": "You are going downstairs with a chair.", + "timestamps": { + "start": "00:07:58", + "end": "00:08:05" + } + } + ], + "question_timestamps": "00:08:22", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You went downstairs once.", + "B": "You went downstairs three times.", + "C": "In total, you went downstairs zero times so far, even though the stairs are in front of you.", + "D": "In total, you went downstairs twice, and the stairs are in front of you.", + "E": "Altogether, you went downstairs four times up to now, and the stairs are still directly ahead." + }, + "answer_label": "D", + "answer_text": "In total, you went downstairs twice, and the stairs are in front of you.", + "q_id": "egolife_A1_D7_008_1_T2", + "correct_answer": "In total, you went downstairs twice, and the stairs are in front of you.", + "original_options": { + "A": "You went downstairs once.", + "B": "You went downstairs three times.", + "C": "In total, you went downstairs zero times so far, even though the stairs are in front of you.", + "D": "In total, you went downstairs twice, and the stairs are in front of you.", + "E": "Altogether, you went downstairs four times up to now, and the stairs are still directly ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 18, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008_4_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How many times did I take out the trash?", + "answer": "You threw out the trash once, and the trash is to your right.", + "question_chinese": "我丢了几次垃圾？", + "answer_chinese": "你丢了一次垃圾，垃圾在你右边。", + "evidence": [ + { + "description": "You threw away the rubbish once.", + "timestamps": { + "start": "00:00:37", + "end": "00:00:44" + } + } + ], + "question_timestamps": "00:00:49", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You threw it out twice.", + "B": "You threw out the trash once, and the trash is to your right.", + "C": "You didn't take it out.", + "D": "You threw out the trash three times, and the trash is positioned to your right.", + "E": "You threw out the trash four times, and the trash is positioned to your right." + }, + "answer_label": "B", + "answer_text": "You threw out the trash once, and the trash is to your right.", + "q_id": "egolife_A1_D7_008_4_T2", + "correct_answer": "You threw out the trash once, and the trash is to your right.", + "original_options": { + "A": "You threw it out twice.", + "B": "You threw out the trash once, and the trash is to your right.", + "C": "You didn't take it out.", + "D": "You threw out the trash three times, and the trash is positioned to your right.", + "E": "You threw out the trash four times, and the trash is positioned to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008_4_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "How many times did I take out the trash?", + "answer": "You threw out the trash twice, and the trash is to your right.", + "question_chinese": "我丢了几次垃圾？", + "answer_chinese": "你丢了两次垃圾，垃圾在你右边。", + "evidence": [ + { + "description": "You threw away the rubbish two.", + "timestamps": { + "start": "00:01:15", + "end": "00:01:24" + } + } + ], + "question_timestamps": "00:01:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You threw out the trash twice, and the trash is to your right.", + "B": "You threw it out once.", + "C": "You threw it out three times.", + "D": "You didn't throw out the trash at all, and the trash is still to your right.", + "E": "You threw out the trash four times, and the trash is still to your right." + }, + "answer_label": "A", + "answer_text": "You threw out the trash twice, and the trash is to your right.", + "q_id": "egolife_A1_D7_008_4_T1", + "correct_answer": "You threw out the trash twice, and the trash is to your right.", + "original_options": { + "A": "You threw out the trash twice, and the trash is to your right.", + "B": "You threw it out once.", + "C": "You threw it out three times.", + "D": "You didn't throw out the trash at all, and the trash is still to your right.", + "E": "You threw out the trash four times, and the trash is still to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008_3_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach the computer?", + "answer": "You can reach the computer; the computer is in front of you.", + "question_chinese": "我现在的位置能够得到电脑吗？", + "answer_chinese": "你能够到电脑，电脑在你前面。", + "evidence": [ + { + "description": "You are using the computer.", + "timestamps": { + "start": "00:07:21", + "end": "00:07:22" + } + } + ], + "question_timestamps": "00:07:27", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can't reach the computer.", + "B": "You can reach the computer; the computer is in front of you.", + "C": "You can reach it on your left.", + "D": "You can reach the computer, but it is behind you rather than in front of you.", + "E": "You can reach the computer, and it is to your right instead of directly in front." + }, + "answer_label": "B", + "answer_text": "You can reach the computer; the computer is in front of you.", + "q_id": "egolife_A1_D7_008_3_T1", + "correct_answer": "You can reach the computer; the computer is in front of you.", + "original_options": { + "A": "You can't reach the computer.", + "B": "You can reach the computer; the computer is in front of you.", + "C": "You can reach it on your left.", + "D": "You can reach the computer, but it is behind you rather than in front of you.", + "E": "You can reach the computer, and it is to your right instead of directly in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008_3_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach the computer?", + "answer": "You can't reach the computer; the computer is behind you.", + "question_chinese": "我现在的位置能够得到电脑吗？", + "answer_chinese": "你不能够到电脑，电脑在你后面。", + "evidence": [ + { + "description": "You are not using the computer.", + "timestamps": { + "start": "00:07:30", + "end": "00:07:34" + } + } + ], + "question_timestamps": "00:07:37", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Yes, you can reach it.", + "B": "You can't reach the computer; the computer is behind you.", + "C": "No, it’s in front of you.", + "D": "Yes, you can reach the computer because it is directly in front of you within arm’s reach.", + "E": "You can reach the computer from here; it is beside you rather than behind you." + }, + "answer_label": "B", + "answer_text": "You can't reach the computer; the computer is behind you.", + "q_id": "egolife_A1_D7_008_3_T2", + "correct_answer": "You can't reach the computer; the computer is behind you.", + "original_options": { + "A": "Yes, you can reach it.", + "B": "You can't reach the computer; the computer is behind you.", + "C": "No, it’s in front of you.", + "D": "Yes, you can reach the computer because it is directly in front of you within arm’s reach.", + "E": "You can reach the computer from here; it is beside you rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008_12", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get a computer at my current location?", + "answer": "You can't reach the computer.", + "question_chinese": "我现在的位置能够得到电脑吗？", + "answer_chinese": "你够不到电脑。", + "evidence": [ + { + "description": "There are two computers in the corner of the living room on the second floor.", + "timestamps": { + "start": "00:10:13", + "end": "00:10:17" + } + }, + { + "description": "You are in the center of the living room on the second floor.", + "timestamps": { + "start": "00:11:05", + "end": "00:11:06" + } + } + ], + "question_timestamps": "00:11:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "You can reach the computer.", + "B": "You can't reach the computer." + }, + "answer_label": "B", + "answer_text": "You can't reach the computer.", + "q_id": "egolife_A1_D7_008_12", + "correct_answer": "You can't reach the computer.", + "original_options": { + "A": "You can reach the computer.", + "B": "You can't reach the computer." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008_7", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the living room on the first floor?", + "answer": "There are two chairs in the living room on the first floor.", + "question_chinese": "现在，一楼客厅里面一共有几把椅子？", + "answer_chinese": "一楼客厅里一共有两把椅子。", + "evidence": [ + { + "description": "There are two chairs in the living room on the first floor.", + "timestamps": { + "start": "00:06:16", + "end": "00:06:18" + } + } + ], + "question_timestamps": "00:06:29", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two chairs in the living room on the first floor.", + "B": "There is one chair.", + "C": "There are three chairs.", + "D": "There are four chairs in the living room on the first floor.", + "E": "There are five chairs in the living room on the first floor, counted across the seating area." + }, + "answer_label": "A", + "answer_text": "There are two chairs in the living room on the first floor.", + "q_id": "egolife_A1_D7_008_7", + "correct_answer": "There are two chairs in the living room on the first floor.", + "original_options": { + "A": "There are two chairs in the living room on the first floor.", + "B": "There is one chair.", + "C": "There are three chairs.", + "D": "There are four chairs in the living room on the first floor.", + "E": "There are five chairs in the living room on the first floor, counted across the seating area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008_8", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the living room on the first floor?", + "answer": "There are two chairs in the living room on the first floor.", + "question_chinese": "现在，一楼客厅里面一共有几把椅子？", + "answer_chinese": "一楼客厅里一共有三把椅子。", + "evidence": [ + { + "description": "There are two chairs in the living room on the first floor.", + "timestamps": { + "start": "00:06:16", + "end": "00:06:18" + } + }, + { + "description": "You brought one chair from the living room on the second floor to the living room on the first floor.", + "timestamps": { + "start": "00:08:19", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:08:32", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One chair.", + "B": "Three chairs.", + "C": "There are four chairs in the living room on the first floor.", + "D": "There are two chairs in the living room on the first floor.", + "E": "There are five chairs in the living room on the first floor." + }, + "answer_label": "D", + "answer_text": "There are two chairs in the living room on the first floor.", + "q_id": "egolife_A1_D7_008_8", + "correct_answer": "There are two chairs in the living room on the first floor.", + "original_options": { + "A": "One chair.", + "B": "Three chairs.", + "C": "There are four chairs in the living room on the first floor.", + "D": "There are two chairs in the living room on the first floor.", + "E": "There are five chairs in the living room on the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008_2_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many chairs are there in the first-floor living room, and where are the chairs relative to me?", + "answer": "There are two chairs in the first-floor living room, and the chairs are in front of you.", + "question_chinese": "现在，一楼客厅里面一共有几把椅子，椅子在我哪个位置？", + "answer_chinese": "一楼客厅里一共有两把椅子，椅子在你前面。", + "evidence": [ + { + "description": "There are two chairs in the living room on the first floor.", + "timestamps": { + "start": "00:06:16", + "end": "00:06:18" + } + } + ], + "question_timestamps": "00:06:29", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One chair, behind you.", + "B": "Three chairs, to your left.", + "C": "There are four chairs in the first-floor living room, and they are behind you.", + "D": "There are two chairs in the first-floor living room, and the chairs are in front of you.", + "E": "There are three chairs in the first-floor living room, and the chairs are to your right." + }, + "answer_label": "D", + "answer_text": "There are two chairs in the first-floor living room, and the chairs are in front of you.", + "q_id": "egolife_A1_D7_008_2_T1", + "correct_answer": "There are two chairs in the first-floor living room, and the chairs are in front of you.", + "original_options": { + "A": "One chair, behind you.", + "B": "Three chairs, to your left.", + "C": "There are four chairs in the first-floor living room, and they are behind you.", + "D": "There are two chairs in the first-floor living room, and the chairs are in front of you.", + "E": "There are three chairs in the first-floor living room, and the chairs are to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_008", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_008_2_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many chairs are there in the first-floor living room, and where are the chairs relative to me?", + "answer": "There are three chairs in the first-floor living room, and the chairs are in front of you.", + "question_chinese": "现在，一楼客厅里面一共有几把椅子，椅子在我哪个位置？", + "answer_chinese": "一楼客厅里一共有三把椅子，椅子在你前面。", + "evidence": [ + { + "description": "There are two chairs in the living room on the first floor.", + "timestamps": { + "start": "00:06:16", + "end": "00:06:18" + } + }, + { + "description": "You brought one chair from the living room on the second floor to the living room on the first floor.", + "timestamps": { + "start": "00:08:19", + "end": "00:08:20" + } + } + ], + "question_timestamps": "00:08:32", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two chairs, in front of you.", + "B": "Four chairs, behind you.", + "C": "There is one chair in the first-floor living room, and the chair is in front of you.", + "D": "There are five chairs in the first-floor living room, and the chairs are to your left.", + "E": "There are three chairs in the first-floor living room, and the chairs are in front of you." + }, + "answer_label": "E", + "answer_text": "There are three chairs in the first-floor living room, and the chairs are in front of you.", + "q_id": "egolife_A1_D7_008_2_T2", + "correct_answer": "There are three chairs in the first-floor living room, and the chairs are in front of you.", + "original_options": { + "A": "Two chairs, in front of you.", + "B": "Four chairs, behind you.", + "C": "There is one chair in the first-floor living room, and the chair is in front of you.", + "D": "There are five chairs in the first-floor living room, and the chairs are to your left.", + "E": "There are three chairs in the first-floor living room, and the chairs are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 17, + 16 + ] + } +] diff --git a/QAs/egolife_A1_D7_009-mcq.json b/QAs/egolife_A1_D7_009-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f9eed02c0d0fc648dd346a8dc51a5b90e29dc080 --- /dev/null +++ b/QAs/egolife_A1_D7_009-mcq.json @@ -0,0 +1,1305 @@ +[ + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the living room on the first floor?", + "answer": "Three.", + "question_chinese": "一楼客厅一共有几把椅子？", + "answer_chinese": "三把", + "evidence": [ + { + "description": "There are three chairs in the living room on the first floor", + "timestamps": { + "start": "00:01:36", + "end": "00:01:40" + } + } + ], + "question_timestamps": "00:02:47", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "There are five chairs in the living room on the first floor.", + "D": "Three.", + "E": "There are six chairs in the living room on the first floor." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "egolife_A1_D7_009_7", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "There are five chairs in the living room on the first floor.", + "D": "Three.", + "E": "There are six chairs in the living room on the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the living room on the first floor?", + "answer": "Five.", + "question_chinese": "一楼客厅一共有几把椅子？", + "answer_chinese": "五把", + "evidence": [ + { + "description": "There are three chairs in the living room on the first floor", + "timestamps": { + "start": "00:01:36", + "end": "00:01:40" + } + }, + { + "description": "I moved a chair from the second floor to the living room on the first floor", + "timestamps": { + "start": "00:02:54", + "end": "00:03:17" + } + }, + { + "description": "I moved another chair from the second floor to the first floor", + "timestamps": { + "start": "00:04:25", + "end": "00:04:44" + } + } + ], + "question_timestamps": "00:06:00", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Five.", + "C": "Six.", + "D": "There are three chairs in the living room on the first floor.", + "E": "There are two chairs in the living room on the first floor." + }, + "answer_label": "B", + "answer_text": "Five.", + "q_id": "egolife_A1_D7_009_8", + "correct_answer": "Five.", + "original_options": { + "A": "Four.", + "B": "Five.", + "C": "Six.", + "D": "There are three chairs in the living room on the first floor.", + "E": "There are two chairs in the living room on the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_9", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me on the first floor, the table or the refrigerator?", + "answer": "Table.", + "question_chinese": "在一楼，桌子和冰箱哪个离我近？", + "answer_chinese": "桌子", + "evidence": [ + { + "description": "The table is in the middle of the living room", + "timestamps": { + "start": "00:01:34", + "end": "00:01:39" + } + }, + { + "description": "The refrigerator is beside the main door", + "timestamps": { + "start": "00:02:01", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:03:30", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Refrigerator.", + "B": "Table." + }, + "answer_label": "B", + "answer_text": "Table.", + "q_id": "egolife_A1_D7_009_9", + "correct_answer": "Table.", + "original_options": { + "A": "Refrigerator.", + "B": "Table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_10", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "On the first floor, which is closer to me, the table or the refrigerator?", + "answer": "Refrigerator?", + "question_chinese": "在一楼，桌子和冰箱哪个离我近？", + "answer_chinese": "冰箱", + "evidence": [ + { + "description": "The table is in the middle of the living room", + "timestamps": { + "start": "00:01:34", + "end": "00:01:39" + } + }, + { + "description": "The refrigerator is beside the main door", + "timestamps": { + "start": "00:02:01", + "end": "00:02:05" + } + } + ], + "question_timestamps": "00:07:11", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Refrigerator?", + "B": "Table?" + }, + "answer_label": "A", + "answer_text": "Refrigerator?", + "q_id": "egolife_A1_D7_009_10", + "correct_answer": "Refrigerator?", + "original_options": { + "A": "Refrigerator?", + "B": "Table?" + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were in the second-floor living room besides me the last time I was there?", + "answer": "Five.", + "question_chinese": "我最近一次到二楼客厅时，二楼客厅里除我之外有几个人？", + "answer_chinese": "五个", + "evidence": [ + { + "description": "The last time I reached the living room on the second floor, there were five people there besides me", + "timestamps": { + "start": "00:02:22", + "end": "00:02:28" + } + } + ], + "question_timestamps": "00:03:00", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "There were three other people in the second-floor living room besides you.", + "E": "There were two other people in the second-floor living room besides you." + }, + "answer_label": "A", + "answer_text": "Five.", + "q_id": "egolife_A1_D7_009_3", + "correct_answer": "Five.", + "original_options": { + "A": "Five.", + "B": "Four.", + "C": "Six.", + "D": "There were three other people in the second-floor living room besides you.", + "E": "There were two other people in the second-floor living room besides you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were in the living room on the second floor when I was last there, besides me?", + "answer": "Zero.", + "question_chinese": "我最近一次到二楼客厅时，二楼客厅里除我之外有几个人？", + "answer_chinese": "零个", + "evidence": [ + { + "description": "The last time I entered the living room on the second floor, there was no one else but me", + "timestamps": { + "start": "00:07:13", + "end": "00:07:22" + } + } + ], + "question_timestamps": "00:07:43", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three people were there besides you in the living room when you last visited the second floor.", + "D": "Zero.", + "E": "Four people were in the second-floor living room with you the last time you were there." + }, + "answer_label": "D", + "answer_text": "Zero.", + "q_id": "egolife_A1_D7_009_4", + "correct_answer": "Zero.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three people were there besides you in the living room when you last visited the second floor.", + "D": "Zero.", + "E": "Four people were in the second-floor living room with you the last time you were there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_1", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the washing machine?", + "answer": "Turn right and go straight to the staircase, then turn left and walk about three steps.", + "question_chinese": "我现在想要使用洗衣机，应该怎么走？", + "answer_chinese": "右转并直行进入楼梯口，然后左转走三步左右", + "evidence": [ + { + "description": "The washing machine is between the staircase on the first floor and the main entrance", + "timestamps": { + "start": "00:01:27", + "end": "00:01:34" + } + } + ], + "question_timestamps": "00:02:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left, then walk straight.", + "B": "Turn right, then turn right again.", + "C": "Turn right and go straight to the staircase, then turn left and walk about three steps.", + "D": "Turn right and walk to the staircase, then turn right and continue about three steps forward.", + "E": "Go straight to the staircase first, then turn left and walk about five steps to reach it." + }, + "answer_label": "C", + "answer_text": "Turn right and go straight to the staircase, then turn left and walk about three steps.", + "q_id": "egolife_A1_D7_009_1", + "correct_answer": "Turn right and go straight to the staircase, then turn left and walk about three steps.", + "original_options": { + "A": "Turn left, then walk straight.", + "B": "Turn right, then turn right again.", + "C": "Turn right and go straight to the staircase, then turn left and walk about three steps.", + "D": "Turn right and walk to the staircase, then turn right and continue about three steps forward.", + "E": "Go straight to the staircase first, then turn left and walk about five steps to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_2", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the washing machine?", + "answer": "Turn left into the stairs, then turn right and go downstairs. After reaching the first floor, walk straight for about four steps.", + "question_chinese": "我现在想要使用洗衣机，应该怎么走？", + "answer_chinese": "左转进入楼梯，然后右转下楼，到一楼后直走四步左右", + "evidence": [ + { + "description": "The washing machine is between the staircase on the first floor and the main entrance", + "timestamps": { + "start": "00:01:27", + "end": "00:01:34" + } + }, + { + "description": "To get to the living room on the second floor from the staircase on the first floor, you need to go upstairs and then turn left", + "timestamps": { + "start": "00:02:12", + "end": "00:02:28" + } + } + ], + "question_timestamps": "00:07:36", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right into the stairs.", + "B": "Turn left into the stairs, then turn right and go downstairs. After reaching the first floor, walk straight for about four steps.", + "C": "Go upstairs instead of downstairs.", + "D": "Turn left into the stairs, then turn left and go downstairs; on the first floor, walk straight about four steps.", + "E": "Turn left into the stairs, turn right to go downstairs, then walk straight for about six steps on the first floor." + }, + "answer_label": "B", + "answer_text": "Turn left into the stairs, then turn right and go downstairs. After reaching the first floor, walk straight for about four steps.", + "q_id": "egolife_A1_D7_009_2", + "correct_answer": "Turn left into the stairs, then turn right and go downstairs. After reaching the first floor, walk straight for about four steps.", + "original_options": { + "A": "Turn right into the stairs.", + "B": "Turn left into the stairs, then turn right and go downstairs. After reaching the first floor, walk straight for about four steps.", + "C": "Go upstairs instead of downstairs.", + "D": "Turn left into the stairs, then turn left and go downstairs; on the first floor, walk straight about four steps.", + "E": "Turn left into the stairs, turn right to go downstairs, then walk straight for about six steps on the first floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 21 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_13", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the microwave the fastest?", + "answer": "Turn back and take two steps.", + "question_chinese": "我现在想要用微波炉，最快应该怎么走？", + "answer_chinese": "向后转，并走两步", + "evidence": [ + { + "description": "There is a microwave oven on the windowsill of the first floor", + "timestamps": { + "start": "00:02:11", + "end": "00:02:14" + } + }, + { + "description": "There is a microwave oven on the windowsill on the second floor.", + "timestamps": { + "start": "00:04:17", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:05:35", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn back and take three steps.", + "B": "Turn back and take two steps.", + "C": "Turn left and take two steps.", + "D": "Turn right first, then take two steps forward to reach the microwave as quickly as possible.", + "E": "Take two steps backward without turning around, keeping your body facing the same direction the whole time." + }, + "answer_label": "B", + "answer_text": "Turn back and take two steps.", + "q_id": "egolife_A1_D7_009_13", + "correct_answer": "Turn back and take two steps.", + "original_options": { + "A": "Turn back and take three steps.", + "B": "Turn back and take two steps.", + "C": "Turn left and take two steps.", + "D": "Turn right first, then take two steps forward to reach the microwave as quickly as possible.", + "E": "Take two steps backward without turning around, keeping your body facing the same direction the whole time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_14", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the microwave the fastest?", + "answer": "Turn right and go straight to the position slightly to the right of the windowsill.", + "question_chinese": "我现在想要用微波炉，最快应该怎么走？", + "answer_chinese": "向右转，并直行到窗台偏右的位置", + "evidence": [ + { + "description": "There is a microwave oven on the windowsill of the first floor", + "timestamps": { + "start": "00:02:11", + "end": "00:02:14" + } + }, + { + "description": "There is a microwave oven on the windowsill on the second floor.", + "timestamps": { + "start": "00:04:17", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:09:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left and go straight.", + "B": "Go straight to the windowsill.", + "C": "Turn right and go straight to a spot slightly to the left of the windowsill instead of the right side.", + "D": "Turn right and go straight to the position slightly to the right of the windowsill.", + "E": "Go straight first, then turn right, stopping at a position slightly behind the windowsill rather than beside it." + }, + "answer_label": "D", + "answer_text": "Turn right and go straight to the position slightly to the right of the windowsill.", + "q_id": "egolife_A1_D7_009_14", + "correct_answer": "Turn right and go straight to the position slightly to the right of the windowsill.", + "original_options": { + "A": "Turn left and go straight.", + "B": "Go straight to the windowsill.", + "C": "Turn right and go straight to a spot slightly to the left of the windowsill instead of the right side.", + "D": "Turn right and go straight to the position slightly to the right of the windowsill.", + "E": "Go straight first, then turn right, stopping at a position slightly behind the windowsill rather than beside it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 20, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_11", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the flower doll now?", + "answer": "No.", + "question_chinese": "我现在能拿到花朵玩偶吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The flower doll is on the sofa in the living room on the second floor", + "timestamps": { + "start": "00:07:40", + "end": "00:07:50" + } + } + ], + "question_timestamps": "00:08:57", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "egolife_A1_D7_009_11", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_12", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the flower doll now?", + "answer": "Yes.", + "question_chinese": "我现在能拿到花朵玩偶吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The flower doll is on the sofa in the living room on the second floor", + "timestamps": { + "start": "00:07:40", + "end": "00:07:50" + } + } + ], + "question_timestamps": "00:09:35", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "egolife_A1_D7_009_12", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_3_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Which side of me is the projector screen on?", + "answer": "The projector screen is to your left.", + "question_chinese": "投影屏幕在我的哪边？", + "answer_chinese": "投影屏幕在你的左边", + "evidence": [ + { + "description": "The projection screen is on your left.", + "timestamps": { + "start": "00:05:00", + "end": "00:05:05" + } + } + ], + "question_timestamps": "00:05:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The projector screen is to your right.", + "B": "The projector screen is in front of you.", + "C": "The projector screen is to your left.", + "D": "The projector screen is behind you, so you would need to turn around to face it directly.", + "E": "The projector screen is above you, positioned overhead rather than on either side of you." + }, + "answer_label": "C", + "answer_text": "The projector screen is to your left.", + "q_id": "egolife_A1_D7_009_3_T1", + "correct_answer": "The projector screen is to your left.", + "original_options": { + "A": "The projector screen is to your right.", + "B": "The projector screen is in front of you.", + "C": "The projector screen is to your left.", + "D": "The projector screen is behind you, so you would need to turn around to face it directly.", + "E": "The projector screen is above you, positioned overhead rather than on either side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 17, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_3_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Which side of me is the projector screen on?", + "answer": "The projector screen is behind you.", + "question_chinese": "投影屏幕在我的哪边？", + "answer_chinese": "投影屏幕在你的后面", + "evidence": [ + { + "description": "The projection screen is behind you.", + "timestamps": { + "start": "00:05:05", + "end": "00:05:10" + } + } + ], + "question_timestamps": "00:05:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The projector screen is behind you.", + "B": "The projector screen is in front of you.", + "C": "The projector screen is to your left.", + "D": "The projector screen is to your right as you face forward, not behind your position.", + "E": "The projector screen is above you, hanging overhead rather than positioned behind you." + }, + "answer_label": "A", + "answer_text": "The projector screen is behind you.", + "q_id": "egolife_A1_D7_009_3_T2", + "correct_answer": "The projector screen is behind you.", + "original_options": { + "A": "The projector screen is behind you.", + "B": "The projector screen is in front of you.", + "C": "The projector screen is to your left.", + "D": "The projector screen is to your right as you face forward, not behind your position.", + "E": "The projector screen is above you, hanging overhead rather than positioned behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_6_T1", + "qid_num": 15, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the table or the air conditioner?", + "answer": "The table is closer to you; it is behind you.", + "question_chinese": "桌子和空调哪个离我近？", + "answer_chinese": "桌子离你更近，它在你后面", + "evidence": [ + { + "description": "The table is closer to you. It's behind you.", + "timestamps": { + "start": "00:01:38", + "end": "00:01:41" + } + } + ], + "question_timestamps": "00:01:43", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The table is closer to you; it is behind you.", + "B": "The air conditioner is closer to you; it is directly above you." + }, + "answer_label": "A", + "answer_text": "The table is closer to you; it is behind you.", + "q_id": "egolife_A1_D7_009_6_T1", + "correct_answer": "The table is closer to you; it is behind you.", + "original_options": { + "A": "The table is closer to you; it is behind you.", + "B": "The air conditioner is closer to you; it is directly above you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_6_T2", + "qid_num": 16, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the table or the air conditioner?", + "answer": "The air conditioner is closer to you; it is directly above you.", + "question_chinese": "桌子和空调哪个离我近？", + "answer_chinese": "空调离你近，在你的正上方", + "evidence": [ + { + "description": "The air conditioner is close to you.", + "timestamps": { + "start": "00:04:55", + "end": "00:04:58" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The air conditioner is closer to you; it is directly above you.", + "B": "The table is closer to you; it is behind you." + }, + "answer_label": "A", + "answer_text": "The air conditioner is closer to you; it is directly above you.", + "q_id": "egolife_A1_D7_009_6_T2", + "correct_answer": "The air conditioner is closer to you; it is directly above you.", + "original_options": { + "A": "The air conditioner is closer to you; it is directly above you.", + "B": "The table is closer to you; it is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_2_T2", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the sofa relative to me?", + "answer": "The sofa is to your left.", + "question_chinese": "沙发在我的哪里？", + "answer_chinese": "沙发在你的左边", + "evidence": [ + { + "description": "The sofa is on your left.", + "timestamps": { + "start": "00:01:02", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The sofa is in front of you.", + "B": "The sofa is to your left.", + "C": "The sofa is behind you.", + "D": "The sofa is to your right rather than positioned on your left side.", + "E": "The sofa is above you, not sitting on the floor to your left." + }, + "answer_label": "B", + "answer_text": "The sofa is to your left.", + "q_id": "egolife_A1_D7_009_2_T2", + "correct_answer": "The sofa is to your left.", + "original_options": { + "A": "The sofa is in front of you.", + "B": "The sofa is to your left.", + "C": "The sofa is behind you.", + "D": "The sofa is to your right rather than positioned on your left side.", + "E": "The sofa is above you, not sitting on the floor to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_2_T1", + "qid_num": 18, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the sofa relative to me?", + "answer": "The sofa is in front of you.", + "question_chinese": "沙发在我的哪里？", + "answer_chinese": "沙发在你的前面", + "evidence": [ + { + "description": "The sofa is in front of you.", + "timestamps": { + "start": "00:02:31", + "end": "00:02:34" + } + } + ], + "question_timestamps": "00:02:36", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The sofa is behind you.", + "B": "The sofa is to your left.", + "C": "The sofa is to your right instead of being positioned straight ahead of you.", + "D": "The sofa is above you, not located at eye level in front of you.", + "E": "The sofa is in front of you." + }, + "answer_label": "E", + "answer_text": "The sofa is in front of you.", + "q_id": "egolife_A1_D7_009_2_T1", + "correct_answer": "The sofa is in front of you.", + "original_options": { + "A": "The sofa is behind you.", + "B": "The sofa is to your left.", + "C": "The sofa is to your right instead of being positioned straight ahead of you.", + "D": "The sofa is above you, not located at eye level in front of you.", + "E": "The sofa is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 14 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_10_T2", + "qid_num": 19, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the windowsill relative to me?", + "answer": "The windowsill is to your right.", + "question_chinese": "窗台在我的哪里？", + "answer_chinese": "窗台在你的右边", + "evidence": [ + { + "description": "The windowsill was on my right", + "timestamps": { + "start": "00:03:10", + "end": "00:03:16" + } + } + ], + "question_timestamps": "00:03:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The windowsill is to your left.", + "B": "The windowsill is to your right." + }, + "answer_label": "B", + "answer_text": "The windowsill is to your right.", + "q_id": "egolife_A1_D7_009_10_T2", + "correct_answer": "The windowsill is to your right.", + "original_options": { + "A": "The windowsill is to your left.", + "B": "The windowsill is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_10_T1", + "qid_num": 20, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the windowsill relative to me?", + "answer": "The windowsill is in front of you.", + "question_chinese": "窗台在我的哪里？", + "answer_chinese": "窗台在你的前面", + "evidence": [ + { + "description": "The windowsill is in front of you.", + "timestamps": { + "start": "00:03:24", + "end": "00:03:26" + } + } + ], + "question_timestamps": "00:03:28", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The windowsill is in front of you.", + "B": "The windowsill is to your left." + }, + "answer_label": "A", + "answer_text": "The windowsill is in front of you.", + "q_id": "egolife_A1_D7_009_10_T1", + "correct_answer": "The windowsill is in front of you.", + "original_options": { + "A": "The windowsill is in front of you.", + "B": "The windowsill is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_5_T2", + "qid_num": 21, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the doorway relative to me?", + "answer": "The doorway is to your left.", + "question_chinese": "门口在我的哪里？", + "answer_chinese": "门口在你的左边", + "evidence": [ + { + "description": "The door is on your left.", + "timestamps": { + "start": "00:01:29", + "end": "00:01:32" + } + } + ], + "question_timestamps": "00:01:34", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The doorway is to your right.", + "B": "The doorway is to your left." + }, + "answer_label": "B", + "answer_text": "The doorway is to your left.", + "q_id": "egolife_A1_D7_009_5_T2", + "correct_answer": "The doorway is to your left.", + "original_options": { + "A": "The doorway is to your right.", + "B": "The doorway is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_5_T1", + "qid_num": 22, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the doorway relative to me?", + "answer": "The doorway is in front of you.", + "question_chinese": "门口在我的哪里？", + "answer_chinese": "门口在你的前面", + "evidence": [ + { + "description": "The door is in front of me.", + "timestamps": { + "start": "00:01:35", + "end": "00:01:38" + } + } + ], + "question_timestamps": "00:01:40", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The doorway is in front of you.", + "B": "The doorway is behind you." + }, + "answer_label": "A", + "answer_text": "The doorway is in front of you.", + "q_id": "egolife_A1_D7_009_5_T1", + "correct_answer": "The doorway is in front of you.", + "original_options": { + "A": "The doorway is in front of you.", + "B": "The doorway is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_11_T2", + "qid_num": 23, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is the dining table in front of me?", + "answer": "It is in front of you.", + "question_chinese": "餐桌在我的前面吗？", + "answer_chinese": "它在你的前面", + "evidence": [ + { + "description": "The table is in front of you.", + "timestamps": { + "start": "00:03:37", + "end": "00:03:40" + } + } + ], + "question_timestamps": "00:03:43", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "No, it’s behind you.", + "B": "No, it’s to your left.", + "C": "No, it isn’t in front of you; it’s behind you when you turn around.", + "D": "It is in front of you.", + "E": "No, it isn’t in front of you; it’s off to your right from where you’re standing." + }, + "answer_label": "D", + "answer_text": "It is in front of you.", + "q_id": "egolife_A1_D7_009_11_T2", + "correct_answer": "It is in front of you.", + "original_options": { + "A": "No, it’s behind you.", + "B": "No, it’s to your left.", + "C": "No, it isn’t in front of you; it’s behind you when you turn around.", + "D": "It is in front of you.", + "E": "No, it isn’t in front of you; it’s off to your right from where you’re standing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 19 + ] + }, + { + "video_uid": "egolife_A1_D7_009", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_009_11_T1", + "qid_num": 24, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Is the dining table in front of me?", + "answer": "No, it isn’t there—it’s behind you.", + "question_chinese": "餐桌在我的前面吗？", + "answer_chinese": "不在，它在你的后面", + "evidence": [ + { + "description": "No, the table is behind you.", + "timestamps": { + "start": "00:04:50", + "end": "00:04:54" + } + } + ], + "question_timestamps": "00:04:57", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Yes, it’s in front.", + "B": "No, it’s to your left.", + "C": "No, it isn’t there—it’s behind you.", + "D": "No, it isn’t behind you; it’s actually in front of you.", + "E": "No, it isn’t behind you; it’s positioned to your right instead." + }, + "answer_label": "C", + "answer_text": "No, it isn’t there—it’s behind you.", + "q_id": "egolife_A1_D7_009_11_T1", + "correct_answer": "No, it isn’t there—it’s behind you.", + "original_options": { + "A": "Yes, it’s in front.", + "B": "No, it’s to your left.", + "C": "No, it isn’t there—it’s behind you.", + "D": "No, it isn’t behind you; it’s actually in front of you.", + "E": "No, it isn’t behind you; it’s positioned to your right instead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 13 + ] + } +] diff --git a/QAs/egolife_A1_D7_010-mcq.json b/QAs/egolife_A1_D7_010-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..0c69abb7bbd53811c4621ef73bffe1183003de39 --- /dev/null +++ b/QAs/egolife_A1_D7_010-mcq.json @@ -0,0 +1,613 @@ +[ + { + "video_uid": "egolife_A1_D7_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_010_7", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the microwave the fastest?", + "answer": "Turn backward and walk straight for about two steps.", + "question_chinese": "我怎么走能最快的使用到微波炉？", + "answer_chinese": "向后转，并直行两步左右", + "evidence": [ + { + "description": "There is a microwave oven on the windowsill on the second floor", + "timestamps": { + "start": "00:00:20", + "end": "00:00:25" + } + }, + { + "description": "There is a microwave oven on the windowsill of the first floor", + "timestamps": { + "start": "00:05:20", + "end": "00:05:24" + } + } + ], + "question_timestamps": "00:06:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn left and walk two steps.", + "B": "Turn right and walk two steps.", + "C": "Turn backward and walk straight for about one step to reach it sooner.", + "D": "Turn backward and walk straight for about three steps before you get to the microwave.", + "E": "Turn backward and walk straight for about two steps." + }, + "answer_label": "E", + "answer_text": "Turn backward and walk straight for about two steps.", + "q_id": "egolife_A1_D7_010_7", + "correct_answer": "Turn backward and walk straight for about two steps.", + "original_options": { + "A": "Turn left and walk two steps.", + "B": "Turn right and walk two steps.", + "C": "Turn backward and walk straight for about one step to reach it sooner.", + "D": "Turn backward and walk straight for about three steps before you get to the microwave.", + "E": "Turn backward and walk straight for about two steps." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_010_8", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How can I get to the microwave the fastest?", + "answer": "Turn left and go straight to the right side of the windowsill.", + "question_chinese": "我怎么走能最快的使用到微波炉？", + "answer_chinese": "左转并直行到窗台右侧", + "evidence": [ + { + "description": "There is a microwave oven on the windowsill on the second floor", + "timestamps": { + "start": "00:00:20", + "end": "00:00:25" + } + }, + { + "description": "There is a microwave oven on the windowsill of the first floor", + "timestamps": { + "start": "00:05:20", + "end": "00:05:24" + } + } + ], + "question_timestamps": "00:10:52", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right to the windowsill.", + "B": "Turn left and go straight to the right side of the windowsill.", + "C": "Go to the left side.", + "D": "Turn left and go straight to the center of the windowsill instead of the right side.", + "E": "Turn right and go straight to the back of the windowsill rather than the right side." + }, + "answer_label": "B", + "answer_text": "Turn left and go straight to the right side of the windowsill.", + "q_id": "egolife_A1_D7_010_8", + "correct_answer": "Turn left and go straight to the right side of the windowsill.", + "original_options": { + "A": "Turn right to the windowsill.", + "B": "Turn left and go straight to the right side of the windowsill.", + "C": "Go to the left side.", + "D": "Turn left and go straight to the center of the windowsill instead of the right side.", + "E": "Turn right and go straight to the back of the windowsill rather than the right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "egolife_A1_D7_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_010_3", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I move after I last entered the living room on the first floor?", + "answer": "The last time I entered the living room, I went around the table and into the bedroom.", + "question_chinese": "我最近一次进入一楼客厅后是怎样移动的？", + "answer_chinese": "我最近一次进入客厅后，绕过桌子进入了卧室", + "evidence": [ + { + "description": "The last time I entered the living room, I went around the table and into the bedroom", + "timestamps": { + "start": "00:03:23", + "end": "00:03:33" + } + } + ], + "question_timestamps": "00:03:38", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The last time I entered the living room, I went around the table and into the bedroom.", + "B": "I went straight into the hallway.", + "C": "I went into the kitchen.", + "D": "The last time I entered the living room, I went around the couch and then into the kitchen.", + "E": "After entering the living room, I walked straight past the table and into the hallway." + }, + "answer_label": "A", + "answer_text": "The last time I entered the living room, I went around the table and into the bedroom.", + "q_id": "egolife_A1_D7_010_3", + "correct_answer": "The last time I entered the living room, I went around the table and into the bedroom.", + "original_options": { + "A": "The last time I entered the living room, I went around the table and into the bedroom.", + "B": "I went straight into the hallway.", + "C": "I went into the kitchen.", + "D": "The last time I entered the living room, I went around the couch and then into the kitchen.", + "E": "After entering the living room, I walked straight past the table and into the hallway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 18, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_010_4", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How did I move after I last entered the living room on the first floor?", + "answer": "I left the bedroom and went to the living room, then walked from one side of the table to the other.", + "question_chinese": "我最近一次进入一楼客厅后是怎样移动的？", + "answer_chinese": "我离开卧室来到客厅，然后从桌子的一边走到另一边", + "evidence": [ + { + "description": "I left the bedroom and went to the living room, then walked from one side of the table to the other", + "timestamps": { + "start": "00:04:48", + "end": "00:05:08" + } + } + ], + "question_timestamps": "00:05:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "I went from living room to bedroom.", + "B": "I left the bedroom and went to the living room, then walked from one side of the table to the other.", + "C": "I headed straight to the kitchen.", + "D": "I left the living room and went to the bedroom, then walked around the couch to the door.", + "E": "I went to the living room from the bedroom, then walked around the table instead of across it." + }, + "answer_label": "B", + "answer_text": "I left the bedroom and went to the living room, then walked from one side of the table to the other.", + "q_id": "egolife_A1_D7_010_4", + "correct_answer": "I left the bedroom and went to the living room, then walked from one side of the table to the other.", + "original_options": { + "A": "I went from living room to bedroom.", + "B": "I left the bedroom and went to the living room, then walked from one side of the table to the other.", + "C": "I headed straight to the kitchen.", + "D": "I left the living room and went to the bedroom, then walked around the couch to the door.", + "E": "I went to the living room from the bedroom, then walked around the table instead of across it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 18, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_010_9", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were in the living room on the first floor, besides me, the last time I entered?", + "answer": "Two.", + "question_chinese": "我最近一次进入一楼客厅时，一楼客厅里除我之外一共有几个人？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "The last time I entered the living room, there were two people in it besides me", + "timestamps": { + "start": "00:03:22", + "end": "00:03:31" + } + } + ], + "question_timestamps": "00:03:45", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four people were in the living room besides you when you last entered.", + "D": "Five people were in the living room besides you the last time you walked in.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "egolife_A1_D7_010_9", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four people were in the living room besides you when you last entered.", + "D": "Five people were in the living room besides you the last time you walked in.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_010_10", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people were in the living room on the first floor, besides me, the last time I entered?", + "answer": "Four.", + "question_chinese": "我最近一次进入一楼客厅时，一楼客厅里除我之外一共有几个人？", + "answer_chinese": "四个", + "evidence": [ + { + "description": "The last time I entered the living room, there were four people in it besides me", + "timestamps": { + "start": "00:04:49", + "end": "00:05:00" + } + } + ], + "question_timestamps": "00:05:23", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Five.", + "C": "Two people were in the living room besides you when you last entered.", + "D": "Six people were in the living room besides you the last time you entered it.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "egolife_A1_D7_010_10", + "correct_answer": "Four.", + "original_options": { + "A": "Three.", + "B": "Five.", + "C": "Two people were in the living room besides you when you last entered.", + "D": "Six people were in the living room besides you the last time you entered it.", + "E": "Four." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "egolife_A1_D7_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_010_5", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How do I get to the yard now?", + "answer": "Turn back, go straight to the staircase, then turn right into the stairs and go downstairs. After reaching the first floor, go straight.", + "question_chinese": "我现在想要去院子里面，应该怎么走?", + "answer_chinese": "向后转，直行到楼梯口，右转进入楼梯然后下楼，到一楼后直走", + "evidence": [ + { + "description": "To get to the courtyard from the second floor staircase, you need to go down the stairs and then walk straight", + "timestamps": { + "start": "00:00:41", + "end": "00:01:08" + } + } + ], + "question_timestamps": "00:01:56", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn back and head to the entrance.", + "B": "Go straight to the stairs, then left.", + "C": "Turn back, go straight to the staircase, then turn right into the stairs and go downstairs. After reaching the first floor, go straight.", + "D": "Turn back, go straight to the staircase, turn left into the stairs, and go downstairs to the first floor.", + "E": "Turn back, go straight to the staircase, turn right into the hallway, and continue downstairs before going straight." + }, + "answer_label": "C", + "answer_text": "Turn back, go straight to the staircase, then turn right into the stairs and go downstairs. After reaching the first floor, go straight.", + "q_id": "egolife_A1_D7_010_5", + "correct_answer": "Turn back, go straight to the staircase, then turn right into the stairs and go downstairs. After reaching the first floor, go straight.", + "original_options": { + "A": "Turn back and head to the entrance.", + "B": "Go straight to the stairs, then left.", + "C": "Turn back, go straight to the staircase, then turn right into the stairs and go downstairs. After reaching the first floor, go straight.", + "D": "Turn back, go straight to the staircase, turn left into the stairs, and go downstairs to the first floor.", + "E": "Turn back, go straight to the staircase, turn right into the hallway, and continue downstairs before going straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 18 + ] + }, + { + "video_uid": "egolife_A1_D7_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_010_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "How should I get to the yard now?", + "answer": "Turn back to the living room, go around the table and go straight to the staircase, then turn left and go straight.", + "question_chinese": "我现在想要去院子里面，应该怎么走?", + "answer_chinese": "向后转来到客厅，绕过桌子后直行到达楼梯口，然后左转并直行", + "evidence": [ + { + "description": "Just step out of the staircase and you'll reach the courtyard", + "timestamps": { + "start": "00:03:07", + "end": "00:03:13" + } + }, + { + "description": "To get to the bedroom from the staircase, you need to first turn right into the living room, go around the table and go straight", + "timestamps": { + "start": "00:03:21", + "end": "00:03:32" + } + } + ], + "question_timestamps": "00:03:38", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn back and go around the couch.", + "B": "Go straight to the staircase, then right.", + "C": "Turn back to the living room, go around the table, reach the staircase, then turn right and go straight.", + "D": "Turn back to the living room, go around the table and go straight to the staircase, then turn left and go straight.", + "E": "Turn back to the living room, go around the table, go to the hallway instead of the staircase, then go straight." + }, + "answer_label": "D", + "answer_text": "Turn back to the living room, go around the table and go straight to the staircase, then turn left and go straight.", + "q_id": "egolife_A1_D7_010_6", + "correct_answer": "Turn back to the living room, go around the table and go straight to the staircase, then turn left and go straight.", + "original_options": { + "A": "Turn back and go around the couch.", + "B": "Go straight to the staircase, then right.", + "C": "Turn back to the living room, go around the table, reach the staircase, then turn right and go straight.", + "D": "Turn back to the living room, go around the table and go straight to the staircase, then turn left and go straight.", + "E": "Turn back to the living room, go around the table, go to the hallway instead of the staircase, then go straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 21 + ] + }, + { + "video_uid": "egolife_A1_D7_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_010_2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which side of me was the washing machine on the last time I passed by it?", + "answer": "Left.", + "question_chinese": "最近一次路过洗衣机时，洗衣机在我的哪一侧？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The last time I passed by the washing machine, it was on my left", + "timestamps": { + "start": "00:01:08", + "end": "00:01:11" + } + } + ], + "question_timestamps": "00:01:46", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right.", + "B": "In front of you.", + "C": "It was on your right side the last time you walked past it.", + "D": "Left.", + "E": "The washing machine was behind you as you passed by it last time." + }, + "answer_label": "D", + "answer_text": "Left.", + "q_id": "egolife_A1_D7_010_2", + "correct_answer": "Left.", + "original_options": { + "A": "Right.", + "B": "In front of you.", + "C": "It was on your right side the last time you walked past it.", + "D": "Left.", + "E": "The washing machine was behind you as you passed by it last time." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 4, + 13, + 13 + ] + }, + { + "video_uid": "egolife_A1_D7_010", + "video_length": "00:12:00", + "qid": "egolife_A1_D7_010_1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which side was the washing machine on the last time I passed by it?", + "answer": "Right.", + "question_chinese": "最近一次路过洗衣机时，洗衣机在我的哪一侧？", + "answer_chinese": "右侧", + "evidence": [ + { + "description": "The last time I passed by the washing machine, it was on my right", + "timestamps": { + "start": "00:02:42", + "end": "00:02:49" + } + } + ], + "question_timestamps": "00:03:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left.", + "B": "In front of you.", + "C": "It was on your left side the last time you passed by it.", + "D": "It was behind you when you went past it the last time.", + "E": "Right." + }, + "answer_label": "E", + "answer_text": "Right.", + "q_id": "egolife_A1_D7_010_1", + "correct_answer": "Right.", + "original_options": { + "A": "Left.", + "B": "In front of you.", + "C": "It was on your left side the last time you passed by it.", + "D": "It was behind you when you went past it the last time.", + "E": "Right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "question_quality": "good", + "question_quality_auto": { + "label": "good", + "score": 100, + "reasons": [] + }, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 4, + 13, + 12 + ] + } +] diff --git a/QAs/f7148e30-2ded-4d14-a8bb-f50813da26df-mcq.json b/QAs/f7148e30-2ded-4d14-a8bb-f50813da26df-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1727105c6b78f276353b7444e2ce6e65ae011961 --- /dev/null +++ b/QAs/f7148e30-2ded-4d14-a8bb-f50813da26df-mcq.json @@ -0,0 +1,969 @@ +[ + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_8", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many traffic lights have I passed in total?", + "answer": "Two.", + "question_chinese": "我一共经过了几个红绿灯？", + "answer_chinese": "2个。", + "evidence": [ + { + "description": "You are passing through the first one.", + "timestamps": { + "start": "00:07:33", + "end": "00:07:34" + } + }, + { + "description": "You are passing through the second one.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:13" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Zero.", + "B": "Four.", + "C": "You have passed three traffic lights in total so far today already.", + "D": "Two.", + "E": "In total, you have gone past five traffic lights during the entire drive." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_8", + "correct_answer": "Two.", + "original_options": { + "A": "Zero.", + "B": "Four.", + "C": "You have passed three traffic lights in total so far today already.", + "D": "Two.", + "E": "In total, you have gone past five traffic lights during the entire drive." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_9", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many traffic lights have I passed in total?", + "answer": "Three.", + "question_chinese": "我一共经过了几个红绿灯？", + "answer_chinese": "3个。", + "evidence": [ + { + "description": "You are passing through the first one.", + "timestamps": { + "start": "00:07:33", + "end": "00:07:34" + } + }, + { + "description": "You are passing through the second one.", + "timestamps": { + "start": "00:08:10", + "end": "00:08:13" + } + }, + { + "description": "You are passing through the third one.", + "timestamps": { + "start": "00:15:00", + "end": "00:15:03" + } + } + ], + "question_timestamps": "00:15:30", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "Five.", + "C": "Three.", + "D": "You have passed four traffic lights in total so far today already.", + "E": "Altogether, you have gone past six traffic lights during the entire drive." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_9", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Five.", + "C": "Three.", + "D": "You have passed four traffic lights in total so far today already.", + "E": "Altogether, you have gone past six traffic lights during the entire drive." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_6", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the trajectory of my driving movement?", + "answer": "You reverse to the right rear.", + "question_chinese": "我开车的移动轨迹是怎样的？", + "answer_chinese": "你往右后方倒车。", + "evidence": [ + { + "description": "You reverse to the right rear.", + "timestamps": { + "start": "00:04:20", + "end": "00:04:30" + } + } + ], + "question_timestamps": "00:04:32", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You reverse to the left rear.", + "B": "You drive forward to the right front.", + "C": "You reverse to the right rear.", + "D": "You reverse straight back without turning toward either rear corner at all.", + "E": "You move forward, arcing left, rather than reversing toward the right rear." + }, + "answer_label": "C", + "answer_text": "You reverse to the right rear.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_6", + "correct_answer": "You reverse to the right rear.", + "original_options": { + "A": "You reverse to the left rear.", + "B": "You drive forward to the right front.", + "C": "You reverse to the right rear.", + "D": "You reverse straight back without turning toward either rear corner at all.", + "E": "You move forward, arcing left, rather than reversing toward the right rear." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 12, + 12 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_7", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is the trajectory of my driving movement?", + "answer": "You reverse to the right rear first.", + "question_chinese": "我开车的移动轨迹是怎样的？", + "answer_chinese": "你先往右后方倒车，然后直行。", + "evidence": [ + { + "description": "You reverse to the right rear first.", + "timestamps": { + "start": "00:04:25", + "end": "00:04:30" + } + }, + { + "description": "Then go straight ahead.", + "timestamps": { + "start": "00:04:33", + "end": "00:04:38" + } + } + ], + "question_timestamps": "00:04:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You reverse straight back first.", + "B": "You reverse to the left rear first.", + "C": "You reverse to the right rear first.", + "D": "You back up straight at first, then angle toward the left rear as you continue reversing.", + "E": "You begin by reversing toward the right front, then straighten out while backing up afterward." + }, + "answer_label": "C", + "answer_text": "You reverse to the right rear first.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_7", + "correct_answer": "You reverse to the right rear first.", + "original_options": { + "A": "You reverse straight back first.", + "B": "You reverse to the left rear first.", + "C": "You reverse to the right rear first.", + "D": "You back up straight at first, then angle toward the left rear as you continue reversing.", + "E": "You begin by reversing toward the right front, then straighten out while backing up afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 16, + 15 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_0", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory like?", + "answer": "Turn around first, then walk straight, and then turn around again.", + "question_chinese": "我的移动轨迹是怎样的？", + "answer_chinese": "你先往后转，然后直走，然后又往后转。", + "evidence": [ + { + "description": "Turn around first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "Then walk straight.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + }, + { + "description": "Then turn around again.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn around first, then walk straight, and then turn around again.", + "B": "Walk straight, then turn around once.", + "C": "Turn right first, then walk straight.", + "D": "You turn around first, walk straight ahead, and then turn left instead of turning around again.", + "E": "You walk straight first, then turn around, and finally keep walking without making another turn." + }, + "answer_label": "A", + "answer_text": "Turn around first, then walk straight, and then turn around again.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_0", + "correct_answer": "Turn around first, then walk straight, and then turn around again.", + "original_options": { + "A": "Turn around first, then walk straight, and then turn around again.", + "B": "Walk straight, then turn around once.", + "C": "Turn right first, then walk straight.", + "D": "You turn around first, walk straight ahead, and then turn left instead of turning around again.", + "E": "You walk straight first, then turn around, and finally keep walking without making another turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory like?", + "answer": "Turn back first, then go straight, then turn back again, and then go straight once more.", + "question_chinese": "我的移动轨迹是怎样的？", + "answer_chinese": "你先往后转，然后直走，然后又往后转,然后再直走。", + "evidence": [ + { + "description": "Turn back first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "Then go straight", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + }, + { + "description": "Then turn back again.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:11" + } + }, + { + "description": "then go straight once more.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:23" + } + } + ], + "question_timestamps": "00:00:25", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight.", + "B": "Go straight, then turn back.", + "C": "You turn back first, go straight, then turn left next, and finally continue straight again.", + "D": "Turn back first, then go straight, then turn back again, and then go straight once more.", + "E": "You start by turning back, walk straight, then turn back again, and stop without going straight afterward." + }, + "answer_label": "D", + "answer_text": "Turn back first, then go straight, then turn back again, and then go straight once more.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_1", + "correct_answer": "Turn back first, then go straight, then turn back again, and then go straight once more.", + "original_options": { + "A": "Turn right, then go straight.", + "B": "Go straight, then turn back.", + "C": "You turn back first, go straight, then turn left next, and finally continue straight again.", + "D": "Turn back first, then go straight, then turn back again, and then go straight once more.", + "E": "You start by turning back, walk straight, then turn back again, and stop without going straight afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df_8_T1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from the doorway did I see the checkout counter?", + "answer": "The checkout counter is to the front-right of the doorway.", + "question_chinese": "我看见收银台在门口的哪个方向？", + "answer_chinese": "收银台在门口的右前方。", + "evidence": [ + { + "description": "The door is there.", + "timestamps": { + "start": "00:24:36", + "end": "00:24:37" + } + }, + { + "description": "The cashier is here.", + "timestamps": { + "start": "00:24:07", + "end": "00:24:08" + } + } + ], + "question_timestamps": "00:24:40", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to the front-left.", + "B": "It is to the back-right.", + "C": "The checkout counter is to the front-right of the doorway.", + "D": "The checkout counter is directly in front of the doorway, not angled to the right.", + "E": "From the doorway, you see the checkout counter behind you and slightly to the left." + }, + "answer_label": "C", + "answer_text": "The checkout counter is to the front-right of the doorway.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df_8_T1", + "correct_answer": "The checkout counter is to the front-right of the doorway.", + "original_options": { + "A": "It is to the front-left.", + "B": "It is to the back-right.", + "C": "The checkout counter is to the front-right of the doorway.", + "D": "The checkout counter is directly in front of the doorway, not angled to the right.", + "E": "From the doorway, you see the checkout counter behind you and slightly to the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df_8_T2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction from the doorway did I see the checkout counter?", + "answer": "The checkout counter is in front of the doorway.", + "question_chinese": "我看见收银台在门口的哪个方向？", + "answer_chinese": "收银台在门口的前面。", + "evidence": [ + { + "description": "The cashier is here.", + "timestamps": { + "start": "00:24:07", + "end": "00:24:08" + } + }, + { + "description": "The entrance is here.", + "timestamps": { + "start": "00:24:36", + "end": "00:24:37" + } + } + ], + "question_timestamps": "00:25:00", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind the doorway.", + "B": "The checkout counter is in front of the doorway.", + "C": "It is to the front-right.", + "D": "The checkout counter is off to the left side of the doorway rather than straight ahead.", + "E": "From the doorway, the checkout counter appears behind you, slightly to the right as you look around." + }, + "answer_label": "B", + "answer_text": "The checkout counter is in front of the doorway.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df_8_T2", + "correct_answer": "The checkout counter is in front of the doorway.", + "original_options": { + "A": "It is behind the doorway.", + "B": "The checkout counter is in front of the doorway.", + "C": "It is to the front-right.", + "D": "The checkout counter is off to the left side of the doorway rather than straight ahead.", + "E": "From the doorway, the checkout counter appears behind you, slightly to the right as you look around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 17 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_14", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the cashier at the entrance?", + "answer": "The cashier is in front of the door.", + "question_chinese": "收银台在门口的哪个方向？", + "answer_chinese": "收银台在门口的前面。", + "evidence": [ + { + "description": "The door is there.", + "timestamps": { + "start": "00:24:36", + "end": "00:24:37" + } + }, + { + "description": "The cashier is here.", + "timestamps": { + "start": "00:24:07", + "end": "00:24:08" + } + } + ], + "question_timestamps": "00:24:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The cashier is behind the door.", + "B": "The cashier is to the left of you.", + "C": "The cashier is in front of the door.", + "D": "The cashier is positioned to the right of the entrance, not directly in front of the door.", + "E": "You see the cashier behind the entrance area, so it is not in front of the doorway." + }, + "answer_label": "C", + "answer_text": "The cashier is in front of the door.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_14", + "correct_answer": "The cashier is in front of the door.", + "original_options": { + "A": "The cashier is behind the door.", + "B": "The cashier is to the left of you.", + "C": "The cashier is in front of the door.", + "D": "The cashier is positioned to the right of the entrance, not directly in front of the door.", + "E": "You see the cashier behind the entrance area, so it is not in front of the doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 17, + 17 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_15", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the cashier at the entrance?", + "answer": "The cashier is in front of the door.", + "question_chinese": "收银台在门口的哪个方向？", + "answer_chinese": "收银台在门口的前面。", + "evidence": [ + { + "description": "The cashier is here.", + "timestamps": { + "start": "00:24:07", + "end": "00:24:08" + } + }, + { + "description": "The entrance is here.", + "timestamps": { + "start": "00:24:36", + "end": "00:24:37" + } + } + ], + "question_timestamps": "00:25:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The cashier is to your left.", + "B": "The cashier is in front of the door.", + "C": "The cashier is beside the door.", + "D": "The cashier is behind the entrance rather than directly in front of the door as you enter.", + "E": "The cashier is on the right side of the entrance, so it is not straight ahead of the door." + }, + "answer_label": "B", + "answer_text": "The cashier is in front of the door.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_15", + "correct_answer": "The cashier is in front of the door.", + "original_options": { + "A": "The cashier is to your left.", + "B": "The cashier is in front of the door.", + "C": "The cashier is beside the door.", + "D": "The cashier is behind the entrance rather than directly in front of the door as you enter.", + "E": "The cashier is on the right side of the entrance, so it is not straight ahead of the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 19 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the sink in for me?", + "answer": "The sink is on your right.", + "question_chinese": "水槽在我的哪个方向？", + "answer_chinese": "水槽在你右方。", + "evidence": [ + { + "description": "The sink is on your right.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The sink is on your left.", + "B": "The sink is on your right.", + "C": "The sink is in front of you.", + "D": "The sink is behind you on the left side, not on your right as you face forward.", + "E": "The sink is straight ahead of you near the counter, rather than positioned to your right." + }, + "answer_label": "B", + "answer_text": "The sink is on your right.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_2", + "correct_answer": "The sink is on your right.", + "original_options": { + "A": "The sink is on your left.", + "B": "The sink is on your right.", + "C": "The sink is in front of you.", + "D": "The sink is behind you on the left side, not on your right as you face forward.", + "E": "The sink is straight ahead of you near the counter, rather than positioned to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 16 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_3", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the sink in for me?", + "answer": "The sink is behind you.", + "question_chinese": "水槽在我的哪个方向？", + "answer_chinese": "水槽在你后面。", + "evidence": [ + { + "description": "The sink is behind you.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "The sink is behind you.", + "timestamps": { + "start": "00:00:20", + "end": "00:00:21" + } + } + ], + "question_timestamps": "00:00:22", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The sink is in front of you.", + "B": "The sink is to your left.", + "C": "The sink is off to your right as you face forward, not directly behind you.", + "D": "The sink is straight ahead of you, positioned slightly to the left rather than behind.", + "E": "The sink is behind you." + }, + "answer_label": "E", + "answer_text": "The sink is behind you.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_3", + "correct_answer": "The sink is behind you.", + "original_options": { + "A": "The sink is in front of you.", + "B": "The sink is to your left.", + "C": "The sink is off to your right as you face forward, not directly behind you.", + "D": "The sink is straight ahead of you, positioned slightly to the left rather than behind.", + "E": "The sink is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 15 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_4", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the piano in for me?", + "answer": "The piano is in front of you.", + "question_chinese": "钢琴在我的哪个方向？", + "answer_chinese": "钢琴在你面前。", + "evidence": [ + { + "description": "The piano is in front of you.", + "timestamps": { + "start": "00:01:48", + "end": "00:01:49" + } + } + ], + "question_timestamps": "00:01:50", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The piano is behind you.", + "B": "The piano is in front of you.", + "C": "The piano is to your left.", + "D": "The piano is to your right side, not directly in front of where you're facing.", + "E": "The piano sits behind and slightly to the right, so you’d need to turn around to see it." + }, + "answer_label": "B", + "answer_text": "The piano is in front of you.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_4", + "correct_answer": "The piano is in front of you.", + "original_options": { + "A": "The piano is behind you.", + "B": "The piano is in front of you.", + "C": "The piano is to your left.", + "D": "The piano is to your right side, not directly in front of where you're facing.", + "E": "The piano sits behind and slightly to the right, so you’d need to turn around to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 19 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_5", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the piano in for me?", + "answer": "The piano is on your left.", + "question_chinese": "钢琴在我的哪个方向？", + "answer_chinese": "钢琴在你左边。", + "evidence": [ + { + "description": "The piano is there.", + "timestamps": { + "start": "00:01:45", + "end": "00:01:50" + } + }, + { + "description": "The piano is on your left.", + "timestamps": { + "start": "00:02:10", + "end": "00:02:12" + } + } + ], + "question_timestamps": "00:02:12", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The piano is on your left.", + "B": "The piano is on your right.", + "C": "The piano is behind you.", + "D": "The piano is directly in front of you, not positioned on your left side.", + "E": "The piano is in front-right of you, so it isn’t located on your left." + }, + "answer_label": "A", + "answer_text": "The piano is on your left.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_5", + "correct_answer": "The piano is on your left.", + "original_options": { + "A": "The piano is on your left.", + "B": "The piano is on your right.", + "C": "The piano is behind you.", + "D": "The piano is directly in front of you, not positioned on your left side.", + "E": "The piano is in front-right of you, so it isn’t located on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 15 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_12", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the entrance in for me?", + "answer": "The door is behind you.", + "question_chinese": "门口在我的哪个方向？", + "answer_chinese": "门口在你后面。", + "evidence": [ + { + "description": "The door is behind you.", + "timestamps": { + "start": "00:22:33", + "end": "00:22:34" + } + }, + { + "description": "The door is there.", + "timestamps": { + "start": "00:22:05", + "end": "00:22:06" + } + } + ], + "question_timestamps": "00:22:35", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The door is in front of you.", + "B": "The door is to your left.", + "C": "The door is to your right side as you face forward, not positioned behind you.", + "D": "The door is behind you.", + "E": "The door is straight ahead of you and slightly offset, so you don’t need to turn around." + }, + "answer_label": "D", + "answer_text": "The door is behind you.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_12", + "correct_answer": "The door is behind you.", + "original_options": { + "A": "The door is in front of you.", + "B": "The door is to your left.", + "C": "The door is to your right side as you face forward, not positioned behind you.", + "D": "The door is behind you.", + "E": "The door is straight ahead of you and slightly offset, so you don’t need to turn around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 18 + ] + }, + { + "video_uid": "f7148e30-2ded-4d14-a8bb-f50813da26df", + "video_length": "00:25:07", + "qid": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_13", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the entrance in for me?", + "answer": "The door is on your right.", + "question_chinese": "门口在我的哪个方向？", + "answer_chinese": "门口在你右边。", + "evidence": [ + { + "description": "The door is on your right.", + "timestamps": { + "start": "00:22:43", + "end": "00:22:44" + } + }, + { + "description": "The door is there.", + "timestamps": { + "start": "00:22:06", + "end": "00:22:08" + } + } + ], + "question_timestamps": "00:22:45", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The door is on your left.", + "B": "The door is behind you.", + "C": "The door is directly in front of you, not located over on your right side.", + "D": "The door is above you rather than to your right, so it isn’t on your right side.", + "E": "The door is on your right." + }, + "answer_label": "E", + "answer_text": "The door is on your right.", + "q_id": "f7148e30-2ded-4d14-a8bb-f50813da26df-mcq_fixed_13", + "correct_answer": "The door is on your right.", + "original_options": { + "A": "The door is on your left.", + "B": "The door is behind you.", + "C": "The door is directly in front of you, not located over on your right side.", + "D": "The door is above you rather than to your right, so it isn’t on your right side.", + "E": "The door is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 18 + ] + } +] diff --git a/QAs/fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq.json b/QAs/fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3643cd9e8d9ac7d0f7401db0008c804e9029b0fb --- /dev/null +++ b/QAs/fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq.json @@ -0,0 +1,609 @@ +[ + { + "video_uid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e", + "video_length": "00:34:20", + "qid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_4", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory after getting on the car?", + "answer": "You reverse back first, then reverse to the right rear.", + "question_chinese": "我上车后的移动轨迹是怎样的？", + "answer_chinese": "你先往后倒车，然后往右后方倒车。", + "evidence": [ + { + "description": "You reverse back first.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:40" + } + }, + { + "description": "Then reverse to the right rear.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:46" + } + } + ], + "question_timestamps": "00:00:47", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You reverse back, then to the left rear.", + "B": "You reverse back first, then reverse to the right rear.", + "C": "You reverse to the right rear first.", + "D": "You back up toward the right rear first, then straighten out by reversing directly backward.", + "E": "You reverse straight back at first, and then angle the car toward the left rear instead of right." + }, + "answer_label": "B", + "answer_text": "You reverse back first, then reverse to the right rear.", + "q_id": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_4", + "correct_answer": "You reverse back first, then reverse to the right rear.", + "original_options": { + "A": "You reverse back, then to the left rear.", + "B": "You reverse back first, then reverse to the right rear.", + "C": "You reverse to the right rear first.", + "D": "You back up toward the right rear first, then straighten out by reversing directly backward.", + "E": "You reverse straight back at first, and then angle the car toward the left rear instead of right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 15, + 18 + ] + }, + { + "video_uid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e", + "video_length": "00:34:20", + "qid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_3", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory after getting on the car?", + "answer": "You reverse backwards first, then to the right rear, and then straight ahead.", + "question_chinese": "我上车后的移动轨迹是怎样的？", + "answer_chinese": "你先往后倒车，然后往右后方倒车，然后直行。", + "evidence": [ + { + "description": "You reverse backwards first.", + "timestamps": { + "start": "00:00:35", + "end": "00:00:40" + } + }, + { + "description": "Then to the right rear.", + "timestamps": { + "start": "00:00:42", + "end": "00:00:46" + } + }, + { + "description": "Then straight ahead.", + "timestamps": { + "start": "00:00:48", + "end": "00:00:51" + } + } + ], + "question_timestamps": "00:01:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You reverse, then left rear, then ahead.", + "B": "You reverse, then right rear, then left.", + "C": "You back up first, then angle to the right rear, and finally turn sharply right instead of going straight.", + "D": "You reverse backwards first, then to the right rear, and then straight ahead.", + "E": "You reverse backward, then swing toward the right front, and then continue straight ahead afterward." + }, + "answer_label": "D", + "answer_text": "You reverse backwards first, then to the right rear, and then straight ahead.", + "q_id": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_3", + "correct_answer": "You reverse backwards first, then to the right rear, and then straight ahead.", + "original_options": { + "A": "You reverse, then left rear, then ahead.", + "B": "You reverse, then right rear, then left.", + "C": "You back up first, then angle to the right rear, and finally turn sharply right instead of going straight.", + "D": "You reverse backwards first, then to the right rear, and then straight ahead.", + "E": "You reverse backward, then swing toward the right front, and then continue straight ahead afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 15 + ] + }, + { + "video_uid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e", + "video_length": "00:34:20", + "qid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_1", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "You go straight ahead.", + "question_chinese": "我的移动轨迹是怎样的？", + "answer_chinese": "你直走。", + "evidence": [ + { + "description": "You go straight ahead.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You step backward.", + "B": "You turn left.", + "C": "You move forward briefly, then veer to the right instead of continuing straight ahead.", + "D": "You go forward for a moment, take a slight detour around an obstacle, and then continue on.", + "E": "You go straight ahead." + }, + "answer_label": "E", + "answer_text": "You go straight ahead.", + "q_id": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_1", + "correct_answer": "You go straight ahead.", + "original_options": { + "A": "You step backward.", + "B": "You turn left.", + "C": "You move forward briefly, then veer to the right instead of continuing straight ahead.", + "D": "You go forward for a moment, take a slight detour around an obstacle, and then continue on.", + "E": "You go straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 17 + ] + }, + { + "video_uid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e", + "video_length": "00:34:20", + "qid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_0", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "You go straight ahead first, then turn left and get on the car.", + "question_chinese": "我的移动轨迹是怎样的？", + "answer_chinese": "你先往直走，然后左转上车。", + "evidence": [ + { + "description": "You go straight ahead first.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "Then turn left and get on the car.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You go straight ahead first, then turn left and get on the car.", + "B": "Go straight, then turn right.", + "C": "Turn left, then go straight.", + "D": "You go straight ahead, then turn right and climb onto the car afterward.", + "E": "You walk straight first, turn left, and then keep walking past the car instead of getting on." + }, + "answer_label": "A", + "answer_text": "You go straight ahead first, then turn left and get on the car.", + "q_id": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_0", + "correct_answer": "You go straight ahead first, then turn left and get on the car.", + "original_options": { + "A": "You go straight ahead first, then turn left and get on the car.", + "B": "Go straight, then turn right.", + "C": "Turn left, then go straight.", + "D": "You go straight ahead, then turn right and climb onto the car afterward.", + "E": "You walk straight first, turn left, and then keep walking past the car instead of getting on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 17 + ] + }, + { + "video_uid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e", + "video_length": "00:34:20", + "qid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_8", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is garlic in for me?", + "answer": "Garlic is on your left.", + "question_chinese": "蒜在我的哪个方向？", + "answer_chinese": "蒜在你的左边。", + "evidence": [ + { + "description": "Garlic is here.", + "timestamps": { + "start": "00:04:48", + "end": "00:04:49" + } + }, + { + "description": "Garlic is on your left.", + "timestamps": { + "start": "00:04:57", + "end": "00:04:58" + } + } + ], + "question_timestamps": "00:05:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Garlic is on your right.", + "B": "Garlic is on your left.", + "C": "Garlic is in front of you.", + "D": "Garlic is behind you, so you would need to turn around to face it.", + "E": "Garlic is above you on a higher shelf, not positioned to either side." + }, + "answer_label": "B", + "answer_text": "Garlic is on your left.", + "q_id": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_8", + "correct_answer": "Garlic is on your left.", + "original_options": { + "A": "Garlic is on your right.", + "B": "Garlic is on your left.", + "C": "Garlic is in front of you.", + "D": "Garlic is behind you, so you would need to turn around to face it.", + "E": "Garlic is above you on a higher shelf, not positioned to either side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 13 + ] + }, + { + "video_uid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e", + "video_length": "00:34:20", + "qid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_9", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is garlic in for me?", + "answer": "Garlic is behind you.", + "question_chinese": "蒜在我的哪个方向？", + "answer_chinese": "蒜在你的后边。", + "evidence": [ + { + "description": "Garlic is behind you.", + "timestamps": { + "start": "00:05:08", + "end": "00:05:09" + } + }, + { + "description": "Garlic is here.", + "timestamps": { + "start": "00:04:40", + "end": "00:04:45" + } + } + ], + "question_timestamps": "00:05:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Garlic is behind you.", + "B": "Garlic is in front of you.", + "C": "Garlic is on your left.", + "D": "Garlic is on your right side, so it is beside you rather than behind you.", + "E": "Garlic is below you near the ground, not located behind your current position." + }, + "answer_label": "A", + "answer_text": "Garlic is behind you.", + "q_id": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_9", + "correct_answer": "Garlic is behind you.", + "original_options": { + "A": "Garlic is behind you.", + "B": "Garlic is in front of you.", + "C": "Garlic is on your left.", + "D": "Garlic is on your right side, so it is beside you rather than behind you.", + "E": "Garlic is below you near the ground, not located behind your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 13 + ] + }, + { + "video_uid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e", + "video_length": "00:34:20", + "qid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_7", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory after entering the supermarket?", + "answer": "You go left first, then turn left.", + "question_chinese": "进超市后我的移动轨迹是怎样的？", + "answer_chinese": "你先往左边走，然后左转。", + "evidence": [ + { + "description": "You go left first.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:50" + } + }, + { + "description": "Then turn left.", + "timestamps": { + "start": "00:04:01", + "end": "00:04:02" + } + } + ], + "question_timestamps": "00:04:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You go right, then turn right.", + "B": "You go straight, then turn left.", + "C": "You turn left first, then continue straight ahead instead of making another left turn.", + "D": "You go left initially, then turn right afterward rather than turning left again.", + "E": "You go left first, then turn left." + }, + "answer_label": "E", + "answer_text": "You go left first, then turn left.", + "q_id": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_7", + "correct_answer": "You go left first, then turn left.", + "original_options": { + "A": "You go right, then turn right.", + "B": "You go straight, then turn left.", + "C": "You turn left first, then continue straight ahead instead of making another left turn.", + "D": "You go left initially, then turn right afterward rather than turning left again.", + "E": "You go left first, then turn left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e", + "video_length": "00:34:20", + "qid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_6", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory after entering the supermarket?", + "answer": "You go left first, then turn left, then turn right and go straight.", + "question_chinese": "进超市后我的移动轨迹是怎样的？", + "answer_chinese": "你先往左边走，然后左转，然后右转直走。", + "evidence": [ + { + "description": "You go left first.", + "timestamps": { + "start": "00:03:35", + "end": "00:03:50" + } + }, + { + "description": "Then turn left.", + "timestamps": { + "start": "00:04:01", + "end": "00:04:02" + } + }, + { + "description": "Then turn right and go straight.", + "timestamps": { + "start": "00:04:18", + "end": "00:04:20" + } + } + ], + "question_timestamps": "00:04:25", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Left, right, then straight.", + "B": "Right, left, then straight.", + "C": "You go left first, then turn left, then turn left again and continue straight ahead.", + "D": "You go left first, then turn left, then turn right and go straight.", + "E": "You go left at first, turn right next, then turn left and go straight afterward." + }, + "answer_label": "D", + "answer_text": "You go left first, then turn left, then turn right and go straight.", + "q_id": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_6", + "correct_answer": "You go left first, then turn left, then turn right and go straight.", + "original_options": { + "A": "Left, right, then straight.", + "B": "Right, left, then straight.", + "C": "You go left first, then turn left, then turn left again and continue straight ahead.", + "D": "You go left first, then turn left, then turn right and go straight.", + "E": "You go left at first, turn right next, then turn left and go straight afterward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e", + "video_length": "00:34:20", + "qid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_12", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the banana in for me?", + "answer": "Bananas are in front of you.", + "question_chinese": "香蕉在我的哪个方向？", + "answer_chinese": "香蕉在你面前。", + "evidence": [ + { + "description": "Bananas are here.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:40" + } + }, + { + "description": "Bananas are in front of you.", + "timestamps": { + "start": "00:05:35", + "end": "00:05:40" + } + } + ], + "question_timestamps": "00:05:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Bananas are behind you.", + "B": "Bananas are in front of you.", + "C": "Bananas are to your right.", + "D": "Bananas are on your left side, so they are beside you rather than straight ahead.", + "E": "Bananas are below you near the floor, not positioned in front of your viewpoint." + }, + "answer_label": "B", + "answer_text": "Bananas are in front of you.", + "q_id": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_12", + "correct_answer": "Bananas are in front of you.", + "original_options": { + "A": "Bananas are behind you.", + "B": "Bananas are in front of you.", + "C": "Bananas are to your right.", + "D": "Bananas are on your left side, so they are beside you rather than straight ahead.", + "E": "Bananas are below you near the floor, not positioned in front of your viewpoint." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 14 + ] + }, + { + "video_uid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e", + "video_length": "00:34:20", + "qid": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_11", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the banana in for me?", + "answer": "The banana is on your right.", + "question_chinese": "香蕉在我的哪个方向？", + "answer_chinese": "香蕉在你的右方。", + "evidence": [ + { + "description": "The banana is on your right.", + "timestamps": { + "start": "00:06:07", + "end": "00:06:08" + } + }, + { + "description": "Bananas are here.", + "timestamps": { + "start": "00:05:32", + "end": "00:05:34" + } + } + ], + "question_timestamps": "00:06:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The banana is on your left.", + "B": "The banana is in front of you.", + "C": "The banana is on your right.", + "D": "The banana is behind you, so you would have to turn around to see it.", + "E": "The banana is above you on a higher shelf, not located to your right side." + }, + "answer_label": "C", + "answer_text": "The banana is on your right.", + "q_id": "fbc425c4-def6-49a7-8b88-5d5d00b5524e-mcq_fixed_11", + "correct_answer": "The banana is on your right.", + "original_options": { + "A": "The banana is on your left.", + "B": "The banana is in front of you.", + "C": "The banana is on your right.", + "D": "The banana is behind you, so you would have to turn around to see it.", + "E": "The banana is above you on a higher shelf, not located to your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 15 + ] + } +] diff --git a/QAs/fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq.json b/QAs/fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..03f53002f77ddd6db60e97db8f995fa6be0c8939 --- /dev/null +++ b/QAs/fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq.json @@ -0,0 +1,630 @@ +[ + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001", + "video_length": "00:20:04", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have I met on the way?", + "answer": "You haven't met anyone.", + "question_chinese": "在路上我遇到过几个人？", + "answer_chinese": "你没遇到人。", + "evidence": [ + { + "description": "You haven't met anyone.", + "timestamps": { + "start": "00:00:00", + "end": "00:02:00" + } + } + ], + "question_timestamps": "00:02:02", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You met one person.", + "B": "You haven't met anyone.", + "C": "You met two people.", + "D": "You met three people while walking, passing them along the way before reaching your destination.", + "E": "You met a few people on the way, encountering several individuals as you moved forward." + }, + "answer_label": "B", + "answer_text": "You haven't met anyone.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_1", + "correct_answer": "You haven't met anyone.", + "original_options": { + "A": "You met one person.", + "B": "You haven't met anyone.", + "C": "You met two people.", + "D": "You met three people while walking, passing them along the way before reaching your destination.", + "E": "You met a few people on the way, encountering several individuals as you moved forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001", + "video_length": "00:20:04", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_0", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many people have I met on the way?", + "answer": "Three.", + "question_chinese": "在路上我遇到过几个人？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "You haven't met anyone.", + "timestamps": { + "start": "00:00:00", + "end": "00:02:00" + } + }, + { + "description": "You have met someone.", + "timestamps": { + "start": "00:04:34", + "end": "00:04:35" + } + }, + { + "description": "You met two people.", + "timestamps": { + "start": "00:05:36", + "end": "00:05:39" + } + } + ], + "question_timestamps": "00:05:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "Four.", + "C": "You met two people while walking along your route to your destination.", + "D": "Three.", + "E": "You met four different people as you went on your way to the next place." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_0", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "You met two people while walking along your route to your destination.", + "D": "Three.", + "E": "You met four different people as you went on your way to the next place." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 15 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001", + "video_length": "00:20:04", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_3", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the trash can in for me?", + "answer": "The trash can is in front of you.", + "question_chinese": "垃圾桶在我的哪个方向？", + "answer_chinese": "垃圾桶在你前面。", + "evidence": [ + { + "description": "The trash can is here.", + "timestamps": { + "start": "00:07:48", + "end": "00:07:49" + } + }, + { + "description": "The trash can is in front of you.", + "timestamps": { + "start": "00:07:48", + "end": "00:07:49" + } + } + ], + "question_timestamps": "00:07:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The trash can is in front of you.", + "B": "Behind you.", + "C": "On your left.", + "D": "The trash can is slightly behind you, off to your right side.", + "E": "The trash can is to your left, a bit off from where you're facing." + }, + "answer_label": "A", + "answer_text": "The trash can is in front of you.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_3", + "correct_answer": "The trash can is in front of you.", + "original_options": { + "A": "The trash can is in front of you.", + "B": "Behind you.", + "C": "On your left.", + "D": "The trash can is slightly behind you, off to your right side.", + "E": "The trash can is to your left, a bit off from where you're facing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 12, + 14 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001", + "video_length": "00:20:04", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_4", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the trash can in for me?", + "answer": "The trash can is on your right.", + "question_chinese": "垃圾桶在我的哪个方向？", + "answer_chinese": "垃圾桶在你右边。", + "evidence": [ + { + "description": "The trash can is here.", + "timestamps": { + "start": "00:07:48", + "end": "00:07:49" + } + }, + { + "description": "The trash can is on your right.", + "timestamps": { + "start": "00:07:55", + "end": "00:07:56" + } + } + ], + "question_timestamps": "00:07:56", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "On your left.", + "C": "The trash can is behind you, not on the side you're facing.", + "D": "The trash can is on your right.", + "E": "The trash can is slightly to your left, rather than on your right." + }, + "answer_label": "D", + "answer_text": "The trash can is on your right.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_4", + "correct_answer": "The trash can is on your right.", + "original_options": { + "A": "In front of you.", + "B": "On your left.", + "C": "The trash can is behind you, not on the side you're facing.", + "D": "The trash can is on your right.", + "E": "The trash can is slightly to your left, rather than on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 13 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001", + "video_length": "00:20:04", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_6", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans have I passed through?", + "answer": "One.", + "question_chinese": "我经过了几个垃圾桶？", + "answer_chinese": "1个。", + "evidence": [ + { + "description": "You are passing by the first trash can.", + "timestamps": { + "start": "00:07:50", + "end": "00:07:55" + } + } + ], + "question_timestamps": "00:08:30", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "You passed by two trash cans during your walk through the area.", + "E": "You went past three trash cans as you moved along your route." + }, + "answer_label": "B", + "answer_text": "One.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_6", + "correct_answer": "One.", + "original_options": { + "A": "Two.", + "B": "One.", + "C": "Three.", + "D": "You passed by two trash cans during your walk through the area.", + "E": "You went past three trash cans as you moved along your route." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001", + "video_length": "00:20:04", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_5", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans have I passed through?", + "answer": "Two.", + "question_chinese": "我经过了几个垃圾桶？", + "answer_chinese": "2个。", + "evidence": [ + { + "description": "You are passing by the first trash can.", + "timestamps": { + "start": "00:07:50", + "end": "00:07:55" + } + }, + { + "description": "You are passing by the second trash can.", + "timestamps": { + "start": "00:08:45", + "end": "00:08:50" + } + } + ], + "question_timestamps": "00:09:00", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You passed by one trash can before moving on to the next part.", + "E": "You passed by three trash cans as you continued along your path." + }, + "answer_label": "C", + "answer_text": "Two.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_5", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Two.", + "D": "You passed by one trash can before moving on to the next part.", + "E": "You passed by three trash cans as you continued along your path." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 12 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001", + "video_length": "00:20:04", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_9", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I pick up things after standing up?", + "answer": "Three times.", + "question_chinese": "站起来后我捡了几次东西？", + "answer_chinese": "三次。", + "evidence": [ + { + "description": "You're picking up something.", + "timestamps": { + "start": "00:15:43", + "end": "00:15:45" + } + }, + { + "description": "You're picking up something.", + "timestamps": { + "start": "00:16:10", + "end": "00:16:13" + } + }, + { + "description": "You're picking up something.", + "timestamps": { + "start": "00:16:55", + "end": "00:16:58" + } + } + ], + "question_timestamps": "00:17:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Four times.", + "C": "You picked things up two times after you stood up from where you were.", + "D": "You picked things up four times after standing up and moving around again.", + "E": "Three times." + }, + "answer_label": "E", + "answer_text": "Three times.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_9", + "correct_answer": "Three times.", + "original_options": { + "A": "Two times.", + "B": "Four times.", + "C": "You picked things up two times after you stood up from where you were.", + "D": "You picked things up four times after standing up and moving around again.", + "E": "Three times." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 13 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001", + "video_length": "00:20:04", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_10", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many times did I pick up things after standing up?", + "answer": "Five times.", + "question_chinese": "站起来后我捡了几次东西？", + "answer_chinese": "五次。", + "evidence": [ + { + "description": "You're picking up something.", + "timestamps": { + "start": "00:15:43", + "end": "00:15:45" + } + }, + { + "description": "You're picking up something.", + "timestamps": { + "start": "00:16:10", + "end": "00:16:13" + } + }, + { + "description": "You're picking up something.", + "timestamps": { + "start": "00:16:55", + "end": "00:16:58" + } + }, + { + "description": "You're picking up something.", + "timestamps": { + "start": "00:17:07", + "end": "00:17:09" + } + }, + { + "description": "You're picking up something.", + "timestamps": { + "start": "00:17:16", + "end": "00:17:19" + } + } + ], + "question_timestamps": "00:17:20", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "Four times.", + "B": "Six times.", + "C": "Five times.", + "D": "You picked things up four times after standing up from your seat or position.", + "E": "You picked things up six times after standing up and continuing what you were doing." + }, + "answer_label": "C", + "answer_text": "Five times.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_10", + "correct_answer": "Five times.", + "original_options": { + "A": "Four times.", + "B": "Six times.", + "C": "Five times.", + "D": "You picked things up four times after standing up from your seat or position.", + "E": "You picked things up six times after standing up and continuing what you were doing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 15 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001", + "video_length": "00:20:04", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_8", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory after standing up?", + "answer": "You go left first, then turn left.", + "question_chinese": "站起来后我的移动轨迹是怎样的？", + "answer_chinese": "你先往左走,然后左转。", + "evidence": [ + { + "description": "You go left first.", + "timestamps": { + "start": "00:15:25", + "end": "00:15:27" + } + }, + { + "description": "Then turn left.", + "timestamps": { + "start": "00:15:30", + "end": "00:15:33" + } + } + ], + "question_timestamps": "00:15:35", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You go right, then turn right.", + "B": "You go left, then turn right.", + "C": "You go right first, then make a left turn after walking a few steps forward.", + "D": "You go left first, then turn left.", + "E": "You go straight ahead first, then turn left after a brief movement." + }, + "answer_label": "D", + "answer_text": "You go left first, then turn left.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_8", + "correct_answer": "You go left first, then turn left.", + "original_options": { + "A": "You go right, then turn right.", + "B": "You go left, then turn right.", + "C": "You go right first, then make a left turn after walking a few steps forward.", + "D": "You go left first, then turn left.", + "E": "You go straight ahead first, then turn left after a brief movement." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 12 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001", + "video_length": "00:20:04", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_7", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory after standing up?", + "answer": "You go left first, Then turn left, then turn right.", + "question_chinese": "站起来后我的移动轨迹是怎样的？", + "answer_chinese": "你先往左走,然后左转，然后右转。", + "evidence": [ + { + "description": "You go left first.", + "timestamps": { + "start": "00:15:25", + "end": "00:15:27" + } + }, + { + "description": "Then turn left.", + "timestamps": { + "start": "00:15:30", + "end": "00:15:33" + } + }, + { + "description": "Then turn right.", + "timestamps": { + "start": "00:15:39", + "end": "00:15:41" + } + } + ], + "question_timestamps": "00:15:50", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You go right first, then turn left.", + "B": "You go left first, then turn right.", + "C": "You go left first, then turn right, then turn left after that.", + "D": "You go right first, then turn right, then turn left to change direction.", + "E": "You go left first, Then turn left, then turn right." + }, + "answer_label": "E", + "answer_text": "You go left first, Then turn left, then turn right.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-001-mcq_fixed_7", + "correct_answer": "You go left first, Then turn left, then turn right.", + "original_options": { + "A": "You go right first, then turn left.", + "B": "You go left first, then turn right.", + "C": "You go left first, then turn right, then turn left after that.", + "D": "You go right first, then turn right, then turn left to change direction.", + "E": "You go left first, Then turn left, then turn right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 13 + ] + } +] diff --git a/QAs/fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq.json b/QAs/fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1230445cb720e43cae5755d001eba7d4c6e7f100 --- /dev/null +++ b/QAs/fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq.json @@ -0,0 +1,760 @@ +[ + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the trash can in for me?", + "answer": "The trash can is on your right.", + "question_chinese": "垃圾桶在我的哪个方向？", + "answer_chinese": "垃圾桶在你右边。", + "evidence": [ + { + "description": "The trash can is on your right.", + "timestamps": { + "start": "00:15:21", + "end": "00:15:22" + } + }, + { + "description": "The trash can is here.", + "timestamps": { + "start": "00:15:18", + "end": "00:15:20" + } + } + ], + "question_timestamps": "00:15:22", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The trash can is on your left.", + "B": "The trash can is in front of you.", + "C": "The trash can is behind you, so you need to turn around to see it.", + "D": "The trash can is above you, not at your side when you look around.", + "E": "The trash can is on your right." + }, + "answer_label": "E", + "answer_text": "The trash can is on your right.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_5", + "correct_answer": "The trash can is on your right.", + "original_options": { + "A": "The trash can is on your left.", + "B": "The trash can is in front of you.", + "C": "The trash can is behind you, so you need to turn around to see it.", + "D": "The trash can is above you, not at your side when you look around.", + "E": "The trash can is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 14 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_6", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the trash can in for me?", + "answer": "The trash can is behind you.", + "question_chinese": "垃圾桶在我的哪个方向？", + "answer_chinese": "垃圾桶在你后面。", + "evidence": [ + { + "description": "The trash can is here.", + "timestamps": { + "start": "00:15:18", + "end": "00:15:20" + } + }, + { + "description": "The trash can is behind you.", + "timestamps": { + "start": "00:15:30", + "end": "00:15:35" + } + } + ], + "question_timestamps": "00:15:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The trash can is in front of you.", + "B": "The trash can is on your right.", + "C": "The trash can is behind you.", + "D": "The trash can is to your left, beside you rather than behind your back.", + "E": "The trash can is above you, so you would look upward instead of turning around." + }, + "answer_label": "C", + "answer_text": "The trash can is behind you.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_6", + "correct_answer": "The trash can is behind you.", + "original_options": { + "A": "The trash can is in front of you.", + "B": "The trash can is on your right.", + "C": "The trash can is behind you.", + "D": "The trash can is to your left, beside you rather than behind your back.", + "E": "The trash can is above you, so you would look upward instead of turning around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 15 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_11", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get on the motorcycle directly?", + "answer": "No,you can't.", + "question_chinese": "我可以直接上摩托车吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The motorcycle is in front of you.", + "timestamps": { + "start": "00:19:45", + "end": "00:19:46" + } + }, + { + "description": "It's too far from you.", + "timestamps": { + "start": "00:19:45", + "end": "00:19:46" + } + }, + { + "description": "You are not allowed to go up.", + "timestamps": { + "start": "00:19:45", + "end": "00:19:46" + } + } + ], + "question_timestamps": "00:19:46", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_11", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_12", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get on the motorcycle directly?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接上摩托车吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The motorcycle is right in front of you.", + "timestamps": { + "start": "00:19:48", + "end": "00:19:49" + } + }, + { + "description": "You can go up directly.", + "timestamps": { + "start": "00:19:48", + "end": "00:19:49" + } + } + ], + "question_timestamps": "00:19:50", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_12", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_7", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I see?", + "answer": "Two.", + "question_chinese": "我看到了几个垃圾桶？", + "answer_chinese": "两个。", + "evidence": [ + { + "description": "You saw a trash can.", + "timestamps": { + "start": "00:15:18", + "end": "00:15:20" + } + }, + { + "description": "You saw a trash can.", + "timestamps": { + "start": "00:16:05", + "end": "00:16:06" + } + } + ], + "question_timestamps": "00:16:10", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Zero.", + "B": "One.", + "C": "Three trash cans, one close by and two farther away as you look around the area.", + "D": "Two.", + "E": "Four trash cans, spaced out so you notice them at different moments while moving through." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_7", + "correct_answer": "Two.", + "original_options": { + "A": "Zero.", + "B": "One.", + "C": "Three trash cans, one close by and two farther away as you look around the area.", + "D": "Two.", + "E": "Four trash cans, spaced out so you notice them at different moments while moving through." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_8", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I see?", + "answer": "Three.", + "question_chinese": "我看到了几个垃圾桶？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "You saw a trash can.", + "timestamps": { + "start": "00:15:18", + "end": "00:15:20" + } + }, + { + "description": "You saw a trash can.", + "timestamps": { + "start": "00:16:05", + "end": "00:16:06" + } + }, + { + "description": "You saw a trash can.", + "timestamps": { + "start": "00:16:50", + "end": "00:16:51" + } + } + ], + "question_timestamps": "00:17:00", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "Five trash cans, appearing one after another at separate points as you walk along the path.", + "E": "Six trash cans, with several visible at once when you pass by the same spot." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_8", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "Five trash cans, appearing one after another at separate points as you walk along the path.", + "E": "Six trash cans, with several visible at once when you pass by the same spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_0", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times have I squatted down?", + "answer": "Three times.", + "question_chinese": "我蹲下去过几次？", + "answer_chinese": "三次。", + "evidence": [ + { + "description": "You're squatting down.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You're squatting down.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "You're squatting down.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two times.", + "B": "Three times.", + "C": "Four times.", + "D": "Five times, with a couple of squats close together during the activity.", + "E": "Six times, counting each squat as you lower down and stand back up." + }, + "answer_label": "B", + "answer_text": "Three times.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_0", + "correct_answer": "Three times.", + "original_options": { + "A": "Two times.", + "B": "Three times.", + "C": "Four times.", + "D": "Five times, with a couple of squats close together during the activity.", + "E": "Six times, counting each squat as you lower down and stand back up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 13 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many times have I squatted down?", + "answer": "Four times.", + "question_chinese": "我蹲下去过几次？", + "answer_chinese": "四次。", + "evidence": [ + { + "description": "You're squatting down.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You're squatting down.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "You're squatting down.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + }, + { + "description": "You're squatting down.", + "timestamps": { + "start": "00:00:25", + "end": "00:00:26" + } + } + ], + "question_timestamps": "00:00:30", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One time.", + "B": "Two times.", + "C": "Four times.", + "D": "Three times total, with one squat happening briefly before you stood back up again.", + "E": "Five times altogether, including an extra squat right at the end before moving on." + }, + "answer_label": "C", + "answer_text": "Four times.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_1", + "correct_answer": "Four times.", + "original_options": { + "A": "One time.", + "B": "Two times.", + "C": "Four times.", + "D": "Three times total, with one squat happening briefly before you stood back up again.", + "E": "Five times altogether, including an extra squat right at the end before moving on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_9", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory after passing through the last trash can?", + "answer": "You go left first, then turn right, and then turn left.", + "question_chinese": "经过最后一个垃圾桶后我的移动轨迹是怎样的？", + "answer_chinese": "你先往左边走，然后右转，然后左转。", + "evidence": [ + { + "description": "You go left first", + "timestamps": { + "start": "00:16:52", + "end": "00:16:54" + } + }, + { + "description": "Then turn right", + "timestamps": { + "start": "00:17:05", + "end": "00:17:10" + } + }, + { + "description": "Then turn left.", + "timestamps": { + "start": "00:17:28", + "end": "00:17:32" + } + } + ], + "question_timestamps": "00:17:40", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You go right, then left, then right.", + "B": "You go left first, then turn right, and then turn left.", + "C": "You go left, then left, then right.", + "D": "You turn right first, continue forward, then turn left as you pass beyond the trash can.", + "E": "You go straight after the trash can, then turn right, and finally turn left near the end." + }, + "answer_label": "B", + "answer_text": "You go left first, then turn right, and then turn left.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_9", + "correct_answer": "You go left first, then turn right, and then turn left.", + "original_options": { + "A": "You go right, then left, then right.", + "B": "You go left first, then turn right, and then turn left.", + "C": "You go left, then left, then right.", + "D": "You turn right first, continue forward, then turn left as you pass beyond the trash can.", + "E": "You go straight after the trash can, then turn right, and finally turn left near the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 17 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_10", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory after passing through the last trash can?", + "answer": "You go left first, then turn right, and then turn left, then turn right.", + "question_chinese": "经过最后一个垃圾桶后我的移动轨迹是怎样的？", + "answer_chinese": "你先往左边走，然后右转，然后左转，然后右转。", + "evidence": [ + { + "description": "You go left first.", + "timestamps": { + "start": "00:16:52", + "end": "00:16:54" + } + }, + { + "description": "Then turn right.", + "timestamps": { + "start": "00:17:05", + "end": "00:17:10" + } + }, + { + "description": "Then turn left.", + "timestamps": { + "start": "00:17:28", + "end": "00:17:32" + } + }, + { + "description": "Then turn right.", + "timestamps": { + "start": "00:17:45", + "end": "00:17:50" + } + } + ], + "question_timestamps": "00:18:00", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Right, left, right, left.", + "B": "Left, right, right, left.", + "C": "You go left first, then turn right, and then turn left, then turn right.", + "D": "You go left first, then turn left, then turn right, and finally turn right again.", + "E": "You go right first, then turn right, then turn left, and finally turn left once more." + }, + "answer_label": "C", + "answer_text": "You go left first, then turn right, and then turn left, then turn right.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_10", + "correct_answer": "You go left first, then turn right, and then turn left, then turn right.", + "original_options": { + "A": "Right, left, right, left.", + "B": "Left, right, right, left.", + "C": "You go left first, then turn right, and then turn left, then turn right.", + "D": "You go left first, then turn left, then turn right, and finally turn right again.", + "E": "You go right first, then turn right, then turn left, and finally turn left once more." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_3", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the bicycle?", + "answer": "The bicycle is in front of you.", + "question_chinese": "自行车在哪里？", + "answer_chinese": "自行车在你面前。", + "evidence": [ + { + "description": "The bicycle is here.", + "timestamps": { + "start": "00:15:03", + "end": "00:15:04" + } + }, + { + "description": "The bicycle is in front of you.", + "timestamps": { + "start": "00:15:03", + "end": "00:15:04" + } + } + ], + "question_timestamps": "00:15:05", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The bicycle is behind you.", + "B": "The bicycle is to your left.", + "C": "The bicycle is off to your right side, not directly in your line of sight ahead.", + "D": "The bicycle is beside you, close enough that you’d need to turn your head rather than look forward.", + "E": "The bicycle is in front of you." + }, + "answer_label": "E", + "answer_text": "The bicycle is in front of you.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_3", + "correct_answer": "The bicycle is in front of you.", + "original_options": { + "A": "The bicycle is behind you.", + "B": "The bicycle is to your left.", + "C": "The bicycle is off to your right side, not directly in your line of sight ahead.", + "D": "The bicycle is beside you, close enough that you’d need to turn your head rather than look forward.", + "E": "The bicycle is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 19 + ] + }, + { + "video_uid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002", + "video_length": "00:21:07", + "qid": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_4", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the bicycle?", + "answer": "The bicycle is behind you.", + "question_chinese": "自行车在哪里？", + "answer_chinese": "自行车在你后面。", + "evidence": [ + { + "description": "The bicycle is behind you.", + "timestamps": { + "start": "00:15:14", + "end": "00:15:15" + } + }, + { + "description": "The bicycle is here.", + "timestamps": { + "start": "00:15:03", + "end": "00:15:04" + } + } + ], + "question_timestamps": "00:15:17", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "The bicycle is in front of you.", + "B": "The bicycle is behind you.", + "C": "The bicycle is to your left.", + "D": "The bicycle is to your right side, not positioned behind you as you face forward.", + "E": "The bicycle is beside you rather than behind, sitting near your position at the same depth." + }, + "answer_label": "B", + "answer_text": "The bicycle is behind you.", + "q_id": "fcc14fe4-35e1-4854-b368-4541ae7bcf58-002-mcq_fixed_4", + "correct_answer": "The bicycle is behind you.", + "original_options": { + "A": "The bicycle is in front of you.", + "B": "The bicycle is behind you.", + "C": "The bicycle is to your left.", + "D": "The bicycle is to your right side, not positioned behind you as you face forward.", + "E": "The bicycle is beside you rather than behind, sitting near your position at the same depth." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 16 + ] + } +] diff --git a/QAs/scene0709_00-0-mcq.json b/QAs/scene0709_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b0e7ef6515c38b621131b3eab3f228c6dde22a64 --- /dev/null +++ b/QAs/scene0709_00-0-mcq.json @@ -0,0 +1,736 @@ +[ + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I want to eat oatmeal . What should I do？", + "answer": "Walk forward, pick up Cheerios cereal next to the sink and get some water , mix them together, and heat it in the microwave on your right.", + "question_chinese": "我想吃麦片，我应该怎么做？", + "answer_chinese": "在你面前拿起Cheerios麦片，接点水，混到一起放在你右后方的微波炉里加热。", + "evidence": [ + { + "description": "The microwave can heat food", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The oatmeal is next to the sink.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Microwave it in front.", + "B": "Heat it in the microwave left.", + "C": "Walk forward, grab Cheerios by the sink, add water, mix, then microwave it behind you.", + "D": "Walk forward, pick up Cheerios cereal next to the sink and get some water , mix them together, and heat it in the microwave on your right.", + "E": "Walk forward, take Cheerios by the sink, add water, mix, and heat it in the microwave in front." + }, + "answer_label": "D", + "answer_text": "Walk forward, pick up Cheerios cereal next to the sink and get some water , mix them together, and heat it in the microwave on your right.", + "q_id": "scene0709_00-0_1", + "correct_answer": "Walk forward, pick up Cheerios cereal next to the sink and get some water , mix them together, and heat it in the microwave on your right.", + "original_options": { + "A": "Microwave it in front.", + "B": "Heat it in the microwave left.", + "C": "Walk forward, grab Cheerios by the sink, add water, mix, then microwave it behind you.", + "D": "Walk forward, pick up Cheerios cereal next to the sink and get some water , mix them together, and heat it in the microwave on your right.", + "E": "Walk forward, take Cheerios by the sink, add water, mix, and heat it in the microwave in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 15, + 18 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I want to eat oatmeal . What should I do？", + "answer": "Pick up Cheerios cereal next to the sink in front of you and get some water, mix them together, and heat it in the microwave behind and to your right.", + "question_chinese": "我想吃麦片，我应该怎么做？", + "answer_chinese": "在你面前拿起Cheerios麦片，接点水，混到一起放在你右后方的微波炉里加热。", + "evidence": [ + { + "description": "Cheerios cereal is next to the sink.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + }, + { + "description": "The microwave can heat food.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Microwave it in front.", + "B": "Heat it in the microwave left.", + "C": "Pick up Cheerios cereal next to the sink in front of you and get some water, mix them together, and heat it in the microwave behind and to your right.", + "D": "Pick up Cheerios by the sink, add water, mix, and heat it in the microwave to your left.", + "E": "Grab Cheerios by the sink, add water, mix, then microwave it in front of you." + }, + "answer_label": "C", + "answer_text": "Pick up Cheerios cereal next to the sink in front of you and get some water, mix them together, and heat it in the microwave behind and to your right.", + "q_id": "scene0709_00-0_2", + "correct_answer": "Pick up Cheerios cereal next to the sink in front of you and get some water, mix them together, and heat it in the microwave behind and to your right.", + "original_options": { + "A": "Microwave it in front.", + "B": "Heat it in the microwave left.", + "C": "Pick up Cheerios cereal next to the sink in front of you and get some water, mix them together, and heat it in the microwave behind and to your right.", + "D": "Pick up Cheerios by the sink, add water, mix, and heat it in the microwave to your left.", + "E": "Grab Cheerios by the sink, add water, mix, then microwave it in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 18, + 15 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_5_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my microwave?", + "answer": "The microwave is on the table in front of you, next to the stove.", + "question_chinese": "我的微波炉在哪里？", + "answer_chinese": "微波炉在你前面炉灶旁边的桌子上。", + "evidence": [ + { + "description": "The microwave is next to the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on the table behind you.", + "B": "It’s on the table to your left.", + "C": "The microwave is on the table to your right, positioned next to the stove.", + "D": "The microwave is on the table in front of you, next to the stove.", + "E": "The microwave is on the table behind you, sitting next to the stove." + }, + "answer_label": "D", + "answer_text": "The microwave is on the table in front of you, next to the stove.", + "q_id": "scene0709_00-0_5_T1", + "correct_answer": "The microwave is on the table in front of you, next to the stove.", + "original_options": { + "A": "It’s on the table behind you.", + "B": "It’s on the table to your left.", + "C": "The microwave is on the table to your right, positioned next to the stove.", + "D": "The microwave is on the table in front of you, next to the stove.", + "E": "The microwave is on the table behind you, sitting next to the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 14, + 13 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_5_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my microwave?", + "answer": "The microwave is on the table to your left, next to the stove.", + "question_chinese": "我的微波炉在哪里？", + "answer_chinese": "微波炉在你左边炉灶旁边的桌子上。", + "evidence": [ + { + "description": "The microwave is next to the stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The microwave is on the table to your left, next to the stove.", + "B": "It’s on the table in front.", + "C": "It’s on the table behind you.", + "D": "The microwave is on the table to your right, next to the stove.", + "E": "The microwave is on the table in front of you, placed away from the stove." + }, + "answer_label": "A", + "answer_text": "The microwave is on the table to your left, next to the stove.", + "q_id": "scene0709_00-0_5_T2", + "correct_answer": "The microwave is on the table to your left, next to the stove.", + "original_options": { + "A": "The microwave is on the table to your left, next to the stove.", + "B": "It’s on the table in front.", + "C": "It’s on the table behind you.", + "D": "The microwave is on the table to your right, next to the stove.", + "E": "The microwave is on the table in front of you, placed away from the stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 15 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_2_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory like?", + "answer": "First you were facing a row of kitchen appliances, then you turned clockwise and saw a window.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先你面对一排厨具，然后顺时针旋转看到了窗户", + "evidence": [ + { + "description": "You first face a row of kitchen utensils.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Rotating clockwise, I saw the window.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned counterclockwise to the window.", + "B": "You walked straight to the table.", + "C": "First you were facing a row of kitchen appliances, then you turned clockwise and saw a window.", + "D": "First you faced the kitchen appliances, then you turned counterclockwise and saw the window.", + "E": "You started facing the appliances, turned clockwise to the refrigerator, then moved toward the table." + }, + "answer_label": "C", + "answer_text": "First you were facing a row of kitchen appliances, then you turned clockwise and saw a window.", + "q_id": "scene0709_00-0_2_T1", + "correct_answer": "First you were facing a row of kitchen appliances, then you turned clockwise and saw a window.", + "original_options": { + "A": "You turned counterclockwise to the window.", + "B": "You walked straight to the table.", + "C": "First you were facing a row of kitchen appliances, then you turned clockwise and saw a window.", + "D": "First you faced the kitchen appliances, then you turned counterclockwise and saw the window.", + "E": "You started facing the appliances, turned clockwise to the refrigerator, then moved toward the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_4", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement status in the room?", + "answer": "First, you are facing a row of kitchen utensils. Then, rotate 180 degrees and move slightly, and you see the window and the refrigerator. Finally, you see the table.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先你面对一排厨具，然后顺时针旋转看到了窗户，继续旋转看到了冰箱，继续旋转看到并走向了桌子。", + "evidence": [ + { + "description": "You first face a row of kitchen utensils.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Rotating clockwise, I saw the window.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:06" + } + }, + { + "description": "Kept spinning and saw the refrigerator.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:13" + } + }, + { + "description": "Keep turning, see the table.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "First, you are facing a row of kitchen utensils. Then, rotate 180 degrees and move slightly, and you see the window and the refrigerator. Finally, you see the table.", + "B": "You turned clockwise and saw a window.", + "C": "You never turned around in place.", + "D": "First you faced the kitchen utensils, then turned clockwise slightly and saw the refrigerator before the window.", + "E": "You began facing the utensils, rotated 90 degrees, saw the window, then walked directly to the table." + }, + "answer_label": "A", + "answer_text": "First, you are facing a row of kitchen utensils. Then, rotate 180 degrees and move slightly, and you see the window and the refrigerator. Finally, you see the table.", + "q_id": "scene0709_00-0_4", + "correct_answer": "First, you are facing a row of kitchen utensils. Then, rotate 180 degrees and move slightly, and you see the window and the refrigerator. Finally, you see the table.", + "original_options": { + "A": "First, you are facing a row of kitchen utensils. Then, rotate 180 degrees and move slightly, and you see the window and the refrigerator. Finally, you see the table.", + "B": "You turned clockwise and saw a window.", + "C": "You never turned around in place.", + "D": "First you faced the kitchen utensils, then turned clockwise slightly and saw the refrigerator before the window.", + "E": "You began facing the utensils, rotated 90 degrees, saw the window, then walked directly to the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 17 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_2_T2", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory like?", + "answer": "First you were facing a row of kitchen appliances, then you turned clockwise and saw a window, kept turning and saw the refrigerator, and kept turning until you saw the table and walked toward it.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先你面对一排厨具，然后顺时针旋转看到了窗户，继续旋转看到了冰箱，继续旋转看到并走向了桌子。", + "evidence": [ + { + "description": "First, you face a row of kitchen utensils.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "you rotate clockwise and see the window.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + }, + { + "description": "Continue rotating and you see the refrigerator.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + }, + { + "description": "Keep rotating and you see and walk toward the table.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned clockwise and stopped at the window.", + "B": "You turned counterclockwise to the refrigerator.", + "C": "First you were facing a row of kitchen appliances, then you turned clockwise and saw a window, kept turning and saw the refrigerator, and kept turning until you saw the table and walked toward it.", + "D": "First you faced the appliances, then turned counterclockwise to the window, and walked straight to the table.", + "E": "You started at the appliances, turned clockwise to the refrigerator first, then saw the window, then approached the table." + }, + "answer_label": "C", + "answer_text": "First you were facing a row of kitchen appliances, then you turned clockwise and saw a window, kept turning and saw the refrigerator, and kept turning until you saw the table and walked toward it.", + "q_id": "scene0709_00-0_2_T2", + "correct_answer": "First you were facing a row of kitchen appliances, then you turned clockwise and saw a window, kept turning and saw the refrigerator, and kept turning until you saw the table and walked toward it.", + "original_options": { + "A": "You turned clockwise and stopped at the window.", + "B": "You turned counterclockwise to the refrigerator.", + "C": "First you were facing a row of kitchen appliances, then you turned clockwise and saw a window, kept turning and saw the refrigerator, and kept turning until you saw the table and walked toward it.", + "D": "First you faced the appliances, then turned counterclockwise to the window, and walked straight to the table.", + "E": "You started at the appliances, turned clockwise to the refrigerator first, then saw the window, then approached the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 17, + 19 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_1_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my cereal?", + "answer": "The cereal is on the countertop to your left, next to the microwave.", + "question_chinese": "我的麦片在哪里", + "answer_chinese": "麦片在你左边的台面上，微波炉的旁边", + "evidence": [ + { + "description": "The cereal is on the counter in front of you next to the microwave.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on the counter in front.", + "B": "It’s on the countertop to your right.", + "C": "The cereal is on the countertop behind you, next to the microwave.", + "D": "The cereal is on the countertop to your left, next to the microwave.", + "E": "The cereal is on the countertop in front of you, placed beside the microwave." + }, + "answer_label": "D", + "answer_text": "The cereal is on the countertop to your left, next to the microwave.", + "q_id": "scene0709_00-0_1_T1", + "correct_answer": "The cereal is on the countertop to your left, next to the microwave.", + "original_options": { + "A": "It’s on the counter in front.", + "B": "It’s on the countertop to your right.", + "C": "The cereal is on the countertop behind you, next to the microwave.", + "D": "The cereal is on the countertop to your left, next to the microwave.", + "E": "The cereal is on the countertop in front of you, placed beside the microwave." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 12, + 14 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_1_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my cereal?", + "answer": "The cereal is on the countertop behind you, next to the microwave.", + "question_chinese": "我的麦片在哪里", + "answer_chinese": "麦片在你后面的台面上，微波炉的旁边", + "evidence": [ + { + "description": "The cereal is on the counter in front of you next to the microwave.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the countertop in front of you.", + "B": "On the countertop to your left.", + "C": "The cereal is on the countertop behind you, next to the microwave.", + "D": "The cereal is on the countertop behind you, far from the microwave.", + "E": "The cereal is on the countertop to your right, right next to the microwave." + }, + "answer_label": "C", + "answer_text": "The cereal is on the countertop behind you, next to the microwave.", + "q_id": "scene0709_00-0_1_T2", + "correct_answer": "The cereal is on the countertop behind you, next to the microwave.", + "original_options": { + "A": "On the countertop in front of you.", + "B": "On the countertop to your left.", + "C": "The cereal is on the countertop behind you, next to the microwave.", + "D": "The cereal is on the countertop behind you, far from the microwave.", + "E": "The cereal is on the countertop to your right, right next to the microwave." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 14 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_3_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the window relative to me?", + "answer": "It’s in front of you.", + "question_chinese": "窗户在我的哪个方位？", + "answer_chinese": "在你的前方", + "evidence": [ + { + "description": "The window is in front of you.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s behind you.", + "B": "It’s in front of you.", + "C": "It’s to your left.", + "D": "It’s to your right, not directly in your line of sight.", + "E": "It’s above you, higher than your eye level rather than ahead." + }, + "answer_label": "B", + "answer_text": "It’s in front of you.", + "q_id": "scene0709_00-0_3_T1", + "correct_answer": "It’s in front of you.", + "original_options": { + "A": "It’s behind you.", + "B": "It’s in front of you.", + "C": "It’s to your left.", + "D": "It’s to your right, not directly in your line of sight.", + "E": "It’s above you, higher than your eye level rather than ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 12, + 12 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_3_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the window relative to me?", + "answer": "It’s to your left.", + "question_chinese": "窗户在我的哪个方位？", + "answer_chinese": "在你的左边", + "evidence": [ + { + "description": "The window is in front of you.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s to your left.", + "C": "It’s behind you.", + "D": "It’s to your right, opposite the side you’re facing toward.", + "E": "It’s above you, not on either side of your position at all." + }, + "answer_label": "B", + "answer_text": "It’s to your left.", + "q_id": "scene0709_00-0_3_T2", + "correct_answer": "It’s to your left.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s to your left.", + "C": "It’s behind you.", + "D": "It’s to your right, opposite the side you’re facing toward.", + "E": "It’s above you, not on either side of your position at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 12, + 13 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_4_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Between the microwave and the oven in front of me, which one is closer to me?", + "answer": "The microwave in front of you was closer to you.", + "question_chinese": "面前的微波炉和烤箱哪个离我更近？", + "answer_chinese": "面前的微波炉离你更近。", + "evidence": [ + { + "description": "The microwave is on the counter in front of me, and the oven is to its right, farther away from me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The oven in front of you was closer to you.", + "B": "The microwave in front of you was closer to you." + }, + "answer_label": "B", + "answer_text": "The microwave in front of you was closer to you.", + "q_id": "scene0709_00-0_4_T1", + "correct_answer": "The microwave in front of you was closer to you.", + "original_options": { + "A": "The oven in front of you was closer to you.", + "B": "The microwave in front of you was closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "scene0709_00-0", + "video_length": "00:00:20", + "qid": "scene0709_00-0_4_T2", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Between the microwave and the oven in front of me, which one is closer to me?", + "answer": "The oven in front of you was closer to you.", + "question_chinese": "面前的微波炉和烤箱哪个离我更近？", + "answer_chinese": "面前的烤箱离你更近。", + "evidence": [ + { + "description": "The microwave is on the counter in front of me, and the oven is to its right, farther away from me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The oven in front of you was closer to you.", + "B": "The microwave in front of you was closer to you." + }, + "answer_label": "A", + "answer_text": "The oven in front of you was closer to you.", + "q_id": "scene0709_00-0_4_T2", + "correct_answer": "The oven in front of you was closer to you.", + "original_options": { + "A": "The oven in front of you was closer to you.", + "B": "The microwave in front of you was closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + } +] diff --git a/QAs/scene0710_00-0-mcq.json b/QAs/scene0710_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2701135d2d9fdfc641acda3e84c8b246d1a25948 --- /dev/null +++ b/QAs/scene0710_00-0-mcq.json @@ -0,0 +1,834 @@ +[ + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_1_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the teddy bear bag relative to me?", + "answer": "The teddy bear bag is in front of you.", + "question_chinese": "小熊包在相对于我的哪里？", + "answer_chinese": "小熊包在相对于你的正前方。", + "evidence": [ + { + "description": "The bear is wrapped in front of you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The teddy bear bag is behind you.", + "B": "The teddy bear bag is to your left.", + "C": "The teddy bear bag is in front of you.", + "D": "The teddy bear bag is to your right, not directly in front of you.", + "E": "The teddy bear bag is above you, positioned higher rather than straight ahead." + }, + "answer_label": "C", + "answer_text": "The teddy bear bag is in front of you.", + "q_id": "scene0710_00-0_1_T1", + "correct_answer": "The teddy bear bag is in front of you.", + "original_options": { + "A": "The teddy bear bag is behind you.", + "B": "The teddy bear bag is to your left.", + "C": "The teddy bear bag is in front of you.", + "D": "The teddy bear bag is to your right, not directly in front of you.", + "E": "The teddy bear bag is above you, positioned higher rather than straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 14, + 13 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_1_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the teddy bear bag relative to me?", + "answer": "The teddy bear bag is to your right.", + "question_chinese": "小熊包在相对于我的哪里？", + "answer_chinese": "小熊包在相对于你的右边。", + "evidence": [ + { + "description": "The bear bag is on the right relative to you.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The teddy bear bag is to your right.", + "B": "In front of you.", + "C": "Behind you.", + "D": "The teddy bear bag is to your left, on the opposite side from your right.", + "E": "The teddy bear bag is above you, not located on either side of you." + }, + "answer_label": "A", + "answer_text": "The teddy bear bag is to your right.", + "q_id": "scene0710_00-0_1_T2", + "correct_answer": "The teddy bear bag is to your right.", + "original_options": { + "A": "The teddy bear bag is to your right.", + "B": "In front of you.", + "C": "Behind you.", + "D": "The teddy bear bag is to your left, on the opposite side from your right.", + "E": "The teddy bear bag is above you, not located on either side of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 15, + 14 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_6", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Can I use the computer directly？", + "answer": "Yes,you can.", + "question_chinese": "我可以直接使用电脑吗？", + "answer_chinese": "你可以。", + "evidence": [ + { + "description": "There is a mouse and a keyboard on the table.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + }, + { + "description": "The computer is turned on.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "There is a router at the back of the computer.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "The router is on as indicated by the blue light.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "No,you cannot.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "scene0710_00-0_6", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No,you cannot.", + "B": "Yes,you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_5", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Can I use the computer directly？", + "answer": "Yes,you can.", + "question_chinese": "我可以直接使用电脑吗？", + "answer_chinese": "你可以。", + "evidence": [ + { + "description": "The computer is turned on.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "There is a router behind the computer.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "The router is on, indicated by the blue light.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "There is a mouse and a keyboard on the table.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No,you cannot." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "scene0710_00-0_5", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No,you cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_3", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the bear directly?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接拿到熊吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "You are standing next to the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The bear is on the table.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No,you cannot." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "scene0710_00-0_3", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No,you cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_4", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the bear directly?", + "answer": "Yes,you can.", + "question_chinese": "我可以直接拿到熊吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The bear is on the table.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "You are standing next to the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,you cannot.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "scene0710_00-0_4", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No,you cannot.", + "B": "Yes,you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_2_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach out and grab the bear right now?", + "answer": "You can’t reach the bear right now; the bear is in front of you.", + "question_chinese": "我现在伸手可以拿到熊吗？", + "answer_chinese": "你现在伸手拿不到熊，熊在你的正前方。", + "evidence": [ + { + "description": "You can't reach the bear now, the bear is right in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the bear right now; the bear is in front of you.", + "B": "You can reach the bear right now; the bear is below your right hand." + }, + "answer_label": "A", + "answer_text": "You can’t reach the bear right now; the bear is in front of you.", + "q_id": "scene0710_00-0_2_T2", + "correct_answer": "You can’t reach the bear right now; the bear is in front of you.", + "original_options": { + "A": "You can’t reach the bear right now; the bear is in front of you.", + "B": "You can reach the bear right now; the bear is below your right hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_2_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach out and grab the bear right now?", + "answer": "You can reach the bear right now; the bear is below your right hand.", + "question_chinese": "我现在伸手可以拿到熊吗？", + "answer_chinese": "你现在伸手可以拿到熊，熊在你的右手下面。", + "evidence": [ + { + "description": "You can reach for the bear now. The bear is under your right hand.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the bear right now; the bear is in front of you.", + "B": "You can reach the bear right now; the bear is below your right hand." + }, + "answer_label": "B", + "answer_text": "You can reach the bear right now; the bear is below your right hand.", + "q_id": "scene0710_00-0_2_T1", + "correct_answer": "You can reach the bear right now; the bear is below your right hand.", + "original_options": { + "A": "You can’t reach the bear right now; the bear is in front of you.", + "B": "You can reach the bear right now; the bear is below your right hand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my bag?", + "answer": "It is on the floor to the right of your chair.", + "question_chinese": "我的包放在哪里？", + "answer_chinese": "它放在你凳子右边的地上。", + "evidence": [ + { + "description": "The bag is on the floor to the right of the green chair.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the floor left of your chair.", + "B": "It is on the floor to the right of your chair.", + "C": "On the shelf behind your chair.", + "D": "It is on the floor in front of your desk, a little farther than your chair.", + "E": "It is on the couch beside you, not on the floor next to your chair." + }, + "answer_label": "B", + "answer_text": "It is on the floor to the right of your chair.", + "q_id": "scene0710_00-0_1", + "correct_answer": "It is on the floor to the right of your chair.", + "original_options": { + "A": "On the floor left of your chair.", + "B": "It is on the floor to the right of your chair.", + "C": "On the shelf behind your chair.", + "D": "It is on the floor in front of your desk, a little farther than your chair.", + "E": "It is on the couch beside you, not on the floor next to your chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 15 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my bag?", + "answer": "It is on the floor to the right of your chair.", + "question_chinese": "我的包放在哪里？", + "answer_chinese": "它放在你凳子右边的地上。", + "evidence": [ + { + "description": "The bag is on the floor to the right of the green chair.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the floor left of your chair.", + "B": "On the floor in front of your chair.", + "C": "It is under the table near your chair, not on the floor beside the chair.", + "D": "It is next to your chair but tucked inside another bag rather than on the floor.", + "E": "It is on the floor to the right of your chair." + }, + "answer_label": "E", + "answer_text": "It is on the floor to the right of your chair.", + "q_id": "scene0710_00-0_2", + "correct_answer": "It is on the floor to the right of your chair.", + "original_options": { + "A": "On the floor left of your chair.", + "B": "On the floor in front of your chair.", + "C": "It is under the table near your chair, not on the floor beside the chair.", + "D": "It is next to your chair but tucked inside another bag rather than on the floor.", + "E": "It is on the floor to the right of your chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 16 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_10", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the robot vacuum cleaner?", + "answer": "The robot vacuum cleaner is on the floor to the left of the electronic piano.", + "question_chinese": "扫地机器人在哪里？", + "answer_chinese": "扫地机器人在电子琴左侧的地面上。", + "evidence": [ + { + "description": "The robot vacuum cleaner is on the floor to the left of the electronic piano.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the floor in front of the piano.", + "B": "On the floor behind the piano.", + "C": "The robot vacuum cleaner is under the table next to the electronic piano, not on the open floor.", + "D": "The robot vacuum cleaner is on a shelf to the right of the electronic piano instead of on the floor.", + "E": "The robot vacuum cleaner is on the floor to the left of the electronic piano." + }, + "answer_label": "E", + "answer_text": "The robot vacuum cleaner is on the floor to the left of the electronic piano.", + "q_id": "scene0710_00-0_10", + "correct_answer": "The robot vacuum cleaner is on the floor to the left of the electronic piano.", + "original_options": { + "A": "On the floor in front of the piano.", + "B": "On the floor behind the piano.", + "C": "The robot vacuum cleaner is under the table next to the electronic piano, not on the open floor.", + "D": "The robot vacuum cleaner is on a shelf to the right of the electronic piano instead of on the floor.", + "E": "The robot vacuum cleaner is on the floor to the left of the electronic piano." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 18, + 20 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_9", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the robot vacuum cleaner?", + "answer": "The robot vacuum cleaner is on the floor to the left of the electronic piano.", + "question_chinese": "扫地机器人在哪里？", + "answer_chinese": "扫地机器人在电子琴左侧的地面上。", + "evidence": [ + { + "description": "The robot vacuum cleaner is on the floor to the left of the electronic piano.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The robot vacuum cleaner is on the floor to the left of the electronic piano.", + "B": "To the right of the piano.", + "C": "In front of the piano.", + "D": "The robot vacuum cleaner is on the floor behind the electronic piano, close to the wall.", + "E": "The robot vacuum cleaner is on the floor near the electronic piano, but positioned on its right side." + }, + "answer_label": "A", + "answer_text": "The robot vacuum cleaner is on the floor to the left of the electronic piano.", + "q_id": "scene0710_00-0_9", + "correct_answer": "The robot vacuum cleaner is on the floor to the left of the electronic piano.", + "original_options": { + "A": "The robot vacuum cleaner is on the floor to the left of the electronic piano.", + "B": "To the right of the piano.", + "C": "In front of the piano.", + "D": "The robot vacuum cleaner is on the floor behind the electronic piano, close to the wall.", + "E": "The robot vacuum cleaner is on the floor near the electronic piano, but positioned on its right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 16, + 18 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_4_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the blue pencil case on the table relative to me?", + "answer": "The blue pencil case on the table is in front of you.", + "question_chinese": "桌子上的蓝色笔袋在相对于我的哪里？", + "answer_chinese": "桌子上的蓝色笔袋在相对于你的正前方。", + "evidence": [ + { + "description": "The pencil case is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you on the table.", + "B": "The blue pencil case on the table is in front of you.", + "C": "To your left on the table.", + "D": "The blue pencil case on the table is to your right, not directly ahead of you.", + "E": "The blue pencil case on the table is behind you, so you would need to turn around to see it." + }, + "answer_label": "B", + "answer_text": "The blue pencil case on the table is in front of you.", + "q_id": "scene0710_00-0_4_T1", + "correct_answer": "The blue pencil case on the table is in front of you.", + "original_options": { + "A": "Behind you on the table.", + "B": "The blue pencil case on the table is in front of you.", + "C": "To your left on the table.", + "D": "The blue pencil case on the table is to your right, not directly ahead of you.", + "E": "The blue pencil case on the table is behind you, so you would need to turn around to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 20 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_4_T2", + "qid_num": 14, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the blue pencil case on the table relative to me?", + "answer": "The blue pencil case on the table is to your right.", + "question_chinese": "桌子上的蓝色笔袋在相对于我的哪里？", + "answer_chinese": "桌子上的蓝色笔袋在相对于你的右边。", + "evidence": [ + { + "description": "The blue pencil case is on your right.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The blue pencil case on the table is to your right.", + "D": "The blue pencil case on the table is behind you, not on the right side.", + "E": "The blue pencil case on the table is in front of you, positioned straight ahead rather than to the right." + }, + "answer_label": "C", + "answer_text": "The blue pencil case on the table is to your right.", + "q_id": "scene0710_00-0_4_T2", + "correct_answer": "The blue pencil case on the table is to your right.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The blue pencil case on the table is to your right.", + "D": "The blue pencil case on the table is behind you, not on the right side.", + "E": "The blue pencil case on the table is in front of you, positioned straight ahead rather than to the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 20 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_3_T1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the electronic piano relative to me?", + "answer": "The electronic keyboard is to your left.", + "question_chinese": "电子钢琴在相对于我的哪里？", + "answer_chinese": "电子钢琴在相对于你的左边。", + "evidence": [ + { + "description": "The electronic piano is on your left.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The electronic keyboard is to your left.", + "B": "The electronic keyboard is in front of you." + }, + "answer_label": "A", + "answer_text": "The electronic keyboard is to your left.", + "q_id": "scene0710_00-0_3_T1", + "correct_answer": "The electronic keyboard is to your left.", + "original_options": { + "A": "The electronic keyboard is to your left.", + "B": "The electronic keyboard is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0710_00-0", + "video_length": "00:00:20", + "qid": "scene0710_00-0_3_T2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the electronic piano relative to me?", + "answer": "The electronic keyboard is in front of you.", + "question_chinese": "电子钢琴在相对于我的哪里？", + "answer_chinese": "电子钢琴在相对于你的正前方。", + "evidence": [ + { + "description": "The electronic piano is in front of you.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The electronic keyboard is to your left.", + "B": "The electronic keyboard is in front of you." + }, + "answer_label": "B", + "answer_text": "The electronic keyboard is in front of you.", + "q_id": "scene0710_00-0_3_T2", + "correct_answer": "The electronic keyboard is in front of you.", + "original_options": { + "A": "The electronic keyboard is to your left.", + "B": "The electronic keyboard is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + } +] diff --git a/QAs/scene0711_00-0-mcq.json b/QAs/scene0711_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5e3284598ac4d6f79a1db755155ff575681a90af --- /dev/null +++ b/QAs/scene0711_00-0-mcq.json @@ -0,0 +1,787 @@ +[ + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_3", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just throw away the garbage in my position?", + "answer": "Yes,you can.", + "question_chinese": "在我这个位置我可以直接丢垃圾吗？", + "answer_chinese": "你可以。", + "evidence": [ + { + "description": "There are two trash cans on the ground in front of you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "There is a garbage bag on a garbage can.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The trash can is not far from you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "No,you cannot.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "scene0711_00-0_3", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No,you cannot.", + "B": "Yes,you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_4", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just throw away the garbage in my position?", + "answer": "Yes,you can.", + "question_chinese": "在我这个位置我可以直接丢垃圾吗？", + "answer_chinese": "你可以。", + "evidence": [ + { + "description": "There are two trash cans on the ground in front of you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "There is a garbage bag on a garbage can.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The trash can is not far from you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No,you cannot." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "scene0711_00-0_4", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No,you cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_9", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit on the chair and use the computer directly?", + "answer": "No,you can't.", + "question_chinese": "我可以直接坐下使用电脑吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "You are too far away from the chair.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "The computer is turned off.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "scene0711_00-0_9", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes,you can.", + "B": "No,you can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_10", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit on the chair and use the computer directly?", + "answer": "No,you can't.", + "question_chinese": "我可以直接坐下使用电脑吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "You are too far away from the chair.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + }, + { + "description": "The computer is turned off.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "scene0711_00-0_10", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_5", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many whiteboard pens do I have?", + "answer": "You have three whiteboard pens.", + "question_chinese": "我有几支白板笔？", + "answer_chinese": "你有三支白板笔。", + "evidence": [ + { + "description": "There are three whiteboard pens.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "You have two whiteboard pens.", + "B": "You have three whiteboard pens.", + "C": "You have four whiteboard pens.", + "D": "You have one whiteboard pen in your hand and no other whiteboard pens with you.", + "E": "You have five whiteboard pens total, counting the ones you’re holding and the extra ones nearby." + }, + "answer_label": "B", + "answer_text": "You have three whiteboard pens.", + "q_id": "scene0711_00-0_5", + "correct_answer": "You have three whiteboard pens.", + "original_options": { + "A": "You have two whiteboard pens.", + "B": "You have three whiteboard pens.", + "C": "You have four whiteboard pens.", + "D": "You have one whiteboard pen in your hand and no other whiteboard pens with you.", + "E": "You have five whiteboard pens total, counting the ones you’re holding and the extra ones nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 17 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_6", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many whiteboard pens do I have?", + "answer": "You have four whiteboard pens.", + "question_chinese": "我有几支白板笔？", + "answer_chinese": "你有4支白板笔。", + "evidence": [ + { + "description": "There are three whiteboard pens.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "There is one red whiteboard pen on the desk.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "You have three whiteboard pens.", + "B": "You have five whiteboard pens.", + "C": "You have six whiteboard pens in total, including the ones you’re holding and the extra ones nearby.", + "D": "You have two whiteboard pens with you, and there aren’t any additional whiteboard pens within reach.", + "E": "You have four whiteboard pens." + }, + "answer_label": "E", + "answer_text": "You have four whiteboard pens.", + "q_id": "scene0711_00-0_6", + "correct_answer": "You have four whiteboard pens.", + "original_options": { + "A": "You have three whiteboard pens.", + "B": "You have five whiteboard pens.", + "C": "You have six whiteboard pens in total, including the ones you’re holding and the extra ones nearby.", + "D": "You have two whiteboard pens with you, and there aren’t any additional whiteboard pens within reach.", + "E": "You have four whiteboard pens." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_4_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Can I see the table?", + "answer": "You can’t see the table; it is to your back-left.", + "question_chinese": "我能看到桌子吗？", + "answer_chinese": "你不能看到桌子，桌子在你的左后方。", + "evidence": [ + { + "description": "You don't see the table.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t see the table; it is to your back-left.", + "B": "You can see a table; the table is in front of you." + }, + "answer_label": "A", + "answer_text": "You can’t see the table; it is to your back-left.", + "q_id": "scene0711_00-0_4_T1", + "correct_answer": "You can’t see the table; it is to your back-left.", + "original_options": { + "A": "You can’t see the table; it is to your back-left.", + "B": "You can see a table; the table is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_4_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Can I see the table?", + "answer": "You can see a table; the table is in front of you.", + "question_chinese": "我能看到桌子吗？", + "answer_chinese": "你能看到桌子，桌子在你的前面。", + "evidence": [ + { + "description": "The table is in front of you.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t see the table; it is to your back-left.", + "B": "You can see a table; the table is in front of you." + }, + "answer_label": "B", + "answer_text": "You can see a table; the table is in front of you.", + "q_id": "scene0711_00-0_4_T2", + "correct_answer": "You can see a table; the table is in front of you.", + "original_options": { + "A": "You can’t see the table; it is to your back-left.", + "B": "You can see a table; the table is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_5_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the computer relative to me?", + "answer": "The computer is to your front-left.", + "question_chinese": "电脑在相对于我的什么位置？", + "answer_chinese": "电脑在你的左前方。", + "evidence": [ + { + "description": "The computer is on your right front.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It is to your front-right.", + "B": "The computer is to your front-left." + }, + "answer_label": "B", + "answer_text": "The computer is to your front-left.", + "q_id": "scene0711_00-0_5_T1", + "correct_answer": "The computer is to your front-left.", + "original_options": { + "A": "It is to your front-right.", + "B": "The computer is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_5_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Where is the computer relative to me?", + "answer": "It is to your front-right.", + "question_chinese": "电脑在相对于我的什么位置？", + "answer_chinese": "在你的右前方。", + "evidence": [ + { + "description": "Front right of you.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It is to your front-right.", + "B": "The computer is to your front-left." + }, + "answer_label": "A", + "answer_text": "It is to your front-right.", + "q_id": "scene0711_00-0_5_T2", + "correct_answer": "It is to your front-right.", + "original_options": { + "A": "It is to your front-right.", + "B": "The computer is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_2_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Relative to me, where is the whiteboard?", + "answer": "The whiteboard is in front of you.", + "question_chinese": "相对于我的位置，白板在哪里？", + "answer_chinese": "白板在你前面", + "evidence": [ + { + "description": "The whiteboard is in front of you.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The whiteboard is behind you.", + "B": "The whiteboard is in front of you." + }, + "answer_label": "B", + "answer_text": "The whiteboard is in front of you.", + "q_id": "scene0711_00-0_2_T2", + "correct_answer": "The whiteboard is in front of you.", + "original_options": { + "A": "The whiteboard is behind you.", + "B": "The whiteboard is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_2_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Relative to me, where is the whiteboard?", + "answer": "The whiteboard is behind you.", + "question_chinese": "相对于我的位置，白板在哪里？", + "answer_chinese": "白板在你的身后。", + "evidence": [ + { + "description": "The whiteboard is behind you.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The whiteboard is behind you.", + "B": "The whiteboard is in front of you." + }, + "answer_label": "A", + "answer_text": "The whiteboard is behind you.", + "q_id": "scene0711_00-0_2_T1", + "correct_answer": "The whiteboard is behind you.", + "original_options": { + "A": "The whiteboard is behind you.", + "B": "The whiteboard is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_1_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the window relative to me?", + "answer": "The window is in front of you.", + "question_chinese": "窗户在我的哪里？", + "answer_chinese": "窗户在你的前面。", + "evidence": [ + { + "description": "The window is in front of you.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The window is in front of you.", + "B": "The window is behind you." + }, + "answer_label": "A", + "answer_text": "The window is in front of you.", + "q_id": "scene0711_00-0_1_T2", + "correct_answer": "The window is in front of you.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_1_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the window relative to me?", + "answer": "The window is to your right.", + "question_chinese": "窗户在我的哪里？", + "answer_chinese": "窗户在你的右边", + "evidence": [ + { + "description": "The window is on your right.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The window is to your left.", + "B": "The window is to your right." + }, + "answer_label": "B", + "answer_text": "The window is to your right.", + "q_id": "scene0711_00-0_1_T1", + "correct_answer": "The window is to your right.", + "original_options": { + "A": "The window is to your left.", + "B": "The window is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Are the door and window all open?", + "answer": "No,they are both closed.", + "question_chinese": "门窗都是开着的吗？", + "answer_chinese": "不，门窗都是关着的。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The window is closed.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, they are both open.", + "B": "No,they are both closed." + }, + "answer_label": "B", + "answer_text": "No,they are both closed.", + "q_id": "scene0711_00-0_1", + "correct_answer": "No,they are both closed.", + "original_options": { + "A": "Yes, they are both open.", + "B": "No,they are both closed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "scene0711_00-0", + "video_length": "00:00:20", + "qid": "scene0711_00-0_2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Are the door and window all open?", + "answer": "No,they are both closed.", + "question_chinese": "门窗都是开着的吗？", + "answer_chinese": "不，门窗都是关着的。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The window is closed.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,they are both closed.", + "B": "Yes, they are both open." + }, + "answer_label": "A", + "answer_text": "No,they are both closed.", + "q_id": "scene0711_00-0_2", + "correct_answer": "No,they are both closed.", + "original_options": { + "A": "No,they are both closed.", + "B": "Yes, they are both open." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + } +] diff --git a/QAs/scene0712_00-0-mcq.json b/QAs/scene0712_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..47213c670cfae722833406553a866ae35fc5f8fe --- /dev/null +++ b/QAs/scene0712_00-0-mcq.json @@ -0,0 +1,828 @@ +[ + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_1_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are my keys?", + "answer": "It was on the top shelf on the shelf to your right, all the way on the right.", + "question_chinese": "我的钥匙在哪里？", + "answer_chinese": "它在右边架子最上面一层的最右边。", + "evidence": [ + { + "description": "The key is on the far right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The key is on the top layer of the shelf.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "The key is on the shelf on the right.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "On the top shelf in front of you.", + "B": "On the top shelf to your left.", + "C": "It was on the top shelf on the shelf behind you, all the way on the right.", + "D": "It was on the top shelf on the shelf in front of you, all the way on the left.", + "E": "It was on the top shelf on the shelf to your right, all the way on the right." + }, + "answer_label": "E", + "answer_text": "It was on the top shelf on the shelf to your right, all the way on the right.", + "q_id": "scene0712_00-0_1_T1", + "correct_answer": "It was on the top shelf on the shelf to your right, all the way on the right.", + "original_options": { + "A": "On the top shelf in front of you.", + "B": "On the top shelf to your left.", + "C": "It was on the top shelf on the shelf behind you, all the way on the right.", + "D": "It was on the top shelf on the shelf in front of you, all the way on the left.", + "E": "It was on the top shelf on the shelf to your right, all the way on the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 19 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_1_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are my keys?", + "answer": "It was on the top shelf of the shelf in front of you, all the way on the right.", + "question_chinese": "我的钥匙在哪里？", + "answer_chinese": "它在前面架子最上面一层的最右边。", + "evidence": [ + { + "description": "The key is on the far right", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The key is on the topmost layer of the shelf.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "The key is on the shelf on the right.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was on the top shelf of the shelf in front of you, all the way on the right.", + "B": "On the top shelf behind you.", + "C": "On the top shelf to your right.", + "D": "It was on the top shelf of the shelf to your left, all the way on the right.", + "E": "It was on the top shelf of the shelf in front of you, all the way on the left." + }, + "answer_label": "A", + "answer_text": "It was on the top shelf of the shelf in front of you, all the way on the right.", + "q_id": "scene0712_00-0_1_T2", + "correct_answer": "It was on the top shelf of the shelf in front of you, all the way on the right.", + "original_options": { + "A": "It was on the top shelf of the shelf in front of you, all the way on the right.", + "B": "On the top shelf behind you.", + "C": "On the top shelf to your right.", + "D": "It was on the top shelf of the shelf to your left, all the way on the right.", + "E": "It was on the top shelf of the shelf in front of you, all the way on the left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are my keys?", + "answer": "It's on the far right side of the topmost shelf on the right.", + "question_chinese": "我的钥匙在哪里？", + "answer_chinese": "它在右边架子最上面一层的最右边。", + "evidence": [ + { + "description": "The key is on the far right", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The key is on the topmost layer of the shelf.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "The key is on the shelf on the right.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "On the far left of the top shelf.", + "B": "On the right side of the bottom shelf.", + "C": "It's on the far right side of the topmost shelf on the right.", + "D": "It's on the far right side of the bottommost shelf on the right, not the topmost shelf.", + "E": "It's on the far left side of the topmost shelf on the right, rather than on the far right." + }, + "answer_label": "C", + "answer_text": "It's on the far right side of the topmost shelf on the right.", + "q_id": "scene0712_00-0_2", + "correct_answer": "It's on the far right side of the topmost shelf on the right.", + "original_options": { + "A": "On the far left of the top shelf.", + "B": "On the right side of the bottom shelf.", + "C": "It's on the far right side of the topmost shelf on the right.", + "D": "It's on the far right side of the bottommost shelf on the right, not the topmost shelf.", + "E": "It's on the far left side of the topmost shelf on the right, rather than on the far right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 17, + 19 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_2_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What do I see on the shelf to my right?", + "answer": "You saw a key, a helmet, and a magazine.", + "question_chinese": "我看到右边架子上有什么？", + "answer_chinese": "你看到有有一把钥匙、一个头盔、一本杂志。", + "evidence": [ + { + "description": "There is a key, a helmet and a magazine.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw a key and a helmet.", + "B": "You saw a key, a helmet, and a magazine.", + "C": "You saw a helmet and a magazine.", + "D": "You saw a key, a helmet, a cup, and a magazine on the shelf to your right.", + "E": "You saw a key, a helmet, a magazine, and a piece of red fabric on the shelf to your right." + }, + "answer_label": "B", + "answer_text": "You saw a key, a helmet, and a magazine.", + "q_id": "scene0712_00-0_2_T2", + "correct_answer": "You saw a key, a helmet, and a magazine.", + "original_options": { + "A": "You saw a key and a helmet.", + "B": "You saw a key, a helmet, and a magazine.", + "C": "You saw a helmet and a magazine.", + "D": "You saw a key, a helmet, a cup, and a magazine on the shelf to your right.", + "E": "You saw a key, a helmet, a magazine, and a piece of red fabric on the shelf to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 20 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_2_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What do I see on the shelf to my right?", + "answer": "You saw a key, a helmet, a magazine, a cup, and a piece of red fabric.", + "question_chinese": "我看到右边架子上有什么？", + "answer_chinese": "你看到有一把钥匙、一个头盔、一本杂志、一个杯子和一块红色布料。", + "evidence": [ + { + "description": "There is a key, a helmet and a magazine.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "There is a glass and a piece of red cloth.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You saw a key and helmet.", + "B": "You saw a magazine and cup.", + "C": "You saw a key, a helmet, a magazine, a cup, and a piece of red fabric.", + "D": "You saw a key, a helmet, a magazine, and a cup, but there was no red fabric on the shelf.", + "E": "You saw a helmet, a magazine, a cup, and a piece of red fabric, but you didn’t see a key there." + }, + "answer_label": "C", + "answer_text": "You saw a key, a helmet, a magazine, a cup, and a piece of red fabric.", + "q_id": "scene0712_00-0_2_T1", + "correct_answer": "You saw a key, a helmet, a magazine, a cup, and a piece of red fabric.", + "original_options": { + "A": "You saw a key and helmet.", + "B": "You saw a magazine and cup.", + "C": "You saw a key, a helmet, a magazine, a cup, and a piece of red fabric.", + "D": "You saw a key, a helmet, a magazine, and a cup, but there was no red fabric on the shelf.", + "E": "You saw a helmet, a magazine, a cup, and a piece of red fabric, but you didn’t see a key there." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 20, + 22 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_3_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Do I see that the door is open?", + "answer": "You saw that the door wasn’t open.", + "question_chinese": "我看到门是打开的吗？", + "answer_chinese": "你看到门不是打开的。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You saw that the door wasn’t open.", + "B": "Yes—you saw that the door was open." + }, + "answer_label": "A", + "answer_text": "You saw that the door wasn’t open.", + "q_id": "scene0712_00-0_3_T1", + "correct_answer": "You saw that the door wasn’t open.", + "original_options": { + "A": "You saw that the door wasn’t open.", + "B": "Yes—you saw that the door was open." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_3_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Do I see that the door is open?", + "answer": "No—you never opened it at any point.", + "question_chinese": "我看到门是打开的吗？", + "answer_chinese": "不是，从始至终你没有打开过它。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—you opened it at some point.", + "B": "No—you never opened it at any point." + }, + "answer_label": "B", + "answer_text": "No—you never opened it at any point.", + "q_id": "scene0712_00-0_3_T2", + "correct_answer": "No—you never opened it at any point.", + "original_options": { + "A": "Yes—you opened it at some point.", + "B": "No—you never opened it at any point." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_9", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I turn on the light directly?", + "answer": "No,you can't.", + "question_chinese": "我能直接打开灯吗？", + "answer_chinese": "你不能。", + "evidence": [ + { + "description": "The switch is at the door, which is too far from you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "scene0712_00-0_9", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_5_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I turn on the light directly right now?", + "answer": "Yes. It was about an arm’s length away from you.", + "question_chinese": "我能直接打开灯吗？", + "answer_chinese": "你可以，离你一臂的距离", + "evidence": [ + { + "description": "The switch is at the door, which is too far from you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. It was too far from you to reach directly.", + "B": "Yes. It was about an arm’s length away from you." + }, + "answer_label": "B", + "answer_text": "Yes. It was about an arm’s length away from you.", + "q_id": "scene0712_00-0_5_T2", + "correct_answer": "Yes. It was about an arm’s length away from you.", + "original_options": { + "A": "No. It was too far from you to reach directly.", + "B": "Yes. It was about an arm’s length away from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_10", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I turn on the light directly?", + "answer": "No,you can't.", + "question_chinese": "我能直接打开灯吗？", + "answer_chinese": "你不能。", + "evidence": [ + { + "description": "The switch is at the door, which is too far from you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "scene0712_00-0_10", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_5_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I turn on the light directly right now?", + "answer": "You can’t.", + "question_chinese": "我能直接打开灯吗？", + "answer_chinese": "你不能。", + "evidence": [ + { + "description": "The switch is at the door, which is too far from you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t.", + "B": "Yes. It was about an arm’s length away from you." + }, + "answer_label": "A", + "answer_text": "You can’t.", + "q_id": "scene0712_00-0_5_T1", + "correct_answer": "You can’t.", + "original_options": { + "A": "You can’t.", + "B": "Yes. It was about an arm’s length away from you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_4_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the keys or the cup?", + "answer": "The key was closer to you.", + "question_chinese": "钥匙和水杯哪个离我更近？", + "answer_chinese": "钥匙离你更近。", + "evidence": [ + { + "description": "The key is on the right side of the shelf.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The key was closer to you.", + "B": "The cup was closer to you." + }, + "answer_label": "A", + "answer_text": "The key was closer to you.", + "q_id": "scene0712_00-0_4_T1", + "correct_answer": "The key was closer to you.", + "original_options": { + "A": "The key was closer to you.", + "B": "The cup was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_7", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the key or the water cup?", + "answer": "The water cup is closer to you.", + "question_chinese": "钥匙和水杯哪个离我更近？", + "answer_chinese": "水杯离你更近。", + "evidence": [ + { + "description": "The key is on the right side of the shelf.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The water cup is on the left side of the shelf.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:00" + } + }, + { + "description": "You are on the left side of the shelf on the right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The water cup is closer to you.", + "B": "The key is closer to you." + }, + "answer_label": "A", + "answer_text": "The water cup is closer to you.", + "q_id": "scene0712_00-0_7", + "correct_answer": "The water cup is closer to you.", + "original_options": { + "A": "The water cup is closer to you.", + "B": "The key is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_4_T2", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the keys or the cup?", + "answer": "The cup was closer to you.", + "question_chinese": "钥匙和水杯哪个离我更近？", + "answer_chinese": "水杯离你更近。", + "evidence": [ + { + "description": "The key is on the right side of the shelf.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The water cup is on the left side of the shelf.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "You are on the left side of the shelf on the right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The key was closer to you.", + "B": "The cup was closer to you." + }, + "answer_label": "B", + "answer_text": "The cup was closer to you.", + "q_id": "scene0712_00-0_4_T2", + "correct_answer": "The cup was closer to you.", + "original_options": { + "A": "The key was closer to you.", + "B": "The cup was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_5", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the door open?", + "answer": "No,it isn't.", + "question_chinese": "门是打开的吗？", + "answer_chinese": "不是。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,it isn't.", + "B": "Yes, it is." + }, + "answer_label": "A", + "answer_text": "No,it isn't.", + "q_id": "scene0712_00-0_5", + "correct_answer": "No,it isn't.", + "original_options": { + "A": "No,it isn't.", + "B": "Yes, it is." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0712_00-0", + "video_length": "00:00:20", + "qid": "scene0712_00-0_6", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the door open?", + "answer": "No,it isn't.", + "question_chinese": "门是打开的吗？", + "answer_chinese": "不是。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, it is.", + "B": "No,it isn't." + }, + "answer_label": "B", + "answer_text": "No,it isn't.", + "q_id": "scene0712_00-0_6", + "correct_answer": "No,it isn't.", + "original_options": { + "A": "Yes, it is.", + "B": "No,it isn't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + } +] diff --git a/QAs/scene0713_00-0-mcq.json b/QAs/scene0713_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a051a906f1ede70457f2532fa242ae49d292424d --- /dev/null +++ b/QAs/scene0713_00-0-mcq.json @@ -0,0 +1,724 @@ +[ + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there any paper left in the toilet?", + "answer": "Yes,there is.", + "question_chinese": "厕所还有纸吗？", + "answer_chinese": "有。", + "evidence": [ + { + "description": "A roll of paper is hung on the wall.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "There is a roll of toilet paper on the floor beneath the toilet.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, there is not.", + "B": "Yes,there is." + }, + "answer_label": "B", + "answer_text": "Yes,there is.", + "q_id": "scene0713_00-0_3", + "correct_answer": "Yes,there is.", + "original_options": { + "A": "No, there is not.", + "B": "Yes,there is." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there any paper left in the toilet?", + "answer": "Yes,there is.", + "question_chinese": "厕所还有纸吗？", + "answer_chinese": "有。", + "evidence": [ + { + "description": "A roll of paper is hung on the wall.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "There is a roll of toilet paper on the floor beneath the toilet.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,there is.", + "B": "No, there is not." + }, + "answer_label": "A", + "answer_text": "Yes,there is.", + "q_id": "scene0713_00-0_4", + "correct_answer": "Yes,there is.", + "original_options": { + "A": "Yes,there is.", + "B": "No, there is not." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_8", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can?", + "answer": "In front of you.", + "question_chinese": "垃圾桶在哪里？", + "answer_chinese": "在你面前。", + "evidence": [ + { + "description": "The trash can is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "On the floor near the door, a few steps away from where you're standing.", + "D": "In front of you.", + "E": "Under the desk by the table, tucked away so you might miss it at first." + }, + "answer_label": "D", + "answer_text": "In front of you.", + "q_id": "scene0713_00-0_8", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "On the floor near the door, a few steps away from where you're standing.", + "D": "In front of you.", + "E": "Under the desk by the table, tucked away so you might miss it at first." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 14, + 15 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_7", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can?", + "answer": "Behind you.", + "question_chinese": "垃圾桶在哪里？", + "answer_chinese": "在你身后。", + "evidence": [ + { + "description": "The trash can is opposite the toilet.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You are facing the toilet.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your right.", + "C": "It was under the table, close to your feet, not behind you.", + "D": "It was off to your left side, not behind your back where you were standing.", + "E": "Behind you." + }, + "answer_label": "E", + "answer_text": "Behind you.", + "q_id": "scene0713_00-0_7", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your right.", + "C": "It was under the table, close to your feet, not behind you.", + "D": "It was off to your left side, not behind your back where you were standing.", + "E": "Behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 15 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_1_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "In which direction is the trash can relative to me?", + "answer": "The trash can was in front of you.", + "question_chinese": "垃圾桶在我的哪个方向？", + "answer_chinese": "垃圾桶在你前面。", + "evidence": [ + { + "description": "The trash can is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The trash can was behind you.", + "B": "The trash can was in front of you." + }, + "answer_label": "B", + "answer_text": "The trash can was in front of you.", + "q_id": "scene0713_00-0_1_T2", + "correct_answer": "The trash can was in front of you.", + "original_options": { + "A": "The trash can was behind you.", + "B": "The trash can was in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_1_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "In which direction is the trash can relative to me?", + "answer": "The trash can was to your right.", + "question_chinese": "垃圾桶在我的哪个方向？", + "answer_chinese": "垃圾桶在你右边。", + "evidence": [ + { + "description": "The trash can is on your right.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The trash can was to your left.", + "B": "The trash can was to your right." + }, + "answer_label": "B", + "answer_text": "The trash can was to your right.", + "q_id": "scene0713_00-0_1_T1", + "correct_answer": "The trash can was to your right.", + "original_options": { + "A": "The trash can was to your left.", + "B": "The trash can was to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_6", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory in the bathroom?", + "answer": "First, walk clockwise towards the left side of the toilet, then return to the original position, walk towards the left side of the toilet again, and return to the original position.", + "question_chinese": "我在卫生间的移动轨迹是什么样的？", + "answer_chinese": "你先顺时针向马桶左侧走去，再回到原位置，再走向马桶左侧，再回到原位置。", + "evidence": [ + { + "description": "First, walk clockwise towards the left side of the toilet.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + }, + { + "description": "Return to the original position.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + }, + { + "description": "Walk towards the left side of the toilet again", + "timestamps": { + "start": "00:00:12", + "end": "00:00:14" + } + }, + { + "description": "Return to the original position.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Walk counterclockwise to the toilet’s right.", + "B": "Walk clockwise to the sink.", + "C": "First, walk clockwise towards the left side of the toilet, then return to the original position, walk towards the left side of the toilet again, and return to the original position.", + "D": "First, walk counterclockwise toward the right side of the toilet, return, then repeat the same move once more.", + "E": "First, walk clockwise toward the left side of the bathtub, return to start, then go there again and return." + }, + "answer_label": "C", + "answer_text": "First, walk clockwise towards the left side of the toilet, then return to the original position, walk towards the left side of the toilet again, and return to the original position.", + "q_id": "scene0713_00-0_6", + "correct_answer": "First, walk clockwise towards the left side of the toilet, then return to the original position, walk towards the left side of the toilet again, and return to the original position.", + "original_options": { + "A": "Walk counterclockwise to the toilet’s right.", + "B": "Walk clockwise to the sink.", + "C": "First, walk clockwise towards the left side of the toilet, then return to the original position, walk towards the left side of the toilet again, and return to the original position.", + "D": "First, walk counterclockwise toward the right side of the toilet, return, then repeat the same move once more.", + "E": "First, walk clockwise toward the left side of the bathtub, return to start, then go there again and return." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 18, + 19 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_5", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory in the bathroom?", + "answer": "First, walk clockwise towards the left side of the toilet, then return to the original position, walk towards the left side of the toilet again, return to the original position, and then walk towards the sink.", + "question_chinese": "我在卫生间的移动轨迹是什么样的？", + "answer_chinese": "你先顺时针向马桶左侧走去，再回到原位置，再走向马桶左侧，再回到原位置，再走向洗手池。", + "evidence": [ + { + "description": "First, walk clockwise towards the left side of the toilet.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + }, + { + "description": "Return to the original position.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + }, + { + "description": "Walk towards the left side of the toilet again.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:14" + } + }, + { + "description": "Return to the original position.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "Walk towards the sink.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "End by walking to the door.", + "B": "Go to the toilet’s right side.", + "C": "First, walk clockwise to the left side of the toilet, return, then go to its right side and return.", + "D": "First, walk counterclockwise toward the right side of the toilet twice with returns, then walk toward the sink.", + "E": "First, walk clockwise towards the left side of the toilet, then return to the original position, walk towards the left side of the toilet again, return to the original position, and then walk towards the sink." + }, + "answer_label": "E", + "answer_text": "First, walk clockwise towards the left side of the toilet, then return to the original position, walk towards the left side of the toilet again, return to the original position, and then walk towards the sink.", + "q_id": "scene0713_00-0_5", + "correct_answer": "First, walk clockwise towards the left side of the toilet, then return to the original position, walk towards the left side of the toilet again, return to the original position, and then walk towards the sink.", + "original_options": { + "A": "End by walking to the door.", + "B": "Go to the toilet’s right side.", + "C": "First, walk clockwise to the left side of the toilet, return, then go to its right side and return.", + "D": "First, walk counterclockwise toward the right side of the toilet twice with returns, then walk toward the sink.", + "E": "First, walk clockwise towards the left side of the toilet, then return to the original position, walk towards the left side of the toilet again, return to the original position, and then walk towards the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 18 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_10", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the paper hanging on the wall?", + "answer": "No,you can't.", + "question_chinese": "我能拿到挂在墙上的纸吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The paper is too far away from you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "scene0713_00-0_10", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes,you can.", + "B": "No,you can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_9", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the paper hanging on the wall?", + "answer": "Yes,you can.", + "question_chinese": "我能拿到挂在墙上的纸吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The paper is very close to you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "scene0713_00-0_9", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_5_T1", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the paper hanging on the wall?", + "answer": "You could reach the paper hanging on the wall; it was to your right.", + "question_chinese": "我能拿到挂在墙上的纸吗？", + "answer_chinese": "你能拿到挂在墙上的纸，它在你右边。", + "evidence": [ + { + "description": "You can reach the paper on the wall.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn’t reach the paper hanging on the wall; it was to your left.", + "B": "You could reach the paper hanging on the wall; it was to your right." + }, + "answer_label": "B", + "answer_text": "You could reach the paper hanging on the wall; it was to your right.", + "q_id": "scene0713_00-0_5_T1", + "correct_answer": "You could reach the paper hanging on the wall; it was to your right.", + "original_options": { + "A": "You couldn’t reach the paper hanging on the wall; it was to your left.", + "B": "You could reach the paper hanging on the wall; it was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_5_T2", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the paper hanging on the wall?", + "answer": "You couldn’t reach the paper hanging on the wall; it was to your left.", + "question_chinese": "我能拿到挂在墙上的纸吗？", + "answer_chinese": "你不能拿到挂在墙上的纸，它在你左边。", + "evidence": [ + { + "description": "The paper is very close to you.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn’t reach the paper hanging on the wall; it was to your left.", + "B": "You could reach the paper hanging on the wall; it was to your left." + }, + "answer_label": "A", + "answer_text": "You couldn’t reach the paper hanging on the wall; it was to your left.", + "q_id": "scene0713_00-0_5_T2", + "correct_answer": "You couldn’t reach the paper hanging on the wall; it was to your left.", + "original_options": { + "A": "You couldn’t reach the paper hanging on the wall; it was to your left.", + "B": "You could reach the paper hanging on the wall; it was to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_2_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "In which direction is the sink relative to me?", + "answer": "The sink was to your left.", + "question_chinese": "洗手池在我哪个方向？", + "answer_chinese": "洗手池在你左边", + "evidence": [ + { + "description": "The sink is on your left.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sink was to your left.", + "B": "The sink was to your right." + }, + "answer_label": "A", + "answer_text": "The sink was to your left.", + "q_id": "scene0713_00-0_2_T1", + "correct_answer": "The sink was to your left.", + "original_options": { + "A": "The sink was to your left.", + "B": "The sink was to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0713_00-0", + "video_length": "00:00:20", + "qid": "scene0713_00-0_2_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "In which direction is the sink relative to me?", + "answer": "The sink was in front of you.", + "question_chinese": "洗手池在我哪个方向？", + "answer_chinese": "洗手池在你前面", + "evidence": [ + { + "description": "The sink is in front of you.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sink was to your right.", + "B": "The sink was in front of you." + }, + "answer_label": "B", + "answer_text": "The sink was in front of you.", + "q_id": "scene0713_00-0_2_T2", + "correct_answer": "The sink was in front of you.", + "original_options": { + "A": "The sink was to your right.", + "B": "The sink was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/scene0715_00-0-mcq.json b/QAs/scene0715_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9f5830f8cd4f3b84eefe42517cb1221a5ed5d92a --- /dev/null +++ b/QAs/scene0715_00-0-mcq.json @@ -0,0 +1,316 @@ +[ + { + "video_uid": "scene0715_00-0", + "video_length": "00:00:20", + "qid": "scene0715_00-0_10", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the vacuum cleaner?", + "answer": "It's on your left front.", + "question_chinese": "吸尘器在哪里？", + "answer_chinese": "在你左前方。", + "evidence": [ + { + "description": "The vacuum cleaner is beside the sofa.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "You are in front of the sofa.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It's on your right front.", + "B": "It's on your left rear.", + "C": "It's on your left front.", + "D": "It's directly in front of you, slightly to the right, near the wall.", + "E": "It's to your right side, a little behind you, along the edge of the room." + }, + "answer_label": "C", + "answer_text": "It's on your left front.", + "q_id": "scene0715_00-0_10", + "correct_answer": "It's on your left front.", + "original_options": { + "A": "It's on your right front.", + "B": "It's on your left rear.", + "C": "It's on your left front.", + "D": "It's directly in front of you, slightly to the right, near the wall.", + "E": "It's to your right side, a little behind you, along the edge of the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 15 + ] + }, + { + "video_uid": "scene0715_00-0", + "video_length": "00:00:20", + "qid": "scene0715_00-0_9", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the vacuum cleaner?", + "answer": "It's behind you.", + "question_chinese": "吸尘器在哪里？", + "answer_chinese": "在你身后。", + "evidence": [ + { + "description": "The vacuum cleaner is beside the sofa.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "You are in front of the sofa with your back facing it.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "It's behind you.", + "B": "It's in front of you.", + "C": "It's to your left.", + "D": "It's on your right side, slightly ahead of you, near where you are standing.", + "E": "It's to your left front, positioned near the wall rather than behind you." + }, + "answer_label": "A", + "answer_text": "It's behind you.", + "q_id": "scene0715_00-0_9", + "correct_answer": "It's behind you.", + "original_options": { + "A": "It's behind you.", + "B": "It's in front of you.", + "C": "It's to your left.", + "D": "It's on your right side, slightly ahead of you, near where you are standing.", + "E": "It's to your left front, positioned near the wall rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 13 + ] + }, + { + "video_uid": "scene0715_00-0", + "video_length": "00:00:20", + "qid": "scene0715_00-0_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sofas are there?", + "answer": "Three.", + "question_chinese": "有几个沙发？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "There are three sofas.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "There are five sofas in total in the room, which is more than three.", + "D": "Three.", + "E": "You can see only one sofa, with no others nearby in the room." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "scene0715_00-0_1", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "There are five sofas in total in the room, which is more than three.", + "D": "Three.", + "E": "You can see only one sofa, with no others nearby in the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 13 + ] + }, + { + "video_uid": "scene0715_00-0", + "video_length": "00:00:20", + "qid": "scene0715_00-0_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sofas are there?", + "answer": "Three.", + "question_chinese": "有几个沙发？", + "answer_chinese": "三个。", + "evidence": [ + { + "description": "There are three sofas.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "There are five sofas in the space altogether, which is more than three.", + "E": "You can count only one sofa, with no additional sofas visible anywhere nearby." + }, + "answer_label": "B", + "answer_text": "Three.", + "q_id": "scene0715_00-0_2", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "Four.", + "D": "There are five sofas in the space altogether, which is more than three.", + "E": "You can count only one sofa, with no additional sofas visible anywhere nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 13 + ] + }, + { + "video_uid": "scene0715_00-0", + "video_length": "00:00:20", + "qid": "scene0715_00-0_3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the door closed?", + "answer": "Yes,it is.", + "question_chinese": "门是关着的吗？", + "answer_chinese": "是的。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,it is not.", + "B": "Yes,it is." + }, + "answer_label": "B", + "answer_text": "Yes,it is.", + "q_id": "scene0715_00-0_3", + "correct_answer": "Yes,it is.", + "original_options": { + "A": "No,it is not.", + "B": "Yes,it is." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "scene0715_00-0", + "video_length": "00:00:20", + "qid": "scene0715_00-0_4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the door closed?", + "answer": "Yes,it is.", + "question_chinese": "门是关着的吗？", + "answer_chinese": "是的。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,it is.", + "B": "No, it is not." + }, + "answer_label": "A", + "answer_text": "Yes,it is.", + "q_id": "scene0715_00-0_4", + "correct_answer": "Yes,it is.", + "original_options": { + "A": "Yes,it is.", + "B": "No, it is not." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + } +] diff --git a/QAs/scene0716_00-0-mcq.json b/QAs/scene0716_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d67824ab5a2245408c746b2fa4c81aae049d9f1f --- /dev/null +++ b/QAs/scene0716_00-0-mcq.json @@ -0,0 +1,587 @@ +[ + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_9", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, washing machine number 7 or washing machine number 11?", + "answer": "Washing machine number eleven is closer to you.", + "question_chinese": "7号洗衣机和11号洗衣机哪个离我更近？", + "answer_chinese": "11号洗衣机离你更近。", + "evidence": [ + { + "description": "Washing machine number 7 is located on the top floor, on the far left.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "Washing machine number 11 is the third one on the upper floor", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Washing machine number seven is closer to you.", + "B": "Washing machine number eleven is closer to you." + }, + "answer_label": "B", + "answer_text": "Washing machine number eleven is closer to you.", + "q_id": "scene0716_00-0_9", + "correct_answer": "Washing machine number eleven is closer to you.", + "original_options": { + "A": "Washing machine number seven is closer to you.", + "B": "Washing machine number eleven is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_10", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, washing machine number 7 or washing machine number 11?", + "answer": "Washing machine number eleven is closer to you.", + "question_chinese": "7号洗衣机和11号洗衣机哪个离我更近？", + "answer_chinese": "11号洗衣机离你更近。", + "evidence": [ + { + "description": "Washing machine number 7 is on the top floor, on the far left.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + }, + { + "description": "Washing machine number 11 is the third one on the upper floor", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Washing machine number eleven is closer to you.", + "B": "Washing machine number seven is closer to you." + }, + "answer_label": "A", + "answer_text": "Washing machine number eleven is closer to you.", + "q_id": "scene0716_00-0_10", + "correct_answer": "Washing machine number eleven is closer to you.", + "original_options": { + "A": "Washing machine number eleven is closer to you.", + "B": "Washing machine number seven is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Which washing machine is currently running?", + "answer": "Washing machines No. ten and No. 11 are currently in operation.", + "question_chinese": "几号洗衣机正在运行？", + "answer_chinese": "10号和11号洗衣机正在运行。", + "evidence": [ + { + "description": "Washing machines No. 10 and No. 11 are currently in operation.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Only No. 10 is running.", + "B": "Only No. 11 is running.", + "C": "Washing machines No. ten and No. 11 are currently in operation.", + "D": "Washing machines No. nine and No. 11 are currently running right now.", + "E": "Washing machines No. ten and No. 12 are currently running right now." + }, + "answer_label": "C", + "answer_text": "Washing machines No. ten and No. 11 are currently in operation.", + "q_id": "scene0716_00-0_3", + "correct_answer": "Washing machines No. ten and No. 11 are currently in operation.", + "original_options": { + "A": "Only No. 10 is running.", + "B": "Only No. 11 is running.", + "C": "Washing machines No. ten and No. 11 are currently in operation.", + "D": "Washing machines No. nine and No. 11 are currently running right now.", + "E": "Washing machines No. ten and No. 12 are currently running right now." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 12 + ] + }, + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Which washing machine is currently running?", + "answer": "Washing machines No. ten 、 No. 11、 No. 16 and No. 18 are currently in operation.", + "question_chinese": "几号洗衣机正在运行？", + "answer_chinese": "10号、11号16号和18号洗衣机正在运行。", + "evidence": [ + { + "description": "Washing machines No. 10 and No. 11 are currently in operation.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "Washing machines No. 16 and No. 18 are currently in operation.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "No. 9, 11, 16, and 18.", + "B": "No. 10, 11, 15, and 18.", + "C": "Washing machines No. twelve, No. 11, No. 16 and No. 18 are currently in operation.", + "D": "Washing machines No. ten 、 No. 11、 No. 16 and No. 18 are currently in operation.", + "E": "Washing machines No. ten, No. 11, No. 16 and No. 17 are currently in operation." + }, + "answer_label": "D", + "answer_text": "Washing machines No. ten 、 No. 11、 No. 16 and No. 18 are currently in operation.", + "q_id": "scene0716_00-0_4", + "correct_answer": "Washing machines No. ten 、 No. 11、 No. 16 and No. 18 are currently in operation.", + "original_options": { + "A": "No. 9, 11, 16, and 18.", + "B": "No. 10, 11, 15, and 18.", + "C": "Washing machines No. twelve, No. 11, No. 16 and No. 18 are currently in operation.", + "D": "Washing machines No. ten 、 No. 11、 No. 16 and No. 18 are currently in operation.", + "E": "Washing machines No. ten, No. 11, No. 16 and No. 17 are currently in operation." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_1_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Can I reach the iron right now?", + "answer": "You can reach the iron right now, and the iron is to your left.", + "question_chinese": "我现在可以拿到熨斗吗？", + "answer_chinese": "你现在可以拿到熨斗，熨斗在你的左边。", + "evidence": [ + { + "description": "You can get the iron now. The iron is on your left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the iron right now, and the iron is behind you.", + "B": "You can reach the iron right now, and the iron is to your left." + }, + "answer_label": "B", + "answer_text": "You can reach the iron right now, and the iron is to your left.", + "q_id": "scene0716_00-0_1_T2", + "correct_answer": "You can reach the iron right now, and the iron is to your left.", + "original_options": { + "A": "You can’t reach the iron right now, and the iron is behind you.", + "B": "You can reach the iron right now, and the iron is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_1_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Can I reach the iron right now?", + "answer": "You can’t reach the iron right now, and the iron is behind you.", + "question_chinese": "我现在可以拿到熨斗吗？", + "answer_chinese": "你现在不可以拿到熨斗，熨斗在你的后面。", + "evidence": [ + { + "description": "You can't get the iron now, the iron is behind you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the iron right now, and the iron is behind you.", + "B": "You can reach the iron right now, and the iron is to your left." + }, + "answer_label": "A", + "answer_text": "You can’t reach the iron right now, and the iron is behind you.", + "q_id": "scene0716_00-0_1_T1", + "correct_answer": "You can’t reach the iron right now, and the iron is behind you.", + "original_options": { + "A": "You can’t reach the iron right now, and the iron is behind you.", + "B": "You can reach the iron right now, and the iron is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_5_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the washbasin relative to me?", + "answer": "The sink is directly in front of you.", + "question_chinese": "洗物池在相对于我的哪里？", + "answer_chinese": "洗物池在相对于你的正前方。", + "evidence": [ + { + "description": "The sink is right in front of you.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sink is directly in front of you.", + "B": "The sink is behind you." + }, + "answer_label": "A", + "answer_text": "The sink is directly in front of you.", + "q_id": "scene0716_00-0_5_T1", + "correct_answer": "The sink is directly in front of you.", + "original_options": { + "A": "The sink is directly in front of you.", + "B": "The sink is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_5_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Where is the washbasin relative to me?", + "answer": "The sink is behind you.", + "question_chinese": "洗物池在相对于我的哪里？", + "answer_chinese": "洗物池在相对于你的后面。", + "evidence": [ + { + "description": "The sink is behind you.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The sink is directly in front of you.", + "B": "The sink is behind you." + }, + "answer_label": "B", + "answer_text": "The sink is behind you.", + "q_id": "scene0716_00-0_5_T2", + "correct_answer": "The sink is behind you.", + "original_options": { + "A": "The sink is directly in front of you.", + "B": "The sink is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_5", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the electric iron?", + "answer": "It is on the ironing board in front of you.", + "question_chinese": "电熨斗在哪里？", + "answer_chinese": "在你面前的熨衣板上。", + "evidence": [ + { + "description": "It is on the ironing board in front of you.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the board beside you.", + "B": "It is on the ironing board in front of you.", + "C": "On the ironing board behind you.", + "D": "It is on the ironing board to your right, not directly in front of you.", + "E": "It is resting on the ironing board to your left, slightly off to the side." + }, + "answer_label": "B", + "answer_text": "It is on the ironing board in front of you.", + "q_id": "scene0716_00-0_5", + "correct_answer": "It is on the ironing board in front of you.", + "original_options": { + "A": "On the board beside you.", + "B": "It is on the ironing board in front of you.", + "C": "On the ironing board behind you.", + "D": "It is on the ironing board to your right, not directly in front of you.", + "E": "It is resting on the ironing board to your left, slightly off to the side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 15 + ] + }, + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_6", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the electric iron?", + "answer": "It is on the ironing board behind you.", + "question_chinese": "电熨斗在哪里？", + "answer_chinese": "在你后面的熨衣板上。", + "evidence": [ + { + "description": "It is on the ironing board behind you.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the ironing board in front.", + "B": "On the ironing board to your right.", + "C": "It is on the chair next to the ironing board, not on the board itself.", + "D": "It is on a small table beside the ironing board, rather than on the board behind you.", + "E": "It is on the ironing board behind you." + }, + "answer_label": "E", + "answer_text": "It is on the ironing board behind you.", + "q_id": "scene0716_00-0_6", + "correct_answer": "It is on the ironing board behind you.", + "original_options": { + "A": "On the ironing board in front.", + "B": "On the ironing board to your right.", + "C": "It is on the chair next to the ironing board, not on the board itself.", + "D": "It is on a small table beside the ironing board, rather than on the board behind you.", + "E": "It is on the ironing board behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 17 + ] + }, + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_4_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the silver triangular table relative to me?", + "answer": "The silver triangular table is to your left.", + "question_chinese": "银色的三角桌在相对于我的那里？", + "answer_chinese": "银色的三角桌在相对于你的左边。", + "evidence": [ + { + "description": "The silver triangle table is on the left relative to you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The silver triangular table is to your left.", + "B": "The silver triangular table is behind you." + }, + "answer_label": "A", + "answer_text": "The silver triangular table is to your left.", + "q_id": "scene0716_00-0_4_T2", + "correct_answer": "The silver triangular table is to your left.", + "original_options": { + "A": "The silver triangular table is to your left.", + "B": "The silver triangular table is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0716_00-0", + "video_length": "00:00:20", + "qid": "scene0716_00-0_4_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the silver triangular table relative to me?", + "answer": "The silver triangular table is behind you.", + "question_chinese": "银色的三角桌在相对于我的那里？", + "answer_chinese": "银色的三角桌在相对于你的后面。", + "evidence": [ + { + "description": "The silver triangle table is behind you.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The silver triangular table is to your left.", + "B": "The silver triangular table is behind you." + }, + "answer_label": "B", + "answer_text": "The silver triangular table is behind you.", + "q_id": "scene0716_00-0_4_T1", + "correct_answer": "The silver triangular table is behind you.", + "original_options": { + "A": "The silver triangular table is to your left.", + "B": "The silver triangular table is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + } +] diff --git a/QAs/scene0717_00-0-mcq.json b/QAs/scene0717_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..093e43e194652cc5c7274bbd96608059a89743c1 --- /dev/null +++ b/QAs/scene0717_00-0-mcq.json @@ -0,0 +1,358 @@ +[ + { + "video_uid": "scene0717_00-0", + "video_length": "00:00:20", + "qid": "scene0717_00-0_1_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hats do I have?", + "answer": "Three, behind you.", + "question_chinese": "我有几个帽子？", + "answer_chinese": "3个,在你的后方。", + "evidence": [ + { + "description": "Two hats are hung on the bed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "A hat is beside the table.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three, behind you.", + "B": "Two, behind you.", + "C": "Four, behind you.", + "D": "One, behind you, but it’s easy to miss because it’s partially hidden from your view.", + "E": "Five, behind you, spread out behind you rather than stacked together in one place." + }, + "answer_label": "A", + "answer_text": "Three, behind you.", + "q_id": "scene0717_00-0_1_T2", + "correct_answer": "Three, behind you.", + "original_options": { + "A": "Three, behind you.", + "B": "Two, behind you.", + "C": "Four, behind you.", + "D": "One, behind you, but it’s easy to miss because it’s partially hidden from your view.", + "E": "Five, behind you, spread out behind you rather than stacked together in one place." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 14 + ] + }, + { + "video_uid": "scene0717_00-0", + "video_length": "00:00:20", + "qid": "scene0717_00-0_1_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many hats do I have?", + "answer": "Three, split between to your right and behind you.", + "question_chinese": "我有几个帽子？", + "answer_chinese": "3个，分别在你的右方和后方。", + "evidence": [ + { + "description": "Two hats are hung on the bed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "A hat is beside the table.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, split right and behind.", + "B": "Four, split right and behind.", + "C": "One hat total, placed to your right, with none of the hats located behind you.", + "D": "Five hats total, with most to your right and the rest behind you.", + "E": "Three, split between to your right and behind you." + }, + "answer_label": "E", + "answer_text": "Three, split between to your right and behind you.", + "q_id": "scene0717_00-0_1_T1", + "correct_answer": "Three, split between to your right and behind you.", + "original_options": { + "A": "Two, split right and behind.", + "B": "Four, split right and behind.", + "C": "One hat total, placed to your right, with none of the hats located behind you.", + "D": "Five hats total, with most to your right and the rest behind you.", + "E": "Three, split between to your right and behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 13 + ] + }, + { + "video_uid": "scene0717_00-0", + "video_length": "00:00:20", + "qid": "scene0717_00-0_2_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What did the layout of the two beds look like when I saw them?", + "answer": "One bed is by the window and placed horizontally; the other bed is opposite it and placed vertically. One is to your left, and the other is to your right.", + "question_chinese": "我看到两张床的布局是什么样的？", + "answer_chinese": "一张床靠窗横着，另一张床在它对面竖着，一张在左边，另一张在右边。", + "evidence": [ + { + "description": "Tnis one bed is positioned horizontally by the window.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "This one bed is positioned vertically opposite the first one.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "One bed is by the window and placed horizontally; the other bed is opposite it and placed vertically. One is to your left, and the other is to your right.", + "B": "Both beds were to your left.", + "C": "Both beds were to your right.", + "D": "The bed by the window is vertical, and the opposite bed is horizontal, one on each side.", + "E": "One bed is by the window horizontally, but the other is next to it vertically, with one left and one right." + }, + "answer_label": "A", + "answer_text": "One bed is by the window and placed horizontally; the other bed is opposite it and placed vertically. One is to your left, and the other is to your right.", + "q_id": "scene0717_00-0_2_T1", + "correct_answer": "One bed is by the window and placed horizontally; the other bed is opposite it and placed vertically. One is to your left, and the other is to your right.", + "original_options": { + "A": "One bed is by the window and placed horizontally; the other bed is opposite it and placed vertically. One is to your left, and the other is to your right.", + "B": "Both beds were to your left.", + "C": "Both beds were to your right.", + "D": "The bed by the window is vertical, and the opposite bed is horizontal, one on each side.", + "E": "One bed is by the window horizontally, but the other is next to it vertically, with one left and one right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 21 + ] + }, + { + "video_uid": "scene0717_00-0", + "video_length": "00:00:20", + "qid": "scene0717_00-0_2_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What did the layout of the two beds look like when I saw them?", + "answer": "One bed is by the window and placed horizontally; the other bed is opposite it and placed vertically. Both are to your right.", + "question_chinese": "我看到两张床的布局是什么样的？", + "answer_chinese": "一张床靠窗横着，另一张床在它对面竖着，都在你的右方。", + "evidence": [ + { + "description": "This one bed is positioned vertically opposite the first one.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "This one bed is positioned horizontally by the window", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Both beds are to your left.", + "B": "One bed is by the window and placed horizontally; the other bed is opposite it and placed vertically. Both are to your right.", + "C": "The beds are side-by-side, not opposite.", + "D": "One bed is by the window and placed vertically; the other is opposite it and placed horizontally, both to your right.", + "E": "One bed is away from the window and placed horizontally; the other is opposite it and placed vertically, both to your right." + }, + "answer_label": "B", + "answer_text": "One bed is by the window and placed horizontally; the other bed is opposite it and placed vertically. Both are to your right.", + "q_id": "scene0717_00-0_2_T2", + "correct_answer": "One bed is by the window and placed horizontally; the other bed is opposite it and placed vertically. Both are to your right.", + "original_options": { + "A": "Both beds are to your left.", + "B": "One bed is by the window and placed horizontally; the other bed is opposite it and placed vertically. Both are to your right.", + "C": "The beds are side-by-side, not opposite.", + "D": "One bed is by the window and placed vertically; the other is opposite it and placed horizontally, both to your right.", + "E": "One bed is away from the window and placed horizontally; the other is opposite it and placed vertically, both to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 21, + 22 + ] + }, + { + "video_uid": "scene0717_00-0", + "video_length": "00:00:20", + "qid": "scene0717_00-0_7", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the calculator or the mobile phone?", + "answer": "The calculator is closer to you.", + "question_chinese": "计算器和手机哪个离我更近？", + "answer_chinese": "计算器离你更近。", + "evidence": [ + { + "description": "The calculator is on the far left side of the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The phone is in the middle of the table.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "You are on the left side of the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:02", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The mobile phone is closer to you.", + "B": "The calculator is closer to you." + }, + "answer_label": "B", + "answer_text": "The calculator is closer to you.", + "q_id": "scene0717_00-0_7", + "correct_answer": "The calculator is closer to you.", + "original_options": { + "A": "The mobile phone is closer to you.", + "B": "The calculator is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0717_00-0", + "video_length": "00:00:20", + "qid": "scene0717_00-0_8", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the calculator or the mobile phone?", + "answer": "The mobile phone is closer to you.", + "question_chinese": "计算器和手机哪个离我更近？", + "answer_chinese": "手机离你更近。", + "evidence": [ + { + "description": "The calculator is on the far left side of the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The phone is in the middle of the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "You are in the middle of the table.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The mobile phone is closer to you.", + "B": "The calculator is closer to you." + }, + "answer_label": "A", + "answer_text": "The mobile phone is closer to you.", + "q_id": "scene0717_00-0_8", + "correct_answer": "The mobile phone is closer to you.", + "original_options": { + "A": "The mobile phone is closer to you.", + "B": "The calculator is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/scene0718_00-0-mcq.json b/QAs/scene0718_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..439bbb93150d798be08179d76b98b558649cc0b8 --- /dev/null +++ b/QAs/scene0718_00-0-mcq.json @@ -0,0 +1,626 @@ +[ + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_5", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the single sofa?", + "answer": "No,you can't.", + "question_chinese": "我可以直接坐在单个沙发上吗?", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The single sofa is too far away from you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "scene0718_00-0_5", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_6", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the single sofa?", + "answer": "No,you can't.", + "question_chinese": "我可以直接坐在单个沙发上吗?", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The single sofa is too far away from you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "scene0718_00-0_6", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes,you can.", + "B": "No,you can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_3", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "First, you face two sofas, and then you rotate clockwise by 90 degrees.", + "question_chinese": "我的运动轨迹是什么样的？", + "answer_chinese": "首先你面对着两个沙发，然后顺时针旋转了90度。", + "evidence": [ + { + "description": "First, you face two sofas.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you rotate clockwise by 90 degrees.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "First, you face two sofas, and then you rotate clockwise by 90 degrees.", + "B": "You rotate counterclockwise by 90 degrees.", + "C": "You rotate clockwise by 180 degrees.", + "D": "First, you face two sofas, and then you rotate counterclockwise by 90 degrees instead of clockwise.", + "E": "First, you face two sofas, and then you rotate clockwise by 90 degrees and take a step forward." + }, + "answer_label": "A", + "answer_text": "First, you face two sofas, and then you rotate clockwise by 90 degrees.", + "q_id": "scene0718_00-0_3", + "correct_answer": "First, you face two sofas, and then you rotate clockwise by 90 degrees.", + "original_options": { + "A": "First, you face two sofas, and then you rotate clockwise by 90 degrees.", + "B": "You rotate counterclockwise by 90 degrees.", + "C": "You rotate clockwise by 180 degrees.", + "D": "First, you face two sofas, and then you rotate counterclockwise by 90 degrees instead of clockwise.", + "E": "First, you face two sofas, and then you rotate clockwise by 90 degrees and take a step forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 18 + ] + }, + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_4", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "First, you face two sofas, and then you rotate clockwise by 270 degrees.", + "question_chinese": "我的运动轨迹是什么样的？", + "answer_chinese": "首先你面对着两个沙发，然后顺时针旋转了270度。", + "evidence": [ + { + "description": "First, you face two sofas.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you rotate clockwise by 270 degrees.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You rotate clockwise by 90 degrees.", + "B": "You rotate counterclockwise by 270 degrees.", + "C": "First, you face two sofas, and then you rotate clockwise by 270 degrees.", + "D": "First, you face two sofas, and then you rotate clockwise by 180 degrees, not 270 degrees.", + "E": "First, you face two sofas, and then you rotate counterclockwise by 90 degrees rather than rotating clockwise." + }, + "answer_label": "C", + "answer_text": "First, you face two sofas, and then you rotate clockwise by 270 degrees.", + "q_id": "scene0718_00-0_4", + "correct_answer": "First, you face two sofas, and then you rotate clockwise by 270 degrees.", + "original_options": { + "A": "You rotate clockwise by 90 degrees.", + "B": "You rotate counterclockwise by 270 degrees.", + "C": "First, you face two sofas, and then you rotate clockwise by 270 degrees.", + "D": "First, you face two sofas, and then you rotate clockwise by 180 degrees, not 270 degrees.", + "E": "First, you face two sofas, and then you rotate counterclockwise by 90 degrees rather than rotating clockwise." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_5_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sofas did I see?", + "answer": "You see three sofas; they are behind you.", + "question_chinese": "我看到几个沙发?", + "answer_chinese": "你看到三个沙发，它们在你的身后。", + "evidence": [ + { + "description": "There are three sofas in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You see three sofas; they are behind you.", + "B": "You see two sofas.", + "C": "You see four sofas.", + "D": "You see one sofa; it is behind you, not three sofas behind you.", + "E": "You see five sofas; they are behind you, not three sofas behind you." + }, + "answer_label": "A", + "answer_text": "You see three sofas; they are behind you.", + "q_id": "scene0718_00-0_5_T2", + "correct_answer": "You see three sofas; they are behind you.", + "original_options": { + "A": "You see three sofas; they are behind you.", + "B": "You see two sofas.", + "C": "You see four sofas.", + "D": "You see one sofa; it is behind you, not three sofas behind you.", + "E": "You see five sofas; they are behind you, not three sofas behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 13 + ] + }, + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_5_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sofas did I see?", + "answer": "You see one sofa; it is behind you.", + "question_chinese": "我看到几个沙发?", + "answer_chinese": "你看到一个沙发，它在你的身后。", + "evidence": [ + { + "description": "You see a sofa.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You see two sofas.", + "B": "You see zero sofas.", + "C": "You see three sofas; two are behind you and one is off to your right.", + "D": "You see four sofas; one is behind you, with the other three spread around the room.", + "E": "You see one sofa; it is behind you." + }, + "answer_label": "E", + "answer_text": "You see one sofa; it is behind you.", + "q_id": "scene0718_00-0_5_T1", + "correct_answer": "You see one sofa; it is behind you.", + "original_options": { + "A": "You see two sofas.", + "B": "You see zero sofas.", + "C": "You see three sofas; two are behind you and one is off to your right.", + "D": "You see four sofas; one is behind you, with the other three spread around the room.", + "E": "You see one sofa; it is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_4_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many doors did I see?", + "answer": "You see two doors; both are behind you.", + "question_chinese": "我看到多少个门？", + "answer_chinese": "你看到两个门，它们都在你的身后。", + "evidence": [ + { + "description": "You see two door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You see two doors; both are behind you.", + "B": "You see one door; it is to your left." + }, + "answer_label": "A", + "answer_text": "You see two doors; both are behind you.", + "q_id": "scene0718_00-0_4_T2", + "correct_answer": "You see two doors; both are behind you.", + "original_options": { + "A": "You see two doors; both are behind you.", + "B": "You see one door; it is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_4_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many doors did I see?", + "answer": "You see one door; it is to your left.", + "question_chinese": "我看到多少个门？", + "answer_chinese": "你看到一个门，它在你的左边。", + "evidence": [ + { + "description": "You see two doors.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You see two doors; both are behind you.", + "B": "You see one door; it is to your left." + }, + "answer_label": "B", + "answer_text": "You see one door; it is to your left.", + "q_id": "scene0718_00-0_4_T1", + "correct_answer": "You see one door; it is to your left.", + "original_options": { + "A": "You see two doors; both are behind you.", + "B": "You see one door; it is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_3_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Relative to me, where is the trash can?", + "answer": "The trash can is directly in front of you.", + "question_chinese": "相对于我的位置，垃圾桶在哪里？", + "answer_chinese": "垃圾桶在你的正前方。", + "evidence": [ + { + "description": "The trash can is directly in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The trash can is behind you.", + "B": "The trash can is directly in front of you." + }, + "answer_label": "B", + "answer_text": "The trash can is directly in front of you.", + "q_id": "scene0718_00-0_3_T1", + "correct_answer": "The trash can is directly in front of you.", + "original_options": { + "A": "The trash can is behind you.", + "B": "The trash can is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_3_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Relative to me, where is the trash can?", + "answer": "The trash can is to your left.", + "question_chinese": "相对于我的位置，垃圾桶在哪里？", + "answer_chinese": "垃圾桶在我的左边。", + "evidence": [ + { + "description": "The trash can is on my left.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The trash can is directly in front of you.", + "B": "The trash can is to your left." + }, + "answer_label": "B", + "answer_text": "The trash can is to your left.", + "q_id": "scene0718_00-0_3_T2", + "correct_answer": "The trash can is to your left.", + "original_options": { + "A": "The trash can is directly in front of you.", + "B": "The trash can is to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the door?", + "answer": "It's on your left.", + "question_chinese": "门在哪里？", + "answer_chinese": "在你左边。", + "evidence": [ + { + "description": "The door is beside the sofa.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "You are facing the door.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "You turned clockwise by 90 degrees.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "It's to your right.", + "B": "It's on your left.", + "C": "It's in front of you.", + "D": "It's behind you, so you’d need to turn around to face it.", + "E": "It's on your right side, but closer than where you are now." + }, + "answer_label": "B", + "answer_text": "It's on your left.", + "q_id": "scene0718_00-0_1", + "correct_answer": "It's on your left.", + "original_options": { + "A": "It's to your right.", + "B": "It's on your left.", + "C": "It's in front of you.", + "D": "It's behind you, so you’d need to turn around to face it.", + "E": "It's on your right side, but closer than where you are now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 12 + ] + }, + { + "video_uid": "scene0718_00-0", + "video_length": "00:00:10", + "qid": "scene0718_00-0_2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the door?", + "answer": "It's behind you.", + "question_chinese": "门在哪里？", + "answer_chinese": "在你后边。", + "evidence": [ + { + "description": "The door is beside the sofa", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You are facing the door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You turned 180 degrees clockwise", + "timestamps": { + "start": "00:00:03", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "It's in front of you.", + "B": "It's to your left.", + "C": "It's to your right side, so you would turn right rather than look behind you.", + "D": "It's straight ahead of you, not behind, visible when you face forward in the room.", + "E": "It's behind you." + }, + "answer_label": "E", + "answer_text": "It's behind you.", + "q_id": "scene0718_00-0_2", + "correct_answer": "It's behind you.", + "original_options": { + "A": "It's in front of you.", + "B": "It's to your left.", + "C": "It's to your right side, so you would turn right rather than look behind you.", + "D": "It's straight ahead of you, not behind, visible when you face forward in the room.", + "E": "It's behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 15 + ] + } +] diff --git a/QAs/scene0719_00-0-mcq.json b/QAs/scene0719_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b4685058fc9500d1afc45003debf92eea1c086e5 --- /dev/null +++ b/QAs/scene0719_00-0-mcq.json @@ -0,0 +1,472 @@ +[ + { + "video_uid": "scene0719_00-0", + "video_length": "00:00:20", + "qid": "scene0719_00-0_4_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I see the door open?", + "answer": "No. The door in front of you is closed.", + "question_chinese": "我看见门是打开的吗？", + "answer_chinese": "不是。门在你正前方是关住的。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes. The door in front of you is open.", + "B": "No. The door in front of you is closed." + }, + "answer_label": "B", + "answer_text": "No. The door in front of you is closed.", + "q_id": "scene0719_00-0_4_T2", + "correct_answer": "No. The door in front of you is closed.", + "original_options": { + "A": "Yes. The door in front of you is open.", + "B": "No. The door in front of you is closed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0719_00-0", + "video_length": "00:00:20", + "qid": "scene0719_00-0_4_T1", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I see the door open?", + "answer": "No. The door is in front of you.", + "question_chinese": "我看见门是打开的吗？", + "answer_chinese": "不是。门在你正前方。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. The door is in front of you.", + "B": "Yes. The door in front of you is open." + }, + "answer_label": "A", + "answer_text": "No. The door is in front of you.", + "q_id": "scene0719_00-0_4_T1", + "correct_answer": "No. The door is in front of you.", + "original_options": { + "A": "No. The door is in front of you.", + "B": "Yes. The door in front of you is open." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0719_00-0", + "video_length": "00:00:20", + "qid": "scene0719_00-0_3_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly grab the desk lamp?", + "answer": "Yes, you can. The desk lamp is on the cabinet to your front-left.", + "question_chinese": "我能直接拿到台灯吗？", + "answer_chinese": "可以，台灯在你左前方的柜子上。", + "evidence": [ + { + "description": "The table lamp is on the cabinet in the corner.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t. The desk lamp is too far to reach.", + "B": "Yes, you can. The desk lamp is on the cabinet to your front-left." + }, + "answer_label": "B", + "answer_text": "Yes, you can. The desk lamp is on the cabinet to your front-left.", + "q_id": "scene0719_00-0_3_T1", + "correct_answer": "Yes, you can. The desk lamp is on the cabinet to your front-left.", + "original_options": { + "A": "No, you can’t. The desk lamp is too far to reach.", + "B": "Yes, you can. The desk lamp is on the cabinet to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "scene0719_00-0", + "video_length": "00:00:20", + "qid": "scene0719_00-0_5", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the desk lamp directly?", + "answer": "No,you can't.", + "question_chinese": "我能直接拿到台灯吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The table lamp is on the cabinet in the corner\n.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "You are in the middle of the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "scene0719_00-0_5", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes,you can.", + "B": "No,you can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0719_00-0", + "video_length": "00:00:20", + "qid": "scene0719_00-0_3_T2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly grab the desk lamp?", + "answer": "No, you can’t.", + "question_chinese": "我能直接拿到台灯吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The table lamp is on the cabinet in the corner.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "You are in the middle of the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t.", + "B": "Yes, you can. The desk lamp is on the cabinet to your front-left." + }, + "answer_label": "A", + "answer_text": "No, you can’t.", + "q_id": "scene0719_00-0_3_T2", + "correct_answer": "No, you can’t.", + "original_options": { + "A": "No, you can’t.", + "B": "Yes, you can. The desk lamp is on the cabinet to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "scene0719_00-0", + "video_length": "00:00:20", + "qid": "scene0719_00-0_4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many clothes hangers are there in the room?", + "answer": "There is no clothes hanger in the room.", + "question_chinese": "房间里有几个衣架？", + "answer_chinese": "房间里没有衣架。", + "evidence": [ + { + "description": "There are two clothes racks in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "There is no clothes hanger on the cabinet.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "There is no clothes hanger on the bed.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "One clothes hanger.", + "B": "There is no clothes hanger in the room.", + "C": "Two clothes hangers.", + "D": "There are three clothes hangers in the room, hanging together on a single rack.", + "E": "There are four clothes hangers in the room, with a couple on the rack and the rest nearby." + }, + "answer_label": "B", + "answer_text": "There is no clothes hanger in the room.", + "q_id": "scene0719_00-0_4", + "correct_answer": "There is no clothes hanger in the room.", + "original_options": { + "A": "One clothes hanger.", + "B": "There is no clothes hanger in the room.", + "C": "Two clothes hangers.", + "D": "There are three clothes hangers in the room, hanging together on a single rack.", + "E": "There are four clothes hangers in the room, with a couple on the rack and the rest nearby." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 18 + ] + }, + { + "video_uid": "scene0719_00-0", + "video_length": "00:00:20", + "qid": "scene0719_00-0_3", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many clothes hangers are there in the room?", + "answer": "There are two clothes racks in the room.", + "question_chinese": "房间里有几个衣架？", + "answer_chinese": "房间里有两个衣架。", + "evidence": [ + { + "description": "There are two clothes racks in the room.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two clothes racks in the room.", + "B": "One clothes rack.", + "C": "Three clothes racks.", + "D": "There are four clothes racks in the room, spaced out along the wall rather than grouped together.", + "E": "There are five clothes racks in the room, lined up so you can hang clothes across multiple sections." + }, + "answer_label": "A", + "answer_text": "There are two clothes racks in the room.", + "q_id": "scene0719_00-0_3", + "correct_answer": "There are two clothes racks in the room.", + "original_options": { + "A": "There are two clothes racks in the room.", + "B": "One clothes rack.", + "C": "Three clothes racks.", + "D": "There are four clothes racks in the room, spaced out along the wall rather than grouped together.", + "E": "There are five clothes racks in the room, lined up so you can hang clothes across multiple sections." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 18 + ] + }, + { + "video_uid": "scene0719_00-0", + "video_length": "00:00:20", + "qid": "scene0719_00-0_1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the computer or the printer?", + "answer": "The computer is closer to you.", + "question_chinese": "电脑和打印机哪个离我更近？", + "answer_chinese": "电脑离你更近。", + "evidence": [ + { + "description": "The computer is on the table in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The printer is on the cabinet in the corner.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "You are closer to the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The computer is closer to you.", + "B": "The printer is closer to you." + }, + "answer_label": "A", + "answer_text": "The computer is closer to you.", + "q_id": "scene0719_00-0_1", + "correct_answer": "The computer is closer to you.", + "original_options": { + "A": "The computer is closer to you.", + "B": "The printer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0719_00-0", + "video_length": "00:00:20", + "qid": "scene0719_00-0_2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the computer or the printer?", + "answer": "The computer is closer to you.", + "question_chinese": "电脑和打印机哪个离我更近？", + "answer_chinese": "电脑离你更近。", + "evidence": [ + { + "description": "The computer is on the desk in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The printer is on the cabinet in the corner.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + }, + { + "description": "You are closer to the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The computer is closer to you.", + "B": "The printer is closer to you." + }, + "answer_label": "A", + "answer_text": "The computer is closer to you.", + "q_id": "scene0719_00-0_2", + "correct_answer": "The computer is closer to you.", + "original_options": { + "A": "The computer is closer to you.", + "B": "The printer is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + } +] diff --git a/QAs/scene0720_00-0-mcq.json b/QAs/scene0720_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c504934b9a4e2d55911686306031564903114bf8 --- /dev/null +++ b/QAs/scene0720_00-0-mcq.json @@ -0,0 +1,521 @@ +[ + { + "video_uid": "scene0720_00-0", + "video_length": "00:00:20", + "qid": "scene0720_00-0_3_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I am right now, which is closer to me, the hat or the shoes?", + "answer": "The hat is closer to you; it’s in front of you.", + "question_chinese": "从我现在的位置，帽子和鞋子哪个离我更近？", + "answer_chinese": "帽子离你更近帽子在你面前。", + "evidence": [ + { + "description": "The shoes is farther from the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The hat is closer to the person.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The hat is closer to you; it’s in front of you.", + "B": "The shoes are closer to you.", + "C": "The hat is farther from you.", + "D": "The shoes are closer to you; they’re right in front of you, not the hat.", + "E": "The hat is farther away, while the shoes are nearer and slightly to your left." + }, + "answer_label": "A", + "answer_text": "The hat is closer to you; it’s in front of you.", + "q_id": "scene0720_00-0_3_T1", + "correct_answer": "The hat is closer to you; it’s in front of you.", + "original_options": { + "A": "The hat is closer to you; it’s in front of you.", + "B": "The shoes are closer to you.", + "C": "The hat is farther from you.", + "D": "The shoes are closer to you; they’re right in front of you, not the hat.", + "E": "The hat is farther away, while the shoes are nearer and slightly to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 15 + ] + }, + { + "video_uid": "scene0720_00-0", + "video_length": "00:00:20", + "qid": "scene0720_00-0_3_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I am right now, which is closer to me, the hat or the shoes?", + "answer": "The shoes are farther from you, and they’re behind you.", + "question_chinese": "从我现在的位置，帽子和鞋子哪个离我更近？", + "answer_chinese": "鞋子离你更远，鞋子在你后边。", + "evidence": [ + { + "description": "The shoes is farther from the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The hat is closer to you.", + "B": "The shoes are farther from you, and they’re behind you.", + "C": "The shoes are closer to you.", + "D": "The hat is farther from you, and it’s in front of you.", + "E": "The shoes are farther away, but they’re in front of you instead of behind you." + }, + "answer_label": "B", + "answer_text": "The shoes are farther from you, and they’re behind you.", + "q_id": "scene0720_00-0_3_T2", + "correct_answer": "The shoes are farther from you, and they’re behind you.", + "original_options": { + "A": "The hat is closer to you.", + "B": "The shoes are farther from you, and they’re behind you.", + "C": "The shoes are closer to you.", + "D": "The hat is farther from you, and it’s in front of you.", + "E": "The shoes are farther away, but they’re in front of you instead of behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 16 + ] + }, + { + "video_uid": "scene0720_00-0", + "video_length": "00:00:20", + "qid": "scene0720_00-0_9", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the bed?", + "answer": "The bed is on your left.", + "question_chinese": "从我现在站的位置，床在哪个方向？", + "answer_chinese": "床在你的左边。", + "evidence": [ + { + "description": "The bed is on the left of the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The bed is in front of you.", + "B": "The bed is on your right.", + "C": "The bed is on your left.", + "D": "The bed is behind you, not on your left from where you’re standing.", + "E": "The bed is off to your right side, rather than to your left." + }, + "answer_label": "C", + "answer_text": "The bed is on your left.", + "q_id": "scene0720_00-0_9", + "correct_answer": "The bed is on your left.", + "original_options": { + "A": "The bed is in front of you.", + "B": "The bed is on your right.", + "C": "The bed is on your left.", + "D": "The bed is behind you, not on your left from where you’re standing.", + "E": "The bed is off to your right side, rather than to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 13 + ] + }, + { + "video_uid": "scene0720_00-0", + "video_length": "00:00:20", + "qid": "scene0720_00-0_10", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the bed?", + "answer": "The bed is behind you.", + "question_chinese": "从我现在站的位置，床在哪个方向？", + "answer_chinese": "床在你的后边。", + "evidence": [ + { + "description": "The bed is behind the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The bed is in front of you.", + "B": "The bed is behind you.", + "C": "The bed is on your left.", + "D": "The bed is to your right side from where you’re standing, not behind you.", + "E": "The bed is positioned in front of you instead of being behind you." + }, + "answer_label": "B", + "answer_text": "The bed is behind you.", + "q_id": "scene0720_00-0_10", + "correct_answer": "The bed is behind you.", + "original_options": { + "A": "The bed is in front of you.", + "B": "The bed is behind you.", + "C": "The bed is on your left.", + "D": "The bed is to your right side from where you’re standing, not behind you.", + "E": "The bed is positioned in front of you instead of being behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 13 + ] + }, + { + "video_uid": "scene0720_00-0", + "video_length": "00:00:20", + "qid": "scene0720_00-0_1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the laptop?", + "answer": "The laptop is in front of you.", + "question_chinese": "从我现在站的位置，笔记本电脑在哪个方向？", + "answer_chinese": "笔记本电脑在你的前方。", + "evidence": [ + { + "description": "he laptop is in front of the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The laptop is behind you.", + "B": "The laptop is on your left.", + "C": "The laptop is off to your right side rather than directly in front of you.", + "D": "The laptop is in front of you.", + "E": "The laptop is behind you near the wall, not positioned in front of you." + }, + "answer_label": "D", + "answer_text": "The laptop is in front of you.", + "q_id": "scene0720_00-0_1", + "correct_answer": "The laptop is in front of you.", + "original_options": { + "A": "The laptop is behind you.", + "B": "The laptop is on your left.", + "C": "The laptop is off to your right side rather than directly in front of you.", + "D": "The laptop is in front of you.", + "E": "The laptop is behind you near the wall, not positioned in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 14 + ] + }, + { + "video_uid": "scene0720_00-0", + "video_length": "00:00:20", + "qid": "scene0720_00-0_2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the laptop?", + "answer": "The laptop is on your left.", + "question_chinese": "从我现在站的位置，笔记本电脑在哪个方向？", + "answer_chinese": "笔记本电脑在你的左边。", + "evidence": [ + { + "description": "The laptop is on the left of the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The laptop is in front of you.", + "B": "The laptop is on your right.", + "C": "The laptop is on your left.", + "D": "The laptop is behind you, not on your left from where you’re standing.", + "E": "The laptop is in front of you on the table rather than positioned to your left." + }, + "answer_label": "C", + "answer_text": "The laptop is on your left.", + "q_id": "scene0720_00-0_2", + "correct_answer": "The laptop is on your left.", + "original_options": { + "A": "The laptop is in front of you.", + "B": "The laptop is on your right.", + "C": "The laptop is on your left.", + "D": "The laptop is behind you, not on your left from where you’re standing.", + "E": "The laptop is in front of you on the table rather than positioned to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 16 + ] + }, + { + "video_uid": "scene0720_00-0", + "video_length": "00:00:20", + "qid": "scene0720_00-0_4", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the hat?", + "answer": "No, it's a little far away from you.", + "question_chinese": "我现在的位置能够到帽子吗？", + "answer_chinese": "不能够到，它理你有点远。", + "evidence": [ + { + "description": "The hat is a little far away from the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, it's a little far away from you.", + "B": "Yes, you can reach it; it's very close to you." + }, + "answer_label": "A", + "answer_text": "No, it's a little far away from you.", + "q_id": "scene0720_00-0_4", + "correct_answer": "No, it's a little far away from you.", + "original_options": { + "A": "No, it's a little far away from you.", + "B": "Yes, you can reach it; it's very close to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "scene0720_00-0", + "video_length": "00:00:20", + "qid": "scene0720_00-0_3", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the hat?", + "answer": "Yes, It's very close to you.", + "question_chinese": "我现在的位置能够到帽子吗？", + "answer_chinese": "能够到，它离你很近。", + "evidence": [ + { + "description": "The hat is very close to the person.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you can’t reach it; it’s a little far away from you.", + "B": "Yes, It's very close to you." + }, + "answer_label": "B", + "answer_text": "Yes, It's very close to you.", + "q_id": "scene0720_00-0_3", + "correct_answer": "Yes, It's very close to you.", + "original_options": { + "A": "No, you can’t reach it; it’s a little far away from you.", + "B": "Yes, It's very close to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "scene0720_00-0", + "video_length": "00:00:20", + "qid": "scene0720_00-0_7", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route should I take to get out of the room?", + "answer": "You should go straight.", + "question_chinese": "我现在离开房间应该走什么路线？", + "answer_chinese": "你应该直走。", + "evidence": [ + { + "description": "The door is in front of the person.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should turn left.", + "B": "You should turn right.", + "C": "You should turn left, then continue forward until you reach the exit.", + "D": "You should back up a little, then turn right and head forward to leave.", + "E": "You should go straight." + }, + "answer_label": "E", + "answer_text": "You should go straight.", + "q_id": "scene0720_00-0_7", + "correct_answer": "You should go straight.", + "original_options": { + "A": "You should turn left.", + "B": "You should turn right.", + "C": "You should turn left, then continue forward until you reach the exit.", + "D": "You should back up a little, then turn right and head forward to leave.", + "E": "You should go straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + }, + { + "video_uid": "scene0720_00-0", + "video_length": "00:00:20", + "qid": "scene0720_00-0_8", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route should I take to get out of the room?", + "answer": "You should turn right first and then go straight.", + "question_chinese": "我现在离开房间应该走什么路线？", + "answer_chinese": "你应该先向右转再直走。", + "evidence": [ + { + "description": "The door is on the right side of this person.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should turn right first and then go straight.", + "B": "You should turn left first.", + "C": "You should go straight first.", + "D": "You should turn right first, then turn left to find the way out.", + "E": "You should go straight, then turn right at the next opening to exit." + }, + "answer_label": "A", + "answer_text": "You should turn right first and then go straight.", + "q_id": "scene0720_00-0_8", + "correct_answer": "You should turn right first and then go straight.", + "original_options": { + "A": "You should turn right first and then go straight.", + "B": "You should turn left first.", + "C": "You should go straight first.", + "D": "You should turn right first, then turn left to find the way out.", + "E": "You should go straight, then turn right at the next opening to exit." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + } +] diff --git a/QAs/scene0721_00-0-mcq.json b/QAs/scene0721_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a621948cfae5b1f402527564b295cf34fb53ef5c --- /dev/null +++ b/QAs/scene0721_00-0-mcq.json @@ -0,0 +1,302 @@ +[ + { + "video_uid": "scene0721_00-0", + "video_length": "00:00:20", + "qid": "scene0721_00-0_1_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the mirror?", + "answer": "The mirror is behind you.", + "question_chinese": "从我现在站的位置，镜子在哪个方向？", + "answer_chinese": "镜子在你的后面。", + "evidence": [ + { + "description": "The mirror is behind the person.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The mirror is in front of you.", + "B": "The mirror is behind you.", + "C": "The mirror is to your left.", + "D": "The mirror is to your right when you face forward from where you are standing.", + "E": "The mirror is above you, higher than eye level from your current position." + }, + "answer_label": "B", + "answer_text": "The mirror is behind you.", + "q_id": "scene0721_00-0_1_T1", + "correct_answer": "The mirror is behind you.", + "original_options": { + "A": "The mirror is in front of you.", + "B": "The mirror is behind you.", + "C": "The mirror is to your left.", + "D": "The mirror is to your right when you face forward from where you are standing.", + "E": "The mirror is above you, higher than eye level from your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 13 + ] + }, + { + "video_uid": "scene0721_00-0", + "video_length": "00:00:20", + "qid": "scene0721_00-0_1_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the mirror?", + "answer": "The mirror is behind you.", + "question_chinese": "从我现在站的位置，镜子在哪个方向？", + "answer_chinese": "镜子在你的后面。", + "evidence": [ + { + "description": "The mirror is behind the person.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The mirror is in front of you.", + "B": "The mirror is behind you.", + "C": "The mirror is to your right.", + "D": "The mirror is above you, so you need to look upward from where you stand.", + "E": "The mirror is to your left, just off to the side from your current view." + }, + "answer_label": "B", + "answer_text": "The mirror is behind you.", + "q_id": "scene0721_00-0_1_T2", + "correct_answer": "The mirror is behind you.", + "original_options": { + "A": "The mirror is in front of you.", + "B": "The mirror is behind you.", + "C": "The mirror is to your right.", + "D": "The mirror is above you, so you need to look upward from where you stand.", + "E": "The mirror is to your left, just off to the side from your current view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 15 + ] + }, + { + "video_uid": "scene0721_00-0", + "video_length": "00:00:20", + "qid": "_5", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the door?", + "answer": "The door is on your left.", + "question_chinese": "从我现在站的位置，门在哪个方向？", + "answer_chinese": "门在你的左边。", + "evidence": [ + { + "description": "The door is on the left of the person.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is in front of you.", + "B": "The door is on your right.", + "C": "The door is directly behind you from where you are standing right now.", + "D": "The door is slightly ahead and to your right when you face forward.", + "E": "The door is on your left." + }, + "answer_label": "E", + "answer_text": "The door is on your left.", + "q_id": "_5", + "correct_answer": "The door is on your left.", + "original_options": { + "A": "The door is in front of you.", + "B": "The door is on your right.", + "C": "The door is directly behind you from where you are standing right now.", + "D": "The door is slightly ahead and to your right when you face forward.", + "E": "The door is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 13 + ] + }, + { + "video_uid": "scene0721_00-0", + "video_length": "00:00:20", + "qid": "_6", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the door?", + "answer": "The door is behind you.", + "question_chinese": "从我现在站的位置，门在哪个方向？", + "answer_chinese": "门在你的后边。", + "evidence": [ + { + "description": "The door is behind the person.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is in front of you.", + "B": "The door is behind you.", + "C": "The door is on your left.", + "D": "The door is to your right side when you face forward from your position.", + "E": "The door is straight ahead of you, so you can walk forward to it." + }, + "answer_label": "B", + "answer_text": "The door is behind you.", + "q_id": "_6", + "correct_answer": "The door is behind you.", + "original_options": { + "A": "The door is in front of you.", + "B": "The door is behind you.", + "C": "The door is on your left.", + "D": "The door is to your right side when you face forward from your position.", + "E": "The door is straight ahead of you, so you can walk forward to it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 14 + ] + }, + { + "video_uid": "scene0721_00-0", + "video_length": "00:00:20", + "qid": "_7", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the shoes?", + "answer": "No, it's a little far away from you.", + "question_chinese": "我现在的位置能够到鞋子吗？", + "answer_chinese": "不能够到，它理你有点远。", + "evidence": [ + { + "description": "The shoes are a little far away from the person.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, it's close enough for you to reach.", + "B": "No, it's a little far away from you." + }, + "answer_label": "B", + "answer_text": "No, it's a little far away from you.", + "q_id": "_7", + "correct_answer": "No, it's a little far away from you.", + "original_options": { + "A": "Yes, it's close enough for you to reach.", + "B": "No, it's a little far away from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0721_00-0", + "video_length": "00:00:20", + "qid": "_8", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the shoes?", + "answer": "Yes, It's very close to you.", + "question_chinese": "我现在的位置能够到鞋子吗？", + "answer_chinese": "能够到，它离你很近。", + "evidence": [ + { + "description": "The shoes are very close to you.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, It's very close to you.", + "B": "No, it's too far for you to reach." + }, + "answer_label": "A", + "answer_text": "Yes, It's very close to you.", + "q_id": "_8", + "correct_answer": "Yes, It's very close to you.", + "original_options": { + "A": "Yes, It's very close to you.", + "B": "No, it's too far for you to reach." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + } +] diff --git a/QAs/scene0722_00-0-mcq.json b/QAs/scene0722_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e20a1d7905ad4dc65f888bfa7e9087bfe709af31 --- /dev/null +++ b/QAs/scene0722_00-0-mcq.json @@ -0,0 +1,560 @@ +[ + { + "video_uid": "scene0722_00-0", + "video_length": "00:00:20", + "qid": "scene0722_00-0_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the bike?", + "answer": "The bike is on your left.", + "question_chinese": "从我现在站的位置，自行车在哪个方向？", + "answer_chinese": "自行车在你的左边。", + "evidence": [ + { + "description": "The bike is on the left of the person.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The bike is in front of you.", + "B": "The bike is on your right.", + "C": "The bike is on your left.", + "D": "The bike is directly ahead of you, lined up with your path.", + "E": "The bike is behind you, slightly off to the right as you stand." + }, + "answer_label": "C", + "answer_text": "The bike is on your left.", + "q_id": "scene0722_00-0_7", + "correct_answer": "The bike is on your left.", + "original_options": { + "A": "The bike is in front of you.", + "B": "The bike is on your right.", + "C": "The bike is on your left.", + "D": "The bike is directly ahead of you, lined up with your path.", + "E": "The bike is behind you, slightly off to the right as you stand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 13 + ] + }, + { + "video_uid": "scene0722_00-0", + "video_length": "00:00:20", + "qid": "scene0722_00-0_8", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the bike?", + "answer": "The bike is behind you.", + "question_chinese": "从我现在站的位置，自行车在哪个方向？", + "answer_chinese": "自行车在你的后边。", + "evidence": [ + { + "description": "The bike is behind the person.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The bike is in front of you.", + "B": "The bike is behind you.", + "C": "The bike is to your right.", + "D": "The bike is off to your left side, next to where you're standing.", + "E": "The bike is straight ahead of you, positioned a few steps in front." + }, + "answer_label": "B", + "answer_text": "The bike is behind you.", + "q_id": "scene0722_00-0_8", + "correct_answer": "The bike is behind you.", + "original_options": { + "A": "The bike is in front of you.", + "B": "The bike is behind you.", + "C": "The bike is to your right.", + "D": "The bike is off to your left side, next to where you're standing.", + "E": "The bike is straight ahead of you, positioned a few steps in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 13 + ] + }, + { + "video_uid": "scene0722_00-0", + "video_length": "00:00:20", + "qid": "scene0722_00-0_3", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the door?", + "answer": "The door is on your left.", + "question_chinese": "从我现在站的位置，门在哪个方向？", + "answer_chinese": "门在你的左边。", + "evidence": [ + { + "description": "The door is on the left of the person.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is in front of you.", + "B": "The door is to your right.", + "C": "The door is on your left.", + "D": "The door is behind you, so you would need to turn around.", + "E": "The door is straight ahead of you, centered in your line of sight." + }, + "answer_label": "C", + "answer_text": "The door is on your left.", + "q_id": "scene0722_00-0_3", + "correct_answer": "The door is on your left.", + "original_options": { + "A": "The door is in front of you.", + "B": "The door is to your right.", + "C": "The door is on your left.", + "D": "The door is behind you, so you would need to turn around.", + "E": "The door is straight ahead of you, centered in your line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 12, + 13 + ] + }, + { + "video_uid": "scene0722_00-0", + "video_length": "00:00:20", + "qid": "scene0722_00-0_4", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the door?", + "answer": "The door is behind you.", + "question_chinese": "从我现在站的位置，门在哪个方向？", + "answer_chinese": "门在你的后边。", + "evidence": [ + { + "description": "The door is behind the person.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is in front of you.", + "B": "The door is behind you.", + "C": "The door is to your left.", + "D": "The door is to your right, so you should pivot slightly to face it.", + "E": "The door is directly ahead of you, not behind your current position." + }, + "answer_label": "B", + "answer_text": "The door is behind you.", + "q_id": "scene0722_00-0_4", + "correct_answer": "The door is behind you.", + "original_options": { + "A": "The door is in front of you.", + "B": "The door is behind you.", + "C": "The door is to your left.", + "D": "The door is to your right, so you should pivot slightly to face it.", + "E": "The door is directly ahead of you, not behind your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 12 + ] + }, + { + "video_uid": "scene0722_00-0", + "video_length": "00:00:20", + "qid": "scene0722_00-0_5", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which route should I take to use the computer?", + "answer": "You should turn left to use the computer.", + "question_chinese": "我应该走什么路线去使用电脑？", + "answer_chinese": "你应该向左走去使用电脑。", + "evidence": [ + { + "description": "The computer is on the left of the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right to use the computer.", + "B": "Walk straight to use the computer.", + "C": "Keep going straight ahead, then look for the computer on your right.", + "D": "You should turn left to use the computer.", + "E": "Turn right first and continue forward until you reach the computer station." + }, + "answer_label": "D", + "answer_text": "You should turn left to use the computer.", + "q_id": "scene0722_00-0_5", + "correct_answer": "You should turn left to use the computer.", + "original_options": { + "A": "Turn right to use the computer.", + "B": "Walk straight to use the computer.", + "C": "Keep going straight ahead, then look for the computer on your right.", + "D": "You should turn left to use the computer.", + "E": "Turn right first and continue forward until you reach the computer station." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 12, + 12 + ] + }, + { + "video_uid": "scene0722_00-0", + "video_length": "00:00:20", + "qid": "scene0722_00-0_6", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Which route should I take to use the computer?", + "answer": "You should turn right to use the computer.", + "question_chinese": "我应该走什么路线去使用电脑？", + "answer_chinese": "你应该向右走以使用电脑。", + "evidence": [ + { + "description": "The computer is on the right of the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left to use the computer.", + "B": "Walk straight to use the computer.", + "C": "Go straight ahead and you’ll reach the computer without making a turn.", + "D": "Turn left at the first opening, then continue forward to get to the computer.", + "E": "You should turn right to use the computer." + }, + "answer_label": "E", + "answer_text": "You should turn right to use the computer.", + "q_id": "scene0722_00-0_6", + "correct_answer": "You should turn right to use the computer.", + "original_options": { + "A": "Turn left to use the computer.", + "B": "Walk straight to use the computer.", + "C": "Go straight ahead and you’ll reach the computer without making a turn.", + "D": "Turn left at the first opening, then continue forward to get to the computer.", + "E": "You should turn right to use the computer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 14 + ] + }, + { + "video_uid": "scene0722_00-0", + "video_length": "00:00:20", + "qid": "scene0722_00-0_1_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see in the room?", + "answer": "There is one chair in the room, to your left.", + "question_chinese": "我看到房间里有几个椅子？", + "answer_chinese": "房间里有一个椅子，在你的左方。", + "evidence": [ + { + "description": "There are three chairs in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are zero chairs in the room.", + "B": "There is one chair in the room, to your left.", + "C": "There are two chairs in the room.", + "D": "There are three chairs in the room, all grouped together to your left.", + "E": "There are four chairs in the room, with two to your left and two in front." + }, + "answer_label": "B", + "answer_text": "There is one chair in the room, to your left.", + "q_id": "scene0722_00-0_1_T2", + "correct_answer": "There is one chair in the room, to your left.", + "original_options": { + "A": "There are zero chairs in the room.", + "B": "There is one chair in the room, to your left.", + "C": "There are two chairs in the room.", + "D": "There are three chairs in the room, all grouped together to your left.", + "E": "There are four chairs in the room, with two to your left and two in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 16 + ] + }, + { + "video_uid": "scene0722_00-0", + "video_length": "00:00:20", + "qid": "scene0722_00-0_1_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see in the room?", + "answer": "There are three chairs in the room, split between in front of you and to your left.", + "question_chinese": "我看到房间里有几个椅子？", + "answer_chinese": "房间里有三个椅子，分别在前方和左方。", + "evidence": [ + { + "description": "There are three chairs in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "There are three chairs in the room.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are three chairs in the room, split between in front of you and to your left.", + "B": "There are two chairs in the room.", + "C": "There are four chairs in the room.", + "D": "There is one chair in the room, positioned directly in front of you.", + "E": "There are five chairs in the room, spread out between in front of you and to your left." + }, + "answer_label": "A", + "answer_text": "There are three chairs in the room, split between in front of you and to your left.", + "q_id": "scene0722_00-0_1_T1", + "correct_answer": "There are three chairs in the room, split between in front of you and to your left.", + "original_options": { + "A": "There are three chairs in the room, split between in front of you and to your left.", + "B": "There are two chairs in the room.", + "C": "There are four chairs in the room.", + "D": "There is one chair in the room, positioned directly in front of you.", + "E": "There are five chairs in the room, spread out between in front of you and to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 18 + ] + }, + { + "video_uid": "scene0722_00-0", + "video_length": "00:00:20", + "qid": "scene0722_00-0_9", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many clothes in the room need to be put away?", + "answer": "There are three clothes in the room need to be put away.", + "question_chinese": "房间里有几件需要整理的衣服？", + "answer_chinese": "房间里有三件需要整理的衣服。", + "evidence": [ + { + "description": "There are three clothes in the room need to be put away.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "There are three clothes in the room need to be put away.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are three clothes in the room need to be put away.", + "B": "There is one clothing item to put away.", + "C": "There are four clothes to put away.", + "D": "There are five clothes in the room that still need to be put away.", + "E": "There are two clothes in the room that need to be put away, not three." + }, + "answer_label": "A", + "answer_text": "There are three clothes in the room need to be put away.", + "q_id": "scene0722_00-0_9", + "correct_answer": "There are three clothes in the room need to be put away.", + "original_options": { + "A": "There are three clothes in the room need to be put away.", + "B": "There is one clothing item to put away.", + "C": "There are four clothes to put away.", + "D": "There are five clothes in the room that still need to be put away.", + "E": "There are two clothes in the room that need to be put away, not three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 15 + ] + }, + { + "video_uid": "scene0722_00-0", + "video_length": "00:00:20", + "qid": "scene0722_00-0_10", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many clothes in the room need to be put away?", + "answer": "There are four clothes in the room need to be put away.", + "question_chinese": "房间里有几件需要整理的衣服？", + "answer_chinese": "房间里有四件需要整理的衣服。", + "evidence": [ + { + "description": "There are four clothes in the room need to be put away.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "There are four clothes in the room need to be put away.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + }, + { + "description": "There are four clothes in the room need to be put away.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "hard", + "qtype": "COUNT", + "options": { + "A": "There are three clothes to put away.", + "B": "There are five clothes to put away.", + "C": "There are six clothes in the room that need to be put away.", + "D": "There are two clothes in the room that need to be put away, not four.", + "E": "There are four clothes in the room need to be put away." + }, + "answer_label": "E", + "answer_text": "There are four clothes in the room need to be put away.", + "q_id": "scene0722_00-0_10", + "correct_answer": "There are four clothes in the room need to be put away.", + "original_options": { + "A": "There are three clothes to put away.", + "B": "There are five clothes to put away.", + "C": "There are six clothes in the room that need to be put away.", + "D": "There are two clothes in the room that need to be put away, not four.", + "E": "There are four clothes in the room need to be put away." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 13, + 15 + ] + } +] diff --git a/QAs/scene0723_00-0-mcq.json b/QAs/scene0723_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b21ce593c82e9dd1cbca64e11308f53c5abd1f04 --- /dev/null +++ b/QAs/scene0723_00-0-mcq.json @@ -0,0 +1,440 @@ +[ + { + "video_uid": "scene0723_00-0", + "video_length": "00:00:20", + "qid": "scene0723_00-0_3_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I’m standing now, which direction is the door?", + "answer": "The door is behind you.", + "question_chinese": "从我现在站的位置，门在哪个方向？", + "answer_chinese": "门在你的后边。", + "evidence": [ + { + "description": "The door.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The door is in front of you.", + "B": "The door is to your left.", + "C": "The door is behind you.", + "D": "The door is to your right when you face forward from where you are standing.", + "E": "The door is slightly ahead and to your left from your current standing position." + }, + "answer_label": "C", + "answer_text": "The door is behind you.", + "q_id": "scene0723_00-0_3_T2", + "correct_answer": "The door is behind you.", + "original_options": { + "A": "The door is in front of you.", + "B": "The door is to your left.", + "C": "The door is behind you.", + "D": "The door is to your right when you face forward from where you are standing.", + "E": "The door is slightly ahead and to your left from your current standing position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 14 + ] + }, + { + "video_uid": "scene0723_00-0", + "video_length": "00:00:20", + "qid": "scene0723_00-0_3_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I’m standing now, which direction is the door?", + "answer": "The door is to your right.", + "question_chinese": "从我现在站的位置，门在哪个方向？", + "answer_chinese": "门在你的右边。", + "evidence": [ + { + "description": "The door is on the right of the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The door is to your left.", + "B": "The door is behind you.", + "C": "The door is directly in front of you from where you’re standing now.", + "D": "The door is to your right.", + "E": "The door is ahead of you and a little to the left, not on your right." + }, + "answer_label": "D", + "answer_text": "The door is to your right.", + "q_id": "scene0723_00-0_3_T1", + "correct_answer": "The door is to your right.", + "original_options": { + "A": "The door is to your left.", + "B": "The door is behind you.", + "C": "The door is directly in front of you from where you’re standing now.", + "D": "The door is to your right.", + "E": "The door is ahead of you and a little to the left, not on your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 16 + ] + }, + { + "video_uid": "scene0723_00-0", + "video_length": "00:00:20", + "qid": "scene0723_00-0_9", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my current position?", + "answer": "You are outside the room.", + "question_chinese": "我现在在什么位置？", + "answer_chinese": "你在房间外面。", + "evidence": [ + { + "description": "The person is outside the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are inside the room.", + "B": "You are outside the room.", + "C": "You are in the hallway.", + "D": "You are standing just inside the room rather than being outside of it.", + "E": "You are positioned near the doorway inside the room, not outside the room." + }, + "answer_label": "B", + "answer_text": "You are outside the room.", + "q_id": "scene0723_00-0_9", + "correct_answer": "You are outside the room.", + "original_options": { + "A": "You are inside the room.", + "B": "You are outside the room.", + "C": "You are in the hallway.", + "D": "You are standing just inside the room rather than being outside of it.", + "E": "You are positioned near the doorway inside the room, not outside the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "scene0723_00-0", + "video_length": "00:00:20", + "qid": "scene0723_00-0_10", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my current position?", + "answer": "You are inside the room.", + "question_chinese": "我现在在什么位置？", + "answer_chinese": "你在房间里面。", + "evidence": [ + { + "description": "The person is inside the room.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are inside the room.", + "B": "You are outside the room.", + "C": "You are in the hallway.", + "D": "You are just outside the room near the entrance, not inside it.", + "E": "You are standing in the hallway close to the room, rather than being inside the room." + }, + "answer_label": "A", + "answer_text": "You are inside the room.", + "q_id": "scene0723_00-0_10", + "correct_answer": "You are inside the room.", + "original_options": { + "A": "You are inside the room.", + "B": "You are outside the room.", + "C": "You are in the hallway.", + "D": "You are just outside the room near the entrance, not inside it.", + "E": "You are standing in the hallway close to the room, rather than being inside the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 16 + ] + }, + { + "video_uid": "scene0723_00-0", + "video_length": "00:00:20", + "qid": "scene0723_00-0_3", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bags are there in the room?", + "answer": "There are three bags in the room.", + "question_chinese": "房间里有几个包？", + "answer_chinese": "房间里有三个包。", + "evidence": [ + { + "description": "There are three bags in the room.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one bag in the room.", + "B": "There are three bags in the room.", + "C": "There are four bags in the room.", + "D": "There are five bags in the room, not three bags as stated.", + "E": "There are two bags in the room total, so the count is less than three." + }, + "answer_label": "B", + "answer_text": "There are three bags in the room.", + "q_id": "scene0723_00-0_3", + "correct_answer": "There are three bags in the room.", + "original_options": { + "A": "There is one bag in the room.", + "B": "There are three bags in the room.", + "C": "There are four bags in the room.", + "D": "There are five bags in the room, not three bags as stated.", + "E": "There are two bags in the room total, so the count is less than three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 15 + ] + }, + { + "video_uid": "scene0723_00-0", + "video_length": "00:00:20", + "qid": "scene0723_00-0_4", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bags are there in the room?", + "answer": "There are six bags in the room.", + "question_chinese": "房间里有几个包？", + "answer_chinese": "房间里有六个包。", + "evidence": [ + { + "description": "There are three bags in the room.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:07" + } + }, + { + "description": "There are three bags in the room.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are five bags in the room.", + "B": "There are seven bags in the room.", + "C": "There are six bags in the room.", + "D": "There are eight bags in the room in total, not six bags.", + "E": "There are nine bags in the room altogether, which is more than six." + }, + "answer_label": "C", + "answer_text": "There are six bags in the room.", + "q_id": "scene0723_00-0_4", + "correct_answer": "There are six bags in the room.", + "original_options": { + "A": "There are five bags in the room.", + "B": "There are seven bags in the room.", + "C": "There are six bags in the room.", + "D": "There are eight bags in the room in total, not six bags.", + "E": "There are nine bags in the room altogether, which is more than six." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 13 + ] + }, + { + "video_uid": "scene0723_00-0", + "video_length": "00:00:20", + "qid": "scene0723_00-0_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many clothes are there in the room?", + "answer": "There are two clothes in the room.", + "question_chinese": "房间里有几件衣服？", + "answer_chinese": "房间里有两件衣服。", + "evidence": [ + { + "description": "There are two clothes in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one piece of clothing.", + "B": "There are three clothes in the room.", + "C": "There are four clothes in the room, with two items on the floor and two draped over furniture.", + "D": "There are two clothes in the room.", + "E": "There are five clothes in the room, spread between the bed area and the space near the doorway." + }, + "answer_label": "D", + "answer_text": "There are two clothes in the room.", + "q_id": "scene0723_00-0_1", + "correct_answer": "There are two clothes in the room.", + "original_options": { + "A": "There is one piece of clothing.", + "B": "There are three clothes in the room.", + "C": "There are four clothes in the room, with two items on the floor and two draped over furniture.", + "D": "There are two clothes in the room.", + "E": "There are five clothes in the room, spread between the bed area and the space near the doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 18 + ] + }, + { + "video_uid": "scene0723_00-0", + "video_length": "00:00:20", + "qid": "scene0723_00-0_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many clothes are there in the room?", + "answer": "There are three clothes in the room.", + "question_chinese": "房间里有几件衣服？", + "answer_chinese": "房间里有三件衣服。", + "evidence": [ + { + "description": "There are three clothes in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + }, + { + "description": "There are three clothes in the room.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are three clothes in the room.", + "B": "There are two clothes in the room.", + "C": "There are four clothes in the room.", + "D": "There are five clothes in the room, with three grouped together and two separated near the side.", + "E": "There are six clothes in the room, with a couple hanging and the rest lying on nearby surfaces." + }, + "answer_label": "A", + "answer_text": "There are three clothes in the room.", + "q_id": "scene0723_00-0_2", + "correct_answer": "There are three clothes in the room.", + "original_options": { + "A": "There are three clothes in the room.", + "B": "There are two clothes in the room.", + "C": "There are four clothes in the room.", + "D": "There are five clothes in the room, with three grouped together and two separated near the side.", + "E": "There are six clothes in the room, with a couple hanging and the rest lying on nearby surfaces." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 18 + ] + } +] diff --git a/QAs/scene0724_00-0-mcq.json b/QAs/scene0724_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c51c0519d646f9f7ec1d251be80daeeab0b49c8b --- /dev/null +++ b/QAs/scene0724_00-0-mcq.json @@ -0,0 +1,644 @@ +[ + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_3", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the window?", + "answer": "The window is in front of you.", + "question_chinese": "从我现在站的位置，窗户在哪个方向？", + "answer_chinese": "窗户在你的前面。", + "evidence": [ + { + "description": "The window is in front of the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is on your left.", + "B": "The window is behind you.", + "C": "The window is to your right, so you’d need to turn to face it.", + "D": "The window is in front of you.", + "E": "The window is above you, positioned higher than eye level from where you’re standing." + }, + "answer_label": "D", + "answer_text": "The window is in front of you.", + "q_id": "scene0724_00-0_3", + "correct_answer": "The window is in front of you.", + "original_options": { + "A": "The window is on your left.", + "B": "The window is behind you.", + "C": "The window is to your right, so you’d need to turn to face it.", + "D": "The window is in front of you.", + "E": "The window is above you, positioned higher than eye level from where you’re standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 15 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the window?", + "answer": "The window is on your left.", + "question_chinese": "从我现在站的位置，窗户在哪个方向？", + "answer_chinese": "窗户在你的左边。", + "evidence": [ + { + "description": "The window is on your left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The window is on your left.", + "B": "The window is on your right.", + "C": "The window is in front of you.", + "D": "The window is behind you, so it’s not in your direct line of sight.", + "E": "The window is above you, higher than where you’re looking from your current spot." + }, + "answer_label": "A", + "answer_text": "The window is on your left.", + "q_id": "scene0724_00-0_4", + "correct_answer": "The window is on your left.", + "original_options": { + "A": "The window is on your left.", + "B": "The window is on your right.", + "C": "The window is in front of you.", + "D": "The window is behind you, so it’s not in your direct line of sight.", + "E": "The window is above you, higher than where you’re looking from your current spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 15 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_5", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my current position?", + "answer": "You are outside the room.", + "question_chinese": "我现在在什么位置？", + "answer_chinese": "你在房间外面。", + "evidence": [ + { + "description": "The person is outside the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are inside the room.", + "B": "You are outside the room.", + "C": "You are in the kitchen.", + "D": "You are in the hallway, a short distance from the room’s doorway.", + "E": "You are next to the bathroom entrance, standing nearby rather than outside the room." + }, + "answer_label": "B", + "answer_text": "You are outside the room.", + "q_id": "scene0724_00-0_5", + "correct_answer": "You are outside the room.", + "original_options": { + "A": "You are inside the room.", + "B": "You are outside the room.", + "C": "You are in the kitchen.", + "D": "You are in the hallway, a short distance from the room’s doorway.", + "E": "You are next to the bathroom entrance, standing nearby rather than outside the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_3_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I right now?", + "answer": "You are outside the room.", + "question_chinese": "我现在在什么位置？", + "answer_chinese": "你在房间外面。", + "evidence": [ + { + "description": "The person is outside the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You are inside the room.", + "B": "You are by the doorway inside.", + "C": "You are outside the room.", + "D": "You are in the hallway, standing a few steps down the corridor away from the room’s entrance.", + "E": "You are next to the bed’s ladder, with the ladder positioned directly in front of you." + }, + "answer_label": "C", + "answer_text": "You are outside the room.", + "q_id": "scene0724_00-0_3_T1", + "correct_answer": "You are outside the room.", + "original_options": { + "A": "You are inside the room.", + "B": "You are by the doorway inside.", + "C": "You are outside the room.", + "D": "You are in the hallway, standing a few steps down the corridor away from the room’s entrance.", + "E": "You are next to the bed’s ladder, with the ladder positioned directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 17 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_3_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where am I right now?", + "answer": "You are next to the bed’s ladder, and the ladder is in front of you.", + "question_chinese": "我现在在什么位置？", + "answer_chinese": "你在床的梯子旁边，梯子在你面前。", + "evidence": [ + { + "description": "You stand in front of the bed ladder.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The ladder is behind you.", + "B": "The ladder is to your left.", + "C": "You are next to the bed’s ladder, and the ladder is in front of you.", + "D": "You are next to the bed’s ladder, but the ladder is positioned to your right.", + "E": "You are next to the bed’s ladder, and the ladder is above you rather than in front." + }, + "answer_label": "C", + "answer_text": "You are next to the bed’s ladder, and the ladder is in front of you.", + "q_id": "scene0724_00-0_3_T2", + "correct_answer": "You are next to the bed’s ladder, and the ladder is in front of you.", + "original_options": { + "A": "The ladder is behind you.", + "B": "The ladder is to your left.", + "C": "You are next to the bed’s ladder, and the ladder is in front of you.", + "D": "You are next to the bed’s ladder, but the ladder is positioned to your right.", + "E": "You are next to the bed’s ladder, and the ladder is above you rather than in front." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 18 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_5_T1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach the ladder?", + "answer": "You can’t reach it; it’s a bit far from you.", + "question_chinese": "我现在的位置能够到梯子吗？", + "answer_chinese": "不能够到，它理你有点远。", + "evidence": [ + { + "description": "The ladder is a little far away from the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach it; it’s close enough to grab.", + "B": "You can’t reach it; it’s a bit far from you." + }, + "answer_label": "B", + "answer_text": "You can’t reach it; it’s a bit far from you.", + "q_id": "scene0724_00-0_5_T1", + "correct_answer": "You can’t reach it; it’s a bit far from you.", + "original_options": { + "A": "You can reach it; it’s close enough to grab.", + "B": "You can’t reach it; it’s a bit far from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_5_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach the ladder?", + "answer": "You can reach it. The ladder is in front of you.", + "question_chinese": "我现在的位置能够到梯子吗？", + "answer_chinese": "能够到，梯子在你面前。", + "evidence": [ + { + "description": "The ladder is in front of you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You cannot reach it. The ladder is in front of you.", + "B": "You can reach it. The ladder is in front of you." + }, + "answer_label": "B", + "answer_text": "You can reach it. The ladder is in front of you.", + "q_id": "scene0724_00-0_5_T2", + "correct_answer": "You can reach it. The ladder is in front of you.", + "original_options": { + "A": "You cannot reach it. The ladder is in front of you.", + "B": "You can reach it. The ladder is in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_1_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pillows did I see in the room?", + "answer": "You saw 1 pillow in the room. The pillow is on the bed to your left.", + "question_chinese": "我看见房间里有几个枕头？", + "answer_chinese": "你看见房间里有1个枕头。枕头在你左边的床上。", + "evidence": [ + { + "description": "There are one pillow in the room.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 0 pillows.", + "B": "You saw 2 pillows.", + "C": "You saw 3 pillows in the room, all on the bed to your left.", + "D": "You saw 1 pillow in the room. The pillow is on the bed to your left.", + "E": "You saw 4 pillows in the room, all on the bed to your left." + }, + "answer_label": "D", + "answer_text": "You saw 1 pillow in the room. The pillow is on the bed to your left.", + "q_id": "scene0724_00-0_1_T2", + "correct_answer": "You saw 1 pillow in the room. The pillow is on the bed to your left.", + "original_options": { + "A": "You saw 0 pillows.", + "B": "You saw 2 pillows.", + "C": "You saw 3 pillows in the room, all on the bed to your left.", + "D": "You saw 1 pillow in the room. The pillow is on the bed to your left.", + "E": "You saw 4 pillows in the room, all on the bed to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_1_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pillows did I see in the room?", + "answer": "There are two pillows in the room.", + "question_chinese": "我看见房间里有几个枕头？", + "answer_chinese": "房间里有两个枕头。", + "evidence": [ + { + "description": "There are two pillows in the room.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + }, + { + "description": "There are two pillows in the room.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are zero pillows.", + "B": "There is one pillow.", + "C": "There are two pillows in the room.", + "D": "There are three pillows in the room, visible together rather than just a pair.", + "E": "There are four pillows in the room, not two, making the set larger than expected." + }, + "answer_label": "C", + "answer_text": "There are two pillows in the room.", + "q_id": "scene0724_00-0_1_T1", + "correct_answer": "There are two pillows in the room.", + "original_options": { + "A": "There are zero pillows.", + "B": "There is one pillow.", + "C": "There are two pillows in the room.", + "D": "There are three pillows in the room, visible together rather than just a pair.", + "E": "There are four pillows in the room, not two, making the set larger than expected." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_4_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I see the curtains open or closed?", + "answer": "You saw that the curtains were open. The curtains are in front of you.", + "question_chinese": "我看见窗帘是拉开的还是合上的？", + "answer_chinese": "你看见窗帘是拉开的。窗帘在你正前方。", + "evidence": [ + { + "description": "The curtains are open.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You saw that the curtains were open. The curtains are in front of you.", + "B": "You saw that the curtains were closed. The curtains are in front of you." + }, + "answer_label": "A", + "answer_text": "You saw that the curtains were open. The curtains are in front of you.", + "q_id": "scene0724_00-0_4_T1", + "correct_answer": "You saw that the curtains were open. The curtains are in front of you.", + "original_options": { + "A": "You saw that the curtains were open. The curtains are in front of you.", + "B": "You saw that the curtains were closed. The curtains are in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_4_T2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I see the curtains open or closed?", + "answer": "The curtains are open.", + "question_chinese": "我看见窗帘是拉开的还是合上的？", + "answer_chinese": "窗帘是拉开的。", + "evidence": [ + { + "description": "The curtains are open.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The curtains are closed.", + "B": "The curtains are open." + }, + "answer_label": "B", + "answer_text": "The curtains are open.", + "q_id": "scene0724_00-0_4_T2", + "correct_answer": "The curtains are open.", + "original_options": { + "A": "The curtains are closed.", + "B": "The curtains are open." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_7", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Are the curtains open or closed?", + "answer": "The curtains are open.", + "question_chinese": "窗帘是拉开的还是合上的？", + "answer_chinese": "窗帘是拉开的。", + "evidence": [ + { + "description": "The curtains are open.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The curtains are open.", + "B": "The curtains are closed." + }, + "answer_label": "A", + "answer_text": "The curtains are open.", + "q_id": "scene0724_00-0_7", + "correct_answer": "The curtains are open.", + "original_options": { + "A": "The curtains are open.", + "B": "The curtains are closed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "scene0724_00-0", + "video_length": "00:00:20", + "qid": "scene0724_00-0_8", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Are the curtains open or closed?", + "answer": "The curtains are open.", + "question_chinese": "窗帘是拉开的还是合上的？", + "answer_chinese": "窗帘是拉开的。", + "evidence": [ + { + "description": "The curtains are open.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The curtains are open.", + "B": "The curtains are closed." + }, + "answer_label": "A", + "answer_text": "The curtains are open.", + "q_id": "scene0724_00-0_8", + "correct_answer": "The curtains are open.", + "original_options": { + "A": "The curtains are open.", + "B": "The curtains are closed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + } +] diff --git a/QAs/scene0725_00-0-mcq.json b/QAs/scene0725_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..c68a814f41af532be33981d77b75281fe6d4e964 --- /dev/null +++ b/QAs/scene0725_00-0-mcq.json @@ -0,0 +1,496 @@ +[ + { + "video_uid": "scene0725_00-0", + "video_length": "00:00:20", + "qid": "scene0725_00-0_3", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I'm standing, which is closer to me, the ladder or the clothes?", + "answer": "The ladder is closer to you.", + "question_chinese": "从我现在的位置，梯子和衣服哪个离我更近？", + "answer_chinese": "梯子离你更近。", + "evidence": [ + { + "description": "The ladder is closer to the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The ladder is closer to you.", + "B": "The clothes are closer to you." + }, + "answer_label": "A", + "answer_text": "The ladder is closer to you.", + "q_id": "scene0725_00-0_3", + "correct_answer": "The ladder is closer to you.", + "original_options": { + "A": "The ladder is closer to you.", + "B": "The clothes are closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0725_00-0", + "video_length": "00:00:20", + "qid": "scene0725_00-0_4", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I'm standing, which is closer to me, the ladder or the clothes?", + "answer": "The clothes are closer to you.", + "question_chinese": "从我现在的位置，梯子和衣服哪个离我更近？", + "answer_chinese": "衣服离你更近。", + "evidence": [ + { + "description": "The clothes are closer to the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The clothes are closer to you.", + "B": "The ladder is closer to you." + }, + "answer_label": "A", + "answer_text": "The clothes are closer to you.", + "q_id": "scene0725_00-0_4", + "correct_answer": "The clothes are closer to you.", + "original_options": { + "A": "The clothes are closer to you.", + "B": "The ladder is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0725_00-0", + "video_length": "00:00:20", + "qid": "scene0725_00-0_5", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the ladder?", + "answer": "The ladder is behind you.", + "question_chinese": "从我现在站的位置，梯子在哪个方向？", + "answer_chinese": "梯子在你的后面。", + "evidence": [ + { + "description": "The ladder is behind the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To your right.", + "B": "The ladder is behind you.", + "C": "In front of you.", + "D": "The ladder is on your left side, not behind you where you are standing.", + "E": "The ladder is in front of you, so you wouldn’t need to turn around to face it." + }, + "answer_label": "B", + "answer_text": "The ladder is behind you.", + "q_id": "scene0725_00-0_5", + "correct_answer": "The ladder is behind you.", + "original_options": { + "A": "To your right.", + "B": "The ladder is behind you.", + "C": "In front of you.", + "D": "The ladder is on your left side, not behind you where you are standing.", + "E": "The ladder is in front of you, so you wouldn’t need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 18 + ] + }, + { + "video_uid": "scene0725_00-0", + "video_length": "00:00:20", + "qid": "scene0725_00-0_6", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the ladder?", + "answer": "The ladder is in front of you.", + "question_chinese": "从我现在站的位置，梯子在哪个方向？", + "answer_chinese": "梯子在你的前方。", + "evidence": [ + { + "description": "The ladder is in front of the person.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "The ladder is in front of you.", + "C": "To your right.", + "D": "The ladder is off to your left rather than directly in front of where you're standing.", + "E": "The ladder is behind your position, so you would have to turn around to face it." + }, + "answer_label": "B", + "answer_text": "The ladder is in front of you.", + "q_id": "scene0725_00-0_6", + "correct_answer": "The ladder is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "The ladder is in front of you.", + "C": "To your right.", + "D": "The ladder is off to your left rather than directly in front of where you're standing.", + "E": "The ladder is behind your position, so you would have to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 16, + 16 + ] + }, + { + "video_uid": "scene0725_00-0", + "video_length": "00:00:20", + "qid": "scene0725_00-0_9", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the green door?", + "answer": "The green door is on your left.", + "question_chinese": "从我现在站的位置，绿色的门在哪个方向？", + "answer_chinese": "绿色的门在你的左边。", + "evidence": [ + { + "description": "The green door is on the left of the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "On your right.", + "C": "The green door is on your left.", + "D": "The green door is behind you, not on your left from where you're currently standing.", + "E": "The green door is straight ahead of you, so it isn’t located to your left side." + }, + "answer_label": "C", + "answer_text": "The green door is on your left.", + "q_id": "scene0725_00-0_9", + "correct_answer": "The green door is on your left.", + "original_options": { + "A": "In front of you.", + "B": "On your right.", + "C": "The green door is on your left.", + "D": "The green door is behind you, not on your left from where you're currently standing.", + "E": "The green door is straight ahead of you, so it isn’t located to your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 17 + ] + }, + { + "video_uid": "scene0725_00-0", + "video_length": "00:00:20", + "qid": "scene0725_00-0_10", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the green door?", + "answer": "The green door is behind you.", + "question_chinese": "从我现在站的位置，绿色的门在哪个方向？", + "answer_chinese": "绿色的门在你的后面。", + "evidence": [ + { + "description": "The green door is behind the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The green door is to your right, not behind you from your current standing position.", + "D": "The green door is ahead of you, so you wouldn’t need to turn around to face it.", + "E": "The green door is behind you." + }, + "answer_label": "E", + "answer_text": "The green door is behind you.", + "q_id": "scene0725_00-0_10", + "correct_answer": "The green door is behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "The green door is to your right, not behind you from your current standing position.", + "D": "The green door is ahead of you, so you wouldn’t need to turn around to face it.", + "E": "The green door is behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 18 + ] + }, + { + "video_uid": "scene0725_00-0", + "video_length": "00:00:20", + "qid": "scene0725_00-0_1", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my current position?", + "answer": "You are outside the room.", + "question_chinese": "我现在在什么位置？", + "answer_chinese": "你在房间外面。", + "evidence": [ + { + "description": "The person is outside the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are inside the room.", + "B": "You are in the hallway.", + "C": "You are standing in the room rather than outside it, with the doorway behind you.", + "D": "You are outside the room.", + "E": "You are positioned just inside the room, not outside beyond the threshold." + }, + "answer_label": "D", + "answer_text": "You are outside the room.", + "q_id": "scene0725_00-0_1", + "correct_answer": "You are outside the room.", + "original_options": { + "A": "You are inside the room.", + "B": "You are in the hallway.", + "C": "You are standing in the room rather than outside it, with the doorway behind you.", + "D": "You are outside the room.", + "E": "You are positioned just inside the room, not outside beyond the threshold." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 12 + ] + }, + { + "video_uid": "scene0725_00-0", + "video_length": "00:00:20", + "qid": "scene0725_00-0_2", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is my current position?", + "answer": "You are inside the room.", + "question_chinese": "我现在在什么位置？", + "answer_chinese": "你在房间里面。", + "evidence": [ + { + "description": "The person is inside the room.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You are inside the room.", + "B": "You are outside the room.", + "C": "You are in the hallway.", + "D": "You are standing outside the room rather than inside it, facing toward the doorway.", + "E": "You are positioned in the corridor instead of inside the room you’re looking toward." + }, + "answer_label": "A", + "answer_text": "You are inside the room.", + "q_id": "scene0725_00-0_2", + "correct_answer": "You are inside the room.", + "original_options": { + "A": "You are inside the room.", + "B": "You are outside the room.", + "C": "You are in the hallway.", + "D": "You are standing outside the room rather than inside it, facing toward the doorway.", + "E": "You are positioned in the corridor instead of inside the room you’re looking toward." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 15 + ] + }, + { + "video_uid": "scene0725_00-0", + "video_length": "00:00:20", + "qid": "scene0725_00-0_4_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach the bag?", + "answer": "You can reach it; it's in front of you.", + "question_chinese": "我现在的位置能够到包吗？", + "answer_chinese": "能够到，它在你前面。", + "evidence": [ + { + "description": "The bag is very close to the person.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach it; it’s too far away in front of you.", + "B": "You can reach it; it's in front of you." + }, + "answer_label": "B", + "answer_text": "You can reach it; it's in front of you.", + "q_id": "scene0725_00-0_4_T1", + "correct_answer": "You can reach it; it's in front of you.", + "original_options": { + "A": "You can’t reach it; it’s too far away in front of you.", + "B": "You can reach it; it's in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "scene0725_00-0", + "video_length": "00:00:20", + "qid": "scene0725_00-0_4_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I am right now, can I reach the bag?", + "answer": "You can’t reach the bag; it’s to your back-left.", + "question_chinese": "我现在的位置能够到包吗？", + "answer_chinese": "不能够到包，它在你左后方。", + "evidence": [ + { + "description": "The bag is very close to the person.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the bag; it’s to your back-left.", + "B": "You can reach the bag; it’s to your front-right." + }, + "answer_label": "A", + "answer_text": "You can’t reach the bag; it’s to your back-left.", + "q_id": "scene0725_00-0_4_T2", + "correct_answer": "You can’t reach the bag; it’s to your back-left.", + "original_options": { + "A": "You can’t reach the bag; it’s to your back-left.", + "B": "You can reach the bag; it’s to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + } +] diff --git a/QAs/scene0730_00-0-mcq.json b/QAs/scene0730_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..60e4e191929bcf3dce071b7cab23e1e520f9b75b --- /dev/null +++ b/QAs/scene0730_00-0-mcq.json @@ -0,0 +1,196 @@ +[ + { + "video_uid": "scene0730_00-0", + "video_length": "00:00:20", + "qid": "scene0730_00-0_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the door?", + "answer": "The door is behind you.", + "question_chinese": "从我现在站的位置，门在哪个方向？", + "answer_chinese": "门在你的后边。", + "evidence": [ + { + "description": "The door is behind the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is in front of you.", + "B": "The door is to your left.", + "C": "The door is slightly to your right when facing forward from where you stand now.", + "D": "The door is behind you.", + "E": "The door is ahead of you, a bit off to the left side." + }, + "answer_label": "D", + "answer_text": "The door is behind you.", + "q_id": "scene0730_00-0_5", + "correct_answer": "The door is behind you.", + "original_options": { + "A": "The door is in front of you.", + "B": "The door is to your left.", + "C": "The door is slightly to your right when facing forward from where you stand now.", + "D": "The door is behind you.", + "E": "The door is ahead of you, a bit off to the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 13 + ] + }, + { + "video_uid": "scene0730_00-0", + "video_length": "00:00:20", + "qid": "scene0730_00-0_6", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the door?", + "answer": "The door is on your right.", + "question_chinese": "从我现在站的位置，门在哪个方向？", + "answer_chinese": "门在你的右边。", + "evidence": [ + { + "description": "The door is on the right of the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is in front of you.", + "B": "The door is behind you.", + "C": "The door is slightly to your left when facing forward from where you stand now.", + "D": "The door is on your right.", + "E": "The door is behind you, a little off to your left side." + }, + "answer_label": "D", + "answer_text": "The door is on your right.", + "q_id": "scene0730_00-0_6", + "correct_answer": "The door is on your right.", + "original_options": { + "A": "The door is in front of you.", + "B": "The door is behind you.", + "C": "The door is slightly to your left when facing forward from where you stand now.", + "D": "The door is on your right.", + "E": "The door is behind you, a little off to your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 12 + ] + }, + { + "video_uid": "scene0730_00-0", + "video_length": "00:00:20", + "qid": "scene0730_00-0_1_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the ladder?", + "answer": "No, it's a little far away from you.", + "question_chinese": "我现在的位置能够到梯子吗？", + "answer_chinese": "不能够到，它理你有点远。", + "evidence": [ + { + "description": "The ladder is a little far away from the person.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can reach the ladder; it’s within arm’s length.", + "B": "No, it's a little far away from you." + }, + "answer_label": "B", + "answer_text": "No, it's a little far away from you.", + "q_id": "scene0730_00-0_1_T1", + "correct_answer": "No, it's a little far away from you.", + "original_options": { + "A": "Yes, you can reach the ladder; it’s within arm’s length.", + "B": "No, it's a little far away from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "scene0730_00-0", + "video_length": "00:00:20", + "qid": "scene0730_00-0_1_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the ladder?", + "answer": "No, it's a little far away from you.", + "question_chinese": "我现在的位置能够到梯子吗？", + "answer_chinese": "不能够到，它理你有点远。", + "evidence": [ + { + "description": "The ladder is a little far away from the person.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, it's a little far away from you.", + "B": "Yes, you can reach the ladder; it’s within arm’s length." + }, + "answer_label": "A", + "answer_text": "No, it's a little far away from you.", + "q_id": "scene0730_00-0_1_T2", + "correct_answer": "No, it's a little far away from you.", + "original_options": { + "A": "No, it's a little far away from you.", + "B": "Yes, you can reach the ladder; it’s within arm’s length." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + } +] diff --git a/QAs/scene0731_00-0-mcq.json b/QAs/scene0731_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1f591224a0c769b439ea404144f75b9e8959f0fd --- /dev/null +++ b/QAs/scene0731_00-0-mcq.json @@ -0,0 +1,553 @@ +[ + { + "video_uid": "scene0731_00-0", + "video_length": "00:00:20", + "qid": "scene0731_00-0_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction are the clothes?", + "answer": "The clothes are on your right.", + "question_chinese": "从我现在站的位置，衣服在哪个方向？", + "answer_chinese": "衣服在你的右边。", + "evidence": [ + { + "description": "The clothes are on the right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The clothes are in front of you.", + "B": "The clothes are behind you.", + "C": "The clothes are slightly to your left when facing forward from where you stand now.", + "D": "The clothes are behind you, a little off to the left side.", + "E": "The clothes are on your right." + }, + "answer_label": "E", + "answer_text": "The clothes are on your right.", + "q_id": "scene0731_00-0_7", + "correct_answer": "The clothes are on your right.", + "original_options": { + "A": "The clothes are in front of you.", + "B": "The clothes are behind you.", + "C": "The clothes are slightly to your left when facing forward from where you stand now.", + "D": "The clothes are behind you, a little off to the left side.", + "E": "The clothes are on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 12 + ] + }, + { + "video_uid": "scene0731_00-0", + "video_length": "00:00:20", + "qid": "scene0731_00-0_8", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction are the clothes?", + "answer": "The clothes are behind you.", + "question_chinese": "从我现在站的位置，衣服在哪个方向？", + "answer_chinese": "衣服在你的后边。", + "evidence": [ + { + "description": "The clothes are behind the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The clothes are behind you.", + "B": "The clothes are in front of you.", + "C": "The clothes are on your right.", + "D": "The clothes are slightly to your left when facing forward from where you stand now.", + "E": "The clothes are in front of you, a bit off to your right side." + }, + "answer_label": "A", + "answer_text": "The clothes are behind you.", + "q_id": "scene0731_00-0_8", + "correct_answer": "The clothes are behind you.", + "original_options": { + "A": "The clothes are behind you.", + "B": "The clothes are in front of you.", + "C": "The clothes are on your right.", + "D": "The clothes are slightly to your left when facing forward from where you stand now.", + "E": "The clothes are in front of you, a bit off to your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 14 + ] + }, + { + "video_uid": "scene0731_00-0", + "video_length": "00:00:20", + "qid": "scene0731_00-0_5", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route should I take to get out of the room?", + "answer": "You should turn around and go straight.", + "question_chinese": "我现在离开房间应该走什么路线？", + "answer_chinese": "你应该向后转然后直走。", + "evidence": [ + { + "description": "The door is behind the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go straight without turning.", + "B": "You should turn around and go straight.", + "C": "Turn left and go straight.", + "D": "Turn around, then take a left before you go straight out through the exit.", + "E": "Turn to your right first, then continue straight until you go out of the room." + }, + "answer_label": "B", + "answer_text": "You should turn around and go straight.", + "q_id": "scene0731_00-0_5", + "correct_answer": "You should turn around and go straight.", + "original_options": { + "A": "Go straight without turning.", + "B": "You should turn around and go straight.", + "C": "Turn left and go straight.", + "D": "Turn around, then take a left before you go straight out through the exit.", + "E": "Turn to your right first, then continue straight until you go out of the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 15 + ] + }, + { + "video_uid": "scene0731_00-0", + "video_length": "00:00:20", + "qid": "scene0731_00-0_6", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route should I take to get out of the room?", + "answer": "You should turn left and walk to the corner, then turn right and go straight.", + "question_chinese": "我现在离开房间应该走什么路线？", + "answer_chinese": "你应该向左走到墙角，然后向右直走。", + "evidence": [ + { + "description": "Here is the corner.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:06" + } + }, + { + "description": "Here is the door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight.", + "B": "You should turn left and walk to the corner, then turn right and go straight.", + "C": "Turn left, then go straight.", + "D": "Turn right and walk to the corner, then turn left and continue straight to exit the room.", + "E": "Go straight to the corner first, then turn left and keep walking straight to get out." + }, + "answer_label": "B", + "answer_text": "You should turn left and walk to the corner, then turn right and go straight.", + "q_id": "scene0731_00-0_6", + "correct_answer": "You should turn left and walk to the corner, then turn right and go straight.", + "original_options": { + "A": "Turn right, then go straight.", + "B": "You should turn left and walk to the corner, then turn right and go straight.", + "C": "Turn left, then go straight.", + "D": "Turn right and walk to the corner, then turn left and continue straight to exit the room.", + "E": "Go straight to the corner first, then turn left and keep walking straight to get out." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 16 + ] + }, + { + "video_uid": "scene0731_00-0", + "video_length": "00:00:20", + "qid": "scene0731_00-0_5_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many windows have appeared in my view?", + "answer": "You saw one window.", + "question_chinese": "我的视野里出现过几个窗户？", + "answer_chinese": "你看见了一个窗户。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one window.", + "B": "You saw zero windows.", + "C": "You saw two windows.", + "D": "You saw three windows appear in your view as you looked around the room.", + "E": "You saw four windows in total come into view during the time you were in the room." + }, + "answer_label": "A", + "answer_text": "You saw one window.", + "q_id": "scene0731_00-0_5_T2", + "correct_answer": "You saw one window.", + "original_options": { + "A": "You saw one window.", + "B": "You saw zero windows.", + "C": "You saw two windows.", + "D": "You saw three windows appear in your view as you looked around the room.", + "E": "You saw four windows in total come into view during the time you were in the room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 17 + ] + }, + { + "video_uid": "scene0731_00-0", + "video_length": "00:00:20", + "qid": "scene0731_00-0_5_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many windows have appeared in my view?", + "answer": "You saw two windows.", + "question_chinese": "我的视野里出现过几个窗户？", + "answer_chinese": "你看见了两个窗户。", + "evidence": [ + { + "description": "the first one", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "the second one", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one window.", + "B": "You saw three windows.", + "C": "You saw zero windows appear in your view during the time you were looking around.", + "D": "You saw four windows in total appear in your view at different moments while moving.", + "E": "You saw two windows." + }, + "answer_label": "E", + "answer_text": "You saw two windows.", + "q_id": "scene0731_00-0_5_T1", + "correct_answer": "You saw two windows.", + "original_options": { + "A": "You saw one window.", + "B": "You saw three windows.", + "C": "You saw zero windows appear in your view during the time you were looking around.", + "D": "You saw four windows in total appear in your view at different moments while moving.", + "E": "You saw two windows." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + }, + { + "video_uid": "scene0731_00-0", + "video_length": "00:00:20", + "qid": "scene0731_00-0_2_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bags are in front of me now?", + "answer": "You saw three bags in front of you.", + "question_chinese": "我的面前现在出现了几个包？", + "answer_chinese": "你看见了面前的三个包。", + "evidence": [ + { + "description": "There are three bags in the room.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two bags.", + "B": "You saw three bags in front of you.", + "C": "You saw four bags.", + "D": "You saw one bag in front of you, with no other bags visible nearby.", + "E": "You saw five bags in front of you at once, lined up together in your view." + }, + "answer_label": "B", + "answer_text": "You saw three bags in front of you.", + "q_id": "scene0731_00-0_2_T1", + "correct_answer": "You saw three bags in front of you.", + "original_options": { + "A": "You saw two bags.", + "B": "You saw three bags in front of you.", + "C": "You saw four bags.", + "D": "You saw one bag in front of you, with no other bags visible nearby.", + "E": "You saw five bags in front of you at once, lined up together in your view." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 16 + ] + }, + { + "video_uid": "scene0731_00-0", + "video_length": "00:00:20", + "qid": "scene0731_00-0_2_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bags are in front of me now?", + "answer": "There are no bags in front of you right now.", + "question_chinese": "我的面前现在出现了几个包？", + "answer_chinese": "你现在的前方没有包", + "evidence": [ + { + "description": "There are three bags in the room.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one bag.", + "B": "You saw two bags.", + "C": "There are no bags in front of you right now.", + "D": "You saw three bags in front of you right now, all visible at the same time.", + "E": "You saw four bags in front of you currently, with multiple bags grouped together in view." + }, + "answer_label": "C", + "answer_text": "There are no bags in front of you right now.", + "q_id": "scene0731_00-0_2_T2", + "correct_answer": "There are no bags in front of you right now.", + "original_options": { + "A": "You saw one bag.", + "B": "You saw two bags.", + "C": "There are no bags in front of you right now.", + "D": "You saw three bags in front of you right now, all visible at the same time.", + "E": "You saw four bags in front of you currently, with multiple bags grouped together in view." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "scene0731_00-0", + "video_length": "00:00:20", + "qid": "scene0731_00-0_1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pillows are there in the room?", + "answer": "There is one pillow in the room.", + "question_chinese": "房间里有几个枕头？", + "answer_chinese": "房间里有一个枕头。", + "evidence": [ + { + "description": "There is one pillow in the room.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are no pillows.", + "B": "There are two pillows.", + "C": "There is one pillow in the room.", + "D": "There are three pillows in the room, visible across the seating area.", + "E": "There are four pillows in the room, with more than one pillow placed on the furniture." + }, + "answer_label": "C", + "answer_text": "There is one pillow in the room.", + "q_id": "scene0731_00-0_1", + "correct_answer": "There is one pillow in the room.", + "original_options": { + "A": "There are no pillows.", + "B": "There are two pillows.", + "C": "There is one pillow in the room.", + "D": "There are three pillows in the room, visible across the seating area.", + "E": "There are four pillows in the room, with more than one pillow placed on the furniture." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 16 + ] + }, + { + "video_uid": "scene0731_00-0", + "video_length": "00:00:20", + "qid": "scene0731_00-0_2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pillows are there in the room?", + "answer": "There are two pillows in the room.", + "question_chinese": "房间里有几个枕头？", + "answer_chinese": "房间里有两个枕头。", + "evidence": [ + { + "description": "There are two pillows in the room.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + }, + { + "description": "There are two pillows in the room.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are two pillows in the room.", + "B": "There is one pillow.", + "C": "There are three pillows.", + "D": "There are four pillows in the room, more than just a pair on the furniture.", + "E": "There are five pillows in the room, with several cushions visible around the seating area." + }, + "answer_label": "A", + "answer_text": "There are two pillows in the room.", + "q_id": "scene0731_00-0_2", + "correct_answer": "There are two pillows in the room.", + "original_options": { + "A": "There are two pillows in the room.", + "B": "There is one pillow.", + "C": "There are three pillows.", + "D": "There are four pillows in the room, more than just a pair on the furniture.", + "E": "There are five pillows in the room, with several cushions visible around the seating area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 15 + ] + } +] diff --git a/QAs/scene0732_00-0-mcq.json b/QAs/scene0732_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6ca1e868dde6aefc6af87401f13d1117d08f20cc --- /dev/null +++ b/QAs/scene0732_00-0-mcq.json @@ -0,0 +1,485 @@ +[ + { + "video_uid": "scene0732_00-0", + "video_length": "00:00:20", + "qid": "scene0732_00-0_3_T2", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I am right now, which is closer to me: the microwave or the chair?", + "answer": "The chair was closer to you.", + "question_chinese": "从我现在的位置，微波炉和椅子哪个离我更近？", + "answer_chinese": "椅子离你更近。", + "evidence": [ + { + "description": "The position of the chair", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The chair was closer to you.", + "B": "The microwave was closer to you." + }, + "answer_label": "A", + "answer_text": "The chair was closer to you.", + "q_id": "scene0732_00-0_3_T2", + "correct_answer": "The chair was closer to you.", + "original_options": { + "A": "The chair was closer to you.", + "B": "The microwave was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0732_00-0", + "video_length": "00:00:20", + "qid": "scene0732_00-0_3_T1", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "From where I am right now, which is closer to me: the microwave or the chair?", + "answer": "The microwave was closer to you.", + "question_chinese": "从我现在的位置，微波炉和椅子哪个离我更近？", + "answer_chinese": "微波炉离你更近。", + "evidence": [ + { + "description": "The chair is far.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The microwave is closer.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The microwave was closer to you.", + "B": "The chair was closer to you." + }, + "answer_label": "A", + "answer_text": "The microwave was closer to you.", + "q_id": "scene0732_00-0_3_T1", + "correct_answer": "The microwave was closer to you.", + "original_options": { + "A": "The microwave was closer to you.", + "B": "The chair was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0732_00-0", + "video_length": "00:00:20", + "qid": "scene0732_00-0_2_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I’m standing right now, which direction is the fridge?", + "answer": "The fridge was directly in front of you.", + "question_chinese": "从我现在站的位置，冰箱在哪个方向？", + "answer_chinese": "冰箱在你的正前方。", + "evidence": [ + { + "description": "The refrigerator is on the right.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The fridge was directly in front of you.", + "B": "Directly behind you.", + "C": "Directly to your left.", + "D": "The fridge was off to your right, so you would need to turn slightly to face it.", + "E": "The fridge was behind you rather than in front, so you would need to turn around to face it." + }, + "answer_label": "A", + "answer_text": "The fridge was directly in front of you.", + "q_id": "scene0732_00-0_2_T2", + "correct_answer": "The fridge was directly in front of you.", + "original_options": { + "A": "The fridge was directly in front of you.", + "B": "Directly behind you.", + "C": "Directly to your left.", + "D": "The fridge was off to your right, so you would need to turn slightly to face it.", + "E": "The fridge was behind you rather than in front, so you would need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 17, + 19 + ] + }, + { + "video_uid": "scene0732_00-0", + "video_length": "00:00:20", + "qid": "scene0732_00-0_2_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I’m standing right now, which direction is the fridge?", + "answer": "The fridge was to your right.", + "question_chinese": "从我现在站的位置，冰箱在哪个方向？", + "answer_chinese": "冰箱在你的右边。", + "evidence": [ + { + "description": "The refrigerator is on the right.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The fridge was to your left, so you would need to turn left to face it.", + "D": "The fridge was above you, not at your side, so you’d need to look upward to see it.", + "E": "The fridge was to your right." + }, + "answer_label": "E", + "answer_text": "The fridge was to your right.", + "q_id": "scene0732_00-0_2_T1", + "correct_answer": "The fridge was to your right.", + "original_options": { + "A": "In front of you.", + "B": "Behind you.", + "C": "The fridge was to your left, so you would need to turn left to face it.", + "D": "The fridge was above you, not at your side, so you’d need to look upward to see it.", + "E": "The fridge was to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 16, + 19 + ] + }, + { + "video_uid": "scene0732_00-0", + "video_length": "00:00:20", + "qid": "scene0732_00-0_7", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the food on the table?", + "answer": "Yes, It's very close to you.", + "question_chinese": "我现在的位置能够到桌上的食物吗？", + "answer_chinese": "能够到，它离你很近。", + "evidence": [ + { + "description": "close to the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, it's very close to you.", + "B": "Yes, It's very close to you." + }, + "answer_label": "B", + "answer_text": "Yes, It's very close to you.", + "q_id": "scene0732_00-0_7", + "correct_answer": "Yes, It's very close to you.", + "original_options": { + "A": "No, it's very close to you.", + "B": "Yes, It's very close to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0732_00-0", + "video_length": "00:00:20", + "qid": "scene0732_00-0_8", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the food on the table?", + "answer": "No, it's a little far away from you.", + "question_chinese": "我现在的位置能够到桌上的食物吗？", + "answer_chinese": "不能够到，它理你有点远。", + "evidence": [ + { + "description": "Far away from the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, it's a little far away from you.", + "B": "No, it's a little far away from you." + }, + "answer_label": "B", + "answer_text": "No, it's a little far away from you.", + "q_id": "scene0732_00-0_8", + "correct_answer": "No, it's a little far away from you.", + "original_options": { + "A": "Yes, it's a little far away from you.", + "B": "No, it's a little far away from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0732_00-0", + "video_length": "00:00:20", + "qid": "scene0732_00-0_2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the bottled alcohol?", + "answer": "No, it's a little far away from you.", + "question_chinese": "我现在的位置能够到酒吗？", + "answer_chinese": "不能够到，它理你有点远。", + "evidence": [ + { + "description": "The bottled alcohol is a little far away from the person.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, it's a little far away from you.", + "B": "No, it's a little far away from you." + }, + "answer_label": "B", + "answer_text": "No, it's a little far away from you.", + "q_id": "scene0732_00-0_2", + "correct_answer": "No, it's a little far away from you.", + "original_options": { + "A": "Yes, it's a little far away from you.", + "B": "No, it's a little far away from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0732_00-0", + "video_length": "00:00:20", + "qid": "scene0732_00-0_1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the bottled alcohol?", + "answer": "Yes, It's very close to you.", + "question_chinese": "我现在的位置能够到酒吗？", + "answer_chinese": "能够到，它离你很近。", + "evidence": [ + { + "description": "The bottled alcohol is very close to the person.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, It's very close to you.", + "B": "No, it's very close to you." + }, + "answer_label": "A", + "answer_text": "Yes, It's very close to you.", + "q_id": "scene0732_00-0_1", + "correct_answer": "Yes, It's very close to you.", + "original_options": { + "A": "Yes, It's very close to you.", + "B": "No, it's very close to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0732_00-0", + "video_length": "00:00:20", + "qid": "scene0732_00-0_9", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route should I take to get out of the kitchen?", + "answer": "You should move a short distance to the left, then go straight.", + "question_chinese": "我现在离开厨房应该走什么路线？", + "answer_chinese": "你应该向左移动一小段距离，然后直走。", + "evidence": [ + { + "description": "The door is on the front left side of the kitchen.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go right, then straight.", + "B": "Go straight, then left.", + "C": "Move a short distance to the right first, then continue straight ahead to reach the exit.", + "D": "You should move a short distance to the left, then go straight.", + "E": "Step backward a little, then turn left and go straight until you leave the kitchen." + }, + "answer_label": "D", + "answer_text": "You should move a short distance to the left, then go straight.", + "q_id": "scene0732_00-0_9", + "correct_answer": "You should move a short distance to the left, then go straight.", + "original_options": { + "A": "Go right, then straight.", + "B": "Go straight, then left.", + "C": "Move a short distance to the right first, then continue straight ahead to reach the exit.", + "D": "You should move a short distance to the left, then go straight.", + "E": "Step backward a little, then turn left and go straight until you leave the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 15 + ] + }, + { + "video_uid": "scene0732_00-0", + "video_length": "00:00:20", + "qid": "scene0732_00-0_10", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What route should I take to get out of the kitchen?", + "answer": "You should turn right and then go straight.", + "question_chinese": "我现在离开厨房应该走什么路线？", + "answer_chinese": "你应该向右拐然后直走。", + "evidence": [ + { + "description": "The door is on the right of the person.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn left, then go straight.", + "B": "Go straight, then turn right.", + "C": "Turn right, then back up a little before continuing straight out of the kitchen.", + "D": "Turn left at the doorway, then walk straight ahead until you reach the exit.", + "E": "You should turn right and then go straight." + }, + "answer_label": "E", + "answer_text": "You should turn right and then go straight.", + "q_id": "scene0732_00-0_10", + "correct_answer": "You should turn right and then go straight.", + "original_options": { + "A": "Turn left, then go straight.", + "B": "Go straight, then turn right.", + "C": "Turn right, then back up a little before continuing straight out of the kitchen.", + "D": "Turn left at the doorway, then walk straight ahead until you reach the exit.", + "E": "You should turn right and then go straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + } +] diff --git a/QAs/scene0733_00-0-mcq.json b/QAs/scene0733_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f486c914a2cb87495790433b23a096c414cba6ae --- /dev/null +++ b/QAs/scene0733_00-0-mcq.json @@ -0,0 +1,320 @@ +[ + { + "video_uid": "scene0733_00-0", + "video_length": "00:00:20", + "qid": "scene0733_00-0_3", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory from the moment I entered the door?", + "answer": "Walk forward first, then step backward.", + "question_chinese": "从进门开始，我的运动轨迹是怎么样的呢？", + "answer_chinese": "先向前走再向后退", + "evidence": [ + { + "description": "Walk forward for 0-12 seconds, then step back for 13-14 seconds.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Step backward first, then walk forward.", + "B": "Walk forward first, then step backward.", + "C": "Walk forward, then turn left.", + "D": "Walk forward first, then take two small steps forward instead of stepping backward.", + "E": "Walk forward first, then pivot around and continue moving forward rather than backing up." + }, + "answer_label": "B", + "answer_text": "Walk forward first, then step backward.", + "q_id": "scene0733_00-0_3", + "correct_answer": "Walk forward first, then step backward.", + "original_options": { + "A": "Step backward first, then walk forward.", + "B": "Walk forward first, then step backward.", + "C": "Walk forward, then turn left.", + "D": "Walk forward first, then take two small steps forward instead of stepping backward.", + "E": "Walk forward first, then pivot around and continue moving forward rather than backing up." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 14 + ] + }, + { + "video_uid": "scene0733_00-0", + "video_length": "00:00:20", + "qid": "scene0733_00-0_4", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory from the moment I entered the door?", + "answer": "First go forward, then step backward, and then turn right.", + "question_chinese": "从进门开始，我的运动轨迹是怎么样的呢？", + "answer_chinese": "先向前走再向后退再向右走", + "evidence": [ + { + "description": "Walk forward for 0-12 seconds, walk backward for 13-15 seconds, walk right for 16-19 seconds.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "First go forward, then step backward, and then turn right.", + "B": "Go forward, then turn right.", + "C": "Go forward, then turn left.", + "D": "First go forward, then step backward, and then turn left instead of turning right.", + "E": "First step backward, then go forward, and finally turn right at the end." + }, + "answer_label": "A", + "answer_text": "First go forward, then step backward, and then turn right.", + "q_id": "scene0733_00-0_4", + "correct_answer": "First go forward, then step backward, and then turn right.", + "original_options": { + "A": "First go forward, then step backward, and then turn right.", + "B": "Go forward, then turn right.", + "C": "Go forward, then turn left.", + "D": "First go forward, then step backward, and then turn left instead of turning right.", + "E": "First step backward, then go forward, and finally turn right at the end." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 13 + ] + }, + { + "video_uid": "scene0733_00-0", + "video_length": "00:00:20", + "qid": "scene0733_00-0_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What were there on the table just now?", + "answer": "Five things.", + "question_chinese": "刚才桌子上有几样东西", + "answer_chinese": "五样东西", + "evidence": [ + { + "description": "There are five items: three books, a bottle of wine, and an envelope.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Six things.", + "B": "Four things.", + "C": "There were six different items placed on the table just a moment ago, not five.", + "D": "Five things.", + "E": "There were four items on the table just now, so the count was less than five." + }, + "answer_label": "D", + "answer_text": "Five things.", + "q_id": "scene0733_00-0_1", + "correct_answer": "Five things.", + "original_options": { + "A": "Six things.", + "B": "Four things.", + "C": "There were six different items placed on the table just a moment ago, not five.", + "D": "Five things.", + "E": "There were four items on the table just now, so the count was less than five." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 16 + ] + }, + { + "video_uid": "scene0733_00-0", + "video_length": "00:00:20", + "qid": "scene0733_00-0_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What were there on the table just now?", + "answer": "Five things.", + "question_chinese": "刚才桌子上有几样东西", + "answer_chinese": "五样东西", + "evidence": [ + { + "description": "There are five items: three books, a bottle of wine, and an envelope.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four things.", + "B": "Six things.", + "C": "There were seven separate items on the table just now, which is more than five.", + "D": "There was only one single item on the table just now, far fewer than five.", + "E": "Five things." + }, + "answer_label": "E", + "answer_text": "Five things.", + "q_id": "scene0733_00-0_2", + "correct_answer": "Five things.", + "original_options": { + "A": "Four things.", + "B": "Six things.", + "C": "There were seven separate items on the table just now, which is more than five.", + "D": "There was only one single item on the table just now, far fewer than five.", + "E": "Five things." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "scene0733_00-0", + "video_length": "00:00:20", + "qid": "scene0733_00-0_7", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the window located at this moment?", + "answer": "on your right side.", + "question_chinese": "此时窗户在什么位置?", + "answer_chinese": "在你的右边", + "evidence": [ + { + "description": "On the right side.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your left side.", + "B": "In front of you.", + "C": "The window is positioned behind you, slightly off to your left side.", + "D": "on your right side.", + "E": "The window is straight ahead of you, not to the right at this moment." + }, + "answer_label": "D", + "answer_text": "on your right side.", + "q_id": "scene0733_00-0_7", + "correct_answer": "on your right side.", + "original_options": { + "A": "On your left side.", + "B": "In front of you.", + "C": "The window is positioned behind you, slightly off to your left side.", + "D": "on your right side.", + "E": "The window is straight ahead of you, not to the right at this moment." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 14 + ] + }, + { + "video_uid": "scene0733_00-0", + "video_length": "00:00:20", + "qid": "scene0733_00-0_8", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the window located at this moment?", + "answer": "Right in front of you.", + "question_chinese": "此时窗户在什么位置?", + "answer_chinese": "在你的正前方", + "evidence": [ + { + "description": "The window is directly in front.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Slightly to your right.", + "B": "Right in front of you.", + "C": "Just behind you.", + "D": "The window is off to your left side rather than directly in front of you.", + "E": "The window is behind your right shoulder, so it is not right in front of you." + }, + "answer_label": "B", + "answer_text": "Right in front of you.", + "q_id": "scene0733_00-0_8", + "correct_answer": "Right in front of you.", + "original_options": { + "A": "Slightly to your right.", + "B": "Right in front of you.", + "C": "Just behind you.", + "D": "The window is off to your left side rather than directly in front of you.", + "E": "The window is behind your right shoulder, so it is not right in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 16 + ] + } +] diff --git a/QAs/scene0734_00-0-mcq.json b/QAs/scene0734_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..67bda869191ced1660fe34d41192effe0e9067ef --- /dev/null +++ b/QAs/scene0734_00-0-mcq.json @@ -0,0 +1,572 @@ +[ + { + "video_uid": "scene0734_00-0", + "video_length": "00:00:20", + "qid": "scene0734_00-0_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From the moment I entered the room, what was my movement trajectory?", + "answer": "As you enter the door, you start walking straight ahead.", + "question_chinese": "从我进门开始我的运动轨迹？", + "answer_chinese": "你进门开始在直行", + "evidence": [ + { + "description": "You enter and start walking straight.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You start to walk left.", + "B": "You step backward first.", + "C": "As you enter the door, you turn right and head that way instead of walking straight ahead.", + "D": "As you enter the door, you start walking straight ahead.", + "E": "As you enter the door, you pause and back up toward the wall rather than moving forward." + }, + "answer_label": "D", + "answer_text": "As you enter the door, you start walking straight ahead.", + "q_id": "scene0734_00-0_6", + "correct_answer": "As you enter the door, you start walking straight ahead.", + "original_options": { + "A": "You start to walk left.", + "B": "You step backward first.", + "C": "As you enter the door, you turn right and head that way instead of walking straight ahead.", + "D": "As you enter the door, you start walking straight ahead.", + "E": "As you enter the door, you pause and back up toward the wall rather than moving forward." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 17, + 17 + ] + }, + { + "video_uid": "scene0734_00-0", + "video_length": "00:00:20", + "qid": "scene0734_00-0_7", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "From the moment I entered the room, what was my movement trajectory?", + "answer": "As you enter the door, you start walking straight and then turn left.", + "question_chinese": "从我进门开始我的运动轨迹？", + "answer_chinese": "你进门开始直行并左转", + "evidence": [ + { + "description": "You enter the door, go straight ahead, and then turn left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk straight, then turn right.", + "B": "Step in, then turn around.", + "C": "After you enter, you head straight ahead and then veer to the right side.", + "D": "As you come through the doorway, you move forward briefly before backing up toward the entrance.", + "E": "As you enter the door, you start walking straight and then turn left." + }, + "answer_label": "E", + "answer_text": "As you enter the door, you start walking straight and then turn left.", + "q_id": "scene0734_00-0_7", + "correct_answer": "As you enter the door, you start walking straight and then turn left.", + "original_options": { + "A": "Walk straight, then turn right.", + "B": "Step in, then turn around.", + "C": "After you enter, you head straight ahead and then veer to the right side.", + "D": "As you come through the doorway, you move forward briefly before backing up toward the entrance.", + "E": "As you enter the door, you start walking straight and then turn left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "scene0734_00-0", + "video_length": "00:00:20", + "qid": "scene0734_00-0_8", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the garbage now?", + "answer": "Right beneath you.", + "question_chinese": "垃圾现在在什么位置？", + "answer_chinese": "在你的正下方", + "evidence": [ + { + "description": "You walked forward, and the garbage is right beneath you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right beneath you.", + "B": "To your left.", + "C": "In front of you.", + "D": "It is behind you near the corner of the room, not directly under your feet.", + "E": "It is off to your right side a short distance away, rather than positioned directly beneath you." + }, + "answer_label": "A", + "answer_text": "Right beneath you.", + "q_id": "scene0734_00-0_8", + "correct_answer": "Right beneath you.", + "original_options": { + "A": "Right beneath you.", + "B": "To your left.", + "C": "In front of you.", + "D": "It is behind you near the corner of the room, not directly under your feet.", + "E": "It is off to your right side a short distance away, rather than positioned directly beneath you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 17 + ] + }, + { + "video_uid": "scene0734_00-0", + "video_length": "00:00:20", + "qid": "scene0734_00-0_9", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the garbage now?", + "answer": "At the bottom right of you.", + "question_chinese": "垃圾现在在什么位置？", + "answer_chinese": "在你的右下方", + "evidence": [ + { + "description": "You walked forward, then turned right, so you are at the lower right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Above you on the shelf.", + "B": "At the bottom right of you.", + "C": "Behind you on the floor.", + "D": "It is on your right side in front of you, instead of down at your bottom right.", + "E": "It is beside you inside the nearby bag, rather than lying at the bottom right of you." + }, + "answer_label": "B", + "answer_text": "At the bottom right of you.", + "q_id": "scene0734_00-0_9", + "correct_answer": "At the bottom right of you.", + "original_options": { + "A": "Above you on the shelf.", + "B": "At the bottom right of you.", + "C": "Behind you on the floor.", + "D": "It is on your right side in front of you, instead of down at your bottom right.", + "E": "It is beside you inside the nearby bag, rather than lying at the bottom right of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "scene0734_00-0", + "video_length": "00:00:20", + "qid": "scene0734_00-0_1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to go to bed, which way should I take?", + "answer": "Walk to your left.", + "question_chinese": "我想上床的话，应该怎么走？", + "answer_chinese": "向你的左边走", + "evidence": [ + { + "description": "The bed is to your left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go straight ahead.", + "B": "Walk to your right.", + "C": "Turn around and walk backward a few steps to reach the bed instead of heading left.", + "D": "Keep moving forward toward the bed in front of you rather than taking a left turn.", + "E": "Walk to your left." + }, + "answer_label": "E", + "answer_text": "Walk to your left.", + "q_id": "scene0734_00-0_1", + "correct_answer": "Walk to your left.", + "original_options": { + "A": "Go straight ahead.", + "B": "Walk to your right.", + "C": "Turn around and walk backward a few steps to reach the bed instead of heading left.", + "D": "Keep moving forward toward the bed in front of you rather than taking a left turn.", + "E": "Walk to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 16, + 16 + ] + }, + { + "video_uid": "scene0734_00-0", + "video_length": "00:00:20", + "qid": "scene0734_00-0_10", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to go to bed, which way should I take?", + "answer": "Walk forward towards you.", + "question_chinese": "我想上床的话，应该怎么走？", + "answer_chinese": "向你的前边走", + "evidence": [ + { + "description": "The bed is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk forward away from you.", + "B": "Walk backward towards you.", + "C": "Walk forward towards you.", + "D": "Move forward slightly to the right rather than walking forward toward you.", + "E": "Step forward and angle to the left a bit instead of going straight forward toward you." + }, + "answer_label": "C", + "answer_text": "Walk forward towards you.", + "q_id": "scene0734_00-0_10", + "correct_answer": "Walk forward towards you.", + "original_options": { + "A": "Walk forward away from you.", + "B": "Walk backward towards you.", + "C": "Walk forward towards you.", + "D": "Move forward slightly to the right rather than walking forward toward you.", + "E": "Step forward and angle to the left a bit instead of going straight forward toward you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 12, + 16 + ] + }, + { + "video_uid": "scene0734_00-0", + "video_length": "00:00:20", + "qid": "scene0734_00-0_2_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many desk lamps can I see in the room right now?", + "answer": "You could see one table lamp in the room.", + "question_chinese": "目前房间里我可以看到几盏台灯？", + "answer_chinese": "你可以在房间里看到一盏台灯", + "evidence": [ + { + "description": "A lamp beside the TV", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You could see one table lamp in the room.", + "B": "Zero table lamps.", + "C": "Two table lamps.", + "D": "You could see three table lamps in the room right now, not just a single one.", + "E": "You could see four table lamps in the room at the moment, rather than only one." + }, + "answer_label": "A", + "answer_text": "You could see one table lamp in the room.", + "q_id": "scene0734_00-0_2_T2", + "correct_answer": "You could see one table lamp in the room.", + "original_options": { + "A": "You could see one table lamp in the room.", + "B": "Zero table lamps.", + "C": "Two table lamps.", + "D": "You could see three table lamps in the room right now, not just a single one.", + "E": "You could see four table lamps in the room at the moment, rather than only one." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "scene0734_00-0", + "video_length": "00:00:20", + "qid": "scene0734_00-0_2_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many desk lamps can I see in the room right now?", + "answer": "You could see three table lamps in the room.", + "question_chinese": "目前房间里我可以看到几盏台灯？", + "answer_chinese": "你可以在房间里看到三盏台灯", + "evidence": [ + { + "description": "A lamp beside the TV", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "Two desk lamps are by the bedhead.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two table lamps.", + "B": "Four table lamps.", + "C": "You could see one table lamp in the room right now, not as many as three.", + "D": "You could see five table lamps in the room at the moment, rather than exactly three.", + "E": "You could see three table lamps in the room." + }, + "answer_label": "E", + "answer_text": "You could see three table lamps in the room.", + "q_id": "scene0734_00-0_2_T1", + "correct_answer": "You could see three table lamps in the room.", + "original_options": { + "A": "Two table lamps.", + "B": "Four table lamps.", + "C": "You could see one table lamp in the room right now, not as many as three.", + "D": "You could see five table lamps in the room at the moment, rather than exactly three.", + "E": "You could see three table lamps in the room." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 16 + ] + }, + { + "video_uid": "scene0734_00-0", + "video_length": "00:00:20", + "qid": "scene0734_00-0_3_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Compared to when I first entered, is the TV closer to me now or farther away?", + "answer": "It got closer to you.", + "question_chinese": "相比较我进门的时候 电视离我远了近了？", + "answer_chinese": "离你变近了", + "evidence": [ + { + "description": "I am walking in, so the TV is closer.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It got closer to you.", + "B": "It got farther from you." + }, + "answer_label": "A", + "answer_text": "It got closer to you.", + "q_id": "scene0734_00-0_3_T1", + "correct_answer": "It got closer to you.", + "original_options": { + "A": "It got closer to you.", + "B": "It got farther from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "scene0734_00-0", + "video_length": "00:00:20", + "qid": "scene0734_00-0_3_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Compared to when I first entered, is the TV closer to me now or farther away?", + "answer": "It got farther from you.", + "question_chinese": "相比较我进门的时候 电视离我远了近了？", + "answer_chinese": "离你变远了", + "evidence": [ + { + "description": "I am walking into the house, so I am getting closer.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + }, + { + "description": "I walk closer to the bed.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It got closer to you.", + "B": "It got farther from you." + }, + "answer_label": "B", + "answer_text": "It got farther from you.", + "q_id": "scene0734_00-0_3_T2", + "correct_answer": "It got farther from you.", + "original_options": { + "A": "It got closer to you.", + "B": "It got farther from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + }, + { + "video_uid": "scene0734_00-0", + "video_length": "00:00:20", + "qid": "scene0734_00-0_5", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Compared to when I first entered, has the TV moved closer or farther from me?", + "answer": "Getting closer to you.", + "question_chinese": "相比较我进门的时候 电视离我远了近了？", + "answer_chinese": "离你变近了", + "evidence": [ + { + "description": "I am walking into the house, so I am getting closer.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Getting closer to you.", + "B": "It moved farther from you." + }, + "answer_label": "A", + "answer_text": "Getting closer to you.", + "q_id": "scene0734_00-0_5", + "correct_answer": "Getting closer to you.", + "original_options": { + "A": "Getting closer to you.", + "B": "It moved farther from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5 + ] + } +] diff --git a/QAs/scene0735_00-0-mcq.json b/QAs/scene0735_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5545643eb20ae9adffc58d795afd9edd0fff6525 --- /dev/null +++ b/QAs/scene0735_00-0-mcq.json @@ -0,0 +1,408 @@ +[ + { + "video_uid": "scene0735_00-0", + "video_length": "00:00:20", + "qid": "scene0735_00-0_5", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Just now when I was looking in the mirror, could I sit directly on the toilet?", + "answer": "No, you can't.", + "question_chinese": "刚才我正对镜子时，我可以直接坐在马桶上吗？", + "answer_chinese": "不，你不可以", + "evidence": [ + { + "description": "At this moment, information about the toilet appeared.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No, you can't.", + "q_id": "scene0735_00-0_5", + "correct_answer": "No, you can't.", + "original_options": { + "A": "No, you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0735_00-0", + "video_length": "00:00:20", + "qid": "scene0735_00-0_6", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Just now when I was looking in the mirror, could I sit directly on the toilet?", + "answer": "No, you can't.", + "question_chinese": "刚才我正对镜子时，我可以直接坐在马桶上吗？", + "answer_chinese": "不，你不可以", + "evidence": [ + { + "description": "A toilet appeared during this period.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "answer_label": "B", + "answer_text": "No, you can't.", + "q_id": "scene0735_00-0_6", + "correct_answer": "No, you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0735_00-0", + "video_length": "00:00:20", + "qid": "scene0735_00-0_8", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to turn on the tap, how should I go?", + "answer": "you should go forward.", + "question_chinese": "如果我想打开水龙头 我应该怎么走？", + "answer_chinese": "你应该向前走", + "evidence": [ + { + "description": "At this moment, the faucet is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "you should go forward.", + "B": "Move slightly left.", + "C": "Step to the right.", + "D": "Back up a little, then reach out to the tap to turn it on.", + "E": "Turn around and approach the tap from behind before trying to switch it on." + }, + "answer_label": "A", + "answer_text": "you should go forward.", + "q_id": "scene0735_00-0_8", + "correct_answer": "you should go forward.", + "original_options": { + "A": "you should go forward.", + "B": "Move slightly left.", + "C": "Step to the right.", + "D": "Back up a little, then reach out to the tap to turn it on.", + "E": "Turn around and approach the tap from behind before trying to switch it on." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 14 + ] + }, + { + "video_uid": "scene0735_00-0", + "video_length": "00:00:20", + "qid": "scene0735_00-0_7", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to turn on the tap, how should I go?", + "answer": "turn left.", + "question_chinese": "如果我想打开水龙头 我应该怎么走？", + "answer_chinese": "向左走", + "evidence": [ + { + "description": "At this time, the faucet is on your left side.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go straight.", + "B": "Turn right.", + "C": "Turn around first, then walk straight until you reach the tap to turn it on.", + "D": "turn left.", + "E": "Go backward a few steps, then adjust your position before turning the tap." + }, + "answer_label": "D", + "answer_text": "turn left.", + "q_id": "scene0735_00-0_7", + "correct_answer": "turn left.", + "original_options": { + "A": "Go straight.", + "B": "Turn right.", + "C": "Turn around first, then walk straight until you reach the tap to turn it on.", + "D": "turn left.", + "E": "Go backward a few steps, then adjust your position before turning the tap." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 13 + ] + }, + { + "video_uid": "scene0735_00-0", + "video_length": "00:00:20", + "qid": "scene0735_00-0_9", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What has been my movement trajectory up to now?", + "answer": "Turned 180 degrees clockwise.", + "question_chinese": "截至目前我的运动轨迹是什么？", + "answer_chinese": "顺时针转了180度", + "evidence": [ + { + "description": "During this period, my movement trajectory appeared, rotating clockwise.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turned 90 degrees clockwise.", + "B": "Turned 180 degrees counterclockwise.", + "C": "Turned 270 degrees clockwise, ending up facing a different direction than when you started.", + "D": "Turned a full 360 degrees counterclockwise, returning to the same facing direction as before.", + "E": "Turned 180 degrees clockwise." + }, + "answer_label": "E", + "answer_text": "Turned 180 degrees clockwise.", + "q_id": "scene0735_00-0_9", + "correct_answer": "Turned 180 degrees clockwise.", + "original_options": { + "A": "Turned 90 degrees clockwise.", + "B": "Turned 180 degrees counterclockwise.", + "C": "Turned 270 degrees clockwise, ending up facing a different direction than when you started.", + "D": "Turned a full 360 degrees counterclockwise, returning to the same facing direction as before.", + "E": "Turned 180 degrees clockwise." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "scene0735_00-0", + "video_length": "00:00:20", + "qid": "scene0735_00-0_10", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my current movement trajectory as of now?", + "answer": "Turned clockwise by 180 degrees and then walked forward a few steps.", + "question_chinese": "截至目前我的运动轨迹是什么？", + "answer_chinese": "顺时针转了180度并向前走了几步", + "evidence": [ + { + "description": "During this time period, my movement trajectory appeared, rotating clockwise and moving forward.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turned 90 degrees clockwise, then forward.", + "B": "Turned 180 degrees counterclockwise, then forward.", + "C": "Turned clockwise by 180 degrees and then walked forward a few steps.", + "D": "Turned clockwise by 180 degrees and then walked backward a few steps instead of moving forward.", + "E": "Turned clockwise by 90 degrees and then walked forward a few steps, not completing a half turn." + }, + "answer_label": "C", + "answer_text": "Turned clockwise by 180 degrees and then walked forward a few steps.", + "q_id": "scene0735_00-0_10", + "correct_answer": "Turned clockwise by 180 degrees and then walked forward a few steps.", + "original_options": { + "A": "Turned 90 degrees clockwise, then forward.", + "B": "Turned 180 degrees counterclockwise, then forward.", + "C": "Turned clockwise by 180 degrees and then walked forward a few steps.", + "D": "Turned clockwise by 180 degrees and then walked backward a few steps instead of moving forward.", + "E": "Turned clockwise by 90 degrees and then walked forward a few steps, not completing a half turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "scene0735_00-0", + "video_length": "00:00:20", + "qid": "scene0735_00-0_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many towels are hung on the railing above the toilet?", + "answer": "three.", + "question_chinese": "马桶上面的栏杆挂了几条毛巾？", + "answer_chinese": "三条", + "evidence": [ + { + "description": "There are three towels on the railing.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "three.", + "B": "two.", + "C": "four.", + "D": "Only two towels hang on the railing above the toilet, not three.", + "E": "Four towels are draped across the railing above the toilet, spaced evenly along the bar." + }, + "answer_label": "A", + "answer_text": "three.", + "q_id": "scene0735_00-0_1", + "correct_answer": "three.", + "original_options": { + "A": "three.", + "B": "two.", + "C": "four.", + "D": "Only two towels hang on the railing above the toilet, not three.", + "E": "Four towels are draped across the railing above the toilet, spaced evenly along the bar." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 15 + ] + }, + { + "video_uid": "scene0735_00-0", + "video_length": "00:00:20", + "qid": "scene0735_00-0_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many towels are hung on the railing above the toilet?", + "answer": "three.", + "question_chinese": "马桶上面的栏杆挂了几条毛巾？", + "answer_chinese": "三条", + "evidence": [ + { + "description": "There are three towels on the railing.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "two.", + "B": "four.", + "C": "Five towels are hung on the railing above the toilet in a single row.", + "D": "Six towels are hanging across the railing above the toilet, covering most of it.", + "E": "three." + }, + "answer_label": "E", + "answer_text": "three.", + "q_id": "scene0735_00-0_2", + "correct_answer": "three.", + "original_options": { + "A": "two.", + "B": "four.", + "C": "Five towels are hung on the railing above the toilet in a single row.", + "D": "Six towels are hanging across the railing above the toilet, covering most of it.", + "E": "three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + } +] diff --git a/QAs/scene0736_00-0-mcq.json b/QAs/scene0736_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..b7ee6ea9d95828a58f38eff372b2a166d70c57d0 --- /dev/null +++ b/QAs/scene0736_00-0-mcq.json @@ -0,0 +1,478 @@ +[ + { + "video_uid": "scene0736_00-0", + "video_length": "00:00:20", + "qid": "scene0736_00-0_5_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the window?", + "answer": "No, you can’t—the window is in front of you.", + "question_chinese": "我可以触摸到窗户吗？", + "answer_chinese": "不，你不能，窗户在你前边。", + "evidence": [ + { + "description": "Information about the window appeared at this time.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t—the window is in front of you.", + "B": "Yes, you can—the window is in front of you." + }, + "answer_label": "A", + "answer_text": "No, you can’t—the window is in front of you.", + "q_id": "scene0736_00-0_5_T1", + "correct_answer": "No, you can’t—the window is in front of you.", + "original_options": { + "A": "No, you can’t—the window is in front of you.", + "B": "Yes, you can—the window is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "scene0736_00-0", + "video_length": "00:00:20", + "qid": "scene0736_00-0_5_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the window?", + "answer": "No, you can’t—the window is to your left.", + "question_chinese": "我可以触摸到窗户吗？", + "answer_chinese": "不，你不能，窗户在你左边。", + "evidence": [ + { + "description": "Information about the window appeared at this time.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can—the window is to your left.", + "B": "No, you can’t—the window is to your left." + }, + "answer_label": "B", + "answer_text": "No, you can’t—the window is to your left.", + "q_id": "scene0736_00-0_5_T2", + "correct_answer": "No, you can’t—the window is to your left.", + "original_options": { + "A": "Yes, you can—the window is to your left.", + "B": "No, you can’t—the window is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0736_00-0", + "video_length": "00:00:20", + "qid": "scene0736_00-0_3_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there a green bag on the table in front of me?", + "answer": "No—there isn’t a green bag on the table in front of you.", + "question_chinese": "我面前桌子上有绿色的袋子吗？", + "answer_chinese": "没有，你面前的桌子上没有绿袋子", + "evidence": [ + { + "description": "There are only red and white items on the table, no green ones.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—there is a green bag on the table in front of you.", + "B": "No—there isn’t a green bag on the table in front of you." + }, + "answer_label": "B", + "answer_text": "No—there isn’t a green bag on the table in front of you.", + "q_id": "scene0736_00-0_3_T2", + "correct_answer": "No—there isn’t a green bag on the table in front of you.", + "original_options": { + "A": "Yes—there is a green bag on the table in front of you.", + "B": "No—there isn’t a green bag on the table in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "scene0736_00-0", + "video_length": "00:00:20", + "qid": "scene0736_00-0_3_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is there a green bag on the table in front of me?", + "answer": "No—the table is to your right.", + "question_chinese": "我面前桌子上有绿色的袋子吗？", + "answer_chinese": "没有，桌子在你的右边。", + "evidence": [ + { + "description": "There are only red and white items on the table, no green ones.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No—the table is to your right.", + "B": "Yes—there’s a green bag on the table in front of you." + }, + "answer_label": "A", + "answer_text": "No—the table is to your right.", + "q_id": "scene0736_00-0_3_T1", + "correct_answer": "No—the table is to your right.", + "original_options": { + "A": "No—the table is to your right.", + "B": "Yes—there’s a green bag on the table in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "scene0736_00-0", + "video_length": "00:00:20", + "qid": "scene0736_00-0_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs have you seen so far?", + "answer": "two.", + "question_chinese": "截至目前看到几把椅子？", + "answer_chinese": "2", + "evidence": [ + { + "description": "There are currently only two chairs in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "two.", + "B": "one.", + "C": "three.", + "D": "You’ve only seen one chair so far in the video up to this point.", + "E": "You have seen three chairs so far, including one earlier and two more later." + }, + "answer_label": "A", + "answer_text": "two.", + "q_id": "scene0736_00-0_1", + "correct_answer": "two.", + "original_options": { + "A": "two.", + "B": "one.", + "C": "three.", + "D": "You’ve only seen one chair so far in the video up to this point.", + "E": "You have seen three chairs so far, including one earlier and two more later." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "scene0736_00-0", + "video_length": "00:00:20", + "qid": "scene0736_00-0_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs have you seen so far?", + "answer": "three.", + "question_chinese": "截至目前看到几把椅子？", + "answer_chinese": "3", + "evidence": [ + { + "description": "There are currently three chairs in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "one.", + "B": "five.", + "C": "You’ve seen seven chairs so far in total across all the rooms visited.", + "D": "You have seen eight chairs so far, with several grouped together and others encountered separately.", + "E": "three." + }, + "answer_label": "E", + "answer_text": "three.", + "q_id": "scene0736_00-0_2", + "correct_answer": "three.", + "original_options": { + "A": "one.", + "B": "five.", + "C": "You’ve seen seven chairs so far in total across all the rooms visited.", + "D": "You have seen eight chairs so far, with several grouped together and others encountered separately.", + "E": "three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "scene0736_00-0", + "video_length": "00:00:20", + "qid": "scene0736_00-0_7", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the whiteboard located relative to me at this moment?", + "answer": "Right behind you.", + "question_chinese": "此时白板在相对于我的什么位置？", + "answer_chinese": "在你的正后方", + "evidence": [ + { + "description": "At this time, a whiteboard appeared on the left wall.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "In front of you.", + "C": "It’s behind you but slightly to the left, not directly centered behind your back.", + "D": "It’s on your right side in front of you, visible when you turn your head slightly right.", + "E": "Right behind you." + }, + "answer_label": "E", + "answer_text": "Right behind you.", + "q_id": "scene0736_00-0_7", + "correct_answer": "Right behind you.", + "original_options": { + "A": "To your left.", + "B": "In front of you.", + "C": "It’s behind you but slightly to the left, not directly centered behind your back.", + "D": "It’s on your right side in front of you, visible when you turn your head slightly right.", + "E": "Right behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 18 + ] + }, + { + "video_uid": "scene0736_00-0", + "video_length": "00:00:20", + "qid": "scene0736_00-0_8", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the whiteboard located relative to me at this moment?", + "answer": "On your right rear side.", + "question_chinese": "此时白板在相对于我的什么位置？", + "answer_chinese": "在你的右后方", + "evidence": [ + { + "description": "At this moment, a whiteboard appeared on the left wall.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On your left rear side.", + "B": "To your right at the front.", + "C": "It’s directly behind you rather than off to your right rear side.", + "D": "On your right rear side.", + "E": "It’s in front of you and slightly left, so you’d face it by turning a bit left." + }, + "answer_label": "D", + "answer_text": "On your right rear side.", + "q_id": "scene0736_00-0_8", + "correct_answer": "On your right rear side.", + "original_options": { + "A": "On your left rear side.", + "B": "To your right at the front.", + "C": "It’s directly behind you rather than off to your right rear side.", + "D": "On your right rear side.", + "E": "It’s in front of you and slightly left, so you’d face it by turning a bit left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 13, + 19 + ] + }, + { + "video_uid": "scene0736_00-0", + "video_length": "00:00:20", + "qid": "scene0736_00-0_2_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the window closer to me now, or farther away?", + "answer": "It’s closer to you now—the window is in front of you.", + "question_chinese": "窗户离我是更近了还是更远了？", + "answer_chinese": "离你更近了，窗户在你的前方", + "evidence": [ + { + "description": "You moved forward.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It’s farther from you now—the window is behind you.", + "B": "It’s closer to you now—the window is in front of you." + }, + "answer_label": "B", + "answer_text": "It’s closer to you now—the window is in front of you.", + "q_id": "scene0736_00-0_2_T1", + "correct_answer": "It’s closer to you now—the window is in front of you.", + "original_options": { + "A": "It’s farther from you now—the window is behind you.", + "B": "It’s closer to you now—the window is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "scene0736_00-0", + "video_length": "00:00:20", + "qid": "scene0736_00-0_2_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Is the window closer to me now, or farther away?", + "answer": "It’s farther from you now—the window is behind you.", + "question_chinese": "窗户离我是更近了还是更远了？", + "answer_chinese": "离你更远了，窗户在你的后方", + "evidence": [ + { + "description": "You moved forward.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "It’s farther from you now—the window is behind you.", + "B": "It’s closer to you now—the window is in front of you." + }, + "answer_label": "A", + "answer_text": "It’s farther from you now—the window is behind you.", + "q_id": "scene0736_00-0_2_T2", + "correct_answer": "It’s farther from you now—the window is behind you.", + "original_options": { + "A": "It’s farther from you now—the window is behind you.", + "B": "It’s closer to you now—the window is in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + } +] diff --git a/QAs/scene0737_00-0-mcq.json b/QAs/scene0737_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..195f64044c58f3ed32d7d9cb4e3d327858205bc9 --- /dev/null +++ b/QAs/scene0737_00-0-mcq.json @@ -0,0 +1,390 @@ +[ + { + "video_uid": "scene0737_00-0", + "video_length": "00:00:20", + "qid": "scene0737_00-0_2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to go out, in which direction should I walk at this moment?", + "answer": "Move backward.", + "question_chinese": "如果我想出去此刻我应该向哪个方向走？", + "answer_chinese": "向后方出去", + "evidence": [ + { + "description": "At this moment, information about the door appeared, and you can go out through the door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Move backward.", + "B": "Move forward.", + "C": "Go to the left.", + "D": "Walk straight ahead past the chair to reach the exit from where you’re standing now.", + "E": "Turn right and walk forward a few steps, since the way out is on that side." + }, + "answer_label": "A", + "answer_text": "Move backward.", + "q_id": "scene0737_00-0_2", + "correct_answer": "Move backward.", + "original_options": { + "A": "Move backward.", + "B": "Move forward.", + "C": "Go to the left.", + "D": "Walk straight ahead past the chair to reach the exit from where you’re standing now.", + "E": "Turn right and walk forward a few steps, since the way out is on that side." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 4, + 16, + 16 + ] + }, + { + "video_uid": "scene0737_00-0", + "video_length": "00:00:20", + "qid": "scene0737_00-0_3", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "If I want to go out, in which direction should I walk at this moment?", + "answer": "Go to the left.", + "question_chinese": "如果我想出去此刻我应该向哪个方向走？", + "answer_chinese": "向左方出去", + "evidence": [ + { + "description": "At this moment, information about the door appeared, and you can go out through the door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk straight ahead.", + "B": "Move to the right.", + "C": "Go to the left.", + "D": "Head backward from here, because the way out is behind your current position.", + "E": "Go forward and angle slightly right, since the exit isn’t on your left side." + }, + "answer_label": "C", + "answer_text": "Go to the left.", + "q_id": "scene0737_00-0_3", + "correct_answer": "Go to the left.", + "original_options": { + "A": "Walk straight ahead.", + "B": "Move to the right.", + "C": "Go to the left.", + "D": "Head backward from here, because the way out is behind your current position.", + "E": "Go forward and angle slightly right, since the exit isn’t on your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 15 + ] + }, + { + "video_uid": "scene0737_00-0", + "video_length": "00:00:20", + "qid": "scene0737_00-0_3_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I just see a trash can a moment ago?", + "answer": "Yes, it’s to your back-right.", + "question_chinese": "我在刚才看到一个垃圾桶吗？", + "answer_chinese": "是的 ，它在你的右后方。", + "evidence": [ + { + "description": "At this moment, a trash can appeared on the ground.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you didn’t see a trash can a moment ago.", + "B": "Yes, it’s to your back-right." + }, + "answer_label": "B", + "answer_text": "Yes, it’s to your back-right.", + "q_id": "scene0737_00-0_3_T1", + "correct_answer": "Yes, it’s to your back-right.", + "original_options": { + "A": "No, you didn’t see a trash can a moment ago.", + "B": "Yes, it’s to your back-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "scene0737_00-0", + "video_length": "00:00:20", + "qid": "scene0737_00-0_3_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Did I just see a trash can a moment ago?", + "answer": "Yes, it’s directly behind you.", + "question_chinese": "我在刚才看到一个垃圾桶吗？", + "answer_chinese": "是的，它在你的正后方", + "evidence": [ + { + "description": "At this moment, a trash can appeared underground.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, it’s directly behind you.", + "B": "No, you didn’t see a trash can a moment ago." + }, + "answer_label": "A", + "answer_text": "Yes, it’s directly behind you.", + "q_id": "scene0737_00-0_3_T2", + "correct_answer": "Yes, it’s directly behind you.", + "original_options": { + "A": "Yes, it’s directly behind you.", + "B": "No, you didn’t see a trash can a moment ago." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "scene0737_00-0", + "video_length": "00:00:20", + "qid": "scene0737_00-0_1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "what has been the trajectory of my movements?", + "answer": "Rotate counterclockwise by 360 degrees.", + "question_chinese": "我的运动轨迹是怎么样的？", + "answer_chinese": "逆时针旋转360度", + "evidence": [ + { + "description": "Counterclockwise rotation of 360 degrees.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Rotate clockwise 360 degrees.", + "B": "Rotate counterclockwise 180 degrees.", + "C": "Turn clockwise in place until you face the original direction, but stop after a half rotation.", + "D": "Spin counterclockwise, then take a step forward and end facing slightly left of where you started.", + "E": "Rotate counterclockwise by 360 degrees." + }, + "answer_label": "E", + "answer_text": "Rotate counterclockwise by 360 degrees.", + "q_id": "scene0737_00-0_1", + "correct_answer": "Rotate counterclockwise by 360 degrees.", + "original_options": { + "A": "Rotate clockwise 360 degrees.", + "B": "Rotate counterclockwise 180 degrees.", + "C": "Turn clockwise in place until you face the original direction, but stop after a half rotation.", + "D": "Spin counterclockwise, then take a step forward and end facing slightly left of where you started.", + "E": "Rotate counterclockwise by 360 degrees." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 16 + ] + }, + { + "video_uid": "scene0737_00-0", + "video_length": "00:00:20", + "qid": "scene0737_00-0_10", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "what has been the trajectory of my movements?", + "answer": "Rotate counterclockwise by 360 degrees and then rotate counterclockwise upwards.", + "question_chinese": "我的运动轨迹是怎么样的？", + "answer_chinese": "逆时针旋转360度然后向上方逆时针旋转", + "evidence": [ + { + "description": "Rotate counterclockwise 360 degrees and then rotate counterclockwise upwards.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Rotate counterclockwise by 360 degrees and then rotate counterclockwise upwards.", + "B": "Rotate counterclockwise by 360 degrees.", + "C": "Rotate clockwise by 360 degrees.", + "D": "Rotate counterclockwise by 360 degrees and then rotate clockwise upwards, reversing the second motion.", + "E": "Rotate counterclockwise by 180 degrees and then rotate counterclockwise upwards, making only a half turn first." + }, + "answer_label": "A", + "answer_text": "Rotate counterclockwise by 360 degrees and then rotate counterclockwise upwards.", + "q_id": "scene0737_00-0_10", + "correct_answer": "Rotate counterclockwise by 360 degrees and then rotate counterclockwise upwards.", + "original_options": { + "A": "Rotate counterclockwise by 360 degrees and then rotate counterclockwise upwards.", + "B": "Rotate counterclockwise by 360 degrees.", + "C": "Rotate clockwise by 360 degrees.", + "D": "Rotate counterclockwise by 360 degrees and then rotate clockwise upwards, reversing the second motion.", + "E": "Rotate counterclockwise by 180 degrees and then rotate counterclockwise upwards, making only a half turn first." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "scene0737_00-0", + "video_length": "00:00:20", + "qid": "scene0737_00-0_5_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the door?", + "answer": "No, you can’t—it’s behind you.", + "question_chinese": "我能摸到门吗？", + "answer_chinese": "不，你不能，它在你的后方。", + "evidence": [ + { + "description": "At this moment, the door and my location information appeared.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can—it’s behind you.", + "B": "No, you can’t—it’s behind you." + }, + "answer_label": "B", + "answer_text": "No, you can’t—it’s behind you.", + "q_id": "scene0737_00-0_5_T1", + "correct_answer": "No, you can’t—it’s behind you.", + "original_options": { + "A": "Yes, you can—it’s behind you.", + "B": "No, you can’t—it’s behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0737_00-0", + "video_length": "00:00:20", + "qid": "scene0737_00-0_5_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the door?", + "answer": "No, you can’t. It’s to your left.", + "question_chinese": "我能摸到门吗？", + "answer_chinese": "不，你不能。它在你的左边", + "evidence": [ + { + "description": "At this moment, information about the door and my distance appeared.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t. It’s to your left.", + "B": "Yes, you can. It’s to your left." + }, + "answer_label": "A", + "answer_text": "No, you can’t. It’s to your left.", + "q_id": "scene0737_00-0_5_T2", + "correct_answer": "No, you can’t. It’s to your left.", + "original_options": { + "A": "No, you can’t. It’s to your left.", + "B": "Yes, you can. It’s to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + } +] diff --git a/QAs/scene0738_00-0-mcq.json b/QAs/scene0738_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..a4f57dc926766f78986df3a3c609501443af4e30 --- /dev/null +++ b/QAs/scene0738_00-0-mcq.json @@ -0,0 +1,440 @@ +[ + { + "video_uid": "scene0738_00-0", + "video_length": "00:00:20", + "qid": "scene0738_00-0_3_T2", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many suitcases do I have in total?", + "answer": "You have one suitcase, in front of you.", + "question_chinese": "我一共有几个行李箱？", + "answer_chinese": "你有一个行李箱，在你的前面.", + "evidence": [ + { + "description": "You saw a suitcase.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You have one suitcase, in front of you.", + "B": "Zero suitcases, in front of you.", + "C": "Two suitcases, in front of you.", + "D": "Three suitcases in total, all positioned together in front of you so they appear grouped in one spot.", + "E": "Four suitcases in total, lined up in front of you within easy view as you look forward." + }, + "answer_label": "A", + "answer_text": "You have one suitcase, in front of you.", + "q_id": "scene0738_00-0_3_T2", + "correct_answer": "You have one suitcase, in front of you.", + "original_options": { + "A": "You have one suitcase, in front of you.", + "B": "Zero suitcases, in front of you.", + "C": "Two suitcases, in front of you.", + "D": "Three suitcases in total, all positioned together in front of you so they appear grouped in one spot.", + "E": "Four suitcases in total, lined up in front of you within easy view as you look forward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 17 + ] + }, + { + "video_uid": "scene0738_00-0", + "video_length": "00:00:20", + "qid": "scene0738_00-0_3_T3", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many suitcases do I have in total?", + "answer": "You have three suitcases: one is to the right in front of you, and two are to the left in front of you.", + "question_chinese": "我一共有几个行李箱？", + "answer_chinese": "你有三个行李箱，一个在你前面的右边，两个在你前面的左边.", + "evidence": [ + { + "description": "You saw a suitcase.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + }, + { + "description": "You saw two suitcases.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two suitcases total: one right, one left.", + "B": "You have three suitcases: one is to the right in front of you, and two are to the left in front of you.", + "C": "Four suitcases total: one right, three left.", + "D": "One suitcase total, placed to the right in front of you with none on the left.", + "E": "Five suitcases total: two to the right in front and three to the left in front." + }, + "answer_label": "B", + "answer_text": "You have three suitcases: one is to the right in front of you, and two are to the left in front of you.", + "q_id": "scene0738_00-0_3_T3", + "correct_answer": "You have three suitcases: one is to the right in front of you, and two are to the left in front of you.", + "original_options": { + "A": "Two suitcases total: one right, one left.", + "B": "You have three suitcases: one is to the right in front of you, and two are to the left in front of you.", + "C": "Four suitcases total: one right, three left.", + "D": "One suitcase total, placed to the right in front of you with none on the left.", + "E": "Five suitcases total: two to the right in front and three to the left in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 16, + 16 + ] + }, + { + "video_uid": "scene0738_00-0", + "video_length": "00:00:20", + "qid": "scene0738_00-0_3", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "clockwise rotation.", + "question_chinese": "我的运动轨迹是什么？", + "answer_chinese": "顺时针旋转", + "evidence": [ + { + "description": "The motion trajectory is a clockwise rotation.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "clockwise rotation.", + "B": "counterclockwise rotation.", + "C": "vertical oscillation.", + "D": "You rotate counterclockwise in place rather than completing a full clockwise rotation around yourself.", + "E": "Your movement traces a back-and-forth sway instead of completing a circular clockwise rotation." + }, + "answer_label": "A", + "answer_text": "clockwise rotation.", + "q_id": "scene0738_00-0_3", + "correct_answer": "clockwise rotation.", + "original_options": { + "A": "clockwise rotation.", + "B": "counterclockwise rotation.", + "C": "vertical oscillation.", + "D": "You rotate counterclockwise in place rather than completing a full clockwise rotation around yourself.", + "E": "Your movement traces a back-and-forth sway instead of completing a circular clockwise rotation." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "scene0738_00-0", + "video_length": "00:00:20", + "qid": "scene0738_00-0_4", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "Rotate clockwise and then walk forward.", + "question_chinese": "我的运动轨迹是什么？", + "answer_chinese": "顺时针旋转然后向前走", + "evidence": [ + { + "description": "Turn clockwise and then walk forward.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Rotate clockwise, then step backward.", + "B": "Rotate counterclockwise and walk forward.", + "C": "You turn to the right and then move sideways instead of rotating and proceeding straight ahead.", + "D": "After rotating, you stop and pivot again rather than walking forward in a straight line.", + "E": "Rotate clockwise and then walk forward." + }, + "answer_label": "E", + "answer_text": "Rotate clockwise and then walk forward.", + "q_id": "scene0738_00-0_4", + "correct_answer": "Rotate clockwise and then walk forward.", + "original_options": { + "A": "Rotate clockwise, then step backward.", + "B": "Rotate counterclockwise and walk forward.", + "C": "You turn to the right and then move sideways instead of rotating and proceeding straight ahead.", + "D": "After rotating, you stop and pivot again rather than walking forward in a straight line.", + "E": "Rotate clockwise and then walk forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 15 + ] + }, + { + "video_uid": "scene0738_00-0", + "video_length": "00:00:20", + "qid": "scene0738_00-0_2_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many bottles of water did I see?", + "answer": "There are 4 bottles of water in front of you. Three are on the cabinet to your left, and one is on the cabinet in front of you.", + "question_chinese": "我看到几瓶水？", + "answer_chinese": "你面前有4瓶水。三瓶在你的左边的柜子上，一瓶在你前面的柜子上。", + "evidence": [ + { + "description": "At this time, there is water, with four bottles.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three bottles total.", + "B": "Five bottles total.", + "C": "There are 3 bottles of water: two on the cabinet to your left and one on the cabinet in front.", + "D": "There are 4 bottles of water in front of you. Three are on the cabinet to your left, and one is on the cabinet in front of you.", + "E": "There are 5 bottles of water: three on the cabinet to your left and two on the cabinet in front." + }, + "answer_label": "D", + "answer_text": "There are 4 bottles of water in front of you. Three are on the cabinet to your left, and one is on the cabinet in front of you.", + "q_id": "scene0738_00-0_2_T1", + "correct_answer": "There are 4 bottles of water in front of you. Three are on the cabinet to your left, and one is on the cabinet in front of you.", + "original_options": { + "A": "Three bottles total.", + "B": "Five bottles total.", + "C": "There are 3 bottles of water: two on the cabinet to your left and one on the cabinet in front.", + "D": "There are 4 bottles of water in front of you. Three are on the cabinet to your left, and one is on the cabinet in front of you.", + "E": "There are 5 bottles of water: three on the cabinet to your left and two on the cabinet in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 20, + 20 + ] + }, + { + "video_uid": "scene0738_00-0", + "video_length": "00:00:20", + "qid": "scene0738_00-0_2_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many bottles of water did I see?", + "answer": "You see eight bottles of water: four are on the table in front of you, and the other four are to your left.", + "question_chinese": "我看到几瓶水？", + "answer_chinese": "你看到八瓶水，四瓶水在你前面的桌子上，另外四瓶在你的左边。", + "evidence": [ + { + "description": "At this moment, water appeared, and there were two bottles.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "There are four bottle of water on the table.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Seven bottles total.", + "B": "Nine bottles total.", + "C": "You see 7 bottles of water: three are on the table in front of you and four are to your left.", + "D": "You see 10 bottles of water: four are on the table in front of you and six are to your left.", + "E": "You see eight bottles of water: four are on the table in front of you, and the other four are to your left." + }, + "answer_label": "E", + "answer_text": "You see eight bottles of water: four are on the table in front of you, and the other four are to your left.", + "q_id": "scene0738_00-0_2_T2", + "correct_answer": "You see eight bottles of water: four are on the table in front of you, and the other four are to your left.", + "original_options": { + "A": "Seven bottles total.", + "B": "Nine bottles total.", + "C": "You see 7 bottles of water: three are on the table in front of you and four are to your left.", + "D": "You see 10 bottles of water: four are on the table in front of you and six are to your left.", + "E": "You see eight bottles of water: four are on the table in front of you, and the other four are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 21, + 21 + ] + }, + { + "video_uid": "scene0738_00-0", + "video_length": "00:00:20", + "qid": "scene0738_00-0_4_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the quilt relative to me?", + "answer": "The quilt is to your left.", + "question_chinese": "被子在我的哪个方位？", + "answer_chinese": "被子在你的左边", + "evidence": [ + { + "description": "At this moment, the blanket appeared, and it was messy.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The quilt is to your right.", + "B": "The quilt is to your left.", + "C": "The quilt is behind you.", + "D": "The quilt is directly in front of you, not positioned off to your left side.", + "E": "The quilt sits to your front-right, so it is not located on your left." + }, + "answer_label": "B", + "answer_text": "The quilt is to your left.", + "q_id": "scene0738_00-0_4_T2", + "correct_answer": "The quilt is to your left.", + "original_options": { + "A": "The quilt is to your right.", + "B": "The quilt is to your left.", + "C": "The quilt is behind you.", + "D": "The quilt is directly in front of you, not positioned off to your left side.", + "E": "The quilt sits to your front-right, so it is not located on your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 15, + 14 + ] + }, + { + "video_uid": "scene0738_00-0", + "video_length": "00:00:20", + "qid": "scene0738_00-0_4_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the quilt relative to me?", + "answer": "The quilt is directly to your left.", + "question_chinese": "被子在我的哪个方位？", + "answer_chinese": "被子在你的正左边。", + "evidence": [ + { + "description": "At this moment, the state of the quilt is messy.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The quilt is directly to your left.", + "B": "The quilt is directly to your right.", + "C": "The quilt is behind you.", + "D": "The quilt is in front of you rather than directly to your left side.", + "E": "The quilt is to your back-left, not aligned directly beside you on the left." + }, + "answer_label": "A", + "answer_text": "The quilt is directly to your left.", + "q_id": "scene0738_00-0_4_T1", + "correct_answer": "The quilt is directly to your left.", + "original_options": { + "A": "The quilt is directly to your left.", + "B": "The quilt is directly to your right.", + "C": "The quilt is behind you.", + "D": "The quilt is in front of you rather than directly to your left side.", + "E": "The quilt is to your back-left, not aligned directly beside you on the left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 14, + 14 + ] + } +] diff --git a/QAs/scene0739_00-0-mcq.json b/QAs/scene0739_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..937d84b8588c3ef1d76f8a2c046e7b3f3fe42014 --- /dev/null +++ b/QAs/scene0739_00-0-mcq.json @@ -0,0 +1,532 @@ +[ + { + "video_uid": "scene0739_00-0", + "video_length": "00:00:20", + "qid": "scene0739_00-0_2_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can relative to me?", + "answer": "The trash can was behind you.", + "question_chinese": "垃圾桶在相对于我的什么位置？", + "answer_chinese": "垃圾桶在你的后面", + "evidence": [ + { + "description": "The trash can is behind you", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can was behind you.", + "B": "The trash can was in front of you.", + "C": "The trash can was to your left.", + "D": "The trash can was to your right, not located behind you at that moment.", + "E": "The trash can was slightly ahead of you rather than positioned behind your back." + }, + "answer_label": "A", + "answer_text": "The trash can was behind you.", + "q_id": "scene0739_00-0_2_T1", + "correct_answer": "The trash can was behind you.", + "original_options": { + "A": "The trash can was behind you.", + "B": "The trash can was in front of you.", + "C": "The trash can was to your left.", + "D": "The trash can was to your right, not located behind you at that moment.", + "E": "The trash can was slightly ahead of you rather than positioned behind your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 14 + ] + }, + { + "video_uid": "scene0739_00-0", + "video_length": "00:00:20", + "qid": "scene0739_00-0_2_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can relative to me?", + "answer": "The trash can was to your back-left.", + "question_chinese": "垃圾桶在相对于我的什么位置？", + "answer_chinese": "垃圾桶在你的左后方", + "evidence": [ + { + "description": "The trash can is behind you to the left.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The trash can was to your back-right.", + "B": "The trash can was to your front-left.", + "C": "The trash can was to your back-left.", + "D": "The trash can was directly behind you rather than offset to the back-left side.", + "E": "The trash can was in front of you and to the left, not behind you on the back-left." + }, + "answer_label": "C", + "answer_text": "The trash can was to your back-left.", + "q_id": "scene0739_00-0_2_T2", + "correct_answer": "The trash can was to your back-left.", + "original_options": { + "A": "The trash can was to your back-right.", + "B": "The trash can was to your front-left.", + "C": "The trash can was to your back-left.", + "D": "The trash can was directly behind you rather than offset to the back-left side.", + "E": "The trash can was in front of you and to the left, not behind you on the back-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 18 + ] + }, + { + "video_uid": "scene0739_00-0", + "video_length": "00:00:20", + "qid": "scene0739_00-0_3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the trash can in the room?", + "answer": "black.", + "question_chinese": "屋子里的垃圾桶是什么颜色？", + "answer_chinese": "黑色", + "evidence": [ + { + "description": "At this moment, a trash can appeared, and it is black.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Red.", + "B": "Green.", + "C": "It’s a dark gray trash can, not black, blending into the room’s shadows.", + "D": "black.", + "E": "The trash can is metallic blue, standing out brightly against the rest of the room." + }, + "answer_label": "D", + "answer_text": "black.", + "q_id": "scene0739_00-0_3", + "correct_answer": "black.", + "original_options": { + "A": "Red.", + "B": "Green.", + "C": "It’s a dark gray trash can, not black, blending into the room’s shadows.", + "D": "black.", + "E": "The trash can is metallic blue, standing out brightly against the rest of the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "scene0739_00-0", + "video_length": "00:00:20", + "qid": "scene0739_00-0_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What color is the trash can in the room?", + "answer": "black.", + "question_chinese": "屋子里的垃圾桶是什么颜色？", + "answer_chinese": "黑色", + "evidence": [ + { + "description": "At this moment, a trash can appeared, and it is black.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Bright red.", + "B": "Metallic blue.", + "C": "It’s forest green, giving the trash can a more noticeable look in the room.", + "D": "The trash can appears vibrant orange, making it easy to spot against the background.", + "E": "black." + }, + "answer_label": "E", + "answer_text": "black.", + "q_id": "scene0739_00-0_4", + "correct_answer": "black.", + "original_options": { + "A": "Bright red.", + "B": "Metallic blue.", + "C": "It’s forest green, giving the trash can a more noticeable look in the room.", + "D": "The trash can appears vibrant orange, making it easy to spot against the background.", + "E": "black." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 14 + ] + }, + { + "video_uid": "scene0739_00-0", + "video_length": "00:00:20", + "qid": "scene0739_00-0_3_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where do I see the electric fan?", + "answer": "It was to your front-right.", + "question_chinese": "我看到电风扇在哪里？", + "answer_chinese": "在你的右前方", + "evidence": [ + { + "description": "The electric fan is in front of me on the right.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It was to your left.", + "B": "It was behind you.", + "C": "It was to your front-right.", + "D": "You saw the electric fan to your front-left when you looked ahead in the room.", + "E": "The electric fan was positioned to your back-right, slightly behind your shoulder as you turned." + }, + "answer_label": "C", + "answer_text": "It was to your front-right.", + "q_id": "scene0739_00-0_3_T1", + "correct_answer": "It was to your front-right.", + "original_options": { + "A": "It was to your left.", + "B": "It was behind you.", + "C": "It was to your front-right.", + "D": "You saw the electric fan to your front-left when you looked ahead in the room.", + "E": "The electric fan was positioned to your back-right, slightly behind your shoulder as you turned." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 15 + ] + }, + { + "video_uid": "scene0739_00-0", + "video_length": "00:00:20", + "qid": "scene0739_00-0_3_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where do I see the electric fan?", + "answer": "It was to your right.", + "question_chinese": "我看到电风扇在哪里？", + "answer_chinese": "在你的右边", + "evidence": [ + { + "description": "The electric fan is on my right.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It was in front of you.", + "B": "It was behind you.", + "C": "You saw the electric fan to your left, not on your right side while facing forward.", + "D": "It was to your right.", + "E": "The electric fan was above you, positioned higher up rather than at your side." + }, + "answer_label": "D", + "answer_text": "It was to your right.", + "q_id": "scene0739_00-0_3_T2", + "correct_answer": "It was to your right.", + "original_options": { + "A": "It was in front of you.", + "B": "It was behind you.", + "C": "You saw the electric fan to your left, not on your right side while facing forward.", + "D": "It was to your right.", + "E": "The electric fan was above you, positioned higher up rather than at your side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 16, + 14 + ] + }, + { + "video_uid": "scene0739_00-0", + "video_length": "00:00:20", + "qid": "scene0739_00-0_6", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the phone relative to me?", + "answer": "The telephone is behind you.", + "question_chinese": "电话在相对于我的什么位置？", + "answer_chinese": "电话在你后方", + "evidence": [ + { + "description": "At this time, the telephone's location status appeared.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The telephone is in front of you.", + "B": "The telephone is to your right.", + "C": "The telephone is behind you.", + "D": "The telephone is above you, positioned higher than your eye level in the room.", + "E": "The telephone is beside you, close to your body rather than positioned behind you." + }, + "answer_label": "C", + "answer_text": "The telephone is behind you.", + "q_id": "scene0739_00-0_6", + "correct_answer": "The telephone is behind you.", + "original_options": { + "A": "The telephone is in front of you.", + "B": "The telephone is to your right.", + "C": "The telephone is behind you.", + "D": "The telephone is above you, positioned higher than your eye level in the room.", + "E": "The telephone is beside you, close to your body rather than positioned behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 14 + ] + }, + { + "video_uid": "scene0739_00-0", + "video_length": "00:00:20", + "qid": "scene0739_00-0_5", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the telephone relative to me?", + "answer": "The telephone is on your left.", + "question_chinese": "电话在相对于我的什么位置？", + "answer_chinese": "电话在你左侧", + "evidence": [ + { + "description": "At this time, the phone's location information appeared.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The telephone is on your right.", + "B": "The telephone is on your left.", + "C": "The telephone is in front of you.", + "D": "The telephone is behind you on the sofa, not positioned on your left side.", + "E": "The telephone is on the shelf above you, higher up rather than to your left." + }, + "answer_label": "B", + "answer_text": "The telephone is on your left.", + "q_id": "scene0739_00-0_5", + "correct_answer": "The telephone is on your left.", + "original_options": { + "A": "The telephone is on your right.", + "B": "The telephone is on your left.", + "C": "The telephone is in front of you.", + "D": "The telephone is behind you on the sofa, not positioned on your left side.", + "E": "The telephone is on the shelf above you, higher up rather than to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 15 + ] + }, + { + "video_uid": "scene0739_00-0", + "video_length": "00:00:20", + "qid": "scene0739_00-0_8", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the status of the door?", + "answer": "The door is closed.", + "question_chinese": "门的状态是什么？", + "answer_chinese": "门的状态是关闭的", + "evidence": [ + { + "description": "At this moment, a door appeared, and it was closed.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is open.", + "B": "The door is ajar.", + "C": "The door is partially open, leaving a small gap between the door and the frame.", + "D": "The door is open wide, so you can clearly see through the doorway into the next area.", + "E": "The door is closed." + }, + "answer_label": "E", + "answer_text": "The door is closed.", + "q_id": "scene0739_00-0_8", + "correct_answer": "The door is closed.", + "original_options": { + "A": "The door is open.", + "B": "The door is ajar.", + "C": "The door is partially open, leaving a small gap between the door and the frame.", + "D": "The door is open wide, so you can clearly see through the doorway into the next area.", + "E": "The door is closed." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "scene0739_00-0", + "video_length": "00:00:20", + "qid": "scene0739_00-0_7", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the status of the door?", + "answer": "The door is closed.", + "question_chinese": "门的状态是什么？", + "answer_chinese": "门的状态是关闭的", + "evidence": [ + { + "description": "At this moment, a door appeared, and it was closed.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is closed.", + "B": "The door is open.", + "C": "The door is ajar.", + "D": "The door is left slightly open, not fully shut against the frame.", + "E": "The door appears open, with the edge angled away from the frame as if it wasn’t closed." + }, + "answer_label": "A", + "answer_text": "The door is closed.", + "q_id": "scene0739_00-0_7", + "correct_answer": "The door is closed.", + "original_options": { + "A": "The door is closed.", + "B": "The door is open.", + "C": "The door is ajar.", + "D": "The door is left slightly open, not fully shut against the frame.", + "E": "The door appears open, with the edge angled away from the frame as if it wasn’t closed." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 18 + ] + } +] diff --git a/QAs/scene0740_00-0-mcq.json b/QAs/scene0740_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..120fc25064e135c9163f511799c1d607ecc5bdf2 --- /dev/null +++ b/QAs/scene0740_00-0-mcq.json @@ -0,0 +1,849 @@ +[ + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_7", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "At the very beginning, could I simply close the door directly?", + "answer": "No，you can't.", + "question_chinese": "在最初的时刻，我可以把门直接关上吗？", + "answer_chinese": "不，你不可以", + "evidence": [ + { + "description": "At this moment, there is a relationship between the door and my distance; I can't reach it, so I can't close it directly.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No，you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No，you can't.", + "q_id": "scene0740_00-0_7", + "correct_answer": "No，you can't.", + "original_options": { + "A": "No，you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_8", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "At the very beginning, could I simply close the door directly?", + "answer": "No，you can't.", + "question_chinese": "在最初的时刻，我可以把门直接关上吗？", + "answer_chinese": "不，你不可以", + "evidence": [ + { + "description": "At this point, the distance information between me and the door appeared; it's too far away to open directly.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No，you can't." + }, + "answer_label": "B", + "answer_text": "No，you can't.", + "q_id": "scene0740_00-0_8", + "correct_answer": "No，you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No，you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_5_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the trash can relative to me?", + "answer": "The trash can is to your front-left.", + "question_chinese": "垃圾桶在相对于我的什么方向？", + "answer_chinese": "垃圾桶在相对于你的左前方", + "evidence": [ + { + "description": "The trash can is on your left front side.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "The trash can is directly to your left, slightly behind your shoulder as you face forward.", + "D": "The trash can sits in front of you but offset to the right, not in the front-left area.", + "E": "The trash can is to your front-left." + }, + "answer_label": "E", + "answer_text": "The trash can is to your front-left.", + "q_id": "scene0740_00-0_5_T1", + "correct_answer": "The trash can is to your front-left.", + "original_options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "The trash can is directly to your left, slightly behind your shoulder as you face forward.", + "D": "The trash can sits in front of you but offset to the right, not in the front-left area.", + "E": "The trash can is to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 18 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_10", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can relative to me?", + "answer": "The trash can is behind you.", + "question_chinese": "垃圾桶在相对于我的什么方向？", + "answer_chinese": "垃圾桶在相对于你的后方", + "evidence": [ + { + "description": "At this time, the location information of the trash can appeared.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The trash can is behind you.", + "B": "It is to your right.", + "C": "It is in front of you.", + "D": "The trash can is on your left side, roughly beside you rather than behind your back.", + "E": "The trash can is across the room ahead of you, not positioned behind you." + }, + "answer_label": "A", + "answer_text": "The trash can is behind you.", + "q_id": "scene0740_00-0_10", + "correct_answer": "The trash can is behind you.", + "original_options": { + "A": "The trash can is behind you.", + "B": "It is to your right.", + "C": "It is in front of you.", + "D": "The trash can is on your left side, roughly beside you rather than behind your back.", + "E": "The trash can is across the room ahead of you, not positioned behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 14 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_5_T2", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the trash can relative to me?", + "answer": "The trash can is behind you.", + "question_chinese": "垃圾桶在相对于我的什么方向？", + "answer_chinese": "垃圾桶在相对于你的后方", + "evidence": [ + { + "description": "At this time, the location information of the trash can appeared.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your right.", + "B": "The trash can is behind you.", + "C": "Directly in front of you.", + "D": "The trash can is on your left side next to you, rather than behind you where you were standing.", + "E": "The trash can is farther down the hall in front of you, not located behind your position." + }, + "answer_label": "B", + "answer_text": "The trash can is behind you.", + "q_id": "scene0740_00-0_5_T2", + "correct_answer": "The trash can is behind you.", + "original_options": { + "A": "To your right.", + "B": "The trash can is behind you.", + "C": "Directly in front of you.", + "D": "The trash can is on your left side next to you, rather than behind you where you were standing.", + "E": "The trash can is farther down the hall in front of you, not located behind your position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 5, + 19, + 17 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_3", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What items appeared on the bed?", + "answer": "Two items: eye makeup case and comb.", + "question_chinese": "床上出现了几样物品？", + "answer_chinese": "两样物品 眼睛盒和梳子", + "evidence": [ + { + "description": "Items appeared on the bed at this moment.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One item: a comb.", + "B": "Two items: eye makeup case and comb.", + "C": "Three items: comb, case, and brush.", + "D": "One item: an eye makeup case was on the bed, with no comb visible beside it.", + "E": "Three items: an eye makeup case, a comb, and a small mirror were laid out on the bed." + }, + "answer_label": "B", + "answer_text": "Two items: eye makeup case and comb.", + "q_id": "scene0740_00-0_3", + "correct_answer": "Two items: eye makeup case and comb.", + "original_options": { + "A": "One item: a comb.", + "B": "Two items: eye makeup case and comb.", + "C": "Three items: comb, case, and brush.", + "D": "One item: an eye makeup case was on the bed, with no comb visible beside it.", + "E": "Three items: an eye makeup case, a comb, and a small mirror were laid out on the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 16, + 18 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_4", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What items appeared on the bed?", + "answer": "Two items: eye makeup case and comb.", + "question_chinese": "床上出现了几样物品？", + "answer_chinese": "两样物品 眼睛盒和梳子", + "evidence": [ + { + "description": "At this moment, items appeared on the bed.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One item: an eye makeup case.", + "B": "Three items: comb, brush, and case.", + "C": "Two items: eye makeup case and comb.", + "D": "One item: only the comb appeared on the bed, without any eye makeup case present.", + "E": "Three items: an eye makeup case, a comb, and a mascara tube were all on the bed together." + }, + "answer_label": "C", + "answer_text": "Two items: eye makeup case and comb.", + "q_id": "scene0740_00-0_4", + "correct_answer": "Two items: eye makeup case and comb.", + "original_options": { + "A": "One item: an eye makeup case.", + "B": "Three items: comb, brush, and case.", + "C": "Two items: eye makeup case and comb.", + "D": "One item: only the comb appeared on the bed, without any eye makeup case present.", + "E": "Three items: an eye makeup case, a comb, and a mascara tube were all on the bed together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 18 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_4_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I close the door directly from where I am?", + "answer": "At the very beginning, you couldn’t—the door was in front of you.", + "question_chinese": "我可以把门直接关上吗？", + "answer_chinese": "在最初时刻，你不可以，门在你前边", + "evidence": [ + { + "description": "At this moment, there is a relationship between the door and my distance; I can't reach it, so I can't close it directly.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "At the very beginning, you couldn’t—the door was in front of you.", + "B": "At the very beginning, you could—the door was in front of you." + }, + "answer_label": "A", + "answer_text": "At the very beginning, you couldn’t—the door was in front of you.", + "q_id": "scene0740_00-0_4_T1", + "correct_answer": "At the very beginning, you couldn’t—the door was in front of you.", + "original_options": { + "A": "At the very beginning, you couldn’t—the door was in front of you.", + "B": "At the very beginning, you could—the door was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_4_T2", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I close the door directly from where I am?", + "answer": "Now you can—the door is to your left.", + "question_chinese": "我可以把门直接关上吗？", + "answer_chinese": "现在你可以，门在你左边", + "evidence": [ + { + "description": "Now you can, the door is on your left", + "timestamps": { + "start": "00:00:14", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Now you can’t—the door is to your left.", + "B": "Now you can—the door is to your left." + }, + "answer_label": "B", + "answer_text": "Now you can—the door is to your left.", + "q_id": "scene0740_00-0_4_T2", + "correct_answer": "Now you can—the door is to your left.", + "original_options": { + "A": "Now you can’t—the door is to your left.", + "B": "Now you can—the door is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "go straight.", + "question_chinese": "我的运动轨迹是什么？", + "answer_chinese": "直行", + "evidence": [ + { + "description": "During this time period, your trajectory is straight.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "go left.", + "B": "move backward.", + "C": "go left, then turn back the way you came, instead of continuing straight forward the whole time.", + "D": "move backward first, then continue forward after a brief pause, rather than walking straight ahead.", + "E": "go straight." + }, + "answer_label": "E", + "answer_text": "go straight.", + "q_id": "scene0740_00-0_1", + "correct_answer": "go straight.", + "original_options": { + "A": "go left.", + "B": "move backward.", + "C": "go left, then turn back the way you came, instead of continuing straight forward the whole time.", + "D": "move backward first, then continue forward after a brief pause, rather than walking straight ahead.", + "E": "go straight." + }, + "optimization_rounds": 4, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 17, + 15 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_2", + "qid_num": 11, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "Go straight, turn left and then turn right.", + "question_chinese": "我的运动轨迹是什么？", + "answer_chinese": "直行左拐然后右拐", + "evidence": [ + { + "description": "During this time period, my trajectory was to go straight, then turn left, and then turn right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go straight, then turn right.", + "B": "Go straight, then turn left.", + "C": "Go straight, turn right, then turn left, instead of turning left and then right as described.", + "D": "Go straight, turn left and then turn right.", + "E": "Go straight, turn left, then continue straight without making the final right turn." + }, + "answer_label": "D", + "answer_text": "Go straight, turn left and then turn right.", + "q_id": "scene0740_00-0_2", + "correct_answer": "Go straight, turn left and then turn right.", + "original_options": { + "A": "Go straight, then turn right.", + "B": "Go straight, then turn left.", + "C": "Go straight, turn right, then turn left, instead of turning left and then right as described.", + "D": "Go straight, turn left and then turn right.", + "E": "Go straight, turn left, then continue straight without making the final right turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 13 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_3_T1", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is my door open or closed?", + "answer": "The door is open, and it’s in front of you.", + "question_chinese": "我门是开着还是关着的？", + "answer_chinese": "你门是开着的，门在你前面", + "evidence": [ + { + "description": "At this moment, the state of the door is open.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The door is closed in front of you.", + "B": "The door is open to your right.", + "C": "The door is open, and it’s in front of you.", + "D": "The door is closed, and it’s behind you, not open in front of you as stated.", + "E": "The door is open, but it’s behind you rather than directly in front." + }, + "answer_label": "C", + "answer_text": "The door is open, and it’s in front of you.", + "q_id": "scene0740_00-0_3_T1", + "correct_answer": "The door is open, and it’s in front of you.", + "original_options": { + "A": "The door is closed in front of you.", + "B": "The door is open to your right.", + "C": "The door is open, and it’s in front of you.", + "D": "The door is closed, and it’s behind you, not open in front of you as stated.", + "E": "The door is open, but it’s behind you rather than directly in front." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 17, + 14 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_3_T2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is my door open or closed?", + "answer": "The door is open, and it’s to your left.", + "question_chinese": "我门是开着还是关着的？", + "answer_chinese": "你门是开着的，门在你左边", + "evidence": [ + { + "description": "At this moment, the door is in the open state.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The door is closed to your left.", + "B": "The door is open in front of you.", + "C": "The door is open, and it’s to your right rather than to your left.", + "D": "The door is open, and it’s to your left.", + "E": "The door is closed, and it’s in front of you instead of to your left." + }, + "answer_label": "D", + "answer_text": "The door is open, and it’s to your left.", + "q_id": "scene0740_00-0_3_T2", + "correct_answer": "The door is open, and it’s to your left.", + "original_options": { + "A": "The door is closed to your left.", + "B": "The door is open in front of you.", + "C": "The door is open, and it’s to your right rather than to your left.", + "D": "The door is open, and it’s to your left.", + "E": "The door is closed, and it’s in front of you instead of to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 15, + 16 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_2_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the air conditioner relative to me?", + "answer": "It’s in front of you.", + "question_chinese": "空调在我的什么方位？", + "answer_chinese": "在你的正前方。", + "evidence": [ + { + "description": "Directly in front of you.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s behind you.", + "B": "It’s in front of you.", + "C": "It’s to your left.", + "D": "It’s to your right, not directly in front of you where you are looking.", + "E": "It’s above you rather than positioned straight ahead in your view." + }, + "answer_label": "B", + "answer_text": "It’s in front of you.", + "q_id": "scene0740_00-0_2_T1", + "correct_answer": "It’s in front of you.", + "original_options": { + "A": "It’s behind you.", + "B": "It’s in front of you.", + "C": "It’s to your left.", + "D": "It’s to your right, not directly in front of you where you are looking.", + "E": "It’s above you rather than positioned straight ahead in your view." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 12 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_2_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the air conditioner relative to me?", + "answer": "It’s above you.", + "question_chinese": "空调在我的什么方位？", + "answer_chinese": "在你的正上方。", + "evidence": [ + { + "description": "Directly above you.", + "timestamps": { + "start": "00:00:012", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s below you.", + "B": "It’s behind you.", + "C": "It’s above you.", + "D": "It’s in front of you, not mounted above your head as it would be if it were overhead.", + "E": "It’s to your left rather than directly above you on the wall." + }, + "answer_label": "C", + "answer_text": "It’s above you.", + "q_id": "scene0740_00-0_2_T2", + "correct_answer": "It’s above you.", + "original_options": { + "A": "It’s below you.", + "B": "It’s behind you.", + "C": "It’s above you.", + "D": "It’s in front of you, not mounted above your head as it would be if it were overhead.", + "E": "It’s to your left rather than directly above you on the wall." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 13 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_5", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the door open or closed?", + "answer": "The door is open.", + "question_chinese": "门是开着还是关着的？", + "answer_chinese": "门是开着的", + "evidence": [ + { + "description": "At this moment, the state of the door is open.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The door is open.", + "B": "The door is closed." + }, + "answer_label": "A", + "answer_text": "The door is open.", + "q_id": "scene0740_00-0_5", + "correct_answer": "The door is open.", + "original_options": { + "A": "The door is open.", + "B": "The door is closed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "scene0740_00-0", + "video_length": "00:00:20", + "qid": "scene0740_00-0_6", + "qid_num": 17, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the door open or closed?", + "answer": "The door is open.", + "question_chinese": "门是开着还是关着的？", + "answer_chinese": "门是开着的", + "evidence": [ + { + "description": "At this moment, the door is in the open state.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "The door is open.", + "B": "The door is closed." + }, + "answer_label": "A", + "answer_text": "The door is open.", + "q_id": "scene0740_00-0_6", + "correct_answer": "The door is open.", + "original_options": { + "A": "The door is open.", + "B": "The door is closed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + } +] diff --git a/QAs/scene0741_00-0-mcq.json b/QAs/scene0741_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3234145b78ea0e6f48bd54f8f390c6b7966ecf71 --- /dev/null +++ b/QAs/scene0741_00-0-mcq.json @@ -0,0 +1,302 @@ +[ + { + "video_uid": "scene0741_00-0", + "video_length": "00:00:20", + "qid": "scene0741_00-0_3_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "At the very beginning, could I touch the desk lamp?", + "answer": "No, you can’t. It’s to your front-left.", + "question_chinese": "在最初的时刻，我可以触摸到台灯吗？", + "answer_chinese": "不你不能，它在你的左前方。", + "evidence": [ + { + "description": "At this moment, a desk lamp appeared on the table, far away from you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t. It’s to your front-left.", + "B": "Yes, you can. It’s within arm’s reach in front-left." + }, + "answer_label": "A", + "answer_text": "No, you can’t. It’s to your front-left.", + "q_id": "scene0741_00-0_3_T1", + "correct_answer": "No, you can’t. It’s to your front-left.", + "original_options": { + "A": "No, you can’t. It’s to your front-left.", + "B": "Yes, you can. It’s within arm’s reach in front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "scene0741_00-0", + "video_length": "00:00:20", + "qid": "scene0741_00-0_3_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "At the very beginning, could I touch the desk lamp?", + "answer": "No, you can’t. It’s directly to your left.", + "question_chinese": "在最初的时刻，我可以触摸到台灯吗？", + "answer_chinese": "不，你不能，他在你的正左方。", + "evidence": [ + { + "description": "At this moment, a desk lamp appeared on the table, far away from you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can. It’s within arm’s reach directly to your left.", + "B": "No, you can’t. It’s directly to your left." + }, + "answer_label": "B", + "answer_text": "No, you can’t. It’s directly to your left.", + "q_id": "scene0741_00-0_3_T2", + "correct_answer": "No, you can’t. It’s directly to your left.", + "original_options": { + "A": "Yes, you can. It’s within arm’s reach directly to your left.", + "B": "No, you can’t. It’s directly to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "scene0741_00-0", + "video_length": "00:00:20", + "qid": "scene0741_00-0_1_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I see appear on the floor next to the bed?", + "answer": "Four items—one suitcase, one backpack, and two bags—all on the left-hand side.", + "question_chinese": "我看到床边的地下出现了几样物品？", + "answer_chinese": "四样物品 一个行李箱 一个书包 两个袋子，都在你的左手边", + "evidence": [ + { + "description": "At this time, these items appeared.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three items total.", + "B": "Five items total.", + "C": "Four items—one suitcase, one backpack, and two bags—all on the left-hand side.", + "D": "Two items—one suitcase and one backpack—both on the left-hand side by the bed.", + "E": "Six items—one suitcase, one backpack, and four bags—clustered on the left-hand side." + }, + "answer_label": "C", + "answer_text": "Four items—one suitcase, one backpack, and two bags—all on the left-hand side.", + "q_id": "scene0741_00-0_1_T2", + "correct_answer": "Four items—one suitcase, one backpack, and two bags—all on the left-hand side.", + "original_options": { + "A": "Three items total.", + "B": "Five items total.", + "C": "Four items—one suitcase, one backpack, and two bags—all on the left-hand side.", + "D": "Two items—one suitcase and one backpack—both on the left-hand side by the bed.", + "E": "Six items—one suitcase, one backpack, and four bags—clustered on the left-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 14 + ] + }, + { + "video_uid": "scene0741_00-0", + "video_length": "00:00:20", + "qid": "scene0741_00-0_1_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many items did I see appear on the floor next to the bed?", + "answer": "Four items: one suitcase, one backpack, and two bags.", + "question_chinese": "我看到床边的地下出现了几样物品？", + "answer_chinese": "四样物品 一个行李箱 一个书包 两个袋子", + "evidence": [ + { + "description": "At this time, these items appeared.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Three items total.", + "B": "Four items: one suitcase, one backpack, and two bags.", + "C": "Five items total.", + "D": "Two items: one suitcase and one backpack, with no extra bags appearing beside the bed.", + "E": "Six items: one suitcase, one backpack, and four bags appearing together on the floor." + }, + "answer_label": "B", + "answer_text": "Four items: one suitcase, one backpack, and two bags.", + "q_id": "scene0741_00-0_1_T1", + "correct_answer": "Four items: one suitcase, one backpack, and two bags.", + "original_options": { + "A": "Three items total.", + "B": "Four items: one suitcase, one backpack, and two bags.", + "C": "Five items total.", + "D": "Two items: one suitcase and one backpack, with no extra bags appearing beside the bed.", + "E": "Six items: one suitcase, one backpack, and four bags appearing together on the floor." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 15, + 14 + ] + }, + { + "video_uid": "scene0741_00-0", + "video_length": "00:00:20", + "qid": "scene0741_00-0_4_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to now, what has my movement trajectory been?", + "answer": "You rotated about 30 degrees toward your right-hand side.", + "question_chinese": "截至到目前我的运动轨迹是什么？", + "answer_chinese": "向你的右手边旋转了大约30度", + "evidence": [ + { + "description": "At this moment, your athletic state appeared.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You rotated about 30 degrees toward your right-hand side.", + "B": "You rotated left about 30 degrees.", + "C": "You rotated right about 60 degrees.", + "D": "You rotated about 15 degrees toward your right-hand side, not a full 30-degree turn.", + "E": "You didn’t rotate at all; you stayed facing the same direction up to this point." + }, + "answer_label": "A", + "answer_text": "You rotated about 30 degrees toward your right-hand side.", + "q_id": "scene0741_00-0_4_T2", + "correct_answer": "You rotated about 30 degrees toward your right-hand side.", + "original_options": { + "A": "You rotated about 30 degrees toward your right-hand side.", + "B": "You rotated left about 30 degrees.", + "C": "You rotated right about 60 degrees.", + "D": "You rotated about 15 degrees toward your right-hand side, not a full 30-degree turn.", + "E": "You didn’t rotate at all; you stayed facing the same direction up to this point." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 16 + ] + }, + { + "video_uid": "scene0741_00-0", + "video_length": "00:00:20", + "qid": "scene0741_00-0_4_T1", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Up to now, what has my movement trajectory been?", + "answer": "You rotated about 90 degrees toward your right-hand side.", + "question_chinese": "截至到目前我的运动轨迹是什么？", + "answer_chinese": "向你的右手边旋转了大约90度", + "evidence": [ + { + "description": "At this moment, your athletic state has appeared.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You rotated left about 90 degrees.", + "B": "You rotated right about 45 degrees.", + "C": "You rotated about 120 degrees toward your right-hand side, turning past a right angle.", + "D": "You didn’t rotate; you stayed facing the same direction instead of turning right.", + "E": "You rotated about 90 degrees toward your right-hand side." + }, + "answer_label": "E", + "answer_text": "You rotated about 90 degrees toward your right-hand side.", + "q_id": "scene0741_00-0_4_T1", + "correct_answer": "You rotated about 90 degrees toward your right-hand side.", + "original_options": { + "A": "You rotated left about 90 degrees.", + "B": "You rotated right about 45 degrees.", + "C": "You rotated about 120 degrees toward your right-hand side, turning past a right angle.", + "D": "You didn’t rotate; you stayed facing the same direction instead of turning right.", + "E": "You rotated about 90 degrees toward your right-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + } +] diff --git a/QAs/scene0742_00-0-mcq.json b/QAs/scene0742_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..15c88ee7855f96dfd0131a5a3ff032912bd753a4 --- /dev/null +++ b/QAs/scene0742_00-0-mcq.json @@ -0,0 +1,690 @@ +[ + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_5", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now, I noticed what color the bath towel was on the shower door?", + "answer": "The bath towel is pink.", + "question_chinese": "刚才看到淋浴间门上的浴巾是什么颜色？", + "answer_chinese": "浴巾是粉色的", + "evidence": [ + { + "description": "A bath towel has appeared, and it is pink.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The bath towel is green.", + "B": "The bath towel is pink.", + "C": "The bath towel is white.", + "D": "The bath towel is orange, hanging over the shower door when you noticed it.", + "E": "The bath towel is purple as it drapes across the shower door in front of you." + }, + "answer_label": "B", + "answer_text": "The bath towel is pink.", + "q_id": "scene0742_00-0_5", + "correct_answer": "The bath towel is pink.", + "original_options": { + "A": "The bath towel is green.", + "B": "The bath towel is pink.", + "C": "The bath towel is white.", + "D": "The bath towel is orange, hanging over the shower door when you noticed it.", + "E": "The bath towel is purple as it drapes across the shower door in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_6", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Just now, I noticed what color the bath towel was on the shower door?", + "answer": "The bath towel is pink.", + "question_chinese": "刚才看到淋浴间门上的浴巾是什么颜色？", + "answer_chinese": "浴巾是粉色的", + "evidence": [ + { + "description": "A bath towel has appeared, and it is pink.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The bath towel is white.", + "B": "The bath towel is red.", + "C": "The bath towel hanging on the shower door is blue, not pink.", + "D": "You saw a green bath towel draped over the shower door when you looked earlier.", + "E": "The bath towel is pink." + }, + "answer_label": "E", + "answer_text": "The bath towel is pink.", + "q_id": "scene0742_00-0_6", + "correct_answer": "The bath towel is pink.", + "original_options": { + "A": "The bath towel is white.", + "B": "The bath towel is red.", + "C": "The bath towel hanging on the shower door is blue, not pink.", + "D": "You saw a green bath towel draped over the shower door when you looked earlier.", + "E": "The bath towel is pink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 15 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_1_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can relative to me?", + "answer": "In front of you.", + "question_chinese": "垃圾桶相对于我时什么方位", + "answer_chinese": "你的前方", + "evidence": [ + { + "description": "There is a trash can under your feet.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "It’s to your right side, not directly in front of you.", + "D": "The trash can is positioned behind you, so you’d need to turn around to see it.", + "E": "In front of you." + }, + "answer_label": "E", + "answer_text": "In front of you.", + "q_id": "scene0742_00-0_1_T2", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "It’s to your right side, not directly in front of you.", + "D": "The trash can is positioned behind you, so you’d need to turn around to see it.", + "E": "In front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 12, + 17 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_1_T1", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can relative to me?", + "answer": "Behind you.", + "question_chinese": "垃圾桶相对于我时什么方位", + "answer_chinese": "你的后面", + "evidence": [ + { + "description": "There is a trash can under your feet right now.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your right.", + "C": "It’s off to your left side rather than directly behind you.", + "D": "The trash can is in front of you, so you can see it without turning around.", + "E": "Behind you." + }, + "answer_label": "E", + "answer_text": "Behind you.", + "q_id": "scene0742_00-0_1_T1", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your right.", + "C": "It’s off to your left side rather than directly behind you.", + "D": "The trash can is in front of you, so you can see it without turning around.", + "E": "Behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 16 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_9", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was facing the shower head just now, could I simply turn it on directly?", + "answer": "No, you can't.", + "question_chinese": "当我刚才面对淋浴头时，我能直接打开他吗？", + "answer_chinese": "不，你不能", + "evidence": [ + { + "description": "At this moment, a showerhead appeared, separated by a door in the middle.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "answer_label": "B", + "answer_text": "No, you can't.", + "q_id": "scene0742_00-0_9", + "correct_answer": "No, you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No, you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_10", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "When I was facing the shower head just now, could I simply turn it on directly?", + "answer": "No, you can't.", + "question_chinese": "当我刚才面对淋浴头时，我能直接打开他吗？", + "answer_chinese": "不你不能", + "evidence": [ + { + "description": "At this time, a showerhead appeared, separated by a door in the middle.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No, you can't.", + "q_id": "scene0742_00-0_10", + "correct_answer": "No, you can't.", + "original_options": { + "A": "No, you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_7", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I first entered the room, in what direction was the faucet relative to me?", + "answer": "The faucet is right in front of you.", + "question_chinese": "当我刚进房间时，水龙头在相对于我的什么方向？", + "answer_chinese": "水龙头在你的正前方", + "evidence": [ + { + "description": "At this time, a faucet appeared.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Slightly to your right.", + "B": "Directly to your left.", + "C": "The faucet was a bit to your left and ahead when you first walked into the room.", + "D": "The faucet is right in front of you.", + "E": "When you entered, the faucet was behind you, so it wasn’t in your forward line of sight." + }, + "answer_label": "D", + "answer_text": "The faucet is right in front of you.", + "q_id": "scene0742_00-0_7", + "correct_answer": "The faucet is right in front of you.", + "original_options": { + "A": "Slightly to your right.", + "B": "Directly to your left.", + "C": "The faucet was a bit to your left and ahead when you first walked into the room.", + "D": "The faucet is right in front of you.", + "E": "When you entered, the faucet was behind you, so it wasn’t in your forward line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 18 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_8", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "When I first entered the room, in what direction was the faucet relative to me?", + "answer": "The faucet is right in front of you.", + "question_chinese": "当我刚进房间时，水龙头在相对于我的什么方向？", + "answer_chinese": "水龙头在你的正前方", + "evidence": [ + { + "description": "At this moment, a faucet appeared.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The faucet is right in front of you.", + "B": "Slightly to your left.", + "C": "Behind you.", + "D": "The faucet was slightly to your right on the counter when you first entered the room.", + "E": "When you walked in, the faucet was off to your left rather than directly in front of you." + }, + "answer_label": "A", + "answer_text": "The faucet is right in front of you.", + "q_id": "scene0742_00-0_8", + "correct_answer": "The faucet is right in front of you.", + "original_options": { + "A": "The faucet is right in front of you.", + "B": "Slightly to your left.", + "C": "Behind you.", + "D": "The faucet was slightly to your right on the counter when you first entered the room.", + "E": "When you walked in, the faucet was off to your left rather than directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 2, + 16, + 18 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_4_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I turn on the faucet directly?", + "answer": "Yes, you can.", + "question_chinese": "我能直接打开水龙头吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The sink is in front of you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No, you can’t." + }, + "answer_label": "A", + "answer_text": "Yes, you can.", + "q_id": "scene0742_00-0_4_T1", + "correct_answer": "Yes, you can.", + "original_options": { + "A": "Yes, you can.", + "B": "No, you can’t." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_4_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I turn on the faucet directly?", + "answer": "No.", + "question_chinese": "我能直接打开水龙头吗？", + "answer_chinese": "不", + "evidence": [ + { + "description": "The sink is in front of you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "scene0742_00-0_4_T2", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_3", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the faucet located relative to me at this moment?", + "answer": "The faucet is behind you.", + "question_chinese": "此时水龙头在我的什么方位？", + "answer_chinese": "水龙头在你后方", + "evidence": [ + { + "description": "At this time, the location information of the faucet appears.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The faucet is behind you.", + "B": "The faucet is in front of you.", + "C": "The faucet is to your left.", + "D": "The faucet is slightly to your right, near your side as you face forward.", + "E": "The faucet is above and a bit in front of you, not positioned behind you." + }, + "answer_label": "A", + "answer_text": "The faucet is behind you.", + "q_id": "scene0742_00-0_3", + "correct_answer": "The faucet is behind you.", + "original_options": { + "A": "The faucet is behind you.", + "B": "The faucet is in front of you.", + "C": "The faucet is to your left.", + "D": "The faucet is slightly to your right, near your side as you face forward.", + "E": "The faucet is above and a bit in front of you, not positioned behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 15 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_4", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the faucet located relative to me at this moment?", + "answer": "The faucet is to your right rear.", + "question_chinese": "此时水龙头在我的什么方位？", + "answer_chinese": "水龙头在你右后方", + "evidence": [ + { + "description": "At this time, the location information of the faucet appeared.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The faucet is to your left rear.", + "B": "The faucet is to your right front.", + "C": "The faucet is behind you on the left side, not on your right rear.", + "D": "The faucet is off to your front-right, closer to your side than to your back.", + "E": "The faucet is to your right rear." + }, + "answer_label": "E", + "answer_text": "The faucet is to your right rear.", + "q_id": "scene0742_00-0_4", + "correct_answer": "The faucet is to your right rear.", + "original_options": { + "A": "The faucet is to your left rear.", + "B": "The faucet is to your right front.", + "C": "The faucet is behind you on the left side, not on your right rear.", + "D": "The faucet is off to your front-right, closer to your side than to your back.", + "E": "The faucet is to your right rear." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 14, + 15 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_1", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Was there a trash can beneath my feet when I entered the door?", + "answer": "Yes.", + "question_chinese": "进门的时候我的脚下有一个垃圾桶吗？", + "answer_chinese": "是的", + "evidence": [ + { + "description": "There is a trash can under your feet right now.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "scene0742_00-0_1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0742_00-0", + "video_length": "00:00:20", + "qid": "scene0742_00-0_2", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Was there a trash can beneath my feet when I entered the door?", + "answer": "Yes.", + "question_chinese": "进门的时候我的脚下有一个垃圾桶吗？", + "answer_chinese": "是的", + "evidence": [ + { + "description": "There is a trash can under your feet.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "scene0742_00-0_2", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/scene0743_00-0-mcq.json b/QAs/scene0743_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..dc6e5fcd1b6116e7fe3c468d83619c9fca79420e --- /dev/null +++ b/QAs/scene0743_00-0-mcq.json @@ -0,0 +1,214 @@ +[ + { + "video_uid": "scene0743_00-0", + "video_length": "00:00:20", + "qid": "scene0743_00-0_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How should I go to the door now?", + "answer": "go straight.", + "question_chinese": "我现在去门应该怎么走？", + "answer_chinese": "直走", + "evidence": [ + { + "description": "The door is directly in front.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "turn left.", + "B": "turn right.", + "C": "go straight.", + "D": "walk backward a few steps, then turn right and head toward the door.", + "E": "sidestep toward the wall first, then angle left to reach the door." + }, + "answer_label": "C", + "answer_text": "go straight.", + "q_id": "scene0743_00-0_5", + "correct_answer": "go straight.", + "original_options": { + "A": "turn left.", + "B": "turn right.", + "C": "go straight.", + "D": "walk backward a few steps, then turn right and head toward the door.", + "E": "sidestep toward the wall first, then angle left to reach the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "scene0743_00-0", + "video_length": "00:00:20", + "qid": "scene0743_00-0_6", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How should I go to the door now?", + "answer": "Turn left and then go straight.", + "question_chinese": "我现在去门应该怎么走？", + "answer_chinese": "左走然后直走", + "evidence": [ + { + "description": "The door is directly on the left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Turn right, then go straight.", + "B": "Turn left and then go straight.", + "C": "Go straight, then turn left.", + "D": "Turn left, then take a right at the corner before continuing straight to the door.", + "E": "Go straight first, then turn left near the end of the hallway to reach the door." + }, + "answer_label": "B", + "answer_text": "Turn left and then go straight.", + "q_id": "scene0743_00-0_6", + "correct_answer": "Turn left and then go straight.", + "original_options": { + "A": "Turn right, then go straight.", + "B": "Turn left and then go straight.", + "C": "Go straight, then turn left.", + "D": "Turn left, then take a right at the corner before continuing straight to the door.", + "E": "Go straight first, then turn left near the end of the hallway to reach the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "scene0743_00-0", + "video_length": "00:00:20", + "qid": "scene0743_00-0_9", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many of the shelves in the room have no doors?", + "answer": "three.", + "question_chinese": "房间里有几格柜子是没有门的？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There are three cubbies in the room without doors.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "three.", + "B": "one.", + "C": "two.", + "D": "Four shelves in the room have no doors at all, unlike the others with doors.", + "E": "Five shelves in the room have no doors at all, leaving the remaining shelves enclosed." + }, + "answer_label": "A", + "answer_text": "three.", + "q_id": "scene0743_00-0_9", + "correct_answer": "three.", + "original_options": { + "A": "three.", + "B": "one.", + "C": "two.", + "D": "Four shelves in the room have no doors at all, unlike the others with doors.", + "E": "Five shelves in the room have no doors at all, leaving the remaining shelves enclosed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "scene0743_00-0", + "video_length": "00:00:20", + "qid": "scene0743_00-0_10", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many of the shelves in the room have no doors?", + "answer": "three.", + "question_chinese": "房间里有几格柜子是没有门的？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "There are three cubbies in the room without doors.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "four.", + "B": "five.", + "C": "Six shelves in the room have no doors at all, while the remaining shelves are enclosed.", + "D": "three.", + "E": "Seven shelves in the room have no doors at all, with only a few shelves having doors." + }, + "answer_label": "D", + "answer_text": "three.", + "q_id": "scene0743_00-0_10", + "correct_answer": "three.", + "original_options": { + "A": "four.", + "B": "five.", + "C": "Six shelves in the room have no doors at all, while the remaining shelves are enclosed.", + "D": "three.", + "E": "Seven shelves in the room have no doors at all, with only a few shelves having doors." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 17 + ] + } +] diff --git a/QAs/scene0744_00-0-mcq.json b/QAs/scene0744_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..f37de90b2ae8e9ccb3852943bca4440ecf892c3c --- /dev/null +++ b/QAs/scene0744_00-0-mcq.json @@ -0,0 +1,408 @@ +[ + { + "video_uid": "scene0744_00-0", + "video_length": "00:00:20", + "qid": "scene0744_00-0_5", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the remote control that was on the table just now?", + "answer": "Yes.", + "question_chinese": "我可以拿到刚才桌上的遥控器吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The table where the remote control is located is very close.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "scene0744_00-0_5", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0744_00-0", + "video_length": "00:00:20", + "qid": "scene0744_00-0_6", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I get the remote control that was on the table just now?", + "answer": "No.", + "question_chinese": "我可以拿到刚才桌上的遥控器吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "The table where the remote control is located is quite far away.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:00" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "scene0744_00-0_6", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0744_00-0", + "video_length": "00:00:20", + "qid": "scene0744_00-0_1_T1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles did I see in the room?", + "answer": "One, to your right.", + "question_chinese": "我看到房间内有几个瓶子？", + "answer_chinese": "一个，在你的右方", + "evidence": [ + { + "description": "saw a bottle", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero, to your right.", + "B": "Two, to your right.", + "C": "Three bottles were to your right in the room, lined up along the wall near the table.", + "D": "Four bottles were to your right side, grouped together on the floor near the doorway.", + "E": "One, to your right." + }, + "answer_label": "E", + "answer_text": "One, to your right.", + "q_id": "scene0744_00-0_1_T1", + "correct_answer": "One, to your right.", + "original_options": { + "A": "Zero, to your right.", + "B": "Two, to your right.", + "C": "Three bottles were to your right in the room, lined up along the wall near the table.", + "D": "Four bottles were to your right side, grouped together on the floor near the doorway.", + "E": "One, to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 15 + ] + }, + { + "video_uid": "scene0744_00-0", + "video_length": "00:00:20", + "qid": "scene0744_00-0_1_T2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles did I see in the room?", + "answer": "One, behind you to the right.", + "question_chinese": "我看到房间内有几个瓶子？", + "answer_chinese": "一个，在你的右后边", + "evidence": [ + { + "description": "saw a bottle", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, behind you to the right.", + "B": "Zero, behind you to the right.", + "C": "Two, behind you to the right.", + "D": "Three bottles were behind you to the right, clustered together near the corner of the room.", + "E": "Four bottles were behind you to the right in the room, spread out near the back wall." + }, + "answer_label": "A", + "answer_text": "One, behind you to the right.", + "q_id": "scene0744_00-0_1_T2", + "correct_answer": "One, behind you to the right.", + "original_options": { + "A": "One, behind you to the right.", + "B": "Zero, behind you to the right.", + "C": "Two, behind you to the right.", + "D": "Three bottles were behind you to the right, clustered together near the corner of the room.", + "E": "Four bottles were behind you to the right in the room, spread out near the back wall." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "scene0744_00-0", + "video_length": "00:00:20", + "qid": "scene0744_00-0_5_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many clothes hangers did I see on the chair?", + "answer": "Three, to your back-right.", + "question_chinese": "我看到椅子上有几个衣架？", + "answer_chinese": "三个，在你的右后方。", + "evidence": [ + { + "description": "There are three hangers on the chair.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One, to your back-right.", + "B": "Two, to your back-right.", + "C": "Four clothes hangers were on the chair to your back-right, stacked together in a small pile.", + "D": "Three, to your back-right.", + "E": "Five clothes hangers were on the chair to your back-right, with the hooks facing upward." + }, + "answer_label": "D", + "answer_text": "Three, to your back-right.", + "q_id": "scene0744_00-0_5_T1", + "correct_answer": "Three, to your back-right.", + "original_options": { + "A": "One, to your back-right.", + "B": "Two, to your back-right.", + "C": "Four clothes hangers were on the chair to your back-right, stacked together in a small pile.", + "D": "Three, to your back-right.", + "E": "Five clothes hangers were on the chair to your back-right, with the hooks facing upward." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 15 + ] + }, + { + "video_uid": "scene0744_00-0", + "video_length": "00:00:20", + "qid": "scene0744_00-0_5_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many clothes hangers did I see on the chair?", + "answer": "Three, behind you to the right.", + "question_chinese": "我看到椅子上有几个衣架？", + "answer_chinese": "三个,在你的右后边。", + "evidence": [ + { + "description": "There are three hangers on the chair.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Two, behind you to the right.", + "B": "Four, behind you to the right.", + "C": "One, behind you to the right, hanging off the chair back and seat edge.", + "D": "Five, behind you to the right, spread across the chair and its arm.", + "E": "Three, behind you to the right." + }, + "answer_label": "E", + "answer_text": "Three, behind you to the right.", + "q_id": "scene0744_00-0_5_T2", + "correct_answer": "Three, behind you to the right.", + "original_options": { + "A": "Two, behind you to the right.", + "B": "Four, behind you to the right.", + "C": "One, behind you to the right, hanging off the chair back and seat edge.", + "D": "Five, behind you to the right, spread across the chair and its arm.", + "E": "Three, behind you to the right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 13 + ] + }, + { + "video_uid": "scene0744_00-0", + "video_length": "00:00:20", + "qid": "scene0744_00-0_4_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the tissue box on the table relative to me right now?", + "answer": "Directly to your left.", + "question_chinese": "桌子上的抽纸现在在我的什么方位？", + "answer_chinese": "在你的正左方", + "evidence": [ + { + "description": "After the tissue appeared on the left, I took a few steps forward.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly to your right.", + "B": "Behind you on the left.", + "C": "Slightly in front of you on the left side of the table.", + "D": "Directly to your left.", + "E": "Behind you and a bit to the right edge of the table." + }, + "answer_label": "D", + "answer_text": "Directly to your left.", + "q_id": "scene0744_00-0_4_T2", + "correct_answer": "Directly to your left.", + "original_options": { + "A": "Directly to your right.", + "B": "Behind you on the left.", + "C": "Slightly in front of you on the left side of the table.", + "D": "Directly to your left.", + "E": "Behind you and a bit to the right edge of the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 12, + 12 + ] + }, + { + "video_uid": "scene0744_00-0", + "video_length": "00:00:20", + "qid": "scene0744_00-0_4_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the tissue box on the table relative to me right now?", + "answer": "To your back-left.", + "question_chinese": "桌子上的抽纸现在在我的什么方位？", + "answer_chinese": "左后方", + "evidence": [ + { + "description": "After the tissue appeared on the left, I took a few steps forward.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-right.", + "B": "Directly to your left.", + "C": "To your back-left.", + "D": "To your front-left, closer to the near edge of the table surface.", + "E": "Directly to your right, positioned along the far side of the table." + }, + "answer_label": "C", + "answer_text": "To your back-left.", + "q_id": "scene0744_00-0_4_T1", + "correct_answer": "To your back-left.", + "original_options": { + "A": "To your back-right.", + "B": "Directly to your left.", + "C": "To your back-left.", + "D": "To your front-left, closer to the near edge of the table surface.", + "E": "Directly to your right, positioned along the far side of the table." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 12, + 12 + ] + } +] diff --git a/QAs/scene0745_00-0-mcq.json b/QAs/scene0745_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..3176e66551e5b55b66f89f5db403f5579adcb33c --- /dev/null +++ b/QAs/scene0745_00-0-mcq.json @@ -0,0 +1,726 @@ +[ + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_4_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the coffee machine relative to me?", + "answer": "It’s directly in front of you.", + "question_chinese": "咖啡机在我的什么方位？", + "answer_chinese": "在你的正前方。", + "evidence": [ + { + "description": "The coffee machine is in the cupboard.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s directly in front of you.", + "B": "It’s directly behind you.", + "C": "It’s directly to your left.", + "D": "It’s off to your right side, not centered in your view.", + "E": "It’s behind you, slightly to the left rather than straight ahead." + }, + "answer_label": "A", + "answer_text": "It’s directly in front of you.", + "q_id": "scene0745_00-0_4_T1", + "correct_answer": "It’s directly in front of you.", + "original_options": { + "A": "It’s directly in front of you.", + "B": "It’s directly behind you.", + "C": "It’s directly to your left.", + "D": "It’s off to your right side, not centered in your view.", + "E": "It’s behind you, slightly to the left rather than straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 12, + 12 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_4_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the coffee machine relative to me?", + "answer": "It’s on the left side.", + "question_chinese": "咖啡机在我的什么方位？", + "answer_chinese": "在你的左侧。", + "evidence": [ + { + "description": "The coffee machine is in the cupboard.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s on the right side.", + "B": "It’s directly in front.", + "C": "It’s behind you, not on either side of your view at the moment.", + "D": "It’s on the left side.", + "E": "It’s in front of you, centered rather than positioned to the side." + }, + "answer_label": "D", + "answer_text": "It’s on the left side.", + "q_id": "scene0745_00-0_4_T2", + "correct_answer": "It’s on the left side.", + "original_options": { + "A": "It’s on the right side.", + "B": "It’s directly in front.", + "C": "It’s behind you, not on either side of your view at the moment.", + "D": "It’s on the left side.", + "E": "It’s in front of you, centered rather than positioned to the side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 13 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_9", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the coffee machine relative to the green doll?", + "answer": "Directly above.", + "question_chinese": "咖啡机在绿色玩偶的什么方位？", + "answer_chinese": "正上方", + "evidence": [ + { + "description": "The coffee machine is directly above the green toy.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To the right.", + "B": "Directly above.", + "C": "Directly below.", + "D": "In front of the green doll on the counter, not positioned over it.", + "E": "Behind the green doll, offset to one side rather than vertically aligned." + }, + "answer_label": "B", + "answer_text": "Directly above.", + "q_id": "scene0745_00-0_9", + "correct_answer": "Directly above.", + "original_options": { + "A": "To the right.", + "B": "Directly above.", + "C": "Directly below.", + "D": "In front of the green doll on the counter, not positioned over it.", + "E": "Behind the green doll, offset to one side rather than vertically aligned." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 13, + 12 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_10", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "What is the position of the coffee machine relative to the green doll?", + "answer": "directly above.", + "question_chinese": "咖啡机在绿色玩偶的什么方位？", + "answer_chinese": "正上方", + "evidence": [ + { + "description": "The coffee machine is directly above the green toy.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "To the left.", + "B": "Directly below.", + "C": "Across from the green doll, separated rather than stacked vertically at all.", + "D": "Behind the green doll and slightly to the right instead of directly overhead.", + "E": "directly above." + }, + "answer_label": "E", + "answer_text": "directly above.", + "q_id": "scene0745_00-0_10", + "correct_answer": "directly above.", + "original_options": { + "A": "To the left.", + "B": "Directly below.", + "C": "Across from the green doll, separated rather than stacked vertically at all.", + "D": "Behind the green doll and slightly to the right instead of directly overhead.", + "E": "directly above." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 12, + 13 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_3_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothing did I see on the chair?", + "answer": "You saw two pieces of clothing.", + "question_chinese": "我看见椅子上有几件衣服？", + "answer_chinese": "你看见有两件衣服", + "evidence": [ + { + "description": "There are two pieces of clothing on the chair.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one piece of clothing.", + "B": "You saw three pieces of clothing.", + "C": "You saw four pieces of clothing laid over the chair back and seat.", + "D": "You saw two pieces of clothing.", + "E": "You saw zero pieces of clothing; the chair didn’t have any clothes on it." + }, + "answer_label": "D", + "answer_text": "You saw two pieces of clothing.", + "q_id": "scene0745_00-0_3_T2", + "correct_answer": "You saw two pieces of clothing.", + "original_options": { + "A": "You saw one piece of clothing.", + "B": "You saw three pieces of clothing.", + "C": "You saw four pieces of clothing laid over the chair back and seat.", + "D": "You saw two pieces of clothing.", + "E": "You saw zero pieces of clothing; the chair didn’t have any clothes on it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 15 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_3_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothing did I see on the chair?", + "answer": "You saw three pieces of clothing.", + "question_chinese": "我看见椅子上有几件衣服？", + "answer_chinese": "你看见有三件衣服。", + "evidence": [ + { + "description": "There are three pieces of clothing on the chair.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two pieces of clothing.", + "B": "You saw four pieces of clothing.", + "C": "You saw three pieces of clothing.", + "D": "You saw one piece of clothing on the chair, with no other garments visible nearby.", + "E": "You saw five pieces of clothing on the chair, piled together rather than spread out." + }, + "answer_label": "C", + "answer_text": "You saw three pieces of clothing.", + "q_id": "scene0745_00-0_3_T1", + "correct_answer": "You saw three pieces of clothing.", + "original_options": { + "A": "You saw two pieces of clothing.", + "B": "You saw four pieces of clothing.", + "C": "You saw three pieces of clothing.", + "D": "You saw one piece of clothing on the chair, with no other garments visible nearby.", + "E": "You saw five pieces of clothing on the chair, piled together rather than spread out." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 15 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_7", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothing are on the chair?", + "answer": "two.", + "question_chinese": "椅子上有几件衣服？", + "answer_chinese": "两件", + "evidence": [ + { + "description": "There are two pieces of clothing on the chair.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four pieces of clothing were on the chair, stacked together rather than separated.", + "D": "two.", + "E": "Five pieces of clothing were on the chair, making the seat look mostly covered." + }, + "answer_label": "D", + "answer_text": "two.", + "q_id": "scene0745_00-0_7", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four pieces of clothing were on the chair, stacked together rather than separated.", + "D": "two.", + "E": "Five pieces of clothing were on the chair, making the seat look mostly covered." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_8", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many pieces of clothing are on the chair?", + "answer": "two.", + "question_chinese": "椅子上有几件衣服？", + "answer_chinese": "两件", + "evidence": [ + { + "description": "There are two pieces of clothing on the chair.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four pieces of clothing were on the chair, leaving little of the seat visible.", + "D": "Five pieces of clothing were on the chair, with multiple items layered on top.", + "E": "two." + }, + "answer_label": "E", + "answer_text": "two.", + "q_id": "scene0745_00-0_8", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four pieces of clothing were on the chair, leaving little of the seat visible.", + "D": "Five pieces of clothing were on the chair, with multiple items layered on top.", + "E": "two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_2_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the blue trash can relative to me?", + "answer": "It was on the right side of you.", + "question_chinese": "蓝色的垃圾桶在我的的哪里？", + "answer_chinese": "在你的右侧。", + "evidence": [ + { + "description": "On your right.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was on the right side of you.", + "B": "It was on your left.", + "C": "It was behind you.", + "D": "It was directly in front of you, centered rather than off to either side.", + "E": "It was slightly behind and to your left, not positioned on your right side." + }, + "answer_label": "A", + "answer_text": "It was on the right side of you.", + "q_id": "scene0745_00-0_2_T2", + "correct_answer": "It was on the right side of you.", + "original_options": { + "A": "It was on the right side of you.", + "B": "It was on your left.", + "C": "It was behind you.", + "D": "It was directly in front of you, centered rather than off to either side.", + "E": "It was slightly behind and to your left, not positioned on your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 14, + 14 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_2_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the blue trash can relative to me?", + "answer": "It’s directly in front of you.", + "question_chinese": "蓝色的垃圾桶在我的的哪里？", + "answer_chinese": "在你的正前方。", + "evidence": [ + { + "description": "The blue trash can is to the left of the gray trash can.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s directly behind you.", + "B": "It’s directly to your left.", + "C": "It’s directly to your right, not positioned in front of you at all.", + "D": "It’s directly in front of you.", + "E": "It’s off to the side of you rather than directly in front, so you’d need to turn." + }, + "answer_label": "D", + "answer_text": "It’s directly in front of you.", + "q_id": "scene0745_00-0_2_T1", + "correct_answer": "It’s directly in front of you.", + "original_options": { + "A": "It’s directly behind you.", + "B": "It’s directly to your left.", + "C": "It’s directly to your right, not positioned in front of you at all.", + "D": "It’s directly in front of you.", + "E": "It’s off to the side of you rather than directly in front, so you’d need to turn." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 14, + 19 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_5", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "On which side of the gray trash can is the blue trash can?", + "answer": "left.", + "question_chinese": "蓝色的垃圾桶在灰色垃圾桶的哪一侧？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The blue trash can is to the left of the gray trash can.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right.", + "B": "In front.", + "C": "left.", + "D": "It’s on the right side of the gray trash can, not on its left.", + "E": "It’s positioned behind the gray trash can instead of being on the left side." + }, + "answer_label": "C", + "answer_text": "left.", + "q_id": "scene0745_00-0_5", + "correct_answer": "left.", + "original_options": { + "A": "Right.", + "B": "In front.", + "C": "left.", + "D": "It’s on the right side of the gray trash can, not on its left.", + "E": "It’s positioned behind the gray trash can instead of being on the left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 15, + 15 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_6", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "On which side of the gray trash can is the blue trash can?", + "answer": "left.", + "question_chinese": "蓝色的垃圾桶在灰色垃圾桶的哪一侧？", + "answer_chinese": "左侧", + "evidence": [ + { + "description": "The blue trash can is to the left of the gray trash can.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "left.", + "B": "Right.", + "C": "Behind it.", + "D": "It’s on the right side of the gray trash can rather than on its left side.", + "E": "It’s in front of the gray trash can, not to the left of it." + }, + "answer_label": "A", + "answer_text": "left.", + "q_id": "scene0745_00-0_6", + "correct_answer": "left.", + "original_options": { + "A": "left.", + "B": "Right.", + "C": "Behind it.", + "D": "It’s on the right side of the gray trash can rather than on its left side.", + "E": "It’s in front of the gray trash can, not to the left of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 17, + 15 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_3", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the door or the chair?", + "answer": "Door.", + "question_chinese": "门和椅子哪一个离我更近？", + "answer_chinese": "门", + "evidence": [ + { + "description": "The door is on the left at a close distance, and the chair is on the right at a far distance.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Door.", + "B": "Chair." + }, + "answer_label": "A", + "answer_text": "Door.", + "q_id": "scene0745_00-0_3", + "correct_answer": "Door.", + "original_options": { + "A": "Door.", + "B": "Chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0745_00-0", + "video_length": "00:00:20", + "qid": "scene0745_00-0_4", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the door or the chair?", + "answer": "Chair.", + "question_chinese": "门和椅子哪一个离我更近？", + "answer_chinese": "椅子", + "evidence": [ + { + "description": "The chair is on the left, and further to the left is the door.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Door.", + "B": "Chair." + }, + "answer_label": "B", + "answer_text": "Chair.", + "q_id": "scene0745_00-0_4", + "correct_answer": "Chair.", + "original_options": { + "A": "Door.", + "B": "Chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/scene0746_00-0-mcq.json b/QAs/scene0746_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..d144e77fcfcdfda30b4517c96c06cc7f078978b9 --- /dev/null +++ b/QAs/scene0746_00-0-mcq.json @@ -0,0 +1,503 @@ +[ + { + "video_uid": "scene0746_00-0", + "video_length": "00:00:20", + "qid": "scene0746_00-0_10", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator?", + "answer": "No.", + "question_chinese": "我能够到冰箱吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The refrigerator is very far away.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "scene0746_00-0_10", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0746_00-0", + "video_length": "00:00:20", + "qid": "scene0746_00-0_9", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the refrigerator?", + "answer": "No.", + "question_chinese": "我能够到冰箱吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The refrigerator is very far away.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "scene0746_00-0_9", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0746_00-0", + "video_length": "00:00:20", + "qid": "scene0746_00-0_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles are in the room?", + "answer": "two.", + "question_chinese": "房间里有多少瓶子？", + "answer_chinese": "2个", + "evidence": [ + { + "description": "saw two bottles", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:9", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "two.", + "B": "One bottle.", + "C": "Three bottles.", + "D": "There are four bottles in the room in total, not counting any outside view.", + "E": "You can see five bottles placed around the room, including those near the table." + }, + "answer_label": "A", + "answer_text": "two.", + "q_id": "scene0746_00-0_1", + "correct_answer": "two.", + "original_options": { + "A": "two.", + "B": "One bottle.", + "C": "Three bottles.", + "D": "There are four bottles in the room in total, not counting any outside view.", + "E": "You can see five bottles placed around the room, including those near the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 14, + 14 + ] + }, + { + "video_uid": "scene0746_00-0", + "video_length": "00:00:20", + "qid": "scene0746_00-0_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles are in the room?", + "answer": "four.", + "question_chinese": "房间里有多少瓶子？", + "answer_chinese": "四个", + "evidence": [ + { + "description": "see two bottles", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + }, + { + "description": "see two bottles", + "timestamps": { + "start": "00:00:10", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Three bottles.", + "B": "four.", + "C": "Five bottles.", + "D": "There are two bottles in the room altogether, with none hidden from view.", + "E": "You have six bottles visible in the room, spread across different surfaces." + }, + "answer_label": "B", + "answer_text": "four.", + "q_id": "scene0746_00-0_2", + "correct_answer": "four.", + "original_options": { + "A": "Three bottles.", + "B": "four.", + "C": "Five bottles.", + "D": "There are two bottles in the room altogether, with none hidden from view.", + "E": "You have six bottles visible in the room, spread across different surfaces." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 12 + ] + }, + { + "video_uid": "scene0746_00-0", + "video_length": "00:00:20", + "qid": "scene0746_00-0_3", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the dishwasher from me?", + "answer": "rear left.", + "question_chinese": "洗碗机在我的什么方向？", + "answer_chinese": "左后方", + "evidence": [ + { + "description": "The dishwasher is at the back left.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Front right.", + "B": "Back right.", + "C": "It is directly behind you, near the center of the space, not to either side.", + "D": "It sits to your left side, slightly in front of you, rather than behind you.", + "E": "rear left." + }, + "answer_label": "E", + "answer_text": "rear left.", + "q_id": "scene0746_00-0_3", + "correct_answer": "rear left.", + "original_options": { + "A": "Front right.", + "B": "Back right.", + "C": "It is directly behind you, near the center of the space, not to either side.", + "D": "It sits to your left side, slightly in front of you, rather than behind you.", + "E": "rear left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 15, + 15 + ] + }, + { + "video_uid": "scene0746_00-0", + "video_length": "00:00:20", + "qid": "scene0746_00-0_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the dishwasher from me?", + "answer": "back.", + "question_chinese": "洗碗机在我的什么方向？", + "answer_chinese": "正后方", + "evidence": [ + { + "description": "The dishwasher is directly at the back.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left.", + "B": "Front.", + "C": "back.", + "D": "It is behind you and slightly to the right, not directly centered behind you.", + "E": "It is in front of you, a bit off to your left, rather than behind you." + }, + "answer_label": "C", + "answer_text": "back.", + "q_id": "scene0746_00-0_4", + "correct_answer": "back.", + "original_options": { + "A": "Left.", + "B": "Front.", + "C": "back.", + "D": "It is behind you and slightly to the right, not directly centered behind you.", + "E": "It is in front of you, a bit off to your left, rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 16 + ] + }, + { + "video_uid": "scene0746_00-0", + "video_length": "00:00:20", + "qid": "scene0746_00-0_5", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is farther from me, the gas stove or the dishwasher?", + "answer": "Dishwasher.", + "question_chinese": "燃气灶和洗碗机哪个离我更远？", + "answer_chinese": "洗碗机", + "evidence": [ + { + "description": "The gas stove is nearby, and the dishwasher is farther away.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Dishwasher.", + "B": "Gas stove." + }, + "answer_label": "A", + "answer_text": "Dishwasher.", + "q_id": "scene0746_00-0_5", + "correct_answer": "Dishwasher.", + "original_options": { + "A": "Dishwasher.", + "B": "Gas stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "scene0746_00-0", + "video_length": "00:00:20", + "qid": "scene0746_00-0_6", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is farther from me, the gas stove or the dishwasher?", + "answer": "Dishwasher.", + "question_chinese": "燃气灶和洗碗机哪个离我更远？", + "answer_chinese": "洗碗机", + "evidence": [ + { + "description": "The gas stove is nearby, and the dishwasher is farther away.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:00" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Dishwasher.", + "B": "Gas stove." + }, + "answer_label": "A", + "answer_text": "Dishwasher.", + "q_id": "scene0746_00-0_6", + "correct_answer": "Dishwasher.", + "original_options": { + "A": "Dishwasher.", + "B": "Gas stove." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "scene0746_00-0", + "video_length": "00:00:20", + "qid": "scene0746_00-0_7", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How do I get to the refrigerator?", + "answer": "Move backward, then turn right.", + "question_chinese": "要如何走到冰箱？", + "answer_chinese": "向后走后右转", + "evidence": [ + { + "description": "The refrigerator is at the back left.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Move forward, then turn right.", + "B": "Move backward, then turn left.", + "C": "Step forward a bit, then turn left to face the refrigerator directly.", + "D": "Move backward, then turn right.", + "E": "Back up slightly and turn left, then continue until you reach the refrigerator." + }, + "answer_label": "D", + "answer_text": "Move backward, then turn right.", + "q_id": "scene0746_00-0_7", + "correct_answer": "Move backward, then turn right.", + "original_options": { + "A": "Move forward, then turn right.", + "B": "Move backward, then turn left.", + "C": "Step forward a bit, then turn left to face the refrigerator directly.", + "D": "Move backward, then turn right.", + "E": "Back up slightly and turn left, then continue until you reach the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "scene0746_00-0", + "video_length": "00:00:20", + "qid": "scene0746_00-0_8", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "How do I get to the refrigerator?", + "answer": "Go right, then turn right.", + "question_chinese": "要如何走到冰箱？", + "answer_chinese": "向右走后右转", + "evidence": [ + { + "description": "The refrigerator is at the back right.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go left, then turn right.", + "B": "Go right, then turn right.", + "C": "Go right, then turn left.", + "D": "Walk straight ahead, then turn right once you reach the end of the counter.", + "E": "Step to the left first, then turn right to line up with the refrigerator." + }, + "answer_label": "B", + "answer_text": "Go right, then turn right.", + "q_id": "scene0746_00-0_8", + "correct_answer": "Go right, then turn right.", + "original_options": { + "A": "Go left, then turn right.", + "B": "Go right, then turn right.", + "C": "Go right, then turn left.", + "D": "Walk straight ahead, then turn right once you reach the end of the counter.", + "E": "Step to the left first, then turn right to line up with the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 14 + ] + } +] diff --git a/QAs/scene0748_00-0-mcq.json b/QAs/scene0748_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8e2b459f72ff3c0fca0d86e0a144c76fed2b79a2 --- /dev/null +++ b/QAs/scene0748_00-0-mcq.json @@ -0,0 +1,390 @@ +[ + { + "video_uid": "scene0748_00-0", + "video_length": "00:00:20", + "qid": "scene0748_00-0_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In what direction is the toilet paper relative to the speaker?", + "answer": "right.", + "question_chinese": "卷纸在音响的什么方向？", + "answer_chinese": "右边", + "evidence": [ + { + "description": "The toilet paper is to the right of the speaker.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "left.", + "B": "behind.", + "C": "It’s in front of the speaker, not on the right side.", + "D": "The toilet paper is behind the speaker rather than positioned to the right.", + "E": "right." + }, + "answer_label": "E", + "answer_text": "right.", + "q_id": "scene0748_00-0_7", + "correct_answer": "right.", + "original_options": { + "A": "left.", + "B": "behind.", + "C": "It’s in front of the speaker, not on the right side.", + "D": "The toilet paper is behind the speaker rather than positioned to the right.", + "E": "right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "scene0748_00-0", + "video_length": "00:00:20", + "qid": "scene0748_00-0_8", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In what direction is the toilet paper relative to the speaker?", + "answer": "right.", + "question_chinese": "卷纸在音响的什么方向？", + "answer_chinese": "右边", + "evidence": [ + { + "description": "The toilet paper is to the right of the speaker.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "right.", + "B": "left.", + "C": "behind.", + "D": "It’s slightly in front of the speaker rather than to the right.", + "E": "The toilet paper is on the left side relative to the speaker, not the right." + }, + "answer_label": "A", + "answer_text": "right.", + "q_id": "scene0748_00-0_8", + "correct_answer": "right.", + "original_options": { + "A": "right.", + "B": "left.", + "C": "behind.", + "D": "It’s slightly in front of the speaker rather than to the right.", + "E": "The toilet paper is on the left side relative to the speaker, not the right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 15 + ] + }, + { + "video_uid": "scene0748_00-0", + "video_length": "00:00:20", + "qid": "scene0748_00-0_10", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the paper that was just on the table?", + "answer": "No.", + "question_chinese": "我能碰到刚在桌子上的纸吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The paper on the table is not close.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "scene0748_00-0_10", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0748_00-0", + "video_length": "00:00:20", + "qid": "scene0748_00-0_9", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the paper that was just on the table?", + "answer": "Yes.", + "question_chinese": "我能碰到刚在桌子上的纸吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The paper on the table is very close.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "scene0748_00-0_9", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0748_00-0", + "video_length": "00:00:20", + "qid": "scene0748_00-0_5", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the door or the speaker?", + "answer": "Door.", + "question_chinese": "门和音响哪个离我更近？", + "answer_chinese": "门", + "evidence": [ + { + "description": "The door is behind me, and the sound system is further back.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Speaker.", + "B": "Door." + }, + "answer_label": "B", + "answer_text": "Door.", + "q_id": "scene0748_00-0_5", + "correct_answer": "Door.", + "original_options": { + "A": "Speaker.", + "B": "Door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0748_00-0", + "video_length": "00:00:20", + "qid": "scene0748_00-0_6", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the door or the speaker?", + "answer": "Speaker.", + "question_chinese": "门和音响哪个离我更近？", + "answer_chinese": "音响", + "evidence": [ + { + "description": "The speakers are behind, and the door is even further back.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Speaker.", + "B": "Door." + }, + "answer_label": "A", + "answer_text": "Speaker.", + "q_id": "scene0748_00-0_6", + "correct_answer": "Speaker.", + "original_options": { + "A": "Speaker.", + "B": "Door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0748_00-0", + "video_length": "00:00:20", + "qid": "scene0748_00-0_3", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the door from me?", + "answer": "Directly behind.", + "question_chinese": "门在我的什么方向？", + "answer_chinese": "正后方", + "evidence": [ + { + "description": "The door is directly behind.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Directly in front.", + "B": "To your left.", + "C": "Slightly to your right, just ahead of where you are standing now.", + "D": "Behind you but offset to the left, so you would need to turn and step sideways to face it.", + "E": "Directly behind." + }, + "answer_label": "E", + "answer_text": "Directly behind.", + "q_id": "scene0748_00-0_3", + "correct_answer": "Directly behind.", + "original_options": { + "A": "Directly in front.", + "B": "To your left.", + "C": "Slightly to your right, just ahead of where you are standing now.", + "D": "Behind you but offset to the left, so you would need to turn and step sideways to face it.", + "E": "Directly behind." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 12, + 19 + ] + }, + { + "video_uid": "scene0748_00-0", + "video_length": "00:00:20", + "qid": "scene0748_00-0_4", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the door from me?", + "answer": "Right side.", + "question_chinese": "门在我的什么方向？", + "answer_chinese": "右方", + "evidence": [ + { + "description": "The door is on the right.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "In front.", + "B": "Behind you.", + "C": "Slightly in front of you, not on your right side at all.", + "D": "Right side.", + "E": "On your left side rather than on your right, away from the right-hand side." + }, + "answer_label": "D", + "answer_text": "Right side.", + "q_id": "scene0748_00-0_4", + "correct_answer": "Right side.", + "original_options": { + "A": "In front.", + "B": "Behind you.", + "C": "Slightly in front of you, not on your right side at all.", + "D": "Right side.", + "E": "On your left side rather than on your right, away from the right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 14 + ] + } +] diff --git a/QAs/scene0749_00-0-mcq.json b/QAs/scene0749_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..bb3800106fa90ef5bf307a74c3fe98abd1542cc2 --- /dev/null +++ b/QAs/scene0749_00-0-mcq.json @@ -0,0 +1,653 @@ +[ + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_5_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the microwave relative to me?", + "answer": "It’s to the left of the gas stove behind you.", + "question_chinese": "微波炉在我的什么方向？", + "answer_chinese": "在你后面的燃气灶的左边", + "evidence": [ + { + "description": "The microwave is to the left of the gas stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your right.", + "C": "It’s to the left of the gas stove behind you.", + "D": "It’s to the right of the gas stove behind you, not to the left.", + "E": "It’s to the left of the gas stove in front of you, not behind you." + }, + "answer_label": "C", + "answer_text": "It’s to the left of the gas stove behind you.", + "q_id": "scene0749_00-0_5_T1", + "correct_answer": "It’s to the left of the gas stove behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your right.", + "C": "It’s to the left of the gas stove behind you.", + "D": "It’s to the right of the gas stove behind you, not to the left.", + "E": "It’s to the left of the gas stove in front of you, not behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 15, + 16 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_5_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the microwave relative to me?", + "answer": "It’s to the left of the gas stove to your right.", + "question_chinese": "微波炉在我的什么方向？", + "answer_chinese": "在你右边的燃气灶的左边", + "evidence": [ + { + "description": "The microwave is to the left of the gas stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s to the left of the gas stove to your right.", + "B": "Behind you.", + "C": "To your left.", + "D": "It’s to the right of the gas stove to your right.", + "E": "It’s to the left of the gas stove behind you, not to your right." + }, + "answer_label": "A", + "answer_text": "It’s to the left of the gas stove to your right.", + "q_id": "scene0749_00-0_5_T2", + "correct_answer": "It’s to the left of the gas stove to your right.", + "original_options": { + "A": "It’s to the left of the gas stove to your right.", + "B": "Behind you.", + "C": "To your left.", + "D": "It’s to the right of the gas stove to your right.", + "E": "It’s to the left of the gas stove behind you, not to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 12, + 15 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_9", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the microwave oven relative to the gas stove?", + "answer": "left.", + "question_chinese": "微波炉在燃气灶的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "The microwave is to the left of the gas stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "left.", + "B": "Right.", + "C": "Above.", + "D": "The microwave is positioned to the right of the gas stove, not on the left.", + "E": "The microwave sits above the gas stove rather than to the left side." + }, + "answer_label": "A", + "answer_text": "left.", + "q_id": "scene0749_00-0_9", + "correct_answer": "left.", + "original_options": { + "A": "left.", + "B": "Right.", + "C": "Above.", + "D": "The microwave is positioned to the right of the gas stove, not on the left.", + "E": "The microwave sits above the gas stove rather than to the left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 13 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_10", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the microwave oven relative to the gas stove?", + "answer": "left.", + "question_chinese": "微波炉在燃气灶的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "The microwave is to the left of the gas stove.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right.", + "B": "Above.", + "C": "left.", + "D": "The microwave is located to the right of the gas stove, not to the left.", + "E": "The microwave is above the gas stove, rather than positioned to its left side." + }, + "answer_label": "C", + "answer_text": "left.", + "q_id": "scene0749_00-0_10", + "correct_answer": "left.", + "original_options": { + "A": "Right.", + "B": "Above.", + "C": "left.", + "D": "The microwave is located to the right of the gas stove, not to the left.", + "E": "The microwave is above the gas stove, rather than positioned to its left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_1_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see next to the table?", + "answer": "You saw two chairs next to the table in front of you.", + "question_chinese": "我看见桌子旁有几个椅子？", + "answer_chinese": "你看见面前的桌子旁有两个椅子", + "evidence": [ + { + "description": "saw two chairs", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw two chairs next to the table in front of you.", + "B": "One chair.", + "C": "Three chairs.", + "D": "You saw four chairs next to the table in front of you.", + "E": "You saw zero chairs next to the table in front of you." + }, + "answer_label": "A", + "answer_text": "You saw two chairs next to the table in front of you.", + "q_id": "scene0749_00-0_1_T2", + "correct_answer": "You saw two chairs next to the table in front of you.", + "original_options": { + "A": "You saw two chairs next to the table in front of you.", + "B": "One chair.", + "C": "Three chairs.", + "D": "You saw four chairs next to the table in front of you.", + "E": "You saw zero chairs next to the table in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 12 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_1_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see next to the table?", + "answer": "You saw four chairs next to the table in front of you.", + "question_chinese": "我看见桌子旁有几个椅子？", + "answer_chinese": "你看见面前的桌子旁有四个椅子", + "evidence": [ + { + "description": "saw two chairs", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "saw two more chairs", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw three chairs.", + "B": "You saw five chairs.", + "C": "You saw two chairs next to the table directly in front of you.", + "D": "You saw six chairs arranged next to the table in front of you.", + "E": "You saw four chairs next to the table in front of you." + }, + "answer_label": "E", + "answer_text": "You saw four chairs next to the table in front of you.", + "q_id": "scene0749_00-0_1_T1", + "correct_answer": "You saw four chairs next to the table in front of you.", + "original_options": { + "A": "You saw three chairs.", + "B": "You saw five chairs.", + "C": "You saw two chairs next to the table directly in front of you.", + "D": "You saw six chairs arranged next to the table in front of you.", + "E": "You saw four chairs next to the table in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 13 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_7", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the gas stove?", + "answer": "Yes.", + "question_chinese": "我能碰到燃气灶吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The gas stove is at close range from the back.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "scene0749_00-0_7", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_8", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the gas stove?", + "answer": "No.", + "question_chinese": "我能碰到燃气灶吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The gas stove is at close range from the back. ...", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "scene0749_00-0_8", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_2_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the gas stove or the trash can?", + "answer": "The trash can is closer to you.", + "question_chinese": "燃气灶和垃圾桶哪个离我更近？", + "answer_chinese": "垃圾桶离你更近", + "evidence": [ + { + "description": "The gas stove is closer than the trash can.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The gas stove is closer to you.", + "B": "The trash can is closer to you." + }, + "answer_label": "B", + "answer_text": "The trash can is closer to you.", + "q_id": "scene0749_00-0_2_T1", + "correct_answer": "The trash can is closer to you.", + "original_options": { + "A": "The gas stove is closer to you.", + "B": "The trash can is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_3", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the gas stove or the trash can?", + "answer": "gas stove.", + "question_chinese": "燃气灶和垃圾桶哪个离我更近？", + "answer_chinese": "燃气灶", + "evidence": [ + { + "description": "The gas stove is closer than the trash can.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "gas stove.", + "B": "trash can." + }, + "answer_label": "A", + "answer_text": "gas stove.", + "q_id": "scene0749_00-0_3", + "correct_answer": "gas stove.", + "original_options": { + "A": "gas stove.", + "B": "trash can." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_2_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the gas stove or the trash can?", + "answer": "The gas stove is closer to you.", + "question_chinese": "燃气灶和垃圾桶哪个离我更近？", + "answer_chinese": "燃气灶离你更近", + "evidence": [ + { + "description": "The gas stove is closer than the trash can.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The trash can is closer to you.", + "B": "The gas stove is closer to you." + }, + "answer_label": "B", + "answer_text": "The gas stove is closer to you.", + "q_id": "scene0749_00-0_2_T2", + "correct_answer": "The gas stove is closer to you.", + "original_options": { + "A": "The trash can is closer to you.", + "B": "The gas stove is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_5", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the door from me?", + "answer": "left.", + "question_chinese": "门在我的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "The door is on my left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "left.", + "B": "right.", + "C": "in front.", + "D": "The door is straight ahead of you, not off to your left.", + "E": "The door is positioned to your right side relative to where you're standing." + }, + "answer_label": "A", + "answer_text": "left.", + "q_id": "scene0749_00-0_5", + "correct_answer": "left.", + "original_options": { + "A": "left.", + "B": "right.", + "C": "in front.", + "D": "The door is straight ahead of you, not off to your left.", + "E": "The door is positioned to your right side relative to where you're standing." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 13 + ] + }, + { + "video_uid": "scene0749_00-0", + "video_length": "00:00:20", + "qid": "scene0749_00-0_6", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the door from me?", + "answer": "behind.", + "question_chinese": "门在我的什么方向？", + "answer_chinese": "后面", + "evidence": [ + { + "description": "The door is behind me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "in front.", + "B": "left.", + "C": "The door is to your left side rather than directly behind you.", + "D": "behind.", + "E": "The door is in front of you, so you wouldn't need to turn around to face it." + }, + "answer_label": "D", + "answer_text": "behind.", + "q_id": "scene0749_00-0_6", + "correct_answer": "behind.", + "original_options": { + "A": "in front.", + "B": "left.", + "C": "The door is to your left side rather than directly behind you.", + "D": "behind.", + "E": "The door is in front of you, so you wouldn't need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 1, + 12, + 17 + ] + } +] diff --git a/QAs/scene0751_00-0-mcq.json b/QAs/scene0751_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e8f1d9a3cb2b56a725aba2658732f59fa75bf8ca --- /dev/null +++ b/QAs/scene0751_00-0-mcq.json @@ -0,0 +1,397 @@ +[ + { + "video_uid": "scene0751_00-0", + "video_length": "00:00:20", + "qid": "scene0751_00-0_7", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the shampoo on the sink?", + "answer": "Yes.", + "question_chinese": "我能碰到洗手台上的洗发水吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The shampoo is very close.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "scene0751_00-0_7", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0751_00-0", + "video_length": "00:00:20", + "qid": "scene0751_00-0_8", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the shampoo on the sink?", + "answer": "No.", + "question_chinese": "我能碰到洗手台上的洗发水吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The shampoo is not close.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "scene0751_00-0_8", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0751_00-0", + "video_length": "00:00:20", + "qid": "scene0751_00-0_9", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the sink from me?", + "answer": "back.", + "question_chinese": "洗手台在我的什么方向？", + "answer_chinese": "在后面", + "evidence": [ + { + "description": "The sink is in the back.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "front.", + "B": "left.", + "C": "The sink is in front of you, not behind you at all.", + "D": "back.", + "E": "It’s positioned off to your left side rather than directly behind you." + }, + "answer_label": "D", + "answer_text": "back.", + "q_id": "scene0751_00-0_9", + "correct_answer": "back.", + "original_options": { + "A": "front.", + "B": "left.", + "C": "The sink is in front of you, not behind you at all.", + "D": "back.", + "E": "It’s positioned off to your left side rather than directly behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "scene0751_00-0", + "video_length": "00:00:20", + "qid": "scene0751_00-0_10", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the sink from me?", + "answer": "right.", + "question_chinese": "洗手台在我的什么方向？", + "answer_chinese": "右边", + "evidence": [ + { + "description": "The sink is on the right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:00" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "right.", + "B": "left.", + "C": "back.", + "D": "The sink is directly in front of you, not on your right.", + "E": "It’s behind you instead of being positioned to your right side." + }, + "answer_label": "A", + "answer_text": "right.", + "q_id": "scene0751_00-0_10", + "correct_answer": "right.", + "original_options": { + "A": "right.", + "B": "left.", + "C": "back.", + "D": "The sink is directly in front of you, not on your right.", + "E": "It’s behind you instead of being positioned to your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "scene0751_00-0", + "video_length": "00:00:20", + "qid": "scene0751_00-0_1_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the toothpaste relative to me?", + "answer": "In front of you.", + "question_chinese": "牙膏相对于我是什么方位？", + "answer_chinese": "你的前面", + "evidence": [ + { + "description": "There are two tubes of toothpaste on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "In front of you.", + "D": "It is positioned to your right side rather than directly in front of you.", + "E": "You would find it above you instead of straight ahead in your view." + }, + "answer_label": "C", + "answer_text": "In front of you.", + "q_id": "scene0751_00-0_1_T1", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "In front of you.", + "D": "It is positioned to your right side rather than directly in front of you.", + "E": "You would find it above you instead of straight ahead in your view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 14, + 13 + ] + }, + { + "video_uid": "scene0751_00-0", + "video_length": "00:00:20", + "qid": "scene0751_00-0_1_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the toothpaste relative to me?", + "answer": "Behind you.", + "question_chinese": "牙膏相对于我是什么方位？", + "answer_chinese": "你的后面", + "evidence": [ + { + "description": "There are two tubes of toothpaste on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + }, + { + "description": "You turn around.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "In front of you.", + "C": "To your right.", + "D": "It is located to your left side rather than behind your body.", + "E": "You would see it above you, not positioned behind you at all." + }, + "answer_label": "A", + "answer_text": "Behind you.", + "q_id": "scene0751_00-0_1_T2", + "correct_answer": "Behind you.", + "original_options": { + "A": "Behind you.", + "B": "In front of you.", + "C": "To your right.", + "D": "It is located to your left side rather than behind your body.", + "E": "You would see it above you, not positioned behind you at all." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 12, + 12 + ] + }, + { + "video_uid": "scene0751_00-0", + "video_length": "00:00:20", + "qid": "scene0751_00-0_4", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer, the toilet or the sink?", + "answer": "Sink.", + "question_chinese": "马桶和洗手台哪个离得近？", + "answer_chinese": "洗手台", + "evidence": [ + { + "description": "The sink is closer than the toilet.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Toilet.", + "B": "Sink." + }, + "answer_label": "B", + "answer_text": "Sink.", + "q_id": "scene0751_00-0_4", + "correct_answer": "Sink.", + "original_options": { + "A": "Toilet.", + "B": "Sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0751_00-0", + "video_length": "00:00:20", + "qid": "scene0751_00-0_5", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer, the toilet or the sink?", + "answer": "Toilet.", + "question_chinese": "马桶和洗手台哪个离得近？", + "answer_chinese": "马桶", + "evidence": [ + { + "description": "The sink is farther away from the toilet.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Toilet.", + "B": "Sink." + }, + "answer_label": "A", + "answer_text": "Toilet.", + "q_id": "scene0751_00-0_5", + "correct_answer": "Toilet.", + "original_options": { + "A": "Toilet.", + "B": "Sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/scene0752_00-0-mcq.json b/QAs/scene0752_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..52efea808ef9a066ea6eac75ce67fff0b72a72f0 --- /dev/null +++ b/QAs/scene0752_00-0-mcq.json @@ -0,0 +1,524 @@ +[ + { + "video_uid": "scene0752_00-0", + "video_length": "00:00:20", + "qid": "scene0752_00-0_2_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the trash can or the pillow?", + "answer": "The trash can is closer to you.", + "question_chinese": "垃圾桶和枕头哪个离我更近？", + "answer_chinese": "垃圾桶离你更近", + "evidence": [ + { + "description": "The trash can is closer than the pillow.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The pillow is closer to you.", + "B": "The trash can is closer to you." + }, + "answer_label": "B", + "answer_text": "The trash can is closer to you.", + "q_id": "scene0752_00-0_2_T1", + "correct_answer": "The trash can is closer to you.", + "original_options": { + "A": "The pillow is closer to you.", + "B": "The trash can is closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0752_00-0", + "video_length": "00:00:20", + "qid": "scene0752_00-0_2_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the trash can or the pillow?", + "answer": "The pillow is closer to you.", + "question_chinese": "垃圾桶和枕头哪个离我更近？", + "answer_chinese": "枕头离你更近", + "evidence": [ + { + "description": "I saw the trash can.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + }, + { + "description": "I saw the pillow.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The pillow is closer to you.", + "B": "The trash can is closer to you." + }, + "answer_label": "A", + "answer_text": "The pillow is closer to you.", + "q_id": "scene0752_00-0_2_T2", + "correct_answer": "The pillow is closer to you.", + "original_options": { + "A": "The pillow is closer to you.", + "B": "The trash can is closer to you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0752_00-0", + "video_length": "00:00:20", + "qid": "scene0752_00-0_7", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the trash can from me?", + "answer": "back.", + "question_chinese": "垃圾桶在我的什么方向？", + "answer_chinese": "后面", + "evidence": [ + { + "description": "The trash can is behind me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "left.", + "B": "right.", + "C": "back.", + "D": "It is in front of you rather than behind you, so it is not at your back.", + "E": "It is above you instead of behind you, so the trash can is not at your back." + }, + "answer_label": "C", + "answer_text": "back.", + "q_id": "scene0752_00-0_7", + "correct_answer": "back.", + "original_options": { + "A": "left.", + "B": "right.", + "C": "back.", + "D": "It is in front of you rather than behind you, so it is not at your back.", + "E": "It is above you instead of behind you, so the trash can is not at your back." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 17 + ] + }, + { + "video_uid": "scene0752_00-0", + "video_length": "00:00:20", + "qid": "scene0752_00-0_8", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the trash can from me?", + "answer": "left.", + "question_chinese": "垃圾桶在我的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "The trash can is on the left", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "left.", + "B": "right.", + "C": "back.", + "D": "It is in front of you instead of on your left, so it is not to your left.", + "E": "It is behind you rather than to your left, meaning it is not positioned on your left." + }, + "answer_label": "A", + "answer_text": "left.", + "q_id": "scene0752_00-0_8", + "correct_answer": "left.", + "original_options": { + "A": "left.", + "B": "right.", + "C": "back.", + "D": "It is in front of you instead of on your left, so it is not to your left.", + "E": "It is behind you rather than to your left, meaning it is not positioned on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 18, + 17 + ] + }, + { + "video_uid": "scene0752_00-0", + "video_length": "00:00:20", + "qid": "scene0752_00-0_1_T2", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many beds are in my room?", + "answer": "One; one is to your right.", + "question_chinese": "我的房间里有几张床？", + "answer_chinese": "一张，一张在你右边", + "evidence": [ + { + "description": "saw a bed", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero beds.", + "B": "Two beds.", + "C": "Two; one is to your right and one is to your left.", + "D": "Three; one is to your right, one is in front of you, and one is to your left.", + "E": "One; one is to your right." + }, + "answer_label": "E", + "answer_text": "One; one is to your right.", + "q_id": "scene0752_00-0_1_T2", + "correct_answer": "One; one is to your right.", + "original_options": { + "A": "Zero beds.", + "B": "Two beds.", + "C": "Two; one is to your right and one is to your left.", + "D": "Three; one is to your right, one is in front of you, and one is to your left.", + "E": "One; one is to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 12, + 18 + ] + }, + { + "video_uid": "scene0752_00-0", + "video_length": "00:00:20", + "qid": "scene0752_00-0_1_T1", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many beds are in my room?", + "answer": "Two: one is in front of you, and one is to your right.", + "question_chinese": "我的房间里有几张床？", + "answer_chinese": "两张，一张在你前面，一张在你右边", + "evidence": [ + { + "description": "saw a bed", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + }, + { + "description": "see one more bed", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero beds.", + "B": "One bed.", + "C": "One: the only bed is to your left, not in front of you.", + "D": "Two: one is in front of you, and one is to your right.", + "E": "Three: one is in front of you, one is to your right, and one is to your left." + }, + "answer_label": "D", + "answer_text": "Two: one is in front of you, and one is to your right.", + "q_id": "scene0752_00-0_1_T1", + "correct_answer": "Two: one is in front of you, and one is to your right.", + "original_options": { + "A": "Zero beds.", + "B": "One bed.", + "C": "One: the only bed is to your left, not in front of you.", + "D": "Two: one is in front of you, and one is to your right.", + "E": "Three: one is in front of you, one is to your right, and one is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 18 + ] + }, + { + "video_uid": "scene0752_00-0", + "video_length": "00:00:20", + "qid": "scene0752_00-0_5_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the red clothes on the chair?", + "answer": "Yes. The red clothes are directly behind you.", + "question_chinese": "我能碰到椅子上的红色衣服吗？", + "answer_chinese": "能，红色衣服在你正后面。", + "evidence": [ + { + "description": "The red clothes on the chair are very close.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. The red clothes are to your back-right.", + "B": "Yes. The red clothes are directly behind you." + }, + "answer_label": "B", + "answer_text": "Yes. The red clothes are directly behind you.", + "q_id": "scene0752_00-0_5_T2", + "correct_answer": "Yes. The red clothes are directly behind you.", + "original_options": { + "A": "No. The red clothes are to your back-right.", + "B": "Yes. The red clothes are directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0752_00-0", + "video_length": "00:00:20", + "qid": "scene0752_00-0_5_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the red clothes on the chair?", + "answer": "No. The red clothes are to your back-right.", + "question_chinese": "我能碰到椅子上的红色衣服吗？", + "answer_chinese": "不能，红色衣服在你的右后方。", + "evidence": [ + { + "description": "The red clothes on the chair are very close.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No. The red clothes are to your back-right.", + "B": "Yes. The red clothes are directly behind you." + }, + "answer_label": "A", + "answer_text": "No. The red clothes are to your back-right.", + "q_id": "scene0752_00-0_5_T1", + "correct_answer": "No. The red clothes are to your back-right.", + "original_options": { + "A": "No. The red clothes are to your back-right.", + "B": "Yes. The red clothes are directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0752_00-0", + "video_length": "00:00:20", + "qid": "scene0752_00-0_3_T1", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the blue bed relative to me?", + "answer": "The blue bed is to your left.", + "question_chinese": "蓝色的床在我的什么方位？", + "answer_chinese": "蓝色的床在你的左边", + "evidence": [ + { + "description": "The blue bed is on the left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The white bed is on the right.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The blue bed is to your left.", + "B": "The blue bed is to your right.", + "C": "The blue bed is in front of you.", + "D": "The blue bed is behind you, not beside you as you face forward.", + "E": "The blue bed sits at your front-right, slightly ahead rather than directly at your side." + }, + "answer_label": "A", + "answer_text": "The blue bed is to your left.", + "q_id": "scene0752_00-0_3_T1", + "correct_answer": "The blue bed is to your left.", + "original_options": { + "A": "The blue bed is to your left.", + "B": "The blue bed is to your right.", + "C": "The blue bed is in front of you.", + "D": "The blue bed is behind you, not beside you as you face forward.", + "E": "The blue bed sits at your front-right, slightly ahead rather than directly at your side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 8, + 13, + 15 + ] + }, + { + "video_uid": "scene0752_00-0", + "video_length": "00:00:20", + "qid": "scene0752_00-0_3_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the blue bed relative to me?", + "answer": "The blue bed is behind you.", + "question_chinese": "蓝色的床在我的什么方位？", + "answer_chinese": "蓝色的床在你的后方。", + "evidence": [ + { + "description": "The blue bed is on the left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The white bed is on the right.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The blue bed is in front of you.", + "B": "The blue bed is behind you.", + "C": "The blue bed is to your left.", + "D": "The blue bed is to your right, not located behind you in your view.", + "E": "The blue bed is at your front-left, slightly ahead instead of directly behind you." + }, + "answer_label": "B", + "answer_text": "The blue bed is behind you.", + "q_id": "scene0752_00-0_3_T2", + "correct_answer": "The blue bed is behind you.", + "original_options": { + "A": "The blue bed is in front of you.", + "B": "The blue bed is behind you.", + "C": "The blue bed is to your left.", + "D": "The blue bed is to your right, not located behind you in your view.", + "E": "The blue bed is at your front-left, slightly ahead instead of directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 14, + 14 + ] + } +] diff --git a/QAs/scene0753_00-0-mcq.json b/QAs/scene0753_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..498378698e20123739ec504ecf0053f15b3bc4f9 --- /dev/null +++ b/QAs/scene0753_00-0-mcq.json @@ -0,0 +1,415 @@ +[ + { + "video_uid": "scene0753_00-0", + "video_length": "00:00:20", + "qid": "scene0753_00-0_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the table with four drawers from me?", + "answer": "left.", + "question_chinese": "带四层抽屉的桌子在我的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "The desk with four drawers is on my left.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Right.", + "B": "In front.", + "C": "left.", + "D": "Behind you, not positioned on your left side where you’re standing.", + "E": "To your front-right, slightly ahead rather than directly to your left side." + }, + "answer_label": "C", + "answer_text": "left.", + "q_id": "scene0753_00-0_5", + "correct_answer": "left.", + "original_options": { + "A": "Right.", + "B": "In front.", + "C": "left.", + "D": "Behind you, not positioned on your left side where you’re standing.", + "E": "To your front-right, slightly ahead rather than directly to your left side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 12 + ] + }, + { + "video_uid": "scene0753_00-0", + "video_length": "00:00:20", + "qid": "scene0753_00-0_6", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the table with four drawers from me?", + "answer": "back.", + "question_chinese": "带四层抽屉的桌子在我的什么方向？", + "answer_chinese": "后面", + "evidence": [ + { + "description": "The desk with four drawers is behind me.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Left.", + "B": "In front.", + "C": "To your right, not located behind you as you face forward now.", + "D": "back.", + "E": "At your front-left, slightly ahead and off to the side rather than directly behind you." + }, + "answer_label": "D", + "answer_text": "back.", + "q_id": "scene0753_00-0_6", + "correct_answer": "back.", + "original_options": { + "A": "Left.", + "B": "In front.", + "C": "To your right, not located behind you as you face forward now.", + "D": "back.", + "E": "At your front-left, slightly ahead and off to the side rather than directly behind you." + }, + "optimization_rounds": 3, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 12, + 15 + ] + }, + { + "video_uid": "scene0753_00-0", + "video_length": "00:00:20", + "qid": "scene0753_00-0_4_T2", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the phone on the table?", + "answer": "You could touch the phone on the table; it was in front of you.", + "question_chinese": "我能碰到桌上的电话吗？", + "answer_chinese": "你能碰到桌上的电话，它在你前面。", + "evidence": [ + { + "description": "You can touch the telephone on the table.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn’t touch the phone on the table; it was in front of you.", + "B": "You could touch the phone on the table; it was in front of you." + }, + "answer_label": "B", + "answer_text": "You could touch the phone on the table; it was in front of you.", + "q_id": "scene0753_00-0_4_T2", + "correct_answer": "You could touch the phone on the table; it was in front of you.", + "original_options": { + "A": "You couldn’t touch the phone on the table; it was in front of you.", + "B": "You could touch the phone on the table; it was in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "scene0753_00-0", + "video_length": "00:00:20", + "qid": "scene0753_00-0_4_T1", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the phone on the table?", + "answer": "You couldn’t touch the phone on the table; it was to your left.", + "question_chinese": "我能碰到桌上的电话吗？", + "answer_chinese": "你不能碰到桌上的电话，它在你左边", + "evidence": [ + { + "description": "You can not touch the telephone.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You couldn’t touch the phone on the table; it was to your left.", + "B": "You could touch the phone on the table; it was to your left." + }, + "answer_label": "A", + "answer_text": "You couldn’t touch the phone on the table; it was to your left.", + "q_id": "scene0753_00-0_4_T1", + "correct_answer": "You couldn’t touch the phone on the table; it was to your left.", + "original_options": { + "A": "You couldn’t touch the phone on the table; it was to your left.", + "B": "You could touch the phone on the table; it was to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "scene0753_00-0", + "video_length": "00:00:20", + "qid": "scene0753_00-0_1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many water bottles are on the table?", + "answer": "two.", + "question_chinese": "桌上有几个水瓶？", + "answer_chinese": "两个", + "evidence": [ + { + "description": "saw two water bottles", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four water bottles are on the table, lined up together rather than just two.", + "D": "two.", + "E": "Five water bottles are on the table, with several grouped closely in the same spot." + }, + "answer_label": "D", + "answer_text": "two.", + "q_id": "scene0753_00-0_1", + "correct_answer": "two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four water bottles are on the table, lined up together rather than just two.", + "D": "two.", + "E": "Five water bottles are on the table, with several grouped closely in the same spot." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "scene0753_00-0", + "video_length": "00:00:20", + "qid": "scene0753_00-0_2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many water bottles are on the table?", + "answer": "three.", + "question_chinese": "桌上有几个水瓶？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "saw two water bottles", + "timestamps": { + "start": "00:00:05", + "end": "00:00:10" + } + }, + { + "description": "saw one more water bottles", + "timestamps": { + "start": "00:00:14", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "One water bottle is on the table, not three bottles spread across the surface.", + "D": "three.", + "E": "Five water bottles are on the table, with multiple bottles clustered together in view." + }, + "answer_label": "D", + "answer_text": "three.", + "q_id": "scene0753_00-0_2", + "correct_answer": "three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "One water bottle is on the table, not three bottles spread across the surface.", + "D": "three.", + "E": "Five water bottles are on the table, with multiple bottles clustered together in view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 14 + ] + }, + { + "video_uid": "scene0753_00-0", + "video_length": "00:00:20", + "qid": "scene0753_00-0_2_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many chairs did I see?", + "answer": "At this moment, you saw 1 chair; it was in front of you.", + "question_chinese": "此时，我看见了几把椅子？", + "answer_chinese": "此时，你看见了1把椅子，它在你前面。", + "evidence": [ + { + "description": "saw a chair", + "timestamps": { + "start": "00:00:09", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 0 chairs.", + "B": "At this moment, you saw 1 chair; it was in front of you.", + "C": "You saw 2 chairs.", + "D": "At this moment, you saw 3 chairs; two were in front of you and one was off to the side.", + "E": "At this moment, you saw 4 chairs; they were grouped together ahead of you rather than a single chair." + }, + "answer_label": "B", + "answer_text": "At this moment, you saw 1 chair; it was in front of you.", + "q_id": "scene0753_00-0_2_T2", + "correct_answer": "At this moment, you saw 1 chair; it was in front of you.", + "original_options": { + "A": "You saw 0 chairs.", + "B": "At this moment, you saw 1 chair; it was in front of you.", + "C": "You saw 2 chairs.", + "D": "At this moment, you saw 3 chairs; two were in front of you and one was off to the side.", + "E": "At this moment, you saw 4 chairs; they were grouped together ahead of you rather than a single chair." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 19 + ] + }, + { + "video_uid": "scene0753_00-0", + "video_length": "00:00:20", + "qid": "scene0753_00-0_2_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many chairs did I see?", + "answer": "At this moment, you saw 2 chairs: one was in front of you, and the other was to your front-right.", + "question_chinese": "此时，我看见了几把椅子？", + "answer_chinese": "此时，你看见了2把椅子，一把在你前面，另一把在你右前方。", + "evidence": [ + { + "description": "saw two chairs.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "At this moment, you saw 2 chairs: one was in front of you, and the other was to your front-right.", + "B": "You saw 1 chair.", + "C": "You saw 3 chairs.", + "D": "At this moment, you saw 0 chairs; there weren’t any in front of you or to your front-right.", + "E": "At this moment, you saw 4 chairs: two were in front of you, and two were to your front-right." + }, + "answer_label": "A", + "answer_text": "At this moment, you saw 2 chairs: one was in front of you, and the other was to your front-right.", + "q_id": "scene0753_00-0_2_T1", + "correct_answer": "At this moment, you saw 2 chairs: one was in front of you, and the other was to your front-right.", + "original_options": { + "A": "At this moment, you saw 2 chairs: one was in front of you, and the other was to your front-right.", + "B": "You saw 1 chair.", + "C": "You saw 3 chairs.", + "D": "At this moment, you saw 0 chairs; there weren’t any in front of you or to your front-right.", + "E": "At this moment, you saw 4 chairs: two were in front of you, and two were to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 19 + ] + } +] diff --git a/QAs/scene0754_00-0-mcq.json b/QAs/scene0754_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..11decb57834547fc83fa04864172e8bd8d537b96 --- /dev/null +++ b/QAs/scene0754_00-0-mcq.json @@ -0,0 +1,634 @@ +[ + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_9", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many photos are on the wall?", + "answer": "five.", + "question_chinese": "墙上有几张照片？", + "answer_chinese": "六张", + "evidence": [ + { + "description": "There are six photos on the wall.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "Seven photos were on the wall in total, arranged in a single line across the wall.", + "D": "five.", + "E": "Eight photos were arranged across the wall, taking up more space than just five pictures." + }, + "answer_label": "D", + "answer_text": "five.", + "q_id": "scene0754_00-0_9", + "correct_answer": "five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "Seven photos were on the wall in total, arranged in a single line across the wall.", + "D": "five.", + "E": "Eight photos were arranged across the wall, taking up more space than just five pictures." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_10", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many photos are on the wall?", + "answer": "five.", + "question_chinese": "墙上有几张照片？", + "answer_chinese": "六张", + "evidence": [ + { + "description": "There are six photos on the wall.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Six.", + "C": "five.", + "D": "Seven photos were on the wall in total, arranged in a single row across the wall.", + "E": "Nine photos were displayed on the wall, spreading wider than a set of five pictures." + }, + "answer_label": "C", + "answer_text": "five.", + "q_id": "scene0754_00-0_10", + "correct_answer": "five.", + "original_options": { + "A": "Four.", + "B": "Six.", + "C": "five.", + "D": "Seven photos were on the wall in total, arranged in a single row across the wall.", + "E": "Nine photos were displayed on the wall, spreading wider than a set of five pictures." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 16, + 15 + ] + }, + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_5_T2", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many photos did I see on the wall?", + "answer": "You saw six photos on the wall on the right side.", + "question_chinese": "我看见墙上有几张照片？", + "answer_chinese": "你看见右侧的墙上有六张照片。", + "evidence": [ + { + "description": "There are six photos on the wall.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw six photos on the wall on the right side.", + "B": "Four photos on the right wall.", + "C": "Seven photos on the right wall.", + "D": "You saw five photos on the wall on the right side, not six.", + "E": "You saw eight photos on the wall on the right side, covering more space than you remembered." + }, + "answer_label": "A", + "answer_text": "You saw six photos on the wall on the right side.", + "q_id": "scene0754_00-0_5_T2", + "correct_answer": "You saw six photos on the wall on the right side.", + "original_options": { + "A": "You saw six photos on the wall on the right side.", + "B": "Four photos on the right wall.", + "C": "Seven photos on the right wall.", + "D": "You saw five photos on the wall on the right side, not six.", + "E": "You saw eight photos on the wall on the right side, covering more space than you remembered." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 17 + ] + }, + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_5_T1", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many photos did I see on the wall?", + "answer": "You saw eight photos on the walls on the right side and on the left side.", + "question_chinese": "我看见墙上有几张照片？", + "answer_chinese": "你看见右侧和左侧的墙上有八张照片。", + "evidence": [ + { + "description": "There are six photos on the wall.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:05" + } + }, + { + "description": "There are two on the opposite side.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Seven photos total.", + "B": "Nine photos total.", + "C": "You saw six photos on the walls on the right side and on the left side altogether.", + "D": "You saw ten photos on the walls on the right side and on the left side in total.", + "E": "You saw eight photos on the walls on the right side and on the left side." + }, + "answer_label": "E", + "answer_text": "You saw eight photos on the walls on the right side and on the left side.", + "q_id": "scene0754_00-0_5_T1", + "correct_answer": "You saw eight photos on the walls on the right side and on the left side.", + "original_options": { + "A": "Seven photos total.", + "B": "Nine photos total.", + "C": "You saw six photos on the walls on the right side and on the left side altogether.", + "D": "You saw ten photos on the walls on the right side and on the left side in total.", + "E": "You saw eight photos on the walls on the right side and on the left side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 17, + 18 + ] + }, + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_3", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the projector on the table?", + "answer": "Yes.", + "question_chinese": "我能碰到桌子上的投影仪吗？", + "answer_chinese": "能", + "evidence": [ + { + "description": "The projector on the table is very close.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "scene0754_00-0_3", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_4", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the projector on the table?", + "answer": "No.", + "question_chinese": "我能碰到桌子上的投影仪吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The projector on the table is out of reach for direct contact.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "scene0754_00-0_4", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "three.", + "question_chinese": "房间里有几个椅子？", + "answer_chinese": "三个", + "evidence": [ + { + "description": "saw three chairs", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "There are two chairs in the room, with one near the table and another by the wall.", + "D": "three.", + "E": "There are four chairs in the room, spread around the table and along the side of the room." + }, + "answer_label": "D", + "answer_text": "three.", + "q_id": "scene0754_00-0_1", + "correct_answer": "three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "There are two chairs in the room, with one near the table and another by the wall.", + "D": "three.", + "E": "There are four chairs in the room, spread around the table and along the side of the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 18 + ] + }, + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "five.", + "question_chinese": "房间里有几个椅子？", + "answer_chinese": "五个", + "evidence": [ + { + "description": "saw three chairs", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "saw two more chairs", + "timestamps": { + "start": "00:00:17", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "five.", + "B": "six.", + "C": "seven.", + "D": "There are six chairs in the room altogether, not counting any stools.", + "E": "You count seven chairs placed around the room, including one near the doorway." + }, + "answer_label": "A", + "answer_text": "five.", + "q_id": "scene0754_00-0_2", + "correct_answer": "five.", + "original_options": { + "A": "five.", + "B": "six.", + "C": "seven.", + "D": "There are six chairs in the room altogether, not counting any stools.", + "E": "You count seven chairs placed around the room, including one near the doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 13 + ] + }, + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_8", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the projector on the table from me?", + "answer": "right.", + "question_chinese": "桌上的投影仪在我的什么方向？", + "answer_chinese": "右边", + "evidence": [ + { + "description": "The projector on the table is on my right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "left.", + "B": "front.", + "C": "right.", + "D": "The projector is positioned directly in front of you on the table.", + "E": "From where you are standing, the projector sits to your left on the table." + }, + "answer_label": "C", + "answer_text": "right.", + "q_id": "scene0754_00-0_8", + "correct_answer": "right.", + "original_options": { + "A": "left.", + "B": "front.", + "C": "right.", + "D": "The projector is positioned directly in front of you on the table.", + "E": "From where you are standing, the projector sits to your left on the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 14 + ] + }, + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_7", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "In which direction is the projector on the table from me?", + "answer": "Front right.", + "question_chinese": "桌上的投影仪在我的什么方向？", + "answer_chinese": "右前方", + "evidence": [ + { + "description": "The projector is in front of me on the right.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "front.", + "B": "back right.", + "C": "It is straight ahead of you, centered on the front edge of the table.", + "D": "Front right.", + "E": "The projector is behind you to the right relative to your current position." + }, + "answer_label": "D", + "answer_text": "Front right.", + "q_id": "scene0754_00-0_7", + "correct_answer": "Front right.", + "original_options": { + "A": "front.", + "B": "back right.", + "C": "It is straight ahead of you, centered on the front edge of the table.", + "D": "Front right.", + "E": "The projector is behind you to the right relative to your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 14, + 13 + ] + }, + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_6", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction relative to the projector is the cup on the table?", + "answer": "inner side.", + "question_chinese": "桌上的水杯在投影仪的什么方向？", + "answer_chinese": "里侧", + "evidence": [ + { + "description": "The water cup on the table is on the inner side of the projector.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "inner side.", + "B": "outer side.", + "C": "near side.", + "D": "The cup is on the outer side of the projector on the table.", + "E": "Relative to the projector, the cup sits on the near side of it." + }, + "answer_label": "A", + "answer_text": "inner side.", + "q_id": "scene0754_00-0_6", + "correct_answer": "inner side.", + "original_options": { + "A": "inner side.", + "B": "outer side.", + "C": "near side.", + "D": "The cup is on the outer side of the projector on the table.", + "E": "Relative to the projector, the cup sits on the near side of it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 2, + 13, + 13 + ] + }, + { + "video_uid": "scene0754_00-0", + "video_length": "00:00:20", + "qid": "scene0754_00-0_5", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction relative to the projector is the cup on the table?", + "answer": "left.", + "question_chinese": "桌上的水杯在投影仪的什么方向？", + "answer_chinese": "左边", + "evidence": [ + { + "description": "The water cup on the table is to the left of the projector.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "right.", + "B": "left.", + "C": "in front.", + "D": "The cup is positioned to the right of the projector on the table.", + "E": "Relative to the projector, the cup sits in front of it on the tabletop." + }, + "answer_label": "B", + "answer_text": "left.", + "q_id": "scene0754_00-0_5", + "correct_answer": "left.", + "original_options": { + "A": "right.", + "B": "left.", + "C": "in front.", + "D": "The cup is positioned to the right of the projector on the table.", + "E": "Relative to the projector, the cup sits in front of it on the tabletop." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 2, + 13, + 14 + ] + } +] diff --git a/QAs/scene0755_00-0-mcq.json b/QAs/scene0755_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5f4a49ff315e0fbca81d9312c92b4fd78c64b5d8 --- /dev/null +++ b/QAs/scene0755_00-0-mcq.json @@ -0,0 +1,543 @@ +[ + { + "video_uid": "scene0755_00-0", + "video_length": "00:00:20", + "qid": "scene0755_00-0_3_T1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are the two blue chairs relative to me?", + "answer": "It was in front of you.", + "question_chinese": "两把蓝色椅子在相对于我的什么位置？", + "answer_chinese": "在你的前面", + "evidence": [ + { + "description": "Two blue chairs are in front of me", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "It was behind you.", + "C": "It was to your left.", + "D": "The two blue chairs were positioned behind you, out of your direct view.", + "E": "The two blue chairs were located off to your right side relative to you." + }, + "answer_label": "A", + "answer_text": "It was in front of you.", + "q_id": "scene0755_00-0_3_T1", + "correct_answer": "It was in front of you.", + "original_options": { + "A": "It was in front of you.", + "B": "It was behind you.", + "C": "It was to your left.", + "D": "The two blue chairs were positioned behind you, out of your direct view.", + "E": "The two blue chairs were located off to your right side relative to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 14 + ] + }, + { + "video_uid": "scene0755_00-0", + "video_length": "00:00:20", + "qid": "scene0755_00-0_3_T2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are the two blue chairs relative to me?", + "answer": "To your back-left.", + "question_chinese": "两把蓝色椅子在相对于我的什么位置？", + "answer_chinese": "在你的左后方", + "evidence": [ + { + "description": "Two blue chairs are in front of me", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "I turned to the right by about 45°", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your back-right.", + "B": "To your back-left.", + "C": "To your front-left.", + "D": "The two blue chairs are positioned to your back-right relative to where you are.", + "E": "The two blue chairs are located in front of you, slightly off to one side." + }, + "answer_label": "B", + "answer_text": "To your back-left.", + "q_id": "scene0755_00-0_3_T2", + "correct_answer": "To your back-left.", + "original_options": { + "A": "To your back-right.", + "B": "To your back-left.", + "C": "To your front-left.", + "D": "The two blue chairs are positioned to your back-right relative to where you are.", + "E": "The two blue chairs are located in front of you, slightly off to one side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 15 + ] + }, + { + "video_uid": "scene0755_00-0", + "video_length": "00:00:20", + "qid": "scene0755_00-0_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in this room?", + "answer": "Two.", + "question_chinese": "房间里有几个椅子", + "answer_chinese": "2个", + "evidence": [ + { + "description": "saw the chairs", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four chairs are in the room, positioned in different parts of the space.", + "E": "Five chairs are in the room, with two placed together and the rest separated." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "scene0755_00-0_1", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Three.", + "D": "Four chairs are in the room, positioned in different parts of the space.", + "E": "Five chairs are in the room, with two placed together and the rest separated." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "scene0755_00-0", + "video_length": "00:00:20", + "qid": "scene0755_00-0_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in this room?", + "answer": "Three.", + "question_chinese": "房间里有几个椅子", + "answer_chinese": "3个", + "evidence": [ + { + "description": "saw the chairs", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + }, + { + "description": "saw the chair", + "timestamps": { + "start": "00:00:15", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "One chair is in the room, sitting by itself near the wall.", + "E": "Five chairs are in the room, with several arranged near each other." + }, + "answer_label": "C", + "answer_text": "Three.", + "q_id": "scene0755_00-0_2", + "correct_answer": "Three.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "One chair is in the room, sitting by itself near the wall.", + "E": "Five chairs are in the room, with several arranged near each other." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 12, + 12 + ] + }, + { + "video_uid": "scene0755_00-0", + "video_length": "00:00:20", + "qid": "scene0755_00-0_4_T2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the painting or the cabinet?", + "answer": "The cabinet was closer to you.", + "question_chinese": "画和柜子哪个离我更近？", + "answer_chinese": "柜子离你更近", + "evidence": [ + { + "description": "The cabinet is closer to this person", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The cabinet was closer to you.", + "B": "The painting was closer to you." + }, + "answer_label": "A", + "answer_text": "The cabinet was closer to you.", + "q_id": "scene0755_00-0_4_T2", + "correct_answer": "The cabinet was closer to you.", + "original_options": { + "A": "The cabinet was closer to you.", + "B": "The painting was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0755_00-0", + "video_length": "00:00:20", + "qid": "scene0755_00-0_9", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, a painting or a cabinet？", + "answer": "Cabinet.", + "question_chinese": "画和柜子哪个离我更近", + "answer_chinese": "柜子", + "evidence": [ + { + "description": "The cabinet is closer to this person", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Painting.", + "B": "Cabinet." + }, + "answer_label": "B", + "answer_text": "Cabinet.", + "q_id": "scene0755_00-0_9", + "correct_answer": "Cabinet.", + "original_options": { + "A": "Painting.", + "B": "Cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0755_00-0", + "video_length": "00:00:20", + "qid": "scene0755_00-0_4_T1", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the painting or the cabinet?", + "answer": "The painting was closer to you.", + "question_chinese": "画和柜子哪个离我更近？", + "answer_chinese": "画离你更近", + "evidence": [ + { + "description": "The cabinet is closer to this person", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + }, + { + "description": "I took a few steps towards where the painting was hung.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The cabinet was closer to you.", + "B": "The painting was closer to you." + }, + "answer_label": "B", + "answer_text": "The painting was closer to you.", + "q_id": "scene0755_00-0_4_T1", + "correct_answer": "The painting was closer to you.", + "original_options": { + "A": "The cabinet was closer to you.", + "B": "The painting was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0755_00-0", + "video_length": "00:00:20", + "qid": "scene0755_00-0_2_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the door or the cabinet?", + "answer": "They were equally close to you.", + "question_chinese": "门和柜子哪个离我更近？", + "answer_chinese": "离你一样近", + "evidence": [ + { + "description": "I am between the door and the cupboard", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "They were equally close to you.", + "B": "The door was closer to you." + }, + "answer_label": "A", + "answer_text": "They were equally close to you.", + "q_id": "scene0755_00-0_2_T2", + "correct_answer": "They were equally close to you.", + "original_options": { + "A": "They were equally close to you.", + "B": "The door was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0755_00-0", + "video_length": "00:00:20", + "qid": "scene0755_00-0_2_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me: the door or the cabinet?", + "answer": "The cabinet was closer to you.", + "question_chinese": "门和柜子哪个离我更近？", + "answer_chinese": "柜子离你更近", + "evidence": [ + { + "description": "I was originally between the door and the cabinet, and I walked a few steps toward the cabinet.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The cabinet was closer to you.", + "B": "The door was closer to you." + }, + "answer_label": "A", + "answer_text": "The cabinet was closer to you.", + "q_id": "scene0755_00-0_2_T1", + "correct_answer": "The cabinet was closer to you.", + "original_options": { + "A": "The cabinet was closer to you.", + "B": "The door was closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0755_00-0", + "video_length": "00:00:20", + "qid": "scene0755_00-0_5", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to the whiteboard, the door or the cabinet", + "answer": "Cabinet.", + "question_chinese": "门和柜子哪个离白板更近", + "answer_chinese": "柜子", + "evidence": [ + { + "description": "To the left of the whiteboard is the cabinet, and to the left of the cabinet is the door", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Cabinet.", + "B": "Door." + }, + "answer_label": "A", + "answer_text": "Cabinet.", + "q_id": "scene0755_00-0_5", + "correct_answer": "Cabinet.", + "original_options": { + "A": "Cabinet.", + "B": "Door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0755_00-0", + "video_length": "00:00:20", + "qid": "scene0755_00-0_6", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to the whiteboard, the door or the cabinet?", + "answer": "Cabinet.", + "question_chinese": "门和柜子哪个离白板更近", + "answer_chinese": "柜子", + "evidence": [ + { + "description": "To the left of the whiteboard is the cabinet, and to the left of the cabinet is the door", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Door.", + "B": "Cabinet." + }, + "answer_label": "B", + "answer_text": "Cabinet.", + "q_id": "scene0755_00-0_6", + "correct_answer": "Cabinet.", + "original_options": { + "A": "Door.", + "B": "Cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/scene0756_00-0-mcq.json b/QAs/scene0756_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..47c273a411a80f402c70dde72459c7daf6c64107 --- /dev/null +++ b/QAs/scene0756_00-0-mcq.json @@ -0,0 +1,323 @@ +[ + { + "video_uid": "scene0756_00-0", + "video_length": "00:00:20", + "qid": "scene0756_00-0_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many posters are hung on the wall?", + "answer": "Four.", + "question_chinese": "墙上挂了几幅海报", + "answer_chinese": "四", + "evidence": [ + { + "description": "saw the paintings", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Two posters are visible, with a third partially hidden behind them on the wall.", + "E": "Six posters are arranged across the wall, filling most of the space above the furniture." + }, + "answer_label": "A", + "answer_text": "Four.", + "q_id": "scene0756_00-0_1", + "correct_answer": "Four.", + "original_options": { + "A": "Four.", + "B": "Three.", + "C": "Five.", + "D": "Two posters are visible, with a third partially hidden behind them on the wall.", + "E": "Six posters are arranged across the wall, filling most of the space above the furniture." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "scene0756_00-0", + "video_length": "00:00:20", + "qid": "scene0756_00-0_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many posters are hung on the wall?", + "answer": "Six.", + "question_chinese": "墙上挂了几幅海报", + "answer_chinese": "六", + "evidence": [ + { + "description": "saw the posters", + "timestamps": { + "start": "00:00:00", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Five.", + "B": "Seven.", + "C": "Four posters are hung in a neat row, leaving a noticeable gap on the wall.", + "D": "Six.", + "E": "Eight posters cover the wall area, with several placed close together side by side." + }, + "answer_label": "D", + "answer_text": "Six.", + "q_id": "scene0756_00-0_2", + "correct_answer": "Six.", + "original_options": { + "A": "Five.", + "B": "Seven.", + "C": "Four posters are hung in a neat row, leaving a noticeable gap on the wall.", + "D": "Six.", + "E": "Eight posters cover the wall area, with several placed close together side by side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 14 + ] + }, + { + "video_uid": "scene0756_00-0", + "video_length": "00:00:20", + "qid": "scene0756_00-0_8", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "One.", + "question_chinese": "房间里有几个椅子", + "answer_chinese": "1个", + "evidence": [ + { + "description": "saw the chair", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Zero.", + "B": "Two.", + "C": "Three chairs are in the room, with two grouped together and one set apart nearby.", + "D": "One.", + "E": "Four chairs are visible, lined up around the room rather than just a single seat." + }, + "answer_label": "D", + "answer_text": "One.", + "q_id": "scene0756_00-0_8", + "correct_answer": "One.", + "original_options": { + "A": "Zero.", + "B": "Two.", + "C": "Three chairs are in the room, with two grouped together and one set apart nearby.", + "D": "One.", + "E": "Four chairs are visible, lined up around the room rather than just a single seat." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "scene0756_00-0", + "video_length": "00:00:20", + "qid": "scene0756_00-0_9", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "Two.", + "question_chinese": "房间里有几个椅子", + "answer_chinese": "2个", + "evidence": [ + { + "description": "saw the chair", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "saw the chair", + "timestamps": { + "start": "00:00:18", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "Four chairs are present, with two on one side and two positioned opposite them.", + "D": "Five chairs are in the room, including a couple pushed near the wall and others closer to you.", + "E": "Two." + }, + "answer_label": "E", + "answer_text": "Two.", + "q_id": "scene0756_00-0_9", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "Four chairs are present, with two on one side and two positioned opposite them.", + "D": "Five chairs are in the room, including a couple pushed near the wall and others closer to you.", + "E": "Two." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 18 + ] + }, + { + "video_uid": "scene0756_00-0", + "video_length": "00:00:20", + "qid": "scene0756_00-0_5", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the window or the door", + "answer": "Window.", + "question_chinese": "窗户和门哪个离我更近", + "answer_chinese": "窗户", + "evidence": [ + { + "description": "saw the window", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + }, + { + "description": "saw the door", + "timestamps": { + "start": "00:00:09", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Window.", + "B": "Door." + }, + "answer_label": "A", + "answer_text": "Window.", + "q_id": "scene0756_00-0_5", + "correct_answer": "Window.", + "original_options": { + "A": "Window.", + "B": "Door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0756_00-0", + "video_length": "00:00:20", + "qid": "scene0756_00-0_6", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the window or the door", + "answer": "Window.", + "question_chinese": "窗户和门哪个离我更近", + "answer_chinese": "窗户", + "evidence": [ + { + "description": "saw the window", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + }, + { + "description": "saw the door", + "timestamps": { + "start": "00:00:09", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Window.", + "B": "Door." + }, + "answer_label": "A", + "answer_text": "Window.", + "q_id": "scene0756_00-0_6", + "correct_answer": "Window.", + "original_options": { + "A": "Window.", + "B": "Door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + } +] diff --git a/QAs/scene0757_00-0-mcq.json b/QAs/scene0757_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..87864834901862a8daa14f324c547d316213bff0 --- /dev/null +++ b/QAs/scene0757_00-0-mcq.json @@ -0,0 +1,316 @@ +[ + { + "video_uid": "scene0757_00-0", + "video_length": "00:00:20", + "qid": "scene0757_00-0_8", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the black schoolbag on the ground now", + "answer": "No, you need to turn left two steps and then right two steps, and then squat down to touch.", + "question_chinese": "我现在能触碰到地上的黑色书包吗", + "answer_chinese": "不能，你需要左转两步后右转两步，然后蹲下触碰", + "evidence": [ + { + "description": "The schoolbag is on the left side of the table", + "timestamps": { + "start": "00:00:03", + "end": "00:00:10" + } + }, + { + "description": "The person is on the right side of the table", + "timestamps": { + "start": "00:00:13", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No, you need to turn left two steps and then right two steps, and then squat down to touch.", + "B": "Yes, you can touch the black schoolbag right now without moving or changing your position." + }, + "answer_label": "A", + "answer_text": "No, you need to turn left two steps and then right two steps, and then squat down to touch.", + "q_id": "scene0757_00-0_8", + "correct_answer": "No, you need to turn left two steps and then right two steps, and then squat down to touch.", + "original_options": { + "A": "No, you need to turn left two steps and then right two steps, and then squat down to touch.", + "B": "Yes, you can touch the black schoolbag right now without moving or changing your position." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "scene0757_00-0", + "video_length": "00:00:20", + "qid": "scene0757_00-0_9", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the black schoolbag on the ground now", + "answer": "No, you need to turn left and go around the table. Your schoolbag is on the left side of the table. Squat down and touch it.", + "question_chinese": "我现在能触碰到地上的黑色书包吗", + "answer_chinese": "不能，你需要左转绕过桌子，书包在桌子的左侧，蹲下触碰", + "evidence": [ + { + "description": "The schoolbag is on the left side of the table", + "timestamps": { + "start": "00:00:03", + "end": "00:00:10" + } + }, + { + "description": "The person is in front of the table", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you can touch it right now; just squat down and reach it.", + "B": "No, you need to turn left and go around the table. Your schoolbag is on the left side of the table. Squat down and touch it." + }, + "answer_label": "B", + "answer_text": "No, you need to turn left and go around the table. Your schoolbag is on the left side of the table. Squat down and touch it.", + "q_id": "scene0757_00-0_9", + "correct_answer": "No, you need to turn left and go around the table. Your schoolbag is on the left side of the table. Squat down and touch it.", + "original_options": { + "A": "Yes, you can touch it right now; just squat down and reach it.", + "B": "No, you need to turn left and go around the table. Your schoolbag is on the left side of the table. Squat down and touch it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 13 + ] + }, + { + "video_uid": "scene0757_00-0", + "video_length": "00:00:20", + "qid": "scene0757_00-0_5", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "Two.", + "question_chinese": "房间里有几个椅子", + "answer_chinese": "2个", + "evidence": [ + { + "description": "saw the chairs", + "timestamps": { + "start": "00:00:00", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Three.", + "C": "There are four chairs in the room, with two on each side of the table.", + "D": "Two.", + "E": "There are five chairs total, including one pushed in near the cabinet." + }, + "answer_label": "D", + "answer_text": "Two.", + "q_id": "scene0757_00-0_5", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Three.", + "C": "There are four chairs in the room, with two on each side of the table.", + "D": "Two.", + "E": "There are five chairs total, including one pushed in near the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 12 + ] + }, + { + "video_uid": "scene0757_00-0", + "video_length": "00:00:20", + "qid": "scene0757_00-0_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "Two.", + "question_chinese": "房间里有几个椅子", + "answer_chinese": "2个", + "evidence": [ + { + "description": "saw the chair", + "timestamps": { + "start": "00:00:00", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Two.", + "C": "Four.", + "D": "There are three chairs in the room, with one beside the table and two near the wall.", + "E": "There are five chairs total, with a couple grouped together near the cabinet." + }, + "answer_label": "B", + "answer_text": "Two.", + "q_id": "scene0757_00-0_4", + "correct_answer": "Two.", + "original_options": { + "A": "One.", + "B": "Two.", + "C": "Four.", + "D": "There are three chairs in the room, with one beside the table and two near the wall.", + "E": "There are five chairs total, with a couple grouped together near the cabinet." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 13 + ] + }, + { + "video_uid": "scene0757_00-0", + "video_length": "00:00:20", + "qid": "scene0757_00-0_1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which side of the table is the chair on?", + "answer": "The white cabinet is on the left side of the table.", + "question_chinese": "白色柜子在桌子的哪一侧", + "answer_chinese": "白色柜子在桌子的左侧", + "evidence": [ + { + "description": "saw the table and the white cabinet", + "timestamps": { + "start": "00:00:00", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the right side.", + "B": "On the front side.", + "C": "The white cabinet is on the left side of the table.", + "D": "The chair is on the right side of the table, opposite where you are standing.", + "E": "The chair is positioned at the back side of the table, away from the table edge closest to you." + }, + "answer_label": "C", + "answer_text": "The white cabinet is on the left side of the table.", + "q_id": "scene0757_00-0_1", + "correct_answer": "The white cabinet is on the left side of the table.", + "original_options": { + "A": "On the right side.", + "B": "On the front side.", + "C": "The white cabinet is on the left side of the table.", + "D": "The chair is on the right side of the table, opposite where you are standing.", + "E": "The chair is positioned at the back side of the table, away from the table edge closest to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 19 + ] + }, + { + "video_uid": "scene0757_00-0", + "video_length": "00:00:20", + "qid": "scene0757_00-0_2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which side of the table is the chair on?", + "answer": "The white cabinet is on the left side of the table.", + "question_chinese": "白色柜子在桌子的哪一侧", + "answer_chinese": "白色柜子在桌子的左侧", + "evidence": [ + { + "description": "saw the table and the white cabinet", + "timestamps": { + "start": "00:00:00", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the right side.", + "B": "On the back side.", + "C": "The chair is on the right side of the table, not on the left where you first looked.", + "D": "The chair sits at the front side of the table, closer to you than the opposite edge.", + "E": "The white cabinet is on the left side of the table." + }, + "answer_label": "E", + "answer_text": "The white cabinet is on the left side of the table.", + "q_id": "scene0757_00-0_2", + "correct_answer": "The white cabinet is on the left side of the table.", + "original_options": { + "A": "On the right side.", + "B": "On the back side.", + "C": "The chair is on the right side of the table, not on the left where you first looked.", + "D": "The chair sits at the front side of the table, closer to you than the opposite edge.", + "E": "The white cabinet is on the left side of the table." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 17 + ] + } +] diff --git a/QAs/scene0758_00-0-mcq.json b/QAs/scene0758_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..430dc4a28d9ad1b41855410ba0ccb56f1918a609 --- /dev/null +++ b/QAs/scene0758_00-0-mcq.json @@ -0,0 +1,302 @@ +[ + { + "video_uid": "scene0758_00-0", + "video_length": "00:00:20", + "qid": "scene0758_00-0_9", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I reach the books on the cabinet now", + "answer": "No.", + "question_chinese": "我现在能否够得到柜子上的书", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The book is rather far from this person", + "timestamps": { + "start": "00:00:03", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "scene0758_00-0_9", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0758_00-0", + "video_length": "00:00:20", + "qid": "scene0758_00-0_10", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Can I reach the books on the cabinet now", + "answer": "Yes.", + "question_chinese": "我现在能否够得到柜子上的书", + "answer_chinese": "能", + "evidence": [ + { + "description": "This person is standing beside the cabinet and can touch the books", + "timestamps": { + "start": "00:00:03", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "scene0758_00-0_10", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0758_00-0", + "video_length": "00:00:20", + "qid": "scene0758_00-0_5", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Depending on where I stand, in which direction is the table on me?", + "answer": "The table is on your right.", + "question_chinese": "根据我站的位置，桌子在我的哪个方位", + "answer_chinese": "桌子在你的右边", + "evidence": [ + { + "description": "The table is on your right", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The table is on your left.", + "B": "The table is in front of you.", + "C": "The table is behind you, so you would need to turn around to face it.", + "D": "The table is on your left side, slightly forward from where you are standing.", + "E": "The table is on your right." + }, + "answer_label": "E", + "answer_text": "The table is on your right.", + "q_id": "scene0758_00-0_5", + "correct_answer": "The table is on your right.", + "original_options": { + "A": "The table is on your left.", + "B": "The table is in front of you.", + "C": "The table is behind you, so you would need to turn around to face it.", + "D": "The table is on your left side, slightly forward from where you are standing.", + "E": "The table is on your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 15, + 14 + ] + }, + { + "video_uid": "scene0758_00-0", + "video_length": "00:00:20", + "qid": "scene0758_00-0_6", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Depending on where I stand, in which direction is the table on me?", + "answer": "The table is in front of you.", + "question_chinese": "根据我站的位置，桌子在我的哪个方位", + "answer_chinese": "桌子在你的前面", + "evidence": [ + { + "description": "The table is in front of you", + "timestamps": { + "start": "00:00:00", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The table is behind you.", + "B": "The table is in front of you.", + "C": "The table is to your left.", + "D": "The table is off to your right side, not directly in front of you as you face forward.", + "E": "The table is directly beside you, so it is not positioned straight ahead in your view." + }, + "answer_label": "B", + "answer_text": "The table is in front of you.", + "q_id": "scene0758_00-0_6", + "correct_answer": "The table is in front of you.", + "original_options": { + "A": "The table is behind you.", + "B": "The table is in front of you.", + "C": "The table is to your left.", + "D": "The table is off to your right side, not directly in front of you as you face forward.", + "E": "The table is directly beside you, so it is not positioned straight ahead in your view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 18, + 16 + ] + }, + { + "video_uid": "scene0758_00-0", + "video_length": "00:00:20", + "qid": "scene0758_00-0_7", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction of the door is the blue bear figurine", + "answer": "The blue bear figurine is on the right side of the door.", + "question_chinese": "蓝色小熊摆件在门的哪个方向", + "answer_chinese": "蓝色小熊摆件在门的右边", + "evidence": [ + { + "description": "The blue bear figurine is on the right side of the door", + "timestamps": { + "start": "00:00:05", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The blue bear figurine is on the right side of the door.", + "B": "On the left side of the door.", + "C": "Directly in front of the door.", + "D": "The blue bear figurine is on the left side of the door, closer to the handle area than the frame.", + "E": "The blue bear figurine is positioned behind the door, not on the right side next to it." + }, + "answer_label": "A", + "answer_text": "The blue bear figurine is on the right side of the door.", + "q_id": "scene0758_00-0_7", + "correct_answer": "The blue bear figurine is on the right side of the door.", + "original_options": { + "A": "The blue bear figurine is on the right side of the door.", + "B": "On the left side of the door.", + "C": "Directly in front of the door.", + "D": "The blue bear figurine is on the left side of the door, closer to the handle area than the frame.", + "E": "The blue bear figurine is positioned behind the door, not on the right side next to it." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 20, + 17 + ] + }, + { + "video_uid": "scene0758_00-0", + "video_length": "00:00:20", + "qid": "scene0758_00-0_8", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction of the door is the blue bear figurine", + "answer": "The blue bear figurine is on the right side of the door.", + "question_chinese": "蓝色小熊摆件在门的哪个方向", + "answer_chinese": "蓝色小熊摆件在门的右边", + "evidence": [ + { + "description": "The blue bear figurine is on the right side of the door", + "timestamps": { + "start": "00:00:05", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the left side of the door.", + "B": "Behind the door.", + "C": "The blue bear figurine is above the door, positioned near the top edge rather than beside it.", + "D": "The blue bear figurine is on the far side of the doorway, not on the right side next to it.", + "E": "The blue bear figurine is on the right side of the door." + }, + "answer_label": "E", + "answer_text": "The blue bear figurine is on the right side of the door.", + "q_id": "scene0758_00-0_8", + "correct_answer": "The blue bear figurine is on the right side of the door.", + "original_options": { + "A": "On the left side of the door.", + "B": "Behind the door.", + "C": "The blue bear figurine is above the door, positioned near the top edge rather than beside it.", + "D": "The blue bear figurine is on the far side of the doorway, not on the right side next to it.", + "E": "The blue bear figurine is on the right side of the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 3, + 17, + 20 + ] + } +] diff --git a/QAs/scene0759_00-0-mcq.json b/QAs/scene0759_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..fc4ce2bd3ec11793a743f02f14b45e2f72d86586 --- /dev/null +++ b/QAs/scene0759_00-0-mcq.json @@ -0,0 +1,539 @@ +[ + { + "video_uid": "scene0759_00-0", + "video_length": "00:00:20", + "qid": "scene0759_00-0_1_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are in front of me?", + "answer": "There were 4 computers in front of you; they were in front of you.", + "question_chinese": "我面前有几个电脑？", + "answer_chinese": "你面前有4个电脑，它们在你前面。", + "evidence": [ + { + "description": "saw four computers", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There were 4 computers in front of you; they were in front of you.", + "B": "Three computers in front of you.", + "C": "Five computers in front of you.", + "D": "There were 2 computers in front of you; they were in front of you.", + "E": "There were 6 computers in front of you; they were in front of you." + }, + "answer_label": "A", + "answer_text": "There were 4 computers in front of you; they were in front of you.", + "q_id": "scene0759_00-0_1_T1", + "correct_answer": "There were 4 computers in front of you; they were in front of you.", + "original_options": { + "A": "There were 4 computers in front of you; they were in front of you.", + "B": "Three computers in front of you.", + "C": "Five computers in front of you.", + "D": "There were 2 computers in front of you; they were in front of you.", + "E": "There were 6 computers in front of you; they were in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 14 + ] + }, + { + "video_uid": "scene0759_00-0", + "video_length": "00:00:20", + "qid": "scene0759_00-0_1_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are in front of me?", + "answer": "There were 0 computers in front of you; they were to your left.", + "question_chinese": "我面前有几个电脑？", + "answer_chinese": "你面前有0个电脑，它们在你左边。", + "evidence": [ + { + "description": "saw the zero computer。", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One computer in front of you.", + "B": "There were 0 computers in front of you; they were to your left.", + "C": "Two computers in front of you.", + "D": "There were 3 computers in front of you; they were to your left.", + "E": "There were 4 computers in front of you; they were to your left." + }, + "answer_label": "B", + "answer_text": "There were 0 computers in front of you; they were to your left.", + "q_id": "scene0759_00-0_1_T2", + "correct_answer": "There were 0 computers in front of you; they were to your left.", + "original_options": { + "A": "One computer in front of you.", + "B": "There were 0 computers in front of you; they were to your left.", + "C": "Two computers in front of you.", + "D": "There were 3 computers in front of you; they were to your left.", + "E": "There were 4 computers in front of you; they were to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 13 + ] + }, + { + "video_uid": "scene0759_00-0", + "video_length": "00:00:20", + "qid": "scene0759_00-0_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are there in the room?", + "answer": "Four.", + "question_chinese": "房间里有几个电脑", + "answer_chinese": "4个", + "evidence": [ + { + "description": "saw the computers", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Six.", + "C": "Four.", + "D": "There is one computer in the room, set up on a single desk with no other stations.", + "E": "There are six computers in the room, with several arranged across desks and work areas." + }, + "answer_label": "C", + "answer_text": "Four.", + "q_id": "scene0759_00-0_1", + "correct_answer": "Four.", + "original_options": { + "A": "One.", + "B": "Six.", + "C": "Four.", + "D": "There is one computer in the room, set up on a single desk with no other stations.", + "E": "There are six computers in the room, with several arranged across desks and work areas." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 17, + 15 + ] + }, + { + "video_uid": "scene0759_00-0", + "video_length": "00:00:20", + "qid": "scene0759_00-0_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many computers are there in the room?", + "answer": "Four.", + "question_chinese": "房间里有几个电脑", + "answer_chinese": "4个", + "evidence": [ + { + "description": "saw the four computers", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + }, + { + "description": "saw the same computers", + "timestamps": { + "start": "00:00:12", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "There are five computers in the room, which is one more than four in total.", + "D": "Four.", + "E": "There are six computers in the room, clearly exceeding four computers present overall." + }, + "answer_label": "D", + "answer_text": "Four.", + "q_id": "scene0759_00-0_2", + "correct_answer": "Four.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "There are five computers in the room, which is one more than four in total.", + "D": "Four.", + "E": "There are six computers in the room, clearly exceeding four computers present overall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 13 + ] + }, + { + "video_uid": "scene0759_00-0", + "video_length": "00:00:20", + "qid": "scene0759_00-0_3", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the door in based on my position?", + "answer": "The door is on my right.", + "question_chinese": "根据我的位置判断门在哪个方向", + "answer_chinese": "门在我的右边", + "evidence": [ + { + "description": "The door is on this person's right", + "timestamps": { + "start": "00:00:04", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is on my left.", + "B": "The door is on my right.", + "C": "The door is behind me.", + "D": "The door is directly in front of me based on where I'm standing, not off to my right side.", + "E": "The door is to my left when I face forward, so I would turn left to reach it." + }, + "answer_label": "B", + "answer_text": "The door is on my right.", + "q_id": "scene0759_00-0_3", + "correct_answer": "The door is on my right.", + "original_options": { + "A": "The door is on my left.", + "B": "The door is on my right.", + "C": "The door is behind me.", + "D": "The door is directly in front of me based on where I'm standing, not off to my right side.", + "E": "The door is to my left when I face forward, so I would turn left to reach it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 19, + 18 + ] + }, + { + "video_uid": "scene0759_00-0", + "video_length": "00:00:20", + "qid": "scene0759_00-0_4", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the door in based on my position?", + "answer": "The door is behind me.", + "question_chinese": "根据我的位置判断门在哪个方向", + "answer_chinese": "门在我的后边", + "evidence": [ + { + "description": "The door is behind this person", + "timestamps": { + "start": "00:00:04", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The door is in front of me.", + "B": "The door is to my left.", + "C": "The door is to your right from where you’re standing, not behind you.", + "D": "The door is beside you at shoulder level rather than positioned behind your back.", + "E": "The door is behind me." + }, + "answer_label": "E", + "answer_text": "The door is behind me.", + "q_id": "scene0759_00-0_4", + "correct_answer": "The door is behind me.", + "original_options": { + "A": "The door is in front of me.", + "B": "The door is to my left.", + "C": "The door is to your right from where you’re standing, not behind you.", + "D": "The door is beside you at shoulder level rather than positioned behind your back.", + "E": "The door is behind me." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 14 + ] + }, + { + "video_uid": "scene0759_00-0", + "video_length": "00:00:20", + "qid": "scene0759_00-0_4_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many chairs did I see?", + "answer": "At this moment, you saw 1 chair; it was to your front-right.", + "question_chinese": "此时，我看见了几把椅子？", + "answer_chinese": "此时，你看见了1把椅子，它在你右前方。", + "evidence": [ + { + "description": "saw one chairs", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 0 chairs.", + "B": "You saw 2 chairs.", + "C": "At this moment, you saw 3 chairs; they were grouped on your front-right side.", + "D": "At this moment, you saw 4 chairs; all of them were positioned to your front-right.", + "E": "At this moment, you saw 1 chair; it was to your front-right." + }, + "answer_label": "E", + "answer_text": "At this moment, you saw 1 chair; it was to your front-right.", + "q_id": "scene0759_00-0_4_T2", + "correct_answer": "At this moment, you saw 1 chair; it was to your front-right.", + "original_options": { + "A": "You saw 0 chairs.", + "B": "You saw 2 chairs.", + "C": "At this moment, you saw 3 chairs; they were grouped on your front-right side.", + "D": "At this moment, you saw 4 chairs; all of them were positioned to your front-right.", + "E": "At this moment, you saw 1 chair; it was to your front-right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "scene0759_00-0", + "video_length": "00:00:20", + "qid": "scene0759_00-0_4_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, how many chairs did I see?", + "answer": "At this moment, you saw 2 chairs; they were to your right.", + "question_chinese": "此时，我看见了几把椅子？", + "answer_chinese": "此时，你看见了2把椅子，它们在你右边。", + "evidence": [ + { + "description": "saw two chairs", + "timestamps": { + "start": "00:00:08", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw 0 chairs.", + "B": "You saw 1 chair.", + "C": "At this moment, you saw 3 chairs; they were all located to your right.", + "D": "At this moment, you saw 2 chairs; they were to your right.", + "E": "At this moment, you saw 4 chairs; they were lined up on your right side." + }, + "answer_label": "D", + "answer_text": "At this moment, you saw 2 chairs; they were to your right.", + "q_id": "scene0759_00-0_4_T1", + "correct_answer": "At this moment, you saw 2 chairs; they were to your right.", + "original_options": { + "A": "You saw 0 chairs.", + "B": "You saw 1 chair.", + "C": "At this moment, you saw 3 chairs; they were all located to your right.", + "D": "At this moment, you saw 2 chairs; they were to your right.", + "E": "At this moment, you saw 4 chairs; they were lined up on your right side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 15 + ] + }, + { + "video_uid": "scene0759_00-0", + "video_length": "00:00:20", + "qid": "scene0759_00-0_3_T2", + "qid_num": 9, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the door relative to me?", + "answer": "The door was to your right.", + "question_chinese": "门在我的哪个方向？", + "answer_chinese": "门在你的右边。", + "evidence": [ + { + "description": "The door is on your right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The door was in front of you.", + "B": "The door was behind you.", + "C": "The door was to your right.", + "D": "The door was to your left, not to your right where you were facing.", + "E": "The door was slightly in front of you rather than on your right-hand side." + }, + "answer_label": "C", + "answer_text": "The door was to your right.", + "q_id": "scene0759_00-0_3_T2", + "correct_answer": "The door was to your right.", + "original_options": { + "A": "The door was in front of you.", + "B": "The door was behind you.", + "C": "The door was to your right.", + "D": "The door was to your left, not to your right where you were facing.", + "E": "The door was slightly in front of you rather than on your right-hand side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 14, + 14 + ] + }, + { + "video_uid": "scene0759_00-0", + "video_length": "00:00:20", + "qid": "scene0759_00-0_3_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction is the door relative to me?", + "answer": "The door was in front of you.", + "question_chinese": "门在我的哪个方向？", + "answer_chinese": "门在你的前面。", + "evidence": [ + { + "description": "The door is in front of you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The door was behind you.", + "B": "The door was to your left.", + "C": "The door was in front of you.", + "D": "The door was to your right, not directly in front of where you were standing.", + "E": "The door was off to your left side rather than straight ahead in your line of sight." + }, + "answer_label": "C", + "answer_text": "The door was in front of you.", + "q_id": "scene0759_00-0_3_T1", + "correct_answer": "The door was in front of you.", + "original_options": { + "A": "The door was behind you.", + "B": "The door was to your left.", + "C": "The door was in front of you.", + "D": "The door was to your right, not directly in front of where you were standing.", + "E": "The door was off to your left side rather than straight ahead in your line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 15, + 17 + ] + } +] diff --git a/QAs/scene0760_00-0-mcq.json b/QAs/scene0760_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..96cbb93d002c427adf02c37c7ca91167aa0d0d9c --- /dev/null +++ b/QAs/scene0760_00-0-mcq.json @@ -0,0 +1,732 @@ +[ + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_2_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the white cup filled with water on the table?", + "answer": "No, you can’t.", + "question_chinese": "我可以碰到桌上装着水的白色杯子吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The distance between the white cup and you was beyond your arm, so I couldn't touch it", + "timestamps": { + "start": "00:00:06", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No, you can’t.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No, you can’t.", + "q_id": "scene0760_00-0_2_T1", + "correct_answer": "No, you can’t.", + "original_options": { + "A": "No, you can’t.", + "B": "Yes, you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_2_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the white cup filled with water on the table?", + "answer": "No, you can’t.", + "question_chinese": "我可以碰到桌上装着水的白色杯子吗？", + "answer_chinese": "不能", + "evidence": [ + { + "description": "The distance between the white cup and you was beyond your arm, so I couldn't touch it", + "timestamps": { + "start": "00:00:06", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No, you can’t." + }, + "answer_label": "B", + "answer_text": "No, you can’t.", + "q_id": "scene0760_00-0_2_T2", + "correct_answer": "No, you can’t.", + "original_options": { + "A": "Yes, you can.", + "B": "No, you can’t." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_2_T4", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the white cup filled with water on the table?", + "answer": "Yes.", + "question_chinese": "我可以碰到桌上装着水的白色杯子吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "You gradually approach the cup and eventually you can touch it", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "You are quite far from the cup", + "timestamps": { + "start": "00:00:06", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "scene0760_00-0_2_T4", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_2_T3", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the white cup filled with water on the table?", + "answer": "Yes.", + "question_chinese": "我可以碰到桌上装着水的白色杯子吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "You gradually approach the cup and eventually you can touch it", + "timestamps": { + "start": "00:00:16", + "end": "00:00:17" + } + }, + { + "description": "You are quite far from the cup", + "timestamps": { + "start": "00:00:06", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "scene0760_00-0_2_T3", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_1_T1", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see in the room?", + "answer": "You saw two chairs in front of you.", + "question_chinese": "我看见房间里有几张椅子？", + "answer_chinese": "你看见了前面有两张椅子", + "evidence": [ + { + "description": "I see two chairs.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one chair.", + "B": "You saw three chairs.", + "C": "You saw two chairs in front of you.", + "D": "You saw zero chairs in the room, with no chairs positioned in front of you.", + "E": "You saw four chairs in the room, all arranged in front of you as you looked ahead." + }, + "answer_label": "C", + "answer_text": "You saw two chairs in front of you.", + "q_id": "scene0760_00-0_1_T1", + "correct_answer": "You saw two chairs in front of you.", + "original_options": { + "A": "You saw one chair.", + "B": "You saw three chairs.", + "C": "You saw two chairs in front of you.", + "D": "You saw zero chairs in the room, with no chairs positioned in front of you.", + "E": "You saw four chairs in the room, all arranged in front of you as you looked ahead." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 17 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_1_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs did I see in the room?", + "answer": "You saw four chairs in front of you.", + "question_chinese": "我看见房间里有几张椅子？", + "answer_chinese": "你看见了前面有四张椅子", + "evidence": [ + { + "description": "saw ftwo chairs", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "the third one", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "the fourth one", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw three chairs.", + "B": "You saw five chairs.", + "C": "You saw two chairs in front of you, with only a pair visible in the room.", + "D": "You saw four chairs in front of you.", + "E": "You saw six chairs in front of you, with more chairs visible than just a small group." + }, + "answer_label": "D", + "answer_text": "You saw four chairs in front of you.", + "q_id": "scene0760_00-0_1_T2", + "correct_answer": "You saw four chairs in front of you.", + "original_options": { + "A": "You saw three chairs.", + "B": "You saw five chairs.", + "C": "You saw two chairs in front of you, with only a pair visible in the room.", + "D": "You saw four chairs in front of you.", + "E": "You saw six chairs in front of you, with more chairs visible than just a small group." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 17 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "Four.", + "question_chinese": "房间里有几个椅子", + "answer_chinese": "4个", + "evidence": [ + { + "description": "saw four different chairs", + "timestamps": { + "start": "00:00:00", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Three.", + "C": "There are five chairs in the room, with one extra chair beyond the four-chair setup.", + "D": "There are six chairs in the room, making it more crowded than a four-chair arrangement.", + "E": "Four." + }, + "answer_label": "E", + "answer_text": "Four.", + "q_id": "scene0760_00-0_1", + "correct_answer": "Four.", + "original_options": { + "A": "Two.", + "B": "Three.", + "C": "There are five chairs in the room, with one extra chair beyond the four-chair setup.", + "D": "There are six chairs in the room, making it more crowded than a four-chair arrangement.", + "E": "Four." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 15 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "Four.", + "question_chinese": "房间里有几个椅子", + "answer_chinese": "4个", + "evidence": [ + { + "description": "saw four chairs", + "timestamps": { + "start": "00:00:00", + "end": "00:00:11" + } + }, + { + "description": "saw the same chairs", + "timestamps": { + "start": "00:00:15", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "There are five chairs in the room, exceeding the four-chair total by one additional chair.", + "E": "There are six chairs in the room, which is more than the four chairs stated in the answer." + }, + "answer_label": "B", + "answer_text": "Four.", + "q_id": "scene0760_00-0_2", + "correct_answer": "Four.", + "original_options": { + "A": "Two.", + "B": "Four.", + "C": "Three.", + "D": "There are five chairs in the room, exceeding the four-chair total by one additional chair.", + "E": "There are six chairs in the room, which is more than the four chairs stated in the answer." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 15, + 18 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_3_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the table or the chair?", + "answer": "The chair is closer to you.", + "question_chinese": "桌子和椅子哪个离我更近？", + "answer_chinese": "椅子离你更近", + "evidence": [ + { + "description": "There are a chair and a table in front of you.The chair is a little closer to you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The table is closer to you.", + "B": "The chair is closer to you." + }, + "answer_label": "B", + "answer_text": "The chair is closer to you.", + "q_id": "scene0760_00-0_3_T1", + "correct_answer": "The chair is closer to you.", + "original_options": { + "A": "The table is closer to you.", + "B": "The chair is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_3_T2", + "qid_num": 10, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the table or the chair?", + "answer": "The table is closer to you.", + "question_chinese": "桌子和椅子哪个离我更近？", + "answer_chinese": "桌子离你更近", + "evidence": [ + { + "description": "There are a chair and a table in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You walk forward and get close to the table.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The chair is closer to you.", + "B": "The table is closer to you." + }, + "answer_label": "B", + "answer_text": "The table is closer to you.", + "q_id": "scene0760_00-0_3_T2", + "correct_answer": "The table is closer to you.", + "original_options": { + "A": "The chair is closer to you.", + "B": "The table is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_4_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the computer relative to me?", + "answer": "The computer is in front of you.", + "question_chinese": "电脑在我的哪个方位？", + "answer_chinese": "电脑在你的前面", + "evidence": [ + { + "description": "The computer is on the table in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The computer is in front of you.", + "D": "The computer is behind you, positioned out of your direct line of sight.", + "E": "The computer is to your right, off to the side rather than straight ahead." + }, + "answer_label": "C", + "answer_text": "The computer is in front of you.", + "q_id": "scene0760_00-0_4_T1", + "correct_answer": "The computer is in front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "The computer is in front of you.", + "D": "The computer is behind you, positioned out of your direct line of sight.", + "E": "The computer is to your right, off to the side rather than straight ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 13, + 14 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_4_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Where is the computer relative to me?", + "answer": "The computer is on your right-hand side.", + "question_chinese": "电脑在我的哪个方位？", + "answer_chinese": "电脑在你的右手边", + "evidence": [ + { + "description": "The computer is on the table in front of you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer is on your right-hand side.", + "B": "The computer is on your left.", + "C": "The computer is in front of you.", + "D": "The computer is directly behind you, opposite where you are facing right now.", + "E": "The computer is slightly to your left and a bit forward from your current position." + }, + "answer_label": "A", + "answer_text": "The computer is on your right-hand side.", + "q_id": "scene0760_00-0_4_T2", + "correct_answer": "The computer is on your right-hand side.", + "original_options": { + "A": "The computer is on your right-hand side.", + "B": "The computer is on your left.", + "C": "The computer is in front of you.", + "D": "The computer is directly behind you, opposite where you are facing right now.", + "E": "The computer is slightly to your left and a bit forward from your current position." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 13, + 15 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_5", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the condition of the long cabinet door?", + "answer": "The doors of the long cabinets are all closed.", + "question_chinese": "长柜子的门是什么状态", + "answer_chinese": "长柜子的门都是关着的", + "evidence": [ + { + "description": "The doors of the long cabinets are all closed", + "timestamps": { + "start": "00:00:09", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The doors of the long cabinets are all closed.", + "B": "Some doors are slightly open.", + "C": "A few doors are fully open.", + "D": "At least one of the long cabinet doors is ajar, leaving a visible gap along the edge.", + "E": "The long cabinet doors are not shut; several are open enough that you could see inside." + }, + "answer_label": "A", + "answer_text": "The doors of the long cabinets are all closed.", + "q_id": "scene0760_00-0_5", + "correct_answer": "The doors of the long cabinets are all closed.", + "original_options": { + "A": "The doors of the long cabinets are all closed.", + "B": "Some doors are slightly open.", + "C": "A few doors are fully open.", + "D": "At least one of the long cabinet doors is ajar, leaving a visible gap along the edge.", + "E": "The long cabinet doors are not shut; several are open enough that you could see inside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 16 + ] + }, + { + "video_uid": "scene0760_00-0", + "video_length": "00:00:20", + "qid": "scene0760_00-0_6", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the condition of the long cabinet door?", + "answer": "The doors of the long cabinets are all closed.", + "question_chinese": "长柜子的门是什么状态", + "answer_chinese": "长柜子的门都是关着的", + "evidence": [ + { + "description": "The doors of the long cabinets are all closed", + "timestamps": { + "start": "00:00:09", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They are partly open.", + "B": "They are fully open.", + "C": "The doors of the long cabinets are all closed.", + "D": "The long cabinet doors are slightly ajar, not completely closed along the top edge.", + "E": "The long cabinet doors are open wide, so the interior space would be visible from where you stand." + }, + "answer_label": "C", + "answer_text": "The doors of the long cabinets are all closed.", + "q_id": "scene0760_00-0_6", + "correct_answer": "The doors of the long cabinets are all closed.", + "original_options": { + "A": "They are partly open.", + "B": "They are fully open.", + "C": "The doors of the long cabinets are all closed.", + "D": "The long cabinet doors are slightly ajar, not completely closed along the top edge.", + "E": "The long cabinet doors are open wide, so the interior space would be visible from where you stand." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 18 + ] + } +] diff --git a/QAs/scene0761_00-0-mcq.json b/QAs/scene0761_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..96c54b0cb82b5e8b249a1008d71940aa2627b227 --- /dev/null +++ b/QAs/scene0761_00-0-mcq.json @@ -0,0 +1,846 @@ +[ + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_2_T1", + "qid_num": 1, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I just woke up, could I touch the table next to the bed?", + "answer": "Yes.", + "question_chinese": "如果我刚起床，可以碰到床旁边的桌子吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "The bed is very close to the table. After getting up, you can touch the table", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "The bed is very close to the table. After getting up, you can touch the table", + "timestamps": { + "start": "00:00:04", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "A", + "answer_text": "Yes.", + "q_id": "scene0761_00-0_2_T1", + "correct_answer": "Yes.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_2_T2", + "qid_num": 2, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "If I just woke up, could I touch the table next to the bed?", + "answer": "Yes, you can. The table is 1 meter to the right of the bed.", + "question_chinese": "如果我刚起床，可以碰到床旁边的桌子吗？", + "answer_chinese": "你可以，桌子在床的右边一米的位置。", + "evidence": [ + { + "description": "The bed is very close to the table. After getting up, you can touch the table", + "timestamps": { + "start": "00:00:04", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes, you can. The table is 1 meter to the right of the bed.", + "B": "No, you can't. The table is 1 meter to the left of the bed." + }, + "answer_label": "A", + "answer_text": "Yes, you can. The table is 1 meter to the right of the bed.", + "q_id": "scene0761_00-0_2_T2", + "correct_answer": "Yes, you can. The table is 1 meter to the right of the bed.", + "original_options": { + "A": "Yes, you can. The table is 1 meter to the right of the bed.", + "B": "No, you can't. The table is 1 meter to the left of the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_10", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to close the curtains, how should I go?", + "answer": "You should first walk forward to the front of the bed, and then turn right to walk to the curtain.", + "question_chinese": "如果我想关闭窗帘，我应该怎么走？", + "answer_chinese": "你应该先向前走到床前面，然后右转走到窗帘", + "evidence": [ + { + "description": "The curtain is on your right front", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Walk forward, then turn left.", + "B": "Turn around and walk straight.", + "C": "You should walk forward to the back of the bed, then turn right to reach the curtain.", + "D": "You should first step to your left side of the bed, then continue straight until you reach the curtain.", + "E": "You should first walk forward to the front of the bed, and then turn right to walk to the curtain." + }, + "answer_label": "E", + "answer_text": "You should first walk forward to the front of the bed, and then turn right to walk to the curtain.", + "q_id": "scene0761_00-0_10", + "correct_answer": "You should first walk forward to the front of the bed, and then turn right to walk to the curtain.", + "original_options": { + "A": "Walk forward, then turn left.", + "B": "Turn around and walk straight.", + "C": "You should walk forward to the back of the bed, then turn right to reach the curtain.", + "D": "You should first step to your left side of the bed, then continue straight until you reach the curtain.", + "E": "You should first walk forward to the front of the bed, and then turn right to walk to the curtain." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 19 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_6_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to close the curtains, how should I walk?", + "answer": "You should first walk forward to the front of the bed, then turn right and walk to the curtain.", + "question_chinese": "如果我想关闭窗帘，我应该怎么走？", + "answer_chinese": "你应该先向前走到床前面，然后右转走到窗帘", + "evidence": [ + { + "description": "The curtain is on your right front", + "timestamps": { + "start": "00:00:00", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Walk forward, then turn left.", + "B": "Go to the back of the bed.", + "C": "You should turn around first, then walk straight ahead because the curtain is behind you.", + "D": "You should walk forward to the front of the bed, then turn left and continue walking to the curtain.", + "E": "You should first walk forward to the front of the bed, then turn right and walk to the curtain." + }, + "answer_label": "E", + "answer_text": "You should first walk forward to the front of the bed, then turn right and walk to the curtain.", + "q_id": "scene0761_00-0_6_T1", + "correct_answer": "You should first walk forward to the front of the bed, then turn right and walk to the curtain.", + "original_options": { + "A": "Walk forward, then turn left.", + "B": "Go to the back of the bed.", + "C": "You should turn around first, then walk straight ahead because the curtain is behind you.", + "D": "You should walk forward to the front of the bed, then turn left and continue walking to the curtain.", + "E": "You should first walk forward to the front of the bed, then turn right and walk to the curtain." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 7, + 15, + 19 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_6_T2", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to close the curtains, how should I walk?", + "answer": "You should turn around, then walk straight. The curtain is behind you.", + "question_chinese": "如果我想关闭窗帘，我应该怎么走？", + "answer_chinese": "你应该向后转，然后直走。窗帘在你后面。", + "evidence": [ + { + "description": "The curtain is behind you.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The curtain is in front of you.", + "B": "The curtain is to your left.", + "C": "You should turn around, then walk straight. The curtain is behind you.", + "D": "You should turn around, then walk straight, because the curtain is positioned off to your right side.", + "E": "You should turn around and walk straight, but the curtain is not behind you; it is directly ahead." + }, + "answer_label": "C", + "answer_text": "You should turn around, then walk straight. The curtain is behind you.", + "q_id": "scene0761_00-0_6_T2", + "correct_answer": "You should turn around, then walk straight. The curtain is behind you.", + "original_options": { + "A": "The curtain is in front of you.", + "B": "The curtain is to your left.", + "C": "You should turn around, then walk straight. The curtain is behind you.", + "D": "You should turn around, then walk straight, because the curtain is positioned off to your right side.", + "E": "You should turn around and walk straight, but the curtain is not behind you; it is directly ahead." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 18 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_5_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to look in the mirror, how should I walk?", + "answer": "You should go straight; the mirror is in front of you.", + "question_chinese": "如果我想照镜子，我应该怎么走？", + "answer_chinese": "你应该直走，镜子在你前面。", + "evidence": [ + { + "description": "The mirror is in front of you on the right", + "timestamps": { + "start": "00:00:00", + "end": "00:00:7" + } + } + ], + "question_timestamps": "00:00:9", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left toward the mirror.", + "B": "Turn right toward the mirror.", + "C": "You should go straight; the mirror is in front of you.", + "D": "Walk straight for a bit, then turn around because the mirror is behind you.", + "E": "Take a few steps forward, then veer left since the mirror sits off to that side." + }, + "answer_label": "C", + "answer_text": "You should go straight; the mirror is in front of you.", + "q_id": "scene0761_00-0_5_T2", + "correct_answer": "You should go straight; the mirror is in front of you.", + "original_options": { + "A": "Turn left toward the mirror.", + "B": "Turn right toward the mirror.", + "C": "You should go straight; the mirror is in front of you.", + "D": "Walk straight for a bit, then turn around because the mirror is behind you.", + "E": "Take a few steps forward, then veer left since the mirror sits off to that side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 16 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_9", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to look in the mirror, how should I walk?", + "answer": "You should first walk straight to the black box and then turn right to the mirror.", + "question_chinese": "如果我想照镜子，我应该怎么走？", + "answer_chinese": "你应该先直走到黑色框子前，然后右转走到镜子前", + "evidence": [ + { + "description": "The mirror is in front of you on the right", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "You should first walk straight to the black box and then turn right to the mirror.", + "B": "Turn left after the black box.", + "C": "Go around the black box first.", + "D": "Walk straight to the black box, then turn left to reach the mirror instead of turning right.", + "E": "Head straight toward the black box, pass it slightly, then turn around to face the mirror." + }, + "answer_label": "A", + "answer_text": "You should first walk straight to the black box and then turn right to the mirror.", + "q_id": "scene0761_00-0_9", + "correct_answer": "You should first walk straight to the black box and then turn right to the mirror.", + "original_options": { + "A": "You should first walk straight to the black box and then turn right to the mirror.", + "B": "Turn left after the black box.", + "C": "Go around the black box first.", + "D": "Walk straight to the black box, then turn left to reach the mirror instead of turning right.", + "E": "Head straight toward the black box, pass it slightly, then turn around to face the mirror." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 17, + 16 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_5_T1", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "If I want to look in the mirror, how should I walk?", + "answer": "You should first walk straight to the black frame, then turn right and walk to the mirror.", + "question_chinese": "如果我想照镜子，我应该怎么走？", + "answer_chinese": "你应该先直走到黑色框子前，然后右转走到镜子前", + "evidence": [ + { + "description": "The mirror is in front of you on the right", + "timestamps": { + "start": "00:00:00", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left at the black frame.", + "B": "Go straight to the mirror.", + "C": "You should first walk straight to the black frame, then turn right and walk to the mirror.", + "D": "Walk straight to the black frame, then turn left and continue walking until you reach the mirror.", + "E": "Go to the black box first, then turn right and walk forward to get to the mirror." + }, + "answer_label": "C", + "answer_text": "You should first walk straight to the black frame, then turn right and walk to the mirror.", + "q_id": "scene0761_00-0_5_T1", + "correct_answer": "You should first walk straight to the black frame, then turn right and walk to the mirror.", + "original_options": { + "A": "Turn left at the black frame.", + "B": "Go straight to the mirror.", + "C": "You should first walk straight to the black frame, then turn right and walk to the mirror.", + "D": "Walk straight to the black frame, then turn left and continue walking until you reach the mirror.", + "E": "Go to the black box first, then turn right and walk forward to get to the mirror." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 17 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_4_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different kinds of items did I see placed on the table next to the bed?", + "answer": "You saw three kinds of items on the table next to the bed: clothes, plastic bags, and a charging cable.", + "question_chinese": "我看见床旁边的桌子上摆了几种物品？", + "answer_chinese": "你看见床旁边的桌子摆了衣服和塑料袋,还有充电线3种物品", + "evidence": [ + { + "description": "Clothes, plastic bags and charging cables are on the table beside the bed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "One kind of item.", + "B": "Two kinds of items.", + "C": "You saw three kinds of items on the table next to the bed: clothes, plastic bags, and a charging cable.", + "D": "You saw two kinds of items on the table: clothes and plastic bags, with no charging cable present.", + "E": "You saw four kinds of items on the table: clothes, plastic bags, a charging cable, and another small item." + }, + "answer_label": "C", + "answer_text": "You saw three kinds of items on the table next to the bed: clothes, plastic bags, and a charging cable.", + "q_id": "scene0761_00-0_4_T2", + "correct_answer": "You saw three kinds of items on the table next to the bed: clothes, plastic bags, and a charging cable.", + "original_options": { + "A": "One kind of item.", + "B": "Two kinds of items.", + "C": "You saw three kinds of items on the table next to the bed: clothes, plastic bags, and a charging cable.", + "D": "You saw two kinds of items on the table: clothes and plastic bags, with no charging cable present.", + "E": "You saw four kinds of items on the table: clothes, plastic bags, a charging cable, and another small item." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_4_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "How many different kinds of items did I see placed on the table next to the bed?", + "answer": "The table next to the bed has two kinds of items on it: clothes and plastic bags.", + "question_chinese": "我看见床旁边的桌子上摆了几种物品？", + "answer_chinese": "床旁边的桌子摆了衣服和塑料袋两种物品", + "evidence": [ + { + "description": "Two items, clothes and plastic bags, were placed on the table beside the bed", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "saw the same items", + "timestamps": { + "start": "00:00:05", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Zero kinds of items.", + "B": "The table next to the bed has two kinds of items on it: clothes and plastic bags.", + "C": "One kind of item.", + "D": "The table next to the bed has three kinds of items on it: clothes, plastic bags, and a charging cable.", + "E": "The table next to the bed has four kinds: clothes, plastic bags, a charging cable, and another item." + }, + "answer_label": "B", + "answer_text": "The table next to the bed has two kinds of items on it: clothes and plastic bags.", + "q_id": "scene0761_00-0_4_T1", + "correct_answer": "The table next to the bed has two kinds of items on it: clothes and plastic bags.", + "original_options": { + "A": "Zero kinds of items.", + "B": "The table next to the bed has two kinds of items on it: clothes and plastic bags.", + "C": "One kind of item.", + "D": "The table next to the bed has three kinds of items on it: clothes, plastic bags, and a charging cable.", + "E": "The table next to the bed has four kinds: clothes, plastic bags, a charging cable, and another item." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 20, + 18 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_1_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I saw it, what state were the curtains in?", + "answer": "When you saw them, the curtains were half open, and the curtains were to your right.", + "question_chinese": "我看见时窗帘是什么状态？", + "answer_chinese": "你看见时窗帘是半拉开的，窗帘在你右边。", + "evidence": [ + { + "description": "The curtain is half drawn.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "When you saw them, the curtains were half open, and the curtains were to your right.", + "B": "The curtains were closed.", + "C": "The curtains were fully open.", + "D": "When you saw them, the curtains were half open, and the curtains were in front of you.", + "E": "When you saw them, the curtains were half open, and the curtains were to your left." + }, + "answer_label": "A", + "answer_text": "When you saw them, the curtains were half open, and the curtains were to your right.", + "q_id": "scene0761_00-0_1_T1", + "correct_answer": "When you saw them, the curtains were half open, and the curtains were to your right.", + "original_options": { + "A": "When you saw them, the curtains were half open, and the curtains were to your right.", + "B": "The curtains were closed.", + "C": "The curtains were fully open.", + "D": "When you saw them, the curtains were half open, and the curtains were in front of you.", + "E": "When you saw them, the curtains were half open, and the curtains were to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 16 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_1_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I saw it, what state were the curtains in?", + "answer": "The curtains are open.", + "question_chinese": "我看见时窗帘是什么状态？", + "answer_chinese": "窗帘是拉开的", + "evidence": [ + { + "description": "The curtain is drawn open", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The curtains are closed.", + "B": "The curtains are open.", + "C": "The curtains are half open.", + "D": "The curtains are mostly closed, leaving only a small gap for light to come through.", + "E": "The curtains are pulled shut rather than open, blocking the view outside." + }, + "answer_label": "B", + "answer_text": "The curtains are open.", + "q_id": "scene0761_00-0_1_T2", + "correct_answer": "The curtains are open.", + "original_options": { + "A": "The curtains are closed.", + "B": "The curtains are open.", + "C": "The curtains are half open.", + "D": "The curtains are mostly closed, leaving only a small gap for light to come through.", + "E": "The curtains are pulled shut rather than open, blocking the view outside." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 12 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_2", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the state of the curtain?", + "answer": "The curtain is drawn open.", + "question_chinese": "窗帘是什么状态", + "answer_chinese": "窗帘是拉开的", + "evidence": [ + { + "description": "The curtain is drawn open", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The curtain is closed.", + "B": "The curtain is half open.", + "C": "The curtain is pulled shut, covering the window rather than being drawn open.", + "D": "The curtain is drawn open.", + "E": "The curtain is only slightly open, leaving most of the window still covered." + }, + "answer_label": "D", + "answer_text": "The curtain is drawn open.", + "q_id": "scene0761_00-0_2", + "correct_answer": "The curtain is drawn open.", + "original_options": { + "A": "The curtain is closed.", + "B": "The curtain is half open.", + "C": "The curtain is pulled shut, covering the window rather than being drawn open.", + "D": "The curtain is drawn open.", + "E": "The curtain is only slightly open, leaving most of the window still covered." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 13, + 13 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What is the state of the curtain?", + "answer": "The curtain is drawn open.", + "question_chinese": "窗帘是什么状态", + "answer_chinese": "窗帘是拉开的", + "evidence": [ + { + "description": "The curtain is drawn open", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The curtain is drawn open.", + "B": "The curtain is drawn closed.", + "C": "The curtain is half-closed.", + "D": "The curtain is pulled shut, blocking most of the window instead of being opened wide.", + "E": "The curtain is only slightly open, leaving the window mostly covered rather than fully uncovered." + }, + "answer_label": "A", + "answer_text": "The curtain is drawn open.", + "q_id": "scene0761_00-0_1", + "correct_answer": "The curtain is drawn open.", + "original_options": { + "A": "The curtain is drawn open.", + "B": "The curtain is drawn closed.", + "C": "The curtain is half-closed.", + "D": "The curtain is pulled shut, blocking most of the window instead of being opened wide.", + "E": "The curtain is only slightly open, leaving the window mostly covered rather than fully uncovered." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 15 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_3_T2", + "qid_num": 15, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the window relative to me?", + "answer": "The window is to your right.", + "question_chinese": "窗户在我的哪个方向？", + "answer_chinese": "窗户在你的右边。", + "evidence": [ + { + "description": "The window is on my right", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is behind you, so you would need to turn around to face it.", + "D": "The window is to your right.", + "E": "The window is above you rather than to the side, requiring you to look upward to see it." + }, + "answer_label": "D", + "answer_text": "The window is to your right.", + "q_id": "scene0761_00-0_3_T2", + "correct_answer": "The window is to your right.", + "original_options": { + "A": "The window is in front of you.", + "B": "The window is to your left.", + "C": "The window is behind you, so you would need to turn around to face it.", + "D": "The window is to your right.", + "E": "The window is above you rather than to the side, requiring you to look upward to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 15, + 18 + ] + }, + { + "video_uid": "scene0761_00-0", + "video_length": "00:00:20", + "qid": "scene0761_00-0_3_T1", + "qid_num": 16, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction is the window relative to me?", + "answer": "The window is behind you.", + "question_chinese": "窗户在我的哪个方向？", + "answer_chinese": "窗户在你的后面", + "evidence": [ + { + "description": "The window is behind me", + "timestamps": { + "start": "00:00:11", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The window is behind you.", + "B": "The window is in front of you.", + "C": "The window is to your right.", + "D": "The window is to your left, positioned beside you instead of behind your viewpoint.", + "E": "The window is above you, so it appears overhead rather than behind you." + }, + "answer_label": "A", + "answer_text": "The window is behind you.", + "q_id": "scene0761_00-0_3_T1", + "correct_answer": "The window is behind you.", + "original_options": { + "A": "The window is behind you.", + "B": "The window is in front of you.", + "C": "The window is to your right.", + "D": "The window is to your left, positioned beside you instead of behind your viewpoint.", + "E": "The window is above you, so it appears overhead rather than behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 14, + 13 + ] + } +] diff --git a/QAs/scene0762_00-0-mcq.json b/QAs/scene0762_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..6a87c5ef73119e106411821872bcd931d310226e --- /dev/null +++ b/QAs/scene0762_00-0-mcq.json @@ -0,0 +1,549 @@ +[ + { + "video_uid": "scene0762_00-0", + "video_length": "00:00:20", + "qid": "scene0762_00-0_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first started shooting videos, in which direction was the computer on me?", + "answer": "The computer is on my right.", + "question_chinese": "我刚开始拍摄视频时，电脑在我的哪个方向？", + "answer_chinese": "电脑在我的右边", + "evidence": [ + { + "description": "When the camera turned right, the computer appeared", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The computer is on my left.", + "B": "The computer is on my right.", + "C": "The computer is in front of me.", + "D": "The computer is behind me, positioned so you would turn around to face it.", + "E": "The computer is below my desk, out of sight unless you look down under the tabletop." + }, + "answer_label": "B", + "answer_text": "The computer is on my right.", + "q_id": "scene0762_00-0_5", + "correct_answer": "The computer is on my right.", + "original_options": { + "A": "The computer is on my left.", + "B": "The computer is on my right.", + "C": "The computer is in front of me.", + "D": "The computer is behind me, positioned so you would turn around to face it.", + "E": "The computer is below my desk, out of sight unless you look down under the tabletop." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 14, + 16 + ] + }, + { + "video_uid": "scene0762_00-0", + "video_length": "00:00:20", + "qid": "scene0762_00-0_3_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first started recording the video, which direction was the computer relative to me?", + "answer": "The computer is directly in front of you.", + "question_chinese": "我刚开始拍摄视频时，电脑在我的哪个方向？", + "answer_chinese": "电脑在你的正前面", + "evidence": [ + { + "description": "The computer is right in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer is directly behind you.", + "B": "The computer is directly to your left.", + "C": "The computer is directly to your right, placed beside you instead of straight ahead in your view.", + "D": "The computer is directly in front of you.", + "E": "The computer is directly above you rather than in front, so you would have to look up to see it." + }, + "answer_label": "D", + "answer_text": "The computer is directly in front of you.", + "q_id": "scene0762_00-0_3_T1", + "correct_answer": "The computer is directly in front of you.", + "original_options": { + "A": "The computer is directly behind you.", + "B": "The computer is directly to your left.", + "C": "The computer is directly to your right, placed beside you instead of straight ahead in your view.", + "D": "The computer is directly in front of you.", + "E": "The computer is directly above you rather than in front, so you would have to look up to see it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 17, + 20 + ] + }, + { + "video_uid": "scene0762_00-0", + "video_length": "00:00:20", + "qid": "scene0762_00-0_3_T2", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "When I first started recording the video, which direction was the computer relative to me?", + "answer": "The computer is to your right.", + "question_chinese": "我刚开始拍摄视频时，电脑在我的哪个方向？", + "answer_chinese": "电脑在我的右边", + "evidence": [ + { + "description": "When the camera turned right, the computer appeared", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The computer is to your left.", + "B": "The computer is in front of you.", + "C": "The computer is behind you, so it would be out of view unless you turned around.", + "D": "The computer is to your right.", + "E": "The computer is above you rather than to your right, making it overhead instead of beside you." + }, + "answer_label": "D", + "answer_text": "The computer is to your right.", + "q_id": "scene0762_00-0_3_T2", + "correct_answer": "The computer is to your right.", + "original_options": { + "A": "The computer is to your left.", + "B": "The computer is in front of you.", + "C": "The computer is behind you, so it would be out of view unless you turned around.", + "D": "The computer is to your right.", + "E": "The computer is above you rather than to your right, making it overhead instead of beside you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "scene0762_00-0", + "video_length": "00:00:20", + "qid": "scene0762_00-0_9", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the white mouse in front of the computer now?", + "answer": "No.", + "question_chinese": "我现在可以触碰到电脑前的白色鼠标吗？", + "answer_chinese": "不可以", + "evidence": [ + { + "description": "The white mouse is quite far from you and out of reach", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes.", + "B": "No." + }, + "answer_label": "B", + "answer_text": "No.", + "q_id": "scene0762_00-0_9", + "correct_answer": "No.", + "original_options": { + "A": "Yes.", + "B": "No." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0762_00-0", + "video_length": "00:00:20", + "qid": "scene0762_00-0_10", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I touch the white mouse in front of the computer now?", + "answer": "Yes.", + "question_chinese": "我现在可以触碰到电脑前的白色鼠标吗？", + "answer_chinese": "可以", + "evidence": [ + { + "description": "You are now very close to the table, and the mouse is placed on the outside of the table and can be touched", + "timestamps": { + "start": "00:00:00", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "B", + "answer_text": "Yes.", + "q_id": "scene0762_00-0_10", + "correct_answer": "Yes.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0762_00-0", + "video_length": "00:00:20", + "qid": "scene0762_00-0_1_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are in my room?", + "answer": "In the room you are in right now, you didn’t see any chairs.", + "question_chinese": "我的房间里有几个椅子？", + "answer_chinese": "你现在所处的房间里没有看见椅子。", + "evidence": [ + { + "description": "saw the black chair", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "In the room you are in right now, you didn’t see any chairs.", + "B": "There is 1 chair.", + "C": "There are 2 chairs.", + "D": "In the room you are in right now, there are 3 chairs grouped together in front of you.", + "E": "In the room you are in right now, there are 4 chairs, with one slightly off to your right." + }, + "answer_label": "A", + "answer_text": "In the room you are in right now, you didn’t see any chairs.", + "q_id": "scene0762_00-0_1_T2", + "correct_answer": "In the room you are in right now, you didn’t see any chairs.", + "original_options": { + "A": "In the room you are in right now, you didn’t see any chairs.", + "B": "There is 1 chair.", + "C": "There are 2 chairs.", + "D": "In the room you are in right now, there are 3 chairs grouped together in front of you.", + "E": "In the room you are in right now, there are 4 chairs, with one slightly off to your right." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 18, + 19 + ] + }, + { + "video_uid": "scene0762_00-0", + "video_length": "00:00:20", + "qid": "scene0762_00-0_1_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are in my room?", + "answer": "In the room you are in right now, there is 1 chair, directly in front of you.", + "question_chinese": "我的房间里有几个椅子？", + "answer_chinese": "你现在所处的房间里有1个椅子，在你的正前方。", + "evidence": [ + { + "description": "saw the black chair", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You didn’t see any chairs.", + "B": "There are 2 chairs.", + "C": "In the room you are in right now, there are 3 chairs directly in front of you.", + "D": "In the room you are in right now, there are 4 chairs, lined up in front of you.", + "E": "In the room you are in right now, there is 1 chair, directly in front of you." + }, + "answer_label": "E", + "answer_text": "In the room you are in right now, there is 1 chair, directly in front of you.", + "q_id": "scene0762_00-0_1_T1", + "correct_answer": "In the room you are in right now, there is 1 chair, directly in front of you.", + "original_options": { + "A": "You didn’t see any chairs.", + "B": "There are 2 chairs.", + "C": "In the room you are in right now, there are 3 chairs directly in front of you.", + "D": "In the room you are in right now, there are 4 chairs, lined up in front of you.", + "E": "In the room you are in right now, there is 1 chair, directly in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 17, + 18 + ] + }, + { + "video_uid": "scene0762_00-0", + "video_length": "00:00:20", + "qid": "scene0762_00-0_2_T2", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What state were the ceiling lights in when I saw them?", + "answer": "The status of the ceiling light above your head is unknown.", + "question_chinese": "我看见天花板上的灯是什么状态？", + "answer_chinese": "你头顶天花板上的灯状态未知。", + "evidence": [ + { + "description": "You never looked up to see the ceiling.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The ceiling light above your head is on.", + "B": "The status of the ceiling light above your head is unknown." + }, + "answer_label": "B", + "answer_text": "The status of the ceiling light above your head is unknown.", + "q_id": "scene0762_00-0_2_T2", + "correct_answer": "The status of the ceiling light above your head is unknown.", + "original_options": { + "A": "The ceiling light above your head is on.", + "B": "The status of the ceiling light above your head is unknown." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0762_00-0", + "video_length": "00:00:20", + "qid": "scene0762_00-0_2_T1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What state were the ceiling lights in when I saw them?", + "answer": "The ceiling light in front of you is off.", + "question_chinese": "我看见天花板上的灯是什么状态？", + "answer_chinese": "你前侧天花板上的灯是关着的", + "evidence": [ + { + "description": "The light on the ceiling wasn't on, so it was turned off", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The ceiling light in front of you is off.", + "B": "The ceiling light in front of you is on." + }, + "answer_label": "A", + "answer_text": "The ceiling light in front of you is off.", + "q_id": "scene0762_00-0_2_T1", + "correct_answer": "The ceiling light in front of you is off.", + "original_options": { + "A": "The ceiling light in front of you is off.", + "B": "The ceiling light in front of you is on." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0762_00-0", + "video_length": "00:00:20", + "qid": "scene0762_00-0_4_T2", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the printer relative to me?", + "answer": "The printer is directly in front of you.", + "question_chinese": "打印机在我的哪个方向？", + "answer_chinese": "打印机在你的正前方。", + "evidence": [ + { + "description": "The printer is right in front of you.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly behind you.", + "B": "The printer is directly in front of you.", + "C": "Directly to your left.", + "D": "The printer is directly to your right, not in front of you at all.", + "E": "The printer is above you, positioned overhead rather than at your eye level." + }, + "answer_label": "B", + "answer_text": "The printer is directly in front of you.", + "q_id": "scene0762_00-0_4_T2", + "correct_answer": "The printer is directly in front of you.", + "original_options": { + "A": "Directly behind you.", + "B": "The printer is directly in front of you.", + "C": "Directly to your left.", + "D": "The printer is directly to your right, not in front of you at all.", + "E": "The printer is above you, positioned overhead rather than at your eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 14, + 13 + ] + }, + { + "video_uid": "scene0762_00-0", + "video_length": "00:00:20", + "qid": "scene0762_00-0_4_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Which direction is the printer relative to me?", + "answer": "The printer is to your front-left.", + "question_chinese": "打印机在我的哪个方向？", + "answer_chinese": "打印机在你的左前方", + "evidence": [ + { + "description": "The printer is on your left front side.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "The printer is directly in front of you, not angled toward your left side.", + "D": "The printer is behind you on the left, so you would need to turn around to face it.", + "E": "The printer is to your front-left." + }, + "answer_label": "E", + "answer_text": "The printer is to your front-left.", + "q_id": "scene0762_00-0_4_T1", + "correct_answer": "The printer is to your front-left.", + "original_options": { + "A": "To your front-right.", + "B": "To your back-left.", + "C": "The printer is directly in front of you, not angled toward your left side.", + "D": "The printer is behind you on the left, so you would need to turn around to face it.", + "E": "The printer is to your front-left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 14, + 18 + ] + } +] diff --git a/QAs/scene0763_00-0-mcq.json b/QAs/scene0763_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..5a0d15dbf09127b83c5162ca981751bbfeb9faa9 --- /dev/null +++ b/QAs/scene0763_00-0-mcq.json @@ -0,0 +1,616 @@ +[ + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_4", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many transparent boxes containing things are placed on the ground?", + "answer": "Three.", + "question_chinese": "地上摆了几个装着东西的透明盒子？", + "answer_chinese": "3个", + "evidence": [ + { + "description": "you saw three transparent boxes on the ground filled with things", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "Five transparent boxes containing items are placed on the ground in front of you.", + "E": "Six transparent boxes containing things are on the ground, spread out across the floor area." + }, + "answer_label": "A", + "answer_text": "Three.", + "q_id": "scene0763_00-0_4", + "correct_answer": "Three.", + "original_options": { + "A": "Three.", + "B": "Two.", + "C": "Four.", + "D": "Five transparent boxes containing items are placed on the ground in front of you.", + "E": "Six transparent boxes containing things are on the ground, spread out across the floor area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 14, + 15 + ] + }, + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_3", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many transparent boxes containing things are placed on the ground?", + "answer": "Three.", + "question_chinese": "地上摆了几个装着东西的透明盒子？", + "answer_chinese": "3个", + "evidence": [ + { + "description": "you saw three transparent boxes on the ground, just like before, filled with things", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + }, + { + "description": "you saw three transparent boxes on the ground filled with things", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "One.", + "B": "Five.", + "C": "Six transparent boxes containing things are placed on the ground near your feet.", + "D": "Three.", + "E": "Eight transparent boxes containing items are on the ground, with several stacked close together." + }, + "answer_label": "D", + "answer_text": "Three.", + "q_id": "scene0763_00-0_3", + "correct_answer": "Three.", + "original_options": { + "A": "One.", + "B": "Five.", + "C": "Six transparent boxes containing things are placed on the ground near your feet.", + "D": "Three.", + "E": "Eight transparent boxes containing items are on the ground, with several stacked close together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1, + 1, + 13, + 14 + ] + }, + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_2_T1", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the transparent box on the floor that has items inside, relative to me?", + "answer": "In front of you.", + "question_chinese": "地上装着东西的透明盒子在我的哪个方位？", + "answer_chinese": "你的前面", + "evidence": [ + { + "description": "you saw three transparent boxes on the ground filled with things", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Behind you.", + "B": "To your left.", + "C": "In front of you.", + "D": "It’s on the floor to your right, not directly in front of you.", + "E": "You left it on the floor behind you, so it isn’t in your forward view." + }, + "answer_label": "C", + "answer_text": "In front of you.", + "q_id": "scene0763_00-0_2_T1", + "correct_answer": "In front of you.", + "original_options": { + "A": "Behind you.", + "B": "To your left.", + "C": "In front of you.", + "D": "It’s on the floor to your right, not directly in front of you.", + "E": "You left it on the floor behind you, so it isn’t in your forward view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 2, + 3, + 14, + 16 + ] + }, + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_2_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the transparent box on the floor that has items inside, relative to me?", + "answer": "To your right.", + "question_chinese": "地上装着东西的透明盒子在我的哪个方位？", + "answer_chinese": "在你的右边", + "evidence": [ + { + "description": "you saw three transparent boxes on the ground filled with things", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "It’s positioned behind you on the floor rather than on your right side.", + "D": "The box sits on the floor in front of you, not off to your right.", + "E": "To your right." + }, + "answer_label": "E", + "answer_text": "To your right.", + "q_id": "scene0763_00-0_2_T2", + "correct_answer": "To your right.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "It’s positioned behind you on the floor rather than on your right side.", + "D": "The box sits on the floor in front of you, not off to your right.", + "E": "To your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 14, + 15 + ] + }, + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_1_T1", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the shoe box on the shelf on the left side right now?", + "answer": "You can’t reach the shoe box in front of you.", + "question_chinese": "我现在可以碰到左侧架子上的鞋盒吗？", + "answer_chinese": "你不可以碰到面前的鞋盒", + "evidence": [ + { + "description": "There are several boxes on the shelf in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the shoe box in front of you.", + "B": "You can reach the shoe box to your left." + }, + "answer_label": "A", + "answer_text": "You can’t reach the shoe box in front of you.", + "q_id": "scene0763_00-0_1_T1", + "correct_answer": "You can’t reach the shoe box in front of you.", + "original_options": { + "A": "You can’t reach the shoe box in front of you.", + "B": "You can reach the shoe box to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_1_T2", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the shoe box on the shelf on the left side right now?", + "answer": "You can reach the shoe box to your left.", + "question_chinese": "我现在可以碰到左侧架子上的鞋盒吗？", + "answer_chinese": "你可以碰到可以左边的鞋盒", + "evidence": [ + { + "description": "The shoe box is very close to you, and you can reach it.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the shoe box to your left.", + "B": "You can reach the shoe box to your left." + }, + "answer_label": "B", + "answer_text": "You can reach the shoe box to your left.", + "q_id": "scene0763_00-0_1_T2", + "correct_answer": "You can reach the shoe box to your left.", + "original_options": { + "A": "You can’t reach the shoe box to your left.", + "B": "You can reach the shoe box to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 10 + ] + }, + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_1_T3", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the shoe box on the shelf on the left side right now?", + "answer": "You can’t reach the shoe box to your left.", + "question_chinese": "我现在可以碰到左侧架子上的鞋盒吗？", + "answer_chinese": "你不可以碰到左边的鞋盒", + "evidence": [ + { + "description": "The shoe box is very close to you, and you can reach it.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can’t reach the shoe box to your left.", + "B": "You can reach the shoe box to your left." + }, + "answer_label": "A", + "answer_text": "You can’t reach the shoe box to your left.", + "q_id": "scene0763_00-0_1_T3", + "correct_answer": "You can’t reach the shoe box to your left.", + "original_options": { + "A": "You can’t reach the shoe box to your left.", + "B": "You can reach the shoe box to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_1_T4", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I reach the shoe box on the shelf on the left side right now?", + "answer": "You can’t reach the shoe box to your left.", + "question_chinese": "我现在可以碰到左侧架子上的鞋盒吗？", + "answer_chinese": "你不可以碰到左边的鞋盒", + "evidence": [ + { + "description": "The shoe box is very close to you, and you can reach it.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the shoe box to your left.", + "B": "You can’t reach the shoe box to your left." + }, + "answer_label": "B", + "answer_text": "You can’t reach the shoe box to your left.", + "q_id": "scene0763_00-0_1_T4", + "correct_answer": "You can’t reach the shoe box to your left.", + "original_options": { + "A": "You can reach the shoe box to your left.", + "B": "You can’t reach the shoe box to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_3_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes did I see?", + "answer": "There are 9 on the shelf in front of you.", + "question_chinese": "我看见几个盒子？", + "answer_chinese": "你面前的架子上有9个", + "evidence": [ + { + "description": "saw nine different shoe boxes", + "timestamps": { + "start": "00:00:02", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 8 boxes.", + "B": "There are 10 boxes.", + "C": "There are 7 on the shelf in front of you, not the number you remembered.", + "D": "There are 9 on the shelf in front of you.", + "E": "There are 11 on the shelf in front of you, which is more than you counted." + }, + "answer_label": "D", + "answer_text": "There are 9 on the shelf in front of you.", + "q_id": "scene0763_00-0_3_T2", + "correct_answer": "There are 9 on the shelf in front of you.", + "original_options": { + "A": "There are 8 boxes.", + "B": "There are 10 boxes.", + "C": "There are 7 on the shelf in front of you, not the number you remembered.", + "D": "There are 9 on the shelf in front of you.", + "E": "There are 11 on the shelf in front of you, which is more than you counted." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 16 + ] + }, + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_3_T1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many boxes did I see?", + "answer": "There are 9 on the shelf to your left, and 1 on the floor in front of you.", + "question_chinese": "我看见几个盒子？", + "answer_chinese": "你左边的架子上有9个，前面的地板上有一个", + "evidence": [ + { + "description": "saw nine different shoe boxes", + "timestamps": { + "start": "00:00:02", + "end": "00:00:09" + } + }, + { + "description": "There is a box on the ground.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are 8 total boxes.", + "B": "There are 11 total boxes.", + "C": "There are 9 on the shelf to your left, and 1 on the floor in front of you.", + "D": "There are 8 on the shelf to your left, and 1 on the floor in front of you.", + "E": "There are 10 on the shelf to your left, and 1 on the floor in front of you." + }, + "answer_label": "C", + "answer_text": "There are 9 on the shelf to your left, and 1 on the floor in front of you.", + "q_id": "scene0763_00-0_3_T1", + "correct_answer": "There are 9 on the shelf to your left, and 1 on the floor in front of you.", + "original_options": { + "A": "There are 8 total boxes.", + "B": "There are 11 total boxes.", + "C": "There are 9 on the shelf to your left, and 1 on the floor in front of you.", + "D": "There are 8 on the shelf to your left, and 1 on the floor in front of you.", + "E": "There are 10 on the shelf to your left, and 1 on the floor in front of you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 18 + ] + }, + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_4_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where were the lights in the room that I saw?", + "answer": "The room light is in front of you.", + "question_chinese": "我看见房间中的灯在哪里？", + "answer_chinese": "房间的灯在你面前", + "evidence": [ + { + "description": "In front of you", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The room light is behind you.", + "B": "The room light is to your left.", + "C": "The room light is to your right as you face into the room from where you were standing.", + "D": "The room light is in front of you.", + "E": "The room light is above your head, centered on the ceiling rather than ahead of you." + }, + "answer_label": "D", + "answer_text": "The room light is in front of you.", + "q_id": "scene0763_00-0_4_T1", + "correct_answer": "The room light is in front of you.", + "original_options": { + "A": "The room light is behind you.", + "B": "The room light is to your left.", + "C": "The room light is to your right as you face into the room from where you were standing.", + "D": "The room light is in front of you.", + "E": "The room light is above your head, centered on the ceiling rather than ahead of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 16 + ] + }, + { + "video_uid": "scene0763_00-0", + "video_length": "00:00:20", + "qid": "scene0763_00-0_4_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where were the lights in the room that I saw?", + "answer": "The room light is above your head.", + "question_chinese": "我看见房间中的灯在哪里？", + "answer_chinese": "房间的灯在你头顶", + "evidence": [ + { + "description": "In front of you", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The room light is above your head.", + "B": "The room light is in front of you.", + "C": "The room light is behind you.", + "D": "The room light is to your left when you look into the room, not directly overhead.", + "E": "The room light is to your right side, positioned off-center instead of above your head." + }, + "answer_label": "A", + "answer_text": "The room light is above your head.", + "q_id": "scene0763_00-0_4_T2", + "correct_answer": "The room light is above your head.", + "original_options": { + "A": "The room light is above your head.", + "B": "The room light is in front of you.", + "C": "The room light is behind you.", + "D": "The room light is to your left when you look into the room, not directly overhead.", + "E": "The room light is to your right side, positioned off-center instead of above your head." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 6, + 16, + 15 + ] + } +] diff --git a/QAs/scene0766_00-0-mcq.json b/QAs/scene0766_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1f9653e77a18412d7ee748c26f128d305b325c6e --- /dev/null +++ b/QAs/scene0766_00-0-mcq.json @@ -0,0 +1,645 @@ +[ + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_6", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction of mine is the cabinet on the left side of the door just entered now?", + "answer": "The cabinet on the left side of the door just entered is now behind you.", + "question_chinese": "刚进门左侧的柜子现在在我的哪个方向", + "answer_chinese": "刚进门左侧的柜子现在在你的后边", + "evidence": [ + { + "description": "The cabinet on the left side of the door just entered is now behind you", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is in front of you.", + "B": "It is on your left.", + "C": "The cabinet on the left side of the door you entered is now to your right as you face forward.", + "D": "The cabinet on the left side of the door just entered is now off to your front-left, not behind you.", + "E": "The cabinet on the left side of the door just entered is now behind you." + }, + "answer_label": "E", + "answer_text": "The cabinet on the left side of the door just entered is now behind you.", + "q_id": "scene0766_00-0_6", + "correct_answer": "The cabinet on the left side of the door just entered is now behind you.", + "original_options": { + "A": "It is in front of you.", + "B": "It is on your left.", + "C": "The cabinet on the left side of the door you entered is now to your right as you face forward.", + "D": "The cabinet on the left side of the door just entered is now off to your front-left, not behind you.", + "E": "The cabinet on the left side of the door just entered is now behind you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 20, + 20 + ] + }, + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_7", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Which direction of mine is the cabinet on the left side of the door just entered now?", + "answer": "The cabinet on the left side of the door just entered is now on your right.", + "question_chinese": "刚进门左侧的柜子现在在我的哪个方向", + "answer_chinese": "刚进门左侧的柜子现在在你的右边", + "evidence": [ + { + "description": "The cabinet on the left side of the door just entered is now on your right", + "timestamps": { + "start": "00:00:00", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The cabinet on the left side of the door just entered is now on your right.", + "B": "It is behind you.", + "C": "It is in front of you.", + "D": "The cabinet on the left side of the door just entered is now on your left, not your right.", + "E": "The cabinet on the left side of the door just entered is now slightly front-left relative to you after entering." + }, + "answer_label": "A", + "answer_text": "The cabinet on the left side of the door just entered is now on your right.", + "q_id": "scene0766_00-0_7", + "correct_answer": "The cabinet on the left side of the door just entered is now on your right.", + "original_options": { + "A": "The cabinet on the left side of the door just entered is now on your right.", + "B": "It is behind you.", + "C": "It is in front of you.", + "D": "The cabinet on the left side of the door just entered is now on your left, not your right.", + "E": "The cabinet on the left side of the door just entered is now slightly front-left relative to you after entering." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 19, + 20 + ] + }, + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_5_T2", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "If I want to get into bed to sleep, how should I walk?", + "answer": "You should just walk straight ahead, and you’ll be at the bedside.", + "question_chinese": "如果我想上床睡觉应该怎么走？", + "answer_chinese": "你应该直走即到床边", + "evidence": [ + { + "description": "The bed is in front of you", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Turn left, then walk forward.", + "B": "Turn right, then walk forward.", + "C": "You should just walk straight ahead, and you’ll be at the bedside.", + "D": "Turn around first, walk straight three steps, then turn left to reach the bedside.", + "E": "Walk straight a few steps, then turn right at the end to approach the bed from the side." + }, + "answer_label": "C", + "answer_text": "You should just walk straight ahead, and you’ll be at the bedside.", + "q_id": "scene0766_00-0_5_T2", + "correct_answer": "You should just walk straight ahead, and you’ll be at the bedside.", + "original_options": { + "A": "Turn left, then walk forward.", + "B": "Turn right, then walk forward.", + "C": "You should just walk straight ahead, and you’ll be at the bedside.", + "D": "Turn around first, walk straight three steps, then turn left to reach the bedside.", + "E": "Walk straight a few steps, then turn right at the end to approach the bed from the side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 18 + ] + }, + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_5_T1", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "Object Trajectory", + "question": "If I want to get into bed to sleep, how should I walk?", + "answer": "You should turn around first, then walk straight three steps, then turn right, and you’ll be able to get onto the bed to sleep.", + "question_chinese": "如果我想上床睡觉应该怎么走？", + "answer_chinese": "你应该先向后转，然后直走三步，然后右转就可以上床睡觉", + "evidence": [ + { + "description": "The bed is behind you to the left", + "timestamps": { + "start": "00:00:00", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Just walk straight ahead.", + "B": "Turn left after three steps.", + "C": "Turn around first, walk straight two steps, then turn right and you’ll reach the bed.", + "D": "Turn around, walk straight three steps, then turn right twice to line up with the bed to sleep.", + "E": "You should turn around first, then walk straight three steps, then turn right, and you’ll be able to get onto the bed to sleep." + }, + "answer_label": "E", + "answer_text": "You should turn around first, then walk straight three steps, then turn right, and you’ll be able to get onto the bed to sleep.", + "q_id": "scene0766_00-0_5_T1", + "correct_answer": "You should turn around first, then walk straight three steps, then turn right, and you’ll be able to get onto the bed to sleep.", + "original_options": { + "A": "Just walk straight ahead.", + "B": "Turn left after three steps.", + "C": "Turn around first, walk straight two steps, then turn right and you’ll reach the bed.", + "D": "Turn around, walk straight three steps, then turn right twice to line up with the bed to sleep.", + "E": "You should turn around first, then walk straight three steps, then turn right, and you’ll be able to get onto the bed to sleep." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 18 + ] + }, + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_4_T1", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the bed relative to me?", + "answer": "The bed was to your front-right.", + "question_chinese": "床在相对于我的什么位置？", + "answer_chinese": "床在你的右前方", + "evidence": [ + { + "description": "The bed is in front of me on the right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bed was to your front-left.", + "B": "The bed was to your back-right.", + "C": "The bed was directly in front of you rather than diagonally to your front-right.", + "D": "The bed was to your front-right.", + "E": "The bed was behind you, so you would need to turn around to face it." + }, + "answer_label": "D", + "answer_text": "The bed was to your front-right.", + "q_id": "scene0766_00-0_4_T1", + "correct_answer": "The bed was to your front-right.", + "original_options": { + "A": "The bed was to your front-left.", + "B": "The bed was to your back-right.", + "C": "The bed was directly in front of you rather than diagonally to your front-right.", + "D": "The bed was to your front-right.", + "E": "The bed was behind you, so you would need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 14, + 15 + ] + }, + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_4_T2", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the bed relative to me?", + "answer": "The bed was behind you.", + "question_chinese": "床在相对于我的什么位置？", + "answer_chinese": "床在我的后面", + "evidence": [ + { + "description": "The bed is in front of me on the right.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:05" + } + }, + { + "description": "I turned 180° to the right", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The bed was behind you.", + "B": "The bed was in front of you.", + "C": "The bed was to your left.", + "D": "The bed was to your front-right, positioned diagonally ahead instead of behind you.", + "E": "The bed was to your right side, so you would look right to see it rather than turning around." + }, + "answer_label": "A", + "answer_text": "The bed was behind you.", + "q_id": "scene0766_00-0_4_T2", + "correct_answer": "The bed was behind you.", + "original_options": { + "A": "The bed was behind you.", + "B": "The bed was in front of you.", + "C": "The bed was to your left.", + "D": "The bed was to your front-right, positioned diagonally ahead instead of behind you.", + "E": "The bed was to your right side, so you would look right to see it rather than turning around." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 13, + 19 + ] + }, + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_2_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What do I see about the state of the curtains right now?", + "answer": "You saw that the curtains were closed.", + "question_chinese": "我看到窗帘的状态是什么？", + "answer_chinese": "你看到窗帘是关闭的", + "evidence": [ + { + "description": "The curtains are drawn", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The curtains were open.", + "B": "You saw that the curtains were closed.", + "C": "The curtains were half-open.", + "D": "You saw the curtains were drawn open, letting in light from outside the window.", + "E": "You noticed the curtains were partially open, with a gap in the middle showing the window." + }, + "answer_label": "B", + "answer_text": "You saw that the curtains were closed.", + "q_id": "scene0766_00-0_2_T2", + "correct_answer": "You saw that the curtains were closed.", + "original_options": { + "A": "The curtains were open.", + "B": "You saw that the curtains were closed.", + "C": "The curtains were half-open.", + "D": "You saw the curtains were drawn open, letting in light from outside the window.", + "E": "You noticed the curtains were partially open, with a gap in the middle showing the window." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 16 + ] + }, + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_2_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "What do I see about the state of the curtains right now?", + "answer": "The curtains were still closed—you never opened them the whole time.", + "question_chinese": "我看到窗帘的状态是什么？", + "answer_chinese": "窗帘还是关闭的，你从始至终没有打开过窗帘", + "evidence": [ + { + "description": "The curtains are drawn", + "timestamps": { + "start": "00:00:01", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The curtains were open.", + "B": "The curtains were partly open.", + "C": "You saw that the curtains were open, as if they had been pulled back earlier.", + "D": "You noticed the curtains were not fully closed, leaving a slit of light coming through.", + "E": "The curtains were still closed—you never opened them the whole time." + }, + "answer_label": "E", + "answer_text": "The curtains were still closed—you never opened them the whole time.", + "q_id": "scene0766_00-0_2_T1", + "correct_answer": "The curtains were still closed—you never opened them the whole time.", + "original_options": { + "A": "The curtains were open.", + "B": "The curtains were partly open.", + "C": "You saw that the curtains were open, as if they had been pulled back earlier.", + "D": "You noticed the curtains were not fully closed, leaving a slit of light coming through.", + "E": "The curtains were still closed—you never opened them the whole time." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 15 + ] + }, + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_1_T2", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I entered the room, what route did I walk?", + "answer": "So far, you’ve been walking straight until you reached next to the foot of the bed.", + "question_chinese": "我进入房间后走了什么路线？", + "answer_chinese": "你目前一直直走到床脚旁边", + "evidence": [ + { + "description": "After you enter the room, walk straight to the bedside.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You turned left immediately.", + "B": "You walked to the headboard.", + "C": "So far, you’ve been walking straight until you reached next to the foot of the bed.", + "D": "You walked straight for a bit, then turned right to stand beside the bed.", + "E": "After entering, you veered left around the bed and stopped near the corner by the wall." + }, + "answer_label": "C", + "answer_text": "So far, you’ve been walking straight until you reached next to the foot of the bed.", + "q_id": "scene0766_00-0_1_T2", + "correct_answer": "So far, you’ve been walking straight until you reached next to the foot of the bed.", + "original_options": { + "A": "You turned left immediately.", + "B": "You walked to the headboard.", + "C": "So far, you’ve been walking straight until you reached next to the foot of the bed.", + "D": "You walked straight for a bit, then turned right to stand beside the bed.", + "E": "After entering, you veered left around the bed and stopped near the corner by the wall." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 14, + 16 + ] + }, + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_1_T1", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "After I entered the room, what route did I walk?", + "answer": "You walked to the corner of the bed and turned one full circle clockwise in place.", + "question_chinese": "我进入房间后走了什么路线？", + "answer_chinese": "你走到床角旁，顺时针原地转了一圈", + "evidence": [ + { + "description": "You walked to the corner of the bed and turned clockwise in place once", + "timestamps": { + "start": "00:00:00", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You walked straight to the foot.", + "B": "You walked to the corner of the bed and turned one full circle clockwise in place.", + "C": "You turned around and backed up.", + "D": "You went to the corner of the bed and turned one full circle counterclockwise in place.", + "E": "After reaching the bed corner, you paused briefly and then turned only half a circle before stopping." + }, + "answer_label": "B", + "answer_text": "You walked to the corner of the bed and turned one full circle clockwise in place.", + "q_id": "scene0766_00-0_1_T1", + "correct_answer": "You walked to the corner of the bed and turned one full circle clockwise in place.", + "original_options": { + "A": "You walked straight to the foot.", + "B": "You walked to the corner of the bed and turned one full circle clockwise in place.", + "C": "You turned around and backed up.", + "D": "You went to the corner of the bed and turned one full circle counterclockwise in place.", + "E": "After reaching the bed corner, you paused briefly and then turned only half a circle before stopping." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 17 + ] + }, + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_9", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction of the bed is the brown throw pillow?", + "answer": "The brown throw pillow is on the right side of the bed.", + "question_chinese": "棕色抱枕在床的哪个方向？", + "answer_chinese": "棕色抱枕在床的右边", + "evidence": [ + { + "description": "The brown throw pillow is on the right side of the bed", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s on the left side.", + "B": "It’s at the foot of the bed.", + "C": "The brown throw pillow is on the right side of the bed.", + "D": "The brown throw pillow is positioned near the head of the bed, closer to the top edge.", + "E": "The brown throw pillow sits centered on the bed, not clearly on either the left or right side." + }, + "answer_label": "C", + "answer_text": "The brown throw pillow is on the right side of the bed.", + "q_id": "scene0766_00-0_9", + "correct_answer": "The brown throw pillow is on the right side of the bed.", + "original_options": { + "A": "It’s on the left side.", + "B": "It’s at the foot of the bed.", + "C": "The brown throw pillow is on the right side of the bed.", + "D": "The brown throw pillow is positioned near the head of the bed, closer to the top edge.", + "E": "The brown throw pillow sits centered on the bed, not clearly on either the left or right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 17, + 18 + ] + }, + { + "video_uid": "scene0766_00-0", + "video_length": "00:00:20", + "qid": "scene0766_00-0_8", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "In which direction of the bed is the brown throw pillow?", + "answer": "The brown throw pillow is on the right side of the bed.", + "question_chinese": "棕色抱枕在床的哪个方向？", + "answer_chinese": "棕色抱枕在床的右边", + "evidence": [ + { + "description": "The brown throw pillow is on the right side of the bed", + "timestamps": { + "start": "00:00:12", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s on the left side.", + "B": "The brown throw pillow is on the right side of the bed.", + "C": "It’s at the head of the bed.", + "D": "The brown throw pillow is placed at the foot of the bed, nearer the bottom edge.", + "E": "The brown throw pillow is in the middle of the bed, rather than on the right side." + }, + "answer_label": "B", + "answer_text": "The brown throw pillow is on the right side of the bed.", + "q_id": "scene0766_00-0_8", + "correct_answer": "The brown throw pillow is on the right side of the bed.", + "original_options": { + "A": "It’s on the left side.", + "B": "The brown throw pillow is on the right side of the bed.", + "C": "It’s at the head of the bed.", + "D": "The brown throw pillow is placed at the foot of the bed, nearer the bottom edge.", + "E": "The brown throw pillow is in the middle of the bed, rather than on the right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 8, + 16, + 17 + ] + } +] diff --git a/QAs/scene0767_00-0-mcq.json b/QAs/scene0767_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..2a13e2f786c8b4bfdc354550a621ebb80c664ae4 --- /dev/null +++ b/QAs/scene0767_00-0-mcq.json @@ -0,0 +1,337 @@ +[ + { + "video_uid": "scene0767_00-0", + "video_length": "00:00:20", + "qid": "scene0767_00-0_1", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are my towel and bath towel?", + "answer": "The towel and bath towel are hung on the towel rack above the bathtub.", + "question_chinese": "我的毛巾和浴巾在哪里？", + "answer_chinese": "毛巾和浴巾挂在浴缸上方的毛巾架上。", + "evidence": [ + { + "description": "The towel and bath towel are hung on the towel rack above the bathtub.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They’re hung over the door.", + "B": "They’re draped on the bathtub edge.", + "C": "The towel and bath towel are hung on the towel rack above the bathtub.", + "D": "The towel and bath towel are hanging on hooks next to the bathroom sink, close to the counter.", + "E": "The towel and bath towel are hung on a towel rack beside the shower area, not above the tub." + }, + "answer_label": "C", + "answer_text": "The towel and bath towel are hung on the towel rack above the bathtub.", + "q_id": "scene0767_00-0_1", + "correct_answer": "The towel and bath towel are hung on the towel rack above the bathtub.", + "original_options": { + "A": "They’re hung over the door.", + "B": "They’re draped on the bathtub edge.", + "C": "The towel and bath towel are hung on the towel rack above the bathtub.", + "D": "The towel and bath towel are hanging on hooks next to the bathroom sink, close to the counter.", + "E": "The towel and bath towel are hung on a towel rack beside the shower area, not above the tub." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 18, + 19 + ] + }, + { + "video_uid": "scene0767_00-0", + "video_length": "00:00:20", + "qid": "scene0767_00-0_2", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are my towel and bath towel?", + "answer": "The towel and bath towel are hung on the towel rack above the bathtub.", + "question_chinese": "我的毛巾和浴巾在哪里？", + "answer_chinese": "毛巾和浴巾挂在浴缸上方的毛巾架上。", + "evidence": [ + { + "description": "The towel and bath towel are hung on the towel rack above the bathtub.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They’re on hooks by the door.", + "B": "They’re on the sink towel bar.", + "C": "The towel and bath towel are draped over the shower curtain rod, hanging down inside the tub.", + "D": "The towel and bath towel are hung on the towel rack above the bathtub.", + "E": "The towel and bath towel are folded on a shelf above the toilet instead of being hung on a rack." + }, + "answer_label": "D", + "answer_text": "The towel and bath towel are hung on the towel rack above the bathtub.", + "q_id": "scene0767_00-0_2", + "correct_answer": "The towel and bath towel are hung on the towel rack above the bathtub.", + "original_options": { + "A": "They’re on hooks by the door.", + "B": "They’re on the sink towel bar.", + "C": "The towel and bath towel are draped over the shower curtain rod, hanging down inside the tub.", + "D": "The towel and bath towel are hung on the towel rack above the bathtub.", + "E": "The towel and bath towel are folded on a shelf above the toilet instead of being hung on a rack." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 20 + ] + }, + { + "video_uid": "scene0767_00-0", + "video_length": "00:00:20", + "qid": "scene0767_00-0_3", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of toiletries are there in the bathroom?", + "answer": "There are a total of three bottles.", + "question_chinese": "浴室里一共有几瓶洗护用品？", + "answer_chinese": "一共有三瓶。", + "evidence": [ + { + "description": "There are three bottles in the corner of the bathtub.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are a total of three bottles.", + "B": "There are a total of two bottles.", + "C": "There are a total of four bottles.", + "D": "There are a total of five bottles of toiletries in the bathroom.", + "E": "There are a total of seven bottles of toiletries in the bathroom." + }, + "answer_label": "A", + "answer_text": "There are a total of three bottles.", + "q_id": "scene0767_00-0_3", + "correct_answer": "There are a total of three bottles.", + "original_options": { + "A": "There are a total of three bottles.", + "B": "There are a total of two bottles.", + "C": "There are a total of four bottles.", + "D": "There are a total of five bottles of toiletries in the bathroom.", + "E": "There are a total of seven bottles of toiletries in the bathroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 12 + ] + }, + { + "video_uid": "scene0767_00-0", + "video_length": "00:00:20", + "qid": "scene0767_00-0_4", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of toiletries are there in the bathroom?", + "answer": "There are a total of eight bottles.", + "question_chinese": "浴室里一共有几瓶洗护用品？", + "answer_chinese": "一共有八瓶。", + "evidence": [ + { + "description": "There are three bottles in the corner of the bathtub.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + }, + { + "description": "There are five bottles on the windowsill.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are a total of seven bottles.", + "B": "There are a total of nine bottles.", + "C": "There are a total of eight bottles.", + "D": "There are a total of six bottles of toiletries in the bathroom.", + "E": "There are a total of ten bottles of toiletries in the bathroom." + }, + "answer_label": "C", + "answer_text": "There are a total of eight bottles.", + "q_id": "scene0767_00-0_4", + "correct_answer": "There are a total of eight bottles.", + "original_options": { + "A": "There are a total of seven bottles.", + "B": "There are a total of nine bottles.", + "C": "There are a total of eight bottles.", + "D": "There are a total of six bottles of toiletries in the bathroom.", + "E": "There are a total of ten bottles of toiletries in the bathroom." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 12, + 12 + ] + }, + { + "video_uid": "scene0767_00-0", + "video_length": "00:00:20", + "qid": "scene0767_00-0_5", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the blue cap white bottle or the purple bottle?", + "answer": "The blue cap white bottle are closer to you.", + "question_chinese": "蓝盖白瓶和紫瓶的哪个离我更近？", + "answer_chinese": "蓝盖白瓶的离你更近。", + "evidence": [ + { + "description": "The blue cap and white bottle are in the upper left corner of the bathtub.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + }, + { + "description": "The purple bottle is on the windowsill.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "You stand at the door on the left side of the bathtub.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The blue cap white bottle are closer to you.", + "B": "The purple bottle is closer to you." + }, + "answer_label": "A", + "answer_text": "The blue cap white bottle are closer to you.", + "q_id": "scene0767_00-0_5", + "correct_answer": "The blue cap white bottle are closer to you.", + "original_options": { + "A": "The blue cap white bottle are closer to you.", + "B": "The purple bottle is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0767_00-0", + "video_length": "00:00:20", + "qid": "scene0767_00-0_6", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which one is closer to me, the blue cap white bottle or the purple bottle?", + "answer": "The blue cap and white bottle are closer to you.", + "question_chinese": "蓝盖白瓶和紫瓶的哪个离我更近？", + "answer_chinese": "蓝盖白瓶的离你更近。", + "evidence": [ + { + "description": "The blue cap and white bottle are in the upper left corner of the bathtub.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + }, + { + "description": "The purple bottle is on the windowsill.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "You stand at the door on the left side of the bathtub.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The blue cap and white bottle are closer to you.", + "B": "The purple bottle is closer to you." + }, + "answer_label": "A", + "answer_text": "The blue cap and white bottle are closer to you.", + "q_id": "scene0767_00-0_6", + "correct_answer": "The blue cap and white bottle are closer to you.", + "original_options": { + "A": "The blue cap and white bottle are closer to you.", + "B": "The purple bottle is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + } +] diff --git a/QAs/scene0773_00-0-mcq.json b/QAs/scene0773_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..4643f6e593daa84e268500ffbbc202dbe6d87587 --- /dev/null +++ b/QAs/scene0773_00-0-mcq.json @@ -0,0 +1,330 @@ +[ + { + "video_uid": "scene0773_00-0", + "video_length": "00:00:20", + "qid": "scene0773_00-0_7", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow paper strips are there on the refrigerator?", + "answer": "There are two yellow paper strips on the refrigerator.", + "question_chinese": "冰箱上有几个黄色的纸条？", + "answer_chinese": "有两个。", + "evidence": [ + { + "description": "There are two yellow paper strips on the refrigerator.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one yellow paper strip.", + "B": "There are three yellow paper strips.", + "C": "There are four yellow paper strips on the refrigerator door, lined up along the left side.", + "D": "There are five yellow paper strips placed on the refrigerator, spread out across the front surface.", + "E": "There are two yellow paper strips on the refrigerator." + }, + "answer_label": "E", + "answer_text": "There are two yellow paper strips on the refrigerator.", + "q_id": "scene0773_00-0_7", + "correct_answer": "There are two yellow paper strips on the refrigerator.", + "original_options": { + "A": "There is one yellow paper strip.", + "B": "There are three yellow paper strips.", + "C": "There are four yellow paper strips on the refrigerator door, lined up along the left side.", + "D": "There are five yellow paper strips placed on the refrigerator, spread out across the front surface.", + "E": "There are two yellow paper strips on the refrigerator." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "scene0773_00-0", + "video_length": "00:00:20", + "qid": "scene0773_00-0_8", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow paper strips are there on the refrigerator?", + "answer": "There are two yellow paper strips on the refrigerator.", + "question_chinese": "冰箱上有几个黄色的纸条？", + "answer_chinese": "有两个。", + "evidence": [ + { + "description": "There are two yellow paper strips on the refrigerator.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one yellow paper strip.", + "B": "There are three yellow paper strips.", + "C": "There are two yellow paper strips on the refrigerator.", + "D": "There are four yellow paper strips on the refrigerator door, lined up along the left side.", + "E": "There are five yellow paper strips placed on the refrigerator, spread out across the front surface." + }, + "answer_label": "C", + "answer_text": "There are two yellow paper strips on the refrigerator.", + "q_id": "scene0773_00-0_8", + "correct_answer": "There are two yellow paper strips on the refrigerator.", + "original_options": { + "A": "There is one yellow paper strip.", + "B": "There are three yellow paper strips.", + "C": "There are two yellow paper strips on the refrigerator.", + "D": "There are four yellow paper strips on the refrigerator door, lined up along the left side.", + "E": "There are five yellow paper strips placed on the refrigerator, spread out across the front surface." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 16, + 16 + ] + }, + { + "video_uid": "scene0773_00-0", + "video_length": "00:00:20", + "qid": "scene0773_00-0_3", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the refrigerator or the water dispenser?", + "answer": "The water dispenser is closer to you.", + "question_chinese": "冰箱和饮水机哪个离我更近？", + "answer_chinese": "饮水机离你更近。", + "evidence": [ + { + "description": "The water dispenser is next to the cabinet in front of you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "The refrigerator is in the corner.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "The water dispenser is closer to you.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The water dispenser is closer to you.", + "B": "The refrigerator is closer to you." + }, + "answer_label": "A", + "answer_text": "The water dispenser is closer to you.", + "q_id": "scene0773_00-0_3", + "correct_answer": "The water dispenser is closer to you.", + "original_options": { + "A": "The water dispenser is closer to you.", + "B": "The refrigerator is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0773_00-0", + "video_length": "00:00:20", + "qid": "scene0773_00-0_4", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the refrigerator or the water dispenser?", + "answer": "The water dispenser is closer to you.", + "question_chinese": "冰箱和饮水机哪个离我更近？", + "answer_chinese": "饮水机离你更近。", + "evidence": [ + { + "description": "The water dispenser is next to the cabinet in front of you.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:02" + } + }, + { + "description": "The refrigerator is in the corner.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The refrigerator is in the corner.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "The refrigerator is closer to you.", + "B": "The water dispenser is closer to you." + }, + "answer_label": "B", + "answer_text": "The water dispenser is closer to you.", + "q_id": "scene0773_00-0_4", + "correct_answer": "The water dispenser is closer to you.", + "original_options": { + "A": "The refrigerator is closer to you.", + "B": "The water dispenser is closer to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0773_00-0", + "video_length": "00:00:20", + "qid": "scene0773_00-0_9", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my white water bottle?", + "answer": "Your white water cup is on the left side of the sink.", + "question_chinese": "我的白色水杯在哪里？", + "answer_chinese": "你的白色水杯在水槽的左边。", + "evidence": [ + { + "description": "Your white water cup is on the left side of the sink.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the right side of the sink.", + "B": "Your white water cup is on the left side of the sink.", + "C": "In the sink basin.", + "D": "Your white water cup is on the counter to the right of the sink, near the edge.", + "E": "Your white water bottle is tucked behind the faucet, resting on the back rim of the sink." + }, + "answer_label": "B", + "answer_text": "Your white water cup is on the left side of the sink.", + "q_id": "scene0773_00-0_9", + "correct_answer": "Your white water cup is on the left side of the sink.", + "original_options": { + "A": "On the right side of the sink.", + "B": "Your white water cup is on the left side of the sink.", + "C": "In the sink basin.", + "D": "Your white water cup is on the counter to the right of the sink, near the edge.", + "E": "Your white water bottle is tucked behind the faucet, resting on the back rim of the sink." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 4, + 17, + 17 + ] + }, + { + "video_uid": "scene0773_00-0", + "video_length": "00:00:20", + "qid": "scene0773_00-0_10", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my white water bottle?", + "answer": "Your white water cup is on the left side of the sink.", + "question_chinese": "我的白色水杯在哪里？", + "answer_chinese": "你的白色水杯在水槽的左边。", + "evidence": [ + { + "description": "Your white water cup is on the left side of the sink.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Your white water cup is on the left side of the sink.", + "B": "On the right side of the sink.", + "C": "In the center of the counter.", + "D": "Your white water bottle is inside the drawer next to the stove, pushed toward the back.", + "E": "Your white water bottle is under the table near the wall, positioned close to the nearest leg." + }, + "answer_label": "A", + "answer_text": "Your white water cup is on the left side of the sink.", + "q_id": "scene0773_00-0_10", + "correct_answer": "Your white water cup is on the left side of the sink.", + "original_options": { + "A": "Your white water cup is on the left side of the sink.", + "B": "On the right side of the sink.", + "C": "In the center of the counter.", + "D": "Your white water bottle is inside the drawer next to the stove, pushed toward the back.", + "E": "Your white water bottle is under the table near the wall, positioned close to the nearest leg." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 16, + 17 + ] + } +] diff --git a/QAs/scene0774_00-0-mcq.json b/QAs/scene0774_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..340b066bc47cc6aeb9be0b1a1f2f40cbc3e2c377 --- /dev/null +++ b/QAs/scene0774_00-0-mcq.json @@ -0,0 +1,855 @@ +[ + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_6", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "First, you face the trash can, then move forward a little, then turn left and walk straight.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先你面对着垃圾桶 ，然后向前走一点，再向左转并直走。", + "evidence": [ + { + "description": "First, you face the trash can.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you move forward a little", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "Then you turn left and walk straight.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Face the trash can, then turn right.", + "B": "Face the trash can, then walk backward.", + "C": "First, you face the trash can, then move forward a little, then turn left and walk straight.", + "D": "First, you face the trash can, move forward a little, then turn right and walk straight ahead.", + "E": "First, you face the trash can, move forward a little, then turn left and walk diagonally instead of straight." + }, + "answer_label": "C", + "answer_text": "First, you face the trash can, then move forward a little, then turn left and walk straight.", + "q_id": "scene0774_00-0_6", + "correct_answer": "First, you face the trash can, then move forward a little, then turn left and walk straight.", + "original_options": { + "A": "Face the trash can, then turn right.", + "B": "Face the trash can, then walk backward.", + "C": "First, you face the trash can, then move forward a little, then turn left and walk straight.", + "D": "First, you face the trash can, move forward a little, then turn right and walk straight ahead.", + "E": "First, you face the trash can, move forward a little, then turn left and walk diagonally instead of straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 17, + 19 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_5", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "First, you face the trash can, then move forward a little, then turn left and walk straight.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先你面对着垃圾桶 ，然后向前走一点，再向左转并直走。", + "evidence": [ + { + "description": "First, you face the trash can.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you move forward a little.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "Then you turn left and walk straight.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Face the trash can, then turn right.", + "B": "Face the chair, then walk straight.", + "C": "First, you face the trash can, move forward a little, then turn right and walk straight ahead.", + "D": "First, you face the trash can, move forward a little, then turn left and walk diagonally rather than straight.", + "E": "First, you face the trash can, then move forward a little, then turn left and walk straight." + }, + "answer_label": "E", + "answer_text": "First, you face the trash can, then move forward a little, then turn left and walk straight.", + "q_id": "scene0774_00-0_5", + "correct_answer": "First, you face the trash can, then move forward a little, then turn left and walk straight.", + "original_options": { + "A": "Face the trash can, then turn right.", + "B": "Face the chair, then walk straight.", + "C": "First, you face the trash can, move forward a little, then turn right and walk straight ahead.", + "D": "First, you face the trash can, move forward a little, then turn left and walk diagonally rather than straight.", + "E": "First, you face the trash can, then move forward a little, then turn left and walk straight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 6, + 17, + 19 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_10", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow tables are there in the room?", + "answer": "There are three yellow tables in the room.", + "question_chinese": "房间里有几张黄色的桌子？", + "answer_chinese": "房间里有三张黄色的桌子。", + "evidence": [ + { + "description": "There are three tables standing upright.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two yellow tables.", + "B": "There are three yellow tables in the room.", + "C": "There are four yellow tables.", + "D": "There is one yellow table in the room, and no other yellow tables are visible.", + "E": "There are five yellow tables in the room, spread around the space rather than clustered together." + }, + "answer_label": "B", + "answer_text": "There are three yellow tables in the room.", + "q_id": "scene0774_00-0_10", + "correct_answer": "There are three yellow tables in the room.", + "original_options": { + "A": "There are two yellow tables.", + "B": "There are three yellow tables in the room.", + "C": "There are four yellow tables.", + "D": "There is one yellow table in the room, and no other yellow tables are visible.", + "E": "There are five yellow tables in the room, spread around the space rather than clustered together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_9", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many yellow tables are there in the room?", + "answer": "There are five yellow tables in the room.", + "question_chinese": "房间里有几张黄色的桌子？", + "answer_chinese": "房间里有五张黄色的桌子。", + "evidence": [ + { + "description": "There are three tables standing upright.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "There are two tables lying horizontally.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are five yellow tables in the room.", + "B": "There are four yellow tables.", + "C": "There are six yellow tables.", + "D": "There are three yellow tables in the room, with only three distinct yellow tabletops visible.", + "E": "There are seven yellow tables in the room, including two additional ones near the far side." + }, + "answer_label": "A", + "answer_text": "There are five yellow tables in the room.", + "q_id": "scene0774_00-0_9", + "correct_answer": "There are five yellow tables in the room.", + "original_options": { + "A": "There are five yellow tables in the room.", + "B": "There are four yellow tables.", + "C": "There are six yellow tables.", + "D": "There are three yellow tables in the room, with only three distinct yellow tabletops visible.", + "E": "There are seven yellow tables in the room, including two additional ones near the far side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 16 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_4", + "qid_num": 5, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Are the black chairs next to the table and against the wall neatly arranged?", + "answer": "No,they aren't.", + "question_chinese": "桌子旁边的黑色椅子和靠墙的黑色椅子摆放的整齐吗？", + "answer_chinese": "不整齐。", + "evidence": [ + { + "description": "The chairs next to the table are not neatly arranged.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "The black chairs against the wall are neatly arranged.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes,they are.", + "B": "No,they aren't." + }, + "answer_label": "B", + "answer_text": "No,they aren't.", + "q_id": "scene0774_00-0_4", + "correct_answer": "No,they aren't.", + "original_options": { + "A": "Yes,they are.", + "B": "No,they aren't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_3", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Are the black chairs next to the table and against the wall neatly arranged?", + "answer": "No,they aren't.", + "question_chinese": "桌子旁边的黑色椅子和靠墙的黑色椅子摆放的整齐吗？", + "answer_chinese": "不整齐。", + "evidence": [ + { + "description": "The chairs next to the table are not neatly arranged.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "The black chairs against the wall are neatly arranged.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + }, + { + "description": "The black chairs against the wall are neatly arranged.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "The chairs next to the table are not neatly arranged.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "No,they aren't.", + "B": "Yes,they are." + }, + "answer_label": "A", + "answer_text": "No,they aren't.", + "q_id": "scene0774_00-0_3", + "correct_answer": "No,they aren't.", + "original_options": { + "A": "No,they aren't.", + "B": "Yes,they are." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_1_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, where is the trash can relative to me?", + "answer": "At this moment, the trash can was in front of you.", + "question_chinese": "此时，垃圾桶在我哪个方位？", + "answer_chinese": "此时，垃圾桶在你前面.", + "evidence": [ + { + "description": "The trash cans are in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "At this moment, the trash can was in front of you.", + "B": "It was behind you.", + "C": "It was to your left.", + "D": "At this moment, the trash can was to your right as you looked around from your spot.", + "E": "At this moment, the trash can was above you, positioned higher than your eye level." + }, + "answer_label": "A", + "answer_text": "At this moment, the trash can was in front of you.", + "q_id": "scene0774_00-0_1_T2", + "correct_answer": "At this moment, the trash can was in front of you.", + "original_options": { + "A": "At this moment, the trash can was in front of you.", + "B": "It was behind you.", + "C": "It was to your left.", + "D": "At this moment, the trash can was to your right as you looked around from your spot.", + "E": "At this moment, the trash can was above you, positioned higher than your eye level." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 15 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_1_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Right now, where is the trash can relative to me?", + "answer": "At this moment, the trash can was to your right.", + "question_chinese": "此时，垃圾桶在我哪个方位？", + "answer_chinese": "此时，垃圾桶在你右边", + "evidence": [ + { + "description": "The trash can are on your right.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "At this moment, the trash can was to your right.", + "B": "It was in front of you.", + "C": "It was behind you.", + "D": "At this moment, the trash can was to your left, not on the side you were facing.", + "E": "At this moment, the trash can was above you, higher up than where you were looking." + }, + "answer_label": "A", + "answer_text": "At this moment, the trash can was to your right.", + "q_id": "scene0774_00-0_1_T1", + "correct_answer": "At this moment, the trash can was to your right.", + "original_options": { + "A": "At this moment, the trash can was to your right.", + "B": "It was in front of you.", + "C": "It was behind you.", + "D": "At this moment, the trash can was to your left, not on the side you were facing.", + "E": "At this moment, the trash can was above you, higher up than where you were looking." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 17, + 16 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_3_T1", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, can I see the trash can?", + "answer": "At this moment, you could see the trash can; it was in front of you.", + "question_chinese": "此时，我能看得见垃圾桶吗？", + "answer_chinese": "此时，你能看得见垃圾桶，它们在你前面。", + "evidence": [ + { + "description": "You can see the trash cans.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "No, you couldn’t see it.", + "B": "At this moment, you could see the trash can; it was in front of you.", + "C": "Yes, it was to your left.", + "D": "At this moment, you could see the trash can, but it was positioned behind you instead of ahead.", + "E": "At this moment, you could see the trash can; it appeared to your right as you faced forward." + }, + "answer_label": "B", + "answer_text": "At this moment, you could see the trash can; it was in front of you.", + "q_id": "scene0774_00-0_3_T1", + "correct_answer": "At this moment, you could see the trash can; it was in front of you.", + "original_options": { + "A": "No, you couldn’t see it.", + "B": "At this moment, you could see the trash can; it was in front of you.", + "C": "Yes, it was to your left.", + "D": "At this moment, you could see the trash can, but it was positioned behind you instead of ahead.", + "E": "At this moment, you could see the trash can; it appeared to your right as you faced forward." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 18 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_3_T2", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Right now, can I see the trash can?", + "answer": "At this moment, you couldn’t see the trash can; it was to your right.", + "question_chinese": "此时，我能看得见垃圾桶吗？", + "answer_chinese": "此时，你不能看得见垃圾桶，它们在你右边", + "evidence": [ + { + "description": "You can not see the trash cans.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "At this moment, you couldn’t see the trash can; it was to your right.", + "B": "You could see the trash can.", + "C": "You couldn’t see it; left.", + "D": "At this moment, you couldn’t see the trash can because it was behind you rather than off to your right.", + "E": "At this moment, you could see the trash can, with it directly in front of you in your view." + }, + "answer_label": "A", + "answer_text": "At this moment, you couldn’t see the trash can; it was to your right.", + "q_id": "scene0774_00-0_3_T2", + "correct_answer": "At this moment, you couldn’t see the trash can; it was to your right.", + "original_options": { + "A": "At this moment, you couldn’t see the trash can; it was to your right.", + "B": "You could see the trash can.", + "C": "You couldn’t see it; left.", + "D": "At this moment, you couldn’t see the trash can because it was behind you rather than off to your right.", + "E": "At this moment, you could see the trash can, with it directly in front of you in your view." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 21, + 19 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_4_T1", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, can I see the things on the table?", + "answer": "At this moment, you couldn’t see the things on the table; the table was to your left.", + "question_chinese": "此时，我能看得见桌子上的东西吗？", + "answer_chinese": "此时，你不能看得见桌子上的东西，桌子在你左边。", + "evidence": [ + { + "description": "You can not see the thing on the table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "At this moment, you could see the things on the table; the table was to your left.", + "B": "At this moment, you couldn’t see the things on the table; the table was to your left." + }, + "answer_label": "B", + "answer_text": "At this moment, you couldn’t see the things on the table; the table was to your left.", + "q_id": "scene0774_00-0_4_T1", + "correct_answer": "At this moment, you couldn’t see the things on the table; the table was to your left.", + "original_options": { + "A": "At this moment, you could see the things on the table; the table was to your left.", + "B": "At this moment, you couldn’t see the things on the table; the table was to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_4_T2", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, can I see the things on the table?", + "answer": "Right now you can see something on the table: it’s a sheet of paper, and it’s in front of you.", + "question_chinese": "此时，我能看得见桌子上的东西吗？", + "answer_chinese": "此时，你能看得见桌子上的东西，它是一张纸，它在你前面。", + "evidence": [ + { + "description": "You can see the thing on the table.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now you cannot see something on the table; it’s to your left.", + "B": "Right now you can see something on the table: it’s a sheet of paper, and it’s in front of you." + }, + "answer_label": "B", + "answer_text": "Right now you can see something on the table: it’s a sheet of paper, and it’s in front of you.", + "q_id": "scene0774_00-0_4_T2", + "correct_answer": "Right now you can see something on the table: it’s a sheet of paper, and it’s in front of you.", + "original_options": { + "A": "Right now you cannot see something on the table; it’s to your left.", + "B": "Right now you can see something on the table: it’s a sheet of paper, and it’s in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 14 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_2_T1", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, in which direction is the chair against the wall relative to me?", + "answer": "At this moment, the chair against the wall was to your left.", + "question_chinese": "此时，靠墙的椅子在我的哪个方向?", + "answer_chinese": "此时，靠墙的椅子在你左边。", + "evidence": [ + { + "description": "The chairs are on your left.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "At this moment, the chair against the wall was to your right.", + "B": "At this moment, the chair against the wall was to your left." + }, + "answer_label": "B", + "answer_text": "At this moment, the chair against the wall was to your left.", + "q_id": "scene0774_00-0_2_T1", + "correct_answer": "At this moment, the chair against the wall was to your left.", + "original_options": { + "A": "At this moment, the chair against the wall was to your right.", + "B": "At this moment, the chair against the wall was to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_2_T2", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Right now, in which direction is the chair against the wall relative to me?", + "answer": "At this moment, the chair against the wall was in front of you.", + "question_chinese": "此时，靠墙的椅子在我的哪个方向?", + "answer_chinese": "此时，靠墙的椅子在你前面。", + "evidence": [ + { + "description": "The black chairs is in front of you.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "At this moment, the chair against the wall was in front of you.", + "B": "At this moment, the chair against the wall was behind you." + }, + "answer_label": "A", + "answer_text": "At this moment, the chair against the wall was in front of you.", + "q_id": "scene0774_00-0_2_T2", + "correct_answer": "At this moment, the chair against the wall was in front of you.", + "original_options": { + "A": "At this moment, the chair against the wall was in front of you.", + "B": "At this moment, the chair against the wall was behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_7", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the door open?", + "answer": "No,it isn't.", + "question_chinese": "门是打开的吗？", + "answer_chinese": "不是。", + "evidence": [ + { + "description": "The door on the right is closed.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "The door on the left is also closed.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,it isn't.", + "B": "Yes, it is." + }, + "answer_label": "A", + "answer_text": "No,it isn't.", + "q_id": "scene0774_00-0_7", + "correct_answer": "No,it isn't.", + "original_options": { + "A": "No,it isn't.", + "B": "Yes, it is." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0774_00-0", + "video_length": "00:00:20", + "qid": "scene0774_00-0_8", + "qid_num": 16, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the door open?", + "answer": "No,it isn't.", + "question_chinese": "门是打开的吗？", + "answer_chinese": "不是。", + "evidence": [ + { + "description": "The door on the right is closed.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + }, + { + "description": "The door on the left is also closed.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, it is.", + "B": "No,it isn't." + }, + "answer_label": "B", + "answer_text": "No,it isn't.", + "q_id": "scene0774_00-0_8", + "correct_answer": "No,it isn't.", + "original_options": { + "A": "Yes, it is.", + "B": "No,it isn't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + } +] diff --git a/QAs/scene0776_00-0-mcq.json b/QAs/scene0776_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..7b399847f40306dcb4f000db8d37aa01dab5d15c --- /dev/null +++ b/QAs/scene0776_00-0-mcq.json @@ -0,0 +1,627 @@ +[ + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_5", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the refrigerator?", + "answer": "The refrigerator is in front of you.", + "question_chinese": "从我现在站的位置，冰箱在哪个方向？", + "answer_chinese": "冰箱在你的前方。", + "evidence": [ + { + "description": "The refrigerator is in front of the person.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s on your left.", + "B": "The refrigerator is in front of you.", + "C": "It’s behind you.", + "D": "The refrigerator is to your right, just off to the side from where you’re standing.", + "E": "The refrigerator is slightly behind you, so you’d need to turn around to face it." + }, + "answer_label": "B", + "answer_text": "The refrigerator is in front of you.", + "q_id": "scene0776_00-0_5", + "correct_answer": "The refrigerator is in front of you.", + "original_options": { + "A": "It’s on your left.", + "B": "The refrigerator is in front of you.", + "C": "It’s behind you.", + "D": "The refrigerator is to your right, just off to the side from where you’re standing.", + "E": "The refrigerator is slightly behind you, so you’d need to turn around to face it." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 16, + 16 + ] + }, + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_6", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the refrigerator?", + "answer": "The refrigerator is on your left.", + "question_chinese": "从我现在站的位置，冰箱在哪个方向？", + "answer_chinese": "The refrigerator is behind you on the left.", + "evidence": [ + { + "description": "The refrigerator is behind you on the left.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s in front of you.", + "B": "It’s behind you.", + "C": "The refrigerator is on your left.", + "D": "The refrigerator is on your right, so you’d turn right to face it from your position.", + "E": "The refrigerator is straight ahead of you, directly in your line of sight as you stand." + }, + "answer_label": "C", + "answer_text": "The refrigerator is on your left.", + "q_id": "scene0776_00-0_6", + "correct_answer": "The refrigerator is on your left.", + "original_options": { + "A": "It’s in front of you.", + "B": "It’s behind you.", + "C": "The refrigerator is on your left.", + "D": "The refrigerator is on your right, so you’d turn right to face it from your position.", + "E": "The refrigerator is straight ahead of you, directly in your line of sight as you stand." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 4, + 17, + 16 + ] + }, + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_9", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the sink?", + "answer": "The sink is on your right.", + "question_chinese": "从我现在站的位置，洗手池在哪个方向？", + "answer_chinese": "洗手池在你的右前方。", + "evidence": [ + { + "description": "The washbasin is in front of you on the right.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "The sink is on your right.", + "B": "It’s on your left.", + "C": "It’s behind you.", + "D": "The sink is in front of you, positioned straight ahead from where you are standing.", + "E": "The sink is slightly to your left, so you’d turn left to face it clearly." + }, + "answer_label": "A", + "answer_text": "The sink is on your right.", + "q_id": "scene0776_00-0_9", + "correct_answer": "The sink is on your right.", + "original_options": { + "A": "The sink is on your right.", + "B": "It’s on your left.", + "C": "It’s behind you.", + "D": "The sink is in front of you, positioned straight ahead from where you are standing.", + "E": "The sink is slightly to your left, so you’d turn left to face it clearly." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 4, + 15, + 16 + ] + }, + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_10", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "From where I'm standing, which direction is the sink?", + "answer": "The sink is on your left.", + "question_chinese": "从我现在站的位置，洗手池在哪个方向？", + "answer_chinese": "洗手池在你的左边。", + "evidence": [ + { + "description": "The sink is on the left of the person.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s on your right.", + "B": "It’s in front of you.", + "C": "The sink is behind you, meaning you’d have to turn around to see it from here.", + "D": "The sink is off to your right side, not on your left from where you’re standing.", + "E": "The sink is on your left." + }, + "answer_label": "E", + "answer_text": "The sink is on your left.", + "q_id": "scene0776_00-0_10", + "correct_answer": "The sink is on your left.", + "original_options": { + "A": "It’s on your right.", + "B": "It’s in front of you.", + "C": "The sink is behind you, meaning you’d have to turn around to see it from here.", + "D": "The sink is off to your right side, not on your left from where you’re standing.", + "E": "The sink is on your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 17 + ] + }, + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_7", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the trash can?", + "answer": "Yes, It's very close to you.", + "question_chinese": "我现在的位置能够到垃圾桶吗？", + "answer_chinese": "能够到，它离你很近。", + "evidence": [ + { + "description": "close to the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No, it's a little far away from you.", + "B": "Yes, It's very close to you." + }, + "answer_label": "B", + "answer_text": "Yes, It's very close to you.", + "q_id": "scene0776_00-0_7", + "correct_answer": "Yes, It's very close to you.", + "original_options": { + "A": "No, it's a little far away from you.", + "B": "Yes, It's very close to you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8 + ] + }, + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_8", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "From where I'm standing, can I reach the trash can?", + "answer": "No, it's a little far away from you.", + "question_chinese": "我现在的位置能够到垃圾桶吗？", + "answer_chinese": "不能够到，它理你有点远。", + "evidence": [ + { + "description": "Far away from the person.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, It's very close to you.", + "B": "No, it's a little far away from you." + }, + "answer_label": "B", + "answer_text": "No, it's a little far away from you.", + "q_id": "scene0776_00-0_8", + "correct_answer": "No, it's a little far away from you.", + "original_options": { + "A": "Yes, It's very close to you.", + "B": "No, it's a little far away from you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6 + ] + }, + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_3", + "qid_num": 7, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What rount should I take to leave the kitchen？", + "answer": "You can go left from your current position.", + "question_chinese": "我要离开厨房应该怎么走？", + "answer_chinese": "你可以从当前位置向左走。", + "evidence": [ + { + "description": "There is a corridor on the left that leads out of the kitchen.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Go right from your position.", + "B": "Go straight ahead to exit.", + "C": "Turn right from where you are, then continue forward until you reach the doorway out.", + "D": "You can go left from your current position.", + "E": "Go forward a few steps, then turn right at the opening to leave the kitchen." + }, + "answer_label": "D", + "answer_text": "You can go left from your current position.", + "q_id": "scene0776_00-0_3", + "correct_answer": "You can go left from your current position.", + "original_options": { + "A": "Go right from your position.", + "B": "Go straight ahead to exit.", + "C": "Turn right from where you are, then continue forward until you reach the doorway out.", + "D": "You can go left from your current position.", + "E": "Go forward a few steps, then turn right at the opening to leave the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 15, + 15 + ] + }, + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_4", + "qid_num": 8, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What rount should I take to leave the kitchen？", + "answer": "You can either go left from your current position or go right from it.", + "question_chinese": "我要离开厨房应该怎么走？", + "answer_chinese": "你可以从当前位置向左走，或者从当前位置向右走。", + "evidence": [ + { + "description": "There is a corridor on the left that leads out of the kitchen.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + }, + { + "description": "There is a corridor on the right that leads out of the kitchen.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "You can either go left from your current position or go right from it.", + "B": "Go straight to exit.", + "C": "Turn around to leave.", + "D": "Go straight ahead from where you are, then turn left at the opening to exit the kitchen.", + "E": "Turn around and walk forward toward the doorway behind you to leave the kitchen." + }, + "answer_label": "A", + "answer_text": "You can either go left from your current position or go right from it.", + "q_id": "scene0776_00-0_4", + "correct_answer": "You can either go left from your current position or go right from it.", + "original_options": { + "A": "You can either go left from your current position or go right from it.", + "B": "Go straight to exit.", + "C": "Turn around to leave.", + "D": "Go straight ahead from where you are, then turn left at the opening to exit the kitchen.", + "E": "Turn around and walk forward toward the doorway behind you to leave the kitchen." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 17, + 14 + ] + }, + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_1", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the room?", + "answer": "There are three trash cans in the room.", + "question_chinese": "房间里有几个垃圾桶？", + "answer_chinese": "房间里有三个垃圾桶。", + "evidence": [ + { + "description": "There are three trash cans in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three trash cans in the room.", + "B": "Two trash cans.", + "C": "Four trash cans.", + "D": "There is one trash can in the room near the wall, with no other bins visible.", + "E": "There are five trash cans spread around the room, including one by the doorway." + }, + "answer_label": "A", + "answer_text": "There are three trash cans in the room.", + "q_id": "scene0776_00-0_1", + "correct_answer": "There are three trash cans in the room.", + "original_options": { + "A": "There are three trash cans in the room.", + "B": "Two trash cans.", + "C": "Four trash cans.", + "D": "There is one trash can in the room near the wall, with no other bins visible.", + "E": "There are five trash cans spread around the room, including one by the doorway." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 14 + ] + }, + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_2", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the room?", + "answer": "There are three trash cans in the room.", + "question_chinese": "房间里有几个垃圾桶？", + "answer_chinese": "房间里有三个垃圾桶。", + "evidence": [ + { + "description": "There are three trash cans in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "Two trash cans.", + "B": "Four trash cans.", + "C": "There is one trash can in the room near the wall, with no other bins visible.", + "D": "There are five trash cans spread around the room, including one by the doorway.", + "E": "There are three trash cans in the room." + }, + "answer_label": "E", + "answer_text": "There are three trash cans in the room.", + "q_id": "scene0776_00-0_2", + "correct_answer": "There are three trash cans in the room.", + "original_options": { + "A": "Two trash cans.", + "B": "Four trash cans.", + "C": "There is one trash can in the room near the wall, with no other bins visible.", + "D": "There are five trash cans spread around the room, including one by the doorway.", + "E": "There are three trash cans in the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 3, + 16, + 14 + ] + }, + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_1_T1", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can in the room relative to me?", + "answer": "To your left.", + "question_chinese": "房间里的垃圾桶在我的哪个方位？", + "answer_chinese": "在你的左边", + "evidence": [ + { + "description": "There are three trash cans in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your right.", + "B": "In front of you.", + "C": "The trash can is behind you, slightly off to one side as you face forward.", + "D": "To your left.", + "E": "The trash can is above you, positioned higher than your eye level in the room." + }, + "answer_label": "D", + "answer_text": "To your left.", + "q_id": "scene0776_00-0_1_T1", + "correct_answer": "To your left.", + "original_options": { + "A": "To your right.", + "B": "In front of you.", + "C": "The trash can is behind you, slightly off to one side as you face forward.", + "D": "To your left.", + "E": "The trash can is above you, positioned higher than your eye level in the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 15 + ] + }, + { + "video_uid": "scene0776_00-0", + "video_length": "00:00:20", + "qid": "scene0776_00-0_1_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Ego-centric Position", + "question": "Where is the trash can in the room relative to me?", + "answer": "Behind you.", + "question_chinese": "房间里的垃圾桶在我的哪个方位？", + "answer_chinese": "在你的后面", + "evidence": [ + { + "description": "There are three trash cans in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "To your left.", + "C": "Behind you.", + "D": "The trash can is to your right, close to you as you face into the room.", + "E": "The trash can is above you, positioned higher up rather than on the floor." + }, + "answer_label": "C", + "answer_text": "Behind you.", + "q_id": "scene0776_00-0_1_T2", + "correct_answer": "Behind you.", + "original_options": { + "A": "In front of you.", + "B": "To your left.", + "C": "Behind you.", + "D": "The trash can is to your right, close to you as you face into the room.", + "E": "The trash can is above you, positioned higher up rather than on the floor." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 3, + 16, + 14 + ] + } +] diff --git a/QAs/scene0780_00-0-mcq.json b/QAs/scene0780_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..34a17a6ce482b0e9235c11322e5ce3e7558f35a3 --- /dev/null +++ b/QAs/scene0780_00-0-mcq.json @@ -0,0 +1,450 @@ +[ + { + "video_uid": "scene0780_00-0", + "video_length": "00:00:20", + "qid": "scene0780_00-0_3", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my photo?", + "answer": "Your photo is on the white shelf above the bucket in front of you.", + "question_chinese": "我的照片在哪里？", + "answer_chinese": "你的照片在你面前桶上面的白色架子上。", + "evidence": [ + { + "description": "Your photo is on the white shelf above the bucket in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "On the shelf below the bucket.", + "B": "On the white shelf beside the bucket.", + "C": "Your photo is on the white shelf above the bucket, but it is to your left rather than in front of you.", + "D": "Your photo is on the white shelf in front of you, positioned below the bucket instead of above it.", + "E": "Your photo is on the white shelf above the bucket in front of you." + }, + "answer_label": "E", + "answer_text": "Your photo is on the white shelf above the bucket in front of you.", + "q_id": "scene0780_00-0_3", + "correct_answer": "Your photo is on the white shelf above the bucket in front of you.", + "original_options": { + "A": "On the shelf below the bucket.", + "B": "On the white shelf beside the bucket.", + "C": "Your photo is on the white shelf above the bucket, but it is to your left rather than in front of you.", + "D": "Your photo is on the white shelf in front of you, positioned below the bucket instead of above it.", + "E": "Your photo is on the white shelf above the bucket in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 22, + 19 + ] + }, + { + "video_uid": "scene0780_00-0", + "video_length": "00:00:20", + "qid": "scene0780_00-0_4", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my photo?", + "answer": "Your photo is on the white shelf above your left bucket.", + "question_chinese": "我的照片在哪里？", + "answer_chinese": "你的照片在你左侧桶上面的白色架子上。", + "evidence": [ + { + "description": "Your photo is on the white shelf above the bucket.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "The bucket is on your left.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "On the black countertop.", + "B": "In the drawer below.", + "C": "Your photo is on the red shelf behind your right chair, not on the white shelf near the bucket.", + "D": "Your photo is on the white shelf above your left bucket.", + "E": "Your photo is on the floor beneath your left window, away from the shelving and buckets." + }, + "answer_label": "D", + "answer_text": "Your photo is on the white shelf above your left bucket.", + "q_id": "scene0780_00-0_4", + "correct_answer": "Your photo is on the white shelf above your left bucket.", + "original_options": { + "A": "On the black countertop.", + "B": "In the drawer below.", + "C": "Your photo is on the red shelf behind your right chair, not on the white shelf near the bucket.", + "D": "Your photo is on the white shelf above your left bucket.", + "E": "Your photo is on the floor beneath your left window, away from the shelving and buckets." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 19, + 16 + ] + }, + { + "video_uid": "scene0780_00-0", + "video_length": "00:00:20", + "qid": "scene0780_00-0_10", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "First, you face two buckets, then turn right and walk straight.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先你面对着两个桶，然后向右转并直走。", + "evidence": [ + { + "description": "First, you face two buckets.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you turn right and walk straight.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "First, you face two buckets, then turn right and walk straight.", + "B": "Face two buckets, then turn left.", + "C": "Face two buckets, then walk backward.", + "D": "First, you face two buckets, then turn left and continue walking straight ahead.", + "E": "First, you face two buckets, then turn right and walk straight back the way you came." + }, + "answer_label": "A", + "answer_text": "First, you face two buckets, then turn right and walk straight.", + "q_id": "scene0780_00-0_10", + "correct_answer": "First, you face two buckets, then turn right and walk straight.", + "original_options": { + "A": "First, you face two buckets, then turn right and walk straight.", + "B": "Face two buckets, then turn left.", + "C": "Face two buckets, then walk backward.", + "D": "First, you face two buckets, then turn left and continue walking straight ahead.", + "E": "First, you face two buckets, then turn right and walk straight back the way you came." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 13, + 16 + ] + }, + { + "video_uid": "scene0780_00-0", + "video_length": "00:00:20", + "qid": "scene0780_00-0_9", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "First, you face two buckets, then turn right and walk straight.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先你面对着两个桶，然后向右转并直走。", + "evidence": [ + { + "description": "First, you face two bucket.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you turn right and walk straight.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Face two barrels, then turn right.", + "B": "Face two buckets, then turn left.", + "C": "First, you face two buckets, then turn right and walk backward instead of continuing forward.", + "D": "First, you face two buckets, then turn right and walk straight.", + "E": "First, you face two barrels, then turn left and walk straight ahead from where you started." + }, + "answer_label": "D", + "answer_text": "First, you face two buckets, then turn right and walk straight.", + "q_id": "scene0780_00-0_9", + "correct_answer": "First, you face two buckets, then turn right and walk straight.", + "original_options": { + "A": "Face two barrels, then turn right.", + "B": "Face two buckets, then turn left.", + "C": "First, you face two buckets, then turn right and walk backward instead of continuing forward.", + "D": "First, you face two buckets, then turn right and walk straight.", + "E": "First, you face two barrels, then turn left and walk straight ahead from where you started." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 15, + 16 + ] + }, + { + "video_uid": "scene0780_00-0", + "video_length": "00:00:20", + "qid": "scene0780_00-0_5", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just throw the garbage into the trash can?", + "answer": "No,you can't.", + "question_chinese": "我能直接把垃圾扔到垃圾桶吗？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The trash can is on the ground next to the window.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + }, + { + "description": "You are too far from the trash can.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "scene0780_00-0_5", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0780_00-0", + "video_length": "00:00:20", + "qid": "scene0780_00-0_6", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I just throw the garbage into the trash can?", + "answer": "No,you can't.", + "question_chinese": "我能直接把垃圾扔到垃圾桶吗？", + "answer_chinese": "不能。", + "evidence": [ + { + "description": "The trash can is on the ground next to your left window.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + }, + { + "description": "You are too far from the trash can.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "medium", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "scene0780_00-0_6", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0780_00-0", + "video_length": "00:00:20", + "qid": "scene0780_00-0_7", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many washing machines are there in the room?", + "answer": "There are fourteen washing machines in the room.", + "question_chinese": "房间里有几个洗衣机？", + "answer_chinese": "房间里有14个洗衣机。", + "evidence": [ + { + "description": "There are 14 washing machines in the room.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are ten washing machines.", + "B": "There are twenty-two washing machines.", + "C": "There are eight washing machines in the room, lined up along one side.", + "D": "There are five washing machines in the room, with only a few visible in view.", + "E": "There are fourteen washing machines in the room." + }, + "answer_label": "E", + "answer_text": "There are fourteen washing machines in the room.", + "q_id": "scene0780_00-0_7", + "correct_answer": "There are fourteen washing machines in the room.", + "original_options": { + "A": "There are ten washing machines.", + "B": "There are twenty-two washing machines.", + "C": "There are eight washing machines in the room, lined up along one side.", + "D": "There are five washing machines in the room, with only a few visible in view.", + "E": "There are fourteen washing machines in the room." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 15 + ] + }, + { + "video_uid": "scene0780_00-0", + "video_length": "00:00:20", + "qid": "scene0780_00-0_8", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many washing machines are there in the room?", + "answer": "There are eighteen washing machines in the room.", + "question_chinese": "房间里有几个洗衣机？", + "answer_chinese": "房间里有18个洗衣机。", + "evidence": [ + { + "description": "There are 14 washing machines on the left.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + }, + { + "description": "There are four washing machines on the right.", + "timestamps": { + "start": "00:00:18", + "end": "00:00:19" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are twelve washing machines.", + "B": "There are eighteen washing machines in the room.", + "C": "There are twenty-one washing machines.", + "D": "There are nine washing machines in the room, arranged in a single row.", + "E": "There are five washing machines in the room, so the space looks mostly empty." + }, + "answer_label": "B", + "answer_text": "There are eighteen washing machines in the room.", + "q_id": "scene0780_00-0_8", + "correct_answer": "There are eighteen washing machines in the room.", + "original_options": { + "A": "There are twelve washing machines.", + "B": "There are eighteen washing machines in the room.", + "C": "There are twenty-one washing machines.", + "D": "There are nine washing machines in the room, arranged in a single row.", + "E": "There are five washing machines in the room, so the space looks mostly empty." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 14 + ] + } +] diff --git a/QAs/scene0784_00-0-mcq.json b/QAs/scene0784_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..1dc83d670b773c1981e56de1c1d8851cbbc8e8a9 --- /dev/null +++ b/QAs/scene0784_00-0-mcq.json @@ -0,0 +1,432 @@ +[ + { + "video_uid": "scene0784_00-0", + "video_length": "00:00:20", + "qid": "scene0784_00-0_2", + "qid_num": 1, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "First, you stand on the left side of the bed and walk half a circle around the tail of the bed.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先你站在床的左侧，然后绕着床尾走了半圈。", + "evidence": [ + { + "description": "First, you stand on the left side of the bed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you walk half a circle around the tail of the bed.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "Stand on the right side of the bed.", + "B": "Walk a full circle around the bed.", + "C": "First, you stand on the left side of the bed and walk half a circle around the tail of the bed.", + "D": "First, you stand at the head of the bed and walk half a circle around the head end.", + "E": "First, you stand on the left side of the bed and walk half a circle around the head of the bed." + }, + "answer_label": "C", + "answer_text": "First, you stand on the left side of the bed and walk half a circle around the tail of the bed.", + "q_id": "scene0784_00-0_2", + "correct_answer": "First, you stand on the left side of the bed and walk half a circle around the tail of the bed.", + "original_options": { + "A": "Stand on the right side of the bed.", + "B": "Walk a full circle around the bed.", + "C": "First, you stand on the left side of the bed and walk half a circle around the tail of the bed.", + "D": "First, you stand at the head of the bed and walk half a circle around the head end.", + "E": "First, you stand on the left side of the bed and walk half a circle around the head of the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 7, + 18, + 21 + ] + }, + { + "video_uid": "scene0784_00-0", + "video_length": "00:00:20", + "qid": "scene0784_00-0_1", + "qid_num": 2, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "First, you stand on the left side of the bed and walk half a circle around the tail of the bed.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先你站在床的左侧，然后绕着床尾走了半圈。", + "evidence": [ + { + "description": "First, you stand on the left side of the bed.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you walk half a circle around the tail of the bed.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:20" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "First, you stand on the left side of the bed and walk half a circle around the tail of the bed.", + "B": "Start at the head of the bed.", + "C": "Walk a quarter circle around the bed.", + "D": "First, you stand on the right side of the bed and walk half a circle around the head of the bed.", + "E": "First, you stand on the left side of the bed and walk a quarter circle around the head of the bed." + }, + "answer_label": "A", + "answer_text": "First, you stand on the left side of the bed and walk half a circle around the tail of the bed.", + "q_id": "scene0784_00-0_1", + "correct_answer": "First, you stand on the left side of the bed and walk half a circle around the tail of the bed.", + "original_options": { + "A": "First, you stand on the left side of the bed and walk half a circle around the tail of the bed.", + "B": "Start at the head of the bed.", + "C": "Walk a quarter circle around the bed.", + "D": "First, you stand on the right side of the bed and walk half a circle around the head of the bed.", + "E": "First, you stand on the left side of the bed and walk a quarter circle around the head of the bed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 21, + 21 + ] + }, + { + "video_uid": "scene0784_00-0", + "video_length": "00:00:20", + "qid": "scene0784_00-0_3", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my black coat?", + "answer": "Your black coat is on your chair.", + "question_chinese": "我的黑色外套在哪里？", + "answer_chinese": "你的黑色外套在你的椅子上。", + "evidence": [ + { + "description": "Your black coat is on your chair.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s on your bed.", + "B": "Your black coat is on your chair.", + "C": "It’s hanging on the door.", + "D": "Your black coat is draped over the couch in the living room, not on the chair.", + "E": "Your black coat is on the floor beside the bed, rather than resting on your chair." + }, + "answer_label": "B", + "answer_text": "Your black coat is on your chair.", + "q_id": "scene0784_00-0_3", + "correct_answer": "Your black coat is on your chair.", + "original_options": { + "A": "It’s on your bed.", + "B": "Your black coat is on your chair.", + "C": "It’s hanging on the door.", + "D": "Your black coat is draped over the couch in the living room, not on the chair.", + "E": "Your black coat is on the floor beside the bed, rather than resting on your chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 16, + 16 + ] + }, + { + "video_uid": "scene0784_00-0", + "video_length": "00:00:20", + "qid": "scene0784_00-0_4", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my black coat?", + "answer": "Your black coat is on your chair.", + "question_chinese": "我的黑色外套在哪里？", + "answer_chinese": "你的黑色外套在你的椅子上。", + "evidence": [ + { + "description": "Your black coat is on your chair.", + "timestamps": { + "start": "00:00:03", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s on your bed.", + "B": "It’s on the floor.", + "C": "Your black coat is in the closet, not placed on your chair where you left it.", + "D": "Your black coat is on your chair.", + "E": "Your black coat is on the rack across the room rather than sitting on the chair." + }, + "answer_label": "D", + "answer_text": "Your black coat is on your chair.", + "q_id": "scene0784_00-0_4", + "correct_answer": "Your black coat is on your chair.", + "original_options": { + "A": "It’s on your bed.", + "B": "It’s on the floor.", + "C": "Your black coat is in the closet, not placed on your chair where you left it.", + "D": "Your black coat is on your chair.", + "E": "Your black coat is on the rack across the room rather than sitting on the chair." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 16 + ] + }, + { + "video_uid": "scene0784_00-0", + "video_length": "00:00:20", + "qid": "scene0784_00-0_9", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the chair?", + "answer": "Yes,you can.", + "question_chinese": "我能直接坐在椅子上吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The chair is right in front of you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "scene0784_00-0_9", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0784_00-0", + "video_length": "00:00:20", + "qid": "scene0784_00-0_10", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the chair?", + "answer": "No,you can't.", + "question_chinese": "我能直接坐在椅子上吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The chair is on your right.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "answer_label": "B", + "answer_text": "No,you can't.", + "q_id": "scene0784_00-0_10", + "correct_answer": "No,you can't.", + "original_options": { + "A": "Yes, you can.", + "B": "No,you can't." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0784_00-0", + "video_length": "00:00:20", + "qid": "scene0784_00-0_4_T2", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which was closer to me, the computer or the TV?", + "answer": "The computer is closer to you. The computer is to your left.", + "question_chinese": "电脑和电视机哪个离我更近？", + "answer_chinese": "电脑离你更近。电脑在你的左边。", + "evidence": [ + { + "description": "The TV is on the wall behind you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The computer is on the desk.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "The computer is closer to you.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:11", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The TV is closer to you. The TV is to your left.", + "B": "The computer is closer to you. The computer is to your left." + }, + "answer_label": "B", + "answer_text": "The computer is closer to you. The computer is to your left.", + "q_id": "scene0784_00-0_4_T2", + "correct_answer": "The computer is closer to you. The computer is to your left.", + "original_options": { + "A": "The TV is closer to you. The TV is to your left.", + "B": "The computer is closer to you. The computer is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "scene0784_00-0", + "video_length": "00:00:20", + "qid": "scene0784_00-0_4_T1", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which was closer to me, the computer or the TV?", + "answer": "The TV is closer to you. The TV is directly behind you.", + "question_chinese": "电脑和电视机哪个离我更近？", + "answer_chinese": "电视机离你更近。电视机在你的正后方。", + "evidence": [ + { + "description": "The TV is on the wall behind you.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The computer is on the desk.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "The TV is closer to you.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The TV is closer to you. The TV is directly behind you.", + "B": "The computer is closer to you. The computer is directly behind you." + }, + "answer_label": "A", + "answer_text": "The TV is closer to you. The TV is directly behind you.", + "q_id": "scene0784_00-0_4_T1", + "correct_answer": "The TV is closer to you. The TV is directly behind you.", + "original_options": { + "A": "The TV is closer to you. The TV is directly behind you.", + "B": "The computer is closer to you. The computer is directly behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + } +] diff --git a/QAs/scene0785_00-0-mcq.json b/QAs/scene0785_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e35248c40c2272bbd59e23013d109a3a0760522f --- /dev/null +++ b/QAs/scene0785_00-0-mcq.json @@ -0,0 +1,436 @@ +[ + { + "video_uid": "scene0785_00-0", + "video_length": "00:00:20", + "qid": "scene0785_00-0_7", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my backpack.", + "answer": "It is on the ground in front of you.", + "question_chinese": "我的书包在哪里。", + "answer_chinese": "在你面前的地上。", + "evidence": [ + { + "description": "The schoolbag is on the ground in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:01", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It is on the ground in front of you.", + "B": "It’s under the table.", + "C": "It’s on the chair to your left.", + "D": "It is on the floor beside the couch, not on the ground in front of you.", + "E": "It is behind the bookshelf in the corner, rather than lying on the ground in front of you." + }, + "answer_label": "A", + "answer_text": "It is on the ground in front of you.", + "q_id": "scene0785_00-0_7", + "correct_answer": "It is on the ground in front of you.", + "original_options": { + "A": "It is on the ground in front of you.", + "B": "It’s under the table.", + "C": "It’s on the chair to your left.", + "D": "It is on the floor beside the couch, not on the ground in front of you.", + "E": "It is behind the bookshelf in the corner, rather than lying on the ground in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 8, + 16, + 18 + ] + }, + { + "video_uid": "scene0785_00-0", + "video_length": "00:00:20", + "qid": "scene0785_00-0_2_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my backpack?", + "answer": "It was on the floor in front of you.", + "question_chinese": "我的书包在哪里？", + "answer_chinese": "在你前面的地上。", + "evidence": [ + { + "description": "The schoolbag is on the ground in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was behind you.", + "B": "It was to your left.", + "C": "It was on the floor to your right, not on the floor in front of you.", + "D": "It was on the floor in front of you.", + "E": "It was on the floor behind you rather than being placed on the floor in front of you." + }, + "answer_label": "D", + "answer_text": "It was on the floor in front of you.", + "q_id": "scene0785_00-0_2_T1", + "correct_answer": "It was on the floor in front of you.", + "original_options": { + "A": "It was behind you.", + "B": "It was to your left.", + "C": "It was on the floor to your right, not on the floor in front of you.", + "D": "It was on the floor in front of you.", + "E": "It was on the floor behind you rather than being placed on the floor in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 16, + 18 + ] + }, + { + "video_uid": "scene0785_00-0", + "video_length": "00:00:20", + "qid": "scene0785_00-0_8", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my backpack.", + "answer": "Your backpack is on the ground next to the dining table.", + "question_chinese": "我的书包在哪里。", + "answer_chinese": "在你面前的地上。", + "evidence": [ + { + "description": "Your backpack is on the ground next to the dining table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It's on the chair by the table.", + "B": "Your backpack is on the ground next to the dining table.", + "C": "It's behind the dining table.", + "D": "Your backpack is on the floor beside the kitchen counter, away from the dining table.", + "E": "Your backpack is on a shelf above the dining table, not resting on the ground." + }, + "answer_label": "B", + "answer_text": "Your backpack is on the ground next to the dining table.", + "q_id": "scene0785_00-0_8", + "correct_answer": "Your backpack is on the ground next to the dining table.", + "original_options": { + "A": "It's on the chair by the table.", + "B": "Your backpack is on the ground next to the dining table.", + "C": "It's behind the dining table.", + "D": "Your backpack is on the floor beside the kitchen counter, away from the dining table.", + "E": "Your backpack is on a shelf above the dining table, not resting on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 15, + 15 + ] + }, + { + "video_uid": "scene0785_00-0", + "video_length": "00:00:20", + "qid": "scene0785_00-0_2_T2", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my backpack?", + "answer": "It was on the floor to your left.", + "question_chinese": "我的书包在哪里？", + "answer_chinese": "在你左边的地上。", + "evidence": [ + { + "description": "Your backpack is on the ground next to the dining table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "I turned 90° to the right", + "timestamps": { + "start": "00:00:04", + "end": "00:00:11" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It was in front of you.", + "B": "It was to your right.", + "C": "It was on the floor to your left.", + "D": "It was on the floor behind you, rather than on the left side.", + "E": "It was on the floor in front of you, not positioned to your left." + }, + "answer_label": "C", + "answer_text": "It was on the floor to your left.", + "q_id": "scene0785_00-0_2_T2", + "correct_answer": "It was on the floor to your left.", + "original_options": { + "A": "It was in front of you.", + "B": "It was to your right.", + "C": "It was on the floor to your left.", + "D": "It was on the floor behind you, rather than on the left side.", + "E": "It was on the floor in front of you, not positioned to your left." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 13, + 14 + ] + }, + { + "video_uid": "scene0785_00-0", + "video_length": "00:00:20", + "qid": "scene0785_00-0_9", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the black chair next to the computer?", + "answer": "Yes,you can.", + "question_chinese": "我能直接坐在电脑旁边的黑色椅子上吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The black chair is right in front of you.", + "timestamps": { + "start": "00:00:07", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "scene0785_00-0_9", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No, you can't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0785_00-0", + "video_length": "00:00:20", + "qid": "scene0785_00-0_10", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I sit directly on the black chair next to the computer?", + "answer": "No,you can't.", + "question_chinese": "我能直接坐在电脑旁边的黑色椅子上吗？", + "answer_chinese": "不可以。", + "evidence": [ + { + "description": "The black chair is next to the computer.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "You're at the dining table.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + }, + { + "description": "You are too far from the black chair.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:13", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "scene0785_00-0_10", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes, you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0785_00-0", + "video_length": "00:00:20", + "qid": "scene0785_00-0_2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "There are three chairs in the room.", + "question_chinese": "房间里有几把椅子？", + "answer_chinese": "房间里有三把椅子。", + "evidence": [ + { + "description": "There are three chairs next to the dining table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two chairs.", + "B": "There are four chairs.", + "C": "There are three chairs in the room.", + "D": "There is one chair in the room, with no other chairs visible nearby.", + "E": "There are five chairs in the room, including several arranged around the main area." + }, + "answer_label": "C", + "answer_text": "There are three chairs in the room.", + "q_id": "scene0785_00-0_2", + "correct_answer": "There are three chairs in the room.", + "original_options": { + "A": "There are two chairs.", + "B": "There are four chairs.", + "C": "There are three chairs in the room.", + "D": "There is one chair in the room, with no other chairs visible nearby.", + "E": "There are five chairs in the room, including several arranged around the main area." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 13, + 14 + ] + }, + { + "video_uid": "scene0785_00-0", + "video_length": "00:00:20", + "qid": "scene0785_00-0_1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many chairs are there in the room?", + "answer": "There are four chairs in the room.", + "question_chinese": "房间里有几把椅子？", + "answer_chinese": "房间里有四把椅子。", + "evidence": [ + { + "description": "There are three chairs next to the dining table.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "There is a black chair next to the computer.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are three chairs.", + "B": "There are five chairs.", + "C": "There are six chairs in the room, spread across different spots in the space.", + "D": "There are four chairs in the room.", + "E": "There are seven chairs in the room, more than a small set grouped together." + }, + "answer_label": "D", + "answer_text": "There are four chairs in the room.", + "q_id": "scene0785_00-0_1", + "correct_answer": "There are four chairs in the room.", + "original_options": { + "A": "There are three chairs.", + "B": "There are five chairs.", + "C": "There are six chairs in the room, spread across different spots in the space.", + "D": "There are four chairs in the room.", + "E": "There are seven chairs in the room, more than a small set grouped together." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + } +] diff --git a/QAs/scene0787_00-0-mcq.json b/QAs/scene0787_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..9aa5a2a09b4e279c2b66dd8aee2b7e190ebddac0 --- /dev/null +++ b/QAs/scene0787_00-0-mcq.json @@ -0,0 +1,777 @@ +[ + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_3", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of orange laundry detergent are there on the shelf?", + "answer": "There are three bottles of orange laundry detergent on the shelf.", + "question_chinese": "在架子上还有几瓶橙色洗衣液？", + "answer_chinese": "在架子上还有三瓶橙色洗衣液。", + "evidence": [ + { + "description": "There are three bottles of orange laundry detergent on the shelf.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:04", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are four bottles.", + "B": "There are five bottles.", + "C": "There are six bottles of orange laundry detergent on the shelf, filling most of the space.", + "D": "There are seven bottles of orange laundry detergent on the shelf, lined up closely together.", + "E": "There are three bottles of orange laundry detergent on the shelf." + }, + "answer_label": "E", + "answer_text": "There are three bottles of orange laundry detergent on the shelf.", + "q_id": "scene0787_00-0_3", + "correct_answer": "There are three bottles of orange laundry detergent on the shelf.", + "original_options": { + "A": "There are four bottles.", + "B": "There are five bottles.", + "C": "There are six bottles of orange laundry detergent on the shelf, filling most of the space.", + "D": "There are seven bottles of orange laundry detergent on the shelf, lined up closely together.", + "E": "There are three bottles of orange laundry detergent on the shelf." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 15 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_4", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many bottles of orange laundry detergent are there on the shelf?", + "answer": "There are three bottles of orange laundry detergent on the shelf.", + "question_chinese": "在架子上还有几瓶橙色洗衣液？", + "answer_chinese": "在架子上还有三瓶橙色洗衣液。", + "evidence": [ + { + "description": "There are three bottles of orange laundry detergent on the shelf.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are two bottles.", + "B": "There are three bottles of orange laundry detergent on the shelf.", + "C": "There is one bottle.", + "D": "There are four bottles of orange laundry detergent on the shelf, more than three in total.", + "E": "There are five bottles of orange laundry detergent on the shelf, forming a longer row." + }, + "answer_label": "B", + "answer_text": "There are three bottles of orange laundry detergent on the shelf.", + "q_id": "scene0787_00-0_4", + "correct_answer": "There are three bottles of orange laundry detergent on the shelf.", + "original_options": { + "A": "There are two bottles.", + "B": "There are three bottles of orange laundry detergent on the shelf.", + "C": "There is one bottle.", + "D": "There are four bottles of orange laundry detergent on the shelf, more than three in total.", + "E": "There are five bottles of orange laundry detergent on the shelf, forming a longer row." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 16, + 15 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_9", + "qid_num": 3, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are my canvas shoes?", + "answer": "Your canvas shoes are on the ground.", + "question_chinese": "我的帆布鞋在哪里？", + "answer_chinese": "你的帆布鞋在地上。", + "evidence": [ + { + "description": "Your canvas shoes are on the ground.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "Your canvas shoes are on the ground.", + "B": "They’re in the closet.", + "C": "They’re under the bed.", + "D": "Your canvas shoes are tucked beside the door, not lying out on the floor.", + "E": "Your canvas shoes are in the car, so you won’t find them anywhere on the ground." + }, + "answer_label": "A", + "answer_text": "Your canvas shoes are on the ground.", + "q_id": "scene0787_00-0_9", + "correct_answer": "Your canvas shoes are on the ground.", + "original_options": { + "A": "Your canvas shoes are on the ground.", + "B": "They’re in the closet.", + "C": "They’re under the bed.", + "D": "Your canvas shoes are tucked beside the door, not lying out on the floor.", + "E": "Your canvas shoes are in the car, so you won’t find them anywhere on the ground." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 14, + 17 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_10", + "qid_num": 4, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where are my canvas shoes?", + "answer": "Your canvas shoes are on the ground.", + "question_chinese": "我的帆布鞋在哪里？", + "answer_chinese": "你的帆布鞋在地上。", + "evidence": [ + { + "description": "Your canvas shoes are on the ground.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:18", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "They’re near the door.", + "B": "They’re under the table.", + "C": "Your canvas shoes are on the ground.", + "D": "Your canvas shoes are beside the couch, not down on the ground where you left them.", + "E": "Your canvas shoes are next to the bed, so they aren’t on the ground in front of you." + }, + "answer_label": "C", + "answer_text": "Your canvas shoes are on the ground.", + "q_id": "scene0787_00-0_10", + "correct_answer": "Your canvas shoes are on the ground.", + "original_options": { + "A": "They’re near the door.", + "B": "They’re under the table.", + "C": "Your canvas shoes are on the ground.", + "D": "Your canvas shoes are beside the couch, not down on the ground where you left them.", + "E": "Your canvas shoes are next to the bed, so they aren’t on the ground in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 19 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_7", + "qid_num": 5, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my white laundry detergent?", + "answer": "It is on the rack above the washing machine.", + "question_chinese": "我的白色洗衣液在哪里？", + "answer_chinese": "在洗衣机上面的架子上。", + "evidence": [ + { + "description": "The white laundry detergent is on the rack above the washing machine.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s on the counter.", + "B": "It is on the rack above the washing machine.", + "C": "It’s in the cabinet.", + "D": "Your white laundry detergent is on the top shelf near the laundry baskets, not above the washer.", + "E": "Your white laundry detergent is on the floor by the laundry baskets instead of on the rack over the machine." + }, + "answer_label": "B", + "answer_text": "It is on the rack above the washing machine.", + "q_id": "scene0787_00-0_7", + "correct_answer": "It is on the rack above the washing machine.", + "original_options": { + "A": "It’s on the counter.", + "B": "It is on the rack above the washing machine.", + "C": "It’s in the cabinet.", + "D": "Your white laundry detergent is on the top shelf near the laundry baskets, not above the washer.", + "E": "Your white laundry detergent is on the floor by the laundry baskets instead of on the rack over the machine." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 20 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_8", + "qid_num": 6, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is my white laundry detergent?", + "answer": "It is on the rack above the washing machine.", + "question_chinese": "我的白色洗衣液在哪里？", + "answer_chinese": "在洗衣机上面的架子上。", + "evidence": [ + { + "description": "The white laundry detergent is on the rack above the washing machine.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:09" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "easy", + "qtype": "OTHER", + "options": { + "A": "It’s on the shelf.", + "B": "It’s below the machine.", + "C": "Your white laundry detergent is on the counter beside the laundry basket, not up on the rack.", + "D": "Your white laundry detergent is on the rack beside the dryer rather than above the washing machine.", + "E": "It is on the rack above the washing machine." + }, + "answer_label": "E", + "answer_text": "It is on the rack above the washing machine.", + "q_id": "scene0787_00-0_8", + "correct_answer": "It is on the rack above the washing machine.", + "original_options": { + "A": "It’s on the shelf.", + "B": "It’s below the machine.", + "C": "Your white laundry detergent is on the counter beside the laundry basket, not up on the rack.", + "D": "Your white laundry detergent is on the rack beside the dryer rather than above the washing machine.", + "E": "It is on the rack above the washing machine." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 17, + 17 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_5", + "qid_num": 7, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly get the fire extinguisher?", + "answer": "Yes,you can.", + "question_chinese": "我能直接拿到灭火器吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The fire extinguisher is on the ground in front of you.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,you can.", + "B": "No, you cannot." + }, + "answer_label": "A", + "answer_text": "Yes,you can.", + "q_id": "scene0787_00-0_5", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "Yes,you can.", + "B": "No, you cannot." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_2_T2", + "qid_num": 8, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the fire extinguisher?", + "answer": "You can reach the fire extinguisher directly; it is to your left.", + "question_chinese": "我能直接拿到灭火器吗？", + "answer_chinese": "你能直接拿到灭火器，灭火器在你的左边。", + "evidence": [ + { + "description": "You can get the fire extinguisher, which is on your left.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the fire extinguisher directly; it is to your left.", + "B": "You can't reach the fire extinguisher directly; it is to your left." + }, + "answer_label": "A", + "answer_text": "You can reach the fire extinguisher directly; it is to your left.", + "q_id": "scene0787_00-0_2_T2", + "correct_answer": "You can reach the fire extinguisher directly; it is to your left.", + "original_options": { + "A": "You can reach the fire extinguisher directly; it is to your left.", + "B": "You can't reach the fire extinguisher directly; it is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 12 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_2_T1", + "qid_num": 9, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I directly reach the fire extinguisher?", + "answer": "You can't reach the fire extinguisher directly; it is behind you.", + "question_chinese": "我能直接拿到灭火器吗？", + "answer_chinese": "你不能直接拿到灭火器，灭火器在你的后面。", + "evidence": [ + { + "description": "The fire extinguisher is on the ground in front of you.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + }, + { + "description": "You turned around and the fire extinguisher became behind you.", + "timestamps": { + "start": "00:00:14", + "end": "00:00:15" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You can reach the fire extinguisher directly; it is behind you.", + "B": "You can't reach the fire extinguisher directly; it is behind you." + }, + "answer_label": "B", + "answer_text": "You can't reach the fire extinguisher directly; it is behind you.", + "q_id": "scene0787_00-0_2_T1", + "correct_answer": "You can't reach the fire extinguisher directly; it is behind you.", + "original_options": { + "A": "You can reach the fire extinguisher directly; it is behind you.", + "B": "You can't reach the fire extinguisher directly; it is behind you." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 11 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_4_T1", + "qid_num": 10, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the canvas bag on the wooden door handle relative to me?", + "answer": "The canvas bag on the wooden door handle is directly in front of you.", + "question_chinese": "木门把手上的帆布包在相对于我的哪里？", + "answer_chinese": "木门把手上的帆布包在相对于你的正前方。", + "evidence": [ + { + "description": "The canvas on the wooden door handle is wrapped in front of you.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It’s directly behind you.", + "B": "It’s directly to your left.", + "C": "The canvas bag on the wooden door handle is directly to your right rather than straight ahead.", + "D": "The canvas bag on the wooden door handle is directly above you, not positioned in front of you.", + "E": "The canvas bag on the wooden door handle is directly in front of you." + }, + "answer_label": "E", + "answer_text": "The canvas bag on the wooden door handle is directly in front of you.", + "q_id": "scene0787_00-0_4_T1", + "correct_answer": "The canvas bag on the wooden door handle is directly in front of you.", + "original_options": { + "A": "It’s directly behind you.", + "B": "It’s directly to your left.", + "C": "The canvas bag on the wooden door handle is directly to your right rather than straight ahead.", + "D": "The canvas bag on the wooden door handle is directly above you, not positioned in front of you.", + "E": "The canvas bag on the wooden door handle is directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 18 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_4_T2", + "qid_num": 11, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the canvas bag on the wooden door handle relative to me?", + "answer": "The canvas bag on the wooden door handle is to your left.", + "question_chinese": "木门把手上的帆布包在相对于我的哪里？", + "answer_chinese": "木门把手上的帆布包在相对于你的左边。", + "evidence": [ + { + "description": "The canvas on the wooden door handle is wrapped on the left relative to you.", + "timestamps": { + "start": "00:00:13", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is to your right.", + "B": "The canvas bag on the wooden door handle is to your left.", + "C": "It is in front of you.", + "D": "The canvas bag on the wooden door handle is behind you, not off to your left side.", + "E": "The canvas bag on the wooden door handle is above you, higher than your usual line of sight." + }, + "answer_label": "B", + "answer_text": "The canvas bag on the wooden door handle is to your left.", + "q_id": "scene0787_00-0_4_T2", + "correct_answer": "The canvas bag on the wooden door handle is to your left.", + "original_options": { + "A": "It is to your right.", + "B": "The canvas bag on the wooden door handle is to your left.", + "C": "It is in front of you.", + "D": "The canvas bag on the wooden door handle is behind you, not off to your left side.", + "E": "The canvas bag on the wooden door handle is above you, higher than your usual line of sight." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 6, + 17, + 18 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_3_T2", + "qid_num": 12, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the white washing machine relative to me?", + "answer": "The white washing machine is directly in front of you.", + "question_chinese": "白色洗衣机在相对于我的哪里？", + "answer_chinese": "白色洗衣机在相对于你的正前方。", + "evidence": [ + { + "description": "The white washing machine is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:04" + } + } + ], + "question_timestamps": "00:00:06", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The white washing machine is to your right, not positioned directly in front of you.", + "D": "The white washing machine is directly in front of you.", + "E": "The white washing machine is above you, not at eye level in front of you." + }, + "answer_label": "D", + "answer_text": "The white washing machine is directly in front of you.", + "q_id": "scene0787_00-0_3_T2", + "correct_answer": "The white washing machine is directly in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "It is to your left.", + "C": "The white washing machine is to your right, not positioned directly in front of you.", + "D": "The white washing machine is directly in front of you.", + "E": "The white washing machine is above you, not at eye level in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 15, + 15 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_3_T1", + "qid_num": 13, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the white washing machine relative to me?", + "answer": "The white washing machine is behind you.", + "question_chinese": "白色洗衣机在相对于我的哪里？", + "answer_chinese": "白色洗衣机在相对于你的后面。", + "evidence": [ + { + "description": "The white washing machine is behind you.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "The white washing machine is behind you.", + "B": "It is in front of you.", + "C": "It is to your left.", + "D": "The white washing machine is to your right, rather than being located behind you.", + "E": "The white washing machine is above you, instead of positioned behind your back." + }, + "answer_label": "A", + "answer_text": "The white washing machine is behind you.", + "q_id": "scene0787_00-0_3_T1", + "correct_answer": "The white washing machine is behind you.", + "original_options": { + "A": "The white washing machine is behind you.", + "B": "It is in front of you.", + "C": "It is to your left.", + "D": "The white washing machine is to your right, rather than being located behind you.", + "E": "The white washing machine is above you, instead of positioned behind your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 14, + 13 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_1_T1", + "qid_num": 14, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the shelf that holds the orange laundry detergent relative to me?", + "answer": "The shelf with the orange laundry detergent is directly in front of you.", + "question_chinese": "装橙色洗衣液的架子在相对于我的哪里？", + "answer_chinese": "装橙色洗衣液的架子在你的正前方 。", + "evidence": [ + { + "description": "The shelf for orange laundry detergent is in front of you.", + "timestamps": { + "start": "00:00:02", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "It is behind you.", + "B": "The shelf with the orange laundry detergent is directly in front of you.", + "C": "It is to your left.", + "D": "The shelf with the orange laundry detergent is to your right, not directly in front of you.", + "E": "The shelf with the orange laundry detergent is above you, rather than straight ahead at your height." + }, + "answer_label": "B", + "answer_text": "The shelf with the orange laundry detergent is directly in front of you.", + "q_id": "scene0787_00-0_1_T1", + "correct_answer": "The shelf with the orange laundry detergent is directly in front of you.", + "original_options": { + "A": "It is behind you.", + "B": "The shelf with the orange laundry detergent is directly in front of you.", + "C": "It is to your left.", + "D": "The shelf with the orange laundry detergent is to your right, not directly in front of you.", + "E": "The shelf with the orange laundry detergent is above you, rather than straight ahead at your height." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 5, + 17, + 17 + ] + }, + { + "video_uid": "scene0787_00-0", + "video_length": "00:00:20", + "qid": "scene0787_00-0_1_T2", + "qid_num": 15, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Where is the shelf that holds the orange laundry detergent relative to me?", + "answer": "The shelf with the orange laundry detergent is behind you.", + "question_chinese": "装橙色洗衣液的架子在相对于我的哪里？", + "answer_chinese": "装橙色洗衣液的架子在你的后面。", + "evidence": [ + { + "description": "The shelf for orange laundry detergent is behind you.", + "timestamps": { + "start": "00:00:16", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "The shelf with the orange laundry detergent is behind you.", + "B": "It is in front of you.", + "C": "It is to your right.", + "D": "The shelf with the orange laundry detergent is to your left, instead of being positioned behind you.", + "E": "The shelf with the orange laundry detergent is above you, rather than located behind your back." + }, + "answer_label": "A", + "answer_text": "The shelf with the orange laundry detergent is behind you.", + "q_id": "scene0787_00-0_1_T2", + "correct_answer": "The shelf with the orange laundry detergent is behind you.", + "original_options": { + "A": "The shelf with the orange laundry detergent is behind you.", + "B": "It is in front of you.", + "C": "It is to your right.", + "D": "The shelf with the orange laundry detergent is to your left, instead of being positioned behind you.", + "E": "The shelf with the orange laundry detergent is above you, rather than located behind your back." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 5, + 17, + 16 + ] + } +] diff --git a/QAs/scene0788_00-0-mcq.json b/QAs/scene0788_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..dd12fde170611005b74e0b698239562a504dc009 --- /dev/null +++ b/QAs/scene0788_00-0-mcq.json @@ -0,0 +1,732 @@ +[ + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many treadmills are there in the gym?", + "answer": "There are three treadmills in the gym.", + "question_chinese": "健身房里有几个跑步机？", + "answer_chinese": "健身房里有三个跑步机。", + "evidence": [ + { + "description": "There are three treadmills in the gym.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three treadmills in the gym.", + "B": "There are two treadmills.", + "C": "There are four treadmills.", + "D": "There are five treadmills in the gym, lined up in a longer row than three.", + "E": "There are six treadmills in the gym, making the set larger than three machines." + }, + "answer_label": "A", + "answer_text": "There are three treadmills in the gym.", + "q_id": "scene0788_00-0_1", + "correct_answer": "There are three treadmills in the gym.", + "original_options": { + "A": "There are three treadmills in the gym.", + "B": "There are two treadmills.", + "C": "There are four treadmills.", + "D": "There are five treadmills in the gym, lined up in a longer row than three.", + "E": "There are six treadmills in the gym, making the set larger than three machines." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 15, + 14 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many treadmills are there in the gym?", + "answer": "There are three treadmills in the gym.", + "question_chinese": "健身房里有几个跑步机？", + "answer_chinese": "健身房里有三个跑步机。", + "evidence": [ + { + "description": "There are three treadmills in the gym.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are four treadmills.", + "B": "There are five treadmills.", + "C": "There are three treadmills in the gym.", + "D": "There are six treadmills in the gym, which is more than the three shown.", + "E": "There are seven treadmills in the gym, exceeding the count of three by several." + }, + "answer_label": "C", + "answer_text": "There are three treadmills in the gym.", + "q_id": "scene0788_00-0_2", + "correct_answer": "There are three treadmills in the gym.", + "original_options": { + "A": "There are four treadmills.", + "B": "There are five treadmills.", + "C": "There are three treadmills in the gym.", + "D": "There are six treadmills in the gym, which is more than the three shown.", + "E": "There are seven treadmills in the gym, exceeding the count of three by several." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 14, + 14 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_4_T1", + "qid_num": 3, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Which side of me are the dumbbells on?", + "answer": "The dumbbells are to your right.", + "question_chinese": "哑铃在我的哪边？", + "answer_chinese": "哑铃在你的右边", + "evidence": [ + { + "description": "The dumbbell is on your right.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The dumbbells are to your right.", + "B": "The dumbbells are on your left-hand side." + }, + "answer_label": "A", + "answer_text": "The dumbbells are to your right.", + "q_id": "scene0788_00-0_4_T1", + "correct_answer": "The dumbbells are to your right.", + "original_options": { + "A": "The dumbbells are to your right.", + "B": "The dumbbells are on your left-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_4_T2", + "qid_num": 4, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Which side of me are the dumbbells on?", + "answer": "The dumbbells are on your left-hand side.", + "question_chinese": "哑铃在我的哪边？", + "answer_chinese": "哑铃在你的左手边。", + "evidence": [ + { + "description": "The dumbbell is on your left hand side.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:14" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "The dumbbells are on your right-hand side.", + "B": "The dumbbells are on your left-hand side." + }, + "answer_label": "B", + "answer_text": "The dumbbells are on your left-hand side.", + "q_id": "scene0788_00-0_4_T2", + "correct_answer": "The dumbbells are on your left-hand side.", + "original_options": { + "A": "The dumbbells are on your right-hand side.", + "B": "The dumbbells are on your left-hand side." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_3_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I use the rowing machine?", + "answer": "You can use the rowing machine; it is to your left.", + "question_chinese": "我可以使用划船机吗？", + "answer_chinese": "你可以使用划船机，划船机在你左边。", + "evidence": [ + { + "description": "The rowing machine is very close to you. It's right in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can’t use it.", + "B": "It’s to your right.", + "C": "You can use the rowing machine; it’s directly in front of you instead of to your left.", + "D": "You can’t use the rowing machine right now, even though it’s positioned to your left.", + "E": "You can use the rowing machine; it is to your left." + }, + "answer_label": "E", + "answer_text": "You can use the rowing machine; it is to your left.", + "q_id": "scene0788_00-0_3_T1", + "correct_answer": "You can use the rowing machine; it is to your left.", + "original_options": { + "A": "You can’t use it.", + "B": "It’s to your right.", + "C": "You can use the rowing machine; it’s directly in front of you instead of to your left.", + "D": "You can’t use the rowing machine right now, even though it’s positioned to your left.", + "E": "You can use the rowing machine; it is to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 18, + 17 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_3_T2", + "qid_num": 6, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "Can I use the rowing machine?", + "answer": "You can’t use the rowing machine; it is behind you.", + "question_chinese": "我可以使用划船机吗？", + "answer_chinese": "你不可以使用划船机，划船机在你身后", + "evidence": [ + { + "description": "This rowing machine is too far from you.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You can’t use the rowing machine; it is behind you.", + "B": "You can use it.", + "C": "It’s in front of you.", + "D": "You can’t use the rowing machine; it’s to your left rather than behind you.", + "E": "You can use the rowing machine because it is in front of you and within reach." + }, + "answer_label": "A", + "answer_text": "You can’t use the rowing machine; it is behind you.", + "q_id": "scene0788_00-0_3_T2", + "correct_answer": "You can’t use the rowing machine; it is behind you.", + "original_options": { + "A": "You can’t use the rowing machine; it is behind you.", + "B": "You can use it.", + "C": "It’s in front of you.", + "D": "You can’t use the rowing machine; it’s to your left rather than behind you.", + "E": "You can use the rowing machine because it is in front of you and within reach." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 6, + 16, + 16 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_1_T2", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Did I see a TV in the gym?", + "answer": "In the gym, you didn’t see a TV; the TV is above the dumbbells, and the dumbbells are to your front-left.", + "question_chinese": "我在健身房里有看到电视吗？", + "answer_chinese": "你在健身房里没有看到电视，电视机在哑铃上方，哑铃在你的左前方。", + "evidence": [ + { + "description": "You don't see any TV in the gym.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:07" + } + } + ], + "question_timestamps": "00:00:09", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "Yes, you saw a TV.", + "B": "No, you saw a TV.", + "C": "In the gym, you didn’t see a TV; the TV is above the dumbbells, and the dumbbells are to your back-right.", + "D": "In the gym, you didn’t see a TV; the TV is above the dumbbells, and the dumbbells are to your front-left.", + "E": "In the gym, you did see a TV; it was above the dumbbells, and the dumbbells were to your front-left." + }, + "answer_label": "D", + "answer_text": "In the gym, you didn’t see a TV; the TV is above the dumbbells, and the dumbbells are to your front-left.", + "q_id": "scene0788_00-0_1_T2", + "correct_answer": "In the gym, you didn’t see a TV; the TV is above the dumbbells, and the dumbbells are to your front-left.", + "original_options": { + "A": "Yes, you saw a TV.", + "B": "No, you saw a TV.", + "C": "In the gym, you didn’t see a TV; the TV is above the dumbbells, and the dumbbells are to your back-right.", + "D": "In the gym, you didn’t see a TV; the TV is above the dumbbells, and the dumbbells are to your front-left.", + "E": "In the gym, you did see a TV; it was above the dumbbells, and the dumbbells were to your front-left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 22, + 20 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_1_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "Did I see a TV in the gym?", + "answer": "In the gym, you saw a TV; the TV is to your left.", + "question_chinese": "我在健身房里有看到电视吗？", + "answer_chinese": "你在健身房里有看到电视，电视机在你的左边。", + "evidence": [ + { + "description": "There are three treadmills in the gym.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "No, you didn’t see one.", + "B": "Yes, it was to your right.", + "C": "In the gym, you saw a TV; the TV is to your left.", + "D": "In the gym, you saw a TV; it was in front of you instead of to your left.", + "E": "In the gym, you didn’t see a TV; there wasn’t one visible from where you were standing." + }, + "answer_label": "C", + "answer_text": "In the gym, you saw a TV; the TV is to your left.", + "q_id": "scene0788_00-0_1_T1", + "correct_answer": "In the gym, you saw a TV; the TV is to your left.", + "original_options": { + "A": "No, you didn’t see one.", + "B": "Yes, it was to your right.", + "C": "In the gym, you saw a TV; the TV is to your left.", + "D": "In the gym, you saw a TV; it was in front of you instead of to your left.", + "E": "In the gym, you didn’t see a TV; there wasn’t one visible from where you were standing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 6, + 18, + 19 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_5", + "qid_num": 9, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "First, you face the rowing machine, then you turn back and walk straight through the treadmill.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先，你面对着划船机，然后你向后转并直行经过跑步机。", + "evidence": [ + { + "description": "First, you face the rowing machine.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you turn back.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + }, + { + "description": "Then you walk straight through the treadmill.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "You face the rowing machine, then turn right.", + "B": "You face the rowing machine, then turn left.", + "C": "First, you face the rowing machine, then you turn around and walk straight toward the wall instead of the treadmill.", + "D": "First, you face the rowing machine, then you turn back and walk straight past the treadmill rather than through it.", + "E": "First, you face the rowing machine, then you turn back and walk straight through the treadmill." + }, + "answer_label": "E", + "answer_text": "First, you face the rowing machine, then you turn back and walk straight through the treadmill.", + "q_id": "scene0788_00-0_5", + "correct_answer": "First, you face the rowing machine, then you turn back and walk straight through the treadmill.", + "original_options": { + "A": "You face the rowing machine, then turn right.", + "B": "You face the rowing machine, then turn left.", + "C": "First, you face the rowing machine, then you turn around and walk straight toward the wall instead of the treadmill.", + "D": "First, you face the rowing machine, then you turn back and walk straight past the treadmill rather than through it.", + "E": "First, you face the rowing machine, then you turn back and walk straight through the treadmill." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 20, + 20 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_6", + "qid_num": 10, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "First, you face the rowing machine, then you turn back and walk straight through the treadmill.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先，你面对着划船机，然后你向后转并直行经过跑步机。", + "evidence": [ + { + "description": "First, you face the rowing machine.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you turn back.", + "timestamps": { + "start": "00:00:01", + "end": "00:00:05" + } + }, + { + "description": "Then you walk straight through the treadmill.", + "timestamps": { + "start": "00:00:06", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "hard", + "qtype": "OTHER", + "options": { + "A": "Face the treadmill, then turn left.", + "B": "Turn back and walk to the wall.", + "C": "First, you face the rowing machine, then you turn right and walk straight past the treadmill.", + "D": "First, you face the rowing machine, then you turn back and walk straight through the rowing machine.", + "E": "First, you face the rowing machine, then you turn back and walk straight through the treadmill." + }, + "answer_label": "E", + "answer_text": "First, you face the rowing machine, then you turn back and walk straight through the treadmill.", + "q_id": "scene0788_00-0_6", + "correct_answer": "First, you face the rowing machine, then you turn back and walk straight through the treadmill.", + "original_options": { + "A": "Face the treadmill, then turn left.", + "B": "Turn back and walk to the wall.", + "C": "First, you face the rowing machine, then you turn right and walk straight past the treadmill.", + "D": "First, you face the rowing machine, then you turn back and walk straight through the rowing machine.", + "E": "First, you face the rowing machine, then you turn back and walk straight through the treadmill." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 16, + 17 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_2_T2", + "qid_num": 11, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many treadmills did I see?", + "answer": "You see three treadmills; all of them are to your left.", + "question_chinese": "我看到多少个跑步机？", + "answer_chinese": "你看到三个跑步机，它们都在你左边。", + "evidence": [ + { + "description": "You see three treadmills.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:08" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You see one treadmill; it is to your left.", + "B": "You see three treadmills; all of them are to your left." + }, + "answer_label": "B", + "answer_text": "You see three treadmills; all of them are to your left.", + "q_id": "scene0788_00-0_2_T2", + "correct_answer": "You see three treadmills; all of them are to your left.", + "original_options": { + "A": "You see one treadmill; it is to your left.", + "B": "You see three treadmills; all of them are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_2_T1", + "qid_num": 12, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "How many treadmills did I see?", + "answer": "You see one treadmill; it is to your left.", + "question_chinese": "我看到多少个跑步机？", + "answer_chinese": "你看到一个跑步机，它在你左边。", + "evidence": [ + { + "description": "You see a treadmill.", + "timestamps": { + "start": "00:00:10", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "You see one treadmill; it is to your left.", + "B": "You see two treadmills; they are to your left." + }, + "answer_label": "A", + "answer_text": "You see one treadmill; it is to your left.", + "q_id": "scene0788_00-0_2_T1", + "correct_answer": "You see one treadmill; it is to your left.", + "original_options": { + "A": "You see one treadmill; it is to your left.", + "B": "You see two treadmills; they are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_7", + "qid_num": 13, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I use the rowing machine directly?", + "answer": "Yes,you can.", + "question_chinese": "我能直接使用划船机吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The rowing machine is right in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:01", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "answer_label": "B", + "answer_text": "Yes,you can.", + "q_id": "scene0788_00-0_7", + "correct_answer": "Yes,you can.", + "original_options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0788_00-0", + "video_length": "00:00:20", + "qid": "scene0788_00-0_8", + "qid_num": 14, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Reachability Judgment", + "question": "Can I use the rowing machine directly?", + "answer": "No,you can't.", + "question_chinese": "我能直接使用划船机吗？", + "answer_chinese": "可以。", + "evidence": [ + { + "description": "The rowing machine is next to the wall.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "You're next to the treadmill.", + "timestamps": { + "start": "00:00:09", + "end": "00:00:10" + } + }, + { + "description": "You are too far from the rowing machine.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:10", + "task_difficulty": "hard", + "qtype": "BINARY", + "options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "answer_label": "A", + "answer_text": "No,you can't.", + "q_id": "scene0788_00-0_8", + "correct_answer": "No,you can't.", + "original_options": { + "A": "No,you can't.", + "B": "Yes,you can." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + } +] diff --git a/QAs/scene0802_00-0-mcq.json b/QAs/scene0802_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..8a1c5e36cffae091e90a8b6f6550416ddceb5fe1 --- /dev/null +++ b/QAs/scene0802_00-0-mcq.json @@ -0,0 +1,676 @@ +[ + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_2_T1", + "qid_num": 1, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I have an empty Coke bottle in my hand— which trash can should I throw it into?", + "answer": "You should throw it into the trash can on the right side with the blue label. The trash cans are to your left.", + "question_chinese": "我手里有个刚喝完的可乐瓶，我应该扔到哪个垃圾桶里？", + "answer_chinese": "你应该扔到最右侧贴着蓝色标识的垃圾桶里。垃圾桶在你左边。", + "evidence": [ + { + "description": "On the far right side of the trash can is written 'cans and bottles'.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use the left trash can.", + "B": "Use the middle trash can.", + "C": "Throw it into the trash can on the left side with the blue label; the trash cans are to your left.", + "D": "You should throw it into the trash can on the right side with the blue label. The trash cans are to your left.", + "E": "Throw it into the trash can on the right side with the blue label; the trash cans are to your right." + }, + "answer_label": "D", + "answer_text": "You should throw it into the trash can on the right side with the blue label. The trash cans are to your left.", + "q_id": "scene0802_00-0_2_T1", + "correct_answer": "You should throw it into the trash can on the right side with the blue label. The trash cans are to your left.", + "original_options": { + "A": "Use the left trash can.", + "B": "Use the middle trash can.", + "C": "Throw it into the trash can on the left side with the blue label; the trash cans are to your left.", + "D": "You should throw it into the trash can on the right side with the blue label. The trash cans are to your left.", + "E": "Throw it into the trash can on the right side with the blue label; the trash cans are to your right." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 21, + 21 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_2_T2", + "qid_num": 2, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "I have an empty Coke bottle in my hand— which trash can should I throw it into?", + "answer": "You should throw it into the trash can on the right side with the blue label.", + "question_chinese": "我手里有个刚喝完的可乐瓶，我应该扔到哪个垃圾桶里？", + "answer_chinese": "你应该扔到最右侧贴着蓝色标识的垃圾桶里。", + "evidence": [ + { + "description": "On the far right side of the trash can is written 'cans and bottles'.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + }, + { + "description": "The finished cola bottle should be thrown into the trash can with a blue label on the far right.", + "timestamps": { + "start": "00:00:04", + "end": "00:00:05" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Use the left trash can.", + "B": "Use the front trash can.", + "C": "You should throw it into the trash can on the right side with the blue label.", + "D": "You should throw it into the trash can on the left side with the blue label.", + "E": "You should throw it into the trash can behind you with the blue label." + }, + "answer_label": "C", + "answer_text": "You should throw it into the trash can on the right side with the blue label.", + "q_id": "scene0802_00-0_2_T2", + "correct_answer": "You should throw it into the trash can on the right side with the blue label.", + "original_options": { + "A": "Use the left trash can.", + "B": "Use the front trash can.", + "C": "You should throw it into the trash can on the right side with the blue label.", + "D": "You should throw it into the trash can on the left side with the blue label.", + "E": "You should throw it into the trash can behind you with the blue label." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 16, + 14 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_10", + "qid_num": 3, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What is my movement trajectory?", + "answer": "First, stand on the left side of the trash can, then walk straight ahead until you reach the door.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先，你站在垃圾桶的左侧，然后一直向前走走到门旁边。", + "evidence": [ + { + "description": "First, stand on the left side of the trash can.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you walk straight ahead until you reach the door.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:16" + } + } + ], + "question_timestamps": "00:00:16", + "task_difficulty": "medium", + "qtype": "OTHER", + "options": { + "A": "First, stand on the left side of the trash can, then walk straight ahead until you reach the door.", + "B": "Stand right of the trash can first.", + "C": "Walk diagonally to the door.", + "D": "First, stand on the left side of the trash can, then walk straight ahead until you reach the window.", + "E": "First, stand on the left side of the trash can, then walk straight back until you reach the door." + }, + "answer_label": "A", + "answer_text": "First, stand on the left side of the trash can, then walk straight ahead until you reach the door.", + "q_id": "scene0802_00-0_10", + "correct_answer": "First, stand on the left side of the trash can, then walk straight ahead until you reach the door.", + "original_options": { + "A": "First, stand on the left side of the trash can, then walk straight ahead until you reach the door.", + "B": "Stand right of the trash can first.", + "C": "Walk diagonally to the door.", + "D": "First, stand on the left side of the trash can, then walk straight ahead until you reach the window.", + "E": "First, stand on the left side of the trash can, then walk straight back until you reach the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 5, + 19, + 19 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_4_T2", + "qid_num": 4, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory like?", + "answer": "First, you were on the left side of the trash cans, then you walked straight forward until you got next to the door.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "首先，你站在垃圾桶的左侧，然后一直向前走走到门旁边。", + "evidence": [ + { + "description": "First, stand on the left side of the trash can.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "Then you walk straight ahead until you reach the door.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:17" + } + } + ], + "question_timestamps": "00:00:19", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You started right of the trash cans.", + "B": "First, you were on the left side of the trash cans, then you walked straight forward until you got next to the door.", + "C": "You started in front of trash cans.", + "D": "First, you were on the right side of the trash cans, then you walked straight forward to the door.", + "E": "First, you were on the left side of the trash cans, then you walked straight backward until next to the door." + }, + "answer_label": "B", + "answer_text": "First, you were on the left side of the trash cans, then you walked straight forward until you got next to the door.", + "q_id": "scene0802_00-0_4_T2", + "correct_answer": "First, you were on the left side of the trash cans, then you walked straight forward until you got next to the door.", + "original_options": { + "A": "You started right of the trash cans.", + "B": "First, you were on the left side of the trash cans, then you walked straight forward until you got next to the door.", + "C": "You started in front of trash cans.", + "D": "First, you were on the right side of the trash cans, then you walked straight forward to the door.", + "E": "First, you were on the left side of the trash cans, then you walked straight backward until next to the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 7, + 7, + 19, + 21 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_4_T1", + "qid_num": 5, + "category": "Embodied trajectory & Movement Memory", + "subcategory": "My Trajectory", + "question": "What was my movement trajectory like?", + "answer": "You were standing in front of the door, then you walked right to the right side of the door.", + "question_chinese": "我的运动轨迹是怎样的？", + "answer_chinese": "你站在门前，然后向右走到了门的右边。", + "evidence": [ + { + "description": "You face the door and walk right to its side.", + "timestamps": { + "start": "00:00:15", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "You were standing in front of the door, then you walked right to the right side of the door.", + "B": "You walked left of the door.", + "C": "You stayed in front of the door.", + "D": "You were standing in front of the door, then you walked left to the left side of the door.", + "E": "You were standing behind the door, then you walked right to the right side of the door." + }, + "answer_label": "A", + "answer_text": "You were standing in front of the door, then you walked right to the right side of the door.", + "q_id": "scene0802_00-0_4_T1", + "correct_answer": "You were standing in front of the door, then you walked right to the right side of the door.", + "original_options": { + "A": "You were standing in front of the door, then you walked right to the right side of the door.", + "B": "You walked left of the door.", + "C": "You stayed in front of the door.", + "D": "You were standing in front of the door, then you walked left to the left side of the door.", + "E": "You were standing behind the door, then you walked right to the right side of the door." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 6, + 7, + 19, + 17 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_1_T2", + "qid_num": 6, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I see in the room?", + "answer": "There are three trash cans in the room.", + "question_chinese": "我看见房间里有几个垃圾桶？", + "answer_chinese": "房间里有三个垃圾桶。", + "evidence": [ + { + "description": "There are three trash cans in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "There are three trash cans in the room.", + "B": "There is one trash can.", + "C": "There are two trash cans.", + "D": "There are four trash cans scattered around the room instead of three.", + "E": "There are five trash cans in the room, not the three you recall seeing." + }, + "answer_label": "A", + "answer_text": "There are three trash cans in the room.", + "q_id": "scene0802_00-0_1_T2", + "correct_answer": "There are three trash cans in the room.", + "original_options": { + "A": "There are three trash cans in the room.", + "B": "There is one trash can.", + "C": "There are two trash cans.", + "D": "There are four trash cans scattered around the room instead of three.", + "E": "There are five trash cans in the room, not the three you recall seeing." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 14 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_1_T1", + "qid_num": 7, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans did I see in the room?", + "answer": "You saw three trash cans in the room. The trash cans are to your left.", + "question_chinese": "我看见房间里有几个垃圾桶？", + "answer_chinese": "你看见房间里有3个垃圾桶。垃圾桶在你左边。", + "evidence": [ + { + "description": "There are three trash cans in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:17", + "task_difficulty": "", + "qtype": "COUNT", + "options": { + "A": "You saw one trash can on your left.", + "B": "You saw two trash cans on your left.", + "C": "You saw four trash cans in the room, and they were positioned to your left.", + "D": "You saw five trash cans in the room, with the trash cans located to your left.", + "E": "You saw three trash cans in the room. The trash cans are to your left." + }, + "answer_label": "E", + "answer_text": "You saw three trash cans in the room. The trash cans are to your left.", + "q_id": "scene0802_00-0_1_T1", + "correct_answer": "You saw three trash cans in the room. The trash cans are to your left.", + "original_options": { + "A": "You saw one trash can on your left.", + "B": "You saw two trash cans on your left.", + "C": "You saw four trash cans in the room, and they were positioned to your left.", + "D": "You saw five trash cans in the room, with the trash cans located to your left.", + "E": "You saw three trash cans in the room. The trash cans are to your left." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 8, + 8, + 15, + 16 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_3_T1", + "qid_num": 8, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I saw it, was the door open?", + "answer": "No—the door in front of you is closed.", + "question_chinese": "我看见时门是开着的吗？", + "answer_chinese": "不是，你面前的门是关着的。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Yes—the door in front of you is open.", + "B": "No—the door in front of you is closed." + }, + "answer_label": "B", + "answer_text": "No—the door in front of you is closed.", + "q_id": "scene0802_00-0_3_T1", + "correct_answer": "No—the door in front of you is closed.", + "original_options": { + "A": "Yes—the door in front of you is open.", + "B": "No—the door in front of you is closed." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 9 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_3_T2", + "qid_num": 9, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "When I saw it, was the door open?", + "answer": "No.", + "question_chinese": "我看见时门是开着的吗？", + "answer_chinese": "不是。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "No.", + "B": "Yes." + }, + "answer_label": "A", + "answer_text": "No.", + "q_id": "scene0802_00-0_3_T2", + "correct_answer": "No.", + "original_options": { + "A": "No.", + "B": "Yes." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 1 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_1", + "qid_num": 10, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the room?", + "answer": "There are three trash cans in the room.", + "question_chinese": "房间里有几个垃圾桶？", + "answer_chinese": "房间里有三个垃圾桶。", + "evidence": [ + { + "description": "There are three trash cans in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:07", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There is one trash can.", + "B": "There are two trash cans.", + "C": "There are four trash cans in the room rather than three in total.", + "D": "There are three trash cans in the room.", + "E": "There are five trash cans in the room, more than the three stated." + }, + "answer_label": "D", + "answer_text": "There are three trash cans in the room.", + "q_id": "scene0802_00-0_1", + "correct_answer": "There are three trash cans in the room.", + "original_options": { + "A": "There is one trash can.", + "B": "There are two trash cans.", + "C": "There are four trash cans in the room rather than three in total.", + "D": "There are three trash cans in the room.", + "E": "There are five trash cans in the room, more than the three stated." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 13, + 13 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_2", + "qid_num": 11, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many trash cans are there in the room?", + "answer": "There are three trash cans in the room.", + "question_chinese": "房间里有几个垃圾桶？", + "answer_chinese": "房间里有三个垃圾桶。", + "evidence": [ + { + "description": "There are three trash cans in the room.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are three trash cans in the room.", + "B": "There is one trash can.", + "C": "There are two trash cans.", + "D": "There are four trash cans in the room, not the three mentioned.", + "E": "There are five trash cans in the room, which is more than three." + }, + "answer_label": "A", + "answer_text": "There are three trash cans in the room.", + "q_id": "scene0802_00-0_2", + "correct_answer": "There are three trash cans in the room.", + "original_options": { + "A": "There are three trash cans in the room.", + "B": "There is one trash can.", + "C": "There are two trash cans.", + "D": "There are four trash cans in the room, not the three mentioned.", + "E": "There are five trash cans in the room, which is more than three." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 5, + 5, + 12, + 13 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_5", + "qid_num": 12, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the door open?", + "answer": "No,it isn't.", + "question_chinese": "门是开着的吗？", + "answer_chinese": "不是。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:14", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "No,it isn't.", + "B": "Yes,it is." + }, + "answer_label": "A", + "answer_text": "No,it isn't.", + "q_id": "scene0802_00-0_5", + "correct_answer": "No,it isn't.", + "original_options": { + "A": "No,it isn't.", + "B": "Yes,it is." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + }, + { + "video_uid": "scene0802_00-0", + "video_length": "00:00:20", + "qid": "scene0802_00-0_6", + "qid_num": 13, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Object Category Recall", + "question": "Is the door open?", + "answer": "No,it isn't.", + "question_chinese": "门是开着的吗？", + "answer_chinese": "不是。", + "evidence": [ + { + "description": "The door is closed.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "easy", + "qtype": "BINARY", + "options": { + "A": "Yes,it is.", + "B": "No,it isn't." + }, + "answer_label": "B", + "answer_text": "No,it isn't.", + "q_id": "scene0802_00-0_6", + "correct_answer": "No,it isn't.", + "original_options": { + "A": "Yes,it is.", + "B": "No,it isn't." + }, + "optimization_rounds": 2, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3 + ] + } +] diff --git a/QAs/scene0804_00-0-mcq.json b/QAs/scene0804_00-0-mcq.json new file mode 100644 index 0000000000000000000000000000000000000000..e78364323972569e1ef337aaa8639c9c009f3211 --- /dev/null +++ b/QAs/scene0804_00-0-mcq.json @@ -0,0 +1,415 @@ +[ + { + "video_uid": "scene0804_00-0", + "video_length": "00:00:20", + "qid": "scene0804_00-0_2_T2", + "qid_num": 1, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can relative to me?", + "answer": "The trash can is directly in front of you.", + "question_chinese": "垃圾桶在我的哪里？", + "answer_chinese": "垃圾桶在你的正前方。", + "evidence": [ + { + "description": "The trash can is in front of you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:02" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly behind you.", + "B": "Directly to your left.", + "C": "The trash can is directly in front of you.", + "D": "The trash can is directly to your right, not in front of you.", + "E": "The trash can is above you rather than positioned directly in front of you." + }, + "answer_label": "C", + "answer_text": "The trash can is directly in front of you.", + "q_id": "scene0804_00-0_2_T2", + "correct_answer": "The trash can is directly in front of you.", + "original_options": { + "A": "Directly behind you.", + "B": "Directly to your left.", + "C": "The trash can is directly in front of you.", + "D": "The trash can is directly to your right, not in front of you.", + "E": "The trash can is above you rather than positioned directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 13, + 14 + ] + }, + { + "video_uid": "scene0804_00-0", + "video_length": "00:00:20", + "qid": "scene0804_00-0_2_T1", + "qid_num": 2, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the trash can relative to me?", + "answer": "The trash can is behind you.", + "question_chinese": "垃圾桶在我的哪里？", + "answer_chinese": "垃圾桶在你的后面。", + "evidence": [ + { + "description": "The trash can is behind you.", + "timestamps": { + "start": "00:00:17", + "end": "00:00:18" + } + } + ], + "question_timestamps": "00:00:20", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "In front of you.", + "B": "Directly to your right.", + "C": "The trash can is behind you.", + "D": "The trash can is to your left, not behind you at all.", + "E": "The trash can is above you rather than located behind you right now." + }, + "answer_label": "C", + "answer_text": "The trash can is behind you.", + "q_id": "scene0804_00-0_2_T1", + "correct_answer": "The trash can is behind you.", + "original_options": { + "A": "In front of you.", + "B": "Directly to your right.", + "C": "The trash can is behind you.", + "D": "The trash can is to your left, not behind you at all.", + "E": "The trash can is above you rather than located behind you right now." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "scene0804_00-0", + "video_length": "00:00:20", + "qid": "scene0804_00-0_1", + "qid_num": 3, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sheets of paper are there on the wall in total?", + "answer": "There are seven sheets of paper pasted on the wall.", + "question_chinese": "墙上一共贴着几张纸？", + "answer_chinese": "墙上贴着七张纸。", + "evidence": [ + { + "description": "There are seven sheets of paper pasted on the wall.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + } + ], + "question_timestamps": "00:00:03", + "task_difficulty": "easy", + "qtype": "COUNT", + "options": { + "A": "There are six sheets.", + "B": "There are seven sheets of paper pasted on the wall.", + "C": "There are eight sheets.", + "D": "There are nine sheets of paper pasted on the wall in total.", + "E": "There are ten sheets of paper pasted on the wall altogether, not seven." + }, + "answer_label": "B", + "answer_text": "There are seven sheets of paper pasted on the wall.", + "q_id": "scene0804_00-0_1", + "correct_answer": "There are seven sheets of paper pasted on the wall.", + "original_options": { + "A": "There are six sheets.", + "B": "There are seven sheets of paper pasted on the wall.", + "C": "There are eight sheets.", + "D": "There are nine sheets of paper pasted on the wall in total.", + "E": "There are ten sheets of paper pasted on the wall altogether, not seven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "scene0804_00-0", + "video_length": "00:00:20", + "qid": "scene0804_00-0_2", + "qid_num": 4, + "category": "Object Recall & Quantity Tracking", + "subcategory": "Quantity Change Tracking", + "question": "How many sheets of paper are there on the wall in total?", + "answer": "There are eleven sheets of paper pasted on the wall.", + "question_chinese": "墙上一共贴着几张纸？", + "answer_chinese": "墙上贴着11张纸。", + "evidence": [ + { + "description": "There are 7 sheets of paper pasted on the left wall.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:01" + } + }, + { + "description": "There are 4 sheets of paper pasted on the right wall.", + "timestamps": { + "start": "00:00:12", + "end": "00:00:13" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "medium", + "qtype": "COUNT", + "options": { + "A": "There are nine sheets.", + "B": "There are fifteen sheets.", + "C": "There are five sheets of paper pasted on the wall in total.", + "D": "There are eleven sheets of paper pasted on the wall.", + "E": "There are three sheets of paper pasted on the wall altogether, not eleven." + }, + "answer_label": "D", + "answer_text": "There are eleven sheets of paper pasted on the wall.", + "q_id": "scene0804_00-0_2", + "correct_answer": "There are eleven sheets of paper pasted on the wall.", + "original_options": { + "A": "There are nine sheets.", + "B": "There are fifteen sheets.", + "C": "There are five sheets of paper pasted on the wall in total.", + "D": "There are eleven sheets of paper pasted on the wall.", + "E": "There are three sheets of paper pasted on the wall altogether, not eleven." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 4, + 4, + 12, + 13 + ] + }, + { + "video_uid": "scene0804_00-0", + "video_length": "00:00:20", + "qid": "scene0804_00-0_4_T2", + "qid_num": 5, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the printer or the black trash can?", + "answer": "Right now, the black trash can is closer to you, and it's directly in front of you.", + "question_chinese": "打印机和黑色垃圾桶哪个离我更近？", + "answer_chinese": "现在黑色垃圾桶离你更近，黑色垃圾桶在你的正前方。", + "evidence": [ + { + "description": "The black trash can is closer to you.", + "timestamps": { + "start": "00:00:00", + "end": "00:00:03" + } + } + ], + "question_timestamps": "00:00:05", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, the printer is closer to you, and it's directly in front of you.", + "B": "Right now, the black trash can is closer to you, and it's directly in front of you." + }, + "answer_label": "B", + "answer_text": "Right now, the black trash can is closer to you, and it's directly in front of you.", + "q_id": "scene0804_00-0_4_T2", + "correct_answer": "Right now, the black trash can is closer to you, and it's directly in front of you.", + "original_options": { + "A": "Right now, the printer is closer to you, and it's directly in front of you.", + "B": "Right now, the black trash can is closer to you, and it's directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 15 + ] + }, + { + "video_uid": "scene0804_00-0", + "video_length": "00:00:20", + "qid": "scene0804_00-0_4_T1", + "qid_num": 6, + "category": "Perceived Proximity & Reachability Memory", + "subcategory": "Distance Comparison Memory", + "question": "Which is closer to me, the printer or the black trash can?", + "answer": "Right now, the printer is closer to you, and it's directly in front of you.", + "question_chinese": "打印机和黑色垃圾桶哪个离我更近？", + "answer_chinese": "现在打印机离你更近,打印机在你的正前方。", + "evidence": [ + { + "description": "The printer is closer to you.", + "timestamps": { + "start": "00:00:11", + "end": "00:00:12" + } + } + ], + "question_timestamps": "00:00:15", + "task_difficulty": "", + "qtype": "BINARY", + "options": { + "A": "Right now, the black trash can is closer to you, and it's directly in front of you.", + "B": "Right now, the printer is closer to you, and it's directly in front of you." + }, + "answer_label": "B", + "answer_text": "Right now, the printer is closer to you, and it's directly in front of you.", + "q_id": "scene0804_00-0_4_T1", + "correct_answer": "Right now, the printer is closer to you, and it's directly in front of you.", + "original_options": { + "A": "Right now, the black trash can is closer to you, and it's directly in front of you.", + "B": "Right now, the printer is closer to you, and it's directly in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 17 + ] + }, + { + "video_uid": "scene0804_00-0", + "video_length": "00:00:20", + "qid": "scene0804_00-0_3_T2", + "qid_num": 7, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the yellow tote bag relative to me?", + "answer": "The yellow tote bag is to your right.", + "question_chinese": "黄色手提袋在我的哪里？", + "answer_chinese": "黄色手提袋在你的右边。", + "evidence": [ + { + "description": "On your right.", + "timestamps": { + "start": "00:00:05", + "end": "00:00:06" + } + } + ], + "question_timestamps": "00:00:08", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "To your left.", + "B": "Behind you.", + "C": "The yellow tote bag is to your right.", + "D": "The yellow tote bag is directly in front of you, not to your right.", + "E": "The yellow tote bag is above you rather than positioned to your right side." + }, + "answer_label": "C", + "answer_text": "The yellow tote bag is to your right.", + "q_id": "scene0804_00-0_3_T2", + "correct_answer": "The yellow tote bag is to your right.", + "original_options": { + "A": "To your left.", + "B": "Behind you.", + "C": "The yellow tote bag is to your right.", + "D": "The yellow tote bag is directly in front of you, not to your right.", + "E": "The yellow tote bag is above you rather than positioned to your right side." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 2, + 14, + 14 + ] + }, + { + "video_uid": "scene0804_00-0", + "video_length": "00:00:20", + "qid": "scene0804_00-0_3_T1", + "qid_num": 8, + "category": "Ego-centric Position & Orientation Memory", + "subcategory": "Relative Position & Orientation Memory", + "question": "Where is the yellow tote bag relative to me?", + "answer": "The yellow tote bag is directly in front of you.", + "question_chinese": "黄色手提袋在我的哪里？", + "answer_chinese": "黄色手提袋在你的正前方。", + "evidence": [ + { + "description": "Right in front of you.", + "timestamps": { + "start": "00:00:08", + "end": "00:00:10" + } + } + ], + "question_timestamps": "00:00:12", + "task_difficulty": "", + "qtype": "OTHER", + "options": { + "A": "Directly behind you.", + "B": "Directly to your right.", + "C": "The yellow tote bag is directly to your left, not directly in front of you.", + "D": "The yellow tote bag is directly in front of you.", + "E": "The yellow tote bag is directly above you rather than being positioned in front of you." + }, + "answer_label": "D", + "answer_text": "The yellow tote bag is directly in front of you.", + "q_id": "scene0804_00-0_3_T1", + "correct_answer": "The yellow tote bag is directly in front of you.", + "original_options": { + "A": "Directly behind you.", + "B": "Directly to your right.", + "C": "The yellow tote bag is directly to your left, not directly in front of you.", + "D": "The yellow tote bag is directly in front of you.", + "E": "The yellow tote bag is directly above you rather than being positioned in front of you." + }, + "optimization_rounds": 1, + "nli_issues_remaining": 0, + "option_build_source": "llm_optimized", + "llm_option_model": "gpt-5.2", + "nli_review_method": "gpt_as_nli", + "option_length_counts": [ + 3, + 4, + 15, + 16 + ] + } +]